Computer Simulations of CO2 Absorption in Amino Acid Ionics & Ion Conduction in Alkali Metal Battery Electrolytes

Abstract

This thesis presents modeling and simulations on absorbents for carbon dioxide capture and electrolytes for Li+/Na+ ion batteries.

In the first part of the thesis, Lysine amino acid based Ionic Liquids (IL) and Salts (AAS) are explored as potential replacements for conventional absorbents for carbon capture like amines, amine-alcohol blends, etc. Classical molecular dynamics (MD) simulations are performed to examine the molecular mechanism of CO2 absorption in [tetrabutylphosphonium+][Lys-] IL. The simulations suggest that an interface of CO2 molecules form at the IL surface within a short span of tens of picoseconds, and attains saturation around ten nanoseconds, where CO2 molecules remain absorbed in the bulk IL layers. The interaction and absorption of CO2 molecules leads a slightly higher mobility of anions than cations due to the preferential interaction of CO2 molecules. Density Functional Theory (DFT) calculations are also employed to understand the mechanism for Lys--CO2 reaction and participation of a single molecule of water in this reaction. The calculations show the existence of various non-bonded complexes and chemical reactions responsible for CO2 absorption and desorption. The reaction mechanisms in each complex are characterized by energy parameters such as binding energy, activation energy, and reaction energy. The competitive reaction pathways which show the dominance of carbamate over bicarbonate products during CO2 absorption are also discussed. The simulations and calculations serve as a predictive tool to develop efficient AAILs for CO2 absorption.

In the second part of the thesis, the thermal stability and mechanism of ion conduction in a new generation of soft-solid cocrystalline electrolytes for Lithium and Sodium Ion Batteries are investigated. Four cocrystalline solid electrolytes composed of alkali metal salts and organic solvents (adiponitrile (ADN) and N,N-dimethylformamide (DMF) solvents) in stoichiometric ratios, DMF∙LiCl, (DMF)3NaClO4, (ADN)3NaClO4 and (ADN)2LiPF6 are examined for their structural complexity and mechanism of ion conduction. MD simulations on model structures of lithium ion electrolyte, DMF∙LiCl, provide an atomic level understanding of various experimental properties: crystal packing arrangement, mechanism of decomposition, existence of a nanolayer of DMF molecules at the surface, and higher mobility of ions at surface compared to bulk. The DFT calculations show that small aggregates on the surface of cocrystal easily decompose compared to the large aggregates in the bulk. In the (DMF)3NaClO4 electrolyte, MD simulations reveal the mechanism of temperature dependent stoichiometric conversion and crystal melting from the analysis of structural properties. The size and number of ion-pair and ion-solvent clusters as a function of temperature provide proofs of stoichiometric conversion and melting of the electrolyte. The calculated diffusion coefficients at different temperatures show a competitive nature of ionic diffusion and the activation energy barrier calculated for Na+ migration agrees with experiments. The minimum energy path calculated using periodic DFT calculations shows a SN2 reaction type, with a planar transition state observed during the Na+ ion migration, which occurs in one-dimension. The behavior of surface and bulk is modeled in (ADN)3NaClO4 and (ADN)2LiPF6 electrolytes, where the thermal stability is found to be in good agreement with TGA and DSC experiments. In contrast to (DMF)3NaClO4, the conduction of Na+ ions in (ADN)3NaClO4 occurs in all the three dimensions via a solvent-anion assisted transition state. The migration of Li+ ion in (ADN)2LiPF6 electrolyte was observed to occur via a solvent-only tetrahedral intermediate. The calculated jump probabilities from van-Hove autocorrelation function in these electrolytes provide insights to the contribution of interstitial dislocation in ion conduction.