Search

International audience; A flexible fiber carrying a liquid drop may coil inside the drop thereby creating a drop-on-fiber system with an ultra-extensible behaviour. During compression, the excess fiber is spooled inside the droplet and capillary forces keep the system taut, while during elongation, the fiber is gradually released and if a large number of spools is uncoiled a high stretchability is achieved. This mechanical behaviour is of interest for stretchable connectors but information, may it be electronic or photonic, usually travels through stiff functional materials and high Young's modulus, leading to large bending rigidity, prevents in-drop coiling. Here we overcome this limitation by attaching a beam of soft elastomer to the functional fiber, thereby creating a composite system which exhibits in-drop coiling and carries information while being ultra-extensible. We present a simple model to explicate the underlying mechanics of the addition of the soft beam and we show how it favors in-drop coiling. We illustrate the method with a two-centimeter long micronic PEDOT:PSS conductive fiber joined to a PVS soft beam, showing the system conveys electricity throughout a 1900% elongation.

The characterization of the biomechanical properties of newly formed bone tissue around implants is important to understand the osseointegration process. The objective of this study is to investigate the evolution of elastic properties of newly formed bone tissue as a function of healing time. To do so, nanoindentation and micro-Brillouin scattering techniques are coupled following a multimodality approach using histological analysis. Coin-shaped implants were placed in vivo at a distance of 200??m from the cortical bone surface, leading to an initially empty cavity. Two rabbits were sacrificed after 7 and 13 weeks of healing time. The histological analyses allow us to distinguish mature and newly formed bone tissue. The bone mechanical properties were measured in mature and newly formed bone tissue. Analysis of variance and Tukey?Kramer tests reveals a significant effect of healing time on the indentation modulus and ultrasonic velocities of bone tissue. The results show that bone mass density increases by 12.2% (2.2% respectively) between newly formed bone at 7 weeks (13 weeks respectively) and mature bone. The dependence of bone properties on healing time may be explained by the evolution of bone microstructure and mineralization.

Ultrafast synchrotron microtomography has been used to study in-situ and in real time the initial stages of silicate glass melt formation from crystalline granular raw materials. Significant and unexpected rearrangements of grains occur below the nominal eutectic temperature, and several drastically different solid-state reactions are observed to take place at different types of intergranular contacts. These reactions have a profound influence on the formation and the composition of the liquids produced, and control the formation of defects., Comment: 4 pages, 4 figures

International audience; Micro-Raman spectroscopy was used to map the Vickers indentation-induced densification on a window glass (soda-lime silicate glass), from the correlation between Raman spectrum and density. Both top surface and cross-section of the residual imprints were investigated. It is shown that the sample preparation needed for the investigation of the in-depth densification does not alter the results. The maximum densification ratio measured is about 3.8%, which is 60% of the maximum densification measured after a high-pressure hydrostatic loading. Within the experimental errors, densification maps are self-similar for the two indentation loads used 10 N and 20 N. A comparison with finite element results based on a constitutive model developed for silica is discussed.

International audience; In addition of a flow, plastic deformation of structural glasses (in particular amorphous silica) is characterized by a permanent densification. Raman spectroscopic estimators are shown to give a full account of the plastic behavior of silica under pressure. While the permanent densification of silica has been widely discussed in terms of amorphous-amorphous transition, from a plasticity point of view, the evolution of the residual densification with the maximum pressure of a pressure cycle can be discussed as a density hardening phenomenon. In the framework of such a mechanical aging effect, we propose that the glass structure could be labelled by the maximum pressure experienced by the glass and that the saturation of densification could be associated with the densest packing of tetrahedra only linked by their vertices.

Raman microspectroscopy was used to characterize amorphous silica plastic behavior. Using a correlation between Raman spectrum and density, a map of the local residual indentation-induced densification is obtained. The existence of a densification-induced hardening is also evidenced through a diamond anvil cell experiment. Such observations are not accounted for by the previously proposed hardening-free pressure-dependent yield criterion based on indentation curves. These results open the way to more accurate description of a constitutive law for amorphous silica.

International audience; Several chemical reactions take place during the vitrification of high-level nuclear waste. The integration of these chemical aspects in CFD codes is today one of the key points in the process of improving the prediction capabilities of numerical tools. In this study, based on simultaneous differential thermal analysis-thermogravimetry (DTA–TGA) and run/rerun method, we propose a modeling of the kinetics of chemical reactions taking place when the nuclear glass precursors are subject to constant heating rates. For the mathematical modeling, a weighted sum of n-order reactions is used to describe the overall mechanism. A hybrid approach coupling Kissinger and least squares method is performed to identify the apparent kinetic parameters. Along with the thermal characterization, we also propose a qualitative analysis of the reactions phenomenology through microstructural evolution and evolved gaz analysis based respectively on scanning electron microscope (SEM) and thermogravimetry–mass spectrometry (TGA–MS).

We analyze particle velocity fluctuations in a simulated granular system subjected to homogeneous quasistatic shearing. We show that these fluctuations share the following scaling characteristics of fluid turbulence in spite of their different physical origins: (i) scale-dependent probability distribution with non-Gaussian broadening at small time scales; (ii) spatial power spectrum of the velocity field showing a power-law decay, reflecting long-range correlations and the self-affine nature of the fluctuations; and (iii) superdiffusion of particles with respect to the mean background flow.

We propose a new method to characterize the geometrical texture of a granular packing at the particle scale including the steric hindrance effect. This method is based on the assumption of a maximum disorder (entropy) compatible both with strain-induced anisotropy of the contact network and steric exclusions. We show that the predicted statistics for the local configurations is in a fairly agreement with our numerical data., Comment: 9 pages, 5 figures

We present the results of a systematic numerical investigation of force distributions in granular packings. We find that all the main features of force transmission previously established for two-dimensional systems of hard particles hold in three-dimensional systems and for soft particles, too. In particular, the probability distribution of normal forces falls off exponentially for forces above the mean force. For forces below the mean, this distribution is either a decreasing power law when the system is far from static equilibrium, or nearly uniform at static equilibrium, in agreement with recent experiments. Moreover, we show that the forces below the mean do not contribute to the shear stress. The subnetwork of the contacts carrying a force below the mean thus plays a role similar to a fluid surrounding the solid backbone composed of the contacts carrying a force above the mean. We address the issue of the computation of contact forces in a packing at static equilibrium. We introduce a model with no local simplifying force rules, that allows for an exact computation of contact forces for given granular texture and boundary conditions. (c) 1999 American Institute of Physics.

The post-growth stress relaxation in thin polycrystalline Ag films, deposited by direct-current magnetron sputtering under different growth conditions, was explored through wafer curvature measurements. It exhibits exponential behaviour with three distinct characteristic time components τ . The slowest one, namely τ t h ≃ 200 s, is ascribed to thermal stress inherent to the deposition method. In fact, the performed temperature measurements match perfectly with an exponential stress variation with heating or cooling, as predicted by a thermal exchange model detailed in this work. Based on a comparison between different deposition conditions (continuous/interrupted sputter deposition and evaporation), the case of Mo deposition and stress relaxation modelling, the remaining components are assigned to the out-diffusion of atoms from grain boundaries ( τ g b ≃ 3 s) and to changes in the grain surface shape induced by grain boundary grooving ( τ s u r f ≃ 20 s). The relaxation amplitude of the first mechanism varies linearly with the steady-state stress at the end of growth in agreement with theoretical expectations. Yet, that of the second one does not. However, clues point to a kinetic limitation of atomic diffusion mechanism along grain boundaries. This study provides proofs of the simultaneous occurrence of several mechanisms of stress relaxation in thin metallic films.

International audience; The lifetimes of single bubbles or foams that are formed in mixtures of liquids can be several orders of magnitude larger than the ones formed in pure liquids. We recently demonstrated that this enhanced stability results from differences between bulk and interfacial concentrations in the mixture, which induce a thickness dependence of the surface tension in liquid films, and thus a stabilizing Marangoni effect. Concentration differences may be associated with nonlinear variations of surface tension with composition and we further investigate their link with foamability of binary mixtures. We show that, for asymmetric binary mixtures, i.e. made of molecules of very different sizes, strong nonlinearities in surface tension can be measured, that are associated with large foam lifetimes. When the molecules that occupy the largest surface areas have the smallest surface tension, the surface tension of the mixture varies sublinearly with composition, reflecting an enrichment in this species at the interface with air, as classically reported in the literature. In contrast, when they exhibit the largest surface tension, superlinear variations of surface tension are observed despite a similar enrichment. We discuss these variations in light of a simple thermodynamic model for ideal mixtures and we demonstrate why foam stability is enhanced for both sublinear and superlinear surface tension variations, thus shedding new light on foamability without added surfactants.

Advanced Materials Technologies 8(3), 2200922 (2022). doi:10.1002/admt.202200922, Published by Wiley, Weinheim

International audience; In this work, a structural characterization of sputtered silica films was carried out using Raman spectroscopy. Due to the low cross-section and the thinness of the silica layer, its Raman signature is dwarfed by that of the glass substrate and is therefore difficult to extract. Overcoming these limitations represents an experimental challenge and requires the development of specific analysis strategies. For this purpose, an integrated approach for extracting and interpreting the Raman signature of amorphous silica films deposited on a soda-lime glass substrate was developed, based on three distinct methods: delamination of the sputtered silica film, creating a reflective mask substrate by depositing a metallic silver coating on the glass substrate and applying a numerical signal analysis (Non-negative matrix factorization) to the multidimensional dataset acquired through depth profile acquisitions on silica films directly deposited on a glass substrate. The reliability of each proposed method is demonstrated for the extraction of the silica thin film Raman spectra. These various methods can be easily extended to other materials, either crystalline or amorphous. Furthermore, we discuss the advantages and the limits of each approach.Applying this methodology allowed us to highlight the structural differences between sputtered silica thin film and bulk vitreous silica glass (v-SiO2). Magnetron sputtering film deposition is shown to form dense silica glass layers, with an estimated densification ratio, measured by x-ray reflectivity, equal to 7%. At the medium distance range, the network connectivity change in v-SiO2 is expressed by an unusually high population of three-membered rings leading to a more compact structure. The short-range order transformation was also studied by deriving the intertetrahedral angle decrease. The present results could be a step towards advanced investigation to gain insights into the structure of films at the atomic level.

Structural properties of aqueous Na-silicates at both room and high temperature (below 500°C) have been investigated to better understand the microscopic evolution of structure and the macroscopic phenomenon of foaming. NMR spectroscopy has provided a direct quantification of structural units defined as Qn. Solid-state NMR along with Raman spectroscopy has indicated the extent of local structural reorganization on heating the silicates. Heating an aqueous Na-silicate leads to structural changes due to the removal of water, where the quantity of leftover water in the system obeys an Arrhenian evolution with an activation energy of 30 kJ.mol−1. TGA, NMR and Raman measurements suggest that the structure becomes similar to the one of hydrated silicate glasses studied in geochemistry. Increasing the Na concentration results in a larger quantity of water retained after a pre-drying treatment, which correlates with a lower softening temperature of the material and is, macroscopically, related to foaming of the silicate.

International audience; Spectroscopies are of fundamental importance but can suffer from low sensitivity. Singular Value Decomposition (SVD) is a highly interesting mathematical tool, which can be conjugated with low-rank approximation to denoise spectra andincrease sensitivity. SVD is also involved in data mining with Principal Component Analysis (PCA). In this paper, we focussed on the optimisation of SVD duration, which is a time-consuming computation. Both Intel processors (CPU) and Nvidia graphic cards (GPU) were benchmarked. A 100 times gain was achieved when combining divide and conquer algorithm, Intel Math Kernel Library (MKL), SSE3 (Streaming SIMD Extensions) hardware instructions and single precision. In such case, the CPU can outperform the GPU driven by CUDA technology. These results give a strong background to optimise SVD computation at the user scale.

International audience; The effect of temperature on multicomponent chemical diffusion in liquids of the quaternary system CaO−Na 2 O−Al 2 O 3 −SiO 2 has been studied. Diffusion-couple experiments were performed around a central composition of 64.5 wt%SiO 2 , 13.3 wt%Na 2 O, 10.8 wt%CaO, 11.4 wt%Al 2 O 3. Experiments were performed for three temperatures far above the glass transition (1200, 1280 and 1360 • C), as well as 30 • C above the glass transition. Strong multidiffusive effects were observed for all temperatures, with significant uphill diffusion of calcium, demonstrating that uphill diffusion happens close to the glass transition as well as at high temperature. For each temperature, we determined the diffusion matrix of the system and quantified its eigenvectors and eigenvalues, which correspond formally to (respectively) exchanges between ions, and associated diffusion coefficients. Little variation of the eigen-vectors of the diffusion matrix was observed as a function of temperature, with a dominant eigenvector corresponding to the exchange of sodium with calcium, the two other eigenvectors corresponding to the exchange of calcium with network formers. For the temperature range 1200-1360°C, the eigenvalues of the diffusion matrix have an Arrhe-nian temperature dependence, with an activation energy consistent with electrical conductivity for the exchange of sodium and calcium, and an activation energy consistent with viscosity for eigenvectors involving network formers. Multicomponent diffusion close to the glass transition is characterized by the same eigenvectors as at higher temperature , but some diffusion profiles are asymmetric due to strong viscosity contrasts resulting in concentration-dependent eigenvalues. Moreover, we observe some departure from Eyring relation close to the glass transition, with diffusion eigenvalues several orders of magnitude greater than the Eyring prediction. Highlights-Multicomponent diffusion matrices were determined experimentally around a central composition of 64.5 wt%SiO 2 , 13.3 wt%Na 2 O, 10.8 wt%CaO, 11.4 wt%Al 2 O 3 , for three superliquidus temperatures and 30 degrees above the glass transition.-Eigenvalues of the diffusion matrix were found to be Arrhenian down to 30 degrees above the glass transition.-Eigenvalues of exchanges involving network formers are in good agreement with Eyring's law at superliquidus temperatures, but a departure from Eyring's law is observed close to the glass transition.

International audience

International audience; Modeling the flow of dry granular materials entering water is crucial to optimize blending processes in the industry or for natural hazard assessment when describing the tsunami waves induced by landslides. In this study, we experimentally investigate the case of a jet of grains entering from the air into a water bath. After an initial transient state, a stationary impregnation front appears between the dry and the wet grains. The wet grains are then dispersed in the liquid. To describe this dry to wet transition, we focus on the first step of the process when the liquid invades the dense granular medium. In this regime, the granular jet is modeled as a translating porous material, and we systematically characterize the impregnation process using a combination of experiments, analytical modeling, and numerical tools. We then compare this approach to the situation of a confined granular jet entering into a water bath. Our approach is a first step toward describing the interplay between dry grains and a liquid and the resulting dispersion of particles.

International audience; A bidisperse medium of large sand particles and small glass beads is mixed in a cylindrical vessel that is put in rotational motion around an eccentric static rod. The subsequent segregation occurring in the system is investigated primarily at the free-surface where the large particles concentration globally increases towards a limit value while oscillating. In order to gain deeper understanding of the intrinsic mechanisms causing the segregation, we also explore the dynamics in the bulk. Average residual bulk displacements fields both vertically and orthoradially could then be measured and show a secondary flow that counteracts the segregation process.

International audience; Wet granulation processes can be driven from low to high water content. In this study, we consider the model situation of the growth of a single wet aggregate rolling in a dry granular flow inside a rotating drum. We measure the time evolution of its diameter for different grains and liquids, as well as various rotation rates of the drum. Using X-ray tomography, we are able to characterize the internal structure of the granular aggregate at different times during the process. We show that the growth rate of the aggregate can be related to the transport of the liquid inside the granule and the capture of grains. We propose a model to rationalize the maximum size of the aggregate and its growth rate.

International audience; This article focuses on the numerical study of the evaporation of sessile dropletsusing a Volume-Of-Fluid (VOF) method in the free-software Basilisk. We con-sider pure liquid droplets forming a spherical-cap onto a smooth or rough andnon-corrosive substrate and we investigate two different modes of evaporation:the unpinned mode where the contact angle is constant and the pinned modewhere the wetting area is constant. The numerical method used to implementthe contact angle and for the reconstruction of the interface is fully described,especially for the pinned mode where we propose a new VOF implementation.In the unpinned mode, we perform parametric studies and we show the influenceof the relative humidity and the contact angle on the evaporation process. Forall the explored parameters, the simulations predict that the volume decreaseslinearly with time, which matches the signature behaviour of evaporating un-pinned droplets, irrespective of the geometrical parameters. In the pinned mode,the contact angle analysis indicates a linear decrease in time which was expectedaccording to the theory and validated with some experiments we performed.

We report on the effect of intermolecular forces on the fluctuations of supported liquid films. Using an optically-induced thermal gradient, we form nanometer-thin films of wetting liquids on glass substrates, where van der Waals forces are balanced by thermocapillary forces. We show that the fluctuation dynamics of the film interface is strongly modified by intermolecular forces at lower frequencies. Data spanning three frequency decades are in excellent agreement with theoretical predictions accounting for van der Waals forces. Our results emphasize the relevance of intermolecular forces on thermal fluctuations when fluids are confined at the nanoscale., Comment: Accepted for publication in Phys. Rev. Lett

© The Author(s), 2020. A fibre withdrawn from a bath of a dilute particulate suspension exhibits different coating regimes depending on the physical properties of the fluid, the withdrawal speed, the particle sizes and the radius of the fibre. Our experiments indicate that only the liquid without particles is entrained for thin coating films. Beyond a threshold capillary number, the fibre is coated by a liquid film with entrained particles. We systematically characterize the role of the capillary number, the particle size and the fibre radius on the threshold speed for particle entrainment. We discuss the boundary between these two regimes and show that the thickness of the liquid film at the stagnation point controls the entrainment process. The radius of the fibre provides a new degree of control in capillary filtering, allowing greater control over the size of the particles entrained in the film.

This work aims at creating nanostrcutures which size can be controlled from a planar glass thin film of 100~nm deposited by reactive magnetron sputtering, thanks to phase separation phenomenon. The model ternary system used is a barium borosilicate because its immicibility domain is large. After demixing the barium rich phase has strong electronic contrast favorable for SEM imaging. Moreover this less polymerized phase can be selectively dissolved to reveal nanostructures. Three different morphologies are finally obtained varying thin films composition : holes, pillars or interconnected roughness. Image processing from SEM and AFM measurements allows quantitative measurements of characteristic lengths such as vertical and horizontal size and lateral correlations. Coarsening kinetics in ultra-confined media are studied for both mechanisms nucleation and growth and spinodal decomposition. Particularly, for droplets assembly with only one object into the thickness, confinement slows down diffusion. This impact is shown experimentally and numerically, it gives a power law slower than classical models. For 2D interconnected network, hydrodynamic growth is quickly interrupted by fragmentation and destabilises the free surface leading to rough samples.; L'objectif de ces travaux est d'obtenir des nanostrcutures de tailles contrôlées en partant d'une couche de verre plane d'environ 100~nm, déposée par pulvérisation cathodique magnétron, ce grâce au phénomène de séparation de phase. Le système modèle utilisé est un borosilicate de baryum dont le ternaire possède une lacune d'immiscibilité large. Après démixtion, la phase riche en baryum possède d'une part un très bon contraste électronique pour des études MEB et d'autre part elle peut être dissoute préférentiellement afin de révéler des nanostrcutures. Trois types de morphologies ont pu être obtenues en maîtrisant la composition des couches minces : des trous, des plots ou une rugosité interconnectée. Le traitement d'images MEB et AFM permet de suivre quantitativement des grandeurs caractéristiques (tailles des objets et corrélations latérales). Ainsi la cinétique de mûrissement en milieu ultra-confiné est étudiée à la fois pour la nucléation-croissance et la décomposition spinodale. En particulier pour une assemblée de gouttes uniques dans l'épaisseur, le confinement ralenti la diffusion. Une loi de puissance plus lente que les modèles classiques est observée expérimentalement et numériquement. Pour un réseau 2D de filaments interconnectés, le mûrissement hydrodynamique est rapidement interrompu par fragmentation et déstabilise la surface engendrant directement une rugosité.

International audience; In this article, we review both theoretical models and experimental results on the motion of micro-and nano-particles that are close to a fluid interface or move in between two fluids. Viscous drags together with dissipations due to fluctuations of the fluid interface and its physicochemical properties affect strongly the translational and rotational drags of colloidal particles, which are subjected to Brownian motion in thermal equilibrium. Even if many theoretical and experimental investigations have been carried out, additional scientific efforts in hydrodynamics, statistical physics, wetting and colloid science are still needed to explain unexpected experimental results and to measure particle motion in time and space scales, which are not accessible so far.

The first chapter presents the basics of plasmonics. The second chapter presents Bedeaux and Vliegler's seductive approach and the notion of excess field s, which allows to define the modified Fresnel coefficients. Chapter 3 presents polarizability calculations for supported nanoparticles, possibly truncated and possibly with a core/shell structure. This involves solving Laplace's equation with boundary conditions imposed by the presence of the substrate and by the geometry of the particle. One of the merits of this thesis is to have been able to synthesize the works of Bedeaux and Vliegler and their formalism. Chapter 4 presents the Granfilm software developed by Rémi Lazzari and Ingve Simonsen, supervisors of Sunniva Indrehus' thesis. Extensions to the proposed models and the further development of a Python-interface (named GranFilmPy) so that the software is more widely used by the community. The examples of simulations obtained by Granfilm (section 4.4) give a good idea of the possibilities of the software. Chapter 5 presents an original way to study the resonance modes in the case of truncated, possibly coated, particles. This consists in arbitrarily strongly reducing the imaginary part of the dielectric function of the metal. The variations of the different resonance modes can then be analyzed very finely. The results are in the form of spectral curves, contour plots of the electrostatic potential in the xz plane, or false color maps (energy in abscissa and geometric parameter in ordinate). The influence of the truncation rate is shown in Figs. 5.10 - 5.12. Chapter 6 considers the polydispersivity of the size and shape of nanoparticles. As in the rest of the manuscript. It is distinguished between the "low coverage regime" and the "finite coverage regime" where the interactions between particles must be taken into account. It also presents the non-negligible influence of the radial particle distribution function (RDF) on the plasmonic response. Implemented in the Granfilm software, taking polydispersivity into account will allow a better interpretation of the widths of resonance peaks., Après quelques rappels d’électromagnétisme et sur la permittivité des métaux, le premier chapitre est dévolu à la présentation des nanoparticules en optique, et comment les modèles analytiques de polarisabilité permettent de prendre en compte la présence d’un substrat et de nanoparticules voisines. Les deux chapitres suivants reprend un certain nombre d’éléments de la théorie de Bedeaux-Vlieger sur la charge excédentaire et les susceptibilités de surface, et sa mise en œuvre pour les nanoparticules sur un substrat plan. Cette théorie, dont les bases ont été établies à la fin des années 70, a depuis connu plusieurs développements spécifiques en intégrant des particules sphériques puis sphéroïdales, puis sphériques tronquées, puis sphéroïdales tronquées, en interaction électrostatique ou non avec leurs voisines mono- disperses. La théorie de Bedeaux-Vlieger et leurs continuateurs permet un calcul particulièrement efficace des propriétés optiques des films métalliques mesurables par ellipsométrie. Rémi Lazzari et Ingve Simonsen, les deux directeurs de thèse de Mme Indrehus, ont ainsi démarré le développement du logiciel GranFilm au début des années 2000. Ce logiciel est aujourd’hui disponible gratuitement sur internet, sous licence GNU. Le chapitre quatre détaille l’organisation de ce logiciel et les différentes librairies numériques sur lesquelles il a été développé. Un guide d’utilisation du logiciel, avec notamment un paragraphe sur les limites du logiciel ; il y est surtout question des approximations du modèle physique. Sont ensuite documentés les apports au logiciel dus à Mme Inderhus : une interface en langage Python, la prise en compte de la taille des nanoparticules pour la correction de leur permittivité, et l’intégration du modèle dans une boucle d’optimisation pour estimer des paramètres physiques à partir de données expérimentales. Le chapitre cinq fait une étude numérique des modes de plusieurs des systèmes plasmoniques élémentaires modélisables à l’aide du logiciel GranFilm. Pour exalter ces modes, elle est amenée à réduire la partie imaginaire de la permittivité des métaux à une fraction inférieure ou égale à 1 % de leur valeur réelle. Ces modes sont représentés par leur carte de potentiel électrostatique. Sont étudiées numériquement l’influence du rapport de troncature pour les nanoparticules déposées sur substrat et l’influence de l’épaisseur de la couche sur les nanoparticules recouvertes. Cette analyse recoupe des résultats expérimentaux de la littérature sur des nanoparticules non recouvertes, et permet de les étendre au cas des nanoparticules recouvertes. GranFilm apparaît ainsi comme un outil capable de prendre en compte des géométries nanoplasmoniques qui sont d’ors et déjà réalisables expérimentalement, et d’en faire l’analyse des propriétés optiques. Les mesures optiques spectrales sur films métalliques donnent des pics de résonance plasmon qui ont tendance à être significativement plus larges que prévu par les premières modélisations. Le chapitre six étudie ce phénomène d’élargissement, et travaille à sa modélisation. La littérature fait état de deux sources d’élargissement : la première, dite intrinsèque, provient de deux effets de taille finie sur la permittivité des nanoparticules, et la seconde, extrinsèque, est associée à la dispersion des nanoparticules en taille et forme. La modélisation de l’élargissement intrinsèque ayant été traitée au chapitre trois, les efforts sont ici concentrés sur l’élargissement extrinsèque. Tant que les nanoparticules ne sont pas en interaction électrostatique, le modèle physique peut être conservé, et la dispersion des nanoparticules prise en compte de manière purement statistique. Ce régime est nommé LCP dans le manuscrit. La mise en œuvre numérique de ce modèle statistique, et les élargissements qui en découlent, sont très détaillés.

International audience; We report a detailed structural investigation of Ca-Na aluminosilicate glasses containing 0-20 wt% ZrO2 using X-ray absorption spectroscopy, Raman spectroscopy and 27 Al and 29 Si NMR. The X-ray absorption spectroscopy data reveal that Zr is predominantly present in ZrO6 octahedra but a significant Zr coordination change occurs with increasing ZrO2 content. Though Al and Zr are competing to be surrounded by charge balancing cations, 27 Al NMR data do not show the presence of high-coordinated Al species; thus charge-balancing cations are preferentially localized close to AlO4 tetrahedra rather than Zr polyhedra, explaining the increase in Zr coordination number that ultimately leads to limit Zr solubility in glasses. Raman and 29 Si NMR data indicates that Zr atoms are linked to the silicate network with a trend for higher degree of network connectivity at high zirconia content. Change in network connectivity and in Zr environment provide insights to understand chemical durability or crystallization in glasses.

Author(s): Raux, Pascal S; Troger, Anthony; Jop, Pierre; Sauret, Alban | Abstract: The fragmentation of liquid sheets produces a collection of droplets. The size distribution of the droplets has a considerable impact on the coating efficiency of sprays and the transport of contaminants. Although many processes commonly used particulate suspensions, the influence of the particles on the spreading dynamics of the sheet and its subsequent fragmentation has so far been considered negligible. In this paper, we consider experimentally a transient suspension sheet that expands radially. We characterize the influence of the particles on the dynamics of the liquid sheet and the fragmentation process. We highlight that the presence of particles modifies the thickness and reduces the stability of the liquid sheet. Our study suggests that particles can significantly modify the dynamics of liquid films through capillary effects, even for volume fractions much smaller than the maximum packing.

International audience; High temperature cation mobilities, based on the multicomponent diffusion method, have been coupled with a structural investigation in the Na2O-CaO-Al2O3-SiO2-ZrO2 (NCASZ) system at 1200°C and 1250°C. From structural investigation using X-ray absorption spectroscopy (XANES), zirconium was determined in six-fold coordinated sites in the reduced glass composition range of our study and no significant structural changes have been evidenced compared to a quaternary Na2O-CaO-Al2O3-SiO2 (NCAS) glass with a similar composition. Additionally, the diffusion experiments revealed that adding zirconium in the quaternary melt has no influence on the dominant eigenvectors that correspond to exchange between calcium and sodium. Despite the expected mobility decrease resulting from a viscosity increase upon the addition of zirconium, the dominant eigenvalue calculated in this study is higher when Zr is introduced in the melt structure. This result strongly suggests that the sodium mobility is enhanced by the presence of Zr, which is explained by a change in the structural role of sodium from network modifier associated to Q3(Si) sites in NCAS melts to charge compensator associated with [ZrO6] 2-sites in NCASZ melts. This modification generates lower bond strengths between sodium and other cations in the melt, thus favoring an enhancement in mobility. Moreover, the diffusion matrix was applied to predict diffusion profiles between zirconium-bearing crystals and melts. We observed that even far from the composition field used for the matrix determination, predictions of zirconium diffusion profiles may be relevant and demonstrate the potential of this approach to evaluate crystal/melt dissolution behavior.; Les mobilités de cation à haute température, basées sur la méthode de diffusion multicomposante, ont été couplées à une étude structurale dans le système Na2O-CaO-Al2O3-SiO2-ZrO2 (NCASZ) à 1200°C et 1250°C. À partir d’une étude structurale utilisant la spectroscopie d’absorption des rayons X (XANES), le zirconium a été déterminé dans des sites de ccoordinence 6 dans la gamme de composition réduite de notre étude et aucun changement structurel significatif n’a été démontré par rapport à un verre quaternaire Na2O-CaO-Al2O3-SiO2 (NCAS) avec une composition similaire. En outre, les expériences de diffusion ont révélées que l’ajout de zirconium dans la fonte quaternaire n’a aucune influence sur les vecteurs propres dominants qui correspondent à l’échange entre le calcium et le sodium. Malgré la diminution prévue de la mobilité résultant d’une augmentation de la viscosité lors de l’ajout de zirconium, la valeur propre dominante calculée dans cette étude est plus élevée lorsque Zr est introduit dans la structure du liquide. Ce résultat suggère fortement que la mobilité en sodium est renforcée par la présence de Zr, ce qui s’explique par un changement dans le rôle structurel du sodium depuis un rôle de modificateur de réseau associé aux sites Q3(Si) dans les liquides NCAS vers un rôle de compensateur de charge associé aux sites [ZrO6]2- dans les liquides NCASZ. Cette modification génère des forces de liaison plus faibles entre le sodium et d’autres cations dans le liquide, favorisant ainsi une amélioration de la mobilité. En outre, la matrice de diffusion a été appliquée pour prédire les profils de diffusion entre les cristaux porteurs de zirconium et les liquides. Nous avons observé que même loin du champ de composition utilisé pour la détermination de la matrice de diffusion, les prédictions des profils de diffusion du zirconium peuvent être pertinentes et démontrer le potentiel de cette approche pour évaluer le comportement de dissolution à l'interface cristal/liquide.

International audience; High temperature cation mobilities, based on the multicomponent diffusion method, have been coupled with a structural investigation in the Na2O-CaO-Al2O3-SiO2-ZrO2 (NCASZ) system at 1200°C and 1250°C. From structural investigation using X-ray absorption spectroscopy (XANES), zirconium was determined in six-fold coordinated sites in the reduced glass composition range of our study and no significant structural changes have been evidenced compared to a quaternary Na2O-CaO-Al2O3-SiO2 (NCAS) glass with a similar composition. Additionally, the diffusion experiments revealed that adding zirconium in the quaternary melt has no influence on the dominant eigenvectors that correspond to exchange between calcium and sodium. Despite the expected mobility decrease resulting from a viscosity increase upon the addition of zirconium, the dominant eigenvalue calculated in this study is higher when Zr is introduced in the melt structure. This result strongly suggests that the sodium mobility is enhanced by the presence of Zr, which is explained by a change in the structural role of sodium from network modifier associated to Q3(Si) sites in NCAS melts to charge compensator associated with [ZrO6] 2-sites in NCASZ melts. This modification generates lower bond strengths between sodium and other cations in the melt, thus favoring an enhancement in mobility. Moreover, the diffusion matrix was applied to predict diffusion profiles between zirconium-bearing crystals and melts. We observed that even far from the composition field used for the matrix determination, predictions of zirconium diffusion profiles may be relevant and demonstrate the potential of this approach to evaluate crystal/melt dissolution behavior.; Les mobilités de cation à haute température, basées sur la méthode de diffusion multicomposante, ont été couplées à une étude structurale dans le système Na2O-CaO-Al2O3-SiO2-ZrO2 (NCASZ) à 1200°C et 1250°C. À partir d’une étude structurale utilisant la spectroscopie d’absorption des rayons X (XANES), le zirconium a été déterminé dans des sites de ccoordinence 6 dans la gamme de composition réduite de notre étude et aucun changement structurel significatif n’a été démontré par rapport à un verre quaternaire Na2O-CaO-Al2O3-SiO2 (NCAS) avec une composition similaire. En outre, les expériences de diffusion ont révélées que l’ajout de zirconium dans la fonte quaternaire n’a aucune influence sur les vecteurs propres dominants qui correspondent à l’échange entre le calcium et le sodium. Malgré la diminution prévue de la mobilité résultant d’une augmentation de la viscosité lors de l’ajout de zirconium, la valeur propre dominante calculée dans cette étude est plus élevée lorsque Zr est introduit dans la structure du liquide. Ce résultat suggère fortement que la mobilité en sodium est renforcée par la présence de Zr, ce qui s’explique par un changement dans le rôle structurel du sodium depuis un rôle de modificateur de réseau associé aux sites Q3(Si) dans les liquides NCAS vers un rôle de compensateur de charge associé aux sites [ZrO6]2- dans les liquides NCASZ. Cette modification génère des forces de liaison plus faibles entre le sodium et d’autres cations dans le liquide, favorisant ainsi une amélioration de la mobilité. En outre, la matrice de diffusion a été appliquée pour prédire les profils de diffusion entre les cristaux porteurs de zirconium et les liquides. Nous avons observé que même loin du champ de composition utilisé pour la détermination de la matrice de diffusion, les prédictions des profils de diffusion du zirconium peuvent être pertinentes et démontrer le potentiel de cette approche pour évaluer le comportement de dissolution à l'interface cristal/liquide.

International audience

Under suitable conditions, an immersed granular bed can be destabilized by local thermal forcing and the induced buoyant force. The destabilization is evident from the triggering and establishment of a dense fluid-like granular plume. Varying the initial granular layer average height $h$, a time series of the free layer surface is extracted, allowing us to dynamically compute the underlying volume of the granular layer. Different observed phenomena, namely the initial interface deformation, the lowering of the average granular interface (i.e. decrease of the granular layer volume) and the emission of a plume, are analysed. We show that the phenomenon is mainly driven by heat transfer, for large $h$ and also involves a variable height thermal boundary condition and Darcy flow triggering, for small $h$. Simple modelling with no adjustable parameters not only allows us to capture the observed scaling power laws but is also in quantitative agreement with the obtained experimental data.

International audience; Lightweight plasterboard is a product composed of plaster foam core whose porosity can reach 75 vol%, lined with two sheets of paper. To optimize the trade-o between thermal resistance and mechanical strength, it is important to understand and characterize the mechanical behavior of plasterboard. Core compaction of plasterboard is one of the prevalent degradation mechanism. This mechanism is studied herein using in-situ indentation experiments performed at two dierent scales. The rst test was carried out in a lab tomograph. The second test was conducted on the high resolution synchrotron ID19 beamline at ESRF. Very early damage is revealed. The basic mechanism of pore collapse and its gradual development is detected in contrast with mesoscale observations where compaction appears very abrupt.

Coating processes allow functionalizing a surface to obtain new properties, as anti-glare or anti-scratch. Amongst the various methods, wet coating, the process of spreading then drying a liquid layer containing a material of interest, is particularly appealing because of its efficiency and low cost. One key hurdle, however, is the possible apparition of defects in the film during the drying process, notably, thickness variations over large areas. These imperfections degrade the aspect and the optical properties of the surface. In this Ph.D. thesis, we shed light on the apparition of these defects, by studying the flows triggered by the evaporation in liquid films of binary mixtures. Indeed, the evaporation of the solvent can induce variations of composition which generate in return gradients of tension at the film surface which destabilize it. The combination of model experiments, theoretical modeling and numerical simulations allowed us to reveal and describe quantitatively several regimes, which correspond to the factor limiting the instability: gravity, Laplace pressure, lateral homogenization by diffusion of the compounds or by their vertical stratification due to the evaporation. An independent study has been lead on the generation of bubble trains when air is slowly injected in a bath.; Revêtir une surface d’un film mince permet de lui conférer de nouvelles propriétés, comme en réduire les reflets ou améliorer sa résistance aux rayures. Une méthode pour produire ces revêtements est le dépôt par voie liquide : elle consiste à couvrir la surface avec un matériau dispersé dans un liquide puis à le sécher. Cette méthode permet de fonctionnaliser efficacement et rapidement de grandes surfaces. Bien qu'attrayante pour l’industrie verrière, la perspective d'une fonctionnalisation par voie liquide se heurte actuellement à l'apparition de défauts dans le film lors de son séchage, notamment des variations d'épaisseurs sur de grandes distances. Ces imperfections détériorent l'esthétique et les propriétés optiques de la surface. Dans cette thèse, nous apportons un éclairage sur l’apparition de ces défauts, en étudiant les écoulements générés lors du séchage de films liquides de mélanges binaires. En effet, l'évaporation du solvant peut induire des variations de composition qui génèrent en retour des gradients de tension à la surface du film qui le déstabilisent. La combinaison d'expériences modèles, de modélisations théoriques et de simulations numériques ont permis de mettre en évidence et de décrire quantitativement plusieurs régimes, qui correspondent aux différents effets limitant l’instabilité : la pesanteur, la pression de Laplace, l’homogénéisation latérale par diffusion des composées dans le film ou au contraire la stratification verticale de ces composés, causée par l’évaporation. Une étude indépendante a été menée sur la génération de trains de bulles lorsque de l’air est lentement injecté dans un bain.

Coating processes allow functionalizing a surface to obtain new properties, as anti-glare or anti-scratch. Amongst the various methods, wet coating, the process of spreading then drying a liquid layer containing a material of interest, is particularly appealing because of its efficiency and low cost. One key hurdle, however, is the possible apparition of defects in the film during the drying process, notably, thickness variations over large areas. These imperfections degrade the aspect and the optical properties of the surface. In this Ph.D. thesis, we shed light on the apparition of these defects, by studying the flows triggered by the evaporation in liquid films of binary mixtures. Indeed, the evaporation of the solvent can induce variations of composition which generate in return gradients of tension at the film surface which destabilize it. The combination of model experiments, theoretical modeling and numerical simulations allowed us to reveal and describe quantitatively several regimes, which correspond to the factor limiting the instability: gravity, Laplace pressure, lateral homogenization by diffusion of the compounds or by their vertical stratification due to the evaporation. An independent study has been lead on the generation of bubble trains when air is slowly injected in a bath.; Revêtir une surface d’un film mince permet de lui conférer de nouvelles propriétés, comme en réduire les reflets ou améliorer sa résistance aux rayures. Une méthode pour produire ces revêtements est le dépôt par voie liquide : elle consiste à couvrir la surface avec un matériau dispersé dans un liquide puis à le sécher. Cette méthode permet de fonctionnaliser efficacement et rapidement de grandes surfaces. Bien qu'attrayante pour l’industrie verrière, la perspective d'une fonctionnalisation par voie liquide se heurte actuellement à l'apparition de défauts dans le film lors de son séchage, notamment des variations d'épaisseurs sur de grandes distances. Ces imperfections détériorent l'esthétique et les propriétés optiques de la surface. Dans cette thèse, nous apportons un éclairage sur l’apparition de ces défauts, en étudiant les écoulements générés lors du séchage de films liquides de mélanges binaires. En effet, l'évaporation du solvant peut induire des variations de composition qui génèrent en retour des gradients de tension à la surface du film qui le déstabilisent. La combinaison d'expériences modèles, de modélisations théoriques et de simulations numériques ont permis de mettre en évidence et de décrire quantitativement plusieurs régimes, qui correspondent aux différents effets limitant l’instabilité : la pesanteur, la pression de Laplace, l’homogénéisation latérale par diffusion des composées dans le film ou au contraire la stratification verticale de ces composés, causée par l’évaporation. Une étude indépendante a été menée sur la génération de trains de bulles lorsque de l’air est lentement injecté dans un bain.

Thin brittle films on compliant substrates are used in many applications, as soft electronics and solar cells. When submitted to large tensile strains, those systems undergo multi-cracking. A saturation of the cracks pattern is observed, i.e. no new crack is formed above a given nominal applied strain. Moreover, a characteristic distance between the cracks is observed at saturation. A mechanical analysis is carried out in this paper in order to quantitatively predict the saturation phenomenon. Fracture in the brittle layer, delamination at the interface and plasticity in the substrate are taken into account. The results of finite elements simulations show that both the plastic deformation pattern inside the substrate and the strength of the brittle layer are key elements for predicting the cracks pattern at saturation.

International audience

International audience; The dynamics of soda‐lime‐silica glass grain melting is investigated experimentally using a nonintrusive technique. A cylindrical alumina crucible is filled with glass cullet and placed into a furnace illuminated by an X‐ray source. This glass granular bed is gradually heated up to 1100°C, leading to its melting and the generation of a size‐distributed population of bubbles rising in the molten glass. An image processing algorithm of X‐ray images of the cullet bed during melting allows the characterization of bubbles size distribution in the crucible as well as their velocity. The introduction of tin dioxide μ‐particles in the glass matrix before melting enhances the texture of the images and makes possible the determination of the bubble‐induced molten glass velocity field by an optical flow technique. The bubble size distribution can be fitted by a log‐normal law, suggesting that it is closely related to the initial size distribution in the cullet bed. The liquid motion induced by the bubbles in Stokes' regime is strongly affected by the flow confinement and the determination of bubble rising velocity along its trajectory unveils the existence of local tiny temperature fluctuations in the crucible. Overall, the measuring techniques developed in this work seem to be very promising for the improvement of models and optimization of industrial glass furnaces.

International audience; We consider the rearrangements of a vertical two-dimensional granular packing induced by the withdrawal of an intruder. Here, we focus on the influence of the size of the intruder on the reorganization process. The long term evolution of the granular packing is investigated as well as the avalanche dynamics that characterize the short term rearrangements around the intruder. For small enough intruder, we observe the formation of arches that periodically destabilize and influence the reorganization dynamics of the twodimensional packing through larger rearrangement events.

International audience; Foam-based solid coatings appear to be a simple solution for giving new properties to solid surfaces. An efficient method is presented for producing open-cell foam coatings having tunable pore radius distribution (i.e. monodisperse within the range 100-1000 µm, bidisperse or fully polydisperse), tunable thickness, and tunable bulk and surface porosities. This is achieved by mixing precursor aqueous foam and particle suspension (here a micrometer-sized polyurethane dispersion), and by spreading with a nozzle the resulting particle-loaded foam on the solid surface to be coated. It is shown that bubble size distribution of the precursor foam can be preserved in the final solid coating. This this highlighted by using monodisperse aqueous foam, for which coatings showed polycrystalline structure, as well as bidisperse or fully polydisperse foams. As a major advantage of our method, bubble size and solid volume fraction are shown to be independent parameters allowing for the size of the microstructural elements to be tuned easily, so are the expected functional properties of the coating. Results obtained with the studied polyurethane dispersion are expected to be reproduced with other dispersions.

International audience; Among the large variety of microfluidic platforms, surface devices are a world apart. Electrowetting systems are used to control the displacement of droplets among predetermined pathways. More confidential, superhydrophobic surfaces are more and more described as new elements to guide spherical droplet reactors. As such, they can exhibit confinement properties analogous to channel-based microfluidics. In this article, we describe a new strategy to use superhydrophobic surfaces as a permanently tilted microfluidic platform, on which droplets containing iron oxide nanoparticles are guided with permanent magnets. These droplets are fed with water through a capillary tube until their weight exceeds the magnetic field force. Thus, the volume at which the droplet rolls off the surface is only governed by the initial quantity of magnetic nanoparticles and the tilting angle of the surface. This phenomenon provides a strategy for droplet dilution in a simple and reproducible manner, which is not that easy in microchannels, and a key advantage of open systems. As a proof of concept, we used this platform to prepare magnetic filaments by a salting-out process already described in large batches. By reducing salt concentration on the platform, we are able to control the electrostatic attractive interactions between iron oxide nanoparticles coated with poly(acrylic acid) and a positively charged polyelectrolyte [poly(diallyldimethylammonium chloride)]. The formation of nanostructured filaments was conducted in 2 min while more than 30 min was required for dialysis. Our results also illustrate the power of microfluidic reaction processes because such magnetic filaments could not be obtained through direct batch dilution because of mixing issues. Such microfluidic platforms could be useful for the efficient and simple dilution of systems where reactivity is controlled by concentration.

International audience; Gypsum-based plasters or stuccoes, in spite of their importance and diffusion, received little attention in cultural heritage materials studies. This work introduces a new, non-destructive methodology, using micro-tomography to measure the water/plaster ratio and the morphology of the hemihydrate powder used to make plasters on < 1 mm3 samples. This methodology give insight in both the raw material (and ultimately provenance) and the technique used to make plaster. The methodology was tested first on mock-up samples of known composition, then in a case study on 13 low-relief cast plaster sculptures from 15th century Florentine artists. Preliminary conclusions on this limited corpus show relative uniformity across most reliefs in terms of raw materials and techniques. The casts of one model (Nativity, attributed to Donatello and B. Bellano) were made with a different raw material, in line with prior geochemical analyses; these results support the previous attribution to a North Italian rather than Florentine origin. The casts of a second model (Virgin and Child, type of Saint Petersburg, attributed to Antonio Rossellino) were prepared with a different technique. This surprising result was not expected from Art History or previous studies.

International audience; Periodic arrays of anisotropic silver nanoparticles having peculiar optical properties are fabricated at a macroscopic scale. The proposed scalable method is based on temperature‐assisted solid‐state dewetting of a continuous thin layer deposited on a silica substrate patterned by the nanoimprint technique. The resulting nanoparticles are shaped like diamonds and are half‐embedded into the patterned silica. A period‐dependent optimum in film thickness for the quality of spatial organization is found and discussed in terms of thermodynamics and, for the first time, in terms of the role of grains in the dewetting process. The optical properties of the arrays are driven by not only simply the particle shape but also the lattice period and the degree of order. A surface lattice resonance that disperses with the underlying period is evidenced experimentally and confirmed by optical simulations. The opportunity to fabricate and tune such an assembly of plasmonic particles on transparent substrate opens interesting perspectives for not only fundamental photonics but also potential optical applications.

An object withdrawn from a liquid bath is coated with a thin layer of liquid. Along with the liquid, impurities such as particles present in the bath can be transferred to the withdrawn substrate. Entrained particles locally modify the thickness of the film, hence altering the quality and properties of the coating. In this study, we show that it is possible to entrain the liquid alone and avoid contamination of the substrate, at sufficiently low withdrawal velocity in diluted suspensions. Using a model system consisting of a plate exiting a liquid bath, we observe that particles can remain trapped in the meniscus which exerts a resistive capillary force to the entrainment. We characterize different entrainment regimes as the withdrawal velocity increases: from a pure liquid film, to a liquid film containing clusters of particles, and eventually individual particles. This capillary filtration is an effective barrier against the contamination of substrates withdrawn from a polluted bath and finds application against biocontamination.

International audience

We report a method to prepare patterned mesoporous silica-based films with fine control of the pore sizes and the porous fraction by combining the sol–gel approach with an alkali ion post-treatment...

Withdrawing a plate from a suspension leads to the entrainment of a coating layer of fluid and particles on the solid surface. In this article, we study the Landau-Levich problem in the case of a suspension of non-Brownian particles at moderate volume fraction $10\% < \phi < 41\%$. We observe different regimes depending on the withdrawal velocity $U$, the volume fraction of the suspension $\phi$, and the diameter of the particles $2\,a$. Our results exhibit three coating regimes. (i) At small enough capillary number $Ca$, no particles are entrained, and only a liquid film coats the plate. (ii) At large capillary number, we observe that the thickness of the entrained film of suspension is captured by the Landau-Levich law using the effective viscosity of the suspension $\eta(\phi)$. (iii) At intermediate capillary numbers, the situation becomes more complicated with a heterogeneous coating on the substrate. We rationalize our experimental findings by providing the domain of existence of these three regimes as a function of the fluid and particles properties.

The transmission of light through low coverage regular and random arrays of glass supported silica micropillars of diameters 10 to 40 \micro\meter and height 10 \micro\meter is studied experimentally. Angle-resolved measurements of the transmitted intensity are performed at visible wavelengths by either a goniospectrophotometer or a multimodal imaging (Mueller) polarimetric microscope. It is demonstrated that for the regular arrays, the angle-resolved measurements are capable of resolving many of the densely packed diffraction orders that are expected for periodic structures of lattice constants 20 to 80 \micro\meter, but they also display features that are due to the scattering and guiding of light in individual micropillars or in the supporting glass slides. These latter features are also found in angle-resolved measurements on random arrays of micropillars of the same surface coverage. Finally we perform a comparison of direct measurements of haze in transmission for our patterned glass samples with what can be calculated from the angle-resolved transmitted intensity measurements. Good agreement between the two types of results are found which testifies to the accuracy of the angle-resolved measurements that we report., Comment: Latex, 11 pages, 10 figures

International audience; Surface differential reflectivity spectroscopy is a fast non-destructive in situ and real-time measurement technique which allows following the first stages of thin film deposition. However, when applied to sputtering technique, spectra can strongly be distorted by residual light coming from plasma in a way, as shown herein, that depends on sample reflectivity. Thanks to suitable measurements, before and after growth with and without plasma or illumination lights, a protocol of signal correction is proposed to get rid of the spurious plasma contribution. The interest of the method is illustrated in the case of silver deposition on a silicon substrate.