Int. J. Electrochem. Sci.

, 6 (2011) 2022 - 2030 International Journal of

ELECTROCHEMICAL SCIENCE
www.electrochemsci.org

Synthesis of Nano-LiMnPO4 from MnPO4H2O Prepared by Mechanochemistry

Received: 8 April 2011 / Accepted: 15 May 2011 / Published: 1 June 2011

Synthesis of nano-LiMnPO4 from MnPO4H2O prepared by mechanochemistry is attempted. SEM, TEM, XRD and TG/DSC are performed to characterize the products. The results show that the primary particles are loosely aggregated with an average particle size of 40-50 nm. The morphology and particle size distribution of nano-LiMnPO4 preserves those of MnPO4H2O. This paves a promising way of high efficiency, easy accessibility and speediness to prepare nano-LiMnPO4.

Keywords: MnPO4, nano-cystalline, mechanochemistery, LiMnPO4, lithium-ion batteries

1. INTRODUCTION The synthesis of inorganic nanocrystals has attracted considerable interest due to dramatic improvements caused by nano-effect [1,2]. Olivine LiMnPO4 is now recognized as a promising cathode material for lithium ion batteries with potential vehicular application because of low cost, environmentally benign constituents and high theoretical capacity. Compared with the popular LiFePO4 cathode material [3-9], LiMnPO4 is of particular interest to the battery community because of the ideal location of the Mn2+/Mn3+ couple at 4.1 V vs. Li/Li+, which is 0.65V higher than LiFePO4, and its theoretical energy density (684Wh/kg = 171mAh/g 4.0V) is 1.2 times larger than that of LiFePO4 (578Wh/kg = 170mAh/g 3.4V). The main drawback of LiMnPO4 is their low electronic conductivity, which results in poor rate capability. Since Li et al.[10] first reported the reversible reaction of Mn(II)Mn(III) in olivine phosphate, researchers have tried various methods to prepare electro-active LiMnPO4 to overcome the limitation on its application. Encouraged by the successful studies on LiFePO4, developing specific synthesis methods appropriating for designing LiMnPO4 in the nano-scale region is in great need to

Int. J. Electrochem. Sci., Vol. 6, 2011

2023

improve the kinetics of LiMnPO4, as it has been proved that decreasing the size of the crystallites to nano-scale offers potential for higher electrode/electrolyte surface contact, less mechanical strain upon lithium (de)intercalation from the lattice than the bulk materials and reduces path length for lithiumion/electron transport through the material [11, 12]. Nano-LiMnPO4 is generally realized by three synthetic routes. Sol-gel method is a typical one, by which nano-LiMnPO4 was realized using precursor with atomic or molecular uniformity [13, 14]. Drezen et al. [15] explored sol-gel method to prepare nano-LiMnPO4. Their results showed that LiMnPO4 with 200nm and 270nm average particles size presented 79 mAh g1 at C/5, while LiMnPO4 with 140nm average particles size presented the reversible capacity up to 116 mAh g1 at C/5. Another synthetic route is direct precipitation of LiMnPO4 in liquid-phase [16-20], among which solvothermal and refluxing using high-boiling-point organic solution are more popular. These methods allow preparing LiMnPO4 with tiny but uniform particle size and controllable morphology. Wang et al. and Grazel [19, 21] presented platelet-like LiMnPO4 with the high specific capacity of 159 mAh g-1 at C/10 and 138 mAh g-1 at 1C. The excellent electrochemical performance was ascribed to the platelet morphology, which gave a very short distance of less than 30nm for Li+ diffusion. Solvothermal method was attracted to synthesis LiMnPO4/C nanocomposite with a uniform particle dimension which delivered an initial discharge capacity of 126.5 mAh g1 [22]. Solid-state reaction using nano-scale precursor is the most widely used synthetic route for industry, due to its advantages such as easy operation, easy adjustment of composition and being appropriate for scale production. Especially, solid-state reaction is generally necessary since it helpful to form effective surface layer on the materials. Rapid and effective method for preparing nanoprecursor is essential. In this study, a mechanochemical method was introduce to prepare uniform nano-MnPO4H2O. On that basis, nano-LiMnPO4 was prepared and evaluated.

2. EXPERIMENTAL 0.08mol 50% Mn(NO3)2 was dissolved in 80ml anhydrous ethanol. Then 20ml 70% H3PO4 was added into the solution and stirred. A suitable amount of urea was used to adjust the PH value. These materials were mixed at room temperature. The following ball milling procedure was carried out in a planetary-type ball mill (500 rpm). After 0.5-1 hour, the products were washed and dried. NanoLiMnPO4 was prepared by solid-state carbon thermal reduction. A stoichiometric amount of MnPO4H2O, LiCOOCH32H2O and 10% excess sucrose were mixed in the agate mortar. Then the mixture was calcined at 500C for 4 h in flowing ultra-pure Ar or N2 to obtain nano-LiMnPO4 material. The products were characterized by X-ray diffraction pattern (XRD, D/max 2550V, Rigaku, Japan) with Cu-K irradiation (=1.5406 nm). The morphology of the samples was inspected using scanning electron microscope (SEM, JSM-5600LV, JEOL, Japan) and transmission electron microscope (TEM, H-800, Hitachi, Japan). A combined differential scanning calorimetry/thermogravimetric analysis (DSC/TGA) instrument (Netzsch, STA 449C) was used to study the decomposition and reaction of the precursors. The powder sample of MnPO4 and the mixture of the starting materials for preparing LiMnPO4 were heated in an argon reducing environment to

Int. J. Electrochem. Sci., Vol. 6, 2011

2024

900C at a ramp rate of 5C/min. The electrochemical characterization was performed using CR2032 coin-type test cells. The test electrode was prepared from a mixture consisting of 80wt% LiMnPO4/C, 10wt% acetylene black, and 10 wt% PVDF. After rolling using rolling machine for several times, the mixture was cut into slices as test electrode. A Li-foil and a polypropylene film (Celgard 2400) were used as the counter electrode and separator, respectively. The electrolyte was 1mol L-1 LiPF6/EC+DMC+DEC (3:3:1 by volume). The prepared electrode pellets were dried at 120 C under vacuum for 48 h. ell assembly was conducted under an Ar atmosphere with a dew point below -70 C. The charge/discharge model was chosen constant current mode (CC mode). In this mode, the cells were chargedischarge cycled over a cell-voltage window of between 3V and 4.5V at a constant current (2.5 mA g-1) at room temperature.

3. RESULTS AND DISCUSSION

Figure 1. XRD pattern of as-prepared MnPO4H2O

Figure 1 shows the XRD pattern of as-prepared MnPO4H2O. All the diffraction peaks in the pattern are in agreement with those of MnPO4H2O (PDF card 00-044-0071). The relative intensity of diffraction peaks of is similar to the standard patterns, indicating that no preferentially growing directions for samples prepared by mechanochemical method. No characteristic peaks of impurity phases are present, indicating the high purity of the samples. It profited from the suitable pH condition. During the soft mechanochemical process, reactions take place as follows: Mn2+ + 4H+ + NO3- + PO43- MnPO4H2O+ NO + H2O (1)

Int. J. Electrochem. Sci., Vol. 6, 2011

2025

Generally, MnPO4H2O can be easily precipitated at room temperature, as long as high acidity can be satisfied since oxidization of Mn2+ to Mn3+ by NO3- can happen only in high acidity conditions. But high acidity also results in uncontrollable reaction rate, so causes too large particle size, random morphology and wide size distribution. So successful preparation of nano-MnPO4H2O crystalline with good uniformity was seldom reported. For example, Zhang et al. [23] prepared rod-like MnPO4H2O single crystals by hydrothermal method. Though the average diameter is about 68 nm, the rods are closely packed. Boonchom et al. [24] used a simple precipitation route at low temperature to synthesis nanocrystalline MnPO4H2O. Various attempts as they tried, the as prepared MnPO4H2O was nonuniform polyhedral grains and badly aggregated. To obtain uniform nano-MnPO4H2O particles, two design are introduced in our work. The first one is using ethanol as inert solvents to reduce both producing rate and growing rate. The second one is using a soft ball-milling to introduce local heating and large amounts of nucleate surface, so that to confine the formation of MnPO4H2O only on the collision surfaces. By tuning aqueous content and milling speed, particle size of nano-MnPO4H2O can be well controlled.

Figure 2. SEM photographs of as-prepared MnPO4H2O

From SEM photographs of the nano-MnPO4H2O, it can be observed that the sample consists of secondary particles which are random in shape. These secondary particles are formed by loose aggregation of primary particles. The TEM image shows that the primary particle size ranges from 40 to 50 nm. It is worth to be noted that the products present both uniform nano-size and pure MnPO4H2O crystalline phase. MnPO4H2O is a good candidate for Mn-contained precursor for LiMnPO4 preparation, since the locations of structural bounded water in MnPO4H2O is close to those of Li atoms in LiMnPO4. This similarity allows a transformation from MnPO4H2O into LiMnPO4 without major structural rearrangements. That is, the size distribution and morphology of LiMnPO4 can be easily controlled by designing those of MnPO4H2O. Xiao et al. [25] synthesized LiMnPO4/C composites with good performance using MnPO4H2O as the precursor which prepared by a precipitation process. As

Int. J. Electrochem. Sci., Vol. 6, 2011

2026

reported, MnPO4H2O will thermally decomposed into Mn2P2O7, the thermal behaviors of nanoMnPO4H2O is essential for preparing nano-LiMnPO4 from nano-MnPO4H2O. Figure 3 shows the TG/DSC results of as-prepared nano-MnPO4H2O and the precursor mixtures for LiMnPO4. It is worthy for confirming the annealing program for LiMnPO4 preparation, as well as getting information about water content in nano-MnPO4H2O precursor which influences the final stoichiometry of the final LiMnPO4 product [26]. According to The TG curve in Fig. 3A, there is only one weight loss of 15.45% by mass lies between 380 and 500C. Besides, the weight loss is accompanied by a change of color from dark greenish black to pale yellowish white. The DSC curve in Fig. 3A also shows one step exothermic process, which lies in the similar temperature range to that of weight loss. Above results both coincide with the description in literatures [24, 26, 27], indicating that thermal decomposition reaction as happened as follows: 4MnPO4H2O2Mn2P2O7 + O2 + 4H2O (2) Especially, the theoretical weight loss is 15.48% according to the reaction, including 4.76% from O2 and 10.72 from H2O. The weight loss calculated based on Fig. 3A is 15.45%, indicating that all the water in the as-prepared MnPO4H2O are existed as bounded water. Moreover, both TG and DSC curves show peak position around 473.3C, which is nearly 40C lower than bulk MnPO4H2O. This can be attributed to nano-effects, and from which it can be deduced that the annealing temperature can be reduced for nano-MnPO4H2O materials when compared with bulk MnPO4H2O materials.

Figure 3. TG/DSC curves of MnPO4H2O powders measured in argon

Int. J. Electrochem. Sci., Vol. 6, 2011

2027

Fig. 3B shows the TG/DSC curves of precursor mixtures. Compared with Fig. 3A, the curves changed greatly both in peak shapes and peak positions. According to the TG curve, there is only one step weight loss, which is 43.59% in mass and lies from 70C to 300C. Exactly, the theoretical weight loss is 43.36% according to LiMnPO4 formation reaction. Though the DSC curve is too complex to be understood at temperatures lower than 300C, it can be known from Fig. 3B that LiMnPO4 formed below 300C and annealing temperature about 500C may be appropriate to promote the crystalline growth of LiMnPO4.

Figure 4. XRD patterns of as-prepared LiMnPO4 powders

Figure 5. SEM photograph of as-prepared LiMnPO4 powders

Int. J. Electrochem. Sci., Vol. 6, 2011

2028

The XRD pattern of as-prepared LiMnPO4 was shown in Fig. 4. All the diffraction peaks in the patterns are in agreement with LiMnPO4 (PDF card #33-0803), which is olivine structure indexed in Pnma of an orthorhombic system. In the traditional solid state reaction process for preparing LiMnPO4, Li3PO4 impurities may easily form during the calcination process. [28] However, the presence of the MnPO4H2O precursor can significantly reduce the impurities produced by the solid state reaction [25]. Because the polyanion PO43 in the MnPO4H2O has formed a stable structure which PO43 tetrahedra and MnO6 octahedra share the corner oxygen ion, and cannot be react with lithium ion alone. The SEM photograph of as-prepared nano-LiMnPO4 powders is shown in Fig.5. A little different from expected, It can be found that the morphology of nano-LiMnPO4 was not as consistent as those of nano-MnPO4H2O precursor. Compared with nano-MnPO4H2O precursor, the average particle size of primary particles of nano-LiMnPO4 are increased to 40-100 nm. Furthermore, though the secondary particles of nano-LiMnPO4 are also formed by aggregation of the primary particles and random in size, while the dispersity of the primary particles obviously decreased. The difference in morphology and dispersity between nano-MnPO4H2O powder and nano-LiMnPO4 powder can be attributed to the nonuniformity of solid-state reaction among nano-MnPO4H2O particles, micro-size sucrose particles and micro-size LiCOOCH32H2O particles. It can be expected that LiMnPO4 powder can keep consistent in morphology and dispersity with nano-MnPO4H2O when all the reactants are in nanosize and well mixed. Further work in this aspect needs to be explored. Besides, the as-prepared nano-LiMnPO4 particles are neat and no amorphous carbon can be observed, indicating that sucrose is almost exhausted during thermo-reduction. Since bare LiMnPO4 can hardly present good electrochemical performances even in nano-size due to poor electronic conductivity, more sucrose should be used to form LiMnPO4/C composites in our future study.

Figure 6. Charge and discharge curves of as-prepared LiMnPO4 powders

Int. J. Electrochem. Sci., Vol. 6, 2011

2029

Fig.6 showed the charge-discharge curves of as-prepared nano-LiMnPO4, with cycling voltage of 3V- 4.5V and a constant current density of 2.5 mA g-1. As can be seen in Fig.4, rather than small voltage plateau at 4.2 V and 3.9 V for charging and discharging curves respectively, the voltage shows a continuous change for both processes. And this is far different with the theoretical behaviors of Mn2+/Mn3+ couple. Besides, the reversible capacity is about 40 mAh g-1. All of the phenomena mean that the as-prepared nano-LiMnPO4 suffers from bad polarizations during the lithiation/delithiation reaction. Keeping in mind that the particle sizes are very small as shown in Fig. 4, it seems more reasonable to attribute low utilization and high polarization of LiMnPO4 to sluggish kinetics of lithium transport within the particles rather than too long diffusion lengths. It is believed that this can be improved by compositing nano- LiMnPO4 particles with conductive carbonous materials. On the other hand, it is reported that for nano-materials, two-phase equilibrium reactions tend to turn into solid solution, presenting a continuous change in voltage should be observed in charge and discharge curves [29, 30]. A large irreversible capacity observed especially during the first cycling is most likely due to the oxidation of the electrolyte at relatively high cut-off voltage. [31] Some reports have demonstrated that the adding of carbon to the reactants and ball-milling before annealing enhances the electrochemical performance. [32] Further research was in progress. Anyway, the above results demonstrated that electrochemically active LiMnPO4 can be obtained by this method.

4. CONCLUSION Nano-MnPO4H2O can be successfully prepared by a soft mechanochemical method for the first time. The as-prepared nano-LiMnPO4 consisted of uniform particles with narrow size distribution. Nano-LiMnPO4 is prepared using this nano-MnPO4H2O as precursor. Though the reversible capacity is only 40 mAh g-1, it is evident that electrochemically active nano-LiMnPO4 can be obtained by this method and applicable electrochemical performance will be reached by optimizing preparation process only. ACKNOWLEDGEMENT This work is supported by the NSFC (Grand No.20901046), the State Key Project of Fundamental Research for Nanoscience and Nanotechnology (Grant No. 2011CB935902) and the State Key Project of International Cooperation (Grant No. 2010DFA72760). References 1. Ganjali M R, Poursaberi T, Khoobi M, Shafiee A, Adibi M, Pirali-Hamedani M, Norouzi P, Int. J. Electrochem. Sci. 6 (2011) 717-726. 2. Mai L Q, Gao Y, Guan J G, Hu B, Xu L, Jin W, Int. J. Electrochem. Sci. 4 (2009) 755-761. 3. B. Ellis, Wang Hay Kan, W. R. M. Makahnouk, L. F. Nazar, J. Mater. Chem. 17 (2007) 3248 3254. 4. Dominko R, Bele M, Gaberscek M, Remskar M, Hanzel D, Pejovnik S, Jamnik J, J. Electrochem. Soc. 152 (2005) A607-A610. 5. Huang H, Yin S C, Nazar L F, Electrochem. Solid-State Lett. 4 (2001) a170-a172. 6. Chung S Y, Bloking J T, Chiang Y M, Nat. Mater.1 (2002) 123-128.

Int. J. Electrochem. Sci., Vol. 6, 2011

2030

7. A.Vadivel Murugan, T. Muraliganth, A. Manthiram, J. Phys. Chem. C. 112 (2008) 14665-14671. 8. Gangulibabu, Kalaiselvi N, Bhuvaneswari D, Doh CH, Int. J. Electrochem. Sci. 5 (2010) 15971604. 9. Jayaprakash N, Kalaiselvi N, Periasamy P, Int. J. Electrochem. Sci. 3 (2008) 476-488. 10. Guohua Li, Hideto Azuma, Masayuki Tohda, Electrochem. Solid-State Lett. 5 (2002) A135-A137. 11. Zhumabay Bakenov, Izumi Taniguchi, Electrochem. Commun. 12 (2010) 7578. 12. Daiwon Choi, Donghai Wang, In-Tae Bae, Jie Xiao, Zimin Nie, Wei Wang, Vilayanur V. Viswanathan, Yun Jung Lee, Ji-Guang Zhang, Gordon L. Graff, Zhenguo Yang, Jun Liu, Nano Lett. 10 (2010) 27992805. 13. V. Koleva, R. Stoyanova, E. Zhecheva, Mater. Chem. Phys. 121 (2010) 370377. 14. Seung-Min Oha, Sung Woo Oha, Seung-Taek Myungb, Sung-Man Leec, Yang-Kook Suna, J. Alloys Compd. 506 (2010) 372376. 15. T. Drezen, N.-H. Kwon, P. Bowen, I. Teerlinck, M. Isono, I. Exnar, J. Power Sources 174 (2007) 949953. 16. C. Delacourt, P. Poizot, M. Morcrette, J.-M. Tarascon, C. Masquelier, Chem. Mater. 16 (2004) 9399. 17. Takayuki Doi, Shota Yatomi, Tetsuya Kida,Shigeto Okada, Jun-ichi Yamaki, Cryst. Growth Des. 9 (2009) 4990-4992. 18. T.R. Kim, D.H. Kim, H.W. Ryu, J.H. Moon, J.H. Lee, S. Boo, J. Kim, J. Phys. Chem. Solids. 68 (2007) 1203-1206. 19. D. Wang, H. Buqa, M. Crouet, G. Deghenghi, T. Dreen, I. Exnar, N.-H. Kwon, J.H. Miners, L. Poletto, M. Graetzel, J. Power Sources, 189 (2009) 624-628. 20. Haisheng Fang, Liping Li, Yong Yang, Guofeng Yan, Guangshe Li, Chem. Commun. (2008) 11181120. 21. N.-H. Kwon, T. Drezen, I. Exnar, I. Teerlinck, M. Isono, M. Grtzel, Electrochem. Solid-State Lett. 9 (2006) A277A280. 22. Yourong Wang, Yifu Yang, Yanbo Yang, Huixia Shao, Solid State Commun. 150 (2010) 81-85. 23. Yuanguang Zhang, Yi Liu, Shengquan Fu, Fan Guo, Yitai Qian, Bull. Chem. Soc. Jpn. 79 (2006) 270-275. 24. Banjong Boonchom, Sujittra Youngme, Santi Maensiri, Chanaiporn Danvirutai, J. Alloys Compd. 454 (2008) 78-82. 25. Jie Xiao, Wu Xu, Daiwon Choi, Ji-Guang Zhang, J. Electrochem. Soc. 157 (2010) A142-A147. 26. Banjong Boonchom, Rattanai Baitahe, Zongporn Joungmunkong, Naratip Vittayakorn, Powder Tech. 203 (2010) 310314. 27. Thomas Witzke, Reinhard Wegner, Thomas Doering, Herbert Pollmann, Walter Schuckmann, American Mineralogist, 85 (2000) 847-849. 28. H. Fang, Z. Pan, L. Li, Y. Yang, G. Yan, G. Li, S. Wei, Electrochem. Commun. 10 (2008) 10711073. 29. P. Ramesh Kumar, M. Venkateswarlu,b Manjusri Misra, Amar K. Mohanty, N. Satyanarayanaa, J. Electrochem. Soc. 158 (2011) A227-A230. 30. Guoying Chen, Thomas J. Richardson, J. Electrochem. Soc. 156 (2009) A756-A762. 31. Haisheng Fang, Liping Li, Guangshe Li, Chem. Lett. 36 (2007) 436-437. 32. N. N. Bramnik, H. Ehrenberg, J. Alloys Compd. 464 (2008) 259-264.

2011 by ESG (www.electrochemsci.org)