Fluorine substituted molybdenum oxide as cathode material for Li-ion battery

Fluorine substituted molybdenum oxide as cathode material for Li-ion battery

Materials Letters 160 (2015) 175–178 Contents lists available at ScienceDirect Materials Letters journal homepage: www.elsevier.com/locate/matlet F...

1MB Sizes 1 Downloads 21 Views

Materials Letters 160 (2015) 175–178

Contents lists available at ScienceDirect

Materials Letters journal homepage: www.elsevier.com/locate/matlet

Fluorine substituted molybdenum oxide as cathode material for Li-ion battery D.P. Opra a,n, S.V. Gnedenkov a, A.A. Sokolov a,b, A.B. Podgorbunsky a, N.M. Laptash a, S.L. Sinebryukhov a a b

Institute of Chemistry, FEB RAS, Vladivostok 690022, Russia Far Eastern Federal University, Vladivostok 690950, Russia

art ic l e i nf o

a b s t r a c t

Article history: Received 6 May 2015 Received in revised form 21 July 2015 Accepted 22 July 2015 Available online 23 July 2015

Molybdenum oxyfluoride MoO2.8F0.2 has been synthesized by solid-phase method. Its morphological and structural features were studied by scanning electron microscopy and X-ray powder diffraction. It was established that the substitution of the O atoms for F atoms in MoO3 structure leads to charge redistribution in the crystal lattice and conductivity increasing from 4.4  10  9 S cm  1 to 1.8  10  6 S cm  1. It was shown that molybdenum oxyfluoride characterized by higher discharge–charge stability in comparison with the oxide itself: 10-fold cycling in the range from 3.5 to 1.5 V yields 160 mA h g  1 and 120 mA h g  1 for MoO2.8F0.2 and MoO3, respectively. & 2015 Elsevier B.V. All rights reserved.

Keywords: Molybdenum oxyfluoride Li-ion battery Cathode Atomic substitution Reversible capacity

1. Introduction Li-ion batteries (LIB) are the commonly used energy sources for state-of-art high-technology applications. At the same time the capacity of the most widespread LiCoO2 cathode is much smaller (120–150 mA h g  1) than that of the graphite anode (350–365 mA h g  1). In addition the Co toxicity and high cost limits the usage of LIBs for high-energy applications, e.g., hybrid and electric vehicles, uninterruptible power supplies, unmanned underwater vehicles [1–3]. Note, that the other popular cathodes, e.g., LiMn2O4, LiNiO2, and LiFePO4 are more cost-effective and environmental friendly in comparison with LiCoO2. However, the low specific capacity is also typical of these cathode materials [4– 6]. In this way, the synthesis of high energy, safety, and low cost cathode materials is a fundamental problem of LIB field. MoO3 with orthorhombic crystal lattice (space group Pbnm) is a promising cathode for LIB due to its thermodynamically stable two-dimensional layered structure. Each layer is formed by MoO6 octahedra that share edges in the (001) direction and are connected by corners in the (100) direction. The interlayer interaction is mainly through the van der Waals forces between oxygen atoms. Two types of vacant sites (intralayers and interlayers) are available in MoO3 crystal structure for hosting foreign ions like lithium [7,8]. n

Corresponding author. E-mail address: [email protected] (D.P. Opra).

http://dx.doi.org/10.1016/j.matlet.2015.07.119 0167-577X/& 2015 Elsevier B.V. All rights reserved.

The theoretical capacity of molybdenum oxide vs. Li þ /Li reaches C ¼1117 mA h g  1, the volume change during Li þ intercalation– deintercalation is calculated to be as much as 104% [9,10]. However, slow solid-state diffusion of Li þ as well as low conductivity decrease the reversible capacity of MoO3 [11–13]. According to the reports [14–16] partial substitution of the O atoms for F atoms is a promising way for improvement of cycle performance of metal oxides. For example, it was established that fluorine substituted LiFe0.4Mn0.6PO4, Li3V2(PO4)3, and LiNi0.8Co0.1Mn0.1O2 provide stable cycling even at high-rate up to 10C. In this work molybdenum oxyfluoride MoO2.8F0.2 was prepared by solid state synthesis resulted in substitution of the O atoms for F atoms (the F/O substitution) in the MoO3 structure. The effects of atomic substitution on structural, morphological and electrochemical parameters were investigated.

2. Experiment Nonstoichiometric molybdenum oxyfluoride MoO2.8F0.2 was synthesized by interaction of two solids: H2MoO4 and NH4HF2 (the mole ratio of 1:0.3–0.5) at 420 710 °C. For comparative analysis of physicochemical and electrochemical parameters of MoO2.8F0.2 the commercial MoO3 (99.9% purity) was investigated also. For the particle size averaging the MoO2.8F0.2 and MoO3 was grinding in a planetary mill Fritch Pulverisette 7 (Germany) during

176

D.P. Opra et al. / Materials Letters 160 (2015) 175–178

20 cycles. Each cycle included an active (grinding during 10 min at 450 rpm) and passive stage (25 min pause). The milling process was conducted using a zirconia container and balls. The morphology was investigated by scanning electron microscopy (SEM) using Hitachi S5500 (Japan) microscope. The crystal structure was characterized by X-ray diffraction (XRD) on Bruker D8-Advance (Germany) diffractometer. The conductivity was determined by electrochemical impedance spectroscopy (EIS) at room temperature using Solartron SI 1260 (Great Britain) analyzer. The working electrode was consisted of 80 wt% MoO2.8F0.2 (or MoO3), 10 wt% carbon black, and 10 wt% polyvinylidene fluoride. Its area was equal to 1.75 cm  2 (round disc). The mass of the active material was approximately 5 mg cm  2. Lithium metal disc (thickness of 0.1 mm) was applied as both counter and reference electrodes. 1 M solution of LiBF4 salt in the miturex of propylene carbonate and dimethoxyethane at a volume ratio of 3:1 was used as an electrolyte. The design of electrochemical half-cell is described in detail in [3–5]. The electrochemical performance was evaluated using a Solartron 1470E (UK) potentiostat/galvanostat. The parameters were measured by the galvanostatic discharge–charge at current density of 30 mA g  1 in the range from 3.5 to 1.5 V during 10 cycles. The measurements were carried out at least on six half-cells.

3. Results and discussions The structure and chemical composition of MoO2.8F0.2 was investigated previously [17,18]. SEM method shows that the morphology of molybdenum oxyfluoride (Fig. 1) and commercial MoO3 after grinding is similar. The high magnification SEM analysis of MoO2.8F0.2 reveals the heterogeneous surface of particles that seems to facilitate the electrochemical reaction with Li þ ions. XRD patterns (Fig. 2) of MoO2.8F0.2 and commercial MoO3 are

Fig. 1. SEM-images for MoO2.8F0.2.

Fig. 2. XRD patterns of MoO2.8F0.2 and commercial MoO3.

similar and correspond to the orthorhombic phase (JSCD no. 01089-5108, space group Pbnm), but their lattice parameters are somewhat different. Partial substitution of the O atoms for F atoms leads to the charge redistribution in the lattice due to the reduction of Mo6 þ to Mo5 þ . Accordingly the X-ray photoelectron spectroscopy data two doublets with the binding energies of 233.5 and 232.0 eV and their intensity ratio [17] confirm the oxyfluoride composition determined also chemically on fluorine content [18], that means the F/O atomic substitution. Impedance spectra (Fig. 3) for both MoO2.8F0.2 and commercial MoO3 consist of high frequency semicircle corresponds to the material conductivity and low frequency arc related to double layer at the sample/electrode interface. According to the EIS results the partial F/O substitution increases the conductivity of material, possibly, due to charge redistribution in the crystal lattice. In particular the conductivity of MoO2.8F0.2 was 1.8  10  6 S cm  1 while the MoO3 yields only 4.4  10  9 S cm  1. The discharge–charge cycling of MoO2.8F0.2 and commercial MoO3 from 3.5 to 1.5 V vs. Li þ /Li at current density of 30 mA g  1 is

Fig. 3. Impedance diagrams for molybdenum oxyfluoride and oxide.

D.P. Opra et al. / Materials Letters 160 (2015) 175–178

177

The 10-fold cycling shows the decrease of specific capacities for both MoO2.8F0.2 and MoO3. At the same time, the intercalation capacity of oxyfluoride (170 mA h g  1) exceeds the capacity of oxide (140 mA h g  1). The reversible capacity after 10th cycle attains to 160 and 120 mA h g  1 for MoO2.8F0.2 and MoO3, respectively. Thus, the Columbic efficiency of the molybdenum oxyfluoride, after the 10th charge–discharge cycle is equal to 94%. The results enable the conclusion about the enhanced reversibility of the Li þ intercalation–deintercalation into the structure of MoO2.8F0.2 in comparison with MoO3. Most likely, partial F/O substitution occurs at the terminal positions of ligand in polyhedron, possibly [22], shielding the electrostatic interaction between the Li þ ions and [MoO6] octahedron layers. At the same time, it should be noted that the observed capacity enhancement is not very high. Thus, more research for synthesis conditions is required to further improve the electrochemical performance of molybdenum oxyfluoride. It seems that the quantitative dependence for F/O atomic substitution should be investigated first and foremost.

4. Conclusions The synthesis, structure, morphology, and electrochemical properties of nonstoichiometric oxyfluoride MoO2.8F0.2 with the lattice parameters close to the layered orthorhombic MoO3 were reported. Partial substitution of the O atoms for F atoms enhances the cycle performance of material as a cathode for Li-ion battery due to shielding the electrostatic interaction between the Li þ ions and [MoO6] octahedron layers. By the galvanostatic 10-fold discharge–charge cycling it was observed that deintercalation capacity for MoO2.8F0.2 (165 mA h g  1) is higher than for commercial MoO3 (120 mA h g  1). Conductivity of material after the F/O atomic substitution leading to charge redistribution in the crystal lattice is increased by three orders of magnitude: from 4.4  10  9 S cm  1 (for molybdenum oxide) to 1.8  10  6 S cm  1 (for oxyfluoride). Thus, the method opens new pathways to the synthesis of prospective cathode materials for Li-ion batteries.

Fig. 4. Discharge–charge curves of MoO2.8F0.2 and commercial MoO3 for 1, 5, and 10 cycles.

shown in Fig. 4. The intercalation capacities during the 1st cycle for the molybdenum oxyfluoride and oxide are equal to 290 and 280 mA h g  1, respectively. According to [19–21], the electrochemical interaction in the voltage range 3.5–1.5 V occurring as follows:

xLi+ + xe – + MoO3 ↔ Lix MoO3 , 0 ≤ x ≤ 1.5 include the Li þ ions intercalation into the [MoO6] octahedron intralayers (vacant sites in MoO6 octahedra) and interlayer spacing between the [MoO6] octahedron layers. On the discharge curves this feature is characterized by change of operating voltage: insertion of Li þ into the intralayers occurs at 2.7 V, while the intercalation into space between layers proceeds from 2.5 to 1.5 V. The charge of the 1st cycle characterizing the Li þ deintercalation yields 245 and 230 mA h g  1 for MoO2.8F0.2 and MoO3, respectively. So deintercalation capacity increased after F/O substitution. The character of 5th discharge as well as of charge curves is similar to that of previous cycles. The 5th intercalation capacity of MoO2.8F0.2 is equal to 230 mA h g  1, while commercial MoO3 yields 210 mA h g  1. Corresponding deintercalation capacities are equal to 220 and 190 mA h g  1 for oxyfluoride and oxide, respectively.

Acknowledgments This work was supported by the Russian Science Foundation (Grant no. 14-33-00009) and Russian Government (Federal Agency of Scientific Organizations).

References [1] S.A. Novikova, S.A. Yaroslavtsev, V.S. Rusakov, T.L. Kulova, A.M. Skundin, A.B. Yaroslavtsev, Mendeleev Commun. 23 (2013) 251–252. [2] M.V. Koudriachova, N.M. Harrison, S.W. de Leeuw, Solid State Ion. 152–153 (2012) 189–194. [3] S.V. Gnedenkov, D.P. Opra, S.L. Sinebryukhov, J. Alloy. Compd. 621 (2015) 364–370. [4] V.G. Kuryavyi, A.Yu. Ustinov, D.P. Opra, G.A. Zverev, T.A. Kaidalova, Mater. Lett. 137 (2014) 398–400. [5] S.V. Gnedenkov, D.P. Opra, V.G. Kuryavyi, S.L. Sinebryukhov, A.Yu. Ustinov, V.I. Sergienko, Nanotechnol. Russ. 10 (2015) 353–356. [6] S.V. Makaev, V.K. Ivanov, O.S. Polezhaeva, Yu.D. Tret’yakov, T.L. Kulova, A.M. Skundin, O.A. Brylev, Russ. J. Inorg. Chem. 55 (2010) 991–994. [7] A.M. Hashem, H. Groult, A. Mauger, K. Zaghib, C.M. Julien, J. Power Sources 219 (2012) 126–132. [8] U.K. Sen, S. Mitra, RSC Adv. 2 (2012) 11123–11131. [9] E. Shembel, R. Apostolova, V. Nagirny, I. Kirsanova, P. Grebenkin, P. Lytvyn, J. Solid State Electrochem. 9 (2005) 96–105. [10] R. Nadimicherla, W. Chen, X. Guo, Mater. Res. Bull. 66 (2015) 140–146. [11] F. Leroux, L.F. Nazar, Solid State Ion. 133 (2007) 37–50. [12] C.-L. Liu, Y. Wang, C. Zhang, X.-S. Li, W.-S. Dong, Mater. Chem. Phys. 143 (2014) 1111–1118.

178

D.P. Opra et al. / Materials Letters 160 (2015) 175–178

[13] M.F. Hassan, Z.P. Guo, Z. Chen, H.K. Liu, J. Power Sources 195 (2010) 2372–2376. [14] B.C. Sin, S.U. Lee, B. Jin, H. Kim, J.S. Kim, S. Lee, J. Noh, Y. Lee, Solid State Ion. 260 (2014) 2–7. [15] S. Zhong, L. Liu, J. Liu, J. Wang, J. Yang, Solid State Commun. 149 (2009) 1679–1683. [16] P. Yue, Z. Wang, H. Guo, X. Xiong, X. Li, Electrochim. Acta 92 (2013) 1–8. [17] E.I. Melnichenko, N.M. Laptash, T.A. Kaidalova, D.G. Epov, Zhurnal Neorg. Khimii 33 (1988) 906–910.

[18] N.M. Laptash, Yu.M. Nikolenko, N.M. Mishchenko, Russ. J. Inorg. Chem. 42 (1997) 681–684. [19] L. Noerochim, J.-Z. Wang, D. Wexler, Z. Chao, H.-K. Liu, J. Power Sources 228 (2013) 198–205. [20] J.S. Chen, Y.L. Cheah, S. Madhavi, X.W. Lou, J. Phys. Chem. C 114 (2010) 8675–8678. [21] U.K. Sen, S. Mitra, Energy Procedia 54 (2014) 740–747. [22] R. Nadimicherla, Y. Liu, K. Chen, W. Chen, Solid State Sci. 34 (2014) 43–48.