Effective Approaches of Improving the Performance of Chalcogenide Solid Electrolytes for All-Solid-State Sodium-Ion Batteries

All-solid-state sodium-ion batteries (SIBs) possess the advantages of rich resources, low price, and high security, which are one of the best alternatives for large-scale energy storage systems in the future. Also, the chalcogenide solid electrolytes (CSEs) of SIBs have the characteristics of excellent room-temperature ionic conductivity (10−3-10−2 S cm−1), low activation energy (<0.6 eV), easy cold-pressing consolidation, etc. Hence, CSEs have become a very active area of all-solid-state SIB research in recent years. In this review, the modification methods and implementation technologies of CSEs are summarized, and the structure and electrochemical performance of the CSEs are discussed. Furthermore, the auxiliary function of first-principle calculations for modification is introduced. Ultimately, we describe the challenges regarding CSEs and propose some strategic suggestions.


INTRODUCTION
Recently, due to the rapidly amplified demand for electric vehicles and power storage systems based on renewable energy, the development of large-scale energy storage systems (ESSs) has aroused the attention of the whole world (Zhong et al., 2014;Dai et al., 2020;Yao et al., 2020). Considering safety and cost of large-scale ESSs, all-solid-state sodium-ion batteries (SIBs) could be an alternative to all-solid-state lithium ion batteries (LIB) (Luo et al., 2016). Additionally, compared with liquid electrolytes, solid electrolytes possess the leak-free state, wide electrochemical windows, and high thermal stability over a broad temperature range (Wang et al., 2019a). Therefore, the development of solid electrolytes is a very meaningful task.
Since the electronegativity of sulfur is lower than that of oxygen, sulfide electrolytes usually exhibit faster sodium ion conduction than oxide electrolytes because the electrostatic force of sulfur and sodium ion is less than that of oxygen and sodium ion. Compared with oxide solid electrolyte, chalcogenide solid electrolytes (CSEs) could be synthesized at low temperature, which reduces the production cost. Due to the deformability of CSEs, it can achieve good interface contact with electrode interface via simple cold pressing to further reduce the actual production cost. Nay, CSEs have high RT ionic conductivity and low activation energy. Among all kinds of solid electrolytes, CSEs with good formability are a good choice for the preparation of bulk-type all-solid-state SIBs. Therefore, CSEs have an important application prospect in the field of bulk-type all-solid-state SIBs for large-scale energy storage. Despite these advantages, there are still some challenges before CSEs are widely used. Initially, the composite electrodes of bulk-type all-solid-state SIBs are in poor contact. Although CSEs have deformability, investigation results demonstrate cold pressing alone is not enough to form compact components without voids. Secondly, CSEs are unstable in the air, which are easy to deliquesce, oxidize, and generate toxic gases, resulting in a sharp drop of conductivity. Therefore, it is very important to improve the performance of CSEs in bulk-type all-solid-state SIBs.
This review will introduce the CSEs from the following aspects: the modification methods and the implementation technologies, the structure and electrochemical performance, and the auxiliary function of FPCs for modification. Ultimately, we describe the challenges of the CSEs and propose some strategic suggestions.

IMPROVING APPROACHES OF THE CSEs
For the CSEs, especially chalcogenide glass-ceramic electrolytes, they not only have high room-temperature ionic conductivity (10 −3 > σ >10 −5 S cm −1 ) but also have a wide electrochemical window (V EW , 6 V> V EW > 4 V) and suitable mechanical properties (Hayashi et al., 2012;Kim et al., 2014;Tanibata et al., 2014a). In recent years, to more effectively enhance the electrochemical properties and mechanical properties of the CSEs, the following effective methods are used in the developments: ion doping and substitution, composite method, crystal transformation method, ceramization, and vitrification, etc. In this section, the effect of modification of different methods on the CSEs is compared vertically, mainly from the aspects of electrochemical window, ionic conductivity, and electrical stability as well as structural stability.

Ion Doping and Substitution
Ion doping and substitution are common methods of material modification (Bo et al., 2016;Kato et al., 2016;Richards et al., 2016;Krauskopf et al., 2017;Rao et al., 2017;Wang et al., 2017;Mahmoud et al., 2019;Wan et al., 2020;Zhang et al., 2020). Cation ion doping into the P site or halogen doping into the S site can introduce vacancies to improve the ionic conductivity of the CSEs (Zhang et al., 2015;Bo et al., 2016;Wang et al., 2016;Krauskopf et al., 2017Krauskopf et al., , 2018aYu et al., 2017). The bond energy between ions and adjacent atoms could be changed by ion doping and substitution to improve the ionic conductivity of the electrolyte (Krauskopf et al., 2018a). Furthermore, the diffusion activation energy of ions can be further changed by this method (Bo et al., 2016;Krauskopf et al., 2017). Additionally, ion doping and substitution not only introduce vacancy but also play an important role in micro diffusion (Yu et al., 2017;Krauskopf et al., 2018b). After doping and substitution, the free volume caused by phonon is large, and the transition probability of sodium ion from one site to the adjacent site is increased. For example, the ionic conductivity of Sb 5+ instead of P 5+ or Se 2− instead of S 2− in Na 3 PS 4 has been improved markedly (Zhang et al., 2015;Bo et al., 2016;Wang et al., 2016). Similarly, Na 3 P 0.62 As 0.38 S 4 with the replacement of P 5+ to As 5+ was developed with an exceptionally high room-temperature ionic conductivity of 1.46 × 10 −3 S cm −1 and enhanced moisture stability (Yu et al., 2017). Peradventure, Cl −1 instead of S 2− in Na 3 PS 4 was obtained by a liquid-phase (suspension) method, Here, we choose one of the most representative examples to exhibit the effect of ion doping and substitution at the same time. The cubic Na 3 PS 4−x Se x (c-Na 3 PS 4−x Se x ) materials were developed by the mechanochemical (BM) approach and high temperature (HT), respectively (Krauskopf et al., 2017). The FPCs illustrate that the replacement S with Se could not only expand the diffusion path but also enhance the ionic conductivity. Recently, Yao et al. synthesized the Na 3 SbS 3.75 Se 0.25 solid electrolyte with less grain-boundary resistance by a liquid/solid fusion technology, as shown in Figures 1a-g. The Na 3 SbS 3.75 Se 0.25 solid electrolyte shows a high RT ionic conductivity of 4.03 × 10 −3 S cm −1 owing to the significantly decreased amorphous phase in the electrolyte, and the small particle size of the solid electrolytes enhances the contact between solid electrolyte and electrode, reducing the interfacial contact resistance (Wan et al., 2019).
Additionally, Krauskopf et al. had investigated the lattice dynamics of the superionic conductor Na 3 PS 4−x Se x by probing the optical phonon modes and the acoustic phonon modes, as well as the phonon density of states via inelastic neutron scattering, as shown in Figures 1h-j (Krauskopf et al., 2018b). The experimental results and FPCs have proved the influence of lattice dynamics and the softness of lattice to ion transport. When studying the change of the phonon spectrum or the density of state of the total phonon, they observed the activation barrier of migration and the decrease in migration entropy and emphasized that the paradigm of "the softer the lattice, the better" is not always true. Additionally, ion doping and substitution can also improve the specific capacity of the battery. Although this change enhances the ionic conductivity and charge/discharge-specific capacity, it does not significantly change the electrochemical window of the electrolyte. In addition, owing to the excessive embedment of sodium ions in the electrolyte, this method will lead to the destruction of the electrolyte structure.

Composite Method
In terms of inorganic-organic composite, due to the strong ability of dissolving sodium salt in the amorphous region of polyethylene oxide (PEO), the PEO-based composite solid electrolyte is considered as an ideal solid electrolyte. The addition of inorganic materials reduces the crystallinity of the polymer matrix and improves the activity of the polymer chain. The addition of inorganic compounds to the polymer acts as a plasticizer, which is conducive to the transition of the polymer to an amorphous structure that is easy to transport ions. Moreover, the existence of active crystal and polymer can make the composite solid electrolyte have relatively high ionic conductivity, good flexibility, stability, and interface compatibility with sodium anode.
Recently, a typical inorganic-organic composite CSE of Na 3 PS 4 -PEO for all-solid-state SIB was obtained by the solution-phase reaction method with room-temperature ionic conductivity of 9.4 × 10 −5 S cm −1 and electrochemical window of ∼4.5 V . Moreover, the PEO coating on the surface of Na 3 PS 4 particles effectively hinders the side reaction between electrolyte and sodium metal electrode. After 40 cycles, the specific capacity of all-solid-state SIB is 230 mAh g −1 . Additionally, the Na 3−x PS 4−x Cl x -PEO-NaClO 4 solid electrolyte with room-temperature ionic conductivity of 8.75 × 10 −4 S cm −1 was obtained by the heat treatment (Wang et al., 2019b).
However, inorganic-inorganic composite materials are rare, mainly related to glass ceramic electrolytes. For example, Na 3 Zr 2 Si 2 PO 12 -Na 3 PS 4 glass ceramics exhibited high total conductivities of 1.1 × 10 −3 at 100 • C; the results suggest that the composite materials make the contact between particles closer, which is conducive to the improvement of ionic conductivity (Noi et al., 2018). Similarly, for the sulfide-based solid electrolyte Na 3 SbS 4 -Na 2 WS 4 with RT ionic conductivity of 4.28 mS cm −1 prepared by the liquid phase method, they point out that the increase in ionic conductivity is caused by sodium vacancies (Yubuchi et al., 2020).

Crystal Transformation Method
High-temperature heat treatment, ion doping, and substitution are widely applied in crystal transformation. The essence of crystal phase transition is that the crystal structure changes at a temperature higher than its phase transition temperature, while the reason for the phase transition caused by doping and substitution is that its process includes high-temperature treatment steps. The experimental results indicate that the tetragonal structure is stable at low temperature, while the cubic structure is more stable at high temperature (Hayashi et al., 2012). As we know, Na 3 PS 4 crystallizes in one of two different polymorphs (i.e., cubic or tetragonal). FPCs of the cubic and tetragonal Na 3 PS 4 (c-Na 3 PS 4 and t-Na 3 PS 4 ) are illustrated in Figure 1 (Zhao et al., 2018).
The results reveal that both c-Na 3 PS 4 and t-Na 3 PS 4 are highly conductive solid electrolytes (De Klerk and Wagemaker, 2016). Although t-Na 3 PS 4 exhibits higher ionic conductivity than c-Na 3 PS 4 (Zhao et al., 2018), the ionic conductivity of c-Na 3 PS 4 formed by ion doping and substitution method is higher than that of t-Na 3 PS 4 (Zhu et al., 2015). Moreover, for the Na 3 PS 4−x Se x materials, it is found that ion doping could be adopted to modify the crystal type due to its process which includes hightemperature treatment steps. Analogously, although RT ionic conductivity of tetragonal Na 3 SbS 4 is high at 3 mS cm −1 , both experiments and theoretical calculations illustrate that the ionic conductivity of cubic Na 3 SbS 4 is higher than that of tetragonal Na 3 SbS 4 (Zhang et al., 2016(Zhang et al., , 2018.

Ceramization and Vitrification
Ceramization and vitrification is also a common approach to modify the solid electrolyte (Tanibata et al., 2014b;Park et al., 2018;Tsuji et al., 2018;Ma et al., 2020). Vitrification is the process of transforming a substance into the glassy state, in which there is no crystal structure. However, there are crystal phases in the ceramics after ceramization. Glass ceramic is a mixture of crystalline phase and residual glass phase, which is formed by crystallization of glass under the action of catalyst or nucleating agent. Ceramization and vitrification could be mainly used to enhance the stability of electrolyte to better resist the effects of harsh environments and simplify the battery packaging process.

CONCLUSIONS AND PERSPECTIVE
In conclusion, ion doping and substitution can raise ionic conductivity of the CSEs. However, its effect on the structure stability needs to be further analyzed. Generally, the FPCs can be used to calculate the phonon spectrum, binding energy, and infrared spectrum of the modified materials, so as to determine the structural stability of the material, and then the experiments of ion doping and substitution could be carried out. The composite method can be widely used to reinforce interface contact. In addition, ceramization and vitrification are also common methods to modify the CSEs, which can improve the ionic conductivity and stability and better resist the influence of a harsh environment. In short, for different electrolytes, it is still necessary to choose appropriate modification methods according to the actual requirement, so as to better augment the performance of electrolytes.
Ultimately, the development of solid-state electrolytes of chalcogenides still faces some challenges, which are summarized as follows: (1) The CSEs are sensitive to air. In a high-humidity environment, they are easy to deliquesce and release toxic H 2 S gas, and they are easy to be oxidized in dry air. Hence, it brings difficulties to the packaging and application of the CSEs.
(2) S or Se in the CSEs is easy to exchange anion with O in oxide anode, and P/Sb/Sn/Ge is easy to be reduced by Na; hence, the interface stability between the electrolyte and the electrode needs to be enhanced.
(3) Although the conductivity of Na 10 GeP 2 S 12 and Na 7 P 3 M 11 (M = O, S, Se) electrolytes is high, it is difficult to synthesize them in the pure phase.
In terms of the above problems, we put forward some suggestions as follows: (1) The theoretical calculation of element doping and surface coating is carried out by the FPCs to improve the efficiency of the experiments.
(2) Coating the CSEs with an oxide-protective layer or compounding with polymer enhances air stability.
(3) The core-shell structure is designed to limit the decomposition of sulfide during volume expansion.
(4) Combined with the phase diagram, the cooling rate is precisely controlled in the process of melting quenching to obtain the electrolyte with excellent performance.

AUTHOR CONTRIBUTIONS
HD conceived the idea and designed this review. WX collected the related literature resources (with assistance from ZH, YC, XW, BY, ZC, JG, and DY). HD and WX contributed to data analysis and interpretation. WW (with assistance from FX, WZ, RG, and GZ) revised the manuscript. HD wrote the paper. All authors provided feedback.