Enhanced kinetics of polysulfide redox reactions on

Download PDF
0 downloads 0 Views 3MB Size Report
Comment
May 26, 2018 - of polysulfide redox reactions is the main hindrance for its practical application in ... Unfortunately, the physical confinement is not sufficient to.
Nanotechnology

Related content

PAPER

Enhanced kinetics of polysulfide redox reactions on Mo2C/CNT in lithium–sulfur batteries To cite this article: Rameez Razaq et al 2018 Nanotechnology 29 295401

View the article online for updates and enhancements.

- Cross-stacked carbon nanotube film as an additional built-in current collector and adsorption layer for high performance lithium-sulfur batteries Li Sun, Weibang Kong, Mengya Li et al. - Synthesis of graphitic ordered mesoporous carbon with cubic symmetry and its application in lithium-sulfur batteries Min-Seop Kim, Jinhoo Jeong, Won Il Cho et al. - Construction of tubular polypyrrolewrapped biomass-derived carbon nanospheres as cathode materials for lithium–sulfur batteries Qiuhong Yu, Yang Lu, Tao Peng et al.

This content was downloaded from IP address 124.128.158.37 on 26/05/2018 at 08:51

Nanotechnology Nanotechnology 29 (2018) 295401 (11pp)

https://doi.org/10.1088/1361-6528/aac060

Enhanced kinetics of polysulfide redox reactions on Mo2C/CNT in lithium–sulfur batteries Rameez Razaq1, Dan Sun1, Ying Xin1, Qian Li1 , Taizhong Huang1, Lei Zheng2, Zhaoliang Zhang1 and Yunhui Huang3 1

School of Chemistry and Chemical Engineering, Shandong Provincial Key Laboratory of Fluorine Chemistry and Chemical Materials, University of Jinan, Jinan 250022, People’s Republic of China 2 Institute of High Energy Physics, Chinese Academy of Sciences, Beijing, 100049, People’s Republic of China 3 School of Materials Science and Engineering, State Key Laboratory of Material Processing and Die & Mould Technology, Huazhong University of Science and Technology, Wuhan, Hubei 430074, People’s Republic of China E-mail: [email protected] Received 15 March 2018, revised 16 April 2018 Accepted for publication 26 April 2018 Published 16 May 2018 Abstract

Among different energy storage devices, the lithium–sulfur (Li–S) battery is the subject of recent attention. However, the capacity decay caused by polysulfide shuttle leading to sluggish kinetics of polysulfide redox reactions is the main hindrance for its practical application in Li–S batteries. Herein, molybdenum carbide nanoparticles anchored on carbon nanotubes (Mo2C/CNT) are reported to serve as an efficient cathode material to enhance the electrochemical kinetics of polysulfide conversion in Li–S batteries. Mo2C/CNT shows strong adsorption and activation of polar polysulfides and therefore accelerates the redox kinetics of polysulfides, reduces the energy barrier, effectively mitigates the polarization and polysulfide shuttle, thus improving the electrochemical performance. The S-Mo2C/CNT composite with 70 wt% sulfur loading exhibits high specific discharge capacity (1206 mA h g−1 at 0.5 C), excellent high-rate performance, long cycle life (900 cycles), and outstanding Coulombic efficiency (∼100%) at a high rate (2 C) corresponding to a capacity decay of only 0.05%. Remarkably, the S-Mo2C/CNT cathode with high areal sulfur loading of 2.5 mg cm−2 exhibits high-rate capacities and stable cycling performance over 100 cycles, offering the potential for use in high energy Li–S batteries. Supplementary material for this article is available online Keywords: lithium–sulfur batteries, polysulfide shuttle, electrochemical kinetics, molybdenum carbides, carbon nanotubes (Some figures may appear in colour only in the online journal) [1–4]. Alternatively, Li–S batteries are promising [5, 6] and have attracted considerable attention due to their high theoretical energy density (2500 Wh kg−1) based on the reaction between Li and S to form Li2S [7]. Furthermore, sulfur is abundant, inexpensive, nontoxic, and has a high theoretical capacity (1675 mA h g−1). However, the commercialization of a Li–S battery is still hindered by several fundamental challenges, for instance, the insulating nature of elemental

1. Introduction Rechargeable batteries with high energy density and long cycle life are highly desirable because of the ubiquitous demand of transportable electronic devices, electric vehicles, and the widespread use of intermittent renewable energy in the modern electrical grid. The current Li–ion batteries cannot fulfill these demands in terms of cost and specific energy 0957-4484/18/295401+11$33.00

1

© 2018 IOP Publishing Ltd Printed in the UK

Nanotechnology 29 (2018) 295401

R Razaq et al

performance for lithium–ion battery anodes [40–45], which is attributed to the synergetic effects between Mo2C and the highly conductive carbon [46]. Inspired by these studies, we found that Mo2C/CNT can effectively suppress the shuttle effect by chemical binding/absorption with polysulfides, meanwhile ensuring fast charge transport to promote the kinetics of polysulfide redox reactions. The integration of Mo2C into CNT can efficiently enhance the redox reactivity and kinetics of lithium polysulfides due to their strong chemical affinity (scheme 1). With these merits, the obtained S-Mo2C/CNT composite with 70 wt% sulfur loading exhibits a high specific discharge capacity of 1206 mA h g−1 at 0.5 C and 802 mA h g−1 at 2 C with a capacity decay of only 0.05% per cycle after 900 cycles with a Coulombic efficiency of ∼100%. Moreover, with high areal sulfur loading of 2.5 mg cm−2 the S-Mo2C/CNT cathode shows discharge capacities of 940, 835, 778, 656, and 552 mA h g−1 at 0.5, 1, 2, 3, and 4 C, respectively.

sulfur and lithium sulfides, volume changes of the cathode during cycling, sluggish electrochemical kinetics, and dissolution of lithium polysulfides (LiPSs) in the electrolyte which triggers a polysulfide shuttle process [8]. Over time, many approaches have been tried to solve these issues by trapping LiPSs within the cathode structure. In the early stage, the main approach was to encapsulate the sulfur into porous carbon materials or a conductive polymer matrix, resulting in interconnected conducting networks and an enhanced physical host of the LiPSs. Examples of such kinds of cathode materials are micro and mesoporous carbon [9–12], carbonspheres [13–15], carbon nanotubes (CNTs)/ fibers [16–18], polyaniline [19], and polypyrrole [20, 21]. Unfortunately, the physical confinement is not sufficient to trap LiPSs over a long cycle life, which is suspected to result in loss of active materials, accumulation of insulating layers on the anode, and finally capacity fading. Recently, sulfur host materials that show robust chemical interactions with LiPSs have been designed and examined to limit the diffusion of soluble polysulfides based on interfacial phenomena rather than spatial confinement, such as polar– polar interactions and Lewis acid-based interactions [22, 23]. These efforts have concentrated on applying nanostructured materials to bind polysulfide redox products, for instance, hollow carbon nanofibers with amphiphilic surface modification [24], N-doped graphene [25], SiOx, and VOx coated on CMK-3 [26], mesoporous SiO2 embedded within the carbon–sulfur composite [27], Ti4O7 [22], Magnéli phase Ti4O7 [28], titanium monoxide@carbon hollow spheres [29], graphene wrapped MIL-101(Cr) [30], MIL-53 (Al), NH2-MIL-53 (Al), HKUST-1 and ZIF-8 [31], titanium disulfide [32], elemental sulfur and molybdenum disulfide composites [33], metal carbide [23, 34–37], mesoporous titanium nitride [38], and conductive porous vanadium nitride/graphene composite [39] have been employed as host materials for Li–S batteries. Although the worth of these elegant systems is clear as demonstrated by improved device performance, the preparation of these nanostructured materials often requires complex and energy intensive synthetic procedures. Thus, the preparation of efficient cathode materials via cost-effective and scalable methods remains a significant objective. Furthermore, the poor electrical conductivity of most metal oxides/sulfides trends to slow down the redox kinetics of absorbed polysulfides. Considering the reaction mechanism, the conversion of surface-bonded polysulfide moieties certainly involves a charge transfer step. If the surface-trapped polysulfide is nonconductive, an additional surface diffusion step will be required for the overall reaction to proceed that further slows down the reaction kinetics. Hence, it is significant to design advanced sulfur hosts with high polysulfide-trapping capability based on hybrid structures of highly conductive materials to improve the sluggish redox kinetics of polysulfides, maximize the cycling life, and enhance the rate performance. To probe the reaction kinetics, metal carbide nanoparticles supported on carbon material have been reported as a highly conductive sulfur-based cathode material for a Li–S battery [23, 34–37]. Actually, Mo2C supported on carbon materials has exhibited enhanced electrochemical

2. Experiment 2.1. Preparation of Mo2C/CNT

CNTs with a carbon purity higher than 99 wt% were purchased from XF NANO Inc. First, the CNT was pre-treated with 3M HCl by placing a suspension of CNT in HCl in an ultrasonic bath for 30 min. After that, the sample was washed several times with distilled water until a neutral pH was reached and the resultant solid was dried in an oven at 110 °C overnight. In the next step, active Mo2C was supported on the pre-treated CNT by a modified incipient wetness impregnation procedure. Dry pre-treated CNT was placed in a flask and put in an ultrasonic bath for 15 min to disperse the particles well. Subsequently, an aqueous solution of ammonium molybdate (0.788 g in 6 ml of distilled water) was added dropwise to the support, and the mixture was then homogenized for 30 min and dried at 110 °C overnight. Last, the sample was annealed in a tube furnace from ambient temperature to 800 °C at a rate of 5 °C min−1 under 100 cm3 min−1 N2 flow for 2 h. The N2 atmosphere was retained throughout the cooling of the reactor to room temperature. For comparison, Mo2C/CNT was also made by simple hand mixing of Mo2C and CNT.

2.2. Sulfur impregnation in Mo2C/CNT

The S-Mo2C/CNT composite electrode was prepared by a solution-based method. To obtain the S-Mo2C/CNT composite, 30 mg of sulfur powder was mixed in 80 ml of absolute ethanol by intensive ultra-sonication until a clear sulfurethanol solution formed. After that 12.85 mg of Mo2C/CNT was added to the sulfur-ethanol solution and the resultant mixture was continuously sonicated for 1 h while dropwise adding the distilled water to allow slow precipitation of the sulfur particles. To remove the solvent, the resulting mixture was dried at 35 °C to form S-Mo2C/CNT. 2

Nanotechnology 29 (2018) 295401

R Razaq et al

Scheme 1. (a) Schematic illustration of the synthesis process of Mo2C/CNT and S-Mo2C/CNT. (b) Illustration of the promoted adsorption and the redox reaction of polysulfides on Mo2C/CNT for the fast conversion of polysulfides.

2.3. Preparation of sulfur electrodes and electrochemical measurements

2.4. Preparation of Li2S electrodes and decomposition energy barrier measurement

Sulfur electrode was prepared by mixing S-Mo2C/CNT (80%), super P (10%) and poly(vinylidene fluoride) (PVDF, Sigma-Aldrich) binder (10%) in N-methyl 2-pyrrolidone (NMP, Sigma-Aldrich) solvent. Typical, S-Mo2C/CNT (80%) and super P (10%) were mixed in a mortar, which were ball milled by adding a PVDF ( Sigma-Aldrich) binder (10%) in NMP for 3 h to form a slurry. The slurry was pasted on aluminum foil with subsequent heating at 80 °C for 12 h. The 2025-type stainless steel coin cells were assembled inside an Ar-filled glovebox with lithium metal foil as the negative electrode. The electrolyte was prepared by dissolving lithium bis-trifluoromethanesulphonylimide (LITFSI, 99%, Acros Organics, 1 M) and lithium nitrate (LiNO3, 99.9%, Alfa Aesar, 0.1 M) in 1,2-dimethoxyethane (DME, 99.5%, Alfa Aesar) and 1,3-dioxolane (DOL, 99.5%, Alfa Aesar) (1:1 ratio, by volume). A LAND galvanostatic charge/discharge system was used at 0.1, 0.5, 1, and 2 C. Stepwise rate performance was examined at elevated current rates of 0.1, 0.2, 0.5, 1, 2, 3, and 4 C on the same battery tester. The charge/discharge voltage range was 1.6–3.0 V. The cyclic voltammetry (CV) tests were performed on an SP-300 at a scan rate of 0.1 mV s−1. Electrochemical impedance spectroscopy (EIS) measurements were conducted on an SP-300 at a frequency range of 100 kHz to 100 mHz.

To check the decomposition energy barrier of the Li2S, a Li2S cathode was prepared by mixing Li2S and Mo2C/CNT (80%), super P (10%), and PVDF binder (10%) in an NMP solvent to form a slurry, which was pasted onto an aluminum foil. After drying, the Li2S electrode was punched out with an 8 mm diameter. To check the decomposition energy barrier of the Li2S, electrochemical measurements were done using CR2025 coin cells with lithium metal as the counter and reference electrodes assembled in an argon-filled glovebox. The electrolyte was prepared by dissolving an appropriate amount of LiTFSI (1 M) in 1:1 v/v DME/DOL containing LiNO3 (0.1 M). To wet the Li2S first, 20 μl of the electrolyte was added, then the Celgard separator was placed over the electrode and a further 20 μl of the electrolyte was added. To perform electrochemical measurements, a LAND galvanostatic charge/discharge instrument was used to charge from open-circuit voltage to 4.0 V (versus Li/Li+) at 0.1 C.

2.5. Symmetrical cell assembly and electrochemical measurements

For symmetrical cells, the electrodes were made by mixing the host material (Mo2C/CNT, Mo2C, and CNT) with a PVDF binder at a 3:1 weight ratio and dispersed in NMP without the presence of elemental sulfur. The slurry was 3

Nanotechnology 29 (2018) 295401

R Razaq et al

stirred and pasted on aluminum foil with subsequent heating at 80 °C for 12 h. After drying, the electrode disks were punched out with a diameter of 8 mm and used as working and reference electrodes. For symmetrical cells, the electrolyte (40 μl) containing 0.5 mol L−1 Li2S6 and 1 mol L−1 LiTFSI dissolved in DOL/DME (v/v=1/1) was used. The Li2S6 solution was prepared by mixing Li2S and sulfur at a molar ratio of 1:5 into the corresponding solvent. The CV tests were performed on an electrochemistry workstation (SP-300) at a scan rate of 50 mV s−1. EIS measurements were conducted on SP-300 at a frequency range of 100 kHz to 100 mHz. 2.6. Characterization

X-ray diffraction (XRD) patterns were recorded on a Rigaku D/max-2500/PC diffractometer employing Cu Kα radiation (λ=1.5418 Å) operating at 50 kV and 200 mA. The Brunauer–Emmett–Teller surface area and pore structure were measured by N2 adsorption/desorption using a Micromeritics 2020 M instrument. Before exposure to N2, the sample was outgassed at 300 °C for 5 h. Field-emission scanning electron microscopy equipped with energy dispersive spectroscopy (EDS) was performed on a Hitachi SU-70 microscope. Transmission electron microscopy (TEM) was conducted on a JEOL JEM-2010 microscope at an accelerating voltage of 200 kV. X-ray photoelectron spectroscopy (XPS) data were obtained on a Thermo Scientific ESCALAB 250 XI electron spectrometer, using monochromatic Al Kα as the exciting radiation at a constant pass energy of 1486.6 eV. To compensate for surface charging effects, the binding energies were calibrated using the C 1s hydrocarbon peak at 284.80 eV. Thermo-gravimetry analyses (TGA) were conducted on a NETZSCH, STA449 from room temperature to 800 °C with a heating rate of 10 K min−1 in Ar. X-ray absorption fine-structure (XAFS) measurements for the S K-edge were performed on the XAFS station at the Beijing synchrotron radiation facility (BSRF, Beijing, China). The S K-edge (3608 eV) data were collected at the 4B7A beam line and analyzed by using the IFEFFIT software package [47].

Figure 1. (a) XRD patterns of S-Mo2C/CNT, Mo2C, and CNT and (b) TGA of S-Mo2C/CNT.

The morphology and microstructure of Mo2C/CNT and S-Mo2C/CNT were characterized by TEM and SEM. CNT was was clearly distinguished in low and high magnification TEM images (figure 2(a)). As to Mo2C/CNT, Mo2C nanoparticles were inhaled on a CNT support (figure 2(b)) [48]. The lattice spacings of 0.28 nm match well with the (211) facets of Mo2C. Mo2C merges with CNT, hindering the accumulation of Mo2C nanoparticles and providing a resistance-less path suitable for fast electron transfer and electrolyte infiltration. EDS spectra of Mo2C/CNT shows almost 51 wt% Mo on the CNT (figure S1(a) and table S1 is available online at stacks.iop.org/NANO/29/295401/mmedia). The TEM image of S-Mo2C/CNT shows sulfur clusters dispersed on Mo2C/CNT (figure 2(c)), which is confirmed by EDS mapping of S-Mo2C/CNT (figures 2(d)–(g) and S1(b)). Furthermore, the similar intensity distribution of S to Mo suggests the bonding of S with Mo on the S-Mo2C/CNT (figures 2(e) and (f)). The chemical states of Mo, C, and S in the as-prepared Mo2C/CNT and S-Mo2C/CNT were investigated by XPS (figures 3(a)–(c) and S2). Mo2C/CNT showed the Mo 3d doublet located at a binding energy of 227.5 and 230.5 eV, characteristic of Mo2C (figure 3(a)) [49]. Besides, the Mo

3. Results and discussion The procedure for fabricating Mo2C/CNT and S-Mo2C/CNT is illustrated in scheme 1(a). Mo2C/CNT was synthesized by simple carburization of CNT-supported ammonium molybdate, and S was loaded into the Mo2C/CNT through a solution-based method [48]. The XRD patterns of CNT, Mo2C/CNT, and S-Mo2C/CNT show a characteristic diffraction peak (002) of CNT at 2θ=26.08° (figure 1(a)). Besides, the Mo2C/CNT shows several sharp peaks corresponding to Mo2C (JCPDS 770720). In the case of S-Mo2C/CNT, all the diffraction peaks were attributed to those of the sulfur (S8, JCPDS 08-0247), suggesting the loading of sulfur into the S-Mo2C/CNT network. As determined by TGA carried out under an Ar atmosphere, the S-Mo2C/CNT sample yields a 70 wt% S loading (figure 1(b)). 4

Nanotechnology 29 (2018) 295401

R Razaq et al

Figure 2. Morphology demonstration of the CNT, Mo2C/CNT, and S-Mo2C/CNT. TEM images of CNT (a), Mo2C/CNT (b), and S-Mo2C/CNT (c); SEM image (d) and elemental mappings corresponding to Mo (e), S (f), and C (g) for the S-Mo2C/CNT.

3d5/2 signals at 231.8 and 228.5 eV were characteristic of +6 and +4 oxidation states, respectively, due to the formation of a passivation layer (oxidation of the sample) when the Mo2C/CNT was exposed to ambient air during the process of the sample preparation [49]. However according to previous results on other metal-oxide host materials, the oxide passivation layer on Mo2C may play a role in adsorbing polysulfides [38]. As expected, S-Mo2C/CNT showed a Mo 3d5/2 peak at 230.9 eV assigned to the S-Mo2C bond with 0.8 eV higher bonding energy than that of MoS2 (figure 3(b)), which is attributed to the asymmetric environment and the electronic effect of Mo atoms at the surface [50, 51]. Primarily, figure 3(c) showed a doublet in the S 2p region at 162.8 and 164 eV corresponding to the S 2p3/2 and 2p1/2, respectively, which is characteristic of S8 [50]. A peak at 166.8 eV is attributed

to thiosulfate species [52]. Furthermore, ex situ XANES spectra of the S K-edge were recorded (figure 3(d)). Elemental sulfur shows a strong peak at 2472.8 eV and a broad peak at 2479.7 eV [53]. Comparatively, an additional peak at 2482.5 eV was observed for S-Mo2C/CNT and MoS2, which could be attributed to sulfate (thiosulfate) species [53], confirming the chemical bonding between sulfur and Mo2C on CNT. Lithium polysulfide adsorption experiments were performed to explore the trapping ability of Mo2C, CNT, and Mo2C/CNT (figure S3). For comparison, these three materials were dispersed into the pristine Li2S4 solution with stirring for 30 min and left undisturbed for 12 h in order to observe the change in the color. Distinctively, the originally yellowcolored pure Li2S4 solution changed into a completely colorless solution after the addition of Mo2C/CNT, suggesting 5

Nanotechnology 29 (2018) 295401

R Razaq et al

Figure 3. XPS spectra of (a) Mo 3d spectrum of Mo2C/CNT; (b) Mo 3d spectrum of S-Mo2C/CNT; (c) S 2p spectrum of S-Mo2C/CNT; and (d) S K-edge XANES spectra of S, MoS2, S-Mo2C/CNT, and S-Mo2C/CNT after discharge.

that Mo2C/CNT possesses higher adsorption capability for Li2S4 compared with Mo2C and CNT. The strong interfacial interaction between Mo2C/CNT and polysulfides may result in two effects. The first is the chemical confinement of polysulfides in the cathode. However, the specific surface areas of Mo2C, CNT, and Mo2C/CNT are ∼2, 109, and 80 m2 g−1, respectively, which implies that Mo2C/CNT can only capture inside a very limited amount of sulfur. Thus, the second effect might be prominent in which the anchored configuration promotes a charge transfer from conductive matrix to adsorbed polysulfides via Mo2C/CNT, subsequently speeding up the polysulfide redox reactions to enhance the redox kinetics of polysulfides. To confirm the redox kinetics of polysulfides, Mo2C/CNT, CNT, and Mo2C symmetrical cells were made using two of the same electrodes in Li2S6-containing electrolytes with 0 V open-circuit voltages [54]. CV tests within a voltage window of −0.7 to 0.7 V show that the current density of Mo2C/CNT considerably increased as compared with Mo2C and CNT (figure 4(a)), suggesting not only the strong interaction of Mo2C/CNT with LiPSs but also promoted speed-up of the electrochemical reactions of LiPSs. As bulk Mo2C is electrochemically active, the Mo2C nanoparticles which are strongly anchored into the CNT guarantee unique coupling effects between them. First, the robust conjugation

helps Mo2C merge strongly with CNT, providing a resistance-less path suitable for fast electron transfer and electrolyte infiltration. Secondly, this conjugation hinders the accumulation of Mo2C nanoparticles, thus smoothing highly reactive sites on the surfaces. Thirdly, the anchored configuration of Mo2C onto CNT provides access for the electric charge to reach the interface of the Mo2C-polysulfide and to prompt polysulfide redox reactions. This in turn enhances the redox kinetics of polysulfides. Such substantially facilitated charge transfer is further confirmed by EIS of symmetrical cells (figure 4(b)). The semicircle in the Nyquist plots, were remarkably reduced in the case of Mo2C/CNT compared with only Mo2C and CNT. The charge transfer resistance is 79.6, 767.7, and 1059.4 Ω for Mo2C/CNT, Mo2C, and CNT respectively, suggesting considerable enhancement of the redox kinetics of polysulfides and a decrease in the charge transfer resistance by introducing Mo2C/CNT. In order to demonstrate the electrocatalytic activity of Mo2C/CNT directly, CV of sulfur-based electrodes were performed (figure 4(c)). All the CV curves of the three materials (Mo2C/CNT, CNT, and Mo2C) show two typical cathodic peaks (reduction) corresponding to the disproportion of long-chain polysulfides and formation of Li2S2, and Li2S respectively. Similarly on the forward scan, the broad oxidation peak is attributed to the conversion of short-chain to 6

Nanotechnology 29 (2018) 295401

R Razaq et al

Figure 4. (a) Polarization curves of Mo2C/CNT, CNT, and Mo2C; (b) EIS spectra of symmetrical Li2S6−Li2S6 cells for Mo2C/CNT, Mo2C, and CNT; (c) CV profiles of S-Mo2C/CNT, S-CNT, and S-Mo2C; and (d) First cycle charge voltage profiles of Mo2C/CNT-Li2S, CNT-Li2S, and Mo2C-Li2S.

long-chain LiPSs and finally elemental sulfur. However, when the CV of Mo2C/CNT is compared with those of Mo2C and CNT, a distinct positive shift in reduction peak and negative shift in oxidation peak were detected, suggesting the superior catalytic activity of Mo2C/CNT towards the LiPS conversion and the resultant decrease in cell polarization. This is the first report on a distinct positive shift in reduction peak and negative shift in oxidation peak for metal carbides [34]. The oxidation of Li2S back to sulfur during battery charging is of crucial importance in achieving high reversible capacity and long cycling life [55]. Figure 4(d) shows the first charge capacity of Mo2C/CNT-Li2S, CNT-Li2S, and Mo2C-Li2S from open-circuit voltage to 4.0 V, in order to elucidate the initial activation energy barrier. Both CNT−Li2S and Mo2C−Li2S exhibit a high potential barrier at about 3.8 V, indicating a sluggish activation process with high charge transfer resistance. However, the addition of Mo2C in CNT significantly reduces the height of the potential barrier to 2.3 V as observed for Mo2C/CNT. The lower potential barrier and longer voltage plateau of the Mo2C/CNT-Li2S electrode compared with CNT-Li2S and Mo2C-Li2S indicate that the overpotential and charge transfer resistance are reduced to a great extent. This is in accordance with the above polarization (figure 4(a)) and EIS (figure 4(b)) results. The oxidation of Li2S back to sulfur during charging,

however, is not provided and is discussed in recently reported metal carbides [34]. The electrochemical performances of S-Mo2C/CNT, S-Mo2C, and S-CNT were measured by keeping the same amount of sulfur loading and electrolyte injection within 2025 coin cells. Figure 5(a) shows the cyclic performance of the S-Mo2C/CNT, S-Mo2C, and S-CNT electrodes at a current rate of 0.5 C (1 C=1672 mA h g−1) with the initial discharge capacities of 1206, 540, and 645 mA h g−1, respectively. After 100 cycles, S-Mo2C/CNT still has a capacity of over 955 mA h g−1, corresponding to a capacity decay of 0.1%. However, S-Mo2C and S-CNT show poor cycling. The rate capability of S-Mo2C/CNT was assessed at different discharge rates from 0.1–1 C rates. As shown in figure 5(b), S-Mo2C/CNT delivered a discharge capacity of 1483 mA h g−1 at 0.1 C, 1285 mA h g−1 at 0.2 C, 1169 mA h g−1 at 0.5 C, and 1011 mA h g−1 at 1 C. When the current density returns to 0.5, 2, and 0.1 C, the electrode shows a reversible capacity of 1124, 1225, and 1298 mAh g−1, respectively, suggesting excellent rate performance and the high stability of S-Mo2C/CNT. The galvanostatic charge/discharge behaviors of S-Mo2C/CNT, S-Mo2C, and S-CNT at charge/discharge rates of 0.1 C are shown in figure 5(c). Among all the charge/ discharge profiles of the three materials, S-Mo2C/CNT 7

Nanotechnology 29 (2018) 295401

R Razaq et al

Figure 6. (a) Cycling performance of S-Mo2C/CNT at 1 C over 100

cycles with a sulfur loading of 2.5 mg cm−2 and (b) Charge/discharge profiles of S-Mo2C/CNT at different current rates.

sulfur loading was increased to approximately 2.5 mg cm−2 by fabricating a thick electrode. The S-Mo2C/CNT cathode exhibits an initial capacity of 835 mA h g−1 at 1 C and still maintains a stable discharge capacity of 718 mA h g−1 over 100 cycle with excellent Coulombic efficiency (∼100%) (figure 6(a)). The rate capability and galvanostatic charge/discharge profiles of the S-Mo2C/CNT cathode with high areal sulfur loading of 2.5 mg cm−2 were performed at different current rates from 0.5–4 C (figure 6(b)). The S-Mo2C/CNT cathode shows discharge capacities of 940, 835, 778, 656, and 552 mA h g−1 at 0.5, 1, 2, 3, and 4 C, respectively. Postcycling characterizations provide the vital evidence on the strong chemical interaction between Mo2C/CNT and polysulfides in eliminating the shuttle effect. After 100 cycles, the Mo 3d spectrum (figure 7(a)) showed some similarity with that of the pristine S-Mo2C/CNT (figure 3(a)). However, the binding energy of Mo 3d5/2 for Mo4+ in figure 7(a) is 1.1 eV higher than that in figure 3(a), which is attributed to MoS2 rather than MoO2 [51]. This confirmed the formation of surface sulfide of Mo after cycled redox reactions, which is feasible considering that sulfur has a strong affinity to Mo as in the HDS catalysts in petroleum refining [57]. In the S 2p spectra after 100 cycles, various sulfur bonds can be observed (figure 7(b)). The newly present peaks in the range of 166–172 eV are attributed to LiTFSI, while those in the range of 159–164 eV are attributed to Li2S, Li2Sx and elemental sulfur S8 (S 2p), respectively [28, 58]. This is confirmed by

Figure 5. Electrochemical performances of Mo2C/CNT, CNT, and Mo2C cathodes: (a) Cycling performance and Coulombic efficiency of S-Mo2C/CNT, S-CNT, and S-Mo2C at 0.5 C for 100 cycles; (b) Rate performance of the S-Mo2C/CNT, S-CNT, and S-Mo2C cathodes at different current densities; (c) Galvanostatic charge/ discharge profiles of the S-Mo2C/CNT, S-CNT, and S-Mo2C cathodes at 0.1 C; (d) Prolonged cycle life and Coulombic efficiency at 2 C for the S-Mo2C/CNT electrode.

showed two typical discharge plateaus and one charge plateau in agreement with the CV curves (figure 4(c)). Figure 5(d) shows the long cyclic performance of S-Mo2C/CNT at the charge/discharge rate of 2 C. A high initial specific capacity of 802 mA h g−1 was obtained with almost 100% Coulombic efficiency. After 900 cycles, S-Mo2C/CNT still possess a capacity of 417 mA h g−1, corresponding to capacity decay of 0.05%. The stable cycling performance of S-Mo2C/CNT at 2 C is of great advantage. Compared with the mechanical mixture of Mo2C and CNT (Mo2C+CNT), Mo2C/CNT showed much better capacity (figure S4) suggesting the strong bonding between Mo2C and CNT as indicated above. For the practical application of Li–S batteries, a threshold of at least 2.0 mg cm−2 sulfur loading is acceptable for addressing the challenges in the way of the commercialization of a Li–S battery [56]. To address this problem, the areal 8

Nanotechnology 29 (2018) 295401

R Razaq et al

Table 1. Comparison of the electrochemical performance of this work with previously reported work related to Li–S batteries.

Catalysts Pt/G Co-N-GC CoS2/G RuO2 TiN/C WS2 MoS2−x/rGO PS/W2C Mo2C/CNT Mo2C/CNT

Initial capacity mAh g−1

Capacity retention mAh g−1

C-rate

No. of Cycles

Capacity decay per cycle %

Coulombic efficiency %

References

1100 1440 1368 912 1069 596 1159.9 1200 1206 802

789 850 1005 513.3 748 542 628.2 1128 866 417

0.2 0.2 0.5 0.5 0.2 0.5 0.5 0.2 0.5 2

100 200 150 400 50 360 500 200 100 900

0.28 0.20 0.26 0.10 0.53 0.02 0.09 0.06 0.1 0.05

99.3 Close to 100 N.A 92.5 N.A 99 99.6 98 Close to 100 Close to 100

[60] [61] [54] [62] [63] [52] [64] [34] This work This work

sulfate after sulfur loading were diminished upon discharge (figure 3(d)), suggesting that the superior cycle performance and high-rate capability of Mo2C/CNT is not derived from a traditional chemical bonding but mainly from a electrocatalytic process as discussed above [52]. To validate the superiority and stability of Mo2C/CNT for long cycling performance, the electrode films were examined after 100 cycles. After complete discharge, the cell was disassembled, and the electrode films were used for XRD and SEM characterization. No structural (figure S5) or morphological changes (figure S6) were observed, indicating the effective suppression of polysulfide shuttling and good mechanical strength in cell fabrication. Based on the characterization results, the well-designed Mo2C/CNT first provided a resistance-less path for fast electron transfer and accelerated the redox reaction kinetics. Secondly, the strong conjugation by supporting Mo2C on CNT hindered the accumulation of Mo2C nanoparticles, and thus provided highly active sites on the surfaces. Thirdly, sulfur loading on Mo2C/CNT by a solution-based process has advantages over ordinary heat treatment. As the ordinary thermal treatment (155 °C or higher) for a long time (12 h or more) requires a large consumption of energy to allow the solid sulfur to melt and diffuse into the pores, which resisted the electrolyte infiltration within the electrodes, limited the kinetic charge/discharge reactions, and decreased sulfur utilization. Last, CNT also exerted its high surface area and large pore volume for accommodating solid deposits of sulfur/Li2S [54]. In order to show the significance of Mo2C/CNT as an electrode for Li–S batteries, the resultant battery is compared with similar systems based on data published recently in leading journals (table 1). The initial capacity of 1206 mA h g−1 at 0.5 C with only a decay rate of 0.1% per cycle after 150 cycles, is the supreme capacity compared with noble metal electrocatalyst [60], metal dichalcogenides [52], and recently developed MoS2−x/rGO [64]. Further, after 900 cycles the fading rate of Mo2C/CNT was only 0.05% per cycle. However, other catalysts such as CoS2/G [54], RuO2 [62], MoS2−x/rGO [64], and recently published PS/W2C [34] after 200, 400 500, and 200 cycles, respectively, showed a capacity decay of 0.26%, 0.1%, 0.09%, and 0.06% per

Figure 7. Ex situ XPS study: (a) Mo 3d spectrum of discharged

S-Mo2C/CNT electrode and (b) S 2p spectrum of discharged S-Mo2C/CNT electrode.

the S K-edge XANES analysis of S-Mo2C/CNT after 100 cycles (figure 3(d)). The low-energy peaks in 2463–2468 eV and those at 2470.8 and 2473.7 eV represent the short-chain polysulfides Li2Sx and Li2S, respectively, as observed in figure 7(b) [59]. Meanwhile, the strong 2472.8 eV and weak 2479.7 eV peaks confirm that element sulfur remains in the cathode after discharging [53], which is also observed in S 2s XPS in figure 7(a). Importantly, the peaks related to (thio) 9

Nanotechnology 29 (2018) 295401

R Razaq et al

[3] Manthiram A 2011 Phys. Chem. Lett. 2 176–84 [4] Ellis B L, Lee K T and Nazar L F 2010 Chem. Mater. 22 691–714 [5] Bruce P G, Freunberger S A, Hardwick L J and Tarascon J M 2012 Nat. Mater. 11 19–29 [6] Yin Y X, Xin S, Guo Y G and Wan L J 2013 Angew. Chem. Int. Ed. 52 13186–200 [7] Yang Y, Zheng G and Cui Y 2013 Chem. Soc. Rev. 42 3018–32 [8] Mikhaylik Y V and Akridge J R 2004 J. Electrochem. Soc. 151 A1969–76 [9] Ji X, Lee K T and Nazar L F 2009 Nat. Mater. 8 500–6 [10] Schuster J, He G, Mandlmeier B, Yim T, Lee K T, Bein T and Nazar L F 2012 Angew. Chem. Int. Ed. 51 3591–5 [11] He G, Ji X and Nazar L 2011 Energy Environ. Sci. 4 2878–87 [12] Liang C, Dudney N J and Howe J Y 2009 Chem. Mater. 21 4724–30 [13] He G, Evers S, Liang X, Cuisinier M, Garsuch A and Nazar L F 2013 ACS Nano 7 10920–30 [14] Jayaprakash N, Shen J, Moganty S S, Corona A and Archer L A 2011 Angew. Chem. Int. Ed. 50 5904–8 [15] Zhang B, Qin X, Li G R and Gao X P 2010 Energy Environ. Sci. 3 1531–7 [16] Elazari R, Salitra G, Garsuch A, Panchenko A and Aurbach D 2011 Adv. Mater. 23 5641–4 [17] Zheng G, Yang Y, Cha J J, Hong S S and Cui Y 2011 Nano Lett. 11 4462–7 [18] Guo J, Xu Y and Wang C 2011 Nano Lett. 11 4288–94 [19] Zhou W, Yu Y, Chen H, DiSalvo F J and Abruna H D 2013 J. Am. Chem. Soc. 135 16736–43 [20] Fu Y and Manthiram A 2012 J. Phys. Chem. C 116 8910–5 [21] Liang X, Liu Y, Wen Z, Huang L, Wang X and Zhang H 2011 J. Power Sources 196 6951–5 [22] Pang Q, Kundu D, Cuisinier M and Nazar L F 2014 Nat. Commun. 5 4759 [23] Liang X, Garsuch A and Nazar L F 2015 Angew. Chem. Int. Ed. 54 3907–11 [24] Zheng G, Zhang Q, Cha J J, Yang Y, Li W, Seh Z W and Cui Y 2013 Nano Lett. 13 1265–70 [25] Qiu Y et al 2014 Nano Lett. 14 4821–7 [26] Lee K T, Black R, Yim T, Ji X and Nazar L F 2012 Adv. Energy Mater. 2 1490–6 [27] Ji X, Evers S, Black R and Nazar L F 2011 Nat. Commun. 2 325 [28] Tao X et al 2014 Nano Lett. 14 5288–94 [29] Li Z, Zhang J, Guan B, Wang D, Liu L-M and Lou X W 2016 Nat.Commun. 7 13065 [30] Zhao Z, Wang S, Liang R, Li Z, Shi Z and Chen G 2014 J. Mater. Chem. A 2 13509 [31] Zhou J, Li R, Fan X, Chen Y, Han R, Li W, Zheng J, Wang B and Li X 2014 Energy Environ. Sci. 7 2715–25 [32] Seh Z W, Yu J H, Li W, Hsu P C, Wang H, Sun Y, Yao H, Zhang Q and Cui Y 2014 Nat. Commun. 5 5017 [33] Dirlam P T et al 2016 ACS Appl. Mater. Interfaces 8 13437–48 [34] Zhou F, Li Z, Luo X, Wu T, Jiang B, Lu L L, Yao H B, Antonietti M and Yu S H 2018 Nano Lett. 18 1035–43 [35] Weizhai B, Dawei S, Wenxue Z, Xin G and Guoxiu W 2016 Adv. Funct. Mater. 26 4746–56 [36] Zhou T, Zhao Y, Zhou G, Lv W, Sun P, Kang F, Li B and Yang Q H 2017 Nano Energy 39 291–6 [37] Bokai C, Yong C, De L, Lihong Y and Yan M 2016 ChemSusChem 9 1–8 [38] Cui Z, Zu C, Zhou W, Manthiram A and Goodenough J B 2016 Adv. Mater. 28 6926–31 [39] Sun Z, Zhang J, Yin L, Hu G, Fang R, Cheng H M and Li F 2017 Nat. Commun. 8 14627 [40] Beibei W, Gang W and Hui W 2015 J. Mater. Chem. A 3 17403–11 [41] Lichun Y, Xiang L, Sina H, Gaohui D, Xiang Y, Jiangwen L, Qingsheng G, Renzong H and Min Z 2016 J. Mater. Chem. A 4 10842–9

cycle, respectively. All of these parameters suggest that our battery with Mo2C/CNT as the cathode material is placed at a high level among the top-performing Li–S batteries with high cycle life at current rate including recently published metal carbides [34] and will be one of the best choices for robust electrochemical performance.

4. Conclusion In summary, we have successfully demonstrated that the kinetics of polysulfide redox reactions in a Li–S battery can be enhanced using a Mo2C/CNT electrode, synthesized by a modified incipient wetness impregnation procedure. Benefiting from the excellent conductivity and strong chemical interaction with LiPSs, Mo2C/CNT was confirmed to be an efficient electrode, which remarkably reduced the energy barrier, completely eliminated polysulfide shuttle, and efficiently improved the redox kinetics of the Li–S battery. The S-Mo2C/CNT composite with areal loadings of sulfur (∼1.5–1.8 mg cm−2) exhibits reversible capacity of 1298, 1225, and 1124 mA h g−1 at 0.1, 0.2, and 0.5 C, respectively. After 900 cycles, S-Mo2C/CNT still had a capacity of 417 mA h g−1 at 2 C, corresponding to capacity decay of 0.05%. The Coulombic efficiency is always ∼100%. More significantly, the S-Mo2C/CNT cathode with a high areal sulfur loading of 2.5 mg cm−2 shows excellent rate capabilities up to 4 C and stable cycling performance over 100 cycles at 1 C. The advantages of Mo2C/CNT for a Li–S battery include the following: (1) strong chemical interaction with LiPSs; (2) high electrocatalytic activity for redox reactions of LiPSs; (3) high conductivity of interconnected carbon networks for increasing sulfur utilization and enhancing redox kinetics of polysulfides; (4) smooth pathways for transportation of electrons and Li ions; and (5) easy electrolyte infiltration throughout the electrode. This study presented a promising Mo2C/CNT electrode which is among the best reported to date.

Acknowledgments This work was supported by the National Natural Science Foundation of China (No. 21477046) and the Key Technology R&D Program of Shandong Province (No. 2016ZDJS11A03).

ORCID iDs Qian Li https://orcid.org/0000-0002-1219-8742 Zhaoliang Zhang https://orcid.org/0000-0002-8048-0600

References [1] Choi N-S et al 2012 Angew. Chem. Int. Ed. 51 9994–10024 [2] Yoo H D, Markevich E, Salitra G, Sharon D and Aurba C D 2014 Mater. Today 17 110–21 10

Nanotechnology 29 (2018) 295401

R Razaq et al

[54] Yuan Z, Peng H J, Hou T Z, Huang J Q, Chen C M, Wang D W, Cheng X B, Wei F and Zhang Q 2016 Nano Lett. 16 519–27 [55] Zhou G et al 2017 Proc. Natl. Acad. Sci. U.S.A. 114 840–5 [56] Peng H-J, Huang J-Q, Cheng X-B and Zhang Q 2017 Adv. Energy Mater. 7 1700260 [57] Bruix A, Fructose H G, Tuxen A K, Walton A S, Andersen M, Porsgaard S, Besenbacher F, Hammer B and Lauritsen J V 2015 ACS Nano 9 9322–30 [58] Demir-Cakan R, Morcrette M, Gangulibabu, Guéguen A, Dedryvère R and Tarascon J-M 2013 Energy Environ. Sci. 6 176–82 [59] Cuisinier M, Cabelguen P E, Adams B D, Garsuch A, Balasubramanian M and Nazar L F 2014 Energy Environ. Sci. 7 2697–705 [60] Al Salem H, Babu G, Rao C V and Arava L M 2015 J. Am. Chem. Soc. 137 11542–5 [61] Li Y-J, Fan J-M, Zheng M-S and Dong Q-F 2016 Energy Environ. Sci. 9 1998–2004 [62] Ding N et al 2016 Sci. Rep. 6 33154 [63] Jeong T-G, Choi D S, Song H, Choi J, Park S-A, Oh S H, Kim H, Jung Y and Kim Y-T 2017 ACS Energy Lett. 2 327–33 [64] Lin H, Yang L, Jiang X, Li G, Zhang T, Yao Q, Zheng G W and Lee J Y 2017 Energy Environ. Sci. 10 1476–86

[42] Ruirui L, Shuguang W, Wei W and Minhua C 2015 Phys. Chem. Chem. Phys. 17 24803–9 [43] Ling C, Hao J, Haibo J, Haoxuan Z, Shaojun G, Yanjie H and Chunzhong L 2017 Adv. Energy Mater. 7 1602782 [44] Tao M, Lirong Z, Jinwen Q, Di Z and Minhua C 2017 J. Mater. Chem. A 5 20228–38 [45] Hailin X, Yang H, Dongzhi L, Zhaozheng Q, Yemao L, Bo Y, Haosen F and Caizhen Z 2018 Appl. Surf. Sci. 441 69–76 [46] Qing G, Xinyu Z, Ying X, Di Z and Minhua C 2014 Nanoscale 6 6151–7 [47] Newville M 2001 J. Synchrotron Radiat. 8 322–4 [48] Chen W F, Wang C H, Sasaki K, Marinkovic N, Xu W, Muckerman J T, Zhu Y and Adzic R R 2013 Energy Environ. Sci. 6 943–51 [49] Pan L F, Li Y H, Yang S, Liu P F, Yu M Q and Yang H G 2014 Chem. Commun. 50 13135–7 [50] Liu G et al 2017 Nat. Chem. 9 810–6 [51] Yan Y, Ge X, Liu Z, Wang J Y, Lee J M and Wang X 2013 Nanoscale 5 7768–71 [52] Babu G, Masurkar N, Al Salem H and Arava L M 2017 J. Am. Chem. Soc. 139 171–8 [53] Zhu P, Song J, Lv D, Wang D, Jaye C, Fischer D A, Wu T and Chen Y 2014 J. Phys. Chem. C 118 7765–71

11