The first thing to remember is, this is a battery replacement and not a cure all for energy generation. Batteries are horribly inefficient in storing energy and thats on top of the whole burning hydrocarbons, converting it to electricity transporting it to the power socket and then converting from AC to DC. You end up using close to 2-4% of the energy you start with.
What they have, from looking at Articles in IEEE and Journal of Applied Physics is a simple chip which uses a phenomenon different than the Peltier/Seebeck effect, as some one else mention thermionic emission, to generate a voltage. Also reading the articles, the efficiency of a single device is pretty horrible around 3-4%. Now, however if they make some sort of multi-segmented device (or holding out something from the papers), that might raise up to 5-10% which puts it in range of regular Seebeck thermoelectrics, but in probably a much more compact and and therefore cheaper setup. The papers talk about using InSb, HgTeCd, PbTe, andInGaSbAs.
If that's the case, you can probably make some pretty good headway in usable energy density compared to a battery, which
Also, again, we're talking about some sort of fuel cell not an open flame and the area of which will probably be very small and probably get as hot as a regular Li-Ion batter would.
Slashdot Mirror
The first thing to remember is, this is a battery replacement and not a cure all for energy generation. Batteries are horribly inefficient in storing energy and thats on top of the whole burning hydrocarbons, converting it to electricity transporting it to the power socket and then converting from AC to DC. You end up using close to 2-4% of the energy you start with. What they have, from looking at Articles in IEEE and Journal of Applied Physics is a simple chip which uses a phenomenon different than the Peltier/Seebeck effect, as some one else mention thermionic emission, to generate a voltage. Also reading the articles, the efficiency of a single device is pretty horrible around 3-4%. Now, however if they make some sort of multi-segmented device (or holding out something from the papers), that might raise up to 5-10% which puts it in range of regular Seebeck thermoelectrics, but in probably a much more compact and and therefore cheaper setup. The papers talk about using InSb, HgTeCd, PbTe, andInGaSbAs. If that's the case, you can probably make some pretty good headway in usable energy density compared to a battery, which Also, again, we're talking about some sort of fuel cell not an open flame and the area of which will probably be very small and probably get as hot as a regular Li-Ion batter would.