It sounds like it is a material that undergoes a phase change from a non-magnetic state to a magnetized state at a useful temperature. That is not direct conversion of heat to useful electricity though. Current can be induced in a conductor by exposing it to a changing magnetic field such as we do here with the wind turbines on this site. I would think that in order to generate electricity with such a material as this, one would have to have a method of cycling the material from a temperature below the phase change point, to one above and back again repeatedly, thereby creating a changing magnetic field that would be useful for inducing a current in a coil exposed to that changing field. The phase change is powered by the random kinetic energy of the individual molecules in its structure as the material warms, and the energy drawn off as electrical energy would only be a small portion of what is contained in the waste heat. Because the temperature of the material would rise and drop more rapidly with a larger temperature change, everything else being equal, the operation of a theoretical generator based on this idea would be more practical if the temperature difference between the hot and cool sides were larger. This is just a different concept of a heat engine, in general, powered by the mixing of warmer and cooler temperatures to a condition with higher entropy. Although the mechanics are different, potentially allowing for different applications, the rules governing how much power could be obtained with 100% efficiency would not be different than for a sterling engine I would think. Rich