Institute of Physics has made progress in the study of thermoelectric transport regulation of semi-Hex alloy materials

Figure 1: Schematic synthesis of a series of semi-Hösler alloy (HH) thermoelectric materials based on the displacement reaction of metal lithium or magnesium with transition metal halides and the vacuum assisted plasma sintering (SPS) process.

Figure 2: Transmission electron microscopy (TEM) photographs show the presence of higher-density dislocation arrays at the grain boundaries of HH materials: (a), (b), (c), and (e) show grain boundaries of Nb0.8Ti0.2FeSb materials, LMTEM and HRTEM, respectively. Twisted-type dislocations; (d), (f) Inverse Fast Fourier Transform (IFFT) images corresponding to (c) and (e), respectively; (g)-(l) Display There is a transitional dislocation array in the grain boundaries of Hf0.25Zr0.75NiSn0.97Sb0.03.

Figure 3: Thermal conductivity vs. temperature curves for typical HH materials: (a) total thermal conductivity κtot; (b) electronic thermal conductivity κe; (c) and (d) are lattice thermal conductivities with bipolar effects (κL+κbip); (e) The theoretically simulated phonon relaxation time versus frequency for a sample of NTFC-20min; (f) The theoretically simulated phonon relaxation time versus frequency for a sample of HZNSS-20min; (g) Comparison of the theoretical thermal conductivity and temperature curves of the lattice thermal conductivity for all Nb0.8Ti0.2FeSb and Hf0.25Zr0.75NiSn0.97Sb0.03 samples.

Figure 4: Analysis of electrical properties of all samples: (a) sample resistivity ρ(T); (b) sample Seebeck coefficient S(T); (c) sample carrier concentration nH; (d) sample Hall mobility μH; (e) and (f) are the theoretical simulations of the Hall mobility versus carrier concentration for the samples NTFC-20min and HZNSS-20min at room temperature. The experimental values ​​are plotted in the figure for comparison; (g) All samples experimental power factor PF The relationship with temperature; (h) Theoretical simulation of PF and carrier concentration at room temperature. The experimental values ​​are plotted in the figure for comparison.

Figure 5: Engineering thermoelectric properties of all HH samples: (a) and (b) are the thermoelectric values ​​of the Nb0.8Ti0.2FeSb and Hf0.25Zr0.75NiSn0.97Sb0.03 series samples, respectively. Values ​​vs. carrier concentration; (c), (d), (e), (f) show (ZT) eng, η, (PF) eng, and Pd for all samples when the cold end of the sample is 50°C. With the sample hot end temperature changes in the curve.

Thermoelectric technology can realize the direct mutual conversion of thermal energy and electric energy. It also has the advantages of small size, no vibration and noise, long service life, and environmental friendliness. It has unique advantages in waste heat power generation and cooling, and it has caused worldwide clean energy. Wide attention in the field. The conversion efficiency of the thermoelectric device is precisely determined by the engineering thermoelectric properties of the material, wherein the energy conversion efficiency η depends on the engineering thermoelectric value (ZT) eng value of the thermoelectric material, which is defined as: (ZT)eng=\ (\frac{\left ( \int_{T_{c}}^{{T_{h}}S\left (T ight )\mathrm{d}T ight )^{2}}{\int_{T_{c} }^{T_{h}}ho \left (T ight )\mathrm{d}T\int_{T_{c}}^{T_{h}}\kappa\left (T ight )\mathrm{d}T }\) ΔT=\(\frac{\left ( PF ight )_{eng}}{\int_{T_{c}}^{T_{h}}\kappa\left (T ight )\mathrm{d} T} \) ΔT. The Seebeek coefficient S(T) and resistivity ρ(T) are collectively referred to as the electrical properties of the material and describe the electrical transport properties of the material, which are closely related to the type, concentration, and mobility of the material carriers and the electronic structure of the material. Determine three parameters of thermoelectric material (ZT) eng: S(T) coefficient, resistivity ρ(T) and thermal conductivity Ƙ(T) (consisting of electron thermal conductivity Ƙe and lattice thermal conductivity ƘL) through the carrier The mutual coupling restriction of transportation, the optimization adjustment of the performance of a single parameter usually causes the deterioration or degeneration of the other two properties, which does not contribute to the improvement of the overall thermoelectric performance. Therefore, how to achieve the decoupling between various transport parameters of thermoelectric properties, in particular to make full use of various phonon scattering mechanisms to reduce the lattice thermal conductivity ƘL of the material without damaging even the electrical input of the enhanced material The operational performance has always been the hotspot and key in the field of thermoelectricity research.

The half-Heusler HH alloy thermoelectric material is an excellent material system operating in a medium-high temperature zone (300-700 °C). It not only has a high value of thermoelectricity (in which NbFeSb-based material has a peak ZT of 1.7), but also has excellent electrical transport properties (the thermoelectric power factor PF up to 106×10-4 W m-1 K-2). In particular, this type of material system has many chemical stability and thermal stability, excellent mechanical properties, and is an ideal thermoelectric power generation material. However, the disadvantage is that most of the HH lattice thermal conductivity (room temperature ~ 10 W m-1 K-1) is significantly higher than other thermoelectric material systems, such as: Bi2Te3, PbTe and MgAgSb and so on. Nanocrystalline materials and alloying means at different atomic positions can significantly reduce the lattice thermal conductivity of HH materials. However, there is still a large gap between the theoretical minimum thermal conductivity Ƙmin~1 W m-1 K-1 of HH materials.

In the HH material system, targeted scattering of medium and low-frequency phonons (≤40 THz) is targeted by the introduction of new types of structural defects, in addition to the use of nano-grain boundary enhancement phonon scattering and alloying methods to enhance high frequencies (≥ 40THz) necessary means other than phonon scattering. High-density grain boundary dislocations belong to this defect type, and the use of high-density grain boundary dislocations to improve the mechanical properties of alloy materials has been a long time. Recently, the introduction of dislocation engineering in the Bi2-xSbxTe3 system has also been proved to improve the material thermoelectricity. The effective means of performance. For HH, a multi-element material system, it is not easy to realize high-density grain boundary dislocations due to the complicated relationship between the phases and the dynamics.

Recently, Zhao Huaizhou, Research Associate, Institute of Physics, Chinese Academy of Sciences/Beijing National Laboratory for Condensed Matter Physics (CIC), Thermoelectricity Team, and Chen Xiaolong and Gu Lin, Research Fellow, Institute of Physics, Ren Zhifeng, Professor, University of Houston, USA, and Northwestern University, USA Professor Jeffrey G. Snyder and others collaborated to propose a composite system of in-situ multi-metal nanoparticles and lithium or magnesium halides based on the substitution reaction of active metallic lithium or magnesium with other halides of main groups or transition elements through vacuum assisted conditions. The pulsed plasma hot press (SPS) technology has synthesized a variety of HH alloy materials with high density and purity. These materials are characterized by the presence of high-density dislocation arrays at grain boundaries within the material, as shown in the following figure. This mechanism of dislocation formation can be simply explained as follows: During SPS hot pressing, the presence of a liquid phase of a lithium halide enhances the diffusion scale of metal elements such as Hf0.25Zr0.75NiSn0.97Sb0.03 and Nb0.8Ti0.2FeSb. The speed is conducive to the mutual slipping and alignment between crystal grains. Crystal grains with similar crystal face index are easy to form low-angle grain boundaries and form semicoherent dislocation arrays, as shown by the Moore rings in the TEM image below. . This dislocation density can reach ~1×1011 cm-2, which has a significant effect on the thermoelectric transport properties of the material. The final study found that high-density dislocations can reduce the ƘL of N-type Hf0.25Zr0.75NiSn0.97Sb0.03 material at 900K to 1 W m-1 K-1, while ZT~1 and η (~11%) are also similar. One of the highest values ​​of the compound. For P-type Nb0.8Ti0.2FeSb material, its power factor reaches 47×10-4 W m-1 K-2, and η-7.5%, which is the highest value of similar components in the literature, as shown in the figure below. After further adjusting the composition to FeNb0.56V0.24Ti0.2Sb, η~11%.

The universal significance of this work is that it can not only extend this method to other famous thermoelectric material systems, but also has guiding significance for the compounds and alloy systems of some non-thermoelectric material systems. Related papers were recently published in Advanced Energy Materials (DOI: 10.1002/aenm.201700446).

The above work was supported by funding from the National Natural Science Foundation of China (51572287) and the Joint Fund Committee-Guangdong Joint Key Fund (U1601213).