Table of contents

Ultraviolet photoemission measurements (h(cross) omega =21.2 eV) on clean liquid copper samples are presented. On melting, the 3d band structure is observed to change in shape in agreement with recent multiple-scattering calculations.

Results concerning the location of unoccupied 4f levels in rare earth metals reported by Padalia et al are re-examined. (See ibid., vol.4, 938-46 (1974)).

From resistance anisotropy measurements it is shown that the critical composition separating strong and weak ferromagnetic regions in the Ni-Fe system is near Ni₃Fe.

The echo effects generated in metal powders in the presence of a steady magnetic field by the application of pulsed radiofrequency magnetic fields are shown experimentally to be due to collective oscillations of groups of about 13 particles.

The magnetic after effect (MAE) and the annealing of the electrical resistivity has been measured simultaneously in the temperature range of stage I of electron irradiated Ni. The results show that the interstitials which become mobile in stage I are responsible for the magnetic relaxation phenomena observed in this temperature range. From a careful comparison of the magnetic relaxation times with the characteristic times observed for the resistivity annealing it can be concluded that the reorientation of (dumb-bell) interstitials during their free migration is responsible for the observed MAE. The long range diffusion of interstitials into the Bloch walls as possible source for the observed MAE can be ruled out.

The angular force model developed by Clark, Gazis, and Wallis (1964) for body centred cubic metals has been modified by including the electron-ion interaction of Sharma and Joshi (1964). The new model thus obtained has been applied to compute the dispersion relations of the alkali metals lithium, potassium, and rubidium and it is found that it gives very good agreement between the theoretical and experimental results.

A theory, based on a generalized Hartree-Fock scheme, is developed, giving both the atomic and the metallic limit of the Hubbard model correctly. In the strong coupling case an insulating state having localized moments is obtained, and an effective Heisenberg Hamiltonian describing the coupling between these moments is deduced. The coupling is antiferromagnetic and a Neel temperature of the order of t²/U is obtained, t representing the hopping matrix elements, U the Coulomb repulsion. Both the magnetically ordered and the magnetically disordered states are insulating having localized moments. Comparison with the functional integral method is made. The qualitative aspects of the theory are applicable to some transition metal oxides (NiO and MnO for example).

In a previous paper a general formalism for the determination of a parametric expression for the Fermi surface of a metal has been described (see abstr. A16088 of 1972). The method of that paper has now been applied to the determination of symmetry adapted expansions, using plane wave type expansions similar in nature to that for f.c.c. metals used previously by Roaf (1962) for Cu, Ag, and Au. The authors have developed similar formulae for b.c.c. and h.c.p. metals and the general forms of the expansions are tabulated for all three structures, not only for expansions about Gamma but also for expansions about other special points of symmetry in the Brillouin zone. General expressions for these symmetrized expansions are tabulated. Results of the application of this method to certain sheets of the Fermi surfaces of group VA and group VIA metals are presented and some advantages and disadvantages of this particular parametrization scheme are discussed.

Quantities associated with soft X-ray emission, the work function, the pseudopotential and Compton profiles for body centred cubic (b.c.c.) and face centred cubic (f.c.c.) lithium crystals have been calculated in the Hartree-Fock approximation of Kumar, Monkhorst and Harris (Phys. Rev. B1974). The width of the occupied part of the valence band is calculated too large by about 3 eV. Combination of soft X-ray emission and work function data indicates the bottom of the valence band is at nearly the correct position. It is found that the energy dependence of the transition dipole matrix element cannot explain prepeaking below the Fermi edge of the soft X-ray line shape. A pseudopotential form factor v(q) is calculated in the on Fermi sphere approximation. Comparison with an 'experimental' pseudopotential reflects the importance of correlation effects for small q values and the correct lattice spacing for large q values.

Results are presented for the density of states of binary alloys from a calculation using the matrix coherent potential approximation. A form of the density of states with realistic features was obtained even though the quasicrystalline assumption for the off-diagonal elements was not employed. These features arise solely from the local information about small clusters of atoms. However, it is felt from these results that a proper density of state can only be calculated from information of larger clusters.

It is shown that, in certain cases, the measured value of the Righi-Leduc coefficient and thermal resistivity might be significantly different from that which would be obtained in the absence of the phonon contribution to the thermal current. Of particular interest is the possibility of measuring the lattice conductivity in pure uncompensated metals even though it might be three orders of magnitude smaller than the electronic thermal conductivity.

Completely random solid solutions have been obtained for the entire range of Au-Ni alloys by the splat quenching technique. In an attempt to compare the Au-Ni system with the similar Cu-Ni system, the electrical resistance and magnetization were measured over the range from 20 to 60 at%Ni. For corresponding composition the Curie temperatures are consistent, in the two systems, while the temperature dependences of electrical resistance are quite different from those of Cu-Ni alloys.

An exact expression for the exchange and correlation energy of an inhomogeneous electron gas, as defined by Hohenberg, Kohn and Sham (1966), is derived. This expression is separated into exchange and correlation terms and a formula linking the surface exchange energy of a half space to the Kohn-Sham one electron potential follows without approximation. For an infinite barrier model, the local density approximation gives a surface exchange energy 50% greater than the exact value, a large and previously unsuspected error. An exact evaluation of the surface correlation energy is not feasible, but the authors argue that the dominant contribution, arising from the difference in zero point energy between bounded and unbounded systems, can be estimated using a simple model. Numerical results, not dependent on the introduction of arbitrary plasmon wavevector cutoffs, give surface correlation energies larger than Lang and Kohn (1970), who work from a local formula, by a factor of six.

The authors find that Fe is magnetic when introduced as an impurity in the A15 superconductor Mo₇₇Ir₂₃. The susceptibility of the Fe impurity follows a Curie-Weiss law, which can be understood in terms of short range interactions. The superconducting properties can be explained in terms of standard pairbreaking theories including a negative exchange field and the effects of the short range ordering. From this influence of the magnetic impurities on the superconducting properties the Maki parameter alpha and the spin-orbit parameter lambda _s0 have been determined for the pure sample.

The problem of calculating the electron-phonon mass enhancement factor, lambda , in transition metal compounds is discussed and a simple expression for lambda is presented. NbN and NbC are treated in detail, using the results of band structure calculation. The calculation, which estimates lambda from first principles without the use of adjustable parameters, gives good quantitative agreement with lambda obtained from measurements of the superconducting transition temperature, T_c. The physical interpretation of the calculation is that, in both the nitride and the carbide, the light atoms scatter electrons strongly at the Fermi energy due to a p resonance, and cause T_c to be higher than in pure Nb.

The authors have used the TDGL equations, derived by Eliashberg (1969) for a superconductor with a low concentration of magnetic impurity ( tau _sT>>1) near T_c, to calculate the entropy and energy currents and the flux flow conductivity associated with a slowly moving isolated vortex. The effect of the anomalous term in the TDGLE is important. The transported entropy (S_d) and the conductivity ( sigma _f) depend on kappa , tau _s and tau ₁ (transport collision time). S_d is decreased by the addition of relatively small amounts of magnetic impurity. The agreement between measurement on dirty nonmagnetic alloys and theory (with tau _sT= infinity ) is good. sigma _f is calculated for tau ₁/ tau _s<<1. In the dirty limit sigma _f/ sigma approximately=2.67 C_alpha tau _sTH_c2/(B); C_alpha is of order unity for all kappa . When tau ₁T>> tau _s/ tau ₁, sigma _f/ sigma approximately=C_beta H_c2/(B); C_beta approximately=(4 kappa ²)^-1 for large kappa , approximately=0.1 for kappa approximately=1/ square root 2.

It was originally suggested by one of the authors (JGP) that the noise voltage associated with magnetic flux in superconductors, when the regions enclosing the flux are set in motion by a transport current, should have a spectrum which depends on the manner in which the leads connecting the voltmeter to the sample are arranged. The authors report the results of some experiments which clearly demonstrate the existence of the effect in both type 1 and type 2 superconductors.

The magnetic properties of the CsCl type compounds GdMg_1-xZn_x were studied. All these compounds show ferromagnetic behaviour with the asymptotic Curie temperatures close to their ferromagnetic Curie temperatures. In spite of the fact that the nominal number of conduction electron concentration does not change in this series, the authors observed a decrease of the asymptotic Curie temperature with decreasing x and in GdMg is 20% lower than the free ion value. In contrast to these results, which are apparently in conflict with the RKKY approach, the authors observed a hyperfine field at the Zn nuclei which remains unaltered with composition x. All these results, together with the complex character of the Gd NMR resonances observed for low x values (even in the binary compound GdMg are interpreted in terms of an increasing interchange of the magnetic and nonmagnetic sites in the CsCl structure.

The theory of Levin, Bass and Bennemann (see abstr. A69754 of 1971 and abstr. A65117 of 1972) for the spin susceptibility of transition metal alloys is generalized to describe the properties of liquid metal alloys. The short range structural order in the liquid alloy is taken into account using a lattice theory with the lattice chosen to match the experimentally measured short range behaviour of the liquids. The theory is applied to the alloy systems Au-Ni and Au-Fe and the results are compared to experimental measurements on these systems, and found to give good agreement.

Measurements have been made of the resistivity of some rhodium-iron alloys, whose concentrations lie between 40 ppm and 2.2 at% or iron, in the temperature range 0.5 to 30K. The resistivities of samples of 0.06% and 0.5% Fe were also measured down to 0.02K. Comparison is made between the data for the more dilute alloys and the spin fluctuation theory of the resistance anomaly which, in contrast with the Kondo resistance minimum behaviour, shows a large positive temperature coefficient. The effects of interactions between iron solute atoms are observed as a function of concentration and the data give strong, if indirect, evidence that magnetic ordering to a spin glass state takes place at low temperatures in alloys containing more than about 0.5%Fe, while at lower concentrations the ground state is nonmagnetic.

The authors have studied the high field susceptibility of nickel-platinum alloys in fields up to 150 kOe, at temperatures between 4.2 and 300K. The high field susceptibility is analysed in the ferromagnetic and paramagnetic range by means of Arrott plots and Landau coefficients. The paramagnetic initial susceptibility follows a Curie-Weiss law, with a Curie constant independent of nickel concentration. The results are compared with those obtained for palladium-nickel alloys. The experimental data are discussed in the framework of recent theories on itinerant ferromagnetism.

The first order antiferromagnetic(D)-antiferromagnetic(F) transition in the ordered Mn₃Pt alloy is investigated thermodynamically. The disorder of some of the magnetic moments in the high temperature phase produces the entropy driving the transition. It is found that the main properties of Mn₃Pt may be described by a single antiferromagnetic exchange constant J=-6.55*10^-22J. The magnetic susceptibility, specific heat and pressure-temperature phase diagram are computed from the model. The volume change at the transition cannot be accounted for by a volume dependence of the exchange interactions, but is discussed in terms of a change, under the action of magnetic disorder, in the occupation of the electronic states. It is also shown that a pressure dependence of the Curie or Neel temperatures is expected in any material presenting a magnetic volume anomaly. Such a dependence is not therefore, in itself, an evidence for a volume dependence of the exchange interactions.

The results of Rice (see abstr. A13376 of 1971) and Malaspinas and Rice (see abstr. A1642 of 1972) on the effects of imperfect Fermi surface nesting in chromium are discussed to find the conditions under which imperfect Fermi surface nesting might give a first order antiferromagnetic-paramagnetic phase transition. These conditions are found to be not conducive to a stable antiferromagnetic state. A new mechanism for such a first order transition is proposed based on the existence of harmonics of the spin density wave. In this theory the transition is predicted to be always second order when T_N approaches zero. Since Kimball and Falicov (see abstr. A31753 of 1968) found the same result in their model it is likely that the authors' mechanism for a first order transition is the same as theirs. The presence of impurities is shown to reduce the tendency for a first order transition for alloys of chromium.

For pt.II see abstr. A16667 of 1972. The electric field gradients (EFG) at host nuclei being neighbours to impurity atoms have been determined in several Al and Cu alloys by the quadrupole effects in nuclear magnetic resonance. The work supplements previous studies by several workers so that neighbour field gradients due to almost all nontransition metal impurities soluble in the two host materials are now available. This body of results has been compared to present theories of impurity scattering and seems to indicate that these are inadequate to describe the symmetries and sizes of the charge distributions close to the impurities.