Silicon nanowires have been investigated as alternatives to the graphite heretofore widely used as an anode material in rechargeable lithium-ion power cells. The theoretical specific charge capacity of graphite, corresponding to the maximum Li content (at a composition of LiC6) is 372 mA•hr/g. In contrast, the theoretical specific charge capacity of Si corresponding to the maximum Li content (at a composition of Li4.4Si) is much greater — 4.2 A•hr/g. In previous studies in which thin films of silicon on substrates were investigated for use as high-capacity anodes, it was found that charge capacities faded rapidly in charge/discharge cycling because large changes in volume (as much as 310 percent) associated with insertion and extraction of lithium ions caused cracking and crumbling of the films, as well as delamination of the films from the substrates. The basic idea of the present nanowire approach is to disperse the active anode material into regions having small volumes in order to reduce the sizes of the changes in volume in order to reduce the adverse effects of those changes.
In the investigation, silicon nanowires were synthesized by a variety of processes from a variety of starting materials. For example, in one experiment, a 5-nm-thick layer of nickel was sputtered onto a p-doped (111)-oriented silicon wafer to serve as a catalyst for growth of silicon nanowires, then the nickel-coated silicon substrate was heated to a temperature of 990 °C in a flowing atmosphere of argon containing 10 percent of water vapor for two hours. This process resulted in the formation of silicon nanowires having diameters from 10 to 50 nm (see figure). The nanowires were dispersed in ethanol by use of sonication, then anodes were fabricated by spraying the suspension onto copper foils that had been cleaned by etching in sulfuric acid, rinsed in deionized water, and dried. Fabrication was completed by drying the anodes in a vacuum oven at 120° for 12 hours.
The specific capacity of the silicon nanowires was measured by charge-discharge testing of one of the anodes at a rate of 0.25 C (where C is a constant electric current that, if integrated for one hour, would amount to the nominal charge capacity). The results of these measurements showed a large irreversible decrease in specific capacity upon the first insertion (charge) followed by smaller irreversible decreases in specific capacity on subsequent cycles (see Figure 2). This behavior has tentatively been attributed to the presence of oxygen on the surfaces of the silicon nanowires: A stable Li2O phase that forms upon exposure to oxygen may be responsible for the high initial irreversible decrease in specific capacity. On the other hand, Li2O is believed to have merit in alleviating the adverse effects of the change in volume as a means of binding silicon particles during cycling, and, hence, may be responsible for limiting the loss of capacity after the first cycle.
Coating the silicon nanowires with carbon has been proposed as a way of reducing the loss of capacity with cycling. Beneficial effects of coating with carbon include stabilization of the surfaces of the silicon nanowires by blocking oxygen, increasing electrical conductivity, and providing a cushion material to relieve the stress of volume expansion. An investigation of the effects of coating with carbon was in progress at the time of reporting the information for this article.
This work was done by KunHong Lee of Pohang University of Science and Technology for the Air Force Research Laboratory.
This Brief includes a Technical Support Package (TSP).
Silicon Nanowires for Anodes of Rechargeable Li Power Cells
(reference AFRL-0078) is currently available for download from the TSP library.
Don't have an account? Sign up here.