Table of contents

The structure of the high-pressure phase of KTP has been determined at 6.2 GPa using Patterson projections, Fourier difference maps and least-squares refinement. Although the phase transition is strongly first-order and involves a 4.2% reduction of the unit-cell volume, the Pna2₁ spacegroup symmetry of the low-pressure phase is retained. The main structural changes involve tilting of the PO₄ tetrahedra relative to the TiO₆ octahedra, while both units remain relatively unchanged in size and shape. In particular, the TiO₆ octahedra remain highly distorted and retain their anomalously short Ti-OT bonds. The potassium atoms are observed to undergo relatively large movements of about 0.5 AA along the polar c-axis.

Experimental results with circularly polarized synchrotron radiation show that magnetic Compton scattering arises solely from the spin magnetization in the sample. This observation disagrees with the cross-section recently derived for bound electrons. The measurements were performed on HoFe₂, a ferrimagnet with dominant orbital magnetization.

The authors calculate the Helmholtz free energies of the liquid alkali metals at various temperatures and those of several polyvalent metals (Mg, Cd, Al, In, Tl and Pb) near freezing, using the variational approach based on the Gibbs-Bogoliubov inequality in conjunction with the charged-hard-sphere reference fluid and with ab initio non-local pseudopotentials for the electron-ion interactions. The reference fluid introduces two variational parameters, i.e. the plasma coupling strength Gamma and the packing fraction eta , and is treated by a thermodynamically self-consistent approach which reduces to a highly accurate description of the one-component classical plasma for eta =0 and of the neutral-hard-sphere fluid for Gamma =0.

The nature of the wetting transition in a binary Sullivan fluid mixture is studied by means of a dynamics approach. Analytical study of the two-phase coexistence system shows that the order of the wetting transition depends mainly on the strength of the fluid-fluid interaction and that the substrate does not play an essential role The results on three-phase coexistence systems obtained by Hauge (1987) through Landau mean field theory are recovered by the authors treatment and some new results are obtained. A thin-thick transition might occur in such a short-range force system; a phase diagram of the wetting transition constructed from the two components of wall-fluid potential is obtained. The method employed in the present paper could be used to investigate the wetting behaviour in systems with more components.

The thermal conductivity ( lambda ) and heat capacity per unit volume ( rho c_p) of amorphous SiO₂ have been measured under hydrostatic pressure up to 1 GPa at room temperature using a transient two-strip method. The thermal conductivity was found to decrease slightly with increasing pressure. For the volume dependence of lambda expressed by the Bridgman parameter, g=-( delta ln lambda / delta ln V)_T, the authors obtained g=-1.5. The experimental value of g is discussed in conjunction with predicted g-values based on theoretical models of lambda (T). The specific heat capacity (c_p) calculated from measured rho c_p was found to be almost constant with increasing pressure.

The two-dimensional (2D) S-1/2 Heisenberg antiferromagnet copper (II) formate tetrahydrate (Cu(HCO₂)₂.4H₂O) has been doped with the diamagnetic ions Mg²⁺ and Zn²⁺ up to levels of 15 at.%. The temperature dependence of the electron spin resonance (ESR) signal provides a way of accurately determining the differences in the Neel temperatures, T_N(x), of samples with different dopant concentrations, x. There is a linear decrease of T_N with increasing dopant concentration: T_N(x)=(1- alpha x)T_N(0) where the constant alpha is 3.4+or-0.1 and T_N(0) is the Neel temperature of the undoped material. This behaviour is compared with that of La₂CuO₄ and other 2D systems. The width of the ESR signal at a particular temperature is also found to decrease with increasing dopant concentration. These features, together with the variation of linewidth with temperature characteristic of 2D systems, are discussed in the light of the interactions between the magnetic ions.

The crystal energy of iron pyrite, ideally of composition FeS₂, is evaluated with an ionic model, assuming atoms in the solid to be ionized as Fe²⁺ and S^-. To calculate the proper crystal energy, terms accounting for the static electric dipole moments of the sulphur ions have to be included. For this purpose, so-called electrostatic lattice constants are introduced, with Madelung's constant being the first term of a Taylor expansion for the electrostatic interaction energy. Following these considerations, a theoretical crystal energy of -28.35 kJ per mole of pyrite is calculated. The formalism gives a first estimation of the strength of the sulphur dipole moment in the FeS₂ lattice: mu s=12.3*10^-30 C m (3.7 D). The thermodynamic Born-Haber cycle is generalized to include polarized species, yielding an 'experimental' value for the crystal energy of -2893 kJ per mole of pyrite, which differs only by 2% from the theoretical one.

The model calculations of lattice dynamics in solid biphenyl and terphenyl for different temperatures and pressures are presented. The authors develop a model that explicitly treats interphenyl motion in the self-consistent phonon approximation. The phase transition is interpreted as an instability (zero frequency at finite wave vector) of the twist mode with weak coupling to the external modes of the molecule. The calculations show fairly good agreement with experimental data.

Using the data from LSDFA LMTO ASA band-structure calculations the authors evaluate the zero-temperature Knight shifts K at the individual atoms of the stoichiometric high-T_c material YBa₂Cu₃O₇. All contributions to K are calculated-without using adjustable parameters-on the level of the previously described RPA-like spin-density-functional formalism approach. As a whole the results for the Cu atoms turn out to be in reasonably good agreement with experiment.

The intrachain and interchain magnetic interactions in the (In;Sc;Lu;Y)₂Cu₂O₅ ternary oxides have been estimated on the basis of the Heisenberg chain model for S=1/2, taking into account only the high-temperature experimental data of the magnetic susceptibility. The authors have calculated the first three coefficients in the high-temperature expansion series of the magnetic susceptibility for the three-dimensional model of weakly coupled Heisenberg chains. The temperature dependence of the magnetic susceptibility has been compared with the predictions of the models mentioned above. The theoretical values of T_N obtained by numerical calculations of the Oguchi expression, are in good agreement with experiment. The applied model of the ferromagnetic chains being mutually weakly coupled antiferromagnetically turns out to be a good approximation for the real magnetic system existing in the (In;Sc;Lu;Y)₂Cu₂O₅ oxides.

It is argued that collision damping of excitations in a modulated (quasi-periodic) system has a pronounced effect on results for the frequency- and wave-vector-dependent response function observed in scattering experiments. Even a relatively small damping can remove all vestige of some of the striking features predicted for the undamped model. Results are obtained for a model that can be solved without approximation in the absence of damping. The latter is introduced by extending the frequency to complex values, and the magnitude of the imaginary component is viewed as the damping control parameter.

Two-magnon bound states are exact eigenstates of the Heisenberg ferromagnet whose energy falls outside the two-magnon continuum. The well-grounded expectation that at least one bound state exists for any wave vector in 2D Heisenberg ferromagnets seems to be violated when numerical calculations are performed in the triangular lattice In contrast the authors show analytically that such a result is a spurious consequence of the numerical computation and they are able to prove that a bound state exists below the two-magnon band for all wave vectors in the triangular ferromagnet.

In order to study the effects of frustration on the magnetic properties of the antiferromagnetic triangular lattice LiNi_1-xCo_xO₂, the authors measured the temperature dependence of both the magnetization and the AC susceptibility. With a random substitution of non-magnetic Co ions for Ni ions, the characteristic temperatures T_N1 ( approximately=240 K) and T_N2( approximately=35 K) gradually shifted to higher and lower temperatures, respectively, and the magnetic moment per Ni ion was drastically enhanced between T_N1 and T_N2. The spin angular momentum of the Ni³⁺ ion, however remains at about 0.5, confirming the low-spin state over the whole range of concentrations. They also performed neutron-scattering experiments and found that the intensity of diffuse scattering in the forward directions at low temperatures decreases on substitution. A new model for the magnetic interactions in LiNiO₂ is proposed on the basis of these results.

Magnetization measurements have been performed on the amorphous alloys RE₄₀Y₂₃Cu₃₇ (RE=Tb, Dy, Ho and Er) over a wide range of magnetic fields (0 to 7 Tesla) and temperatures. These systems show no spontaneous magnetization and are characterized as disordered magnets with random magnetic anisotropy (RMA). Although the low field susceptibility does not diverge it is close to the demagnetizing limit for T<or=T_SG. The ratio D/J of the strength of the RMA axial crystal field parameter D to the ferromagnetic exchange parameter J, is large for these systems. The authors have performed a ferromagnetic-like scaling analysis of the magnetization data for large applied fields, thereby determining the critical RMA exponents, beta _a, gamma _a and delta _a The non-linear susceptibility, chi _nl, is taken to be the order parameter, as for canonical spin glasses; chi _nl also shows critical behaviour. Both of these results point to a phase transition at T_SG. The thermal variation of the high-field magnetization and also of the magnetostriction support this conclusion, as does the existence of transition lines of the de Almeida-Thouless type.

Positron lifetime measurements were used for studying decomposition during post-quench in situ annealing of binary Al-7.35 at.% Li and ternary Al-7.3 at.% Li-X (X=1.1 at.% Cu, 5.2 at.% Mg or 0.18 at.% Sc) alloys. A typical lifetime of 185+or-5 ps observed in Al-Li is attributed to coherent delta '(Al₃Li) particles. In ternary alloys positrons show a strong response to the formation of semicoherent and incoherent phases such as T₁(Al₂LiCu) and Al₂LiMg. Positron trapping in Al-Li-Sc is attributed to composite delta ' (Al₃Li)/Al₃Sc particles associated with vacancies. Quenched-in vacancies were detected in Al-Li-Sc and Al-Li-Mg but not in Al-Li-Cu and Al-Li.

Based on the dichroism of the FA (II) KCl:Li centres the off-centre angle of the Li⁺ ion was measured to be 13.5+or-1 degrees at 26 K. Furthermore, the birefringence of the centres was verified and the difference between the refractive indexes relating to the two preferred directions was directly derived from the absorption spectra measured with polarized wave.

The lifetime of luminescence in KI:Sn²⁺ crystals has been investigated using a streak camera. By excitation in the absorption band B due to a dipole-forbidden ¹S_o to ³P₂ transition in the Sn²⁺ centre, an emission band with a peak at 446 nm is produced by the reverse transition ³P₂ to ¹S₀. The emission B has a lifetime of 27.4 ns at 15 K and its lifetime decreases with increasing temperature. Such a short lifetime, which was unexpected for the originally forbidden band, is explained by a displacement of the vacancies around Sn²⁺ ions in the relaxed excited state which gives rise to a mixing of the forbidden ³P₂ states with the allowed ¹P₁ state.

The vibrational densities of states for an octagonal quasiperiodic lattice are calculated. Crossovers from phonon excitations to fracton-like excitations are found in the low-frequency parts of the spectra (this part is referred to as the continuous spectrum); meanwhile, localized vibrational modes are observed in the high-frequency parts (this part is referred to as the characteristic spectrum). The spectral features are governed by alpha and beta , the central and non-central force constants of the Born model. For the low-frequency continuous spectrum when alpha / beta equals unity (isotropic model) and when alpha / beta is not very large, a power law is satisfied in the fracton-like excitation frequency regime but, when alpha / beta approaches infinity, the spectrum deviates from the original power law. For the high-frequency characteristic spectrum, in an isotropic situation, i.e. in the case when alpha / beta =1, the localized mode is isolated and a gap separates it from other modes but, when there is no longer isotropy, the gap disappears.