Table of contents

The a and c lattice parameters of tetragonal PbTiO₃ have been measured by X-ray diffraction techniques, using a diamond-anvil cell, up to a pressure of 6.35(5) GPa at room temperature. The a lattice parameter is found to show a small expansion with pressure up to approximately 1.7 GPa, beyond which da/dP then becomes negative (as is more usual). The compressibility is highly anisotropic: between atmospheric pressure and 6.35 GPa the c parameter is compressed more than ten times as much as the a parameter. The spontaneous strain, z¹2/=(c/a-1)¹2/, exhibits significantly anomalous behaviour: the magnitude of d(z¹2/)/dP decreases with pressure, from -3.4(2)*10^-2 GPa^-1 at low pressures to about -1.3*10^-2 GPa^-1 at 6.0 GPa. There is some evidence that this can be interpreted as a discontinuity in the behavior of PbTiO₃ at approximately 1.7 GPa, close to the pressure where da/dP passes through zero.

The author shows how the kinetic energy factor for isotope diffusion may be calculated within the quasi-harmonic approximation using methods devised to calculate the vibrational contribution to defect entropies. The results are compared with experiment in the cases of CoO and NiO and good agreement found.

The f-d hybridisation and the appearance of two 4f structures close to the Fermi level in the golden phase of SmS are analysed. These two 4f structures are split by a pseudo-gap which is justified by the coupling of the XYZ orbital with the d-band states. The coupling ratio depends on the hydrostatic pressure effects and the value of the Coulomb correlation energy.

A theoretical study of the inter- and intra-chain photo-excitation of electron-hole pairs by light polarised both parallel (E/sub ///) and perpendicular (E_{perpendicular to} ) to the chain direction is presented and compared with experimental results for the anisotropic absorption coefficients, alpha /sub /// and alpha /sub ///. It is found that to account for the form of alpha _{perpendicular to} adjacent chains must be aligned out of phase, the resulting anisotropy being alpha /sub //// alpha _{perpendicular to} approximately=200, in reasonable agreement with experiment. It is also shown that alpha _{perpendicular to} is dominated by inter-chain transitions near the band gap and by intra-chain transitions further into the band, supporting the model for the anisotropic photo-excitation where electron-hole pairs initially created on separate chains are viewed as one of the most important sources of long-lived charge carriers.

For the FCC Co_pMg_1-pO system (0.22<or=p<or=1), Neel temperatures T_N(p) are determined using temperature-dependent (4.2-300K), low-field (50 Oe) DC susceptibility ( chi ) measurements. The variations of t=T_N(p)/T_N(1) and Dt/dp with p reveal a crossover near p=0.45; from p=1 to 0.45, dt/dp increases from 1 to 2.2, whereas from p=0.45 to 0.22, dt/dp decreases from 2.2 to 0.3. For p<0.45, chi is irreversible below T_N between the zero-field-cooled and field-cooled modes. Calculations of t versus p for FCC antiferromagnets, with NN and NNN couplings included, are needed to compare with the data.

The antiferromagnetic ground state of beta -(BEDTTTF)₂ICl₂ and AuCl₂ is probed by antiferromagnetic resonance. The parameters deduced (position of the magnetic axes and magnitude of the anisotropy) are discussed using a strongly localised picture of the spin ordering as suggested by the strongly dimerised structure of the ICl₂ salt and the low electrical conductivity of these compounds. An excellent agreement is found between experimental and calculated results, assuming the dipole-dipole interactions are at the origin of the observed anisotropy.

The temperature variation of the absorption spectrum in the energy range from 3.40 to 5.0 eV is measured at temperatures from 15 to 291 degrees C for the rubidium nitrate crystal. Anomalous behaviour in the absorption spectrum has been observed near the phase transition points. It is found from profile analyses of the spectra at various temperatures that the interaction between the electronic process and the lattice system in the crystal changes at the transition points.

A consistent description of the rotational excitations in (NH₄)₂PdCl₄ and (NH₄)₂PtCl₄ is given in the model of a one-dimensional rotation in a nearly fourfold rotational potential, which is supported by the crystal structure. In such a potential, large tunnel splittings are combined with high-lying librations and moderate activation energies. The estimated model potentials are V( phi )=70 cos 4 phi +0.0004 cos 2 phi (meV) for (NH₄)₂PdCl₄ and V( phi )=66 cos 4 phi +0.0008 cos 2 phi (meV) for (NH₄)₂PtCl₄.

A new method is described for obtaining information on the phonon frequency spectrum g( omega ) from the phonon specific heat C^ph(T). A rapidly convergent function G(T) is first generated from C^ph(T), (G(T) approximately 1/T¹¹ at high temperature T), and a spectrum g( omega ) is obtained from a least-squares fit to G(T) of a set of functions similarly derived from the Einstein specific heat function. The resolution (half width at half height) of 0.18 of the frequency is comparable to that of existing direct methods for the determination of g( omega ) such as incoherent inelastic neutron scattering. The geometric mean frequency and integer and non-integer negative and positive moments of g( omega ) may be accurately determined from suitable temperature integrals of G(T). The method is simple to apply and can be used if specific heat measurements of reasonable accuracy are available up to theta _D/2.

The theory of the X-ray, Raman and neutron scattering cross sections is evaluated for incommensurately modulated phases. The theory is worked out in detail in the small harmonic distortion limit, but includes the coupling of the amplitude, phase and acoustic fluctuations. The effect of the coupling is very dependent on the wave-vector and the form of the Raman tensor probed. The effect of the corrections to mean-field theory are discussed and in particular the phason renormalisation of the amplitude modes is shown to contribute to the neutron and X-ray cross sections but not to the Raman cross section.

Calculations of local and gap mode frequencies of monovalent impurities, including H^-, at substitutional and interstitial sites in alkali halide host crystals are presented. The calculations are based on previously published inter-ionic potentials for alkali halides and hydrides and involved the determination of normal modes for a spherical 'crystallite' containing the impurities and up to around 200 polarised host atoms at relaxed positions. The crystallite is embedded in a matched dielectric continuum and the point group symmetries of the defects, either single impurities or impurity pairs, are fully exploited. The level of reliability of the method and the inter-ionic potentials input data is demonstrated by comparing the results with infrared measurements for single impurity centres. New interpretations and symmetry assignments for some existing experimental data are proposed and several unmeasured (and principally infrared inactive) gap modes are predicted.

The author constructs and examines the properties of a simulation model for Pd:H. The metal-metal potential is based on the polynomial model of van Heugten (1979), and the metal-hydrogen potential is represented by a parameterised exponential form. The molecular dynamics calculations show that the temperature-dependent tracer diffusion coefficient of hydrogen is in semi-quantitative agreement with experiment, and that the residence site of the hydrogens is the octahedral position, as required by diffraction measurements. Diffusion occurs by jumps between these sites, but the jumps are not uncorrelated, so that the Chudley-Elliott model does not provide a satisfactory description. The relation between the model and the real system is discussed with particular reference to the correlation of jumps, and a physical mechanism which might reduce this correlation is suggested.

The authors consider the kinetics of the diffusion process of an assembly of N₀ initially aggregated particles. The particles are non-interacting except that double occupancy is forbidden. A computer simulation provides the number of distinct visited sites after a given number of steps performed by the particles on a 2D square grid. The particles are supposed to diffuse according to a random-walk process on the lattice where all jumps to the nearest-neighbour sites are allowed. As the system exhibits a centrifugal diffusion the simulation shows that the number of distinct visited sites obeys three distinct successive processes. The various singular points of these three processes have been determined analytically as functions of N₀ and of the exponent d, where the N₀ dependence of d is derived from the computer simulation. These results, which are relevant to several significant physical situations, shed a new light on the processes where the fractal nature of the kinetics depends on the initial conditions of the problem rather than on the mere geometry of the diffusion paths.

The authors show how the equation of motion method for solving Schrodinger's equation numerically may be employed in k-space and present calculations of the density of states and spectral function for fully bonded models of amorphous Si containing 216 atoms. The magnitude of the 'gap' in the density of states depends sensitively on the strength of the pseudopotential and the structure. The method provides a simple method for investigation of the effect of defects in amorphous semiconductors and may also be used to look at individual energy levels. This method has considerable potential for the investigation of electron conduction in the vicinity of the gap in the density of the states and for self-consistent calculations.

Germanium impurity in alpha -quartz as a well known deep electron trap and hence makes possible the study of hole trapping processes by EPR techniques. The authors have carried out an EPR study of SiO₂:Ge X-irradiated in a microwave cavity at 4K and show that Ge is an amphoteric impurity at low temperatures. In addition to the trapped electron centres they find three EPR spectra, designated H(I), H(II) and H(III), associated with hole trapping by Ge. The spectra H(I) and H(II) are the ground states and an excited state of a hole in a nonbonding 2p orbital in an oxygen ion next to a single substantial germanium ion. This hole centre decays at 35K. The weak spectrum H(III) is due to a hole in a non-bonding 2p orbital in an oxygen ion between two substitution Ge ions and it decays at 58K. The results are compared with the hole-trapping properties of aluminium impurity ions and the question of intrinsic self-trapping of holes in crystalline quartz is discussed.

The temperature dependence of the magnitude of the optical birefringence of RbHSeO₄ has been studied between room temperature and 170 degrees C by the rotating-analyser technique. The phase transition at T_c-95 degrees C is marked by singularities in the Delta n_i(T) curves obtained for the principal directions. A transition, originally reported by Suzuki et al. (1979) to occur at 173 degrees C, has been found to be the result of a decomposition process.

Random-field models are considered with long-range exchange, J(r) approximately 1/r^{d+ sigma} , and/or long-range correlated random fields, (h(r)h(r')) approximately 1/ mod r-r' mod ^{d- rho} . A general scaling analysis is presented, in which the existence of three independent critical exponents is emphasized. Four different universality classes are identified, according to whether the fixed point controlling the critical behaviour is of short-range or long-range character for the exchange and/or the random-field correlations. The lines in the rho - sigma plane at which the long-range couplings first become relevant are expressed in terms of the exponents eta and eta of the corresponding short-range problem: long-range exchange is relevant for sigma <2- eta , long-range field correlations for rho >2 eta - eta . The scaling results are confirmed and amplified by explicit calculations of three independent exponents of the n-component spin model, correct to order 1/n. The results reveal errors in published expansions around the upper critical dimension.

Proton and deuteron magnetic resonance experiments have been carried out on Sr(BrO₃)₂.H₂O at RT. From PMR data the hydrogen positions have been calculated by means of a simple method which makes use of the crystal symmetry. The EFG parameters obtained from the DMR data indicate that the D₂O molecules are executing rapid 180 degrees flips at RT giving rise to an averaged EFG tensor. From the results of a point charge calculation of the EFGs, the validity of hydrogen coordinates derived from PMR is confirmed. Again, making use of the symmetry considerations, component tensors are extracted from the room-temperature-averaged tensors. The hydrogen bonding in isostructural Ba(ClO₃)₂.H₂O, Ba(BrO₃)₂.H₂O and Sr(BrO₃)₂.H₂O is discussed in relation to the presence of superfine interactions observed in EPR studies of gamma -irradiated Ba(ClO₃)₂.H₂O and Ba(BrO₃)₂.H₂O.

ENDOR measurements of ³¹P and ⁵¹V in LnPO₄ and LnVO₄ have shown three features. First they have shown that there are large non-dipolar contributions to the transferred hyperfine interaction between a Gd³⁺ impurity ion and the nearest neighbour positive ion, as was also found in alkaline earth fluorides by Baker and Bluck (1985). Secondly, they have shown that changes in the electric quadrupole interaction of ⁵¹V may be used to deduce local distortions produced by a Gd³⁺ impurity. Thirdly, they have demonstrated a large deviation in the angular dependence of the ENDOR spectra from the commonly used first-order expressions, due to effects of fine structure.

Self-healing dielectric breakdown in thin oxide films has been shown to obey Weibull statistics after allowance has been made for both left and right censoring that is intrinsic to the measuring technique. Two overlapping distributions have been found under constant DC voltage stress testing but under ramp stress conditions these were not always observed. Nevertheless it is shown that the two types of test gave consistency in their analyses. Two particular features of the results reported are that the number of potential events, in some cases, was an exponential function of the applied field and that under ramp test conditions the magnitude of the applied potential at the breakdown estimator time appears to be constant and independent of the sample thickness. The implications of these observations are discussed.

The impact excitation and Auger quenching rates are calculated for a luminescent centre in a solid. The approximations used are a perturbation approach (equivalent to the Born approximation) and a long-range dipole interaction (equivalent to the Bethe approximation), together with the neglect of exchange. It is shown that the calculation is inapplicable to ZnS:Mn in which the exchange process is enhanced by a matching of energy spacings within manganese with energy spacings within the conduction band. This provides an explanation of the pre-eminence of ZnS:Mn as a material for high-field electroluminescence.