Thermoelectric power factor in nanostructured materials with randomized nanoinclusions

We investigate the electric and thermoelectric transport coefficients of nanocomposites using the Non-Equilibrium Greens Function (NEGF) method, which can accurately capture the details of geometry and disorder in these structures. We consider here two dimensional (2D) channels with embedded nanoinclusions (NIs) modelled as potential barriers of cylindrical shape and height VB. We investigate the effect of randomness of the NIs on the thermoelectric power factor by varying the positions, diameter, and heights of the barriers according to a Gaussian probability distribution. We find that the power factor shows indications of tolerance to variations in the parameters of the NIs when the Fermi level is placed into the bands and VB approx. kBT. These results could be experimentally relevant in the design of nanocomposites for thermoelectric applications.