Modeling Multi-Physics in PEM Water Electrolyzers using LBM
Date Issued
August 2026
Author(s)
The University of Tennessee
Advisor(s)
Matthew M Mench
Additional Advisor(s)
Seungha Shin
Anna Herring
Douglas S. Aaron
Abstract
The increasing demand for renewable energy has created a need for efficient and reliable energy storage and conversion technologies. Polymer Electrolyte Water Electrolyzers (PEWEs) offer a promising solution, but require performance enhancements for economic
viability. In particular, a better understanding of multiphase transport within porous
transport layers (PTLs) is essential for improving cell efficiency. This work develops a
multi-scale modeling framework that links pore-scale transport phenomena to full-cell PEWE
performance predictions. Conventional PEWE continuum-scale computational fluid dynamics models simplify
porous transport layers using homogenized properties such as porosity, permeability, and
tortuosity, limiting their ability to resolve local transport phenomena. In contrast, pore
scale models capture detailed multiphase physics but are computationally restricted to small
domains. This work aims to bridge these scales through a physically grounded modeling
framework. A segmented-cell diagnostic system was used to obtain in-situ measurements of local current density and high-frequency resistance, providing insight into transport mechanisms and the effects of diffusion media and flow-field designs. These experimental results guided both component optimization and model development.
The primary contribution of this work is the development of a lattice Boltzmann method (LBM) framework for simulating two-phase flow and fluid-solid interactions at large density ratios. The coupled Cahn-Hilliard and Navier-Stokes formulation provides improved numerical stability while enabling physically meaningful material and boundary parameters, allowing prediction of PTL transport properties including permeability, tortuosity, and
capillary transport characteristics. A generalized moment-based boundary condition framework was also developed for coupled multiphysics LBM simulations. Applicable within the single-relaxation-time (SRT) framework for both Navier-Stokes and advection-diffusion equations, the method reconstructs unknown boundary distribution functions directly from macroscopic moments. The incorporation of ghost-node extensions further improves accuracy for complex and curved geometries. The overall objective of this work is to capture the fundamental multiphase transport processes occurring within PEWE porous transport layers and to integrate these pore scale insights into a continuum-scale performance model. By combining advances in pore-scale LBM modeling, generalized boundary-condition development, and experimental informed electrochemical modeling, this work establishes a physically grounded multi-scale framework for predicting PEWE performance and guiding the design of next-generation
porous transport layers.
Subjects
Degree
Doctor of Philosophy
Major
Mechanical Engineering
File(s)![Thumbnail Image]()
Name
ROENNING-PRIMARY-2026.pdf
Size
45.63 MB
Format
Adobe PDF
Checksum (MD5)
9d234baf91e9a61d3dc1466e8e523b3f