Table of contents

Electron beam irradiation is used to make electronic walls intersecting the chains of the single crystal polymer PDADCH. Photocarriers are shown to travel along the polymer chains until stopped at the walls. The carrier motion is consistent with that of the acoustic solitary wave polaron. Photocarriers of opposite sign accumulating at the walls are shown to recombine. Intrinsic recombination centres are shown to be spaced 70 mu m apart on each chain.

Polyacetylene prepared by the Durham precursor route can be obtained in a highly oriented form if the transformation region is carried out with the polymer in a stress field. The authors report here a comparison of the stretched and unstretched films. For stretched films the resonant Raman phonon frequencies are lower, the DC conductivity is a factor of 1000 higher and the spins detected in ESR experiments have a room temperature linewidth of 4.25 G; approximately one half the linewidth measured for unstretched films. They interpret these results as indicating that the chain conjugation length in stretched films is significantly longer than in unstretched films. Disorder and chain entanglement in the precursor polymer, which can be reduced by the stretching process, is therefore the principal factor limiting the conjugate length of polyacetylene prepared by this route.

A model of the superionic conductor Ag₂S is proposed and studies by the molecular dynamics technique using effective pair potentials. The self-diffusion for silver, D_Ag(T), is in good agreement with experiment. The Ag density map and its temperature dependence is calculated. The authors also calculate the X-ray and neutron structure factors which explain the anomalous intensity at Q=(1.6, 1, 0) observed by Cava and McWhan (1980).

The recently predicted interlayer state in graphite is observed, using synchrotron radiation excited photoemission constant-final-state spectroscopy on a single crystal of natural graphite, to lie where calculated in a recent theoretical band structure already known to give good agreement with the experimental band structure determined by angle-resolved photoemission. This result highlights discrepancies between photoemission and inverse photoemission over the position of conduction band features in graphite.

The primary difficulty in developing a homogeneous deformation theory for piezoelectric crystals is that the strain energy density contains lattice sums whose Coulomb parts do not converge. It is shown that this difficulty can be resolved by separating out the macroscopic electric field from these lattice sums. This separation is achieved quite simply by using the concept of an effective electric field. In contrast, using the long-wave theory developed by Born and Huang requires a complicated procedure involving the use of a Ewald theta -transformation to achieve the separation. Using the microscopic expressions for polarisation and stress, the authors have obtained rather simple expressions for the second-order elastic constant, direct and converse piezoelectric constants and the dielectric susceptibility for the rigid-ion model, two-body central interaction and an arbitrary crystal structure. Finally, an expression for strain energy density is derived from microscopic theory in terms of the macroscopic parameters and is seen to agree with that obtained from the usual macroscopic theory.

The momentum eigenfunctions Phi (k+K) of copper Bloch functions were evaluated based on a scheme allowing both small and large values of mod k+K mod . As is to be expected, Phi (k+K) derived from MAPW functions have, in comparison with the results of other schemes, the most rapid decrease at large values of mod k+K mod . However, even this behaviour does not allow the matrix elements of the momentum operator to be determined with sufficient accuracy by folding the plane wave expansion. In the case of the matrix elements of the plane wave operator exp(iq.r) an absolute accuracy of 10^-3 may be achieved by summing over up to 30000 plane waves.

A general method is presented for approximating continuous densities of states, and wavefunctions over infinite regions, within the recursion method by means of a terminator which is an infinite set of approximate basis states and their Hamiltonian matrix elements. This is an analytical and computational solution to the problem of approximating the projected density of states for a Hamiltonian having an arbitrary number of bands with van Hove singularities. It also enables calculation of charge densities and other quantities related to the wavefunctions. The method is applied to many-band problems. Criteria are developed for the convergence of densities of states, and the asymptotic convergence of the terminator is related to singularities in the spectrum.

The concentration dependence of the critical temperature for the transition from the paramagnetic phase to the ferromagnetic phase (or antiferromagnetic phase) and the possibility of the spin-glass phase in the system of the randomly diluted quenched mixture of Ising ferromagnetic and antiferromagnetic bonds with first- and second-neighbour interactions is discussed using the bond coherent-potential approximation (BCPA). The phase diagram in the (concentration, interaction ratio I₂/I₁) parameter space is given and compared with the result of the modified Bethe-Peierls approximation theory of Katsura and Shimada (1980).

The author studies the effects of the spin-orbit interaction on the competing pairing states of a model quasi-one-dimensional conductor with hopping-type interchain coupling. Within a ladder approximation scheme use of the renormalisation-group method allows him to account explicitly for anisotropy effects involving the intrachain and interchain transfer integrals. Parametrising the spin-orbit interaction by a backward, g_1n<0, and forward, g_2n<0, scattering coupling constant, he concludes that its main effect is to stabilize the triplet-pairing states of spin density wave and triplet superconductivity over the corresponding singlet-pairing states of charge density wave and singlet superconductivity. In particular the boundary line between the density-wave-type states and the superconducting states is determined as: g₁+g_1n=2g₂, where g₁ and g₂ are the usual spin-independent backward- and forward-scattering coupling constants. The boundary line between the triplet- and singlet-pairing states is determined as: 2g₁=g_1n+g_2n- mod g_1n-g_2n mod . The direction of the triplet-pairing polarization axis m with respect to the spin-orbit-anisotropy axis n is shown to be dependent only upon the sign of the difference: g_1n-g_2n. The implications of these results of the (TMTSF)₂X and (TMTTF)₂X families of organic conductors is discussed. Finally, it is shown that the spin-dependent exchange interaction introduced recently by Abrikosov (1983) can be reduced to a simple definition of the spin-independent couplings g₁ and g₂. He concludes that a ferromagnetic type of exchange also favours the triplet-pairing states over the singlet-pairing states.

The electrical conduction in CdF₂ crystals doped with Sm or Gd is studied in both semi-insulating and semiconducting states. In the temperature range 60-300K the transport process is attributed to electrons moving in the conduction band, the injected carriers being successively trapped in the coulombic impurity centres and released by Poole-Frenkel emission. Activation energies of about 0.2 and 0.1 eV are found in semi-insulating and semiconducting crystals respectively. The fact that there are two levels is interpreted as arising from the contributions of the electron-phonon interactions being different.

The authors apply percolation theory to the high-temperature small-polaron hopping regime including correlation between bonds due to the energy in the common site. This drastically affects the temperature dependence of the conductivity. Thus, despite the fact that the average transition rates, and consequently the percolation condition, for the few-phonon low-T and the multiphonon high-T regimes are different, they obtain the same temperature dependence for the conductivity, namely Mott's law sigma approximately exp(-(T₀/T)¹4/).

The results of experimental and theoretical studies of the angular dependence of the photoelectric voltage (PEV) arising under dynamical X-ray diffraction conditions in semiconductor crystals containing a p-n junction are presented. The cases of diffraction both in the Bragg and the Laue geometries for samples with the p-n junction at the crystal entrance and crystal exit surfaces are considered. In the case of Bragg diffraction a strong dependence of the form of the PEV curve, while rotating the sample, on the thickness, d, of the upper n side for dL_ex(where L_ex is the extinction length) was found. The general form of the response function P(z) is obtained. This describes the contribution to the PEV from electrons created in the bulk of the crystal at depth z. The effect of both diffraction and electrophysical parameters, such as diffusion lengths of minority carriers in the p and n sides, on the form of the PEV angular dependence is analysed in detail theoretically. The experimental curves are well correlated with the ones calculated on the basis of the theory developed.

The measurements of the loss parameter in terms of the effective linewidth ( Delta H_eff) in two MgMn ferrite samples have been reported at the X band by using the magneto-microwave Kerr effect (MMKE). The effect of the sample thickness on Delta H_eff has been examined experimentally at low power levels. It has been found that Delta H_eff has a linear dependence on the thickness. The value of Delta H_eff extrapolated to zero thickness is the same as that expected from conventional cavity techniques.

The NMR lineshape and spin-lattice relaxation rate in incommensurate systems are evaluated for the case where, in addition to local contributions to the NMR frequency, non-local contributions also have to be taken into account. The case of a static wave as well as the case of a floating modulation wave are treated and the theoretical results are compared with experimental data.

A general theoretical treatment is developed for NMR spin-lattice (T₁ and T_{1 rho} ) and spin-spin (T₂) times for relaxation due to modulation for dipolar interactions by diffusion of atoms on inequivalent lattice sites. This extends previous calculations of T₂ to the other relaxation parameters and gives rate equations for a single nuclear species moving over several inequivalent lattice sites and interacting with other (static) nuclear species. The temperature dependences of the relaxation times are shown to have specified characteristics, such as minima in T₂ and multiple minima in T_{1 rho} . The formalism is then used to analyse data obtained for LaF₃ to evaluate details of fluorine-ion motion. Its applicability is successfully demonstrated.

The electronic and vibrational properties of group IVB transition-metal dichalcogenides are investigated by near-normal-incidence reflectivity measurements. The experimental spectra are discussed in terms of a dielectric function expressed as epsilon ( omega )=1+ chi _VE+ chi _PH+ chi _FC, where chi _VE represents the contribution from the valence electrons, chi _PH the phonon and chi _FC the free-carrier susceptibility. The free carriers are interpreted using a single-carrier Drude term with the spectroscopic parameters omega _p (plasma frequency) and omega _tau (collision frequency). The basic properties of the lattice vibrations are discussed. From the measured phonon parameters, values for the Born effective charges and for the Szigeti charges are deduced. The results indicate that the sulphides are more ionic than the selenides and that the ionicity increases in the sequence Ti to Zr to Hf.

The infrared transmission spectra of small GaAs crystals (0.5 mu m in average size) are reported. The samples are produced by mechanical grinding down from the bulk crystals with known carrier concentrations from 5.3*10^-17 to 2.4*10¹⁸ cm^-3. The spectrum is constructed from a surface phonon mode and coupled modes between surface phonons and plasmons. The two coupled modes are clearly separated and their frequencies agree very well with the theoretical dispersion curves. On the basis of the theory of an average dielectric function the observed spectra are reproduced from the calculations, in which the particle is assumed to be a sphere composed of a shell with no carriers and a core containing high carrier concentration.

Natural brown diamonds that emit yellow luminescence under 3.40 eV (365 nm) excitation are known to exhibit also a red luminescence band when excited with 2.69 eV (460 nm) radiation. Four zero-phonon red-luminescent lines have been studied under uniaxial stress. One of them, at 2.145 eV, believed to be the origin of the major vibronic progression known as the red band, is found to be a transition between A states at a monoclinic I centre. The other lines at 2.133, 2.156 and 2.166 eV are shown to be transitions between A and B states at monoclinic II centres, with the C₂ point group. It is the first time that centres with this point group have been reported in diamond. The authors suggest that a relationship between the centres giving rise to the yellow and to the red luminescent bands, by comparison of their phonon sidebands, symmetries, responses to stress and zero-phonon energies.

Crystals of LiF and NaF doped with uranium and grown in oxygen contain centres with strong fluorescent lines at 527.8 nm and 563.6 nm respectively. The spectroscopic properties of these tetragonal centres have been studied at liquid helium temperature using polarised selective excitation and magnetic circular dichroism techniques. This has enabled the degeneracy of levels to be determined and the phonon coupling from localised vibrations to be described. An E electronic level has been found in each case fairly close in energy to the fluorescent level. The g-values for these E levels and several others in the higher-energy region have been measured.

Ground-state studies are carried out for a two-dimensional helium liquid using a Jastrow-type wavefunction. The two-body distribution and liquid structure functions are determined by solving an Euler-Lagrange equation within the HNC approximation and have the expected long-range and short-range behaviours respectively. Calculations for the energy, pressure, compressibility and the sound velocity have been performed for various densities. The condensate fraction is also evaluated using the HNC treatment of the one-particle distribution function and the relative momentum distribution formalism. The results of these two indicate the existence of more complicated condensate mechanisms.