High-Performance All-Solid-State Lithium–Sulfur Batteries Enabled by Slurry-Coated Li6PS5Cl/S/C Composite Electrodes

Among many lithium secondary batteries, lithium–sulfur batteries stand out because of their high theoretical specific energy, low cost, non-toxicity and the fact that they cause no environmental pollution. However, due to poor electronic and ionic conductivity, shuttle effect, lithium dendrites and other defects, it remains a big challenge to achieve large-scale application of lithium-sulfur batteries. Here we report an all-solid-state lithium–sulfur battery based on Li-argyrodite Li6PS5Cl solid-state electrolytes through a slurry-coating method. Li6PS5Cl with a high ionic conductivity of 1.3 × 10–3 S cm−1 at room temperature is used as the solid electrolyte and the ion conductive additive in the electrode. The sulfur-based composite cathode is fabricated through a slurry-coating process by dispersing sulfur, Li6PS5Cl, ethyl cellulose, and carbon black in 1,3-dioxolane (DOL). This method can disperse the Li6PS5Cl around sulfur particles well, and the solvent does not react with any component of composite cathodes during preparation. The battery delivers a high discharge capacity of 962 mA h g−1 at room temperature for the first cycle at 80 mA g−1. While the Coulombic eﬃciency is approximately 99.5% during 100 cycles. This work provides a new insight into the combination method between the sulfide-type SSEs and sulfur cathodes, which is critical to the electrochemical performance of all-solid-state lithium-sulfur batteries.


INTRODUCTION
With the rapid increase in the energy requirements for energy storage devices, it has been difficult to meet these needs using traditional lithium-ion batteries (LIBs) due to low energy density, poor cycle stability, and high cost. In the future, energy devices will require higher and higher energy density (Armand and Tarascon, 2008;Nitta et al., 2015;Liu et al., 2018;Wang et al., 2018b). When lithium metal (low density and high electronegativity) is paired with elemental sulfur (theoretical capacity of up to 1,672 mA h g −1 ) to form a lithium-sulfur battery, the theoretical capacity density of the battery can reach 2600 W h kg −1 or 2800 W h L −1 (Ji et al., 2009;Yin et al., 2013). Moreover, compared with traditional LIBs, lithium-sulfur batteries have many unmatched advantages, such as being low cost, non-toxic, causes no environmental pollution and has a safer working voltage (Lochala et al., 2017;Qu et al., 2018). Based on the above advantages, lithium-sulfur batteries are regarded as the most promising next-generation energy storage product. However, there are still some bottlenecks in the application of lithium-sulfur batteries: 1) Liquid electrolytes are easy to flow, and flammable; 2) Sulfur is an insulator, and its poor conductivity and low active material utilization seriously affect the battery rate performance; 3) Polysulfide is easily dissolved in the electrolyte and the polysulfide shuttle effects lead to poor Coulombic efficiency and cycle performance; 4) Lithium dendrites can pierce the separator, causing short circuits (Balakumar and Kalaiselvi, 2015;Pang et al., 2016;Chen et al., 2018b).
Aiming to solve the problems mentioned above, many studies mainly focus on the optimization of the electrode. Currently, the most common solution is to combine carbon materials (mesoporous carbon (Ji et al., 2009;Han et al., 2020), hollow porous carbon (Jayaprakash et al., 2011), carbon nanotube (Guo et al., 2011;Zheng et al., 2019), and graphene Liu et al., 2020), etc.) with sulfur to improve sulfur utilization and to fix the polysulfide. But the surface of the carbon material is non-polar, it therefore cannot exert a strong anchoring effect on the polar polysulfide. In addition to carbon materials, metal oxides (TiO 2 (Wei Seh et al., 2013), MnO 2 (Liang et al., 2015), Al 2 O 3 (Han et al., 2013), Fe 2 O 3 (Zheng et al., 2017), and V 2 O 5 (Carter et al., 2017), etc.) can also fix polysulfide and conductive polymers (polyaniline (PANI) , and polypyrrole (PPy) (Zhang et al., 2016;Geng et al., 2019), etc.) can improve electronic conductivity of composite sulfur electrodes. The low electronic conductivity of metal oxides and the high cost of conductive polymers are still not optimal solutions to the serious capacity decay during cycling for a lithium-sulfur battery. Replacing the liquid electrolyte with solid-state electrolytes to achieve an all-solid-state lithium-sulfur battery is one of the most promising strategies to solve these issues (Lin et al., 2013;Han et al., 2016a;Zhang et al., 2017;Fan et al., 2018;Zhang et al., 2018b). The conversion of sulfur does not generate soluble polysulfide, so it can fundamentally solve the polysulfide shuttle effect . At the same time, the ultra-high mechanical modulus of SSEs can also effectively inhibit the growth of lithium dendrites, thereby improving the Coulombic efficiency and safety of the battery (Chen et al., 2018a;Cheng et al., 2019;Shen et al., 2019;Zhao et al., 2020). Unlike traditional lithium-sulfur batteries that rely on the infiltration of liquid electrolytes to achieve rapid ion migration, in the all-solidstate lithium-sulfur batteries, the ion migration is mainly achieved through solid electrolytes (Kato et al., 2016;Yan et al., 2019;Zhao et al., 2019;Bai et al., 2020;Ding et al., 2020).
Among various SSEs, sulfide SSEs stand out because of their high ionic conductivity, low grain boundary resistance, compatible interface with sulfur-based cathodes, and their easy processability (Chen et al., 2018a;Ma et al., 2018). The most well-studied sulfide SSEs are the binary xLi 2 S-(100-x)P 2 S 5 system (x 0.4-0.8) (Zhang and Kennedy, 1990;Kanno and Murayama, 2001;Ohtomo et al., 2013). The ionic conductivity of β-Li 3 PS 4 synthesized via wetchemistry is 1.6 × 10 -4 S cm −1 with an activation energy of 0.36 eV . Crystalline Li 7 P 3 S 11 is also an important member of the binary system, because of its very high ionic conductivity (1.7 × 10 -2 S cm −1 at room temperature (RT)) (Seino et al., 2014). In 2011, Li 10 GeP 2 S 12 (LGPS) was found to have a an extremely high ionic conductivity of 1.2 × 10 -2 S cm −1 at RT (Kamaya et al., 2011). As an isostructural alternative to LGPS, Li 10 SnP 2 S 12 has the advantages of being low cost and easily promoted, but it shows a lower ionic conductivity of 4 × 10 -3 S cm −1 , because of a slightly different lithium ion disorder (Bron et al., 2013). The highest reported ionic conductivity was found in the doping binary system, Li 9.54 Si 1.74 P 1.44 S 11.7 Cl 0.3 , 2.5 × 10 -2 S cm −1 at RT (Kato et al., 2016). Li-argyrodite Li 6 PS 5 Cl was reported as a promising electrolyte with a high ionic conductivity of 1.3 × 10 -3 S cm −1 at RT (Boulineau et al., 2012;Zhang et al., 2018a;Zhang et al., 2019). The disadvantages of most sulfide electrolytes are their narrow electrochemical window and the electrochemical instability between the electrolytes and lithium metal (Han et al., 2016b). The theoretical calculation results (Zhu et al., 2015;Han et al., 2016b;Zhu et al., 2016) show that LGPS begins to be lithiated and reduced at 1.7 V. The phase equilibrium components of LGPS at 0 V consist of Li 3 P, Li 2 S, and Li 15 G 4 , which have also been observed in experiments. The interphase has a higher growth rate between LGPS and lithium, which means the presence of cations in SSEs such as Ge, Sn, Ti is detrimental to the formation of stable interface (Sakuma et al., 2016). But Li 6 PS 5 Cl can form a stable SEI, the thickness of SEI and interfacial resistance become stable after a certain time and no longer increase over time (Wenzel et al., 2016a;Wenzel et al., 2016b;Wenzel et al., 2018). Many cathode active materials such as LiCoO 2 , LiNi 0.8 Co 0.1 Mn 0.1 O 2 , S, and Li 2 S have been used in all-solidstate lithium batteries (ASSLBs) with Li 6 PS 5 Cl (Boulineau et al., 2012;Boulineau et al., 2013;Huang et al., 2015;Yubuchi et al., 2015;Han et al., 2016a;Yu et al., 2016;Zhang et al., 2018a;Zhao et al., 2019;Zhang et al., 2020). Han et al. used a novel bottom-up method to synthesize nanocomposite cathode by dissolving Li 2 S and Li 6 PS 5 Cl in ethanol, which showed a large reversible capacity of 830 mAh g −1 at 50 mA g −1 for 60 cycles (Han et al., 2016a). Yu et al. fabricated the S/Li 6 PS 5 Cl/Li-In ASSLBs with a ball-milled S-Li 6 PS 5 Cl composite cathode . The batteries displayed large capacity around 1,400 mAh g −1 during the first cycle and decayed rapidly 400 mA h g −1 after 20 cycles. In 2018, Nan et al. employed the nano-sulfur/multiwall carbon nanotube composites combined with Li 6 PS 5 Cl as the cathode, which delivered a high discharge capacity of 1850 mA h g −1 at 0.1 C for the first cycle and 1,393 mAh g −1 after 50 cycles (Wang et al., 2018a).
However, there are very few works on the synthesis of sulfur composite cathodes for ASSLBs by wet chemical methods. In this work, we proposed a new preparation method about sulfurcomposite electrodes for ASSLBs using a liquid-phase process with a DOL solution. The sulfur-composite electrodes consist of sulfur as the active material, Li 6 PS 5 Cl as the solid electrolyte, ethyl cellulose as the binder, and Super P (SP) as the conductive additive. The microstructure, ionic and electronic conductivity, and electrochemical stability of Li 6 PS 5 Cl were systematically investigated. The electrochemical property of the sandwichtype S/Li 6 PS 5 Cl/Li cell was tested at 80 mA g −1 and 30°C. Finally, to further investigate how interface affects battery performance, S/Li 6 PS 5 Cl interface properties before and after cycling were examined by ex situ characterizations including scanning electron microscopy (SEM), energy-dispersive spectra (EDS), and electrochemical impedance spectra (EIS).

Material Synthesis
Synthesis of Li 6 PS 5 Cl: The Li 6 PS 5 Cl was prepared using a method described in our previous work Zhang et al., 2019;Zhang et al., 2020). In brief, a stoichiometric mixture of laboratory-grade Li 2 S, P 2 S 5 (99%, Aladdin), and LiCl (99.99%, Aladdin) was milled at 500 rpm for 24 h. Subsequently, the mixture was sealed in a stainless-steel tube and annealed at 500°C for 2 h to obtain argyrodite Li 6 PS 5 Cl. All the experiments were carried out in an Ar-filled glove box (H 2 O and O 2 < 0.5 ppm).

Fabrication of All-Solid-State Lithium-Sulfur Batteries
The sulfur-composite electrodes (sulfur powder (99%, Alfa Aesar): Li 6 PS 5 Cl: SP (SCM industrial Chemical Co., Ltd.): ethyl cellulose (Aladdin)) were dispersed with anhydrous DOL (≥99.8%, Aladdin) in a mass ratio of 60 : 25: 10 : 5 under continuous stirring for 6 h. The slurry was then coated onto carbon-coated aluminum foil with a glass rod and dried at 55°C for 24 h. The average sulfur loading in each electrode is 0.95 mg cm −2 . For the assembly of S/Li 6 PS 5 Cl/Li cell, the sulfur-composite electrode was covered with 120 mg Li 6 PS 5 Cl powder and pressed together under 350 MPa in a stainless-steel tank with a diameter of 12 mm. After that, lithium metal as the counter electrode was attached on the other side of the Li 6 PS 5 Cl layer. Finally, three-layered pellets (shown in Figure 1) were assembled with a 2032 coin-type cell in an Ar-filled glove box. For the SS/Li 6 PS 5 Cl/Li, Li/Li 6 PS 5 Cl/Li cell and SS/Li 6 PS 5 Cl/SS, lithium foil or stainless steel (SS) as the working electrode, the assembling procedure remained exactly the same. All the procedures were carried out in an Ar-filled glove box (H 2 O and O 2 < 0.5 ppm).

Characterization and Electrochemical Measurements
X-ray diffraction (XRD) patterns of samples were recorded by an X-ray diffractometer (Rigaku Ultima IV) with Cu K α radiation (λ 0.15418 nm). Data was recorded in the 2-theta range of 10-80°. The Raman spectra in a range of 100-2,500 cm −1 were collected using a DXR Raman microscope (Renishaw InVia Raman spectrometer) with He-Ne 532 nm laser excitation and a ×50 objective. The laser beam with a laser spot size of ca. 1 μm was focused on each sample and the acquisition time for each spectrum was 20 s. The morphology of samples was observed by SEM (SEM, Hitachi S4700). An elemental analysis was conducted on EDS attached to SEM. The cyclic voltammetry (CV) measurement, the linear sweep voltammetry (LSV) measurement, and the direct-current (DC) polarization test was conducted on a CHI660D electrochemical workstation (Chenhua, Shanghai). The CV curve of S/Li 6 PS 5 Cl/Li cell was recorded between 0.4 V and 3 V with a sweep rate of 0.1 mV s −1 . The LSV curve of SS/Li 6 PS 5 Cl/Li cell was recorded in the potential range from 0 to 6 V with a sweep rate of 0.1 mV s −1 . DC polarization test of a blocking SS/Li 6 PS 5 Cl/SS cell at 2 V to determine the electronic conductivity. The ionic conductivity of Li 6 PS 5 Cl was calculated by the equation: where R is the total resistance of the electrolyte, d is the sample thickness, and S is the area of the electrolyte. The total resistance of Li 6 PS 5 Cl was measured on the SS/Li 6 PS 5 Cl/SS cell using the Zennium electrochemical workstation (ZAHNER, Germany), in which the frequency ranges from 100 mHz to 4 MHz with an amplitude of 10 mV and the testing temperature ranges from 25°C to 60°C in stepwise increments of 5°C. The EIS measurements of the S/Li 6 PS 5 Cl/Li cell before and after the cycle were measured using the same instrument in the same frequency. A Galvanostatic discharge-charge test of S/Li 6 PS 5 Cl/Li cell was performed in a potential range from 0.4 to 3 V (vs. Li + /Li) at 80 mA g −1 using a Neware battery test system (Neware, Shenzhen). The stability of lithium against Li 6 PS 5 Cl was tested by the Neware using symmetric Li/Li 6 PS 5 Cl/Li cell at a current density of 0.1 mA cm −2 .

RESULT AND DISCUSSION
The crystal structure, composition, and morphology of the sulfurcomposite electrode were characterized by XRD, Raman, and SEM, respectively. Figure 2A shows the typical XRD patterns of the sulfur-composite electrode and all the materials used in the synthesis process. After sintering at 500°C for 2 h, the peaks of Li 6 PS 5 Cl can be indexed to the crystalline Li 7 PS 6 structure, where the Cl atoms can replace S atoms in Li 7 PS 6 (Deiseroth et al., 2008). Regarding the sulfur-composite electrode, there are no obvious peaks of sulfur; the main diffraction peaks and the broad diffraction peaks are indexed as Li 6 PS 5 Cl and amorphous carbon materials with low crystallization degree, respectively. It indicates that Li 6 PS 5 Cl can distribute evenly on the surface of sulfur particles. By comparing the Raman spectra ( Figure 2B) of all materials, the sulfur-composite electrode can be further analyzed.
The weak peak at 425 cm −1 associated with PS 4 3− (ortho- thiophosphate) from the ionic formula (Li + ) 6 (PS 4 3− )S 2− Cl − (Deiseroth et al., 2008;Ohara et al., 2016) and peaks at 153 cm −1 , 219 cm −1 and 474 cm −1 associated with S-S bond were observed, indicating no chemical reaction occurred in DOL solvent during the synthesis process. As shown in Supplementary Figure S1, the irregular shape of the Li 6 PS 5 Cl particle is beneficial to increasing the contact area between the active material and electrolyte in the composite cathode. As seen  from the morphology and elemental distribution in Figure 2C, the sulfur particles are well embedded in the solid electrolyte and conductive additives, resulting in a dense composite and a low interfacial resistance.
The temperature dependence of the ionic conductivity of Li 6 PS 5 Cl was systematically investigated. Supplementary Figure S2 shows the different impedance spectra of Li 6 PS 5 Cl at different temperatures. Through calculations based on resistance values, the ionic conductivity of Li 6 PS 5 Cl exhibits a gradual increase as the temperature increases, as shown in Figure 3A. It shows the ionic conductivities of 1.30 × 10 -3 S cm −1 and 1.05 × 10 -2 S cm −1 at 25°C and 60°C, respectively. The activation energy E a is calculated from the slope of the linear Arrhenius plot according to the Arrhenius equation: where T is the absolute temperature, A is a pre-exponential factor, and k is the Boltzmann constant. According to Figure 3A, the E a value of Li 6 PS 5 Cl is 0.32 eV, which is low in comparison to most sulfide SSEs. In addition to ionic conductivity, the electronic conductivity of Li 6 PS 5 Cl has also been investigated by the DC polarization test ( Figure 3B). It shows a low electronic conductivity of 3.56 × 10 -8 S cm −1 , which is beneficial for suppressing lithium dendrite growth. LSV was used to determine the electrochemical window, which is important in evaluating the stability of Li 6 PS 5 Cl against a lithium anode and cathode. As shown in Figure 3C, the decomposition current corresponding to Li stripping (Li → Li + + e − ) is observed around 0 V vs Li/Li + . There is no significant current change in the potential range from 0 to 6 V vs Li/Li + . Assembling the symmetric Li/Li 6 PS 5 Cl/Li cell is also an important experimental method to investigate the stability between Li 6 PS 5 Cl and lithium. The voltage profile of Li 6 PS 5 Cl remains at a lower voltage (≈10mV) during 400 h at 0.1 mA cm −2 . The results of LSV and symmetric battery prove that the Li 6 PS 5 Cl is suitable to pair with the lithium anode. Considering the high ionic conductivity, low electronic conductivity, and wide electrochemical window, Li 6 PS 5 Cl is an appropriate candidate for an ionic conductor and battery separator. Figure 4A shows the charge-discharge profiles of the S/Li6PS5Cl/Li cell between 0.4 and 3.0 V during the first three cycles. The current density is 80 mA g-1 and the test temperature is 30°C. The initial discharge capacity of ASSLBs is 962 mAh g-1 and the corresponding Coulombic efficiency is as low as 56%, which are ascribed to the activation process for the sulfur cathode, interfacial decomposition, and the interfacial chemical reaction between the electrolyte and electrode (Zhang et al., 2015;Han et al., 2016a). As shown in Supplementary Figure S3, the CV was used to study the mechanism of the electrode reaction. During the first cycle, one reduction peak can be observed at 1.8 V due to the reduction of S to Li2S and the oxidation peak at 2.5 V can be attributed to the oxidation of Li2S to S (Takeuchi et al., 2010;Wang et al., 2018a). During the subsequent cycles, both reduction and oxidation peaks slightly shift, which means a good reversibility of the composite cathode. Figure 4B presents the cycling performance and Coulombic efficiency, the cell delivers a high Coulombic efficiency of nearly 100% after three cycles, specific capacity decreases to 387 mAh g-1, and capacity retention reaches 66% after 100 cycles. The impedance changes for the S/Li6PS5Cl/Li cell before and after cycling are shown in Figure 4C. Two Nyquist plots were fitted into one equivalent circuit. The equivalent circuit is constructed to distinguish common resistances: Rse, Rct1, Rct2, and Zw, representing the bulk resistance, two interfacial resistances, and Warburg impedance, respectively (Deng et al., 2015;Takada et al., 2015). Based on our previous work , the Rct1 and Rct2 represent the interfacial resistance between lithium and the electrolyte, and the interfacial resistance between the cathode and electrolyte, respectively. The bulk and interfacial resistance values increase after cycling. But the variations of Rse and Rct1 are very small (≈40 Ω). There is a large difference in value (≈675 Ω) of Rct2, which is attributed to interfacial reactions or a chemo-mechanical failure (Ohno et al., 2019). These negative effects may be the main reason for the low initial Coulombic efficiency and the subsequent capacity loss.
To investigate the interfacial reactions or chemo-mechanical failure in the S/Li 6 PS 5 Cl/Li cell after cycling, a cross-sectional SEM was conducted, and the elemental distributions of Al, S, P, Cl, and C elements in the red selected area were assessed by EDS, as depicted in Figure 5. Since both active material and electrolytes are sulfur-based materials, it is difficult to determine where the chemical decomposition products at the interface come from. Compared to the morphologies and elemental distribution of the C element at the S/Li 6 PS 5 Cl interface, it was found that the thickness of the sulfur-composite electrode after cycling is more than twice what it was before. This significant change can be explained by the volume change (chemo-mechanical failure) because of the conversion reaction of S and a small number of Li 6 PS 5 Cl. There is about 80% added volume in the cathode after discharging, when 1 mole of S 8 converted into 8 moles of Li 2 S. The loose distribution of the C element after cycling, confirms that the volume increase stresses the surrounding matrix of solid electrolytes and carbon materials, resulting in a lot of pores and voids in the cathode. Thus, at this scale, such a large volume change has a much greater impact on interfacial resistance than on chemical reactions. In other words, chemo-mechanical failure is an urgent issue that needs to be resolved in all-solid-state lithium-sulfur batteries.

CONCLUSION
Argyrodite Li 6 PS 5 Cl is an ionic conductor with high ionic conductivity and a wide electrochemical window. The Li 6 PS 5 Cl possesses good potential for applications in lithium-sulfur batteries. We successfully demonstrated a new strategy to fabricate a sulfur cathode for ASSLBs via the solution coating method by mixing sulfur materials, Li 6 PS 5 Cl solid electrolyte, ethyl cellulose binder, and SP conductive additive in a DOL solution. The slurry coating achieved intimate contact between the S/C particles and the Li 6 PS 5 Cl solid electrolyte, guaranteeing adequate lithium-ion pathways. The battery shows a high initial discharge capacity of 962 mAh g −1 at 80 mA g −1 and a specific capacity of 387 mAh g −1 . After 100 cycles, the capacity retention reaches 66%. Although the Li 6 PS 5 Cl solid electrolyte can eliminate the shuttle effect and alleviate interfacial reaction, the volume change of sulfur still seriously affects the performance of ASSLBs.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

AUHTOR CONTRIBUTIONS
Conceptualization, JZ, HH, and WK; experimental, CZ, KW, and LL; data analysis, CZ, YG, YX, and CL; writing-original draft preparation, ZC; writing-review and editing, JZ; supervision, JZ, and WZ. All authors have read and agreed to the submitted version of the manuscript.

FUNDING
The authors acknowledge the support of the Zhejiang Provincial Natural Science Foundation of China under grant No.