Candidates For Thermoelectric Materials Research Articles

Melt growth of crystalline α-SrSi2 by the vertical Bridgman method and its thermoelectric characteristics

As a candidate thermoelectric material operating in the low-temperature regime (room temperature to 473 K), we have examined α-SrSi2: a narrow bandgap semiconductor material, which is composed of environmentally benign elements. We are able to report the successful production of single-crystalline-like α-SrSi2 with clear semiconducting properties by melting synthesis, a thermodynamically stable and thermally equilibrium process that reduces unexpected process contamination and improves crystallinity.The crystals up to 20 × 20 × 5 mm3 have been obtained for a solid–liquid phase reaction temperature of 1308 K for 10 h with a growth rate averaging at 0.617 K/h in the temperature range from 1403 K to 1373 K. The band gap of this material was measured at 48 meV, which is higher than the values previously reported. The power factor was measured at 2.9 mW/mK2 at 300 K, which is the highest value ever reported for undoped α-SrSi2.Based on the experimental bandgap values, the hybrid functional was used to correct the first-principles calculation method, and the Seebeck coefficient obtained from the first-principles calculation was compared with experimental values. Which values showed good agreement with experimental values.

Advancing thermoelectric materials discovery through semi-supervised learning and high-throughput calculations

The data-driven machine learning technique is widely used to assist in accelerating the design of thermoelectric materials. In this study, we proposed a positive and unlabeled learning (PU learning) method, a semi-supervised learning, to train a classifier to distinguish the positive samples from the unlabeled samples, in which the positive class was labeled by matching the formulas in our dataset with the published article titles. The probabilities that the unlabeled materials belong to the positive class were predicted by PU learning, and 40 candidate thermoelectric materials were determined. The transport properties were calculated by high-throughput first-principles calculations, among which 8 p-type and 12 n-type materials have the maximum theoretical zT values greater than 1. Specifically, a series of AX2 binary compounds, (Cd/Zn)(GaTe2)2 ternary compounds, and Cs(Dy/Ho/Tb)2Ag3Te5 quaternary compounds deserve further investigations in the future.

Low-Dimensionalization Enhancing the Thermoelectric Performance of Higher Manganese Silicide

Higher manganese silicide (HMS) is a candidate thermoelectric (TE) material at medium temperature due to its non-toxicity, abundance, and competitive price. The focus on improving the TE performance of HMS is to decrease the thermal conductivity. Low-dimensionalization techniques, such as nanocrystallization, embedding quantum dots (QDs) and thin film formation are effective strategies to decrease the lattice thermal conductivity by enhancing the phonon scattering on interfaces. Additionally, the Seebeck coefficients also can be improved due to the energy filtering effect via the interface barrier, and correspondingly increasing the power factor of HMS. The TE performance of HMS can be enhanced due to synergistically optimized electrical and thermal properties.

Preparation and Thermoelectric Properties of Nano-Bismuth Telluride Film by High-Temperature Thermal Shock

Thermoelectric materials can directly convert thermal energy into electrical energy, realizing the recovery of waste heat. Bismuth Telluride (Bi-Te) is considered as a perfect candidate thermoelectric material which has great potential in the field of refrigeration and temperature sensor. However, in the field of intelligent wearable devices and integrated circuits, traditional Bi-Te block material is difficult to be directly used due to its poor flexibility. In this paper, a series of Bi-Te thin films were prepared by a self-designed high-temperature thermal shock equipment. This equipment can heat up the reduced graphene oxide strip to 1750 K in 20 ms, which features both high heating rate (8.2 × 104 K/s) and cooling rate (1.5 × 104 K/s). Thermoelectric films on different substrates were prepared via high-temperature thermal shock. Through regulating the temperature of evaporation source, the particle size and composition of Bi-Te thin films can be precisely modulated, thus optimizing the Seebeck coefficient of the films. The Seebeck coefficient of copper foil (Cuf)-based Bi-Te film can reach 345.41 μV/K, which was prepared by thermal-shocking the Bi-Te powders for 30 s at 900°C.

Preparation and thermoelectric properties of Mn<sub>3</sub>As<sub>2</sub>-doped Cd<sub>3</sub>As<sub>2</sub> nanostructures

Cd<sub>3</sub>As<sub>2</sub>, especially its various nanostructures, has been considered as an excellent candidate for application in novel optoelectronic devices due to its ultrahigh mobility and good air-stability. Recent researches exhibited Cd<sub>3</sub>As<sub>2</sub> as a candidate of thermoelectric materials by virtue of its ultralow thermal conductivity in comparison with other semimetals or metals. In this work, at first <b>(</b> Cd<sub>1–<i>x</i></sub>Mn<sub><i>x</i></sub>)<sub>3</sub>As<sub>2</sub> (<i>x</i> = 0, 0.05, 0.1) bulk alloys are prepared by high-pressure sintering to suppress the volatilization of As element, and then several kinds of Mn<sub>3</sub>As<sub>2</sub>-doped Cd<sub>3</sub>As<sub>2</sub> nanostructures are obtained on mica substrates by chemical vapor deposition (CVD), with bamboo-shoot-nanowire structure forming in a high-temperature region and films in a low-temperature region. Effects of Mn<sub>3</sub>As<sub>2</sub> doping on the crystalline structure, phase compositions, microstructures and thermoelectric properties of the Cd<sub>3</sub>As<sub>2</sub> nanostructures are systematically studied. Energy-dispersive spectrometer (EDS) analysis at various typical positions of the Mn<sub>3</sub>As<sub>2</sub>-doped Cd<sub>3</sub>As<sub>2</sub> nanostructures shows that the Mn content in these nanostructures is in a range of 0.02%–0.18% (atomic percent), which is much lower than the Mn content in <b>(</b> Cd<sub>1–<i>x</i></sub>Mn<sub><i>x</i></sub>)<sub>3</sub>As<sub>2</sub> (<i>x</i> = 0, 0.05, 0.1) parent alloys. The main phases of these nanostructures are all body centered tetragonal <i>α</i> phase with a small amount of primitive tetragonal <i>α</i>′ phase. Doping results in the <i>α</i>″ phase and Mn<sub>2</sub>As impurity phase occurring. The Cd<sub>3</sub>As<sub>2</sub> film presents a self-assembled cauliflower microstructure. Upon Mn<sub>3</sub>As<sub>2</sub> doping, this morphology finally transforms into a vertical-growth seashell structure. In a high temperature region of the mica substrate, a unique bamboo-shoot-nanowire structure is formed, with vertical-growth bamboo shoots connected by nanowires, and at the end of these nanowires grows a white pentagonal flower structure with the highest Mn content of 0.18% (atomic percent) for all the nanostructures. Conductivity of the Cd<sub>3</sub>As<sub>2</sub> film and the bamboo-shoot-nanowire structure are ～20 and 320 S/cm, respectively. The remarkable conductivity enhancement can be attributed to higher crystallinity and the formation of nanowire conductive network, which significantly increase carrier concentration and Hall mobility. The Hall mobility values of the nanowire structures range from 2271 to 3048 cm<sup>2</sup>/(V·s) much higher than the values of 378–450 cm<sup>2</sup>/(V·s) for the films. The Seebeck coefficient for the bamboo-shoot-nanowire structure is in a range of 59–68 µV/K, which is about 15% higher than those for the films (50–61 µV/K). Although maximal power factor of the bamboo-shoot-nanowire structure is 14 times as high as that of the thin film, reaching 0.144 mW/(m·K<sup>2</sup>) at room temperature, this value is still one order of magnitude lower than the previously reported value of 1.58 mW/(m·K<sup>2</sup>) for Cd<sub>3</sub>As<sub>2</sub> single crystal.

Structural Analysis, Phase Stability, Electronic Band Structures, and Electric Transport Types of (Bi2)m(Bi2Te3)n by Density Functional Theory Calculations

Thermoelectric power generation is a promising candidate for automobile energy harvesting technologies because it is eco-friendly and durable owing to direct power conversion from automobile waste heat. Because Bi−Te systems are well-known thermoelectric materials, research on (Bi2)m(Bi2Te3)n homologous series can aid the development of efficient thermoelectric materials. However, to the best of our knowledge, (Bi2)m(Bi2Te3)n has been studied through experimental synthesis and measurements only. Therefore, we performed density functional theory calculations of nine members of (Bi2)m(Bi2Te3)n to investigate their structure, phase stability, and electronic band structures. From our calculations, although the total energies of all nine phases are slightly higher than their convex hulls, they can be metastable owing to their very small energy differences. The electric transport types of (Bi2)m(Bi2Te3)n do not change regardless of the exchange–correlation functionals, which cause tiny changes in the atomic structures, phase stabilities, and band structures. Additionally, only two phases (Bi8Te9, BiTe) became semimetallic or semiconducting depending on whether spin–orbit interactions were included in our calculations, and the electric transport types of the other phases were unchanged. As a result, it is expected that Bi2Te3, Bi8Te9, and BiTe are candidates for thermoelectric materials for automobile energy harvesting technologies because they are semiconducting.

Study of Electrical and Thermoelectrical Properties of One Strand DNA Chain for Nanoscale Thermoelectric Applications

The concept of using DNA molecules for designing nano-scale electronic systems has attracted researcher’s attention due to the unique properties of DNA, such as self-assembly and self-recognition. Thus, increased number of studies, theoretically and experimentally, have been carried out to study the possibility of adopting DNA molecules in designing nanoscale thermoelectric devices. In this work, a general expression of the electron transmission probability that describes the electron transfer through one strand DNA chain has been derived using the steady-state-formalism by assuming one strand of DNA molecules as line model. The energy-dependent transmission was studied, then energy-and temperature-dependent Seebeck coefficient, and thermoelectric characteristics of four one strand DNA sequences: (A-A)10, (C-C)10, (G-G)10 and (T-T)10 are theoretically studied. According to the obtained results, it is found that the transmission behavior (magnitude and position) is varying with the type of DNA sequence. Also, the energy dependent Seebeck coefficient (S-E) curves clearly show a nonlinear energy-dependence, while the relationship between Seebeck coefficient and temperature (S-T) is linear. Thermoelectric power factor as a function of temperature was found to be enhanced with the temperature increment for the four types of DNA nucleobases. The highest values of thermoelectric power factor belong to thymine (120Wm-1K-2) and cytosine (60 Wm-1K-2), that nominate them as outstanding candidate thermoelectric materials to be adopted in the fabrication of one strand DNA-base nanoscale thermoelectric devices.

Carrier Concentration of CeFe<sub>2</sub>Al<sub>10</sub> as a Candidate of Thermoelectric Material

The study of Hall effect of Kondo semiconductor CeFe2Al10 is reported as a candidate of thermoelectric material used at low temperatures. Single crystals of CeFe2Al10 with orthorhombic crystal structure were grown by Al self-flux method. An anisotropy of the Hall effect is clarified by measuring Hall resistance by changing the direction of electrical current, magnetic field, and voltage respect to all the three crystal axes of orthorhombic crystal structure. The Hall effect of CeFe2Al10 has a strong anisotropy against the direction of magnetic field but weak anisotropy against the directions of current and voltage. The value of carrier concentration indicates that CeFe2Al10 is matallic, which causes a low performance as a thermoelectric material. In order to improve the value of dimensionless figure of merit, the electrons should be doped to CeFe2Al10.

Novel metal oxides with promising high-temperature thermoelectric performance

By performing high-throughput calculations, we have successfully screened out a novel class of metal oxides MTa2O6 (M = Mg, Ca) as promising candidate thermoelectric materials for high-temperature applications.

Improved thermoelectric properties of doped A0.5B0.5NiSn (A, B = Ti, Zr, Hf) with a special quasirandom structure

MNiSn (Ti, Hf, Zr) is used as the parent material in constructing special quasirandom structure-based half-Heusler compounds that experimentally mimic atomic occupation disorders. We randomly doped Ti, Hf, and Zr with an occupation probability of 50% to obtain Ti0.5Zr0.5NiSn, Ti0.5Hf0.5NiSn, and Zr0.5Hf0.5NiSn, respectively. We systematically studied the electronic structure and thermoelectric transport properties of these compounds by referring to density functional theory and the semiclassical Boltzmann transport equation. Compared with the parent compound MNiSn (M = Ti, Zr, Hf), the substituted compounds retain the characteristic of semiconductivity. The lattice thermal conductivity of the three compounds is about 2.5 Wm−1 K−1 at about 1000 K, which is lower than that of their corresponding parent materials (5.0–6.0 Wm−1 K−1). The maximum thermoelectric figures of merit for Ti0.5Zr0.5NiSn (0.57), Ti0.5Hf0.5NiSn (0.55), and Zr0.5Hf0.5NiSn (0.58) are higher than those of their parent compounds TiNiSn (0.35), ZrNiSn (0.496), and HfNiSn (0.473). The increase in thermoelectric figure of merit is due to low lattice thermal conductivity and high power factor. Therefore, the three predicted compounds can be considered candidate thermoelectric materials that can be used at a carrier concentration of 1020–1022 cm−3 and a temperature of 900–1200 K.

Phonon transport properties of NbCoSb compound

We present a first-principles study of the phonon transport properties of the Half-Heusler NbCoSb compound. The single crystal elastic constants have been calculated using the stress- strain method. The mechanical properties such as bulk modulus, shear modulus, Young’s modulus, Poisson’s ratio, and anisotropy factor are estimated using the Voigt–Reuss–Hill approximation. The Grüneisen parameters, phonon life time, and lattice thermal conductivity have been studied in detail. The elastic and phonon properties reveal that this compound is mechanically and dynamically stable. An analysis of the harmonic and anharmonic phonon dispersion and vibrational density of states reveal that, NbCoSb has a very low lattice thermal conductivity, hence, it is a candidate thermoelectric material.

Improved thermoelectric performance of p-doped half-Heusler Ti0.5Zr0.5CoSb0.5P0.5, Ti0.5Hf0.5CoSb0.5P0.5, and Zr0.5Hf0.5CoSb0.5P0.5 compounds

We substituted the M site with two atoms of Ti, Zr, Hf and Sb atom with P atom in MCoSb half-Heusler compounds to obtain Ti0.5Zr0.5CoSb0.5P0.5, Ti0.5Hf0.5CoSb0.5P0.5, and Zr0.5Hf0.5CoSb0.5P0.5. Then, we systematically studied the electronic structure and TE transport properties of the three compounds by using density functional theory and the semiclassical Boltzmann transport equation. In contrast to their parent compound MCoSb (M = Ti, Zr, Hf), the substituted compounds retained the characteristic of indirect bandgap semiconductivity. The lattice thermal conductivity of these three compounds was lower than that of their corresponding parent material. The thermoelectric figure of merit of the three compounds was approximately 0.85, which was remarkably higher than that of their corresponding parent compounds (0.25). The increased in the thermoelectric figure of merit may be attributed to a reduction in lattice thermal conductivity and high power factor. Therefore, the three compounds with carrier concentrations of 1020 and 1021 cm−3 are candidate thermoelectric materials that can be applied at the temperature of 900–1500 K.

Achieving high thermoelectric properties of Bi2S3 via InCl3 doping

Bi2S3, with earth-abundant compositions and a low thermal conductivity, is regarded as a candidate thermoelectric material. In this work, Bi2S3 samples that were doped with x % mol InCl3 were successfully fabricated via a mechanical alloying and spark plasma sintering process. InCl3, as an n-type donor dopant, was added to the Bi2S3 system to improve its electrical transport properties and optimize its thermal conductivity. Upon doping, the electrical conductivity of Bi2S3 doped with 1 mol% InCl3 reaches up to 62 Scm−1, and the Seebeck coefficient maintains a relatively large value of −244 μV K−1 at 673 K, which results in a maximum power factor of 363 μW m−1 K−2. Furthermore, due to a simultaneously reduced thermal conductivity at high temperature, a ZT peak of 0.57 is obtained at 673 K along the parallel to the press direction for the sample doped with 1.0 mol% InCl3, which is almost four times higher than that of pristine Bi2S3 (0.14 at 673 K). The elastic properties and Debye temperature of Bi2S3 are also calculated to analyze the origin of the intrinsically low thermal conductivity and are compared to those of other thermoelectric materials with a low thermal conductivity.

Effect of Functional Groups on the Thermoelectric Performance of Carbon Nanotubes

Single-wall carbon nanotubes (SWCNTs) have received much attention in the past decade due to their excellent electric and mechanical properties. The quantum confinement effect of charge carriers in an individual carbon nanotube (CNT), together with the flexible topology and high tensile strength, make it a potential candidate thermoelectric (TE) material. However, the low Seebeck coefficient and high thermal conductivity of SWCNTs limits further development for a good TE material with high performance. Although many efforts have been focused on the improvement of TE performance for SWCNTs by doping and composites, few works have been devoted to investigation of functional groups in SWCNTs. In this work, we investigated the TE performance of SWCNT films in detail with different functional groups (hydroxyl, carboxyl, and amino). It is found that the Seebeck coefficient of SWCNTs with different functional groups have an obvious improvement with the decrease in electrical conductivity. An optimal power factor of 47.8 μW m−1 K−2 was obtained for SWCNTs with a hydroxyl group comparable to pure SWCNTs. Significantly, the introduction of functional groups results in a marked reduction in thermal conductivity and an enhanced TE figure of merit (ZT). This work provides an alternative strategy to optimize the TE performance of SWCNTs.

Ab initiothermal conductivity of thermoelectricMg3Sb2: Evidence for dominant extrinsic effects

The lattice thermal conductivity of the candidate thermoelectric material Mg$_3$Sb$_2$ is studied from first principles, with the inclusion of anharmonic, isotope, and boundary scattering processes, and via an accurate solution of the Boltzmann equation. We find that the anomalously low observed conductivity is due to grain-boundary scattering of phonons, whereas the purely anharmonic conductivity is an order of magnitude larger. Mass disorder due to alloying and off-stoichiometry is also found to contribute significantly to its decrease. Combining ab initio values vs sample size with measured grain-size distributions, we obtain an estimate of $\kappa$ vs T in nano-polycrystalline material in good agreement with typical experiments, and compute the ZT figure of merit in the various cases.

Effects of Y, GdCu, and Al Addition on the Thermoelectric Behavior of CoCrFeNi High Entropy Alloys

Thermoelectric (TE) materials can interconvert waste heat into electricity, which will become alternative energy sources in the future. The high-entropy alloys (HEAs) as a new class of materials are well-known for some excellent properties, such as high friction toughness, excellent fatigue resistance, and corrosion resistance. Here, we present a series of HEAs to be potential candidates for the thermoelectric materials. The thermoelectric properties of YxCoCrFeNi, GdxCoCrFeNiCu, and annealed Al0.3CoCrFeNi were investigated. The effects of grain size and formation of the second phase on thermoelectric properties were revealed. In HEAs, we can reduce the thermal conductivity by controlling the phonon scattering due to the considerable complexity of the alloys. The Y, Gd-doped HEAs are competitive candidate thermoelectric materials for energy conversion in the future.

Prediction of fully compensated ferrimagnetic and nonmagnetic semiconductors with promising thermoelectric properties through the Mo substitution of Cr for Ti2CrZ (Z=Ge, Sn) Heusler alloys

The electronic, magnetic and thermoelectric properties of Heusler alloys Ti2Cr1-x/4Mox/4Z (Z = Ge, Sn) (x = 1, 2, 3, 4) have been investigated by the density functional theory calculations and Boltzmann transport equation. The electronic band structures show that Ti2Cr1-x/4Mox/4Ge (x = 1, 2) and Ti2Cr1-x/4Mox/4Sn (x = 1, 2, 3) have the fully compensated ferrimagnetic semiconducting nature with totally zero magnetism, belonging to promising spintronic materials. The two full substitution alloys Ti2MoGe and Ti2MoSn belong to nonmagnetic semiconductors with both the atomic and total nonmagnetic moments. The total moment Mt of all the predicted semiconductors satisfy the rule of Mt = Zt-18 per unit cell since their valence electron numbers Zt are all 18, and the reason has been demonstrated according to the feature of fully occupied bands. Moreover, the three predicted semiconductors Ti2Cr1/2Mo1/2Ge, Ti2MoGe and Ti2MoSn have high thermoelectric power factor, indicating that they are the candidate thermoelectric materials for moderate temperature application.

Discovery of Intermetallic Compounds from Traditional to Machine-Learning Approaches

Intermetallic compounds are bestowed by diverse compositions, complex structures, and useful properties for many materials applications. How metallic elements react to form these compounds and what structures they adopt remain challenging questions that defy predictability. Traditional approaches offer some rational strategies to prepare specific classes of intermetallics, such as targeting members within a modular homologous series, manipulating building blocks to assemble new structures, and filling interstitial sites to create stuffed variants. Because these strategies rely on precedent, they cannot foresee surprising results, by definition. Exploratory synthesis, whether through systematic phase diagram investigations or serendipity, is still essential for expanding our knowledge base. Eventually, the relationships may become too complex for the pattern recognition skills to be reliably or practically performed by humans. Complementing these traditional approaches, new machine-learning approaches may be a viable alternative for materials discovery, not only among intermetallics but also more generally to other chemical compounds. In this Account, we survey our own efforts to discover new intermetallic compounds, encompassing gallides, germanides, phosphides, arsenides, and others. We apply various machine-learning methods (such as support vector machine and random forest algorithms) to confront two significant questions in solid state chemistry. First, what crystal structures are adopted by a compound given an arbitrary composition? Initial efforts have focused on binary equiatomic phases AB, ternary equiatomic phases ABC, and full Heusler phases AB2C. Our analysis emphasizes the use of real experimental data and places special value on confirming predictions through experiment. Chemical descriptors are carefully chosen through a rigorous procedure called cluster resolution feature selection. Predictions for crystal structures are quantified by evaluating probabilities. Major results include the discovery of RhCd, the first new binary AB compound to be found in over 15 years, with a CsCl-type structure; the connection between "ambiguous" prediction probabilities and the phenomenon of polymorphism, as illustrated in the case of TiFeP (with TiNiSi- and ZrNiAl-type structures); and the preparation of new predicted Heusler phases MRu2Ga and RuM2Ga (M = first-row transition metal) that are not obvious candidates. Second, how can the search for materials with desired properties be accelerated? One particular application of strong current interest is thermoelectric materials, which present a particular challenge because their optimum performance depends on achieving a balance of many interrelated physical properties. Making use of a recommendation engine developed by Citrine Informatics, we have identified new candidates for thermoelectric materials, including previously unknown compounds (e.g., TiRu2Ga with Heusler structure; Mn(Ru0.4Ge0.6) with CsCl-type structure) and previously reported compounds but counterintuitive candidates (e.g., Gd12Co5Bi). An important lesson in these investigations is that the machine-learning models are only as good as the experimental data used to develop them. Thus, experimental work will continue to be necessary to improve the predictions made by machine learning.

Eutectoid transformations in Fe-Si Alloys for thermoelectric applications

Semiconducting β-FeSi2 is a candidate thermoelectric material whose constituents are abundant and eco-friendly, but significant improvements in the relevant properties are needed. This work investigates eutectoid decomposition, α-FeSi2 → β-FeSi2 + Si, as a means to modify microstructure and control thermal transport. Process conditions are adjusted to strongly affect both the microstructural lengthscales and morphology, and hence the thermal conductivity. Low temperature annealing of a hypoeutectic sample produces cooperatively-grown Si lamellae, which then decompose into Si nanowires by Rayleigh instability upon further aging. We show that nucleation of eutectoid colonies occurs preferentially on cracks, while at smaller undercooling, nucleation also occurs on eutectic Si particles. The growth velocity, v, and interlamellar spacing, λ, of the pearlitic colonies obey a relation of the type vλn = f(T). This sets the bounds of the activation energy for the diffusion mechanism, although the exact mechanism cannot be specified. Nanostructuring of eutectoid Si increases heterointerface density by 40x, with a concomitant reduction in thermal conductivity of 2x. The thermal boundary conductance is determined for the β-FeSi2/Si heterointerface, which shows that this interface only weakly scatters phonons.

A temperature-variant method for performance modeling and economic analysis of thermoelectric generators: Linking material properties to real-world conditions

A new methodology for the systematic study of thermoelectric generator (TEG) design and economic analysis is presented, with the objective of assessing the performance and financial feasibility of small-scale TEG installations, for 4 leading candidate thermoelectric materials. Temperature of a steam trap pipe surface were measured at the University of California Davis Pilot Brewery, and device performance was modeled using the finite-element modeling software ANSYS. The model integrated temperature-dependent material properties from leading candidate thermoelectric materials and experimental time-variant temperature data. Calculated power outputs were utilized in a net present value (NPV) framework to assess the financial feasibility and economic implications of small scale TEG installations, as well as to address the aspects of TEG research, design and implementation which have potential for rapid and substantive improvement. This model, along with case study results, provides a powerful platform for analyzing the performance of real-world systems and can be used to predict where further technological development on TEG materials and devices would be most effective. It is found that a BiSbTe based TEG generated the highest power output at the measured temperatures and consequently resulted in the highest NPV at the end of 25years. Sensitivity analysis of the NPV revealed a strong dependence on the heat-exchanger cost, highlighting the importance of efficient heat transfer design. The zT necessary for a 7-year payback period as a function of the capital cost and hot-side temperature was also calculated for a SiGe based TEG.