betaFeedback

WorksPhysics and Astronomy

Electronic transport and thermoelectricity in selenospinel Cu6−x_{6-x}Fe4+x_{4+x}Sn12_{12}Se32_{32}

Abstract

We report a study of selenospinel Cu6−x_{6-x}Fe4+x_{4+x}Sn12_{12}Se32_{32} (xx = 0, 1, 2) single crystals, which crystalize in a cubic structure with the Fd3‾mFd\overline{3}m space group, and show typical semiconducting behavior. The large discrepancy between the activation energy for electrical conductivity EρE_ρ (32.3 ∼\sim 69.8 meV), and for thermopower ESE_\textrm{S} (3.2 ∼\sim 11.5 meV), indicates a polaronic transport mechanism between 350 and 50 K. With decreasing temperature, it evolves into variable-range hopping conduction. Furthermore, the heat capacity shows a hump around 25(5) K and diverges from the Debye T3T^3 law at low temperatures, indicating the observation of structural glassy features in these crystalline solids.

Introduction

New transition metal chalcogenides may enable transformative changes in thermoelectric energy creation and conversion [1, 2, 3]. The Kondo-insulator-like semiconductor FeSb2_2 hosts not only strong electronic correlation but also the highest thermoelectric power factor in nature along with thermopower up to ∼ 45 mV K−1^{-1} [4, 5, 6, 7, 8, 9]. The ternary CoSbS also features high value of thermopower ∼ 2.5 mV K−1^{-1} around 40 K but also relatively large value of thermal conductivity ∼ 100 W K−1^{-1} m−1^{-1} near the thermopower peak temperature [10]. This inhibits the thermoelectric figure of merit ZT=(S2/ρκ)TZT = (S^2/\rho\kappa)T, where SS is thermopower, ρ\rho is electrical resistivity, κ\kappa is thermal conductivity, and TT is temperature, respectively. All these parameters are interdependent and related to the electronic structure and charge carrier scattering of the material. To improve a material for thermoelectricity it is important to achieve high thermoelectric power factor (S2/ρS^2/\rho) and low thermal conductivity κ\kappa.

Quaternary materials, especially transition metal chalcogenides, are more tunable than binary and ternary compounds [11]. Quaternary spinel Cu6_6Fe4_4Sn12_{12}Se32_{32} features a large slope of density of states near the Fermi energy, favorable for high value of thermopower [12]. This compound crystallizes in cubic yet complex crystal structure of high symmetry with many atoms where high degeneracy of band extrema are commonly found [13, 14, 15, 16]. Numerous atoms in the high symmetric unit cell along with the Fe/Sn site disorder contribute to relatively low thermal conductivity κ\kappa in Cuy_yFe4_4Sn12_{12}Se32_{32} polycrystals [17, 18]. In addition, Cuy_yFe4_4Sn12_{12}Se32_{32} is stable at y=6y = 6 without impurity, where a large SS of ∼ 200 μ\muV K−1^{-1} and a rather low ρ\rho coexist at 340 K as well as low κ\kappa of ∼ 1.5 W K−1^{-1} m−1^{-1} at the room temperature.

Here we fabricated a series of Cu6−x_{6-x}Fe4+x_{4+x}Sn12_{12}Se32_{32} (x=0,1,2x = 0, 1, 2) single crystals to shed light on the electronic transport mechanism. X-ray photoemission spectroscopy (XPS) measurements were performed to investigate the valencies of interpolated cations. We observe a relatively high value of thermopower ∼ 0.18 mV/K at 300 K for x=0x = 0, which increases to ∼ 0.36 mV/K for x=2x = 2. The large discrepancy between the activation energy for electrical conductivity and for thermopower above 50 K points to the polaronic transport mechanism. With decreasing temperature, polaronic transport evolves into the dominant variable-range hopping (VRH) mechanism.

Experimental Details

Single crystals of Cu6−x_{6-x}Fe4+x_{4+x}Sn12_{12}Se32_{32} (x=0,1,2x = 0, 1, 2) were grown by melting stoichiometric mixture of Cu and Fe (4N, Alfa Aesar) powder, Sn and Se (5N, Alfa Aesar) shot [12]. The starting materials were mixed and loaded in 1 cm diameter Al2_2O3_3 crucibles, vacuum-sealed in quartz tubes, heated to 650 °C over 10 h and dwelled for 12 h, then cooled to 550 °C at a rate of 1 °C/h, and finally quenched in ice water. Crucible-size-limited single crystals were obtained with shining surface and confirmed by back-reflection Laue x-ray photographs.

The powder x-ray diffraction (XRD) data were taken on crushed crystals with Cu Kα_\alpha (λ=0.15418\lambda= 0.15418 nm) radiation of a Rigaku Miniflex powder diffractometer. The element analysis was performed using energy-dispersive x-ray spectroscopy (EDX) in a JEOL LSM-6500 scanning electron microscope. Multiple points on several samples were examined and the average actual chemical composition of samples are listed in Table 1. XPS measurements were carried out in an ultrahigh-vacuum (UHV) system with base pressure of ∼ 2×10−102 \times10^{-10} Torr and equipped with a hemispherical electron energy analyzer (SPECS, Phoibos 100) and twin anode x-ray source (SPECS, XR50). Al-Kα_\alpha (1486.6 eV) radiation was used at 15 kV and 20 mA. The angle between the analyzer and x-ray source is 45° and photoelectrons were collected along the sample surface normal. In order to remove potential surface contaminations and oxygen layers, each samples were cleaned in UHV by Ar+^{+} sputtering for 60 minutes under conditions of Ar gas pressure of 2×10−52 \times10^{-5} Torr and Ar+^{+} kinetic energy of 2 keV. XPS data was analyzed using Casa XPS and peak positions were calibrated using residual adventitious carbon C 1s at 284.8 eV. The electrical resistivity, thermopower, and thermal conductivity were measured in the quantum design thermal transport option (TTO) in PPMS-9 with standard four-probe method and in continuous measuring mode. The maximum heater power and period were set as 50 mW and 1430 s along with the maximum temperature rise of 3%. The crystals were cut and polished into rectangular bars with typical dimensions of 4×1×0.84 \times1 \times0.8 mm3^3. Epoxy and copper leads were used for TTO contacts. The relative error in our measurement for thermopower was below 5% based on the Ni standard measured under identical conditions. Sample dimensions were measured by an optical microscope Nikon SMZ-800 with resolution of 10 µm.

NominalActualaa (Å)
Cu6_6Fe4_4Sn12_{12}Se32_{32}Cu5.9(2)_{5.9(2)}Fe4.0(1)_{4.0(1)}Sn13(1)_{13(1)}Se32(1)_{32(1)}10.774
Cu5_5Fe5_5Sn12_{12}Se32_{32}Cu5.3(2)_{5.3(2)}Fe3.6(2)_{3.6(2)}Sn14(1)_{14(1)}Se31(1)_{31(1)}10.786
Cu4_4Fe6_6Sn12_{12}Se32_{32}Cu5.1(2)_{5.1(2)}Fe4.2(2)_{4.2(2)}Sn13(1)_{13(1)}Se32(1)_{32(1)}10.796

Table I. The actual chemical composition and lattice parameter aa for Cu6−x_{6-x}Fe4+x_{4+x}Sn12_{12}Se32_{32} (x=0x = 0, 1, 2).

Results and Discussions

Figure 1(a) shows the crystal structure of selenospinel Cu6−x_{6-x}Fe4+x_{4+x}Sn12_{12}Se32_{32} (x=0x = 0, 1, 2). The spinel-type compounds have a general formula of AB2_2X4_4, where A and B are metals encapsulated by tetrahedra and octahedra, and X is oxygen or chalcogen elements. Herein, the unit cell contains numerous 54 atoms [Figure 1(a)] with A-site Cu-deficiencies; Cu atoms are coordinated tetrahedrally by X = Se; B-site Fe/Sn atoms are randomly distributed in the same crystallographic site and are coordinated octahedrally by Se [12]. We note that Cuy_yFe4_4Sn12_{12}Se32_{32} is stable at y=6y = 6 [17, 18]. Figure 1(b) shows the structural refinement of powder XRD for Cu6−x_{6-x}Fe4+x_{4+x}Sn12_{12}Se32_{32} (x=0x = 0, 1, 2), indicating that all reflections can be well indexed in the Fd3ˉ\bar{3}m space group. No impurity peaks are observed, confirming higher tolerance of Cu deficiencies in Cu6−x_{6-x}Fe4+x_{4+x}Sn12_{12}Se32_{32}. The determined lattice parameter aa increases slightly from 10.774(2) to 10.796(2) Å with increasing xx [Figure 1(c)]. The value of aa is close to the previous result for Cu6_6Fe4_4Sn12_{12}Se32_{32} [17], whilst the rise in xx suggests larger ionic size of Fe than Cu ions. To investigate the valences of interpolated cations, we measured the XPS where all three samples show similar

Crystal structure, powder x-ray diffraction patterns, and evolution of lattice parameter

Figure 1. (Color online) (a) Crystal structure of selenospinel Cu6−x_{6-x}Fe4+x_{4+x}Sn12_{12}Se32_{32} (x=0x = 0, 1, 2). (b) Powder x-ray diffraction patterns and (c) the evolution of lattice parameter aa.

valence states for each elements. As shown in Figure 2(a), the Cu 2p3/2_{3/2} binding energy of 932.9 eV suggest the Cu in metallic state Cu0^0 or Cu1+^{1+}. Auger spectra shows Cu LMM at kinetic energy of 916.8 eV [Figure 2(b)], which exclude the possibility of Cu0^0 and directly confirm Cu1+^{1+} state. The highest intensity of Fe 2p are coincidentally overlapping with Sn 3p3/2_{3/2} [Figure 2(c)]. By deconvoluting each components, the binding energy of Fe 2p3/2_{3/2} is located at 708.2 eV and shifted toward high binding energy about 1.5 eV when compare to the binding energy of pure metal Fe 2p3/2_{3/2} at 706.7 eV [Figure 2(d)]. Compared to the binding energy of Sn 3d5/2_{5/2} at 485.2 eV for pure metal, the binding energy of Sn 3d5/2_{5/2} is located at 486.5 eV [Figure 2(e)] and shifted toward high binding energy about 1.3 eV. These observations suggest that the electropositive Fe [mostly Fe2+^{2+}] and Sn [mostly Sn4+^{4+}] ions in these compounds [19, 20]. Moreover, the binding energy of Se 3d5/2_{5/2} is located at 54.5 eV and shifted toward to lower binding energy about -0.9 eV when compare to the binding energy of pure metal Se 3d5/2_{5/2} at 55.4 eV [Figure 2(f)], suggesting that the electronegative Se [mostly Se2−^{2-}] ions in the compounds. Subtle difference in binding energy of Sn and Se indicates slight local structural change induced by Fe substitution, calling for further synchrotron x-ray diffraction study on the site occupancies.

Typical XPS spectra for the indicated elements of Cu6Fe4Sn12Se32, Cu5Fe5Sn12Se32, and Cu4Fe6Sn12Se32 samples

Figure 2. (Color online). Typical XPS spectra for the indicated elements of Cu6_6Fe4_4Sn12_{12}Se32_{32} (bottom), Cu5_5Fe5_5Sn12_{12}Se32_{32} (middle), and Cu4_4Fe6_6Sn12_{12}Se32_{32} (top) samples.

Temperature dependence of electrical resistivity ρ(T)\rho(T) for Cu6−x_{6-x}Fe4+x_{4+x}Sn12_{12}Se32_{32} (x=0x = 0, 1, 2) is depicted in Figure 3(a), showing an obvious semiconducting behavior. The value of room temperature resistivity (ρ300K\rho_{300K}) is about 2.2×10−4 Ω m2.2 \times10^{-4}\ \Omega\ \mathrm{m} for Cu6_6Fe4_4Sn12_{12}Se32_{32}, which monotonically increases to 8.0×10−48.0 \times10^{-4} and 5.1×10−3 Ω m5.1 \times10^{-3}\ \Omega\ \mathrm{m} for x=1x=1 and 2, respectively. In such a complex disordered system, the electronic conduction mechanism is of high interest. In general, three typical models are considered to describe the semiconducting behavior: (i) thermally activated model ρ(T)=ρ0exp⁡(Eρ/kBT)\rho(T)=\rho_{0}\exp(E_{\rho}/k_{\mathrm{B}}T), where EρE_{\rho} is activation energy and kBk_{\mathrm{B}} is the Boltzmann constant; (ii) adiabatic small polaron hopping model ρ(T)=ATexp⁡(Eρ/kBT)\rho(T)=AT\exp(E_{\rho}/k_{\mathrm{B}}T); and (iii) variable-range hopping (VRH) model ρ(T)=ρ0exp⁡(T0/T)ν\rho(T)=\rho_{0}\exp(T_{0}/T)^{\nu} [21, 22, 23, 24, 25, 26, 27, 28], where T0T_{0} is a characteristic temperature and is related to density of states available at the Fermi level and carrier localization length, ν\nu depends on the dimensionality. To understand the transport mechanism, it is necessary to fit the temperature-dependent resistivity curves based on these three formulas.

Temperature dependence of electrical resistivity and fitting curves

Figure 3. (Color online). (a) Temperature dependence of electrical resistivity ρ(T)\rho(T) for the indicated samples. (b) ln⁡(ρ/T)\ln(\rho/T) versus 1000/T1000/T curves fitted by the adiabatic small polaron hopping model ρ(T)=ATexp⁡(Eρ/kBT)\rho(T)=AT\exp(E_{\rho}/k_{\mathrm{B}}T). (c) ln⁡(−δln⁡ρ/δln⁡T)\ln(-\delta\ln\rho/\delta\ln T) versus ln⁡(T)\ln(T), where the slopes give the exponents for variable-range hopping model ρ(T)=ρ0exp⁡(T0/T)ν\rho(T)=\rho_{0}\exp(T_{0}/T)^{\nu}.

Sampleν\nuβ\betaΘD\Theta_{\mathrm{D}}νs\nu_{\mathrm{s}}
Cu6_6Fe4_4Sn12_{12}Se32_{32}0.38(2)87(1)107(1)1030
Cu5_5Fe5_5Sn12_{12}Se32_{32}0.29(2)53(1)125(1)1210
Cu4_4Fe6_6Sn12_{12}Se32_{32}0.44(3)81(1)109(1)1050

Table 2. Variable range hopping exponents and parameters from specific heat measurements for indicated samples.

Figure 3(b) exhibits the fitting result of the adiabatic small polaron hopping model from 50 to 350 K. The extracted activation energy EρE_{\rho} is about 32.3(4) meV for x=0x=0, in agreement with the previous result [17], and gradually increases to 37.3(5) and 69.8(9) meV for x=1x=1 and 2, respectively. However, the ρ(T)\rho(T) curves can also be fitted

by the thermally activated model (see discussion below). With decreasing temperature, the resistivity increases abruptly below 50 K, indicating a strong localization regime with a VRH conductivity, ρ(T)=ρ0exp⁡(T0/T)ν\rho(T)=\rho_{0}\exp(T_{0}/T)^{\nu}. We use the logarithmic-derivative method for an accurate estimate of the ν\nu exponents, i.e. plot ln⁡(−δln⁡ρ/δln⁡T)\ln(-\delta\ln\rho/\delta\ln T) versus ln⁡(T)\ln(T) [Figure 3(c)], where the slopes of the fit give the values of exponent ν\nu. The obtained ν\nu (0.29 ∼\sim 0.44) situates between 0.25 for the Mott’s VRH and 0.5 for the Efros-Shklovskii-type VRH conductivity.

To distinguish the thermally activated model and polaron hopping model, we further measured temperature-dependent thermopower S(T)S(T). The S(T)S(T) exhibits large positive values in the whole temperature range [Fig. 4(a)], indicating dominant hole-type carriers. The room temperature value S300KS_{300\mathrm{K}} of x=0x=0 is about 180 μV K−1\mu\mathrm{V}\ \mathrm{K}^{-1}; the large value of SS is attributed to the small number of holes. The rise in xx further increases thermopower, reaching S300K=360 μV K−1S_{300\mathrm{K}}=360\ \mu\mathrm{V}\ \mathrm{K}^{-1} for x=2x=2. The S(1000/T)S(1000/T) curves of all samples are shown in Figure 4(b) and can be fitted with the equation S(T)=(kB/e)(α+ES/kBT)S(T)=(k_{\mathrm{B}}/e)(\alpha+E_{\mathrm{S}}/k_{\mathrm{B}}T) from 50 to 350 K [22], where ESE_{\mathrm{S}} is activation energy and α\alpha is constant. The obtained activation energy for thermopower, ESE_{\mathrm{S}} (3.2 ∼\sim 11.5 meV), are much smaller than those for conductivity, EρE_{\rho} (32.3 ∼\sim 69.8 meV), as shown in Figure 4(c). This large discrepancy between ESE_{\mathrm{S}} and EρE_{\rho} indicates a polaron transport mechanism of carriers. According to the polaron model, the ESE_{\mathrm{S}} is the energy required to activate the hopping of carriers, while EρE_{\rho} is the sum of the energy needed for the creation of carriers and activating the hopping of carriers [22]. Therefore, ESE_{\mathrm{S}} is smaller than EρE_{\rho}. The weak temperature-dependent S(T)S(T) at high temperatures also supports the small polaron conduction, however, the S(T)S(T) changes its slope at lower temperatures where it evolves into the VRH conduction.

Temperature dependence of thermopower and evolution of activation energies

Figure 4. (Color online) (a) Temperature dependence of thermopower S(T)S(T) for the indicated samples. (b) S(T)S(T) vs 1000/TT curves fitted using S(T)=(kB/e)(α+ES/kBT)S(T) = (k_{\mathrm{B}}/e)(\alpha+ E_{\mathrm{S}}/k_{\mathrm{B}}T) from 50 to 350 K. (c) The evolution of EρE_{\rho} and ESE_{\mathrm{S}}.

Figure 5(a) shows the temperature dependence of thermal conductivity κ(T)\kappa(T) for Cu6−x_{6-x}Fe4+x_{4+x}Sn12_{12}Se32_{32} (x=0x = 0, 1, 2). In general, κtotal=κe+κL\kappa_{\mathrm{total}} = \kappa_{\mathrm{e}} + \kappa_{\mathrm{L}}, consists of the electronic part κe\kappa_{\mathrm{e}} and the phonon term κL\kappa_{\mathrm{L}}. Herein, the κe\kappa_{\mathrm{e}} estimated from the Wiedemann-Franz law is negligibly small due to large electrical resistivity, indicating a predominantly phonon contribution. At room temperature, the κ\kappa shows relatively low value of 1.32 ∼\sim 1.57 W K−1^{-1} m−1^{-1}, arising from its structural complexity, i.e., large Cu deficiencies and Fe/Sn site disorder as phonon scattering centers. Interestingly, the low-temperature κ(T)\kappa(T) follows a quasi-linear TT-dependence [inset in Figure 5(a)]. This deviates from the common κ∼T3\kappa\sim T^3 usually observed in bulk crystals or thin films [29, 30], implying nanostructural differences that are induced by different vacancies in particular grains and associated phonon frequency changes [31, 32]. The Fe-substituted samples show slightly larger values of κ(T)\kappa(T) when compared to Cu6_6Fe4_4Sn12_{12}Se32_{32}. This also leads to the increase in phonon mean free path lκl_{\kappa} [Figure 5(b)] estimated from the heat capacity and thermal conductivity via κL=Cpvslκ/3\kappa_{\mathrm{L}} = C_{\mathrm{p}}v_{\mathrm{s}}l_{\kappa}/3, where κL\kappa_{\mathrm{L}} is the lattice thermal conductivity, CpC_{\mathrm{p}} is the heat capacity, vsv_{\mathrm{s}} is the average sound speed calculated via Debye temperature [33, 34], and lκl_{\kappa} is the phonon mean free path.

Temperature dependence of thermal conductivity, phonon mean free path, figure of merit, and specific heat

Figure 5. (Color online) Temperature dependence of (a) thermal conductivity κ(T)\kappa(T), (b) phonon mean free path lκl_{\kappa}, (c) the figure of merit ZTZT, and (d) specific heat CpC_{\mathrm{p}} divided by T3T^3 for the indicated samples.

The figure of merit ZTZT shows maxima around the room temperature, reaching ∼0.03\sim0.03 for Cu6_6Fe4_4Sn12_{12}Se32_{32} at 300 K [Figure 5(c)]. Note that the increase of κe\kappa_{\mathrm{e}} above 200 K can be due to radiation and as a consequence the real ZTZT can be ∼0.044\sim0.044 for Cu6_6Fe4_4Sn12_{12}Se32_{32} at 300 K. Although the thermopower is enhanced by Fe-substitution, the electrical resistivity and thermal conductivity are also increased in Fe-substituted samples. Further efforts in nanostructuring of single crystal alloys and/or doping with heavier element would be helpful to obtain better thermoelectricity performance at room temperature.

Temperature dependence of specific heat CpC_{\mathrm{p}} divided by T3T^3 of the indicated samples is depicted in Figure 5(d). Two interesting features are observed: a hump around 25(5) K and, it diverges from the Debye T3T^3 law at low temperatures, similar to those observed in amorphous solids arising from the enhanced density of states of acoustic phonons caused by the disordered structure [35, 36, 37, 38, 39, 40]. The coefficients β=53(1)∼87(1)\beta= 53(1) \sim87(1) mJ mol−1^{-1} K−4^{-4} were estimated from the local minimum values of Cp/T3C_{\mathrm{p}}/T^3, as shown by dashed lines in Figure 4(d). The next derived Debye temperature ΘD=107(1)∼125(1)\Theta_{\mathrm{D}} = 107(1) \sim125(1) K by using the equation ΘD=[12π4NR/(5β)]1/3\Theta_{\mathrm{D}} = [12\pi^4NR/(5\beta)]^{1/3}, which implies the average sound velocity of vs≈1030∼1210v_{\mathrm{s}} \approx1030 \sim1210 m s−1^{-1} [Table II]. We should note that the large values of γ\gamma can also be from magnetic spin glassy state in this high Fe-content disordered system. Further synchrotron local structural (pair distribution function analysis) and neutron scattering experiments will be helpful to unveil its origin.

Conclusions

In summary, we synthesized and studied single crystals of Cu6−xFe4+xSn12Se32\mathrm{Cu}_{6-x}\mathrm{Fe}_{4+x}\mathrm{Sn}_{12}\mathrm{Se}_{32} (x=0x = 0, 1, 2). Electronic transport mechanism on cooling is dominated by polaron effect down to 50 K. On further cooling the VRH mechanism dominates. The Fe substitution enhances the thermopower possibly due to an increase of the slope at and shift of the density of states near the Fermi level. The figure of merit ZTZT reaches ∼\sim 0.03 - 0.044 around the room temperature for Cu6Fe4Sn12Se32\mathrm{Cu}_{6}\mathrm{Fe}_{4}\mathrm{Sn}_{12}\mathrm{Se}_{32}, calling for further carrier optimization.

Acknowledgements

Work at BNL is supported by the Office of Basic Energy Sciences, Materials Sciences and Engineering Division, U.S. Department of Energy (DOE) under Contract No. DE-SC0012704. A portion of this work was performed at the National High Magnetic Field Laboratory, which was supported by the National Science Foundation Cooperative Agreement No. DMR-1644779 and the State of Florida.

References

  1. [1]S. Roychowdhury, M. Samanta, A. Banik, and K. Biswas, Thermoelectric energy conversion and topological materials based on heavy metal chalcogenides, J. Solid State Chem. 275, 103 (2019).DOI
  2. [2]Y. Ouyang, Z. Zhang, D. Li, J. Chen, and G. Zhang, Emerging theory, materials, and screening methods: new opportunities for promoting thermoelectric performance, Ann. Phys. 531, 1800437 (2019).DOI
  3. [3]A. Banik, S. Roychowdhury, and K. Biswas, The journey of tin chalcogenides towards high-performance thermoelectrics and topological materials, Chem. Comm. 54, 6573 (2018).DOI
  4. [4]C. Petrovic, J. W. Kim, S. L. Bud’ko, A. I. Goldman, P. C. Canfield, W. Choe, and G. J. Miller, Anisotropy and large magnetoresistance in the narrow-gap semiconductor FeSb2, Phys. Rev. B 67, 155205 (2003).DOI
  5. [5]C. Petrovic, Y. Lee, T. Vogt, N. D. Lazarov, S. L. Budko, and P. C. Canfield, Kondo insulator description of spin state transition in FeSb2, Phys. Rev. B 72, 045103 (2005).DOI
  6. [6]R. Hu, K. Wang, H. Ryu, H. Lei, E. S. Choi, M. Uhlarz, J. Wosnitza, and C. Petrovic, Electronic Griffiths Phase in the Te-Doped Semiconductor FeSb2, Phys. Rev. Lett. 109, 256401 (2012).pubmed.ncbi.nlm.nih.gov/23368482
  7. [7]A. Bentien, S. Johnsen, G. K. H. Madsen, B. B. Iversen, and F. Steglich, Colossal Seebeck coefficient in strongly correlated semiconductor FeSb2, Europhys. Lett. 80, 17008 (2007).
  8. [8]Q. Jie, R. Hu, E. Bozin, A. Llobet, I. Zaliznyak, C. Petrovic, and Q. Li, Electronic thermoelectric power factor and metal-insulator transition in FeSb2, Phys. Rev. B 86, 115121 (2012).
  9. [9]H. Takahashi, R. Okazaki, S. Ishiwata, H. Taniguchi, A. Okutani, M. Hagiwara, and I. Terasaki, Colossal Seebeck effect enhanced by quasi-ballistic phonons dragging massive electrons in FeSb2, Nat. Commun. 7, 12732 (2016).
  10. [10]Q. Du, M. Abeykoon, Y. Liu, G. Kotliar, and C. Petrovic, Low-temperature thermopower in CoSbS, Phys. Rev. Lett. 123, 076602 (2019).DOI
  11. [11]C. Han, Q. Sun, Z. Li, and S. Xue, Thermoelectric Enhancement of Different Kinds of Metal Chalcogenides, Adv. Energy Mater. 6, 1600498 (2016).DOI
  12. [12]K. Suekuni, K. Tsuruta, H. Fukuoka, and M. Koyano, Structural and thermoelectric properties of Cu6Fe4Sn12Se32 single crystal, J. Alloys Compd. 564, 91 (2013).DOI
  13. [13]F. J. DiSalvo, Thermoelectric Cooling and Power Generation, Science 285, 703 (1999).pubmed.ncbi.nlm.nih.gov/10426986
  14. [14]P. Larson, S. D. Mahanti, M. G. Kanatzidis, Electronic structure and transport of Bi2Te3 and BaBiTe3, Phys. Rev. B 61, 8162 (2000).DOI
  15. [15]J. R. Sootsman, D. Y. Chung, M. G. Kanatzidis, New and old concepts in thermoelectric materials, Angew. Chemie Int. Ed. 48, 8616 (2009).DOI
  16. [16]G. J. Snyder and E. S. Toberer, Complex thermoelectric materials, Nat. Mater. 7, 105 (2008).
  17. [17]K. Suekuni, K. Tsuruta, T. Ariga, M. Koyano, Variable-range-hopping conduction and low thermal conductivity in chalcogenide spinel Cu3Fe4Sn12X32 (X = S, Se), J. Appl. Phys. 109, 083709 (2011).
  18. [18]K. Suekuni, M. Koyanii, H. Nishiate, M. Ohta, A. Yamamoto, M. Koyano, Thermoelectric Properties of Selenospinel Cu6Fe4Sn12Se32, J. Electron. Mater. 41, 1130 (2012).DOI
  19. [19]T. Yamashita and P. Hayes, Analysis of XPS spectra of Fe2+ and Fe3+ ions in oxide materials, Appl. Surf. Sci. 254, 2441 (2008).DOI
  20. [20]K. Watanable, T. Kawaguchi, and S. Aikawa, p-type conversion of distorted SnOx thin film by mild thermal annealing treatment in pure N2 environment, AIP Advances 12, 105012 (2022).
  21. [21]B. Raquet, M. N. Baibich, J. M. Broto, H. Rakoto, S. Lambert, and A. Maignan, Hopping conductivity in one-dimensional Ca3Co2O6 single crystals, Phys. Rev. B 65, 104442 (2002).
  22. [22]I. G. Austin, and N. F. Mott, Polarons in crystalline and non-crystalline materials, Adv. Phys. 50, 757 (2001).
  23. [23]Y. Liu, C.-J. Kang, E. Stavitski, Q. Du, K. Attenkofer, and C. Petrovic, Polaronic transport and thermoelectricity in Fe1−xCoxSb2S4 (x = 0, 0.1, and 0.2), Phys. Rev. B 97, 155202 (2012).
  24. [24]Y. Liu, Z. Hu, M. Abeykoon, E. Stavitski, K. Attenkofer, E. D. Bauer, and C. Petrovic, Polaronic transport and thermoelectricity in Mn3Si2Te6 single crystals, Phys. Rev. B 103, 245122 (2021).DOI
  25. [25]Y. Liu, A. M. M. Abeykoon, N. Aryal, D. Graf, Z. Hu, W. Jin, and C. Petrovic, Thermal transport properties of IrSbSe, Phys. Rev. B 106, 195116 (2022).DOI
  26. [26]Y. Liu, S. B. Zhang, L. J. Li, W. J. Lu, B. C. Zhao, P. Tong, W. H. Hong, S. Lin, Y. N. Huang, Z. H. Huang, S. G. Tan, and Y. P. Sun, Synthesis, structure and properties of the new layered manganese oxyselenide Sr₂Fe₂MnSe₂O, J. Alloys Compd. 580, 211 (2013).
  27. [27]R. Hu, V. F. Mitrovic, and C. Petrovic, Anisotropy in the magnetic and electrical transport properties of Fe1−xCrxSb2, Phys. Rev. B 76, 115105 (2007).DOI
  28. [28]J. R. Friedman, Y. Zhang, P. Dai, and M. P. Sarachik, Magnetic-field-induced crossover from Mott variable-range hopping to weakly insulating behavior, Phys. Rev. B 53, 9528 (1996).pubmed.ncbi.nlm.nih.gov/9982491
  29. [29]Thermophysical Properties of Matter; Touloukian, Y. S., Ed.; IFI/Plenum: New York, 1970–1979.
  30. [30]A. D. McConnell, K. E. Goodson, Thermal conduction in silicon micro- and nanostructures, Annu. Rev. Heat Transfer 14, 129 (2005).DOI
  31. [31]Z. Wang, J. E. Alaniz, W. Jang, J. E. Garay, and C. Dames, Thermal conductivity of nanocrystalline silicon: importance of grain size and frequency-dependent mean free paths, Nano Lett. 11, 2206 (2011).DOI
  32. [32]H. Zhao, M. Pokharel, G. Zhu, S. Chen, K. Lukas, Q. Jie, C. Opeil, G. Chen, and Z. Ren, Dramatic thermal conduction reduction by nanostructures for large increase in thermoelectric figure-of-merit of FeSb2, Appl. Phys. Lett. 99, 163101 (2011).DOI
  33. [33]Orson L. Anderson, A simplified method for calculating the Debye temperature from elastic constants, J. Phys. Chem. Solids 24, 909 (1963).DOI
  34. [34]Q. Du, X. Tong, Y. Liu, and C. Petrovic, Suppression of thermal conductivity and electronic correlations in Fe1−xRuxSb2 (0 ≤ x ≤ 0.6), Appl. Phys. Lett. 118, 171904 (2021).
  35. [35]R. C. Zeller and R. O. Pohl, Thermal conductivity and specific heat of noncrystalline solids, Phys. Rev. B 4, 2029 (1971).DOI
  36. [36]L. Gil, M. A. Ramos, A. Bringer, and U. Buchenau, Low-temperature specific heat and thermal conductivity of glasses, Phys. Rev. Lett. 70, 182 (1993).pubmed.ncbi.nlm.nih.gov/10053723
  37. [37]D. R. Queen, X. Liu, J. Karel, T. H. Metcalf, and F. Hellman, Excess specific heat in evaporated amorphous silicon, Phys. Rev. Lett. 110, 135901 (2013).pubmed.ncbi.nlm.nih.gov/23581344
  38. [38]Y. Ishii, Y. Ouchi, S. Kawaguchi, H. Ishibashi, Y. Kubota, and S. Mori, Glassy anomalies in the lattice heat capacity of a crystalline solid caused by ferroelectric fluctuations, Phys. Rev. Mater. 3, 084414 (2019).DOI
  39. [39]H. Shintani and H. Tanaka, Universal link between the boson peak and transverse phonons in glass, Nat. Mater. 7, 870 (2008).DOI
  40. [40]A. I. Chumakov, G. Monaco, A. Monaco, W. A. Crichton, A. Bosak, R. Rüffer, A. Meyer, F. Kargl, L. Comez, D. Fioretto, H. Giefers, S. Rositsch, G. Wortmann, M. H. Manghnani, A. Hushur, Q. Williams, J. Balogh, K. Parliński, P. Jochym, and P. Piekarz, Equivalence of the Boson peak in glasses to the transverse acoustic van Hove singularity in crystals, Phys. Rev. Lett. 106, 225501 (2011).

Paper details

vlnc:W_ebjv8ven
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents