Approximating Clarke’s subgradients of semismooth functions by divided differences

We show that the algorithm presented in an earlier paper by Studniarski (Numer. Math., 55:685–693, 1989) can be applied, after only a small modification, to approximate numerically Clarke’s subgradients of semismooth functions of two variables. Results of computational testing of this modified algorithm are also reported.      

Structural and lithium intercalation studies of Mn(0.5−x)CaxTi2(PO4)3 phases (0≤x≤0.50)

Materials from the Mn_(0.5−x)Ca_xTi₂(PO₄)₃ (0≤x≤0.50) solid solution were obtained by solid-state reaction in air at 1000 °C. Selected compositions were investigated by powder X-ray diffraction analysis, ³¹P nuclear magnetic resonance (NMR) spectroscopy and electrochemical lithium intercalation. The structure of all samples determined by Rietveld analysis is of the Nasicon type with the R space group. Mn²⁺/Ca²⁺ ions occupy only the M1 sites in the Ti₂(PO₄)₃ framework. The divalent cations are ordered in one of two M1 sites, except for the Mn_0.50Ti₂(PO₄)₃ phase, where a small departure from the ideal order is observed by XRD and ³¹P MAS NMR. The electrochemical behaviour of Mn_0.50Ti₂(PO₄)₃ and Mn_(0.5−x)Ca_xTi₂(PO₄)₃ phases was characterised in Li cells. Two Li ions can be inserted without altering the Ti₂(PO₄)₃ framework. In the 0≤y≤2 range, the OCV curves of Li//Li_yMn_0.50Ti₂(PO₄)₃ cells show two main potential plateaus at 2.90 and 2.50–2.30 V. Comparison between the OCV curves of Li//Li_(1+y)Ti₂(PO₄)₃ and Li//Li_yMn_0.50Ti₂(PO₄)₃ shows that the intercalation occurs first in the unoccupied M1 site of Mn_0.50Ti₂(PO₄)₃ at 2.90 V and then, for compositions y>0.50, at the M2 site (2.50–2.30 V voltage range). The effect of calcium substitution in Mn_0.50Ti₂(PO₄)₃ on the lithium intercalation is also discussed from a structural and kinetic viewpoint. In all systems, the lithium intercalation is associated with a redistribution of the divalent cation over all M1 sites. In the case of Mn_0.50Ti₂(PO₄)₃, the stability of Mn²⁺ either in an octahedral or tetrahedral environment facilitates cationic migration.     

Polarization inscription in ferroelectric (111) PZT and (100) SBT films

Ferroelectric thin films form an equilibrium domain structure compatible with their respective crystallographic symmetry. In tetragonal (111) PZT, 90° domains prevail; in (pseudo-tetragonal) (100) SBT both 90° and 180° domains are present. The size of 90° domains has been measured for e.g., PZT as slabs of 15 nm width. Domain size is a result of stress minimization in the film during the paraelectric (PE) → ferroelectric (FE) transition. A precise and regular domain pattern for (111) PZT and (100) SBT films has been investigated in detail by TMSFM. Single domains can be addressed mechanically with the tip of an AFM. Such single domain switching corresponds to a data storage density of 200 Gbit/inch². Applications of ferroelectric and high- paraelectric materials for e.g., non-volatile data storage replacing DRAM devices or as sensors in infrared cameras are increasingly becoming popular.     

Reactivity of a Ti–45.9Al–8Nb alloy in air at 700–900°C

A Ti–45.9Al–8Nb (at%) alloy with a lamellar structure (γ+α₂) was oxidised in air at 700, 800, 850 and 900°C in isothermal and thermal cycling conditions. The reaction progress was followed by thermogravimetric measurements. In isothermal conditions the oxidation kinetics followed approximately a parabolic rate law and the rate constants ranged from about 10^–12 kg² m^–4 s^–1 at 700°C to 10^–10 kg² m^–4 s^–1 at 900°C. The oxide scales were built of Al₂O₃ and TiO₂, the former being the main component of the outermost layer. The oxidation behaviour of Ti–45.9Al–8Nb was referred to a commercial titanium alloy, WT4 (Ti–6Al–1Mn), and selected oxidation-resistant alloys.     

Convergent Algorithm Based on Progressive Regularization for Solving Pseudomonotone Variational Inequalities

In this paper, we extend the Moreau-Yosida regularization of monotone variational inequalities to the case of weakly monotone and pseudomonotone operators. With these properties, the regularized operator satisfies the pseudo-Dunn property with respect to any solution of the variational inequality problem. As a consequence, the regularized version of the auxiliary problem algorithm converges. In this case, when the operator involved in the variational inequality problem is Lipschitz continuous (a property stronger than weak monotonicity) and pseudomonotone, we prove the convergence of the progressive regularization introduced in Refs. 1, 2.     

Nonlinear electrohydrodynamic instability of a finitely conducting cylinder: Effect of interfacial surface charges

The nonlinear electrohydrodynamic stability of cylindrical interface, supporting surface charge, among two conducting fluids is investigated. The two fluids are subjected to a radial electric field. The analysis based on the multiple scale technique. It is shown that the evolution of the amplitude is governed by two partial differential equations. These equations are combined to yield two alternate Schrödinger equations with cubic nonlinearity. One of which calculates the nonlinear cutoff electric field, separating stable and unstable disturbances, while the other is used to analyze the stability of the system. The stability criteria are analytically discussed and numerically confirmed. Numerical calculations resulted in set of graphs to indicate the stability picture of the considered system.     

Robust H∞ control for linear Markovian jump systems with unknown nonlinearities

This paper studies the problem of stochastic stability and disturbance attenuation for a class of linear continuous-time uncertain systems with Markovian jumping parameters. The uncertainties are assumed to be nonlinear and state, control and external disturbance dependent. A sufficient condition is provided to solve the above problem. An H_∞ controller is designed such that the resulting closed-loop system is stochastically stable and has a disturbance attenuation γ for all admissible uncertainties. It is shown that the control law is in terms of the solutions of a set of coupled Riccati inequalities. A numerical example is included to demonstrate the potential of the proposed technique.     

Short-range order and F ion diffusion inside the Pb1 − xAlxF2 + x solid solution Part II: Al NMR investigation and proposal of clustering process

A ²⁷Al nuclear magnetic resonance (NMR) investigation at room temperature of the Pb_{1 − x}Al_xF_{2 + x} (0 ≤ x ≤ 0.12) solid solution and of the ordered Pb₉Al₂F₂₄ phase is carried out. Two different types of aluminium ions are identified and their ratio is determined as a function of x. A clustering process based on the formation of column clusters, more and more extended when x increases, is proposed on the basis of results issued from the ¹⁹F (Part I) and ²⁷Al NMR investigations.     

On Subspaces and Quotients of Banach Spaces C(K, X)

 We study the relation of to the subspaces and quotients of Banach spaces of continuous vector-valued functions , where K is an arbitrary dispersed compact set. More precisely, we prove that every infinite dimensional closed subspace of totally incomparable to X contains a copy of complemented in . This is a natural continuation of results of Cembranos-Freniche and Lotz-Peck-Porta. We also improve our result when K is homeomorphic to an interval of ordinals. Next we show that complemented subspaces (resp., quotients) of which contain no copy of are isomorphic to complemented subspaces (resp., quotients) of some finite sum of X. As a consequence, we prove that every infinite dimensional quotient of which is quotient incomparable to X, contains a complemented copy of . Finally we present some more geometric properties of spaces. Received 8 November 2000; in revised form 7 December 2001