UC Berkeley

Lithium sulfur batteries have garnered a significant amount of attention as a next-generation energy storage technology. They have a theoretical specific capacity of 1672 mAh/g and a theoretical specific energy density of 2600 Wh/kg, which is five times greater than current lithium ion battery standards. Unfortunately, Li-S cells are plagued with numerous scientific problems that make practical implementation of the technology impossible. The overall reaction mechanism for the battery is given by S8 +16 e- + 16 Li+ → 8 Li2S. However, it is well-known that the actual reaction mechanism is much more complex, involving a multistep series of reactions through which lithium polysulfide reaction intermediates are formed. Lithium polysulfides are highly soluble in common battery electrolytes, and as a result, their formation during charge/discharge leads to their dissolution out of the cathode and into the cell electrolyte separator. This results in a direct loss of cell capacity, detrimental reactions at the cell anode, and ultimately, cell failure.

Despite over four decades of research, the redox reaction mechanisms that govern the Li-S charge/discharge processes are still unclear. This is primarily due to challenges associated with obtaining spectral ‘fingerprints’ for the lithium polysulfide intermediates (Li2Sx, 2 ≤ x ≤ 8, referred to as polysulfide dianions; or LiSx, 3 ≤ x ≤ 5, referred to as polysulfide radical anions). Numerous spectroscopy and characterization techniques have been used to study the Li-S redox reactions, but all have had issues obtaining unambiguous spectral standards for the different polysulfide dianion species. In this work, X-ray absorption spectroscopy at the sulfur K-edge is used to study Li-S battery reaction mechanisms and lithium polysulfide mixtures. First principles calculations of theoretical spectra of lithium polysulfide species are used to interpret results obtained for experimentally measured Li-S battery cells. These theoretical calculations circumvent the issues associated with obtaining spectral standards for polysulfide species experimentally. Fundamental studies of Li-S chemistry are a necessity to our ability to rationally address and overcome the obstacles that Li-S batteries face.

To begin, X-ray absorption spectroscopy at the sulfur K-edge was used to probe chemically synthesized mixtures of lithium polysulfide species dissolved in a block copolymer of poly(styrene)-poly(ethylene oxide) (SEO), and a homopolymer of poly(ethylene oxide) (PEO). For both solvents, a series of spectra were gathered for polysulfide mixtures that had stoichiometric Li2Sx ‘x’ values of 2, 4, 6 and 8. The system of experimental spectra obtained from XAS was analyzed using a statistical technique called principal component analysis. This analysis revealed that the polysulfide mixtures contained only three species: Li2S, Li2S4, and Li2S8. The parsimonious interpretation of these results suggests that in PEO-based solid electrolytes containing chemically synthesized polysulfide species, Li2S6 and Li2S2 disproportionate to form binary mixtures of Li2S4/Li2S8, and Li2S/Li2S4, respectively.

Next, XAS at the sulfur K-edge was used to examine Li-S cells that were discharged to different depths of discharge and allowed to reach equilibrium. The experimental geometry and novel cell construction was such that incoming X-rays primarily probed the lithium polysulfide species dissolved in the cell electrolyte. Analysis of the experimental spectra using theoretically calculated spectra from first principles revealed that polysulfide radical anions were present in the Li-S cell electrolyte after discharge. However, evidence of radical polysulfide species was only obtained for a cell that was stopped at the midpoint of the first discharge plateau. No evidence of polysulfide radical species was found at increased depths of discharge. This suggests that polysulfide radical species are formed during early stages of discharge, or that polysulfide radical species are formed through chemical disproportionation reactions involving polysulfide dianion species electrochemically created during the initial stages of discharge. The detection of radical species was especially notable given that the electrolyte used in the Li-S cell was an ether-based polymer electrolyte (SEO). While it had already been established that radicals were stable in electrolytes with high electron pair donor numbers, it was unclear whether or not radical species could be stabilized in ether-based solvent (which have low electron pair donor numbers).

The appearance of polysulfide radical species in the electrolyte of partially discharged Li-S cells motivated a further examination of the stability of radical species in ether-based electrolytes. Lithium polysulfide species dissolved in PEO and a PEO oligomer of tetraethylene glycol dimethyl ether (TEGDME) were probed using a combination of ultraviolet-visible (UV-vis) spectroscopy and electron paramagnetic resonance (EPR) spectroscopy. EPR results unambiguously confirmed the presence of radical species in ether-based electrolytes. Comparison of the EPR spectra to corresponding UV-vis spectra established that the UV-vis absorbance signature for radical species in ether-based solvents occurs at a wavelength of 617 nm. Additionally, analysis of the UV-vis spectra using the Beer Lambert law allowed for the determination of polysulfide radical concentration and the fraction of sulfur that was present in the form of radical species. As sulfur concentration increased, the fraction of sulfur (on an atomic basis) present in the form of radical species decreased. That is, polysulfide radical species are less stable at higher concentrations of sulfur (and lithium) and likely recombine to form dianion species (e.g. through reactions of the kind: 2 LiS3 → Li2S6).

Multiple authors have shown that in order for Li-S batteries to succeed, Li-S cathodes need to be thicker than what is typically used in Li-S battery research. Little is known about the fundamental reaction mechanisms and chemical processes that take place in thick cathodes, as most research has focused on studying thinner cathodes that enable high performance. In this part of the dissertation work, in situ XAS at the sulfur K-edge was used to probe the back of a thick Li-S cathode during discharge. Interpretation of the experimental spectra using theoretically derived spectra, and analysis of the fluorescence intensity revealed that lithium polysulfide dianion species formed in the front of the cathode during discharge diffused to the back of the cathode during discharge. Additionally, high conversion of elemental sulfur in the back of the cathode is achieved through chemical disproportionation reactions between elemental sulfur and polysulfide dianion species.