Search

In this work, we elucidated the electronic structure and thermoelectric properties of intermetallic XSi (X = Co, Rh) compounds. By combining first-principles calculations and the Boltzmann equation within electron–phonon average approximation, thermoelectric properties of XSi are well reproduced compared to experimental observations. We found that the considerable Seebeck coefficient and low electrical resistivity give XSi a large power factor, which has the potential for thermoelectric applications. We also discussed the electron mean free paths, a possible factor affecting electrical conductivity when using the nanostructuring approach to optimize the thermoelectric performance.

Deformation of silicene buckled structure attracts great interest for the possibility of ultrahigh thermoelectric power factor. Therefore, the control method of silicene buckled structure is needed. Here, we developed the method to control composition ratio in epitaxial Ca intercalated layered silicene (CaSi2) film formed by solid phase epitaxy through an atomic interdiffusion between Ca films and Si substrate because of the possible existence of the relation between silicene buckled structure in CaSi2 film and the composition ratio. The interdiffusion is controlled by introducing hydrogen-terminated layer as an interface layer between Ca and Si substrate, resulting in the control of the composition ratio in CaSi2 film. Moreover, we find that the CaSi2 films with different composition ratio exhibit different thermoelectric power factors. This study reveals that introducing the interface layer for interdiffusion control is an effective way to control the composition ratio and to form metastable high-buckled silicene with high power factor.

In order to propose a new dilute magnetic semicoductors (DMS) other than II–VI and III–V semiconductor based DMS, magnetism of CuAlO2 based DMS is investigated by the Korringa–Kohn–Rostoker method combined with the coherent potential approximation within the local spin density approximation and Monte Carlo simulations (MCS). We found that effective exchange interactions (Jij) between magnetic ions are short-ranged and Jij's between magnetic ions in the same Cu-plane are ferromagnetic and ones between the different Cu-planes are nearly negligible in CuAlO2-based DMS. In comparison to this, Jij's between magnetic impurities occupied at Al-sites are slightly longer-ranged due to the stronger hybridization effect. According to MCS calculations, it is found that the value of the Curie temperature (TC) exceeds 80 K in (Cu,Fe)AlO2- and (Cu,Co)AlO2-DMS and that the TC is suppressed due to the strong percolation effect. This effect also appears in Cu(Al,TM)O2, where TM denote 3d-transition metal elements.

We propose that InGaN is superior to GaN as a host material for GaN-based red-light-emitting diodes (LEDs). In our previous paper, we proposed that codoping of Eu and a Mg and O pair generates an efficiently luminescent center in GaN. This is caused by the quantum confinement of the quantum dot constructions generated by the codoping method. The present report illustrates that InGaN allows the expansion of such electronic structures throughout the crystal owing to spontaneous phase decomposition. This can be used for self-organized fabrication and self-regenerated products.

Based on first-principles calculations, we propose co-doping method for increasing solubility of magnetic impurities in dilute magnetic semiconductors (DMSs). The concentration dependences of the mixing energy of DMS, such as (Ga,Mn)N, (Ga,Cr)N, (Ga,Mn)As, and (Zn,Cr)Te, show large convexity and these systems have a tendency toward spinodal decomposition. By introducing compensating impurities into these DMS, the mixing energy shows gradual transition from convex to concave concentration dependence resulting in negative mixing energy of magnetic impurities. This observation suggests that the co-doping method dramatically increases the solubility of magnetic impurities in DMS, thus high concentration doping of magnetic impurities into DMS becomes possible.

In addition to the conventional Peltier effect, we propose a new class of thermoelectric-cooling mechanism based on the adiabatic spin-entropy expansion in a quasi-one-dimensional nano-superstructure by injecting the spin current from the ferromagnetic metal to paramagnetic one. The spin-entropy expansion mechanism dominates and enhances the thermoelectric-cooling power dramatically in current perpendicular to plane-giant magneto-resistance (CPP-GMR) Co/Au nano-interface. Based upon the spin-entropy expansion mechanism, we design the new thermoelectric-cooling nano-superstructures using the newly designed half-Heusler ferromagnets NiMnSi (TC=1050 K) and self-organized quasi-one-dimensional Konbu-phase (Zn,Cr)Te with very high blocking temperature (>1000 K) by spinodal nano-decomposition.

We present computer simulations on the super-paramagnetic blocking phenomena in wide band-gap dilute magnetic semiconductors (DMSs) in particular (Ga, Mn)N. It is shown that due to the formation of magnetic nano-clusters of magnetic impurities under the spinodal decomposition, super-paramagnetic blocking temperature (TB) of the DMS is enhanced and hysteretic magnetic response can be observed at finite temperature even if Curie temperature of the whole system is nearly zero. TB higher than room temperature can be expected and the blocking phenomena explains reported ferromagnetic behaviour of wide band-gap DMS at high temperature.

We show that spinodal decomposition under layer by layer crystal growth condition leads to characteristic quasi-one-dimensional nano-structures in dilute magnetic semiconductors (DMS). It is found that the DMS systems can form rather large clusters with highly anisotropic shape even for low concentrations. It is suggested that the blocking phenomena in the super-paramagnetism, the magnetic dipole–dipole interaction and the network of the one-dimensional structures should be considered to understand the magnetism in DMS. Based on the present simulations, we propose that the delta-doping method can be effective approach to realize high Curie temperature.

The electronic structure and the magnetic properties of (Zn,Cr)S, (Zn,Cr)Se and (Zn,Cr)Te are calculated by using the Korringa-Kohn-Rostoker method combined with the coherent potential approximation (KKR-CPA) within the local density approximation. Curie temperatures are calculated by the mean field approximation (MFA) and by Monte Carlo simulation. It is found that the MFA strongly overestimates the Curie temperatures for low concentrations due to shortrangeness of interactions in (Zn,Cr)S, (Zn,Cr)Se and (Zn,Cr)Te. The Curie temperatures of (Zn,Cr)Te calculated by Monte Carlo simulation agree very well with recent experimental values.

We propose that nanostructures spontaneously generated by spinodal decomposition can be used as an efficiently luminescent material. The doping of Eu into GaN beyond the solubility limit forms EuN nanostructures, whose forms depend on the crystal growth method and conditions. The three-dimensional crystal growth generates the Dairiseki phase constructed of EuN quantum dots. These nanostructures are suitable for emission of red light and laser. The two-dimensional layer-by-layer crystal growth leads to the Konbu phase consisting of nanorods. The Konbu phase can be applied to the bottom-up construction of distributed feedback semiconductor lasers, which is currently built by the top-down nanotechnology such as photolithography.

We have recently implemented a new version of the quasiparticle self-consistent GW (QSGW) method in the ecalj package released at http://github.com/tkotani/ecalj. Since the new version of the ecalj is numerically stable and accurate compared to the previous versions, we can perform calculations easily without being bothered with setting input parameters. Here we examine its ability to describe energy band properties, e.g., band-gap energy, eigenvalues at special points and effective mass, for variety of semiconductors and insulators. We treat C, Si, Ge, Sn, SiC (in 2H, 3C, and 4H structures), (Al, Ga, N)x(N, P, As, Pb), (Zn, Cd, Mg)x(O, S, Se, Te), SiO2, HfO2, ZrO2, SrTiO3, PbS, PbTe, MnO, NiO, and HgO. We propose that a hybrid QSGW method, where we mix 80 percent of QSGW and 20 percent of LDA, gives universally good agreement with experiments for these materials., 8 pages, 3 figures, 4 tables

New functional 4d-transition-metal (4d-TM)-doped K2S diluted magnetic semiconductors (DMSs) with transparent and half-metallic ferromagnetism are designed based upon the first principles calculations. We have systematically investigated the magnetism in 4d-TM-doped K2S DMSs. K2S is a transparent semiconductor with anti-CaF2 crystal structure and has large lattice spacing due to its large ionic radius of K+. It is found that Zr-, Nb-, Tc-, Ru- and Rh-doped K2S show the half-metallic and high-spin ferromagnetism and that Zr- or Nb-doped K2S are promising candidate for high-TC (TC ~350 K for 5 at% doping of 4d-TM) ferromagnetic DMSs with transparecy and large magneto-optical effect. Mo-doped K2S shows the spin-glass state since anti-ferromagnetic super-exchange interaction dominates the system without ferromagnetic double-exchange interaction.

The computational materials design of n-type filled tetrahedral compound magnetic semiconductors is proposed on the basis of first-principles calculations within the density functional theory. Using the Korringa–Kohn–Rostoker coherent potential approximation, the electronic structures of electron-doped Li(Zn,Mn)As, Li(Zn,Mn)P, and Li(Zn,Mn)N are calculated. By estimating free energy, the phase diagrams of these systems are predicted. It is shown that these systems are phase-separated and favor spinodal decomposition when the electrons are not doped. By introducing Li interstitials as donor impurities, spinodal decomposition is strongly suppressed and Mn can be doped at a high concentration. Moreover, ferromagnetic interactions between Mn atoms are induced by electron doping. Thus, we can expect electron-mediated ferromagnetism in these systems with a reasonable Curie temperature.

Using the first-principles calculation method, we study the ferromagnetic states of electron-doped (In,Fe)As by the Korringa–Kohn–Rostoker coherent potential approximation method and calculate Curie temperature (TC) by Monte Carlo simulation. We show that the chemical pair interactions and magnetic exchange interactions between Fe atoms depend on the concentration of the doped interstitial Be donor. We obtain the so-called ordered phase of Fe-rich regions in this material. With this type of structure, Be-doped (In,Fe)As can exhibit ferromagnetism and its TC could be up to approximately 85 K. We propose that the ordered phase is necessary for enhancing ferromagnetism and increasing TC.

Using ab initio electronic structure calculations by the Korringa–Kohn–Rostoker coherent potential approximation and Monte Carlo simulation of the two-dimensional spinodal nano-decomposition, we propose a new materials design for ferromagnetic Fe–Cu alloys. We demonstrate that Fe nano-wires of the size of 3–5 nm are surrounded by Cu matrix under a modified layer by layer crystal growth. Ferromagnetism is enhanced by the inhomogeneous distribution of Fe atoms in the alloy.

We investigate GaN codoped with Eu and Mg from calculations on the density functional theory. As a result, we obtain the total energies that are almost independent of Mg positions relative to Eu positions. The result suggests the coexistence of the luminescent centers that are Eu–Mg complexes and typical Eu atoms at least. Eu ions are trivalent in both centers; hence, the magnetization in the ground state 7F0 is zero. Considering the low excited states via the Boltzmann distribution, we can expect the finite magnetization on each Eu atom, but the Curie temperature is unlikely to be room temperature.

We propose the possibility of spinodal nanodecomposition in Cu2ZnSn[Se1-x S x ]4 (CZTSeS) for high efficiency photovoltaic solar cells, based on the first-principles calculations within the self-interaction-corrected local density approximation. By using the Korringa–Kohn–Rostoker coherent potential approximation method, electronic structures of CZTSeS are calculated. Due to the calculated positive mixing energy and type II band alignment between Cu2ZnSnSe4 and Cu2ZnSnS4, we can expect that the efficiency of CZTSeS becomes higher by spinodal nano-decomposition. Then we simulate the self-organized two-dimensional spinodal nanodecomposition by Monte Carlo method using the Ising model with chemical pair interactions calculated from the first principles.

We propose a new mechanism explaining the magnetic properties of MgO-based d0 ferromagnets determined from multi-scale simulations. Chemical pair interactions between N atoms in Mg(O,N) and Mg vacancies (VMg) in (Mg,VMg)O were calculated using a generalized gradient approximation and the VASP code. Monte Carlo simulations of the crystal growth were performed, using the Ising model, to predict the favored configurations of dopant distribution. It was found that self-organized nanowires can be formed both in Mg(O,N) and (Mg,VMg)O under layer-by-layer crystal growth, which suggests high blocking temperatures can be obtained in these d0 ferromagnets by spinodal nanodecomposition.

The electronic structures and the magnetic properties of dilute magnetic semiconductors (DMSs) based on transparent semiconducting oxide CuAlO2 are calculated by using the Korringa–Kohn–Rostoker method combined with the coherent potential approximation within the local density approximation. We see from the results that we can expect to obtain the half-metallic and high-spin ferromagnetic state is expected to be stable in Mn-, Fe-, Co- and Ni-doped CuAlO2.

We show that spinoadal decomposition phase in dilute magnetic semiconductors (DMS) offers the possibility to have high Curie temperatures (T C) even if the magnetic exchange interaction is short ranged. The spinodal decomposition is simulated by applying the Monte Carlo method to the Ising model with realistic (ab initio) chemical pair interactions between magnetic impurities in DMS. Curie temperatures are estimated by the random phase approximation with taking disorder into account. It is found that the spinodal decomposition phase inherently occurs in DMS due to strong attractive interactions between impurities. This phase decomposition supports magnetic network over the dimension of the crystal resulting in a high-T C phase.

We propose materials design of ferromagnetic diamond without any transition metal elements based on first principles calculations. The electronic structure and the magnetic properties of impurities-doped diamond are calculated by using the Korringa-Kohn-Rostoker method within the local spin density approximation with taking into account disorder using coherent potential approximation. It is found that H atoms which are doped into tetrahedral interstitial sites show finite local magnetic moments. Moreover, the impurities-doped diamond show ferromagnetism with half-metallic density of states.

We propose a new class of diluted magnetic semiconductors based on CaO without transition metal elements. The electronic structure and the magnetic properties of B-, C- or N-doped CaO are calculated by using the Korringa-Kohn-Rostoker method within the local spin density approximation. The substitutional and magnetic disorder is taken into account by the coherent potential approximation. It is found that B, C and N impurities show finite local magnetic moments in CaO at the oxygen-substitutional site. Moreover, these C- and N-doped CaO show the room-temperature ferromagnetism with half-metallic density of states.

A new class of 4d transition-metal-doped anti-CaF2 I2-VI diluted magnetic semiconductors (DMSs) is presented based on first-principles calculations. We have systematically investigated the stability of the ferromagnetic state in K2O-, K2S-, K2Se-, K2Te-, Li2S-, Na2S- and Rb2S-based DMSs. Anti-CaF2 I2-VI compounds have a large lattice spacing due to their large ionic radius of cations and many of them are wide-bandgap semiconductors. From the total energy differences between the ferromagnetic state and the spin-glass state, ferromagnetic solutions are derived and it is found that Zr-, Tc- and Ru-doped K2O, Zr-, Nb-, Tc- and Ru-doped K2S, and Zr-, Nb-, and Ru-doped K2Se, K2Te, and Rb2S are good candidates for transparent, half-metallic and room-temperature ferromagnetic DMSs with a large magnetoptical effect.

Curie temperatures are evaluated using first principles in (K,Zr)2S and (K,Nb)2S diluted magnetic semiconductors which have been designed as half-metallic and transparent materials with a large magnetoptical effect. From the total energy differences between the ferromagnetic state and the spin-glass state, estimations of Curie temperatures are achieved by mapping to the Heisenberg model within the mean-field approximation. According to the results, it is possible to fabricate transparent, half-metallic and room-temperature ferromagnets with Zr- and Nb-concentrations of around 5 at%. Effects of additional carrier doping for (K,Mo)2S are also evaluated and it is found that (K,Mo)2S with hole doping (∼5 at%) exhibits transparency, is half-metallic and has room-temperature ferromagnetism.

We propose a new route for ferromagnetic diluted magnetic semiconductors by controlling the impurity-band width (W) and the electron-correlation energy (U) in the partially occupied impurity band in the condition of highly correlated electron system (U>W). Based upon first-principles calculations of K2(S,Si) and K2(S,Ge), we demonstrate that transparent, half-metallic and room-temperature ferromagnetic DMS could be designed even without transition metal elements. The results show that it is possible to fabricate the room-temperature ferromagnets in K2(S,Si) and K2(S,Ge) around 8 at%-impurity concentration.

We propose materials design to fabricate the transparent and half-metallic ferromagnets in V or Cr-doped ZnS, ZnSe and ZnTe based upon ab initio electronic structure calculation with the local-spin-density-functional approximation. The electronic structure and the magnetic properties of 3d-transition-metal-atom-doped II-VI compound semiconductors of ZnS, ZnSe and ZnTe are systematically calculated using the Korringa-Kohn-Rostoker method with taking into account the substitutional disorder by the coherent potential approximation. It is shown that V or Cr-doped ZnS, ZnSe and ZnTe are ferromagnetic without p- or n-type doping treatment, however, Mn-, Fe-, Co- or Ni-doped ZnS, ZnSe and ZnTe are the spin-glass states.

Material design of GaN-based ferromagnetic diluted magnetic semiconductors is given based on ab initio calculations within the local spin density approximation. The electronic structure of 3d-transition-metal-atom-doped GaN was calculated by the Korringa-Kohn-Rostoker method combined with the coherent potential approximation. It was found that the ferromagnetic ground states were readily achievable in V-, Cr- or Mn-doped GaN without any additional carrier doping treatments. A simple explanation on the systematic behavior of the magnetic states in GaN-based diluted magnetic semiconductors is also given.

Detailed guidelines for controlling magnetic states in ZnO-based diluted magnetic semiconductors are given based on ab initio electronic structure calculations within the local spin density approximation using the Korringa-Kohn-Rostoker method. Effects of disorder were taken into account by the coherent potential approximation. It was found that the ferromagnetic state was stabilized by electron doping in the case of Fe-, Co- or Ni-doped ZnO. From the view point of practical applications, it is possible to realize a high-Curie-temperature ferromagnet, because n-type ZnO is easily available.

Ramo's theorem is generalized for the current statically induced in neighboring conductors by a specified movement of a charge carrier to include the higher frequency domain. As an example of application of the generalized theory, the current associated with carriers generated in a rectangular semiconducting material is computed. When taking the velocity difference between electrons and holes into account, the current normalized at low frequencies due to spatially uniform generation of carrier pairs in an intrinsic semiconductor is formulated as follows: I(ω)=∑ n=1,k (n/ωτ)2[(1+ jωτ/n)exp (- jωτ/n)-1], where k(>1) is the ratio of the electron velocity to the hole velocity, ω is the angular frequency, and τ is the transit time for a hole traversing through the device.

An expression is developed for the suppression factor of noise due to spatiaily exponential carrier generation in the depletion region of linearly graded p-n junctions. Under the assumption that the generation rate is expressed asU(x)=U0exp (-x/a) and the noise spectrum asS(ω)=2qIΓ2, it is found that the value of the noise suppression factor Γ2lies between 26/35 and unity, and depends only on the ratio of characteristic lengthato the width of the depletion region of the junction.

A theory of noise due to carrier generation in the space charge region of one-sided abrupt junctions with exponentially graded generation rate is presented. Assuming that the generation rate is expressed as C 0 exp (-x/a) and the noise spectrum as S(ω)=2q IΓ2, we find that the noise reduction factor Γ2 lies between 0.68 and 1, depending upon the ratio of the characteristic length, a, to the width of the space charge region.

Measurements of noise spectral density in the avalanche multiplication region have been made from 1 MHz to 4 GHz for four silicon photodiodes. The noise spectral density has been compared with the existing theories. At low frequencies the microplasma on-off noise, the theory of which has been successfully applied to one diode containing a single bistable microplasma, is pronounced. At high frequencies (≥100 MHz) one of the photodiodes exhibited a uniform-avalanche shot noise, which has been well described by McIntyre's theory with k=0.1, where k is the ratio of the hole ionization coefficient to the electron ionization coefficient. The rf (1 MHz~1 GHz) bias has been found to give rise to shifts of the I–V curves to higher voltages and changes in the noise spectral density. In some cases the noise spectral density is reduced. However, whether the noise is reduced or not depends on the property of microplasmas contained in the diode.

An improved method is presented for the extraction of the differential capacitance of a forward-biased Schottky-barrier diode. The present method has been applied to the capacitance measurement of Mo-nSi diodes with great success. The correction of the capacitance is much appreciable in the higher-conductance region compared to the results obtained by the existing theory.