Table of contents

Fe_xAg_1-x (x=0.20-0.65) nano-multilayers were prepared by alternate deposition of pure Fe and Ag metals. The multilayers were then irradiated with 200 keV xenon ions at 77 K. A new metastable crystalline phase with cubic structure was observed in the irradiated samples when the xenon ion dose exceeded 1*10¹⁶ Xe⁺ cm^-2 and the fraction of this new phase increased with ion dose.

It is shown numerically that the distribution of squared components of eigenvectors of the Anderson 1D tight binding equation on lattices of finite lengths, is parametrized by the single scaling parameter x= zeta infinity /N, where zeta infinity is the localization length for the infinite lattice and N is the number of sites of the finite lattice.

The authors discuss some of the fundamental properties unique to magnetic multilayers. Complex spin configurations are examined for many different systems and are shown to arise from a simple competition between exchange and Zeeman energies. The spin configurations found in multilayer systems determine macroscopic properties such as the static susceptibility and magnetization, and can lead to anomalous field and temperature behaviour. The authors also discuss the dynamic behavior of magnetic multilayers. Emphasis is placed on spin waves in magnetic multilayers with canted spin configurations and the softening of modes at magnetic phase transitions. Furthermore they show that spin wave excitations provide a powerful method for studying exchange interactions and spin configurations. Finally, the phenomenon of giant magnetoresistance in magnetic multilayers where the resistivity of the metallic structure can be changed by over 60% at room temperature, is discussed. Simple theoretical approaches are used to understand and predict the properties of the multilayer systems and comparisons between theory and experiment are stressed.

The energy spectra of Ne⁺ ions backscattered from a nickel surface were measured for chosen values of incidence angles and with two different target arrangements. It was found that the ratio of isotope components of the quasi-single-scattering peak reaches a maximum at some incidence angle. The maximum is attributed to the fact that ions doubly scattered at the atoms with smaller masses can have the same energy and exit angle as ions singly scattered at the atoms with larger masses. From geometrical considerations, the mean distance between two atoms of the topmost atomic layer can be found. This method may be used also for metallic alloys at low temperatures.

Optical second-harmonic generation (SHG) from mixed liquid crystal-stearic acid molecular monolayers, as a function of liquid crystal concentration, is reported. It is shown that, in concentrated monolayers, a substantial local-field effect exists which can be described, to a first approximation, by a point dipole model. The apparent molecular hyperpolarizability in pure monolayers of these liquid crystals is reduced by a factor of between six and eight, due to this local field. This is an important consideration in the design of materials for ultra-thin-film nonlinear optical devices. Hyperpolarizability measurements should be made using monolayer films where the active species is diluted by about a factor of ten by a species with negligible polarizability, such as a fatty acid, having first taken care to establish that the active species disperses uniformly. In this way the experimental hyperpolarizability may be directly compared to theoretical values from calculations of isolated molecular clusters. These LB monolayers are ideal model systems for testing theoretical approaches to the local-field problem, and progress in this area is important in understanding the optical response of surfaces and interfaces in general, and those of semiconductors in particular.

The energy of a plain boundary between a superconducting and normal region for a superconductor with a tricritical point is found as a function of the parameters entering the Ginzburg-Landau expansion of the free energy of such a superconductor. The authors define the regions of these parameters over which this energy is positive and negative. The structure of a quantized vortex is defined in the limit of large values of the parameter of Ginzburg and Landau, and the characteristic features of a temperature dependence of H_cl are discussed in comparison to those of usual superconductors.

The authors determine the bulk and surface canted-paramagnetic phase boundary of uniaxial antiferromagnets as a function of temperature for a semi-infinite system. Within the Green-function formalism, they show that the canted-paramagnetic critical field at the surface, which depends on the ratio the surface to bulk exchange coupling, is larger than that corresponding to the bulk. This critical field at the surface does not follow a T^3/2 Bloch law, but instead decreases linearly with temperature. At a given temperature, which also depends on the ratio of the surface to bulk exchange coupling, the surface critical field becomes identical to that of the bulk.

The misfit layer compounds with a general formula (MX)_{1+ delta} TX₂ (with M=Sn, Pb, Bi or rare earth metal; X=S, Se; T=Ti, V, Cr, Nb, Ta and 0.05< delta <0.25) have a stacking of MX double layers with a NaCl-type structure alternated by TX₂ sandwiches. The unit cell axes in the plane of the layers of both subsystems are equal in the b direction but are incommensurate in the a direction. To understand the stability of this remarkable and ordered stacking of the two different layers, charge transfer from the MX subsystem to the TX₂ subsystem has been suggested. In this study misfit layer compounds with Sn, Pb and Bi as the M element have been investigated with X-ray photoelectron spectroscopy. The information obtained from the core levels shows that for the Sn and Pb containing compounds the Pb and Sn atoms are divalent, and no significant charge transfer takes place between the layers. The spectra of the Bi misfit layer compound indicate a valency close to three for Bi. The stability of the misfit layer compounds is attributed to the presence of covalent interlayer bonds.

The authors have measured angle-resolved ultraviolet photoemission spectra from epitaxially grown thin films of CuBr with (111) orientation. Their analysis allows the determination of an experimental valence band dispersion very near Gamma -L-direction of the bulk Brillouin zone. These results show fair agreement with a KKR band structure calculation performed by Overhof. The authors' results, combined with earlier results obtained by 'constant-initial-state' photoelectron spectroscopy, are in agreement with all available optical data to within typically +or-0.1 eV.

Perturbed gamma gamma -angular correlation (PAC) experiments have been performed to study the occupation of different surface sites by ¹¹¹In probe atoms at vicinal Pd surfaces with (100) terrace orientation. Within an isochronal annealing sequence In probes are observed to dwell in adatomic and substitutional step and terrace sites, in agreement with earlier experiments on Cu, Ag and Ni surfaces. The temperature dependence of the electric field gradient for ¹¹¹In atoms occupying substitutional terrace sites deviates from the behaviour detected on Cu(100), Ag(100) and Ni(100) surfaces. The observed deviation in the low-temperature regime is attributed to the adsorption of hydrogen.

The values of atomic parameters, k and dk/dN, used in the analysis of Auger parameter shifts have been determined from the results of atomic structure calculations for the elements K to Kr. The values of the parameters for free atoms have been corrected for the effect of valence electron compression in solids. Calculations for Ni and Ge shows that the values obtained for the atomic parameters are independent of the configuration of passive valence electrons and of the character of the inner shell core hole.

Spin-resolved photoemission spectra from solid Xe(111) have been measured using circularly polarized synchrotron radiation with photon energies ranging from 11-18 eV. Corresponding calculations by a fully relativistic one-step-model layer KKR formalism produce fairly good agreement with the experimental data. The spectra can be interpreted in terms of direct interband transitions in the real part of a complex-potential band structure. This provides detailed information on the quasi-particle band structure including real and imaginary self-energy corrections to the valence and conduction bands. Observed 'opposite-spin features', which are absent in the calculated spectra, are tentatively interpreted as due to electron-hole scattering processes.

To study the electronic structure of ferromagnetic crystalline compounds, their surfaces and adsorbed ultrathin films, a fully relativistic Green function formalism has been developed. For a semi-infinite system, which is described by a complex effective potential the single-particle Green function and thence the layer-, k/sub ///- and spin- or symmetry-resolved densities of states-at the surface and in the bulk-are obtained using a layer Korringa-Kohn-Rostoker method. This Green function-for quasi-hole states-is employed in a relativistic one-step model formalism to yield spin- and angle-resolved photoemission intensities.

In the present paper the authors study the evolution of the ferroelastic domains and domain wall interactions of KSCN with temperature using a polarizing microscope. The typical lens shape of the ferroelastic domains in KSCN and the change of their shape (i.e. the width and length) with temperature are explained in terms of the dislocation theory for twin boundaries. The interface between the high- and low-symmetry phases was found to be at 45 degrees orientation with respect to the ferroelastic domain walls, in agreement with recent theoretical results.

For pt.III see ibid. vol.5, p.823 (1993). The author derives equations of motion for a new model to introduce temperature effects into Davydov's mod D₁) theory for solitons in proteins. The temperature model used so far suffers from conceptual difficulties, although it can reproduce exact quantum Monte Carlo (QMC) results qualitatively correctly as was shown previously. In their model they populate the lattice with a thermal phonon distribution and derive equations of motion in the presence of these thermal phonons, instead of deriving equations of motion from a thermally averaged Hamiltonian as in Davydov's theory. This procedure seems to be more reliable than Davydov's averaged Hamiltonian approach. For derivation of the equations of motion the Euler-Lagrange method is used. Numerical applications to soliton dynamics at 300 K are reported and comparisons of the results obtained with available exact QMC data are discussed. It is found that Davydov's method agrees better with the QMC results than does the newly derived method, although the latter method seems to be physically more reliable.

For pt.IV see ibid. vol.5, p.3883 (1993). For the two approximations commonly used in the Davydov soliton theory (called |D₁> and |D₂> ansatz), expressions for states which give the deviation of the approximation from the Schrodinger equation, i.e. [ih(cross)(partial/partial t)-H(hat)] |D₁> = J|delta>, can be derived. The author presents numerically calculated expectation values of various operators formed with the deviation states and compare them with the corresponding expectation values formed with H(hat)|D₁>. Together with the fact that the basis space of the |D₁> state is sufficient to reproduce the exactly solvable small-polaron limit, the author concludes from his results that the |D₁> model should be an at least qualitatively correct approximation.

The authors have determined the equilibrium configurations of O⁺, O⁰ and O^- states for O in GaP using a self-consistent semi-empirical method. A centred configuration is found for both the O⁺ and the O⁰ states, consistent with experimental results while the O^- state has the O atom heavily latticerelaxed from its original O⁰ state. The energy barrier between the metastable centred O^- and the stable heavily lattice-relaxed original off-centred O^- is only 0.12 eV, small enough to be compatible with experimental findings, with the defect levels (+/0) and (0/-) exhibiting an inversion in the ordering. Thus, a negative-U behaviour is demonstrated here and a charge-transfer reaction is postulated for O^- in the GaP system to account for the large lattice relaxation upon capture of the second electron.

The accurately determined energy levels at the W point of the FCC Brillouin zone obtained from low-temperature galvanometric measurements of the Fermi surface of aluminium are used to determine an empirical local pseudopotential and to test the precision of the standard scheme for calculations of band structure. While satisfactory for cohesive and structural properties, the standard scheme appears to reproduce the third-zone Fermi surface dimensions and topology rather poorly. Since there still remains some flexibility in the definition of ab initio pseudopotentials, it is suggested that agreement with the major cohesive and structural properties can be maintained, but that with relatively minor adjustment Fermi surface properties can also be reproduced. This notion is illustrated for the empirical local pseudopotential which is applied to the evaluation of the interatomic pair potential (which is found to be in good agreement with other calculated pair potentials) and also to a redetermination of the conductivity of molten aluminium which reveals rather clearly the limitations of the first-order Born approximation.

Optical-RF double-resonance techniques have been used to measure the superhyperfine transition frequencies of two Ho³⁺ centres of C_4v symmetry present in SrF₂:0.001%Ho³⁺ crystals containing about 0.2% of Ba²⁺ and Ga²⁺ as inadvertent impurities. One centre has a charge-compensating F^- ion located in the nearest-neighbour interstitial position along the (100) axis while the other has an additional Ba²⁺ ion also located on this axis. The superhyperfine transition frequencies are about 60% smaller than those reported for CaF₂:Ho³⁺ and these decreases are attributed principally to the decreased Ho³⁺ ground-state magnetic moment resulting from the increased energy separation between the ground and first excited state Ho³⁺ electronic energy levels. Another contribution arises from the larger host lattice dimension for SrF₂. The two sets of nearest-neighbour F^- ions have almost identical resonance frequencies for both SrF₂ centres while the interstitial F^- ion has a substantially reduced resonance frequency in the Ba²⁺ modified centre. These trends correlate with those in the relevant crystal-field parameters. Some qualitative conclusions are drawn about the positions of the Ho³⁺ and F^- ions in these centres.

Inter-Landau-level relaxation in two-dimensional electron and hole gases in high magnetic fields is strongly affected by the extent, lambda _z,/sub /of the electronic wavefunction perpendicular to the two-dimensional channel. As the field is increased, the suppression of the one-acoustic-phonon (cyclotron phonon) emission process that occurs when the phonon wavenumber at the cyclotron frequency exceeds lambda _z^-1 causes two-acoustic-phonon emission to replace one-phonon emission as the dominant relaxation process. The field at which this occurs has been calculated. The net reduction in relaxation rate is believed to be partly responsible for the sensitivity of the optical detection of cyclotron resonance in two-dimensional electron gas systems. In zero magnetic field, the two-phonon process should also contribute significantly to the energy and momentum relaxation of hot electrons.

The propagation of the circularly polarized ultrashort video pulse in a spin system with an anisotropic spin-spin interaction has been studied. A few physical realizations of this model have been suggested. Steady-state soliton-like solutions without using a slowly varying envelope and rotating-wave approximations have been obtained. The conditions of two-component video pulse bound-state formation have been investigated.

Despite the absence of a static spin-density wave (SDW), the cross section of paramagnetic chromium alloys still peaks near the wavevectors of the SDW in pure chromium. An analysis of the paramagnetic fluctuations in doped chromium reveals that the cross section vanishes at zero temperature and zero frequency, in agreement with recent experiments by Fawcett et al. The authors also predict that the peak elastic cross section will reach a maximum at a temperature which increases with the impurity concentration and that impurity scattering will eventually broaden the sidepeaks into a single central peak.

The metamagnetic transition and paramagnetic susceptibility for the pseudo-binary compound (Y, Lu)(Co, Al), with the cubic Laves-phase structure are measured. The observed results are analysed on the basis of phenomenological spin-fluctuation theory for itinerant-electron metamagnetism. A satisfactory agreement between experiment and theory is obtained qualitatively.

Two new axial Fe³⁺ centres. arising after reduction of stoichiometric lithium niobate crystals grown from melts containing K₂O, were discovered and investigated by EPR. Their b₂⁰ crystal field parameters are equal to 495 and 688 (*10^-4 cm^-1) and are thus much smaller than b₂⁰ for the axial Fe_Li³⁺ centre, 1660*10^-4 cm^-1, which had been studied before. The models of Fe³⁺ centres in lithium niobate and lithium tantalate crystals are compared and the possible structures-Fe_Nb³⁺ and Fe_Nb³⁺-K_Li-of the new centres are discussed.

The optical reflection spectrum of incommensurately modulated calaverite is presented. Using a Kramers-Kronig analysis the dielectric function is derived from the reflection spectrum. The results for the imaginary part of the dielectric function are qualitatively discussed in terms of absorption by phonon, free electron and interband scattering. The phonon contribution clearly shows the influence of the incommensurability by the activity of phonons with k not=0. The free electron scattering can be understood in terms of a simple Drude model. The interband scattering is found at a lower frequency than predicted from electronic band structure calculations.

The defect structures in deformed and helium-irradiated refractory metals, vanadium, niobium and tantalum have been investigated by means of positron lifetime and Doppler broadening techniques. The evolution of the defects was studied by performing isochronal annealing programmes. In all three metals after deformation a single, well defined lifetime was observed which was 30 ps to 40 ps lower than that reported for vacancies. The recovery of the deformation-induced defects however, is remarkably different for niobium and tantalum as compared to vanadium. In helium-irradiated vanadium and niobium two distinctly different defect types were observed with characteristic lifetimes of about 190 ps and 375 ps, respectively. The longer lifetime is attributed to large helium bubbles. Both metals showed a similar behaviour in the annealing studies. All measured lifetime data are consistent with the simple trapping model. The influence of oxygen impurities and oxygen resorption during the isochronal annealing procedure is discussed.