LLM-Assisted Generation of Transparent, Open-Source Multiphysics Models of Electrochemical Devices
Abstract
Multiphysics continuum models are powerful tools for studying electrochemical devices, enabling in silico reactor design and resolution of local pH, potential, and concentration fields that govern device performance but are difficult to measure experimentally. However, constructing such models requires substantial numerical expertise or reliance on proprietary software. Here, we show that frontier large language model agents can remove this implementation burden while keeping the underlying physics under researcher control. Using one-dimensional electrochemical CO2 reduction to CO in a porous gas diffusion electrode as a test case, we develop a machine-readable, human-specified modeling harness containing governing equations, parameters, numerical methods, logical build stages, and human-verifiable checkpoints. From this specification, the agent reproducibly constructs complete multiphysics models in open-source Julia. Independently built models, including fully autonomous agent-built models, agree with an equivalent COMSOL implementation to within 0.7% of the peak CO partial current density, and with one another to within 0.04%. Systematically planted errors demonstrate the importance of explicit specifications for reproducibility and reveal the agent's capabilities and limitations in debugging model physics. This framework establishes a more transparent approach to multiphysics modeling in which physical descriptions and governing equations, rather than specialized code, become the primary inputs for computational model development.
Related
- AI-assisted analytical theory in soft matter and multiphysics: new results and lessons from diffusiophoresis and bipolar membranes
- Recent advances in high surface area porous materials for electrochemical energy storage devices
- Lignin as a Sustainable Building Block for Electrochemical Energy Devices: A Critical Review
- Deciphering Electrochemical Generation, Transport, and Reactivity of Hydrogen Species in the Electrochemical Palladium Membrane Reactor.
- High Fidelity Stochastic Electrode Microstructure Generation for Low Temperature Electrochemical Systems
BibTeX
@misc{castro2026llmassisted,
title = {{LLM-Assisted Generation of Transparent, Open-Source Multiphysics Models of Electrochemical Devices}},
author = {Sebastian Castro and Maya F. Schuchert and Spencer A. McCluskey and Eric W. Lees and Justin C. Bui},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.10320},
eprint = {2610.10320},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.10320},
}
RIS
TY - UNPB
TI - LLM-Assisted Generation of Transparent, Open-Source Multiphysics Models of Electrochemical Devices
AU - Sebastian Castro
AU - Maya F. Schuchert
AU - Spencer A. McCluskey
AU - Eric W. Lees
AU - Justin C. Bui
PY - 2026
DA - 2026/10/07
PB - arXiv
DO - 10.48550/arxiv.2610.10320
UR - https://arxiv.org/abs/2610.10320
AB - Multiphysics continuum models are powerful tools for studying electrochemical devices, enabling in silico reactor design and resolution of local pH, potential, and concentration fields that govern device performance but are difficult to measure experimentally. However, constructing such models requires substantial numerical expertise or reliance on proprietary software. Here, we show that frontier large language model agents can remove this implementation burden while keeping the underlying physics under researcher control. Using one-dimensional electrochemical CO2 reduction to CO in a porous gas diffusion electrode as a test case, we develop a machine-readable, human-specified modeling harness containing governing equations, parameters, numerical methods, logical build stages, and human-verifiable checkpoints. From this specification, the agent reproducibly constructs complete multiphysics models in open-source Julia. Independently built models, including fully autonomous agent-built models, agree with an equivalent COMSOL implementation to within 0.7% of the peak CO partial current density, and with one another to within 0.04%. Systematically planted errors demonstrate the importance of explicit specifications for reproducibility and reveal the agent's capabilities and limitations in debugging model physics. This framework establishes a more transparent approach to multiphysics modeling in which physical descriptions and governing equations, rather than specialized code, become the primary inputs for computational model development.
ER -