Finite element method in the solution of the Euler and Navier-Stokes equations for internal flow

Finite element solutions of the Euler and Navier-Stokes equations are presented, using a simple dissipation model. The discretization is based on the weak-Galerkin weighted residual method and equal interpolation functions for all the unknowns are permitted. The nonlinearity is iterated upon using a Newton method and at each iteration the linear algebraic system is solved by a direct solver with all unknowns fully coupled. Results are presented for two-dimensional transonic inviscid flows and two- and three-dimensional incompressible viscous flows. Convergence of the algorithm is shown to be quadratic, reaching machine accuracy in very few iterations. The inviscid results demonstrate the existence of nonunique numerical solutions to the steady Euler equations.     

Substitution of calcium by strontium within selected calcium phosphates

Sr incorporation in the molecules of amorphous calcium phosphate, apatitic tricalcium phosphate, hydroxyapatite, octacalcium phosphate and dicalcium phosphate dihydrate was investigated. The concentration of Sr ranged from 225 to 1010 μ g / g, i.e. it overlapped with the physiological range of Sr concentrations in human bone. The leading experimental technique was extended X-ray absorption fine structure (EXAFS) at the Sr K edge. Results of these studies demonstrated the following: (1) Sr incorporation in the calcium phosphates is compound-dependent, (2) the coordination of incorporated Sr atoms in the Ca-P molecules is similar to that of Ca atoms, but interatomic distances are ≈0.015 nm larger, (3) in apatitic tricalcium phosphate, hydroxyapatite and octacalcium phosphate lattices Sr atoms may occupy selected Ca sites, which was not the case for dicalcium phosphate dihydrate, (4) in the apatite lattice Sr atoms are coordinated by 6 PO₄ tetrahedrals and (5) EXAFS spectra at the K edge of the incorporated Sr may be used to distinguish the structures of amorphous calcium phosphate, dicalcium phosphate dihydrate as well as apatite and its derivatives (apatitic tricalcium phosphate, octacalcium phosphate).     

X-ray induced fluorescence spectrometry at grazing incidence for quantitative surface and layer analysis

Summary Single layers and layer systems on diverse substrates were measured by Total Reflection X-Ray Fluorescence (TXRF) spectrometry. The angular dependence of the fluorescence intensities at grazing incidence allows the elemental composition, density and thickness of the layers to be evaluated using model calculations.     

A note on dephasing and pair-breaking

It is shown that the dephasing which suppresses the weak localization correction to the conductivity has the same physical origin than pair-breaking in superconductors and thus-following de Gennes-may be expressed in terms of the correlation function of the operator for time reversal.     

Theoretical and experimental investigation of the linear optical response of YBa2Cu4O8 The relevance of LDA calculations

We report results of ellipsometric measurements of the dielectric tensor of YBa₂Cu₄O₈ as well as calculations employing density functional theory in the local density approximation (LDA). For the computation of the one-electron eigenvalues and states the linear-muffin-tin-orbital method (LMTO) was used. Measured and calculated results show good agreement. In particular, the calculation predicts different band structures for YBa₂Cu₃O₇ and YBa₂Cu₄O₈ in the vicinity of the Fermi energy which lead to significant differences in the optical spectra. One such difference is a region of small ε₂ in the near infrared which is a fingerprint of the high (relative to the top of the highest fully occupied band) Fermi energy as compared to the situation in YBa₂Cu₃O₇. Our experimental results confirm the theoretical findings.     

Input control in nonnegative matrix equations

In this paper we consider the nonnegative matrix system c_iA+u_iB=c_i+1 where the nonnegative matrix A is allowed to vary, within bounds. The cone control problem is to find a nonnegative matrix B such that if C_i is a nonnegative vector in a specified cone, then there is a nonnegative vector u_i such that c_i+1 is in that cone. We extend this problem to input control by finding a B such that the cone, generated by the rows of B, is as small as possible. Thus, the percent distribution of ∣u_iB∣ through the states of the sustem by u_iB is either constant or varies little.     

Flow and reactive transport in porous media induced by well injection: Similarity solution

The authors study concentration profiles of solutes undergoingequilibrium absorption in the vicinity of a water well. Forthe case of a contamination event, the limit problem of vanishingwell radius, which is of self-similar nature, is analysed indetail. Existence, uniqueness, and qualitative properties ofsolutions of the corresponding ordinary differential equationsare shown. Some numerical examples are also presented.     

A globally convergent multi-grid algorithm for moving boundary problems of two-phase Stefan type

In this paper we deal with the numerical solution of movingboundary problems of two-phase Stefan type. Based on an implicitdiscretization in time and the use of continuous, piecewiselinear finite elements in the space variables with respect tothe weak formulation of the problem, a globally convergent multi-gridalgorithm is developed. That algorithm strongly relies on thevariational characterization of the fully discretized problemas the unconstrained minimization of a subdifferentiable convexobjective functional. Numerical results indicate a significantimprovement in efficiency compared with previous multi-gridapproaches.