Search

This work describes the continued examination of the low-temperature performance of the Li–CFx electrochemical couple in a −40°C or colder environment. Previously, the efficacy of subfluorinated CFx (SFCFx) cathode active materials was demonstrated; preliminary results indicated that the material was functional at rates up to C/10 at −40°C. However, there were often substantial voltage fluctuations during discharge, accompanied with inconsistent capacity yields and sometimes dramatic polarization events. In the research described herein, an investigation of various electrolyte and cathode compositions was conducted in an effort to optimize performance. In particular, several different electrolyte solvent formulations were examined using a salt content of either 1 or 0.5 M LiBF4. A further modification consisted of addition of an anion receptor to the electrolyte that was intended to be a LiF-solvating agent. The most promising electrolyte was tested with several different SFCFx compositions created using either graphitic or multiwall nanotube precursor materials. Electrochemical evaluation showed that the best SFCFx-based test cells were able to deliver specific capacity values up to five times greater than control cells (containing conventional CF1.08 powder) at −40°C under discharge currents as high as C/5 with composite electrodes thicker than 100 µm.

The temperature dependence of the open-circuit potential of lithium half-cells was measured for electrodes of carbon materials having different amounts of structural disorder. The entropy of lithium intercalation, DeltaS, and enthalpy of intercalation, DeltaH, were determined over a broad range of lithium concentrations. For the disordered carbons, DeltaS is small. For graphite, an initially large DeltaS decreases with lithium concentration, becomes negative, and then shows two plateaus associated with the formation of intercalation compounds. For all carbons DeltaH is negative, and decreases in magnitude with increased lithium concentration. For lithium concentrations less than x = 0.5 in LixC6, for the disordered carbons the magnitude of DeltaH is significantly more negative than for graphite (i.e., intercalation is more exothermic). The measurements of DeltaH provide an energy spectrum of chemical environments for lithium. This spectrum can be used to understand some of the concentration dependence of configurational entropy, but the negative values of DeltaS require another contribution to entropy, perhaps vibrational in origin.

Germanium nanocrystals (12 nm mean diam) and amorphous thin films (60-250 nm thick) were prepared as anodes for lithium secondary cells. Amorphous thin film electrodes prepared on planar nickel substrates showed stable capacities of 1700 mAh/g over 60 cycles. Germanium nanocrystals showed reversible gravimetric capacities of up to 1400 mAh/g with 60% capacity retention after 50 cycles. Both electrodes were found to be crystalline in the fully lithiated state. The enhanced capacity, rate capability (1000C), and cycle life of nanophase germanium over bulk crystalline germanium is attributed to the high surface area and short diffusion lengths of the active material and the absence of defects in nanophase materials.

Anode materials of nanostructured silicon have been prepared by physical vapor deposition and characterized using electrochemical methods. The electrodes were prepared in thin-film form as nanocrystalline particles (12 nm mean diameter) and as continuous amorphous thin films (100 nm thick). The nanocrystalline silicon exhibited specific capacities of around 1100 mAh/g with a 50% capacity retention after 50 cycles. The amorphous thin-film electrodes exhibited initial capacities of 3500 mAh/g with a stable capacity of 2000 mAh/g over 50 cycles. We suggest that the nanoscale dimensions of the silicon circumvents conventional mechanisms of mechanical deterioration, permitting good cycle life.

Lithium transition-metal oxides (LiTMO₂) are currently the preferred cathode materials for secondary Li ion batteries. Several of these materials have a layered structure suitable for intercalation compounds owing to their good kinetics and ability to withstand internal strains of Li intercalation. Despite extensive research on this system, it is still not understood how the electronic charge is redistributed during lithiation. Traditionally it has been accepted that the charge on the intercalated Li is accommodated by the transition metal, allowing the O valence to remain 2-. However, recent computational studies suggest that there is substantial charge transfer from the ionic Li to the anion [1,2]. Empirical data from x-ray photoemission spectroscopy (XPS) and x-ray-absorption spectroscopy (XAS) show that LiTMO₂ is strongly covalent with available O 2p holes lying in the TM 3d band [3]. These techniques also show a decrease in O 2p holes during Li intercalation indicating an increase in electron density around the O ion [3,4].