Table of contents

Density functional theory (DFT), when applied to systems with T≠0, is based on the grand canonical extension of the Hohenberg-Kohn-Sham theorem due to Mermin (HKSM theorem). While a straightforward canonical ensemble (CE) generalization fails, work in nanopore systems could certainly benefit from a mesoscopic DFT in the CE. We show that, if the asymptotic behaviour of the canonical distribution functions is taken into account, the HKSM theorem can be extended to the CE. We generate N-modified correlation and distribution functions hierarchies, show that their functional relationship is equivalent to the one holding between the more conventional ones and prove that, if they are employed, either a modified external field or the density profiles can be indistinctly used as independent variables. We also write down the N-modified free energy functional and prove that its minimum is reached when the equilibrium values of the new hierarchy are used. This completes the extension of the HKSM theorem.

Abnormal structural changes were observed in the equilibrium immiscible Co₅₀Cu₅₀ multilayers induced by 200 keV xenon-ion irradiation at 77 K. First, a dodecagonal quasicrystal phase was formed at an irradiation dose of 1×10¹⁵ Xe⁺ cm^-2. Secondly, series of selected diffraction patterns taken later showed that the dodecagonal phase formed evolved to feature a high degree of ordering with increasing ion dose. Moreover, molecular dynamics simulations verified the possibility of formation of a metastable phase in the Co-Cu system. A mechanism possibly responsible for the observed structural changes upon ion irradiation is also discussed.

Recent work on magnetic properties of transition-metal nanowire arrays produced by electro-deposition is reviewed. The wires, which are electro-deposited into self-assembled porous anodic alumina, form nearly hexagonal arrays characterized by wire diameters down to less than 10 nm, wire lengths up to about 1 µm, and variable centre-to-centre spacings of the order of 50 nm. The fabrication and structural characterization of the arrays is summarized, magnetic data are presented and theoretical explanations of the behaviour of the wires are given. Emphasis is on extrinsic phenomena such as coercivity, magnetization reversal and interactions of the magnetic nanowires. In particular, we analyse how wire imperfections give rise to magnetic localization and dominate the hysteresis behaviour of the wires. Potential applications are outlined in the last section.

This article reviews our recent experimental studies of domain wall (DW) resistivity in epitaxial transition metal ferromagnetic thin film microstructures with stripe domains. The results are presented and analysed in the context of models of DW scattering and conventional magnetoresistance (MR) effects in ferromagnetic metals. Microstructures of progressively higher magnetic anisotropy and thus smaller DW widths have been studied, including; bcc Fe, hcp Co and L1_o FePt. The magnetic domain structure of these materials have been investigated using magnetic force microscopy and micromagnetic simulations. In Fe and Co the dominant sources of low-field MR are ferromagnetic resistivity anisotropy, due to both anisotropic MR (AMR) and the Lorentz MR. In Fe, at low temperature, a novel negative DW contribution to the MR has been found. Hcp Co microstructures show a greater resistivity for current perpendicular to DWs than for current parallel to DWs, that is consistent with a small (positive) DW resistivity and a Hall effect mechanism. High anisotropy L1_o FePt microstructures show strong evidence for an intrinsic DW contribution to the resistivity. Related studies and future directions are also discussed.

A comparison of orthogonal and non-orthogonal localized wavefunctions for Si in the diamond structure is carried out. We have used a real-space grid formulation of density functional theory in combination with the local density approximation for exchange and correlation to describe the energetics. Maximally localized wavefunctions, obtained from the extended Kohn-Sham states with and without an orthogonality constraint, are calculated and it is found that the wavefunctions calculated without any orthogonality constraint are the most localized. When solving directly for localized states, by applying a localization constraint to each electronic state, we find that there is a large difference between orthogonal and non-orthogonal states: when the localization region is a sphere with a radius of 3.0 Å, we get an error in the total energy due to the localization constraint of 0.2 and 2.7 eV/atom for non-orthogonal and orthogonal wavefunctions respectively.

The compression behaviour of a Pd₃₉Ni₁₀Cu₃₀P₂₁ bulk metallic glass is investigated at room temperature up to 23.5 GPa using in situ high pressure energy dispersive x-ray diffraction with a synchrotron radiation source. Pressure induced structural relaxation of the bulk metallic glass is exhibited within the pressure range. It is found that below about 5 GPa, the existence of excess free volume contributes to rapid structural relaxation, which gives rise to rapid volumetric change. Under higher pressure, further relaxation results in structural stiffness.

We use an alternative operator transform to study hole motion against a frustrated antiferromagnet background described by the t-J model with next-nearest-neighbour exchange energy. Using the self-consistent Born approximation, the optical conductivity and direct-current resistivity are evaluated, and their dependences on the next-nearest-neighbour exchange energy are discussed.

The multiplet structures of tetrahedrally coordinated Cr⁴⁺ in the three silicate crystals Mg₂SiO₄ (forsterite), Ca₂MgSi₂O₇ (åkermanite), and Y₂SiO₅ (yttrium orthosilicate (YSO)) were calculated by the many-electron electronic structure calculation method developed by the authors. The method is a hybrid of the molecular orbital method based on the density functional theory and the configuration interaction approach. For every crystal, the calculations were conducted by using cluster models with the three sizes: (A) (CrO₄)^4- (without point charges) models, (B) (CrO₄)^4- (with point charges) models, and (C) (CrMg₉Si₂O₃₇)^44- (forsterite), (CrCa₆Mg₂SiO₃₈)^52- (åkermanite), and (CrY₈O₃₇)^46- (YSO) models. The calculated multiplet energies of the triplet states agreed with the experimentally obtained peak energies in the absorption spectra in the literature. The theoretical spectra showed polarization dependence of the peak intensity. The best agreement was found in the results obtained from the largest models C. The difference in polarization dependence between Cr⁴⁺:forsterite and Cr⁴⁺:åkermanite was related to the different mixing of the many-electron wave functions as regards the ³T₂(et₂) and ³T₁(et₂) triplet terms. The covalency of the impurity-level molecular orbitals was also analysed. The results of models C indicated that the wave functions of the atoms outside the CrO₄ tetrahedron should not be neglected. Both the degree of covalency and the correlation-correction factor, which was introduced in the method, were regarded as reduction factors of two-electron repulsion. The two factors were multiplied together, and the reduction factor was a convenient indicator for simply evaluating the magnitude of the reduction. The traditional nephelauxetic parameter was obtained as 0.49. Some empirical values given recently in the literature were confirmed to have appropriate magnitude.

The transport and magnetic behaviours of Ar-annealed (O-depleted) bulk samples of an archetypal, `optimally' doped manganese perovskite, La_0.67Ca_0.33MnO₃, are reported, analysed and compared with those of an untreated specimen. These indicate that the transport data for both the paramagnetic and ferromagnetic phases of the O-depleted system are consistent with the predictions based on polaronic hopping, specifically non-adiabatic small-polaron hopping (on the basis of the internal consistency of estimates for the characteristic frequency Ω₀ from both the adiabatic and non-adiabatic formalisms). These results yield indirect support for models which attribute transport in doped perovskites in general, and the colossal-magnetoresistance phenomenon in particular, to polaronic behaviour.

The ground-state energies of several interacting electrons (N⩽12) confined in a parabolic quantum dot with an impurity ion at the centre are obtained by numerical diagonalizations. Series of magic values of angular momentum are determined. The rules for identifying the magic numbers are established.

Electron absorption, excitation and emission as well as electron spin-resonance spectra were measured for orthorhombic (K, Rb, Cs) and monoclinic (Li, Na) M^ICr_xIn_1-x(MoO₄)₂ (x = 0.5-2%, M^I = Li, Na, K, Rb, Cs) in the temperature range 5-300 K. The local structure of the Cr³⁺ ions is discussed on the basis of spectroscopic results and the crystal-field parameters are derived for all of the materials. The possible application of the compounds studied as laser materials is discussed.

The low temperature TL spectra of undoped alkaline earth fluorides consistently show a continuous broad band emission around 280-300 nm. The origin of this emission is related to the relaxation of the self-trapped exciton (STE) in the form of an F-H pair. In doped samples, the broad emission is quenched in favour of emissions from the rare earth (RE) impurity sites. The degree of quenching varies between the REs. The spectral measurements showed that the temperature of the glow peaks below room temperature is to a large extent independent of the RE impurity, although some effects are related to the concentration. Additionally, the host material has minimal effect on the glow peak temperatures, T_max. The scale of the observed differences in T_max between the three hosts is similar to the differences in annealing temperatures of the V_k and H centres in these hosts. Above room temperature, the glow peaks are specific to the added RE ions and do not show common peaks.

We investigate the statistical properties of the conductance through a quantum dot in the Coulomb blockade regime. By taking into account the charging energy we calculate the correlation function of conductance by the use of two types of energy level distributions of the dot, the Poisson distribution and the Wigner–Dyson distribution. In both cases, the conduction correlation obtained as a function of the difference in occupation number shows similar behaviour, decaying in the range where occupation differences are small, in agreement with experiment, but exhibits different tail behaviour where occupation differences are large.

The crystal and magnetic structures of Y_n+1Co_3n+5B_2n (n = 2, 3 and ∞) have been studied by high-resolution powder neutron diffraction. The results are compared to earlier measurements on YCo₅ and YCo₄B. A change in the regular stacking of the boron-containing plane along the c axis has been observed in the Y₂Co₇B₃ sample. Very short Co–B distances are observed, indicating that strong bonds are formed between cobalt and boron. The YCo₃B₂ compound is paramagnetic down to 2 K. The magnetic structures of Y₃Co₁₁B₄ and Y₂Co₇B₃ confirm the large variety of cobalt magnetic moments obtained in these compounds. The magnetic behaviour of the Co(2c) atoms is not significantly affected by the substitution of boron for cobalt. Cobalt atoms with significantly reduced magnetic moments are found on the 3g and 6i₂ sites in both Y₃Co₁₁B₄ and Y₂Co₇B₃. A relationship between the magnitude of the Co magnetic moment and the presence of boron in the neighbourhood of the cobalt atoms is proposed. The hybridization of the cobalt 3d electronic state with the boron 2p state is found to play a major role in the determination of the magnitude of the Co magnetic moment in the Y_n+1Co_3n+5B_2n compounds.