Prof Neophytou described his method to determine if nanostructures can improve Seebeck Coefficient. In the simple model of electron filtering there is actually no improvement. depending on lambda_E there can be a modest improvement. need quantum reflections not tunneling. Avoid ionized impurity scattering. dislocation networks can increase power factor.
Teng Fang described computational efforts to find new p-type Half Heusler by looking for compounds with low effective mass. Calculations confirmed that 5d have lower effective mass than 4d transition metals. multiband effects are also calculated. Predicting RuTaSb as good TE materials.
H.Takaki TE transport properties (thermal and electrical) of Magnetic Semiconductor CuFeS2 is calculated. Fe doping lowers lattice thermal conductivity. lattice thermal conductivity is larger along a-axis than c=axis. predicts more dilute doping would make better zT.