Table of contents

The reference interaction site method (RISM) may be applied to polyatomic fluids sorbed by quenched materials. Formal aspects of such applications are discussed. Connections with and generalizations of the theory of Madden and Glandt are made.

It has often been said that 'the structure factors of all liquids (and glasses) look the same'. Those working in the field know that there is a certain truth to this, as many of the basic features are determined by hard sphere packing. However they also know that there are differences, sometimes subtle, which depend on details of the bonding and local structural arrangement. The initial requirement for understanding these details is an accurate experimental measurement. This is not a trivial task, but even when it has been done the interpretation of the structure factor(s) or radial distribution function(s) often only use gross features such as peak positions and coordination numbers. This approach makes poor use of the data obtained and it can be misleading. RMC is a method of structural modelling which is based on the experimental data and so makes full use of all the available information. The ability to create a three-dimensional model leads to a greater understanding of the relationship between structure and structure factor, independent of their relationship to the interatomic potential.

The state of the art in the calculation of interatomic forces in s, p and d-bonded metals and alloys, and their application in the atomistic simulation of the structural and electronic properties of liquid and amorphous phases are reviewed.

The solid state amorphizing transformations during mechanical alloying and mechanical grinding of a Ni-V binary system were observed by neutron total scattering using a spallation pulsed neutron facility. The topological mechanisms of crystal-to-amorphous solid structure transition are interpreted in terms of the transformation of octahedral structure units into tetrahedral structure units and the modification of ways of connecting tetrahedral structure units from vertex-sharing into face- and/or edge-sharing ones.

The author will summarize studies on the dynamics of monatomic liquids, mainly liquid metals using inelastic neutron scattering techniques. Many microscopic properties of disordered systems like liquids which are not accessible by other techniques can be investigated by properly designed neutron scattering experiments. Details of the physics of diffusion processes, that is of single-particle motion, can be revealed by incoherent quasi-elastic scattering. coherent scattering can explore relative motions or collective modes and their damping mechanism. both aspects will be illustrated with a few examples of recent experiments.

Numerical solutions of the thermodynamically self-consistent RY integral equation for the pair structure of a binary fluid hard sphere mixture, of size ratio sigma ₁/ sigma ₂=0.1, exhibit divergent concentration fluctuations when the partial packing fractions of the two species comparable, i.e. at low concentration of large spheres. This spinodal instability is studied at constant pressure, and it is shown that phase separation sets in for a reduced pressure P*=P alpha ₁³/k_BT in the range 10^-2<P*<10^-1.

Brillouin scattering experiments with neutrons are discussed and recent instrumental developments are presented. The most recent results of such experiments performed on compressed Ar and N₂ gases at temperatures well above the critical temperature T_c are summarized. The results are discussed within the framework of hydrodynamic theory and it is shown that thermodynamic conditions can be found under which the newly developed experimental techniques allow to intrude into the region of pure hydrodynamics, up to those now reserved for light scattering techniques. For atomic and molecular fluids a common limit between the pure hydrodynamic and the generalized hydrodynamic regimes is found between 0.25 and 0.3 l_EQ, where l_E is the mean free path between atomic collisions in the fluid and h(cross)Q is the momentum transfer. This experimentally determined limit is about a factor of two to three higher than the values of l_EQ predicted by generalized Enskog theories for hard-sphere fluids.

A review of the application of neutron scattering to the study of ionic solvation is presented. Particular emphasis is placed on the use of isotropic substitution and incoherent quasi-elastic scattering.

The authors have made an extensive investigation of the phase diagrams and the associated micro-structure of a pseudo-ternary system AOT/water (0.6 wt.% NaCl)/decane. This micro-emulsion system shows, for a surfactant concentration in excess of 6 wt.%, a characteristic phase progression from a water-in-oil (W/O) micro-emulsion in coexistence with excess water at low temperatures to an oil-in-water (O/W) micro-emulsion in coexistence with excess oil at high temperatures through a one-phase micro-emulsion in the intermediate temperatures. Thus, one expects a structural inversion to occur somewhere in the one-phase channel. They have performed extensive small angle neutron scattering (SANS) measurements to study the pathway of such a structural inversion in this micro-emulsion system when there are comparable amounts of oil and water in the system, and when the volume fraction of water is much higher than that of oil. In the former case, the authors observed the inversion of the W/O to the O/W micro-emulsions through an intermediate bicontinuous one-phase micro-emulsion. In the latter case, they show that the inversion takes place through a pathway of L₃-L_alpha -L₁ as temperature increases. Particular emphasis is put on analyses of the interfacial structure in different micro-emulsion phases.

Theoretical and experimental results for the distribution of molecular orientations at the liquid/vapour interface of water are compared. The present calculations are based on an extended mean-field density functional theory applied to an intermolecular potential model containing dipolar and quadrupolar interactions, with parameters chosen to agree with the TIP4P model of water. The results for the orientational order parameters at the interface are used to calculate several observable quantities as functions of temperature, namely: (i) the surface electrostatic potential, (ii) the coefficient of ellipticity; (iii) the nonlinear susceptibility measured by second-harmonic generation. Comparisons are made with both the experimental and computer simulation data for water. The limitations of current methods for revealing the preferred molecular orientations at an interface are discussed.

The perovskite structure of the relaxor PbMg_1/3Nb_2/3O₃ (PMN) is studied using X-ray and neutron powder diffraction data. The static diffuse scattering (SDS) observed in the diffraction patterns at low temperatures is interpreted using a two-phase Rietveld analysis. The structural model is based on a long-range structure with an average cubic symmetry, and a short-range order due to atomic shifts involved by the formation of polar regions. The two-phase model provides a good improvement in reliability factors. The correlation length and the coexistence of different phases at low temperatures are discussed.

An X-ray single-crystal determination of the CuO structure has been made at 196 K, i.e. below the Neel temperature 230 K, and, as a check, the crystal structure at room temperature was also determined. The correct space group for the structure at both temperature was found to be Cc. Earlier results of magnetic and neutron diffraction measurements can be explained as antiferromagnetic coupling between copper atoms via oxygen (superexchange) in chains running in the (1 0-1) direction. The structural results show changes with temperature in Cu-O distance in these chains: in each -Cu-O-Cu-group the longer is increased and the shorter decreased when passing from 196 K to room temperature. This implies a weaker antiferromagnetic coupling at room temperature. The refinement of CuO-structure at room temperature reported earlier by Asbrink and Norrby (1970) showed the symmetry to be C2/c. An attempt to refine CuO in the space group Cc with the old data was not successful. The different results obtained with different crystals are tentatively explained from published observations regarding valence fluctuations in CuO and non-stoichiometry caused by cation vacancies.

Neutron inelastic scattering determination of the phonon dispersion curves in high symmetry directions for iron-aluminium alloys are presented. The alloys have compositions near 25 atomic percent aluminium, and were prepared in three states of ordered BCC, B2 and D0₃ structures. The data were analysed by fitting sixth-nearest neighbour Born-von Karman models. The phonon frequencies follow a consistent trend in the ordering sequence BCC to B2 to D0₃. A similar trend is apparent in the values of the atomic force constants for the nearest neighbours but is less apparent for the force constants for more distant neighbours. Of the low frequency modes, the soft (qq0) transverse mode is the most sensitive to changes in the atomic force constants. High frequency modes exhibit large changes in frequency as a result of ordering because of the large difference in the masses of iron and aluminium atoms, as well as the changes in force constants.

The electronic band structure of rutile TiO₂ has been calculated using the linear muffin-tin orbital (LMTO) method. The X-ray absorption spectrum has also been obtained from the calculated band structure and self-consistent potential. The dipole as well as the quadrupole contribution to the X-ray absorption spectrum has been calculated. The calculated band structure shows that there is a significant 2p (O)-3d (Ti) mixing in the valence band which, however, becomes weaker in the Ti d-dominated conduction band. The valence band is well separated from the conduction band by a direct energy gap which is almost degenerate with the indirect energy gap. A comparison with experiments suggests that the majority of the observed pre-edge and edge features in the Ti K-edge X-ray absorption spectrum are due to 1s to 4p dipole transitions, a result of the small Ti-Ti (3d-4p) hybridization in the conduction band region. The quadrupole transition contribution is found to be insignificant.

The magnetocrystalline anisotropy energies of one Fe monolayer in Cu (Ag,Au) have been calculated from first principles. The spin-polarized relativistic linear muffin-tin orbitals (SPR-LMTO) method is used. The electronic contribution to the magnetic anisotropy energy is found to dominate in these systems and it favours an easy axis of magnetization perpendicular to the Fe monolayer plane. In contrast, the calculated magnetic-dipolar interaction contribution favours an in-plane magnetization orientation but its magnitude is, respectively, about three, six and five times smaller than the electronic contribution.

Optical absorption studies in the infra-red region around the fundamental energy gap have been carried out on both n- and p-Si samples in the range of concentrations from 3*10¹⁴ cm^-3 to 7.3*10¹⁸ cm^-3 for p-Si and from 6.5*10¹⁶ cm^-3 to 9.2*10¹⁸ cm^-3 for n-Si at room temperature (294 K). The transmission ratios were first measured for each sample in the range of wavelengths from about 0.9 mu m to 2.5 mu m. These ratios were then applied to calculate the reflectivity and the absorption coefficients for the samples. After correcting for free-carrier absorption, the band-gap shrinkage was then determined from the absorption coefficients for each sample. These results were then compared with those determined from other optical absorption studies, with those found by electrical device measurements and with those predicted from theories. It was found that even well below 3*10¹⁸ cm^-3, a reduction in band-gap energy Delta E_g could be determined for both n- and p-Si, in contrast to the earlier optical absorption studies which reported that none could be detected. Between 10¹⁸ cm^-3 and 10¹⁹ cm^-3 the authors experimental values are consistent with those determined by other optical absorption studies. Their results also show that even in the moderately doped regime, their values of mod Delta E_g mod are less than those found by electrical device measurements and those predicted by various recent theories. This is consistent with the findings of the other optical studies in the heavily doped regime. The authors also found that the gap shrinkage mod Delta E_g mod depends on the dopant concentration as N¹3/. This is in agreement with the experimental results of Wagner and Vol'fson and Subashiev as well as with the theoretical results of Mahan and Berggren and Sernelius.

Double and multiple irreducible products of irreducible tensor operators O^(ki)(J) obeying special commutation relations arising from the properties of angular momentum operators are considered. It is shown that the double product can be expressed in terms of another set O^(k) associated with a coupling coefficient epsilon _k^(k1,k2) defined by (O^(k1)*O^(k2))^(k)= epsilon _k^(k1,k2)O^(k). An analytical expression has been derived for the coupling coefficient epsilon _k^(k1,K2) and its values are tabulated for k₁ and k₂ up to 3. It is proved that (O^(k1)*O^(k2))^(k)=(O^(k2)*O^(k1))^(k). Multiple irreducible products are also discussed. The coupling coefficients for triple products are tabulated for Sigma _ik_i up to 6. Some important special cases of quadruple products are dealt with explicitly. Applications of the present results to high-order effects in spectroscopy are investigated.

The energy spectrum the density of states, the electron probability distributions and the magnetization of a Ga_1-xAl_xAs-GaAs-Ga_1-xAl_xAs quantum well in strong tilted magnetic fields B=(B_x,0,B_z) are calculated. It is shown that the spectrum and the probability densities of the states bound in the quantum well would change little if the coupling Hamiltonian approximately=B_xB_z were to be neglected. Localized states appear above the barrier edge. The magnetization parallel to the layer interfaces exhibits oscillations and sharp jumps as a function of the chemical potential. It remains to be seen whether these jumps may serve as the basis for a dissipation free switching device.

Results are presented of extensive numerical simulations of electron wave packets transmitted by horns. A detailed quantitative analysis is given of the collimation of the electron wave by horn-like devices. It is demonstrated that the electron wave collimation effect cannot be described in terms of adiabatic wave expansion and semi-classical considerations. It is argued that, in the context of atomic-size electron beam sources, the collimation effect due to a possible horn-like structure in the metal-vacuum potential is much less than the effect stemming from the tunnel barrier itself.

For an Ising ferromagnetic with the bilinear exchange interaction J and the three-site four-spin interaction J', the ground-state spin structures are determined from the J'-dependence of the energy of a single spin in various spin structures. It is pointed out that the phase transitions occur at J'=-J/4 for the linear-chain Ising ferromagnet of spin S=1 and at J'=-J/9 and J'=-3J/22 for the Ising ferromagnet of spin S=1 on a two-dimensional square lattice. For the Ising ferromagnet of spin S=3/2 on a two-dimensional square lattice, it turns out that the phase transitions also occur at J'=-2J/45, J'=-4J/75 and J'=-5J/78. Furthermore, it is made clear that the phase transitions at J'=-J/9 and J'=-2J/45 agree well with those obtained from the anomalous behaviour of the magnetization for ising ferromagnets of S=1 and S=3/2, respectively.

The authors have applied resonant Raman spectroscopy to two p-type modulation-doped GaAs-GaAlAs multiple-quantum wells (MQW) with well widths d_W=10 and 30 nm. In both samples one of the hole intersubband transitions observed is coupling to the quantum-well LO (longitudinal optical) phonon. The influence of an external magnetic field on this coupling phenomenon was investigated. A shift of the intersubband transition energy with excitation energy due to well width fluctuations results in an anticrossing of the two branches of the coupled modes. From the strength of the coupling as well as from the small depolarization shift of the intersubband transitions, the conclusion must be drawn that depolarization field effects are of low importance in p-doped MQW structures. The coexistence of a non-interacting quantum-well LO phonon is discussed in detail.

A spectroscopic study of chromium-doped RbCdF₃ crystals has been performed. Cubic, tetragonal and trigonal Cr³⁺ centres have been detected by EPR measurements. The spin Hamiltonian parameters of trigonal and one type of tetragonal Cr³⁺ ions, not previously reported, have been found to be: g/sub ///=1.969, g_{perpendicular to} =1.971 and mod D mod =0.1576 cm^-1 for the trigonal centre and g/sub ///=1.970, g_{perpendicular to} =1.972 and mod D mod =0.0722 cm^-1 for the tetragonal one. Crystal field and Racah parameters of cubic Cr³⁺ derived from the optical absorption spectrum are: D_q=1420 cm^-1, B=800 cm^-1 and C=3290 cm^-1. Photoluminescence has been measured at different temperatures and the phonon structure observed at 15 K has been associated with different Cr³⁺ centres. The lifetimes of the emitting levels of cubic and trigonal Cr³⁺ ions measured at 15 K have been found to be 470 mu s and 220 mu s respectively. At 300 K the lifetime of cubic centres decreases to about 180 mu s. No direct influence of the phase transition on the optical properties of Cr³⁺ ions, when the authors go through the transition temperature, has been detected.

It was found that the energy spectrum of secondary electrons emitted under ion bombardment of the cold (below 100 K) metal targets contains an additional maximum, the position of which on the energy scale depends on the ion incidence angle and the angle of electron detection. This maximum is also observed at room temperature but only at grazing exist angles. The additional maximum is supposed to be a result of the creation of electrons on the surface. Such electrons can be distinguished from the bulk electron emission at the special experimental arrangement (low target temperature and small exist angle of electrons).

The authors report calculations of the liquid pair structure of YCl₃ as a prototype for strongly ionic melts of lanthanide metal trichlorides. The calculations adopt an ionic model involving a pair potential of the Busing type, with parameters for the Y³⁺ ion adjusted to the Y-Cl bond length in the YCl₃ crystal and to the breathing mode frequency of the (YCl₆)^3- octahedral unit. The model is solved in the hypernetted chain approximation and the results are tested against neutron diffraction data on the Faber-Ziman structure factor and the total pair correlation function of molten YCl₃. The calculated partial structure factors and pair correlation functions provide insight into the origin and nature of the short-range and intermediate-range order in the melt.

The vibrons nu ₁ and nu ₂ of N₂ have been measured around the epsilon - delta phase transition by Raman spectroscopy. Thus, for the first time the epsilon - delta transition could be detected in this way. The Raman shift of the nu ₁-peak has been observed to be larger in the epsilon -phase than in the delta -phase and the Raman shift of the nu ₂-peak smaller. More importantly, this feature has also been observed in the mixture N₂-He with 9.2 mol.% He around the phase transition at p=60 kbar and T=245 K, which is new evidence for a considerable temperature and pressure shift of the epsilon - delta phase transition line in the mixture and also for the solubility of fluid He in solid epsilon -N₂ at high pressures.

The coupling between the longitudinal and transverse fluctuations in a spin s ferromagnet becomes significant above a crossover temperature T approximately=T_c/s. This coupling produces a splitting of the spin-wave resonance near the energy zJs, where z is the coordination number and J the coupling constant. The authors calculate the transverse mode frequencies and propose an experiment to detect the longitudinal fluctuations.