Table of contents

Structures of Mn/Sb multilayered films have been investigated by X-ray diffraction. MnSb/Sb superlattice films are obtained when the nominal thicknesses are (Mn(1 AA)/Sb(50 AA))₉₀ and (Mn(2 AA)/Sb(50 AA))₉₀. Ultrathin layers of MnSb are formed by interfacial reaction in these samples with keeping an epitaxial orientation of (00.1)_MnSb//(00.1)_Sb; (11.0)_MnSb//(11.0)_Sb. These films have fibre texture structures with an in-plane coherence length of about 300 AA. The distribution of growth orientation is approximated by a Gaussian distribution function with a half-width of about 10 degrees . The analyses of diffraction intensities indicate that (i) the texture structure causes a monotonic reduction of peak intensity with increase of scattering vector including fundamental reflection and (ii) that layer thickness fluctuations in the range of one or two monolayers cause peak broadening and intensity reduction only for the higher-order satellite reflections. Observed satellite intensities could be reproduced by a model calculation which takes into consideration these two effects.

A new formalism for the calculation of elastic multiple-scattering processes of electrons is presented, which takes the spherical character of the electron waves fully into account. It is based on a magnetic quantum number expansion of the electron Green's function and makes extensive use of recurrence relations. In the present paper this method is applied to the escaping electron in the final state of Auger electron spectroscopy (AES). Formulae for an analytic integration of the electron intensity over the aperture of the detector are derived and inserted into this calculation scheme. Within this framework angle-resolved M₂₃VV Auger electron intensities from a clean Ni(111) surface are calculated for the first time and compared with experimental data.

A 3D aperiodic structure model with cubic point symmetry is established by projection from a lattice in 6D space. Its Fourier transform describes well the diffraction patterns of the aperiodic phase in rapidly solidified V-Ni-Si and other transition-metal-silicon alloys. This aperiodic phase is suggested to be a quasi-crystal with cubic point group symmetry.

The authors have made a detailed study of the thermalization of positrons implanted into metals, using aluminium as an example. This is done by solving the Boltzmann equation for the positron momentum distribution in a homogeneous medium, allowing the positrons to scatter off electrons and phonons. They obtain both the time-dependent and steady-state solutions. The former gives the time evolution of the positron momentum distribution and the average energy and energy loss rate as functions of time after implantation. The full statistical description of the slowing-down process and the inclusion of both electron and phonon scattering mean that the authors' energy loss rates are more accurate than earlier results, which are considerably lower than the authors'. The steady-state solution gives the momentum distribution from which positrons annihilate. Their formulation allows them to evaluate the influence of non-thermal trapping into defects, such as vacancies, on measurable parameters in positron experiments. The results show that, even if resonances are present in the momentum-dependent trapping rates, the differences between the results from the full calculation and for trapping from a thermal momentum distribution are exceedingly small. Thus the authors' conclude that non-thermal trapping is not important for positron studies of vacancies in metals.

The theory of thermally activated motion of dislocations yields a cusp in the graph of activation free enthalpy Delta G versus temperature when applied to FeNi-invar alloys. This is attributed to a magnetic induced hardening of FeNi. A theory of thermal activation is developed taking into account the magnetic hardening and the results are compared with CRSS and yield stress measurements of FeNi and FeNiMn alloys.

The semiclassical cluster model for the propagation of shock waves in solids is applied to a linear molecular lattice with Morse interaction potentials. The exact quantum eigenstates for a cluster are first found and the interactions between neighbouring clusters are then considered as time-dependent perturbations which induce transitions among the cluster eigenstates. Calculations using the semiclassical cluster model qualitatively agree with classical calculations, although quantitative differences do arise in the atomic coordinates and molecular energies because of the anharmonicity of the potentials.

A simple mean-field lattice gas model is used to predict the hydride formation properties of Cr, Mn, Fe, Co, Ni, Mo, Ru, Rh, Ir, W, Pt, Cu, Ag and Au, which at low hydrogen pressures absorb only small amounts of hydrogen. It is shown that with thermodynamic parameters determined from low-pressure data only (the molar standard heat of formation Delta H_x⁰ and the Einstein temperature Theta _E of the interstitial hydrogen vibration) it is also possible to describe their high-pressure behaviour. As a result, the pressure-composition isotherms of these 14 metals are given for several temperatures and for pressures up to 1 Mbar. Where possible, comparison is made with experimental data.

Experimental observation of a (Pb_1-xCa_x)TiO₃ complex perovskite ceramic near x=0.5 has suggested that the Pb and Ca atoms are ordered alternately on a particular A site (111) plane. Furthermore, a body-centred superlattice ordered structure has been observed in (Ba_1-xPb_x)TiO₃ ceramic. In this paper, the eight-point cluster variation method, in which the second- and third-neighbour interactions are considered, has been employed to study these ordered structures and calculate the phase diagrams as a function of reduced temperature and composition for various interaction parameters. Theoretical results explained well the formation mechanism of these ordered structures.

Magnetic properties of stage 2 Co_cNi_1-cCl₂ graphite intercalation compounds (GICs) for 0<or=c<or=1 have been studied by DC magnetic susceptibility. These compounds can be considered to approximate two-dimensional randomly mixed ferromagnets with XY spin symmetry. As a function of Co concentration c the average effective magnetic moment P_eff, and the Curie-Weiss temperature Theta have been determined. The measurements indicate that Co²⁺ and Ni²⁺ spins are distributed randomly on triangular lattice sites of each intercalate layer. It is found that the intraplanar exchange interaction J(Co-Ni) between the different spins is larger than that between like spins, J(Co-Co) or J(Ni-Ni):J(Co-Ni):=1.2(J(Co-Co)J(Ni-Ni))¹2/. As a function of concentration, the susceptibility shows a rounding effect at the magnetic phase transition for a series of compounds with c=0.1, 0.19, 0.35, 0.40, 0.52, 0.63, 0.80 and 1.0. The authors have assumed a Gaussian distribution of transition temperatures to determine the average value of transition temperature (T_c), the distribution of transition temperature sigma and the critical exponent beta . It appears that the critical exponent beta is unchanged for 0.35<or=c<or=1.0: beta =0.09+or-0.01. They obtain for the initial slope of (T_c) versus c at c=0 the value -1.21+or-0.02. The ratio sigma /(T_c) shows a broad peak centred at c=0.5 which is explained here qualitatively by both finite size effect and macroscopic Co concentration gradient within the intercalate layer.

A fast method is presented for the calculation of total energies within the density functional formalism for systems that require a large number of k-points. Approximate solutions at a large number of k-points are obtained from exact self-consistent solutions at a smaller number of k-points using an extension of the k.p method. The method is demonstrated by a calculation of the total energy of FCC aluminium at a series of lattice parameters. An analysis is presented that partitions the errors in the calculations into those inherent in any finite sampling procedure and those that are specific to the k.p method. The extra error due to the k.p method is shown to be around 10% of energy differences for these calculations but a simple modification to the calculations requiring no more computational time reduces this error to 2%. For these simple calculations the time saved by using the k.p method is fairly small but for larger calculations the method is several orders of magnitude quicker. Errors from either source are shown to be far less important in the electronic potentials than in the eigenvalue sums. It is concluded that in any total energy calculation fewer k-points are required to describe the potential than to calculate the eigenvalue sum. This could be exploited to gain an order of magnitude saving in computational time in any calculation.

Photoconductivity, absorption and temperature dependence of conductivity were studied in CdTe:Fe crystals. The position of the Fe^2+/3+ donor level has been determined. It lies 130-150 meV above the top of the valence band. The value of the optical ionization energy has been confirmed as 1455 meV. The lattice relaxation energy has been found to be below 20 meV. The Auger photoconductivity originated by the ⁵E to ⁵T₂ intracentre transitions within the Fe²⁺ centres has been observed.

A theory is developed for dispersion characteristics of spin waves in ferromagnetic films taking into account both dipole-dipole and exchange interactions, crystallographic anisotropy and mixed exchange boundary conditions on the film surfaces. An arbitrary orientation of the external bias magnetic field with respect to the orientation of crystallographic axes and the film normal is assumed. The influence of crystallographic anisotropy on the spin wave spectrum of a ferromagnetic film is discussed. The theoretical results obtained are compared with the results of experiments performed in yttrium-iron garnet (YIG) and Mn-ferrospinel films.

A Monte Carlo computing procedure is presented for evaluating the set of parameters which characterize the superhyperfine interaction in electron-nuclear spin coupled systems. The method is illustrated in the determination of the components T/sub /// and T_{perpendicular to} of the interaction tensor T^{to or from} together with the linewidth Delta H_p-p of each of the superhyperfine transitions present in the electron paramagnetic resonance spectrum given by a cubic Gd³⁺ centre in CaF₂ recorded at room temperature. The different steps followed to simulate the 'annealing' of the system are described in detail together with the shortcuts proposed in order to minimize the computing time taken for the procedure to converge to the best set of parameters. This method is applicable even if no good estimates are available for the initial set of parameters as these are usually taken as zero. This modern approach to the complicated problem of fitting the superhyperfine structure to the theoretical expressions is shown to be the best alternative and even if it is time consuming it can be easily run on a microcomputer.

The authors calculate the Knight shift K in simple metals using a general expression for K and pseudopotential techniques. K is expressed as the sum K=K_s+K_o+K_so, where K_s, K_o and K_so are the spin, orbital and spin-orbit contributions to the Knight shift, respectively. Many-body and spin-orbit interactions are considered and their importance is emphasized. Metals with different crystal structures are considered. Results of the present calculations are compared with those obtained in some recent works. In contrast to other calculations, the authors obtain good to excellent agreement with experimental results in most of the cases. A remarkable feature of the authors' results is the explanation of the negative Knight shift in beryllium. The fact that the theory explains the Knight shift in several metals suggests that various important mechanisms are accounted for adequately.

Transmission electron microscopy and Mossbauer effect measurements were made on a coevaporated Fe₆₀Mg₄₀ alloy in order to investigate both the initial amorphous state and the amorphous-to-crystalline transition. In the initial state, the nearest-neighbour spacing was determined to be 0.257 nm while the Mossbauer spectroscopy exhibited a doublet, the average quadrupolar separation of which was 0.54 mm s^-1 at 293 K. A superparamagnetic-like behaviour was also detected at 77 K. The temperature range for the amorphous-to-crystalline transition was found to be 464-482 K. The index mu and activation energy Q in the Johnson-Mehl-Avrami plot were found to be 1 and 310 kJ mol^-1, respectively. The transition was clearly a eutectic crystallization of Fe+Mg.

The photoluminescence of Ti-doped sapphire (Ti:Al₂O₃) crystals excited using the second (532 nm) and fourth (266 nm) harmonics of a high power pulsed Nd:YAG laser has been investigated. Excitation at 532 nm into the crystal field band of Ti³⁺ produces the well known broad infrared emission band corresponding to the e_g to t_2g transition in the octahedral field approximation. However, this emission can also be excited with ultraviolet light together with an additional blue emission via a two-photon excitation process. This has two lifetime components of 600 and 21 mu s. The mechanism of excitation for the infrared emission band when it is excited via ultraviolet radiation has been studied: it is shown that the excited state of the Ti³⁺ ions is populated via an intermediate state (labelled as X) which clearly correlates with the shortest lifetime component of the blue emission.

A reconstruction of 3D electron-positron densities from 2D ACPAR experimental spectra for Mg was performed using Cormack's method and compared with theoretical results. In theoretical calculations the LMTO band structure method was applied. Many-body effects have been incorporated in a local density approximation both for valence and core electrons. These results lead to improved agreement between theory and experiment in comparison to the independent particle model. For the first time an over-enhancement of the higher momentum components was obtained both theoretically and experimentally.

A positron annihilation study of the icosahedral Al-Cu-Fe alloy system has been carried out using both Doppler broadening and lifetime measurement techniques. Isochronal annealing studies show that a nearly unchanged two-component lifetime spectrum is obtained up to the highest annealing temperature of 450 degrees C used in the present work, indicating the stability of the structure and the defect state in this quasicrystalline phase up to this annealing temperature. The vacancy defect concentration is about 8 ppm which is higher than that observed in the authors' earlier study on icosahedral Al-Mn-Si, though the vacancy size is lower-approximately a monovacancy size on average in Al-Cu-Fe against a divacancy size in Al-Mn-Si. The existence of such a high concentration of vacancy defects even in quasicrystalline Al-Cu-Fe, which has little phason disorder, should be significant from the point of view of developing a structural model for quasicrystals. It appears that a two-component positron lifetime spectrum indicative of a high concentration of vacancy-type defects is characteristic of the icosahedral quasicrystalline phase. The possibility of a central hole in the icosahedral building block in this phase is ruled out from the authors' positron annihilation experiments.

The KL_2.3 to L_2.3L_2.3 V Auger vacancy satellite transitions of Mg and Al, which lie to high kinetic energy of the KL_2.3 V Auger profile, are analysed using a combination of self-consistent densities of states (DOS) and atomic-like Auger transition rate calculations. The transition rate calculations demonstrate that previous work based on statistical weight arguments is incorrect. The DOS calculations show that the valence s contribution to the local DOS forms a bound state in the presence of a double core-hole state, though the major contribution to the core-hole screening is provided by the local p DOS. The calculation successfully reproduces the spectral profile for the Al transitions, but is unable to account for one of the features in the Mg spectrum.

The authors carried out a tunnelling spectroscopy measurement on a single crystal of superconducting oxide Bi₂Sr₂CaCu₂O_{8- delta} using a low temperature scanning tunnelling microscope. The superconducting gap parameter shows a spatial variation over a relatively small area ( approximately 10 nm). The authors obtained the energy gap parameter Delta (4.2 K)=26 meV, the most observable value, and correspondingly 2 Delta /k_BT_c=7, larger than the BCS value. The shape of the (dI/dV)-V curve suggests an anisotropy of the energy gap. Thermal smearing of the gap structure is observed below T_c. Temperature dependence of the gap parameter is consistent with BCS. Even above T_c the conductance curve is not flat.