Manabu Inukai

Thermoelectric properties of supersaturated Re solid solution of higher manganese silicides

In this study, we developed a higher manganese silicide (HMS) that possesses a high dimensionless figure of merit ZT exceeding unity. HMSs containing a larger amount of Re than its solubility limit were prepared by the liquid quenching technique, and the obtained metastable HMSs showed good thermal stability to enable pulse current sintering at 1240 K. The lattice thermal conductivity was effectively reduced with increasing Re concentration, whereas the electron transport properties were not greatly affected. Consequently, the ZT of p-type HMS increased to 1.04 at 6 at. % Re from 0.4 of the Re-free sample.

Thermoelectric properties of Al?Mn?Si C40 phase containing small amount of W or Ta

The lattice thermal conductivity of the Al–Mn–Si C40 phase was effectively reduced by substituting a small amount of W for Mn while maintaining a large Seebeck coefficient |S| > 100 µV/K and a low electrical resistivity ρ < 1.7 mΩ cm. The Al–Mn–Si-based C40 phase containing a small amount of W showed essentially the same behaviors of both Seebeck coefficient and electrical resistivity as those of a W-free sample at the same carrier concentration. This experimental result was consistent with the first principles calculations that predicted the absence of impurity states in the energy gap in association with the W atom at the Mn site. The lattice thermal conductivity, on the other hand, markedly decreased owing to the heavy element impurity scattering effect of phonons. Consequently, the ZT of the n-type Al27.5Mn29.0W3.0Fe1.0Si39.5 C40 phase increased to a value that was ~4 times larger than that of the W-free sample.

Hume-Rothery stabilisation mechanism and d-states-mediated Fermi surface―Brillouin zone interactions in structurally complex metallic alloys

The stability of Co2Zn11 and Al8V5 gamma-brasses, both of which are composed of a transition metal element and polyvalent elements Zn or Al, can be discussed in terms of d-states-mediated Fermi surface–Brillouin zone (FsBz) interactions in the context of first-principles full-potential linearised augmented plane wave (FLAPW) band calculations. A FsBz-induced pseudogap is revealed in the FLAPW-Fourier spectrum, though it is hidden behind a much larger d-band in the total density of states. The stability range of three families of complex metallic alloys (CMAs) that include gamma-brasses, RT-, MI- and Tsai-type 1/1-1/1-1/1 approximants and 2/1-2/1-2/1 approximant, each of which is characterised by = 18, 50 and 125, respectively, can be well scaled in terms of the number of electrons per unit cell (e/uc) given by the product of the number of atoms per unit cell and the e/a value determined by the Hume-Rothery plot on the basis of the FLAPW-Fourier method. This is taken as the evidence for the justification of the Hume-Rothery stabilisation mechanism for all these CMAs having a pseudogap at the Fermi level.

Fermi surface–Brillouin-zone-induced pseudogap in γ-Mg17Al12 and a possible stabilization mechanism of β-Al3Mg2

The electronic structure of γ phase in the system Mg(17)Al(12) containing 58 atoms per unit cell with space group I43m has been calculated by using the WIEN2k-FLAPW program package. A pseudogap is found across the Fermi level. The FLAPW-Fourier spectra at the symmetry points N and Γ of the bcc Brillouin zone revealed that electronic states across the Fermi level at these symmetry points are dominated by |G|(2) = 26 and 24 states corresponding to centers of {510} + {431} and {422} zone planes, respectively. The 1253-wave nearly-free-electron (NFE) band calculations identified that a combination of the two Fermi surface-Brillouin-zone (FsBz) interactions associated with |G|(2) = 26 and 24 account well for the observed DOS pseudogap in γ-Mg(17)Al(12), most likely leading to the stabilization of this complex metallic compound. The β-Al(3)Mg(2) containing 1178 atoms per cubic unit cell is suggested to be stabilized by satisfying the Hume-Rothery matching condition expressed in terms of e/uc, the number of electrons per unit cell, versus critical |G|(2). A critical |G|(2) is predicted to be 200 in β-Al(3)Mg(2), which results in 84 Brillouin zone planes interacting almost simultaneously with a more or less spherical Fermi surface.

e/adetermination for 4d- and 5d-transition metal elements and their intermetallic compounds with Mg, Al, Zn, Cd and In

The present work is devoted to the determination of the effective electrons per atom ratio e/a by means of first-principles full-potential linearized augmented plane wave-Fourier method for elements from Rb to Ag in Period 5 and from Cs to Au in Period 6 of the periodic table and is regarded as a continuation of the preceding work done for elements from K to Cu in Period 4. The value of e/a was determined by reading off the square of the Fermi diameter, from the dispersion relation for electrons outside the Muffin-Tin spheres. A straightforward reading of the ordinate at the Fermi level, i.e. local reading method was validated for Rb and Cs in Group 1, Sr in Group 2, Y in Group 3, Pd and Pt in Group 10 and Ag and Au in Group 11. Instead, the nearly free electron (NFE) method was found to be indispensable for TM elements from Zr to Rh in Period 5 and those from Ba to Ir in Period 6. The composition dependence of e/a values for intermetallic compounds in X–TM (X = Mg, Al, Zn, Cd and In) alloy systems was also studied. The new Hume–Rothery electron concentration rule was established by constructing e/uc, the number of electrons per unit cell, vs. square of critical reciprocal lattice vector, , diagram for structurally complex metallic alloys having a pseudogap at the Fermi level. A proper use of either the local reading- or the NFE-e/a for the elements as indicated above is found to be essential.

Electrons per Atom Ratio Determination and Hume-Rothery Electron Concentration Rule for P-Based Polar Compounds Studied by FLAPW–Fourier Calculations

The extent to which reliable electrons per atom ratio, e/a, are determined and the validity of the Hume-Rothery stabilization mechanism are ensured upon increasing ionicity are studied by applying first-principles full potential linearized augmented plane wave (FLAPW)-Fourier band calculations to as many as 59 binary compounds formed by adding elements from periods 2-6 to phosphorus in group 15 of the Periodic Table. Van Arkel-Ketelaar triangle maps were constructed both by using the Allen electronegativity data and by using an energy difference between the center-of-gravity energies of FLAPW-derived s and p partial densities of states (DOSs) for the equiatomic compounds studied. The determination of e/a and the test of the interference condition, both of which play a key role in the Hume-Rothery stabilization mechanism, were reliably made for all intermetallic compounds, as long as the ionicity is less than 50%. In the A-P (A = Li, Na, K, Rb, and Cs) compounds with ionicity exceeding 50%, however, e/a determination becomes unstable, as reflected in its P concentration dependence. New Hume-Rothery electron concentration rules were theoretically found in two families of polar compounds: skutterudite compounds TMP(3), TMAs(3), and TMSb(3) (TM = Co, Ni, Rh, and Ir; cI32) with e/a = 4.34 and TM(3)P (TM = Cr, Mn, Fe, and Ni; tI32) with e/a = 2.20.

NFEapproximation for the e/a determination for 3d-transition metal elements and their intermetallic compounds with Al and Zn

First-principles full-potential linearized augmented plane wave (FLAPW) band calculations with subsequent FLAPW-Fourier analyses have been performed for elements from K to Cu in period 4 of the periodic table to determine the effective electrons per atom ratio (e/a). For the series of 3d-transition metals (TM), the determination of the square of the Fermi diameter , from which e/a is derived, has been recognized not to be straightforward because of the presence of a huge anomaly associated with the TM-d states across the Fermi level in the energy dispersion relation for electrons outside the muffin-tin sphere. The nearly free electron (NFE) approximation is newly devised to circumvent this difficulty. The centre of gravity energy is calculated from the energy distribution of the square of the Fourier coefficients for the FLAPW state . The NFE dispersion relation is constructed for the set of and in combination with the tetrahedron method. The resulting e/a values are distributed over positive numbers in the vicinity of unity for elements from Ti to Co. Instead, the e/a values for the early elements K, Ca and Sc and the late TM elements Ni and Cu were determined to be close to one, two, three, 0.50 and unity, respectively, using our previously designed local reading method. In addition, the composition dependence of e/a values for intermetallic compounds in X-TM (X = Al and Zn) alloy systems was studied to justify an appropriate choice between the local reading and NFE methods for respective elements.

Correlation between structure and mixed ionic–electronic conduction mechanism for (La1−xSrx)CoO3−δ using synchrotron X-ray analysis and first principles calculations

The mechanism of mixed ionic–electronic conduction (MIEC) for (La1−xSrx)CoO3−δ (LSC) (x = 0.00–0.50) was investigated by experimental as well as theoretical studies which included Rietveld refinements, maximum entropy method (MEM) analysis, X-ray absorption spectroscopy, and first principles calculations (FPCs). The correlations among MIEC, structural parameters, i.e., the bond length, the bond angle, electron density, and the spin configuration, were discussed from the practical results. Moreover, these correlations were verified by FPC, and interpreted by the band structure and the total energy of LSC. The electronic conductivity was affected by the Co–O–Co bond angle, electron densities, and the spin state. In particular, the spin transition from the low to high spin state was accompanied by the cross-over from semiconductor to metallic at x = 0.15. Concerning the oxide ion diffusion, LSC has large anisotropic atomic displacement parameters of the oxide ion sites, in the vertical direction of the Co–O bond, which associate with the high oxide ion diffusion. The total energy of LSC during the migration of the oxide ion was also calculated with varying Sr content, lattice parameters, and symmetries by using FPC. Consequently, the symmetry of LSC is more effective to reduce the activation energy of oxide ion diffusion than Sr content and lattice parameters.

Electronic structure of Ag5Zn8, Ag9In4 and Mn3In gamma-brasses studied by FLAPW band calculations

First-principles FLAPW (full-potential linearized augmented plane wave) band calculations were performed, using the WIEN2k program package, for three isostructural gamma-brasses, Ag5Zn8, Ag9In4 and Mn3In. The FsBz (Fermi surface–Brillouin zone) interactions were successfully extracted from the case.output1 file generated from WIEN2k. This has opened a way for any users to perform FLAPW-Fourier analyses using the commercially available WIEN2k package. From the FLAPW-Fourier spectrum, it was found that the square of the reciprocal lattice vector,  = 18, for the set of {411} and {330} lattice planes is responsible for opening a pseudogap in Ag5Zn8 and Ag9In4 gamma-brasses through FsBz interactions. In the case of Mn3In, the importance of the d-states mediated FsBz interaction involving  = 18 is emphasized.

SYNCHROTRON RADIATION PHOTOELECTRON STUDY OF HEUSLER-TYPE Fe2VAl-BASED ALLOYS

The electronic structures of Heusler-type Fe2VAl-based alloys, n-type thermoelectric alloys Fe2VAl1-zXz(X = Si, Ge) and p-type one Fe2(V1-yTiy)Al, have been investigated by photoelectron spectroscopy with use of synchrotron radiation as an excitation photon source. Observed valence-band spectra of Fe2VAl agree quite well with the predicted density of states with a sharp pseudogap across the Fermi energy EF. Small substitution in Fe2VAl1-zXz and Fe2(V1-yTiy)Al reveals a rigid-band-like energy shift of the valence band, although the observed shift is smaller than expected in the rigid band model. Further substitution causes the modification of the electronic structure near EF and the appearance of mid-pseudogap states. However, the dependence of the thermoelectric properties of these alloys on substitution can be qualitatively explained by the observed change in the electronic structure near EF.