Synthesis of Promising Cathode Material for Lithium Polymer Batteries

. Original method for synthesis of lithium vanadium phosphate was developed. The method includes two stages: 1st, synthesis of iron phosphate from a mixture of ammonium dihydrophosphate and metal oxide; and 2st, synthesis of lithium vanadium phosphate by thermal lithiation of the product obtained in the 1st stage, with mechanical activation of the precursor in the course of plastic deformation. Our results would provide some basis for further improvement on the Li 3 V 2 (PO 4 ) 3 electrode materials for advance lithium-ion batteries.


Introduction
The modern energy, in particular, hydrogen, requires the development of new efficient storage systems of energy generation and accumulating. The electricity accumulat ion to create power plants based on renewable energy sources (RES) is relevant due to the variability derived energy. The energy storage generated by small power plants for the subsequent smoothing of peak loads is very important task. In addition to small and large energy, a significant need for highly efficient generators of electricity and batteries demand on transport, portable (mobile phones, gadgets, laptops, etc.) technology, aviation, space and other fields [1][2][3]. Fo r the autonomous wind and solar power sources it is appropriate to use electrochemical batteries. They are mostly limit the cost performance, reliability and efficiency of wind and solar power plants with a capacity up to 100 kW. At the mo ment lithiu m poly mer batteries are the most promising rechargeable chemical current sources: they dominate due to their light weight, high density electrical energy.
Recently, the demand for lithiu m poly mer battery has increased, which is due both to the tendency toward miniaturization of electronic boards and to the increased requirements imposed by power consumers [4][5][6][7]. The development of lithiu m poly mer battery substantially expands the opportunities of modern miniature devices, such as smart cards, imp lanted medical devices, memo ry units, various sensors, and converters. One of the main difficult ies in creation of film batteries consists in development of efficient cathode materials.
Particularly, monoclinic Li 3 V 2 (PO 4 ) 3 has emerged as one of the pro mising cathode candidates for h igh power lithiu m-ion batteries due to its high theoretical capacity, high operate voltage (3.6 V, 4.1 V) and good ion mobility [10]. The large interstitial spaces created by units allows fast ion migration in three d imensions, and three reversible lithiu m ions can be totally extracted fro m the lattice of Li 3 V 2 (PO 4 ) 3 with in a range of 3.0 and 4.8 V with the highest theoretical capacity o f 197 mA*h/g obtained . However, the power perfo rmance of Li 3 V 2 (PO 4 ) 3 is seriously limited by the poor electronic conductivity (2.4 × 10 -7 S/cm) .
Up to now, tremendous effective approaches have been investigated to overcome these obstacles by minimizing the particle size, coating with carbon and doping with metal ions [10].
In the methods known from the literature, synthesis of lithiu m metal phosphates is a double-stage thermal synthesis of ternary mixtures: ammoniu m dihydrophosphate, metal o xide, and lithiu m co mpounds. However, it has been found that its mechanis m is rather complicated and presumably includes several parallel processes. Therefore, the following two-stage process model has been suggested: 1st, synthesis of metalphosphate from a mixtu re of ammoniu m dihydrophosphate; and 2nd, synthesis of lithiu m metal phosphate by thermal lithiation of the product obtained in the 1st stage [6,8].
It has been shown previously that the mechanical activation of a precursor in a high-pressure apparatus of the Bridg man anvil type can be successfully used to synthesize h ighly dispersed cathode materials for lithiu m batteries [11][12][13]. Therefore, we studied the effect of mechanical activation on synthesis and electrochemical properties of lithium vanadium phosphate.

Experimental section
We chose NH 4 H 2 PO 4 , Li 2 CO 3 , and V 2 O 3 of chemically pure grade as objects of study. Starting mixtures of powdered components were prepared by mixing in a mortar. In the first stage of synthesis, a mixtu re of NH 4 H 2 PO 4 and V 2 O 3 was annealed in a mu ffle furnace at a temperature of 750°C for 6h. In the second stage, 20% Li 2 CO 3 was added to the product obtained and the mixtu re was thermally treated at temperatures of 600, 700, and 800°C for 4-10h. The plastic deformation of the precursors under a pressure of 1.5 GPa was performed at room temperature on anvils made of a VK6 hard alloy, with the working surfaces of the anvils having a diameter o f 15 mm and anvil rotation angle equal to 300°. The heat effects that occur in the prepared materials at temperatures fro m room temperature to 800ºC were studied by scanning differential calorimet ry on a TA Instruments model Q100 instrument at a scanning rate of 20 K* min -1 ; the sample weight was fro m 1 to 3 mg. Thermograv imetric analysis was performed on TA Instrument model Q500 thermogravimet ric analy zer; the scan rate was 20 K* min -1 and the sample weight was 2-4 mg . X-ray diffraction (XRD) measurements were performed on an Empyrean diffracto meter using Cu K radiat ion (t wo wavelengths -1.5406 and 1.5444E were used for calculations, considering a 2:1 rat io of their intensities in the doublet) and scanning over a 2Ө range of 5°-100°. The calculated phase composition was verified by dual phase Rietveld refinement using the MRIA software program [14].
A composite electrode was prepared by mixing 80 wt % Li 3 V 2 (PO 4 ) 3 , 10 wt % PVDF and 10 wt % acetylene black with NMP (1-methyl-2-pyrrolidone) to form the slurry, which was then spread on to a alu minum foil and dried at 120ºC for 24h in a vacuum oven. The batteries were assembled in an argon-filled gloved-box, in which o xygen and mo isture level less than 1 pp m, and the electrolyte was 1 M LiPF 6 in a mixture of EC (ethylene carbonate), DMC (dimethyl carbonate) and EM C (ethylmethyl carbonate) (1:1:1 by weight). Typically, a working electrode of 1.5 cm 2 was prepared with the active material mass loading of 3.0 mg per cm 2 . The coin cell was fabricated using the lithiu m metal as a counter electrode. Electrochemical measurements were conducted with galvanostatic charge and discharge on a Elins P-20X8 cell testing apparatus in the voltage range of 3.0 -4.3 V at room temperature. The discharge-rate range is fro m 0.5 to 1 C to 3.0 V and charge is at 0.5 C to 4.3 V. The C-rates and storage capacities were calculated fro m the mass of Li 3 V 2 (PO 4 ) 3 with the amount of carbon being subtracted (1C = 130 mA*h/g). Cyclic voltammetry (CV) measurements were performed on a Elins P-20X8 electrochemical workstation. CVs were conducted in the cut-off voltage range of 3.0-4.3 V versus Li/Li + at a scan rate of 0.1 mV/s.

Results and discussion
The thermogravimetric curves for the init ial mixture of V 2 O 3 and NH 4 H 2 PO 4 and the same mixture after being subjected to mechanical activation were similar. In both cases, we observed a process spanning the temperature range of 100-800°C featuring a characteristic exothermic peak (Fig. 1). The enthalpy associated with the exothermic peak was 2177 and 2519 J/g for the initial mixture and for the mechanically activated one, respectively. The exothermic processes in the specimens being heated were acco mpanied by the decrease in sample weight by 30.9 and 32.1% for the in itial mixture and for the mechanically act ivated one, respectively (Fig.  2).   [8] for the phases VPO 4 : V 2 O 3 , the following ratios were obtained: 11: 1 (Fig. 3).
The thermogravimetric curves for the initial mixture of VPO 4 and Li 2 CO 3 and the same mixture after being subjected to mechanical activation were similar. In both cases, we observed a process spanning the temperature range of 150-800°C featuring a characteristic exothermic peak (Fig. 4). The enthalpy associated with the exothermic peak was 3796 and 3946 J/g for the initial mixture and for the mechanically activated one, respectively. The exothermic processes in the specimens being heated were acco mpanied by the decrease in sample weight by 17.4 and 15.6% for the in itial mixture and for the mechanically act ivated one, respectively (Fig.  5). For the initial mixture, weight loss occurred as a onestep process, and a major fract ion of the sample weight was lost with in the range of temperature of 200 to 650°C. For the mechanically activated sample, weight loss started 30°C lower and ended 50°C lo wer co mpared to the initial mixture. Th is difference can be explained by the fact that the chemical reactions partially occurred during mechanical activation. Structurization processes, which may not exhib it associated heat effects, can proceed in parallel with the physicochemical processes in our samples.
We identified four peaks of Li 3 V 2 (PO 4 ) 3 and LiVP 2 O 7 peaks in XRD patterns of the initial VPO 4 -Li 2 CO 3 mixture annealed at 700°C for 10h. By applying dual phase Rietveld refinement to our XRD data [8], we established that the phase ratio Li 3 V 2 (PO 4 ) 3 :LiVP 2 O 7 was 8:2 for the init ial mixture and 9:1 fo r the mixture annealed at 800°C for 6h. It was found that the plastic deformation of the precursor is effective in the second stage of Li 3 V 2 (PO 4 ) 3 synthesis. XRD patterns of samples subjected to mechanical activation and annealing at 600°C for 7h featured peaks due to Li 3 V 2 (PO 4 ) 3 and LiVP 2 O 7 phases, as shown in Fig. 6. Applying dual phase Rietveld refinement yielded the phase ratio Li 3 V 2 (PO 4 ) 3 :LiVP 2 O 7 = 9:1. With annealing temperatures raised to 750°C (4 h ) the phase ratio was 11:1.  [5,9,10].
We thus see that the mechanical activation of the precursor in the Bridg man anvil apparatus shortened the annealing time required to achieve the desired nanodispersed material. We exp lain these results by considering the follo wing processes. Plastic deformation induces numerous structural defects in individual solids with different chemical natures. These processes are particularly active in binary mixtures; namely, mass transfer processes resulting in the formation of solid solutions are very intense under these conditions. As was established earlier, structure format ion processes proceed considerably more facile in mixtures subjected to deformations [15].

Fig. 6. X-ray diffraction pattern of lithium vanadium phosphate
Electrochemical tests were carried out for electrodes prepared using the active mass based on the VPO 4 -Li 2 CO 3 mixture subjected to mechanical activation followed by annealing at 750°C. The tests showed that our cathodes based on lithiu m vanadium phosphate displayed reversible cycling at current densities of 0.2-1.0 mA/cm 2 . Figure 7 shows the CV curves of Li 3 V 2 (PO 4 ) 3 electrode at a scan rate of 0.1 mV/s fro m 3.0 to 4.3 V.

Fig. 7. Results of cyclic voltammetry analysis
Each of the CV curves includes three oxidation peaks and three reduction peaks, which is consistent with the galvanostatic charge/discharge curves. The good overlap of CV cycles and the symmetry of the o xidation and reduction peaks in the CV curves indicate the good reversibility of lithium insertion/deinsertion reactions.
A comparison of the results we obtained with published data demonstrated that electrodes based on Li 3 V 2 (PO 4 ) 3 synthesized in the study compare well in specific capacity and stability with the known foreign and domestic analogues [5,9,10].

Conclusions
An original method for synthesis of lithiu m vanadium phosphate was developed. The method includes two stages: 1st, synthesis of vanadium phosphate fro m a mixtu re of ammon iu m d ihydrophosphate and metal oxide; and 2st, synthesis of lithiu m iron phosphate by therma l lithiation of the product obtained in the 1st stage, which includes mechanical activation of the precursor in the course of plastic deformation. Our results would provide some basis for further improvement on the Li 3 V 2 (PO 4 ) 3 electrode materials.