Trimethylsilyl Compounds for the Interfacial Stabilization of Thiophosphate‐Based Solid Electrolytes in All‐Solid‐State Batteries

Abstract

Argyrodite‐type Li6PS5Cl (LPSCl) has attracted much attention as a solid electrolyte for all‐solid‐state batteries (ASSBs) because of its high ionic conductivity and good mechanical flexibility. LPSCl, however, has challenges of translating research into practical applications, such as irreversible electrochemical degradation at the interface between LPSCl and cathode materials. Even for Li‐ion batteries (LIBs), liquid electrolytes have the same issue as electrolyte decomposition due to interfacial instability. Nonetheless, current LIBs are successfully commercialized because functional electrolyte additives give rise to the formation of stable cathode‐electrolyte interphase (CEI) and solid‐electrolyte interphase (SEI) layers, leading to supplementing the interfacial stability between electrolyte and electrode. Herein, inspired by the role of electrolyte additives for LIBs, trimethylsilyl compounds are introduced as solid electrolyte additives for improving the interfacial stability between sulfide‐based solid electrolytes and cathode materials. 2‐(Trimethylsilyl)ethanethiol (TMS‐SH), a solid electrolyte additive, is oxidatively decomposed during charge, forming a stable CEI layer. As a result, the CEI layer derived from TMS‐SH suppresses the interfacial degradation between LPSCl and LiCoO2, thereby leading to the excellent electrochemical performance of Li | LPSCl | LiCoO2, such as superior cycle life over 2000 cycles (85.0% of capacity retention after 2000 cycles).

Interfacial Degradation Mechanism of Nanostructured LiCoO2 for Li6PS5Cl-Based All-Solid-State Batteries

“…Figure S3e presents the XPS spectra of the S 2p peak for LPSCl powder and its mixed pellets with Li 2 SnO 3 and LiCoO 2 after storage under an Ar atmosphere at room temperature for 48 h. When LPSCl was in contact with Li 2 SnO 3 , no significant side reactions were observed on the surface. In contrast, upon contact with LiCoO 2 , numerous byproducts, such as SO 3 , SO 4 , and P 2 S x , were detected at the interface. ,, This suggests that Li 2 SnO 3 suppresses the decomposition of LPSCl on the surface of S-LiCoO 2 compared to bare LiCoO 2 .…”

“…Figure compares the ex situ TOF-SIMS spectra of bare LiCoO 2 and S-LiCoO 2 before and after cycling. The relative amounts of decomposed species of LPSCl were estimated from the normalized peak intensities of each species, where the intensity of each fragment was divided by the total intensity of all fragments. ,, Before cycling, both bare LiCoO 2 and S-LiCoO 2 showed negligible amounts of oxygen-containing compounds, such as SO x – and PO y – (with 1 ≤ x ≤ 4 and 1 ≤ y ≤ 4) fragments. However, the intensities of these oxidized fragments increased significantly after cycling.…”

“…In contrast, upon contact with LiCoO 2 , numerous byproducts, such as SO 3 , SO 4 , and P 2 S x , were detected at the interface. 6,15,16 This suggests that Li 2 SnO 3 suppresses the decomposition of LPSCl on the surface of S-LiCoO 2 compared to bare LiCoO 2 .…”

“…To tackle these issues, there is a growing interest in transitioning from conventional LIBs to the next generation of all-solid-state batteries (ASSBs) using solid electrolytes. , Although various solid electrolytes, including oxide, sulfide, and polymer-based ionic conductors, have been explored for ASSBs, thiophosphate-based solid electrolytes, such as Li 6 PS 5 Cl (LPSCl), have attracted intensive attention due to their mechanical flexibility, high ionic conductivity, and manufacturability. − However, the transition to ASSBs also encounters difficulties related to structural and mechanical degradation at the interface between thiophosphate solid electrolytes and layered oxide-based cathode materials. This gradual interfacial degradation during cycling gives rise to an increasing charge-transfer resistance and, eventually, poor capacity retention. − …”

“…Figure 5 compares the ex situ TOF-SIMS spectra of bare LiCoO 2 and S-LiCoO 2 before and after cycling. The relative amounts of decomposed species of LPSCl were estimated from the normalized peak intensities of each species, where the intensity of each fragment was divided by the total intensity of all fragments 6,12,52. Before…”