Optimization analysis of a coupled carbonation-moisture transport model using COMSOL

Concrete porosity, surface microcracks, and material properties significantly influence carbon dioxide diffusion, leading to inconsistent carbonation depths even under identical mix proportions or environmental conditions. Based on accelerated carbonation tests of pile cores from a silo project in Lianyungang, this study optimizes the liquid CO₂ diffusion coefficient formula D C O 2 (m 2 /s) by introducing the connected porosity φ (c) (instead of total porosity) and microcrack density λ crack , grounded in cement hydration-carbonation theory, concrete pore fractality, and the cubic law. Through secondary development of COMSOL software, this study established a dynamic carbonation-moisture coupled transport model. The final conclusions are as follows: The content of dissolved CO₂ in concrete increases over time, while its transport rate declines; the precipitation of calcium carbonate (CaCO 3 , CC) generation is higher in shallow concrete regions and decreases inward; by accounting for calcium hydroxide (Ca(OH) 2 , CH) and calcium silicate hydrate (C-S-H, CSH), the model can better simulate actual CC generation; longer carbonation time reduces CH content but increases CC content. When pH= 9.3 is used as the carbonation depth criterion, simulations align closely with field data (<10 % error) with enhanced accuracy. Validation against Papadakis’ rapid carbonation data (1989) and comparison with M. Valcuende’s natural carbonation samples (2010) confirm that the optimized model achieves superior agreement with experimental results. This study may provide a foundation for future concrete carbonation prediction models in civil engineering applications.