Ni₀.₀₅Mo₃Sb₅.₄Te₁.₆ Based Thermoelectric Nanocomposites
MetadataShow full item record
Thermoelectric (TE) materials have the capability to convert thermal energy into useful electrical energy. Sustainable energy production and its utilization are among the many challenges that humankind is facing today. In 2016 and 2017, the expected global production of hydrocarbon based automotive vehicles is expected to be 97.8 and 101.8 million respectively, and is expected to rise. The collective thermal energy losses from radiators and exhausts from these automotive vehicles are enormous. This is a big bottleneck in sustainable energy production and utilization. In mitigating this hurdle, TE materials will play a very important role. However, TE materials have low efficiency in thermal to electrical energy conversion owing to the reciprocal relation between the thermal and the electrical transport properties. Recent advances in nanotechnology tools have given a new dimension to decouple this relation. Back in 2003, our group reported a very promising TE material, NiyMo3Sb5.4Te1.6 (y < 0.1). Improving the figure-of-merit of Ni0.05Mo3Sb5.4Te1.6 (“bulk”) material through nanocomposite synthesis is one of the goals of my research. The main outcome of nanocomposite synthesis is the reduced thermal conductivity through arresting the coherent propagation of heat carrying acoustic waves in TE materials. To this end, I synthesized and characterized the transport properties of various nanocomposites. I have used fullerenes, oxides, carbides, and metal particles to fabricate nanocomposites. Chapter 3 addresses the effect of multi-wall carbon nanotubes (MWCNT) when added to Ni0.05Mo3Sb5.4Te1.6. We characterized these samples for their TE properties, and addressed the effect of porosity on transport properties. The effect of ball-milling on MWCNT was studied. Scanning and transmission electron microscopy were used to study the microstructural and nanostructural features of the samples. In a sample with 3 mass-% of MWCNT, the main contributing factor in elevating the figure-of-merit by 25% was the reduction in the thermal conductivity by 40%. In Chapter 4, we reported the results of Ni0.05Mo3Sb5.4Te1.6 /SiC and Ni0.05Mo3Sb5.4Te1.6/Al2O3 composites consolidated through hot-pressing and spark-plasma sintering respectively. Samples with different volume fractions of SiC were prepared and characterized. Thermoelectric transport properties of these composites were characterized from 325 K to 740 K. For the sample with 0.01volume fraction of SiC, there was an enhancement in the figure-of-merit by an 18% compared to the reference sample, mainly due to an 18% reduction in the thermal conductivity. Microstructural information obtained through SEM, TEM, and BET was used to elucidate phase and transport properties. Spark-plasma sintered bulk sample has exhibited the highest figure-of-merit, which is 35% higher than the bulk consolidated through hot-pressing. Pore effect on thermal conductivity and electrical conductivity were investigated. In Chapter 5, we covered various properties of bulk/NiSb composite. In this study, requisite amounts of bulk Ni0.05Mo3Sb5.4Te1.6 thermoelectric (TE) material and NiSb nanoparticles of ~60 nm were synthesized through solid-state reactions and solvothermal routes, respectively. NiSb nanoparticles were then manually mixed with Ni0.05Mo3Sb5.4Te1.6 in volume fractions of 0, 0.034, 0.074, and 0.16. All samples were consolidated by hot-pressing, and then their TE properties were characterized. The addition of NiSb nanoparticles elevated both the electrical conductivity and the thermal conductivity, but reduced the Seebeck coefficient. TEM images show highly electrically and thermally conductive NiSb nanoparticles settled within interstitial voids between bulk particles, thus, forming bridges between the balk particles. These networks of NiSb particles facilitate the electrical and thermal conductivity. Variation in lattice thermal conductivity from 330 K to 755 K for all samples shows that phonons are undergoing the same scattering mechanism. By adopting Callaway formalism for the lattice thermal conductivity, we made attempts to delineate the contribution of various scattering mechanism to the enhanced lattice thermal conductivity in bulk/NiSb composites. We concluded that electron-phonon scattering, which is the main factor for the increased lattice thermal conductivity in Mo3Sb7-xTex (0.0 < x ≤ 1.8) series was not the reason in our case, but rather that an overlap in the DOS of phonons of bulk and NiSb particles, was primarily responsible for the high phonon-phonon transfer efficiency from bulk to NiSb particles and vice versa. We also speculated that there was a reduction in the umklapp process. Contribution from boundaries and interfaces was less in scattering phonons. Increase in hardness with respect to NiSb content provided an indication of increased Debye temperature of composites. Irrespective of extensive interfaces created by NiSb nanoparticles, there was no reduction in the thermal conductivity. Contrary to this, electrical conductivity of the bulk, 0.034, and 0.074 samples showed the same temperature dependence, whereas the 0.16 sample did not, revealing that the charge carrier scattering for the 0.16 sample was different from the rest. The existence of the electrical percolation phenomena was analyzed in detail and compared with other composites. SEM images of composites show that NiSb nanoparticles distributed homogeneously in 0.16 composite. HRTEM images of 0.16 composite shows crystalline nature of bulk and NiSb particles; we also used this information to interpret the enhanced lattice thermal conductivity. Furthermore, larger NiSb particles of 100 m 00 m size were synthesized through solid-state reaction and consolidated by hot-pressing. Their TE properties were characterized as well, in order to use these parameters to interpret the transport properties of the nanocomposites. Finally, there was a systematic improvement in the hardness of the composites with increasing NiSb content. In Chapter 6, we extensively investigated the effective thermal properties and thermal boundary resistance of Ni0.05Mo3Sb5.4Te1.6/SiC and Ni0.05Mo3Sb5.4Te1.6/Al2O3 composites at 325K. Different effective media approximations (EMA) models were utilized to predict thermal boundary resistance (Kapitza resistance). The salient feature of this study is the comparison of these two composites based on their microstructure. This study shows that it is important to give credence to microstructures while addressing the effective thermal properties. Bounds for the effective thermal conductivity were determined using the Lipton–Vernescu model. The effect of SiC and Al2O3 nanoparticles on mechanical properties were analyzed and interpreted. Thermal properties in relation to mechanical properties were discussed. In Chapter 7, we engineered the grain boundaries of bulk particles through a process called nanocoating. NiSb nanoparticles of 60 nm – 80 nm were coated on the bulk particles through the solvothermal process. The actual process of solvothermal coating did not affect the bulk material. Layers of 300 nm – 500 nm NiSb nanoparticles were observed on the bulk sample with the 0.16 vol% NiSb nanoparticles. In Chapter 8, we covered the effect of addition of C60 on the TE properties of bulk material. C60 was added in 1, 2, and, 3 mass%, respectively. Each part was hot-pressed at 150 MPa and 923 K. The sample with 1% C60 was characterized via a Rietveld refinement and TEM analysis. Measurements of the three thermoelectric key properties revealed that the Seebeck coefficient barely depends on the carbon amount added, while both the electrical and the thermal conductivity decrease with increasing amount of carbon. Depending on the amount of C60 used and on the temperature, the thermoelectric performance was either enhanced or decreased, depending on whether the electrical conductivity decreased less or more than the thermal conductivity. At the highest temperature measured, all carbon-containing samples performed better than the unmodified bulk sample, namely up to 14%. These improvements are within the error margin, however.