Novel multiscale model for grain-packed inorganic salt hydrate-based open thermochemical storage for low-temperature space heating applications

This study presents the development, validation, and application of a multiscale numerical model for an open thermochemical energy storage (TCES) reactor using SrBr2·6H2O as the representative reactive medium. The objective is to accurately capture and predict the coupled phenomena of heat and mass transfer occurring across two physical scales—the reactor bed and individual salt grains—during hydration and dehydration cycles. Experiments are conducted on a rectangular stainless-steel reactor filled with SrBr2·6H2O under controlled inlet air conditions. SEM images are used to estimate grain-level porosity and the particle size distribution. Grain diameters range from 0.2 mm to 4.0 mm, and porosities vary from 0.13 to 0.22 across cycles. The multiscale model is developed in MATLAB, where the bed-level routine solves conservation equations using Darcy's law and volume-averaged properties. Conversely, the grain-level routine solves unsteady diffusion-reaction equations inside the spherical salt grains. Reaction advancement modifies the grain size and the porosity dynamically, and volume-averaged source terms aid in coupling the two scales. A representative grain approach is adopted for better computational efficiency without compromising the accuracy of the results. The developed model is validated against in-house experiments and prominent literature with outlet temperature predictions, achieving R2 values of 0.93 and 0.98 for hydration and dehydration, respectively. A parametric study reveals that smaller grain sizes (0.2 mm) reduce reaction time by 47 % but increase pressure drop up to 2843 Pa. Higher energy densities (∼1.6 GJ/m3) enhance storage while increasing flow resistance. The novel framework helps evaluate TCES reactor performance with evolving material properties across different scales, and enables performance optimisation by balancing energy density, thermal output, and auxiliary power demand.