Table of contents

Metals with closed orbits typically show a nonsaturating magnetoresistance. It is now well understood that macroscopic spatial inhomogeneities can cause this behaviour. The spatial variations about point defects resemble those caused by macroscopic inhomogeneities in several ways. It is shown, however, that point defects cannot cause the observed lack of saturation.

Corrections from the non-Hermiticity of the pseudopotential, constructed from first principles, on the phonon-limited electrical resistivities of metals are theoretically investigated. It appears that such corrections to the standard calculation are significant, particularly at very low temperatures.

The effect of pressure on the superconducting transition temperature T_c and the fluctuation conductivity Delta sigma above T_c is studied for the bulk amorphous superconductor La₇₈Au₂₂ at pressures up to 137 kbar. It is found that the normalised pressure dependence of T_c, T_c(P)/T_c(0), is nearly identical to that for FCC La and that Delta sigma is a universal function of T/T_c.

It is shown that recent high-field magnetisation data of YCo₂ and LuCo₂ by Schinkel (1978), which he calls anomalous because neither band theory nor spin fluctuation theory accounts for them, can be explained by the Fermi liquid model of magnetism through the logarithmic field dependence of the magnetisation.

Eight de Haas-van Alphen effect frequencies, ranging from 50 T to 8000 T, are reported for a specimen of praseodymium which was refined by solid state electro-transport. The frequencies are consistent with a model Fermi surface based on a relativistic APW calculation for paramagnetic neodymium.

The ¹¹⁵In and ⁷Li Knight shifts and the electrical resistivities of a number of liquid Li-In alloys have been determined experimentally. The possibility of Li₃In compound forming is discussed.

The author examines in a Hartree-Fock approximation, the influence of carbon interstitials on the transition temperature T^{alpha gamma} between the alpha and gamma phases of iron. The author shows that the electronic free energy resulting from the difference of the screening potentials in the alpha and gamma phases is considerable. It is also shown that the solubility of carbon at T^{alpha gamma} is much higher in the gamma phase than in the alpha phase.

In some metals the specific heat Delta c_P (per single vacancy) associated with the temperature variation of the activation enthalpy h^act is large and hence an Arrhenius curvature is detectable. It is shown that, at high pressures, the value of Delta c_P reduces by a large factor and therefore the Arrhenius plot becomes a straight line. Furthermore, it is found that whenever Delta c_P exceeds 35 k the presence of a second diffusion mechanism is obligatory.

The rate at which directional order is established in a Ag-24 at.% Zn alloy exposed to gamma rays or to 2 MeV electrons was studied by in-flux anelasticity measurements. These were performed as a function of dose rate and irradiation temperature for steady state conditions corresponding to a dynamic point defect supersaturation. Comparison of the ordering rates yielded information on the respective production rates of freely migrating defects induced by gamma photons and fast electrons. For gamma photons, the displacement cross section inferred from the results was 0.015+or-0.007 barn. An evaluation of the mean threshold energy in the alloy gave a value of 35+or-2 eV.

The self-consistent density functional approach has been applied to a study of electronic properties of vacancies and vacancy clusters in simple metals. The electron density profiles and potentials have been obtained for spherical voids of varying size. The formation energies and residual resistivities have been calculated for vacancies using both the perturbational and variational inclusion of discrete lattice effects. The relation of the void properties to the plane surface properties is studied, and the inadequacy of the jellium-based methods to high-index faces is demonstrated.

Using elasticity as an illustration a recursion formula is established which gives the relevant upper (+) and lower (-) bounds C^(+or-n), (n=2,3,4... infinity ) for the tensor C^eff of effective elastic moduli of the overall grade n material. It follows from this formula that for n to infinity both C⁽⁺ⁿ⁾ and C^(-n) converge towards the well known self-consistent tensor C^sc which therefore represents a material which, in a statistical sense, is perfectly homogeneous, isotropic and disordered. All these results also have implications for other material properties and for disordered structures other than polycrystals. Since the self-consistent approach is isomorphic to the coherent potential approximation, some of the conclusions should also be relevant for other areas of solid state physics.

Young's modulus and internal friction have been measured between 6K and room temperature for several (Hf_1-xZr_x)V₂ alloys (x=0, 0.25, 0.5, 0.75 and 1.0). A V shaped minimum is found in Young's modulus for each of the alloys. For HfV₂ and ZrV₂ the minimum is centred about a temperature at which, as has been confirmed by X-ray diffraction, there is a structural distortion. A peak in internal friction near this temperature is observed and discussed on the basis of current understanding of the structural transformation.

In cubic crystals not subject to inner displacement, the linear combinations of elastic constants Q₃=C₁₂-C₄₄+C₁₄₄-C₄₅₆ and Q₆=2(C₁₂-C₄₄)-C₁₂₃+C₁₄₄+C₁₁₂-C₁₅₅ are devoid of contributions from central or isotropic interactions. In HCP crystals the combination Q₁=¹/₂(3C₁₂-C₁₁)-C₁₃+C₄₄ has a similar property but is partly influenced by central interactions via an inner displacement term. A special combination R=C₁₃-C₄₄+¹/₂(C₁₄₄+C₂₄₄)-C₁₁₃+C₁₃₃-C₃₄₄+C₃₆₆ has the additional property that it is entirely free from inner displacement effects. A systematic derivation of these extended Cauchy discrepancies is given and their usefulness as direct measures of non-central, non-isotropic interactions is discussed with reference to all cubic and HCP metals whose third-order elastic constants have been determined.

The specific heat from 1.6 to 18K of a lutetium sample which has been well characterised with respect to impurities is reported. Values for the electronic coefficient ( gamma =8.30+or-0.08 mJ mol^-1 K^-1) and the Debye temperature at 0K ( theta _D(0)=185+or-1K) were obtained. These results are compared to previously reported values.

A modified LMTO method is used for self-consistent local density functional calculations of the electronic structure of ScRu, ScRh, ScPd and ScAg, which form ordered compounds in the (cubic) CsCl structure. Energy bands, local partial and total densities of states are presented and compared with experimental results. The lattice constants, heat of formation and the bulk modulus are computed for ScRu, ScRh, ScPd, ScAg as well as ScMo. The bulk modulus is decomposed into angular momentum contributions and the different roles played by s and p electrons on the one hand and d electrons on the other are elucidated.

Pressure derivatives of extremal cross sections of the Fermi surface of palladium have been determined by means of de Haas-van Alphen measurements under pressures up to 3.7 kbar. The experimental pressure derivatives are found to be in good agreement with theoretical values obtained from a self-consistent bandstructure calculation based on the linear muffin-tin orbital (LMTO) method using the local density approximation for exchange and correlation. The calculated extremal areas of the X centred Fermi surface pocket are found to be in better agreement with experiment than those obtained from the standard potential construction. The electronic Gruneisen parameter and the pressure dependence of the electron effective masses have also been calculated.

The detailed bandstructure of Mg having the HCP lattice structure has been calculated using the KKR method for the three different potentials V_s, V_H and V_vs. The potential V_s includes only the exchange part through Slater's formula whereas the other two potentials V_H and V_vs include exchange as well as correlation. The energy bands, density of states, electronic specific heat, magnetic susceptibility and Fermi surface cross sections are calculated for all the three potentials and compared with the experimental data. The results show that the potentials which include both exchange and correlation are in better agreement with the experimental values. However the results obtained with the potential V_s may still be considered reliable and do not lead to any violent disagreement with the experimental values.

The frequency and temperature dependence of the effective mass, m*, and relaxation time, tau , of electrons on the lenticular orbit on the 3rd band Fermi surface sheet of cadmium have been obtained from high-frequency Azbel'-Kaner cyclotron resonance experiments. The measurements were carried out in the frequency range omega /2 pi =80-460 GHz (wavelengths of 4-0.6 mm) and in the temperature range T=1.5-7K. In agreement with theory the effective masses vary quadratically with frequency and temperature.

Defining an effective relaxation time and then using the Boltzmann transport equation, analytical expressions have been derived, in the case of nearly specular scattering on external surfaces (p>or=0.5), for the Hall coefficient and conductivity in thin metallic films subjected to a transverse magnetic field. The results for moderately high magnetic field agree well with previous theoretical works; at low magnetic field the Hall coefficient in thin films is greater than the bulk value R_H0 and becomes identical with R_H0 in strong magnetic field. The theoretical predictions agree well with experimental data on copper and potassium thin films.

The authors have measured the electrical resistivity between 4.2K and 300K, and the longitudinal and transverse magnetoresistance at 4.2K, of several dilute Co and Ni based alloys containing various s, p impurities such as Sn and Sb. The deviations from Matthiessen's rule (DMR) derived from the resistivity data have been analysed in terms of the two current model of Campbell et al. (1968) and values for alpha (the ratio of spin up to spin down resistivities at the impurity site), mu (the same ratio at host sites) and D (the coefficient of the spin mixing term) estimated. The bulk of these data yield alpha approximately=1 for these impurities, as expected on the basis of a simple virtual bound state picture. The magnetoresistance data also yield values of alpha close to 1, but there are systematic differences between experiment and the predictions of the two current model.

The magnetisation of Dy_xCe_1-xRu₂ (x=0 to 1) has been measured at temperatures ranging from 1.5 to 300K. In the alloys with concentration in the range of 0.15<or=x<0.24, two critical temperatures, i.e. a superconducting transition temperature and a paramagnetic Curie temperature have been observed at low temperatures, but no ferromagnetic hysteresis has been found that the observed total magnetisation can be expressed by the sum of the superconducting magnetisation and the normal state (nonsuperconducting) magnetisation and that this expression holds whether the field is smaller than H_c1 or not. The magnitude of the initial slope of the magnetisation against magnetic field curve decreases with increasing x and is smaller than that of the perfect diamagnet. A domain structure model composed of superconducting domains and nonsuperconducting domain boundaries, is proposed to explain the present experimental results.

For pt.I see Act Crystallogr., vol.B32, p.1348 (1976). The intermetallic compound Fe₃Sn₂ has been studied by neutron diffraction and magnetisation measurements. This compound is ferromagnetic (T_C=657K). At room temperature the spin direction is along the c axis. A spin rotation is observed between approximately 250 and 60K. Below 60K the spin direction lies in the (001) plane. A consistent fit of the neutron diffraction pattern is only obtained by a modification of the magnetic form factor of Shull and Yamada (1962). This modification means that the spin carriers are delocalised in the 'Fe-Sn' planes while they are localised along the c axis. The variations of the ⁵⁷Fe anisotropic hyperfine fields (from Mossbauer effect measurements) and the mechanisms which predominantly determine the tin hyperfine fields in the series Fe₃Sn-Fe₃Sn₂-FeSn are discussed.

The room-temperature observation of extensive spin wave resonance spectra in (100) single-crystal nickel films is reported including the presence of surface spin wave modes. The variation of the spectrum with the angle between the applied magnetic field and the film normal is interpreted semiquantitatively in terms of Puszkarski's surface inhomogeneity models (1972). The value of the exchange constant D, is found to be 47+or-5*10^-30 erg cm^-2 for the applied field normal to the film surface.

The magnetic hyperfine field distributions (P(H)) of amorphous (Fe_xNi_1-x)₈₀P₁₄B₆ have been measured by ⁵⁷Fe Mossbauer spectroscopy. The saturation P(H) and their concentration dependences have been determined using a Fourier series method. The reliability and shortcomings of this method are illustrated. For all samples of (Fe_xNi_1-x)₈₀P₁₄B₆, the P(H) are well defined, single-maximum and asymmetric functions of H with a higher population at the low-field side. The P(H) shifts toward smaller H values for samples with decreasing Fe content, but the extrapolated P(H) for dilute Fe in amorphous Ni₈₀P₁₄B₆ remains finite. The shape of P(H) is found to be insensitive to the Fe content and practically temperature-independent. For amorphous Fe₈₀P₁₄B₆, the ratio of the average hyperfine field and the average magnetic moment is 150 kOe/ mu _B, which is essentially the same as those observed in a large number of crystalline Fe metalloid compounds. This suggests that the Fe hyperfine field is likely to be proportional to its moment, due to a negligible conduction electron contribution. An appreciable amount of Fe atoms in samples with x>or approximately=0.3 has been found to have hyperfine fields larger than 340 kOe, the saturation value of alpha Fe.

A theory of spin waves in a metallic spin glass is studied using a simple analytical model which describes an important aspect of an equilibrium state of the system. For a sufficiently small wavenumber k the real frequency of the spin wave is proportional to k if correlations between the spins and their interactions are taken into account. Effect of the inhomogeneity in the spin systems is examined perturbationally. It gives rise to the damping rate of the spin wave proportional to k⁴, which is shown to be quite small. The author also discusses a possible energy gap in the excitation spectrum due to the magnetic dipole interactions.

The 9.1 GHz microwave transmission against magnetic field spectrum of a high-purity tungsten single crystal coated with ferromagnetic layers is investigated. A strong transmission peak at the ferromagnetic resonance of the coating layers is attributed to longitudinal single electron spin waves (Larmor waves). The authors determined the wavelength of the spin waves from the magnetic field angle dependence of the transmission peak and by applying simple considerations the Fermi velocity around the (111) direction was derived.