# Resolve Species in MEA/MDEA Capture Solvents Using Multiple Analytical Signals and Evaluate Absorption–Regeneration Operating Campaigns

A pilot-scale unit uses a mixed MEA/MDEA amine solvent to absorb CO2. The laboratory independently prepared three replicates each of the lean and rich solvents from two operating campaigns and performed quantitative 13C NMR, dual-endpoint acid titration, total inorganic carbon analysis, and ion chromatography. Calibrated analytical readings, measurement uncertainties, the response relationships between each signal and the solution species, sample pairing information, and operating acceptance criteria are provided in /app/input.

Integrate these analytical results to resolve free/protonated MEA, MEA carbamate, free/protonated MDEA, dissolved CO2, bicarbonate, carbonate, and HSS equivalents in each replicate. Summarize species concentrations and repeatability for each sample; check total amine, total inorganic carbon, and charge closure; and calculate CO2 loading, carbon-containing species fractions, and free-amine fractions. Then compare the paired lean/rich samples from the two operating campaigns in terms of working capacity, HSS level, and the species-based heat-release proxy, and provide operating conclusions and the reasons for them.

Generate analysis.py, replicate_speciation.csv, measurement_fit.csv, species_summary.csv, sample_balance.csv, campaign_performance.csv, co2_speciation.png, campaign_performance.png, analytical_fit_residuals.png, and amine_speciation_report.md under /app/output. All summary results must retain units and the means and standard deviations of the three independently prepared replicates. The figures, tables, and report must be consistent with the CSV files. Do not read data from outside /app/input or hard-code results in the script.

## Deliverable Files

This section defines only the submission format and does not prescribe the analytical method or expected results. All files listed below are required and must be placed directly under `/app/output/`.

### `/app/output/analysis.py`

- Format/encoding: text, UTF-8.

### `/app/output/replicate_speciation.csv`

- Format/encoding: comma-separated CSV, UTF-8; use a period as the decimal separator and include no index column.
- Row semantics: each row corresponds to one species in one independently prepared replicate of one sample, for a total of 108 rows; the stable key is `sample_id` + `replicate` + `species`.
- Required fields:

| Standard field | Type | Unit/meaning |
|---|---|---|
| `sample_id` | string | Sample identifier |
| `campaign_id` | string | Operating campaign identifier |
| `stream_role` | string | Sample stream-role identifier |
| `replicate` | integer | Independently prepared replicate identifier |
| `species` | string | Species identifier |
| `concentration_mmol_g_solution` | number | mmol/g solution |

### `/app/output/measurement_fit.csv`

- Format/encoding: comma-separated CSV, UTF-8; use a period as the decimal separator and include no index column.
- Row semantics: each row corresponds to one analytical signal in one independently prepared replicate of one sample, for a total of 144 rows; the stable key is `sample_id` + `replicate` + `signal_id`.
- Required fields:

| Standard field | Type | Unit/meaning |
|---|---|---|
| `sample_id`, `campaign_id`, `stream_role` | string | Sample, operating campaign, and stream-role identifiers |
| `replicate` | integer | Independently prepared replicate identifier |
| `method`, `signal_id`, `unit` | string | Analytical category, signal identifier, and unit of the values in that row |
| `observed_value`, `predicted_value`, `measurement_sd`, `residual` | number | As specified by `unit` in that row |
| `standardized_residual` | number | Dimensionless |

### `/app/output/species_summary.csv`

- Format/encoding: comma-separated CSV, UTF-8; use a period as the decimal separator and include no index column.
- Row semantics: each row corresponds to the summary of one species in one sample, for a total of 36 rows; the stable key is `sample_id` + `species`.
- Required fields:

| Standard field | Type | Unit/meaning |
|---|---|---|
| `sample_id`, `campaign_id`, `stream_role`, `species` | string | Sample, operating campaign, stream-role, and species identifiers |
| `n_replicates` | integer | Number of independently prepared replicates |
| `concentration_mean_mmol_g_solution` | number | mmol/g solution |
| `concentration_sd_mmol_g_solution` | number | mmol/g solution |

### `/app/output/sample_balance.csv`

- Format/encoding: comma-separated CSV, UTF-8; use a period as the decimal separator and include no index column.
- Row semantics: each row corresponds to the summary and closure results for one sample, for a total of 4 rows; the stable key is `sample_id`.
- Required fields:

| Standard field | Type | Unit/meaning |
|---|---|---|
| `sample_id`, `campaign_id`, `stream_role` | string | Sample, operating campaign, and stream-role identifiers |
| `n_replicates` | integer | Number of independently prepared replicates |
| `total_MEA_mmol_g_mean`, `total_MEA_mmol_g_sd`, `total_MDEA_mmol_g_mean`, `total_MDEA_mmol_g_sd`, `total_amine_mmol_g_mean`, `total_amine_mmol_g_sd`, `total_CO2_mmol_g_mean`, `total_CO2_mmol_g_sd` | number | mmol/g solution |
| `CO2_loading_mol_per_mol_amine_mean`, `CO2_loading_mol_per_mol_amine_sd` | number | mol CO2/mol amine |
| `carbamate_fraction_of_CO2_mean`, `carbamate_fraction_of_CO2_sd`, `bicarbonate_fraction_of_CO2_mean`, `bicarbonate_fraction_of_CO2_sd`, `carbonate_fraction_of_CO2_mean`, `carbonate_fraction_of_CO2_sd`, `free_amine_fraction_mean`, `free_amine_fraction_sd` | number | Dimensionless |
| `HSS_equivalent_per_mol_amine_mean`, `HSS_equivalent_per_mol_amine_sd` | number | mol equivalent/mol amine |
| `charge_residual_mmol_g_mean`, `charge_residual_mmol_g_sd` | number | mmol/g solution |
| `maximum_abs_standardized_fit_residual_mean`, `maximum_abs_standardized_fit_residual_sd` | number | Dimensionless |

### `/app/output/campaign_performance.csv`

- Format/encoding: comma-separated CSV, UTF-8; use a period as the decimal separator and include no index column.
- Row semantics: each row corresponds to one operating campaign, for a total of 2 rows; the stable key is `campaign_id`.
- Required fields:

| Standard field | Type | Unit/meaning |
|---|---|---|
| `campaign_id`, `lean_sample_id`, `rich_sample_id` | string | Operating campaign and paired-sample identifiers |
| `lean_loading_mol_per_mol`, `rich_loading_mol_per_mol`, `working_capacity_mol_per_mol` | number | mol CO2/mol amine |
| `maximum_HSS_per_mol_amine` | number | mol equivalent/mol amine |
| `maximum_abs_standardized_fit_residual` | number | Dimensionless |
| `maximum_abs_charge_residual_mmol_g` | number | mmol/g solution |
| `speciation_heat_proxy_kJ_mol_CO2` | number | kJ/mol CO2 |
| `campaign_decision`, `failure_reasons` | string | Operating campaign decision and reason codes |

The required fields in the five CSV files above may appear in any column order. Additional columns are permitted but must not replace the required fields.

### `/app/output/co2_speciation.png`, `/app/output/campaign_performance.png`, and `/app/output/analytical_fit_residuals.png`

- Format: PNG image.
- Image categories: carbon-containing species results, operating campaign results, and analytical fit residuals, respectively.

### `/app/output/amine_speciation_report.md`

- Format/encoding: Markdown, UTF-8.
- Content categories: integration of analytical signals, species balances, repeatability, closure checks, and the differences between and reasons for the decisions on the two operating campaigns.