Magnetism and metal-insulator transition in<mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"><mml:mrow><mml:mtext>Fe</mml:mtext><mml:msub><mml:mrow><mml:mrow><mml:mo>(</mml:mo><mml:mrow><mml:msub><mml:mrow><mml:mtext>Sb</mml:mtext></mml:mrow><mml:mrow><mml:mn>1</mml:mn><mml:mo>−</mml:mo><mml:mi>x</mml:mi></mml:mrow></mml:msub><mml:msub><mml:mrow><mml:mtext>Te</mml:mtext></mml:mrow><mml:mi>x</mml:mi></mml:msub></mml:mrow><mml:mo>)</mml:mo></mml:mrow></mml:mrow><mml:mn>2</mml:mn></mml

Søndergaard

Applied Physics Letters

et al. 2011

Substitution of Sb in FeSb 2 by less than 0.5% of Te induces a transition from a correlated semiconductor to an unconventional metal with large effective charge carrier mass m ‫ء‬ . Spanning the entire range of the semiconductor-metal crossover, we observed an almost constant enhancement of the measured thermopower compared to that estimated by the classical theory of electron diffusion. Using the latter for a quantitative description one has to employ an enhancement factor of 10-30. Our observations point to the importance of electron-electron correlations in the thermal transport of FeSb 2 , and suggest a route to design thermoelectric materials for cryogenic applications. © 2011 American Institute of Physics. ͓doi:10.1063/1.3556645͔Current material science has not yet found efficient thermoelectric ͑TE͒ materials for cryogenic ͑T Ͻ 100 K͒ applications such as spot cooling microelectronic superconducting devices. 1 In this context recent observations of a colossal thermopower S of -͑6-45͒ mV/K below 30 K in FeSb 2 have stimulated a great interest in the underlying physics and also practical applications. [2][3][4] Original interest in this compound stems from its narrow energy gap and strong electronic correlations as well as the thus induced unusual physical properties. 5-8 For a long time, systems with electron correlations have been expected to be candidates for TE applications. 1 Following this line are correlated semiconductors, e.g., FeSi, and correlated metals, such as CePd 3 , which do show promising TE properties at cryogenic range. 9 Though the figure of merit zT ͑=TS 2 / , with being the electrical resistivity and the thermal conductivity͒ is only ϳ0.01 at 50 K for FeSi, it can be enhanced to slightly below 0.1 at 100 K by 5% Ir doping. 9 CePd 3 , on the other hand, an intermediate-valence metal with low carrier density, exhibits a zT = 0.23 at around 200 K. 9 Our intensive investigations on FeSb 2 have already clarified the interplay between the enhanced thermopower and electron correlations, 2,3,10,11 nevertheless the microscopic mechanism of the enhancement is far from being understood. 4 Here, we report TE properties of FeSb 2−x Te x with a narrow doping range, 0 Ͻ x Ͻ 0.16, where a semiconductormetal ͑SM͒ transition occurs. Magnetism and electrical resistivity over a wide x range ͑0-1.2͒ have already been reported by Hu et al. 12 Our emphasis is aimed at learning to what extent the enhanced S in a correlated semiconductor can be kept when crossing a SM transition. FeSb 2 and FeTe 2 are isostructural ͑marcasite type͒ semiconductors having rather different energy gaps. The former shows a transport gap of E g = 26-36 meV together with an even smaller one of ϳ6 meV, 11 contrasting to a much larger E g of the latter, 0.2-0.5 eV. 13 The most striking finding of our study is that the enhancement to S͑T͒ ͑by a factor of 10-30͒ relative to the classical expectation, is rather robust against doping, spanning the entire range of the SM transition. This feature provides a convincing link between the enhan...

mentioning

confidence: 76%

Unchanged thermopower enhancement at the semiconductor-metal transition in correlated FeSb2−xTex

Søndergaard

Applied Physics Letters

et al. 2011

Journal of Applied Physics

“…[17][18][19] Powder X-ray diffraction (XRD) spectra of the ground samples were taken with Cu Kα radiation (λ = 1.5418Å) using a Rigaku Miniflex x-ray machine. The lattice parameters were obtained by fitting the XRD pattern using the RIETICA software.…”

Section: Methodsmentioning

confidence: 99%

“…[17][18][19][20] Forming solid solution, i.e., maximizing the influence of point-defect scattering by creating large mass difference and elastic strain at the lattice sites, proved to be an efficient approach to suppress the thermal conductivity and enhance ZT of some thermoelectric materials, such as filled skutterudites materials Co(Sb 1−x As x ) 3 , Co …”

Section: Introductionmentioning

confidence: 99%

Enhancement of the thermoelectric properties in doped FeSb2 bulk crystals

Warren

Petrović

2012

Self Cite

Kondo insulator FeSb 2 with large Seebeck coefficient would have potential in thermoelectric applications in cryogenic temperature range if it had not been for large thermal conductivity κ. Here we studied the influence of different chemical substitutions at Fe and Sb site on thermal conductivity and thermoelectric effect in high quality single crystals. At 5% of Te doping at Sb site thermal conductivity is suppressed from ∼ 250 W/Km in undoped sample to about 8 W/Km. However, Cr and Co doping at Fe site suppresses thermal conductivity more slowly than Te doping, and even at 20% Cr/Co doping the thermal conductivity remains ∼ 30 W/Km. The analysis of different contributions to phonon scattering indicates that the giant suppression of κ with Te is due to the enhanced point defect scattering originating from the strain field fluctuations. In contrast, Te-doping has small influence on the correlation effects and then for small Te substitution the large magnitude of the Seebeck coefficient is still preserved, leading to the enhanced thermoelectric figure of merit (ZT ∼ 0.05 at ∼ 100 K) in Fe(Sb 0.9 Te 0.1 ) 2 .

“…With the origin of the enhanced TEP of FeSi in debate for decades [7][8][9], the observations in FeSb 2 , with the thermoelectric power factor the largest ever reported [3,10,11], have added timely interest to this concern. Accumulating experimental facts have confirmed the significant effect of electronelectron correlations in FeSb 2 [3][4][5][6][10][11][12][13][14][15][16][17]. These include the large spectral weight redistribution up to higher energies [14,15] and the occurrence of electronic Griffiths phases at low temperatures [17].…”

Section: Introductionmentioning

confidence: 95%

Highly dispersive electron relaxation and colossal thermoelectricity in the correlated semiconductor FeSb2

Tomczak

et al. 2013

Phys. Rev. B

We show that the colossal thermoelectric power, S(T ), observed in the correlated semiconductor FeSb2 below 30 K is accompanied by a huge Nernst coefficient ν(T ) and magnetoresistance MR(T ). Markedly, the latter two quantities are enhanced in a strikingly similar manner. While in the same temperature range, S(T ) of the reference compound FeAs2, which has a seven-times larger energy gap, amounts to nearly half of that of FeSb2, its ν(T ) and MR(T ) are intrinsically different to FeSb2: they are smaller by two orders of magnitude and have no common features. Underlying the essentially different thermoelectric properties between FeSb2 and FeAs2, a large mismatch between the electrical and thermal Hall mobilities was found only in the former compound. With the charge transport of FeAs2 successfully captured by the density functional theory, we emphasize a significantly dispersive electron-relaxation time τ (ǫ k ) related to electron-electron correlations to be at the heart of the peculiar thermoelectricity and magnetoresistance of FeSb2.