Batteries have great promise for facilitating the grid integration of renewable energy and powering electric vehicles. One critical concern for the scale-up of battery production is the availability of elements used in battery couples. We provide the first systematic comparison of supply limits and extraction costs of the elements in battery couples against short- and long-term scaling goals. Several couples can scale well beyond short- and long-term grid-storage goals, including: Na/S, Zn/Cl-2,Cl- and FeCl2/CrCl3. Li-based couples currently have the performance characteristics most suitable for electric vehicles, yet scaling beyond 10MM vehicles per year will demand significant increases in Li production. We also provide a framework to evaluate new couples, such as those based on Mg, which may be an alternative to Li-based couples. While the extraction costs of the elements used in current battery couples are, in many cases, below 105$kWh(-1), the cost of finished battery cells is in the range of 150-1000$KWh(-1), well above cost targets of 100$kWh(-1) for both grid and transportation applications. Currently high costs remain a critical barrier to the widespread scale-up of battery energy storage. (C) 2010 Elsevier B.V. All rights reserved.