Bulletin ofElectrochemistry 14 (10) October 1998, pp 329-331 0256-1654/98/$ 3-50 @ 1998 CEC'RI

SYNTHESIS OF LITHIUM VANADATE AND ITS CHARACTERISATION

T SARAVANAN, R THIRUNAKARAN, A M STEPHAN, N G RENGANATIIAN

K R MURAU, V SUBRAMANIAN AND N MUNTYANDI

Central Electrochemical Research Institute, Karaikudi 630 006, INDIA

[Received: 1 June 1998 Accepted: 22 June 1998}

In view of h~b specific capacity, possibility of lithium uptake to 3.0 Li/mol with respect to cycleability, structural stability, high rate capability (due to high diffusion of lithium) and deep dischllrgeability below 1.6 V lithium vanadate is preferred to V,013 In lithium secondary cells. To have materials of less particle size, sol-gel synthesis coupled with proper dehydration processes are followed to synthesize LiV308 and it is characterized by XRD and FfIR analysis. Results are presented in this communication.

Keywords: Lithium vanadate, sol-gel synthesis, structural stability.

INTRODITCTTON

Much effort has been made in 80's to develop suitable cathode material for secondary lithium cells. TiS2 and V6013 bave been employed extensively (1). TiS2 bad indeed impressive features in tenns of reversibility and rate c~pability 12-3). Howeve.r,reduced energy density and inslability in moist air were the main drawbacks, which limited its application in practical cells. V6013 was thought of suhstituting TiS2 in lithium ct'.lIs and bad been tried as positive electrode 14,5). This material has greater energy density than TiS2 and strict stoicbiometry need not be mentioned for optimum perfonnance.

Their cycling be.haviour at current densities around 1 mNcm2 is cbaracterize.d by bigb and stable specific ca pac ities afte.r the first few cycles 16) that is 0.16­0.20 Ah/g. However, an important drawback is reported for both \'6013 (stoicbiometric) and espe.cially V60 n (non-stoicbiometric). At 1.6 V, a reduction process occurs whicb inhihited further rcchargeahiJity (tl). This is probably duc to structural instability ill the reorganizations connected with the high Li content (2.37 Li+N). This drawback was the main reason to discourage the use of these materials ill practical lithium cells. An attempt was made to overcome this drawback by using a mate.rial whicb, while relaining tbe basic electrochemical fealUres of V60!3' would be able to undergo overdischarges without structural damage. Tbe

Li J+XV:Ps phase investigated in (7) is closely related ill

structure to V6013'

This bronze, a layere.d compound wbose layers are held together by Li+ ions 171 bas proven to be able to insert lithium in its structure with outstanding energy, power and cycling capability. This bronze is superior to V6 in such aspects 0 13 as energy density, rate c~pahility, cycle life and resistance to overdischarge. On this basis this ternary oxide did seem worthy of a further investigation. Early in the development it was realized (8) tbat tbe methods used to prepare oxide strongly influenced its electrochemical properties. It was demonstrat.ed that [81 LiV.Ps prepared in an amorphous glassy slate by rapid quenching from tbe mell had a higher initial capacity than the crystalline analogue. These findings werc, however, not pursued further t"ven though several rese.arch groups reported on the use of crystallinc LiV30 S as host material for lithium inte.rcalation [9- 181.

Several preparation proct'.durt"s have been devised to improve the perfonnance of LiV.Ps including control of stoich iometry by ra pid cooling (I81, more efficient grind ing (191 and addition of incrt nucleation ccntres like silica or alumina to the melt [201. The main problem in all tht"se st"t"nlS to be tbat on slow cool ing, LiV30g cryslallized as a very bard and tough material, which is difficult to process into proper electrode structures that can ma intain their integrity during deep cyc.J ing. It is recently reported that [2l) fully amorphous LiV.Pg oblained from a precipitation technique showed significantly higber capacity, bettrr rate

329

SARAVANAN et al - Synthesis of lithium vanadate and its characterisatrion.

281- 2' s I-:===============::::::Jn 5-16' Ir

1{1·' -1·9 1(;-62 f:============_-=::=J

I-----,;;,------,;J;--__~--*"--~ o 10 20 30 40 50

Vot ume J n band (O/.)

Fig. J: Particle ize analysis of LiVJOS

capability and longer cycle life than conventionally made crystalline UV~Pg.

In the present paper an adoption of sol-ge.l technique has been made to get materials with a controlled degree of crystallinity and a particle morphology well suited to the processing steps involved in electrode fabrication. The method is based on the dehydration of stable gels prepared from LiOH and V20 S as report d ill r20). These gels are vacuum dried and further dehydrated ill a heat treatment step. Powders ohtailled arc characte.riud by XRD, FTIR etc.

EXPERIMENTAL

An aqueous LiV:Ps gel was prepared by mlxmg LiOH, V205 in 1:3 mole ratio. To UOH aqueous solution V20S

suspended solution was slowly added with con tant stirring. The gelatinous precipitate was dtied at 353 K. This powder is then vacuum dried at 393 K for 4 hours. The vacuum dried powder was then he.ated to 873 K for 5-6 hours to get LiV~08 powder.

eo TO

90

60 70

40 50

2030

DTA 010

0 100 200 300 400 500 600 700 800

Deg C

Fig.2: TG &: DTA of LiVps

400

1II

C ::J

..0

0 ~

>,

<ii c Q.>

C 0

10 30 50 70

Scattering angle. 26

Fig.3: X-ray diffractogram of synthesised powder of L1VPS

RESULTS AND DISCUSSION

LiV~08 powder syntbe..<;ized is analyzed using Malvern Instruments for particle size and the result is shown in Fig. 1. Particle size is in tbe range of 8 to 28.1 f.1m. Particle size could have been reduced to still lower range if freeze drying is adopted.

From the thennogravimetric (TG) curves reported r20] it is seen that water is lost in two distinct steps below 423 to 448 K (Fig. 2). Whereas a smaller part of the water roughly corresponding to 0.5 H20 per LiV~08 unit is more strongly bound to the oxide and only released at temperatures above 473 K. Based on these observations and vacuum drying was adopted at 393 K for 4 hours. Even after drying at 873 K for 5 to 6 hours, debydration due to water adsorption was observed and this indicates that initial drying bas to be done for more hours in vacuum.

Fig. 3 shows the X-ray diffractogram of the synthesized powder of UV30 S' The powder is not perfectly crystalline as it is evident from the figure and identification of tbe powder as LiV30g has been made from the 100% peak and it has he,en compared with data r221 and the unit cell parameter has been calculated and it is found to be a = 8.132 A. Specific resistivity was found out from I-V characteristics and it is of the order of 103 ohm-cm and tbis is increased to 107 ohm-cm upon litbiatioll done in a cell constructed using lithium vanadate as cathode, pure I itbium as anode and separated by a polymer electrolyte comprising PMMA/PVC with lithium salt.

From the IR spectra as shown ill Fig. 4 it is evident that th absorption bands ncar 1000 and 950 em-I are associated with stretching vihration of V (2) = 0 (4) and those of V (3) = 0(7) and V (1) = 0 (5) respectively /22] where number in parentheses represents the position of the atoms labeled in

r23 ].

330

SARAVANAN ct al - Synthesis of lithium vanadate and its characterisatrion.

57·68 ,--------------;---------,

38·01 '---7,v;r;---"'.-c;:~--~2000~------:;:;1000~--J

Fig.4: FTTR spectra of LiV30 8

CONCLUSION

Lithium vanadate. bas been prepared by sol-gel method as this material is preferred to V60 L\ in lithium secondary cells due 10 its high cycJebility, slruchJral stability high rale capacity and deep dischargeabilily below 1.6 V. The material prepa red by sol-gel has be.en characterized.

REFERENCES

I. M S Whittingham, Prog Solid Slate Chem, 12 (1978) 41 2. M S Whittingham and J A Panella, Mat Res Bull, 16 (1981)

37 3. G L Hollcch, K M Abraham and S A Rrummer, Proc Symp

Power Sources for Biomedical Implantable Applications and Ambient Temperature Lithium Batteries, Eds B B Owens and N Margali (Electrochem Soc, Princeton, 1980) 384

4. G Pistoia, J Power Sources, 9 (1983) 307 5. D W Murphy, P A Christian, F J Disalvo ,md J N Carides,

J Electrochcm Soc, 126 (1971)) 41)7 6. K M Ahrllham J L Goldman and M D Dempsey,

J EllXtrochem Soc, 128 (1981) 2493 7. A D Wadsley, Acta Cryst, 10 (1957) 261 8. K Nassau and D W Murphy, J Non (ryst Solids, 44 (l981)

297 I). S Pllnero, M PasqulIli and G Pistoia, J ElectrocJrem Soc, 130

(1983) 1226

10. G Pistoill, M Panero, M Tocci, R V Moshtev and V Maney, Solid State Ionics, 13 (11)84) 311

11. G Pistoia, M P<lsqullli, M Tocci, R V Moshtev <lnd V Manev,

J Electrochem Soc, 132 (1985) 281 12. G Pistoia, M PasqU<lli, M Tocci, R V Moshtev, J Power

Sources, 15 (11)85) 13 13. M Pllsquali, G Pistoi<l, M Manev and R V Mostev,

J EI ctrochem Soc, 133 (1986) 2454 14. F Bonino, M Ott<lviani, B Scrosati and G Pistoia,

J Electmchem Soc, 135 (11)88) 12 15. B Scrosati, A Selvaggi, F Croce and W Gang, J Power

Sources, 24 (1988) 287 16. 1 D Raistrich, Solid State lonics, 9 & 10 (1983) 425 17. A Hammou and A Hammouche, Electrochimica Acta, 33

(1988) 1719 18. K West, B achan-Chiri!.tiansen, M J L Ostergard and

T Jacobsen, J Power Sources, 20 (1987) 165 19. V Manev, A Momchilov, A Nassalevsha, G Pistoia and

M Pasquali, J Power Sources, 54 (1995) 501 20. K West, B Zachen-Olristiansen, S Shaarup, Y Saidi,

J Barher, I I Olsen, R Pynenberg and R Koksbang,

J Electrochem Soc, 143 (1996) 820 21. G Pistoia, M PasqU<lli, G Wang and L Li,J Electrochem Soc,

137 (1990) 2365 22. Jin Kawahita, M Majima, T Miura and T Kishi, J Power

Sources, 66 (t 997) 135 23. L A De Picciotto, K l' Adendorff, D C Liles and

M M Thackeray, Solid State Tonics, 62 (1993) 21)7