Table of contents

The structure factors S( kappa ) under a pressure P are calculated for five alkali metals near the melting curve T=T_m(P) within the interval of compressions u identical to 1- Omega (P)/ Omega (0) from 0 to 0.3. The Percus-Yevick approximation and the pseudopotential model are used which give a rather accurate description of S( kappa , T) in the metals at P=0. The calculated structure factors vary with u much more sharply than those calculated using the hard-sphere approximation and the variational method for free energy. For example, the height S_p of the S( kappa ) main peak increases by 20-25% under a compression u=0.1 and by more than 1.5 times at u=0.2. The conjecture of Ferraz and March (1980) on the existence of the 'crystallisation criterion' of the type S_p(T_m)=constant has not been confirmed by these calculations. Comparison with experiment is hindered by the spread in data available on S( kappa , u). However, the results of the virial equation of state appear to imply that the calculated S( kappa , u) are precise enough for u<or approximately=0.1 and become less accurate under higher compressions.

A theoretical investigation has been carried out to determine the mechanical stability of a perfect crystal in the presence of applied forces by applying the Born stability criteria. A face-centred-cubic (FCC) perfect crystal of aluminium is subjected to tensile and compressive stresses parallel to the (100) crystallographic axis. The total energy of the solid consists of the kinetic, exchange and correlation energies of the conduction electron gas, along with the ion-ion and electron-ion interactions which are estimated following a calculation based on perturbation theory. Specifically, the latter is obtained under the pseudopotential formalism in the Ashcroft local pseudopotential model (1966). The ranges of stability in tension and compression are located and a theoretical tensile strength of 0.4560 GPa (4.56*10⁸ N m^-2) is calculated which compares fairly well with the experimentally measured tensile strength of 0.1655 GPa obtained for high-purity aluminium in the form of a fine wire (which although not in a perfect crystalline form is chosen for comparison in the absence of better experimental data).

A Monte Carlo technique is used to calculate the spectral distribution of multiply scattered electrons in a typical inelastic electron scattering experiment. The various contributions to multiple scattering are classified, including double and triple scattering, and good agreement is found with published data for two different thicknesses of aluminium samples.

The authors extend earlier soft X-ray emission measurements on the AlCu system for metastable states which depend on the heat treatment at a comparatively low Cu concentration, namely 4 wt%. The soft X-ray emission spectra of the alloy, after solution heat treatment and quenching, show a shift of the Fermi edge to higher energy (by +0.3 eV) compared with the Fermi edge of pure Al. After annealing at 200 degrees C for two hours the emission spectrum of the alloy has an increased second maximum at 71.5 eV; thus the density of electron states also has a higher value corresponding to the K and/or W₃ points in kappa space, and the Fermi energy has a minimum value which is approximately 0.2 eV lower than the Fermi energy of pure Al (72.8 eV). After annealing at 350 degrees C for half an hour the emission spectra and the Fermi edge are similar to those of pure Al.

High-precision measurements of the temperature dependence of the (111) extremal cross-section area of the neck of the Fermi surface of gold have been carried out by means of the de Haas-van Alphen effect in the temperature and field ranges 4<or approximately=T<or approximately=17K and 48<or approximately=H<or approximately=71.5 kG, respectively. The area is found to decrease approximately as the fourth power of the temperature with a coefficient which is over twice that expected from the effect of lattice expansion alone. The difference is attributed to the modification of the effective potential seen by the carriers due to thermally excited lattice vibrations. The observed area shifts corrected for constant volume provide an estimate of the electron-phonon interaction parameter for electronic states in the vicinity of the L(2') point in the Brillouin zone at the Fermi level.

The pressure dependence of the Fermi surfaces in vanadium is calculated within the framework of the Slater-Koster interpolation of the augmented-plane-wave local-density bands. The results are compared with the experimental and theoretical data for niobium; a significantly small effect of pressure is predicted for the areas in the alpha orbit on the jungle-gym arms and in the gamma ₁ orbits on the octahedral surface.

Polycrystalline nickel samples containing up to 700 at PPM hydrogen have been prepared by quenching from equilibrium in the temperature range between 150 degrees C and 600 degrees C. The analysis of the quenched-in resistivity and solubility data gives an enthalpy of solution of Delta H=(0.13+or-0.01) eV and a specific resistivity of Delta rho =(0.67+or-0.02) mu Omega cm (at% H)^-1. Isochronal annealing of the resistivity of pure Ni and NiSi alloys up to 460K shows two recovery stages. The lower temperature stage is attributed to the trapping of the hydrogen atoms at the impurities. The higher recovery stage is due to hydrogen escaping from the sample. Transmission electron microscopy of pure Ni and NiSi alloys gives no evidence for hydrogen precipitation into bubbles upon annealing.

The resistivity and thermopower of amorphous Mg-Zn alloys have been computed as a function of temperature and composition. The diffraction model incorporating the dynamical partial structure factors is applied. The effect of the electron mean free path is investigated. The authors find that the resistivity is well described by the model, and that the inclusion of the mean free path does not change the results considerably. In the case of thermopower the diffraction model turns out to be inadequate: it gives a composition dependence which is against the experimental evidence. This suggests that there exists another scattering mechanism, which is not accounted for by the diffraction model. The choice of the various parameters of the model is discussed.

A reanalysis of available data on the influence of surface scattering upon the resistivity of high-purity tungsten is presented. It is shown that some of the data appear to fit quite well to Soffer's model (1967) of surface scattering. From this new analysis values for the product rho _infinity lambda _infinity (bulk resistivity times bulk mean free path) may be obtained lying in the range 1.0-1.7 f Omega m². Early data, which could not be interpreted using Fuchs' theory (1938) of surface scattering, may now be explained and yield surface roughness ratios (ratio of root mean square height of deviations from the mean surface to the Fermi wavelength) of the order of 0.2-2.5, typical of the value expected for atomically smooth to atomically rough surfaces. However, more recent data violate Soffer's theory in a convincing fashion. They illustrate fully the severe dangers of using a theory to interpret data for which it is not at all applicable. The Fermi surface of tungsten by no means approximates a simple sphere (one of the basic assumptions of the Soffer model). Hence such a naive treatment as Fuchs' leads to meaningless parameters such as quoted above. The authors present a warning to all experimentalists who attempt to interpret data from metals with complex Fermi surfaces in a simple-minded fashion (e.g. using Fuchs' model).

For pt.I see ibid., vol.10, p.893 (1980). Predicts the existence of a novel collective mode in a dense Kondo system. Its influence on the shapes of the densities of one-particle states of the system is studied, focusing especially on the widths of the peak and dips found previously at the Fermi level. The collective mode narrows these features (they tend to zero) for an extremely high concentration of localised magnetic moments, whereas the Kondo resonance scattering mechanism makes them maximal in the less doped region only. The self-energy of an individual conduction electron is found to decrease asymptotically as omega ^-5 when leaving the Fermi level ( omega =0). In a uniform Kondo lattice system the collective mode vanishes, together with the resonance scattering mechanism.

Based on the all-electron local density functional theory, the electronic structures and cohesive energies of close-packed hexagonal monolayers of the alkali atoms Li-Cs have been calculated using the self-consistent full-potential linearised augmented plane wave method for thin films. The energy band structures of the monolayers of the 'light' alkali atoms Li and Na exhibit wide occupied bands of almost pure s character and do not show any sharp spectral features in the unoccupied part of their densities of states up to the vacuum level. In contrast, the monolayers of the 'heavy' alkali atoms K, Rb and Cs are found to have markedly smaller widths of their occupied bands and increasingly sharper structures in their densities of states above the Fermi energy. These structures originate from d-like states which, in the case of Cs, are present even at the Fermi energy. The theoretical work functions obtained for the monolayers are consistently larger by about 0.5 eV than the experimental values for clean metal surfaces, but about 2 eV smaller than the first ionisation potentials of the free atoms. In accordance with the trend observed for bulk alkali metals, the cohesive energies in the monolayers decrease from 1.38 eV for Li to 0.71 eV for Cs and are found to be surprisingly similar to the theoretical values for the bulk metals, if the number of nearest neighbours is taken into account.

A model is developed for the change in superconducting transition temperature of metallic glasses due to the presence of two-level systems (TLS). The effect of such low-energy excitations is described by an electron-TLS interaction, which results in the addition of a TLS-mediated term to the usual phonon-mediated BCS electron-electron interaction. The model gives good qualitative agreement between calculated and measured superconducting T_c; in particular, it leads to a nonnegligible enhancement by the TLS, and corroborates the hypothesis of a reduction of the TLS density of states upon thermal relaxation.

A reanalysis of existing relevant experimental data shows that at low temperatures (T<or=0.4T_C) the variation of magnetisation with temperature can be well explained, for both Ni and Fe, by adding a T² term to the usual Bloch T^3/2 spin-wave term. In the particular case of Ni such a procedure now reconciles the values of the spin-wave stiffness constant as deduced from magnetisation and neutron measurements. The T² term is thought to arise from Stoner nonspin-flip excitations rather than Stoner spin-flip excitations since the latter are found to be unimportant at low temperatures. At higher temperatures collective excitations in the molecular field (as distinct from single-site spin reversals in such a field) and Stoner spin-flip excitations contribute to the decrease of the spontaneous magnetisation. Regarding the former as some sort of an 'optic' spin-wave mode with a quadratic dispersion relation, a cross-over point between the resulting two spin-wave modes in Ni is obtained at about the same wavevector (0.47 AA^-1) and energy (120 meV or 29 THz) as has been found in recent neutron measurements of the spin-wave dispersion curves. A similar cross-over is predicted to occur for pure Fe at a wavevector of about 0.74 AA^-1 and an energy of about 166 meV (40 THz). It is further suggested that the intersection of the acoustic spin-wave mode with the Stoner spin-flip excitations should occur at energies much higher than the cross-over energies.

A recent generalised perturbation method starting from the completely disordered state is used for the computation of interaction potentials responsible for ordering effects in ferromagnetic Fe alloys. It is found that the 'effective' pair potential at the first neighbour position varies with the impurity concentration and even changes its sign in FeCr. A comparison with experimental data is tentatively made and discussed, in spite of the lack of experimental data at sufficiently low quenched state temperatures.

The magnetisation of polycrystalline Er and alpha ErH_x solid solutions (0<or=x<or=0.03) was measured between 4.2 and 200K in applied fields 0.2<or=H<or=55 kOe. The authors have observed a drop in the paramagnetic Curie temperature, theta _p, from 40K in Er to 31.5K in ErH_x (with no significant x dependence); a decrease of the antiferromagnetic and the helical transition temperatures, T_N and T_H, with increasing x; an increase in the Curie temperature, T_C, and in another critical point, T_Cl, with increasing x; and a strong magnetic hardening of the solid solutions, with the appearance of an x-dependent critical field, H_c=18, 16 and 13 kOe for x=0.01, 0.02 and 0.03. The decrease in T_N and T_H is attributed to a weakening RKKY exchange interaction due to a pumping of conduction electrons by hydrogen. The other results and, in particular, the striking similarity between the present data and those reported by other workers using Er single crystals measured in fields parallel to the basal plane are explained through an anisotropic magnetoelastic interaction of hydrogen atoms with the crystal field, leading to an opening-up of the top angle of the ferromagnetic cone structure thus turning the total moment towards the basal plane.

Magnetisation and specific heat measurements are reported for ferromagnetic samples Eu_xSr_1-xS_0.50Se_0.5O. Using modified Arrott plots the critical exponents beta , gamma , and delta are determined for different concentrations x down to the critical concentration x_c. Whereas beta and gamma increase continuously with disorder up to beta =0.44 and gamma =1.84, delta remains at about delta =5.0. The Widom relation gamma = beta ( delta -1) is satisfied and universal functions describing the dependence of the scaled magnetisation on the scaled magnetic field exist for all concentrations. The authors show that the strong broadening of the specific heat anomaly close to xi _c is consistent with a critical exponent of the specific heat alpha , as derived from the Rushbrooke relation (1972) 2= alpha +2 beta + gamma . The critical behaviour within the spin-glass regime x<x_c is discussed.

By combining ¹⁵¹Eu Mossbauer spectroscopy (on the Eu²⁺ S-state ion) and EXAFS (extended X-ray absorption fine structure) measurements in a series of amorphous RE_1-xX_x (RE=rare earth, X='sp' metal) the authors have investigated the angular correlations and the radial distribution functions (RDF) of the local atomic arrangement restricted to the first neighbour shell. The electric field gradient (EFG) parameters, including the modulus of q_z, its balance in sign and the marginal distribution of the asymmetry parameter eta , display strong dependences on the nature of the element X. These various distributions rule out a single description of short-range order and they are discussed by reference to models of structure. The shape of the first peak of the partial RDF centred on the major (RE) and on the minor (X) atomic components is asymmetric, with the extent of asymmetry depending upon the element X. No significant structural change is detected as a function of the rare-earth element, including rare earths with different valence (Eu²⁺, Gd³⁺). The degree of structural short-range order is evaluated consistently from the EFG distribution and from the asymmetry of the RDF as determined from EXAFS. The decrease of the short-range order in the sequence Au, Ga, Al ... is correlated with a decreasing enthalpy of formation of the corresponding crystalline compounds. On the other hand, no clear relationship can be established between short-range order in amorphous and crystalline phases. In the new amorphous metals Eu₇₀Mg₃₀, Eu₇₀Zn₃₀ and Eu₇₅Cd₂₅ which have been studied here, one observes similarly (from the EFG distribution) that the order is correlated with the enthalpy of formation.

The hyperfine interactions and Mossbauer parameters of a 'Pd₆Sn' complex in iron have been measured as a function of temperature and pressure. The pressure coefficient of the hyperfine field is found to be -0.32+or-0.04 Oe bar^-1. This result is discussed in the light of models for the generation of hyperfine fields at nonmagnetic elements dissolved in iron, and is found to indicate changes in the charge screening process as being mainly responsible for the pressure dependence.

Using the model interpolation scheme band structure of paramagnetic iron given recently by Baker and Smith (1977) and the special direction method for performing the Brillouin zone integrations the authors report calculations of the imaginary part of the dielectric function and optical conductivity. The effect of including matrix elements and a phenomenological relaxation time are also investigated. The results are compared with the experimental data.

The authors have made quantitative calculations of the elastic interactions between interstitial hydrogen atoms in Pd metal. These calculations use the Harwell HADES code, and hence go beyond the usual harmonic models. Results have been obtained for several potentials and, where appropriate, agree well with those of previous workers. They find (i) that the absolute values are sensitive to assumptions for the potentials, suggesting caution in the prediction of thermodynamic properties, and (ii) that there are significant few-body terms not included in the usual approaches. These extra terms affect the equilibrium structure, for example by removing the symmetry between fractional occupancies theta and (1- theta ), and may lead to the initial nucleation of metastable structures during hydrogenation. The present results suggest that corner-sharing tetrahedra are favoured.

The precipitation kinetics of Au-2 at.% Co are investigated by simultaneous measurement of the saturation magnetisation and the electrical resistivity during isothermal annealing. The degree of precipitation determined using the two methods shows deviations of up to 15%; this discrepancy can be explained through the properties of the matrix and not the precipitates: Kondo scattering from cobalt atoms dissolved in gold as well as typical deviations from Matthiessen's rule must be considered for the interpretation of the results.

From the RDF of liquid argon, liquid chlorine and amorphous silicon it is concluded that chemical bonding between nearest neighbours exists whenever the minimum following the first peak tends to zero. From a reexamination of recently published data on Fe₈₀B₂₀, the evaluated RDF shows a well defined minimum after the first maximum. This directly implies chemical bonds between Fe-Fe atoms in this metal-metalloid system, excluding the description of the iron matrix as DRP based upon frozen liquid structures, and favours cluster models such as that proposed by Gaskell (1978).

de Haas-van Alphen (DHVA) measurements have been carried out in fatigued copper crystals with average dislocation densities of approximately 10¹¹ cm^-2. The signals are unexpectedly large because the long-range strain fields of the individual dislocations are suppressed by correlations in their spatial distribution. This emphasises that dislocation arrangement as well as density is important in the DHVA effect. If the results are interpreted in terms of scattering, the ratio of the total (DHVA) scattering rate to the momentum-weighted (resistive) scattering rate is about ten, only four times larger than for impurity scattering. This implies that the dislocation scattering mechanism important in resistivity is large angle.

The mean free path dependence of the RKKY indirect exchange interaction is applied in a nearest-neighbour mean-field approach to give a simple universal relation between initial ordering temperature and resistivity. This is found to be consistent with most of the data on heavy rare-earth metals and alloys in the literature and gives an estimate of 4.7 AA for the mean free path of Gd at 300K.