Table of contents

We report, in this letter, a simple thermal oxidation approach to growing large-scale arrays of aligned tungsten oxide nanorods on planar substrates. By passing a current of ∼50 A through a tungsten spiral coil in a vacuum of ∼3 × 10⁻² Torr, single-crystalline WO_2.9 nanorods, ∼20 nm in diameter and ∼400 nm long, were deposited on planar substrates, e.g. Si(001), Si(111), SiO₂/Si, glass, and Ag-coated silicon substrates at temperatures below 200°C. The nanorods were vertically aligned on the substrates. The structure, optical properties and alignment of the nanorods was investigated by x-ray diffraction and transmission electron microscopes, micro-Raman spectrometer, and scanning electron microscope and x-ray pole figure measurement, respectively. The present study provides an easy way to grow large-scale arrays of aligned tungsten oxide nanorods at relatively low temperatures, which might also be applicable to the growth of other metal oxides.

We have measured the electrical resistivity under pressure up to 2 GPa for an antiferromagnet, CeRhGe. The resistivity at ambient pressure decreases monotonically below a Néel temperature T_N = 10 K. With increasing pressure P, the resistivity has a peak around 10 K, which is anomalously enhanced up to 1.2 GPa. With further increasing pressure, the resistivity peak is diminished; the Néel temperature becomes zero at a critical pressure P_c = 1.9 GPa. The A-value of the resistivity ρ in the Fermi liquid relation ρ = ρ₀ + AT² is also found to increase with increasing pressure and has a maximum at P_c which corresponds to the value for the heavy-fermion superconductor CeCu₂Si₂.

We have studied second-harmonic generation (SHG) from quasi-one-dimensional aperiodic optical superlattices (AOSs) of finite lateral width by inverting poled ferroelectric domains. The search for optimal AOS structures corresponds with solving a difficult inverse source problem. We describe the design principle in real-space representation and undertake model designs. The numerical simulations show that the constructed AOSs can implement multiple-wavelength SHG with identical effective nonlinear coefficients at the pre-assigned wavelengths of the incident light. We investigate the effects of mode–mode coupling and the lateral width of the superlattice on the SHG for two cases: incident light beams of plane-wave and Gaussian profiles. When the number of modes increases, the effective nonlinear coefficient decreases in an oscillatory fashion at the beginning and then tends to a constant. For an incident plane-wave beam, the dependence of the effective nonlinear coefficient on the width of the sample is quite weak, while for an incident Gaussian beam this dependence exhibits a rapid decrease at the beginning and then tends to a constant. We display the variation in the effective nonlinear coefficients with the distance of propagation of the optical wave from where the incident light beam impinges on the sample surface and find that this variation exhibits monotonically increasing behaviour. This clearly infers that the contribution of every block to the optical SHG process takes the form of constructive addition. It is expected that this new design method may provide an effective and useful technique for constructing nonlinear optical material to match various practical applications.

An extensive investigation of structure, surface morphology and composition in Ni_75−xFe_xAl₂₅ alloys with nominal composition x = 0, 5, 10 and 20, deliberately 'prepared' in different states of site disorder, has been carried out by x-ray diffraction, scanning electron microscopy, x-ray fluorescence (energy dispersive absorption of x-rays) and inductively coupled plasma optical emission spectroscopy. The results of an elaborate analysis of the present x-ray diffraction data on Ni_75−xFe_xAl₂₅ (Ni-deficit) alloys and those reported previously on Ni₇₅Al_25−xFe_x (Al-deficit) alloys vindicate the recent theoretical prediction that the composition of the alloy plays a decisive role in determining the site preference of Fe by clearly demonstrating that Fe has nearly exclusive Ni (Al) site preference in the Ni-deficit (Al-deficit) alloys and thereby resolves a long-standing controversy surrounding the site preference of the ternary Fe addition in the γ^'-Ni₃Al phase in the Ni–Al phase diagram. However, contrary to the recent theoretical claim that entropy promotes Al site preference of Fe in Ni-deficit alloys, we find that Fe exclusively occupies the Ni sites in those Ni-deficit alloys that have a greater degree of site disorder.

Synthetic amorphous (Si_0.71Ge_0.29)O₂ has been investigated by high-energy synchrotron x-ray diffraction (E = 130.1 keV), in order to achieve a better understanding of amorphous structures built from the two network-forming oxides SiO₂ and GeO₂. The interatomic distances and bond angles that were determined are predominantly the same as in pure a-SiO₂ and a-GeO₂, respectively. According to the first coordination shell, the total coordination number CN(TO) = 3.4(5) (T = Si, Ge) indicates the existence of a considerable amount of cations that are connected to less than four oxygen ions. This result is clearly different with respect to the coordinations found in the structures of the respective pure amorphous oxides. If the second coordination shell is also taken into account, a second average (Ge–O) interatomic distance at r = 2.23(1)Å can be determined and CN(TO) increases to 4.2(1.1), indicating that all cations are actually connected to four anions. However, regular [GeO₄] tetrahedra are no longer supposed to be the only possible germanium–oxygen coordination polyhedra in the structure investigated. Based on these findings, we infer that SiO₂ and GeO₂ predominantly form a homogeneous but not completely continuous random network.

The low-and high-temperature structures of La₂CoMnO₆ and La₂NiMnO₆ have been refined using powder neutron diffraction. At low temperatures the materials adopt a charge-ordered monoclinic structure which transforms to a rhombohedral structure at elevated temperature without loss of charge ordering. This charge ordering scheme allows us for the first time to rationalize the relationships between the physical and structural properties of these materials.

It is shown that a contribution to the total scattering of neutrons by a single crystal of KHCO₃ held at 15 K arises from a purely quantum-mechanical and diffuse addition to scattering from identical particles which, in this instance, are protons in small dimers in a configuration described by a monoclinic crystal structure. The addition to scattering engages the total single-atom cross-section, and this feature explains why it is not visible in deuterated crystals where the relevant cross-section is down by an order of magnitude. Moreover, two other features of KHCO₃ make it a favourable case in which to observe the quantum-mechanical addition to scattering. First, the addition increases with decreasing nuclear spin I and I = 1/2 for protons. Secondly, there are two dimer orientations in KHCO₃ and their contributions to scattering overlap in reciprocal space and produce high intensity in the form of rods. Our successful model of the observed scattering represents the protons as an ensemble of independent dimers with the appropriate crystal arrangement. Scattering from a dimer is calculated including quantum-mechanical correlations of the nuclei (protons) that in a homonuclear diatomic molecule produce ortho- and para-states.

The atomic structure of the perfect AlPdMn icosahedral phase has been studied on a single-crystal specimen using the quantitative convergent beam electron diffraction (QCBED) technique combined with a description of the shapes of atomic surfaces using a symmetry-adapted series of surface harmonics. The spherical model (Boudard et al 1992 J. Phys.: Condens. Matter4 10149) was used as a starting point for the refinement. By fitting the calculated electron diffraction intensities in the CBED line profile to the experimental electron diffraction intensities, the coefficients in the surface harmonic expansion of the atomic surfaces' boundaries are refined. The refined parameters show that the fluctuations of the external boundary of atomic surface for Pd at n₀ can be as large as 0.2 nm. The boundaries of the atomic surfaces for Mn show little fluctuation. Compared to the spherical model, in the present model the number of unphysically short interatomic distances is reduced significantly. The three-dimensional atomic clusters with centres located successively at the six-dimensional lattice nodes n₀, n₁ and bc₁ are generated by means of the cut method.

The melting temperature–pressure (T_m–P) phase diagram for both bulk and nanocrystalline silicon is predicted by using the Clapeyron equation, where the pressure-dependent volume difference is modelled by introducing the effect of surface stress induced pressure. The predictions are found to be consistent with other theoretical results.

The large number of self-interstitials created during implantation mediate the fast transient diffusion of implanted boron, leading to clustering. Sophisticated annealing strategies based on knowledge of the formation energy of the clusters are required to achieve full activation of the implant. In recent years attempts have been made to determine these data a priori from theoretical calculations. However, energy calculations alone are not sufficient to establish the key players in the clustering process of boron. The present paper describes a systematic first-principles quantum mechanical study of the characteristic vibration frequencies of a large number of boron–interstitial clusters (including possible configurational isomers). Comparison with the first Raman spectra obtained on B-implanted samples after high temperature annealing is presented.

In spite of the large number of reports on the physical properties of LiKSO₄, its low-temperature phase diagram is still not well defined. One possible reason for this lack of reliable data below 100 K might be that LiKSO₄ crystals often break into many pieces when cooled below 80 K under atmospheric pressure. We have found that it is possible to thermally cycle LiKSO₄ crystals, particularly at temperatures below 80 K, as long as a minimum pressure of about 0.5 kbar is maintained. Thus, we successfully measured the temperature dependence of the sound velocity between 4 and 300 K for pressures up to 7 kbar. Over that temperature range, we clearly identify five different phase transitions (37, 48, 65, 185, 195 K) which correspond to those observed by other groups using different techniques. However, our results also show that both phase transitions below 50 K are strongly suppressed at pressures greater than 3 kbar. A Landau model of the free energy, based on the group theory, is also presented in order to explain the elastic and dielectric properties of LiKSO₄ above 100 K. To support our analysis, we show how this model accounts for the temperature dependence of the strains, polarizations, dielectric susceptibility and elastic constants.

This paper shows that an exact calculation for transition probability can make some systems deviate seriously from Fermi's golden rule. The paper also shows that the corresponding exact calculation of the phonon-induced hopping rate for a deuteron in the Pd–D system plus many-body electron screening, proposed by Ichimaru, can explain experimental factors observed in the Pd–D system. It also predicts that the perfect and low dimensional forms of the Pd lattice are very important factors for the phonon-induced hopping rate enhancement of deuterons in the Pd–D system.

We report the first straightforward measurements of the diffusion coefficient of hydrogen migrating in icosahedral (i-) quasicrystal ribbons of Ti₄₅Zr₃₈Ni₁₇ and in two-phase bulk samples containing both the i-phase and the Laves phase. It is found that at 273–353 K diffusion obeys Arrhenius's law with activation energy 0.24 eV and that the ratio of the volumes of the phases changes at hydrogenization.

The full-potential linearized augmented plane-wave method within the local-density approximation is used to investigate the effect of hydrostatic pressure on the structure of zirconium metal. Three phases are considered: hexagonal close-packed (hcp); body-centred cubic (bcc); and ω. For each of them the Gibbs free energy G = E + pV is calculated as a function of hydrostatic pressure. The three free energies as functions of pressure show three phase transitions at 0 K: hcp to bcc at 23 kbar; ω to bcc at 50 kbar; and ω to hcp at 87 kbar. The ground state at 0 K is ω up to 50 kbar and bcc above 50 kbar. Anomalies in the hcp structure and elastic constants at 0 K are found from 0 to 50 kbar.

First-principles calculations within density-functional theory are used to investigate zinc-blende-structured CrBi. Our calculation predicts ZB-type CrBi to be a true half-metallic ferromagnet with a magnetic moment of 3 μ_B per formula. Its half-metallicity can be maintained over a relatively wide range of lattice constant (from −8 to 8%). The trends with varying lattice constant for the ZB-type half-metallic system CrM (M  =  P, As, Sb, Bi) are also studied.

We derive exact quantum expressions for the localization length L_c for weak disorder in two- and three-chain tight-binding systems coupled by random nearest-neighbour interchain hopping terms and including random energies of the atomic sites. These quasi-1D systems are the two- and three-channel versions of Dorokhov's model of localization in a wire of N periodically arranged atomic chains. We find that, for weak disorder, L_c⁻¹ = Nξ⁻¹ for the systems considered with N = (1, 2, 3), where ξ is Thouless' quantum expression for the inverse localization length in a single 1D Anderson chain. The inverse localization length is defined from the exponential decay of the two-probe Landauer conductance, which is determined from an earlier transfer matrix solution of the Schrödinger equation on a Bloch basis. Our exact expressions above differ qualitatively from Dorokhov's localization length, identified as the length-scaling parameter in his scaling description of the distribution of the participation ratio. For N = 3 we also discuss the case where the coupled chains are arranged on a strip rather than periodically on a tube. From the transfer matrix treatment we also obtain matrices of reflection coefficients, which allow us to find mean free paths and to discuss their relation to localization lengths in the two- and three-channel systems.

Polycrystalline samples of the partially filled skutterudites Nd_xCo₄Sb₁₂ have been prepared and characterized by x-ray powder diffraction and differential thermal analysis. The saturation limit of the Nd void filling in CoSb₃ was found to be around 13%. All samples decompose incongruently at a temperature of 1149 ± 6 K. Room temperature Hall measurements show that each Nd atom donates approximately 0.8 electrons, which is significantly less than the Nd oxidation state (3+). The temperature dependence of the electrical and thermal transport properties has been measured over the range of 11–700 K. The electrical resistivity and absolute value of the Seebeck coefficient decrease with increasing Nd content and for samples with x > 0.02 the temperature dependence is typical of heavily doped semiconductors. Filling CoSb₃ with Nd causes a rapid initial decrease in the lattice thermal conductivity with a minimum at the composition Nd_0.1Co₄Sb₁₂. Nd-filled skutterudites exhibit the lowest value of the lattice thermal conductivity in comparison with other partially filled skutterudites at x < 0.1, which could be attributed to a smaller radius of Nd than that of other filling elements. At high temperature the ZT value of the Nd-filled skutterudites is limited due to intrinsic conduction caused by the relatively low carrier concentration. The effect of the partial Nd filling on the transport properties of the filled skutterudite compounds is discussed in the context of potential thermoelectric materials.

We present a theoretical study of the ballistic magnetoresistance in Ni contacts using first-principles, atomistic, electronic structure calculations. In particular we investigate the role of defects in the contact region with the aim of explaining the recently observed spectacular magnetoresistance ratio. Our results predict that the possible presence of spin-polarized oxygen in the contact region could explain conductance changes by an order of magnitude. Electronic transport essentially occurs through spin-polarized oxygen p states, and this mechanism gives a much higher magnetoresistance than that obtained assuming clean atomically sharp domain walls alone.

AgI–anatase composites, (x)AgI–(1 − x)anatase, were fabricated over a wide composition range of 0–100% AgI for the first time. The electrical conductivity at 300 K increases with increasing AgI content and reaches a maximum (about 3 × 10⁻² S m⁻¹) at about 40% AgI. The conductivity is enhanced by about three orders of magnitude in comparison with that of pristine AgI. These composite specimens have been characterized by both scanning electron microscopy and x-ray diffractometry. The observations with a scanning electron microscope show that anatase fine particles are densely packed in an AgI particle of several micrometres in size and such composite particles are three-dimensionally connected to each other to form a composite specimen. The photoluminescence (PL), PL excitation and time-resolved PL spectra of these composites have also been measured at different temperatures between 8 and 278 K. The exciton spectra obtained related to the anatase particles and the AgI domains in anatase-packed AgI particles are discussed in the light of the morphological, structural and conductivity data.

In this report we demonstrate that the width of the spin-split plateaus of the integer quantum Hall effect are determined by the quantum lifetime of the electrons via the exchange interaction. The widths of the adjacent Landau-split plateaus are a consequence of the pronounced changes of the spin-split plateaus. In contrast, the transport lifetime and thus the electron mobility are not directly related to the width of the quantum Hall plateaus.

Barium hexaferrite was synthesized by chemical co-precipitation. Its Mössbauer spectra were obtained. A semi-empirical model, based on the Phillips theory of bonding, has been developed for quantitative explanation of the Mössbauer isomer shifts of Fe ions in BaFe₁₂O₁₉ crystals. The results show that, using the relationship between isomer shifts and covalency, the site assignments in hexaferrites will be resolved easily. This paper provides a powerful tool for studying other members of the hexagonal ferrimagnetic oxides family.

Using an accurate density-functional method, we systematically explore vanadium chalcogenides in various crystal structures in order to find half-metallic (HM) ferromagnets which could possibly be fabricated as thin films with a thickness sufficient for real spintronic applications. The zincblende and wurzite (WZ) phases of VTe and the WZ phase of VSe are found to be HM ferromagnets with a magnetic moment of 3.000 μ_B/formula unit. Their HM gaps of, respectively, 0.31, 0.60, and 0.25 eV are 0.53, 0.56, and 0.52 eV higher in total energy/formula unit than the ground-state NiAs phase of VTe and therefore could possibly be realized in the form of films with great enough thickness. Their HM ferromagnetism persists even when the volume is compressed by 10, 20, and 5%. As they are compatible with the well-known binary semiconductors, these HM ferromagnetic phases, when realized experimentally, may be useful in spintronic applications.

The perovskite La_1/3Sr_2/3FeO_3−δ was investigated by neutron diffraction, magnetic and Mössbauer spectroscopy measurements. La_1/3Sr_2/3FeO_3−δ undergoes magnetic ordering at T = 190–200 K accompanied by charge disproportionation. Magnetic peaks due to charge ordering are observed below 200 K. The charge ordering is gradually developed below 200 K along with a charge disproportionation, 2Fe⁴⁺ ⇒ Fe³⁺ + Fe⁵⁺. La_1/3Sr_2/3FeO_3−δ shows an antiferromagnetic structure at low temperature. Magnetic moments of about 3 and 1.3 μ_B were obtained from the neutron diffraction data refinement for Fe³⁺ and Fe⁵⁺ at 15 K, respectively, which suggest that both Fe ions are in a low spin state. These values are significantly lower than those reported by Battle et al for La_1/3Sr_2/3FeO_2.98. Mössbauer spectra indicate that full charge ordering might be reached below 20 K with no Fe⁴⁺.

⁵⁷Fe Mössbauer spectroscopy of the metal-assembled layered ferrimagnetic material {N(n-C₄H₉)₄[Fe^IIFe^III(C ₂O₄)₃]}_∞ was performed under a high external magnetic field at 4.2 K. This study successfully detected different types of Fe^II and Fe^III high spin sites in the lattices. The hyperfine field values estimated for all the Fe^III high spin sites are close to the expected values, whereas those for Fe^II high spin sites are substantially different to those expected usually. A large fraction of magnetically disordered Fe^III high spin sites were found. It is understood that the layered structure of the compound may have two kinds of sub-lattices—(Fe_A^II–ox–Fe _A^III–···) and (Fe_B^II–ox–Fe _B^III–···)—along with randomly placed magnetically disordered Fe_C^III sites, where the Fe_A^III and Fe_B^III are antiferromagnetically aligned. The existence of the magnetically and/or structurally non-equivalent spin sites in the same lattice has been held responsible for the coexistence of different ferrimagnetic interactions leading to the spin glass state, as observed through earlier magnetic and calorimetric measurements.

Electron paramagnetic resonance (EPR) studies of Er³⁺ of single-crystal KY(WO₄)₂ have been carried out at X-band frequencies over a temperature range from 4.2 to 300 K. The main x- and z-axes of the g-tensor in the ac plane do not coincide with the crystallographic a-and c-axes, while the y- and b-axes are parallel. The EPR spectrum is characterized by a strong anisotropy of the g-factors (g_z = 13.25 ± 0.05, g_x = 0 ± 0.1, g_y = 3.378 ± 0.005) and of the hyperfine interaction parameters (A_z = −469 ± 3, A_x = 0 ± 3.5, A_y = −120.75 ± 0.8 in 10⁻⁴ cm⁻¹). The temperature dependence of the g-factors is caused by an anharmonic part of the spin–phonon Hamiltonian as well as by a rotational contribution to this Hamiltonian. A spin–lattice relaxation is shown to result from two Orbach processes through the first and second excited levels of an electronic spectrum.

The electron spin resonance line-width anisotropy and intensity are analysed for the quasi-one-dimensional organic conductors (fluoranthene)₂PF₆ and (perylene)₂PF₆·2/3 tetrahydrofurane in the metallic phase above the Peierls transition temperature. Based on the bottleneck model of relaxation, the temperature dependence of the intra-stack exchange constant between conduction-electron spins and localized defect spins is derived and discussed.

Optical pump–probe measurements have been performed on three materials with pump pulses of 120 fs duration and variable helicity. The samples were probed in a reflection geometry. We compare the transient intensity and rotation signals induced in intrinsic GaAs wafers and sputtered polycrystalline Ni and Al thin films. An initial peak is observed in the transient rotation signal. The dependence of the peak height on the polarization of the pump is found to be qualitatively different for the three materials, indicating different amounts of linear and circular birefringence in each case. For ferromagnetic Ni the peak is superimposed on a longer-lived ultra-fast demagnetization signal. For GaAs and Al we also observe a tail in the rotation signal that decays on time-scales of several picoseconds. In the case of GaAs, we attribute this to the relaxation of a spin polarization that is induced by the pump beam within the thermal electron population. In the case of Al, the dependence of the amplitude of the tail on the pump beam's polarization is characteristic of linear birefringence. We suggest that this signal is associated with an ultra-fast excitation of the lattice rather than with optical orientation of spin.

Three series of nanocrystalline powders of Lu₂O₃:Eu with Eu concentration varying from 0.05 to 13% were prepared. All specimens were obtained through combustion synthesis using urea or glycine as the fuel. The powders of the first series consisted of materials with crystallites about 13 nm in size and were prepared with urea. For the next series of powders, made with glycine, the crystallites were about 30 nm in size. The powders of the third series differed from those of the second one in that they were co-doped with 1% of Ca. For the series made with urea the quantum efficiency was highest for 3% of Eu and never exceeded about 30%. For glycine-prepared specimens the highest quantum efficiency was about 90%. Without the Ca co-doping such a value could be obtained for specimens doped with 5% of Eu, while in the series co-doped with Ca 85–90% quantum efficiency could be maintained for all concentrations in the range 3–10%. A significant number of OH groups were proved to be left in the final product obtained with urea. The low quantum efficiency of these powders is attributed to this effect. The results prove that properly prepared nanocrystalline phosphors can produce luminescence efficiently.

CaWO₄ phosphor films doped with rare-earth ions (Eu³⁺, Dy³⁺, Sm³⁺, Er³⁺) were prepared by the Pechini sol–gel process. X-ray diffraction (XRD), Fourier transform infrared spectroscopy, thermogravimetric and differential thermal analysis, atomic force microscopy, and photoluminescence spectra, as well as lifetimes, were used to characterize the resulting powders and films. The results of the XRD analysis indicated that the films began to crystallize at 400°C and that the crystallinity increased with elevation of the annealing temperature. The doped rare-earth ions showed their characteristic emissions in crystalline CaWO₄ phosphor films due to energy transfer from WO₄²⁻ groups to them. Both the lifetimes and PL intensities of the doped rare-earth ions increased with increasing annealing temperature, from 500 to 900°C, and the optimum concentrations for Eu³⁺, Dy³⁺, Sm³⁺, Er³⁺ were determined as 30, 1.5, 1.5, 0.5 at.% of Ca²⁺ in CaWO₄ films annealed at 900°C, respectively.

The photosensitivity of the field emission current from a GaAs crystal with an atomically clean emitting surface has been studied. Measurements of the optical and emission characteristics showed that the quantum yield can reach a magnitude considerably above 100%. A physical mechanism that could possibly lead to this abnormally large quantum yield (for a semiconductor field emitter) is discussed. The feasibility of using a GaAs field emitter as an efficient detection structure is demonstrated.

A new series of oxides, Ce_6−xEr_xMoO_15−δ (0.0 ≤ x < 1.5), was synthesized using wet-chemistry techniques. The precursors and resultant oxide powders were characterized by differential thermal analysis/thermogravimetry, x-ray diffraction, and IR, Raman and x-ray photoelectron spectroscopy. The formation temperature of the powders was found to be as low as 350°C. Ce_6−xEr_xMoO_15−δ crystallized to a fluorite-related cubic structure. The electrical conductivity of the samples was investigated by using ac impedance spectroscopy. This showed that the presence of Er was related to the oxygen-ion conductivity, and that the highest oxygen-ion conductivity was found in Ce_6−xEr_xMoO_15−δ (x = 0.4), ranging from 5.9 × 10⁻⁵ S cm⁻¹ at 300°C to 1.26 × 10⁻² S cm⁻¹ at 700°C, respectively. This kind of material shows a potential application in intermediate-temperature solid oxide fuel cells.