Ionic conductivity of LiHf2(PO4)3 with NASICON-type structure and its possible application as electrolyte in lithium batteries

Article

Ionic conductivity of LiHf₂(PO₄)₃ with NASICON-type structure and its possible application as electrolyte in lithium batteries

A. Martínez-Juárez, J.M. Amarilla, J.E. Iglesias, and J.M. Rojo^*

Instituto Ciencia de Materiales de Madrid, CSIC, Cantoblanco, 28049 Madrid, Spain

Received: August 10, 1996; October 10, 1996

Este trabalho reporta medidas de condutividade iônica realizadas para o LiHf₂(PO₄)₃ calcinado a 1100 °C. As respostas devidas aos grãos - interior e junção -puderam ser identificadas tanto nas curvas de impedância, como na parte real das curvas de condutividade vs. freqüência. A energia de ativação, associadaao movimento dos íons Li⁺ no interior dos grãos, é 0,33 eV, enquanto que aquela associada à condutividade total cc, está na faixa de 0,36-0,47 eV. os resultados desta última dependem da contribuição relativa devida ao interior e à junção de grão. A possível aplicação do LiHf₂(PO₄)₃ como eletrólito foi testada para a pilha Li/LiHf₂(PO₄)₃/LiMn₂O₄. Observou-se que o potencial de equilíbrio aumenta de 0,076 V a 2,217 V, quando a temperatura varia de 28 a 148 °C.

The ionic conductivity of LiHf₂(PO₄)₃ calcined at 1100 °C has been measured. Grain interior and grain boundary responses can be distinguished in the impedance plots as well as in the real part of conductivity vs frequency plots. The activation energy associated with the motion of Li⁺ ions inside the grains is 0.33 eV while the activation energy corresponding to the total dc conductivity changes from 0.36 to 0.47 eV, depending on the relative contribution of grain interior and grain boundary. The possible application of LiHf₂(PO₄)₃ as an electrolyte has been tested in the Li/LiHf₂(PO₄)_3/LiMn₂O₄ cell. The equilibrium potential increases from 0.076 V to 2.217 V when the temperature is raised from 28 to 148 ºC.

Keywords: NASICON, ionic conductivity, lithium hafnium phosphate, solid state lithium batteries

Introduction

The worldwide increase in consumer electronics and the challenge of the electrical car account for the significant effort to develop high energy rechargeable batteries. Among the batteries those based on lithium show the best performance. The solid electrolytes have several advantages as compared with liquid electrolytes. In particular, they do not produce either corrosion or passivation of the electrodes, and the safety of the battery is improved.

LiHf₂(PO₄)₃ is a NASICON-type compound which shows a relatively high Li⁺ conduction as compared with other compounds of the LiM₂(PO₄)₃, M^IV = Ge,Ti family^1-10. The detailed crystal structure of this compound is not available yet, but the powder X-ray diffraction pattern has been indexed^11-16 on the basis of a rhombohedral lattice in space group , which is the usual one found for the NASICON-type structure^1,17-22. However, it has been also reported that LiHf₂(PO₄)₃ can exhibit a lower symmetry as other NASICON-type compounds do^23-27. Thus, a monoclinic P2₁/n group was found²⁸ for a sample sintered below 1100 ºC, while the symmetry was rhombohedral when the sample was sintered above that temperature.

The ionic conductivity reported for this compound by several authors is rather variable. Some of them^15,29 have ascribed the conductivity to motion of Li⁺ ions inside the grains, while others^12-14,16 have reported the total conductivity, i.e. that associated with grain interior and grain boundary, or else they do not identify which of these contributions is dominant. In any case the conductivity values seem to be affected by the lack of resolution between grain interior and grain boundary response, and significant differences in the activation energy (0.32-0.44 eV) have been also reported^12-16,28,29.

The aim of this paper is to study the ionic conductivity of LiHf₂(PO₄)₃ as well as the possible application of this material as an electrolyte in solid state batteries.

Experimental

LiHf₂(PO₄)₃ was prepared by calcination of a stoichiometric mixture of Li₂CO₃ (Fluka, >99%), HfO₂ (Aldrich, 99.9%), and (NH₄)₂HPO₄ (Fluka, >99%). The reagents were previously dried at 100 ºC for 12 h, then stoichiometric amounts of these compounds were thoroughly mixed and calcined in a platinum crucible at the following temperatures: 300, 600, 700, 800, 900, and 1000 ºC. Thermal treatments were accumulative up to 1000 ºC. At each defined temperature the time spent was 10 h except at 300 ºC where the time was 6 h. The mixture was ground before each thermal treatment.

X-ray powder diffraction patterns were taken at room temperature in a PW-1710 Philips diffractometer with Cu Ka radiation. The scan was carried out in steps of 0.02º, counting for 0.5 s at each step. The peaks were fitted with Ka₁-Ka₂ doublets, and the position of each peak was taken to be that of the Ka₁ component, l =1.5405981 Å.

Electrical conductivity measurements were carried out by the complex impedance method using a 1174 Solartron frequency response analyzer. The pellets (ca. 6 mm diameter and 1 mm thickness) obtained in the last thermal treatment of the preparation procedure were additionally calcined at 1100 ºC for 10 h. Gold electrodes were deposited on the two faces of the pellets by vacuum evaporation. The frequency range used was 10^-1-10⁵ Hz. The measurements were carried out on the pellets heated at increasing temperature in the range 40-300 ºC.

The usefulness of LiHf₂(PO₄)₃ as a solid electrolyte in lithium batteries has been tested by using the Li/LiHf₂(PO₄)₃/LiMn₂O₄ cell. The composite cathode is formed by LiMn₂O₄ as active material (50% by mass), carbon black as electronic conductor (25% by mass), ethylene propylene diene monomer, EPDM, as binder (10% by mass) and LiHf₂(PO₄)₃ (15% by mass). This composite electrode was made by mixing the powders in cyclohexane, evaporating the solvent, and pressing the mixture at 1 Ton cm^-2 to obtain a disk of 0.28 cm². As electrolyte, a sintered pellet of the NASICON compound was used. A lithium foil was used as anode. The different components were placed in a sealed Swagelock cell under a dried atmosphere containing <2ppm H₂O. The cell equilibrium potential was measured in the range 25-150 ºC. To that end the sealed cell was immersed in a silicone bath and the measurements were carried out at each temperature after half an hour of stabilization.

Results and discussion

The LiHf₂(PO₄)₃ sample prepared in this work showed a powder X-ray diffraction pattern which was indexed on the basis of a rhombohedral cell with hexagonal parameters a_H = 8.8285(5), c_H = 22.019(1) Å. These parameters agree well with other reported for the rhombohedral phase^11-16. Our parameters were obtained from a least squares refinement procedure in which the 2Q values of 41 unambiguously indexed reflections were used as observable data.

The impedance plots (imaginary -Z" vs. real Z’) recorded at different temperatures are shown in Fig. 1. At low temperatures (40 ºC) two arcs and one spike are observed. The arcs found at high and low frequency are ascribed to grain interior and grain boundary response, respectively, in agreement with the assignement reported for other ceramic materials³⁰ . The capacitance associated with these arcs is 30 pF for the high-frequency arc and 64 nF for the low-frequency one. The spike is ascribed to the blocking effect of Li⁺ ions at the electrode surface. The capacitance of the spikes is in the range 1-10 mF. When the sample is heated at higher temperatures the arcs disappear from the plots, first the high-frequency arc and then the low-frequency one. Above 200 ºC only the spike are clearly observed.

The real part of the conductivity (s) vs angular frequency (w=2pf) is plotted in Fig. 2. A dispersive regime (I) due to the electrode response is observed at low frequency. At intermediate and high frequency two plateaus (II and III) are detected. The plateau II is affected by the grain boundary dc response, while the other (III) is ascribed to the dc response of the grain interior, in agreement with the information deduced from the impedance plots. The experimental data have been fitted by using the power expressions: i) s = Ew^e for the electrode response, ii) s = s'_dc + Bw^P for the grain boundary response, and iii) s = s_dc + Awⁿ for the grain interior response. s_dc and s'_dc are the dc conductivities corresponding to each plateau. Ew^e, BwP, and Awⁿ are the ac conductivities of the three responses. A and E are temperature-dependent parameters , and e, p, and n are the slopes of the three dispersive regimes. The solid curve in Fig. 2 shows a good fit of the experimental data. Then, by this procedure, the dc conductivity corresponding to motion of Li⁺ ions inside the grains, through the border of the grains, as well as the total dc value of the pellet, have been determined.

Variation of dc conductivity (log₁₀ s_dc) vs inverse temperature (1000/T) is shown in Fig. 3. In this figure we have included the values of the grain interior (open circles) and total (open triangles) conductivity. All the experimental data are well fitted by Arrhenius equations, sa = s₀ exp(-E/KT), where s₀ is a pre-exponential factor, E the activation energy, and K the Boltzmann constant. The total dc conductivity (solid line) is close to that of the grain interior (dotted line) for temperatures above 160 ºC, indicating that the total conductivity is dominated by the grain interior response. However, below that temperature the total conductivity differs from the grain interior one because of the grain boundary contribution. The activation energy associated with the total conductivity is also a parameter whose value depends on the relative grain interior and grain boundary contributions. Thus, a value (0.36 eV) close to that corresponding to the grain interior response (0.33 eV) is obtained when the total conductivity is dominated by the grain interior contribution (above 160 ºC); the activation energy is higher (0.47 eV) when the total conductivity is clearly affected by the grain boundary response (below 160 ºC).

Finally, we have tested the possible application of LiHf2(PO₄)₃ as an electrolyte in solid state lithium batteries. Fig. 4 shows the variation of the equilibrium potential (E_eq.) of the Li/LiHf₂(PO₄)₃/LiMn₂O₄ cell vs. temperature. At 28 ºC the E_eq. value is quite small (0.076 V). When the cell temperature increases E_eq. increases monotonically. The slope is larger in the range 60-110 ºC. Above 110 ºC the slope trends to level, and at 150 ºC E_eq. reaches 2.217 V. Although this value is lower than that normally found (ca. 3.2 V) when a liquid electrolyte, such as 1M LiClO₄ dissolved in propylene carbonate, is used^31,32, the observed increase in E_eq. shows that the response of the LiHf₂(PO₄)₃ electrolyte is improved with the cell temperature, and hence it is potentially useable in solid state lithium batteries. The increase with temperature of E_eq. is consistent with the increase in Li⁺ conductivity, as deduced from impedance data.

Conclusions

Grain interior and grain boundary response are distinguished in a LiHf2(PO₄)₃ pellet previously calcined at 1100 ºC. The total dc conductivity and its associated activation energy are very sensitive to the relative contribution of grain interior and grain boundary. For Li⁺ motion inside the grains the dc conductivity at rt and the activation energy are 1.3 x 10^-5 ohm^-1 cm^-1 and 0.33 eV, respectively. These data together with the observed increase in the equilibrium potential with the cell temperature show that LiHf₂(PO₄)₃ can be used as an electrolyte in solid state Li-batteries.

Acknowledgement

Financial support by CICYT (Project MAT 95-0899) is gratefully acknowledged.

• 1. Petit, D.; Colomban, Ph.; Collin, G.; Boilot, J.P. Mat. Res. Bull. 1986, 21, 365.
• 2. Li, S.; Cai, J.; Lin, Z. Solid State Ionics 1988, 28-30, 1265.
• 3. Sudreau, F.; Petit, D.; Boilot, J.P. J. Solid State Chem. 1989, 83, 78.
• 4. Casciola, M.; Costantino, U.; Krogh Andersen, I.G.; Krogh Andersen, E. Solid State Ionics 1990, 37, 281.
• 5. Colomban, Ph.; Mouchon, E.; Belhadj-Tahar, N.; Badot, J.C. Solid State Ionics 1992, 53-56, 813.
• 6. Nomura, K.; Ikeda, S; Ito, K.; Einaga, H. Solid State Ionics 1993, 61, 293.
• 7. Ando, Y.; Hirose, N.; Kuwano, J.; Kato, M., Otsuka, H. In Ceramic Today - Tomorrows Ceramics, Proc. 7th Internat. Meeting Modern Ceramics Technologies, Montecatini Term; Vincenzini, P. Ed.; Elsevier, 1991, Italy, p. 2245.
• 8. Aono, H.; Sugimoto, E.; Sadaoka, Y.; Imanaka, N.; Adachi, G. Solid State Ionics 1991, 47, 257.
• 9. Aono, H.; Sugimoto, E.; Sadaoka, Y.; Imanaka, N.; Adachi, G. J. Electrochem. Soc. 1993, 140, 1827.
• 10. Paris, M.A.; Martinez-Juarez, A.; Rojo, J.M.; Sanz, J. J. Phys.: Condens. Matter 1996, 8, 1.
• 11. Sljukic, M.; Matkovic, B.; Prodic, B.; Scavnicar, S. Croat. Chem Acta 1967, 39, 145.
• 12. Taylor, B.E.; English, A.D.; Berzins, T. Mat. Res. Bull. 1977, 12, 171.
• 13. Shanon, R.D.; Taylor, B.E.; English, A.D.; Berzins, T. Electrochim. Acta 1977, 22, 783.
• 14. Subramanian, M.A.; Subramanian, R.; Clearfield, A. Solid State Ionics 1986, 18/19, 562.
• 15. Chowdari, B.V.R.; Radhakrishnan, K.; Thomas, K.A.; Subba Rao, G.V. Mat. Res. Bull. 1989, 24, 221.
• 16. Winand, J.M.; Rulmont, A.; Tarte, P. J. Solid State Chem. 1991, 93, 341.
• 17. Hagman, L.; Kierkegaard, P. Acta Chem. Scand. 1968, 22, 1822.
• 18. Hong, H.,Y-P. Mat. Res. Bull. 1976, 11, 173.
• 19. Goodenough, J.B.; Hong, H.Y-P.; Kafalas, J.A. Mat. Res. Bull. 1976, 11, 203.
• 20. Alamo, J.; Roy, R. J. Mater. Sci. 1986, 21, 444.
• 21. Tran Qui, D.; Hamdoune, S.; Soubeyroux, J.L.; Prince, E. J. Solid State Chem. 1988, 72, 309.
• 22. Alami, M.; Brochu, R.; Soubeyroux, J.L.; Gravereau, P.; Le Flem, G.; Hagenmuller, P. J. Solid State Chem. 1991, 90, 185.
• 23. Rudolf, P.R.; Clearfield, A.; Jorgensen, J.D. Solid State Ionics 1986, 21, 213.
• 24. Rodrigo, J.L.; Alamo, J. Mat. Res. Bull. 1991, 26, 475.
• 25. Angenault, J.; Couturier, J.C.; Souron, J.P.; Siliqi, D.; Quarton, M. J. Mater. Sci. Lett. 1992, 11, 1705.
• 26. Sanz, J.; Rojo, J.M.; Jimenez, R.; Iglesias, J.E.; Alamo, J. Solid State Ionics 1993, 62, 287.
• 27. Martinez-Juarez, A.; Rojo, J.M.; Iglesias, J.E.; Sanz, J. Chem. Mater. 1995, 7, 1857.
• 28. Aono, H.; Sugimoto, E.; Sadaoka, Y.; Imanaka, M.N.; Adachi, G. Solid State Ionics 1993, 62, 309.
• 29. Kuwano, J.; Sato, N.; Kato, M.; Takano, K. Solid State Ionics 1994, 70/71, 332.
• 30. Macdonald, J.R. In Impedance Spectroscopy; John Wiley and Sons; USA, 1987.
• 31. Gummow, R.J.; de Kock, A.; Thackeray, M.M. Solid State Ionics 1994, 69, 59.
• 32. Koksbang, R.; Barker, J.; Shi, H.; Saidi, M.Y. Solid State Ionics 1996, 84, 1.

Publication Dates

• Publication in this collection
  31 Jan 2011
• Date of issue
  June 1997

History

• Accepted
  10 Oct 1996
• Received
  10 Aug 1996