Sr/Eu cation doping and thermoelectric properties of the compound Ca<sub>9</sub>Zn<sub>4.5</sub><sub>−</sub><sub><italic>δ</italic></sub>Sb<sub>9</sub> (0<<italic>δ</italic><0.5)

Abstract

Previous studies suggested that Zintl phases related to the Ca9Zn4.5− δ Sb9 (0 δ x Cu x Sb9 indicated that they could achieve an optimized zT around 0.7. Especially for Ca9Zn4.5− δ Sb9, a very recent report suggested that a high zT of 1.1 at 875 K could be obtained in combination of various secondary phases introduced by the variation of Zn contents, which provided a chemical potential to change the composition. However, to change the composition of Ca9Zn4.5− δ Sb9 is actually very difficult to implement owing to the narrow homogeneity range. In this work, with Ca substituted by big cations such as Eu2+ and Sr2+, the unit cell of Ca9Zn4.5− δ Sb9 was expanded, and similar significant enhancement on the thermoelectric performance was resulted. All syntheses were handled inside a glovebox, then the reactants were loaded in the Nb tube and arc-welded, which were enclosed in a fused silica tube. The reactions were firstly heated to 1173 K in 6 h and then maintained at this temperature for 24 h. After the homogeneity process, the furnace was slowly cooled down to 873 K at a rate of 6 K/h, followed by another dwelling for 6 hours and finally cooled down to 573 K at a rate of 10 K/h. At last, the furnace was shut down and the reactions were opened in the glovebox. The compound were fully characterized by single-crystal X-ray diffraction (SXRD) and powder X-ray diffraction(PXRD), Energy dispersive spectrometer (EDS) and Hall effect measurement, etc. Single-crystal X-ray diffraction (SXRD) proves that with the increase of Sr component, the crystal parameters are slightly increasing. Hall effect measurement show that the substitution of Ca by Eu or Sr can both lead to an obvious decreasing on the carrier (holes) concentration, and the resultant mobility for Sr/Eu-doped materials is very similar if compared by the same doping level. It is also noted that neither the carrier concentration nor the mobility seems to vary linearly with the Eu/Sr-doping contents, which means a saturated state will be eventually approached. This phenomenon is consistent with the speculation above that the changes on the properties should predominantly originate from the change of interstitial Zn content, which the size effect dominated. While the decreasing carrier concentrations as well as the increasing mobility and lattice thermal conductivity all point to a structure modification, the crystal structures of these doped materials were systematically verified through the single crystal X-ray diffraction. In conclusion, although a direct measurement on such a small composition variation is very difficult (the X-ray diffraction results prove it), the significantly reduced carrier concentration and increased mobility as well as lattice thermal conductivity can still provide some useful hints on understanding the property optimization of these materials. For material Ca8.2Eu0.8Zn4.5− δ Sb9, high figure of merit with ZT ~0.81 has been achieved at 873 K.