README.md

---
title: LiB Simulation AI Engine
emoji: 🔋
colorFrom: blue
colorTo: purple
sdk: docker
pinned: false
short_description: ReaxFF reparameterization & LiF SEI analytics dashboard
---

# 🔋 Lithium-Ion Battery Simulation AI Engine

A comprehensive analytics dashboard for ReaxFF force-field reparameterization,
LiF solid-electrolyte interphase (SEI) modelling, and Li⁺ transport prediction
in lithium-ion batteries.

## Features

- **Force Field Performance** — R² / RMSE comparison across Yun et al., Wang et al., and new reparameterization
- **Li Diffusion & Arrhenius** — Interactive diffusion coefficient plots, Arrhenius analysis at 300/400/500 K
- **Crystal Stability** — Energy-strain curves and Murnaghan equation of state for LiF
- **RDF Evolution** — Radial distribution function time series showing crystalline-to-amorphous transition
- **SEI Component Analysis** — Multi-criteria ranking of LiF, Li₂CO₃, Li₂O, and other SEI components
- **ML Force Field Comparison** — Benchmarking of M3GNet, CHGNet, NequIP, DeepMD, SchNet, ALIGNN-FF
- **Simulation Campaign** — Full database and CI-NEB energy barrier summary

## Data Sources

All quantitative data comes directly from:

> De Angelis P., Cappabianca R., Fasano M., Asinari P., Chiavazzo E.
> *Enhancing ReaxFF for molecular dynamics simulations of lithium-ion batteries:
> an interactive reparameterization protocol.*
> **Scientific Reports** 14:978 (2024). DOI: 10.1038/s41598-023-50978-5

Key values:
- Diffusion coefficient for LiF at 300 K (new FF): **3.44 × 10⁻⁸ cm²/s**
- Improvement over prior FFs: **~1000×**
- CMA-ES iterations to convergence: **13,000**
- DFT simulations in training database: **300+**

## Running Locally

```bash
pip install -r requirements.txt
streamlit run app.py
```