Approximate inverse computation using Frobenius inner product

Parallel preconditioners are presented for the solution of general linear systems of equations. The computation of these preconditioners is achieved by orthogonal projections related to the Frobenius inner product. So, min_M∈??∥AM?I∥ _F and matrix M₀∈?? corresponding to this minimum (?? being any vectorial subspace of ??_n(?)) are explicitly computed using accumulative formulae in order to reduce computational cost when subspace ?? is extended to another one containing it. Every step, the computation is carried out taking advantage of the previous one, what considerably reduces the amount of work. These general results are illustrated with the subspace of matrices M such that AM is symmetric. The main application is developed for the subspace of matrices with a given sparsity pattern which may be constructed iteratively by augmenting the set of non‐zero entries in each column. Finally, the effectiveness of the sparse preconditioners is illustrated with some numerical experiments. Copyright © 2002 John Wiley & Sons, Ltd.     

Lagrangian for the t–J Model Constructed from the Generators of the Supersymmetric Hubbard Algebra

In order to describe the dynamics of the t–J model, two different families of first-order Lagrangians in terms of the generators of the Hubbard algebra are found. Such families correspond to different dynamical second-class constrained systems. The quantization is carried out by using the path-integral formalism. In this context the introduction of proper ghost fields is needed to render the model renormalizable. In each case the standard Feynman diagrammatics is obtained and the renormalized physical quantities are computed and analyzed. In the first case a nonperturbative large-N expansion is considered with the purpose of studying the generalized Hubbard model describing N-fold-degenerate correlated bands. In this case the 1/N correction to the renormalized boson propagator is computed. In the second case the perturbative Lagrangian formalism is developed and it is shown how propagators and vertices can be renormalized to each order. In particular, the renormalized ferromagnetic magnon propagator coming from our formalism is studied in details. As an example the thermal softening of the magnon frequency is computed. The antiferromagnetic case is also analyzed, and the results are confronted with previous one obtained by means of the spin-polaron theories.     

Pulse radiolysis study of the initial steps of the polymerization of cyclohexyl acrylate and cyclohexyl methacrylate

Pulse radiolysis with kinetic spectroscopic detection was applied to study the kinetics of the first steps of radiation induced polymerization of cyclohexyl acrylate and cyclohexyl methacrylate in cyclohexane solvent. The reactions were initiated by cyclohexyl radicals produced in the radiolysis of the solvent. The transient absorption spectra of the -carboxyalkyl type radicals produced in addition reaction show maxima around 300 nm. This shifts to longer wavelength with time after the pulse. This phenomenon was explained by the oligomerization reaction. From the kinetic curves average rate coefficients of termination for the oligomer radicals (2k_t) were determined as a function of time elapsed after the electron pulse. The values obtained were compared with those calculated for other (acrylate and methacrylate type) monomers.     

Phase separation in blends of two homopolymers and a diblock copolymer during film casting

Demixing during film casting of blends of polystyrene, polymethylmethacrylate, and a symmetric diblock copolymer of styrene and methylmethacrylate is discussed. The concentration fluctuations in the homogeneous solutions were calculated in mean field approximation. The structures in the homogeneous and demixed solutions and in the dry films were measured by small-angle x-ray scattering, and the morphologies of the dry films were characterized by transmission electron microscopy. The structure of the dry blends is evidently already pre-formed in solution.     

Viscoelastic properties of amorphous polymers 3: low molecular weight poly(methylphenylsiloxane)

By making creep and recoverable creep measurements of a nearly monodisperse low molecular weight poly(methyl phenyl siloxane) sample, we have found on decreasing temperature towardsT _g that there is continuously a change in the viscoelastic spectrum concomitant with a decrease of the steadystate recoverable compliance. This behavior is exactly the same as previously observed in low molecular weight poly(styrene), proving that this spectacular anomaly in the viscoelasticity of low molecular weight polymers is general and deserves an explanation. Photon correlation spectroscopic measurements performed on the same sample have extended the observation of the viscoelastic response to shorter times and the result corroborates the trend of variation established by the creep data.Dedicated to Prof.Dr. E. W. Fischer on his 65th Birthday. Prof.Dr. Fischer is known for his valuable contribution to fosterine, international collaboration of research in polymer science. This work is an example of his contribution because it would not be possible without him bringing us together. One of us (KLN) would like to take this opportunity to thank Prof. Dr. Fischer for his unwaiving support of the 1st (Crete) and the 2nd (Alicante) International Discussion Meeting on Relaxations in Complex Systems     

Correlation of modified crack tip opening distance with heat input to the heat affected zone of high-strength low-alloy steels

Investigated is the dependency of the crack tip opening displacement (CTOD) on the local microstructure of the heat affected zone in high-strength low-alloy (HSLA) microalloyed steel. Since the initiation of the crack tip location could not be controlled in fatigue, any possible correlation between heat iput in welding and fracture toughness could be smeared. Modified CTOD data are defined; they show that the fracture resistance of the weld joint decreased as the heat input increased.     

On S-Normal Spaces and Pre GS-Closed Functions

In this paper the notion of pre gs-closed functions is introduced and investigated. Presemiclosed functions are pre gs-closed but not conversely. It is shown that s-normality is preserved under continuous pre gs-closed surjections.     

Structure and stability of 0° domain walls localized at defects in a (001)-crystal plate with combined anisotropy

An analysis is made of the conditions of formation and the stability region of static-soliton-type magnetic nonuniformities in a (001)-crystal plate with combined anisotropy. It is shown that if demagnetizing fields in the plate are taken into account, static solitons can localize at certain defects for appropriate parameters of the material. The soliton stability region was found to be bounded by two extreme values of the material parameters, namely, those at which the soliton is unstable against collapse and at which it expands. It was also found that the soliton stability region differs considerably from that predicted theoretically in the model disregarding the finite size of the sample and the presence of defects. Fiz. Tverd. Tela (St. Petersburg) 40, 1498–1502 (August 1998)