Synthesis of molybdenum disulfide (MoS2) for lithium ion battery applications

S ip i Un g, N 252 252 Materials Research Bulletin 44 (2009) 1811–1815 of er X vati es. also ith hig g e to Contents lists available at ScienceDirect Materials Rese journa l homepage: www.e lse 1. Introduction During the past decade, the properties of molybdenum disulfide, especially those relating to its uses in such applications as hydrogen storage [1], catalysts [2], and lubricants [3,4], and its electrochemical double-layer capacitance [5], have been exten- sively investigated. Molybdenum disulfide features a layered structure, in which the atoms are covalently bonded to form two- dimensional layers that are stacked together throughweak van der Waals interactions [6]. The weak interlayer interaction allows foreign ions or molecules to be introduced between the layers through intercalation. Thus, MoS2 could be developed as an intercalation host to form a promising electrode material in high energy density batteries [7–9]. The electrochemical performance of transition metal disulfides, including MoS2, for Li-ion batteries has been investigated, and the results showed that the particle size and morphology of materials have a great influence on their electrochemical properties. For example, Julien [10] reported on the electrochemical behavior of crystalline WS2 powders and found that the lithium insertion capacity was only 0.6 mol Li+ per mole of crystalline WS2, while Dominko et al. [7] found that 1.7– 3.0 mol of Li+ could be inserted into onemole ofMoS2�xIy nanotube electrode, depending on the quality of the MoS2�xIy nanotubes. More recently, work from our group [8] demonstrated that WS2 nanotube electrode could deliver a reversible capacity of above 500 mAh/g, corresponding to 4.7 mol lithium per mole of WS2 nanotubes. We attributed such a high capacity to lithium intercalation into intratubular and intertubular sites of MS2 (M =W, Mo, etc.) nanotubes, as well as diffusion into the layered MS2 structure to form LixMS2 intercalation compounds. Since lithium ions can intercalate into small holes/channels, it may be imagined thatMS2 nanostructuredmaterials with plenty of defects can deliver even higher lithium intercalation capacity. In this work, we report a simple synthesis method (rheological phase reaction) to synthesize MoS2 nanoflakes [11,12]. The samples prepared can reversibly store lithium with a capacity of 1175 mAh/g in the voltage range of 0.01–3.0 V vs. Li/Li+, corresponding to 8 mol lithium per mole of MoS2, which is the highest capacity reported for MoS2 electrodes so far. Moreover, the MoS2 exhibited good cycling performance as an electrodematerial. 2. Experimental The MoS2 material was synthesized by using analytically pure (NH4)6Mo7O24�4H2O, sulfocarbamide (CS(NH2)2), and oxalic acid (H2C2O4�2H2O) as starting materials. The Mo/S/H2C2O4 molar ratio was 1:2:1. These powders were mixed and thoroughly ground in an agate mortar, and a few drops of water were added to form a rheological state mixture. The mixture was then put into a 50 ml sealed Teflon-lined autoclave and maintained at 200 8C for 24 h to B. Chemical synthesis B. Electrochemical properties � 2009 Elsevier Ltd. All rights reserved. * Corresponding author at: Institute for Superconducting & Electronic Materials, University of Wollongong, Wollongong, NSW 2522, Australia. Tel.: +61 2 4221 5225; fax: +61 2 4221 5731. E-mail address: zguo@uow.edu.au (Z. Guo). 0025-5408/$ – see front matter � 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.materresbull.2009.05.018 Synthesis of molybdenum disulfide (Mo Chuanqi Feng a,c, Jun Ma a, Hua Li a, Rong Zeng c, Za aKey Laboratory for Synthesis and Applications of Organic Functional Molecules, Hube b School of Mechanical, Materials & Mechatronic Engineering, University of Wollongon c Institute for Superconducting & Electronic Materials, University of Wollongong, NSW dARC Centre of Excellence for Electromaterials Science, University of Wollongong, NSW A R T I C L E I N F O Article history: Received 29 October 2008 Received in revised form 29 March 2009 Accepted 29 May 2009 Available online 6 June 2009 Keywords: A. Inorganic compounds A. Layered compounds A. Nanostructures A B S T R A C T This paper reports the use characterization by powd electronmicroscopy obser partly form MoS2 nanotub nanoflake electrode was higher specific capacity, w possible reasons for the discussed. The outstandin make it possible for MoS2 2) for lithium ion battery applications ing Guo b,c,d,*, Huakun Liu c,d iversity, Wuhan 430062, PR China SW 2522, Australia 2, Australia 2, Australia a rheological phase reaction method for preparing MoS2 nanoflakes. The -ray diffraction indicated that MoS2 had been formed. High resolution on revealed that the as-preparedMoS2 nanoflakes had started to curve and The lithium intercalation/de-intercalation behavior of as-prepared MoS2 investigated. It was found that the MoS2 nanoflake electrode exhibited very high cycling stability, compared to MoS2 nanoparticle electrode. The h electrochemical performance of the nanoflakes electrodes are also lectrochemical properties of MoS2 nanoflakes obtained by this method be used as a promising anode material. arch Bulletin vier .com/ locate /mat resbu form a dark gray solid. After that, the solidmixturewas packed into alumina crucibles and calcined in a tube furnace at 500 8C for 2 h under a flow of argon gas. Powder X-ray diffraction (XRD; Rigaku D/max-ra), using Cu Ka radiation (l = 1.5406 A˚) with a graphite monochromator, was employed to identify the crystalline phase of the synthesized materials. The morphology of the resulting compound was observed using a transmission electron microscope (TEM). The electrochemical characterizationswere performed using coin cells. The anode was prepared by dispersing 70 wt% as-prepared MoS2 powder and 20 wt% carbon black in 10 wt% polytetrafluoroethy- lene (PTFE) solution. TheMoS2 and carbon black powderswere first added to a solution of PTFE in isopropanol to form a homogeneous slurry. The slurry was then spread onto copper foil. The coated electrodes were dried at 125 8C for 24 h in vacuum and then pressed to enhance the contact between the active materials and the conductive carbon. Coin test cells were assembled in an argon filled glove box, where the counter electrode was Li metal and the electrolyte was 1 mol L�1 LiPF6 dissolved in a 50/50 vol% mixture of ethylene carbonate (EC) and diethyl carbonate (DEC). These cells were galvanostatically charged and discharged in the voltage range of 0.01–3.0 V to measure the electrochemical response at room temperature. hydrolysis of CS(NH2)2; (b) the reduction of Mo(VI) and the Fig. 2. Transmission electron micrographs of MoS2 sample. Fig. 3. SEM of MoS2 sample. C. Feng et al. /Materials Research Bulletin 44 (2009) 1811–18151812 3. Results and discussion 3.1. Reaction mechanism and structure characterization X-ray diffraction (XRD) was performed on the intermediate product (before the 600 8C calcination) and on the MoS2 final product (Fig. 1). The results show that the MoS2 phase is nearly completely formed after hydrothermal treatment in an autoclave at 200 8C for 24 h (Fig. 1(a)), but the precursor still contains some impurities resulting from the reactants used during the reaction process. A lot of small peaks appear in the precursor before calcinations. These peaks are resulted from impurities (raw material: H2C2O4, CS(NH2)2), when temperature arrived at 500 8C, these impurities were decomposed, so these small peaks were disappeared. The hexagonal phase of MoS2 was formed after sintering the intermediate product at 500 8C for 2 h under a flow of argon. All diffraction lines can be readily indexed to the hexagonal phase of MoS2 (JCPDS file No. 170744). The strong and clear peaks Fig. 1. XRD patterns of as-prepared MoS2 sample: (a) precursor before calcination; (b) sample formed after calcination at 500 8C. indicate that the MoS2 was highly crystalline. The crystal cell parameters were calculated by refining the XRD data: a = 0.3125 nm, c = 1.2302 nm. Fig. 2(a) shows a typical TEM image of the as-prepared MoS2 samples. It is obvious that some of the products have taken on the shapes of nanoflakes. The TEM images (see Fig. 2) of the samples revealed that the MoS2 nanoflakes had started to partly form nanotubes which have average diameter of about 240 nm. Fig. 3 contains an SEM image of a MoS2 sample. From the SEM, it can be seen that at least part of the nanotubes had to have been formed from nanoflakes. The growth process of the as-prepared MoS2 nanotubes is to some extent similar to themechanism proposed by Ye et al. for the tube-formation process in materials with layered structures [13], which involves curving followed by seaming of molecular layers. This indicates that MoS2 nanotubes could be formed under proper conditions by this synthesis method. Both Figs. 2 and 3 show some partially formed MoS2 nanotubes. Based on the literature [9,14–16] and the experimental conditions that we used, the formation of MoS2 and then MoS2 nanotubes may involve a complex process, which contains four steps: (a) the formation of MoO2; (c) the formation of MoS2; and (d) curving of nanoflakes to form tubes under heat treatment. The oxalic acid plays a key role as the reducing reagent, as well as a pH adjustment agent during the reaction process. Before reaction, the pH of reactants is about 1, after reaction, the pH of precursor in water is about 5. The one of roles of water is to react with CS(NH2)2 to produce H2S (middle product), the other is to speed up diffusion of ions in reactant. While the sulfocarbamideworks as a sulfurization reagent. The reaction process for the synthesis of MoS2 could be expressed as follows: CSðNH2Þ2þ2H2O ! 2NH3þCO2þ2H2S (1) ðNH4Þ6Mo7O24þ7H2C2O4 ! 6NH3þ7MoO2þ14CO2þ10H2O (2) MoO2þ2H2S ! MoS2þ2H2O (3) The overall reactions could be expressed as (4): into the layer sites ofMoS2 nanoflakes to form LixMoS2 compounds. Once these sites are occupied with Li+ ions, lithium ions will intercalate into MoS2 defect sites in nanoclusters, intertubular sites, and intratubal sites (the hollow core) in the MoS2 nanotubes. During the de-intercalation process, lithium ions are extracted from the intratubal sites, the defect sites in nanoclusters, and the layer sites of LixMoS2 in the nanoflakes, successively, which deliver the reduction/oxidation peaks in the CV curves at lower potentials. The CV curves become stable after the 5th cycle, in the 5th cycle and the 8th cycle, two CV curves are nearly overlapped, which verified that MoS2 material has excellent cycle performances as anode material. The electrochemical properties of the as-prepared MoS2 were also measured via coin cell testing. Fig. 5 shows the charge/discharge profiles of as-prepared MoS2 electrode. In the first cycle, the MoS2 electrode delivered a lithium insertion capacity of about 1174.7 mAh/g when the discharge current density was 60 mA/g, which is much higher than in the reported data on MoS2 nanopowder [8,10]. The as-prepared MoS2 electrode retained a reversible capacity of 851.5 mAh/g after 20 cycles. The first discharge curve shows three insertion plateaus at �1.0 V, �0.8 V, and �0.4 V. In the second cycle, three lithium insertion plateaus at �2.1 V, �1.1 V and �0.4 V were observed, as well as a C. Feng et al. /Materials Research Bulletin 44 (2009) 1811–1815 1813 ðNH4Þ6Mo7O24þ7H2C2O4þ14CSðNH2Þ2þ4H2O ! 7MoS2þ28CO2þ34NH3 (4) During these reactions, the increased reaction entropy facil- itates the formation of the expected product (MoS2). This facile method provides a simple and easily applicable route to synthesize MoS2 nanoflakes at moderate temperature. It can also be used to synthesize other transition metal sulfides. 3.2. Electrochemical properties Cyclic voltammetry (CV) was performed on the MoS2 electrode, and selected cyclic voltammograms are shown in Fig. 4. During the first cycle, there are three reduction peaks (at �1.0 V, �0.8 V, and �0.2 V) and three corresponding oxidation peaks (at �1.5 V, �1.8 V, and �2.25 V). The reduction peaks at �1.0 V and �0.8 V overlap with each other, while the two oxidation peaks at the corresponding potentials (�1.5 V, �1.8 V) are also joined together. From the second CV cycle, three pairs of reduction and oxidation peaks were observed. There was no significant change in the potentials of the oxidation peaks, but the potentials of the reduction peaks shifted from their original positions (�1.0 V, �0.8 V, and �0.2 V) to �2.0 V, �1.1 V, and �0.3 V, respectively. Based on the above situation and our previous research [8], we suggest that, in the first lithiation process, lithium ions intercalate Fig. 4. Cyclic voltammograms of theMoS2 electrode in a coin cell (vs. Li) for selected cycles. Fig. 5. Typical charge and discharge curves for selected cycles of an as-prepared MoS2 electrode. (a) For voltage range from 0.01 V to 3.0 V; (b) for a voltage range from 0.3 V to 3.0 V. Current density was 40 mA/g for both (a) and (b). slope starting from 0.4 V and running down to the cut-off voltage of 0.01 V. TheMoS2 nanoflake electrodes were also investigated for different voltage ranges. When the voltage range was from 0.3 V to 3.0 V, the capacity of the electrode decreased, however, it shows slightly better cycling performances compared to the electrode cycled within the voltage range of 0.01–3 V, as shown in Fig. 5(b). During the first charge, the charge voltage increased gradually at first, and then a short plateau at about 2.1 V vs. Li/Li+ was observed. From the second cycle, the discharge plateau is located at about 2.1 V, while the discharge plateaus at lower potential that were observed in the first cycle have disappeared. The charge/ discharge behavior of the as-prepared MoS2 electrode is similar to that of MoS2 nanotube electrodes, which is described in other reports [7,8]. It should be noted that the irreversible capacity in the first cycle is high (about 184 mAh/g), which may caused by (1) the decomposition of electrolyte on the surface of the MoS2 to form a passivation layer on the electrode; and (2) the fraction of Li+ ions that were trapped in the nanoclusters or defect sites/intratubal sites in the MoS2 nanoflakes and nanotubes induced irreversible capacity. From the second cycle, only one plateau (2.2 V for the charge curve and 2.1 V for the discharge curve) connected with a Acknowledgments Financial support provided by the Australian Research Council (ARC) through an ARC Discovery project (DP0878611) and by the Ministry of Education, China, through the Key Laboratory for Synthesis and Applications of Organic Functional Molecules is gratefully acknowledged. Moreover, the authors would like to thank Dr. Tania Silver at the University of Wollongong for critical reading of the manuscript. References [1] J. Chen, N. Kuriyama, H.T. Yuan, H.T. Takeshita, T. Sakai, J. Am. Chem. Soc. 123 (2001) 11813–11814. [2] M.Y. Sun, J. Adjaye, A.E. Nelson, Appl. Catal. A: Gen. 263 (2004) 131–143. [3] M. Chhowalla, G.A.J. Amaratunga, Nature 407 (2000) 164–167. [4] Y.Q. Zhu, T. Sekine, Y.H. Li, W.X. Wang, M.W. Fay, H. Edwards, P.D. Brown, N. Fleischer, R. Tenne, Adv. Mater. 17 (2005) 1500–1503. [5] J.M. Soon, K.P. Loh, Electrochem. Solid-State Lett. 10 (2007) A250–A254. [6] B.G. Sibernagel, Solid State Commun. 17 (1975) 361. [7] R. Dominko, D. Arcon, A. Mrzel, A. Zorko, P. Cevc, P. Venturini, M. Gaberscek, M. Remskar, D. Mihailovic, Adv. Mater. 14 (2002) 1531. [8] G.X. Wang, S. Bewlay, J. Yao, H.K. Liu, S.X. Dou, Electrochem. Solid-State Lett. 7 (2004) A321. C. Feng et al. /Materials Research Bulletin 44 (2009) 1811–18151814 slope can be observed for both charge and discharge curves, indicating that the intercalation reaction dominates the electro- chemical processes after the first cycle. This is in agreement with the cyclic voltammograms shown in Fig. 4. Although the MoS2 electrode has a relatively high discharge potential, it still can be accepted as an anode material for most electronic devices. The cycling behavior within 40 cycles is shown in Fig. 6. As mentioned above, we can observe a large irreversible capacity in the first cycle, but the coulombic efficiency in the subsequent cycles is high (96%), and the average fade in specific capacity is just 1.5% per cycle. When the coin cell was operated at different current densities, the first discharge capacities were different. This resulted from the different quantities of Li ions inserted per mole of MoS2, as shown in Fig. 7. The quantity of inserted Li ions reached as high as 8 mol per mole of MoS2 during the initial discharge process. To speculate as to why the MoS2 nanoflake electrode showed such a high capacity, we propose that lithium ions not only intercalate into layers of the MoS2 structure to form LixMoS2 compounds, but also intercalate into defect sites of nanoclusters and intratubal sites (the hollow core) in MoS2 nanoflakes and nanotubes. So the morphology and size particles of MoS2 have effects to its capacity when they are used as electrodes [17–19]. Considering the high reversible Fig. 6. Typical cycling performance of an as-prepared MoS2 electrode for different ranges of voltages under current density 40 mA/g: (a) 0.3–3.0 V and (b) 0.01–3 V. capacity and good cycling stability of the as-prepared MoS2 nanoflake electrodes, MoS2 nanoflakes could be a promising alternative anode material for lithium ion batteries. 4. Conclusions In this study, a novel method (the rheological reaction method) was used to synthesize MoS2 nanoflakes (partly formed nano- tubes). The synthesized MoS2 shows stable cyclability over a wide voltage range. (The reversible capacity remains 840 mAh/g after 20 cycles, which is 84% of the initial reversible capacity.) There are four possibilities for lithium intercalation in the MoS2 nanoflake electrodes: (1) lithium ions intercalate into nanoflake clusters; (2) lithium ions intercalate into defect sites in nanoflakes (partly formed nanotubes); (3) lithium ions intercalate into intratubal sites (the hollow core) through the open end; and (4) lithium ions intercalate into the MoS2 layer sites to form LixMoS2. All these four possibilities contribute to the high lithium insertion capacity of MoS2 nanoflakes electrodes. Nanoflakes, i.e. partly formed nanotube materials, may indicate a new possible direction to further improve the electrochemical performance of electroche- mically active materials. Fig. 7. The quantity of Li ions inserted at different working current densities: (a) 40 mA/g and (b) 60 mA/g. [9] X.L. Li, Y.D. Li, J. Phys. Chem. B 108 (2004) 13893. [10] C.M. Julien, Mater. Sci. Eng. R 40 (2003) 47. [11] J.T. Sun, W. Xie, L.J. Yuan, K.L. Zhang, Mater. Sci. Eng. B 64 (1999) 157. [12] C.Q. Feng, H. Tang, K.L. Zhang, J.T. Sun, Mater. Chem. Phys. 80 (2003) 573. [13] C.H. Ye, G.W. Meng, Z. Jiang, Y.H. Zhang, G.Z. Wang, L.D. Zhang, J. Am. Chem. Soc. 124 (2002) 15180. [14] J.A. Wilson, A.D. Yoffe, Adv. Phys. 18 (1969) 193. [15] X.L. Li, J.P. Ge, Y.D. Li, Chem. Eur. J. 10 (2004) 6163. [16] Y.Q. Zhu, T. Sekine, K.S. Brigatti, et al. J. Am. Chem. Soc. 125 (2003) 1329. [17] H. Li, W. Li, L. Ma, W. Chen, J. Wang, J. Alloys Compd. 471 (2009) 442–447. [18] L.P. Xu, Y.S. Ding, C. Chen, L.L. Zhao, C. Rimkus, R. Joesten, S.L. Suib, Chem. Mater. 20 (2008) 308–316. [19] Q. Pan, H. Jin, H. Wang, G. Yin, Electrochim. Acta 53 (2007) 951–956. C. Feng et al. /Materials Research Bulletin 44 (2009) 1811–1815 1815 Synthesis of molybdenum disulfide (MoS2) for lithium ion battery applications Introduction Experimental Results and discussion Reaction mechanism and structure characterization Electrochemical properties Conclusions Acknowledgments References