Molecular Dynamics Simulations of Structure, Ion Transport and Interphase Stability in Alkali- Metal-ion Battery Electrolytes

Abstract

The development of next-generation rechargeable batteries is crucial to address the limitations of current commercial options, such as low energy density, safety concerns, and inadequate performance for energy storage demands. A key challenge in battery design lies in making the best choice of electrolytes that are cost-effective, have low flammability and reactivity, and offer excellent rate capability and power. This thesis describes structural and dynamics in Li+ and Na+ ion battery electrolytes using classical Molecular Dynamics (MD) simulations. For example, the effect of concentration and temperature on ion-ion interactions, ion solvent interactions, self diffusion coefficients and ionic conductivity in NaPF6/diglyme is discussed. The simulations show that interactions characterized using the radial distribution functions (RDF) remains invariant with concentration and temperature. In contrast, the diffusion coefficients of the ions decreases with concentration and increases with temperature. The distribution of solvated ion pairs, contact ion pairs and aggregate ion pairs, and their life times are also presented. In a Li+-ion based battery electrolyte, the structure, Li-ion transport in a Polyoligomeric silsesquioxanes (POSS) solvated in tetraglyme (G4) solvent is explored. The simulations show that Li+ ions interact preferably with the sulfonyl oxygen and nitrogen atoms on the POSS compared to G4. The presence of larger number of solvent-separated ion pairs with increase in G4 leads to higher cationic mobility. The cation transference number calculated from simulations is in agreement with experimental data. The development of a reactive force field (FF) to investigate the electrolyte decomposition and formation of solid electrolyte interphase is also presented in this thesis. The reactive FF for Na/O/C/H was developed and trained specifically to model reductive decomposition reactions between Na metal surface and water/organic solvents (water, ethylene carbonate and dimethyl ether). The FF was trained using the interaction energies and structures obtained from the density functional theory calculations. The training data from FF closely resembles the binding and formation energies for various electrolyte decomposition products such as NaOH, Na2CO3, and Na alkoxides. The ReaxFF-MD simulations using the FF is employed to explore the dissociation of water/ organic molecules on the Na(001) slab. The collective variable hyperdynamics (an accelerated MD technique) simulations is used to observe the dissociation of water and ethylene carbonate on Na surface. The thesis concludes by exploring the future directions in understanding the decomposition and formation of SEI in ester and ether-based electrolytes.