Reduction mechanism of sulfur in lithium–sulfur battery: From elemental sulfur to polysulfide

Highlights

• ${{\text{S}}_4}^{2-}$ and ${{\text{S}}_5}^{2-}$ were the major species at the first reduction wave of elemental sulfur.

• The polysulfides during the discharge of Li–S batteries were captured instantly.

• The polysulfides were accurately in-situ determined.

Abstract

The polysulfide ions formed during the first reduction wave of sulfur in Li–S battery were determined through both in-situ and ex-situ derivatization of polysulfides. By comparing the cyclic voltammetric results with and without the derivatization reagent (methyl triflate) as well as the in-situ and ex-situ derivatization results under potentiostatic condition, in-situ derivatization was found to be more appropriate than its ex-situ counterpart, since subsequent fast chemical reactions between the polysulfides and sulfur may occur during the timeframe of ex-situ procedures. It was found that the major polysulfide ions formed at the first reduction wave of elemental sulfur were the ${{\text{S}}_4}^{2-}$ and ${{\text{S}}_5}^{2-}$ species, while the widely accepted reduction products of ${{\text{S}}_8}^{2-}$ and ${{\text{S}}_6}^{2-}$ for the first reduction wave were in low abundance.

Introduction

Recently, rechargeable lithium sulfur (Li–S) and lithium air (Li-Air) batteries have drawn significant attention due to their high theoretical energy density [1]. Both batteries are considered to be potential candidates to replace state-of-art Li-ion batteries in electric vehicles (EVs). Although oxygen and sulfur are in the same group on the periodic table, the electrochemical redox reactions of these two are quite different. It's well known that the redox reaction of sulfur is one of the most complicated redox reactions and its mechanism is still not fully understood. It's believed that the reduction of the most stable form of elemental sulfur, cyclooctasulfur (S_8c), is a multistep reduction. The cyclooctasulfur is first electrochemically reduced into long chain linear polysulfides by the cleavage of the sulfur ring. The long chain sulfides are then further reduced into shorter chain polysulfides at a different potential [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12]. Correspondingly there are two reduction waves observed in the cyclic voltammetry and two discharge plateaus observed between 1.5 and 3.0 V vs. Li/Li⁺ in the discharge profile of a Li–S battery. We demonstrate in this paper that the stepwise sulfur reduction mechanism may not be accurate.

Unlike Li-ion cathode materials which are based on Li ion insertion, the redox reaction of sulfur involves polysulfides dissolution and re-deposition. Therefore, the sulfur cathode experiences compositional, structural, and surface alteration during charge and discharge. Due to the lack of a reliable analytical method for the quantitative and qualitative determination of the soluble polysulfides formed during the various stages of cell operation, the mechanism for the sulfur redox reaction is still debatable; for example, whether the very first charge transfer reaction is the electrochemical reduction of S_8c yielding linear ${{\text{S}}_8}^{2-}$ [8]. To fully understand the stepwise reduction of elemental sulfur in the Li–S battery, substantial fundamental research has been done by means of electrochemistry [4], [6], [13], [14], [15], UV–Vis spectroscopy [2], [3], [5], [7], [8], [9], [12], [16], [17], Raman [10], [11], [18], ESR [19], XRD [10], [19], [20], [21], [22], XAS [19], [22], as well as the theoretical calculations [23], [24], [25]. These studies revealed the complexity associated with the reduction of sulfur. Different mechanisms of reduction were proposed although most of them lacked direct and clear experimental proof. There are three mainstream mechanisms proposed to explain the first reduction wave at around 2.3 V vs. Li as shown in Scheme 1, Scheme 2, Scheme 3: 1) 2-electron electrochemical reduction process followed by chemical reactions (abbreviated 2EC) [5], [6], [9], [13], [16], [18]; 2) 1-electron electrochemical reduction process followed first by a chemical reaction and then by another one-electron electrochemical reduction (abbreviated ECE) [7], [17]; and 3) two successive 1-electron electrochemical reduction processes followed by a chemical reaction (abbreviated EEC) [4], [8], [14], [15]. Although similar analytical and electrochemical methods were used, different observations were reported, thus the different mechanisms described above were proposed according to the corresponding results. It is worth to emphasizing that the existence of ${{\text{S}}_8}^{\mathbf{.}-}$ radicals have never been proven experimentally, although the radicals with shorter sulfur chain length e.g. ${{\text{S}}_3}^{\mathbf{.}-}$ were detected by ESR experiments [19].

In the investigation of polysulfide species in an aqueous system, Lev et al. reported a derivatization method tandem with separation and identification by HPLC [26], [27], [28], [29]. Through reaction with methyl triflate, or methyl iodide, the polysulfide anions were derivatized into different dimethyl polysulfides. Based on the sulfur chain length in the dimethyl polysulfides, the baseline separation of different dimethyl polysulfides can be achieved by HPLC and each dimethyl polysulfide in the chromatogram can be identified based on the relationship between retention time and sulfur chain length. Both Barchasz et al. [30] and our group [31] recently reported the investigation of the discharge mechanism of the Li–S battery using ex-situ derivatization coupled with HPLC-MS. All eight polysulfide ions and elemental sulfur can be separated and identified by HPLC-MS after derivatization.

In this work, for the first time in-situ derivatization with HPLC was used to investigate the mechanism of sulfur reduction. By comparing the in-situ and ex-situ derivatization results under potentiostatic condition, the electrochemical mechanism of the first reduction wave of sulfur was discussed.