Molecular dynamics simulation of water molecules adsorption by different cations based montmorillonite

Abstract

Using molecular mechanics and molecular dynamics methods, Li-, Na-, K-, Ca-cations based montmorillonite models are built in Materials Studio software. After absorbing 16, 32, 48, 64, 80, 96 water molecules, geometry optimization is performed to get the optimum configuration, the influences of geometry optimization on the total energy of the system is analyzed. Further, molecular dynamic simulations are carried out to investigate the influences of different cations on the swelling behavior of montmorillonite when absorbing water. After geometry optimization, cations are moving close to the position of isomorphism substitution, the initial regular structure changed, the total energy of the system also changed greatly, the total energy of the system is reduced, both bond energy and non-bond energy are reduced, the reduction of the bond energy is about 56% of the reduced total energy. Torsional energy of dihedral angle is increased a bit, indicating the increased structural distortion. The bond stretching energy is reduced about 90%. Molecular dynamic simulations indicate the montmorillonite volume and layer spacing are increased by multi-steps, but the density is reduced by multi-steps. Water molecule concentration profiles in lithium-, sodium-, potassium- and calcium-montmorillonite adsorption models show that the montmorillonite can form one, two and three layers of water molecules. The radial distribution function of Li-, Na-, K-, Ca-cations based montmorillonite is obtained for absorbing 16 to 96 water molecules. The results indicate that Li+ is easier to hydrate, follows by Na+, while K+ and Ca2+ are not easy to hydrate. Different types of cations between layers have different effects on the adsorption ability of montmorillonite. After obtaining the mean squared distance of different cations and water molecules, the self-diffusion coefficients are calculated by fitting mean squared distance curves with the least square method. Compared with the literature data by Wang, Chang, and Na, it indicates that different cations, simulation methods, potential function, and simulation temperature all affect the diffusion of the middle layer cations and water molecules. MD simulations revealed the mechanism of chemical stabilization of expansive soils from molecular level.