Experimental and numerical study of seawater freezing coupled with ice growth dynamics and heat/mass transfer

Abstract

Seawater cold thermal energy storage offers high efficiency, environmental benefits, and abundant availability, making it promising for load shifting and energy system optimization. However, current seawater freezing models struggle to accurately capture the multiscale dynamics of ice formation and associated heat and mass transfer, limiting system performance improvement. To address this, a multiscale seawater freezing model is developed by coupling microscopic ice crystal growth mechanisms with macroscopic refrigeration system behavior. The model investigates seawater’s thermophysical transport and freezing kinetics. Ice crystal growth parameters are extracted through microscale simulations, and a kinetic model is constructed using the Arrhenius equation and hybrid particle swarm optimization (HPSO). This kinetic model is integrated into a system-level simulation to predict the freezing process in a cold storage tank under realistic conditions. The model achieves high accuracy, with maximum errors of 0.9 °C for water temperature, 4 mm for ice thickness, 1.5 °C for refrigerant outlet temperature, and 0.02 MPa for outlet pressure. Compared with the traditional enthalpy-porosity model (EPM), the proposed model reduces the water temperature error by 40 % and improves ice thickness accuracy, reducing the maximum error from 15 mm to 4 mm. Furthermore, salinity significantly influences salt rejection, temperature evolution, heat transfer coefficient, cooling capacity, and thermal storage performance. A 3 % salinity seawater achieves the best performance due to a favorable balance between temperature difference and heat transfer efficiency. This study provides theoretical insights and technical support for enhancing seawater freezing models and optimizing system-level cold thermal storage design.