Table of contents

A graphical overview of the bifurcation characteristics of the generalized Lorenz equations obtained by Stenflo (1996 Physica Scripta53 83) for nonlinear acoustic gravity waves in a rotational system is presented.

Monoenergetic unpolarised Kα-radiation is produced by filtering the Kβ radiation emitted by the secondary target in transmission and reflection modes. In the transmission mode photons with low energy and photons with energy larger than that of the discontinuity are absorbed considerably by the filter. Only photons with energy approximately in the energy interval [(E_k – 5 keV) − (E_k)] are able to cross the filter, giving rise to a peak of about 2 keV full width at half maximum. In the reflection mode the secondary target is placed at an angle of 45°. The variation of the Kα₂/Kα₁ intensity ratio at various concentrations and the homogeneity of the filters have been studied.

We analyze the consequences of the appearance of more than one resonance in the evolution equations of neutrinos interacting with a magnetic field in a scenario in which this field locally fluctuates around an equilibrium profile. We use time ordered exponential operators and the stationary phase method to solve the involved integrals to show that in this case neutrino left-right conversion probability is larger than the same probability calculated assuming only one resonance. Furthermore, quantum interference terms can substantially modify the left-right conversion probability exceeding, in certain cases, by around 100% the averaged classical probability. Concerning the solar neutrino case, such consequences of multiple resonances have to be taken into account in any attempt of solving the solar neutrino problem through a nonvanishing neutrino magnetic moment.

Stark widths and shifts for eight lines of Tl I, arising from the np ²P⁰_{1/2, 3/2} (n = 7−9) and 7s ²S_1/2 levels have been calculated using semiempirical formulas, and compared with those obtained by other authors.

Upper bounds to the electron momentum density of atomic systems are derived and expressed entirely in terms of the expectation values ⟨r²⟩, ⟨p²⟩ or ⟨p⁻²⟩ in atoms in an extremely simple manner. A short discussion is carried out.

Rigorous lower bounds for the first-gradient corrections to the kinetic and exchange-energy functionals in atoms are derived using Benson's inequality. They are formulated in terms of the expectation values and the momentum expectation value in a simple manner namely

T₂[ρ] = (1/72)∫(|∇ρ(r)|²/ρ(r)) dr ⩾ (n²/72) ⟨r^{n − 3}⟩²⟨r^{2n − 4}⟩⁻¹

and

|K₂[ρ]| = β ∫(|∇ρ(r)|²/ρ^4/3(r)) dr ⩾ 0.05105 ⟨r⁻¹⟩²⟨p⟩⁻¹,

where n, the natural number = 1, 2, 3,..., ⟨r^{n − 3}⟩ and ⟨r^{2n − 4}⟩ are the expectation values, ⟨p⟩ is the momentum value, ρ(r) is the electron density with the normalization condition ∫ ρ(r) dr = N, the number of electrons. The bounds derived in this work are tested for the atoms Z = 3 to 36. A comparison is also made with the previously derived lower bound estimates for T₂[ρ] and K₂[ρ]. The bounds presented are sharper than the previous ones given by Pathak and Gadre (atomic units are used throughout the paper).

A method of the quantitative estimation of the configuration mixing influence on the parameters of the hyperfine structure of atomic spectra is suggested. The hyperfine structure of the 3d4s and 4d5s configurations in the spectra of Sc II and Y II correspondingly is considered. The contribution of the 3d4s + 3d4d, 3d4s + 3d² and 4d5s + 4d5d, 4d5s + 4d² configuration mixing is investigated.

Transition rates are calculated for decays from the z ⁴D_J^o, z ⁴F_J^o and z ⁴G_J^o levels of neutral vanadium to the levels of the a ⁴F^e ground term. Up to 62 configurations are used to construct the wave functions in a configuration interaction calculation. In comparison with accurate laboratory measurements, the multiplets involving the ⁴F^o and ⁴G^o levels are reasonably good while the ⁴D^o decay rate values are rather poorer. Several alternative calculations for this multiplet are described and quantitative measures of the effects of different calculations and approximations are given.

The R-matrix method is used to calculate electron collision strengths for the forbidden and allowed transitions among the seven lowest states of Ni XVIII. Configuration interaction wave-functions are used to represent the target states. The standard and no-exchange R-matrix codes have been used to calculate the contribution of partial waves with L ⩽ 9 and L > 9 respectively. Collision strengths are tabulated at selected energies in the range 45–120 Ryd for all 21 transitions among the lowest seven target states. Effective collision strengths are tabulated for electron temperature in the range log T_e = 4.00 to log T_e = 6.00 K. This is the first detailed calculation on this ion in which the effects of exchange, channel couplings and short-range correlation effects are taken into account.

Effective collision strengths for electron-impact excitation of S II from the ground 3p³⁴S^o state to the higher lying 3p⁴⁴P^e, 4s ⁴P^e, 3d ⁴F^e, 3d ⁴D^e and 3d ⁴P^e excited states are presented over a wide range of temperatures of astrophysical interest (Log (T_e) = 3.5–5.1 K). The collision strengths are calculated in the close-coupling approximation using the R-matrix method and sophisticated configuration-interaction wavefunctions are used to represent the 18 target state which are included in the calculation. Convergence behaviors are examined by comparing the present 18-state work with earlier 12-state, 6-state, 5-state and 2-state calculations. Significant deviations are, however, found to occur for all five optical transitions considered, and in particular the present results indicate that the effective collision strengths in low impact energies are greatly enhanced by the presence of resonances converging to the higher lying thresholds. The accurate effective collision strengths are fitted to interpolation formulae which are valid for the range of temperatures 3200 to 100 000 degrees Kelvin. Interpolated values may readily be obtained using a Numerical Algorithm Group (NAG) library routine or by an easily coded algorithm.

Using nearly monoenergetic unpolarised Kα x-ray photons coherent (Rayleigh) and incoherent (atomic Compton) scattering cross sections in the atomic region 13 ⩽ Z ⩽ 82 have been measured at 23.18, 24.14, 25.19, 26.27 and 30.85 keV. Experimental scattering cross sections in the momentum transfer region 1.322 h Å⁻¹ ⩽ q ⩽ 1.759 h Å⁻¹ [q = sin (θ/2)λ(Å)] were obtained at an angle of 90°. Experimental coherent scattering cross sections are in good agreement with second order S-matrix calculations, compared to theoretical estimates based on form factor approximations. Experimental incoherent scattering cross sections are higher than the theoretical estimates calculated using non-relativistic incoherent scattering functions for low and medium Z elements and lower for heavy elements. The latter is due to the binding effects at low photon energies.

The usual two coupled equations for the resonance wave-particle are solved for the slide-away regime. After suggesting that the wave grows and damps like the lower hybrid and the ion sound modes, it is shown that the electron beam velocity and the wave energy density spectrum exhibits relaxational oscillations whereby the latter is interpreted as lower hybrid bursts.

L'influence du champ électromagnétique en rotation sur le tenseur diélectrique du plasma complétement ionisé est présentée. Notre expression du tenseur diffère de celle trouvée par Klimontovich pour un champ magnétique constant par la présence d'un terme nouveau d à la rotation du champ électromagnétique. Il est montré aussi que trois éléments du tenseur ont changé de positions qu'ils occupaient précédemment dans le tenseur de Klimontovich.

The second-order adiabatic invariant of particle motion in electromagnetic fields is found within the context of the quasi-particle description. Macroscopic invariants for the magnetized plasma are derived, and the constraints on stable plasma states which follow from them are discussed.

Macroscopic invariants for the plasma in a non-uniform magnetic field are derived and the constraints on stable equilibrium plasma states, which follow from such invariants are studied.

The evolution of fusion plasma temperature profiles in the presence of spatially localized or distributed sources is investigated with particular emphasis on Electron Cyclotron Heating. A central expansion technique is used, for which the time dependence of the expansion parameters (amplitude, width and shape) are found to be sensitive to the width of the temperature source and to the position of the boundary. The results obtained are in qualitative agreement with recent experiments on Electron Cyclotron heating.

Linearized Augmented Plane Wave (LAPW) calculations of the Pt and Pt₃Ti bulk electron structures are presented. We emphasis on hybridization, charge transfer effects and the impact on chemisorption properties Densities of States (DOS) and novel band-structures are also provided. The delocalized character of free electron states documents itself by sp-orbitals largely unaffected by alloy formation i.e. the Pt and Ti projected sp-densities are similar and also akin to the distribution of sp-states for pure platinum. The strong Pt-Ti interaction comes from d and f orbitals with a significantly lowered Pt d center-of-mass for the alloy and an intense Ti d structure above the Fermi level. The f-projected density of states gives two narrow peaks: a Ti peak below the Fermi level and an unoccupied Pt peak near E_f. Together the d and f interactions result in a much lowered DOS(E_f). Our results are of direct relevance for adsorption systems modelling titania supported platinum, a well known catalyst for CO hydrogenation. The ordered alloy, Pt₃Ti, forms in the industrial catalyst under reducing conditions at elevated temperatures, a phenomenon related to the Strong Metal Support Interaction (SMSI) effect in catalysis, and much work has addressed special adsorption sites for enhanced methanation rates.

We study the Josephson current through a superconducting or normal grain coupled by two tunnel junctions to macroscopic superconducting leads and capacitively to a gate. For a superconducting grain the critical Josephson current has a threshold behavior - due to the Coulomb blockade - when the superconducting gap Δ is suppressed below the charging energy E_c by a magnetic field. This effect is a direct consequence of a discrimination by the superconducting ground state between odd and even number of electrons on the grain. The critical current is periodically modulated by the gate voltage; it is 2e-periodic in the gate-induced grain charge if Δ > E_c, but tends to become e-periodic as Δ → 0. In the intermediate region the gate-modulated critical current develops a double-peak structure; the gap between peaks increases as Δ decreases. If the gap energy on the grain Δ_g is smaller than in the leads, the Josephson current is reduced, but reaches a finite, temperature dependent asymptotic value corresponding to a normal grain if Δ_g → 0.

Y-Co/Nd-Co/Y-Co and Nd-Co/Y-Co/Nd-Co amorphous sandwiches have been prepared by d.c. triode sputtering. The chemical composition of the constituent layers is R_0.33Co_0.67 (R = Y, Nd). In such systems, the Co moments are coupled parallel through the whole sandwich thickness by strong positive 3d-3d exchange interactions. But, the coercive fields of the constituent layers taken separately differ. In the sandwiches as well, the reversal of magnetization in the different layers occurs at different values of the applied magnetic field. This phenomenon has been analysed qualitatively in terms of creation and annihilation of walls at the interfaces between layers for sandwiches where the thicknesses of the constituent layers are about 1000 Å. In sandwiches where the thickness of the constituent layers is 100 Å, the wall width available is very weak and the value of the applied magnetic field necessary for the creation of such a wall is higher than the coercive field of the entire sandwich system. As a result, the magnetization of the sandwich system reverses as a whole like in homogeneous systems.

The flux phase solutions in the t-J model, the gauge theories with Chern-Simons term, promising new exciting physics, are so far obtained only within slave-boson (or fermion) approaches, where local gauge invariance is not taken into account exactly. We inspect the possibility to do this, mainly, in the example of the t-J model. The analysis of different parametrizations in the functional representation shows that (i) calculations which take into account constraints for the number of on-site-available states are describing quasiparticles in terms of the wrong local statistics, and contain gauge non-invariant objects; (ii) application of the projection technique in the slave-boson(fermion) representation reproduces the correct statistics, and is exactly equivalent to the conventional diagram technique for Hubbard operators. In the language of slave bosons this means that the Fourier transformation leading to Bloch states should be made only after accounting constraints in the measure of functional integral.