Titanium disulfide has been found to be attractive as an alternative to graphite as the anode material in rechargeable lithium-ion electrochemical cells that are required to operate at temperatures below -20 °C. By using TiS2 as the anode material, LiCoO2 as the cathode material, and a suitable low-temperature electrolyte described below, it is possible to construct cells that exhibit superior low-temperature characteristics, including relatively high charge/discharge capacities, capabilities for charging and discharging at relatively high rates, and excellent retention of capacity after repeated charge/discharge cycling.

Specific Charge Capacities of TiS2 and Graphite anode materials were measured at three different temperatures. One measurement (the one on graphite at –30 °C) was made at a current density of 0.04 mA/cm2; all the other measurements were made at a current density of 0.4 mA/cm2.

The three immediately preceding articles report improvements in electrolytes for low-temperature, rechargeable lithium-ion cells with carbon (usually graphite) anodes. Unfortunately, in cells with carbon anodes, improvements in electrolytes may not be sufficient by themselves; this is because the performances of such cells are limited by high polarization resistances at the carbon anode surfaces. It has been conjectured that the high polarization resistances are due to the slowness, at low temperatures, of diffusion of ions through the graphite bulk and through surface films (solid/electrolyte interphase) that form on carbon anodes and freeze at low temperatures. Among other things, high polarization resistances make it necessary to charge and discharge at low rates — in some cases as low as I = C/100, where I is the charge or discharge current in amperes and C is the nominal charge capacity in ampere—hours.

Two major reasons for choosing TiS2 as a candidate alternative anode material are that (1) the diffusivity of Li in TiS2 is high and (2) solid/electrolyte interphases are not expected to form on the surfaces of TiS2 electrodes because the voltage of TiS2 versus Li in the fully discharged condition (1.7 V) lies within the stability window of state-of-the-art electrolytes for lithium-ion cells. Because of reason (2), it is possible to use an electrolyte that contains high concentrations of low-freezing-temperature solvents that would be unsuitable for carbon anodes. One particularly attractive electrolyte turns out to consist of 1 M LiPF6 in a solvent that comprises ethylene carbonate (EC), diethyl carbonate (DEC), dimethyl carbonate (DMC), and ethyl methyl carbonate (EMC) in volume proportions of 3:5:4:1. This electrolyte was developed by carefully selecting proportions of low-freezing-point solvents to suppress the freezing point; in this case, the specific volume proportions were chosen because the resulting solvent mixture remains liquid down to a temperature of –42.5 °C — below the freezing temperatures of other candidate carbonate solvent mixtures that were also tested.

For experiments to quantify the performances of TiS2 and graphite anodes, half cells with these anodes and lithium counter/reference electrodes were constructed. The cells were tested electrochemically at room temperature (+20 °C) and at temperatures of –30 and –40 °C. The results of the tests (see figure) show that while the graphite anode material exhibits higher specific charge capacity at room temperature, TiS2 exhibits higher specific charge capacity at the lower test temperatures.

In another experiment, a full cell containing a TiS2 anode and an LiCoO2 cathode was constructed and demonstrated to operate at an average potential of about 1.8 V at room temperature. No information on low-temperature tests of this cell was available at the time of writing this article.

This work was done by Chen-Kuo Huang and Jeffrey Sakamoto of Caltech for NASA's Jet Propulsion Laboratory. For further information, access the Technical Support Package (TSP) free on-line at www.nasatech.com/tsp under the Materials category.