Study of concentration dependent microstructures and dynamic behaviors of aqueous CaCl2 solutions via deep potential with electrostatic interactions

Salt solutions have long been a subject of scientific interest owing to their significant role in various applications. Despite advances in experimental techniques and simulation methods, contradictions persist in describing the solvation structure and dynamic behavior of ions and water molecules in salt solutions, especially for non-monovalent salt solutions. In this study, we focus on the aqueous CaCl2 system and developed an advanced machine learning force field that incorporates electrostatic interactions, enabling high-accuracy molecular dynamics simulations at the SCAN functional level on nanosecond timescales. Benchmark tests confirmed that our model closely matches both density functional theory results and experimental data. Through comprehensive analysis, the concentration dependence of the microscopic structure and dynamics of CaCl2 solutions was revealed, highlighting significant effects arising from ion species. In addition, we performed the first high-precision infrared spectroscopy simulation of CaCl2 solutions, validating changes in hydrogen bond networks, ion solvation shells, and water molecule dynamics. Our results describe the dependence of solvation structure and diffusion behavior in CaCl2 solutions, offering the theoretical foundation for future research in this field.