A Modified Quasi-Newton Method for Structured Optimization with Partial Information on the Hessian

This paper develops a modified quasi-Newton method for structured unconstrained optimization with partial information on the Hessian, based on a better approximation to the Hessian in current search direction. The new approximation is decided by both function values and gradients at the last two iterations unlike the original one which only uses the gradients at the last two iterations. The modified method owns local and superlinear convergence. Numerical experiments show that the proposed method is encouraging comparing with the methods proposed in [4] for structured unconstrained optimization Presented at the 6th International Conference on Optimization: Techniques and Applications, Ballarat, Australia, December 9–11, 2004     

Illumination dependence of I-V and C-V characterization of Au/InSb/InP(1 0 0) Schottky structure

The effects of surface preparation and illumination on electric parameters of Au/InSb/InP(100) Schottky diode were investigated, in the later diode InSb forms a fine restructuration layer allowing to block In atoms migration to surface. In order to study the electric characteristics under illumination, we make use of an He-Ne laser of 1 mW power and 632.8 nm wavelength. The current-voltage I(V_G), the capacitance-voltage C(V_G) measurements were plotted and analysed. The saturation current I_s, the serial resistance R_s and the mean ideality factor n are, respectively, equal to 2.03 × 10⁻⁵ A, 85 Ω, 1.7 under dark and to 3.97 × 10⁻⁵ A, 67 Ω, 1.59 under illumination. The analysis of I(V_G) and C(V_G) characteristics allows us to determine the mean interfacial state density N_ss and the transmission coefficient θ_n equal, respectively, to 4.33 × 10¹² eV⁻¹ cm⁻², 4.08 × 10⁻³ under dark and 3.79 × 10¹² eV⁻¹ cm⁻² and 5.65 × 10⁻³ under illumination. The deep discrete donor levels presence in the semiconductor bulk under dark and under illumination are responsible for the non-linearity of the C⁻²(V_G) characteristic.     

The orientation-selective growth of LaNiO3 films on Si(100) by pulsed laser deposition using a MgO buffer

Highly (100)-oriented, (110)-oriented and polycrystalline LaNiO₃ (LNO) films were successfully prepared on Si(100) using an oriented MgO film as a buffer. It was somewhat surprising to find that that the orientation relation between the LNO film and the corresponding MgO buffer was: LNO(100)\MgO(110), LNO(110)\MgO(111) and LNO(polycrystalline)\MgO(100). The crystalline quality of the LNO films was shown to be sensitive to the preparation conditions of the MgO buffer. The film surface was very smooth, without micrometer-sized droplets being observed. All LNO films were of metallic conductivity, with a room-temperature resistivities of approximately 250, 280 and 420 μΩ cm for the (110)-oriented, (100)-oriented and polycrystalline LNO, respectively. Received: 2 April 2001 / Accepted: 23 October 2001 / Published online: 3 June 2002     

Nonlinear Theory of Spectra of Thermally Stimulated Currents in Complex Crystals with Hydrogen Bonds

The nonlinear theory of thermally stimulated depolarization currents is developed. The theory explains the processes of hetero- and homocharge relaxation in complex crystals with hydrogen bonds and allows the relaxation oscillator parameters to be calculated using the quadratic approximation for the external electric field.     

Stoner-Wohlfarth-type behavior of a close-packed array of high-anisotropy hexaferrite nanoparticles

Magnetization and remagnetization processes in a close-packed nanodispersed barium hexaferrite powder sample in the magnetically stable state were analyzed. Reversibility effects were discussed in terms of interparticle interaction. Judging from the magnetization curve and the parameters characterizing remagnetization irreversibility, the sample under study is a model system of small Stoner-Wohlfarth particles.     

A High Power Conversion Efficiency L-Band Erbium Doped Fiber Amplifier

By using two sections of erbium doped fiber and a fiber optical reflector, a novel, highly efficient L bandamplifier is demonstrated with significantly power-conversion-efficiency enhancement and the gain increasing of as much as 13 dB.     

Electron–hole complexes in self-assembled quantum dots in strong magnetic fields

We present results of calculations and experiments on electron–hole complexes in InGaAs/GaAs self-assembled quantum dots in high magnetic field (B). Due to hidden symmetries, the chemical potential of an N-exciton system at special B fields becomes insensitive to the exciton number as well as the magnetic field. This results in plateau regions of high intensity in measured magneto-PL spectrum. Theoretical calculations using exact diagonalization techniques successfully explain the measured magneto-photoluminescence spectrum with B fields up to 28 T.     

Remarks on exact controllability for Stokes and Navier–Stokes systems

This Note deals with the controllability of Stokes and Navier–Stokes systems with distributed controls with support in possibly small subdomains. We first present a new global Carleman inequality for the solutions to Stokes-like systems that leads to the null controllability at any time T>0. Then, we present a local result concerning exact controllability to trajectories of the Navier–Stokes system. To cite this article: E. Fernández-Cara et al., C. R. Acad. Sci. Paris, Ser. I 338 (2004).     

Comparison of two models for managing reliability growth during product design

Relying on reliability growth testing to improve system designis neither usually effective nor efficient. Instead it is importantto design in reliability. This requires models to estimate reliabilitygrowth in the design that can be used to assess whether goalreliability will be achieved within the target timescale forthe design process. Many models have been developed for analysisof reliability growth on test, but there has been much lessattention given to reliability growth in design. This paperdescribes and compares two models: one motivated by the practicalengineering process; the other by extending the reasoning ofstatistical reliability growth modelling. Both models are referencedin the recently revised edition of international standard IEC61164. However, there has been no reported evaluation of theirproperties. Therefore, this paper explores the commonalitiesand differences between these models through an assessment oftheir logic and their application to an industrial example.Recommendations are given for the use of reliability growthmodels to aid management of the design process and to informproduct development.