The battery companies are discussing maximum power draw, which is measured in watts, not energy storage, which is measured in joules or watt hours (1 Wh = 3600 Ws = 3600 J). A standard A123 cell has 2.3 amp hours at 3.3 volts, for an energy content of 7.6 watt hours. This comes from a package with a mass of 70 grams. Allowing a 10% mass increase for wires and connections (which is highly optimistic), that brings us close to 76 g / 7.6 Wh = 10 kg/kWh.

The ability to move power in and out of the A123 batteries quickly is really mind-boggling. You could ride for 120 miles on a 17 kWh pack in our benchmark design, or you could expend it all in one two minute, 510 kW (682 hp) burst! That is why drag racing is a natural sport for electric vehicles.

You mentioned “each kWh would have about 10 kg mass… “. A follow up question; looking at the chart on http://www.a123systems.com/html/technology.html# and in this paper http://www.altairnano.com/documents/altair_anoder_way.pdf, it seems that the nano-li-ion startups are claiming 3kw to 4kw/kg versus “Conventional lithium ion batteries result in power levels of 1kW per kilogram” Does this jive with your 1kWh=10KG rule of thumb? I note they refer to kW and not kWh. I understand the difference between kilowatt and kilowatt hours but, for the intended purposes, it their metric would seem be of little use unless they meant the same application as your metric.

If they do mean 3+kWh/kg, if their numbers are to be believed, and if the technology is capable of efficient mass production, then we seem to be closer to battery packs producing the same power and range per KG as we get from gas.