We have studied the ageing effects on the photoluminescence (PL) of porous silicon (PS) prepared by anodisation in HF solution of low concentration (3% ). It was found that the PL behaviour within oxidation is different from that elaborated in habitual condition. The natural oxidation of PS layers shows two opposite behaviours of the PL before stabilisation. The PL intensity decreases significantly after an initial strong and fast increase. At low temperature, the PL exhibits large band and multi-peaks structures. It was shown by electron paramagnetic resonance (EPR) and Raman measurements that a possible phase transition from crystalline to amorphous phase can occur. We show that each transformation is related to preparation conditions.

We study the conditions to perform micro-spectroscopy with a sub-wavelength lateral resolution in the wavelength spectral range from 3 to 20 μm, taking advantage of the infrared spectral signature of different chemical species. We utilised CLIO, a free electron laser, as the photon source. The transmitted photons were collected by either fluoride or chalcogenide glass fibres. Fibre tips were obtained through chemical etching by organic solvents. Metallisation of the tips permits to achieve lateral resolution of the order of the tip size. However, parasitic propagation of the light in the film reduces the contrast between irradiated and non-irradiated zones. We exemplify our set up with near-field infrared spectra obtained for polymer thin films deposited onto silicon wafers.

Stoichiometric In2S3 films were prepared by thermal evaporation technique onto clean glass substrates. According to X-ray investigations, the as-deposited films were in amorphous state. Both the ac conductivity and dielectric constants were measured in the frequency range 100 Hz−100 kHz at different temperatures. Different parameters such as the frequency exponent parameter s, the density of states near the Fermi level $N(E_{\rm F})$ , the activation energy $(\Delta E)$ and the optical band gap E g of In2S3 amorphous thin films were estimated. The hopping conduction was recognized as the conduction mechanism for the investigated films.

Polypropylene metallizedcapacitors are of general use in power electronics because of their reliability, theirself-healing capabilities, and their low price. Though the behavior of metallized coiledcapacitors has been discussed, no work has been carried out on stacked and flattenedmetallized capacitors. The purpose of this article is to suggest an analytical model ofresonance frequency, stray inductance and impedance of stacked capacitors. We first solve theequation of propagation of the magnetic potential vector (A) in the dielectric of anhomogeneous material. Then, we suggest an original method of resolution, like the one usedfor resonant cavities, in order to present an analytical solution of the problem. Finally, we give some experimental results proving that the physical knowledge of the parameters of the capacitor (dimension of the component, and material constants), enables us to calculate an analytical model of resonance frequency, stray inductance and impedance of stacked capacitors.

A fully silicon CMOS compatible high voltage (H-V) integrated circuit has been developed that features 5-V high performance digital CMOS with H-V devices. The high voltage device has to support voltage drop upper 50 V between drain and source, both for NMOS and PMOS transistors. It will be placed only in the input and output circuit, for interface application. A lateral diffusion MOS (LDMOS) structure has been chosen, for its compatibility with 5 V CMOS devices. Two specific implants are introduced into the standard process. They create the high voltage N and P junctions. Numerical simulations are performed to determine specific implant characteristics. Moreover, a two-dimensional simulator gives best LDMOS dimensions. A process control monitor has been done according to these results. After the technological realization, a quantitative electrical characterization, as maximal breakdown voltage, determines the best architecture. These devices are modeled by the MOS SPICE2G Level3, which gives sufficient results for digital applications. A simple circuit, a 5 V-50 V buffer, is simulated, realized and characterized, to conclude this work.

A two-dimensional numerical code, previously developed for the modelling of a-Si:H deposition in a SiH4/H2 radio frequency cylindrical PECVD reactor, has been modified in order to perform the modelling of CH4/H2 mixtures in the same experimental configuration. This model, which includes electrical and chemical modules, takes into account the transport and the chemistry of the charged and neutral species. The results of the models have been compared to measurements. The calculated electrical power coupled to the plasma and the self-bias voltages are compared to the experimental ones. The calculated radical densities are compared to those measured by threshold ionization mass spectroscopy at the substrate centre. Moreover, the calculated radial distribution of the deposition rate is compared to profilometry measurements. Eventually, results obtained here in CH4/H2 are compared with those previously obtained in SiH4/H2 plasmas under similar experimental conditions.

The transmission of breakdown plasma generated in water during laser shock processing experiments was investigated theoretically. A numerical model based on a rate equation formalism has been developed to calculate the characteristics (peak irradiance and duration) of the laser pulses transmitted through the breakdown plasmas generated in water. Results are in good agreement with previous experimental data for 25 ns−1064 nm laser pulses. Above 10 GW/cm2, the transmitted peak irradiance saturates and the transmission starts to decrease. The results have been extended to shorter wavelengths (532 and 355 nm). At 1064 nm, the breakdown process is dominated by avalanche ionization, whereas for shorter wavelengths, multiphoton ionization plays the major role.

A systematic parameters study is performed in a DC pulsed glow discharge in oxygen in order to improve the wettability of thin polystyrene layers. The experimental parameters considered are the pressure P, the gas flow rate Q, the interelectrode voltage V, the frequency ν of the pulsed power supply and the treatment time t t. The wettability is characterized by the contact angle technic and evaluated by the determination of the $\Delta\Theta/\Theta_{\rm i}=(\Theta_{\rm i}-\Theta_{\rm f})/\Theta_{\rm i}$ ratio ( $\Theta_{\rm i}$ and $\Theta_{\rm f}$ are respectively the initial and final contact angles). In the first part of the paper, the relative contact angle variations $\Delta\Theta/\Theta_{\rm i}$ as a function of the treatment duration time t t are investigated. In the second part, the specific energy ε and the economical criterion γare introduced to deduce the best running conditions through a chemical engineering approach. Taking into account the weak duty factor of the power supply (1% ), it is shown that the polymer surface is treated with a minimum of energy. Finally, it is deduced that the duration time necessary to reach the same $\Delta\Theta/\Theta_{\rm i}$ value is shorter for an oxygen plasma than for a nitrogen plasma.

Experiments have shown that highly intense and collimated cluster beams can be produced by a simple aerodynamic lens coupled to the nozzle of a pulsed microplasma cluster source. The mechanism of the observed cluster focusing is here presented. We discuss, as a case example, a supersonic beam of helium seeded by carbon clusters. The laminar flow of the helium-clusters mixture through a focalizing nozzle assembly has been numerically simulated and compared to the experiments. A three-dimensional steady compressible flow model has been considered for the simulation. Carbon clusters have been modeled by rigid spheres with uniform density. The trajectories of the particles are calculated during their travel through the nozzle. The simulations show that the effect of the focalizing nozzle is to divert the particles from their streamlines towards the center of the beam, thus narrowing the spatial and velocity cluster distribution. The dependence of these effects on the nozzle geometry and on the beam parameters is reproduced by the simulations in good agreement with the experimental findings.

The formation of defects during the LCM processing of composite parts depends on various injection parameters. Industrial users need to realize pieces with good physical and mechanical properties and appearance. This requires to predict what is named a “processability window”. This term defines a range of parameters which will ensure a nearly absence of defects. Knowing that most of the defects created during an injection are voids, a bibliographical background about the formation and removal of these voids is presented. An original model of void fraction prediction is developed. This model is based on an empirical analysis of void formation and of the flow behaviour. An experimental qualitativevalidation of the model is presented.The proposed model can be used as an effective prediction tool at the design stage of a composite part.