A strategy to optimize the thermoelectric performance in a spark plasma sintering process

www.nature.com/scientificreports OPEN received: 27 October 2015 accepted: 29 February 2016 Published: 15 March 2016 A strategy to optimize the thermoelectric performance in a spark plasma sintering process Wan-Ting Chiu1, Cheng-Lung Chen1 & Yang-Yuan Chen1,2 Spark plasma sintering (SPS) is currently widely applied to existing alloys as a means of further enhancing the alloys’ figure of merit. However, the determination of the optimal sintering condition is challenging in the SPS process. This report demonstrates a systematic way to independently optimize the Seebeck coefficient S and the ratio of electrical to thermal conductivity (σ/κ) and thus achieve the maximum figure of merit zT = S2(σ/κ)T. Sb2−xInxTe3 (x = 0–0.2) were chosen as examples to validate the method. Although high sintering temperature and pressure are helpful in enhancing the compactness and electrical conductivity of pressed samples, the resultant deteriorated Seebeck coefficient and increasing thermal conductivity eventually offset the benefit. We found that the optimal sintering temperature coincides with temperatures at which the maximum Seebeck coefficient begins to degrade, whereas the optimal sintering pressure coincided with the pressure at which the σ/κ ratio reaches a maximum. Based on this principle, the optimized sintering conditions were determined, and the zT of Sb1.9In0.1Te3 is raised to 0.92 at 600 K, showing an approximately 84% enhancement. This work develops a facile strategy for selecting the optimal SPS sintering condition to further enhance the zT of bulk specimens. Thermoelectric (TE) materials have progressively attracted interest in recent years for their potential in our quest for a sustainable energy solution1. A measure of the performance of thermoelectric materials is the figure of merit: zT =  σ S2T/κ , where S is the Seebeck coefficient, σ is the electrical conductivity, κ is the thermal conductivity, T is the absolute temperature, and the product (σ S2) is known as the power factor2. Specifically, a good TE material should have both a high Seebeck coefficient and a high σ /κ ratio. In the last decade, several strategies such as band structure engineering3, hierarchical architecture structuring4, and nanostructuring5 have been developed in efforts to overcome thermoelectric performance bottlenecks by reducing thermal conductivity and/or increasing the power factor to effectively enhance zT. The thermoelectric performance of materials is highly sensitive to the materials’ microstructures and fabrication methods6. An ideal synthetic approach should produce thermoelectric materials with the following features: high and consistent quality, low cost, scalability, machinability, and good thermal and chemical stability. With regard to these qualities, the spark plasma sintering (SPS) technique has distinct advantages, which include rapid heating and electric current activation; the method also produces well-controlled microstructures. For these reasons, SPS has become a popular tool in thermoelectric research7–9. The SPS densification process is a temperature-dependent mass transfer process that entails a complex working mechanism. In brief, the sintering mechanism involves surface diffusion, evaporation, grain boundary diffusion and interparticle neck formation10,11. To achieve effective densification, the sintering parameters should be tuned to favour densification over coarsening, but elevated temperatures tend to cause particle coarsening and thereby adversely change the chemical composition and Seebeck coefficient of materials. The densification process is effectively a sliding and rearrangement of particles (or crystallites); therefore, with the assistance of an applied external pressure, the sintering temperature required for the densification can be significantly reduced. Despite the fact that the SPS technique possesses great potential in creating high-performance composited bulk thermoelectric materials, very few studies have investigated the criterion for optimal sintering temperature and pressure in SPS processes. In this work, we offer a facile strategy in determining the optimal SPS sintering condition—specifically, by judging both the σ /κ ratio and the sintering temperatures at which the maximum 1 Institute of Physics, Academia Sinica, Taipei 11529, Taiwan. 2Graduate Institute of Applied Physics, National Chengchi University, Taipei 11605, Taiwan. Correspondence and requests for materials should be addressed to C.L.C. (email: ) or Y.Y.C. (email: ) Scientific Reports | 6:23143 | DOI: 10.1038/srep23143 1 www.nature.com/scientificreports/ Figure 1. (a) Powder XRD patterns of Sb2−xInxTe3 (x =  0–0.2) samples, (b) lattice parameters as a function of x at 300 K. The inset of (a) shows the crystal structure of Sb2Te3. Seebeck coefficient begins to degrade, thus proving the effectiveness of this strategy in the zT enhancement of bulk Sb2−xInxTe3 (x =  0–0.2) alloys. We also examined the applicability of our approach to other thermoelectric material systems and confirmed its generalisability to other systems that can be densified by the SPS process. At 300–500 K, several potential thermoelectric materials are found in V–VI compounds such as Sb2Te3, Bi2Te3 and Bi2Se312. To achieve higher zT, Bi-Sb-Te and Bi-Se-Te ternary alloys are developed by optimizing the carrier concentrations and thermal properties accordingly13. In contrast, Sb2Te3 attracts less attention because of its low figure of merit (zT ~ 0.3)13. Previous studies have indicated the presence of antisite defects in Sb-Te or Bi-Te alloys owing to their bond polarity similarity14. Because Sb-Te bond has relatively lower bond polarity compared with Bi-Te, the antisite defects are more severe in Sb-Te. More defects in the lattice certainly contribute extra carriers, too many of which inevitably lead to a poorer Seebeck coefficient and higher thermal conductivity15. Doping effects on Sb2Te3 have been intensively studied for theoretical and applicational purposes. Dopant candidates such as bismuth16, selenium17, titanium18, vanadium19, and indium20,21 have been examined. Indium is a promising dopant because it not only effectively suppresses the formation of antisite defects but also widens the bandgap of Sb2Te3, which is helpful in reducing the detrimental bipolar conduction at high temperatures. Under optimal SPS conditions, the In-doped Sb2Te3 alloys exhibited good thermoelectric performance with a zT value of approximately 0.73 at 600 K22. In this study, p-type polycrystalline Sb2−xInxTe3 (x =  0, 0.05, 0.10, 0.15, 0.20) were prepared by melting and annealing, followed by the spark plasma sintering procedure. Six combinations of sintering temperature and pressure were performed on testing specimens of Sb1.85In0.15Te3 to identify the optimal sintering temperature and pressure by measuring their thermoelectric properties. The optimal SPS criterion was then applied to all other specimens to validate the m (...truncated)