Computational modelling in fuel cell research: From density functional theory to computational fluid dynamics

Fuel cells are pivotal to the global transition toward a sustainable hydrogen economy, offering high efficiency and carbon-free energy conversion. However, large-scale deployment remains constrained by the high cost and limited availability of platinum-group metal catalysts, as well as performance losses associated with sluggish oxygen reduction and hydrogen oxidation kinetics, and performance losses associated with transport limitations and degradation under realistic operating conditions. Experimental characterisation alone often struggles to resolve internal gradients, transient behaviour, and multiscale interactions, motivating the growing role of computational modelling as a complementary design and diagnostic tool. This review critically examines the integration of density functional theory (DFT) and computational fluid dynamics (CFD) for multiscale modelling of fuel cells, highlighting how atomistic insights can be systematically translated into continuum-scale performance predictions. DFT provides fundamental understanding of catalytic activity, charge transport, defect chemistry, and degradation mechanisms, while CFD captures coupled transport, electrochemical, thermal, and water management phenomena at the device scale. We synthesise recent strategies for mapping DFT-derived descriptors—such as reaction energetics, transport coefficients, and degradation kinetics—into CFD frameworks, with particular emphasis on uncertainty propagation, validation practices, and computational efficiency. The review covers applications across proton exchange membrane fuel cells, solid oxide fuel cells, and protonic ceramic fuel cells, illustrating representative DFT-informed modelling approaches rather than exhaustive quantitative benchmarking. Special attention is given to validation against experimental polarisation, impedance, and degradation trends, as well as to emerging surrogate and reduced-order modelling techniques that mitigate computational cost. Finally, future research directions are discussed, focusing on DFT-guided catalyst design, defect engineering, and robust multiscale coupling strategies to support predictive and scalable fuel-cell modelling.