Phase Stability and Defect Behavior in Complex Thermoelectric Zinc-Antimonides

Citation

Pomrehn, Gregory Schoelerman (2013) Phase Stability and Defect Behavior in Complex Thermoelectric Zinc-Antimonides. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/0AWH-4P69. https://resolver.caltech.edu/CaltechTHESIS:01252013-135311552

Abstract

The Zn-Sb binary phase system has been of interest for many years in the search for efficient and low-cost thermoelectric materials. Of primary interest has been the Zn₄Sb₃ phase which exhibits a thermoelectric figure of merit, zT, in excess of 1 in an intermediate temperature range. In this study, Zn₄Sb₃ is shown to be entropically stabilized with respect to decomposition to Zn and ZnSb through the effects of configurational disorder and phonon free energy. Single-phase stability is predicted for a range of compositions and temperatures. Retrograde solubility of Zn is predicted on the two-phase boundary region between Zn₄Sb₃ and Zn. The complex temperature-dependent solubility can be used to explain the variety of nanoparticle formation observed in the system: formation of ZnSb on the Sb-rich side, Zn on the far Zn-rich side, and nano-void formation due to Zn precipitates being reabsorbed at lower temperatures.

A new binary compound, Zn₈Sb₇, known only in nanoparticulate form, is also studied using density functional calculations. The free energies of formation, including effects from vibrations and configurational disorder, are calculated to compare with the relevant phases ZnSb, Zn, and Zn₄Sb₃, yielding insight into the phase stability of Zn₈Sb₇. Band structure calculations predict Zn₈Sb₇, much like ZnSb and Zn₄Sb₃, to be an intermetallic semiconductor with similar thermoelectric properties. If sufficient entropy or surface energy exists to stabilize the bulk material, it would be stable in a limited temperature window at high temperature.

In the AZn₂Sb₂ series of materials—A = Ca, Sr, Yb, and Eu—I show that a large concentration of thermodynamically stable cation vacancies leads to high extrinsic carrier concentrations. The stable defect level depends on the choice of A, and is consistent with experimentally observed carrier concentrations in these materials. These results demonstrate that point defects are the primary mechanism by which the covalency of the cation bond can influence carrier concentration in nominally valence-precise AZn₂Sb₂compounds. This mechanism may be generally applicable to other Zintl phases, perhaps explaining similar trends seen in A₁₄MSb₁₁, A₂MSb₂ (A=2+ cation, M = 2+ or 3+ metal),and similar materials.

Thesis Files