(604b) Novel Design Routes for High Performance Hybrid Thermoelectric Nanocomposites

Authors: 
Sahu, A., New York University
Russ, B., University of California at Berkeley
Coates, N., University of California at Berkeley
Forster, J., Lawrence Berkeley National Laboratory
Su, N., University of California, Berkeley
Urban, J., Lawrence Berkeley National Laboratory
Segalman, R., UCSB

Hybrid materials can be broadly defined as a class of composites with organic or biological and inorganic components intimately mixed with one another where atleast one of the components has dimensions ranging from a few angstroms to several nanometers. In contrast to mixtures whose properties can often accurately be described as an arithmetic average of the properties of their components, the distinctive feature of a truly hybrid material is that material synergies lead to performance that is greater than the sum of its parts, which can happen when there are strong, non-linear interactions between the constituent components and the role of their interfaces become predominant. Hybrid inorganic–organic materials offer a greater degree of tunability than single component materials due to the considerable structural and electronic diversity of the available building blocks. Additionally, one can easily control the relative volume fraction of constituents in the hybrid, the shapes/structures of the components (e.g., nanoscale inorganic semiconductor spheres or cubes in a polymer matrix), as well as the electronic/bonding interactions between components. While extensively used for fuel cells, photovoltaics, membranes and separation devices, smart coatings, catalysts, sensors etc., very few hybrid organic inorganic materials exist for thermoelectric applications. Here, we demonstrate a novel approach to create a myriad of thermoelectric materials comprising of a conductive polymer and a range of different inorganic components spanning from nanodots to nanoplatelets to nanowires. Judicious tuning of the electronic nature of the components allows us to obtain both n-type and p-type composites. By varying the loading of the constituents, we observe interesting non-monotonic electrical conductivity at intermediate loadings suggesting non-effective medium behavior. The surprising electrical conductivity behavior can be explained with a model where carrier transport is primarily through a highly conductive volume of polymer that exists at the nanoparticle-polymer interface. Additionally, by doping the individual components separately, we can boost the electrical conductivities even further and obtain higher power factors. Thus, a mix of traditional doping mechanisms and innovative interface engineering at the nanoscale allows us to generate high performance thermoelectric materials.