Synthesis, crystal chemistry, and properties of skutterudites

Skutterudites are a class of materials that have been studied for a number of years, and are considered to be promising for thermoelectric applications. Previous research has shown that it is possible to synthesize metastable skutterudites using low-temperature synthetic methods. In this dissertation we examine the extended compositional ranges over which several skutterudites will form using the ultra-thin layer, elemental deposition method, and explore the effects of composition on the material's properties. We have found that the amount of filler atoms in filled skutterudites, the range of possible metal to pnictide ratios, and how the properties respond depend on the system. For example, Co-rich samples of CoSb3 were n-type, and the number of charge carriers decreased with annealing. Samples that had the ideal stoichiometry were n-type, but the charge carriers were relatively stable. Sb-rich samples started n-type, and became p-type with annealing. This indicates how sensitive the properties are to changes in composition.; We have also discovered the presence of secondary phases in samples that appear single phase when analyzed using X-ray diffraction. We used Rietveld refinement and quantitative analysis on the samples to determine the percent crystallinity. Some of them were only 5% crystalline, whereas others were nearly 100% crystalline, depending on the system and the composition. The amount of crystallinity was larger in the more stable samples and smaller in the less stable ones. We used several methods to confirm the crystallinity and the presence of secondary phases. In the case of the CexCo 4Sb12 samples, magnetic measurements confirmed the presence of a secondary phase that contributed a ferromagnetic signal. Mossbauer spectroscopy on the FeSb3 samples revealed that the majority phase had the chemical environment of FeSb2. Finally, we used electron back-scatter diffraction to image these samples and gain insight into the mechanism by which they crystallize. EBSD data indicated that the growing grains exclude elements that are not soluble in the crystal. We presume that crystal growth stops because the grain boundaries have become depleted in the elements required to allow continued growth in the crystal grains.