Challenges and Improvements of All-Solid-State Batteries

Lithium-ion batteries (LIBs) have become a crucial component of modern technologies and gained widespread popularity in recent years due to their high energy density and power density, long cycle-life performance, and low self-discharge.^1,2 They are used in a wide range of applications, including portable devices, electric vehicles (EVs), and energy storage systems (ESSs). The research trend in LIBs is focused on improving their energy density, reducing cost, and increasing their safety. The quest for higher energy density has led to the development of new cathode/anode materials, while efforts to reduce costs have resulted in the replacement of rare materials with earth-abundant materials as well as the improvement/optimization of the processes for the material synthesis, electrode fabrications, and battery production. The importance of ensuring the safety of LIBs has also led to the development of new designs that prevent thermal runaways and other safety issues.³ One of the promising alternative systems for safe batteries is all-solid-state batteries (ASSBs), which replace liquid or gel-like electrolytes with solid-state electrolytes (Fig. 1a).⁴ It offers several advantages including the safety improvement of the battery as it eliminates the risk of environmentally-polluting liquid leakage and reduces the risk of thermal runaway. Additionally, ASSBs have a higher energy density (Fig. 1b) and can operate at higher temperatures, which makes them ideal for use in demanding applications such as EVs and ESSs.⁵ In this mini-review, we will break down each component of ASSB as the solid-state electrolyte, cathode composite, and anode, and focus on their current challenges and how those have been addressed and the performance has been eventually improved.

Figure1.

(a) Change in energy density upon replacing liquid electrolytes in conventional LIB with solid-state electrolytes (ASSBs) and Li metal anode (all-solid-state Li metal batteries). The volumetric and gravimetric energy densities are represented by E_vol and E_grav, respectively. Reproduced with permission from ref 4. Copyright 2020 Wiley-VCH GmbH. (b) Calculated volumetric and gravimetric energy densities of ASSBs as a function of cell parameters. Reproduced with permission from ref 5. Copyright 2022 Elsevier Inc.

Solid-state electrolytes are the most important key component in terms of the realization of ASSB, where a lot of research and development is being conducted for the stable energy storage and conversion system. The ignition of the battery can occur when it is exposed to an abnormal environment including unexpected overcharge and exposure to the extremely high temperature. In this case, the organic liquid electrolyte can be used as fuel for this ignition and thermal runaway. By replacing this organic liquid electrolyte with a solid-state electrolyte, the safety of the battery could be dramatically improved. However, in order to use a non-flammable solid material as an electrolyte material, it must have sufficient lithium-ion conductivity to perform its role as an electrolyte. The development of solid-state electrolyte which has ionic conductivity corresponding to liquid electrolyte enables ASSB to be considered at a commercial level.^6,7 Various solid-state electrolytes with conductivities of 1-10 mS cm⁻¹ have been reported (Fig. 2a), not far behind organic liquid electrolytes with ionic conductivities of approximately 10 mS cm⁻¹.⁸ In order to work as an electrolyte in a battery, not only ionic conductivity but also electrochemical stability at the working voltage range of the anode and cathode must be satisfied. As shown in Fig. 2b, it plays a role as an electrolyte stably in the voltage or chemical potential range where the anode and cathode operate, or it must have the ability to create a stable passivation layer at the interface of the cathode or anode.⁹ Solid-state electrolytes that have been developed are mainly oxides, sulfides, and halides. The electrochemical stability windows of those electrolytes are calculated through the computational simulation, which shows that they have various ranges depending on the type of material and the core elements (Fig. 2c).¹⁰ As comprehensively examining parameters such as ion conductivity and thermal stability as well as redox stability of these electrolytes, each type has distinct advantages and disadvantages, so it can be applied in various ways depending on the purpose of the battery (Fig. 2d).¹¹ Among them, the solid-state electrolyte that is currently used the most in many research institutes and industries and is most likely to be applied to commercial batteries is the sulfide-based Li₆PS₅Cl (LPSCl) argyrodite series whose crystal structure is shown in Fig. 2e.¹² Although its electrochemical redox stability is not outstanding compared to other solid-state electrolytes, it has moderate ionic conductivity (1-3 mS cm⁻¹) and has physical properties that can make good physical contact with the cathode or anode owing to the softness or ductile property of sulfide material. In addition, it shows the excellent passivation ability of interphase (cathode electrolyte interphase (CEI) at the surface of the cathode and solid electrolyte interphase (SEI) at the surface of the anode), which enables the long cycle-life performance of the battery.^13,14 With these compromising advantages, it is believed that LPSCl will be the solid-state electrolyte for the first commercial ASSB.

Figure2.

(a) Arrhenius of solid-state electrolytes compared to organic liquid electrolytes. Reproduced with permission from ref 8. Copyright 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. (b) Schematic diagram of the electrochemical window (color bars) and the Li chemical potential profile (black line) in the ASSBs. Reproduced with permission from ref 9. Copyright 2015 American Chemical Society. (c) Electrochemical stability ranges of various electrolyte materials grouped by anion, with corresponding binary for comparison. Reproduced with permission from ref 10. Copyright 2016 American Chemical Society. (d) Comparison of various solid-state electrolyte compounds regarding the requirements for ASSB applications. Reproduced with permission from ref 11. Copyright 2021 Elsevier Ltd. (e) Crystal structure of Li₆PS₅X. In the ordered structure, X⁻ anions form a face-centered cubic lattice with the free S²⁻ in the tetrahedral sites and PS₄³⁻ in the octahedral voids. Reproduced with permission from ref 12. Copyright 2018 Royal Society of Chemistry.

The redox reaction of LPSCl has been investigated in various previous studies. Elemental S and P₂S₅ are formed in the oxidation process, while the reduction reaction generates Li₂S and Li₃P, which are the components of stable interphases on the surface of the cathode and anode active materials, respectively (Fig. 3a).¹⁵ It is one of the main factors contributing to stable charge/discharge cycling characteristics. The electrochemical decomposition reaction of LPSCl is greatly influenced by the type of conductive carbon material used for the cathode composite. In particular, when carbon black with a very large surface area is used as shown in Fig. 3b, the side reaction becomes severe. Therefore, a conductive agent that provides an electronic conduction pathway while having a relatively small surface area, such as vapor-grown carbon fiber (VGCF), must be used for the cathode composite of ASSB (Fig. 3b).¹⁵ LPSCl is usually synthesized by applying mechanical energy such as a ball-milling of stoichiometric amounts of Li₂S, P₂S₅, and LiCl precursors, and the following heat-treatment process makes the high crystalline LPSCl with higher ionic conductivity. If the particle size of LPSCl is too large, it is difficult to maintain good contact with the cathode active material in the cathode composite which results in high resistance and hinders the increase in the ratio of the cathode active material making it very difficult to achieve high energy density. Therefore, it is necessary to implement a particle size of several hundred nanometers to perform the role of an electrolyte with a small amount. As such, various post-processing ways have been developed to reduce the particle size of LPSCl (Fig. 3c).⁵ In order to use a solid-state electrolyte as a key component of ASSB, not only the ionic conductivity must be high, but both electrochemical redox stability and physical/mechanical properties must be optimized. The various physical properties of solid-state electrolyte candidates such as oxides and sulfides should be investigated, and the optimal electrolyte should be selected according to each purpose (Fig. 3d).⁴

Figure3.

(a) Cyclic voltammograms for the first two cycles; the voltage was swept between 0 and 4.2 V (vs. Li/Li⁺) at 100 μV s⁻¹, and first and second voltage profiles of the Li-In | LPSCl | LPSCl−C half-cell at a current density of 0.25 mA cm⁻². (b) SEM images, charge voltage profiles of Li-In | LPSCl | LPSCl−C, and charge/discharge voltage profiles of Li-In | LPSCl | NCM811 cells of carbon black and VGCF. Reproduced with permission from ref 15. Copyright 2019 American Chemical Society. (c) Schematic of single-step room temperature dry chemo-mechanical synthesis of LPSCl and post-synthesis conditioning steps to reduce particle size. Reproduced with permission from ref 5. Copyright 2022 Elsevier Inc. (d) Overview of transport, electrochemical, mechanical stabilities, and processing temperature of oxides and sulfide solid-state electrolytes. Reproduced with permission from ref 4. Copyright 2020 Wiley-VCH GmbH.

Within the configuration of a representative ASSB system, the cathode component consists of a composite form of cathode active material used in conventional LIB, solid-state electrolyte, and conductive carbon agent. It uses spherical lithium transition metal oxide active materials including high-Ni NCA and NCM which are similar to the conventional LIB cathode electrode, but the difference is that the liquid electrolyte wetted in the electrode is replaced with a solid-state electrolyte. Therefore, different from LIB whose processes are the fabrication of a porous electrode followed by the assembly of the cell and the injection of a liquid electrolyte so that the liquid is impregnated into the pores in the electrode, the solid-state electrolyte must exist from the electrode fabrication. As shown in Fig. 4a, the cross-sectional image of the cathode composite electrode shows the active material and solid-state electrolyte distributed evenly and well-contacted.^5,16 At this time, all parts that are in contact between the cathode active materials and the solid-state electrolytes are the interfaces where lithium ions move. Fig. 4b indicates that an electrochemical reaction initiates at the interface between the cathode transition metal oxide and the solid-state electrolyte. Since it is exposed to a highly oxidizing environment during the charging process, it is very likely for the solid-state electrolyte to oxidatively decompose, which makes it no longer function as an electrolyte and leads to a large increase in interfacial resistance.¹⁷ To prevent this formation of poor CEI and degradation, a similar coating strategy used for surface protection in LIB cathode active materials has also been used in ASSB. One difference here is that unlike the cathode coating layer used in conventional LIBs, the coating layer itself must also have somewhat lithium-ion conductivity. Therefore, as the ASSB cathode coating layer, oxides containing lithium ions have been developed such as Li₂ZrO₃ and LiNiO₃, and in particular, amorphous ones are mainly used because of relatively high lithium-ion conductivity.¹⁷ Fig. 4c shows the chemical compatibility with cathode active materials and electrochemical redox stability windows derived from the computational calculation of representative oxides considered as coating layers.¹⁸

Figure4.

(a) FIB-SEM cross-section image of dry-processed solid-state electrolyte separator laminated on a dry-processed cathode composite. Reproduced with permission from ref 5. Copyright 2022 Elsevier Inc. (b) False-color scanning electron microscopic image showing the contact loss between NCM and a sulfide electrolyte. Reproduced with permission from ref 16. Copyright 2019 WILEY?VCH Verlag GmbH & Co. KGaA, Weinheim. (c) Interfacial contact between uncoated cathode oxide and sulfide electrolyte. Thicker CEI at the surface of the charged cathode oxide. A thin conformal coating layer to prevent chemical reactions and mitigate electrochemical decomposition. Reproduced with permission from ref 17. Copyright 2020 American Chemical Society. (d) Properties of possible coating materials at the NCA/LPSCl interface. From left to right, reaction energies with pristine NCA (E_rxnNCA) and with LPSCl (E_rxnSE) in eV/atom, percentages of volume change (ΔV_NCA and ΔV_SE), diffusion channel radius (R_c) in Å, band gap (E_g) in eV, and energy above hull (E_hull) in eV/atom. The electrochemical windows of selected coatings. Reproduced with permission from ref 18. Copyright 2019 American Chemical Society.

Several studies have confirmed that highly advanced cathode active materials that showed excellent LIB performance show improved performance in the ASSB systems as well. Compared to transition metal oxides having a general secondary particle structure, transition metal oxide particles with a radially oriented primary particle is advantageous including a shorter lithium-ion pathway and reduced volume change which makes the contact between the cathode and solid-state electrolyte maintain (Fig. 5a).¹⁹ It results in lower resistance and thus improved cycle-life performance is achieved. In addition, by introducing a single crystal cathode material that does not have the secondary particle structure, the contact losses within the cathode particle and between the cathode and solid-state electrolyte are significantly reduced through the mitigation of volume change even after continuous charge and discharge (Fig. 5b).²⁰ It has been reported to exhibit excellent ASSB performance.

Figure5.

(a) Schematic illustrating the different microstructural and interfacial evolutions in two different cathode electrodes in all-solid-state cells. Reproduced with permission from ref 19. Copyright 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. (b) Schematic illustrating the different microstructural and interfacial evolutions in the NCA electrodes in all-solid-state cells, affected by the type of NCAs (single- vs. poly-crystalline) and solid-state electrolytes (halides (LYC) vs. sulfides (LPSX)). Reproduced with permission from ref 20. Copyright 2021 Wiley-VCH GmbH.

High Energy Density Anode Materials

One of the expected values of the ASSB system was that lithium metal with high energy density, which had not yet reached the commercial level in conventional LIBs, could be used as an anode material. Dendrite growth on the surface of lithium metal, which is one of the main obstacles in lithium metal anodes, and the resulting short-circuit in which the anode and cathode are electrically connected makes the battery no longer safe. Since the liquid electrolyte and mechanically weak separator are replaced with a solid layer, it is expected that this dendritic growth can be inhibited physically and thus does not occur. In the case of a solid-state electrolyte that is not electrochemically stable at the operating voltage of lithium metal, because of continuous reduction decomposition on the surface, the SEI becomes very thick and the resistance of the cell continuously increases, eventually making the cell unable to operate (Fig. 6a).²¹ As mentioned earlier, since LPSCl forms a stable SEI, there is no abnormal increase in resistance, enabling this metallic lithium/LPSCl combination to work. However, it has been reported that dendrite growth also exists in this ASSB system, causing a similar short-circuit phenomenon (Fig. 6a). It has been also confirmed that this dendrite growth behavior is significantly affected by the amount of pressure applied when charging and discharging the ASSB. As can be seen in Fig. 6b, when a pressure of around 75 MPa was applied, a short occurred after several tens of hours without lithium plating and stripping on the lithium metal anode.²² This is because soft lithium metals can easily penetrate/creep into the porous space between grains within the solid-state electrolyte layer. As such, this phenomenon has a high correlation with pressure. The symmetric cell (both working and counter electrodes composed of lithium metal), which is mainly used when evaluating the characteristics of lithium metal anode, experience much less pressure change than an actual full-cell (cathode composite working electrode and lithium metal counter electrode) during the charging and discharging (Fig. 6c).²³ Therefore, the short-circuit behavior can occur more easily in an actual full cell than in the case of reported symmetric cells in most studies. In addition to this dendrite growth issue of lithium metal, voids inside lithium metal generated during the continuous plating and stripping lead to contact loss and thus resistance increase, which is another cause of degradation of long-term performance (Fig. 6d).²⁴

Figure6.

(a) Metallic lithium anodes with Li₁₀SnP₂S₁₂ and Li₆PS₅Cl solid-state electrolytes. Reproduced with permission from ref 21. Copyright 2021 American Chemical Society. (b) Schematic of the effect of the stack pressure on the shorting behavior of lithium metal solid-state batteries. Reproduced with permission from ref 22. Copyright 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. (c) Schematic of the pressure change during cycling in a lithium metal symmetric cell and a full cell. Reproduced with permission from ref 23. Copyright 2022 Elsevier B.V. (d) Sequence from pristine to after several strippings and platings of lithium metal. Reproduced with permission from ref 24. Copyright 2019 Springer Nature Limited.

The creep and dendritic growth of lithium metal problems have been addressed with various strategies of high-energy-density anode materials and systems. As mentioned earlier, these drawbacks of lithium metal are very dependent on the cell pressure, so if the pressure change that occurs during the charging and discharging of the battery can be mitigated, the performance of the lithium metal anode will be improved. The specific current density value when a short-circuit occurs on a lithium metal anode is critical current density (CCD), and the CCD of a lithium metal anode is greatly improved with a cell configuration that can alleviate pressure changes during the charging and discharging of a battery (Fig. 7a).²³ In addition, a new type of anode called anode-free has also been proposed, which is one of the most creative ways to address the problem of lithium metal. As shown in Fig. 7b, when a thin layer composed of sliver/carbon is introduced between the anode current collector and the solid-state electrolyte layer, it can be seen that a very dense lithium layer is formed rather than dendrite-shaped lithium plating during lithium metal deposition.²⁵ This significantly improved the performance of ASSBs with lithium metal anodes and, as the name of the concept implies, the energy density could be further enhanced since it is a system without a real heavy anode. In another research, pure silicon anode revealed that it can show extremely stable performance in ASSB with LPSCl solid-state electrolyte (Fig. 7c).²⁶ Conventional LIBs cannot use almost 100% silicon with anode material since it generates SEI continuously with liquid electrolyte due to the large volume changes during cycling, resulting in poor performance caused by unlimited loss of active lithium ions and high impedance. However, since the pure silicon proposed in this study does not have a solid-state electrolyte inside the anode, the electrolyte decomposition reaction and the resulting SEI formation occur only in the apparent 2-dimensional area where each layer is in contact, so this amount is very small. In addition, the lithiated silicon generated at the interface also plays the role of lithium-ion transport, leading to a system in which the entire anode can participate in the charge (lithiation) reaction. This allowed us to significantly increase the performance of ASSB. As such, the development of anode materials and systems to realize high energy density ASSB has progressed considerably, and the remaining task is to study how to implement it in a large-scale cell.

Figure7.

(a) The schematic of the cell cycling setups for fixed gap and constant pressure. The NCM811 loading of both cells was 25.5 mg cm⁻². The operando pressure monitoring and corresponding voltage profiles of NCM811 | LPSCl | Li metal cells with fixed gap and constant pressure setup at ramping current densities. Reproduced with permission from ref 23. Copyright 2022 Elsevier B.V. (b) Schematic of Li plating–stripping on the current collector with an Ag–C nanocomposite layer during charging and discharging processes and cross-sectional SEM images for the Ag–C nanocomposite layer. Reproduced with permission from ref 25. Copyright 2020 Springer Nature Limited. (c) Schematic of 99.9 wt.% micro-Si electrode in an ASSB full cell and its high current density and wide temperature range tests. Reproduced with permission from ref 26. Copyright 2021 The American Association for the Advancement of Science.

In this mini-review, we systematically summarized the challenges of each ASSB material and how they are being addressed based on recently reported research results. ASSB research began in earnest with the development of a solid-state electrolyte with lithium-ion conductivity comparable to that of liquid electrolytes, and in particular, sulfide-based LPSCl is currently establishing itself as a leading electrolyte in next-generation ASSBs. In the cathode composite containing LPSCl, high-capacity lithium transition metal oxide, which has been developed in current LIBs, is used as a cathode active material. The main direction of research so far was to analyze what happened at the interface between cathode active materials and LPSCl and how to address these interfacial problems. Various protection coating layers which have lithium-ion conduction ability have been suggested, as well as efforts to replace LPSCl electrolytes with other solid-state electrolytes that are stable in a highly oxidizing atmosphere. In the case of the anode, despite of high energy density of metallic lithium, it is difficult to obtain stable long-term performance due to its inherent dendritic growth phenomenon. However, a cell configuration that can control the cycling pressure which is one of the important factors in ASSB, a new interlayer that can eliminate dendrite growth of lithium metal anode, and a pure silicon anode that was difficult to apply in LIBs enable to implement high-energy-density ASSB with stable performance. The remaining obstacles for ASSBs are to devise a plan to apply ASSB to large-scale EVs or ESSs with these materials and cell concepts developed through various efforts in academia.