Li Ion Diffusivity and Improved Electrochemical Performances of the Carbon Coated LiFePO 4

Transcription

1 836 Bull. Korean Chem. Soc. 2011, Vol. 32, No. 3 ChangKyoo Park et al. DOI /bkcs Li Ion Diffusivity and Improved Electrochemical Performances of the Carbon Coated LiFePO ChangKyoo Park,, Sung Bin Park, Si Hyung Oh, Ho Jang, and Won Il Cho,* Advanced Battery Center, Korea Institute of Science and Technology 39-1, Seoul , Korea * wonic@kist.re.kr Department of Materials Science and Engineering, Korea University, Seoul , Korea Received October 8, 2010, Accepted December 29, 2010 This study examines the effects of a carbon coating on the electrochemical performances of LiFePO. The results show that the capacity of bare LiFePO decreased sharply, whereas the LiFePO /C shows a well maintained initial capacity. The Li ion diffusivity of the bare and carbon coated LiFePO is calculated using cyclic voltammetry (CV) to determine the correlation between the electrochemical performance of LiFePO and Li diffusion. The diffusion constants for LiFePO and LiFePO /C measured from CV are and cm 2 s 1, respectively, indicating considerable increases in diffusivity after modifications. The Li ion diffusivity (D Li ) values as a function of the lithium content in the cathode are estimated by electrochemical impedance spectroscopy (EIS). The effects of the carbon coating as well as the mechanisms for the improved electrochemical performances after modification are discussed based on the diffusivity data. Key Words : Lithium iron phosphate, Carbon coating, Chemical diffusion coefficient, Cyclic voltammetry, Electrochemical impedance spectroscopy Introduction Olivine-type lithium iron phosphate (LiFePO ) has attracted great attention as a cathode material in lithium-ion batteries due to high theoretical capacity (170 mah g 1 ), low manufacturing cost and good thermal stability. 1,2 However, its low electronic conductivity and low lithium ion diffusivity limited its wide commercialization. 3-6 To overcome the drawbacks of LiFePO, carbon coating on the LiFePO surface 7-10 and metal elements doping, such as Cr, Mg, Ni, and Nb, were performed to improve the electrical conductivity In particular, lithium diffusion within the particle is a key factor that determining the electrochemical performances of LiFePO because it influences the phase transformation between triphylite and heterosite during charge-discharge cycling. Therefore, the investigation of the lithium ion chemical diffusion coefficient, D Li, was critical; Franger et al. 15 reported the D Li values of LiFePO /C in a range of to 10 1 cm 2 s 1 by using electrochemical impedance spectroscopy (EIS). Prosini et al. 16 measured D Li using a galvanostatic intermittent titration technique (GITT) in the range of to 10 1 cm 2 s 1. By using cyclic voltammetry (CV), Xie et al. 17 also reported the D Li values as cm 2 s 1. However, there are only a little knowledge is available on LiFePO that demonstrates the relation between Li ion diffusivity and modification such as carbon coating. Therefore, in this paper, technique of cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) were used to discover the role of carbon coating on the electrochemical performances in terms of lithium diffusion coefficient. Experimental A mixture of Li 2 CO 3 (Aldrich, 99%), FeC 2 O 2H 2 O (Junsei, 99%), and (NH )H 2 PO (Junsei, 99%) was placed in a zirconia bowl and the mechanochemical reaction was carried out for 3 h using a planetary mill (FRITSCH Pulverisette 5). The rotation speed was 250 rpm and the ballto-powder weight ratio was 20:1. The resulting powder mixture was heat treated at 600 o C for 10 hours under an Ar+5% H 2 atmosphere. For the carbon coating, 3 wt % of Ensaco 350G were added to the starting materials before mechanochemical activation. The LiFePO /C was heat-treated at 750 o C for 10 hours. The crystal structure of LiFePO was analyzed by X-ray diffraction (XRD; D/MAX-II A) using Cu K α radiation between 15-5 (2θ). The morphology of the LiFePO was examined by field emission scanning electron microscopy (FE-SEM, Hitachi, S-200, Japan). The cathode of the bare and carbon coated LiFePO (LiFePO /C) was composed of the active material, acetylene black, and polyvinylidene fluoride (PVDF) at a weight ratio of 85:10:5 and coated onto an Al foil. The cathode was held in a vacuum oven at 80 o C for 12 h. After drying, the cathode was 60 μm thick and contained approximately 5-7 mg cm 2 of the active materials. The electrolyte was 1 M LiPF 6 in an ethylene carbonate/dimethyl carbonate/ethylmethyl carbonate (EC/DMC/EMC) solution, and lithium foil was used as the counter electrode. A standard coin cell (2032 type) was used to examine the charging and discharging activities of the cathode. The Maccor 000 battery cycler with cut-off voltages of V was used to analyze the electrochemical performances. Cyclic voltammetry (CV) was carried out between V at various scan rates from 0.01 to 0.5 mv

2 Bull. Korean Chem. Soc. 2011, Vol. 32, No. 3 Li Ion Diffusivity and Electrochemical Performances 837 s 1. An impedance/gain-phase analyzer (Schlumberger model SI 1260) connected to a Schlumberger model SI 1286 electrochemical interface was used to assess the electrochemical impedance of the cell. The amplitude of the AC signal was 5 mv over the frequency range between 100 khz and 10 mhz. Results and Discussion Crystal Structure and Surface Morphology. The structure and size distribution of the bare and modified LiFePO particles were analyzed by XRD (Fig. 1). The diffraction peaks indicated a typical single phase olivine structure after heat-treatment. This suggests that the olivine structure was well maintained after the carbon coating. The (101), (111)/ (201), (211)/(020), and (311) peaks of bare LiFePO and LiFePO/C were selected and average particle sizes of the samples were calculated using the Scherrer s formular (D = 0.9λ/βcosθ, where λ = 15.2 nm (Cu Kα) and β = full width half maximum at the diffraction angle of θ). They were 1.25 nm and nm for LiFePO and LiFePO/C, respectively. The particle size was controlled by heat treating them at different temperatures. To produce similar size particle, the bare sample was heat-treated at 600 oc, while modified sample was heat treated at 750 oc. This is because a large particle can deteriorate electrochemical performance of the cathode rapidly18,19 due to the extended Li diffusion path.20,21 Figure 2 shows SEM images of the bare LiFePO and LiFePO/C samples. The bare LiFePO powder showed clean particle surfaces, whereas the LiFePO/C exhibited small adherents to the surface, which appeared to be carbon. This carbon coating layer suppressed particle growth during heattreatment. In addition, the particle size of the samples identified by SEM is consistent with the data from XRD peak analysis. Electrochemical Properties. The cyclability of the bare LiFePO and LiFePO/C was obtained at 0.5 C rate, as shown in Figure 3(a). The carbon coating had strong effects Figure 1. XRD profiles of LiFePO and LiFePO /C. LiFePO /C was heat-treated at 750 C for 10 h. LiFePO was heat-treated at 600 C for 10 h. o o Figure 2. SEM micrographs of (a) the bare LiFePO and (b) LiFePO /C. Figure 3. (a) Cyclabilities and (b) rate performances of LiFePO and LiFePO /C during discharging.

3 838 Bull. Korean Chem. Soc. 2011, Vol. 32, No. 3 ChangKyoo Park et al. on the electrochemical performance of LiFePO. The initial capacity of the bare LiFePO delivered 59.9 mah g 1 at the first cycle, and it showed severe capacity reduction in the subsequent cycles. Eventually, it delivered only 33.6 mah g 1 after 50 cycles. On the other hand, LiFePO /C delivered 1.7 mah g 1 at the first cycle, and it exhibited excellent capacity retention. After 50 cycles, LiFePO /C maintained its initial capacity and delivered 13.6 mah g 1. The rate performance of the bare and modified LiFePO was obtained at discharge rates ranging from 0.1 C to 20 C, as shown in Figure 3(b). The initial capacity of the bare LiFePO delivered 8.5 mah g 1 at 0.1 C, and despite its considerably low C rate, it showed severe capacity reduction in the subsequent four cycles. Moreover, as the discharge current density was increased, significant decreases in capacity were observed at high C rates, eventually showing almost no capacity at 20 C. On the other hand, LiFePO /C delivered mah g 1 at the first cycle at 0.1 C and maintained a capacity of 89. mah g 1 at the first cycle at 20 C with excellent capacity retention. These results indicate that the carbon coating improves the electrochemical performances significantly compared to the bare LiFePO. Similar results have been reported and attributed to the improvement in electrical conductivity These results demonstrate the positive role of carbon coating which prevents the fading of LiFePO. The CV profiles from the bare LiFePO and LiFePO /C were performed at a scan rate of 0.05 mv s 1 during the redox reaction, as shown in Figure (a). LiFePO showed one distinct anodic peak (charge) and cathodic peak (discharge) which is comprised by a two phase system. The figure demonstrates that the intensity and shape of the peak current depend strongly on carbon coating. The LiFePO /C cathode resulted in increase peak current than the bare LiFePO, which indicated that Li ions and electrons considerably contributed to redox reactions as a result of the carbon coating on LiFePO. In addition, CV profile shifted toward open-circuit voltage (OCV) after carbon coating, which suggested that carbon coating significantly decreased the nonequilibrium condition of the cathode. LiFePO cathodes were carried out cyclic voltammetry at various scan rates in the range of mv s 1 (Fig. (b)) to obtain Li ion diffusivity. The peak separation and intensity increased with increasing scan rate. Randles-Sevcik equation (Eq. 1), which describes the relationship between the peak current (i p ) and square root of the scan rate, 22 can be applied to determine the chemical diffusion coefficient of the Li ion in LiFePO. I p /m = 0.63F(F/RT) 1/2 C Li V 1/2 AD 1/2 (1) Figure. CV profiles of (a) the bare LiFePO and LiFePO /C at a scan rate of 0.05 mv s 1 and (b) the LiFePO /C with various scan rates. Figure 5. Graphs of the (a) bare LiFePO and (b) LiFePO /C with normalized peak current vs square root of the scan rate.

4 Li Ion Diffusivity and Electrochemical Performances Bull. Korean Chem. Soc. 2011, Vol. 32, No Table 1. Chemical diffusion coefficients of LiFePO and LiFePO / C obtained from CV Cathode materials LiFePO LiFePO /C Anodic D (cm 2 s 1 ) Cathodic D (cm 2 s 1 ) where i p is the peak current (A), m is the mass of electrode, F is the Faraday constant, C Li is the concentration of Li ion in the cathode, V is the scan rate, D is the diffusion constant, and A is the contact area between electrode and electrolyte. In particular, Li ions in LiFePO were intercalated and deintercalated along the [010] direction. 23,2 Therefore, the entire electrode area of parameter A in Eq. (1) was substituted for one-third of the total Brunauer-Emmett-Teller (BET) surface area. 25 A plot of normalized peak current with the square root of the scan rate (V 1/2 ) of LiFePO and LiFePO /C is shown Figure 5 to calculate the diffusion constants in the electrodes, as listed in Table 1. The diffusion constant at the cathodic peak increased approximately 3.8 times after coating the bare LiFePO with carbon. The value of Li ion diffusivity are well consistent with those measured by Denis et al. 25 using CV and Srinivasan and Newman 26 with porous electrode theory. To understand carbon coating effect intensively, the EIS experiments performed on bare LiFePO and LiFePO /C cathodes at various lithium contents in cathode electrode. Before each EIS measurement, for the stable SEI film formation and the good permeation of electrolyte into the active material, several galvanostatical cycles were performed. Impedance was measured at various lithium contents in LiFePO particle until fully discharged. Figure 6 shows an equivalent circuit and Nyquist plots of the samples at Li content of 30% in LiFePO. R s represents the ohmic resistance between the electrolyte and electrode, which corresponds to the intercept at the Re(Z) in the high frequency region. In addition, the radius of the semicircle in a medium frequency range indicated the charge transfer Figure 6. The equivalent circuit and Nyquist plot of LiFePO electrodes for lithium content at 30%. Frequency range: 0.01 Hz- 100 khz. Figure 7. The relationship of imaginary resistance with inverse square root of angular speed for Li 0.3FePO and Li 0.3FePO /C. resistance (R ct ). The inclined line in the low frequency region corresponds to the Warburg impedance Z w, which is related to Li-ion diffusion within the LiFePO particle. The model proposed by Ho et al. 27 was used to obtain the diffusion coefficients of the bare LiFePO and LiFePO /C, as described in Eq. (2). D Li = 1/2 [(V M /SFA)(δ E/δ x)] 2 (2) where V M is the molar volume of the LiFePO, S is the surface area of the cathode, F is the Faraday constant, A is obtained from the Warburg impedance, and δ E/δ X is the slope of the coulometric titration curve. Figure 7 demonstrates the plot of the imaginary resistance with the inverse square root of angular speed in the low frequency range for the bare and modified LiFePO when Li content is 30%. The slopes of these plots were substituted in Eq. (2) to determine the diffusion coefficient of lithium at different x values in the LiFePO cathodes. Table 2 shows the diffusion coefficient of Li ions obtained from the EIS method. When the Li content in LiFePO was 30%, the diffusion coefficient of carbon coated Li 0.3 FePO /C was approximately 3.7 times higher than that of bare Li 0.3 FePO. The Li ion diffusivity is well consistent with the diffusion data obtained from Franger et al. 15 and Zhu et al.. 28

5 80 Bull. Korean Chem. Soc. 2011, Vol. 32, No. 3 ChangKyoo Park et al. Table 2. Chemical diffusion coefficients of Li in Li xfepo obtained from EIS at various Li contents X in Li xfepo (cm 2 s 1 ) LiFePO LiFePO /C Figure 8. Schematic diagram of lithium diffusion in the electrolyte and particle. The results from CV and EIS suggested that LiFePO /C exhibited higher Li ion diffusivity than that of LiFePO. The significant difference in chemical diffusion coefficient indicates that the modification by the carbon coating influenced D Li in the particle. The effect of the carbon coating on the electrochemical performances of LiFePO was attributed to the low resistance to charge transfer at the particle surface (R ct ). As shown in Figure 6, the R ct value of the bare LiFePO was approximately 2 ohm, whereas that of LiFePO / C was 9 ohm. This indicates that the R ct value of LiFePO /C was approximately one third that of bare LiFePO due to the electrically conducting carbon on the surface. Therefore, the improved electrochemical performances of the LiFePO /C sample were attributed to the high electrical conductivity on the particle surface, which leads to faster Li ion diffusion from the electrolyte (route B in Fig. 8). Conclusions The chemical diffusion coefficient of bare LiFePO and LiFePO /C were studied with CV and EIS techniques for investigating improvement in electrochemical performances after carbon coating. CV analysis revealed an approximately 3.8 fold increase in diffusion constant when the bare LiFePO was coated with carbon. EIS technique gave more detailed results of chemical diffusion coefficients that LiFePO / C showed much faster Li ion diffusion in the particle than bare LiFePO. The results from CV and EIS suggested that the carbon coating reduced the electrical resistance (R ct ) at the particle surface during the charge-discharge cycles which led to enhance Li ion diffusion and electrochemical performances. Acknowledgments. This work was supported by Energy Resource R&D Programs ( B and 2008EEL11P ) under the Ministry of Knowledge Economy, Republic of Korea. References 1. MacNeil, D. D.; Lu, Z.; Chen, Z.; Dahn, J. R. J. Power Sources 2002, 108, Takahashi, M.; Tobishima, S. I.; Takei, K.; Sakurai, Y. Solid State Ionics 2002, 18, Padhi, A. K.; Nanjundaswamy, K. S.; Masquelier, C.; Okada, S.; Goodenough, J. B. J. Electrochem. Soc. 1997, 1, Takahashi, M.; Tobishima, S.; Takei, K.; Sakurai, Y. J. Power Sources 2001, 97/98, Barker, J.; Saidi, M. Y.; Swoyer, J. L. Electrochem. Solid-State Lett. 2003, 6, A Andersson, A. S.; Kalska, B.; Haggstrom, L. Thomas, J. O. Solid State Ionics 2000, 130, Ravet, N.; Chouinard, Y.; Magnan, J. F.; Besner, S.; Gauthier, M.; Armand, M. J. Power Sources 2001, 97-98, Prosini, P. P.; Zane, D.; Pasquali, M. Electrochim. Acta 2001, 6, Huang, H.; Yin, S. C.; Nazar, L. F. Electrochem, Solid State Lett. 2001,, A Chen, Z.; Dahn, J. R. J. Electrochem. Soc. 2002, 19, A Chung, S. Y.; Bloking, J. T.; Chiang, Y. M. Nat. Mater. 2002, 1, Shi, S.; Liu, L.; Ouyang, C.; Wang, D. S.; Wang, Z.; Chen, L.; Huang, X. Phys. Rev. B 2003, 68, Wang, D.; Li, H.; Shi, S.; Huang, X.; Chen, L. Electrochem. Acta 2005, 50, Wang, G. X.; Bewlay, S.; Needham, S. A.; Liu, H. K.; Liu, H. K.; Drozd, V. A.; Lee, J. F.; Chen, J. M. J. Electrochem. Soc. 2006, 153, A Franger, S.; Le Cras, F.; Bourbon C.; Rouault, H. Electrochem. Solid-State Lett. 2002, 5, A Prosini, P. P.; Lisi, M.; Zane, D.; Pasquali, M. Solid State Ionics 2002, 18, Xie, J.; Imanishi, N.; Zhang, T.; Hirano, A.; Takeda, Y.; Yamamoto, O. Electrochim. Acta 2009, 5, Shin, H. C.; Cho, W. I.; Jang, H. J. Power Sources 2006, 159, Shin, H. C.; Cho, W. I.; Jang, H. Electrochim. Acta 2006, 52, Guo, Z. P.; Liu, H.; Bewlay, S.; Liu, H. K.; Dou, S. X. J. New Mater. Electrochem. Syst. 2003, 6, Lee, J.; Teja, A. S. Mater. Lett. 2006, 60, Bard, A. J.; Faulkner, L. R. Electrochemical Methods; John Wiley & Sons Inc.: New York, Morgan, D.; Van der Ven, A.; Cedar, G. Electrochem. Solid-State Lett. 200, 7, A Ouyang, C.; Shi, S.; Wang, Z.; Huang, X.; Chen, L. Phys. Rev. B 200, 69, Yu, D. Y. W.; Fietzek, C.; Weydanz, W.; Donoue, K.; Inoue, T.; Kurokawa, H.; Fujitani, S. J. Electrochem. Soc. 2007, 15, A Srinivasan, V.; Newman, J. J. Electrochem. Soc. 200, 151, A Ho, C.; Raistrick, I. D.; Huggins, R. A. J. Electrochem. Soc. 1980, 127, Zhu, Y.; Wang, C. J. Phys. Chem. C 2010, 11, 2830.