sample_number int64 8 203k | abstract stringlengths 21 17.4k | article stringlengths 0 18.9k | compressed_article stringlengths 0 10.9k | compression_time float64 0 0.78 | original_tokens int64 0 3.79k | compressed_tokens int64 0 2.29k | token_saving stringclasses 2 values |
|---|---|---|---|---|---|---|---|
154,579 | in the @xmath0-user single - input single - output ( siso ) frequency - selective fading interference channel , it is shown that the maximal achievable multiplexing gain is almost surely @xmath1 by using interference alignment ( ia ) . however , when the signaling dimensions are limited , allocating all the resources to all users simultaneously is not optimal .
so , a group based interference alignment ( gia ) scheme is proposed , and it is formulated as an unbounded knapsack problem .
optimal and greedy search algorithms are proposed to obtain group patterns .
analysis and numerical results show that the gia scheme can obtain a higher multiplexing gain when the resources are limited .
interference channel , interference alignment , multiplexing gain , knapsack problem . | management is an important problem in wireless system design . as an effective technique for interference management , interference alignment ( ia )
is first considered in @xcite , @xcite as a coding technique for the two - user multiple - input multiple - out ( mimo ) x channel . using this scheme
, it is shown that each user can obtain almost surely a multiplexing gain ( mg ) of @xmath2 per channel use in the @xmath0-user siso interference channel ( ic ) @xcite .
a beamforming matrices optimized ia ( called bf - ia ) scheme is proposed in @xcite which can obtain a higher mg than the ia scheme in @xcite at any given number of channel realizations .
ia scheme is also applied in cellular networks in @xcite which can boost system performance in some scenarios .
however , when the signaling dimensions are limited , allocating all the resources to all users simultaneously is not optimal . in this letter , a gia scheme is proposed based on the bf - ia scheme which can obtain a higher mg when the resources are limited .
consider the @xmath0-user frequency - selective fading ic model : @xmath3}=\sum_{l=1}^k{\mathbf{h}^{[kl]}\mathbf{x}^{[l]}+\mathbf{z}^{[k]}},~ \forall k\in \{1,\dots , k\}.\ ] ] @xmath4}$ ] is the @xmath5 input signal vector of the @xmath6 transmitter , and @xmath7}$ ] is the channel output at the @xmath8 receiver , where @xmath9 is the number of available frequency - selective channel realizations .
@xmath10}$ ] is the diagonal channel matrix between transmitter @xmath11 and receiver @xmath12 .
we assume that all @xmath10}$ ] s are known in advance at all the transmitters and all the receivers , and assume the channel is time - invariant .
@xmath13}$ ] is @xmath5 additive white gaussian noise ( awgn ) vector at the @xmath8 receiver , where all noise terms are independent identically distributed ( i.i.d . )
zero - mean complex gaussian with unit variance .
the definition of achievable sum - rate follows from @xcite .
define @xmath14 as the mg @xcite , where @xmath15 is the achievable sum - rate in the @xmath0-user ic , and @xmath16 is defined as the total transmit power across all transmitters .
the frequency - selective channel realizations are called channel uses or resources for convenience in this letter .
let @xmath17 , @xmath18 ( @xmath19 is a positive integer ) , and let @xmath20},d^{[2]},\dots , d^{[k]})=((n+1)^n , n^n,\dots , n^n)$ ] be the numbers of streams allocated to the @xmath0 users respectively .
then @xmath21},r^{[2]},\dots , r^{[k]}\}=\bigg\{\frac{(n+1)^n}{(n+1)^n+n^n},\\ \frac{n^n}{(n+1)^n+n^n},\dots,\frac{n^n}{(n+1)^n+n^n}\bigg\}\end{gathered}\ ] ] are the achievable mgs of the @xmath0-user ic over @xmath9 channel uses .
so , the total achievable mg over @xmath9 channel uses is @xcite @xmath22}}=\frac{(n+1)^n+(k-1)n^n}{(n+1)^n+n^n}.\ ] ] when @xmath34 , we have @xmath35 hence , allocating all the resources to all users simultaneously is mg optimal . using ( [ eq6 ] ) ,
[ fig1 ] illustrates the achievable mg when @xmath36 and @xmath27 .
it can be seen that when @xmath37 allocating all the resources to all users simultaneously ( @xmath27 ) can obtain more mg , and when @xmath38 allocating them to partial users @xmath39 can obtain more mg .
for example , when @xmath40 , @xmath41 while @xmath42 . for general values of @xmath9 and @xmath0 ,
a natural question is how to allocate resources among users can obtain more mg . in the following section a gia scheme is proposed based on the bf - ia scheme .
let @xmath9 be the total resources that can be used for ia in the @xmath0-user ic .
let @xmath43 , @xmath44 and @xmath45 , and let @xmath46 & k\geq 3\\ 0 & k < 3 \end{cases}\ ] ] be the relative mg that the bf - ia scheme obtained compared to orthogonal multiplexing scheme ( when @xmath47 , orthogonal multiplexing is mg optimal ) . define @xmath48 as a group pattern . for example , in fig .
[ fig2]a we have @xmath49 while in fig .
[ fig2]b we have @xmath50 , @xmath51 , and @xmath52 .
let @xmath53 be the set of all the group patterns over @xmath12 users exactly , where @xmath54 , @xmath55 .
define the relative mg obtained per dimension as @xmath56 & k\geq 3\\ 0 & k<3 \end{cases}\ ] ] which is the _ efficiency _ of a group pattern .
for example , in fig .
[ fig2]a we have @xmath57 while in fig .
[ fig2]b we have @xmath58 , @xmath59 , and @xmath60 .
let @xmath61 be the set of all the group patterns over any @xmath12 users when @xmath62 .
we denote the elements of @xmath63 as @xmath64 , @xmath65 , and @xmath66 respectively . if there exist two group patterns @xmath63 and @xmath67 with @xmath68 and @xmath69 in the set @xmath70 , then it would always be better ( or at least not worse ) to choose @xmath63 .
hence , @xmath67 is removed from the set @xmath70 , and we obtain @xmath71 , @xmath72 . for example , if @xmath73 and @xmath74 , then @xmath67 is removed from the set @xmath70 .
the elements of @xmath75 are sorted by @xmath76 in non - increasing order , and we obtain @xmath77 .
for example , when @xmath78 and @xmath79 , we have @xmath80 , @xmath81 , @xmath82 , @xmath83 where @xmath84 , @xmath85 , @xmath86 , @xmath87 , @xmath88 , and @xmath89 . given @xmath0 and @xmath9 , we have obtained all the group patterns sorted by efficiency , and the question becomes how to choose among them so as to obtain more mg .
it is modeled as an unbounded knapsack problem which is np - hard @xcite .
the corresponding integer programming formulation is given as follows .
@xmath90 we denote the solution values by @xmath91 ( optimal algorithm ) and @xmath92 ( greedy algorithm ) , and denote the solution sets by @xmath93 ( optimal algorithm ) and @xmath94 ( greedy algorithm ) respectively . and @xmath27 .
] as a standard dynamic programming algorithm , unbounded - dp @xcite is adapted to evaluate our problem . +
@xmath95 + @xmath96 for @xmath97 to @xmath9 do @xmath98 , @xmath99 % _ initialization _ for @xmath100 to @xmath101 do for @xmath102 to @xmath9 do % _ @xmath103 may be packed _
if @xmath104 then @xmath105 @xmath106 @xmath107 , @xmath108 repeat % _ recover the optimal solution set _ @xmath109 @xmath110 @xmath111 until @xmath112 @xmath113 ,
@xmath114 + @xmath95 let @xmath115 , @xmath116 ( @xmath117 ) .
[ fig2]b shows the search results of the optimal algorithm .
the total mg obtained is @xmath118 . using ( [ eq6 ] )
, the mg obtained by the bf - ia scheme is about @xmath119 .
so , about @xmath120 more mg is obtained by making full use of @xmath121 channel uses .
the computation complexity of this optimal algorithm is @xmath122 , and it is a pseudopolynomial algorithm @xcite . when @xmath9 is large , the optimal algorithm will need prohibitive time to obtain the solutions even using a powerful computer .
so , a greedy algorithm is proposed in the following : + @xmath95 + @xmath123 @xmath97 , @xmath124 , @xmath107 for @xmath100 to @xmath101 do if @xmath125 then @xmath126 @xmath127 @xmath128 @xmath129 @xmath130 ,
@xmath131 + @xmath95 the computation complexity of the greedy algorithm is @xmath132 .
however , as the greedy algorithm does best every step which is locally optimal , global optimum is not confirmed .
according to _ theorem _ 8.5.1 in @xcite , the relative performance guarantee ( @xmath133 ) of the greedy algorithm is bounded by @xmath2 .
[ fig2]c shows the search results of the greedy algorithm when @xmath78 and @xmath79 . in this example , the total mg obtained is @xmath134 , and the relative performance guarantee is about @xmath135 .
when each user is equipped with @xmath136 antennas , let @xmath137 , @xmath138 , and @xmath139 . using _ corollary _
@xmath140 in @xcite , the achievable mg is @xmath141 .
so , the gia scheme and the search algorithms can be readily extended to this scenario .
the stream allocations among users are non - uniform both in the bf - ia and the gia schemes .
hence , dynamic algorithm should be considered to balance unfairness among users .
there is a simple solution , that is choosing the user served with more streams from all users periodically .
also , sophisticated schemes can be designed to incorporate other factors .
however , we leave it for future work .
the comparison of the gia and the bf - ia schemes in the @xmath0-user ic is presented in fig .
we only compare the achievable mg when @xmath142 in fig .
[ fig3]a ( or @xmath143 in fig .
[ fig3]b ) , as the bf - ia scheme has no solution when @xmath144 ( or @xmath145 ) .
and we also only compare the greedy algorithm of the gia scheme with the bf - ia scheme , as the optimal algorithm is only feasible when @xmath9 is small .
[ fig3 ] shows that the proposed gia scheme obtains a higher mg when the resources below a certain value ( e.g. , about @xmath146 in fig .
[ fig3]a ) .
when the resources are large , as discussed above , allocating all them to all users simultaneously is mg optimal .
the unsteadiness of the curves of the gia scheme is because the discrete nature of the problem .
the gia scheme is preferred to be used when the resources are limited .
in this letter , under the same ia conditions as in @xcite , a group based ia scheme is proposed .
analysis and numerical results show that the gia scheme obtains a higher mg in comparison with the bf - ia scheme when the resources are limited . and @xmath79 .
( a ) the bf - ia scheme , which allocates all the resources to @xmath147 users simultenously .
( b ) the gia scheme ( optimal algorithm ) , where @xmath9 is divided into 4 parts ( @xmath148 , @xmath149 , and @xmath150 ) .
( c ) the gia scheme ( greedy algorithm ) , where @xmath9 is divided into 6 parts ( @xmath148 , @xmath151 , @xmath152 , @xmath153 , @xmath154 , and @xmath155).,width=291,height=113 ] m. maddah - ali , a. motahari , and a. khandani , communication over mimo x channel : interference alignment , decomposition , and performance analysis , " _ ieee trans .
inf . theory _ ,
54 , no . 8 , pp . 3457 - 3470 , aug . 2008 . | management important problem in wireless system design technique interference management alignment considered in @xcite coding technique two - user multiple - input - out x channel each user multiplexing gain of @xmath2 per channel use in @xmath0-user siso interference channel @xcite beamforming matrices optimized ia bf - ia scheme proposed in @xcite higher mg than at channel realizations applied in cellular networks system performance signaling dimensions limited allocating resources to users simultaneously not optimal gia scheme proposed based bf - ia higher mg when resources limited @xmath0-user frequency - selective fading ic model @xmath3} @xmath4}$ @xmath5 input signal vector of @xmath6 transmitter @xmath7} channel output at @xmath8 receiver @xmath9 number of available frequency - selective channel realizations @xmath10}$ diagonal channel matrix between transmitter @xmath11 and receiver @xmath12 assume all @xmath10 s known at transmitters receivers channel time - invariant@xmath13} @xmath5 additive white gaussian noise vector @xmath8 receiver noise terms independent distributed zero - mean complex gaussian unit variance definition achievable sum - rate from @xcite @xmath14 mg @xcite @xmath15 achievable sum - rate @xmath0-user ic @xmath16 total transmit power across transmitters frequency - selective channel realizations channel uses resources @xmath17 @xmath18 positive integer @xmath20}=((n+1)^n numbers streams allocated to @xmath0 users @xmath21},r{[2]}{(n+1)^n^n+n^n achievable mgs @xmath0-user ic over @xmath9 channel uses total achievable mg over @xmath9 channel uses is @xcite @xmath22}}=\frac{(n+1)^n+^n}{(n+1)^n+n^n} @xmath34 @xmath35 allocating resources to users simultaneously optimaleq6 fig1 ] illustrates achievable mg @xmath36 @xmath27 @xmath37 allocating resources to users more mg @xmath38 allocating partial users @xmath39 more mg @xmath40 @xmath41 @xmath42 general values @xmath9 @xmath0 question allocate resources among users obtain more mg gia scheme proposed bf - ia scheme @xmath9 be total resources for ia in @xmath0-user ic @xmath43 @xmath44 @xmath45 @xmath46 & k\geq 3\\ 0 & k < 3 be relative mg bf - ia scheme compared to orthogonal multiplexing scheme @xmath47 orthogonal multiplexing mg optimal define @xmath48 as group pattern fig @xmath49 @xmath50 @xmath52 @xmath53 set group patterns over @xmath12 users @xmath54 @xmath55 relative mg obtained per dimension as @xmath56 & k\geq 3\\ 0 & k<3 efficiency of group pattern @xmath57 @xmath58 @xmath61 set group patterns over @xmath12 users @xmath62 denote elements @xmath63 as @xmath64 two group patterns @xmath63 and @xmath67 with @xmath68 in set @xmath70 better choose @xmath63 @xmath67 removed from set @xmath70 obtain @xmath71 @xmath72 @xmath73 and @xmath74 @xmath67 removed from set elements @xmath75 sorted by @xmath76 non - increasing obtain @xmath77 @xmath78 and @xmath79 have @xmath80 where @xmath84 @xmath86 @xmath89 @xmath0 and @xmath9 obtained group patterns sorted by efficiency question how to choose obtain more mg modeled as unbounded knapsack problem np - hard @xcite integer programming formulation @xmath90 denote solution values by @xmath91 @xmath92 solution sets by @xmath93 @xmath27 standard dynamic programming algorithm unbounded - dp @xcite adapted to evaluate problem@xmath95 @xmath96 @xmath97 @xmath9 @xmath98 @xmath99 initialization @xmath100 to @xmath101 @xmath102 @xmath9 @xmath103 packed @xmath104 @xmath105 @xmath106 repeat recover optimal solution @xmath109 @xmath110 @xmath112 @xmath95 @xmath115 @xmath116 @xmath117 fig2]b search results optimal algorithm total mg @xmath118 mg bf - ia scheme @xmath119 @xmath120 more mg @xmath121 channel uses computation complexity @xmath122 pseudopolynomial algorithm @xcite @xmath9 large prohibitive time solutions greedy algorithm proposed @xmath95 @xmath123 @xmath97 @xmath124 @xmath107 @xmath100 to @xmath101 @xmath125 @xmath126 @xmath127 @xmath128 @xmath129 @xmath130 @xmath131 @xmath95 computation complexity greedy algorithm @xmath132 locally optimal global optimum not confirmed theorem 8.5.1 relative performance guarantee @xmath133 greedy algorithm bounded by @xmath2 fig2]c search results greedy algorithm @xmath78 @xmath79example total mg is @xmath134 relative performance guarantee @xmath135 each user equipped with @xmath136 antennas let @xmath137 @xmath138 @xmath139 using @xmath140 in @xcite achievable mg is @xmath141 gia scheme search algorithms extended scenario stream allocations non - uniform in bf - ia gia schemes dynamic algorithm balance unfairness simple solution choosing user served with more streams periodically sophisticated schemes incorporate other factors future work comparison of gia bf - ia schemes in @xmath0-user ic presented in fig . compare achievable mg when @xmath142 or @xmath143 bf - ia scheme no solution when @xmath144 or @xmath145 ) compare greedy algorithm gia scheme with bf - ia scheme optimal algorithm feasible when @xmath9 small proposed gia scheme obtains higher mg when resources below certain value @xmath146 when resources large allocating all to users simultaneously optimal unsteadiness of curves gia scheme discrete nature problem preferred when resources limited same conditions group based ia scheme proposedanalysis results show gia scheme obtains higher mg bf - ia scheme resources limited @xmath79 bf - ia scheme allocates resources to @xmath147 users gia scheme optimal algorithm @xmath9 divided into 4 parts gia scheme greedy algorithm @xmath9 divided 6 parts,width=291,height=113 m. maddah - ali a. motahari a. khandani communication over mimo x channel interference alignment decomposition performance analysis trans . inf theory 54 , no . 8 pp . 3457 - 3470 aug . 2008 | 0.434877 | 2,824 | 1,446 | , Saving $0.1 in GPT-4. |
154,580 | in this paper , i review the results of 3-d evolution of the inner heliosphere over the solar cycle # 23 , based on observations of interplanetary scintillation ( ips ) made at 327 mhz using the ooty radio telescope .
the large - scale features of solar wind speed and density turbulence of the current minimum are remarkably different from that of the previous cycle .
the results on the solar wind density turbulence show that ( 1 ) the current solar minimum is experiencing a low level of coronal density turbulence , to a present value of @xmath050% lower than the previous similar phase , and ( 2 ) the scattering diameter of the corona has decreased steadily after the year 2003 .
the results on solar wind speed are consistent with the magnetic - field strength at the poles and the warping of heliospheric current sheet . | in this study , a large amount of interplanetary scintillation ( ips ) data obtained from the ooty radio telescope ( ort ) , operating at 327 mhz ( swarup et al .
1971 ) , has been employed to study the 3-d evolution of the heliosphere over the period 19892009 .
the ips observations made with the ort can provide the velocity of the solar wind and the scintillation index ( _ m _ ) in the heliocentric distance range of @xmath110250 solar radii ( @xmath2 ) and at all heliographic latitudes .
the value of _ m _ is a measure of electron - density turbulence in the solar wind ( @xmath3 ) , along the line of sight ( _ z _ ) to the radio source ( e.g. , manoharan et al .
2000 ) . the normalized scintillation index , @xmath4 ( i.e. , observed index normalized by its long - term average ) , enables the comparison of levels of density turbulence obtained from different sources . however , the value of _ g _ is linearly related to @xmath5 only in the weak - scattering region at distances @xmath640 @xmath2 .
for example , an _
m - r _ profile attains the peak value at the strong - to - weak scattering transition point , which typically occurs @xmath040 @xmath2 for ips at 327 mhz ( e.g. , manoharan 1993 ; 2006 ) . in this study
, the solar wind velocity and turbulence images have been exclusively obtained from weak - scattering data . however
, the contour of constant level of turbulence in a year at different latitudes has been determined using peaks of several _ m - r _ profiles .
figure 1a shows the latitudinal distributions of solar wind speed and density turbulence ( _ g _ ) observed at ooty over the solar cycle # 23 .
these plots are similar to the well - known ` _ butterfly diagram _ ' of photospheric magnetic field intensity .
they have been made by tracing backward / forward from the measurement location onto a sphere of radius @xmath0100 r@xmath7 , which approximately corresponds to the mid range of distances covered in the observations utilized to generate the plots . .
( b ) the stack plot ( _ right _ ) shows geo - magnetic disturbance index , kp , intensity of polar magnetic field , tilt angle of heliospheric current sheet ( hcs ) , and solar activities ( i.e. , sunspot number and solar radio flux density at 10.7 cm ) .
, title="fig:",width=253 ] .
( b ) the stack plot ( _ right _ ) shows geo - magnetic disturbance index , kp , intensity of polar magnetic field , tilt angle of heliospheric current sheet ( hcs ) , and solar activities ( i.e. , sunspot number and solar radio flux density at 10.7 cm ) .
, title="fig:",width=226 ] it is evident in the ` _ latitude - year _ ' speed plot that during minimum of the solar cycle , polar regions are dominated by high speed streams ( @xmath0600800 kms@xmath8 ) from open - field coronal holes and low and variable flow speeds ( @xmath9500 kms@xmath8 ) are observed at the low- and mid - latitude regions of the complex / closed field corona .
but , there are marked differences in latitudinal extents of low- and high - speed flow regions between the current and previous minimum phases .
for example during 199697 , the low - speed flow is confined to @xmath0@xmath10 of the equatorial belt ; whereas at the current minimum , it extends to a latitude range of @xmath0@xmath11 .
these low - speed wind widths also correlate with the latitudinal warping ( tilt angle ) of heliospheric current sheet ( hcs ) , respectively , a small amplitude , @xmath1215@xmath13 , at the previous minimum and a moderate amplitude , @xmath0@xmath1430@xmath13 , at the current minimum ( figure 1b ) .
these results suggest ( i ) a near - dipole magnetic field of the sun for the previous minimum , around the year 1997 and ( ii ) a never - approached dipole - field geometry during the current minimum .
it is in good agreement with the result from other independent ips speed measurements obtained from the solar - terrestrial environment laboratory ( e.g. , tokumaru et al .
thus , the hcs tilt of the corona of the current phase tends to resemble a condition similar to that of moderate activity , but without activity ! these changes in the latitudinal extents of low - speed wind
have also influenced the high - speed flows from the polar regions . in the current minimum phase ,
high - speed regions at the poles have remarkably shrunk towards the poles ( figure 1a , _ top _ ) . moreover ,
the speed of high - latitude ( @xmath650@xmath13 ) wind is considerably less for the current minimum than that of the previous minimum .
these findings nicely correlate with the polar field strength , which is @xmath04050% weaker for the current minimum phase .
the magnetic pressure associated with the polar coronal holes seems to determine the acceleration of the high - speed wind .
the weak field may be due to the fact that the polar field has not fully developed after the field reversal around the year 2000 ( figure 1b ) .
the drifting of density structures from high to low latitudes , seen in figure 1a ( _ bottom _ ) , is caused by the slow and gradual movement of concentrated magnetic - field regions of the corona .
it is likely due to the migration of small / medium - size coronal holes from polar to low latitude regions and the high - speed wind from these coronal holes interacting with the low - speed wind , causing compression in front of the high - speed stream .
the latitudinal spread of density patterns is also consistent with the hcs tilt angle ( figure 1b ) , which is maximum at the time of polarity reversal of the cycle and during when a large number of coronal mass ejections ( cmes ) dominate the heliosphere ( e.g. , yashiro et al . 2004 ) .
the density plot is also consistent with the solar wind disappearance period , around mid 1999 ( low level of density turbulence ) and co - rotating interaction regions ( cirs ) dominated heliosphere in the first half of 2003 ( intense density turbulence along the latitudinal direction ) . it may be noted that during late october and early november 2003 , a number of cme events prevailed the interplanetary medium .
however , the effect of a cme in the sun - earth distance is rather limited to 24 days after its onset and weakens with solar distance .
whereas , in the case of cir events , the influence of each event is seen for several days , with a systematic increase in density turbulence with radial distance .
moreover , they show a latitudinal pattern . for cir events during 2003 , the latitudinal distribution and the radial evolution of enhancement of turbulence have been observed ( manoharan 2008 ) .
the cir - dominated period , in the first half of 2003 , is in agreement with the moderate - to - severe storms observed at the earth ( figure 1b , refer to kp index plot ) .
as observed in the speed plot , the latitudinal extents of low - turbulence regions at the poles also show remarkable changes between the current and previous minimum phases .
the average level of turbulence at the current minimum seems to be considerably lower than that of the previous cycle .
figure 2 displays the constant @xmath15 plots , at different phases of the solar cycle .
the plot for the year 1997 has been limited by weak - scattering observations and the last plot includes data up to may 2009 . in general , a given level of turbulence is observed closer to the sun at the poles than at the equator . however , depending on the phase of the solar cycle , the diameter of the contour can vary along the poles , but , remains nearly the same along the equator ( e.g. , manoharan 1993 ) . the important point to note in this analysis is that after the year 2003 , the overall diameter of @xmath15 contour has gradually decreased with respect to the sun s center . in other words ,
the same level of turbulence seems to move close to the sun .
thus , the radial dependence of turbulence ( @xmath16@xmath0@xmath17 ^ 2 $ ] , which typically varies as @xmath18 ) suggests that the scattering diameter of the corona has gradually shrunk towards the sun . in other words ,
the scattering power ( @xmath16 ) has remained nearly same at all latitudes between 1989 and 2003 , but , decreased @xmath050% around middle of 2009 at low - latitudes .
the present large - scale 3-d features of solar wind speed and density turbulence are remarkably different from that of the previous cycle . in the current minimum phase ,
the extent of low - speed region along the equatorial belt is considerably wider than that of the previous cycle ; whereas the high - speed regions have shrunk towards the poles in contrast to low - latitude extent of the previous cycle .
the other important result of this study is that after the year 2003 , the overall scattering diameter of the corona has gradually decreased with respect to the sun s center . these results are consistent with the ecliptic and off - ecliptic studies ( e.g. , lee et al .
2009 ; mccomas et al . 2008
; smith & balogh 2008 ; tokumaru et al .
2009 ) .
the weak fields observed at the poles , as well as corresponding solar wind speed and density turbulence for the current low activity , are possibly caused by the changes in the movement of large - scale fields , as the reversal of polarity progresses .
it is linked to the rate of poleward and equatorward meridional flows , which transport the unbalanced magnetic flux ( e.g. , sheeley 2008 ) .
moreover , the flux - transport dynamo has predicted weak polar fields and a long solar cycle ( e.g. , choudhuri et al . 2007 ) .
i thank all the members of the radio astronomy centre for making the ooty radio telescope available for ips observations .
i also thank national space science data center for omni data and the wilcox solar observatory for the magnetic - field data .
this work is partially supported by the cawses - india program , which is sponsored by isro . | study interplanetary scintillation ips data from ooty radio telescope at 327 mhz 3-d evolution heliosphere 19892009 ips observations provide velocity solar wind scintillation index in heliocentric distance @xmath110250 solar radii all heliographic latitudes value _ m _ electron - density turbulence in solar wind @xmath3 along line of sight radio source 2000 normalized scintillation index @xmath4 enables comparison of density turbulence value _ g _ linearly related to @xmath5 in weak - scattering region at distances @xmath640 @xmath2 _ m - r _ profile peak at strong - to - weak scattering transition point @xmath040 @xmath2 327 mhz solar wind velocity turbulence images from weak - scattering data contour of constant level turbulence at different latitudes determined using peaks of several _ m - r _ profiles figure 1a shows latitudinal distributions of solar wind speed density turbulence at ooty over solar cycle # 23 plots similar to butterfly diagram of photospheric magnetic field intensitymade by tracing from measurement location sphere radius @xmath0100 r@xmath7 corresponds to mid range distances observations plots stack plot shows geo - magnetic disturbance index intensity polar magnetic field tilt angle heliospheric current sheet solar activities solar radio flux density at 10.7 stack plot shows geo magnetic disturbance index intensity tilt angle solar activities latitude - year speed plot during minimum solar cycle polar regions dominated by high speed streams @xmath0600800 kms@xmath8 from open - field coronal holes low variable flow speeds ( @xmath9500 kms@xmath8 at low- mid - latitude regions of complex closed field corona marked differences in latitudinal extents low- high - speed flow regions between current previous minimum phases during 199697 low - speed flow confined to @xmath0@xmath10 equatorial belt current minimum extends to latitude @xmath0@xmath11 low - speed wind widths correlate with latitudinal warping of heliospheric current sheet small amplitude at previous minimum moderate amplitude at current minimum results suggest near - dipole magnetic field of sun for previous minimum 1997 never - approached dipole - field geometry current minimum agreement with independent ips speed measurements from solar - terrestrial environment laboratory hcs tilt of corona current phase moderate activity changes in latitudinal extents low - speed wind influenced high - speed flows from polar regions current minimum phase high - speed regions at poles shrunk towards poles speed of high - latitude ) wind less for current minimum than previous minimum correlate with polar field strength weaker for current minimum phase magnetic pressure polar coronal holes acceleration of high - speed wind weak field may due to polar field not fully developed after field reversal around 2000 drifting of density structures from high to low latitudes caused by slow movement of concentrated magnetic - field regions of coronalikely due to migration of coronal holes from polar to low latitude high - speed wind with causing compression stream latitudinal spread density consistent with hcs tilt angle figure 1b maximum at polarity reversal coronal mass ejections dominate heliosphere density plot consistent with solar wind disappearance period mid 1999 low density turbulence co - rotating interaction regions first half 2003 intense density turbulence late october early november 2003 cme events prevailed interplanetary medium effect of cme in sun - earth distance limited to 24 days after weakens with solar distance cir events influence seen for several days increase in density turbulence with radial distance show latitudinal pattern for cir events 2003 latitudinal distribution radial evolution of enhancement turbulence observed ( manoharan 2008 ) cir - dominated period first half 2003 with moderate - to - severe storms at earth ( figure 1b plot latitudinal extents of low - turbulence regions at poles show changes between current previous minimum phases average level turbulence at current minimum lower than previous cyclefigure 2 displays constant @xmath15 plots at different phases solar cycle plot for 1997 limited by weak - scattering observations last plot includes data up to may 2009 turbulence observed closer to sun at poles than equator on phase diameter contour along poles remains nearly same along equator manoharan 1993 after 2003 overall diameter @xmath15 contour decreased sun s center same level turbulence close to sun radial dependence of turbulence ( @xmath16@xmath0@xmath17 varies as @xmath18 ) suggests scattering diameter corona shrunk towards sun scattering power @xmath16 ) remained nearly same at all latitudes between 1989 and 2003 decreased @xmath050% around 2009 at low - latitudes present large - scale 3-d features of solar wind speed density turbulence different from previous cycle current minimum phase low - speed region along equatorial belt wider than high - speed regions shrunk towards poles after 2003 overall scattering diameter corona gradually decreased sun s center results consistent with ecliptic and off - ecliptic studies ( lee 2009 mccomasweak fields at poles solar wind speed density turbulence low activity caused by changes movement large - scale fields reversal polarity linked to poleward equatorward meridional flows unbalanced magnetic flux sheeley 2008 flux - transport dynamo weak polar fields long solar cycle choudhuri 2007 thank radio astronomy centre ooty radio telescope for observations national space science data center wilcox solar observatory magnetic - field data supported by cawses - india program sponsored by isro | 0.360916 | 2,294 | 1,120 | , Saving $0.1 in GPT-4. |
154,591 | we present the results obtained by the cdf and d0 collaborations on the extraction of the ckm matrix element @xmath0 , the @xmath1 helicity measurements , and the constraints obtained on the @xmath2 vertex .
[ 1999/12/01 v1.4c il nuovo cimento ] | in the standard model ( sm ) , the top quark is predicted to decay almost exclusively to a @xmath1 boson and a @xmath3 quark , and the @xmath2 coupling is supposed to be purely v - a .
however , physics beyond the sm could be visible in the top decays by changing the branching fraction of the top quark , or by altering the @xmath2 coupling
. this would be the case for example with a fourth generation of quarks or supersymmetric particles .
a precise measurement of the top and @xmath1 properties can be used to probe such models of new physics .
we will first describe the extraction of the cabibbo - kobayashi - maskawa ( ckm ) matrix element @xmath0 .
then we will present the measurement of the @xmath1 helicity , and the constraints on anomalous couplings at the @xmath2 vertex .
the extraction of @xmath0can be done both from the @xmath4decays and the single top production . in the @xmath4decays , @xmath0can be extracted if the ckm matrix unitarity is assumed .
since the single top production is directly proportional to @xmath0 , it can be extracted directly from the cross - section measurement without such assumptions , but the precision reached is lower because of the small statistics available . using @xmath4events we can perform a measurement of @xmath5 which is defined as : @xmath6 where q = d , s , b .
the extraction of @xmath0can be done by assuming @xmath7 .
we then have : @xmath8 .
the measurement has been performed recently at d0 with a data sample corresponding to an integrated luminosity of 5.4 fb@xmath9 , both in the semileptonic ( @xmath10+jets ) and dileptonic ( @xmath11 ) channels .
the selected events contain high energetic jets ( at least 2 in the @xmath11 channel , at least 3 in the @xmath10+jets channel ) and lepton(s ) ( exactly one for @xmath10+jets and two for the @xmath11 channel ) . in the @xmath10+jets channel , an additional cut on the missing transverse energy
is applied to reject the background , which is mostly composed of @xmath1+jets and multijet events . in the dilepton channel ,
a topological selection is applied , based on the missing @xmath12 significance of the event , and the scalar sum of the transverse energies of the leptons and leading jet .
the measurement relies on the identification of b - jets , which is done with a neural network algorithm ( nn ) using the impact parameters of the tracks associated to jets and the properties of secondary vertices . in the @xmath11 channel ,
the nn is applied to the two leading jets , and the lowest output of the two is kept to build templates . in the @xmath10+jets channel ,
the number of jets passing a cut on the nn output is counted .
this requirement has an efficiency of 55 @xmath13 4@xmath14 for b - jets and of 1.5 @xmath13 0.1 @xmath14 for light jets .
three decay modes of the @xmath4pair are simulated using pythia : @xmath15 , @xmath16 and @xmath17 where @xmath18 .
templates are built for each of these decay channels and the background ( see figure [ fig : templates ] ) , and data is fitted to the templates using a binned likelihood .
a nuisance parameter approach is used to take into account the systematic uncertainties . :
number of identified b - jets in the @xmath10 + 3jets ( left ) and @xmath10 + 4 jets channels ( middle ) , and output of the b identification algorithm for the @xmath11 channel ( right ) . ] the @xmath4production cross - section and @xmath5 were simultaneously extracted with this method .
d0 obtains @xcite : @xmath19 ( stat+syst ) pb and @xmath20 ( stat+syst ) . using the modified frequentist feldman - cousins method , the interval in @xmath0at
the 95@xmath14 confidence level was derived : 0.90 @xmath21 0.99 .
cdf obtained a result using a similar method with 160 pb@xmath9 of integrated luminosity @xcite : @xmath22 ( stat ) @xmath23 ( syst ) , and @xmath24 0.78 at 95@xmath14 c.l .
this measurement can be performed by assuming the sm single - top cross - sections , a branching fraction of @xmath25 of 1 and the v - a coupling at the @xmath2 vertex .
the tevatron combination using results of d0 and cdf with 2.3 to 3.2 fb@xmath9 of luminosity yields @xcite : + @xmath26 and @xmath27 at the 95@xmath14 c.l .
+ at d0 , a measurement updated with 5.4 fb@xmath9 yields @xcite : + @xmath28 ( stat+syst ) and @xmath29 at the 95@xmath14 c.l .
the @xmath1 boson has three distincts polarizations : it can be longitudinal , left- or right - handed . because of
the v - a coupling at the @xmath2 vertex and the small mass of the @xmath3 quark compared to the top quark mass , b - quarks from top decays are almost only left - handed . since the angular momentum is conserved at the vertex , almost no right - handed @xmath1 bosons are produced from a top decay . in the sm , the different helicity fractions
are predicted to be : @xmath30 ( longitudinal ) , @xmath31 ( right - handed ) and @xmath32 ( left - handed ) @xcite . to measure these fractions ,
two main methods are used : a template method and a matrix element method .
the template method has been used by cdf in the @xmath11 channel with 5.1 fb@xmath9 and by d0 in the @xmath10+jets and @xmath11 channels with 5.4 fb@xmath9 .
both measurements use the @xmath33 variable which is very sensitive to the helicity , and where @xmath34 is defined as the angle between the direction of the incoming top and the charged lepton or down - type quark in the @xmath1 rest - frame ( see figure [ fig : wtemp ] ) .
templates are built for the three different polarizations and the background , taking into account reconstruction effects .
we present here the results obtained with two - dimensional fits of ( @xmath35,@xmath36 ) , with the constraint @xmath37 . with 4.8 fb@xmath9 in the dilepton channel
, cdf obtains @xcite : + @xmath38 ( stat ) @xmath39 ( syst ) and @xmath40 ( stat ) @xmath41 ( syst ) .
+ d0 results in the @xmath10+jets and @xmath11 channels with 5.4 fb@xmath9 @xcite are : + @xmath42 ( stat ) @xmath43 ( syst ) and @xmath44 ( stat ) @xmath45 ( syst ) .
cdf also used the matrix element method in the @xmath10+jets channel with 2.7 fb@xmath9 .
the probability for each event to correspond to a certain set of parameters ( @xmath35,@xmath36 ) is computed , by integrating over the pdf , the jet transfer functions , and the matrix element which is proportional to the helicity fractions .
the fractions obtained are @xcite : + @xmath46 ( stat ) @xmath47 ( syst ) and @xmath48 ( stat ) @xmath49 ( syst ) .
the cdf and d0 measurements have been combined using the blue method ( best linear unbiased estimator ) , and the result is @xcite : + @xmath50 ( stat ) @xmath51 ( syst ) and @xmath52 ( stat ) @xmath53 ( syst ) . + figure [ fig : whel_tev ] shows the individual and combined results with the sm prediction .
the most general lagrangian including operators up to dimension five can be written as : @xmath54 where @xmath55 and @xmath56 are the @xmath1 boson mass and four - momentum , @xmath57 and @xmath58 are left- and right - handed vector components ( @xmath59 and @xmath60 ) , and @xmath61 and @xmath62 are left- and right - handed tensor components ( @xmath63 and @xmath64 ) . the sm predicts : @xmath65 , @xmath66 , @xmath67 and @xmath68 .
a non - vanishing anomalous coupling would change the rate and kinematics of the single - top production and the @xmath1 helicity fractions . in the measurements ,
two couplings are investigated simultaneously : @xmath57 and one of the anomalous ones , the others being set to zero . for the single - top , multivariate discriminants ( bayesian neural networks ) have been used to distinguish between signals obtained from the different couplings .
inputs to these discriminants are kinematic variables such as the lepton and jets four - vectors , the missing @xmath12 and b - tagging informations .
d0 obtains limits with 5.4 fb@xmath9 shown in table [ tab : anom ] @xcite .
the combination of single - top results with @xmath1 helicity measurements can bring an even better discrimination between left- and right - handed couplings .
it has been done at d0 with 2.7 fb@xmath9 of integrated luminosity @xcite ( see table [ tab : anom ] ) .
.limits on anomalous couplings from the single - top channel ( first row ) and the combination of single - top and @xmath1 helicity ( second row ) . [ cols="<,<,<,<,<",options="header " , ]
top decays can be used for many searches of physics beyond the standard model .
the cdf and d0 collaborations have been able to put several limits on new physics models .
some of these results are still statistically limited , and would benefit from the analysis of the full tevatron dataset or from a combination with the lhc .
99 the d0 collaboration , phys .
107 , 121802 ( 2011 ) the cdf collaboration , phys
95 , 102002 ( 2005 ) the cdf and d0 collaborations , fermilab - tm-2440-e ( 2009 ) the d0 collaboration , phys . rev .
d 84 , 112001 ( 2011 ) m. fischer _
et al . _ ,
phys . rev .
d 63 , 031501(r ) ( 2001 ) the cdf collaboration , phys .
105 , 042002 ( 2010 ) the d0 collaboration , phys . rev .
d 83 , 032009 ( 2011 ) the cdf collaboration , conf . note 10333 ( 2010 ) the cdf and d0 collaborations , conference notes 6231 ( d0 ) and 10622 ( cdf ) ( 2011 ) the d0 collaboration ,
arxiv:1110.4592 submitted to phys .
b ( 2011 ) the d0 collaboration , d0 conference note 5838 ( 2009 ) | standard model top quark decay to @xmath1 boson @xmath3 quark @xmath2 coupling v - a physics beyond sm visible top decays changing branching fraction top quark altering @xmath2 coupling fourth generation quarks supersymmetric particles precise measurement top @xmath1 properties new physics extraction cabibbo - kobayashi - maskawa matrix element @xmath0 measurement @xmath1 helicity constraints on anomalous couplings at @xmath2 vertex extraction @xmath0can from @xmath4decays single top production @xmath4decays extracted if ckm matrix unitarity assumed single top production proportional to @xmath0 extracted from cross - section measurement without assumptions precision lower small statistics @xmath4events measurement @xmath5 defined as @xmath6 where q = d , s , b extraction @xmath0can assuming @xmath7 @xmath8 measurement performed at d0 data sample integrated luminosity 5.4 fb@xmath9 in semileptonic dileptonic channelsselected events contain high energetic jets 2 @xmath11 3+jets channel lepton(s one @xmath10+jets two @xmath10+jets channel additional cut on missing transverse energy reject background @xmath1+jets multijet events dilepton channel topological selection applied based on missing @xmath12 significance scalar sum transverse energies leptons leading jet measurement relies on identification b - jets with neural network algorithm nn using impact parameters properties secondary vertices @xmath11 channel nn applied to two leading jets lowest output kept to build templates @xmath10+jets channel number jets passing cut on nn output counted requirement efficiency 55 4@xmath14 for b - jets 1.5 0.1 for light jets three decay modes of @xmath4pair simulated using pythia @xmath15 templates built for decay channels background figure data fitted to templates using binned likelihood nuisance parameter approach used systematic uncertainties number of identified b - jets in @xmath10 + 3jets + 4 jets channels output b identification algorithm @xmath11 channelcross - section @xmath5 extracted d0 obtains @xcite @xmath19 @xmath20 feldman - cousins method interval @xmath0at 95@xmath14 confidence level 0.90 @xmath21 0.99 cdf 160 pb@xmath9 luminosity @xcite @xmath22 0.78 95@xmath14 c.l single - top cross - sections branching fraction @xmath25 1 v - a coupling @xmath2 vertex tevatron combination 2.3 to 3.2 fb@xmath9 luminosity yields @xcite @xmath26 @xmath27 95@xmath14 c.l measurement 5.4 fb@xmath9 yields @xcite @xmath28 stat+syst @xmath29 95@xmath14 c.l @xmath1 boson polarizations longitudinal left- right - handed v - a coupling @xmath2 vertex small mass @xmath3 quark b - quarks top decays left - handed angular momentum conserved vertex no right - handed @xmath1 bosons top decay helicity fractions @xmath30 longitudinal @xmath31 right - handed @xmath32 left - handedmeasure fractions two methods template matrix element method template method used by cdf @xmath11 channel 5.1 fb@xmath9 d0 @xmath10+jets 5.4 fb@xmath9 measurements use @xmath33 variable sensitive to helicity @xmath34 angle between incoming top charged lepton quark @xmath1 rest - frame figure templates built for three polarizations background reconstruction effects results with two - dimensional fits @xmath35,@xmath36 ) constraint @xmath37 4.8 fb@xmath9 dilepton channel cdf obtains @xcite : + @xmath38 @xmath39 @xmath40 d0 results @xmath10+jets channels 5.4 fb@xmath9 @xcite + @xmath42 @xmath43 @xmath44 @xmath45 cdf used matrix element method @xmath10+jets channel 2.7 fb@xmath9 probability for each event parameters @xmath35,@xmath36 ) computed integrating pdf jet transfer functions matrix element proportional to helicity fractions fractions obtained are @xcite : + @xmath46 @xmath47 @xmath48cdf d0 measurements combined blue method result is @xcite @xmath50 syst figure shows individual combined results with sm prediction general lagrangian operators up to dimension five written as @xmath54 @xmath55 are @xmath1 boson mass four - momentum @xmath57 left- right - handed vector components @xmath61 left- right - handed tensor components sm predicts @xmath65 @xmath66 @xmath67 @xmath68 non - vanishing anomalous coupling rate kinematics single - top production @xmath1 helicity fractions two couplings investigated simultaneously @xmath57 one anomalous others set to zero single - top multivariate discriminants used distinguish signals couplings inputs are kinematic variables lepton jets four - vectors missing @xmath12 b - tagging informations d0 obtains limits with 5.4 fb@xmath9 combination single - top results with @xmath1 helicity measurements better discrimination between left- right - handed couplings done at d0 with 2.7 fb@xmath9 integrated luminosity @xcitelimits anomalous couplings single - top channel first row combination @xmath1 helicity second row top decays searches physics beyond standard model cdf d0 collaborations limits new physics models results statistically limited benefit full tevatron dataset combination lhc . 99 d0 collaboration phys . 107 121802 2011 cdf collaboration 95 102002 2005 fermilab - tm-2440-e 2009 d0 collaboration phys . rev . 84 112001 2011 fischer phys . rev . 63 031501 2001 cdf collaboration phys . 105 042002 2010 d0 rev 83 032009 2011 cdf collaboration conf note 10333 2010 cdf d0 conference notes 6231 d0 10622 cdf 2011 d0 collaboration arxiv:1110.4592 phys . 2011 d0 collaboration conference note 5838 2009 | 0.356142 | 2,557 | 1,369 | , Saving $0.1 in GPT-4. |
154,603 | six glitches have been recently observed in the rotational frequency of the young pulsar psr b1737@xmath030 ( j1740@xmath03015 ) using the 25-m nanshan telescope of urumqi observatory . with a total of 20 glitches in 20years ,
it is one of the most frequently glitching pulsars of the @xmath11750 known pulsars .
glitch amplitudes are very variable with fractional increases in rotation rate ranging from @xmath2 to @xmath3 .
inter - glitch intervals are also very variable , but no relationship is observed between interval and the size of the preceding glitch . there is a persistent increase in @xmath4 , opposite in sign to that expected from slowdown with a positive braking index , which may result from changes in the effective magnetic dipole moment of the star during the glitch .
stars : neutron pulsars : individual : psr b1737@xmath030 | glitches in pulsars are rare and extraordinary events .
the study of such glitches and their post - glitch recoveries can give an insight into the interior of neutron stars , the physics of ultradense matter and provide limits on the equation of state @xcite .
pulsar glitches are usually detected from long - term and frequent timing observations .
the number of observed glitches has increased dramatically in the past few years , which probably is due to both the increased continuous observations of active pulsars and a large number of new young pulsars detected by the pulsar surveys ( e.g. , the parkes surveys , see * ? ? ? * ) over the last decade .
glitches are characterized as sudden increases in the rotation frequency @xmath5 ( where @xmath6 is the pulsar period ) , often followed by an interval of approximately exponential recovery or relaxation back towards the pre - glitch frequency .
the post - glitch relaxation can have timescale of days to years .
frequency jumps with magnitudes @xmath7 are recognized as `` giant '' glitches , and have been observed mostly in pulsars with characteristic ages @xmath8 yr , such as psrs b0833@xmath045 ( the vela pulsar ) , b1046@xmath058 and b1338@xmath062 @xcite .
giant glitches have not been observed in the youngest radio pulsars , such as psr b0531 + 21 ( the crab pulsar ) and psr b1509@xmath058 .
detected glitches are generally of magnitude @xmath9 , and the relative increment in slow - down rate @xmath10 is in the range @xmath11 to @xmath12 .
very small glitches with @xmath13 are difficult to distinguish from timing noise in young pulsars @xcite . of the 1750 known pulsars , about 170 glitches in 54 pulsars have been observed so far .
half of the published glitches have fractional amplitudes @xmath14 .
most glitches detected in the vela pulsar are giant glitches , and the largest glitch , with @xmath15 , was observed in psr j1806@xmath02125 @xcite .
about 60% of the 54 glitching pulsars have glitched only once , and 16% have glitched twice .
observations with rossi x - ray timing explorer ( rxte ) have revealed that psr j0537@xmath06910 , the 16-ms pulsar associated with the supernova remnant n157b in the large magellanic cloud , is the most frequently glitching known pulsar with 23 glitches detected in seven years of monitoring @xcite .
most of the glitches have @xmath16 of a few@xmath17 and , interestingly , a strong correlation is observed between the amplitude of a glitch and the time to the next one . furthermore , despite the usual post - glitch recovery ( decrease ) in @xmath4 , a persistent long - term increase in this parameter is observed , corresponding to a negative braking index @xmath18 .
this increase is most probably related to changes on or in the neutron star that result from the glitch activity . in this paper
we report on six recent glitches detected in the period of psr b1737@xmath030 and discuss the implications of these results .
psr b1737@xmath030 is a young radio pulsar with a characteristic age of @xmath19 yr which glitches frequently , with 20 glitches of all sizes ranging from `` small '' to `` giant '' ( @xmath20 to @xmath3 ) observed since its discovery in 1986 ( * ? ? ?
* ; * ? ? ?
* this paper ) . as for psr j0537@xmath06910
, there appears to be a cumulative shift in the spin - down rate of psr b1737@xmath030 resulting from its frequent glitches , which most likely accounts for its long - term braking index of @xmath21 @xcite .
the timing observations of psr b1737@xmath030 using the 25-m nanshan radio telescope of urumqi observatory started in 2000 january using a room - temperature dual - channel receiver . from 2002 july a dual - channel cryogenic receiver system sensitive to two orthogonal polarizations was used at a central observing frequency of 1540 mhz .
two polarizations , each of bandwidth 320 mhz , are fed to a filterbank consisting of @xmath22 channels of width 2.5 mhz .
the data are digitized to one - bit precision with a sampling interval of 1 ms .
time is provided by a hydrogen maser clock calibrated using the global positioning system .
observations times for psr b1737 - 30 were generally 16 min with the room - temperature receiver before 2002 july and 4 min with the cryogenic receiver after that date .
the data are dedispersed with off - line programs to remove the effects of interstellar dispersion , and then folded at the pulsar s nominal topocentric period with four sub - integrations per observation .
the pulse profiles obtained by summing an observation were cross - correlated with a high signal - to - noise template of the pulsar profile to produce accurate pulse topocentric times of arrival ( toas ) , which are then processed with the standard timing program tempo to convert them to barycentric arrival times at infinite frequency .
the toas refer to the peak of the main pulse .
the jet propulsion laboratory ephemeris de405 is used to correct the toas to the solar system barycenter .
the folded profiles are stored on disk for subsequent processing and analysis .
timing observations are usually made approximately three times per month and the glitching pulsars are observed more frequently .
we use the corrected barycentric toas to determine the basic parameters of the pulsar .
the basic timing model gives the predicted rotational pulse phase @xmath23 , as a function of time , t : @xmath24 timing irregularities appear as phase residuals ( @xmath25 ) , which are usually divided by @xmath26 to place them in time units .
for a glitching pulsar the residuals suddenly develop a negative slope at the time of the glitch as illustrated in fig .
[ fg : gl_phs ] . the frequency perturbation due to the glitch
can usually be described by : @xmath27 where @xmath28 is the change in pulse frequency relative to the pre - glitch model , @xmath29 and @xmath30 are the permanent changes in frequency and frequency derivative respectively and @xmath31 is the amplitude of the exponential recovery with a decay time constant of @xmath32 . the total frequency jump at the time of the glitch @xmath33 and the degree of recovery is often described by the parameter @xmath34 . because of the decaying component , the instantaneous change in @xmath35 at the glitch differs from @xmath30 : @xmath36 the glitch model of equation [ eq : gl_model ] describes the post - glitch behaviour fairly well for most glitches @xcite . [ cols="^,<,<,<,^,^,^,^ " , ] the glitch activity , @xmath37 , defined as the mean fractional change in period per year owing to glitches @xcite , is given by the simple expression : @xmath38 where @xmath39 is the total fractional increase of frequency owing to all of the glitches over an interval of @xmath40 .
psr b1737@xmath030 has a relatively high glitch activity with @xmath41 yr@xmath42 .
an advantage of @xmath37 as a long - term indicator of glitch effects is that it is relatively insensitive to the additional discovery of smaller glitches as the data quality improves @xcite .
pulsars can be grouped into three classes : pulsars with low glitch activity ( e.g. , psr b0525 + 21 ) , high glitch activity ( e.g. , vela pulsar ) and no glitch activity .
psr b1737@xmath030 belongs among those with high glitch activity .
the pulsars with characteristic ages between @xmath43 and @xmath44 yr are more likely to have higher glitch activity than the younger ones ( @xmath45yr ) or the older ones ( @xmath46yr ) as shown by statistical studies of pulsar glitches @xcite .
[ fg : spin - rate ] shows the evolution of the spin frequency @xmath26 and the slow - down rate @xmath35 over 20 years .
clearly , the long - term evolution of @xmath26 is dominated by the regular spin - down the glitches are not even visible on this plot . the lower part of fig .
[ fg : spin - rate ] shows that there has been a long - term decrease in @xmath35 ( increase in @xmath4 ) of about @xmath1 0.1% over the 20 years .
this corresponds to a value of @xmath47 .
the braking index is therefore @xmath48 for the long - term evolution of psr b1737@xmath030 .
inter - glitch data spans are frequently affected by relaxation from large glitches and generally have large positive braking indices @xcite .
for example , we obtain a braking index of @xmath49 by fitting the data starting 500 days after the last glitch ( mjd 53550 54095 ) . as mentioned in the introduction , a negative long - term braking index
is also observed for psr j0537@xmath06910 @xcite .
permanent increases in @xmath4 associated with glitches are also observed in the crab pulsar @xcite . in this case
though , since the glitches are much smaller , they do not bias the observed braking index so much .
the vela pulsar also has an anomalously low value of @xmath50 @xcite which may be likewise related to the frequent glitches .
a possible interpretation of these observations is that the component of the magnetic dipole moment which is perpendicular to the spin axis increases at the time of a glitch @xcite .
[ fg : spin - rate](b ) also shows what appears to be a cyclic variation in @xmath51 superimposed over the linear trend .
such a cyclic variation would have @xmath1 3000d period and needs at least another cycle to confirm .
[ fg : ppdot ] is a plot of pulsar period p versus period derivative @xmath52 of all known 1700 pulsars which distinguishes the glitching pulsars and other different classes of pulsars .
the most frequently glitching pulsars are young and have relatively strong dipole fields .
some glitching pulsars are older , including psr b1821@xmath024 , which is a recycled pulsar in the globular cluster m28 @xcite . at present ,
20 pulsars have detected giant glitches and , except psr b2224 + 65 whose age is @xmath53 yr , all the pulsars which have giant glitches are in the younger group .
our observations of psr b1737@xmath030 show six glitches in psr b1737@xmath030 over a period of seven years . in total ,
20 glitches have been observed over a 20-year period .
glitch sizes cover a wide range from just a few parts in @xmath54 to large glitches comparable to those seen in the vela pulsar .
the inter - glitch intervals are highly variable , ranging from 3 weeks to more than 3 years .
unlike in psr j0537@xmath06910 , there is no clear relationship of glitch interval with size of the preceding glitch .
although the age of psr b1737@xmath030 is comparable to that of the vela pulsar , their glitch behaviours are different , with glitch sizes and intervals being much more variable in psr b1737@xmath030 .
exponential and quasi - linear relaxations in @xmath35 are observed in both pulsars , but with quite different parameters .
it is remarkable that the three most highly glitching pulsars have such different glitch and post - glitch properties , yet another example of the diversity and complexity of magnetised neutron stars .
we acknowledge the support from the nnsfc under the project 10673021 ; and the key directional project program of cas .
wz is very grateful to j. yang for his valuable suggestions that led to an improvement on an earlier draft of this paper .
jou is grateful for the cas
twas visiting scholar fellowship that enabled him visit the urumqi observatory of naoc .
m. , bell j. f. , manchester r. n. , lyne a. g. , camilo f. , stairs i. h. , damico n. , kaspi v. m. , hobbs g. , morris d. j. , crawford f. , possenti a. , joshi b. c. , mclaughlin m. a. , lorimer d. r. , faulkner a. j. , 2003 , mnras , 342 , 1299 | glitches in pulsars rare study recoveries into neutron stars physics ultradense matter limits equation of state @xcite pulsar glitches detected from long - term frequent timing observations observed glitches increased due to increased observations active pulsars new young pulsars glitches sudden increases in rotation frequency @xmath5 @xmath6 pulsar period followed by exponential recovery pre - glitch frequency post - glitch relaxation days to years frequency jumps with magnitudes @xmath7 giant '' glitches observed in pulsars with characteristic ages @xmath8 yr b0833 b1046@xmath058 b1338@xmath062 giant glitches not observed in youngest pulsars b0531 21 b1509@xmath058 detected glitches magnitude @xmath9 increment slow - down rate @xmath10 range @xmath11 to @xmath12 small glitches with @xmath13 difficult distinguish from timing noise in young pulsars of 1750 known pulsars 170 glitches in 54 pulsars observed half published glitches have fractional amplitudes @xmath14glitches in vela pulsar are giant largest glitch @xmath15 in psr j1806@xmath02125 @xcite 60% of 54 glitching pulsars glitched once 16% twice observations with rossi x - ray timing explorer psr j0537@xmath06910 16-ms pulsar supernova remnant n157b large magellanic cloud most frequently glitching 23 glitches in seven years monitoring @xcite most glitches have @xmath16 few@xmath17 strong correlation between amplitude glitch time to next post - glitch recovery in @xmath4 persistent long - term increase in parameter negative braking index @xmath18 increase related to changes neutron star glitch activity on six recent glitches in psr b1737@xmath030 implications b1737@xmath030 young radio pulsar age @xmath19 yr glitches frequently 20 glitches small to giant to since discovery 1986 psr j0537@xmath06910 cumulative shift in spin - down rate of frequent glitches accounts for long - term braking index of @xmath21 @xcitetiming observations psr b1737@xmath030 25-m nanshan radio telescope 2000 january room - temperature dual - channel receiver 2002 july dual - channel cryogenic receiver polarizations used 1540 mhz two polarizations 320 mhz fed filterbank @xmath22 channels 2.5 mhz data digitized one - bit precision sampling interval 1 ms time hydrogen maser clock global positioning system observations times b1737 16 min before 2002 july 4 min after data dedispersed folded pulsar s topocentric period four sub - integrations per observation pulse profiles cross - correlated high signal - to - noise template pulsar profile accurate pulse topocentric times arrival processed barycentric arrival times infinite frequency peak main pulse jet propulsion laboratory ephemeris de405 correct toas solar system barycenter folded profiles stored disk processing analysis timing observations three times per month glitching pulsars observed more frequently corrected barycentric toas determine basic parameters pulsartiming model gives predicted rotational pulse phase @xmath23 function of time @xmath24 timing irregularities appear as phase residuals divided by @xmath26 time glitching pulsar residuals develop negative slope glitch in fig . frequency perturbation due to glitch described by @xmath27 change in pulse frequency pre - glitch @xmath29 permanent changes in frequency derivative @xmath31 amplitude exponential recovery with decay time constant @xmath32 total frequency jump at glitch @xmath33 degree of recovery described by @xmath34 decaying component instantaneous change in @xmath35 at glitch differs from @xmath30 @xmath36 glitch model describes post - glitch behaviour for most glitches glitch activity @xmath37 mean fractional change in period per year to glitches by @xmath38 @xmath39 total fractional increase of frequency glitches over interval @xmath40@xmath030 high glitch activity with @xmath41@xmath42 @xmath37 insensitive to discovery smaller glitches as data quality improvespulsars grouped into three classes low glitch activity high glitch no glitch activity psr b1737@xmath030 high glitch activity pulsars ages between @xmath43 and yr likely higher glitch activity younger older studies spin - rate shows evolution spin frequency @xmath26 slow - down rate @xmath35 over 20 years long - term evolution @xmath26 dominated by regular spin - down glitches not visible long - term decrease in @xmath35 increase in @xmath4 ) of 0.1% over 20 years corresponds to value @xmath47 braking index is @xmath48 for long - term evolution b1737@xmath030 inter - glitch data spans affected by relaxation from large glitches large positive braking indices braking index of @xmath49 by fitting data 500 days after last glitch negative long - term braking index observed for psr j0537@xmath06910 @xcite increases in @xmath4 glitches observed in crab pulsar @xcite glitches smaller not bias braking indexvela pulsar low value @xmath50 @xcite related to frequent glitches interpretation magnetic dipole moment perpendicular to spin axis increases glitch spin - rate](b shows cyclic variation in @xmath51 over linear trend variation @xmath1 3000d period needs another cycle to confirm ppdot ] plot of pulsar period p versus derivative @xmath52 of 1700 pulsars distinguishes glitching pulsars most glitching pulsars young strong dipole fields some older including psr b1821@xmath024 recycled pulsar in globular cluster m28 @xcite 20 pulsars detected giant glitches psr b2224 + 65 all pulsars giant glitches in younger group observations psr b1737@xmath030 show six glitches over seven years total 20 glitches observed over 20-year period glitch sizes to large glitches inter - glitch intervals variable 3 weeks to 3 years j0537@xmath06910 no clear relationship glitch interval with size preceding glitchage of psr b1737@xmath030 comparable vela pulsar glitch behaviours different glitch sizes intervals more variable in exponential quasi - linear relaxations in @xmath35 observed in both pulsars different parameters three highly glitching pulsars different glitch post - glitch properties example diversity complexity of magnetised neutron stars acknowledge support nnsfc project 10673021 key directional project program cas grateful to j. yang for suggestions improvement earlier draft paper jou grateful for cas twas visiting scholar fellowship urumqi observatory bell manchester lyne camilo stairs damico kaspi hobbs morris crawford possenti joshi mclaughlin lorimer faulkner 2003 mnras , 342 , 1299 | 0.438072 | 2,807 | 1,415 | , Saving $0.1 in GPT-4. |
154,605 | we discuss the impact of a hidden sector consisting of minicharged particles ( mcps ) and massless hidden photons on the expansion history of our universe .
we present parameter scans for the amount of extra relativistic particles ( @xmath0 ) and the abundance of light nuclei for fermionic mcps with masses between @xmath1100 kev and 10 gev and minicharges in the range @xmath2 .
current cmb and bbn data significantly constrain the available parameter space of mcps .
the shown results are a valuable indicator for future experimental searches and are presented in a flexible way so that more accurate results on @xmath3 can be easily interpreted . | minicharged particles ( mcps ) naturally arise in extensions of the standard model ( sm ) with an additional local hidden gauge group @xmath4 .
the hidden photon ( hp ) associated with this @xmath4 does not couple to any sm particle but mixes kinetically with the sm photon @xmath5 where @xmath6 is the kinetic mixing parameter , @xmath7 is the photon field strength tensor and @xmath8 is the field strength tensor of the hp .
we assume that the @xmath4 is unbroken .
hence , the hp is massless .
it would be unobservable because the mixing can be removed through a field redefinition .
the kinetic mixing becomes unphysical .
+ the hp field has observable consequences in a minimal extension of this model with a ( fermionic ) field @xmath9 , the hidden fermion .
this fermion is only charged under the @xmath4 with a coupling strength @xmath10 .
this makes @xmath6 a physical parameter because transforming the kinetic mixing away induces an effective coupling of the hidden fermion to the sm photon with a coupling strength @xmath11 .
since the hidden fermion couples to the sm photon , it carries an effective electric charge .
it is useful to define @xmath12 where @xmath13 is the electron charge and @xmath14 is the minicharge .
since @xmath14 can be small , @xmath9 is called a minicharged particle .
it is fully characterized by @xmath10 , @xmath14 and its arbitrary mass @xmath15 .
at the cmb epoch ( left ) and primordial helium yield @xmath16 ( right ) as a function of @xmath15 and @xmath14 , for @xmath17 .
numbers correspond to the value of the closest contour line .
_ left : _ dark green : @xmath18 , light green and yellow @xmath19 , orange and red @xmath20 .
the red dashed line shows the 95% upper exclusion limit @xmath21 ( planck+wp+highl+bao ) by planck @xcite .
the blue dashed line gives the best fit value @xmath22 from a combination of bicep2 and planck data @xcite .
_ right : _ dark green coloring denotes regions far away from the upper limit @xmath23 @xcite .
the limit is given by the red dashed contour line .
orange and red regions are excluded on more than a 95% cl.,title="fig:",scaledwidth=45.0% ] at the cmb epoch ( left ) and primordial helium yield @xmath16 ( right ) as a function of @xmath15 and @xmath14 , for @xmath17 .
numbers correspond to the value of the closest contour line .
_ left : _ dark green : @xmath18 , light green and yellow @xmath19 , orange and red @xmath20 .
the red dashed line shows the 95% upper exclusion limit @xmath21 ( planck+wp+highl+bao ) by planck @xcite .
the blue dashed line gives the best fit value @xmath22 from a combination of bicep2 and planck data @xcite .
_ right : _ dark green coloring denotes regions far away from the upper limit @xmath23 @xcite .
the limit is given by the red dashed contour line .
orange and red regions are excluded on more than a 95% cl.,title="fig:",scaledwidth=45.0% ] [ sec : figures1 ] due to the small charge , hps and mcps will be produced in the early universe to some extent .
both of them will then contribute to the energy density of radiation during the formation of the cmb anisotropies .
the impact of light degrees of freedom on the cmb is parametrized with the effective neutrino degrees of freedom @xmath3 .
they are defined as @xmath24 where @xmath25 is the energy density of dark radiation ( dr ) , i.e. all relativistic particles that are not strongly coupled to the sm photon at the cmb epoch , @xmath26 is the energy density of one neutrino species which instantly decoupled from the electron - photon plasma before bbn .
all densities are evaluated at times corresponding to the formation of the cmb anisotropies . @xmath3
scales like @xmath27 .
it is , hence , very sensitive to temperature changes .
+ in the sm the three neutrinos contribute @xmath28 @xcite after correcting for non - instantaneous decoupling .
the most precise observations to date come from the planck collaboration .
they set an upper limit of @xmath29 at 95 % cl on the number of relativistic degrees of freedom at the cmb epoch .
this value is obtained by fitting a cosmological model to the data .
the limit is conservative since it does not use the controversial value for the hubble rate today , @xmath30 , obtained from local astrophysical measurements .
local measurements prefer a higher value for @xmath3 . also the disputed result obtained by bicep2 @xcite suggests a higher value , @xmath31 @xcite .
+ in order to compare the predictions of models with minicharged particles to the observations , we have to accurately compute the energy density of dark radiation at the cmb epoch .
we numerically solve a set of boltzmann - like equations that parametrize the energy transport between sm particles and the dark sector ( ds ) consisting of hps and mcps @xmath32 where @xmath33 is the energy density ( pressure ) of the ds and the sm particles , @xmath34 is the hubble parameter , the dot is a derivative with respect to physical time and @xmath35 is a generalized collision term that gives the energy transported from one sector to the other .
+ for the computation we assume that the ds is always in thermal equilibrium with itself .
it is fully characterized by a common temperature @xmath36 and @xmath15 . following the spirit of a dark , weakly coupled sector we initially set @xmath37 .
the ds is then produced by the sm particles during the evolution of the universe . in order to compute this thermalization as accurately as possible
we include all sm particles and light mesons and all relevant processes in our computations .
it turns out that for @xmath38 ds particles are most efficiently produced by sm particle pair annihilation ( @xmath39 ) .
when @xmath36 approaches the sm temperature coulomb - like scattering ( @xmath40 ) , regularized by plasma screening , and ( for large @xmath10 ) compton - like scattering ( @xmath41 ) become the most important processes .
+ we scan over a wide range of parameters @xmath42 for @xmath43 .
the constraints for @xmath17 are the most conservative .
for @xmath44 the results are indistinguishable from the results with @xmath17 because in this case the dominant processes are @xmath45 .
a change in @xmath10 can be compensated by a corresponding change in @xmath6 but leaves @xmath14 unchanged .
the bounds for @xmath46 are stronger because compton - like scattering ( @xmath41 ) , which is @xmath47 , becomes dominant for high @xmath10 .
increasing @xmath10 can not be compensated by a shift in @xmath6 anymore if @xmath14 is fixed . for @xmath46
the ds and sm particles equilibrate faster .
+ our result for @xmath48 is presented in fig .
[ fig : cmb ] ( left ) .
the sudden increase for small kinetic mixing @xmath49 corresponds to partial thermalization of the ds with the sm photon .
the behavior for large @xmath14 can be understood analytically .
the ds fully thermalizes and after decoupling its temperature at the cmb epoch is given by entropy conservation . for more details
see @xcite . +
another primordial probe is bbn
. the ds changes the formation of nuclei .
the most sensitive probe is the yield of primordial helium @xmath16 . by adjusting the bbn code of @xcite
, we computed @xmath16 using the data of our cmb simulation . the result is shown in fig . [ fig : cmb ] ( right ) .
we use @xmath23 at 95% cl @xcite as a conservative upper bound .
figure [ fig : cmb ] shows that , using planck s limit , the cmb anisotropies allow for light minicharged particles in the range @xmath50 mev for a wide range of minicharges @xmath14 .
this parameter space is , however , disfavored by the abundance of primordial helium . using these two results we set a lower limit on the hidden fermion mass of @xmath51 mev for @xmath52 . a compilation of this result with earlier approaches using astrophysics , cosmology or laboratory experiments is shown in fig . [ fig : completeresult ] .
+ if bicep2 has observed primordial gravitational waves , the limit on mcps from the cmb anisotropies vanishes and bbn gives the most stringent bounds .
further research is needed to clarify the role of bicep s observations . since the question of the existence of additional light degrees of freedom can not be settled with the current data , we present our results in a flexible way .
future more accurate values of @xmath3 are to be expected . using the predicted accuracy of cmbpol @xcite , planck s current mean value @xmath53
@xcite could be detected at @xmath54 .
et al . _ , `` planck 2013 results .
cosmological parameters `` , astron . astrophys .
2014 [ arxiv:1303.5076 [ astro-ph.co ] ] .
e. giusarma _
et al . _ ,
`` relic neutrinos , thermal axions and cosmology in early 2014 '' , phys .
d90 043507 ( 2014 ) , [ arxiv:1403.4852 [ astro-ph.co ] ] h. vogel and j. redondo , `` dark radiation constraints on minicharged particles in models with a hidden photon '' , jcap 1402 029 ( 2014 ) , [ arxiv:1311.2600 [ hep - ph ] ] g. mangano and p. serpico , `` a robust upper limit on @xmath3 from bbn , circa 2011 '' , phys.lett .
b701 296 - 299 ( 2011 ) , [ arxiv:1103.1261 [ astro-ph.co ] ] g. mangano _ et al . _
`` relic neutrino decoupling including flavor oscillations '' , nucl.phys .
b729 221 - 234 ( 2005 ) , [ arxiv : hep - ph/0506164 [ hep - ph ] ] p.a.r .
_ , detection of b - mode polarization at degree angular scales by bicep2 , phys.rev.lett . 112 241101 ( 2014 ) , [ arxiv:1403.3985 [ astro-ph.co ] ] d. cadamuro and j. redondo , ' ' cosmological bounds on pseudo nambu - goldstone bosons '' , jcap ( 1202 ) , [ arxiv:1110.2895 ] s. galli _
et al . _ ,
`` constraining fundamental physics with future cmb experiments '' phys.rev .
d82 123504 ( 2010 ) , [ arxiv:1005.3808 [ astro-ph.co ] ] | minicharged particles arise in extensions standard model hidden gauge group @xmath4 hidden photon sm mixes with sm photon @xmath5 @xmath6 kinetic mixing parameter @xmath7 photon field strength strength tensor hp @xmath4 unbroken hp massless unobservable mixing removed field redefinition kinetic mixing unphysical hp field observable consequences in minimal extension fermionic field @xmath9 hidden fermion charged under @xmath4 coupling strength @xmath10 @xmath6 physical parameter transforming kinetic mixing induces coupling hidden fermion to sm photon coupling strength @xmath11 fermion couples to sm photon carries electric charge define @xmath12 @xmath13 electron charge @xmath14 minicharge @xmath14 small @xmath9 minicharged particle characterized by @xmath10 @xmath14 arbitrary mass @xmath15 cmb epoch primordial helium yield @xmath16 function of @xmath15 @xmath14 @xmath17 numbers correspond closest contour line dark green @xmath18 light green yellow @xmath19 orange red @xmath20red dashed line shows 95% exclusion limit @xmath21 ( planck+wp+highl+bao by planck blue line best fit value @xmath22 from bicep2 planck data dark green regions upper limit @xmath23 limit by red contour line orange red regions excluded on than 95% cmb epoch primordial helium yield @xmath16 of @xmath15 @xmath14 @xmath17 numbers closest contour line dark green @xmath18 light green yellow @xmath19 orange red @xmath20 red line 95% exclusion limit @xmath21 planck+highl by planck blue line best fit value @xmath22 from bicep2 planck data dark green regions upper limit @xmath23 limit red line orange red regions excluded on more than 95% cl small charge hps mcps produced in early universe contribute to energy density radiation during formation cmb anisotropies impact of light degrees freedom on cmb parametrized with neutrino degrees freedom @xmath3 defined as @xmath24 @xmath25 energy density of dark radiationrelativistic particles not coupled to sm photon at cmb epoch @xmath26 energy density of one neutrino decoupled from electron - photon plasma before bbn densities evaluated formation cmb anisotropies @xmath3 scales like @xmath27 sensitive to temperature changes three neutrinos contribute @xmath28 @xcite non - instantaneous decoupling precise observations from planck collaboration upper limit @xmath29 at 95 % cl relativistic degrees freedom cmb epoch value obtained cosmological model limit conservative controversial value hubble rate @xmath30 local measurements prefer higher value for @xmath3 disputed result bicep2 @xcite suggests higher value @xmath31 compare predictions models minicharged particles observations compute energy density of dark radiation at cmb epoch solve boltzmann - equations energy transport between sm particles dark sector ds @xmath32 @xmath33 energy density of ds sm particles @xmath34 hubble parameter dot derivative physical time @xmath35 generalized collision term energy transported ds in thermal equilibrium characterized by common temperature @xmath36 and @xmath15dark coupled sector set @xmath37 ds produced by sm particles evolution universe compute thermalization include sm particles light mesons relevant processes @xmath38 ds particles produced by sm particle pair annihilation @xmath39 @xmath36 sm temperature coulomb - scattering @xmath40 plasma screening @xmath10 compton - like scattering @xmath41 ) important processes scan parameters @xmath42 @xmath43 constraints @xmath17 conservative @xmath44 results indistinguishable from @xmath17 dominant processes @xmath45 change in @xmath10 compensated change @xmath6 leaves @xmath14 unchanged bounds for @xmath46 stronger compton - like scattering @xmath41 dominant high @xmath10 increasing @xmath10 compensated shift @xmath6 @xmath14 fixed @xmath46 ds sm particles equilibrate faster result @xmath48 in fig . increase small kinetic mixing @xmath49 partial thermalization ds with sm photon behavior large @xmath14 understood analytically ds thermalizes temperature cmb epoch entropy conservation details see @xcite primordial probe bbn ds changes formation nuclei sensitive probe yield primordial helium @xmath16 bbn code @xcite computed @xmath16 using cmb simulation result shown in fig . use @xmath23 at 95% cl @xcite conservative upper bound planck s limit cmb anisotropies allow light minicharged particles range @xmath50 mev wide range minicharges @xmath14 disfavored by abundance primordial helium set lower limit hidden fermion mass @xmath51 @xmath52 compilation result with earlier approaches shown in fig . if bicep2 observed primordial gravitational waves limit mcps cmb anisotropies vanishes bbn gives stringent bounds further research needed role bicep s observations additional light degrees freedom data present results flexible future accurate values @xmath3 expected predicted accuracy cmbpol @xcite planck s current mean value @xmath53 @xcite detected at @xmath54 al planck 2013 results cosmological parameters astron astrophys . 2014 giusarma relic neutrinos thermal axions cosmology early 2014 phys .043507 2014 arxiv:1403.4852 astro-ph.co h vogel j redondo dark radiation constraints minicharged particles hidden photon jcap 1402 029:1311.2600 g mangano serpico upper limit @xmath3 bbn 2011 phys.lett b701 296 - 299 arxiv:1103.1261-ph mangano relic neutrino decoupling flavor oscillations nucl.phys b729 221 - 234 2005 hep - ph/0506164 detection b - mode polarization angular scales bicep2 phys.rev.lett 112 241101 2014:1403.3985 astro-ph.co d. cadamuro j redondo cosmological bounds nambu - goldstone bosons jcap 1202:1110.2895 s galli constraining fundamental physics future cmb experiments phys.rev d82 123504 2010 arxiv:1005.3808 astro-ph | 0.357376 | 2,528 | 1,329 | , Saving $0.1 in GPT-4. |
154,619 | the recombination model as a model for hadronization from a quark - gluon plasma has been recently revived since it has advantages in explaining several important features of the final state produced in heavy - ion collisions at rhic , such as the constituent quark number scaling of the elliptic coefficient versus the transverse energy of identified hadrons , the bending shape of the @xmath0 spectrum of hadrons near 5 gev / c , and the measured large value of baryon to meson ratio(of the order of unity ) in the same @xmath0 range .
we have developed a dynamic simulation model of heavy - ion collisions in which a quark - gluon plasma , starting from a certain initial condition , evolves hydrodynamically until it reaches the phase boundary , and then hadronizes by valence quark recombination .
rescattering after hadronization is described by urqmd .
we discuss some details of the model and report first , preliminary results . | the quark recombination model of hadron formation , which was originally proposed in 1970s@xcite and studied actively in 1980s@xcite , has been revived@xcite as it naturally explains some of the features of the produced hadrons in relativistic heavy - ion collisions .
the hydrodynamic analysis of the measured elliptic flow coefficient of hadrons , which signals the collective behavior of the produced system , could explain the magnitude and the ordering of the magnitude of the coefficient @xmath1 of various hadrons and thus showed that the system produced during the heavy - ion collisions behaves as a strongly interacting , nearly perfect fluid . if one plot the elliptic coefficient per constituent quarks , @xmath2 versus the transverse kinetic energy per constituent quarks , @xmath3 , then the data for various hadrons converge into a single line , suggesting that the collectivity is the property not of the hadrons but also of the constituent ( valence ) quarks .
a natural explanation is that the hadrons are produced by recombining the quarks which are collectively moving in the quark - gluon plasma . a second set of data supporting the recombination of quarks to form hadrons are the transverse momentum spectra of hadrons , which shows bending behavior near @xmath4 gev .
hadrons with small momentum ( @xmath5 gev ) are thermal ones and large momentum hadrons are those from jets .
fragmentation of jets may produce hadrons with @xmath6 gev while the recombination of quarks with thermal distribution in the quark - gluon plasma can produce the hadrons with @xmath7 .
the two mechanisms combined have shown to fit the hadron spectra in the mid-@xmath0 range well .
another experimental observation supporting the recombination model is the large baryon - to - meson ratio , _ e.g. _
@xmath8 , near @xmath5 gev .
if the hadrons are formed by recombining quarks with a thermal distribution , the probabilities of forming mesons or baryons are determined only by the probabilities to find quarks , and the ratio @xmath8 can be naturally explained .
the recombination calculations reported so far were mostly done in the framework of schematic evolution models .
the only dynamical calculation is that by ko _ et .
al . _ : the initial partonic matter evolves through the parton cascade and at the phase boundary hadrons are formed by recombination . thus fixing the initial state just after the the nuclear impact , they can simulate the main experimental features described above . here
we report a dynamic simulation method describing the heavy - ion collisions where the evolution of the quark - gluon plasma is described by ideal hydrodynamics up to the phase boundary . at the phase boundary , hadronization is simulated by the recombination mechanism and the produced hadrons may or may not further rescatter until thermal freeze - out . in our implementation of this model ,
quark - gluon evolution is described by the hydrodynamic equations . for the hydrodynamic cells at the phase boundary
, hadronization occurs via recombination model and thus the monte carlo program creates hadrons with their definite position and momentum at each cell whose expansion velocity is explicitly known .
the hadrons may further undergo hadronic rescattering and this process is described by urqmd .
we use the three - dimensional ideal relativistic hydrodynamic program by c. nonaka @xcite which uses comoving ( lagrangian ) grid points rather than the commonly used fixed(in some coordinate system ) eulerian grid .
the advantage of the lagrangian grid for the calculation the hypersurface integral of the type appearing in the cooper - frye formula will be explained below . each quark - gluon plasma fluid cell at @xmath9
is specified by the temperature , @xmath9 , the baryon chemical potential , @xmath10 , position , @xmath11 , and the spatial length , @xmath12 , where @xmath13 is the proper time and @xmath14 is the space - time rapidity .
the cell is moving with its 4-velocity @xmath15 .
the quark distribution in the cell at hadronization is given as @xmath16 the invariant cross sections for recombined hadrons are given as @xcite : @xmath17 @xmath18 for mesons , and @xmath19 @xmath20 denotes the integration over the hadronization hypersurface .
inserting the quark distribution into the above equations , one gets @xmath21 for mesons , where @xmath22 and @xmath23 are the energies of the quark and antiquark , respectively .
similarly for baryons : @xmath24 since the wave functions for hadrons are best known in the light - front frame , it is convenient to use the light - cone variable , @xmath25 , which denotes the longitudinal momentum fraction carried by the quark in the infinite momentum frame . if one neglects the transverse momentum of quarks relative to the hadron momentum @xmath26 , the equations become simplified .
when the wave functions for hadrons are known , one can explicitly decompose the quark momentum into the longitudinal transverse components , @xmath27 , and evaluate the integrals exactly . in the above equations , integration over the hadronization hypersurface @xmath28
is done numerically .
the hypersurface areas are given by the cell surfaces between adjacent cells , one of which has a temperature @xmath29 and the other has @xmath30 . in the case of a first - order phase transition with phase coexistence , also pairs of cells with a different fraction of quark - gluon plasma contribute .
a single grid cell may have space - like as well as time - like surfaces , contributing to the hadronization hypersurface .
it is known that the hydrodynamic programs using an eulerian grid sometimes give negative values of the expression @xmath31 .
however , in the hydrodynamic program which uses the lagrangian coordinate such as developed by c. nonaka and is used in this calculation , @xmath31 is positive definite . during the hadronization
energy and momentum should be conserved , equivalent to energy conservation in the rest frame of a cell .
we enforce energy conservation by continuing the creating process of hadrons , using the monte carlo method , until all the energy of the cell is exhausted for hadronization .
however , since our grid is so fine that the total energy of each individual fluid cell is smaller than the energy required to create a hadron , we add the energies of all fluid cells within a small @xmath32 interval and randomly create hadrons from this ensemble of cells . the monte carlo implementation of this scheme works as follows .
consider a fluid cell at @xmath9 in the mixed phase whose degree of hadronization is characterized by the qgp volume fraction @xmath33 .
if @xmath34 is the total energy in the fluid cell @xmath35 and @xmath36 is the negative of the change of @xmath37 during the evolution in one time step , then @xmath38 is the energy used for the hadronization during the time interval . summing the energy for all the cells in a given @xmath32 interval , one finds the total energy to hadronize among the fluid cells in @xmath39 .
hadronization continues until all the energy has been used to create hadrons . in the sudden recombination model the probability to create a hadron with certain momentum
@xmath26 depends only on the number and flavor of its constituent quarks but is independent of the hadron mass . in order to implement these constraints
, we dived the mesons into four groups with different flavor contents : @xmath40 similarly , there are a total of eight groups for baryons , four for baryons and four for antibaryons : @xmath41 to obtain the probability to hadronize into a hadron with certain flavor content and momentum , one randomly choses a cell among those in the same @xmath32 interval , the flavor group , the momentum and a random number in the interval @xmath42 $ ] .
when the random number exceeds the probability to create the hadron given by the recombination model equations , a hadron with certain momentum is created in that fluid cell .
this procedure is repeated until all the energy available is exhausted . finally , in order to determine the hadron species inside a given group ( e.g. @xmath43 versus @xmath44 , or some other resonance included in the urqmd model ) , we assume a relative boltzmann weight , @xmath45 , where @xmath46 is the degeneracy and @xmath47 the hadron mass . as a result of hadronization
, we create hadrons moving with a thermal velocity spectrum of a fluid cell superimposed on the collective flow velocity of the cell given by the hydrodynamics .
rescattering among those hadrons is then simulated with the help of urqmd . by running urqmd code , we may check whether the chemical freeze - out coincides with the time of hadronization , and we can track the cooling of the system until the final thermal freeze - out .
in fig . [ fig : pt ] the transverse momentum spectra , @xmath48 vs. @xmath49 of pions and protons are shown .
those particles are hadronized from a qgp with an initial state same as in ref .
[ 9 ] for the u+u collisions at 200 gev / a , and experienced the final state interaction by running urqmd .
the results are obtained from 20 events and in order to make the systematic error smaller one has to repeat the simulation for many events . in fig .
[ fig : dndy ] the rapidity spectra of pions are shown .
as emphasized before , hadrons are generated from a qgp according to the sudden recombination model , and rescattering among hadrons is simulated by urqmd .
the creation of hadrons with their position and momentum enables us to``measure '' any physical observables of interest .
the only inputs are those of the initial state of the hydrodynamic evolution and the hadronization temperature @xmath9 ( or the equation of state ) .
the comparison of the result with the statistical model will be interesting . in the future we hope to calculate of the elliptic flow coefficients of various hadrons in order to study how the flow anisotropy is related to the initial momentum anisotropy of the qgp .
_ acknowledgements : _ this work was supported in parts by chonnam national university , 2007 , the korea science and engineering foundation(kosef ) grant funded by the korea government(most ) ( no .
r0120070002107402008 ) , and doe grant de - fg02 - 05er41367 .
k. s. lee wish to thank s. bass and b. mller for their hospitality during his stay at duke university .
das and r.c .
hwa , phys .
b*68 * , 459 ( 1977 ) .
biro , p. levai and j. zimanyi , phys .
b*347 * , 6(1995 ) , and references therein .
hwa and c.b .
yang , phys .
c*66 * , 025205 ( 2002 ) .
r. fries , b. mller , c. nonaka and s.a .
bass , phys .
lett . * 90 * , 202303 ( 2003 ) .
r. fries , b. mller , c. nonaka and s.a .
bass , phys .
c*68 * , 044902 ( 2003 ) . v. greco and c.m .
ko , phys .
* 90 * , 202302 ( 2003 ) .
z. w. lin and c.m .
ko , phys .
lett . * 89 * , 202302 ( 2002 ) .
s. a. voloshin , nucl .
a*715 * , 379c ( 2003 ) .
c. nonaka and s. a. bass , phys .
c*75 * , 014902 ( 2007 ) . | quark recombination model hadron formation proposed 1970s studied 1980s revived explains features produced hadrons in heavy - ion collisions hydrodynamic analysis of elliptic flow coefficient of hadrons explain magnitude ordering coefficient @xmath1 hadrons system strongly interacting nearly perfect fluid elliptic coefficient per constituent quarks @xmath2 versus transverse kinetic energy per constituent quarks @xmath3 data hadrons converge into single line collectivity property hadrons constituent quarks hadrons produced by recombining quarks moving in quark - gluon plasma second data supporting recombination transverse momentum spectra of hadrons bending behavior near @xmath4 gev small momentum gev are thermal large momentum from jets fragmentation of jets produce hadrons with @xmath6 gev recombination of quarks with thermal distribution hadrons with @xmath7 two mechanisms fit hadron spectra in mid-@xmath0 range observation recombination large baryon - to - meson ratio @xmath8 near @xmath5 gevhadrons formed recombining quarks thermal distribution probabilities forming mesons baryons determined by quarks ratio @xmath8 explained recombination calculations schematic evolution models dynamical calculation ko _ et initial partonic matter evolves through parton cascade phase boundary hadrons formed by recombination fixing initial state after nuclear impact experimental features dynamic simulation method heavy - ion collisions evolution quark - gluon plasma ideal hydrodynamics phase boundary hadronization simulated by recombination produced hadrons rescatter until thermal freeze - out quark - gluon evolution described hydrodynamic equations hydrodynamic cells phase boundary hadronization occurs via recombination monte carlo program creates hadrons definite position momentum cell expansion velocity known hadrons undergo hadronic rescattering described by urqmd three - dimensional ideal relativistic hydrodynamic program by c. nonaka uses comoving lagrangian grid points fixed eulerian grid advantage lagrangian grid hypersurface integral cooper - frye formula explainedquark - gluon plasma fluid cell at @xmath9 specified by temperature baryon chemical potential position spatial length @xmath13 proper time space - time rapidity cell moving with 4-velocity @xmath15 quark distribution at hadronization @xmath16 invariant cross sections for recombined hadrons as integration over hadronization hypersurface inserting quark distribution equations @xmath21 for mesons @xmath22 @xmath23 energies of quark antiquark for baryons @xmath24 wave functions for hadrons use light - cone variable @xmath25 longitudinal momentum fraction by quark in infinite momentum frame neglects transverse momentum of quarks relative to hadron momentum equations simplified wave functions hadrons known decompose quark momentum into longitudinal transverse components evaluate integrals integration over hadronization hypersurface @xmath28 done numerically hypersurface areas by cell surfaces between adjacent cells temperature @xmath30 first - order phase transition pairs cells with different fraction of quark - gluon plasma contributesingle grid cell space time - like surfaces hadronization hypersurface hydrodynamic programs eulerian grid give negative values @xmath31 hydrodynamic program lagrangian coordinate @xmath31 positive during hadronization energy momentum equivalent to energy conservation rest frame cell enforce energy conservation creating hadrons monte carlo method until energy exhausted for hadronization grid fine total energy fluid cell smaller than hadron add energies fluid cells within @xmath32 interval randomly create hadrons monte carlo implementation consider fluid cell at @xmath9 mixed phase hadronization characterized by qgp volume fraction @xmath33 if @xmath34 total energy @xmath35 @xmath36 negative of change @xmath37 evolution @xmath38 is energy used for hadronization summing energy cells in @xmath32 interval total energy to hadronize among fluid cells in @xmath39 hadronization continues until all energy used to hadrons in sudden recombination model probability to create hadron with momentum @xmath26 depends on number flavor constituent quarks independent of hadron massimplement constraints dived mesons into four groups different flavor contents @xmath40 eight groups for baryons four baryons four antibaryons @xmath41 probability to hadronize into hadron flavor content momentum randomly choses cell same @xmath32 interval flavor group momentum random number interval @xmath42 when random number exceeds probability create hadron hadron with momentum created cell procedure repeated until energy exhausted determine hadron species group assume relative boltzmann weight @xmath45 @xmath46 degeneracy @xmath47 hadron mass hadronization hadrons moving with thermal velocity spectrum fluid cell superimposed on collective flow velocity rescattering hadrons simulated with urqmd check chemical freeze - out with time hadronization track cooling system until final thermal freeze - out fig . transverse momentum spectra @xmath48 vs. @xmath49 of pions protons shown particles hadronized from qgp initial state same ref 9 for u+u collisions at 200 gev / a experienced final state interaction by running urqmd results obtained from 20 events systematic error smaller simulation for many events fig .fig : dndy rapidity spectra pions shown hadrons generated from qgp sudden recombination model rescattering among hadrons simulated by urqmd creation hadrons position momentum physical observables inputs initial state hydrodynamic evolution hadronization temperature comparison result with statistical model interesting future calculate elliptic flow coefficients hadrons study flow anisotropy initial momentum anisotropy qgp work supported chonnam national university korea science engineering foundation grant korea government r0120070002107402008 doe grant de - fg02 - 05er41367 lee s. bass b. mller hospitality duke university das hwa . b*68 *, 459 ( 1977 ) biro. levai j. zimanyi phys . b*347 * 6(1995 ) hwa yang phys . c*66 * 025205 ( 2002 ) r. fries b. mller c. nonaka bass phys . 90 * 202303 ( 2003 ) phys c*68 * 044902 ( 2003 ) v. greco c.m . ko phys .90 202302 2003 ) lin c ko phys lett 89 202302 2002 a. voloshin nucl a*715 379c 2003 ) nonaka s a. bass c*75 014902 2007 ) | 0.432825 | 2,615 | 1,343 | , Saving $0.1 in GPT-4. |
154,621 | we calculated spectral energy distribution of the helium strong chemically peculiar star hd 37776 using adequate model atmospheres . we show that the chemical peculiarity influences the stellar energy distribution .
consequently , spots of peculiar chemical composition on the surface of a rotating star may cause detectable light variations of the star .
however , the observed light variations are not likely caused by an uneven surface distribution of helium , but may be due to spots of metals ( mainly carbon ) . | magnetic chemically peculiar ( mcp ) stars exhibit periodic light variations , that used to be explained by presence of photometric " spots with uneven energy distribution on the surface of rotating stars .
these photometric spots are likely connected with the spectroscopic " spots of a peculiar abundance of various chemical elements .
the departures of energy distribution in spectra of mcp stars from normal energy distribution are explained as a consequence of line and continuum blanketing originating in the spectroscopic spots with peculiar chemical composition ( e.g. * ? ? ?
@xcite @xcite ) . to test whether the photometric spots are related to the spectroscopic ones we calculated light variations of the helium strong mcp star hd37776 , for which @xcite by means of doppler analysis derived a spot model with a helium - rich ring surrounded by two helium - poor spherical caps ( fig .
[ povrch ] ) .
for lte atmosphere modelling we opted code tlusty @xcite .
the models correspond to the helium rich / poor surface elements on hd37776 with @xmath0 and @xmath1 respectively @xcite .
the effective temperature @xmath2 and the surface gravity @xmath3 of all these models are assumed to be the same and equal to @xmath4k and @xmath5 , respectively @xcite .
the emergent radiative flux was calculated with synspec code .
the spectral energy distribution derived for individual surface elements was used to obtain corresponding fluxes in individual colours . finally , these fluxes are integrated over all visible surface and the observed magnitude difference is calculated .
the radiative fluxes calculated for atmospheres with different chemical composition are displayed on ( fig .
[ toky ] ) .
the flux calculated was convolved with a gauss function to better demonstrate possible light variations .
apparently , the he - poor surface elements with solar metallicity produce a flux that is lower in the visible region than the flux produced by the he - strong ones .
the regions with enhanced metallicity produces a flux highest in the visible region .
from the spectrum and light variability observations follows that at the maximum stellar brightness , the helium equivalent width have its minimum , whereas the si equivalent widths have their maximum .
clearly , the he - rich ring is dimmer and mostly si - poor , whereas the he - poor spots are brighter and mostly si - rich .
however , the radiative flux from the solar metallicity helium - poor surface elements in the visible region is lower than that from the helium - strong ones ( fig .
[ toky ] ) .
consequently , the observed light variability of hd37776 can not be explained merely by altering of emergent flux due to the uneven surface distribution of helium .
also surface variations of metallicity should be taken in to account .
therefore , we calculated the predicted light variability of hd37776 assuming that he - poor spots have metallicity thirty times higher relative to the solar one and that the he - strong ring consists from h and he only ( fig .
[ hvvel ] ) .
we calculated light curves of the he chemically peculiar star hd37776 .
it is possible to explain its observed light variability provided that the he - poor spots have enhanced metallicity whereas the he - strong ring has low metallicity . in such case
the observed light variations are mostly due to bound - free transitions of carbon .
although this picture roughly corresponds to the observed data , detailed spectrum observations and model improvement are necessary to test our ideas .
this work was supported by grants gar 205/03/ d020 , 205/04/1267 , vega 3014 and mvts sr - r 128/04 .
adelman , s. j. , & pyper , d. m. , 1985 , a&as , 62 , 279 groote , d. , & kaufmann , j. p. , 1981
, in : upper ms cp stars , 23d lige p. 435 hubeny , i. , & lanz , t. 1995 , apj , 439 , 875 mikulek , z. , krtika , j. , zverko , j. , iovsk , j. , & jank , j. , 2006 , these proceedings peterson , d.m .
1970 , apj , 161 , 685 townsend , r. h. d. , owocki , s. p. , & groote d. , 2005 , apj , in press vet , b. , hempelmann , a. , schneich , w. , & stahlberg , j. , 1991 , astr .
nachr . , 312 , 2 | peculiar stars exhibit light variations explained by photometric spots uneven energy distribution on spots connected with spectroscopic spots chemical elements departures energy distribution from line continuum blanketing in spectroscopic spots with chemical composition test photometric spots calculated light variations of helium strong mcp star hd37776 derived spot model with helium - rich ring two helium - poor spherical caps for atmosphere modelling opted code tlusty @xcite models correspond to helium rich / poor surface elements on hd37776 with @xmath0 @xmath1 effective temperature @xmath2 surface gravity @xmath3 of same equal to @xmath4k @xmath5 emergent radiative flux calculated with synspec code spectral energy distribution for surface elements used fluxes in colours fluxes integrated over observed magnitude difference calculated radiative fluxes calculated for atmospheres different chemical composition displayed on fig . flux calculated convolved with gauss function light variations he - poor surface elements with solar metallicity produce flux lower in visible region than he - strong ones regions with enhanced metallicity flux highest inspectrum light variability observations maximum stellar brightness helium minimum si equivalent widths maximum he - rich ring dimmer mostly si - poor he - poor spots brighter mostly si - rich radiative flux from solar metallicity helium - poor surface elements lower helium - strong ones light variability of hd37776 explained by altering flux uneven surface distribution helium surface variations metallicity calculated light variability hd37776 assuming he - poor spots metallicity thirty times higher solar he - strong ring h he only calculated light curves of star hd37776 possible explain light variability he - poor spots enhanced metallicity he - strong ring low metallicity light variations due to bound - free transitions carbon picture corresponds data detailed spectrum observations model improvement necessary work supported by grants 205/03/ d020 205/04/1267 vega 3014 mvts sr - r 128/04 adelman pyper 1985 a&as 62 , 279 groote kaufmann 1981 upper ms cp stars 23d. 435 hubeny lanz 1995 439 , 875 mikulek krtikazverko j iovsk jank j 2006 proceedings peterson 1970 161 , 685 townsend r h d. owocki s p groote d. 2005 press vet b hempelmann schneich stahlberg j. 1991 astr. nachr 312 , 2 | 0.182627 | 990 | 494 | , Saving $0.0 in GPT-4. |
154,633 | modern cosmic ray transport codes , that are capable of use for a variety of applications , need to include all significant atomic , nuclear and particle reactions at a variety of energies .
lepton pair production from nucleus - nucleus collisions has not been included in transport codes to date . using the methods of baur , bertulani and baron , the present paper provides estimates of electron - positron pair production cross sections for nuclei and energies relevant to cosmic ray transport .
it is shown that the cross sections are large compared to other typical processes such as single neutron removal due to strong or electromagnetic interactions .
therefore lepton pair production may need to be included in some transport code applications involving mev electrons . | cosmic radiation transport codes @xcite find use in many diverse areas of physics , astrophysics and space applications . by transporting protons and nuclei through the interstellar medium and comparing the results to the cosmic ray spectrum observed in the solar system , one can deduce properties of interstellar space and also learn about the primary cosmic ray sources . in turn by propagating particles through the earth atmosphere one can relate ground based observations to the solar system cosmic ray spectrum .
this is especially true of extremely high energy cosmic rays where the flux is so low that only ground based observations are feasible .
( actually , a downward looking telescope viewing the fluorescence radiation of a cosmic ray shower trail from the vantage of the space station or from a free flyer gives a much better way to detect these highest energy cosmic rays . )
cosmic transport codes also find many applications in aerospace .
it is important to be able to predict radiation environments inside satellites so that sensitive electronics can be protected .
the same is true for robot missions to planets such as mars , saturn and jupiter .
human space flight in low earth orbit and future lunar and martian missions also require an excellent knowledge of radiation environments inside the spacecraft . again these environments are predicted by transporting cosmic ray particles through spacecraft walls .
the incident cosmic ray spectrum @xcite consists of protons , leptons and heavier nuclei .
the peak of the spectrum lies in the mev - gev region .
typical targets through which particles are transported contain protons and heavier nuclei .
therefore a cosmic ray transport code must include all the atomic , nuclear and particle physics pertaining to an incident nucleus impinging upon any target nucleus over a wide range of energies .
a huge knowledge database is therefore required which must include many reactions . of course , depending on the applications , some reactions may not be as important as others .
however , the best transport codes available today , such as hzetrn @xcite , geant @xcite and fluka @xcite aim at including all possible physics so that they can be used for any application ranging from design of accelerator experiments to radiation therapy . including as large a knowledge database as possible is also important for space applications because the code may be used for such wide ranging topics as calculating the radiation environment inside a spacesuit to the environment on the moon callisto .
a reaction that has not been included in transport codes up to now is lepton pair production in nucleus - nucleus collisions , denoted as @xmath0 for the case where the lepton pair consists of electrons and positrons . here
@xmath1 and @xmath2 are the mass numbers of the projectile and target nucleus respectively .
diagrams illustrating this process are shown in figure 1 .
+ * figure 1 .
* some feynman diagrams for electron - positron pair production via two - photon interactions generated through peripheral nuclear collisions .
the projectile and target nuclei are shown as horizontal straight lines . + this reaction has been studied in detail during the last two decades @xcite and has lead to the realizations that the cross sections are very large @xcite . the spectrum of the resulting electrons has also been calculated and typical electron energies are in the mev region @xcite . applications to spaceflight immediately come to mind .
it is known @xcite that a 10 mev proton can successfully penetrate a spacesuit ( typically of density @xmath3 of equivalent al ) and a 30 mev proton can penetrate the mid - deck of the space shuttle . in comparison electrons
are easily shielded and electrons in the mev range do nt penetrate spacecraft such as the shuttle .
however it is also known that a 0.5 mev electron can penetrate a spacesuit and therefore in calculating radiation environments for extra - vehicular activity ( eva ) , or other applications involving mev electrons , one may need to consider the reaction of equation ( [ reaction ] ) . the aim of the present paper is to give some typical cross section estimates for pair production processes that are relevant to the nuclei and energies found in the cosmic ray spectrum .
these are shown in table 1 .
for the sake of comparison , the cross sections for single neutron removal via strong and electromagnetic interactions are also shown .
@xmath4 , @xmath5 and @xmath6 are calculated according to the methods of references @xcite , @xcite and @xcite respectively .
the equation for lepton pair production is @xcite @xmath7 where @xmath8 and @xmath9 are the projectile and target charges , @xmath10 is the fine structure constant , @xmath11 is the lepton mass and @xmath12 is the relativistic gamma factor of the projectile .
the various factors can be understood from the first order diagram as follows .
each photon - nucleus vertex contains @xmath13 , where @xmath10 is the fine structure constant , and each electron - photon vertex contains @xmath14 . from table 1 it can be seen that in comparison to the neutron removal cross sections which are typically in the millibarn range , the pair production cross sections are much larger , typically in the barn region .
in fact the pair cross section for au + fe at 50 gev is about 200 barn .
the reason is that pair production cross sections scale as @xmath15 @xcite .
the cross sections for production of heavier leptons , such as muons , are much smaller , due to the factor @xmath16 .
lepton pair production cross sections have been calculated in the past and have been shown to be large @xcite . however these calculations have typically been performed at very high energies and for very heavy nuclei .
the aim of the present work has been to present these cross sections for the nuclei and energies typically found in the cosmic ray spectrum and to compare them to other typical reaction cross sections . compared to single nucleon removal
, it is seen that electron - positron pair production cross sections from nuclear collisions are large for the nuclei and energies typically found in the cosmic ray spectrum .
therefore it might be necessary to include this reaction in cosmic ray transport codes involving mev electrons .
this will depend on the particular application , the particles of interest and the incident spectrum that one is considering .
+ acknowledgements : jwn was supported by nasa grant nnl05aa05 g .
* table 1 .
* pair production cross sections @xmath6 from nuclear collisions . @xmath4 and @xmath5 denote the strong and electromagnetic cross sections for single neutron removal .
@xmath17 is the kinetic energy of the projectile .
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxprojectile target t @xmath18 @xmath19@xmath20 + ( agev ) ( mb ) ( mb ) ( mb ) xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx@xmath21c @xmath21c 3 64 0.6 0.7 + 5 64 0.7 2.8 + 10 64 0.9 10 + 50 64 1.5 65 + + @xmath22al 3 77 2.4 3 + 5 77 3 13 + 10 77 4 46 + 50 77 6 305 + + @xmath23fe 3 92 8.5 14 + 5 92 11 52 + 10 92 15 186 + 50 92 25 1220 + + @xmath24si @xmath21c 3 73 1.1 4 + 5 73 1.4 15 + 10 73 1.8 54 + 50 73 3 354 + + @xmath22al 3 86 5 19 + 5 86 6 71 + 10 86 8 253 + 50 86 14 1661 + + @xmath23fe 3 100 17 75 + 5 100 22 282 + 10 100 30 1011 + 50 100 52 6643 + + @xmath23fe @xmath21c 3 89 7 14 + 5 89 8 52 + 10 89 11 186 + 50 89 17 1220 + + @xmath22al 3 102 27 65 + 5 102 36 243 + 10 102 47 872 + 50 102 77 5728 + + @xmath23fe 3 116 104 259 + 5 116 132 973 + 10 116 178 3487 + 50 116 298 22910 + + @xmath25au @xmath21c 3 128 46 127 + 5 128 56 478 + 10 128 73 1714 + 50 128 118 11264 + + @xmath22al 3 141 201 598 + 5 141 250 2245 + 10 141 330 8048 + 50 141 541 52879 + + @xmath23fe 3 156 749 2391 + 5 156 946 8980 + 10 156 1263 32193 + 50 156 2108 211516
10 wilson j w , townsend l w , schimmerling w s , khandelwal g s , khan f , nealy j e , cucinotta f a , simonsen l c , shinn j l and norbury j w 1991 _ transport methods and interactions for space radiations_. nasa reference publication , no . 1257 | cosmic radiation transport codes @xcite use in physics astrophysics space applications transporting protons nuclei interstellar medium comparing results to cosmic ray spectrum solar system deduce properties interstellar space primary cosmic ray sources propagating particles earth atmosphere ground observations to solar system cosmic ray spectrum especially true of high energy cosmic rays flux low ground observations downward telescope viewing from space station free flyer highest energy cosmic rays transport codes applications in aerospace important predict radiation environments inside satellites sensitive electronics for robot missions to mars saturn jupiter human space flight in low earth orbit future lunar martian missions require knowledge of radiation environments spacecraft predicted by transporting cosmic ray particles through spacecraft walls incident cosmic ray spectrum @xcite consists protons leptons heavier nuclei peak in mev - gev region targets particles contain protons heavier nuclei cosmic ray transport code must include atomic nuclear particle physics incident nucleus target nucleus knowledge database required many reactions some reactions not importantbest transport codes hzetrn geant fluka aim all physics for any application from accelerator experiments to radiation therapy large knowledge database important for space applications code used for topics calculating radiation environment inside spacesuit to environment on moon reaction not included in transport codes is lepton pair production in nucleus - nucleus collisions denoted as @xmath0 pair electrons and positrons @xmath1 and @xmath2 are mass numbers of projectile and target nucleus diagrams shown in figure 1 diagrams for electron - positron pair production via interactions collisions projectile and target nuclei as horizontal straight lines reaction studied in cross sections are large spectrum of resulting electrons calculated typical electron energies in mev region applications to spaceflight 10 mev proton can penetrate spacesuit 30 mev proton can penetrate mid - deck of space shuttle electrons easily shielded in mev range do nt penetrate spacecraft 0.5 mev electron can penetrate spacesuit calculating radiation environments for extra - vehicular activity other applications involving mev electrons may consider reaction of equation (present paper typical cross section estimates for pair production processes relevant to nuclei energies cosmic ray spectrum shown in table 1 cross sections for single neutron removal via electromagnetic interactions shown @xmath4 @xmath5 @xmath6 calculated @xcite equation for lepton pair production is @xcite @xmath7 @xmath8 projectile target charges @xmath10 fine structure constant lepton mass @xmath12 relativistic gamma factor projectile each photon - nucleus vertex contains @xmath13 fine structure constant electron - photon vertex @xmath14 neutron removal pair production cross sections larger in barn region pair cross section for au + fe at 50 gev about 200 barn pair production cross sections scale as @xmath15 @xcite cross sections for production heavier leptons smaller due to factor @xmath16 lepton pair production cross sections calculated large @xcite performed at high energies heavy nuclei aim present work cross sections for nuclei energies cosmic ray spectrum to other reaction cross sectionsto single nucleon removal electron - positron pair cross sections from nuclear collisions large for nuclei energies cosmic ray spectrum reaction in cosmic ray transport codes mev electrons on application particles incident spectrum supported by nasa grant table 1 pair production cross sections from nuclear collisions @xmath4 strong electromagnetic cross sections for single neutron removal @xmath17 kinetic energy projectiletarget 64 0.6 0.7 0.7 2.8 0.9 1.5 65 @xmath22al 3 77 2.4 3 77 13 4 46 6 305 @xmath23fe 3 92 8.5 14 5 92 11 52 15 186 25 1220 3 73 1.1 4 73 1.4 15 1.8 54 73 3 354 @xmath22al 3 86 5 19 86 6 71 8 253 14 1661 @xmath23fe 3 100 17 75 + 5 100 22 282 30 1011 52 6643 3 89 7 14 5 89 8 52 89 11 186 89 17 1220 3 102 27 65 102 36 243 47 872 77 5728 @xmath23fe 3 116 104 259 132 973 178 3487 298 22910 3 128 46 127 128 56 478 73 1714 118 11264 @xmath22al 3 141 201 598 141 250 2245 330 8048 141 541 52879 @xmath23fe 3 156 749 2391 156 946 8980 156 1263 32193 156 2108 211516 wilson townsend schimmerling khandelwal khan nealy cucinotta simonsen shinn norbury1991 transport interactions space radiations nasa 1257 | 0.346196 | 2,069 | 1,037 | , Saving $0.1 in GPT-4. |
154,641 | in fritz john s famous paper ( 1979 ) , he discovered that for the wave equation @xmath0 , where @xmath1 and @xmath2 denoting the dalembertian , there is no global solution for any nontrivial and compactly supported initial data .
this paper is intended to simplify his proof by applying a gronwall s type inequality .
blow - up , wave equations , gronwall s type inequality | in 1979 , fritz john published his pioneering work @xcite , which was the first one that discovered the critical power of the blow - up phenomenon for wave equations .
after this article , many people worked on this kind of blow - up problem . for details and many other related references ,
see @xcite .
these work generalize the critical power to other dimensions , some of them also provide simpler proof by imposing additional assumptions or by applying different methods . the paper @xcite claims the following well - known theorem . [ mainthm ]
let @xmath3 be a continuous function which satisfies @xmath4 moreover , suppose there exists @xmath5 and @xmath1 such that for all @xmath6 . then for any function @xmath7 that solves @xmath8 where @xmath9 and both of them have compact support , we have @xmath10 in @xmath11 . here
@xmath2 denotes the dalembertian operator : @xmath12 .
the key step to prove this theorem is the statement as following .
[ keythm ] let @xmath5 , @xmath1 , let @xmath13 be a @xmath14 solution of @xmath15 .
moreover , suppose there exists a point @xmath16 such that @xmath17 for @xmath18 , then @xmath13 has compact support and @xmath19 [ decomp . of u ]
if @xmath13 satisfies @xmath20 with initial data @xmath21 and @xmath22 , then one decomposes @xmath13 by @xmath23 , where @xmath24 solves @xmath25 and @xmath26 suffices @xmath27 for any @xmath16 , the forward cone @xmath28 and the backward cone @xmath29 are defined as @xmath30 and @xmath31 . in the proof of theorem [ keythm ]
, @xcite employs a technical induction which requires complicated calculations . by introducing a suitable nonlinear functional
, this paper gives a much more succinct proof which follows a gronwall s type inequality .
the organization of this paper is as following : in section [ preliminaries ] , it is shown how theorem [ keythm ] implies theorem [ mainthm ] , the argument is from @xcite .
in addition , some notations and a basic lemma are introduced , where the lemma is the key technique used in section [ using g s type ineq . ] . in section [ proof of keythm ] , we prove theorem [ keythm ] .
more precisely , section [ set - up steps ] , the first part of the proof , follows from @xcite with modifications while section [ using g s type ineq . ] , the rest part of the proof , comes from our own observation .
firstly , one can assume that both @xmath21 and @xmath22 have support in @xmath32 , then by huygens principle , @xmath33 in @xmath34 .
it follows from theorem [ keythm ] that @xmath35 .
secondly , one considers the function @xmath36 for @xmath37 . using the assumptions on @xmath38 in theorem [ mainthm ]
, one can see @xmath39 for some fixed @xmath40 depending on @xmath13 and @xmath38 . then by energy estimate , @xmath41 in @xmath42 . thus theorem [ mainthm ]
is verified . for any function @xmath43 , its radial average function ( with respect to spatial variable ) @xmath44
is defined by @xmath45 now let s calculate the radial average of the solution @xmath13 to the wave equation @xmath20 with zero initial data . to do so , one defines @xmath46 by @xmath47 , then we will have @xmath48 hence @xmath49 which implies @xmath50 for convenience , one defines the operator @xmath51 acting on @xmath52 with domain @xmath53 by @xmath54 where @xmath55 ( see figure [ fig first ] ) . ]
it is clear that @xmath51 is a positive operator and now we can rewrite ( [ radial ave . of inhomo .
soln ] ) as @xmath56 whenever @xmath13 solves @xmath20 with initial data .
next , we introduce a lemma which is a generalized gronwall s type inequality with weight .
[ intineqlemma ] let @xmath57 with @xmath58 , suppose @xmath59 to be continuous and @xmath60 for any @xmath61 .
then there does not exist constants @xmath62 such that @xmath63 suppose there exist @xmath62 such that ( [ intineq ] ) holds , then one defines @xmath64 by @xmath65 now for any @xmath66 , it follows from ( [ intineq ] ) that @xmath67 and therefore @xmath68 as a result , for any @xmath69 , @xmath70 now the left hand side is bounded by @xmath71 which is a fixed number . however , the right hand side @xmath72 when @xmath73 , since @xmath74 .
thus , the lemma follows . later in section [ using g s type ineq .
] , in order to prove the finite time blow - up , the goal is to construct a function @xmath75 , associated with the solution of ( [ prob ] ) , which satisfies ( [ intineq ] ) with @xmath76 related to the exponent @xmath77 .
to verify theorem [ keythm ] , first of all , without loss of generality , one can assume @xmath78 to be the origin in @xmath79 , otherwise just doing a translation .
in addition , we can suppose @xmath80 , otherwise just doing a dilation . now using proof by contradiction
, one assumes @xmath81 is not in @xmath82 , then there exists @xmath83 but @xmath84 . set @xmath85 , then @xmath86 and therefore @xmath87 . since @xmath88 in @xmath89 and @xmath87 , then @xmath88 in @xmath90 . as a consequence , for any @xmath91
, @xmath92 noticing the fact that @xmath51 is positive and the assumption @xmath93 in theorem [ keythm ] , it follows from ( [ inhomo .
soln . rep . by p
] ) that @xmath94 because of the simple fact @xmath95 , one has that @xmath96 from ( [ rad .
point est . 1 ] ) , we have @xmath97 since the point @xmath98 lies in @xmath99 and @xmath100 .
now one fixes @xmath101 and a positive number @xmath102 , then considers the regions ( see figure [ fig second ] ) and @xmath103 @xmath104 @xmath105 it is easy to check that the fixed region @xmath106 for any @xmath107 .
then it follows from ( [ rad .
ave . point est .
2 ] ) that for any @xmath107 , @xmath108 where @xmath40 is a positive constant due to ( [ positive on s = t_2+lambda ] ) . let @xmath109 , where @xmath110 .
for any @xmath111 , one defines the sets ( see figure [ fig third ] ) and @xmath112 @xmath113 @xmath114 then for any @xmath111 , we get @xmath115 thus from ( [ rad .
point est . 2 ] ) , ( [ averpointest ] ) , ( [ graphinfo3 ] ) , one obtains @xmath116 the area of @xmath117 is @xmath118 and for any point @xmath119 in it , @xmath120 . as a result ,
@xmath121 where @xmath122 is a positive constant . until now
, the ideas are natural and all the estimates are not difficult to get . but after this step , @xcite claims an induction : @xmath123 where @xmath124 satisfy complex recurrence formulas . by tedious computations , one finds that @xmath125 is the dominant term and @xmath126 , then the blow - up follows when taking @xmath127 to be a fixed large number and @xmath128 . in the following , we will carry out a much more concise argument by introducing a suitable nonlinear functional and making use of lemma [ intineqlemma ] .
firstly , we observe from ( [ rad .
ave . point est . 2 ] ) and ( [ graphinfo3 ] ) that for any @xmath111 , @xmath129 which implies @xmath130 on @xmath131 and especially by ( [ graphinfo3 ] ) , @xmath130 on @xmath132 .
so we can remove the absolute value sign in the last term of ( [ ineq 1 ] ) to get @xmath133 by change of variable : @xmath134 and @xmath135 , we obtain @xmath136 here comes an important observation by considering @xmath137 , where @xmath138 and @xmath139 corresponding to the definition of @xmath131 . now ( [ betteraverpointest ] ) becomes @xmath140 moreover , ( [ posaverineqcov ] ) becomes @xmath141 from here , it attempts to employ the gronwall s inequality technique , which reduces the blow - up problem to a pure analysis technique .
this is our motivation .
however we can not apply it directly since gronwall s inequality only deals with single variable and the right hand side being a double integral .
in addition , ( [ pointestintcov ] ) involves some weight functions . to overcome these difficulties ,
we introduce some new functions and make use of lemma [ intineqlemma ] together with ( [ pointestcov ] ) . for @xmath142 to be determined later
, we define @xmath143 by @xmath144 from ( [ pointestintcov ] ) , @xmath145 we define @xmath146 by @xmath147 and integrate ( [ g1ineq ] ) for @xmath148 from @xmath149 to @xmath150
. then the left hand side of ( [ g1ineq ] ) becomes @xmath151 . in order to exploit lemma [ intineqlemma ] , the right hand side ,
hopefully after changing the order of integration , can become a single integral of @xmath151 .
we calculate as follows , @xmath152 \,d\b\,d\a.\end{aligned}\ ] ] since @xmath142 , then @xmath153 . as a result , @xmath154 we choose @xmath155 and denote @xmath156 to be a constant which is independent of the variable @xmath150 but may be different from line to line , then @xmath157 using holder s inequality , @xmath158 so @xmath159 plugging in ( [ h1ineq ] ) gives @xmath160 now it is almost done !
only trouble for applying lemma [ intineqlemma ] is that @xmath161 may be less than @xmath162 . in order to raise the power of @xmath163
, we borrow the idea from @xcite .
namely , we write @xmath164 and hope to find @xmath165 for some @xmath166 , which can increase the power of @xmath167 by @xmath168 . from ( [ pointestcov ] ) and the definition of @xmath169 and @xmath75 , we have for any @xmath170 , @xmath171 getting back to ( [ hineqsingle ] ) , for any @xmath172 , one gets @xmath173 where @xmath174 since @xmath1 , then @xmath175 and therefore it is possible to choose small @xmath176 in @xmath177 such that @xmath178 . now applying lemma [ intineqlemma ] with @xmath179
, we get contradiction .
thus we reject the assumption @xmath180 , hence theorem [ keythm ] follows .
the authors thank the referees for their suggestions which make this work more clearly .
hengli jiao and zhengfang zhou , an elementary proof of the blow - up for semilinear wave equation in high space dimensions . _ j. differential equations _ ,
* 189 * ( 2003 ) , no . 2 , 355 - 365 .
doi : 10.1016/s0022 - 0396(02)00041 - 4 | 1979 fritz john published work @xcite discovered critical power blow - up phenomenon for wave equations many people worked problem details see @xcite critical power dimensions provide simpler proof assumptions methods @xcite claims theorem @xmath3 continuous function satisfies @xmath4 @xmath5 @xmath1 @xmath6 function @xmath7 solves @xmath8 @xmath9 compact support @xmath10 in @xmath11 @xmath2 denotes dalembertian operator @xmath12 key step prove theorem let @xmath5 @xmath1 @xmath13 be @xmath14 solution of @xmath15 point @xmath16 @xmath17 for @xmath18 @xmath13 has compact support if @xmath13 satisfies @xmath20 @xmath21 @xmath22 decomposes @xmath13 by @xmath23 @xmath24 solves @xmath25 @xmath26 suffices @xmath27 for @xmath16 forward cone @xmath28 backward cone @xmath29 defined as @xmath30 and @xmath31 @xcite employs technical induction complicated calculationsintroducing nonlinear functional paper gives succinct proof gronwall s type inequality organization section preliminaries theorem [ keythm ] implies theorem mainthm argument from @xcite notations basic lemma introduced lemma key technique section using g s type ineq section of keythm prove theorem keythm section set - up steps first proof follows @xcite modifications using g s type ineq rest from observation assume @xmath21 @xmath22 support @xmath32 huygens principle @xmath33 @xmath34 follows from theorem keythm @xmath35 considers function @xmath36 for @xmath37 assumptions @xmath38 @xmath39 fixed @xmath40 depending @xmath13 @xmath38 energy estimate @xmath41 @xmath42 theorem [ mainthm ] verified function @xmath43 radial average function spatial variable @xmath44 defined by @xmath45 calculate radial average solution @xmath13 to wave equation @xmath20 zero initial datadefines @xmath46 by @xmath47 @xmath48 @xmath49 implies @xmath50 defines operator @xmath51 acting on @xmath52 with domain @xmath53 by @xmath54 where @xmath55 figure @xmath51 is positive operator rewrite as @xmath56 whenever @xmath13 solves @xmath20 with initial data introduce lemma generalized gronwall s type inequality with weight intineqlemma @xmath57 with @xmath58 @xmath59 continuous and @xmath60 for @xmath61 exist constants @xmath62 @xmath63 exist @xmath62 defines @xmath64 by @xmath65 @xmath66 follows @xmath67 and @xmath68 @xmath69 , @xmath70 left hand side bounded by @xmath71 fixed number right hand side @xmath72 when @xmath73 since @xmath74 lemma follows prove finite time blow - up goal construct function @xmath75 associated with solution of prob satisfies intineq with @xmath76 related to exponent @xmath77 verify theorem assume @xmath78 origin in @xmath79 translationsuppose @xmath80 dilation using proof by contradiction assumes @xmath81 not in @xmath82 exists @xmath83 but @xmath84 set @xmath85 @xmath86 @xmath87 @xmath88 in @xmath89 and @xmath87 @xmath88 in @xmath90 @xmath91, @xmath92 @xmath51 positive assumption @xmath93 theorem follows from @xmath94 fact @xmath95 has @xmath96 point have @xmath97 point @xmath98 in @xmath99 and @xmath100 fixes @xmath101 positive number @xmath102 considers regions figure fig second @xmath103 @xmath104 @xmath105 check fixed region @xmath106 for @xmath107 follows from @xmath107, @xmath108 @xmath40 positive constant to positive on s = t_2+lambda @xmath109 @xmath110 @xmath111 defines sets figure fig third @xmath112 @xmath113 @xmath114 @xmath111 get @xmath115 from rad . point est .2 averpointest graphinfo3 ] obtains @xmath116 area @xmath117 @xmath118 point @xmath119 @xmath120 @xmath121 @xmath122 positive constant ideas natural estimates not difficult after @xcite claims induction @xmath123 @xmath124 satisfy complex recurrence formulas computations @xmath125 dominant term @xmath126 blow - up follows @xmath127 fixed large number @xmath128 concise argument introducing nonlinear functional lemma [ intineqlemma ] observe rad . ave . point est . 2 ] @xmath111 @xmath129 implies @xmath130 @xmath131 @xmath130 @xmath132 remove absolute value sign last term ineq 1 ] get @xmath133 change variable @xmath134 @xmath135 obtain @xmath136 observation considering @xmath137 @xmath138 @xmath139 definition @xmath131 [ betteraverpointest ] becomes @xmath140 [ posaverineqcov ] ) becomes @xmath141 gronwall s inequality technique reduces blow - up problem pure analysis technique motivation not apply directly gronwall s inequality deals single variable right hand side double integralpointestintcov ] involves weight functions introduce new functions lemma [ intineqlemma ] with pointestcov ] @xmath142 define @xmath143 by @xmath144 @xmath145 @xmath146 by @xmath147 integrate [ g1ineq ] for @xmath148 from @xmath149 to @xmath150 left side of g1ineq becomes @xmath151 exploit lemma intineqlemma right side single integral of @xmath151 calculate @xmath152{aligned since @xmath142 then @xmath153 @xmath154 choose @xmath155 denote @xmath156 constant independent of variable @xmath150 different line @xmath157 using holder s inequality @xmath158 @xmath159 h1ineq gives @xmath160 intineqlemma @xmath161 may be less than @xmath162 raise power of @xmath163 borrow idea from @xcite write @xmath164 find @xmath165 for @xmath166 increase power of @xmath167 by @xmath168 pointestcov definition @xmath169 @xmath75 @xmath170 @xmath171 hineqsingle @xmath172 @xmath173 @xmath174 @xmath1 @xmath175 choose @xmath176 @xmath177 @xmath178 applying lemma @xmath179 contradiction reject assumption @xmath180 theorem keythm follows authors thank referees suggestions hengli jiao zhengfang zhou elementary proof blow - up semilinear wave equation high space dimensions differential equations 189 2003 ) no . 2 355 - 365 doi 10.1016/s0022 - 0396(02)00041 - 4 | 0.431005 | 2,744 | 1,536 | , Saving $0.1 in GPT-4. |
154,644 | we give a simple proof of an explicit formula for kerov polynomials . this explicit formula
is closely related to a recent formula of goulden and rattan . | kerov polynomials are universal polynomials which express the characters of symmetric groups evaluated on cycles , in terms of quantities known as the free cumulants of a young diagram .
we now explain these notions .
let @xmath0 be a young diagram , to which we associate a piecewise affine function @xmath1 , with slopes @xmath2 , such that @xmath3 for @xmath4 large enough , as in fig . 1 below , which corresponds to the partition @xmath5 .
we can encode the young diagram using the local minima and local maxima of the function @xmath6 , denoted by @xmath7 and @xmath8 respectively , which form two interlacing sequences of integers .
these are ( -3,-1,2,4 ) and ( -2,1,3 ) respectively in the picture .
@xmath9 @xmath10 @xmath11 to the young diagram we associate the rational fraction @xmath12 and the formal power series @xmath13 , inverse to @xmath14 for composition , @xmath15 the quantities @xmath16 are called the free cumulants of the diagram @xmath17 .
note that @xmath18 for any young diagram , so we do not include it in the series of free cumulants .
these quantities arise in the asymptotic study of representations of symmetric groups , see @xcite .
it turns out that there exists universal polynomials @xmath19 in the variables @xmath20 such that for any young diagram @xmath17 the normalized character @xmath21 evaluated on a cycle of length @xmath22 is given by @xmath23 the remarkable fact here is that these polynomials do not depend on the size of the symmetric group .
we list the first such polynomials below .
@xmath24 we refer to @xcite and @xcite for more information about results and conjectures on the coefficients of these polynomials .
we take from @xcite , section 5 , the following expression for kerov polynomials . here
@xmath25\,f(z)$ ] denotes the coefficient of @xmath26 ( the residue if @xmath27 ) of a laurent series @xmath28 .
consider the formal power series @xmath29 define @xmath30\ , h(z)h(z-1)\ldots h(z - k+1)\qquad ( 1)\ ] ] and @xmath31\ , h(z)^k\ ] ] then the expression of @xmath32 in terms of the @xmath33s is given by kerov s polynomials .
recently i. p. goulden and a. rattan @xcite have given an explicit expression for kerov polynomials , from which they have deduced a certain number of positivity properties of the coefficients of these polynomials .
their proof uses the lagrange inversion formula . in the next section we use the invariance of residue under change of variables to derive in a simple way a closely related formula , and
show how to recover goulden and rattan s formula .
we use the notations of proposition 1.1 above . let us introduce the power series @xmath34 one has @xmath35 , by lagrange inversion formula .
we use the invariance of the residue under change of variables , namely if @xmath36 are laurent series , and @xmath37 is invertible for composition , then @xmath38\,f(z)=[\zeta^{-1}]\,u'(\zeta)f\circ u(\zeta).\ ] ] using taylor s formula as well as the change of variables @xmath39 in the residue , one gets from ( 1 ) @xmath40\,\prod_{j=0}^{k-1 } \left(\sum_{r=0}^{\infty}\frac{(-j)^r}{r!}h^{(r)}(z)\right ) \\ & = & -{1\over k}\,[\zeta^{-1}]\,l'(\zeta)\prod_{j=0}^{k-1 } \left(\sum_{r=0}^{\infty}\frac{(-j)^r}{r!}h^{(r)}\circ l(\zeta)\right ) . \end{array}\ ] ] using @xmath41 one gets @xmath42 therefore @xmath43\,l'(\zeta)\prod_{j=0}^{k-1 } \left(\zeta+\sum_{r=1}^{\infty}\frac{(-j)^r}{r ! } \left(\frac{1}{l'(\zeta)}\frac{d}{d\zeta}\right)^{r-1 } \frac{1}{l'(\zeta)}\right).\ ] ] putting @xmath44 we obtain the following proposition .
let @xmath45 then kerov s polynomials are given by the following expression @xmath46\,\frac{1}{f(\zeta)}\prod_{j=0}^{k-1 } \left(\zeta+\sum_{r=1}^{\infty}\frac{(-j)^r}{r ! }
\left(f(\zeta)\frac{d}{d\zeta}\right)^{r-1 } f(\zeta)\right).\qquad ( 2)\ ] ]
goulden and rattan give various equivalent formulas for @xmath32 .
they introduce the series @xmath47 , and define polynomials @xmath48\ldots\right]\ ] ] where @xmath49 .
the generating series form of their formula now reads @xmath50\frac{1}{c(z)}\prod_{j=1}^{k-1}(1+\sum_{i=1}^\infty j^ip_i(z)z^i).\qquad ( 3)\ ] ] we recover this formula using @xmath51 . for this
we factor out @xmath52 in the expression in the rhs of @xmath51 , and use the change of variable @xmath53 .
this gives @xmath54\,\frac{1}{f(\zeta)}\prod_{j=0}^{k-1 } \left(\zeta+\sum_{r=1}^{\infty}\frac{(-j)^r}{r ! } \left(f(\zeta)\frac{d}{d\zeta}\right)^{r-1 } f(\zeta)\right)\\ & = & -{1\over k}\,[\zeta^{-1-k } ] \,\frac{1}{f(\zeta)}\prod_{j=1}^{k-1 } \left(1+\sum_{r=1}^{\infty}\frac{(-j)^r}{r!}\zeta^{-1 } \left(f(\zeta)\frac{d}{d\zeta}\right)^{r-1 } f(\zeta)\right)\\ & = & -{1\over k}\,[z^{k+1}]\,\frac{1}{c(z)}\prod_{j=1}^{k-1 } \left(1+\sum_{r=1}^{\infty}\frac{(-j)^r}{r!}z \left(-c(z)z^2\frac{d}{dz}\right)^{r-1 } c(z)\right)\\ & = & -{1\over k}\,[z^{k+1}]\,\frac{1}{c(z)}\prod_{j=1}^{k-1 } \left(1-\sum_{r=1}^{\infty}\frac{j^rz}{r!}(c(z)zd)^{r-1}c(z ) \right).\\ \end { array}\ ] ] now remark that @xmath55 to get @xmath56\ldots\right]\\ & = & -r!p_r(z ) .
\end{array}\ ] ] p. biane , _ characters of symmetric groups and free cumulants . _ asymptotic combinatorics with applications to mathematical physics ( st . petersburg , 2001 ) , 185200 , lecture notes in math . , 1815 , springer , berlin , 2003 .
i. p. goulden , a. rattan , _ an explicit form for kerov s character polynomials . | kerov polynomials express characters symmetric groups evaluated cycles free cumulants of young diagram @xmath0 be young diagram function @xmath1 slopes @xmath2 @xmath3 @xmath4 large fig . 1 partition @xmath5 encode diagram using local minima maxima of function @xmath6 @xmath7 @xmath8 form two interlacing sequences of integers ( -3,-1,2,4 ) ( -2,1,3 ) picture @xmath9 @xmath10 @xmath11 associate rational fraction @xmath12 formal power series @xmath13 @xmath14 @xmath15 quantities @xmath16 free cumulants of diagram @xmath17 @xmath18 include in free cumulants quantities arise in asymptotic study of symmetric groups universal polynomials @xmath19 in variables @xmath20 young diagram @xmath17 normalized character @xmath21 cycle @xmath22 given by @xmath23 polynomials depend on size symmetric group first polynomials @xmath24 refer to @xcite for information results coefficients from @xcite section 5 expression for kerov polynomials .@xmath25\,f(z)$ denotes coefficient @xmath26 residue @xmath27 laurent series @xmath28 power series @xmath29 @xmath30\ h(z)h(z-1)\ldots h(z - k+1)\qquad ( @xmath31\ h(z)^k\ expression @xmath32 @xmath33s kerov s polynomials goulden a. rattan expression kerov polynomials deduced positivity properties coefficients proof uses lagrange inversion formula invariance residue change variables derive closely related formula recover goulden rattan s formula notations proposition 1.1 introduce power series @xmath34 @xmath35 lagrange inversion formulause invariance residue change variables @xmath36 laurent series @xmath37 invertible composition @xmath38(z)=[\zeta^{-1}](\zeta)f\circ(\zeta). taylor s formula change variables @xmath39 residue gets @xmath40\prod_{j=0}{k-1\sum{r=0\infty\frac{(-j)^r}{r{(r} -{1 k}[\zeta^{-1}](\zeta)\prod_{j=0}^{k-1\frac{(-j)^r}{r{(r}(\zeta)\right @xmath41 gets @xmath42 @xmath43(\zeta)\prod_{j=0}{k-1\sum{r=1\infty}\frac{(-j)^r}{r{1(\zeta\zeta}^{r-1(\zeta @xmath44 obtain proposition@xmath45 kerov s polynomials expression @xmath46\frac{1}(\zeta)\prod_{j=0}{k-1(\zeta+\sum_{r=1}\infty\frac{(-j)^r}{r(\zeta)\frac{d}\zeta^{r-1 goulden rattan formulas @xmath32 introduce series @xmath47 define polynomials @xmath48 @xmath49 generating series formula @xmath50\frac{1}{c(z)}\prod_{j=1}^{k-1}(1\sum_{i=1}\infty(z recover formula @xmath51 factor @xmath52 expression @xmath51 change variable @xmath53gives @xmath54\frac{1\zeta\prod{j=0{k-1+\sum{r=1\infty\frac{(-j)^r}{r\frac\right^{r-1 -{1\over k{-1-k\frac{1}{f(\zeta\prod{j=1}^{k-1+\sum_{r=1\infty}\frac{(-j)^r}{r\zeta{-1\zeta)\frac{d\zeta\right^{r-1\right -{1\over k}\frac{1}{c(z}\prod{j=1}^{k-1+\sum{r=1\infty}\frac{(-j)^r}{r\frac{d{dz\right)^{r-1\right -{1\over k\frac{1}{c(z}\prod_{j=1}^{k-1-\sum_{r=1}\infty}\frac{j^rz}{r^{r-1}c(z array remark @xmath55 get @xmath56\ldots\right -r_r(z{arraybiane symmetric groups free cumulants asymptotic combinatorics mathematical physics petersburg 2001 185200 lecture notes math 1815 springer berlin 2003 goulden rattan kerov s character polynomials | 0.338943 | 1,811 | 1,032 | , Saving $0.0 in GPT-4. |
154,658 | we study the scattering of an electron by a ferromagnetic domain wall of the quantum heisenberg - ising model ( xxz model ) with certain boundary conditions .
the spin of the electron interacts with the spins of the xxz model by the hund coupling . using the exact domain wall ground states of the xxz model ,
we analytically obtain the exact effective schrdinger equation for conduction electrons .
this equation coincides with a conventional phenomenological schrdinger equation which was derived in a classical treatment of spins of a domain wall . by solving the schrdinger equation numerically ,
we have calculated the transmission coefficient which is a function of the hund coupling and of the anisotropy of the xxz model .
it turns out that the transmission coefficient is vanishing in the low energy limit for the electron .
* key words : * ferromagnetic metals , ferromagnetic domain walls , magnetoresistance , quantum heisenberg - ising model , xxz model , domain wall ground states # 1@xmath0#1 # 1#2#3 | the early experiments of ferromagnetic materials @xcite show that the resistivity increases with the increase of the number of the domain walls .
this result seems to admit of no doubt because the conduction electrons are scattered by the domain walls @xcite .
an opposite result , however , was reported recently , i.e. , the existence of a _ negative _ domain wall contribution to the resistivity was proposed from experimental @xcite and theoretical @xcite studies . in particular , tatara and
fukuyama @xcite stated that the existence of impurities plays an important roll for the negative domain wall contribution .
we should note that some other recent studies @xcite show that a domain wall contribution to resistivity is _
positive_. as a starting point , it is necessary to study the problem of the scattering of conduction electrons by a single domain wall .
this problem was considered in the early works @xcite by cabrera and falicov .
but they treated the spins of the domain wall classically . in this paper , we study the scattering of an electron by a _
quantum _ domain wall which is realized as a ground state of the quantum heisenberg - ising model ( xxz model ) with certain boundary conditions @xcite .
the spin of the conduction electron interacts with the spins of the xxz model by the hund coupling .
our main results are the following two : ( i ) we analytically obtain the exact effective schrdinger equation for the conduction electron .
more precisely the spin of the conduction electron feels the effective magnetic field of @xmath1 , and the amplitude of the spin flip is proportional to @xmath2 . here
@xmath3 is the one - dimensional coordinate perpendicular to the domain wall centered at @xmath4 , and @xmath5 is the width of the domain wall .
our effective schrdinger equation coincides with a conventional phenomenological schrdinger equation by cabrera and falicov @xcite . from this equation
, one can immediately notice that the domain wall never transmits an electron with low energy without a spin flip of the electron .
( ii ) by solving the effective schrdinger equation numerically , we have calculated the transmission coefficient which is a function of the hund coupling and of the width of the domain wall .
( the width is determined by the anisotropic exchange of the xxz model . )
it turns out that the transmission coefficient is vanishing in the low energy limit for the electron . as an example of a real system ,
consider a quantum wire with a single domain and with low density impurities .
then the transmission coefficient we obatined leads to the conductance by relying on the landauer formula .
unfortunately it is not so easy to realize this situation in an experiment . throughout the present paper ,
we assume that the systems we consider are spatially homogeneous except for the @xmath3 direction perpendicular to domain walls . having this assumption in mind
, we treat the systems in any dimension as a one - dimensional one in this paper .
before discussing a quantum domain wall , we briefly review a classical treatment of a domain wall .
the spin configuration of a classical domain wall ( bloch wall ) is represented by a vector - valued function @xmath6 of the position @xmath3 . in @xcite , the authors assumed @xmath7 , where @xmath5 is the width of the domain wall , and @xmath8 is a real constant .
the interaction between @xmath6 and the spin @xmath9 of a conduction electron is given by the hund coupling @xmath10 with the anisotropy @xmath11 .
then the schrdinger equation for the conduction electron is given by @xmath12 \psi_{\uparrow } = \frac{b}{2}{\rm sech}\left(\frac{x}{\lambda}\right)\psi_{\downarrow } } , \nonumber\\ \quad \nonumber\\ \displaystyle{\left[-\frac{\hbar^2}{2 m } \frac{d^2}{d x^2 } + \frac{b\delta}{2}\tanh\left(\frac{x}{\lambda}\right)-e\right ] \psi_{\downarrow } = \frac{b}{2}{\rm sech}\left(\frac{x}{\lambda}\right)\psi_{\uparrow } , } \end{array } \right .
\label{phenoschro}\ ] ] where @xmath13 and @xmath14 are , respectively , the up ( @xmath15 ) and the down ( @xmath16 ) components of the wavefunction of the electron , and @xmath17 is the mass of the electron . in the region
far right from the domain wall ( @xmath18 ) , a up - spin electron has lower energy of the hund coupling than an down - spin electron . in the opposite region ( @xmath19 ) ,
up and down spins interchange their situation .
if an electron goes thorough the domain wall without a spin flip of the electron , then the electron feels the potential barrier .
a spin flip of the electron occurs only near the domain wall . in this treatment , the effective magnetic potential for the conduction electrons
is assumed to be identical to the profile of the domain wall .
clearly the potential should be derived from a microscopic hamiltonian because the spins of the conduction electrons interact with _
quantum spins _ of a domain wall .
the domain wall also should be realized in the microscopic hamiltonian .
thus the phenomenological schrdinger equation by cabrera and falicov needs to be justified from a microscopic level .
in order to introduce a quantum domain wall , we consider the spin-1/2 heisenberg - ising model ( xxz model ) on the one - dimensional lattice @xmath20 $ ] .
the hamiltonian is given by @xmath21 , \label{hamhei}\ ] ] where @xmath22 is the spin-1/2 operator at the site @xmath23 $ ] , and @xmath24 are the exchange integral and the anisotropy , respectively .
we assume @xmath25 and @xmath26 .
we apply the boundary fields to make a single domain wall as a ground state .
the sector of the ground states @xcite of @xmath27 for a finite size @xmath28 is spanned by the set of the product states @xcite @xmath29 where @xmath30 is a complex number , and @xmath31 here @xmath32 is defined as @xmath33 with @xmath34 , and @xmath35 and @xmath36 are the spin up and down states , respectively .
write @xmath37 with real numbers @xmath38
. then @xmath39 specifies the position of the domain wall and the angle @xmath40 is a quantum mechanical phase corresponding to the degrees of freedom of the rotation about the third axis of the spin .
for simplicity we set @xmath41 .
then we have , from ( [ eta ] ) , @xmath42 in the thermodynamic limit @xmath43 , and from ( [ equation : productstate ] ) , @xmath44 where @xmath45 is the lattice constant , and we have used @xmath46 .
thus the expectation about the _ quantum _ domain wall ground state @xmath47 gives the profile of the classical domain wall .
now we consider the problem of the scattering of a conduction electron by a single quantum domain wall . to treat this problem
, we introduce a heisenberg - kondo model .
see fig .
[ figuremodel ] for the schematic structure .
the hamiltonian is given by @xmath48 with the kinetic energy of the conduction electron @xmath49 and the hund coupling @xmath50{\cal p } \times\ \chi_y(x),\end{aligned}\ ] ] between the spin @xmath51 of the electron and the spins @xmath52 of the domain wall of @xmath27 of ( [ hamhei ] ) . here
@xmath53 are real parameters , @xmath54 is the projection operator onto the sector spanned by the ground states of @xmath27 , and @xmath55 is a characteristic function defined by @xmath56\\ 0 , & \mbox{otherwise . }
\end{array } \right.\end{aligned}\ ] ] by introducing the projection @xmath54 , we have neglected the excitations above the domain wall ground states of the xxz model . in pure one dimension , this approximation is justified because of the energy gap above those ground states @xcite .
but , in higher dimensions there exist gapless excitations above the domain wall ground states @xcite .
the treatment of the excitations is left for future studies @xcite . using the domain wall
ground state @xmath47 of ( [ equation : productstate ] ) , a single - electron eigenstate " can be written as @xmath57 where @xmath58 is the @xmath59 component of the wave function of the conduction electron . then the effective schrdinger equation for the conduction electron is given by @xmath60 where we have used the single - electron schrdinger equation @xmath61 for the hamiltonian @xmath62 of ( [ eq : hamiltonian ] ) and with the energy eigenvalue @xmath63 . using ( [ eq : elementz ] ) and @xmath64 , the equation ( [ eq : schroedinger ] ) is written as @xmath65 \psi_{\uparrow } = \frac{j_k}{4}{\rm sech}\left(\frac{x}{\lambda}\right)\psi_{\downarrow } , } \nonumber\\ \quad \nonumber\\
\displaystyle{\left [ -\frac{\hbar^2}{2m}\frac{d^2}{dx^2 } + \frac{j_k \delta_k}{4 } \tanh\left(\frac{x}{\lambda}\right)-e\right ] \psi_{\downarrow } = \frac{j_k}{4}{\rm sech}\left(\frac{x}{\lambda}\right)\psi_{\uparrow}. } \end{array}\right .
\label{eq : schrodingereq}\ ] ] this coincides with the phenomenological equation ( [ phenoschro ] ) by cabrera and falicov . although the vector ( [ eq : oneparticle ] ) is _ not _ an eigenvector of the hamiltonian @xmath62 of ( [ eq : hamiltonian ] ) in a usual sense , we can justify the above treatment for the present system in any dimension @xcite .
suppose that an @xmath15 electron comes from the right side of the domain wall .
then the spin of the electron feels the potential barrier @xmath66 and flips with the transition amplitude @xmath67 near the domain wall .
since the height of the potential wall is order of @xmath68 , a conduction electron with low energy @xmath69 never goes to the other side without a spin flip .
we solved the schrdinger equation ( [ eq : schrodingereq ] ) numerically and obtained the transmission coefficient as a function of the energy @xmath63 of the conduction electron and of the width @xmath5 of the domain wall .
the results for the transmission coefficients with a spin flip are shown in fig .
[ figuretransmission ] .
the remarkable feature is that the transmission coefficients are vanishing in the low energy limit for the conduction electron . as we expected ,
our results indicate that there appears no negative domain wall contribution to resistivity without introducing a new mechanism such as an effective interaction between domain walls and impurities .
actually such mechanisms were proposed by some authors @xcite for explaining the recent experiments @xcite .
however , it is very difficult to take account of the effect of impurities in a mathematically rigorous manner as in the present paper .
* acknowledgements * the authors are grateful to koh mibu , taro nagahama , naoto nagaosa , yoshiko nakamura , yoshichika otani , teruya shinjo , and gen tatara for useful discussions .
a part of the numerical calculations was performed by using facilities of the supercomputing center in issp , university of tokyo . | early experiments of ferromagnetic materials show resistivity increases with number domain walls conduction electrons scattered by domain walls opposite result reported recently negative _ domain wall contribution to resistivity proposed from experimental theoretical studies tatara and fukuyama stated impurities for negative domain wall contribution other recent studies show domain wall contribution to resistivity positive necessary to study problem scattering of conduction electrons by single domain wall problem considered in early works by cabrera and falicov treated in this paper study scattering of electron by quantum _ domain wall ground state quantum heisenberg - model with certain boundary conditions spin conduction electron interacts with spins xxz model by main results exact effective schrdinger equation for conduction electron spin feels magnetic field of @xmath1 amplitude of spin flip proportional to @xmath2 @xmath3 is coordinate perpendicular to domain wall at @xmath4 @xmath5 is width of domain wall schrdinger equation coincides with conventional schrdinger equation by cabrera and falicov domain wall never transmits electron with low energy without spin flipsolving schrdinger equation calculated transmission coefficient function of hund coupling width domain wall width determined by anisotropic exchange xxz model transmission coefficient in low energy limit for electron consider quantum wire single domain low density impurities transmission coefficient leads to conductance landauer formula not easy to realize in experiment assume systems spatially homogeneous except @xmath3 direction perpendicular to domain walls treat systems as one - dimensional before discussing quantum domain wall review classical treatment domain wall spin configuration classical domain wall represented by vector - valued function @xmath6 of position @xmath3 authors assumed @xmath7 @xmath5 width domain wall @xmath8 real constant interaction between @xmath6 spin @xmath9 of conduction electron given by hund coupling @xmath10 with anisotropy @xmath11schrdinger equation for conduction electron by @xmath12 \psi = \frac{b}{2}{\rm sech}\left({x\lambda\right)\nonumber\\\displaystyle\left[-\frac{\hbar^2}{2 m } \frac{d^2}{d x^2 } + \frac{b\delta}{2}\tanh\left({x\lambda}\right)-e\right \psi\downarrow } = \frac{b}{2}{\rm sech}\left(\lambda{array{phenoschro} @xmath13 and @xmath14 up down components of wavefunction electron @xmath17 mass electron region from domain wall @xmath18 ) up - spin electron has lower energy than down - spin opposite region @xmath19 up down spins interchange electron domain wall without spin flip feels potential barrier spin flip occurs near domain wall effective magnetic potential for electrons identical to profile domain wall potential derived from microscopic hamiltonian interact with spins domain wall domain wall in microscopic hamiltonianphenomenological schrdinger equation by cabrera falicov microscopic level introduce quantum domain wall consider spin-1/2 heisenberg - ising model on one - dimensional lattice @xmath20 hamiltonian by @xmath21 @xmath22 spin-1/2 operator @xmath23 @xmath24 exchange integral anisotropy assume @xmath25 @xmath26 apply boundary fields single domain wall ground state sector ground states @xmath27 finite size @xmath28 spanned by product states @xmath29 @xmath30 complex number @xmath31 @xmath32 @xmath33 @xmath34 @xmath35 @xmath36 spin up down states write @xmath37 with real numbers @xmath38 @xmath39 specifies position domain wall angle @xmath40 quantum mechanical phase degrees freedom rotation third axis spin set @xmath41 @xmath42 thermodynamic limit @xmath43 @xmath45 lattice constant used @xmath46 expectation quantum domain wall ground state @xmath47 profile classical domain wall consider problem scattering conduction electron by single quantum domain wall introduce heisenberg - kondo model see fig . for schematic structurehamiltonian given by @xmath48 with kinetic energy conduction electron @xmath49 hund coupling @xmath50 between spin @xmath51 electron spins @xmath52 domain wall @xmath27 @xmath53 real parameters @xmath54 projection operator sector ground states @xmath27 @xmath55 characteristic function defined by @xmath56\\ introducing projection @xmath54 neglected excitations above domain wall ground states xxz model dimension approximation justified energy gap above ground states higher dimensions gapless excitations above domain wall ground states treatment for future studies using domain wall ground state @xmath47 single - electron eigenstate written as @xmath57 @xmath58 @xmath59 component of wave function conduction electron effective schrdinger equation conduction electron by @xmath60 used single - electron schrdinger equation @xmath61 for hamiltonian @xmath62 energy eigenvalue @xmath63elementz @xmath64 equation schroedinger written @xmath65\psi\uparrow = \frac{j_k}{4}\rm sech}\left({x}{\lambda}\right)\downarrow\nonumber\\\displaystyle\left -\frac{\hbar^2}{2m}{d^2}{dx^2 } + \frac{j_k \delta_k}{4 }\left({x}{\lambda}\right)-e\right\psi\downarrow } = \frac{j_k}{4}{\rm sech}\left\lambda\psi{\uparrow}{array}\right schrodingereq} coincides phenomenological equation phenoschro cabrera falicov vector oneparticle not eigenvector hamiltonian @xmath62 justify treatment system dimension @xcite @xmath15 electron right side domain wall spin feels potential barrier @xmath66 flips with transition amplitude @xmath67 near domain wall height potential wall @xmath68 conduction electron low energy @xmath69 other side spin flipsolved schrdinger equation numerically obtained transmission coefficient function energy @xmath63 conduction electron width domain wall results transmission coefficients with spin flip shown in fig . transmission coefficients vanishing in low energy limit conduction electron results indicate no negative domain wall contribution to resistivity without new mechanism effective interaction between domain walls impurities mechanisms proposed by authors experiments difficult to effect impurities mathematically present paper authors grateful to koh mibu taro nagahama naoto nagaosa yoshiko nakamura yoshichika otani teruya shinjo gen tatara for discussions numerical calculations performed facilities supercomputing center issp university of tokyo | 0.432735 | 2,654 | 1,379 | , Saving $0.1 in GPT-4. |
154,664 | we measure the inclusive branching fractions of charm mesons into three mesons with large @xmath0 content , namely the @xmath1 , @xmath2 and @xmath3 .
data were accumulated with the cleo - c detector . for @xmath4 and @xmath5 rates ,
we use 281 pb@xmath6 taken on the @xmath7 resonance , and for @xmath8 rates , we use 195 pb@xmath6 taken at 4170 mev . we find that the production rates of these particles are larger in @xmath8 decays than in @xmath4 and @xmath5 decays .
the @xmath3 rate , in particular , is 15 times greater .
these branching fractions can be used to measure @xmath9 yields either at the @xmath10 resonance or at hadron colliders . | inclusive decay rates of charm mesons into mesons with large @xmath0 inherent quark content , the @xmath1 , @xmath2 and @xmath3 mesons , are important for studies of both charm and @xmath11 decays .
non - strange @xmath12 mesons are generally expected not to decay into such objects , while the @xmath8 is likely to have both @xmath13 and @xmath14 quarks present after the primary @xmath15 transition , resulting in many more such particles .
this is particularly useful in distinguishing between @xmath16 and @xmath9 mesons as the @xmath16 decays into @xmath12 s with a large rate of @xmath1790% , while the decays into @xmath8 are at the @xmath1710% level .
we expect that the reverse is true for @xmath9 mesons .
knowledge of the charm yields into these mesons would allow alternative analyses of @xmath9 rates at the @xmath10 or at hadron colliders @xcite .
in this analysis we use 281 pb@xmath6 integrated luminosity of cleo - c data produced in @xmath18 collisions and recorded at the peak of the @xmath19 resonance ( 3.770 gev ) to study the @xmath1 , @xmath2 and @xmath3 yields in @xmath4 and @xmath5 decays .
production in @xmath8 decays is studied at 4170 mev , where the cross - section for @xmath20 is @xmath171 nb @xcite .
the cleo - c detector is equipped to measure the momenta and directions of charged particles , identify charged hadrons , detect photons , and determine with good precision their directions and energies .
it has been described in more detail previously @xcite .
fully reconstructed charged or neutral @xmath12 meson candidates are selected from the data at 3.770 gev , where pairs of @xmath23 or @xmath24 mesons are produced .
the decay modes used are listed in table [ tab : d0dprecon ] . in general , in this paper
, @xmath12 refers to either a @xmath4 or @xmath5 meson and its anti - particle , and @xmath25 refers to @xmath8 meson and its anti - particle .
( also , mention of a flavor specific state also implies use of the charge - conjugate state . )
.tagging modes and numbers of signal and background events determined from the fits to the @xmath21 and @xmath26 beam constrained mass distributions , after making the mode - dependant @xmath27 cuts .
the error on the summed signal yield is obtained by adding the errors on the individual yields in quadrature . [ cols="<,<,^,^,^",options="header " , ] the large asymmetry in the yields of these particles between @xmath25 and the lighter @xmath12 mesons will permit further studies of @xmath9 decays at the @xmath28(5s ) , and will be most useful for separating @xmath9 from @xmath16 decays at hadron colliders .
we gratefully acknowledge the effort of the cesr staff in providing us with excellent luminosity and running conditions .
d. cronin - hennessy and a. ryd thank the a.p . sloan foundation .
this work was supported by the national science foundation , the u.s .
department of energy , and the natural sciences and engineering research council of canada .
the function is @xmath30 ^ 2\right)}$ ] . here
@xmath31 is the overall normalization and @xmath16 , @xmath32 and @xmath12 are parameters that govern the shape .
see h. albrecht _
et al . _ ( argus ) , phys .
b * 229 * , 304 ( 1989 ) .
lange , nucl .
methods phys .
res . , sect .
a * 462 * , 152 ( 2001 ) ; r. brun _ et al . _ ,
geant 3.21 , cern program library long writeup w5013 ( 1993 ) , unpublished . for final state photon radiation
see e. barberio , b. van eijk , z. was , comput .
. commun . * 66 * , 115 ( 1991 ) ; e. barberio and z. was , comput .
phys . commun .
* 79 * , 291 ( 1994 ) . | decay rates of charm mesons into large @xmath0 quark content important for studies decays non - strange @xmath12 mesons not decay into objects @xmath8 likely @xmath13 quarks after @xmath15 transition more particles distinguishing @xmath16 @xmath9 mesons @xmath16 decays into @xmath12 rate @xmath1790% @xmath8 @xmath1710% reverse for @xmath9 mesons knowledge charm yields alternative analyses @xmath9 rates at @xmath10 hadron colliders 281 pb@xmath6 luminosity cleo - c data @xmath18 collisions @xmath19 resonance 3.770 gev study @xmath1 yields in @xmath4 @xmath5 decays production in @xmath8 decays studied at 4170 mev cross - section for @xmath20 is @xmath171 nb @xcite cleo - c detector momenta directions charged particles charged hadrons photons directions energies reconstructed charged or neutral @xmath12 meson candidates selected at 3.770 gev @xmath23 or @xmath24 mesonsdecay modes listed in table d0dprecon @xmath12 refers @xmath4 or meson anti - particle @xmath25 @xmath8 meson anti - particle flavor specific state implies charge - conjugate state modes signal background events from @xmath21 @xmath26 beam constrained mass distributions after mode - dependant @xmath27 cuts error summed signal yield adding errors individual yields in quadrature large asymmetry in yields between @xmath25 @xmath12 mesons studies @xmath9 decays at @xmath28(5s useful separating @xmath9 from @xmath16 decays at hadron colliders cesr staff excellent luminosity running conditions d. cronin - hennessy a. ryd thank a.p . sloan foundation supported by national science foundation u.s . department of energy natural sciences engineering research council of canada function is @xmath30 @xmath31 normalization @xmath16 @xmath12 parameters shape h albrecht et al phys . b 229 304 1989 lange methods phys 462 152 2001 r. brungeant 3.21 cern program library writeup w5013 1993 unpublished final state photon radiation barberio van eijk z. comput commun 66 115 1991 ) barberio z. commun 79 291 1994 ) | 0.182605 | 995 | 524 | , Saving $0.0 in GPT-4. |
154,666 | let @xmath0 be the mapping class group of a genus @xmath1 surface .
the group @xmath0 has virtual cohomological dimension @xmath2 . in this note
we use a theorem of broaddus and the combinatorics of chord diagrams to prove that @xmath3 . | let @xmath0 be the mapping class group of a closed , oriented , genus @xmath1 surface , and let @xmath4 be the moduli space of genus @xmath5 riemann surfaces .
it is well - known that for each @xmath6,@xmath7 it is a fundamental open problem to determine the maximal @xmath8 for which these vector spaces are nonzero .
harer @xcite proved that the _ virtual cohomological dimension _
@xmath9 equals @xmath2 .
more precisely , he proved that @xmath10 for a certain @xmath0-module @xmath11 ( see below for details ) and that @xmath12 for all @xmath13 and all @xmath0-modules @xmath14 .
thus the first step of the problem above is to determine whether @xmath15 .
the purpose of this note is to answer this question .
let @xmath16 ( resp .
@xmath17 ) denote the mapping class group of the genus @xmath5 surface with one marked point ( resp .
one boundary component ) .
[ theorem : main ] for any @xmath1 , @xmath18 further , the rational cohomology of @xmath16 ( resp .
the integral cohomology of @xmath17 ) vanishes in its virtual cohomological dimension .
this theorem was announced some years ago by harer , but he has informed us that his proof will not appear .
we recently learned that morita
suzuki @xcite have independently found a proof of theorem [ theorem : main ] using a completely different method .
they apply a theorem of kontsevich on graph homology to their computation of a generating set for a certain symplectic lie algebra .
our proof combines some results about the combinatorics of chord diagrams with the work of broaddus @xcite on the steinberg module of @xmath0 .
we thank allen hatcher and takuya sakasai for their comments on an earlier version of this paper , and john harer for informing us about the paper @xcite and his own work .
theorem [ theorem : main ] is consistent with the well - studied analogy between mapping class groups and arithmetic groups .
for example , theorem 1.3 of lee
szczarba @xcite states that the rational cohomology of @xmath19 vanishes in its cohomological dimension .
we begin by briefly summarizing previous results that make our computation possible ; for details see broaddus @xcite .
let @xmath20 be a connected , closed orientable surface of genus @xmath1 .
let @xmath21 be the _ curve complex _ of @xmath20 defined by harvey @xcite , i.e. the flag complex whose @xmath22-simplices are the @xmath23-tuples of distinct free homotopy classes of simple closed curves in @xmath20 that can be realized disjointly .
harer @xcite proved that @xmath21 is homotopy equivalent to a wedge of spheres @xmath24 .
there exists a constant @xmath25 such that any two closed geodesics on a hyperbolic surface of length @xmath26 are disjoint ( the _ margulis constant _ for hyperbolic surfaces ) .
let @xmath27 be the teichmller space of marked hyperbolic surfaces diffeomorphic to @xmath20 having no closed geodesic of length @xmath28 .
it is known that @xmath27 is a @xmath29-dimensional manifold with corners .
ivanov @xcite proved that @xmath30 is contractible and that its boundary @xmath31 is homotopy equivalent to @xmath21 .
briefly , for each simplex @xmath32 of @xmath21 , let @xmath33 be the subset of @xmath31 consisting of surfaces where each curve in @xmath32 has length @xmath34 .
each @xmath35 is contractible , and @xmath36 unless @xmath37 is a simplex of @xmath21 , in which case @xmath38 .
the mapping class group @xmath0 acts properly discontinuously on @xmath27 with finite stabilizers .
defining @xmath39 , it follows that @xmath40 .
mumford s compactness criterion states that @xmath41 is compact . combining this with the previous two paragraphs , the work of bieri eckmann ( * ? ? ?
* theorem 6.2 ) shows that @xmath42 and that @xmath43 in fact , we can say more
. let @xmath11 denote the _ steinberg module _
, i.e. the @xmath0-module @xmath44 . then @xmath45 is the rational _ dualizing module _ for @xmath0 , meaning that @xmath46 for any @xmath22 and any @xmath47 .
moreover @xmath11 is also the dualizing module for @xmath48 and @xmath17 , which act on @xmath11 via the natural surjections @xmath49 and @xmath50 @xcite .
this implies that for @xmath51 we have @xmath52 .
for @xmath17 we obtain a similar result with @xmath53 , except that since @xmath17 is torsion - free the result holds integrally : @xmath54 .
fix a finite - volume hyperbolic metric on @xmath55 .
another model for @xmath11 comes from the _ arc complex _
@xmath56 , the flag complex whose @xmath22-simplices are the disjoint @xmath23-tuples of simple geodesics on @xmath55 beginning and ending at the cusp @xmath57
. let @xmath58 be the subcomplex consisting of collections of geodesics @xmath59 for which @xmath60 has some non - contractible component .
harer proved that @xmath58 is homotopy equivalent to @xmath21 @xcite , and that @xmath56 is contractible @xcite ( see also @xcite ) .
thus @xmath61 by examining how the geodesics are arranged in a neighborhood of @xmath57 , an @xmath62-simplex of @xmath56 can be encoded by a @xmath63-chord diagram ; see @xcite .
an _ ordered @xmath63-chord diagram _ is an ordered sequence @xmath64 , where @xmath65 is an unordered pair of distinct points on @xmath66 ( a _ chord _ ) and @xmath67 if @xmath68 .
we will visually depict @xmath69 by drawing arcs connecting the points in each @xmath65 ( see figure [ figure:1 ] for examples ) .
two ordered chord diagrams are identified if they differ by an orientation - preserving homeomorphism of the circle .
an unlabeled _ @xmath22-filling system _ of genus @xmath5 is a @xmath70-chord diagram satisfying the conditions described in @xcite : no chord should be parallel to another chord or to the boundary circle , and the chords should determine exactly @xmath71 boundary cycles .
these conditions , which guarantee that these chords define a simplex of @xmath72 , have the following simple combinatorial formulation .
given @xmath64 , consider two permutations of the @xmath73 points @xmath74 : let @xmath75 be the @xmath73-cycle which takes each point to the point immediately adjacent in the clockwise direction , while @xmath76 exchanges the two points of each chord @xmath65 and thus is a product of @xmath63 transpositions . then a @xmath70-chord diagram is a @xmath22-filling system of genus @xmath5 if @xmath77 has @xmath71 orbits , none of which have length 1 or 2 .
finally , let @xmath78 be the straight line in @xmath79 connecting the two points of @xmath65 .
then we say that @xmath69 is _ disconnected _ if the set @xmath80 is not connected .
fix a genus @xmath5 , and set @xmath81 .
let @xmath82 be the free abelian group spanned by ordered @xmath22-filling systems of genus @xmath5 modulo the following relation .
for @xmath83 and @xmath64 , define @xmath84 .
we impose the relation @xmath85 .
the differential @xmath86 is defined as follows .
consider an ordered @xmath22-filling system @xmath64 of genus @xmath5 .
for @xmath87 , let @xmath88 equal @xmath89 if this is an ordered @xmath90-filling system of genus @xmath5 ; otherwise , let @xmath91 . then @xmath92 we will need the following theorem of broaddus @xcite . recall that if @xmath93 is a group and @xmath47 is a @xmath93-module , then the _ module of coinvariants _ , denoted @xmath94 , is the quotient @xmath95 .
let @xmath96 be the @xmath97-filling system of genus @xmath5 depicted in figure [ figure:1]a .
[ theorem : broaddus]for each @xmath98 , the following hold .
a. @xmath99 .
b. the abelian group @xmath100 is spanned by the image @xmath101\in \u_0/\partial(\u_1)$ ] of @xmath102 . c. if @xmath103 is a disconnected @xmath97-filling system of genus @xmath5 , then the image of @xmath103 in @xmath100 is @xmath97 .
for part ( i ) of theorem [ theorem : broaddus ] , see ( * ? ? ?
* proposition 3.3 ) together with the remark preceding ( * ? ? ?
* example 4.1 ) ; for part ( ii ) , see ( * ? ? ?
* theorem 4.2 ) ; and for part ( iii ) , see ( * ? ? ?
* proposition 4.5 ) .
for any group @xmath93 and any @xmath93-module @xmath47 , recall that @xmath104 . since the actions of @xmath16 and @xmath17 on @xmath11 factor through @xmath0 , to prove theorem [ theorem : main ] it suffices by to show that @xmath105 . by theorem [ theorem : broaddus](i ) , this is equivalent to showing that @xmath106 . for @xmath107 , let @xmath108 $ ] denote the associated element of @xmath100 .
let @xmath109 be the @xmath97-filling system depicted in figure [ figure:1](a ) . by theorem [ theorem : broaddus](ii ) , it is enough to show that @xmath101=0 $ ]
. let @xmath110 be the @xmath111-filling system depicted in figure [ figure:1](b ) .
observe that @xmath112 where the second equality holds since the indicated chord diagrams differ by an orientation preserving homeomorphism of @xmath66 . similarly , @xmath113 .
also , @xmath114 ( resp .
@xmath115 ) by definition , since the chord @xmath116 ( resp .
@xmath117 ) becomes parallel to the boundary .
we thus have @xmath118 for @xmath119 , the @xmath97-filling system @xmath120 is disconnected , so theorem [ theorem : broaddus](iii ) implies that @xmath121=0 $ ] .
we conclude that @xmath122=0 $ ] .
now consider the @xmath111-filling system @xmath123 depicted in figure [ figure:1](c ) . removing any chord from figure [ figure:1](c ) yields figure [ figure:1](a ) up to rotation ,
so @xmath124 for each @xmath8 .
in fact , it is clear that @xmath125 , that @xmath126 , that @xmath127 , and so on , with @xmath128 .
this shows that @xmath129 so @xmath130 = 0 $ ] .
w. j. harvey , boundary structure of the modular group , in _ riemann surfaces and related topics :
proceedings of the 1978 stony brook conference _ , 245251 , ann . of math . stud .
, 97 , princeton univ . press , princeton , n.j . , 1981 | @xmath0 mapping class group closed genus @xmath1 surface @xmath4 moduli space genus @xmath5 riemann surfaces @xmath6,@xmath7 open problem determine maximal @xmath8 vector spaces nonzero harer @xcite proved virtual cohomological dimension @xmath9 equals @xmath2 proved @xmath10 certain @xmath0-module @xmath11 @xmath12 @xmath13 @xmath0-modules @xmath14 first step determine @xmath15 purpose note answer question @xmath16 @xmath17 ) denote mapping class group genus @xmath5 surface one marked point boundary component theorem @xmath1 @xmath18 rational cohomology @xmath16 integral cohomology @xmath17 vanishes virtual cohomological dimension theorem announced years ago by harer proof appear morita suzuki @xcite found proof theorem different method apply theorem kontsevich graph homology computation generating set symplectic lie algebra proof combines results combinatorics chord diagrams work broaddus @xcite steinberg module @xmath0 thank allen hatcher takuya sakasai comments earlier version paper john harer informing paper @xcite worktheorem consistent with analogy between mapping class groups arithmetic groups theorem 1.3 szczarba states rational cohomology of @xmath19 vanishes in cohomological dimension results broaddus @xmath20 connected closed surface of @xmath1 @xmath21 curve complex of @xmath20 harvey flag complex @xmath22-simplices @xmath23-tuples homotopy classes closed curves harer @xmath21 homotopy equivalent to wedge spheres @xmath24 constant @xmath25 two closed geodesics on hyperbolic surface @xmath26 disjoint @xmath27 space of hyperbolic surfaces diffeomorphic to @xmath20 no closed geodesic @xmath27 @xmath29-dimensional manifold with corners ivanov @xmath30 contractible boundary @xmath31 homotopy equivalent to @xmath21 simplex @xmath32 @xmath21 @xmath33 subset of @xmath31 curve @xmath32 length @xmath34 @xmath35 contractible @xmath36 unless @xmath37 simplex of @xmath21 mapping class group @xmath0 acts discontinuously on @xmath27 with finite stabilizers@xmath39 follows @xmath40 mumford s compactness criterion states @xmath41 compact bieri eckmann theorem 6.2 shows @xmath42 @xmath43 @xmath11 steinberg module @xmath0-module @xmath44 @xmath45 dualizing module @xmath0 @xmath46 @xmath22 @xmath47 @xmath11 dualizing module for @xmath48 @xmath17 @xmath11 natural surjections @xmath49 @xmath50 @xcite implies @xmath51 @xmath52 @xmath17 similar result @xmath53 @xmath17 torsion - free result holds @xmath54 finite - volume hyperbolic metric on @xmath55 model @xmath11 from arc complex @xmath56 @xmath22-simplices disjoint @xmath23-tuples simple geodesics @xmath55 @xmath57 @xmath58 subcomplex collections geodesics @xmath59 @xmath60 non - contractible component @xmath58 homotopy equivalent to @xmath21 @xcite @xmath56 contractible @xcite @xmath61 geodesics @xmath57 @xmath62-simplex of @xmath56 encoded by @xmath63-chord diagramordered @xmath63-chord diagram ordered sequence @xmath64 @xmath65 unordered points on @xmath66 @xmath67 @xmath68 depict @xmath69 arcs connecting points @xmath65 figure ordered chord diagrams by orientation preserving homeomorphism circle unlabeled @xmath22-filling system @xmath5 @xmath70-chord diagram no chord parallel boundary circle chords determine @xmath71 boundary cycles conditions define simplex @xmath72 combinatorial formulation consider two permutations @xmath73 points @xmath74 @xmath75 @xmath73-cycle point adjacent clockwise @xmath76 exchanges two points chord @xmath65 product of @xmath63 transpositions @xmath70-chord diagram @xmath22-filling system @xmath5 if @xmath77 has @xmath71 orbits length 1 or 2 @xmath78 straight line in @xmath79 connecting two points @xmath65 @xmath69 disconnected if @xmath80 not connected fix genus @xmath5 set @xmath81 @xmath82 free abelian group by ordered @xmath22-filling systems @xmath5 @xmath83 and @xmath64 define @xmath84impose relation @xmath85 differential @xmath86 defined consider ordered @xmath22-filling system @xmath64 genus @xmath5 @xmath87 @xmath88 equal @xmath89 ordered @xmath90-filling system genus @xmath5 otherwise let @xmath91 @xmath92 need theorem broaddus @xcite @xmath93 group @xmath47 @xmath93-module module coinvariants @xmath94 @xmath95 @xmath96 @xmath97-filling system genus @xmath5 figure theorem broaddus each @xmath98 hold @xmath99 abelian group @xmath100 spanned image @xmath101 @xmath102 @xmath103 disconnected @xmath97-filling system genus @xmath5 image @xmath103 @xmath100 is @xmath97 part i theorem see proposition 3.3 ) remark example 4.1 ) part ii ) see theorem 4.2 iii proposition 4.5 ) group @xmath93 @xmath93-module @xmath47 recall @xmath104 actions @xmath16 @xmath17 @xmath11 factor through @xmath0 prove theorem main suffices show @xmath105 .theorem broaddus(i ) equivalent @xmath106 @xmath107 @xmath108 associated element @xmath100 @xmath109 @xmath97-filling system figure:1](a ) theorem(ii ) show @xmath101=0 $ @xmath110 @xmath111-filling system figure:1](b ) @xmath112 second equality holds chord diagrams differ preserving homeomorphism @xmath66 @xmath113 @xmath114 @xmath115 chord @xmath116 parallel boundary @xmath118 @xmath119 @xmath97-filling system @xmath120 disconnected theorem broaddus(iii implies @xmath121=0 $ conclude @xmath122=0 $ @xmath111-filling system @xmath123 figure:1](c ) removing chord figure:1](c ) yields figure:1](a ) rotation @xmath124 each @xmath8 @xmath125 @xmath126 @xmath127 @xmath128 shows @xmath129 @xmath130 = 0 $ w. j. harvey boundary structure modular group riemann surfaces related topics proceedings 1978 stony brook conference 245251 ann . math . stud . 97 princeton univpress princeton 1981 | 0.430599 | 2,741 | 1,559 | , Saving $0.1 in GPT-4. |
154,671 | the giant hii region ngc604 constitutes a complex and rich population to studying detail many aspects of massive star formation , such as their environments and physical conditions , the evolutionary processes involved , the initial mass function for massive stars and star - formation rates , among many others . here , we present our first results of a near - infrared study of ngc604 performed with niri images obtained with _
gemini north_. based on deep _ jhk _ photometry , 164 sources showing infrared excess were detected , pointing to the places where we should look for star - formation processes currently taking place .
in addition , the color - color diagram reveals a great number of objects that could be giant / supergiant stars or unresolved , small , tight clusters .
a extinction map obtained based on narrow - band images is also shown . | ngc604 is a giant hii region ( ghr ) located in an outer spiral arm of m33 , at a distance of 840 kpc .
it is the second most luminous hii region in the local group , after 30 doradus in the lmc .
both are nearby examples of giant star - forming regions whose individual objects can be spatially resolve for further study .
ngc604 has been the target of many studies during the past few decades .
a brief summary of known facts about ngc604 includes the following .
this ghr is ionized by a massive , young cluster , with at least 200 o stars ( some as early as o3-o4 ) .
the cluster does not exhibit a central core distribution .
instead , the stars are widely spread over its projected area in a structure called scaled ob association ( soba ; @xcite ; @xcite ; @xcite ) .
wolf - rayet stars , a confirmed and many candidate red supergiant stars , a luminous blue variable and a supernova remnant are all part of ngc604 s stellar population ( @xcite ; @xcite ; @xcite ; @xcite ; @xcite ; @xcite ; @xcite ) . the age of the central ionizing cluster has been determined by different authors as between 3 and 5 myr ( @xcite ; @xcite ; @xcite ; @xcite ; @xcite ) . + the interestellar medium reveals a complex structure with a high - excitation central region ( made up of multiple two - dimensional structures ) , asymmetrically surrounded by a low - excitation halo .
the whole region shows a very complex geometry of cavities , expanding shells and filaments , as well as dense molecular regions .
all of these structures show different kinematic behaviour ( @xcite ; @xcite ; @xcite ; @xcite ) .
+ aiming to characterize the youngest stellar population and its environment , we performed near - infrared ( nir ) photometry ( _ jhk _ ) and analysed narrow - band images in pa@xmath0 , br@xmath1 and h@xmath2(2 - 1 ) .
taking into account that nir observations are less affected by dust extinction characteristic in star - fomation environments , we can take a deep dive into ngc604 to study those very young objects which are still immersed in their parental clouds at the sites of current star formation .
the images were obtained with the near infrared imager and spectrometer ( niri ) at _ gemini north_. the resulting plate scale is 0.117 arcsecpixel@xmath3 , with a field of view of 120 @xmath4 120 arcsec@xmath5 . the filters used and their main characteristics are listed in table 1 .
.main characteristics of broad - band and narrow - band filters used in our gemini / niri observations . [ cols="^,^,^,^,^,^ " , ] the images were taken under excellent seeing conditions , on average @xmath6 in the j , h , and k@xmath7 images .
a set of approximately 10 individual exposures was taken in each band to combine into the final image .
data reduction and processing were performed with specific tasks using the gemini - niri iraf package .
images were sky subtracted and flat fielded and short darks exposures were used to identify bad pixels .
stellar magnitudes were obtained by point - spread - function ( psf ) fitting in crowded fields using daophot software ( stetson 1987 ) in iraf .
although a standard procedure , psf construction and fitting involves an iterative and careful process in which several tries were made to get the best results .
the effective area covered by our photometry is @xmath8 107 @xmath4 107 arcsec@xmath5 ( @xmath8 430 @xmath4 430 pc@xmath5 at the distance of m33 ) .
the average photometric errors are 0.09 , 0.11 and 0.21 mags in the h , j and k@xmath7 filters , respectivelly , and the completeness limits are 22 mags in j and 21 mags in h and k@xmath7 .
magnitudes in the individual filters were matched in a unique list containing 5566 objects in the field in which all three j , h and k@xmath7 magnitudes were measured .
astrometry was derived using 35 objects in common in our field of ngc604 and the gsc - ii catalog , version 2.3.2 ( 2006 ) in the icrs , equinox j2000.0 .
) composite image of ngc604 with 8.4 ghz radio - continuun countours overlaid ( adapted from @xcite ) .
red circles are objects that show ir excess in our photometry . _ ( right ) _ extinction map from br@xmath1/h@xmath9 , regions with higher extinction are in red / white.,title="fig:",scaledwidth=45.0% ] ) composite image of ngc604 with 8.4 ghz radio - continuun countours overlaid ( adapted from @xcite ) .
red circles are objects that show ir excess in our photometry . _ ( right ) _ extinction map from br@xmath1/h@xmath9 , regions with higher extinction are in red / white.,title="fig:",scaledwidth=53.0% ] the resulting color - magnitude ( cm ) and color - color ( cc ) diagrams are shown in figures 1 and 2 , respectively .
we have included @xmath8 2000 selected objects , located within a radius of 48 arcsec ( @xmath8 200 pc ) and centered on ngc604 , meeting a certain photometric quality level ( magnitude error @xmath10 median(error)+ 1.0 @xmath4 standard deviation(error ) ) .
red symbols in both plots are objects that lie on the right side of the reddening line for a o6-o8 v star.for each of these objects , the error in ( h - k@xmath7 ) color is smaller than its distance to the reddening line , so that we can ensure that they undoubtedly show an ir excess . + among the objects with intrinsic ir excesses in ghrs we expect to find wolf - rayet stars , early supergiants and of stars , and massive young stellar objects ( mysos ) .
many of these types of objects ( or candidates ) have already been found in ngc604 . based on our deep _
photometry we found a total of 164 objects that show ir excesses .
as can be expected , @xmath8 70% of these lie in a small area near the region s center and a large proportion are tightly grouped in regions coincident with the radio - continuum peaks at 8.4 ghz from @xcite , as shown in figure 3 ( left panel ) .
these results are in complete agreement with the analysis of @xcite based on hst / nicmos nir images , where myso candidates also appeared aligned with the radio peak structures .
+ what we found supports the idea mentioned by many authors , that those areas may be embedded star - forming regions , also taking into account that regions coincident with the radio knots show conditions of high temperature and density ( @xcite ) , and in two of them @xcite identified compact hii regions .
most objects with ir excesses will be the targets for further observing programs ( making use of the integral - field spectroscopic facilities at _ gemini north _ ) to elucidate their nature and derive accurate properties by means of spectroscopic study . also , those objects that are located at the bright , massive end on the cm diagram deserve further observation and study .
+ on the basis of our narrow - band images we generated an extinction map , derived from the observed variations of the br@xmath1 to h@xmath9 ratio ( with the h@xmath9 image taken from @xcite ) .
the color scale used in the map shows regions with higher extinction in red / white .
the present and future results of our nir study will be placed in the context of previous studies of ngc604 to complete the picture of the overall formation and evolution scenario of this ghr . + * acknowledgements * + rb acknowledges partial support from the universidad de la serena , project diuls cd08102 . | ngc604 is giant hii region in outer spiral arm of m33 distance 840 kpc second most luminous hii region in local group after 30 doradus in lmc both nearby examples giant star - forming regions individual objects spatially resolve for further study ngc604 target of many studies past decades summary facts ngc604 ghr ionized by massive young cluster with 200 o stars o3-o4 ) cluster central core distribution stars spread over projected area in scaled ob association wolf - rayet stars confirmed candidate red supergiant stars luminous blue variable supernova remnant part of ngc604 s stellar population age of central ionizing cluster between 3 and 5 myr interestellar medium reveals complex structure with high - excitation central region multiple two - dimensional structures surrounded by low - excitation halo region shows complex geometry of cavities expanding shells filaments dense molecular regions structures show different kinematic behaviourcharacterize youngest stellar population performed near - infrared nir photometry analysed narrow - band images pa@xmath0 br@xmath1 h@xmath2(2 nir observations less affected dust extinction star - environments deep dive ngc604 study young objects immersed parental clouds star formation images obtained near infrared imager spectrometer at gemini north_ plate scale 0.117 arcsecpixel@xmath3 field of view 120 @xmath4 120 arcsec@xmath5 filters main characteristics listed in table 1 broad - narrow - band filters gemini niri observations images taken under excellent seeing conditions @xmath6 j h k@xmath7 images 10 exposures each band final image data reduction processing gemini - niri iraf package images sky subtracted flat fielded short darks exposures identify bad pixels stellar magnitudes obtained by point - spread - function psf ) fitting crowded fields daophot software psf construction fitting iterative process effective area covered photometry @xmath8 107 @xmath4 107 arcsec@xmath5 430 @xmath4 430 distance m33average photometric errors 0.09 0.11 0.21 mags in h j k@xmath7 filters completeness limits 22 mags j 21 mags in h k@xmath7 magnitudes filters matched in unique list 5566 objects field j h k@xmath7 magnitudes measured astrometry derived using 35 objects ngc604 gsc - ii catalog version 2.3.2 2006 equinox j2000.0 composite image of ngc604 ghz radio - continuun countours red circles objects show ir excess extinction map higher extinction red / white composite image ngc604 8.4 ghz radio countours red circles objects show ir excess extinction map higher extinction red / white color - magnitude cc diagrams in figures 1 2 included @xmath8 2000 selected objects within radius 48 arcsec centered on ngc604 meeting photometric quality level ( magnitude error @xmath10 median(error)+ 1.0 @xmath4 standard deviation(error red symbols objects right side of reddening line for o6-o8 v starobjects error in h - k@xmath7 ) color smaller than distance to reddening line show ir excess objects intrinsic ir excesses in ghrs wolf - rayet stars early supergiants stars massive young stellar objects ( mysos objects found in ngc604 deep _ photometry found 164 objects show ir excesses @xmath8 70% small area near region s center large proportion grouped in regions with radio - continuum peaks at 8.4 ghz from @xcite in figure 3 results agreement with analysis @xcite hst / nicmos nir images myso candidates aligned with radio peak structures areas be star - forming regions regions with radio knots show high temperature density two identified compact hii regions objects with ir excesses targets for further observing programs integral - field spectroscopic facilities at gemini north elucidate nature properties objects at bright massive end on cm diagram deserve further observation narrow - band images generated extinction map from variations br@xmath1 to h@xmath9 ratio image from @xcite color scale shows regions higher extinction in red / whitepresent future results nir study context previous studies ngc604 formation evolution scenario ghr acknowledgements rb acknowledges partial support universidad la serena project diuls cd08102 | 0.316958 | 1,795 | 898 | , Saving $0.1 in GPT-4. |
154,677 | thermodynamic - driven filament formation in redox - based resistive memory and the impact of thermal fluctuations on switching probability of emerging magnetic switches are probabilistic phenomena in nature , and thus , processes of binary switching in these nonvolatile memories are stochastic and vary from switching cycle - to - switching cycle , in the same device , and from device - to - device , hence , they provide a rich in - situ spatiotemporal stochastic characteristic .
this work presents a highly scalable neuromorphic hardware based on crossbar array of 1-bit resistive crosspoints as distributed stochastic synapses .
the network shows a robust performance in emulating selectivity of synaptic potentials in neurons of primary visual cortex to the orientation of a visual image .
the proposed model could be configured to accept a wide range of nanodevices . | cognitive computing with nanoelectronics is an emerging field of research that aims to fill the gap between cpus performance and mammalian brains .
cpus outperform human brain in almost all tasks that involve deterministic computation .
they , however , lack a very important feature , which is the ability to learn and work with unreliable building blocks .
this unique characteristic of mammalian brains has attracted attention of scientists from different fields including electronic engineering . in the late 1980s ,
caver mead , and later eric vittoz , envisioned the use of very - large - scale integration ( vlsi ) systems to mimic neuro - biological architectures of nervous system @xcite .
it is believed that the hippocampus of a mammalian brain is responsible for memory and learning .
it consists of neurons and chemical synapses @xcite . on the contrary to the common belief that neural systems , including hippocampus , are analog systems , there are several neurobiological evidence for existence of discrete changes of synaptic strength at least in some subfields in hippocampus @xcite .
there are also a number of research that suggest efficient learning with digital synapses @xcite .
therefore , even though an analog approach promises more power efficiency , a digital implementation improves design simplicity , it does not have limited scalability of analog implementations , it is less costly , small , and by far more integrable , hence , practically the best way , if not the only way , to implement highly scalable neuromorphic systems @xcite .
for instance , analog designs allow low - voltage operation , which then requires larger sensing circuitry and also does not necessarily support reasonable throughputs ( spikes / s ) .
it is expected that massively parallel , reliable , scalable , and potentially energy efficient cognitive systems could be implemented using 1-bit stochastic synapses .
a deterministic approach , has its focus on more number of bits per synapse to allow a more precise implementation of a learning rule , and generally suffers multiple data conversion ( adcs and dacs ) . on the other side ,
emerging artificial synapses , such as rram ( resitive ram ) , cbram ( conductive bridge ram ) , pcm ( phase change memory ) and atomic switches , in their analog operating mode , have issues like ( i ) complicated programming pulse - schemes to achieve more intermediate states , which sometimes only exists in one switching direction ( either off - on or on - off ) , and ( ii ) the problem of resistance drift or state retention @xcite . technologies like stt - mram ( spin - transfer torque magnetoresistive ram ) and meram ( magnetoelectric ram ) also show existence of multi - stable states , but their low signal - to - noise ( snr ) properties @xcite suggests them impractical for conventional analog neuromorphic computing .
a workaround to these problems is to make the artificial synaptic transmission a probabilistic process , which can be achieved either in - situ ( randomness within the device ) or ex - situ ( external random number generator ) @xcite .
this work uses the in - situ stochastic switching characteristic of rram devices to present a technology independent approach to implement a highly scalable neuromorphic hardware that is capable of emulating sensitivity of synaptic potentials in neurons of primary visual cortex to the orientation of a visual image .
considering the underlying physical mechanisms of binary switching in rram , cbram and pcm it can be envisioned that their yield - loss and variation remain substantially larger than their cmos counterparts .
therefore , a probabilistic approach is one of the most practical and highly scalable designs for neuromorphic hardware and logic operations with nanoelectronics .
stochastic binary switching is common between the emerging memory technologies .
this behavior normally occurs under certain condition , which is called `` weak '' programming .
although , the weak programming condition is different across technologies , a technology independent approach is presented to pave the way for further unification of system - level behavior based on above mentioned technologies .
stochastic switching of a nanodevice could be treated as a black - box ; a generalized model , which is identified by a set of inputs and outputs .
[ fig : func ] provides an overview of the model .
outputs @xmath0 and @xmath1 represent next state and next conductance of the device , which both are probabilistic and their values are defined by the functions @xmath2 and @xmath3 , respectively . once @xmath0 is determined , @xmath1 is calculated using an experimentally verified distribution ( in this paper , log - normal ) of on and off state resistances .
the function @xmath2 is a function of voltage / current , time ( here pulse - width ) and operating temperature .
small updates in @xmath3 is independent of function @xmath2 output , which means if @xmath4 , @xmath5 and @xmath1 may not be equal .
( or current , @xmath6 ) , operating temperature , @xmath7 , signal pulse - width , @xmath8 , initial state @xmath9 and initial conductance @xmath10 .
device type and relevant fitting parameters are given by @xmath11 .
the model accepts cross - point yield as a defect rate input @xmath12 ( not to be confused with the materials point defect ) .
function @xmath2 defines switching probability and function @xmath3 defines conductance value randomness under a given c2c variation distribution . switching probabilities are defined as @xmath13 and @xmath14 for switching from logic state 0 , s@xmath15 , to logic state 1 , s@xmath16 , and from logic state 1 to logic state 0 , respectively .
outputs are current and voltage responses , next state ( @xmath0 ) , next conductance ( @xmath1 ) , as well as total consumed energy @xmath17 of the whole operation to performing a programming or sensing , regardless of its success.,scaledwidth=45.0% ] experimental data on voltage , pulse - width and temperature dependency of a resistive memory systems is gathered from @xcite .
device switching probability is defined as a strong function of applied voltage , @xmath18 , and time , @xmath19 . for simplicity , temperature dependency and switching threshold variation of the device
are considered to have an overall impact on the switching probability , therefore , their impact are not discussed as independent variables .
[ fig_swprob ] demonstrates switching probability of the set ( s@xmath20s@xmath21 ) and reset ( s@xmath22s@xmath23 ) . under similar condition ,
a set is more likely than a reset . given that the s@xmath20s@xmath21 seems more likely , it is possible to modulate the probabilistic ltp using a complicated learning rule , @xmath24 , where @xmath25 represents synaptic weight that is bounded between a maximum and a minimum limits , through a different programming pulse - scheme for ltp , which makes the design more complex .
experimentally verified information rgarding the switching probability model are given in @xcite .
the switching probability is achieved by integrating a poisson - like distribution , which is in an excellent agreement with the hypothesis of thermodynamic activation during filament formation over a dominant energy barrier in rram devices . .
( a ) demonstrates set switching probability and ( b ) shows reset switching probability .
@xmath26 represents applied voltage pulse - width.,scaledwidth=70.0% ] the mentioned experimental studies suggest that the switching probability of redox - based memory is a cumulative phenomenon , which links the digital and analog characteristic of the device
. the probabilistic set and reset , under such circumstance , can therefore be shown as a function of number of applied pulses with a given amplitude - time combination to model the history dependent effects .
one could argue that such spike - history - dependent - plasticity provides further phenomenological similarities between these artificial synapses and the chemical synapses , where history dependent effects also play an important role @xcite .
this stochastic behavior in combination with @xmath24 results a probabilistic ltp ( long - term potentiation ) and ltd ( long - term depression ) , which is a function of time difference ( @xmath19 ) between a pre- and a post - synaptic spike ( probabilistic spike timing - dependent plasticity , pstdp ) or their difference in voltage ( @xmath18 ) .
stochastic change of the device state can also occur under a sequence of applied pulses with a fixed @xmath27 .
the probabilistic plasticity , which defines the probability of change in the device state , is shown in fig .
[ fig_plasticity ] . here
@xmath24 is defined as @xmath28 it can be shown that a simple form of @xmath24 , like eq .
( [ equ : deltaw ] ) , is enough to successfully implement a robust and reliable selectivity task .
there is a significant cycle - to - cycle resistance variation .
this uncertainty is defined and verified using the function @xmath3 .
the rram and cbram devices have experimentally shown log - normal distribution of on and off state resistances . fig .
[ fig_res ] illustrates high and low resistance variation of a rram device for more than @xmath29 switching cycles . in some devices , like @xcite , off
state resistances may shows wider variation due to the uncontrolled metallic filament dissolution .
@xmath29 switching cycles .
this behavior is captured by @xmath3.,scaledwidth=95.0% ]
a two layer neural network is implemented with @xmath30-pixel binary visual images .
inputs were generated using a gabor - like function to produce more than a thousand , randomly centered , receptive fields .
each input pixel drives an input neuron that spikes if the pixel represents 1 .
there are nine output neurons that are fully connected to all input neurons through one - bit resolution stochastic rram connections . therefore , @xmath31 synapses are available that are randomly programed and some are faulty .
faulty device patterns , random initial state , and some of the visual inputs are shown in fig .
[ fig_result1 ] .
output neurons are integration - and - fire neurons that are connected to each other through inhibitory connections , which form a winner - takes - all network .
output neurons could be configured to implement hebbian , anti - hebbian , stdp , and anti - stdp learning with a support for a programmable refractory period .
the system implements an unsupervised competitive learning that is capable of accepting a wide range of input patterns , including visual and auditory with or without preprocessing . and @xmath32 of faulty devices are stuck - at - on .
( b ) demonstrates initial random patterns of on and off devices .
black pixels show off state devices and white pixels are on device .
half of the functioning devices are randomly programmed as on , and the remaining as off .
( c ) some of the randomly generated input patterns are shown here.,scaledwidth=90.0% ] energy per spike is an important factor .
the model is able to estimate both dynamic switching energy and static leakage power through unselected devices and sneak current paths with or without selector devices .
although , there is little experimental information available regarding spatial variation ( device - to - device ) , the model aims to consider overall impact of such variation with some worst - case assumptions . during training
, inputs were shown to the network and each was repeated several times .
after training , as shown in fig .
[ fig_result2]a , output neurons become sensitive to several orientations . fig .
[ fig_result2]b shows that the system was not sensitive to any angle prior to learning . the system sensitivity to particular angles of the visual input patterns
is demonstrated in fig .
[ fig_result2]c .
one of the main parameters of designing the system is the dynamic tuning of the output neurons firing threshold , which is not the focus of this paper .
the training process were repeated with different threshold tuning approaches and a range of initial conditions , device variation , and defect patterns .
the system continues to perform reliably under extreme yield ( defect rate is @xmath33@xmath34 ) and variability assumptions , where the function @xmath2 output varies within @xmath35 probability ( @xmath36 ) around its nominal value . implementing such system with low - power synapses would produce an ultra - high energy efficient and highly scalable system with an energy consumption of @xmath37@xmath38 pj / spike .
@xmath39 test patterns before training ( in response to fig . [ fig_result1]b ) and after training
are shown in ( b ) and ( c ) , respectively.,scaledwidth=90.0% ]
the implemented neuromorphic system is ( i ) highly scalable , ( ii ) easily mappable on the current crossbar memory architecture , ( iii ) not sensitive to device variation , ( iv ) working reliably with or without selector devices , ( v ) potentially very high energy efficient , and ( vi ) potentials for high throughput applications . each training process , in this case , demands for an endurance requirement in the order of @xmath33@xmath40k cycles , which is well within the current reported endurances of rram , cbram , pcm , stt - mram , meram , and atomic switches .
cost - wise , implementing such network with rram , cbram , pcm , and atomic switches may be cheaper relative to stt - mram and meram .
however , rapid progress of these technologies promises high energy efficient memory systems , which makes the cost - energy trade - off of these devices more attractive in the near future . modeling accuracy can be further improved through systematic studies like @xcite .
omid kavehei would like to thank dr shimeng yu for his insightful discussions on memory technologies and ms nafise erfanian saeedi for her comments on the neural network implementation and selectivity . | cognitive computing with nanoelectronics emerging research gap between cpus performance mammalian brains cpus outperform human brain in tasks deterministic computation lack feature ability to learn work with unreliable building blocks unique characteristic mammalian brains attracted attention scientists electronic engineering late 1980s caver mead eric vittoz envisioned - large - scale integration ( vlsi ) systems to mimic neuro - biological architectures nervous system hippocampus of mammalian brain responsible for memory learning consists of neurons chemical synapses belief systems analog neurobiological evidence for discrete changes of synaptic strength in hippocampus efficient learning with digital synapses analog approach promises power efficiency digital implementation improves design simplicity scalability less costly small more integrable best way to implement highly scalable neuromorphic systems analog designs allow low - voltage operation requires larger sensing circuitry not support reasonable throughputs expected parallel reliable scalable energy efficient cognitive systems could be implemented using 1-bit stochastic synapses deterministic approach on more bits per synapse precise suffers multiple data conversionemerging artificial synapses rram cbram pcm atomic switches analog operating mode have issues like complicated programming pulse - schemes intermediate states in one switching direction problem of resistance drift state retention technologies like stt - mram meram magnetoelectric show multi - stable states low signal - to - noise properties impractical for conventional analog neuromorphic computing workaround is artificial synaptic transmission a probabilistic process achieved in - situ or ex - situ work uses in - situ stochastic switching of rram devices technology independent approach scalable neuromorphic hardware emulating sensitivity synaptic potentials in neurons primary visual cortex to orientation visual image mechanisms of binary switching in rram cbram pcm yield - loss variation remain larger than cmos counterparts probabilistic approach is practical highly scalable for neuromorphic hardware logic operations with nanoelectronics stochastic binary switching common between emerging memory technologies occurs under `` weak '' programming across technologies technology independent approach for unification of system - level behavior based on technologiesstochastic switching nanodevice black - box generalized model identified by inputs outputs fig : func overview model outputs @xmath0 @xmath1 represent next state conductance device probabilistic values defined by functions @xmath2 @xmath3 @xmath0 determined @xmath1 calculated using verified distribution log - normal on off state resistances function @xmath2 of voltage / current time pulse - operating temperature updates in @xmath3 independent of @xmath2 output @xmath4 equal current operating temperature signal pulse - width initial state conductance @xmath10 device type fitting parameters by @xmath11 model accepts cross - point yield as defect rate input @xmath12 not materials point defect function @xmath2 defines switching probability @xmath3 defines conductance value randomness under c2c variation distribution switching probabilities defined as @xmath13 @xmath14 for switching from logic state 0 to 1 outputs are current voltage responses next state conductance total consumed energy @xmath17 operation programming sensing success0% experimental data on voltage pulse - width temperature dependency of resistive memory systems from @xcite device switching probability as function of applied voltage time temperature dependency switching threshold variation on switching probability not independent variables fig_swprob demonstrates switching probability of set reset set more likely than reset to modulate probabilistic ltp using complicated learning rule @xmath24 represents synaptic weight between maximum minimum limits different programming pulse - scheme for design complex verified information switching probability model in @xcite achieved by integrating poisson - like distribution with hypothesis of thermodynamic activation during filament formation over energy barrier in a demonstrates set switching probability b shows reset switching probability @xmath26 represents applied voltage pulse - width switching probability of redox - based memory is cumulative phenomenon links digital analog characteristic device probabilistic set reset as function of number of applied pulses with amplitude - time combination effectsspike - history - dependent plasticity provides similarities between artificial synapses chemical synapses @xcite stochastic behavior with @xmath24 results probabilistic ltp potentiation ltd depression ) function of time difference between pre- post - synaptic spike difference voltage @xmath18 stochastic change device state under applied pulses with fixed @xmath27 probabilistic plasticity state shown in fig . @xmath24 defined as @xmath28 simple form of @xmath24 selectivity task significant cycle - to - cycle resistance variation uncertainty defined verified using function @xmath3 rram cbram devices shown log - normal distribution of on off state resistances fig . illustrates high low resistance variation of rram device for @xmath29 switching cycles some devices off state resistances variation to uncontrolled metallic filament dissolution behavior captured by @xmath3 two layer neural network implemented with @xmath30-pixel binary visual images inputs generated using gabor - like function thousand randomly centered receptive fields each input pixel drives neuron spikes if represents 1 .nine output neurons connected to input neurons through one - bit resolution stochastic rram connections @xmath31 synapses randomly programed faulty faulty device patterns random initial state visual inputs shown in fig ._result1 output neurons integration - and - fire neurons connected through inhibitory connections winner - takes - all network implement hebbian anti - hebbian stdp anti - stdp learning programmable refractory period system implements unsupervised competitive learning input patterns visual auditory without preprocessing faulty devices stuck - at - on initial random patterns of on and off devices black pixels show off white pixels on half functioning devices randomly programmed as on remaining off randomly generated input patterns shown energy per spike important model dynamic switching energy static leakage power through unselected devices current paths little experimental information spatial variation model impact with worst - case assumptions during training inputs shown repeated several times after training output neurons sensitive to orientations system not sensitive to angle prior to learning system sensitivity to angles visual input patterns demonstrated in fig main system dynamic tuning output neurons firing threshold not focus paper training repeated with threshold tuning approaches initial conditions device variation defect patterns system reliably under extreme yield defect rate @xmath33@xmath34 variability assumptions function @xmath2 output varies within @xmath35 around nominal value system with low - power synapses ultra - high energy efficient highly scalable system energy consumption @xmath37@xmath38 pj / spike @xmath39 test patterns before after training shown in ( b ) c ) neuromorphic system highly scalable easily mappable on crossbar memory architecture not sensitive to device variation with without selector devices potentially high energy efficient high throughput applications training process demands endurance requirement @xmath33@xmath40k cycles within current endurances of rram cbram pcm stt - mram meram atomic switches network with rram cbram pcm atomic switches cheaper rapid progress promises high energy efficient memory systems cost - energy trade - off attractive modeling accuracy improved through studies like @xciteomid kavehei dr shimeng yu memory technologies nafise erfanian saeedi comments neural network selectivity | 0.506334 | 2,930 | 1,390 | , Saving $0.1 in GPT-4. |
154,680 | a simple iteration methodology for the solution of a set of a linear algebraic equations is presented .
the explanation of this method is based on a pure geometrical interpretation and pictorial representation .
convergence using this method is obtained and a simple numerical example is provided . | in various scientific fields there is a need to reduce the initial problem to a set of linear algebraic equations .
this is true in physics when solving eigenstates or eigenfunction problems in quantum mechanics . also , it is applicable in applied physics and mathematics in the areas of linear programming , optimization methods and the root mean square method
. therefore this reduction of the initial problem to a set of linear algebraic equations can be viewed as a critical step in obtaining the correct solution to the initial problem .
the methods used in obtaining a solution for a set of the linear algebraic equations are well known .
typically , the process begins with cramer s formulas , deriving a final solution from the linear systems coefficients or similar numerical methods , then using a gaussian method to obtain further improvement and modifications or even applying the jacobi method to obtain further refinement .
however , the real problem starts with the attempt to effectively solve a set of linear equations containing a large number of variables in which cramer s method is not effective , because we need to calculate determinants , which in itself is a problem when a large number of variables are involved because there will be a loss of accuracy and the numerical stability of the calculation procedure is reduced .
this would also apply in the application of both the gaussian and jacobi methods as well . in obtaining practical solutions
we may also have observed linear algebraic set coefficients with error bars which makes the solution obtained even more sensitive to the method used in obtaining a numerical solution . in practice one of the main requirements desired in the numerical method selected
is its stability and fast convergence in obtaining the final solution . in this paper
the goal is to present a simple and rather convenient blueprint to describe the programming iteration method that could be used to solve the linear algebraic equation problem mentioned above .
in this section we want reformulate the algebraic problem involving the linear system solution using a geometrical language . therefore , we need to obtain a number such as @xmath0 , which satisfies the set of the linear system : @xmath1 which can be written in a more convenient way , as : @xmath2 where @xmath3 and @xmath4 completely define the set of linear equations and have the following bounds @xmath5 . from a geometrical point of view
we can consider every line in the set of linear equations , for any fixed number @xmath6 , as a hyperplane in the @xmath7 - dimensional space of the solution set .
this means , that we can consider the solution as a vector , defined by @xmath8 , such that the matrix @xmath3 transforms the vector @xmath9 into the vector @xmath10 which can be defined as @xmath11 .
for instance , let s take some fixed point @xmath12 , where @xmath13 is a @xmath6-th hyperplane , such that @xmath14 . then for the arbitrary point @xmath15 which also belongs to this hyperplane , we can write , that the scalar product @xmath16 , where @xmath17 , is defined as @xmath18 is @xmath19 to the @xmath20 vector . this case can be described as : @xmath21 therefore , we can see , that from a geometrical point of view the solution of a linear set of algebraic equations is some point in @xmath7 space , which is an intersection of all @xmath7 hyperplanes .
one of the possibly solutions therefore , is the construction of an iterative process , which will allow convergence at this point .
for the simplification of the numerical procedure , we can rewrite the set of the linear equations in the following way : @xmath22 where we delineate , that : @xmath23 here we want to underline , that within the normalization procedure , the @xmath6 index is fixed .
thus , we can now rewrite the set of the linear equations using the same form , but now our coefficients are normalized which means that for any indices pair of @xmath6 , and @xmath24 we observe : @xmath25 , which leads to the conclusion that all of our vectors which are perpendicular to their hyperplanes are interrelated .
for the iteration procedure we need to choose the first point , from which the iteration procedure will start . it is obvious , that the iteration procedure convergence will be dependent on this starting point .
it would be preferred to derive this point as close as possible to the final solution . which of course is impossible to do in a general case .
however , at a minimum we can require that within the iteration procedure itself that the chosen initial starting point must be numerically stable . in this case
, we can start from a point which for example might belong to the first hyperplane and have coordinates:@xmath26 , where @xmath27 .
for the start of the iteration procedure we need in that procedure a method for the projection of any arbitrary point @xmath28 ( defined by vector @xmath29 ) on the @xmath6 -th hyperplane . in this case
it is easy to see , that if the projection of this point is some point @xmath30 ( defined by vector @xmath31 ) , which belongs to the @xmath6-th hyperplane , then : @xmath32 here @xmath33 is a unit vector , which is @xmath19 to the @xmath6-th hyperplane . because @xmath34 , then we can state that : @xmath35 from where we determine : @xmath36 and it follows that @xmath37 after determination of the projection parameter @xmath38 we can now describe the components of the projection point @xmath30 : @xmath39 where our iteration procedure consists of the following steps : + 1 .
the first step starts from the procedure of choosing the 1-st point , which will belong to the 1-st hyperplane and we will project it on to the next or second hyperplane .
the second step consists of the projection of the 2-nd point which belongs to the 2-nd hyperplane , on to the next or 3-rd hyperplane and so forth until the last @xmath7 hyperplane is reached .
the third step then consists of the projection of the iteration point from the @xmath7-th hyperplane onto the first hyperplane , in other words the one which began our iteration procedure .
after that we then can repeat all of the above listed procedures again until a complete convergence to the final solution is reached .
proving the convergence in the numerical procedure described above can be accomplished in the following way .
if we take some arbitrary initial point @xmath28 , which belongs to a hyperplane and its projection point @xmath30 on the @xmath6-th hyperplane , then it is clear , that the distance between projection point @xmath30 and our final point of solution , noted as @xmath40 , is : @xmath41 therefore for the @xmath24-th number of iterations we can to write , that : @xmath42 .
then the distance between solution @xmath40 and our projection point @xmath30 after @xmath7 projections can be written as : @xmath43 for example , in the case , when all our hyperplanes are mutually orthogonal , then we have , for any number @xmath24 a value of @xmath44 and then we can automatically obtain a solution after , at least for @xmath45 iterations .
however , in a case in which the difference between vectors , which happen to be @xmath19 to the hyperplanes is @xmath46 for any two given hyperplanes then in this case the number of iteration procedures becomes @xmath47 .
it is clear , from a geometric point of view that in this case we have almost parallel hyperplanes and the convergence will proceed slowly .
it is possible however , to roughly estimate the approximate number of iterations : @xmath48 from which we arrive at : @xmath49 here @xmath50 - is the accuracy of the solution .
this example simply shows how the number of iterations depends on the choice of the first point selected for any arbitrary system , in general .
however , the main result is that this iteration procedure is numerically stable , it is not dependent on the choice of the initial starting point to achieve convergence , if a solution to the set of linear systems exists .
let s consider a simple numerical example .
we will use our method to solve the following set of the linear equations : @xmath51 the exact solution for this set of the linear equations is : @xmath52 for the first starting point we have : @xmath53 the result of this iteration procedure is presented in table 1 .
based on these results we are able to state that @xmath54 is the maximum deviation from the solution at the given number of iterations when @xmath55 .
ccccccc & & & number of iterations & + & 10 & 30 & 50 & 70 & 100 & 200 + @xmath56 & 6.36729084 & 0.24225408 & 1.00692936 & 1.00648839 & 0.99979049 & 1.0 + @xmath57 & -1.70446795 & 2.26483449 & 2.011421866 & 1.99653770 & 2.00010026 & 2.0 + @xmath58 & -5.01380157 & 3.74870823 & 3.02248973 & 2.99148333 & 3.00025466 & 3.0 + @xmath59 & 13.2914602 & 4.28298482 & 3.89357617 & 4.00503095 & 3.99990957 & 4.0 + @xmath60 & 1.96186307 & 4.89304449 & 5.03535119 & 4.99844825 & 5.00002627 & 5.0 + @xmath61&5.9 & 0.3677 & 0.0377 & 0.0046 & @xmath62 & @xmath63 +
in the current mathematical literature there exists many different and useful methods for obtaining a solution for a set of the linear algebraic equations .
the most notable include works by cramer , gauss , jacobi , gauss - seidel .
these methods provide a very useful solution to the problems described in the works @xcite .
however , a major problem arises in attempts to solve problems involving a very high number of variables , for example when the number of variables is equal to @xmath64 or more . in this case
we need to allow more time for convergence , especially if we use an iteration method .
however , at the same time many of these existing methods become numerically unstable . for example , even a solution of obtaining the inverse matrix to @xmath3 becomes a non trivial problem .
the application of our method is rather simple at the programming level and it has a transparent geometrical explanation and easy to understand pictorial representation .
the efficiency of this method is much higher in solutions involving problems with large numbers of variables .
the iteration procedure works rather quickly , for example in our case when the number of iterations is equal to @xmath65 it takes less then 5 sec .
in other methods the number of iterations required are sometimes much less than in our method , but they may require some special properties and/or conditions .
this is the case of the jacobi method .
our approach is not sensitive to these problems . a.k .
express his gratitude to the bogoliubov laboratory of theoretical physics ( jinr , dubna , russia ) for the invitation and warm hospitality . a grant from osp ,
cloud state university
( st . cloud , minnesota , usa ) is also gratefully acknowledged .
99 ilyin v.a .
, pozniak e.g. ( 2004 ) . _ linear algebra _ ,
fizmatlit , moscow .
maksimov , y. a. ( 1982 ) .
_ alhoritms of the solution of nonlinear programming problems _ , mifi publisher , moscow .
volkov , e. a. ( 2003 ) . _
numerical methods _ ,
fizmatlit , moscow . | scientific fields need to reduce initial problem to linear algebraic equations true in physics solving eigenstates eigenfunction problems quantum mechanics applicable in applied physics mathematics linear programming optimization methods root mean square method reduction initial problem to linear algebraic equations critical step in correct solution methods obtaining solution for linear algebraic equations known process begins with cramer s formulas final solution from linear systems coefficients similar numerical methods gaussian method improvement or jacobi method refinement real problem starts with attempt to solve linear equations containing large number variables cramer s method not effective need to calculate determinants problem when large variables loss of accuracy numerical stability calculation reduced in gaussian and jacobi methods practical solutions observed linear algebraic set coefficients with error bars solution more sensitive to method numerical solution main requirements desired in numerical method stability fast convergence final solution paper goal to present simple convenient blueprint describe programming iteration method to solve linear algebraic equation problem section reformulate algebraic problem linear system solution using geometrical languageneed obtain number @xmath0 satisfies linear system : @xmath1 written as @xmath2 @xmath3 and @xmath4 define linear equations bounds @xmath5 . consider every line in linear equations fixed number @xmath6 as hyperplane in @xmath7 - dimensional space solution set consider solution as vector defined by @xmath8 matrix @xmath3 transforms vector @xmath9 into vector @xmath10 defined as @xmath11 . take fixed point @xmath12 @xmath13 is @xmath6-th hyperplane @xmath14 arbitrary point @xmath15 belongs to hyperplane write scalar product @xmath16 @xmath17 defined as @xmath18 is @xmath19 to @xmath20 vector case described as : @xmath21 solution of linear set algebraic equations is point in @xmath7 space intersection of all @xmath7 hyperplanes construction of iterative process convergence rewrite linear equations : @xmath22 delineate @xmath23 within normalization procedure @xmath6 index fixed .rewrite linear equations same form coefficients normalized for indices pair of @xmath6 @xmath24 observe : @xmath25 conclusion vectors perpendicular to hyperplanes interrelated for iteration procedure choose first point convergence dependent on starting point preferred to derive point close to final solution impossible require initial starting point must be numerically stable start from point to first hyperplane coordinates:@xmath26 , @xmath27 for start iteration procedure need method for projection of arbitrary point @xmath28 ( defined by vector @xmath29 ) on @xmath6 -th hyperplane if projection of point is point @xmath30 ( defined by vector @xmath31 ) belongs to @xmath6-th hyperplane : @xmath32 @xmath33 is unit vector @xmath19 to @xmath6-th hyperplane @xmath34 state : @xmath35 determine : @xmath36 follows @xmath37 after determination projection parameter @xmath38 describe components of projection point @xmath30 : @xmath39 iteration procedure consists steps : + 1 .first step starts choosing 1-st point to 1-st hyperplane project next or second hyperplane second step projection 2-nd point to 2-nd hyperplane next or 3-rd hyperplane until last @xmath7 hyperplane reached third step projection of iteration point from @xmath7-th hyperplane onto first hyperplane began iteration procedure repeat procedures until complete convergence to final solution reached proving convergence numerical procedure arbitrary initial point @xmath28 to hyperplane and projection point @xmath30 on @xmath6-th hyperplane distance between projection point @xmath30 and final point of solution @xmath40 is @xmath41 for @xmath24-th number of iterations write : @xmath42 distance between solution @xmath40 and projection point @xmath30 after @xmath7 projections written as : @xmath43 hyperplanes mutually orthogonal for any number @xmath24 value of @xmath44 automatically obtain solution after for @xmath45 iterations difference between vectors @xmath19 to hyperplanes is @xmath46 for two hyperplanes number of iteration procedures becomes @xmath47 almost parallel hyperplanes convergence slowlypossible estimate number iterations @xmath48 @xmath49 @xmath50 accuracy solution shows number iterations depends first point arbitrary system iteration procedure numerically stable not dependent on initial starting point if solution linear systems exists numerical example method solve linear equations @xmath51 exact solution @xmath52 first starting point @xmath53 result iteration in table 1 @xmath54 maximum deviation from solution given number iterations @xmath55 number iterations 10 30 50 70 100 200 + @xmath56 6.36729084 0.24225408 1.00692936 1.00648839 0.99979049 1.0 + @xmath57 -1.70446795 2.26483449 2.011421866 1.99653770 2.00010026 2.0 + @xmath58 -5.01380157 3.74870823 3.02248973 2.99148333 3.00025466 3.0 + @xmath59 13.2914602 4.28298482 3.89357617 4.00503095 3.99990957 4.0 + @xmath60 1.96186307 4.89304449 5.03535119 4.99844825 5.00002627 5.0 + @xmath61&5.9 0.3677 0.0377 0.0046 @xmath62 @xmath63 current literature many methods obtaining solution linear algebraic equationsnotable include works by cramer gauss jacobi gauss - seidel methods provide useful solution to problems major problem arises in problems high number of variables when equal to @xmath64 or more need more time for convergence especially iteration method existing methods become numerically unstable obtaining inverse matrix to @xmath3 non trivial application of our method simple transparent geometrical explanation easy pictorial representation efficiency higher in solutions with large numbers variables iteration procedure works quickly iterations equal to @xmath65 takes less then 5 sec other methods iterations required less may require special properties conditions case of jacobi method our approach not sensitive to these problems gratitude to bogoliubov laboratory of theoretical physics for invitation hospitality grant from osp , cloud state university acknowledged ilyin v pozniak. ( 2004 ) _ linear algebra fizmatlit moscow maksimov, y. ( 1982 ) alhoritms of solution of nonlinear programming problems mifi publisher moscow volkov e. ( 2003 ) numerical methods _ fizmatlit moscow | 0.361227 | 2,592 | 1,417 | , Saving $0.1 in GPT-4. |
154,691 | boomerang is a balloon - borne telescope designed for long duration ( ldb ) flights around antarctica .
the second ldb flight of boomerang took place in january 2003 .
the primary goal of this flight was to measure the polarization of the cmb .
the receiver uses polarization sensitive bolometers at 145 ghz .
polarizing grids provide polarization sensitivity at 245 and 345 ghz .
we describe the boomerang telescope noting changes made for 2003 ldb flight , and discuss some of the issues involved in the measurement of polarization with bolometers .
lastly , we report on the 2003 flight and provide an estimate of the expected results . | the 1998 flight of boomerang ( b98 ) provided a measurement of the angular power spectrum of cmb temperature anisotropies from @xmath0 to @xmath1 @xcite .
boomerang made its second long duration balloon ( ldb ) flight ( boom03 ) in january 2003 with a receiver configured to simultaneously measure cmb temperature and polarization anisotropies .
the new receiver used pairs of polarization sensitive bolometers ( psb s ) at 145 ghz . at 245 ghz and 345 ghz ,
spider web bolometers are used with polarization sensitivity provided by polarizing grids mounted at the front of the cryogenic feed horns .
the b98 results were partially limited by pointing reconstruction error ( @xmath2 rms ) . for the boom03 flight , we added a pointed sun sensor and a tracking star camera which should reduce our pointing reconstruction error to less than @xmath3 rms . from the boom03 flight , we have 11.7 days of data .
an analysis effort is underway , with the goal of producing measurements of @xmath4 , @xmath5 and @xmath6 .
the boomerang telescope ( figure [ fig : tele ] ) was designed specifically for the harsh conditions of antarctic ballooning @xcite .
the extensive shielding prevents the contamination by stray - light , and protects other sensitive components from the sun .
this is especially important for daytime ballooning over antarctica , where thermal management is vital .
the back of the telescope is constantly illuminated by the sun , reaching temperatures of @xmath7 c , while the front is shaded and can cool to less than @xmath8 c. boomerang scans in azimuth , with the elevation kept constant for at least one hour .
sky rotation turns this one - dimensional azimuthal scan strategy into a cross - linked pattern on the celestial sphere .
b98 and boom03 both used a differential gps array , a fixed sun sensor and rate gyros for pointing reconstruction .
for boom03 , the pointed sun sensor and tracking star camera should improve the pointing reconstruction error .
the boomerang primary mirror is an off - axis parabola with a diameter of 1.3 m. it is 45@xmath9 degrees off axis and has a focal length of 1280 mm .
the primary feeds a pair of cold re - imaging mirrors which are kept at 1.65 k inside the cryostat ( figure [ fig : optics ] ) .
the tertiary forms an image of the primary and acts as the lyot stop for our system .
it controls the illumination on the primary , limiting the effective diameter to about 80 cm .
there is a 1 cm hole in the tertiary behind which sits a calibration lamp .
it fires once every 15 minutes allowing us to monitor any calibration drift .
the boomerang cryostat can keep the detectors at 0.275 k for 15 days @xcite .
toroidal nitrogen and helium tanks ( with the helium tank inside the nitrogen tank ) are suspended via 1.6 mm kevlar ropes .
the insert , which contains the cold mirrors and the focal plane , is bolted to the helium tank .
the insert also contains a single stage @xmath10he refrigerator which can keep the focal plane at 0.275 k for 12 - 13 days @xcite .
since the @xmath11he and ln@xmath12 stages have a longer hold time ( 18 and 16 days respectively ) than the @xmath10he stage , we added capability for an in - flight cycle of the @xmath10he refrigerator .
the boom03 receiver was designed for measurement of cmb temperature and polarization anisotropies .
there are 8 pixels , and each pixel contains two detectors .
four of the pixels contain pairs of polarization sensitive bolometers ( psb s ) operating at 145 ghz .
the other four pixels are 2-color photometers operating at 245 and 345 ghz .
both the psb s and the photometers use corrugated feed horns .
table [ tab : rec ] summarizes the properties of the boom03 receiver .
figure [ fig : focalplane ] shows the layout of the focal plane projected onto the sky .
.summary of the properties of the boom03 receiver .
the fwhm is based on a pre - flight measurement using a tethered thermal source placed 1 km from the telescope .
the @xmath13 is calculated using a model of our detectors based on pre - flight loadcurve data .
preliminary analysis of flight data provides a similar answer at 145 ghz [ tab : rec ] [ cols="^,^,^,^,^",options="header " , ]
with our flight data in hand , we can project our sensitivity to @xmath4 , @xmath5 , and @xmath6 .
figure [ fig : b2k2_forecast ] shows what we can expect statistically with eight detectors at 145 ghz .
these error bars are estimated using the formulas presented in @xcite .
the combined net assumes that all channels have an net@xmath14 .
no allotment is made for calibration uncertainty , beam uncertainty , pointing error , systematics or the fact that long time constants at 145 ghz will decrease signal - to - noise at high-@xmath15 .
the boomerang project has been supported by nasa , nsf - opp and nersc in the u.s .
, by pnra , universit `` la sapienza '' and asi in italy , by pparc in the uk , and by ciar and nserc in canada . t.m . acknowledges support from a nasa gsrp fellowship . the authors would like to thank the national scientific balloon facility ( nsbf ) and the u.s . antarctic program for excellent field and flight support
. w. c. jones , et al .
`` a polarization sensitive bolometric receiver for observations of the cosmic microwave background '' , in proceedings of spie vol . * 4855 * _ millimeter and submillimeter detectors for astronomy _ , edited by t.g .
phillips , j. zmuidzinas , ( spie , bellingham , wa , 2003 ) page 227 . | 1998 flight boomerang b98 angular power spectrum cmb temperature anisotropies @xmath0 to @xmath1 @xcite second long duration balloon flight boom03 january 2003 receiver cmb temperature polarization anisotropies new receiver used polarization sensitive bolometers at 145 ghz 245 ghz 345 ghz spider web bolometers grids front cryogenic feed horns b98 results limited pointing reconstruction error @xmath2 rms ) boom03 flight added pointed sun sensor tracking star camera reduce reconstruction error less than @xmath3 rms boom03 11.7 days data analysis effort underway goal measurements @xmath4 @xmath5 @xmath6 boomerang telescope designed for harsh conditions antarctic ballooning @xcite shielding prevents contamination protects sensitive components sun ballooning back telescope illuminated sun @xmath7 c front shaded less than @xmath8 c boomerang scans azimuth elevation constant one hour sky rotation turns scan cross - linked pattern celestial sphere b98 boom03 used differential gps array fixed sun sensor rate gyros for pointing reconstructionboom03 pointed sun sensor star camera improve reconstruction error boomerang primary mirror off - axis parabola diameter 1.3 m 45@xmath9 degrees off axis focal length 1280 mm primary feeds cold re - imaging mirrors kept at 1.65 k inside cryostat tertiary image primary lyot stop controls illumination primary diameter to 80 cm 1 cm hole in tertiary calibration lamp fires every 15 minutes calibration drift boomerang cryostat detectors at 0.275 k for 15 days toroidal nitrogen helium tanks suspended via 1.6 mm kevlar ropes insert cold mirrors focal plane bolted to helium tank single stage @xmath10he refrigerator focal plane at 0.275 k for 12 - 13 days @xmath11he@xmath12 stages longer hold time 18 16 days added capability in - flight cycle refrigerator boom03 receiver for cmb temperature polarization anisotropies 8 pixels two detectors four polarization sensitive bolometers at 145 ghz other four 2-color photometers at 245 345 ghz photometers use corrugated feed horns table summarizes properties boom03 receiverfigure focalplane shows layout focal plane projected sky properties boom03 receiver fwhm based on pre - flight measurement tethered thermal source 1 km from telescope @xmath13 calculated model detectors pre - flight loadcurve data preliminary analysis provides similar answer at 145 ghz flight data project to @xmath4 @xmath6 figure b2k2_forecast shows eight detectors at 145 ghz error bars estimated formulas @xcite combined net assumes all channels have net@xmath14 no allotment for calibration uncertainty beam uncertainty pointing error systematics long time constants at 145 ghz decrease signal - to - noise high-@xmath15 boomerang project supported by nasa nsf - opp nersc pnra universit asi pparc ciar nserc canada acknowledges support nasa gsrp fellowship authors thank national scientific balloon facility nsbf u.s . antarctic program support jones polarization sensitive bolometric receiver for observations cosmic microwave background proceedings spie 4855 millimeter detectors astronomy edited t phillipszmuidzinas bellingham 2003 page 227 | 0.259298 | 1,335 | 678 | , Saving $0.0 in GPT-4. |
154,700 | we present simulation results addressing the phenomena of colloidal gelation induced by attractive interactions .
the liquid - gas transition is prevented by the glass arrest at high enough attraction strength , resulting in a colloidal gel .
the dynamics of the system is controlled by the glass , with little effect of the liquid - gas transition .
when the system separates in a liquid and vapor phases , even if the denser phase enters the non - ergodic region , the vapor phase enables the structural relaxation of the system as a whole . | in colloids with short range attractions , the system `` gels '' at high enough attraction strength . this colloidal gel is a solid , with a low colloid density , stabilized by the bonds between particles .
usually , gelation occurs above crystallization and liquid - gas transition , i.e. increasing the attraction strength , the system enters the fluid - crystal coexistence region , at higher strengths it separates into a liquid and a gas , and finally it gels @xcite . in some cases ,
gels are found outside the liquid - gas separation boundary @xcite , although the latter can not be identified and a clear independence between gelation and phase separation can not be claimed .
thus , it has been argued that gels are indeed states with arrested phase separation @xcite . the mechanism for arrest can be either the dense phase crossing a glass transition ( either attraction or repulsion driven ) @xcite , or a glass transition driven by the particle bonding that prevents the phase separation . in this work
we have simulated a system that mimics the mixture of a colloid with a non - adsorbing polymer ( which induces attraction between the colloids ) , in three states beyond the liquid - gas transition .
the two states with stronger attractions indeed show arrested phase separation , and the system has formed a percolating network of particles with voids and tunnels .
the dynamics of the system is studied in the three states , and we find that the system with arrested phase separation show many properties typical of glass aging .
the system with the lowest attraction strength separates into a dense and a dilute phase , but the liquid phase is non - ergodic , i.e. it has reached the glass region at higher density .
the dynamics of the system with a long range repulsive barrier , which suppresses the liquid - gas separation , has been studied previously @xcite , what allows us to conclude here that gels are indeed caused by a quench to the attractive glass region .
quenches to lower attraction strength result in phase separation the liquid may undergo a glass transition and become non - ergodic , but the system does not end up with the typical gel properties .
newtonian dynamics simulations were run for a system comprised by 1000 quasi hard particles with an attractive interaction mimicking the effective interaction between colloids in colloid - polymer mixtures due to polymer depletion .
the system is polydisperse with radii distributed according to a flat distribution of width 10% of the average radius , @xmath0 .
the core - core repulsion is given by @xmath1 , where @xmath2 is the thermal energy and @xmath3 .
the attractive interaction is given by the simple depletion model developed by asakura and oosawa @xcite , corrected to consider a polydisperse system @xcite ; the attraction strength is given by the polymer volume fraction , @xmath4 , and the range by the size of the polymers @xmath5 .
further details of the total interaction potential can be found in previous works @xcite .
the colloidal density is reported as volume fraction , @xmath6 , and the attraction strength in units of @xmath4 ; dimensionless units are used by setting the mean radius @xmath7 , the thermal velocity @xmath8 and the mass @xmath9 . for the attraction range and volume fraction used in this work , @xmath10 and @xmath11 ,
respectively , the system undergoes liquid - gas separation at @xmath12 ( the crystallization transition occurs in monodisperse systems at lower @xmath4 ) .
the system is equilibrated without attraction and instantaneously quenched to three different states : @xmath13 , @xmath14 and @xmath15 .
previous works where the liquid - gas transition is inhibited by means of a long range repulsive barrier showed that there is an attractive glass transition at @xmath16 @xcite .
thus , the first state is below the glass point , whereas the other two are above .
the structure and dynamics are studied below as a function of the time elapsed since the quench , termed _
waiting time_. phase separation order parameter , @xmath17 , for different states along the isochore @xmath11 .
the system is divided in @xmath18 boxes , and @xmath19 , where @xmath20 is the density in every box , and the summation runs over all the boxes . the black circles mark the states studied in detail , @xmath13 , @xmath14 and @xmath15 . the vertical dashed line shows the state where the glass transition was found without liquid - gas separation.,scaledwidth=60.0% ] structure factor for the three states as labeled and at the waiting times shown.,scaledwidth=60.0% ]
fig . [ psi4 ] shows the evolution of the density inhomogeneity in the system as a function of the attraction strength , @xmath4 , along @xmath11 .
homogeneous fluids are obtained at low @xmath4 , and liquid - gas separation at @xmath21 .
however , instead of observing increasing inhomogeneity of the system with @xmath4 , due to denser liquids and more dilute vapors , we note that the phase separation is impeded by an additional mechanism .
we show here that this mechanism is the glass transition already studied in the same system without phase separation ( vertical dashed line in the figure ) .
the inhibition of liquid - vapor separation is also noticeable in the evolution of the structure factor , shown in fig .
whereas the state @xmath13 shows the typical behaviour of spinodal decomposition , with a peak at low wavevectors that grows and moves to lower @xmath22 , the states at higher attraction strength show a more homogeneous structure .
only a peak at low @xmath22 is observed , which grows much slower than at @xmath13 , and which is smaller the higher @xmath4 .
a similar peak at low @xmath22 is obtained in the system without phase transition , due to local compaction of the system @xcite .
the arrested phase separation scenario is fully in agreement with fig .
[ psi4 ] and with previous findings in simulations and experiments ; the resulting gels are heterogeneous locally and show arrested dynamics @xcite .
incoherent density correlation functions for the three states and at different waiting times , as labeled .
the right hand panels show the incoherent non - ergodicity parameter , @xmath23 , as a function of @xmath22 ( the dashed line is the gaussian approximation with a localization length equal to the attraction range ) and the structural relaxation time , @xmath24 as a function of the waiting time ( @xmath24 is defined as @xmath25 , for @xmath26).,scaledwidth=85.0% ] the dynamics of the system is studied in fig .
[ fsqt ] by means of the incoherent part of the density correlation function , @xmath27 , where the summation runs over the @xmath28 particles in the system and @xmath29 .
the density correlation functions in the left panels show the typical features of a glass transition , i.e. a two step decay , where the second , structural , relaxation increases with waiting time .
the height of the intermediate plateau , the non - ergodicity parameter , is presented in the right upper - most panel as a function of @xmath22 , and the evolution of the relaxation time with waiting time is given in the lower panel .
the state with @xmath13 shows the fastest relaxations and the lowest non - ergodicity parameters ( yielding a localization length longer than the attraction range ) , but more importantly , the time scale for structural relaxation saturates . on the other hand , the dynamics of the states for higher @xmath4 do not saturate and
@xmath24 follows a power law with waiting time , with exponents larger than one , and the localization length is smaller than the interaction range .
similar behaviour was observed in aging of the attractive glass obtained in this system without phase separation @xcite .
left panels : bond correlation functions for the three states and waiting times labeled .
right panels : correlation between the squared displacement between @xmath30 and @xmath31 and the average number of neighbours in this time interval for @xmath32 . ,
scaledwidth=85.0% ] the dynamical arrest is caused by the bonds between particles due to the attraction , as observed by the non - ergodicity parameters .
it is thus interesting to study the bond correlation function , i.e. the fraction of bonds that have existed uninterruptedly since @xmath33 until @xmath34 , fig .
[ bonds ] ( two particles are bonded when their separation is smaller than the attraction range ) . in agreement with the dynamics studied in fig . [ fsqt ] , the bonds are stronger and live longer for high @xmath4 .
note that even at the highest @xmath4 studied , @xmath15 , the bonds are still reversible , and @xmath35 does not show any plateau .
the correlation between the displacement of a given particle and its mean number of bonds ( neighbours ) , shown in the right - hand panels , indicates that particles with a small number of neighbours ( in average ) move much longer than average , but this population of mobile particles is only apparent in the state @xmath13 .
these particles , thus , comprise the vapor phase which is in coexistence with the liquid phase , with much lower mobility .
the overall relaxation of the system observed in the density correlation function for @xmath13 is therefore caused by the vapour phase , due to the exchange between particles in both phases , although the liquid one can be itself inside the non - ergodic region ( most probably in the attractive glass , according to the non - ergodicity parameter ) .
a population of _ fast _ particles was also observed in the system without phase separation and aided the structural relaxation below the glass transition .
the states @xmath14 and @xmath15 , on the other hand , are quenched above the glass transition ( found in the system without phase separation @xcite ) and the liquid - gas transition is inhibited . here , no vapor phase is present , and the system can not relax structurally , also in agreement with previous findings in the system without phase separation ( there , the population of fast particles vanished in the glass states @xcite ) .
thus , the dynamics of the system is controlled by the glass transition , with only small effects due to the phase separation .
our results show , therefore , that systems undergoing phase separation , where the liquid phase enters the non - ergodic region , may still relax and appear ergodic due to the vapor phase . in order to observe
really arrested dynamics and phase separation , the system must be quenched beyond the glass transition , to prevent the formation of the gas phase .
we can not say , however , if the crossover from the `` apparent ergodic '' regime to the `` arrested '' one , is abrupt or continuous , although our results of static properties point to the latter ( fig .
[ psi4 ] ) .
in addition , the glass transition is found with the aid of local compaction of the system , both in the system with and without phase transition , which is also found in other studies of the attractive glass @xcite .
our results , however , can not elucidate whether this compaction is a real necessity for the glass transition or only favors it .
financial support is acknowledged from the m.e.c .
projects mat2003 - 03051-co3 - 01 and ha2004 - 0022 ( a.m.p . ) .
we thank f. sciortino and e. zaccarelli for useful discussions . | colloids short range attractions system gels at high attraction strength colloidal gel solid low colloid density stabilized by bonds between particles gelation occurs above crystallization liquid - gas transition increasing attraction strength system enters fluid - crystal coexistence region higher strengths separates into liquid gas finally gels @xcite some gels found outside liquid - gas separation boundary independence between gelation phase separation argued gels are states arrested phase separation mechanism arrest dense phase crossing glass transition or particle bonding simulated system colloid non - adsorbing polymer induces attraction in three states beyond liquid - gas transition two states stronger attractions show arrested phase separation system percolating network of particles with voids tunnels dynamics studied three states arrested phase separation properties typical glass aging system lowest attraction strength separates into dense dilute phase liquid phase non - ergodic reached glass region at higher density dynamics system with long range repulsive barrier suppresses liquid - gas separation studied previously gels caused by quench to glass regionquenches lower attraction strength phase separation liquid glass transition non - ergodic system typical gel properties newtonian dynamics simulations run for system 1000 quasi hard particles attractive interaction mimicking colloids - polymer mixtures polymer depletion system polydisperse radii flat distribution width 10% average radius core - core repulsion by @xmath1 @xmath2 thermal energy @xmath3 attractive interaction by simple depletion model asakura oosawa attraction strength by polymer volume fraction range by size of polymers @xmath5 details total interaction potential in previous works colloidal density as volume fraction @xmath6 attraction strength in units @xmath4 dimensionless units mean radius @xmath7 thermal velocity mass @xmath9 attraction range volume fraction system liquid - gas separation at @xmath12 crystallization transition at lower @xmath4 system equilibrated without attraction quenched to three states @xmath13 @xmath15 previous works liquid gas transition inhibited repulsive barrier attractive glass transition at @xmath16 first state below glass point other two abovestructure dynamics studied function time since quench waiting time phase separation order parameter @xmath17 different states isochore @xmath11 system divided in @xmath18 @xmath19 @xmath20 density in every box summation over boxes black circles mark states studied @xmath13 vertical dashed line shows state glass transition without liquid - gas separation,scaledwidth=60.0% structure factor for three states waiting times fig . psi4 shows evolution density inhomogeneity function attraction strength @xmath4 @xmath11 homogeneous fluids at low @xmath4 liquid - gas separation at @xmath21 increasing inhomogeneity with @xmath4 denser liquids dilute vapors phase separation impeded by additional mechanism glass transition studied without phase separation inhibition liquid - vapor separation noticeable in evolution structure factor fig state @xmath13 shows spinodal decomposition peak at low lower @xmath22 states higher attraction strength show more homogeneous structure peak at low @xmath22 grows slower than smaller higher @xmath4 peak @xmath22 obtained without phase transition due to local compaction arrested phase separation scenario with fig . psi4 previous findings resulting gels heterogeneous show arrested dynamics @xcite incoherent density correlation functions three states different waiting times panels show incoherent non - ergodicity parameter @xmath23 function of @xmath22 localization length equal attraction range structural relaxation time @xmath24 function waiting time defined @xmath25 dynamics studied in fig . [ fsqt incoherent density correlation function @xmath27 summation runs over @xmath28 particles @xmath29 density correlation functions show features glass transition two step decay second relaxation increases with waiting time height intermediate plateau non - ergodicity parameter function of @xmath22 evolution relaxation time with waiting time lower panel state with @xmath13 shows fastest relaxations lowest non - ergodicity parameters localization length longer than attraction range time scale for structural relaxation saturatesdynamics states higher @xmath4 saturate @xmath24 follows power law with waiting time exponents larger than one localization length smaller than interaction range similar behaviour observed in aging attractive glass system without phase separation left bond correlation for three states waiting times right correlation between squared displacement between @xmath30 and @xmath31 average number of neighbours time interval @xmath32 dynamical arrest caused by bonds between particles due to attraction non - ergodicity parameters study bond correlation function fraction of bonds existed uninterruptedly since @xmath33 until @xmath34 separation smaller than attraction range bonds stronger live longer for high @xmath4 at highest @xmath4 @xmath15 bonds reversible @xmath35 plateau correlation between displacement particle mean number of bonds indicates particles with small number of neighbours move longer population mobile particles apparent in state @xmath13 particles comprise vapor phase with liquid phase lower mobilityrelaxation system in density correlation function for @xmath13 caused by vapour phase due to exchange between particles phases liquid inside non - ergodic region probably attractive glass population of fast particles observed in system without phase separation aided structural relaxation below glass transition states @xmath14 and @xmath15 quenched above glass transition liquid - gas transition inhibited no vapor phase present system can not relax structurally agreement with previous findings without phase separation fast particles vanished in glass states dynamics controlled by glass transition small effects due to phase separation results show systems undergoing phase separation liquid phase enters non - ergodic region may relax appear ergodic due to vapor phase to observe arrested dynamics phase separation system must be quenched beyond glass transition to prevent formation gas phase if crossover from apparent ergodic regime to arrested abrupt or continuous results static properties point to latter glass transition found with local compaction system and results can not elucidate compaction necessity for glass transition or favors financial support acknowledged from m . projects mat2003 - 03051-co3 - 01 and ha2004 - 0022 (thank f. sciortino e. zaccarelli discussions | 0.432882 | 2,468 | 1,214 | , Saving $0.1 in GPT-4. |
154,702 | we present an xmm newton observation of the nearby dwarf galaxy ic 2574 .
this galaxy is part of our ensemble observed within a multi frequency campaign to study the physical conditions of the interstellar medium in dwarf galaxies in general . here
, we focus on the diffuse soft ( @xmath0 ) x ray emission associated with ic 2574 . since this emission
is expected to be quite faint we have to reach the true x ray background intensity level . for this aim
, we have to account for unrelated diffuse x ray emission and for the vignetting of the x
ray mirrors aboard the xmm newton satellite .
# 1_#1 _ # 1_#1 _ = # 1 1.25 in .125 in .25 in | according to the standard model the interstellar medium of dwarf galaxies is thought to be strongly shaped by stellar evolution .
supernovae and/or stellar winds produce holes and bubbles in the interstellar medium of their host galaxies .
the linear extent of these structures reach @xmath1 sizes . to illustrate this , fig .
1 shows the distribution of ic 2574 measured by the vla .
it is very likely that hot gas produced by stellar activity is expelled into the halo of the galaxy because the gravitational potential of low mass dwarf galaxies is very shallow .
therefore , we expect to find hot coronal gas in the vicinity of the galaxy body . in the case of ic 2574
these hypothetical plasma structures are expected on a tens of arcmin scale .
henceforward , the xmm
newton satellite with its 30 fov and angular resolution of 14 ( half energy width ) is the ideal instrument to detect coronal gas in dwarf galaxies .
the diffuse soft x ray emission is a very faint signal in comparison to the contribution of x ray point sources .
therefore , the reduction of the x ray data is a crucial point . the epic instruments aboard xmm
newton are affected by proton flares originating from the sun .
these proton flares show up as peaks in the lightcurves ( total counts vs. exposure time ) of the epic data , producing an artificial count rate across the whole fov , which is by a factor of up to @xmath2 higher than the true x ray signal .
it is essential to flag the time intervals affected by proton flares . for this aim
, we developed an algorithm which rejects the flares on a user definable @xmath3level .
we compute the mean ( @xmath4 ) of the lightcurve and flag the photon counts which exceed a certain threshold ( depending on the chosen @xmath3level ) .
then the new mean ( @xmath5 ) is calculated which is smaller than @xmath4 and the photons exceeding the adapted @xmath3level are flagged as well .
the iteration stops when @xmath6 , i.e. the statistical uncertainty of the data .
the flagged time intervals are excluded from further analysis .
this is the first major step in preparing the xmm
newton epic data for the investigation of diffuse soft x ray emission .
like all imaging x ray telescopes the xmm
newton epic instruments show the effect of _ vignetting_. this is the reduction of effective area which depends on the photon energy and off axis angle .
the effect is clearly visible as a gradual decrease of sensitivity towards the edges of the fov .
it is possible to correct for this effect by calculating a so called _ exposure map_. using the sas5.3.3 expmap task results in exposure maps which are inconsistent with the observational data .
this is exemplified in fig .
2 . the sas exposure maps lead to an over correction of the data even in the center region of the fov . in order to correct for the vignetting we produce
improved exposure maps for the epic mos instruments using five deep observations ( @xmath7 total ) obtained from the xmm
newton science archive . combining the individual deep field data is sufficient to sample the full fov and moreover , it minimizes the contribution of point sources .
this approach is more confident than the expmap task because we make use of the xmm
newton inflight performance .
after the calibration of the deep field data we remove all point sources down to a flux limit of @xmath8 ( in the energy range @xmath0 ) using the eboxdetect and emldetect tasks . these yield so
called _ cheese images _ which show holes at the point source positions . since the sas calculated cut out radii are too small , we developed software which calculates the correct radii and fills up the holes with a statistical photon noise map including the intensity normalization for each hole . after combining ( i.e. adding ) all hole filled images of the multiple deep pointings for each energy band , we smooth these merged exposure maps and multiply by the detector mask .
the resulting exposure map is displayed in fig .
2 ( right ) .
the flare filter algorithm and the improved exposure maps are applied to the calibrated xmm
newton data of ic 2574 . after filtering the lightcurves and dividing the images by the exposure maps ,
both mos1 and mos2 are combined to achieve a better s / n .
the brightest point sources down to @xmath9 are removed and the image is adaptively smoothed envoking csmooth of the ciao2.2.1 software package . the resulting image in the energy range of @xmath0 is shown in fig .
3 . with this technique
the detection limits of the ic 2574 observation for point sources and diffuse emission are @xmath10 and @xmath11 respectively .
it turns out that diffuse soft x ray emission emerges from the galaxy .
since the emission is visible in a very soft energy window ( @xmath0 ) it is likely that the emission originates from in front of the distribution
. otherwise it would not be detectable due to the photoelectric absorption of the .
however , a better s / n can be achieved when the epic pn instrument is included . | standard model interstellar medium of dwarf galaxies shaped by stellar evolution supernovae stellar winds produce holes bubbles in medium linear structures reach @xmath1 sizes fig . 1 shows distribution of ic 2574 measured vla likely hot gas stellar activity expelled into halo galaxy gravitational potential low mass dwarf galaxies shallow expect hot coronal gas in vicinity of galaxy body ic 2574 plasma structures expected on tens of arcmin scale xmm newton satellite 30 fov angular resolution 14 ideal instrument detect coronal gas in dwarf galaxies diffuse soft x ray emission faint signal x ray sources reduction of x ray data crucial instruments affected by proton flares from sun flares show as peaks in lightcurves artificial count rate to @xmath2 higher than true x ray signal essential to flag time intervals affected by proton flares developed algorithm rejects flares on user definable @xmath3level compute mean @xmath4 ) lightcurve flag photon counts threshold new mean @xmath5 ) calculated smaller than @xmath4 photons exceeding @xmath3level flaggediteration stops when @xmath6 statistical uncertainty flagged time intervals excluded from analysis first step preparing xmm newton epic data for diffuse soft x ray emission xmm newton epic instruments show vignetting_ reduction of effective area depends on photon energy off axis angle visible as decrease sensitivity towards edges by calculating exposure map sas5.3.3 expmap task results in exposure maps inconsistent with data exemplified in fig . 2 exposure maps lead to over correction correct vignetting produce improved exposure maps for instruments using five deep observations @xmath7 total from newton science archive combining deep field data sample full fov minimizes contribution point sources approach more confident than expmap task xmm newton inflight performance after calibration remove point sources to flux limit @xmath8 @xmath0 using eboxdetect emldetect tasks yield cheese images holes at point source positions calculated cut out radii small developed software calculates correct radii fills up holes with statistical photon noise map intensity normalization after combining hole filled images smooth merged exposure maps multiply by detector maskexposure map fig . 2 flare filter algorithm improved exposure maps applied to xmm newton data ic 2574 after filtering lightcurves dividing images mos1 mos2 combined better s / n brightest point sources to @xmath9 removed image smoothed csmooth ciao2.2.1 software resulting image energy range @xmath0 fig . 3 detection limits ic 2574 for point sources diffuse emission are @xmath10 @xmath11 diffuse soft x ray emission emerges from galaxy emission visible soft energy window @xmath0 likely originates from distribution detectable photoelectric absorption better s / n epic pn instrument included | 0.186805 | 1,124 | 539 | , Saving $0.0 in GPT-4. |
154,726 | we show a purely electronic cooling scheme to cool a charged mechanical resonator ( mr ) down to nearly the vibrational ground state by elaborately tuning bias gate voltages on the electrodes , which couple the mr by coulomb interaction . the key step is the modification of time - dependent effective eigen - frequency of the mr based on the lewis - riesenfeld invariant . with respect to a relevant idea proposed previously [ li _ et al_. , phys . rev .
a * 83 * , 043803 ( 2011 ) ] , our scheme is simpler , more practical and completely within the reach of current technology . micro- and nano - mechanical resonators ( mrs ) have attracted much research interest due to their combination of both classical and quantum properties @xcite together with broad impact on fundamental researches and applications @xcite , such as ultra - sensitive measurements approaching quantum limit @xcite , observation of continuous - variable entanglement with mesoscopic objects pra-84 - 024301,prl-101 - 200503 , quantum information processing prl-105 - 220501 , and biological sensing @xcite .
the prerequisite of the research in these aspects is to cool the mrs down to their ground states in order to suppress detrimental influence from thermal fluctuations . up to now , there have been many proposals for cooling the mrs @xcite in either optomechanical or electromechanical systems , in which the resolved sideband cooling method enables the vibrational ground - state cooling of mrs .
so far , the ground - state cooling of mrs has been achieved experimentally nature-464 - 697 , nature-478 - 89 , natphys-5 - 485 , nature-463 - 72 .
very recently , a different approach to fast ground - state cooling of mrs has been proposed @xcite with time - dependent optical driving in a three - mirror cavity optomechanical system , in which the effective frequency of the mr can be changed like an ` optical spring ' and the mr can be cooled down to nearly its ground- state under the control of time - dependent external optical driving fields .
but the scheme seems challenging experimentally due to the requirement for both adiabatic evolution under born - oppenheimer approximation and very strong optical powers .
couples to two identical electrodes via coulomb interaction .
the two electrodes are distant by @xmath0 and @xmath1 is the deviation of the mr from the equilibrium position .
@xmath2 is the capacitance of gate , and @xmath3 is the tunable bias gate voltage.,width=226 ] we present in this work an alternative method using a purely electronic way , which is simpler but more practical , for cooling a charged mr by two electrodes via coulomb interaction . under the control of bias gate voltages
, the mr behaves as a single - mode harmonic oscillator with its effective frequency tunable like an ` electrical spring ' .
our method is somewhat similar to that given in ref .
@xcite but without involvement of born - oppenheimer approximation .
moreover , in contrast to ref .
@xcite , our idea is more feasible using current experimental techniques due to manipulation of bias gate voltages . furthermore , different from the traditional cooling schemes in electromechanical systems based on superconductor qubits @xcite or microwave photons prb-78 - 134301,prb-76 - 205302 , our scheme can cool a mr down to its vibrational ground state by simply tuning the bias gate voltages . to the best of our knowledge ,
this is the first cooling scheme with purely electronic way to cool the mr system without any assistance from auxiliary qubits or additional photons . as schematically shown in fig .
[ fig b1 ] , we consider a system where a charged mr is placed in the middle of two identical electrodes , and couples electrostatically to these two electrodes .
each electrode takes the charge @xmath4 with @xmath2 and @xmath3 being the capacitance and voltage of the bias gate , respectively .
the coulomb force between the mr and each electrode varies with the bias gate voltage , and the coulomb potential is written by @xmath5where @xmath6 is the equilibrium separation between the mr center - of - mass position and the electrodes with @xmath1 the deviation of the mr from the equilibrium position , @xmath7 ( @xmath8 ) is the coulomb constant in vacuum with @xmath9 being the vacuum dielectric constant , and @xmath10 is the charge of the mr . after defining @xmath11 with the dimensionless time - dependent factor @xmath12
satisfying @xmath13 , we may reduce eq .
( [ c1 - 1 ] ) , under the condition of @xmath14 , to @xmath15where the term @xmath16 has been ignored since it commutes with both the position and the momentum operators of the mr , and makes no contribution to following discussions .
supposing the charged mr in the absence of the electrodes takes the hamiltonian @xmath17where @xmath1 and @xmath18 are the position and momentum operators of the mr with the bare eigen - frequency @xmath19 and the effective mass @xmath20 , we have the modified hamiltonian under the control of the bias gate as @xmath21where the effective frequency is @xmath22 with @xmath23
. generally speaking , the dynamics of a harmonic oscillator ( e.g. mr ) governed by eq .
( [ c1 - 3 ] ) can be obtained by the method of lewis - riesenfeld invariants @xcite .
particularly , for some special intermediate trajectories , the instantaneous quantum state of the harmonic oscillator at initial time instant can be the same as that at the final time instant , although the effective eigen - frequency of the harmonic oscillator has been changed significantly @xcite .
that is to say , we may have the final mean phonon number of the harmonic oscillator to be the same as the initial one , but with significant change of the effective temperature .
this idea holds for arbitrary initial state , and has been used to cool the spatial motion of atoms @xcite and to cool the mr in cavity optomechanical system @xcite .
our present idea is to achieve cooling of the mr by modifying time - dependent eigen - frequency via tuning the bias gate voltages on external electrodes .
we note that the effective frequency @xmath24 of the mr can be easily changed by tuning the bias gate voltages @xmath3 .
in fact , @xmath25 can be much smaller than @xmath26 under negative @xmath3 .
moreover , @xmath24 can be even imaginary in order to accelerate the cooling process @xcite . specifically , considering some experimentally achievable parameters in charged mr systems @xcite , such as @xmath27n@xmath28m@xmath29c@xmath30 , @xmath31 khz , @xmath32 pg , @xmath33 @xmath34 m , @xmath35 v , @xmath36 , @xmath37/cm@xmath30 , @xmath39 nf , and @xmath40 @xmath34m@xmath30 , we have @xmath41 , meeting the approximate condition in eq .
( vc - approx1 ) .
similar to ref .
@xcite , our protocol for cooling mr comprises following two steps . _ _ step i : decreasing the mean phonon number with the increase of the mr effective frequency__in the absence of bias gate voltages ( i.e. , @xmath42 ) , we assume the mr with eigen - frequency @xmath43 initially in a thermal equilibrium state @xmath44 at temperature @xmath45 mk .
so the corresponding mean thermal phonon number is given by @xmath46 - 1\right\}\simeq 3100.\ ] ] with an arbitrary trajectory of @xmath47 [ or @xmath48 , the mr is assumed to be in a new thermal state @xmath49,\ ] ] at a later time @xmath50 with @xmath51 , where the effective frequency is @xmath52 and the mean thermal phonon number is @xmath53 - 1}% \simeq 0.47.\ ] ] it is clear that the temperature of the mr at the time @xmath50 remains unchanged compared with the initial one , while the mean thermal phonon number of the mr is much less than the one at the initial time @xmath54 due to enlargement of the effective frequency @xmath24 .
the fact @xmath55 means that the mr with enlarged effective frequency has been cooled down to nearly the vibrational ground state .
_ _ step ii : decreasing the mr effective frequency by keeping the low mean phonon number__this step is accomplished via a special trajectory of @xmath56 evolving to the final time @xmath57 under the control of bias gate voltages satisfying two conditions : i ) the final effective frequency is equal to the bare one ( i.e. , @xmath58 ) , implying that the bias gate voltages are absent again at the final time @xmath57 ( i.e. , @xmath59 ) ; ii ) during the cooling process , the mean phonon number remains unchanged compared with that at time @xmath50 ( i.e. , @xmath60 ) .
consequently , we have the mr with the bare frequency cooled to nearly the vibrational ground state .
the solution to such a special trajectory of @xmath56 between @xmath50 and @xmath61 governed by eq .
( [ c1 - 3 ] ) follows the lewis - riesenfeld invariant @xcite . using the inverse - invariant method @xcite in appendix
, we obtain the control function of the trajectory of @xmath56 by tracing the lewis - riesenfeld invariant , @xmath62 where the dimensionless function @xmath63 is defined in appendix .
the trajectory of @xmath56 , starting from @xmath64 and ending with @xmath65 , corresponds to the effective frequency of the mr decreasing from the end frequency in the first step ( @xmath66 ) to the final frequency @xmath67 .
we plot in fig .
[ fig b2 ] the control function @xmath56 , the corresponding instantaneous eigen - frequency @xmath68 and the effective temperature @xmath69 .
using eq .
( 5 ) , we know that our cooling enables the mr from the initial temperature @xmath70 mk down to the final temperature @xmath71 @xmath34k , as plotted in fig .
[ fig b2](c ) . here
the effective temperature is defined through @xmath72 - 1\}$ ] . to estimate the imperfection in the trajectory of @xmath56 , which corresponds to the experimental error in tunable bias gate voltages on electrodes , we give the trajectory of @xmath56 a ten percent fluctuation , i.e. , @xmath73 , as in fig .
[ fig b3](a ) , and simulate in fig . [ fig b3](b ) and ( c ) the changes of the instantaneous eigen - frequency @xmath74 of the mr and the corresponding effective temperature of the mr .
we find that , although the experimental deviation is as large as @xmath75 , the effective eigen - frequency can still reach @xmath76 for @xmath77 ( @xmath78 for @xmath79 ) deviation with the corresponding effective temperature 7 @xmath34k for @xmath77 ( 5 @xmath34k for @xmath79 ) deviation .
in other words , the mr in such cases can still be cooled down to nearly its vibrational ground state .
moreover , we note that the final effective temperature 5 @xmath80k for @xmath79 deviation ( see fig . [ fig b3](c ) ) seems lower than 6 @xmath80k under the ideal condition . but
this is the case of fluctuation , which is uncontrollable .
in addition , in this deviating case , the effective frequency is @xmath81 , rather than the desired effective frequency @xmath26 . therefore , our choice of trajectory in fig .
[ fig b3](b ) is definitely optimal .
we would like to point out that our scheme is only for an instantaneous cooling , which is different from the usual sideband cooling scheme achievable in steady state .
this is because the mr is decoupled from electrodes at the end of our cooling process .
so the mr can not be kept in low temperature for very long time but be heated again by environment . however , after the cooling , the mr would not return to the bath temperature suddenly due to its high q factor ( @xmath82 ) @xcite which makes sure the small decay of the mr . as a result
, there should be a long enough time to finish the scheduled experimental works , such as quantum information processing @xcite before the mr is heated to the bath temperature .
even if the mr is thermalized to the bath temperature , we may repeat the same cooling process to put the mr down to the vibrational ground state again .
therefore , both the cooling process and the experimental work should be done alternately in the implementation of the scheduled experimental works .
this is similar to that in ion trap system , wherein one carries out the sideband cooling and the operations of quantum algorithms alternately @xcite . in summary , we have proposed a practical protocol for cooling a mr near to its vibrational ground state by controlling bias gate voltages on two nearby electrodes . using the achievable experimental parameters and considering experimental imperfection , we have shown the feasibility of the proposal by currently available technology . as a result ,
our work is not only a practical cooling scheme , but also very promising to verify the lewis - resisenfeld invariant method experimentally . | we present below how to obtain the special trajectory of @xmath56 between @xmath83(@xmath84 ) and @xmath85 by means of the lewis - riesenfeld invariant .
based on the inverse - invariant method @xcite , eq .
( [ c1 - 3 ] ) satisfies the time - dependent invariant @xcite , @xmath86 , \label{c1 - 4}\ ] ] where @xmath87 for simplicity , @xmath88 takes the role of the momentum conjugate to @xmath89 , and the dimensionless real function @xmath63 follows @xmath90 according to the boundary conditions at time @xmath50 and the finial time @xmath61 , we have @xmath91 with @xmath92 .
so the simplest polynomial @xmath63 between @xmath50 ( @xmath84 ) and @xmath57 is given by @xmath93 with @xmath94 .
the work is supported by the national fundamental research program of china ( grant nos .
2012cb922102 , 2012cb922104 , 2009cb929604 and 2007cb925204 ) and the national natural science foundation of china ( grant nos . 10974225 , 60978009 and 11174027 ) . | present trajectory @xmath56 between lewis - riesenfeld invariant inverse - invariant method @xcite c1 - 3 satisfies time - dependent invariant @xcite @xmath86 - 4 @xmath87 @xmath88 momentum conjugate @xmath89 dimensionless function @xmath63 follows @xmath90 boundary conditions time @xmath50 finial time @xmath61 @xmath91 @xmath92 simplest polynomial @xmath63 between @xmath50 @xmath57 @xmath93 @xmath94 supported national fundamental research program china grant 2012cb922102 2009cb929604 2007cb925204 national natural science foundation china grant 10974225 11174027 | 0.070631 | 278 | 152 | , Saving $0.0 in GPT-4. |
154,739 | this paper addresses the problem of summarizing decisions in spoken meetings : our goal is to produce a concise _ decision abstract _ for each meeting decision .
we explore and compare token - level and dialogue act - level automatic summarization methods using both unsupervised and supervised learning frameworks . in the supervised summarization setting , and given true clusterings of decision - related utterances , we find that token - level summaries that employ discourse context can approach an upper bound for decision abstracts derived directly from dialogue acts . in the unsupervised summarization setting , we find that summaries based on unsupervised partitioning of decision - related utterances perform comparably to those based on partitions generated using supervised techniques ( 0.22 rouge - f1 using lda - based topic models vs. 0.23 using svms ) . | meetings are a common way for people to share information and discuss problems .
and an effective meeting always leads to concrete decisions . as a result
, it would be useful to develop automatic methods that summarize not the entire meeting dialogue , but just the important decisions made .
in particular , decision summaries would allow participants to review decisions from previous meetings as they prepare for an upcoming meeting .
for those who did not participate in the earlier meetings , decision summaries might provide one type of efficient overview of the meeting contents . for managers , decision summaries could act as a concise record of the idea generation process .
while there has been some previous work in summarizing meetings and conversations , very little work has focused on decision summarization : and investigate the use of a semantic parser and machine learning methods for phrase- and token - level decision summarization .
we believe our work is the first to explore and compare token - level and dialogue act - level approaches using both unsupervised and supervised learning methods for summarizing decisions in meetings .
.a clip of a meeting from the ami meeting corpus @xcite .
a , b , c and d refer to distinct speakers ; the numbers in parentheses indicate the associated meeting decision : decision 1 , 2 , 3 or 4
. also shown is the gold - standard ( manual ) abstract ( summary ) for each decision .
[ cols= " < " , ] 0.9 0.9 [ [ dialogue - act - based - summarization . ] ] dialogue act - based summarization .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + previous research ( e.g. , , , ) has shown that drda - level extractive summarization can be effective when viewed as a binary classification task . to implement this approach , we assume that the drda to be extracted for the summary is the one with the largest vocabulary overlap with the cluster s gold - standard decision abstract .
this da - level summarization method has an advantage that the summary maintains good readability without a natural language generation component .
[ [ token - based - summarization . ] ] token - based summarization .
+ + + + + + + + + + + + + + + + + + + + + + + + + + as shown in table 1 , some decision - related das contain many useless words when compared with the gold - standard abstracts . as a result
, we propose a method for token - level decision summarization that focuses on identifying critical keywords from the cluster s drdas .
we follow the method of , but use a larger set of features and different learning methods . [ [ adding - discourse - context . ] ] adding discourse context .
+ + + + + + + + + + + + + + + + + + + + + + + + + for each of the supervised da- and token - based summarization methods , we also investigate the role of the discourse context .
specifically , we augment the drda clusterings with additional ( not decision - related ) das from the meeting dialogue : for each decision partition , we include the da with the highest tf - idf similarity with the centroid of the partition
. we will investigate the possible effects of this additional context on summary quality . in the next subsection
, we describe the features used for supervised learning of da- and token - based decision summaries .
different from text , dialogues have some notable features that we expect to be useful for finding informative , decision - related utterances .
this section describes some of the dialogue - based features employed in our classifiers .
the full lists of features are shown in table 4 and table 5 . 0.9 [ [ structural - information - adjacency - pairs . ] ] structural information : adjacency pairs .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + an _ adjacency pair _
( ap ) is an important conversational analysis concept ; aps are considered the fundamental unit of conversational organization @xcite . in the ami corpus ,
an ap pair consists of a source utterance and a target utterance , produced by different speakers .
the source precedes the target but they are not necessarily adjacent .
we include features to indicate whether or not two das are aps indicating question+answer or positive feedback . for these features
, we use the gold - standard ap annotations .
we also include one feature that checks membership in a small set of words to decide whether a da contains positive feedback ( e.g. ,
yeah " , yes " ) .
[ [ discourse - information - review - and - closing - indicator . ] ] discourse information : review and closing indicator .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + another pragmatic cue for dialogue discussion is terms like wrap up " or recap " , indicating that speakers will review the key meeting content .
we include the distance between these indicators and das as a feature .
[ [ grammatical - information - dependency - relation - between - words . ] ] grammatical information : dependency relation between words .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + for token - level summarization , we make use of the grammatical relationships in the das . as in and , we design features that encode ( a ) basic predicate - argument structures involving major phrase types ( s , vp , np , and pp ) and ( b ) additional typed dependencies from .
we use the stanford parser .
experiments based on supervised learning are performed using 3-fold cross - validation .
we train two different types of classifiers for identifying informative das or tokens : conditional random fields ( crfs ) ( via mallet ) and support vector machines ( svms ) ( via @xmath0 ) .
we remove function words from das before using them as the input of our systems .
the ami decision abstracts are the gold - standard summaries .
we use the rouge @xcite evaluation measure .
rouge is a recall - based method that can identify systems producing succinct and descriptive summaries .
[ [ results - and - analysis . ] ] results and analysis .
+ + + + + + + + + + + + + + + + + + + + + results for the unsupervised and supervised summarization methods are shown in tables 6 and 7 , respectively . in the tables ,
true clusterings means that we apply our methods on the gold - standard drda clusterings .
system clusterings use clusterings obtained from the methods introduced in section 4 ; we show results only using the best unsupervised ( using lda ) and supervised ( using svms ) drda clustering techniques .
both table 6 and 7 show that some attempt to cluster drdas improves the summarization results vs. no clustering . in table 6
, there is no significant difference between the results obtained from the longest da and prototype da for any experiment setting .
this is because the longest da is often selected as the prototype .
an upper bound result is listed for comparison : for each decision cluster , this system selects all words from the drdas that are part of the decision abstract ( discarding duplicates ) .
table 7 presents the results for supervised summarization .
rows starting with da or token indicate results at the da- or token - level .
the + context rows show results when discourse context is included .
we see that : ( 1 ) svms have a superior or comparable summarization performance vs. crfs on every task . ( 2 ) token - level summaries perform better than da - level summaries only using true clusterings and the svm - based summarizer .
( 3 ) discourse context generally improves token - level summaries but not da - level summaries .
( 4 ) drda clusterings produced by ( unsupervised ) lda lead to summaries that are quite comparable in quality to those generated from drda clusterings produced by svms ( supervised ) . from table 6 , we see that f1 is 0.2214 when choosing longest das from lda - generated clusterings , which is comparable with the f1s of 0.1935 and 0.2349 , attained when employing crf and svms on the same clusterings . the results in table 7 are achieved by comparing abstracts having function words with system - generated summaries without function words . to reduce the vocabulary difference as much as possible , we also ran experiments that remove function words from the gold - standard abstracts , but no significant difference is observed . finally , we considered comparing our systems to the earlier similar work of @xcite and @xcite , but found that it would be quite difficult because they employ a different notion from drdas which is decision dialogue acts(ddas ) .
in addition , they manually annotate words from their ddas as the gold - standard summary , guaranteeing that their decision summaries employ the same vocabulary as the ddas .
we instead use the actual decision abstracts from the ami corpus . here
we show sample summaries produced using our methods ( table 8) .
we pick one of the clusterings generated by lda consisting of four das which support two decisions and take svms as the supervised summarization method .
we remove function words and special markers like [ disfmarker ] " from the das .
the outputs indicate that either the longest da or prototype da contains part of the decisions in this
mixed " cluster .
adding discourse context refines the summaries at both the da- and token - levels .
in this work , we explore methods for producing decision summaries from spoken meetings at both the da - level and the token - level .
we show that clustering drdas before identifying informative content to extract can improve summarization quality .
we also find that unsupervised clustering of drdas ( using lda - based topic models ) can produce summaries of comparable quality to those generated from supervised drda clustering .
token - level summarization methods can be boosted by adding discourse context and outperform da - level summarization when true drda clusterings are available ; otherwise , da - level summarization methods offer better performance .
satanjeev banerjee , carolyn penstein ros , and alexander i. rudnicky .
the necessity of a meeting recording and playback system , and the benefit of topic - level annotations to meeting browsing . in _ interact _ , pages 643656 .
trung h. bui , matthew frampton , john dowding , and stanley peters .
2009 . extracting decisions from multi - party dialogue
using directed graphical models and semantic similarity . in _ proceedings of the sigdial 2009 conference _ , pages 235243 .
anne hendrik buist , wessel kraaij , and stephan raaijmakers
automatic summarization of meeting data : a feasibility study . in _ in proc .
meeting of computational linguistics in the netherlands ( clin_. jean carletta , simone ashby , sebastien bourban , mike flynn , thomas hain , jaroslav kadlec , vasilis karaiskos , wessel kraaij , melissa kronenthal , guillaume lathoud , mike lincoln , agnes lisowska , and mccowan wilfried post dennis reidsma . 2005 .
the ami meeting corpus : a pre - announcement .
in _ in proc . mlmi _ ,
pages 2839 .
freddy y. y. choi . 2000 .
advances in domain independent linear text segmentation . in _ proceedings of the 1st north american chapter of the association for computational linguistics conference _ , pages 2633 .
raquel fernndez , matthew frampton , john dowding , anish adukuzhiyil , patrick ehlen , and stanley peters .
identifying relevant phrases to summarize decisions in spoken meetings .
interspeech-2008 , pages 7881 .
raquel fernndez , matthew frampton , patrick ehlen , matthew purver , and stanley peters .
2008b . modelling and detecting decisions in multi - party dialogue
. in _ proceedings of the 9th sigdial workshop on discourse and dialogue _
, pages 156163 .
matthew frampton , jia huang , trung huu bui , and stanley peters .
real - time decision detection in multi - party dialogue . in _ proceedings of the 2009 conference on empirical methods in natural language processing
: volume 3 - volume 3 _ , pages 11331141 . michel galley .
2006 . a skip - chain conditional random field for ranking meeting utterances by importance . in _ proceedings of the 2006 conference on empirical methods in natural language processing _ , pages 364372 .
pei - yun hsueh and johanna d. moore .
2008 . automatic decision detection in meeting speech . in _ proceedings of
the 4th international conference on machine learning for multimodal interaction _ ,
pages 168179 .
chin - yew lin and eduard hovy .
automatic evaluation of summaries using n - gram co - occurrence statistics . in _ proceedings of the 2003 conference of the north american chapter of the association for computational linguistics on human language technology - volume 1 _ , pages 7178 .
gabriel murray , steve renals , and jean carletta . 2005 .
extractive summarization of meeting recordings .
in _ in proceedings of the 9th european conference on speech communication and technology _ , pages 593596 . | meetings common share information discuss problems effective meeting leads to concrete decisions useful to develop automatic methods entire meeting dialogue important decisions decision summaries allow participants review decisions previous meetings for upcoming meeting for decision summaries provide efficient overview of meeting contents for managers decision summaries concise record of idea generation process previous work summarizing meetings little work focused on decision summarization semantic parser machine learning for phrase- token - level decision summarization our work first to explore compare token - level dialogue act - level approaches using unsupervised supervised learning methods for summarizing decisions meetings clip of meeting from ami meeting corpus @xcite a b c d refer to distinct speakers numbers indicate associated meeting decision decision 1 2 3 4 shown gold - standard manual ) abstract summary ) for each decision dialogue - act - based - summarization previous research drda - level extractive summarization effective as binary classification task approach assume drda extracted for summary with largest vocabulary overlap with cluster s gold - standard decision abstract da - level summarization method maintains good readability without natural language generation component token - based - summarization token - based summarization . + in table 1 decision - related das contain useless words compared with gold - standard abstracts propose method for token - level decision summarization identifying critical keywords from cluster s drdas follow method use larger set of features different learning methods adding - discourse - context . discourse context for supervised da- token - based summarization methods investigate role of discourse context augment drda clusterings with additional not decision - related das from meeting dialogue each decision include da with highest tf - idf similarity with centroid partition investigate effects of additional context on summary quality next describe features for supervised learning of da- and token - based decision summaries different from text dialogues have notable features useful for finding informative decision - related utterances describes dialogue - based features in classifiers full lists of features shown in table 4 and table 5 . structural - information - adjacency - pairs structural information : adjacency pairs . adjacency pair ap ) important conversational analysis concept fundamental unit of conversational organization ap pair consists of source utterance and target utterance produced by different speakerssource precedes target not necessarily adjacent include features indicate two das are aps indicating question+answer or positive feedback use gold - standard ap annotations include one feature checks membership in small set of words decide da contains positive feedback ( e.g. yeah " , yes " ) discourse - information - review - and - closing - indicator discourse information : review and closing indicator pragmatic cue for dialogue discussion is terms like wrap up " or recap " speakers review key meeting content include distance between indicators and das as feature grammatical - information - dependency - relation - between - words . grammatical information : dependency relation between words for token - level summarization use grammatical relationships in das design features encode basic predicate - argument structures major phrase types and additional typed dependencies use stanford parser experiments based on supervised learning performed using 3-fold cross - validation train two types of classifiers for identifying informative das or tokens : conditional random fields ( crfs ) and support vector machines ( svms ) remove function words from das before using input systems ami decision abstracts are gold - standard summariesuse rouge @xcite evaluation measure rouge recall - based method systems producing succinct descriptive summaries results - analysis results for unsupervised supervised summarization methods shown in tables 6 7 true clusterings methods on gold - standard drda clusterings system clusterings use clusterings from methods section 4 results using best unsupervised supervised svms drda clustering techniques cluster drdas improves summarization results vs no clustering no significant difference between results longest da prototype da longest da selected as prototype upper bound result listed for comparison decision cluster system selects words from drdas decision abstract duplicates table 7 results for supervised summarization rows starting with da token indicate results da- token - level + context rows show results discourse context included svms superior summarization performance vs crfs task token - level summaries perform better than da - level using true clusterings svm - based summarizer discourse context improves token - level summaries not da - level drda clusterings unsupervised lda lead to summaries comparable quality to drda clusterings svms supervisedtable 6 f1 0.2214 choosing longest das from lda - clusterings comparable with f1s 0.1935 0.2349 crf svms same clusterings results table 7 comparing abstracts function words with summaries without function words reduce difference experiments function words from gold - standard abstracts no significant difference considered comparing systems to @xcite @xcite difficult employ different notion from drdas decision dialogue acts(ddas manually annotate words ddas gold - standard summary same vocabulary use decision abstracts from ami corpus sample summaries methods table 8) pick lda four das two decisions take svms supervised summarization method remove function words special markers from das outputs longest da or prototype da contains decisions mixed cluster discourse context refines summaries da- token - levels methods producing decision summaries from spoken meetings da - token - level clustering drdas before identifying informative content summarization quality unsupervised clustering drdas lda - models summaries comparable quality to supervised drda clustering token - level summarization methods adding discourse context outperform da - level summarization when true drda clusterings available da - level summarization better performancesatanjeev banerjee carolyn penstein ros alexander rudnicky necessity meeting recording playback system topic - level annotations interact pages 643656 trung bui matthew frampton john dowding stanley peters 2009 extracting decisions multi - party dialogue graphical models semantic similarity proceedings sigdial 2009 conference pages 235243 hendrik buist wessel kraaij stephan automatic summarization meeting data feasibility study computational linguistics netherlands jean carletta ashby sebastien bourban flynn karaiskos kronenthal guillaume lathoud lincoln agnes lisowska wilfried dennis reidsma 2005 meeting corpus pre - announcement pages 2839 freddy choi 2000 domain independent linear text segmentation proceedings 1st north american association computational linguistics conference pages 2633 raquel fernndez matthew frampton john dowding patrick ehlen stanley peters identifying phrases summarize decisions spoken meetings interspeech-2008 pages 7881 peters modelling detecting decisions multi - party dialogueproceedings 9th workshop discourse dialogue 156163 matthew frampton jia huang trung huu bui stanley peters real - time decision detection multi - party dialogue 2009 conference language processing 3 11331141 michel galley 2006 skip - chain random field ranking utterances proceedings 2006 conference 364372 pei - yun hsueh johanna moore 2008 automatic decision detection meeting speech 4th international conference machine learning multimodal interaction 168179 chin - yew lin eduard hovy automatic evaluation summaries n - gram co - occurrence statistics proceedings 2003 conference language technology 7178 gabriel murray steve renals jean carletta 2005 extractive summarization meeting recordings 9th european conference speech communication technology 593596 | 0.444532 | 2,994 | 1,447 | , Saving $0.1 in GPT-4. |
154,762 | using density functional theory , we determine parameters of tight - binding hamiltonians for a variety of fabre charge transfer salts , focusing in particular on the effects of temperature and pressure . besides relying on previously published crystal structures , we experimentally determine two new sets of structures ; ( tmttf)@xmath0sbf@xmath1 at different temperatures and ( tmttf)@xmath0pf@xmath1 at various pressures .
we find that a few trends in the electronic behavior can be connected to the complex phase diagram shown by these materials .
decreasing temperature and increasing pressure cause the systems to become more two - dimensional .
we analyze the importance of correlations by considering an extended hubbard model parameterized using wannier orbital overlaps and show that while charge order is strongly activated by the inter - site coulomb interaction , the magnetic order is only weakly enhanced .
both orders are suppressed when the effective pressure is increased . | quasi - one dimensional organic salts formed from tetramethyltetrathiafulvalene ( tmttf ) molecules - also known as fabre charge - transfer ( ct ) salts - have been intensively investigated in the last two decades since they exhibit a rich variety of phases like antiferromagnetism , superconductivity , charge ordering , spin - density wave ordering or spin - peierls behavior .
@xcite such phases can be driven both by external ( physical ) pressure as well as by chemical pressure ( see fig . [
fig : phasediagram ] ) . even though a few successful models have been proposed for the description of these systems ,
constructing a consistent microscopic picture of the relationships between the various phases remains a challenge.@xcite clo@xmath2 is not well known . in the phase
diagram its approximate position has been indicated with a dashed arrow .
the possible phases are charge ordered ( co ) , mott insulating ( mi ) , antiferromagnetic ( af ) , spin peierls ( sp ) , spin density wave ( sdw ) , superconducting ( sc ) and 1d , 2d or 3d metal . ]
the primary avenue of the present work is to understand the microscopic origin of the close competition between the different phases in these compounds as a function of chemical and external pressure as well as temperature .
for that , we performed _
ab initio _ density functional theory ( dft ) @xcite and model hamiltonian calculations for several fabre ct salts whose crystal structures were determined at different temperatures and pressure and investigated variations of their electronic properties with temperature and pressure.[multiblock footnote omitted ] by computing the real - space overlaps of wannier orbitals for the bands near the fermi level , we parametrize a two - band tight - binding hamiltonian model for the various systems and examine the differences in their electronic hopping parameters . in this way , we can connect structural and chemical modifications with changes in the electronic properties .
furthermore , in order to analyze some of the preferred orderings we consider a description of the fabre ct salts in terms of an extended hubbard hamiltonian including on - site and inter - site coulomb interaction terms .
the kinetic part of this model is given by the computed hopping parameters for the various compounds .
we discuss spin - spin and charge - charge correlation properties by diagonalizing the model .
the work is organized as follows .
sections [ secii ] and [ seciii ] are dedicated to the description of the computational details as well as the crystal structure of the fabre ct salts . in sections
[ seciv ] , [ secv ] and [ secvi ] we present , respectively , our results on the electronic structure , tight - binding models as well as extended hubbard models for a few members of fabre ct salts .
discussion and conclusions are given in sections [ secvii ] and [ secviii ] .
the electronic structure calculations presented here were performed in an all - electron full - potential local orbital basis using the fplo package.@xcite the densities were converged on an @xmath3 @xmath4 mesh , using a generalized gradient approximation ( gga ) functional.@xcite for materials without published hydrogen coordinates , hydrogen atoms were placed according to the expected bond lengths and angles of a methyl group ( c h distance 1.1 , c c h angle 109@xmath5 ) . with the bond length and angle fixed , one has the freedom to choose the rotation angle of the set of hydrogen atoms on each methyl group around the c c bond .
we chose this angle such that one hydrogen is as far out of the plane of the molecule as possible .
we tested the effect of this choice on the band structure and found no contribution to the bands of interest near the fermi level since in this energy region only tmttf bands of @xmath6 origin are involved .
some of the structures had suspicious bond lengths and angles , therefore the atomic coordinates were relaxed with dft using the vienna _ ab initio _ simulation package ( vasp , version 5.2.11 ) , @xcite with a projector - augmented wave basis .
@xcite we used the gga functional,@xcite and included van der waals corrections @xcite for the relaxations .
we performed two kinds of relaxations ; in one relaxation we kept sulfur and the heavy anion atoms coordinates fixed , and in the other relaxation all atom positions were relaxed .
the differences between the two relaxed structures were minimal .
the atomic coordinates were converged to an energy difference of 1 mev on a @xmath7 @xmath4-mesh , with a plane wave cutoff energy of 500 ev .
the tight - binding parameters were obtained by constructing wannier orbitals for the tmttf bands at the fermi level and computing real - space overlaps , as implemented in fplo .
another way to generate these parameters is to fit the band structure of the model hamiltonian to the dft bands .
the latter method can become difficult when many hopping parameters need to be fitted ; there can be a number of solutions which reproduce the dft bands equally well , but differ in physical details ( such as relative strengths of certain bonds ) . by using wannier orbital overlaps we can be sure that our parameter values have a clear physical interpretation .
the exact diagonalization of the extended hubbard model was performed by considering system sizes of @xmath8 tmttf sites with periodic boundary conditions .
the fabre ct salts consist of alternate layers of tmttf molecules ( cations ) and monovalent anions , stacked in the @xmath9 direction ( see fig . [
fig : tmttfcrystal ] ) . in between the cation layers ,
the planar tmttf molecules form @xmath6-stacked one dimensional chains in the @xmath10 direction with a slight ` zig - zag ' arrangement .
there is a charge transfer of one electron from each ( tmttf)@xmath0 dimer to each anion , _
i.e. _ the tmttf molecules carry half a hole on average .
there are two classes of anions : those that conform to the @xmath11 symmetry of the tmttf part of the crystal ( such as pf@xmath1 ) , and those that break that symmetry ( such as clo@xmath2 ) . also , the anion species influence the proportions of the unit cell , as well as the intra- and inter - chain hopping strengths .
the inter - chain hopping strengths are not only determined by the distance between the tmttf molecules , but also by changes in their zig - zag arrangement ( that is to say , how far away from a perfectly aligned stack they are , and in what direction ) . plane .
the organic molecules form @xmath6-stacked 1d chains along the crystal @xmath10 direction ( with a slight zig - zag pattern ) , and form layers parallel to the @xmath12 plane .
these organic layers alternate with anion layers ( with the anions centered on the pink as sites ) stacked in the @xmath9 direction .
grey atoms are carbon , yellow are sulfur , while hydrogen atoms are shown in white . ]
tmttf molecules within a chain show a slight dimerization along the chain .
we can quantify this structural dimerization as the difference between the larger dimerization distance of adjacent tmttf molecules , @xmath13 , and the shorter dimerization distance , @xmath14 , normalized by the sum of the two distances : @xmath15 these distances are defined as the distances between the centres of mass of the c and s atoms in each tmttf molecule .
table [ tab : properties ] shows the structural dimerization of the materials investigated in this work .
this table also includes the electronic dimerization , which is introduced in sec .
v. in general , the structural dimerization increases slightly with increasing temperature .
cooling from room temperature to t = 4 k , ( tmttf)@xmath0asf@xmath1 and ( tmttf)@xmath0pf@xmath1 both show charge ordering phase transitions and spin - peierls transitions .
these ordering transitions are not visible in the crystal structures ; there are no significant changes in the structural dimerization from t= 300 k to t= 4 k. ( tmttf)@xmath0sbf@xmath1 shows similar changes in the structural dimerization between 100 k and room temperature , and does not go through any ordering transitions in this range .
.structural and electronic dimerization of the fabre ct salts considered in the present work .
the structural dimerization of the tmttf molecules is defined in eq . , and the electronic dimerization is defined in eq . in sec
. v. references marked with * have no ( or unrealistic ) published hydrogen coordinates .
note that clo@xmath2 and bf@xmath2 are tetrahedral anions , and so do not conform to the reported @xmath11 symmetry ( the anions do not have the required inversion symmetry ) .
the anion ordering transition only occurs in these systems where the anion does not have inversion symmetry .
the change in sign of @xmath16 indicates that the shorter bond has the smaller @xmath17 . [ cols="<,^,^,^",options="header " , ]
44 natexlab#1#1bibnamefont # 1#1bibfnamefont # 1#1citenamefont # 1#1url # 1`#1`urlprefix[2]#2 [ 2][]#2 | - one dimensional organic salts from tetramethyltetrathiafulvalene molecules known as fabre charge - transfer salts intensively investigated two decades exhibit phases like antiferromagnetism superconductivity charge ordering spin - density wave spin - peierls behavior phases driven by external chemical pressure fig phasediagram successful models proposed constructing consistent microscopic picture relationships between phases remains challenge clo@xmath2 not known phase diagram position indicated with dashed arrow possible phases are charge ordered insulating antiferromagnetic spin peierls spin density wave superconducting sc 1d 2d 3d metal primary avenue work understand microscopic origin of competition between phases compounds chemical external pressure temperature performed density functional theory model hamiltonian calculations for fabre ct salts structures at different temperatures pressure investigated variations electronic properties with temperature pressure computing real - space overlaps of wannier orbitals for bands near fermi level two - band tight - binding hamiltonian model for systems differences in electronic hopping parameters connect structural chemical modifications with changes electronic propertiesanalyze preferred orderings description fabre ct salts extended hubbard hamiltonian on - inter - site coulomb interaction terms kinetic part model given computed hopping parameters compounds discuss spin - charge - charge correlation diagonalizing model work organized sections secii seciii computational details crystal structure fabre ct salts sections seciv secv secvi ] results electronic structure tight - binding models extended hubbard models fabre ct salts discussion conclusions in sections secvii secviii electronic structure calculations performed all - electron full - potential local orbital basis fplo package densities converged on @xmath3 @xmath4 mesh generalized gradient approximation materials without hydrogen coordinates hydrogen atoms placed expected bond lengths angles methyl group c h distance 1.1 angle 109@xmath5 bond length angle fixed freedom choose rotation angle hydrogen atoms each methyl group c bond chose angle one hydrogen far out of plane molecule tested effect band structure no contribution to bands near fermi level tmttf bands @xmath6 origin involvedstructures had suspicious bond lengths angles atomic coordinates relaxed with dft using vienna initio simulation package version 5.2.11 projector - augmented wave basis used gga functional van der waals corrections for relaxations performed two relaxations one kept sulfur heavy anion atoms coordinates fixed all atom positions relaxed differences between structures minimal atomic coordinates converged to energy difference 1 mev on @xmath7 @xmath4-mesh plane wave cutoff energy 500 ev tight - binding parameters obtained by constructing wannier orbitals for tmttf bands computing real - space overlaps band structure model hamiltonian to dft bands difficult when parameters solutions reproduce dft bands differ in physical details wannier orbital overlaps parameter values clear physical interpretation exact diagonalization of extended hubbard model system sizes of @xmath8 tmttf sites with periodic boundary conditions fabre ct salts consist of layers tmttf molecules monovalent anions stacked in @xmath9 direction planar tmttf molecules form-stacked chains in @xmath10 direction zig - zag arrangementcharge transfer of one electron from each tmttf)@xmath0 dimer to each anion tmttf molecules carry half a hole average two classes of anions conform to @xmath11 symmetry tmttf pf@xmath1 symmetry clo@xmath2 ) anion species influence proportions unit cell intra- inter - chain hopping strengths strengths determined by distance between tmttf molecules changes zig - zag arrangement organic molecules form @xmath6-stacked 1d chains along crystal @xmath10 direction slight zig - zag pattern form layers parallel to @xmath12 plane organic layers alternate with anion layers anions centered on pink sites in @xmath9 direction grey atoms are carbon yellow sulfur hydrogen in white tmttf molecules within chain show slight dimerization along chain dimerization as difference between larger distance and shorter @xmath14 normalized by sum of two distances distances defined as between centres of mass of c and s atoms in each molecule table shows structural dimerization of materials investigated includes electronic dimerization structural dimerization increases slightly with temperaturecooling room temperature to t = 4 k@xmath0asf@xmath1 and show charge ordering phase spin - peierls transitions not visible in crystal structures no significant changes in structural dimerization from t= 300 k to t 4 k@xmath0sbf@xmath1 shows similar changes between 100 k and room temperature ordering transitions electronic dimerization of fabre ct salts considered work structural dimerization tmttf molecules defined in eq . electronic dimerization references marked with * have no published hydrogen coordinates clo@xmath2 bf@xmath2 are tetrahedral anions conform to @xmath11 symmetry anion ordering transition occurs in systems where inversion symmetry change in sign of @xmath16 indicates shorter bond has smaller @xmath17 natexlab#1#1bibnamefont | 0.318867 | 2,046 | 999 | , Saving $0.1 in GPT-4. |
154,765 | we have obtained measurements of the total cross section for @xmath0 annihilation into hadronic final states for 6 energy points ( 2.6 , 3.2 , 2.4 , 3.55 , 4.6 , and 5.0 gev ) with the upgraded beijing spectrometer ( besii ) .
we report preliminary values from this data and outline future plans for a finer scan in the 2 - 5 gev energy range . | two topics that depend on measurements in the center of mass ( cm ) energy range of a few gev are of current interest .
one is the precision of the qed coupling constant evaluated at the mass of the @xmath1 boson , @xmath2 , and the other is the anomalous magnetic moment of the muon , @xmath3 .
both are accessible via a measurement of the lowest order cross - section for @xmath4 hadrons , which is usually parameterized in terms of the ratio @xmath5 , defined as @xmath6 where the denominator is the lowest - order qed cross section and equals @xmath7 .
the @xmath5 ratio has been measured by many experiments over a large cm energy range from the hadron production threshold to the @xmath8 mass @xcite .
experimentally determined @xmath5 values are , in general , consistent with theoretical predictions , and provide an impressive confirmation of the hypothesis of three color degrees of freedom for quarks .
the current values for r below 5 gev have experimental uncertainties on the order of 15%@xcite .
these uncertainties limit the precision of @xmath9 and @xmath3 and in turn the determination of the higgs mass from radiative corrections in the standard model@xcite .
for example , the contributions to the value of @xmath9 and the error in @xmath9 are shown in figures [ fig : alphval ] and [ fig : alpherr ] .
the bes collaboration has started a program of r measurements over the 2 - 5 gev energy range with the goal of reducing the present uncertainties by a factor of two or more .
= 2.5 in = 2.5 in
the measurements were carried out with the upgraded bes detector ( besii ) , which is a conventional solenoidal detector that is described in some detail in ref .
upgrades included the replacement of the central drift chamber with a vertex chamber ( vc ) composed of 12 tracking layers organized around a beryllium beam pipe .
this chamber provides a spatial resolution of about 90 @xmath10 m .
the barrel time - of - flight counter ( btof ) was replaced with a new array of 48 plastic scintillators that are read out by fine mesh photomultiplier tubes situated in the 0.40 t magnetic field volume , providing 180 ps resolution .
a new main drift chamber ( mdc ) has 10 superlayers , each with four sublayers of sense wires .
it provides @xmath11 information for particle identification and has a momentum resolution of @xmath12 for the charged tracks with momentum @xmath13 in gev .
the sampling - type barrel electromagnetic calorimeter ( bsc ) , which covers 80% of 4@xmath14 solid angle , consists of 560 cells along the @xmath15 direction , with each cell containing 24 layers in the radial direction .
the bsc operates in the self - quenching streamer mode with an energy resolution of @xmath16 ( e in gev ) and a spatial resolution of 7.9 mrad in @xmath15 and 3.6 cm in @xmath17 .
the outermost component of besii is a @xmath10 identification system consisting of three double layers of streamer tubes interspersed in the iron flux return of the magnet .
these measure coordinates along the muon trajectories with resolutions of 3 cm and 5.5 cm in @xmath18 and @xmath17 , respectively .
a first scan of @xmath5 measurements was performed this spring with data samples of about 1000 - 2000 hadronic events collected each at 2.6 , 3.2 , 3.4 , 3.55 , 4.6 and 5.0 gev . to understand beam - associated backgrounds ,
separated beam data were taken at each energy point , and single beam data were accumulated at 3.55 gev .
additionally , data were taken at the @xmath19 and @xmath20 resonances to provide a precise energy calibration as well as a cross check of the trigger efficiencies for the bhabha , dimuon and hadronic events measured in the r scan .
the 3.4 gev point was remeasured after a time interval of about two weeks to test the stability of the detector .
the value of @xmath5 is determined from the number of observed hadronic events ( @xmath21 ) by the relation @xmath22 where @xmath23 is the tree - level cross - section of the particular process , @xmath24 is the number of beam associated background events ; @xmath25 and @xmath26 are the numbers of misidentified lepton - pair and two - photon processes events ; @xmath27 is the integrated luminosity ; @xmath28 is the correction for initial state radiation ; and @xmath29 and @xmath30 represents the detection and trigger efficiency for the hadronic events . in the following sections , we examine each of these factors in detail .
a 1.5 m @xmath31 event sample was used for detector calibration and for monitoring the data quality and the stability of the detector performance .
these data indicated that the besii detector performance was stable and the data quality was good .
figure [ fig : stab ] shows the time dependence of mdc momentum resolution , bsc energy resolution , btof time resolution and @xmath11 pulse heights for bhabha events .
= 2.9 in the goal of hadronic event selection was to distinguish single - photon hadron production from other processes .
the following track - level selection criteria were used to define good charged tracks : * @xmath32 , where @xmath33 is the track polar angle ; * the track must have a reasonable three - dimensional helix fit ; * distance of closest approach to the beam in the transverse plane and along the beam axis are less than 2.0 and 15 cm , respectively ; * @xmath34 , where @xmath13 and @xmath35 are the momenta of the track and the beam , respectively , and @xmath36 is the momentum resolution of the beam ; * @xmath37 , where @xmath38 and @xmath39 are the energy of the track ( as measured in the bsc ) and beam , respectively , and @xmath13 is again the momentum of the track ; * a track must not be definitely identified as an electron via de / dx information ; * a track must not be definitely identified as a muon ; * @xmath40 ( in ns ) , where @xmath41 and @xmath42 are the time - of - flight for the track and a nominal time - of - flight calculated for the track assuming a proton hypothesis , respectively , and @xmath43 is the btof time resolution .
after the track - level selection , a further event level selection was applied via the following : * at least 2 good charged tracks , with at least one track having a good helix fit ; * a total deposited energy in the bsc @xmath44 .
a further selection scheme was required based on the number of good tracks in the event . for three or more prong events ,
the only additional requirement was that all the charged tracks not be positive ( to remove beam - gas events ) .
however , two - prong events needed to be distinguished from cosmic ray and lepton pair events , requiring a further selection scheme : * the two tracks must not have been back - to - back : + @xmath45 * at least two isolated neutral tracks fulfilled the following : + @xmath46 and @xmath47 are the @xmath33 and @xmath15 angle of either of the two charged tracks .
the performance of the event selection routines were checked by hand scans , which were carried out for both the selected and rejected hadronic events , the separated - beam data , and the monte carlo events . a typical hadronic event that passed the selection cuts
is shown in figure [ fig : event ] .
we determined the error associated with the hadronic selection criteria from changes in the event yield caused by varying the selection requirements and from the results of the hand scan .
= 2.5 in there were three major types of background to be considered .
one type , consisting of cosmic rays , bhabha , dimuon events and some two - photon process events , was directly selected out during the event selection routine .
the second , consisting of tau - pair production and residual two - photon processes , was subtracted out statistically via a monte carlo simulation . finally , the most serious sources of background in the hadronic event sample were beam - gas and beam - wall interactions .
most of the beam - gas and beam - wall background events were rejected by a vertex cut .
the salient features of the beam associated background were that their tracks were very much along the beam pipe direction , the energy deposited in bsc was small , and most of the tracks were protons .
the same hadronic event selection criteria were applied to the separated - beam data , and the number of separated - beam events @xmath48 surviving these criteria were obtained .
the number of the beam associated background events @xmath24 in the corresponding hadronic event sample was given by @xmath49 , where @xmath50 was determined by the ratio of the products of the pressure p at the collision region times the integrated beam currents i , i.e. @xmath51
. the integrated luminosity was determined using large - angle bhabha events with the following selection criteria , using only bsc information : * two clusters in the bsc with largest deposited energy in the polar angle @xmath52 ; * each cluster with energy @xmath53 gev ( for 3.55 gev data , scaled for other energy points ) ; * @xmath54 , where @xmath55 and @xmath56 are the azimuthal angles of the clusters to account for radiative events .
a cross check using only mdc information ( de / dx ) was generally consistent with the bsc measurement ; the difference was taken into account in the systematic error .
the detection efficiency for hadronic events was determined via a monte carlo simulation using the jetset7.4 event generator@xcite .
parameters in the jetset7.4 generator were tuned using more than 40,000 hadronic events selected from the tau mass measurement data sample@xcite .
the parameters of the generator were adjusted to reproduce distributions of kinematic variables such as multiplicity , sphericity , transverse momentum , etc .
figure [ fig : lund ] shows the charge multiplicity , the sphericity , the aplanarity and the transverse momentum distributions for the real and simulated event samples .
= 2.6 in a two - level trigger system retained good events while rejecting background @xcite .
the trigger was derived from signals from a beam pickup electrode located upstream of the detector and from each sub - detector .
event categories were classified according to numbers of charged and neutral tracks seen at the trigger level . for beam crossings with charged tracks ,
two trigger channels were utilized : in the first , we required at least one hit in the 48 btof counter array , one track in the vc and mdc , and at least 150 mev of energy deposited in the bsc ; in the second , we required back - to - back hits in the btof counter with one track in the vc and two tracks in the mdc .
for the neutral track trigger , we required that the sum of the deposited energy of the tracks in two cells of the bsc was greater than 80 mev and the total energy deposited in bsc from all sources was greater than 800 mev .
all coincidences were formed using 40 , 120 , 630 , 640 ns wide gates derived from the beam pickup signal for the tof , vc , mdc and bsc , respectively .
event candidates fulfilling the two - level criteria were accepted .
the trigger efficiencies were measured run by run by comparing the responses to different trigger requirements . as a cross check
, data taken in special runs made at the @xmath31 resonance was used to provide independent measurements of the trigger efficiencies .
table [ tab : trigeff ] lists the measured efficiencies for different event types in the r scan data , as well as the cross - check performed using @xmath19 data .
the errors in the trigger efficiencies for bhabha and hadronic events were less than @xmath570.5% .
.trigger efficiencies .
[ cols= " < , > , > , > " , ] [ tab : rsyst ]
the main goal of the first scan was a careful understanding of the trigger and hadronic event acceptances , as well as the hadronic event selection and the background subtraction , which are central to a total cross section measurement . in 1999
, bes - ii will carry out a finer scan from 2 to 5 gev as indicated by figure [ fig : future ] .
our goal is to reduce the present uncertainties of @xmath5 by at least a factor of two over this entire energy region .
= 2.5 in -.1 cm = 2.5 in -.1 cm
particle data group , eur .
j. * c3 * 204 ( 1998 ) .
f. ceradini et al . , phys . lett .
* b47 * ( 1973 ) 80 ; + b. batoli et al .
, phys . rev .
* d6 * ( 1972 ) 2374 ; + m. bernardini et al .
, phys . lett . * b51 * ( 1974 ) 200 .
g. cosme et al . , phys . lett .
* b40 * ( 1972 ) 685 .
m. kurdadze et al . , phys . lett .
* b42 * ( 1972 ) 515 .
a. litke et al . , phys .
lett . * 30 * ( 1973 ) 1189 .
a. blondel , plenary talk at ichep96 , warsaw .
b. pietrzyk , rad .
, cracow 1996 . m. davier and a. hoeker , lal 97 - 85 . h. burkhardt and b. pietrzyk , phys . lett . * b356 * ( 1995 ) 398 .
bai et al . , nucl .
instr . and methods in phys .
research * a344 * ( 1994 ) 319 .
torbjoen sjoestrand , lu tp 95 - 20 .
qi et al . , to be published on high energy physics and nuclear physics ( in chinese ) .
f.a . berends and r. kleiss , nucl . phys .
* b178 * ( 1981 ) 141 .
e. a. kuraev and v. s. fadin , sov
* 41 * ( 3 ) ( 1985 ) 3 .
g. bonneau and f. martin , nucl . phys .
* b27 * ( 1971 ) 387 . c. edwards et al .
, slac - pub 5160 ( 1990 ) | topics measurements center mass energy range few gev precision qed coupling constant at mass @xmath1 boson anomalous magnetic moment of muon @xmath3 accessible via measurement lowest order cross - section for @xmath4 hadrons parameterized ratio @xmath5 @xmath6 denominator lowest - order qed cross section equals @xmath7 ratio measured large energy range from hadron production threshold to @xmath8 mass @xcite @xmath5 values consistent with theoretical predictions hypothesis three color degrees of freedom for quarks current values r below 5 gev experimental uncertainties 15%@xcite uncertainties limit precision of @xmath9 @xmath3 determination higgs mass from radiative corrections standard model contributions to value @xmath9 error shown in figures alphval alpherr bes collaboration started program r measurements over 2 - 5 gev energy range reducing uncertainties by factor two or 2.5 measurements with upgraded bes detector conventional solenoidal upgrades included replacement central drift chamber with vertex chamber 12 tracking layers beryllium beam pipe chamber spatial resolution 90 @xmath10 m barrel time flight counter replaced with 48 plastic scintillators by fine mesh photomultiplier tubes 0.40 t magnetic field volume 180 ps resolution new main drift chamber has 10 superlayers four sublayers sense wires provides @xmath11 information particle identification momentum resolution @xmath12 charged tracks sampling barrel electromagnetic calorimeter covers 80% 4@xmath14 solid angle 560 cells @xmath15 direction 24 layers radial operates self - quenching streamer mode energy resolution @xmath16 spatial resolution 7.9 mrad in @xmath15 3.6 cm @xmath17 @xmath10 identification system three double layers streamer tubes iron flux return measure coordinates muon trajectories resolutions 3 cm 5.5 cm in @xmath18 @xmath17 first scan @xmath5 measurements spring 1000 - 2000 hadronic events at 2.6 3.2 3.4 3.55 4.6 5.0 gev separated beam data at each energy point single beam at 3.55 gev data at @xmath19 @xmath20 resonances precise energy calibration cross check trigger efficiencies3.4 gev point remeasured after two weeks stability detector value of @xmath5 determined from observed hadronic events by relation @xmath22 tree - level cross - section process beam background events misidentified lepton - pair two - photon processes events integrated luminosity correction for initial state radiation detection trigger efficiency hadronic events examine factors 1.5 m @xmath31 event sample used for detector calibration monitoring data quality stability data indicated detector performance stable data quality good figure shows time dependence of mdc momentum bsc energy resolution btof time resolution @xmath11 pulse heights for bhabha events goal hadronic event selection distinguish single - photon hadron production from processestrack - level selection criteria good charged tracks @xmath32 @xmath33 track polar angle reasonable three - dimensional helix fit distance closest approach to beam transverse plane axis less than 2.0 15 cm @xmath34 @xmath13 @xmath35 momenta of track beam @xmath36 momentum resolution of beam @xmath37 @xmath38 @xmath39 energy of track beam @xmath13 momentum track track not identified as electron dx not as muon @xmath40 @xmath41 @xmath42 time - of - flight for track nominal time - of - flight @xmath43 btof time resolution after further event level selection least 2 good charged tracks one track good helix fit total deposited energy in bsc @xmath44 further selection required based on number of good tracks for three or more prong events additional requirement all charged tracks not positive remove beam - gas events two - prong events distinguished from cosmic ray and lepton pair events further selection scheme two tracks not back - to - back two isolated neutral tracks @xmath46 @xmath47 @xmath33 and @xmath15 angle of two charged tracksevent selection routines checked by hand scans for selected rejected hadronic events separated - beam data monte carlo events typical hadronic event passed selection cuts shown in figure error from changes event yield results hand scan three major types of background considered one type cosmic rays bhabha dimuon events two - photon process events selected during routine second tau - pair production residual two - photon processes subtracted statistically via monte carlo simulation serious sources background were beam - gas beam - wall interactions most background events rejected by vertex cut beam background tracks along beam pipe direction energy deposited in bsc small most tracks were protons same selection criteria applied to separated - beam data number of events @xmath48 surviving criteria obtained number beam associated background events @xmath24 in given by @xmath49 @xmath50 determined by ratio of pressure p at collision region times integrated beam currents @xmath51integrated luminosity determined large - angle bhabha events selection criteria bsc information two clusters bsc largest deposited energy polar angle @xmath52 cluster energy @xmath53 gev 3.55 gev data @xmath54 @xmath55 @xmath56 azimuthal angles clusters radiative events cross check mdc information consistent with bsc measurement difference systematic error detection efficiency hadronic events determined monte carlo simulation jetset7.4 event generator parameters tuned 40,000 hadronic events tau mass data parameters adjusted reproduce kinematic variables multiplicity sphericity transverse momentum shows charge multiplicity sphericity aplanarity transverse momentum distributions real simulated event samples two - level trigger system retained good events rejecting background @xcite trigger derived from signals beam pickup electrode detector sub - detector event categories classified charged neutral tracks trigger level beam crossings with charged tracks two trigger channels one hit in 48 btof counter array track vc mdc 150 mev energy deposited bsc second back - to - back hits btof counter one track vc two tracks mdcneutral track trigger required sum deposited energy tracks in two cells bsc greater than 80 mev total energy bsc all sources greater than 800 mev coincidences formed using 40 120 630 640 ns wide gates beam pickup signal tof vc mdc bsc event candidates two - level criteria accepted trigger efficiencies measured run by run comparing responses trigger requirements data special runs @xmath31 resonance used independent measurements table lists measured efficiencies different event types r scan data cross - check @xmath19 data errors in trigger efficiencies for bhabha hadronic events less than @xmath570.5% main goal first scan understanding trigger hadronic event acceptances hadronic event selection background subtraction central total cross section measurement 1999 bes - ii finer scan from 2 to 5 gev figure goal reduce uncertainties of @xmath5 by factor of two over entire energy region 2.5 in -.1 cm particle data group 204 ( 1998 ) ceradini et al phys . 1973 ) 80 batoli et al phys 1972 ) 2374 bernardini et alphys . lett b51 1974 ) 200 cosme b40 1972 ) 685 kurdadze b42 1972 ) 515 litke 30 1973 ) 1189 blondel plenary talk ichep96 warsaw pietrzyk cracow 1996 davier hoeker lal 97 - 85 burkhardt pietrzyk . b356 1995 ) 398 bai nucl . instr. methods . research a344 1994 ) 319 torbjoen sjoestrand tp 95 - 20 qi et al high energy physics nuclear physics chinese berends kleiss nucl b178 1981 ) 141 kuraev s fadin 41 1985 ) 3 bonneau f. martin nucl . phys . b27 1971 ) 387 edwards slac - pub 5160 ( 1990 ) | 0.505195 | 3,266 | 1,671 | , Saving $0.1 in GPT-4. |
154,766 | we propose a new benchmark corpus to be used for measuring progress in statistical language modeling . with almost one billion words of training data
, we hope this benchmark will be useful to quickly evaluate novel language modeling techniques , and to compare their contribution when combined with other advanced techniques .
we show performance of several well - known types of language models , with the best results achieved with a recurrent neural network based language model .
the baseline unpruned kneser - ney 5-gram model achieves perplexity 67.6 .
a combination of techniques leads to 35% reduction in perplexity , or 10% reduction in cross - entropy ( bits ) , over that baseline .
the benchmark is available as a code.google.com project ; besides the scripts needed to rebuild the training / held - out data , it also makes available log - probability values for each word in each of ten held - out data sets , for each of the baseline n - gram models . | statistical language modeling has been applied to a wide range of applications and domains with great success . to name a few , automatic speech recognition , machine translation ,
spelling correction , touch - screen `` soft '' keyboards and many natural language processing applications depend on the quality of language models ( lms ) .
the performance of lms is determined mostly by several factors : the amount of training data , quality and match of the training data to the test data , and choice of modeling technique for estimation from the data .
it is widely accepted that the amount of data , and the ability of a given estimation algorithm to accomodate large amounts of training are very important in providing a solution that competes successfully with the entrenched n - gram lms . at the same time , scaling up a novel algorithm to a large amount of data involves a large amount of work , and provides a significant barrier to entry for new modeling techniques . by choosing one billion words as the amount of training data we hope to strike a balance between the relevance of the benchmark in the world of abundant data , and the ease with which any researcher can evaluate a given modeling approach .
this follows the work of goodman , who explored performance of various language modeling techniques when applied to large data sets .
one of the key contributions of our work is that the experiments presented in this paper can be reproduced by virtually anybody with an interest in lm , as we use a data set that is freely available on the web .
another contribution is that we provide strong baseline results with the currently very popular neural network lm @xcite .
this should allow researchers who work on competitive techniques to quickly compare their results to the current state of the art .
the paper is organized as follows : section 2 describes how the training data was obtained ; section 3 provides a short overview of the language modeling techniques evaluated ; finally , section 4 presents results obtained and section 5 concludes the paper . [ cols="<,^,^,^,^ " , ] another option to reduce training complexity of the maxent models is to use a hierarchical softmax @xcite . the idea is to estimate probabilities of groups of words , like in a class based model only the classes that contain the positive examples need to be evaluated . in our case , we explored a binary huffman tree representation of the vocabulary , such that evaluation of frequent words takes less time .
the idea of using frequencies of words for a hierarchical softmax was presented previously in @xcite .
the recurrent neural network ( rnn ) based lm have recently achieved outstanding performance on a number of tasks @xcite .
it was shown that rnn lm significantly outperforms many other language modeling techniques on the penn treebank data set @xcite .
it was also shown that rnn models scale very well to data sets with hundreds of millions of words @xcite , although the reported training times for the largest models were in the order of weeks .
we cut down training times by a factor of 20 - 50 for large problems using a number of techniques , which allow rnn training in typically 1 - 10 days with billions of words , @xmath0 vocabularies and up to 20b parameters on a single standard machine without gpus .
these techniques were in order of importance : a ) parallelization of training across available cpu threads , b ) making use of simd instructions where possible , c ) reducing number of output parameters by 90% , d ) running a maximum entropy model in parallel to the rnn .
because of space limitations we can not describe the exact details of the speed - ups here they will be reported in an upcoming paper .
we trained several models with varying number of neurons ( table [ results - table ] ) using regular sgd with a learning rate of 0.05 to 0.001 using 10 iterations over the data .
the maxent models running in parallel to the rnn capture a history of 9 previous words , and the models use as additional features the previous 15 words independently of order .
while training times approach 2 weeks for the most complex model , slightly worse models can be trained in a few days . note that we did nt optimize for model size nor training speed , only test performance .
results achieved on the benchmark data with various types of lm are reported in table [ results - table ] .
we focused on minimizing the perplexity when choosing hyper - parameters , however we also report the time required to train the models .
training times are not necessarily comparable as they depend on the underlying implementation .
mapreduces can potentially process larger data sets than single - machine implementations , but come with a large overhead of communication and file i / o .
discussing details of the implementations is outside the scope as this paper .
the best perplexity results were achieved by linearly interpolating together probabilities from all models .
however , only some models had significant weight in the combination ; the weights were tuned on the held - out data . as can be seen in table [ combination - table ] ,
the best perplexity is about 35% lower than the baseline - the modified kneser - ney smoothed 5-gram model with no count cutoffs .
this corresponds to about 10% reduction of cross - entropy ( bits ) .
somewhat surprisingly the sbo model receives a relatively high weight in the linear combination of models , despite its poor performance in perplexity , whereas the kn baseline receives a fairly small weight relative to the other models in the combination .
we introduced a new data set for measuring research progress in statistical language modeling .
the benchmark data set is based on resources that are freely available on the web , thus fair comparison of various techniques is easily possible .
the importance of such effort is unquestionable : it has been seen many times in the history of research that significant progress can be achieved when various approaches are measurable , reproducible , and the barrier to entry is low .
the choice of approximately one billion words might seem somewhat restrictive .
indeed , it can be hardly expected that new techniques will be immediately competitive on a large data set .
computationally expensive techniques can still be compared using for example just the first or the first 10 partitions of this new dataset , corresponding to approx .
10 million and 100 million words .
however , to achieve impactful results in domains such as speech recognition and machine translation , the language modeling techniques need to be scaled up to large data sets .
another contribution of this paper is the comparison of a few novel modeling approaches when trained on a large data set .
as far as we know , we were able to train the largest recurrent neural network language model ever reported .
the performance gain is very promising ; the perplexity reduction of 35% is large enough to let us hope for significant improvements in various applications . in the future
, we would like to encourage other researchers to participate in our efforts to make language modeling research more transparent .
this would greatly help to transfer the latest discoveries into real - world applications . in the spirit of a benchmark
our first goal was to achieve the best possible test perplexities regardless of model sizes or training time . however , this was a relatively limited collaborative effort , and some well known techniques are still missing .
we invite other researchers to complete the picture by evaluating new , and well - known techniques on this corpus .
ideally the benchmark would also contain asr or smt lattices / n - best lists , such that one can evaluate application specific performance as well . | statistical language modeling applied to applications domains with success automatic speech recognition machine translation spelling correction touch - screen soft keyboards natural language processing applications depend on quality of language models ( lms ) performance of lms determined by factors training data quality match training data to test data choice of modeling technique for estimation amount of data ability estimation algorithm to accomodate large amounts training important in with n - gram lms scaling up novel algorithm to large data involves work barrier to entry for new modeling techniques choosing one billion words as training data balance between relevance benchmark in abundant data ease modeling approach follows work of goodman explored performance of language modeling techniques large data sets experiments can be reproduced by anybody lm data set freely available on web provide strong baseline results with popular neural network lm @xcite researchers to quickly compare results to current state art paper organized section 2 describes training data section 3 overview of language modeling techniques evaluated section 4 presents results section 5 concludes paper option to reduce training complexity use hierarchical softmax @xcite idea estimate probabilities of groups words class model positive examples explored binary huffman tree representation vocabulary evaluation frequent words less time idea using frequencies for hierarchical softmax presented in @xcite recurrent neural network rnn lm achieved outstanding performance @xcite rnn lm outperforms other language modeling techniques on penn treebank data set rnn models scale well to data sets with hundreds of millions of words @xcite training times largest models weeks cut down training times by factor 20 - 50 for large problems techniques rnn training in 1 - 10 days with billions of words 20b parameters on single standard machine without techniques parallelization training across cpu threads simd instructions reducing output parameters by 90% running maximum entropy model in parallel to rnn describe details speed - ups reported upcoming paper trained models with varying neurons using regular sgd learning rate 0.05 to 0.001 10 iterations maxent models parallel rnn capture history of 9 previous words use additional features previous 15 words training times approach 2 weeks complex model models in few days nt optimize for model size training speed performance results on benchmark data with types lm reported in table [ results - ]focused minimizing perplexity choosing hyper - parameters report time to train models training times not comparable depend on implementation mapreduces process larger data sets single - machine implementations large overhead communication file i / o discussing details outside scope best perplexity results achieved linearly interpolating probabilities from all models some models had significant weight combination weights tuned on held - out data table best perplexity 35% lower than baseline - modified kneser - ney smoothed 5-gram model no count cutoffs corresponds 10% reduction of cross - entropy sbo model high weight in linear combination despite poor performance perplexity kn baseline small weight introduced new data set for measuring research progress statistical language modeling benchmark data set based on resources available web fair comparison techniques possible importance of effort significant progress achieved when approaches measurable reproducible barrier to entry low choice of one billion words restrictive new techniques immediately competitive on large data set expensive techniques compared using first 10 partitions of new dataset 10 million 100 million words impactful results speech recognition machine translation language modeling techniques need to large data setscontribution paper is comparison of novel modeling approaches trained on large data set largest recurrent neural network language model reported performance gain promising perplexity reduction 35% hope for significant improvements in applications future encourage researchers to participate make language modeling research transparent transfer discoveries into real - world applications benchmark goal achieve best test perplexities regardless of model sizes or training time limited collaborative effort known techniques missing invite researchers to evaluating new techniques on corpus ideally benchmark contain asr or smt lattices / n - best lists evaluate application specific performance | 0.269596 | 1,541 | 734 | , Saving $0.0 in GPT-4. |
154,773 | the angle @xmath0 is the most accurately measured quantity that determines the unitarity triangle . in this article
i review the various measurements of this angle performed by the babar collaboration , and discuss their implications in the search for new physics . | the babar detector@xcite is located at the @xmath1 asymmetric b - factory at slac .
the main physics goal of babar is to study cp - violation in the @xmath2 system and provide a quantitative test of the @xmath3 sector of the standard model .
the determination of the angle @xmath0 through the study of decays plays a key role in this test .
the measurement of @xmath0 from decays mediated by @xmath4 tree level diagrams allows for a precise test of cp violation in the standard model and provides the most precise constraint in the determination of the parameters @xmath5 and @xmath6 .
in addition , the measurement of the same angle in final states mediated by penguin decays and @xmath7 diagrams can be used to look for new physics .
all measurements presented in this article are based on a dataset of @xmath8 events .
at the @xmath2-factories , @xmath3 violation is studied through the measurement of the time dependent @xmath3 asymmetry , @xmath9 .
this quantity is defined as @xmath10 where @xmath11 is the number of that decay into the @xmath3-eigenstate @xmath12 after a time @xmath13 .
in general , this asymmetry can be expressed as the sum of two components : @xmath14 where @xmath15 is the difference in mass between @xmath2 mass eigenstates .
when only one diagram contributes to the final state , the cosine term vanishes . for decays such as @xmath16 , @xmath17 , where @xmath18 is the @xmath3 eigenvalue of the final state , negative for charmonium + @xmath19 and positive for charmonium + @xmath20 .
it follows that @xmath21 which shows how the angle @xmath0 is directly and simply measured by the amplitude of the time dependent @xmath3 asymmetry .
the measurement of @xmath9 utilizes decays of the @xmath22 into two neutral @xmath2 mesons , of which one ( @xmath23 ) can be completely reconstructed into a @xmath3 eigenstate , while the decay products of the other ( @xmath24 ) identify its flavor at decay time .
the time @xmath13 between the two @xmath2 decays is determined by reconstructing the two @xmath2 decay vertexes .
the @xmath3 asymmetry amplitudes are determined from an unbinned maximum likelihood fit to the time distributions separately for events tagged as and .
the angle @xmath0 can be independently measured through the three types of decays illustrated in fig .
[ fey ] and discussed in the following .
these decays , known as `` golden modes '' , are dominated by a tree level diagram @xmath4 with internal @xmath26 boson emission ( fig . [ fey]-a ) . the leading penguin diagram contribution to the final state
has the same weak phase as the tree diagram , and the largest term with different weak phase is a penguin diagram contribution suppressed by @xmath27 .
this makes @xmath28 in equation [ acpt2 ] a very good approximation .
besides the theoretical simplicity , these modes also offer experimental advantages because of their relatively large branching fractions ( @xmath29 ) and the presence of the narrow @xmath30 resonance in the final state , which provides a powerful rejection of combinatorial background .
the @xmath3 eigenstates considered for this analysis are @xmath31 , @xmath32(2s)@xmath19 , @xmath33 , @xmath34 and @xmath35 .
the asymmetry between the two distributions , clearly visible in figure [ babarsin2b-2 ] , is a striking manifestation of @xmath3 violation in the @xmath2 system .
the same figures also display the corresponding raw @xmath3 asymmetry with the projection of the unbinned maximum likelihood fit superimposed .
the results of the fit is @xmath36@xcite .
the main sources of systematic errors are uncertainties in the background level and characteristics , in the parameterization of the time resolution , and in the measurement of the mis - tag fractions .
most of these uncertainties will decrease with additional statistics .
[ babarsin2b-2 ] the world average value for @xmath37 , heavily dominated by the results from babar@xcite and belle@xcite , is @xmath38 .
this value can be compared with the indirect constraints on the apex of the unitarity triangle originating from measurements of @xmath39 , @xmath40 , @xmath41 , and @xmath42 mixing as described , for example , in reference @xcite .
the comparison , illustrated in figure [ ckmfitter1 ] , shows excellent agreement between the measurements , indicating that the observed @xmath3 asymmetry in @xmath25 is consistent with the predictions of the ckm mechanism .
the measurement of @xmath43 leads to a four - fold ambiguity in the determination of the angle @xmath0 .
this ambiguity can be reduced to a two - fold ambiguity measuring the sign of @xmath44 , which can be measured in a time - dependent angular analysis of the decays @xmath45(@xmath46 ) .
an analysis @xcite based on 81.9 fb@xmath47 measures @xmath48 , which determines the sign of @xmath49 to be positive at the 86.6% c.l . , in agreement with the standard model expectation .
the decays @xmath50 are dominated by a tree level diagram @xmath7 with external @xmath26 boson emission ( fig . [ fey]-b ) . in the standard model , the leading penguin diagram contribution to this final state
is expected to be small .
the penguin - induced correction has been estimated using factorization and heavy quark symmetry to be about 2% @xcite .
new physics could enhance the penguin contribution and would lead to a measurement of time dependent @xmath3 asymmetry substantially different from that measured in @xmath25 decays .
babar measures @xmath9 in three different channels : @xmath51@xcite , @xmath52 and @xmath53@xcite .
since two different diagrams can contribute to the final state in these decays , both the sine ( @xmath54 ) and cosine ( @xmath55 ) terms have to be extracted in the fit for @xmath9 .
the extraction of the coefficients s and c is straightforward for the channel @xmath56 , which is a pure cp eigenstate .
instead , the decay @xmath57 is an admixture of cp - odd and cp - even components .
the cp - odd fraction is measured by means of a transversity analysis@xcite to be @xmath58 .
the results of the cp fit for @xmath2 decays to open charm reported in table [ tabdd ] are in agreement with the standard model expectations .
.measurements of the sine ( s ) and cosine ( c ) coefficient in the fit of @xmath9 in @xmath7 decays . [ cols="<,<,<",options="header " , ] + in the standard model , final states dominated by @xmath60 or @xmath61 decays offer a clean and independent way of measuring @xmath62@xcite .
examples of these final states are @xmath63 , @xmath64 , @xmath65 , @xmath66 , @xmath67 , @xmath68 and @xmath69 .
these decays are mediated by the gluonic penguin diagram illustrated in figure [ fey]-c . in presence of physics beyond the standard model , new particles such as squarks and gluinos , could participate in the loop and affect the time dependent asymmetries@xcite .
the decays @xmath70 are ideal for these studies . in the standard model ,
these decays are almost pure @xmath60 penguin decays , and their cp asymmetry is expected to coincide with the one measured in charmonium + k@xmath71 decays within a few percent @xcite .
experimentally , this channel is also very clean , thanks to the powerful background suppression due to the narrow @xmath72 resonance .
unfortunately , the branching fraction for this mode is quite small ( @xmath73 ) , therefore the measurement is affected by a large statistical error .
the decays @xmath74 are favored by a larger branching fraction ( @xmath75 ) . in the standard model ,
these decays are also dominated by penguin diagrams ; other contributions are expected to be small @xcite .
a summary of the measurements of @xmath9 in penguin modes @xcite by the babar experiment is reported in figure [ figurepenguin ] .
the average of all the penguin modes @xcite , shown in yellow in figure [ figurepenguin ] , is 2.8@xmath76 away from the value of @xmath37 measured in the golden mode .
this discrepancy , however , has to be interpreted with caution since each mode is theoretically affected by new physics in different ways .
the measurement of time - dependent cp asymmetry in decays have provided a crucial test of @xmath3 violation in the standard model .
the parameter @xmath37 is now measured in @xmath4 decays by babar with a precision of 5% .
measurements of time - dependent @xmath3 violation asymmetries in @xmath7 and in penguin - dominated modes are sensitive to contributions from physics beyond the standard model .
these measurements are still heavily dominated by statistical errors and will benefit greatly from additional data .
babar is planning to double its dataset by 2006 and quadruple it by 2008 .
9 b. aubert _ et al .
_ , nucl .
instrum .
a * 479 * , 1 ( 2002 ) phys .
94 , 161803 ( 2005 ) k. abe _ et al .
_ , phys .
d * 71 * , 072003 ( 2005 ) b. aubert _ et al .
_ , prd 71 ( 2005)032005 . see http://ckmfitter.in2p3.fr .
x. y. pham and z. z. xing , phys .
b 458 , 375 ( 1999 ) . d. london and r. d. peccei , phys . lett .
b * 223 * , 257 ( 1989 ) ; n. g. deshpande and j. trampetic , phys .
d * 41 * , 895 ( 1990 ) ; r. fleischer , z. phys .
c * 62 * , 81 ( 1994 ) ; n. g. deshpande and x. g. he , phys .
b * 336 * , 471 ( 1994 ) ; y. grossman , z. ligeti , y. nir and h. quinn , phys .
d * 68 * , 015004 ( 2003 ) ; m. gronau and j. l. rosner , phys .
b * 564 * , 90 ( 2003 ) .
a.b . carter and a.i .
sanda , phys .
d * 23 * , 1567 ( 1981 ) ; i.i .
bigi and a.i .
sanda , nucl .
b * 193 * , 85 ( 1981 ) ; r. fleischer and t. mannel , phys .
b * 511 * , 240 ( 2003);y .
grossman , g. isidori and m.p .
worah , phys .
d * 58 * , 057504 ( 1998 ) ; y. grossman , z. ligeti , y. nir and h. quinn , phys .
d * 68 * , 015004 ( 2003 ) ; y. grossman and m.p .
worah , phys .
lett . b * 395 * , 241 ( 1997 ) ; r. fleischer , int . j. mod
. phys . a * 12 * , 2459 ( 1997 ) ; d. london and a. soni , phys . lett . b * 407 * , 61 ( 1997 ) .
m. gronau _ et .
b 596 , 107 ( 2003 ) ; m. beneke and m. neubert , nucl .
b675 , 333 ( 2003 ) .
b. aubert _ et al .
_ , phys . rev .
d 71 , 091102(r ) ( 2005 ) b. aubert _ et al .
94 , 191802 ( 2005 ) b. aubert _ et al .
_ , arxiv : hep - ex/0503011 . b. aubert _ et al .
_ , arxiv : hep - ex/0408095 . b. aubert _ et al .
_ , arxiv : hep - ex/0503018 .
the heavy flavor averaging group , http://www.slac.stanford.edu / xorg / hfag/. | babar detector at @xmath1 asymmetric b - factory at slac cp - violation in @xmath2 system quantitative test @xmath3 sector standard model determination angle @xmath0 through decays measurement @xmath0 from decays @xmath4 tree level diagrams precise test cp violation precise constraint parameters @xmath5 @xmath6 measurement same angle in final states penguin decays @xmath7 diagrams new physics measurements based on dataset @xmath8 events @xmath2-factories @xmath3 violation studied through time dependent @xmath3 asymmetry quantity defined as @xmath10 number decay into @xmath3-eigenstate after asymmetry expressed as sum of two components difference in mass between @xmath2 mass eigenstates one diagram contributes final state cosine term vanishes decays @xmath16 @xmath18 @xmath3 eigenvalue final state negative for charmonium + @xmath19 positive for charmonium + @xmath20 angle @xmath0 measured by amplitude time dependent @xmath3 asymmetrymeasurement @xmath9 utilizes decays @xmath22 into two neutral @xmath2 mesons one @xmath23 reconstructed into @xmath3 eigenstate decay products other @xmath24 identify flavor at decay time time between two decays determined by reconstructing decay vertexes @xmath3 asymmetry amplitudes determined from unbinned maximum likelihood time distributions for events angle @xmath0 measured through three decays fig . decays golden modes dominated by tree level diagram @xmath4 with internal @xmath26 boson emission leading penguin diagram contribution same weak phase as tree diagram largest term with different weak phase is penguin diagram contribution suppressed by @xmath27 makes @xmath28 in equation good approximation modes offer experimental advantages large branching fractions narrow @xmath30 resonance in final state rejection combinatorial background @xmath3 eigenstates considered are @xmath31 asymmetry between distributions visible in figure @xmath3 violation in @xmath2 systemfigures display @xmath3 asymmetry unbinned maximum likelihood fit superimposed fit is @xmath36@xcite main sources errors are uncertainties background level characteristics time resolution measurement mis - tag fractions uncertainties decrease with additional statistics world average value for @xmath37 dominated results babar belle@xcite is @xmath38 compared with indirect constraints unitarity triangle from measurements @xmath39 mixing comparison shows agreement measurements @xmath3 asymmetry in @xmath25 consistent with predictions ckm mechanism measurement @xmath43 leads four - ambiguity angle @xmath0 ambiguity reduced to two - fold ambiguity measuring @xmath44 measured in time - dependent angular analysis decays @xmath45(@xmath46 ) analysis based 81.9 fb@xmath47 measures @xmath48 determines sign @xmath49 positive at 86.6% c.l . agreement standard model expectation decays @xmath50 dominated by tree level diagram @xmath7 external @xmath26 boson emission standard model leading penguin diagram contribution to final state smallpenguin - induced correction estimated about 2% @xcite new physics could enhance penguin contribution time dependent @xmath3 asymmetry different from @xmath25 decays babar measures @xmath9 in three channels two diagrams contribute final state sine and cosine terms extracted in fit for @xmath9 extraction of coefficients s and c straightforward for channel @xmath56 pure cp eigenstate decay @xmath57 admixture of cp - odd cp - even components cp - odd fraction measured transversity analysis @xmath58 results of cp fit for @xmath2 decays agreement with standard model expectations.measurements of sine cosine c ) coefficient in fit of @xmath9 in @xmath7 decays standard model final states dominated by @xmath60 or @xmath61 decays offer measuring @xmath62@xcite examples final states are @xmath63 decays mediated by gluonic penguin diagram figurephysics beyond standard model new particles squarks gluinos could participate affect time dependent asymmetries@xcite decays @xmath70 ideal for studies standard model pure @xmath60 penguin decays cp asymmetry coincide with charmonium + k@xmath71 decays experimentally channel clean background suppression narrow @xmath72 resonance branching fraction small @xmath73 affected by large statistical error decays @xmath74 favored by larger branching fraction @xmath75 standard model decays dominated by penguin diagrams other contributions small summary measurements @xmath9 penguin modes babar experiment reported in figure average penguin modes 2.8@xmath76 from @xmath37 golden mode discrepancy interpreted caution each mode affected by new measurement time - dependent cp asymmetry in decays test of @xmath3 violation standard model parameter @xmath37 measured in @xmath4 decays precision 5% time - dependent @xmath3 violation asymmetries in @xmath7 penguin - dominated modes sensitive to physics beyond standard model measurements dominated by statistical errors benefit from additional datababar double dataset 2006 quadruple 2008 aubert nucl 479 2002 phys . 94 161803 2005 abe phys . d * 71 * 072003 2005 aubert 71 2005 http://ckmfitter.in2p3.fr pham xing . 458 375 1999 ) london peccei . b 223 257 ( 1989 ) deshpande trampetic phys . d * 41 * 895 ( 1990 ) fleischer . c 62 * 81 1994 ) deshpande he phys b 336 471 ( 1994 ) grossman ligeti nir quinn phys . d * 68 * 015004 2003 ) gronau rosner . b 564 * 90 ( 2003 ) carter sanda . d * 23 * 1567 ( 1981 ) bigi sanda nucl . b * 193 * 85 ( 1981 ) fleischer mannel phys . b * 511 * 240 ( grossman isidori worah . d * 58 * 057504 1998 ) grossman ligeti nirquinn phys . 68 015004 2003 ) grossman worah phys . 395 * 241 ( 1997 ) fleischer mod phys . 12 * 2459 ( 1997 ) d london soni . lett 407 * 61 ( 1997 ) gronau 596 , 107 ( 2003 ) beneke neubert nucl b675 333 ( 2003 ) aubert . rev d 71 , 091102(r ) ( 2005 aubert 94 , 191802 ( 2005 ) arxiv hep - ex/0503011 hep - ex/0408095 ex/0503018 heavy flavor averaging group http://www.slac.stanford.edu xorg hfag | 0.435044 | 2,936 | 1,464 | , Saving $0.1 in GPT-4. |
154,791 | the existence of complex stellar populations in some star clusters challenges the understanding of star formation .
e.g. the onc or the sigma orionis cluster host much older stars than the main bulk of the young stars .
massive star clusters ( @xmath0 cen , g1 , m54 ) show metallicity spreads corresponding to different stellar populations with large age gaps . we show that ( i ) during star cluster formation field stars can be captured and ( ii ) very massive globular clusters can accrete gas from a long - term embedding inter stellar medium and restart star formation . | the collapsing cloud is described by a plummer potential with a fixed total gas mass and a time - dependent plummer parameter , infinite at the beginning and decreasing within the collapse time - scale .
the field stars are simulated by test particles which are initially uniformly distributed in a sphere and have a gaussian velocity distribution .
@xcite found 4 stars out of 84 low mass stars in the onc being 10 myr older than the main bulk of stars and followed that star formation is prolonged .
extrapolating , @xmath1 53 such older stars are expected in the whole onc @xcite . the number of simulated captured stars within a radius of 2.5 pc of the centre of the onc is plotted in fig . [ fig01_02 ] ( left ) for different collapse time - scales and initial background velocity dispersions .
age spreads in young star clusters may therefore not be due to prolonged star formation but can be explained by stellar capture .
@xcite found 110 faint red sources in a field of 3@xmath23 pc@xmath3 of the centre of r136 in the lmc of unknown physical nature and excluded that they can be red giants as their required age would be larger than 350 myr .
as membership probabilities can not be determined for stars in r136 all stars within the central sphere with a radius of 1.7 pc are calculated after the collapse with a time - scale of 10 myr has stopped @xcite .
the faint red sources can indeed be captured old red giants . , and collapse time - scales , @xmath4 , @xcite .
_ right _ : field stars in the central sphere of 1.7 pc radius in r136 for different velocity dispersions and final plummer parameters , @xmath5 , @xcite.,title="fig:",scaledwidth=50.0% ] , and collapse time - scales , @xmath4 , @xcite .
_ right _ : field stars in the central sphere of 1.7 pc radius in r136 for different velocity dispersions and final plummer parameters , @xmath5 , @xcite.,title="fig:",scaledwidth=50.0% ]
globular clusters more massive than @xmath1 10@xmath6 m@xmath7 such as @xmath0 cen , g1 and m54 show a spread in metallicity corresponding to different stellar population with age gaps of several hundred myr or a few gyr .
m54 and @xmath0 cen are confirmed and supposed , respectively , to have been embedded in the ism of a dwarf galaxy . to explore the effect of the cluster potential on a long - term embedding co - moving ism we calculate the hydrostatic solution of an isothermal non - self - gravitating gas with an additional plummer potential @xcite . the central particle density in the cluster potential is plotted in fig . [ fig03 ] for different temperatures and densities of the warm component of the ambient ism .
star clusters more massive than @xmath1 10@xmath6 m@xmath7 may cause an instability of the ism , start gas accretion and form stars . gas accretion can explain multiple stellar populations in massive star clusters and the change of the mass - radius relation at cluster masses of @xmath1 10@xmath6 m@xmath7 . | collapsing cloud described plummer potential fixed gas mass time - dependent plummer parameter infinite beginning decreasing collapse time scale field stars simulated by test particles uniformly distributed sphere gaussian velocity distribution @xcite found 4 stars out of 84 low mass stars onc 10 myr older star formation prolonged 53 older stars expected in whole onc simulated captured stars within radius 2.5 pc centre onc plotted in fig for collapse time - scales initial background velocity dispersions age spreads in young star clusters not due to prolonged star formation explained by stellar capture found 110 faint red sources in centre r136 unknown nature excluded red giants required age larger than 350 myr membership probabilities r136 stars central sphere with radius 1.7 pc calculated after collapse time - scale 10 myr faint red sources be captured old red giants collapse time - scales field stars central sphere 1.7 pc radius different velocity dispersions final plummer parameters collapse time - scales field stars central sphere 1.7 radius different velocity dispersions final parameters.0% globular clusters massive @xmath1 @xmath0 cen g1 m54 show spread metallicity different stellar population age gaps myr gyr m54 cen confirmed embedded in ism dwarf galaxy effect cluster potential long - term embedding - ism calculate hydrostatic solution isothermal non - self - gravitating gas additional plummer potential @xcite central particle density cluster potential plotted in fig . temperatures densities warm component ambient ism star clusters massive cause instability ism start gas accretion form stars gas accretion multiple stellar populations clusters change mass - radius relation at cluster masses @xmath1 | 0.133635 | 722 | 325 | , Saving $0.0 in GPT-4. |
154,793 | we present here a study of the infrared properties of asymptotic giant branch stars ( hereafter agb ) based on existing databases , mainly from space - borne experiments .
preliminary results about c and s stars are discussed , focusing on the topics for which future infrared surveys from antarctica will be crucial .
this kind of surveys will help in making more quantitative our knowledge of the last evolutionary stages of low mass stars , especially for what concerns luminosities and mass loss . | towards the end of their life , stars of low and intermediate mass ( @xmath0 ) evolve along the asymptotic giant branch ( agb ) phase [ see busso _ et al .
_ @xcite and references herein for more details ] . in this stage
they experience extensive phenomena of mass loss that affect deeply their evolution .
sedlmayr @xcite shows that stellar winds are also fundamental for the enrichment of the interstellar medium that is replenished by these stars with about 70% of all the matter returned after stellar evolution .
moreover , agb stars provide the starting conditions for the formation of planetary nebulae .
the radiative emission of the cool dust in the infrared ( ir ) normally dominates the energy distribution of agb stars ( particularly for the most evolved ones ) and this fact is mainly due to their strong stellar winds . until recently the bolometric magnitude of the most evolved agb stars was difficult to derive , due to insufficient photometric coverage of the mid - ir range of the electromagnetic spectrum ( the importance of mid - ir observations for agb stars is clearly shown in guandalini _ et al . _
@xcite , figure 1 ) .
the availability of large ir databases from space - borne telescopes like iso , irts , msx has substantially improved this situation . at the same time , hipparcos distances for agb stars have been corrected from various biases in works like the one from bergeat & chevallier @xcite and the period - luminosity relations found for miras have been drastically improved as shown in whitelock _
the study of fundamental physical parameters of these stars ( like luminosity , ir colors and mass loss ) can now be be performed in a rather quantitative way .
however , ir space - borne observations of agb stars present some disadvantages . in particular
, the duration of the operational period of the telescope is quite limited , observations with a long time of integration are difficult and agb stars are generally observed at a single epoch .
all these facts hinder our understanding of several basic physical parameters , which are fundamental in the study of agb stars . in this respect a complementary role in solving these problems
could be played by ground - based observations at ir wavelengths , especially from antarctica .
the antarctic plateau ( in particular dome c ) presents the best conditions available on earth from the point of view of infrared observations as shown in several other contributions from this conference : therefore , antarctica is the best place where ground - based observations in the ( mid-)ir can be performed .
an antarctic ir observatory might be crucial for clarifying the final stages of stellar evolution by : * observing properties of evolved stars in the magellanic clouds , at known distance and metallicity different from the milky way ; * doing the same for the galactic center , where the metal blend is different and the extreme o - enhancement prevents the formation of c stars as shown in uttenthaler _
_ @xcite ; * looking for mass loss calibrations in the ir .
our efforts are addressed to the preparation for ir observations of agb stars from antarctica through the irait telescope , presented by tosti _
@xcite in this conference . with this aim
we are trying to understand which kind of observations from antarctica is the best and most promising from the point of view of agb stars . in guandalini _
@xcite,@xcite we are making for these sources an overview of the ir data available from existing catalogues that will be completed in future works . in this note
we address some interesting issues concerning agb stellar variability from an ir point of view , and the relevance of antarctic observations for them .
m all show a constant flux in mid - ir . ]
m appear to be variable . ]
agb stars present strong variability at optical wavelengths and are roughly divided into three subclasses according to their variability type : miras , semiregulars and irregulars .
effects due to variability at ir wavelengths are expected to be less relevant when compared with the optical ones .
figures [ fig1 ] and [ fig2 ] present the available information on the ir seds for two groups of agb sources that are discussed in detail in busso _ et al . _ @xcite .
figure [ fig1 ] shows distributions that share the property of being non - variable at ir wavelengths over an elapse of time of almost 20 years .
they include semiregular sources with minimal ir excess , in which the sed is peaked in near - ir .
they also include evolved ( post - agb ) objects in which the maximum emission is at very long wavelengths ( from 20 to more than 40 @xmath1 m ) .
instead , figure [ fig2 ] shows a group of mira variables , in which the emission peaks near 10 @xmath1 m : they present the common property of a long - term mid - ir variability that seems to be restricted to this special class of sources .
moreover , figure [ fig3 ] shows two of the few available agb sources observed several times by iso - sws and confirms the same behavior : mira variables present an ir variability even over a relatively short time interval .
m region appears to be variable . ]
there is not yet an agreement on the origin and properties of this phenomenon .
it needs to be examined in more detail to understand its nature .
it could be perhaps a variability induced by shock waves caused by dynamic events in the photosphere or by magneto - hydrodynamical modes ( and magnetic storms ) .
otherwise , it could be a `` simpler '' mid - ir variability , originating in the emission of the circumstellar envelopes and caused by modulations in the efficiency of dust formation .
this would be expected to be more typical of the mira sources ; semiregulars have thinner circumstellar envelopes , while post - agb stars should have detached shells not strongly influenced by the variability of the central star .
the best way to examine this variability is that of observing agb stars at different epochs also in mid - ir and this task can be performed by a ground - based telescope placed in antarctica .
moreover , simultaneous observations in the near- and mid - ir and correlations with optical variability could be fundamental to understand these phenomena .
important ir studies for agb stars can be made in the best way from ground - based locations like dome c and antarctica .
the study of the main features regarding ir variability ( and also optical variability ) is one of them .
it can be fulfilled with , surveys through wide field ( in the future ) or small area ( irait ) imaging of interesting stellar systems , multiple observations of chosen agb sources at different epochs and wavelengths and observation of agb sources from magellanic clouds , whose distance is well - estimated . * light curves and therefore good knowledge of variability ; * an accurate study of the luminosity variations over a wide region of the electromagnetic spectrum , including optical and ir ; * finally , a reliable comparison between observations and models of agb stars
. 99 bergeat , j. , & chevallier , l. 2005 , a&a , 429 , 235 busso , m. , gallino , r. , & wasserburg , g. j. 1999 , ara&a , 37 , 239 busso , m. , guandalini , r. , persi , p. , corcione , l. , & ferrari - toniolo , m. 2006 , aj , submitted guandalini , r. , busso , m. , ciprini , s. , silvestro , g. , & persi , p. 2006
, a&a , 445 , 1069 guandalini , r. , busso , m. , & cardinali , m. 2007 , in preparation sedlmayr , e. 1994 , lecture notes in physics , 428 , 163 straniero , o. , domnguez , i. , cristallo , s. , & gallino , r. 2003 , pasa 20 , 389 tosti , g. 2007 , this conference uttenthaler , s. , hron , j. , lebzelter , t. , busso , m. , schultheis , m. , & kaeufl , h. u. 2006 , astro - ph/0610500 whitelock , p. a. , feast , m. w. , marang , f. , & groenewegen , m. a. t. 2006 , mnras , 369 , 751 | end life stars low intermediate mass ( @xmath0 ) evolve asymptotic giant branch ( agb ) phase see busso _ et al . _ @xcite references details experience mass loss evolution sedlmayr @xcite shows stellar winds fundamental for enrichment interstellar medium replenished 70% matter returned after evolution agb stars provide starting conditions for formation planetary nebulae radiative emission cool dust infrared ir dominates energy distribution agb stars evolved due to strong stellar winds bolometric magnitude of evolved agb stars difficult derive due insufficient photometric coverage mid - ir range importance observations shown guandalini _ et al . _ @xcite figure 1 availability large ir databases space - borne telescopes iso irts msx improved situation hipparcos distances agb stars corrected biases period - luminosity relations for miras improved study fundamental physical parameters stars luminosity ir colors mass loss performed ir space - borne observations of agb stars present disadvantages duration operational period telescope limited observations long time integration difficult agb stars observed single epochfacts hinder understanding physical parameters fundamental in study agb stars complementary role solving problems by ground - based observations at ir wavelengths especially from antarctica antarctic plateau dome c presents best conditions infrared observations antarctica best place ground - based observations in mid-)ir antarctic ir observatory might crucial for final stages stellar evolution by observing properties evolved stars in magellanic clouds distance metallicity different from milky way galactic center metal blend different extreme o - enhancement prevents formation of c stars looking for mass loss calibrations in ir efforts preparation for ir observations of agb stars from antarctica through irait telescope presented by trying to understand which observations from antarctica best most promising agb stars overview of ir data from existing catalogues future address issues concerning agb stellar variability ir relevance of antarctic observations show constant flux in mid - ir variable agb stars present strong variability at optical wavelengths divided into three subclasses miras semiregulars irregulars effects due to variability at ir wavelengths expected less relevant compared with opticalfigures fig1 and fig2 ] present ir seds for two groups agb sources discussed in busso et al . @xcite fig1 shows distributions non - variable at ir wavelengths over 20 years include semiregular sources with minimal ir excess sed peaked in near - ir evolved post - agb objects maximum emission at long wavelengths 20 to 40 @xmath1 m ) figure fig2 ] shows mira variables emission peaks near 10 @xmath1 m long - term mid - ir variability restricted to figure fig3 ] shows two agb sources observed by iso - sws confirms mira variables present ir variability over short time interval region variable agreement on origin properties phenomenon needs induced by shock waves events photosphere magneto - hydrodynamical modes storms simpler mid - ir variability in emission circumstellar envelopes by modulations dust formation typical of mira sources semiregulars have thinner circumstellar envelopes post - agb stars detached shells not influenced by variability central star variability agb stars at different epochs in mid - ir by ground - based telescope in antarcticasimultaneous observations near- mid - ir correlations optical variability fundamental phenomena ir studies for agb stars from ground locations like dome c antarctica study main features ir variability optical variability surveys wide field small area imaging stellar systems multiple observations of agb sources different epochs wavelengths sources from magellanic clouds light curves knowledge variability accurate study luminosity variations electromagnetic spectrum including optical ir reliable comparison between observations models agb stars bergeat chevallier . 2005 a&a , 429 , 235 busso gallino wasserburg , g. 1999 ara&a , 37 , 239 busso guandalini persi corcione ferrari - toniolo . 2006 guandalini busso ciprini silvestro persi 2006 a&a , 445 , 1069 guandalini busso cardinali . 2007 sedlmayr . 1994 lecture notes physics , 428 , 163 straniero domnguez cristallo gallino . 2003 pasa 20 , 389 tosti ,.2007 conference uttenthaler s hron j lebzelter busso m schultheis kaeufl h.. 2006 astro - ph/0610500 whitelock feast marang f. groenewegen m 2006 mnras 369 751 | 0.313571 | 1,824 | 922 | , Saving $0.1 in GPT-4. |
154,795 | the quantum mechanical approach to the well known prisoners dilemma , one of the basic examples to illustrate the concepts of game theory , is implemented with a classical optical resource , nonquantum entanglement between spin and orbital degrees of freedom of laser modes .
the concept of entanglement is crucial in the quantum version of the game , which brings novel features with a richer universe of strategies . as we show
, this richness can be achieved in a quite unexpected context , namely that of paraxial spin - orbit modes in classical optics .
numerous quantum information protocols rely on entanglement , a property usually attributed to composite quantum systems that can not be described by a tensor product state vector .
quantum cryptography and teleportation are popular examples of protocols relying on quantum entanglement . while this property is considerably sensitive to local measurements performed in each party of the composite system , it is unaffected by local unitary operations .
teleportation protocols , for example , are achieved by classical communication followed by local unitary operations and measurements .
this framework is where quantum mechanics meets an important area of applied mathematics , the game theory , a powerful tool for decision making @xcite . here
, two or more agents ( players ) take their decisions by acting on a quantum system with unitary operations .
these decisions or conflict situations can be as simple as tossing a coin@xcite or rather involved like the so - called minority game @xcite , where one models a competition among several players for a limited resource . in this sense games
can be cooperative or non - cooperative like the prisoners dilemma , and with complete ( incomplete ) information where one player knows ( does not know ) all strategies his opponent can choose .
quantum versions of this game was realized experimentally both with nuclear magnetic resonance @xcite and entangled photon pairs @xcite .
although frequently attributed to quantum systems , entanglement has been recently identified in classical optics as the coherent superposition of paraxial modes with orthogonal spatial profiles and orthogonal polarizations .
we shall refer to such superpositions as _ spin - orbit modes_. these modes were used in our group as a classical optical resource to investigate the topological phase acquired by an entangled state following a cyclic evolution under local unitary operations @xcite and to demonstrate alignment free bb84 quantum cryptography ref.@xcite .
also , a spin - orbit bell inequality has been investigated both in the quantum @xcite and classical @xcite domains .
an important tool for spin - orbit coupling was used in ref.@xcite to exchange quantum information between these two degrees of freedom . in a recent work @xcite the term _ nonquantum entanglement _
has been coined to designate the spin - orbit inseparability of paraxial modes in connection with the mueller matrices employed in polarization optics . in this work
we demonstrate how this kind of nonquantum entanglement can be used to evaluate the performance of quantum strategies in a classical example of game theory , the prisoners dilemma .
a laser beam propagating along the @xmath0 direction is usually described by its polarization unit vector @xmath1 and its spatial mode @xmath2 .
the spatial modes in rectangular coordinates are hermite - gaussian ( hg ) solutions of the paraxial wave equation described in many text books @xcite .
the subspace of first order spatial modes has a qubit structure similar to the polarization mode space , where hg modes along different orientations play the role of linear polarizations and the laguerre - gaussian ( lg ) modes are equivalent to circular polarization @xcite . by combining the two mode spaces ,
a general spin - orbit mode can be written as @xmath3 where @xmath4 are linear polarization unit vectors along the horizontal ( vertical ) directions , and @xmath5 are hg spatial modes along these same directions . in analogy to the usual entanglement measure used for bipartite quantum states @xcite , we can define the spin - orbit mode concurrence , which for eq.([spmode ] ) is @xmath6 .
we shall refer to the spin - orbit modes with maximal concurrence @xmath7 as maximally entangled modes . in these modes neither the polarization nor the spatial profile is well defined , in fact they correspond to polarization vortices which have been extensively studied due to their potential applications to high resolution microscopy @xcite .
in this work we demonstrate another appealing feature in the unexpected context of a quantum game .
an example of such entangled spin - orbit mode is @xmath8 we use this mode to implement a classical optical version of the well known prisoners dilemma , in which two players , alice and bob , accused of a felony , have to decide whether they cooperate ( @xmath9 ) or defeat ( @xmath10 ) each other .
in classical game theory , each agent decision is represented by one bit of information with possible states @xmath9 and @xmath10 . depending on their decision
, a reduced penalty may be applied to each one of them .
the penalty reduction is the payoff each player gets from their combined decisions .
the penalty reductions for both players are shown in table [ redtab ] for all possible strategies adopted by alice ( rows ) and bob ( columns ) .
.penalty reduction table [ cols="^,^,^",options="header " , ] from the table we see that both players are tempted to choose @xmath10 , although their added reduction would be maximized by @xmath11 .
here comes the dilemma , the players are isolated without the permission to negotiate .
each player is left to his own and has to decide whether to defeat or cooperate with the other , a bad choice may cost his freedom . for a cooperative game , players would choose strategies which maximize both payoffs , i.e. , they would search for pareto optimal strategies ; on the other hand , since prisoners dilemma is a non - cooperative game , each player will try to maximize solely his own payoff , i.e. , the intelligent choice is the nash equilibrium @xmath12 . at this point ,
the concepts of game theory are due .
suppose this situation is repeated many times and the players adopt probabilistic strategies , that is , they randomly choose between @xmath9 and @xmath10 with prestablished probabilities .
then , the payoff function of each player is given by the average penalty reduction obtained : @xmath13 where @xmath14 and @xmath15 is the joint probability that alice chooses @xmath16 and bob chooses @xmath17 . in the classical approach to the problem , one shows that the payoff as a function of @xmath18 and @xmath19 has an absolute minimum at @xmath20 , that is the best the players can do is to defeat , as a consequence of the severe cost brought by a possible betrayment . in ref.@xcite eisert proposes an ingenuous alternative to the classical approach , employing the concept of quantum entanglement .
briefly , in this approach the prisoners share a pair of entangled qubits and rather than making a definite @xmath9 or @xmath10 statement , they are allowed to perform single qubit unitary operations ( strategies ) , each one on the qubit in his possession .
the entangled two - qubit state is prepared by a nonlocal operation @xmath21 \label{u}\ ] ] acting on an initial state @xmath22 , so that @xmath23 .
after each player has applied his own strategy @xmath24 ( @xmath14 ) , the qubits are nonlocally operated with @xmath25 and separately measured .
the payoff function ( [ payoff ] ) is evaluated with the probabilities @xmath26 ( @xmath27 ) associated with the two possible outcomes ( @xmath9 or @xmath10 ) in each qubit .
therefore , the strategy space is much larger in this quantum approach , it corresponds to the space of @xmath28 matrices , apart from irrelevant global phases .
this quantum version has been implemented experimentally with quantum correlated photon pairs generated by parametric down conversion @xcite .
we now demonstrate the implementation of the quantum game strategies with the nonquantum entanglement provided by the spin - orbit classical mode ( [ psi0 ] ) . in the game language ,
we shall make the identification @xmath29 and @xmath30 .
the experimental setup is shown in fig.([setup ] ) .
the tem@xmath31 output of a he - ne laser is diffracted by a hologram that generates an hg mode on the first diffraction order , so that the initial spin - orbit mode is @xmath32 .
this mode is first sent to a quarter wave plate ( qwp ) rotated at @xmath33 , which makes the transformation @xmath34 , and then to a mach - zehnder ( mz ) interferometer with input and output polarizing beam splitters ( pbs ) .
a dove prism ( dp ) is inserted in one arm of the mz interferometer , and the relative phase @xmath35 between the two arms is controlled by a piezoelectric transducer ( pzt ) .
the dp oriented at @xmath33 makes the transformation @xmath36 , so that this mz interferometer coherently superposes modes @xmath37 and @xmath32 at its output with an adjustable phase @xmath35 . in the basis @xmath38 ,
the matrix representation of the mz transformation is : @xmath39\ ; , \label{mz}\ ] ] so that after passing through the qwp and the balanced interferometer ( @xmath40 ) , the beam is prepared in mode ( [ psi0 ] ) .
this mode is the object on which the players will implement their strategies . in order to continue our experimental description ,
it is essential to define the mode converter operators which will be used to implement the players strategies and @xmath25
. we will be dealing with either polarization ( wave plates ) or spatial ( dp and cylindrical lenses ) mode converters , that is , elements acting on one degree of freedom only .
we shall inform rotation angles in degrees and phase retardations in radians for immediate identification of the physical meaning all over the experimental description .
when oriented along horizontal and vertical directions , they introduce a retardation phase @xmath35 between @xmath41 and @xmath42 modes .
a mode converter rotated by the angle @xmath43 is described by the @xmath44 matrix : @xmath45\;. \label{c}\ ] ] for example , quarter wave plates correspond to @xmath46 and half wave plates ( hwp ) to @xmath47 .
spatial mode converters can be made with cylindrical lenses @xcite for variable retardation @xmath35 , or the dp for @xmath47 .
now , alice is equipped with polarization elements and realizes strategies of the kind @xmath48 , whereas bob is equipped with dps and cylindrical lenses , and his strategies are @xmath49 .
after the players have made their choices , the spin - orbit mode of the laser beam is : @xmath50 \mathbf{\psi}_0(\mathbf{r } ) \;.
\label{psi1}\ ] ] the mode converters are also used together with @xmath51 to implement @xmath25 .
indeed , one can easily show that : @xmath52\,mz(\pi/2)\ ; , \label{udagger}\ ] ] which physically means that , after passing through the players strategies , the beam is sent through a mz interferometer with a @xmath53 phase shift , a quarter wave plate ( qwp ) rotated at @xmath54 , and another mz interferometer with balanced arms ( @xmath40 ) , where it gets transformed to mode @xmath55 where @xmath56 and @xmath57 .
after this transformation , we arrive at the measurement stage where the probabilities @xmath58 used in the payoff function eq.([payoff ] ) are obtained . in our setup , these probabilities are given by the projected intensities : @xmath59\right\rvert}^2 \ ; , \label{pmn}\end{aligned}\ ] ] which , in fact , correspond to photodetection probabilities when the beam is attenuated down to the single photon regime . in order to measure the projected intensities ,
the @xmath60 mode is first sent to a pbs where polarization projection is performed .
then , each pbs output is sent to a balanced mach - zehnder interferometer with an additional mirror in one arm ( mzim ) , where spatial mode projection occurs @xcite .
the projected intensities are measured at the four outputs of the two mzims either with a ccd camera or with photodetectors .
the payoff function is evaluated with the intensities measured with four photodetectors .
the background noise is subtracted and the partial intensities are then normalized to the total intensity so that @xmath61 . for each player , we implemented five different strategies : @xmath62 , @xmath63 , @xmath64 , @xmath65 , and @xmath66 . in this notation @xmath67 and @xmath68
are the usual pauli matrices .
strategies @xmath69 and @xmath70 are equivalent to the classical ones where the players can only cooperate or defeat , while the other strategies are intrinsically quantum mechanical since they involve rotations and phase retardations not available in the classical scenario .
a table with the coefficients @xmath71 resulting from these strategies is shown in fig.([tables]a ) .
also , the images obtained with the ccd camera are displayed in the same table format in fig.([tables]b ) . as expected ,
only those output ports corresponding to nonzero coefficients are illuminated . in this sense ,
the qualitative agreement between the two tables is clear .
we have also evaluated alice s payoff as a function of the strategy parameters @xmath72 and @xmath73 in the domain @xmath74 and @xmath75 .
the analytical result is shown in fig.([figpayoff ] ) together with the points corresponding to all possible combinations of the experimental strategies .
the experimental values were obtained from the intensity measurements , where the relative intensities play the role of measurement probabilities .
we observe , from both theoretical and experimental results , that the quantum move @xmath76 proposed originally in @xcite dominates all classical ones and it is also a nash equilibrium with a better outcome than @xmath77 since no player can improve his respective payoff by changing unilaterally his strategy .
note also that if bob could choose only between classical @xmath78 or @xmath79 , and alice keep @xmath80 then it would be better for him to cooperate since his payoff would be increased . and @xmath73 .
dots correspond to the experimental values obtained with the intensity measurements . ] in conclusion , we have used the concept of nonquantum entanglement to implement a game theory task in the context of the well known prisoners dilemma .
the advantages offered by the quantum mechanical approach could be realized in the classical optics framework .
nonseparable spin - orbit modes corresponding to polarization vortices were used .
this implementation opens promising perspectives regarding potential applications of nonquantum entanglement to the investigation of quantum information protocols . | funding was provided by coordenao de aperfeioamento de pessoal de nvel superior ( capes ) , fundao de amparo pesquisa do estado do rio de janeiro ( faperj - br ) , and instituto nacional de cincia e tecnologia de informao quntica ( inct - cnpq ) . | funding provided coordenao aperfeioamento pessoal superior fundao amparo pesquisa rio janeiro faperj instituto nacional cincia tecnologia quntica inct cnpq | 0.066397 | 74 | 41 | , Saving $0.0 in GPT-4. |
154,799 | this paper presents an interpretation based on gamma - ray burst source kinematics for the relationship found by norris et al . between peak luminosity and energy - dependent pulse lag .
i argue that the correlation should instead be between _ number _ luminosity and pulse lag .
this interpretation improves the least - squares fit of this correlation for the known bursts by 25 percent or more .
it also suggests a distance estimation scheme .
i propose that this relationship is due to the variation in line - of - sight velocity among bursts .
this interpretation allows one to speculate on the range of gamma - ray burst expansion velocities or the size of their jet opening angles . | it has recently been reported by @xcite that there is a relationship between the peak luminosity of gamma - ray bursts ( grb ) and the pulse time lag between batse energy channels . specifically , they find that the peak luminosity @xmath0 is anticorrelated with the duration of the lag , @xmath1 , according to @xmath2 ergs / sec .
this intriguing relationship is useful because it probes the properties of an ensemble of bursts and thereby can provide clues of the global dynamics of grbs .
it also has the potential for distance estimation to grbs .
several workers have attempted to explain the origin of the pulse lag , or so called pulse paradigm @xcite . @xcite were able to produce a pulse lag by modeling the burst as a series of internal shocks , but obtained timescales an order of magnitude too large .
however , as @xcite point out , the lag timescale is much longer than the synchrotron cooling time .
@xcite modeled the kinematics of internal shocks and found no evidence of pulse peak lag .
as of yet , the source of the pulse paradigm remains a mystery .
this paper does not attempt to explain the origin of the pulse lag , but assumes that it derives from a process common to all bursts and has some proper decay timescale @xmath3 in the reference frame of the emitter .
in this paper i will present an interpretation of the luminosity - lag relationship as being due to relativistic motion of the emitting region toward the observer .
thus it is a purely kinematic effect .
in @xcite a peak @xmath4-ray luminosity @xmath0 was defined from the observed peak _ number _ flux @xmath5 , a luminosity distance @xmath6 derived from the observed redshift @xmath7 and a given cosmology , and a mean emitted photon energy @xmath8 , constant for all bursts : @xmath9 however , one does not know the intrinsic energy of the photons emanating from the pulses
. therefore multiplication of all grb fluxes by an average photon energy @xmath8 , while useful to estimate grb peak luminosites , obfuscates the underlying relationships in the dynamics of grbs .
in other words , photon number flux @xmath5 is the observed quantity and thus photon number luminosity @xmath10 can be the only calculable luminosity . the luminosity distance @xmath6 is defined so that redshift of photons is accounted for . the photon number luminosity does not depend on photon energy redshift and thus is given by @xmath11 in figure [ numberluminosity ] is shown the peak number luminosity @xmath10 versus spectral lag @xmath1 for six bursts with known redshifts .
i fit the data by minimizing @xmath12 where @xmath13 .
the best fit for lags ccf31 0.1 ( see fig . [ numberluminosity ] ) is @xmath14 with @xmath15 . by comparison , a similar fit of @xmath0 from eqn ( [ lumeqn ] ) versus @xmath1 reproduces the @xcite result @xmath16 with @xmath17 . for ccf31 0.5
, @xmath18 with @xmath19 contrasts with @xmath20 with @xmath21 . with six datapoints ,
these are small number statistics to be sure .
however , the improvement in correlation @xmath10 vs. @xmath1 over that of @xmath0 vs. @xmath1 via reductions of @xmath22 by 25% or greater is clear .
the relation @xmath23 , where 0.98 is almost 1 , is a strikingly simple statistical property for an ensemble of grbs to obey . in fact
this relationship suggests a simple explanation .
i propose that this proportionality is a result of variations in line - of - sight relativistic velocities of grbs . for a burst expanding with lorentz factor @xmath24 , @xmath25 and at redshift @xmath7
, the observed number luminosity @xmath26 ( photons sec@xmath27 ) varies as @xmath28 where @xmath29 is the proper isotropic number luminosity for a non - expanding source .
this expression is valid for emission that appears isotropic to the observer ( i.e. if the emission is in the form of a jet , then the opening angle @xmath30 , where @xmath31 is the relativistic beaming angle ) .
also , as mentioned previously , if one assumes that the spectral pulse lag is due to some proper decay timescale @xmath3 , possibly due to cooling or deceleration , then we observe a lag @xmath32 for @xmath33 , where the angular dependence is removed by the strong effect of attenuation of number flux @xmath34 received from emission moving at inclination angles greater than @xmath31 ; @xmath35 .
thus one can focus one s attention only on emitters moving directly toward the observer . combining these then yields the inverse relationship @xmath36 this relationship is redshift independent and thus is valid for luminosity @xmath37 and spectral lag @xmath38 as would be observed local to the burst .
thus we have an explanation of the relation shown in figure [ numberluminosity ] .
i propose that the variety in the spectral lags and number luminosities of bursts depend solely upon the relativistic motion of the emitter along the line of sight .
certainly one expects variation among bursts to cause deviations from this relationship , however the current data suggests this effect to be dominant .
there are two possibilities for the nature of this variation ( see section [ sectionbeaming ] ) .
first , we are observing an intrinsic range of burst expansion speeds , ranging from fast bursts having little or no spectral lag to slower bursts having longer lags .
the second possibility is that emitting material is in the form of a jet with some angular dependence on the emitter s lorentz factor @xmath39 where the lorentz factor would have a peak at the centroid of the jet @xmath40 and monotonically decrease on some angular scale greater than @xmath31 for increasing inclination angles @xmath41 . in either case
we observe only the emission dominated by those emitters moving with trajectories inclined within an angle @xmath31 from our line - of - sight .
the above discussion allows the estimation of redshift @xmath7 from the observed peak number flux @xmath5 and the observed spectral lag @xmath42 . @xcite
also estimate redshifts using grb variability to indicate luminosity . from eqn ( [ numlumeqn ] )
one has @xmath43 where @xmath44 is the angular diameter distance ( e.g. * ? ? ?
* ) which is a function of the cosmology and redshift @xmath7 , but does not vary much over the interval @xmath45 .
for example , here i follow @xcite and choose ( @xmath46,@xmath47 ) = ( 0.3,0.7 ) , for which @xmath48 to within 10% over this range of @xmath7 , where @xmath49 is the hubble distance ( choose @xmath50 km sec@xmath27 mpc@xmath27 ) .
also , an observed spectral lag @xmath51 varies from the local spectral lag @xmath1 by @xmath52 one can put these components into the @xmath53 curve of bursts with known redshifts @xmath54 where @xmath55 photons as taken from the fit in figure [ numberluminosity ] . thus one can estimate a redshift for a given observed peak number flux and spectral lag @xmath56 where flux @xmath5 is measured in photons @xmath57 sec@xmath27 and @xmath51 is in seconds .
it is worth noting that a well determined cosmology gives a well determined @xmath44 and thus a refined redshift @xmath7 .
conversely , many independently determined redshifts might give cosmological information .
however , the primary source of error in this relation is not the uncertainty in @xmath44 , but in the deviations of burst luminosities and lags from the relation of eqn ( [ ntrelation ] ) as can be seen in figure [ numberluminosity ] .
in order for there to be a lag @xmath1 of the arrival of a pulse over an energy range fixed with respect to the observer ( e.g. the range between batse channels 1 and 3 ) that varies inversely with lorentz factor ( eqn [ dtobseqn ] ) , the peak energy of the pulse must decay exponentially as @xmath58 where @xmath59 is the peak energy in the emitter frame , @xmath60 is observer time , @xmath4 is the lorentz factor of the emitter and @xmath61 is the proper decay timescale in the emitter frame .
thus we infer an exponential decay law for the pulse peak energy .
it is important to note that this exponential evolution can be very rapid ( @xmath62 msec ) @xcite and is distinct from the evolution of the pulse envelope ( the so - called fred ) that has been studied by many authors ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
one outstanding question in the study of grb pulses is the apparent invariance among bursts of the peak energy @xmath63 of the @xmath64 flux spectrum ( @xmath65 a few @xmath66 kev ) where @xmath67 is number flux per photon energy .
one might expect this value to vary with burst luminosity , but it remains roughly constant ( to a factor of a few ) over a wide variety of ( long ) grbs .
if the pulse photon number flux @xmath68 ( which , for demonstration , we assume to be peaked at @xmath69 ) increases as the peak photon energy @xmath69 of eqn ( [ epkeqn ] ) decreases , then there will be a maximum in @xmath70 .
this may happen as follows : the nascent pulse emitter is extremely hot and compact .
as it evolves , it cools ( @xmath69 drops ) and it expands or becomes progressively more optically thin , releasing an ever increasing photon rate @xmath68 .
this is analogous to early - time supernova lightcurves @xcite . in this picture , uncovering the flux dependence @xmath71 will give insight into why @xmath72 constant .
for instance @xmath73 with eqn ( [ epkeqn ] ) demonstrates this behavior .
in this letter i have argued that the correlation of @xcite is between number luminosity and spectral lag and i interpret this as being due to the variety , among bursts , of relativistic velocities at which emitting regions move toward the observer .
suppositions as to the physical emission process have been deliberately avoided in order to highlight this purely kinematic effect .
there are two possible scenarios by which this might happen .
the first is simply that the variation is due to the variation of expansion lorentz factor among bursts . from eqn ( [ dtobseqn ] ) , this implies that the ratio of maximum to minimum spectral lags is the same as the ratio of maximum to minumum lorentz factors @xmath74 where observed lags are seen to range over roughly two orders of magnitude .
thus if the fastest bursts have @xmath75 , then the minimum bursts would have @xmath76 and thus may be only mildly relativistic .
the second possibility is that grb ejecta is directed in a jet such that the lorentz factor @xmath39 has some maximum @xmath40 and monotonically decreases with increasing angle .
thus the function @xmath39 will determine the relative numbers of bursts observed with given spectral lags .
@xcite has noted that the distribution of bursts is strongly peaked at small lags indicating that the lorentz factor of emitting ejecta , @xmath77 , is basically constant over the jet opening angle @xmath78 , with a fairly narrow edge region @xmath79 of decreasing @xmath4 .
thus let us estimate a characteristic opening angle @xmath78 by assuming that @xmath80 is constant for @xmath81 and decreases to order unity within a narrow boundary or `` edge '' region @xmath82 .
thus the solid angle subtended by `` face '' bursts ( i.e. observed at inclinations @xmath83 ) is @xmath84 , and that for `` edge '' bursts is @xmath85 and the ratio is : @xmath86 from fig . 3 of @xcite , one can estimate this ratio , by defining short lags @xmath87 sec as `` face '' and the longer lags as `` edge '' bursts , to be @xmath88 .
assuming @xmath89 , then this gives @xmath90 . with more statistics of the distribution of bursts along the curve of eqn .
( [ nteqn ] ) one can fit the function @xmath39 much more precisely . | reported by @xcite relationship between peak luminosity of gamma - ray bursts pulse time lag between energy channels peak luminosity @xmath0 anticorrelated with duration lag @xmath2 ergs / sec relationship probes properties of bursts clues global dynamics of grbs potential for distance estimation to grbs workers attempted explain origin of pulse lag pulse paradigm @xcite @xcite pulse lag burst as internal shocks obtained timescales too large lag timescale longer than synchrotron cooling time found no evidence of pulse peak lag source of pulse paradigm remains mystery paper origin lag assumes derives from process common to all bursts decay timescale in frame emitter interpretation luminosity - lag relationship due to relativistic motion of emitting region toward observer purely kinematic effect peak @xmath4-ray luminosity @xmath0 defined from observed peak _ _ flux @xmath5 luminosity distance @xmath6 from redshift @xmath7 cosmology mean emitted photon energy @xmath8 constant for all bursts know intrinsic energy of photons from pulsesmultiplication grb fluxes by average photon energy @xmath8 obfuscates relationships photon flux @xmath5 is observed quantity luminosity @xmath10 only calculable luminosity luminosity distance @xmath6 defined redshift accounted photon number luminosity on energy redshift given by @xmath11 peak luminosity @xmath10 versus spectral lag @xmath1 for six bursts with redshifts minimizing @xmath12 where @xmath13 best fit for lags ccf31 0.1 is @xmath14 with @xmath15 similar fit @xmath0 versus @xmath1 reproduces result @xmath16 with @xmath17 for ccf31 0.5 @xmath18 with @xmath19 contrasts with @xmath20 with @xmath21 six datapoints small number statistics improvement in correlation @xmath10 vs. @xmath1 via reductions of @xmath22 by 25% relation @xmath23 0.98 is almost 1 simple statistical property for explanation proportionality result of variations in velocities grbsburst expanding with lorentz factor @xmath24 redshift @xmath7 observed luminosity @xmath26 photons sec@xmath27 varies @xmath28 @xmath29 isotropic luminosity for non - expanding source valid for emission isotropic to observer jet opening angle @xmath30 @xmath31 relativistic beaming angle spectral pulse lag due to decay timescale @xmath3 cooling or deceleration lag @xmath32 @xmath33 angular dependence removed by attenuation of number flux @xmath34 from emission inclination angles greater than @xmath31 ; @xmath35 focus attention on emitters moving toward observer combining yields inverse relationship @xmath36 redshift independent valid for luminosity @xmath37 spectral lag @xmath38 local to burst explanation of relation in figure [ numberluminosity ] variety in spectral lags number luminosities of bursts depend relativistic motion emitter line sight variation among bursts deviations current data suggests effect dominant two possibilities for variation range of burst expansion speeds fast lag to slower longer lagssecond possibility emitting material jet angular dependence on emitter s lorentz factor @xmath39 factor peak at centroid jet @xmath40 decrease greater than @xmath31 for increasing inclination angles @xmath41 observe emission emitters inclined within angle @xmath31 from line - - sight discussion allows estimation redshift @xmath7 from peak number flux @xmath5 spectral lag @xmath42 estimate redshifts using grb variability luminosity eqn @xmath43 @xmath44 angular diameter distance function of cosmology redshift @xmath7 vary over interval @xmath45 @xmath46,@xmath47 ) = ( 0.3,0.7 ) @xmath48 within 10% over range @xmath7 @xmath49 hubble distance @xmath50 sec@xmath27 spectral lag @xmath51 varies from local spectral lag @xmath1 @xmath52 components into @xmath53 curve of bursts redshifts @xmath54 estimate redshift for peak number flux spectral lag @xmath56 flux @xmath5 measured in photons secondsdetermined cosmology gives @xmath44 refined redshift @xmath7 many determined redshifts might give cosmological information primary source error not uncertainty in @xmath44 but deviations of burst luminosities lags from relation eqn ( ntrelation in figure [ numberluminosity ] lag @xmath1 of arrival pulse over energy range between channels 1 3 with lorentz factor eqn dtobseqn peak energy pulse must decay exponentially as @xmath58 @xmath59 is peak energy in emitter frame @xmath60 observer time @xmath4 lorentz factor emitter @xmath61 proper decay timescale in emitter frame exponential decay law for pulse peak energy exponential evolution rapid ( @xmath62 msec distinct from evolution pulse envelope fred ) studied by authors outstanding question pulses is invariance among bursts of peak energy @xmath63 of @xmath64 flux spectrum ( @xmath65 @xmath66 kev ) @xmath67 is number flux per photon energy value vary with burst luminosity remains roughly constant over long grbs pulse photon flux @xmath68 peaked at @xmath69 increases peak photon energy @xmath69 decreases maximum in @xmath70 nascent pulse emitter hot compact evolves cools @xmath69 expands becomes optically thin releasing increasing photon rate @xmath68 analogous to early - time supernova lightcurves @xcite uncovering flux dependence @xmath71 @xmath72 constant @xmath73 with eqn epkeqn demonstrates behavior correlation of @xcite between number luminosity spectral lag due to variety of relativistic velocities emitting regions move toward observer suppositions physical emission process avoided highlight kinematic effect two scenarios first variation due to variation expansion lorentz factor among bursts ratio of maximum to minimum spectral lags same as ratio maximum to minumum lorentz factors @xmath74 lags range over two orders of magnitude fastest bursts have @xmath75 minimum bursts have @xmath76 mildly relativistic second possibility grb ejecta directed in jet lorentz factor @xmath39 maximum @xmath40 decreases with increasing anglefunction @xmath39 numbers bursts spectral lags noted distribution bursts peaked at small lags lorentz factor @xmath77 constant over jet opening angle @xmath78 narrow edge region @xmath79 decreasing @xmath4 estimate opening angle @xmath78 @xmath80 constant for @xmath81 decreases within narrow edge region @xmath82 solid angle subtended by face bursts at inclinations @xmath83 is @xmath84 edge '' bursts @xmath85 ratio @xmath86 fig estimate ratio defining short lags @xmath87 as face '' longer lags edge '' bursts @xmath88 assuming @xmath89 gives @xmath90 more statistics distribution bursts along curve eqn function @xmath39 precisely | 0.433508 | 2,796 | 1,461 | , Saving $0.1 in GPT-4. |
154,805 | continuing our work from paperi ( southworth et al . , 2006 ) we present medium - resolution spectroscopy and broad - band photometry of seven cataclysmic variables ( cvs ) discovered by the sloan digital sky survey . for six of these objects
we derive accurate orbital periods , all which are measured for the first time . for sdssj013132.39@xmath0090122.2 , which contains a non - radially pulsating white dwarf , we find an orbital period of @xmath1min and a low radial velocity variation amplitude indicative of an extreme mass ratio . for sdssj205914.87@xmath0061220.4
, we find a period of @xmath2min .
this object is a dwarf nova and was fading from its first recorded outburst throughout our observations .
int photometry of sdssj155531.99@xmath0001055.0 shows that this system undergoes total eclipses which are 1.5mag deep and occur on a period of @xmath3min .
a not light curve of sdssj075443.01@xmath4500729.2 shows that this system is also eclipsing , on a period of @xmath5min , but here the eclipses are v - shaped and only 0.5mag deep .
its low emission - line strengths , orbital period and v - shaped eclipse unambiguously mark it as a novalike object .
wht photometry of sdssj005050.88@xmath4000912.6 and sdssj210449.94@xmath4010545.8 yields periods of @xmath6 and @xmath7min , respectively .
photometry of the seventh and final system , sdssj165658.12@xmath4212139.3 , shows only flickering .
our results strengthen the conclusion that the faint magnitude limit of the sdss spectroscopic database implies that the sample of cvs contained in it has quite different characteristics to previously studied samples of these objects .
five of the six orbital periods measured here are shorter than the observed 23hr cv period gap .
two systems have periods very close to the minimum orbital period for hydrogen - rich cvs .
stars : novae , cataclysmic variables stars : binaries : close stars : binaries : eclipsing stars : binaries : spectroscopic stars : white dwarfs stars : dwarf novae stars :
individual : sdss j005050.88@xmath4000912.6 stars : individual : sdss j013132.39@xmath0090122.2 stars : individual : sdssj075443.01@xmath4500729.2 stars : individual : sdss j155531.99@xmath0001055.0 stars : individual : sdss j165658.12@xmath4212139.3 stars : individual : sdss j205914.87@xmath0061220.4 stars : individual : sdss j210449.94@xmath4010545.8 | cataclysmic variables ( cvs ) are interacting binary stars containing a white dwarf primary star in a close orbit with a low - mass secondary star which fills its roche lobe . in most of these systems
the secondary component is unevolved and is transferring material to the white dwarf primary via an accretion disc .
comprehensive reviews of the properties of cvs have been given by @xcite and @xcite .
the evolution of cvs is governed primarily by angular momentum loss ( aml ) from the binary orbit .
this is important during the phase of common - envelope evolution , where an initially wide orbit is dramatically shrunk due to the ejection of the envelope .
the efficiency of this process is poorly understood , although recent advances @xcite are promising ( see @xcite for further discussion ) . for many objects
this results in a close but detached binary system .
further aml causes the binary to become semidetached and undergo mass transfer .
this aml is thought to occur mainly in two ways : by the emissison of gravitational radiation @xcite and due to magnetic braking via the stellar wind @xcite which acts on the secondary star @xcite .
according to the standard theory of cv evolution , magnetic braking is the dominant aml mechanism for the longer - period systems but abruptly terminates when the secondary star becomes completely convective .
this disruption occurs when the orbital period has decreased to approximately three hours @xcite .
the distended secondary star is then able to shrink until it returns to thermal equilibrium , so no longer fills its roche lobe .
this disrupted magnetic braking scenario was formulated to explain the 23hr ` gap ' in the orbital period distribution of cvs @xcite , which contains far fewer systems than which occur at adjacent periods .
whilst the cv is in the period gap , aml continues more slowly via gravitational radiation until a period of about two hours , at which point the roche lobe of the secondary shrinks onto the star s surface and mass transfer recommences .
aml continues to shorten the orbital period until a minimum is reached at about 76min @xcite , where structural changes in the secondary star mean that further aml causes the orbital period to increase again .
this results in faint ` period bounce ' systems with very low mass transfer rates @xcite .
theoretical studies of the evolution of cvs @xcite are able to reproduce some of the features of the observed orbital period distribution of cvs @xcite , but there are several discrepancies between the predicted and observed distributions @xcite .
firstly , it is predicted that the vast majority of cvs have orbital periods shorter than 3hr , whereas in the observed sample the numbers of systems above and below this value are roughly equal .
secondly , the predicted minimum period is around 65min and there should be a large number of systems at this point due to the very slow rate of change of period with time .
the minimum period in the observed cv sample is at approximately 76min ( excluding a small number of systems which contain evolved secondary stars ) and the observational sample does not contain a ` spike ' at this period .
thirdly , population synthesis studies consistently find that at the present age of the galaxy about 70% of cvs should have evolved beyond the minimum period .
whilst there are over six hundred known cvs @xcite there are very few candidate period bouncers ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , and there are only _ two _ systems with a measured secondary star mass which is definitely below the hydrogen - burning limit : the eclipsing cvs sdssj103533.02 + 055158.3 @xcite and sdssj150722.30 + 523039.8 @xcite .
a further problem with the standard disrupted magnetic braking evolutionary scenario is that there are many observations of very - low - mass fully convective stars which have strong magnetic fields @xcite , and no detections of a sudden cessation of magnetic braking in mdwarfs below a certain mass @xcite .
the known population of cvs is extremely heterogeneous due to the number of different ways in which individual objects have been discovered @xcite .
most of the methods of discovery are strongly biased towards detecting longer - period systems with higher mass transfer rates , because these are on average intrinsically brighter and bluer than short - period cvs .
these effects mean that when comparing the observed population of cvs to theoretical predictions it is currently impossible to determine whether any discrepancies are due to the predictions being incorrect or to the large selection biases affecting the observed sample . the selection effects affecting observational studies of cvs have been considered in detail by @xcite for the case of the palomar - green survey @xcite .
the selection biases in the known population of cvs are so strong and complex that this problem can only be reliably solved by studying a sample of cvs which has been homogeneously selected and is not strongly biased towards intrinsically bright or variable systems .
recent studies of an homogeneous sample of cvs have included the population of objects spectroscopically identified in the hamburg quasar survey ( hqs ; * ? ? ?
* ) , which is a complete magnitude - limited sample conducted using objective prism observations .
@xcite have found that the hqs cvs are on average photometrically less variable than the overall sample of known cvs , confirming that spectroscopic identification is important for avoiding some selection biases .
these authors also found an unexpected deficiency of short - period dwarf novae in their sample , resulting in a low number ratio of short - period to long - period systems for the newly - discovered hqs cvs .
the corresponding space density of short - period hqs cvs is much lower than theoretically predicted ( e.g. * ? ? ?
* ; * ? ? ?
another sample arises from the 2df quasar survey , which has serendipitously identified a number of cvs via their spectroscopic characteristics .
@xcite studied six of these objects and presented evidence that four of them had short periods , but the size of this sample must be increased before any detailed conclusions can be drawn .
[ cols="<,<,<,^,^,^,^,^,^,^",options="header " , ] we are engaged in a project to categorise the sample of cvs discovered by the sdss , the primary goal being measurement of the orbital period and other available quantities for each object . in this work
we have presented spectroscopy and photometry from four telescopes , and measured the orbital periods of six sdss cvs ( table[tab : result ] ) . of the six cvs for which we have measured orbital periods ,
two have values close to the minimum period for cvs with unevolved secondary stars : sdssj0050 with 80.3min and sdssj0131 with 80.54min .
three objects have slightly longer periods but still qualify as short - period systems : sdssj1555 with 113.54min , sdssj2059 with 107.518min and sdssj2104 with 103.62min .
we observed sdssj2059 during the decline from its first recorded outburst so are able to classify it as a dwarf nova .
only one system , sdssj0754 , has an orbital period ( @xmath5min ) which is longer than the 23hr period gap .
it also has a high mass transfer rate .
the orbital velocity amplitude for sdssj0131 is only @xmath8kms@xmath9 .
using several constraints we find that this system has an extremely low mass ratio and contains a massive white dwarf and/or a very low - mass secondary star .
the mass donor is a good candidate for being a cv secondary star with a mass below the minimum required for core hydrogen burning .
our photometry of sdssj0754 and sdssj1555 show that they are eclipsing systems with 0.5mag and 1.5mag deep eclipses ( in unfiltered light ) , respectively .
sdssj1555 is a good candidate for accurate measurement of the absolute properties of a short - period cv , as the functional form of its eclipse is well understood ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
a similar analysis for sdssj0754 is not currently possible due to the complexities of the novalike class of cvs .
a recurrent theme in the previous paragraphs is that the orbital periods we are finding for the sdss cvs are almost all shorter than the 23hr period gap , and in many cases adjacent to the observed 76min minimum period .
this is consistent with the expectation that shorter - period cvs are not only intrinsically fainter but more populous that their longer - period counterparts @xcite .
moreover , the orbital period distribution of those sdss cvs with known periods is much closer to theoretical predictions than the distribution for the complete sample of known cvs .
we would like to warmly thank david boyd and tom krajci for their speed in responding to our request for observations of sdssj2059 after our spectroscopic observations indicated that it might be in outburst .
we are also grateful for a positive and timely review from the anonymous referee .
this work is based on observations made with the william herschel telescope and the isaac newton telescope , which are operated on the island of la palma by the isaac newton group in the spanish observatorio del roque de los muchachos ( orm ) of the instituto de astrofsica de canarias ( iac ) .
observations were also made with the nordic optical telescope , operated on the island of la palma jointly by denmark , finland , iceland , norway , and sweden , also in the orm of the iac .
these observations used alfosc , which is owned by the instituto de astrofsica de andaluca ( iaa ) and operated at the nordic optical telescope under agreement between iaa and the nbifafg of the astronomical observatory of copenhagen . js acknowledges financial support from stfc in the form of a postdoctoral research assistant position .
btg acknowledges financial support from pparc in the form of an advanced fellowship .
trm was supported by a pparc senior fellowship during the course of this work .
ddm ddm acknowledgs financial support by asi and inaf via grant i/023/05/0 .
aa thanks the royal thai government for a studentship .
the following internet - based resources were used in research for this paper : the eso digitized sky survey ; the nasa astrophysics data system ; the simbad database operated at cds , strasbourg , france ; and the ar@xmath11iv scientific paper preprint service operated by cornell university . funding for the sloan digital sky survey ( sdss ) has been provided by the alfred p. sloan foundation , the participating institutions , the national aeronautics and space administration , the national science foundation , the u. s. department of energy , the japanese monbukagakusho , and the max planck society .
the sdss website is http://www.sdss.org/. | cataclysmic variables interacting binary stars white dwarf primary star close orbit with low - mass secondary star roche lobe secondary unevolved transferring material to white dwarf primary via accretion disc reviews properties cvs by @xcite and @xcite evolution governed by angular momentum loss aml ) from binary orbit important during common - envelope evolution wide orbit shrunk due to ejection envelope efficiency poorly understood advances @xcite promising results in close detached binary system aml causes binary semidetached mass transfer aml gravitational radiation @xcite magnetic braking via stellar wind @xcite secondary star magnetic braking dominant aml mechanism for terminates when secondary star becomes convective disruption occurs when orbital period decreased to three hours @xcite distended secondary star until to thermal equilibrium no longer fills roche lobe disrupted magnetic braking scenario 23hr gap in orbital period distribution of cvs @xcite fewer systems than adjacent periods aml continues slowly via gravitational radiation until two hours roche lobe secondary shrinks star s surface mass transfer recommencesaml orbital period until minimum 76min @xcite structural changes in secondary star orbital period results in faint period bounce systems with low mass transfer rates @xcite theoretical studies evolution of cvs @xcite reproduce observed orbital period distribution discrepancies between predicted and observed distributions majority of cvs have orbital periods shorter than 3hr observed sample systems above and below equal predicted minimum period around 65min large number of systems due to slow change period minimum period in observed cv sample approximately 76min observational sample spike at this period population synthesis studies find 70% of cvs should evolved beyond minimum period over six hundred known cvs @xcite few candidate period bouncers only two systems with secondary star mass below hydrogen - burning limit eclipsing cvs sdssj103533.02 + 055158.3 @xcite and sdssj150722.30 + 523039.8 @xcite problem with disrupted magnetic braking evolutionary scenario many observations of - low - mass convective stars strong magnetic fields @xcite no detections sudden cessation of magnetic braking in mdwarfs below certain mass @xciteknown population of cvs heterogeneous due to different discovered methods discovery biased towards longer - period systems with higher mass transfer rates brighter bluer than short - period cvs comparing observed population cvs to theoretical predictions impossible to determine discrepancies due to predictions or large selection biases sample selection effects affecting observational studies considered by for palomar - green survey selection biases in cvs strong complex solved by studying sample homogeneously selected not biased towards bright or variable systems recent studies of homogeneous sample included spectroscopically identified in hamburg quasar survey found hqs cvs photometrically less variable than overall sample spectroscopic identification important for avoiding selection biases found unexpected deficiency of short - period dwarf novae in sample low number ratio of short - period to long - period systems for newly - discovered hqs cvs space density of short - period hqs cvs lower than theoretically predicted another sample from 2df quasar survey serendipitously identified cvs via spectroscopic characteristics@xcite studied six objects four had short periods size sample must before detailed conclusions engaged in project to categorise sample cvs sdss goal measurement orbital period quantities for each object presented spectroscopy photometry from four telescopes measured orbital periods of six sdss cvs two values close to minimum period for sdssj0050 80.3min sdssj0131 80.54min three objects longer periods short - period systems sdssj1555 113.54min sdssj2059 107.518min sdssj2104 103.62min observed sdssj2059 during decline as dwarf nova one system sdssj0754 orbital period @xmath5min longer than 23hr period gap high mass transfer rate orbital velocity amplitude for sdssj0131 @xmath8kms@xmath9 system low mass ratio contains massive white dwarf low - mass secondary star mass donor good candidate for cv secondary star mass below minimum for core hydrogen burning photometry of sdssj0754 sdssj1555 show eclipsing systems with 0.5mag 1.5mag deep eclipsessdssj1555 good candidate for accurate measurement properties short - period cv functional form eclipse understood similar analysis for sdssj0754 not possible due to complexities novalike class cvs orbital periods for sdss cvs shorter than 23hr period gap adjacent to 76min minimum period consistent with expectation shorter - period cvs fainter more populous longer - period counterparts orbital period distribution of sdss cvs with known periods closer to theoretical predictions than complete sample known cvs thank david boyd tom krajci for responding to request for observations sdssj2059 after spectroscopic observations outburst grateful for positive review from anonymous referee work based on observations with william herschel telescope isaac newton telescope operated on island of la palma newton group spanish observatorio del observations with nordic optical telescope operated by denmark finland iceland norway sweden observations used alfosc owned by instituto de astrofsica de andaluca operated at nordic optical telescope under agreement between iaa nbifafg astronomical observatory of copenhagenjs support stfc postdoctoral assistant btg support pparc advanced fellowship trm supported pparc senior fellowship ddm support asi inaf grant i/023/05/0 aa thanks royal thai government studentship internet resources used eso digitized sky survey nasa astrophysics data system simbad database ar@xmath11iv scientific paper preprint service cornell university funding sloan digital sky survey provided alfred p. sloan foundation participating institutions national aeronautics space administration national science foundation u. s. department energy japanese monbukagakusho max planck society website http://www.sdss.org | 0.357293 | 2,357 | 1,180 | , Saving $0.1 in GPT-4. |
154,822 | we present observations which probe the small - scale structure of the interstellar medium using diffuse interstellar bands ( dibs ) . towards hd 168075/6 in the eagle nebula , significant differences in dib absorption
are found between the two lines of sight , which are separated by 0.25 pc , and @xmath0 exhibits a velocity shift .
similar data are presented for four stars in the @xmath1 sgr system .
we also present a search for variations in dib absorption towards @xmath2 vel , where the atomic lines are known to vary on scales of @xmath3 au .
observations separated by @xmath4 yr yielded no evidence for changes in dib absorption strength over this scale , but do reveal an unusual dib spectrum . | early theories of the diffuse interstellar medium predicted that there should be no significant structure on scales below @xmath5 pc .
however , observations have shown that there can be large variations in interstellar column densities between lines of sight separated by as little as @xmath6 au ( @xmath7 pc ) .
previous observations have relied on narrow atomic and molecular lines , requiring high - resolution observations at high signal - to - noise , which is very demanding in terms of telescope time .
dibs have recently been shown to exhibit similar variations on small scales @xcite , and can be observed at lower resolution . they can be used to identify promising lines of sight for high - resolution follow - up and to probe the spatial distribution of the dib carriers . by comparing the variations seen in different dibs , it may also be possible to further constrain theories of the carriers .
as a proof of concept , we observed dibs towards four stars in the @xmath1 sgr system ( @xmath1 sgr a , d , e and f ) . using the robert stobie spectrograph ( rss ) on the southern african large telescope ( salt ) at a resolving power @xmath9
, we were able to obtain high quality spectra of dibs even in short integration times . in fig .
[ fig : msgr ] significant differences are seen between dibs towards @xmath1 sgr e and d , which are separated by @xmath10 pc .
full results are presented in @xcite .
and @xmath0 dibs towards four stars of the @xmath1 sgr system .
top panel shows the dibs towards @xmath1 sgr d , whilst the other three panels are the spectra of @xmath1 sgr e , a and f. respectively , each divided by the spectrum of @xmath1 sgr d. dotted lines illustrate the dib rest wavelengths from @xcite . reproduced from @xcite , with permission.,scaledwidth=70.0% ]
hd 168075 and hd 168076 are a pair of o - type stars which provide most of the uv photons that excite the eagle nebula . using rss on salt , we observed hd 168075/6 at a resolving power of @xmath11 , with both stars on the slit simultaneously .
the integration time was just 570 s , despite thin cloud , poor seeing , and a narrow ( @xmath12 ) slit .
and @xmath0 dibs .
upper panel : reduced spectra .
middle panel : after removal of the stellar c iv line at 5802 to provide a clean continuum .
lower panel : ratio of the two spectra .
dotted lines illustrate the dib rest wavelengths from @xcite .
@xmath13 is stronger towards hd 168076 , whilst @xmath0 exhibits a velocity shift .
reproduced from @xcite , with permission.,scaledwidth=70.0% ] fig .
[ fig : hd168 ] shows the @xmath13 and @xmath0 dibs towards hd 168075/6 .
there are highly significant variations between the two lines of sight .
@xmath13 is stronger towards hd 168076 , whilst @xmath0 appears to shift in velocity .
full results for the @xmath14 , 5797 , 5850 , 6196 , 6203 , 6269 , 6283 and 6613 dibs are presented in @xcite .
the interstellar absorption lines towards @xmath2 velorum are known to be time variable . for example , the equivalent width of the k i 7698 line has increased by @xmath16 between 1994 and 2006 , during which proper motion of the star has carried it 30 au perpendicular to the line of sight ( fig . [
fig : kvel ] ) .
vel between 1994 and 2006 .
right : the @xmath13 dib towards @xmath2 velorum in 1995 and 2004 .
@xmath175797 is not detected at either epoch , and there is no evidence for changes in @xmath13 between the two epochs .
both panels reproduced from @xcite , with permission.,title="fig:",scaledwidth=55.0% ] vel between 1994 and 2006 .
right : the @xmath13 dib towards @xmath2 velorum in 1995 and 2004 .
@xmath175797 is not detected at either epoch , and there is no evidence for changes in @xmath13 between the two epochs .
both panels reproduced from @xcite , with permission.,title="fig:",scaledwidth=45.0% ] using observations from mount stromlo in 1995 and the anglo australian telescope in 2004 , we searched for variations in the dibs towards this star .
none were found , with an upper limit of @xmath18 on changes in @xmath13 ( fig .
[ fig : kvel ] ) . the @xmath19 equivalent width ratio observed in 2004 is at least @xmath20 ( @xmath21 ) , which is the highest ever reported .
full results have been published in @xcite .
significant differences in dib absorption have been found towards the @xmath1 sgr and hd 168075/6 systems , on scales of less than a parsec .
these results confirm the discovery by @xcite that dibs can be used as probes of small - scale interstellar structure , and demonstrate that moderate - resolution observations of dibs may be used to identify promising targets for high - resolution studies .
although the observations of @xmath2 vel found no variation in dibs on scales of @xmath22 au , the dib spectrum itself was found to be unusual , with a very high @xmath19 ratio . | theories interstellar medium predicted no significant structure below @xmath5 pc observations shown large variations in interstellar column densities between lines separated @xmath6 au ( @xmath7 pc observations relied on narrow atomic lines high - resolution observations high signal - - noise demanding telescope time dibs exhibit variations on small scales @xcite observed at lower resolution identify promising lines high - resolution probe spatial distribution dib carriers comparing variations dibs constrain theories observed dibs towards four stars in @xmath1 sgr system stobie spectrograph large telescope resolving power @xmath9 high quality spectra of dibs short integration times fig differences between dibs towards @xmath1 sgr e and d separated by @xmath10 pc full results in @xcite @xmath0 dibs towards four stars sgr top panel shows dibs towards @xmath1 sgr d spectra @xmath1 sgr e a f divided by spectrum @xmath1 sgr d dotted lines illustrate dib rest wavelengths from @xcite reproducedhd 168075 168076 o - type stars provide uv photons eagle nebula rss observed hd 168075/6 resolving power @xmath11 both stars slit simultaneously integration time 570 s thin cloud poor seeing narrow @xmath12 slit @xmath0 dibs upper reduced spectra middle removal stellar c iv line at 5802 clean continuum lower ratio two spectra dotted lines dib rest wavelengths @xmath13 stronger towards hd 168076 @xmath0 velocity shift fig @xmath13 @xmath0 dibs towards hd 168075/6 significant variations between lines @xmath13 stronger towards hd 168076 @xmath0 velocity results for @xmath14 5797 5850 6196 6203 6269 6283 6613 dibs in @xcite interstellar absorption lines towards @xmath2 velorum time variable width k 7698 line increased by @xmath16 between 1994 2006 au perpendicular line sight @xmath13 dib towards @xmath2 velorum in 1995 2004 @xmath175797 not detected epoch no evidence changes in @xmath13 between two epochspanels from @xcite=55.0% 1994 2006 @xmath13 dib towards @xmath2 velorum in 1995 2004 @xmath175797 detected epoch no evidence changes @xmath13 epochs panels @xcite=45.0% observations mount stromlo 1995 anglo australian telescope 2004 variations dibs towards star none found upper limit @xmath18 on changes @xmath13 @xmath19 equivalent width ratio 2004 @xmath20 ( @xmath21 ) highest reported results published in @xcite differences in dib absorption found @xmath1 sgr hd 168075/6 systems less than parsec results confirm dibs probes small - scale interstellar structure moderate - resolution observations identify targets for high - resolution studies observations @xmath2 vel no variation dibs on scales @xmath22 au dib spectrum unusual high @xmath19 ratio | 0.186279 | 1,261 | 643 | , Saving $0.0 in GPT-4. |
154,830 | the raw light curves of both kic 10684673 and kic 12216817 show variability .
both are listed in the kepler eclipsing binary catalog ( hereafter kebc ) , however both are flagged as uncertain in nature . in the present study
we show their light curves can be modeled by considering each target as single , multi - modal @xmath0 scuti pulsator . while this does not exclude the possibility of eclipsing systems , we argue , while spectroscopy on the systems is still lacking , the @xmath0 scuti model is a simpler explanation and therefore more probable .
+ 2 | while the original kepler mission was in operation , it continually monitored over 150,000 stars in a region of sky containing parts of the cygnus and lyre constellations . with unprecedented photometric precision
, a great amount of work has been done to categorize , model , and determine the nature of the many new variables discovered ( for an overview of mission characteristics and objectives , see borucki ( 2008 ) ; for an overview of the kepler asteroseismology group , see gilliland ( 2010 ) .
two categories which have had several members added as a result of the mission are eclipsing binaries and intrinsic pulsating variables .
however , as with any rapid increase of such categories , the potential for initial misclassification of a system is possible .
both targets are listed in the kebc ( 10684673 with a period of 0.1925563 days and 12216817 with a period of 0.2462731 days ; kirk et al .
2016 ) , however both are flagged as uncertain in nature .
the light curves of both targets are shown in figure 1 .
while both light curves show variability , they do not share characteristics of classical eclipsing binaries , such as algol type systems .
for example , there are no obvious eclipse features .
rather , the structures in the light curve seem to be a superposition of several sinusoidal signals . as will be shown below
, the frequencies and amplitudes of all the dominant frequencies in each target are consistent with scuti - type pulsations .
one specific class of eclipsing binaries , w uma eclipsing binaries , have the same depth for both the primary and secondary eclipse , and can therefore produce , under the right circumstances , a sinusoidal signal . with the presence of the additional signals , there are two logical possibilities .
either the targets could be w uma systems , in which one of the components is also pulsating , or the light curve can be interpreted as a classic single @xmath0 scuti pulsating in multiple modes .
+ w uma eclipsing binaries are characterized by near equal minima , variability as low as a few tenths of a magnitude , and short periods of about 0.25 to about 1.0 days .
because the periods are short , most w uma variables are assumed to be near contact or overcontact binaries .
w uma systems can range from a through k spectral types .
see binnendijk ( 1970 ) and li et al .
( 2007 ) for an overview of w uma systems .
in this study , we use exclusively the one short - cadence ( sc ) data set obtained by the kepler space telescope and retrieved from the mast archive(aura 2015 ) .
sc data sets are approximately one month in duration ( rather than a full three - month quarter ) yet with the shortened image collection and integration times , data sets regularly have more than 40,000 data points ( as opposed to 7,000 data points for each long cadence data set ) . because the sc data sets have such short time intervals , they are the most useful for studying @xmath0 scuti - type variability .
both targets have one sc data set publicly available .
kepler began sc data collection for kic 10684673 at 12:15:49 on 2009 - 05 - 13 and ended at 11:47:11 2009 - 06 - 15 ( taken during quarter 1 of observations ) .
the start and end collection times for kic 12216817 are 20:38:32 2009 - 08 - 20 and 23:23:42 2009 - 09 - 23 , respectively ( taken during quarter 2 of the observational cycle ) . the data set for kic 10684673 had a total of 49,030 usable datapoints , while kic 12216817 had a total of 38271 , each after the removal of the outliers .
sample pdcsap flux light curves ( after the removal of the null data points when the telescope was offline ) are shown in figure 1 . as can be seen from figure 1 , no obvious eclipse features are prominent in the light curves of either target .
the kebc gives an orbital period of 0.1925563 days for kic 10684673 , and an orbital period of 0.2462731 days for kic 1226817 .
both targets , while in the catalog , carry an unc flag for uncertain whether the object is an eclipsing binary . +
to prepare the data for analysis , the first step taken involved removing any obvious outlier data points . secondly , the fluxes were converted to a magnitude scale via the equation@xmath1 where m is the magnitude , and f is the flux of each individual pdcsap measurement . for convenience ,
the conventional minus sign was omitted , as we are only interested in @xmath2 m ; the minus sign is unnecessary and this equation preserves the orientation of the data as presented in figure 1 . once converted to the magnitude scale ,
the data were frequency - searched using the period04 software package ( lenz and breger 2005 ) between the values of 3 c / d and 50 c / d , the typical range of frequencies for @xmath0 scuti variables .
the power spectra for both targets are shown in figure 2 in terms of the amplitude in magnitudes .
as can be seen , both targets have several signals in their respective power spectra . to model the data , only frequencies with a signal - to - noise ratio of @xmath3 5
were used to save on computational time .
table 1 gives the values of the frequencies used for the modeling and their associated uncertainties .
+ [ cols="^,^,^,^,^",options="header " , ]
to model the data , we performed a fitting routine in period04 using the frequencies of table 1 . from the fit ,
the amplitudes and phases were determined ( along with their uncertainties ) .
these parameters were then taken as the initial estimates for our model which was then inserted into a reduced @xmath4 calculation , as outlined in bevington and robinson ( 2003 ) .
the reduced @xmath4 can be written as @xmath5 where n is the number of data points , y@xmath6 represents the observed magnitudes associated with each observation time , @xmath7 represents the uncertainty associated with the measurements , and y(t@xmath6 ) represents the value of the model at each observation time .
each pdcsap flux measurement comes with an associated uncertainty , which must be converted to the uncertainties on the magnitudes . to convert the uncertainties from fluxes to magnitudes , we use the usual procedure @xmath8 where @xmath9 and @xmath10 are the uncertainties on the magnitudes and fluxes , respectively .
the model y(t@xmath6 ) in the reduced @xmath4 calculation are given by@xmath11 where @xmath12 , @xmath13 , and @xmath14 are the amplitude , angular frequency , and phase of each sine wave given and @xmath15 is the total number of frequencies for each object used in the study ( coincidentally @xmath15 = 5 in both cases ) .
it should be noted that in the case of kic 10684673 , there is a definite systematic and positive trend in the data . the entire data set was fit with a linear function to determine the slope and intercept of the trend .
finally , it is customary to define the reduced @xmath4 function as the original @xmath4 function divided by the difference between the number of data points and the number of degrees of freedom .
however , because the number of data points for both target are @xmath16 40,000 , and the number of parameters ( p ) being fit for both targets are p < 10 , the difference becomes @xmath17 .
+ the model was refined by running a nonlinear least - squares fit using a marquardt - levenburg algorithm .
the routine allowed the amplitudes and phases to vary while holding the frequencies of table 1 fixed until the set of amplitudes and phases which minimized the reduced @xmath4 were found . in the case of kic 10684673 , the linear trend was also incorporated into the routine .
the values of the parameters ( and associated uncertainties ) obtained from the fitting procedure are given in table 2 , as is the overall value for the reduced @xmath4 calculation .
the results of the fitting for both targets are shown in figure 3 .
this shows representative samples of the fit over the entire sc data set for both targets in order to show the quality of the fit with the data .
+ ideally , the expectation value of the reduced @xmath4 for a very good fit would be very close to one ( bevington and robinson , 2003 ) . as can be seen from table 2 , our values are somewhat high .
however , this can be explained by the extreme smallness of the photometric uncertainties associated with each measurement .
they are typically about 3 or more orders of magnitude smaller than the measurements . this being the case , even though our reduced @xmath4 values seem a little high for typical photometric measurements , kepler s precision is very much atypical
. therefore we conclude , given how well the data is replicated by the model , that the fit is satisfactory , especially since the model replicates every major feature throughout the entire data set .
it is also at this point that we can feel confident , given the quality of the models , that the frequencies reported in table 1 are genuine signals and not spurious .
in this study we have demonstrated that the sc light curves of kic 10684673 and kic 12216817 can both be modeled as single @xmath0 scuti - type pulsating systems pulsating in several modes . both
are included in the kebc , however both are flagged as uncertain in nature . while our analysis does not completely rule out the possibility of either of these systems being eclipsing binaries , this study gives strong support for the hypothesis that all of the variation in each case can be explained by a single , pulsating object . to completely determine the nature of the system
, spectroscopy would need to be performed on both targets .
+ 1 association of universities for research in astronomy , inc .
2015 , mikulski archive for space telescopes ( mast ; https://archive.stsci.edu/ ) bevington , p. r. , robinson , d. k. 2003 , data reduction and error analysis for the physical sciences , ed .
bevington , p. r. , robinson , d. k. ( boston , ma : mcgraw - hill ) binnendijk , l. 1970 , va , 17 , 217b .
borucki , w. , et al .
2008 , in exoplanets : detection , formation and dynamics , eds .
sun , s. f. mello , and j .- l .
zhou , iau symp .
249 , cambridge univ . press , cambridge , 17 .
breger , m. 2000 , in delta scuti and related stars , reference handbook and proceedings of the 6th vienna workshop in astrophysics , eds . m. breger and m. h. montgomery , asp conf .
210 , astron .
pacific , san francisco , 3 .
gilliland r. l. et al . , 2010 ,
pasp , 122 , 131 kirk , b. et al . , 2016 ,
aj , 151 , 68k lenz p. , breger m. 2005 , coast , 146 , 53 li l. , zhang f. , han z. , jiang d. , jiang t. , 2008 , mnras , 387 , 97 . | original kepler mission monitored over 150,000 stars in cygnus lyre constellations unprecedented photometric precision work done to categorize model determine nature of new variables discovered overview mission characteristics objectives see borucki ( 2008 ) kepler asteroseismology group see gilliland ( 2010 ) two categories members added are eclipsing binaries intrinsic pulsating variables potential for initial misclassification possible both targets listed in kebc ( 10684673 0.1925563 days and 12216817 0.2462731 days 2016 ) both flagged as uncertain in nature light curves of targets shown in figure 1 show variability share characteristics of classical eclipsing binaries no obvious eclipse features structures light curve superposition of sinusoidal signals frequencies amplitudes dominant frequencies target consistent with scuti - type pulsations class eclipsing binaries w uma eclipsing binaries have same depth for primary secondary eclipse can produce sinusoidal signal additional signals two possibilities targets be w uma systems pulsating or light curve as classic single @xmath0 scuti pulsating in multiple modesw uma eclipsing binaries near equal minima variability tenths magnitude short periods 0.25 to 1.0 days variables assumed near contact or overcontact binaries systems range a through k spectral types see binnendijk 1970 li et al. 2007 overview w uma systems study use one short - cadence sc data set kepler space telescope from mast archive 2015 sc data sets one month duration shortened image collection integration times more 40,000 data points 7,000 long cadence sc data sets short useful for studying @xmath0 scuti - type variability both targets have one sc data set kepler began sc data collection for kic 10684673 12:15:49 2009 - 05 - 13 ended 11:47:11 2009 - 06 - 15 start end collection times for kic 12216817 20:38:32 2009 - 08 - 20 23:23:42 2009 - 09 - 23 data set kic 10684673 49,030 datapoints kic 12216817 38271 after sample flux light curves shown in figure 1 no obvious eclipse features in light curves kebc orbital period 0.1925563 days for kic 10684673 0.2462731 days for kic 1226817targets carry unc flag for uncertain eclipsing binary data for analysis step removing outlier data points fluxes converted to magnitude scale via equation@xmath1 m is magnitude f flux of minus sign omitted interested in @xmath2 m preserves orientation data in figure 1 converted to data frequency - searched using period04 software package 2005 between 3 c / d and 50 c / d typical for @xmath0 scuti variables power spectra for targets shown in figure 2 in magnitudes targets have several signals in power spectra model only frequencies with signal - to - noise ratio of @xmath3 5 used to save computational time table 1 gives values frequencies used for modeling uncertainties model performed fitting routine in period04 using frequencies table 1 amplitudes and phases determined uncertainties parameters as initial estimates for model inserted into reduced @xmath4 calculation outlined in bevington and robinson ( 2003 )reduced @xmath4 written as @xmath5 n data points y@xmath6 observed magnitudes @xmath7 uncertainty y(t@xmath6 value model each each pdcsap flux measurement uncertainty converted to uncertainties magnitudes convert uncertainties to magnitudes procedure @xmath8 @xmath9 @xmath10 uncertainties on magnitudes fluxes model y(t@xmath6 ) in reduced @xmath4 calculation given by@xmath11 @xmath12 amplitude angular frequency phase of each sine wave @xmath15 total frequencies for each object @xmath15 = 5 both cases case of kic 10684673 positive trend in data data set fit with linear function determine slope intercept trend define reduced @xmath4 function as original @xmath4 function divided by difference between number data points degrees of freedom number data points 40,000 parameters p fit p < 10 difference becomes @xmath17 model refined by nonlinear least - squares fit marquardt - levenburg algorithm routine allowed amplitudes phases vary frequencies table 1 fixed until amplitudes phases reduced @xmath4 foundcase of kic 10684673 linear trend incorporated into routine values parameters uncertainties ) from fitting procedure given in table 2 overall value for reduced @xmath4 calculation results fitting for both targets shown in figure 3 shows representative samples of fit over entire sc data set for quality fit expectation value of reduced @xmath4 for good fit close to one ( bevington and robinson 2003 ) values somewhat high explained by extreme smallness of photometric uncertainties typically about 3 or more orders of magnitude smaller than measurements reduced @xmath4 values high for kepler s precision atypical fit is satisfactory model replicates every major feature data set confident frequencies reported in table 1 are genuine signals not spurious sc light curves of kic 10684673 and kic 12216817 can be modeled as single @xmath0 scuti - type pulsating systems in several modes both included in kebc flagged as uncertain in nature analysis not rule out possibility of systems eclipsing binaries for hypothesis variation can be explained by single pulsating object to determine nature system spectroscopy on both targetsassociation universities astronomy 2015 mikulski archive space telescopes https://archive.stsci.edu bevington robinson 2003 data reduction error analysis physical sciences bevington robinson boston mcgraw - hill binnendijk 1970 va 17 , 217b borucki 2008 exoplanets detection formation dynamics eds sun mello zhou symp. 249 cambridge univ . press cambridge 17 breger 2000 delta scuti stars proceedings 6th vienna workshop astrophysics breger m h montgomery conf . 210 astron . pacific san francisco 3 gilliland r. l. 2010 pasp , 122 , 131 kirk 2016 151 , 68k lenz breger 2005 coast , 146 , 53 zhang han jiang jiang. 2008 mnras , 387 , 97 | 0.359709 | 2,552 | 1,335 | , Saving $0.1 in GPT-4. |
154,832 | we propose a computational framework for identifying linguistic aspects of politeness .
our starting point is a new corpus of requests annotated for politeness , which we use to evaluate aspects of politeness theory and to uncover new interactions between politeness markers and context .
these findings guide our construction of a classifier with domain - independent lexical and syntactic features operationalizing key components of politeness theory , such as _ indirection _ , _ deference _ , _ impersonalization _ and _ modality_. our classifier achieves close to human performance and is effective across domains .
we use our framework to study the relationship between politeness and social power , showing that polite wikipedia editors are more likely to achieve high status through elections , but , once elevated , they become less polite .
we see a similar negative correlation between politeness and power on stack exchange , where users at the top of the reputation scale are less polite than those at the bottom .
finally , we apply our classifier to a preliminary analysis of politeness variation by gender and community . | politeness is a central force in communication , arguably as basic as the pressure to be truthful , informative , relevant , and clear @xcite .
natural languages provide numerous and diverse means for encoding politeness and , in conversation , we constantly make choices about where and how to use these devices .
observes that `` people desire to be _ paid _ respect '' and identifies honorifics and other politeness markers , like _ please _ , as `` the coin of that payment '' . in turn ,
politeness markers are intimately related to the power dynamics of social interactions and are often a decisive factor in whether those interactions go well or poorly @xcite .
the present paper develops a computational framework for identifying and characterizing politeness marking in requests .
we focus on requests because they involve the speaker imposing on the addressee , making them ideal for exploring the social value of politeness strategies @xcite . requests also stimulate extensive use of what call _ negative politeness _ : speaker strategies for minimizing ( or appearing to minimize ) the imposition on the addressee , for example , by being indirect ( _ would you mind _ ) or apologizing for the imposition ( _ i m terribly sorry , but _ ) @xcite .
our investigation is guided by a new corpus of requests annotated for politeness .
the data come from two large online communities in which members frequently make requests of other members : wikipedia , where the requests involve editing and other administrative functions , and stack exchange , where the requests center around a diverse range of topics ( e.g. , programming , gardening , cycling ) .
the corpus confirms the broad outlines of linguistic theories of politeness pioneered by , but it also reveals new interactions between politeness markings and the morphosyntactic context .
for example , the politeness of _ please _ depends on its syntactic position and the politeness markers it co - occurs with . using this corpus ,
we construct a politeness classifier with a wide range of domain - independent lexical , sentiment , and dependency features operationalizing key components of politeness theory , including not only the negative politeness markers mentioned above but also elements of _ positive politeness _
( gratitude , positive and optimistic sentiment , solidarity , and inclusiveness ) .
the classifier achieves near human - level accuracy across domains , which highlights the consistent nature of politeness strategies and paves the way to using the classifier to study new data .
politeness theory predicts a negative correlation between politeness and the power of the requester , where power is broadly construed to include social status , authority , and autonomy @xcite . the greater the speaker s power relative to her addressee , the less polite her requests are expected to be : there is no need for her to incur the expense of paying respect , and failing to make such payments can invoke , and hence reinforce , her power .
we support this prediction by applying our politeness framework to wikipedia and stack exchange , both of which provide independent measures of social status .
we show that polite wikipedia editors are more likely to achieve high status through elections ; however , once elected , they become less polite . similarly , on stack exchange ,
we find that users at the top of the reputation scale are less polite than those at the bottom .
finally , we briefly address the question of how politeness norms vary across communities and social groups .
our findings confirm established results about the relationship between politeness and gender , and they identify substantial variation in politeness across different programming language subcommunities on stack exchange .
requests involve an imposition on the addressee , making them a natural domain for studying the inter - connections between linguistic aspects of politeness and social variables . [
[ sub : request_data ] ] requests in online communities + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + we base our analysis on two online communities where requests have an important role : the wikipedia community of editors and the stack exchange question - answer community . on wikipedia , to coordinate on the creation and maintenance of the collaborative encyclopedia , editors can interact with each other on user talk - pages ; requests posted on a user talk - page , although public , are generally directed to the owner of the talk - page . on stack exchange , users often comment on existing posts requesting further information or proposing edits ; these requests are generally directed to the authors of the original posts .
both communities are not only rich in user - to - user requests , but these requests are also part of consequential conversations , not empty social banter ; they solicit specific information or concrete actions , and they expect a response . [ [ sub : politeness_annotation ] ] politeness annotation + + + + + + + + + + + + + + + + + + + + + computational studies of politeness , or indeed any aspect of linguistic pragmatics , demand richly labeled data .
we therefore label a large portion of our request data ( over 10,000 utterances ) using amazon mechanical turk ( amt ) , creating the largest corpus with politeness annotations ( see table [ tab : data ] for details ) .
we choose to annotate requests containing exactly two sentences , where the second sentence is the actual request ( and ends with a question mark ) .
this provides enough context to the annotators while also controlling for length effects .
each annotator was instructed to read a batch of 13 requests and consider them as originating from a co - worker by email . for each request ,
the annotator had to indicate how polite she perceived the request to be by using a slider with values ranging from `` very impolite '' to `` very polite . ''
each request was labeled by five different annotators .
we vetted annotators by restricting their residence to be in the u.s . and by conducting a linguistic background questionnaire .
we also gave them a paraphrasing task shown to be effective for verifying and eliciting linguistic attentiveness @xcite , and we monitored the annotation job and manually filtered out annotators who submitted uniform or seemingly random annotations . because politeness is highly subjective and annotators may have inconsistent scales , we applied the standard z - score normalization to each worker s scores .
finally , we define the politeness score ( henceforth _ politeness _ ) of a request as the average of the five scores assigned by the annotators .
the distribution of resulting request scores ( shown in figure [ fig : annotation ] ) has an average of 0 and standard deviation of 0.7 for both domains ; positive values correspond to polite requests ( i.e. , requests with normalized annotations towards the `` very polite '' extreme ) and negative values to impolite requests . a summary of all our request data is shown in table [ tab : data ] .
.summary of the request data and its politeness annotations . [ cols="<,^,^,^",options="header " , ]
politeness has been a central concern of modern pragmatic theory since its inception @xcite , because it is a source of pragmatic enrichment , social meaning , and cultural variation @xcite . the starting point for most research
is the theory of .
aspects of this theory have been explored from game - theoretic perspectives @xcite and implemented in language generation systems for interactive narratives @xcite , cooking instructions , @xcite , translation @xcite , spoken dialog @xcite , and subjectivity analysis @xcite , among others . in recent years , politeness has been studied in online settings .
researchers have identified variation in politeness marking across different contexts and media types @xcite and between different social groups @xcite .
the present paper pursues similar goals using orders of magnitude more data , which facilitates a fuller survey of different politeness strategies .
politeness marking is one aspect of the broader issue of how language relates to power and status , which has been studied in the context of workplace discourse @xcite and social networking @xcite .
however , this research focusses on domain - specific textual cues , whereas the present work seeks to leverage domain - independent politeness cues , building on the literature on how politeness affects worksplace social dynamics and power structures @xcite .
study the question of how and why specific individuals rise to administrative positions on wikipedia , and show that power differences on wikipedia are revealed through aspects of linguistic accommodation .
the present paper complements this work by revealing the role of politeness in social outcomes and power relations .
we construct and release a large collection of politeness - annotated requests and use it to evaluate key aspects of politeness theory .
we build a politeness classifier that achieves near - human performance and use it to explore the relation between politeness and social factors such as power , status , gender , and community membership .
we hope the publicly available collection of annotated requests enables further study of politeness and its relation to social factors , as this paper has only begun to explore this area .
we thank jean wu for running the amt annotation task , and all the participating turkers .
we thank diana minculescu and the anonymous reviewers for their helpful comments .
this work was supported in part by nsf iis-1016909 , cns-1010921 , iis-1149837 , iis-1159679 , aro muri , darpa smisc , okawa foundation , docomo , boeing , allyes , volkswagen , intel , alfred p.
sloan fellowship , the microsoft faculty fellowship , the gordon and dailey pattee faculty fellowship , and the center for advanced study in the behavioral sciences at stanford .
penelope brown and stephen c. levinson .
universals in language use : politeness phenomena . in esther
n. goody , editor , _ questions and politeness : strategies in social interaction _ , pages 56311 ,
cambridge university press .
robert munro , steven bethard , victor kuperman , vicky tzuyin lai , robin melnick , christopher potts , tyler schnoebelen , and harry tily .
2010 . . in _ proceedings of the naacl hlt 2010 workshop on creating speech and language data with amazon s mechanical turk
_ , pages 122130 .
william yang wang , samantha finkelstein , amy ogan , alan w. black , and justine cassell . 2012 .
`` love ya , jerkface '' : using sparse log - linear models to build positive and impolite relationships with teens . in _ proceedings of sigdial _ , pages 2029 . | politeness central force in communication as basic as pressure to be truthful informative relevant clear natural languages provide diverse means for encoding politeness in conversation make choices about use devices. people desire _ paid _ respect '' identifies honorifics politeness markers like _ please _ as coin of payment '' politeness markers related to power dynamics social interactions decisive factor in interactions or present paper develops computational framework for identifying characterizing politeness marking in requests focus on requests speaker imposing on addressee ideal for exploring social value of politeness strategies requests stimulate use of negative politeness _ speaker strategies for minimizing imposition indirect apologizing for imposition investigation guided by new corpus of requests annotated for politeness data from two online communities wikipedia stack exchange diverse topics programming gardening corpus confirms linguistic theories of politeness reveals new interactions between politeness markings morphosyntactic context politeness of _ please _ depends on syntactic position politeness markers co - occurs withcorpus construct politeness classifier with domain - independent lexical sentiment dependency features operationalizing components politeness theory including negative politeness markers positive politeness gratitude sentiment solidarity inclusiveness classifier achieves near human - level accuracy across domains highlights consistent politeness strategies paves way classifier new data politeness theory predicts negative correlation between politeness power requester power social status authority autonomy greater speaker s power less polite requests no need respect failing payments power support prediction politeness framework to wikipedia stack exchange measures social status polite wikipedia editors likely achieve high status through elections once elected become less polite stack exchange users top of reputation scale less polite bottom address question politeness norms vary across communities social groups findings confirm results relationship politeness gender identify substantial variation in politeness across programming language subcommunities on stack exchange requests involve imposition on addressee natural domain for studying inter - connections between linguistic politeness social variablessub : request_data ] ] requests in online communities + base analysis on two online communities : wikipedia community and stack exchange question - answer community on wikipedia editors interact on user talk - pages requests directed to owner on stack exchange users comment on posts requesting information or proposing edits requests directed to authors original posts both communities rich in user - to - user requests part of consequential conversations not social banter solicit specific information or concrete actions expect response sub : politeness_annotation ] ] politeness annotation computational studies of politeness demand richly labeled data label large portion of request data ( over 10,000 utterances ) using amazon mechanical turk ( amt ) creating largest corpus with politeness annotations ( see table [ tab : data ] for details choose to annotate requests containing two sentences second sentence is actual request ends with question mark provides context length effects each annotator instructed to read batch of 13 requests from co - worker by email for each request slider with values from very impolite to very polite each request labeled by five annotators vetted annotators residence to in u.conducting linguistic background questionnaire gave paraphrasing task effective for verifying linguistic attentiveness monitored annotation job filtered out annotators uniform or random annotations politeness subjective annotators inconsistent scales applied standard z - score normalization to each worker s scores politeness score of request as average of five scores assigned by annotators distribution of request scores in figure average 0 standard deviation 0.7 for both domains positive values correspond to polite requests negative values to impolite requests summary of request data shown in table [ data data politeness annotations politeness central concern of modern pragmatic theory source of pragmatic enrichment social meaning cultural variation starting point for research theory of aspects theory explored from game - theoretic perspectives implemented in language generation systems for interactive narratives cooking instructions translation spoken dialog subjectivity analysis politeness studied in online settings identified variation in politeness marking across contexts media types social groups present paper pursues similar goals using more data facilitates fuller survey of politeness strategiespoliteness marking language power status studied workplace discourse social networking research focusses on domain - specific textual cues present work domain - independent politeness cues literature politeness social dynamics power structures individuals rise administrative positions wikipedia power differences linguistic accommodation paper role politeness in social outcomes power relations construct release collection politeness - annotated requests evaluate politeness theory build politeness classifier near - human performance explore relation between politeness social factors power status gender community membership hope collection requests enables further study politeness social factors thank jean wu annotation task participating turkers diana minculescu anonymous reviewers comments work supported by nsf iis-1016909 cns-1010921 iis-1149837 iis-1159679 aro muri darpa smisc okawa foundation docomo boeing allyes volkswagen intel alfred p. sloan fellowship microsoft faculty fellowship gordon dailey pattee faculty fellowship center for advanced study behavioral sciences stanford penelope brown stephen c. levinson universals in language use : politeness phenomenaeditor questions politeness strategies social interaction pages 56311 cambridge university press robert munro steven bethard kuperman robin melnick christopher potts tyler schnoebelen harry tily 2010 proceedings naacl workshop speech language data amazon pages 122130 william wang samantha finkelstein amy ogan alan black justine cassell 2012 love jerkface linear models positive relationships teens proceedings sigdial pages 2029 | 0.359046 | 2,202 | 1,058 | , Saving $0.1 in GPT-4. |
154,835 | in the framework of the @xmath0 meson decay model , the strong decays of the @xmath1 and @xmath2 @xmath3 states are investigated .
it is found that in the presence of the initial state mass being @xmath4 gev , the total widths of the @xmath1 and @xmath2 @xmath3 states are about @xmath5 mev and @xmath6 mev , respectively .
also , when the initial state mass varies from 2220 to 2400 mev , the total width of the @xmath2 @xmath3 state varies from about 100 to 132 mev , while the total width of the @xmath1 @xmath3 state varies from about 400 to 594 mev .
a comparison of the predicted widths and the experimental result of @xmath7 gev , the width of the @xmath8 with a mass of @xmath9 gev recently observed by the bes collaboration in the radiative decay @xmath10 , suggests that it would be very difficult to identify the @xmath8 as the @xmath1 @xmath3 state , and the @xmath8 seams a good candidate for the @xmath2 @xmath3 state .
-1 cm mesons , @xmath0 model * pacs numbers:*14.40 .
cs , 12.39.jh | experimentally , a low - mass enhancement in @xmath11 radiative decays @xmath12 at @xmath13 gev with a clear pseudoscalar assignment was first reported by the dm2 collaboration@xcite .
subsequently , the dm2 collaboration@xcite and the mark iii collaboration@xcite gave the evidence of a resonant @xmath14 production around @xmath15 gev , preferably pseudoscalar , also in @xmath16 .
a fit to the @xmath14 invariant - mass spectrum gave a mass of @xmath17 mev and a width of @xmath18 mev@xcite .
an angular analysis of the @xmath14 signal found it to be consistent with a @xmath19 [ @xmath20 $ ] assignment .
the nature of the @xmath8 is unclear .
possibilities of the nature of the @xmath8 include the second and third radial excitations of the pseudoscalar meson @xmath21 , hybrid , glueball or multiquark state .
however , the large uncertainty of the width of the @xmath8 leads to that the theoretical interpretations perhaps remain open .
recently , based on the @xmath22 @xmath11 events collected in the besii detector , the radiative decay @xmath10 was analyzed by the bes collaboration , and a near - threshold enhancement was found in the @xmath14 invariant mass distribution at @xmath4 gev with a statistical significance larger than @xmath23 .
a partial wave analysis shows that this structure is dominated by a @xmath19 [ @xmath8 ] with a mass of @xmath9 gev and a width of @xmath25 gev , and the production branching fraction is @xmath26@xcite .
the improved measurements of the @xmath8 performed by the bes collaboration maybe open a window for revealing the nature of the @xmath8 .
it is very important to exhaust possible conventional @xmath27 description of the @xmath8 before resorting to more exotic interpretations such as hybrid , glueball or multiquark state as mentioned above . in the present work ,
we shall focus on the possibility of the @xmath8 being the ordinary pseudoscalar @xmath27 state . from pdg2006@xcite , the @xmath28 meson nonet ( @xmath29 , @xmath30 , @xmath21 and @xmath31 ) as well as the @xmath32 members [ @xmath33 , @xmath34 and @xmath35 have been well established . in our previous work@xcite , we suggested that the @xmath36 and @xmath37 , together with the @xmath38 and @xmath39 observed by the bes collaboration@xcite , constitute the @xmath1 meson nonet .
theoretically , both the second@xcite and third@xcite radial excitations of the @xmath21 are predicted to lie in the mass range of the @xmath8 .
the main purpose of this work is to evaluate the widths of the @xmath1 and @xmath2 @xmath3 states in the @xmath0 meson decay model , and then check which of these two pictures can reasonably account for the total width of the @xmath8 .
the organization of this paper is as follows . in section 2 ,
the brief review of the @xmath0 decay model is given ( for the detailed review see _
e.g. _ refs.@xcite . ) in section 3 , the decay widths of the @xmath1 and @xmath2 @xmath3 states are presented , and the summary and conclusion are given in section 4 .
the @xmath0 decay model , also known as the quark - pair creation model , was originally introduced by micu@xcite and further developed by le yaouanc et al.@xcite .
the @xmath0 decay model has been widely used to evaluate the strong decays of hadrons@xcite , since it gives a good description of many of the observed decay amplitudes and partial widths of the hadrons .
the main assumption of the @xmath0 decay model is that strong decays take place via the creation of a @xmath0 quark - antiquark pair from the vacuum . the new produced quark - antiquark pair , together with the @xmath27 within the initial meson regroups into two outgoing mesons in all possible quark rearrangement ways , which corresponds to the two decay diagrams as shown in fig.1 for the meson decay process @xmath40 . the transition operator @xmath41 of the decay @xmath42 in the @xmath0 model is given by @xmath43 where @xmath44 is a dimensionless parameter representing the probability of the quark - antiquark pair @xmath45 with @xmath46 creation from the vacuum , @xmath47 and @xmath48 are the momenta of the created quark @xmath49 and antiquark @xmath50 , respectively .
@xmath51 , @xmath52 , and @xmath53 are the flavor , color , and spin wavefunctions of the @xmath45 , respectively .
the solid harmonic polynomial @xmath54 reflects the momentum - space distribution of the @xmath45 . for the meson wavefunction
, we adopt the mock meson @xmath55 defined by@xcite @xmath56 where @xmath57 and @xmath58 are the masses of the quark @xmath59 with a momentum of @xmath60 and the antiquark @xmath61 with a momentum of @xmath62 , respectively .
@xmath63 is the radial quantum number of the meson @xmath64 composed of @xmath65 .
@xmath66 , @xmath67 , @xmath68 ( @xmath69 ) is the spin of @xmath59 ( @xmath70 ) , @xmath71 is the relative orbital angular momentum between @xmath59 and @xmath70 .
@xmath72 , @xmath73 .
@xmath74 is a clebsch - gordan coefficient , and @xmath75 is the total energy of the meson @xmath64 .
@xmath76 , @xmath77 , @xmath78 , and @xmath79 are the spin , flavor , color , and space wavefunctions of the meson @xmath64 , respectively .
the mock meson satisfies the normalization condition @xmath80 the @xmath81-matrix of the process @xmath42 is defined by @xmath82 with @xmath83 where @xmath84 is the helicity amplitude of @xmath42 . in the center of mass frame of meson @xmath64 , @xmath84
can be written as @xmath85{l } \scriptstyle m_{l_a},m_{s_a},\\\scriptstyle m_{l_b},m_{s_b},\\ \scriptstyle m_{l_c},m_{s_c},m \end{array}}\renewcommand{\arraystretch}{1}\!\ !
\langle l_am_{l_a}s_am_{s_a}|j_am_{j_a}\rangle\nonumber\\ & & \times\langle l_bm_{l_b}s_bm_{s_b}|j_bm_{j_b}\rangle\langle l_cm_{l_c}s_cm_{s_c}|j_cm_{j_c}\rangle\nonumber\\ & & \times\langle 1m1-m|00\rangle\langle \chi^{14}_{s_bm_{s_b}}\chi^{32}_{s_cm_{s_c}}|\chi^{12}_{s_am_{s_a}}\chi^{34}_{1-m}\rangle\nonumber\\ & & \times[f_1i(\vec{p},m_1,m_2,m_3)+(-1)^{1+s_a+s_b+s_c}f_2i(-\vec{p},m_2,m_1,m_3)],\end{aligned}\ ] ] with @xmath86 and @xmath87 , corresponding to the contributions from figs . 1 ( a ) and 1 ( b ) , respectively , and @xmath88 where @xmath89 , @xmath90
, @xmath91 is the mass of the created quark @xmath49 .
the spin overlap in terms of winger s @xmath92 symbol can be given by @xmath93 in order to compare with experiment conventionally , @xmath94 can be converted into the partial amplitude by a recoupling calculation@xcite @xmath95{l } \scriptstyle m_{j_b},m_{j_c},\\\scriptstyle m_s , m_l \end{array}}\renewcommand{\arraystretch}{1}\!\ ! \langle lm_lsm_s|j_am_{j_a}\rangle\langle j_bm_{j_b}j_cm_{j_c}|sm_s\rangle\nonumber\\ & & \times\int d\omega\,\mbox{}y^\ast_{lm_l}{\cal{m}}^{m_{j_a}m_{j_b}m_{j_c } } ( \vec{p}).\end{aligned}\ ] ] if we consider the relativistic phase space , the decay width @xmath96 in terms of the partial wave amplitudes is @xmath97 here @xmath98=@xmath99[m^2_a-(m_b - m_c)^2]}}{2m_a}$ ] , @xmath100 , @xmath101 , and @xmath102 are the masses of the meson @xmath64 , @xmath103 , and @xmath104 , respectively . the decay width can be derived analytically if the simple harmonic oscillator ( sho ) approximation for the meson space wave functions is used . in momentum - space ,
the sho wave function is @xmath105 where the radial wave function is given by @xmath106 here @xmath107 is the sho wave function scale parameter , and @xmath108 is an associated laguerre polynomial .
the sho wave functions can not be regarded as realistic , however , they are a _
de facto _ standard for many nonrelativistic quark model calculations .
moreover , the more realistic space wave functions such as those obtained from coulomb , plus the linear potential model do not always result in systematic improvements due to the inherent uncertainties of the @xmath0 decay model itself@xcite .
the sho wave function approximation is commonly employed in the @xmath0 decay model in literature . in the present work , the sho wave function approximation for the meson space wave functions is taken .
under the sho wave function approximation , the parameters used in the @xmath0 decay model involve the @xmath27 pair production strength parameter @xmath44 , the sho wave function scale parameter @xmath107 , and the masses of the constituent quarks . in the present work ,
we take @xmath109 and @xmath110 gev , the typical values used to evaluate the light meson decays@xcite is higher than that used by other groups such as @xcite by a factor of @xmath111 due to different field conventions , constant factor in @xmath41 , etc .
the calculated results of the widths are , of course , unaffected . ] , and @xmath112 gev , @xmath113 gev@xcite . based on the partial wave amplitudes listed in the appendixes a and b , and the flavor and charge multiplicity factors shown in table 2 , from ( [ width1 ] ) ,
the numerical values of the partial decay widths of the @xmath2 and @xmath1 @xmath3 states are listed in table 1 .
the initial state mass is set to 2.24 gev and masses of the final mesons are taken from pdg2006@xcite except for the @xmath114 mass as the @xmath115 kaon is problematic@xcite .
quark model@xcite and other phenomenological approaches@xcite consistently suggest the @xmath115 kaon has a mass about 1580 mev , here we take 1580 mev as the mass of the @xmath115 kaon . ] . | low - mass enhancement in @xmath11 radiative decays gev pseudoscalar assignment reported by dm2 collaboration@xcite mark iii resonant @xmath14 production around @xmath15 gev pseudoscalar @xmath16 fit @xmath14 invariant - mass spectrum mass @xmath17 mev width @xmath18 mev@xcite angular analysis signal consistent with @xmath19 $ assignment nature @xmath8 unclear possibilities include second third radial excitations pseudoscalar hybrid glueball multiquark state uncertainty of width theoretical interpretations open @xmath22 @xmath11 events detector radiative decay @xmath10 analyzed by bes collaboration near - threshold enhancement in @xmath14 invariant mass distribution at @xmath4 gev statistical significance larger than @xmath23 partial wave analysis structure dominated by @xmath19 [ ] mass @xmath9 gev width @xmath25 gev production branching fraction @xmath26@xcite improved measurements @xmath8 bes collaboration open nature important to exhaust conventional @xmath27 description before exotic interpretations hybrid glueball multiquarkpresent work focus on possibility @xmath8 being ordinary pseudoscalar @xmath27 state pdg2006@xcite @xmath28 meson nonet ( ) @xmath32 members @xmath33 established previous work suggested @xmath36 @xmath37 @xmath38 @xmath39 observed bes collaboration@xcite constitute @xmath1 meson nonet second third@xcite radial excitations of @xmath21 predicted in mass range @xmath8 main purpose work evaluate widths @xmath1 @xmath2 @xmath3 states in @xmath0 meson decay model check which two pictures account for total width @xmath8 organization paper section 2 brief review @xmath0 decay model section 3 decay widths of @xmath1 @xmath2 @xmath3 states presented summary conclusion section 4 @xmath0 decay model quark - pair creation model introduced by micu@xcite developed by le yaouanc et al.@xcite used evaluate strong decays of hadrons@xcite good description observed decay amplitudes partial widths hadronsassumption @xmath0 decay model decays creation quark - antiquark pair from vacuum new pair @xmath27 regroups into two mesons corresponds decay diagrams fig.1 meson decay process @xmath40 transition operator @xmath41 decay @xmath42 by @xmath43 @xmath44 parameter probability quark - antiquark pair @xmath45 @xmath46 creation @xmath47 @xmath48 momenta of created quark @xmath49 antiquark @xmath50 @xmath51 flavor color spin wavefunctions of @xmath45 solid harmonic polynomial @xmath54 reflects momentum - space distribution @xmath45 meson wavefunction mock meson @xmath55 defined @xmath56 @xmath57 masses of quark @xmath59 momentum @xmath60 antiquark @xmath61 @xmath62 @xmath63 radial quantum number meson @xmath64 @xmath65 @xmath66 spin of @xmath59 @xmath71 relative orbital angular momentum between @xmath59 @xmath70 @xmath72 @xmath73@xmath74 clebsch - gordan coefficient total energy meson @xmath64 @xmath76 spin flavor color space wavefunctions mock meson satisfies normalization condition @xmath80 @xmath81-matrix process @xmath42 defined @xmath82 @xmath83 @xmath84 helicity amplitude @xmath42center mass frame meson @xmath64 @xmath84 written @xmath85 m_{l_a}_{s_a} m_{l_b}{s_b} m_{l_c}_{s_c}\renewcommand l_am_{l_a{s_a}_am_{j_a} l_bm_{l_b{s_b}_bm{j_b} l_cm_{l_c}s_{s_c}_cm{j_c} 1m1-m{14}_{s_bm_{s_b}}{32}{s_cm_{s_c}}{12}_{s_am_{s_a}}{34}{1-m}[f_1i({p},m_1,m_2,m_3){1+s_a+s_b+s_c}f_2i_2_3){aligned @xmath86 @xmath87 contributions figs 1 a 1 b ) @xmath88 @xmath89 @xmath90 @xmath91 mass quark @xmath49spin overlap winger s @xmath92 symbol given by @xmath93 @xmath94 converted into partial amplitude recoupling calculation @xmath95 m_{j_b},m_{j_c}_s_l{array}}\renewcommand\arraystretch}{1}_lsm_{j_a}{j_b_{j_c}{lm_l}_{j_a{j_b{j_c relativistic phase space decay width @xmath96 partial wave amplitudes is @xmath97 @xmath98=@xmath99[m^2_a-(m_b - m_c)^2]}}_a @xmath100 @xmath101 @xmath102 masses of meson @xmath64 @xmath103 @xmath104 decay width derived harmonic oscillator sho approximation meson space wave functions momentum - space sho wave function is @xmath105 radial wave function @xmath106 @xmath107 sho wave function scale parameter @xmath108 laguerre polynomialsho wave functions realistic standard for nonrelativistic quark model calculations realistic space wave functions from coulomb linear potential model improvements uncertainties @xmath0 decay model sho wave function approximation employed in @xmath0 decay model present work for meson space wave functions parameters involve @xmath27 pair production strength parameter @xmath44 sho wave function scale parameter @xmath107 masses of constituent quarks @xmath109 and @xmath110 gev typical values light meson decays higher than other groups by factor @xmath111 due to different field conventions constant factor @xmath41 calculated results of widths unaffected @xmath112 gev partial wave amplitudes in appendixes a b flavor charge multiplicity factors in table 2 numerical values of partial decay widths of @xmath2 @xmath1 @xmath3 states listed in table 1 initial state mass to 2.24 gev masses final mesons taken from pdg2006@xcite except @xmath114 mass @xmath115 problematicquark model@xcite phenomenological approaches suggest @xmath115 kaon 1580 mev take 1580 mass @xmath115 kaon | 0.483799 | 2,602 | 1,389 | , Saving $0.1 in GPT-4. |
154,859 | influence of the damping on collective oscillations of a one - dimensional trapped bose gas in the mean field regime has been studied . using the phenomenological damping approach developed by l.p .
pitaevskii , modified variational equations for the parameters of the condensate wave function is derived .
analytical expressions for the condensate parameters in equilibrium state have been obtained .
bistability in nonlinear oscillations of the condensate under periodic variations of the trap potential is predicted .
the predictions of the modified variational approach are confirmed by full numerical simulations of the 1d gp equation with the damping . | the dynamics of a one - dimensional trapped ultra - cold bose gas has attracted considerable attention for last years @xcite .
recently has been achieved regimes where 1d bose gas with a zero - range two - body interaction in the tonks -girardeau ( tg)regime @xcite with fermionic behavior becomes visible .
an investigation of the transition from the mean field regime to the tg regime was performed in @xcite .
experimentally 1d regime has been realized in @xcite .
measurements of the collective oscillations of such a system should give a lot of information about the bec dynamics . in particular
this is important in the analysis of the condensate dynamics in a magnetic waveguide , being a fundamental atom optical element @xcite
. performed by this time theoretical descriptions have mainly dealt with conservative systems , e.g. see @xcite , where collective oscillations of a quasi 1d bose - einstein condensate(bec ) in the low and high density regimes were investigated .
the damping of the radial bec oscillations in a cylindric trap , connected with the parametric resonance and leading to the energy transfer from collective oscillations to longitudinal sound waves has been studied in @xcite .
the dissipation inheres in real systems .
so it is of interest to investigate theoretically effect of damping on collective oscillations of a one - dimensional trapped repulsive bose gas .
we consider here the problem using the phenomenological approach developed by l.p .
pitaevskii @xcite and employed later in @xcite
the dynamics of a trapped one dimensional repulsive bose gas with the damping is described in the framework of the 1d gross - pitaevskii equation @xcite @xmath0 with the total number of atoms @xmath1 .
this equation is obtained in the case of a highly anisotropic external potential under the assumption that the transversal trapping potential is harmonic : @xmath2 and @xmath3 under such conditions we can consider the solution of 3d equation to have the form @xmath4 where @xmath5 . averaging in the radial direction ( i.e. integrating over the transversal variables ) we come to equation ( [ gpe ] ) describing the dynamics of the gas in longitudinal direction .
the potential @xmath6 is assumed to be @xmath7 , where @xmath8 describes the time dependence of the potential , which we consider here as @xmath9 .
the effective one dimensional mean field nonlinearity coefficient @xmath10 , where @xmath11 is the atomic scattering length .
@xmath12 corresponds to the bose gas with a repulsive interaction between atoms and @xmath13 to an attractive interaction .
we have the following dimensionless form of the equation ( [ gpe ] ) @xmath14 by setting : @xmath15 @xmath16 @xmath17 @xmath18 with @xmath19 for the repulsive and attractive two - body interactions respectively .
to describe collective oscillations of a bose gas under damping we employ the variational approach . for the wavefunction @xmath20 we use the gaussian trial function @xmath21 where a , a , b and @xmath22 are the amplitude , width , chirp and linear phase , respectively .
equation ( [ gpe1 ] ) can be obtained from the variational equations @xmath23 where l is the lagrangian density , @xmath24 , of a conservative system , given by @xmath25 and @xmath26 is defined as @xmath27 , where @xmath28 is the right side of equation ( [ gpe1 ] ) . inserting trial function([gaussian ] ) into equation ( [ lagden ] ) and averaging it as @xmath29 we obtain the averaged lagrangian of the conservative system in terms of the trial function parameters : @xmath30 using equation ( [ vareq ] ) and its conjugate , we obtain a system of equations for the variational parameters @xmath31 @xcite : @xmath32 inserting ( [ gaussian ] ) and ( [ avlag ] ) into equation ( [ varpareq ] ) we derive the following system of ordinary differential equations ( ode ) : @xmath33 this system of equations can also be obtained from the modified form of the conservation law for the number of atoms and by using the moments method , as shown in appendix .
let us rewrite the system using the notations @xmath34 : @xmath35 this ode system is the main result of this section .
we have carried out a series of time dependent simulations of the system within the variational approach using equation ( [ rewode ] ) and also by performing exact numerical calculations using equation ( [ gpe1 ] ) . in our numerical calculations
we discretize the problem in a standard way , with the time step @xmath36 , and spatial step @xmath37 , so @xmath38 approximates @xmath39 .
more specifically we approximate the governing equation ( [ gpe1 ] ) with the following semi - implicit crank - nickolson scheme using split - step method @xcite @xmath40 where @xmath41 @xmath42 @xmath43 and @xmath44 @xmath45 are defined so that @xmath46 @xmath47 @xmath48 solving ( [ splstep1 ] ) and ( [ splstep2 ] ) first at time @xmath49 we produce intermediate solutions @xmath45 and @xmath50 .
the final solution for one time step dt is obtained from ( [ splstep3 ] ) .
the results of numerical simulations of both pde and ode models are presented below .
figure [ equilstate ] shows time evolution of the width and the norm of the condensate for the case of @xmath51 , i.e. when the external potential has no time perturbations .
the figure is presented to compare the character of the width oscillations and the behavior of the norm depending on the values of the chemical potential @xmath52 and the dissipative constant @xmath53 .
as seen , the frequency of the oscillations does not depend on the dissipative constant at all and depends weakly on the chemical potential , e.g. when @xmath54 , @xmath55 while when @xmath56 , @xmath57 . for greater values of @xmath53 ,
the oscillations damp faster .
we see that the ode leads to an equilibrium state the norm of which is 3 - 4 percent less than that of the pde results . from figure [ equilstate ] it can be seen that damping process of the condensate eventually leads to the equilibrium state .
this equilibrium state can also be obtained analytically by solving equation ( [ rewode ] ) .
taking into consideration that in an equilibrium state @xmath58 , @xmath59 and @xmath60 the following expressions can be obtained : @xmath61 putting the value of chemical potential e.g. for @xmath54 we find @xmath62 , @xmath63 and @xmath64 which are confirmed by the pde results . in figure
[ norm ] we can observe an interesting behavior of the norm . if an external perturbation is applied to a trapped bec which is already in the equilibrium state then the norm of the condensate starts decreasing . in the figure
the frequency of the periodical trap perturbation is taken to be equal to the eigenfrequency of the system determined from the figure [ equilstate ] and the amplitude of time perturbations is taken as @xmath65 .
we see that for smaller values of the dissipative constant @xmath53 the norm goes farther from the equilibrium state . that is , the dissipative constant is not simply the quantity which is responsible for diminishing the norm , but it is the constant keeping a condensate in the equilibrium state .
the width dynamics under main resonance with the initial wave packet taken in the equilibrium state is depicted in figure [ widthres ] . as shown , in contrast to the norm the width oscillates near the previous point .
as seen from the figure the amplitude of the width oscillations becomes stable by the time @xmath66 .
performing ode and pde calculations for the frequencies which lie around the eigenfrequency of the bec and measuring the amplitudes of the above - said stable oscillations we have plotted the values of the oscillation amplitude as a function of the frequency of the periodical trap perturbations with @xmath67 and @xmath65 in the cases @xmath68 and @xmath69 in figure [ ampfrchar ] .
we see that e.g. when @xmath65 and @xmath69 the highest amplitude of oscillations is driven in the trap perturbation with the frequency @xmath70 which is more than the eigenfrequency @xmath71 .
bistability appears with the smaller values of @xmath53 in the vicinity of this critical frequency .
for @xmath72 we observe large oscillations with @xmath73 , while at @xmath74 we observe much smaller width oscillations with @xmath75 .
it can be seen that with the growth of trap perturbations the value of the critical frequency becomes greater .
let us estimate parameters for the experiment .
the magnetic trap can be taken with parameters @xmath76 hz , @xmath77 hz , and the number of atoms of @xmath78rb @xmath79 .
for the external field @xmath80 , @xmath81 ( repulsive gas ) .
then , the eigenfrequency of the harmonically trapped bec @xmath82 hz . by applying the external perturbation @xmath83 with @xmath84 hz to the trapped bec
one will observe decreasing of the number of atoms by 20 , 13 and 7 percent for @xmath53 = 0.1 , 0.2 and 0.3 respectively .
for the external perturbation with @xmath85 hz , large oscillations will be observed with @xmath86 , while at @xmath87 hz one will observe much smaller oscillations with @xmath88 .
in this paper we study the collective oscillations of a quasi one dimensional bose gas in the presence of dissipative effects .
the modified gross - pitaevskii equation in the framework of the phenomenological approach @xcite has been used . to describe evolution of oscillations
we employ the modified variational approach taking into account the dissipation .
we confirm results for the oscillations damping obtained from the system of equations for the wave function parameters , by the moments method and direct numerical simulations of the full gp equation .
the calculations show that damping oscillations of a bec which eventually comes to an equilibrium state occur with the eigenfrequency which does not depend on the value of the dissipative constant @xmath53 and only depends on the chemical potential @xmath52 .
the expressions for the width and the norm of a condensate in an equilibrium state have been derived analytically .
we study the main resonance in condensate oscillations . for a bec in an equilibrium state it is shown that making the external potential oscillate leads to decreasing of the condensate norm and the norm begins oscillating around the point which is less than the previous stationary one in the equilibrium state , whereas the condensate width oscillates near the previous point .
we have also shown that in resonances the bistability appears with smaller values of the dissipative constant in the vicinity of the critical frequency which is above the eigenfrequency of the bec .
the work was partially supported by a fund for fundamental research support from the uzbek academy of sciences ( award n 17 - 04 ) .
fkha is grateful to the physics department of the university of salerno for the research grant .
differentiating the norm of the condensate by time we can obtain the first equation of the system ( [ rewode ] ) as @xmath89 inserting @xmath90 from equation ( [ gpe1 ] ) into ( [ app1 ] ) we derive the modified form of the conservation law for the norm of the condensate [ 29 ] .
@xmath91 substituting the gaussian trial function ( [ gaussian ] ) into this equation we obtain equation ( [ rewode]a ) . equations ( [ rewode]b ) and ( [ rewode]c ) can be obtained by the moments method .
we obtain equation ( [ rewode]b ) by calculating @xmath92 the substitution of the @xmath90 from equation ( [ gpe1 ] ) leads to the following equation : @xmath93 substituting trial function ( [ gaussian ] ) we get equation ( [ rewode]b ) . equation ( [ rewode]c ) is derived by calculating @xmath94 integrating by parts and substituting @xmath90 from equation ( [ gpe1 ] ) we obtain @xmath95 inserting the gaussian ansatz into this equation and using results ( [ rewode]a ) and ( [ rewode]b ) we obtain equation ( [ rewode]c ) . | dynamics one - dimensional trapped ultra - cold bose gas attracted attention achieved regimes 1d bose gas zero - range two - body interaction tonks -girardeau tg)regime fermionic behavior visible transition from mean field to tg regime in @xcite experimentally 1d regime realized in @xcite measurements collective oscillations information bec dynamics important analysis condensate dynamics magnetic waveguide fundamental atom optical element theoretical descriptions dealt conservative systems collective oscillations quasi 1d bose - einstein condensate low high density regimes investigated damping radial bec oscillations in cylindric trap energy transfer sound waves studied in @xcite dissipation real systems investigate effect damping on collective oscillations one - dimensional trapped repulsive bose gas phenomenological approach by l.p . pitaevskii @xcite dynamics trapped one dimensional repulsive bose gas damping described 1d gross - pitaevskii equation @xcite total number atoms equation obtained anisotropic external potential transversal trapping potential harmonic solution 3d equation form @xmath4 where @xmath5 .averaging radial direction integrating transversal variables equation gpe ] ) dynamics gas longitudinal direction potential @xmath6 @xmath7 @xmath8 time dependence @xmath9 effective one dimensional mean field nonlinearity coefficient @xmath10 @xmath11 atomic scattering length @xmath12 bose gas repulsive interaction @xmath13 attractive interaction form equation @xmath14 @xmath15 @xmath16 @xmath17 @xmath19 repulsive attractive two - body interactions collective oscillations bose gas damping variational approach wavefunction @xmath20 gaussian trial function @xmath21 a b @xmath22 amplitude width chirp linear phase equation gpe1 ] ) variational equations @xmath23 l lagrangian density @xmath24 conservative system @xmath25 @xmath26 @xmath27 @xmath28 right side equation gpe1inserting trial function[gaussian into equation lagden averaging @xmath29 obtain averaged lagrangian conservative system trial function parameters @xmath30 using equation vareq conjugate obtain system equations variational parameters @xmath31 @xmath32 inserting gaussian avlag into equation varpareq derive system ordinary differential equations ode ) @xmath33 system equations obtained from modified conservation law number atoms moments method appendix rewrite system using notations @xmath34 @xmath35 ode system main result section carried time dependent simulations system variational approach using equation rewode ] numerical calculations using equation gpe1 ] ) numerical calculations discretize problem time step @xmath36 spatial step @xmath37 @xmath38 approximates @xmath39 approximate governing equation [ gpe1 ] ) semi - implicit crank - nickolson scheme split - step method @xmath40 @xmath41 defined solving splstep1 splstep2 ] ) intermediate solutions final solution time step from splstep3 ] ) results of numerical simulations pde ode models presented belowfigure [ equilstate ] shows evolution width norm condensate @xmath51 external potential no perturbations figure width oscillations behavior norm depending on chemical potential @xmath52 dissipative constant @xmath53 frequency oscillations on dissipative constant depends weakly on chemical potential @xmath54 greater values @xmath53 oscillations damp faster ode leads to equilibrium state norm 3 - 4 percent less than pde results damping condensate leads to equilibrium state equilibrium state obtained analytically by solving equation ( [ rewode ] ) in equilibrium state @xmath58 @xmath59 @xmath60 expressions obtained : @xmath61 value chemical potential @xmath54 @xmath62 @xmath63 @xmath64 confirmed by pde results figure [ norm ] behavior norm if external perturbation applied to trapped bec in equilibrium state norm condensate starts frequency periodical trap perturbation equal to eigenfrequency system amplitude of time perturbations as @xmath65 smaller values dissipative constant @xmath53 norm goes farther from equilibrium statedissipative constant not diminishing norm keeping condensate in equilibrium state width dynamics with initial wave packet equilibrium state depicted in figure [ widthres width oscillates near previous point amplitude stable by @xmath66 calculations eigenfrequency bec oscillations plotted oscillation amplitude function of frequency periodical trap perturbations with @xmath67 @xmath65 @xmath68 @xmath69 in figure @xmath65 highest amplitude oscillations driven in trap perturbation with frequency @xmath70 more than eigenfrequency @xmath71 bistability appears with smaller values of @xmath53 critical frequency @xmath72 large oscillations with @xmath73 @xmath74 smaller width oscillations with @xmath75 growth of trap perturbations value critical frequency becomes greater estimate parameters for experiment magnetic trap with parameters @xmath76 hz @xmath77 hz number of atoms of @xmath78rb @xmath79 external field @xmath80 , @xmath81 repulsive gas eigenfrequency of harmonically trapped bec @xmath82 hz applying external perturbation @xmath83 @xmath84 hz to trapped bec atoms 20 13 7 percent for @xmath53 = 0.1 0.2 0.3 external perturbation @xmath85 hz large oscillations with @xmath86 @xmath87 hz smaller oscillations @xmath88 study collective oscillations quasi one dimensional bose gas dissipative effects modified gross - pitaevskii equation phenomenological approach @xcite used evolution oscillations modified variational approach dissipation confirm results oscillations damping equations wave function parameters moments method direct numerical simulations full gp equation calculations show damping oscillations bec equilibrium state occur eigenfrequency depend on dissipative constant @xmath53 depends chemical potential @xmath52 expressions width norm condensate in equilibrium state derived analytically study main resonance condensate oscillations equilibrium external potential leads condensate norm point less condensate width oscillates near previous point resonances bistability appears with smaller values dissipative constant critical frequency above eigenfrequency bec work supported by fund fundamental research uzbek academy of sciences award n 17 - 04 )fkha grateful physics department university salerno research grant differentiating norm condensate by time first equation [ rewode ] inserting @xmath90 app1 modified conservation law norm condensate 29 @xmath91 substituting gaussian trial function equation [ rewode]a ) equations rewode]b]c obtained moments method obtain equation [ rewode]b calculating @xmath92 substitution @xmath90 equation gpe1 leads equation @xmath93 substituting trial function gaussian equation [ rewode]b ) equation [ rewode]c derived calculating @xmath94 integrating parts substituting @xmath90 equation gpe1 obtain @xmath95 inserting gaussian ansatz using results rewode]a rewode]b obtain equation ( [ rewode]c ) | 0.434906 | 2,779 | 1,451 | , Saving $0.1 in GPT-4. |
154,867 | in multiquark correlator analyses with @xmath0 classifications , it is possible to separate the scattering background and to justify the factorization of condensates , which allows us to achieve an _ isolated peak saturation _ in the qcd sum rules for multiquark currents
. then we can extract leading @xmath1 properties of the ground state .
an application to the @xmath2 meson is demonstrated . | the multiquark states have been intriguing subjects of researches to study the infrared aspects of qcd beyond those manifested in the usual hadron spectra .
one of the conventional approaches from qcd is to analyze correlators of multiquark currents , which is related to the hadronic spectral functions through the dispersion relation . compared to analyses for usual meson and baryon correlators ,
the multiquark cases are expected to have the following difficulties : ( i ) the complications associated with the many - body system , which make the calculation of correlators much involved .
( ii ) the decays into usual hadrons , whose backgrounds contaminate the signal from multiquark states
. we will show that the use of @xmath0 expansion @xcite greatly reduces both difficulties , and , more importantly , enables to make logically clear statements @xcite .
we perform the correlator analyses using the borel transformed qcd sum rules ( qsr ) for the hadronic correlators @xmath3 @xcite : @xmath4 where lhs is calculated by the operator product expansion ( ope ) and rhs is the integral of the hadronic spectrum . here
@xmath5 characterizes the scale of momentum flow between interpolating fields , and @xmath6 is factorization scale separating high energy calculable part and low energy condenstate part .
usually we are interested in the ground state properties of rhs in eq.([qsr ] ) , and they can be investigated by taking small @xmath7 due to the exponential factor @xmath8 .
but we can not take arbitrarily small @xmath7 value since the ope is basically expansion of @xmath9 and is violated in small @xmath7 region .
thus we must take intermediate region where the integral is well - saturated by ground state contribution and the ope is well - converged .
these are quantified as : pole contribution @xmath10 50% of the total @xcite ( for upperbound @xmath11 ) and highest dimension terms @xmath12 10% of whole ope ( for lowerbound @xmath13 ) .
practically , without these @xmath7 constraints , we will be stuck with the _ pseudo - peak _ artifacts leading to wrong conclusions , as discussed in @xcite . keeping these constraints in mind ,
let us consider how the first difficutly ( i ) is manifested in the qsr analyses .
the multiquark correlators involve many quark lines , sharing only small fraction of the momentum with a scale relevant to the multiquark states . as a result
, we have to incorpolate more non - perturbative diagrams in the ope calculations than usual hadron analyses , to include relevant low energy correlations . otherwise , we can not find the @xmath7 window due to the lack of sufficient low energy contributions .
here we shall add one remark about the low energy dominance . in the case of multiquark correlator analyses
, it is frequently said that the pole - dominace is spoiled by large continuum ( or high energy part of spectrum ) contributions peculiar to multiquark correlators @xcite .
this is a somewhat misleading expression .
a real problem in the early studies is the lack of low energy correlation .
in fact , in certain tetraquark correlators , inclusion of higher dimension terms enables to take sufficiently small @xmath7 value , achieving the low energy dominance not much worse than usual meson sum rules , and even better than baryon sum rules @xcite .
thus we conclude that a relevant problem is numerical ambiguities in values of higher dimension condensates . the second difficulty ( ii ) peculiar to multiquark cases arises as follows : even if we take into account the sufficient low energy correlations , they might be contributions from scattering backgrounds which are ground states in most of multiquark correlators .
this prevents us from investigating the multiquark states of our interests .
both difficulties seem profound . however , as we will show
, leading @xmath1 properties of multiquark states are tractable and , in principle , the corrections to them can be treated step by step with identifying origins of contaminations .
in this section , we discuss how useful @xmath0 expansion is in the application of the qsr .
the first merit is presented in the evaluation of the higher dimension condensates .
it is well - known that in large @xmath1 they can be factorized into the products of known condensates , @xmath14 , @xmath15 , and @xmath16 .
thus we can extend the ope calculations to sufficiently higher dimension terms needed to include relevant low energy contributions . here
we must make one assumption on the @xmath1 scaling of these low dimension condensates . since @xmath17(@xmath18 , @xmath19 , @xmath16 ) are known to be @xmath20 , we simply assume @xmath21 . next , we discuss how to separate the scattering background from the multiquark correlators .
the point is that @xmath1 counting of the quark - gluon graphs is directly related to the qualitative picture of hadronic states .
we will show that with taking appropriate interpolating fields , the multiquark state and background can be assigned in the different order of @xmath1 , and their mixing occurs only in the higher order of @xmath0 than those studied in this work .
then we can concentrate on the quark - gluon graphs which are saturated by the _ isolated _ poles which reflect leading @xmath1 properties of multiquark states . to make discussions concrete
, we consider the @xmath2 meson with @xmath22 as a candidate of the tetraquark ( 4q ) state .
the 4q operators with the @xmath2 quantum number are given ( assuming the ideal mixing for the @xmath2 meson ) by @xmath23 as products of meson operators @xmath24 , where dirac matrix @xmath25 and @xmath26 are the pauli matrices acting on @xmath27 .
let us start the @xmath0 classifications of quark - gluon graphs in the 2-point correlators , graphs ( a)-(c ) in fig.[fig:2pointgraph ] , where planar - gluon lines are not explicitly drawn .
the leading @xmath1 diagrams start from fig.[fig:2pointgraph ] a ) , which include only 2 planar loops and thus reflect irrelevant free 2-meson scattering states .
this indicates that the studies of 4q components require systematic steps beyond the leading @xmath28 .
thus we must proceed to the next leading order of @xmath29 , @xmath20 diagrams which could include the multiquark states . and
@xmath20 quark - gluon diagrams for 2 and 3 point correlators.,width=491,height=90 ] [ fig:2pointgraph ] for systematic classifications , it is useful to first investigate the overlap strength of the operator @xmath30 with hadronic states by @xmath0 , then to use them in the classification of subleading diagrams of 2-point correlators . for this purpose , we consider the 3-point correlator among @xmath30 and two separated @xmath31 ( fig.[fig:2pointgraph ] , d - f ) .
simple calculations indicate that the overlap strength of the 4q field with 2 meson states is @xmath32 since the leading order diagrams are @xmath33 ( @xmath20 ) for @xmath34 ( @xmath35 ) and the overlap strength of @xmath36 with 2q meson state @xmath37 is @xmath38 ) @xcite .
now we can classify the hadronic states in 2-point correlators @xmath39 based on @xmath29 ( see , fig.[fig:2pointgraph ] , a - c ) : ( a ) if @xmath34 , @xmath33 quark - gluon graphs include only the free 2 m scattering states in the region @xmath40 . if @xmath35 , the contributions from these quark - gluon diagrams vanish , indicating the absence of free 2 meson scattering states .
( b ) @xmath20 graphs include the 2 m or 2m scattering and could include 4q poles .
note that if @xmath41 are different from pseudoscalar or axial vector , 2@xmath42 scattering intermediate states are not included up to @xmath20 diagrams .
interactions are needed to transfer initial states into the @xmath43 intermediate states , but such interactions are suppressed by @xmath0 .
then the resonance peaks ( if exist ) _ below _
@xmath44 are _ isolated and have no width _ since the decay channels are absent at this order .
therefore , now we can reduce the @xmath2 spectrum in the 4q correlator into peak(s ) plus continuum _ if we retain only diagrams up to @xmath20 _ for the appropriate currents .
we will study these cases .
this separate investigation of the @xmath33 and @xmath20 parts enables to perform the step by step analyses for each hadronic state .
we relate the ope , _ term by term of @xmath0 _ , to the _ integral _ of the hadronic spectral function through the dispersion relation : @xmath45 the ground state in the reduced @xmath20 spectra @xmath46 can be described as the sharp peak due to the absence of the decay channels . applying the usual quark - hadron duality to the higher excited states , @xmath47 , and taking the moment of borel transform of eq.([disp ] )
, we can express the effective mass as @xmath48 @xmath49 can be uniquely fixed to satisfy the least sensitivity @xcite of the expression ( [ eq : peakmass ] ) against the variation of @xmath5 , since the physical peak should not depend on the artificial expansion parameter @xmath5 .
this criterion is justified only when the peak is very narrow , and our @xmath0 reduction of spectra is essential for its application to allow the qsr framework to determine all physical parameters ( @xmath50 ) in a self - contained way .
[ cols="^,^,^ " , ] in this section , we show the results of borel analyses .
we concentrate on the @xmath20 results of the off diagonal correlator @xmath51 , whose leading order is @xmath20 thus without the factorization violations at the @xmath20 ope . the ope is carried out up to dimension 12 .
the numerical analyses are performed with the values for @xmath52 case , @xmath53 , @xmath54 , @xmath55 , and @xmath56 , respectively .
most results shown below will use the central value .
first we show in the left panel of fig.[fig : borelfig ] the results of the large @xmath1 2q correlators ( expanded up to dimension 6 ) for the vector meson as a reference and the scalar meson as the 2q state in the @xmath2 meson .
the downarrow ( upperarrow ) indicates the values of @xmath13 ( @xmath57 ) . following the @xmath58 ( @xmath59 ) fixing criterion
, we fix @xmath58 to 1.0 ( 1.4 ) gev for vector ( scalar ) mesons , and determine the mass as 0.65 ( 1.10 ) gev .
now we turn to the @xmath20 part of the 4q correlator , @xmath51 , to investigate the possiblity of 4q states . shown in the middle panel of fig.[fig : borelfig ] are the effective masses for @xmath58=1.0 , 1.2 , and 1.4 gev .
we select the @xmath58=1.2 gev case and evaluate the mass as @xmath600.90 gev , which is obviously lower than that of the 2q scalar meson case , @xmath601.10 gev in large @xmath1 limit , and thus is considered as the mass of 4q state . although these absolute values depend on the details of condensate values , the inequality @xmath61 is insensitive to them . in the right panel of fig.[fig : borelfig ] , we show the @xmath62 and @xmath63 dependence of 2q vector , scalar meson masses , and of the 4q mass deduced from @xmath51 . we have also investigated both @xmath33 and @xmath20 parts of @xmath64 ( @xmath65 ) . as for @xmath33 part
, low energy contribution is almost absent below two meson thresholds .
more precisely speaking , acceptable effective mass plots can be obtained only if we take much larger @xmath58 value than the two meson thresholds , giving large effective masses .
this indicates that @xmath33 part includes only free scattering states consistently with @xmath0 arguments . as for @xmath20 part ,
@xmath64 ( @xmath65 ) give similar values @xmath66 gev , as @xmath51 case , despite of the possible factorization violation in the formers .
we have also investigated 4q correlators with @xmath67 .
they do not allow us to find any stable effective mass plots or any reasonably wide @xmath7 window .
this would indicate the absence of 4q pole in @xmath67 channel , consistenly with experimental observations . in conclusion
, it is found that the @xmath0 expansion is very useful to analyze the leading @xmath1 properties of multiquark states with separating scattering backgrounds .
the qsr results show consistencies with expectations from @xmath0 arguments , and further provide the prediction on 4q components in the @xmath2 meson without @xmath68 correlations .
the leading @xmath1 analyses for multiquark correlators may be also useful to distinguish two meson molecules and tetraquarks , which is discussed for the recently found charmed mesons ( x , y , z ) .
since initial states are controled by interpolating fields , and @xmath0 countings give observations on intermediate states , the combination of them enables to separate the molecular type contributions in the spectral functions .
further discussions will be made elsewhere .
thanks the organizers of _ hsqcd _ for their hospitality during the workshop .
this work is supported by riken ,
brookhaven national laboratory and the u. s. department of energy [ contract no . de - ac02 - 98ch10886 ] , and by the grant for scientific research ( no .
20028004 ) in japan .
99 g. t hooft , nucl . phys .
* b724 * , 61 ( 1974 ) .
e. witten , nucl .
b160 * , 57 ( 1979 ) .
t. kojo and d. jido , arxiv:0807.2364 [ hep - ph ] .
shifman , a.i .
vainshtein , and v.i .
zakharov , nucl .
b147 * , 385 ( 1979 ) .
t. kojo , a. hayashigaki , and d. jido , phys .
c * 74 * , 045206 ( 2006 ) .
t. kojo and d. jido , arxiv:0802.2372 [ hep - ph ] . | multiquark states subjects infrared aspects qcd beyond hadron spectra conventional analyze correlators multiquark currents related to hadronic spectral functions through dispersion relation meson baryon multiquark cases expected difficulties complications many - body system calculation decays into hadrons signal @xmath0 expansion @xcite reduces difficulties enables clear statements perform correlator analyses using borel transformed qcd sum rules for hadronic correlators @xmath3 @xcite @xmath4 lhs calculated by operator product expansion ope rhs integral of hadronic spectrum @xmath5 momentum flow between interpolating fields @xmath6 factorization scale high energy calculable low energy condenstate part interested in ground state properties of rhs in eq. investigated small @xmath7 to exponential factor @xmath8 take arbitrarily small @xmath7 value ope expansion of @xmath9 violated in small @xmath7 region take intermediate region where integral saturated by ground state contribution ope well - convergedquantified pole contribution @xmath10 50% total @xcite upperbound @xmath11 ) highest dimension terms @xmath12 10% ope lowerbound @xmath13 ) without @xmath7 constraints stuck with pseudo - peak artifacts wrong conclusions discussed @xcite consider first difficutly ( i ) in qsr analyses multiquark correlators involve many quark lines small fraction momentum scale relevant multiquark states incorpolate more non - perturbative diagrams ope calculations include low energy correlations find @xmath7 window due lack low energy contributions remark low energy dominance multiquark correlator analyses pole - dominace spoiled by large continuum high energy contributions misleading expression problem lack of low energy correlation certain tetraquark correlators inclusion higher dimension terms enables small @xmath7 value low energy dominance worse than meson sum rules better than baryon sum rules @xcite relevant problem is numerical ambiguities in values higher dimension condensates second difficulty ( ii ) multiquark cases low energy correlations might contributions from scattering backgrounds ground states prevents investigating multiquark statesdifficulties profound leading @xmath1 properties of multiquark states tractable corrections treated identifying origins contaminations discuss @xmath0 expansion in application qsr first merit in evaluation higher dimension condensates large @xmath1 factorized into products condensates @xmath14 extend ope calculations to higher dimension terms low energy contributions assumption on @xmath1 scaling low dimension condensates @xmath17 ) @xmath20 assume @xmath21 discuss separate scattering background from multiquark correlators @xmath1 counting quark - gluon graphs related to qualitative picture of hadronic states interpolating fields multiquark state and background assigned different order of @xmath1 mixing occurs in higher order of @xmath0 concentrate on quark - gluon graphs saturated by isolated poles reflect leading @xmath1 properties of multiquark states consider @xmath2 meson with @xmath22 as candidate of tetraquark ( 4q ) state 4q operators @xmath2 quantum number given by @xmath23 products meson operators @xmath24 dirac matrix @xmath25 @xmath26 pauli matrices @xmath27 start @xmath0 classifications quark - gluon graphs 2-point correlators- planar - gluon lines not drawn @xmath1 diagrams start from include 2 planar loops reflect irrelevant 2-meson scattering states studies 4q components require systematic steps beyond @xmath28 proceed next order @xmath29 @xmath20 diagrams multiquark states @xmath20 quark - gluon diagrams for 2 3 point correlators systematic classifications investigate overlap strength operator @xmath30 with hadronic states @xmath0 classification subleading diagrams 2-point correlators consider 3-point correlator among @xmath30 @xmath31 overlap strength 4q field with 2 meson states is @xmath32 diagrams @xmath33 overlap strength @xmath36 with 2q meson state @xmath37 @xmath38 classify hadronic states in 2-point correlators @xmath39 based on @xmath29:2pointgraph - c @xmath34 graphs include free 2 m scattering states @xmath40 @xmath35 contributions vanish absence free 2 meson scattering states @xmath20 graphs include 2 m scattering include 4q poles if @xmath41 different from pseudoscalar axial vector@xmath42 scattering intermediate states not included up to diagrams interactions needed transfer initial states into @xmath43 states suppressed by @xmath0 resonance peaks below @xmath44 isolated no width decay channels absent reduce @xmath2 spectrum 4q correlator into peak(s plus continuum diagrams up to @xmath20 study cases separate investigation @xmath33 @xmath20 step by step analyses for each hadronic state relate @xmath0 to hadronic spectral function through dispersion relation @xmath45 ground state reduced @xmath20 spectra @xmath46 sharp peak due absence decay channels quark - hadron duality to higher excited states @xmath47 borel transform eq.express effective mass as @xmath48 @xmath49 fixed least sensitivity @xcite peakmass against variation @xmath5 physical peak depend on artificial expansion parameter @xmath5 criterion justified when peak narrow @xmath0 reduction of spectra essential qsr framework determine physical parameters results borel analyses concentrate on @xmath20 results off diagonal correlator @xmath51 leading order without factorization violations @xmath20 ope up to dimension 12 numerical analyses with values for @xmath52 @xmath53 @xmath54 @xmath55 @xmath56 results use central value left results of large @xmath1 2q correlators expanded to dimension 6 for vector meson reference scalar meson 2q state @xmath2 meson downarrow indicates values of @xmath13 ( @xmath57 ) fixing criterion fix @xmath58 to 1.0 ( 1.4 ) gev for vector scalar mesons determine mass as 0.65 ( 1.10 ) gev turn to @xmath20 part 4q correlator @xmath51 investigate of 4q states middle panel fig.borelfig effective masses @xmath58=1.0 1.2 1.4 gev select @xmath58=1.2 gev evaluate mass @xmath600.90 gev lower than 2q scalar meson case @xmath601.10 gev @xmath1 limit considered mass 4q state absolute values depend condensate values inequality @xmath61 insensitive right panel @xmath62 @xmath63 2q vector scalar meson masses 4q mass deduced @xmath51 investigated @xmath33 @xmath20 parts @xmath64 @xmath65 ) @xmath33 low energy contribution absent below two meson thresholds acceptable effective mass plots larger @xmath58 value meson thresholds large effective masses @xmath33 part includes free scattering states @xmath0 arguments @xmath20 part @xmath64 similar values @xmath66 gev @xmath51 possible factorization violation investigated 4q correlators @xmath67 stable effective mass plots wide @xmath7 window absence 4q pole in @xmath67 channel @xmath0 expansion useful analyze @xmath1 properties multiquark states separating scattering backgroundsqsr results show @xmath0 arguments provide prediction 4q components @xmath2 meson without @xmath68 correlations @xmath1 analyses multiquark correlators distinguish meson molecules tetraquarks charmed mesons x y z initial states controled by interpolating fields @xmath0 countings observations intermediate states molecular type contributions spectral functions discussions thanks organizers hsqcd hospitality supported by riken brookhaven national laboratory u s. department of energy contract - grant scientific research 20028004 japan hooft . b724 *, 61 ( 1974 ) witten nucl b160 * 57 ( 1979 ) kojo d. jido arxiv:0807.2364 hep - ph shifman vainshtein zakharov nucl . b147 * 385 ( 1979 ) kojo hayashigaki jido phys . 74 * 045206 ( 2006 ) arxiv:0802.2372 hep - ph | 0.505704 | 3,325 | 1,755 | , Saving $0.1 in GPT-4. |
154,878 | to study the infrared behaviour of the propagator , exponentiation of the lowest order spectral function has been known.we show this method is helpful in super renormalizable theory with dimension - full coupling constant.in the @xmath0 approximation anomalous dimension is independent of @xmath1,which plays an important role for confinement and pair condensation . | it has long been known that the infrared behavour of the propagator in the theory with massless particle as photon is determined by low - energy theorem or renormalization group analysis[1,2,3].the answer is known as @xmath2 near @xmath3where @xmath4 is a covariant gauge fixing parameter and @xmath5 is a fine structure constant.@xmath6 is an anomalous dimension of wave function renormalization constant.this formula is valid in the infrared region.but the super renormalizable model as qed@xmath7 suggests the validity of the above type of formula in all region of momentum.if we sum all ladder type diagrams which is dominant in the infrared we obtain the similar form of the propagator with dimensionless coupling constant @xmath8including vacuum polarization the infrared divergences turns out to be soft .
the spectral representation of the propagator is given[1 ] @xmath9 if the intermediate states contain massless particle as photon there are infrared divergences.in the quenched approximation the state @xmath10 stands for a fermion and arbitrary numbers of photons,@xmath11 we have the solution for the spectral function @xmath12 which is written symbolically@xmath13 here the notations @xmath14 have been introduced to show the phase space of each photons.to evaluate the contribution of soft photons ,
first we consider the situation when only the @xmath15-th photon is soft.we define the matrix element @xmath16 we consider @xmath17 for @xmath18continue off the photon mass shell @xmath19 @xmath17 satisfies ward - takahashi - identity@xmath20 formal proof of ward - takahahsi - identity is provided with lsz reduction formula and subsidary condition for photon[4 ] @xmath21 by the low - energy theorem the fermion pole term in @xmath22 is dominant for the infrared singularity in @xmath23.these arise from diagrams in which the soft - photon line with momentum @xmath23 and the incoming fermion line can be separated from the remainder of the diagram by cutting a single fermion line.inclusion of regular terms and their contribution to ward - identities are given for the scalar case[1].hereafter we consider the one - photon matrix element @xmath24 which is given in [ 1,4,5]@xmath25 where @xmath26 is a four component free particle spinor with positive energy.if we sum infinite numbers of photon in the final state as in ( 5),assuming pole dominance for @xmath23 we have a simplest solution to @xmath17 in ( 9 ) @xmath27 from this relation we obtain the @xmath15-photon matrix element @xmath17 as the direct products of @xmath24@xmath28 in this way we have an approximate solution of ( 5 ) by exponentiation of the one - photon matrix element @xmath29@xmath30 the fermion propagator is written explicitly in the following form in quenched case@xmath31),\nonumber\\ \left\vert x\right\vert & = \sqrt{-x^{2}}.\end{aligned}\ ] ] the function @xmath32 is evaluated with exponential cut - off for quenched case[1,3,4 ] @xmath33 where we find mass shift @xmath34here we apply the spectral function of dressed photon in the landau gauge[2,6 ] to evaluate the unquenched fermion proagator .
@xmath35 in this case wave function renormalization is given@xmath36 we have an improved @xmath32 @xmath37 here linear divergent term is regularized by cut - off @xmath38the fermion propagator with @xmath1 flavours in position space is modified to @xmath39@xmath40 it is known@xmath41 if we assume the spectral density for position dependent mass which makes fermion fat@xmath42 we get the propagator in euclid space @xmath43@xmath44 explicit evaluation of @xmath45 is not easy.in fig.1 and fig.2 we see the profile of the scalar part of the propagator in position and momentum space respectively for @xmath46 and quenched linear approximation to dyson - schwinger equation[7].@xmath47 effects of position dependent mass with vacuum polarization correction can be seen by comparison with the quenched linear approximation .
we have the renormalization constant and bare mass in our approximation from ( 22 ) @xmath48@xmath49 there is no pole and it shows the confinement for @xmath50.order parameter @xmath51 is given in position space directly from ( 20)@xmath52 from the above equation we see that @xmath51 @xmath53 and finite only if @xmath54 here we consider the condition of @xmath46 is satisfied or not.in our approximation there is not a definite way to determine physical mass @xmath55the linear approximation to the dyson - schwinger equation in quenched case leads @xmath46[7].in our case for vanishing bare mass physical mass is assumed to be @xmath56 the condition is approximately satisfied.if we add small but finite bare mass it may be satisfied.for @xmath57 case we have@xmath58 , which is the same order as in ref[8,9 ] .
we evaluated the non - perturabative fermion propagator in qed@xmath7 by the method of spectral function.infrared behaviour was significantly modified by position dependent mass for @xmath59 case in comparison with quenched linear aaproximation to dyson - scwinger case.it is very slow to reach high energy behaviour @xmath60at large @xmath1 these effects are suppresed and damps fast as @xmath60
[ 1]r.jackiw , l.soloviev , phys.rev.*137*.3(1968)1485 . | infrared propagator massless particle photon determined by low - energy theorem renormalization group analysis answer @xmath2 @xmath4 covariant gauge fixing parameter @xmath5 fine structure constant anomalous dimension wave function renormalization constant formula valid infrared region super renormalizable model@xmath7 suggests validity momentum ladder diagrams infrared similar form propagator dimensionless coupling constant @xmath8including vacuum polarization infrared divergences soft spectral representation propagator @xmath9 intermediate states contain massless particle photon infrared divergences @xmath10 fermion arbitrary numbers photons@xmath11 solution spectral function @xmath12 written symbolically@xmath13 notations @xmath14 phase space photons evaluate contribution soft photons consider @xmath15-th photon soft define matrix element @xmath16 @xmath17 satisfies ward - takahashi - identity@xmath20 proof lsz reduction formula subsidary condition photon low - energy theorem fermion pole term @xmath22 dominant infrared singularity @xmath23diagrams soft - photon line momentum incoming fermion line separated single fermion line regular terms contribution ward - identities scalar consider one - photon matrix element @xmath24 1,4,5]@xmath25 @xmath26 four component particle spinor positive energy sum infinite numbers photon pole dominance @xmath23 simplest solution @xmath17 9 @xmath15-photon matrix element products @xmath24@xmath28 approximate solution ( 5 ) exponentiation one - photon matrix element @xmath29@xmath30 fermion propagator written quenched case{-x^{2}} function @xmath32 evaluated exponential cut - off quenched case[1,3,4 @xmath33 mass shift apply spectral function dressed photon landau gauge[2,6 evaluate unquenched fermion proagatorwave function renormalization improved @xmath32 linear divergent term regularized by cut - off fermion propagator position space modified @xmath39 known spectral density position dependent mass fermion fat propagator in euclid space evaluation @xmath45 easy.1.2 profile scalar part propagator in position momentum space @xmath46 quenched linear approximation dyson - schwinger@xmath47 effects position dependent mass vacuum polarization correction quenched linear approximation renormalization constant bare mass 22 )@xmath49 no pole confinement @xmath50.order parameter @xmath51 position space ( 20)@xmath52 @xmath51 @xmath53 finite @xmath54 condition @xmath46 satisfied definite determine physical mass linear approximation dyson - schwinger equation leads @xmath46[7] vanishing bare mass physical mass @xmath56 condition satisfied add small finite bare mass satisfied @xmath57@xmath58 same order ref[8,9 evaluated non - perturabative fermion propagator@xmath7 spectral function.infrared behaviour modified by position dependent mass @xmath59 linear dyson - scwingerslow reach high energy behaviour effects suppresed damps fast.jackiw l.soloviev.rev.*137*.3(1968)1485 | 0.258246 | 1,269 | 649 | , Saving $0.0 in GPT-4. |
154,887 | the possible occurence of highly deformed configurations is investigated in the @xmath0ca and @xmath1ni di - nuclear systems as formed in the @xmath2si+ @xmath3c,@xmath2si reactions by using the properties of emitted light charged particles .
inclusive as well as exclusive data of the heavy fragments and their associated light charged particles have been collected by using the icare charged particle multidetector array .
the data are analysed by monte carlo cascade statistical - model calculations using a consistent set of parameters with spin - dependent level densities .
significant deformation effects at high spin are observed as well as an unexpected large @xmath4be cluster emission of a binary nature .
* keywords : * fusion - fission , nuclear deformation , exclusive light charge particle measurements | extensive efforts have been made in recent years , to understand the decay of * light * di - nuclear systems ( a@xmath5 @xmath6 60 ) formed through low - energy ( e@xmath7 @xmath6 10 mev / nucleon ) heavy - ion reactions [ 1 ] . in most of the reactions
studied , the observed fully energy - damped yields of the fragments have been successfully explained in terms of a fusion - fission mechanism [ 1 ] with the noticeable exception of @xmath2si+@xmath3c reaction for which the deep - inelastic orbiting mechanism has been found to be particularly competitive [ 2 ] .
strong resonance - like structures have been observed in elastic and inelastic excitation functions of some specific reactions ( such as @xmath8mg+@xmath8 mg or @xmath2si+@xmath2si ) indicating the presence of shell stabilized , highly deformed configurations in the ( @xmath9cr and @xmath1ni ) compound systems @xcite .
the investigation of the structure of the doubly - magic @xmath1ni nucleus has recently become a subject of great interest @xcite . in a recent experiment using the eurogam multidetector array ,
the present collaboration studied the possibility of preferential population of highly deformed bands in the symmetric fission channel of the @xmath1ni compound nucleus ( cn ) , produced through the @xmath2si+@xmath2si @xcite reaction at e@xmath10 mev , which corresponds to the energy of the conjectured j@xmath11 = 38@xmath12 quasi - molecular resonance @xcite .
the study of light charged particle ( lcp ) emission is a good tool in exploring nuclear deformations and other properties of hot rotating nuclei at high spins and angular momenta @xcite .
the present work has been performed with the aim to investigate the possible occurence of highly deformed configurations of the @xmath0ca and @xmath1ni di - nuclear systems as formed in the @xmath2si+@xmath3c and @xmath2si+@xmath2si reactions , by using the properties of emitted lcp .
inclusive as well as exclusive data of the heavy fragments ( a @xmath13 6 ) and their associated light charged particles ( p , d , t , and @xmath14-particles ) emitted in both reactions at @xmath15(@xmath2si ) = 112 , 180 mev have been measured @xcite using the * icare * multidetector array @xcite .
the lcp s emitted from ff fragments may provide the deformation properties of these fragments .
moreover , the in - plane angular correlations data will be used to extract the emitters temperatures . in this paper
we will present the exclusive data of lcp s in coincidence with er .
the two experiments were performed at the ires strasbourg vivitron tandem facility using 112 mev and 180 mev @xmath2si beams on @xmath3c ( 160 @xmath16g/@xmath17 thick ) and @xmath2si ( 230 @xmath16g/@xmath17 thick ) targets respectively @xcite .
both the heavy ions ( a @xmath13 6 ) and their associated lcp s ( p , d , t , and @xmath14 ) were detected using the * icare * charged particle multidetector array @xcite which consists in nearly 40 telescopes in coincidence .
the heavy fragments , i.e. er , quasi - elastic ( qe ) , di , and ff fragments , were detected in 10 telescopes , each consisting of an ionisation chamber ( ic ) followed by a 500 @xmath16 m si detector .
the in - plane coincident lcp s were detected using either 3 three - elements telescopes ( si 40 @xmath16 m , si 300 @xmath16 m , 2 cm csi(tl ) ) or 24 two - elements telescopes ( si 40 @xmath16 m , 2 cm csi(tl ) ) and two double telescopes ( ic , si 500 @xmath16 m ) located at the most backward angles .
the ic s were filled with isobutane and the pressures were kept at 30 torrs and at 60 torrs for detecting heavy fragments and light fragments , respectively .
the acceptance of each telescope was defined by thick aluminium collimators .
the lcp spectra obtained in both of the reactions have maxwellian shapes with an exponential fall - off at high energy ( their shape and high - energy slopes are essentially independent of c.m .
angle ) which reflects a relatively low temperature of the decaying nucleus .
this is the signature of a statistical de - excitation process arising from a thermalized compound nucleus like source .
the energy spectra of the coincident @xmath14-particles , for the reactions @xmath2si+@xmath2si and @xmath2si+@xmath3c at e@xmath7=180mev , have been displayed in figs . 1 and 2 respectively @xcite .
+ the analysis of the data has been performed using cacarizo @xcite , the monte carlo version of the statistical - model code cascade @xcite .
the informations on the main ingredients of the statistical description , _
i.e. , _ the nuclear level densities and the barrier transmission probabilities , are usually obtained from the study of the evaporated light particle spectra . in recent years , it has been observed that the standard statistical model could not predict satisfactorily , the shape of the evaporated @xmath14-particle energy spectra @xcite and the measured average energies of the @xmath14-particles were much lower than the corresponding theoretical predictions .
several attempts have been made in the past few years to explain this anomaly either by changing the emission barrier or using a spin - dependent level density @xcite .
in hot rotating nuclei formed in heavy - ion reactions , the energy level density at higher angular momentum is spin dependent .
the level density , @xmath18 , for a given angular momentum @xmath19 and energy @xmath20 is given by the well known fermi gas expression : where @xmath22 the level density parameter is taken equal to a/8 , t is the nuclear " temperature , and @xmath23 is the pairing correction , e@xmath24 = @xmath25j(j+1 ) is the rotational energy , @xmath26 is the effective moment of inertia , @xmath27 is the rigid body moment of inertia , and @xmath28 and @xmath29 are deformation parameters @xcite . by changing the deformability parameters @xmath28 and
@xmath29 one can simulate the spin - dependent level density @xcite .
the cacarizo calculations have been performed using two sets of input parameters : one with a standard set consistent with the deformation of the finite range liquid drop model ( frldm ) @xcite , and another with a spin - dependent moment of inertia and non - zero values for the deformability parameters .
as observed at e@xmath7= 112 mev data @xcite , the energy spectra of the @xmath14 spectra is well reproduced after including the deformation .
the dashed lines in fig . 1 show the predictions of cacarizo using the standard frldm deformation parameter set .
it is clear that the average energies of the measured @xmath14 energy spectra are lower than those predicted by the standard statistical - model calculations .
the solid lines show the predictions of cacarizo using the paramater set with non- zero values @xmath28 and @xmath29 .
the exclusive energy spectra of @xmath14-particle measured in coincidence with individual s and p er s , which are shown in fig .
2 for the reaction @xmath2si+@xmath3c at e@xmath7 = 180 mev , are quite interesting .
the spectra associated with s are completely different from those associated with p. the latter are reasonably well reproduced by the cacarizo curves whereas the model could not predict the shape of the spectra obtained in coincidence with s. an additional non - statistical components appear to be significant in this case .
the similar results have been observed at the lower bombarding energy e@xmath7 = 112 mev @xcite .
these non - statistical components may come from the decay of unbound @xmath4be ( in g.s . and the two first excited states ) produced in the binary reaction @xmath0ca
@xmath30 @xmath31s+@xmath4be of a massive transfer type @xcite
. + s.j .
sanders , a. szanto de toledo , and c. beck , phys .
rep . * 311 * , 487 ( 1999 ) ; and references therein .
d. shapira et al .
lett.*53 * , 1634 ( 1984 ) .
zurmhle et .
al . , phys . lett . * 129b , 384 ( 1983 )
al . , phys .
lett . * 47 , 23 ( 1981 ) ; r. r.betts et.al.,phys . lett . * 100b , 117 ( 1981 ) .
d. rudolph et .
al . , phys .
lett . * 82 , 3763 ( 1999 ) .
t. mizusaki et .
al . , phys .
* t88 * , 107 ( 2000 ) .
svensson et .
* 82 , 3400 ( 1999 ) . c. h. yu et .
al . , phys .
c * 60 , 031305 ( 1999 ) .
a.o . macchiavelli et .
al . , phys .
b * 480 , 1 ( 2000 ) .
r. nouicer et .
al . , phys .
c * 60 * , 41303 ( 1999 ) . c. beck et .
al . , phys .
c * 63 * ( 2001 ) ( in press jan .
g.viesti et .
al . , phys .
c * 38 * , 2640 ( 1988 ) .
al . , phys .
b * 197 * , 515 ( 1987 )
. g. la rana et .
al . , phys .
c * 37 * , 1920 ( 1988 ) ; _ ibidem _ c * 40 , 2425(1989 ) .
huizenga et .
c * 40 * , 668 ( 1989 ) .
al . , phys .
c * 57 * , 1269 ( 1998 ) ; _ ibidem _ phys . lett . * b307 , 283 ( 1993 ) . c. bhattacharya et .
al . , nucl . phys . * a654 * , 841 ( 1999 ) .
m. rousseau et al .
, proceed . of _
7@xmath32 cluster99 conference _ , rab , croatia , june 14 - 19 , 1999 , eds .
m. korolija , z. basrak , and r. caplar , _ world scientific _ , 189 ( 2000 ) . * * * * * * * * * m. rouseeau , ph .
d. thesis , strasbourg university , 2000 ( work in progress ) .
t. bellot , ph .
d. thesis , strasbourg university , report * ires 97 - 34*. f. phlhofer , nucl .
* a280 , 267 ( 1977 ) .
sierk , phys .
c * 33 * , 2039 ( 1986 ) . n. arena et al . ,
c * 50 * , 880 ( 1994 ) . * | efforts understand decay of light di - nuclear systems@xmath5 @xmath6 60 low - energy@xmath7 @xmath6 10 mev / heavy - ion reactions energy - damped yields explained fusion - fission mechanism exception @xmath2si+@xmath3c reaction deep - inelastic orbiting mechanism competitive strong resonance - like structures observed in elastic inelastic excitation functions reactions @xmath8mg+ @xmath2si+ indicating shell stabilized deformed configurations in @xmath9cr @xmath1ni compound systems investigation structure doubly - magic @xmath1ni nucleus studied population highly deformed bands in symmetric fission channel @xmath1ni compound nucleus @xmath2si+@xmath2si @xcite reaction at e@xmath10 mev corresponds energy j@xmath11 = 38@xmath12 quasi - molecular resonance study of light charged particle ) emission exploring nuclear deformations properties hot rotating nuclei at high spins angular momentapresent work deformed configurations @xmath0ca @xmath1ni di - nuclear systems in @xmath2si+@xmath3c reactions using properties emitted lcp data heavy fragments @xmath13 6 ) light charged particles p d @xmath14 reactions @xmath15 ) = 112 180 mev measured @xcite * icare * multidetector array lcp s from ff fragments may provide deformation properties in - plane angular correlations data extract emitters temperatures exclusive data of lcp s coincidence with er experiments performed at ires strasbourg vivitron tandem facility using 112 mev 180 mev @xmath2si beams on @xmath3c @xmath2si 230 targets heavy ions 6 associated lcp s p d t @xmath14 ) detected using * icare * charged particle multidetector array @xcite 40 telescopes heavy fragments er ff fragments detected in 10 telescopes ionisation chamber 500 @xmath16 m si detectorin - plane coincident lcp s detected using 3 three - elements telescopes 300 or 24 two - elements telescopes two double telescopes 500 at backward angles ic s filled with isobutane pressures kept at 30 and 60 torrs for detecting heavy and light fragments acceptance telescope defined by thick aluminium collimators lcp spectra in reactions have maxwellian shapes exponential fall - off at high energy independent of c.m angle reflects low temperature of decaying nucleus signature of statistical de - excitation process from thermalized compound nucleus energy spectra of coincident @xmath14-particles for reactions @xmath2si+ and+@xmath3c at=180mev displayed in figs . 1 and 2 analysis using cacarizo @xcite version statistical - model code cascade informations on statistical description nuclear level densities barrier transmission probabilities obtained from study evaporated light particle spectra standard statistical model predict shape of evaporated @xmath14-particle energy spectra measured average energies lower than theoretical predictionsattempts explain anomaly changing emission barrier or using spin - dependent level density @xcite in hot rotating nuclei heavy ion energy level density at higher angular momentum spin dependent level density @xmath18 for momentum energy @xmath20 given by fermi gas expression @xmath22 level density parameter equal to a/8 t nuclear temperature pairing correction@xmath24 =(j+1 rotational energy effective moment inertia rigid body moment inertia deformation parameters changing deformability parameters simulate spin - dependent level density cacarizo calculations performed using two input parameters standard set deformation spin - dependent moment inertia non - zero values deformability parameters 112 mev data energy spectra @xmath14 spectra reproduced after deformation dashed lines fig 1 show predictions cacarizo using standard frldm deformation parameter set average energies of measured @xmath14 energy spectra lower than statistical - model calculations solid lines show predictions cacarizo using non- zero values @xmath28 @xmath29energy spectra @xmath14-particle measured with s p shown fig . 2 reaction @xmath2si+@xmath3c at e@xmath7 = 180 mev interesting spectra s different from p reproduced by cacarizo curves model predict shape spectra s additional non - statistical components significant similar results observed lower bombarding energy e@xmath7 = 112 mev @xcite components from decay unbound @xmath4be first excited states binary reaction @xmath0ca @xmath30 @xmath31s+@xmath4be massive transfer type @xcite sanders szanto de toledo beck . rep. 311 * , 487 ( 1999 ) shapira*53, 1634 ( 1984 ) zurmhle 129b , 384 ( 1983 ) 47 , 23 ( 1981 ) r.betts 100b , 117 ( 1981 ) rudolph phys 82 , 3763 ( 1999 ) mizusaki phys t88, 107 ( 2000 ) svensson 82 , 3400 ( 1999 ) yu phys 60 , 031305 ( 1999 )macchiavelli phys . b * 480 1 ( 2000 ) nouicer phys c 60 * 41303 ( 1999 ) beck phys c 63 * 2001 ).viesti c 38 2640 ( 1988 ) b 197 515 ( 1987 ) la rana phys c 37 * 1920 ( 1988 ) 40 2425(1989 ) huizenga 40 668 ( 1989 ) c 57 * 1269 ( 1998 ) phys b307 283 ( 1993 ) bhattacharya . a654 841 ( 1999 ) rousseau 7@xmath32 cluster99 conference croatia june 14 - 19 1999 korolija basrak caplar world scientific 189 ( 2000 ) rouseeau ph . d. thesis strasbourg university 2000 bellot ph d. thesis university report 97 - 34* phlhofer nucl . a280 , 267 ( 1977 ) sierk phys . c * 33 * 2039 ( 1986 ) arena c * 50 * 880 ( 1994 ) | 0.358457 | 2,619 | 1,335 | , Saving $0.1 in GPT-4. |
154,894 | high energy photon colliders ( , ) based on backward compton scattering of laser light is a very natural addition to linear colliders . in this report
we consider mainly this option for the tesla project .
recent study has shown that the horizontal emittance in the tesla damping ring can be further decreased by a factor of four . in this case the luminosity luminosity in the high energy part of spectrum can reach 0.30.5 @xmath0 .
typical cross sections of interesting processes in collisions are higher than those in collisions by about one order of magnitude , so the number of events in collisions will be more that in collisions .
the key new element in photon colliders is a very powerful laser system .
the most straightforward solution is `` an optical storage ring ( optical trap ) '' with diode pumped laser injector which is today technically feasible .
this paper briefly review the status of a photon collider based at tesla , its possible parameters . | over the last decade , several laboratories in the world have been working on linear collider projects with an energy from several hundreds gev up to several tev . beside collisions
, linear colliders can `` convert '' electrons to high energy photons using the compton backscattering of laser light , thus obtaining and @xmath1 collisions with energies and luminosities close to those in collisions @xcite .
recently the international workshop on high energy photon colliders has been held at desy .
there are very good news , which i will discuss below . the basic scheme of a photon collider is shown in fig .
[ ggcol ] .
the maximum energy of the scattered photons ( in direction of electrons ) is @xcite @xmath2 \left[\frac{\omega_0}{ev}\right],\ ] ] where @xmath3 is the electron energy , @xmath4 the energy of the laser photon , @xmath5 is the angle between electron and laser beam ( see , fig.[ggcol]a ) .
for example : @xmath3 = 250 gev , @xmath6 ev ( @xmath7 ) ( nd : glass and other powerful solid state lasers ) @xmath8 @xmath9 and @xmath10 .
for an introduction to photon colliders see refs @xcite . below
i will discuss most important issues connected with a photon collider based on the tesla @xcite .
the following essential topics are discussed : physics motivation , parameters of the photon collider at tesla , lasers , optics .
in general , physics in and , collisions is quite similar because the same particles can be produced .
however , it is always better to study new phenomena in various reactions because they give complementary information .
some phenomena can best be studied at photon colliders due to better accuracy or larger accessible masses .
the second aspect important for physics motivation is the luminosity .
typical luminosity distribution in and collisions has a high energy peak and some low energy part ( see the next section ) .
this peak has a width at half maximum of about 15% ( in collision ) and photons here can have a high degree of polarization . in the next section
we will see that in the current tesla designs the luminosity in the high energy peak can be up to 30 - 40% of the luminosity at the same beam energy .
* higgs boson * the present standard model ( sm ) assumes existence of a very unique particle , the higgs boson .
it has not been found yet , but from existing experimental information it follows that , if it exists , its mass is higher then 112 gev ( lep200 ) and is below 200 gev , i.e. lays in the region of the next linear colliders . in the simplest extensions to the sm
the higgs sector consists of five physics states : @xmath11 and @xmath12 .
all these particles can be studied at photon colliders and some characteristics can be measured better than in collisions .
the process @xmath13 goes via the loop with heavy virtual charged particles and its cross section is very sensitive to the contribution of particles with masses far beyond the energies covered by present and planned accelerators . for the integrated luminosity 50 fb@xmath14 ( in the peak )
the number of produced higgs will be 50150 thousands ( depending on the mass ) . as a result
, one can measure the @xmath15 width at photon colliders with an accuracy better than 2 - 3% @xcite .
this is sufficient for distinguishing between higgs models @xcite .
moreover , in the models with several neutral higgs bosons , heavy h@xmath16 and a@xmath16 bosons have almost equal masses and at certain parameters in the theory are produced in collisions only in associated production @xmath17 , while in collision they can be produced singly with sufficiently high cross section .
correspondingly , in collisions one can produce higgs bosons with about 1.5 times higher masses . * charge pair production * the second example is the charged pair production .
cross sections for the production of charged scalar , lepton , ww pairs in collisions are larger than those in collisions by a factor of approximately 520 .
the corresponding graphs can be found elsewhere @xcite .
note , that in collisions two charged pairs are produced both via annihilation diagram with virtual @xmath18 , @xmath19 and also via exchange diagrams where some new particles can give contributions , while in collisions it is pure a qed process which allows the charge of produced particles to be measured unambiguously .
* accessible masses * in collisions , charged particle with a mass higher than that in collisions can be produced ( a heavy charged particle plus a light neutral ) ; for example , supersymmetric charged particle plus neutralino or new w boson and neutrino .
also , collisions provide higher accessible masses for particles which are produced as a single resonance in collisions ( such as neutral higgs bosons ) . * search for anomalous interactions * precise measurement of cross sections allow the observation of effects of anomalous interactions .
the process @xmath20 ww has large cross section ( about 80 pb ) and it is one of most sensitive processes for a search for a new physics .
the vertex @xmath21 can be studied much better than in collisions because in the latter case the cross section is much smaller and this vertex gives only 10% contribution to the total cross section .
the two factors together give about 40 times difference in the cross sections .
besides that , in collisions the @xmath22 vertex can be studied . * quantum gravity effects with extra dimensions * this new theory suggests a possible explanation of why gravitation forces are so weak in comparison with electroweak forces .
it is suggested that the gravitation constant is equal to the electroweak but in a space with extra dimensions .
it can be tested at photon colliders and a two times higher mass scale than in collisions can be reached @xcite .
many other examples can be found in proceedings of gg2000 workshop @xcite .
as it is well known in collisions the luminosity is restricted by beamstrahlung and beam instabilities . due to the first effect
the beams should be very flat . in collisions
these effects are absent , therefore one can use beams with much smaller cross section . at present tesla beam parameters
the luminosity is determined only by the attainable geometric luminosity .
so , the luminosity depends on emittances of electron beams .
specially for photon collider the tesla group has studied the possibility of decreasing emittances at the tesla damping ring .
the conclusion is that the horizontal emittance can be reduced by a factor of 4 in comparison with the previous design .
the resulting parameters of the photon collider at tesla for 2e=500 gev and h(130 ) are presented in table 1 @xcite .
it is assumed that electron beams have 85% longitudinal polarization and laser photons have 100% circular polarization and that @xmath23 .
the thickness of laser target is one collision length ( so that @xmath24 ) .
.parameters of the collider based on tesla . left column for 2e=500 gev , next two columns for higgs with m=130 gev , two options . [ cols="<,^,^,^ " , ] [ table1 ] for these luminosities the rate of production of the sm higgs boson with m@xmath25=130(160 ) gev in collisions is 0.9(3 ) of that in collisions at 2e = 500 gev ( both reactions , zh and h@xmath26 ) . comparing the luminosity with the luminosity ( @xmath27 for @xmath28 gev ) we see that for the same energy @xmath29 for example , the cross section for production of @xmath30 pairs in collisions of polarized photons is higher than that in collisions by a factor of 20 ( not far from the threshold ) ; this means 8 times higher production rate for the luminosities given above .
the relation ( [ lgge+e- ] ) is valid only for the considered beam parameters .
a more universal relation is ( for @xmath31 ) @xmath32 figures for the luminosity distribution in and collisions can be found elsewhere @xcite .
a key element of photon colliders is a powerful laser system which is used for e@xmath33 conversion .
lasers with the required flash energies ( several j ) and pulse duration @xmath34 1 ps already exist and are used in several laboratories .
the main problem is the high repetition rate , about 1015 khz , with a pulse structure repeating the time structure of the electron bunches .
the most attractive and reliable solution at this moment is an `` optical storage ring '' with a diode pump laser injector .
this new approach can be considered as a base - line solution for the tesla photon collider @xcite .
this scheme with multiple use of each laser bunch is shown in fig .
[ loop ] . [ loop ]
the laser pulses are send to the interaction region where they are trapped in an optical storage ring .
this can be done using pockels cells ( p ) , thin film polarizers ( tfp ) and 1/4-wavelength plates ( @xmath35 ) .
each bunch makes several ( n ) round trips ( i assume n=10 for further example ) and then is deleted from the ring .
all these tricks can be done by switching one pockels cell . during one total loop each bunch
is used for conversion twice . to avoid problems of non - linear effects ( self - focusing ) in optical elements , laser pulses which are compressed before collisions down to about 2 ps using grating pairs
are then stretched again up to previous length before passing through optical elements .
such system can be done in the best way only at the tesla due to a large spacing between bunches and a long train . a laser system required for a such optical storage ring with 10 round trips can consist of about 8 lasers of 0.9 kw average powereach . due to the high average power
the lasers should be based on diode pumping .
diodes have much higher efficiency than flash lamps and much more reliable .
this technology is developed very actively for other application , such as inertial fusion .
present cost of diodes for such laser system is about 15 m$ ( for 10-fold use of one laser bunch as described above ) and it is expected that their cost will be further decreased several times .
such system can be done now : all technologies exist .
the luminosity in collisions ( in the high energy peak ) can reach about 40% of luminosity . since cross sections in collisions are typically higher by one order of magnitude than those in collisions and because of access to higher masses for some particles , the photon collider now has very serious physics motivation .
v. telnov , proc . of the international conference on the structure and interactions of the photon ( photon 99 ) , freiburg , germany , 23 - 27 may 1999 , to be published in nucl .
b , e - print : hep - ex/9908005 . proc . of intern .
workshop on high energy photon colliders , 14 - 17 june , 2000 , desy , hamburg , germany , to be published in nucl .
inst . and methods a. thomas rizzo , _ proceedings of 4th international workshop on linear colliders ( lcws 99 ) _ , sitges , barcelona , spain , 28 apr - 5 may 1999 , slac - pub-8204 , e - print archive : hep - ph/9907401 , also these proceedings . | last decade laboratories on linear collider projects energy hundreds gev to tev linear colliders convert electrons to high energy photons using laser light obtaining collisions with energies luminosities close to @xcite international workshop on high energy photon colliders held at desy basic scheme photon collider shown in fig . [ ggcol ] maximum energy of scattered photons is @xcite @xmath2 @xmath3 electron energy @xmath4 energy laser photon @xmath5 angle between electron and laser beam example @xmath3 = 250 gev @xmath6 ev @xmath7 @xmath8 @xmath9 @xmath10 introduction to photon colliders see refs @xcite important issues photon collider based tesla @xcite topics physics motivation parameters photon collider tesla lasers optics physics collisions similar same particles produced better to study new phenomena in reactions complementary information some phenomena best studied at photon colliders better accuracy larger accessible masses aspect luminosity typical luminosity distribution in collisions high energy peak low energy partpeak width half maximum 15% in collision photons high polarization current tesla designs luminosity in high energy peak 30 - 40% same beam energy higgs boson present standard model assumes unique particle higgs boson not found if exists mass higher then 112 gev ( lep200 ) below 200 gev region next linear colliders higgs sector five physics states @xmath11 @xmath12 particles studied at photon colliders characteristics measured better than collisions process @xmath13 loop with heavy virtual charged particles cross section sensitive to particles with masses beyond energies accelerators integrated luminosity 50 fb@xmath14 peak number produced higgs 50150 thousands depending on mass @xmath15 width at photon colliders accuracy better than 2 - 3% @xcite sufficient for distinguishing between higgs models models neutral higgs bosons heavy h@xmath16 a@xmath16 bosons equal masses produced in collisions only production @xmath17 collision produced singly with high cross section collisions produce higgs bosons with 1.5 times higher masses charge pair production second example charged pair productioncross sections for charged scalar lepton ww pairs in collisions larger by approximately 520 graphs found @xcite in collisions two charged pairs produced via annihilation diagram @xmath18 @xmath19 exchange diagrams collisions qed process allows charge particles measured unambiguously accessible masses collisions charged particle mass higher produced heavy light neutral supersymmetric charged particle neutralino new boson neutrino collisions provide higher masses for particles single resonance neutral higgs bosons search for anomalous interactions precise measurement cross sections allow observation effects anomalous interactions process @xmath20 ww large cross section 80 pb sensitive for new physics vertex @xmath21 studied better collisions cross section smaller 10% contribution to total cross section two factors give 40 times difference in cross sections in collisions @xmath22 vertex studied quantum gravity effects with extra dimensions new theory suggests gravitation forces weak electroweak forces gravitation constant equal to electroweak in space with extra dimensions tested at photon colliders two times higher mass scale than collisions @xcite examples in proceedings gg2000 workshop @xcitein collisions luminosity restricted by beamstrahlung instabilities beams should be flat collisions effects absent beams smaller cross section tesla beam parameters luminosity determined by attainable geometric luminosity depends on emittances electron beams for photon collider tesla group studied decreasing emittances at tesla damping ring horizontal emittance reduced by factor of 4 parameters of photon collider at tesla for 2e=500 gev h(130 ) in table 1 assumed electron beams have 85% longitudinal polarization laser photons 100% circular polarization thickness of laser target is one collision length.parameters of collider based tesla left column for 2e=500 gev columns for higgs with m=130 gev rate of production of higgs boson with m@xmath25=130(160 ) gev in collisions is 0.9(3 ) at 2e = 500 gev luminosity with same energy cross section for production of pairs in collisions polarized photons higher by factor of 20 8 times higher production rate for luminositiesrelation ( [ lgge+e- ] ) valid for beam parameters universal relation @xmath31 ) @xmath32 figures luminosity distribution collisions found @xcite key element photon colliders powerful laser system for e@xmath33 conversion lasers with required flash energies pulse duration @xmath34 1 ps exist used laboratories problem high repetition rate 1015 khz pulse structure repeating time structure electron bunches solution optical storage ring with diode pump laser injector new approach base - line solution for tesla photon collider @xcite scheme multiple use laser bunch shown in fig . loop laser pulses send to interaction region trapped in optical storage ring pockels cells thin film polarizers 1/4-wavelength plates each bunch makes n round trips deleted from ring switching one pockels cell each bunch used for conversion twice non - linear effects laser pulses compressed before collisions to 2 ps stretched previous length before optical elements system at tesla large spacing long train laser system for optical storage ring with 10 round trips 8 lasers 0.9 kw average powereach lasers based on diode pumping diodes higher efficiency than flash lamps more reliabletechnology developed for inertial fusion cost of diodes laser system 15 m$ 10-fold use laser bunch expected cost system technologies exist luminosity in collisions high energy peak 40% luminosity cross sections collisions higher higher masses photon collider has serious physics motivation v. telnov international conference on structure interactions photon 99 freiburg germany 23 - 27 may 1999 published in nucl . e - print hep - ex/9908005 workshop on high energy photon colliders 14 - 17 june 2000 hamburg germany published in nucl . inst methods thomas rizzo proceedings 4th international workshop on linear colliders 99 sitges barcelona spain 28 apr - 5 may 1999 slac - pub-8204 e - print hep - ph/9907401 | 0.35879 | 2,548 | 1,245 | , Saving $0.1 in GPT-4. |
154,897 | the vortex states in a thin mesoscopic disk are investigated within the phenomenological ginzburg - landau theory in the presence of a step - like external magnetic field with zero average which could model the field resulting from a ferromagnetic disk or a current carrying loop .
the regions of existence of the multi - vortex state and the giant vortex state are found .
we analyzed the phase transitions between them and found regions in which ring - shaped vortices contribute .
furthermore , we found a vortex state consisting of a central giant vortex surrounded by a collection of anti - vortices that are located in a ring around this giant vortex . | recent progress in microfabrication and measurement techniques makes it possible to study the properties of mesoscopic superconducting samples , and , in particular , disks with sizes comparable to the penetration depth @xmath0 and the coherence length @xmath1 @xcite .
depending on the disk size various superconducting states can exist in such samples in an external magnetic field . in small disks ,
where the confinement dominates , there exist only giant vortex ( @xmath2 ) states ( circular symmetric states with a fixed value of angular momentum @xmath3 ) @xcite . for sufficiently large disks the @xmath4 state can break up into the multi - vortex ( @xmath5 ) state ( a mixture of giant vortices with different angular momentum ) @xcite .
also ring - shaped vortex states with larger energy than @xmath2 and @xmath5 ones were predicted @xcite .
these ring - shaped two - dimensional vortex states have a cylindrically symmetric magnetic field profile and they are different from the ring - vortices that were e.g. found in three - dimensional superfluid liquid helium .
but the analysis of ref . @xcite showed that in a superconducting disk in the presence of a homogeneous magnetic field the ring - shaped vortices are unstable . in the present paper
we study the @xmath5 states of a thin superconducting disk in an inhomogeneous magnetic field @xmath6 with step - like profile .
this magnetic field profile could approximate the field resulting from a magnetic dot ( they are usually used in experiments , e.g. see ref .
@xcite ) as well as the field of a current loop ( fig . [ fig1 ] ) .
we found that this model field profile captures a lot of interesting physics and that the step height and width strongly influence the superconducting phase diagram .
the details of our approach can be found in refs .
@xcite and , therefore , we briefly sketch only the necessary steps .
we restrict ourselves to sufficiently thin disks such that @xmath7 . in this case , to a first approximation , the magnetic field due to the circulating superconducting currents may be neglected and the magnetic field inside the disk equals the external one .
following refs .
@xcite , the order parameter of the @xmath5 state is written as a linear combination of eigenfunctions of the _ linearized _ ginzburg - landau equation @xmath8 where @xmath9 is the radial distance from the disk center , @xmath10 is the azimuthal angle , and @xmath11 enumerates the different radial states for the same @xmath12 . in the homogeneous magnetic field case @xmath3 is the value of the effective total angular momentum which is equal to the number of vortices in the disk .
later we will see that , for our inhomogeneous magnetic field case , the assignment of the total vorticity can be tricky . substituting eq .
( [ lin ] ) in the free energy ( measured in @xmath13 units ) @xmath14 \right\ } \text { , } \label{free_en}\ ] ] ( here @xmath15 is the local magnetic field , @xmath16 is the vector potential ) we obtain @xmath17 as a function of the complex parameters @xmath18 .
minimization of @xmath17 with respect to these parameters allows to find the equilibrium vortex configurations and to determine their stability .
we restrict ourselves to states built up by only two components in eq .
( [ lin ] ) .
it brings quantitative bounds in our analysis but , nevertheless , gives the correct qualitative behavior and facilitates the physical insight into the problem . using two different approaches from refs .
@xcite @xmath5 configurations with three and five components in eq .
( [ lin ] ) were checked .
this resulted in 1 ) a reduction of the region of existence of metastable @xmath5 states at the low magnetic field limit , 2 ) an appearance of additional unstable states corresponding to the saddle points of the free energy function , and 3 ) no new vortex configurations in the ground state .
thus , two components in eq .
( [ lin ] ) are sufficient to describe the vortex structure in the ground state of our system .
depending on the radial state of the components we defined two different configurations : _ giant - giant _ ( @xmath19 ) and _ giant - ring _ ( @xmath20 @xmath21 ) @xmath5 states .
the first represents a combination of two @xmath22 states while the second is composed of one giant and one ring vortex .
first , we investigate the case of a centered step - like magnetic field corresponding to a magnetic disk . the effect of the size of the superconducting disk on the vortex state is shown in the phase diagram of fig .
the parameters considered are @xmath23 and @xmath24 .
the solid lines indicate where the ground state of the free energy changes from one state to another and dashed lines correspond to transitions between different @xmath2 states as metastable states .
the right axis of the diagram shows the corresponding positive flux through the region @xmath25 .
one can clearly see the _ re - entrant _ behavior .
for example , for @xmath26 we observe the following change of total vorticity @xmath27 .
notice that the ground state with total vorticity @xmath28 and @xmath29 covers the largest part of the phase diagram . with increasing disk size ,
all other @xmath2 states are strongly suppressed in favor of the various @xmath5 states . for small disk sizes different @xmath2 states are present as the ground state . but with increase of disk size , islands with different @xmath5 configurations ( _ giant - ring _ and _ giant - giant _ @xmath5 states ) dominate the ground state diagram . for @xmath30 ,
a giant - ring @xmath5 state appears , and for @xmath31 we obtain giant - giant states . as one can see in fig .
[ fig2 ] , for example , for @xmath32 and with increasing magnetic field @xmath33 , giant - giant combinations @xmath34 , @xmath35 , @xmath36 etc .
become the ground state . in the contour plot of the @xmath37-distribution for these states we observe single vortices which are arranged on a ring with a low density area ( see fig .
[ fig3](a ) , blue color - zero density , red - highest density ) situated in the center of the disk ( see fig .
[ fig3](a ) , for the @xmath38 state ) .
this central area is associated with a giant vortex , and encircling it leads to a phase change showing the vorticity @xmath39 .
[ fig3](b ) shows the contour plot of the corresponding phase with seven anti - vortices arranged in a circle around the giant vortex and the _ total _ vorticity is @xmath28 .
it is important to emphasize that for the disk size @xmath40 and with changing of @xmath41 _ giant - giant _ @xmath5 states consist only of vortices arranged radially symmetric with no anti - vortices present .
thus , it is obvious that enlarging the superconducting disk enhances the influence of the negative part of the step - like magnetic field . as a result
we obtain a new giant - giant form called _ giant multi anti - vortex _ states .
the total vorticity is now equal to the lowest vorticity of the giant vortex states of which the multi - vortex consists of .
this inevitably leads to a re - entrant behavior in the total vorticity , which is in agreement with ref .
@xcite where it was found numerically that several vortices can enter ( or exit ) at once for disks with sufficiently large radius . for @xmath40 disks we only found that @xmath42 transitions are possible . with increase of the disk radius the @xmath35 and @xmath36 _ giant - multi anti - vortex
_ states become @xmath43 and @xmath44 , respectively . for @xmath45 ,
the multi - vortex states become metastable again and the @xmath28 and @xmath29 states are now the only ground states . moreover , for @xmath46 we have the meissner state for all values of the applied magnetic field .
next , we move the step - like field profile along the radius of the superconducting disk , i.e. consider a ring magnetic field profile as in the case of a current loop . in fig .
[ fig4 ] , we present the phase diagram as function of @xmath47 . the parameters of the magnetic field profile are @xmath48 , @xmath49 , and @xmath32 .
thick solid curves indicate the transitions between different vortex states and dashed lines denote transitions between metastable @xmath2 states .
the _ re - entrant _ behavior , as in previous case , is clearly visible .
the most important result is that the @xmath5 states dominate this diagram .
moreover , we have both types of multi - vortices , i.e. _ giant - giant _ and _ giant - ring _ , as ground state , and with shifting the field profile towards the disk periphery , giant - giant @xmath5 configurations cover most of the superconducting region .
the giant - giant @xmath5 states in this case appear both as states with no anti - vortices present , and as _ giant multi anti - vortex _ states . following the transition lines ,
a correspondence between different giant - multi anti - vortex states @xmath50 can be seen : @xmath51 remains the same , while @xmath52 increases from @xmath53 to @xmath54 .
as one can see , the latter is strongly correlated with _
re - entrant _ behavior . in the rest of the diagram ,
the @xmath5 states with the classical geometry ( both _ giant - giant _ and _ giant - ring _ ) dominate . however , a difference between those states exists . as shown in fig .
[ fig5 ] , in the case of a _ giant - ring _ state , vortices are preferentially distributed within a ring - shaped lower density area .
further , considering the high density areas at the disk periphery , a shift in phase of @xmath55 is observed , where @xmath3 is the total vorticity .
although these states have a different origin sometimes they can exhibit a similar distribution of vortices but the phase is always able to distinguish between them .
this work was supported by the flemish science foundation ( fwo - vi ) , the belgian inter - university attraction poles ( iuap - iv ) , the `` onderzoeksraad van de universiteit antwerpen '' ( goa ) , and the esf programme on `` vortex matter '' . | progress in microfabrication measurement techniques properties of mesoscopic superconducting samples disks with sizes comparable to penetration depth coherence length @xmath1 depending on disk size superconducting states exist in in external magnetic field in small disks confinement giant vortex ( @xmath2 ) states circular symmetric fixed angular momentum large disks @xmath4 state can break up into multi - vortex ( @xmath5 ) state ring - shaped vortex states with larger energy than predicted have cylindrically symmetric magnetic field profile different from ring - vortices in three superfluid liquid helium analysis showed in superconducting disk homogeneous magnetic field ring - shaped vortices unstable present paper study @xmath5 states of thin superconducting disk in inhomogeneous magnetic field @xmath6 with step - like profile magnetic field profile approximate field from magnetic dot field of current loop model field profile captures interesting physics step height width influence superconducting phase diagram details of approach in refs . @xcite sketch necessary stepsrestrict to thin disks @xmath7 magnetic field due circulating superconducting currents neglected field inside disk equals external @xcite order parameter @xmath5 state linear combination eigenfunctions ginzburg - landau equation @xmath8 @xmath9 radial distance from disk center @xmath10 azimuthal angle @xmath11 radial states for @xmath12 homogeneous magnetic field case @xmath3 effective total angular momentum equal to number vortices disk inhomogeneous magnetic field case assignment total vorticity tricky substituting eq . lin in free energy @xmath13 units @xmath14 @xmath15 local magnetic field @xmath16 vector potential obtain @xmath17 function of complex parameters @xmath18 minimization of @xmath17 allows find equilibrium vortex configurations determine stability restrict to states two components in eq . ( [ lin brings quantitative bounds gives correct qualitative behavior facilitates physical insight two approaches @xcite @xmath5 configurations with three and five components in eq checkedresulted in reduction metastable @xmath5 states at low magnetic field limit appearance additional unstable states free energy no new vortex configurations in ground state two components in describe vortex structure ground state radial state defined two configurations giant - giant @xmath19 giant - ring _ @xmath5 states first combination two @xmath22 states second one giant one ring vortex investigate centered step - like magnetic field magnetic disk effect size superconducting disk on vortex state shown in phase diagram parameters @xmath23 @xmath24 solid lines indicate ground state changes dashed lines transitions between @xmath2 states metastable right axis shows positive flux through region @xmath25 re - entrant _ behavior @xmath26 change total vorticity @xmath27 ground state with total vorticity @xmath28 and @xmath29 covers largest part phase diagram increasing disk size other @xmath2 states suppressed @xmath5 states small disk sizes different @xmath2 states present ground state increase disk size islands with different @xmath5 configurations giant - ring dominate ground state diagram@xmath30 giant - ring @xmath5 state @xmath31 giant - giant states fig . fig2 @xmath32 increasing magnetic field @xmath33 giant - giant combinations @xmath34 become ground state contour plot @xmath37-distribution single vortices arranged ring low density area fig fig3](a ) blue zero density red - highest density center disk fig fig3](a @xmath38 state central area associated giant vortex leads phase change vorticity @xmath39 fig3](b ) contour plot phase seven anti - vortices circle around giant vortex total vorticity @xmath28 disk size @xmath40 @xmath41 giant - giant _ @xmath5 states vortices radially symmetric no anti - vortices enlarging superconducting disk enhances influence negative part step - like magnetic field new giant - giant form giant multi anti - vortex _ states total vorticity equal to lowest vorticity giant vortex states leads re - entrant behavior total vorticity agreement ref. @xcite several vortices enter exit once disks large radius@xmath40 disks @xmath42 transitions possible increase disk radius @xmath35 giant - multi anti - vortex states become @xmath43 @xmath45 multi - vortex states metastable @xmath28 states ground states @xmath46 meissner state values applied magnetic field move step - like field profile radius superconducting disk ring magnetic field profile current loop fig phase diagram function @xmath47 parameters magnetic field profile @xmath48 @xmath49 @xmath32 solid curves transitions between vortex states dashed lines transitions metastable @xmath2 states re - entrant _ behavior visible @xmath5 states dominate diagram both multi - vortices giant - giant giant - ring _ ground state shifting field profile disk periphery giant - @xmath5 configurations cover superconducting region giant - giant @xmath5 states appear no anti - vortices giant multi anti - vortex states correspondence giant - multi anti - vortex states @xmath50 @xmath51 remains same @xmath52 increases to correlated with _ re - entrant _ behaviordiagram @xmath5 states with classical geometry _ giant - giant and giant - ring _ dominate difference between states exists in fig _ giant - ring _ state vortices distributed within ring - shaped lower density area high density areas disk periphery shift in phase of @xmath55 observed @xmath3 total vorticity states different origin similar distribution vortices phase work supported by flemish science foundation belgian inter - university attraction poles onderzoeksraad van universiteit antwerpen esf programme on vortex matter '' | 0.354867 | 2,334 | 1,177 | , Saving $0.1 in GPT-4. |
154,900 | using the recently proposed model of combinatorial landscapes : _ local optima networks _ , we study the distribution of local optima in two classes of instances of the _ quadratic assignment problem_. our results indicate that the two problem instance classes give rise to very different configuration spaces . for the so - called real - like class ,
the optima networks possess a clear modular structure , while the networks belonging to the class of random uniform instances are less well partitionable into clusters .
we briefly discuss the consequences of the findings for heuristically searching the corresponding problem spaces . | we have recently introduced a model of combinatorial landscapes : _ local optima networks _ ( lon ) @xcite , which allows the use of complex network analysis techniques @xcite for studying fitness landscapes and problem difficulty in combinatorial optimization .
the model , inspired by work in the physical sciences on energy surfaces@xcite , is based on the idea of compressing the information given by the whole problem configuration space into a smaller mathematical object which is the graph having as vertices the local optima and as edges the possible transitions between them .
this characterization of landscapes as networks has brought new insights into the global structure of the landscapes studied . moreover ,
some network features have been found to correlate and suggest explanations for search difficulty on the studied domains .
our initial work considered binary search spaces and the nk family of abstract landscapes @xcite .
recently , we have turned our attention to more realistic combinatorial spaces ( permutation spaces ) , specifically , the _ quadratic assignment problem _ ( qap ) @xcite . in this article , we focus on a particular characteristic of the optima networks using the qap , namely , the manner in which local optima are distributed in the configuration space .
several questions can be raised .
are they uniformly distributed , or do they cluster in some non - homogeneous way ?
if the latter , what is the relation between objective function values within and among different clusters and how easy is it to go from one cluster to another ?
knowing even approximate answers to some of these questions would be very useful to further characterize the difficulty of a class of problems and also , potentially , to devise new search heuristics or variation to known heuristics that take advantage of this information .
this short paper starts to address some of these questions .
the sections below summarize our methodology and preliminary results .
the qap is a combinatorial problem in which a set of facilities with given flows has to be assigned to a set of locations with given distances in such a way that the sum of the product of flows and distances is minimized .
a solution to the qap is generally written as a permutation @xmath0 of the set @xmath1 .
the cost associated with a permutation @xmath0 is : @xmath2 , where @xmath3 denotes the number of facilities / locations and @xmath4 and @xmath5 are referred to as the distance and flow matrices , respectively .
the structure of these two matrices characterizes the class of instances of the qap problem . for the statistical analysis conducted here ,
the two instance generators proposed in @xcite for the multi - objective qap were adapted for the single - objective qap .
the first generator produces uniformly random instances where all flows and distances are integers sampled from uniform distributions .
the second generator produces flow entries that are non - uniform random values .
the instances produced have the so called `` real - like '' structure since they resemble the structure of qap problems found in practical applications . for the purpose of community detection ,
@xmath6 instances were produced and analyzed with size @xmath7 for the random uniform class , and @xmath6 of size @xmath8 for the real - like instances class .
problem size @xmath8 is the largest one for which an exhaustive sample of the configuration space was computationally feasible in our implementation . in order to define the local optima network of the qap instances
, we need to provide the definitions for the nodes and edges of the network .
the vertexes of the graph can be straightforwardly defined as the local minima of the landscape , w.r.t . the neighborhood defined by a @xmath9-opt swap in the permutation space . in this work ,
we select small qap instances such that it is feasible to obtain all the nodes exhaustively by running a best - improvement hill - climbing algorithm from every configuration of the search space .
the edges account for the transition probability between basins of attraction of the local optima .
more formally , the edges reflect the total probability of going from basin @xmath10 to basin @xmath11 , which is the average over all @xmath12 of the transition probabilities to solutions @xmath13 .
the reader is referred to @xcite for a more detailed exposition .
we define a _ local optima network _
( lon ) as being the graph @xmath14 where the set of vertices @xmath15 contains all the local optima , and there is an edge @xmath16 with weight @xmath17 between two nodes @xmath18 and @xmath19 iff @xmath20 .
notice that since each maximum has its associated basin , @xmath21 also describes the interconnection of basins .
the study of lons for the qap instances @xcite , showed that the networks are dense .
indeed , they are complete or almost complete graphs , which is inconvenient for cluster detection algorithms . moreover , from the perspective of heuristic search algorithms , only the most likely transitions play a role . therefore , we opted for filtering out the networks edges keeping the more likely transitions .
in filtering , we first replace the directed graph by an undirected one ( @xmath22 ) , and then suppress all edges that have @xmath23 smaller than the value making the @xmath24-quantile ( @xmath25 in experiments ) in the weights distribution . such a less dense network provides a coarser but clearer view of the fitness landscape backbone , and can be used for minima cluster analysis .
clusters or communities in networks can be loosely defined as being groups of nodes that are strongly connected between them and poorly connected with the rest of the graph .
community detection is a difficult task , but today several good approximate algorithms are available @xcite . here
we use two of them : ( i ) a method based on greedy modularity optimization , and ( ii ) a spin glass ground state - based algorithm , in order to double check the community partition results .
figure [ fig : box ] shows the modularity score ( @xmath26 ) distribution calculated for each algorithm / instance - class . in general , the higher the value of @xmath26 of a partition , the crisper the community structure @xcite .
the plot indicates that the two instance classes are well separated in terms of @xmath26 , and that the community detection algorithm does not seem to have any influence on such a result . on the y - axis with respect to class problem ( rl stands for real - like and uni stands for random uniform ) and community detection algorithm ( 1 stands for fast greedy modularity optimization and 2 stands for spin glass search algorithm ) .
, scaledwidth=55.0% ] the modularity measurements ( fig .
[ fig : box ] ) indicate that real - like instances have significantly more minima cluster structure than the class of random uniform instances of the qap problem .
this can be appreciated visually by looking at fig .
[ fig : comm ] where the community structures of the lon of two particular instances are depicted .
although these are the two particular cases with the highest @xmath26 values of their respective classes , the trends observed are general . for the real - like instance ( fig .
[ fig : comm ] , left ) one can see that groups of minima are rather recognizable and form well separated clusters ( encircled with dotted lines ) , which is also reflected in the high corresponding modularity value @xmath27 .
contrastingly , the right plot represents a case drawn from the class of random uniform instances .
the network has communities , with a @xmath28 , although they are hard to represent graphically , and thus are not shown in the picture .
the clustering results discussed above may have deep consequences on the heuristic algorithms used to search the corresponding landscapes .
for example , on the random uniform instances a simple local heuristic search , such as hill - climbing , should be sufficient to quickly find satisfactory solutions since they are homogeneously distributed .
in contrast , in the real - like case they are much more clustered in regions of the search space .
this leads to more modular optima networks and using multiple parallel searches , or large neighborhood moves would probably be good strategies .
these ideas clearly deserve further investigation .
future work will confirm the statistical significance of our results , consider larger instances using appropriate sampling , and explore additional combinatorial problems .
knowles , j. , corne , d. : instance generators and test suites for the multiobjective quadratic assignment problem . in : evolutionary multi - criterion optimization , emo 2003 .
. 295310 .
2632 in lncs , springer ( 2003 ) | introduced model of combinatorial landscapes : _ local optima networks _ ( lon ) @xcite allows complex network analysis techniques for studying fitness landscapes problem difficulty in combinatorial optimization model inspired by work physical sciences energy surfaces based on compressing information problem configuration space into smaller mathematical object graph vertices local optima edges possible transitions characterization of landscapes as networks brought insights into global structure network features correlate suggest explanations for search difficulty domains initial work considered binary search spaces nk family of abstract landscapes turned attention to realistic combinatorial spaces spaces ) _ quadratic assignment problem _ ( qap ) @xcite focus on characteristic optima networks local optima distributed in configuration space questions raised uniformly distributed or non - ? relation between objective function values clusters easy to from one cluster to another ? approximate answers useful to characterize difficulty class problems devise new search heuristics short paper questions sections summarize methodology preliminary results qap is combinatorial problem facilities with given flows assigned to locations with given distances sum of product of flows distances minimizedsolution to qap as permutation @xmath0 of set @xmath1 cost permutation @xmath0 is @xmath2 @xmath3 number facilities locations @xmath4 @xmath5 distance flow matrices structure matrices characterizes class instances qap problem two instance generators multi - adapted for single - objective qap first generator produces uniformly random instances flows distances integers from uniform distributions second generator produces non - uniform random values instances real - like '' structure resemble qap problems @xmath6 instances produced analyzed with size @xmath7 for random uniform class @xmath8 for real - like instances problem size @xmath8 largest exhaustive sample of configuration space feasible define local optima network of qap definitions for nodes edges network vertexes graph defined as local minima of landscape neighborhood defined by @xmath9-opt swap permutation space select small qap instances obtain all nodes exhaustively by best - improvement hill - climbing algorithm from edges account for transition probability between basins of attraction local optimaedges reflect probability from basin @xmath10 to @xmath11 average over @xmath12 transition probabilities to solutions @xmath13 to @xcite for detailed exposition local optima network graph @xmath14 where vertices @xmath15 contains local optima edge @xmath16 with weight @xmath17 between nodes @xmath18 and @xmath19 @xmath20 each maximum has associated basin @xmath21 describes interconnection of basins study of lons instances showed networks are dense complete or almost complete graphs inconvenient for cluster detection algorithms search likely transitions play role for networks edges keeping likely transitions replace directed graph by undirected suppress edges @xmath23 smaller than value @xmath24-quantile ( @xmath25 in weights distribution less dense network provides clearer view of fitness landscape backbone used for minima cluster analysis clusters groups of nodes strongly connected poorly connected with rest graph community detection difficult good approximate algorithms available @xcite use two method greedy modularity optimization spin glass ground state - based algorithm to double check community partition resultsfigure fig box ] shows modularity score ( @xmath26 ) distribution for each algorithm / instance - class higher value of @xmath26 partition crisper community structure @xcite plot indicates two instance classes well separated @xmath26 community detection algorithm result y - axis class problem rl real - like uni random uniform community detection algorithm 1 fast greedy modularity optimization 2 spin glass search algorithm scaledwidth=55.0% modularity measurements fig indicate real - like instances have more minima cluster structure than random uniform instances problem visually fig . comm community structures of two instances depicted cases with highest @xmath26 values trends general for real - like instance left groups of minima recognizable form well separated clusters reflected in high modularity value @xmath27 right plot represents case from random uniform instances network has communities with @xmath28 hard to represent graphically not shown in picture clustering results may have consequences on heuristic algorithms search landscapes on random uniform instances simple local heuristic search hill should sufficient to find satisfactory solutions homogeneously distributedcontrast real - like case clustered search space leads modular optima networks multiple parallel searches large neighborhood moves good strategies ideas deserve investigation future work confirm statistical significance consider larger instances sampling explore combinatorial problems knowles corne d. instance generators test suites multiobjective quadratic assignment problem evolutionary multi - criterion optimization 2003 springer | 0.316489 | 1,805 | 885 | , Saving $0.1 in GPT-4. |
154,911 | we demonstrate that there are systematic scale errors in the [ fe / h ] values determined by the hamburg / eso survey ( and by inference by the hk survey in the past ) for certain extremely metal poor highly c - enhanced giants .
the consequences of these scale errors are that a ) the fraction of carbon stars at extremely low metallicities has been overestimated in several papers in the recent literature b ) the number of extremely metal poor stars known is somewhat lower than has been quoted in the recent literature c ) the yield for extremely metal poor stars by the he s survey is somewhat lower than is stated in the recent literature . a preliminary estimate for the frequency of carbon stars among the giants in the he s sample with @xmath0 [ fe / h ] @xmath1 dex is 7.4@xmath2% ; adding an estimate for the c - enhanced giants with [ c / fe ] @xmath3 dex without detectable c@xmath4 bands raises the fraction to 14@xmath5% .
we rely on the results of an extensive set of homogeneous detailed abundance analyses of stars expected to have [ fe / h ] @xmath6 dex selected from the he s to establish these claims .
we have found that the fe - metallicity of the cooler ( @xmath7@xmath8 k ) c - stars as derived from spectra taken with hires at keck are a factor of @xmath910 higher than those obtained via the algorithm used by the he s project to analyze the moderate resolution follow - up spectra , which is identical to that used until very recently by the hk survey .
this error in fe - abundance estimate for c - stars arises from a lowering of the emitted flux in the continuum bandpasses of the kp ( 3933 line of ca ii ) and particularly the hp2 ( h@xmath10 ) indices used to estimate [ fe / h ] due to absorption from strong molecular bands . | we are engaged in a large scale project to find additional extremely metal poor stars in the halo of our galaxy .
the major existing survey for very metal - poor stars is the hk survey described in detail by beers , preston & shectman ( 1985 , 1992 ) .
the stellar inventory of this survey has been scrutinized with considerable care over the past decade , but , as summarized by @xcite , only roughly 100 are believed to be extremely metal poor ( henceforth emp ) , with [ fe / h ] @xmath6 dex is given by @xmath11 on a scale where @xmath12 h atoms . then [ x / h ] = log@xmath13[n(x)/n(h ) ] @xmath14 log@xmath13[n(x)/n(h ) ] , and similarly for [ x / fe ] . ] .
we are therefore exploiting the database of the hamburg / eso survey ( he s ) for this purpose .
the he s is an objective prism survey from which it is possible to efficiently select emp stars @xcite .
the existence of a new list of candidates for emp stars with [ fe / h ] @xmath15 dex selected in an automated and unbiased manner from the he s , coupled with the very large collection area and efficient high resolution echelle spectrographs of the new generation of large telescopes , offers the possibility for a large increase in the number of emp stars known and in our understanding of their properties .
we have obtained and analyzed spectra with hires @xcite at the keck i telescope of a large number of emp candidates selected from the he s .
the normal procedures outlined by @xcite to isolate emp stars from the candidate lists produced by the he s were followed . in an effort to avoid selection biases ,
these differ from the criteria adopted by the hk survey ; see @xcite for details .
candidate emp stars selected from the he s were vetted via moderate resolution spectroscopy at large telescopes to eliminate the numerous higher abundance interlopers .
most of the follow up spectra for the stars discussed here were obtained with the double spectrograph @xcite on the hale telescope at palomar mountain , denoted p200 , ( a few are from the boller and chivens spectrograph on the clay and baade telescopes at the las campanas observatory ) during the period from 2001 to the present .
we intend to observe all candidates to the magnitude limit of the he s ( b @xmath9 17.5 ) in our fields ; observations are now complete in @xmath9990 deg@xmath16 , complete to b=16.5 in an additional @xmath9700 deg@xmath16 , and approaching completion in the remaining fields .
these follow up spectra are used to determine an accurate measure of the metallicity of the star , much more so than is possible to derive from the low resolution objective prism spectra of the he s itself .
this is accomplished via a combination of strength of absorption in h@xmath10 ( determining @xmath7 ) and in the ca ii line at 3933 ( the kp index , which determines [ fe / h ] , once @xmath7 and hence log(@xmath17 ) are specified ) .
a calibration between the strength of the indices and metallicity is required , and is generally derived from literature searches for high resolution abundance studies of relevant stars .
we denote the resulting metallicity value as [ fe / h](hes ) . the specific algorithm adopted by the he s is described in @xcite and is essentially identical to that used by the hk survey until recently ;
the latest updates to the algorithm as used by the hk survey are described in @xcite .
stars were chosen for observation at high resolution with hires primarily on the basis of low predicted metallicity ; every star with [ fe / h](hes ) @xmath18 dex north of @xmath19 was put on the hires observing list .
spectra have now been obtained for more than 55 emp candidates from the he s . in all cases
the stellar parameters have been determined by jgc from broad band ( v - i , v - j and v - k ) photometry and theoretical isochrones with no reference to the spectra themselves .
insofar as possible , the procedures , the codes , the model atmospheres ( we use those of * ? ? ?
* ) and the atomic data used to reduce the hires echelle spectra and to carry out the detailed abundance analysis are identical , and the analyses are thus as homogeneous as possible .
@xcite present full details of the analysis and results for a large sample of emp dwarfs from the he s , while @xcite will present abundance analyses for 13 of the 16 known carbon stars from this he s sample ; two of these have not yet been observed with hires .
fifteen c - normal giants have been analyzed to date , and these will appear in a future publication @xcite .
our operational definition of a carbon - star ( c - star ) is one whose spectrum shows bands of c@xmath4 .
the p200 dbsp spectra mostly extend to 5300 , hence the prominent c@xmath4 band at 5160 is included .
we are reasonably certain , as will be discussed in @xcite , through inspection of the regions of both c@xmath4 and ch in these followup spectra , that there are no additional giant c - stars in the palomar sample .
if no c@xmath4 bands are detected , but [ c / fe]@xmath20 dex , we denote a star to be c - enhanced . also we denote stars with @xmath7 @xmath21 k as `` dwarfs '' , while all cooler stars are called `` giants '' .
the difference between the [ fe / h ] derived from analysis of the hires spectrum versus that obtained by applying the algorithm of @xcite to the moderate resolution follow up spectra for the set of 497 candidate emp stars with moderate resolution follow up p200 spectra is shown in fig .
[ fig_delta_feh ] .
the giants with normal c and the warmer c - giants show good agreement ; [ fe / h](hes ) inferred from the moderate resolution spectra is a reliable indicator of the fe - metallicity found from analysis of hires spectra .
getting the dwarfs correct is harder as metal lines become weaker at their higher @xmath7 .
also we have adopted a @xmath7 scale for them which is hotter than that used in most earlier analyses which provided the calibration of the @xcite relation for emp dwarfs ( see , e.g. * ? ? ?
* ; * ? ? ?
in the mean , [ fe / h](hes ) systematically underestimates our hires - based fe - metallicity by 0.37 dex for emp dwarfs , corresponding to a systematic difference in adopted @xmath7 of @xmath9400 k. however , fig .
[ fig_delta_feh ] shows that for the cooler c - giants ( @xmath7 @xmath8 k ) , [ fe / h](hes ) substantially underestimates our hires - based fe - metallicity by @xmath91 dex .
@xcite suggested the presence of a systematic error of comparable size ( up to 1 dex ) for the sample of c - stars they analyzed from the hk survey .
this is a very large systematic error , much too large to be caused by problems in the @xmath7 scale , and so we attempt to understand what might be causing it . in fig .
[ fig_spec ] we show sections of the hires spectra of three emp candidates from the he s shifted into the rest frame .
these stars all have @xmath7@xmath22 k. two are c - stars , the third is a genuine emp giant with weak ch .
each of these stars has [ fe / h](hes ) @xmath23 dex .
the red and blue continuum and the feature bandpass are shown for the kp and the hp2 indices used to determine [ fe / h](hes ) , with a feature bandpass 12 wide for each ; see @xcite for details of the index definitions .
the figure clearly shows the source of the problem afflicting the c - stars the `` continuum '' bands are full of strong molecular absorption , particularly the red continuum band for the hp2 index .
if the hp2 index is underestimated because the continuum is depressed , then the star is assumed to be cooler than it actually is , and the resulting [ fe / h](hes ) for a fixed kp index will be too low .
furthermore , the blue continuum region of the kp index also shows strong ch absorption ( the big chunks missing from the spectra of the c - stars in the left column blueward of the 3933 caii line ) , and hence the abundance indicator kp will also be underestimated .
the derived [ fe / h](hes ) obtained using the calibration of @xcite will thus be substantially reduced below its true value for such c - stars . because the absorption in the relevant spectral regions arises from both cn and ch , the magnitude of this effect depends on additional factors such as the c - enhancement and the c / n ratio as well as on @xmath7 .
several tests have been performed to verify this .
first we checked that the measured kp and hp2 indices for c - stars ( and for c - normal stars ) can be reproduced to within their uncertainties from the much more precise hires spectra .
we also checked that adding back the missing flux removed from the continuum by absorption in the sidebands significantly increases kp and particularly the hp2 indices above the measured values , by factors of two or more .
finally we checked that by so altering the kp and hp2 indices we derive a significantly higher value for [ fe / h](hes ) which is much closer to the that obtained by the detailed abundance analyses for the cooler c - stars .
an underestimate of a factor of @xmath91 dex in the deduced value of [ fe / h](hes ) for the cooler c - stars will have significant effects .
[ fig_feh_vmk ] shows [ fe / h](hes ) versus v - k for a sample of 489 emp candidates from the he s with moderate resolution spectra from the double spectrograph at the hale telescope ( details will appear in * ? ? ?
the 10 known c - stars and the 1 c - enhanced star with hires analyses from this sample are indicated . in the upper panel , these stars are plotted at their [ fe / h](hes ) values , while in the lower panel they are plotted at their [ fe / h](hires ) as determined from detailed abundance analyses . ] .
although at their nominal fe - metallicities the c - stars dominate the population of the giants below [ fe / h](hes ) @xmath24 dex , using the results from analysis of high resolution spectra in the lower panel the frequency of c - stars among the most metal poor emp stars is considerably reduced .
it is our contention that , as shown in fig .
[ fig_feh_vmk ] , this underestimate of the metallicity of the cooler c - stars by the algorithm of @xcite , used by both the he s and in the past by the hk survey , produces a spurious high frequency of c - stars among emp stars ( see fig . [ fig_delta_feh ] ) . using the [ fe / h](hes ) values for
the c - stars would yield an apparent c - star fraction of 33% for [ fe / h ] @xmath25 dex , while using the hires fe - metallicities , a value a factor of 2.4 smaller is obtained .
there are 122 he s giants with [ fe / h](hes ) @xmath26 dex in our sample , suggesting a c - star frequency of 7.4@xmath2% for emp stars . adding in the fraction of of c - enhanced stars among giants with [ fe / h](hes ) @xmath26 dex found by @xcite of 6.5@xmath27% , one obtains a total fraction of c - rich stars with [ c / fe ] @xmath3 dex of 14@xmath5% among the giants in our he s emp sample with [ fe / h](hes ) @xmath26 dex , smaller than the value of 25% for stars with [ fe / h ] @xmath28 dex given by @xcite .
( we have derived this fraction for the c - enhanced stars among the dwarfs in our sample with [ fe / h](hes ) @xmath26 dex ; it has the same value , as will be reported in cohen et al 2006a . )
we are currently analyzing larger samples to refine this fraction .
it will probably be necessary to include an indicator of the strength of the molecular bands ( most easily the g band of ch , i.e. the gp index already introduced in * ? ? ?
* ) with the standard kp and hp2 indices in the calibration algorithm to obtain valid fe - abundances and maximum information from the he s and hk survey samples and/or to replace the hp2 index with a combination of v from the he s and j or k from 2mass ; the latter was not available at the time the hk survey defined their metallicity calibration algorithms , ( see , e.g. * ? ? ?
the metallicity distribution function is very sharply declining among halo stars at the lowest fe - metallicities .
thus the systematic errors we have found in the calibration of the he s and by inference the hk metallicity scale of @xcite , at least until quite recently ( see * ? ? ?
* ) , will also lead directly to systematic overestimates of the number of emp stars and of the yield for emp stars by these two major surveys .
we are currently evaluating in detail the impact of these calibration errors on such issues .
the entire keck / hires user community owes a huge debt to the many other people who have worked to make the keck telescope and hires a reality and to operate and maintain the keck observatory .
jgc and jm are grateful for partial support from nsf grant ast-0205951 .
jgc is grateful for support from the ernest fullam award of the dudley observatory which helped initiate this work .
the work of n.c . and fjz is supported by deutsche forschungsgemeinschaft ( grants ch 214 - 3 and re 353/44 ) .
n.c . acknowledges support through a henri chretien international research grant administered by the american astronomical society . | engaged in large project to find metal poor stars in halo galaxy major survey is hk survey by beers preston shectman 1985 1992 stellar inventory scrutinized summarized roughly 100 extremely metal poor emp ) with [ fe / h ] @xmath6 dex given by @xmath11 scale @xmath12 h atoms [ x / h ] = log@xmath13[n(x)/n(h ) ] @xmath14 log@xmath13[n(x)/n(h ) ] x / fe ] exploiting database hamburg / eso survey objective prism survey efficiently emp stars new list of candidates for emp stars with [ fe / h ] @xmath15 dex selected large collection area high resolution echelle spectrographs possibility for increase in number emp stars known understanding of properties obtained analyzed spectra with @xcite at keck i telescope of large number of emp candidates selected procedures to isolate emp stars followed selection biases differ from criteria hk survey see @xcite details emp stars selected vetted via moderate resolution spectroscopy at large telescopes to eliminate higher abundance interlopersfollow up spectra for stars obtained with double spectrograph @xcite hale telescope at palomar mountain p200 few from boller and chivens spectrograph clay baade telescopes las campanas observatory 2001 to intend to observe all candidates to magnitude limit he s ( b @xmath9 17.5 ) observations complete in @xmath9990 deg@xmath16 b=16.5 @xmath9700 deg@xmath16 approaching completion remaining fields spectra determine metallicity star low resolution objective prism spectra he s accomplished via strength of absorption in h@xmath10 ca ii line at 3933 kp index calibration between strength indices metallicity required derived from literature high resolution abundance studies resulting metallicity value as [ fe / h](hes ) algorithm by he s described in @xcite identical to hk survey updates stars chosen for observation high resolution low predicted metallicity every star with [ fe / h](hes ) @xmath18 dex north of @xmath19 on hires observing list spectra obtained for 55 emp candidates from he s stellar parameters determined by jgc broad band v j k photometry theoretical isochrones no reference to spectra procedures codes model atmospheres atomic data to hires echelle spectra detailed abundance analysis identical analyses homogeneous @xcite details analysis results for large sample emp dwarfs from s @xcite abundance analyses for 13 of 16 carbon stars two not observed with hires fifteen c - normal giants analyzed in future publication definition of carbon - star c - star is spectrum shows bands of c@xmath4 p200 dbsp spectra extend to 5300 prominent c@xmath4 band at 5160 included no additional giant c - stars in palomar sample if no c@xmath4 bands detected but [ c / fe]@xmath20 dex star c - enhanced stars with @xmath7 @xmath21 k as `` dwarfs '' cooler stars `` giants '' difference between fe / h ] analysis hires spectrum algorithm @xcite to moderate resolution follow up spectra for 497 candidate emp stars with p200 spectra shown in fig .giants normal c warmer c show agreement [ fe / h](hes ) moderate resolution spectra reliable indicator fe - metallicity dwarfs harder metal lines weaker at higher @xmath7 adopted @xmath7 scale hotter calibration @xcite relation for emp dwarfs [ fe / h](hes ) underestimates hires - fe - metallicity 0.37 dex for emp dwarfs difference @xmath7 @xmath9400 k fig . shows cooler c - giants @xmath7 @xmath8 k [ fe / h](hes ) underestimates hires - based fe - metallicity by @xmath91 dex @xcite systematic error up to 1 dex ) for sample c - stars survey large systematic error problems @xmath7 scale attempt understand fig . hires spectra three emp candidates stars have @xmath7@xmath22 k two c - stars third genuine emp giant weak ch each has [ fe / h](hes ) @xmath23 dex red blue continuum feature bandpass shown for kp hp2 indices fe / h](hes ) feature bandpass 12 wide each see @xcite index definitionsfigure shows problem c - stars continuum bands strong molecular absorption particularly red continuum band hp2 index if hp2 index underestimated continuum depressed star assumed cooler [ fe / h](hes ) for fixed kp index too low blue continuum region kp index shows strong ch absorption missing from c - stars left column line abundance indicator kp underestimated derived [ fe / h](hes ) calibration @xcite reduced below true value for c - stars absorption arises from cn and ch magnitude effect depends on factors c - enhancement c / n ratio @xmath7 tests performed measured kp hp2 indices for c - stars reproduced from precise hires spectra adding back missing flux continuum increases kp hp2 indices above measured values by factors two or more altering kp and hp2 indices higher value for [ fe / h](hes ) closer to abundance analyses for cooler c - stars underestimate of factor @xmath91 dex in deduced value [ fe / h](hes ) for cooler c - stars significant effectsfig_feh_vmk shows fe / h](hes ) versus v - k 489 emp candidates moderate resolution spectra double spectrograph hale telescope 10 c - stars 1 c - enhanced star with hires upper panel stars plotted at fe / h](hes ) values lower panel fe / h](hires ) abundance analyses c - stars dominate giants below fe / h](hes ) @xmath24 dex high resolution spectra frequency c - stars among metal poor emp stars reduced underestimate of metallicity cooler c - stars by algorithm @xcite produces high frequency of c - stars among emp stars fig_delta_feh fe / h](hes ) values c - stars apparent c - star fraction 33% for fe / h ] @xmath25 dex hires fe - metallicities factor 2.4 smaller 122 he s giants with fe / h](hes ) @xmath26 dex c - star frequency 7.4@xmath2% for emp starsadding fraction c - enhanced stars among giants with @xmath26 dex @xcite of 6.5@xmath27% obtains total fraction c - rich stars with @xmath3 dex of 14@xmath5% among giants in emp sample with dex smaller than 25% for stars with @xmath28 dex @xcite derived fraction for c - enhanced stars among dwarfs in sample dex same value reported in cohen et al 2006a analyzing larger samples to refine fraction necessary to include indicator of strength molecular bands g band of ch gp index with standard kp and hp2 indices in calibration algorithm obtain valid fe - abundances maximum information from hk survey samples replace hp2 index with v from he s and j or k from 2mass not available hk survey metallicity calibration metallicity distribution function declining among halo stars at lowest fe - metallicities systematic errors in calibration of he s hk metallicity scale of @xcite lead to overestimates of number of emp stars yield for emp stars by surveys evaluating impact calibration errorskeck / hires user community debt keck telescope hires observatory jgc jm grateful support nsf grant ast-0205951 jgc support ernest fullam award dudley observatory n.c . fjz supported by deutsche forschungsgemeinschaft grants ch 214 - 3 re 353/44 n.c . acknowledges support henri chretien research grant american astronomical society | 0.505249 | 3,076 | 1,557 | , Saving $0.1 in GPT-4. |
154,915 | we shall use symmetry breaking as a tool to attack the problem of identifying the topology of chaotic scatteruing with more then two degrees of freedom .
specifically we discuss the structure of the homoclinic / heteroclinic tangle and the connection between the chaotic invariant set , the scattering functions and the singularities in the cross section for a class of scattering systems with one open and two closed degrees of freedom .
* keywords : * chaotic scattering , cross sections , classical mechanics , rainbow singularities pacs : 45.20.df 05.45.-a 45.50.-j 05.45.-a 05.45.ac 05.45.pq | in a hamiltonian system with two degrees of freedom the poincare map acts on a two dimensional domain .
the chaotic invariant set is represented by the well known horseshoe construction of smale @xcite . in the meantime
it has been worked out quite well , how in the case of open systems ( scattering systems ) the properties of this chaotic set are seen in scattering functions @xcite and in scattering cross sections @xcite .
the corresponding investigations become a lot more complicated when we proceed to systems with more degrees of freedom . therefore relatively little
has been done for more degrees of freedom . in the present article we present some results for one class of higher dimensional scattering systems , namely the one with one open and two closed degrees of freedom .
our strategy to attack this higher dimensional case is based on an idea introduced to treat elliptical orbits in reduced three - body systems @xcite , which is actually much in the spirit of moshinskys work in nuclear physics .
we start from a system with a symmetry and a conserved quantity ( in the former case the jacobi integral ) .
then the three degrees of freedom system can be reduced to a continuum of systems with two degrees of freedom where the numerical value of the conserved quantity serves as parameter .
we use the knowledge of the techniques for the two degree of freedom case , to treat the reduced system .
then we embed the continuum of the reduced systems into the full six - dimensional phase space to assemble the corresponding higher dimensional structures still for the symmetric case . in other words
, we find the solution for the symmetric case .
finally we break the symmetry to get to the true three degrees of freedom case . here
we show that the breaking of the symmetry does not have important effects on the qualitative properties of the structure , at least as long as the symmetry breaking is weak .
the argument is supported by numerical calculations .
for hamiltonian systems with three degrees of freedom the poincar map acts on a four dimensional domain .
the first thing to do is to describe in the domain of the four dimensional poincar map the higher dimensional analog of the two dimensional horseshoe we obtain , when a symmetry reduces the effective dimensionality of the system . as representative example of the class of system we want to treat we imagine the motion of a particle in a channel with an obstacle .
the empty channel is cylindrical and thereby has a rotational symmetry in the azimuth angle @xmath0 .
the obstacle is represented by a short range potential depending on a parameter a which describes the deviation of the obstacle from rotational symmetry .
for a=0 the obstacle is symmetric in azimuth direction and the conjugate angular momentum @xmath1 is conserved and the rotational symmetry allows the reduction to two degrees of freedom .
a natural choice for the intersection condition for the poincare map is the choice of relative maxima of the cylindrical radius @xmath2 .
every times when the particle runs through a relative maximum of @xmath2 we measure the values of @xmath3 , @xmath4 , @xmath0 and @xmath1 where @xmath3 and @xmath4 are position and conjugate momentum along the axis of the channel .
the space of these four coordinates is the domain of the poincar map .
one possibility to proceed is to set up particular models for the channel and the potential of the obstacle and to study solutions of hamilton s equations of motion and to obtain the poincare map from them .
we do something simpler and more direct .
we set up a model function for the poincare map which serves as prototype example ( paradigm ) for the whole class of systems .
we compose it from three steps : first , half of a free flight second , a kick generated by the generating function @xmath5 here @xmath6 is the maximally possible value of @xmath1 , indirectly @xmath6 fixes the value of the total energy @xmath7 .
third , again half a free flight for the potential function @xmath8 appearing in eq .
[ genfun ] we take @xmath9 for @xmath10 the map becomes independent of @xmath0 and reduces to a map acting on the @xmath3 , @xmath4 plane .
this reduced map depends on the conserved value of @xmath1 as parameter . because the potential of eq .
[ potv ] is purely attractive the points @xmath11 act as outer fixed points of the map and we use their invariant manifolds to construct the horseshoe of the reduced system . [
cols="^ " , ] in figs .
[ fundisp1 ] , [ fundisp2 ] and [ fundisp3 ] we show an example of this function for three values of a , namely @xmath12 in parts a and b , @xmath13 in parts c and d and @xmath14 in parts e and f. the scattering flow casts a kind of shadow image of the homoclinic tangle into the outgoing asymptotic region and in the scattering functions we see exactly this shadow image .
singularities of the scattering function , i.e. boundaries of intervals of continuity of this function indicate initial conditions on the stable manifold of the nhim at infinity .
thereby an analysis of the singularities of scattering functions allows the reconstruction of all the important information concerning the chaotic set .
observe that for @xmath10 the scattering function has a factorization into a fractal structure in @xmath15 direction and constancy in @xmath16 direction .
accordingly the function shows intervals of continuity which are the product of an interval in @xmath15 direction with a circle in @xmath16 direction . for @xmath17 small only very tiny intervals of continuity of high level of hierarchy
change their qualitative structure .
the larger intervals keep their basic topology of a strip running around in @xmath16 direction .
we have to go to the very strong symmetry breaking of @xmath18 to see the largest intervals change their qualitative structure .
in this section we explain how the fractal chaotic set in phase space or in the poincar map causes a fractal pattern of rainbow singularities in the cross section . for the case of three degrees of freedom
we generalize the ideas developed in @xcite for the case of two degrees of freedom .
usually in scattering experiments we do not have any control over the relative phase shift variables @xmath15 and @xmath16 .
we keep @xmath7 , @xmath19 and @xmath20 fixed and the angles @xmath15 and @xmath16 have a random distribution with constant density on a two dimensional torus .
we measure the relative probability to find values @xmath21 and @xmath22 .
this quantity properly normalized with respect to the incoming flow is the doubly differential cross section . to calculate this cross section for a particular pair of values of @xmath21 and @xmath22 we have first to find all the preimages @xmath23 of the scattering function which lead to these particular values of the final variables .
then we have to calculate the determinant @xmath24\ ] ] in the particular initial preimage point k. the cross section is given by @xmath25 we see immediately that the cross sections has singularities for such values of @xmath21 and @xmath22 where the determinant of eq .
[ detk ] is zero . these are values where several preimages collide and disappear , they are projection singularities of the scattering function .
these singularities are of one over square root type , therefore the integral over the cross section is finite and there are no problems with flux conservation . as long as the system is close to the symmetric case the scattering function in each interval of continuity is similar to a ring shaped mountain and accordingly the projection singularity of this function looks like the one of half a torus ( for details see @xcite ) . fig .
[ unasing ] shows the cross section contribution from a single interval of continuity .
fig.[varsing ] shows the singularities from the whole toroidal domain of the scattering function .
the pattern seen in fig.[varsing ] coincides with what a detector in a real experiment might see .
it is a fractal repetition of the basic singularity structure from fig.[unasing ] . .
note that this figure coincides with the projection of half a torus or of a ring shaped mountain .
the caustics of this projection lead to one over square root singularities ( rainbow singularities ) in the cross section.,scaledwidth=85.0% ] .
in contrast to fig.7 here the contributions from the complete domain of the scattering functions are included .
this pattern of singularity lines is a fractal repetition of the basic pattern created by a single interval of continuity as shown in fig.7.,scaledwidth=85.0% ] we can change a parameter of the system , this might be the symmetry breaking parameter , or a action like initial condition , this might be the incoming angular momentum @xmath26 .
when we measure the cross section singularities as function of this parameter then we can follow in the cross section the changes of the chaotic set under these changes of the system .
this is an interesting contribution to the inverse chaotic scattering problem .
the stable manifold of a nhim at infinity consists of trajectories which go out to infinity while at the same time the longitudinal momentum goes to zero , they are trajectories which arrive at infinity with velocity zero .
these manifolds are separatrix surfaces between trajectories going out to infinity and trajectories which return to the interaction region at least once more before eventually also going out to infinity .
the trajectories on the unstable manifolds show the same behaviour under time reversal .
the homoclinic / heteroclinic tangle between the stable and unstable manifolds of the nhim at @xmath27 are the higher dimensional generalizations of the horseshoes in two dimensional maps . in our class of systems
this chaotic structure exists for all values of the total energy @xmath7 .
the reason is that in our class of systems for any value of @xmath7 there are trajectories going out to infinity with longitudinal velocity zero .
this happens thanks to the closed degrees of freedom which can swallow up any amount of energy such that only energy zero is left for the open degree of freedom . in this sense
our system is at a channel threshold of the inelastic scattering for any value of @xmath7 . because of these considerations it is clear that we do not expect a similar behaviour for scattering systems with open degrees of freedom only . for a discussion of this point for scattering off four hard spheres
see @xcite .
the poincare map of this system has hyperbolic fixed points but does not have any nhim . for systems with more than three degrees of freedom similar considerations hold . for each additional degree of freedom the nhim and
its manifolds acquire two additional neutral directions .
the domains of the scattering functions and of the cross section acquire one additional dimension .
this work has been supported by conacyt under grant number 79988 and by dgapa under grant number in-110110 | hamiltonian system with two degrees freedom poincare map acts on two dimensional domain chaotic invariant set represented by horseshoe construction of smale @xcite open systems properties chaotic set seen in scattering functions cross sections investigations complicated systems more degrees freedom little done for more degrees freedom article results for higher dimensional scattering systems with one open two closed degrees freedom strategy based on idea treat elliptical orbits in reduced three - body systems @xcite spirit moshinskys work nuclear physics start from system with symmetry conserved quantity jacobi integral three degrees freedom system reduced to continuum systems with two degrees freedom numerical value conserved quantity parameter use techniques for two degree freedom case to treat reduced system embed continuum reduced systems into six - dimensional phase space to assemble higher dimensional structures for symmetric case find solution for symmetric case break symmetry true three degrees freedom case breaking symmetry effects qualitative properties structure symmetry breaking weak argument supported by numerical calculations hamiltonian systems with three degrees freedom poincar map acts on four dimensional domaindescribe four dimensional poincar map higher analog two dimensional horseshoe symmetry reduces effective dimensionality system imagine motion particle in channel with obstacle empty channel cylindrical rotational symmetry azimuth angle @xmath0 obstacle represented by short range potential parameter a deviation from rotational symmetry for a=0 obstacle symmetric azimuth direction conjugate angular momentum @xmath1 conserved rotational symmetry allows reduction to two degrees freedom natural choice intersection condition poincare map relative maxima cylindrical radius @xmath2 particle runs through relative maximum @xmath2 measure values of @xmath3 @xmath0 @xmath1 position conjugate momentum along axis channel space of four coordinates domain poincar map set up models for channel potential obstacle study solutions hamilton s equations of motion obtain poincare map set up model function for poincare map prototype example for class systems compose from three steps half free flight kick generated by function @xmath5 @xmath6 maximally possible value of @xmath1 fixes value total energy @xmath7 half free flight for potential function @xmath8 in eq .take @xmath9 @xmath10 map independent of @xmath0 reduces to map @xmath3 @xmath4 plane reduced map depends on conserved value @xmath1 potential of eq potv attractive points @xmath11 act outer fixed points use invariant manifolds construct horseshoe reduced system figs fundisp1 example function for three values a @xmath12 parts a b @xmath13 c d @xmath14 e f scattering flow casts shadow image homoclinic tangle into asymptotic region functions singularities scattering function boundaries intervals continuity indicate initial conditions on stable manifold nhim at infinity analysis singularities allows reconstruction information chaotic set @xmath10 scattering function factorization into fractal structure in @xmath15 constancy in @xmath16 direction function shows intervals continuity product of interval @xmath15 circle in @xmath16 direction @xmath17 small intervals continuity change qualitative structure larger intervals keep basic topology strip in @xmath16 direction strong symmetry breaking @xmath18 largest intervals change qualitative structuresection explain fractal chaotic causes pattern rainbow singularities cross section three degrees freedom generalize ideas @xcite two degrees freedom scattering experiments control over phase shift variables @xmath15 @xmath16 keep @xmath7 @xmath20 fixed angles random distribution constant density on two dimensional torus measure probability find values @xmath21 @xmath22 quantity normalized doubly differential cross section calculate cross section find preimages @xmath23 scattering function lead to values final variables calculate determinant @xmath24\ in initial preimage point k cross section given by @xmath25 cross singularities for values @xmath21 @xmath22 where determinant of eq . [ detk ] zero values preimages collide disappear projection singularities of scattering function singularities one over square root type integral cross section finite no problems with flux conservation system close to symmetric case scattering function each interval continuity similar to ring shaped mountain projection singularity looks like half torus details see @xcite fig . shows cross section contribution from single interval continuity fig. shows singularities from whole toroidal domain scattering function pattern fig.[varsing coincides with detector experiment fractal repetition of singularity structure from fig. figure coincides with projection of half torus ring shaped mountain caustics lead to square root singularities rainbow singularities in cross section=85.0% contrast to fig.7 contributions from scattering functions included pattern of singularity lines fractal repetition of basic pattern single interval continuity fig.7. change parameter system symmetry breaking parameter incoming angular momentum @xmath26 measure cross section singularities function parameter follow changes chaotic set system contribution to inverse chaotic scattering problem stable manifold of nhim at infinity trajectories out to infinity longitudinal momentum zero arrive at infinity with velocity zero manifolds separatrix surfaces between infinity return interaction region before infinity trajectories unstable manifolds same behaviour under time reversal homoclinic / heteroclinic tangle between stable unstable manifolds at @xmath27 higher dimensional generalizations of horseshoes in two dimensional maps class systems chaotic structure exists for all values total energy @xmath7 any value trajectories out to infinity with longitudinal velocity zerohappens thanks closed degrees freedom swallow energy zero for open freedom system at channel threshold inelastic scattering for value @xmath7 expect similar behaviour scattering systems with open degrees freedom scattering off four spheres see @xcite poincare map has hyperbolic fixed points nhim systems more than three degrees freedom similar considerations hold each additional degree freedom nhim manifolds acquire two neutral directions domains scattering functions cross section acquire one additional dimension supported by conacyt grant 79988 dgapa grant-110110 | 0.356601 | 2,329 | 1,151 | , Saving $0.1 in GPT-4. |
154,917 | we consider a gaussian process formulation of the multiple kernel learning problem . the goal is to select the convex combination of kernel matrices that best explains the data and by doing so improve the generalisation on unseen data .
sparsity in the kernel weights is obtained by adopting a hierarchical bayesian approach : gaussian process priors are imposed over the latent functions and generalised inverse gaussians on their associated weights .
this construction is equivalent to imposing a product of heavy - tailed process priors over function space .
a variational inference algorithm is derived for regression and binary classification . | kernel - based methods are well - established tools for supervised learning , allowing to perform various tasks , such as regression or binary classification , with linear and non - linear predictors . like most statistical models ,
kernel - based methods can be considered within two frameworks : in the frequentist approach , estimators are obtained by minimizing a regularized empirical risk , leading e.g. to kernel ridge regression or the support vector machine @xcite ; in the bayesian approach , gaussian processes ( gps ) provide a bayesian interpretation to kernel - based methods @xcite , with the potential to learn the kernel parameters from the data without having to use cross - validation .
crucial to the predictive performance of kernel methods is the choice of the kernel function . in the bayesian
setting , the kernel function ( often called covariance function ) determines the correlations between the predictions we make . assuming that the predictor s smoothness is fully specified by these correlations can be formalised by a gaussian process imposed over function space .
techniques based on automatic relevance determination have been successful at learning the parameters of kernel functions such as the individual length scales of the squarred exponential kernel @xcite . in the frequentist
setting , a specific parameterization of kernel functions has led to a significant amount of work , namely positive linear combination of pre - defined kernel functions ( or kernel matrices ) , leading to the multiple kernel learning ( mkl ) framework @xcite .
the first contribution of this paper is to propose a gaussian process ( gp ) formulation of the multiple kernel learning framework , which we refer to as _ multiple gaussian process _ ( mgp ) models .
its second contribution is to provide a framework to consider all @xmath0-norms at once and to determine _ from data _ whether we should use a sparsity - inducing prior or not .
currently , there is no consensus in the frequentist community on how to choose the type of regularization . in practice
, however , the choice of the regulariser leads to solutions of very different kinds .
for example , when considering an @xmath1-norm a sparse solution will be obtained whether or not it is supported by the data .
obviously , if all kernels are important for prediction , this will be detrimental @xcite .
let @xmath2 be the set of noisy targets and @xmath3 the set of features , which are assumed to be non - random column vectors .
we consider a weighted linear model of the @xmath4 feature vectors with i.i.d .
gaussian noise : @xmath5 where @xmath6 and @xmath7 is the identity matrix of dimension @xmath8 .
the weights associated to the feature matrix @xmath9 are denoted by @xmath10 and the residual precision by @xmath11 .
the case of interest is the one where the weight vectors are sparse , i.e. , many of their elements are ( nearly ) zero .
however , we do not know a priori the degree of sparsity . from a bayesian perspective
, the spike and slab prior is the golden standard for inducing sparsity . here ,
follow a different approach and choose the prior @xmath12 to be a gaussian scale mixture @xcite centred at zero . in effect
we approximate the spike and slab prior by a continuous prior which favours sparse solutions .
although zero probability mass is put on exact zero values , the use of heavy - tailed priors allows us to infer large , as well as quasi zero values for the kernel weights .
formally , we impose the following product of zero - mean gaussian scale mixtures on the weights : @xmath13 where @xmath14 is the vector of unobserved scale variables on which independent generalised inverse gaussian densities ( see appendix [ sect : inverse gaussian ] ) are imposed .
the marginal @xmath15 is then a symmetric generalised hyperbolic density @xcite , which has fat tails compared to the gaussian .
this family contains the student-@xmath16 , the laplace , the gamma - variance and jeffrey s as special cases .
given this probabilistic model , one can integrate out @xmath17 , leading to a closed form expression for the marginal density of the observations : @xmath18 where @xmath19 is the kernel matrix associated with the @xmath20-th feature matrix . clearly , the marginal density is well - defined for any set of valid kernel matrices since @xmath21 for all @xmath20 .
the linear additive model defined in ( [ eq : like lin ] ) corresponds to the _ weight - space _ view representation of the multiple gaussian process model ( mgp ) .
from ( [ eq : marginal likelihood ] ) , however , we see that the marginal density only depends on the linear kernel matrices @xmath22 .
hence , the model can be generalised to a non - linear additive model by replacing these linear kernel matrices by non - linear ones .
this new representation corresponds to the multiple _ function - space _ view representation .
let @xmath23 be a set of @xmath4 latent functions on which we impose scaled gaussian process priors : @xmath24 for all @xmath20 .
the functions @xmath25 are covariance functions , which are also valid kernel functions @xcite .
again we consider i.i.d .
gaussian noise , but assume @xmath2 are noisy observations of a sum of @xmath4 latent _ function _ values @xmath26 .
the likelihood function and the mgp prior are given by @xmath27 where @xmath28 , @xmath29 and @xmath30 .
vector @xmath31 is the unit vector of dimension @xmath4 and the operator @xmath32 denotes the kronecker product .
the prior on @xmath33 is still given by ( [ eq : gam prior ] ) and the marginal @xmath34 has the same form as in the weight - space view representation .
the mgp model corresponds to imposing @xmath4 independent _ non - gaussian _ process priors over function space .
if we condition on the corresponding scale variable , any finite subset of latent function values is distributed according to a multivariate gaussian marginal . for any of these marginals
one can integrate out the scale variable , such that any finite set of latent function values is distributed according to a product of @xmath4 independent multivariate gaussian scale mixture densities : @xmath35 where @xmath36 is the modified bessel function of the second kind .
hence , the prior measure imposed over function space is a heavy - tailed one , known as the generalised hyperbolic measure @xcite .
the gaussian process is recovered for @xmath37 and the symmetric multivariate zero - mean hyperbolic process is obtained for @xmath38 .
other special cases include the multivariate gamma - variance process ( @xmath39 and @xmath40 ) , the multivariate laplace process ( @xmath38 and @xmath40 ) , the multivariate student-@xmath16 process ( @xmath41 and @xmath42 ) and the multivariate cauchy process ( @xmath43 and @xmath42 ) .
the most straightforward approach for the estimation of @xmath33 is to use type ii maximum a postriori ( or type ii maximum likelihood in absence of prior on @xmath33 as adopted in @xcite ) .
the optimisation can be performed using standard nonlinear optimisation tools , but the regulariser needs to be chosen in advance .
instead , we turn our attention to the inference problem of these parameters from data .
we view @xmath33 as a latent variable and the desired level of sparsity is learnt from the data by optimising the hyperparameters by type ii maximum likelihood ( ml ) .
we follow a mean field approach @xcite . in order to find an analytically tractable solution ,
the posterior over the latent function values @xmath44 and the scale vector @xmath33 is assumed to factorise given the data , that is @xmath45 .
it can be shown that the variational posteriors maximising the negative variational free energy ( a lower bound to the log - marginal likelihood ) are given by @xmath46 and @xmath47 .
the parameters are defined as @xmath48 where @xmath49 .
the predictive mgp is obtained by assuming that @xmath50 is peaked around its mean such that @xmath51 .
the true predictive density can then be approximated by the following analytically tractable integral : @xmath52 where @xmath53 with @xmath54 . from these expression
we see that the posterior mean and variance have the same form as in standard gp regression ; the kernel is simply replaced by a convex combination of kernels .
note , moreover , that the expression @xmath55 has the same form as the one we would obtain with a frequentist method such as kernel ridge regression .
the ml ii updates for the hyperparameters are obtained by solving the following expressions ( which are simple root finding equations , with unique solutions , hence easily solved by binary search ) : @xmath56 where @xmath57 .
these updates are obtained by direct maximising of the variational bound .
the update for @xmath11 is obtained in the same manner .
we restrict ourselves to binary classification and consider a scaled probit model in which the likelihood is derived from the gaussian cumulative density .
a probit model is equivalent to a gaussian noise and a step function likelihood @xcite .
let @xmath58 be the class labels , with @xmath59 for all @xmath60 .
the likelihood ( [ eq : likelihood ] ) is replaced by @xmath61 where @xmath62 for @xmath63 and @xmath64 otherwise . as in the case of regression
, we consider a mean field approximation and assumes the posterior is of the form @xmath65 .
we further assume the variational posterior @xmath66 is a product of truncated gaussians ( see appendix [ sect : truncated gaussian ] ) : @xmath67 where @xmath68 and @xmath69 . the posterior mean and the posterior covariance of @xmath44 are unchanged , except that @xmath70 is replaced by @xmath71 .
the elements of @xmath71 are defined in ( [ eq : trun mean ] ) .
the posterior @xmath50 and the updates for the hyperparameters are identical to the ones in mgp regression . in bayesian classification the label with highest probability @xmath72 is selected . since an exact computation is analytically intractable , we assume the posteriors @xmath66 and @xmath50 are highly peaked around their mean leading to the following classification rule : @xmath73 where @xmath55 and @xmath74 are as before with @xmath70 replaced by @xmath71 .
deciding whether the label is @xmath75 or @xmath76 is equivalent to using the sign of @xmath55 as the decision rule .
we compare frequentist and bayesian approaches to kernel combination .
we demonstrate the flexibility and the performance of the mgp models on the following two detat sets : * toy regression data . *
we generate random functions from the hilbert spaces induced by 10 laplacian kernels and add gaussian i.i.d .
noise ; we show results on three different settings : a sparse problem , where only one kernel is used to generate the response , a semi - sparse problem with 3 functions are used and a non - sparse problem where all ten functions are active .
table [ table : regression ] compares several mgp models and several cross - validated mkl models with fixed regularisation norms .
[ fig : regression student ] in the appendix shows that the hierarchical bayesian approach is able to adapt to the sparsity of the data .
lcccr number of active kernels & 1 out of 10 & 3 out of 10 & 10 out of 10 + + + multiple student-@xmath16 & .033 ( @xmath77 ) & .067 ( @xmath78 ) & .719 ( @xmath79 ) + multiple laplace & .034 ( @xmath80 ) & .076 ( @xmath81 ) & .704 ( @xmath82 ) + multple gamma - variance & .033 ( @xmath77 ) & .067 ( @xmath78 ) & .719 ( @xmath83 ) + ard & .033 ( @xmath77 ) & .066 ( @xmath84 ) & .746 ( @xmath83 ) + mkl @xmath1 & .037 ( @xmath78 ) & .066 ( @xmath85 ) & .720 ( @xmath86 ) + mkl @xmath87 & .830 ( @xmath88 ) & .831 ( @xmath89 ) & .762 ( @xmath90 ) + mkl @xmath91 & .097 ( @xmath92 ) & .233 ( @xmath93 ) & .719 ( @xmath94 ) + * * flowers data set . * * due to a lack a space we do not describe the data and the features , but only mention it is a standard mkl benchmark for multi - class image classification . for each of the @xmath95 flower classes we learn a one - versus - all classifier .
[ fig : roc student ] shows the roc curve for two classes when considering a student-@xmath16 process , for which we obtained an average @xmath96 .
the average @xmath97 for gamma - variance process and ard are respectively given by @xmath98 and @xmath99 .
all are better than state - of - the - art mkl results @xcite .
the generalised inverse gaussian density is defined as follows @xcite : @xmath100 where @xmath101 and @xmath102 is the modified bessel function of the second kind with index @xmath103 . depending on the value taken by @xmath104
, we have the following constraints on @xmath105 and @xmath106 : @xmath107 let us define @xmath108 .
the following expectations are useful : @xmath109 when @xmath42 and @xmath41 , the generalised inverse gaussian density reduces to the gamma density .
when @xmath40 and @xmath39 , it reduces to the inverse gamma density .
the expectations simplify also .
the ( positive / negative ) truncated gaussian density is defined as follows : @xmath110 where @xmath111 is the cumulative density of the unit gaussian .
let @xmath112 . the mean and variance are given by @xmath113 | kernel - based methods are tools for supervised learning tasks regression binary classification with linear and non - linear predictors like statistical models methods considered within two frameworks frequentist approach estimators obtained by minimizing regularized empirical risk to kernel ridge regression support vector machine bayesian approach gaussian processes provide bayesian interpretation to kernel - methods potential to learn kernel parameters from data without cross - validation crucial to predictive performance is choice of kernel function bayesian kernel function covariance function determines correlations between predictions predictor s smoothness specified correlations formalised by gaussian process function techniques automatic relevance determination successful at learning parameters of kernel functions in frequentist setting specific parameterization of kernel functions led to work positive linear combination of pre - defined kernel functions to multiple kernel learning ( framework first contribution paper gaussian process formulation of multiple kernel learning framework multiple gaussian process _ ) models second contribution framework to consider all @xmath0-norms determine sparsity - inducing prior no consensus in frequentist community on type regularization choice leads to solutions differentconsidering @xmath1-norm sparse solution obtained data all kernels important for prediction detrimental @xcite @xmath2 noisy targets @xmath3 features non - random column vectors consider weighted linear model @xmath4 feature vectors with gaussian noise : @xmath5 @xmath6 @xmath7 identity matrix dimension @xmath8 weights feature matrix @xmath9 denoted by @xmath10 residual precision @xmath11 case interest weight vectors sparse elements nearly zero know degree sparsity spike and slab prior golden standard for inducing sparsity choose prior @xmath12 gaussian scale mixture @xcite centred at zero approximate spike slab prior continuous prior favours sparse solutions zero probability mass zero values heavy - tailed priors infer large quasi zero values for kernel weights impose zero - mean gaussian scale mixtures on weights : @xmath13 @xmath14 vector of unobserved scale variables inverse gaussian densities imposed marginal @xmath15 symmetric generalised hyperbolic density @xcite fat tails compared gaussianfamily contains student-@xmath16 laplace gamma - variance jeffrey s special cases probabilistic model @xmath17 closed form expression marginal density observations @xmath18 @xmath19 kernel matrix @xmath20-th feature matrix marginal density defined for valid kernel matrices since @xmath21 @xmath20 linear additive model corresponds weight - space view representation multiple gaussian process model marginal density depends on linear kernel matrices @xmath22 model non - linear additive model replacing linear kernel matrices new representation corresponds multiple function - space view representation @xmath23 @xmath4 latent functions scaled gaussian process priors @xmath24 @xmath20 functions @xmath25 covariance functions valid kernel functions @xmath2 noisy observations sum @xmath4 latent function values @xmath26 likelihood function mgp prior given by @xmath27 @xmath28 @xmath30 vector @xmath31 unit vector of dimension @xmath4 operator @xmath32 kronecker product prior on @xmath33 given by gam prior marginal @xmath34 same form weight - space view representationmgp model @xmath4 independent non - gaussian process priors over function space condition on scale variable finite latent function values distributed according to multivariate gaussian marginal integrate scale variable latent function values distributed according to @xmath4 independent multivariate gaussian scale mixture densities @xmath35 @xmath36 modified bessel function second kind prior measure over function space heavy - tailed generalised hyperbolic measure @xcite gaussian process recovered for @xmath37 symmetric multivariate zero - mean hyperbolic process for @xmath38 special cases include multivariate gamma - variance process laplace process student-@xmath16 process cauchy process approach for estimation @xmath33 type ii maximum postriori optimisation using standard nonlinear optimisation tools regulariser chosen in advance attention to inference problem of parameters from data view @xmath33 as latent variable desired level sparsity learnt from data by optimising hyperparameters by type ii maximum likelihood follow mean field approach @xcitefind analytically tractable solution posterior over latent function values @xmath44 scale vector @xmath33 assumed factorise data @xmath45 variational posteriors maximising negative variational free energy lower bound log - marginal likelihood given by @xmath46 @xmath47 parameters defined as @xmath48 @xmath49 predictive mgp @xmath50 peaked around mean @xmath51 true predictive density approximated by integral : @xmath52 @xmath53 @xmath54 posterior mean variance same form as standard gp regression kernel replaced by convex combination of kernels expression @xmath55 same form as method kernel ridge regression ml ii updates for hyperparameters obtained solving expressions simple @xmath56 @xmath57 updates obtained direct maximising variational bound update for @xmath11 same restrict to binary classification scaled probit model likelihood derived from gaussian cumulative density probit model equivalent to gaussian noise step function likelihood @xmath58 class labels @xmath59 @xmath60 likelihood replaced by @xmath61 @xmath62 @xmath63 @xmath64regression consider mean field approximation assumes posterior form @xmath65 assume variational posterior @xmath66 product of truncated gaussians appendix @xmath67 @xmath68 @xmath69 posterior mean covariance @xmath44 unchanged @xmath70 replaced by @xmath71 elements @xmath71 defined in trun mean ] posterior @xmath50 updates hyperparameters identical mgp regression bayesian classification label highest probability @xmath72 selected exact computation intractable assume posteriors @xmath66 @xmath50 peaked around mean classification rule @xmath73 @xmath55 @xmath74 @xmath70 replaced by @xmath71 deciding label @xmath75 or @xmath76 sign @xmath55 decision rule compare frequentist bayesian approaches kernel combination demonstrate flexibility performance mgp models regression data generate random functions from hilbert spaces 10 laplacian kernels add gaussian i noise show results three settings sparse problem one kernel semi - sparse problem 3 functions non - sparse problem ten functions active table compares mgp models cross - validated mkl models with fixed regularisation norms hierarchical bayesian approach to sparsity datalcccr active kernels 1 out 10 3 out of 10 multiple student-@xmath16 .033 .067 .719 @xmath79 ) multiple laplace .034 @xmath80 .076 .704 @xmath82 ) gamma - variance .033 .067 .719 ) + ard .033 .066 .746 @xmath83 ) mkl @xmath1 .037 .066 .720 @xmath86 ) mkl @xmath87 .830 @xmath88 ) .831 @xmath89 ) .762 @xmath90 ) mkl @xmath91 .097 @xmath92 ) .233 @xmath93 .719 @xmath94 ) flowers data set standard mkl benchmark multi - class image classification @xmath95 flower classes one - versus - all classifier fig roc student roc curve two classes student-@xmath16 process average @xmath96 average @xmath97 gamma - variance process ard @xmath98 @xmath99 better than state - - - art mkl results generalised inverse gaussian density defined @xmath100 @xmath101 @xmath102 modified bessel function second kind index @xmath103value @xmath104 constraints @xmath105 @xmath106 define @xmath108 expectations useful @xmath109 @xmath42 @xmath41 inverse gaussian density reduces gamma density @xmath40 @xmath39 reduces inverse gamma density expectations simplify truncated gaussian density defined @xmath110 @xmath111 cumulative density unit gaussian @xmath112 mean variance @xmath113 | 0.506655 | 3,091 | 1,590 | , Saving $0.1 in GPT-4. |
154,929 | a preliminary design of the pep - n electromagnetic calorimeter is given .
the spatial , energy and time resolutions achievable using a kloe type electromagnetic calorimeter are presented . 0bar@xmath0 | 3 truemm .,width=288,height=288 ] the main goals of the pep - n physics programs are : * a measurement of r with a 2 @xmath1 or better error.this condition requests the measurement of all exclusive channels . * a measurement of the form factors .
* do spectroscopy in the energy range covered by the machine in order to study new possible final states . at pep - n energies , photons with energies lower that 100 mev are produced , as shown in fig .
[ fig:2dim ] for one of the channels with higher cross section @xmath2 . in order to achieve these goals the electromagnetic calorimeter should be : * hermetic . *
have high efficiency , till 20 mev , for charged and neutral particles .
* have good ( few @xmath1 ) energy resolution for photons . *
have good time resolution in order to separate events from other final states .
the kloe electromagnetic calorimeter@xcite was designed to detect photons in the ( 20 - 500 ) mev energy range with good time resolution , so its performances have been taken as guidelines in the design of the pep - n electromagnetic calorimeter .
this detector can provide a fast and unbiased first level trigger , with high acceptance for final states with low energy photons.a good k@xmath3 separation is achievable and also some kind of @xmath4 discrimination is possible .
the kloe calorimeter is a fine sampling lead and scintillating fibers calorimeter .
the barrel modules have trapezoidal cross section , 4.3 m long , 60 cm wide and 23 cm thick .
each module is obtained gluing 0.5 mm thick lead foils worked to house the 1 mm diameter fibers .
the resulting structure ( fig .
[ fig : pbscifi ] ) has fiber : lead : glue volume ratio of 42:48:10 , an average density of @xmath5 , a mean radiation length @xmath6 , and a sampling fraction of @xmath7 for minimum ionizing particles .
the readout granularity is @xmath8 ( 4.4 @xmath9 4.4 ) @xmath10 , for a total number of 4880 read - out channels .
a precision in measuring the photon conversion point in the transverse plane of @xmath11 cm has been achieved.the coordinate along the fiber is measured using the relation : z = v@xmath12 , with @xmath13 the time difference at the two module ends and @xmath14 the effective light propagation speed in the fibers .
the measured effective light propagation speed is @xmath15/ns .
the z resolution is @xmath16 .
the obtained performances ( fig .
[ fig : resolutions]),are summarized as follows : * detection efficiency for photons with energy between 20 mev and 500 mev of about 99@xmath1 .
* resolution in the photon conversion point position in the transverse plane of @xmath11 cm and in the z coordinate @xmath16 .
* energy resolution @xmath17 . *
time resolution @xmath18 .
the calorimeter is inside a 0.6 t magnetic field with a consequent residual magnetic field up to 0.2 t in the photomultipliers ( pm ) area and with an angle with respect to pm axis up to 25@xmath19 ; for this reason fine mesh photomultipliers , hamamatsu r5946 , were specially designed for kloe .
in fig . [ fig : exploded ] an exploded view of the calorimeter design is shown .
it consists of barrel , forward and backward detectors .
* each of the vertical sides of the _ barrel _ have 3 modules with rectangular cross section , 220 cm long , 55 cm high and 25 cm thick , with fibers parallel to the beams ( bcal detector ) .
also the horizontal sides have 3 modules 220 cm long , 50 cm wide , 15 cm thick , positioned over and under the tpc chamber , in order to complete the coverage of the azimuthal acceptance ( pcal detector ) .
the angular region covered by the barrel modules is 27@xmath20 . due to lack of space
pcal detector is only 15 cm thick .
the efficiency and energy resolution , simulated with montecarlo , are shown in fig .
[ fig : pcaleff ] and fig .
[ fig : pcalres ] . the efficiency is greater than 99@xmath1 for energies higher
then 40 mev .
the gamma energy resolution is @xmath21 ( fig .
[ fig : pcalres ] ) .
+ energies are in mev.,width=288,height=288 ] + * the _ forward detector _
( fcal ) is made of modules ( table 1 ) with the same thickness as bcal .
they are located at ( 130@xmath22 ) cm , cover an area of 280 * 180@xmath23 , with polar angle range 6@xmath24 ( fig .
[ fig : forward ] ) .
+ * the _ backward detector _ ( rcal ) is composed of 2 modules positioned at z= -140 cm .
they are 120 cm long , 54 cm high and 15 cm thick ( fig .
[ fig : back ] ) , specifically designed to detect very low energy photons .
the total area covered is 120 * 120 @xmath10 ( 135@xmath25 ) . in table
[ tab : unfolded ] the dimensions , number of modules and number of photomultipliers requested are summarized .
.sizes(cm),modules and photomultipliers for each detector .
[ tab : unfolded ] [ cols="^,^,^,^,^,^",options="header " , ] we have assumed the same readout granularity as kloe , that is @xmath8 ( 4.4 @xmath9 4.4 ) @xmath10 , so that the total number of photomultipliers is 1350 . with this configuration a precision in measuring the photon conversion point in the transverse plane of @xmath11 cm should be achieved .
the inclusive photon energy distributions in the four detectors , at e@xmath26 = 2.25 gev , is shown in fig .
[ fig : gammaenergy ] .
the pep - n detector is equipped with an aerogel detector for k-@xmath27 separation in the ( 0.6@xmath28)gev / c momentum range .
the tpc chamber can be used to separate particles with momenta lower then 0.6 gev / c ( de / dx measurement).unfortunately the tpc measures badly the de / dx of particles hitting the pole calorimeter pcal where , due to lack of space , it is difficult to insert a specific particle identification detector . because of very good time resolution , better then 0.2 ns for m.i.p .
, pcal can supply these informations .
[ fig : deltatov ] shows the k-@xmath27 separation as function of the momentum after 1 m of path lenght .
pions and kaons are well separated till 1 gev / c momenta after 1 m of path .
separation as function of momentum after 1 m of path lenght . ] in fig .
[ fig : time ] the time - momentum separation for the process kk@xmath29 is shown for fcal , bcal and pcal at e@xmath26 = 2.25 gev.a 3 @xmath30 separation is shown for these process till 0.8 gev / c momenta .
events for fcal , bcal and pcal at e@xmath26 = 2.25 gev.,width=288,height=288 ]
the present preliminary design of the electromagnetic calorimeter fulfills all the physics requirements for pep-n.the energy resolution is acceptable , even compared to more expensive crystal detectors.the excellent timing resolution is a very nice feature that could be used to reduce background contributions and do also particle identification .
the very high efficiency for very low energy photons gives the possibility to build a minimum bias first level trigger . | =288 main goals pep - n physics programs measurement r with 2 @xmath1 or error channels form factors spectroscopy energy range study final states pep - n energies photons lower 100 mev produced fig . cross section electromagnetic calorimeter should hermetic high efficiency till 20 mev neutral particles good energy resolution good time resolution separate events final states kloe electromagnetic calorimeter photons 20 - 500 ) mev energy range good time resolution performances guidelines design pep - n calorimeter fast unbiased first level trigger high acceptance for final states low energy photons good k@xmath3 separation @xmath4 discrimination possible kloe calorimeter fine sampling lead scintillating fibers barrel modules trapezoidal cross section 4.3 m long 60 cm wide 23 cm thick module gluing 0.5 mm thick lead foils 1 mm diameter fibers structure fig . fiber : lead : glue volume ratio 42:48:10 average density @xmath5 mean radiation length @xmath6 sampling fraction of @xmath7 for minimum ionizing particlesreadout granularity @xmath8 ( 4.4 @xmath10 4880 read - out channels precision measuring photon conversion point transverse plane @xmath11 cm achieved coordinate fiber measured z = v@xmath12 @xmath13 time difference module ends @xmath14 effective light propagation speed measured light propagation speed @xmath15/ns z resolution @xmath16 performances detection efficiency photons 20 mev and 500 mev 99@xmath1 resolution photon conversion point position transverse plane @xmath11 cm z coordinate @xmath16 energy resolution @xmath17 time resolution @xmath18 calorimeter inside 0.6 t magnetic field residual magnetic field 0.2 t photomultipliers area angle pm axis up to 25@xmath19 fine mesh photomultipliers hamamatsu r5946 designed for kloe calorimeter design barrel forward backward detectors vertical sides 3 modules 220 cm long 55 cm high 25 cm thick fibers parallel beams horizontal sides 3 modules 220 cm long 50 cm wide 15 cm thick over under tpc chamber azimuthal acceptance angular region covered barrel modules 27@xmath20 lack space pcal detector 15 cm thick efficiency energy resolution simulated with montecarlo shown in fig . efficiency greater than 99@xmath1 for energies higher 40 mev gamma energy resolution @xmath21 energies in mev.,width=288,height=288 forward detector ( fcal modules same thickness as bcal located at 130@xmath22 ) cm area 280 * 180@xmath23 polar angle range 6@xmath24 backward detector ( rcal 2 modules at z= -140 cm 120 cm long 54 cm high 15 cm thick detect low energy photons total area 120 * 120 @xmath10 ( 135@xmath25 ) dimensions number of modules number photomultipliers summarized for assumed same readout granularity as kloe @xmath8 ( 4.4 @xmath9 4.4 ) @xmath10 total number photomultipliers 1350 precision in measuring photon conversion point in transverse plane of @xmath11 cm achieved inclusive photon energy distributions in four detectors at e@xmath26 = 2.25 gev shown in fig .pep - n detector with aerogel detector for k-@xmath27 separation 0.6@xmath28)gev / c momentum range tpc chamber particles momenta lower then 0.6 gev / c tpc measures badly de / dx particles hitting pole calorimeter difficult to insert specific particle identification detector good time resolution better 0.2 ns pcal supply informations fig k-@xmath27 separation momentum after 1 m path lenght pions kaons separated till 1 gev / c momenta after 1 m path time - momentum separation kk@xmath29 for fcal bcal pcal at e@xmath26 = 2.25 gev 3 @xmath30 separation till 0.8 gev / c momenta preliminary design electromagnetic calorimeter fulfills physics requirements for pep-n energy resolution acceptable detectors excellent timing resolution background contributions particle identification high efficiency for low energy photons minimum bias first level trigger | 0.267923 | 1,760 | 889 | , Saving $0.1 in GPT-4. |
154,933 | in this letter , we propose a new generalized ricci dark energy ( ngr ) model to unify ricci dark energy ( rde ) and xcdm .
our model can distinguish between rde and xcdm by introducing a parameter @xmath0 called weight factor .
when @xmath1 , ngr model becomes the usual rde model .
the xcdm model is corresponding to @xmath2 .
moreover , ngr model permits the situation where neither @xmath1 nor @xmath2 .
we then perform a statefinder analysis on ngr model to see how @xmath0 effects the trajectory on the @xmath3 plane . in order to know the value of @xmath0
, we constrain ngr model with latest observations including type ia supernovae ( sne ia ) from union2 set ( 557 data ) , baryonic acoustic oscillation ( bao ) observation from the spectroscopic sloan digital sky survey ( sdss ) data release 7 ( dr7 ) galaxy sample and cosmic microwave background ( cmb ) observation from the 7-year wilkinson microwave anisotropy probe ( wmap7 ) results . with markov chain monte carlo ( mcmc ) method , the constraint result is @xmath0=@xmath4 , which manifests the observations prefer a xcdm universe rather than rde model .
it seems rde model is ruled out in ngr scenario within @xmath5 regions .
furthermore , we compare it with some of successful cosmological models using aic information criterion .
ngr model seems to be a good choice for describing the universe . | various cosmic observations suggest our universe is undergoing an accelerated expansion @xcite .
to explain this phenomenon , people introduce an exotic component with negative pressure known as dark energy .
the simplest dark energy model is cosmological constant ( @xmath7 ) @xcite or xcdm model where dark energy has an arbitrary equation of state ( eos ) @xmath8 .
it fits all kinds of observational data well while it is confronted with theoretical problems such as `` coincidence '' problem and `` fine - tuning '' problem @xcite . as a result
, other dark energy models have been widely proposed including quintessence @xcite , quintom @xcite , phantom @xcite , gcg @xcite and so on . in principle , dark energy is related to quantum gravity @xcite .
but until now , a self - consistent quantum gravity theory has not established .
nevertheless , the holographic principle @xcite is thought to be a reflection of quantum gravity . motivated by this , holographic dark energy has been proposed .
it embodies the relation between uv cut - off and ir cut off .
however , how to choose the ir cut - off is a problem .
cohen et al .
@xcite first chose hubble scale as ir cut - off .
hsu and li @xcite pointed out it can not give an acceleration solution .
li then suggested the future event horizon as ir cut - off @xcite .
basing on causality , cai proposed agegraphic dark energy @xcite and new agegraphic dark energy @xcite . furthermore , gao et al .
@xcite proposed a holographic dark energy from ricci scalar curvature . in rde model ,
the ir cut - off is determined by a local quantity .
nowadays , all the models above seem to be consistent with current observations . usually , we estimate models through the @xmath9 or information criteria like bic and aic @xcite . in this letter
, we find xcdm model and rde model can be related by a parameter @xmath0 , thus we can estimate them through constraining @xmath0 .
the distribution of @xmath0 can reflect which model is better .
for example , if the best - fit value of @xmath0 is close to 1 and 0 is not within 2-@xmath10 range , we can say the observations support rde model rather than xcdm model .
we now give some similar examples . in order to know whether @xmath7cdm is right , people free the eos parameter and constrain it with observations
. if the result is close to -1 , we can say @xmath7cdm is still a good choice
. however , if the eos parameter tends to -2 , then @xmath7cdm should be suspected .
likewise , for purely dimensional reasons , granda and oliveros @xcite proposed a new ir cut - off .
wang and xu @xcite give the constraint results which suggest the coefficient of @xmath11 is two times larger than the one of @xmath12 , thus ruling out the srde model @xcite .
in rde model , the density of dark energy is proportional to ricci scalar or the sum of traces of energy - momentum tensors of each component .
since the trace of radiation is 0 , we can ignore its impact on space - time curvature .
rde model suggests the weights of dark energy and matter are the same , while xcdm model suggests only the trace of dark energy can affect its density .
therefore , what on earth is the weight of matter ( 0.5 or 0 ? ) is an interesting thing we want to know .
motivated by this , we free the weight of matter as an arbitrary parameter called weight factor . the rest of the letter is organized as follows . in section 2 ,
we give the dynamics of the new generalized ricci dark energy model . in section 3
, we give a statefinder diagnostic . in section 4
, we introduce the observational data we use .
the constraint results are shown in section 5 .
at last , we give the discussion and conclusion in section 6 . throughout the letter ,
the unit with light velocity @xmath13 is used .
we assume the universe is flat and described by friedmann - robertson - walker ( frw ) metric . for rde model ,
the density of dark energy is proportional to ricci scalar @xmath15 from this equation , the coefficients of @xmath18 and @xmath19 are both 1 , which means the weights of dark energy and matter are the same .
we now change the weight of matter , the equation becomes @xmath20 @xmath0 here is the weight factor we introduce which reflects the relative weight of matter to dark energy .
if @xmath1 , it becomes the usual rde model .
when @xmath2 @xmath21 equivalently , @xmath22 the ngr model becomes xcdm model . for simplicity , we define dimensionless quantities @xmath23 , @xmath24 , where @xmath25 is the critical density of the universe .
@xmath26 is hubble parameter , subscript `` 0 '' represents the quantity today .
the friedmann equation can be expressed as @xmath27 where @xmath28 .
the energy - momentum conservation equation can be expressed as @xmath29 subscript `` i '' represents dark energy or matter
. then we get @xmath30 where @xmath31 . with the initial condition @xmath32
we can obtain the evolution of @xmath33 with respect to redshift z @xmath34 in order to exhibit the effects of @xmath0 , we fix @xmath37 and @xmath38 and plot the evolutions of @xmath39 , @xmath40 , hubble parameter @xmath41 and density parameters defined as @xmath42 in fig . [ fig1 ] and fig .
[ figh ] . * acknowledgments * this work was supported by the national natural science foundation of china under the distinguished young scholar grant 10825313 , the ministry of science and technology national basic science program ( project 973 ) under grant no.2012cb821804 , the fundamental research funds for the central universities and scientific research foundation of beijing normal university .
riess et al , astron .
j. * 116 * 1009 ( 1998 ) [ arxiv : astro - ph/9805201 ] ; s. perlmutter et al , astrophys .
j. * 517 * 565 ( 1999 ) [ arxiv : astro - ph/9812133 ] ; e. komatsu et al .
, astrophys .
* 192 * , 18 ( 2011 ) , arxiv:1001.4538 ; r. amanullah et al .
, [ supernova cosmology project collaboration ] , astrophys . j. * 716 * , 712 ( 2010 ) .
u. alam et al . , mon . not .
. soc . * 344 * 1057 ( 2003 ) ; v. gorini et al . , phys .
d. * 67 * 063509 ( 2003 ) ; x. zhang , phys .
b , * 611 * 1 ( 2005 ) ; m. malekjani et al . , astrophys . space .
* 334 * 193 ( 2011 ) . | cosmic observations suggest universe accelerated expansion introduce exotic component negative pressure dark energy simplest dark energy model is cosmological constant @xmath7 ) or xcdm model arbitrary equation of state fits observational data confronted with theoretical problems coincidence fine - tuning other dark energy models proposed including quintessence quintom phantom gcg dark energy related to quantum gravity self - consistent quantum gravity theory established holographic principle reflection of quantum gravity holographic dark energy proposed embodies relation between uv cut - off ir cut off ir cut - off problem cohen et al chose hubble scale as ir cut - off hsu li suggested future event horizon as ir cut - off cai proposed agegraphic dark energy new agegraphic dark energy gao et al . proposed holographic dark energy from ricci scalar curvature model ir cut - off determined by local quantity models consistent with observations estimate models through @xmath9 or information criteria like bic and aic @xciteletter xcdm rde model related by parameter @xmath0 estimate constraining @xmath0 distribution @xmath0 model better if best - fit value @xmath0 close to 1 0 not within 2-@xmath10 range observations support rde model xcdm similar examples @xmath7cdm free eos parameter constrain with observations if result close to -1 @xmath7cdm good choice if eos parameter tends to -2 @xmath7cdm suspected granda oliveros proposed new ir cut - off wang and xu constraint results coefficient of @xmath11 two times larger than @xmath12 ruling out srde model rde model density of dark energy proportional to ricci scalar traces energy - momentum tensors trace of radiation 0 ignore impact on space - time curvature rde model suggests weights of dark energy matter same xcdm model trace of dark energy density weight of matter 0.5 or 0 free weight of matter as arbitrary parameter weight factor letter organized section 2 dynamics of new generalized ricci dark energy model section 3 statefinder diagnostic section 4 observational data constraint results in section 5discussion conclusion section 6 unit light velocity @xmath13 used universe flat described friedmann - robertson - walker metric density dark energy proportional to ricci scalar @xmath15 coefficients @xmath18 @xmath19 1 weights dark energy matter same change weight matter equation becomes @xmath20 @xmath0 weight factor reflects relative weight matter to dark energy @xmath1 usual rde model @xmath2 @xmath21 @xmath22 model becomes xcdm model define dimensionless quantities @xmath23 @xmath24 @xmath25 critical density universe @xmath26 hubble parameter 0 quantity friedmann equation @xmath27 @xmath28 energy - momentum conservation equation @xmath29 i represents dark energy matter @xmath30 @xmath31 initial condition @xmath32 evolution of @xmath33 redshift z @xmath34 effects fix @xmath37 @xmath38 plot evolutions @xmath39 @xmath40 hubble parameter @xmath41 density parameters @xmath42 in fig .work supported national natural science foundation china young scholar grant 10825313 ministry science technology national basic science program project 973 grant.2012cb821804 fundamental research funds central universities scientific research foundation beijing university riess astron j. 116 * 1009 1998 ) ph/9805201 perlmutter astrophys . 517 * 565 1999 )/9812133 komatsu astrophys. 192 18 2011 ) arxiv:1001.4538 amanullah supernova cosmology project collaboration astrophys . j 716 * 712 2010 ) alam 344 * 1057 2003 ) gorini . 67 * 063509 2003 ) zhang phys 611 * 1 2005 ) malekjani astrophys . 334 * 193 ( 2011 ) | 0.265941 | 1,591 | 822 | , Saving $0.0 in GPT-4. |
154,946 | ghmann , klmper and seel derived the multiple integral formula of the density matrix of the @xmath0 heisenberg chain at finite temperatures .
we have applied the high temperature expansion ( hte ) method to isotropic case of their formula in a finite magnetic field and obtained coefficients for several short - range correlation functions .
for example , we have succeeded to obtain the coefficients of the hte of the 3rd neighbor correlation function @xmath1 for zero magnetic field up to the order of 25 .
these results expand our previous results on the emptiness formation probability [ z.tsuboi , m.shiroishi , j. phys.a : math .
gen .
38(2005 ) l363 ] to more general correlation functions .
_ msc : _ 82b23 ; 45g15 ; 82b20 ; 82b80 + _ pacs : _
75.10.jm , 02.30.ik , 05.70.-a , 05.30.-d + _ key words : _ correlation function ; density matrix ; high temperature expansion ; nonlinear integral equation ; quantum transfer matrix + _ report - no : _ * oiqp-06 - 10 * + * to appear in physica a * | ghmann , klmper and seel derived @xcite ( see also @xcite ) a multiple integral formula of matrix elements of a density matrix of a finite segment of arbitrary length @xmath2 of the anti - ferromagnetic spin @xmath3 @xmath0 heisenberg infinite chain at finite temperature in a finite magnetic field by combining the quantum transfer matrix approach @xcite-@xcite and the algebraic bethe ansatz technique .
their formula generalizes the multiple integral formulae for zero temperature @xcite to finite temperature case .
this is a fundamental quantity since thermal average of any operators acting non - trivially on the segment of length @xmath2 can be expressed in terms of their formula .
thus it is an important problem to perform this multiple integral and extract concrete numbers from it .
their formula contains an auxiliary function , which is a solution of a nonlinear integral equation .
this nonlinear integral equation is essentially same as the one for the free energy @xcite .
thus to evaluate their formula consists of two non - trivial tasks : to solve the nonlinear integral equation and to integrate the multiple integrals . in our previous paper @xcite , we applied the high temperature expansion ( hte ) method to a multiple integral formula @xcite of the emptiness formation probability @xmath4 for the @xmath5 model , which is the probability of @xmath2 adjacent spins being aligned upward , and succeeded to obtain the coefficients of @xmath6 up to the order of 42 .
as for zero magnetic field case , there is also numerical calculation @xcite for the multiple integral for @xmath7 .
the purpose of this paper is to expand our previous results on the hte for @xmath4 @xcite to more general correlation functions for the spin @xmath3 isotropic heisenberg chain in a magnetic field @xmath8 . in section 2
, we introduce the multiple integral formula of the matrix elements of the density matrix @xcite . in section 3 , we evaluate this multiple integral by the hte method . based on the result of the hte of the density matrix , we will calculate the hte of two point correlation functions ( [ 2ndzz])-([2ndpm ] ) .
in particular for zero magnetic field case , we have succeeded to obtain the coefficients of the hte of a 3rd neighbor correlation function up to the order of 25 ( cf .
( [ 3rdh=0 ] ) ) .
section 4 is devoted to concluding remarks .
the hamiltonian of the spin-@xmath3 isotropic heisenberg chain in a magnetic field @xmath8 is given as @xmath9 where @xmath10 , @xmath11 , @xmath12 are the pauli matrices which act non - trivially on the @xmath13-th lattice site in a chain of length @xmath14 . here
the periodic boundary condition @xmath15 is assumed .
let us introduce @xmath16 matrices : @xmath17 these matrices are embedded into the space @xmath18 on which the hamiltonian ( [ hamil ] ) acts : @xmath19 where @xmath20 , @xmath21 and @xmath22 .
the above pauli matrices can be written in terms of these matrices : @xmath23 , @xmath24 , @xmath25 .
we also put @xmath26 and @xmath27 .
ghmann , klmper and seel derived @xcite an integral representation of the density matrix of the @xmath0 chain at finite temperature @xmath28 .
the isotropic ( @xmath5 ) limit of their formula is given as follows .
@xmath29 where @xmath30 and @xmath31 are solutions of the following integral equations .
@xmath33 here @xmath34 and @xmath35 are sequences of @xmath36 or @xmath37 .
we define the number of @xmath36 in @xmath34 as @xmath38 and the position @xmath39 of @xmath13-th @xmath36 in @xmath34 as @xmath40 : @xmath41 , @xmath42 .
we also define the number of @xmath37 in @xmath35 as @xmath43 and the position @xmath39 of @xmath13-th @xmath37 in @xmath35 as @xmath44 : @xmath45 , @xmath46 .
we shall put @xmath47 for @xmath48 and @xmath49 for @xmath50 .
the contour @xmath51 surrounds the real axis anti - clockwise manner .
@xmath52 is defined as @xmath53 .
the emptiness formation probability is a special case of this density matrix element : @xmath54 . in this case
, the multiple integral formula ( [ density ] ) reduces to the one in @xcite .
in this paper , we will calculate the hte of the following two - point correlation functions for finite magnetic field @xmath8 : @xmath55 at @xmath56 , one can express nearest neighbor and 2nd neighbor correlation functions in terms of @xmath57 : @xmath58 , @xmath59
. then we can immediately calculate the hte of these correlation functions up to the order of 42 from the results in @xcite .
we can also calculate the hte of @xmath60 up to the order of 42 from the relation @xmath61 which holds for isotropic model ( @xmath5-model ) at @xmath56 .
moreover one can calculate the nearest neighbor correlation functions by taking the derivative of the free energy of the @xmath0-chain with respect to the anisotropy parameter . on the other hand
, @xmath1 can not be expressed only in terms of @xmath4 even at @xmath56 . at first
, we will calculate the hte of @xmath30 from the nlie ( [ nlie1 ] ) . note that this calculation is similar to the one for the hte for the free energy @xcite .
we assume the following expansion for small @xmath62 : @xmath63 substituting ( [ a - ex ] ) into ( [ nlie1 ] ) , and comparing coefficients of @xmath64 on both sides , we obtain the following integral equation for each @xmath2 ( @xmath65 ) : @xmath66 where @xmath67 is made of @xmath68 : @xmath69 we can solve ( [ nlie1-mth ] ) recursively .
the first few terms of @xmath70 are @xmath71 one sees that only @xmath72 has a pole at the origin .
one can also solve the integral equation ( [ nlie2 ] ) based on the results of ( [ nlie1 ] ) .
we assume the following expansion for small @xmath62 : @xmath73 substituting ( [ ex - g ] ) into ( [ nlie2 ] ) , we can obtain the coefficients .
the first few terms of @xmath74 are @xmath75 one see that only @xmath76 has a pole at @xmath77
. taking note on the above results and substituting ( [ a - ex ] ) and ( [ ex - g ] ) into ( [ density ] ) , we can calculate the coefficients of the hte for the density matrix in a finite magnetic field @xmath8 only by taking residues at the origin .
then we can calculate the hte of the two point functions ( [ 2ndzz])-([2ndpm ] ) by using the result on the hte of matrix elements of the density matrix .
for example , the hte of a 3rd neighbor correlation function at @xmath56 up to the order of 25 is given as follows : @xmath78 where we put @xmath79 .
we also calculated the hte of @xmath80 for finite magnetic fields @xmath81 up to the order of 30 ; @xmath82 for @xmath81 up to the order of 20 ; @xmath83 for @xmath81 up to the order of 30 . in @xcite ,
the hte of correlation functions for zero magnetic field were calculated up to the order of 19 by other method .
it is remarkable that we can also treat nonzero magnetic field case and the orders of our hte are higher than the ones in @xcite .
we have plotted the pad approximations of the hte of these correlation functions for finite magnetic fields @xmath8 in figures [ zz2af]-[htegreenf ] .
once we introduce the hte , we can also consider @xmath84 case by analytic continuation , though original multiple integral formula ( [ density ] ) was defined for @xmath85 case .
we have also plotted quantum monte carlo simulation ( qmc ) data by shiroishi @xcite based on the open source software in the alps project @xcite .
the system size of the simulations is @xmath86 .
our hte results agree well with these qmc data except in the very low temperature regions .
at least in the region where the pad approximation seems to converge , @xmath80 for @xmath85 at @xmath56 and for @xmath84 , @xmath82 for @xmath84 and @xmath83 at @xmath56 monotonously decrease with respect to temperature ( here we omit a figure of @xmath80 for @xmath84 ) ; @xmath82 for @xmath85 at @xmath56 monotonously increases with respect to temperature . when @xmath87 in figures [ zz2af],[hteaf],[htegreenf ] , peaks appear at non - trivial temperatures .
dependence of @xmath80 for @xmath85 with a magnetic field @xmath8 .
we have plotted the pad approximations of order @xmath88 $ ] for @xmath56 and @xmath89 $ ] for @xmath81 ( [ @xmath39,@xmath90 means that the numerator is a degree @xmath39 polynomial of @xmath91 and the denominator is a degree @xmath92 polynomial of @xmath91 ) .
an exact value of @xmath80 at @xmath93 was calculated in @xcite . ,
scaledwidth=100.0% ] dependence of @xmath82 for @xmath85 with a magnetic field @xmath8 .
we have plotted the pad approximations of order @xmath94 $ ] for @xmath56 and @xmath95 $ ] for @xmath81 .
an exact value of @xmath96 at @xmath93 was calculated in @xcite .
, scaledwidth=100.0% ] dependence of @xmath82 for @xmath84 with a magnetic field @xmath8 .
we have plotted the pad approximations of order @xmath94 $ ] for @xmath56 and @xmath95 $ ] for @xmath81 .
, scaledwidth=100.0% ] dependence of @xmath83 for @xmath85 with a magnetic field @xmath8 .
we have plotted the pad approximations of order @xmath88 $ ] for @xmath56 and @xmath89 $ ] for @xmath81 .
an exact value of @xmath97 at @xmath93 was calculated in @xcite .
, scaledwidth=100.0% ] dependence of @xmath83 for @xmath84 with a magnetic field @xmath8 .
we have plotted the pad approximations of order @xmath88 $ ] for @xmath56 , @xmath89 $ ] for @xmath98 and @xmath99 $ ] for @xmath100 .
, scaledwidth=100.0% ]
in this paper , we have evaluated the multiple integral formula of the density matrix by using the hte method , and thereby obtained the hte of several two point correlation functions . together with our previous paper @xcite on @xmath4
, we recognized that the hte method is efficient to evaluate the multiple integral formulae of correlation functions @xcite-@xcite . in @xcite ,
the multiple integral formula ( [ density ] ) has been reduced to finite sums over products of single integrals for short segments of length 2 and 3 .
if a similar reduction is done for length 4 , one will be able to calculate the hte whose order is higher than the one in this paper . + * acknowledgments * we would like to thank masahiro shiroishi for discussions and providing us his unpublished qmc data @xcite .
mathematica computation in this work was carried out at the yukawa institute computer facility .
f. alet , p. dayal , a. grzesik , a. honecker , m. koerner , a. laeuchli , s.r .
manmana , i.p .
mcculloch , f. michel , r.m .
noack , g. schmid , u. schollwoeck , f. stoeckli , s. todo , s. trebst , m. troyer , p. werner , s. wessel : j. phys .
74 ( 2005 ) 30 - 35 ; cond - mat/0410407 .
( see also http://alps.comp-phys.org ) | ghmann klmper seel derived @xcite multiple integral formula matrix elements density matrix finite segment arbitrary length anti - ferromagnetic spin heisenberg infinite chain finite temperature finite magnetic field combining quantum transfer matrix approach algebraic bethe ansatz technique formula generalizes multiple integral formulae for zero temperature finite temperature case fundamental quantity thermal average operators acting segment length expressed formula important problem perform multiple integral extract concrete numbers formula contains auxiliary function solution nonlinear integral equation same as free energy @xcite formula tasks solve integrate multiple integrals applied high temperature expansion hte method multiple integral formula emptiness formation probability @xmath5 model probability adjacent spins aligned upward coefficients @xmath6 up to 42 zero magnetic field case numerical calculation for multiple integral @xmath7 purpose paper expand previous results hte @xmath4 to general correlation functions spin @xmath3 isotropic heisenberg chain in magnetic field @xmath8 section 2 introduce multiple integral formula matrix elements density matrix @xcite section 3 evaluate multiple integral by hte methodresult density matrix calculate two point correlation functions 2ndzz])-([2ndpm ] zero magnetic field case coefficients 3rd neighbor correlation function up 25 3rdh=0 section 4 concluding remarks hamiltonian spin-@xmath3 isotropic heisenberg chain in magnetic field @xmath8 @xmath9 pauli matrices act on @xmath13-th lattice site chain @xmath14 periodic boundary condition @xmath15 assumed introduce @xmath16 matrices embedded into space @xmath18 hamiltonian acts @xmath19 @xmath20 pauli matrices @xmath23 @xmath25 @xmath26 @xmath27 ghmann klmper seel derived integral representation density matrix @xmath0 chain at finite temperature @xmath28 isotropic @xmath5 ) limit formula @xmath29 @xmath30 solutions integral equations @xmath33 @xmath34 sequences of @xmath36 or @xmath37 define number @xmath36 in @xmath34 as @xmath38 position @xmath39 of @xmath13-th @xmath36 in @xmath34 as @xmath40 @xmath41 @xmath42define number @xmath37 @xmath35 as @xmath43 position @xmath39 @xmath13-th @xmath37 as @xmath44 @xmath47 for @xmath48 @xmath49 @xmath50 contour @xmath51 surrounds axis anti - clockwise @xmath52 @xmath53 emptiness formation probability special case density matrix element @xmath54 multiple integral formula density reduces to @xcite calculate two - point correlation functions finite magnetic field @xmath8 : @xmath55 @xmath56 express nearest neighbor 2nd neighbor correlation functions @xmath57 @xmath58 @xmath59 calculate hte correlation functions 42 from results @xcite calculate hte @xmath60 up 42 from relation @xmath61 isotropic model @xmath5-model @xmath56 calculate nearest neighbor correlation functions derivative free energy @xmath0-chain anisotropy parameter @xmath1 not expressed @xmath4 at @xmath56 calculate hte @xmath30 from nlie nlie1 calculation similar to free energy @xcite assume expansion @xmath62 : @xmath63 substituting a - ex ] into [ nlie1 ] comparing coefficients @xmath64 obtain integral equation @xmath2 @xmath65 : @xmath66 @xmath67 @xmath68 @xmath69 solve [ nlie1-mth ] ) recursively terms @xmath70 @xmath71 @xmath72 pole origin solve integral equation ( [ nlie2 ] ) results nlie1 expansion @xmath62 : @xmath73 substituting [ ex - g ] ) into [ nlie2 ] obtain coefficients terms @xmath74 @xmath75 @xmath76 pole @xmath77 substituting [ a - ex ] [ ex - g ] into [ density ] ) calculate coefficients density matrix finite magnetic field @xmath8 taking residues origin calculate hte two point functions [ 2ndzz])-([2ndpm ] ) result hte matrix elements density matrix hte 3rd neighbor correlation function @xmath56 25 @xmath78 @xmath79 calculated hte @xmath80 magnetic fields @xmath81 30 @xmath82 20 @xmath83 30@xcite hte correlation functions zero magnetic field calculated up to 19 treat nonzero magnetic field case orders hte higher than plotted pad approximations hte correlation functions for finite magnetic fields @xmath8 in figures zz2af]-[htegreenf introduce hte consider @xmath84 case analytic continuation original multiple integral formula density defined for @xmath85 case plotted quantum monte carlo simulation qmc data by shiroishi @xcite open source software alps project system size @xmath86 hte results agree with qmc data except low temperature regions @xmath80 @xmath85 @xmath56 @xmath84 @xmath82 decrease temperature @xmath80 @xmath82 increases @xmath87 figures zz2af peaks appear at non - trivial temperatures @xmath80 for @xmath85 with magnetic field @xmath8 plotted pad approximations of order @xmath88 @xmath56 @xmath89 @xmath81 @xmath39,@xmath90 numerator degree @xmath39 polynomial of @xmath91 denominator degree @xmath92 polynomial of @xmath91value @xmath80 at @xmath93 calculated @xcite scaledwidth=100.0% dependence @xmath82 @xmath85 magnetic field plotted pad approximations @xmath94 $ @xmath56 @xmath95 $ @xmath81 @xmath96 at @xmath93 calculated scaledwidth=100.0% dependence @xmath82 @xmath84 magnetic field plotted pad approximations @xmath94 $ @xmath56 $ @xmath81 scaledwidth=100.0% dependence @xmath83 @xmath85 magnetic field plotted approximations @xmath88 $ @xmath56 @xmath89 $ @xmath81 value @xmath97 at @xmath93 calculated scaledwidth=100.0% dependence @xmath83 @xmath84 magnetic field @xmath8 plotted pad approximations @xmath88 $ @xmath56 @xmath89 $ @xmath98 @xmath99 $ @xmath100 scaledwidth=100.0% evaluated multiple integral formula density matrix hte method obtained two point correlation functions hte method efficient evaluate multiple integral formulae correlation functions multiple integral formula reduced finite sums single integrals short segments length 2 3similar reduction length 4 calculate hte order higher paper thank masahiro shiroishi discussions unpublished qmc data mathematica computation yukawa institute computer facility f. alet p. dayal. grzesik honecker m. koerner laeuchli manmana mcculloch michel noack g. schmid u. schollwoeck stoeckli s. todo m. troyer p. werner s. wessel j. . 74 ( 2005 ) 30 - 35 cond - mat/0410407 http://alps.comp-phys.org | 0.430317 | 2,867 | 1,539 | , Saving $0.1 in GPT-4. |
154,950 | we show the usefulness of the fisher - shannon information product in the study of the sequence of collapses and revivals that take place along the time evolution of quantum wavepackets .
this fact is illustrated in two models , the quantum bouncer and a graphene quantum ring . | sequences of collapses and revivals in the wavepackets temporal evolution are a well known aspect of quantum dynamics .
this phenomenon has been theoretically understood @xcite and to date it has been observed in striking experiments with atoms and molecules @xcite , bose - einstein condensates @xcite and recently in coherent states in a kerr medium @xcite .
moreover , quantum revivals have been studied theoretically in low - dimensional quantum structures as graphene , graphene quantum dots and rings in perpendicular magnetic fields @xcite in this paper we show that the analysis of wavepacket quantum revivals can be carried out using the fisher - shannon product @xmath0 , defined as @xcite : @xmath1 with @xmath2 being the fisher information , @xmath3 the shannon entropy , and @xmath4 the so - called entropy power @xcite .
it is known that these information measures show complementary descriptions of the spreading or concentration of the probability density , where the fisher information gives a local measure of the spreading ( due to the gradient in the functional form ) , whereas the shannon entropy provides a global one .
this entropic product has proved useful in the analysis of different physical situations , i. e. , electronic correlation @xcite , atomic physics @xcite , chemical reactions @xcite , quantum phase transitions @xcite , astrophysics @xcite or in the study of geophysical phenomena @xcite .
there is a generalization of the fisher - shannon product , the so - called fisher - rnyi product @xcite .
note here the existence of other important complexity measures which have also been used in the description of a great variety of systems ( see @xcite and references therein ) .
we shall consider the fisher - shannon information product as it applies to quantum revival phenomena .
in particular , we shall show the role of this quantity in the dynamics of two model systems that exhibits sequences of quantum collapses and revivals : the so - called quantum ` bouncer ' ( that is a quantum particle bouncing against a hard surface under the influence of gravity ) and a graphene quantum ring model .
it is well known that the temporal evolution of localized bound states @xmath5 for a time independent hamiltonian is given in terms of the eigenvectors @xmath6 and eigenvalues @xmath7 as @xmath8 where @xmath9 are the fourier components of the vector @xmath5 , and @xmath10 is the main quantum number of the system ( in general one has to consider the set of quantum numbers corresponding to the system , see @xcite , but in this paper we will consider only systems with one quantum number ) .
now , a wavepacket is constructed with the coefficients @xmath11 tightly centered around a large value of @xmath12 , with @xmath13 . the exponential factor in ( [ expansion ] )
can then be written as a taylor expansion around @xmath14 ( within this approximation , @xmath10 is a continuous variable ) as @xmath15 \notag \\ = & \exp\left(-i\omega_0 t - 2\pi i(n - n_0 ) t / t_{\rm cl } -2\pi i ( n - n_0)^2 t / t_{\rm r } + \cdots \right)\end{aligned}\ ] ] where each term in the exponential ( except for the first one , which is a global phase ) defines a characteristic time scale , that is , @xmath16 and @xmath17 ( see @xcite for more details ) .
the so called fractional revival times can be given in terms of the quantum revival time - scale by @xmath18 , where @xmath19 and @xmath20 are mutually prime .
next , we study the wave packet dynamics by means of the so - called entropy product , i.e. , the product of the fisher information and the shannon entropy power , @xmath21 , to conclude that it provides another framework for visualizing fractional revival phenomena . again
, we expect that the formation of a number of minipackets of the original packet will correspond to relative minima of the information product . before proceeding ,
recall that @xmath0 satisfies the isoperimetric inequality @xcite @xmath22 the equality is reached for gaussian densities . by combining the above inequality with the stam uncertainty principle @xcite @xmath23 and the power entropy inequality @xcite @xmath24 leads to the usual formulation of the uncertainty principle in terms of the variance in conjugate spaces , @xmath25 .
it is straightforward to show that the equality limit of these four inequalities is reached for gaussian densities .
quantum states of matter in a gravitational field have been recently realized experimentally with neutrons @xcite .
these were allowed to fall towards a horizontal mirror which , together with the earth s gravitational field , provided the necessary confining potential well . from a theoretical point of view , this constitutes an example of the quantum variant of a classical particle subject to a uniform downward force , above an impermeable flat surface .
the revival behavior of quantum bouncers has been discussed in @xcite and an entropy - based approach was carried out in @xcite . here , we shall evaluate the goodness of the entropy - product when applied to a quantum bouncer .
the time dependent solution of the schrdinger equation for the potential @xmath26 if @xmath27 and @xmath28 otherwise reads @xmath29 where the eigenfunctions and eigenvalues are given by @xcite @xmath30 in the above equation , position and energy variables have been rescaled , @xmath31 , @xmath32 , and are denoted by primed symbols .
@xmath33 is a characteristic gravitational length , @xmath34 is the airy function , @xmath35 denotes its zeros , and @xmath36 is the @xmath37 normalization factor . @xmath38 and
@xmath36 were determined numerically by using scientific subroutine libraries for the airy function , although accurate analytic approximations for them can be found in @xcite . in the remainder of this paper , the primes on the energy and position variables will be omitted and we shall assume initial conditions that correspond to gaussian wave packets localized at a height @xmath39 above the surface , with a width @xmath40 and an initial momentum @xmath41 , @xmath42 the corresponding coefficients @xmath11 of the wave function ( [ wavefunction ] ) can be obtained analytically as @xcite , @xmath43 \nonumber \\ & & \times { \rm ai}\left(z_0-z_n+\frac{\sigma^4}{16 } \right),\end{aligned}\ ] ] with @xmath44 .
it is now a straightforward calculation to obtain for the classical period and the revival time @xmath45 and @xmath46 , respectively @xcite .
the temporal evolution of the wave packet in momentum - space is obtained numerically by the fast fourier transform method .
we have computed the temporal evolution of the entropy product ( [ fs ] ) for the initial conditions @xmath47 , @xmath41 , and @xmath48 .
figure [ fig1 ] displays @xmath49 and the location of the main fractional revivals .
it can be neatly observed that the entropic product decreases and reaches a minimum at most of the fractional revivals , where the quasiclassical behavior and a gaussian shape are recovered .
notice how the initial value of unity for the information product is approximately recovered at the full revival , when the gaussian form of the wavepacket is roughly restored .
, and @xmath50 in a quantum bouncer .
the main fractional revivals are indicated by vertical dotted lines .
, scaledwidth=85.0% ] we shall consider the behavior of the fisher - shannon information in another physical situation .
let us consider a graphene quantum ring within a simplified model @xcite which is described by the hamiltonian @xmath51 where @xmath52 corresponds to each of the two inequivalent corners @xmath53 and @xmath54 of the first brillouin zone , @xmath55 is the momentum measured relatively to the @xmath53 ( @xmath56 ) point , @xmath57 a finite mass term , @xmath58 m / s is the fermi velocity , and @xmath59 where the components are the pauli matrices .
we shall work using a geometrical approximation of a zero width ring with radius @xmath60 ( used in @xcite ) .
the eigenstates and eigenfunctions are given in polar coordinates by @xcite @xmath61 @xmath62 @xmath63 where @xmath64 , @xmath65 and @xmath66 , and with @xmath67 with @xmath68 , @xmath69 , @xmath70 being the eigenvalue of the total angular momentum @xmath71 .
now we shall construct the initial wave packet centered around an eigenvalue @xmath72 . in fig .
[ ring ] the fisher - shannon product for a ring of @xmath73 nm , @xmath74 mev and @xmath75 mev is depicted .
we can observe a quasiclassical evolution ( with a period of @xmath76 ns ) at early times , which corresponds to a minimum in the fisher - shannon product .
we can see that at @xmath77 ( and at multiples times of it ) , where the wave - packet recovers its quasiclassical behavior , we have a relative minimum again .
mev , @xmath78 , for a quantum ring with @xmath73 nm , @xmath74 mev . ]
we have presented an analysis of quantum wavepacket revival phenomena in the quantum bouncer and a graphene ring , based on the information product .
we have shown that this theoretical tool ( which has proved to be useful for the analysis of different phenomena in atomic physics , molecular reactions , solids , and even in geophysics ) appropriately describes the dynamics of wavepackets .
in particular , we have found that the revivals and fractional revivals of wavepackets correspond to relative minima in the entropic product , signaling the recovery of the quasiclassical behavior of the wavepacket .
this work was supported by the spanish projects no .
micinn fis2009 - 08451 , no .
fqm-02725 ( junta de andaluca ) , and no .
micinn fis2011 - 24149 .
g. rempe , h. walther , and n. klein , phys .
lett . * 58 * , 353 ( 1987 ) ; j. a. yeazell , m. mallalieu , and c. r. stroud , jr .
lett . * 64 * , 2007 ( 1990 ) ; t. baumert _
et al . _ ,
. lett . * 191 * , 639 ( 1992 ) ; m.j.j .
vrakking , d. m. villeneuve , and a. stolow , phys .
a * 54 * , r37-r40 ( 1996 ) ; a. rudenko _
et al _ t. ergler , and b. feuerstein et al . , chem .
* 329 * , 193 ( 2006 ) .
a. lpez , z. z. sun , and j. schliemann , phys .
b , * 85 * 205428 ( 2012 ) . tomasz m. rusin and wlodek zawadzki , phys .
b , * 78 * , 215419 ( 2008 ) .
john schliemann , new j. phys .
* 10 * , 043024 ( 2008 ) .
demikhovskii , g. m. maksimova , a. a. perov and a. v. telezhnikov , phys . rev .
a , * 85 * , 022105 ( 2012 ) .
e. romera , and j. s. dehesa , j. chem . phys . *
120 * , 8906 ( 2004 ) .
nesvizhevsky and k.v .
protasov , quantum states of neutrons in the earth s gravitational field : state of the art , applications , perspectives , in _ trends in quantum gravity research _
editor david c. moore , pp . 65 - 107 nova science publishers , inc . ( 2006 ) . | collapses revivals in wavepackets temporal evolution are aspect of quantum dynamics phenomenon theoretically understood observed in experiments with atoms molecules bose - einstein condensates in coherent states in kerr medium quantum revivals studied in low - dimensional quantum structures as graphene quantum dots rings in perpendicular magnetic fields analysis of wavepacket quantum revivals using fisher - shannon product @xmath0 @xmath1 @xmath2 fisher information @xmath3 shannon entropy @xmath4 entropy power these information measures show complementary descriptions of spreading probability density fisher information local measure spreading shannon entropy global entropic product useful in analysis of physical situations electronic correlation atomic physics chemical reactions quantum phase transitions astrophysics study geophysical phenomena generalization of fisher - shannon product fisher - rnyi product of other important complexity measures used in description variety of systems consider fisher - shannon information product to quantum revival phenomenashow role quantity in dynamics of two model systems quantum collapses revivals : quantum ` bouncer ' particle bouncing against surface gravity and graphene quantum ring model temporal evolution of localized bound states @xmath5 for time independent hamiltonian given in eigenvectors @xmath6 and eigenvalues @xmath7 as @xmath8 where @xmath9 are fourier components of vector @xmath5 @xmath10 is main quantum number system consider set quantum numbers system consider only systems with one quantum number wavepacket constructed with coefficients @xmath11 centered around large value of @xmath12 with @xmath13 exponential factor in expansion written as taylor expansion around @xmath14 @xmath10 continuous variable as @xmath15 - 2\pi - each term in exponential first global phase defines characteristic time scale @xmath16 and @xmath17 see @xcite for fractional revival times given quantum revival time - scale by @xmath18 @xmath19 and @xmath20 are mutually prime study wave packet dynamics entropy product product fisher information shannon entropy power @xmath21 provides framework visualizing fractional revival phenomena formation minipackets original packet correspond minima information product @xmath0 satisfies isoperimetric inequality @xmath22 equality reached for gaussian densities combining inequality stam uncertainty principle @xmath23 power entropy inequality @xmath24 formulation uncertainty principle variance in conjugate spaces @xmath25 equality limit reached for gaussian densities quantum states matter in gravitational field realized with neutrons @xcite fall towards horizontal mirror earth s gravitational field confining potential quantum variant classical particle uniform downward force impermeable flat surface revival behavior of quantum bouncers discussed @xcite entropy - based approach evaluate goodness entropy - product applied quantum bouncer time dependent solution schrdinger equation potential @xmath26 @xmath27 @xmath28 reads @xmath29 eigenfunctions eigenvalues given by @xcite @xmath30 position energy variables rescaled @xmath31 @xmath32 denoted by primed symbols@xmath33 gravitational length @xmath34 airy function zeros @xmath36 @xmath37 normalization factor @xmath38 @xmath36 determined scientific subroutine libraries for airy function accurate analytic approximations in @xcite primes on energy position variables omitted assume initial conditions gaussian wave packets height @xmath39 above surface width @xmath40 initial momentum @xmath41 @xmath42 coefficients @xmath11 of wave function obtained analytically as @xcite @xmath43 with @xmath44 straightforward calculation for classical period revival time @xmath45 @xmath46 @xcite temporal evolution of wave packet in momentum - space obtained by fast fourier transform method computed temporal evolution entropy product for initial conditions @xmath47 @xmath41 @xmath48 figure displays @xmath49 location main fractional revivals entropic product decreases minimum at most fractional revivals quasiclassical behavior gaussian shape recovered initial value of unity information product recovered at full revival gaussian form wavepacket restored@xmath50 quantum bouncer fractional revivals vertical dotted lines scaledwidth=85.0% consider behavior fisher - shannon information physical situation graphene quantum ring simplified model @xcite described hamiltonian @xmath51 @xmath52 corners @xmath53 first brillouin zone @xmath55 momentum @xmath53 point @xmath57 finite mass term @xmath58 m s fermi velocity @xmath59 components pauli matrices geometrical approximation zero width ring radius @xmath60 @xcite eigenstates eigenfunctions polar coordinates @xcite @xmath61 @xmath62 @xmath63 @xmath64 @xmath66 @xmath67 @xmath68 @xmath70 eigenvalue total angular momentum @xmath71 construct initial wave packet eigenvalue @xmath72 fig . fisher - shannon product @xmath73 nm @xmath74 mev @xmath75 mev depicted quasiclassical evolution period @xmath76 ns early times minimum fisher - shannon product @xmath77 wave - packet recovers quasiclassical behavior relative minimum @xmath78 quantum ring @xmath73 nm @xmath74 mevpresented analysis quantum wavepacket revival phenomena quantum bouncer graphene ring information product shown theoretical tool useful atomic physics molecular reactions solids geophysics describes dynamics wavepackets found revivals fractional revivals wavepackets correspond minima entropic product signaling recovery quasiclassical behavior wavepacket supported spanish projects micinn fis2009 - 08451 fqm-02725 fis2011 - 24149 g. rempe h walther n. klein phys . lett 58 *, 353 ( 1987 ). yeazell mallalieu stroud jr lett 64 * 2007 ( 1990 ) t baumert lett 191 639 ( 1992 ) vrakking villeneuve stolow phys . 54 r37-r40 ( 1996 ) rudenko ergler feuerstein chem . 329 *, 193 ( 2006 ) a. lpez sun j. schliemann phys . 85 * 205428 ( 2012 ) tomasz rusin wlodek zawadzki phys . 78 215419 ( 2008 ) john schliemann . 10 * 043024 ( 2008 )demikhovskii maksimova perov telezhnikov phys . rev. 85 022105 2012 ) romera j dehesa chem . 120 8906 2004 ) nesvizhevsky protasov quantum states neutrons earth gravitational field state art applications perspectives trends quantum gravity research editor david c. moore 65 - 107 nova science publishers 2006 | 0.435433 | 2,673 | 1,382 | , Saving $0.1 in GPT-4. |
154,952 | how to describe a black hole embedded in an expanding universe is an important topic . some models about this issue
are suggested by assuming that the metric is a conformal transformation of the schwarzschild metric or of the isotropic black hole metric .
however , there exists naked singularities in the two metrics .
recently , it is argued that the singularity in the sultana - dyer space - time is covered by an apparent horizon surface .
but we find that such an apparent horizon does not exist if the null energy condition holds .
pacs : 04.70.bw 04.70.-s 95.30.sf * key words : * black hole , frw universe , metric | isolated black holes have been investigated in great depth and detail for more than forty years . on the other hand ,
black holes embedded in the background of the expanding universe are also important and even more realistic situations .
some works on the issue have been carried out . in @xcite , mcvitie found a celebrated space - time describing a black hole embedded in the friedman - robertson - walker universe , which is generalized to the reissner - nordstrm case in @xcite . in @xcite , the model of swiss cheese black holes
is shown . in @xcite ,
the author suggested the vaidya s space - time describing a frw universe with a schwarzschild - like black hole that does not expand with the rest of the universe , which is generalized to the kerr - newman case in @xcite . in @xcite ,
the thakurta s black hole is shown . in @xcite ,
the sultana - dyer black hole metric is suggested .
recently , in @xcite , the authors suggested new solutions describing black holes embedded in the expanding universe .
most of the models with the spherical symmetry on the topic are based on the assumption @xmath0 or @xmath1 where @xmath2 . here and after , we take @xmath3 .
however , the space - time manifold described by the line element of eq.([isotropicfrwbl ] ) is singular at surfaces @xmath4 . and there
also exists a singularity in the spacetime of eq.([frwschwbl ] ) at the surface @xmath5
. then we may ask whether the singularity in the spacetime ( [ isotropicfrwbl ] ) or ( [ frwschwbl ] ) is a spacetime singularity or not ? if it is a proper singularity , is the singularity naked or covered by an apparent horizon ? in ref.@xcite , the authors claimed that , in the case of constant @xmath6 , the singularity is covered by an apparent horizon . however , we find such an apparent horizon does not exist if the null energy condition is satisfied . in the paper ,
firstly , we show that the space - time ( [ isotropicfrwbl ] ) or ( [ frwschwbl ] ) is properly singular at the surface @xmath7 or @xmath8 respectively .
secondly , we show the apparent horizon suggested in ref.@xcite does not exist .
section [ conclusion ] contains conclusions and discussion .
the metric of the mcvittie space - time @xcite is assumed as eq.([isotropicfrwbl ] ) with @xmath9 , where @xmath10 is a constant parameter . in @xcite ,
the metric is assumed as eq.([isotropicfrwbl ] ) , but @xmath11 is taken to be constant . here , we note that , if @xmath12 , the metrics in eq.([frwschwbl ] ) and eq.([isotropicfrwbl ] ) define the same space - time .
generally , for @xmath13 , the two metrics represent different spacetimes . in ref.@xcite ,
the imperfect fluid stress - energy tensor is used to solve the einstein equations . in these models , at the surface @xmath7 , the spactimes are singular .
let us show this by calculating the scalar curvature of the metric in eq.([isotropicfrwbl ] ) , @xmath14 where @xmath15 and @xmath16 and @xmath17 obviously , in the limiting cases corresponding to the standard frw universe , the standard schwarzschild black hole or the schwarzchild - de sitter black hole , the scalar curvature @xmath18 is finite@xcite .
however , in this paper , we exclude these limiting cases . then for @xmath19 , @xmath18 is infinity at the surface @xmath4 . even for @xmath13
, we find that it is almost impossible to find a time - dependent function @xmath11 to avoid the divergence at @xmath7 .
the divergence of @xmath18 at @xmath4 implies the singularity is proper .
the thakurta spacetime @xcite is based on the line element in eq.([frwschwbl ] ) .
the scalar curvature of the metric ( [ frwschwbl ] ) is @xmath20 where @xmath21 still , we exclude the limiting cases corresponding to the standard frw universe , the standard schwarzschild black hole and the schwarzschild - de sitter black hole .
then , similarly to the metric ( [ isotropicfrwbl ] ) , the scalar curvature @xmath22 at the surface @xmath8 is divergent .
so the singularity at @xmath5 is proper .
in@xcite , the authors showed that , in the case of @xmath23=0 , the singularity at the surface @xmath4 or @xmath5 in the space - time ( [ isotropicfrwbl ] ) or ( [ frwschwbl ] ) respectively is covered by an apparent horizon . in this section
we will show such an apparent horizon does not exist for the case of @xmath23=0 if the null energy condition holds .
first , we show the outgoing light emitted from the surface @xmath24 in eq.([isotropicfrwbl ] ) or @xmath25 in eq.([frwschwbl ] ) can reach the future null infinity asymptotically if @xmath19 . considering the world line of a photon propagating along the direction of radius emitted from the point @xmath26 with constant @xmath27 and @xmath28
, we have @xmath29 or @xmath30 for an outgoing photon which arrives @xmath31 at the moment @xmath32 , we obtain respectively @xmath33 or @xmath34 of course , we assum @xmath35 in eq.([timeandradiusi ] ) and @xmath36 in eq.([timeandradiuss ] ) , and @xmath37 in the two equations . generally , since the einstein equations which determine @xmath37 and @xmath11 are unknown , nothing can be concluded from the equation ( [ timeandradiusi ] ) and ( [ timeandradiuss ] ) .
but for the special case of @xmath19 , the integral on the right - hand sides of eq.([timeandradiusi ] ) is finite for finite @xmath38
. then , in the spacetime of eq.([isotropicfrwbl ] ) , the light emitted from the point at the surface @xmath39 will reach the surface @xmath31 in the finite time .
this implies that _ outgoing photons emitted outside the surface @xmath40 can reach the future null infinity asymptotically in the spacetime ( [ isotropicfrwbl ] ) if @xmath19_. in fact , this conclusion can be obtained easily from the conformal diagram of the sultana - dyer space - time , eq.([isotropicfrwbl ] ) with @xmath19 , given in fig.1 of ref.@xcite which is displayed as fig.[sultanadyer ] in our paper ( see ref.@xcite for details ) .
the conformal diagram of the sultana - dyer spacetime .
the horizontal wiggly line at the bottom describes the big bang singularity , the wiggly line at 45 degrees denotes the @xmath41 null singularity , and the solid line at 45 degrees describes future null infinity .
null geodesics end at future null infinity or at the black singularity ( either when it is naked if started early on , or crossing the timelike black hole apparent horizon labelled ah).,width=245 ] for the space - time ( [ frwschwbl ] ) , similar conclusion can be made from eq.([timeandradiuss ] ) . here
, we note that , generally , to obtain our conclusion , we should use the affine parameter @xmath42 in eq.([timeandradiusi ] ) or eq.([timeandradiuss ] ) instead of the coordinate time @xmath43 .
but , fortunately , here the coordinate time @xmath43 is appropriate for our conclusion , since finite @xmath43 always indicates finite @xmath42 along the worldline of an outgoing photon outside the surface @xmath4 or @xmath5 in the spacetime ( [ isotropicfrwbl ] ) or ( [ frwschwbl ] ) respectively .
now let s assume that , outside the surface @xmath4 in the spacetime ( [ isotropicfrwbl ] ) or the surface @xmath5 in the spacetime ( [ frwschwbl ] ) respectively , there exists an apparent horizon defined by the equation @xmath44 , if @xmath19 .
then at the moment @xmath45 , a space - like two surface is defined by @xmath46 , and the expansion of the null generators of the future causal boundary of the two surface , @xmath27 , is zero at the two surface . due to the analysis in the last paragraph
, we know outgoing photons emitted from the surface @xmath46 will reach the future null infinity asymptotically if @xmath19 .
then , naturally , the outgoing future causal boundary of the two surface intersects the future null infinity .
however , on the other hand , it is known that the raychaudhuri equation @xcite implies that the future causal boundary of the two surface @xmath46 can not intersect the future null infinity because @xmath27 is zero at the two surface @xmath46 and is nonincreasing if the null energy condition holds @xmath47 @xmath48 is the energy - momentum tensor and @xmath49 is a null vector .
then we obtain two contrary conclusions from the assumption of the existence of the apparent horizon .
this indicates that _ in the spacetime ( [ isotropicfrwbl ] ) or ( [ frwschwbl ] ) , outside the surface @xmath4 or @xmath5 respectively , there does not exist any apparent horizon and the singularity at @xmath4 or @xmath5 is naked , if the null energy condition holds and @xmath19 .
_ in particular , in the model of the sultana - dyer black hole @xcite , the energy - momentum tensor is @xmath50 where @xmath51 describes an ordinary massive dust and @xmath52 describes a null dust with density @xmath53 and @xmath54 .
obviously , the null energy condition is satisfied for any null vector @xmath49 .
then we conclude that in the space time of the sultana - dyer black hole , there is no apparent horizon outside the surface @xmath4 in eq.([isotropicfrwbl ] ) and the singular surface @xmath4 is naked .
the conclusion is different to that in @xcite .
the issue on the solutions of the einstein equation describing black holes embedded in the frw universe is very important .
some spacetime models on this issue @xcite have been suggested .
most of the models are based on the assumption of eq.([frwschwbl ] ) or eq.([isotropicfrwbl ] ) .
however , there exists singularity in the spacetime of eq.([frwschwbl ] ) or eq.([isotropicfrwbl ] ) .
the singularity can not be eliminated even for the imperfect fluid stress - energy tensor to be used to solve the einstein equations @xcite .
but it is argued that the singularity in the spacetime ( [ isotropicfrwbl ] ) at @xmath4 is gravitationally weak @xcite in the sense that it does not crush extended bodies which falls into it to zero volume @xcite . in this view , the line elements ( [ isotropicfrwbl ] ) and ( [ frwschwbl ] ) may still make some sense in physics . in @xcite , the authors claimed that , for @xmath19 , there exists a regular apparent horizon @xmath55 inside which the surface @xmath4 in eq.([isotropicfrwbl ] ) is contained .
however , our analysis above shows that photons emitted outside the surface @xmath4 can reach the future null infinity if @xmath19 .
so the outgoing future causal boundary of the apparent horizon will intersect the future null infinity .
however , if the null energy condition holds , it is impossible for the future causal boundary of the apparent horizon to intersect the future null infinity because of the non - decreasing expansion of the null generators of the causal boundary indicates by the raychaudhuri equation @xcite .
then such an apparent horizon does not exist , and the singular surface is naked still if @xmath19 .
the conclusion in @xcite may be due to the special coordinate @xmath56 .
actually , the coordinate transformation between @xmath56 and the ones in eq.([isotropicfrwbl ] ) is not diffeomorphic globally , which is singular at the apparent horizon @xmath55 .
our conclusion for the case of @xmath19 can not be applied to that of @xmath13 directly . however , since our analysis is based on two conditions : the null energy condition holds and outgoing photons emitted from the apparent horizon ( if exists ) can reach the future null infinity .
then it is still possible for the existence of the apparent horizon outside the surface @xmath4 or @xmath5 in the space - time ( [ isotropicfrwbl ] ) or ( [ frwschwbl ] ) respectively , if the null energy condition is violated or outgoing photons emitted from the apparent horizon ( if exists ) can not reach the future null infinity .
this work is supported by the natural science foundation of northwest university of china under grant no . | black holes investigated for forty years black holes in expanding universe important realistic works on issue carried out mcvitie found space - time black hole in friedman - robertson - walker universe generalized to reissner - nordstrm case model of swiss cheese black holes shown author suggested vaidya s space - time frw universe with schwarzschild - like black hole expand generalized to kerr - newman case thakurta s black hole shown sultana - dyer black hole metric suggested authors suggested new solutions describing black holes in expanding universe models with spherical symmetry based on @xmath0 or @xmath1 @xmath2 @xmath3 space - time manifold line element eq.([isotropicfrwbl ] singular at surfaces @xmath4 singularity in spacetime of eq.([frwschwbl ] ) at surface @xmath5 singularity in spacetime isotropicfrwbl or frwschwbl spacetime singularity naked or covered by apparent horizon ? authors claimed constant @xmath6 singularity covered by apparent horizonapparent horizon exist if null energy condition satisfied paper show space - time isotropicfrwbl or frwschwbl singular at surface @xmath7 or @xmath8 apparent horizon in ref.@xcite exist section conclusion ] contains conclusions discussion metric mcvittie space - time @xcite assumed as eq.([isotropicfrwbl ] with @xmath9 @xmath10 constant in @xcite assumed as eq.([isotropicfrwbl @xmath11 constant @xmath12 metrics in eq.([frwschwbl and([isotropicfrwbl define same space - time for @xmath13 two metrics represent different spacetimes in ref.@xcite imperfect fluid stress - energy tensor used solve einstein equations at surface @xmath7 spactimes singular show calculating scalar curvature of metric in eq.([isotropicfrwbl ] @xmath14 @xmath15 limiting cases standard scalar curvature @xmath18 finite@xcite exclude limiting cases @xmath19, @xmath18 is infinity at surface @xmath4@xmath13 almost impossible find time - dependent function @xmath11 avoid divergence at @xmath7 divergence @xmath18 at @xmath4 implies singularity proper thakurta spacetime @xcite based on line element eq.([frwschwbl ] scalar curvature metric frwschwbl @xmath20 @xmath21 exclude limiting cases standard frw universe schwarzschild black hole schwarzschild - de black hole scalar curvature @xmath22 surface @xmath8 divergent singularity at @xmath5 proper authors @xmath23=0 singularity at surface @xmath4 or @xmath5 space - time covered by apparent horizon apparent horizon exist @xmath23=0 if null energy condition holds outgoing light surface @xmath24 or @xmath25 reach future null infinity asymptotically @xmath19 world line photon propagating radius point @xmath26 constant @xmath27 and @xmath28 @xmath29 or @xmath30 outgoing photon @xmath31 @xmath32 obtain @xmath33 or @xmath34 assum @xmath35 in eq.[timeandradiusi ] @xmath36 in eq.([timeandradiuss @xmath37 in two equations einstein equations @xmath37 @xmath11 unknown nothing concluded from equation timeandradiusi ] timeandradiuss ] special case @xmath19 integral right - hand sides of eq.[timeandradiusi finite for @xmath38 spacetime eq.[isotropicfrwbl light from surface @xmath39 reach surface @xmath31 in finite time implies outgoing photons outside surface @xmath40 reach future null infinity in spacetime isotropicfrwbl @xmath19 conclusion from conformal diagram of sultana - dyer space - time eq.([isotropicfrwbl ] with @xmath19.1 of ref.@xcite horizontal wiggly line bottom big bang singularity line at 45 degrees @xmath41 null singularity solid line at 45 degrees future null infinity null geodesics end at future null infinity or black singularity crossing timelike black hole horizon space - time frwschwbl similar conclusion from eq.([timeandradiuss ] )conclusion use affine parameter @xmath42 in eq.([timeandradiusi or instead of coordinate time @xmath43 coordinate time @xmath43 appropriate for conclusion finite @xmath43 indicates finite @xmath42 worldline outgoing photon outside surface @xmath4 or @xmath5 in assume outside surface @xmath4 or exists apparent horizon defined by equation @xmath44 if @xmath19 @xmath45 space - like two surface defined by @xmath46 expansion of null generators future causal boundary @xmath27 is zero at two surface outgoing photons from surface @xmath46 reach future null infinity if @xmath19 outgoing future causal boundary intersects future null infinity raychaudhuri equation @xcite implies future causal boundary @xmath46 can not intersect future null infinity because @xmath27 is zero at surface @xmath46 nonincreasing if null energy condition holds @xmath47 @xmath48 is energy - momentum tensor and @xmath49 is null vector obtain two contrary conclusions from assumption existence apparent horizonindicates spacetime isotropicfrwbl or frwschwbl outside surface @xmath4 or @xmath5 apparent horizon singularity at @xmath4 or naked if null energy condition holds @xmath19 model sultana - dyer black hole @xcite energy - momentum tensor @xmath50 @xmath51 describes ordinary massive dust @xmath52 null dust with density @xmath53 and @xmath54 null energy condition satisfied for null vector @xmath49 conclude space time sultana - dyer black hole no apparent horizon outside surface @xmath4 singular surface naked conclusion different in @xcite issue solutions einstein equation describing black holes frw universe important spacetime models @xcite suggested based on assumption of eq.[frwschwbl or[isotropicfrwbl exists singularity in spacetime singularity can not be eliminated for imperfect fluid stress - energy tensor argued singularity in spacetime at @xmath4 gravitationally weak @xcite does not crush extended bodies zero volumeline elements isotropicfrwbl frwschwbl ] may make sense in physics in @xcite authors claimed for @xmath19 regular apparent horizon @xmath55 surface @xmath4[isotropicfrwbl contained analysis shows photons outside surface @xmath4 can reach future null infinity if @xmath19 future causal boundary intersect future null infinity if null energy condition holds impossible for future causal boundary to intersect future null infinity non - decreasing expansion of null generators apparent horizon exist singular surface naked if @xmath19 conclusion due to special coordinate @xmath56 coordinate transformation between @xmath56[isotropicfrwbl not diffeomorphic globally singular at horizon @xmath55 conclusion for @xmath19 not applied to @xmath13 analysis based on two conditions null energy condition holds outgoing photons horizon can reach future null infinity possible existence apparent horizon outside surface @xmath4 or @xmath5 in space - time isotropicfrwbl or frwschwbl ] if null energy condition violated or outgoing photons horizon reach future null infinitysupported natural science foundation northwest university china grant | 0.49805 | 2,957 | 1,555 | , Saving $0.1 in GPT-4. |
154,955 | the article describes the identification of hadronically decaying @xmath0 leptons in @xmath1 collisions at @xmath2 tev collected by the d detector at the fermilab tevatron . after a brief description of the motivations and the challenges of considering @xmath0 leptons in high energy hadronic collisions , details of the @xmath0 reconstruction and identification
will be discussed .
the challenges associated for @xmath0 energy measurements in an hadronic environment will be presented including approaches to deal with such measurements . | leptons are of particulary interest in high energy hadronic collisions because they provide clean event signatures in a complex hadronic environment .
electrons and muons are often considered because of such detector signatures .
in addition , @xmath0 leptons ( called simply @xmath0 here after ) brings a potential acceptance gain of @xmath3 @xmath4 according to the lepton multiplicity of the analyzed final state .
moreover , studies with @xmath0 leptons provide an interesting area to test the consistency of the standard model ( sm ) through lepton universality at high energies by measuring the branching ratio of electroweak ( ew ) bosons decaying into @xmath0 leptons @xcite .
one promising candidate for extension to the sm is the minimal supersymmetric standard model ( mssm ) which predicts new particles that can decay into @xmath0 leptons .
the existence of such scenario can be probed within @xmath0 final states @xcite .
finally , many decay chains initiated by the higgs boson involve @xmath0 leptons and considering them can then increase the experimental sensitivity to understand the origin of electroweak symmetry breaking ( ewsb ) @xcite .
considering @xmath0 leptons thus appears to encompass many virtues .
however , understanding such an object in a hadronic environment is very challenging .
the neutrino(s ) from @xmath0 final states escape the detector without interacting and forces a fraction of the @xmath0 energy to be invisible .
hence , the resulting visible energy is relatively soft and includes contamination from soft qcd interactions , which becomes a primary component of the hadronic final state .
further , the various decay mode of @xmath0 must be considered . on the one hand
, the identification of @xmath0 from its leptonic decay ( @xmath5 ) suffers from electrons and muons originating from direct ew bosons decay as well as a poor statistic due to @xmath6 . on the other hand , @xmath0 identification from its
hadronic decays ( @xmath7 ) have different detector signature according to the hadronic resonance involved .
moreover , direct qcd interactions from hadrons collisions produce a lot of hadronic final states which can mimic the hadronically decaying @xmath0 . to be optimal ,
each of the different hadronic @xmath0 final states requires dedicated analysis and a combination is performed to extract the full information .
the complexity of hadronic @xmath0 final states require sophisticated algorithms based on multivariate technics applied after event reconstruction .
the next section describes the @xmath0 reconstruction at the d experiment .
the subsequent section discusses multivariate analysis ( mva ) discriminant , which allows separation of @xmath0 from jet fakes .
in this section , we describe how the @xmath0 candidate is built from elementary reconstructed objects within the calorimeter and tracking systems @xcite .
three objects are potentially used to build reconstructed @xmath0 object : 1 .
calorimeter cluster found by a simple cone algorithm in a @xmath8 cone .
electromagnetic subcluster found using a nearest neighbour algorithm with a seed in the third layer of the electromagnetic calorimeter ( i.e. the layer containing the finest segmentation ) .
the clustered energy is required be greater than @xmath9 mev .
all the tracks found by the d tracking algorithm in a @xmath10 cone around the calorimeter cluster are considered .
an invariant mass requirement is applied to keep only tracks compatible with the @xmath0 decay . in the next section
we discuss the use of these objects to optimize the @xmath0 reconstruction according to its hadronic decay .
the d experiment bases its @xmath0 lepton reconstruction on specific signatures of the hadronic system present in the @xmath0 final state .
three types of candidate are defined and depends on the reconstructed objects signature .
there are categorized as : * type @xmath11 : one calorimeter cluster and one track ( optimized for @xmath12 decay ) ; * type @xmath13 : one calorimeter cluster , one track and one calorimeter em subcluster ( optimized for @xmath14 decay ) ; * type @xmath15 : one calorimeter cluster , at least two tracks ( optimized for @xmath15-prong @xmath16 decay ) .
figure [ fig : recoeff ] shows the reconstruction algorithm for type @xmath13 @xmath0 .
one can see that almost all real @xmath0 above @xmath17 gev are selected by the d reconstruction but a large fraction of jets ( or fake @xmath0 ) remains .
indeed , jets from direct strong interaction can easily mimic hadronically decaying @xmath0 lepton and therefore , sophisticated methods is needed to separate @xmath0 from jets .
this is the subject of the next section .
hadronic final states of @xmath0 leptons can be faked by jets arising from direct qcd interactions .
however , jets signatures from @xmath0 are narrower and have lower tracks multiplicity .
these properties can be used to reject qcd jets and thereby increase the @xmath0 identification efficiency .
in fact , other specificities of hadronic @xmath0 decays can also be exploited and the full information can be combined in a neural network ( nn ) @xcite . in particular , a total of @xmath18 observables ( depending on the @xmath0 type ) are considered based on a)the isolation in the tracking system as well as in the calorimeter b)shower shape and composition c)correlations between the tracking system and the calorimeter .
figure [ fig : varnn ] provides an example of two such nn input variables . the nn output peaks at @xmath19 for the qcd jets and at @xmath11 for the @xmath0 jets as shown in figure [ fig : nnoutput ] .
the @xmath20 electroweak process is clearly sizeable and data ( black dots ) are well understood by the sm prediction ( filled histograms ) . although @xmath0 reconstruction algorithm is efficient , the signal over background ratio @xmath21 is poor .
in fact , the order of magnitude of @xmath22 is @xmath13 at the reconstruction level and reaches almost @xmath23 after the nn selection .
table [ table:1 ] summarizes efficiencies for each @xmath0 types after reconstruction and subsequent nn identification . [ cols="<,<,<,<,<",options="header " , ] + the optimization strategy
is based on how the nn treats the information provided by the physics . by denoting @xmath24 as a point in the discriminating variables space
, the nn output @xmath25 converges to the true likelihood function ( which is the best classifying function ) : @xmath26 here , @xmath27 and @xmath28 are the probability density function of the genuine @xmath0 ( signal ) and qcd jets ( background ) respectively in the discriminating observables space .
one can find two strategies to optimize the separation between jets and @xmath0 .
a first approach is to improve @xmath29 by adding more observables based on physical properties of @xmath0 leptons .
a separate method would be to minimize @xmath30 by performing the nn training in specific phase space regions or by increasing the statistics of the training samples to describe the differences in more details . in practice ,
the first approach was tested by trying to include information from the d preshower system @xcite which has a better segmentation than the calorimeter and allows an improved separation of the @xmath31 from the @xmath32 ( for type @xmath13 decay ) .
after several studies , it has been shown that this approach does not significantly help to separate @xmath0 from qcd jets .
new physical information can be used such as the long lifetime of the @xmath0 lepton which can lead to displaced tracks from the primary vertex similar to the case of b - jets .
figure [ fig : nnbtagvar ] shows a new discriminating variable based on the impact parameter of each track for type @xmath15 @xmath0 candidates .
this new observable increases the @xmath0 identification efficiency of about @xmath33 for the same jet rejection .
a second approach consisting in helping the nn convergence was applied by using larger training samples and tuning specific nn parameters such as the numbers of nodes , epochs and minimization algorithm @xcite .
moreover , the difference between @xmath0 and qcd jets can evolve with the candidate transverse energy . in order to help the nn to exploit this behaviour ,
dedicated trainings were done for events with high ( @xmath34 gev ) and low ( @xmath35 gev ) @xmath36 @xmath0 candidate .
overall , the @xmath0/jet discrimination is improved by @xmath37 at low @xmath36 and @xmath38 at high @xmath36 as shown in figure [ fig : nnptdep ] .
in order to perform physical measurements in @xmath0 lepton final states , the identification of @xmath0 leptons is necessary but not sufficient .
one needs to have a good understanding of @xmath0 kinematic properties .
this section deals with the @xmath0 energy measurement at the d experiment . the object energy calibration at collider experiments is usually performed with the help of well known physical processes . at high energies ,
the production and decay of @xmath39 bosons allow to perform such calibrations .
two important challenges arise for the hadronically decaying @xmath0 .
firstly , the @xmath20 visible mass events is defined by @xmath40 , where @xmath41 is the transverse missing energy .
] peak is broad and shifted by the neutrinos present in the @xmath0 decay which makes this observable less sensitive than for the @xmath42 decay .
secondly , the various @xmath0 decay modes implie lower statistics compared with the @xmath42 decay ( @xmath43 events versus @xmath44 events ) .
these situations require to find another strategy to calibrate the visible energy of hadronic @xmath0 lepton .
if one considers a type @xmath13 candidate having visible energy from @xmath45 , the true energy can be obtained from the measured energy using the hadronic calorimeter response ( @xmath46 ) and the electromagnetic calorimeter response @xmath47 : @xmath48 however , the d calorimeter is not compensated ( @xmath49 ) and its segmentation does not allow to separate the @xmath32 shower from the @xmath50 one .
thus , the only measurable energy is @xmath51 . in order to deal with these experimental constraints
, one must use the track energy as reference and propagate it to the calorimeter in two ways : the absolute correction and the relative correction .
both methods are described below .
the idea of the absolute correction is to use the tracker to measure the @xmath32 energy . in order to avoid double counting ,
the average of the energy deposited by charged pion in the calorimeter is subtracted : @xmath52 figure [ fig : energycorrabs ] shows that method brings a better measurement of the @xmath0 visible energy .
another strategy to propagate the track energy is to correct the simulation event - by - event using the @xmath53(@xmath54 ) distribution for @xmath20 events selected in data : @xmath55 here , @xmath56 is the average value over the events .
such a method is relative due to the fact that the simulation is adjusted to the data but the true energy is not known either in the data or in the simulation .
however , the @xmath46 measurement is not needed for this approach .
in high energy hadronic collisions , @xmath0 leptons require more sophisticated tools than electrons and muons because of their various hadronic decay modes and the escaping neutrino .
such @xmath0 objects are of particular interest to probe the standard model as well as physics beyond the sm including many areas such as the electroweak symmetry breaking origin or supersymmetric scenarios . in spite of the various experimental challenges ,
the d experiment has developed an effective algorithm to identify hadronically decaying @xmath0 leptons and has performed several physics measurements and searches employing such technics .
in addition , some recent progress has been achieved on the @xmath0 identification improving the potential sensitivity of the experiment with @xmath0 leptons .
9 d collab .
plb 670 , 292 ( 2009 ) .
d collab .
plb 680 , 24 - 33 ( 2009 ) .
d collab .
plb 682 , 278 - 286 ( 2009 ) .
d collab .
prl 102 , 251801 ( 2009 ) .
d collab .
nucl instrum . method .
a565 : 463 - 537(2006 ) christopher m. bishop , _ neural networks for pattern recognition _ , oxford university press ( 1995 ) .
galea , phd thesis , fermilab - thesis-2008 - 22 . | leptons in high energy hadronic collisions provide clean event signatures in complex hadronic environment electrons muons considered detector signatures @xmath0 leptons potential acceptance gain of @xmath3 @xmath4 lepton multiplicity studies with leptons test consistency standard model lepton universality at high energies branching ratio of electroweak bosons decaying into leptons candidate for extension minimal supersymmetric standard model ) predicts new particles decay into @xmath0 leptons existence probed within @xmath0 final states @xcite decay chains by higgs boson involve leptons experimental sensitivity origin of electroweak symmetry breaking @xcite leptons virtues understanding in hadronic environment challenging neutrino(s from states escape detector fraction @xmath0 energy invisible resulting visible energy soft includes contamination from soft qcd interactions primary component of hadronic final state decay mode of @xmath0 considered identification of @xmath0 from leptonic decay suffers from electrons muons from direct ew bosons decay poor statistic due to @xmath6@xmath0 from hadronic decays different detector signature hadronic resonance direct qcd interactions from hadrons collisions produce hadronic final states mimic hadronically decaying @xmath0 hadronic @xmath0 final states requires analysis combination to extract full information complexity of hadronic @xmath0 final states require sophisticated algorithms multivariate technics reconstruction next section describes @xmath0 reconstruction at d experiment multivariate analysis ) discriminant separation @xmath0 from jet fakes @xmath0 candidate built from reconstructed objects calorimeter tracking systems three objects used @xmath0 calorimeter cluster electromagnetic subcluster in third layer calorimeter clustered energy greater than @xmath9 mev tracks by tracking algorithm in @xmath10 cone around calorimeter considered invariant mass requirement tracks compatible with @xmath0 decay use objects to optimize @xmath0 reconstruction hadronic decay d experiment bases @xmath0 lepton reconstruction on signatures hadronic system in @xmath0 final state three types of candidate defined on reconstructed objects signaturecategorized @xmath11 one calorimeter cluster one track for @xmath12 decay @xmath13 one calorimeter cluster one track subcluster for @xmath14 decay @xmath15 one calorimeter cluster two tracks for @xmath15 @xmath16 decay figure shows reconstruction algorithm @xmath13 @xmath0 real @xmath0 above @xmath17 gev selected large jets fake @xmath0 remains jets from interaction mimic hadronically decaying @xmath0 lepton sophisticated methods needed to separate @xmath0 from jets hadronic final states @xmath0 leptons faked by jets direct qcd interactions jets signatures from @xmath0 narrower lower tracks multiplicity reject qcd jets increase @xmath0 identification efficiency other specificities of hadronic @xmath0 decays exploited full information combined in neural network @xcite @xmath18 observables @xmath0 type considered based on isolation tracking system calorimeter shape composition)correlations between tracking system calorimeter figure nn input variables nn output peaks at @xmath19 qcd jets @xmath11 @xmath0 jets@xmath20 electroweak process sizeable data black dots understood by sm prediction histograms @xmath0 reconstruction algorithm efficient signal over background ratio @xmath21 poor order magnitude of @xmath22 is @xmath13 at reconstruction level reaches almost @xmath23 after nn selection table summarizes efficiencies @xmath0 types after reconstruction nn identification optimization strategy based on nn information @xmath24 point in discriminating variables space nn output @xmath25 converges to likelihood function @xmath26 @xmath27 @xmath28 probability density function of @xmath0 signal qcd jets background in discriminating observables space two strategies to optimize separation between jets @xmath0 first improve @xmath29 adding more observables properties @xmath0 leptons method minimize @xmath30 nn training in specific phase space regions or increasing statistics training samples first approach tested information from d preshower system @xcite better segmentation improved separation @xmath31 from @xmath32 approach help separate @xmath0 from qcd jetsphysical information long lifetime @xmath0 lepton displaced tracks vertex b - jets figure shows new discriminating variable impact parameter track for @xmath15 @xmath0 candidates increases @xmath0 identification efficiency @xmath33 for same jet rejection second approach nn convergence larger training samples tuning nn parameters epochs minimization algorithm difference between @xmath0 and qcd jets with candidate transverse energy dedicated trainings for events with high low @xmath36 @xmath0 candidate @xmath0/jet discrimination improved by @xmath37 at low @xmath36 @xmath38 at high physical measurements in @xmath0 lepton final states identification @xmath0 leptons necessary not sufficient understanding @xmath0 kinematic properties @xmath0 energy measurement at d experiment object energy calibration at collider experiments physical processes high energies production decay of @xmath39 bosons allow calibrations challenges for hadronically decaying @xmath0 @xmath20 visible mass events defined by @xmath40 @xmath41 transverse missing energypeak broad shifted by neutrinos in @xmath0 decay less sensitive than @xmath42 decay @xmath0 decay modes implie lower statistics @xmath42 decay @xmath43 versus @xmath44 events require strategy to calibrate visible energy of hadronic @xmath0 lepton type @xmath13 candidate visible energy from @xmath45 true energy obtained from measured energy using hadronic calorimeter response @xmath46 ) electromagnetic calorimeter response @xmath47 : @xmath48 d calorimeter not compensated @xmath49 ) segmentation separate @xmath32 shower from @xmath50 one only measurable energy is @xmath51 use track energy propagate to calorimeter absolute correction relative correction absolute correction tracker measure @xmath32 energy average energy charged pion in calorimeter subtracted @xmath52 figure better measurement of @xmath0 visible energy strategy correct simulation event - by - event using @xmath53(@xmath54 ) distribution for @xmath20 events @xmath56 average value over events method relative simulation adjusted to data true energy not known @xmath46 measurement not neededhigh energy hadronic collisions @xmath0 leptons require sophisticated tools electrons muons hadronic decay modes escaping neutrino @xmath0 objects standard model physics beyond sm electroweak symmetry breaking supersymmetric scenarios d experiment developed algorithm identify decaying leptons performed physics measurements searches progress @xmath0 identification improving sensitivity experiment plb 670 , 292 2009 plb 680 , 24 - 33 2009 plb 682 , 278 - 286 2009 prl 102 , 251801 2009 nucl instrum method a565 : 463 - 537(2006 ) christopher m. bishop neural networks pattern recognition oxford university press 1995 phd thesis fermilab - thesis-2008 - 22 | 0.435093 | 2,826 | 1,463 | , Saving $0.1 in GPT-4. |
154,964 | we consider the cmb bispectrum signal induced by structure formation through the correlation between the integrated sachs - wolfe and the weak lensing effect .
we investigate how the bispectrum knowledge can improve our knowledge of the most important cosmological parameters , focusing on the dark energy ones .
the bispectrum signal arises at intermediate redshifts , being null at present and infinity , and is characterized by a large scale regime ( dominated by linear dynamics of cosmological perturbation ) and a small scale one ( dominated by density perturbations in a non - linear regime ) ; on the other hand , the effect induced by dark energy on the power spectrum is mostly geometrical and imprinted at redshift close to the present . because of this , the knowledge of power spectrum and bispectrum yield two complementary informations at very different cosmological epochs , particularly suitable to extract informations about the onset of the cosmic acceleration and dark energy properties that provide it . in order to quantify
how much the bispectrum can help the power spectrum in constraining the dark energy parameters , we choose a fiducial model on a three - dimensional space including the following dark energy parameters : dark energy density @xmath0 ; dark energy equation of state today @xmath1 and dark energy equation of state in the past @xmath2 ( @xmath3 is related to the first derivative of equation of state ) .
then we simulate a likelihood analysis showing how contour levels become narrower when bispectrum is included .
preliminary results suggest a consistent improvement on the estimation of dark energy abundance and on dynamical properties of the equation of state .
this indicates that the knowledge of the bispectrum in future high resolution and high sensitivity cmb observations could yield a substantial improvement with respect to the traditional analysis based on the power spectrum only . | the combination of several independent cosmological datasets , namely type ia supernovae ( @xcite ) , cmb ( @xcite ) and large scale structure ( see e.g. @xcite ) indicate that the universe is presently accelerating .
the dark energy responsible for the acceleration appears to be a fraction of about 73% of the cosmic energy density today .
the main properties of the dark energy are described in terms of its equation of state @xmath4 , i.e. the ratio between pressure and energy density ; its value , for a pure cosmological constant , is -1 .
a current experiments indicate that the present value of @xmath5 should be in the range @xmath6 ( @xcite ) .
the cosmological constant as an explanation to cosmic acceleration has two well - known problems : the coincidence problem ( why cosmological constant density and matter density are comparable today ? ) and the fine tuning problem ( i.e. why cosmological constant is 123 orders of magnitude less than planck scale ? ) . to solve the latter problem , a dynamical scalar field , known as quintessence ( @xcite ) , has been introduced as a minimal extension of the cosmological constant ; the dark energy equation of state gets dynamical alleviating the fine tuning problem . in most models
the dark energy equation of state can be easily parameterized with only two parameters : @xmath7 and its first derivative with respect to the scale factor ( @xcite ) .
the next challenge in cosmology is to constrain the time evolution of @xmath5 ; this can be done with future snia observations and future cmb experiments like planck ( @xcite ) . as we shall see in the next section ,
the cmb power spectrum alone is limited to constrain dark energy .
here we study the improvement which might be achieved by taking into account the non - gaussian distortion induced by the correlation between weak lensing and integrated sachs - wolfe effect ( isw ) .
such signal is suitable studied in the higher order statistic of cmb anisotropies .
we choose here the cmb bispectrum as an estimator of third order statistics ( see e. g. @xcite ) .
the weak lensing effect on cmb anisotropies has been studied ( see e. g. @xcite and references therein ) and the third order statistics , the bispectrum , has been used to constrain the effective dark energy equation of state by @xcite .
the main problem of the cmb power spectrum in studying the dark energy is the degeneracy that affects the distance of last scattering surface with respect the dark energy abundance and , in particular , its equation of state .
its variation produces a change to the distance at last scattering surface ( see e. g. @xcite ) ; unfortunately such distance is degenerate with respect the main dark energy parameters .
this can be easily understood writing the formula of the comoving distance to the last scattering surface ( we restrict our analysis to the flat case and we neglect the radiation contribution ) : @xmath8 in the previous equation @xmath9 is the speed of light , @xmath10 is the hubble constant today , @xmath11 is the redshift of last scattering surface , @xmath12 is the matter density today , @xmath13 ( because we consider only the flat case ) is the dark energy density and @xmath14 depends on the equation of state of dark energy @xmath15 and is defined as @xmath16 analyzing eq .
( [ e : dist ] ) and ( [ e : fz ] ) , we can see that the time dependence on equation of state is washed out by two redshift integrations . in most models
the dark energy equation of state can be parameterized with the following relation ( @xcite ) @xmath17 where @xmath1 and @xmath2 are respectively its present and the asymptotic values .
the difference @xmath18 represents the time - variation of dark energy equation of state .
several different combinations of these dark energy parameters can produce the same comoving distance to the last scattering surface ; for example a comoving distance of about 13900 mpc can be obtained from these three sets of dark energy parameters : @xmath19 ; @xmath20 and @xmath21 .
the net effect of this distance degeneracy is reflected in the cmb power spectrum ; different cosmological models with the same @xmath22 , will produce very similar power spectra ( see figure [ f : cmb ] , right panel ) .
together with degeneracy in the projection effect , there is another physical motivation for which the cmb power spectrum is limited in its capability to constrain the dark energy : the power spectrum is mostly injected at decoupling , and at that time the dark energy density was negligible with respect to the matter density and could nt produce remarkable signatures in the cmb . a way to include the cmb sensitivity on the dark energy is to consider the signatures in the cmb produced by the correlation between the integrated sachs - wolfe effect and the weak lensing on the cmb .
the isw takes into account the rees - sciama effect that arises when the non linear growth of structures is included .
the isw effect affects the cmb photons with a reddening ; the photon acquires a blueshift when it falls down into the potential well of growing structures and it acquires a redshift when it climbs out , but these two contributions are not balanced because the perturbations change in time .
the gravitational lensing effect is the well - known deflection of light due to the gravitational potential of clump of matter .
both effects arise at the epoch of structure formation , thus we can try to exploit them to alleviate the distance degeneracy . the correlation between isw and weak lensing
induce a non vanishing power in the cmb higher order statistics , which we study it considering the cmb bispectrum .
the latter is the harmonic transform of three point correlation function and it is null , within cosmic variance , if the cmb anisotropies are gaussian . since the isw and the weak lensing
are produced by the same physical entity ( growing perturbations in the matter distribution ) , these secondary anisotropies are correlated and induce a non - vanishing bispectrum with exceeds the cosmic variance .
when we consider the isw and weak lensing effect , the cmb anisotropies in a direction @xmath23 in the sky can be decomposed as @xmath24 where @xmath25 includes the primordial and isw anisotropy contributions , and the last term is the contribution from the weak lensing re - mapping ; @xmath26 is the deflection angle .
following @xcite and expanding eq .
( [ e : dt ] ) in spherical harmonics , from the general bispectrum definition @xmath27 , we can build the quantity @xmath28 where @xmath29 indicates the permutations over the three multipoles , the parenthesis are the wigner s 3j symbols , @xmath30 is the primordial cmb power spectrum , and @xmath31_{k=\frac{l}{r(z)}}\ , \ ] ] is the correlation between lensing and isw . in eq .
( [ e : ql ] ) @xmath32 is the gravitational potential power spectrum ; to evaluate the non linear contribution to density power spectrum we have used the existing semi - analytical approach ( @xcite ) . the asymptotic redshift behavior of the integrand of eq .
( [ e : ql ] ) , for @xmath33 and @xmath34 , is vanishing ( @xcite ) ; in fact , fixing the multipole , at low redshift the gravitational potential power spectrum probes infinite wavenumbers where the power is vanishing , while at high redshift the gravitational potential is constant since the universe approach the standard cdm .
therefore , the bispectrum signal is acquired at intermediate redshifts only , and is expected to reflect the cosmological expansion rate at that epoch .
the dark energy domination occurs approximately at the same time , and therefore the bispectrum can be used as a tool to investigate the dark energy properties and to alleviate the distance degeneracy ( @xcite ) .
this can be seen clearly in figure [ f : cmb ] where we compare the cmb power spectrum with the equilateral bispectrum @xmath35 for three dark energy models with the same comoving distance at last scattering surface : the degeneracy is removed at the bispectrum level ; we describe in the next section the shape of bispectrum curves .
the bispectrum has some peculiar features that can be used to discriminate between different cosmological models . plotting the absolute value of bispectrum we have a _ cusp _
, for which the bispectrum is vanishing , and its position depends on cosmological model ( see right panel in figure [ f : cmb ] ) .
the position of this zero - crossing point is the value of the multipole @xmath36 for which the integrand of @xmath37 is null as shown @xcite : the positive contribution ( due to linear growth ) balances exactly the negative one ( due to non linear growth ) .
another relevant feature that must be taken into account is the amplitude of the bispectrum in the linear part ( @xmath38 ) : different cosmological models produce different amplitudes in the linear regime .
the linear power affect @xmath36 : the higher is the power in the linear regime ( @xmath38 ) , the higher is the value of @xmath36 .
.position and shift of multipole @xmath36 of bispectrum zero - crossing varying the main cosmological parameters . [ cols="^,^,^,^",options="header " , ] in table [ t : shift ] we analyze the variation of @xmath36 with respect to the main cosmological parameters : dark energy equation of state today @xmath1 , dark energy density @xmath0 , baryons energy density @xmath39 , hubble constant @xmath40 and spectral index of primordial fluctuations @xmath41 .
the values of @xmath36 and its shift are evaluated when we vary each parameter , fixing the others to a reference model . at first sight @xmath36
is more sensitive to parameters which affect geometrically distances @xmath42 , with the exception of @xmath1 at least in the interval considered .
unfortunately the bispectrum is much more noisy than the cmb power spectrum already at the level of cosmic variance , since it is a higher order effect .
we can increase the signal to noise ratio including all triangles configurations in @xmath43-space ( i.e. considering all possible multipoles triplets ) in our analysis .
we have tested the improvement that the bispectrum can bring to the power spectrum alone building a three - dimensional likelihood on dark energy parameters @xmath44 and choosing a fiducial model for an ideal ( cosmic variance limited ) experiment . in figure
[ f : like ] we show our results marginalizing each time over one dark energy parameter ; as we can see in all cases the contours of the joint likelihoods are narrower than the contours of the power spectrum likelihoods alone . this first , and very preliminary result ,
is encouraging : a gain is expected on the estimation of dark energy equation of state at the beginning of structure formation .
we have discussed how to alleviate the distance degeneracy in the cmb power spectrum using the isw and weak lensing induced cmb bispectrum ; we have shown some preliminary results about the sensitivity of the bispectrum with respect to the main cosmological parameters .
we have simulated a joint likelihood of power spectrum and bispectrum , limited to only dark energy parameters and fixing a fiducial cosmological model .
our preliminary results indicate that adding the bispectrum likelihood to the power spectrum one , the contour levels are narrower and a gain in the estimation of dynamics of dark energy equation of state is expected .
further study is needed to assess the magnitude of this improvement .
future precision cosmology data from high resolution experiments like planck and future measures of tridimensional matter power spectrum with ly-@xmath45 forest ( @xcite ) and cosmic shear ( @xcite ) will help in the use of the bispectrum since they will allow on a better knowledge of the matter power spectrum affecting the bispectrum signal . | combination of independent cosmological datasets type ia supernovae cmb large scale structure indicate universe accelerating dark energy responsible for acceleration fraction of 73% cosmic energy density main properties dark energy described equation of state @xmath4 ratio between pressure and energy density value for is -1 experiments indicate present value @xmath5 should be in range @xmath6 ( ) cosmological constant acceleration has two problems coincidence problem fine tuning problem 123 orders less than planck scale dynamical scalar field quintessence ( ) introduced as minimal extension of cosmological constant dark energy equation of state gets dynamical fine tuning problem equation parameterized with two parameters : @xmath7 and first derivative scale factor ( @xcite ) next challenge to constrain time evolution of @xmath5 with future snia observations cmb experiments like planck cmb power spectrum alone limited to constrain dark energy study improvement non - gaussian distortion correlation between weak lensing integrated sachs - wolfe effect signal studied in higher order statistic of cmb anisotropies choose cmb bispectrum estimator third order statistics @xcite weak lensing effect on cmb anisotropies studied @xcite third order statistics bispectrum used constrain dark energy equation of state @xcite main problem cmb power spectrum dark energy degeneracy distance last scattering surface dark energy abundance equation of state variation change distance last scattering surface distance degenerate main dark energy parameters understood writing formula comoving distance to last scattering surface flat case radiation contribution : @xmath8 @xmath9 speed of light @xmath10 hubble constant @xmath11 redshift of last scattering surface @xmath12 matter density @xmath13 dark energy density @xmath14 depends on equation of state dark energy @xmath15 @xmath16 time dependence on equation of state washed out by two redshift integrations dark energy equation of state parameterized with relation ( @xcite ) @xmath17 @xmath1 @xmath2 are present asymptotic values difference @xmath18 represents time - variation of dark energy equation of statecombinations dark energy parameters produce same comoving distance to last scattering surface distance 13900 mpc from three sets dark energy parameters @xmath19 effect distance degeneracy reflected in cmb power spectrum cosmological models with same @xmath22 produce similar power spectra see figure f : cmb right panel degeneracy in projection effect cmb power spectrum limited constrain dark energy power spectrum injected at decoupling dark energy density negligible produce remarkable signatures in cmb cmb sensitivity on dark energy consider signatures correlation between integrated sachs - wolfe effect weak lensing on cmb isw rees - sciama effect non linear growth of structures affects cmb photons with reddening acquires blueshift redshift climbs out not balanced perturbations change in time gravitational lensing effect deflection of light due to gravitational potential matter both effects arise at epoch structure formation alleviate distance degeneracy correlation between isw weak lensing induce non vanishing power in cmb higher order statistics latter is harmonic transform of three point correlation function null within cosmic variance if cmb anisotropies are gaussian isw weak lensing produced same entity growing perturbations matter distribution secondary anisotropies correlated induce non - vanishing bispectrum cosmic variance isw weak lensing effect cmb anisotropies direction @xmath23 decomposed as @xmath24 @xmath25 includes primordial isw anisotropy contributions last term contribution weak lensing re - mapping @xmath26 deflection angle following @xcite expanding eq spherical harmonics from bispectrum definition @xmath27 build quantity @xmath28 @xmath29 permutations over three multipoles parenthesis wigner s 3j symbols @xmath30 primordial cmb power spectrum @xmath31 correlation between lensing isw @xmath32 gravitational potential power spectrum evaluate non linear contribution to density power spectrum used semi - approach asymptotic redshift behavior of @xmath33 @xmath34 vanishing low redshift gravitational potential power spectrum probes infinite wavenumbers power vanishing high redshift gravitational potential constant standard cdm bispectrum signal acquired at intermediate redshifts reflect cosmological expansion ratedark energy domination occurs same time bispectrum used dark energy properties alleviate distance degeneracy figure f : cmb cmb power spectrum with equilateral bispectrum @xmath35 for three dark energy models same distance degeneracy removed at bispectrum level describe shape of bispectrum curves bispectrum peculiar features discriminate cosmological models plotting absolute value bispectrum _ cusp _ bispectrum vanishing position depends on cosmological model figure position of zero - crossing point is value of multipole @xmath36 integrand of @xmath37 null positive contribution linear growth balances negative non linear growth feature amplitude of bispectrum in linear part @xmath38 ) different cosmological models produce different amplitudes in linear regime linear power affect @xmath36 higher power higher value of @xmath36.position shift of multipole @xmath36 of bispectrum zero - crossing main cosmological parameters table t : shift analyze variation of @xmath36 main cosmological parameters dark energy equation of state dark energy density energy density hubble constant spectral index of primordial fluctuations @xmath41values @xmath36 shift evaluated parameter fixing reference model @xmath36 sensitive to parameters distances @xmath42 exception @xmath1 bispectrum more noisy than cmb power spectrum cosmic variance higher order effect increase signal to noise ratio including triangles configurations in @xmath43-space multipoles triplets tested improvement bispectrum power spectrum building three - likelihood on dark energy parameters @xmath44 fiducial model for ideal experiment figure f results marginalizing over one dark energy parameter contours joint likelihoods narrower than power spectrum likelihoods result encouraging gain expected on estimation dark energy equation state structure formation discussed alleviate distance degeneracy in cmb power spectrum using isw weak lensing induced cmb bispectrum shown preliminary results sensitivity bispectrum main cosmological parameters simulated joint likelihood of power spectrum bispectrum limited dark energy parameters fixing fiducial cosmological model preliminary results indicate adding bispectrum likelihood to power spectrum contour levels narrower gain in estimation dynamics dark energy equation state expected further study needed assess magnitude improvementprecision cosmology data experiments planck measures tridimensional matter power spectrum ly-@xmath45 forest cosmic shear help use bispectrum better knowledge matter power spectrum bispectrum signal | 0.432985 | 2,568 | 1,288 | , Saving $0.1 in GPT-4. |
154,967 | we explore the ziptronix direct bond interconnect ( dbi ) technology @xcite for the integration of sensors and readout integrated circuits ( roics ) for high energy physics .
the technology utilizes an oxide bond to form a robust mechanical connection between layers which serves to assist with the formation of metallic interlayer connections .
we report on testing results of sample sensors bonded to roics and thinned to 100 @xmath0 m . | 3d integration of a multilayer stack of ics has emerged as a solution to the demand of cost reduction and increase in functionality and performance in semi - conductor markets @xcite .
the 3d integration technology includes alignment and bonding to stack the ics , electrical interconnection between the ics , and thinning to minimize the volume and facilitate electrical interconnections .
we have explored two 3d techniques for sensor and roic integration : cu - sn @xcite and ziptronix dbi bonding .
we describe studies of the latter in these proceedings .
the ziptronix dbi technology utilizes direct oxide bonding where initial wafer bonds are formed by placing two surfaces into contact at room temperature .
the technology includes formation of electrical interconnects integrated with the bond . a process flow to bond pixel sensors and roics and make the electrical interconnects is the following .
patterned metal contact structures are deposited on sensor / roic pixels .
silicon oxide is then deposited to cover the contact structures and wafer surface .
chemical - mechanical polish is used to planarize the wafer surface and expose the metal contacts . to form the initial oxide bond , wafers
are carefully cleaned and their surfaces are activated by a chemical treatment .
they are then placed together with the metal contacts opposed .
a final heating improves the inter - wafer bond strength and the metal - metal contact quality .
alignment accuracies within @xmath11 @xmath0 m have been achieved .
unlike bump - bonding where the interconnect pitch is limited by bump - to - bump shorts , the dbi technology allows very fine pitch .
three micron pitch has been demonstrated .
the dbi technology is also low mass .
the dbi metal contacts have a height less than 1 @xmath0 m and cover a very small fraction of the surface area .
to demonstrate the ziptronix dbi technology for sensor and roic integration , we have contracted with ziptronix to bond sample pixel sensors to btev fpix2 @xcite roic wafers .
the sensor mask was designed at fermilab and the sensors were fabricated at the mit lincoln labs .
these are 1@xmath20.7 @xmath3 , 4-side abuttable sensors with an array of 22@xmath2128 p@xmath4 diodes on 5 k@xmath5 cm n - bulk silicon .
the sensors employ 50 @xmath0 m deep trench etch at the edge followed by doping of the sidewalls using polysilicon to provide a high quality doped surface @xcite .
this allows the anode and thus the charge collection field to be extended to the edge without excessive leakage current .
the fpix2 chip was designed for use with the btev n+-in - n pixel detector , but can be operated with degraded performance with the opposite sign input pulse . when used in this mode the dynamic range of the fpix2 amplifier
is severely limited .
the dbi technology can be used either to bond wafers together or to bond dice to a wafer .
one would normally bond known good roic dices to a sensor wafer .
however , our sensor wafer layout made this problematic , so we chose instead to bond fifty sensors to each of two fpix2 wafers .
being conservative at the beginning of the project , the sensors after bonding were ground to a thickness of 100 @xmath0 m , instead of being ground to the polysilicon trench etch .
the sensor diodes are connected to the inputs of the corresponding fpix2 pixels by dbi metal contacts .
the trench etch filled with polysilicon is also connected by dbi metal contacts to a pad on the fpix2 chip .
this allows biasing the sensor through the fpix2 chip by the electrical connection between the polysilicon and the sensor backside .
such an connection should exist either through the surface and the sensor cut edge , or by punch through 50 @xmath0 m silicon bulk @xcite .
bond voids corresponding to areas without bonding have been found in sixteen integrated devices by ziptronix with scanning acoustic microscopy .
we note that according to ziptronix , void - free bonding yields are 80 - 90@xmath6 when processing a large number of wafers .
excluding also known bad fpix2 chips , we get thirty good integrated devices .
the integrated devices were first tested by pulse injection to check the functionality of the fpix2 chips after the bonding and thinning .
the input capacitance of the sensor to the fpix2 chip is found to be small by comparing the measured noise level and a simulation .
the leakage currents of the sensors were measured as a function of bias voltage .
the individual electrical interconnect per pixel unit was then examined by a laser and a @xmath7 source .
the results indicate that most of the dbi electrical interconnects between the sensor and the fpix2 chip are formed properly . in preparation for beam tests , the response of a device
was calibrated by using a variable x - ray source .
these tests are summarized below .
pulses with a sharp negative step followed by a slow recovery slope were injected into the charge injection capacitors of the selected fpix2 pixels .
the `` hit '' signals were received from the pixels whose analog outputs were above a certain threshold .
the amplitude of the pulses were scanned in a step of 1 mv . at each step , a fixed number of pulses were injected .
assuming gaussian noise , the threshold and noise level of the pixels were then extracted by a cumulative gaussian fit to the number of hits versus the amplitude of the injected pulses .
the noise distribution of a bare fpix2 chip and one bonded to a sensor are shown in figure [ fig : charge]a . by comparing the average noise to a simulation for the fpix2 chip , the sensor input capacitance to the fpix2 chip
is estimated to be around 0.10 - 0.15 pf per pixel unit .
the p@xmath4 diodes of the sensors are connected to the fpix2 inputs near ground potential .
a positive bias voltage was provided through the fpix2 chips to the trench etch filled with polysilicon .
the leakage currents of the sensors were measured as a function of the bias voltage .
a typical measurement is shown in figure [ fig : charge]b .
the rapid increase of the leakage current beyond the full depletion voltage of the sensor ( @xmath88 v ) is due to the bare ground backside and the associated crystal damage .
we plan to fine grind the sensors down to 50 @xmath0 m , followed by backside implantation and laser annealing .
we expect that the leakage current beyond the full depletion voltage will be greatly reduced after this @xcite .
we would like to note another way we have tested to make the sensor backside contact for the bias voltage . in this approach , indium - gallium eutectic was applied and formed a very thin layer which wet the sensor backside .
the backside contact was then made through the in - ga layer .
we tried this first on bare silicon and measured an ohmic contact between the indium - gallium layer and bare silicon .
we then tried this on an integrated device with bond voids at the place where the polysilicon trench etch is connected to the fpix2 chip .
the measured current for this device was zero when applying the bias voltage to the polysilicon trench etch . in the approach with in - ga ,
the current showed a similar behavior as figure [ fig : charge]b .
the surfaces of the integrated sensors were scanned by a laser with single photon energy of 1.13 ev .
the hit map from one of the devices is shown in figure [ fig : charge]c , in which `` hit '' signals were received from more than 99.9% of the pixels .
the analog signal responses of the pixels near the laser spot were probed by an oscilloscope . as can be seen in figures [ fig : charge]d and [ fig : charge]e , the amplitude ( rise time ) of these signals increases ( decreases ) as the bias voltage increases .
this behavior is consistent with charge collection switching from diffusion to drifting .
the integrated devices were also tested using a sr-90 @xmath7 source .
the hit maps are consistent with efficient dbi electrical connections in all the pixel units .
all these tests indicate that the dbi interconnects between the fpix2 chips and the sensors are properly formed .
we plan to study the charge collection efficiency profile inside each pixel unit , as well the profile near the trench in the fermilab testbeam .
as mentioned above , the dynamic range of the fpix2 chip is greatly reduced when it works with p@xmath4-on - n detectors . in order to know in advance whether or not the integrated devices will respond to a `` mip '' signal ,
the response of an integrated device was calibrated by charge injection and a variable x - ray source .
test pulses with a sharp positive step followed by a slow recovery slope were injected into the charge injection capacitor of a fpix2 pixel in the device .
debugging pads on the fpix2 allow one access to the analog signals from one row of pixels .
the analog signal response of the pixel was measured by an oscilloscope as a function of the amplitude of the test pulses .
the analog signal response of the same pixel was also measured when the device was exposed to a variable x - ray source .
the source is primarily an am-241 source inceident on different metal foils , which fluoresce in the x - ray region with different energy .
the results are shown in figure [ fig : charge]f , in which the amplitude of the test pulse and the x - ray energy have been converted to numbers of electron / holes . in the conversion from the amplitude of the test pulse to charge , the nominal value of the charge injection capacitance of the fpix2 chip is assumed . in the conversion from the x - ray energy to charge , we assume the sensor is fully depleted with a sensitive thickness of 100 @xmath0 m .
we find that for a `` mip '' signal , the amplitude of the analog signal response is within the dynamic range of the fpix2 chip .
we explore the dbi technology for the integration of pixel sensors and roics .
this technology provides small pitch , low mass electrical interconnects and allows thinning after bonding .
we have tested sample pixel sensors dbi - bonded with btev fpix2 chips and thinned to 100 @xmath0 m .
we find that the sensor input capacitance to the fpix2 chip is small .
we also find that most if not all of the integral dbi electrical interconnections are formed properly .
we plan to study the charge collection efficiency of the sensors in beam tests . 99 presentation : + ` http://ilcagenda.linearcollider.org/contributiondisplay.py?contribid=207&sessionid=21&confid=2628 ` .
ziptronix , inc . , 800 perimeter park drive , suite b , morrisville , nc 27560 ; + http://www.ziptronix.com/techno/dbi.html ibm journal of research and development volume * 52 * , no . 6 , 2008 , issue devoted to 3d technology .
p. garrou , c. bower and p. ramm , _ handbook of 3d integration technology and applications of 3d integrated circuits _ , wiley - vch , 2008 .
r. lipton , _ et al .
_ , contributions to the same conference , arxiv:0901.4741 .
d. christian , _ et al .
instr . and method * a549 * 165 ( 2005 )
. v. suntharalingam , _
_ , _ megapixel cmos image sensor fabricated in three - dimensional integrated circuit technology _ , ieee sscc 2005 , pp356 - 7 . | 3d integration multilayer stack ics solution cost reduction functionality performance in semi - conductor markets includes alignment bonding electrical interconnection thinning volume explored two 3d techniques cu - sn @xcite ziptronix dbi bonding studies ziptronix dbi technology utilizes direct oxide bonding wafer bonds formed placing surfaces contact at room temperature includes electrical interconnects bond process pixel sensors roics patterned metal contact structures deposited on sensor roic pixels silicon oxide deposited cover contact structures wafer surface chemical - mechanical polish planarize wafer surface expose metal contacts initial oxide bond wafers cleaned activated chemical treatment placed together metal contacts opposed final heating improves wafer bond strength metal contact quality alignment accuracies within @xmath11 @xmath0 m achieved bump - bonding dbi technology allows fine pitch three micron pitch demonstrated low mass dbi metal contacts height less than 1 @xmath0 m cover small fraction surface area dbi contracted with ziptronix to bond sample pixel sensors to btev fpix2 @xcite roic waferssensor mask designed at fermilab sensors fabricated at mit lincoln labs 1@xmath20.7 @xmath3 , 4-side abuttable sensors 22@xmath2128 p@xmath4 diodes on 5 k@xmath5 cm n - bulk silicon sensors employ 50 @xmath0 m deep trench etch edge doping sidewalls using polysilicon high quality doped surface allows anode charge collection field to edge without leakage fpix2 chip for btev n+-in - n pixel detector degraded performance with opposite sign input pulse dynamic range amplifier limited dbi technology bond wafers or dice to wafer sensor wafer wafer layout problematic fifty sensors to two fpix2 wafers sensors after bonding ground to thickness 100 @xmath0 m polysilicon trench etch sensor diodes connected to inputs fpix2 pixels by dbi metal contacts trench etch filled polysilicon connected by to pad on fpix2 chip allows biasing sensor through fpix2 chip by electrical connection connection 50 @xmath0 m silicon bulk bond voids found in sixteen integrated devices by ziptronixziptronix void - free bonding yields 80 - 90@xmath6 processing large number wafers excluding bad fpix2 chips thirty good integrated devices tested by pulse injection functionality fpix2 after bonding thinning input capacitance sensor to fpix2 chip small measured noise level simulation leakage currents measured bias voltage electrical interconnect per pixel unit examined by laser @xmath7 source results dbi electrical interconnects between sensor fpix2 chip formed properly response calibrated variable x - ray source pulses with sharp negative step slow recovery slope injected into charge injection capacitors fpix2 pixels hit signals from pixels analog outputs above threshold amplitude pulses scanned 1 mv fixed number of pulses injected threshold noise level extracted by cumulative gaussian fit to number hits versus amplitude injected pulses noise distribution of bare fpix2 chip one bonded to sensor shown in figure simulation sensor input capacitance to around 0.10 - 0.15 pf per pixel unit p@xmath4 diodes sensors connected to fpix2 inputs near ground potentialpositive bias voltage provided through fpix2 chips to trench etch filled polysilicon leakage currents sensors measured bias voltage typical measurement shown in figure]b rapid increase leakage current beyond full depletion voltage ( @xmath88 v ) due to bare ground backside crystal damage plan to fine grind sensors to 50 @xmath0 m backside implantation laser annealing expect leakage current beyond full depletion voltage after @xcite tested sensor backside contact for bias voltage indium - gallium eutectic applied layer sensor backside backside contact through in - ga layer tried on bare silicon measured ohmic contact then tried on integrated device with bond voids polysilicon trench etch fpix2 chip measured current zero when applying bias voltage approach with in - ga current similar behavior as figure surfaces integrated sensors scanned by laser with single photon energy 1.13 ev hit map shown in figure]c hit signals received from 99.9% pixels analog signal responses of pixels near laser spot probed by oscilloscope amplitude signals increases as bias voltage increasesbehavior consistent with charge collection switching from diffusion to drifting integrated devices tested using sr-90 @xmath7 source hit maps consistent with efficient dbi electrical connections in pixel units tests indicate dbi interconnects between fpix2 chips sensors properly formed plan to study charge collection efficiency profile inside pixel unit profile near trench in fermilab testbeam dynamic range fpix2 chip reduced with p@xmath4-on - n detectors to `` mip '' signal response calibrated by charge injection variable x - ray source test pulses with sharp positive step slow recovery slope injected into charge injection capacitor fpix2 pixel debugging pads allow access to analog signals from pixels analog signal response measured by oscilloscope amplitude test pulses measured to variable x - ray source source am-241 source metal foils results shown in figure amplitude pulse x - ray energy converted to numbers electron / holes conversion to charge nominal value charge injection capacitance fpix2 chip assumed sensor fully depleted with sensitive thickness of 100 @xmath0 m for `` mip '' signal amplitude analog signal response within dynamic range of fpix2 chipexplore dbi technology pixel sensors roics provides small pitch low mass interconnects allows thinning after bonding tested pixel sensors dbi - bonded btev fpix2 chips thinned 100 @xmath0 m sensor input capacitance fpix2 chip small most integral dbi electrical interconnections formed properly plan study charge collection efficiency beam tests presentation http://ilcagenda.linearcollider.org/contributiondisplay=21=2628 ziptronix 800 perimeter park drive suite b morrisville nc 27560 http://www.ziptronix.com/techno/dbi ibm journal research development volume 52 6 2008 3d technology p. garrou bower ramm handbook 3d integration technology circuits 2008. lipton contributions conference arxiv:0901.4741 d. christian method 2005. suntharalingam megapixel cmos image sensor three - dimensional integrated circuit technology 2005 pp356 - 7 | 0.36296 | 2,551 | 1,281 | , Saving $0.1 in GPT-4. |
154,973 | we investigate the robustness of multiparty nonlocality under local decoherence , acting independently and equally on each subsystems . to be specific , we consider an @xmath0-qubit ghz state under depolarization , dephasing , or dissipation channel , and tested the nonlocality by violation of mermin - klyshko inequality , which is one of bell s inequalities for multi - qubit systems .
the results show that the robustness of nonlocality increases with the number of qubits , and that the nonlocality of an @xmath0-qubit ghz state with even @xmath0 is extremely persistent against dephasing . | nonlocality is one of the most surprising and important features of quantum physics .
it is firstly referred by einstein , podolsky , and rosen , and has been receiving an enormous attention and interest of scientists since bell@xcite designed an inequality to expose it .
now , it is an essential part of quantum information science and quantum physics foundation .
however , nonlocality , like other quantum features , is easily destroyed by decoherence , and the concrete pictures of its destruction is not well known .
one might gain insight into it by analyzing nonlocal states under decoherence .
werner@xcite studied a maximally entangled two qubits under white noise , and kazlikowski et .
al.@xcite extended werner s work to a two @xmath1-dimensional system .
in particular , we are interested in the robustness of nonlocality according to the number of consisting particles .
nonlocality is hardly seen at macroscopic level because of strong decoherence , caused by many interactions with environment . in a previous study@xcite
, it was shown that entanglement of an @xmath0-qubit ghz state is more robust to depolarization as the number of qubits increases .
the authors examined persistency of inseparability , which is the mathematical definition of entanglement . in this paper , we are interested in persistency of nonlocality , which is physically meaningful nature of entanglement .
we examine the decoherence process under realistic noise .
besides depolarization we consider two other common decoherence processes , dissipation and dephasing , which are called @xmath2 and @xmath3 process , respectively . for simplicity , we assume equal and independent local decoherence on each sub - particles .
cabello and feito@xcite studied bell s inequality with depolarization and dephasing , and found that violations of bell s inequality for a two qubit system are extremely robust to dephasing .
although they extended their work to three and four qubits , they failed to show this extreme robustness , because they searched the violations of bell s inequality over limited hilbert space . in this paper , we study the nonlocality of multi - qubit ghz states under three decoherences . the paper is organized as follows .
section ii.a introduces our nonlocality criteria , mermin - klyshoko(mk ) inequality , section ii.b describes three decoherence models and decohered ghz states , and section ii.c construct the inequalities of the states . in section iii
, we numerically test the inequalities and discuss the results .
finally , section iv summarizes the conclusions .
or @xmath4 on her qubit @xmath5 , bob performs a measurement @xmath6 or @xmath7 , and charles performs a measurement @xmath8 or @xmath9 , respectively.,width=302 ] we assume the following situation .
parties are sharing multi - qubit ghz states , where individual qubit suffers equal and independent local decoherence . to estimate nonlocality
, parties measure their qubits in one of two promised measurements and consult the results .
1 shows the case where the number of party is three .
we will use the term `` a ghz state '' not only for a three qubit state but also for a multi - qubit ghz state .
we select mermin - klyshko(mk)@xcite inequalities as nonlocality tester .
we want to test the nonlocality of symmetric @xmath0-qubit states by dichotomic measurements(two different measurements per party ) . in that situations ,
the mk inequality has many advantages ; it is simpler than other bell s inequalities for multi - qubit systems@xcite , invariant under permutation , and maximally violated by a ghz state .
however , the inequality also has a limitation ; for some nonlocal states , it can not determine the nonlocality .
mk inequality of an @xmath0-qubit system is written as @xmath10 where , the operator @xmath11 is an @xmath0-qubit _ bell operator _ , and a _ bell value _
@xmath12 is the expectation value of it , given a state @xmath13 . although the bound is @xmath5 as seen in eq .
( [ mk1 ] ) , the bell values of mk inequalities can rise to @xmath14 , quantum mechanically .
the bell operator is recursively defined as follows .
@xmath15 where the symbol @xmath16 is vector product .
the capital letters stand for individual party s measurements(observables ) , and @xmath17th party s two measurements are presented by the primed and non - primed @xmath17th capital ( @xmath18 ) . the letter @xmath19(also the lower - case @xmath20 ) refers to first party and the letter @xmath21(also @xmath22 ) refers to @xmath0th party . the primed bell operator @xmath23 is obtained from @xmath11 by exchanging every primed and non - primed capitals .
the starting of recursion is @xmath24 , which is a chsh inequality , eq .
( [ b2 ] ) below .
for example , the bell operators of two , three , and four mk inequality are the following .
@xmath25 @xmath26 @xmath27 depolarization , dephasing , and dissipation on a qubit are our decoherence models .
let s define @xmath28 as _ the degree of decoherence of an individual qubit _ , which lies between @xmath29 and @xmath5 , where the value @xmath29 means no decoherence , and @xmath5 means complete decoherence .
the depolarization process to a state with a decoherence degree @xmath28 is represented by @xmath30 the dephasing process is described by @xmath31 the dissipation is the process losing energy , and thus it changes a state to a specific state ( e.g. , a ground state ) .
we choose @xmath32 as the ground state .
the dissipation process is then described by @xmath33 these three decoherences can be mathematically simplified to two processes , population transfer and dephasing .
an initial ghz state suffers one of the above decoherence processes , which act equally and independently on each qubits . if every individual qubit of a ghz state is partially depolarized as eq .
( [ depol ] ) , then the density matrix of the depolarized ghz state becomes @xmath34 , \label{depol - ghz } \end{array}\ ] ] where @xmath35 .
the partially dephased ghz state by an amount @xmath28 is @xmath36 , \label{dephase - ghz}\ ] ] and the partially dissipated ghz state is @xmath37 .
\label{dissip - ghz } \end{array}\ ] ] every projection measurement over a qubit can be represented by a unit vector in the bloch sphere .
an arbitrary observable @xmath19 can be written as @xmath38 @xmath39 , @xmath40 , and @xmath41 are pauli spin matrices , and @xmath42 and @xmath43 are the polar and azimuthal angles corresponding to @xmath19 in the bloch sphere representation .
an eigenvalue of the observable(i.e .
a measurement result ) is @xmath5 or @xmath44 .
the measurements conducted by the participating parties construct _ correlation _ , which is the ensemble average of the multiplication of every party s observable @xmath45 .
the correlation of a ghz state is @xmath46 , \label{correl - ghz}\ ] ] the correlation of a depolarized ghz state ( [ depol - ghz ] ) is @xmath47 , \label{correl - depol}\ ] ] the correlation of a dephased ghz state ( [ dephase - ghz ] ) is @xmath48 , \label{correl - dephase}\ ] ] and the correlation of a dissipated ghz state ( [ dissip - ghz ] ) is @xmath49 .
\label{correl - dissip}\ ] ] since mk inequalities are sum of correlations , as seen in ( [ b2])-([b4 ] ) , the inequalities of a pure ghz state and those of the state under depolarization(under dephasing or dissipation ) can be constructed from ( [ correl - ghz ] ) and ( [ correl - depol])(eq .
( [ correl - dephase ] ) or eq .
( [ correl - dissip ] ) ) , respectively . comparing the correlations eq .
( [ correl - ghz]),([correl - depol ] ) and ( [ correl - dephase ] ) , we observe that @xmath50 where the last equality holds when the number of qubits is odd .
therefore , bell inequalities made from the above correlations satisfy @xmath51
under dephasing , eq . ( [ pmax1 ] ) . the points
are for a ghz state under dissipation by numerical calculations , and fits well to eq .
( [ pmax2 ] ) , which is represented by the dashed line .
, width=302 ] we can easily observe the robustness of multi - qubit nonlocality of a ghz state under depolarization and the state of odd @xmath0 under dephasing . the robustness is judged by the amount of the maximum noise for which the bell inequality is still violated , i.e. nonlocality prolongs . from eq .
( [ trivial - mk ] ) the violation condition of the states is @xmath52 in the above inequality , the noise @xmath28 becomes maximal at the highest bell value of a ghz state @xmath53 , which is @xmath14 .
consequently , the maximum noise @xmath54 is given by @xmath55 eq .
( [ pmax1 ] ) is a monotonically increasing function of n that asymptotically becomes close to @xmath56 , and is shown by the solid line of fig . 2 .
in a ghz state under depolarization and the state of odd @xmath0 under dephasing , the nonlocality becomes more robust as the number of qubits increases .
we numerically investigated violations of mk inequalities for nontrivial cases : a ghz state under dissipation and the state of even n under dephasing .
we used mathematica as a simulation tool , and evaluated the system up to five qubits .
the nonlocality of a ghz state also shows persistency against dissipation with the increasing number of qubits .
the maximum noise @xmath54 for which the bell inequality is still violated is numerically calculated and is plotted by the points in fig . 2 .
the numerical result fits well to the equation @xmath57 eq .
( [ pmax2 ] ) is also a monotonically increasing function of n and is drawn by the dashed line of fig . 2 .
to see how nonlocality changes , we plot the maximum bell value of a ghz state under dissipation in fig . 3 .
the curve falls below the inequality bound by dissipation , and it return to the bound at complete dissipation , because the fully dissipated state is a ground state @xmath58 , which has a classical correlation .
4 shows that the nonlocality of a ghz state with even number of qubits is extremely robust against dephasing .
dephasing is considered to be the main cause of destroying quantum coherence .
although dephasing is very fast in a real system , nonlocality always survives , if the dephasing is not complete .
this effect was already noted by cabello and feito @xcite but not for more than two qubits .
the extreme robustness of a ghz state under dephasing critically depends on the parity of the number of qubits .
whether this parity dependence is real or accidental is not clear yet . to study this question further we could explore other nonlocality criteria .
we remark that this feature is very closely related to entanglement purification .
one can get a pure ghz state by consuming many partially dephased states by the purification , if the dephasing does not happen completely @xcite ; this is not the case for depolarization . for states with dissipation given by eq .
( [ dissip ] ) .
the number of qubits is from two to five and is distinguished by line style.,width=302 ] for states under dephasing given by eq .
( [ dephase ] ) .
the number of qubits is from two to five and is distinguished by line style.,width=302 ]
in summary , we studied the persistency of nonlocality over @xmath0-qubit ghz states under realistic local decoherences : depolarization , dephasing , and dissipation .
we observed nonlocality by testing violations of mk inequality of decohered ghz states , using numerical simulations .
this study revealed that multi - qubit ghz states have much robust nonlocality with increasing number of qubits , and showed extreme persistency of a ghz state of even @xmath0 against dephasing .
this strong persistency can not be seen for a ghz state of odd @xmath0 . whether this parity dependence is real nature of nonlocality or due to the incompleteness of nonlocality criterion needs further investigation .
the states , having robust nonlocality , will be very useful in quantum communication and quantum information with realistic noisy circumstances .
the authors thank professor jinhyoung lee for helpful discussions .
s. s. jang and j. w. kim appreciate the financial support from the korea ministry of information and communication .
h. w. lee was supported by korea research institute of standard and science(kriss ) .
j. s. bell , physics * 1 * , 195 ( 1964 ) .
j. f. clauser , m. a. horne , a. shimony , and r. a. holt , phys.rev.lett .
* 23 * , 880 ( 1969 ) .
r. f. werner , phys.rev.a * 40 * , 4277 ( 1989 ) .
d. kazlikowski , p. ginacinski , m. zukowski , w. miklaszwski , and a. zeilinger , phys.rev.lett .
* 85 * , 4418 ( 2000 ) . c. simon and j. kempe , phys.rev.a * 65 * , 052327 ( 2002 ) .
a. cabello and a. feito , phys.rev.a * 72 * , 052112 ( 2005 ) .
n. d. mermin , phys.rev.lett . * 65 * , 1838 ( 1990 ) ; | nonlocality important quantum physics referred by einstein podolsky rosen attention since bell@xcite designed inequality essential part of quantum information science physics foundation nonlocality like destroyed by decoherence concrete pictures destruction not known analyzing nonlocal states under decoherence werner@xcite studied entangled two qubits under white noise kazlikowski extended work to two @xmath1-dimensional system interested in robustness of nonlocality according to number of particles hardly seen at macroscopic level strong decoherence interactions environment previous study entanglement of @xmath0-qubit ghz state more robust to depolarization as number qubits increases authors examined persistency of inseparability interested in persistency of nonlocality entanglement examine decoherence process under realistic noise consider decoherence processes dissipation dephasing called @xmath2 @xmath3 process assume equal independent local decoherence on each sub - particles cabello feito@xcite studied bell s inequality with depolarization dephasing violations for two qubit system robust to dephasingextended work to three four qubits failed show extreme robustness searched violations bell s inequality over limited hilbert space paper study nonlocality multi - qubit ghz states under three decoherences paper organized section ii.a nonlocality criteria mermin - klyshoko(mk ) inequality.b describes three decoherence models decohered ghz states.c inequalities section iii test inequalities discuss results section iv summarizes conclusions bob charles assume situation parties sharing multi - qubit ghz states individual suffers equal local decoherence estimate nonlocality parties measure qubits consult results 1 shows case number party three term ghz state three multi - qubit select mermin - klyshko(mk)@xcite inequalities as nonlocality tester test nonlocality of symmetric @xmath0-qubit states by dichotomic measurements mk inequality advantages simpler than bell s inequalities for multi - qubit systems invariant under permutation maximally violated by ghz state inequality limitation some nonlocal states can not determine nonlocalitymk inequality @xmath0-qubit system written @xmath10 operator @xmath11 @xmath0-qubit bell operator bell value @xmath12 expectation value state @xmath13 bound @xmath5 bell values rise to @xmath14 bell operator defined @xmath15 symbol @xmath16 vector product capital letters party measurements @xmath17th party s two measurements presented primed non - primed @xmath17th capital @xmath18 letter @xmath19 refers first party @xmath21 @xmath0th party primed bell operator @xmath23 obtained from @xmath11 exchanging primed non - primed capitals starting recursion @xmath24 chsh inequality bell operators of two three four mk inequality @xmath25 @xmath26 @xmath27 depolarization dephasing dissipation qubit decoherence models define @xmath28 degree decoherence individual qubit between @xmath29 and @xmath5 value @xmath29 no decoherence @xmath5 complete decoherencedepolarization to decoherence @xmath28 represented by @xmath30 dephasing by @xmath31 dissipation energy changes state to ground state choose @xmath32 as ground state dissipation by @xmath33 three decoherences simplified to population transfer dephasing initial ghz state suffers decoherence processes act on qubits if qubit partially depolarized density matrix of depolarized state becomes @xmath34 partially dephased ghz state @xmath28 is @xmath36 partially dissipated ghz state @xmath37 projection measurement over qubit represented by unit vector in bloch sphere arbitrary observable @xmath19 written as @xmath38 @xmath39 @xmath40 @xmath41 are pauli spin matrices @xmath42 and @xmath43 are polar azimuthal angles to @xmath19 eigenvalue of observable measurement result is @xmath5 or @xmath44 measurements participating parties construct correlation _ ensemble average of multiplication of every party s observable @xmath45 correlation ghz state @xmath46{correl - ghz} depolarized ghz state depol - ghz @xmath47{correl - depol} dephased ghz state dephase - ghz @xmath48 - dephase dissipated ghz state dissip - ghz @xmath49 {correl - dissip} mk inequalities sum of correlations b2]-[b4 inequalities pure ghz state state under depolarization constructed from [ correl - ghz ] [ correl - depol])(eq correl - dephase correl - dissip ] comparing correlations correl - ghz - depol dephase @xmath50 last equality holds when number qubits odd bell inequalities correlations satisfy @xmath51 under dephasing pmax1 ] points for ghz state under dissipation numerical calculations fits to eq . ( [ pmax2 ] ) represented dashed line observe robustness multi - qubit nonlocality ghz state under depolarization odd @xmath0 under dephasing robustness judged by maximum noise bell inequality violated nonlocality prolongseq trivial - mk violation condition @xmath52 inequality noise @xmath28 maximal at highest bell value ghz state @xmath53 @xmath14 maximum noise @xmath54 @xmath55 pmax1 monotonically increasing function n close to @xmath56 shown solid line fig . 2 ghz state depolarization @xmath0 under dephasing nonlocality robust number qubits increases investigated violations mk inequalities cases ghz state under dissipation even n under dephasing evaluated system up to five qubits nonlocality ghz state shows persistency against dissipation increasing number qubits maximum noise @xmath54 bell inequality violated calculated plotted points fig . 2 result equation @xmath57 pmax2 ] monotonically increasing function n dashed line fig . 2 maximum bell value ghz state under dissipation fig . 3 curve falls below inequality bound by dissipation at complete dissipation fully dissipated state ground state @xmath58 classical correlation nonlocality ghz state even number qubits robust against dephasing dephasing main cause quantum coherencedephasing fast nonlocality survives if not complete effect noted by cabello feito @xcite not for more than two qubits robustness of ghz state under dephasing depends on parity number qubits parity dependence real or accidental not clear explore other nonlocality criteria feature related to entanglement purification pure ghz state by consuming partially dephased states if dephasing completely not for depolarization states with dissipation number qubits two to five distinguished by line style under dephasing two to five studied persistency of nonlocality over @xmath0-qubit ghz states under local decoherences depolarization dephasing dissipation observed nonlocality testing violations of mk inequality of decohered ghz states multi - qubit ghz states have robust nonlocality with increasing number qubits extreme persistency of ghz state @xmath0 against dephasing persistency for state of odd @xmath0 parity dependence real nature nonlocality or needs investigation states robust nonlocality useful in quantum communication information noisy circumstances authors thank professor jinhyoung lee for discussionsjang kim financial support korea ministry information communication lee supported korea research institute standard science j s. bell physics 1 195 ( 1964 ). clauser horne shimony holt phys.rev.lett 23 880 ( 1969 ) werner.rev.a 40 4277 ( 1989 ) kazlikowski ginacinski zukowski miklaszwski zeilinger.rev.lett 85 4418 ( 2000 ) c. simon j. kempe.rev.a 65 052327 ( 2002 ) a. cabello feito phys.rev.a 72 052112 2005 ) mermin phys.rev.lett 65 1838 ( 1990 ) | 0.503544 | 3,237 | 1,666 | , Saving $0.1 in GPT-4. |
154,975 | compass is a fixed target experiment at cern .
part of its physics programme is dedicated to study the transverse spin and the transverse momentum structure of the nucleon using sidis . for these measurements ,
data have been collected using transversely polarised proton and deuteron targets .
a selection of recent measurements of azimuthal asymmetries using data collected with transversely polarised protons is presented . | the description of the partonic structure of the nucleon remains one of the main challenges in hadron physics . in the last few decades
, a considerable theoretical and experimental progress has been accomplished and the relevance of the quark transverse spin and transverse momentum to study its structure has been assessed .
the nucleon constituents are not only collinear moving objects but they may also have a momentum component transverse to the nucleon direction of motion . in the present theoretical framework ,
eight transverse momentum dependent parton distribution functions ( tmd pdfs ) are required for each quark flavour at leading twist , describing all possible correlations between the transverse momentum and spin of the quarks , and the spin of the nucleon .
the most famous and studied one is the sivers pdf .
integrating over the quark intrinsic transverse momentum , five among these functions vanish and three survive giving the well known number ( @xmath0 ) , helicity ( @xmath1 ) and transversity ( @xmath2 ) distribution functions .
experimentally , the latter is the least known one . beside these , many other twist-2 distributions can be introduced , correlating the spin and the transverse momentum .
many processes are being , and will be , studied to access the tmd pdfs , namely transversely polarised hard proton - proton collisions , drell - yan processes and semi - inclusive deep inelastic scattering ( sidis ) .
although they are complementary , the last channel is nowadays the major source of information .
its main advantage is that tmd effects generate different azimuthal modulations in its cross section , which can be separately explored and extracted from the same experimental data set .
the modulations depend on two angles , @xmath3 and @xmath4 which define the azimuthal angle of the initial nucleon spin and the produced hadron momentum respectively .
these angles are defined in the so called gamma nucleon system in which the direction of the virtual photon is the @xmath5 axis and the @xmath6 plane is defined by the lepton scattering plane .
the modulation amplitudes are different structure functions , proportional to the convolution of the tmd pdfs and fragmentation functions ( ffs ) .
the transversity distributions @xmath7 can not be measured in inclusive dis due to their chirally odd nature .
they can instead be extracted from measurements of single - spin azimuthal asymmetries in cross - sections for sidis of leptons on transversely polarised nucleons , in which a hadron is also detected .
the measurable asymmetry , the collins asymmetry , is due to a combined effect of @xmath8 and the chiral - odd collins tmd - ff @xmath9 , which describes the spin - dependent part of the hadronization of a transversely polarised quark into a hadron with transverse momentum @xmath10 . at leading order , the collins mechanism leads to a modulation in the azimuthal distribution of the produced hadrons given by : @xmath11 where @xmath12 contains the apparatus efficiency and acceptance , @xmath13 is the target polarisation , @xmath14 is the spin transfer coefficient and @xmath15 is the fraction of polarisable nuclei in the target , @xmath16 is the collins angle , with @xmath4 the hadron azimuthal angle , @xmath17 the final azimuthal angle of the quark spin and @xmath3 the azimuthal angle of the nucleon spin in the gamma - nucleon system .
compass @xcite ( common muon and proton apparatus for structure and spectroscopy ) is a fixed target experiment located at the cern sps taking data since 2002 .
semi - inclusive dis data have been collected using a 160 gev longitudinally polarised muon beam and longitudinally or transversely polarised proton ( nh@xmath18 ) and deuteron ( @xmath19lid ) targets .
the spectrometer comprises a variety of different tracking detectors , and allows to detect charged tracks in a broad momentum and angular range .
calorimeters , muon filters and a gas radiator rich detector are available for particle identification .
collins and sivers asymmetries have been extracted as a function of @xmath20 , @xmath5 and @xmath10 for positive and negative hadrons , pions and kaons , using lepton scattering off transversely polarised deuterons ( 2002 - 04 ) and protons ( 2007,2010 ) . using a deuteron target ,
the resulting collins and sivers asymmetries turned out to be compatible with zero @xcite,@xcite , an observation which has been interpreted as a cancellation between the @xmath21 and @xmath22 quark contributions in the isoscalar target . using a proton target ,
a first measurement was performed separately versus @xmath20 , @xmath5 and @xmath10 for unidentified hadrons , pions and kaons @xcite,@xcite,@xcite .
the collins asymmetry is small , compatible with zero , except for @xmath23 where a significant signal ( up to 10 % ) appears with opposite sign for positive and negative hadrons , pions and kaons .
the results for the sivers asymmetry are compatible with zero for negative hadrons and exhibit small positive values ( up to 3% ) for positive hadrons both at small and at large @xmath20 . compared with hermes results measured at smaller @xmath24
, the results were found to be slightly smaller .
a further investigation showed that the signal is concentrated at small @xmath25 while at larger @xmath25 , it tends to zero .
thus compass data highlights a possible @xmath25 dependence of the sivers asymmetry for positive hadrons .
the other six asymmetries were extracted from deuteron and proton data and were found to be compatible with zero .
recently , further investigation of the previous observations has been performed by studying the @xmath20 , @xmath5 , @xmath10 and @xmath25 dependencies in different @xmath24 regions : @xmath26 [ 1,4 ] , [ 4,6.25 ] , [ 6.25,16 ] and @xmath27 ( gev / c)@xmath28 , using the data set collected in 2010 on a transversely polarised proton . , @xmath5 , @xmath10 , @xmath25 in four @xmath24 ranges .
] , @xmath5 , @xmath10 , @xmath25 in four @xmath24 ranges . ] these new results for collins and sivers asymmetries are shown in fig .
[ fig : col]-[fig : siv ] versus @xmath20 , @xmath5 , @xmath10 and @xmath25 and the mean of the eight asymmetries are shown in fig .
[ fig : all ] .
error bars show only statistical uncertainties .
the systematic uncertainties , estimated separately for each asymmetry and hadron charge , are shown by the bands .
the increase of collins asymmetry at large @xmath20 and small @xmath25 increases with @xmath24 . for sivers asymmetry ,
the non - negligible signal for positive hadrons at large @xmath20 is clearly visible , also at large @xmath24 , which is important for the drell - yan measurement foreseen by compass in 2015 .
the other six asymmetries are compatible with zero .
a first study of the dependence of collins , sivers and the other six asymmetries upon kinematic variables @xmath20 , @xmath5 , @xmath10 and @xmath25 in different @xmath24 regimes is performed , showing a non negligible @xmath24 dependence of the collins asymmetry .
a more detailed multidimensional analysis of the eight asymmetries in simultaneous bins of ( @xmath20,@xmath24 ) and ( @xmath20,@xmath5,@xmath10 ) is close to be finalised .
99 p. abbon _
[ compass collaboration ] , http://www.sciencedirect.com/science/article/pii/s0168900207005001[nucl .
instrum . meth .
a * 577 * 455 ( 2007 ) ] .
[ compass collaboration ] , http://www.sciencedirect.com/science/article/pii/s0370269307002419[nucl .
b * 766 * 31 ( 2007 ) ] .
[ compass collaboration ] , http://www.sciencedirect.com/science/article/pii/s0370269309000884[phys .
b * 673 * 127 ( 2009 ) ] . c. adolph _ et al .
_ [ compass collaboration ] , http://www.sciencedirect.com/science/article/pii/s0370269312010027[phys .
b * 717 * 376 ( 2012 ) ] . c. adolph _ et al .
_ [ compass collaboration ] , http://www.sciencedirect.com/science/article/pii/s0370269312010039[phys .
b * 717 * 383 ( 2012 ) ] . c. adolph _ et al .
_ [ compass collaboration ] , http://arxiv.org/abs/arxiv:1408.4405[hep-ex/1408.4405 ] . | description partonic structure nucleon in hadron physics decades theoretical experimental progress accomplished relevance quark transverse spin momentum to study structure assessed nucleon constituents collinear moving objects momentum transverse to nucleon direction motion theoretical framework eight transverse momentum dependent parton distribution functions required for each quark flavour at leading twist correlations between momentum spin quarks nucleon famous sivers pdf momentum five functions vanish three survive number helicity @xmath1 transversity @xmath2 ) distribution functions latter least known many other twist-2 distributions introduced correlating spin transverse momentum processes studied access tmd pdfs transversely polarised proton - proton collisions drell - yan processes semi - inclusive deep inelastic scattering last channel major source of information tmd effects generate different azimuthal modulations in cross section separately explored from experimental data set modulations depend on two angles @xmath3 @xmath4 angle initial nucleon spin produced hadron momentumangles defined in gamma nucleon system direction virtual photon @xmath5 axis @xmath6 plane defined lepton scattering plane modulation amplitudes structure functions proportional to convolution tmd pdfs fragmentation functions transversity distributions @xmath7 measured inclusive dis chirally odd extracted from measurements single - spin azimuthal asymmetries cross - sections sidis leptons on transversely polarised nucleons hadron detected asymmetry collins asymmetry due to effect @xmath8 chiral - odd collins @xmath9 spin - dependent hadronization transversely polarised quark into hadron transverse momentum @xmath10 collins mechanism leads to modulation azimuthal distribution hadrons by apparatus efficiency acceptance target polarisation spin transfer coefficient fraction of polarisable nuclei collins angle hadron azimuthal angle final azimuthal angle quark spin nucleon spin gamma - nucleon system compass @xcite fixed target experiment cern sps since 2002 semi - inclusive dis data collected using 160 gev longitudinally polarised muon beam polarised proton deuteron targetsspectrometer tracking detectors charged tracks broad momentum angular range calorimeters muon filters gas radiator detector available for particle identification collins sivers asymmetries extracted @xmath20 for positive negative hadrons pions kaons using lepton scattering off deuterons protons deuteron target collins sivers asymmetries compatible with zero @xcite interpreted cancellation between @xmath21 @xmath22 quark contributions proton target first measurement performed versus for unidentified hadrons pions kaons collins asymmetry small compatible with zero except @xmath23 significant signal 10 % with opposite sign for positive negative hadrons pions kaons sivers asymmetry compatible with zero for negative hadrons small positive values 3% for positive hadrons at small large @xmath20 smaller @xmath24 results slightly smaller signal concentrated at small @xmath25 larger tends to zero highlights possible @xmath25 dependence of sivers asymmetry for positive hadrons other six asymmetries extracted from deuteron proton data compatible with zeroinvestigation observations studying @xmath20 dependencies in @xmath24 regions : @xmath26 [ 1,4 4,6.25 6.25,16@xmath28 using data set 2010 on transversely polarised proton @xmath5 in four @xmath24 ranges new results for collins sivers asymmetries shown in fig . versus @xmath20 mean of eight asymmetries fig . error bars show statistical uncertainties systematic uncertainties shown by bands increase of collins asymmetry at @xmath20 @xmath25 increases with @xmath24 sivers asymmetry non - negligible signal for positive hadrons at @xmath20 visible @xmath24 important for drell - yan measurement compass 2015 other six asymmetries compatible with zero first study of dependence of collins sivers other six asymmetries upon kinematic variables @xmath20 in @xmath24 regimes performed showing non negligible @xmath24 dependence of collins asymmetrydetailed multidimensional analysis eight asymmetries bins @xmath20 close finalised 99 p. abbon compass collaboration.com/science/article/pii/s0168900207005001[nucl instrum meth a * 577 * 455 2007 ) compass collaboration/s0370269307002419[nucl. b * 766 * 31 2007 )/s0370269309000884[phys . b * 673 * 127 2009 ) adolph et al compass collaboration/s0370269312010027[phys . b * 717 * 376 2012 ) adolph compass collaboration/s0370269312010039[phys . b * 717 * 383 2012 ) compass collaboration http://arxiv.org/abs/arxiv:1408.4405-ex/1408.4405 | 0.313516 | 2,009 | 998 | , Saving $0.1 in GPT-4. |
154,977 | we present three - dimensional numerical simulations of the interaction of a circular - orbit planet with a protoplanetary disk .
we calculate the flow pattern , the accretion rate , and torques on a planet .
we consider planet masses ranging from 1 earth mass to 1 jupiter mass .
# 1_#1 _ # 1_#1 _ = # 1 1.25 in .125 in .25 in | planet formation is believed to involve accretion from the surrounding disk of material .
we consider this gas accretion onto a planet .
a non - keplerian flow pattern develops near the planet .
furthermore , the planet is subject to gravitational torques due to its interaction with the gas disk which result in planetary migration .
we have extended the 2d simulations by lubow , seibert , and artymowicz ( 1999 ) to 3d , in order to analyze these effects more realistically .
we expect 3d effects to be important for planet masses below @xmath0m@xmath1 , since the hill sphere radius becomes smaller than the disk thickness .
the calculations were performed by using the 3d , spherical version of the zeus finite - difference code ( stone et al .
the disk model is similar to that of lubow et al .
( 1999 ) , except that we include the third dimension .
the disk is locally isothermal with @xmath2 , and disk thickness ratio @xmath3 .
the initial surface density is taken to be @xmath4 away from the planet .
the total disk mass is assumed to be 0.0075 solar masses in the radial interval from 1.6 to 21 au .
the planet s orbital radius is taken to be 5.2 au .
thus , the surface density of the unperturbed disk at the planet s orbital radius is 75 g @xmath5 .
we apply a constant kinematic viscosity @xmath6 which , at the planet s radius , corresponds to viscosity parameter @xmath7 .
we adopt a grid @xmath8 that is non - uniform in @xmath9 and @xmath10 .
the highest resolution is achieved close to the planet .
zones near the planet measure 0.002875 in @xmath9 and @xmath10 , and twice this value in @xmath11 .
we model 4 vertical scale heights above the midplane .
we consider separately planets of masses 1 , 0.1 , 0.03 , 0.01 , and 0.003 m@xmath1 which orbit about a 1 @xmath12 star .
the latter planet mass corresponds to 1 earth mass . in all cases ,
the flow is resolved within the planet s hill sphere .
the planet - disk interaction modifies the disk in the vicinity of the planet ( fig . 1 ) .
in fig . 2 , we plot the azimuthally - averaged disk surface density .
only a 1@xmath13 planet opens a deep gap and a mass of at least 0.1@xmath13 is required for the surface density to be perturbed significantly .
lower mass planets result in much less severe surface - density perturbations .
in fig . 3 , left panel , we plot the migration rates of the 5 planets . we find that the migration rate increases linearly with planet mass , as expected for type i migration where the planet does not open a gap in the disk .
we also plot the migration rates expected from the linear theories of ward ( 1997 ) and tanaka et al .
ward s calculations approximate the disk as two - dimensional and do not include co - rotation resonances .
tanaka et al .
consider three - dimensional disks and both lindblad and co - rotation resonances .
our results are in excellent agreement with tanaka et al . for @xmath14 jupiter masses ( i.e. in the type i regime ) .
in fig . 3 , right panel , we plot the accretion rates onto the planet against planet mass for the 5 planets ( filled circles ) . also plotted are the rates obtained for higher mass planets from the two - dimensional calculations of lubow et al ( 1999 ) .
* the migration rate scales linearly with @xmath15 for planets that do not open a gap .
the results are in excellent agreement with the linear analysis of tanaka et al .
planets with masses @xmath16 m@xmath1 migrate on the disc s viscous timescale . * the mass accretion rate increases linearly with planet mass @xmath15 for low - mass planets and peaks at @xmath17 m@xmath1 . at higher masses ,
the mass accretion rate declines as a gap is opened in this disk . | planet formation accretion from gas accretion onto planet non - keplerian flow pattern near planet subject to gravitational torques gas disk planetary migration extended 2d simulations to 3d 3d effects important for planet masses below @xmath0m@xmath1 hill sphere radius smaller than disk thickness calculations 3d zeus finite - difference code disk model similar to lubow third dimension disk locally isothermal @xmath2 thickness ratio @xmath3 initial surface density @xmath4 away from planet total disk mass 0.0075 solar masses radial interval 1.6 to 21 au planet s orbital radius 5.2 au surface density unperturbed radius 75 g @xmath5 constant kinematic viscosity @xmath6 corresponds to viscosity parameter @xmath7 grid @xmath8 non - uniform in @xmath9 and @xmath10 highest resolution close to planet zones near planet measure 0.002875 in @xmath9 @xmath10 twice in @xmath11 model 4 vertical scale heights above midplane planets masses 1 0.1 0.03 0.01 0.003 m@xmath1 orbit about 1 @xmath12 starlatter planet mass to 1 earth mass flow resolved within planet s hill sphere planet - disk interaction modifies disk vicinity fig . 1 fig . 2 - averaged disk surface density 1@xmath13 planet opens deep gap mass 0.1@xmath13 required for surface density lower mass planets less severe surface density perturbations fig . 3 migration rates of 5 planets migration rate increases with planet mass type i migration gap migration rates from theories ward 1997 tanaka et al s calculations co - rotation resonances tanaka three - dimensional disks results for @xmath14 jupiter masses type i regime fig . 3 accretion rates planet against planet mass for 5 planets rates for higher mass planets from calculations lubow et al 1999 migration rate scales linearly with @xmath15 for planets gap with analysis tanaka al planets with masses @xmath16 m@xmath1 migrate on disc s viscous timescale mass accretion rate increases with planet mass @xmath15 for low - mass planets peaks at @xmath17 m@xmath1higher masses accretion rate declines gap opened disk | 0.178559 | 933 | 490 | , Saving $0.0 in GPT-4. |
154,980 | the particle density in extensive air showers fluctuates at the ground level .
these fluctuations , at the scale of the scintillator detector size ( several meters ) , lead to the diversity of the individual detector responses . therefore , small scale fluctuations contribute to the error in the estimation of the primary energy by a ground array .
this contribution is shown to be non - gaussian .
the impact on the primary energy spectrum measured by a ground array is estimated .
ir is argued that super - gzk events observed by agasa experiment do not result from the energy overestimation , due to small scale fluctuations , of lower energy events . | a significant part of conclusions on ultra - high energy cosmic rays ( uhecrs ) is made today on the basis of the quantities observed by ground detector arrays .
typical ground detector registers only small fraction @xmath0 of the shower particles at the ground level .
the reading of the individual detector in the array is determined by the local density of particles in the shower .
the latter is affected by the small scale fluctuations within the shower . a typical eas , induced by a vertical proton with a primary energy of about @xmath1 ev contains about 20 billions of particles with energies above @xmath2 mev at the ground level .
the modelling of such a huge number of particles is time and resource consuming process .
monte - carlo simulations of air showers induced by ultra - high energy cosmic rays often involve procedures like `` thinning '' @xcite , aimed to reduce the effective number of particles in a calculation .
these procedures make it impossible to estimate small scale fluctuations in a reliable way . suggested `` unthinning '' procedure @xcite washes out small scale fluctuations by its definition . in the present work we use some artificial vertical proton - induced air - showers with energies up to @xmath3 ev ,
simulated without thinning , to estimate the impact of the small scale fluctuations on the energy spectrum in the ultrahigh - energy region , observed by a ground detector array with detectors simular to ones used in agasa experiment .
the fluctuations were earlier estimated experimantally by the akeno experiment @xcite .
the standard deviation value for the fluctuations was obtaned but the fluctuation distribution has not been discussed .
ground array experiments traditionally use average density of scintillator signal @xmath4 , measured in units of vertical equivalent muons ( vems ) , at a fixed constant core distance @xmath5 as the energy estimator .
the @xmath4 is usually obtained by fitting the detectors readings by the empirical lateral distribution function ( ldf ) @xmath6 .
for example , agasa experiment uses @xmath7 and the following empirical ldf @xcite , @xmath8 where @xmath9 only detectors with 300 m@xmath10 1000 m are used for the fit .
average scintillator detector responses for agasa experiment are published in @xcite . to ensure fit quality , agasa implements
the following procedure : the worst detector is excluded from the fit in the case of bad @xmath11 ( @xmath12 ) @xcite and the procedure is repeated until @xmath13 .
detectors experiencing largest local fluctuations are expected to be excluded by this procedure .
we have generated several showers without thinning with primary energies ranging from @xmath14 to @xmath3 ev .
the simulations were preformed with corsika @xcite .
qgsject 01c @xcite , qgsjet ii @xcite , gheisha @xcite and egs4 @xcite models were used in simulations . the simulations without thinning are time consuming and cpu - time and storage required grows nearly linearly with primary energy .
this is why we limit our simulations up to energies of @xmath3 ev .
all datafiles are made publicly available within the public library of artificial air showers called `` livni '' , so all the results of this and the following works may be confirmed using the same dataset .
the library may be used for any other studies of the structure of air showers .
the detailed information on the library content and access rules are available at the website @xcite .
hereafter we use references to the library showers in the form of livni : codename , e.g. reference to the shower , named in the library as `` 18 - 3 '' will be livni:18 - 3 .
in the present work we intend to study fluctuations in air showers induced by the highest energy primary particles .
unfortunately , simulations of artificial air showers of such a high energy is impractical . in order to make statements on @xmath1 ev showers we consider scintillator detectors with an area 100 times larger than normal detector area ( @xmath15 in agasa ) and utilize simulated showers with a primary energy of @xmath3 ev .
this of course make our analysis approximate .
the procedure is justified by the fact that the lateral distribution functions of scintilation signal density for energies @xmath3 ev and @xmath1 have a similar shape @xcite .
furthermore , we have calculated the cross - correlation function , @xmath16 , of scintillation density at the ground level .
we found that cross - correlation function is close to zero on the scale of the detector size , @xmath17 . the scintillation density distribution over detectors centered at the core distance between 595 and 605 meters is shown in fig .
[ single ] . to produce this plot
we assume the ground to be fully covered by the detectors .
it can be seen that an individual detector may be exposed to a larger or smaller density than an average one and the central part of the distribution obeys the gaussian law in @xmath18 scale .
let us note that the plot refers to 600 m core distance where the fluctuations are small . at larger distances an individual detector may be exposed to up to 100 times larger particle density than an average , though the probability is small . scintillation density distribution over detectors ( size of @xmath19 ) centered at [ 595 m;605 m ] core distance .
horizontal axis : scintillator density , normalized to average , vertical axis : number of detectors with a signal in a bin centered in @xmath20 ( livni:18 - 3 shower ) . ] in order to reconstruct the observables we assume the ground array detector to consist of 100 plastic 5 cm thick scintillators ( @xmath21 @xmath22 @xmath21 ) forming square lattice covering the area of @xmath23 .
our intent is to make our detection procedure close to one used for the analysis of agasa experimental data @xcite .
the detector responses are estimated using agasa average detector response functions @xcite .
each simulated shower has been detected 30000 times with different core positions within the ground array and azimuthal angles with respect to the array .
fitting responses of the detectors at core distances from 300 to 1000 meters with the agasa experimental ldf eq .
( [ agasaldf ] ) , we obtain @xmath20 . following the agasa procedure to ensure fit quality we exclude the worst detector from the fit in the case of bad @xmath11 ( @xmath12 ) @xcite .
the distribution of the number of excluded detectors is presented in table 1 does not include detector fluctuations and therefore is different from the experimental one . ] . .
the distribution of the number of excluded detectors in the fit quality assurance procedure ( livni:18 - 3 shower ) [ cols="^,^ " , ] distribution of reconstructed s(600 ) normalized to the average value ( livni:18 - 3 shower ) .
estimated energy distribution has the same form as an estimated energy is nearly proportional to @xmath20 . ]
the resulting distribution of reconstructed @xmath20 calculcated for one artificial air shower is shown on fig .
. the reconstruction error may depend on the first interaction producing a shower , as we discuss later in sec .
[ sec : lim ] .
the main part of the distribution may be fit with the gaussian in @xmath18 scale .
the same type of profile is suggested by agasa for @xmath20 experimental error distribution @xcite . finally , we estimate one - sigma error for @xmath20 reconstruction for @xmath1 ev air showers due to small scale fluctuations as @xmath24 . in rare cases
@xmath20 may be overesimated by factor of @xmath25 , hovewer , the probabilty of this is less than @xmath26 .
we may also see that the part of the distribution which corresponds to energy overestimation is broader than the underestimation part .
the estimate does not include the fluctuations of the detector response , which are present in the experiment and are of the same order of magnitude @xcite .
the fluctuations discussed above may affect the primary spectrum observed by a ground array .
significant energy overestimation even in a relatively small number of cases may influence the experimental conclusion on the presence of the gzk cut - off @xcite .
let us assume a toy primary spectrum with a spectral density proportional to @xmath27 , @xmath28 up to the energy of @xmath1 ev and equal to zero for higher energies .
we have calculated the convolution of our toy spectrum and energy estimation fluctuation distribution ( which is the same as @xmath20 fluctuation distribution , presented in fig .
[ sdistr ] ) . the resulting spectrum is shown in fig . [ spec ] .
we see that the fluctuation contribution to the spectrum may be considered minor for the gzk - predictions : the probability of energy overestimation by a factor of @xmath25 is less than @xmath26 .
we conclude that super - gzk events observed by agasa experiment do not result from energy overestimation , due to small scale fluctuations , of lower energy events . the impact of the small scale fluctuations on the toy primary spectrum with a cut - off .
dashed line shows the convolution of a toy spectrum and an energy estimation error presented in fig .
[ sdistr ] ]
one limitation of the above analysis is that is does not include the fluctuations of the detector response which are of the same order of magnitude @xcite .
the fluctuations within the detector may be accounted for by combining the library showers with relevant monte - carlo simulations of the detector .
the second problem one should care about is that the magnitude of small - scale fluctuations may be different for showers having different first interactions . to estimate this effect
we analysed 20 showers with primary energy of @xmath14 ev and 3 showers with primary energy of @xmath3 ev .
for each shower we have estimated @xmath29 , where @xmath30 is a standard deviation of the detector response , measured in vems , at distance @xmath31 ( calculated for the ensemble of detectors centered at core distances close to @xmath31 ) and @xmath32 is the average detector response . as long as the correlation function @xmath33 is zero @xmath34 does nt depend on the detector size .
this motivates the choice of @xmath34 as an estimator for the fluctuation magnitude in a particular artificial shower . for pure poisson distribution ( which would be the case for equivalent independent particles ) @xmath35 .
average and maximum values of @xmath36 for @xmath14 ev showers are 0.47 and 0.51 , respectively , with a standard deviation of @xmath24 .
three studied @xmath3 showers have @xmath36 equal to 0.48 , 0.57 and 0.69 .
these numbers imply that the difference between statistical properties of different showers is not very substantial . to interpret these data further
, we assume that the character of the fluctuations does not change dramatically when the primary energy changes from @xmath3 to @xmath1 ev . as the study is based on a small number of artificial showers , we also have to assume practical inexistence of air showers with extremely large fluctuations .
the first assumption may be checked by a simulation of @xmath1 ev artificial air shower without thinning and the second by simulating hundreds of showers without thinning .
both simulations are extremely resource consuming and yet are expected to be possible in the nearest future .
thus , the difference in the fluctuation magnitude in different studied artificial showers is not substancial for the conclusions of the work , although our study is limited by the following : * the primary energy of artificial showers does not exceed @xmath3 ev . *
the number of artificial events studied is relatively small and does not exclude the existence of paricular air showers with extremely large fluctuations in them . the above limitations may be checked in the future .
we calculated the cross - correlation function of the scintillation signal density and found it to be close to zero at the detector scale @xmath37 .
we estimated a contibution of small scale fluctuations on the detector scale to the energy reconstruction error by a ground array at the level of about @xmath24 for primary energy of about @xmath1 ev . the contribution , although found to be non - gaussian , is minor for gzk predictions .
the study , however , has certain limitations ; we discussed ways to get rid of them .
we are indebted to l. g. dedenko , d. s. gorbunov , m. kachelrie , a. kusenko , r. peccei , c. rebbi , v. a. rubakov , s. shinozaki , m. teshima , g. thomson , i. i. tkachev , and s. .v troitsky for useful discussions and criticism .
g.r . is grateful to the boston university , where the work has been started .
this work was supported in part by the intas grant 03 - 51 - 5112 , by the russian foundation of basic research grant 05 - 02 - 17363 ( gr ) , by the grants of the president of the russian federation ns-7293.2006.2 ( government contract 02.445.11.7370 ) , by the fellowships of the `` dynasty '' foundation ( awarded by the scientific council of icfpm , gr ) .
numerical part of the work was performed at the computer cluster of the theoretical division of inr ras .
d. heck _ et al .
_ , report fzka-6019 ( 1998 ) , forschungszentrum karlsruhe ; d. heck and j. knapp , fzka-6097 _ prepared for spring meeting of the germany physical society ( dpg ) on particle physics ( in german ) ,
freiburg , germany , 23 - 27 mar 1998 _
n. n. kalmykov , s. s. ostapchenko and a. i. pavlov , nucl .
. suppl .
* 52b * , 17 ( 1997 ) . s. ostapchenko , nucl .
* 151 * , 143 ( 2006 ) [ arxiv : hep - ph/0412332 ] . h. fesefeldt , pitha 85/02 . | conclusions on ultra - high energy cosmic rays ground detector arrays typical ground detector registers small fraction shower particles at ground level reading determined by local density particles in shower affected by small scale fluctuations typical eas induced by vertical proton primary energy @xmath1 ev contains 20 billions particles energies above @xmath2 mev at ground level modelling time resource consuming simulations of involve thinning @xcite particles to estimate small scale fluctuations unthinning washes out small scale fluctuations work use artificial vertical proton - induced air - showers with energies up to @xmath3 ev without thinning to estimate impact small scale fluctuations on energy spectrum in ultrahigh - energy region by ground detector array experiment fluctuations estimated by akeno experiment standard deviation value for obtaned fluctuation distribution not discussed ground array experiments use average density of scintillator signal @xmath4 at fixed constant core distance @xmath5 as energy estimator @xmath4 obtained by fitting detectors readings by empirical lateral distribution function @xmath6 agasa experiment uses @xmath7 ldf @xcite @xmath8 @xmath9 detectors with 300 m@xmath10 1000 m used for fit average scintillator detector responses published in @xcite fit quality procedure worst detector excluded from bad @xmath11 @xcite repeated until @xmath13 detectors largest local fluctuations excluded generated showers without thinning with primary energies @xmath14 to @xmath3 ev simulations with corsika @xcite qgsject 01c qgsjet ii gheisha egs4 @xcite models used simulations without thinning time consuming cpu - time storage grows with primary energy limit simulations to energies @xmath3 ev datafiles publicly available within public library of artificial air showers livni '' results confirmed using same dataset library used for other studies detailed information library content access rules at website @xcite references to library showers in livni : codename study fluctuations in air showers highest energy primary particles simulations of artificial air showers high energy impracticalstatements @xmath1 ev showers consider scintillator detectors area 100 times larger normal @xmath15 simulated showers primary energy @xmath3 ev analysis approximate procedure justified lateral distribution functions scintilation signal density @xmath3 ev @xmath1 similar shape calculated cross - correlation function @xmath16 scintillation density at ground level close to zero on scale detector size @xmath17 scintillation density distribution over detectors core distance between 595 and 605 meters shown in fig . ground covered by detectors individual detector exposed larger or density distribution obeys gaussian law in @xmath18 scale plot refers to 600 m core distance fluctuations small larger distances detector 100 times larger particle density probability small scintillation density distribution over detectors size @xmath19 ) at 595 m;605 m core distance horizontal axis : scintillator density normalized average vertical axis number of detectors signal bin centered @xmath20 ground array detector 100 plastic 5 cm thick scintillators @xmath21 square lattice covering area @xmath23 intent detection procedure close to agasa experimental data @xcitedetector responses estimated using agasa average functions each simulated shower detected 30000 times different core positions azimuthal angles fitting responses detectors at distances 300 to 1000 meters with agasa experimental ldf eq obtain @xmath20 exclude worst detector bad @xmath11 ) distribution excluded detectors in table 1 include fluctuations different from experimental distribution in fit quality assurance procedure distribution of reconstructed s(600 ) normalized to average value estimated energy distribution proportional to @xmath20 resulting distribution of reconstructed @xmath20 for one artificial air shower shown on fig . reconstruction error may depend on first interaction shower main part distribution may be fit with gaussian in @xmath18 scale same profile suggested by agasa for @xmath20 experimental error distribution estimate one - sigma error for @xmath20 reconstruction for @xmath1 air showers due to small scale fluctuations as rare cases @xmath20 may be overesimated by factor of @xmath25 probabilty less than @xmath26 distribution to energy overestimation broader than underestimationestimate include fluctuations detector response present same magnitude @xcite fluctuations may affect primary spectrum ground array energy overestimation may influence experimental conclusion gzk cut - off @xcite assume toy primary spectrum spectral density proportional to @xmath27 @xmath28 up @xmath1 ev equal to zero for higher energies calculated convolution spectrum energy estimation fluctuation distribution same as @xmath20 fluctuation distribution fig . resulting spectrum shown in fig . spec fluctuation contribution minor for gzk - predictions probability energy overestimation by @xmath25 less than @xmath26 super - gzk events experiment result from energy overestimation small scale fluctuations lower energy events impact small scale fluctuations on toy primary spectrum cut - off line shows convolution energy estimation error fig . limitation analysis include fluctuations detector response same magnitude @xcite fluctuations detector accounted for by combining library showers with monte - carlo simulations magnitude small - scale fluctuations may different for showers different first interactions analysed 20 showers energy @xmath14 ev 3 @xmath3 eveach shower estimated @xmath29 @xmath30 standard deviation detector response measured in vems at distance @xmath31 @xmath32 average detector response correlation function @xmath33 zero @xmath34 depend on detector size motivates choice @xmath34 estimator fluctuation magnitude artificial shower pure poisson distribution @xmath35 average maximum values @xmath36 @xmath14 ev showers 0.47 0.51 standard deviation @xmath24 three @xmath3 showers have @xmath36 equal 0.48 0.57 0.69 difference between statistical properties showers not substantial character fluctuations change primary energy changes from @xmath3 to @xmath1 ev study small number artificial showers inexistence of air showers with extremely large fluctuations checked simulation @xmath1 ev artificial air shower without thinning simulating hundreds of showers without thinning simulations resource consuming expected possible future difference in fluctuation magnitude in artificial showers not substancial conclusions study limited by primary energy artificial showers exceed @xmath3 ev number of artificial events studied small exclude existence air showers with extremely large fluctuations limitations may checked futurecalculated cross - correlation scintillation signal density found close to zero at detector scale @xmath37 estimated small scale fluctuations to energy reconstruction error ground array @xmath24 for primary energy @xmath1 ev contribution non - gaussian minor for gzk predictions study has limitations discussed ways indebted to. dedenko gorbunov kachelrie kusenko peccei rebbi rubakov shinozaki teshima. thomson tkachev troitsky for discussions criticism grateful boston university work supported by intas grant 03 - 51 - 5112 russian foundation basic research grant 05 - 02 - 17363 grants president russian federation ns-7293.2006.2 fellowships dynasty '' foundation scientific council numerical part work performed at computer cluster theoretical division inr ras d. heck et al. report fzka-6019 1998 forschungszentrum karlsruhe heck j. knapp fzka-6097 spring meeting germany physical society particle physics freiburg germany 23 - 27 mar 1998 n. kalmykov s s. ostapchenko a. i. pavlov nucl suppl 52b 17 1997 ostapchenko 151 143 2006 ) arxiv hep ph/0412332 fesefeldt pitha 85/02 | 0.498881 | 3,169 | 1,557 | , Saving $0.1 in GPT-4. |
154,988 | a transformed auto - correlation method is presented here , where a received signal is transformed based on a priori reflecting model , and then the transformed signal is cross - correlated to its original one .
if the model is correct , after transformation , the reflected signal will be coherent to the transmitted signal , with zero delay .
a map of transformed auto - correlation function with zero delay can be generated in a given parametric space .
the significant peaks in the map may indicate the possible reflectors nearby the central transmitter .
the true values of the parameters of reflectors can be estimated at the same time . | echo signals probably exist in the light curves of some sorts of astronomical objects , such as supernovae ( ( * ? ? ?
* rest 2005 ) , ( * ? ? ?
* krause 2008a ) , ( * ? ? ?
* krause 2008b ) ) , x - ray binaries ( ( * ? ? ? * greiner 2001 ) ) and agns ( ( * ? ? ?
* perterson 1993 ) ) etc .. if the relative location between a central transmitter and a reflecting object is fixed , then auto - correlation ( ( * ? ? ?
* edelson & krolik 1988 ) ) can be used to detect the reflected signal and estimate the distance thereafter . however , in common situation , the nearby reflecting object is always moving ( ( * ? ? ? * perterson 2008 ) ) , therefore , the delay between transmitted and reflected signal is variable , which means the coherence is broken and no echo signals could be detected by auto - correlation . here
, we are going to introduce a method which is called transformed auto - correlation .
its aim is to rebuild the coherence and perform cross - correlation between the reflected and transmitted signals .
suppose there are only one transmitter and one reflector , and @xmath0 is a transmitted signal , and @xmath1 is the received signal . since astronomical objects are usually far far away , the received light curve is the combination of transmitted and reflected signals , i.e. , @xmath2 where @xmath3 is reflectance , and @xmath4 is the variable delay which is model dependent , @xmath5 are the parameters of the model . due to variable delay , a reflected signal will probably loss the coherence to the transmitted signal .
in such situation , it is impossible to detect a reflected signal by auto - correlation .
a transformation is needed to rebuild the coherence .
let s set : @xmath6 where @xmath7 is the inversion function of @xmath8 in term of @xmath9 . applying equation .
[ eqn2 ] and [ eqn3 ] into equation [ eqn1 ] , then we obtain a transformed received signal : @xmath10 * the definition of transformed auto - correlation is the cross - correlation between received signal @xmath1 and its transformation @xmath11 , i.e. * , @xmath12 if the parameters are correct , then after transformation , the reflected signal is coherent to transmitted signal now , with zero delay .
so , only @xmath13 is important .
we can calculate all @xmath13 in a given parametric space .
the significant peaks of @xmath13 may indicate possible reflectors existed nearby the central transmitter , and the true parameters of these reflectors cab be estimated at the same time .
here , an one - dimension simulation is used as an example , where a transmitter is fixed at @xmath14 , and several reflectors are moving or fixed in x axis .
the transmitted signal , generated by convolving a gaussian white noise with a gaussian point spread function ( @xmath15 ) , is stochastic and band - limited .
the sampling interval is @xmath16 . and 2 , 048
, 000 sampled data are generated .
four reflectors are set in the simulation .
the model parameters including initial position , velocity and reflectance are listed in table [ tab1 ] , columns 2 - 4 .
.the parameters of the reflectors .
columns 2 - 4 are values of a model .
columns 5 - 7 are estimated values by transformed auto - correlation . [ cols="^,^,^,^,^,^,^ " , ] a part of received signal and its auto - correlation function are displayed in figure [ fig1 ] . as shown in the auto - correlation function , although there are four reflectors , only one reflector , which is fixed at position of 0.9 light second , has a fringe .
other reflectors have variable delays , so the reflected signals are incoherent to the transmitted signals , and no fringes exist .
now , it is able to search the possible reflectors and their parameters by transformed auto - correlation . here , @xmath17 is normalized by @xmath18
. thus the intensity of the peak of a reflector is equal to reflectance @xmath3 .
the generated map in parametric space ( initial position and velocity ) , which is called reflectance map , is shown in figure [ fig2 ] .
it is clearly seen that all of four reflectors have been detected .
the estimated parameters ( initial position , velocity and reflectance ) are listed in table [ tab1 ] , columns 5 - 7 .
the standard deviation of the background in reflectance map is about 0.0036 .
transformed auto - correlation is able to detect reflected signals when there are constant or variable delays . with a priori reflecting model , the parameters ( such as position , velocity etc .. ) of reflectors can also be estimated . | echo signals exist in light curves of astronomical objects supernovae rest 2005 krause 2008a x - ray binaries greiner 2001 agns perterson 1993 if location between central transmitter and reflecting object fixed auto - correlation edelson krolik 1988 can to detect reflected signal estimate distance nearby reflecting object moving perterson 2008 delay between transmitted and reflected signal variable coherence broken no echo signals detected by auto - correlation method transformed auto - correlation aim to rebuild coherence perform cross - correlation between reflected transmitted signals one transmitter one reflector @xmath0 is transmitted signal @xmath1 received signal astronomical objects far away received light curve is combination of transmitted and reflected signals @xmath2 @xmath3 is reflectance @xmath4 is variable delay model dependent @xmath5 are parameters of model due to variable delay reflected signal coherence to transmitted impossible to detect reflected signal by auto - correlation transformation needed to rebuild coherence set @xmath6 where @xmath7 is inversion function of @xmath8 @xmath9 applying equationeqn2 eqn3 into equation eqn1 obtain transformed received signal @xmath10 definition transformed auto - correlation cross - correlation between received signal @xmath1 transformation @xmath11 @xmath12 parameters correct after transformation reflected signal coherent to transmitted signal zero delay @xmath13 important calculate all @xmath13 parametric space significant peaks @xmath13 indicate possible reflectors nearby central transmitter true parameters estimated one - dimension simulation transmitter fixed at @xmath14 reflectors moving in x axis transmitted signal gaussian white noise gaussian point spread function @xmath15 stochastic band - limited sampling interval @xmath16 2 , 048 , 000 sampled data generated four reflectors set model parameters initial position velocity reflectance listed in table tab1 columns 2 - 4 values model 5 - 7 estimated values by transformed auto - correlation received signal auto - correlation function displayed in figure fig1 four reflectors one fixed 0.9 light second has fringe other reflectors variable delays reflected signals incoherent to transmitted signals no fringesreflectors parameters by transformed auto - correlation @xmath17 normalized by @xmath18 intensity of peak reflector equal to reflectance @xmath3 generated map reflectance map shown in figure fig2 four reflectors detected estimated parameters velocity reflectance listed in table [ tab1 ] columns 5 - 7 standard deviation of background in reflectance map about 0.0036 transformed auto - correlation reflected signals constant or variable delays with priori reflecting model parameters position velocity of reflectors estimated | 0.182423 | 1,062 | 529 | , Saving $0.0 in GPT-4. |
154,993 | the new nhmfl 60 t quasi - continuous magnet produces a flat - top field for a period of 100 ms at 60 tesla , and for longer time at lower fields , e.g. 0.5 s at 35 tesla .
we have developed for the first time the capability to measure heat capacity at very high magnetic fields in the nhmfl 60 t quasi - continuous magnet at lanl , using a probe built out of various plastic materials .
the field plateau allows us to utilize a heat - pulse method to obtain heat capacity data .
proof - of - principle heat capacity experiments were performed on a variety of correlated electron systems .
both magnet performance characteristics and physical properties of various materials studied hold out a promise of wide application of this new tool .
# 1#2#3#4#1 * # 2 * , # 3 ( # 4 ) | the 60 tesla long - pulse ( 60tlp ) magnet was recently commissioned at the los alamos national laboratory .
this magnet produces a flat - top field for a period of 100 ms at 60 tesla , and for longer time at lower fields , e.g. 0.5 s at 35 tesla . during the entire pulse ,
the magnetic field varies at a maximum ramp rate of @xmath0 tesla / sec .
together , these properties allow for the development of brand new tools to study materials in pulsed magnetic fields .
heat capacity measurement is one of these tools .
we have built a probe made of plastic materials that allows us to perform heat capacity measurements at temperatures between 1.6 k and 20 k in fields up to 60 tesla . to maximize the available experimental space a novel vacuum tapered seal was developed .
the conical plug part of the joint is made out of g-10 , and the matching vacuum can is made out of 1266 stycast epoxy .
the differential thermal contraction between the parts aids in producing a superfluid - tight joint .
the simple construction of the joint resulted in a 16 mm diameter experimental region .
the main parts of the probe inside the vacuum can are the temperature regulated block ( trb ) and the silicon heat capacity platform with sample , thin film resistive heater , and bare chip cernox thermometer .
the trb is made of 2850 stycast epoxy chosen for its fair thermal conductivity , and is thermally connected to the bath via two dozen thin ( gauge 44 and 38 ) 4-inch long copper wires .
the heat capacity platform is suspended with nylon strings , and electrical connections to thermometer and heater provide a thermal link as well .
the resulting thermal equilibrium time constants for the trb and the platform were measured to be on the order of few minutes .
therefore , during the magnetic field pulse , which lasts for about 2 seconds , both the trb and the heat capacity platform can be regarded as thermally isolated from the bath and each other , and under adiabatic condition . the third time constant @xmath1 is that of the heat capacity stage including sample , platform , thermometer , and heater . the sample s internal thermal relaxation time constant @xmath2 can be less than millisecond at a temperature of a few kelvin .
however it grows rapidly as the temperature is increased . at the low temperature end @xmath1
can increase substantially due to either increase in @xmath2 ( electronic or nuclear magnetic entropy ) or boundary thermal resistance between different constituents of the stage .
the temperature interval between 1 k and 20 k is therefore a convenient starting point for developing heat capacity measurements in pulsed magnetic fields .
we use a heat pulse method to measure heat capacity , where a known amount of heat is delivered to the sample using a chip resistor as a heater element .
the heat capacity stage must come to equilibrium both before and after the heat pulse is delivered while the magnetic field remains constant .
the flat field plateau of the 60tlp magnet allows this to occur .
the temperature of the stage is measured with a cernox chip resistance thermometer , which was calibrated in both dc field up to 30 tesla and pulsed fields up to 60 tesla .
the heat capacity of the sample is then determined as the ratio of the heat delivered to the sample to the change in its temperature .
the low ramp rate of the long - pulse magnet ( in comparison with short pulse , capacitively driven magnet with the total magnet pulse time of about 10 ms ) reduces the unavoidable eddy current heating in metallic samples . however , during magnetic field sweep the temperature does not stay constant even in the total absence of eddy current heating , due to the magnetocaloric effect .
the heat capacity stage is thermally isolated from the bath , and remains in adiabatic condition during the magnetic field pulse .
the dependence of the temperature of the stage on the magnetic field during the pulse is then given by the expression @xcite @xmath3 where t is temperature , h is magnetic field , m is magnetization , and c is the specific heat of the sample , and subscripts @xmath4 and @xmath5 indicate constant entropy and magnetic field , respectively .
( [ eq : dtdh ] ) is used to describe the process of adiabatic demagnization cooling , where @xmath6 is negative and therefore @xmath7 is positive .
such a system warms as the field is ramped up , and cools during the ramp down portion of the magnetic field pulse _ reversibly_. however , magnetization in general can also increase with temperature .
the sample then would cool during the ramp up , and warm _ reversibly _ during the ramp down of the magnetic field .
below we show examples of both types of behavior .
the first heat capacity experiments in the 60tlp magnet were performed on a single crystal of metallic ybincu@xmath8 , grown from an in - cu flux as described previously .
@xcite this system undergoes a first order valence transition at 42 k in zero field .
the specific volume is increased by 0.5% upon cooling through the phase transition , @xcite with accompanying rise in the kondo temperature @xmath9 from 25 k to 500 k. @xcite it is believed that unlike in the case of ce , where the phase transition is described within a kondo - collapse scenario , the valence transition ybincu@xmath8 it is driven by the band structure effects . @xcite the complete magnetic field - temperature phase diagram was obtained in dc bitter magnets at nhmfl / tallahassee and in capacitor - driven pulsed magnets at nhmfl / los alamos .
@xcite this work showed that the transition can be suppressed down to t = 0 k with an applied field of 34.3 tesla .
the length of the flat top can be close to 0.5 s in the 60tlp magnet for magnetic fields less than or equal to 35 tesla .
when @xmath1 is much smaller than the length of the plateau , a sequence of heat pulses can be delivered to it within the flat portion of the field profile , with sufficient time for the calorimeter to come to equilibrium before and after each of the heat pulses .
this situation is illustrated by fig .
[ train](a ) for a field of 20 tesla , where a sequence of five 10 ms long heat pulses were delivered to the heat capacity stage during the plateau of a single magnetic field pulse .
the thermometer comes to equilibrium after the heat pulse is delivered well before the next heat pulse , and temperature is determined before and after each of the pulses . in this way a series of five c@xmath10(t )
data points is collected in a single shot experiment , as the initial temperature for each of the heat capacity experiments is increased due to the previous heat pulse . the data from this and one zero field
shot is shown in fig .
[ train](b ) .
we fit the data with a sum of t - linear ( electronic ) and t - cubic ( phononic ) terms @xmath11 . for zero field
we obtain @xmath12 @xmath13 and @xmath14 @xmath15 . the value of a is in excellent agreement with available data in the literature .
@xcite at 20 tesla we obtain @xmath16 @xmath17 and @xmath18 @xmath19 .
the magnitude of the cubic term due to phonons is field - independent as expected .
the increase of the linear term with field is likely due to scaling with magnetic field observed for various properties of ybincu@xmath20 .
@xcite another way to increase the data acquisition rate relies on the programmable nature of the 60tlp magnet .
a series of plateaus at different magnetic fields can be produced within a single experiment .
[ multi ] displays the data for one such experiment on ybincu@xmath20 , with four plateaus at 25 , 30 , 35 , and 40 tesla , each 130 ms long . at each of the magnetic field
plateaus the heat pulse is applied to the stage , and heat capacity experiment is performed .
in addition , as the field was changed between 30 and 35 tesla through the first order phase boundary , the temperature of the sample was observed to go down on the up - sweep , and up on the down - sweep , due to the magnetocaloric effect .
these features are very sharp , and allow direct determination of the phase diagram of ybincu@xmath20 .
this is yet another complementary way to collect data using the heat capacity apparatus .
it was not possible to measure heat capacity in the high field phase due to a large increase in @xmath1 .
preliminary measurements of heat capacity of ube@xmath21 and ce@xmath22bi@xmath20pt@xmath23 were performed up to 60 tesla .
[ mc2 ] shows temperature variation of ube@xmath21 and ce@xmath22bi@xmath20pt@xmath22 as a function of field during magnetic field pulses to 60 t. this figure illustrates the magnetocaloric effect under adiabatic conditions of our apparatus .
the temperature of the ube@xmath24 sample increases from 4 k up to 10 k during the ramp up , and returnes to 4 k during the ramp down in a very reversible fashion .
the opposite is true for ce@xmath22bi@xmath20pt@xmath22 , which is colder at 60 t than at 0 t. with available magnetization data it should be possible to calculate the specific heat of these compounds during the ramp portions of the field pulse , providing information in addition to the direct heat capacity measurements at field plateaus .
we have demonstrated the feasibility of heat capacity measurements in the pulsed magnetic fields provided with the 60 tesla long pulse magnet at nhmfl / lanl .
direct measurement of heat capacity at field plateaus was clearly demonstrated for a variety of compounds .
it appears that thermal equilibrium can be achieved in some compounds even during the field sweep , given the low ramp rates which the 60tlp magnet is capable of achieving . in this situation
specific heat data can be obtained via the magnetocaloric effect from the temperature vs. field traces and magnetization data for such compounds .
other types of thermal relaxation - related experiments like thermal conductivity and seebeck effect are also under development .
this work was conducted under the auspices of the department of energy .
it was also supported by the in - house research program of the nhmfl . | 60 tesla long - pulse magnet commissioned at los alamos national laboratory produces flat - top field 100 ms at 60 tesla longer at lower fields 0.5 s at 35 tesla magnetic field varies maximum rate @xmath0 tesla / sec properties allow development new tools study materials pulsed magnetic fields heat capacity measurement built probe plastic heat capacity measurements at temperatures 1.6 k and 20 k up to 60 tesla novel vacuum tapered seal developed conical plug g-10 vacuum can 1266 stycast epoxy differential thermal contraction superfluid - tight joint 16 mm diameter experimental region main parts temperature regulated block trb silicon heat capacity platform with sample thin film resistive heater bare chip cernox thermometer trb of 2850 stycast epoxy connected to bath via 4-inch copper wires heat capacity platform suspended with nylon strings electrical connections to thermometer heater provide thermal link thermal equilibrium time constants for trb platform measured few minutes during magnetic field pulse 2 seconds trb thermally isolated from bath under adiabatic conditionthird time constant @xmath1 heat capacity stage sample platform thermometer heater sample s internal thermal relaxation @xmath2 less than millisecond at kelvin grows as temperature low temperature @xmath1 due to increase @xmath2 magnetic entropy or boundary thermal resistance constituents temperature interval between 1 k and 20 k starting point for heat capacity measurements in pulsed magnetic fields heat pulse method heat capacity heat delivered to sample chip resistor heater element heat capacity stage equilibrium before after pulse magnetic field constant flat field plateau 60tlp magnet allows temperature measured with cernox chip resistance thermometer calibrated in dc field 30 tesla pulsed fields 60 tesla heat capacity determined as ratio heat delivered to change temperature low ramp rate of long - pulse magnet reduces eddy current heating in metallic samples during magnetic field sweep temperature constant due to magnetocaloric effect heat capacity stage thermally isolated from bath remains in adiabatic condition during magnetic field pulsedependence temperature stage on magnetic field during pulse given by expression @xmath3 t temperature h magnetic field m magnetization c specific heat sample @xmath4 @xmath5 indicate constant entropy magnetic field adiabatic demagnization cooling @xmath6 negative @xmath7 positive system warms as field ramped up cools during down magnetization with temperature sample during ramp up warm during ramp down examples behavior first heat capacity experiments in 60tlp magnet performed on single crystal metallic ybincu@xmath8 from in - cu flux system undergoes valence transition at 42 k in zero field volume increased by 0.5% upon cooling rise in kondo temperature from 25 k to 500 k valence transition driven by band structure effects complete magnetic field - temperature phase diagram obtained in dc bitter magnets at nhmfl / tallahassee capacitor - driven pulsed magnets at nhmfl / los alamos transition suppressed to t = 0 k with applied field 34.3 tesla length flat top close to 0.5 s in 60tlp magnet for magnetic fields less than 35 tesla@xmath1 smaller than plateau heat pulses delivered within flat field profile sufficient time for calorimeter after illustrated by fig .(a ) for field 20 tesla five 10 ms heat pulses delivered heat capacity stage during single magnetic field pulse thermometer equilibrium after temperature determined before after five c@xmath10(t ) data points collected in single shot experiment initial temperature increased due to previous heat pulse data from zero field shot shown in fig . train](b ) data with t - linear electronic t - cubic phononic terms @xmath11 for zero field obtain @xmath12 @xmath13 @xmath14 @xmath15 value a with data at 20 tesla obtain @xmath16 @xmath17 magnitude cubic term due to phonons field - independent increase linear term with field due to scaling with magnetic field ybincu@xmath20 increase data acquisition rate programmable nature 60tlp magnet plateaus at different magnetic fields produced within single experiment displays data experiment on ybincu@xmath20 four plateaus at 25 30 35 40 tesla each 130 ms longmagnetic field plateaus heat pulse applied to stage heat capacity experiment performed field changed between 30 35 tesla phase temperature sample down up down due to magnetocaloric effect features allow determination phase diagram ybincu@xmath20 complementary collect data heat capacity apparatus measure heat capacity high field phase increase in @xmath1 preliminary measurements heat capacity ube@xmath21 performed up to 60 tesla mc2 shows temperature variation of ube function field during magnetic field pulses to 60 t figure illustrates magnetocaloric effect under adiabatic conditions temperature ube@xmath24 sample increases from 4 to 10 k ramp up returnes to 4 k ramp down opposite for ce colder at 60 t than 0 t data calculate specific heat compounds during ramp portions field pulse heat capacity measurements plateaus demonstrated feasibility heat capacity measurements in pulsed magnetic fields with 60 tesla long pulse magnet nhmfl / lanl direct measurement heat capacity at field plateaus demonstrated for variety compounds thermal equilibrium in some compounds during field sweep low ramp rates 60tlp magnetheat data obtained magnetocaloric effect temperature field traces magnetization data compounds thermal relaxation experiments thermal conductivity seebeck effect under development work conducted department of energy supported by in - house research program nhmfl | 0.353193 | 2,269 | 1,127 | , Saving $0.1 in GPT-4. |
154,996 | we invert directly the redshift - luminosity distribution of observed long _ swift _ grbs to obtain their rate and luminosity function . our best fit rate is described by a broken power law that rises like @xmath0 for @xmath1 and decrease like @xmath2 for @xmath3 .
the local rate is @xmath4 $ ] .
the luminosity function is well described by a broken power law with a break at @xmath5 $ ] and with indices @xmath6 and @xmath7 .
the recently detected grb 090423 , with redshift @xmath8 , fits nicely into the model s prediction , verifying that we are allowed to extend our results to high redshifts . while there is a possible agreement with the star formation rate ( sfr ) for @xmath9 ,
the high redshift slope is shallower than the steep decline in the sfr for @xmath10 .
however we can not rule out a grb rate that follows one of the recent sfr models .
[ firstpage ] | gamma - ray bursts ( grbs ) are short and intense pulses of soft @xmath11-rays . in this work
we study their luminosity function and their cosmic rate .
these functions are essential to understand the nature of grbs and to determine their progenitors
. they may also shed light on the still mysterious physics of the central engine .
the luminosity function and the cosmic grb rate are observationally entangled as the observed rate is a convolution of the luminosity function with the cosmic rate .
for this reason , almost every work before this paper have made an a - priori assumptions on at least one of these functions . at first - having no motivation to believe otherwise -
the rate was assumed to be constant .
the simplest form for the luminosity function was a standard candle i.e. a constant luminosity .
these early studies used the measured value to find the typical luminosity @xcite and correspondingly a maximal distance from which grbs were observed . later , using the flux distribution ( @xmath12 relation ) , @xcite and @xcite showed how relaxing the standard candle assumption allows a corresponding relaxation of the constant rate assumption .
@xcite noticed that the hosts of the bursts are in star forming regions and suggested that grbs follow the sfr ( see also * ? ? ?
* ; * ? ? ?
the detection of a supernova associated with grb980425 @xcite , strengthened the expectation that the grb rate should follow the sfr . using this proportionality
, numerous studies examined the typical luminosity assuming at first standard candles ( e.g. * ? ? ?
* ; * ? ? ?
* ) and later more elaborate shapes of the luminosity function ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
_ swift _ that discovers routinely grbs afterglows detected grbs from higher redshifts than was previously possible sparked renewed interest in the grb redshift distribution ( e.g. * ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
more and more signs appeared suggesting that the rate of grbs does not simply follow the global star formation rate , as was believed earlier .
@xcite conclude that the grb rate differs from the sfr .
in particular we observe more high redshift bursts than what is expected for a rate that follows the sfr .
@xcite uses the high luminosity subsample of bursts to obtain the grb rate without assuming any luminosity function .
they find that the grb rate at high redshifts ( @xmath13 ) is higher on than @xcite sfr .
alternatively luminosity evolution , has been suggested by some authors : ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
@xcite used a sample of long bursts with redshift and found evidence for luminosity evolution .
other papers , including this one , find self consistency without a luminosity evolution .
we introduce here a new method for determining the rate and the luminosity , by inverting the observations without making any assumptions on the functional form of the luminosity and rate functions .
this direct inversion allows us to use most of the available redshift data and to obtain robust estimation of both functions . in [ sec :
sample ] we describe the sub - sample of bursts we analyze and its advantages over other samples . in [ sec : direct ] we introduce a new formalism allowing us to directly invert the observed distribution and obtain the intrinsic luminosity function and rate . in
[ sec : results ] we present the results obtained using this method and the confidence ranges associated with them . in [ sec : contosfr ] we address the question whether the grb rate is consistent with the sfr .
we discuss a few implications of the results in [ sec : summary ] .
we consider long bursts ( @xmath14 ) detected by _ swift _ from the beginning of its operation until burst 090726 , information page @xmath15 with a measured peak - flux and a measured redshift .
generally grbs redshifts are obtained from the optical afterglow spectrum using absorption lines or photometry , or from the spectrum of the host galaxy using emission lines .
however in the global sample we find different redshift distributions for the different detection methods , namely : absorption , emission and photometry ( see figure [ fig : zmethodsdist ] ) . for redshifts determined using the hosts emission lines , we do not detect high - redshift events , whereas absorption lines redshifts extend over the entire range of redshifts .
furthermore , emission lines are more susceptible to a selection effect known as the redshift desert in the range @xmath16 , ] ( @xcite , see also @xcite ) .
this is in line with the fact that we have also found that the probability to measure the redshift using emission lines strongly depends on the gamma ray flux , favoring high fluxes .
this effect is mild for absorption lines and photometry ( see appendix b ) . to obtain a more uniform sample we consider therefore only bursts whose
redshift was measured using the afterglow . for each burst
we calculate the isotropic equivalent peak - luminosity : @xmath17 ( see appendix a. for details ) using the peak - flux and redshift .
we use standard @xmath18 cosmology with @xmath19 .
we assume that the luminosity function is redshift independent . in this case
the number of bursts at a given redshift and with a given luminosity is the product of the luminosity function , @xmath20 , that depends only on the luminosity @xmath21 and the grb rate , @xmath22 , that depends only on the redshift @xmath23 .
this common assumption is reasonable since a - priori there is no competing reason why the luminosity function should depend on the redshift .
we test later the validity of this assumption and show that it is accepted with a high statistical significance .
the isotropic peak luminosity function , @xmath20 , is defined traditionally as the fraction of grbs with isotropic equivalent luminosities in the interval @xmath24 and @xmath25 .
the rate , @xmath22 , is defined as the co - moving space density of grbs in the interval @xmath23 to @xmath26 .
the distribution density : @xmath27 is given by : @xmath28 where @xmath29 is the differential co - moving rate of bursts at a redshift z , @xmath30 is the derivative of the volume element and the factor @xmath31 reflects the cosmological time dilation .
we consider now a direct method to invert the observed @xmath32 distribution and obtain the functions @xmath33 and @xmath34 .
to do so we approximate @xmath20 and @xmath35 as @xmath36 whose range is divided to bins with a constant value within each bin .
these functions can be expressed as a sum of heaviside functions .
first , we divide both luminosity and redshift intervals into bins .
we denote @xmath37 @xmath38 we also define the _ weights factors _ @xmath39 as the _ probability for detecting a burst with a measured redshift
z and luminosity l _ where @xmath40 is the probability to detect and measure redshift for a burst with a luminosity @xmath21 at a redshift @xmath23 ( see appendix b ) .
we denote the observed number of events per bin as : @xmath41 @xmath42 @xmath43 @xmath44 next , we determine @xmath45 and @xmath46 and the error estimates , using the maximum - likelihood formalism .
we define @xmath47 the ( logarithmic ) likelihood of the model given the observations as : @xmath48
- n \ln [ \sum_{ij } \phi_i r_j w_{ij } ] \ .\ ] ] at the maximum all partial derivatives of @xmath47 with respect to @xmath45 and with respect to @xmath46 vanish , leading to @xmath49 and @xmath50 notice that the second term on the right hand side of each of the equations [ eqn : rate1 ] and [ eqn : lumf ] is the same normalization factor .
we have a set of non - linear equations with as many equations as variables .
we solve numerically these non linear equations using successive iterations until convergence .
a - priori it is not clear whether @xmath33 and @xmath51 are uniquely determined and whether there is a solution at all .
however , we find good convergence . we have examined a large set ( @xmath52 ) of initial guesses where each component was randomly drawn from a uniform distribution .
we found a rapid convergence to a unique solution for all initial guesses , all reaching the requested accuracy of @xmath53 with less than 25 iterations .
we thus conclude that the existence of other stable solutions is very unlikely .
we approximate the error as the value for which m deviate by -1 from it s maximum : ( i.e. the likelihood is smaller by a factor @xmath54 ) .
this reflects an @xmath55 error for a normal - distribution . @xmath56
@xmath57 the two solutions , i.e. the positive one and the negative one , give an upper and a lower bounds on the error respectively . for small deviations we can approximate the error using the second derivatives of m : @xmath58 @xmath59 to estimate the uncertainty induced by the specific bins choice , we preform all the analysis for a 1/2 unit redshift and @xmath60 binning and repeat for a 1/3 unit binning ( where all bins widths are 1/3 unit , except the last two redshift bins which we can not change because they contain too few data points ) . in the following ,
unless otherwise stated , we use the 1/2 unit binning for all further analysis .
clearly , the results with different binning are slightly different , but they are all within each other s error range . when we include the uncertainty induced by the binning the error ranges
become only slightly wider .
so far , we have not assumed any functional form for the luminosity function or for the rate as our method does not require such an assumption .
raw `` results are depicted in figures [ fig : rate ] and [ fig : luminosity ] . later on
we will compare these step functions with models for the grb rate that follow the sfr .
however , in order to easily characterized the result we need to approximate our raw '' step functions with simple functional forms . therefore ,
after obtaining the results in the form of step functions we approximate them in term of broken power laws : @xmath61 @xmath62 we obtain the parameters of the best fit functions by minimizing the @xmath63 values .
upper frame : the resulting step function and the best fit model : a broken power law with indices @xmath64 , @xmath65 and with a break at @xmath66 ( solid line ) .
the @xmath63 values for the models is 2.3 at 10 d.o.f . giving a rejection probability of 0.007 .
lower frame : the number of detected bursts for each redshift bin and the rates expected from the model fitted above .
the two upper boxes in each column represent the statistical error range .
[ fig : rate ] upper frame : the resulting step function and the best fit model for the luminosity function : a broken power law with a break at @xmath67 , a low luminosity index @xmath68 and a high luminosity index @xmath69 .
the @xmath63 value for this model is 0.63 at 4 d.o.f . , giving a rejection probability of 0.04 .
lower frame : the number of detected bursts for each luminosity bin and the rates expected from the model fitted above .
the two upper boxes in each column , represents the statistical error range .
[ fig : luminosity ] the best fit functions are shown ( together with the step functions ) in figures [ fig : rate ] and [ fig : luminosity ] .
the parameters of the best fit models are summarized in table [ tbl : mc2 ] . to estimate the statistical errors involved in the parameter
estimates we preformed a monte - carlo simulation . in this simulation
we use the model with the best fit parameters to draw random sets of data ( with same size as the real sample ) .
we then carry out the analysis on this mock data sets and obtain new best fit parameters . repeating this process many times
we obtain a scatter of points in the parameters plane around the original best fit parameters .
the central @xmath70 and @xmath71 ranges for each of the parameters separately , are also shown in table [ tbl : mc2 ] .
for both corrections , we find a non - negligible effect on the results , although the deviations induced on the parameters are in most cases within the statistical error ranges of the analysis . when taking both effects into account , the models give somewhat better results in the various statistical tests .
therefor , the function @xmath72 that take both effects into account is used in this work . | gamma - ray bursts ( grbs ) are short intense pulses of soft @xmath11-rays we study their luminosity function cosmic rate functions essential to understand nature grbs determine progenitors shed light on physics central engine luminosity function cosmic grb rate observationally entangled observed rate is convolution of luminosity function with cosmic rate work before paper made a - priori assumptions on functions first rate was assumed to constant simplest form for luminosity function was standard candle constant luminosity early studies used measured value to find typical luminosity maximal distance from grbs observed later flux distribution ( @xmath12 relation ) @xcite showed relaxing standard candle assumption allows relaxation of constant rate assumption @xcite noticed hosts of bursts in star forming regions suggested grbs follow sfr detection of supernova associated with grb980425 @xcite strengthened expectation grb rate should follow sfr studies examined typical luminosity assuming standard candles elaborate shapes of luminosity function afterglows detected grbs from higher redshifts sparked interest in grb redshift distribution* ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? * ; * ? ? * ; * ? ? * ; * ? ? ? * ; * ? ? ? signs rate grbs follow global star formation rate @xcite grb rate differs from sfr more high redshift bursts than @xcite uses high luminosity subsample grb rate without assuming luminosity function grb rate at high redshifts higher than @xcite sfr luminosity evolution suggested by authors ? @xcite used sample long bursts redshift found evidence for luminosity evolution other papers find consistency without luminosity evolution introduce new method for determining rate luminosity inverting observations without assumptions luminosity rate direct inversion redshift data robust estimation of functions sec : sample ] sub - sample advantages over direct ] new formalism invert observed distribution obtain intrinsic luminosity function and rate results ] present results confidence ranges contosfr grb rate consistent with sfr discuss implications results in : summary ]long bursts @xmath14 ) detected by swift until burst 090726 @xmath15 measured peak - flux redshift grbs redshifts obtained from optical afterglow spectrum absorption lines photometry host galaxy emission lines global sample different redshift distributions for detection methods absorption emission photometry figure redshifts hosts emission lines detect high - redshift events absorption lines redshifts extend range emission lines susceptible to selection redshift desert in range @xmath16 probability to measure redshift using emission lines depends on gamma ray flux favoring high fluxes effect mild for absorption lines photometry appendix b uniform sample consider bursts redshift measured using afterglow calculate isotropic equivalent peak - luminosity @xmath17 appendix a. using peak - flux redshift use standard @xmath18 cosmology with @xmath19 luminosity function redshift independent number of bursts at given redshift luminosity product of luminosity function depends on luminosity @xmath21 grb rate @xmath22 redshift @xmath23 assumption reasonable no reason luminosity function redshift test validity high statistical significance isotropic peak luminosity function @xmath20 defined fraction grbs isotropic equivalent luminosities @xmath24 @xmath25 rate @xmath22 co - moving space density of grbs @xmath23 to distribution density @xmath27 given by @xmath28 @xmath29 differential co - moving rate bursts redshift @xmath30 derivative volume element reflects cosmological time dilation invert @xmath32 distribution obtain functions @xmath33 @xmath34 approximate @xmath20 @xmath35 as @xmath36 range divided to bins constant value functions expressed sum of heaviside functions divide luminosity redshift intervals into bins denote @xmath37 @xmath38 define @xmath39 probability for detecting burst redshift @xmath40 probability detect redshift burst luminosity @xmath21 redshift @xmath23 denote observed number of events per bin as @xmath41 @xmath42 @xmath44 determine @xmath45 @xmath46 error estimates maximum - likelihood formalismdefine @xmath47 logarithmic likelihood model observations @xmath48 - n \ln [ \sum_{ij } \phi_i r_j w_{ij } ] maximum partial derivatives @xmath47 @xmath45 @xmath46 vanish leading to @xmath49 @xmath50 second term right side equations eqn : rate1 : lumf ] same normalization factor set non - linear equations many variables solve numerically iterations until convergence not clear @xmath33 @xmath51 uniquely determined solution find good convergence examined large set @xmath52 initial guesses randomly drawn uniform distribution rapid convergence unique solution guesses reaching accuracy @xmath53 less than 25 iterations conclude other stable solutions unlikely approximate error as value m deviate by -1 from maximum likelihood smaller factor @xmath54 reflects @xmath55 error normal - distribution @xmath56 @xmath57 two solutions positive negative give upper lower bounds errorsmall deviations approximate error using second derivatives m : @xmath58 @xmath59 estimate uncertainty specific bins choice preform analysis for 1/2 unit redshift @xmath60 binning repeat 1/3 unit binning all bins widths 1/3 unit except last two redshift bins change few data points use 1/2 unit binning for further analysis results with different binning slightly different within error range include uncertainty binning error ranges slightly wider not assumed functional form for luminosity function rate method require raw results depicted in figures rate luminosity compare step functions with models grb rate need approximate raw step functions with simple functional forms results approximate broken power laws : @xmath61 @xmath62 obtain parameters best fit functions minimizing @xmath63 values resulting step function best fit model broken power law with indices @xmath64 @xmath65 break at @xmath66 @xmath63 values 2.3 at 10 d.o.f. rejection probability 0.007 lower number of detected bursts for each redshift bin rates expected from model two upper boxes represent statistical error rangefig : rate ] upper frame step function best fit model for luminosity function broken power law break at @xmath67 low luminosity index @xmath68 high luminosity index @xmath69 @xmath63 value 0.63 at 4 d.o.f rejection probability 0.04 lower frame number detected bursts each luminosity bin rates expected from model two upper boxes each column statistical error range fig luminosity best fit functions shown step functions in figures fig rate luminosity parameters best fit models summarized in table [ tbl : mc2 ] estimate statistical errors monte - carlo simulation model best fit parameters draw random sets data same size analysis mock data obtain new best fit parameters scatter of points in parameters plane around original best fit parameters central @xmath70 @xmath71 ranges shown in table tbl mc2 both corrections non - negligible effect on results deviations within statistical error ranges both effects models give better results in statistical tests function @xmath72 both effects used | 0.441563 | 2,887 | 1,454 | , Saving $0.1 in GPT-4. |
155,004 | an exploratory search for an anomalous energy deposition by neutrinos in a germanium crystal was performed in the cern high energy @xmath0 beam . no signal was found and a limit is set at a level of about @xmath1 of the normal de / dx for a minimum ionizing particle . | we present here the result of a first test looking for energy deposition of neutrinos in matter , other than via weak interactions .
this test consisted in exposing a high purity germanium crystal to the cern high energy @xmath0 beam .
such a device offers an easy - to - use , high - sensitivity active medium well suited for an exploratory search . with the very intense neutrino fluxes available in a beam , it is possible to achieve a meaningful limit even if each neutrino deposits only a very small amount of energy .
if detected , neutrino energy losses in matter would have important consequences in several astrophysical contexts .
in particular , an energy loss of neutrinos in the interior of the sun could explain the various deficits observed in solar neutrino experiments .
the new super - kamiokande measurement of the solar neutrino spectrum @xcite excludes a relative energy loss greater than a few per cent for 10 @xmath2 neutrinos traversing 600000 @xmath3 of matter with an average density of 1 @xmath4 .
the result discussed here reaches a limit lower by two orders of magnitude , when one assumes an energy deposition proportional to the initial neutrino energy .
it is also relevant in the case of energy transfers in a supernova explosion .
we used a high - purity germanium ( hpge ) coaxial diode from eurisys mesures@xcite with an effective volume of 140 @xmath5 cooled at liquid nitrogen temperature .
germanium is a semiconductor having an intrinsic band gap of 0.74 @xmath6 and a mean electron - hole pair creation work of 2.9 @xmath6 .
this device gives the lowest energy threshold that can be obtained in an active target of sufficient mass without requiring heavy cryogenic environment .
the detector was installed in the cern high - energy wide band neutrino beam on a platform between the chorus and nomad oscillation experiments , 800 @xmath7 away from the primary beryllium target . the beam is essentially composed of @xmath0 of average energy 24 @xmath8 , with a 1% contamination of @xmath9 .
it is produced in two spills of 6 @xmath10 duration extracted at the beginning and the end of the 2.2 @xmath11 so - called flat top of the sps accelerator .
the machine cycle is repeated every 14.4 @xmath11 .
synchronisation signals came from the nomad experiment , defining 4 gates ( we keep nomad s nomenclature ) : -calib gate , @xmath12 before the first neutrino spill , -nu1 and nu2 gates , covering the neutrino spills , -mu gate , during the flat - top in between the neutrino spills .
if the neutrino energy loss was caused by rare but individually detectable effects ( i.e. releasing more than a few @xmath13 per ` interaction ' ) , it would be impossible to disentangle them from ordinary background ( such as charged particles correlated with the neutrino beam or neutrinos weak interactions ) .
thus , for this first test , we concentrated on hypothetical processes where neutrinos would each leave a tiny amount of energy , the neutrino flux resulting in a global increase of the leakage current during the @xmath14 accelerator extraction duration .
as a consequence , the read - out was carried out by an infinite time constant charge - preamplifier , dc - coupled to the hpge diode . the output of the preamplifier was sampled and digitized by a sigma - delta adc . on each gate ,
the sigma - delta loop was reset , and sampling points were accumulated during @xmath15 , constituting a ` row ' .
this arrangement allowed to integrate the energy deposition during the spill , whilst keeping a sufficient resolution on impulsional energy deposition ( @xmath16 fwhm on the @xmath17 @xmath18 line ) , thus allowing rough spectrometric monitoring and identification of photon or charged particle interactions in the crystal .
a p - type ge crystal was chosen , the peripheric li - doped zone giving some protection against alphas and low - energy beta and gamma radiations from natural radioactivity .
this protection was enforced by a @xmath19 thick lead shield surrounding the detector .
the whole set - up was supported by 4 pneumatic insulators .
special care was taken to enforce electromagnetic compatibility ( emc ) .
electron - hole pairs are produced when energy is deposited in the crystal , thus creating a current through the diode .
the leakage current was measured to be @xmath20 after a 24-hour warm - up period .
when an ionizing particle crosses the crystal , it gives a fast pulse of @xmath21 duration .
typically , a minimum ionizing particle crossing the whole thickness of the diode ( 5 @xmath22 ) deposits 30 @xmath2 .
as explained in the previous section , the expected signal behaves very differently : we are looking for a continuous current excess , building up during the neutrino spill and proportional to the neutrino flux ( see figure [ rows ] ) .
the first step in the analysis is to identify and reject rows contaminated by photons or charged particles in the crystal .
the discriminating cut is set to @xmath23 over at least two sampling points .
this results in rejecting 41% of the rows in the nu1 gate , 38% in the nu2 gate and 12% in the calib and mu gates .
the time distribution of these rejected energy depositions inside the @xmath24 nu1 or nu2 gates clearly exhibits the neutrino pulse time structure ( figure [ beamprof ] ) .
[ cols="^,^ " , ]
we performed a first test to search for a continuous energy loss of neutrinos in matter .
this was done by exposing a high - sensitivity germanium crystal to the cern high - energy @xmath0 beam .
it was found that a neutrino of type @xmath0 loses less than @xmath25@xmath6 per @xmath22 of germanium ( about @xmath1 of the normal de / dx characterizing a minimum ionizing particle ) .
incidently , this would mean an energy loss of @xmath26 for @xmath0 traversing the whole diameter of the earth ; thus an anomalous energy loss of @xmath0 can not explain the deficit of upgoing neutrinos seen by the superkamiokande detector @xcite . for the sake of comparison , the energy loss due to weak interactions amounts to @xmath27 per @xmath22 of germanium for neutrinos of the energy considered here . if the loss is predominantly attributed to @xmath9 s ( which could be more probable for an interaction of electromagnetic origin ) , then the limit is @xmath28 per @xmath22 of germanium and per @xmath9 .
this result applies to a continuous interaction having an energy quantum in the range 1 @xmath6 to 1 @xmath13 . to improve the significance of this result
we should carry out the search again in better experimental conditions ( improved mechanical stability , enlarged statistics ... ) .
we are also planning to continue the search at a nuclear reactor where the neutrino flux is higher and the energies are more relevant for astrophysical considerations .
finally , we should investigate the use of other devices ( bolometers , rf cavities , etc . ) in order to alleviate the limitation on the minimum interaction quantum , which is intrinsic to the present technique .
we would like to thank p. repain and l. serot for their help in the realization of the experimental set - up .
we thank the nomad collaboration for providing the necessary interfaces with the accelerator informations .
special thanks to the icarus 50l group who kindly shared their already - crowded platform .
y. suzuki , presented at the `` neutrino 98 '' conference , takayama ( japan ) , june 1998 , ( to be published ) egpc30p ; eurisys mesures , st quentin en yvelines , france m.j .
scherwish , theory of statistics ( 1995 ) , springer verlag y. fukuda _ et al _ , phys .
lett . * 81*,1562 - 1567 ( 1998 ) | present result first test energy deposition neutrinos matter interactions test high purity germanium crystal cern high energy @xmath0 beam device offers easy high - sensitivity active medium exploratory search intense neutrino fluxes achieve limit neutrino deposits small energy detected neutrino energy losses consequences astrophysical contexts energy loss sun explain deficits solar neutrino experiments super - measurement solar neutrino spectrum excludes energy loss few per cent 10 @xmath2 neutrinos traversing 600000 @xmath3 matter average density 1 @xmath4 result reaches limit two orders magnitude energy deposition proportional initial neutrino energy relevant energy transfers supernova explosion used high - purity germanium coaxial diode effective volume 140 @xmath5 cooled liquid nitrogen germanium intrinsic band gap 0.74 @xmath6 mean electron - hole pair creation work 2.9 @xmath6 device gives lowest energy threshold active target heavy cryogenic environment detector installed cern high - energy band neutrino beam chorus nomad oscillation experiments 800 @xmath7 primary beryllium targetbeam composed of @xmath0 average energy 24 @xmath8 1% contamination @xmath9 produced in two spills 6 @xmath10 duration extracted beginning end 2.2 @xmath11 flat top sps accelerator machine cycle repeated every 14.4 @xmath11 synchronisation signals from nomad experiment 4 gates -calib gate @xmath12 before -nu1 nu2 gates -mu gate flat - top spills if neutrino energy loss caused by rare effects releasing @xmath13 per interaction impossible to disentangle from background charged particles weak interactions test concentrated on hypothetical processes neutrinos leave tiny energy leakage current during @xmath14 accelerator extraction duration read - out by infinite time constant charge - preamplifier coupled to hpge diode output sampled digitized by sigma - delta adc each gate sigma - delta loop reset sampling points accumulated during @xmath15 row energy deposition during spill sufficient resolution on impulsional energy deposition @xmath16 @xmath17 @xmath18 line spectrometric monitoring identification photon charged particle interactions crystalp - type ge crystal chosen li - doped zone against alphas low - energy beta gamma radiations protection enforced by @xmath19 lead shield detector supported by 4 pneumatic insulators electromagnetic compatibility electron - hole pairs produced when energy deposited in crystal creating current through diode leakage current measured @xmath20 after 24-hour warm - up ionizing particle crosses crystal fast pulse @xmath21 duration minimum ionizing particle ( 5 @xmath22 deposits 30 @xmath2 expected signal continuous current excess during neutrino spill proportional to neutrino flux identify reject rows contaminated by photons charged particles discriminating cut set to @xmath23 over two points 41% rows in nu1 gate 38% in nu2 gate 12% in calib mu gates time distribution of rejected energy depositions gates exhibits neutrino pulse time structure first test for continuous energy loss of neutrinos in matter high - sensitivity germanium crystal to cern high - energy @xmath0 beamfound neutrino @xmath0 loses less than @xmath25@xmath6 per @xmath22 germanium @xmath1 normal de / dx minimum ionizing particle energy loss of @xmath26 for @xmath0 traversing diameter earth anomalous energy loss @xmath0 explain deficit upgoing neutrinos superkamiokande detector @xcite energy loss weak interactions @xmath27 per @xmath22 germanium if loss attributed to @xmath9 s limit is @xmath28 per @xmath22 germanium @xmath9 result applies to continuous interaction energy quantum range 1 @xmath6 to 1 @xmath13 improve significance search in better conditions stability enlarged statistics planning continue search at nuclear reactor neutrino flux higher energies relevant for astrophysical considerations investigate use of other devices bolometers rf cavities alleviate limitation minimum interaction quantum thank p. repain l. serot experimental set - up nomad collaboration interfaces accelerator informations thanks to icarus 50l group platform y. suzuki neutrino 98 conference takayama japan june 1998 published egpc30p ; eurisys mesures st quentin en yvelines francescherwish theory statistics 1995 springer verlag fukuda phys . lett.,1562 - 1567 1998 | 0.311636 | 1,778 | 913 | , Saving $0.1 in GPT-4. |
155,008 | we present results for the equation of state of a graphene - like model in an effort to understand the properties of its quantum phase transition .
the @xmath0 fermion species interact through a three dimensional instantaneous coulomb potential .
since there are no reliable analytical tools that work for all values of @xmath0 and the coupling constant @xmath1 , we rely on monte carlo simulations to calculate the critical properties of the model near the phase transition .
we consider the four - component formulation for the fermion elds , which arises naturally as the continuum limit of the staggered fermion construction in ( 2 + 1 ) dimensions . in the limit of infinitely strong coulomb interaction
, the system undergoes a quantum phase transition at a critical number of fermion species @xmath2 .
we also calculate the values of the critical exponents at the quantum phase transition . | there has been considerable recent interest in graphene sparked by its discovery and subsequent experimental study . in brief , for a carbon monolayer having one mobile electron per atom , a simple tight - binding model shows that the spectrum of low - energy excitations exhibits a linear dispersion relation centred on zeroes located at the six corners of the first brillouin zone ( see e.g. @xcite ) . using a linear transformation among the fields at two independent zeroes it is possible to recast the hamiltonian in dirac form with @xmath3 flavors of four - component spinor @xmath4 , the counting of the massless degrees of freedom coming from 2 carbon atoms per unit cell @xmath5 2 zeroes per zone @xmath5 2 physical spin components per electron .
electron propagation in the graphene layer is thus relativistic , albeit at a speed @xmath6 .
although this is a factor of @xmath7 slower than the speed of light in vacuum , it is still much faster than the speed of electrons in an ordinary conductor .
the implications for the high mobility of the resulting charge carriers ( which may be negatively - charged `` particles '' or positively - charged `` holes '' depending on doping ) is the source of the current excitement .
the stability of the zero - energy points is topological in origin , as emphasised by creutz @xcite .
while the above considerations apply quite generally , a realistic model of graphene must incorporate interactions between charge carriers .
one such model , due to son @xcite , has @xmath0 massless fermion flavors propagating in the plane , but interacting via an instantaneous 3@xmath8 coulomb interaction . in euclidean metric and static gauge @xmath9
the action reads @xmath10 where @xmath11 is the electron charge , @xmath12 , and the @xmath13 dirac matrices satisfy @xmath14 , @xmath15 . in our notation @xmath16
is a vector in the 2@xmath8 plane while the index @xmath17 runs over all three spatial directions . in the large-@xmath0
limit the dominant quantum correction @xmath18 comes from a vacuum polarisation fermion- antifermion loop and the resummed @xmath19 propagator becomes @xmath20 where @xmath21 , and @xmath22 corresponds to the classical propagation of v. in either the strong coupling or large-@xmath0 limits @xmath23 is thus dominated by @xmath18 , the relative importance of the original coulomb interaction being governed by a parameter @xmath24 .
the chiral symmetry breaking , due to the spontaneous condensation of particle - hole pairs , is signalled by an order parameter @xmath25 .
physically the most important outcome is the generation of a gap in the fermion spectrum , implying the model is an insulator .
son @xcite postulates that this insulating phase exists in the corner of the phase diagram corresponding to large @xmath26 and small @xmath0 , and in particular that the insulator - conductor phase transition taking place at @xmath27 in the strong - coupling limit @xmath28 is a novel quantum critical point .
the value of @xmath29 , and the issue of whether it is greater or less than the physical value @xmath3 , must be settled by a non - perturbative calculation . a recent estimate obtained by a renormalisation group treatment of radiatively - induced four - fermion contact interactions ,
is @xmath30 @xcite .
the proposed physics is very reminiscent of the three dimensional thirring model , which is analytically tractable at large @xmath0 , but exhibits spontaneous chiral symmetry breaking at small @xmath0 and strong coupling @xcite .
arguably the thirring model is the simplest field theory of fermions requiring a computational solution : the location of the phase transition at @xmath27 in the strong coupling limit has recently been determined by lattice simulations to be @xmath31 @xcite .
the apparent similarity of the two systems has led us to propose a thirring - like model pertinent to graphene , with lagrangian @xmath32 .
\label{eq : model2}\ ] ] as for ( [ eq : model ] ) we assume a large-@xmath0 limit to estimate the dominant vacuum polarisation correction ; the resultant propagator for @xmath19 is @xmath33 in the strong - coupling or large-@xmath0 limits , @xmath34 coincides with @xmath35 , implying that the fermion interactions are equivalent .
it is also the case that @xmath36 .
this last limit is important because critical behaviour in the thirring model is governed by a uv - stable fixed point of the renormalisation group @xcite .
we thus expect predictions made with the model ( [ eq : model2 ] ) , and in particular critical behaviour such as the value of @xmath29 , to be generally valid for son s model ( [ eq : model ] ) in the limit of large @xmath37 .
the lattice model studied in this paper is related to the lattice thirring model studied in @xcite , except that in this case we used the compact formulation for the auxiliary boson field .
the simulations were performed with the standard hybrid molecular dynamics algorithm .
@xmath38 vs. @xmath39 near @xmath40 for @xmath3 . ]
the novelty of @xcite is that it is the first study of the thirring model by lattice simulations in the strong - coupling limit @xmath41 .
since we aim to repeat the strategy here we should discuss how this was done .
first , note that the vacuum polarisation calculation leading to the results ( [ eq : d_gr],[eq : d_th ] ) does not go through in quite the same way for the lattice regularised model ; rather , there is an additive correction which is momentum independent and uv - divergent : @xmath42 where @xmath43 comes from incomplete cancellation of a lattice tadpole diagram @xcite .
this extra divergence not present in the continuum treatments can be absorbed by a wavefunction renormalisation of @xmath19 and a coupling constant renormalisation @xmath44 in the large-@xmath0 limit we thus expect to find the strong coupling limit of the lattice model at @xmath45 implying @xmath46
. for @xmath47 @xmath48 becomes negative , and @xmath49 no longer describes a unitary theory .
away from the large-@xmath0 limit , where chiral symmetry may be spontaneously broken , there is no analytical criterion for identifying @xmath50 ; however in this case a numerical calculation of @xmath51 shows a clear peak at @xmath52 , whose location is approximately independent of both volume and @xmath53 , indicating an origin at the uv scale @xcite .
[ gr : beta_peak_nf2 ] exemplifies this behaviour in the lattice model with @xmath3 on system volumes @xmath54 : for @xmath55 we identify @xmath56 .
since for orthodox chiral symmetry breaking the magnitude of the condensate is expected to increase monotonically with the coupling strength , we interpret the peak as the point where unitarity violation sets in , ie .
we use simulations performed at @xmath58 to explore the strong coupling limit , and find clear evidence for a chiral symmetry restoring phase transition at a well - defined @xmath29 . as shown in fig .
[ gr : beta_peak ] @xmath59 decreases as @xmath0 increases from @xmath60 to @xmath61 . near @xmath62 the curve reaches a minimum , implying a significant change in the strong coupling behavior of the model .
the rapid change in the value of the condensate at @xmath63 implies a signifacant change in the strong coupling behavior of the model .
this can be a phase transition that separates a chirally broken from a chirally symmetric phase .
equation of state fits to data generated with different fermion masses on different lattice sizes . ]
equation of state fits to data generated with different fermion masses on different lattice sizes . ]
since our model has anisotropic interactions in the spatial and temporal directions the correlation lengths near the transition diverge with different exponents , @xmath64 in the spatial directions , and @xmath65 in the temporal direction . in this case
a modified hyperscaling relation holds : @xmath66 we carefully monitored the finite size effects in the spatial directions by comparing the values of the condensate at @xmath67 , @xmath68 extracted from simulations on @xmath69 and @xmath70 lattices .
our results show that finite @xmath71 effects are negligible for @xmath72 .
we fitted the data generated on several lattices with @xmath73 , @xmath74 and @xmath75 to the following rg - inspired equation of state @xcite : @xmath76\langle\bar\psi\psi\rangle^p + b\langle\bar\psi\psi\rangle^\delta,\ ] ] where @xmath77 .
the results show clearly that the model undergoes a second order quantum phase transition and the values of the most significant parameters are : @xmath78 , @xmath79 , @xmath80 , and @xmath81 . the data and the fitted curves are shown in fig .
[ gr : eos ] .
the value of @xmath78 is close to the value of @xmath63 , where a rapid change in the @xmath59 vs @xmath0 occurs .
we presented results from a non - perturbative monte carlo study of a thirring - like model pertinent to graphene .
our analysis of its equation of state at strong coupling reveals that the model undergoes a second order quantum phase transition at @xmath82 , which is greater than the physical @xmath3 value .
this result implies that freely suspended graphene may be an insulator for couplings @xmath1 larger than a critical coupling @xmath83 .
9 r. jackiw and s. y. pi , phys .
rev . lett .
* 98 * ( 2007 ) 266402 .
m. creutz , jhep * 0804 * ( 2008 ) 017 .
son , phys . rev .
b * 75 * ( 2007 ) 235423 .
drut and d.t .
son , phys . rev .
b * 77 * ( 2008 ) 075115 . l. del debbio , s.j .
hands and j.c .
mehegan , nucl . phys .
b * 502 * ( 1997 ) 269 .
l. del debbio and s.j .
hands , nucl .
b * 552 * ( 1999 ) 339 ; s.j .
hands and b. lucini , phys .
b * 461 * ( 1999 ) 263 .
s. christofi , s.j . hands and c. strouthos , phys . rev .
d * 75 * ( 2007 ) 101701 . | interest in graphene sparked by discovery experimental study carbon monolayer one mobile electron per atom tight - binding model shows low - energy excitations linear dispersion on zeroes at six corners of first brillouin zone linear transformation at two zeroes recast hamiltonian in dirac form with @xmath3 flavors four - component spinor massless degrees freedom from 2 carbon atoms per unit cell 2 zeroes per zone spin components per electron electron propagation in layer relativistic at speed @xmath6 slower than light in vacuum faster than electrons in ordinary conductor high mobility of charge carriers negatively - charged or positively - charged source of current excitement stability of zero - energy points topological emphasised by realistic model of graphene interactions between charge carriers model massless fermion flavors propagating in plane via instantaneous 3@xmath8 coulomb interaction euclidean metric static gauge action reads @xmath10 electron charge dirac matrices satisfy @xmath14 @xmath15@xmath16 vector 2@xmath8 plane @xmath17 three directions large-@xmath0 limit quantum correction @xmath18 from vacuum polarisation fermion- antifermion loop @xmath19 propagator @xmath20 @xmath21 @xmath22 classical propagation v strong coupling large-@xmath0 limits @xmath23 dominated by @xmath18 original coulomb governed parameter @xmath24 chiral symmetry breaking particle - signalled by parameter @xmath25 important outcome gap fermion spectrum model insulator insulating phase corner phase diagram large @xmath26 small @xmath0 insulator - conductor phase transition at @xmath27 strong - coupling limit @xmath28 quantum critical point value @xmath29 value @xmath3 non - perturbative calculation estimate radiatively - induced four - fermion contact interactions @xmath30 @xcite proposed physics three dimensional thirring model tractable at large @xmath0 spontaneous chiral symmetry breaking at small @xmath0 strong coupling @xcitethirring model simplest field theory fermions computational solution location phase transition at @xmath27 strong coupling limit determined by lattice simulations @xmath31 @xcite similarity systems thirring - like model graphene with lagrangian @xmath32 assume large-@xmath0 limit dominant vacuum polarisation correction resultant propagator for @xmath19 is @xmath33 strong - limits @xmath34 coincides with @xmath35 fermion interactions equivalent @xmath36 limit important critical behaviour model governed by uv - stable fixed point renormalisation group @xcite expect predictions critical behaviour value of @xmath29 valid for son s model limit large @xmath37 lattice model studied related to thirring model @xcite compact formulation for auxiliary boson field simulations performed with standard hybrid molecular dynamics algorithm @xmath38 vs. @xmath39 near @xmath40 for @xmath3 first study of thirring model by lattice simulations in strong - coupling limit @xmath41 strategy discussvacuum polarisation calculation results d_gr_th lattice regularised model additive correction momentum independent uv - divergent : @xmath42 @xmath43 from cancellation lattice tadpole diagram divergence absorbed by wavefunction renormalisation @xmath19 coupling constant renormalisation @xmath44 large-@xmath0 limit strong coupling limit at @xmath45 @xmath46 @xmath47 @xmath48 becomes negative @xmath49 unitary theory-@xmath0 chiral symmetry no analytical criterion for identifying @xmath50 numerical calculation of @xmath51 shows peak at @xmath52 independent of volume @xmath53 origin at uv scale @xcite beta_peak_nf2 ] behaviour in lattice model with @xmath3 on system volumes @xmath54 @xmath55 @xmath56 chiral symmetry breaking condensate coupling strength interpret peak as unitarity violation sets use simulations at @xmath58 explore strong coupling limit evidence for chiral symmetry restoring phase transition at @xmath29 @xmath59 decreases as @xmath0 increases from @xmath60 to @xmath61@xmath62 curve minimum significant change strong coupling behavior rapid change condensate at @xmath63 implies change coupling behavior phase transition separates chirally broken from symmetric phase equation of state fits data different fermion masses lattice sizes model has anisotropic interactions spatial temporal correlation lengths transition diverge with exponents @xmath64 @xmath65 modified hyperscaling relation holds @xmath66 monitored finite size effects comparing values condensate at @xmath67 @xmath68 @xmath69 @xmath70 lattices finite @xmath71 effects negligible for @xmath72 fitted data lattices @xmath73 @xmath74 @xmath75 to rg - equation of state @xmath76\langle\bar\psi\psi\rangle^p + b\rangle^\delta @xmath77 results model undergoes second order quantum phase transition significant parameters are @xmath78 @xmath79 @xmath80 @xmath81 data fitted curves shown in fig . value @xmath78 close to value @xmath63 rapid change in @xmath59 vs @xmath0 occurspresented results non - perturbative monte carlo study thirring - model graphene analysis equation strong coupling reveals model second order quantum phase transition @xmath82 greater physical @xmath3 value implies suspended graphene insulator couplings @xmath1 larger critical coupling @xmath83 jackiw y pi phys . rev 98 2007 ) 266402 creutz 0804 2008 ) 017 son rev 75 2007 235423 drut son phys 77 2008 ) 075115 del debbio s.j hands mehegan phys . 502 1997 ) 269 del debbio s.j hands 552 1999 ) 339 s hands lucini phys 461 1999 ) 263 christofi s hands strouthos phys . rev 75 2007 ) 101701 | 0.355113 | 2,399 | 1,259 | , Saving $0.1 in GPT-4. |
155,016 | we consider the repulsive hubbard model on a square lattice with an additional term , @xmath0 , which depends upon the square of a single - particle nearest - neighbor hopping . at half - band
filling , constant @xmath0 , we show that enhancing @xmath1 drives the system from a @xmath2-wave superconductor to an antiferromagnetic mott insulator . at zero temperature in the superconducting phase ,
spin - spin correlations follow a powerlaw : @xmath3 is in the range @xmath4 and depends upon the coupling constants @xmath0 and @xmath5 .
this results is reached on the basis of large scale quantum monte - carlo simulations on lattices up to @xmath6 , and is shown to be independent on the choice of the boundary conditions .
we define a pairing ( magnetic ) scale by the temperature below which _ short range _ @xmath2-wave pairing correlations ( antiferromagnetic fluctuations ) start growing . with finite temperature quantum monte carlo simulations , we demonstrate that both scales are identical over a large energy range .
those results show the extreme compatibility and close interplay of antiferromagnetic fluctuations and @xmath2-wave superconductivity .
the understanding of the interplay between @xmath2-wave superconductivity and antiferromagnetism is a central issue for the understanding of the phase diagram of high-@xmath7 superconductors @xcite .
the aim of this work is to further study a model which shows a quantum transition between a @xmath2-wave superconductor and an antiferromagnetic mott insulator .
it thus enables us to address the above question .
the model we consider , has been introduced in ref .
@xcite .
it is defined by : @xmath8 with the hopping kinetic energy @xmath9 here , @xmath10 , @xmath11 , and @xmath12 where @xmath13 ( @xmath14 ) creates ( annihilates ) an electron on site @xmath15 with z - component of spin @xmath16 .
we impose twisted boundary conditions : @xmath17 with @xmath18 the flux quanta and @xmath19 the linear length of the square lattice .
the boundary conditions given by eq .
( [ bound ] ) account for a magnetic flux threading a torus on which the lattice is wrapped . one major advantage of the above model consists in the fact that at half - band filling , the sign problem in the quantum monte - carlo ( qmc ) method may be avoided .
this statement is valid even for the above boundary conditions .
hereafter , we will only consider the half - filled case .
previously we have considered the model of eq .
( [ tuw ] ) at fixed values of @xmath5 and as a function of @xmath0 . at @xmath20
we have shown that the quantum transition between the mott insulator and @xmath2-wave superconductor occurs at @xmath21 @xcite .
we have equally considered the model at finite doping and given numerical evidence of the occurrence of a doping induced quantum transition between the mott insulator and @xmath2-wave superconductor @xcite . here , we fix @xmath22 and vary @xmath1 for the half - filled case .
the advantage of this choice of parameters is that it provides us with a _
large _
parameter range in the superconducting phase where qmc simulations are extremely precise , thus allowing us to reach large lattices .
this allows us to reliably study the nature of the spin degrees of freedom in the superconducting state .
we have used two qmc algorithms .
i ) the projector qmc algorithm which produces zero temperature results in a canonical ensemble @xcite and ii ) the finite temperature grand canonical qmc method @xcite . the application of those algorithms for the above model has been discussed in reference @xcite . both algorithms , generate a systematic error proportional to @xmath23 where @xmath24 denotes the imaginary time step . to determine the exponent of the spin - spin correlations
we have extrapolated @xmath24 to zero .
we first concentrate of the charge degrees of freedom at zero temperature . to distinguish between a superconductor and insulator
, we compute the ground state energy as a function of the twist in the boundary condition : @xmath25 . for an insulator ,
the wave function is localized and hence , an exponential decay of @xmath26 as a function of lattice size is expected @xcite . in the hartree - fock spin density wave ( sdw ) approximation for the half - filled hubbard model , one obtains @xmath27 where @xmath28 is the localization length of the wave function . on the other hand , for a superconductor
, @xmath29 shows anomalous flux quantization : @xmath30 is a periodic function of @xmath31 with period @xmath32 and a non vanishing energy barrier is to be found between the flux minima @xcite so that @xmath33 remains finite as @xmath34 .
fig .
( [ sut0.fig ] ) plots @xmath35 versus @xmath36 at @xmath22 and for various values of @xmath1 . at values of @xmath37 ,
the data is consistent with a @xmath36 form and scales to a finite value . at @xmath38
, the data may be fitted to the above sdw form .
thus , for values of @xmath22 , the data is consistent with the occurrence of a quantum transition between a superconductor and a mott insulator at @xmath39 . in order to determine the symmetry of the superconducting state ,
we have computed equal time pairing correlations in the extended @xmath40-wave and @xmath41 channels : @xmath42 with @xmath43 here , @xmath44 and @xmath45 for @xmath46 ( @xmath47 . the vertex contribution to the above quantity is given by : @xmath48 per definition , @xmath49 in the absence of interactions .
fig .
[ pair.fig ] plots @xmath50 along the diagonal of the lattice .
we consider lattices ranging up to @xmath51 . for a fixed lattice size ,
one notices a plateau structure as a function of distance .
the extrapolation of this plateau value to the thermodynamic limit is hard and the above criterion of flux quantization , proves to be a more efficient method to conclude superconductivity @xcite . in comparison to the @xmath2-wave signal
, the extended @xmath40-wave signal at _ large _ distances ( data not shown ) may not be distinguished from zero within our accuracy .
the extended @xmath40-wave pair - field correlation dominate at short distances .
the data confirming this statement may be found in the table .
the added @xmath0-term , contains no processes which explicitly favor @xmath2-wave superconductivity . on the contrary
, one would expect the @xmath0-term to favor extended @xmath40-wave symmetry , and this shows up on short length scales .
the fact that @xmath2-wave symmetry dominates at long - range is a result of the underlying magnetic structure . to study the spin degrees of freedom ,
we have computed equal time spin - spin correlations @xmath52 where @xmath53 denotes the spin operator on site @xmath54 .
fig .
[ spt0.fig ] , plots @xmath55 versus @xmath36 where @xmath19 corresponds to the linear size of the lattice .
we consider periodic boundary conditions , and various values of @xmath1 .
@xmath56 is constant and set to @xmath57 . for values of @xmath58
the data is consistent with a powerlaw decay : @xmath59 at @xmath38 extrapolation of the data leads to a finite staggered moment , and thus the presence of antiferromagnetic long - range order .
the statement in eq .
( [ s_alpha ] ) is surprising and deserves confirmation .
to cross - check the validity of the above equation , we demonstrate numerically that it is independent on the choice of the boundary condition .
fig .
[ spin_ln.fig ] plots @xmath55 at @xmath60 and @xmath61 for @xmath22 . at @xmath62 ,
the solid lines corresponds to least square fits to the form @xmath63 .
we thus determine the exponent @xmath64 .
the data at @xmath65 is consistent at large distances with the form @xmath66 thus showing that the exponent @xmath67 is independent on the choice of the boundary .
for the @xmath65 simulations we were able to reach lattices up to @xmath6 . at @xmath61 and for the boundary conditions set by @xmath65 , @xmath68 ( see fig .
[ pair.fig ] ) decays slower than @xmath55 ( see fig .
[ spin_ln.fig ] ) thus confirming that the pairing correlations are dominant @xcite .
having put eq .
( [ s_alpha ] ) on a numerically firm basis , we now argue why it is a surprising result . in two dimensions , correlation functions which decay slower than @xmath69 lead to divergences in fourier space .
the fourier transform of the spin - spin correlations at @xmath61 is plotted in fig .
[ sq.fig ] .
one sees a systematic increase of @xmath70 as a function of system size @xcite .
in contrast , a mean field @xmath2-wave bcs calculation , yields spin - spin correlations which decays as a powerlaw with @xmath71 .
this mean - field result leads to a finite @xmath72 in the thermodynamic limit .
another surprise comes from the dependence of the exponent @xmath64 on the coupling constants , @xmath5 and @xmath0 .
this is a feature which occurs in one dimensional quantum systems such as the @xmath73 or hubbard models .
we now consider the model at finite temperatures with the use of the grand - canonical qmc algorithm . in fig .
[ tjtp.fig ] ( a ) we plot s(q ) for values of @xmath1 ranging from @xmath74 to @xmath75 .
we define the magnetic scale , @xmath76 , as the temperature scale where @xmath77 starts growing . in figs .
[ tjtp.fig ] ( b ) and ( c ) we consider the vertex contribution to the @xmath2-wave paring correlations at distance @xmath78 : @xmath79 . here
, we consider an @xmath80 lattice . at this distance
the @xmath40-wave pairing correlations are negligible .
we define the @xmath2-wave pairing scale , @xmath81 , by the temperature scale at which @xmath79 starts growing . from the data ,
we conclude that @xmath82 that is , antiferromagnetic fluctuations as well as @xmath2-wave pairing fluctuations occur hand in hand at the same energy scale .
if @xmath83 ( @xmath84 ) antiferromagnetic correlations ( @xmath2-wave pair - field correlations ) will dominate at low temperatures .
we note that in the large @xmath1 limit , @xmath76 should scale as @xmath85 .
that up to @xmath75 we still do not see this behavior , may be traced back to the fact that the @xmath0-term enhances the band - width @xcite .
@xmath86 is a natural consequence of the assumption of @xmath87 symmetry which unifies antiferromagnetism with @xmath2-wave superconductivity @xcite .
the data , however , does not demonstrate the presence of this symmetry in this model and further work is required . in conclusion ,
we have considered aspects of the @xmath88 model defined in eq.([tuw ] ) .
the model , at half - band filling , shows rich physics and allows us to study the interplay between magnetism and @xmath2-wave superconductivity .
we have pointed out the surprising nature of the spin - spin correlations in the superconducting state : latexmath:[$ s(\vec{r } )
\sim e^{-i \vec{r } \cdot \vec{q } } @xmath64 is in the range @xmath90 and depends on the coupling constants @xmath91 and @xmath5 .
those conclusions are based on large scale calculations for system sizes up to @xmath92 .
we have equally shown that the energy scales at which @xmath2-wave pairing and antiferromagnetic fluctuations occur are identical .
the further understanding of spin and charge dynamics , as well as the doping of the model remains for further studies . | m. imada , d.j . scalapino and
m. muramatsu are thanked for many instructive conversations .
the computations were carried out on the t3e of the hlrs , stuttgart , as well as on the t90 and t3e of the hlrz , jlich .
flux quantization , measures directly the superfluid density which does not require the knowledge of the symmetry of pairing correlations , the _ form factor _ of the cooper pair , or the nature of the _ quasiparticle _ which constitute the cooper pair .
we note however that we not yet established firmly the occurrence of a flux minima at @xmath93 @xcite . at @xmath94 and @xmath95 ,
one may show numerically that the vertex contribution to the @xmath2-wave pairing susceptibility grows quicker than the vertex contribution to the staggered spin susceptibility as a function of system size @xcite .
we note that obtaining the exponent @xmath64 from the size scaling of @xmath96 requires larger lattice sizes than the ones considered here .
short range spin - spin correlations included in @xmath96 are still important for the considered lattice sizes . .
short range vertex contribution of pair - field correlations in the extended @xmath40- and @xmath2-wave channels .
here we consider an @xmath97 lattice at @xmath98 , @xmath99 and @xmath100 .
the boundary conditions are set by @xmath101 .
the distance @xmath102 is in units of the lattice constant . [
cols="^,^,^",options="header " , ] | imada scalapino muramatsu thanked for conversations computations carried on t3e hlrs stuttgart t90 t3e hlrz jlich flux quantization measures superfluid density knowledge symmetry pairing correlations form factor cooper pair nature quasiparticle flux minima at @xmath93 @xcite @xmath94 @xmath95 vertex contribution to @xmath2-wave pairing susceptibility grows quicker than staggered spin susceptibility system size obtaining exponent @xmath64 from size scaling @xmath96 requires larger lattice sizes short range spin - correlations in @xmath96 important for lattice sizes short range vertex contribution pair - field correlations in extended @xmath40- @xmath2-wave channels consider @xmath97 lattice at @xmath98 @xmath99 @xmath100 boundary conditions set by @xmath101 distance @xmath102 in units lattice constant | 0.07411 | 344 | 177 | , Saving $0.0 in GPT-4. |
155,023 | this paper deals with efficient numerical representation and manipulation of differential and integral operators as symbols in phase - space , i.e. , functions of space @xmath0 and frequency @xmath1 .
the symbol smoothness conditions obeyed by many operators in connection to smooth linear partial differential equations allow to write fast - converging , non - asymptotic expansions in adequate systems of rational chebyshev functions or hierarchical splines .
the classical results of closedness of such symbol classes under multiplication , inversion and taking the square root translate into practical iterative algorithms for realizing these operations directly in the proposed expansions . because symbol - based numerical methods handle operators and not functions , their complexity depends on the desired resolution @xmath2 very weakly , typically only through @xmath3 factors .
we present three applications to computational problems related to wave propagation : 1 ) preconditioning the helmholtz equation , 2 ) decomposing wavefields into one - way components and 3 ) depth - stepping in reflection seismology
. * acknowledgements . * the first author is partially supported by an nsf grant .
the second author is partially supported by an nsf grant , a sloan research fellowship , and a startup grant from the university of texas at austin . | a typical problem of interest in this paper is the efficient representation of functions of elliptic linear operators such as @xmath4 where @xmath5 is smooth , and @xmath6 ^ 2 $ ] with periodic boundary conditions .
we have in mind the inverse , the square root , and the exponential of @xmath7 as important examples of functions of @xmath7 .
while most numerical methods for inverting @xmath7 , say , would manipulate a right - hand side until convergence , and leverage sparsity or other properties of @xmath7 in doing so , the scope of this paper is quite different . indeed , we present expansions and iterative algorithms for manipulating operators as `` symbols '' , which make no or little reference to the functions of @xmath0 to which these operators may later be applied .
the central question is of course that of choosing a tractable representation for differential and integral operators . if a function @xmath8 has @xmath2 degrees of freedom if for instance it is sampled on @xmath2 points then a direct representation of operators acting on @xmath8 would in general require @xmath9 degrees of freedom .
there are many known methods for bringing down this count to @xmath10 or @xmath11 in specific cases , such as leveraging sparsity , computing convolutions via fft , low - rank approximations , fast summation methods @xcite , wavelet or x - let expansions @xcite , partitioned svd and h - matrices @xcite .
the framework presented in this paper is different , in the sense that we aim for a complexity essentially independent of @xmath2 , i.e. , at most a low - degree polynomial of @xmath3 , for representing and combining operators belonging to standard classes . like the methods above ,
the operator is `` compressed '' in such a way that applying it to functions remains simple ; it is only for this operation that the complexity needs to be greater than @xmath2 , in our case @xmath12 .
let @xmath7 denote a generic differential or singular integral operator , with kernel representation @xmath13 expanding the distributional kernel @xmath14 in some basis would be cumbersome because of the presence of a singularity along the diagonal @xmath15 . for this reason we choose to consider operators as _ pseudodifferential symbols _
@xmath16 , by considering their action on the fourier transform ] ] @xmath17 of @xmath8 ; @xmath18 one often writes @xmath19 for @xmath7 , where @xmath20 . in this representation ,
the singularities of @xmath14 are turned into the oscillating factor @xmath21 , which can be discounted by focusing on the symbol @xmath16 .
the latter is usually smooth , and in a very peculiar way .
it is the special form of the smoothness estimates for @xmath22which we now describe that guarantees the efficiency of the discretizations proposed in this paper .
a symbol defined on @xmath23 is said to be pseudodifferential of order @xmath24 ( and type @xmath25 ) if it obeys @xmath26 .
this symbol class is denoted @xmath27 ; the operator corresponding to some @xmath22 in this class is denoted @xmath19 , and belongs by definition to the class @xmath28 .
manifestly , one power of @xmath29 is gained for each differentiation , meaning that the larger @xmath29 , the smoother @xmath22 .
for instance , the symbols of differential operators are polynomials in @xmath1 and obey ( [ eq : type10 ] ) when they have @xmath30 coefficients . large classes of singular integral operators also have symbols in the class @xmath27 @xcite .
the standard treatment of pseudodifferential operators makes the further assumption that some symbols can be represented as polyhomogeneous series , such as @xmath31 which defines the `` classical '' symbol class @xmath32 when the @xmath33 are of class @xmath30 .
corresponding operators are said to be in the class @xmath34 . the series should be understood as an asymptotic expansion ; it converges only when adequate cutoffs smoothly removing the origin multiply each term .
only then , the series does not converge to @xmath16 , but to an approximation that differs from @xmath22 by a smoothing remainder @xmath35 , smoothing in the sense that @xmath36 .
for instance , an operator is typically transposed , inverted , etc .
modulo a smoothing remainder @xcite .
the subclass ( [ eq : classical ] ) is central for applications
it is the cornerstone of theories such as geometrical optics but the presence of remainders is a nonessential feature that should be avoided in the design of efficient numerical methods .
the lack of convergence in ( [ eq : classical ] ) may be acceptable in the course of a mathematical argument , but it takes great additional effort to turn such series into accurate numerical methods ; see @xcite for an example . in a sense
, the goal of this paper is to find adequate substitutes for ( [ eq : classical ] ) that promote asymptotic series into fast - converging expansions .
there are in general no explicit formulas for the symbols of functions of an operator .
fortunately , some results in the literature guarantee exact closedness of the symbol classes ( [ eq : type10 ] ) or ( [ eq : classical ] ) under inversion and taking the square root , without smoothing remainders . a symbol @xmath37 , or an operator @xmath38 , is said to be elliptic when there exists @xmath39 such that @xmath40 * it is a basic result that if @xmath41 , then @xmath42 .
see for instance theorem 18.1.8 in @xcite , volume 3 .
* it is also a standard fact that if @xmath43 , then its adjoint @xmath44 . *
if @xmath45 , and @xmath7 is elliptic and invertible is a bijection from @xmath46 to @xmath47 , hence obeys @xmath48 .
ellipticity , in the sense in which it is defined for symbols , obviously does not imply invertibility . ] on @xmath49 , then @xmath50 .
this result was proved by shubin in 1978 in @xcite .
* for the square root , we also assume ellipticity and invertibility .
it is furthermore convenient to consider operators on compact manifolds , in a natural way through fourier transforms in each coordinate patch , so that they have discrete spectral expansions .
a square root @xmath51 of an elliptic operator @xmath7 with spectral expansion @xmath52 , where @xmath53 are the spectral projectors , is simply @xmath54 with of course @xmath55 . in 1967 , seeley @xcite studied such expressions for elliptic @xmath56 , in the context of a much more general study of complex powers of elliptic operators . if in addition @xmath24 is an even integer , and an adequate choice of branch cut is made in the complex plane , then seeley showed that @xmath57 ; see @xcite for an accessible proof that involves the complex contour `` dunford '' integral reformulation of ( [ eq : as ] ) .
we do not know of a corresponding closedness result under taking the square root , for the non - classical class @xmath28 . in practice
, we will also manipulate operators that come from pde on bounded domains with certain boundary conditions ; the extension of the theory of pseudodifferential operators to bounded domains is a difficult subject that this paper has no ambition of addressing .
let us also mention in passing that the exponential of an elliptic , non - self - adjoint pseudodifferential operator is not in general pseudodifferential itself .
numerically , it is easy to check that smoothness of symbols is remarkably robust under inversion and taking the square root of the corresponding operators , as the following simple one - dimensional example shows .
let @xmath58 on the periodic interval @xmath59 $ ] where @xmath60 is a random bandlimited function shown in figure [ fig : sec11](a ) .
the symbol of this operator is @xmath61 which is of order 2 . in figure
[ fig : sec11](b ) , we plot the values of @xmath62 for @xmath0 and @xmath1 on a cartesian grid . since @xmath7 is elliptic and invertible , its inverse @xmath63 and square root @xmath64 are both well defined .
let use @xmath65 and @xmath66 to denote their symbols . from the above theorems , we know that the orders of @xmath65 and @xmath66 are respectively @xmath67 and @xmath68 .
we do not believe explicit formulae exist for these symbols , but the numerical values of @xmath69 and @xmath70 are shown in figure [ fig : sec11](c ) and ( d ) , respectively .
these plots demonstrate regularity of these symbols in @xmath0 and in @xmath1 ; observe in particular the disproportionate smoothness in @xmath1 for large @xmath71 , as predicted by the class estimate ( [ eq : type10 ] ) . [
cols="^,^ " , ]
an interesting question is what form discrete symbol calculus should take when other boundary conditions than periodic are considered , or on more general domains than a square .
one can speculate that the discrete sine transform ( dst ) should be used as @xmath72 for dirichlet boundary conditions on a rectangle , or the discrete cosine transform ( dct ) for neumann on a rectangle .
whatever choice is made for @xmath72 should dictate the definition of the corresponding frequency variable @xmath1 .
a more robust approach could be to use spectral elements for more complicated domains , where the spectral domain would be defined by chebyshev expansions .
one may also imagine expansions in prolate spheroidal wavefunctions .
regardless of the type of expansions chosen , the theory of pseudodifferential operators on bounded domains is a difficult topic that will need to be understood .
another interesting problem is that of designing absorbing boundary conditions in variable media .
we hope that the ideas of symbol expansions will provide new insights for this question .
symbol - based methods may help solve other equations than elliptic pde .
the heat equation in variable media comes to mind : its fundamental solution has a nice pseudodifferential smoothing form that can be computed via scaling - and - squaring .
a more challenging example are hyperbolic systems in variable , smooth media .
the time - dependent green s function of such systems is not a pseudodifferential operator , but rather a fourier integral operator ( fio ) , where @xmath73 needs to be replaced by @xmath74 .
we regard the extension of discrete symbol calculus to handle such phases a very interesting problem , see @xcite for preliminary results on fast application of fio .
as previously , write @xmath75 for the fourier series coefficients of @xmath16 in @xmath0
. then we can express ( [ eq : psido - sum ] ) as @xmath76 we seek to interchange the two sums .
since @xmath16 is differentiable @xmath77 times , we have @xmath78^d } e^{-2 \pi i x \cdot \lambda } ( 1 + ( - \delta_x)^{d'/2 } ) a(x,\xi ) \ , dx,\ ] ] hence @xmath79 . the exponent @xmath77 is chosen so that @xmath80 is absolutely summable in @xmath81 .
if in addition we assume @xmath82 , then we can apply fubini s theorem and write @xmath83 where @xmath84 , and @xmath85 is the operator of multiplication by @xmath86 on the @xmath1 side . by plancherel ,
we have @xmath87 therefore , by the triangle inequality , @xmath88 as we have seen , the sum over @xmath89 converges .
this proves the theorem when @xmath90 is sufficiently smooth ; a classical density argument shows that the same conclusion holds for all @xmath91^d)$ ] . | typical problem is representation of functions of elliptic linear operators @xmath4 smooth @xmath6 2 $ with periodic boundary conditions inverse square root exponential of @xmath7 important examples functions methods for inverting @xmath7 manipulate right - hand side until convergence leverage sparsity properties scope paper different present expansions iterative algorithms for manipulating operators as symbols no reference to functions of @xmath0 central question is choosing tractable representation for differential integral operators if function @xmath8 has @xmath2 degrees freedom on direct representation of operators on require @xmath9 degrees freedom methods for down count to @xmath10 or @xmath11 leveraging sparsity computing convolutions via fft low - rank approximations fast summation methods wavelet x - expansions partitioned svd h - matrices framework paper different aim for complexity independent of @xmath2 low - degree polynomial of @xmath3 for representing combining operators standard classes operator is compressed applying to functions simple only for operation complexity needs greater than @xmath2 @xmath12 .@xmath7 generic differential singular integral operator kernel representation @xmath13 expanding distributional kernel @xmath14 cumbersome singularity along diagonal @xmath15 consider operators pseudodifferential symbols @xmath16 action on fourier transform @xmath17 @xmath8 @xmath18 @xmath19 for @xmath7 @xmath20 singularities @xmath14 turned into oscillating factor @xmath21 discounted focusing symbol @xmath16 latter smooth peculiar special form smoothness estimates @xmath22which guarantees efficiency discretizations symbol @xmath23 pseudodifferential of order @xmath24 @xmath25 obeys @xmath26 symbol class denoted @xmath27 operator corresponding @xmath22 denoted @xmath19 belongs class @xmath28 one power of @xmath29 gained for each differentiation larger @xmath29 smoother @xmath22 symbols of differential operators polynomials in @xmath1 obey @xmath30 coefficients large classes singular integral operators symbols in class @xmath27 pseudodifferential operators symbols represented polyhomogeneous series @xmath31 defines classical symbol class @xmath32 @xmath33 class @xmath30 corresponding operators in class @xmath34series as asymptotic expansion converges when cutoffs multiply each term converge to @xmath16 approximation differs from @xmath22 by smoothing remainder @xmath35 @xmath36 operator transposed inverted modulo smoothing remainder @xcite subclass ( [ eq : classical ] central for cornerstone of theories geometrical optics presence remainders nonessential in efficient numerical methods lack of convergence in classical acceptable effort to turn into accurate numerical methods @xcite goal paper find substitutes for classical asymptotic series into fast - converging expansions no explicit formulas for symbols of functions of operator some results guarantee closedness of symbol classes ( [ eq : type10 or classical under inversion square root without smoothing remainders symbol @xmath37 or operator @xmath38 elliptic when @xmath39 @xmath40 if @xmath41 , then @xmath42 theorem 18.1.8 standard fact if @xmath43 adjoint @xmath44 if @xmath45 @xmath7 elliptic and invertible bijection from @xmath46 to @xmath47 obeys @xmath48 ellipticity defined for symbols imply invertibility @xmath49 @xmath50 proved by shubin 1978 in @xcite square root assume ellipticity invertibility consider operators on compact manifolds fourier transforms coordinate discrete spectral expansions square root @xmath51 of elliptic operator @xmath7 with spectral expansion @xmath52 @xmath53 spectral projectors @xmath54 with @xmath55 1967 seeley @xcite studied expressions for elliptic @xmath56 study complex powers elliptic operators if @xmath24 is even integer adequate choice of branch cut in complex plane seeley showed @xmath57 @xcite proof complex contour dunford integral reformulation of corresponding closedness result under square root for non - classical class @xmath28 manipulate operators from pde on bounded domains with boundary conditions extension theory pseudodifferential operators to bounded domains difficult exponential of elliptic non - self - adjoint pseudodifferential operator not pseudodifferential check smoothness of symbols robust under inversion square root of operators one - dimensional example shows@xmath58 periodic interval @xmath59 @xmath60 random bandlimited function figure fig sec11](a ) symbol operator @xmath61 order 2 figure fig sec11](b ) plot values @xmath62 @xmath0 @xmath1 cartesian grid @xmath7 elliptic invertible inverse @xmath63 square root @xmath64 defined @xmath65 @xmath66 denote symbols orders @xmath65 @xmath66 @xmath67 @xmath68 explicit formulae symbols numerical values @xmath69 @xmath70 figure [ fig sec11](c ) ( d ) plots demonstrate regularity symbols @xmath0 @xmath1 disproportionate smoothness @xmath1 large @xmath71 predicted class estimate type10 question form discrete symbol calculus other boundary conditions periodic general domains square discrete sine transform dst ) @xmath72 dirichlet boundary conditions rectangle discrete cosine transform dct ) neumann rectangle choice @xmath72 definition corresponding frequency variable @xmath1 robust approach use spectral elements complicated domains spectral domain defined chebyshev expansions imagine expansions prolate spheroidal wavefunctionsexpansions theory pseudodifferential operators on bounded domains difficult problem designing absorbing boundary conditions in variable media symbol expansions insights symbol - based methods solve other equations elliptic pde heat equation in variable media solution pseudodifferential smoothing form computed via scaling - and - squaring challenging example hyperbolic systems in variable smooth media time - dependent green s function not pseudodifferential fourier integral operator fio @xmath73 by @xmath74 extension of discrete symbol calculus phases interesting problem see @xcite for results fast application fio write @xmath75 for fourier series coefficients of @xmath16 in @xmath0 express psido - sum as @xmath76 interchange two sums @xmath16 differentiable @xmath77 times @xmath78 @xmath79 exponent @xmath77 chosen @xmath80 summable in @xmath81 assume @xmath82 apply fubini s theorem write @xmath83 where @xmath84 @xmath85 operator of multiplication by @xmath86 on @xmath1 sideplancherel @xmath87 triangle inequality @xmath88 sum @xmath89 converges proves theorem when @xmath90 smooth classical density argument same conclusion holds @xmath91^d) | 0.434206 | 2,617 | 1,394 | , Saving $0.1 in GPT-4. |
155,029 | we have determined the land factor @xmath0 in self - organized inas quantum dots using resonant - tunnelling experiments . with
the magnetic field applied parallel to the growth direction @xmath1 we find @xmath2 for the specific dot investigated . when the magnetic field is tilted away by @xmath3 from the growth axis
, @xmath0 gradually increases up to a value @xmath4 when @xmath5 .
its angular dependence is found to follow the phenomenological behaviour @xmath6 .
epsf.tex subject classification : 73.40 gk , 73.23 hk , 73.61 ey ; s7.12 | resonant tunnelling experiments through zero - dimensional structures are an efficient tool to access their quantized energy levels . in recent years
several groups succeeded in performing such experiments with self - assembled inas quantum dots ( qds ) embedded in the barrier of a single - barrier tunnelling device @xcite .
low - temperature experiments allowed to measure directly the land factor @xmath0 of inas dots @xcite . in this note we will report on resonant tunnelling experiments through inas qds in tilted magnetic fields . we will show that @xmath0 depends on the orientation of the magnetic field with respect to the growth direction and can be described phenomenologically by a tensor with two independent components .
the samples were grown by molecular - beam epitaxy on a highly si - doped gaas substrate with a donor concentration @xmath7 @xmath8 .
first we grew a 1-@xmath9 m thick gaas buffer layer with the same doping level followed by two 10-nm thick n - doped gaas layers with @xmath10 @xmath8 and @xmath11 @xmath8 and an 15-nm thick undoped gaas spacer . on this bottom electrode
a 10-nm thick alas barrier was deposed .
the growth of the barrier was interrupted at a thickness of 5 nm where 1.8 mono - layers of inas were embedded in the barrier . with such an inas coverage
self - assembled inas qds are formed .
the structure was terminated with a top electrode symmetric to the bottom electrode finishing with 1 @xmath9 m highly n - doped gaas ( @xmath7 @xmath8 ) . as a consequence of the high doping three - dimensional electrodes
are present on both sides of the alas barrier .
subsequent to the growth of the wafer macroscopic augeni contacts with a typical diameter of 50
@xmath9 m were annealed into the top electrode and vertical tunnelling diodes with the same diameter were processed using wet - chemical etching .
when a bias voltage is applied between the top and the bottom electrode our devices show @xmath12-@xmath13-characteristics which can be described by tunnelling through a single alas barrier @xcite .
superimposed on this coarse @xmath12-@xmath13-curve we observe distinct current steps assigned to resonant tunnelling through individual inas quantum dots @xcite , for more details see @xcite . a typical step for @xmath14 k
is shown in fig .
[ step ] . in high magnetic fields the current step observed at @xmath15
t develops into two spin - split steps with a voltage separation @xmath16 . here
@xmath17 denotes the lever factor between the energy separation emitter - dot , @xmath18 , and the total voltage drop , @xmath13 . from the temperature
dependent smearing of the current step we deduced @xmath19 .
the spin splitting of the current step is shown in fig .
[ split]a where the @xmath12-@xmath13-characteristic of the tunnelling structure at @xmath20 t is displayed . for the top curve
the magnetic field @xmath21 is oriented along the growth direction @xmath1 , the bottom trace was measured in a magnetic field perpendicular to @xmath1 . as can be seen in fig .
[ split]b , @xmath22 indeed increases linearly with magnetic field confirming the scenario of a simple zeeman splitting of the qd s energy level .
however , the land factor @xmath0 as deduced from the slope is considerably different for the two field orientations @xmath23 and @xmath5 where we find land factors @xmath24 and @xmath25 . from the temperature dependence of the step heights in high magnetic fields
we find that the low - voltage step can be assigned to tunnelling of spin - down electrons@xcite resulting in a positive @xmath26-factor as also found in @xcite . at first sight
it is quite astonishing that the land factor is far away from that of bulk inas ( @xmath27 ) .
this discrepancy can be explained qualitatively considering effects of size quantization , strain and possible other effects @xcite .
such effects may be the leakage of the electronic wave - function in the inas quantum dot into the alas barrier and the space dependent alloying of the inas dots with alas . in a simple picture
this can be described by an ( rather complex ) admixture of the land factor in alas and strained inalas to the ( size - quantized ) @xmath26-factor in the inas - dot .
in order to clarify the dependence of @xmath0 on the orientation of the magnetic field more clearly we have performed experiments in tilted magnetic fields at a temperature @xmath28 k. the angle between the magnetic field and the growth direction was stepped from @xmath29 to @xmath30 . for each angle i - v - curves were recorded at @xmath31 t and the voltage splitting between two spin - split steps was determined . from this
we deduced the angular dependent land factor , @xmath32 .
the results are shown in fig .
[ gvontheta ] . as can be seen in the figure
, @xmath33 gradually increases when tilting the magnetic field away from the growth axis . as indicated with the solid line it follows an angular dependence @xmath34 with @xmath35 and @xmath36 .
this phenomenological behaviour can be understood when regarding the zeeman contribution to the total energy in a size quantized structure . for the most general case
the corresponding hamiltonian can be written in the form @xcite @xmath37 were @xmath38 are the three spatial directions , @xmath39 are the components of the magnetic field along @xmath40 , @xmath41 is the bohr magneton and @xmath42 are the pauli spin - matrices . in the most general form
the tensor @xmath43 contains nine independent real components @xcite . to get a better physical insight we model our dot by a flat disc with a height @xmath44 nm and a diameter @xmath45 nm @xcite . in this case
the land tensor reduces to a diagonal tensor with two independent components @xmath46 and @xmath47 . in a tilted magnetic field with @xmath48 , @xmath49 and @xmath50 equation ( [ gtheta ] ) then directly follows from ( [ h ] ) .
our experiment clearly shows the anisotropic nature of spin splitting in inas quantum dots . comparing with the measured @xmath26-factor anisotropy in quantum - wires @xcite and quantum wells @xcite one would expect the largest @xmath26-factor when the magnetic field is applied in the direction of the strongest confinement .
our experiments , however , indicate that this simple relation for the @xmath26-factor anisotropy does not hold .
we observe a smaller @xmath0 if the magnetic field is applied in the growth direction where the quantum confinement is the strongest .
therefore , it is probable that the effects responsible for the @xmath26-factor anisotropies are not dominated by quantum confinement but by more complicated mechanisms such as alas alloying into the inas quantum dots , leakage of the electronic wave function into the alas barriers , spatial dependent strain , spin interactions between the dot and the electrodes etc . without any doubt
more elaborate theoretical models are necessary to clarify the possible origins of @xmath26-factor anisotropies in self - assembled inas quantum dots .
using resonant tunnelling experiments we have measured the land factor tensor of inas quantum dots .
the @xmath26-factor tensor was described by two independent tensor components @xmath51 and @xmath52 . , and p. c. main , phys .
b * 54 * , 16401 ( 1996 ) ; + t. suzuki , k. nomoto , k. taira , and i. hase , jpn .
j. appl .
phys . * 36 * , 1917 ( 1997 ) ; + m. narihiro , g. yusa , y. nakamura , t. noda , and h. sakaki , appl .
lett . * 70 * , 105 ( 1997 ) . | resonant tunnelling experiments zero - dimensional structures quantized energy levels groups self - assembled inas quantum dots barrier tunnelling device @xcite low - temperature experiments land factor @xmath0 dots experiments tilted magnetic fields @xmath0 depends on orientation magnetic field growth tensor two independent components samples grown molecular - beam epitaxy si - doped gaas substrate donor concentration @xmath7 @xmath8 grew 1-@xmath9 m thick gaas buffer layer doping level two 10-nm thick n - doped gaas layers 15-nm thick undoped gaas spacer 10-nm thick barrier deposed growth interrupted at 5 nm 1.8 mono - layers inas embedded self - assembled inas qds formed structure terminated top electrode symmetric bottom electrode 1 @xmath9 m n - doped gaas high doping three - dimensional electrodes present both sides alas barrier macroscopic augeni contacts diameter 50 @xmath9 m annealed into top electrode vertical tunnelling diodes diameter processed wet - chemical etchingbias voltage between top bottom electrode devices show @xmath12-@xmath13-characteristics tunnelling through single barrier distinct current steps resonant tunnelling through inas quantum dots typical step for @xmath14 k in fig . high magnetic fields current step @xmath15 t develops into two spin - split steps voltage separation @xmath16 @xmath17 lever factor between energy separation emitter - dot total voltage drop temperature dependent smearing current step deduced @xmath19 spin splitting shown in fig . split @xmath12-@xmath13-characteristic tunnelling structure at @xmath20 t top curve magnetic field @xmath21 oriented along growth direction @xmath1 bottom trace measured field perpendicular to @xmath1 @xmath22 increases linearly with magnetic field zeeman splitting energy level land factor @xmath0 slope different for field orientations @xmath23 @xmath5 land factors @xmath24 @xmath25 temperature dependence step high fields low - voltage step tunnelling spin - down electrons positive @xmath26-factor land factor from bulk inas @xmath27 )discrepancy explained considering size quantization strain effects @xcite leakage electronic wave - function inas dot into alas barrier space alloying inas dots alas described admixture land factor strained inalas to size - quantized @xmath26-factor in inas - dot dependence @xmath0 on orientation magnetic field experiments in tilted magnetic fields at temperature @xmath28 k angle between magnetic field growth direction stepped from @xmath29 to @xmath30 i - v - curves recorded at @xmath31 t voltage splitting between spin - split steps determined deduced angular dependent land factor @xmath32 results shown in fig . @xmath33 increases when tilting magnetic field away from growth axis follows angular dependence @xmath34 @xmath35 @xmath36 zeeman contribution to total energy in size quantized structure hamiltonian written in form @xcite @xmath37 @xmath38 three spatial directions components magnetic field bohr pauli spin - matrices tensor @xmath43 contains nine components @xcite model dot by flat disc height @xmath44 nm diameter @xmath45 nmland tensor reduces to diagonal tensor with components @xmath46 @xmath47 tilted magnetic field with @xmath48 @xmath49 @xmath50 equation gtheta ] ) follows from [ h ] ) experiment shows anisotropic spin splitting in inas quantum dots @xmath26-factor anisotropy in quantum - wires quantum wells largest @xmath26-factor when magnetic field applied in strongest confinement experiments indicate-factor anisotropy hold smaller @xmath0 if magnetic field applied in growth direction quantum confinement strongest probable effects-factor anisotropies not dominated by quantum confinement but complicated mechanisms alas alloying into inas quantum dots leakage electronic wave function into barriers spatial dependent strain spin interactions between dot electrodes elaborate theoretical models necessary clarify origins of-factor anisotropies in self - assembled inas quantum dots resonant tunnelling experiments measured land factor tensor of inas quantum dots @xmath26-factor tensor described by two independent tensor components @xmath51 @xmath52 16401 1996 ) 1997sakaki appl lett 70 105 1997 ) | 0.311648 | 1,825 | 900 | , Saving $0.1 in GPT-4. |
155,030 | we study the influence of surface anisotropy on the zero - temperature hysteretic properties of a small single - domain magnetic particle , and give an estimation of the anisotropy constant for which deviations from the stoner - wohlfarth model are observed .
we consider a spherical particle with simple cubic crystalline structure , a uniaxial anisotropy for core spins and radial anisotropy on the surface , and compute the hysteresis loop by solving the local landau - lifshitz equations for classical spin vectors .
we find that when the surface anisotropy constant is at least of the order of the exchange coupling , large deviations are observed with respect to the stoner - wohlfarth model in the hysteresis loop and thereby the limit - of - metastability curve , due to the non - uniform cluster - wise reversal of the magnetisation .
= -3 cm | in this work , we deal with the effect of surface anisotropy on the hysteretic properties ( hysteresis loop and limit - of - metastability curve , the so - called sw astroid @xcite ) of a single - domain spherical particle of simple cubic crystalline structure , uniaxial anisotropy in the core , and radial single - site anisotropy for spins on the boundary . we compute the hysteresis loop and thereby the critical field of this particle by solving , at zero temperature , the local landau - lifshitz equation ( lle ) derived from the classical anisotropic dirac - heisenberg model in field ( [ dh ] ) .
doing this for different values of the polar angle @xmath0 between the applied field and the core easy axis , renders the critical field as a function of @xmath0 , that is the sw astroid .
finally , our model hamiltonian reads @xcite @xmath1 where @xmath2 is the unit spin vector on site @xmath3 @xmath4 is the uniform magnetic field applied in a direction @xmath0 with respect to the reference @xmath5 axis , @xmath6 is the total number of spins ; @xmath7 is the nearest - neighbour exchange coupling ; the last term in ( [ dh ] ) is the uniaxial anisotropy energy with easy axis @xmath8 and constant @xmath9 .
core spins have an easy axis along @xmath5 and constant @xmath10 , while surface spins have radial easy axes and constant @xmath11 . in the sequel , we use the reduced parameters , @xmath12 the only free parameter of our model is the surface anisotropy constant @xmath13 .
so upon varying it we estimate its value at which deviations from the ( macroscopic ) stoner - wohlfarth model are observed .
details of the method used here and more results can be found in ref.@xcite .
fig . [ fig1_jap ] shows that when @xmath13 becomes comparable with @xmath14 , the competition between exchange coupling and surface anisotropy produces large deviations from the sw model .
namely , the hysteresis loop exhibits multiple jumps , which can be attributed to the switching of different spherical shells of spins starting from surface down to the centre .
the hysteresis loop is then characterised by two field values : one that marks the limit of metastability , called the _ critical field _ or the saturation field , and the other that marks the magnetization switching , and is called the _ switching field _ or the coercive field . in fig .
[ fig2_jap]a we present the variation with the particle s diameter of the critical field @xcite ( in diamonds ) obtained from the numerical solution of the lle for @xmath15 , and ( in circles ) the sw critical field multiplied by the core - to - volume ratio , i.e. @xmath16 , analytically obtained upon assuming infinite exchange coupling .
this figure also shows that for such a value of @xmath13 , when the field is applied along the core easy axis , the critical field increases with the particle s size , and that for @xmath17 all hysteresis loops can be scaled with those rendered by the sw model .
( 9,8.5)(-1,-9.0 ) ( 9,8.5)(0,-9.0 ) on the other hand , these results clearly show that even in a large particle there still exist some magnetic inhomogeneities that lead to deviations from the sw hypothesis of uniform rotation , as this is only rigorously realized in an infinite system . moreover , fig .
[ fig2_jap ] shows that the critical field of a spherical particle with @xmath18 can be obtained from the sw model through a scaling with constant @xmath16 .
one should also note that the astroid for all particle sizes falls inside that of sw , see fig .
[ fig2_jap]b .
therefore , for @xmath19 our model renders hysteresis loops and limit - of - metastability curves that scale with the sw results for all values of the angle @xmath0 , the scaling constant being @xmath16 , which is smaller than unity for a particle of any finite size . on the other hand ,
the critical field increases with the particle s size and tends to the sw critical field in infinite systems .
( 9,8.5)(0,-9.0 ) for larger values of @xmath20 , but @xmath21 , we still have some kind of scaling but the corresponding constant now depends on @xmath0 .
this is reflected by a deformation of the limit - of - metastability curve .
more precisely , as shown by fig .
[ fig3_jap]a , the latter is depressed in the core easy direction and enhanced in the perpendicular direction . however , there is still only one jump in the hysteresis loop implying that the magnetization reversal can be considered as uniform . on the other hand , fig .
[ fig3_jap]b shows that for @xmath22 there are two new features in comparison with the previous case ( compare fig .
[ fig2_jap]a ) : the values of the switching field are much higher , and more importantly , its behaviour as a function of the particle s size is opposite to that of the previous case , as now the field increases when the particle s size is decreased . for such high values of @xmath13 ( @xmath23 )
surface spins are aligned along their easy axes , and because of the strong exchange coupling they also drive core spins in their switching process .
thus the smaller the particle the larger the surface contribution , and the larger the field required for a complete reversal of the particle s magnetization .
( 9,8.5)(-1,-9.0 ) in fig . [ fig4_jap ] @xmath24 first slightly decreases for @xmath25 and then increases , and when @xmath13 becomes of the order of @xmath14 it jumps to large values .
this plot illustrates the different regimes discussed above according to the value of the ratio @xmath20 . for an order of magnitude estimate of @xmath11 and the critical field @xmath26 at the point @xmath27 , consider a 4 nm spherical cobalt particle for which @xmath28 8 mev , @xmath29 erg/@xmath30 .
then , @xmath31 erg/@xmath32 , and @xmath33 t , when the field is applied along the core easy axis .
considering the fact that experiments on nanoparticles show that the switching field does increase when the particle s size decreases ( see e.g. @xcite for cobalt particles ) , we may conclude that the anisotropy constant @xmath11 is at least of the order of the exchange coupling @xmath34 , inasmuch as we can assume radial anisotropy on the surface , as is usually done in the literature . then
, as discussed above , for such values of @xmath11 , large deviations are observed with respect to the sw model in the hysteresis loop and thereby the limit - of - metastability curve , since in this case the magnetisation reverses its direction in a non - uniform manner via a progressive switching of spin clusters .
so to deal with these features one has to resort to microscopic approaches such as the one used in this work .
a work in progress applies the present technique to the cubo - octahedral cobalt particles with a diameter of about 3 nm recently reported on in @xcite . | effect surface anisotropy on hysteretic properties hysteresis loop limit - - metastability curve sw astroid single - domain spherical particle cubic crystalline structure uniaxial anisotropy core radial single - site anisotropy spins boundary compute hysteresis loop critical field solving zero temperature local landau - lifshitz equation classical anisotropic dirac - heisenberg model values polar angle @xmath0 applied field core easy axis critical field function of @xmath0 sw astroid model hamiltonian reads @xcite @xmath1 @xmath2 unit spin vector site uniform magnetic field direction axis total number spins @xmath7 nearest - neighbour exchange coupling last uniaxial anisotropy energy easy axis @xmath8 constant @xmath9 core spins easy axis @xmath5 constant @xmath10 surface spins radial easy axes constant @xmath11 reduced parameters free parameter surface anisotropy constant @xmath13 estimate value deviations from stoner - wohlfarth model observed details method results in ref.@xcite figfig1_jap ] shows @xmath13 comparable with @xmath14 competition between exchange coupling surface anisotropy produces deviations from sw model hysteresis loop exhibits multiple jumps attributed to switching spherical shells spins to loop characterised by two field values limit of metastability critical field magnetization switching switching field coercive field fig fig2_jap variation with particle s diameter of critical field @xcite obtained from numerical solution for @xmath15 sw critical field multiplied by core - to - volume ratio @xmath16 infinite exchange coupling @xmath13 field applied along core axis critical field increases with particle s size for @xmath17 all hysteresis loops scaled with sw model results show large particle magnetic inhomogeneities to deviations from sw hypothesis of uniform rotation in infinite system fig . fig2_jap shows critical field of spherical particle with @xmath18 obtained from sw model through scaling with constant @xmath16 astroid for all particle sizes falls inside sw see fig . fig2_jap]b@xmath19 model renders hysteresis loops limit - of - metastability curves scale with sw results for values angle @xmath0 scaling constant @xmath16 smaller than unity for particle finite size critical field increases with particle s size tends to sw critical field in infinite systems 9,8.5)(0,-9.0 ) larger values of @xmath20 @xmath21 scaling constant depends on @xmath0 reflected by deformation limit - of - metastability curve depressed in core direction enhanced in perpendicular direction one jump in hysteresis loop magnetization reversal uniform @xmath22 two new features values switching field higher behaviour particle s size opposite field increases when particle size decreased high values of @xmath13 ( @xmath23 ) surface spins aligned along easy axes strong exchange coupling drive core spins switching process smaller particle larger surface contribution larger field required for complete reversal of particle magnetization ( 9,8.5)(-1,-9.0 ) fig .fig4_jap @xmath24 decreases for @xmath25 increases @xmath13 @xmath14 jumps to large values plot illustrates regimes according to ratio @xmath20 order magnitude estimate of @xmath11 critical field @xmath26 point @xmath27 consider 4 nm spherical cobalt particle @xmath28 8 mev @xmath29 erg/@xmath30 @xmath31 erg/@xmath32 @xmath33 t field applied along core axis experiments nanoparticles show switching field when particle s size decreases @xcite cobalt anisotropy constant @xmath11 order exchange coupling @xmath34 radial anisotropy surface values @xmath11 large deviations observed sw model hysteresis loop limit - of - metastability curve magnetisation reverses direction via progressive switching of spin clusters deal to microscopic approaches work work in progress applies technique to cubo - octahedral cobalt particles diameter 3 nm reported in @xcite | 0.266696 | 1,625 | 837 | , Saving $0.0 in GPT-4. |
155,031 | we describe techniques for measuring the central masses of galaxies using emission - line kinematics observed with the _ hubble space telescope_. for accurate results , it is necessary to model various instrumental effects , particularly the blurring due to the telescope psf and the width of the spectroscopic aperture .
observations of nuclear gas disks often reveal substantial internal velocity dispersions in the gas , suggesting that the disks may be partially pressure - supported .
we also describe a technique for fitting 2-dimensional spectroscopic data directly in pixel space .
this method may be useful for objects such as m84 that show highly complex and asymmetric line profiles . | supermassive black holes are thought to occur in the nuclei of all massive galaxies , or at least in those galaxies having a bulge component .
recent studies have shown that the black hole mass is correlated loosely with the host galaxy bulge luminosity ( kormendy & richstone 1995 ; magorrian et al .
1998 ) and correlated tightly with the stellar velocity dispersion of the bulge ( ferrarese & merritt 2000 ; gebhardt et al .
however , the slope of the @xmath0 relation , and the amount of intrinsic scatter in the correlation , remain somewhat controversial . for further progress , more black hole masses
must be measured , and with the highest accuracy possible .
disks of ionized gas and dust with radii of tens to hundreds of parsecs are found in the nuclei of @xmath1 of nearby galaxies . with _ hst _
spectra , it is possible to map out the velocity structure of these disks and determine the central masses . in principle , the measurement is straightforward : the goal is simply to measure the rotation speed of a circular disk .
however , the region with the greatest diagnostic power for determining the central mass is generally at or near the resolution limit of _
hst_. near the nucleus , there are steep gradients in rotation velocity and emission - line surface brightness across the spectroscopic aperture . in addition , the gas often has a substantial internal velocity dispersion , which may affect the dynamical properties of the disks .
all of these effects must be taken into account in order to derive accurate black hole masses from the data .
* instrumental effects . * the velocity fields in circumnuclear disks can be measured with the stis spectrograph on _ hst _ , usually with a slit width of 01 or 02 . due to the telescope psf and the nonzero aperture size
, the innermost keplerian rotation curve is blurred into a smooth velocity gradient across the nucleus . only for the largest black holes
will the keplerian portion of the velocity field be partly resolved .
kinematic analyses which do not take into account this instrumental blurring ( e.g. , ferrarese et al .
1996 ) will tend to underestimate the true black hole masses , because an unblurred model will need a smaller central mass to reach a given rotation velocity near the nucleus .
we have developed modeling software that performs a complete simulation of the stis observation , so that kinematic models for the rotating disk can be compared with the data in detail .
the code calculates the propagation of full line profiles through the spectrograph slit , accounting for psf blurring , aperture size , and the apparent wavelength shift of light that enters the slit off - center .
these techniques have been applied to measure the central mass of the s0 galaxy ngc 3245 ( barth et al .
. * emission - line surface brightness . *
observations of the velocity field are weighted by the surface brightness of emission lines , which sometimes show a sharp , nearly unresolved `` spike '' at the nucleus .
the measured velocities near the nucleus can be very sensitive to the brightness profile of this central spike .
in addition , a nonuniform distribution of emission - line surface brightness can distort the shape of the radial velocity curves if a patch of emission is located off - center in the slit ( figure 1 ) .
* disk orientation parameters . *
at least 3 slit positions are needed to fully constrain the orientation of the disk . for ngc 3245
, we used 5 parallel positions covering the inner arcsecond .
the off - nuclear slit positions give enough kinematic information to tightly constrain the disk inclination and the major axis position angle , eliminating what would otherwise be a significant source of uncertainty in the analysis .
* intrinsic velocity dispersion and asymmetric drift . *
our modeling code includes all of the major sources of line broadening that contribute to the observed linewidths .
these include the point - source line - spread function , the broadening due to the nonzero width of the slit , the diffusion of charge on the ccd , and the rotational broadening for the portion of the disk subtended by the slit aperture .
the models for ngc 3245 demonstrate that the disk can not be dynamically cold : the gas has an intrinsic velocity dispersion which rises to @xmath2 km s@xmath3 at the nucleus .
substantial internal velocity dispersions are observed in most circumnuclear disks of this type .
this suggests that the disks may be partially pressure - supported , and that the black hole mass inferred from pure circular rotation models may be an underestimate . for ngc 3245 , we apply a correction for this effect using the asymmetric drift equation of stellar dynamics ; in this case the effect on the black hole mass is only 12% , but it may be larger for other galaxies with more turbulent disks . for ngc 3245 ,
we find @xmath4 @xmath5 .
the error budget includes uncertainties due to the kinematic model fitting , the measurement of the stellar luminosity profile , the distance to the galaxy , and differences between the kinematics of h@xmath6 and [ n ii ] .
the analysis technique described above , based on fitting the measured velocity curves , only works well if the emission - line profiles are close to gaussian .
this is not always the case ; for example , the galaxy m84 ( ngc 4374 ) has extremely complex line profiles near the nucleus .
bower et al . (
1998 ) found that the line profiles in m84 show two kinematic components , a fast keplerian component and a slowly rotating one .
they fitted keplerian disk models to the fast component and derived a central mass of @xmath7 @xmath5 .
more recently , maciejewski & binney ( 2001 ) showed that the two kinematic components actually both arise from the disk itself , because the slit is wide enough to subtend different portions of the disk having a wide range of velocities . as a result ,
the position and velocity information become badly entangled in the central regions , and it is effectively impossible to measure a well - defined rotation curve for the inner disk . to circumvent this problem ,
we have adapted our code to fit the emission profiles of the rotating disk _ directly _ in pixel space .
this method makes use of all available information in the data , including the high - velocity wings of the line profiles .
preliminary results for m84 are shown in figure 2 . with @xmath8 @xmath5 ,
the model naturally accounts for much of the structure in the line profiles : narrow cores and broad wings at the nucleus , extreme asymmetries , and even double - peaked profiles at some locations . while the fits are by no means perfect , these preliminary results demonstrate that direct profile fitting is a promising technique for extracting black hole masses from data showing highly asymmetric emission lines | supermassive black holes occur in nuclei galaxies studies shown black hole mass correlated with host galaxy bulge luminosity stellar velocity dispersion slope of @xmath0 relation intrinsic scatter remain controversial more black hole masses must be measured highest accuracy disks of ionized gas dust with radii tens to hundreds of parsecs found in nuclei @xmath1 nearby galaxies with _ hst _ spectra possible to map velocity structure disks determine central masses measurement rotation speed of circular disk region greatest diagnostic power for central mass at or near resolution limit of _ hst_ near nucleus steep gradients in rotation velocity emission - line surface brightness across spectroscopic aperture gas has internal velocity dispersion affect dynamical properties disks effects derive accurate black hole masses velocity fields in circumnuclear disks measured with stis spectrograph on _ hst _ slit width of 01 or 02 due to telescope psf nonzero aperture size innermost keplerian rotation curve blurred into smooth velocity gradient across nucleuslargest black holes keplerian portion velocity field partly resolved kinematic analyses instrumental blurring ferrarese et 1996 underestimate true black hole masses unblurred model smaller central mass rotation velocity near nucleus developed modeling software simulation stis observation kinematic models rotating disk data code calculates propagation full line profiles through spectrograph slit for psf blurring aperture size wavelength shift light slit off - center techniques applied measure central mass s0 galaxy ngc 3245 barth et al emission - line surface brightness observations velocity field weighted by surface brightness emission lines sharp unresolved spike at nucleus velocities near nucleus sensitive to brightness central spike nonuniform distribution emission - line surface brightness distort radial velocity curves patch emission off - center slit disk orientation parameters 3 slit positions needed constrain orientation disk ngc 3245 used 5 parallel positions inner arcsecond off - nuclear slit positions constrain disk inclination major axis position angle eliminating uncertainty intrinsic velocity dispersion asymmetric drift modeling code includes major sources line broadening linewidthsinclude point - source line - spread function broadening nonzero width slit diffusion charge ccd rotational broadening disk subtended by slit aperture models ngc 3245 disk dynamically cold gas intrinsic velocity dispersion rises to @xmath2 s@xmath3 at nucleus internal velocity dispersions in circumnuclear disks disks partially pressure - supported black hole mass circular rotation underestimate ngc 3245 correction asymmetric drift equation dynamics effect black hole mass 12% larger other galaxies ngc 3245 @xmath4 @xmath5 error budget includes uncertainties kinematic model fitting stellar luminosity profile distance to galaxy differences kinematics h@xmath6 [ n ii ] analysis technique works if emission - line profiles close to gaussian not galaxy m84 ( ngc 4374 ) complex line profiles near nucleus bower line profiles show two kinematic components fast keplerian slowly rotating one fitted keplerian disk models fast component derived central mass of @xmath7 @xmath5 maciejewski & binney 2001 ) showed two kinematic components arise from disk slit velocities position and velocity information entangled in central regions impossible to measure rotation curve for inner disk adapted code to fit emission profiles rotating disk directly in pixel space method information data including high - velocity wings line preliminary results for m84 shown in figure 2 with @xmath8 @xmath5 model accounts for structure line profiles narrow cores broad wings at nucleus extreme asymmetries double - peaked profiles perfect results demonstrate direct profile fitting promising technique for extracting black hole masses from asymmetric emission lines | 0.264426 | 1,430 | 694 | , Saving $0.0 in GPT-4. |
155,044 | a _ unified _ description of the near - side and away - side structures observed in two - particle correlations as function of @xmath0 is proposed for low to moderate transverse momentum .
it is based on the combined effect of tubular initial conditions and hydrodynamical expansion . | one of the most striking results in relativistic heavy ion collisions , at rhic and the lhc , is the existence of structures in the two - particle correlations plotted as function of the pseudorapidity difference @xmath1 and the angular spacing @xmath2 .
these structures may have a common hydrodynamic origin : the combined effect of longitudinal high energy density tubes ( leftover from initial particle collisions ) and transverse expansion . in order to compare with the data mentioned above , event - by - event 3 + 1 hydrodynamics must be used . in this type of approach ,
some initial conditions are generated , an hydrodynamics code is run and results are stored ; this is done many times thus mimicking experience .
this method has been developed since 2001 by the brazilian collaboration spherio @xcite ( some typical results can be seen in @xcite ) .
it has been also studied since 2007 by of h.petersen et al .
@xcite . in 2010 , k.werner et al . @xcite and b.schenke et al . @xcite
have also started to use this method .
all groups except the last one assume an ideal fluid .
( recently , various groups have also been working on event - by - event 2 + 1 hydrodynamics , see e.g. @xcite . ) in the nexspherio approach , initial conditions , generated with the nexus code , have tubular structures and two particle correlations exhibit _ near _ and _ away - side _
ridges @xcite as can be seen in figure [ ic2d ] . in the calculation by k.werner et al .
, initial conditions , obtained with the epos code , also have tubular structures and two particle correlations exhibit near - side @xcite and _ small _ away - side ridges @xcite .
( the other two groups mentioned above have no result on two - particle correlations . )
nexspherio central au+au collisions at 200 gev a : initial energy density and example of resulting two - particle correlation.,title="fig:",width=132 ] nexspherio central au+au collisions at 200 gev a : initial energy density and example of resulting two - particle correlation.,title="fig : " ] nexspherio central au+au collisions at 200 gev a : initial energy density and example of resulting two - particle correlation.,title="fig:",width=170 ] in addition to reproducing both the near and away - side structures , nexspherio leads to good qualitative agreement with various data @xcite : 1 ) for fixed @xmath3 and increasing @xmath4 , the near - side and away - side peaks decrease for central collisions while for fixed @xmath4 and increasing @xmath3 , the peaks increase , 2 ) when going from central to peripherical collisions , the near - side ridge decreases and the away - side ridge changes from double to single hump , 3 ) for a mid - central window , the away - side ridge changes from single peak for in - plane trigger to double peak for out - of - plane trigger while for central collisions , it is always double - peaked .
when using nexspherio , it is not clear how the various structures in the two - particle correlations are generated . to investigate this , we study the transverse expansion of a _ realistic _
slice of matter with only one tube ( cf .
figure [ tube ] ) .
longitudinal expansion is assumed to be boost invariant and freeze out to occur at some constant temperature . as seen in figure [ 1dist ]
, the single particle angular distribution has two peaks located on both sides of the position of the tube , more or less independently of the transverse momentum value .
the peak spacing is @xmath5 ( this is not a parameter . ) the resulting two - particle angular correlation has a large central peak at @xmath6 ( corresponding to the near - side ridge ) and two smaller peaks respectively at @xmath7 ( associated to the double - hump ridge ) .
we have checked that this structure is robust by studying the effect of the height and shape of the background , initial velocity , height , radius and location of the tube @xcite .
the occurrence of the two - peak emission can be understood from figure [ 1dist ] .
as time goes on , as a consequence of the tube expansion , a hole appears at the location of the high - energy tube ( see also @xcite ) .
this hole is surrounded by matter that piles up in a roughly semi - circular cliff of high energy density matter .
the two extremities of the cliff emit more fast particles than the background giving rise to the two - peaks in the single - particle angular distribution . left : choice of a typical nexus tube located near the border .
center : choice of the energy density profile for the tube and averaged background ( solid lines ) .
right : resulting transverse slice of matter.,title="fig : " ] left : choice of a typical nexus tube located near the border . center : choice of the energy density profile for the tube and averaged background ( solid lines ) .
right : resulting transverse slice of matter.,title="fig:",width=143 ] left : choice of a typical nexus tube located near the border . center : choice of the energy density profile for the tube and averaged background ( solid lines ) .
right : resulting transverse slice of matter.,title="fig : " ] left and center : single particle angular distributions and resulting two - particle correlation .
right : temporal evolution of the energy density in the slice .
, title="fig:",height=136 ] left and center : single particle angular distributions and resulting two - particle correlation .
right : temporal evolution of the energy density in the slice .
, title="fig:",height=166 ] left and center : single particle angular distributions and resulting two - particle correlation .
right : temporal evolution of the energy density in the slice .
, title="fig:",height=136 ] we have also generalized this one tube model to non - central collisions @xcite . to get final results
, one must sum on all tube positions . for a given tube position ,
the single particle angular distribution is more complicated ( the double - peak signature of the tube may be hidden by the elliptic flow ) and the shape of the two - particle angular distribution depends on the trigger angle with respect to the reaction plane . after subtracting the elliptic flow and averaging on the tube angular position ,
the two - particle correlation is found to be single peaked for in plane trigger and double peaked for out - of - plane trigger in agreement with data . with these information
, we can discuss what happens in a more complex event such as a nexus event .
only the outer tubes need to be considered and will contribute with rather similar two - peak emission pattern at various angles in the single particle angular distribution .
the two - particle correlation has a well - defined main structure similar to that of a single tube surrounded by several other peaks and depressions due to trigger and associated particles coming from different tubes . when averaging over many events they disappear and only the main one - tube like structure is left .
event - by - event 3 + 1 hydrodynamic with tubular initial conditions predicts two - particle correlations as function of @xmath0 in qualitative agreement with data . using a simpler model ,
we have seen that near - side and away - side structures are related to a `` two - horn '' emission from each tube .
it has been suggested @xcite that both these near - side and away - side structures are due to the triangular flow ( @xmath8 ) produced by a triangular shape ( or triangularity @xmath9 ) present in the initial conditions .
the one - tube model does have triangularity and triangular flow .
however , one can add three inner tubes to cancel @xmath9 and still have the same @xmath8 because only the outer tube contributes to this quantity . by removing this outer tube , leaving only the three inner tubes
, one can have @xmath10 while @xmath9 decreases a little but is non - zero .
both @xmath9 ( due to @xmath11 ) and @xmath8 are sensitive to external tubes however @xmath8 more so . as a consequence , on an event - by - event basis @xmath12 will not be satisfied .
a more systematic study is in progress ( see also @xcite ) .
we acknowledge funding from cnpq and fapesp .
99 aguiar c e , hama y , kodama t and osada t 2001 _ j.phys .
g _ * 27 * 75 ; 2001 _ j.phys . g _ * 27 * 551 ; 2002 _ nucl . phys .
a _ * 698 * 639c @xmath13 socolowski jr o , grassi f , hama y and kodama t 2004 _ phys . rev .
lett . _ * 93 * 18230 andrade r p g , grassi f , hama y , socolowski jr o and kodama t 2006 _ phys .
lett . _ * 97 * 202302 andrade r p g , grassi f , hama y , kodama t and qian w l 2008 _ phys .
* 101 * 112301 takahashi j , tavares b m , qian w l , andrade r p g , grassi f , hama y , kodama t and xu n 2009 _ phys .
lett . _ * 103 * 242301 petersen h , steinheimer j , burau g , bleicher m , stcker h 2008 _ phys .
c _ * 78 * 044901 werner k , karpenko iu , pierog t , bleicher m and mikhailov k ( 2010 ) _ phys.rev.c _ * 82 * 044904 schenke b , jeon s and gale c 2011 _ phys .
lett . _ * 106 * 042301 holopainen h , niemi h and eskola k j 2011 _ phys . rev .
c _ * 83 * 034901 qiu z and heinz y arxiv:1104.0650 .
werner k _ private communication _
andrade r p g , grassi f , hama y and qian w l 2011 _ nucl phys a _ * 854 * 81 hama y , andrade r p g , grassi f and qian w l 2010 _ nonlin . phenom .
complex sys . _
* 12 * 446 staig p and shuryak e arxiv:1105.0676 andrade r p g , grassi f , hama y and qian w l arxiv:1012.5275 alver b and roland g 2010 _ phys .
c _ * 81 * 054905 qin g y , petersen h , bass s a and mller b 2010 _ phys .
c _ * 82 * 064903 | striking results in relativistic heavy ion collisions at rhic lhc structures in two - particle correlations pseudorapidity difference angular spacing @xmath2 structures common hydrodynamic origin longitudinal high energy density tubes transverse expansion compare event - by - event 3 + 1 hydrodynamics initial conditions generated hydrodynamics code run results stored experience method developed since 2001 by brazilian collaboration spherio @xcite results studied since 2007 by h.petersen et al . @xcite 2010 k.werner et al . @xcite b.schenke et al . @xcite method all groups assume ideal fluid groups on event - by - event 2 + 1 hydrodynamics nexspherio approach initial conditions nexus code have tubular structures two particle correlations exhibit near away - side ridges figure calculation by k.werner et al . initial conditions epos code tubular structures particle correlations - ridges other two groups no result on two - particle correlations nexspherio central au+au collisions at 200 gev a initial energy density resulting two - particle correlation=132 nexspherio central au+au collisions at 200 gev a initial energy density two - particle correlation collisions 200 gev initial energy density two - particle correlation=170 reproducing near away - side structures nexspherio qualitative agreement with data fixed @xmath3 @xmath4 near - side away - side peaks decrease central collisions fixed @xmath4 @xmath3 peaks increase central to peripherical collisions near - side ridge decreases away - side ridge changes from double to single hump mid - central window away - side ridge changes single to double peak central collisions always double - peaked nexspherio clear structures two - particle correlations generated study transverse expansion of slice matter one tube figure longitudinal expansion boost invariant freeze out constant temperature figure single particle angular distribution two peaks both sides tube independently transverse momentum value peak spacing @xmath5 parameter two - particle angular correlation large central peak at @xmath6 near - side ridge two smaller peaks at @xmath7 double - hump ridgechecked structure robust studying effect height shape background initial velocity height radius location of tube @xcite occurrence two - peak emission understood from figure [ 1dist ] time tube expansion hole appears at location high - energy tube @xcite hole surrounded by matter in semi - circular cliff of high energy density matter two extremities cliff emit more fast particles rise two - peaks in single - particle angular distribution left choice typical nexus tube near border choice energy density profile averaged background solid lines transverse slice of matter choice typical nexus tube border energy density profile background transverse slice of matter choice typical nexus tube near border energy density profile averaged background solid transverse slice of matter left center single particle angular distributions two - particle correlation right temporal evolution of energy density in slice single particle angular distributions two - particle correlation temporal evolution of energy density in slice single particle angular distributions two - particle correlation right temporal evolution of energy density in slicetitle="fig:",height=136 generalized one tube model to non - central collisions final results sum all tube positions tube position single particle angular distribution complicated double - peak signature hidden by elliptic flow shape depends on trigger angle reaction plane subtracting elliptic flow averaging tube angular position two - particle correlation single peaked in plane trigger double peaked out - of - plane trigger data discuss complex event nexus event outer tubes considered contribute similar two - peak emission pattern various angles distribution two - particle correlation defined main structure single tube surrounded by peaks depressions due to trigger particles different tubes averaging over many events disappear main one - tube structure left event - by - event 3 + 1 hydrodynamic with tubular initial conditions predicts two - particle correlations agreement data near - side away - side structures related to two - horn '' emission from each tube due to triangular flow triangular shape @xmath9 initial conditions one - tube model triangularity triangular flow add three inner tubes to cancel @xmath9 same @xmath8 outer tube contributesremoving outer tube leaving three inner tubes @xmath10 @xmath9 decreases non - zero @xmath9 @xmath8 sensitive to external tubes @xmath8 more event event @xmath12 not satisfied systematic study progress @xcite acknowledge funding cnpq fapesp aguiar c e hama y kodama t osada t 2001 j.phys . g 27 * 75 27 * 551 2002 698 * 639c @xmath13 socolowski jr grassi f hama y kodama t 2004 phys 93 * 18230 andrade grassi 2006 97 * 202302 andrade qian w l 2008 101 * 112301 takahashi j tavares b qian w l andrade xu n 2009 103 * 242301 petersen h steinheimer j burau g bleicher m stcker h 2008 phys 78 * 044901 werner k karpenko iu pierog t bleicher m mikhailov k 2010 phys.rev.c 82 * 044904 schenke b jeon s gale c 2011 106 * 042301 holopainen h niemi h eskola k j 201183 034901 qiu z heinz:1104.0650 werner private communication andrade grassi hama qian w l 2011 nucl phys 854 81 hama andrade grassi qian w l 2010 nonlin complex sys 12 446 staig p shuryak e:1105.0676 andrade grassi hama qian w.5275 alver roland g 2010. 81 054905 qin g petersen bass s mller b 2010 phys 82 064903 | 0.352731 | 2,365 | 1,191 | , Saving $0.1 in GPT-4. |
155,048 | there is growing evidence that the majority of the energy density of the universe is not baryonic or dark matter , rather it resides in an exotic component with negative pressure .
the nature of this ` quintessence ' influences our view of the universe , modifying angular diameter and luminosity distances . here
, we examine the influence of a quintessence component upon gravitational lens time delays . as well as a static quintessence component , an evolving equation of state
is also considered .
it is found that the equation of state of the quintessence component and its evolution influence the value of the hubble s constant derived from gravitational lenses
. however , the differences between evolving and non - evolving cosmologies are relatively small .
we undertake a suite of monte carlo simulations to examine the potential constraints that can be placed on the universal equation of state from the monitoring of gravitational lens system , and demonstrate that at least an order of magnitude more lenses than currently known will have to be discovered and analysed to accurately probe any quintessence component .
cosmology : theory cosmological parameters gravitational lensing | the searches for supernovae at cosmological distances have proved very successful , providing evidence that , while topologically flat , the majority of energy in the universe is in the form of an exotic component with negative pressure ( riess et al .
1999 ; perlmutter et al . 1999 ) .
the recent identification of a supernova at @xmath0 ( riess et al .
2001 ) has provided further weight to these claims ( turner & riess 2001 ) , which suggest that this component may differ from the classical cosmological constant @xmath1 . termed ` quintessence ' , or more colloquially ` dark energy ' , this has an equation of state of the form @xmath2 , where @xmath3 is the pressure and @xmath4 the density .
a @xmath5 opposes the action of gravity and drives the cosmological expansion to accelerate .
linder ( 1988a ; 1988b ) has examined the physical nature of various quintessence components ; with @xmath6 equating to non - relativistic matter ( dust ) , @xmath7 being radiation and @xmath8 , a classical cosmological constant .
more exotic components are ; massless scalar fields @xmath9 , cosmic string networks @xmath10 , and two - dimensional topological defects @xmath11 . as well as the supernova programs , other approaches , such as gravitational lensing statistics ( cooray & huterer 1999 ) , geometrical probes of the @xmath12 forest ( hui , stebbins & burles 1999 ) and galaxy distributions ( yamamoto & nishioka 2001 ) , and classical angular - size redshift tests ( lima & alcaniz 2001 ) , will provide complementary probes of the universal equation of state .
the value of the quintessence component , @xmath13 , influences our view of the universe , modifying the various distances used in mapping the cosmos . this paper concerns itself with the influence of @xmath13 on angular diameter distances , especially in relation to the determination of the hubble s constant from the measurement of time delays in gravitational lens systems . unlike local determinations of hubble s constant ( e.g. freedman et al . 2001 ) , the cosmological nature of gravitational lenses means that they are more sensitive to the underlying cosmological parameters .
section [ background ] briefly covers the basic formulae for generalized angular diameter distances in quintessence cosmologies , while in section [ timedelay ] we consider the influence of @xmath13 on the determination of @xmath14 from lensed systems .
section [ evolve ] extends this analysis to simple models of an evolving quintessence component . in section [ montecarlo ]
a series of monte carlo simulations are undertaken to estimate the efficacy of this approach in probing the cosmological equation of state , while in section [ speculate ] we speculate on the possability that current observations of gravitational lens systems may suggest that @xmath15 .
the conclusions of this study are presented in section [ conclusions ] .
while there has been a resurgence in quintessence cosmology , the generalized cosmological equations for such universes were presented more than a decade ago by linder ( 1988a;1988b ) , including a generalized form of the dyer - roeder equation for the evolution of a bundle of rays traveling from a distant source ( dyer & roeder 1973 ) . expressing the angular diameter distance as @xmath16 , the generalized beam equation is given by @xmath17\dot{r } + \sum_w \frac{3 r ( 1+w)\alpha_w(z)\omega_w(z)}{2(1+z)^2 } = 0 \label{linder}\ ] ] where @xmath18 is the density , in units of the critical density @xmath19 , of a contributor to the total energy - density of the universe with an equation of state @xmath20 , and where @xmath3 is its pressure and @xmath4 is its density .
this is given by @xmath21 ^ 2 \label{omega}\ ] ] where @xmath22 is the contribution of this component to the present energy - density budget at the present epoch , and the critical density is given by @xmath23 .
here , @xmath24 is a generalized form of the hubble parameter and is given by ; @xmath25^\frac{1}{2 } \label{hubble}\ ] ] where @xmath26 , and @xmath27 , is related to the overall curvature of the universe .
@xmath28 is a generalized form of the deceleration parameter and is given by @xmath29 finally , equation [ linder ] also contains the parameter @xmath30 which represents how much of the fluid lies in the beam and influences the the evolution of a ray bundle . for a universe containing matter , the solution to equation [ linder ] with @xmath31 represents the classic dyer - roeder ` filled beam ' distance , while @xmath32 is the ` empty beam ' distance ( dyer & roeder 1973 ) . when solving equation [ linder ] , the boundary conditions need to be defined .
these are @xmath33 \label{boundary}\end{aligned}\ ] ] equation [ linder ] was integrated using a runge - kutta scheme ( the rksuite package from www.netlib.org ) and compared to both analytic results and the minimum angular extent redshifts in quintessence cosmologies as tabulated in lima & alcaniz ( 2000 ) ; excellent agreement was found . throughout this work ,
filled - beam distances @xmath34 are employed .
refsdal ( 1964 ) was the first to note that cosmological parameters could be determined from the measurement of a time delay between the relative paths taken by light though a gravitational lens system .
since the discovery of multiply - imaged quasars , this has become the goal of a number of monitoring campaigns ( e.g. cohen et al .
2000 ; oscoz et al . 2001 ; patnaik & narashima 2001 ) , although these analyses are frustrated by degeneracies in the derived mass models .
the cosmological model simply enters the determination of hubble s constant ; @xmath35 where @xmath36 are the normalized angular diameter distances between and observer @xmath37 , lens @xmath38 and source @xmath39 , and @xmath40 is the measured time delay between an image pair .
giovi & amendola ( 2001 ) examined the influence on a quintessence component on gravitational lens time delays and the determination of hubble s constant .
their analysis , however , was mainly concerned with the influence of the clumping of material and the dependence on @xmath14 of whether distances are empty - beam or full - beam . here , a different approach is considered ; solving equation [ linder ] , equation [ hubbletime ] is evaluated for a range of quintessence components . for the study ,
the redshifts of the lens and source in seven gravitational lens systems that are favourable for time delay measures were considered [ table 2 in giovi & amendola ( 2001 ) , q0957 + 561 and b1608 + 656 ] , plus two fiducial redshift pairs of @xmath41 and @xmath42 .
while there are currently no lensed systems with established time delays at these particular redshifts , there are several potential systems ; e.g. the quadruple lens h1413 + 117 at a redshift of 2.55 , with a lens redshift , established from prominent absorption features , at @xmath43 and he1104 - 1805 at a redshift of 2.31 and an estimated lens redshift of @xmath44 .
it is assumed throughout that the universe is flat , @xmath45 .
.[table1 ] the minima for the curves in figure [ fig1 ] for the various cosmologies under consideration . [ cols="^,^,^",options="header " , ] figure [ fig5 ] presents the results of undertaking this procedure for various samples of gravitational lens systems .
the left panel presents the probability distribution for the quintessence equation of state @xmath13 while the right hand panel is the corresponding distribution in hubble s constants ; note that all the distributions have been normalized to a peak value of one .
the top row is for a sample of ten gravitational lens systems , akin to the situation today , followed by 50 , 100 and 500 systems .
ten thousand realizations were undertaken for each sample size .
each line on the plot corresponds to a different level of uncertainty in the gravitational lens time delay as outlined in the previous paragraph .
clearly today , where there are but a handful of gravitational lens systems for which we have determined the time delay , with overall uncertainties exceeding 5 - 10% , then the resultant hubble s constant and @xmath13 that can be derived from the sample are unlikely to accurately represent the underlying values .
table [ table3 ] presents the 95% confidence interval for the estimation of @xmath13 and @xmath14 for the various samples . increasing the sample size
greatly improves the situation , although even with 500 lenses and 15% noise , the values of @xmath13 and @xmath14 are not strongly constrained .
one is led to conclude , therefore , that a large sample of lenses with very accurately determined time delays and lens models is required to significantly determine the the underlying cosmological parameters . given the observational effort in such a task , other approaches to probing the cosmological equation of state are likely to prove more fruitful .
given this , the analysis was not extended to consider the smaller influences of quintessence evolution ( see section [ evolve ] ) .
can the current observations of time delays in gravitational lens systems tell us anything about the equation of state of the quintessence component ? in recent years , dedicated monitoring of a number of lensed systems has provided accurate @xmath46 time delay determinations ( e.g. fassnacht et al .
1999 ; koopmans et al . 2000 ) .
an examination of the hubble s constant derived from such studies reveals that , typically , it is less than the value determined from local studies , even accounting for standard cosmological differences ( e.g. impey et al . 1998 ; koopmans et al .
2001 ; winn et al .
this very question was also recently addressed by kochanek ( 2002 ) who suggests that this discrepancy is potentially due to galaxies possessing concentrated dark matter halos with a constant mass - to - light ratios , at odds with expectations from cold dark matter structure models . here , an alternative solution is considered . examining the panel in figure 1 corresponding to an @xmath47 cosmology
, it is apparent that choosing @xmath48 ( equivalent to setting @xmath6 ) results in almost the lowest possible determination of @xmath14 .
considering a cosmology with a classical cosmological constant ( @xmath8 ) increases the determined value of @xmath14 by @xmath49 ( dependent upon the source and lens redshifts ) , but as noted above , the currently determined values still tend to lie below the 72@xmath50 derived locally ( freedman et al .
2001 ) . assuming that the gravitational lens models are correct , one way to reach concordance between the two approaches is that @xmath15 ; such a conclusion is consistent with the recently derived limit of @xmath51 from an analysis of a combination of cosmic microwave background , high redshift supernovae , cluster abundances and large scale structure data ( wang et al .
2000 ; bean & melchiorri 2001 ) .
while tantalizing , however , it must be conceded that the current differences in the approaches are not statistically significant , especially given the relatively large uncertainty in the modeling of gravitational lens mass distributions ( see , for example , the range of @xmath14 values obtained from the modeling of pg 1115 + 080 ; kochanek 2002 ) .
given a large sample of gravitational lens systems , however , the determination of a systematic difference in the derived hubble s constant could be made .
this paper has investigated the role of a quintessence component on angular diameter distances , specifically their influence on the determination of hubble s constant from the measurement of a time delay in multiply imaged quasars . for flat universes , with an unevolving quintessence component , its seen that , for a gravitational lens system in which the time delay has been measured , the resultant hubble s constant is dependent upon the value of the equation of state parameter @xmath13 .
interestingly , the dependence of the determined value of the hubble s constant as a function of @xmath13 possesses a minimum which is independent on the lens and source redshift .
several models of evolving quintessence were also examined , consisting of a linear evolution of the equation of state with redshift .
the cosmologies resulted in significantly different forms of the angular diameter distance .
hence , our view of the cosmos would be different in the various cosmologies .
when considering the specific combination of angular diameter distances that constitute the cosmological contribution to the gravitational lensing determination of hubble s constant , it is seen that the resulting variations between cosmologies is very small , a matter of only a few percent , relative to an unevolving case with the same present day constitution .
a number of monte carlo simulations of the determination of hubble s constant and the quintessence equation of state , @xmath13 , were undertaken to explore the efficacy of this approach .
these revealed that the present situation with only a handful of lensed systems does not allow an accurate determination of the cosmic equation of state , and that at least an order of magnitude more lenses are truly required to provide a reasonably robust determination of the underlying cosmology .
the next generation of all - sky surveys are presently underway ( e.g. sloan digital sky survey ) or are being planned ( e.g. vista , prime ) , and these datasets will greatly increase the number of lensed quasars available for monitoring studies .
cooray & huterer ( 1999 ) estimate that @xmath52 lensed quasars will be identified from the sdss database alone , and a much larger number can be expected from the deeper vista and prime surveys .
the number of these sources amenable for follow - up monitoring campaigns will naturally be much smaller , but one can confidently expect a sample of several hundred systems to eventually become available .
however , given the effort required to first find such systems , as well as monitor them to determine the time delays and the modeling procedure , it is likely that @xmath13 will be first determined using one of the other various techniques currently being proposed .
we conclude , therefore , that gravitational lens time delays are likely to prove poor probes of the universal equation of state .
the anonymous referee is thanked for comments that improved the paper .
gfl acknowledges using david w. hogg s wonderful `` distance measures in cosmology '' cheat sheet ( astro - ph/9905116 ) , and thanks the gorillaz for their self - titled album .
terry bridges is thanked for providing the computational cycles on * odin*. | searches for supernovae at distances successful evidence topologically flat majority of energy universe exotic component with negative pressure ( riess 1999 recent identification of supernova at @xmath0 riess 2001 ) weight to claims component may differ from classical constant @xmath1 termed quintessence dark energy equation of state of form @xmath2 @xmath3 pressure @xmath4 density @xmath5 opposes gravity drives cosmological expansion linder examined physical nature of quintessence components @xmath6 to non - relativistic matter dust @xmath7 radiation @xmath8 classical cosmological constant exotic components are massless scalar fields @xmath9 cosmic string networks @xmath10 two - dimensional topological defects @xmath11 programs other approaches gravitational lensing statistics geometrical probes of @xmath12 forest galaxy distributions classical angular - size redshift tests provide probes of universal equation of state value of quintessence component @xmath13 influences view of universe modifying distances mapping cosmospaper concerns influence @xmath13 on angular diameter distances determination hubble s constant time delays in gravitational lens systems unlike local determinations hubble s constant cosmological nature gravitational lenses sensitive to underlying cosmological parameters section background covers basic formulae for generalized angular diameter distances in quintessence cosmologies timedelay influence @xmath13 on determination @xmath14 from lensed systems section evolve extends analysis to models evolving quintessence component montecarlo simulations estimate efficacy approach cosmological equation state speculate observations gravitational lens systems suggest @xmath15 conclusions presented in section [ conclusions ] resurgence in quintessence cosmology generalized cosmological equations presented ago by linder including generalized dyer - roeder equation for evolution of bundle rays from distant sourceangular diameter distance @xmath16 generalized beam equation by @xmath17\dot{r + \sum_w \frac{3 r ( 1+w)\alpha_w(z)\omega_w(z)}{2(1+z)^2 } = 0{linder @xmath18 density critical density @xmath19 contributor to total energy - density equation state @xmath20 @xmath3 pressure @xmath4 density given by @xmath21 ^ 2{omega} @xmath22 contribution to energy - density budget critical density by @xmath23 @xmath24 generalized form hubble parameter by @xmath25^\frac{1}{2 }{hubble} @xmath26 @xmath27 related to curvature universe @xmath28 deceleration parameter by @xmath29 equation [ linder ] contains parameter @xmath30 fluid in beam influences evolution ray bundle universe matter solution with @xmath31 dyer - roeder ` filled beam ' distance @xmath32 empty beam ' distance boundary conditions @xmath33 \label{boundary}\end{aligned} equation integrated using runge - kutta scheme package www.netlibcompared analytic results minimum angular redshifts quintessence cosmologies lima alcaniz 2000 agreement found filled - beam distances @xmath34 employed refsdal 1964 cosmological parameters time delay between paths light gravitational lens system discovery multiply - imaged quasars goal monitoring campaigns cohen 2000 oscoz 2001 patnaik narashima 2001 analyses frustrated by degeneracies mass models cosmological model enters determination hubble s constant @xmath35 @xmath36 normalized angular diameter distances between observer lens source @xmath39 @xmath40 measured time delay image pair giovi amendola 2001 examined influence quintessence component gravitational lens time delays determination hubble s constant clumping material dependence @xmath14 distances empty - full - beam different approach solving equation hubbletime evaluated for quintessence components redshifts lens source in seven gravitational lens systems favourable time delay measures considered table 2 giovi amendola 2001 q0957 + 561 + 656 two fiducial redshift pairs @xmath41 @xmath42no lensed systems with time delays at redshifts several potential systems quadruple lens h1413 + 117 at redshift 2.55 lens redshift at @xmath43 he1104 - 1805 redshift 2.31 estimated lens redshift @xmath44 assumed universe is flat @xmath45 minima for curves in figure for cosmologies figure results procedure for samples gravitational lens systems left panel probability distribution for quintessence equation of state @xmath13 right distribution in hubble s constants distributions normalized to peak value one top row sample ten gravitational lens systems followed by 50 100 500 systems ten thousand realizations for each sample size each line different level uncertainty in gravitational lens time delay handful of gravitational lens systems time delay uncertainties exceeding 5 - 10% resultant hubble s constant and @xmath13 unlikely to represent underlying values table 95% confidence interval for estimation of @xmath13 @xmath14 for samples increasing sample size improves situation 500 lenses 15% noise values of @xmath13 @xmath14 not constrainedconclude large sample of lenses with accurately determined time delays models required to determine cosmological parameters observational effort other approaches cosmological equation state fruitful analysis not extended to consider smaller influences of quintessence evolution evolve current observations of time delays in gravitational lens systems equation of state quintessence component ? dedicated monitoring lensed systems provided accurate @xmath46 time delay determinations fassnacht hubble s constant from studies reveals less than local studies differences by kochanek 2002 suggests discrepancy due to galaxies concentrated dark matter halos with constant mass - to - light ratios odds with expectations cold dark matter structure models alternative solution considered panel in figure 1 @xmath47 cosmology choosing @xmath48 equivalent @xmath6 ) results in lowest determination of @xmath14 cosmology with classical cosmological constant ( @xmath8 ) increases determined value of @xmath14 by @xmath49 dependent source lens redshifts determined values below 72@xmath50 derived locally ( freedman et al 2001 )assuming gravitational lens models correct concordance @xmath15 consistent with limit @xmath51 from analysis cosmic microwave background high redshift supernovae cluster abundances large scale structure data ( wang et al . 2000 ; bean & melchiorri 2001 ) current differences not statistically significant large uncertainty in gravitational lens mass distributions range @xmath14 values pg 1115 + 080 ; kochanek 2002 ) large sample gravitational lens systems determination systematic difference in hubble s constant could be made paper investigated role quintessence component on angular diameter distances influence hubble s constant time delay in multiply imaged quasars flat universes unevolving quintessence component gravitational lens system time delay measured hubble s constant dependent upon equation of state parameter @xmath13 dependence @xmath13 minimum independent on lens source redshift models of evolving quintessence examined linear evolution equation state with redshift cosmologies different forms angular diameter distance view of cosmos different in cosmologiesconsidering angular diameter distances cosmological contribution to gravitational lensing determination of hubble s constant variations between cosmologies small few percent relative to unevolving case with constitution monte carlo simulations of hubble s constant quintessence equation of state @xmath13 undertaken efficacy approach revealed present situation with handful of lensed systems allow accurate determination cosmic equation of state more lenses required robust determination cosmology next generation of all - sky surveys underway or planned increase lensed quasars for cooray & huterer 1999 estimate @xmath52 lensed quasars identified from sdss database larger number expected from deeper vista and prime surveys sources for follow - up monitoring campaigns smaller expect sample of several hundred systems available effort to find systems monitor time delays modeling likely @xmath13 determined using other techniques gravitational lens time delays likely prove poor probes of universal equation of state anonymous referee thanked for comments paper gfl acknowledges using david w. hogg s `` distance measures in cosmology '' cheat sheet astro - thanks gorillaz for self - titled albumterry bridges thanked cycles odin | 0.56454 | 3,271 | 1,662 | , Saving $0.1 in GPT-4. |
155,049 | optical excitations in hexagonal nanonetwork materials , for example , boron - nitride ( bn ) sheets and nanotubes , are investigated theoretically . a permanent electric dipole moment , whose direction is from the b site to the n site , is considered along the bn bond . when the exciton hopping integral is restricted to the nearest neighbors
, the flat band of the exciton appears at the lowest energy .
the symmetry of this exciton band is optically forbidden , indicating that the excitons relaxed to this band will show quite long lifetime which will cause luminescence properties .
# 1 electron density , excitation spectra calculations ; many - body and quasiparticle theories ; insulating films ; photoluminescence ; graphite and related compounds | the hexagonal nanonetwork materials composed of atoms with ionic characters , for example , boron - nitride ( bn ) sheets and nanotubes [ 1 ] , have been investigated intensively .
they are intrinsically insulators with the energy gap @xmath0 of about 4 ev as the preceding band calculations have indicated [ 2,3 ] .
the possible photogalvanic effects depending on the chiralities of bn nanotubes have been proposed by the model calculation [ 4 ] .
1 . optical excitations along the bn alternations . in this paper ,
we investigate optical excitation properties in bn systems .
there is a permanent electric dipole moment along the bn bond , which will give rise to strong excitonic properties as illustrated in fig .
low energy optical excitations are the excitations of the electron - hole pairs between the higher occupied states of n and the lower unoccupied states of b atoms .
we consider exciton interactions among nearest neighbor dipoles . in fig .
2 ( a ) , the b and n atoms are represented by full and open circles , respectively .
we assume one orbital hubbard model with the hopping integral of electrons @xmath1 , the onsite repulsion @xmath2 , and the energy difference @xmath0 between the b and n sites .
after second order perturbations , we obtain the following forms of the nearest neighbor interactions : @xmath3 for the case of conserved excited spin ( type-1 interaction ) and @xmath4 for the case that spin of the excited electron flips ( type-2 interaction ) .
the condition @xmath5 means that @xmath6 and @xmath7 are positive .
the interactions are present along the thin lines of fig .
2 ( a ) . after the extraction of the interactions @xmath6 and @xmath7
, there remains the two - dimensional kagom lattice which is shown in fig .
2 ( b ) . therefore
, the optical excitation hamiltonian becomes : h & = & _ i , j _
= , j_1 ( |i , j,| + h.c . )
+ & + & _ i , j j_2 ( |i , j,| + |i , j,| + h.c . ) , where the indices @xmath8 and @xmath9 mean the vertex points of the kagom lattice , and the sum is taken over the nearest neighbor pairs @xmath10 and the excited spin @xmath11 .
the unit cell has three lattice points , namely , 1 , 2 , and 3 , as shown in fig .
2 ( b ) . fig .
the hexagonal nanonetwork of boron ( full circles ) and nitrogen sites ( open circles ) .
several arrows indicate the directions of dipole moments .
( b ) the kagom lattice extracted from fig .
the shaded area is the unit cell .
the energy dispersions of the model are given in terms of wavenumbers @xmath12 : e = \ { l - 2 ( j_1 + j_2 ) , + ( j_1 + j_2 ) [ 1 + ] , + 2 ( - j_1 + j_2 ) , + ( j_1 - j_2 ) [ 1 + ] , . where the two dimensional x - y axes are defined as usual in fig . 2 , and @xmath13 is the unit cell length of the kagom lattice in fig
2 ( b ) , and @xmath14 is the bond length of fig .
2 ( a ) . there appears a dispersionless band ( triplet state ) with the lowest energy @xmath15 .
there is another dispersionless band ( singlet state ) at the higher energy @xmath16 .
such the appearance of the flat band has been discussed with the possibility of ferromagnetism in the literatures [ 5 ] . in the present case ,
the lowest optical excitation band becomes flat in the honeycomb bn plane .
when the bn plane is rolled up into nanotubes , the flat band is dispersionless too .
the flat exciton band will have strong optical density originating from the huge density of states due to the weak dispersive character .
symmetries of two wavefunctions at @xmath17 .
the full and open circles indicate positive and negative values at the lattice point , respectively .
we look at symmetries of the lowest excitons with the energy @xmath18 and the wavenumber @xmath19 .
the symmetries of the twofold degenerate solutions are shown in fig .
both wavefunctions have the symmetry _ gerade_.
the transition to the lowest exciton is optically forbidden , which indicates that excitons relaxed to this lowest exciton band will show quite long lifetime which will cause luminescence properties .
in addition , the lowest energy excitons will have huge density of state due to the flatness of the band .
these properties might result in interesting optical measurements in hexagonal nanonetwork materials .
the flat band of the optically forbidden exciton appears at the lowest energy in the optical excitations of bn systems .
the excitons relaxed to this band might show quite long lifetime which will cause luminescence properties .
d. golberg , y. bando , k. kurashima , and t. sato , solid state commun .
116 ( 2000 ) 1 .
a. rubio , j. l. corkill , and m. l. cohen , phys .
b 49 ( 1994 ) 5081 .
x. blase , a. rubio , s. g. louie , and m. l. cohen , europhys .
( 1994 ) 335 .
p. krl , e. j. mele , and d. tomnek , phys .
85 ( 2000 ) 1512 .
a. mielke , j. phys .
a 24 ( 1991 ) 3311 ; _ ibid .
_ 25 ( 1992 ) 4335 . | hexagonal nanonetwork materials ionic characters boron - nitride bn ) sheets nanotubes investigated intensively insulators energy gap @xmath0 4 ev possible photogalvanic effects chiralities bn nanotubes proposed by model calculation optical excitations along bn alternations properties in bn systems permanent electric dipole moment along bn bond strong excitonic properties fig low energy optical excitations electron - hole pairs between higher states n lower states b atoms interactions among nearest neighbor dipoles fig . 2 b n atoms represented by full open circles orbital hubbard model hopping integral of electrons @xmath1 onsite repulsion @xmath2 energy difference @xmath0 between b n sites perturbations forms nearest neighbor interactions @xmath3 conserved excited spin @xmath4 condition @xmath5 @xmath6 @xmath7 positive interactions along thin lines of fig . 2 ( a ) after extraction of interactions @xmath6 remains two - dimensional kagom lattice shown in fig . 2 ( b )optical excitation hamiltonian becomes h & = i j j_1 ( |i , j,| + h.c ) + _ i j j_2 ( |i j,| + + h.c indices @xmath8 @xmath9 vertex points kagom lattice sum over pairs @xmath10 spin @xmath11 unit cell has three lattice points 1 2 3 . 2 b hexagonal nanonetwork boron full nitrogen sites open circles arrows indicate directions dipole moments kagom lattice shaded area unit cell energy dispersions wavenumbers @xmath12 : e = l - 2 ( j_1 + j_2 ) , + ( j_1 + j_2 ) [ 1 + ( - j_1 + j_2 - two dimensional x - y axes defined fig . 2 @xmath13 unit cell length kagom lattice @xmath14 bond length fig . 2 dispersionless band lowest energy @xmath15 another band higher energy @xmath16 flat band discussed ferromagnetism lowest optical excitation band becomes flat in honeycomb bn planebn plane rolled into nanotubes flat band dispersionless flat exciton band strong optical density huge density states weak dispersive character symmetries wavefunctions at @xmath17 full open circles indicate positive negative values lattice point symmetries lowest excitons energy @xmath18 wavenumber @xmath19 twofold degenerate solutions fig wavefunctions symmetry gerade transition to lowest exciton optically forbidden excitons relaxed long lifetime luminescence lowest energy excitons huge density state flatness band optical measurements hexagonal nanonetwork materials flat band optically forbidden exciton lowest energy optical excitations bn systems excitons relaxed long lifetime luminescence d. golberg bando kurashima sato solid state commun . 116 ( 2000 ) 1 rubio corkill l cohen . 49 ( 1994 ) 5081 blase rubio europhys . 1994 ) 335 krl mele tomnek phys . 85 ( 2000 ) 1512 mielke phys . 24 ( 1991 ) 3311 25 ( 1992 ) 4335 | 0.187573 | 1,314 | 674 | , Saving $0.0 in GPT-4. |
155,051 | we have measured the energy spectrum of ultra - high energy cosmic rays ( uhecr ) with the hires fadc detector ( hires-2 ) in monocular mode .
a detailed monte carlo simulation of the detector response to air showers has been used to calculate the energy dependent acceptance of the air fluorescence detector .
the measured spectrum complements the measurement by the hires-1 detector down to lower energies .
systematic effects of the assumed input spectrum and composition on the aperture are presented , as well as systematics due to the atmosphere . | the two air fluorescence detectors of the hires experiment provide stereoscopic observation of extensive air showers ( eas ) , initiated in the atmosphere by ultra - high energy cosmic rays .
the stereo view leads to an improved resolution in the reconstructed arrival direction and energy of the primary cosmic ray particle due to better constraints on the location of the shower axis .
however , even if observation with both eyes is the preferred method , there are good reasons for the analysis of monocular data .
since the hires-1 detector began operation two years before hires-2 , the hires-1 dataset provides larger statistics than the stereo data .
this is especially significant in the measurement of the cosmic ray energy spectrum at energies around the gzk flux suppression .
the hires-2 monocular dataset , which will be presented here , offers a different advantage : it contains well reconstructed events at energies lower than hires-1 mono and stereo events .
the hires-2 detector uses fadc electronics to record signals at a frequency of 10 mhz , which leads to a better time resolution than is achieved with the sample - and - hold electronics of the hires-1 detector .
furthermore , the two rings of mirrors of hires-2 provide twice the coverage in elevation angle that is achieved with the single ring of mirrors of the hires-1 detector .
events recorded with hires-2 can be reliably reconstructed down to energies of about @xmath0 .
this is also about a decade lower in energy than the lower limit for stereo events , which is constrained by the separation between the two detectors : events with the lowest observable energy lie half - way between the two detectors and have thus a distance of about @xmath1 from each detector , which sets a lower limit to their observable energies . the hires-2 measurement of the uhecr spectrum is intended to complement the hires-1 measurement at lower energies , where the features of the second knee and the ankle provide recognizable signatures of the cosmic ray spectrum , which can be used to compare the hires measurements to other experiments .
the calculation of the cosmic ray energy spectrum from the measured event distribution is a problem of unfolding the true spectrum of cosmic rays at their arrival in the earth s atmosphere from the distorsions of the detector response , i.e. the acceptance of the detector and its limited resolution .
the purpose of the monte carlo ( mc ) simulation in the hires experiment is to give an accurate description of the geometry and energy dependent acceptance of the detector and of its resolution .
the simulation consists basically of two parts : an air shower generator and a detector response mc . in the first part of the simulation ,
large sets of air showers are generated with several discrete energies and with different primary particles , using corsika @xcite and qgsjet @xcite .
their profiles are saved in a library of air showers . from this library
, the individual profiles can be read into the detector response mc and used to simulate eas at different geometries .
energies in the detector response mc are chosen from a given continuous input spectrum .
profiles from library showers generated at a nearby , discrete energy are scaled to the chosen energy .
the detector response simulation includes : generation of fluorescence and erenkov light at the shower , propagation of light through the atmosphere , ray tracing of photons through the optical path of the detector , pmt response to the signal , simulation of noise , electronics and trigger simulation .
events accepted by the trigger are written out in the same format as the data .
two databases have been generated and are read by the detector response mc in order to allow event simulation under the exact data taking conditions .
a trigger database contains information on the livetime of the detector , dead mirrors and variable trigger settings .
an atmospheric database provides hourly measurements of the aerosol content of the atmosphere .
an average atmosphere has been used in this analysis to allow a direct comparison with the hires-1 measurement .
as the air shower develops in the atmosphere , its image sweeps across the photomultiplier tube clusters , causing a track of triggered phototubes .
the first step in the reconstruction of an event is the determination of the shower - detector plane from a fit to this track .
the geometry of the shower axis within this plane and the distance of the shower from the detector are found in the hires-2 monocular reconstruction with a fit of the times of triggered phototubes versus their angle along the shower track .
once the location and geometry of the shower axis have been determined , the charged particle profile of the air shower is reconstructed from the distribution of recorded fadc counts in triggered phototubes .
absolute calibration of the photomultiplier tubes with a calibrated light source and measurements of the aerosol density of the atmosphere with steerable lasers ensure the accurate calculation of the shower profile from the fadc signals . using the atmosphere as a calorimeter
, we can calculate the total deposited energy of the shower from the charged particle profile after subtraction of erenkov light .
adding about @xmath2 for missing energy , which is deposited into the ground in muons and neutrinos or lost in nuclear excitations , yields the total energy of the primary cosmic ray particle .
the same quality cuts and the same reconstruction procedure are applied to both data and simulated events , which allows us to directly compare all the details of our mc simulation against the actual experiment .
an extensive set of data - mc comparison plots including e.g. the distance of air showers , their angular distributions , number of triggered mirrors and phototubes , reconstructed profiles and energies , etc.tests how well our mc simulates the actual detector .
an example for the overall very good agreement between data and mc is shown in figure [ datamc ] : the distributions of photoelectrons per degree of track are measures of the amount of light that is seen in the data and generated in the mc .
the simulation closely follows the data distribution .
more details on the hires-2 analysis can be found in @xcite . with our mc simulation proven to be realistic
, we can calculate the acceptance of the hires-2 detector , which is given by the ratio of fully reconstructed events ( @xmath3 ) over all generated events ( @xmath4 ) in each energy bin .
multiplying the acceptance by the geometrical aperture @xmath5 and the livetime @xmath6 of the detector ( about 531 hours of analyzed data ) yields the exposure of the detector .
we fit the simulated exposure to an appropriately chosen function in order to smooth out statistical fluctuations .
the result is shown in figure [ exposure ] .
the differential flux @xmath7 in each energy bin can now be calculated as : @xmath8 where @xmath9 is the number of data events in the energy bin and @xmath10 is the binwidth . ) measured by the hires-2 detector in monocular mode . ]
the energy spectrum measured by the hires-2 detector on clear , moonless nights from december 1999 to september 2001 ( @xmath11 events ) is shown in figure [ hr2_spec ] .
we observe the ankle at about @xmath12 .
our updated measurement is consistent with the gzk flux suppression , but statistics at the high energy end are low .
figure [ hr12fe_spec ] shows the spectra measured by hires-1 , hires-2 and the fly s eye stereo experiment superimposed .
the two monocular hires spectra agree with each other and with our predecessor experiment .
there is a slight difference in the slopes of the hires and fly s eye spectra above the ankle , which is most likely due to a different atmospheric model used in the fly s eye experiment
. the low energy end of the hires-2 spectrum is consistent with the fly s eye spectrum .
however , due to low statistics , hires can not claim detection of the second knee
, seen by fly s eye at @xmath13 . ) measured by hires-1 ( squares ) , hires-2 ( circles ) and fly s eye stereo ( triangles ) . ]
the main systematic uncertainties in the hires monocular spectra have been reported in @xcite : uncertainties in the absolute phototube calibration , fluorescence yield , missing energy correction and atmospheric calibration add up to a total uncertainty in the flux of @xmath14 .
here , we want to discuss a potential impact on the calculated acceptance from our assumptions about the mc input energy spectrum and input composition , and the assumption of an average atmosphere in our simulation and analysis . using mc simulations to determine the acceptance of the detector could potentially introduce a bias into the analysis due to a wrong model used in the mc .
this potential bias can be estimated by calculating the acceptance for different model assumptions in the mc @xcite .
we have varied the shape of the input spectrum , from which mc event energies are chosen .
figure [ e3inp ] shows a plot comparing the energy distributions of a mc set using a simple @xmath15 power law as input with about half of our analyzed data .
the disagreement of the data with this assumption can be seen clearly from the tilted ratio plot . input spectrum . *
bottom : * ratio of data over mc . ]
our regular mc uses an input spectrum whose shape fits the fly s eye stereo spectrum below the ankle and follows a straight line fit to the hires-1 data for higher energies .
the gzk feature has not been included .
a comparison plot for this input spectrum is shown in figure [ ourinp ] . here
, the agreement is very good .
we have calculated acceptances for both mc sets .
the bias we are avoiding by using a realistic input spectrum is of the order of @xmath16 .
this bias would translate directly into the measured spectrum , if one assumed a wrong ( i.e. @xmath15 ) input spectrum in the mc simulation .
we determine the fraction of showers initiated by light and heavy , i.e. proton and iron , cosmic rays in the mc from composition measurements by the hires / mia @xcite and hires stereo @xcite experiments .
the uncertainties in the hires / mia measurement that translate into our spectrum calculation add up to @xmath17 .
their sources are the detector calibration , the aerosol component of the atmosphere and the statistical uncertainty of a fit to the hires / mia data .
a @xmath18 uncertainty in the fluorescence yield is common to both hires and hires / mia and was therefore not included .
the difference in the predictions of pure iron and pure proton maximum shower depths ( @xmath19 ) between different hadronic interaction models was not taken into account either , since we are not concerned about the fraction of real proton and iron showers here , but only about the fraction of showers with a certain @xmath19 . by generating two mc sets with pure proton and pure iron showers
, we can calculate the effect a @xmath20 change in the proton fraction would have on the acceptance .
the uncertainty in the final spectrum from such a variation is shown in figure [ compsys ] .
at the low energy end of the spectrum , the acceptance for iron cosmic rays is lower because iron showers develop higher up in the atmosphere and are more likely to be outside of our elevation coverage than proton showers .
this leads to larger uncertainties at lower energies . for energies above @xmath21 , where we use the hires stereo composition
measurement , no difference was seen in the acceptances for iron and proton showers .
change in the proton fraction of the mc . ]
our current analysis uses a measurement of the average aerosol content of the atmosphere .
we have studied the effect on the energy resolution of using an atmospheric database with hourly entries instead of the average .
a mc set has been generated with use of the database to simulate data from september 2000 to march 2001 .
our database covers @xmath22 of all nights in this period .
the mc set has been reconstructed first with the average and then with the database
. a gaussian fit to the energy resolution has a @xmath23 of @xmath24 for the reconstruction using the database and @xmath25 when using the average .
an acceptance has been calculated in both cases .
the ratio of acceptances can be seen in figure [ apratio_adb ] : the acceptance for the mc set reconstructed with database is in the numerator , the acceptance for the same mc set , but reconstructed with the average , in the denominator .
using the database in the reconstruction does not have a significant impact on the calculated aperture .
data from september 2000 to march 2001 have been analyzed with the average and with the atmospheric database .
we have not found any systematic shift in the reconstructed energies . when we examined the highest energy events ( @xmath26 ) , we found that using the database shifts one event from the bin centered at @xmath27 to the bin centered at @xmath28 .
this leads to a reduction of the events above @xmath29 from 7 to 6 in this dataset .
* our measurement of the uhecr spectrum with the hires fadc detector in monocular mode is consistent with the gzk flux suppression at @xmath30 and with the fly s eye stereo spectrum .
the ankle is observed at an energy of about @xmath12 . * detailed monte carlo simulation
is used to calculate the acceptance of the detector , after being tested with data - mc comparisons . *
we avoid a bias of about @xmath16 in our spectrum by using a realistic assumption about the input energy spectrum in the mc . * systematic uncertainties caused by uncertainties of the input composition in the mc
do not exceed our statistical errors .
* we have tested the impact of using an atmospheric database with hourly entries rather than an average atmosphere in the reconstruction for about half of our analyzed data .
we have not found a significant difference . | two air fluorescence detectors hires experiment provide stereoscopic observation of air showers initiated by ultra - high energy cosmic rays stereo view leads to improved resolution in reconstructed arrival direction energy primary cosmic ray particle due shower axis observation both eyes good reasons for analysis monocular data hires-1 detector dataset provides larger statistics than stereo data significant in measurement cosmic ray energy spectrum at energies around gzk flux suppression hires-2 monocular dataset offers advantage contains well reconstructed events at energies lower than stereo detector uses fadc electronics record signals at 10 mhz better time resolution two rings of mirrors hires-2 provide twice coverage in elevation angle single events recorded reconstructed down to energies about @xmath0 about a decade lower in energy than lower limit for stereo events constrained by separation between detectors events with lowest observable energy lie half - way between distance about @xmath1 from each lower limit to observable energies hires-2 measurement complement hires-1 measurement at lower energies provide recognizable signatures of cosmic ray spectrum compare measurements tocalculation cosmic ray energy spectrum from event distribution problem unfolding true spectrum at arrival earth s atmosphere from distorsions detector response limited resolution purpose monte carlo ( mc ) simulation in experiment accurate description geometry energy dependent acceptance detector resolution simulation two parts air shower generator detector response mc large air showers generated with discrete energies different primary particles using corsika @xcite qgsjet @xcite profiles saved in library air showers profiles read into detector response mc simulate geometries energies detector chosen from continuous input spectrum profiles from library showers scaled to chosen energy detector response simulation includes generation fluorescence erenkov light propagation light ray tracing photons pmt response simulation noise electronics trigger simulation events accepted by trigger written in same format as data two databases generated read by detector response mc event simulation under exact data taking conditions trigger database contains information livetime detector dead mirrors variable trigger settings atmospheric database provides hourly measurements aerosol content average atmosphere used direct comparison with hires-1 measurementair shower develops image sweeps across photomultiplier tube clusters causing track of triggered phototubes first step reconstruction shower - detector plane track geometry shower axis distance from detector found in hires-2 monocular reconstruction triggered phototubes versus angle shower track location geometry determined charged particle profile reconstructed from distribution recorded fadc counts in phototubes calibration photomultiplier tubes measurements aerosol density with lasers ensure accurate calculation shower profile from fadc signals atmosphere as calorimeter calculate total deposited energy shower from charged particle profile after subtraction erenkov light adding @xmath2 for missing energy deposited yields total energy of primary cosmic ray particle same quality cuts reconstruction procedure applied to data simulated events compare details mc simulation against experiment data - mc comparison plots including distance of air showers angular distributions number of triggered mirrors phototubes reconstructed profiles energies mc simulates detector example agreement between data mc shown in figure distributions of photoelectrons per degree of track measures of light seen in data generated in mc simulation follows data distributiondetails hires-2 analysis in @xcite mc simulation realistic acceptance hires-2 detector ratio reconstructed events over generated events in each energy bin multiplying acceptance by geometrical aperture livetime detector 531 hours yields exposure simulated exposure to function statistical fluctuations result in figure exposure differential flux @xmath7 in each energy bin calculated as @xmath8 number data events binwidth measured by hires-2 detector monocular mode energy spectrum measured on clear moonless nights december 1999 to september 2001 events in figure hr2_spec observe ankle at @xmath12 measurement consistent with gzk flux suppression statistics high energy end low figure hr12fe_spec shows spectra measured by hires-1 hires-2 fly s eye stereo experiment spectra agree predecessor experiment slight difference in slopes fly s eye spectra above ankle due to different atmospheric model fly s eye experiment low energy end hires-2 spectrum consistent with fly s eye spectrum low statistics hires claim detection second knee fly s eye at @xmath13measured by hires-1 squares hires-2 circles fly s eye stereo triangles uncertainties in hires monocular spectra reported in @xcite uncertainties absolute phototube calibration fluorescence yield missing energy correction atmospheric calibration total uncertainty in flux @xmath14 potential impact on acceptance from assumptions mc input energy spectrum composition average atmosphere in simulation analysis mc simulations introduce bias due wrong model bias estimated by calculating acceptance for different model assumptions in mc @xcite varied shape of input spectrum event energies figure comparing energy distributions mc set using @xmath15 power law with half analyzed data disagreement with assumption seen from tilted ratio plot regular mc uses input spectrum shape fits fly s eye stereo spectrum follows straight line fit to hires-1 data for higher energies gzk feature not included comparison plot in figure agreement good calculated acceptances for both mc sets bias avoiding by input spectrum @xmath16 bias translate into measured spectrum if wrong @xmath15 input spectrum in mc simulation determine fraction of showers initiated by light and heavy proton iron cosmic rays in mc from measurements hires / mia stereo experiments uncertainties in hires / mia measurement spectrum calculation to @xmath17 sources are detector calibration aerosol component statistical uncertainty mia data @xmath18 uncertainty in fluorescence yield common to hires mia not included difference in predictions pure iron proton maximum shower depths @xmath19 between hadronic interaction models not not real proton iron showers certain @xmath19 generating two mc sets with pure proton iron showers calculate effect change in proton fraction on acceptance uncertainty in final spectrum from variation shown in figure low energy end spectrum acceptance for iron cosmic rays lower iron showers develop higher outside elevation coverage proton leads larger uncertainties at lower energies energies above @xmath21 no difference in acceptances for iron proton showers change in proton fraction current analysis uses average aerosol content studied effect on energy resolution atmospheric database with hourly entries mc set generated database data from september 2000 to march 2001 database covers @xmath22 all nightsmc set reconstructed with average then database gaussian fit to energy resolution has @xmath23 @xmath24 for reconstruction database @xmath25 average acceptance calculated both cases ratio of acceptances in figure_adb acceptance for mc set database in numerator average in denominator database significant impact on calculated aperture data from september 2000 to march 2001 analyzed with average atmospheric database systematic shift in reconstructed energies highest energy events @xmath26 database shifts one event from @xmath27 to @xmath28 to reduction events above @xmath29 from 7 to 6 measurement uhecr spectrum with fadc detector consistent with gzk flux suppression at @xmath30 fly s eye stereo spectrum ankle observed at energy @xmath12 detailed monte carlo simulation acceptance detector tested data - mc comparisons avoid bias of @xmath16 in spectrum realistic assumption about input energy spectrum mc systematic uncertainties input composition mc exceed statistical errors tested impact atmospheric database with hourly entries average atmosphere reconstruction for half of analyzed data significant difference | 0.439907 | 2,880 | 1,353 | , Saving $0.1 in GPT-4. |
155,055 | we present a study of the globular cluster ( gc ) systems of nearby elliptical and s0 galaxies at a variety of wavelengths from the x - ray to the infrared .
our analysis shows that roughly half of the low mass x - ray binaries ( lmxbs ) , that are the luminous tracers of accreting neutron star or black hole systems , are in clusters .
there is a surprisingly strong correlation between the lmxb frequency and the metallicity of the gcs , with metal - rich gcs hosting three times as many lmxbs as metal - poor ones , and no convincing evidence of a correlation with gc age so far . in some of the galaxies
the lmxb formation rate varies with gc color even within the red peak of the typical bimodal cluster color distribution , providing some of the strongest evidence to date that there are metallicity variations within the metal - rich gc peak as is expected in a hierarchical galaxy formation scenario .
we also note that any analysis of subtler variations in gc color distributions must carefully account for both statistical and systematic errors .
we caution that some published gc correlations , such as the apparent blue - tilt or mass - metallicity effect might not have a physical origin and may be caused by systematic observational biases . | high resolution chandra x - ray images of nearby ellipticals and s0s have resolved large numbers of point sources , confirming a long - standing suggestion that the hard x - ray emission in many of these galaxies is predominantly from x - ray binaries . in early type galaxies
most of the bright , l@xmath0@xmath110@xmath2 erg s@xmath3 sources seen in typical chandra observations must be low mass x - ray binaries , binary systems comprising a neutron star or black hole accreting via roche lobe overflow from a low mass companion , since they generally have stellar populations that are at least a few gyrs old . an important characteristic of lmxbs
is that they are disproportionately abundant in globular clusters .
even though gcs account for @xmath40.1% of the stellar mass in the galaxy , they harbor about 10% of the l@xmath0@xmath110@xmath5 erg s@xmath3 lmxbs , indicating a probability of hosting a lmxb that is at least two orders of magnitude larger than for field stars .
this has long been attributed to efficient formation of lmxbs in clusters due to dynamical interactions in the core .
early type galaxies are ideal for studies of the lmxb - gc link as they are particularly abundant in globular clusters .
the identification of lmxbs with these simple stellar systems that have well defined properties such as metallicity and age provides a unique opportunity to probe the effects of these parameters on lmxb formation and evolution .
fig 1 plots the colors , magnitudes , half - light radii , and galactocentric distances of globular clusters identified in hst - wfpc2 images of ngc 4472 ( ( * ? ? ?
* kundu , maccarone & zepf 2002 ) [ kmz02 ] ) , the brightest elliptical in the virgo cluster .
the well known gaussian globular cluster luminosity function and the bimodal globular cluster color ( metallicity ) distribution are apparent .
the large symbols indicate the gcs that host lmxbs .
lmxbs are preferentially found in the most luminous and red ( metal - rich ) gcs .
statistical tests ( kmz02 ) reveal a marginal tendency of lmxbs to favor gcs with smaller half light radius , and no convincing correlation with galactocentric distance .
these correlations have subsequently been confirmed in other galaxies ( ( * ? ? ?
* kim et al . 2006 ) ; ( * ? ? ?
* kundu , maccarone , & zepf 2007 ) [ kmz07 ] ; ) .
the presence of bright lmxbs in @xmath64% of gcs and the association of @xmath640% of lmxbs with gcs in ngc 4472 is also typical of the values in other early type galaxies .
the underlying reasons for these correlations provide a window into both the dynamics of gcs and the physics of lmxbs .
since luminous clusters are known to be denser than less luminous ones and obviously have more stars , the higher rate of lmxb formation in these clusters due to the consequently higher dynamical interaction rate is not surprising .
one the other hand there are no obvious dynamical reasons for the three times larger rate of lmxbs in the red , metal - rich , globular clusters as compared to the metal - poor ones .
lmxbs are similarly found preferentially in the metal - rich galactic gcs , but in the past this was often dismissed as the consequence of the location of these clusters in the bulge , where the dynamical evolution of gcs may be accelerated .
the lack of correlation of the lmxbs with galactocentric distance in the large gc samples of ellipticals ( kmz02 ; kmz07 ) argues against this possibility and establishes a primary correlation with gc color .
it has also been suggested that enhanced rate of lmxbs in metal - rich gcs may be because of the younger ages of these clusters
. however infrared imaging , which yields more accurate constraints on the metallicities of gcs and helps disentangle the age - metallicity degeneracy , has established that the optical colors of globular cluster systems that have clearly bimodal distributions are primarily a tracer of metallicity ( ( * ? ? ?
* kundu & zepf 2007 ) ) and lmxbs are indeed preferentially located in metal - rich gcs ( ( * ? ? ?
* kundu et al . 2003 ) ; ( * ? ? ?
* hempel et al . 2007 ) ) .
fig 2 shows the color magnitude distributions of five galaxies with confirmed bimodal color distributions and the factor of three enhancement with metallicity .
a very interesting feature of the ngc 1399 distribution is that not only is there a larger fraction of lmxbs in the metal - rich sub - population of clusters , but in fact lmxbs are preferentially located in the most metal - rich gcs .
this is the most convincing evidence to date that there is metallicity structure within the metal - rich peak of gcs , as is expected from hierarchical models of galaxy and globular cluster system formation .
it is not clear if the reason that this feature is obvious only in ngc 1399 is because of the location of this galaxy at the center of the fornax cluster which leads to a particularly efficient enrichment history , or because of the larger color baseline of this data set .
although the sivakoff et al . ( 2007 ) analysis is broadly in agreement about the metallicity trend and finds a linear increase in the fraction of lmxbs in gcs with metallicity , it suggests that the rate of lmxbs in the reddest , and consequently most metal - rich gcs actually drops .
we note that lmxbs are found preferentially in the brightest globular clusters , and if the gc sample in sivakoff et al .
( 2007 ) were to be restricted to similar luminosities this apparent discrepancy would disappear .
in other words the reddest gcs in the sivakoff sample are the faint ones which have scattered to these colors due to observational uncertainties and are not representative of the true underlying color / metallicity of the gcs .
there have been some theoretical attempts to explain the metallicity effect .
* maccarone , kundu , & zepf ( 2004 ) ) suggest that irradiation of the donor star by the lmxb is key .
higher metallicity objects can dissipate this energy through line cooling while a wind is generated in low metallicity star , thus lowering the lmxb lifetime .
* ivanova ( 2006 ) ) on the other hand suggests that the absence of an outer convective layer in solar mass metal - poor stars limits magnetic braking and the formation of mass transferring lmxbs .
various groups have attempted to derive the dependence of the observed lmxb rate on the globular cluster metallicity and mass by assuming a specific m / l ratio and color - metallicity correlation ( ( * ? ? ?
* jordan et al . 2004 ) ; ; sivakoff et al .
these studies generally agree that the mass dependence is roughly linear and the metallicity dependence is approximately z@xmath7 .
jordan et al . ( 2004 ) and sivakoff ( 2007 ) further attempt to link the probability of finding a lmxb in a gc to the rate of stellar interactions in the core of a gc .
however , this requires measuring the core radii of gcs which is a small fraction of even a hst acs pixel for these galaxies .
thus the core radius is derived by extrapolation of other measured gc properties .
smits et al .
( 2006 ) show that the jordan et al .
( 2004 ) results are as expected if there is no information about the core radius of the clusters in their sample . while the core radii of extragalactic gcs are a challenge for the present data sets , the half - light radii can indeed be measured and its effect on other measured gc parameters must be accounted for carefully .
some of the consequences of not doing so are outlined next .
many recent hst - based studies of globular cluster systems have reported the discovery of color - metallicity trends in the blue , metal - poor sub - population of globular clusters ( ( * ? ? ?
* strader et al . 2006 ) ; ( * ? ? ?
* mieske et al . 2006 ) ; ( * ? ? ?
* harris et al . 2006 ) ) .
this small trend of brighter blue clusters appearing a few hundredths of a magnitude redder per magnitude of brightness has been dubbed the blue - tilt , and is viewed as evidence of a mass - metallicity trend .
this has been interpreted as evidence of self - enrichment of the more massive clusters either because they started out with dark matter halos that have since been stripped , or because they formed in large gas clouds .
this effect is larger in more massive galaxies , and more apparent when the color magnitude diagram is plotted using the redder of the two filters under consideration for the magnitude axis ( mieske et al .
2006 ) .
we have reanalyzed some of these data sets and find that there is a small , but measurable , mass - radius relationship for gcs in these galaxies ( kundu & zepf 2007b in preparation ) .
since the blue gcs are on average larger , this trend is better defined for the metal - poor clusters .
studies of extragalactic cluster systems typically assume either a uniform aperture correction or attempt to fit the profile within a small aperture in order to minimize the uncertainties introduced by the galaxy background . we show in kundu & zepf ( 2007b ) that the blue tilt is likely the consequence of photometric bias introduced by the effect of the mass - radius relationship on such photometric techniques .
it is also important to note that the uncertainties in color and magnitude are not independent , and hence not orthogonal on a color magnitude diagram .
in fact when the redder of the two filters is used for the magnitude axis in a color - magnitude diagram the uncertainties are parallel to the direction of the blue - tilt and hence amplify the effect .
conversely the uncertainties tend to negate the blue - tilt when the blue filter is plotted on the magnitude axis of a color magnitude plot , thus explaining the filter effect .
moreover , the underlying galaxy light is known to be redder for more massive galaxy .
this explains the apparent variation of the strength of the blue - tilt with galaxy mass .
finally we note that various studies have suggested that there is a small color metallicity trend in the mean colors of the metal - poor globular clusters with host galaxy mass ( e.g. strader et al .
we find a trend of larger mean sizes of blue gcs with larger galaxy mass , which may cause this effect .
thus , within the observational uncertainties the properties of metal - poor halo clusters appear to be remarkably uniform everywhere .
harris , w. e. , et al .
2006 , _ apj _ , 636 , 90 hempel , m. , zepf , s. , kundu , a. , geisler , d. , & maccarone , t. j. 2007 , _ apj _ , 661 , 768 ivanova , n. 2006 , _ apj _ , 636 , 979 jordn , a. , et al . 2004 , _
apj _ , 613 , 279 kim , e. , et al . 2006 , _
apj _ , 647 , 276 kundu , a. , maccarone , t. j. , & zepf , s. e. 2002 , _ apj _ ( letters ) , 574 , l5 kundu , a. , maccarone , t. j. , & zepf , s. e. 2007 , _ apj _ , 662 , 525 kundu , a. , maccarone , t. j. , zepf , s. e. , & puzia , t. h. 2003 , _ apj _ ( letters ) , 589 , l81 kundu , a. , & zepf , s. e. 2007 , _ apj _ ( letters ) , 660 , l109 maccarone , t. j. , kundu , a. , & zepf , s. e. 2004 , _ apj _ , 606 , 430 mieske , s. , et al . 2006 , _
apj _ , 653 , 193 sivakoff , g. r. , et al . 2007 , _ apj _ , 660 , 1246 smits , m. , maccarone , t. j. , kundu , a. , & zepf , s. e. 2006 , _ a&a _ , 458 , 477 strader , j. , brodie , j. p. , spitler , l. , & beasley , m. a. 2006 , _ aj _ , 132 , 2333 | chandra x - ray images ellipticals resolved point sources confirming hard x - ray emission galaxies from x - ray binaries early type galaxies bright sources low mass x - ray binaries neutron star black hole roche lobe overflow from low mass companion stellar populations few gyrs old lmxbs disproportionately abundant in globular clusters gcs account for.1% stellar mass harbor 10%@xmath0 lmxbs probability of hosting lmxb two orders magnitude larger than field stars attributed to efficient formation of lmxbs clusters due dynamical interactions core early type galaxies ideal for studies lmxb - gc link abundant in globular clusters identification of lmxbs with stellar systems metallicity age opportunity probe effects on lmxb formation evolution fig 1 plots colors magnitudes half - light radii galactocentric distances of globular clusters in hst - wfpc2 images of ngc 4472 brightest elliptical in virgo cluster gaussian globular cluster luminosity function bimodal globular cluster color metallicity ) distribution apparentlarge symbols indicate gcs lmxbs lmxbs found in luminous red metal - rich gcs statistical tests reveal marginal tendency favor gcs smaller half light radius no correlation with galactocentric distance correlations confirmed in other galaxies kim et 2006 kundu maccarone zepf 2007 bright lmxbs in @xmath64% gcs association @xmath640% lmxbs with gcs in ngc 4472 typical other early type galaxies correlations provide window into dynamics gcs physics lmxbs luminous clusters denser have more stars higher rate lmxb formation higher dynamical interaction rate not surprising no obvious dynamical reasons for three times larger rate of lmxbs in red metal - rich globular clusters to metal - poor ones lmxbs found in metal - rich galactic gcs dismissed consequence location in bulge evolution lack of correlation with galactocentric distance in large gc samples argues against possibility establishes primary correlation with gc color enhanced rate lmxbs in metal - rich gcs may younger ages clustersinfrared imaging accurate constraints on metallicities gcs age - metallicity degeneracy established optical colors of globular cluster systems bimodal distributions tracer of metallicity 2007 lmxbs preferentially located in metal - rich gcs fig 2 shows color magnitude distributions of five galaxies with bimodal color distributions factor of three enhancement with metallicity ngc 1399 distribution larger fraction of lmxbs in metal - rich sub - population clusters preferentially located in most metal - rich gcs convincing evidence metallicity structure within metal - rich peak of gcs clear if feature obvious in ngc 1399 location galaxy at center of fornax cluster efficient enrichment history or larger color baseline data set sivakoff et al . 2007 analysis about metallicity trend finds linear increase in of lmxbs in gcs with metallicity suggests rate of lmxbs in reddest most metal - rich gcs drops lmxbs found preferentially in brightest globular clusters if gc sample in sivakoff 2007 restricted to similar luminosities discrepancy would disappearreddest gcs in sivakoff sample are faint ones scattered due to observational uncertainties not representative of true color metallicity theoretical attempts to explain metallicity effect maccarone kundu zepf 2004 suggest irradiation of donor star by lmxb key higher metallicity objects dissipate energy through line cooling wind generated in low metallicity star lmxb lifetime ivanova 2006 suggests absence of outer convective layer in solar mass metal - poor stars limits magnetic braking formation mass transferring lmxbs groups derive dependence of lmxb rate on globular cluster metallicity mass specific m / l ratio color - metallicity correlation jordan 2004 sivakoff mass dependence linear metallicity dependence approximately z@xmath7 sivakoff 2007 link probability of finding lmxb in gc to stellar interactions core requires measuring core radii core radius derived by extrapolation other measured gc properties smits et al . 2006 jordan et al . 2004 results expected if no information about core radius core radii of extragalactic gcs challenge half - light radii can be measured effect on gc parametersconsequences of not doing so outlined next recent hst studies of globular cluster systems reported color - metallicity trends in blue metal - poor sub - population ( strader 2006 mieske 2006 harris 2006 small trend of brighter blue clusters redder dubbed blue - tilt as evidence of mass - metallicity trend interpreted as evidence of self - enrichment of massive clusters started with dark matter halos or in large gas clouds effect larger in more massive galaxies apparent when color magnitude diagram plotted using redder filters ( mieske 2006 reanalyzed data sets small measurable mass - radius relationship for gcs in these galaxies ( kundu & zepf 2007b blue gcs average larger trend better defined for metal - poor clusters studies of extragalactic cluster systems assume uniform aperture correction or fit profile within small aperture to minimize uncertainties galaxy background zepf blue tilt likely of photometric bias mass - radius relationship uncertainties in color and magnitude not independent not orthogonal on color magnitude diagramredder filters used for magnitude axis in color - diagram uncertainties parallel to direction blue - tilt amplify effect uncertainties negate blue - tilt when blue filter plotted on magnitude axis explaining filter effect underlying galaxy light redder for massive galaxy explains variation strength blue - tilt with galaxy mass studies small color metallicity trend in colors metal - poor globular clusters with host galaxy mass larger mean sizes of blue gcs with larger galaxy mass may cause effect observational properties of metal - poor halo clusters uniform everywhere harris , w. et al . 2006 636 , 90 hempel zepf kundu geisler maccarone t 2007 661 , 768 ivanova 2006 636 , 979 jordn ,.. 2004 613 , 279 kim . 2006 647 , 276 kundu a. maccarone t zepf 2002 574 , l5 kundu . maccarone zepf 2007 662 , 525 kundu , a. maccarone t. jzepf s. e. puzia h. 2003 letters 589 , l81 kundu zepf s. e. 2007 660 , l109 maccarone kundu zepf s. e. 2004 606 , 430 mieske . 2006 653 , 193 sivakoff g. r. al . 2007 , 1246 smits maccarone. j. kundu zepf s. e. 2006 458 , 477 strader brodie j. p. spitler l beasley m. a. 2006 132 , 2333 | 0.445338 | 2,868 | 1,440 | , Saving $0.1 in GPT-4. |
155,057 | a model of electromagnetic interaction of relativistic heavy ions which makes it possible to calculate the inclusive and exclusive characteristics of such interactions is presented .
the multistep model includes ( 1 ) the single and double virtual photon absorption by a nucleus , ( 2 ) intranuclear cascades of produced hadrons and ( 3 ) statistical decay of excited residual nuclei described by evaporation , fission and multifragmentation mechanisms .
components of the model were verified by comparison of calculation results with experimental data obtained with monoenergetic photons and heavy ions at kek - tanashi synchrotron and cern sps accelerator , respectively . | the relativistic coulomb excitation of nuclei is a well - known phenomenon @xcite .
it is expected to play an important role in collisions of ultrarelativistic heavy nuclei at new colliders , rhic and lhc @xcite .
the process may be described in the framework of the weizscker - williams method of equivalent photons . according to this method the impact of lorentz - boosted coulomb field of a nucleus on the collision partner
may be treated as the absorption of equivalent photons .
due to the coherent nature of photon emission by the nucleus , such photons are almost real , @xmath0 , @xmath1 is the nuclear radius .
therefore photonuclear reaction data obtained in experiments with real monoenergetic photons and appropriate theoretical models describing photonuclear reactions may be used to predict the properties of electromagnetic dissociation process . since
an ultrarelativistic nucleus moving at the impact parameter @xmath2 with the velocity @xmath3 , ( @xmath4 ) spends a short time @xmath5 near the collision partner , the virtual photon spectrum contains all the frequencies up to the maximum energy @xmath6 : @xmath7 where @xmath8 is the fine structure constant , @xmath9 and @xmath10 are the modified bessel functions and @xmath11 .
the mean number of photons absorbed by the collision partner of mass @xmath12 is defined by : @xmath13 where @xmath14 is the appropriate total photoabsorption cross section obtained starting from the photoneutron threshold at @xmath15 mev . assuming the poison distribution for the multiphoton absorption with the mean multiplicity @xmath16 , eq.([eq:2 ] ) ,
the integral cross sections are calculated for the first- and second - order processes @xcite for a particular dissociation channel @xmath17 : @xmath18 @xmath19 with the following single and double photon spectral functions : @xmath20 the values @xmath21 and @xmath22 defined as the branching ratios for the considered disintegration channel @xmath17 in the single and double photon absorption , respectively , have to be calculated by our multistep model .
similar expressions may be written for the differential distributions of produced particles on the rapidity , transverse momentum and other variables . in order to obtain @xmath23 and
@xmath24 several mechanisms are included in a new relativistic electromagnetic dissociation ( reldis ) code for the monte carlo simulation , namely , the intranuclear cascade of fast particles produced after the photon absorption on a nucleon or nuclear pair @xcite , and the evaporation of nucleons and lightest fragments , binary fission or multifragmentation @xcite at a later stage of interaction .
the multifragmentation process dominates when the excitation energy of residual nucleus , @xmath25 , exceeds @xmath26 mev / nucleon and is described by the statistical multifragmentation model ( smm ) @xcite . for fissile nuclei
the evaporation may take place before or after the fission .
the competition of evaporation and fission is also described with the smm .
the lightest fragments may be also created via coalescence of fast nucleons into d , t , @xmath27 or @xmath28 @xcite .
the decay of highly excited light residual nuclei with @xmath29 is treated by the fermi break - up mechanism @xcite
. other details of the calculation scheme may be found in refs .
depending on the photon energy , @xmath30 , and mass number , @xmath12 , different processes take place in the nuclear photoabsorption . due to the contributions of several mechanisms : @xmath31 , @xmath32 and @xmath33 the calculated double differential spectra @xmath34 of pions and protons have complex shapes above the pion production threshold @xcite .
the spectra of fast particles @xmath35 and @xmath36 predicted by the model were compared in ref .
@xcite with available sets of experimental data at @xmath37 mev including the data obtained with kek - tanashi , the 1.3-gev electron synchrotron .
a satisfactory description of the spectra was obtained .
fast hadrons produced after the photon absorption initiate a cascade of subsequent collisions with the intranuclear nucleons leading to the heating of a residual nucleus .
later the nucleus undergoes de - excitation by means of the emission of nucleons and fragments . because of a low coulomb barrier in light nuclei the rates of proton and neutron emission are comparable . fragment spectra in the photoabsorption on a carbon nucleus
are given in fig .
[ fig : c ] .
the low energy part of the deuteron spectra is explained by the explosive fermi break - up while the high energy part is attributed to the coalescence mechanism .
since the main part of fast nucleons is emitted in the forward direction , the coalescence contribution dominates at small angles . on the contrary ,
the distribution of deuterons from fermi break - up is nearly isotropic . and 200 mev / c , respectively.,title="fig : " ] because of the option to accelerate also light oxygen ions at rhic , the calculations of the electromagnetic dissociation contribution for such ions should take into account both of the considered mechanisms .
the photoabsorption scenario is very different for heavy nuclei , gold and lead . in this case
neutrons are the most abundant particles produced in the electromagnetic collisions of ultrarelativistic nuclei @xcite .
when a heavy nucleus absorbs photons in the giant resonance region , @xmath38 mev , the evaporation model may be used with the assumption @xmath39 which leads to the emission of one or two neutrons . on the other side of the equivalent photon spectrum ,
when @xmath30 reaches the value of several gev , the multiple pion photoproduction on intranuclear nucleons becomes the main absorption mechanism . in this case up to 95% of the photon energy is released in the fast particles on average .
nevertheless , the remaining energy deposited in the residual nucleus is sufficient for evaporating many neutrons or even multifragmentation @xcite . as it was found in ref .
@xcite for pbpb collisions at sps ( @xmath40 gev beam ) and lhc ( @xmath41 tev beams ) the mean neutron multiplicities are 4.2 and 8.8 , respectively .
the same value for auau collisions at rhic ( @xmath42 gev ) is equal to 7.2 .
the neutron multiplicity distributions are shown in fig .
[ fig : nmult ] .
they are strongly peaked at the single neutron emission channel due to gdr decay , while there is a long tail of the multiple neutron emission originating from high excitations .
these results provide important information for designing large - rapidity detectors and zero - degree calorimeters at rhic and lhc .
[ fig : nmult ]
cern sps accelerator currently supplies the highest energy available for heavy ion studies .
a detailed understanding of fragmentation mechanisms at sps energies provides an important point for the extrapolation of the theoretical and experimental results to rhic and lhc energies . produced in the electromagnetic dissociation of @xmath43 ions at sps.,title="fig:",scaledwidth=90.0% ] produced in the electromagnetic dissociation of @xmath43 ions at sps.,title="fig:",scaledwidth=95.0% ]
partial break - up charge changing cross sections ( @xmath44 ) for 158 agev @xmath43 ions interacting with various targets , from h to pb , were measured recently @xcite .
although the contributions to fragmentation due to the electromagnetic and nuclear forces were not separated in the experiment @xcite , one can use reldis code to evaluate the role of the electromagnetic dissociation for each particular dissociation mode . as it is shown in fig .
[ fig : zdis ] , the main part of charge changing cross sections in pbpb collisions may be explained by the electromagnetic interaction .
calculations show that multiple neutron emission also plays a role in each of the partial charge changing channels with @xmath45 for pbpb collisions , see fig .
[ fig : zdis_neut ] .
the lost of each proton is accompanied with a high probability by the multiple neutron emission .
for example , up to 20 neutrons may be emitted in @xmath46 fragmentation channel , and the emission of 810 neutrons is the most probable process .
this is evident from the fact that due to a high coulomb barrier of pb nucleus , photoemission of protons takes place well above the gdr region , particularly in the region of quasideuteron absorption , where the excitation energy , @xmath47 mev , is sufficient to evaporate many neutrons . on the contrary ,
the mass distribution for @xmath48 channel is completely different since in this case @xmath49 and @xmath50 emission in gdr region strongly dominates .
however , a long tail of the multiple neutron emission also exists in this channel .
our expectation of the intense neutron emission in the charge changing electromagnetic collisions of heavy nuclei is very different from a simple picture assumed in ref .
@xcite , where the rates of neutron and proton emission above the photoproton threshold were determined by @xmath51 ratio of the nucleus which undergoes fragmentation .
a detailed understanding of the fragmentation mechanism may be obtained by measurements of fragment masses and detection of neutrons produced in collisions of ultrarelativistic heavy ions .
a variety of fragmentation mechanisms takes place in the electromagnetic dissociation of ultrarelativistic heavy ions , namely the coalescence , fermi break - up , evaporation , fission and multifragmentation .
a partial or complete disintegration of the colliding nuclei is possible despite the absence of geometrical overlap of the nuclear densities .
comparison with cern sps data on pbpb collisions demonstrates the dominance of electromagnetic dissociation in the partial charge - changing fragmentation channels with @xmath45 .
reldis model predicts also a very intense neutron emission in such channels . + the author gratefully acknowledges the fruitful collaboration with a.s.botvina , j.p.bondorf , a.s.iljinov and i.n.mishustin on the subjects of this talk .
discussions with a.b.kurepin , g.giacomelli , m.giorgini , l.patrizii and p.serra are greatly appreciated .
the author is very indebted to the organizing committee of the kek - tanashi symposium for the kind hospitality and financial support . | relativistic coulomb excitation of nuclei known phenomenon expected important in collisions ultrarelativistic heavy nuclei at new colliders rhic lhc process described weizscker - williams method of equivalent photons impact of lorentz - boosted coulomb field nucleus on collision partner absorption of equivalent photons coherent photon emission photons almost real nuclear radius photonuclear reaction data monoenergetic photons models predict electromagnetic dissociation ultrarelativistic nucleus moving at impact parameter @xmath2 velocity spends short time near collision partner virtual photon spectrum contains frequencies up to maximum energy fine structure constant modified bessel functions mean number of photons absorbed by collision partner mass defined by total photoabsorption cross section from photoneutron threshold at @xmath15 mev poison distribution for multiphoton absorption with mean multiplicity @xmath16integral cross sections calculated first second - order processes dissociation channel @xmath17 @xmath18 @xmath19 single double photon spectral functions values @xmath21 @xmath22 branching ratios disintegration channel @xmath17 single double photon absorption calculated by multistep model similar expressions differential distributions particles on rapidity transverse momentum variables obtain @xmath23 @xmath24 mechanisms new relativistic electromagnetic dissociation code monte carlo simulation intranuclear cascade of fast particles after photon absorption evaporation of nucleons lightest fragments binary fission multifragmentation later stage interaction multifragmentation excitation energy residual nucleus exceeds mev / nucleon described statistical multifragmentation model fissile nuclei evaporation before or after fission lightest fragments created via coalescence fast nucleons @xmath27 @xmath28 decay of highly excited light residual nuclei @xmath29 treated by fermi break - up mechanism details calculation scheme in refs photon energy mass number different processes in nuclear photoabsorptionmechanisms @xmath31 double differential spectra @xmath34 of pions protons complex shapes above pion production threshold spectra fast particles @xmath35 compared with experimental data @xmath37 mev - 1.3-gev electron synchrotron satisfactory description spectra obtained fast hadrons after photon absorption initiate collisions with intranuclear nucleons heating residual nucleus de - excitation emission nucleons fragments low coulomb barrier in light nuclei proton neutron emission comparable fragment spectra photoabsorption carbon nucleus in fig . low energy deuteron spectra explosive fermi break - up high energy coalescence mechanism fast nucleons emitted forward direction coalescence contribution dominates at small angles distribution deuterons from fermi break - up nearly isotropic 200 mev / c option accelerate light oxygen ions at rhic calculations electromagnetic dissociation contribution mechanisms photoabsorption scenario different for heavy nuclei gold lead neutrons most abundant particles in electromagnetic collisions ultrarelativistic nucleiheavy nucleus absorbs photons resonance region @xmath38 evaporation model used with assumption @xmath39 leads to emission one or two neutrons @xmath30 several gev multiple pion photoproduction on intranuclear nucleons main absorption mechanism 95% photon energy released in fast particles remaining energy residual nucleus sufficient for neutrons multifragmentation pbpb collisions at sps @xmath40 lhc mean neutron multiplicities 4.2 8.8 auau collisions at rhic @xmath42 7.2 neutron multiplicity distributions shown in fig . peaked at single neutron emission channel due to gdr decay long tail multiple neutron emission from high excitations results information for designing large - rapidity detectors zero - degree calorimeters at rhic lhc cern sps accelerator supplies highest energy for heavy ion studies understanding fragmentation mechanisms at sps energies for extrapolation theoretical experimental results to rhic lhc energies electromagnetic dissociation of @xmath43 ions at spspartial break - up charge changing cross sections @xmath44 158 agev @xmath43 ions interacting targets h to pb measured @xcite contributions to fragmentation electromagnetic nuclear forces not separated reldis code evaluate role electromagnetic dissociation each mode shown fig . charge changing cross sections pbpb collisions explained by electromagnetic interaction multiple neutron emission partial charge changing channels @xmath45 pbpb collisions lost each proton high probability multiple neutron emission up to 20 neutrons emitted in @xmath46 fragmentation channel emission 810 neutrons most probable high coulomb barrier pb nucleus photoemission of protons above gdr region quasideuteron absorption excitation energy @xmath47 mev evaporate neutrons mass distribution @xmath48 channel different @xmath49 and @xmath50 emission gdr region dominates long tail multiple neutron emission exists channel expectation intense neutron emission in charge changing electromagnetic collisions heavy nuclei different from @xcite rates neutron proton emission above photoproton threshold determined by @xmath51 ratio nucleus fragmentationunderstanding fragmentation mechanism by fragment masses neutrons collisions ultrarelativistic heavy ions fragmentation mechanisms coalescence fermi break - up evaporation fission multifragmentation partial or complete disintegration colliding nuclei possible overlap nuclear densities comparison with cern sps data electromagnetic dissociation in partial charge - changing fragmentation channels with @xmath45 reldis model predicts intense neutron emission author acknowledges collaboration with a.botvina j.bondorf.iljinov i.n.mishustin discussions with a.kurepin g.giacomelli m.giorgini l.patrizii p.serra appreciated indebted to committee kek - tanashi symposium for hospitality financial support | 0.361411 | 2,267 | 1,101 | , Saving $0.1 in GPT-4. |
155,063 | the quantum mechanical effect on the plasma two - stream instability is studied based on the dielectric function approach .
the analysis suggests that the degenerate plasma relevant to the inertial confinement fusion behaves differently from classical plasmas when the electron drift velocity is comparable to the fermi velocity . for high wave vector comparable to the fermi wave vector ,
the degenerate quantum plasma has larger regime of the two - stream instabilities than the classical plasma .
a regime , where the plasma waves with the frequency larger than 1.5 times of the langmuir wave frequency become unstable to the two - stream instabilities , is identified . | dense plasmas become important research subjects as plasmas with extremely high density are achieved in the laboratories @xcite and as our understanding of the astrophysical entities such as the white dwarf and super dense stars gets deeper .
the two - stream instability in dense plamsas is especially important to understand since dense electron beams are common phenomenon in the inertial confinement fusion ( icf ) @xcite and the violent astrophysical events in the dense astrophysics , involving the two - stream instabilities , are also being observed from the phenomena such as the gamma ray burst @xcite . it becomes a challenge to understand the dense plasmas , as the quantum effects begin to modify the physical process at high density ; a few physical processes deviating from the classical prediction have been identified @xcite .
the main goal of this paper is to study the electron quantum effect on the two stream instabilities in dense quantum plasmas .
there has been a general theoretical development in addressing the quantum effects on the two - stream instabilities by employing the fluid - type equation @xcite . in this paper , however , the author will utilize the more realistic lindhard dielectric function @xcite and will focus on the completely degenerate electron plasma with the density from @xmath0 to @xmath1 .
two cases are studied ; the fist case is one in which the plasma has two different group of electrons with different drift velocities and the second case is one in which the electrons have different velocity with the ions .
the findings are as follows .
when the electron drift velocity is comparable to the fermi velocity , the quantum plasma and classical plasma behave very differently . for high wave vector comparable to the fermi vector ,
the two - stream instability regime from the quantum plasmas is larger than the classical plasmas with the temperature comparable to the fermi energy .
a regime is identified where the unstable langmuir wave can have higher frequency than the langmuir wave frequency by 1.5 times .
this paper is organized as follows . in section
ii , the dielectric function approaches in the two - stream instability is introduced . in section iii
, we consider the case when there are two different group of degenerate electron with different drift velocities . in section iv , we consider the case when the degenerate electrons have different drift velocity with the ions . in section
v , the summary and discussion are provided .
the longitudinal dielectric function of a plasma is given as @xmath2 where the summation is over the group of particle species and @xmath3 is the particle susceptibility .
given the dielectric function @xmath4 , the analysis of the two - stream instability can be obtained by finding the root of @xmath5 as a function of @xmath6 ; if the dielectric function has a root @xmath7 as a function of @xmath6 in the complex upper - half plane , the plasma is unstable to the two - stream instabilities . in classical plasmas ,
the susceptibility is given as @xmath8d^3 \mathbf{v}\ ] ] where @xmath9 ( @xmath10 , @xmath11 ) is the particle mass ( charge , density ) and @xmath12 is the distribution with the normalization @xmath13 . for the degenerate electrons ,
the susceptibility @xmath14 of the free electron plasma has been computed by lindhard @xcite and is given as @xmath15 where @xmath16 is the fermi velocity , @xmath17 ( @xmath18 is the fermi energy ( fermi wave vector ) , @xmath19 , @xmath20 , and @xmath21 .
the fermi energy is given as @xmath22 where @xmath23 .
the real part of @xmath24 is given as @xmath25 in the next two sections , we consider the two cases of two - stream instabilities .
the first case is when there are two groups of electrons .
each group is the maxwellian distribution ( the degenerate fermi distribution ) with the same density and temperature but with different drift .
the second case is when the maxwellian ( complete degenerate ) electrons have different drift velocity with the ions .
the ions is assumed to be the maxwellian with the temperature comparable to the fermi energy . in the first case ,
the dielectric function is given as @xmath26 where @xmath27 is the drift velocity .
in the second case , the dielectric function is given as @xmath28
in this section , we analyze eq .
( [ eq : ele ] ) . in the conventional classical two - stream instability
, the threshold condition for the two stream instabilities is that there should be a local maxima between @xmath29 and that local maxima should be less than 0 . in fig .
( [ fig1 ] ) , we plot the classical dielectric function as a function of @xmath30 for a particular @xmath31 .
the hump at @xmath32 is the local maxima and it is less than 0 ; the plasma langmuir wave is unstable to the two stream instabilities of the classical plasmas . in fig .
( [ fig2 ] ) , we plot @xmath33 as a function of @xmath30 for a fixed @xmath31 in an example plasma .
the classical plasma does not have the local maxima .
the degenerate plasma has a local maxima at @xmath34 but is larger than zero . for this particular @xmath31 ,
the plasma langmuir wave is stable to the two - stream instabilities . in fig .
( [ fig3 ] ) , the only difference in the physical parameter with fig .
( [ fig2 ] ) is the wave vector . in this case
, the classical plasma does not have the local maxima and the degenerate plasma has the local maxima smaller than 0 ; the degenerate plasma ( classical plasma ) langmuir wave is unstable ( stable ) to the two - stream instability .
for a fixed electron density and temperature ( or completely degenerate ) , we plot @xmath33 as a function of @xmath30 and determine the critical @xmath35 values for which the plasma becomes unstable to the two - stream instabilities .
the necessary condition for the two - stream instabilities is @xmath36 . by varying the drift velocity @xmath37
, we can determine the boundary of the two - stream instabilities . in fig .
( [ fig4 ] ) , we do this for a particular plasma with @xmath38 , @xmath39 .
the fermi energy is @xmath40 and we choose the electron temperature of the classical plasma as @xmath41 since the average kinetic energy of the electron in the completely degenerate case is @xmath42 . from fig .
( [ fig4 ] ) , it can be concluded that the regime of the two - stream instability is larger in quantum prediction than classical prediction .
as the drift velocity becomes larger than @xmath43 , the difference between the quantum plasma and the classical plasma is small .
similar analysis suggests that , for more dense plasma , the regime of the quantum deviation , where the degenerate plasma ( classical plasma ) is unstable ( stable ) , widen further .
in this section , we analyze eq .
( [ eq : ele ] ) .
we employ the same criteria that was used in sec .
ii . in fig . ( [ fig5 ] )
, we plot the real part of the dielectric function of the classical plasmas and the degenerate plasmas .
the condition for the two - stream instability is the existence of the local maxima of the dielectric function that is lower than 0 . in this section
, we assume that the ion temperature is zero and the ion can be treated as the classical particle as the de broglie wave length of the ion is smaller than any other length scale in the regime of our interest .
as well - known in the conventional lindhard function @xcite , the degenerate plasma can support the collective langmuir waves whose wave length is much smaller than the classical debye length .
the regime of the two stream instability is higher in the degenerate plasma than the classical plasmas due to this fact . in fig .
( [ fig6 ] ) , we plot the threshold condition for the two - stream instability in the degenerate plasma and classical plasmas ; the necessary condition is @xmath36 . for the classical plasmas ,
the regime of the instabilities is smaller than the degenerate plasma . for the plasma with the lower density ,
the difference between the classical plasma and the degenerate plasma is smaller than the example that we have shown .
the example in fig .
( [ fig5 ] ) is the case when the classical plasmas are stable regardless of the wave vector .
as shown in the figure , the frequency of the unstable langmuir wave is @xmath44 .
such high frequency wave can not be sustained in the classical plasmas and can only be supported via the diffraction and degeneracy effect of the electrons . for the same plasma in fig .
( [ fig5 ] ) , the unstable langmuir wave frequency can reach @xmath45 .
in this paper , the author has analyzed the effect of the electron degeneracy and diffraction on the two - stream instability , focused on the plamsas relevant to the inertial confinement fusion and the astrophysical system ranging in density from @xmath0 to @xmath1 .
our approach is based on the lindhard random phase approximation , which takes into accounts the quantum degeneracy and diffraction .
we compare our prediction based on the lindhard approaches with the classical plasma .
the analysis suggests that the quantum effect become more pronounced as the density gets higher and that the classical dielectric function describes , as accurately as the lindhard function , the two - stream instabilities when the drift velocity is higher than 3.5 times of the fermi velocity .
however , when the drift velocity is smaller than 3.5 times of the fermi velocity , the quantum plasma and classical plasmas behave quite differently to the two - stream instability .
prominently , for the both cases we consider in secs .
ii and iii , the regime of the instability predicted by the lindhard approach is considerably larger than the classical plasma as illustrated in figs .
( [ fig4 ] ) and ( [ fig6 ] ) .
we also have shown that the unstable langmuir wave due to the two - stream instability can have the frequency as large as @xmath46 , which is impossible in the classical plasmas .
theses findings in this paper can have major implications on the backward raman scattering @xcite and the beam stopping by the dense background plasmas . | dense plasmas important research high density achieved in laboratories understanding of astrophysical white dwarf super dense stars deeper two - stream instability in dense important dense electron beams common in inertial confinement fusion violent astrophysical events two - stream instabilities observed gamma ray burst challenge to understand dense plasmas quantum effects modify physical process at high density few processes deviating from classical prediction identified main goal paper study electron quantum effect on two stream instabilities in dense quantum plasmas general theoretical development quantum effects two - stream instabilities fluid - type equation realistic lindhard dielectric function focus on degenerate electron plasma density from @xmath0 to @xmath1 two cases studied electrons different drift velocities different velocity ions findings electron drift velocity comparable to fermi velocity quantum plasma classical plasma behave differently high wave vector comparable two - stream instability regime from quantum plasmas larger than classical plasmas temperature comparable to fermi regime identified unstable langmuir wave higher frequency than langmuir wave by 1.5 times paper organizedsection ii dielectric function approaches two - stream instability introduced section iii two degenerate electron different drift velocities section iv degenerate electrons different drift velocity ions section v summary discussion provided longitudinal dielectric function plasma @xmath2 summation particle species @xmath3 particle susceptibility dielectric function @xmath4 analysis two - stream instability finding root @xmath5 function @xmath6 dielectric function root @xmath7 function @xmath6 complex upper - half plane plasma unstable two - stream instabilities classical plasmas susceptibility @xmath8d^3 @xmath9 particle mass charge density @xmath12 distribution normalization @xmath13 degenerate electrons susceptibility @xmath14 free electron plasma computed lindhard @xcite @xmath15 @xmath16 fermi velocity @xmath17 fermi energy wave vector @xmath19 @xmath21 fermi energy @xmath22 @xmath23 real part @xmath24 @xmath25 next two sections consider two cases two - stream instabilities first case two groups electrons each group maxwellian distribution fermi distribution same density temperature different driftsecond case maxwellian electrons different drift velocity with ions ions assumed be maxwellian temperature comparable to fermi energy first case dielectric function as @xmath26 @xmath27 is drift velocity second case dielectric function @xmath28 analyze eq . conventional classical two - stream instability threshold condition local maxima between @xmath29 maxima less than 0 fig classical dielectric function function of @xmath30 for @xmath31 hump at @xmath32 is local maxima less than 0 plasma langmuir wave unstable to two stream instabilities classical plasmas fig @xmath33 function of @xmath30 for fixed @xmath31 in example plasma classical plasma local maxima degenerate plasma has local maxima at @xmath34 larger than zero for @xmath31 plasma langmuir wave stable to two - stream instabilities only difference in physical parameter wave vector classical plasma local maxima degenerate plasma local maxima smaller than 0 langmuir wave unstable to two - stream instabilityfixed electron density temperature completely degenerate plot @xmath33 function of @xmath30 determine critical @xmath35 values plasma unstable to two - stream instabilities necessary condition two - stream instabilities @xmath36 varying drift velocity @xmath37 determine boundary two - stream instabilities fig . fig4 plasma with @xmath38 @xmath39 fermi energy @xmath40 choose electron temperature classical plasma @xmath41 average kinetic energy electron degenerate @xmath42 regime two - stream instability larger in quantum prediction classical prediction drift velocity larger than @xmath43 difference quantum plasma classical plasma small more dense plasma regime quantum deviation degenerate plasma classical plasma unstable widen analyze eq . ele employ same criteria sec . ii fig . fig5 plot real part dielectric function classical plasmas degenerate plasmas condition two - stream instability local maxima dielectric function lower than 0 assume ion temperature zero ion treated as classical particle de broglie wave length ion smaller than other length scale regimeknown lindhard function degenerate plasma langmuir waves wave length smaller than classical length regime two stream instability higher in degenerate plasma than fig threshold condition for two - stream instability in degenerate classical plasmas necessary condition is @xmath36 for classical plasmas regime instabilities smaller than degenerate plasma plasma lower density difference between classical degenerate plasma smaller example fig fig5 classical plasmas stable regardless wave vector frequency unstable langmuir wave is @xmath44 high frequency wave sustained in classical plasmas supported via diffraction degeneracy effect of electrons same plasma unstable langmuir wave frequency reach @xmath45 author analyzed effect electron degeneracy diffraction on two - stream instability focused on plamsas relevant to inertial confinement fusion astrophysical system density from @xmath0 to @xmath1 approach based on lindhard random phase approximation quantum degeneracy diffraction prediction with classical plasmaanalysis suggests quantum effect pronounced density higher classical dielectric function describes two - stream instabilities when drift velocity higher than 3.5 times fermi velocity drift velocity smaller than 3.5 quantum plasma classical plasmas behave differently to two - stream instability regime instability predicted by lindhard approach larger than classical plasma illustrated in figs fig6 unstable langmuir wave due to two - stream instability frequency large as @xmath46 impossible in classical plasmas findings on backward raman scattering @xcite beam stopping by dense background plasmas | 0.350319 | 2,305 | 1,151 | , Saving $0.1 in GPT-4. |
155,085 | accurate rotation - vibration line lists are calculated for silicon monoxide .
line lists are presented for the main isotopologue , @xmath0si@xmath1o , and for four monosubsituted isotopologues ( @xmath2si@xmath1o , @xmath3si@xmath1o , @xmath0si@xmath4o and @xmath0si@xmath5o ) , in their ground electronic states .
these line lists are suitable for high temperatures ( up to 9000 k ) including those relevant to exoplanetary atmospheres and cool stars .
a combination of empirical and _ ab initio _ methods is used : the potential energy curves are determined to high accuracy by fitting to extensive data from the analysis of both laboratory and sunspot spectra ; a high quality _ ab initio
_ dipole moment curve is calculated at the large basis set , multi - reference configuration interaction level .
a partition function plus full line lists of rotation - vibration transitions are made available in an electronic form as supplementary data to this article and at www.exomol.com .
molecular data ; opacity ; astronomical data bases : miscellaneous ; planets and satellites : atmospheres ; stars : low - mass [ firstpage ] | silicon monoxide ( sio ) is a widely observed astronomical species and appears to be ubiquitous in our galaxy . since its initial detection by @xcite
, sio has proved to be a key astrophysical maser and its very bright maser emissions are the subject of continued study @xcite .
sio was originally detected in the interstellar medium by @xcite .
its emissions were subsequently observed in the envelopes of oxygen - rich giant and super - giant stars @xcite and the remnants of supernova 1987a @xcite .
sio was detected in absorption in cool giant stars @xcite including @xmath6 tau @xcite .
sio absorptions are prominent in sunspots which have proved to be a fruitful source of spectroscopic data on highly excited states @xcite which we exploit below
. furthermore recent detections of hot , dense exoplanets have led to the speculation that their atmospheres might contain significant quantities of sio from vaporised silicates @xcite .
these astronomical applications , combined with technological uses of sio spectra @xcite , have motivated a number of laboratory studies .
@xcite,@xcite , @xcite , @xcite and @xcite have all produced molecular constants characterising sio vibration - rotation states .
transition probabibilities or einstein - a coefficients have also been provided by @xcite , @xcite and @xcite .
transition line lists have been constructed using these data @xcite , but none of them appears to be particularly complete .
for example no single line list combines a comprehensive set of transition frequencies with an accurate model for the transition intensities .
it is this that we aim to do here .
@xcite reported transition probabilities for a large number of transitions based on a semi - empirical dipole moment which they claim should give results accurate to about 10 % .
@xcite computed an accurate electric dipole moment function for the sio molecule which not only reproduces accurate dipole moments , but also reproduces line strengths for vibrational - rotational transitions within the ground - state manifold .
the resulting line lists are for @xmath0si@xmath1o , @xmath2si@xmath1o and @xmath3si@xmath1o for @xmath7 but @xmath8 is limited to 15 .
intensities were calculated using band intensities and hnl - london factors .
@xcite calculated an _ ab initio _ @xmath0si@xmath1o line list for @xmath9 and @xmath10
although the publically available line list from the strasbourg data centre contains only 503 transitions .
@xcite s intensities are significantly larger than those of @xcite and they estimate their intensities are only accurate to about 20 % .
the exomol project @xcite aims to provide line lists of spectroscopic transitions for key molecular species which are likely to be important in the atmospheres of extrasolar planets and cool stars ; it aims , scope and methodology have been summarised by @xcite .
line lists for @xmath11 xh molecules , x = be , mg , ca , have already been published @xcite . in the present paper ,
we present rotation - vibration transition lists and associated spectra for the five major isotopologues of sio .
these line lists are particulary comprehensive and should be valid for temperatures up to 9000 k.
rotation - vibration line lists for the ground electronic state of sio were obtained by direct solution of the nuclear motion schrdinger equation using the program level8.0 @xcite . in principle
the calculations were initiated using a potential energy curve ( pec ) calculated _ ab initio_. in practice , as detailed below , there are sufficient experimental data available for sio that the pec was actually characterised by fitting to these data using the program dpotfit 1.1 @xcite .
there appears to be no experimental measurements of any sio transition dipoles and a single measurement of its permanent dipole moment as a function of the vibrational state by @xcite .
we determined a new dipole moment curve ( dmc ) using high level _ ab initio _ calculations .
these are compared to previous , high - level _ ab initio _
determinations @xcite given below .
the _ ab initio _
calculations were performed using molpro @xcite ; we tested both the coupled cluster ( ccsd(t ) ) and multi - reference configuration interaction ( mrci ) methods with large basis sets .
our largest , and best , calculations used an aug - cc - pcv5z basis set for the ccsd(t ) study and aug - cc - pwcv5z for mrci ; in both cases a correction due to the correlation of the core electrons was included but the corresponding correction due to relativistic effects was found to be very small and was neglected .
these calculations give a value for the dipole at equilibrium of 3.10 d and 3.07 d , respectively , which bracket the experimentally determined value of 3.088 d @xcite .
the _ ab initio _
dmc grid points were used directly in level .
figure [ fig : dipole ] compares the dmc s arising from our calculations with those obtained by @xcite ( cck ) using an mrci method .
other previous studies based on the use of similar ccsd(t ) method gave curves similar to our ccsd(t ) calculation .
the ccsd(t ) dipoles appear correct at short bondlengths but behave in an unphysical fashion for @xmath12 .
ccsd(t ) is known to have problems as molecules are dissociated and so this model was not pursued . again at short @xmath13 our mrci dipoles are similar to those of cck .
it is unclear why cck s dipoles do not vary smoothly as @xmath13 is increased but such behaviour is not consistent with obtaining reliable transition intensities .
all calculations presented below therefore use our mrci dipole moments .
llllll reference & method & transitions & temperature & frequency range & uncertainty + & & & ( k ) & ( @xmath14 ) & ( @xmath14 ) + @xcite & laboratory & @xmath15 , @xmath16 & 1000 9000 & 0.13 1.67 & @xmath17 + & & @xmath0si@xmath1o @xmath18 + & & @xmath2si@xmath1o @xmath19 + & & @xmath0si@xmath4o @xmath20 + + @xcite & sunspots & @xmath21 , @xmath22 & 3200 & 900 1300 & 0.006 + & & @xmath0si@xmath1o @xmath23 @xmath24 + & & @xmath2si@xmath1o @xmath25 @xmath26 + & & @xmath3si@xmath1o @xmath25 @xmath27 + the potential energy curve is essentially characterised by fitting to spectroscopic data . measurements with very small estimated uncertainties were chosen for the fit such that any uncertainty in the fitted surface depends only on the accuracy of the fit and , for higher energies , the extrapolation beyond the experimental input values .
measurements for multiple isotopologues were used both to maximise the input data and to ensure that the resulting curve is applicable to all isotopologues , including the yet - to - be observed spectrum of @xmath0si@xmath5o .
the most comprehensive and accurate sets of available measurements are the infrared ro - vibrational sunspot lines detected by @xcite and the microwave rotational laboratory lines recorded by@xcite .
the potential was expressed as an extended morse oscillator potential function ( emop ) : @xmath28 ^ 2,\ ] ] where @xmath29 @xmath30 and @xmath31 was set to 2 and @xmath32 to 6 .
spectroscopically derived values for the dissociation energy , @xmath33 @xmath14 @xcite and @xmath34 @xcite were used .
.fitting parameters used in the extended morse oscillator potential , see eq .
( uncertainties are given in parentheses in units of the last digit . ) [ cols="<,<",options="header " , ] since there are no measured intensities to compare with , we compare the results of our calculations with astronomical spectra .
sunspots display many sio features over a wide range of wavelengths .
high resolution sunspot spectra have been compiled by @xcite .
figure [ sun.eps ] compares small regions of these spectra with our line lists .
note that only the theoretical spectra , generated at @xmath35=3200 k , are given in the absolute units of @xmath36molecule ) .
these spectra , which are chosen to cover regions where @xmath37 is both 1 and 2 , illustrate the excellent agreement between our line positions and the observed features .
our intensities are also well correlated with the observed ones .
we note that the sunspot spectra show many other features in this figure .
these features are either due to water @xcite or so to some so - far unidentified species .
figure [ giant.eps ] compares the spectra of four red giant @xcite with our simulated spectrum of @xmath0si@xmath1o at a temperature of 3200 k. these spectra are for the @xmath38 overtone spectrum . to fully reproduce
the observed stellar spectrum would require running a stellar model which is beyond the scope of this paper .
however it clear that both the line positions and general band structure predicted by our calculations are in very good agreement with the observations .
we present comprehensive line lists for the five most important isotopologues of sio .
these are based on the direct solution of the nuclear motion schrdinger equation using a potential energy curve obtained by fitting to an extensive data set of measured transitions .
these data are reproduced to near experimental accuracy resulting in high accuracy line positions . a new _ ab initio _
dipole moment is computed , which appears to behave more physically at large internuclear separations .
this dipole is used to compute einstein a coefficients for all possible dipole - allowed transitions within each sio isotopologue .
the result is a comprehensive line list for each species , including the first data for @xmath0si@xmath5o .
the line lists can be downloaded from the cds , via ftp://cdsarc.u-strasbg.fr/pub/cats/j/mnras/ , or from http://cdsarc.u-strasbg.fr/viz-bin/qcat?j/mnras/ , or from www.exomol.com .
this work is supported by erc advanced investigator project 267219 .
r. j. , 2006 , dpotfit 1.1 a computer program for fitting diatomic molecule spectral data to potential energy functions .
university of waterloo chemical physics research report cp-662r , http://leroy.uwaterloo.ca/programs/ r. j. , 2007 , level 8.0 a computer program for solving the radial schrdinger equation for bound and quasibound levels .
university of waterloo chemical physics research report cp-663 , http://leroy.uwaterloo.ca/programs/ | silicon monoxide ( sio ) observed astronomical species ubiquitous in our galaxy detection by @xcite key astrophysical maser bright emissions subject of study sio detected in interstellar medium by @xcite emissions observed in oxygen - rich giant super - giant stars remnants of supernova 1987a sio detected in cool giant stars including @xmath6 tau sio absorptions prominent in sunspots source spectroscopic data on excited states detections of hot exoplanets atmospheres contain quantities sio from vaporised silicates astronomical applications technological uses of sio spectra motivated laboratory studies @xcite produced molecular constants characterising sio vibration - rotation states transition probabibilities coefficients provided by @xcite transition line lists constructed using data none complete no line list combines comprehensive transition frequencies with accurate model transition intensities @xcite reported transition probabilities for transitions - empirical dipole moment to 10 % @xcite computed accurate electric dipole moment function for sio molecule reproduces dipole moments line strengths for vibrational - rotational transitionsline lists for @xmath0si @xmath8 limited to 15 intensities calculated using band intensities hnl - london factors @xcite calculated ab initio @xmath0si line list for @xmath9 @xmath10 line list strasbourg data centre contains 503 transitions @xcite s intensities larger @xcite accurate to 20 % exomol project @xcite aims line lists spectroscopic transitions for key molecular species extrasolar planets cool stars summarised by line lists for @xmath11 xh molecules published paper rotation - vibration transition lists spectra for five major isotopologues of sio line lists comprehensive valid for temperatures up to 9000 k lists ground electronic state sio obtained by direct solution nuclear motion schrdinger equation program level8.0 @xcite calculations potential energy curve pec ) calculated ab initio sufficient experimental data for sio no experimental measurements of sio transition dipoles measurement permanent dipole moment vibrational state by @xcite determined new dipole moment curve ) using high level ab initio calculationscompared previous high determinations @xcite calculations performed using molpro @xcite tested coupled cluster multi - reference configuration interaction mrci methods large basis sets largest calculations used aug - cc - pcv5z basis set for ccsd(t aug - cc - pwcv5z mrci correction due correlation core electrons included correction relativistic effects small neglected calculations give value dipole at equilibrium 3.10 d 3.07 d experimentally determined value 3.088 d @xcite ab initio dmc grid points used in level figure compares dmc s calculations with @xcite mrci method previous studies gave curves similar ccsd(t ) calculation ccsd(t dipoles correct at short bondlengths behave unphysical for @xmath12 problems molecules model not pursued @xmath13 mrci dipoles similar to cck unclear why cck s dipoles vary smoothly as @xmath13 increased not consistent reliable transition intensities calculations use mrci dipole momentsreference method transitions temperature frequency range uncertainty @xmath14 @xcite laboratory @xmath15 @xmath16 1000 9000 0.13 1.67 @xmath17 @xmath0si@xmath1o @xmath2si @xmath19 @xmath0si@xmath4o @xmath20 @xcite sunspots @xmath21 @xmath22 3200 900 1300 0.006 @xmath0si @xmath23 @xmath24 @xmath2si @xmath26 potential energy curve fitting to spectroscopic data measurements small uncertainties for fit uncertainty depends on accuracy higher energies extrapolation beyond input values measurements for multiple isotopologues maximise input data curve all including spectrum @xmath0si@xmath5o comprehensive measurements infrared - vibrational sunspot lines @xcite microwave rotational laboratory lines@xcite potential expressed as extended morse oscillator potential function @xmath28 2 @xmath29 @xmath30 @xmath31 set to 2 @xmath32 to 6 spectroscopically derived values for dissociation energy @xmath33 @xmath14 @xcite @xmath34 usedparameters extended morse oscillator potential see eq uncertainties parentheses units last digit cols=" no measured intensities compare compare calculations with astronomical spectra sunspots display sio features wavelengths high resolution sunspot spectra compiled by @xcite figure sun.eps compares spectra with line lists theoretical spectra @xmath35=3200 k absolute units @xmath36molecule spectra regions @xmath37 1 and 2 illustrate agreement between line positions observed features intensities correlated with observed sunspot spectra show features due to water @xcite or unidentified species figure giant.eps compares spectra red giant @xcite with simulated spectrum @xmath0si@xmath1o at temperature 3200 k spectra @xmath38 overtone spectrum reproduce observed stellar spectrum stellar model beyond line positions general band structure calculations with observations comprehensive line lists for five important isotopologues of sio based on direct solution nuclear motion schrdinger equation potential energy curve extensive data set measured transitions data reproduced near experimental accuracy high accuracy line positionsnew dipole moment computed large internuclear separations dipole einstein coefficients dipole - transitions each sio isotopologue result comprehensive line list each species including first data @xmath0si@xmath5o line lists downloaded from cds.u-strasbg.fr/mnras www work supported by erc advanced investigator project 267219. 2006 dpotfit 1.1 computer program fitting diatomic molecule spectral data energy functions university waterloo chemical physics report cp-662r 2007 level 8.0 program solving radial schrdinger equation bound quasibound levels university waterloo physics report cp-663 | 0.352115 | 2,442 | 1,209 | , Saving $0.1 in GPT-4. |
155,091 | we performed n - body / sph simulations of merging gas - rich disk galaxies with mass ratios of 1:1 and 3:1 .
a stellar disk and bulge component and a dark halo was realized with collisionless particles , the gas was represented by sph particles . since we did not include star formation we focused on the gas dynamics and its influence on the structure of merger remnants .
we find that equal - mass mergers are in general more effective in driving gas to the center . around 50% of the gas
resides in the central regions after the merger is complete .
the gas in unequal - mass mergers keeps a large amount of its initial angular momentum and settles in a large scale disk while only 20% -30% of the gas is driven to the center .
this general result is almost independent of the merger geometry .
the gas in the outer regions accumulates in dense knots within tidal tails which could lead to the formation of open clusters or dwarf satellites .
later on , the gas knots loose angular momentum by dynamical friction and successively sink to the center of the remnant thereby increasing the total gas content of the disk . due to the influence of the gas
the simulated merger remnant becomes more oblate than its pure stellar counterpart .
the shape of the stellar losvd changes and the third - order gauss - hermite coefficient @xmath0 is in good agreement with observations . | the disk - galaxies are constructed in dynamical equilibrium ( hernquist 1993 ) and consist of an exponential stellar disk , a bulge with a hernquist profile and a pseudoisothermal dark halo .
the smaller galaxy has 1/3 of the mass and 1/3 of the particles of the massive one .
the scale length is reduced by @xmath1 . in total
we used 88888 particles for the collisionless component and 26666 gas particles simulated with sph and an isothermal equation of state .
both galaxies approach each other on parabolic orbits with slightly inclined disks relative to the orbital plane .
all simulations were performed on a sun ultra 60 workstation .
recent simulations have shown that pure stellar 3:1 mergers of disk - galaxies lead to fast rotating ( barnes , 1998 ) and isotropic , disky galaxies whereas 1:1 mergers result in anisotropic , boxy ellipticals ( naab , burkert & hernquist 1999 ) .
we tested this hypothesis with a model that also follows the evolution of a massive gas component ( see figure 1 ) . during the mergers a large amount of gas falls to the central region of the remnants ( inside 300pc ) .
part of the gas is driven to the center by the formation of a tidally induced bar during the first interaction of the two disks .
the accretion rate in all cases peaks when the two galaxy centers merge ( figure 3 ) . there
the gas could experience a central starburst or fuel a central black hole .
equal - mass mergers are in general more effective in driving gas to the center ( barnes & hernquist 1996 ; mihos & hernquist 1996 ) .
+ the accretion history for one selected example of a 1:1 merger is shown in figure 3 .
it can be divided in two major accretion events .
first the gas flows to the center along a tidally induced bar and then the two galaxy centers merge . in the end 50% of the gas
is funneled into the central 300pc of the remnant .
the centrifugal support of the gas in 3:1 mergers seems to be much stronger and prevents the gas from clumping at the center .
instead the gas in unequal - mass mergers settles in a large scale disk ( figure 2 ) and only around 30% of the gas is driven to the center , almost independent of the merger geometry ( figure 3 ) .
the gas in the outer regions accumulates in dense knots within tidal tails ( see figure 1 : c and c ) which could lead to the formation of open clusters or dwarf satellites .
later on , the gas knots loose angular momentum by dynamical friction and successively sink to the center of the remnant thereby increasing the gas content of the disk . as in the equal - mass case
the main accretion event is the merger of the two galaxy centers .
+ the resulting stellar remnant has an oblate shape with disky isophotes and kinematical properties in good agreement with observations of faint ellipticals .
in addition , we investigate the losvd of one unequal - mass gas - rich merger remnant .
simulations of pure collisionless unequal - mass mergers indicate a positive value of @xmath0 , in contrast to observations ( naab & burkert , 2001 ) . in gas - rich mergers
the sign of @xmath0 changes due to the influence of the gas disk and is in good agreement with observations of massive elliptical galaxies ( bender , saglia & gerhard 1994 ) .
a comparison between simulated data and observations is shown in figure 3 .
the stellar particles alone and together with the gas particles give the right correlation .
therefore the presence of gas and the late formation of central disks might have played an important role during the formation of elliptical galaxies .
+ simulations like this can help to understand the evolution of major starbursts , agns and central power - law or disk - like structures that are observed in fast rotating disky ellipticals ( faber et al .
we have to note that the influence of gas on the global structure of elliptical galaxies is not well understood since this influence is sensitive to some certain details about star formation which is not included in the simulations we presented here .
future investigations with better resolution ( in combination with grape5 ) and carefully implemented recipes for starformation will help to understand the role of gas dynamics and starformation in the formation history of elliptical galaxies .
barnes , j. & hernquist , l. , 1996 , , 471 , 115 barnes , j. , 1998 , in galaxies : interactions and induced star formation , ed .
d. friedli , l. mertinet , & d.pfenniger ( saas - fee advanced course lecture notes , no . 26 : berlin : springer ) , 275 bender , r. , saglia , r. p. , & gerhard , o. e. 1994 , , 269 , 785 faber , s. m. , 1997 , , 114 , 1771 hernquist , l. , 1993 , , 86 , 389 mihos , j. c. & hernquist , l. , 1996 , , 464 , 119 naab , t. , burkert , a. & hernquist , l. , 1999 , , 523 , l133 naab , t. & burkert , a. 2001 , , 555 , l91 | disk - galaxies constructed in dynamical equilibrium hernquist 1993 ) consist exponential stellar disk bulge hernquist profile pseudoisothermal dark halo smaller galaxy has 1/3 mass 1/3 particles massive scale length reduced by @xmath1 used 88888 particles for collisionless component 26666 gas particles with sph isothermal equation state galaxies approach parabolic orbits slightly inclined disks orbital plane simulations performed on sun ultra 60 workstation pure stellar 3:1 mergers lead to fast rotating isotropic disky galaxies 1:1 mergers result in anisotropic boxy ellipticals tested hypothesis with model evolution massive gas component figure 1 during mergers gas falls to central region remnants driven to center by tidally induced bar accretion rate peaks when galaxy centers merge figure 3 gas could experience central starburst fuel central black hole equal - mass mergers effective gas to center accretion history for 1:1 merger shown in figure 3 divided in two major accretion events gas flows to center bar galaxy centers merge 50% gas funneled into central 300pc remnantcentrifugal support gas in 3:1 mergers stronger prevents clumping at center gas in unequal - mass mergers settles in large scale disk 30% driven to center independent of merger geometry gas in outer regions accumulates in knots within tidal tails could lead to open clusters dwarf satellites gas knots loose momentum sink to center increasing gas content main accretion event is merger of two galaxy centers resulting stellar remnant has oblate shape disky isophotes kinematical properties with observations faint ellipticals of unequal - mass gas - rich merger remnant simulations of collisionless mergers indicate positive value of @xmath0 observations in gas - rich mergers sign @xmath0 changes to gas disk with observations massive elliptical galaxies comparison between simulated data observations in figure 3 stellar particles gas particles give right correlation presence gas late formation of central disks might during formation elliptical galaxies simulations understand evolution of major starbursts agns central power - law disk - like structures in fast rotating disky ellipticalsinfluence gas on structure elliptical galaxies not understood star formation not included simulations future investigations better resolution grape5 recipes starformation understand gas dynamics starformation formation galaxies barnes , j. hernquist , l. 1996 471 , 115 barnes 1998 galaxies interactions induced star formation mertinet d.pfenniger saas - fee advanced course lecture notes 26 berlin springer 275 bender r. saglia gerhard 1994 269 , 785 faber 1997 114 , 1771 hernquist l 1993 86 , 389 mihos j hernquist 1996 464 , 119 naab burkert hernquist 1999 523 , l133 2001 555 , l91 | 0.185539 | 1,188 | 598 | , Saving $0.0 in GPT-4. |
155,094 | we discuss the properties of an object in the subaru deep field ( sdf ) classified as a galaxy in on - line data bases and revealed on the subaru images as a genuine polar - ring galaxy ( prg ) candidate .
we analyse available photometric data and conclude that this object consists of a @xmath0 gyr old early - type central body surrounded by a faint , narrow inner ring tilted at a @xmath1 angle relative to the polar axis of the host galaxy .
the halo surrounding the main stellar body exhibits a diversity of spatially extended stellar features of low surface brightness , including a faint asymmetric stellar cloud and two prominent loops .
these faint features , together with the unperturbed morphology of the central host , are clear signs of a recent coalescence of two highly unequal mass galaxies , most likely a pre - existing early - type galaxy and a close - by gas - rich dwarf galaxy .
the presumed stellar remnants observed near the edges of the ring , including possibly the surviving captured companion itself , indicate that the merger is still taking place .
[ firstpage ] galaxies : peculiar - galaxies : individual : sdss j132533.22 + 272246.7 - galaxies : photometry | mutual interactions between galaxies are probably one of the main processes leading to the formation of presently observed galaxies . the discovery of shells and ripples around elliptical galaxies ( kormendy & djorgovski 1989 ; tal et al . 2009 ) and of extended stellar tidal features around normal disc galaxies ( martnez - delgado et al .
2010 ) are among the most compelling evidence for this scenario . in more extreme and rare cases the remnants of the interacting galaxies
are not mixed in one smooth object , but some of the material settles quasi - statically into the equatorial plane of the main body , forming a polar - ring galaxy ( prg ; whitmore et al . 1990 ) .
alternatively , simulations show that the polar structure can form as the result of a `` secondary event '' during the evolution of the galaxy , involving the accretion of gas from a close , gas - rich dwarf companion ( reshetnikov & sotnikova 1997 ) . however , both formation scenarios are difficult to validate observationally since they predict the formation of a polar ring on short time scales of less than 1 - 2 gyr ( bekki 1998 ; bournaud & combes 2003 ) . since only about 0.5 per cent of all early - type galaxies have observable polar rings ( whitmore et al
. 1990 ) , on - going galaxy interactions that form a prg are probably extremely rare events .
a galaxy tidally disrupted while still in orbit around a much more luminous and massive companion should leave distinctive tidal stellar features before merging with the central galaxy or mixing into a seemingly smooth component .
these stellar debris can extend well beyond the central galaxy and mostly occur at a very low surface brightness level , making it difficult to detect in sky survey images .
in this paper we present an interpretation of existing observational data of the object sdss j132533.22 + 272246.7 ( hereafter referred to as rg ) identified as a ring galaxy during a supernova survey in the subaru deep field ( sdf ; graur et al .
, in preparation ) .
the object is centred at @xmath2 ( j2000 ) at a redshift of @xmath3 , which puts it at a distance of about 250 mpc , assuming @xmath4 km sec@xmath5 mpc@xmath5 .
this paper is organized as follows : section [ obs ] gives a description of the observations and data reduction ; in section [ results ] we present the observational results along with our interpretation of the object and the main conclusions are summarized in section [ conc ] .
[ cols="^,^ " , ] prgs are believed to form as a result of a dissipative galaxy merger between two disc galaxies , or by the accretion of gas from a small gas - rich companion by a pre - existing s0 galaxy ( see bekki 1998 ; bournaud & combes 2003 and references therein ) .
distinguishing between these two kinds of scenarios can be addressed by examining their morphology and by observing polar - ring formation in action in the local universe .
prgs are typically divided into two groups : ( 1 ) galaxies with an unusually compact and faint bulge and an extended disc - like ring with the central region cut out ; ( 2 ) galaxies with a dominant bulge and a narrow ring which is not extended in radius ( whitmore 1991 ; reshetnikov & sotnikova 1997 ) .
rg seems to fit into the second category . among known prgs , rg resembles eso 415-g26 by its faint narrow ring ( see e.g. fig . 2 of iodice et al .
2002a ) , central body morphology , red colours and the presence of tidal features extending beyond the host ( e.g. loops and debris , see whitmore et al .
1987 ) .
the morphology of a prg formed in a merger event depends not only on the orbital parameters , but also on the initial mass ratio of the two disc galaxies .
for instance , a mass ratio larger than @xmath6 can reproduce both the central discy s0 host and the inclined narrow ring of eso 415-g26 ( bekki 1998 ) .
in this view , the diffusely dispersed material around rg can be explained as stars pushed out from the victim galaxy during a head - on collision with a larger , more massive galaxy .
the faint stellar features around rg are due to polar material oscillating at large and small radii before gathering in the ring , and the interacting companion is expected to eventually merge with the central object ( bournaud & combes 2003 ) .
this scenario , previously proposed by whitmore , mcelroy & schweizer ( 1987 ) for eso 415-g26 , places strong constraints on the age of the system . considering the off - centre position of the ring and the distinctive structural features surrounding the main body ,
we conclude that rg is only a few dynamical periods old and is still developing ( see also schweizer , whitmore & rubin 1983 ) .
a scenario in which gas was accreted from a gas - rich companion onto rg can be safely ruled out .
in such a scenario , no clear sign relating the ring and the accreted companion is expected to be left even 1 or 2 gyr after the ring has formed .
the accreted matter , including only few stars , falls directly into a polar orbit and therefore a stellar halo around the polar ring is not expected , contrary to the merging scenario ( bournaud & combes 2003 ) .
understanding the role played by different mechanisms in prg formation requires a deep photometric study of a large sample of such objects at various redshifts .
the redshift of rg places it between the high-@xmath7 prg - candidates found in _
hst _ observations ( reshetnikov 1997 ; lavery et al . 2004 ; reshetnikov & dettmar 2007 ) and the low-@xmath7 prgs published and studied by whitmore et al .
( 1990 ) and others . however , observed prgs that could bridge the distance gap are very rare since they are difficult to detect by large and shallow surveys such as the sdss ( see brosch et al .
in particular , detecting tidal features or merger remnants in such systems often requires exquisite surface brightness sensitivity ( tal et al .
2009 ; smirnova , moiseev & afanasiev 2010 ; martnez - delgado et al .
rg is an excellent example , since the polar ring and the complex stellar halo structure could probably not have been detected by a cursory examination of sky survey images .
we present a photometric study of a candidate prg in the sdf .
the object has a complex structure consisting of : + 1 . a central host
which we identify as a @xmath0 gyr old early - type galaxy with an exponential bulge and a nearly edge - on disc ; + 2 . a faint off - centre narrow ring at a @xmath1 angle to the polar axis of the host ; + 3 .
a low surface brightness stellar halo extending asymmetrically on both sides of the ring ; + 4 .
at least two disrupted stellar debris in intermediate position angles between the ring and the polar axis of the host ; + 5 .
two outer loops on both sides in the direction of the stellar debris . +
our morphological study suggests that this galaxy belongs to the group of bulge - dominated prgs and that the ring was probably formed around the pre - existing host during a merger with a gas - rich dwarf companion .
this event probably took place less than 1 - 2 gyr ago , since it appears that the captured material has not yet fully dispersed into the halo or settled into the polar structure .
the authors wish to recognize and acknowledge the very significant cultural role and reverence that the summit of mauna kea has always had within the indigenous hawaiian community .
we are most fortunate to have the opportunity to conduct observations from this mountain .
this publication makes use of data products from the two micron all sky survey , which is a joint project of the university of massachusetts and the infrared processing and analysis center / california institute of technology , funded by the national aeronautics and space administration and the national science foundation .
funding for the sdss and sdss - ii has been provided by the alfred p. sloan foundation , the participating institutions , the national science foundation , the u.s .
department of energy , the national aeronautics and space administration , the japanese monbukagakusho , the max planck society , and the higher education funding council for england .
the sdss web site is http://www.sdss.org/. the sdss is managed by the astrophysical research consortium for the participating institutions .
the participating institutions are the american museum of natural history , astrophysical institute potsdam , university of basel , university of cambridge , case western reserve university , university of chicago , drexel university , fermilab , the institute for advanced study , the japan participation group , johns hopkins university , the joint institute for nuclear astrophysics , the kavli institute for particle astrophysics and cosmology , the korean scientist group , the chinese academy of sciences ( lamost ) , los alamos national laboratory , the max - planck - institute for astronomy ( mpia ) , the max - planck - institute for astrophysics ( mpa ) , new mexico state university , ohio state university , university of pittsburgh , university of portsmouth , princeton university , the united states naval observatory , and the university of washington .
99 bekki k. , 1998 , apj , 499 , 635 bournaud f. & combes f. , 2003 , a&a , 401 , 817 brosch n. , kniazev a.y . , moiseev a. , pustilnik s.a . , 2010 , mnras , 401 , 2067 bruzual g. & charlot s. 2003 , mnras , 344 , 1000 iodice e. , arnaboldi m. , sparke l.s . ,
gallagher j.s . ,
freeman k.c . , 2002a , a&a , 391 , 103 iodice e. , arnaboldi m. , sparke l.s . , freeman k.c
, 2002b , a&a , 391 , 117 kashikawa n. et al . , 2004 , pasj , 56 , 1011 kent s.m . , 1985 , apjs , 59 , 115 kormendy j. & djorgovski s. , 1989 , ara&a , 27 , 235 lavery r.j . , remijan a. , charmandaris v. , hayes r.d . ,
ring a.a . , 2004 , apj , 612 , 679 martnez - delgado d. et al .
2010 , aj , 140 , 962 ouchi m. et al .
, 2004 , apj , 611 , 660 reshetnikov v.p . , 1997,a&a , 321 , 749 reshetnikov v.p .
& dettmar r .-
j . , 2007 ,
astl , 33 , 222 reshetnikov v.p . & sotnikova n. , 1997 , a&a , 325 , 933 schweizer f. , whitmore b.c .
, rubin v.c . , 1983 ,
aj , 88 , 909 smirnova a.a .
, moiseev a.v .
, afanasiev v.l . , 2010 , mnras , 408 , 400 tal t. , van dokkum p.g . , nelan j. , bezanson r. , 2009 , aj , 138 , 1417 yagi m. , kashikawa n. , sekiguchi m. , doi m. , yasuda n. , shimasaku k. , okamura s. , 2002 , aj , 123 , 66 whitmore b.c . , 1991 , warped disks and inclined rings around galaxies , ed .
s. casertano , p. sackett and f. briggs ( new york : cambridge univ .
press ) , 60 whitmore b.c .
, lucas r.a . , mcelroy d.b .
, steiman - cameron t.y .
, sackett p.d . , olling r.p . , 1990 ,
aj , 100 , 1489 whitmore b.c . , mcelroy d.b . , schweizer f. , 1987 , apj , 314 , 439 | interactions between galaxies main to formation of observed galaxies discovery of shells ripples around elliptical galaxies extended stellar tidal features around normal disc galaxies evidence for scenario in extreme cases remnants interacting galaxies not mixed in object material settles into equatorial plane main body forming polar - ring galaxy simulations show polar structure can form secondary event during evolution galaxy accretion gas from gas - rich dwarf companion both formation scenarios difficult to predict formation polar ring on short time scales less than 1 - 2 gyr ( 1998 2003 only 0.5 per cent of early - type galaxies have observable polar rings galaxy interactions form prg rare galaxy tidally disrupted orbit around luminous companion should leave distinctive tidal stellar features before merging with central galaxy or mixing into smooth component stellar debris can extend beyond central galaxy occur at low surface brightness level difficult to detect in sky survey images paper interpretation of observational data of object sdss j132533.22 + 272246.7 rg ) identified as ring galaxy during supernova survey in subaru deep fieldobject centred at @xmath2 ( j2000 ) redshift of @xmath3 distance about 250 mpc assuming @xmath4 km sec@xmath5 mpc@xmath5 paper organized section [ obs ] observations data reduction section [ results ] observational results interpretation of object main conclusions summarized in section [ conc ] prgs form dissipative galaxy merger between two disc galaxies or accretion of gas from small gas - rich companion pre - existing s0 galaxy bekki 1998 2003 distinguishing scenarios examining morphology observing polar - ring formation local universe prgs divided into two groups 1 ) galaxies compact faint bulge extended disc - like ring central region cut out ; 2 ) galaxies dominant bulge narrow ring not extended radius 1991 1997 ) rg second category prgs resembles eso 415-g26 faint narrow ring central body morphology red colours tidal features extending morphology of prg in merger event depends on orbital parameters initial mass ratio of two disc galaxiesmass ratio larger than @xmath6 reproduce central discy s0 host and ring of eso 415-g26 ( bekki 1998 ) dispersed material around rg as stars pushed out from victim galaxy during collision with larger galaxy faint stellar features around rg due to polar material oscillating at before gathering in ring interacting companion expected to merge with central object ( bournaud combes 2003 ) scenario proposed by whitmore mcelroy schweizer 1987 for eso 415-g26 constraints on age system off - centre position of ring distinctive structural features surrounding main body rg few dynamical periods old still developing scenario gas accreted from companion onto rg ruled out no clear sign relating ring accreted companion 1 or 2 gyr after ring formed accreted matter few stars falls into polar orbit stellar halo around polar ring not expected contrary to merging scenario understanding in prg formation requires deep photometric study of at redshifts redshift of rg between high-@xmath7 prg - candidates and low-@xmath7 prgs by whitmore et 1990observed prgs bridge distance gap rare difficult to detect by large surveys sdss brosch et detecting tidal features merger remnants requires surface brightness sensitivity rg excellent example polar ring complex stellar halo structure detected by sky survey images photometric study of candidate prg in sdf object complex structure central host @xmath0 gyr old early - type galaxy exponential bulge nearly edge - on disc faint off - centre narrow ring at @xmath1 angle to polar axis low surface brightness stellar halo extending asymmetrically both sides two disrupted stellar debris in intermediate angles between ring polar axis two outer loops on both sides direction stellar debris morphological study suggests galaxy belongs to group bulge - dominated prgs ring formed around host during merger with gas - rich dwarf companion event less than 1 - 2 gyr ago captured material not fully dispersed into halo settled into polar structure authors recognize cultural role reverence summit of mauna kea within indigenous hawaiian community fortunate conduct observations from mountainpublication two micron sky survey joint project university massachusetts infrared processing analysis center california institute technology funded national aeronautics space administration national science foundation funding sdss sdss - ii provided alfred p. sloan foundation participating institutions national science foundation u.s department of energy national aeronautics space administration japanese monbukagakusho max planck society higher education funding council england sdss web site http://www.sdss.org managed astrophysical research consortium participating institutions american museum natural history astrophysical institute potsdam university basel cambridge western reserve chicago drexel university fermilab advanced study japan participation group johns hopkins university joint institute nuclear astrophysics kavli institute particle astrophysics korean scientist group chinese academy sciences los alamos national laboratory new mexico state university ohio state university pittsburgh portsmouth princeton university united states naval observatory university washington bekki k 1998 apj , 499 , 635 bournaud combes 2003 401 817 pustilnik2010 mnras 401 2067 bruzual charlot s 2003 344 1000 iodice arnaboldi sparke gallagher freeman&a 391 , 103 iodice sparke freeman 2002b 391 117 kashikawa n. 2004 56 1011 kent s 1985 apjs 59 , 115 kormendy j djorgovski s. 1989 ara&a 27 , 235 lavery r remijan charmandaris v hayes r ring 2004 apj 612 , 679 martnez - delgado d. 2010 140 , 962 ouchi m. 2004 611 , reshetnikov 1997 321 , 749 dettmar 2007 33 , 222 sotnikova 1997&a 325 , 933 schweizer f whitmore b rubin v 1983 88 , 909 smirnova moiseev afanasiev v 2010 mnras 408 , 400 tal dokkum nelan j bezanson r2009 138 1417 yagi kashikawa sekiguchi doi yasuda shimasaku okamura 2002 123 , 66 whitmore b 1991 warped disks inclined rings galaxies casertano sackett f briggs new cambridge univ press 60 whitmore lucas r mcelroy steiman - cameron sackett olling r.p 1990 100 , 1489 whitmore mcelroy d.b schweizer f 1987 314 , 439 | 0.436843 | 2,779 | 1,371 | , Saving $0.1 in GPT-4. |
155,097 | we briefly discuss our survey of extremely red objects ( eros ) with hst / nicmos and deep ground - based images and summarize its results .
we use the high - resolution nicmos images to sub - divide eros into morphological classes , tentatively arguing that about 60 % of eros ( 18 out of 30 ) are early - type galaxies .
about 20 % ( 6 out of 30 objects ) appear to be flattened disk systems .
another 10 % ( 3 out of 30 ) have an irregular morphology .
thus , combining disk and irregular systems , our data are compatible with 30 % of eros being dusty starbursts . a small fraction ( 3 out of 30 )
appear point - like even on the hst images and could therefore be either compact high redshift galaxies or stars in the milky way halo .
# 1_#1 _ # 1_#1 _ = # 1 1.25 in .125 in .25 in | the advent of high quality near - infrared detectors has allowed a veritable breakthrough in the study of extremely red objects ( hereafter eros ) , _ i.e. _ galaxies characterized by @xmath0 colors of 5 or redder .
clearly , a natural interpretation of their red colors is that eros are elliptical galaxies with relatively old stellar populations at a redshift between 1 and 2 .
this has been confirmed by spinrad et al .
( 1997 ) for the ero with regular morphology lbds 53w091 .
however , graham & dey ( 1996 ) have shown that another ero , hr10 , actually has a spectrum with emission lines .
cimatti et al .
( 1998 ) definitely confirmed the dusty nature of hr10 .
clearly , on the basis of colors alone it is not possible to discriminate between a dusty starburst or an old stellar population at @xmath1 . on the other hand , spectroscopy of a large sample of eros
is , unfortunately , prohibitive even with 8-meter class telescope ( _ e.g. _ cimatti et al .
2000 ) .
one important difference between lbds 53w091 and hr 10 is their morphology . at hst resolution ,
the former appears regular while the latter has an irregular morphology .
this has led us to adopt as a working hypothesis the assumption that morphology with hst can help identify the nature of eros .
therefore , we have carried out a survey of eros in fields where hst / nicmos images were available . in section 2
, we describe our survey and some basic results .
section 3 is devoted to a discussion of morphologies .
section 4 sums up .
we have carried out a survey of eros over an area of 13.74 arcmin@xmath2 using hst / nicmos f160w images , for a total of 23 fields .
we have complemented the near - ir data with either wfpc2 f606w images or ground - based r - band images ( see treu & stiavelli 1999 , hereafter ts99 , for more details ) .
we built a number of bruzual & charlot ( 1993 , and 1996 unpublished ) models to assess the range of colors displayed by a variety of star formation histories .
we focussed on two star formation histories that may be relevant for high - redshift elliptical galaxies : _ i ) _ a stellar population 1 gyr old at any given redshift , and _
ii ) _ a stellar population formed as a single burst at z=10 .
eros were selected according to two color criteria determined on the basis of these models .
the first , broader , criterion is aimed at selecting elliptical galaxy candidates at @xmath3 and defines our basic ero sample .
its color criteria are r - h@xmath4 and v@xmath5-h@xmath6 ( here and in the following we will use magnitudes in the ab system . ) the second criterion selects candidate elliptical galaxies at @xmath7 and corresponds to r - h@xmath8 and v@xmath5-h@xmath9 .
we have identified a total of 30 eros .
the completeness of our data set was verified by carrying out a large set of montecarlo simulations showing that at the faintest level we are still more than 90 per cent complete .
the luminosity function that we derive extends down to h@xmath10 ( only for a fraction of the field , see ts99 for a detailed list of the area covered at each depth . ) at this depth we detect @xmath11 eros arcmin@xmath12 .
each ero was morphological classified by visual inspection independently by the two authors . in general
there was a good agreement between the two independent classifications by the authors but in four cases there was a discrepancy .
these cases were looked in more detail until a common classification could be agreed upon .
the final breakdown of morphologies are given in table 1 .
those objects classified as early - type galaxies ( e / s0 ) were then used to derive a luminosity function to be compared with the one observed at low redshift under the assumption of passive evolution .
the number density of _ regular _ eros in our survey at the limit of h@xmath10 is @xmath13 arcmin@xmath12 .
the derived luminosity function is compared to the observed one in figure 1 .
the number of red galaxies that we find requires the early formation of at least some early - type galaxies .
however , our data do not support the formation at high - redshift of all early - type galaxies , _ i.e. _ we see a deficit of red galaxies .
each panel in the figure indicates the fraction of early - type that could have formed at the given redshift .
this fraction ranges from 15 to 66 per cent depending on cosmology and formation redshift ( ts99 ) .
this discrepancy could be interpreted in a number of ways : _ i ) _ all ellipticals are already in place at high redshift but we are only seeing those that have not been polluted by younger populations .
in fact , only a very modest fraction of young stars would be sufficient to make an elliptical appear bluer than our selection criteria ; _ ii ) _ only some of the early - type galaxies were in place at @xmath14 .
as an example , lenticular galaxies could have formed later while giant ellipticals were already present at @xmath15 ; _ iii ) _ our error bars are underestimated due to the clustering of eros seen , _
e.g. _ by daddi et al .
( 2000 ) , and our results are thus compatible with all early - type galaxies forming at high redshift .
clearly it is difficult at this stage to determine which of these alternatives ( or their combinations ) is correct .
however , our data strongly suggest that at least some elliptical galaxies were already in place at relatively high redshift .
in the course of our morphological analysis we have binned eros in four classes : _
i ) _ e / s0 , characterized by regular morphology , _ ii ) _ disk , characterized by high apparent flattening , _ iii ) _ irregular , and _
iv ) _ point - like .
the point - like objects could be red stars in the milky way halo , small elliptical galaxies , or compact regions embedded in low surface brightness objects .
the number of point - like objects that we detect ( 3 ) is well within the limits expected for brown dwarfs in the galactic halo ( _ e.g. _ flynn et al .
1996 ) . in the analysis of the luminosity function of early - type galaxies we have included only those objects belonging to the e / s0 category .
clearly there could be a contamination introduced by face - on disk eros .
however , the e / s0 eros tend to be characterized by a r@xmath16 light profile ( _ e.g. _ stiavelli et al .
1999 ) ; this is in contrast to the low redshift luminous disk galaxies none of which has an r@xmath16 light profile .
this argues in favor of little or no contamination of the e / s0 sample from disk systems .
this also implies that face - on disk systems would have not been classified as e / s0 eros . in general , these face - on disks
would likely be observed as irregular eros or even drop out from the eros sample if , _
e.g. _ , the red colors of edge - on disk eros are due to dust absorption through the line of sight of the disk . lc class & number + e / s0 & 18 + disks & 6 + irregular & 3 + point like & 3 +
our results suggest that some early - type galaxies were in place at high redshift .
we detect a deficit of red objects which may imply either that some early - type galaxies formed later , or that observational effects or interactions interve to make the colors bluer and thus cause some early - type galaxies to drop from the e / s0 ero sample .
we find a number density of about 0.5 arcmin@xmath12 ( at h@xmath17 ) of eros that appear to be as flattened as a disk .
they could be massive disks already in place at @xmath1 .
we expect these objects as well as the irregular eros to be characterized by a dusty starburst spectrum .
the major limitation of our study is the relatively small area of the survey especially in light of the large field - to - field fluctuations discovered by daddi et al .
( 2000 ) for the bright - end of the ero distribution . for this reason
, we are planning to increase the area of our survey by a factor four in the near future . | - infrared detectors breakthrough in study extremely red objects eros galaxies @xmath0 colors 5 or redder eros are elliptical galaxies with old stellar populations at redshift between 1 and 2 confirmed by spinrad et al 1997 for lbds 53w091 graham & dey 1996 ) another ero hr10 has spectrum with emission lines cimatti et al . 1998 confirmed dusty nature of hr10 colors discriminate between dusty starburst or old stellar population at @xmath1 spectroscopy of large sample eros prohibitive even with 8-meter class telescope difference between lbds 53w091 and hr 10 morphology at hst resolution former appears regular latter irregular morphology hst identify nature eros survey of eros in hst / nicmos images available section 2 survey results section 3 morphologies section 4 survey of eros over area 13.74 arcmin@xmath2 using hst / nicmos f160w images 23 fields complemented data with wfpc2 f606w images or ground - based r - band images see treu & stiavelli 1999 for detailsbuilt bruzual & charlot 1993 1996 models assess colors star formation histories focussed on two histories relevant for high - redshift elliptical galaxies : stellar population 1 gyr old at redshift stellar population single burst at z=10 eros selected two color criteria first criterion elliptical galaxy candidates at @xmath3 defines basic ero sample color criteria are r - h@xmath4 v@xmath5-h@xmath6 magnitudes second criterion selects elliptical galaxies at @xmath7 corresponds to r - h@xmath8 and v@xmath5-h@xmath9 identified 30 eros completeness data set verified by montecarlo simulations faintest level more than 90 per cent complete luminosity function extends to h@xmath10 see ts99 @xmath11 eros arcmin@xmath12 each ero morphological classified by inspection independently by two authors good agreement in four cases discrepancy cases looked in until common classification agreed final breakdown of morphologies in table 1 objects classified as early - type galaxies ( s0 ) used luminosity function compared with low redshift passive evolutionnumber density of regular eros in survey at limit h@xmath10 is @xmath13 arcmin@xmath12 derived luminosity function compared to observed in figure 1 number red galaxies requires early formation early - type galaxies data support formation at high - redshift all early - type galaxies deficit of red galaxies each panel indicates fraction early - type formed at redshift ranges 15 to 66 per cent depending on cosmology formation redshift discrepancy interpreted all ellipticals in place at high redshift only seeing not polluted by younger populations modest fraction of young stars make elliptical bluer some early - type galaxies in place at @xmath14 lenticular galaxies could have formed later giant ellipticals present at @xmath15 error bars underestimated due to clustering of eros results compatible with all early - type galaxies forming at high redshift difficult to determine data suggest some elliptical galaxies in place at high redshiftanalysis binned eros in four classes e / s0 regular morphology ii disk high flattening iii irregular iv point - like point - like objects red stars milky way halo small elliptical galaxies compact regions in low surface brightness objects number point - like objects ( 3 ) within limits for brown dwarfs in galactic halo analysis luminosity function of early - type galaxies included objects e / s0 category contamination by face - on disk eros e / s0 eros characterized r@xmath16 light profile contrast to low redshift luminous disk galaxies r@xmath16 light profile little or no contamination e / s0 sample from disk systems implies face - on disk systems not classified as e / s0 eros face - disks likely irregular eros drop out from eros sample red colors edge - disk eros due to dust absorption class e / s0 18 disks 6 irregular 3 point like results suggest some early - type galaxies at high redshiftdetect deficit of red objects imply early - type galaxies formed later or observational effects colors bluer early galaxies drop from e / s0 ero sample find number density 0.5 arcmin@xmath12 h@xmath17 ) eros flattened as disk could be massive disks at @xmath1 expect objects irregular eros characterized dusty starburst spectrum limitation small area survey large field - to - field fluctuations by daddi et al . 2000 bright - end ero distribution planning increase area survey factor four | 0.31517 | 1,867 | 943 | , Saving $0.1 in GPT-4. |
155,105 | hot surface ignition and subsequent flame propagation of premixed n - hexane air mixtures are shown in this fluid dynamics video .
high speed schlieren photography revealed 3 distinct behaviors of ignition and propagation as a function of mixture composition and initial pressure . | experiments were performed in a closed 2 liter vessel , filled to a precision of 0.1 torr using the method of partial pressures .
the video shows schlieren images from 3 fuel rich mixtures of n - hexane in air , 3.4% , 4.74% , and 6.23% , which elucidate the three different behaviors : single flame propagation , multiple flames , and puffing flames .
the first mode involves a single flame propagating until it reaches the vessel walls as shown in the first sequence . in the second mode ,
two to three flames ignite sequentially , as shown in the second schlieren video .
the final mode corresponds to a continuously puffing flame as seen throughout the duration of the video ( about six seconds ) , of which a portion is shown here .
all three experiments were performed at 101kpa initial pressure and the mixtures ignite when the glow plug reaches around 900 k. | experiments performed in closed 2 liter vessel filled 0.1 torr partial pressures video shows images from 3 fuel mixtures n - hexane in air 3.4% 4.74% 6.23% three behaviors single flame propagation multiple flames puffing flames first mode single flame until vessel walls second mode two to three flames ignite sequentially final mode continuously puffing flame all three experiments performed at 101kpa initial pressure mixtures ignite when glow plug reaches 900 k. | 0.069353 | 199 | 102 | , Saving $0.0 in GPT-4. |
155,108 | the _ chandra _ image of the central region of the coma cluster reveals that both its dominant galaxies , ngc4874 and ngc4889 , retain the central parts of their x - ray gas coronae .
the interstellar gas with a temperature of 12 is confined by the hot intergalactic medium of the coma cluster into compact clouds ( only 3 kpc in radius ) containing @xmath0 of gas . the physical state of the gas in these clouds appears to be determined by a delicate balance between radiative cooling and suppressed ( by a factor of 30100 ) heat conduction through the interface between these clouds and the hot cluster gas . | _ rosat _ observations showed that the coma cluster , once thought to be an archetype of a relaxed system , contains substructures on a wide range of linear scales ( briel , henry & bhringer 1992 , white , briel & henry 1993 , vikhlinin , forman & jones 1997 ) .
the detected structures include 100 kpc - scale x - ray enhancements around the two dominant cluster galaxies , ngc4874 and ngc4889 ( vikhlinin , forman , & jones 1994 ) .
recent xmm observations confirmed these findings and also revealed unresolved x - ray sources coincident with these galaxies ( arnaud et al .
2001 , briel et al .
2001 ) . in this _ letter _
, we resolve these sources with _ chandra_. both sources are extended arcsec - scale ( @xmath1 kpc ) remnants of galactic x - ray coronae with temperatures 12 kev , compressed by the hot , 9 kev , gas of the coma cluster .
their very existence poses interesting physical questions .
the problem of thermal evaporation of cold clouds embedded in a hot medium has been considered in a number of papers ( cowie & mckee 1977 , mckee & cowie 1977 , balbus & mckee 1982 , mckee & begelman 1990 , bandiera & chen 1994 among others ) .
this problem is relevant for the survival of cold clouds in multi - phase cluster cooling flows ( e.g. , bhringer & fabian 1989 , fabian , canizares , & bhringer 1994 ) .
the general consensus emerging from these studies , as summarized by fabian ( 1994 ) , is that heat conduction must be suppressed by one to two orders of magnitude below the spitzer value , otherwise the radiative cooling of the gas will be overcome by heat conduction .
however , the direct evidence for suppression of heat conductivity in the intracluster medium ( icm ) is scarce ( one of a few examples can be found in bechtold et al . 1983 ) .
the _ chandra
_ observations of the interstellar medium in the coma galaxies presented here allow a direct study of cold clouds embedded in a hot intracluster medium .
we use @xmath2 km s@xmath3 kpc@xmath3 , which implies a linear scale of 0.68 kpc per arcsec at the coma distance ( @xmath4mpc ) .
the coma cluster was observed with acis in the fall of 1999 in a series of 6 pointings with individual exposures of @xmath5 ksec .
the pointing position was kept the same during these observations , with ngc4874 and ngc4874 located at off - axis distances of 2.8 and 4.0 , respectively .
the detector was moved in the focal plane between the pointings , so that the galaxies were observed by both acis - i ( 4 times ) and acis - s ( 2 times ) .
there were no background flares in the coma observations .
in one pointing , ngc4874 was located in the gap between the acis - i chips .
the total exposure was 48 ksec for ngc4874 and 56 ksec for ngc4889 .
the central part of the composite acis - i image is shown in fig .
[ fig : big ] .
notice the faint arcmin - scale x - ray enhancements around ngc4874 and 4889 , previously detected by _
rosat_. at the center of these extended structures there are bright ( 500 counts ) compact sources coincident with the optical positions of the galaxies .
the observed 0.52 kev flux is @xmath6 erg s@xmath3 @xmath7 for ngc4874 and @xmath8 erg s@xmath3 @xmath7 for ngc4889 , corresponding to luminosities in the same band of @xmath9 and @xmath10 erg s@xmath3 .
the enlarged x - ray and optical images of ngc4874 and 4889 are shown in fig .
[ fig : panels ] .
the x - ray sources associated with the galaxies are extended , but unlike most ellipticals , the x - ray emission is much more compact than the distribution of optical light .
below we discuss these interesting objects in detail .
the spectra of the ngc4874 and 4889 sources were extracted within a radius of 7@xmath11 ( which encompasses most of the observed x - ray flux ) .
since we are interested in the intrinsic properties of the galaxies , we collected the background spectra in the 15@xmath1130@xmath11 annuli around them , so it includes the true background as well as the emission of the coma cluster .
the source , background spectra , and the response files were generated individually for each pointing .
we then co - added the observed spectra and averaged the response files for each source . in the spectra for both galaxies ( fig .
[ fig : panels ] ) , one immediately notices strong emission lines near 1 kev a key signature of a 1 kev - temperature plasma .
indeed , the spectra are well fit by the mewe - kaastra - liedahl model in xspec with the absorption fixed at the galactic value @xmath12 @xmath7 .
therefore , the observed x - ray emission is most likely produced by warm interstellar gas
. the x - ray emission can not be due to the integrated emission of stars because its spatial distribution does not follow the optical light ( fig .
[ fig : panels ] and [ fig : profile:4874 ] ) .
any significant contribution by low - mass x - ray binaries ( lmxb ) can be ruled out because it also should follow the starlight and have a very different ( hard ) spectrum ( sarazin , irwin , & bregman 2001 ) .
the best - fit temperatures and metal abundances of these sources are listed in table [ tab : spectra ] ( the uncertainties hereafter are at the 68% confidence level for one interesting parameter ) .
the average temperature is 1.0 kev for ngc4874 and 1.8 kev for ngc4889 .
the average abundance of heavy elements is close to the solar value in both galaxies .
this is significantly higher than the metal abundance in the surrounding coma gas ( @xmath13 , arnaud et al .
2001 ) .
_ chandra _ s angular resolution permits spectral fitting in several concentric annuli in ngc4874 .
this analysis reveals a marginally significant ( at the 99% confidence level ) decrease of the gas temperature towards the center of the galaxy ( fig .
[ fig : tprofile:4874 ] and table [ tab : spectra ] ) .
ngc4889 does not show any non - isothermality , but the statistical errors are much larger . [
cols="<,<,<,^",options="header " , ] @xmath14 fixed .
the results of the surface brightness profile modeling are summarized in table [ tab : profiles ] .
we were unable to obtain meaningful constraints on @xmath15 because of the parameter correlation and finite psf .
generally , all @xmath16 are allowed .
the cutoff and core - radii are well - constrained for @xmath17 . for larger @xmath15 ,
the cutoff radius becomes much larger than the core - radius and hence has almost no effect on the observed profile .
for a pure @xmath15-model without a cutoff , @xmath15 must be in excess of @xmath18 , mimicking the cutoff at @xmath19 , which is consistent with the best - fit @xmath20 .
of course , different models correspond to very similar absolute gas density distributions in the radial range of interest .
it is interesting to compare the pressure of the ism at the best - fit cutoff radius with the external pressure of the cluster gas .
even though the three - dimensional location of the galaxies within the cluster is uncertain , the icm density in their vicinity can be determined rather accurately because each galaxy has an associated arcmin - scale x - ray enhancement ( vikhlinin et al .
1994 ) : @xmath21 @xmath22 around ngc4874 and @xmath23 @xmath22 around ngc4889 .
the temperature of the cluster gas around both galaxies is @xmath24 kev ( see above ) .
remarkably , we find that for the observed ism parameters , the pressure ratio at @xmath25 is @xmath26 for ngc4874 and @xmath27 for ngc4889 .
therefore , the pressure of the ism in both galaxies equals the external pressure at a radius where the _ chandra _ image independently indicates an ism density cutoff .
this is a clear indication that the distribution of the ism is affected by the external pressure .
the derived central ism densities are higher by a factor of 210 than those normally found in isolated elliptical galaxies ( @xmath28 @xmath22 ; forman , jones & tucker 1985 ) , which also suggests compression by an external pressure .
stellar ejecta is the most likely source of interstellar gas since the metal abundance in the ism of both galaxies is significantly higher than that in the coma icm .
the observed gas mass ( @xmath29 and @xmath30 in ngc4874 and ngc4889 , respectively ) would be ejected by stars over the time @xmath31 , where @xmath32 is the stellar mass loss rate per unit optical luminosity . using the generally adapted value @xmath33 yr@xmath3
@xmath34 ( faber & gallagher 1976 ) and the optical luminosity within @xmath20 ( faber et al .
1997 ) , we find @xmath35 yr in both galaxies .
therefore , stellar ejecta easily produce the observed ism over the galaxy lifetime .
the ism is subject to radiative cooling , thermal evaporation due to the surrounding hot gas , heating by stellar winds and supernovae , and possibly to ram pressure stripping .
_ ram pressure stripping_. the observed x - ray centroids are within 0.5 kpc of the optical centers ( limited by the _
chandra _ aspect accuracy ) .
this and the azimuthally symmetric x - ray morphology probably exclude any significant ongoing stripping .
stripping will be unimportant if the galaxies move together with their surrounding 100 kpc scale gas density enhancements .
_ radiative cooling_. the average cooling time of the ism is @xmath36 . for the observed ism parameters , we find @xmath37 yr for ngc4874 and @xmath38 yr for ngc4889 , with a @xmath39 uncertainty .
_ stellar winds and supernovae_. after @xmath40 years past the burst of star formation , the average temperature of ejecta from normal stars and supernovae is @xmath41 kev ( david , forman , & jones 1990 ) .
therefore , stellar mass loss can not be a source of heating for the ism with @xmath42 kev .
_ thermal evaporation_. following cowie & mckee ( 1977 ) , one can show that classical heat conduction through the ism - icm boundaries of our galaxies should be saturated . from their equation
( 64 ) , we find the saturated evaporation timescales @xmath43 yr for ngc4874 and @xmath44 for ngc4889 , with a 20% formal statistical uncertainty .
the timescales for radiative cooling and classical evaporation are both very short and the ism would be destroyed quickly by either mechanism , unless the radiative cooling and the heat influx that drives evaporation balance each other .
this would be achieved if the conductivity through the icm - ism interface is a factor of @xmath45 below the classical value .
a plausible reason for such a suppression is a disjoint magnetic structure of the ism and icm .
if heat influx from the icm does balance radiative cooling of the ism , one expects that at each radius , the x - ray volume emissivity of the ism equals the divergence of the heat flux ( i.e. , the local heating rate ) .
interestingly , this can be verified for ngc4874 , where we observe a temperature gradient and therefore can estimate the heat flux .
assuming that conductivity in the ism is classical , the heat flux is @xmath46 erg @xmath7 s@xmath3 where @xmath47 is the plasma temperature in degrees ( mckee & cowie 1977 ) .
for the observed radial dependence of temperature in ngc4874 which can be modeled as @xmath48 $ ] ( fig .
[ fig : tprofile:4874 ] ) , the divergence of @xmath49 is indeed within a factor of 1.5 of the observed x - ray volume emissivity . for ngc4889 ,
the uncertainty in the temperature profile is too large to compute the heat flux .
however , we can say that its ism temperature is too high to be plausibly maintained by internal heat sources such as stellar mass loss and supernovae , so there should be external heating , most likely due to conduction through the ism boundary .
_ chandra _ observations show that the dominant coma galaxies retain a fraction of their x - ray coronae in the form of compact ( @xmath50 kpc in size ) clouds with temperatures of @xmath51 and @xmath52 kev which are in the pressure equilibrium with the hot intracluster gas .
time scales for radiative cooling and classical evaporation are very short but the ism observed in two galaxies is unlikely to be a transient phenomenon .
we propose that cooling in the ism is balanced by heat influx through the cluster / ism interface .
for this , thermal conductivity through the interface should be 30100 times below the classical value , perhaps because of the disjoint magnetic field structure .
furthermore , in ngc4874 , the observed radial temperature gradient implies that if the conductivity _ inside _ the ism cloud is classical , then heat from outside the cloud is distributed through the cloud volume in such a way that it precisely balances radiative cooling at each radius .
this leaves little or no energy for the ism evaporation , preserving the ism clouds for a sufficiently long time . | rosat observations showed coma cluster relaxed system contains substructures on linear scales ( briel bhringer 1992 1993 1997 detected structures include 100 kpc - scale x - ray enhancements around dominant cluster galaxies ngc4874 and ngc4889 1994 recent xmm observations confirmed revealed unresolved x - ray sources coincident with galaxies ( arnaud 2001 briel et 2001 ) resolve sources with chandra_ both sources are extended arcsec - scale remnants of galactic x - ray coronae with temperatures 12 kev compressed by hot 9 kev gas coma cluster existence poses physical questions problem of thermal evaporation of cold clouds in hot medium considered in 1977 1982 1994 problem relevant for survival of cold clouds in multi - phase cluster cooling flows 1989 1994 consensus heat conduction must be suppressed by to two orders of magnitude below spitzer value otherwise radiative cooling gas overcome by heat conduction direct evidence for suppression of heat conductivity in intracluster medium scarce in bechtold et al . 1983 )chandra _ observations interstellar coma galaxies allow study cold clouds hot intracluster medium use @xmath2 km s@xmath3 kpc scale 0.68 kpc per arcsec coma distance @xmath4mpc coma cluster observed with acis fall 1999 6 pointings exposures @xmath5 ksec pointing position same ngc4874 ngc4874 off - axis distances 2.8 4.0 detector moved between pointings galaxies observed acis - i 4 acis - s 2 no background flares ngc4874 gap between acis - i chips total exposure 48 ksec ngc4874 56 ksec ngc4889 central composite acis - i image fig . faint arcmin - scale x - ray enhancements around ngc4874 4889 bright 500 compact sources optical positions galaxies observed 0.52 kev flux @xmath6 erg s@xmath3 @xmath7 ngc4874 @xmath8 erg ngc4889 luminosities same band @xmath9 @xmath10 erg enlarged x - ray optical images ngc4874 4889 fig .x - ray sources galaxies extended unlike ellipticals emission compact than distribution optical light discuss objects spectra of ngc4874 4889 sources extracted within radius 7@xmath11 most observed x - ray flux collected background spectra in 15@xmath1130@xmath11 annuli around includes true background emission coma cluster source background spectra response files generated individually for each pointing co - added observed spectra averaged response files each source spectra both galaxies strong emission lines near 1 kev signature 1 - temperature plasma spectra well fit by mewe - kaastra - liedahl model in xspec absorption fixed at galactic value @xmath12 @xmath7 observed x - ray emission likely produced by warm interstellar gas emission not due to integrated emission stars spatial distribution follow optical light contribution by low - mass x - ray binaries ruled out starlight different spectrum best - fit temperatures metal abundances of sources listed in table [ spectra uncertainties at 68% confidence level for parameter average temperature 1.0 kev for ngc4874 1.8 kev for ngc4889average abundance heavy elements close to solar value in both galaxies higher than metal abundance in surrounding coma gas @xmath13 arnaud 2001 chandra s angular resolution permits spectral fitting in concentric annuli in ngc4874 analysis reveals significant decrease gas temperature towards center galaxy fig table ngc4889 non - isothermality statistical errors larger @xmath14 fixed results surface brightness profile modeling summarized in table profiles constraints on @xmath15 parameter correlation finite psf all @xmath16 allowed cutoff core - radii - constrained for @xmath17 larger @xmath15 cutoff radius larger than core no effect on profile pure @xmath15-model without cutoff @xmath15 must excess of @xmath18 cutoff at @xmath19 consistent with best - fit @xmath20 different models correspond similar gas density distributions radial range compare pressure ism at best - fit cutoff radius with external pressure cluster gas location galaxies uncertain icm density vicinity determined each galaxy x - ray enhancement vikhlinin et1994 @xmath21 around ngc4874 @xmath23 around ngc4889 temperature cluster gas around galaxies is @xmath24 kev pressure ratio at @xmath25 is for ngc4874 for ngc4889 pressure ism in galaxies equals external pressure at radius chandra image indicates ism density cutoff distribution ism affected by external pressure central ism densities higher by factor 210 isolated elliptical galaxies suggests compression by external pressure stellar ejecta likely source of interstellar gas metal abundance in ism galaxies higher than coma icm observed gas mass @xmath29 @xmath30 in ngc4874 ngc4889 ejected by stars over @xmath31 @xmath32 is stellar mass loss rate per unit optical luminosity value @xmath33 yr@xmath3 @xmath34 optical luminosity @xmath20 @xmath35 yr in both galaxies stellar ejecta produce observed ism over galaxy lifetime ism subject to radiative cooling thermal evaporation heating by stellar winds supernovae possibly ram pressure strippingobserved x - ray centroids within 0.5 kpc optical centers limited by chandra aspect accuracy symmetric x morphology exclude significant stripping stripping unimportant if galaxies move together with 100 kpc gas density enhancements radiative cooling average cooling time ism @xmath36 @xmath37 yr for ngc4874 @xmath38 yr for ngc4889 @xmath39 uncertainty stellar winds supernovae after @xmath40 years star formation average temperature ejecta from normal stars supernovae @xmath41 kev stellar mass loss source heating for ism with @xmath42 kev thermal evaporation classical heat conduction through ism - icm boundaries should be saturated saturated evaporation timescales @xmath43 yr for ngc4874 @xmath44 for ngc4889 20% statistical uncertainty timescales for radiative cooling classical evaporation short ism destroyed quickly unless balance if conductivity through icm - ism interface @xmath45 below classical value reason disjoint magnetic structureheat influx from icm radiative cooling ism each radius x - ray volume emissivity equals divergence heat flux local heating rate verified for ngc4874 temperature gradient heat flux conductivity classical heat flux is @xmath46 erg @xmath7 s@xmath3 @xmath47 plasma temperature degrees 1977 radial dependence temperature in ngc4874 modeled as @xmath48 divergence of @xmath49 within factor of 1.5 of x - ray volume emissivity ngc4889 uncertainty in temperature profile too large to compute heat flux ism temperature too high by internal heat sources external heating due to conduction through ism boundary dominant coma galaxies retain x - ray coronae compact @xmath50 clouds with temperatures of @xmath51 and @xmath52 kev in pressure equilibrium with hot intracluster gas time scales for radiative cooling classical evaporation short ism in two galaxies unlikely transient cooling balanced by heat influx through cluster / ism interface thermal conductivity through interface 30100 times below classical value disjoint magnetic field structurein ngc4874 radial temperature gradient implies conductivity inside ism cloud classical heat outside cloud distributed through volume balances radiative cooling each radius leaves little energy for ism evaporation preserving ism clouds long time | 0.497946 | 3,152 | 1,565 | , Saving $0.1 in GPT-4. |
155,109 | lyman break analogs ( lbas ) , characterized by high far - uv luminosities and surface brightnesses as detected by galex , are intensely star - forming galaxies in the low - redshift universe ( @xmath0 ) , with star formation rates reaching up to 50 times that of the milky way .
these objects present metallicities , morphologies and other physical properties similar to higher redshift lyman break galaxies ( lbgs ) , motivating the detailed study of lbas as local laboratories of this high - redshift galaxy population .
we present results from our recent integral - field spectroscopy survey of lbas with keck / osiris , which shows that these galaxies have the same nebular gas kinematic properties as high - redshift lbgs .
we argue that such kinematic studies alone are not an appropriate diagnostic to rule out merger events as the trigger for the observed starburst .
comparison between the kinematic analysis and morphological indices from hst imaging illustrates the difficulties of properly identifying ( minor or major ) merger events , with no clear correlation between the results using either of the two methods .
artificial redshifting of our data indicates that this problem becomes even worse at high redshift due to surface brightness dimming and resolution loss .
whether mergers could generate the observed kinematic properties is strongly dependent on gas fractions in these galaxies .
we present preliminary results of a carma survey for lbas and discuss the implications of the inferred molecular gas masses for formation models . | the paradigm for the formation of galaxies has changed dramatically over the course of the last decade .
new integral field units ( ifus ) have allowed the detailed study of star - forming galaxies at @xmath1 , when the universe was young and forming the bulk of its current stellar mass .
these studies ( law et al . 2009 , frster - schreiber et al . 2009 ) have revealed new details about the properties of the nebular gas in starburst galaxies , showing that these objects , unlike local star - forming spirals , present strong random dynamical components , with velocity dispersion reaching values well above 100 km s@xmath2 . however , the interpretation of these data is difficult , mainly because of the loss of surface brightness due to cosmic dimming and the loss of spatial resolution .
therefore , the importance of a comparison sample at low redshift , one with similar physical properties but less subject to observational biases , becomes evident . in an attempt to find such a sample , heckman et al .
( 2005 ) and hoopes et al .
( 2007 ) have selected the galaxies in the low redshift universe ( @xmath0 ) with highest values of fuv luminosity and surface brightness ( @xmath3 , @xmath4 kpc@xmath5 ) , based on galex data and with thresholds determined as to mimic the properties of typical lyman break galaxies ( lbgs ) at @xmath6 . even though these were the only selection criteria , subsequent studies have shown that these galaxies strongly resemble lbgs in most of their physical properties , including metallicity , radio , morphology , extinction and outflow properties ( hoopes et al .
2007 , basu - zych et al .
2007 , overzier et al .
2009 , overzier et al .
2010 , overzier et al .
2011 , heckman et al .
therefore , we refer to this sample as the lyman break analogs ( lbas ) . in this work we describe our ifu observations of the emission line gas kinematics in lbas , comparing our results to those found for high - redshift star - forming galaxies .
these results , along with a detailed description of the observations , are presented in full in basu - zych et al .
( 2009 ) and gonalves et al .
we have observed 19 lbas with the osiris instrument at keck .
the compact sizes and strong hydrogen line emission of these galaxies are ideal for this purpose , since we can achieve high signal - to - noise ratios and still fit within the small field of view of the instrument . by using laser guide star adaptive optics , we are able to resolve approximately 200 pc in each galaxy , which represents the diffraction limit of keck at these redshifts .
we have also artificially redshifted our data to @xmath1 , to allow for a direct comparison with the results from the aforementioned ifu studies at high redshift selected from the literature .
the first important result is that the gas kinematics in lbas is similar to that found in the star - forming galaxies at high - redshift .
we also find velocity dispersion values above 100 km s@xmath2 in some cases , which generate low @xmath7 values , that is , the ratio between velocity shear and velocity dispersion . in figure 1 , we show these values as a function of stellar mass .
intrinsic values are slightly higher than typical ratios found at high redshift ( black triangles ) , but once we redshift our data , we lose information at the outer radii , where the gas is most offset from the central velocity .
the result is that @xmath7 appears smaller ( red and blue symbols ) , with values similar to those found in law et al .
( 2009 ) and frster - schreiber et al .
( 2009 ) ( hollow symbols ) .
we also show that intrinsic values depend on stellar mass , with more massive galaxies presenting higher @xmath7 ratios , indicating a stronger rotational dynamical component , as observed also at @xmath6 .
furthermore , we have analyzed the asymmetry of the kinematic maps , as described by shapiro et al .
( 2008 ) using vlt / sinfoni data .
this kinemetry technique consists of decomposing the velocity and velocity maps into fourier components in order to compare the purely symmetrical component with higher - order ones .
shapiro et al . (
2008 ) have used this technique to differentiate between rotating disks ( low asymmetry ) and mergers ( high asymmetry ) .
we find that galaxies with higher stellar masses present lower asymmetry values , reinforcing the idea of more massive galaxies more closely resembling local spirals .
in addition , we have shown that the loss of spatial resolution at high redshift makes galaxies appear smoother , rendering asymmetry values smaller ; in this way , 50% of our galaxies would be wrongly classified as relatively undistorted disks at higher redshift .
this is shown in figure 1 , where we compare intrinsic asymmetry values with those determined for the artificially redshifted observations .
approximately half the galaxies fall inside the gray - shaded area , indicating galaxies that would be classified as `` mergers '' at low - z but as `` disks '' at high - z .
ratios for starburst galaxies .
black triangles indicate intrinsic values for our lba data , while downward blue and red triangles represent the artificially redshifted data .
hollow symbols indicate actual high - redshift data from law et al .
( 2009 ) and frster - schreiber et al .
the dashed line is a fit to the intrinsic lba data , evidencing the stellar mass dependence .
_ right _ : comparison of kinematic asymmetry indices between intrinsic ( @xmath8-axis ) and artificially redshifted data ( @xmath9-axis ) .
dashed lines represent the threshold between mergers and disks as in shapiro et al .
the gray shaded area shows galaxies classified as `` mergers '' at low redshift but as `` disks '' at high @xmath10 .
most points lie below the solid line , indicating that galaxies look less asymmetric at high redshift .
figures from gonalves et al .
( 2010).,title="fig:",width=220 ] ratios for starburst galaxies .
black triangles indicate intrinsic values for our lba data , while downward blue and red triangles represent the artificially redshifted data .
hollow symbols indicate actual high - redshift data from law et al .
( 2009 ) and frster - schreiber et al .
the dashed line is a fit to the intrinsic lba data , evidencing the stellar mass dependence . _
right _ : comparison of kinematic asymmetry indices between intrinsic ( @xmath8-axis ) and artificially redshifted data ( @xmath9-axis ) .
dashed lines represent the threshold between mergers and disks as in shapiro et al .
the gray shaded area shows galaxies classified as `` mergers '' at low redshift but as `` disks '' at high @xmath10 .
most points lie below the solid line , indicating that galaxies look less asymmetric at high redshift .
figures from gonalves et al .
( 2010).,title="fig:",width=220 ] particularly striking examples of these effects are shown in gonalves et al .
( 2010 , see figure 10 of that work ) , where we show a case having the expected kinematic profile of a rotating disk , while careful examination of the hst @xmath11-band morphology reveals clear signs of a recent merger event as evidenced by clumpy starburst , tidal tails , and an asymmetric outer isophote .
another example is shown in which evidence for a rotating disk only appears after redshifting the lba to @xmath1 , while no such evidence exists in the full resolution , full sensitivity data .
in addition to these technical problems , there are also problems related to the timescale of merger events . for example
, robertson & bullock ( 2008 ) have shown an example of a gas - rich merger remnant rapidly cooling and forming a rotating disk that would be difficult to identify as such through ifu observations .
the amount of cold molecular gas in the reservoir available for star formation is a crucial ingredient in understanding the starbursts in star forming galaxies , and in particular whether a merger could generate a rotating disk in such short timescales . with this in mind , we have initiated a survey searching for co(1 - 0 ) emission in lbas .
we have thus far been awarded 92 hours of time with the carma array .
we present preliminary results in the following paragraphs . out of three objects targeted
, we have two detections and one upper limit , after approximately 4 hours of integration in each case . as further evidence for the analogy between our sample and high redshfit lbgs , we present in figure 2 the gas mass fraction of galaxies as a function of stellar mass .
we note that the expected gas fractions in lbas from the schmidt - kennicutt relation are very similar to expected values for lbgs .
moreover , these expectations are confirmed by actual carma measurements : gas fractions vary between @xmath12% at intermediate stellar masses to @xmath13% at the high - mass end .
, hence the range in @xmath14 values .
_ right _ : the schmidt - kennicutt relation for different galaxy samples .
lbas ( blue circles ) follow the relation for local spirals . for more details on all samples , see daddi et al .
( 2010).,title="fig:",width=230 ] , hence the range in @xmath14 values .
_ right _ : the schmidt - kennicutt relation for different galaxy samples
. lbas ( blue circles ) follow the relation for local spirals . for more details on all samples , see daddi et al .
( 2010).,title="fig:",width=230 ] another interesting finding is that lbas appear to follow the local schmidt - kennicutt relation , similar to what is found for bzk galaxies at @xmath1 by daddi et al .
( 2010 ) , only at higher surface densities .
this reinforces the idea of a bimodal schmidt - kennicutt relation , since our sample is distinct from more infrared bright galaxies such as ulirgs and smgs , which present a higher star formation surface density for a given surface density of gas ( see daddi et al .
2010 and references therein ) . with strong co(1 - 0 ) line emission ( with fluxes up to @xmath15 jy
km s@xmath2 ) , lbas will be ideal candidates for further studies with alma once the instrument comes online .
not only will detections be possible within minutes , but we will be able to measure the surface density of gas at a detail level only achievable for local galaxies so far
. we will thus be able to measure the spatially resolved schmidt - kennicutt relation at scales below 100 pc , providing invaluable insight into the interplay between the gas reservoir and star formation in starburst galaxies . | paradigm for formation of galaxies changed last decade new integral field units allowed detailed study of star - forming galaxies at @xmath1 universe young forming current stellar mass studies ( law et 2009 frster - schreiber 2009 revealed details properties nebular gas in starburst galaxies present strong random dynamical components velocity dispersion above 100 km s@xmath2 interpretation data difficult loss of surface brightness due to cosmic dimming spatial resolution importance of comparison sample at low redshift similar physical properties less subject observational biases evident heckman et al . 2005 hoopes et al. ( 2007 ) selected galaxies low redshift universe @xmath0 with highest values fuv luminosity surface brightness ( @xmath3 @xmath4@xmath5 ) based on galex data mimic properties typical lyman break galaxies at @xmath6 studies shown galaxies resemble lbgs in physical properties including metallicity radio morphology extinction outflow properties ( refer to this sample as lyman break analogs ( lbas ) work describe ifu observations emission line gas kinematics in lbas comparing high - redshift star - forming galaxies results detailed presented in basu - zych et al . 2009 gonalves et al observed 19 lbas with osiris instrument at keck compact sizes strong hydrogen line emission ideal achieve high signal - to - noise ratios fit small field of view laser guide star adaptive optics resolve 200 pc each galaxy diffraction limit of keck at redshifts artificially redshifted data to @xmath1 direct comparison with ifu studies high redshift gas kinematics in lbas similar to star - forming galaxies at high - redshift velocity dispersion values above 100 km s@xmath2 generate low @xmath7 values velocity figure 1 values function of stellar mass intrinsic values higher than ratios high redshift redshift lose information at outer radii gas velocity @xmath7 appears smaller values similar to law et al . 2009 frster - schreiber et al . ( 2009intrinsic values depend on stellar mass massive galaxies higher @xmath7 ratios stronger rotational dynamical component analyzed asymmetry kinematic maps shapiro et al . 2008 using vlt / sinfoni data velocity maps symmetrical with higher rotating disks mergers high galaxies higher stellar masses present lower asymmetry values massive galaxies resembling local spirals loss of spatial resolution at high redshift makes galaxies appear smoother asymmetry values smaller 50% galaxies wrongly classified as undistorted disks at higher redshift shown in figure 1 intrinsic asymmetry values with artificially redshifted observations half galaxies fall inside gray - shaded area as mergers at low - z disks at high - z starburst galaxies black triangles indicate intrinsic values data blue red triangles artificially redshifted data hollow symbols indicate high - redshift data from law et al 2009 frster - schreiber et al dashed line to intrinsic lba data stellar mass dependencecomparison kinematic asymmetry indices intrinsic artificially redshifted data-axis dashed lines represent threshold between mergers disks gray shaded area shows galaxies mergers low disks high points below solid line galaxies less asymmetric high redshift figures from gonalves et al . ratios starburst galaxies black triangles indicate intrinsic values lba data blue red triangles artificially redshifted data hollow symbols indicate high - redshift data law et 2009 frster dashed line fit intrinsic lba data stellar mass dependence comparison kinematic asymmetry indices intrinsic artificially redshifted data-axis dashed lines threshold mergers disks gray shaded area shows galaxies mergers low redshift disks at high points below solid line galaxies less asymmetric at high redshift figures gonalves et al . striking examples effects gonalves et al 2010 figure 10 case expected kinematic profile rotating disk hst @xmath11-band morphology reveals signs recent merger event clumpy starburst tidal tails asymmetric outer isophote example evidence rotating disk after redshifting lba to @xmath1 no evidence full resolution sensitivity data technical problems problems timescale merger events robertson & bullock 2008 ) gas - rich merger remnant cooling forming rotating disk difficult identify ifu observations cold molecular gas reservoir star formation crucial understanding starbursts merger generate rotating disk short timescales initiated survey for co(1 - 0 ) emission in lbas awarded 92 hours time carma array preliminary results three objects targeted two detections one upper limit after 4 hours integration each analogy sample high redshfit lbgs figure 2 gas mass fraction of galaxies function stellar mass expected gas fractions in lbas schmidt - kennicutt relation similar to lbgs expectations confirmed by carma measurements gas fractions vary between @xmath12% intermediate stellar masses to @xmath13% high - mass end range in @xmath14 valuesright schmidt - kennicutt relation galaxy samples lbas blue circles follow local spirals details see daddi et al . 2010) range @xmath14 values right schmidt - kennicutt relation galaxy samples lbas blue circles follow local spirals details see daddi et al . lbas follow local schmidt - kennicutt relation similar bzk galaxies @xmath1 daddi et al 2010 higher surface densities reinforces bimodal schmidt - kennicutt relation our sample distinct from infrared bright galaxies ulirgs smgs higher star formation surface density daddi al 2010 strong co(1 - 0 ) line emission fluxes up @xmath15 s@xmath2 lbas ideal candidates studies alma instrument online detections possible within minutes measure surface density gas detail level local galaxies measure spatially resolved schmidt - kennicutt relation scales below 100 pc insight interplay gas reservoir star formation starburst galaxies | 0.35575 | 2,368 | 1,172 | , Saving $0.1 in GPT-4. |
155,110 | we consider any dynamical system that starts from a given ensemble of configurations and evolves in time until the system reaches a certain fixed stopping criterion , with the mean first - passage time the quantity of interest .
we present a general method , projective dynamics , which maps the multi - dimensional dynamics of the system onto an arbitrary discrete set of states @xmath0 , subject only to the constraint that the dynamics is restricted to transitions not further than the neighboring states @xmath1 .
we prove that with this imposed condition there exists a master equation with nearest - neighbor coupling with the same mean first - passage time as the original dynamical system .
we show applications of the method for brownian motion of particles in one and two dimensional potential energy landscapes and the folding process of small bio - polymers .
we compare results for the mean first passage time and the mean folding time obtained with the projective dynamics method with those obtained by a direct measurement , and where possible with a semi - analytical solution . | the mean first - passage time ( mfpt ) is the time required for an ensemble of identical systems starting from a given initial distribution of configurations to reach a fixed criterion . being able to obtain
the mfpt is important in a wide array of physical applications .
these include the mfpt when a star of a certain mass goes supernova , to neuron firing dynamics , spreading of diseases , chemical kinetics , folding processes of polymers , the death of living creatures , and the decay of metastable states via nucleation and growth processes @xcite .
the projective dynamics ( pd ) method to calculate mfpt for systems with a discrete state space @xcite mapped the mfpt problem for metastable decay of magnetic states associated with ising - type ( discrete spin states ) models onto a nearest - neighbor coupled master equation .
however , the pd method has not previously been shown to be valid for other dynamical systems , for example for systems with continuous state spaces @xcite .
the calculation of the mfpt for systems with long time scales has previously required that the dynamic system satisfy certain conditions .
for example , for the milestoning method @xcite to work the system must be close to equilibrium or have separation of time scales or committer surfaces .
we prove in this letter that the pd method has no such constraints , and hence it is generally applicable to obtain the mfpt of _ any _ dynamical system .
we prove that for any mfpt problem ( with a finite mfpt ) there exists a discrete master equation with nearest - neighbor coupling that has the same mfpt as the original system .
this master equation has time - independent rates , even if the underlying system is non - markovian .
every point in phase space in the original system must be mapped onto a state in the discrete master equation .
we show that _ any _ mapping that leads to only nearest - neighbor coupling in the master equation preserves the mfpt provided the growing ( @xmath2 ) and shrinking ( @xmath3 ) rates between all master equation states @xmath4 are correct .
therefore , whether the @xmath2 and @xmath3 are known theoretically or are measured , the correct mfpt is obtained . for metastable decay in the ferromagnetic @xmath5@xmath6@xmath7 ising case such knowledge enabled calculations of mfpt of about @xmath8 monte carlo steps per spin @xcite .
it is anticipated that the knowledge that the pd method gives the correct mfpt will enable long - time - scale simulations for other physical models .
an example of where the pd method may be useful for long - time simulations is studies of the intercalation time of li ions in molecular dynamics studies modeling charging of batteries @xcite .
consider any ensemble of systems evolving in phase space @xmath9 under some dynamic .
they start with some given distribution , and we desire to calculate the mfpt until each system stops after it meets some fixed stopping criterion .
the dynamics of the ensemble can be mapped onto a set of discrete non - overlapping states @xmath0 , with @xmath10@xmath11@xmath4@xmath11@xmath12 , which cover the entire domain .
we impose the restriction that the states @xmath13 can make transitions only to the states @xmath1 .
this condition can be met for many systems by adjusting the ` length ' @xmath14 of the states , so that no transitions within a single time step @xmath15 further than to the neighboring states exist .
the states that form the stopping citerion will all be mapped to @xmath16 .
the time - evolution of the system can be described by a master equation with nearest neighbor coupling , which describes the change of occupation @xmath17 of state @xmath4 by the probability flows between adjacent states .
namely , @xmath18 p_k(t)+s_{k-1}p_{k-1}(t ) \mbox { , } \label{master1}\ ] ] where the growing @xmath2 ( shrinking @xmath3 ) rates represent the directional rates at which the system transits from state @xmath4 to state @xmath19 ( @xmath20 ) .
the initial conditions @xmath21 are specified by the ensemble of initial conditions of the original system .
let @xmath22 denote the number of times the system makes a transition from state @xmath4 to its neighboring states ( @xmath23 ) in the time interval @xmath24 $ ] and @xmath25 the number of times the system resides in state @xmath4 at time @xmath26 .
then the growing rate for the time interval @xmath27 $ ] can be obtained by measuring the system at time intervals @xmath15 .
explicitly @xmath28 a similar expression exists for the shrinking rate @xmath3 . assuming that the ensemble of systems has been fully absorbed at @xmath29 after starting at @xmath30
, we write in shorthand the master equation @xmath31 where the elements of @xmath32 are the growing and shrinking rates , which have been measured from @xmath30 up to some point in time when all systems in the ensemble are fully absorbed . by using the fundamental relation @xcite @xmath33 the general expression for the mfpt @xmath34 for a system starting fully in state @xmath35
can be obtained as @xmath36 \label{001}\ ] ] where @xmath35 denotes the starting state and @xmath12 the total number of states within the domain .
we prove that eq .
( [ 001 ] ) gives the same mfpt independent of the particular choice of the states .
for this purpose , let @xmath37 be the joined interval of state @xmath4 and @xmath20 , then for a single time step it follows that the transitions from state @xmath38 to @xmath19 ( @xmath39 ) in the new system are equal to the transitions from state @xmath4 to @xmath19 ( @xmath20 to @xmath39 ) in the old system namely @xmath40 and @xmath41 , whereas the probability of occupation of state @xmath38 rescales as @xmath42/\sum_k n_k(t)$ ] .
thus the growing rate at time @xmath26 in the old and new system are related as @xmath43 and similarly for the new shrinking rate @xmath44 . that this procedure leaves the mfpt unchanged
can be understood by noting the effect of joining two states on the master equation .
the joining of adjacent states corresponds to the addition of terms of the master equation of the same adjacent states , namely @xmath18 p_k(t)+s_{k-1}p_{k-1}(t ) \mbox{. } \label{master2}\ ] ] and @xmath45 p_{k+1}(t)+s_{k}p_k(t ) \mbox{. } \label{master3}\ ] ] it is easily seen that the cross terms vanish and we are left with an expression of the form @xmath46 using ( [ gres ] ) and similarly for @xmath44 in ( [ masterend ] ) for a single time step , leads to @xmath47 \tilde{p}_{m}(t)+s_{k-1}p_{k-1}(t ) \mbox{.}\ ] ] hence the general form [ compare eq .
( [ master1 ] ) and ( [ master2 ] ) ] is restored in the reduced system of equations . performing this joining repeatedly ,
finally only one state is left .
this gives the same mfpt as the original ensemble of systems @xcite .
-coordinate of some fixed length @xmath48 . ]
to illustrate the theoretical framework , we demonstrate diffusion in both dimensions @xmath5@xmath6@xmath49 and @xmath5@xmath6@xmath50 , as well as the folding process of a model for a small linear polymer chain .
we choose these examples to specifically address the independence of the method from any existence or knowledge of a free energy landscape or most probable escape pathway . as a @xmath5@xmath6@xmath49 historical example
, we studied the diffusion process of particles subjected to a kramers or to a rough @xcite potential ( fig .
[ kramer ] ) .
the trajectories were integrated along 5120 individual paths starting at @xmath51@xmath6@xmath52@xmath53 and terminated once they leave the domain @xmath54 $ ] at the absorbing boundary @xmath55 .
the mfpts were calculated using eq .
( [ 001 ] ) .
in order to show the freedom in choosing the length of the intervals , we show examples of bins with different lengths @xmath48 .
starting at the absorbing boundary @xmath4@xmath6@xmath10 for @xmath51@xmath56@xmath10 , the system resides in state k when @xmath57 . in both cases ( smooth and rough ) , we chose @xmath58 [ fig .
( [ kramer ] ) ] .
table [ tab2a ] shows the results obtained with the pd method are not changed with different interval lengths . a comparison with a direct measurement ( on the underlying dynamics ) and the semi - analytical solution further verifies that the mfpts obtained with the pd method gives reliable results ( provided that sufficient statistics are obtained to get accurate estimates of the growing and shrinking rates ) .
note that to illustrate the method we chose examples of the starting position being at the boundary between two states ( @xmath59 ) , in the midrange ( @xmath60 ) , and away from the midrange ( @xmath61 ) of the starting state .
the results do not change with any of these starting positions , which further indicates the independence of the pd method . [
cols="<,<,<,<,<,<,<",options="header " , ] as a @xmath5@xmath6@xmath50 example we applied the pd method to the diffusion process at the entropic barrier @xmath62 of ref .
we chose the coordinate @xmath63 in two different ways .
the trajectories are started at @xmath64@xmath6@xmath65 and terminated once they reach @xmath51@xmath66@xmath10 .
we obtained results for defining the states both with the coordinate @xmath63@xmath6@xmath51 and @xmath63@xmath6@xmath67 [ fig .
( [ entropic ] ) ] .
the first represents a binning along the x - coordinate , interpreted as the progress towards the absorbing boundary , while the circular binning has no such interpretation . for @xmath63@xmath6@xmath51 the states were @xmath68@xmath11@xmath51@xmath69@xmath70 , where we chose @xmath71@xmath6@xmath72 and @xmath73 . for @xmath63@xmath6@xmath67 , the states are defined as @xmath74@xmath11@xmath67@xmath75 for @xmath76 , where @xmath67 denotes the distance from @xmath64 .
we chose @xmath77@xmath6@xmath78 and @xmath79 .
for the circular binning , the bin @xmath4@xmath6@xmath49 is comprised of all points with @xmath51@xmath11@xmath10 outside the circular annulii .
the results in table [ tab2a ] showing independence of the mfpt illustrate the strength of the pd method .
( representing a progress toward the absorbing boundary ) and @xmath80 ( representing circular intervals ) of some fixed interval length @xmath81 . ] as a polymeric example , we applied the pd method to the folding process of a model of a small linear polymer chain having the chemical structure @xmath82 , where @xmath83 stands for the hydrophobic and @xmath84 for the polar monomers .
the structure of the individual chain is maintained by three interatomic potentials : a bond - stretching potential , a bond - angle bending potential , and a rejecting potential avoiding an overlap of @xmath85 or @xmath86 monomers .
the chain undergoes a folding process from the fully elongated state to a compact near native state caused by attractive lennard - jones type of interactions between @xmath87 monomers .
the pd method is given by choosing the coordinate @xmath63 as the potential energy of the native contacts ( the @xmath87 interaction energy ) .
although this choice seems apparent , it does not correspond to an actual dynamical coordinate . starting with the fully elongated chain , for different temperatures of the external bath
, we simulated the folding process due to the brownian motion .
we compared the mfpts of the chain of reaching a potential energy deep enough in the basin of attraction of the energy ground state ( i.e. the native state ) so that no escape ( i.e. unfolding ) from there is likely to occur .
the results of the pd method as well as a direct measurement are shown in table [ tab2a ] .
for all simulated temperatures the pd method gives the same results ( within statistics ) as a direct measurement .
the projective dynamics ( pd ) method , as a tool for projecting multi - dimensional systems onto one set of states @xmath0 and mapping the time - evolution onto a discrete master equation with nearest neighbor coupling , was shown to be independent of the choice in the states @xmath13 and the ` length ' of the intervals which define the states .
provided only transitions between adjacent states occur , the pd method correctly obtains mfpts .
this result makes the pd method generally applicable to the dynamics of any system ( discrete or continuous , markovian or non - markovian , stochastic or deterministic ) .
subject only to the nearest - neighbor coupling constraint , the states @xmath0 can be chosen in multiple ways , without changing the mean first - passage time ( mfpt ) . in many cases of interest ,
the average occupation probability for every state @xmath13 is also the same as for the original dynamical system .
the calculation of the growing and shrinking rates required in the pd method may for some models and dynamics be obtained by using other methods , such as calculations of histories @xcite or forward flux sampling @xcite or accelerated molecular dynamics @xcite .
useful discussions with jerzy blawzdziewicz , weinan e , ron elber , miroslav kolesik , cory ohern , per arne rikvold , and eric vanden - eijnden are gratefully acknowledged , as well as useful interactions at a 2008 summer school at the aspen center for physics .
this research was supported in part by the national science foundation through teragrid resources provided by the texas advanced computing center ( tacc ) under grant number tg - dmr090092 .
m.a . novotny , ` a tutorial on advanced dynamic monte carlo methods for systems with discrete state spaces ' , in _ annual reviews of computational physics ix _ , ed .
d. stauffer ( world scientific , singapore 2001 ) , pp . | mean first - passage time ( mfpt ) is for identical systems to reach fixed criterion mfpt important in physical applications include star goes supernova neuron firing dynamics spreading diseases chemical kinetics folding processes polymers death creatures decay of metastable states via nucleation growth projective dynamics pd ) method mfpt for discrete state space mapped mfpt problem for metastable decay magnetic states onto nearest - neighbor coupled master equation pd method not valid for other dynamical systems continuous state spaces calculation mfpt for systems long time scales required conditions system close to equilibrium separation of time scales committer surfaces pd method has no constraints applicable to mfpt of any dynamical system for any mfpt problem discrete master equation with nearest - coupling same mfpt as original system equation has time - independent rates even if system non - markovian every point in phase space in original system must be mapped onto state discrete master equationmapping to nearest - neighbor coupling preserves mfpt growing and shrinking rates between states correct @xmath2 known or measured correct mfpt obtained metastable decay in ferromagnetic knowledge enabled calculations mfpt of @xmath8 monte carlo steps per spin knowledge pd method gives correct mfpt long - time simulations for physical models pd method useful for simulations studies intercalation time of li ions in molecular dynamics modeling charging batteries consider ensemble of systems evolving in phase space start with distribution calculate mfpt until system stops after fixed stopping criterion dynamics mapped onto discrete non - overlapping states @xmath0 cover entire domain restriction states @xmath13 make transitions only to states @xmath1 met by adjusting length @xmath14 of states no transitions time step further than to neighboring states states stopping citerion mapped to @xmath16 time - evolution system described by master equation with nearest neighbor coupling change of occupation of state @xmath4 by probability flows between adjacent states@xmath18 p_k(t)+s_{k-1}p(t ) growing shrinking @xmath3 ) rates represent from state @xmath4 to @xmath19 initial conditions @xmath21 specified by original system @xmath22 from @xmath4 to states @xmath23 time interval @xmath24 @xmath25 in state @xmath4 @xmath26 growing rate for time interval @xmath27 $ obtained measuring system at time intervals @xmath15 @xmath28 similar expression for shrinking rate @xmath3 fully absorbed at @xmath29 after @xmath30 master equation @xmath31 elements of @xmath32 are growing shrinking rates measured from @xmath30 to systems absorbed fundamental relation @xcite @xmath33 general expression for mfpt @xmath34 for system starting in state @xmath35 obtained as @xmath36 @xmath35 starting state @xmath12 total number of states within domain eq . gives same mfpt independent of choice states@xmath37 joined interval of state @xmath4 @xmath20 single time step transitions from state @xmath38 to @xmath19 new system equal to transitions @xmath4 to old system @xmath40 and @xmath41 probability of occupation of state @xmath38 rescales as @xmath42/\sum_k n_k(t)$ growing rate @xmath26 old new system related as @xmath43 new shrinking rate @xmath44 procedure leaves mfpt unchanged effect of joining two states on master equation joining of adjacent states corresponds to addition of terms of master equation states @xmath18 p_k(t)+s_{k-1}p_{k-1}(t and @xmath45 p_{k+1}(t)+s_{k}p_k cross terms vanish expression of form @xmath46 using [ gres ] @xmath44 single time step leads to @xmath47(t)+s_{k-1}p_{k-1}(t general form restored in reduced system of equations joining repeatedly one state left gives same mfpt as original ensemble systems @xcite-coordinate fixed length @xmath48 illustrate theoretical framework demonstrate diffusion both dimensions @xmath5@xmath49 folding process model small linear polymer chain examples address independence method from free energy landscape escape pathway historical example studied diffusion process particles subjected kramers rough @xcite potential fig . kramer trajectories integrated along 5120 paths starting @xmath51@xmath6@xmath52@xmath53 terminated domain @xmath54 absorbing boundary @xmath55 mfpts calculated using eq . ( [ 001 ] ) freedom choosing length intervals show examples bins different lengths @xmath48 starting absorbing boundary @xmath4@xmath6@xmath10 @xmath51 system resides in state k @xmath57 both cases chose @xmath58 fig kramer table tab2a shows results pd method not changed different interval lengths comparison direct measurement semi - analytical solution verifies mfpts pd method reliable results sufficient statistics accurate estimates growing shrinking ratesmethod chose starting position boundary between states @xmath59 midrange @xmath60 away from midrange results change starting positions independence pd method @xmath5 example applied pd method diffusion process at entropic barrier @xmath62 chose coordinate @xmath63 trajectories started at @xmath64@xmath6@xmath65 terminated @xmath51@xmath66@xmath10 obtained results defining states coordinate @xmath63@xmath6@xmath51@xmath67 first represents binning along x - coordinate progress towards absorbing boundary circular binning no interpretation @xmath63@xmath6@xmath51 states @xmath68@xmath11@xmath51@xmath69@xmath70 chose @xmath71@xmath6@xmath72 @xmath73 @xmath63@xmath67 states defined as @xmath74@xmath11@xmath67@xmath75 distance from @xmath64 chose @xmath77@xmath6@xmath78 @xmath79 circular binning bin @xmath4@xmath6@xmath49 points with @xmath51@xmath11@xmath10 outside circular annuliiresults in table tab2a ] independence mfpt illustrate strength pd method progress toward absorbing boundary @xmath80 circular intervals fixed interval length @xmath81 applied pd method to folding process small linear polymer chain chemical structure @xmath82 @xmath83 hydrophobic @xmath84 polar monomers structure maintained by three interatomic potentials bond - stretching bond - angle bending rejecting potential avoiding overlap @xmath85 or @xmath86 monomers chain folding from fully elongated to compact near native state by interactions between @xmath87 monomers pd method coordinate @xmath63 as potential energy of native contacts @xmath87 interaction energy correspond to actual dynamical coordinate different temperatures simulated folding process due to brownian motion compared mfpts chain reaching potential energy in basin attraction energy ground state native state no escape results pd method direct measurement shown in table [ tab2a ] all simulated temperatures pd method gives same results as direct measurementprojective dynamics method projecting multi - dimensional systems states @xmath0 mapping time - evolution discrete master equation nearest neighbor coupling independent of choice states @xmath13 intervals transitions between adjacent states obtains mfpts applicable to dynamics any system discrete markovian non - markovian stochastic deterministic subject to nearest - neighbor coupling constraint states @xmath0 chosen multiple ways without changing mean first - passage time mfpt average occupation probability for state @xmath13 same as original system calculation growing shrinking rates may obtained other methods forward flux sampling accelerated molecular dynamics discussions with jerzy blawzdziewicz weinan ron elber miroslav kolesik cory ohern arne rikvold eric vanden - eijnden interactions at 2008 summer school aspen center for research supported by national science foundation teragrid resources texas advanced computing center under grant number tg - dmr090092 novotny tutorial on advanced dynamic monte carlo methods for systems with discrete state spaces annual reviews of computational physics ix stauffer | 0.503107 | 3,256 | 1,649 | , Saving $0.1 in GPT-4. |
155,115 | it is well documented that sdw in ( tmtsf)@xmath0pf@xmath1 undergoes another phase transition at @xmath2k , though the nature of the new low temperature phase is controversial .
we have shown recently that the new phase is well described in terms of unconventional sdw ( usdw ) which modifies the quasiparticle spectrum dramatically . in this paper
we show that the same model describes consistently the hall resistivity observed in ( tmtsf)@xmath0pf@xmath1 . , ,
, transport measurements , magnetotransport , organic superconductors | since the discovery of superconductivity in ( tmtsf)@xmath0pf@xmath1 in 1979 @xcite , the bechgaard salts or the highly anisotropic organic superconductors ( tmtsf)@xmath0x ( where tmtsf is tetramethyltetraselenfulvalene and x is anion pf@xmath3 , asf@xmath3 , clo@xmath3 ) are one of the most well studied systems @xcite . the quasi - one - dimensionality ( 1d ) is a consequence of the crystal structure , where the tmtsf molecules are stacked in columns in the @xmath4 direction ( along which the highest conductivity occurs ) , and the resulting anisotropy in conductivity is commonly taken to be @xmath5 .
the rich phase diagram of ( tmtsf)@xmath0x salts exhibits various low temperature phases under pressure and/or in magnetic field , among which the spin density wave ( sdw ) , field induced sdw ( fisdw ) with quantum hall effect and spin triplet superconductivity are very intriguing @xcite .
( tmtsf)@xmath0pf@xmath1 is metallic down to @xmath6k , where the transition into the semiconducting sdw takes place .
it is known that sdw in ( tmtsf)@xmath0pf@xmath1 undergoes another transition at @xmath7 ( at 3.54@xmath8k at ambient pressure ) @xcite .
the indication of the subphase was first seen by nmr @xcite , where @xmath9 diverges and the spin susceptibility changes at @xmath10 .
the transition at @xmath10 is preserved through the entire @xmath11 phase diagram .
furthermore a calorimetric transition at 3.5@xmath8k with a large hysteretic phenomenons in the temperature range 2.54@xmath8k ( caused by the sample history ) has been observed and interpreted as an indication of a glass transition @xcite . on the other hand , the low frequency dielectric relaxation of sdw in
( tmtsf)@xmath0pf@xmath1 did not show the existence of the glass transition @xcite .
since then the sdw state was widely investigated , but the nature of the subphase remained controversial .
recently we have studied the @xmath12 axis magnetoresistance ( mr ) of ( tmtsf)@xmath0pf@xmath1 at ambient pressure and with magnetic field rotated within the @xmath13plane .
the mr has different behaviour for @xmath14k and @xmath15k @xcite .
for @xmath14k mr is described in terms of the quasiparticle in a magnetic field , where the imperfect nesting term @xcite plays the crucial role . however , in order to describe mr below 4@xmath8k we have introduced a rather artificial scattering term .
more recently , unconventional density waves ( ucdw or usdw ) have been proposed as a possible ground state in electronic systems in organic conductors and heavy fermions @xcite . unlike the conventional dw , the udw is defined as the dw where the order parameter @xmath16 depends on the quasi - particle momentum @xmath17 . in particular , ucdw appears to describe the striking angular dependent magnetoresistance ( admr ) found in the low temperature phase of @xmath18(et)@xmath0khg(scn)@xmath3 @xcite . on the other hand
, we have shown that the remarkable features of admr in ( tmtsf)@xmath0pf@xmath1 below @xmath19k ( the decrease in the quasiparticle energy gap for @xmath20 and the sudden change in the angular dependence of the energy gap in the presence of magnetic field @xmath21 as the temperature crosses @xmath22 ) can be described within the model sdw plus usdw using the usdw order parameter @xmath23 , where @xmath24 with @xmath25 @xcite . in this paper we shall present the hall resistivity data in ( tmtsf)@xmath0pf@xmath1 for @xmath14k and @xmath15k and discuss them within the model of sdw+usdw .
the hall resistivity @xmath26 in two crystals of ( tmtsf)@xmath0pf@xmath1 with dimension @xmath27mm@xmath28 and @xmath29mm@xmath28 was measured with 6 contact method as shown in inset of fig.[fig.fig1 ] . the results shown and discussed here were obtained on one of them , and similar qualitative behaviour was observed on another sample , too .
the measurements were performed between 2.0@xmath8k and 6.3@xmath8k , with magnetic field up to 9@xmath8 t .
the @xmath4 direction of the monocrystal is the highest conductivity direction , the intermediate conductivity @xmath12 is perpendicular to @xmath4 in the @xmath30 plane and the lowest conductivity @xmath31 direction is perpendicular to the @xmath30 ( and @xmath32 ) plane . versus magnetic field @xmath21 at several fixed temperatures ( @xmath15k ) .
inset : the contacts arrangement.,width=340 ] the current flow was along the @xmath4 axis , the magnetic field @xmath33 along @xmath31 direction and the hall voltage was detected along the @xmath12 axis .
the magnetic field dependence of the hall resistivity is shown in fig.[fig.fig1 ] ( @xmath15k ) and fig.[fig.fig2 ] ( @xmath14k ) . versus magnetic field @xmath21 at several fixed temperatures ( @xmath14k).,width=340 ] the general @xmath34 and @xmath21
dependencies are consistent with earlier data by uji _
et al . _
@xcite , although their main objective is the study of the rapid quantum oscillation .
as is readily seen from fig.[fig.fig1 ] and [ fig.fig2 ] , the negative hall resistivity is much smaller for @xmath14k than for @xmath15k . according to @xcite ,
the hall resistivity in the quasi one dimensional system is given by @xmath35 where @xmath36 is the conductivity tensor .
if we neglect the quantum hall effect , which contributes a new term in @xmath37 @xcite , we obtain @xmath38 where @xmath39 is the quasiparticle density and @xmath40 is the quasiparticle energy gap in the presence of magnetic field @xmath21 . on the other hand
, it is well known that @xmath41 has no activation form where @xmath42 is parallel to the @xmath43 axis @xcite .
a possible explanation is that the conductivity parallel to the @xmath43 axis has another channel of which quasiparticle has no energy gap .
a similar approach has been used in the quantitative analysis of @xmath44 in the fisdw state in ( tmtsf)@xmath0pf@xmath1 under high pressure @xcite .
therefore , the fitting of our hall data is done with @xmath45 where @xmath46 and @xmath47 are temperature dependent constants , and @xmath48 we took these expressions from @xcite , with @xmath49 and @xmath50 . here , we limit ourselves to the case @xmath51
. the results of the fitting procedure are shown on both figures as the dotted lines ( the values of @xmath46 and @xmath47 are given in table [ table ] ) .
we have an excellent agreement with the experimental data .
we note that @xmath46 is almost independent of temperature , while @xmath47 decreases as temperature decreases .
the values of @xmath52 s used in the present fitting ( @xmath53 and @xmath54 ) are somewhat smaller than the ones used earlier @xcite , but they are of the same order of magnitude . also , the exponent @xmath55 is somewhat strange ( naturally , we expect @xmath56 ) , but the similar exponent has been found in fitting the diagonal component of the magnetoresistance tensor @xcite .
.table with fitting constants .
[ cols="^,^,^",options="header " , ] the agreement between the model and experimental data implies that the appearance of usdw below @xmath22 , over the preexisting sdw , with a new quasiparticle energy gap appears to describe the hall resistivity consistently
. in particular , the rapid increase of the hall resistivity below @xmath22 testifies the rapid change in the quasiparticle energy gap across @xmath22 . in order to further test the present model the hall resistivity data with the magnetic field away from the @xmath31 axis are highly desirable
we have completed the study of the resistivity tensor in ( tmtsf)@xmath0pf@xmath1 below @xmath22 . for @xmath51
we have shown that an approach with usdw+sdw below @xmath10 gives an excellent fit of the hall resistivity data .
this further supports our proposal that usdw appearing on top of existing sdw in ( tmtsf)@xmath0pf@xmath1 below @xmath22 gives a consistent description of the resistivity tensor .
this experimental work was performed on samples prepared by k. bechgaard .
we acknowledge the participation of n. franceti in the experiment and useful discussions with b. dra , a. virosztek and s. tomi . | discovery superconductivity in tmtsf)@xmath0pf@xmath1 1979 bechgaard salts anisotropic organic superconductors tmtsf)@xmath0x tmtsf tetramethyltetraselenfulvalene x anion pf@xmath3 studied systems quasi - one - crystal structure tmtsf molecules stacked columns @xmath4 direction highest conductivity anisotropy conductivity @xmath5 phase diagram tmtsf@xmath0x salts exhibits low temperature phases under pressure magnetic field spin density wave field induced sdw quantum hall effect spin triplet superconductivity intriguing tmtsf)@xmath0pf@xmath1 metallic down to @xmath6k transition into semiconducting sdw sdw transition at @xmath7 3.54@xmath8k ambient pressure indication subphase seen by nmr @xcite @xmath9 diverges spin susceptibility changes at @xmath10 transition @xmath10 preserved @xmath11 phase diagram calorimetric transition at 3.5@xmath8k hysteretic phenomenons temperature range 2.54@xmath8k glass transition @xcitelow frequency dielectric relaxation of sdw in@xmath0pf@xmath1 show glass transition @xcite sdw state investigated subphase controversial studied @xmath12 axis magnetoresistance of at ambient pressure magnetic field rotated @xmath13plane mr different behaviour for @xmath14k @xmath15k @xcite @xmath14k mr quasiparticle in magnetic field imperfect nesting term @xcite mr below 4@xmath8k introduced artificial scattering term unconventional density waves ( ucdw or usdw proposed possible ground state in electronic systems organic conductors heavy fermions @xcite order parameter @xmath16 depends on quasi - particle momentum @xmath17 ucdw angular dependent magnetoresistance low temperature phase of @xmath18(et)@xmath0khg@xmath3 @xcite features admr in@xmath0pf below @xmath19k decrease quasiparticle energy gap change angular dependence energy gap magnetic field temperature crosses @xmath22 described model sdw plus usdw using usdw order parameter @xmath23 @xmath25 @xcitepresent hall resistivity data in@xmath0pf@xmath1 for @xmath14k @xmath15k discuss model sdw+usdw hall resistivity @xmath26 in two crystals @xmath27mm measured with 6 contact method fig. results obtained on one similar another measurements between 2.0@xmath8k and 6.3@xmath8k magnetic field up to 9@xmath8 t @xmath4 direction highest conductivity intermediate conductivity @xmath12 perpendicular to @xmath4 lowest conductivity @xmath31 perpendicular to @xmath30 plane versus magnetic field @xmath21 at fixed temperatures ( @xmath15k ) current flow along @xmath4 axis magnetic field @xmath33 @xmath31 direction hall voltage detected along @xmath12 axis magnetic field dependence hall resistivity shown in fig. @xmath14k versus magnetic field @xmath21 at fixed temperatures @xmath34 @xmath21 dependencies consistent with data by uji et al @xcite main objective study rapid quantum oscillation figfig2 negative hall resistivity smaller for @xmath14k than @xmath15k hall resistivity system by @xmath35 @xmath36 conductivity tensor neglect quantum hall effect @xmath37 obtain @xmath38 @xmath39 quasiparticle density @xmath40 quasiparticle energy gap in magnetic field @xmath21 @xmath41 no activation form @xmath42 parallel to @xmath43 axis explanation conductivity parallel to axis another channel no energy gap similar approach in analysis of @xmath44 fisdw state@xmath0pf@xmath1 under high pressure fitting hall data with @xmath45 @xmath46 @xmath47 temperature dependent constants @xmath48 expressions from @xcite @xmath49 @xmath50 limit to @xmath51 results fitting shown on figures as dotted lines values of @xmath46 @xmath47 in table agreement with experimental data @xmath46 independent of temperature @xmath47 decreases as temperature decreases values of @xmath52 s present fitting @xmath53 @xmath54 smaller than same order of magnitudeexponent @xmath55 strange expect @xmath56 similar exponent found diagonal component magnetoresistance tensor @xcite fitting constants agreement model experimental data implies usdw below @xmath22 over preexisting sdw new quasiparticle energy gap hall resistivity rapid increase hall resistivity below @xmath22 testifies rapid change quasiparticle energy gap test model hall resistivity data magnetic field away from @xmath31 axis desirable completed study resistivity tensor@xmath0pf@xmath1 below @xmath22 @xmath51 approach usdw+sdw below @xmath10 excellent fit hall resistivity data supports proposal usdw on top existing sdw below @xmath22 consistent description resistivity tensor experimental work performed samples k. bechgaard participation n. franceti discussions with b. dra a. virosztek s. tomi | 0.315012 | 2,119 | 1,110 | , Saving $0.1 in GPT-4. |
155,138 | gas velocity dispersion measures the amount of disordered motions of a rotating disk .
accurate estimates of this parameter are of the utmost importance because it is directly linked to disk stability and star formation .
a global measure of the gas velocity dispersion can be inferred from the width of the atomic hydrogen ( hi ) 21 cm line .
we explore how several systematic effects involved in the production of hi cubes affect the estimate of hi velocity dispersion .
we do so by comparing the hi velocity dispersion derived from different types of data cubes provided by the hi nearby galaxy survey ( things ) .
we find that residual - scaled cubes best recover the hi velocity dispersion , independent of the weighting scheme used and for a large range of signal - to - noise ratio . for hi observations where the dirty beam is substantially different from a gaussian ,
the velocity dispersion values are overestimated unless the cubes are cleaned close to ( e.g. , @xmath01.5 times ) the noise level . | the random motion of gas in galaxies can be traced from the width of the ubiquitous 21-cm line of atomic hydrogen ( hi ) .
the amount of the disordered motion , the gas velocity dispersion , provides valuable information on the physical properties of the gas such as its ability to collapse and form stars , the energy balance and the phase structure of the gas @xcite .
this parameter is usually determined by fitting hi profiles with an assumed model , e.g. , a single or a double gaussian function , a gauss - hermite polynomial , or a lorentzian @xcite . instead of assuming a model ,
the intensity - weighted standard deviation of the line - of - sight velocities has also been used as a measure of hi velocity dispersion ( e.g. , * ? ? ?
an accurate determination of the hi velocity dispersion is limited by the ability of the imaging process to recover the intrinsic shapes of the hi line profiles . in radio interferometry ,
the source intensity distribution is not completely sampled by the telescope array . therefore , to recover the missing values in the @xmath1 data and reconstruct the true image of the source , the telescope point spread function ( psf or dirty beam )
needs to be deconvolved from the observed ( dirty ) image . as will be explained in detail later , the recovered image from a deconvolution algorithm
is obtained by adding a deconvolved map of the model of the source brightness distribution ( derived down to a certain flux ) to a map containing the remaining emission , usually assumed to be noise ( the residual map ) .
the width of the hi velocity profiles means that typically they will be spread over several velocity channels .
the faint outer wings of these profiles can thus appear in adjacent channels but with intensities below the threshold for cleaning .
the residual image will then still contain faint residual emission corresponding to these wings . in this paper
, we investigate what the effect of this uncleaned , residual emission is on measurements of the velocity dispersion and find that it leads to an overestimate of this parameter .
here we focus on data obtained as part of the things survey @xcite , but our conclusions will apply to all hi observations where the dirty beam is substantially different from a gaussian , such as is the case with multi - configuration very large array ( vla ) observations . this includes recent surveys such as , local irregulars that trace luminosity extremes : the hi nearby galaxy survey ( little things ; * ? ? ?
* ) , vla survey of advanced camera for surveys nearby galaxy survey treasury galaxies ( vla - angst ; * ? ? ?
* ) , survey of hi in extremely low - mass dwarfs ( shield ; * ? ? ?
this study is also relevant for other arrays that produce a non - gaussian dirty beam , e.g. , the atacama large millimeter / submillimeter array ( alma ) , meaning the results could also be relevant for molecular line velocity dispersions . in section [ sec : background ] , we give background on the weighting of visibilities and cleaning processes . in section
[ sec : goal ] , we describe the analysis goal and method . in section [ sec:1 ] , we compare hi velocity dispersion from different types of data cubes . in section [ sec:2 ]
, we investigate the type of data cubes that should be used in hi velocity dispersion analysis . in section [ sec : discussion ] , we discuss our results and conclusions .
interferometric observations of radio astronomical sources provide noisy samples of the complex visibility function of the sources at discrete locations in the @xmath2 plane @xcite , where @xmath3 and @xmath4 are the baseline vectors projected onto a plane perpendicular to the source direction .
deconvolution algorithms attempt to recover the image information ( missing values ) in the un - sampled @xmath2 plane .
the inverse fourier transform of the sampled visibility function , @xmath5 , is called _ the dirty image _ , @xmath6 : @xmath7 where @xmath8 is a sampling function and @xmath9 is the visibility function . using the fourier convolution theorem , which states that the fourier transform of the product of two functions is the convolution of the fourier transform of the two functions , equation [ eq : num1 ] becomes : @xmath10 in the radio astronomy jargon ,
@xmath11 is called _ the dirty beam_. the true image is the inverse fourier transform of the visibility function , @xmath12 .
a deconvolution algorithm involves solving for the true image given the dirty beam and the dirty image . to get the dirty image from the sampled visibility functions , a fast fourier transform ( fft )
is usually preferable over the _
simple _ fourier transform ( ft ) due to its computational speed .
however , fft requires that the sample visibilities are on a regularly spaced grid ( cell ) ( e.g. , * ? ? ?
* ) . in the image reconstruction and gridding process , a weighting function
is usually applied to the sampled visibility function to , e.g. , maximise point source sensitivity or to minimise sidelobe levels .
if @xmath13 is a weighting function , then equation [ eq : num1 ] becomes : @xmath14 many forms of weighting functions have been developed for different science goals .
this paper discusses the effects of using different weighting schemes in the derivation of hi velocity dispersion .
natural , uniform and robust ( or briggs ) weightings are widely known weighting schemes in radio interferometry .
the natural weighting gives a weight of 1/@xmath15 on visibility @xmath16 , where @xmath15 is the noise variance of that visibility .
in most cases , the noise is roughly similar for all visibilities and thus each @xmath2 points gets approximately the same weight . due to the gridding , @xmath2 cells at shorter baselines
get more weight as the inner parts of the @xmath1 plane have a higher density of @xmath1 points ( i.e. , measured visibilities ) .
thus , natural weighting offers the best surface brightness sensitivity and minimum noise , but at the cost of a larger beam size . for uniform weighting , a given cell in the @xmath2 plane
gets a weight of 1/@xmath17 , where @xmath17 is the density of the @xmath2 points in the cell . given that the density of the @xmath2 points is usually lower at larger baselines , the latter are weighted more , resulting in better resolution but lower sensitivity ( short baselines are de - emphasized ) , and therefore an increased rms noise .
the uniform weighting gives better ( i.e. , lower ) side lobes than the natural weighting .
robust weighting combines the advantage offered by the natural weighting and the uniform weighting .
it gives a resolution approaching that of a uniform weighting , with only a modest cost in sensitivity @xcite .
after gridding the sample visibilities and performing an fft on equation [ eq : weighting ] , the obtained dirty image needs to be deconvolved to get an estimate of the true image .
clean , first described by @xcite , is among the widely used algorithms for this purpose .
there have been many variants of clean since its inception , most of them aiming to improve computational efficiency and performance with respect to extended structure .
the original clean algorithm , also called classical clean or delta function clean , assumes that the true image brightness distribution can be well represented by a collection of point sources .
the algorithm iteratively searches for the locations and strengths of the point sources and iteratively subtracts them using the shape of the dirty beam to generate a _ residual map _ and a _ clean component list_. it is common practice to convolve the clean component list with an idealized `` clean beam '' ( e.g. , usually a gaussian with a fwhm matching that of the central component of the dirty beam * ? ? ?
the convolved clean component list is then added to the residual map to obtain the final restored image , which is a plausible representation of the true image .
thus , the clean - ed map is a combination of two maps , imaged with two different beams .
this paper explores the effects of this on measurements of the hi velocity dispersion .
naively one would expect that the flux of a channel map is obtained by simply adding the flux of the cleaned map to that of the residual map : @xmath18 where @xmath19 is the cleaned flux and @xmath20 is the residual flux .
however , @xmath19 and @xmath20 have different units ( @xmath19 is in jy / clean beam and @xmath20 in jy / dirty beam ) and depending on the configuration of the array , the shape of the dirty beam can be strongly non - gaussian and more extended than the clean beam .
this mismatch in shape between the dirty beam and the clean beam leads to an overestimate of the flux density if the clean beam parameters are used to determine the flux of sources containing both cleaned and residual emission .
we illustrate in figure [ fig : crosscut ] , how the dirty beam and the clean beam differ from each other in typical multi - configuration observations with the vla . to account for the difference in area between the clean beam and the dirty beam ,
the flux of the residual must be scaled by a scaling factor @xmath21 : @xmath22 where @xmath23 is the corrected flux and @xmath21 is a correction factor that accounts for the ratio between the clean beam area and the dirty beam area @xcite
. theoretically , due to the absence of zero spacing in interferometric data , the dirty beam integral and the total flux density of the dirty image are zero as there are equal amounts of positive and negative flux in the dirty image . in practice ,
the positive flux is mostly concentrated in the main lobe of the dirty beam and , for the purposes of residual - scaling , one has to choose a region ( box ) centered on the main lobe over which to integrate the dirty beam and ensure most of the flux in the main lobe is captured .
the residual scaling method is included in the aips task imagr and more technical aspects are presented there .
as described in @xcite , the calculation of the flux in the main lobe of the things dirty beam was done inside a box with a half - width of 50 pixels ( 75@xmath24 ) in r.a . and in dec .
this box was chosen as it encompasses the larger part of the main lobe of the things dirty beam while still well within its first negative sidelobe .
this choice is appropriate for all of the things galaxies as they have all been observed in the same manner and with the same set - up , meaning their dirty beams are all very similar
. a quantitative analysis of the uncertainties associated with this choice of box size and with the use of a different dirty beam area has not been performed , and is beyond the scope of this paper .
however , initial testing prior to the reduction of the things data as presented in @xcite indicates that residual - scaling results do not critically depend on the box size ( f. walter , priv . com . ) note that the residual scaling affects the noise and thus it should only be used for flux measurements in areas with genuine emission @xcite .
for this reason , the things residual - scaled cubes are blanked in areas judged to be devoid of emission .
the criteria used by things is that genuine emission must be present in at least three consecutive channels at or above a level of 2@xmath25 in standard cubes convolved to 30 resolution .
these cubes are then used as masks to blank areas containing only noise . while residual - scaling has been widely used to get more accurate flux values ( e.g. , * ? ? ?
* ) , the effects its presence or absence has on hi velocity dispersion have not yet been well explored .
@xcite first noticed the difference between hi velocity dispersion derived from flux - rescaled cubes and standard cubes obtained using multi - configuration vla observations .
this paper makes a thorough comparison of hi velocity dispersions derived from residual - scaled cubes and non - residual - scaled cubes from things .
working visit to the netherlands was supported by the south african national research foundation - netherlands organisation for scientific research ( nrf - nwo ) exchange programme in `` astronomy , and enabling technologies for astronomy '' .
braun , r. 1997 , , 484 , 637 briggs , d. s. ( 1995 ) , high fidelity deconvolution of moderately resolved radio sources , ph.d .
thesis , new mexico inst . of mining & technology .
cannon , j. m. , giovanelli , r. , haynes , m. p. , et al . 2011 , , 739 , l22 cornwell , t. j. 2008 , ieee journal of selected topics in signal processing , 2 , 793 hgbom , j. a. 1974 , , 15 , 417 hunter , d. a. , ficut - vicas , d. , ashley , t. , et al .
2012 , aj , 144 , 134 ianjamasimanana , r. , de blok , w. j. g. , walter , f. , et al .
2012 , , 144 , 96 jorsater , s. , & van moorsel , g. a. 1995 , , 110 , 2037 obrien , j. c. , freeman , k. c. , & van der kruit , p. c. 2010 , , 515 , 62 ott , j. , stilp , a. m. , warren , s. r. , et al .
2012 , , 144 , 123 petric , a. o. , & rupen , m. p. 2007
, , 134 , 1952 rich , j. w. , de blok , w. j. g. , cornwell , t. j. , et al .
2008 , , 136 , 2897 schaye , j. 2004 , , 609 , 667 stilp , a. m. , dalcanton , j. j. , warren , s. r. , et al .
2013 , , 765 , 136 tamburro , d. , rix , h .- w . ,
leroy , a. k. , et al .
2009 , , 137 , 4424 walter , f. , brinks , e. , de blok , w. j. g. , et al .
2008 , , 136 , 2563 yatawatta , s. 2014 , , 444 , 790 young , l. m. , van zee , l. , lo , k. y. , et al .
2003 , , 592 , 111 | random motion of gas in galaxies traced from width 21-cm line of atomic hydrogen hi ) disordered motion gas velocity dispersion provides information on physical properties gas collapse stars energy balance phase structure determined by fitting hi profiles with assumed model gaussian function gauss - hermite polynomial lorentzian @xcite intensity - weighted standard deviation of line - of - sight velocities used as measure of hi velocity dispersion accurate determination hi velocity dispersion limited by imaging process recover shapes of hi line profiles in radio interferometry source intensity distribution not completely sampled by telescope array to recover missing values data reconstruct image telescope point spread function needs deconvolved from image recovered image obtained by adding deconvolved map source brightness distribution to map containing remaining emission noise residual width of hi velocity profiles spread over several velocity channels faint outer wings can appear in adjacent channels with intensities below threshold for cleaning residual image faint residual emission corresponding to these wingspaper investigate effect uncleaned residual emission on velocity dispersion leads to overestimate focus on data things survey @xcite conclusions apply to all hi observations where dirty beam different from gaussian multi - configuration very large array vla ) observations includes recent surveys local irregulars luminosity extremes hi nearby galaxy survey vla survey advanced camera galaxy survey hi in extremely low - mass dwarfs study relevant for other arrays non - gaussian dirty beam atacama large millimeter / submillimeter array results could relevant for molecular line velocity dispersions section background background on weighting of visibilities cleaning processes analysis goal method compare hi velocity dispersion from different types of data cubes investigate type of data cubes in hi velocity dispersion analysis discussion discuss results conclusions interferometric observations of radio astronomical sources provide samples complex visibility function at locations in @xmath2 plane @xcite @xmath3 @xmath4 baseline vectors projected plane perpendicular to source direction deconvolution algorithms recover image information missing values in un - sampled @xmath2 planeinverse fourier transform sampled visibility function @xmath5 dirty image _ @xmath6 : @xmath7 @xmath8 sampling function @xmath9 visibility function fourier convolution theorem fourier transform two functions convolution equation [ eq : num1 ] becomes : @xmath10 @xmath11 dirty beam true image inverse fourier transform visibility function @xmath12 deconvolution algorithm solving true image dirty beam dirty image dirty image fast fourier transform fft ) preferable over simple fourier transform computational speed fft requires sample visibilities regularly spaced grid image reconstruction gridding weighting function applied to sampled visibility function maximise point source sensitivity minimise sidelobe levels @xmath13 weighting function equation [ eq : num1 ] : @xmath14 weighting functions developed for science goals paper discusses effects weighting schemes derivation hi velocity dispersion natural uniform robust weightings radio interferometry natural weighting weight 1/@xmath15 on visibility @xmath16 @xmath15 noise variance visibility noise similar for all visibilities each @xmath2 points same weightdue to gridding @xmath2 cells at shorter baselines get more weight plane higher density points natural weighting offers best surface brightness sensitivity minimum noise larger beam size for uniform weighting cell in @xmath2 plane gets weight of 1/@xmath17 @xmath17 is density of @xmath2 points density lower at larger baselines weighted more in better resolution lower sensitivity increased rms noise uniform weighting gives better lower side lobes robust weighting combines natural uniform gives resolution approaching uniform weighting modest cost in sensitivity after gridding sample visibilities equation weighting obtained dirty image needs to deconvolved estimate of true image clean first described by @xcite widely used for this purpose many variants of clean improve computational efficiency performance extended structure original clean algorithm classical clean or delta function clean assumes true image brightness distribution represented by collection point sources algorithm searches for locations strengths point sources subtracts them using shape dirty beam to generate residual map clean component list common to convolve clean component list with idealized clean beamusually gaussian fwhm matching central component dirty beam convolved clean component list added to residual map final restored image plausible representation true image clean - ed map combination of two maps two different beams paper explores effects on hi velocity dispersion flux channel map obtained by adding flux cleaned map to residual map @xmath18 @xmath19 cleaned flux @xmath20 residual flux @xmath19 @xmath20 have different units jy / clean beam jy / dirty beam configuration shape dirty beam can non - gaussian more extended than clean beam mismatch shape leads to overestimate flux density clean beam parameters flux sources cleaned residual emission illustrate in figure dirty beam clean beam differ in multi - configuration observations with vla account for difference flux residual must be scaled by scaling factor @xmath21 : @xmath22 @xmath23 corrected flux @xmath21 correction factor ratio between clean beam area dirty beam area zero spacing in interferometric data dirty beam integral total flux density dirty image are zero equal amounts positive and negative flux in dirty imagepositive flux concentrated in main lobe dirty beam for residual - scaling region ( box ) on main lobe to integrate dirty beam ensure flux captured residual scaling method included in aips task imagr technical aspects presented calculation of flux in main lobe inside box with half - width of 50 pixels ( 75@xmath24 ) in r dec box chosen encompasses larger part of main lobe within first negative sidelobe choice appropriate for all things galaxies observed same same set - up dirty beams similar quantitative analysis of uncertainties box size different dirty beam area not performed beyond scope paper initial testing indicates residual - scaling results not depend on box size residual scaling affects noise only used for flux measurements in areas with genuine emission residual - scaled cubes blanked in areas devoid of emission criteria genuine emission must be present in three consecutive channels at or above level of 2@xmath25 in standard cubes convolved to 30 resolution cubes used as masks to blank areas noise residual - scaling used to accurate flux valueseffects on hi velocity dispersion explored @xcite noticed difference between velocity dispersion flux - rescaled cubes standard cubes multi - configuration vla observations paper velocity dispersions residual - scaled non - residual - scaled cubes visit to netherlands supported by south african national research foundation - netherlands exchange programme astronomy braun . 1997 , 484 , 637 briggs d 1995 high fidelity deconvolution of moderately resolved radio sources ph.d . thesis new mexico inst. of mining & technology cannon j giovanelli haynes 2011 , 739 , l22 cornwell t. j. 2008 journal signal processing 2 , 793 hgbom j 1974 15 , 417 hunter ficut - vicas ashley 2012 144 , 134 ianjamasimanana de blok walter f 2012 144 , 96 jorsater van moorsel 1995 110 , 2037 obrien freeman van der kruit. 2010 , 515 , 62 ott stilp warren2012 144 , 123 petric rupen. 2007 134 , 1952 rich de blok j.. cornwell t. j. 2008 136 , 2897 schaye 2004 609 , 667 stilp dalcanton j. j. warren s. r. 2013 765 , 136 tamburro rix leroy a. k. 2009 137 , 4424 walter brinks de blok w. j. g. 2008 136 , 2563 yatawatta 2014 444 , 790 young . van zee lo k. y. 2003 592 , 111 | 0.508497 | 3,170 | 1,560 | , Saving $0.1 in GPT-4. |
155,139 | as part of a broader study of kaon photoproduction on deuterium , the @xmath0 channel was investigated .
the data were acquired at jefferson lab using the clas detector and the photon - tagging facility installed in hall b. the photon energy range covered was from 0.50 to 2.95 gev .
for the present analysis , the @xmath1 channel was identified by detecting the positive kaon in coincidence with both decay products of the @xmath2 hyperon , @xmath3 and @xmath4 .
preliminary differential cross
sections are shown as a function of the invariant energy w and the kaon polar angle in the center of mass system . | the electromagnetic strangeness production is an important part of the jefferson lab s experimental program .
several experiments have been approved to run in all three experimental halls .
these experiments include kaon electro- and photoproduction on hydrogen , deuterium , @xmath5he , and @xmath6he .
kaon photoproduction on deuterium is governed by three main mechanisms : * the elementary amplitudes of the six kaon production reactions possible on the nucleon ( @xmath7 , @xmath8 , @xmath9 , @xmath0 , @xmath10 , @xmath11 ) . * the fermi motion of the proton and neutron inside the deuteron .
the momentum distribution of nucleon momenta can be calculated using the deuteron wave function . *
the interaction between the final state hadrons .
experimental information exists for the first three exclusive kaon photoproduction channels . as shown in fig .
[ fig : gamkelem ] @xcite , there are no previous data for the kaon photoproduction on the neutron .
one of the goals of experiment e89 - 045 was to investigate open strangeness photoproduction on the neutron .
these studies will provide additional information about isospin dependence since the elementary operator for the @xmath2 production could be quite different than the operator for @xmath12 production @xcite .
the data presented here were obtained in hall b in august
september 1999 .
the total running time was 23 days .
real photons were produced by tagged bremsstrahlung provided by the photon - tagging facility of hall b @xcite .
experimental data were obtained for two incident electron energies : 2.47 and 3.11 gev . since the tagging range is 20% 95% of the electron - beam energy , the photon energy range covered was from 0.50 to 2.95 gev .
the rate of tagged photons was approximately 10@xmath13/sec .
the outgoing particles were detected in the cebaf large acceptance spectrometer ( clas ) @xcite , a magnetic toroidal multi
gap spectrometer , covering a range of polar angles from @xmath14 to @xmath15 , with almost @xmath16 azimuthal coverage .
charged particle identification is achieved in clas using information about the flight time and momentum of the particle obtained using the time of flight ( tof ) detectors and drift chambers , respectively . a spectrum for positive hadrons as detected in clas
is shown in panel a ) of figure [ fig : kp_pid ] .
the peak corresponding to the positive kaons is barely visible between the pion and proton peaks .
the dashed curve in the bottom panel shows the kaon peak after filtering the data by applying a mass cut from 0.3 to 0.7 .
there is still a significant background underneath the kaon peak , coming mostly from misidentified pions and protons .
the solid curve represents the same spectrum after applying additional timing cuts to minimize the number of pions and protons produced by photons coming from neighbouring beam bunches .
= 4.5 in in the present analysis the reaction @xmath17 was selected by detecting the positive kaon and the decay products of the @xmath2 , the neutron and the negative pion .
the pion was detected using the time of flight counters and the drift chambers ( figure [ fig : kp_pid ] c ) ) , while the neutron was detected in the electromagnetic calorimeter .
the neutron momentum was determined from the time of flight information given by the electromagnetic calorimeter .
the efficiency for detecting neutrons in clas was studied using the reaction @xmath18 .
the results are shown in figure[fig : n_pid ] compared to a geant based monte carlo simulation of the clas detector ( gsim ) .
the agreement between the data and monte carlo , while not perfect , was deemed reasonable for the limited precision required for the present experiment .
detection inefficiencies and geometric acceptance corrections for charged particles were also obtained using gsim .
the events for the reaction @xmath19 were generated as quasifree production on a neutron with initial momentum distribution calculated using the bonn potential @xcite .
= 4.5 in in figure [ fig : inv_mass]a the invariant mass of the @xmath20 system reconstructed using the energy and momentum of the negative pion and neutron is shown , integrated over all photon energies ( incident electron beam energy was 2.47 gev ) and @xmath21 angles .
the expected value for the @xmath2 mass is indicated by the dashed vertical line . for the present analysis
the data were binned in a 2d grid with 100 mev bins in incident photon energy and five bins in the center of mass polar angle of the @xmath21 ( @xmath22 ) . for each bin a cut around the @xmath2 peak was used to identify the @xmath23 events . due to the large uncertainty in the determination of the neutron momentum ,
this cut was kept loose , so as to minimize the rejection of good @xmath2 events .
the experimental yield was extracted by fitting ( using a using a log - likelihood method ) the shape of the invariant mass , as obtained from the simulation to the experimental data , allowing for a small ammount of residual background .
this procedure is illustrated in figure [ fig : inv_mass]b for events having photon energies between 1.4 and 1.5 gev and @xmath24 : the dashed line shows the simulation ( signal ) plus the background , while the dotted line represents the background alone .
the data are represented by the solid line .
= 4.5 in in order to obtain the differential cross section the extracted yield was corrected for all known detector inefficiencies and geometric acceptance effects .
the target density and length as well as the tagged photon flux were also taken into account 20% uncertainty in the determination of the photon flux .
final normalization constants will be available in the near future . ] .
figure [ fig : myresults ] shows preliminary differential cross sections for four of the five angular bins .
the uncertainties shown are statistical only .
extensive studies of the systematic uncertainties are well underway ( current estimates indicate a systematic uncertainty of 15% or less for most bins ) . for all angles one sees that the cross
section rises rather sharply from threshold , reaching a maximum for w between 1.8 and 1.9 gev .
the data do not rule out the existence of structures ( bumps ) in the higher w range , although further investigation is needed .
the reaction @xmath25 was studied using tagged photons and the clas detector in hall b at jefferson lab .
the strange @xmath21 meson was detected before its in flight decay , while the associated @xmath2 hyperon was identified by observing both its decay products ( @xmath3 and @xmath4 ) .
preliminary differential cross sections were presented .
the current analysis provides complementary information to that obtained in photoproduction off the proton , furthering our understanding of the reaction mechanism that governs open strangeness production . future work , besides finalizing the present analysis , will focus on the extraction of the final state yn interaction .
this work was supported in part by the u. s. department of energy under grant no .
de - fg02 - 95er40901 . | electromagnetic strangeness production jefferson lab s program experiments approved in three halls include electro photoproduction on hydrogen deuterium @xmath5he @xmath6he photoproduction on deuterium governed by three mechanisms amplitudes of six kaon production reactions on nucleon motion of proton neutron inside deuteron momentum distribution wave function interaction between final state hadrons information for first three kaon photoproduction channels no previous data for kaon photoproduction on neutron experiment e89 - 045 investigate open strangeness photoproduction on neutron studies provide information about isospin dependence operator for @xmath2 production different @xmath12 data obtained in hall b august september 1999 running time 23 days photons produced by tagged bremsstrahlung photon - tagging facility hall b data for electron energies 2.47 3.11 gev tagging range 95% electron energy photon energy range 0.50 to 2.95 gev rate of tagged photons approximately 10@xmath13/sec outgoing particles detected in cebaf large acceptance spectrometer @xcite toroidal polar angles from @xmath14 to @xmath15 almost @xmath16 azimuthal coverage charged particle identification in using flight time momentum time of flight detectors drift chambers spectrum for positive hadrons shown in panel a ) figure peak to positive kaons barely visible between pion proton peaks dashed curve shows kaon peak after filtering data mass cut from 0.3 to 0.7 significant background underneath kaon peak from misidentified pions protons solid curve represents same spectrum after additional timing cuts to reaction @xmath17 positive kaon decay products of @xmath2 neutron negative pion pion detected using time of flight counters drift chambers neutron detected in electromagnetic calorimeter neutron momentum determined from time of flight information efficiency for detecting neutrons in studied using reaction @xmath18 results shown in compared to geant based monte carlo simulation of clas detector agreement between data monte carlo not perfect reasonable for limited precision experimentdetection inefficiencies geometric acceptance corrections for charged particles obtained using gsim events reaction @xmath19 generated quasifree production on neutron initial momentum distribution calculated using bonn potential @xcite invariant mass @xmath20 system reconstructed using energy momentum negative pion neutron integrated over photon energies energy 2.47 gev @xmath21 angles expected value @xmath2 mass indicated by dashed vertical line data binned in 2d grid 100 mev bins incident photon energy five bins center mass polar angle @xmath21 cut around @xmath2 peak @xmath23 events uncertainty neutron momentum cut kept loose minimize rejection @xmath2 events experimental yield extracted by fitting log - likelihood method shape invariant mass to experimental data small residual background procedure illustrated in figure for events photon energies between 1.4 1.5 gev @xmath24 dashed line shows simulation signal background dotted line background data represented by solid line 4.5 extracted yield corrected for detector inefficiencies geometric acceptance effects target density length tagged photon flux taken 20% uncertainty final normalization constants available futurefigure shows differential cross sections four five angular bins uncertainties statistical studies systematic uncertainties underway estimates indicate uncertainty 15% or less most bins angles cross section rises from threshold maximum between 1.8 1.9 gev data rule out structures in higher w range further investigation needed reaction @xmath25 studied using tagged photons clas detector hall b jefferson lab strange @xmath21 meson detected before flight decay @xmath2 hyperon identified decay products preliminary differential cross sections presented current analysis provides complementary information photoproduction proton understanding reaction mechanism open strangeness production future work extraction final state yn interaction work supported by u. s. department of energy grant - fg02 - 95er40901 | 0.265394 | 1,583 | 767 | , Saving $0.0 in GPT-4. |
155,164 | many networks are characterized by highly heterogeneous distributions of links , which are called scale - free networks and the degree distributions follow @xmath0 .
we study the robustness of scale - free networks to random failures from the character of their heterogeneity .
entropy of the degree distribution can be an average measure of a network s heterogeneity .
optimization of scale - free network robustness to random failures with average connectivity constant is equivalent to maximize the entropy of the degree distribution . by examining the relationship of the entropy of the degree distribution ,
scaling exponent and the minimal connectivity , we get the optimal design of scale - free network to random failures .
we conclude that the entropy of the degree distribution is an effective measure of network s resilience to random failures .
scale - free networks ; information theory ; entropy ; random failures _ pacs _ : 89.75.-k ; 89.75.fb | many complex systems are characterized by the network of interactions among its components .
it has been shown that most of the networks are highly heterogeneous in their connectivity patterns .
heterogeneity can be easily identified by looking at the degree distribution @xmath1 , which gives the probability of having a node with @xmath2 links .
most complex networks can be described by degree distributions @xmath0 , with @xmath3 .
these networks include social networks(such as movie - actor networks , science citations and cooperation networks ) , the internet and the world wide web , metabolic networks and protein interaction networks , etc @xcite . since albert _ et al .
_ raised the question of random failures and intentional attack on networks @xcite , enormous interest has been devoted to the study of the resilience of networks to failures of nodes or to intentional attacks @xcite .
magoni studied a general strategy of attacks on the internet @xcite .
it is important to understand how to design networks which are optimally robust against both failures and attacks .
most of researchers use percolation theory to study this problem @xcite .
a fraction @xmath4 of the nodes and their connections are removed randomly@xmath5the integrity might be compromised : for @xmath6 , there exists a certain threshold value @xmath7 . when @xmath8 , the network disintegrates into smaller , disconnected parts . below that critical threshold value
, there still exists a connected cluster . for @xmath9 ,
the networks are more resilient and the threshold value @xmath7 approaches 1 @xcite . also , some researchers work on optimizing network robustness to both random failures and attacks with percolation theory @xcite .
a simple but essential character of scale - free network is its heterogeneous link distribution .
moreover , heterogeneity is in direct relationship with the network s resilience to attacks .
many real - world networks are scale - free and they are robust to random failures but vulnerable to targeted attacks .
heterogeneity can be measured by entropy @xcite .
used entropy of the remaining degree and mutual information to study some model networks of different heterogeneity and randomness @xcite . in this paper , different from percolation theory
, we try an alternative point of view@xmath5the entropy of the degree distribution to describe scale - free network s heterogeneity . to determine the optimal design of scale - free network to random failures , we maximize the robustness of scale - free network to random failures while keeping the cost constant ( that is , the average number of links per node remains constant ) .
we find that optimization of scale - free network s robustness to random failures is equivalent to maximize the entropy of the degree distribution . by optimizing the entropy of the degree distribution
, we get the scenario of optimal design of scale - free network to random failures .
in general , scale - free networks have the degree distributions @xmath1 , @xmath0 , @xmath10 , where @xmath11 is the minimal connectivity and @xmath12 is an effective connectivity cutoff present in finite networks .
then the entropy of the degree distribution can be defined as follows : @xmath13 where @xmath14 is the total number of nodes in the network and the restriction to the degree distribution is @xmath0 . within the context of complex networks
, the entropy of the degree distribution provides an average measure of network s heterogeneity , since it measures the diversity of the link distribution @xcite .
two extreme cases are the maximal value and the minimal one .
the maximum value is @xmath15 obtained for @xmath16 .
@xmath17 occurs when @xmath18 . for the power - law degree distribution of scale - free network ,
the impact of diversity ( long tails ) is obvious , increasing the uncertainty . as the scaling exponent increases or the cut - off decreases
, the network becomes less heterogeneous and as a result a lower entropy is observed @xcite . in this paper , we study the entropy of the degree distribution for scale - free networks . with continuous approximation , for the power law degree distribution of scale - free network
, entropy @xmath19 can be expressed as follows : @xmath20\nonumber\\ & = & -\frac{c\log{c}}{1-\alpha}\cdot{(k^{1-\alpha}-m^{1-\alpha})}+\frac{c\alpha}{1-\alpha } \cdot[\log{k}\cdot{k^{1-\alpha}}-\log{m}\cdot{m^{(1-\alpha)}}\nonumber\\ & & -\frac{1}{1-\alpha}\cdot{(k^{1-\alpha}-m^{1-\alpha})}]\end{aligned}\ ] ] the largest connectivity @xmath21 @xcite , the coefficient @xmath22 is obtained from @xmath23 and thus @xmath24 then entropy @xmath19 can be expressed as a function of scaling exponent @xmath25 , the minimal connectivity @xmath11 and network scale @xmath14 . @xmath26 fig .
1 displays the relationship between entropy @xmath19 and @xmath25 for different @xmath14 when the minimal connectivity @xmath11 is defined . we know that with the increase of scaling exponent @xmath25 , the cut - off connectivity @xmath12 decreases , which makes the network less heterogeneous . as a result , a lower entropy is observed .
so different scaling exponent @xmath25 brings the network different degree of heterogeneity .
the average connectivity @xmath27 is obtained with continuous approximation , @xmath28 and thus @xmath29 according to the above results , we can formulate the scale - free network s resilience to random failures problem as the following nonlinear mixed integer programming .
@xmath30 the problem mentioned above is different from the usual maximizing the entropy of the network problem .
note that the highest entropy for a network with a given @xmath27 is exponential random graph @xcite . to get the optimal design of scale - free networks , we restrict the degree distribution to be a power - law distribution .
the problem can be rewritten as follows : @xmath31 by solving the mixed integer programming problem , we get the optimal value @xmath19 , the optimal solutions @xmath25 and @xmath11 for different average connectivity @xmath27 .
the results are shown in fig .
2 . from fig .
2 , we can see that , in general , with the increase of average connectivity @xmath27 , entropy @xmath19 as well as the minimal connectivity @xmath11 increases , too . in part
, the optimal minimal connectivity @xmath11 keeps constant for a certain range of @xmath27 , where scaling exponent @xmath25 decreases with the increase of @xmath27 . as a result ,
the optimal network becomes much more heterogeneous than the one with bigger scaling exponent . to prove the entropy of the degree distribution is indeed an effective measure of network robustness under node removal
, we examine the relationship between the entropy of the degree distribution and the network threshold value @xmath7 , which was presented by cohen _
with percolation theory @xcite .
the criterion to calculate the percolation critical threshold value @xmath7 of a randomly connected network is : @xmath32 where @xmath33 is calculated from the original distribution before the random failures . with @xmath34 , @xmath35 can be expressed as follows : @xmath36 in fig .
3 , @xmath7 is calculated from the optimal value @xmath25 and @xmath11 which are obtained from maximizing @xmath19 for different average connectivity @xmath27 .
the results indicate that for some constant @xmath11 , @xmath7 increases with the increase of @xmath27 .
since both @xmath25 and @xmath14 affect network s resilience to random failures , we may ask how the network s robustness changes with @xmath14 .
we just examine the relationship between @xmath7 and @xmath25 for different @xmath14 .
the results are shown in fig .
4 . from fig .
4 , we can see that with the increase of @xmath25 , @xmath7 decreases , while the threshold value @xmath7 becomes bigger with the increase of network scale @xmath14 , which are consistent with the results shown in fig.1 . to get an intuitive understanding of the relationship between the two measures ,
we show the results of the threshold value @xmath7 and @xmath19 for different average connectivity @xmath27 in fig .
it can be seen that @xmath7 and @xmath19 vary consistently , that is , @xmath7 increases with @xmath19 for different average degree @xmath27 . also , with the increase of @xmath27 , both @xmath7 and @xmath19 get improved .
in real - world networks one is often interested in the question of how to design a robust network while keeping the cost constant .
it is well known that scale - free networks are resilient to random failures for their heterogeneous links distributions . in this paper , different from percolation theory to evaluate the threshold value @xmath7
, we use the entropy of the degree distribution to study the resilience of scale - free networks to random failures .
heterogeneity is a simple but essential character of scale - free networks and it is in direct relationship with the network s resilience to random failures .
this character motivates us to use the entropy of the degree distribution , which provides an average measure of heterogeneity to study the effect of random failures on scale - free networks . by optimizing the entropy of the power - law degree distribution
, we get the optimal value of entropy @xmath19 , the optimal solutions of @xmath25 and @xmath11 for different @xmath27 .
the results indicate that when the network scale @xmath14 is given , the optimal solution @xmath11 is constant for a certain range of @xmath27 , in which range the scaling exponent @xmath25 decreases .
so a higher entropy is obtained .
of course , large average connectivity brings high cost .
we then examine the relationship between threshold value @xmath7 and @xmath14 for different @xmath25 .
the results are also consistent with that of the entropy of the degree distribution .
we also show the relationship of the two measures@xmath5@xmath7 and @xmath19 .
the results indicate that @xmath7 and @xmath19 vary consistently , that is , @xmath7 increases with @xmath19 for a certain average degree @xmath27 .
also , both @xmath7 and @xmath19 get improved with the increase of the average degree @xmath27 .
we just try an alternative point of view to analyze the robustness of the network from its heterogeneity .
the results indicate that the entropy of the degree distribution is an effective measure of scale - free network s resilience to random failures .
deep discussions on the resilience of scale - free networks to targeted attacks will be made in the future .
we are grateful to the anonymous referees for their insightful and helpful comments .
the work was supported by nsfc(90103033 ) .
00 r. albert , a .-
barabsi , rev .
newman , siam review 45(2003)167 .
dorogovtsev , f.f .
mendes , adv . phys .
51(2002)1079 . h. jeong , b. tombor , r. albert , z.n .
oltvai , nature 407(2000)651 .
h. jeong , s.p .
mason , a .-
barabsi , z.n .
oltvai , nature 411(2001)41 .
r. albert , h. jeong , a .-
barabsi , nature 406(2000)378 .
r. cohen , k. erez , d. ben - avraham , s. havlin , phys .
r. cohen , k. erez , d. ben - avraham , s. havlin , phys .
r. albert , h. jeong , a .-
barabsi , nature 409(2001)542 .
gallos , p. argyrakis , a. bunde , r. cohen , s. havlin , physica a 344(2004)504 .
p. crucitti , v. latora , m. marchiori , a. rapisarda , physica a 340(2004)388 .
d. magoni , ieee journal on selected areas in communications , 21(2003)949 .
tao zhou , gang yan , bing - hong wang , phys .
e 71(2005)046141 .
valente , a. sarkar and h.a .
stone , phys .
t. tanizawa , g. paul , r. cohen , s. havlin and h.e .
stanley , phys .
e. 71(2005)047101 .
g. paul , t. tanizawa , s. havlin and h.e .
stanley , eur .
j. b 38(2004)187 . | complex systems characterized by network interactions networks heterogeneous in connectivity patterns heterogeneity identified by degree distribution @xmath1 probability of node with @xmath2 links complex networks described by degree distributions @xmath0 @xmath3 networks include social networks internet web metabolic networks protein interaction networks since albert raised question of random failures intentional attack networks interest devoted to resilience of networks to failures attacks magoni studied strategy of attacks on internet important to understand design networks optimally robust against failures attacks researchers use percolation theory fraction nodes connections removed randomly integrity might be compromised threshold value @xmath7 network disintegrates into parts below threshold value connected cluster for @xmath9 networks more resilient threshold value @xmath7 approaches 1 some researchers network robustness to random failures attacks with percolation theory character of scale - free network is heterogeneous link distribution heterogeneity with network s resilience to attacks real - world networks are scale - free robust to random failures vulnerable to targeted attacksheterogeneity measured by entropy used entropy study model networks heterogeneity this paper alternative point of view entropy of degree distribution describe scale - free network s heterogeneity optimal design of - free network to random failures maximize robustness cost constant average links per node optimization of robustness to failures equivalent to maximize entropy degree distribution optimizing entropy scenario optimal design scale free network to random failures scale - free networks have degree distributions @xmath1 @xmath0 @xmath11 minimal connectivity @xmath12 effective connectivity cutoff in finite networks entropy of degree distribution defined @xmath13 @xmath14 is total number of nodes restriction to degree distribution is @xmath0 complex networks entropy provides average measure of network heterogeneity measures diversity of link distribution extreme cases maximal minimal maximum value is for @xmath16 @xmath17 occurs when @xmath18 power - law degree distribution network impact of diversity increasing uncertainty scaling exponent increases cut - off decreases network less heterogeneous lower entropy observed paper study entropy of degree distribution for scale - free networkscontinuous approximation power law degree distribution scale - free network entropy @xmath19 expressed @xmath20\nonumber\\{1-\alpha}{(k{1-\alpha}-m^{1-\alpha{k{1-\alpha}}-{m}{(1-\alpha{1-\alpha}-m{aligned largest connectivity @xmath21 coefficient @xmath22 @xmath23 @xmath24 entropy @xmath19 function scaling exponent @xmath25 minimal connectivity @xmath11 network scale @xmath14 @xmath26 fig 1 relationship entropy @xmath19 @xmath25 different @xmath14 minimal connectivity @xmath11 increase scaling exponent @xmath25 cut - off connectivity @xmath12 decreases network less heterogeneous lower entropy different scaling exponent @xmath25 different heterogeneity average connectivity @xmath27 continuous approximation @xmath28 @xmath29 scale - free network s resilience random failures nonlinear mixed integer programming @xmath30 problem different maximizing entropy network problemhighest entropy network @xmath27 exponential random graph @xcite optimal design scale - free networks restrict degree distribution power - law distribution problem rewritten @xmath31 mixed integer programming optimal value @xmath19 solutions @xmath25 @xmath11 average connectivity @xmath27 results fig . 2 increase average connectivity @xmath27 entropy @xmath19 minimal connectivity @xmath11 increases optimal minimal connectivity @xmath11 constant range @xmath27 scaling exponent @xmath25 decreases increase @xmath27 optimal network heterogeneous bigger scaling exponent entropy measure network robustness under node removal relationship entropy network threshold value @xmath7 cohen percolation theory @xcite criterion calculate percolation critical threshold value @xmath7 randomly connected network @xmath32 @xmath33 calculated from original distribution before random failures @xmath34 @xmath35 expressed @xmath36 @xmath7 calculated from optimal value @xmath25 @xmath11 maximizing @xmath19 different average connectivity @xmath27 constant @xmath11 @xmath7 increases with increase @xmath27@xmath25 @xmath14 affect network resilience to random failures network robustness changes with @xmath14 examine relationship between @xmath7 @xmath25 for different @xmath14 results in fig . 4 increase @xmath25 @xmath7 decreases threshold value @xmath7 bigger with increase network scale @xmath14 consistent with results.1 show results threshold value @xmath7 @xmath19 for different average connectivity @xmath27 in fig @xmath7 @xmath19 vary consistently @xmath7 increases with @xmath19 for different average degree @xmath27 increase of @xmath27 @xmath7 @xmath19 improved real - world networks design robust network cost constant scale - free networks resilient to random failures heterogeneous links distributions paper use entropy of degree distribution study resilience of scale - free networks to random failures heterogeneity essential character scale networks with resilience to random failures entropy of degree distribution effect random failures on scale networks optimizing entropy distribution optimal value of entropy @xmath19 solutions of @xmath25 @xmath11 for different @xmath27results indicate network scale @xmath14 optimal solution @xmath11 constant for range @xmath27 scaling exponent @xmath25 decreases higher entropy obtained large connectivity brings high cost relationship threshold value @xmath7 @xmath14 for different @xmath25 results consistent with entropy degree distribution show relationship two measures@xmath5@xmath7 @xmath19 results indicate @xmath7 @xmath19 vary consistently @xmath7 increases with @xmath19 average degree @xmath27 @xmath7 @xmath19 improved with increase average degree @xmath27 alternative analyze robustness network heterogeneity results indicate entropy degree distribution effective measure scale - free network s resilience to random failures deep discussions on resilience scale networks targeted attacks future grateful to anonymous referees for comments work supported by nsfc(90103033 ) albert barabsi newman siam review 45(2003)167 dorogovtsev mendes phys 51(2002)1079 h jeong tombor albert oltvai nature 407(2000)651 nature 411(2001)41barabsi nature 406(2000)378 r cohen k erez ben - avraham s havlin albert h jeong barabsi nature 409(2001)542 gallos p. argyrakis bunde r. cohen s. havlin 344(2004)504 crucitti v. latora m. marchiori rapisarda 340(2004)388 magoni journal selected areas 21(2003)949 tao zhou gang yan bing - hong wang phys 71(2005)046141 valente sarkar stone tanizawa g. paul r. cohen s. havlin h stanley 71(2005)047101 g. paul tanizawa s. havlin stanley 38(2004)187 | 0.433814 | 2,860 | 1,446 | , Saving $0.1 in GPT-4. |
155,165 | the theoretical relation between the number of post main sequence stars from stellar populations and their total mass is investigated .
this is used to derive some basic relations between the stellar number counts of stellar populations with an age spread and the star formation history which produced them .
the main purpose of this investigation is to offer a guideline for the construction of simulations aimed at reproducing color - magnitude diagrams . | theoretical simulations of color - magnitude diagrams ( cmd ) have proven very successful to investigate the star formation history ( sfh ) of nearby galaxies ( see the contributions by aparicio and by tosi et al . in this volume ) .
in essence , the method works because , on the basis stellar evolution theory , we relate star counts in selected regions of the cmd to the mass of the stellar population which generated them .
since different regions of the cmd are populated by stars of different ages , the stellar counts speak for the mass that went into stars as a function of time . however ,
there are regions in the cmd which , sampling stars from a large age range , only yield integrated information . in this paper
i will anticipate some results of an investigation of the potentials and the limitations of the stellar evolution theory for reconstructing the sfh .
i will consider stellar populations with an age spread , but a fixed metallicity .
i will also just focus on the number counts of stars in post main sequence ( pms ) evolutionary phases .
often theses star occupy the brightest portions of the observational cmds .
in addition , for this kind of stars a straightforward relation applies between the counts and the mass of the parent population .
for an simple stellar population ( ssp ) , i.e. a collection of single stars all with the same age ( @xmath0 ) and metallicity ( @xmath1 ) , the number of stars in the @xmath2-th pms evolutionary phase is well approximated by ( renzini 1981 ) : @xmath3 where @xmath4 to is the turn off mass , @xmath5 is its time derivative , @xmath6 is the lifetime of the pms phase considered , and @xmath7 is the initial mass function ( imf ) by number evaluated at the turn - off mass .
i characterize the ssp for its total mass in the range 0.6@xmath8120 @xmath9 ( @xmath10 ) : @xmath11 having assumed a single slope power law for the imf in the range 0.6@xmath8120 @xmath9 . in this notation
, salpeter imf corresponds to @xmath12=2.35 the scaling of the number of pms stars with the mass of the parent population then becomes : @xmath13 where @xmath14 is the _ specific evolutionary flux _ ,
i.e. the number of stars leaving the ms per unit time per unit mass of the parent population .
the quantity @xmath15 is the _ specific production of pms stars _ in the @xmath2-th evolutionary phase of the ssp with age @xmath0 and metallicity @xmath1 , such that its turn off mass is @xmath4 .
all factors in equation [ eq : dnjs ] depend on @xmath0 and @xmath1 ; the dependence on the imf is included in @xmath16 .
the behaviour of the specific production of pms stars as a function of the ssp age is shown in fig .
[ f : nmass ] for the set of tracks by fagotto et al .
( 1994 ) with z=0.004 .
the trend with age results from two competing effects : the temporal decrease of the specific evolutionary flux as the population ages ( dotted line in the left panel ) , and the increase of the duration of specific pms stages .
for young ssps ( up to @xmath17 1gyr ) the specific production of pms objects increases : the lengthening of the lifetimes prevails , and it s by far driven by the helium burning lifetime .
there is a sharp maximum in @xmath15 at @xmath17 1 gyr : this is due to the fact that ssps around this age have the longest helium burning lifetime ( e.g. sweigart , greggio , & renzini 1990 ; girardi & salaris 2001 ) . a second maximum appears at @xmath17 3 gyr , caused by the lengthening of the rgb lifetime .
after this epoch the specific production of pms stars mildly decreases with increasing age , as the rgb lifetime levels off .
the right panel of fig .
[ f : nmass ] illustrates the impact of the imf slope on @xmath15 .
older than a few hundred myr , steeper slopes correspond to a larger number of pms stars at fixed @xmath10 , reflecting the larger stellar evolutionary flux .
i now pass to consider the location in the cmd where these pms stars are to be counted .
[ f : taumag ] shows the magnitude range covered by pms stars of ssps with ages from 10 myr to 10 gyr .
the fagotto z=0.004 tracks in combination with the bessel , castelli and pletz ( 1998 ) model atmospheres with @xmath18=-0.5 $ ] have been used to produce this plot .
different shading refer to different pms phases ( see caption ) .
there is a well defined @xmath19 relation for helium burners up to @xmath17 1 gyr , that is the age at which the red giant branch develops to full extension ( rgb phase transition ) .
thereafter the he burning is spent at virtually the same luminosity , indipendent on age . in the intermediate age regime
, the luminosity at the 1st agb thermal pulse can be used to age - date the ssps .
this is at the basis of the agb - tip age dating , although the location of the tip itself is uncertain due to our poor knowledge of the mass loss processes during the agb evolution . beyond 12 gyrs there
s no bright feature that betrayes the age of the ssp . to age - date
the ssp we need to sample as deep as the rgb base .
a notable exception is the presence of blue horizontal branch stars , but the corresponding age - dating is uncertain because the color distribution of these stars is sensitive to various poorly understood effects ( see e.g. the second parameter problem ( renzini & fusi pecci 1988 ) ) .
having introduced the basic ingredients , i now pass to examine the connection between stellar counts in magnitude bins on the cmd of stellar populations with an age range , and their mass . for such composite stellar population ( csp )
( [ eq : dnjs ] ) becomes : where the subscript @xmath2 now refers to the magnitude bin , @xmath21 is the star formation rate ( sfr ) at the age @xmath0 , and the integration is performed over the age range contributing to the counts in the @xmath2-th bin . where the sfr and the specific productions are evaluated at the typical age @xmath23 sampled by the @xmath2-th magnitude bin , and @xmath24 is the intercepted age range .
as long as @xmath24 is small ( i.e. the position of the stars on the cmd varies steeply with their age ) the luminosity function ( lf ) gives a detailed description of the star formation rate .
inspection of the left panel of fig .
[ f : taumag ] shows that this is the case only for the brightest portion of the red lf .
magnitude bins fainter than @xmath25 already collect e - agb stars from a large age range .
this problem is not present in the blue part of the cmd ( right panel of fig . [
f : taumag ] ) , where @xmath24 remains small up to a few hundred myrs .
thus counts of blue helium burners give a fair indication of the sfr over this age range ( dohm - palmer et al .
1997 ) . for older stellar populations , the bright portion of the cmd yields basically integrated information , as can be appreciated from fig .
[ f : taumag ] .
i turn then to consider the specific production of objects in a region of the cmd from a csp : the specific production of @xmath2-type stars of a composite stellar population with a range of ages is the weighted mean of the specific productions of the ssps , with the weights given by the sfr in the relevant age range .
the star counts in a region of the cmd divided by @xmath27 yields the mass that went into stars in the age range corresponding to the selected region of the cmd .
it is thus important to see how this quantity varies under different assumptions for the sfh . in the following
i will consider the stellar counts in the red portion of the cmd .
the left panel of fig .
[ f : nspec ] illustrates the specific production of e - agb stars brighter than the rgb tip ( in the i band ) for a variety of sf histories in the age range from 0.1 to 0.9 gyr .
the thick horizontal lines show the level of @xmath27 for bursts of sf , modelled as gaussian distributions centered on progressively older ages , all with a 0.1 gyr age dispersion ( dotted lines ) .
the thick continuous line , instead , shows @xmath27 as a function of the maximum age of the csp , for a family of sf episodes all occurred at a constant rate up to @xmath28 gyr . the ratio between the mass of the csp and the star counts depends on the age and the age limits of the sf episode considered .
on the other hand the luminosity and colour of the agb stars can be used to rank their age distribution , so that the uncertainty in the derived csp mass can be estimated within a factor of a few .
the right panel of fig .
[ f : nspec ] shows the specific production of stars in the bin @xmath30 , which collects young helium burners , e - agb intermediate age and old ( agb + rgb ) bright stars ( see fig . [
f : taumag ] ) . in this magnitude bin , @xmath29 is very large for the young helium burners , temporarily drops at intermediate ages ( where the ssp only yield agb stars ) , and has a second peak at the rgb phase transition .
for ages older than @xmath17 1gyr , @xmath29 is only mildly varying over the whole range of ages considered .
it follows that , in @xmath30 , the specific production of stars from csp with @xmath31 1 gyr only marginally depends on the age range .
the impact of the presence of young stars ( @xmath32gyr ) in this magnitude bin is illustrated by the thick lines for a variety of sfhs .
specifically , the dashed lines correspond to an age distribution @xmath33 ; the solid line to a constant sfr ; the dotted line to an age distribution @xmath34 .
the ratio of mass in stars younger than 1 gyr to the mass in stars with ages between 1 and 10 gyr is 5 ( upper dashed ) , 1 ( lower dashed ) , 0.1 ( solid ) , 0.05 ( upper dotted ) and 0.02 ( lower dotted ) . it can be seen that , if the csp maximum age is older that the rgb phase transition ( which gives well defined signature on the cmd , i.e. the rgb tip ) , the specific production is not very sensitive to the sf history and age limits .
again , the uncertainty of the conversion of stellar counts to total mass of the parent population can be estimated around a factor of a few .
similar considerations hold for a magnitude range which samples the helium burners ( e.g. @xmath35 in fig .
[ f : taumag ] ) . *
the bright portion of the lf of pms stars allows to recover the sfh with a fair degree of detail , up to ages of a few 10@xmath36 years . * for older ages ,
the specific production of stars in the various regions of the cmd depends on the age limits and on the sfr law of the csp .
it is however possible to estimate the total mass of the csp within a factor of a few , for an assumed imf slope . on the average , for salpeter imf
one gets approximately 1 e - agb stars brighter than the rgb tip every 20000 @xmath9 , 1 ( red ) star in @xmath37 every 2000 @xmath9 and 1 ( red ) star star in @xmath35 every 200 @xmath9 of the parent population ( in stars more massive than 0.6 @xmath9 ) .
the best way to decode the stars distribution on the cmd in terms of sfh is by computing synthetic cmds , which account for the stellar evolution , number statistics and observational uncertainties .
in addition , the simulations allow to benefit of the information from the ms phase , and to cross check the star counts in different cmd regions which sample the same ages .
nevertheless , the number - mass relation illustrated here offers a quick - look tool to picture the sf history from an observational cmd , as well as theoretical support for the results we achieve through the construction of synthetic cmds .
bessel , m. s. , castelli , f. , & pletz , b. 1998 , , 337 , 321 dohm - palmer , r. c. , skillman , e. d. , saha , a. , tolstoy , e. , mateo , m. , gallagher , j. , hoessel , j. , chiosi , c. , & doufour , r. j. 1997 , , 114 , 2527 fagotto , f. , bressan , a. , bertelli , g. , & chiosi , c. 1994 , , 105 , 29 girardi , & salaris , m. 2001 , , 323 , 109 renzini , a. 1981 , ann .
fr . , 6 , 87 renzini , a. , & fusi pecci f. 1988 , , 26 , 199 sweigart , a. v. , greggio , l. , & renzini , a. 1990 , , 364 , 527 | theoretical simulations of color - magnitude diagrams successful investigate star formation history of nearby galaxies see contributions by aparicio tosi et al method works stellar relate star counts in regions cmd to mass stellar population different regions populated by stars different ages stellar counts speak for mass function of time regions large age range yield integrated information paper anticipate results of investigation potentials limitations stellar evolution theory for reconstructing sfh consider stellar populations with age spread fixed metallicity focus on number counts of stars in post main sequence evolutionary phases occupy brightest portions of observational cmds straightforward relation applies between counts and mass of parent population for simple stellar population single stars same age metallicity number of stars in @xmath2-th pms evolutionary phase approximated by renzini 1981 ) : @xmath3 @xmath4 is turn off mass @xmath5 time derivative lifetime of pms phase @xmath7 initial mass function at turn - off masscharacterize ssp total mass range 0.6@xmath8120 @xmath9 @xmath10 @xmath11 single slope power law imf range 0.6@xmath8120 @xmath9 salpeter imf corresponds @xmath12=2.35 scaling number pms stars mass parent population becomes @xmath13 @xmath14 specific evolutionary flux number stars leaving ms per unit time mass parent population quantity @xmath15 specific production pms stars @xmath2-th evolutionary phase ssp age @xmath0 metallicity @xmath1 turn off mass @xmath4 factors equation depend @xmath0 @xmath1 dependence imf @xmath16 behaviour specific production pms stars function ssp age shown fig . nmass set tracks fagotto et al . 1994 ) z=0.004 trend age results temporal decrease specific evolutionary flux population ages increase duration specific pms stages young ssps up to @xmath17 1gyr specific production pms objects increases lengthening lifetimes prevails driven helium burning lifetime maximum @xmath15 @xmath17 1 gyr due ssps longest helium burning lifetimesecond maximum at @xmath17 3 gyr caused lengthening rgb lifetime after production pms stars decreases with age rgb lifetime levels off right panel impact imf slope on @xmath15 steeper slopes larger pms stars at @xmath10 larger stellar evolutionary flux location in cmd pms stars counted f : taumag shows magnitude range pms stars ssps ages 10 myr to 10 gyr fagotto z=0.004 tracks bessel , castelli pletz 1998 ) model atmospheres with @xmath18=-0.5 $ ] used plot different shading different pms phases defined @xmath19 relation for helium burners up to @xmath17 1 gyr red giant branch develops to full extension rgb phase transition burning spent at same luminosity indipendent on age intermediate age regime luminosity at 1st agb thermal pulse to age - date ssps basis of agb - tip age dating location uncertain mass loss beyond 12 gyrs no bright feature age ssp to age - date sample deep as rgb base exception blue horizontal branch stars age - dating uncertain color distribution sensitivesecond parameter problem renzini & fusi pecci 1988 introduced basic ingredients examine connection between stellar counts in magnitude bins cmd stellar populations age range mass composite stellar population csp ) dnjs subscript @xmath2 refers magnitude bin @xmath21 star formation rate sfr at age @xmath0 integration over age range counts @xmath2-th bin sfr specific productions evaluated at typical age @xmath23 @xmath2-th magnitude bin @xmath24 intercepted age range @xmath24 small position varies age luminosity function lf ) detailed description star formation rate brightest portion of red lf magnitude bins fainter than @xmath25 collect e - agb stars from large age range problem not present in blue part cmd @xmath24 small up to few hundred myrs counts of blue helium burners give fair indication sfr over age range dohm - palmer et al . 1997 older stellar populations bright portion of cmd yields integrated information fig .consider specific production of objects cmd from csp production @xmath2-type stars composite stellar population ages is weighted mean of specific productions ssps weights sfr relevant age range star counts in region cmd divided by @xmath27 yields mass stars in age range selected region important to see quantity varies under assumptions sfh consider stellar counts in red portion of cmd left panel fig illustrates specific production of e - agb stars brighter than rgb tip for sf histories age range 0.1 to 0.9 gyr horizontal lines show level of @xmath27 for bursts sf older ages 0.1 gyr age dispersion continuous line shows @xmath27 function of maximum age csp for sf episodes constant up to @xmath28 gyr ratio between mass csp star counts depends on age age limits sf episode luminosity colour of agb stars rank age distribution uncertainty in csp mass estimated within factor few right panel shows specific production of stars in bin @xmath30 young helium burners e - agb intermediate age old ( agb + rgb bright starsmagnitude bin @xmath29 large for young helium burners drops at intermediate ages peak at rgb phase transition ages older than @xmath17 1gyr @xmath29 mildly varying over ages @xmath30 specific production of stars from csp with @xmath31 1 gyr marginally depends on age range impact young stars ( @xmath32gyr ) illustrated by thick lines for sfhs dashed lines correspond to age distribution @xmath33 solid constant sfr dotted line age distribution @xmath34 ratio of mass in stars younger than 1 gyr to stars ages between 1 and 10 gyr is 5 1 lower 0.1 solid 0.05 0.02 csp maximum age older rgb phase transition specific production not sensitive to sf history age limits uncertainty of conversion of stellar counts to total mass parent population estimated around factor of a few similar considerations for magnitude range helium burners @xmath35 bright portion of lf pms stars sfh detail up to ages 10@xmath36 years older ages specific production of stars depends on age limits sfr law of csppossible estimate total mass csp factor few assumed imf slope average salpeter imf 1 e - agb stars brighter than rgb tip every 20000 @xmath9 1 red star @xmath37 every 2000 @xmath9 1 @xmath35 every 200 @xmath9 parent population massive than 0.6 @xmath9 decode stars distribution cmd sfh computing synthetic cmds stellar evolution number statistics observational uncertainties simulations benefit information ms phase cross check star counts different cmd regions same ages number - mass relation offers quick tool sf history observational cmd theoretical support results construction synthetic cmds bessel castelli pletz 1998 , 337 , 321 dohm - palmer skillman saha tolstoy mateo gallagher hoessel chiosi doufour 1997 , 114 , 2527 fagotto bressan bertelli chiosi 1994 , 105 , 29 girardi salaris 2001 , 323 , 109 renzini 1981, 6 , 87 renzini fusi pecci f.1988 26 199 sweigart v greggio renzini 1990 364 527 | 0.512496 | 3,017 | 1,564 | , Saving $0.1 in GPT-4. |
155,200 | we consider a non - linear extension of biot s model for poromechanics , wherein both the fluid flow and mechanical deformation are allowed to be non - linear .
we perform an implicit discretization in time ( backward euler ) and propose two iterative schemes for solving the non - linear problems appearing within each time step : a splitting algorithm extending the undrained split and fixed stress methods to non - linear problems , and a monolithic l - scheme .
the convergence of both schemes is shown rigorously .
illustrative numerical examples are presented to confirm the applicability of the schemes and validate the theoretical results .
* _ index terms _ * biot s model l - schemes mfem convergence analysis coupled problems poromechanics | poromechanics , that is to say the coupled flow and mechanics of porous media , plays a crucial role in many societal relevant applications . these include geothermal energy extraction , energy storage in the subsurface , @xmath0 sequestration and understanding of biological tissues . the increased role played by computers for the development and optimisation of ( industrial ) technologies for these applications
enhances the need for improved mathematical models and robust numerical solvers for poromechanics .
the most common mathematical model for coupled flow and mechanics in porous media is the linear , quasi - stationary biot model @xcite .
the model consists of two fully coupled partial differential equations , representing balance of forces for the mechanics and conservation of mass for ( single - phase ) flow in porous media . in terms of modelling ,
non - linear extensions are considered in @xcite or non - stationary biot , i.e. biot - allard model in @xcite .
these coupled ( linear or non - linear ) equations are in practice impossible to solve analytically , and very challenging to solve numerically . it has been widely acknowledged that only simulation with fully coupled fluid potential and mechanical deformation is accurate for non - linear poro - mechanical processes @xcite . nevertheless , due to the complexities associated with monolithic solvers for the full non - linear problem , industry standard remains to use so - called weakly , or iteratively , coupled approaches @xcite .
these de - coupled approaches lead to significant numerical diffusion , which may also mask fundamental numerical incompatibilities between the spatial discretizations . as a response to this current status ,
the objective of this paper is to show how a simple linearization technique , i.e. the @xmath1-scheme ( see @xcite for application of this method to richards equation or two - phase flow in porous media ) can be combined with a splitting algorithm ( known both as the undrained split or the fixed stress method @xcite ) to obtain a robust and efficient iterative scheme for solving a non - linear biot model . in this paper
, we use for concreteness linear conformal galerkin elements for the discretization in space of the mechanics equation and mixed finite elements for the flow equation @xcite .
precisely , the lowest order raviart - thomas elements are used @xcite .
we expect , however , that the solution strategy discussed herein will be applicable to other combinations of spatial discretizations such as those discussed in @xcite and the references therein .
backward euler is used for the temporal discretization .
multirate time discretizations or higherorder space - time galerkin method have been also proposed for the linear biot model in @xcite and @xcite , respectively .
we propose two new iterative methods for solving the resulting non - linear equations at the new time - level : a splitting algorithm based on a combination between the undrained split and the fixed stress methods ( as mentioned above ) and a monolithic approach based on the same linearization technique .
the existence and uniqueness of a solution for the both formulations , as well as their global , linear convergence are rigorously shown . to the best of our knowledge ,
these are the first rigorous convergence results for iterative ( monolithic or splitting ) schemes in the non - linear case . at the same time
, we also acknowledge that while the non - linearities considered here are representative for compressible materials , they are simpler than those encountered in more complex applications such as multiphase flow .
to summarise , the new contributions of this paper are * we propose splitting and a monolithic l - scheme for a non - linear biot model . * the linear convergence of both schemes is rigorously shown in energy norms .
* we provide a benchmark for the convergence of splitting algorithms for the non - linear biot model , including a comprehensive comparison between the splitting and monolithic l - scheme .
the paper is structured as follows . in the next section we present the mathematical model and the discretization , including the fully discrete computational schemes . in section [ sec : convergence ] we analyse rigorously the proposed schemes .
numerical results are presented in section [ sec : numericalresults ] , and we conclude the paper in section [ sec : conclusions ] .
in this paper we use typical notations from functional analysis .
let @xmath2 be an open and bounded domain with a lipschitz continuous boundary @xmath3 , with @xmath4 being the dimension of the space .
we denote by @xmath5 the space of square integrable functions and by @xmath6 the sobolev space @xmath7 furthermore , @xmath8 will be the space of functions in @xmath6 vanishing on @xmath3 and @xmath9 the space of vector valued function having all the components and the divergence in @xmath5
. we will use bold face notation to specify when dealing with vectors .
we denote by @xmath10 the inner product on @xmath5 and @xmath11 the associated norm .
let further @xmath12 $ ] be a time interval , with @xmath13 denoting the final computational time .
the notations for the variables and parameters of biot s model are summarized in table [ table : nomenclature ] .
.nomenclature [ cols="<,>,^,<,>",options="header " , ] the influence of nonlinearities on the convergence is smaller for the test problem 1 , see figures [ mfl1l2_iterations]-[mfl1l2_iterations3 ] .
moreover , both schemes are even more sensitive on the choice of the tuning parameter @xmath14 .
we point out that the schemes are converging also for @xmath15 , but the convergence is slow ( around 200 iterations ) .
the schemes are performing similarly .
figures [ dh_3 ] - [ afl1l2_iterations10 ] illustrates the influence of the mesh size , time step , value of permeability @xmath16 and biot s coupling coefficient on the convergence of the schemes for test problem 2 .
we observe , again according to the theory , that a higher time step or higher permeability imply a faster convergence .
the schemes are converging faster when the time step increases and that the convergence is not depending of the mesh diameter , confirming again the theory .
we remark a slight increase in the number of iterations for an increasing biot coefficient @xmath17 , see figures [ afl1l2_iterations3 ] and [ afl1l2_iterations10 ] . 0.5 0.5 and @xmath14 for test problem 2 , case 1 : @xmath18 @xmath19.,title="fig : " ] 0.5 0.5 and @xmath14 for test problem 2 ,
case 2 : @xmath20{p}}{m};\ $ ] @xmath21{({\nabla\cdot{\vec{u}}})^5}$].,title="fig : " ] 0.5 0.5 and @xmath14 for test problem 2 , case 3 : @xmath22 @xmath21{({\nabla\cdot{\vec{u}}})^5}$].,title="fig : " ] 0.5 0.5 @xmath23.,title="fig : " ] 0.5 0.5 { p}}{m};\ $ ] @xmath24{({\nabla\cdot{\vec{u}}})^5 } $ ] .,title="fig : " ] 0.5 0.5 , @xmath16 for test problem 2 , case 2 : @xmath20{p}}{m};\ $ ] @xmath25{({\nabla\cdot{\vec{u}}})^5})$].,title="fig : " ] 0.5 0.5 , @xmath16 for test problem 2 , case 3 : @xmath22 @xmath21{({\nabla\cdot{\vec{u}}})^5}$].,title="fig : " ] 0.5 0.5 , @xmath17 for test problem 2 , case 2 : @xmath20{p}}{m};;\ $ ] @xmath26{({\nabla\cdot{\vec{u}}})^5}$].,title="fig : " ] 0.5 0.5 , @xmath17 for test problem 2 , case 3 : @xmath22 @xmath21{({\nabla\cdot{\vec{u}}})^5}$].,title="fig : " ]
we have proposed two linearisation schemes for the non - linear biot model , a monolithic l - scheme and a splitting l - scheme .
the convergence of both schemes has been rigorously shown , similar techniques as in @xcite and @xcite being involved .
the schemes are linearly , but global convergent and they are not involving the computations on any derivatives ( as in the case of the newton method ) . two illustrative numerical examples , an academic one and a non - linear extension of mandel s problem were implemented for testing the performance of the schemes . to summarise , we make the following remarks : * both schemes are very robust with respect to the choice of the tuning parameter , the mesh size and time step size . *
the tuning parameters @xmath27 have a strong influence on the speed of the convergence , with @xmath14 being the dominant one .
* the two schemes ( splitting and monolithic ) performed similarly with respect to parameters @xmath27 , cpu time and number of iterations when no preconditioning was applied .
when preconditioned , the splitting l - scheme is faster . *
the splitting l - scheme can be used both as a robust solver or as a preconditioner to improve the performance of a monolithic solver * the convergence of the schemes is faster for higher permeability . *
the convergence of the schemes is almost independent of the mesh size , and varies only slightly with the biot coupling parameter . according to the results in section [ sec : convergence ]
, we should observe a faster convergence for the pressure variable when we increase the time step size .
nevertheless , there is no indication on how the time step size is affecting the coupled problem .
numerically , we observed decreasing the time step only for the mandel s problem .
it was not observed a clear tendency for example one .
the research was supported by the university of bergen in cooperation with the fme - success center ( grant 193825/s60 ) funded by the research council of norway .
the work has also been partly supported by the nfr - daadppp grant 255715 and the nfr - toppforsk project 250223 .
almani , t. , kumar , k. , dogru , a.h . , singh , g. , wheeler , m.f .
: convergence analysis of multirate fixed - stress split iterative schemes for coupling flow with geomechanics .
comput . methods .
mech . eng . * 311 * , 180207 ( 2016 ) .
armero , f. : formulation and finite element implementation of a multiplicative model of coupled poro - plasticity at finite strains under fully saturated conditions .
comput . methods .
eng . * 171*(3 ) , 205241 ( 1999 ) .
castelletto , n. , white , j.a . ,
tchelepi , h.a . : accuracy and convergence properties of the fixed - stress iterative solution of two - way coupled poromechanics .
geomech . *
39*(14 ) , 15931618 ( 2015 ) .
chin , l.y . , thomas , l.k . , sylte , j.e . ,
pierson , r.g .
: iterative coupled analysis of geomechanics and fluid flow for rock compaction in reservoir simulation .
oil & gas sci . technol . * 57*(5 ) , 485497 ( 2002 ) .
doster , f. , nordbotten , j.m . : full pressure coupling for geo - mechanical multi - phase multi - component flow simulations , paper spe 173232 presented at the spe reservoir simulation symposium , houston ( 2015 ) .
haga , j.b . ,
osnes , h. , langtangen , h.p . :
efficient block preconditioners for the coupled equations of pressure and deformation in highly discontinuous media .
geomech . * 35*(13 ) , 14661482 ( 2011 ) .
kim , j. , tchelepi , h. , juanes , r. : stability and convergence of sequential methods for coupled flow and geomechanics : drained and undrained splits .
comput . methods .
. eng . * 200*(2324 ) , 20942116 ( 2011 ) .
kim , j. , tchelepi , h. , juanes , r. : stability and convergence of sequential methods for coupled flow and geomechanics : fixed - stress and fixed - strain splits .
comput . methods .
. eng . * 200*(1316 ) , 15911606 ( 2011 ) .
radu , f.a . ,
nordbotten , j.m . ,
pop , i.s . ,
kumar , k. : a robust linearization scheme for finite volume based discretizations for simulation of two - phase flow in porous media .
j. comput .
appl . math . *
289 * , 134141 ( 2015 ) .
wan , j. , durlofsky , l. , hughes , t. , aziz , k. : stabilized finite element methods for coupled geomechanics -reservoir flow simulations , paper spe 79694 presented at the spe reservoir simulation symposium , houston ( 2003 ) . | poromechanics coupled flow mechanics of porous media crucial in societal applications include geothermal energy extraction energy storage sequestration understanding biological tissues increased role computers for industrial enhances need for improved mathematical models numerical solvers for poromechanics common model for flow mechanics is linear quasi - stationary biot model @xcite two coupled partial differential equations balance of forces for mechanics conservation of mass for single phase flow non - linear extensions considered in @xcite non - stationary biot - allard model coupled equations impossible to solve analytically challenging numerically only simulation with fully coupled fluid potential mechanical deformation accurate for non - linear poro - mechanical processes @xcite complexities monolithic solvers linear industry standard weakly coupled approaches de - coupled approaches lead to numerical diffusion mask numerical incompatibilities between spatial objective paper show simple linearization technique@xmath1-scheme application richards equation two - phase flow porous media combined with splitting algorithm undrained split fixed stress method robust efficient iterative scheme for solving non - linear biot model linear conformal galerkin elements for space mechanics equation mixed finite elements for flow equation @xcite lowest order raviart - thomas elements used @xcite solution strategy applicable to other spatial discretizations backward euler used for temporal discretization multirate time discretizations higherorder space - time galerkin method proposed for linear biot model in @xcite propose two new iterative methods for solving non - linear equations time - level splitting algorithm undrained split fixed stress methods monolithic approach same linearization technique uniqueness solution for both formulations global linear convergence shown first rigorous convergence results for iterative monolithic splitting schemes in non - linear case non - linearities representative for compressible materials simpler than complex applications multiphase flow new contributions propose splitting monolithic l - scheme for non - linear biot model linear convergence of shown in energy normsprovide benchmark for convergence splitting algorithms non - linear biot model comparison between splitting monolithic l - scheme paper structured next section present mathematical model discretization fully discrete computational schemes section convergence analyse proposed schemes numerical results section conclude paper conclusions use typical notations functional analysis @xmath2 open bounded domain lipschitz continuous boundary @xmath3 @xmath4 dimension space denote @xmath5 space of square integrable functions @xmath6 sobolev space @xmath7 @xmath8 space of functions @xmath6 @xmath3 @xmath9 space vector valued function components divergence in @xmath5 use bold face notation dealing vectors denote @xmath10 inner product @xmath5 @xmath11 associated norm @xmath12 time interval @xmath13 final computational time notations variables parameters biot s model summarized in table nomenclature influence of nonlinearities on convergence smaller for test problem 1 figures- both schemes sensitive on choice tuning parameter @xmath14schemes converging for @xmath15 slow around 200 iterations schemes performing similarly figures dh_3 - afl1l2_iterations10 ] influence mesh size time step permeability @xmath16 biot s coupling coefficient on convergence test problem 2 higher time step higher permeability faster convergence faster time step increases not mesh diameter theory slight increase in iterations for increasing biot coefficient @xmath17 figures afl1l2_iterations3 afl1l2_iterations10 0.5 0.5 @xmath14 test problem 2 case 1 : @xmath18 @xmath19. 0.5 0.5 @xmath14 problem 2 2 @xmath20{p}}{m @xmath21{^5} 0.5 0.5 @xmath14 test problem 2 case 3 : @xmath22 @xmath21 0.5 0.5 @xmath23. 0.5 0.5 @xmath24{ 0.5 0.5 @xmath16 test problem 2 case 2 : @xmath20{p}}{m} @xmath25{^5}$0.5 0.5 @xmath16 test problem 2 case 3 @xmath22 @xmath21{\cdot\vec{u)^5}$ 0.5 0.5 @xmath17 test problem 2 case 2 @xmath20{p}}{m} @xmath26{{\nabla\cdot{\vec{u}}})^5}$ 0.5 0.5 @xmath17 test problem 2 case 3 @xmath22 @xmath21{{u}}})^5}$ proposed two linearisation schemes non - linear biot model monolithic l - scheme splitting l - scheme convergence schemes shown similar @xcite schemes linearly global convergent not computations derivatives newton method two numerical examples academic one non - linear extension mandel s problem testing performance both schemes robust tuning parameter mesh size time step size tuning parameters @xmath27 speed convergence @xmath14 dominant two schemes splitting monolithic performed similarly parameters @xmath27 cpu time number iterations no preconditioning preconditioned splitting l - scheme fastersplitting l - scheme used robust solver or preconditioner performance monolithic solver convergence faster for higher permeability independent of mesh size varies slightly with biot coupling parameter faster convergence for pressure variable time step size no indication time step size coupled problem decreasing time step only for mandel s problem clear tendency for example one research supported by university of bergen fme - success center grant 193825/s60 ) funded research council of norway partly supported by nfr - daadppp grant 255715 nfr - toppforsk project 250223 almani kumar dogru wheeler convergence analysis of multirate fixed - stress split iterative schemes for coupling flow with geomechanics 311 180207 2016 ) armero formulation multiplicative model of coupled poro - plasticity at finite strains under fully saturated conditions 171 205241 1999 castelletto white tchelepi accuracy convergence properties fixed - stress iterative solution of two - way coupled poromechanics 39*(14 15931618 ( 2015 ) chinthomas l sylte pierson r.g iterative analysis geomechanics fluid flow rock compaction reservoir simulation oil gas sci technol 57*(5 ) 485497 2002 ) doster nordbotten j.m full pressure coupling geo - mechanical - phase flow simulations paper spe 173232 reservoir simulation symposium houston 2015 haga j osnes langtangen h.p efficient block preconditioners equations pressure deformation discontinuous media geomech 35*(13 ) 14661482 2011 ) kim j tchelepi juanes stability convergence methods coupled flow geomechanics drained undrained splits methods 200*(2324 ) 20942116 2011 ) kim tchelepi juanes stability methods flow fixed - stress fixed - strain splits 200*(1316 ) 15911606 2011 radu nordbotten j pop kumar k robust linearization scheme finite volume discretizations two - phase flow porous media comput appl math, 134141 ( 2015 ) wan durlofsky hughes t.aziz stabilized element methods geomechanics flow simulations paper 79694 reservoir simulation symposium houston 2003 | 0.504607 | 2,980 | 1,498 | , Saving $0.1 in GPT-4. |
155,216 | we show that if a flavour - less vector meson remains bound after deconfinement , and if its limiting velocity in the quark - gluon plasma is subluminal , then this meson produces a distinct peak in the spectrum of thermal photons emitted by the plasma . we also demonstrate that this effect is a universal property of all strongly coupled , large-@xmath0 plasmas with a gravity dual . for the @xmath1
the corresponding peak lies between 3 and 5 gev and could be observed at the lhc . | in recent years several experimental and theoretical considerations indicate that the qgp right above the deconfining phase transition does not behave like a weakly coupled gas of quarks and gluons but that the interactions among plasma constituents are not small .
this observation has lead to increasing interest in the dynamics of strongly coupled plasmas analyzed via the ads / cft correspondence . in these proceedings
we report about a generic feature of all strongly coupled plasmas with a gravity dual @xcite .
our discussion is based on two assumptions about the properties of heavy mesons in a deconfined medium which are fulfilled by all gauge theories with a gravity dual : 1 .
heavy vector mesons such as the @xmath2 remain bound above the deconfinement transition . even though this is the subject of intense theoretical research ,
studies of lattice data indicate that this might be the case @xcite 2 .
the in - medium dispersion relation of these mesons is significantly modified and they posses a ( sub - luminal ) maximum speed of propagation .
we will also need to assume that the dispersion relation extends to sufficiently large momentum and that the meson states remain sufficiently narrow .
these assumptions are enough to describe the main features of the in - medium dispersion relation ; its generic form is shown in a ) . at high momentum ,
the dispersion relation approaches a constant slope , which determines the in - medium limiting velocity @xmath3 . at small momentum
the dispersion relation approaches a finite value of @xmath4 , the in medium mass . as a consequence of this asymptotic behavior ,
the dispersion relation crosses the light - cone at a position @xmath5 .
note that this crossing does not mean that heavy quark becomes massless at this point ; the dispersion relation is very different from the vacuum one and there is no reason why the invariant @xmath6 should be considered as the mass . at the point
@xmath7 the in - medium @xmath2 has the same quantum numbers of the photon ; thus , the electromagnetic interaction leads to mixing of these two degenerate states .
what this means is that at @xmath7 the in - medium @xmath2 can decay into an _ on shell _ photon .
this is different from the standard @xmath2 decay into dileptons , in which the @xmath2 leads to an _ off shell _ photon of virtuality of the @xmath2 mass .
the mixing between the in - medium @xmath2 and the on shell photon leads to a peak in the current - curren correlator at null momentum @xmath8 at @xmath9 .
a direct consequence of this observation is that if the dispersion relation of mesons is of the type described , a quark gluon plasma at finite temperature with a @xmath2 in kinetic equilibrium would lead to an almost monochromatic peak of photons at @xmath7 with a width given by the in - medium width of the @xmath2 .
sym plasma at strong coupling @xcite .
the black straight line corresponds to @xmath10 .
right : thermal photon spectrum for lhc energies with arbitrary normalization .
the continuous , monotonically decreasing , blue curve is the background from light quarks .
the continuous , red curve is the signal from @xmath2 mesons .
the dashed , black curve is the sum of the two.[thefigures],title="fig : " ] sym plasma at strong coupling @xcite .
the black straight line corresponds to @xmath10 .
right : thermal photon spectrum for lhc energies with arbitrary normalization .
the continuous , monotonically decreasing , blue curve is the background from light quarks .
the continuous , red curve is the signal from @xmath2 mesons .
the dashed , black curve is the sum of the two.[thefigures],title="fig : " ]
we can find an example in which the assumed features of the meson dispersion relation are realized : @xmath11 super - symmetric yang mills with @xmath12 flavors in the @xmath13 , @xmath14 limit with @xmath15 at finite temperature . in this case , the dynamics can be described by a dual gravitational theory with @xmath16 d7 probe branes ; the fact that @xmath15 implies that the back - reaction of the branes can be neglected and the metric is that of thermal @xmath11 without d7 branes which in suitable coordinates is given by ds^2= ( -f(z ) dt^2 + + dx^2 ) + l^2 d_5 with @xmath17 and t is the temperature .
the d7 branes span the coordinates @xmath18 and a 3-sphere within the @xmath19 .
the d7 branes can end `` in thin air '' at a scale @xmath20 ; if @xmath21 the branes are outside the black hole ( minkowski embedding ) . in this case , the small excitations of the brane correspond to ( infinitely narrow ) mesons in the dual gauge theory and the scale @xmath22 determines their mass .
if the temperature is higher , the brane falls inside of the black - hole and there are no mesons excitations ( black - hole embedding ) @xcite .
the d3-d7 model in the minkowski phase constitutes an interesting model to study the dynamics of meson in a strongly coupled , deconfined gauge theory plasma . in the minkowski embedding , due to the gravitational attraction
, the meson wave function is concentrated around the tip of the brane , @xmath20 . as a consequence of the gravitational redshift in the presence of the black hole , the speed of light in the @xmath23 hyperplane is reduced and it is given by [ eqvl ] = = 1-(t ) ^4 thus , meson excitations in this theory can not move faster than @xmath3 .
this observation provides a simple geometrical interpretation for the meson dispersion relation computed in the d3/d7 model a ) .
we now note that discussion above is common to all strongly coupled , large @xmath0 gauge theory plasmas with a gravity dual since in these theories a finite number of flavors @xmath12 is described by @xmath12 d - brane probes and since the deconfined phase of these theories is described by background with a black - hole .
since at the black - hole horizon the time component of the metric @xmath24 the limiting velocity of the mesons excitations is sub - luminal .
thus , if qcd has a gravity dual in the large @xmath0 limit , the dispersion relation of heavy mesons ( and in particular heavy vector mesons ) must cross the lightcone leading to a peak on the photon spectrum as discussed in the previous section .
we now address whether such a peak could be observed in a relativistic nucleus - nucleus collision . the natural vector meson candidates to consider are the @xmath2 and the @xmath25
since they may survive the deconfining transition ( we have not considered the @xmath26 since its survival is more controversial ) .
the main difference with the analysis above is that in a collision the temperature of the plasma is not fixed , but it evolves as the plasma expands and cools . as a consequence , the energy at which the peak would occur changes : it grows as the temperature is reduced . depending on
how fast the expansion happens , the peak may be transformed into an enhancement in the neighborhood of the meson mass .
another difficulty for the observation of the peak is the existence of more sources for photons such as the thermal emission from light degrees of freedom and the prompt photon emission ( perturbative contribution ) .
addressing the exact magnitude of the effect in a nuclear collision is a complicated task that demands an accurate model of the dynamics of the medium , the dynamics of mesons in the plasma and the meson production mechanism .
however there are significant theoretical uncertainties in these issues , specially in the interaction of meson with the plasmas ; there are different theoretical models that address the @xmath2 suppression pattern .
thus , we will not try to improve the current description of in - medium meson , but we will construct a model in which the peak can be observed in the final photon distribution .
instead of embedding the thermal rate into a hydrodynamical calculation , for this estimate we use a fireball model in which the volume of the system expands as @xmath27 and , in the qgp phase the product @xmath28 is constant .
we will assume that the initial temperature of the fireball is larger than the meson dissociation temperature .
we first consider the @xmath29 since the thermal photon emission in the region of @xmath30 gev is dominant at the lhc .
inspired by statistical recombination @xcite , we assume that the @xmath31-pairs produced in the initial collisions become kinetically equilibrated in the qgp but their total number , @xmath32 , stays constant
. this condition is implemented via a fugacity factor g_c(t ) = , where @xmath33 is the in - medium charm - quark mass . at @xmath34
the temperature reaches the dissociation temperature of the @xmath29 in the medium and a fraction of the @xmath31-pairs recombine forming @xmath29 mesons .
their contribution to the total number of photons is s ( ) _ t_diss^t_hadro dt v(t ) g_c(t(t))^2 e^-/t(t ) _ j/ ( , t(t ) ) , [ signal ] where the ` s ' stands for ` signal ' .
the overall normalization is not important since it is the same as that for the background of thermal photons emitted by the light quarks , which is given by b ( ) _ 0^t_hadro dt v(t ) e^-/t(t ) .
[ b ] we have omitted @xmath35 since , guided by the results for plasmas with a gravity dual , we assume it is structureless . we model the spectral function for the @xmath2 by a unit - area gaussian distribution of width @xmath36 mev centered at @xmath37 .
the dispersion relation we use is a fit to a ) and the @xmath38 .
the @xmath2 dissociation temperature is @xmath39 .
the results of numerically evaluating and for lhc values of the parameters are plotted in b ) where the effect of the @xmath2 peak is clearly seen on the spectrum . however
, this figure should be taken as illustrative , since the precise form is very sensitive to the value of the parameter used .
the enhancement is , in particular , is quadratically sensitive to @xmath38 ; thus , our estimates indicate that at rhic , where @xmath40 is ten times smaller , the enhancement can not be observed .
we have also explored the possibility of a similar enhancement for the @xmath25 meson at @xmath41 gev ; however at these energies the thermal photon contribution is exponentially suppressed as compare to pqcd photons . as a conclusion
, we have seen that a modified dispersion relation of the type shown in a ) for the @xmath2 leads , under certain assumptions , to a distinct peak in the thermal photon yield at the lhc in the region of @xmath42 gev .
this dispersion relation is expected in all strongly coupled gauge theory plasma , so its observation would be in agreement with the interpretation of the qgp as strongly coupled . unfortunately , there are many model assumptions in the analysis and not observing this peak would not necessarily imply a weakly coupled picture of the qgp .
this work was done in collaboration with d. mateos and it has been supported by a marie curie intra - european fellowship of the european community s seventh framework programme under contract number ( pief - ga-2008 - 220207 ) | experimental considerations indicate qgp above deconfining phase transition like weakly coupled gas quarks gluons interactions among plasma constituents not small interest in dynamics strongly coupled plasmas / cft correspondence proceedings report generic feature strongly coupled plasmas gravity dual @xcite discussion based on two assumptions heavy mesons in deconfined medium fulfilled by theories heavy vector mesons @xmath2 remain bound above deconfinement transition data indicate might be case @xcite 2 in - medium dispersion relation modified sub - luminal maximum speed propagation dispersion relation extends to large momentum meson states remain narrow assumptions describe main features - medium dispersion relation form shown high momentum dispersion relation approaches constant slope determines - medium limiting velocity @xmath3 small momentum approaches finite value @xmath4 dispersion relation crosses light - cone at @xmath5 crossing heavy quark becomes massless dispersion relation different from vacuum no reason invariant @xmath6 considered as mass at point @xmath7 in - medium @xmath2 has same quantum numbers of photon electromagnetic interaction leads to mixing degenerate statesat @xmath7 in - medium @xmath2 into on shell photon different from standard @xmath2 decay into dileptons leads to off shell photon mixing between @xmath2 on shell photon leads to peak in current - curren correlator at null momentum @xmath8 @xmath9 dispersion relation of quark gluon plasma at finite temperature with @xmath2 in kinetic equilibrium to monochromatic peak photons at @xmath7 with width by in - medium width @xmath2 sym plasma at strong coupling @xcite black straight line to @xmath10 thermal photon spectrum for continuous blue curve background from light quarks red signal from @xmath2 mesons black sum sym plasma at strong coupling @xcite black straight line @xmath10 thermal photon spectrum blue curve background from light quarks signal from @xmath2 mesons black curve sum of twoexample assumed features meson dispersion relation @xmath11 super - symmetric yang mills @xmath12 flavors limit @xmath15 finite temperature dynamics described dual gravitational theory @xmath16 d7 probe branes @xmath15 implies back - reaction neglected thermal @xmath11 without d7 branes ds^2= ( -f(z ) dt^2 + dx^2 ) + l^2 d_5 @xmath17 t temperature d7 branes span coordinates @xmath18 3-sphere @xmath19 d7 branes end thin air @xmath20 @xmath21 branes outside black hole minkowski embedding small excitations brane correspond narrow mesons dual gauge theory scale @xmath22 determines mass temperature higher brane falls inside black - hole no mesons excitations black - hole embedding d3-d7 model minkowski phase dynamics meson coupled deconfined gauge theory plasma minkowski embedding gravitational attraction meson wave function concentrated around tip brane gravitational redshift black hole speed light @xmath23 hyperplane reduced given = 1-(t ) ^4 meson excitations move faster than @xmath3 observation provides geometrical interpretation for meson dispersion relation in d3/d7 model discussion common to large @xmath0 gauge theory plasmas with gravity dual finite number flavors @xmath12 described by d - brane probes deconfined phase described by background black - hole at black - hole horizon time component of metric @xmath24 limiting velocity mesons excitations sub - luminal if qcd has gravity dual in large @xmath0 limit dispersion relation of heavy mesons must cross lightcone leading to peak on photon spectrum address peak in relativistic nucleus - nucleus collision natural vector meson candidates are @xmath2 @xmath25 may survive deconfining transition not considered @xmath26 survival controversial in collision temperature plasma not fixed evolves as expands cools energy peak changes grows as temperature reduced expansion peak may transformed into enhancement in meson mass difficulty for observation peak more sources for photons thermal emission prompt photon emission addressing exact magnitude effect in nuclear collision complicated demands accurate model of dynamics medium mesons plasma meson production mechanismsignificant theoretical uncertainties specially interaction meson with plasmas different models address @xmath2 suppression pattern improve description in - medium meson construct model peak in final photon distribution thermal rate fireball model volume expands as @xmath27 product @xmath28 constant initial temperature fireball larger than meson dissociation temperature consider @xmath29 thermal photon emission @xmath30 gev dominant at lhc @xmath31-pairs initial collisions kinetically equilibrated in qgp total number @xmath32 stays constant implemented via fugacity factor g_c(t ) = , @xmath33 in - medium charm - quark mass at @xmath34 temperature reaches dissociation temperature @xmath29 fraction @xmath31-pairs recombine @xmath29 mesons contribution to total number photons is s ( ) _ t_diss^t_hadro dt v(t ) g_c(t^2 e^-/t(t ) _ signal overall normalization not important same as background thermal photons light quarks given by b ( _^t_hadro dt v(t ) e^-/t(t ) omitted @xmath35 structureless model spectral function @xmath2 by unit - area gaussian distribution of width @xmath36 mev centered at @xmath37 dispersion relation to a ) @xmath38 @xmath2 dissociation temperature is @xmath39 results numerically evaluating lhc values plotted in b ) effect @xmath2 peak seen on spectrum figure illustrative form sensitive to value parameter enhancement quadratically sensitive to @xmath38 estimates indicate at rhic @xmath40 ten times smaller enhancement not observed explored similar enhancement for @xmath25 meson at @xmath41 gev thermal photon contribution suppressed modified dispersion relation for @xmath2 leads to distinct peak in thermal photon yield at lhc region @xmath42 gev dispersion relation expected in strongly coupled gauge theory plasma observation with interpretation qgp strongly coupled many model assumptions not observing peak imply weakly coupled picture qgp work with d. mateos supported by marie curie intra - european fellowship s seventh framework programme contract number ( pief - ga-2008 - 220207 ) | 0.361471 | 2,508 | 1,283 | , Saving $0.1 in GPT-4. |
155,218 | a current status of the observations with interferometer fabry - perot ( ifp ) at the 6 m telescope are described .
a reduction of the data obtaining with a new focal reducer scorpio in ifp mode are outlined . | the 2d spectroscopy observations with ifp at the 6 m telescope using the cigale system were begun in early 80s by teams from marseille observatory ( france ) and from sao ( boulesteix et al . , 1982 ; amram et al . ,
1992 ) . in 1997
a ccd was attached on the old focal reducer instead of photon counter .
a new multi - mode focal reducer scorpio ( spectral camera with optical reducer for photometrical and interferometrical observations ) was developed in the sao ras in 2000 ( afanasiev et al .
, this proceeding ) .
here we describe only fabry - perot mode of the scorpio . for more details
see our www - site + ` http://www.sao.ru/~moisav/scorpio/scorpio.html ` .
the optics of scorpio compensates the aberrations of the main telescope mirror and provide a total focal ration @xmath0 .
the queensgate ifps et-50 installs into the 35 mm collimated beam .
scorpio are under full remote control from a pc .
a total dqe ( telescope@xmath1filter@xmath1ifp@xmath1ccd ) is about @xmath2 in the @xmath3 spectral region .
scorpio equipped with narrow filters ( @xmath4 ) for ifp observations in the @xmath3 line for object s velocities @xmath5 .
focal reducer mounted on the prime focus universal adapter which has two movable probes for offset guiding and continuum and spectral lamps for calibration . a he - ne - ar lamp and continuum spectrum lamp mounted in the integration sphere which produced a uniform lighting .
the optics of calibration system forms the @xmath6 beam on the reducer entrance .
there is no systematic shift in velocities because telecentrical beams used for wavelength calibration .
a ccd has a large advantage in quantum effeciency in comparison with a photon counter .
however a readout noise and a reading time do not allow to do very short exposures of the object channels .
therefore exist a problem of the photometry correction of flux and seeing in the individual channels because only 1 - 2 scanning cycles per channel may be accepted on ccd observations , we should take in account a time dependence of a sky brightness and of an atmospheric extinction .
therefore wavelength object cube contains a set of narrow rings ( fig .
this pattern limits the velocities measurements for the low brightness objects .
we suggest a method to resolve this problem .
main idea is removing the sky counts _ before _ cube linerization .
after standard ccd - reduction a mean radial profile of the sky lines is constructed by the azimuthal averaging of the interferometrical rings ( fig .
the sky radial profile are subtracted from the object .
1b shows example of the sky subtracted images .
variations of an atmospheric extinction , seeing and channel offset shifts could be measured from the photometry of background stars and corrected in the cube .
unfortunately our method requires that about half of field of view will be free from the object emission , but it allows to obtain a good results in the object cube . | 2d spectroscopy observations ifp 6 m telescope cigale system begun early 80s marseille observatory sao 1997 ccd attached old focal reducer photon counter new multi - mode focal reducer scorpio photometrical interferometrical observations developed sao ras 2000 fabry - perot mode scorpio details http://www.ru/scorpio optics scorpio compensates aberrations telescope provide total focal ration @xmath0 queensgate ifps et-50 installs into 35 mm collimated beam scorpio remote control from pc total dqe telescope @xmath2 @xmath3 spectral region scorpio narrow filters @xmath4 for ifp observations @xmath3 line object velocities focal reducer prime focus universal adapter two movable probes offset guiding continuum spectral lamps calibration he - ne - ar lamp continuum spectrum lamp integration sphere uniform lighting optics calibration system forms @xmath6 beam on reducer entrance no systematic shift in velocities telecentrical beams wavelength calibration ccd quantum effeciency photon counter readout noise reading time allow short exposuresproblem photometry correction flux seeing channels 1 - 2 scanning cycles per channel accepted ccd observations time dependence sky brightness atmospheric extinction wavelength object cube narrow rings limits velocities measurements low brightness objects suggest method resolve removing sky counts before cube linerization ccd - reduction mean radial profile sky lines constructed averaging interferometrical rings sky radial profile subtracted from object 1b sky subtracted images variations atmospheric extinction seeing channel offset shifts measured from photometry background stars corrected cube method requires half field of view free from object emission allows good results object cube | 0.132967 | 664 | 317 | , Saving $0.0 in GPT-4. |
155,221 | a dispersive - optical - model analysis has been performed for both protons and neutrons on @xmath0ca isotopes .
the fitted potentials describe accurately both scattering and bound quantities and extrapolate well to other stable nuclei .
further experimental information will be gathered to constrain extrapolations toward the driplines . | experiments with radioactive beams are unraveling the properties of nuclei at the limits of stability @xcite .
studies of low - energy processes and transfer reactions require optical potentials ( op ) suitable for dripline isotopes .
this is also important for astrophysics applications , where capture and decay processes require a consistent determination of both scattering and bound states .
dispersive optical models ( dom ) provide a useful tool that allows one to describe the mean field consistently over a broad energy range .
the dom is grounded in the theory of many - body green s functions ( or propagators ) @xcite and in the proven equivalence between the many - body self - energy ( of propagator theory ) and the feshbach theory of the microscopic optical model @xcite . in practice
the dom can be thought as a parametrization of the self - energy .
this has several advantages .
it allows us to develop an op that describes consistently both elastic scattering and bound - state properties .
the theory directly imposes important analytic relation between the real and imaginary components ( see below ) .
also microscopic calculations @xcite can be employed to determine the physical ingredients of the potentials .
the dom was developed by mahaux and collaborators for a number of individual magic or near magic nuclei @xcite .
recently , the authors of refs .
@xcite have performed a dom study for the chain of ca isotopes .
this employed the world data on @xmath1ca , including _ both _ elastic scattering ( up to 200 mev ) and bound - state information from ( @xmath2,@xmath3 ) experiments .
this analysis also discussed possible extrapolations toward the driplines and suggested new experiments to constrain the predictions for exotic nuclei @xcite .
a global parametrization of the coupled - channel dom was also reported in ref .
@xcite for stable nuclei .
the present talk discusses briefly the analysis of refs .
@xcite and points to open issues that are being addressed to refine the predictions for unstable isotopes .
ca , the data and curves are scaled down by an additional factor of 100.,title="fig:",scaledwidth=85.0% ] -.3 cm
while the nucleon self - energy is an _ energy dependent _ and _ non - local _ potential , the vast majority of dom applications employ a local approximation of it .
the same approach is followed here , although this approximation may be removed in future works .
mahaux and sartor @xcite start by separating the real part of the optical potential at the fermi surface ( @xmath4 ) and write the dom as @xmath5 where @xmath6 is the local equivalent of the self - energy , @xmath7 , at the fermi energy ( @xmath8 ) is not an hartree - fock potential but it does describe the effects of the nuclear mean - field .
we maintain this notation for consistency with the rest of the literature . ] .
normally , this would be a static potential but it acquires an energy - dependence to account for the non - locality of @xmath9 .
the real and imaginary parts of the dynamic components of the dom are linked by the subtracted _ dispersion relation _
, @xmath10 this relation is a direct consequence of the causality principle , which is therefore embedded in the model .
thus , the dom is fully determined by parametrizing only the @xmath6 and @xmath11 components . the explicit form employed in this work
is reported in detail in ref .
@xcite . , 1@xmath12 , and 0@xmath13 proton hole states for @xmath14ca ( left panels ) and @xmath15ca ( right panels ) .
the fitted quantities include the root - mean - squared radius @xmath16 , the spectroscopic factors @xmath17 and the widths @xmath18 of these states.,scaledwidth=100.0% ] , 1@xmath12 , and 0@xmath13 proton hole states for @xmath14ca ( left panels ) and @xmath15ca ( right panels ) .
the fitted quantities include the root - mean - squared radius @xmath16 , the spectroscopic factors @xmath17 and the widths @xmath18 of these states.,scaledwidth=100.0% ] [ fig : ay ] -0.3 cm the potential ( [ eq : dom ] ) was fitted to published data sets at both positive and negative energies .
these included elastic scattering data for @xmath19ca and @xmath20ca , with energies up to 200 mev , total reaction cross sections , experimental single - particle levels and spectroscopic factors and radii of proton bound states measured in ( @xmath2,@xmath3 ) reactions .
the quality of the fit to the differential cross sections and analysing powers is shown in figs .
[ fig : ds ] and [ fig : ay ] ( left ) . for proton bound states ,
only one data point could not be reproduced , i.e. , the @xmath12 orbit in @xmath14ca as shown in fig .
[ fig : ds ] ( right ) .
an independent re - analysis of ( @xmath2,@xmath3 ) data has later corrected this value and resolved this discrepancy @xcite .
for such a large body of data ( 81 data sets comprising 3569 data points ) , the excellent agreement of the fit with just 25 free parameters provides confidence in the predictive power of the dom calculations .
-.3 cm figure [ fig : ni ] shows a _
preliminary _ comparison between the elastic proton scattering on stable ni sotopes and the prediction of the above dom , which was fitted to ca data only .
the overall agreement is quite satisfying and shows the potentiality for global fits with this approach .
however , proper extrapolations to unstable isotopes ( and in particular close to the driplines ) require particular care for the n / z asymmetry dependence of the surface part of the imaginary potential .
this is usually assumed to behave as @xmath21 the choice for protons is @xmath22=+1 , which works quite well for @xmath23+@xmath1ca scattering . for neutrons ,
the standard assumption in the literature ( @xmath24=-1 ) leads to inconsistencies for neutron - rich isotopes .
similar difficulties have also been reported in a recent global op fit @xcite . unfortunately , the n+@xmath14ca scattering data solely available to ref .
@xcite are not sufficient to constrain @xmath24 . choosing @xmath24=0
was found to be consistent with spectroscopic factors from ( @xmath25,@xmath23 ) reactions but does not rule out other possibilities .
new data for the n+@xmath15ca have recently been taken to verify this assumption @xcite .
it is expected that this will allow new fits that improve the description of dripline isotopes .
this work is supported by the u.s .
department of energy , division of nuclear physics under grant de - fg02 - 87er-40316 and the u.s .
national science foundation under grant phy-0652900 .
c.b . acknowledges a kakenhi grant ( no .
21740213 ) form the japanese mext .
a. gade and t. glasmacher , prog .
. nucl . phys . * 60 * , 161 ( 2008 ) .
w. h. dickhoff and d. van neck , _ many - body theory exposed ! _
( 2nd edition , world scientific , singapore , 2008 ) .
f. capuzzi and c. mahaux , ann . phys . * 245 * , 147 ( 1996 ) . c. barbieri and and b. k. jennings , phys .
c * 72 * , 014613 ( 2005 ) . c. mahaux and r. sartor , adv .
. phys . * 20 * , 1 ( 1991 ) .
r. j. charity , l. g. sobotka , w. h. dickhoff , phys .
lett . * 97 * , 162503 ( 2006 ) .
r. j. charity , j. m. mueller , l. g. sobotka , w. h. dickhoff , phys .
c * 76 * , 044314 ( 2007 ) .
r. j. charity , l. g. sobotka _ et al .
_ , in preparation
. l. hao , w. sun , e. sh .
soukhovitskii , j. phys .
g * 35 * 095103 ( 2008 ) .
l. lapiks , priv .
comm . and to be published .
weppner , phys .
c * 80 * , 034608 ( 2009 ) . | experiments with radioactive beams properties nuclei at limits stability studies low - energy processes transfer reactions require optical potentials for dripline isotopes important for astrophysics applications capture decay require consistent determination scattering bound states dispersive optical models mean field over broad energy range grounded in theory of many - body green s functions equivalence between - body self - energy feshbach theory microscopic optical model dom of self - energy allows op elastic scattering bound - state properties theory imposes analytic relation between real imaginary components microscopic calculations determine physical ingredients of potentials dom developed by mahaux collaborators for individual magic near nuclei authors of performed dom study for chain of ca isotopes employed world data on @xmath1ca elastic scattering bound - state information from experiments analysis discussed extrapolations toward driplines suggested new experiments to predictions for exotic nuclei global parametrization of coupled - channel dom reported in for stable nuclei present talk discusses analysis of points to open issues to refine predictions for unstable isotopesdata curves scaled by factor 100:",scaledwidth=85.0% -.3 cm nucleon self - energy energy dependent non - local potential majority dom applications employ local approximation same approach followed may removed future works mahaux sartor optical potential at fermi surface @xmath4 write dom as @xmath5 @xmath6 local equivalent self - energy @xmath7 fermi energy @xmath8 not hartree - fock potential effects nuclear mean - field maintain notation for consistency literature static potential acquires energy - dependence non - locality @xmath9 real imaginary dynamic components dom linked by subtracted dispersion relation @xmath10 consequence causality principle embedded in model dom determined by parametrizing @xmath6 @xmath11 components explicit form reported ref. @xcite 1@xmath12 0@xmath13 proton hole states for @xmath14ca @xmath15ca fitted quantities include root - mean - radius @xmath16 spectroscopic factors @xmath17 widths @xmath18,scaledwidth=100.0% 1@xmath12 0@xmath13 proton hole states for @xmath14cafitted quantities include root - mean - radius spectroscopic factors widths=100.0% fig -0.3 cm potential dom fitted to data sets at positive negative energies included elastic scattering data for @xmath19ca @xmath20ca energies up to 200 mev reaction cross sections experimental single - particle levels spectroscopic factors radii of proton bound states in @xmath2,@xmath3 reactions quality fit differential cross sections analysing powers shown in figs . ds ay proton bound states one data point reproduced @xmath12 orbit in @xmath14ca fig independent re - analysis corrected resolved discrepancy large data 81 data sets 3569 data points agreement fit with 25 free parameters provides confidence in predictive power dom calculations -.3 cm figure shows comparison between elastic proton scattering on stable sotopes prediction dom fitted to ca data only agreement satisfying shows for global fits extrapolations to unstable isotopes require care for n / z asymmetry dependence surface imaginary potentialassumed @xmath21 protons @xmath22=+1 works for @xmath23+@xmath1ca scattering neutrons standard assumption @xmath24=-1 ) leads inconsistencies neutron - rich isotopes difficulties global op fit @xcite n+@xmath14ca scattering data not sufficient constrain @xmath24 @xmath24=0 consistent with spectroscopic factors @xmath25,@xmath23 reactions rule out possibilities new data n+@xmath15ca verify assumption expected new fits description dripline isotopes supported by u.s. department of energy division nuclear physics grant - fg02 - 87er-40316 u national science foundation grant phy-0652900 acknowledges kakenhi grant 21740213 ) japanese a. gade t. glasmacher nucl . 60 161 ( 2008 ) dickhoff d. van neck many - body theory exposed 2nd scientific 2008 f. capuzzi c. mahaux phys . 245 147 ( 1996 ) c barbieri jennings phys 72 014613 2005 mahaux r. sartor phys. 20 1 1991 charity sobotkadickhoff phys . 97 162503 ( 2006 ) j. charity j mueller l g. sobotka dickhoff phys . 76 044314 2007 ) j. charity l sobotka preparation. hao sun soukhovitskii . 35 * 095103 ( 2008 ) lapiks priv. comm published weppner . 80 * 034608 ( 2009 ) | 0.313515 | 1,951 | 957 | , Saving $0.1 in GPT-4. |
155,225 | motivated by experiments in nanoscopic systems , we study a generalized anderson , which consists of two spin degenerate doublets hybridized to a singlet by promotion of an electron to two conduction bands , as a function of the energy separation @xmath0 between both doublets . for @xmath0=0 or very large , the model is equivalent to a one - level su(n ) anderson model , with n=4 and 2 respectively .
we study the evolution of the spectral density for both doublets ( @xmath1 and @xmath2 ) and their width in the kondo limit as @xmath0 is varied , using the non - crossing approximation ( nca ) . as @xmath0 increases , the peak at the fermi energy in the spectral density ( kondo peak ) splits and the density of the doublet of higher energy @xmath2 shifts above the ferrmi energy . the kondo temperature @xmath3 ( determined by the half width at half maximum of the kondo peak in density of the doublet of lower energy @xmath1 ) decreases dramatically .
the variation of @xmath3 with @xmath0 is reproduced by a simple variational calculation . anderson model , spectral density , kondo temperature , non - crossing approximation 75.20.hr , 75.10.jm | the kondo effect , found originally for systems with magnetic impurities in metals is now present in a variety of nanoscopic systems , including semiconducting quantum dots @xcite , magnetic adatoms on surfaces @xcite and carbon nanotubes @xcite . in the latter , in addition to the spin kramers degeneracy ,
there is in addition orbital degeneracy due to the `` pseudospin '' degree of freedom related with the particular band structure of graphene .
this leads to an su(4 ) kondo effect which has been observed experimentally @xcite and also discussed theoretically @xcite .
in particular , lim _
et al . _ have studied the spectral density when the su(4 ) symmetry is reduced to su(2 ) , mainly by a change in the tunneling matrix elements @xcite .
our main motivation in the problem arises from interference phenomena in systems of quantum dots @xcite or molecules @xcite .
for example depressions in the integrated conductance through a ring described by the hubbard or @xmath4 model , pierced by an aharonov - bohm magnetic flux , related with spin - charge separation @xcite , are due to a partial destructive interference when the energy of two doublets cross @xcite .
similar interference effects were predicted in molecular transistors @xcite .
the effective hamiltonian near the crossing is discussed in the next section . to our knowlege
, the conductance in these systems has so far been calculated using approximate expressions or a slave - boson formalism @xcite , which is valid only for very low temperatures and applied bias voltages .
this work is a step towards a more quantitative theory to describe the transport through similar systems , treating the effective hamiltonian within the non - crossing approximation ( nca ) @xcite .
work is in progress to deal with the non - equilibrium situation , which is necessary within our formalism to calculate the current . in this paper
, we report on our study of the spectral density of the model as a function of the splitting @xmath5 between both doublets in the kondo regime .
we also calculate the dependence of the kondo temperature @xmath6 with @xmath5 and analyze the validity of the friedel sum rule @xcite .
we start from a model in which two doublets of an interacting system are hybridized with a singlet by promotion of an electron to two conducting leads .
this is the low - energy effective hamiltonian for several systems with partial destructive interference , such as aharonov - bohm rings @xcite or aromatic molecules @xcite connected to conducting leads .
the hamiltonian can be written as @xcite @xmath7 where the singlet @xmath8 and the two doublets @xmath9 ( @xmath10 ; @xmath11 or @xmath12 ) denote the localized states ( representing for example the low - energy states of an isolated molecule ) , @xmath13 create conduction states in the left ( @xmath14 ) or right ( @xmath15 ) lead , and @xmath16 describe the four hopping elements between the two leads and both doublets , assumed independent of @xmath17 . changing the phase of the conduction states and the relative phase between both doublets , three among the four @xmath16 can be made real and positive .
the phase @xmath18 of the remaining hopping @xmath19 , depends on the particular system and its symmetry .
for example in molecules with rotational symmetry @xmath20 , where @xmath21 is the distance between the sites connected to the left and right leads , and @xmath22 is the wave vector of the state @xmath23 , which can be modified with an applied magnetic flux @xcite . in absence of magnetic flux and when the relevant states have wave vector @xmath24 , as in rings with a number of atoms multiple of four , @xmath25 and there is complete destructive interference in transport @xcite . in the following
we will assume this case , with states 1 and 2 related by symmetry implying @xmath26 .
we further assume symmetric leads , @xmath27 , @xmath28 .
then , without loss of generality we can take @xmath29 , @xmath30 , and @xmath31 . for these parameters ,
changing basis @xmath32 , @xmath33 , the hamiltonian takes the form of an su(4 ) anderson model with a `` field '' @xmath34 and on - site hybridization @xmath35 @xmath36 interchanging the doublet index ( 1 or 2 ) with the spin index @xmath37 , one realizes that this model also describes transport trough a carbon nanotube with electron density depleated at two points ( so as to created an su(4 ) quantum dot in the middle ) under a real applied magnetic field .
@xmath38 , @xmath39 .
the lowest temperature is @xmath40 . , width=283 ] in fig .
[ dens1 ] we present numerical results for the spectral density @xmath41 of the su(4 ) anderson model ( @xmath42 ) in the kondo regime @xmath43 , where the hybridization function @xmath44 , assumed independent of energy .
we set @xmath45 as the unit of energy .
we also assume a conduction band symmetric around the fermi level @xmath46 of half width @xmath38 .
the spectral density shows a broad charge transfer peak near @xmath47 . for temperatures below a characteristic energy scale @xmath3 ( defined below )
, @xmath41 develops a narrow peak around the fermi level .
in contrast to the better known one - level su(2 ) case , this peak is displaced towards positive energies and is much broader , as discussed in section 5 . and @xmath48 .
other parameters as in fig .
1 . dot - dashed line corresponds to @xmath49 for @xmath50 , 0.3 and 0.6.,width=283 ] the evolution of the spectral densities as @xmath51 is displaced to larger energies , breaking the su(4 ) symmetry is shown in fig .
the peak near the fermi energy of @xmath52 is displaced towards positive energies ( near @xmath53 ) .
in contrast , the corresponding peak in @xmath54 narrows significantly and displaces towards the fermi energy .
this implies that the kondo temperature @xmath3 defined as the half width at half maximum of this peak , also decreases strongly .
the evolution of @xmath3 with @xmath0 is discussed in the section 5 .
in addition @xmath55 develops a broad peak near energy @xmath56 which becomes visible when @xmath0 becomes greater than @xmath3 .
the anderson model studied has a fermi liquid ground state which satisfies well known relationships at zero temperature .
one of them is the friedel sum rule which relates the spectral density at the fermi level for each `` pseudospin '' channel with the occupation of that channel @xcite . for the simplest case of a constant density of conduction states ,
this rule reads @xmath57 where @xmath58 .
this is an exact relationship for a fermi liquid , which is not necessarily satisfied by approximations . in particular , it is known that at very low temperatures , the nca has a tendency to develop spurious spikes in @xmath59 at the fermi energy , while thermodynamic properties , such as expectation values are accurately reproduced @xcite . in fig .
[ friedel ] , we compare both members of eq .
( [ fsr ] ) for the lowest lying doublet , at a temperature @xmath60 @xcite low enough so that no further increase in @xmath61 takes place as the temperature is lowered ( according to physical expectations and eq .
( [ fsr ] ) ) , but high enough to prevent the presence of spurious spikes .
the disagreement lies below 20 % .
the agreement improves as the parameters are moved deeper in the kondo regime @xmath62 .
thus , while the spectral density at zero temperature is not well represented by the nca results at @xmath63 , one can take the values at @xmath60 as a reasonable description of the correct @xmath63 ones .
this statement is supported by the comparison of results obtained by nca and numerical renormalization group for the one - level su(2 ) case @xcite . .
circles : corresponding ( more accurate ) result given by eq .
( [ fsr ] ) .
lines are guides to the eye .
parameters are @xmath60 and the rest as in fig .
from the half width at half maximum of the peak nearest to the fermi energy of the spectral density of the lowest level ( @xmath54 ) , we have calculated the kondo temperature of the system @xmath6 for several values of @xmath0 .
this requires to solve the self - consistent nca equations up to low enough temperatures ( about 0.1 @xmath3 as discussed above ) so that the height of the peak does not increase significantly with further lowering of the temperature @xcite .
fortunately , the result is not very sensitive to the ratio @xmath64 .
the results are shown in fig .
[ kondo ] and compared with eq .
( [ tk ] ) obtained from a variational calculation as explained below .
we see that except for an overall multiplicative factor , the agreement is quite good , in spite of the fact that @xmath6 changes by nearly two orders of magnitude . .
the temperatures used were @xmath65 , @xmath66 , and @xmath67 depending on @xmath0 dashed line : corresponding variational result eq .
( [ tk ] ) multiplied by a factor 0.606.,width=283 ] to provide an independent estimate of @xmath6 , we have calculated the stabilization energy of the following variational wave function @xmath68 where @xmath69 is the many - body singlet state with the filled fermi sea of conduction electrons and the state @xmath8 at the localized site , while @xmath70 and @xmath71 are variational parameters . from the resulting optimized energy @xmath72 , we can define the stabilization energy as @xmath73 .
the kondo energy scale defined in this way becomes @xmath74 + \delta ^{2}/4\right\ } ^{1/2}-\delta /2 .
\label{tk}\ ] ] this expression interpolates between the su(4 ) result ( @xmath75 $ ] for @xmath76 ) and the su(2 ) one for one doublet only ( @xmath77 $ ] for @xmath78 ) .
we have studied an impurity anderson model containing two doublets , which interpolates between the cases for one level with su(4 ) and su(2 ) symmetry , and is of interest for several nanoscopic systems , using the non - crossing approximation ( nca ) .
we have shown that the nca provides reasonable results for the equilibrium spectral density .
the values of the spectral density for both doublets agree within 20% with the predictions of the friedel sum rule , in spite of the fact that it is not expected to satisfy fermi liquid relationships at zero temperature .
in addition , the kondo temperature scale @xmath3 obtained from the width of the peak in spectral density near the fermi energy agrees very well with the stabilization energy of a variational calculation , in spite of the change in several orders of magnitude of @xmath3 when the splitting between both doublets is changed .
the approach seems promising for studying transport properties within a non - equilibrium formalism .
work in this direction is in progress .
one of us ( a. a. a. ) is partially supported by conicet , argentina .
this work was partially supported by pip no 11220080101821 of conicet , and pict nos 2006/483 and r1776 of the anpcyt . | kondo effect for magnetic impurities in metals present in nanoscopic systems including semiconducting quantum dots magnetic adatoms surfaces carbon nanotubes spin kramers degeneracy orbital degeneracy due to pseudospin freedom band structure graphene leads to su(4 ) kondo effect observed experimentally discussed theoretically lim studied spectral density when su(4 ) symmetry reduced to su(2 ) by change in tunneling matrix elements motivation from interference in quantum dots molecules depressions in integrated conductance through ring model by aharonov - bohm magnetic flux due to partial destructive interference when two doublets cross similar interference effects predicted in molecular transistors effective hamiltonian near crossing discussed in next section conductance in calculated using approximate expressions slave - boson formalism valid for low temperatures applied bias voltages work step towards quantitative theory transport through similar systems treating effective hamiltonian within non - crossing approximation work progress to with non - equilibrium situation to calculate current paper report study spectral density model splitting @xmath5 between doublets kondo regime calculate dependence kondo temperature @xmath6 with @xmath5 analyze validity friedel sum rule @xcite start model two doublets hybridized singlet promotion electron to two conducting leads low - energy hamiltonian for systems partial destructive interference aharonov - bohm rings aromatic molecules connected conducting leads hamiltonian written as @xcite @xmath7 singlet @xmath8 two doublets @xmath9 denote localized states @xmath13 create conduction states left or right lead @xmath16 describe four hopping elements between two leads doublets independent @xmath17 changing phase conduction states relative phase doublets three among four @xmath16 made real positive phase @xmath18 remaining hopping @xmath19 depends on system symmetry molecules rotational symmetry @xmath21 distance between sites connected left right leads @xmath22 wave vector state @xmath23 modified applied magnetic flux absence magnetic flux relevant states wave vector @xmath24 rings four complete destructive interference in transport @xciteassume case states 1 2 symmetry @xmath26 assume symmetric leads @xmath27 take @xmath29 changing basis @xmath32 hamiltonian su(4 ) anderson model field @xmath34 on - site hybridization @xmath35 @xmath36 interchanging doublet index 1 2 spin index @xmath37 model describes transport carbon nanotube electron density depleated su(4 ) quantum dot magnetic field @xmath38 lowest temperature @xmath40 fig . dens1 numerical results spectral density @xmath41 su(4 ) anderson model @xmath42 kondo regime @xmath43 hybridization function @xmath44 independent energy set @xmath45 unit energy assume conduction band symmetric fermi level @xmath46 half width @xmath38 spectral density broad charge transfer peak near @xmath47 temperatures below characteristic energy scale @xmath3 @xmath41 narrow peak around fermi level - su(2 ) peak displaced positive energies broader section 5 @xmath48 parameters fig . 1 dot - dashed line @xmath49 @xmath50 0.3 0.6.evolution spectral densities @xmath51 displaced to larger energies breaking su(4 ) symmetry shown in fig . peak near fermi energy @xmath52 displaced towards positive energies @xmath53 peak @xmath54 narrows displaces towards fermi energy kondo temperature @xmath3 decreases evolution of @xmath3 with @xmath0 discussed in section 5 @xmath55 develops broad peak near energy @xmath56 visible when @xmath0 greater than @xmath3 anderson model has fermi liquid ground state satisfies relationships at zero temperature friedel sum rule relates spectral density at fermi level for each pseudospin channel with occupation @xcite rule reads @xmath57 where @xmath58 exact relationship for fermi liquid not satisfied by approximations at low temperatures nca spurious spikes in @xmath59 at fermi energy thermodynamic properties reproduced @xcite in fig . compare members of eq . for lowest lying doublet at temperature @xmath60 @xcite low no increase in @xmath61 temperature spurious spikes disagreement below 20 %agreement improves parameters deeper in kondo regime @xmath62 spectral density at zero temperature not represented by nca results at @xmath63 values @xmath60 description correct @xmath63 ones supported by comparison results nca numerical renormalization group for one - level su(2 ) case @xcite circles result eq . fsr ] lines guides eye parameters @xmath60 rest fig . half width maximum peak energy spectral density lowest level @xmath54 calculated kondo temperature system @xmath6 for values @xmath0 requires solve self - consistent nca equations low temperatures 0.1 @xmath3 height peak increase lowering temperature @xcite result not sensitive to ratio @xmath64 results shown in fig . compared with eq . ( [ tk ] variational calculation agreement good @xmath6 changes two orders of magnitude temperatures used @xmath65 @xmath66 @xmath67 depending @xmath0 corresponding variational result eq .tk multiplied factor 0.606.,width=283 independent estimate @xmath6 calculated stabilization energy variational wave function @xmath68 @xmath69 many - body singlet state filled fermi sea conduction electrons @xmath8 localized site @xmath70 @xmath71 variational parameters optimized energy @xmath72 stabilization energy @xmath73 kondo energy scale becomes @xmath74 + \delta-\delta /2 expression interpolates between su(4 ) result @xmath75 @xmath76 su(2 ) doublet @xmath77 @xmath78 ) studied impurity anderson model two doublets interpolates cases one level su(4 ) su(2 ) symmetry interest nanoscopic systems non - crossing approximation nca nca provides reasonable results equilibrium spectral density values both doublets agree within 20% with predictions friedel sum rule not satisfy fermi liquid relationships at zero temperature kondo temperature scale @xmath3 peak spectral density near fermi energy agrees with stabilization energy variational calculation change @xmath3 splitting doublets changed approach promising studying transport properties non - equilibrium formalism work progresspartially supported conicet argentina supported pip 11220080101821 conicet pict 2006/483 r1776 anpcyt | 0.430984 | 2,589 | 1,333 | , Saving $0.1 in GPT-4. |
155,231 | linear predictive coders form an important class of speech coders .
this paper describes the software level implementation of linear prediction based vocoders , viz .
code excited linear prediction ( celp ) , low - delay celp ( ld - celp ) and mixed excitation linear prediction ( melp ) at bit rates of 4.8 kb / s , 16 kb / s and 2.4 kb / s respectively .
the c programs of the vocoders have been compiled and executed in linux platform .
subjective testing with the help of mean opinion score test has been performed .
waveform analysis has been done using praat and adobe audition software .
the results show that melp and celp produce comparable quality while the quality of ld - celp coder is much higher , at the expense of higher bit rate .
_ keywords - vocoder , linear prediction , code excited , low delay , mixed excitation , celp , ld - celp , melp , praat , linux _ | speech coding is the encoding of speech signals to enable transmission at bit rates lower than that of the original digitized speech .
the human auditory system can capture only certain aspects of a speech signal .
thus , perceptually relevant information of a speech signal can be extracted to produce an equivalent - sounding wave at a much lower bandwidth .
linear prediction is a widely used compression technique in which past samples of a signal are stored and used to predict the next sample @xcite . in the basic linear predictive coder prototype
, prediction is done over a time interval of one pitch period using adaptive linear delay and gain factors .
this basic prototype produces intelligible but artificial - sounding speech output , and various techniques have been researched to improve the perceptual quality .
variants to the linear prediction coders are code excited linear prediction ( celp ) and low - delay celp ( ld - celp ) which use forward and backward linear prediction respectively , along with the _ codebooks _ ,
i.e. lookup tables with codevectors corresponding to speech residual signals @xcite . in mixed excitation linear prediction ( melp )
@xcite , an additional classification of speech is introduced - the jittery voiced speech .
mixed excitation , i.e the mixing of periodic and noise excitation , is another distinguishing feature of the melp model .
this paper aims at comparing the three types of linear prediction based vocoders in terms of their bit rate and perceptual quality . in comparing vocoders ,
subjective testing of the voice quality is a major step in the evaluation of a vocoder .
a vocoder will finally be accepted in the market only if humans are satisfied with the voice quality .
keeping this fact in consideration , subjective evaluation using the mean opinion score ( mos ) test has been conducted .
the paper is organized as follows : section ii gives a brief overview of the vocoder algorithms .
the bit allocation and bit rate calculations are also described .
section iii describes the method adopted in implementing and testing the vocoder .
the results obtained and their implications are discussed in section iv , followed by concluding remarks .
the linear prediction model forms the core of the celp , ld - celp and melp algorithms , as depicted in fig . [ fig : core ] . in the lpc model
, speech signals are classified into voiced and unvoiced signals .
voiced signals are generated when the vocal cords vibrate and are represented in the lpc source - filter model as periodic excitation .
unvoiced signals are generated by turbulence of air in the vocal tract and do not involve the vocal cords .
these signals are usually represented as white gaussian noise .
the lpc coder consists of a linear predictor having adaptive delay and gain factors @xcite .
since there are sounds in speech that are produced by a combination of voiced and unvoiced signals , it has been observed that important perceptual speech information gets missed out from the predictor output .
also a slight error in the predictor coefficients will lead to more speech information being missed out .
this unaccounted residual output of the lpc coder contains important data on how the sound signal is perceived by the human auditory system .
it was proposed in @xcite that the prediction residual signal could be used to enhance the perceptual quality of the coder output .
a codebook containing a list of codevectors is searched to obtain a closest match to the residual signal .
the index of the codevector is selected such that minimization of the perceptually weighted error metric is obtained .
the codec earned its name due to the use of codebooks to obtain _ codes _ for modeling the speech signal .
the same codebook is available at both the transmitter and the receiver . at the receiver
, the index is used to obtain the codevector and use it in the synthesis filters .
celp uses the analysis - by - synthesis ( abs ) method in which the transmitter _ analyzes _ the signal to produce linear prediction coefficients , and then uses these coefficients to _ synthesize _ the speech signal within the transmitter itself .
an error signal is generated and codevectors are selected from the codebook in order to minimize the perceptually weighted mean square error . in the celp transmitter ,
the transmitter first splits the input speech into frames of around 30 ms .
short - term linear prediction is performed , i.e. formants ( peaks of the spectral envelope ) are estimated for each input frame .
the transfer function of the perceptual weighting filter of celp is given by the following equation .
@xmath0 where @xmath1 @xmath2 where m is the lpc predictor order and @xmath3 s are the quantized lpc coefficients . after finding the short term lpc coefficients , each frame
is split into four subframes , i.e. 7.5 ms each , which are given as input to the long - term prediction filter .
the pitch and the intensity of the speech signal are estimated .
an optional post filtering stage may be added after decoding to enhance the quality of the output signal .
the celp bit allocation @xcite is shown below : [ cols="^,^,^",options="header " , ] from the time domain waveforms , it can be concluded that the overall shape of the original wave has been preserved .
however peaks have been clipped at certain portions which result in decrease in clarity of the speech output .
the spectrograms show the frequency content of the speech waveforms as a function of time .
the pitch and intensity graphs also show slight variations in the output of the vocoders .
reliable estimates of the perceptual quality can be made only by conducting subjective tests using human listeners .
all recorded input speech files were sampled at a rate of 8 khz .
the details of speech input files are shown in table [ tab : in_speech_files ] .
the average input speech file size was 393.6 kb and average duration 24.59s .
the mos scores corresponding to each input speech file and the average mos score obtained for each vocoder are shown in table [ tab : mos ] .
the mos scores in table [ tab : mos ] show that ld - celp has the highest perceptual quality ( toll quality ) among the three vocoders .
the perceptual quality of celp and melp vocoders is rated less , with celp scoring slightly higher than melp .
the results of the comparison indicate that a choice can be made only based on the application of the vocoder . in applications where the focus is on low delay and high perceptual quality , as in two - way communication systems , the ld - celp algorithm at 16
kb / s is the ideal candidate . in areas where low bit rate is essential
, melp is the best candidate because it can work at bit rates as low as 2.4 kb / s and gives intelligible output .
when both low bit rate and good quality are required , the celp coder at 4.8 kb / s seems to be the most suitable coder . in this study ,
the number of evaluators for the mos test was limited to 10 due to time constraints . for more accurate results ,
the number of evaluators needs to be increased .
the authors would like to thank the students and faculty of the department for providing speech samples and for their participation in the mean opinion score testing .
special thanks goes to karthika balan for her valuable help in the mos testing process .
m. schroeder and b. s. atal , `` code - excited linear prediction ( celp ) : high - quality speech at at very low bit rates , '' _ ieee international conference on acoustics , speech and signal processing _ ,
10 , pp . 937940 , 1985 .
juin - hwey chen , r. v. cox , y. c. lin , n. jayant and m. j. melchner , `` a low - delay celp coder for the ccitt 16 kb / s speech coding standard , '' _ ieee journal on selected areas in communications _ ,
10 , issue 5 , pp . 830849 , 1992 .
alan mccree , kwan truong , e. b. george , t. p. barnwell and v. viswanathan , `` a 2.4 kbit / s melp coder candidate for the new u.s .
federal standard , '' _ ieee international conference on acoustics , speech , and signal processing _ , vol . 1 , pp . 200203 , 1996 .
lynn m. supplee , ronald p. cohn , john s. collura and alan v. mccree,``melp : the new federal standard at 2400 bps , '' _ ieee international conference on acoustics , speech , and signal processing _
, vol . 2 , pp . 15911594 , 1997 .
l. mathew , a. anselam and s. s. pillai , `` analysis of ld - celp coder output with sound exchange and praat software , '' unpublished .
wai c. chu , `` speech coding algorithms : foundation and evolution of standardized coders , '' wiley interscience , 2003 .
olivier hersent , jean - pierre petit and david gurle , `` beyond voip protocols : understanding voice technology and networking techniques for ip telephony , '' john wiley & sons , 2005 . | speech coding speech signals transmission at bit rates lower than original digitized speech human auditory system certain aspects speech signal perceptually relevant information extracted equivalent - sounding wave at lower bandwidth linear prediction compression technique past samples stored predict next sample basic linear predictive coder prototype prediction over one pitch period using adaptive linear delay gain factors produces intelligible artificial - sounding speech output improve perceptual quality variants are code excited linear prediction ) low - delay celp ) use forward backward linear prediction codebooks lookup tables with codevectors speech residual signals in mixed excitation linear prediction melp ) @xcite additional classification of speech jittery voiced speech mixed excitation noise distinguishing feature melp model paper aims three types linear prediction based vocoders bit rate perceptual quality subjective testing voice quality major step in evaluation vocoder vocoder accepted if satisfied with voice quality subjective evaluation using mean opinion score mos ) test conducted paper organized section ii vocoder algorithms bit allocation bit rate calculations described section iii describes method implementing testing vocoderresults implications discussed in section iv followed concluding remarks linear prediction model forms core of celp ld - celp melp algorithms depicted in fig . lpc model speech signals classified into voiced unvoiced signals voiced signals generated vocal cords vibrate represented as periodic excitation unvoiced signals generated by turbulence air vocal tract cords represented as white gaussian noise lpc coder linear predictor adaptive delay gain factors @xcite sounds speech produced combination voiced unvoiced signals important perceptual speech information from predictor output slight error in predictor coefficients more speech information unaccounted residual output contains data on sound signal human auditory system proposed prediction residual signal enhance perceptual quality coder output codebook codevectors searched closest match to residual signal index codevector selected minimization perceptually weighted error metric codec earned name due to use codebooks modeling speech signal same codebook available at transmitter receiver receiver index used to obtain codevector in synthesis filterscelp uses analysis - by - synthesis method transmitter analyzes signal linear prediction coefficients synthesize speech signal error signal generated codevectors selected from codebook minimize perceptually weighted mean square error splits input speech into frames 30 ms short - term linear prediction performed formants peaks spectral estimated for each input frame transfer function of perceptual weighting filter given by equation @xmath0 m lpc predictor order s quantized lpc coefficients after short term lpc coefficients each frame split into four subframes 7.5 ms each input to long - term prediction filter pitch intensity of speech signal estimated optional post filtering stage added after decoding enhance quality output signal celp bit allocation @xcite shown below time domain waveforms shape of original wave preserved peaks clipped at clarity speech spectrograms show frequency content speech waveforms function of time pitch intensity graphs show variations in output vocoders reliable estimates perceptual quality subjective tests using human listeners all recorded input speech files sampled at 8 khz details shown in tableaverage input speech file size 393.6 kb duration 24.59s mos scores each score each vocoder shown in table [ tab : mos ] show ld - celp highest perceptual quality among three vocoders quality of celp melp vocoders rated less celp higher than indicate choice based on application vocoder low delay high perceptual quality ld - celp algorithm at 16 kb / s ideal low bit rate essential melp best at bit rates 2.4 kb / s intelligible output low bit rate good quality celp coder at 4.8 kb / s most suitable evaluators for mos test limited to 10 time constraints for results needs authors thank students faculty for speech samples participation in opinion score testing thanks to karthika balan for help mos testing m. schroeder b. s. atal code - excited linear prediction ( celp ) : high - quality speech at low bit rates ieee international conference on acoustics speech signal processing 10 pp 937940 1985 juin - hwey chen v. cox lin n jayant j.melchner low - delay celp coder ccitt 16 kb / s speech standard journal communications 10 issue 5 830849 1992 alan mccree kwan truong george barnwell viswanathan 2.4 kbit / s melp coder u.s federal standard international conference acoustics speech signal processing vol 1 200203 1996 lynn supplee ronald cohn john collura alan mccree new federal standard 2400 bps conference vol 2 15911594 1997 mathew anselam pillai analysis ld - celp coder output sound exchange software wai chu speech coding algorithms foundation evolution standardized coders wiley interscience 2003 olivier hersent petit david gurle beyond voip protocols voice technology networking techniques ip telephony john wiley sons 2005 | 0.322448 | 2,008 | 967 | , Saving $0.1 in GPT-4. |
155,234 | the cleo - c experiment has made several recent contributions to the study of charmonium decays .
this review briefly outlines the cleo - c analyses of charmonium that were completed or made public during 2010 .
special attention is given to the discovery of the process @xmath0 at a center of mass energy of @xmath1 . | despite the conclusion of data - taking in 2008 , the cleo - c experiment which utilizes symmetric @xmath2 collisions in the charmonium region provided by the cornell electron storage ring ( cesr ) continues to have an active program in the analysis of charmonium data .
recent cleo - c studies include : ( 1 ) measurements of the branching fractions of @xmath3 , @xmath4 , and @xmath5 @xcite ; ( 2 ) studies of @xmath6 decays to @xmath7 , @xmath8 , and @xmath4 and a confirmation of the besii observation of the @xmath9 , a @xmath10 threshold enhancement in @xmath11 @xcite ; ( 3 ) a search for the radiative transition @xmath12 @xcite ; ( 4 ) measurements of the higher order multipoles in the radiative transitions @xmath13 and @xmath14 @xcite ; and ( 5 ) the discovery of the process @xmath0 at a center of mass energy of @xmath1 ( preliminary ) . in the following , each study is briefly reviewed .
starting with the full sample of 25.9 million @xmath6 decays , cleo - c measured the branching fractions for @xmath3 , @xmath4 , and @xmath5 @xcite .
these results are important inputs to predictions of the related charmonium production rates ( @xmath15 , where @xmath16 is a hadron ) that will be studied by the panda experiment at gsi .
in addition , the dalitz plots of @xmath17 have been studied ( figure [ fig : pppi ] ) .
an enhancement in @xmath10 mass appears , which is inconsistent with both predictions from the meson emission model @xcite and the @xmath9 threshold enhancement @xcite .
in addition to the @xmath18 decays discussed above , cleo - c has also measured the branching fractions of @xmath6 decays to final states involving a @xmath10 pair @xcite .
the dalitz plots of @xmath8 and @xmath4 show evidence for the production of a number of @xmath19 resonances in @xmath20 and @xmath21 , but do not show evidence for a @xmath10 threshold enhancement consistent with the @xmath9 .
the @xmath9 is also not seen in the @xmath6 radiative decay to @xmath10 .
however , evidence is found for the @xmath9 through the process @xmath22 , which provides confirmation of the original besii discovery @xcite .
in the spectroscopy of charmonium below @xmath24 threshold , the @xmath23 remains one of the least measured states .
cleo - c performed a search for the @xmath23 in the allowed @xmath25 transition , @xmath12 , where the potential @xmath23 was fully reconstructed in sixteen of its possible decay modes to light quark systems @xcite .
upper limits on the product branching fractions , @xmath26 , were separately set for each possible decay mode of the @xmath23 , @xmath16 .
the upper limits ranged from approximately @xmath27 to @xmath28 .
[ fig : multipoles ] the radiative decays @xmath13 and @xmath14 are dominated by electric dipole ( e1 ) amplitudes , but are expected to have small additional contributions from the higher - order magnetic quadrupole ( m2 ) amplitudes .
the relative contributions of the m2 amplitudes are , at least in principle , sensitive to the anomalous magnetic moment of the charm quark .
cleo performed a high - statistics analysis of this process using the decay chains @xmath29 for @xmath30 , using clean samples of approximately @xmath31 events for @xmath32 and approximately @xmath33 events for @xmath34 @xcite , significantly larger event samples than previous measurements . using an unbinned maximum likelihood fit to angular distributions , cleo found the following normalized m2 admixtures : for @xmath35 @xmath36 , @xmath37 ; for @xmath38 @xmath39 , @xmath40 ; for @xmath41 @xmath42 , @xmath43 ; and for @xmath44 @xmath45 , @xmath46 . for the quoted @xmath34 measurements ,
the electric octupole ( e3 ) moments were fixed to zero .
there is clear evidence for non - zero m2 amplitudes ( left plot of figure [ fig : multipoles ] ) .
these new measurements agree well with theoretical expectations when the anomalous magnetic moment of the charm quark is assumed to be zero and the mass of the charm quark is assumed to be @xmath47 , as shown in the right plot of figure [ fig : multipoles ] .
finally , cleo - c has presented preliminary results on the observation of @xmath0 at a center of mass energy of @xmath1 .
this builds upon a previous analysis of cleo - c to study the @xmath49 @xmath50 cross sections as a function of center of mass energy @xcite . in that analysis a rise in the cross section at @xmath51 for both the @xmath52 and @xmath53 reactions was attributed to the @xmath54 , a charmonium hybrid meson candidate first discovered by babar @xcite . since that analysis , cleo - c has collected a total of @xmath55 of data at @xmath1 , about twenty times larger than the data sets used to search for @xmath56 between 4120 and 4200 @xmath57 . using this larger data set ,
cleo - c has made the first observation of @xmath0 at @xmath1 and has searched for @xmath0 in the original scan data , which spans energies from @xmath58 to @xmath51 .
the reaction @xmath0 was exclusively reconstructed using @xmath59 and twelve decay modes of the @xmath60 ( @xmath61 , @xmath62 , @xmath63 , @xmath64 , @xmath65 , @xmath66 , @xmath67 , @xmath68 , @xmath69 , @xmath70 , @xmath71 , @xmath72 ) .
all final state particles were detected and the sum of their four - momenta was constrained to the four - momentum of the initial @xmath2 system .
the quality ( @xmath73 ) of this kinematic fit was used to select events .
the @xmath60 was selected by requiring the recoil mass of the @xmath74 system be within @xmath75 of the @xmath60 mass .
then the @xmath76 recoil mass was used to isolate the @xmath77 signal .
the left plot of figure [ fig : hcsignal ] shows the @xmath76 recoil mass when all twelve decay modes of the @xmath60 are summed together .
there is a peak at the @xmath48 mass , which contains @xmath78 events and has a @xmath79 significance . to convert this to a cross section , however , one must know the @xmath60 branching fractions .
the right plot of figure [ fig : hcsignal ] shows the same figure except now restricted to those decay modes of the @xmath60 that have measured branching fractions .
here there are @xmath80 events with a @xmath81 significance .
this corresponds to a cross section of @xmath0 of @xmath82 , where the first error is statistical , the second is systematic , and the third is due to the errors on the external measurements of the @xmath60 branching fractions and the branching fraction of @xmath59 . for reference ,
the cross section for @xmath83 at @xmath1 is @xmath84 . to obtain a scan of the @xmath77
cross section as a function of center of mass energy , the same procedure was followed using the smaller samples of scan data . following the @xmath56 scan @xcite ,
the data is divided into three bins : @xmath85 ; @xmath86 ; and @xmath51 .
figure [ fig : scan ] shows the resulting @xmath77 cross section as a function of center of mass energy .
the @xmath52 results are included for reference .
there is perhaps a suggestive rise at @xmath51 , but more data is clearly needed . | conclusion data 2008 cleo - c experiment utilizes symmetric @xmath2 collisions charmonium region continues active analysis charmonium data recent studies include measurements branching fractions @xmath3 @xmath5 @xcite studies @xmath6 decays to @xmath7 confirmation observation @xmath9 @xmath10 threshold enhancement in @xmath11 @xcite search for radiative transition @xmath12 @xcite measurements higher order multipoles in radiative transitions @xmath13 @xmath14 @xcite discovery process @xmath0 at center of mass energy @xmath1 each study briefly reviewed starting full sample 25.9 million @xmath6 decays measured branching fractions for @xmath3 @xmath5 @xcite results important inputs predictions related charmonium production rates @xmath16 hadron studied panda experiment gsi dalitz plots of @xmath17 studied enhancement in @xmath10 mass appears inconsistent with predictions meson emission model @xcite @xmath9 threshold enhancement @xcite measured branching fractions of @xmath6 decays to final states @xmath10 pair @xcitedalitz plots @xmath8 show @xmath19 resonances in @xmath10 threshold enhancement @xmath9 @xmath9 not seen in @xmath6 radiative decay to @xmath10 evidence for @xmath9 through @xmath22 original besii discovery spectroscopy charmonium below @xmath24 threshold @xmath23 least measured cleo - c for @xmath23 in @xmath25 transition @xmath12 potential reconstructed in sixteen decay modes to light quark systems upper limits product branching fractions @xmath26 set for each decay mode limits ranged @xmath27 to @xmath28 radiative decays @xmath13 @xmath14 dominated by electric dipole amplitudes small contributions from higher - order magnetic quadrupole m2 ) amplitudes relative contributions m2 amplitudes sensitive to anomalous magnetic moment charm quark cleo performed high - statistics analysis using decay chains @xmath29 @xmath30 clean samples @xmath31 @xmath33 @xmath34 larger samplesunbinned maximum likelihood angular distributions cleo found normalized m2 admixtures @xmath35 @xmath34 measurements electric octupole e3 moments fixed to zero evidence for non - zero m2 amplitudes left plot figure new measurements agree with theoretical expectations anomalous magnetic moment charm quark zero mass charm quark @xmath47 right plot figure cleo - c presented preliminary results observation @xmath0 at center of mass energy @xmath1 builds previous analysis @xmath49 @xmath50 cross sections function center mass energy rise in cross section at @xmath51 @xmath52 reactions attributed to @xmath54 charmonium hybrid meson candidate discovered babar @xcite cleo c collected @xmath55 data at @xmath1 twenty times larger than data sets @xmath56 between 4120 4200 @xmath57 larger data set first observation of @xmath0 at @xmath1 searched for @xmath0 in original scan data energies @xmath58 to @xmath51reaction @xmath0 reconstructed using @xmath59 twelve decay modes @xmath60 ( final state particles detected four - momenta constrained to four initial @xmath2 system quality @xmath73 ) used to select events @xmath60 selected recoil mass @xmath74 system within @xmath75 @xmath60 mass @xmath76 recoil mass used to isolate @xmath77 signal left plot shows @xmath76 recoil mass when twelve decay modes summed together peak at @xmath48 mass contains @xmath78 events @xmath79 significance to convert to cross section must know @xmath60 branching fractions right plot shows same figure restricted to decay modes @xmath60 fractions @xmath80 events with @xmath81 significance corresponds to cross section of @xmath0 of @xmath82 first error statistical second systematic third due to errors on external measurements @xmath60 branching fractions @xmath59 cross section for @xmath83 at @xmath1 is @xmath84 obtain scan @xmath77 cross section center mass energy same procedure followed smaller samples scan data following @xmath56 scan data divided three bins @xmath85 @xmath86 @xmath51 figure shows @xmath77 cross section center mass energy @xmath52 results included reference suggestive rise @xmath51 more data needed | 0.308728 | 1,858 | 942 | , Saving $0.1 in GPT-4. |
155,237 | much of our knowledge about the solar dynamo is based on sunspot observations .
it is thus desirable to extend the set of positional and morphological data of sunspots into the past .
gustav sprer observed in germany from anklam ( 18611873 ) and potsdam ( 18741894 ) .
he left detailed prints of sunspot groups , which we digitized and processed to mitigate artifacts left in the print by the passage of time .
after careful geometrical correction , the sunspot data are now available as synoptic charts for almost 450 solar rotation periods .
individual sunspot positions can thus be precisely determined and spot areas can be accurately measured using morphological image processing techniques .
these methods also allow us to determine tilt angles of active regions ( joy s law ) and to assess the complexity of an active region . | the first continuous observations of sunspots were made by samuel heinrich schwabe 18251867 . in this time , richard carrington started his observations in 1853 , followed by sprer s first observations in january 1861 during carrington rotation period no .
96 . in total 445 rotation periods were recorded .
the images were published as lithographs , which reproduce many details of the sunspot fine structure , e.g. , umbra and penumbra @xcite .
each page contains five rotation periods .
the abscissa shows the heliographic longitude @xmath0 from @xmath1 to @xmath2 and on the ordinate is the heliographic latitude @xmath3 from @xmath4 to @xmath5 for one rotation period .
the sunspot drawings were collated in hardbound books .
consequently , pages will not be perfectly flat when opening the books .
binder clips and metal coins were used to get a flat image of the drawings .
whole pages were photographed with a digital camera .
the drawings were published nearly 150 years ago , thus , the pages are yellowish , stained , and corrugated .
these artifacts have to be removed because they would severely affect the analysis of sunspot properties .
all algorithms were developed in the interactive data language ( idl ) .
after converting the color images to gray - scale images , they are rotated by @xmath6 to have them in portrait format as in the books .
the images are cropped so that only the coordinate grids with sunspot data and their annotation remain . to remove a small rotation angle error of the camera
, the standard deviation of the intensity traces of all rows and all columns has to reach a maximum .
then , the angular offset is zero .
an order - statistics filter @xcite is used to compute a background intensity map .
the filter ranks the intensities within a @xmath7 pixel neighborhood and replaces the value of the center pixel by the @xmath8 percentile of the intensity distribution .
the 16-pixel wide borders of the full background intensity map are replaced by the mean background intensity .
division of an image by its background intensity map ( i.e. , a pseudo flat - fielding procedure ) results in an artificially bleached and flattened image . to get the exact coordinates of the grid lines ,
a @xmath9-pixel kernel is used , where the value of two intersections will be set to unity and all other pixels at zero .
a good approximation of the grid coordinates is the average of the line profiles of rows and columns , which were smoothed with a @xmath10-pixel lee - filter and afterwards normalized .
good estimates of the @xmath11- and @xmath12-grid coordinates yields a parabolic fit to local maxima of row and column profile .
to improve the grid coordinates , from the binary mask , a region of @xmath13 pixels will be extracted and enlarged by factor of five . afterwards
, the region will be smoothed with a @xmath14-pixel boxcar filter .
nevertheless , sunspots near to intersecting grid lines can lead to erroneous results . as a precaution , all @xmath11- and @xmath12-coordinates
are checked against the positions of their immediate neighbors and replaced , if a significant offset is detected .
the heliographic longitude and latitude are given in increments of @xmath15 , so that we have a grid of ( @xmath16 ) @xmath11- and @xmath12-coordinates for each rotation period .
we also included a @xmath17-wide boundary to minimize sampling errors at the periphery .
subsequently , delaunay triangulations were used to interpolate the sunspot data to the new grid ( the result is shown in fig .
[ fig : map ] ) . after image processing ,
standardized sunspot maps are at hand for further analysis .
the sunspot drawings are made of dark - inked areas for the umbrae and many single points to indicate the penumbral fine structure .
a more realistic ` intensity distribution ' can be achieved using some gaussian smoothing . even though blurred , the images are now suitable for image processing techniques such as intensity thresholding .
morphological opening and closing operations can then be applied to the binary masks , in which individual sunspots are labeled .
thus , ` holes ' in the sunspot area can be filled and spot contours become less rough .
we used standard tools for ` blob analysis ' @xcite to derive parameters describing the morphology of sunspots and to determine their positions .
in general , this approach leads to very good results but in some cases , the spoke - like structure used to indicate penumbral fine structure results in artifacts .
another problem are the numbers , which label the sunspot groups . however , since they consist of thin lines as compared to the more compact sunspots , morphological image processing is again capable of indentifying and removing the labels . in summary ,
sprer s sunspot drawing
are of high quality facilitating the application of modern feature recognition and image processing methods . 3 sprer , g. : 1874 , pub .
. ges . 13 fanning , d.w . : 2011 , _
coyote s guide to traditional idl graphics _ , coyote book publ .
, fort collins , co gonzalez , r.c . , woods , r.e .
: 2002 , _ digital image processing _ , prentice - hall , upper saddle river , nj | first observations of sunspots by samuel heinrich schwabe richard carrington started observations 1853 followed sprer s observations 1861 carrington rotation period 96 445 rotation periods recorded images published as lithographs reproduce sunspot structure umbra penumbra each page contains five rotation periods abscissa shows heliographic longitude from to ordinate heliographic latitude from to for one rotation period sunspot drawings collated in hardbound books pages not flat binder clips metal coins used flat image pages photographed with digital camera drawings published 150 years ago pages yellowish stained corrugated artifacts affect analysis sunspot properties algorithms developed in interactive data language idl ) after converting color images to gray - scale rotated by @xmath6 to portrait format images cropped coordinate grids with sunspot data annotation remain remove rotation angle error standard deviation of intensity traces maximum angular offset is zero order - statistics filter background intensity map ranks intensities within pixel neighborhood replaces center pixel by @xmath8 percentile of intensity distribution16-pixel borders replaced by mean intensity division image by intensity results in bleached flattened image exact coordinates grid lines @xmath9-pixel kernel used value two intersections set to unity other pixels at zero approximation grid coordinates is average line profiles of rows columns smoothed with @xmath10-pixel lee - filter normalized estimates @xmath11- @xmath12-grid coordinates parabolic fit to local maxima row column profile coordinates region of @xmath13 pixels extracted enlarged by factor five smoothed with @xmath14-pixel boxcar filter sunspots near intersecting grid lines lead to erroneous results @xmath11- @xmath12-coordinates checked against neighbors replaced if offset heliographic longitude latitude given in increments of @xmath15 grid of @xmath11-coordinates for each rotation period included @xmath17-wide boundary minimize sampling errors delaunay triangulations interpolate sunspot data to new grid result shown in fig . after standardized sunspot maps at hand for further analysissunspot drawings dark - inked areas for umbrae single points penumbral fine structure realistic intensity distribution achieved using gaussian smoothing blurred images suitable for image processing techniques intensity thresholding morphological opening closing operations applied to binary masks sunspots labeled holes sunspot filled spot contours less rough used standard tools for blob analysis parameters morphology sunspots determine positions approach leads good results some spoke - like structure results in artifacts problem numbers label sunspot groups thin lines compact sunspots morphological image processing removing labels sprer s sunspot drawing high quality modern feature recognition image processing methods sprer , g. : 1874 pub ges 13 fanning d : 2011 coyote s guide to traditional idl graphics book publ fort collins gonzalez woods 2002 digital image processing prentice - hall upper saddle river nj | 0.188061 | 1,157 | 566 | , Saving $0.0 in GPT-4. |
155,238 | in this paper , we propose a blind signature scheme and three practical educed schemes based on elliptic curve discrete logarithm problem .
the proposed schemes impart the gost signature structure and utilize the inherent advantage of elliptic curve cryptosystems in terms of smaller key size and lower computational overhead to its counterpart public key cryptosystems such as rsa and elgamal .
the proposed schemes are proved to be secure and have less time complexity in comparison with the existing schemes .
blind signature , elliptic curve , gost signature , unforgeability , blindness . | blind signature is a form of digital signature in which the message is blinded before it is signed , in order to allow the requester to get a signature without giving the signer any information about the actual message or the resulting signature .
blind signatures are used to build practical offline and online untraceable electronic cash schemes @xcite and widely employed in privacy - related cryptographic protocols , such as electronic election systems @xcite . the paper analogy to the blind signature is enclosing a ballot in a carbon paper lined envelope ; in this way , the signer does not view the message content , and also everyone can later check the validity of the signature .
several blind signature schemes are proposed in the literature .
the first scheme , proposed by chaum @xcite , was based on rsa signature . in @xcite , okamoto proposed the blind schnorr signature and pointcheval et al . proved its security in @xcite . in 1995 , camenisch et al .
proposed a blind signature scheme based on the discrete logarithm problem ( dlp ) @xcite and later , in 2005 , wu et al .
proved its untraceability @xcite .
pointcheval developed a blinding scheme for okamoto s signature in @xcite . in @xcite ,
huang et al . presented a blind signature scheme based on gost signature , which is the russia s digital signature algorithm @xcite . in @xcite ,
an efficient blind signature scheme is presented based on the elliptic curve discrete logarithm problem . in this paper
, we propose a gost - like blind signature scheme and three efficient educed schemes based on elliptic curve discrete logarithm problem .
the schemes utilize the inherent advantage of elliptic curve cryptosystems in terms of smaller key size and lower computational overhead compared to its counterpart public key cryptosystems such as rsa and elgamal .
the schemes are proved to be correct and secure .
they can be used in various cryptographic protocols where the anonymity of the requester is required .
the remainder of this paper is organized as follows . in section 2 ,
basic concepts of elliptic curves are presented .
the gost digital signature scheme is described in section 3 . in section 4 ,
the generalized scheme and three other educed schemes are elaborated and the security and performances are discussed .
finally , section 5 concludes the paper .
the elliptic curve analogues of dlp - based schemes was independently proposed by koblitz @xcite and miller @xcite , in 1985 . since then , several cryptosystems are developed based on elliptic curve computations . a non - super singular elliptic curve @xmath0 over a finite field @xmath1 is as follows : @xmath2 where @xmath3 . the point @xmath4 , where @xmath5 satisfy equation [ eq : first ] , together with a point at infinity , denoted by @xmath6 , form an abelian group @xmath7 whose identity element is @xmath6 .
the negative of @xmath8 is @xmath9 .
let @xmath8 and @xmath10 be two distinct points on an elliptic curve such that @xmath11
. then @xmath12 , where : @xmath13 where @xmath14 . doubling a point @xmath15 , in case of @xmath16 , results in @xmath17 , where : @xmath18 where @xmath19 .
* definition : * let @xmath0 be an elliptic curve over a finite field @xmath1 and let @xmath20 be a point of order @xmath21 . given another point @xmath22 , the elliptic curve discrete logarithm problem ( ecdlp ) is to find the integer @xmath23 $ ] , such that @xmath24 @xcite .
in this section , we describe the gost digital signature scheme @xcite .
let @xmath25 and @xmath26 be large primes that satisfy @xmath27 , and @xmath28 be an element in @xmath29 with order @xmath26 .
let @xmath30 be a secure hash function .
the signer s secret and public key pair is @xmath31 , where @xmath32 and @xmath33 .
let @xmath34 be the message to be signed .
* signing : * the signer chooses random number @xmath35 and computes : @xmath36 the signature on message @xmath34 is @xmath37 . *
verification : * the verifier computes : @xmath38 and checks whether @xmath39 .
in @xcite , a blind signature scheme based on the gost signature is presented . here
, we propose a gost - like blind signature scheme based on ecdlp .
there are two participants in a blind signature scheme : a signer and a group of requesters .
initially , the signer publishes the necessary information .
then , the user sends a blinded version of the message to the signer .
the signer signs the blinded message , and sends the result back to the user .
afterwards , the user extracts the signature . at the end
, the validity of the signature is verified .
the details of these phases are described below .
* initialization : * first , the curve parameters must be agreed upon by signer and requester .
let @xmath0 be the used elliptic curve over @xmath1 and suppose that the number of @xmath1-rational points on @xmath0 is divisible by a sufficiently large prime @xmath40 .
let @xmath41 be a point on @xmath0 of order @xmath21 .
signer must have a key pair suitable for elliptic curve cryptography , consisting of a private key @xmath42 ( a randomly selected number in the interval @xmath43 $ ] ) and a public key @xmath44 where @xmath45
. then the signer chooses random number @xmath46 in the interval @xmath43 $ ] , computes @xmath47 and sends @xmath48 to the requester .
* requesting : * the requester chooses random numbers @xmath49 , @xmath50 and @xmath51 in the interval @xmath43 $ ] and computes : @xmath52 then sends @xmath53 to the signer .
@xmath53 is an encrypted version of the message , i.e. the blinded message . *
signing : * signer computes the signature of the blinded message as : @xmath54 and sends the result back to the requester . * extraction : * requester extracts the signature of the message from the signature of the blinded message , by computing : @xmath55 and declares the pair @xmath56 as the signature on @xmath34 . * verification : * the legitimacy of the signature @xmath56 for the message @xmath34 is verified by examining : @xmath57 the various phases of the proposed scheme are summarized in figure 1 .
the correctness can be easily proved as follows : @xmath58 the security of blind signature schemes is defined by unforgeability and blindness . here , we discuss these properties of the proposed blind signature scheme .
* unforgeability : * forgery is an attack trying to fabricate a digital signature for a message without having access to the respective signer s private key .
the security requirement of unforgeability of digital signatures is also called non - repudiation . to forge a valid blind signature ,
the adversary should obtain the signature @xmath59 or the signer s private key @xmath42 to fabricate the signature @xmath60 .
it is impossible to obtain @xmath42 from the public key @xmath44 using the equation @xmath45 , because it is based on ecdlp . to forge @xmath59 , a dishonest requester ( as an adversary )
must calculate @xmath61 .
the requester knows the parameters @xmath44 and @xmath48 and can compute @xmath62 , which is equal to @xmath63 . again finding @xmath59 from @xmath63 is impossible , because it is based on ecdlp .
thus , the unforgeability of the scheme is assured .
* blindness : * a signature scheme is called blind , if the signer s view and the resulting signature are statistically independent .
the signer s view is the set of all values that the signer gets during the execution of the signature issuing protocol , which in the proposed scheme is the tuple @xmath64 .
the three blinding functions are : @xmath65 it can be seen that , there always exists a tuple of random numbers @xmath66 which maps any @xmath64 to any @xmath56 , because there are three random parameters in the three blinding functions . thus , the scheme is blind . as in @xcite ,
three educed schemes are derived from the generalized scheme .
in fact , two random parameters are sufficient to provide blindness . the tuple of random parameters @xmath66 for the three educed schemes are @xmath67 , @xmath68 and @xmath69 .
the security of the educed schemes is discussed below .
* case i : @xmath70 + in this case , the blinding functions are : @xmath71 + the correctness and the unforgeability are the same as the generalized scheme and the blindness can be proved as follows .
+ let @xmath72 be the data appearing in the signer s view during the execution of the signature and @xmath73 be the corresponding data at the verifier .
it is sufficient to show that there exist a pair of random numbers @xmath74 that maps @xmath75 to @xmath76 , for @xmath77 .
we define : @xmath78 by using equations 6 , 8 , 10 and 14 , we have : @xmath79 thus , the tuples @xmath75 and @xmath76 have exactly the same relation defined by the signature issuing protocol , thus the scheme is blind .
* case ii : @xmath80 + in this case , the blinding functions are : @xmath81 + the correctness and the unforgeability are also the same as the generalized scheme and the blindness is proved similar to the case i , by defining : @xmath82 * case iii : @xmath83 + in this case , the blinding functions are : @xmath84 + the correctness and the unforgeability are also the same as the generalized scheme and the blindness is proved similar to the case i , by defining : @xmath85 the time complexity of the proposed schemes is compared with a recently proposed ecdlp - based blind signature @xcite and the scheme proposed by camenisch et al .
@xcite , which is declared to have superior performance than other dlp - based blind signatures @xcite .
table [ tb : notations ] defines the notations . in this table
, the sub - index ( @xmath25 ) denotes a prime field of order @xmath86 .
the time complexity of various operation units in terms of the time complexity of a modular multiplication is illustrated in table [ tb : conversion ] @xcite .
comparisons are based on the fact that an elliptic curve @xmath87 with a point @xmath20 whose order is a 160-bit prime offers approximately the same level of security as dsa with a 1024-bit modulus @xmath25 @xcite .
.definition of notations [ cols="^,^ " , ]
this paper suggested a secure and efficient gost - like blind signature scheme and three practical educed schemes based on the elliptic curve discrete logarithm problem .
the schemes utilize the inherent advantage of elliptic curve cryptosystems in terms of smaller key size and lower computational overhead compared to its counterpart public key cryptosystems such as rsa and elgamal .
we proved the security of the proposed schemes is based on ecdlp and the time complexity is lower than the existing blind signature schemes .
the schemes are applicable in the cryptographic services that emphasize the privacy of users , such as electronic voting over internet and untraceable payment services . | blind signature digital signature message blinded before signed requester signature without signer information about message blind signatures used build untraceable electronic cash schemes in privacy - related cryptographic protocols electronic election systems analogy signature ballot in carbon paper lined envelope signer view message content check validity signature several blind signature schemes proposed first scheme chaum based on rsa signature okamoto proposed blind schnorr signature pointcheval et proved security 1995 camenisch et al proposed blind signature scheme based on discrete logarithm problem 2005 wu et al . proved untraceability pointcheval developed blinding scheme for okamoto s signature huang et al presented blind signature scheme based on gost signature russia s digital signature algorithm efficient blind signature scheme based on elliptic curve discrete logarithm problem paper propose gost - like blind signature scheme three efficient educed schemes based on elliptic curve discrete logarithm problem schemes utilize elliptic curve smaller key size lower computational overhead schemes proved correct secure used in cryptographic protocols where anonymity requester required remainder of paper organizedsection 2 concepts elliptic curves gost digital signature scheme section 3 section 4 generalized scheme three schemes elaborated security performances discussed section 5 concludes elliptic curve analogues dlp - schemes proposed koblitz miller 1985 cryptosystems developed elliptic curve computations non - super singular elliptic curve @xmath0 finite field @xmath1 @xmath2 @xmath3 point @xmath4 @xmath5 equation point infinity @xmath6 form abelian group @xmath7 identity element @xmath6 negative @xmath8 @xmath9 @xmath8 @xmath10 distinct points elliptic curve @xmath11 @xmath13 @xmath14 doubling point @xmath15 results @xmath17 @xmath18 @xmath19 @xmath0 elliptic curve finite field @xmath1 @xmath20 point order @xmath21 point @xmath22 elliptic curve discrete logarithm problem find integer @xmath23 $ @xmath24 gost digital signature scheme @xmath25 @xmath26 large primes @xmath27 @xmath28 element @xmath29 @xmath26 @xmath30 secure hash functionsigner s secret public key pair @xmath31 @xmath32 @xmath33 @xmath34 message signed signing signer chooses random number @xmath35 computes @xmath36 signature on message @xmath34 @xmath37 verification verifier computes @xmath38 checks @xmath39 blind signature scheme gost signature propose gost - like blind signature scheme ecdlp two participants signer requesters signer publishes information user sends blinded version message to signer signer signs sends result back to user user extracts signature validity verified details phases initialization curve parameters agreed by signer requester @xmath0 elliptic curve over @xmath1 @xmath1-rational points @xmath0 divisible by large prime @xmath40 @xmath41 point on @xmath0 order @xmath21 signer key pair for elliptic curve cryptography private key @xmath42 randomly selected number @xmath43 public key @xmath44 @xmath45 signer chooses random number @xmath46 computes @xmath47 sends @xmath48 to requesterrequesting requester chooses random numbers interval computes @xmath52 sends @xmath53 to signer @xmath53 is encrypted version message blinded message signing signer computes signature as @xmath54 sends result to requester extraction requester extracts signature from blinded @xmath55 declares @xmath56 as signature on @xmath34 verification legitimacy signature @xmath56 verified examining @xmath57 phases proposed scheme summarized in figure 1 correctness proved @xmath58 security of blind signature schemes defined by unforgeability blindness discuss properties proposed signature scheme unforgeability forgery digital signature without access signer s private key non - repudiation forge valid signature adversary obtain signature @xmath59 or signer s private key @xmath42 fabricate signature @xmath60 impossible to obtain @xmath42 from public key @xmath44 based on ecdlp to forge @xmath59 dishonest requester must calculate @xmath61 requester knows parameters @xmath44 @xmath48 compute @xmath62 equal to @xmath63 finding @xmath59 from @xmath63 impossible based on ecdlpunforgeability scheme assured blindness signature scheme blind if signer s view signature statistically independent signer s view is values during signature issuing protocol proposed scheme tuple @xmath64 three blinding functions are @xmath65 tuple random numbers @xmath66 maps @xmath64 to @xmath56 three random parameters in blinding functions scheme blind three educed schemes derived from generalized scheme two random parameters blindness tuple random parameters @xmath66 for three educed schemes are @xmath67 @xmath68 @xmath69 security educed schemes discussed case i : @xmath70 blinding functions are @xmath71 correctness unforgeability same as generalized scheme blindness proved @xmath72 data in signer s view @xmath73 data at verifier pair of random numbers @xmath74 maps @xmath75 to @xmath76 for @xmath77 define @xmath78 equations 6 8 10 14 @xmath79 tuples @xmath75 and @xmath76 same relation defined by signature issuing protocol scheme blindcase ii : @xmath80 + blinding functions @xmath81 + correctness unforgeability same as scheme blindness similar to case i defining @xmath82 case iii : @xmath83 + blinding functions @xmath84 + correctness unforgeability same blindness similar to case i @xmath85 time complexity proposed schemes compared with ecdlp - based blind signature @xcite scheme camenisch et al . superior performance than dlp - based blind signatures table notations defines notations sub - index ( @xmath25 ) denotes prime field of order @xmath86 time complexity of operation units modular multiplication illustrated in table conversion comparisons elliptic curve @xmath87 with point @xmath20 160-bit prime same security as dsa with 1024-bit modulus @xmath25 paper suggested secure efficient gost - like blind signature scheme three practical educed schemes elliptic curve discrete logarithm problem schemes smaller key size lower computational overhead security proposed schemes based on ecdlp time complexity lower than existing blind signature schemesschemes applicable cryptographic services privacy electronic voting internet untraceable payment services | 0.432164 | 2,554 | 1,310 | , Saving $0.1 in GPT-4. |
155,245 | we highlight distinct and systematic observational features of magnetic field morphologies in polarized submm dust continuum .
we illustrate this with specific examples and show statistical trends from a sample of 50 star - forming regions . | dust polarization observations in the mm / submm continuum reveal plane - of - sky projected magnetic field morphologies . here , dust grains are expected to be aligned with their shorter axes parallel to the magnetic field , thus the detected polarized emission is perpendicular to the b field . despite the recent growing observations , e.g. , carma ( hull _ et al . _
@xcite ) , sma ( zhang _ et al . _
@xcite ) , magnetic field strengths can unfortunately not be directly inferred from these observed magnetic field morphologies . through a series of publications ( koch _ et al . _
@xcite , @xcite , @xcite , @xcite ) we introduced @xmath0 the local angle between an observed magnetic field orientation and the gradient of the underlying dust continuum stokes @xmath1 emission ( figure [ example_delta ] ) . with the help of a magnetohydrodynamics ( mhd ) force equation , @xmath0 can be interpreted as a magnetic field alignment factor , where @xmath2 measures the fraction of the magnetic field tension force that is directed along an emission ( density ) gradient .
we demonstrate clear systematic features found in @xmath3 across two regions observed with the sma at 345 ghz ( qiu _ et al . _
@xcite , @xcite ) in figure [ example_delta ] .
strikingly , higher - resolution observations with the sma at 1.3 mm ( qiu _ et al . _
@xcite ) show condensations along a mid - plane in g240.31 + 0.07 where @xmath3 is maximised , with regions above and below this mid - plane showing small @xmath3-values that allow the material to be channelled to the mid - plane .
opposite trends in @xmath3 structures are seen in g35.2 - 0.74n .
visually , these sources can clearly be categorised as type iia ( g240 ) and iib ( g35.2n ) as described below .
maps of @xmath3 ( color scale , @xmath4 ) for g240 ( left ; sma , @xmath5 ) and g35.2 ( right ; sma , @xmath5 ) .
contours show dust continuum .
magnetic field and emission gradient orientations are displayed with white and black dotted segments .
the dense condensations ( white filled circles ) in the center of g240 fall precisely onto regions with large @xmath3 , towards which material is channeled from the lower @xmath3 zones .
, title="fig : " ] maps of @xmath3 ( color scale , @xmath4 ) for g240 ( left ; sma , @xmath5 ) and g35.2 ( right ; sma , @xmath5 ) .
contours show dust continuum .
magnetic field and emission gradient orientations are displayed with white and black dotted segments .
the dense condensations ( white filled circles ) in the center of g240 fall precisely onto regions with large @xmath3 , towards which material is channeled from the lower @xmath3 zones .
, title="fig : " ] the examples depicted above are not unique
. from our sample of 50 targets ( sma , zhang _ et al .
_ @xcite ; cso , dotson _
@xcite ) most of the sources show systematic features in @xmath3 as soon as polarization is detected over a large enough area .
our sample covers both low- and high - mass star - formation sites over scales of 0.1 to 0.01 pc ( @xmath6 , @xmath7 m ; sma ) and larger scales around 1pc ( @xmath8 , @xmath9 m ; cso ) .
figure [ hist_delta ] shows histograms for the source - averaged @xmath10 ( 50 data points ) and for @xmath3 ( based on 4000 independent beams across all 50 sources ) .
the @xmath3-histogram of each source is normalized first before co - adding , in order to avoid the statistics being dominated by sources with the largest numbers of independent beams .
both histograms show a broad shoulder peaking around small @xmath10 and @xmath3 , falling off with a tail towards larger values .
this indicates that the prevailing source - magnetic field orientation statistically prefers small misalignments with the emission gradient .
thus , magnetic fields are found to be more often within small deviations of a source minor axis .
histograms for the source - averaged @xmath10 ( left panel , 50 sources ) and @xmath3 ( right panel , about 4000 independent measurements from all 50 sources ) .
color - coded in red , blue and black are the sources that can visually easily be identified as type iia or iib , and all iia , iib and iii sources together ] if @xmath3 is further interpreted with an mhd force equation , the ratio between the magnetic field tension force and gravity , @xmath11 , can be calculated from @xmath3 and @xmath12 as @xmath13 , where @xmath12 is the local angle between the gradient of the dust continuum and the direction of local gravity ( koch _ et al . _
this finding reflects the idea that an observed molecular cloud morphology with its characteristic angles @xmath0 and @xmath12 is a measure of the imprint of all the relevant forces in a system .
figure [ sigma_b_vs_delta ] displays source - averaged force ratios @xmath14 versus source - averaged @xmath10 .
the force ratio increases with growing @xmath10 .
source - averaged force ratios @xmath15 vs source - averaged @xmath16 for the combined sma and cso samples .
red and blue filled symbols show the visually identified iia- and iib - type sources from both the sma ( circles ) and cso ( squares ) .
empty symbols correspond to sources with less clear features .
black filled circles are the visually identified sma - iii sources .
schematic magnetic field structures are illustrated for iia- ( red segments ) , iib- ( blue segments ) and iii - type sources ( white segments ) .
white contours display the dust continuum emission .
the black - to - white color grading on the top indicates the local @xmath3 .
clear opposite trends are apparent for type - iia and -iib sources .
the solid gray line is the log - linear best fit @xmath17 with @xmath18 and @xmath19 .
the dashed and dotted gray lines indicate mean error ( @xmath20 ) and @xmath21 ( @xmath22 ) bounds . ]
the local angle @xmath3 is found to be a prime observable across a sample of 50 star - forming regions . it can further be linked to a magnetic field - to - gravity force ratio , @xmath11 .
maps of @xmath11 are closely correlated to features in @xmath3-maps , making @xmath3 a tracer for the local role of the magnetic field . while figure [ sigma_b_vs_delta ] shows a source - averaged correlation , it is important to note that our approach also provides a local assessment of the magnetic field .
thus , @xmath3 can explain where condensations , and potentially star formation , occur ( figure [ example_delta ] ) .
our finding of magnetic fields being more aligned parallel to emission gradients in clouds and cores ( figure [ hist_delta ] ) connects to _ planck _ results that observe a transition from fields being parallel to ridges in the diffuse interstellar medium to more orthogonal as densities towards molecular clouds increase ( planck collaboration xxxii @xcite ) .
dotson , j. l. , vaillancourt , j. e. , & kirby , l. 2010 , apjs , 186 , 406 hull , c.l.h . ,
plambeck , r.l .
, kwon , w. 2014 , apjs , 213 , 13 koch , p.m. , tang , y .- w .
, & ho , p.t.p .
2012a , apj , 747 , 79 koch , p.m. , tang , y .- w .
, & ho , p.t.p .
2012b , apj , 747 , 80 koch , p.m. , tang , y .- w .
, & ho , p.t.p .
2013 , apj , 775 , 77 koch , p.m. , tang , y .- w . , ho , p.t.p .
2014 , apj , 797 , 99 planck collaboration xxxii .
2014 , arxiv:1409.6728 qiu , k. , zhang , q. , wu , j. 2009 , apj , 696 , 66 qiu , k. , zhang , q. , menten , k.m .
2013 , apj , 779 , 182 qiu , k. , zhang , q. , menten , k. m. 2014 , apj , 794 , l18 zhang , q. , qiu , k. , girart , j. m. 2014 , apj , 792 , 116z | dust polarization observations in mm / submm continuum reveal plane - of - sky magnetic field morphologies dust grains aligned shorter axes parallel to magnetic field detected polarized emission perpendicular to b field despite recent observations magnetic field strengths not directly inferred from morphologies publications introduced @xmath0 local angle between magnetic field orientation and gradient dust continuum stokes @xmath1 emission @xmath0 interpreted as magnetic field alignment factor @xmath2 measures fraction magnetic field tension force directed along emission density gradient systematic features in @xmath3 across two regions observed with sma at 345 ghz higher - resolution observations with sma at 1.3 mm show condensations along mid - plane in g240.31 + 0.07 where @xmath3 maximised regions above and below plane showing small @xmath3-values material to mid - plane opposite trends in @xmath3 structures in g35.2 - 0.74n sources categorised as type iia ( g240 ) and iib ( g35.2n )maps @xmath3 ) for g240 g35.2 contours show dust continuum magnetic field emission gradient orientations displayed white black dotted segments dense condensations white circles in center of g240 fall onto regions with large @xmath3 material channeled from lower zones maps @xmath3 ) for g240 g35.2 contours show dust continuum magnetic field emission gradient orientations white black dotted segments dense condensations white circles center of g240 fall onto regions with large @xmath3 towards material channeled from lower zones examples not unique sample of 50 targets most sources show systematic features in @xmath3 polarization detected large area sample covers low- high - mass star - formation sites scales 0.1 to 0.01 pc larger scales around 1pc ( @xmath8 @xmath9 figure hist_delta shows histograms for source - averaged @xmath10 50 data points @xmath3 4000 independent beams 50 sources @xmath3-histogram normalized first before co - adding avoid statistics dominated by sources largest numbers independent beamshistograms show shoulder around @xmath10 @xmath3 falling towards larger values prevailing source - magnetic field orientation prefers misalignments with emission gradient magnetic fields within deviations of source axis histograms for source - averaged @xmath10 50 @xmath3 4000 measurements color - coded in red blue black sources identified as type iia or iib if @xmath3 interpreted with mhd force equation ratio between magnetic field tension force gravity @xmath11 calculated from @xmath3 as @xmath13 @xmath12 local angle between gradient dust continuum direction local gravity reflects molecular cloud morphology with angles @xmath0 @xmath12 measure of imprint forces in system figure displays source - averaged force ratios @xmath14 versus @xmath10 force ratio increases with @xmath10 force ratios @xmath15 vs @xmath16 for combined sma cso samples red blue symbols show identified iia- iib - type sources empty symbols sources less clear features black filled circles are identified sma - iii sources schematic magnetic field structures illustrated for iia iib- sourcescontours display dust continuum emission black - to - white grading indicates local @xmath3 opposite trends for type - iia -iib sources solid gray line log - linear fit @xmath17 @xmath18 @xmath19 dashed dotted gray lines indicate mean error @xmath20 @xmath21 bounds local angle @xmath3 prime observable across 50 star - forming regions linked to magnetic field - to - gravity force ratio @xmath11 maps @xmath11 correlated to @xmath3-maps tracer for local role magnetic field figure [ sigma_b_vs_delta ] shows source - averaged correlation approach provides local assessment magnetic field @xmath3 explain condensations star formation magnetic fields aligned parallel to emission gradients in clouds cores connects to planck results transition from fields parallel to ridges to orthogonal molecular clouds dotson j. l. vaillancourt kirby 2010 apjs , 186 , 406 plambeck kwon 2014 213 , 13 koch tang ho 2012a 747 , 79 kochtang y w ho p.t.p 2012b apj 747 80 koch p.m tang ho p.t.p 2013 775 77 koch p.m tang ho p.t.p 2014 apj 797 99 planck collaboration xxxii arxiv:1409.6728 qiu k. zhang q. wu 2009 apj 696 66 qiu zhang q. menten 2013 779 182 qiu zhang menten. 2014 794 l18 zhang qiu girart j m. 2014 apj 792 116z | 0.317305 | 2,000 | 1,010 | , Saving $0.1 in GPT-4. |
155,255 | in this paper we discuss the dynamics of the cosmological bartnick - mckinnon analogue with @xmath0 and @xmath1 .
we derive boundary conditions from energy considerations .
numerical simulations are carried out to show the existence of a non - stationary solution . | in general numerical simulations of the dynamics of soliton solutions in particle physics are of fundamental interest to test their stability .
specifically the dynamics of einstein - yang - mills systems help to understand the interplay of gravity and non - abelian gauge fields . as analytical solutions for the fully nonlinear equations are usually not available in these cases , only numerical simulations can provide further insight .
bartnik and mckinnon @xcite found that particle - like solutions for the einstein - yang - mills system exist , while that is not the case for either pure yang - mills or the pure einstein systems @xcite .
volkov et al .
@xcite investigated the einstein - yang - mills equations with cosmological constant @xmath2 to find an analogous class of solutions . in @xcite brodbeck et al .
performed the respective stability analysis of these static bartnick - mckinnon analogues .
first steps in simulating the dynamical evolution of the the ground state @xmath0 in the case of @xmath1 has been taken in @xcite .
different to the approach of @xcite we use a finite element scheme for our numerical simulations and a reformulated differential equation .
special emphasis is on the boundary conditions in relation of the conservation of energy . in our approach
we consider the static de sitter manifold only .
the paper is organized as follows . in section [ sec : eq ] we discuss the physical field equation that governs the dynamics of the monopole . in section [ sec : bnd ]
we derive the initial- and boundary conditions in relation to the particle energy .
the discretisation and related stability issues are discussed in section [ sec : disc ] .
finally numerical results are given in section [ sec : results ] .
in their analysis of einstein - yang - mills fields with a cosmological constant volkov et al .
@xcite found that three different classes of solutions exist .
these are static , spherically symmetric and are characterized by their value of @xmath2 and the number of nodes @xmath3 .
the three different classes can be distinguished by their values of @xmath2 .
for @xmath4 the solutions approach asymptotically the de sitter geometry .
the solutions with @xmath5 are the so called `` bag of gold solutions '' .
they have finite size , but still an horizon .
all the solutions with non - vanishing @xmath6 have non - vanishing magnetic charge . at @xmath7
the solutions are compact solutions on a de sitter manifold , where magnetic charges are no longer defined .
the smooth , spherically symmetric yang - mills field has a su(2 ) gauge group .
the components of the gauge field in spherical coordinates for the abelian gauge are @xmath8 [ feq ] where @xmath9 is the coupling constant of the yang - mills ( ym ) field and the @xmath10 are the generators of the su(2 ) group , @xmath11 being the pauli matrices .
this gauge has been chosen as the yang - mills field in this case only depends on @xmath12 and @xmath13 , which reflects the spherical symmetry of the field .
the de sitter manifold is defined by its metric . in conformal
coordinates @xcite it is given by @xmath14 the coordinates ranges are @xmath150,\pi[$ ] , @xmath16 , @xmath17 and @xmath18 .
its curvature is given by @xmath19 , where @xmath2 is the cosmological constant .
the field equations can be derived through using the variational principle on the appropriate lagrangian .
the yang - mills lagrangian is given by @xmath20.\end{aligned}\ ] ] @xmath21 is the associated 2-form field of the gauge field and is also called the field strength tensor and @xmath22 $ ] is the product of the @xmath23 lie algebra .
the equations for the coupled einstein - yang - mills system have been derived in @xcite , as well as the equations for the decoupled yang - mills field on a static de sitter manifold @xmath24 here the dot denotes the derivative with respect to time and prime the derivative with respect to @xmath12 .
this is a semi - linear partial differential equation of hyperbolic type .
in this paper we investigate the numerical solution of equation .
it is defined on the compact domain @xmath25_{(\mathbb{r } } , \notag\end{aligned}\ ] ] and requires initial conditions for both @xmath26 and @xmath27 . for the initial conditions for @xmath26 we will chose the stationary solution given in equation , for the initial conditions for @xmath27
we introduce a perturbation of the stationary system .
boundary conditions follow from energy considerations , see below . as we have already mentioned before we are analysing the dynamical evolution of the static bartnik - mckinnon analogue with @xmath0 and @xmath28 .
it is given by @xmath29 thus it represents our initial condition for @xmath26 . in the following we set the coupling constant @xmath30 . as the initial condition for @xmath27
we choose @xmath31 with an arbitrary function @xmath32 , which was chosen in different numerical experiments differently .
the boundary conditions for the system will be derived from the finiteness of the energy . for systems on a minkowski space - times , these types of boundary conditions are related to the asymptotic behavior of the field at infinity .
these conditions are implemented by an appropriate choice of the functions - space , e.g. @xmath33 , etc . as for the minkowski space - times
, for the compact de sitter manifold the energy finiteness of @xmath26 will be related to an appropriate choice of the function space . as we are investigating a spherical symmetric yang - mills field
the energy is independent of these variables . in @xcite chrusciel et al .
derived a priori estimates of the solution space of yang - mills equations on globally hyperbolic four dimensional lorentz manifolds .
they show that the yang - mills gauge field @xmath36 belongs to the sobolev space @xmath37 with @xmath38 and an appropriate constant @xmath39 . from the assumed symmetry ( @xmath36 is independent of @xmath34 , @xmath35 ) and
it follows @xmath40 with @xmath41 the energy of the yang - mills field in the decoupled case is given by @xmath42 where @xmath43is the energy momentum tensor and @xmath44 , @xmath45 are the normal of the domain pointing outside and a killing vector respectively . with the constant of proportionality @xmath46 we rewrite the energy as @xmath47 due to the continuity of @xmath48 with respect to @xmath12 , the finiteness of @xmath49 implies @xmath50 , @xmath51 and @xmath52 .
we choose @xmath53 assuming the boundary conditions and a field @xmath48 governed by equation , it is easily varified that the energy is conserved at all times . with it
follows @xmath54 as for the initial condition , we assume vanishing spatial derivatives of the solution at @xmath55 for all times .
this follows from the assumption that the solution should describe a smooth solution on the three - dimensional sphere requiring tangential planes at the poles corresponding to @xmath55 .
we therefore make the ansatz @xmath56 with the unknown function @xmath57 . from
follows the equation for @xmath58 @xmath59 with the boundary and initial conditions @xmath60 the following two terms in equation need special attention : the scaling factor of dalembert operator ( @xmath61 vanishes at the boundary @xmath55 and the source term linear in @xmath58 has infinite weights at the boundaries .
the first difficulty is in view of a non - vanishing source at the boundary unproblematic .
assuming the same regularity for @xmath58 as we have for @xmath26 , i.e. @xmath62 , the linearly diverging weight of the linear source term is also not problematic .
to integrate forwards in time , we transformed the equation to a set of first order equations , which we integrate by two different implicit schemes .
the first is the classical implicit euler methods , the second a modification of a two - stage runge - kutta method ( alexander scheme)@xcite , first introduced by in @xcite . for the discretisation of the spatial dimension we chose the conforming finite element discretisation with continuous , piece - wise linear basis functions .
we used a equidistant grid with mesh - size @xmath63 , @xmath64 .
the spatial integrals have been carried out with _
maple _ with the exception of the fourth term in equation .
this term can not be integrated analytically and has been treated by appropriate integration rules ( simpson rule ) .
the numerical approximation of these integrals has been investigated and chosen appropriately to avoid numerical artifacts .
additionally it was necessary to use taylor expansions for the sine and cosine terms in the discretisation of the two non - linear terms in f. as an illustration for this issue we discuss the term cubic in f. its discretisation is given by @xmath65 where @xmath66 denotes the @xmath67 scalar product , @xmath68 and @xmath69 denote the piecewise linear basis functions .
for @xmath70 it has the form @xmath71 .
all the terms are similar , so we concentrate on @xmath72.\end{aligned}\ ] ] here special attention is required to control round - off errors . to accurately account for the difference
@xmath73 an expansion of the trigonometric functions had to be carried out .
if was found that an expansion up to the sixth order in @xmath74 was sufficient , leading to an evaluation of @xmath75 according to @xmath76 numerical experiments have shown that a further increase of the order were not necessary .
the simulations have been carried out with a software tool _ seyms-1d _ ( imple instein - ang - ills olver for one dimension ) , a c++-code .
a documentation is available with @xcite .
several numerical experiments have been carried out . to validate to code , the conservation of energy has been considered .
its behavior over time is displayed in fig .
[ fig : enhv ] on a sequence of successively refined grids . as a second numerical example we investigated the behavior of the field resulting from the initial conditions with @xmath77
the amplitude has been chosen large enough in order to show the full non - linear dynamics of the system without destroying the original structure of the field .
the simulations has been carried out on a @xmath78 grid ( using same notation as above ) using the implicit euler method .
it is displayed in fig .
[ fig : s10 ] .
the result in terms of the original field @xmath26 is shown in fig . [
fig : s10w ] .
as can be seen clearly , energy transfer takes place between the different modes resulting in a strongly perturbed steady - state solution .
no plane wave solutions can be observed .
this is a property intrinsic to yang - mills fields and has already been seen on flat manifolds .
firstly the authors would like to thank the referee for his valuable comments .
furthermore we would like to thank matthias bartelmann , jrg frauendiener and ralf peter for continuous support and helpful discussions .
additionally we would like to thank jonathan coles for finding the `` last '' programming error and thomas gehrmann for his helpful remarks .
r. stevenson , an overview and reproduction of bartnick and mckinnon s discovery of particlelike solutions to the einstein - yang - mills equations , ( unpublished , laplace.physics.ubc.ca/people/roland/papers/bartnik/bmsols.ps ) r. peter , kugelsymmetrische einstein - yang - mills - systeme auf de sitter - artigen mannigfaltigkeiten ( unpublished , diploma thesis , university of tbingen , 2005 ) , http://www.tat.physik.uni-tuebingen.de/@xmath79rpeter/ k. johannsen , numerische aspeckte dichtegetriebener strmungen in porsen medien ( unpublished , professorial dissertation , university of heidelberg , 2004 ) , http://www.klaus-johannsen.net/publications.html h. lux , numerical simulations of spherically symmetric einstein - yang - mills - systems on a de sitter mannifold ( unpublished , diploma thesis , university of heidelberg , 2006 ) , http://www.itp.uzh.ch/@xmath79lux/ | numerical simulations soliton solutions particle physics stability einstein - yang - mills systems interplay gravity non - abelian gauge fields analytical solutions nonlinear equations numerical simulations insight bartnik mckinnon found particle - like solutions for einstein - yang - mills system exist not pure yang - mills einstein systems @xcite volkov investigated einstein - yang - mills equations with cosmological constant @xmath2 analogous solutions brodbeck et al performed stability analysis static bartnick - mckinnon analogues first steps simulating dynamical evolution ground state @xmath0 @xmath1 @xcite finite element scheme simulations reformulated differential equation emphasis on boundary conditions conservation energy consider static de sitter manifold paper organized section sec : eq physical field equation dynamics monopole bnd initial- boundary conditions particle energy discretisation stability issues sec : disc ] numerical results section sec : results ] analysis einstein - yang - mills fields cosmological constant volkov found three classes of solutions static spherically symmetric characterized by value @xmath2 number of nodes @xmath3three classes distinguished by values @xmath2 @xmath4 solutions approach de sitter geometry solutions @xmath5 are bag of gold solutions finite size horizon solutions with non - vanishing @xmath6 have magnetic charge @xmath7 solutions compact on de sitter manifold magnetic charges no longer defined spherically symmetric yang - mills field has su(2 ) gauge group components are @xmath8 ] @xmath9 coupling constant @xmath10 generators su(2 ) group @xmath11 pauli matrices gauge depends on @xmath12 @xmath13 spherical symmetry de sitter manifold defined by metric given by @xmath14 coordinates ranges @xmath150 @xmath16 @xmath18 curvature given by @xmath19 @xmath2 cosmological constant field equations derived variational principle on lagrangian yang - mills lagrangian given by @xmath20 @xmath21 associated 2-form field gauge field field strength tensor @xmath22 product of @xmath23 algebraequations coupled einstein - yang - mills system derived in @xcite decoupled yang - mills field on static de sitter manifold @xmath24 dot denotes derivative time prime @xmath12 semi - linear partial differential equation hyperbolic type numerical solution defined on compact domain @xmath25 requires initial conditions for @xmath26 @xmath27 @xmath26 chose stationary solution @xmath27 introduce perturbation stationary system boundary conditions follow from energy considerations analysing dynamical evolution static bartnik - mckinnon analogue with @xmath0 and @xmath28 given by @xmath29 represents initial condition for @xmath26 set coupling constant @xmath30 initial condition @xmath27 choose @xmath31 with arbitrary function @xmath32 chosen boundary conditions derived from finiteness energy systems minkowski space - times boundary conditions related to asymptotic behavior field at infinity implemented by appropriate choice functions - space @xmath33 energy finiteness of @xmath26 related to appropriate choice function space investigating spherical symmetric yang - mills field energy independent of variableschrusciel derived estimates solution space yang - mills equations hyperbolic four dimensional lorentz manifolds yang - mills gauge field @xmath36 belongs sobolev space @xmath37 @xmath38 constant @xmath39 assumed symmetry @xmath36 independent of @xmath34 @xmath40 energy yang - mills field given by @xmath42 energy momentum tensor normal domain pointing outside killing vector constant proportionality @xmath46 rewrite energy as @xmath47 continuity @xmath48 finiteness @xmath49 implies @xmath50 @xmath51 @xmath52 choose @xmath53 boundary conditions field @xmath48 governed equation energy conserved @xmath54 assume vanishing spatial derivatives solution at @xmath55 solution smooth solution three - dimensional sphere tangential planes poles @xmath55 ansatz @xmath56 unknown function @xmath57 equation for @xmath58 @xmath59 boundary initial conditions @xmath60 terms attention scaling factor dalembert operator vanishes at boundary source term linear @xmath58 infinite weights at boundariesfirst difficulty non - vanishing source boundary unproblematic regularity @xmath58 @xmath26 @xmath62 linearly diverging weight linear source term not problematic integrate forwards transformed equation to first order equations two implicit schemes first classical euler methods second modification two - stage runge - kutta method alexander scheme spatial dimension conforming finite element discretisation continuous piece - wise linear basis functions used equidistant grid mesh - size @xmath63 @xmath64 spatial integrals carried with maple _ exception fourth term term integrated analytically treated by integration rules simpson rule numerical approximation integrals investigated avoid numerical artifacts taylor expansions for sine cosine terms discretisation two non - linear terms f term cubic in f discretisation by @xmath65 @xmath66 @xmath67 scalar product @xmath68 @xmath69 piecewise linear basis functions @xmath70 form @xmath71 terms similar concentrate on @xmath72 attention control round - off errors account difference @xmath73 expansion trigonometric functionsfound expansion to sixth order in @xmath74 sufficient evaluation of @xmath75 according to @xmath76 experiments shown further increase not necessary simulations carried out with software tool seyms-1d _ c++-code documentation available with @xcite numerical experiments carried out code conservation of energy considered behavior over time displayed in fig . [ ] on refined grids investigated behavior field from initial conditions with @xmath77 amplitude chosen large enough show full non - linear dynamics without destroying original structure simulations carried on @xmath78 grid implicit euler method displayed in fig . [ s10 ] result original field @xmath26 shown in fig . [ fig s10w ] energy transfer between modes strongly perturbed steady - state solution no plane wave solutions observed property intrinsic to yang - mills fields seen on flat manifolds authors thank referee for comments matthias bartelmann jrg frauendiener ralf peter for support discussions jonathan coles programming error thomas gehrmann for remarksstevenson overview bartnick mckinnon particlelike solutions einstein yang mills equations unpublished.physics.ubc peter kugelsymmetrische einstein yang mills systeme mannigfaltigkeiten unpublished diploma university tbingen 2005-tuebingen johannsen numerische aspeckte dichtegetriebener strmungen porsen medien unpublished dissertation university heidelberg 2004-johannsen lux numerical simulations spherically einstein - yang - mills systems sitter mannifold unpublished diploma thesis university heidelberg 2006.uzh | 0.43506 | 2,765 | 1,397 | , Saving $0.1 in GPT-4. |
155,257 | we give a parameterization , using jacobi s elliptic functions , of alfred gray s elliptical catenoid and elliptical hellicoid that avoids some problems present in the original depiction of these surfaces . | on of the alfred gray s favorite tools was the solution to bjrling s problem : _ given a planar analytic curve , find a minimal surface in @xmath0 that contains it as a geodesic_. the weierstrass representation provides an explicit solution : if @xmath1 is a parameterization of the curve , the parameterization @xmath2 gives the solution , where @xmath3 , @xmath4 are extensions of @xmath5 , @xmath6 to functions of a complex variable @xmath7
. we will call it the bjrling surface of the curve . when one takes , instead of the real part , the imaginary part of the expression , one gets another minimal surface , known as the conjugated surface of the first one .
it is well know how alfred used this to produce many beautiful surfaces .
but he could use it also as a theoretical tool : one day we asked alfred what order of contact could a minimal surface have at a self - intersection point ; without a second of thought he replied : `` any order of contact : take a planar curve with a self - contact of order @xmath8 and solve bjrling s problem '' .
an instant theorem!. formula ( [ eqn : bjor ] ) can be directly given to the computer .
however , the fact that the integrand is in many cases multivaluated poses some problems when we try to get the global picture .
take , for instance , the parabola @xmath9 .
the formula gives the following parameterization of its bjrling surface @xmath10 which the reader can immediately give to his favorite graphics package and get the wrong picture !
( figure 1a , where the thick line shows the parabola ) .
the problem has to do with the integrand that is multivalued and branches at the points @xmath11 and @xmath12 .
when we integrate from 0 to , say , @xmath13 along differents paths we can get different values , all with the same imaginary part .
this produces a sharp edge , which is impossible in a minimal surface .
the problem is more evident in the conjugate surface : we get a discontinuity which the computer fills with a planar face ( figure 1b ) .
the problem can be easily solved by a simple sustitution @xmath14 which makes the function single valued and the surface becomes : @xmath15 which gives the correct picture ( figure 2 ) .
the surface continues beyond the sharp edges which becomes lines of self - intersection and then takes a turn , giving a periodic pattern .
the surface turns out to be the well - know catalan s surface that contains a cycloid perpendicular to the parabola which is another planar geodesic of the surface .
figure 2b shows the correct conjugate surface .
the above relation shows an interesting relation between the parabola and the cycloid : they have the same bjrling surface .
recall that a plane cuts a minimal surface along a geodesic , if and only if , it is a plane of symmetry of the surface .
now , the bjrling surface of a curve that has a line of symmetry has to planes of symmetry : the original one containing the curve and the plane perpendicular to it that contains the line of symmetry . in formula
( ? ) the original curve is obtained by restricting to real values of @xmath7 and the orthogonal geodesic by restricting to purely imaginary values of @xmath7 .
the two curves are in a sense , dual to each other . to be more precise : the duality is between objects consisting each of a curve and a point of intersection of the curve with a line of symmetry . we can call those pairs _ bjrling duals _ to each other . for example , the circle and the catenary are bjrling duals ( the common surface is the catenoid ) and so are the parabola and the cycloid with its line of symmetry that cuts t on one of the highest ( smooth ) points of the cycle .
let us apply ( 1 ) to the ellipse with semi - axes @xmath16 and @xmath17 @xmath18 , and excentricity @xmath19 .
we obtain @xmath20 and with it the computer produces figure 3 .
one observes that this surface has various _ channels _ where one of the principal curvatures is close to zero and the other one is very big .
again , this can not be a minimal surface !
( one obtains a channel or an edge depending on the parity of the size of the grid used to plot the surface ) .
the conjugate surface presents again a discontinuity which if filled by the computer with a flat face ( figure 3b ) alfred s depiction in @xcite of these surfaces , called by him _ elliptical catenoid _ and _ elliptical hellicoid _ , have the same problems .
( those two loves of his , the computer and the bjrling problem , did not get togheter as smoothly as he thought ) . both maple ( which we use ) and mathematica give the same picture .
maybe alfred was aware of the problem since in this case he used an elliptic function instead of applying directly the formula .
the analysis must , however , be carried some steps further : we use jacobi s elliptic functions ( for details , see @xcite , @xcite ) : jacobi s elliptic sine function @xmath21 , being the inverse function of the integral @xmath22 jacobi s elliptic cosine @xmath23 and jacobi s _ delta _ function @xmath24 like the trigonometric funtions ( which we obtain as the particular case @xmath25 ) they can be extended to complex functions , but the extensions are now meromorphic and doubly periodic instead of being holomorphic and periodic .
the integrand is still a meromirphic function , but its residues are all @xmath29 @xcite , so the integral is a well - defined function of @xmath26 .
this parameterization produces the correct picture of the elliptical catenoid .
the ellipse has two lines of symmetry and , correspondly , two bjrling duals : one resembles a cycloid , the other one a catenary .
figure 4 shows the correct pictures of the elliptical catenoid and the elliptical hellicoid : 99 a. gray , _ modern differential geometry of curves and surfaces _ , crc press , 2nd edition , 1998 .
d. f. lawden , _ elliptic functions and applications _ ,
applied mathematical sciences 98 , springer - verlag , 1989 .
markusevich , _ the remarkable sine functions _ , elsevier , 1966 . | alfred gray s solution to bjrling s problem given planar analytic curve find minimal surface in @xmath0 geodesic weierstrass representation provides explicit solution @xmath1 parameterization of curve @xmath2 gives solution @xmath3 @xmath4 extensions of @xmath5 @xmath6 to functions complex variable @xmath7 bjrling surface of curve takes imaginary part expression gets another minimal surface conjugated surface of first one alfred used beautiful surfaces theoretical tool asked order of contact minimal surface at self - intersection point any order contact take planar curve with self - contact of order @xmath8 solve bjrling s problem instant theorem formula ( [ eqn : bjor ] ) given to computer integrand multivaluated poses problems global picture parabola @xmath9 formula gives parameterization of bjrling surface @xmath10 reader wrong picture figure 1a line shows parabola problem integrand multivalued branches at points @xmath11 and @xmath12 integrate from 0 to @xmath13 paths different values with same imaginary part produces sharp edge impossible in minimal surface problem evident in conjugate surface discontinuity computer fills with planar face ( figure 1b ) solved by sustitution @xmath14 function single valued surface becomes @xmath15 correct picture ( figure 2 ) surface continues beyond sharp edges lines of self - intersection takes turn periodic pattern surface catalan s surface cycloid perpendicular to parabola planar geodesic figure 2b shows correct conjugate surface between parabola cycloid same bjrling surface plane cuts minimal surface along geodesic if plane of symmetry bjrling surface of curve line of symmetry has planes of symmetry original perpendicular line symmetry formula original curve obtained by restricting to real values of @xmath7 orthogonal geodesic imaginary values of @xmath7 two curves dual to each other duality between objects curve point of intersection curve with line of symmetry pairs bjrling duals circle catenary are bjrling duals common surface catenoid parabola and cycloid with line of symmetry on highest cycle apply ( 1 ) to ellipse with semi - axes @xmath16 @xmath17 @xmath18 excentricity @xmath19 obtain @xmath20 computer produces figure 3 surface has various _ channels _ one principal close to zero other big not minimal surface obtains channel or edge depending on size grid conjugate surface presents discontinuity filled computer with flat face ( figure 3b ) alfred s depiction in @xcite surfaces _ elliptical catenoid _ and elliptical hellicoid _ have same problems computer bjrling problem smoothly maple and mathematica give same picture alfred used elliptic function instead of formula analysis further use jacobi s elliptic functions @xcite s elliptic sine function @xmath21 inverse function of integral @xmath22 s elliptic cosine @xmath23 s delta _ function @xmath24 trigonometric funtions extended to complex functions extensions meromorphic doubly periodic instead holomorphic periodic integrand meromirphic function residues are all @xmath29 @xcite integral well - defined function of @xmath26 parameterization produces correct picture of elliptical catenoid ellipse has two lines of symmetry two bjrling duals one resembles cycloid other catenaryfigure 4 elliptical catenoid hellicoid gray differential geometry curves surfaces crc press 2nd edition 1998 lawden elliptic functions applications applied mathematical sciences springer - verlag 1989 markusevich sine functions elsevier 1966 | 0.264062 | 1,436 | 744 | , Saving $0.0 in GPT-4. |
155,260 | saturation of gluon distribution is a consequence of the non - linear evolution equations of qcd .
saturation implies the existence of so called saturation momentum which is defined as a gluon density per unit rapidity per transverse area . at large energies for certain kinematical domains
saturation momentum is the only scale for physical processes . as a consequence different observables exhibit geometrical scaling ( gs ) .
we discuss a number of examples of gs in different reactions . | at the eqcd meeting in 2013 @xcite we have discussed the emergence of geometrical scaling @xcite for @xmath0 in deep inelastic scattering ( dis ) @xcite and for charged particle distributions in proton collisions @xcite . here , after short reminder , we extend this analysis to @xmath1 correlation @xcite and to heavy ion collisions ( hi ) @xcite .
references @xcite , @xcite@xcite include a more complete bibliography of the subject .
geometrical scaling hypothesis means that some observable @xmath2 that in principle depends on two independent kinematical variables , say @xmath3 and @xmath4 , in fact depends only on a specific combination of them denoted as @xmath5 : @xmath6 here function @xmath7 in eq .
( [ gsdef ] ) is a dimensionless function of scaling variable @xmath8 and @xmath9 is the saturation scale .
@xmath10 is a transverse area that corresponds to the overlap of hadrons colliding at fixed impact parameter @xmath11 ( or integrated over @xmath12 ) , or like in the case of dis it is a cross section for large dipole scattering on a proton .
@xmath13 and @xmath14 in eq .
( [ qsat ] ) are free parameters , which can be extracted from the data within some specific model for @xmath2 , and parameter @xmath15 is a dynamical quantity of the order of @xmath16 . here
we shall test the hypothesis whether different pieces of data can be described by formula ( [ gsdef ] ) with _ constant _ @xmath15 , and what is the range of transverse momenta where gs is working satisfactorily . throughout this paper
we shall be neglecting logarithmic energy dependence due to the running of @xmath17 .
let us start with dis where the relevant scaling observable is @xmath18 @xcite . in fig .
[ gsdis ] we plot @xmath19 as a function of @xmath4 ( left panel ) and in terms of @xmath5 for @xmath20 ( right panel ) for combined hera data @xcite .
different points correspond to different bjorken @xmath3 s .
we see from fig .
[ gsdis ] that points of different bjorken @xmath3 s scale very well with some exception in the right part of fig .
[ gsdis].b .
these points , however , correspond to large bjorken @xmath3 s where gs is supposed to break . .
different points forming a wide band as a function of @xmath4 in the left panel correspond to different bjorken @xmath3 s .
they fall on a universal curve when plotted in terms of @xmath5 ( right panel ) .
( figure from the first paper of ref .
@xcite).,title="fig:",width=235 ] .
different points forming a wide band as a function of @xmath4 in the left panel correspond to different bjorken @xmath3 s .
they fall on a universal curve when plotted in terms of @xmath5 ( right panel ) .
( figure from the first paper of ref .
in hadronic collisions at c.m .
energy @xmath22 particles are produced in the scattering process of two patrons characterized by bjorken @xmath3 s @xmath23 for central rapidities @xmath24 .
geometrical scaling in this case means simply that @xcite : @xmath25 where @xmath7 is a universal dimensionless function of the scaling variable @xmath26 in fig .
[ gsalice ] we plot alice pp data @xcite in terms of @xmath21 ( left panel ) and in terms of scaling variable @xmath5 ( right panel ) for @xmath27 .
we have found by a model independent analysis that the optimal exponent @xmath28 @xcite , which is smaller than in the case of dis .
why this so , remains to be understood . and
full ( black ) circles correspond to @xmath30 tev , down ( red ) triangles to 2.76 tev and up ( blue ) triangles to 0.9 tev.,title="fig:",width=226 ] and @xmath29 .
full ( black ) circles correspond to @xmath30 tev , down ( red ) triangles to 2.76 tev and up ( blue ) triangles to 0.9 tev.,title="fig:",width=226 ] an immediate consequence of gs for the @xmath21 spectra is a power - like growth of multiplicity with energy .
indeed , since @xmath31 where the _ average _ saturation scale is defined as@xmath32 one arrives at @xmath33 .\ ] ] data indeed support the power - like growth of inelastic multiplicity as @xmath34 as predicted by gs by eq .
( [ qbarsat ] ) for @xmath28 .
in pp collisions at 7 tev ( full black circles ) , 2.76 tev ( full red down - triangles ) , 0.9 tev ( full blue up - triangles ) and in ppb collisions at 5.02 tev ( full brown diamonds ) plotted in terms of scaling variable @xmath35 . for pp @xmath36 tev and for ppb @xmath37 tev .
( figure from the second paper of ref .
@xcite . ) ] another consequence of eq .
( [ gsinpp ] ) is that @xcite @xmath38 which means that @xmath39 rises with energy as @xmath40 , which is indeed seen in the data . on the other hand , since the saturation momentum is by eq .
( [ qbarsat ] ) equal to the gluon density per transverse area , one easily derive the correlation between mean @xmath21 and charged particles multiplicity at given energy @xmath41 @xcite : @xmath42 by fixing multiplicity , one is probing some fixed impact parameter that corresponds to the overlap transverse area @xmath43 that itself is by construction both multiplicity and energy dependent .
therefore one needs a model for @xmath43 . to this end
we have used the color glass condensate result for pp and pa collisions @xcite .
the result is plotted in fig .
[ fig : scaled ] where we plot alice data @xcite as a function of scaling variable defined in eq .
( [ ptwdep ] )
. in the right panel the same distributions are scaled according to eq . ( [ multhi ] ) . ,
title="fig:",width=230 ] . in the right panel
the same distributions are scaled according to eq .
( [ multhi ] ) . ,
title="fig:",width=230 ] gs for particle spectra in hi collisions has been already discussed in ref .
@xcite and in ref .
@xcite for photons .
hi data are divided into centrality classes that select events within certain range of impact parameter @xmath11 . in this case both transverse area @xmath10 and the saturation scale @xmath44 acquire additional dependence on centrality that is characterized by an average number of participants @xmath45 .
we have @xcite:@xmath46 therefore in hi collisions@xmath47 in fig .
[ fig : all ] we plot lhc and rhic data in terms of @xmath48 ( left panel ) and @xmath29 for @xmath49 ( right panel ) .
one can see an approximate scaling of , however , worse quality than in the pp case . to summarize : a wealth of data in hadronic collisions exhibit gs .
this may be interpreted as a signature of saturation .
however some details , like the non - universality of the value of @xmath15 , remain to be understood . | eqcd meeting 2013 discussed geometrical scaling @xmath0 deep inelastic scattering charged particle distributions proton collisions analysis to @xmath1 correlation heavy ion collisions references include complete bibliography geometrical scaling hypothesis means observable @xmath2 depends on two independent kinematical variables @xmath3 @xmath4 depends on specific combination @xmath5 : @xmath6 function @xmath7 dimensionless function of scaling variable @xmath8 @xmath9 saturation scale @xmath10 transverse area overlap of hadrons colliding fixed impact parameter @xmath11 cross section for large dipole scattering on proton @xmath13 @xmath14 free parameters extracted from data specific model @xmath2 parameter @xmath15 dynamical quantity order @xmath16 test hypothesis different pieces data described by formula ( gsdef ] ) with constant @xmath15 range of transverse momenta where gs working satisfactorily neglecting logarithmic energy dependence running @xmath17 start with dis relevant scaling observable @xmath18 @xcite fig .plot @xmath19 function @xmath4 left panel @xmath5 @xmath20 right panel combined hera data @xcite different points correspond different bjorken @xmath3 s fig points different bjorken @xmath3 s scale well exception right part points correspond large bjorken @xmath3 s gs break points wide band @xmath4 left panel correspond different bjorken @xmath3 s fall universal curve plotted @xmath5 right panel ) figure first paper points wide band @xmath4 left panel correspond different bjorken @xmath3 s fall universal curve plotted @xmath5 right panel ) hadronic collisions c energy @xmath22 particles produced scattering process two patrons bjorken @xmath3 s @xmath23 central rapidities @xmath24 geometrical scaling means @xcite : @xmath25 @xmath7 universal function scaling variable @xmath26 fig plot pp data @xcite @xmath21 left panel scaling variable @xmath5 right panel @xmath27 found analysis optimal exponent @xmath28 @xcite smaller dis remains understoodfull black circles correspond @xmath30 tev red triangles 2.76 tev blue triangles 0.9 tev @xmath29 full black circles correspond @xmath30 tev red triangles 2.76 blue 0.9 tev consequence gs @xmath21 spectra power - like growth multiplicity with energy @xmath31 average saturation scale@xmath32 arrives at @xmath33 data support power - like growth inelastic multiplicity @xmath34 predicted by gs eq @xmath28 pp collisions at 7 tev black circles 2.76 tev red triangles 0.9 tev blue triangles ppb collisions 5.02 tev brown diamonds scaling variable @xmath35 pp @xmath36 tev ppb @xmath37 tev figure second paper consequence eq @xcite @xmath38 @xmath39 rises with energy @xmath40 seen in data saturation momentum equal to gluon density per transverse area correlation between mean @xmath21 charged particles multiplicity at given energy @xmath41 @xmath42 fixing multiplicity fixed impact parameter overlap transverse area @xmath43 construction multiplicity energy dependent needs model for @xmath43used color glass condensate result for pp pa collisions @xcite result plotted in fig . alice data @xcite function of scaling variable defined in eq right panel distributions scaled according to eq distributions scaled according eq gs for particle spectra in hi collisions discussed in ref . @xcite for photons hi data divided into centrality classes events within range impact parameter @xmath11 transverse area @xmath10 saturation scale @xmath44 acquire dependence on centrality by average number of participants @xmath45 @xcite:@xmath46 in hi collisions@xmath47 fig plot lhc rhic data @xmath48 left @xmath29 @xmath49 approximate scaling worse quality than pp case wealth data in hadronic collisions exhibit gs signature of saturation details non - universality of value of @xmath15 to be understood | 0.265836 | 1,698 | 870 | , Saving $0.0 in GPT-4. |
155,268 | the present - day state of the milky way disk can tell us much about the history of our galaxy and provide insights into its formation . we have constructed a high - precision catalogue of disk stars using data from the sloan digital sky survey ( sdss ) and use these stars to probe the heating history as well as investigating the detailed phase - space distribution .
we also show how this sample can be used to probe the global properties of the milky way disk , employing the jeans equations to provide a simple model of the potential close to the disk .
our model is in excellent agreement with others in the literature and provides an indication that the disk , rather than the halo , dominates the circular speed at the solar neighborhood .
the work presented in these proceedings has been published as `` slicing and dicing the milky way disc in sdss '' @xcite . | large surveys such as sdss are beginning to enable detailed studies to be made of the properties of the milky way .
it is now possible to construct high - precision catalogues of stars with full 3d kinematics and from these analyse the structure and evolution of our galaxy . in the following proceedings
i summarise our study of the disk of the milky way ( presented in full in * ? ? ?
* ) , highlighting two of the most important results .
we use data from the 7th public data release of the sdss , concentrating on the 250 square degree stripe 82 region for which it is possible to obtain high precision photometry and proper motions @xcite .
this catalogue has been exploited in a variety of works , covering white dwarfs @xcite , halo kinematics @xcite and rr lyrae in the outer halo @xcite .
we take all dwarf stars with good quality spectra @xcite , estimating distances using a slightly modified version of the @xcite photometric parallax relation .
our modification concerns the turn - off correction ( equation a6 of * ? ? ?
* ) , which we argue is not applicable to disk stars ( see the appendix of * ? ? ?
we then calculate the three - dimensional velocities for our sample , propagating all uncertainties , and obtain final errors of around 20 to 30 @xmath0 for each component of the velocity . as we are interested in the kinematics of the disk as a function of height from the plane , we restrict ourselves to @xmath1 .
we also split the data into three ranges in metallicity ( @xmath2}\le -0.8 , -0.8 \le { [ { \rm fe / h}]}\le -0.5 $ ] and @xmath3}$ ] ) , and then for each metallicity bin we further divide the data into four ranges in z out to a maximum distance of 2 kpc . the stars are equally divided between the four distance bins , resulting in in around @xmath4 to @xmath5 stars per bin from our full sample of 7280 stars .
this binning is necessary in order to avoid having to model the sdss spectroscopic selection function .
once we have divided up our stars , we then proceed to fit the distribution of kinematics using maximum likelihood methods ( incorporating the uncertainties on the individual velocities ) .
both @xmath6 and @xmath7 are fit using gaussian distributions , but since it is well known that the distribution of @xmath8 for the disk is highly skewed and non - gaussian we adopt an asymmetric functional form ( equation ( 15 ) of * ? ? ?
when carrying out this analysis it is crucial to remove contamination from halo stars , particularly for the most metal - poor bins .
we do this by including a non - rotating gaussian component into the velocity fits , where the level of contamination ( and hence amplitude of the gaussian ) is assumed to be twice the number of counter - rotating stars .
the halo dispersions are fixed according to the values found in @xcite .
a comprehensive analysis of the results are presented in @xcite . here
we concentrate on the most important properties , starting with rotation lag .
we are able to trace the lag ( fig .
[ fig : lag ] ) to a couple of kpc outside the plane and find that it follows the well - known asymmetric drift relation .
the hotter ( metal - poor ) stars exhibit greater lag than the colder ( metal - rich ) populations , resulting in a clear correlation between lag and metallicity .
the gradient of the lag with respect to @xmath9z@xmath9 varies from around 15 to 40 @xmath10 , depending on metallicity .
[ fig : lag ] also shows how the lag is correlated with @xmath11 . for the solar - neighbourhood
it is know that this is a linear relation @xcite , but as we can see from our figure this is no longer true once we move beyond 0.5 kpc .
however , it can be seen that there is still a relatively tight correlation with @xmath11 , which is independent of metallicity . .
figure taken from @xcite.,title="fig:",width=173 ] . figure taken from @xcite.,title="fig:",width=173 ]
as has been shown by many authors , most notably in the seminal work of @xcite , the vertical velocity dispersion profile can be used to constrain the gravitational potential of the disk .
we show the results derived from our data in fig .
[ fig : potential ] , where we have included an additional solar - neighbourhood metal - rich data point derived from the geneva - copenhagen survey @xcite .
we take a simplified two - parameter model for the potential , consisting of an infinite razor - thin sheet embedded in a constant background .
our best - fit model estimates that the mass density of these two components are @xmath12 and @xmath13 , respectively . as can be seen from fig .
[ fig : potential ] , our measurement of the potential is in good agreement with existing models , especially model 1 of @xcite .
this is the least halo - dominated model of @xcite , with the disk ( which has a relatively short scale - length ) dominating the circular speed at the solar neighborhood .
interested readers should consult section 2.7 of @xcite for a detailed comparison of these models .
note that this background mass density is close to the @xmath14 predicted using isothermal spherical halo models ( equation 4.279 of * ? ? ?
if we assume our background mass represents the dark halo , it corresponds to a local dark matter density of 0.57 @xmath15 , which is noticeably larger than the canonical value of 0.30 @xmath15 typically assumed ( e.g. * ? ? ?
our analysis adds still more weight to the argument the local halo density may be substantially underestimated by this generally accepted value .
perhaps more robust than the local mass density is the surface mass density . by integrating our mass distribution
we obtain a total surface mass density of @xmath16 , which agrees well with the classical value of @xmath17 from @xcite .
if we integrate beyond 1.1 kpc , we find @xmath18 and @xmath19 .
our results ( presented in full in * ? ? ?
* ) can be used to address the global properties of the disk and also the heating processes that have shaped its evolution . in order to fully understand the nature of disk heating and to disentangle the contributions from various mechanisms
, one needs to go beyond the work presented here .
the most crucial improvement will be the ability to make direct estimates for stellar ages , rather than relying on correlations with metallicity .
one aspect that will help us in this effort is by folding in measurements of alpha - element abundances that are now being determined routinely for vast numbers of stars ( e.g. * ? ? ?
* question : * ( nicolas martin ) can you develop your model to test for the presence of a dark disk ( e.g. * ? ? ?
* reply : * due to the limitations of the data that we have used there are a number of uncertainties in our modelling . as a consequence of these limitations ,
i think it would be difficult to include extra parameters .
mcs acknowledges financial support from the peking university one hundred talent fund ( 985 ) and nsfc grants 11043005 and 11010022 .
this work was also supported by the european science foundation ( esf ) for the activity entitled gaia research for european astronomy training. binney , j.j . , & tremaine , s. 2008 , galactic dynamics .
princeton univ . press ,
princeton , nj bramich , d.m . , vidrih , s. , wyrzykowski , l. , et al .
2008 , mnras , 386 , 887 cuddeford , p. , & binney , j. 1994 , mnras , 266 , 273 dehnen , w. , & binney , j.j .
1998a , mnras , 298 , 387 dehnen , w. , & binney , j.j .
1998b , mnras , 294 , 429 ivezi , z. , sesar , b. , juri , m. , et al .
2008 , apj , 684 , 287 jungman , g. , kamionkowski , m. , griest , k. , 1996 , phys .
, 267 , 195 kuijken , k. , & gilmore , g. 1991 , apj , 367 , l9 lee , y.s . , beers , t.c . , sivarani , t. , et al .
2008 , aj , 136 , 2022 lee , y.s . , beers , t.c . , a. , d. , et al .
2011 , apj , 738 , 187 nordstrm , b. , mayor , m. , andersen , j. , et al .
2004 , a&a , 418 , 989 read , j.i . , lake , g. , agertz1 , o. , debattista , v.p .
2008 , mnras , 389 , 1041 smith , m.c . ,
evans , n.w . , belokurov , v. , et al .
2009a , mnras , 399 , 1223 smith , m.c . , evans , n.w . , an , j.h .
2009b , apj , 698 , 1110 smith , m.c . , whiteoak , s.h . ,
evans , n.w .
2012 , apj , in press ( arxiv:1111.6920 ) vidrih , s. , et al .
2007 , mnras , 382 , 515 watkins , l.l .
, et al . 2009 , mnras , 398 , 1757 | surveys sdss enable detailed studies properties milky way possible to construct high - precision catalogues stars with full 3d kinematics analyse structure evolution galaxy summarise study of disk milky way highlighting two important results use data from 7th public data release sdss concentrating on 250 square degree stripe 82 region high precision photometry proper motions @xcite catalogue exploited in works covering white dwarfs halo kinematics rr lyrae in outer halo take all dwarf stars with good quality spectra estimating distances using modified @xcite photometric parallax relation modification concerns turn - off correction ( equation a6 not applicable to disk stars calculate three - velocities for sample obtain final errors of around 20 to 30 @xmath0 for each component velocity interested in kinematics disk function height from plane restrict to @xmath1 split data into three ranges in metallicity -0.8 each divide data into four ranges z maximum distance 2 kpc stars equally divided between four distance bins around @xmath4 to @xmath5 stars per bin from full sample of 7280 starsbinning necessary to avoid model sdss spectroscopic selection function divided stars fit distribution kinematics using maximum likelihood methods incorporating uncertainties on individual velocities @xmath6 and @xmath7 fit using gaussian distributions distribution of @xmath8 skewed non - gaussian adopt asymmetric functional form ( equation 15 ) * crucial to remove contamination from halo stars particularly metal - poor bins including non - rotating gaussian component into velocity fits level contamination amplitude gaussian assumed twice counter - rotating stars halo dispersions fixed according to values in @xcite comprehensive analysis results presented in @xcite concentrate on important properties rotation lag lag fig to couple kpc outside plane follows asymmetric drift relation hotter ( metal - poor ) stars exhibit greater lag than colder metal - rich clear correlation between lag and metallicity gradient lag @xmath9z@xmath9 varies 15 to 40 @xmath10 depending on metallicity fig lag with @xmath11 linear relation @xcite no longer true beyond 0.5 kpc tight correlation with @xmath11 independent of metallicity figure from @xcite.,title="fig:",width=173 figure from @xcite shown by authors vertical velocity dispersion profile gravitational potential disk results data in fig . included additional solar - neighbourhood metal - rich data point from geneva - copenhagen survey simplified two - parameter model for potential infinite razor - thin sheet in constant background best - fit model estimates mass density of components are @xmath12 and @xmath13 measurement with existing models especially model 1 of @xcite least halo - dominated model disk short scale - length dominating circular speed at solar neighborhood consult section 2.7 of @xcite for detailed comparison background mass density close to @xmath14 predicted using isothermal spherical halo models dark halo corresponds to local dark matter density of 0.57 @xmath15 larger than canonical value of 0.30 @xmath15 analysis local halo density may underestimated more robust than local mass density is surface mass density integrating mass distribution total surface mass density of @xmath16 agrees with classical value of @xmath17 from @xciteintegrate beyond 1.1 kpc find @xmath18 @xmath19 results presented * ? ? * address global properties disk heating processes shaped evolution understand disk heating disentangle contributions mechanisms go beyond work crucial improvement direct estimates for stellar ages correlations with metallicity folding in measurements alpha - element abundances determined for vast numbers stars question nicolas martin develop model test for presence dark disk due to limitations data uncertainties in modelling difficult to include extra parameters mcs acknowledges financial support from peking university one hundred talent fund nsfc grants 11043005 11010022 work supported by european science foundation ) gaia research for european astronomy training binney , j.j . tremaine , s. 2008 galactic dynamics princeton univ . press princeton bramich d.m vidrih s wyrzykowski l.. 2008 mnras , 386 , 887 cuddeford p binney , j. 1994 mnras , 266 , 273 dehnen binney 1998a mnras , 298 , 3871998b mnras 294 , 429 ivezi sesar juri m 2008 apj 684 , 287 jungman kamionkowski griest k 1996 phys 267 , 195 kuijken gilmore g 1991 apj 367 , l9 lee y beers t sivarani t 2008 aj 136 , 2022 lee beers d 2011 apj 738 , 187 nordstrm mayor andersen j. 2004 a&a 418 , 989 read j lake agertz1 debattista v 2008 mnras 389 , 1041 smith m evans n belokurov v. 2009a mnras 399 , 1223 smith evans n 2009b apj 698 , 1110 smith whiteoak s evans n.w 2012 apj press arxiv:1111.6920 ) vidrih s. 2007 mnras 382 , 515 watkins l.l . 2009 mnras 398 , 1757 | 0.325034 | 2,148 | 1,097 | , Saving $0.1 in GPT-4. |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.