Table of contents

The effects of alloying elements Al, Cr, Si, V, Ni and Co on the Neel temperatures and lattice parameters of gamma -Fe-Mn alloys are analysed. Our experimental results fail to confirm the universal relation between Neel temperature and lattice parameter proposed by Adachi et al (1973). It is suggested that the Neel temperature is not a unique and monotonic function of the lattice parameter for gamma -Fe-Mn host alloys.

We present the polarized far-infrared and Raman spectra of CuGeO₃ single crystals in the temperature range between 10 and 300 K. Assignations of vibrational modes were performed on the basis of factor-group and normal-coordinate analyses; The frequencies of the infrared-active modes are determined using an oscillator-fitting procedure of reflectivity data. In the Raman-scattering spectra, besides phonon modes, two broad structures at about 500 and 1600 cm^-1 are observed. The temperature dependences of the frequency, full width at half maximum (FWHM) and integrated intensities of these modes suggest their magnetic origin. In the temperature range we considered, no evidence of the spin-Peierls-transition contribution to the phonon and magnon spectra is found.

A new interstitial Cu²⁺ centre was found in NH₄Br crystals grown in acidic solution (pH<0.2) that was not observed in our previously studied sample (pH approximately=1.6). X-band EPR at 293 K indicates that the symmetry is tetragonal and the spin-Hamiltonian parameters are g/sub ///=2.044+or-0.002, A/sub ///=2.188+or-0.002, A/sub ///=0+or-5, A_{perpendicular to} =83+or-3 and A_{perpendicular to} '=30+or-1(A is given in units of 10^-4 cm^-1). A model H₂O-Cu²⁺-H₂O was assigned after studying the appearance of three types of centre in crystals grown with pH between approximately 8 and <0.2. Reexamination of the previous study on NH₄Br (pH approximately=1.6) has indicated that the centre obtained there may be of type NH₃-CU²⁺-H₂O, which can also be ascribed to the centre in NH₄Cl known as centre III.

We have performed positron lifetime experiments on semi-insulating (SI) GaAs in darkness at 20 K, before and after illumination with light of photon energy 1.32 eV. EL2 concentrations have been measured by infrared absorption (FTIR). After illumination, positrons detect a metastable vacancy whose trapping rate correlates with the total EL2 concentration. We conclude that this vacancy is included in the metastable configuration of EL2 (EL2*). This correlation allows us to propose a value of (3.0+or-0.3)*10¹⁶ S^-1 for the positron trapping coefficient at EL2* at 20 K.

High-resolution neutron powder diffraction measurements on Fe₃Ga_1.7As_0.3 and Fe₃Ga_1.15Sb_0.85 are reported. Substantial differences are observed in the refined crystal and magnetic structures of the two compounds. The compound Fe₃Ga_1.15Sb_0.85 is a disordered effect structure. B8₂ isotopic with the Ni₂In (p6₃/mmc) structure, whilst Fe₃Ga_1.7As_0.3 is a B8₂ variant ordered superstructure with a doubled basal plane lattice constant of the parent Ni₂In subcell. The Fe site occupancies have been refined for the subcell and the superstructure. A low-temperature magnetic structure, with moments orientated in the hexagonal basal plane, is found to give best agreement with the observed data for both compounds. The refined Fe magnetic moments for the superstructure are higher than those for the subcell.

Perturbed gamma - gamma angular correlation measurements with ¹¹¹In(EC) ¹¹¹Cd probes have been performed in the monoclinic (B-phase) rare earth sesquioxides Sm₂O₃, Eu₂O₃ and Gd₂O₃. The spectra in all three oxides show a dominating electric field gradient of high asymmetry, eta approximately=1, and strong coupling constant, nu _Q approximately=208 MHz, at room temperature. A slight modification of the structure in this phase with temperature was observed, but no phase transition occurred. Unlike the other structural isomorphs of the rare earth oxides, the cationic lattice sites in the B phase are not equally occupied by the implanted ¹¹¹In ions in the temperature range investigated (10-900 K). The point-charge model and symmetry considerations were used to associate the dominating electric field gradient with substitutional ¹¹¹In on the crystallographic site M(3). Two additional electric field gradients populated in small fractions were found in limited temperature ranges and tentatively attributed to defects.

We have calculated the electronic structure of amorphous hydrogenated Si-rich Si_1-xC_xH_y alloys paying particular attention to methylated Si, i.e. alloys produced at low deposition power with CH₃ configurations only and no C-C bonds. The configurational averaging method we use produces densities of states (DOS) which are not only a function of the compositional indices but also of the short-range order (SRO) parameters. By varying the DOS with respect to these parameters we are led to deduce the origin of all the experimental XPS peaks including the one at -14 eV which is assigned to Si(s) and C(s) orbitals in Si-C bonds. The calculated band gaps of the alloys produced at high deposition power are in good agreement with experiment and tentatively point to a small to medium degree of disorder. On the other hand the calculated band gaps of methylated Si is in remarkable agreement for x<0.25 when the CH₃ configuration is known to prevail but deviates considerably from the very high recently published experimental band gaps near x approximately 0.4 suggesting that the model of methylated Si may require modification for these values of x.

We report detailed band structure results on the series LnNiO₃ with Ln=La, Pr, Nd, Sm and Ho. LaNiO₃ is shown to be a pd metal, expected to be close to localization due to correlation effects; the other four compounds are suggested to be covalent insulators. There is a systematic reduction in the bandwidths with increasing atomic number of the rare-earth ion. Using a tight-binding analysis of the ab initio results, the reduction in the bandwidth is related to the changes in the intercluster hopping interaction strengths arising from the structural changes driven by the lanthanide contraction across the series.

We examine equivalent forms of the current-voltage relationship applicable to mesoscopic superconductors and use one of these to derive general conditions for the sign of the conductance in a four-probe measurement. For disordered systems, with well separated voltage probes, where the normal-state conductance is positive, we predict that the sign of the longitudinal four-probe conductance of superconducting wires can be reversed by varying an applied magnetic field, and that at certain critical values of the field, the conductance passes through a singularity.

The temperature dependences of bulk positron annihilation parameters for pure and Fe doped YBa₂Cu₃O_{7- delta} are considered in the context of temperature dependent positron trapping. Data for samples of composition YBa₂(Cu_1-xFe_x)₃O_{7- delta} with x=0.0, 0.005, 0.01, 0.02 and 0.03 are reported. Shallow traps with positron binding energies of 0.06+or-0.01 eV and 0.16+or-0.01 eV were identified for samples with x<or=0.01 and x=0.02 respectively. It is demonstrated that the data in the literature can be explained qualitatively by the presence of such traps and that the initial observation of a correlation between bulk positron annihilation parameters and superconductivity was coincidental.

The electrical conductivities of FeCl₃-doped polypyrrole samples with different dopant concentrations have been measured in the temperature range 25-300 K. Thermoelectric power measurements have been performed on two polypyrrole samples at low doping levels between 50 and 300 K. The variation in DC conductivity can be explained by Mott's variable-range hopping (VRH) model. Fitting with Mott's VRH theory, the conductivity data provide the values of the density N( epsilon _F) of states at the Fermi level and the hopping distance R. The low value and the sublinear character of the thermoelectric power is consistent with the VRH theory.

The problem of the so-called constant phase angle (CPA) behaviour in the frequency dependent impedance of an electrolyte in contact with a rough metallic electrode is studied by a numerical treatment and by real-space renormalization of two-dimensional electrical networks. Deterministic and disordered fractal boundaries based on quadratic Koch curves are considered. Both methods indicate that the CPA exponent eta can be estimated qualitatively from both the fractal dimension of the interface and from the scaling of the high-frequency response with system size.

We present a development of the dipole-trap model for the description of the field and temperature dependences of the non-dispersive charge-carrier transport in disordered organic matrices. Very good agreement between our numerical results and the phenomenological Gill-Pfister equation is obtained. Our consideration reveals the physical meaning of the parameters of this equation.

This paper presents a discussion about a spin model arising out of singleand double-electron exchange processes. The Hamiltonian which consists of the usual Heisenberg exchange, biquadratic exchange and three-atom coupling is studied using the Green function equation-of-motion method within the random-phase approximation. The effects of biquadratic exchange and three-atom coupling on the magnetization curves, quadrupolar ordering and the transition temperature T_C are studied. It is shown that the magnetization changes from second to first order at T_C depending on the strengths of the biquadratic and three-atom coupling. It is also seen that T_C becomes equal to zero at certain values of the parameters. The results are compared with those of molecular-field theory.

Using neutron inelastic scattering techniques we investigated the magnetic excitations in a diluted Heisenberg antiferromagnet (RbMn_cMg_1-cF₃ with c=0.39) close to the percolation threshold of c_p=0.312. No indication of propagating spin waves was observed throughout the Brillouin zone, but a broad peak superimposed on Ising-cluster excitations was observed. The origin of this broad peak is attributed to the excitation of fractons in a percolating network.

The magnetic anisotropy of the DyFe₁₁Ti intermetallic has been studied by measuring the angular dependence of the components of the magnetization parallel and perpendicular to the applied magnetic field. The measurements were performed on two plate-shaped samples with the surfaces parallel to the (110) and (001) planes, respectively. From the measurements in the (110) plane the thermal dependence of the spin reorientation angle has been obtained. The magnetic field dependence of M/sub /// and M_{perpendicular to} showed a first-order magnetization process (FOMP). A single-ion model for the crystal-electric-field interaction and a mean-field model for the exchange interaction have been used to explain theoretically the experimental results. An in-depth study of the character of the spin reorientation transition from easy cone to easy plane has been undertaken. From our study we show that for temperatures near this spin reorientation temperature, two magnetic phases (planar and conical phases) coexist in the sample, giving rise to an average angle that evolves continuously with temperature. A similar behaviour has been also observed for magnetic fields near the critical field of the FOMP, producing a continuous change of the magnetization instead of a jump at the critical field.

The magnetic properties of UCu₅ with the C15b-type cubic crystal structure, which shows an antiferromagnetic transition at 15 K and, further, an unidentified transition at 1 K, were investigated by using the nuclear magnetic resonance (NMR) and nuclear quadrupole resonance (NQR) techniques. The measurements were performed for both crystallographically inequivalent Cu sites with cubic and trigonal symmetry in all the paramagnetic, antiferromagnetic and unidentified states. The results clearly indicate that the low-temperature state below I K is another antiferromagnetic state with a different spin structure from the high-temperature state. From the temperature variation of the field-swept spectrum, we propose a non-collinear quadruple-q (4-q) structure for the state between 15 and 1 K, which is different from the structure previously proposed by Murasik et al based on their neutron diffraction result. Furthermore, we propose a collinear single-q (1-q) structure for the state below 1 K, which cannot be distinguished from the 4-q structure by neutron diffraction. Assuming a simple model for hyperfine interactions and the magnetic structures, we consistently explain the complicated zero-field spectra obtained both above and below 1 K. The analysis shows clearly the spin reorientation at 1 K and that the variation of the local field at Cu sites is reasonably interpreted in terms of only the spin reorientation. The analysis also indicates that the electric field gradient changes significantly at 1 K, which is in contrast to no change at 15 K. The relation between the spin reorientation and the gap formation at 1 K is discussed.

The electric-dipole polarizabilities of all the ions in the solid group IV dioxides ZrO₂, CeO₂, ThO₂ and UO₂ are deduced from experimental high-frequency dielectric-constant data. The resulting polarizabilities of 2.76+or-0.15 au, 5.54+or-0.15 au and 8.82+or-0.15 au for Zr⁴⁺, Cd⁴⁺ and Th⁴⁺, being unaffected by the crystalline environment, pertain also to these free cations. The polarizabilities of the O^2- ions are found to depend on their in-crystal environment, as parameterized by the closest cation-anion separation, in the same way as those of the anions in the M₂O alkali oxides. The ionic polarizabilities are used to derive values with an estimated error of 5% for the coefficients governing the dipole-dipole dispersive attractions between all ion pairs in all four crystals. Taken in conjunction with ion wavefunctions calculated elsewhere, these coefficients are used to compute values, with estimated errors of about 20%, for the coefficients governing the dipole-quadrupole dispersive attractions between all the ions in ThO₂ and UO₂. Values are presented for the parameters governing the damping of these dispersive attractions when the overlap between the wavefunctions of the interacting ions is not negligible.

The 2D scaling approach for the optical properties of semicontinuous metal films is generalized to thick (3D) percolative films. Model calculations demonstrate that the scaling model is qualitatively different from any effective medium approach, and should be used whenever the inhomogeneity length scale is larger than the optical one.

The spin- 1/2 ferromagnetic Heisenberg Hamiltonian with a Dzyaloshinski-Moriya interaction and in the presence of a strong magnetic field is studied in one dimension. Using the exact Bethe ansatz analysis, the excitation spectrum of one- and two-magnon states is derived. It is further shown that in certain parameter regimes the two-magnon bound state exists in the full range of allowed values of the centre-of-mass momentum wavevector k.

In this work, the luminescent quantum efficiency of Eu²⁺ ions in alkali halides is determined by using a method based on the simultaneous and multiwavelength measurement of photoacoustic and luminescent signals after pulsed laser excitation. This method, which was first demonstrated in Eu²⁺-doped KCl, is now extended in order to include the possibility of direct non-radiative relaxation from upper excited states to the ground state. A high quantum efficiency is found for Eu²⁺ ions in different alkali halides (NaCl, KCl, KBr and KI).