Ab Initio Study On The Cathode/Electrolyte Interface In Li-O₂ Battery

Li-O₂ battery is one of the most potential energy storage and conversion systems,whose theoretical capacity is up to 3500 Wh/kg.However,the discharge products,such as the Li₂O₂/LiOH are all insulate,leading to the super high overpotential(1?2V)during the charging process.Redox mediators(RMs),due to their flexible mass transfer,are elected as the electron transfer agent to resolve the cathode irreversibility.Though the over potential has been reduced effectively by the RM,the sluggish reaction kinetics between the oxidized RM and the discharge products remains a problem which hinders the rate capability.Also,the halide's effect on the solvent stability in terms of the acid dissociation is still in dispute.Based on the ab initio calculations,machine learning and experiment,this thesis explores the structural effect of the discharge products on the decomposition kinetics with halide RM present,the solvent effect on the decomposition of the discharge products,and the solvent acidity in the mixed electrolyte with RM(Li I?Li Br)present.These explorations provide theoretical basis for improving the kinetics of the oxygen evolution reaction(OER)process and guide the screening of highly stable solvents in the practical batteries.The main studies and results are listed as follows,(i)The ab initio calculations are applied to investigate the reaction kinetics between LiOH and I₂.The latter is considered as the oxidized state of the RM Li I.First,we explored three decomposition path and determine the optimized path.Also,the Li⁺desorption is identified as the rate determining step(rds)of the reaction.Based on the optimized path,we further construct the disordered/amorphous structure and define the disorder parameter(?)based on the bond length deviation against the crystalline structure.The disordering(?)of the(001),(100)and amorphous structure ascends.While LiOH turns from the crystalline to disordered/amorphous structure,the rds energy barrier will be reduced by?500meV,demonstrating the faster reaction kinetics for the disordered/amorphous structure.The results suggest that the OER kinetics can be greatly improved by regulating the structure of the discharge products,which can be achieved via the electrolyte design or current tuning.(ii)The ab initio calculations and machine learning method are applied to explore the solvent effect on the OER kinetics.Based on the rds(the desorption of Li⁺)in Li-O₂ battery determined above,we investigate the participation of 33 solvents in the decomposition process of the LiOH and compared their solvation effect on the reaction kinetics.To evaluate the solvation capability of the solvent,the binding energy(E_b)between the solvent and the LiOH molecule is determined as the descriptor.The functional group of the solvent is detected as the key to regulating the solvation effect and phosphate-based solvent is predicted to accelerate the decomposition kinetics most with the strongest binding energy.(iii)To predict the pK_a in the practical electrolyte with additives,we combine the ab initio calculations with the experiment and start from the solvation structure to predict the pK_a of 43 molecules in Li I/DMSO and Li Br/DMSO electrolyte.The solvation free energy of the dissociated proton is determined as the key factor to predict pK_a.And we propose that the acidity in the practical electrolyte is discrete.While the solvation capability differs for the additive and the solvent,more than one pK_a are needed to describe the solute acidity.For example,two pK_a values are needed for the LiI/DMSO and only one value is enough for the LiBr/DMSO.The concept of the discrete properties of the solute in the practical electrolyte is crucial for the solution electrochemistry and relevant applications.This protocol can also be extended to predict the other critical properties of the solvent in the practical electrolyte and guide the screening of stable solvents for the energy storage systems such as aprotic Li-ion battery.