An electrochemical cell based on the lithium/sulfur couple is attractive as an electric vehicle (EV) power source because of its very high theoretical specific energy (2600 Wh/kg). A primary obstacle in producing a functional Li/S cell is the poor conductivity of sulfur. This difficulty led other researchers to utilize sulfur compounds instead of elemental sulfur, a choice which results in significantly lower theoretical specific energies. We are developing ambient-temperature solid-state Li/S cells using solid polymer electrolytes. The use of a solid polymer electrolyte which is ionically conductive below the melting point of sulfur alleviates some problems encountered by researchers studying high-temperature lithium-sulfur cells (e.g. Li/FeS2 cells). It also mitigates problems associated with the use of solid lithium metal in liquid electrolytes.
Sulfur electrodes were initially prepared from suspensions of elemental sulfur powder, carbon (graphite), polyethylene oxide (PEO), lithium trifluoromethanesulfonate (LiTf), and Brij 35 surfactant in acetonitrile. These suspensions were cast onto Teflon-coated plates to produce 100-200 micron thick films. Galvanostatic cycling of cells fabricated from these electrodes and PEO-LiTf electrolyte films showed good charge-discharge behavior at low currents (~10 microA/cm2), but resulted in poor utilization of the active material (of order 1-2%). Scanning electron microscopy (SEM) in conjunction with electron spectroscopy for chemical analysis (ESCA) revealed the presence of 10-micron sulfur islands in the as-prepared sulfur electrodes. Based on our electrochemical and morphological studies, we have formulated a preliminary phenomenological model of the sulfur electrode. The model postulates the creation of a Li2S reaction zone which is characterized by slightly higher ionic conductivity but equally poor electronic conductivity, compared to elemental sulfur. In addition, the volume changes that accompany the electrochemical reaction are incorporated into the model, which predicts a loss of available active material. This model makes apparent the need for the use of smaller sulfur particles in the electrode, and various fabrication methods were used to achieve this.
Further work on this system has resulted in the achievement of sulfur utilizations of nearly 100% at temperatures near 80C, and long cycle lives in excess of 600 charge-discahrge cycles at 25C.