How it works

Finding a better battery electrolyte additive is a search problem over a space too large to enumerate and too expensive to sample: every candidate you want to test costs weeks of bench time. So the useful thing to build is not a better predictor, it is a loop that proposes, scores, and justifies candidates before anyone picks up a pipette.

BatteryGen is that loop. A conditional VAE trained on 7.1 million molecules generates candidates in SELFIES, which was chosen over SMILES specifically because essentially every decoded SELFIES string is a chemically valid molecule by construction. A paired ExtraTrees model predicts Coulombic efficiency. A multi-LLM stage then reasons over a retrieval-augmented literature base to argue for or against each finalist and score how novel it actually is against published work.

The whole thing trained in 16 hours on a single RTX 3060, which took several tuning rounds to get right. It ships as a pip-installable package with CI that fails the build on regression, and one config block retargets it from zinc to lithium, sodium, or potassium without touching code.

• PyTorch
• Conditional VAE
• SELFIES
• RDKit
• xTB
• XGBoost
• Optuna
• RAG

Publications

• published2026

  Kapadia, N., Ke, J., & Su, L. A transferable data-driven framework for electrolyte discovery.

  Discover Energy 6, 10

  Peer-reviewed Perspective, open access. Argues for data-centric standards in electrolyte machine learning: minimum information reporting standards, FAIR community databases, literature-scale extraction, and shared benchmark tasks.

  Read it
• in preparation2026

  Kapadia, N. et al. BatteryGen manuscript.

  Target venue: iScience

  Experimental validation ongoing.

• under review2026

  Kapadia, N. & Su, L. Designing Next-Generation Batteries with Generative and Predictive Artificial Intelligence.

  UT Dallas Exley Research Journal

• poster2025

  Kapadia, N., Ke, J., & Su, L. A Generative-Predictive AI Pipeline for Accelerating Discovery of High-Efficiency Additives for Aqueous Zinc-Ion Batteries.

  Summer Platform for Undergraduate Research (SPUR)