Conductivity of Thionyl Chloride-Lithium Tetrachloroaluminate Solutions

The specific conductivity of solutions of LiA1C14 dissolved in SOC12 was determined as a function of composition and temperature. An analytical expression from which the conductivity can be calculated is given as a function of the mole fraction of LiA1C14 and temperature in the ranges from 0 ~ to 0.37~ and from -20 ~ to +70~ respectively. LiAIC14-SOC14 mixtures are used commercially as electrolytes in lithium batteries (1). Thionyl chloride itself has a low melting point ( 104.5~ a low vapor pressure (ca. 15 kPa at 25~ a high boiling point (77~ but a low electrical conductivity (ca. 3.5 10 -9 S cm -~ at room temperature). To be useful in a battery, salts must be dissolved in the SOC12. One common choice of such a salt with a large solubility in thionyl chloride is LiA1C14. However, the conductivities of these solutions are only approximately known (1-8). We decided to check and extend the available conductivity data. Apparently at all temperatures studied in the liter* Electrochemical Society Active Member. ature there is an optimum concentration of LiA1C14 for maximum conductivity. During the present work we found the position of this maximum conductivity and derived an analytical expression from which the specific conductivity of any LiA1C14SOC12 solution can be calculated with a high degree of precision. The high conductivity of LiA1C14-SOC12 solutions compared with other lithium inorganic room temperature electrolytes was pointed out by Auborn and co-workers (2), who reported among other things the conductivity as a function of concentration of LiA1C14 in SOC12 up to 2.0M at 25~ They found that a maximum conductivity of 8.0204 S cm -1 was reached in ca. 1.7M solution. Very dilute soluDownloaded 05 Oct 2009 to 192.38.67.112. Redistribution subject to ECS license or copyright; see http://www.ecsdl.org/terms_use.jsp 324 J. Electrochem. Sac., Vol. 136, No. 2, February 1989 9 The Electrochemical Society, Inc. tions of LiA1CI4 in SOC12 have also been studied by others (3). The combined temperature and concentration dependences of the conductivity, density, and viscosity were studied extensively by Venkatasetty and Szpak (5-7). In these works, conductivities are reported as a function of concentration. Because of volume changes, concentration itself is a function of temperature. This function has only been reported indirectly in density vs. temperature relations at certain concentrations. In our estimation this is not adequate. Either a solution must be prepared at the temperature in question or a correction procedure is required. This procedure, involving knowledge of concentration changes cannot be easily performed because the previous authors (5-7) have not supplied the data required nor described exactly how they obtained their corrected concentrations. Without corrections, only limited accuracies can be attained. To allow for high precision and simple calculations