Using an Atomic Force Microscope (AFM) operated in contact mode under ambient conditions, we have tried to resolve atomically sharp cracks perpendicular to the basal plane of muscovite mica. However, although tip radius of probes of less than 10 nm allowed a lateral resolution of 0.7 nm and the tip-sample adhesive force has been minimized, our experimental results demonstrate that the molecular structure of crack tips can't still be resolved. Even worse, sometimes the tip-sample friction can disrupt the region around the crack tip. This thus leads to the conclusion that one will fail to resolve atomically sharp cracks under ambient conditions by contact AFM. The inability is due to low lateral resolution of contact AFM and strong tip-sample interaction force under ambient conditions. New techniques of scanning samples in liquids and UHV non-contact AFM both of which have true atomic resolution are strongly recommended for future studies.

In this paper, we present a numerical simulation of three metal cathode (silver, copper and nickel) submitted to a constant flux power flux density ranging between $1\times 10^{11}$ and $5\times 10^{12}~\rm W\,m^{-2}$ . The goal is to compare the interface evolution (vaporization and liquefaction rate, appearance time of liquid and vapour, energetic repartition) to predict the behaviour of the cathodes during an electric arc.

A new laboratory apparatus devoted to the characterization of various devices for the X-UV range (100–5000 eV), such as mirrors, diffraction gratings, spectrometers or detectors is described. The apparatus includes open X-ray tubes as X-ray sources, a two-crystal monochromator for wavelength selection and a goniometer. Various examples of its use are presented: dispersive mode where the radiation coming from the X-ray tube is dispersed by the two-crystal monochromator, spectrometric mode where the goniometer is used as a plane X-ray spectrometer and reflectometric mode where a selected wavelength is used to perform absolute reflectivity measurements.

The present work aims at studying mainly the wall shear stress ofa laminar steady flow of an incompressible Newtonian fluid whichis conveyed through a collapsed tube with a straight centreline.This tube is composed of a tapered segment, a contact segmentwhere the opposite walls touch and a reopening segment. The tubegeometry and steady flow characteristics are obtained frommeasurements in a collapsed tube. The Navier-Stokesequations associated with the classical boundary conditions aresolved using the finite element method. The tridimensional flowresults from the tube configuration. In particular, the flowconsists of two side-jets due to two tear-drop shaped outerpassages in the downstream contact segment associated withreversed flow. In order to compute both the stream-wise andcross-wise components of the shear stress on the wall, a localbasis is defined in each wall node. Downstream of the contactsegment, flow is separated in two jets which are studied thoughthe help of the velocity field and the wall shear stress.

Textile cardiovascular prostheses are tubular structures made of polyester filaments. They present particular mechanical properties linked to wavy form of their walls allowing them to stretch under pressure. Pulsatile blood flow was studied in a moving walls vascular prosthesis. First, an image processing device was used to measure prosthesis displacement under air pressure in an free end impregnated textile prosthesis. Then, fluid-structure interaction is simulated with a numerical computation code allowing to couple prosthesis walls motion with blood flow. Navier-Stokes equations governing fluid flow are numerically solved with N3S code based on finite elements method. The numerical process is based on the Arbitrary Lagrangian Eulerian (ALE) formulation allowing moving domains. The obtained results showed a particular distribution of blood flow velocities and shear stress near the graft walls. The flow velocity distribution near a prosthetic surface is strongly influenced by the crimping morphology and deformation. A local flow analysis is imperative to understanding pathologies implying hemodynamic factors and to optimize the prosthesis design.

This paper presents a simple and general direct modulation strategy that enables to copy directly modulated waveforms onto output voltages of a multilevel three-phase Diode Clamped Inverter (DCI). A general modelling of this converter is presented. A space vector scheme is developed without using Park transforms. Based on this algorithm, the location of the reference voltage vector is determined and the voltage vectors for the modulation are deduced. Simultaneously, their durations are calculated. The proposed algorithm is general and can be directly applied to a (n+1) levels inverter independently on its topology (Diode Clamped Inverter, Neutral Point Clamped, Flying Capacitor Inverter...). To verify this algorithm, both control algorithms of a 5-level DCI and a 11-level DCI are considered and simulation results are given.

We present a first theoretical evaluation of a new optical technique for diffraction grating metrology. While well-known spectroscopic ellipsometry (SE) is based on classical ellipsometric spectra taken in the usual planar diffraction geometry, we propose to use spectrally resolved full Mueller matrices of the gratings measured in the most general geometry of conical diffraction. We simulate the case of two superimposed one-dimensional (1D) gratings with an overlay defect, i.e. a relative shift of the two gratings with respect to each other. We show that the proposed new technique is sensitive to both the magnitude and sign of the shift, and thus it should be more efficient than usual SE for overlay characterization in real cases.

Composites studied in this work are the associations of aluminosilicates and 13% of calcium phosphates. These composites present great interest. They are destined to be applied in biomedical field, particularly in orthopedic or jawbone surgery. Calcium phosphates are composed of HA (hydroxyapatite) and TCP (tricalcic phosphate). The success of synthesised bony biomaterials depends on two determinant factors: the porosity (which facilitate the cells deposition and the vascularisation) and the compressive strength (which permits the support of body charge). In this way, a statistical experimental design was employed to quantify the influence of these two synthesis parameters. It concerns the effect of the K2O/SiO2 molecular ratio (X 1) and the effect of the calcium phosphate (HA/TCP) weight % (X 2). The K2O/SiO2 molecular ratio characterises the synthesis of the aluminosilicate. It varies between two limit levels: the stoichiometric ratio K2O/SiO2 = 0.54 corresponding to: $X_{1 }= - 1$ and the ratio K2O/SiO2 = 0.80 corresponding to $X_{1 }= 1$ . In bony biomaterials field, various calcium phosphates are commonly used as biomaterials. In our previous works, the influence of the commercial hydroxyapatite HA and tri-calcium phosphate TCP (13 wt%) addition was investigated. To study the effect of calcium phosphate composition, the weight percentage of mixing HA and TCP varied between two levels: the composite aluminosilicate with 13 wt% of HA ( $X_{2 }= -1$ ) and the composite aluminosilicate with 13 wt% of TCP ( $X_{2 }= 1$ ). Eight samples were studied. The statistical experimental design predicted answer surfaces for compressive strength and percentage of porosity. After the validation of models, it was possible to determine composite which presents best compromise between percentage of porosity and compressive strength. This composite will be evaluated by “in-vitro” and “in-vivo” studies to investigate its potential for forthcoming applied as biomaterial.

Characterization of optical gratings by resolution of inverse scattering problem has become a widely used tool. Indeed, it is known as a non-destructive, rapid and non-invasive method in opposition with microscopic characterizations. Use of a neural model is generally implemented and has shown better results by comparison with other regression methods. The neural network learns the relationship between the optical signature and the corresponding profile shape. The performance of such a non-linear regression method is usually estimated by the root mean square error calculated on a data set not involved in the training process. However, this error estimation is not very significant and tends to flatten the error in the different areas of variable space. We introduce, in this paper, the calculation of local error for each geometrical parameter representing the profile shape. For this purpose a second neural network is implemented to learn the variance of results obtained by the first one. A comparison with the root mean square error confirms a gain of local precision. Finally, the method is applied in the optical characterization of a semi-conductor grating with a 1 $\mu $ m period.