In the future, predominantly solar or wind power systems will likely require energy storage for days to approximately a week, with or without connections to the 2 electric grid. If so, conversion of electricity to chemical energy is potentially attractive, since chemicals are inexpensive to store and turnaround efficiency is less critical for storage periods of a week or more. The most attractive chemical for this process is likely to be hydrogen (H2) generated locally by electrolysis of water using intermittent excess solar or wind power. Later, when combined with air or oxygen (O2) in engines or fuel cells, H2 can regenerate electricity on demand. Although H2 electricity storage is less energy efficient (40-50%) than compressed air storage, H2 has far lower costs of storage capacity, since H2 is a chemical fuel and air is not. For example, a 250-liter pressure vessel designed to store 10-20 kWh of compressed air could store enough H2 to provide 150-300 kWh of electricity, reducing the cost of storage capacity by more than a factor of 10. H2 energy storage is therefore economically best suited to situations where the total amount of energy stored is more valuable than efficiency. This should be the case for electricity stored longer than 1-2 days. H2 may also be stored cheaply without high pressure as a very low temperature (-453 oF) liquid (LH2) or by absorption in powders of abundant metals (e.g. iron, titanium, aluminum, and sodium) that release H2 upon moderate heating (<200 oF).