Evolution of Thermoelectric Properties in the Triple Cation Zintl Phase: Yb_13–xCa_xBaMgSb₁₁ (x = 1–6)

Abstract

The band structure of Yb14MgSb11 is tuned by substituting the more earth-abundant cations, Ca and Ba, on the four crystallographically distinct Yb sites (Yb13–x Ca x BaMgSb11 (x = 1, 2, 3, 4, 5, 6)). Single crystals of composition Yb9.7(2)Ca3.85(5)Ba0.29(4)Mg1.13(3)Sb11.0(1) were grown from Sn flux revealing the cation site preferences. Magnetic measurements on this crystal show paramagnetic behavior consistent with the presence of ∼0.85 Yb3+. High-purity samples (>98%) with compositions close to nominal of Yb13–x Ca x BaMgSb11 (x = 1–6) were prepared by ball milling and spark plasma sintering. The carrier concentration can be rationalized with the presence of Yb3+ for all samples and decreases as a function of x in a systematic fashion at room temperature and increases above ∼600 K for x = 3–6. The temperature dependence of the carrier concentration can be understood considering the electronic structure with a light and heavy band valence band contributing to the properties and suggests the involvement of a localized flat band or impurity state that is active with increasing amounts of Ca. The effect of temperature leads to sustained high Seebeck coefficients with low electrical resistivity arising from the transitioning of the light to heavy band with localization of carriers in the flat band or impurity state for Ca-rich compositions. Speed of sound measurements show that the lattice stiffens with increasing x. Despite the stiffening lattice, the thermal conductivity decreases until x = 3, 4 at which point it increases slightly. The x = 4 sample reaches a peak figure of merit (zT) of 1.32 at 1273 K while being 16% lighter by the molar mass compared to Yb14MnSb11 thereby providing a more power dense material.

Evolution of Thermoelectric and Oxidation Properties in Lu-Substituted Yb14MnSb11

“…9 After this point phonon scattering dominates, and there is a decrease in the thermal conductivity to a minimum of 7.7 mW cm −1 K −1 at 1073 K. As the temperature increases, there is an upturn in the thermal conductivity which corresponds to the onset of bipolar conduction and an increase electronic contribution due to the activation of minority carriers. 9,11,24,25 This agrees well with previously reported thermal conductivities of polycrystalline samples of Yb 14 MnSb 11 made from both elemental and binary precursors. 11,12 The temperature-dependent behavior of this sample (x = 0.9) is distinct and likely a result of increased contributions from the larger amounts of LuSb impurity seen in both PXRD and SEM.…”

“…The thermal conductivity of the sample of Yb 14 MnSb 11 begins at 8.4 mW cm –1 K –1 at room temperature and increases to a maximum of 8.9 mW cm –1 K –1 at 573 K due to an increase in electronic contribution as carriers move between the bands of the complex structure . After this point phonon scattering dominates, and there is a decrease in the thermal conductivity to a minimum of 7.7 mW cm –1 K –1 at 1073 K. As the temperature increases, there is an upturn in the thermal conductivity which corresponds to the onset of bipolar conduction and an increase electronic contribution due to the activation of minority carriers. ,,, This agrees well with previously reported thermal conductivities of polycrystalline samples of Yb 14 MnSb 11 made from both elemental and binary precursors. , The temperature-dependent behavior of this sample ( x = 0.9) is distinct and likely a result of increased contributions from the larger amounts of LuSb impurity seen in both PXRD and SEM. LuSb is also seen in the PXRD and SEM for x = 0.7 at lesser amounts and may explain the trend of increasing thermal conductivity at x = 0.7 and 0.9.…”

“…The density of the disks was determined using the Archimedes method with toluene as the liquid. The heat capacities of the samples of Yb 14– x Lu x MnSb 11 were estimated based on the heat capacity of Yb 14 MnSb 11 which was then modified for the molecular weight of the doped compound where C p (Yb 14– x Lu x MnSb 11 ) = C p (Yb 14 MnSb 11 ) × MW­(Yb 14– x Lu x MnSb 11 )/MW­(Yb 14 MnSb 11 ). ,,− Here, MW stands for the molecular weight of the respective compounds. The thermal expansion of Yb 14 MnSb 11 was used to estimate the temperature dependence of density in the Lu-substituted compounds. ,, These were all combined to give the total thermal conductivity according to the equation κ = D × C p × d ( D = measured diffusivity; C p = heat capacity adjusted for molar mass; d = the temperature-adjusted density from Yb 14 MnSb 11 ).…”

2 + 2 = 3: Making Ternary Phases through a Binary Approach

“…An improved synthetic route is shown to allow for efficient synthesis of stoichiometric and extremely high-purity products. Through this synthesis, a sample of Yb 14 MgSb 11 was produced, which exhibited the material leading efficiencies of zT = 1.28 at 1175 K. The slight improvement to overall zT increased the zT avg of Yb 14 MgSb 11 in the temperature window 875 to 1275 K from 1.10 to 1.15, making this phase competitive with Yb 14 MnSb 11 (1.07). ,, The increase in average zT is important when considering these materials for incorporation into high-temperature thermoelectric generators, such as a radioisotope thermoelectric generator (RTG). The impact of the experimental method in acquiring the Seebeck coefficient was also investigated and showed that while the commercial four-probe experimental setup provides reliable electrical resistivity; Seebeck coefficients are exaggerated due to a cold-finger effect at high temperatures.…”

Influence of Cu insertion on the thermoelectric properties of the quaternary cluster compounds Cu3M2Mo15Se19 (M = In, K) and Cu4In2Mo15Se19

“…Furthermore, because the conversion efficiency and output power of thermoelectric generators depends on the temperature difference across the thermoelectric compounds, 5–8 identifying novel materials with high ZT values and high melting points is also of particular interest. However, to date, only a handful of compounds are able to operate above 1000 K; a restricted list that includes the state-of-the-art Si–Ge alloys, 9,10 Zintl phases based on Yb 14 MnSb 11 and RE 3− x Te 4 (RE = La, Nd, Pr, Gd), 11–17 and half-Heusler alloys. 18–20…”