Rietveld Refinement Of In-doped β-Zn₄Sb₃ Crystal Structure

is a promising thermoelectric material in the moderate temperature range. However, recent researches revealed that both the structural parameters for the three-interstitial model and the mechanism of low thermal conductivity ofβ-Zn4Sb3 compound are under debate. Meanwhile, most studies of In-dopedβ-Zn4Sb3 focused on the effect of In substitution for Zn on the thermoelectric properties, while In substitution for Sb has not been investigated. In this thesis, three series of In-dopedβ-Zn4Sb3 compounds with nominal composition of Zn4-mInmSb3, Zn4Sb3-mInm, and Zn4Sb3lnm were investigated, which have In doping of Zn site, Sb site, and interstitial site, respectively. Emphasis was placed on the effects of different In doping on the crystal structure ofβ-Zn4Sb3 according to Rietveld refinement results of powder X-ray diffraction data of three series of In-dopedβ-Zn4Sb3 compounds based on the three-interstitial model.For Zn4-mInmSb3 compounds with In substitution for Zn, In impurity preferentially occupied Zn(2) and Zn(3) interstitial sites and the solid-solution limit of In was about 4.07 at%, namely m≈0.163. As the m increased in the range of 0-0.16, the lattice parameters a, c, and V of Zn4-mInmSb3 linearly increased, the fractional coordinate x, y, and z of 36f Zn site gradually increased, and the fractional coordinate x of 18e Sb(1) site and z of 12c Sb(2) site linearly increased. The occupancy of In while that of Zn decreased at the interstitial site, the total occupancy at interstitial site slightly decreased from 0.1724 to 0.1698. However, the occupancy of Zn at 36f Zn site, and that of Sb at 18e Sb(1) and 12c Sb(2) sites were nearly unchanged. The substitution of In for Zn at interstitial sites led to the shorten of bond length of Sb-Sb dimer and caused the Zn at framework site slightly deviate from Sb(2) and get close to Sb(1).For Zn4Sb3-mInm compounds with In substitution for Sb, In impurity preferentially occupied 12c Sb site and the solid-solution limit of In at 12c Sb site was about 3.00 at%, namely m≈0.09. As the m increased in the range of 0-0.09, the lattice paremeters a and V of Zn4Sb3-mInm linearly increased while c linearly decreased, the fractional coordinate x, y of 36f Zn site and x of 18e Sb(1) linearly increased, and the fractional coordinates z of both 36f Zn and 12c Sb(2) sites decreased. The occupancy of In increased while that of Sb decreased at 12c Sb(2) site. However, the total occupancy at 12c Sb(2) site and the occupancy of Sb at 18e Sb(1) site were nearly unchanged. The total occupancy of Zn at interstitial sites increased slightly from 0.1785 to 0.2017. The substitution of In for Sb at 12c Sb(2) sites led to the elongate of bond length of Sb-Sb dimer, and caused the Zn at framework site slightly deviate from Sb(1) and get close to Sb(2).For Zn4Sb3Inmcompounds with In doping at interstitial site, In impurity preferentially occupied the interstitial site and the doping limit of In was about 0.54 at%, namely m≈0.038. With increasing the m in the range of 0-0.04, the lattice of parameters a, c and V linearly increased, the fractional coordinate x, y, and z of 36f Zn site gradually increased, and the fractional coordinate x of 18e Sb(1) site and z of 12c Sb(2) site linearly increased. The occupancy of In increased at interstitial site, leading to slightly decrease of the total occupancy from 0.1833 to 0.1970 at interstitial site. However, the occupancy of Zn at 36f Zn site, and that of Sb at 18e Sb(1) and 12c Sb(2) sites were nearly unchanged. The In doping at interstitial site led to the slight elongate of bond length of Sb-Sb dimer.