I ended up not doing this because the Seebeck effect isn't really a useful power-plant (in the broad sense of the word) when you have a small temperature gradient because it is really inefficient, although it is useful for thermometers and other low power items, and for some exotic things like Radioisotope Thermoelectric Generators, which is what powers Voyager and other long distance spacecraft. It becomes worth using when you have a large, consistent energy source, which is usually geothermal energy. But the small temperature difference between the human body and the flashlight and the low efficiency of thermocouples means that this project won't seriously compete with battery powered flashlights. However, it is a cool idea and could have some niche uses.