Table of contents

The authors present the theory of a method for calculating the surface tension gamma of liquid simple metals. The method uses the direct perturbation expansion to second order of the electron-ion pseudopotential. It is found that the second order correction to gamma is significant for liquid metals and that certain important pseudo-atom models which have previously been proposed (Evans and Kumaravadivel (1976), Mon and Stroud (1980)) are inadequate, at least in principle, from a microscopic point of view. The calculated values of gamma for Na and Al are in good agreement with experiment.

The composition dependence of the change in the hard-sphere diameters component metal, due to the charge transfer on alloying, is obtained from the entropy and from the structure factor data on the basis of the hard-sphere model. The results obtained in these two different ways agree well with each other and are consistent with the result of the first principles calculation and with the result deduced from the electronegativity.

Ion channelling measurements in an Al-0.04 at.% Sn crystal have shown that the Sn atoms trap irradiation induced vacancies to create trivacancy-Sn, tetravacancy-Sn and hexavacancy-Sn complexes. The Sn atoms in these complexes occupy interstitial positions at the centre of a triangle of nearest neighbour vacancies, at the tetrahedral site and at the octahedral site respectively. Quantitative measurements were made of (i) the growth of these complexes caused by vacancy trapping at 200K following 35K irradiation, and (ii) the decay of these complexes caused by self-interstitial trapping during irradiation at 70K.

An explanation is given of the discrepancy which is observed in aluminium between the value of R₁ (equal to r_t/r_v, the ratio of the capture radius of a self-interstitial atom (SIA) by an impurity to that by a vacancy) determined by radiation damage rate measurements performed above stage I and by a kinetic analysis of stage I recovery for low-temperature irradiated samples. It is shown that the multiple trapping ceases to be negligible in the damage rate method as soon as the induced defect concentration, C₁, becomes significant compared to the trapping impurity concentration, C₁, i.e. as soon as C₁>or=R₁C₁/10. The effect of the multiple trapping is more important when r_t is low compared to r_i, the capture radius of a SIA by another SIA.

A breakdown of the additivity of the specific heat is found in liquid Na-Tl alloys. The maximum deviation of the additional specific heat occurs near the equiatomic composition.

If the EMF of a concentration cell of the type A(liquid) mod electrolyte (A⁺) mod A-B (liquid) is measured as a function of pressure as well of temperature and alloy composition then in principle all the thermodynamic properties of a binary liquid alloy A_cB_1-c can be determined if they are known for the pure components A and B. A function eta (c)= Delta V_i/(1-c_i)² is defined and used together with the analogous function beta (c)= Delta S_i/(1-c_i)² to obtain expressions for the alloy molar volume, expansivity, heat capacity and compressibility. Some results for the sodium-mercury system obtained using a beta-alumina solid electrolyte are presented.

The authors have measured the thermoelectric powers of a series of amorphous transition metal alloys and find, contrary to expectation, that they are not simply proportional to temperature below about 200K. They argue that the results can be best understood as being due to electron-phonon enhancement effects and show that there is agreement between the experimentally deduced electron-phonon enhancement factors and those calculated from superconductivity data.

The authors study the longitudinal and transverse spin-correlation functions for an amorphous rare-earth alloy in the random anisotropy model within the Oguchi pair approximations. Both ferromagnetic and antiferromagnetic interactions are considered, as well as correlated and uncorrelated nearest-neighbour anisotropy axes. They relate their results to the elastic coherent spin-flip neutron scattering cross section, which should provide a test for the model.

Following studies of soft X-ray emission from very dilute solutions of Mg in Li, the authors report here investigations on the analogous, but metallurgically less favourable, case of Al in Li. The findings substantiate the prediction made by Inglesfield (1972) of a band of screening states lying below the host Li band.

The displacement of atoms from their normal positions, which is often observed in long-period superlattice (LPS) alloys, is investigated theoretically using the virtual crystal model. It is shown that the charge density wave (CDW) which exists in the LPS is the possible cause of these atomic displacements. If atomic potentials in binary alloys are divided into (i) the mean potential of the two kinds of atom and (ii) the potential difference, then the displacement waves of the optical mode are produced by the interaction between the CDW and the mean potential, while those of the acoustical mode are produced by the interaction between the CDW and the potential difference. The theory, when applied to the specific cases of L1_0-s and L1_2-s type of LPS alloys, gives results which are in qualitative agreement with the atomic displacements observed for CuAu and Au₃Cd.

The single-crystal elastic moduli of several Nb-V and Nb-Ta solid solutions were measured at 300K with, and without, an H or D interstitial solute. The intent was to determine whether the enhancement of the terminal solubility for H(D) in these alloys is related to any of the following properties: (a) absolute values of the C_ij, (b) elastic anisotropy, (c) relaxation of the C' shear modulus caused by H or its isotopes, or (d) the increase in the C₄₄ shear modulus caused by solute H(D) and presumed to result from electron band shifts. The variations in C_ij for the binary alloys of Nb-V suggest and atomic ordering that may be related to the H solubility. The magnitude of the effect of H on C₄₄ varies with alloy composition and may indicate that indicate that singularities in the electron band-structure of Nb are removed by adding V or Ta, thereby significantly changing the electronic effects of adding H.

The Gibbs-Bogoliubov variational method based on the hard-sphere reference system is considered to calculate the heat and entropies of mixing of equiatomic AlMg, CdIn, CdTl, InZn, CdZn and MgZn liquid alloys. A Heine-Abarenkov form of model potential is used in conjunction with the Vashishta-Singwi dielectric function. The results of internal energies and entropies of divalent and trivalent simple liquid metals are also presented. The alloying properties are found to be sensitive to the structural part of the free energy.

A new self-consistent cluster theory for studying the electronic structure of substitutional alloys exhibiting both long- and short-range order is developed and examined in detail. The formalism is based on an extension of the self-consistent boundary-site approximation (SCBSA) introduced by Butler (1973) for a linear chain. General off-diagonal disorder in the tight-binding picture is also included within the framework of the renormalisation propagator formalism (RPF) proposed by Niizeki (1975). The effectiveness of this method is evaluated from numerical studies of NaCl structure fictitious alloys as well as CsCl structure order-disorder alloys. A recent paper by Kumar and Joshi (1978) reported that the application of the SCBSA to the three-dimensional simple cubic systems yielded some unphysical features. It is, however, shown that the authors do not encounter any such difficulty throughout the present investigation if they pay attention to computational accuracy.

The field dependence of the hyperfine interaction of ¹⁶¹Dy in Al has been studied in external magnetic fields up to 3.21 T by means of Mossbauer spectroscopy. ¹⁶¹Dy was introduced into a single crystal of Al by means of low-temperature ion implantation. From the measurements the authors determine the parameters of the cubic crystalline electric field acting on the Dy nuclei: A₄(r⁴)=-32+or-32K and A₆(r⁶)=14.1+or-1.4K. These parameters correspond to a Gamma ₇ ground doublet separated from a Gamma ₈ excited quartet by Delta =26+or-6K and an overall multiplet splitting of 143+or-25K.

A theory of the electrical conductivity of liquid non-simple metals is formulated based on the tight-binding model by extending the effective medium approximation which is known to be the best single-site theory for the electronic density of states. Special emphasis is given to the consistency between the calculation of the one-particle Green function and that of the response function and the Ward identities connecting both the functions and the related conservation laws prove to be satisfied in the present scheme. Also the formulation is performed in such a way that the extensions to the multi-band, the muffin-tin and the s-d mixing models can be done straightforwardly.

Previous data on the resistivity of the pure noble metals has been reanalysed in the light of a recent paper by Bergmann, Kaveh and Wiser (1980). New data on strained pure metal and alloy samples have been similarly analysed and show the variations in electron-phonon and electron-electron scattering predicted by theories based on anisotropic scattering on the Fermi surface.

The dispersion equation for a helicon-phonon interaction has been derived for the case when non-local effects are significant. Numerical studies have been carried out for potassium. The nature of the variation of the damping of the four modes with the magnetic field strength has been studied. The results confirm that in the interaction region, mode conversion takes place and the waves are of a hybrid nature.

The electronic structure and spin-polarisation of five-layer vanadium (100) thin films have been calculated using the self-consistent-charge spin-polarised discrete variational X alpha method. The calculations were made for the three thin-film spacings between the surface and the inner adjacent layers at the bulk value, at a 10% contracted value and at a 10% expanded value. The bandstructures the layer density of states and the charge and spin densities are presented. It is shown that the spin polarisation appears on the surface layer and that the magnetic moment is more than 0.2 mu _B for each atom on the surface. The surface relaxation affects the distribution of the spin polarisation on the inner layers as well as on the surface layer.

Values for the magnetic Gruneisen parameter gamma _M for CuMn, (Cu-10 at.% Al)Mn and AgMn alloys determined form low-temperature thermal expansion and specific heat capacity measurements are reported. mu _M is found to be both temperature- and concentration-dependent in contrast to the simple RKKY model for the spin-glass ordering. The concentration dependence of mu _M differs significantly from that reported earlier based upon a combination of thermal expansion and specific heat capacity measurements made independently on different samples. Dipolar anisotropy is discussed as a possible cause of the dependence of mu _M on temperature and magnetic impurity concentration.

The hyperfine field (H_tau ) of V, Mn, Sb and Au in Fe and of Mn in Ni has been measured as a function of pressure to 15 kbar at room temperature and from 4.2K to room temperature as a function of temperature at atmospheric pressure. The magnitude of ( theta H_T/ theta P) _T for Mn in Fe is much larger than that of other 3d impurities, supporting the view that the well-known anomalous temperature dependence of the HFF of this impurity is due to a narrow bound state at the Fermi level. In the low-temperature limit only the HFF of Mn and Au in Fe exhibit a temperature dependence significantly different from that of the host magnetisation. The anomalous temperature dependence of the HFF of Sb in Fe near room temperature is removed by correcting to constant volume.

Spin-density functional theory, in the local spin-density approximation (LSDA) was used to calculate the total Fermi contact contribution K_s, to the Knight shift in solid Mg. The self-consistent spin-polarised Kohn-Sham equations were solved for the spin and number densities by the linear augmented plane wave method in the muffin-tin approximation. For the exchange-correlation energy functional the authors used the recent accurate results of Vosko et al. (1980), K_s was found to consist of the contributions 0.138%, -0.003% and -0.007% due to the Fermi surface, core and valence electrons below the Fermi surface respectively, yielding a total which is 14% larger than the experimental value. The authors compare their results previous LSDA calculations which used an ion-in-jellium model.

A detailed comparison is made between the experimental results on Alfven waves propagating through Bi samples and the numerical results obtained by a recently developed non-local formalism. At both of the frequencies 45 and 300 GHz, samples with trigonal and binary surfaces were used. It is shown that a reasonably accurate value for the relaxation time may be derived from the experimental data. In contrast to the local calculations, the non-local calculations lead to good overall agreement with the experimentally observed dispersion and damping of the Alfven waves, the deviations being explicable in terms of the de Haas-van Alphen oscillations of the Alfven wave velocity.

Recently improved techniques have led to theoretical calculations which give excellent agreement with soft X-ray emission (SXE) and X-ray photoemission (XPS) from alloys. The authors present here experimental results for SXE from very dilute solutions of Mg (about 1 at.%) in BCC Li. At these very low concentrations the emitting Mg atoms should have only Li atoms as nearest neighbours, and indeed the L₂₃ spectrum bears no resemblance to that from HCP Mg, being dominated by a large density of screening s states, as previously seen in more concentrated alloys. An emission edge persists, at a lower energy.

The authors present the results of XPS (X-ray photoemission spectroscopy) and XRA (X-ray absorption) studies performed on several intermetallic cerium compounds (CeRh₃, CeCo₅, CeNi₅, CeNi₂, CeRu₄B₄, CePd₃ and CeBe₁₃). In CePd₃ and CeBe₁₃, the cerium is known to be in the so-called intermediate valence state; in CeRh₃ and CeCo₅, the cerium is thought to be in the tetravalent state: Ce⁴⁺ (4f⁰5d²). Finally for CeNi₅, CeNi₂ and CeRu₄B₄ the situation is unclear. The main purpose of this paper is to show that the valence of cerium in these compounds cannot be deduced from XPS and XRA measurements in the same way as it is for other rare-earth compounds (Tm or Sm for instance). Indeed pure tetravalent compounds like CeRh₃, CeCo₅ or even insulating CeO₂, present XPS and XRA spectra which are similar to those obtained for intermediate valence compounds like TmSe, SmB₆, etc.... If one supposes that the final states starting from a Ce⁴⁺ initial state are essentially made of configurations which contain one more electron in the 4f shell, the authors show that the valencies of cerium at the surface and in the bulk, deduced from their XPS and XRA experiments, are in approximate agreement with those deduced from lattice parameter measurements. Then, the direct observation of replicates in the XPS or XRA spectra of these cerium intermetallic compounds cannot be considered as a proof of the existence of an intermediate valence state.

The 2p and 3p core level widths for the metals ²²Ti-³⁰Zn have been studied by photoelectron spectroscopy. The L₂L₃M₄₅ Coster-Kronig rate is found to have a maximum for Co and Ni. The Coster-Kronig broadening seems to be a simple function of the kinetic energy of the emitted electron. Absolute 2p level widths have been determined which in general show much higher accuracy than previously reported values. The presently determined widths are found to agree well with theoretical predictions. The M₂M₃M₄₅ super Coster-Kronig decay channel is open in Co but not in Ni.