Atomistic Modelling Approaches to the Challenges in Lithium-Sulfur Batteries

Abstract

Internal combustion engine vehicles (ICEVs) rely on petroleum-based fuels; these fuels have limited reserves and their combustion is a major contributor to climate change. These issues have triggered the development of electric vehicles (EVs) that use electricity stored in Li-ion batteries as a fuel. However, the performance of current Li-ion technology has largely plateaued, yet the driving range of EVs remains below that of conventional ICEVs. This limitation has sparked a search for safer battery chemistries with higher capacities than Li-ion at lower cost. The lithium-sulfur (Li-S) battery is one promising chemistry due to its high theoretical energy density and potential for low cost. Despite these benefits, Li-S batteries are not commercially feasible yet because their cycleability and practical energy density do not meet expectations. The limited performance is attributed to deficient understanding of several phenomena, including slow transport in the positive electrode, dissolution of polysulfide species, and the cycleability of Li metal anodes. This thesis aims to provide insight into these mechanisms using density functional theory (DFT) calculations, and thereby provide information aiding the design of Li-S systems. As a first step in understanding the fundamental features of Li-S batteries, the state-of-art DFT techniques were applied to predict several properties of redox end members in Li-S batteries. More specifically, the relative stabilities of the α-and β- sulfurs were confirmed by calculating the phase diagram. Similarly, the stability analysis of Li2S2 suggests that this is a metastable phase. The equilibrium crystallites of Li2S are predicted to be comprised entirely of stoichiometric (111) surfaces, while for α-S a mixture of several facets is predicted. Finally, α-S, β-S, Li2S, and Li2S2 are predicted to be insulators with band gaps greater than 2.5 eV. Regarding PS dissolution from the cathode, metal-organic frameworks (MOFs) are explored as cathode support materials. MOFs combine encapsulation and chemical adsorption as strategies for minimizing PS dissolution. Optimal MOF compositions are pinpointed by computationally screening 16 metal-substituted variants of M2(dobdc) for their ability to chemically anchor prototypical PS species. Ti2(dobdc), Ni2(dobdc), and Mo2(dobdc) are identified as the compositions with the largest affinities for Li2S4 and Li2S. As Ni2(dobdc) has been synthesized previously, this MOF is proposed as a promising cathode support for Li-S batteries. Charge transport limitations through insulating Li2S and α-S have the potential to constrain the capacity of Li-S batteries. Therefore, understanding these charge transport mechanisms is a prerequisite for enhancing cell performance. Charge transport mechanisms were explored in both the adiabatic and nonadiabatic regimes. Charge transfer in Li2S is predicted to be adiabatic. In sulfur, however, transitions between S8 rings are nonadiabatic and conventional DFT overestimates charge transfer rates. Delocalized holes are predicted to be the most mobile charge carriers in α-S; in Li2S hole polarons dominate. Finally, the last portion of this dissertation examines sulfide-based solid electrolytes (SSEs). These electrolytes are expected to enable the use of a high-capacity Li-metal anode. Realizing these systems requires stability between the SSEs and Li metal. This stability is investigated computationally by predicting band edge positions. Based on this information we predict the likelihood charge injection from Li anodes. Our calculations reveal that reduction by Li is expected for all SSEs examined. The position of the CBM is sensitive to the surface features of the SSE and varying the surface composition of Li3BS3 can stabilize the Li/Li3BS3 interface.