As you are talking about the reduction of thermal conductivity. Basically most of thermoelectric scientists are working on nanocomposites and doping of rattler atoms in the matrix to increase the phonon scattering. Are you doing the same ?
It has been well known that the figure of merit ZT = ZT=σS2T/ κ determines the effectiveness of the thermoelectricity. In principle the electrical conductivity should be high and the thermal conducvitiy should be low. In metals where the conducivity is determined by the electrons the two are usually proportional. Additionally the Seebeck coefficient goes through a maxiumum as a function of the charge carrier concentration. That is why most research has been done on doped semiconductor materials (such as Bi-Te, Pb-Te and the like). Heusler alloys are aos popular. The problem with Bi-Te based systems is high toxicity and the extreme rarity of Te in the Earth crust. I did not follow the the state of the art recently, but if I remember well nano-composites with low dimensionality (esepecially nanowires) proved to be promising. I hope that real experts in this field can give you more precise answer.
Nanomaterials describe, in principle, materials of which a single unit is sized (in at least one dimension) between 1 to 1000 nanometres (10−9 meter) but usually is 1 to 100 nm. Nanomaterials research takes a materials science-based approach to nanotechnology, leveraging advances in materials metrology and synthesis which have been developed in support of microfabrication research. Materials with structure at the nanoscale often have unique optical, electronic, or mechanical properties.
Nanomaterials are slowly becoming commercialized and beginning to emerge as commodities. Kindly go through these links:-
It has been well known that the figure of merit ZT = ZT=σS2T/ κ determines the effectiveness of the thermoelectricity. In principle the electrical conductivity should be high and the thermal conducvitiy should be low. In metals where the conducivity is determined by the electrons the two are usually proportional. Additionally the Seebeck coefficient goes through a maxiumum as a function of the charge carrier concentration. That is why most research has been done on doped semiconductor materials (such as Bi-Te, Pb-Te and the like). Heusler alloys are aos popular. The problem with Bi-Te based systems is high toxicity and the extreme rarity of Te in the Earth crust. I did not follow the the state of the art recently, but if I remember well nano-composites with low dimensionality (esepecially nanowires) proved to be promising. I hope that real experts in this field can give you more precise answer.