Thermoelectricity offers a sustainable path to recover and convert waste heat into readily available electric energy, and has been studied for more than two centuries. From the controversy between Galvani and Volta on the Animal Electricity, dating back to the end of the XVIII century and anticipating Seebeck’s observations, the understanding of the physical mechanisms evolved along with the development of the technology. In the XIX century Orsted clarified some of the earliest observations of the thermoelectric phenomenon and proposed the first thermoelectric pile, while it was only after the studies on thermodynamics by Thomson, and Rayleigh’s suggestion to exploit the Seebeck effect for power generation, that a diverse set of thermoelectric generators was developed. From such pioneering endeavors, technology evolved from massive, and sometimes unreliable, thermopiles to very reliable devices for sophisticated niche applications in the XX century, when Radioisotope Thermoelectric Generators for space missions and nuclear batteries for cardiac pacemakers were introduced. While some of the materials adopted to realize the first thermoelectric generators are still investigated nowadays, novel concepts and improved understanding of materials growth, processing, and characterization developed during the last 30 years have provided new avenues for the enhancement of the thermoelectric conversion efficiency, for example through nanostructuration, and favored the development of new classes of thermoelectric materials. With increasing demand for sustainable energy conversion technologies, the latter aspect has become crucial for developing thermoelectrics based on abundant and non-toxic materials, which can be processed at economically viable scales, tailored for different ranges of temperature. This includes high temperature applications where a substantial amount of waste energy can be retrieved, as well as room temperature applications where small and local temperature differences offer the possibility of energy scavenging, as in micro harvesters meant for distributed electronics such as sensor networks. While large scale applications have yet to make it to the market, the richness of available and emerging thermoelectric technologies presents a scenario where thermoelectrics is poised to contribute to a future of sustainable future energy harvesting and management. This work reviews the broad field of thermoelectrics. Progress in thermoelectrics and milestones that led to the current state-of-the-art are presented by adopting an historical footprint. The review begins with an historical excursus on the major steps in the history of thermoelectrics, from the very early discovery to present technology. Then, the most promising thermoelectric material classes are discussed one by one in dedicated sections and subsections, carefully highlighting the technological solutions on materials growth that have represented a turning point in the research on thermoelectrics. Finally, perspectives and the future of the technology are discussed in the framework of sustainability and environmental compatibility. An appendix on the theory of thermoelectric transport in the solid state reviews the transport theory in complex crystal structures and nanostructured materials.

/pdf/thermoelectrics-from-history-a-window-to-the-future-36gnx77fbv.pdf

Thermoelectrics: From history, a window to the future

Thermoelectric high ZT half-Heusler alloys Ti1−x−yZrxHfyNiSn (0 ≤ x ≤ 1; 0 ≤ y ≤ 1)

Thermoelectric materials are capable of converting heat and electricity to each other Thermoelectric devices can be miniaturized and highly integrated with existing semiconductor chip systems with microgenerators or microrefrigerators After years of research and accumulation, BiTe series, SnSe series, CuSe series, half-Heusler series, multicomponent oxides series, organic–inorganic composites series, and GeTe/PbTe series have been found to have excellent thermoelectric properties According to theoretical calculation, when the diameter of Bi2Te3 nanowires is 5 A, the ZT value reaches 14, and graphdiyne has a ZT value of 48 at 300 K Experimental measurements revealed that the ZT value of n-type SnSe reached 28 This review would focus on the updated experimental and theoretical achievements of seven kinds of materials, including BiTe series, SnSe series, CuSe series, multicomponent oxides, half-Heusler alloys, organic–inorganic composites, and GeTe/PbTe series The preparation method, microstructure characteristics, device structure, and thermoelectric properties of each material will be described in detail By analyzing the performance of these materials, three possible development directions are put forward for how to further improve the thermoelectric properties of materials

Review of current high-ZT thermoelectric materials

Thermoelectric properties of the TiX(Zr0.5Hf0.5)1 − XNiSn half-Heusler compounds

(V,Nb)-doped half Heusler alloys based on {Ti,Zr,Hf}NiSn with high ZT

Functionally graded materials (FGMs), which feature a continuous variation in microstructure and properties, have potential applications in engineering. Much interest has been shown in the processing methods that can be employed to produce such FGMs. One interesting method is that reported by T. Jungling et al. at the FGM’94 [1]. In this work, a compositional gradient was produced by the settling of particles under gravity in a vessel. Subsequently, J. Huang et al. and G. Kapelski et al. also used this method to produce SUS316/3Y-PSZ and Al2O3/C FGM with smoothly-varying composition, respectively [2, 3]. Based on these researches, another settling process has been proposed to form FGMs [4, 5], as shown in Fig. 1. The significant feature of the latter process is that an additional settling distance is added below the suspension. For convenience, the latter process is called the “process 2” and the former process the “process 1” in the following description. Until now few people have noticed that there are great differences between the above two processes in forming FGMs. In the present investigation, an attempt has been made to theoretically analyze the characteristics of the above two settling processes. Under the influence of gravity, particles in a dilute suspension often fall and stack at the bottom of a stationary vessel. The particle-depositing mass fraction is a function of time t and frequency mass distribution f (D) of the powder. It was given by Oden [6] as follows,

Theoretical study on two sedimentation processes used to form functionally graded materials

Thermoelectric properties of ZrNiSn-based half-Heusler compounds by solid state reaction method

ZrB2-60 mol%SiC composite with a eutectic microstructure was oxidized at 1573 to 1873 K with reduced total pressures (Ptot) and low oxygen partial pressures (PO2). The mass change was continuously measured by a thermobalance, and then fit with a multiple paralinear model. Oxidation scale of SiO2/ZrO2+SiO2/ZrO2/ZrB2 was formed at PO2 > 0.13 kPa, whereas only porous ZrO2 remained at PO2 < 0.13 kPa, Ptot < 1.33 kPa and higher than 1773 K. With increasing PO2, the parabolic oxidation constant decreased, whereas the linear oxidation constant increased.

Oxidation Behavior of ZrB2–SiC Composites at Low Pressures

Metal/ceramic type functionally gradient materials [1 ] (FGMs) are new-generation composite materials for future use in aerospaee, nuclear engineering, etc. The key problem for their practical use is how to ensure no damage is caused by the thermal stresses in the preparation and working processes. In analysing the residual and working stresses of ceramic/metal FGMs, Kawasaki [2], Kato [3], and Yuan et al. [4] have done some interesting work. However, work which optimizes the composition and structure by comprehensively surveying the residual and working stresses has not been done until now. In fact, these studies are very important to the thermal stress relaxation design of FGM. I n this paper, a TiC/Ni3A1 FGM is studie& As a ceramic phase, TiC has a high electronic expansion coefficient and high-temperamre strength; as a metal phase, the intermetallic Ni3A1 has a fine creepresistance at elevated temperature, and an excellent interface bondability with TiC [5, 6]. The thermomechanical properties of TiC/Ni3A1 composites with different TiC-Ni3A1 ratios are first tested. The tested properties are then used as a basis of axisyrnmetric finite element calculation for the two different types of stresses. For a disc-like TiC/Ni3A1 FGM, by surveying the stresses, the stress sites, and the stress state at the pure ceramic side, the second optimization for the FGM compositional distribution and structure is obtained. TiC/Ni3A1 composites with Ni3A1 volume fraction of 0, 20%, 40%, 60%, 80% and 100% were sintered by the HP method for 2 h at 1300 °C and.300 MPa under Ar gas protection. These sintering conditions are the same as the conditions for sintering TiC/ Ni3A1 FGM. The sintered composites were cut and ground into 36 x 4 x 3 m m strip samples; the properties were then tested with a four-point bending test method in a materials test system. In the test, the top span of the samples was 3 0 m m and the underpart span 20 mm. Strain ganges were fixed at the centre of the samples. Young's modulus E was calculated through the tested load-defiection curve; t h e Poisson ratio /z was determined by the ratio of the tested cross-strain to the longitudinal strain; and the fracture strength of the samples was indicated as the four-point bending strength o-. The average values of E and /z were obtained from 30 samples and the average ofrom seven samples. In addition, the thermal expansion coefficient was tested by a nonloaded thermal dilatometer and the thermal diffusion coefficient ff and specific heat Cp were tested by a laser perturbation method (SHINKO RIKO TC-700). The samples were 10 mm long and 2 mm wide; they were tested at 25 °C, 300 °C, 600 °C, 800 °C and 1200 °C, respectively. The thermal conductivity A was calculated from A = ff x Cp x d, where d is the sample density. The tested properties are indicated in Table I. F ig . 1 is a finite element model for the thermal stress calculation of a disc-like TiC/Ni3A1 FGM. The material is 30 mm in diameter, 6 mm in thickness, and has 1 l gradient layers. The top side is the metal phase, the bottom is the ceramic phase, and between the two sides are the FGM interlayers. Due to the

Residual and working stresses of a TiC/Ni3Al FGM and its structure optimization

3C-SiC (111) thick films were grown on Si (110) substrate via laser chemical vapor deposition (laser CVD) using hexamethyldisilane (HMDS) as precursor and argon (Ar) as dilution gas The 3C-SiC (111) polycrystalline films were prepared at deposition temperature (Tdep) of 1423-1523 K, whereas the 3C-SiC (111) epitaxial films were obtained at 1573-1648 K with the thickness of 540 to 932 μm The in-plane relationship was 3C-SiC [-1-12] // Si [001] and 3C-SiC [-110] // Si [-110] The deposition rates (Rdep) were 162-280 μm·h−1, which are 2 to 100 times higher than that of 3C-SiC (111) epi-grown on Si (111) by conventional CVD The growth mechanism of 3C-SiC (111) epitaxial films has also been proposed

This article is protected by copyright All rights reserved

Lianmeng Zhang

Papers

Theoretical study on two sedimentation processes used to form functionally graded materials

Thermoelectric properties of ZrNiSn-based half-Heusler compounds by solid state reaction method

Oxidation Behavior of ZrB2–SiC Composites at Low Pressures

Residual and working stresses of a TiC/Ni3Al FGM and its structure optimization

High‐speed heteroepitaxial growth of 3C‐SiC (111) thick films on Si (110) by laser chemical vapor deposition