Thermoelectric Stability of Cu2Se and Magnetic Phenomena in (Ti, Zr, Hf)NiSn Half-Heusler Alloys

Abstract

Vast amounts of heat are emitted by industrial factories, automobiles and geologic formations. To convert the abundance of heat into usable electricity, stable thermoelectric materials with high conversion efficiencies (or equivalently, large zT values) are essential. In this thesis, I first focus on a superionic conductor, Cu2Se, that is an excellent thermoelectric material, yet known to have dissociation issues in real application conditions. By doping the structure with Sn, I was able to demonstrate improved zT values and increased chemical stability. I describe the homebuilt current stress test that I implemented for emulating a high-temperature power generation setting, which has now become more common in thermoelectric studies of Cu2Se-based materials.

In a second project, I formed part of a team that showed that magnetic dopants can augment the thermoelectric conversion efficiencies in a half-Heusler alloy, (Ti, Zr, Hf)NiSn. Here, I detail my study of the low-temperature magnetic properties of the composite materials, including the cluster-glass behavior that is visible in the bifurcation of the zero-field-cooled and field-cooled temperature-dependent magnetic susceptibility. I applied the AC susceptibility technique to witness the evolution of the cluster-glass freezing temperatures as a function of driving frequency and strength of the applied magnetic field. The extensive magnetic characterization is important for the recent acceleration of research on magnetic compounds within the thermoelectrics community.

I further discovered that the same half-Heusler composites exhibit a strongly non-monotonic dependence of the heat capacity on applied magnetic field at temperatures below 10 K. In the final project of this thesis, I describe the fitting of multi-level paramagnon and volumetric magnon models to the heat capacity data. Using statistical analysis tools, I show that the two-level paramagnon effect on the heat capacity is the most significant, and could be due to trace impurities or defects within the highly disordered compounds. The Fe added samples exhibit a unique trend of the heat capacity with magnetic field at temperatures lower than 4 K. I derive a magnon model as a possible explanation, yet it overfits the data. In the end, I qualitatively attribute the lowest temperature effect in the Fe-added samples to a magnetotransport effect observed in the second project.

Overall, the thesis is primarily motivated by thermoelectrics research, which is seen most directly in the high-temperature thermoelectric property study of Cu2Se. The magnetic half-Heusler alloys that were shown to possess improved thermoelectric properties at high temperatures are then studied for their magnetic properties, magnetotransport and magnetothermodynamics at low temperatures as a way of discovering more fundamental aspects of the compounds.