Grafting polymeric sulfur onto carbon nanotubes as highly-active cathode for lithium–sulfur batteries

Grafting polymeric sulfur onto carbon nanotubes as highly-active cathode for lithium–sulfur batteries

Journal of Energy Chemistry 42 (2020) 27–33 Contents lists available at ScienceDirect Journal of Energy Chemistry journal homepage: www.elsevier.com...

3MB Sizes 0 Downloads 13 Views

Journal of Energy Chemistry 42 (2020) 27–33

Contents lists available at ScienceDirect

Journal of Energy Chemistry journal homepage: www.elsevier.com/locate/jechem

Grafting polymeric sulfur onto carbon nanotubes as highly-active cathode for lithium–sulfur batteries Junfeng Wu a, Siyu Ding b, Shihai Ye a,∗, Chao Lai b,∗ a

Institute of New Energy Material Chemistry, School of Materials Science and Engineering, National Institute for Advanced Materials, Nankai University, Tianjin 300350, China b School of Chemistry and Materials Science, Jiangsu Normal University, Xuzhou 221116, Jiangsu, China

a r t i c l e

i n f o

Article history: Received 20 April 2019 Revised 15 May 2019 Accepted 23 May 2019 Available online 29 May 2019 Keywords: Lithium–sulfur batteries Polymeric sulfur Carbon nanotubes DFT calculations High capacity

a b s t r a c t Lithium–sulfur (Li–S) batteries are being explored as promising advanced energy storage systems due to their ultra-high energy density. However, fast capacity fading and low coulombic efficiency, resulting from the dissolution of polysulfides, remain a serious challenge. Compared to weak physical adsorptions or barriers, chemical confinement based on strong chemical interaction is a more effective approach to address the shuttle issue. Herein, we devise a novel polymeric sulfur/carbon nanotube composite for Li–S battery, for which 2, 5-dithiobiurea is chosen as the stabilizer of long-chain sulfur. This offers chemical bonds which bridge the polymeric sulfur and carbon nanotubes. The obtained composite can deliver an ultra-high reversible capacity of 1076.2 mAh g−1 (based on the entire composite) at the rate of 0.1 C with an exceptional initial Coulombic efficiency of 96.2%, as well as remarkable cycle performance. This performance is mainly attributed to the reaction reversibility of the obtained polymeric sulfur-based composite during the discharge/charge process. This was confirmed by density functional theory calculations for the first time. © 2019 Science Press and Dalian Institute of Chemical Physics, Chinese Academy of Sciences. Published by Elsevier B.V. and Science Press. All rights reserved.

1. Introduction The element sulfur (S) is a star cathode material for lithium battery due to its ultrahigh specific capacity of 1675 mAh g−1 and high energy density of 2600 Wh kg−1 after pairing with lithium anode [1–7]. Combining with its additional advantages of earthabundance, low-cost and environmentally benign nature, lithium– sulfur (Li–S) batteries have been regarded as the most promising next-generation batteries. However, the practical application of Li– S batteries has been severely impeded by the poor conductivity and large volume expansion of sulfur, as well as the “shuttle effect” caused by the dissolution of discharge–charge intermediates (polysulfides) [1–10]. According to the “dissolving-reaction” mechanism of sulfur cathode in ether-based electrolyte [11], the dissolution of polysulfides is inevitable during the cycling process. Thus, the key issue is how to prohibit the intermediates from migrating from the cathode to anode, and then prevents the intermediates from reacting with the lithium anode via a parasitic reaction. To suppress the shuttle effect, a variety of strategies have been developed, including physical and chemical confinement [2,12].



Corresponding author. E-mail addresses: [email protected] (S. Ye), [email protected] (C. Lai).

Compared with weak physical adsorptions or barriers, chemical confinement based on strong chemical interactions between polysulfides and polar compounds (e.g., metal-based compounds and carbon dopants) [2,13–18] can effectively prohibit the dissolution of polysulfides and consequently prolong the cycle life of Li–S batteries. In particular, Pyun and co-workers recently proposed a novel sulfur-based copolymer that consists of the linear long-chain sulfur and 1, 3-diisopropenylbenzene. The chemical bonds (S–C) between the polysulfides and the co-polymer are much stronger than the physical adsorptions [19]. As a result, the polymeric sulfur presents a high initial capacity of 1100 mAh g−1 at 0.1 C with excellent cycling stability. However, poor conductivity of polymeric sulfur also limits its performance, rate capability and energy density. In further advancement of polymeric sulfur-based cathode, Park and co-workers devised a three-dimensionally interconnected sulfur-rich polymer and the obtained copolymers can present a high initial capacity of 1210 mAh g−1 at 0.1 C and enhanced rate performance. It should also be noted that the specific capacity based on the composite was only 762.3 mAh g−1 [20]. Other sulfur stabilizers, such as covalent triazine, 1, 3-diethynylbenzene and 3-butylthiophene were also employed in the literature [21–27]. Though the resultant specific capacities seemed remarkable, it was calculated based on the active material (i.e., pure sulfur). The specific capacities based on the entire active composite still need great

https://doi.org/10.1016/j.jechem.2019.05.020 2095-4956/© 2019 Science Press and Dalian Institute of Chemical Physics, Chinese Academy of Sciences. Published by Elsevier B.V. and Science Press. All rights reserved.

28

J. Wu, S. Ding and S. Ye et al. / Journal of Energy Chemistry 42 (2020) 27–33

Fig. 1. Schematic illustration of the synthesis of SDCD sample (a); SEM images of CC-DTB (b) and SDCD samples (c); element mapping of SDCD sample (d).

improvement. In addition, In addition, the polymeric sulfur will break down into shorter-chain sulfides during discharge process, and the regeneration of polymeric sulfur during charge process, but not ring S8, is essential to maintain the cycle stability for such restricting strategy. Nevertheless, there are still no reports on the electrochemical-reaction reversibility of polymeric sulfur. Herein, we developed a novel polymeric sulfur/carbon nanotubes composite cathode to address all the problems discussed earlier. 2, 5-Dithiobiurea (DTB) was chosen as the stabilizer of long-chains sulfur to offer chemical bonds bridging the polymeric sulfur and carbon nanotubes. As presented in Fig. 1(a), polymeric sulfur was first produced via ring-opening reaction, and DTB can react with the chain-ends of the sulfur polymer to hinder the reformation of rings, which is designated as S-DTB sample. Then carbon nanotubes functionalized with carboxylic acid group (CNT-CA) and DTB were connected via an amidation reaction, which is marked as CC-DTB. Finally, S-DTB was further bound to the surface of CCDTB via the same reaction between DTB and polymeric sulfur, and the obtained coaxial composite of carbon nanotubes and polymeric sulfur is marked as SDCD. In our design concept, polymeric sulfur can be steadily anchored on the surface of conductive carbon nanotubes via chemical bonds, which effectively prohibited the dissolution of polysulfides, and significantly enhanced the utilization of active materials. Naturally, ultra-high discharge capacity and stable cycle performance were obtained for the as-prepared composites. Furthermore, we also investigated the reaction reversibility of

polymeric sulfur via the first-principle calculations based on density functional theory (DFT). 2. Experimental 2.1. Material synthesis and characterization S-DTB was synthesized via a facile melting route. Briefly, 2 g pure sulfur was firstly added into a crucible and 400 mg DTB was added into the crucible when sulfur became yellow couloured liquid under stirring. Then the crucible was sealed and heated to 185 °C for 30 min, upon cooling to room temperature. The final product was ground into powder. For the synthesis of CC-DTB, 400 mg CNT-CA was added into a 100 mL beaker and 50 mL dimethyl formamide was poured into the beaker. After ultrasonic dispersion for 30 min, 200 mg 1-ethyl3-(3-dimethylaminopropyl) carbodiimide (EDC) (as carboxyl activator) and 200 mg 2,5-dithiobiurea were added under stirring for 10 min. Then, 1.5 mL ammonium hydroxide and 1.5 mL Nhydroxysuccinimide (as linking agent) was added, and the sealed beaker was put on a magnetic stirring apparatus for an 8 h reaction. The obtained sample was washed with deionized water for 3 times and freeze dried for about 24 h. 800 mg S-DTB and 200 mg CC-DTB were uniformly mixed, sealed and put in an argon gas filled autoclave. The autoclave was then put into a Muffle furnace and kept at 185 °C for 12 h, and finally, 957 mg SDCD

J. Wu, S. Ding and S. Ye et al. / Journal of Energy Chemistry 42 (2020) 27–33

smple was produced. For comparsion, S-CNT was also prepared through molten sulfur method. To be specific, 100 mg of commercial MWCNT (XF NANO, INC, Multi-walled carbon nanotubes OD < 8 nm, length 10–30 μm, purity > 95 wt%, Ash < 1.5 wt%, SSA > 50 0 m2 g−1 , EC > 10 0 s cm−1 ) and 400 mg sulfur powder were uniformly mixed and put into a sealed argon filled crucible and then heated at 155 °C for 12 h. XRD patterns were confirmed using X-ray powder diffractometer (Rigaku MiniFlex II, XRD) in the diffraction angle range of 15°–45° with a scan rate of 4°/min. X-ray photoelectron spectra (XPS) were conducted on X-ray photoelectron spectrometer (PHI50 0 0VersaProbe) with an aluminum monochromator at a power of 44.9 W. Raman spectra were attained using Raman spectrometer (RTS-HiR-AM) with a 532 nm laser in Raman frequency shift of 50–60 0 0 cm−1 at more than 100 mW. TGA was performed with a thermogravimetric analyzer (Mettler-Toledo) in argon with a heating rate of 10 °C min−1 from ambient temperature to 600 °C. Fourier transform infrared spectroscopy (FTIR) was conducted with an infrared spectrometer (Nicolet 6700) using KBr pellets in the transmission mode. The microstructure and morphology of the as-prepared materials were acquired from field-emisson scanning electron microscope (SEM, JSM-7800) with an X-ray energy dispersive spectrometer (EDS) and transmission electron microscope (TEM, FEI, Tecnai F20). 2.2. Electrochemical measurements For working electrodes, 70 wt% as-prepared electrode material, 20 wt% Super P as conductive carbon and 10 wt% PVDF as binder were dispersed in NMP as a dispersing agent and stirred on the magnetic stirrer for 4 h. Subsequently, the slurry was coated on an aluminum foil using a doctor blade and then dried at 60 °C overnight in a vacuum oven. The obtained electrode foil was pressed and cut into circular disks with a diameter of 10 mm, where the areal mass is around 1.5–1.8 mg cm−2 . Metallic lithium was used as the anode, and the electrolyte was 1 M lithium bis (trifluoromethane)sulfonamide (LiTFSI, 98.0%, TCI) and 0.2 M lithium nitrate (99.999%, Alfa Aesar) in a mixture of 1,3-dioxolane (DOL) and dimethoxyethane (DME)(v:v, 1:1). 40 wt% DTB, 50 wt% Super P carbon as the conductive agent and 10 wt% PVDF as the binder were compressed to prepare DTB cathode material. The cells of DTB were assembled in the same method above. The galvanostatic discharge/charge tests on the assembled cells were performed between 1.7 and 2.8 V (vs. Li+ /Li) at different current rates using a LAND battery test system (CT2001C). The specific capacities were calculated based on the weight of composite as the cathode active material. Electrochemical impedance spectra (EIS) were measured by using electrochemical impedance test instrument unit (ZAHNER elektrik IM6ex) at different states within the frequency range of 100 KHz–10 mHz with 5 mV AC amplitude. 2.3. DFT calculations The DFT calculations were carried out by using the Vienna Ab-initio Simulation Package (VASP) with an exchange-correlation functional described by the Perdew–Burke–Ernzerhof generalized gradient approximation (PBE-GGA) method [28,29], and interaction between core electrons and valence electrons was performed using the frozen-core projector-augmented wave (PAW) method [30,31]. Wave functions were expanded in a plane wave basis with a high energy cut-off of 400 eV and 5 × 5 × 5 for the numbers of k-point can ensure the convergence for the total energy. The convergence criterion was set to 10−5 eV between two ionic steps for the selfconsistency process, and 0.02 eV A˚ −1 was adopted for the total energy calculations. In our calculations, we set the lattice parameter

29

15 × 15 × 15 A˚ 3 to fit the molecule (2, 5-Dithiobiurea), which is coupling with polysulfides. 3. Results and discussion The morphology of the obtained composites was first studied by scanning electron microscopy (SEM) and elemental mappings. As shown in Fig. 1(b) and (c) and S1, both CC-DTB and SDCD have a three-dimensional inter-linked structure. For SDCD, there is an obvious coating layer on the surface of carbon nanotubes (Fig. 1(c)). However, such a “coaxial structure” is hardly detected by HRTEM (Fig. S1(d)) due to the evaporation of the coating layer under the high-energy electron beams. Fig. 1(d) shows the elemental mapping of the SDCD sample, where the elements of carbon (C), nitrogen (N) and sulfur are clearly present, suggesting that DTB and sulfur are uniformly anchored on the surface of carbon nanotubes. The existences of carbon, nitrogen, and sulfur in different samples were also confirmed by the X-ray photoelectron spectroscopy (XPS) results in Fig. S2. To study the interaction between DTB and polymeric sulfur and disclose the detailed structure of SDCD, X-ray diffraction (XRD), Fourier Transform infrared spectroscopy (FTIR), and XPS were conducted. Fig. 2(a) shows the XRD patterns of sulfur, DTB, S-DTB and SDCD samples. For S-DTB, all diffraction peaks match well with the characteristic peaks of sulfur [13–18], while obvious peak shift was observed as compared to pure sulfur and no characteristic peaks of DTB were detected. This might be attributed to the fact that all C=S bonds have reacted with polymeric sulfur, and there are still element sulfur residues in S-DTB. After further reaction with CCDTB, the peaks of sulfur disappeared, suggesting that the residual sulfur is further consumed by C=S bonds and bound on the surface of carbon nanotubes. Detailed reaction mechanism can be illustrated from the FTIR spectra in Fig. 2(b). For DTB sample, the characteristic peaks at 1045.8 and 1467.1 cm−1 can be attributed to the C=S bond and mixed vibrations of N–C=S bands, respectively. These bands are almost disappeared in S-DTB and SDCD samples [32–35]. There are two new peaks located at 1094.8 and 1676.4 cm−1 both in S-DTB and SDCD samples, and they can be attributed to the C−S bonds and C=N bonds in S–C=N groups, respectively [32–35]. In addition, as present in the infrared spectra of CC-DTB in Fig. S3, the peak located at 1620.8 cm−1 is the characteristic band of stretching vibration of CONH. This was also confirmed by the slight appearance of the characteristic peaks at 1567.5 cm−1 of −COOH [36,37], indicating that DTB was chemically bound to the surface of the carbon nanotubes. Based on the above results, the detailed formation process of SDCD can be clarified as follow: (1) firstly, the C=S bonds in DTB break and bind to long-chain sulfur resulting in formation of N=C–S groups; (2) DTB chemically connects with carbon nanotubes via the reaction between −NH2 and −COOH groups to form O=C–N bond; (3) compounds generated from step 1 are chemically bound to the surface of the carbon nanotubes via the same reaction between polymeric sulfur and C=S groups. A detailed structure can be found in Fig. 1(a). Fig. 2(c) is the C 1s core spectrum of CC-DTB and SDCD samples. For both samples, the deconvolution of the C 1s spectrum are almost identical and shows four peaks at 284.4, 284.6, 285.5 and 288.7 eV assigned to C–C, C=C, C–S/C=N and C=O bonds, respectively [38–41]. It is difficult to distinguish the two samples in C 1s spectra, while the S 2p core spectrum (Fig. 2(d)) is different. The S 2p core level signals of SDCD sample show five peaks at 162.5, 163.6, 164.8, 165.0 and 168.4 eV, which can be attributed to C–S (2p1/2), C-S (2p3/2), S-S (2p1/2), S-S (2p3/2), sulfate, respectively [2,42]. While only a broad peak is observed for the CC-DTB sample, generated from the C=S bonds. Besides, the formation of polymeric sulfur can be directly proved via dissolution experiments (Fig. 2(e)). As presented, sulfur can dissolve into carbon disulfide

30

J. Wu, S. Ding and S. Ye et al. / Journal of Energy Chemistry 42 (2020) 27–33

Fig. 2. XRD patterns (a) and FTIR spectra (b) of the different as-prepared samples; C 1s (c) and S 2p (d) core level spectra of CC-DTB and SDCD; dissolution experiments of sulfur and S-DTB in CS2 solvent (e); TGA curves of S-CNT, S-DTB and SDCD samples (f).

solvent, but S-DTB is basically insoluble. Based on all above results, it can be concluded that long-chain sulfur is stabilized and chemically bound to the surface of carbon nanotubes via DTB to produce a three-dimensionally inter-linked composite. The content of active sulfur in the composite was also investigated via thermogravimetric analysis (TGA) instruments. For comparison, the sulfur/carbon nanotubes composite (S-CNT) was also prepared via the same experimental process. As shown in Fig. 2(f), an obvious weight loss process, occurring at temperature above 200 °C, can be observed for all the samples, and the weight loss was 75.0%, 92.0%, 75.4% corresponding to SDCD, S-DTB, S/CNT, respectively. It is obvious that weight loss of SDCD includes the loss of sulfur and 2, 5-dithiobiurea, and it also should be noted that there are basically no weight loss below 200 °C for S-DTB. Thus, the content of active polymeric sulfur in the SDCD can be obtained from the weight change after the thermal-treating process at 185 °C, and it is calculated about 65.2 wt%. As presented in Fig. S4, DTB is electrochemical inactive, and thus the theoretical capacity of SDCD is calculated to be 1092.1 mAh g−1 based on the content of sulfur. The initial discharge–charge curves of SDCD and S-CNT cathode at the rate of 0.1 C are first given in Fig. 3(a) to illustrate the electrochemical behaviour of the obtained composite. The SDCD sample shows an ultra-high initial discharge capacity of 1118.9 mAh g−1 and a reversible charge capacity of 1076.2 mAh g−1 based on the composite, which is almost equal to the theoretical capacity of SDCD and much higher than that of S-CNT composite. To the best of our knowledge, this is the best result when compared to previous reports (Table S1) [19–26]. Especially, the initial coulombic efficiency which is up to 96.2%, suggesting the high reversibility of the SDCD cathodes. Two obvious discharge plateaus are found from the curve of both samples, corresponding to the transformation of polymeric sulfur or elemental sulfur (S8 ) to lithium polysulfides and then to Li2 S2 /Li2 S [43–46]. It can be seen that SDCD exhibited very similar voltage curves compared to the characteristic peaks of elemental sulfur, which is consistent with other reports of

polymeric sulfur cathodes [19–26]. In addition, the overpotential of SDCD cathode is much lower than that of S–CNT sample, indicating a better kinetic process. Furthermore, the direct evidence of excellent conductivity for SDCD cathode can be investigated by the electrochemical impedance spectroscopy (EIS) measurements and the Nyquist plots of both samples are given in Fig. S5. It is obvious that the charge transfer impedance associated with semi-circular arc in the high frequency range of SDCD is much smaller than that of S-CNT, corresponding to a fast electron transfer process [47]. Thus the SDCD cathode can demonstrate a much better rate performance as shown in Fig. 3(b). It should be noted that the capacity calculation were all based on the composite. With the C-rate increasing from 0.2, 0.5 and 1.0 C, high reversible capacity of 820.6, 699.4 and 585.5 mAh g−1 still can be retained, and when the C-rate return to 0.1 C, the discharge capacity can be recovered. In addition, the SDCD cathode showed a stable cycle performance with a high Coulombic efficiency close to 100% from the second cycle onwards. For the controlled sample, S–CNT, the discharge capacity is much lower, and an overcharge phenomenon can be observed at the rate of 1.0 and 0.1 C after initial cycles due to the gradual dissolution of polysulfides during cycling. The long cycle curves at the rate of 0.2 and 0.5 C of SDCD electrodes further display its excellent structural stability. As present in Fig. 3(c), the SDCD can show excellent cycle performance with high Coulombic efficiency at 0.2 C, after 200 cycles, the discharge capacity can be stably retained at 586.3 mAh g−1 (based on the composite). In the case of S–CNT, fast decay of discharge capacity can be observed, and overcharge and short circuit are detected again after cycling as shown in Fig. S6. The discharge–charge curves of SDCD cathode with 1st, 10th, 10 0th, 20 0th cycle number at 0.2 C are also given in Fig. 3(d). From the initial cycle, the potential plateau was well retained after different cycles, indicating the high stability of electrode structure during the discharge–charge process. Fig. 4(a) is the cycle curves of SDCD at the rate of 0.5 C. After activation at 0.1 C for one cycle the SDCD electrode showed a

J. Wu, S. Ding and S. Ye et al. / Journal of Energy Chemistry 42 (2020) 27–33

31

Fig. 3. Initial discharge–charge curves at the rate of 0.1 C (a) and cycling performance at various rates (b) of SDCD and S-CNT cathodes; cycling performance of SDCD and S-CNT cathodes at the rate of 0.2 C (c); discharge–charge curves of SDCD at the rate of 0.2 C (d). The current density is calculated based on the composite.

ultra-stable cycle performance up to 200 cycles, a high capacity of 545.5 and 836.6 mAh g−1 (based on the composite and sulfur) was still retained, respectively. Such excellent electrochemical performance, including ultra-high utilization of active materials, high reversibility and stable cycle performance, can be mainly attributed to the unique structure of the polymeric sulfur/carbon nanotubes composite. As shown in Fig. 4(b), polymeric sulfur was stably anchored on the surface of the carbon nanotubes via the formation of C–S–S bonds, which is not only beneficial for the fast transfer of electrons from carbon nanotubes to polymeric sulfur, but also firmly restricts the dissolution of polysulfides during the cycling process. Moreover, the loose inter-linked carbon nanotubes facilitate the transport of lithium-ions throughout the whole composite. Apart from the enhanced kinetic properties, the prior ringopening reaction of sulfur also contributes to the high capacity of the obtained composite. As shown in Fig. S7, the S-DTB electrodes prepared by compressing S-DTB, carbon black and binder with a weight ratio of 7:2:1, showed a high initial discharge capacity of 905.9 mAh g−1 , and even after 100 cycles, it stably retained 606.8 mAh g−1 at the rate of 0.1 C. In contrast, pure sulfur electrode experiences poor discharge/charge and delivers a much low capacity below 150 mAh g−1 . It is obvious that polymeric sulfur is a more electroactive material as compared to elemental sulfur (S8 ). Meanwhile, it should be pointed out that previous materials designed to produce high performance cathodes are based on the hypothesis of reversible reaction of polymeric sulfur. The binding of polysulfides with DTB during the charging process is the key to stabilizing the electrode structure and delivering a stable cycle

performance. To illustrate the detailed reaction mechanisms, DFT calculations are employed to calculate the Gibbs free energy of different reaction routes. As shown in Fig. 5, S2− and S4 2− are more inclined to form high-order polysulfides, while the S8 2− reacts easily with DTB to form polymeric sulfur, but not ring S8 , as the reaction Gibbs free energy from S8 2− to S8 is −0.446 eV which is higher than that of forming polymeric sulfur (−0.680 eV). The theoretical prediction can be further verified by the XPS analysis before cycling, discharge and charge states after 5 cycles, and the results are given in Fig. 6. For the electrode before cycling, two peaks at 163.6 and 164.8 eV can be assigned to C–S and S–S [–(Sn)–] bonds of the polymeric sulphur, respectively. This is consistent with the results of Fig. 2(d) [2,48]. The peak of S–S [–(Sn)–] bond at 164.8 eV disappears at the discharge state after five cycles, accompanied with two other peaks arising at 162.8 and 161.6 eV due to the presence of polysulfides and Li2 S2 /Li2 S, respectively. The obvious specie transformation reveals the reduction reaction from polymeric sulphur to Li2 S2 /Li2 S and partial dissolution of polysulfides in the discharge process [49,50]. Other peaks between 166.0 and 171.0 eV are the signals of thiosulfate and polythionate resulting from the side reaction between polysulfides and electrolytes or the oxidation of polysulfides in air [38,50]. In addition, the characteristic peaks of C−S and S−S [–(Sn)–] bonds at 163.6 and 164.8 eV can be also detected at the charged state after five cycles, indicating that the polymeric sulphur reappears and its redox reaction is reversible. It should be emphasized that the peak of polysulfides at 162.8 eV still exists at the charged state owing to the partial dissolution of polysulfides, which is inevitably in Li−S batteries. FTIR

32

J. Wu, S. Ding and S. Ye et al. / Journal of Energy Chemistry 42 (2020) 27–33

Fig. 4. Cycling performance of SDCD and S-CNT cathodes at the rate of 0.5 C after activations at the rate of 0.1 C for the first cycle (a) and schematic illustration of the electrochemical reaction of SDCD (b). The current density is calculated based on the composite.

Fig. 6. High resolution S 2p core level XPS spectra of SDCD electrode before cycling, as well as discharge and charge states after 5 cycles. The bold black line is the raw data. Fig. 5. The reaction Gibbs free energy (G) obtained via DFT calculations.

J. Wu, S. Ding and S. Ye et al. / Journal of Energy Chemistry 42 (2020) 27–33

measurements of SDCD electrode were also conducted before and after cycling, and the results are given in Fig. S8. As presented, the C−S and S–C=N bonds are regenerated after 5 cycles, indicating the reversibility of polymeric sulfur binding on DTB. Both calculation and experimental results confirm that the polymeric sulfur can be re-anchored on the surface of carbon nanotubes during the charging process, and thus excellent cycle performance is obtained. 4. Conclusions In summary, a novel polymeric sulfur/carbon nanotube composite for Li−S battery, for which polymeric sulfur is stably anchored on the surface of carbon nanotubes, is successfully synthesized. Ultra-high reversible capacity of 1076.2 mAh g−1 is obtained for SDCD cathodes at the rate of 0.1 C, as well as high initial coulombic efficiency up to 96.2%. To the best of our knowledge, this is the highest capacity (based on the composite) that has been reported for polymeric sulfur. When the current density was increased to 0.2, 0.5 and 1.0 C, a corresponding high discharge capacity of 820.6, 699.4 and 585.5 mAh g−1 can still be retained with excellent cycle performance. DFT calculations were also conducted to illustrate the reaction mechanisms of the composite. During the charging process, low-order polysulfides, such as Li2 S, tend to transform into high-order polysulfides, and then the generated S8 2− reacts with DTB preferentially to form polymeric sulfur, but does not transform into S8 , producing a reversible reaction process. These results can offer new insights into the development of highly efficient and stable electrode materials for Li–S batteries. Acknowledgments This work is financially supported by the National Natural Science Foundation of China (No. 51572116 and 51871113) and Key Research and Development Program of Xuzhou (KC17004). Thanks for David Adekoya (Griffith University) for his contribution to the language revision. Supplementary material Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.jechem.2019.05.020. References [1] [2] [3] [4] [5] [6] [7] [8] [9]

R.V. Noorden, Nature 507 (2014) 26–28. X. Liu, J.Q. Huang, Q. Zhang, L.Q. Mai, Adv. Mater. 29 (2017) 1601759. X. Zhang, X. Cheng, Q. Zhang, J. Energy Chem. 6 (2016) 967–984. A. Manthiram, Y. Fu, S.H. Chung, C.X. Zu, Y.S. Su, Chem. Rev. 114 (2014) 11751–11787. Z.-W. Zhang, H.-J. Peng, M. Zhao, J.-Q. Huang, Adv. Funct. Mater. 28 (2018) 1707536. H. Chen, Ling, M. Ling, L. Hencz, H.Y. Ling, G. Li, Z. Lin, G. Liu, S. Zhang, Chem. Rev. 118 (2018) 8936–8982. H. Yuan, J.-Q. Huang, H.-J. Peng, M.-M. Titirici, R. Xiang, R. Liu, Q. Chen, Q. Zhang, Adv. Energy Mater. 8 (2018) 1802107. M. Zhao, H.-J. Peng, Z.-W. Zhang, B.-Q. Li, X. Chen, J. Xie, X. Chen, J.-Y. Wei, Q. Zhang, J.-Q. Huang, Angew. Chem. Int. Ed. 131 (2019) 3819–3823. L. Hencz, H. Chen, H.Y. Ling, Y. Wang, C. Lai, H. Zhao, S. Zhang, Nano-Micro Lett. 1 (2019) 11–17.

33

[10] J.-L. Shi, C. Tang, J.-Q. Huang, W. Zhu, Q. Zhang, J. Energy Chem. 27 (2018) 167–175. [11] J.Y. Koh, M.-S. Park, E.H. Kim, T.J. Kim, S. Kim, K.J. Kim, Y.-J. Kim, Y. Jung, J. Electrochem. Soc. 161 (2014) A2117–A2120. [12] S. Evers, L.F. Nazar, Acc. Chem. Res. 46 (2013) 1135–1143. [13] X.Y. Tao, J.G. Wang, C. Liu, H.T. Wang, H.B. Yao, G.Y. Zheng, Z.W. Seh, Q.X. Cai, W.Y. Li, G.M. Zhou, C.X. Zu, Y. Cui, Nat. Commun. 7 (2016) 11203. [14] C. Dai, J.-M. Lim, M. Wang, L. Hu, Y. Chen, Z. Chen, H. Chen, S.-J. Bao, B. Shen, Y. Li, G. Henkelman, M. Xu, Adv. Funct. Mater. 28 (2018) 1704443. [15] X. Liang, C.Y. Kwok, F. Lodi-Marzano, Q. Pang, M. Cuisinier, H. Huang, C.J. Hart, D. Houtarde, K. Kaup, H. Sommer, T. Brezesinski, J. Janek, L.F. Nazar, Adv. Energy Mater. 6 (2016) 1501636. [16] Q.F. Zhang, Y.P. Wang, Z.W. Seh, Z.H. Fu, R.F. Zhang, Y. Cui, Nano Lett. 15 (2015) 3780–3786. [17] X. Liang, A. Garsuch, L.F. Nazar, Angew. Chem. Int. Ed. 54 (2015) 3907–3911. [18] S.Q. Li, T. Mou, G.F. Ren, J. Warzywoda, B. Wang, Z.Y. Fan, ACS Energy Lett. 1 (2016) 481–489. [19] W.J. Chung, J.J. Griebel, E.T. Kim, H. Yoon, A.G. Simmonds, H.J. Ji, P.T. Dirlam, R.S. Glass, J.J. Wie, N.A. Nguyen, B.W. Guralnick, J. Park, A. Somogyi, P. Theato, M.E. Mackay, Y.E. Sung, K. Char, J. Pyun, Nat. Chem. 5 (2013) 518–524. [20] H. Kim, J. Lee, H. Ahn, O. Kim, M.J. Park, Nat. Commun. 6 (2015) 7278. [21] J. Park, E.T. Kim, C.J. Kim, J. Pyun, H.S. Jang, J. Shin, J.W. Choi, K. Char, Y.E. Sung, Adv. Energy Mater. 7 (2017) 1614. [22] G.J. Hu, Z.H. Sun, C. Shi, R.P. Fang, J. Chen, P.X. Hou, C. Liu, H.M. Cheng, F. Li, Adv. Mater. 29 (2017) 1603835. [23] S.B. Zeng, L.G. Li, D.K. Zhao, J. Liu, W.H. Niu, N. Wang, S.W. Chen, J. Phys. Chem. C 121 (2017) 2495–2503. [24] B. Ding, Z. Chang, G.Y. Xu, P. Nie, J. Wang, J. Pan, H. Dou, X.G. Zhang, ACS Appl. Mater. Interface 7 (2015) 11165–11171. [25] S.N. Talapaneni, T.H. Hwang, S.H. Je, O. Buyukcakir, J.W. Choi, A. Coskun, Angew. Chem. Int. Ed. 55 (2016) 3106–3111. [26] Z.J. Sun, M. Xiao, S.J. Wang, D.M. Han, S.Q. Song, G.H. Chen, Y.Z. Meng, J. Mater. Chem. A 2 (2014) 9280–9286. [27] S.-Z Zeng, Y. Yao, X. Zeng, Q. He, X. Zheng, S. Chen, W. Tu, J. Zou, J. Power Sources 357 (2017) 11–18. [28] G. Kresse, J. Hafner, Phys. Rev. B 47 (1993) 558. [29] J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 77 (1996) 3865. [30] G. Kresse, D. Joubert, Phys. Rev. B 59 (1999) 1758. [31] G. Kresse, J. Furthmüller, Comput. Mater. Sci. 6 (1996) 15. [32] R.A. Bailey, S.L. Kozak, T-W. Michelsen, W.N. Mills, Coord. Chem. Rev. 6 (1971) 407–445. [33] C.N.R. Rao, R. Venkataraghavan, T.R. Kasturi, Can. J. Chem. 42 (1964) 36–42. [34] V. Kvishnakumar, V. Balachandran, Indian J. Pure Ap. Phys. 42 (2004) 313–318. [35] K.A. Jensen, P.H. Nielsen, Acta Chem. Scand. 20 (1966) 597–629. [36] V. Yatsyna, D.J. Bakker, R. Feifel, A.M. Rijs, V. Zhaunerchyk, J. Chem. Phys. 145 (2016) 104309. [37] R. Imani, S.H. Emami, S. Faghihi, Phys. Chem. Chem. Phys. 17 (2015) 6328–6339. [38] L. Sun, D.T. Wang, Y.F. Luo, K. Wang, W.B. Kong, Y. Wu, L.N. Zhang, K.L. Jiang, Q.Q. Li, Y.H. Zhang, J.P. Wang, S.S. Fan, ACS Nano 10 (2016) 1300–1308. [39] S.-K. Park, J.-K. Lee, Y.C. Kang, Adv. Funct. Mater. 28 (2018) 1705264. [40] S.Y. Yuan, J.W. Lucas Bao, L.N. Wang, Y.Y. Xia, D.G. Truhlar, Y.G. Wang, Adv. Energy Mater. 6 (2016) 1501733. [41] J. Liu, T. Qian, M.F. Wang, X.J. Liu, N. Xu, Y.Z. You, C.L. Yan, Nano Lett. 17 (2017) 5064–5070. [42] X.H. Zhao, H.J. Ahn, K.W. Kim, K.K. Cho, J.H. Ahn, J. Phys. Chem. C 119 (2015) 7996–8003. [43] Y. Liu, G. Li, J. Fu, Z. Chen, X. Peng, Angew. Chem. Int. Ed. 56 (2017) 6176–6180. [44] L. Qie, A. Manthiram, ACS Energy Lett. 1 (2016) 46–51. [45] H.-J. Peng, G. Zhang, X. Chen, Z.W. Zhang, W.T. Xu, J.-Q. Huang, Q. Zhang, Angew. Chem. Int. Ed. 55 (2016) 12990–12995. [46] L.-B. Xing, K. Xi, Q. Li, Z. Su, C. Lai, X. Zhao, R.V. Kumar, J. Power Sources 303 (2016) 22–28. [47] J. Jiang, Z. Lin, Q. Ju, Z. Ma, C. Zheng, Z. Wang, Energy Proc. 105 (2017) 844–849. [48] H. Pan, K.S. Han, M.H. Engelhard, R. Cao, J. Chen, J.-G. Zhang, K.T. Mueller, Y. Shao, J. Liu, Adv. Funct. Mater. 28 (2018) 1870275. [49] R.X. Chu, J. Lin, C.Q. Wu, J. Zheng, Y.L. Chen, J. Zhang, R.H. Han, Y. Zhang, H. Guo, Nanoscale 9 (2017) 9129–9138. [50] X. Liang, C. Hart, Q. Pang, A. Garsuch, T. Weiss, L.F. Nazar, Nat. Commun. 6 (2015) 5682.