Reductive regeneration of iron-based fuels in a fixed-bed reactor for renewable energy storage: multiphysics simulation and comparative case evaluation

Iron-based metal fuels offer a recyclable route for storing renewable electricity as a dense and transportable chemical energy carrier. However, the reactor-scale coupling of gas transport, heat transfer, product-gas removal, and local solid conversion during Fe 2 O 3 regeneration remains difficult to observe directly. Here, an experimentally validated COMSOL multiphysics model was developed for Fe-based fuel production from Fe 2 O 3 in a fixed-bed tubular reactor using an H 2 /CO mixed reducing gas. The baseline case employed 0.5 g Fe 2 O 3 at 800 °C with 80 mL/min Ar, 10 mL/min H 2 , and 10 mL/min CO. The model was validated against experimentally measured outlet CO concentrations obtained by GC/MS, yielding an RMSE of 0.48%, a MAPE of 5.99%, and relative errors below 10% at selected validation points. In addition, the simulated products showed good agreement with the experimental XRD patterns, confirming the model's capability to reproduce the major phase composition after regeneration. Based on the validated model, the effects of reactor geometry, total inlet flow rate, and H 2 /CO ratio were systematically evaluated, and sensitivity analysis was performed to identify the dominant operating parameters affecting regeneration performance. Six indicators, including Fe 2 O 3 conversion, Fe yield, gas utilization, outlet gas composition, residual Fe 2 O 3 fraction, and average product gas fraction, were integrated using the EWM-TOPSIS method. The results showed that the residual Fe 2 O 3 fraction exerted the greatest influence on the comprehensive evaluation. Among all cases, G3, with a 20% increase in equivalent reactor/sample diameter, achieved the highest EWM-TOPSIS comprehensive score of 0.9364 owing to its highest Fe 2 O 3 conversion and lowest residual Fe 2 O 3 fraction. This work provides guidance for the design and selection of fixed-bed Fe-based fuel regeneration systems.