# Fileset

[PhysRevMaterials.7.085402.pdf](https://mdr.nims.go.jp/filesets/b51ceee6-7ab0-4f8b-ba49-5d8a08106b8c/download)

## Creator

Kouta Kazama, Masato Sakano, Kohei Yamagami, Takuo Ohkochi, Kyoko Ishizaka, [Terumasa Tadano](https://orcid.org/0000-0002-8132-2161), [Yusuke Kozuka](https://orcid.org/0000-0001-7674-600X), Hidetoshi Yoshizawa, [Yoshihiro Tsujimoto](https://orcid.org/0000-0003-2140-3362), [Kazunari Yamaura](https://orcid.org/0000-0003-0390-8244), Jun Fujioka

## Rights

[In Copyright](http://rightsstatements.org/vocab/InC/1.0/)

## Other metadata

[Charge transport and thermopower in the electron-doped narrow gap semiconductor Ca1−xLaxPd3O4](https://mdr.nims.go.jp/datasets/e7d0cc5e-9530-4573-bd2b-80e37744d507)

## Fulltext

Charge transport and thermopower in the electron-doped narrow gap semiconductor ${\rm Ca}_{1-x}{\rm La}_x{\rm Pd}_3{\rm O}_4$PHYSICAL REVIEW MATERIALS 7, 085402 (2023)Charge transport and thermopower in the electron-doped narrow gap semiconductorCa1−xLaxPd3O4Kouta Kazama,1 Masato Sakano,2 Kohei Yamagami,3 Takuo Ohkochi,3 Kyoko Ishizaka,2 Terumasa Tadano ,4Yusuke Kozuka,5 Hidetoshi Yoshizawa ,1 Yoshihiro Tsujimoto ,5 Kazunari Yamaura,5 and Jun Fujioka1,61Graduate School of Science and Technology, University of Tsukuba, Tsukuba, Ibaraki 305-8573, Japan2Quantum-Phase Electronics Center and Department of Applied Physics, University of Tokyo, Bunkyo-ku, Tokyo 113-8656, Japan3Japan Synchrotron Radiation Research Institute, Kouto Sayo, Hyogo 679-5198, Japan4Research Center for Magnetic and Spintronic Materials (CMSM), National Institute for Material Science (NIMS), Sengen,Tsukuba 305-0047, Japan5Research Center for Materials Nanoarchitechtonics (MANA), National Institute for Material Science (NIMS),Namiki, Tsukuba 305-0044, Japan6Department of Materials Science, University of Tsukuba, Tsukuba, Ibaraki 305-8573, Japan(Received 13 April 2023; accepted 24 July 2023; published 11 August 2023)We have investigated the charge transport property, thermoelectric effect, and electronic state forthe electron-doped narrow gap semiconductor Ca1−xLaxPd3O4 by means of the transport measurement,optical/photoemission spectroscopy, and ab initio calculation. The high-quality polycrystalline samples ofCa1−xLaxPd3O4 were synthesized by using the high-pressure synthesis technique. In the undoped system x = 0,the optical conductivity spectra show a charge gap of about 0.1 eV, which is qualitatively consistent with theresults of ab initio calculation. The electron doping causes the rapid reduction of resistivity ρxx , resulting in themetallic state at the doping level of as small as x = 0.01. On the other hand, the magnitude of Seebeck coefficient|S| moderately decreases with increasing x. Consequently, the power factor S2/ρxx reaches about 7 µW/K2cm at350 K at x = 0.03, which is much higher than that of hole-doped analog of CaPd3O4. Combined with the resultsof photoemission spectra and ab initio calculation, the large power factor in the present material likely originatesfrom the relatively low resistivity, probably high electron mobility (∼30 cm2/Vs at 350 K), due to the dispersivePd 4dx2−y2 conduction band.DOI: 10.1103/PhysRevMaterials.7.085402I. INTRODUCTIONThe Seebeck effect is a typical thermoelectric effect, whichhas been utilized as a principle of commercial thermoelec-tric devices. The power factor is a measure of efficiency forthe energy conversion, which is defined as σS2 with S andσ being the Seebeck coefficient and electrical conductivity,respectively. The semimetals or narrow gap semiconductorsare known as typical thermoelectric materials with potentiallyhigh power factor [1–12]. In particular, they often show themetallic state with small carrier density and high carrier mo-bility, which is advantageous for simultaneously realizing thelarge Seebeck coefficient and high electrical conductivity. Forexample, there are a number of narrow gap semiconductorswith power factor more than 30 µW/K2cm at room tempera-ture, regardless of their quite different band structure [2,3,6,7].In fact, some of them have already been incorporated intopractical thermoelectric devices. However, they often includetoxic elements or are chemically unstable at the operatingtemperature above room temperature.In this regard, the semimetals or narrow gap semiconduc-tors of transition metal oxide have a potential to overcomethese issues. The APd3O4 (A = Ca and Sr) is a candidateof narrow gap semiconductor with potentially high powerfactor. As shown in Fig. 1(a), this material crystallizes in theNaPt3O4-type crystal structure of cubic symmetry, which iscomposed of the three-dimensional network of corner sharedPdO4 plaquette [13,14]. The Pd ion takes nominally divalentstate in the 4d8-electronic configuration, and the electronicstates near the Fermi energy EF is mainly composed of thePd 4d eg state. So far, there are several theoretical proposalsabout the possible electronic structure near EF. On the basisof electronic structure calculation, the narrow gap opening hasbeen proposed [15,16], but the possibility of Dirac semimetal-lic state protected by the crystal symmetry is also argued[17–19]. On the other hand, there are few experimental studieson the electronic state near EF [20], which has not yet beenelucidated so far.Nevertheless, the charge transport and thermoelectric prop-erty have been extensively explored in the hole-doped analogof CaPd3O4 [14,21–23]. For example, the relatively largeSeebeck coefficient more than 100 µV/K is observed at roomtemperature in the metallic state of Ca1−xLixPd3O4 [21]. Un-fortunately, the resistivity ρxx is not sufficiently low due tothe small carrier mobility about 1−10 cm2/Vs. Consequently,the power factor is at most about 1.6 µW/K2cm at x = 0.4.A strategy to further improve the thermoelectric property isthe electron doping into CaPd3O4. However, the electronicproperty for the electron-doped analog of CaPd3O4 has beenrarely explored due to the difficulty of preparing high quality2475-9953/2023/7(8)/085402(6) 085402-1 ©2023 American Physical Societyhttps://orcid.org/0000-0002-8132-2161https://orcid.org/0000-0003-0436-6029https://orcid.org/0000-0003-2140-3362http://crossmark.crossref.org/dialog/?doi=10.1103/PhysRevMaterials.7.085402&domain=pdf&date_stamp=2023-08-11https://doi.org/10.1103/PhysRevMaterials.7.085402KOUTA KAZAMA et al. PHYSICAL REVIEW MATERIALS 7, 085402 (2023)FIG. 1. (a) The crystal structure of CaPd3O4 [47]. The dashedline highlights the PdO4 plaquette, which forms a three-dimensionalnetwork by sharing the corner oxygen ions. (b) Photograph of poly-crystalline sample of CaPd3O4 synthesized by the solid state reactionunder 2.5 GPa and 1300 ◦C. (c) Powder x-ray diffraction (XRD)for Ca1−xLaxPd3O4 with x = 0 − 0.07 at room temperature. Eachpattern is offset for clarity. The triangles indicate the diffraction fromimpurity phase (PdO). (d) The lattice constant derived from XRD forCa1−xLaxPd3O4.samples. In this study, we synthesized the polycrystalline sam-ple of electron-doped Ca1−xLaxPd3O4 by means of the highpressure synthesis technique and explored their charge trans-port, thermoelectric property, and electronic structure. Wefound that the doping level of x = 0.01 induces the metallicstate with relatively high electrical conductivity and Seebeckcoefficient, resulting in the power factor of about 7 µW/K2cmat 350 K.II. EXPERIMENTAL METHODS AND ELECTRONICSTRUCTURE CALCULATIONTo obtain polycrystalline samples of Ca1−xLaxPd3O4 withminimal impurity phases such as PdO and Pd, we employedthe high-pressure synthesis techniques. The starting materi-als CaO, PdO, and La2O3 were reacted under 2.5 GPa and1300 ◦C by using a cubic anvil type facility. The samplesare in the form of densely packed pellets with a few mil-limeters in size as shown in Fig. 1(b) [see also Fig. S1][24]. We performed the x-ray diffraction by using com-mercial diffractometer (Rigaku MiniFlex) with Cu-Kα1 and-Kα2 as sources. The results of x-ray diffraction patternsshow that the contamination of PdO and Pd is not signifi-cant [see Fig. 1(c)] and the lattice constant slightly increaseswith increasing x [see Fig. 1(d)]. These are suitable forexploring the fine doping dependence of charge transport,thermoelectric property, and optical reflectivity. The nominaldoping level x at the time of synthesis is nearly identical tothe results of energy dispersive x-ray spectroscopy (EDX)[see Fig. S1(e)]. Resistivity measurements were performedby the four-probe method in a physical property measure-ment system (Quantum Design). The Seebeck coefficientwas measured while applying the temperature gradient byusing a chip heater attached on one side of the sample[see Fig. S4]. The temperature gradient was measured bycommercial thermocouples (chromel-constantan). The heaterpower was set so that the typical temperature difference be-tween the thermocouple was about 1 K.The optical reflectivity spectra were measured in the geom-etry of nearly normal incidence by using a Fourier transformspectrometer (grating-type monochromator) equipped with amicroscope in the energy region of 0.02−0.7 eV (0.5−5 eV).In the region of 4−30 eV, we carried out the measurementat room temperature with use of synchrotron radiation atUV-SOR, Institute for Molecular Science (Okazaki). The opti-cal conductivity spectra were derived by the Kramers-Kroniganalysis. For the analysis, we adopted the Hagen-Rubens-typeextrapolation below 0.02 eV and ω4-type extrapolation above30 eV.High-resolution photoemission spectroscopy (PES) mea-surements were performed under an ultrahigh vacuum lessthan 2 × 10−8 Pa by using a He discharge lamp (hν =21.2 eV). The total energy resolution was set to 15 meV. Sam-ples were fractured in situ and measured at 16 K. Core-levelPES measurements were performed by using a monochroma-tized soft x-ray source (hν = 900 eV) at BL25SU of SPring-8[25,26]. The total energy resolution was set to 110 meV.Samples were fractured in situ and measured at 20 K. TheFermi level was determined by measuring the Fermi cutoff ofpolycrystalline gold electrically connected to each sample.We performed electronic structure calculations based ondensity functional theory using the Vienna ab initio simula-tion package (VASP) [27], which implements the projectoraugmented wave (PAW) method [28]. The plane-wave kineticenergy cutoff was set to 400 eV, and 6 × 6 × 6 �-centeredk points were used for the Brillouin zone integration. TheHeyd-Scuseria-Ernzerhof (HSE06) screened hybrid func-tional [29–31] was used, and the spin-orbit coupling wasincluded in the calculation. To draw HSE06 band structuresalong high-symmetry lines, we performed Wannier interpola-tion using the Wannier90 code [32], where the Kohn-Shamstates in the energy range of [−9:4] eV relative to the valenceband maximum were used to construct maximally localizedWannier functions for the Pd 4d and O 2p orbitals. For thesecalculations, we assumed the crystal structure determined bythe experiment.III. RESULTS AND DISCUSSIONFigure 2(a) shows the temperature dependence of resis-tivity for Ca1−xLaxPd3O4 with x = 0. The resistivity showsthe insulating behavior and divergently increases at low tem-perature. At high temperatures, the resistivity shows theactivation-type behavior ρxx ∝ exp(−Eg/2kBT ) with Eg andkB being the effective band gap and Boltzmann constant, re-spectively [see Fig. 2(b)]. From the fitting, the effective bandgap is estimated to be 0.11−0.13 eV. As shown in Fig. 2(c),the Hall resistivity ρyx for x = 0 is nearly linear as a functionof magnetic field, which shows a sizable temperature depen-dence. The Hall coefficient RH derived from the slope of ρyxis negative in the measured temperature range (100−350 K),indicating that the electron-type carriers dominate the chargetransport above 100 K. Assuming the single carrier model,the carrier density nH (= |1/eRH|) appears to decrease from085402-2CHARGE TRANSPORT AND THERMOPOWER IN THE … PHYSICAL REVIEW MATERIALS 7, 085402 (2023)FIG. 2. (a) The temperature dependence of resistivity ρxx . (b) ln(ρxx) plotted as a function of 1/T . (c) The Hall resistivity (ρyx) forCa1−xLaxPd3O4 with x = 0. The dashed line in (b) is the fitting by the thermal activation model (see also the text). (d) Temperature dependenceof ρxx for Ca1−xLaxPd3O4 with x = 0.01 − 0.07. (e) Temperature dependence of carrier density derived from the Hall coefficient nH. (f) nHplotted as a function of nominal carrier density ncal expected from the La doping. The dashed line denotes the nH = ncal.1.5 × 1019 cm−3 at 300 K to 1.1 × 1018 cm−3 at 100 K withdecreasing temperature [see Fig. 2(e)].To clarify the electronic state, we investigated the opticalconductivity σ (ω) spectra for x = 0. As shown in Fig. 3(a),σ (ω) spectra exhibit several peaks due to the optical phononsbelow 0.1 eV and the absorption band due to the interbandtransition above 0.12 eV. Moreover, a sharp peak with sizabletemperature dependence is observed at 0.12 eV. The temper-ature variation of spectral shape for interband transition ismoderate, suggesting that the electronic state does not signifi-cantly change with temperature. By linearly extrapolating therising part of interband transition at 10 K, the charge gap isroughly estimated to be about 0.1 eV. Considering that thepeak at 0.12 eV is located near the onset of interband transi-tion and the peak width is much broader than that of opticalFIG. 3. (a) Optical conductivity σ (ω) spectra for CaPd3O4. Theclosed triangle denotes the peak at 0.12 eV. (b) The electronic struc-ture for CaPd3O4. The color denotes the weight of Pd 4d3z2−r2 orbital(magenta) and Pd 4dx2−y2 orbital (blue). The illustration of (c) Pd4d3z2−r2 orbital and (d) Pd 4dx2−y2 orbital in a PdO4 plaquette. Theclosed circles denote the oxygen ions.phonons, it may be attributed to the electronic excitation dueto the impurity state located near the valence band top orexciton absorption.More detailed insight into the electronic state can be ob-tained by the ab initio calculation. Figure 3(b) shows thecalculated band structure for CaPd3O4. The calculated resultsshow the band gap about 0.24 eV at the � point in the mo-mentum space. The magnitude of band gap is slightly larger,but is of the same order as that determined from the opticalconductivity spectra. The valence bands and conduction bandsare mainly composed of Pd 4d3z2−r2 state and Pd 4dx2−y2 state,respectively [see also Fig. S2] [24]. In particular, near the �point, the conduction band is dispersive along all of the � − R,� − M, and � − X lines. On the contrary, the dispersion ofvalence band of the highest energy is relatively flat along the� − X and � − M lines, while that along the � − R line isdispersive. This difference can be qualitatively understood interms of the orbital hybridization between the Pd 4d state andO 2p state in the PdO4 plaquette; the Pd 4dx2−y2 state is hy-bridized with surrounding O 2p state more significantly thanthe Pd 4d3z2−r2 state [see also Figs. 3(c) and 3(d)], resultingin the larger band dispersion for the former [33]. In otherwords, the spatial anisotropy of Pd 4d orbital is likely crucialfor the difference in the band dispersion between the conduc-tion band and valence band, i.e., the electron-hole asymmetryin this material, which is rarely seen among conventionalnarrow gap semiconductors.Next, we explored the charge transport and electronicstate for the electron-doped Ca1−xLaxPd3O4. As shown inFig. 2(d), ρxx rapidly decreases with increasing x, and themetallic state emerges at the doping level of as small asx = 0.01. In particular, the resistivity is reduced down toabout 2−3 m�cm for x = 0.01 and further decreases forx � 0.03. Such low resistivity is in contrast with the caseof hole-doped analogs, wherein the resistivity is more than085402-3KOUTA KAZAMA et al. PHYSICAL REVIEW MATERIALS 7, 085402 (2023)FIG. 4. Seebeck coefficient S (a) for CaPd3O4 (x = 0) and (b) for Ca1−xLaxPd3O4 with x = 0.01 − 0.06. (c) The absolute value of Seebeckcoefficient |S| and electrical conductivity σ at 350 K. The solid curve is guide to the eyes. (d) The power factor σS2 for Ca1−xLaxPd3O4 at350 K as well as that near the room temperature (290 − 350 K) for various n-type oxides. (e) The carrier mobility for Ca1−xLaxPd3O4 withx = 0.01 − 0.07.10 m�cm in the lightly doped region of x � 0.05 [21–23].Figure 2(e) shows the temperature dependence of carrier den-sity for Ca1−xLaxPd3O4 with various x. In the doped systems,the carrier density is nearly temperature independent in con-trast with the case of x = 0 [see also Fig. S3] [24]. In Fig. 2(f),the carrier density at 300 K is plotted against the nominalcarrier density assuming that a trivalent La ion partially substi-tuted into a divalent Ca ion introduces a conduction electron.The carrier density is nearly identical to the nominal carrierdensity, suggesting that the electron density is controlled bythe doping level of La ions.Figure 4(a) shows the temperature dependence of See-beck coefficient for CaPd3O4. The sign of Seebeck coefficientis positive below 175 K, but becomes negative at highertemperatures. This suggests the so-called bipolar effect; thecontribution from hole-type carriers (electron-type carriers) isdominant at low (high) temperatures. As shown in Fig. 4(b),for x � 0.01, the sign of Seebeck coefficient remains to benegative in all the temperature range from 10 K to 350 K. Weplot the absolute value of Seebeck coefficient |S| at 350 Kas a function of carrier density in Fig. 4(c). |S| moderatelydecreases with increasing the carrier density. In fact, |S| forx = 0.06 (n = 7 × 1020 cm−3) is more than one-third of thevalue for x = 0. On the contrary, the electrical conductiv-ity σ rapidly increases with increasing the carrier density[see Fig. 4(c)]; the electrical conductivity for x = 0.03 − 0.04(n = 3 − 4 × 1020 cm−3) is one order larger than that forx = 0. As a result, the power factor σS2 reaches a maxi-mum around n = 3 × 1020 cm−3 (x = 0.03), wherein the peakvalue is about 7 µW/K2cm as shown in Fig. 4(d). This valueis about four times the maximum in the hole-doped analog ofCaPd3O4 [21,22], which is also relatively large compared withother n-type transition metal oxide semiconductors [34–42].To clarify the electronic state in the electron-doped region,we explored the photoemission spectra for Ca1−xLaxPd3O4with x = 0, 0.03, 0.05, and 0.07. As shown in Fig. 5(a), thephotoemission spectrum from the valence band onsets imme-diately below EF for x = 0. On the contrary, for x � 0.03, thethreshold energy seems to be slightly lower, and a small kinkdue to the Fermi edge is observed at EF [see also Fig. S5][24]. Here, we defined the valence band maximum (VBM) bylinearly extrapolating the rising part of photoemission spec-trum to the baseline [see Fig. 5(a)]. Figure 5(b) shows theVBM plotted as a function of x. The relative energy of VBMmeasured from EF, i.e., E − EF monotonically decreases withincreasing x.We also explored the photoemission spectra of Pd 3d5/2core level, which is located in a much deeper energy region.As shown in Fig. 5(c), the photoemission spectrum from thePd 3d5/2 core level is composed of a peak and shoulder. Thepeak is assigned to the signal from the nominally divalentstate Pd2+, while the shoulder is likely attributed to that fromthe surface state and/or the tetravalent state Pd4+ [43,44]. Byfitting the spectra, we determined the relative energy of Pd3d5/2 core level. As shown in Fig. 5(b), with increasing x, therelative energy of Pd 3d5/2 core level decreases similarly withthat of the VBM. These results suggest that the photoemissionpeaks nearly uniformly shifts to deeper relative energy withincreasing x. In other words, this means that EF increases inthe scheme of rigid band picture as illustrated in Fig. 5(d); EF085402-4CHARGE TRANSPORT AND THERMOPOWER IN THE … PHYSICAL REVIEW MATERIALS 7, 085402 (2023)FIG. 5. (a) Valence band photoemission spectra forCa1−xLaxPd3O4 with x = 0 − 0.07. Each spectrum is offsetfor clarity. The horizontal (vertical) solid line denotes the baseline(Fermi energy). The dashed line is the extrapolation to estimate thevalence band maximum (VBM), which is shown by arrows. (b) Theenergy of VBM and that of Pd 3d5/2 plotted as a function of x.(c) The core level photoemission spectra of Ca1−xLaxPd3O4 withx = 0 − 0.05. The circles and triangles denote the main peak fromPd 3d5/2 state and the shoulder (see the text), respectively. Eachspectrum is offset for clarity. (d) The illustration of density of statenear the Fermi energy.is almost located at the valence band top for x = 0, but rapidlymoves above the conduction band minimum for x � 0.01.Such doping variation of electronic state is in contrast withthe case of doped Mott insulator, in which the electronic statenear EF significantly changes upon the electron/hole doping[45,46].On the basis of these results, we consider the doping vari-ation of charge transport and thermoelectric effect. For x = 0,it is likely that the hole-type carriers in the valence band orimpurity state located near the VBM govern the transport atlow temperatures. This is consistent with the positive signof the Seebeck coefficient observed at low temperatures. Onthe contrary, near the room temperature, the thermally excitedelectrons in the conduction band likely dominate the transportproperty, resulting in the negative sign of the Seebeck coeffi-cient and Hall coefficient. This is also consistent with the factthat the effective band gap is nearly identical to the chargegap determined from the optical conductivity spectra. We notethat the previous study reports the positive Seebeck coefficientup to 500 K [21]. This discrepancy may be attributed tothe difference in the Fermi energy, which is sensitive to thechemical off-stoichiometry of samples.For x � 0.01, the doped electrons govern the charge trans-port in all the temperature range below 350 K. It is likely thatthe enhancement of power factor is attributed to the relativelylarge Seebeck coefficient while keeping the high electricalconductivity, probably the relatively high electron mobilityin the lightly doped region. Indeed, as shown in Fig. 4(e),the electron mobility remains to be about 30 cm2/Vs even at350 K for x = 0.03, which however decreases with increasingthe carrier density due to the scattering from the doped Laions or degradation of crystallinity by chemical substitution.Note that the value in the lightly doped region is one or-der of magnitude higher than the value in the hole-dopedanalog about 1−10 cm2/Vs [21]. The larger dispersion ofconduction Pd 4dx2−y2 band may be crucial for the relativelyhigh electron mobility and power factor in the electron dopedCa1−xLaxPd3O4.IV. CONCLUSIONIn conclusion, we have investigated the charge transport,electronic state, and thermoelectric effect of electron-dopedCa1−xLaxPd3O4 by means of transport measurements, op-tical spectroscopy, ab initio calculation, and photoemissionspectroscopy. We synthesized the high quality samples ofCa1−xLaxPd3O4 by using the high pressure synthesis tech-nique. For CaPd3O4 (x = 0), the temperature dependenceof resistivity shows the insulating behavior, and the opticalconductivity spectra show a charge gap of about 0.1 eV.The ab initio calculation suggests that the conduction bandand valence band are mainly composed of the Pd 4dx2−y2state with larger band dispersion and Pd 4d3z2−r2 state withsmaller band dispersion, respectively. Upon the electron dop-ing, the resistivity rapidly decreases with increasing x, andthe metallic state emerges even at the small doping levelof x = 0.01. Indeed, owing to the relatively high electronmobility (∼30 cm2/Vs), the electrical conductivity exceeds1000 �−1cm−1 in the lightly doped region for x = 0.03−0.05even at 350 K. On the contrary, the magnitude of Seebeckcoefficient moderately decreases with increasing x. Conse-quently, the power factor σS2 reaches about 7 µW/K2cm at350 K for x = 0.03 − 0.04. Furthermore, the results of pho-toemission spectra suggest that the doped electrons occupythe conduction band in the scheme of the rigid band picture. Itis likely that the enhanced power factor is attributed to the rel-atively high electron mobility due to the dispersive Pd 4dx2−y2conduction band in this electron-doped oxide semiconductor.ACKNOWLEDGMENTSWe thank A. Oshima, T. Nakamura, and T. Koyano forthe technical supports for measurements of the photoemissionspectra and transport property. The synchrotron radiation ex-periments were performed at SPring-8 with the approval ofthe Japan Synchrotron Radiation Research Institute (JASRI)(Proposals No. 2022B1332 and No. 2022B2106). This workwas partly supported by Grant-In-Aid for Science Research(Grants No. 18H01171, No. 19H01851, No. JP20H01834,No. 21K18813, and No. 22H01177) from the MEXT, byNippon Sheet Glass Foundation for Materials Science andEngineering, Japan, by JST FOREST Program (Grant No. JP-MJFR203D) and by the University of Tsukuba, Pre-StrategicInitiatives Development Center for High-Function and High-Performance Organic-Inorganic Spin Electronics. MANA issupported by World Premier International Research CenterInitiative (WPI), MEXT, Japan.085402-5KOUTA KAZAMA et al. PHYSICAL REVIEW MATERIALS 7, 085402 (2023)[1] J. W. Sharp, E. C. Jones, R. K. Williams, P. M. Martin, and B. C.Sales, J. Appl. Phys. 78, 1013 (1995).[2] T. Caillat, M. Carle, P. Pierrat, H. Scherrer, and S. Scherrer,J. Phys. Chem. Solids 53, 1121 (1992).[3] J. W. G. Bos, H. W. Zandbergen, M.-H. Lee, N. P. Ong, and R. J.Cava, Phys. Rev. B 75, 195203 (2007).[4] Y. Pei, X. Shi, A. LaLonde, H. Wang, L. Chen, and G. J. Snyder,Nature (London) 473, 66 (2011).[5] C.-R. Wang, W.-S. Lu, L. Hao, W.-L. Lee, T.-K. Lee, F. Lin,I.-C. Cheng, and J.-Z. Chen, Phys. Rev. Lett. 107, 186602(2011).[6] L. D. Zhao, G. Tan, S. Hao, J. He, Y. Pei, H. Chi, H. Wang, S.Gong, H. Xu, V. P. Dravid, C. Uher, G. J. Snyder, C. Wolverton,and M. G. Kanatzidis, Science 351, 141 (2016).[7] T. Inohara, Y. Okamoto, Y. Yamakawa, A. Yamakage, and K.Takenaka, Appl. Phys. Lett. 110, 183901 (2017).[8] J. M. Tomczak, J. Phys.: Condens. Matter 30, 183001 (2018).[9] A. Nakano, U. Maruoka, F. Kato, H. Taniguchi, and I. Terasaki,J. Phys. Soc. Jpn. 90, 033702 (2021).[10] K. Tsuruda, K. Nakagawa, M. Ochi, K. Kuroki, M. Tokunaga,H. Murakawa, N. Hanasaki, and H. Sakai, Adv. Funct. Mater.31, 2102275 (2021).[11] T. Zhou, C. Zhang, H. Zhang, F. Xiu, and Z. Yang, Inorg. Chem.Front. 3, 1637 (2016).[12] J. Fujioka, M. Kriener, D. Hashizume, Y. Yamasaki, Y. Taguchi,and Y. Tokura, Phys. Rev. Mater. 5, 094201 (2021).[13] P. L. Smallwood, M. D. Smith, and H.-C. zur Loye, J. Cryst.Growth 216, 299 (2000).[14] K. Itoh and N. Tsuda, Solid State Commun. 109, 715 (1999).[15] I. Hase and Y. Nishihara, Phys. Rev. B 62, 13426 (2000).[16] A. Khan, Z. Ali, I. Khan, S. J. Asadabadi, and I. Ahmad, Bull.Mater. Sci. 39, 1861 (2016).[17] G. Li, B. Yan, Z. Wang, and K. Held, Phys. Rev. B 95, 035102(2017).[18] S. M. L. Teicher, L. K. Lamontagne, L. M. Schoop, and R.Seshadri, Phys. Rev. B 99, 195148 (2019).[19] F. Tang and X. Wan, Front. Phys. 14, 43603 (2019).[20] B. H. Reddy, A. Ali, and R. S. Singh, J. Phys.: Condens. Matter33, 185502 (2021).[21] S. Ichikawa and I. Terasaki, Phys. Rev. B 68, 233101 (2003).[22] T. C. Ozawa, A. Matsushita, Y. Hidaka, T. Taniguchi, S.Mizusaki, Y. Nagata, Y. Noro, and H. Samata, J. Alloys Compd.448, 77 (2008).[23] L. K. Lamontagne, G. Laurita, M. Knight, H. Yusuf, J. Hu, R.Seshadri, and K. Page, Inorg. Chem. 56, 5158 (2017).[24] See Supplemental Material at http://link.aps.org/supplemental/10.1103/PhysRevMaterials.7.085402 for the sample characteri-zation, transport property, and electronic state.[25] Y. Senba, H. Ohashi, Y. Kotani, T. Nakamura, T. Muro, T.Ohkochi, N. Tsuji, H. Kishimoto, T. Miura, M. Tanaka, M.Higashiyama, S. Takahashi, Y. Ishizawa, T. Matsushita, Y.Furukawa, T. Ohata, N. Nariyama, K. Takeshita, T. Kinoshita,A. Fujiwara et al., AIP Conf. Proc. 1741, 030044 (2016).[26] T. Muro, Y. Senba, H. Ohashi, T. Ohkochi, T. Matsushita,T. Kinoshita, and S. Shin, J. Synchrotron Rad. 28, 1631(2021).[27] G. Kresse and J. Furthmuller, Phys. Rev. B 54, 11169 (1996).[28] G. Kresse and D. Joubert, Phys. Rev. B 59, 1758 (1999).[29] J. Heyd, G. E. Scuseria, and M. Ernzerhof, J. Chem. Phys 118,8207 (2003).[30] J. Heyd, G. E. Scuseria, and M. Ernzerhof, J. Chem. Phys. 124,219906 (2006).[31] A. V. Krukau, O. A. Vydrov, A. F. Izmaylov, and G. E. Scuseria,J. Chem. Phys. 125, 224106 (2006).[32] G. Pizzi, V. Vitale, R. Arita, S. Blügel, F. Freimuth,G. Géranton, M. Gibertini, D. Gresch, C. Johnson, T. Koretsune,J. Ibañez-Azpiroz, H. Lee, J.-M. Lihm, D. Marchand,A. Marrazzo, Y. Mokrousov, J. I. Mustafa, Y. Nohara,Y. Nomura, L. Paulatto et al., J. Phys.: Condens. Matter 32,165902 (2020).[33] M.-L. Doublet, E. Canadell, and M.-H. Whangbo, J. Am. Chem.Soc. 116, 2115 (1994).[34] T. Okuda, K. Nakanishi, S. Miyasaka, and Y. Tokura, Phys. Rev.B 63, 113104 (2001).[35] A. Sakai, T. Kanno, S. Yotsuhashi, H. Adachi, and Y. Tokura,Jpn. J. Appl. Phys. 48, 097002 (2009).[36] T. Tsubota, M. Ohtaki, K. Eguchi, and H. Arai, J. Mater. Chem.8, 409 (1998).[37] K.-H. Jung, K. H. Lee, W.-S. Seo, and S.-M. Choi, Appl. Phys.Lett. 100, 253902 (2012).[38] D. Berardan, C. Byl, and N. Dragoe, J. Am. Ceram. Soc. 93,2352 (2010).[39] M. Ohtaki, H. Koga, T. Tokunaga, K. Eguchi, and H. Arai,J. Solid State Chem. 120, 105 (1995).[40] I. Terasaki and T. Nonaka, J. Phys.: Condens. Matter 11, 5577(1999).[41] H. Liu, H. Ma, T. Su, Y. Zhang, B. Sun, B. Liu, L. Kong, B. Liu,and X. Jia, J. Materiomics 3, 286 (2017).[42] I. Terasaki, S. Ichikawa, and S. Shibasaki, Proceedings of the23rd International Conference on Thermoelectrics (Organizingcommittee of 23rd International Conference on Thermo-electrics, ICT2004, 2005), p. 094.[43] L. P. A. Guerrero-Ortega, E. Ramírez-Meneses, R. Cabrera-Sierra, L. M. Palacios-Romero, K. Philippot, C. R. Santiago-Ramírez, L. Lartundo-Rojas, and A. Manzo-Robledo, J. Mater.Sci. 54, 13694 (2019).[44] L. S. Kibis, A. A. Simanenko, A. I. Stadnichenko, V. I.Zaikovskii, and A. I. Boronin, J. Phys. Chem. C 125, 20845(2021).[45] M. Imada, A. Fujimori, and Y. Tokura, Rev. Mod. Phys. 70,1039 (1998).[46] H. Wadati, D. Kobayashi, H. Kumigashira, K. Okazaki, T.Mizokawa, A. Fujimori, K. Horiba, M. Oshima, N. Hamada,M. Lippmaa, M. Kawasaki, and H. Koinuma, Phys. Rev. B 71,035108 (2005).[47] K. Momma and F. Izumi, J. Appl. Cryst. 44, 1272 (2011).085402-6https://doi.org/10.1063/1.360402https://doi.org/10.1016/0022-3697(92)90087-Thttps://doi.org/10.1103/PhysRevB.75.195203https://doi.org/10.1038/nature09996https://doi.org/10.1103/PhysRevLett.107.186602https://doi.org/10.1126/science.aad3749https://doi.org/10.1063/1.4982623https://doi.org/10.1088/1361-648X/aab284https://doi.org/10.7566/JPSJ.90.033702https://doi.org/10.1002/adfm.202102275https://doi.org/10.1039/C6QI00383Dhttps://doi.org/10.1103/PhysRevMaterials.5.094201https://doi.org/10.1016/S0022-0248(00)00432-2https://doi.org/10.1016/S0038-1098(98)00549-3https://doi.org/10.1103/PhysRevB.62.13426https://doi.org/10.1007/s12034-016-1322-8https://doi.org/10.1103/PhysRevB.95.035102https://doi.org/10.1103/PhysRevB.99.195148https://doi.org/10.1007/s11467-019-0902-7https://doi.org/10.1088/1361-648X/abf0c7https://doi.org/10.1103/PhysRevB.68.233101https://doi.org/10.1016/j.jallcom.2007.03.137https://doi.org/10.1021/acs.inorgchem.7b00307http://link.aps.org/supplemental/10.1103/PhysRevMaterials.7.085402https://doi.org/10.1063/1.4952867https://doi.org/10.1107/S1600577521007487https://doi.org/10.1103/PhysRevB.54.11169https://doi.org/10.1103/PhysRevB.59.1758https://doi.org/10.1063/1.1564060https://doi.org/10.1063/1.2204597https://doi.org/10.1063/1.2404663https://doi.org/10.1088/1361-648X/ab51ffhttps://doi.org/10.1021/ja00084a057https://doi.org/10.1103/PhysRevB.63.113104https://doi.org/10.1143/JJAP.48.097002https://doi.org/10.1039/a706213chttps://doi.org/10.1063/1.4729560https://doi.org/10.1111/j.1551-2916.2010.03751.xhttps://doi.org/10.1006/jssc.1995.1384https://doi.org/10.1088/0953-8984/11/29/304https://doi.org/10.1016/j.jmat.2017.06.002https://doi.org/10.1007/s10853-019-03843-8https://doi.org/10.1021/acs.jpcc.1c04646https://doi.org/10.1103/RevModPhys.70.1039https://doi.org/10.1103/PhysRevB.71.035108https://doi.org/10.1107/S0021889811038970