Long‐life high‐capacity lithium battery with liquid organic cathode and sulfide solid electrolyte

Electrochemical batteries with organic electrode materials have attracted worldwide attention due to their high safety, low cost, renewability, low contamination, and easiness of recycling. However, the practical application of such system is limited by low density, low electronic/ionic conductivity, and the dissolution of organic electrode materials in conventional liquid electrolytes. Herein, a novel battery configuration is proposed to replace liquid electrolyte/solid organic cathode with solid electrolyte/liquid organic cathode to ultimately solve the shuttle effect and dissolution problem of organic cathodes. More importantly, this configuration combines room‐temperature high‐safety liquid lithium metal anode Li‐BP‐DME that can essentially inhibit lithium dendrite nucleation/growth and sulfide SE with ultrahigh room‐temperature ionic conductivity for facilitated ion‐conduction.


| INTRODUCTION
Inorganic transition-metal oxide and phosphate materials (such as LiCoO 2 , LiMn 2 O 4 , LiNi 1-x-y Co x Mn y O 2 , or LiFePO 4 ) are currently the major cathode materials for commercial lithium-ion batteries, but the mineral resources used to produce these materials are nonrenewable and limited. The global energy market will face a new round of resource crises with the dramatically increased production of electric vehicles and portable electronics as well as the complicated/costly battery recycling technology. Therefore, recyclable organic electrode materials have received considerable attention in recent years [1][2][3][4] due to their high safety, low cost, renewability, low pollution, and ease to recycle, thus the development of recyclable organic electrode materials has become the research frontier in the battery field. [5][6][7][8][9][10] Organic electrode materials are rich in carbon, hydrogen, oxygen, and other elements. With the advantages of renewability, low cost, and high capacity, organicelectrode lithium-ion batteries are expected to be a very promising candidate for the energy-storage system. Although organic electrode materials have many advantages, their applications in batteries still face many technical challenges. 11 First of all, the low density of organic electrode materials (<2 g cm −3 ) leads to a low bulk energy density of the battery. Moreover, the low electronic and ionic conductivities of organic electrode materials require the use of a large number of conductive agents, also resulting in low energy densities. In addition, the dissolution of organic electrode materials in conventional electrolytes leads to a severe loss of active materials and the shuttle effect, which lead to capacity decay and low cycle efficiency.
To solve the above difficulties, this work innovatively replaces the conventional organic electrolyte/solid organic cathode with a solid electrolyte (SE)/liquid organic cathode to successfully solve the shuttle effect and dissolution problem. Replacing the liquid electrolyte with SE not only provides advantages in battery design but also avoids catastrophic consequences such as outgassing, thermal runaway, and combustion caused by short circuits in conventional carbonate liquid-electrolyte batteries. Since organic quinones are generally soluble in liquid organic electrolytes, the cycle performance of lithium-ion batteries with quinone organic compounds as cathodes is not good. In this work, an organic solvent with higher solubility is adopted to dissolve the solid organic cathode for the liquid organic cathode, which successfully avoids the capacity degradation caused by solid anthraquinone dissolved in liquid organic electrolytes. This work also mitigates bulk density and conductivity issues through electrode design and battery configuration design. The new battery uses anthraquinone (AQ, a carbonyl compound with a high theoretical specific capacity, 256 mAh/g), as the active material for cathode, liquid-lithium metal (Li-BP-DME) anode, and sulfide SE. Liquid lithium metal Li-BP-DME [12][13][14][15] has the advantages of low cost, simple/efficient preparation, high conductivity of 12 mS cm −1 , and low potential of 0.3 V versus Li/Li + . Most importantly, it can essentially inhibit the nucleation and growth of lithium dendrites and prevent the safety hazard of internal short circuits caused by dendrites growth. In addition, compared with lithium metal, the reaction between liquid lithium metal and water is mild without any observable flame. Sulfide SEs have the highest room-temperature ionic conductivity among all types of SEs, ideal mechanical ductility, and good thermal stability. Moreover, a fully compact sulfide SE membrane with low porosity can be achieved by simple room-temperature cold pressing, [16][17][18][19][20][21][22][23][24][25][26][27][28][29] which is almost void-free to effectively prevent the shuttle effect and short circuit. Furthermore, Li 7 P 3 S 11 was found to be the most suitable sulfide SE through compatibility experiments of various sulfide SEs with organic cathodes and anodes. The organic lithium battery assembled with Li 7 P 3 S 11 shows longer cycle life and higher capacity compared with the organic lithium battery using liquid electrolytes. These results corroborate that this new secondary battery has the advantages of desirable electrochemical performance and low cost, which provides a new idea for the development of energy-storage devices.
2 | EXPERIMENTAl SECTION 2.1 | Li 1.5 BP 3 DME 10 solution preparation The Li 1.5 Bp 3 DME 10 solution was prepared in a glove box filled with argon (O 2 < 0.1 ppm, H 2 O < 0.1 ppm). Different amounts of lithium metal (China Energy Lithium Co. Ltd), biphenyl (alfa Aesar), and DME (alfa Aesar) were prepared according to the molar ratio of the Li 1.5 Bp 3 DME 10 solution. Then, biphenyl was dissolved in DME at ambient temperature in glass bottles to form a transparent, colorless solution. Finally, the lithium metal pieces were put into the Bp 3 DME 10 solution with heating for 10 min in a 60°C water bath.

| AQ solution preparation
Five milligrams of anthraquinone were dissolved in 10 mL of propylene carbonate (PC) solution; 2.87 g of Lithium bis (trifluoromethanesulphonyl)imide LiTFSI and 400 mg of conductive carbon Super P were added into the PC solution. Finally, the mixture was well stirred.

| Preparation of the Li 7 P 3 S 11 pellet
First, amorphous Li 7 P 3 S 11 precursor powder was synthesized by two methods. One is liquid phase synthesis. The precursor was obtained by dissolving Li 2 S (99.9%, Ganfeng Lithium) and P 2 S 5 (99.9%, Macklin) with a molar ratio of 7:3 in 20 mL Acetonitrile (ACN) at 50°C for 48 h. The ACN solution was evaporated and the precursor powder was placed in a vacuum oven and treated at 100°C for 24 h. The other method is solid phase synthesis. The precursor was obtained by ball-milling Li 2 S (99.9%, Ganfeng Lithium) and P 2 S 5 (99.9%, Macklin) with a molar ratio of 7:3. Then the precursor was pressed into pellets, which were vacuum-sealed in a quartz tube, and heated at 260°C for several hours. 150 mg of the Li 7 P 3 S 11 power was pressed into a pellet under a pressure of 65 MPa. The diameter of the Li 7 P 3 S 11 pellet was 15 mm and the thickness was 500 μm.

| Preparation of home-made battery
Five milliliters of Li 1.5 Bp 3 DME 10 solution and 5 ml AQ cathode solution were added into two tanks of the homemade cell separately. The Li 7 P 3 S 11 pellet was placed in the middle of two tanks. The schematic illustration of the battery system is shown in Figure 1. The crystal structure of synthesized materials was investigated by X-ray diffraction (XRD, Beijing Purkinje XD2), equipped with Cu Kα radiation (λ = 1.54056 A) and operated at 36 kV and 20 mA in the 2θ range of 10°-80°. The ionic conductivity was measured by electrochemical impedance spectroscopy (EIS) by using an electrochemical workstation (Zahner IM6e). EIS data were recorded in the frequency range from 100 mHz to 8.0 MHz by applying an alternating voltage of 5 mV at room temperature. The galvanostatic charge-discharge test was carried out on a battery tester (CT2001, Land Ins.) at a cutoff voltage between −5 V and 5 V.

| Compatibility study of different sulfide solid electrolytes with organic electrodes
Sulfide SE has poor chemical stability with a polar solvent, and the contact between them will lead to chemical side reactions. As a result, on the surface of sulfide SE, an interfacial layer is generated, which determines the performance of the battery. Therefore, the selection of sulfide SEs that are compatible with organic liquid electrodes is a prerequisite for the operation of the battery. Comparative experiments were conducted on four mainstream sulfide SEs (Li 6  assembled with Li 6 PS 5 Cl and Li 7 SiPS 8 have no voltage plateau for charging and discharging, and the voltage values jump between cutoff voltages (Figure 2A,B). This indicates that the ionic conductivity of the interface formed by the reaction of the Li 6 PS 5 Cl and Li 7 SiPS 8 in contact with the organic liquid electrode (AQ-LiTFSI-PC and Li-BP-DME solution) is very poor, and therefore the battery cannot operate properly. The voltage-time curve of the battery assembled with Li 3 PS 4 shows that the voltage decreases slowly with time and the battery remains in a discharged state without normal charging and discharging ( Figure 2C). This indicates that the interface formed by the reaction between Li 3 PS 4 and the organic liquid electrode after contact is unstable and a continuous interfacial reaction occurs, so the battery also cannot operate properly. The voltage-time curve of the battery assembled with Li 7 P 3 S 11 shows a normal charge-discharge curve as well as a discharge voltage plateau of~1.9 V and a charging voltage plateau of~2.2 V ( Figure 2D). It indicates that the interface formed by the reaction of Li 7 P 3 S 11 in contact with the organic liquid electrode is more stable and the ionic conductivity of the interfacial layer is better, so the battery can operate normally. The different crystal structures of Li 3 PS 4 and Li 7 P 3 S 11 are used to explain the possible reasons for the different battery behaviors of Li 3 PS 4 and Li 7 P 3 S 11 . It is well known that Li 3 PS 4 is mainly composed of PS 4 3− units, while the structure of Li 7 P 3 S 11 is based on PS 4 3− and P 2 S 7 4− units. Bridging sulfur exists in the P 2 S 7 4− unit, but not in PS 4 3− unit. The different chemical bond strengths and chemical stabilities of these two structural units result in their different chemical compatibilities with liquid cathode/anode materials, which may be the reason for the differences in batteries with Li 3 PS 4 and Li 7 P 3 S 11 . Through the above comparison experiments, it is found that Li 7 P 3 S 11 electrolyte has better compatibility with the organic liquid electrode (AQ-LiTFSI-PC and Li-BP-DME solution), and the organic lithium battery assembled with it can operate normally.

| Batteries electrochemical performance of Li 7 P 3 S 11 electrolytes prepared by different methods
After screening the Li 7 P 3 S 11 electrolyte as a sulfide SE of with better compatibility with organic liquid electrodes, the effects of Li 7 P 3 S 11 electrolytes prepared by different preparation methods (solid phase synthesis and liquid phase synthesis) on the battery performance were further investigated. Figure 3 shows the electrochemical performance of the battery with Li 7 P 3 S 11 synthesized by the solid phase method. Figure 3A shows the 1st, 2nd, 10th, and 14th cycle charge/discharge profiles of the batteries in the voltage range of 1.6−2.5 V. The specific capacity in the first cycle was 395.6 mAh g −1 , which exceedes AQ's theoretical specific capacity of 256 mAh g −1 . This may be due to the interfacial reaction between the Li 7 P 3 S 11 electrolyte and the organic liquid electrode upon contact during the initial discharge process, and there was a partial decomposition of Li 7 P 3 S 11 , which contributed a part of the specific capacity. In the later cycles, the battery capacity gradually decreased, and in the 10th cycle, the discharge-specific capacity decreased to 241.1 mAh g −1 , which was close to the theoretical specific capacity of AQ. The charge and discharge voltage plateau were~2.2 and~1.9 V, respectively. Figure 3B shows the cycling performance of the battery. The first-cycle Coulombic efficiency was 88.2% and maintained~90% until the 14th cycle. After the 14th cycle, the battery capacity decreased rapidly and the battery failed. The reason may be that the interface layer generated during the initial cycle has good ionic conductivity, which enables the battery to operate, but the electrochemical stability of this interface layer is poor, and poor ionic conductivity is continuously generated during the battery cycle and accumulates at the interface, leading to an increase in the internal resistance of the battery, and the battery eventually fails. Figure 4 shows the electrochemical performance of the battery with Li 7 P 3 S 11 synthesized by the liquid phase method. Figure 4A shows the 1st, 2nd, 10th, 20th, and 50th cycle charge/discharge profiles of the batteries in the voltage range of 1.6−2.5 V. The specific capacity in the first cycle is 251.7 mAh g −1 , and this specific capacity value is close to the theoretical specific capacity of anthraquinone (256 mAh g −1 ). The battery capacity gradually decreased in the later cycles, the dischargespecific capacity decreased to 161.1 mAh g −1 at the 10th cycle and 90.1 mAh g −1 at the 50th cycle. The charge and discharge voltage plateau were~2.2 and~1.8 V, respectively. From the charge/discharge curves, the polarization voltage of the battery increased with the increase of the cycle. This may be due to the deterioration of the interface or the denaturation of the organic liquid electrode due to the lack of sealing of the battery. Figure 4B shows the cycling performance of the battery. The first-cycle Coulombic efficiency was 58.6% and maintains~90% until the 77th cycle. The battery capacity gradually decreased (from 251.7 mAh g −1 [1st cycle] to 43.5 mAh g −1 [77th cycle]). Compared with the liquid phase-synthesized Li 7 P 3 S 11 -assembled batteries, the solid phase-synthesized Li 7 P 3 S 11 -assembled batteries showed better cycling performance, and no sudden failure was observed.
To investigate the reasons for the different performance of the batteries assembled from Li 7 P 3 S 11 prepared by the two methods (solid phase synthesis and liquid phase synthesis), XRD characterizations of Li 7 P 3 S 11 synthesized by the two methods were carried out and corresponding ionic conductivities were tested by EIS spectra. The XRD patterns of the synthesized Li 7 P 3 S 11 solid electrolytes via the solid-state synthesis method and liquid phase synthesis are shown in Figure 5A,B, which corresponds well with the indexed diffraction pattern of the Li 7 P 3 S 11 phase (ICSD #157654) without the impurity phase. This indicates that the electrolyte material synthesized by two methods is Li 7 P 3 S 11 . The electrochemical performances of the synthesized Li 7 P 3 S 11 via two routes were also compared. Figure 5C,D shows the Nyquist plots. The plots both consist of a semicircle and a spike. The line in the low-frequency region represents the diffusion impedance, and the incomplete semicircle contains the grain boundary resistance (R SE,gb ) of the solid electrolyte in the intermediate frequency region and the bulk resistance (R SE,bulk ) of the solid electrolyte in the high-frequency region. After calculation, the ionic conductivity of the Li 7 P 3 S 11 electrolyte synthesized in the liquid phase method was 0.0045 mS cm −1 and that of the Li 7 P 3 S 11 electrolyte prepared by the ball milling method was 0.64 mS cm −1 . The high ionic conductivity of the Li 7 P 3 S 11 electrolyte synthesized in the solid phase is consistent with the lower polarization voltage of the batteries assembled with the Li 7 P 3 S 11 electrolyte synthesized in the solid phase. As for the liquid phase-synthesized Li 7 P 3 S 11 electrolyte with low ionic conductivity but relatively better cycling performance, it may be because the solvent involvement in the electrolyte preparation process affects not only the electrolyte material but also the electrochemical performance of the battery.
The cycle performance of batteries assembled with the Li 7 P 3 S 11 electrolyte prepared by liquid-phase synthesis is better than the AQ cathode batteries of the liquid electrolyte without additives such as LiNO 3 . [30][31][32][33][34] Comparison of the cycling performance in this study with the previous reports about AQ for lithium batteries is shown in Figure 6 and Table 1. Due to the high solubility and shuttle effect of carbonyl compounds in conventional liquid electrolytes, the cycling performance of AQ cathode batteries is poor, and the dischargespecific capacity decreases rapidly with increasing cycles, with the specific capacity of the battery dropping to 80-125 mAh g −1 after 10 cycles and to about 33-80 mAh g −1 after 50 cycles. [30][31][32][33][34] However, it can be seen that cycling performance in this study is higher than those reported. The specific capacity of the battery is 161 mAh g −1 after 10 cycles and 90 mAh g −1 after 50 cycles. The above results indicate that this sulfide-based organic lithium battery can solve the problem of the high solubility of AQ in liquid electrolyte and shuttle effect, therefore improving the cycling performance of the AQ organic lithium battery. F I G U R E 5 X-ray diffraction (XRD) patterns of the synthesized Li 7 P 3 S 11 solid electrolytes via solid-state synthesis method (A), liquid phase synthesis (B), and the indexed diffraction pattern of the Li 7 P 3 S 11 phase (ICSD #157654). Impedance spectroscopy of Li 7 P 3 S 11 solid electrolytes via liquid synthesis method (C) and solid-state phase synthesis (D).
F I G U R E 6 Comparison of the cycling performance in this study with the previous reports about anthraquinone (AQ) for lithium batteries.
T A B L E 1 Comparison of the cycling performance in this study with the previous reports about anthraquinone (AQ) for lithium batteries.

| CONCLUSION
In summary, a novel battery configuration is proposed to replace liquid electrolyte/solid organic cathode with solid electrolyte/liquid organic cathode to ultimately solve the shuttle effect and dissolution problem of organic cathodes. Li 7 P 3 S 11 is found to be the most suitable sulfide SE through compatibility experiments of various sulfide SEs with organic cathodes and anodes, and the electrochemical performances of batteries assembled with Li 7 P 3 S 11 prepared by liquid-phase synthesis are better than solid-phasesynthesized counterparts. Moreover, the organic lithium battery assembled with Li 7 P 3 S 11 and roomtemperature high-safety dendrite-free liquid lithium metal anode Li-BP-DME shows longer cycle life and higher capacity compared with the organic lithium battery using the liquid electrolyte. These results show that this new secondary battery has the advantages of long cycle life and low cost, which provides a new idea for the development of energy-storage batteries.