A New Anode Material LiVMoO 6 for Use in Rechargeable Li-Ion Batteries

Size: px
Start display at page:

Download "A New Anode Material LiVMoO 6 for Use in Rechargeable Li-Ion Batteries"

Transcription

1 Tamkang Journal of Science and Engineering, Vol. 5, No. 2, pp (2002) 107 A New Anode Material LiVMoO 6 for Use in Rechargeable Li-Ion Batteries Ru-Shi Liu 1, Chien-Yuan Wang 1, Ling-Yun Jang 2 and Jyh-Fu Lee 2 1 Department of Chemistry National Taiwan University Taipei, Taiwan 106, R.O.C. 2 Synchrotron Radiation Research Center Hsinchu, Taiwan 300, R.O.C. Abstract The lithiated transition metal oxide LiVMoO 6 has been synthesized by solid state reaction. This is the first report of this compound to be studied as an anode material for use in secondary batteries. The synthesized LiVMoO 6 powder has been studied by means of X-ray diffraction (XRD) and X-ray absorption near edge structure (XANES) spectroscopy. The electrochemical characteristics of the prepared electrodes assembled in coin cells were also investigated in terms of half-cell performance. It is observed that the cell exhibits three stages of discharge plateaus. The total discharge capacity, averaged over several test runs, is about 1250 mah/g. This value is much higher than the capacities exhibited by many kinds of anode materials. Key Words: Li-ion Batteries, Anode, X-ray Absorption, XANES, LiVMoO 6 1. Introduction Rechargeable batteries have been considered an attractive power source for a wide variety of applications and in particular, lithium-ion batteries are emerging as the technology of choice for portable electronics. One of the main challenges in the design of these batteries is to ensure that the electrodes maintain their integrity over many discharge and recharge cycles. Michael et al. have reported that LiVMoO 6 can be synthesized by a soft-combustion (wet chemical) method [1]. However, the XRD pattern of LiVMoO 6 reported by them was not indexed. On the other hand, Gopalakrishnan et al. have shown that LiVMoO 6 as well as its reduced product, LiVMoO 5, obtained by the solid-state reaction method can be identified by their XRD patterns [2]. Moreover, the authors performed many measurements on the physical properties of these materials. In the present work, we focus on finding a simple method for preparing LiVMoO 6 and on understanding the factors that influence its electrochemical properties. 2. Experimental The LiVMoO 6 sample was synthesized by solid state reaction of Li 2 CO 3, V 2 O 5 and MoO 3. Well ground mixtures of the starting materials were sintered at 550 C in air for 24 h. X-ray diffraction (XRD) analyses were carried out with a SCINTAG (X1) diffractometer (Cu K α radiation) at 40 kev and 30 ma. Data for the Rietveld refinement were collected in the 2θ range with a step size of 2 and a count time of 10 s per step. The program GSAS was used for the Rietveld refinement in order to obtain information about the crystal structure of LiVMoO 6 [3]. The valences of V and Mo were determined by the X-ray absorption technique. The experiments for V and Mo were carried out by using synchrotron radiation with the electron beam

2 108 Ru-Shi Liu et al. energy of 1.5 GeV at the Synchrotron Radiation Research Center (SRRC) in Taiwan and the 8 GeV Spring-8 facility in Japan, respectively. The spectra were recorded by measuring the ratio I/I 0, where I 0 is the intensity of the incident beam. According to the attenuation law I = I 0 exp ( µx), ln (I/I 0 ) is proportional to the absorption function µx where µ is the absorption cross-section of the element of interest and x is the thickness of the sample. Furthermore, to avoid the sample thickness effect, the condition µx 1 must be satisfied, where µx is the edge step. Therefore, the thickness of the samples was manipulated by folding the sample-coated Scotch tape to achieve µx = 1. The incident photon flux (I 0 ) was monitored simultaneously by an ion-chamber which was positioned after the exit slit of the monochromator. The intensity of the transmitted X-ray monitored in the same way was considered as I 0 of the standard metal foil for calibrating the energy of the beam. All the measurements were performed at room temperature. The photon energies were calibrated to an accuracy of 0.1 ev via the theoretical values of the V and Mo metal K-edge absorption energies. Electrochemical characterization was performed using the coin-type cells. The method of assembling the cell was as follows. The sheets were prepared by spreading a slurry mixture of 85 wt.% LiVMoO 6, 9 wt.% carbon black and 6 wt.% polyvinylidene fluoride (PVDF) dissolved in 1-methyl-2-pyrolidinone (NMP) on an aluminum or copper foil. The prepared sheets were then placed into a vacuum oven to evaporate the solvent at 90~100 C for 12 h. The electrode disks (1/2 1.0 inch o.d.) were punched from the sheets, with an average weight of 3 mg of active material. The cell consisted of an electrode disk and a lithium metal foil with a porous polyethylene film as a separator. The electrolyte used was 1 M LiPF 6 in a mixture of ethylene carbonate (EC) and dimethyl carbonate (DMC) in the volume ratio of 1:1. The cell was assembled in an argon-filled dry box and tested at room temperature. The charge and discharge experiments were performed with a Maccor battery cycling instrument. The cells were cycled at current density values of 0.3 ma/cm 2 in the voltage range V or V. 3. Results and Discussion The powder XRD pattern of LiVMoO 6 and the structural parameters calculated with Rietveld refinement are presented in Figure 1 and Table 1, respectively. The XRD pattern of LiVMoO 6 could be indexed on the basis of a monoclinic cell [a = (4) Å, b = (14) Å, c = (25) Å, α = γ = 90 and β = (10) ]. The space group of crystal structure is C12/m1. The sample could be synthesized successfully in single phase, as evidenced by the XRD refinement results. The layered structure of LiVMoO 6 is shown in Figure 2. LiVMoO 6 crystallizes in the brannerite (ThTi 2 O 6 ) structure which consists of edge- and corner-sharing MO 6 (M = V or Mo) octahedra. The negative charges on the M 2 O 6 sheets are compensated by additional cations (Li) which reside in the interlayer space [2]. 0.8 Counts ( ) θ Figure 1. Observed (cross), calculated (solid line) X-ray powder diffraction pattern of LiVMoO 6. Small bars indicate the positions of Bragg reflections for LiVMoO 6. The difference between the calculated and experimental patterns is plotted along the bottom.

3 A New Anode Material LiVMoO 6 for Use in Rechargeable Li-Ion Batteries 109 Table 1. Refined fractional atomic positions, unit cell parameters and reliability factors (%) of LiVMoO 6 having space group C12/m1 at room temperature atoms x y z fraction Uiso (Å 2 ) Li (17) V (15) (21) 252(6) Mo (15) (21) 252(6) O (7) (8) 1 385(26) O (7) (9) 1 420(26) O (7) (9) 1 524(30) space group: C12/m1 reliability factor bond distances (Å) lattice parameters: a = (4) Å b = (1) Å c = (3) Å R p = 7.25% R wp = 10.69% χ 2 = 2.11 V(Mo)-O Figure 2. Ideal crystal structure of LiVMoO 6 with monoclinic cell (space group: C12/m1). Unit cell is shown in the center. Part of each V(Mo)O 6 octahedra is shaded. The X-ray absorption near edge structure (XANES) of LiVMoO 6 and the standard samples at the V and Mo K-edges are shown in Figures 3 and 4, respectively. The differences between the energy values corresponding to half height of normalized absorption ( µx) can usually be used to compare the oxidation states of the metal cations. Hence, according to the chemical shift of LiVMoO 6 spectra, the oxidation number of the vanadium ion is about +4.5 (as shown in Figure 3) while that of molybdenum ion is +6 (as shown in Figure 4). Normalized absorption 1.0 V 2 O 3 VO 2 V 6 O 13 V 2 O 5 LiVMoO Energy (ev) Figure 3. Normalized V K-edge XANES spectra of LiVMoO 6 and those of the standard samples Normalized absorption 1.0 Mo MoO 2 MoO 3 LiVMoO Energy (ev) Figure 4. Normalized Mo K-edge XANES spectra of LiVMoO 6 and those of the standard samples

4 110 Ru-Shi Liu et al. Moreover, the pre-edge feature can be assigned to the forbidden transition 1s to 3d, the lower-energy shoulder to the 1s to 4p shakedown transition and the strong peak to the dipole-allowed transition 1s to 4p [4,5]. The initial 1s state being a gerade state, the 1s to 3d transition is dipole forbidden in the regular octahedral VO 6 units with a center of inversion. When the symmetry of the VO 6 is lowered to distorted octahedral (as in V 2 O 3 ) or distorted square-pyramidal (as in V 2 O 5 ), the inversion center is broken. The pre-edge absorption becomes dipole allowed due to a combination of stronger 3d-4p mixing and overlap of the vanadium 3d orbital with the 2p orbital of oxygen. The intense dipole-allowed absorption occurring in the pre-edge region of the vanadium K-edge in LiVMoO 6 is shown in Figure 3. This is consistent with the result of the XRD refinement (Table 1.) in the manner of bond lengths between metal and oxygen. The charge, discharge and cycling plotted in the form of potential vs. capacity are shown in Figure 5. It is observed that although the ratio of the discharge capacity to charge in each cycle remains almost 95% for 30 cycles, the absolute values of the capacities were very low, 0.12 mah/g. + /Li) Charge Potential (V vs. Li Potential (V + /Li) Discharge cycle cycle number increasing Capacity (mah/g) Figure 5. Typical charge and discharge curves of LiVMoO 6 at a voltage of V Based on the chemical formula of LiVMoO 6 that is used for a cathode material, the theoretical capacity is calculated to be 107 mah/g, while the practical utilizable capacity was 92 mah/g for the first cycle, which was reported by Michael et al. [1]. However, based on our experimental results presented above, we feel that LiVMoO 6 is not suitable as a cathode material in rechargeable Li-ion batteries. The reasons are as follows. First, since de-intercalation of positive lithium ions from the crystal structure must be electrically compensated by oxidation of V 4.5+ or Mo 6+, this suggests that even for layered structure phase, only the amount of V 4.5+ contributes to the charge-discharge capacity. As a result, the initial capacity of LiVMoO 6 is limited by the intrinsic amount of V 4.5+ in the material. On the other hand, the expected capacity owing to V ions that can be oxidized seem to be much less than the experimental result of Michael et al. [1]. Second,

5 A New Anode Material LiVMoO 6 for Use in Rechargeable Li-Ion Batteries 111 as we know, the LiVMoO 6 phase has a layered structure. When it is used for the cathode material in the Li-ion batteries and the charge current is passed, the Li-ions are extracted during this period. But it is reasonable to expect that the layered structure needs a sufficient amount of Li ions in the lattice to maintain the structure or else, the structure would break down. Furthermore, it might be an irreversible process. Therefore, coin type cells were also assembled and cycled between 3.00 and 1 V. When the cell was discharged, the potential rapidly drops to reach a plateau, and then continuously decreases down to 1 V as shown in Figure 6. The amplitudes of three plateaus are about V, 0.6- V and V, respectively. The total discharge capacity, averaged over several test runs, is about 1250 mah/g which is much higher than the capacities, about mah/g, exhibited by many kinds of anode materials. that the well crystallized LiVMoO 6 powders were decomposed to nano-scale particles or even to amorphous phase, and the organic electrolyte molecules were also deposited on the surface of LiVMoO 6 by some electrocatalytic reactions. Although the LiVMoO 6 material cannot maintain its crystallinity after discharge, the new phase, i.e. the products of decomposed LiVMoO 6, still has significant reversibility of cycling. The assumption, however, that the nano-sized materials can exhibit quite different properties from bulk materials is now widely accepted [6]. Therefore, on this basis, the electrochemically driven size confinement of the decomposed particles is thus believed to enhance their electrochemical activity towards lithium metal. (a) 3.0 /Li) 2.5 Potential (V vs. Li Capacity (m A h/g) (b) Figure 6. Typical discharge curves of LiVMoO 6 at a voltage of V The differences of surface morphology of the anode electrodes containing LiVMoO 6 between as-prepared and after cycling were examined by scanning electron microscope (SEM; Philips XL30) are shown in Figure 7. The particle size distribution of LiVMoO 6 before discharge is broadened; the estimated grain size varies from 1 to 10 µm. As seen from the micrograph in Figure 7(a), almost all the particles have clear grain boundaries after sintering at 550 C. Therefore, it is plausible to suppose that the synthesized powders are very suitable for use as an electrode-active material in rechargeable lithium ion batteries. It is interesting to note that the clear grain boundaries were not exhibited in the sample after discharge, as shown in Figure 7(b), and some translucent materials were observed on the surface of the powders. It seems reasonable to conclude Figure 7. SEM micrographs of the LiVMoO 6 electrode (a) before and (b) after discharge 4. Conclusion LiVMoO 6 was successfully synthesized using the conventional solid-state reaction method. We have shown that LiVMoO 6 does not possess good structural characteristics for a lithium half cell (Li/ LiVMoO 6 ) as a cathode in non-aqueous electrolyte environment. Furthermore, we suggest that LiVMoO 6 may instead be considered as an anode

6 112 Ru-Shi Liu et al. material of choice for developing rechargeable lithium-ion battery technology. Acknowledgment This work was supported by the Synergy Corporation and the National Science Council of R.O.C. under the grant number NSC M References [1] Michael, M. S.; Fauzi, A.; Prabaharan, S. R. S. Int. J. Inorg. Mater. 2000, 2, 261. [2] Gopalakrishnan, J.; Bhuvanesh, N. S. P.; Vijayaraghavan, R.; Vasanthacharya, J. Mater. Chem. 1997, 7, 307. [3] Larson, A. C.; von Dreele, R. B. Generalized Structure Analysis System; Los Alamos National Laboratory, Los Alamos, New Mexico, U.S.A., [4] Bair, R. A.; Goddard III, W. A. Phys. Rev. B 1980, 22, [5] Wang, J.; Lytle, F. W.; Messmer, R. P.; Maylotte, D. H. Phys. Rev. B 1984, 30, [6] Poizot, P.; Laruelle, S.; Grugeon, S.; Dupont, L.; Tarascon, J.-M. Nature 2000, 407, 496. Manuscript Received: Apr. 15, 2002 and Accepted: May 20, 2002