Ternary Sulfur/Polyacrylonitrile/SiO2 Composite Cathodes for High-Performance Sulfur/Lithium Ion Full Batteries

In the present study, a novel sulfur/lithium-ion full battery was assembled while using ternary sulfur/polyacrylonitrile/SiO2 (S/PAN/SiO2) composite as the cathode and prelithiated graphite as the anode. For anode, Stabilized Lithium Metal Powder (SLMP) was successfully transformed into lithiated graphite anode. For cathode, scanning electron microscopy (SEM) and transmission electron microscopy (TEM) revealed that SiO2 was uniformly distributed on S/PAN composites, where SiO2 served as an effective additive due to its ultra high absorb ability and enhanced ability in trapping soluble polysulfide. The tested half-cell based on S/PAN/SiO2 composite revealed high discharge capacity of 1106 mAh g−1 after 100 cycles at 0.2 C. The full cell based on prelithiated graphite//S/PAN/SiO2 composite system delivered a specific capacity of 810 mAh g−1 over 100 cycles.


Introduction
Battery storage systems have been widely used in various applications, such as wind and solar energy storage, emergency back-up power, peak shaving, load-leveling, and transportation electrification [1,2]. Lithium-ion secondary batteries are widely used in mobile communications, notebook computers [3,4], digital cameras, and other small electronic devices due to several interesting features [5], including high energy storage density, long service life, and elevated rated voltage [6]. However, the energy densities of lithium-ion batteries do not currently meet the demand, hence the development of new energy storage systems becomes important [7][8][9][10][11][12]. Lithium/sulfur (Li/S) batteries have high theoretical energy densities reaching up 2600 Wh kg −1 , making them the most promising energy chemical power systems. In regard to this, elemental sulfur as cathode has low cost when combined with its environmental friendliness and high theoretical specific capacity (1672 mAh g −1 ) [13][14][15][16][17][18]. However, despite these criteria, the development of Li/S batteries still faces numerous challenges. Elemental sulfur is electrically insulating and the polysulfide generated during charging and discharging processes (Li+S 8 →Li 2 S x (4 ≤ x ≤ 8)) is highly soluble in electrolytes [19,20].
The above issues could be solved by adding various types of conductive carbon materials [21][22][23]. Also, some polymer and oxide additives might enhance the structural stability of composites and improve the conductivity of matrix materials. On the other hand, it is important to suppress the "shuttle effect" of polysulfides [24].
The S/PAN/SiO 2 and S/PAN electrodes were prepared by evenly grinding 80 wt % S/PAN/SiO 2 (S/PAN) composites, 10 wt % Super P, and 10 wt % PVDF (50 mm diameter, Shanghai Xingya, Shanghai, China). The mixture was then dropwise added to NMP to yield a slurry, which was coated on nickel foam and dried in 60 • C oven for 12 h. the obtained platforms were then cut into circular electrodes. The graphite anode was comprised of graphite, Super P, and PVDF at the mass ratio of 90:3:7. A surface application technique was employed to apply SLMP suspended in xylene slurry onto the prefabricated graphite anodes. An SLMP: graphite molar ratio of 11:60 was utilized to compensate for the irreversible capacity and lithiating the graphite. Upon solvent evaporation, the anode sheets were calendered using manual rolling mill at 3 MPa. The cathode loading of each cell was 2.5 mg cm −2 and the graphite active material loading was about 3 mg cm −2 .
The assembled 2025 button cell was completed in an argon-filled glove box. In half cells, a lithium chip was used as anode. In full cells, lithium was substituted by the lithiated graphite anode. The electrolyte was composed of 1M LiPF 6 (Li zhiyuan, Shanghai, China) solution dissolved in ethyl carbonate (EC), dimethylcarbonate (DMC), and diethyl carbonate (DEC) at volume ratio of 1:1:1. Coin cells were assembled and pre-conditioned for 24 h at room temperature. The galvanostatic charge/discharge tests were conducted at different current densities (1 C = 1672 mA g −1 ) and voltage of 1.0-3.0 V Li + /Li. Figure 1a represents the XRD patterns of S/PAN/SiO 2 composite. The characteristic Fddd orthorhombic crystal structure peaks of elemental sulfur vanished from the XRD patterns of S/PAN composite. This could be due to trapping of S in the internal structure of the composite to form highly dispersed state, with crystalline sulfur and PAN most likely being converted to amorphous sulfurized PAN by heat-treatment [10,20]. Comparison of the characteristic bands between S/PAN/SiO 2 and S/PAN composites revealed a broad peak at 23 • , which can be indexed to mixed peaks of S/PAN and amorphous SiO 2 [32,33]. The latter was probably induced by the dispersion of SiO 2 on the S/PAN surface and a slight shift in characteristic peaks of the ternary composites to the left. It also shows that SiO 2 and other components did not react during ball milling and heat treatment. Chemical analysis has shown that the sulfur content in the S/PAN/SiO 2 ternary composite was 45 wt %.

Results and Discussion
In order to determine the various functional groups in the S/PAN/SiO 2 composite, the FTIR analysis of the sample presents in Figure 1b. The characteristic peaks at 805, 1252, 1365 and 1502 cm −1 indicated the presence of C=C and C=N bonds in the composite [30]. The peaks at 882 cm −1 can be attributable to the S-S bonds, and those at 1046 and 1092 cm −1 can be assigned to the C-S stretching. The presence of the three peaks suggested that sulfur particles and PAN formed sulfurized-polyacrylonitrile [34]. The peak at 1116 cm −1 was associated with the asymmetric stretching vibration of Si-O-Si in SiO 2 [33]. The assembled 2025 button cell was completed in an argon-filled glove box. In half cells, a lithium chip was used as anode. In full cells, lithium was substituted by the lithiated graphite anode. The electrolyte was composed of 1M LiPF6 (Li zhiyuan, Shanghai, China) solution dissolved in ethyl carbonate (EC), dimethylcarbonate (DMC), and diethyl carbonate (DEC) at volume ratio of 1:1:1. Coin cells were assembled and pre-conditioned for 24 h at room temperature. The galvanostatic charge/discharge tests were conducted at different current densities (1 C = 1672 mA g −1 ) and voltage of 1.0-3.0 V Li + /Li. Figure 1a represents the XRD patterns of S/PAN/SiO2 composite. The characteristic Fddd orthorhombic crystal structure peaks of elemental sulfur vanished from the XRD patterns of S/PAN composite. This could be due to trapping of S in the internal structure of the composite to form highly dispersed state, with crystalline sulfur and PAN most likely being converted to amorphous sulfurized PAN by heat-treatment [10,20]. Comparison of the characteristic bands between S/PAN/SiO2 and S/PAN composites revealed a broad peak at 23°, which can be indexed to mixed peaks of S/PAN and amorphous SiO2 [32,33]. The latter was probably induced by the dispersion of SiO2 on the S/PAN surface and a slight shift in characteristic peaks of the ternary composites to the left. It also shows that SiO2 and other components did not react during ball milling and heat treatment. Chemical analysis has shown that the sulfur content in the S/PAN/SiO2 ternary composite was 45 wt %.

Results and Discussion
In order to determine the various functional groups in the S/PAN/SiO2 composite, the FTIR analysis of the sample presents in Figure 1b. The characteristic peaks at 805, 1252, 1365 and 1502 cm −1 indicated the presence of C=C and C=N bonds in the composite [30]. The peaks at 882 cm −1 can be attributable to the S-S bonds, and those at 1046 and 1092 cm −1 can be assigned to the C-S stretching. The presence of the three peaks suggested that sulfur particles and PAN formed sulfurizedpolyacrylonitrile [34]. The peak at 1116 cm −1 was associated with the asymmetric stretching vibration of Si-O-Si in SiO2 [33]. The BET specific surface area of S/PAN/SiO2 composite was estimated to 33.75 m 2 g −1 , which was significantly higher than those of S/PAN binary composites that were published previously [20,31]. Therefore, the surface area was increased significantly by adding the additive with a nano structure. This should facilitate the contact between the electrolyte and electrodes. Moreover,  (Figure 2c,d), resulting in a rough surface of the ternary composite. The BET specific surface area of S/PAN/SiO 2 composite was estimated to 33.75 m 2 g −1 , which was significantly higher than those of S/PAN binary composites that were published previously [20,31]. Therefore, the surface area was increased significantly by adding the additive with a nano structure. This should facilitate the contact between the electrolyte and electrodes. Moreover, the ternary composite contained many nano-sized particles, creating a three-dimensional (3D) porous nanostructure, it is beneficial to ion diffusion in Li/S battery and the SiO 2 suppresses the separation and agglomeration of active materials in the composite. The TEM image of S/PAN/SiO 2 composite were shown in Figure 2e,f, the "dark dots" were the amorphous SiO 2 particles well-dispersed in the composite bulk, which agrees well with the wide peaks in the XRD patterns. The as-prepared ternary composite was enabled to maintain the homogeneous distribution of its components and unchanged morphology during discharge-charge cycling, and retain the reactive sites in its nanosized structure [35].
Polymers 2018, 10, x FOR PEER REVIEW 4 of 9 the ternary composite contained many nano-sized particles, creating a three-dimensional (3D) porous nanostructure, it is beneficial to ion diffusion in Li/S battery and the SiO2 suppresses the separation and agglomeration of active materials in the composite. The TEM image of S/PAN/SiO2 composite were shown in Figure 2e,f, the "dark dots" were the amorphous SiO2 particles well-dispersed in the composite bulk, which agrees well with the wide peaks in the XRD patterns. The as-prepared ternary composite was enabled to maintain the homogeneous distribution of its components and unchanged morphology during discharge-charge cycling, and retain the reactive sites in its nanosized structure [35].  To further determine the chemical bonds in the S/PAN/SiO 2 composite, the XPS analysis of the sample is shown in Figure 3a-d. The C 1s, S 2p, and Si 2p peaks were all detected in the spectra. Figure 3b revealed the presence of a high-resolution peak of C 1s, as well as two peaks corresponding to C-C (284.6 eV) and C-N/C-S. The XPS C 1s spectra of S/PAN/SiO 2 composite confirmed the presence of a distinct peak at 285.8 eV, corresponding to the C-S/C-N bonds. This indicated the existence of certain chemical bonding between S and PAN in the S/PAN/SiO 2 composite. Figure 3c reveals the high-resolution S 2p peak. The S 2p 3/2 peak located at 161.4 and 163.3 eV were associated with C-S bond, attributed to single C-S bond and C-S bond in short-chain sulfide. The S 2p 1/2 peak at 164.6 eV was assigned to the S-S bond [21]. In XPS spectrum of Si 2p (Figure 3d), the peak at 104.8 eV was also present in the characteristic peak of SiO 2 [36]. These data were in accordance with the XRD patterns.
Polymers 2018, 10, x FOR PEER REVIEW 5 of 9 Figure 3b revealed the presence of a high-resolution peak of C 1s, as well as two peaks corresponding to C-C (284.6 eV) and C-N/C-S. The XPS C 1s spectra of S/PAN/SiO2 composite confirmed the presence of a distinct peak at 285.8 eV, corresponding to the C-S/C-N bonds. This indicated the existence of certain chemical bonding between S and PAN in the S/PAN/SiO2 composite. Figure 3c reveals the high-resolution S 2p peak. The S 2p3/2 peak located at 161.4 and 163.3 eV were associated with C-S bond, attributed to single C-S bond and C-S bond in short-chain sulfide. The S 2p1/2 peak at 164.6 eV was assigned to the S-S bond [21]. In XPS spectrum of Si 2p (Figure 3d), the peak at 104.8 eV was also present in the characteristic peak of SiO2 [36]. These data were in accordance with the XRD patterns. To figure out the electrochemical performances, the Li/S half cells were assembled while using lithium metal as anode, S/PAN/SiO2 composites as cathode, and 1M LiPF6 as electrolyte. Figure 4a shows the discharge/charge curves of S/PAN/SiO2 half batteries. A method of containment at the cathode is to provide additives in the cathode matrix that can attract and hold polysulfides so that they do not diffuse to the negative electrode. We utilized SiO2 as an additive to S/PAN electrode. The main interaction that the polysulfides have with the additive is through surface sorption, and therefore the surface area is increased significantly by synthesizing the additive with a nano structure. the SiO2 additive was able to sorb polysulfides during the intermediate discharge and release them near the end of discharge so that they could be further reduced in the S/PAN matrix with most of the sulfur being reversibly sorbed in the SiO2. The kinetics improvement and the polarization decrease achieved in the system by the addition of SiO2 could be very beneficial for the utilization of the lowconductive sulfur active material in the composite cathode, and consequently improved the energy and power density of the battery [20,30]. The initial discharge curve consisted of a rapid drop curve around 2.4 V and flat discharge curve around 1.6 V. The initial charge-discharge process was typical of lithium-sulfur battery reactions [10]. During the following two cycles, the discharge curve was mainly composed of two lines with different inclinations. The first part of the curve looked relatively flat (from 2.2 V to 1.6 V) and the second was steep (from 1.6 V to 1.0 V). The two curves indicated the To figure out the electrochemical performances, the Li/S half cells were assembled while using lithium metal as anode, S/PAN/SiO 2 composites as cathode, and 1M LiPF 6 as electrolyte. Figure 4a shows the discharge/charge curves of S/PAN/SiO 2 half batteries. A method of containment at the cathode is to provide additives in the cathode matrix that can attract and hold polysulfides so that they do not diffuse to the negative electrode. We utilized SiO 2 as an additive to S/PAN electrode. The main interaction that the polysulfides have with the additive is through surface sorption, and therefore the surface area is increased significantly by synthesizing the additive with a nano structure. the SiO 2 additive was able to sorb polysulfides during the intermediate discharge and release them near the end of discharge so that they could be further reduced in the S/PAN matrix with most of the sulfur being reversibly sorbed in the SiO 2 . The kinetics improvement and the polarization decrease achieved in the system by the addition of SiO 2 could be very beneficial for the utilization of the low-conductive sulfur active material in the composite cathode, and consequently improved the energy and power density of the battery [20,30]. The initial discharge curve consisted of a rapid drop curve around 2.4 V and flat discharge curve around 1.6 V. The initial charge-discharge process was typical of lithium-sulfur battery reactions [10]. During the following two cycles, the discharge Polymers 2018, 10, 930 6 of 10 curve was mainly composed of two lines with different inclinations. The first part of the curve looked relatively flat (from 2.2 V to 1.6 V) and the second was steep (from 1.6 V to 1.0 V). The two curves indicated the available capacity of about 1266 mAh g −1 . Figure 4b confirmed a coulombic efficiency of binary and ternary composite electrode half cells close to 100%. However, the S/PAN/SiO 2 half-cell showed a higher specific discharge capacity, which can be attributed to added nanosized SiO 2 . Moreover, the S/PAN/SiO 2 electrode maintained a discharge capacity of 1106 mAh g −1 at 0.2 C after 100 cycles. The S/PAN/SiO 2 half-cell exhibited specific discharge capacities of 1268, 969, 806, and 622 mAh g −1 at current densities of 0.2, 0.5, 1 and 2 C, respectively (Figure 4c). By comparison, the S/PAN half-cell was also tested under the same conditions and the data are gathered in Figure 4d. Obviously, the cell with the S/PAN/SiO 2 composite cathode showed an enhanced rate capability. This is, again, due to the significant improvement of the charge transfer properties of the composite cathode and its stability by the SiO 2 additive that was observed in this work [37]. available capacity of about 1266 mAh g −1 . Figure 4b confirmed a coulombic efficiency of binary and ternary composite electrode half cells close to 100%. However, the S/PAN/SiO2 half-cell showed a higher specific discharge capacity, which can be attributed to added nanosized SiO2. Moreover, the S/PAN/SiO2 electrode maintained a discharge capacity of 1106 mAh g −1 at 0.2 C after 100 cycles. The S/PAN/SiO2 half-cell exhibited specific discharge capacities of 1268, 969, 806, and 622 mAh g −1 at current densities of 0.2, 0.5, 1 and 2 C, respectively (Figure 4c). By comparison, the S/PAN half-cell was also tested under the same conditions and the data are gathered in Figure 4d. Obviously, the cell with the S/PAN/SiO2 composite cathode showed an enhanced rate capability. This is, again, due to the significant improvement of the charge transfer properties of the composite cathode and its stability by the SiO2 additive that was observed in this work [37]. As shown in Figure 5, the prelithiated graphite//S/PAN/SiO2 composite system exhibited an initial capacity of 804 mAh g −1 and specific capacity of 810 mAh g −1 after 100th cycle. Hence, extremely low capacity decay rate was registered. Figure 5a depicts the coulomb efficiency of the full cell, which was close to 100%. The cycling curve of the full cell indicated a slight decrease from the initial cycle up to the 10th cycle then slowly increased up to 100th cycle. Although the amorphous S/PAN can provide good mechanical support for SiO2 spheres, the structural stability cannot be guaranteed due to repeated volume changes during cycling. This led to decrease in capacity. Meanwhile, the added amorphous nano-sized SiO2 absorbed more electrolyte and polysulfide on the surface of rough ternary composite. The discharge capacity also gradually increased in subsequent cycles, which is may due to that the pre-lithiated graphite was not fully lithiated, meaning that the electrode underwent slow activation. When the electrode was in contact with the electrolyte, the SLMP in the anode released lithium ions, making it slow to activate. As the battery cycles increases, the anode side of the graphite was completely lithiated, and the battery exhibited a gradual increase in discharge specific capacity [28,29,38].
As shown in Figure 5, the prelithiated graphite//S/PAN/SiO 2 composite system exhibited an initial capacity of 804 mAh g −1 and specific capacity of 810 mAh g −1 after 100th cycle. Hence, extremely low capacity decay rate was registered. Figure 5a depicts the coulomb efficiency of the full cell, which was close to 100%. The cycling curve of the full cell indicated a slight decrease from the initial cycle up to the 10th cycle then slowly increased up to 100th cycle. Although the amorphous S/PAN can provide good mechanical support for SiO 2 spheres, the structural stability cannot be guaranteed due to repeated volume changes during cycling. This led to decrease in capacity. Meanwhile, the added amorphous nano-sized SiO 2 absorbed more electrolyte and polysulfide on the surface of rough ternary composite. The discharge capacity also gradually increased in subsequent cycles, which is may due to that the pre-lithiated graphite was not fully lithiated, meaning that the electrode underwent slow activation. When the electrode was in contact with the electrolyte, the SLMP in the anode released lithium ions, making it slow to activate. As the battery cycles increases, the anode Polymers 2018, 10, 930 7 of 10 side of the graphite was completely lithiated, and the battery exhibited a gradual increase in discharge specific capacity [28,29,38].
When compared to other related work ( When compared to other related work (Table 1), our S/PAN/SiO2 composite exhibited superior electrochemical performance.

Conclusions
Sulfur/lithium-ion full batteries were assembled while using ternary S/PAN/SiO2 composite as cathode and a prelithiated graphite as anode. SLMP was successfully applied to lithiation of graphite anode. Nano-sized SiO2 was found to be uniformly distributed on S/PAN composites, which served as an effective additive due to its ultra high absorbtion ability and enhanced trapping soluble polysulfide. The S/PAN/SiO2 composite cathode half-cell showed a high discharge capacity of 1106 mAh g −1 after 100 cycles at 0.2 C. The pre-lithiated graphite//S/PAN/SiO2 composite full cell system delivered a specific capacity of 810 mAh g −1 over 100 cycles. These findings look promising for future use in energy conversion and storage devices.

Conclusions
Sulfur/lithium-ion full batteries were assembled while using ternary S/PAN/SiO 2 composite as cathode and a prelithiated graphite as anode. SLMP was successfully applied to lithiation of graphite anode. Nano-sized SiO 2 was found to be uniformly distributed on S/PAN composites, which served as an effective additive due to its ultra high absorbtion ability and enhanced trapping soluble polysulfide. The S/PAN/SiO 2 composite cathode half-cell showed a high discharge capacity of 1106 mAh g −1 after 100 cycles at 0.2 C. The pre-lithiated graphite//S/PAN/SiO 2 composite full cell system delivered a specific capacity of 810 mAh g −1 over 100 cycles. These findings look promising for future use in energy conversion and storage devices.