Artificial time integration

Many recent algorithmic approaches involve the construction of a differential equation model for computational purposes, typically by introducing an artificial time variable. The actual computational model involves a discretization of the now time-dependent differential system, usually employing forward Euler. The resulting dynamics of such an algorithm is then a discrete dynamics, and it is expected to be “close enough” to the dynamics of the continuous system (which is typically easier to analyze) provided that small – hence many – time steps, or iterations, are taken. Indeed, recent papers in inverse problems and image processing routinely report results requiring thousands of iterations to converge. This makes one wonder if and how the computational modeling process can be improved to better reflect the actual properties sought. In this article we elaborate on several problem instances that illustrate the above observations. Algorithms may often lend themselves to a dual interpretation, in terms of a simply discretized differential equation with artificial time and in terms of a simple optimization algorithm; such a dual interpretation can be advantageous. We show how a broader computational modeling approach may possibly lead to algorithms with improved efficiency. AMS subject classification (2000)  65L05, 65M32, 65N21, 65N22, 65D18     

Estimation of the convergence rate in a transfer theorem for centered random sums

Estimates of the convergence rate in a transfer theorem for random sums centered by constants are presented. The estimates are obtained as a consequence of the estimates of the stability of the distributions of the sums represented in the form of special mixtures to small variations in the mixing and miscible distributions.     

Existence Results for Semilinear Neutral Functional Differential Inclusions via Analytic Semigroups

In this paper we prove existence results for semilinear neutral functional differential inclusions with finite or infinite delay in Banach spaces. Our theory makes use of analytic semigroups and fractional powers of closed operators, integrated semigroups and cosine families.      

Stored energy release behaviour of disordered carbon

The use of graphite as a moderator in a low temperature thermal nuclear reactor is restricted due to accumulation of energy caused by displacement of atoms by neutrons and high energetic particles. Thermal transients may lead to a release of stored energy that may raise the temperature of the fuel clad above the design limit. Disordered carbon is thought to be an alternative choice for this purpose. Two types of disordered carbon composites, namely, CB (made up of 15 wt. % carbon black dispersed in carbonized phenolic resin) and PAN (made up of 20 vol. % chopped polyacrylonitrile carbon fibre dispersed in carbonized phenolic resin matrix) have been irradiated with 145 MeV Ne⁶⁺ ions at three fluence levels of 1.0×10¹³, 5.0×10¹³ and 1.5×10¹⁴ Ne⁶⁺/cm², respectively. The XRD patterns revealed that both the samples remained disordered even after irradiation. The maximum release of stored energy for CB was 212 J/g and that of PAN was 906 J/g. For CB, the release of stored energy was a first order reaction with activation energy of 2.79 eV and a frequency factor of 3.72×10²⁸ per second. 13% of the defects got annealed by heating up to 700 °C. PAN showed a third-order release rate with activation energy of 1.69 eV and a frequency factor of 1.77×10¹⁴ per second. 56% of the total defects got annealed by heating it up to 700 °C. CB seems to be the better choice than PAN as it showed less energy release with a slower rate. PACS 61.80.Jh; 61.80.-x; 61.43Er; 61.43.-j; 68.43.Vx     

ZnO:Mn as a member of II-VI:Mn family

ZnO:Mn thin films are grown by the metal organic chemical vapor deposition technique. Mn (x) varies in the 0<x<0.44 range. Vegard’s law has been verified for the lattice parameters. Electron paramagnetic resonance (EPR) measurements prove the substitutional incorporation of Mn²⁺ on zinc sites. The behavior of the EPR line width regarding temperature is discussed. All ZnO:Mn layers show antiferromagnetic interaction and a J₁/k_B=-15 K effective exchange constant. The optical band gap of ZnO:Mn increases with the manganese concentration. Raman spectroscopy reveals a Mn-related scattering band. PACS 71.55.Gs; 75.50.Pp; 61.10.Nz; 76.30.Fc; 75.30.Et; 78.40.-q     

A few points on omnidirectional band gaps in one-dimensional photonic crystals

The universal condition for the formation of omnidirectional band gaps (OBG) in photonic crystal (PC) was derived with consideration of permeability of the materials. And it was found that there are four kinds of PCs: one of them has no OBG, and one always possesses OBG. For the other two kinds of PCs, there are OBG for only TM or TE waves respectively. Moreover, in all PCs, the OBG can be broadened by decreasing the refractive index of the ambient medium or/and increasing the contrast between the wave impedances of the component materials of the PC. PACS 42.70.Qs; 71.20.Tx     

Giant shot-like voltage noise in high current density Nb–AlOx–Nb superconducting tunnel junctions

The flat voltage noise component of current biased, high-transparency Nb/AlO_x/Nb superconducting tunnel junctions has been investigated at frequencies up to 70 kHz. Several aspects of the analyzed phenomena suggest the presence of current noise effects induced by the discreteness of the charge carriers. At subgap voltages, where excess currents occur, a behavior coherent with a multiple Andreev reflection-assisted transport through the tunnel barrier has been found. However, the measured charge values exceed any theoretical prediction.     

Evaluation of Coulomb Energy Difference for Light Mirror Nuclei Using Slater—Type Orbitals

Behavior of the Coulomb energy difference for light nuclei is explained in terms of the different values of the average Coulomb interaction between two particles.Coulomb energy difference according to shell model of light mirror nuclei in the Coulomb and exchange integrals in the formula can be explained with exponential-type wavefunctions.In this study,using the one-center expansion of exponential-type wavefunctions in terms of Slater-type orbitals with the same center,we derived formula for Coulomb energy difference of light mirror nuclei.     

Mode beating in spot-size converter lasers

An anomalous modulation in the wavelength spectrum has been observed in lasers with spot-size converters. This intensity modulation is shown to be caused by beating between the fundamental lasing mode and radiation modes in the taper. This results in a periodic modulation in the net gain spectrum, which causes wavelength jumps between adjacent net gain maxima, and a drive current dependent spectral width that is expected to affect system performance. The amplitude of this spectral modulation is reduced significantly by either using an angled rear-facet which reflects the beating radiation modes away from the laser axis, or by using a nonlinear, adiabatic taper.     

Magnetic anisotropy of textured CrO2 thin films investigated by X-ray magnetic circular dichroism

Highly a-axis-textured CrO₂ films have been deposited on Al₂O₃ (0001) substrates by chemical vapor deposition. CrO₂ has been found to have highly a-axis (010)-oriented columnar growth on a Cr₂O₃ (0001) initial layer. The six-fold surface symmetry of the Cr₂O₃ initial layer leads to three equivalent in-plane orientations of the a-axis-oriented CrO₂ unit cell. We report Cr L_2,3 X-ray magnetic circular dichroism data along the surface normal and at 60° off-normal sample orientation. For a 60° sample alignment, a strong increase of the projected orbital moment could be observed for unoccupied majority t_2g states using moment analysis. Therefore, the c axis is identified as the intrinsic magnetic easy axis of CrO₂. In addition, a small spin moment and a very strong magnetic dipole term T_z have been found. Received: 8 January 2002 / Accepted: 8 January 2002