Search

In this work, we examine the relationships between the microstructure and the mechanical properties of glass fiber–reinforced polyamide 6,6 composite materials ( V f = 54%). These materials made by thermocompression incorporate different grades of high fluidity polyamide-based polymers and two types of quasi-UD glass fiber reinforcement. One is a classic commercial fabric, while the other specially designed and manufactured incorporates weaker tex glass yarns (the spacer) to increase the planar permeability of the preform. The effects of the viscosity of the polymers and their composition on the wettability of the reinforcements were analyzed by scanning electron microscopy observations of the microstructure. The respective influences of the polymers and the spacer on the mechanical performance were determined by uniaxial tensile and compression tests in the directions parallel and transverse to the warp yarns. Not only does the spacer enhance permeability but it also improves physical and mechanical properties: tensile longitudinal Young’s modulus increased from 38.2 GPa to 42.9 GPa (13% growth), tensile strength increased from 618.9 MPa to 697 MPa (3% growth), and decrease in ultimate strain from 1.8% to 1.7% (5% reduction). The correlation of these results with the damage observed post mortem confirms those acquired from analyses of the microstructure of composites and the rheological behaviors of polymers.

Hypothesis Computer fluid dynamics simulations of dynamic wetting are often performed using a slip model on the substrate. In previous studies, the generalized Navier boundary condition (GNBC) has shown promising results and could help clear the gap between molecular and continuum scales, but lacks quantitative comparisons to experiments. We seek to investigate the dependence between the contact-line velocity and the slip length in a GNBC, by confronting numerical simulations to experimental data. Experiments The physical properties of a molten polymer (polyethylene glycol) were assessed thoroughly. Its dynamic contact angle on a cellulosic substrate was measured carefully using the Wilhelmy method. The experiment was reproduced in a finite elements model using a GNBC. It was repeated for capillary numbers between 10−6 and 10−1, and slip lengths ranging from 1 μm to 1 mm. Findings A realistic value of the slip length was selected by matching the dynamic contact angles issued from numerical simulations and their experimental counterparts. The slip length behavior as a function of contact line velocity displayed a clear transition. The model also reproduced a dynamic wetting transition between frictional and viscous dissipations, which seems to be linked to an increasing difference between microscopic and macroscopic contact angles.

Hypothesis The temperature dependence of the static contact angle could a priori be predicted by using surface tension partitioning. An original model based on the transition state theory is also introduced. This model considers thermocapillary fluctuations on the droplet surface near the triple line and the self-affine pinning of this triple line against a solid substrate modeled with a pseudo-periodic distribution of adsorption sites. Experiments The temperature dependence of the static contact angle was studied for a representative range of liquids with different polarities and on a wide array of solid substrates for temperatures ranging from 25 to 240 °C. Atomic force microscopy (AFM) was also used to quantify the surface roughness of the solid substrates. Findings Whereas the surface tension partitioning failed to bring consistent results above room temperature, the transition state model proved very useful, thereby opening a way to yield predictive contact angle values with temperature variations. The introduction of a topological dimension in the equations yields a unified model that covers normal wetting (perfectly bonded liquids on smooth surfaces) but also the onset of Cassie-Baxter and Wenzel states on real surfaces. Moreover, the model encompasses the transition to complete wetting.

International audience; This work consists in an experimental investigation of forced dynamic wetting of molten polymers on cellulosic substrates and an estimation of models describing this dynamic. A previous work of Pucci et al. (2018) showed that for totally wetting liquids (as paraffin oils), temperature-induced variations in dynamic wetting are included into the capillary number (Ca) and then a master curve of dynamic contact angle (θ d) as a function of Ca can be obtained. The hydrodynamic theory (HDT) correctly describes the dynamic wetting for Ca > 2 · 10 −3. For lower Ca, a change in the dynamic wetting behavior was observed. Here, partially wetting liquids (polyethylene glycols, a.k.a. PEGs) at different molecular weight (Mn) were used at temperatures above their melting point to investigate the dynamic wetting behavior on cellulosic substrates for a large range of Ca. It was found that the dynamic curves of θ d vs. Ca depend on Mn. Moreover, the HDT correctly describes the experimental measurements for Ca > 2 · 10 −3. Below this threshold the dynamic contact angle decreases toward the static one. A linear correlation between parameters obtained fitting the HDT and the molecular weight of polymer was found. The prediction of dynamic wetting for low Ca (Ca < 2 · 10 −3) with the molecular kinetic theory (MKT) was also evaluated and discussed.

International audience; Flax fibres are a promising reinforcement in the development of biocomposites and are finding new applications in transport structures. However, there is a perceived problem with plant fibres related to the variability of the properties of these natural materials. This paper describes the factors which affect variability, from plant growth conditions to fibre sampling and testing. A large number of test results are presented (characterization of elementary fibres, bundles, assemblies of bundles, and unidirectional composites), and it is shown that provided fibre supply is carefully controlled, characterization procedures are appropriate, and manufacturing processes are optimal then excellent composite properties can be achieved with low variability.

International audience; The use of composite materials reinforced by flax fibres has been increasing steadily over the last 20 years. These fibres show attractive mechanical properties but also some particularities (naturally limited length, presence of a lumen, fibres grouped in bundles in the plant, complex surface properties and composition). An analysis of the available literature indicates that the quality of the composite materials studied is not always optimal (high porosity, incomplete impregnation, heterogeneous microstructure, variable fibre orientation). This paper reviews published data on the specific nature of flax fibres with respect to manufacturing of biocomposites (defined here as polymers reinforced by natural fibres). All the important steps in the process which influence final properties are analyzed, including the plant development, retting, fibre extraction, fibre treatment, preform preparation, available manufacturing processes, the impregnation step, fibre cell wall changes during processing and fibre/matrix adhesion. © 2018, Springer-Verlag France SAS, part of Springer Nature.

International audience; The aim of this study was to set an experimental protocol in order to identify key parameters in wetting dynamics during liquid composite manufacturing. To validate the procedure, model materials were studied : cellulose films as substrates,and paraffins and polyethylene glycols as liquids. The experimental procedure consisted of characterizing the substrates interms of morphology and surface chemistry, showing that, as a first approximation, it was possible to consider substrates as model surfaces without physico-chemical defects. The characterization of liquids as a function of temperature and molar mass was then performed, considering an additional issue : the molten state of polymers. Temperature conditions and molar mass were then considered in wetting and dewetting dynamic. The results were well described by the hydrodynamic approach for a range of capillary numbers. Using these models, physical parameters were estimated, explaining the influence of temperature and polymer chain lengths on wetting dynamics.; Ce travail a pour objectif de calibrer un protocole de mesure permettant d’identifier les paramètres clés dans les phénomènes de dynamique de mouillage lors de la mise en oeuvre de composites. Pour valider la procédure, des matériaux modèles ont été étudiés : des films cellulosiques en tant que substrats, et des paraffines et polyéthylène glycols comme liquides.La procédure expérimentale a consisté à caractériser le substrat en termes de morphologie et de chimie de surface, en montrant qu’en première approximation on peut s’affranchir des hétérogénéités de surface. La caractérisation de liquides en fonction de la température et en fonction de la masse molaire a été ensuite analysée en prenant en compte l’état fondu des polymères. Les conditions de température et de masse molaire ont été ensuite considérées en dynamique de mouillage et de mouillage. Les résultats ont bien été décrit par l’approche hydrodynamique pour un certain intervalle de valeur du nombre capillaire. Sur la base de ces modèles, des paramètres physiques ont été estimés, en précisant l’influence de la température et des longueurs de chaine de polymères sur la dynamique de mouillage.

International audience; All-cellulose nanocomposites reinforced by cellulose nanocrystals (CNC) were produced using a solvent consisting of 1-butyl-3-methylimidazolium chloride and dimethyl sulfoxide. Microcrystalline cellulose (MCC) was pre-dissolved at high temperature in the solvent. Freeze-dried CNC were then added to the slurry at room temperature, thereby avoiding complete CNC dissolution. Solid all-cellulose composite films were obtained by film casting, solvent exchange and drying. The MCC to CNC ratio was kept constant while the solvent content was incremented. The short-range and long-range cellulose–cellulose interactions in the solid materials were respectively assessed by Fourier-transform infrared spectroscopy and X-ray diffraction. The CNC used in this work contained both cellulose I and cellulose II. The cellulose concentration in the mixture drastically changed the overall crystallinity as well as the cellulose I to cellulose II ratio in the ACC. Cellulose II was formed by recrystallisation of the dissolved fractions. These fractions include the pre-dissolved MCC and the cellulose II portion of the CNC. Cocrystallisation with the cellulose I CNC acting as a template was also evidenced. This phenomenon was controlled by the initial solvent content. The correlation between the hygromechanical properties and the nanostructure features of the ACC was investigated by humidity-controlled dynamic mechanical analysis (RH-DMA). The introduction of the cocrystallisation and percolation concepts provided a thorough explanation for the humidity dependency of the storage modulus.

Resin Transfer Molding (RTM) is among the most commonly used fabrication processes for producing high quality and complex composite structural parts. RTM process consists of placing a dry fibrous preform into a mold cavity. A liquid resin is subsequently injected into that cavity. The consolidation of the part is then obtained by crosslinking in case of a thermosetting resin or by crystallization in case of thermoplastic one. Voids can be created in the porous medium during the flow of the resin. Presence of residual voids in the composite part at the end of the filling drastically affect mechanical performances. Even if several authors have contributed to a better understanding and modeling of the mechanisms of formation and transport of voids during injection, few experimental approaches allowed a direct measurement of the saturation curve. The aim of this study is then to identify the saturation of a fibrous preform by a liquid through thermal analysis. To address this issue, an experimental bench that allows the injection of a fluid into a textile preform has been used. This apparatus combines the measurement of temperatures and wall heat flux densities at several locations. A simplified modeling of the filling front has been performed with FEM using Comsol Multiphysics™. The saturation curve is modeled using several geometric parameters. Saturation is taken into account through the evolution of thermophysical properties. Effective thermophysical properties of the dry and completely-saturated porous medium in transverse and longitudinal directions have been measured by several methods, and their results have been then cross-checked and compared with good accuracy. The evolution between these two states has been modeled. A particular attention has been paid for the modeling of the transverse thermal conductivity. This parameter has been modeled using a periodic homogenization method as a function of the micro- and macro-saturation. The saturation curve parameters are determined by minimizing the cost function defined as the square difference between the measured and computed heat flux. The obtained saturation curve is finally compared with the one measured by a conductometric sensor.

Mesoscopic simulations of the transverse compression of textile preforms are presented in this paper. They are based on 3D FE models of each yarn in contact with friction with its neighbours. The mesoscopic simulations can be used as virtual compression tests. In addition they determine the internal geometry of the reinforcement after compaction. The internal geometry can be used to compute the permeability of the deformed reinforcement and to calculate the homogenised mechanical properties of the final composite part. A hypoelastic model based on the fibre rotation depicts the mechanical behaviour of the yarn. The compression responses of several layer stacks with parallel or different orientations are computed. The numerical simulations show good agreement when compared to compaction experiments.

Liquid composite molding processes are widely accepted in the aeronautic industry to manufacture large and complex structural parts. In spite of their cost-effectiveness, void defects created durin...

This paper presents the results of a numerical simulation of the ultimate flax fibre ( Linum usitatissimum) tensile mechanical behaviour using finite element analysis. Experimental data were used to develop a numerical multilayer model of the flax fibre. Thus, the influence of some parameters, such as cell wall thicknesses, microfibrils angles (MFAs), biochemical composition and mechanical properties of the biochemical components, on the flax fibre tensile mechanical behaviour has been investigated. Results show that the typical stress–strain curve profile of the flax fibre could be due to the mechanical properties of hydrophilic components (hemicelluloses) and thus to the environmental conditions. A parameter sensitivity study reveals that ultrastructural parameters (hemicelluloses and cellulose Young’s modulus) strongly influence the flax fibre mechanical behaviour and structural parameters (S2 cell wall layer MFA and thickness) significantly influence the fibre longitudinal Young’s modulus. Thus, the knowledge of the fibre ultrastructure seems to be the key of the understanding of the flax fibre mechanical behaviour.

The aim of this work is to underline the influence of the wetting behavior on bubble formation and transport during the impregnation of fibrous preforms for Liquid Composite Molding processes. The void prediction within the elaboration of composite material sparks off interest because it represents a significant issue to keep the expected mechanical properties of the final product. Voids or bubbles are mainly formed due to the resin flow competition inside and outside the tows. However, the experimental characterization of void formation and transport during the flow inside fibrous media remains delicate. Since the direct visualization of the liquid flow and the entrapped bubbles in a practical material is uneasy, we used model networks to study the formation of bubbles and their transport. Thus, our experimental study deals with a simple configuration of connected pores: two capillaries converging at a node perpendicularly (T-junction). We emphasized on microfluidic and millifluidic approaches where the wetting is significant during bubble formation mechanisms. For the same flow conditions, the experimental results reveal that the bubble lengths are higher for partially wetting liquids than for completely wetting ones. In a wetting case, it was demonstrated that the lubrication hypothesis can be a good approximation to describe the bubble transport in fibrous media.

We described an experimental technique, which can be used to measure simultaneously the three principal permeabilities of fiber reinforcement made of short or long fibers. This original technique utilizes the X-Ray radioscopy for detecting the position of the liquid front inside of the preform. An analysis of the different experimental results allowed us to measure, explain and justify the influence of the various parameters related to the flow conditions, the mold geometry and the reinforcement material structure. The experimental study proves that this technique is not limited by high anisotropy factors. We obtain good agreement between results of this visualization technique and numerical simulation of the resin flow.

International audience; This experimental work addresses the mechanical behavior of weaven UD glass fiber-thermoplastic composites designed for layered materials to be used in the automotive industry. The investigation was implemented in terms of the formulation of the PA66-based thermoplastic resins, the architecture of the fibrous reinforcement (warp spacer) and the glass fiber content. Longitudinal and transverse tensile tests results demonstrate the excellent mechanical behavior of these composites, which correlates with the rheological/permeability properties and wettability behaviour. © 2016 Author(s).

A simple model is used to explain the decrease of the Young’s modulus of plant fibers as the apparent diameter is increased. The assumption made in this work is that the fiber consists of a very thin but stiff outer shell, and a thicker but less stiff inner shell which defines the load bearing area. The model is favorably compared with the experimental data reported in the literature for flax, hemp and stinging nettle fibers.

International audience; The biosourced Polyamide 11 (PA11) is envisaged as a matrix in structural ecocomposites in combination with long flax fibers. Herein we compare the thermal, rheological, physical, mechanical and hydrothermal behavior of PA11 available as film or powder. These data are used for choosing between the two forms of the PA11 and to determine the temperature, holding time and pressure ranges suited for the elaboration of biocomposites by induction heating. © Lavoisier 2016.

Fiber-reinforced composite materials are often composed of fibers collected in bundles that are stitched together. During the impregnation of a fibrous preform by a liquid resin, the multiscale porous medium leads to an heterogenous flow front, and therefore bubbles may be created and entrapped. Indeed, for a wetting system, capillary pressure is higher inside bundle, due to the microspace between fibers, than outside the bundles that represent the macrospace, thus, inducing an overflow between both pore scales. Motivated by the prediction of bubble formation during fiber fabric infiltration for composite materials, we attempt to determine the bubble rate in imbibition through a simple model network with two connected capillaries, called “Pore Doublet Model” (PDM). Our system is composed of two parts: a first part, continuously interconnected, in which the suppling mass to the microchannel from the macrochannel occurs, and a second part connected only by nodes. To quantify the leading flow front, a theoretical model based on the supplying principle and arranged Washburn equation is proposed. This approach has been conducted for wetting liquids, Newtonian flows, incompressible fluids and pores, no inertial and gravitational forces and no dynamic contact angle. The geometrical variability (channel radius and length) and the different configuration of connections (continuous and discrete) influence the entrapped bubble rate, leading to either microbubble in the microchannel or macrobubble in the macrochannel. The outcomes can contribute to the knowledge of void formation especially during the filling of fibrous preforms and may extend the previous works on the PDM in general.

In this work, a train of confined bubbles inside a millimetric cylindrical T-junction device is studied. Calibrated bubbles need to be known precisely for multiphase phase flow applications. Experimental results of bubble formation and transport processes will be presented and discussed. For a liquid-gas flow, we will demonstrate how the wetting behaviour plays a significant role on the formation and motion of the bubble.

In this international permeability benchmark exercise, in-plane permeability data for two reinforcement fabrics, obtained using a total of 16 different experimental procedures, were compared. Although, for each procedure, the results appear consistent, different procedures result in a scatter of up to one order of magnitude in principal permeability values for each fabric at any given fibre volume fraction. The ratio of the principal permeability values varies by factors of up to 2. While experimental uncertainties and variability of the specimens affect the scatter in results for any single series of experiments, it is suspected that the main source of scatter in results from different procedures is related to human factors. Aiming at standardisation of measurement methods and interchangeability of results, "good practice" guidelines will be formulated in order to eliminate sources of scatter. (C) 2011 Elsevier Ltd. All rights reserved.

International audience

Dans ce travail, nous recherchons des correlations entre la morphologie, la biocomposition et les proprietes mecaniques de la fibre de lin. Pour ce faire, nous avons selectionne sept varietes de fibres parmi celles les plus cultivees en Normandie. Les resultats indiquent que ces varietes ne presentent pas de differences significatives en termes de leur biocomposition; par contre les differences des proprietes mecaniques d’une variete a une autre sont plus importantes et sont comparables a celles relevees au sein d’une meme variete. Il ressort que les ecarts entre les proprietes mecaniques des differentes varietes ne s’expliquent pas par la morphologie ou la biocomposition des fibres.

Flax fibres can be considered as natural composites made of concentric layers in which cellulose microfibrils are embedded helicoidally in a polysaccharidic matrix. Since these fibres possess good mechanical properties and a low density, they are more and more foreseen as a plausible reinforcement material in polymer-matrix composites. In the present work hundreds of flax fibres were analysed in terms of their morphology (by image analysis) and mechanical behaviour (by tensile testing). The common feature between microstructure analyses and mechanical properties is the large scattering of the experimental values. Moreover, the correlation of the morphological analysis with the mechanical properties shows that the scattering of the microstructural figures at this scale of observation is not responsible for the scattering of the mechanical properties.

In this study, investigations were carried out on flax at different levels of observation. Stems, bundles and elementary fibres were studied in order to correctly understand the fibre structure and deformation so that the mechanical behaviour of derived flax fibre-based composite materials could be modelled. New structural information was obtained and correlated to mechanical properties. The main results are that fibres taken near the root are the largest of the stem, that both intra-bundle size variability and inter-fibre size scattering are linked to large intra-fibre size dispersion, that fibre mechanical properties should be calculated on the basis of their diameter near their rupture point rather than their mean diameter, and that a new schema of the fibre internal structure could be drawn from focused ion beam experiments.

Flax fibers were used to process unidirectional composites by two different methods. Their mechanical properties obtained by tensile testing are discussed with respect to the properties of the fibers and those of the matrix (unsatured polyester). The similarity of the tensile curves of the composites and of the elementary fibers is attributed to the good adhesion of the fibers with the matrix. Moreover, as there is almost a linear evolution of the composite properties with the fiber volume fraction, these properties are used to estimate those of the real reinforcement material, that is, the flax bundles: the calculations lead to a fiber strength of 500-800 MPa and a fiber modulus of roughly 30 GPa, which is half the values obtained by tensile testing elementary fibers. These data may be helpful when trying to model the deformation behavior of flax fiber-reinforced composites.

International audience; The modelling of composite manufacturing processes where hydro-mechanical coupling takes place depends on the validity of compressibility and permeability models. In this work, the computer code initially used to simulate the effect of coupled hydro-mechanical load on composite preform (Ouahbi et al. Composites Part A, 38:1646-1654, 2007) is integrated into an inverse method to predict the compaction behaviour of the reinforcements. An experimental device developed at Le Havre is used to apply hydro-mechanical loads to the preforms. Two ramps of stress are imposed to the preform and the thickness evolution is measured as a function of time. The speed of thickness reduction is not constant and varies in the range of 0.1 to 12 mm/min. The effect of compression speed upon the saturated fabrics is investigated. For a fixed fibre volume fraction, an increase in stress is observed in increasing compression speed. The experimental results are compared to the compressibility curves determined by an inverse method. The calculated curves correspond to the compressibility curves experimentally obtained with low compression speed (∼0.25 mm/min). As a consequence, this suggests that a low compression speed should be applied when investigating the compressibility behaviour of composite preform with a view of modelling resin infusion processes.

Fibrous composites have very specific mechanical behaviours due to their compositions. They exhibit not only the strain in the fibre direction but also the shear caused by the relative movement of weft and warp during the deformation. To address these behaviours, in this paper, a constitutive model has been proposed for the simulation of fabric shaping process, which is based on a meso-macro approach. The strain energy by tension is considered in the field of large strain, and the shear behaviour is also taken into account. The bias-test experiment is employed to characterize the shear behaviour. The model is implemented in the user subroutine ‘Vuel’ of Abaqus/Explicit and it is applicable either to four nodes or to three nodes element. A set of elementary tests and a simulation of bias-test have been performed to validate this approach. Finally, a composite forming of real part is simulated with this method. In the simulation results, we can observe the winkle that takes place in the experiment test which corresponds well to the experimental.

In this paper we investigate the tensile properties of a natural composite material: the flax fiber. The beginning of the stress-strain curve of a flax fiber upon tensile loading appears markedly non-linear. The hypothesis of a progressive alignment of the cellulose microfibrils with the tensile axis provides a quantitative explanation of this departure from linearity. This hypothesis is confirmed by the similar behavior which characterises hemp and ramie fibers. Besides, it has long been recognized that the natural character of flax fibers induces a large scatter of their mechanical properties. This scatter is shown not to be associated with the pronounced cross-section size variation observed along the fiber profiles. Two fiber size measurement methods are compared in order to check their influence on the property scatter and the decrease of the fiber mechanical properties as a function of the fiber diameter.

Textile composites become more and more popular in aeronautic industries, due to their high performances. In this work, based on the model shown in the paper [9], a solid-shell element is used to simulate the woven composite forming. Comparing with other shell elements, a distinctive advantage of solid-shell elements is that the complication on handing finite rotations does not exist. Accounting the specific mechanical behaviors of woven composite material, the tensile, in-plane shear, bending and compressive energies are taken into account depending on the fiber direction, and then the total strain energy is computed to be the sum of those energies. The necessary material data for the simulation come from standard tensile, compressive and bias test experiments. Some forming simulations are performed with this method, and we can observe the wrinkles that exhibit the influence of bending stiffness. The results show the efficiency of the approach.

National audience

The topic of this paper concerns void defects. We introduce a model to describe the macroscopic resin flow considering the compression of residual air in the preform. A microscopic flow model is also developed to account for the microvoid formation due to the variation of resin velocity at the flow front. According to these mathematical models, a numerical simulation code was developed to perform a combined macro and micro flow analysis with the saturation of the medium. The results of these simulations are compared with experimental results obtained by the laboratory device. We present the real-time measurement of the void content evolutions during the injection, using an original sensor. Our technique is based on the electrical conductivity of the injected liquid. After the material and method presentation, some results obtained during injection are shown. The developed analysis technique and the numerical simulation code can be used for obtaining the optimal processing conditions and design parameters in manufacturing of composites by Liquid Composite Molding (LCM) processes.

In Liquid Composite Molding (LCM) processes the saturation of the reinforcement by the resin may induce the creation of porosity in the preform affecting the final properties of the composite. The purpose of this work concerns the development of an experimental protocol and the associated modeling to identify the dynamic saturation curve during filling by taking advantage of sharp contrasts of thermal properties existing between dry and fully-saturated reinforcement. To identify saturation, several injections were performed with a laboratory RTM mold for which thermal design allows accurate control of heat transfer. Several heat flux sensors were used to identify the saturation curve. Sensitivity analysis proves the feasibility of the method. The results are compared with a conductometric method with good agreement. Evolution of residual voids identified for several flow rates are also consistent with those expected according to the capillary number.

Resin transfer molding (RTM) has become one of the most widely used processes to manufacture medium size reinforced composite parts. To further enhance the process yield while ensuring the best possible quality of the produced parts, physically based optimization procedures have to be devised. The filling of the mold remains the limiting step of the whole process, and the reduction of the filling time has an important impact on the overall cost reduction. On the other hand, the injection cycle has to be appropriately carried out to ensure a proper fiber impregnation. Indeed, a partial fiber impregnation leads to the creation of micro-scopic and macro-scopic voids. In the present work, based on a double scale flow model and the capillary number Ca , an optimization algorithm is proposed to minimize the micro/macro-voids in RTM composite parts. The optimized injection flow rate ensures an optimum Ca at the flow front during part filling. The implemented algorithm allows the use of various constraints such as maximum capabilities of the injection equipment (i.e., maximum pressure or flow rate at the injection gates) or maximum velocity to avoid fiber washing. Bounded by these constraints, the optimization procedure is devised to handle any injection configuration (i.e., injection gates or vents locations) for two or three-dimensional parts. The numerical model is based on a mixed (FE/CV) formulation that uses non-conforming elements to ensure mass conservation. The proposed algorithm is tested for two and three-dimensional parts while emphasizing the important void reduction that results from the optimized injection cycle.

A structure-process integrated optimization methodology is proposed considering both the structural properties and manufacturing process for the weight minimization of composite laminated plates manufactured by Resin Transfer Molding (RTM) process. In the optimization problem, the structural weight of composite structures is minimized under the stiffness constraint and the mold filling time constraint which is a part of process cycle time. With the results obtained, it is investigated how the weight and material cost are traded-off. The optimization methodology suggests a guide of material selection in the preliminary conceptual design stage.

This work proposes to perform an analysis about the numerical simulation of LCM processes. After a description of the modelling context of these processes, our first part will describe the formulation of the laws, and a second step the application with the Thermo- Hydro-Mechanical coupling. We present a finite element procedure to simulate mold filling in the three dimensional case. The method presented in the paper is div-conform i.e. the mass of injected fluid is proved to be perfectly conserved. Numerical results are compared with experiments. In order to take into account the micro-macro voids during the filling of the reinforcement, we introduce a phenomenon of transport in saturation corresponding to a mechanism of hydrodynamic dispersion. This study is supplemented by a study of coupling between processes and mechanical properties, insisting on the advantage of introducing the concept of saturation.

Porous media are encountered in numerous applications from Soil Mechanics to composite materials manufacturing. Part I of this article investigated the flow behavior in a double-scale porous medium, i.e., in a material with a bimodal distribution of pore sizes. Typical examples of such medium are the dry fibrous reinforcements used in composite manufacturing through which resin is injected. This type of porous medium contains two types of pores: microscopic voids inside the fiber tows and macroscopic ones between the tows. Because of this double-scale structure, the saturation of such reinforcing materials is not instantaneous during resin injection. Part I has shown that the notion of saturation must be taken into account in order to describe accurately transient unsaturated flows during the filling of double-scale fibrous materials. Part II of this investigation is concerned with the transient flow behaviors that can be observed once the reinforcement is saturated. In permanent regime, saturated resin flows are governed by Darcy's law, which describes the purely linear pressure drop observed in fully saturated reinforcements. However, this model cannot be applied to transient saturated flows unless they are considered as a succession of quasi-stationary flows. Simple macroscopic measurements during transient flow regimes permit us to derive important information on the microscopic characteristics of the flow in connection with the deformations of the porous medium under the effect of fluid pressure. The pressure variations in time observed in permanent flow regimes, when the flow rate is changed, are connected with the double-scale porosity structure of the fibrous reinforcement. Therefore the second part of this analysis is focused on the stress-strain behavior of fibrous reinforcements deformed by the pore pressure. A physical model based on Kelvin-Voigt visco-elastic material is proposed here to describe the dynamic compaction and relaxation behaviors of saturated fibrous reinforcements. This approach is not only in good agreement with experiments, but it shows also that double-scale effects are intimately connected with the pressure variations observed in transient saturated flows.

This article presents a general approach to model flows through unsaturated porous media as they occur in Liquid Composite Molding (LCM). Saturated and unsaturated flows will be studied here both from the experimental and theoretical points of view. It is indeed important to distinguish between these two flow behaviors in order to understand the interactions between the three phases that coexist in a fibrous reinforcement: the solid and fluid phases on one hand, and the air content on the other. The experimental work presented here includes the study of permanent and transient flow regimes, both for saturated and unsaturated porous media. The dynamic effects that occur during fluid injection through fibrous reinforcements highlight the double scale, structure of their pore volume. The ratio between saturated and unsaturated permeabilities appears to be connected to the degree of saturation and to the porosity of the part. Given the importance of permeability as a key input parameter in process simulation, this article proposes to introduce the degree of saturation in the equations that govern the flow in order to increase the accuracy of numerical predictions. This will not only provide a better understanding of the underlying physical phenomena during the fluid impregnation of a fibrous preform, but will also ultimately allow the study of air entrapment mechanisms that govern the quality of composite parts.

International audience; Les paramètres tels que le taux et l'orientation des fibres, et des phénomènes tels que l'endommagement de la préforme et la formation de bulles couplent le contrôle des procédés CRTM (Compression Resin Transfer Molding) et les propriétés mécaniques de la structure finale. Ce travail étudie l'effet de la prise en compte du couplage procédé d'élaboration/structure dans les plaques composites. Des modèles semi-analytiques sont utilisés pour simuler le CRTM et estimer les rigidités et la charge critique de flambement. L'effet du couplage est quantifié en comparant les solutions de quatre problèmes d'optimisation sous contraintes. Cette analyse montre qu'un procédé d'injection à pression imposé et de compression successive est optimal. Le coût des hautes performances structurales en termes de temps d'élaboration est important.

Numerical simulations on confined bubble trains formed by cross-flowing streams are carried out with the numerical code THETIS which is based on the Volume of Fluid (VOF) method and has been developed for two phase flow studies and especially for a gas-liquid system. The surface tension force, which needs particular attention in order to determine the shape of the interface accurately, is computed using the Continuum Surface Force model (CSF). Through the coupling of a VOF-PLIC technique (Piecewise-Linear Interface Calculation) and a smoothing function of adjustable thickness, the Smooth Volume of Fluid technique (SVOF) is intended to capture accurately strong interface distortion, rupture or reconnection with large density and viscosity contrasts between phases. This approach is extended by using the regular VOF-PLIC technique, while applying a smoothing procedure affecting both physical characteristics averaging and surface tension modeling. The front-capturing strategy is extended to gas injection. We begin by introducing the main physical phenomena occurring during bubble formation in microfluidic systems. Then, an experimental study performed in a cylindrical T-junction for different wetting behaviors is presented. For the wetting configuration, Cartesian 2D numerical simulations concerning the gas-liquid bubble production performed in a T-junction with rectangular, planar cross sections are shown and compared with experimental measurements. Finally, the results obtained of bubble break-up mechanism, shape, transport and pressure drop along the channel will be presented, discussed and compared to some experimental and numerical outcomes given in the literature.