DOI: /D0SCF (Minireview) Chem. Sci., , 11,
Tengfei Zhang†*a, Wenjie He†a, Wei Zhang *b, Tao Wang a, Peng Li a, ZhengMing Sun b and Xuebin Yu *c
aJiangsu Key Laboratory of Electrochemical Energy-Storage Technologies, College of Materials Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing , China. E-mail: firstname.lastname@example.org
bJiangsu Key Laboratory of Advanced Metallic Materials, School of Materials Science and Engineering, Southeast University, Nanjing, , China. E-mail: email@example.com
cDepartment of Materials Science, Fudan University, Shanghai , China. E-mail: firstname.lastname@example.org
Received 4th June , Accepted 18th July
First published on 20th July
Solid-state electrolytes (SSEs) are capable of inhibiting the growth of lithium dendrites, demonstrating great potential in next-generation lithium-ion batteries (LIBs). However, poor room temperature ionic conductivity and the unstable interface between SSEs and the electrode block their large-scale applications in LIBs. Composite solid-state electrolytes (CSSEs) formed by mixing different ionic conductors lead to better performance than single SSEs, especially in terms of ionic conductivity and interfacial stability. Herein, we have systematically reviewed recent developments and investigations of CSSEs including inorganic composite and organic–inorganic composite materials, in order to provide a better understanding of designing CSSEs. The comparison of different types of CSSEs relative to their parental materials is deeply discussed in the context of ionic conductivity and interfacial design. Then, the proposed ion transfer pathways and models of lithium dendrite growth in composites are outlined to inspire future development of CSSEs.
Tengfei Zhang is an associate professor in the College of Materials Science and Technology at Nanjing University of Aeronautics and Astronautics (NUAA). He obtained his BE from the School of Materials Science and Engineering, Central South University, China and PhD in Advanced Materials from Hokkaido University, Japan. He then worked as a JSPS fellow at Hokkaido University and an assistant professor at Hiroshima University. His current research interests are in the areas of solid-state electrolytes, hydrogen storage materials, and in situ TEM.
Wenjie He received his M.S. degrees in material engineering from Shaanxi University of Science and Technology in He is currently pursuing his PhD degree under the supervision of Prof. Xiaogang Zhang in applied chemistry at Nanjing University of Aeronautics and Astronautics. His current research focuses on advanced materials for electrochemical energy storage devices, such as lithium-ion batteries and all-solid-state batteries.
Wei Zhang is an associate professor in School of Materials Science and Engineering at Southeast University. He obtained his MASc in Chemistry and PhD in Nanotechnology from the University of Waterloo. After joining Southeast University in , he established the Surface Science and Bionanomaterials Laboratory, and is working on both fundamental and applied research to meet the growing need of nanotechnology in advanced materials, e.g. multifunctional and smart hydrogels and aerogels, flexible energy storage devices or electronic packaging. He was awarded the Periodic Table of Younger Chemists in by the Chinese Chemical Society.
Xuebin Yu received his PhD degree from the Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Science in He then worked as a postdoctoral fellow at the University of Nottingham and the University of Wollongong from January to December and March to March , respectively. In , he joined Fudan University and now works as a professor in the Department of Materials. His research interests cover hydrogen storage, fuel-cell integration with hydrogen systems, hydride-based solid-state electrolytes, lithium/sodium-ion batteries, and preparation of nanomaterials for energy storage.
As efficient energy storage devices, batteries, including nickel-metal hydride (Ni-MH) batteries, lead-acid batteries and lithium-ion batteries (LIBs), can be effectively combined with renewable energy sources such as solar energy, wind energy and hydrogen energy, and such batteries are expected to be advanced energy storage systems and reduce fossil fuel dependence.1
In the past few decades, the energy density of LIBs has been higher than that of many other types of batteries; they have been widely used in the fields of personal electronics, hybrid and electrical vehicles, and grid energy conversion–storage systems.2,3
Although LIBs have appeared in daily life in many forms, currently available commercial LIBs still cannot meet the stringent or increasing demand for high-performance equipment in modern society.4–16
There is an urgent need to develop batteries with higher energy density and longer cycle life, along with an acceptable level of safety and an affordable price. Liquid electrolytes demonstrate high conductivity and excellent electrode surface wettability, but they often suffer from the problems of insufficient electrochemical and thermal stability, low ion selectivity and poor safety.17–25
Concurrently, lithium metal, as a negative electrode, exhibits low redox potential (− V) and high theoretical specific capacity ( mA h g−1), while also being light-weight (M = g mol−1, D = g cm−3).26,27 However, as the charging and discharging process continues, the gradual growth of lithium dendrites can pierce the separator, leading to a short circuit in the battery.28 Furthermore, the electrolyte is a flammable substance. If leaked, it can cause serious safety accidents. In addition, an unstable solid electrolyte interphase (SEI) is formed between the liquid electrolyte and the electrode material.27,29–35 The appearance of an SEI will reduce battery capacity and shorten the cycle life of the battery. Therefore, LIBs using lithium metal as the negative electrode have poor cycling performance and serious safety problems. The emergence of SSEs is expected to enable lithium metal to be used as a negative material in all solid-state battery systems, resulting in a higher energy density than what is currently available. Replacing liquid electrolytes with SSEs can also effectively inhibit the generation of SEI films and improve the cycle performance of batteries. At the same time, the size of the battery is reduced, and its application scope is expanded. More importantly, compared to liquid electrolytes, SSEs (e.g., inorganic ceramic electrolytes) do not leak and are non-flammable, and thus, safety is greatly enhanced. The use of solid-state electrolytes instead of liquid electrolytes can not only overcome the liquid electrolyte durability problem, but also provide an important avenue for developing next-generation LIBs.
With the emergence of LIBs, research on SSEs continued. Manthiram et al. provided a detailed account of the development of SSEs in 11,36 Solid electrolytes can be divided into two categories: solid polymer materials and inorganic materials. Typical conductive polymers are polyethylene oxide (PEO),37 polyacrylonitrile (PAN),38,39 polymethyl methacrylate (PMMA),40,41 and polyvinylidene fluoride (PVDF).42,43 In polymer electrolytes, the polymer long chains are partially shifted above the glass transition temperature, thereby creating a binding site for ion hopping, and the constant hopping of ions in a specific direction can achieve an ion transport effect. Typical inorganic lithium-ion conductive materials include lithium phosphorus oxynitride (LiPON),44 perovskite,45 sodium superionic conductor (NASICON),46–48 garnet,15,49–56 sulfide,10,17,57–64 halide,65–70 and hydride based materials.71–76 Among them, the high ionic conductivity of solid crystalline materials can be attributed to the large number of structural defects or special crystal structures. The former involves point defect-based ion diffusion mechanisms, including simple vacancy mechanisms and relatively complex diffusion mechanisms, while the latter involves usually two types of sublattices, which are composed of immobile and mobile ions. The ionic diffusion process in glass materials is similar to that of crystalline materials in that ions move between active sites.
High ionic conductivity, low ionic area specific resistance, high electronic area specific resistance, high ionic selectivity, a wide electrochemical stability window, good chemical compatibility, excellent thermal stability, excellent mechanical properties, simple fabrication processes, and environmental friendliness are the main properties of a good solid-state electrolyte.11 Much progress has been made in improving the above-mentioned properties. Perovskite-type oxide compounds, in the form of ABO3, show a lithium ionic conductivity as high as 10−3 S cm−1 in the bulk portion, which has been attributed to the presence of a significant amount of equivalent deficient sites for lithium ions to substitute and freely move in bulk crystals. Additionally, these compounds (e.g., Li3xLa2/3xTiO3) have a high grain boundary resistance that is related to the reduction of Ti4+ to Ti3+ upon contact with Li metal.77,78 Garnet-type materials, such as Li5La3Ta2O12 and Li5La3Nb2O12, have also been considered fast lithium-ion conductors since 79 Since then, a new garnet-type ceramic of Li7La3Zr2O12 has been discovered with a high ionic conductivity ( × 10−4 S cm−1).80 Although these ceramics are chemically stable with electrodes, their poor interfacial compatibility with lithium metal limits their applications in the field of solid-state batteries. Furthermore, the volume expansion and shrinkage of the electrode during the charging/discharging process will cause the electrolyte to crack and thereby lose its capacity. The mechanical flexibility of SSEs determines whether they easily crack.81 Therefore, electrolytes should have a moderate elastic modulus. In general, the mechanical flexibility of sulfide-type materials is better than that of oxide-type materials. Additionally, the replacement of oxygen ions with sulfur ions, which have a larger radius, can not only provide more migration space for lithium ions, but also reduce bonding strength. As a result, compared with oxides, sulfides exhibit higher lithium-ion conductivity with a value of approximately 10−2 to 10−4 S cm−1 at room temperature (RT).82 For example, Li10GeP2S12 exhibits an extremely high lithium ionic conductivity of × 10−2 S cm−1 at RT owing to its three-dimensional framework structure.16 Notably, the low output and high cost of germanium limit the possibility of its large-scale production and application. Moreover, most sulfide solid electrolytes are not stable and can easily react with H2O, releasing a highly toxic gas: H2S.83 More recently, a complex hydride lithium superionic conductor, Li(CB9H10)–Li(CB11H12), has been developed with excellent stability against lithium metal and a high conductivity of × 10−3 S cm−1 at K.84 However, the compatibility of the above material with cathode materials is poor due to the reducibility of complex hydrides. Unlike brittle crystalline solid inorganic electrolytes (SIEs), solid polymer electrolytes (SPEs) are highly flexible. Nonetheless, SPEs have not been applied in commercial batteries due to their low ionic conductivities (10−6 to 10−8 S cm−1) at RT and poor electrochemical stability (<4 V).85
Based on the excellent work of many researchers in analyzing various electrolytes, it is difficult for a single type of SSE to fully satisfy the challenging requirements mentioned above. This has led to a growing research interest in CSSEs, which aims to develop promising SSEs by combining the advantages and eliminating the drawbacks of both inorganic and organic solid electrolytes. A brief overview of CSSEs is first introduced (Fig. 1).11 Inspired by the radar plot from the work of Manthiram et al.11 (Fig. 1a), it is necessary to summarize the recent progress of composite solid-state electrolytes. Herein, this review systematically surveys recent CSSE progress, with special emphasis on the following aspects: polymer-based, oxide-based, hydride-based, sulfide-based, and halide-based SSEs. The conductivity, interphase behavior, electrochemical stability and properties of these electrolytes, along with their associated all-solid-state LIBs are also systematically summarized and characterized (Fig. 1b). We focus on the design of the electrolyte itself and discuss the challenges in developing these materials. Additionally, CSSEs are discussed with regard to their application in many Li battery systems, including Li-ion, Li-metal, and Li–sulfur batteries.
|Fig. 1 (a) The superposed radar plot of different types of SSEs and requirements for SSEs. ASR, area-specific resistance. Reproduced from ref. 11 with permission from Springer Nature, copyright (b) Design of CSSEs from single type SSEs and corresponding physicochemical characteristics.|
Progress of CSSEs
Solid-state ionics is the science of ion transport in solids. The term “ionic” has been used for ionic conduction since the early days of electrochemical research, but mainly for liquid electrolytes. The field of solid-state ionics first started with the work of Faraday in heated solid electrolytes of Ag2
S and PbF2
In general, the mobility of ions in solids is very slow, and hardly contributes to electrical conductivity. However, high ionic mobility is observed in certain types of inorganic ceramics, organic polymers, and composite materials, and their ionic conductivities are comparable to that of a liquid. In the s, a ceramic-based β-alumina (Na2
) was found to possess a remarkable sodium-ion transport characteristic, and was successfully used in a high-temperature sodium–sulfur (Na–S) battery for grid energy conversion–storage systems.87
It is marked as a milestone and subsequently boosted the increase in practical applications of SSEs. For example, improved Na–S battery modules were developed and made commercially available by NGK Insulators Ltd. in 88
Recently, increasing research has been made to realize the application of solid electrolytes. In particular, increasing attention has been turned from purely inorganic SSEs or purely organic SPEs to CSSEs. The research of these CSSEs has been focused on designing innovative superionic conductors, understanding the ion transport mechanism at the interface, and improving the electrochemical performance based on CSSEs. In the following sections, a comprehensive introduction to different CSSEs, including polymer-based, oxide-based, hydride-based, sulfide-based and halide-based CSSEs, will be presented and discussed.
Since , significant attention has been paid to SPEs owing to their ease of synthesis, low shear modulus, low cost, compatibility with large-scale manufacturing processes and inherent mechanical toughness. However, SPEs exhibit a low oxidation voltage and poor thermal stability. Additionally, SPEs show particularly low ionic conductivity (10−6
) at room temperature because the polymer chains are locked in a crystal lattice which hinders ion-pair dissociation. Improvements in ionic conductivity and interfacial resistance between the electrolyte and electrodes are still unable to satisfy the requirements for practical applications. To address these problems, various physical approaches and chemical strategies, such as polymer–inorganic material blending, architectural design of inorganic fillers, copolymerization, crosslinking and the introduction of ionic side groups, have been adopted.
Researchers began to invest more effort to balance electrochemical stability and mechanical robustness in the development of composite polymer electrolytes (CPEs). The ionic conductivities of composite membranes were optimized as the content of Li7La3Zr2O12 changes. The composite membranes exhibited the highest ionic conductivity ( × 10−4 S cm−1 at 55 °C) and maintained mechanical flexibility, consisting of an organic PEO matrix with the optimum composition (% Li7La3Zr2O12).89 In contrast to conventional blending methods, Cui et al. reported a new approach for the preparation of ceramic–polymer electrolytes via in situ synthesis of ceramic filler particles in polymer electrolytes. The improved distribution of monodisperse ultrafine SiO2 particles in PEO helped increase the effective surface area and suppress the crystallization of PEO, thus facilitating polymer segmental motion for ionic conduction. All of these factors led to good ionic conductivity ( × 10−3 S cm−1 at 60 °C, × 10−5 S cm−1 at 30 °C) and greatly extended the electrochemical stability window up to V.90 Chung et al. reported a novel composite electrolyte fabricated by a simple solution casting method. The composite electrolyte (LiLa3ZrTaO12 (LLZTO) fillers in a PEO/LiClO4 matrix) exhibited low interfacial resistance and good Li-ion conductivity ( × 10−4 S cm−1 at 60 °C).91
Li+-conducting oxides are considered better fillers than Li+-insulating oxides for improving Li+ conductivity and distribution in a composite electrolyte. To explore this possibility, a PEO/perovskite Li3/8Sr7/16Ta3/4Zr1/4O3 (LSTZ) electrolyte was prepared. An interphase layer was in situ formed during cycling with the electrolyte membrane, which indicated the strong interaction between F− and the surface Ta5+. This strong interaction improved Li-ion transport at the PEO/perovskite interface and suppressed lithium dendrite growth.92 More recently, Goodenough et al. used two Li+-insulating oxides, fluorite GdCeO (GDC) and perovskite LaSrGaMgO (LSGM), as ceramic fillers in the PEO matrix. Density functional theory (DFT) calculations and 7Li nuclear magnetic resonance (NMR) measurements confirmed the interaction between the oxide surface and the Li-salt anion in the polymer, which modified the activation energy for Li+ transport to obtain a high Li+ conductivity that was above 10−4 S cm−1 at 30 °C.93 Cui et al. reported the facile synthesis of Al3+/Nb5+ co-doped Li7La3Zr2O12 (LLZO) nanoparticles. The substitution of Li+ by Al3+ enhanced the stabilization of cubic LLZO at RT, and the substitution of Zr4+ by Nb5+ improved the ion conductivity. After optimization, the polymer electrolyte with 15 wt% LLZO showed an improved conductivity of × 10−6 and × 10−4 S cm−1 at 20 and 40 °C, respectively.94 There was a higher Li-ion conductivity in one/two-dimensional fillers than in zero-dimensional particles. One-dimensional LiLaTiO3 (LLTO) nanofiber embedded in a PEO matrix, as shown in Fig. 2a–c, provided continuous ionic transport pathways and reduced interfacial resistance.95 Moreover, two-dimensional garnet nanosheets were first reported via co-precipitation with a graphene oxide (GO) template. The specially designed CPE containing garnet nanosheets could robustly isolate Li dendrites and exhibited a conductivity of × 10−4 S cm−1 at RT.96
|Fig. 2 (a) Schematic illustration of the carbon nanofiber (CNF)/S-PEO/LLTO CSSE. (b) SEM images and (c) cross-sectional SEM image of the PEO/LLTO CSSE. Reproduced from ref. 95 with permission from Elsevier Inc., copyright (d) Graphical representation of the 3D framework in cross-linked nanocomposite polymer electrolytes (CNPEs). Reproduced from ref. 97 with permission from Elsevier Inc., copyright (e) Schematic illustration of the different interfacial characteristics between the PVDF-based CPE and PVDF/PVAC-based CPE. (f) Cross-sectional SEM images of the symmetric battery and Li metal after Li plating/stripping h. Reproduced from ref. with permission from Wiley-VCH, copyright |
Currently, the general strategies are adding integrated inorganic fillers to the SPE matrix or preparing polymer electrolytes with specific intermolecular interactions, which can not only improve the ionic conductivity but also sustain a high working voltage. Mai et al. provided a new approach to prepare cross-linked nanocomposite polymer electrolytes based on hydrophobic–hydrophilic–hydrophobic triblock copolymers (PPO–PEO–PPO) in Fig. 2d. In this enhanced framework, polymer-based composite electrolytes were generated from copolymers and surface-modified SiO2 nanoparticles, which led to the effective solvation of lithium salts and encapsulation of organic solvents. Thus, the electrolytes exhibited high ionic conductivity and their electrochemical stability window was extended to V.97 The experiments, together with theoretical calculations, demonstrated that the gas releasing behaviour of PEO may be ascribed to the high oxidizing ability of delithiated LiCoO2. PEO was reported to decompose and release oxygen from LiCoO2 when more than Li+ was removed.98 The surface catalytic effect of delithiated LiCoO2 caused oxidation/dehydration of PEO-based SPBs and unexpected H2 generation at V. To mitigate the surface catalytic effect, the surface of LiCoO2 was coated with a stable solid electrolyte LiAlTi(PO4)3 (LATP), thus avoiding direct contact with PEO and therefore extending the stable working voltage to over V.99 Park et al. synthesized a self-standing and flexible CPE through the introduction of poly-(ethylene glycol)-dimethyl ether (PEGDME) to be plasticized in a PEO matrix with a uniformly distributed ceramic filler. After the addition of PEGDME, the stable working voltage of the PEO matrix was extended to V. Besides PEO-based polymer electrolytes, a PVDF/polyvinyl acetate (PVAC)/LLZTO CPE was first fabricated to achieve high RT ionic conductivity and a high electrochemical stability window ( V) (Fig. 2e). More importantly, the intermolecular interactions of tetramethylene sulfone (TMS) combined with PVAC or PVDF showed distinct differences. The Li/PVDF/PVAC-based CPE/Li symmetric battery shows a good inhibition of lithium dendrite growth in Fig. 2f. The PVAC/TMS layer formed on both the cathode and anode interfaces constructed an effective sulfurous Li+ transport pathway. Therefore, TMS with low flammability and excellent stability was able to selectively interact with only PVAC, which was helpful to enhance lithium-ion conductivity and electrode/electrolyte interfacial compatibility.
Recently, sulfide electrolytes, such as Li2S–GeS2, Li2SeP2S5, Li2SeB2S3 and Li2SeSiS2, have attracted increasing attention due to their superior ion conductivity (∼10−2 S cm−1) and wide potential window (>10 V). Furthermore, sulfide electrolytes have cheaper precursors and simpler processes to provide the electrolyte with a large potential in all-solid-state lithium batteries (ASSLBs). However, little success has been achieved in adopting lithium metal anodes with sulfide-based electrolytes in ASSLBs. The main challenges are the interfacial instability and Li dendrite formation between Li metal and SSEs.
To solve these issues, Li10GeP2S12 (LGPS) was dispersed in a PEO-based polymer to fabricate SPE membranes. The inorganic LGPS in the organic PEO matrix impeded crystallization and weakened the interactions between the Li+ and PEO chains. The optimal SPE containing 1% LGPS electrolyte exhibited a maximum ionic conductivity of × 10−3 S cm−1 at 80 °C and a broadened electrochemical window up to V. Inspired by the similarity between the H bond and Li bond, hybrid solid electrolytes were prepared via an in situ coupling reaction. A commercialized silane coupling agent was used as a bridge builder to realize the chemical bonding interaction between LGPS, polyethylene glycol (PEG), and PEO (Fig. 3a and b). Hence, the optimal ceramic/polymer hybrid solid electrolyte (HSE) membrane provided an expressway for the transport of Li+, and the growth of lithium dendrites was suppressed. Li6PS5Cl is a promising solid electrolyte in ASSLBs. In the preparation process, S and Cl easily formed chemical bonds with the polymer groups and replaced other anion sites. A Li6PS5Cl/PEO composite electrolyte with enhanced mechanical properties and a stable interface was fabricated by a liquid-phase process. In particular, with an optimal value of 5 wt% PEO, the CSEs show an improved ionic conductivity and electrochemical window. Sun et al. reported a plastic crystal electrolyte (PCE) interlayer to address the interfacial challenge and lithium dendrite formation between sulfide electrolytes and Li metal. Using PCE as an interlayer, interfacial reactions could be avoided by preventing contact between the sulfide electrolytes and Li metal (Fig. 3c–e). In addition, submerging the cathode in a PCE matrix forms a continuous 3D-conduction pathway for Li+ on the cathode side. The synchrotron-based X-ray absorption spectra in Fig. 3e were used to analyze the interface between LGPS and Li metal, which suggests that using the PCE interlayer can prevent the reduction of LGPS by Li metal.
|Fig. 3 (a) Schematic illustration of the as-prepared HSE structure. Electrochemical characterization of the HSE membranes. (b) Arrhenius plot of different electrolytes from 0 to 50 °C. Reproduced from ref. with permission from Wiley-VCH, copyright Schematic diagrams of (c) ASSLMBs and (d) ASSLMBs with the PCE interlayer. X-ray absorption (e) P K-edge and S K-edge spectra of LGPS before cycling, LGPS on the Li surface after cycling, and LGPS with the PCE interlayer after cycling, respectively. Reproduced from ref. with permission from Wiley-VCH, copyright |
Oak Ridge Laboratory first synthesized LiPON in the s.
Since then, oxide electrolytes have attracted increasing attention. Compared to SPEs, oxide-based electrolytes are becoming a research hotspot owing to their chemical stability in air, good ion selectivity and wide electrochemical window.15
Unfortunately, an increasing number of studies have demonstrated that lithium dendrites occur in most oxide-based electrolytes owing to poor interfacial stability, large interfacial resistance and voids and cracks inside the electrolytes. To develop an interface with chemical and mechanical stability, researchers have adopted interfacial engineering by introducing artificial buffer layers to protect these SSEs against lithium dendrite growth.,
Nevertheless, simple physical contact could not solve the problems of large area specific resistance and dendrite formation. Due to a high shear modulus, the problem of interfacial contact was initially solved by designing a soft interface, such as polymer layers, amorphous oxides and metal coatings.–
In practice, Ding et al. demonstrated that the coating layer of PEO (LiTFSI) could effectively prevent side reactions between the LiAlGe(PO4)3 (LAGP) and Li anode. It was exciting that the CSSE remained stable even at a high potential of V. Considering the excellent performance of the polymer layer, Goodenough et al. developed a low-cost composite ceramic/polymer solid-state electrolyte (CPSE) containing up to 80 wt% garnet LiLa3ZrTaO12 (LLZTO).56 Composites consisting of compositions ranging from “ceramic-in-polymer” to “polymer-in-ceramic” were found to be flexible and mechanically robust (Fig. 4a–d). The crystalline LLZTO particles not only increased the chain segment motion in a PEO matrix but also afforded an alternative Li+-conduction pathway. Compared with the “ceramic-in-polymer” electrolyte with high flexibility, the “polymer-in-ceramic” electrolyte was more suitable for large-scale application in electric vehicles owing to its high mechanical strength and safety. The PEO–LLZTO composite electrolyte showed the highest ionic electrolyte conductivity (above 10−4 S cm−1 at 55 °C) and suppressed Li dendrite growth. Coincidentally, to design an effective approach to achieve dendrite-free CSSEs, Sun et al. investigated the performance of composite electrolytes from “ceramic-in-polymer” to “polymer-in-ceramic” (Fig. 4e–g). The different sizes of garnet particles embedded in the electrolyte improved ionic conductivity and tensile strength. The satisfactory sandwich-type composite electrolyte with hierarchical garnet particles simultaneously achieved dendrite suppression and excellent interfacial contact with the Li metal in Fig. 4g. The “polymer-in-ceramic” interlayer with 80 vol% 5 μm LLZTO showed a high mechanical strength of MPa and hindered Li dendrite propagation due to physical obstacles. The “ceramic-in-polymer” thin-film outer layers with 20 vol% nm LLZTO particles created a smooth and flexible surface with a high t+ (Li+ transference number) of Symmetric solid-state cells with Li maintained highly stable plating/stripping cycling for h under mA cm−2 at 30 °C. Full SSBs with a LiFePO4 cathode and Li metal anode delivered a RT specific capacity of mA h g−1 with a good capacity retention of % after cycles at C.
|Fig. 4 (a) Schematic illustration of PEO–LLZTO CSSE: “ceramic-in-polymer”; “intermediate”; “polymer-in-ceramic”, and corresponding (b–d) SEM images. Reproduced from ref. 56 with permission from Elsevier Inc., copyright The schematic illustrations of (e) PIC-5 μm, (f) CIP nm, and (g) hierarchical sandwich-type CSSEs, and corresponding SEM images. Reproduced from ref. with permission from Wiley-VCH, copyright |
The design of artificial structures can not only be beneficial for interfacial contact but also improve mechanical properties. Composite electrolytes with high loadings of ceramic fillers (>50 vol%) are necessary to achieve the required mechanical modulus. However, high loading can lead to a large interparticle contact resistance and insufficient particle–particle contact area, thus resulting in greatly decreased ionic conductivity. Therefore, designing a composite with an interconnected ceramic network takes advantage of the high ionic conductivity of the ceramic. Inspired by the structure of natural nacre, Yang et al. fabricated a “brick-and-mortar” arrangement of solid electrolytes with ceramic electrolyte microplatelets as the “brick” and polymer electrolytes as the “mortar” (Fig. 5a and b). Compared to pure ceramic electrolytes, the nacre-like ceramic/polymer composite electrolyte simultaneously exhibited high fracture strain and an ultimate flexural modulus to accommodate external deformation. The staggered microstructure provided a high fracture strain of % and a flexural modulus of GPa. An ice template method was used to build a vertically aligned ceramic/polymer composite electrolyte (Fig. 5c and d), which was composed of LAGP, with a high ionic conductivity, and PEO. The vertical LAGP walls in the polymer provided fast ionic transport channels while retaining matrix flexibility. The ideal structure maximized the ionic conductivity of the composite electrolyte ( × 10−4 S cm−1 at RT and × 10−3 S cm−1 at 60 °C). In Chen's work, a composite solid electrolyte with a 3D-interconnected ceramic network was fabricated using a readily scalable processing method, as shown in Fig. 5e and f. The composite electrolyte had high ceramic loadings (77 wt% and 61 vol%) and demonstrated high ionic conductivity ( × 10−5 S cm−1 at RT) as well as good mechanical strength ( MPa).
|Fig. 5 (a) Schematic of the staggered “brick-and-mortar” microstructure of the LAGP–PEO NCPE film, and the corresponding (b) cross-sectional SEM image. Reproduced from ref. with permission from Wiley-VCH, copyright (c) The schematic of the preparation process of the ice-templated LAGP/PEO CSSE, and the corresponding (d) SEM images. Reproduced from ref. with permission from Elsevier Inc., copyright (e) Schematic illustration of the fabrication procedure of the composite electrolyte film, and the corresponding (f) SEM images. Reproduced from ref. with permission from Elsevier Inc., copyright |
In addition to excellent performance, the cost of processing should be considered in practical applications. Oxide electrolytes with superior ionic conductivity are synthesized at a high temperature, increasing the cost of the preparation process as well as the risk to the operation process. Therefore, developing a feasible method to address energy consumption and safety issues is reasonably significant for SSEs in practical applications. Liquid-phase sintering additives have received considerable attention because they have a low melting point and form a liquid phase. Li-based glass ceramics, such as Li2O, Li3BO3 (LBO), and LiSiO4, are utilized for densifying LLZO at low temperatures.– In addition to borate electrolytes, halide electrolytes Li3OX (X = Cl and Br) also display good electrochemical stability with Li metal and a low melting temperature (Tm ≈ °C)., Taking advantage of the above merits, Li3OCl was introduced into the voids and boundaries of Ta-doped LiLa3ZrTaO12 (LLZTO) particles at °C (Fig. 6). In LLZTO–xLi3OCl CSSEs, amorphous Li3OCl, as a binder, filler and bridge, promoted the formation of an integrated continuous ion-conductive network among the LLZTO particles. Furthermore, Li3OCl, with excellent affinity, in situ reacted to form a stable and dense interfacial layer, which greatly decreased the interfacial resistance (from to 90 Ω cm2) and effectively suppressed lithium dendrite growth. The integrated CSSEs (LLZTO–2wt% Li3OCl) with compact and stable structures presented optimal ionic conductivity ( × 10−4 S cm−1), and their electrochemical stability window was up to 10 V at RT. The above studies highlighted a novel strategy for developing integrated and compact CSSEs at ultralow temperature for high-performance ASSLBs.
|Fig. 6 Schematics of Li deposition behavior using (a) LLZTO SSE and (b) LLZTO–2wt% Li3OCl CSE. Reproduced from ref. with permission from Elsevier Inc., copyright |
In contrast to oxide and polymer SSEs, hydride SSEs belong to the complex hydride family and have been widely investigated as solid-state hydrogen-storage materials.,
To date, there has been little research on ionic conduction in complex hydrides owing to their poor ionic conductivity at moderate temperatures. In , LiBH4
was reported as a fast-ion conductor by Orimo et al., and this is regarded as a turning point in the development of hydride SSEs.73
The ionic conductivity dramatically jumped by 3 orders of magnitude from 10−7
( K) to 10−3
( K), accompanied by a structural transition from orthorhombic to hexagonal (Fig. 7a).
Furthermore, no polarization has been detected at the LiBH4
–Li metal interface due to good reduction stability and low grain boundary resistance. The interfacial stability was even tested under a high current density of mA cm−2
However, the fast 2D-Li+
conduction phenomenon could only occur in the high-temperature phase, which was a serious constraint for applying hydride SSEs in practical solid-state batteries. To improve the ionic conductivity of orthorhombic LiBH4
or prepare RT-stabilized hexagonal LiBH4
, great efforts have been made, such as second-phase compositions, anion doping, and interfacial modifications. Therefore, a series of CSSEs have been synthesized based on hydrides.71
|Fig. 7 (a) Structural transition of LiBH4 from low temperature to high temperature phases triggered by the reorientation of complex anions. Reproduced from ref. with permission from Springer Nature, copyright (b) Li+ ionic conductivity in complex hydrides. Reproduced from ref. 84 with permission from Springer Nature, copyright (c) Schematic diagram of the synthetic process of Li4(BH4)3I@SBA (d) Full cell test based on the Li4(BH4)3I@SBA composite electrolyte. Reproduced from ref. with permission from Wiley-VCH, copyright |
A larger ionic radius (such as I−, nm) can help increase the distance between alkali-metal ions and [BH4]− ions, which exhibit a low transition temperature.74 With efforts to obtain high conductivity at moderate temperature, the concept of partially replacing [BH4]− complex ions with I− ions was proposed and investigated. Among these conceptual materials, a LiBH4–LiI system was proven to be the most successful due to showing a high Li+ conductivity in the order of 10−5 S cm−1 for 3LiBH4–LiI at RT (Fig. 7b). No significant structural change was observed from differential scanning calorimetry (DSC) profiles, which revealed good structural stability and thermodynamic properties of the composite electrolyte. After a halide modification, the stability against Li was demonstrated by SEM and Li plating/stripping experiments. An all-solid-state Li–S battery with a 3LiBH4–LiI electrolyte exhibited high reversible capacity. Afterwards, researchers found that nanoconfined LiBH4 in the pores of ordered mesoporous silica scaffolds led to high Li+ conductivity due to the fast Li+ mobility at the interface between LiBH4 and SiO2.72,, The above result also led to a lower phase transition temperature than for bulk LiBH4. More recently, inspired by the success of hydride–halide composites and nanoconfinement through mesoporous materials, Li4(BH4)3I was nanoconfined into SBA via a two-step process by Zheng et al.(Fig. 7c and d). Li4(BH4)3I@SBA exhibited a high conductivity of × 10−4 S cm−1 at 35 °C with a Li-ion transference number of Furthermore, the formation of a stable SEI between Li and the electrolyte could effectively suppress the growth of Li dendrites. More significantly, the good compatibility of Li4(BH4)3I@SBA for ASSLBs was investigated with different cathodes, i.e., Li4Ti5O12, S and LiCoO2. On the other hand, the origin of dendrite growth in LiBH4 was clarified by Sun et al. in Li+ combines with electrons within the grain boundary/pore of the SSE, reducing to Li0 and eventually leading to a short circuit. Herein, the electronic conductivity of SSEs plays a significant role in dendrite formation. In this case, LiF was employed as an insulator for the LiBH4 SSE. The LiBH4–LiF CSSE exhibited amazing stability at mA cm−2 for over cycles, successfully inhibiting the growth of Li dendrites.
A low activation energy is also needed to obtain a superionic conductor. For example, the Ea of eV for Li4(BH4)3I@SBA was lower than that of eV for Li4(BH4)3I, which contributed to the high conductivity of the former. Thus, to weaken the electrostatic interaction between Li+ and [BH4]1−, sulfide materials were introduced into LiBH4 to form a hybrid system. A pseudo-binary system composed of a complex hydride and sulfide was reported by Orimo et al. in A new crystalline phase of 90LiBH4P2S5 with a possible orthorhombic structure was achieved with a high ionic conductivity of log(σ/S cm−1) = − at K. A lower Ea for the new structure was calculated to be eV compared with eV for HT-LiBH4. A smooth charge transfer between the SSE and TiS2 electrode was certified to occur through a charge/discharge process. Unlike single complex hydrides, the new composite electrolyte exhibited no phase transition before K and a wide electrochemical window of 0–5 V. Hydride-based CSSEs also experience stability issues as many candidates decompose on the cathode side. Many approaches have been developed to enhance the interfacial stability of CSSEs against cathodes, such as designing stable interfaces between the hydride SSEs and cathodes. To accommodate high-voltage cathodes, such as commercial LiCoO2, sulfides have recently been introduced to decrease the interfacial resistance between LiCoO2 and hydride-based SSEs by forming a stable Li-ion conductive cathode electrolyte interphase (CEI) layer., These hydride-based CSSEs can work well in all-solid-state LIBs with a LiCoO2 cathode, which exhibits typical high-voltage plateaus for charge/discharge.
Additionally, a novel SSE from a solid solution of two closo-type complex hydrides, videlicet Li(CB9H10)–Li(CB11H12), was reported with a high ionic conductivity of over 10−3 S cm−1 at K (Fig. 8a–e).84 High stability against lithium metal was certified by lithium plating/stripping for cycles with an extremely stable lithium-ion transfer capability at mA cm−2. Based on a cathode loading of mg cm−2, all-solid-state Li–S batteries (ASSLSBs) presented a good reversible capacity of mA h g−1 after 20 cycles at 1C ( K); the notable energy densities were calculated to be over W h kg−1 at high current densities (1–3C). Moreover, based on the SEM observation it was found that dendrite growth was suppressed at the interface between Li(CB9H10)–Li(CB11H12) and Li metal. Overall, these CSSE-based SSLSBs exhibited outstanding electrochemical performance at K. Alternatively, approaches to improve the conductivity of LiBH4-based composites in regard to inducing defects and changing the atomic arrangement have been attempted. To decrease the volume density of Li ions in LiBH4, neutral molecules were brought into LiBH4. This new concept was first achieved by the utilization of ammonia absorption into LiBH4, leading to a structural transition and reducing the activation energy of Li-ion mobility (Fig. 8f).76 Li(NH3)nBH4 (0 < n ≤ 2) exhibited high ionic conductivity ( × 10−3 S cm−1). A drastic increase in ionic conductivity occurred at approximately 37 °C, which is close to human body temperature, which shows its potential to be utilized in wearable devices. More recently, the Li+-conduction mechanism in LiBH4·1/2(NH3) was systematically investigated through crystal structure analysis and DFT calculations by Jensen et al. The molecular volume of NH3BH3 (v = Å3 per unit cell) was much larger than that of NH3 (v = Å3 per unit cell), which would intrinsically increase the cell volume of LiBH4 and lower the volume density of Li ions (Fig. 8g). Remarkably, a novel RT ultrafast CSSE (LiBH4·NH3BH3) was demonstrated with a conductivity up to × 10−4 S cm−1 at K.
|Fig. 8 (a) Discharge–charge profiles at K. (b) Stability of the composite electrolyte with lithium metal. (c) Cycling performance of the closo-type complex hydride-based Li–S battery. (d) FE-SEM image of the electrolyte/Li interface. (e) Ionic conductivity properties. Reproduced from ref. 84 with permission from Springer Nature, copyright (f and g) Crystal structures of Li(NH3)BH4, LiBH4, (LiBH4)2AB, and LiBH4AB. Reproduced from ref. 76 and with permission from Elsevier Inc., copyright and permission from American Chemical Society, copyright |
In short, the formation of composites with other materials can effectively promote high ionic conductivity in LiBH4. Additionally, the original shortcomings of hydride-based SSEs, such as the ionic area specific resistance, thermal stability of composites, and oxidation stability with cathodes, have been overcome accordingly. Inspired by these studies, other nano/composites have been investigated, e.g., × 10−5 S cm−1 for the LiBH4–C60 composite at K. There are also several reports on ionic conductivities in hydride–nitride/imide systems, including but not limited to a LiBH4–Li3N system and a LiBH4–Li2NH system.
By replacing oxygen ions with sulfur ions, owing to their lower electronegativity and larger radius, sulfide SSEs present a weaker bonding strength between the sulfur and lithium ions and provide a wider migration tunnel for lithium ions. As a result, sulfide SSEs exhibit higher ionic conductivities, approximately 10−3
at RT, which are almost comparable to those of liquid electrolytes. In , Li2
was first synthesized by twin roller quenching.
Then, through further doping with a Li3
electrolyte, the conductivity increased to × 10−3
In , a thio-lithium superionic conductor (thio-LISICON) Li10
was first reported to have a high ionic conductivity of × 10−2
at RT (Fig. 9a), which greatly promoted the development of sulfide SSEs.16
To improve conductivity, the same group discovered a novel lithium superionic conductor, Li
, based on a double substitution with aliovalent-ion doping (Fig. 9b).10
An exceptionally high conductivity of × 10−2
|Fig. 9 (a) Framework structure of Li10GeP2S12 and conduction pathways of lithium ions. Reproduced from ref. 13 with permission from Nature Publishing Group, copyright (b) Crystal structure of LiSiPSCl and nuclear distributions of Li atoms. Reproduced from ref. 7 with permission from Nature Publishing Group, copyright (c) Schematic of the formation of the solid electrolyte in a polymer matrix membrane. Reproduced from ref. with permission from Wiley-VCH, copyright (d) Schematic illustration of the preparation process of the LPS–LLZO CSSE. (e) The ionic conductivity properties of the LPS:LLZO mixture. (f) TEM image of the LLZO–LPS composite electrolyte (the core–shell structure) and the EELS map show a higher Li concentration across the LLZO–LPS interface (the bright part has a high concentration of Li). Reproduced from ref. with permission from the Royal Society of Chemistry, copyright |
However, under current laboratory conditions, the cell-based energy density of ASSLBs is much lower than that of their competitors with liquid electrolytes. Because of the brittleness of electrolyte materials, the thickness of the electrolyte layer is often more than 1 mm to avoid the formation of cracks during high pressure stress. Synthesizing a sulfide/polymer composite electrolyte is an effective method to slim the electrolyte layer and maintain a high ionic conductivity. Kanno et al. prepared a thio-LISICON/silicone composite electrolyte sheet. Recently, an increasing number of compliant sulfide–polymer composite electrolytes have been successfully developed. Whiteley et al. produced a 64 μm-thick membrane with a sulfide loading of 77 wt% that exhibited a low shear modulus and high ionic conductivity of 10−4 S cm−1 at RT (Fig. 9c). An ultrathin solid-state membrane ( μm) was fabricated based on Li2S–P2S5 with a self-healing polymer matrix. It was applied as a separator (80 wt%) in ASSLBs with an FeS2-based cathode and achieved excellent rate capability and stable cycling for over cycles. More recently, Nan et al. prepared a free-standing composite solid electrolyte membrane with 78Li2S–22P2S5 concentrations from 80 to 97 wt%. The sulfide/PEO and sulfide/PVDF composites showed ionic conductivities of 4–7 × 10−4 S cm−1 with a thickness of μm.58 A moderate PEO content could help the composite electrolyte achieve good mechanical properties and a stable electrolyte/lithium interface.,– For example, with 5 wt% PEO, the proportional limit of Li6PS5Cl composite solid electrolytes was enhanced to a value of 60 MPa. The as-assembled cell exhibited a good capacity retention rate of 91% over cycles at C and K. Further characterization indicated that lithium dendrite growth could be effectively inhibited after the PEO modification. The formation of P–O–C bonds between the sulfide glass and oligomers was the key factor to ensure a high conductivity. It was also noted that the addition of small amounts of polymers improved ion conduction by lowering the glass transition temperature.
In addition to combining with organic polymers, inorganic materials can act as fillers in sulfide-based electrolytes. Hood et al. reported the effect of oxide fillers in composites with β-Li3PS4 for the enhancement of the parent electrolyte. For example, 2 wt% Al2O3 increased the ionic conductivity of the parent electrolyte to × 10−4 S cm−1 while maintaining electrochemical stability against metallic lithium up to 5 V. Similarly, the same group examined the composite electrolyte of a “hard” oxide (Li7La3Zr2O12, LLZO) and a “soft” sulfide (β-Li3PS4, LPS), which demonstrated an excellent conductivity of × 10−4 S cm−1 at K; the above conductivity was higher than that of its parent electrolytes (Fig. 9d and e). The improved conductivity was considered an effect of the space charge layer at the interface between the LLZO and LPS particles, which was believed to redistribute the ionic and electronic point defects (Fig. 9f). As both sulfide and hydride SSEs are very sensitive to humid air, the handling process is always conducted in a glove box in an inert gas atmosphere. Due to the similarity of the chemical stability of these two types of electrolytes, a composite of hydride (LiBH4) and sulfide (Li3PS4) showed the potential to improve the total lithium-ion conductivity. Tatsumisago et al. examined the effects of the addition of LiBH4 on the structures and properties of sulfide electrolytes. The conductivities of the composite SSEs increased with increasing LiBH4 content. The CSSEs had a wide electrochemical window up to 5 V vs. Li+/Li−. The glass with a composition of 2Li3PS4·LiBH4 showed the highest conductivity of × 10−3 S cm−1 at RT.
To realize the application of ASSLBs at RT, current research efforts focus mostly on ionic conductivity and a wide electrochemical stability window. Compared to other SSEs, halide SSEs have had a relatively delayed development before because of their low ionic conductivity and low oxidation voltage.,
With the tireless efforts of countless researchers, a large breakthrough in halide SSEs occurred in 68
Tetsuya Asano et al. successfully synthesized Li3
with high ionic conductivities of – × 10−3
Afterwards, other halide SSEs with high ionic conductivity, such as Li3
(ref. 70 and ) and Li3−x
(M = Y, Er),
were also successfully fabricated. In addition, recent experimental and theoretical results have further demonstrated that halide SSEs are quite promising owing to their wide electrochemical windows, good electrode stability, high humidity tolerance, and simple production processes. Halide SSEs and their application in ASSLBs are relentlessly advancing. These new developments make it necessary to revisit halide SSEs for potential applications in ASSLBs. Among halide SSEs, Li3
(M = Sc, Y, Ho, Er, X = Cl, Br)-type SSEs have received wide attention. However, there is still a large gap between their experimental and theoretical results. To date, only a few halide SSEs, such as Li3
, have achieved a high ionic conductivity over 10−3
at RT. In addition, chloride-based SSEs showed an oxidation onset voltage of approximately 4 V, which cannot fully meet the electrolyte needs of high-voltage cathodes. Therefore, a tremendous amount of work is urgently required to improve ionic conductivity and optimize the electrochemical stability window in chloride-based SSEs.
The Li3YBr6, Li3InCl6, and Li3InBr6 SSEs, which possess a cubic close-packed (ccp)-like anion arrangement, display relatively high ionic conductivities of 1–2 × 10−3 S cm.67,68,, Furthermore, theoretical calculations demonstrated that halide SSEs with ccp anion sublattices could display high ionic conductivities. Inspired by the possibility of achieving fast Li+ migration in ccp halide SSEs, new routes have opened in the development of ASSLBs. Recently, LixScCl3+x SSEs (x = , 3, , and 4) were synthesized by a simple co-melting strategy from LiCl and ScCl3. The structural evolution and ionic diffusion mechanisms in LixScCl3+x were also systematically explored, and it was found that the vacancy concentration has the opposite trend with increasing x in LixScCl3+x (Fig. 10). The Li+ probability density and migration pathways were fitted based on AIMD simulations (Fig. 10a–c). The site occupations of metal/vacancies favoured Li+ migration within the local structure (Fig. 10e–g). As a result, the obtained Li3ScCl6 showed a high ionic conductivity of 3 × 10−3 S cm−1. Moreover, the all-solid-state LiCoO2/Li3ScCl6/In full cell exhibited a long cycle life and a wide electrochemical window of – V. Note that Li3ScCl6 was not stable towards Li in the initial cycles of plating/stripping. To enhance the ionic conductivity, the covalent substitution of metal ions was also an effective strategy to introduce vacancies in the mobile ion sublattice. Nazar et al. prepared a class of mixed-metal chloride solid-state electrolytes, Li3−xM1−xZrxCl6 (M = Y, Er) SSEs. These new halide SSEs exhibited high ionic conductivities (up to mS cm−1 at RT) due to their unique new structures (Fig. 10h and i). In Fig. 10j, combining neutron and single-crystal X-ray diffraction methods, the evolution of new structures after a Zr substitution revealed trigonal to orthorhombic phase transition processes. Most importantly, chloride SSEs without any protective coating showed excellent oxidation stability. When using unprotected LiCoO2 as the cathode material, ASSLBs exhibited exceptional electrochemical oxidation stability up to V at RT.
|Fig. 10 (a and b) The Li+ probability density based on AIMD simulations. (c) The Li+ migration pathways of the Li3ScCl6 structure. (d) The Li+ probability density marked by yellow isosurfaces of Li5ScCl8 (x = 5), Li3ScCl6 (x = 3), LiScCl (x = ) and LiScCl4 (x = 1) structures along the a axis. (e) The blocking effect of Sc. (f) Radial distribution function (rdf) of Sc–Li ions in the LixScCl3+x (x = 1, , 3, and 5) SSEs. (g) Arrhenius plot of Li+ diffusivity in LixScCl3+x (x = 1, , 3, and 5) from AIMD simulations. Reproduced from ref. with permission from the American Chemical Society, copyright (h) The crystal structure of LiErZrCl. (i) Li connectivity along the  direction. (j) Phase evolution of Li3M1−xZrxCl6 (M = Er, Y) upon Zr substitution. Reproduced from ref. with permission from the American Chemical Society, copyright |
Remarkably, halide SSEs based on close-packed anion arrangements or covalent substitution of metal ions exhibited a high ionic conductivity. However, the chemical and electrochemical stability was difficult to satisfy with both high oxidation and low reduction voltages in full battery applications. One of the possible solutions is to combine halide SSEs and other SSEs to be compatible with cathode and anode materials, respectively. Unlike chloride SSEs, fluoride SSEs exhibit a wide electrochemical stability window (∼6 V). However, they have the lowest ionic conductivity among halide SSEs. To solve this problem, tuning the chemical composition of lithium difluoro(oxalato)borate (LiDFOB) and PEO can heighten the stability and improve the ionic conductivity in CSSEs,, which undoubtedly proves the possibility of modifying CSSEs with other types of SSEs. Therefore, the lower performance compared to other SSEs can be enhanced by various physical approaches and chemical strategies. It is predicted that halide SSEs will become more deserving of in-depth attention, and their commercialization in ASSLBs will be realized in the near future.
Mobile ion migration mechanism in CSSEs
A comprehensive summary of different CSSEs, including polymer-based, oxide-based, hydride-based, sulfide-based and halide-based CSSEs, is provided in Table 1. For practical application in ASSBs, in addition to the ionic conductivity, electrochemical stability window, chemical compatibility, and mechanical properties, other properties such as thermal stability, fabrication processes, cost, device integration and environmental friendliness are also important.11,,
In , Manthiram et al. summarized the properties of the existing solid electrolyte materials and visualized those properties in radar plots.11
It is clear that single solid-state electrolytes have difficulties satisfying the increasing demands in our daily life. Even though significant progress has been achieved, huge challenges still remained for CSSEs to seek an equilibrium relationship, which minimizes the weaknesses of individual SSEs to enhance the overall performance. Therefore, it is important to summarize previous achievements in order to understand, design, and fabricate novel CSSEs. Fig. 11 briefly describes the migration mechanism of lithium ions in different kinds of composite solid-state electrolytes. For polymer-based electrolytes, the diffusion in the polymeric host is based on ether linkages. However, with the increase of inorganic fillers, the ion transport gradually transits to the newly formed interphase between the dispersed crystalline and polymer matrix (Fig. 11a).
It is noticed that many solid-state electrolytes turn out to be superionic phases only at high temperature, such as hydride-based electrolytes (Fig. 2b).