Enhanced electrochemical performances of multi-walled carbon nanotubes modified Li3V2(PO4)3/C cathode material for lithium-ion batteries
yu qing qiao,J.P. Tu,Y.J. Mai,L.J. Cheng,X.L. Wang,C.D. Gu
DOI: https://doi.org/10.1016/j.jallcom.2011.04.048
IF: 6.2
2011-01-01
Journal of Alloys and Compounds
Abstract:Highlights ► MWCNTs were employed as a highly conductive agent in Li 3 V 2 (PO 4 ) 3 /C. The MWCNTs and the residual C will link the Li 3 V 2 (PO 4 ) 3 particles effectively. The Li 3 V 2 (PO 4 ) 3 /(C + MWCNTs) showed a discharge capacity of 122.9 mAh g −1 at 10 C. Abstract The multi-walled carbon nanotubes (MWCNTs) modified Li 3 V 2 (PO 4 ) 3 /C composite is synthesized by polyvinyl alcohol (PVA) based carbon-thermal reduction method using MWCNTs as a highly conductive agent. PVA mainly supplies a reductive atmosphere to reduce V 5+ and provides a network of carbon to inhibit the aggregation of Li 3 V 2 (PO 4 ) 3 particles. The amorphous carbon coating and MWCNTs co-modified composite shows excellent high-rate lithium intercalation/deintercalation property and cycling performance between 3.0 and 4.3 V. The discharge capacities of 131.7 and 122.9 mAh g −1 are obtained at rates of 1 C and 10 C, respectively, for the Li 3 V 2 (PO 4 ) 3 /(C + MWCNTs). These improvements are attributed to the valid conducting networks of C + MWCNTs and the reduced Li 3 V 2 (PO 4 ) 3 particle size by the network carbon from the pyrolysis of PVA. Keywords Lithium vanadium phosphate Multi-walled carbon nanotubes Carbon coating Lithium-ion battery 1 Introduction In recent years, lithium ion batteries are considered to have great potential application for electric vehicles (EVs) and hybrid electric vehicles (HEVs) [1–8] . The performance of these batteries depends critically on the properties of their cathode materials. Among cathode materials, monocline Li 3 V 2 (PO 4 ) 3 is considered to be one of the most promising materials because of its good ion mobility, relatively high operate voltage and large theoretical capacity [9–12] . However, the main disadvantage of Li 3 V 2 (PO 4 ) 3 is its poor intrinsic conductivity, which presents a major obstacle to practical implementation. Many efforts have been made over the past few years to improve the low conductivity by alien metal doping [13–15] and coating electronically conductive materials [16–20] . It is well known that MWCNTs have many advantages, such as excellent electronic conductivity, small specific surface area and tubular shape [21–24] . It is suggested that the MWCNTs are good additives in the electrode because a more integrated conductive network can be formed between the active materials and the substrate. The batteries with CNT additives in the electrodes presented higher total capacity and better high-rate discharge performance [22,23] . Recently, MWCNTs have introduced as a conductive carbon additive in the preparation of LiFePO 4 /MWCNTs composites [24,25] and LiCoO 2 /MWCNTs [26] . Li et al. [24] applied MWCNTs as the conducting additive in LiFePO 4 and they found that MWCNTs addition was an effective way to increase rate capability and cycle efficiency. Jin et al. [25] reported that the added MWCNTs in LiFePO 4 not only increase the electronic conductivity and lithium-ion diffusion coefficient, but also decrease crystallite size and charge-transfer resistance. In the present work, polyvinyl alcohol (PVA) based carbon-thermal reduction route by firstly using MWCNTs as a highly conductive agent was adopted to prepare MWCNTs-modified Li 3 V 2 (PO 4 ) 3 /C. Herein, PVA is mainly supplying a reductive atmosphere to reduce V 5+ and providing a network of carbon to inhibit the aggregation of Li 3 V 2 (PO 4 ) 3 particles. Foremost, the MWCNTs and the network carbon from the pyrolysis of PVA will link the Li 3 V 2 (PO 4 ) 3 particles effectively, which expect the MWCNTs-modified Li 3 V 2 (PO 4 ) 3 /C composite to achieve a high-rate lithium intercalation/deintercalation property. 2 Experimental MWCNTs-modified Li 3 V 2 (PO 4 ) 3 /C composite was prepared by carbon-thermal reduction method, both PVA and MWCNTs were used as carbon source in this study (denoted as Li 3 V 2 (PO 4 ) 3 /(C + MWCNTs)). The details about the synthesis of MWCNTs were introduced elsewhere [27] . Stoichiometric of Li 2 CO 3 , NH 4 VO 3 , NH 4 H 2 PO 4 and appropriate amount of PVA + MWCNTs (PVA:MWCNTs = 1:1, wt.%) were mixed in ethanol and ball milled for 4 h. After ethanol was evaporated, the precursor was heated at 850 °C for 10 h in argon flowing. For comparison, Li 3 V 2 (PO 4 ) 3 /C composite without MWCNTs was also prepared using the same process. The amounts of residual carbon in the Li 3 V 2 (PO 4 ) 3 /(C + MWCNTs) and Li 3 V 2 (PO 4 ) 3 /C were 4.71 and 5.27 wt.%, respectively, measured by elemental analyzer (EA, Flash EA1112). Electrical conductivity measurements were carried on CON510, the powders were hot-pressed respectively in a graphite die for 1 h with T = 500 °C, P = 80 MPa in a vacuum uni-axial compressor. The structure and morphology of the as-prepared powders were characterized using X-ray diffraction (XRD, Philips PC-APD with Cu K radiation) and high-resolution transmission electron microscopy (TEM, Tecnai G2 F30 S-Twin). Electrochemical performances of the composites were characterized using CR2025 coin-type cell. A metallic lithium foil served as the anode electrode. The cathode consisted of 85 wt.% active material, 10 wt.% carbon conductivity (residual carbon and acetylene black) and 5 wt.% polyvinylidene fluoride (PVDF) on aluminum foil. 1 M LiPF 6 in ethylene carbonate (EC)-dimethyl carbonate (DMC) (1:1 in volume) as the electrolyte, and a polypropylene micro-porous film (Cellgard 2300) as the separator. The cells were assembled in a glove box filled with high-purity argon. The charge–discharge tests were carried out on LAND battery test system (Wuhan, China) between 3.0 and 4.3 V by applying from 0.5 C to 10 C. Cyclic voltammetry (CV) test was carried out using the CHI660 C electrochemical workstation in a potential range of 3.0–4.3 V (vs. Li/Li + ) at a scan rate of 0.1 mV s −1 . For electrochemical impedance spectroscopy (EIS) measurements, the test cells were with the metallic lithium foil as both the reference and counter electrodes. EIS measurements were performed on the CHI660 C over a frequency range of 100 kHz to 10 mHz at a stage of charge (3.6 V vs. Li/Li + ) by applying an AC signal of 5 mV. 3 Results and discussion Fig. 1 shows the XRD patterns of Li 3 V 2 (PO 4 ) 3 /(C + MWCNTs) and Li 3 V 2 (PO 4 ) 3 /C composites. All the peaks can be assigned to monoclinic Li 3 V 2 (PO 4 ) 3 (space group P 2 1 / n (14), ICSD #96962) and no impurities can be observed. Additionally, the added MWCNTs and PVA do not change the crystal structure of Li 3 V 2 (PO 4 ) 3 which is similar to those in previous reports [3,9,19] . Otherwise, no additional diffraction peak related-carbon is detected in the XRD patterns, which is attributed to the amorphous state of the residual carbon, or the amount of carbon is too low to be detected. Fig. 2 a shows the TEM image of Li 3 V 2 (PO 4 ) 3 /C. It can be observed that the Li 3 V 2 (PO 4 ) 3 particles are coated with a uniform carbon layer about 8 nm in thickness. The 4.08 Å spacing corresponds to the (0 1 2) plane of monoclinic Li 3 V 2 (PO 4 ) 3 . Dissociative amorphous carbon can also be found between the particles. It is well known that the residual carbon from pyrolysis of organics or polymers is able to inhibit the aggregation of Li 3 V 2 (PO 4 ) 3 particles and provide good contact between the particles [16–18] . Electronic conductivity measurement illustrates that the electrical conductivity of the as-prepared Li 3 V 2 (PO 4 ) 3 /C composite is 4.95 × 10 −3 S cm −1 , which is much higher than that of the uncoated Li 3 V 2 (PO 4 ) 3 (2.3 × 10 −7 S cm −1 ) [28] . In order to distinguish between the MWCNTs and Li 3 V 2 (PO 4 ) 3 /C, TEM analysis were also conducted. As illustrated in Fig. 2 b and c, the Li 3 V 2 (PO 4 ) 3 /C particles are well modified by MWCNTs with the help of the PVA. On the assumption that PVA was not used in Figs. 2 b and c, only two or even three particles could be connected by MWCNTs. However, when the combination of PVA and MWCNTs was employed during the process, the network carbon from the pyrolysis of PVA is able to link the MWCNTs to the Li 3 V 2 (PO 4 ) 3 particles effectively, making more particles interlink with each other. On the other hand, the electrical conductivity of the as-prepared Li 3 V 2 (PO 4 ) 3 /(C + MWCNTs) composite is increased to 1.13 S cm −1 when MWCNTs was added to the mixture, which is two orders of magnitude higher than that of Li 3 V 2 (PO 4 ) 3 /C. Fig. 2 d shows an enlarged TEM image of the Li 3 V 2 (PO 4 ) 3 /(C + MWCNTs) composite. It is clearly that the Li 3 V 2 (PO 4 ) 3 particles are coated with carbon, and the dissociative amorphous carbon can link the MWCNTs to the Li 3 V 2 (PO 4 ) 3 particles effectively, forming a valid conducting network. Therefore, it can be speculated that this valid conducting networks will result in excellent electrochemical performance of Li 3 V 2 (PO 4 ) 3 /(C + MWCNTs) composite. Fig. 3 shows the charge–discharge curves of Li 3 V 2 (PO 4 ) 3 /(C + MWCNTs) and Li 3 V 2 (PO 4 ) 3 /C electrodes in the potential range of 3.0–4.3 V at 0.5 C rate. Both the cells exhibit three charge plateaus and correspondingly three discharge plateaus, which are identified as the two-phase transition processes during the electrochemical reactions [6,7] . Compared to the Li 3 V 2 (PO 4 ) 3 /C, the Li 3 V 2 (PO 4 ) 3 /(C + MWCNTs) electrode has longer charge/discharge curves and smaller potential differences of the plateaus, which manifests that the MWCNTs-modified Li 3 V 2 (PO 4 ) 3 /C composite has lower electrochemical polarization and leads to better reversibility in the charge–discharge processes. In the potential range of 3.0–4.3 V, two Li-ions in the Li 3 V 2 (PO 4 ) 3 can be removed, and its theoretical capacity is 133 mAh g −1 between 3.0 and 4.3 V. It can be found that the Li 3 V 2 (PO 4 ) 3 /(C + MWCNTs) electrode exhibits a high initial discharge capacity of 132.3 mAh g −1 (very close to its theoretical capacity), which is higher than that of Li 3 V 2 (PO 4 ) 3 /C (131.1 mAh g −1 ). What is more, the effect of the C + MWCNTs is even more significant from the rate performance at different current densities, as shown in Fig. 3 b. The discharge capacities of 131.7 and 122.9 mAh g −1 are obtained at rates of 1 C and 10 C, respectively, for the Li 3 V 2 (PO 4 ) 3 /(C + MWCNTs), whereas the values drop from 131.0 mAh g −1 (1 C) to 91.9 mAh g −1 (10 C) for the Li 3 V 2 (PO 4 ) 3 /C. It is also observed that about 92.9% of its capacity remains on going from 0.5 C to 10 C rate, for the Li 3 V 2 (PO 4 ) 3 /(C + MWCNTs) electrode, such an improved kinetics can be attributed to the fast charge transport between the electrolyte and the particles on account of the valid conducting networks of C + MWCNTs [23,24] . It has been reported that the carbon coating can retard formation of the SEI film in the surface of the particles and also decrease the charge-transfer impedance [16] . The C + MWCNTs can provide a good network for low-conductance Li 3 V 2 (PO 4 ) 3 , thus the electronic conductivity was improved significantly due to the valid conducting networks in Li 3 V 2 (PO 4 ) 3 , making an enhanced rate capability of Li 3 V 2 (PO 4 ) 3 /(C + MWCNTs). Fig. 4 shows the CV curves of Li 3 V 2 (PO 4 ) 3 /(C + MWCNTs) and Li 3 V 2 (PO 4 ) 3 /C. Noticeably, the anodic/cathodic peaks of MWCNTs modified Li 3 V 2 (PO 4 ) 3 /C are more symmetrical and sharper than that of Li 3 V 2 (PO 4 ) 3 /C composite, which indicates the better electrochemical reversibility. Furthermore, the potential separations of Li 3 V 2 (PO 4 ) 3 /(C + MWCNTs) between the anodic and cathodic peaks are less than that of Li 3 V 2 (PO 4 ) 3 /C, which also demonstrates the reversibility and reactivity of Li 3 V 2 (PO 4 ) 3 /(C + MWCNTs) are enhanced due to the presence of valid conducting networks of C + MWCNTs. Fig. 5 a shows the Nyquist plots of Li 3 V 2 (PO 4 ) 3 /(C + MWCNTs) and Li 3 V 2 (PO 4 ) 3 /C measured at the charge potential of 3.6 V after 5 cycles. In order to clearly illuminate the AC impedance spectra of the electrodes, an equivalent circuit has been built in Fig. 5 a. R e represents the solution resistance of the electrolyte; R sl and CPE sl designate the migration resistance of lithium ions and the capacity of surface layer, respectively; R ct and CPE ct stand for the charge-transfer resistance and double-layer capacitance, respectively; W represents the diffusion-controlled Warburg impedance [24,29–31] . It is found that the R e is similar for the two electrodes, and the R e values of Li 3 V 2 (PO 4 ) 3 /(C + MWCNTs) and Li 3 V 2 (PO 4 ) 3 /C are 4.52 Ω and 4.73 Ω, respectively, which are much lower than those of R sl and R ct . The migration resistance of lithium ions R sl of Li 3 V 2 (PO 4 ) 3 /(C + MWCNTs) and Li 3 V 2 (PO 4 ) 3 /C is 12.31 Ω and 19.38 Ω, respectively. The simulated charge-transfer resistance R ct of Li 3 V 2 (PO 4 ) 3 /(C + MWCNTs) and Li 3 V 2 (PO 4 ) 3 /C is 22.36 Ω and 41.12 Ω, respectively. It can be found that the Li 3 V 2 (PO 4 ) 3 /(C + MWCNTs) electrode exhibits a smaller charge-transfer resistance, which indicates the formation of conductive network by C + MWCNTs can increase the conductivity of active materials and reduce the impedance remarkably. The diffusion coefficient of Li + can be derived from the following Eq. (1) [32,33] or Eq. (2) [25,32] . (1) D L i = 1 2 V m F A σ w 2 d E d x 2 (2) D L i = R 2 T 2 2 A 2 n 4 F 4 σ w 2 C 2 where V m is the molar volume, F is Faraday constant, A is the active surface area, σ w is the Warburg factor, d E /d x is the slope of the open-circuit voltage vs. mobile Li + concentration x, D Li is the apparent diffusion coefficient, R is the gas constant, T is the absolute temperature, n is the number of electrons per molecule during oxidization, and C is the molar concentration of Li + . At the charge state of 3.6 V, the electrochemical reaction is in the two-phase region. It can easily find that the value of d E /d x is almost zero [32,33] , thus without any meaning for Eq. (1) . Therefore, we select Eq. (2) to calculate the apparent D Li values. The σ w was determined as the value of the slope of Z re vs. ω −1/2 for the Warburg region [25,32] . The relationship plots between Z re and ω −1/2 at low-frequency region are shown in Fig. 5 b. Thus, based on Eq. (2) , the apparent diffusion coefficient of Li-ions for Li 3 V 2 (PO 4 ) 3 /(C + MWCNTs) and Li 3 V 2 (PO 4 ) 3 /C is 6.2 × 10 −9 and 2.4 × 10 −9 , respectively. It is noticed that the diffusion coefficient of Li 3 V 2 (PO 4 ) 3 /(C + MWCNTs) is enhanced by MWCNTs adding, which can be attributed to its higher electronic conductivity [18,31,34] . 4 Conclusions MWCNTs-modified Li 3 V 2 (PO 4 ) 3 /C composite was successfully synthesized by PVA based carbon-thermal reduction method. Compared with the Li 3 V 2 (PO 4 ) 3 /C and Li 3 V 2 (PO 4 ) 3 /MWCNTs composites, the formation of conductive network by C + MWCNTs on Li 3 V 2 (PO 4 ) 3 particles has been greatly improved by using of PVA. This co-modified material exhibited improved cycling performance and good high-rate lithium intercalation/deintercalation properties. At a charge–discharge rate of 10 C, it can deliver a capacity of 122.9 mAh g −1 , much higher than that of Li 3 V 2 (PO 4 ) 3 /C. References [1] X.-L. Wu L.-Y. Jiang F.-F. Cao Y.-G. Guo L.-J. Wan Adv. Mater. 21 2009 2710 [2] M. Yang Q.M. Gao J. Alloys Compd. 509 2011 3690 [3] H. Huang T. Faulkner J. Barker M.Y. Saidi J. Power Sources 189 2009 748 [4] C.G. Son H.M. Yang G.W. Lee A.R. Cho V. Aravindan H.S. Kim W.S. Kim Y.S. Lee J. Alloys Compd. 509 2011 1279 [5] J. Zhong X.L. Wang X.H. Xia C.D. Gu J.Y. Xiang Y. Zhou J. Zhang J.P. Tu J. Alloys Compd. 509 2011 3889 [6] M.X. Gao Y. Lin Y.H. Yin Y.F. Liu H.G. Pan Electrochim. Acta 55 2010 8043 [7] X.Y. Wang S.Y. Yin K.L. Zhang Y.X. Zhang J. Alloys Compd. 486 2009 L5 [8] X.H. Huang J.P. Tu B. Zhang C.Q. Zhang Y. Li Y.F. Yuan H.M. Wu J. Power Sources 161 2006 541 [9] H. Huang S.-C. Yin T. Kerr N. Taylor L.F. Nazar Adv. Mater. 14 2002 1525 [10] B. Huang X.P. Fan X.D. Zheng M. Lu J. Alloys Compd. 509 2011 4765 [11] E. Kobayashi L.S. Plashnitsa T. Doi S. Okada J.-I. Yamaki Electrochem. Commun. 12 2010 894 [12] J. Zhai M.S. Zhao D.D. Wang Y.Q. Qiao J. Alloys Compd. 502 2010 401 [13] J.S. Huang L. Yang K.Y. Liu Y.F. Tang J. Power Sources 195 2010 5013 [14] S.-Q. Liu S.-C. Li K.-L. Huang B.-L. Gong G. Zhang J. Alloys Compd. 450 2008 499 [15] Q. Kuang Y.M. Zhao X.N. An J.M. Liu Y.Z. Dong L. Chen Electrochim. Acta 55 2010 1575 [16] Y.Z. Li Z. Zhou X.P. Gao J. Yan Electrochim. Acta 52 2007 4922 [17] Y.Q. Qiao X.L. Wang Y. Zhou J.Y. Xiang D. Zhang S.J. Shi J.P. Tu Electrochim. Acta 56 2010 510 [18] T. Jiang Y.J. Wei W.C. Pan Z. Li X. Ming G. Chen C.Z. Wang J. Alloys Compd. 488 2009 L26 [19] L. Zhang J.P. Tu J.Y. Xiang Y. Zhou X.L. Wang S.J. Shi J. Power Sources 195 2010 5057 [20] Y.Q. Qiao X.L. Wang J.Y. Xiang D. Zhang W.L. Liu J.P. Tu Electrochim. Acta 56 2011 2269 [21] D.J. Yang S.G. Wang Q. Zhang P.J. Sellin G. Chen Phys. Lett. A 329 2004 207 [22] J.B. Wu J.P. Tu Z. Yu X.B. Zhang J. Electrochem. Soc. 153 2006 A1847 [23] J.Y. Xiang J.P. Tu J. Zhang J. Zhong D. Zhang J.P. Cheng Electrochem. Commun. 12 2010 1103 [24] X.L. Li F.Y. Kang X.D. Bai W.C. Shen Electrochem. Commun. 9 2007 663 [25] B. Jin E.M. Jin K.-H. Park H.-B. Gu Electrochem. Commun. 10 2008 1537 [26] Q.-T. Zhang M.-Z. Qu H. Niu Z.-L. Yu New Carbon Mater. 22 2007 361 [27] J.P. Cheng X.B. Zhang Y. Ye J. Solid State Chem. 179 2006 91 [28] S.-C. Yin P.S. Strobel H. Grondey L.F. Nazar Chem. Mater. 16 2004 1456 [29] X.K. Zhi G.C. Liang L. Wang X.Q. Ou L.M. Gao X.F. Jie J. Alloys Compd. 503 2010 370 [30] T. Jiang C.Z. Wang G. Chen H. Chen Y.J. Wei X. Li Solid State Ionics 180 2009 708 [31] T. Jiang F. Du K.J. Zhang Y.J. Wei Z. Li C.D. Wang G. Chen Solid State Sci. 12 2010 1672 [32] T. Jiang W.C. Pan J. Wang X.F. Bie F. Du Y.J. Wei C.Z. Wang G. Chen Electrochim. Acta 55 2010 3864 [33] X.H. Rui J. Liu C. Li C.H. Chen Electrochim. Acta 55 2010 2384 [34] L.A. Montoro J.M. Rosolen Electrochim. Acta 49 2004 3243