Table of contents

A new vector-tensor model of classical gravity, which contains coupling between the field strength of the vector field and the curvature tensors in six dimensions, is proposed. Cosmological solutions of the scale factors in this model with the compactified space are obtained. The generalization of the model is also considered.

Thick-target recoil properties of deep spallation and fragmentation products of the interaction of tantalum with 3.65 AGeV ¹²C-ions and 3.65 GeV protons have been studied. The kinematic parameters such as mean product kinetic energies and velocities of the remnant have been deduced from the data by means of the two-step vector velocity model of high-energy reactions. The results have also been used to test the applicability of the factorization hypothesis to the kinematic properties.

A method is presented that accounts for the effects of singlet-triplet mixing in measured lifetimes of resonance and intercombination lines in alkaline-earthlike and inert-gaslike isoelectronic sequences. The method uses measured lifetimes and spectroscopic energy levels as its only inputs, and yields mixing-reduced line strength factors that can be isoelectronically linearized for precision interpolation, extrapolation, and smoothing. The method is applied to predictively systematize lifetimes in the Be, Mg and Ne iso-electronic sequences.

Recent calculations of electron impact excitation rates for He-like S XV are used to derive the theoretical electron temperature and density sensitive emission line ratios G( = (f + i)/r) and R( = f/i), where f, i and r are the forbidden 1s²¹S-1s2s³S, intercombination 1s²¹S-1s2p³P_1,2 and resonance 1s²¹S-1s2p¹P transitions, respectively. Our results for R are found to be significantly different from previous calculations, due primarily to the adoption of improved electron impact excitation rates in the present paper. We estimate that the present diagnostics should be accurate to approximately ±15%.

Recent R-matrix calculations of electron excitation rates for transitions in Ti XIV are used to derive the theoretical emission line ratios R1 = I(2s²2p⁵²P_3/2-2s²2p⁵²P_1/2)/I(2s²2p⁵²P_3/2-2s2p⁶²S_1/2) = I(2117.07 Å)/(121.98 Å) and R2 = I(2s²2p⁵²P_3/2-2s²2p⁵²P_1/2)/I(2s²2p⁵²P_1/2-2s2p⁶²S_1/2) = I(2117.07 Å)/I(129.44 Å) as a function of electron density for values of N_e (=10¹⁰ − 10¹⁴ cm⁻³) applicable to astrophysical and tokamak plasmas. The results are found to be significantly different from those deduced if the distorted-wave electron rates of Bhatia et al. are adopted in the calculations, and lead to electron density estimates between a factor of 1.6 larger and 30% smaller for electron temperatures in the range T_e = 1.5 × 10⁶ to 6 × 10⁶ K.

The n = 3-4 transitions observed in the X-ray spectra of nickel-like ions (Ag XX-Pb LV) have been investigated theoretically by the multiconfigurational Dirac-Fock approach. The effect of the level crossings on the transitions rates of low excitation depopulation transitions is considered both for forbidden and for allowed transitions. A comparison of the ad initio wavelengths with the available observations for the resonance transitions shows that some experimental data must be refined.

Absolute cross sections for K-shell ionization of free metal atoms K, Ca, Rb and Sr by electrons with energies up to 45 keV have been measured with uncertainty ± 15% using a crossed-beams technique. The results are compared with other measurements for free atoms and theoretical calculations.

Transition probabilities of 20 Te I principal spectral lines have been obtained by means of the Ladenburg method (combining "branching ratios" with line absorption measurements). "Branching ratios" have been measured for line groups with common upper levels 5s²5p³6s(⁵S₂^o, ³S₁^o, ¹D₂^o, ³D₁^o, ³D₂^o, ³D₃^o). Relative probabilities for line group with common lower levels 5s²5p⁴(³P₂, ³P₁¹D₂) have been obtained from line absorption measurements. Combining the results of these measurements a common scale of relative transition probabilities for all 20 studied lines and relative lifetime values for the 6 above mentioned upper levels have been obtained. Using previously known lifetimes for levels 5s²5p³6s(³S₁^o, ⁵S₂^o), all absolute values of transition probabilities have been determined.

We report a beam-foil study of multiply ionized bromine in the region 400-1300 Å, performed with 6 and 8 MeV Br ions from a tandem accelerator. At these energies transitions belonging to Fe-like Br X and Mn-like Br XI are expected to be prominent. We have identified 31 lines as 3d⁷4s-3d⁷4p transitions in Br X, from which 16 levels of the previously unknown 3d⁷4s configuration could be established. We have also added 6 new 3d⁷4p levels to the 99 previously known. For Br XI we have classified 9 lines as 3d⁶4s-3d⁶4p combinations. The line identifications have been corroborated by isoelectronic comparisons and theoretical calculations using the superposition-of-configurations technique.

The oscillator strengths (and corresponding wavelengths) for stronger n = 3 to 2 and 3 transitions in F-like Selenium have been computed in both Coulomb gauge and length gauge using the Multi-Configurational Dirac-Fock (MCDF) code [Grant, I. P. et al., Comput. Phys. Commun. 21, 207 (1980)] and the successive subroutines developed by ourselves for calculating electric dipole transition probability rates. For most n = 3 to 2 transitions, our results from both gauges are in good agreement with those of Hagelstein's [Phys. Rev. A34, 924 (1986)], but for n = 3 to 3 transitions, only in length gauge, our results are close to his. It is shown that the discrepancies between our results from different gauges are large for n = 3 to 3 transitions.

The first part of this report deals with a new linear model of the capacitor. It is based entirely on old concepts and results that have not been combined before. The model current after a voltage step at t = 0 is proportional to t⁻ⁿ where n is a number less than but close to one. In the frequency domain losses are distributed to all frequencies. Both of these qualities adhere to real capacitors.

The major result can be expressed in the integral equation

q(t) = C₀ · u(t) + C_χ · (d^n-1u(t)/dt^n-1) t > 0

which replaces the traditional equation q(t) = C · u(t). C_χ is a constant and the time/frequency dependence lies in the fractional derivative, eq. (15).

Application of these concepts imply neglibible changes to the majority of the results obtained hitherto by conventional methods but several effects that are not explicable by these methods follow from the model. The most important of these is the memory effect and that particular discussion ends in the conclusion that dielectrics have memory. The memory involves a weight factor, t⁻ⁿ, which makes it impossible to replace the memory of past voltages with a single initial value in order to predict the future behaviour of the dielectric.

In the last part of the report it is found that the same model applies to the theory of elasticity and to many, if not all, other systems. The conclusion is that matter has memory. It appears that there is a common memory that can be influenced by many means, e.g., electric fields, mechanical stress, magnetic fields, temperature etc.

The first and second order dielectric functions are calculated for an inhomogeneous magnetized Vlasov-Poisson plasma. The expression derived for the linear dielectric function generalizes previous results by including higher order terms in the background inhomogeneity. The corresponding quadratic response function is derived in a form exhibiting the Manley-Rowe symmetry when resonant wave-particle interactions are neglected. Particularly interesting applications of these results would be the inclusion of modes which only exist in the presence of the inhomogeneity (like the various drift modes) in the nonlinear theory.

A fluid model which includes impurity ions is used to study the particle diffusion due to η_i -modes in a toroidally confined plasma. The diffusion coefficients are derived by quasi-linear theory. The diffusion of the main ion species, which in the present system is nonzero only due to the presence of impurities, may be directed inwards for a negative radial impurity ion density gradient or, under special conditions, for a flat or a positive gradient. In the latter case the impurity ion temperature must have a faster radial variation than the main ion temperature. In the typical case, the impurity concentration and charge usually tend to increase the magnitude of the particle transport in the studied region.

We consider the problem of the lifetime of a metastable atomic state in a nonequilibrium plasma. We have obtained a region of parameters, in which metastability breakdown by the Langmuir turbulence spectrum is a very important agent capable of leading to observed effects under laboratory and astrophysical conditions.

A historical review is given of the main stages of the development of neoclassic theory starting from its beginning and up to nowadays; the scientists are mentioned who contributed much in the development of the theory. Although almost none of the mathematical expositions are given in the paper, however, the physical nature of new effects and the results obtained in the cited papers are discussed in detail.

A lattice size dependence of the helicity modulus for the two-dimensional XY-model is inferred by aid of the Coulomb gas-vortex analogy and an explicit form of this dependence is obtained from a set of self-consistent equations for the Coulomb gas. High precision Monte Carlo simulations in conjunction with this inferred lattice size dependence is used to determine the critical temperature. Some other recent results in the literature are discussed in the light of the present investigation.

The effect of the second order elasticity on the critical thickness of Hybrid Aligned Nematic cells possessing either weak planar or weak homeotropic anchoring has been analysed. It is found that the critical thickness of such cells is strictly dependent on the surface elastic constant K₁₃ and on the second order bulk elastic constant K*. Moreover, an experimental method for an indirect measurement of K₁₃ and K* is proposed.

The problem of metallization of semiconductor surfaces by deposition of a layer of alkali atoms has originated a recent controversy about the ionicity of the alkali-semiconductor bond and the associated charge transfer from the alkali layer to the empty dangling bonds of the reconstructed semiconductor surface. In this work we report calculations about the 2x1 reconstructed Si(100)/K interface (symmetric dimers model) for coverages θ = 1 and 2 potassium atoms per dimer. Two new ingredients are introduced: (1) The empty band of Si dangling bonds is a narrow Hubbard band (large U/W ratio) of antibonding orbitals and (2) the low dimensionality (1D or 2D) of the Si-supported potassium band may also play an important role. Large charge transfers are hindered by the large U/W ratio. Although the electron charge sits mainly on the Si surface for small U, it tends to be distributed evenly between both the Si surface and the K overlayer as U increases. For θ = 1 the system is metallic because both components metallize. In the θ = 2 case, however, a gap which develops at the Fermi level for large U confers the interface its observed semiconducting character.

A formula of diamagnetic susceptibility of metallic thin films was obtained using a general formula for the static magnetic susceptibility. This expression is used to calculate the static diamagnetic susceptibility of Cu thin films in case of pure and decorated surfaces. The calculations are done for electrons in both of s-band and s + d bands.