Ultra-High-Capacity Lithium Metal Batteries Based on Multi-Electron Redox Reaction of Organopolysulfides including Conductive Organic Moieties

Recently, organic polysulfides have been synthesized as cathode active materials exceeding the battery performance of sulfur. However, the conventional organic polysulfides have exhibited capacities lower than the theoretical capacity of sulfur because the π-organic moieties do not conjugate with the sulfur chains. In this work, the organopolysulfides, synthesized via inverse vulcanization using disulfide compounds, exhibited higher capacities equal to the theoretical capacity of sulfur because of enhanced electronic conductivity based on the conjugation between organic moieties and sulfur chains. Furthermore, the organopolysulfide including 1,3-dhitiol-2-thione moiety exhibited the highest capacity because of the enhanced electronic conductivity. This finding will pave the way to develop next-generation rechargeable batteries.


Introduction
Lithium-ion batteries (LIBs), which were composed of graphite anodes (capacity of 372 mAh g −1 ) and inorganic cathode active materials, such as LiCoO 2 , LiMnO 2 , and LiFePO 4 (capacity of 140-170 mAh g −1 ) [1,2], have been used as electric sources in mobile electric devices and electric vehicles because of their capacities and energy densities (250-400 Wh kg −1 ) [3]. Recently, to develop rechargeable batteries with higher capacity and energy density than LIBs for longer-life electric devices and longer-range mobile electric vehicles, anode and cathode active materials have been proposed to develop rechargeable batteries with higher capacity and energy density than LIBs. The candidates with high theoretical capacities have been widely investigated based on Equation (1) (where n and M w represent reactive electrons and the molecular weight of active materials, respectively) because rechargeable batteries exhibit a lower capacity of either cathode or anode active materials.
Theoretical capacity = 26, 800 One criterion is that redox-active materials were composed of lightweight atoms. The other is that redox-active materials show multi-electron redox reaction. Among the alternative anode materials, lithium metal is one of the powerful materials for the breakthrough of the capacity and energy density limitation of current LIBs because of its light-weight, ultra-high theoretical capacity of 3860 mAh g −1 and low redox potential (-3.04 V vs. standard hydrogen electrode) [4,5]. On the other hand, conventional inorganic cathode active materials have not reached a capacity of more than 200 mAh g −1 because of heavy-weight metal ions and one-electron redox reaction.
As the candidates for cathode active materials of high-capacity batteries, compounds with disulfide (S-S), such as organodisulfides, organosulfides, and sulfur ring S 8 , have been used. They have been studied as alternative cathode materials because S-S bonds, composed of lighter element S than transition metals, show two-electron redox reaction reversibly. Organodisulfides [6] and organosulfides [7] have been synthesized as cathode active materials with high capacities between 300-1200 mAh g −1 since Visco and co-authors reported the electrochemical redox reaction of organodisulfides and S-S bond formation (oxidation)/cleavage (reduction) [8] (Equation (2), where R represents organic moieties).
In particular, lithium-sulfur (Li-S) batteries, using sulfur ring S 8 and lithium metal as the cathode and the anode active materials, respectively, have been investigated as one of the next-generation batteries because of the high theoretical capacity of sulfur ring S 8 (1675 mAh g −1 ) based on the 16-electron redox reaction (Equation (3)). [9,10] Nevertheless, the intermediate lithium polysulfides (Li 2 S 8 , Li 2 S 6 , and Li 2 S 4 ) were dissolved into organic electrolyte continuously and migrate between cathode and lithium metal anode during the discharge-charge process, which results in large, irreversible capacities because the lithium polysulfides react with the lithium metal anode (shuttle effect) [9,10]. To overcome the shuttle effect, lithium polysulfides should be converted into insoluble low-order sulfides (Li 2 S 3 and Li 2 S 2 ) before lithium polysulfides are dissolved into the electrolyte.
Recently, inverse vulcanization methods have been reported for the improvement of the cycle performance of sulfur-based cathode active materials. The conventional inverse vulcanization of organosulfides proceeds through the attack of sulfur radicals in the edge of the opened chain toward the electrophilic sites of organic moieties, such as carbon-carbon double bonds (C=C) in 1,3-diisopropenylbenzene (DIB) [11,12]. The organopolysulfide discharge products, such as DIB and styrene units, suppress the irreversible deposition of the insoluble lower-order lithium sulfides (Li 2 S 3 , Li 2 S 2 , and Li 2 S), and as a result, compared to elemental sulfur, the cycle performance is enhanced by the suppression of mechanical damage during discharge-charge process [13,14]. However, the π-organic moieties cannot contribute to enhancement of electronic conductivity of the organopolysulfides without conjugation between π-organic moieties and polysulfide chains. This results in capacities of organosulfides that are lower than the theoretical capacity of elemental sulfur. To overcome the low capacities, it is necessary to develop the synthesis of organopolysulfides in which polysulfide chains are directly bonded to π-organic moieties.
In this work, we report the battery performances of the organopolysulfides synthesized via inverse vulcanization using disulfide compounds from the viewpoint of inverse vulcanization (synthesis) and redox reaction (electronic conductivity). As shown in Scheme 1, the disulfide conjugated with π-organic moieties were cleaved easily by the attack of sulfur radicals, and the inverse vulcanization proceeded repeatedly [15,16]. This strategy is effective for increasing the loading of ring sulfur, meaning that the organopolysulfides with high capacities based on the redox reaction of the long sulfur chain were easily obtained. In addition, the electronic-conductive organopolysulfides were obtained by using electronicconductive 1,3-dithiol-2-thione [17,18]. Moreover, the hybridization between cathode active materials and conductive carbon materials, such as carbon nanotube (CNT) and graphene, has been reported as a method to improve the battery performance [19,20]. In this study, π-organic moieties interact with the π-conjugated ring of conductive carbon materials, suggesting that organopolysulfides would exhibit a more excellent battery performance. tube (CNT) and graphene, has been reported as a method to improve the battery perfor-mance [19,20]. In this study, π-organic moieties interact with the π-conjugated ring of conductive carbon materials, suggesting that organopolysulfides would exhibit a more excellent battery performance.

Nucler Magnetic Resonance (NMR)
Proton NMR ( 1 H NMR) and carbon NMR ( 13 C NMR) charts were recorded on BRUKER AVANCE III HD NMR spectrometers (Yokohama, Japan). 1 H NMR measurements were performed at 400 MHz using tetramethylsilane (TMS) and CDCl 3 as a standard material and solvent, respectively. Proton chemical shift values were recorded in parts per million (ppm, δ scale) downfield from TMS (δ 0) and CDCl 3 (δ 7.27). 13 C NMR charts were recorded at 101 MHz and also were performed using tetramethylsilane (TMS) and CDCl 3 as a standard material and solvent, respectively. Carbon chemical shift values are also reported in ppm (δ scale) downfield from TMS (δ 0) and CDCl 3 (δ 77). Data are presented as chemical shift, multiplicity (s = singlet and m = multiplet), and signal area integration in natural numbers.

Syntheses of Organopolysulfides
The organopolysulfides 3a and 3ab were synthesized via inverse vulcanization using 2a and 2b (Scheme 3). 3a was obtained as follows: Elemental sulfur (1.03 mg, 32.2 µmol) was added to a glass test tube equipped with a magnetic stir bar and heated to 120 • C in an aluminum block until a clear, orange-colored molten phase was formed. 2a was then directly added into the molten sulfur medium. The resulting mixture was stirred at 120 • C for 8-10 min, which resulted in vitrification of the reaction media. After allowing the reaction mixture to cool to room temperature, the product was taken directly from the test tube using a metal spatula. 3ab was obtained by the above inverse vulcanization method using 2a (469 mg) and 2b (581 mg).

Syntheses of organodisulfide polymers
The organodisulfide polymer 3'ab was obtained via two-step synthesis (Scheme 4). [21] First, p-toluene (15.2 g, 80 mmol) was added to BTBA-BDTD Zn complex (4.71 g, 5 mmol) in acetone (91.9 g). After stirring the solution at room temperature for 4 h, the yellow, crystalline precipitation 2'b (1.79 g, 3.5 mmol) was obtained by filtration and dried at room temperature. Second, the yellow, amorphous solid 3'ab was obtained by grinding 2'a (77.7 mg, 0.55 mmol) and 2'b (304.4 mg, 0.6 mmol) in a mortar for 5 min. 3'ab (116.1 mg, yield: 63.1%) was purified through extraction and silicagel chromatography using CH 2 Cl 2 , accompanied by evapolation. 3'ab: 1  To investigate polymerization of 3ab based on inverse vulcanization, TGA and DSC measurements were undertaken. The TGA measurement was carried out with TG8120 (Rigaku) between 27 and 307 • C at a heating rate of 10 • C min −1 under a flow of N 2 gas (20 mL min −1 ). Aluminum sample pans were used for the measurement. The amount of sample used in TGA measurement was 1.41 mg. As the reference, an empty aluminum pan was used. The DSC measurements were performed at a heating rate of 10 • C min −1 from -70 to 40 • C after cooling at 10 • C min −1 on a PerkinElmer DSC8500 under a flow of N 2 gas (20 mL min −1 ). Sealed aluminum sample pans and covers were used for the measurement. The amount of sample used in DSC measurement was 2.25 mg. As the reference, an empty aluminum pan was used.

Diffuse Reflectance Ultraviolet-Visible-Near-Infrared (UV-vis-NIR) Spectroscopy
To identify the aromatic moieties in 3ab, which was insoluble to organic solvents, diffuse reflectance UV-vis-NIR spectroscopy was performed using a UV-3600 UV-vis-NIR spectrophotometer (SHIMADZU) in the range of 300-800 nm wavelength with calcium sulfate as the standard. The samples, namely monomers (2a and 2b) or 3ab, were ground with calcium sulfate (WAKO) in agate mortars. The molar ratio of samples and calcium sulfate was 1:100. The spectra were obtained as average of 100 sequential measurements.

Battery Fabrication
The thin-film cathode was fabricated as follows. First, an organosulfide (3a, 3ab, or 3'ab) was added into carbon black (Toka black 5500, Tokai carbon), and it was ground in an agate mortar for 30 min. After that, polyvinylidene difluoride (PVDF, Kishida chemicals) binder (Kishida chemicals) was added and ground again for 1 h. The weight ratios of an organosulfide, carbon black, and PVDF were 30:50:20 and 70:20:10. The mixture was stirred in N-methylpyrrolidone. The prepared slurry was coated evenly onto aluminum foil (thickness: 20 µm) using a doctor blade technique and dried at room temperature under vacuum overnight at room temperature. The foil was cut into a disc with a diameter ϕ of 15.95 mm. CR2032 coin-type cells were assembled by using the cathode, lithium foil (ϕ: 15.50 mm) as an anode, polypropylene film separator, and 1 M LiClO 4 in TEGDME in an Ar-filled glovebox. Further, to investigate the effect of the π-π interaction between organic moieties in organopolysulfides and CNT [22], the cathodes including CNT were fabricated. The weight ratio of an organopolysulfide (3a or 3ab), carbon black, CNT, and PVDF was 30:40:10:20.

Results and Discussion
3'ab, composed of m-phenylene disulfide and dithiole-2-thione moieties, was characterized by 1 H and 13 C NMR measurements. As shown in Figure S7, the 1 H NMR chart of 3ab shows four peaks (δ 1.00-2.50) attributed to the methyl groups in the terminal thiol thiosulfonic acid ester ( Figure S1) and three peaks (δ 7.25-7.31) in the aromatic region. Moreover, the peaks in the 13 C NMR chart of 3ab were observed at the positions near to those of m-phenylene disulfide (2a, Figure S4) and dithiole-2-thione (2b, Figure S5). These indicate that polymerization of 2'a and 2'b occurred through formation of disulfide while maintaining the skeletons of m-phenylene disulfide and dithiole-2-thione. Although no 1 H and 13 C NMR data of 3ab and 3a were obtained due to their insolubility, the skeletons of m-phenylene disulfide and dithiole-2-thione in 3ab and 3a because of no nucleophilic attack of sulfide toward C=C and C=S bonds. In addition, the peak of S-S bonds at 470 cm −1 was observed in the Raman spectrum of 3ab [23], indicating that the inverse vulcanization proceeded while maintaining the skeletons of m-phenylene disulfide and dithiole-2-thione. Figure 1a,b show the TGA and DSC traces of 3ab. As shown in Figure 1a, the weight of 3ab decreased to 40% of the initial state between 120-300 • C, whereas the weight of sulfur dropped near the decomposition temperature of ring sulfur [24]. This means that sulfur chains in 3ab were decomposed, while the organic group remained. Furthermore, Figure 1b shows a DSC curve of 3ab measured at a heating rate of 10 • C min −1 . An endothermic baseline shift due to the glass transition was observed clearly. The glass transition temperature T g was~15.8 • C, which was determined as the temperature at the half height of the heat capacity difference ∆C p step. This indicates that 3ab had a polymer network composed of sulfur chain and organic moieties. Furthermore, Figure 1c shows that the peaks were observed in the UV-vis-NIR spectrum of 3ab at 331 and 460 nm. The one peak was near the peak in the spectrum of 2a (317 nm). The other was near the peak in the spectrum of 2b (460 nm). As shown in Figure S9a,b, the diffuse reflectance UV-vis-NIR spectra of 2a and 2b were broadened and similar to those of 2a and 2b in the CH 2 Cl 2 solution, indicating the two following points: (1) the aromatic moieties in 3ab can be detected by the diffuse reflectance UV-vis-NIR spectroscopy; (2) the spectra of 2a and 2b were broadened in a constrained environment, indicating that 3ab included m-phenylene and 1,3-dithiole-2-thione moieties. As mentioned above, it was found that 3ab was composed of sulfur chain, m-phenylene, and 1,3-dithiole-2-thione moieties. Figure 2a-c show the charge-discharge curves of 3a, 3ab, and 3'ab in the lithium batteries during 20 cycles. In the case of 3a and 3ab, short (2.5 V) and long (2.0 V) plateaus were observed. These plateaus indicated the initial S-S chain cleavage at 2.5 V accompanied by the subsequent S-S chain cleavage at 2.0 V, and as a result, 3a and 3ab exhibited high initial capacities of 750 and 1096 mAh g −1 , respectively. The initial capacities of 3a and 3ab were equal to those of previously reported conductive polymers with sulfur chains (Table S1) [25][26][27]. The reaction mechanism of 3ab was revealed by ex situ Raman spectra. As shown in Figure S10, the intensity of the broad peak around 424 cm −1 attributed to S-S bonds in 3ab was small because it was overlapped by that of carbon black. However, the peak around 424 cm −1 disappeared and appeared in the discharge and charge state, respectively, indicating that S-S bonds in 3ab were cleaved and recovered reversibly during the battery reaction. This indicates that 3ab exhibited a high capacity based on the reversible S-S bond cleavage/formation [23]. Furthermore, the battery including 70 wt% of 3ab exhibited a capacity of 712 mAh g −1 , which corresponds to 65.0% of the capacity of 30 wt% of 3ab ( Figure S11). In the case of 3'ab, the short plateaus were observed at 2.5 and 2.0 V, resulting in the small initial capacity of 309 mAh g −1 . This explains the two-electron redox reaction per sulfur atom of the sulfur chains in 3a and 3ab and the one-electron redox reaction per sulfur atom of the disulfide in 3'ab. The difference between the capacities of organopolysulfides (3a and 3ab) and those of the disulfide polymer 3a was caused by the length of S-S chain. However, the capacity retentions of organopolysulfides (3a: 59.3% and 3ab: 62.1%) were lower than 3'ab (68.9%) after 20 cycles. This indicates that high-order lithium polysulfides (Li 2 S x ) produced from the sulfur chains in 3a and 3ab dissolved into electrolyte during discharge process [6,7] although the disulfide in 3'ab showed a reversible redox reaction without more dissolution into electrolyte than 3a and 3ab. In addition, in comparison to 3a, 3ab exhibited the higher capacity because 1,3-dithiol-2-thione could improve the electrical conductivity, which induced a redox reaction of the sulfur chain. Despite the high capacity, the capacity retention of 3ab was lower than 3a because more high-order lithium polysulfide produced from 3ab was dissolved.   high-order lithium polysulfides (Li2Sx) produced from the sulfur chains in 3a and 3ab dissolved into electrolyte during discharge process [6,7] although the disulfide in 3'ab showed a reversible redox reaction without more dissolution into electrolyte than 3a and 3ab. In addition, in comparison to 3a, 3ab exhibited the higher capacity because 1,3-dithiol-2-thione could improve the electrical conductivity, which induced a redox reaction of the sulfur chain. Despite the high capacity, the capacity retention of 3ab was lower than 3a because more high-order lithium polysulfide produced from 3ab was dissolved. The battery performances of 3a and 3ab including CNT are shown in Figure 3a,b. Although 3a and 3ab including CNT exhibited higher initial capacities of 1082 and 1399 mAh g −1 compared to 3a and 3ab without CNT, 3a (661 mAh g −1 , 61.1%) and 3ab (622 mAh g −1 , 44.4%) showed capacity retentions nearly equal to those of 3a and 3ab without CNT. This indicates that the π-π interaction between CNT [22] and organic moieties in 3a and 3ab improved the electrical conductivities of 3a and 3ab, which induced a multi-electron redox reaction of the sulfur chain of 3a and 3ab. In addition, the π-π interaction between CNT and dithiole-2-thione moieties, which enhanced the conductivity of 3ab more effectively than that of 3a, might induce release of a larger amount of soluble LixS from 3ab, resulting in the shorter sulfur chain and the lower capacity of 3ab. However, the capacity retentions of 3a or 3ab including CNT were nearly equal to those of 3a or 3ab without CNT because organic moieties and CNT could not suppress the dissolution of Lewis basic lithium polysulfides into electrolyte. The battery performances of 3a and 3ab including CNT are shown in Figure 3a,b. Although 3a and 3ab including CNT exhibited higher initial capacities of 1082 and 1399 mAh g −1 compared to 3a and 3ab without CNT, 3a (661 mAh g −1 , 61.1%) and 3ab (622 mAh g −1 , 44.4%) showed capacity retentions nearly equal to those of 3a and 3ab without CNT. This indicates that the π-π interaction between CNT [22] and organic moieties in 3a and 3ab improved the electrical conductivities of 3a and 3ab, which induced a multi-electron redox reaction of the sulfur chain of 3a and 3ab. In addition, the π-π interaction between CNT and dithiole-2-thione moieties, which enhanced the conductivity of 3ab more effectively than that of 3a, might induce release of a larger amount of soluble Li x S from 3ab, resulting in the shorter sulfur chain and the lower capacity of 3ab. However, the capacity retentions of 3a or 3ab including CNT were nearly equal to those of 3a or 3ab without CNT because organic moieties and CNT could not suppress the dissolution of Lewis basic lithium polysulfides into electrolyte.
This indicates that the π-π interaction between CNT [22] and organic moieties in 3a and 3ab improved the electrical conductivities of 3a and 3ab, which induced a multi-electron redox reaction of the sulfur chain of 3a and 3ab. In addition, the π-π interaction between CNT and dithiole-2-thione moieties, which enhanced the conductivity of 3ab more effectively than that of 3a, might induce release of a larger amount of soluble LixS from 3ab, resulting in the shorter sulfur chain and the lower capacity of 3ab. However, the capacity retentions of 3a or 3ab including CNT were nearly equal to those of 3a or 3ab without CNT because organic moieties and CNT could not suppress the dissolution of Lewis basic lithium polysulfides into electrolyte.

Conclusions
In conclusion, the inverse vulcanization using electronic conductive disulfide compounds could produce the organopolysulfides with a higher capacity than the counterpart organodisulfide polymers. The sulfur/phenyl/1,3-dithiol-2-thione organopolysulfide exhibited a higher initial capacity than that of sulfur/phenyl copolymer-type organopolysulfide. Furthermore, CNT-hybridization could improve the capacity of organopolysulfides based on the interaction between π-organic moieties and the π-conjugated ring of CNT. However, capacity retentions of organopolysulfides were lower than that of organodisulfide polymer. This indicates that high-order lithium polysulfides produced from the sulfur chains during the discharge process dissolved into organic electrolyte, which caused the poor cycle performances [14]. To solve this problem, it is important to introduce Lewis acidic, such as iron phthalocyanine-based complexes [28,29], because it interacts with Lewis basic lithium polysulfides. Work addressing this challenge is currently underway to contribute to the development of next-generation rechargeable batteries with high capacities and excellent cycle performance.

Conclusions
In conclusion, the inverse vulcanization using electronic conductive disulfide compounds could produce the organopolysulfides with a higher capacity than the counterpart organodisulfide polymers. The sulfur/phenyl/1,3-dithiol-2-thione organopolysulfide exhibited a higher initial capacity than that of sulfur/phenyl copolymer-type organopolysulfide. Furthermore, CNT-hybridization could improve the capacity of organopolysulfides based on the interaction between π-organic moieties and the π-conjugated ring of CNT. However, capacity retentions of organopolysulfides were lower than that of organodisulfide polymer. This indicates that high-order lithium polysulfides produced from the sulfur chains during the discharge process dissolved into organic electrolyte, which caused the poor cycle performances [14]. To solve this problem, it is important to introduce Lewis acidic, such as iron phthalocyanine-based complexes [28,29], because it interacts with Lewis basic lithium polysulfides. Work addressing this challenge is currently underway to contribute to the development of next-generation rechargeable batteries with high capacities and excellent cycle performance.