Tuning pseudocapacitance via C-S bonding in WS2 nanorods

Mar 27, 2018 - Pseudocapacitance plays an important role in high-power lithium-ion batteries. However, it is still lack of effective methods to tailor...
2 downloads 5 Views 3MB Size
Subscriber access provided by UNIV OF DURHAM

Article

Tuning pseudocapacitance via C-S bonding in WS2 nanorods anchored on N, S co-doped graphene for high power lithium batteries Yun Song, Shuo Bai, Lin Zhu, Mingyu Zhao, Dawei Han, Suhua Jiang, and Yong-Ning Zhou ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b02506 • Publication Date (Web): 27 Mar 2018 Downloaded from http://pubs.acs.org on March 27, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Tuning pseudocapacitance via C-S bonding in WS2 nanorods anchored on N, S co-doped graphene for high power lithium batteries Yun Song, Shuo Bai, Lin Zhu, Mingyu Zhao, Dawei Han, Suhua Jiang, Yong-Ning Zhou* This article is dedicated to the memory of Prof. Suhua Jiang (1976-2017) for her contribution in Materials Science Department of Materials Science, Fudan University, Shanghai 200433, P. R. China *Corresponding author: E-mail: [email protected]

Abstract: Pseudocapacitance plays an important role in high-power lithium-ion batteries. However, it is still lack of effective methods to tailor the pseudocapacitance contribution in electrode materials for lithium-ion batteries (LIBs). Herein, pseudocapacitance tuned by the strength of C-S bonding has been rendered in WS2 nanorods anchored on N, S co-doped three dimensional graphene hybrid (WS2@N,S-3DG) for the first time. The pseudocapacitive contributions in the charge storage can be enhanced effectively with the increased strength of C-S bonding. As expected, the enhanced extrinsic pseudocapacitance makes WS2@N,S-3DG a fascinating electrode material for high power LIBs, with a high reversible capacity of 509 mAh g-1 over 500 cycles at a current density as high as 2 A g-1. These encouraging results of pseudocapacitance tailored by chemical bonding provide new opportunities for designing advanced electrode materials.

Keywords: WS2; lithium-ion batteries; C-S bonding; graphene; pseudocapacitance

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 24

1. Introduction Transition-metal disulfides (TMDs) have drawn remarkable attention as electrode materials for lithium-ion batteries (LIBs), owing to their layered structure, low cost and high theoretical capacities.1-8 Among these layered metal sulfide materials, Tungsten disulfide (WS2) is covalently bonded to form 2D layers, stacking together through weak Van der Waals interaction.9,10 This structure makes Li+ very easy to be inserted/extracted, and facilitates subsequent lithium storage based on conversion reaction between Li2S and W. Furthermore, compared with the widely studied layered MoS2, WS2 has a higher intrinsic electrical conductivity, which should be beneficial for rate capability.11,12 However, recent studies showed that the application of WS2 as an anode material for LIBs suffered from severe capacity fading due to the large volume changes upon cycling, dissolution of the polysulfide intermediates and the poor Li+ diffusion kinetics. 13-15 A viable way to solve these problems is to synthesize small mechanically isolated but electrically connected WS2 particles which are confined in suitable carbon-based scaffold. 16-19 The mechanical stress generated upon charge-discharge cycling could be alleviated by reducing the particle size, and consequently prevent the particle pulverization. Introducing suitable carbon-based scaffold can not only serve as low resistance networks for electron transfer, but also adsorb polysulfide intermediates, consequently improving rate capability and cycle stability of WS2 electrode. On the basis of above conception, Shiva et al. fabricated WS2@RGO nanocomposite with a specific capacity of 349 mAh g-1 at a high current density of 1 A g-1. 20 Although progress has been achieved, the high rate performance of WS2 is still unsatisfying, especially for high energy and high power applications. To boost the rate capability, the critical step lies in radically changing the kinetics of charge transfer upon lithiation/delithiation. Augustyn et al. have done the pioneering work in this field, demonstrating that capacitive charge storage is advanced in rendering high charging rate, compared with the diffusion-controlled process in conventional LIBs materials. 21,22 Our previous study revealed that pseudocapacitance effect tends to arise in sheet-like structure coupled with graphene to improve the rate capability significantly. ACS Paragon Plus Environment

23,24

Page 3 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Furthermore, Shen et al have firstly proposed that pseudocapacitive contributions can be tuned by controlling the thickness of SnS.25 Apart from the size/thickness of the material, is there any other factor can tune the pseudocapacitive contributions? Herein, we have demonstrated that the pseudocapacitive contribution can be controlled by the strength of C-S bonding between WS2 and graphene, as confirmed by quantitative kinetics analysis. The control of C-S bonding strength can be simply realized by physical milling and chemical reaction. The pseudocapacitive contributions enhanced from 59 % to 67 % with the increased strength of C-S bonding at the same test condition. As expected, pseudocapacitance enhanced by robust C-S bonding has been rendered in WS2 nanorods anchored on N, S co-doped three dimensional graphene hybrid (WS2@N,S-3DG), exhibiting a high reversible capacity of 509 mAh g-1 over 500 cycles at a current density as high as 2 A g-1. This encouraging result of pseudocapacitance tailored by chemical bonding can be extended to other metal sulfides for highrate lithium- or sodium-ion storage. 2. Experimental Sections 2.1 Materials synthesis. WS2@N,S-3DG with strong C-S bonds is synthesized by a simple two-step method. Firstly, the WO3@N-3DG composite was synthesized by a hydrothermal synthesis and a heating process, as reported previously.26 Graphene oxide (GO) was prepared by a modified Hummers method, and subsequently dispersed in hydrochloric acid (pH=1.0). Afterwards, 1 g of ammonium tungstate ((NH4)10H2(W2O7)6), 2 g of ammonium sulfate ((NH4)2SO4), and 8g of oxalic acid (H2C2O4) were dissolved into the solution. The resulting solution was then mixed with the pyrrole, which was simultaneously used as the nitrogen source and the swelling agent. The resultant mixture solution was transferred for hydrothermal treatment at 180 °C for 24 h. After calcining at a high temperature 800 °C under argon, the composite was annealed at 450 °C in air. The prepared product is noted as WO3@N-3DG. Finally, the prepared WO3@N-3DG precursors were sulfided at different temperature (300 ℃, 400 ℃, 500 ℃) for 6 h in the presence of H2S gas to obtain WS2@N,S-3DG. To synthesize the pure WS2, 1 g of ammonium tungstate ((NH4)10H2(W2O7)6), 2 g of ammonium sulfate ((NH4)2SO4), and 8g of oxalic acid (H2C2O4) were dissolved into deionized water. The ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 24

mixture solution was transferred for hydrothermal treatment at 180 °C for 24 h. After calcining at a high temperature 800 °C under argon, the composite was annealed at 500 °C in H2S. The N,S-3DG was prepared by the same procedure as

synthesizing WS2@N,S-3DG without the introduction of tungsten-based

compounds. WS2+N,S-3DG with weak C-S bond was obtained by milling WS2 and N,S-3DG mixture with the same stoichiometric ratio as WS2@N,S-3DG. 2.2 Characterization. The morphology of the prepared samples were characterized using field-emission scanning electron microscopy (FE-SEM, LEO 1530 Gemini) and field-emission transmission electron Microscopy (FE-TEM, Tecnai G2 F20 S-Twin). To determine the crystalline structure, an X-ray diffractometer (D2 PHASER, Bruker AXS) was employed. Thermal gravimetric analysis (TGA) was carried out by a thermal analyzer (TG, Netzsch STA 409 PC) at a heating rate of 10 °C min-1. The chemical states of the samples were investigated by X-ray photoelectron spectroscopy (XPS) (Kratos UK). Raman spectra was carried out a LabRAM HR800 spectrograph and Fourier-transform infrared spectra (FT-IR) was recorded on a Nicolet iS50 spectrophotometer. 2.3 Electrochemical Measurements. The working electrodes were prepared by mixing the active powder (80 wt%), Super P (10 wt%) and polyvinylidene fluoride (PVDF, 10 wt%), then 1-methyl-2-pyrrolidinone (Aldrich, 99%) was added in the mixture. The formed slurry was then spread onto a copper foil by a doctor blade and subsequently dried at 120 °C. Half cells using Li metal as the counter electrode were assembled in an Ar filled glovebox, with polypropylene membrane as the separator. The electrolyte is the mixture of 1.0 M LiPF6 in ethylene carbonate and dimethyl carbonate (1:1). Galvanostatic cycling measurements were carried out on a LAND CT2001A battery test instrument. The cyclic voltammogram (CV) and electrochemical impedance spectroscopy (EIS) measurements were performed on the CHI660 electrochemical working station (CHI Instruments, TN). All the current densities and specific capacities are calculated on the basis of the total mass of the composite materials. 3. Results and discussions 3.1 Preparation and Characterization of WS2@N,S-3DG ACS Paragon Plus Environment

Page 5 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

WS2 nanorods anchored on N, S co-doped three dimensional graphene hybrid (WS2@N,S-3DG) has been successfully synthesized by a simple two-step method, as shown in Figure 1a. Firstly, WO3/3D nitrogendoped graphene (WO3@N-3DG) is prepared using a facile hydrothermal method. Then, WS2@N,S-3DG is synthesized by sulfidation of WO3@N-3DG by annealing in H2S atmosphere. The XRD patterns of the pristine WO3 sulfided at various temperatures are shown in Figure S1. The diffraction peaks of the pristine WO3 can be indexed to a monoclinic phase (JCPDS No.20-1324). The phase transition from monoclinic WO3 to hexagonal WS2 does not happen at a sulfidation temperature up to 300 °C. When the temperature increases to 400 and 500 °C, all peaks can be assigned to pure hexagonal WS2 phase (JCPDS No.08-0237), without any impurity peaks. The XRD patterns of pure WS2, WS2@N,S-3DG and WS2+N,S-3DG composites are compared in Figure 1b, from which two features can be observed: i) with respect to main peak of (002) reflection at 2θ = 14.5°, we can see that the peak intensity of them is in the trend of pure WS2 > WS2+N,S3DG > WS2@N,S-3DG, which can be attributed to the increase of interaction (bonding) between WS2 and N,S-3DG from pure WS2 to WS2@N,S-3DG. It is mainly because the graphene can prevent particles growing bigger during the thermal annealing process. 27, 28 In addition, based on theoretical calculation on sulfide and sulfide@rGO, when the chemical environment changes, the charge transfer will be rebalanced and reduced, leading to the increase of formation energy (>250 meV) per sulfide formula. The increased formation energy indicates the sulfide@rGO is thermodynamically less stable and more reactive, hence, will significantly enhance the reversible electrochemical reaction.29 ii) no XRD peak for graphene is observed, indicating the amorphous nature of N,S-3DG. The existence of N,S-3DG is further confirmed by TGA test as shown in Figure 1c. The mass ratio of WS2 to N,S-3DG in WS2@N,S-3DG is determined to be 68.8 : 31.2, as initially designed. The molar ratio of W and S in WS2@N,S-3DG is about 1:2.1, obtained by ICP measurement. In addition, the onset oxidation temperature for WS2@N,S-3DG in Figure 1c is approximately 100 °C, much lower than 450 °C for pure WS2, further confirming strong interfacial interactions between WS2 and N,S3DG are involved in WS2@N,S-3DG.24 The thermal by-product of WS2@N,S-3DG is found to be WO3, which is same as that of pure WS2 (Figure S2). Raman spectra in Figure 1d demonstrates the existence of WS2 and graphene in both WS2@N,S-3DG and ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 24

WS2+N,S-3DG hybrid. Two peaks at 350.4 and 416.6 cm-1 are attributed to the typical vibrations of W-S bond of WS2 phase. Peaks in ranges of 1217-1467 cm-1 and 1505-1669 cm-1 are assigned to the D band (sp3 hybridization) and G band (sp2 hybridization) of graphitic carbon.30 The ID/IG ratio of WS2@N,S-3DG, WS2+N,S-3DG and N,S-3DG are calculated to be 1.06, 1.08 and 1.17 respectively. The small ID/IG ratio of WS2@N,S-3DG and WS2+N,S-3DG composite means an enhanced graphitization degree of N,S-3DG in the composites, suggesting the electron transfer effect between WS2 and N,S-3DG.30 The electron transfer from WS2 to N,S-3DG, could reduce the N,S-3DG, thus resulting in more sp2 components in graphene, so enhancing the graphitization.29,31 In addition, the C-S bond located at 623 cm-1 in FT-IR spectrum of WS2@N,S-3DG gives a strong evidence that S bonded with N,S-3DG covalently and bridging WS2 and N,S3DG, thus enabling high stability of WS2@N,S-3DG.32 However, the C-S bond is too weak to be detected by FT-IR in WS2+N,S-3DG.

Figure 1. (a) Schematic illustration for the synthesis process of WS2@N,S-3DG composite, phase change from WO3@N-3DG to WS2@N,S-3DG; (b) XRD patterns of WS2, WS2+N,S-3DG and WS2@N,S-3DG; (c) TGA data of WS2@N,S-3DG and WS2; (d) Raman spectra and (e) FT-IR spectra of WS2, WS2+N,S-3DG and WS2@N,S-3DG and N,S-3DG, respectively.

ACS Paragon Plus Environment

Page 7 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

To further reveal how S atoms bridge N,S-3DG sheets with WS2, X-ray photoelectron spectroscopy (XPS) is employed to characterize the pure WS2 and WS2@N,S-3DG, as shown in Figure 2a. For pure WS2, three major peaks from W are located at 33.7, 35.9 and 39.2 eV, as presented in Figure 2b, while two major peaks from S are allocated at 163.3, and 164.5 eV, as shown in Figure 2c. Compared to pure WS2, both W (33.5, 35.5 and 39.0 eV) and S peaks (162.8 and 163.7 eV) of WS2@N,S-3DG (Figure 2d and e) show obvious shift to the lower energy, indicating an increased density of electron clouds around WS2. Given that chemically bonding with S makes graphene more electron-rich, and that WS2 is a p-type semiconductor, the electron clouds bias to WS2 from N,S-3DG, forming a strong electronic coupling between them. Furthermore, XPS spectra of S in WS2@N,S-3DG (Figure 2e) can be mainly de-convoluted into two doublets. The couple of strong peaks at 162.8 and 163.7 eV are attributed to S bonded with W. The other couple of peaks at 163.9 and 165.1 eV are attributed to S bonded with graphene covalently in a heterocyclic structure. Another peak at a higher energy of 168.1 eV could be attributed to a little carbon bonded with SOx.33 Compared with the XPS spectra of N,S-3DG (Figure S3), the peaks of C-S interaction in S spectra of WS2@N,S-3DG are much stronger than those of N,S-3DG, suggesting enhanced C-S bonding in WS2@N,S-3DG. The XPS analysis is well consistent with previous FT-IR results, further confirming the existence of C-S bond. For comparison, XPS spectra of WS2+N,S-3DG are shown in Figure S4. Both W (33.6, 35.7 and 39.1 eV) and S (163.3 and 164.1 eV) peaks of WS2+N,S-3DG show shifts to the lower energy, compared with pure WS2, but the shift amount is less than that of WS2@N,S-3DG, implying weaker electronic coupling between WS2 and N,S-3DG in WS2+N,S-3DG.

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 24

Figure 2. (a) XPS full survey spectra of pure WS2 and WS2@N,S-3DG. High-resolution XPS spectra of (b) W and (c) S for pure WS2. High-resolution XPS spectra of (d) W, (e) S, (f) C and (g) N for WS2@N,S-3DG.

ACS Paragon Plus Environment

Page 9 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

The C1s spectrum of WS2@N,S-3DG in Figure 2f can be de-convoluted into four peaks at 285.7, 286.8, 288.1 and 290.3 eV, assigning to C-C, C-S, C=O and HO-C=O bonds, respectively. The peak intensity of CS bond for WS2@N,S-3DG are much stronger than that for N,S-3DG (Figure S3), suggesting enhanced C-S bonding in WS2@N,S-3DG. N1s spectrum of WS2@N,S-3DG shown in Figure 2g is fitted by three peaks (401.1 eV, 400.1 and 398.9 eV), indicating the different existence of nitrogen in 3DG (graphitic, pyrrolic and pyridinic nitrogen, respectively). The pyrrolic and pyridinic nitrogen have been proved to introduce surface defects to form disordered honeycomb carbon structures, providing channels for fast Li+diffusion.26,30 The morphology and detailed structure of the WS2@N,S-3DG are examined by SEM and TEM characterization, as shown in Figure 3. The WS2@N,S-3DG has the similar morphology with its precursor WO3@N-3DG (Figure S5 and S6). As shown in Figure 3a and b, graphene randomly intersect to form fungus-like mesoporous N,S-3DG, on which WS2 nanorods lie flat. WS2 nanorods with lengths of proximately 200 nm are spread uniformly on the surface of voile-like N,S-3DG sheets, which are lined out in Figure 2c. In the corresponding HRTEM image (Figure 3d), the lattice fringe is clearly observed, which can be assigned to the (111) planes of the hexagonal WS2. These TEM observations are well consistent with the XRD data in Figure 1b.

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 24

Figure 3. (a) Low magnified and (b) high magnified SEM images of WS2@N,S-3DG; (c) TEM and (d) HRTEM images of WS2@N,S-3DG. 3.2 Electrochemical performance of WS2@N,S-3DG composite The electrochemical performance of WS2@N,S-3DG are characterized in coin cells. Figure 4a shows the cyclic voltammogram (CV) of WS2@N,S-3DG, which are collected at a scan rate of 0.1 mV s−1 for the first five cycles between 0.01 and 3.0 V. In the initial cathodic sweeping process, a small cathodic peak at about 1.68 V indicates an insertion process of lithium ions into WS2 lattice to form LixWS2.16,34 Two peaks can be seen at 0.69 and 0.52 V, ascribed to a multi-step electrochemical reaction process, including the formation of solid electrolyte interface (SEI) layers and the reduction of Wx+ to W0. 18,20 These two peaks of WS2@N,S3DG is lower than those of the pure WS2 (0.73 and 0.58 V) (Figure. S6), which may result from tuning effects of N,S-3DG to the formation potential of SEI layers and conversion reaction of WS2 phase.24 ACS Paragon Plus Environment

Page 11 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Moreover, the intense cathodic peak around 0.3 V is assigned to the intercalation of Li+ into N,S-3DG, as evidenced by the CV curves of N,S-3DG (Figure S7). In the first anodic sweep, a peak located at 2.43 V refers to the conversion of Li2S to S. It has been reported that the electrode after the initial cycle mainly consisted of elementary substance W and S rather than WS2.13,35 During the subsequent cycles, all the redox peaks overlap with those in the second cycle very well, suggesting good reversibility of WS2@N,S-3DG. Figure 4b shows the galvanostatic discharge-charge (GDC) curves of WS2@N,S-3DG at the current density of 2 A g-1 in the voltage range of 0.01-3 V versus Li+/Li. In the first discharge process, an obvious plateau of approximately 0.6 V is attributed to conversion reaction from WS2 to W. In the subsequent cycles, the plateaus of discharge/charge process represent the typical reaction between S and Li. The subsequent discharge/charge curves are in high coincidence, indicating a reversible and stable electrochemical performance. These typical plateaus for the WS2@N,S-3DG are well maintained upon charge-discharge cycling, but those of the pure WS2 vanish after the second cycle (Figure S8). To directly compare the cycle performance of WS2@N,S-3DG, WS2+N,S-3DG, N,S-3DG and pure WS2, specific capacities of them are obtained as a function of cycle number up to 500 cycles at a high current density of 2 A g-1 (Figure 4c). For WS2+N,S-3DG, pure WS2 and N,S-3DG, low specific capacities of about 140, 100 and 280 mAh g-1 are delivered after 500 cycles, respectively. In contrast, WS2@N,S-3DG yields a reversible capacity of 509 mAh g-1 upon 500 cycles. It should be mentioned here that the obtained capacity of WS2@N,S-3DG electrode is higher than the theoretical capacity of WS2 (432 mAh g-1). This phenomenon is quite common in nanostructured composite materials.36,37 It can be ascribed to two reasons. First, the pseudocapacitance behavior during electrochemical reaction38, which will be discussed in details later; Second, extra charge can be stored in the interface, which has been reported in many nanostructured anode materials. It has a typical capacity of 100-300 mAh g-1.

39,40

The cycling performance of WS2@N,S-3DG, WS2+N,S-3DG,

pure WS2 and N,S-3DG at different rates are compared in Figure 4d. The current density is increased stepwise from 0.1 to 2 A g-1 and turned back to 0.1 A g-1. Compared with the pure WS2 and N,S-3DG, the WS2@N,S-3DG composite exhibits much higher reversible capacity of 721, 686, 606, 543 and 490 mAh g-1 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 24

at current densities of 0.1, 0.2, 0.5, 1 and 2 A g-1, respectively. In addition, when the current density is reduced back to 0.1 A g-1 after high rate test, the capacity of WS2@N,S-3DG recovers to 688 mAh g-1, whereas those of the WS2+N,S-3DG, pure WS2 and N,S-3DG can only recover to 381, 248 and 376 mAh g-1, respectively. The rate capability of WS2@N,S-3DG are extraordinary compared with the reported WS2-based electrode materials for LIBs in the literatures, as shown in Table S1.

Figure 4. (a) Representative CV curve of the WS2@N,S-3DG for 1st – 4th cycles at a scan rate of 0.1 mV s-1; (b) galvanostatic charge-discharge curves of the WS2@N,S-3DG at a current density of 2 A g-1; (c) cycle performance of pure WS2, N,S-3DG, WS2+N,S-3DG and WS2@N,S-3DG at a current density of 2 A g-1; (d) rate capability of the pure WS2, N,S-3DG, WS2+N,S-3DG and WS2@N,S-3DG at different current densities. To investigate the electrode kinetics, EIS measurements are employed for WS2@N,S-3DG and pure WS2. As shown in Figure S9 (including detailed calculation methods), the Nyquist plots consist of depressed semicircles and inclined lines. The semi-circle in the middle and high frequency corresponds to charge transfer ACS Paragon Plus Environment

Page 13 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

resistance (Rct) at the interface of electrodes.41,42 After simulating the semi-circles of two samples, values of Rct of WS2@N,S-3DG and pure WS2 are calculated to be 53.6 and 78.1 Ω, respectively. The smaller Rct of WS2@N,S-3DG indicates that the electron transportation inside WS2@N,S-3DG is faster than that of pure WS2. The inclined lines in the low frequency are attributed to the Warburg impedance, generally reflecting lithium-ion diffusion within bulk electrode. After linear fitting and calculating, the lithium diffusion coefficients of WS2@N,S-3DG and pure WS2 are determined to be 1.31 × 10-12 and 8.13 × 10-13 cm2 s-1, respectively. This result clearly suggests that Li ion diffusion is facilitated in WS2@N,S-3DG compared with pure WS2. Additionally, SEM measurement is carried out to check the structure stability of WS2@N,S-3DG upon cycling. Figure S10 shows the SEM image of WS2@N,S-3DG after 500 cycles. It is observed that the mesoporous layered structure of N,S-3DG can be well maintained, implying the robust structure of the 3D graphene framework to endure the volume change upon cycling. However, the rod shape of WS2 cannot be observed after cycling, indicating the nanorod morphology of WS2 cannot be maintained, while well dispersed nanoparticles are observed on the 3D graphene framework, suggesting that WS2 nanorods transform to WS2 nanoparticles after conversion reaction. The stable mesoporous layered structure of N,S3DG and homogeneous WS2 nanoparticles without aggregation ensure the high capacity retention of the WS2@N,S-3DG. 3.3 Lithium storage mechanism of WS2@N,S-3DG From the slopy charge-discharge curves and the unique nanostructure of WS2@N,S-3DG, we believe the capacitance-related effect contributes an important part of rate capability for WS2@N,S-3DG. To study extrinsic pseudocapacitance effect of WS2@N,S-3DG, WS2+N,S-3DG and pure WS2, the CV at different scan rates from 0.1 to 1.2 mV s-1 have been tested. As shown in Figure 5a-c, the redox peaks show little shift with the increase of sweep rate, indicating reversible reactions and fast charge transfer. It has been demonstrated that Li storage mechanism (diffusion-controlled electrochemical reaction or capacitive process) could be estimated based on the following equations:23,24 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

i = avb

(1)

log(i) = b × log(v) + log(a)

(2)

Page 14 of 24

where a and b are adjustable parameters, i and v are the current and scan rate, respectively. When b = 1.0, the process is controlled by capacitive effect (extrinsic pseudocapacitance effect); When b = 0.5, the process is controlled by ion diffusion. When 0.5< b