Intercell nonuniformity and dominant mechanisms analysis in industrial-scale water electrolyzer stacks: a unified comparative multiphysics modeling across AWE, AEM and PEM

Low temperature water electrolysis is a key route for converting renewable electricity into green hydrogen, but industrial-scale stacks are prone to intercell nonuniformity during scale up. The origins and dominant mechanisms of this nonuniformity remain unclear across different electrolyzer technologies. To address this issue, this study develops a unified stack-scale multiphysics lumped-parameter model for industrial-scale alkaline water electrolysis (AWE), anion exchange membrane water electrolysis (AEM), and proton exchange membrane water electrolysis (PEM) stacks and uses it to compare the dominant mechanisms of intercell nonuniformity across the three technologies. The model is formulated on a common stack-level basis and couples flow distribution, shunt current, bubble coverage, temperature distribution, and electrochemical reaction under a common external flow network. The common stack-level framework is first validated using an AWE experimental platform. The model results show that AWE exhibits the strongest intercell nonuniformity, mainly due to shunt current induced redistribution of effective reaction current. AEM shows intermediate behavior governed by weak shunt current and gas-phase effects, whereas PEM is influenced mainly by bubble coverage and temperature development. Load dependent and layered physical field analyses further reveal that AWE is dominated by shunt current, causing a 12.38% rated current loss and a 7.21% energy-efficiency drop, whereas the responses of AEM and PEM are governed mainly by voltage efficiency degradation, driven by gas-phase and temperature effects. The proposed framework provides a unified basis for mechanism identification and differentiated optimization of industrial-scale low temperature electrolyzer stacks.