Table of contents

The partial radial distribution functions g_ij(r) are obtained for a liquid Li_0.61Na_0.39 alloy at 590 K by the molecular dynamics (MD) simulation method. The phase separation or the homocoordination tendency is characterised by the fact that the first peaks of g_LiLi(r) and g_NaNa(r) are higher than that of g_LiNa(r). The concentration-concentration structure factor S_cc(q) derived from the MD g_ij(r) by Fourier transformation is in good agreement with the experimental one as well as with that calculated recently by the hypernetted-chain approximation.

It is shown that the binding energy per valence electron of a simple metal is well described, in atomic units, experimentally and theoretically by the formula a+bz/4r_c. Here z is the valency, r_c is the empty-core radius, a approximately=0.1 and, in all except the group 1A and 2A cases, 0.9<b<1.0 (In the group 1A and 2A systems, b is a little larger than unity.) The present study clearly demonstrates the presence, role and contribution of the free electrons.

The authors report self-consistent energy band calculations for UFe₂ and UCo₂ in four approximations: both semi-relativistic and fully relativistic for the paramagnetic ground states; both spin-polarised and spin-polarised with spin-orbit coupling included self consistently for the ferromagnetic ground state of UFe₂. They compare ab initio calculated and measured lattice constants, bulk moduli, linear specific heat coefficients, pressure dependence of the magnetic moment and magnetic form factor. UCo₂ is found to be an enhanced paramagnet and UFe₂ an itinerant ferromagnet with a magnetic moment due mainly to the iron moments. However, the magnetism on the uranium sites is unusual in that they calculate both spin and orbital contributions to the moment that are large, antiparallel and almost equal.

Measurements have been made at low temperatures of the heat capacity (2 to 20 K) and magnetic susceptibility (4.2 to 300 K) in the cubic compounds YbPdSb and YbPdBi. From the linear portion of plots of C/T against T² the authors obtain gamma values of 470 and 240 mJ mol^-1 K^-2 for YbPdBi and YbPdSb respectively. C/T exhibits a sharp upturn at low temperatures and exceeds 1 J mol^-1 K^-2 at 2 K. The magnetic susceptibility results reflect the fluctuating nature of the valence of Yb ions in both compounds.

The magnetic susceptibility of liquid K-Sn alloys has been measured as a function of temperature and concentration. The curve of the susceptibility isotherm exhibits a distinct minimum around the composition K_0.55Sn_0.45. This anomaly in the concentration dependence of the electronic behaviour is discussed in terms of compound formation.

For pt.I see Hafner et al., ibid., vol.14, p.1139, (1984). A simple model used by Hafner et al. to describe the structure and thermodynamics of liquid alloys with chemical short-range ordering is extended to describe liquid Li-Na. The model is based on simple approximations for the electron-gas and pseudoatom self-energy contributions to the heat of mixing and model-pair interactions of a hard-core Yukawa form. It achieves a reasonably consistent description of the energetic conditions for phase separation and of the critical divergence of the long-wavelength concentration fluctuations.

The properties of large vacancy clusters, containing from 10 to 40 vacancies, were investigated for copper using an interatomic potential derived from the pseudopotential method. The stacking fault energy of copper was calculated to be 65.3 mJ m^-2 for a potential cut-off radius of 8.5155 AA, which is in good agreement with the experimental value of approximately 70 mJ m^-2. For longer cut-off radii, however, the calculated stacking energy decreased and became negative. It is suggested that this results from the analytical fit of the numerically calculated interatomic potential being inaccurate at relatively long interatomic radii. Using the cut-off radius of 8.5155 AA, the minimum energy configurations of hexagonal and triangular vacancy platelets of 10 or more vacancies were determined, and it was found that they collapsed into hexagonal faulted loops and stacking fault tetrahedra, respectively. This agrees with X-ray diffuse scattering experiments in which loops of this size have been observed, and with previous calculations for copper using empirical interatomic potentials. The calculated displacement field at short distances above the small hexagonal loops differed significantly from linear isotropic inelasticity theory predictions. The stacking fault tetrahedra were found to be more stable than the hexagonal loops, although both were locally stable, which is in accord with experimental observations of both defects in copper.

The positron annihilation characteristics, namely the positron lifetime and the angular correlation curve, has been calculated for small Zn clusters in an FCC Al matrix. The positron potential is obtained from the full lattice electrostatic potential and correlation potential, within the local density approximation. The wavefunction and eigenvalue of the lowest-lying positron state is determined by solving the Schrodinger equation in the finite different scheme. The results show that the positron forms a bound state for clusters containing as few as nine Zn atoms. The positron lifetime decreases with increasing Zn cluster size with a tendency to saturate for clusters larger than 13 atoms. The angular correlation curves are much more sensitive to the variation in cluster size, especially the S parameter which measures the ratio of core annihilations to valence annihilations. The results of the calculation suggest that the kinetics of the very early stages of precipitation in alloys can be analysed by careful positron annihilation experiments.

Small loops and stacking fault tetrahedra of sizes 1 to 10 nm were created during irradiation of copper with 14 MeV neutrons at room temperature. The electrical resistivity of these defect clusters has been investigated using a combination of resistivity and electron microscopy techniques. The dislocation specific resistivity of the loops was determined to be rho _d approximately=1.4*10^-25 Omega m³ at a temperature of 4.2 K, in fair agreement with the literature value for straight dislocations in copper. The effective Frenkel pair resistivity of the point defects in the loops was constant over the fluence range 1*10²⁰ to 2*10²¹ n m^-2, with a value of about 1.7 mu Omega m per unit concentration of Frenkel pairs. This is only slightly less than the specific resistivity of isolated Frenkel pairs in copper, and suggests that clustering of point defects does not necessarily lead to a large decrease in the Frenkel pair resistivity. The ratio of surviving to created point defects ( eta ) decreased from 11 to 4.7% as the neutron fluence increased from 1*10²⁰ to 3*10²¹ n m^-2.

The low-temperature lattice specific heat of palladium-based dilute alloys PdAg, PdRh, PdRu, PdMo and PdNb has been studied theoretically on the basis of Green function theory. A nearest-neighbour impurity model with central and non-central force constant changes has been employed to discuss the crystal impurity problem. The effect of volume changes has also been considered. The lattice specific heat in these alloys largely depends on the central force-constant change which seems to be quite high in a few cases. The possible origin of such high force-constant changes in these alloys has been discussed.

The authors report measurements of the ⁶Li and ⁷Li diffusion coefficient D^NMR in both natural Li and enriched ⁶Li by the NMR pulsed magnetic field gradient techniques. Between 350 K and the melting point (454 K) D^NMR obeys an Arrhenius law with an average activation energy of 0.561+or-0.002 eV. The isotope effect can be explained in terms of classical statistics (mass dependence of the attempt frequency). By combining their D^NMR data with the uncorrelated self-diffusion coefficient D^SD determined by NMR spin-lattice relaxation measurements they obtain the effective correlation factor which decreases with rising temperature.

To change the elastic moduli of a material by several hundred per cent is a noteworthy achievement. In metallic multilayer thin film structures, this phenomenon, 'the supermodulus effect', can be selectively incorporated. To date, this effect has only been observed in man-made structures fabricated with an artificial one-dimensional composition modulation. It has not yet been found in naturally occurring systems. To truly understand the cause of the supermodulus effect will prove useful, in concept, in elaborating the inter-relation between microstructure and physical properties and, in application, in the search for structures as yet unidentified. Some insight into the cause of this effect has been obtained for single unbounded layers using a rather practical 'coherency strain' model. This approach will be expanded upon in this paper to model the effect in the actual multilayer systems measured experimentally. The role of the interface in these materials of nanometric-scaled periodic structures will be seen to play a key role in understanding and modelling the supermodulus effect.

The wavelength dependence of the X-ray diffraction peak intensity in the small-angle region has been measured on two Fe/Mn superlattice films using synchrotron radiation. Peak intensities are found to be greatly enhanced at a wavelength slightly longer than the K-absorption edge of the Mn atom (1.896 AA), where the contrast in X-ray scattering power of the adjacent Fe and Mn layers becomes significantly large because of the effect of anomalous dispersion. Higher-order reflections up to the seventh are observed for (Fe(15 AA)/Mn(50 AA))₄₀ and up to the third for (Fe(20 AA)/Mn(20 AA))₁₀₀. Analysis of the peak width and intensity reveals intermixing of the atoms extending over several atom layers ( approximately 10-12 AA) at Fe/Mn interfaces and the fluctuation of the superlattice period within approximately 3-4 AA.

Multilayered films composed of iron metal and amorphous carbon have been investigated by Mossbauer spectroscopy. When iron layers are very thin (8 AA and 15 AA), reduced magnetic hyperfine fields with distributions are observed, which indicate that amorphous Fe layers containing carbon atoms are formed. Mossbauer spectra of a sample with 30 AA Fe layers and of an interface-selectively ⁵⁷Fe-doped sample indicate that the structure of the major portion of the Fe layer is BCC but amorphous Fe-C alloy layers are formed at BCC-Fe/amorphous-carbon interfaces, the thickness of which is estimated to be about 5 AA.

A general expression for the 'on the energy shell' relativistic t-matrix for non-spherically-symmetric potentials is derived. It is shown that, in the special case of a spin-polarised target, the resulting expression is the same as that obtained recently by Strange et al. (1984).

The recent data of Bass et al. (1986) for the electrical resistivity of concentrated KRb alloys at very low temperatures were found to be 'anomalous'. A calculation of the resistivity is presented here that includes the Pippard ineffectiveness condition for inelastic electron-impurity scattering and the contribution of the electron-phonon scattering term with material parameters appropriate to concentrated alloys. The resulting calculated values are in excellent agreement with the data over the entire temperature range measured.

The thermopower of pure iron and dilute iron-vanadium alloys (0-1.8 at.% V) has been measured in the temperature range 300-1750 K. The temperature dependence of the thermopower for the BCC phases above and below the FCC regime has been clearly established for the first time. The large differences between the BCC and FCC values reflect the difference in band structures, and the large value for the lattice indicates the presence of fine structure in the energy spectrum and hence a strong energy dependence of scattering.

The authors formulate a density functional theory (DF) to describe non-collinear magnetism. Self-consistent, spin-polarised energy-band calculations based on the local approximation to DF theory are presented in which the magnetisation associated with different atoms in a unit cell is allowed to point along different, non-collinear directions. Non-self consistent calculations employing non-collinear quantisation axes have been presented before; the present calculations are, they believe, distinguished by: first, being self-consistent; second, providing the total energy; and third, providing the spin-quantisation axes. In their first applications they deal with the non-collinear antiferromagnets gamma -FeMn, RhMn₃, and PtMn₃ and show that their total energies are minimised in the tetrahedral (FeMn) or triangular (RhMn₃, PtMn₃) magnetic structures first proposed by Kouvel and Kasper (1963).

The authors develop two alternative reciprocal space methods applicable to the calculation of the spin susceptibility of transition metals. The first, a complex energy method, uses the tau matrices of scattering theory, while the second is based on the one-particle energy eigenvalues and the wavefunction coefficients. They demonstrate that both methods may be used to evaluate the non-interacting susceptibility by performing Brillouin zone integrations. As a second step the interacting quantities may be obtained using the methods described in previous work. Their applications to Pd yield a remarkable similarity between results gained in these different ways.

To investigate the magnetism and the stability of the crystal structure of the Laves phase compounds AB₂, the total energy was systematically calculated by the LMTO method within the framework of the LDS approximation. For the non-magnetic compounds the crystal structure is determined by the B atom: the stable structure is the C15 structure when the B atom is Cr, Co or Ni(Mo, Rh or Pd) and the C14 structure when the B atom is Mn or Fe (Tc or Ru). In the Fe Laves phase compounds AFe₂ the electronic structures change slightly with a change in the lattice spacing, thus inducing a varied magnetic behaviour. The theoretical predictions about the stability of the crystal structure and the magnetism are in good agreement with the experimental results.

The elastic constants and ultrasonic attenuation of single-Q and of single-S single-Q chromium are measured through the spin-flip transition. Comparison of the attenuation data with behaviour at the spin-flop transition in MnF₂ suggests by analogy that the peak seen at the spin-flip transition in single-Q Cr is due to coexisting domains of the longitudinal and transverse states of the spin density wave.

In an attempt to understand the nature of the magnetisation at and around Fe sites in dilute CrFe antiferromagnetic alloys, neutron polarisation analysis has been used to isolate the magnetic diffuse scattering in a single crystal of Cr-2.8 at.% Fe and a polycrystalline sample of Cr-4 at.% Fe. The magnetic diffuse scattering gives a direct measure of the spatial magnetic moment distribution at and around impurity sites in a magnetic host. It has been found that the Fe possesses an ordered antiferromagnetic moment which is coupled into the SDW in both alloys. For the 2.8% Fe sample, the iron moment is (2.13+or-0.05) mu _B, and the chromium moment is (0.72+or-0.02) mu _B. For the 4% Fe sample, the moments are (1.8+or-0.1) mu _B and (0.72+or-0.02) mu _B for iron and chromium respectively. No significant disturbances on the Cr neighbours of Fe atoms have been found.

The magnetic phase diagram of the hexagonal polymorph of FeGe has been determined by means of neutron diffraction on a single crystal at low temperature and for magnetic fields applied perpendicular to the c axis. Between 410 and approximately 60 K the magnetic order is c-axis collinear antiferromagnetic. In zero magnetic field the structure changes to a c-axis double cone antiferromagnetic structure at approximately 60 K. In a magnetic field of 9.4 T, applied perpendicular to the c axis, the corresponding transition is observed at approximately 43 K. At 30 K (25 K in an applied field B₁ of 9.4 T) another phase transition is observed which displays itself by a kink in the temperature of the conc half angle. The interlayer turn angle is almost independent of temperature and applied field. At 4.2 K an anomalous decrease of the basal plane component is observed at a critical field B_cl-1.4 T (B perpendicular to c) and at a second critical field B'_{c perpendicular to} =4.8 T another transition is observed. The critical fields B_{c perpendicular to} and B'_{c perpendicular to} and B'_{c perpendicular to} and the anomaly at B_{c perpendicular to} decrease as the temperature is increased and above 30 K no field-induced transitions (B perpendicular to c) are observed. The cone structure is found to persist up to at least 9.4 T, which was the upper limit of the applied field. Results from previously published macroscopic data (e.g. magnetisation, magnetoresistance, torsion and susceptibility data) are discussed and related to the neutron diffraction data.

The critical magnetic properties of the amorphous Ni-Fe-B-Si system have been studied using AC susceptibility techniques. Alloys of the composition Ni_80-xFe_xB₁₂Si₈ with x>or=5 show re-entrant ferromagnetic behaviour. The critical exponents gamma , associated with the magnetic susceptibility, and delta , associated with the critical isotherm, have been determined. Values of the exponents are consistent with those predicted for three-dimensional Heisenberg systems. The critical susceptibility of the alloy with x=4 in the region above the spin-glass transition is found to follow a power law with an exponent gamma '₀=1.5.

The critical resolved stress tau _M for the martensitic transformation from beta to orthorhombic 9 R has been measured at liquid air temperature for Cu-Zn alloys between 40.5 and 43 at.% Zn. Whereas alloys with a zinc content below 41 at.% Zn also transform spontaneously on cooling, those for zinc concentrations between 41 and 43 at.% transform only when stresses are applied. At 42.5 and 43 at.% Zn the alloys can deform plastically in the beta phase prior to the transformation, depending on the orientation of the stress axis. Measurements have been performed which permit extrapolation of tau _M to the undeformed state. The composition dependence of tau _M between 41 and 43 at.% Zn agrees within 20% with that predicted from data obtained for alloys with a zinc concentration below 40 at.%. This suggests that the same mechanism controls the relative phase stability of beta and 9 R in the whole composition range, and that the energy from the monoclinic distortion of 9 R martensite plays only a minor role.

The CuZn alloys and their derivative CuZnAl have been extensively studied because of their reverse thermoelastic martensitic transformation which leads to shape memory devices. This displacive, diffusionless phenomenon is explained in terms of shear systems. Various models attempt to describe the B2(CsCl type)-9 R transformation. In order to contribute to this controversy the authors investigate CuZn alloys using X-ray absorption spectroscopy as a technique able to follow the changes of local atomic order during martensitic transformation. Investigations start from room temperature down to -231 degrees C. The martensitic transformation starts at -90 degrees C for this alloy. Premartensitic effects have often been reported at room temperature, and the Cu and Zn sets of data are consistent with a crystal deformation coherent with homogeneous sets of strains within two (110)-type BCC planes separated by 60 degrees along (110) directions.

The relaxation spectrum of the twin boundary is studied in the Mn(10-40) wt.% Cu alloys by examining characteristics of the twin relaxation peak which was caused by stress relaxation across the twin boundaries in the FCT martensite. It is concluded that there is a close connection between the tetragonality (1-a/c) and the relaxational characteristics, and the distribution parameter beta obeys beta = mod beta ₀- beta _Q/k_BT mod .

Weiler and Schaefer (1985) have recently published vacancy formation parameters for indium from a detailed analysis of positron lifetime spectra. Flower et al. (1985) have simultaneously published the pressure variation of the elastic constants. It is shown that these two sets of data when combined with earlier self-diffusion studies are interconnected through a thermodynamic relation published by Varotsos and Alexopoulos (1986).