Sustainable Sulfur-rich Copolymer/Graphene Composite as Lithium-Sulfur Battery Cathode with Excellent Electrochemical Performance

A sulfur-rich copolymer, poly(S-r-C-a) has been synthesized via a sustainable route, showing the utility of two major industrial wastes- elemental sulfur (petroleum waste) and cardanol (agro waste), to explore its potential as cathode material for Li-S batteries. The sulfur-rich copolymer exhibited a reduction in the active material dissolution into the electrolyte and a low self-discharge rate behavior during the rest time compared to an elemental sulfur cathode, indicating the chemical confinement of sulfur units. The presence of organosulfur moieties in copolymer suppress the irreversible deposition of end-discharge products on electrode surfaces and thus improve the electrochemical performances of Li-S batteries. This sulfur copolymer offered a reversible capacity of 892 mA h g−1 at 2nd cycle and maintained the capacity of 528 mA h g−1 after 50 cycles at 200 mA g−1. Reduced graphene oxide (rGO) prepared via a sustainable route was used as a conductive filler to extract the better electrochemical performances from this sulfur copolymer. Such sustainable origin batteries prepared via economically viable showed an improved specific capacity of ~975 mA h g−1 after 100 cycles at 200 mA g−1 current rate with capacity fading of 0.15% per cycle and maintained a stable performance over 500 cycles at 2000 mA g−1.

dispersion of reduced graphene oxide throughout the sulfur copolymer, resulting to the excellent electrochemical performances. For comparison, reduced graphene oxide was also synthesized via a non-sustainable hydrazine route (abbreviated as rGO (NS) ) and used as conductive additive in S90 copolymer cathode. Figure 1a represents the synthetic scheme of cardanol benzoxazine monomer C-a and the preparation of S90 copolymer via inverse vulcanization route. The monomer C-a was synthesized by Mannich-like condensation reaction of aniline with cardanol. Elemental sulfur exhibited melting at ~130 °C followed by free radical ring opening polymerization to form linear polysulphane diradicals, which were trapped by C-a. The growing radicals reacted at the alkylene chain and benzoxazine ring 50 of C-a to form S90 copolymer. Proton nuclear magnetic resonance ( 1 H-NMR) spectrum of S90 copolymer ( Figure S1) confirmed the absence of any unreacted C-a monomer due to disappearance of signals of benzoxazine ring and double bonds (4.5-5.5 ppm) and simultaneously appearance of signals due to methylene hydrogens in Mannich bridge Ar-CH 2 -N and N-CH 2 -S at 3.5-4.5 ppm indicative of copolymerization. In addition, a relative shift in signals and disappearance of protons in aliphatic region (2.5-3.0 ppm) was also observed. Raman spectroscopy ( Figure S2) of S90 copolymer also showed the signatures 500-800 cm −1 and 2800-3100 cm −1 corresponding to S-S, C-S and C-H linkages confirming the formation of Scientific RepoRts | 6:25207 | DOI: 10.1038/srep25207 copolymer 51 . The stability of the S90 copolymer [weight average molecular weight of 5.3 × 10 3 g mol −1 and a polydispersity index (PDI) of 3.7] in solution was confirmed by performing gel permeation chromatography (GPC) analysis ( Figure S3) at different periods of storage (0 th and 17 th day). The GPC traces clearly suggest the stability of the polymer architecture under solvated conditions. The S90 copolymer exhibited two melting endotherms at 104 and 116 °C ( Figure S4) as determined using differential scanning calorimetry (DSC) suggesting the existence of different crystalline domains in the polymer architecture.

Results and Discussion
GO was prepared using modified Hummer's method which was then reduced to rGO using two alternative synthetic methodology, namely gallic acid, a natural occurring phenolic acid reductant and hydrazine hydrate (Fig. 1b) to form rGO (S) and rGO (NS) respectively. FTIR spectra of GO, rGO (S) and rGO (NS) are shown in Figure S5. The peaks centered at 3100, 1718, 1615, 1350, 1220, 1050 cm −1 in the FT-IR spectra of GO are attributed to hydroxyl, carboxyl, carbonyl, aromatic C = C, epoxy, ester stretching vibrations respectively. Both rGO (S) and rGO (NS) showed the absence of oxygenated functionalities peaks suggested successful reduction. The spectra of rGO (S) showed no IR frequencies corresponding to gallic acid suggesting its absence and purity of the synthesized reduced graphene prepared via sustainable route. Raman spectra of rGO (S) and rGO (NS) (Fig. 1c) showed a strong band at 1589 and 1595 cm −1 (G band) corresponding to the first order scattering of the E 2g phonon mode of sp 2 bonded C atoms and a weak band at 1348 and 1354 cm −1 (D band) arising from breathing mode of A 1g symmetry at K point. The I D /I G intensity ratio was calculated as 1.42 rGO (S) and 1.81 rGO (NS) suggesting the development of more sp 2 structural domains in the former with appreciable synthetic viability. In addition to D and G peaks, some other peaks are clearly visible at 2688/2711 (2D-mode) and 2933/2941 (D + G-mode) for rGO (S) /rGO (NS) as shown in Fig. 1c (inset). In particular, the 2D and G bands represent the key features of Raman spectroscopy for the identification of graphene sheets. The 2D band peak at 2690 cm −1 is due to the highest optical branch phonons near the K point at the Brillouin zone boundary and the G band peak at 1575 cm −1 is due to the two-fold degenerate E 2g mode at the C-point. The relative intensity ratio of I 2D /I G can be used to distinguish the number of layers of graphene sheets. The integral intensity ratio I 2D /I G of > 2, 1-2, and < 1 correspond to single-layered, double-layered and many-layered graphene, respectively. With increasing number of layers, the intensity ratio of I 2D /I G decreases whilst the FWHMs of the 2D peaks increases. For a good comparison, the spectra were normalized to give approximately the same relative intensity for all the samples. The I 2D /I G intensity ratio is less than 1 verifying existence of many layered graphene sheet and no single-layered graphene sheet was identified by the Raman spectroscopy 52 .
To understand the microstructural properties of bulk rGO prepared via sustainable route (i.e., rGO (S) ) scanning electron microscopic (SEM) as well as transmission electron microscopic (TEM) images were taken. Figure 2 represented the morphological features of rGO prepared via sustainable route. Both SEM (Fig. 2a) and TEM ( Fig. 2b) images, showed the presence of wrinkles, ripples and scrolls in the rGO (S) suggesting the occurrence of few-layered graphene sheets. Furthermore, selected area electron diffraction (SAED) patterns of GO and rGO (S) (Fig. 2c,d) were compared to understand the successful reduction. A diffused diffraction ring pattern typical of a disordered structure is observed in GO illustrating the disruption of conjugated structure of graphite due to chemical oxidation. However, SAED image of rGO (S) showed set of six-fold symmetric diffraction points of a typical hexagonal configuration 53 . This illustrates the formation of crystalline states in rGO (S) representing the restoration of aromatic graphene structure and successful chemical reduction of GO using gallic acid. The thickness of the rGO (S) was determined using atomic force microscopy (AFM) as shown in Fig. 3. The as-synthesized rGO (S) consisted mainly of a corrugated surface with a height range between 18-26 nm nanometers and a lateral size of several micrometers. AFM analysis identified the absence of single-layered graphene (less than 1 nm) domains 53 . The existence of multi-layered graphene architecture in rGO (S) was also confirmed by TEM, SEM and Raman analysis. Figure 4a shows SEM image of S90-2.5% rGO (S) composite. A zoom-in SEM image (Fig. 4b) shows the uniform wrapping of rGO (S) on sulfur copolymer surface. To verify the composition of our S90-2.5% rGO (S) composite energy dispersive X-ray spectroscopic (EDX) characterization and elemental mapping of our material were carried out. From EDX spectra and elemental mapping of S90-2.5% rGO (S) (Fig. 4c-f), uniform distribution of sulfur atoms throughout S90 copolymer and conformal wrapping of graphene sheets on the copolymer were confirmed.
The electrochemical characterizations of S90 copolymer were evaluated using 2016 coin cells with lithium foil as anode. Figure 5a, represents the low scan rate (20 μ V s −1 ) CV curve of bulk S90 copolymer cathode for the first cycle showing three reduction peaks and one oxidation peak. The reduction peak at 2.35-2.4 V is attributed to the conversion of sulfur to higher order linear polysulfides. The peak at 2.0-2.05 V is related to the further reduction of higher order polysulfides to lower order polysulfides, indicating multiple reaction feasibility of sulfur with lithium ions. The third peak in the potential range of 1.65-1.7 V is attributed to the irreversible reduction of LiNO 3 additive in the electrolyte 54 . In addition, only one oxidation peak was observed at 2.3-2.35 V in the subsequent charge process, corresponding to the transformation of lower ordered polysulfides to higher ordered polysulfides. To study the lithium ion storage capacity of the bulk 90 copolymer, charge/discharge performances were performed within the potential window of 1.5-2.6 V at a current density of 200 mA g −1 . Figure 5b illustrates the initial discharge profile of the S90 cathode at 200 mA g −1 . As it is obvious that the positions of charge/ discharge plateaus exactly corresponds to the peaks obtained during cyclic voltammetry, the initial discharge profile also exhibits three discharge plateaus at around 2.28, 2.0 and 1.7 V, respectively. Figure 5c demonstrates the charge-discharge profiles corresponding to 1 st , 2 nd , 10 th and 20 th cycles of electrode-A at 200 mA g −1 rate. Figure 5d shows the cycling performance of S90 cathode at the current density of 200 mA g −1 . It is notable that there is a huge capacity loss at the 2 nd cycle. This extra capacity during the first discharge of the electrodes comes from the irreversible reduction of LiNO 3 at the potential range of 1.5-1.7 V to form solid electrolyte interphase (SEI), resulting in low coulombic efficiency of the first cycle (ca. 85%). However, from 2 nd cycle onwards the coulombic efficiency was > 98.5%. S90 electrode exhibited a reversible discharge capacity of 892 mA h g −1 at 2 nd cycle which degrades rapidly in first 20 cycles and becomes stable after 30 cycles. After 50 cycles the S90 copolymer delivers a discharge capacity of 528 mA h g −1 . The initial capacity fading for S90 copolymer could be explained based on the low electronic conductivity and polymeric nature of the sample that may block both electronic as well as ionic pathway resulting to the poor Li + ions and electrons accessibility towards active sulfur in S90 copolymer. However, the S90 electrode exhibited better cycling stability compare to the electrode made from pure sulfur. The better cycling performance of S90 copolymer could be due to the chemical attachment of sulfur atoms/chains with organic part (i.e., Bzc) which can reduce active sulfur dissolution into electrolyte. In addition, the "plasticizing effect" of lower ordered organosulfides which get formed along with lower order lithium polysulfides (Li 2 S 3 / Li 2 S 2 ) at lower potential during initial discharge of sulfur copolymer 14 could also be another reason for improved performances of S90 copolymer.
However, to achieve better electrochemical performances from this copolymer, conductive fillers with excellent conductivity (like: carbon nanotubes or graphene) need to incorporate in the S90 copolymer before using as cathode active material. Graphene has been used as a conductive filler in this work due to its extremely high surface area and high electronic conductivity which can easily create even better percolation network throughout the copolymer with its minimum presence. A reduced graphene oxide was synthesized via green and sustainable route (rGO (S) ) using gallic acid as reductant as well as stabilizer, and 2.5% (by mass) of it was dispersed into the copolymer. For comparison, the same amount of reduced graphene oxide synthesized though "non-sustainable" hydrazine process (rGO (NS) ) was dispersed into the S90 copolymer and used as cathode material for Li-S batteries. Figure 6a displays the CV curves for S90-2.5% rGO (S) composite electrode. The almost equal area of first five curves certifies the excellent electrochemical stability and reversibility of the S90-2.5% rGO (S) composite. The S90-2.5% rGO (S) composite exhibited a reversible capacity of 1156 mA h g −1 at 2 nd cycle at 200 mA g −1 rate (Fig. 6b). In terms of active material utilization rate at the 2 nd cycle, the S90-2.5% rGO (S) composite had a higher utilization rate of 69.14% (1156 mA h g −1 ) of theoretical value than that of bulk S90 copolymer at 53.35% (892 mA h g −1 ). A reversible capacity up to 975 mA h g −1 was retained after 100 cycles, which is much higher than that of conventional sulfur-2D graphene composites (100 th cycle capacity = 910 mA h g −1 , current rate = 167.2 mA g −1 , sulfur loading = 40% 55 or 100 th cycle capacity = 476 mA h g −1 , current rate = 167.2 mA g −1 , sulfur loading = 57.5% 56 ), sulfur-3D graphene composites (100 th cycle capacity = 700 mA h g −1 , current rate = 167.2 mA g −1 , sulfur loading = 73% 57 ) and the cathode made of electrodeposited sulfur onto graphene sheets (900 mA h g −1 after 60 cycles at 100 mA g −1 , sulfur loading = 70.2% 58 ). The superior performances of the S90-2.5% rGO (S) composite material in terms of improved specific capacity and improved capacity retention is presumably caused by the use of high electronic conductivity and high surface area of rGO (S) which induces an improved percolation network throughout S90 copolymer, facilitating Li + ion transport through the electrode, as a result of which Li + ions accessibility towards active sulfur increases. However, from Fig. 6b it is visible that S90-2.5% rGO (S) composite exhibited almost same performance as S90-2.5% rGO (NS) composite over 100 cycles, indicating that the similar type of percolation network as S90-2.5% rGO (NS) composite was achieved for S90-2.5% rGO (S) composite. Figure 6c, represents the cycling performances of S90-2.5% rGO (S) and S90-2.5% rGO (NS) composites at 1000 mA g −1 current rate. Both the composites exhibited an initial reversible capacity around 810 mA h g −1 . A reversible capacity up to 700 mA h g −1 (86.42% of initial capacity) was retained after 300 cycles. It can be observed from Fig. 6d that the rate capability of S90 copolymer was also increased upon graphene incorporation. The corresponding charge-discharge profiles of S90-2.5% rGO (S) at different current rates are shown in Figure S6. From Fig. 6d, it can be observed that S90-2.5% rGO (S) composite exhibited almost similar rate performance as S90-2.5% rGO (NS) . However, to gain insight into the reason why both S90-2.5% rGO (S) and S90-2.5% rGO (NS) composites exhibited such similar electrochemical performances, the kinetics of both S90-2.5% rGO (S) and S90-2.5% rGO (NS) composite cathodes have been investigated through in-situ electrochemical impedance spectroscopy (EIS) measurements. The electrochemical impedance spectroscopy results for both the cells cycled over 100 cycles were compared. Generally, an EIS spectra is composed of a high-frequency semicircle related to the charge transfer process and low frequency straight line corresponding to a semi-infinite Warburg diffusion process. Both the composites exhibited almost identical sized semi-circle at low frequency region ( Figure S7), indicating almost same amount of charge transfer resistance (R ct ), as expected. The long term cycling performance of S90-2.5% rGO (S) composite at 2000 mA g −1 current rate after the initial five cycles of activation at 200 mA g −1 is shown in Fig. 6e.
Considering the view point from commercialization of Li-S battery in future, low self-discharge behavior (or long shelf-life behavior) of the cell is necessary. Most commonly, LiNO 3 is used as an electrolyte additive to protect the lithium anode (by forming a stable solid electrolyte interface) from higher order polysulfides and dissolved sulfur molecules, thus preventing the self-discharge of Li-S batteries 59 . However, to investigate the answer of the question, "can the active sulfur dissolution in Li-S battery be suppressed through chemical confinement strategy?", self-discharge studies have been carried out for both S90 copolymer and elemental sulfur cathode using an electrolyte containing no LiNO 3 additive. Initially, both the cells were cycled for 5 cycles and then kept in rest for 120 h. The cells were charged to 2.6 V before keeping in rest. After 120 h of rest time, the cells were discharged as usual and recharged to 2.6 V. Then they were further cycled for another 20 cycles. From Fig. 7a, it can be observed that both the electrodes showed stable cycling performances for initial five cycles. During rest period the open circuit voltage (OCV) gradually decreases for both the cells. After 120 h rest time, OCV became very stable at 2.33 V for S90 copolymer cell, while for pure sulfur cell it was 2.20 V (Fig. 7b). After 120 h rest time, the S90 copolymer cathode exhibited two distinct discharge plateaus, while the pure sulfur cathode showed only lower discharge plateau (since the OCV for pure sulfur cathode was less than the potential at which the reactions corresponding to upper discharge plateau usually take place). This observation manifests that the self-discharge rate for S90 copolymer electrode was significantly lower compared to elemental sulfur cathode. The rapid fall in OCV and disappearance of upper discharge plateau for pure sulfur cathode can be related to the formation of lower order polysulfides (i.e., Li 2 S 3 /Li 2 S 2 ) from the reaction between rapidly dissolved unbound sulfur and bare lithium 60 . In brief, the dissolved S 8 molecules migrate towards anode and chemically reacts with unprotected lithium metal to produce initial discharge product Li 2 S 8 . This unstable Li 2 S 8 further reacts with lithium to form more stable lower order polysulfides. On other hand, due to chemical attachment of sulfur with the organic part in S90 copolymer, the dissolution of sulfur for S90 cathode was less compare to pure sulfur cathode. Figure 7c shows the discharge profiles of S90 copolymer electrode for 1 st , 5 th (before rest), 6 th (after rest), and 7 th cycles. From Fig. 7c, it is noteworthy to mention that there was 98.6% retention of 5 th cycle (the cycle before rest time) discharge capacity (458 mA h g −1 ) at 7 th cycle (the cycle after rest time, discharge capacity = 452 mA h g −1 ) for  Fig. 7d reveals that for S90 electrode the 6 th cycle charge capacity was very much higher (447 mA h g −1 , almost same as next cycle discharge capacity) as compare to the discharge capacity of the same cycle (the discharge after rest time). To investigate the reason behind these observations, electrochemical impedance measurements were taken for both pure sulfur and bulk S90 copolymer electrodes at different stages. Both the electrodes made of pure sulfur and S90 copolymer, exhibited almost same charge transfer resistance (R ct ) at their open circuit voltage (OCV) before charge-discharge was performed (Fig. 7e). A remarkable difference in R ct was observed when EIS were taken for both the cells after 120 h of rest time (Fig. 7f). The S90 cathode showed very much lower charge transfer resistance (R ct = 97 Ω) than pure sulfur cathode (R ct = 135 Ω). The lower charge transfer resistances for S90 electrode after 120 h of rest time could be due to the presence of organosulfur units in S90 copolymer. During initial electrochemical discharge, the organosulfur units in the copolymer get reduced to form lower order lithium polysulfides (Li 2 S 2 /Li 2 S) and lithium organosulfides 22 . These lithium organosulfides then co-deposited within the matrix of insoluble/insulating lower order lithium polysulfides and react to form higher order lithium organosulfides which is soluble into electrolyte. Thus the organosulfur units present in copolymer acts as a "plasticizers" in the insoluble/insulating Li 2 S/Li 2 S discharge product phase, preventing their irreversible deposition on electrode surfaces (during cycling or during rest time) to reduce cell resistance as well as capacity fading. To interrogate the "plasticizing effect" of organosulfur units on the deposition of Li 2 S 3 /Li 2 S 2 on the electrode surfaces, morphological characterizations of the electrodes after cycling (at fully charged state) were also conducted. From Figure S8, dense discharge products (Li 2 S 3 /Li 2 S 2 ) could be observed on the surface of elemental sulfur electrode. These insoluble and insulating products, formed from deep discharge of elemental sulfur, "paste" on sulfur electrode which not only lead to the blockage of ionic/electronic transport, but also lead to loss of active material, both of which could result in the capacity fading. In contrast, the S90 copolymer electrode showed no such type of insulating product deposition. Hence the regeneration of capacity was observed at 6 th charge (Fig. 7d) for S90 copolymer while for elemental sulfur no such capacity retention was observed after 6 th cycle of charge-discharge. Therefore, the presence of organosulfur units in the sulfur copolymer could significantly suppress the irreversible deposition of insulated end discharge products on the electrode surface during cycling as well as during the rest time.
In summary, a novel sulfur-rich (90% sulfur) copolymer has been synthesized via a facile, sustainable and scalable process utilizing two industrial waste-sulfur, a petroleum waste and cardanol, an agro waste. One of the monomers, cardanol benzoxazine (C-a, the organic part of copolymer) was extracted from cardanol and synthesized by a solventless, one pot and atom economized process with water as the only by-product of the reaction. The solubility of the copolymer allowed structural characterization using solution-based techniques. We Scientific RepoRts | 6:25207 | DOI: 10.1038/srep25207 demonstrate that the chemical attachment of sulfur with the organic part (C-a) and the presence of organosulfur units in this copolymer can significantly compensate the negative aspects of elemental sulfur to improve the Li-S battery performances without the need of any nanoscopic synthesis. However, the large structural property of the organic part in the copolymer impeded it to perform the best electrochemical activities. Therefore, in addition reduced graphene oxide (rGO) has also been prepared in a green, scalable route utilizing a naturally occurring compound, gallic acid and used as conductive agent to get the better electrochemical performances. Finally, considering the synthetic approaches for sulfur copolymer and reduced graphene oxide and the intriguing features of their composite cathode in Li-S battery, we believe that this work can provide a valuable insight in the utilization of wastes as energy storage materials and the usage of sustainable carbon based nanomaterial as a conductive additive without compromising on cost or performance. Methods Materials. Cardanol was procured from Satya Cashew Chemicals Pvt. Ltd. (India), paraformaldehyde, sulfuric acid, sodium nitrate, potassium permanganate, hydrogen peroxide, hydrazine hydrate, aniline and liquor ammonia from Fisher scientific, gallic acid from Sigma Aldrich, chloroform from Finar, anhydrous sodium sulphate from Chemlabs, graphite flakes and sulfur (S 8 , 325 mesh, reagent grade) from Alfa Aesar. All the reagents were used as received.

Synthesis of sulfur-rich copolymer, poly (sulfur-random-cardanol benzoxazine) (S90).
Elemental sulfur (1.8 g) was added to a 15 mL glass vial equipped with a magnetic stir bar and heated to a temperature of 185 °C in an oil bath equipped with thermostat to obtain a clear orange coloured viscous molten phase. Cardanol benzoxazine (C-a) (0.2 g) was then added directly to the molten sulfur via a syringe and this mixture was stirred further at the temperature of 185 °C for 10 minutes. Mixture was cooled to room temperature slowly and poly(S-r-C-a) (S90) was obtained in quantitative yields.
Synthesis of Graphene Oxide (GO). GO (Graphite oxide) was prepared by a modified Hummer's method 61 . Briefly, graphite flakes (2.0 g) were added to concentrated sulfuric Acid (46.0 mL) and sodium Nitrate (1.0 g) while keeping the temperature at 0 °C with continuous stirring. This was followed by a gradual addition of potassium permanganate (6.0 g) while maintaining the temperature below 20 °C. Upon completion of addition, ice bath was removed and resultant suspension was stirred. The solution turned a brownish gray in 30 minutes after which distilled water (92.0 mL) was added to the paste. The solution was finally treated with warm water (280.0 mL) and hydrogen peroxide (3%) to give brown colored GO. Hence, the formed GO was washed with water (8 × 500 mL) to remove all the acidic content and dried under vacuum to obtain yellowish brown solid (1.2 g, 60%).

Synthesis of reduced Graphene Oxide (rGO).
As prepared graphene oxide (GO) was reduced following a green method using gallic acid as reducing agent instead of hydrazine. GO (260 mg) was dispersed in deionized water (65 mL) by sonication. Liquor Ammonia (25%) was added to this aqueous dispersion to adjust the pH value of the mixture to 12. Gallic acid (2.607 g, 15.3 mmol) was then added to the mixture followed by stirring at 95 °C for 10 h in a nitrogen atmosphere. rGO was filtered and subjected to washing with deionized water followed by acetone. Hence, obtained product was dried and weighed to give as black solid (150 mg, 58%).
Synthesis of S90-rGO composite. S90 (1.46 g) was well-dispersed in THF (30 mL) by ultra-sonication for 1 h. This was followed by addition of rGO (37.5 mg) and resulting suspension was sonicated for 2 h for efficient blending. The solvent was removed under vacuo to give a gray powder in quantitative yield.
Characterizations. Fourier Transform Infrared (FT-IR) spectra were recorded on a Nicolet iS5 spectrometer equipped with iD5-ATR accessory, in the range of 4000 to 400 cm −1 . Raman spectra were obtained using a JobinYvon HR800 Raman Microscope at 514.5 nm. Bruker AC 500 MHz FT-NMR spectrometer was used to record the 1 H-NMR of the samples in CDCl 3 using tetramethylsilane as the internal standard. Agilent HRMS 6540 Series Q-TOF was used for mass spectrometry of C-a monomer. GPC was recorded on a Viscotek Model 305 TDAmax with refractive index detector. Samples were solvated in tetrahydrofuran (THF) solvent (4-5 mg mL −1 ) and filtered through 0.2 μ m PTFE filter before injection. GPC system was calibrated with polystyrene standards and data was analyzed using Omnisec software. Calorimetric studies were performed on a Differential Scanning Calorimeter (DSC, TA instruments Q 20). For dynamic DSC scans, samples (10 ± 2 mg) were sealed in aluminum pans, and heated from 50 to 160 °C at 10 °C min −1 under nitrogen atmosphere. X-Ray Diffraction studies were performed on Rigaku SmartLab X-Ray Diffractometer, CuK α radiation (λ = 1.5406 Å). The surface morphology of samples was studied using a Scanning Electron Microscope (Zeiss EVO MA15) under an acceleration voltage of 1 kV. Samples were mounted on aluminium stubs and sputter-coated with gold and palladium (10 nm) using a sputter coater (Quorum-SC7620) operating at 10-12 mA for 120 s. The morphology and microstructure of bulk rGO (S) and S90-2.5% rGO (S) composite were characterized through field emission scanning electron microscopy (FESEM, Carl Zeiss Ultra 55) and high resolution transmission electron microscopy (HRTEM, JEOL-2100F). Energy-dispersive X-ray spectroscopy (EDX) and elemental analysis were recorded on a Carl-Zeiss Ultra 55 FE-SEM with Oxford EDX system. Cells fabrication and electrochemical measurements. The S90 copolymer cathode was prepared by mixing S90 copolymer as active material, super C-65 carbon black as conductive additive and polyethylene oxide (PEO) as binder with an overall ratio of 7:2:1 (by wt.) in acetonitrile. After stirring, the slurry was blade cast onto carbon-coated aluminium foil and then dried at 60 °C for 48 h. Electrodes using S90 copolymer with 2.5% rGO (S) and 2.5% rGO (NS) composites were prepared separately in the same way. The electrodes were cut into disks with a diameter of 12 mm. The average sulfur loading densities on the electrodes were ~1.0 (for S90) and ~0.95 mg cm −2 (for S90-2.5% rGO composite) respectively. Coin-type (CR2016) cells were assembled in an argon-filled glovebox. Lithium metal was used as both reference and counter electrode. Borosilicate glass microfiber filters (Whatmann) was used as separator soaked with 40 μ L of electrolyte composed of 0.25 M LiTFSI and 0.1 M LiNO 3 in 1:1 (v/v) mixture 1,3-dioxolane (DOL) and 1,2-dimethoxy ethane (DME). Cyclic voltammetry was carried out at a scan rate of 20 μ V s −1 within the potential window 1.2-3.0 V vs. Li + /Li using Biologic VMP-3. The Galvanostatic charge-discharge experiments were performed using an Arbin BT-2000 within the potential range 1.5-2.6 V. The capacities were evaluated based on the mass of sulfur present in the S90 copolymer (90% sulfur) and S90-2.5% rGO composites (87.75% sulfur). Electrochemical impedance spectroscopy (EIS) measurements were carried out using Bio-logic VMP-3 over the frequency range of 10 mHz to 1 M Hz and with a voltage amplitude of 5 mV. All electrochemical measurements were carried out at a constant temperature of 20 °C.