Search

International audience; The control of biofouling is required to improve membranes performances for water filtration. In this context, developing bioinspired approaches to modify membranes and give them antifouling properties is an interesting alternative. In this study, we elaborated new anti-biofouling membranes by coating them with vanillin, a natural bioactive molecule acting as a Quorum Sensing Inhibitor (QSI). First, the adsorption of vanillin by filtration was optimized to obtain homogeneous surface modifications. The modified membranes were then characterized in terms of physical-chemical properties: the results showed that they were more hydrophobic and exhibited a 14% decrease in pure water flux. Then, their anti-fouling properties were evaluated using complementary multiscale characterization methodologies; AFM single-cell experiments showed that vanillin adsorption had an effect on both the probability of bacterial cells to adhere to the surface, and on the force of adhesion. Direct EFM and SEM observation and adhesion assays further confirmed the effect of vanillin-modified membranes at the population-scale which showed a decrease in the bacterial coverage rate up to 50%. Altogether, it confirms the multifunctionality of vanillin-coated membranes against biofouling: reduction of bacterial initial adhesion by changing the membranes surface properties and reduction of biofilm formation by quorum sensing inhibition

International audience; The behavior of a particle mixture composed of short fibers and silicon carbide powder in variable proportions has been experimentally investigated. The study shows that the hydrodynamic behavior of a bed initially composed of powder is affected as soon as a few percent of short fibers are added. The minimum fluidization velocity (Umf) is globally lowered and the intrinsic properties of the porous medium (equivalent particle diameter, porosity, flow coefficients…) are altered. Several hydrodynamic equations describing Umf as a function of the pressure are considered and the intrinsic properties of the bed such as the hydraulic diameter or the Darcy and Knudsen coefficients are fitted. The Darcy-Klinkenberg equation on one hand and a semi-empirical correlation based on Ergün’s equation on the other hand, are both adapted to describe the Umf behavior as a function of a limited number of initial parameters. The semi-empirical correlation leads to the best Umf prediction, with an absolute error of less than 10%, using only the temperature, pressure and percentage of fibers in the mixture as input parameters.

Photocatalysis and Plasma are widely considered as efficient and eco-friendly methods for the air treatment at domestic and industrial scales. In fact, the respective application of titanium dioxide (TiO2) and Dielectric-Barrier-Discharge(DBD) drew increasing attention in the last decades for their, flexible, non-selective and high efficiencies for the mineralization and the degradation of large selection of molecules. Nevertheless, the application of these techniques in separate conditions might generate undesired by-products that might hinder their treatment efficiencies. In this study, the effect of combining photocatalysis and Dielectric-Barrier-Discharge (DBD)techniques was investigated for the treatment of livestock building air at laboratory and industrial scale.. of Experiments were conducted in separate (i.e., Photocatalysis or plasma) and combined (i.e., Photocatalysis and plasma) systems to assess the degradation efficiency of two molecules, namely (NH3) and Propionaldehyde (ProPA) under various operating conditions. Studies onducted on laboratory scale showed a interesting level of treatment when applying either photocatalysis or plasma, with degradation yields reaching 29 and 36% for NH3 and 37 and 42% for ProPA, respectively. These degradation efficiencies were further enhanced when combining the processes, thus reaching 72 and 83% for NH3and ProPA, respectively. Moreover, investigations conducted on the generation of ozone during photocatalysis / Plasma treatment showed a decrease in the O3 concentration when compared to separate Plasma treatment. The photocatalytic degradation of ozone under UV radiation and in presence of TiO2 led to Low concentrations of ozone were detected at joint processas a consequence ofits catalytic decomposition under UV-light into HO2°, OH°,O2°-radicals at high humid conditions. The synergetic effect during the pollutant degradation was explained by the association of discharge plasma, presence of O3, UV-light and TiO2..Additionally,The implementation of the photocatalytic-plasma discharge reactor was performed to treat livestock sector air effluent loaded with ammonia.The effect of airflow and relative himidity were assessed. Results showed a significant NH3 removal efficiency in real conditions where it varied between59 and 75% for a flowrate of 2 m2/h and a relative humidity of 100%. Additonnlay, chemical and olfactometric studies were conducted to assess the performance of cimined sytem with real efflluent of livestock building air.

International audience; Freshwater supply is declining in the context of climate change, pollution, and soil salinization, calling for sustainable methods to produce drinking water. For instance, salt water can be converted into pure water by steam generation. Interfacial solar steam generation involves photoabsorbers consisting of a photothermal material with broad solar absorption and a porous substrate with a thermal insulating character. Nonetheless, scaling up of classical devices for interfacial solar steam generation is actually limited by cost, biofilm formation, salt fouling, complicated fabrication processes, and toxicity. Alternatively, wood-based devices are cheap, biodegradable, abundant, and display high fluxes of evaporation compared with other non-biodegradable photoabsorbers. Here we review the design and applications of wood-based solar steam generation devices, with focus on wood structure and properties, different types of devices, and factors controlling the evaporative performance.

International audience; Cyclohexane oxidation chemistry was investigated using a near-atmospheric pressure JSR at 570 K and ϕ = 0.8. Numerous intermediates including hydroperoxides and highly oxygenated molecules were detected using synchrotron vacuum ultraviolet photoelectron photoion coincidence spectroscopy. Supported by high-level quantum calculations the analysis of photoelectron spectra allowed the firm identification of molecular species formed during the oxidation of cyclohexane. Besides, this work validates recently published gas chromatography and synchrotron vacuum ultraviolet photoionization mass spectrometry data. Unambiguous detection of characteristic hydroperoxides (e.g. γ-ketohydroperoxides) and their respective decomposition products provide support for the conventional O2-addition channels up to the third addition and their relative contribution to the cyclohexane oxidation.The results were also compared to the predictions of a recently proposed new detailed kinetic model of cyclohexane oxidation. Most of the predictions are in line with the current experimental findings highlighting the robustness of the kinetic model. However, the analysis of the recorded slow photoelectron spectra indicating the possible presence of C5 species in the kinetic model provides hints that substituted cyclopentyl radicals from cyclohexyl ring opening might play a minor role in cyclohexane oxidation. Potentially important missing reaction are discussed.

International audience; Autothermal operation is a big challenge for a chemical looping combustion (CLC) demonstration plant. One of the key points of CLC autothermal operation is to achieve a high solid circulation rate and enough solid residence time. To address this challenge, a dedicated lifter is proposed to control the solid circulation rate and residence time in the dual fluidized bed reactor system. This dedicated lifter is integrated into a 1.5 MWth CLC cold flow model to evaluate the control of the solid circulation rate and residence time of this new CLC scheme. Continuous and stable circulation of solid particle operation is achieved in the cold flow model, and it has been demonstrated that the dedicated lifter can effectively control the solid circulation rate. The effect of gas velocity, bed inventory, and particle size on the fluidization in the whole loops is studied. The dedicated lifter can control the solid circulation rate in the range of 20–68 kg m–2 s–1, and residence time in the fuel reactor is maintained in the range of 180–610 s. The solid circulation rate and residence time obtained in this work satisfy the theoretical requirements of CLC autothermal and char gasification, respectively, which indicates that the autothermal operation of the CLC unit as well as high carbon capture efficiency can be realized in this new CLC scheme.

International audience; Macroscopic models for diffusion and heterogeneous reversible reaction of two species in porous media are developed by using coupled homogenization technique and spectral approach. Three representative cases related to the order of magnitude of the macroscopic Damköhler number DaL, namely predominating reaction, diffusion-reaction of the same order and dominating diffusion, are considered. The concentrations are developed as time series in an eigenfunctions basis associated with periodic spectral problems formulated in the unit-cell, thus forming a new microscopic problem to be homogenized. Such an approach represents a powerful tool to upscale diffusion-reaction microscopic problems, especially for high Damköhler number values where classical asymptotic development fails. It enables to capture the physics at very short times, when the characteristic time of reaction involved is much faster than the diffusion one. This work allows us to formulate the complex macroscopic laws describing the heterogeneous diffusion/reaction problem for two species in high Damköhler number regime.

International audience; Syngas biomethanation is a promising technology for waste to energy conversion. However, it had not yet been tested at high syngas flow rates. The aim of this study was to assess the possibility for syngas biomethanation to reach high methane productivity at higher syngas inflow rate. A pressurized stirred column was implemented. The syngas inflow rate was gradually increased, and two different increase strategies were compared. The highest methane productivity achieved yet with syngas-biomethanation was obtained, with 23.2 LCH4/L/d, with high conversion efficiencies of 89% for H2 and 82% for CO. The mass transfer performances of the process were investigated, and the existence of a biological enhancement factor was observed. Considering an enhancement factor in bioprocesses is a pioneering concept that could change the way we design bioreactor to improve mass transfer. The high methane productivity obtained in this study paves the way for the process industrialization.

Capturing and utilising carbon dioxide (CO2) is a major challenge for developing a low- carbon economy. In addition, the reduction of CO2 allows the synthesis of platform molecules for the chemical and energy industry. Anaerobic mixed cultures contain homoacetogenic microorganisms (HAC) capable of reducing CO2 to acetate. However, one of the obstacles to their use is the understanding and control of their functional diversity. In particular, managing the competition between HAC and hydrogenotrophic methanogens (HM) that convert CO2 into methane is crucial to selectively produce acetate.This study contributes to bring new knowledge on the competition between HAC and HM. In this sense, mass transfer between the gas phase where the substrates are located, and the liquid phase which contains the microbial catalysts, as well as kinetic and thermodynamic aspects of biological reactions have been integrated in this work. The microbial competition between HM and HAC was studied in successive batches. The effect of temperature between 25 °C and 35 °C was investigated, as well as the time of each batch, leading to different states of mass transfer limitation in the system. A clear effect of temperature between 25 °C and 35 °C on the outcome of the competition between HM and HAC was highlighted, as well as the effect of mass transfer limitation. This study contributes to identify specific process parameters influencing the selection of HAC over HM and should help in the design of experiments depending on the target product from H2/CO2.In successive batches, lifting of mass transfer limitation, as well as lowering of the temperature to 25 °C made it possible to select HAC in mixed anaerobic cultures for acetate production by eliminating the methanogens, after four successive batches of 4 to 6 days each.

For approximately five decades, bulk niobium resonators have been the best RF accelerating cavities available for accelerators. Thanks to the huge research efforts made by the SRF community, niobium cavities provide today reliably exceptional performances with accelerating gradient up to 40 MV/m. Nonetheless, niobium cavities still witness performance limitations when exposed to high fields which are essentially due to current vortices that appear in the superconductors when a magnetic field is applied. These vortices start dissipating energy and creating heat as soon as they are set to motion by the RF field. Moreover, their interactions with native oxides and the material defects drastically demeans superconducting performances. This work aims at exploring a new technological pathway to enhance the performances of SRF cavities through functionalizing their inner surface with thin film coatings. Due to its complex shape, coating an SRF cavity demands a finetuning of the film’s chemical and structural properties and a thickness uniformity down to the nanometric scale. To our knowledge, atomic layer deposition is a very well suited deposition technique to fulfill these requirements. This Project is a synergistic approach between synthesis, characterization and performance tests of the most effective ALD deposited thin films to address some of the most detrimental phenomena to RF cavities operation and to explore new technological routes to boost their performances.; Depuis environ cinq décennies, les résonateurs en niobium massif ont été les meilleures cavités radiofréquences disponibles pour les accélérateurs. Grâce aux énormes efforts de recherche fournis par la communauté SRF, les cavités en niobium offrent aujourd'hui des performances exceptionnelles et fiables avec des gradients accélérateur allant jusqu'à 40 MV/m. Néanmoins, les cavités en niobium connaissent encore des limitations de performances lorsqu'elles sont exposées à des champs élevés qui sont essentiellement dus aux tourbillons de courant qui apparaissent dans les supraconducteurs lorsqu'un champ magnétique est appliqué. Ces tourbillons commencent à dissiper de l'énergie et à créer de la chaleur dès qu'ils sont mis en mouvement par le champ radiofréquence. De plus, leurs interactions avec les oxydes natifs et les défauts du matériau dégradent considérablement les performances supraconductrices. Ce travail vise à explorer une nouvelle voie technologique pour améliorer les performances des cavités SRF en fonctionnalisant leur surface interne avec des revêtements en couches minces. En raison de sa forme complexe, le revêtement d'une cavité SRF nécessite un contrôle fin des propriétés chimiques et structurelles du film et une uniformité d'épaisseur jusqu'à l'échelle nanométrique. Le dépôt par couche atomique (ALD) est une technique de dépôt particulièrement adaptée pour répondre à ces exigences. Ce projet est une approche synergique entre la synthèse, la caractérisation et les tests de performance sur cavités recouvertes par ALD de couches minces les plus efficaces pour traiter certains des phénomènes les plus limitants au fonctionnement des cavités RF et pour explorer de nouvelles voies technologiques pour améliorer leurs performances.

International audience; Biological methanation is an emerging Power-to-Gas technology that provides a flexible, efficient, and long-term storage system to address the renewable energy intermittency issue. However, scaling-up a biological methanation plant requires overcoming the bottleneck of H2 gas–liquid mass transfer. The current work demonstrates analytically, experimentally and numerically the importance of H2 mass transfer on CH4 purity and productivity. The influence of sparger design was investigated in a pilot-scale bubble column bioreactor. An increase of 265 % in kLa and subsequently 81 % in CH4 purity is obtained when a uniform porous plate is used over a 4-points porous sparger. Numerical simulation of the bioreactor was performed using a multiscale modeling approach coupling 1D and CFD models. The numerical simulations were validated satisfactorily by the experimental data. Parametric studies were also performed using the 1D model to reveal the impact of operating conditions on the bioprocess. An analytical solution for the productivity of biomethanation reactors at steady-state is proposed based on mass balances and provides new insights into biological methanation. It is found that any modification of operating parameters that improve H2 mass transfer, such as higher gas recirculation rate and smaller bubble size, leads to a better bioreactor performance.

International audience; All-solid-state lithium metal batteries (ASSLMBs) have received much attention because of their potential for high energy density and high safety performance. However, the poor compatibility of the lithium/solid electrolytes (SEs) interface and the penetration of lithium dendrites during cycling prevent them from achieving current densities that are capable of commercialization. In this work, a lithium-metastable solid electrolyte and a lithium-unstable solid electrolyte have been mixed to obtain mix-structured solid electrolyte. In particular, Li5.5PS4.5Cl1.5 acts as a lithium-compatible matrix, and Li10SnP2S12 as a dendrite-scavenger. The synergy between the two generates a stable interface with the Li anode and effectively suppresses the penetration of lithium dendrites. Li symmetric cells can operate stably for 160 h at a current density of 0.5 mA cm-2 at room temperature. The critical current density can reach 5 mA cm-2 and the overpotential is less than 0.5 V. The Li/SEs/Li4Ti5O12 all-solid-state cells show a high capacity of 174 mAh g-1 at 0.2 C and 148 mAh g-1 at 2 C. High-capacity retentions of 97.3% and 94.4% are shown after 75 cycles at 1 C and 2 C at room temperature, respectively. This work highlights the advantages of the electrolyte with mixed structure and provides a new strategy to resist lithium dendrites and improve the critical current density and cycle life of ASSLMBs.

International audience; The fluidization of short Hi Nicalon S fibers was investigated before considering CVD coatings with controlled composition and thicknesses. The fluidization of a short fiber batch alone being delicate and not reproducible, the study consisted in comparing the hydrodynamic behavior of a powder alone, to that of a short fiber/powder mixture. The fluidization behavior of the particle beds was studied by following their pressure drop as a functionof pressure and temperature. The minimum fluidization velocity (U$_{mf}$) was evaluated at room temperature (T$_{amb}$) at different reduced pressures, and at 200 mbar as a function of temperature. The results are in good agreement with the theoretical equations. At low pressures and T$_{amb}$, U$_{mf}$ increases rapidly. In contrast, U$_{mf}$ varies only slightly with temperature. While the pressure does not seem to affect the fluidization behavior of the powder alone, it displays a clear optimum at 200 mbar for the short fiber/powder mixture.

International audience; The prediction of solvation quantities is an important issue in several fields of chemical and process engineering. In this regard, this work was conducted to increase the prediction power of the equation-of-state (EoS) approach by coupling the tc-PR (translated-consistent Peng-Robinson) EoS with COSMO-RS (Conductor-like Screening Model for Real Solvents) as gE model through an advanced mixing rule. Since the only input parameter of the COSMO-RS model is the charge density (σ) profile of each molecule, the resulting EoS (called tc-PR/COSMO-RS) does not require any binary interaction parameter contrary to other predictive cubic EoS, such as VTPR, PPR78, PSRK, and UMR-PRU. This feature allows the quantification of solvation data of a great number of systems from readily available pure-component properties (critical temperature and pressure, acentric factor and σ -profile). Satisfactory results were obtained for the prediction of solvation Gibbs energy data that were extracted from the extensive CompSol database (the average absolute deviation is 0.30 kcal/mol). The proposed model represents a very flexible alternative for the prediction of solvation data, since it can deal with a large variety of solutes in either a pure or mixed solvent, and for the entire phase diagram (including the supercritical region).

International audience; The capability of supercritical CO2 to extract edible oil from handmade Argan press cake was investigated. The aim is to enable Moroccan cooperatives to improve their economic situation by valuing the handmade argan press cake, which is nowadays considered as a waste, applying a clean extraction process. Extraction experiments were conducted at 300 and 400 bar, 333 K and 0.14 kg/h on dried biomass. The fatty acids and tocopherols compositions of the extracted oil were found similar to previous studies and correspond to a commercial edible oil. The air flow dried biomass exhibits a higher extraction yield compared to the freeze-dried biomass. A pressure of 300 bar seems to be suffcient to allow the extraction of oil with a satisfactory extraction kinetic.

The objective of this study was to evaluate the antioxidant (AA), antibacterial, and antifungal activity of fresh and pre-treated apricot extracts, dried by microwave at different powers (200, 400, and 800 W), extracted by the cold soaking method, and ultrasound-assisted extraction (UAE). Biological activity (bacterial and fungal) was estimated by agar disk diffusion test against four bacterial strains (Escherichia coli, Pseudomonas aeruginosa, Staphylococcus aureus, and Streptococcus sp.), and two fungal strains (Candida spp. and Geotrichum capitatum). Methanolic extracts of apricot fruits: fresh, dried processed, and apricot dough extracted by cold soaking showed a higher AA ranging from 34.22 to 96 % than the other extracts extracted by UAE with values ranging from 14.37 to 66.88 %. The results of tested extracts from fruits (Prunus armeniaca L.) extracted by both extraction methods showed the highest inhibitory activity against most of the tested bacterial and fungal strains with inhibition zones ranging from 4 to 45 mm. The biological activity (antibacterial and antifungal activity) has been improved using different treatments and microwave powers for drying apricots. In addition, the results of the biological activity of the extracts obtained by UAE are the best compared to cold soaking. However, it was determined that the UAE extraction method of cold soaking and drying of apricot fruits was more appropriate for the food industry due to the obtaining of extracts with good antibacterial and antioxidant potential, and their incorporation in foods would increase their production and uses, thus leading to the development of superior biofunctional foods, and the use of bio-solvent, such as methanol, which is easily available in high purity, highly safe, and completely biodegradable., Cilj ovog istraživanja bio je procijeniti antioksidativno, antibakterijsko i antifungalno djelovanje ekstrakata svježih i obrađenih marelica (Prunus armeniaca L.), sušenih u mikrovalnoj pećnici pri različitim snagama (200, 400 i 800 W), dobivenih metodom hladnog namakanja i ekstrakcijom potpomognutom ultrazvukom. Biološka aktivnost (antibakterijska i antifungalna) procijenjena je difuzijskim testom na agar disku koristeći četiri bakterijska soja (Escherichia coli, Pseudomonas aeruginosa, Staphylococcus aureus i Streptococcus sp.) i dva gljivična soja (Candida spp. i Geotrichum capitatum). Metanolni ekstrakti svježih i sušenih plodova te tijesta marelice dobiveni hladnim namakanjem pokazali su antioksidativno djelovanje u rasponu od 34,22 do 96 %, što je više od ekstrakata dobivenih ekstrakcijom potpomognutom ultrazvukom s vrijednostima u rasponu od 14,37 do 66,88 %. Ekstrakata iz plodova marelice dobiveni objema ekstrakcijskim metodama pokazali su najveću inhibitornu aktivnost prema većini ispitivanih sojeva bakterija i gljiva sa zonama inhibicije u rasponu od 4 do 45 mm. Biološka aktivnost (antibakterijska i antifungalna) poboljšana je primjenom različitih intenziteta sušenja marelica. Osim toga, rezultati biološke aktivnosti ekstrakata dobivenih ekstrakcijom potpomognutom ultrazvukom bolji su u usporedbi s hladnim namakanjem.

This study was carried out to verify the influence of drying parameters on phenolic and flavonoid compounds of apricots (Prunus armeniaca L.) treated with sucrose, NaCl, and sodium bisulphite solutions dried by microwave at different powers (200, 400, and 800 W). We used two extraction methods, namely, cold soaking and ultrasound-assisted extraction (UAE). Total phenolics and flavonoids in fresh and dried apricots and apricot dough were estimated using the Folin-Ciocalteu reagent and the aluminium trichloride method, respectively. Fresh apricot contained considerable amounts of polyphenols and flavonoids by the cold soaking and UAE (285.43 and 165.49 mg GAE/100 g DM and 48.57 and 12.11 mg QE/100 g DM, respectively). Analysis of the data showed that the decrease in polyphenol and flavonoid contents of the dried treated apricots compared to the fresh material was significant. The greatest losses of these nutrients were recorded when applying the ultrasonic extraction method., U ovom radu ispitan je utjecaj parametara sušenja na fenolne i flavonoidne spojeve marelica (Prunus armeniaca L.) tretiranih otopinama saharoze, natrijeva klorida i natrijeva bisulfita te sušene u mikrovalovoj pećnici pri različitim snagama (200, 400 i 800 W). Primijenjene su dvije metode ekstrakcije, hladno namakanje i ekstrakcija potpomognuta ultrazvukom (UAE). Ukupni fenoli u svježim i suhim marelicama te tijestu marelica određeni su primjenom Folin-Ciocalteu reagensa, dok su flavanoidi određeni metodom s aluminijevim(III) kloridom. Obje metode ekstrakcije pokazale su da je svježa marelica sadržavala znatne količine polifenola i flavonoida. U suženim marelicama zabilježene su znatno manje količine polifenola i flavanoida. Najveći gubici tih nutrijenata zabilježeni su prilikom primjene ekstrakcije potpomognute ultrazvukom.

International audience; To address the problem of pollution caused by antibiotics in black water, we synthesized membranes containing the g-C3N4/TiO2 photocatalysts and tested them for the removal of sulfamethoxazole and tetracycline in pure water conditions and black water. We compared the basic membrane filtration and photocatalytic performance of the g-C3N4/TiO2 and the PVDF membranes, and investigated the influencing factors and application aspects of membrane filtration-photocatalytic systems for antibiotic removal. The anti-fouling performance and re-usability of g-C3N4/TiO2 membranes were investigated by evaluating the fouling reversibility of photocatalytic membranes. The results showed that g-C3N4/TiO2 improved the porosity, hydrophilicity and permeability of the membranes significantly. PgT-3 (PVDF/g-C 3 N 4 /TiO 2) membrane with 0.03 wt% of g-C3N4/TiO2 has the best overall performance with 72.8 % and 63.9 % removal efficiency for sulfamethoxazole and tetracycline respectively. Neutral or weakly acidic solution (pH = 5.0-7.0) is favorable for the removal of both study antibiotics. The complex composition of black water increased the adsorption load on the membrane and caused the inhibition of the photocatalysis of the g-C3N4/TiO2 membrane. The absorption of visible light by g-C3N4 accelerates the electron transfer rate and promotes the separation of electrons from holes. The oxidation-active substance h + produced in the system plays an important role in the removal of sulfamethoxazole and tetracycline.

International audience; Switchable solvents are changing properties when exposed to an external stimulus. As example, for thermomorphic multiphasic system (TMS), temperature changes can induce phase separations. Aqueous systems presented here are formulated with a carboxylic acid (weak acid) and an amphiphilic tertiary amine (weak base). They show thermo-switchable hydrophobicity and after separation, the weak acid and base are mainly found in the organic phase while the aqueous phase is mostly composed of water, up to 99 wt%. The phase separation temperature (cloud point) can be tuned from 20 to 80 °C depending on the choice of the carboxylic acid and on the acid: base molar ratio. A collection of saturated aliphatic carboxylic acids and benzoic acid derivatives have been investigated to evaluate the impact of different physico-chemical properties on the cloud point. The octanol/water partition coefficient (log P) and acidic constant (pKa) were identified as key physico-chemical properties to adjust the phase separation temperature. Finally, as illustration of the efficiency of these systems to remove a chemical from water, it has been shown that 99 % extraction of an anionic dye, methyl orange, could be achieved with most of the tested TMS.

The future of combustion processes lies on a significant reduction of pollutant emissions, as well as increased efficiencies. The control of the ignition phase is a critical step towards this goal, in many different configurations. Whatever the application, reproductible initiation of the combustion phase is sought. It is well known that the fuel ignition is largely dependent on low-temperature combustion (LTC) mechanisms, responsible for the occurrence of cool flames. These mechanisms involve the formation and decomposition of highly reactive intermediates, peroxides, whose structure reflect the initial fuel structure. Therefore, the fuel structure strongly influences its reactivity. Furthermore, the widespread use of bio-based or synthetic fuels requires a detailed understanding of their reaction pathway in this temperature domain, which may differ from those of fossil fuels. Finally, following the recent demonstration of easier cool flame ignition through the use of dielectric barrier plasma discharge or ozone-seeding stresses the need for fundamental studies on these low-temperature combustion kinetics. Within this work performed at the PC2A laboratory, a stagnation plate burner dedicated to the low-temperature combustion kinetics studies of fuels inside stabilized cool flames was developed and characterized.The detailed chemical structures of cool flames were established by coupling different analytic and optical techniques, namely formaldehyde Planar Laser Induced Fluorescence (CH2O-PLIF), excited formaldehyde chemiluminescence, temperature measurements by thermocouple, (micro-) gas chromatography and mass spectrometry. Three different fuels were selected in this work: dimethyl ether (DME), diethyl ether (DEE) and oxymethylene ether-2 (OME-2). Cool flames were generally stabilized under lean conditions (varying between 0.2 and 0.5) by means of ozone-seeding (varying between 1 and 2% mol. of the mixture), except for diethyl ether, for which a stoichiometric self-sustaining cool flame was investigated without ozone-addition. Comparing measured and simulated cool flames chemical structures, using various detailed kinetic mechanisms from the literature, allowed to assess the performance of these models in these specific conditions. Furthermore, the ozone influence as in-situ radical provider was investigated. While its use only slightly affects the products distribution downstream of the cool flame, this species highly impacts some reaction pathways in the cool flame, promoting chain-propagating over chain-branching reactions.Finally, Particle Image Velocimetry (PIV) technique was used in the stagnation plate burner configuration, allowing the measurement of cool flames laminar burning velocities (LBVs) at atmospheric pressure for the first time. The LBV is an intrinsic data of the fuel, depending on the initial mixture composition, temperature and pression. This data is particularly valuable for kinetic studies, but was almost never investigated in the case of cool flames. Its measurement was successfully achieved inside lean, ozone-seeded dimethyl ether cool flames, the experimental propagation speeds varying between 14 and 18 cm.s-1. This study notably allowed to highlight the most important reactions for the prediction of cool flames LBVs.The newly acquired data in this work constitute a strong foundation to further work on the low-temperature combustion mechanisms in the presence of ozone, in novel experimental conditions complementary to those met in typical kinetic reactors.; L'avenir des processus de combustion est tributaire d'une réduction significative des émissions polluantes, ainsi que de l'augmentation des rendements énergétiques. Le contrôle de l'initiation de la combustion est une étape critique dans cette direction, et ce dans de nombreuses configurations. Quelle qu'en soit l'application, l'initiation reproductible de la phase de combustion est recherchée. Cependant, l'initiation de l'inflammation des carburants est largement dépendante des mécanismes chimiques de basses températures, dits LTC (Low-Temperature Combustion), responsables de la formation de flammes froides. Ces mécanismes impliquent la formation et décomposition de composés hautement oxygénés instables, des peroxydes, dont la structure reflète celle du carburant initial. Celle-ci conditionne donc fortement sa réactivité. De plus, la généralisation de carburants biosourcés ou de synthèse nécessite une compréhension fine de leurs chemins réactionnels, ceux-ci pouvant être spécifiques aux fonctions chimiques absentes des carburants fossiles. Enfin, les récentes démonstrations de la facilitation de l'initiation d'une flamme froide par interaction avec un plasma froid ou par ensemencement d'ozone souligne le besoin d'études fondamentales de ces mécanismes chimiques de combustion de basses températures. Dans le cadre de cette thèse effectuée au laboratoire PC2A, un brûleur à plan de stagnation dédié à l'étude de la cinétique des flammes froides a été conçu et caractérisé.La structure chimique détaillée de flammes froides est obtenue par la mise en commun de différentes techniques analytiques et optiques, respectivement la fluorescence induite par plan laser du formaldéhyde, la chimiluminescence du formaldéhyde excité, la mesure de température par thermocouple, l'analyse chimique par (micro-)chromatographie en phase gaz et spectrométrie de masse. Trois carburants ont été sélectionnés, respectivement le diméthyl éther (DME), le diéthyl éther (DEE) et l'oxyméthylène éther-2 (OME-2). Les flammes froides étudiées sont généralement stabilisées à basse richesse (0.2 < E.R. < 0.5) à l'aide d'ajout d'ozone (variant entre 1 et 2% mol. du prémélange), à l'exception du diéthyl éther, où une flamme froide a été stabilisée à la stœchiométrie sans ozone. L'établissement de structures détaillées de ces flammes froides a notamment permis, lorsque comparées aux prédictions de divers modèles cinétiques, de constater les manques de certains de ces modèles quant à la prédiction de la réactivité dans ces conditions spécifiques. De plus, l'influence de l'ozone comme générateur de radicaux in-situ a été étudié. Bien que son utilisation n'affecte que marginalement la composition chimique après la flamme froide, son influence est importante sur certains branchements de réactions primordiaux à basses températures, favorisant les réactions de propagation par rapport à celles de branchement.Finalement, l'utilisation de la Vélocimétrie par Imagerie de Particules (PIV - Particle Image Velocimetry) dans le brûleur à plan de stagnation a permis, pour la première fois, la mesure de la vitesse de propagation de flammes froides à pression atmosphérique. Cette donnée, intrinsèque au carburant étudié, dépendant des composition, température et pression du mélange initial, est particulièrement importante pour la modélisation cinétique de systèmes réactifs. L'utilisation de flammes froides de DME ensemencées à l'ozone a permis sa mesure à basse richesse, les vitesses de propagation mesurées variant entre 14 et 18 cm.s-1. Cette étude a notamment permis de mettre en lumière les réactions les plus importantes pour la prédiction d'une telle grandeur.Les nouvelles données expérimentales obtenues dans le cadre de ces travaux de thèse constituent les fondations de travaux plus larges sur l'influence de l'ozone sur la chimie de basses températures, dans des conditions expérimentales originales et complémentaires de celles généralement étudiées en laboratoire.

In France, High-level nuclear wastes are vitrified. The resulting packages will eventually be stored in a deep geological formation. After corrosion of the canister and overpack, the groundwater will interact with the glass, resulting in the mobilization of soluble radionuclides. Therefore, researchers must design the glass with an optimal composition and durability to limit the radionuclides release in the environment. Al is an important constituent of nuclear waste glasses; however, its role on glass durability remains somewhat of a mystery! During the initial dissolution regime, at low concentration in the glass, it increases the resistance to hydrolysis of the glass, whereas at high concentration, an opposite effect is observed. To understand this phenomenon, we performed a large number of Molecular Dynamics simulations coupled with the potential mean force technique (PMF) to estimate the activation barriers for hydrolysis of cation-oxygen bonds in pure silica and in alumino silicate glasses. In parallel, experiments were performed to validate the simulation results. This study revealed that the Al-O bonds are easier to dissociate than the Si-O bonds. However, Al increases the glass chemical durability by significantly increasing both the strength of the Si-O bonds and network connectivity in alumino silicate glass.In contrast, at high Al concentration, preferential dissolution of Al weakens the silicate network because of the percolation of the weak Al-O bonds, so the glass resistance becomes poor. Through PMF calculations, we evaluated the activation barriers for dissociating bonds around Al as 0.49 eV, which is less than half of the energy to dissociate bonds around Si in pure silicate (1.22 eV) and around Si in aluminosilicate glass (1.34 eV). We have shown that these energy differences are statistically significant. Molecular structural investigation revealed that Si with Al as a second neighbor in the glass network has significantly higher activation energy for dissociation than Si in pure silicate glass. Then the strengthening of the Si-O bonds when Al is present in the glass is mainly a local effect. The role of Al in strengthening the Si is qualitatively validated by ReaxFF calculations, another molecular modelling technique but more precise. Then we have continued the investigations with glasses containing Ca or Na. Mixing both Ca and Na in aluminosilicate glass made the strengthening effect of Si by Al insignificant because the strengthening effect is compensated by an increase of the structural disorder.Finally we have investigated the role of Al during the residual alteration rate. For this, we designed six sodium borosilicate glasses with varying concentrations of Al2O3 from 0% up to 9.6 wt%. Three glasses have small to no Al content, and the three other glasses have higher Al contents. It is observed that glasses with small quantities of Al2O3 quickly released a large amount of Si into the solution. On the other hand, glasses with a larger quantity of Al2O3 released less amount of Si. This is because the Al content controls Si release into the solution under dilute conditions. Eventually, the solution is saturated with Si, and the glass develops a protective layer on its surface.A new Monte Carlo model was developed to reproduce and better understand these experimental results. Glasses releasing more Si into the solution favored the formation of a thick Si-enriched layer on the gel surface in contact with water. By investigating experimentally the alteration layer of one of the six glasses by ToF-SIMS, we observed that B begins to be retained in the young gel developed during one month. This retention is correlated to the decrease of the residual alteration rate. By coupling Monte Carlo simulations and experiments, it is proposed that the formation of an external layer enriched in Si induces B retention in the gel, which, in turn, could limit further glass alteration.; En France, les déchets nucléaires de haute activité sont vitrifiés. Ils seront à terme stockés en formation géologique profonde. Après corrosion des enveloppes métalliques, l’eau souterraine arrivera au contact du verre et interagira chimiquement avec ce dernier, conduisant à la mobilisation des radionucléides solubles. Il est donc important de définir des verres avec des compositions et des durabilités optimales pour limiter ce relâchement. L’aluminium est un constituant majeur des verres nucléaires. Mais son rôle sur l’altération est encore mal compris. Pendant la phase de régime initial, à faible concentration, il conduit à accroître la résistance du verre à l’hydrolyse, alors qu’à plus forte concentration, un effet opposé est observé. Pour comprendre ce phénomène, nous avons réalisé un grand nombre de simulations de dynamique moléculaire, couplées à la technique de “Potential Mean Force” (PMF) pour estimer les barrières d’activation pour l’hydrolyse dans la silice pure et dans des verres d’alumino-silicates. En parallèle, des expériences ont été faites pour valider les simulations. Cette étude a révélé que les liaisons Al-O sont aisément dissociables comparées aux liaisons Si-O. Cependant la présence de Al augmente significativement la durabilité du verre en renforçant les liaisons Si-O et en augmentant la connectivité du réseau vitreux.Au contraire, à plus forte concentration, la dissolution préférentielle des Al affaiblit le réseau silicaté en raison de la percolation des liaisons faibles Al-O. La résistance du verre à l’altération s’affaiblit. Grâce à la technique PMF, nous avons évaluer à 0.49eV l’énergie d’activation moyenne pour l’hydrolyse des liaisons autour des Al, ce qui représente moins de la moitié de l’énergie nécessaire pour dissocier les liaisons autour des Si dans la silice pure (1.22eV) ou dans un verre alumino-silicaté (1.34eV). Nous avons montré que ces différences d’énergie sont statistiquement significatives. Cet effet est un processus principalement local. Des calculs de dynamique moléculaire par la méthode ReaxFF ont permis de valider les résultats. Nous avons poursuivi les investigations avec des verres contenant du Ca et du Na. Le mélange Ca/Na dans un verre alumino-silicaté conduit à rendre insignifiant le renforcement des liaisons Si-O par le Al du fait de la compensation de cet effet par l’accroissement du désordre structural.Finalement, nous avons étudié le rôle du Al pendant la phase d’altération résiduelle, c’est-à-dire quand un gel passivant se forme. Pour cela, six verres d’alumino-borosilicate avec des teneurs variables en Al2O3 (de 0% à 9.6% massiques) ont été préparés. Trois verres contiennent une concentration en Al2O3 faible voire nulle et les trois autres ont des concentrations en Al2O3 plus élevées. Les verres avec de faibles quantités de Al2O3 relâchent rapidement une grande quantité de Si en solution. En revanche, les verres avec une quantité plus importante de Al2O3 relâchent moins de Si. Ceci confirme le contrôle du relâchement des Si en solution en conditions diluées par le Al. Au final, quand la solution devient saturée en Si, une couche protectrice se forme à la surface du verre.Un nouveau modèle Monte Carlo a été développé pour reproduire et mieux comprendre ces résultats expérimentaux. Les verres relâchant davantage de Si en solution forment une couche épaisse enrichie en Si à la surface du gel en contact avec l’eau. En analysant expérimentalement le couche d’altération de l’un des six verres par Tof-SIMS, nous avons observé que le B, élément soluble et généralement traceur de l’altération est partiellement retenu dans le gel formé après un mois de réaction. Cette rétention est corrélée à la décroissance de la vitesse d’altération. En couplant les simulations Monte Carlo et les expériences, il a été proposé que la formation d’une couche externe enrichie en Si conduit à la rétention du B dans le gel, ce qui, à son tour, pourrait limiter la poursuite de l’altération.

The formation of soot particles in combustion is a major research issue due to their negative impact on our health and our environment. The link between the inhalation of these particles and certain pathologies affecting mainly the respiratory tract is indeed well established and it is also proven that these particles have an impact on climate change. For these reasons, their emission rate from devices involving combustion processes (aircraft engines, thermal or hybrid engines of land or sea vehicles, wood boilers...) is increasingly regulated. However, the state of the art concerning the formation mechanisms of soot particles reveals many deficiencies, notably regarding the nature of the chemical species and reaction pathways involved. The nucleation step, which is the crucial step of these mechanisms since it corresponds to the transformation of gaseous precursors into soot particles, raises many questions and interrogations. Currently, several hypotheses are considered in the literature involving in particular the formation of polycyclic aromatic hydrocarbons (PAHs) and potentially that of intermediate species (PAH dimers, resonantly stabilized radicals of PAHs) between PAHs and soot particles. However, there is very little direct experimental evidence in the literature on the formation of these compounds in flames.The work carried out in this thesis focused on the characterization of the key species involved in the soot nucleation process in a methane diffusion flame. In that context, we have implemented a set of in situ and ex situ experimental methodologies (ToF-SIMS, Raman, LII, LIF and EPR) which have allowed us to obtain a large panel of experimental data able to provide relevant and innovative information on the main molecular species and reaction pathways involved in the nucleation process. The coupling of these different methods and the analysis of the obtained data, in particular by means of multivariate data analysis, reveals important correlations between the nucleation process and the successive formation of different species (PAHs, resonance stabilized radical of PAHs, incipient soot particles, mature particles). This work brings crucial information on the size, mass, structure and nature of the species involved and their evolution in the nucleation process, which will allow the development of kinetic models and significantly improve our knowledge of this complex physicochemical process.; La formation des particules de suie dans les processus de combustion est une problématique de recherche majeure du fait de l'impact négatif de ces composés sur notre santé et notre environnement. Le lien entre l'ingestion de ces particules et certaines pathologies touchant principalement les voies respiratoires est en effet bien établi et il est également avéré que ces particules ont un impact sur le changement climatique. C'est pourquoi leur taux d'émission au sortir de dispositifs mettant en jeu des processus de combustion (réacteurs d'avions, moteurs thermiques ou hybrides des véhicules terrestres ou maritimes, chaudières à bois…) est de plus en plus règlementé. Or, l'état de l'art concernant ces mécanismes chimiques fait apparaître de nombreuses carences notamment dans la nature des espèces chimiques impliqués et des voies réactionnels mises en jeu. Plus particulièrement, l'étape de nucléation, qui est l'étape cruciale de ces mécanismes puisqu'elle correspond à la transformation des précurseurs gazeux en particules de suie, soulèvent de nombreuses questions et interrogations. Actuellement, plusieurs hypothèses sont envisagées dans la littérature mettant notamment en jeu la formation d'hydrocarbures aromatiques polycycliques (HAPs) et potentiellement celle d'espèces intermédiaires (dimères de HAPs, HAPs radicalaires stabilisés par résonance) entre les HAPs et les particules. Cependant, il n'existe que très peu de mesures directes ou de preuves expérimentales de la formation de ces composés dans les flammes.Le travail de cette thèse a porté sur la mise en évidence et la caractérisation des espèces clefs impliquées dans le processus de nucléation des particules de suie dans une flamme de diffusion de méthane. Dans le cadre cette étude, nous avons mis en œuvre un ensemble de dispositifs expérimentaux in situ et ex situ (ToF-SIMS, Raman, LII, LIF et RPE) qui ont permis l'obtention d'un large panel de données expérimentales à même d'apporter des informations pertinentes et novatrices sur les principales espèces moléculaires et voies réactionnelles impliquées dans le processus de nucléation. Le couplage de ces différentes méthodes et l'analyse des données obtenues, notamment au moyen de la méthode PCA, fait apparaître un séquençage du processus de nucléation corrélé à la formation successive de différentes espèces (HAPs, HAPs radicalaires stabilisés par résonance, particules de suie naissantes, particules matures). Ce travail apporte des informations cruciales sur la taille, la masse, la structure et la nature des espèces impliquées et leur évolution dans le processus dans le processus de nucléation, de nature à permettre le développement de modèles cinétiques et améliorer sensiblement nos connaissances de ce processus physicochimique complexe.

This work uses small angle neutron scattering and electrochemical characterizations to study the structure and properties of ionomer in proton exchange membrane fuel cell (PEMFC) electrodes. The final objective of this thesis is to establish the link between ionomer degradation and the decrease of electrochemical performances. Small Angle Neutron Scattering (SANS) allows to study the electrode structure for characteristic sizes ranging from the nanometre to several hundreds of nanometres. The analysis method developed in this thesis gives access to the distribution of the ionomer, the size of the primary carbon particles and their porous structure as well as the adsorption and localization of water in the electrode. The analysis of SANS profiles shows that the ionomer disperses as a thin film of 1.5 to 2.5 nm thickness on the surface of the Pt/C catalyst, or forms aggregates (>50 nm) depending on the electrode fabrication process. Electrochemical characterizations are developed to study the role of the ionomer in the electrode operation. Measurements of performance, electrochemical impedance spectroscopy, double layer capacitance, CO oxidation and proton desorption are carried out. These measurements underline the importance of adsorbed water on the surface of the Pt/C catalyst for proton transport and show that the ionomer allows to keep water in the vicinity of active sites at low relative humidity. However, ionomer limits the accessibility of platinum to the gas, or even prevents it completely. SANS and electrochemical measurements are applied to new and aged electrodes to study the impact of aging on the properties and structure of the ionomer. The degradation of protonic conductivity in the electrode contributes to the drop in performance and seems to be related to the degradation of the chemical structure of the ionomer.; Ce travail de thèse utilise la diffusion de neutrons aux petits angles et les caractérisations électrochimiques pour étudier la structure et les propriétés du ionomère dans les électrodes de pile à combustible à membrane échangeuse de protons (PEMFC). L'objectif final de cette thèse est de faire le lien entre la dégradation du ionomère et la diminution des performances électrochimiques. La diffusions de neutrons aux petits angles, ou SANS pour Small Angle Neutron Scattering, permet d'étudier la structure de l'électrode pour des tailles caractéristiques allant du nanomètre à plusieurs centaines de nanomètres. La méthode d'analyse développée dans cette thèse permet d'accéder à la distribution du ionomère, aux taille des particules primaires de carbone et à leurs porosités ainsi qu'à l'adsorption et à la localisation de l'eau dans l'électrode. L'analyse des profils met en évidence que le ionomère se disperse en film fin de 1,5 à 2,5 nm d'épaisseur à la surface du catalyseur Pt/C, ou forme des agrégats (>50 nm) en fonction du procédé de fabrication des électrodes. Des caractérisations électrochimiques sont développées pour étudier le rôle du ionomère dans le fonctionnement de l'électrode. Des mesures de performances, de spectroscopie d'impédances électrochimiques, de capacité de double couche, d'oxydation de CO et de désorptions de protons sont mises en œuvres. Ces mesures soulignent l'importance de l'eau adsorbé à la surface du catalyseur Pt/C pour le transport des protons et montrent que le ionomère permet de conserver de l'eau à proximité des sites actifs aux faibles humidités relatives mais qu'il limite l'accessibilité du platine au gaz, voir l'empêche complétement. Les mesures SANS et électrochimiques sont appliquées à des électrodes neuves et vieillis pour étudier l'impact du vieillissement sur les propriétés et la structure du ionomère. La dégradation de la conduction protonique dans l'électrode contribue à la chute des performances et semble reliée à la dégradation de la structure chimique du ionomère.

International audience

The multiple aims of this thesis fall within the implementation of a new HT-SMART-FORMU technological platform dedicated to formulation science. The focus is set on the development of reliable experimental methods, theoretical and predictive tools, so as to establish relations between chemical composition, physicochemical properties and applicative properties.The first type of systems investigated in this work consists of solid dispersions. Their stability has long been approached by the DLVO theory, but more recent studies suggest the use of Hansen Solubility Parameters to describe their stability in non-aqueous media. In the first chapter, a systematic analytical method based on zeta potential measurement and light scattering stability monitoring allows deducing a complementarity of both theories to describe the stability of TiO2 dispersions in non-aqueous solvents.The problematic of oil hydrophobicity prediction through EACN values modelling, which is a key parameter to design surfactant/oil/water (SOW) dispersed systems, is addressed in the second chapter. Two mathematical models, designed using machine-learning methods, are proposed for the rapid prediction of the EACN of oils, namely Neural Networks (NN) and Graph Machines (GM). While the GM model is implemented from the SMILES codes, the NN model is fed with σ-moments descriptors computed with the COSMOtherm software. The prediction reliability of both models is discussed based on a complex 10-molecule test set.In chapters 3 and 4, the scope of applicative properties of a nonionic glycerol-based surfactant are investigated. Firstly, chapter 3 focuses on its aggregation behavior in aqueous solutions and the formation of liquid crystals (LC) at low concentration, in comparison with the benchmark polyethoxylated fatty alcohols and alkyl polyglucosides. The influence of its physicochemical properties, in particular the air / water interface dilational viscoelasticity, is put in relation with the observed poor foamability and long-lasting foam stability.Finally, C12Gly2 properties in SOW systems are investigated in chapter 4. The Normalized Hydrophilic-Lipophilic Deviation (HLDN), a powerful theoretical tool, is regarded as a way to rationalize the characteristics of both emulsions and microemulsions. In this way, a thorough quantification of surfactants amphiphilicity, temperature sensitivity and salt-tolerance are presented. The potential use of C12Gly2 as O/W and W/O emulsifier is then investigated: the granulometry and stability of emulsions obtained by varying the oil concurs with HLDN values. A minimum is observed at HLDN = 0 and increases for negative and positive HLDN values.; Les objectifs de la thèse sont multiples et s'inscrivent dans le cadre de la mise en place d'une nouvelle plateforme technologique HT-SMART-FORMU dédiée à la formulation. L'accent est mis sur le développement de méthodes expérimentales fiables et d'outils théoriques et prédictifs, afin d'établir des relations entre la composition chimique, les propriétés physicochimiques et les propriétés applicatives.Les dispersions de particules sont le premier type de systèmes étudiés. Leur stabilité a longtemps été abordée par la théorie DLVO, mais des études plus récentes suggèrent l'utilisation des Paramètres de Solubilité de Hansen pour décrire leur stabilité en milieu non aqueux. Dans le premier chapitre, une méthode analytique systématique basée sur la mesure du potentiel zêta et le suivi de la stabilité par diffusion de la lumière permet de déduire une complémentarité des deux théories pour décrire la stabilité des dispersions de TiO2 en solvants non aqueux.La problématique de la prédiction de l'hydrophobie des huiles par la modélisation d'EACN, qui est un paramètre clé pour concevoir des systèmes dispersés tensioactif/huile/eau (SOW), est abordée dans le deuxième chapitre. Deux modèles mathématiques, conçus à l'aide de méthodes de machine-learning, sont proposés pour la prédiction rapide de l'EACN des huiles, à savoir les réseaux de neurones (NN) et les machines graphiques (GM). Alors que le modèle GM est implémenté à partir des codes SMILES, le modèle NN est alimenté par des descripteurs σ-moments calculés avec le logiciel COSMOtherm. La fiabilité des prédictions des deux modèles est discutée sur la base d'un ensemble de test de 10 molécules complexes.Dans les chapitres 3 et 4, l'étendue des propriétés applicatives d'un tensioactif non ionique à base de glycérol (C12Gly2) est étudiée. Le chapitre 3 se porte sur son agrégation en solution aqueuse et la formation de cristaux liquides (CL) à faible concentration, en comparaison avec les alcools gras polyéthoxylés et les alkyl polyglucosides. L'influence de ses propriétés physico-chimiques, en particulier la viscoélasticité de dilatation de l'interface air/eau, est mise en relation avec la faible moussabilité et la stabilité de mousse durable observées.Enfin, les propriétés du C12Gly2 en systèmes SOW sont étudiées dans le chapitre 4. La déviation hydrophile-lipophile normalisé (HLDN), un outil théorique puissant, est considéré comme un moyen de rationaliser les caractéristiques des émulsions et des microémulsions. Ainsi, une quantification approfondie de l'amphiphilie des tensioactifs, de leur sensibilité à la température et de leur tolérance au sel est présentée. L'utilisation du C12Gly2 comme émulsifiant H/E et E/H est ensuite étudiée : la granulométrie et la stabilité des émulsions obtenues en faisant varier l'huile concordent avec les valeurs du HLDN. Un minimum est observé à HLDN = 0, puis la granulométrie et la stabilité augmentent pour des valeurs de HLDN négatives et positives.

The grain-size control of aluminum alloys ensures good mechanical properties. In presence of are inforcing phase such as TiB2, these properties are directly influenced by the coherent interfaces respecting specific orientation relationships. In this context, the objectives of our work were (i) to achieve a sufficiently reliable description of the Al|TiB2 interfaces including an evaluation of their energies, (ii) to specify the bulk characteristics of the various Al-B-Ti phases which can influence the interfacial properties. The understanding of the underlying mechanisms being difficult by experimental techniques at atomic scale, we have opted for an approach based on numerical simulations at this scale, mainly using ab initio methods.The first part of this work deals with the study of the properties of heterophase Al|TiB2 interfaces. By focusing on basal Al|TiB2 interfaces, our analysis allowed to clarify various uncertainties found in the bibliography. Firstly, we have taken care to rigorously consider the role of elasticity, whose effect has been mostly neglected in previous works on Al|TiB2. Moreover, another critical point concerns the uncertainties of these previously calculated interface energies, due to the insufficient knowledge of the chemical potentials of the elements present in the involved ordered phases. We have also clarified this point by a rigorous analysis ensuring the stability of TiB2 and the absence of other undesirable phases. Finally, the comparison of the different systems used for the simulation of heterophase interfaces allowed us to conclude that periodic boundary conditions should be preferred, since free surface boundary conditions, mostly used in previous works, lead to less reliable results. Under these conditions, the interface energies obtained using our methodology recover a satisfactory agreement with the range of values expected for coherent interfaces. Our approach could be fruitfully applied to other systems currently suffering from the same deficiencies (Al|SiO2, Mg|TiB2, Al|Mg5Si6).In the second part of this thesis, we have investigated the bulk properties of the various Al-B-Ti phases relevant for the previous interfaces, in particular TiB2 and AlB2. To this end, we have determined the thermodynamic properties of these phases via the IPDA (independent point defect approximation). We first performed a detailed study of the TiB2 compound, including the effects of temperature, interstitials, and aluminum. We have shown that Ti and B vacancies are the major defects, that interstitials have no effect on the thermodynamic properties of this compound, and that Al is present in substitution in TiB2, this presence being limited to B-rich compounds. After completing this study of metal borides by a brief analysis of the solid solution of Ti and B in Al, we were able to partially build the Al-B-Ti ternary phase diagram at 1000 K, in particular in the two-phase domain (Al)+TiB2. In this domain, we have highlighted a characteristic "double fan" tie-line structure, thus providing complementary information to those previously obtained byphenomenological thermodynamic simulations (CALPHAD approach).; Le contrôle granulométrique des alliages d'aluminium permet de garantir de bonnes propriétés mécaniques. En présence d'une phase de renfort telle que TiB2, ces propriétés sont directement influencées par les interfaces cohérentes respectant certaines relations d'orientations. Dans ce contexte, les objectifs de notre travail étaient (i) de parvenir à une description suffisamment fiable des interfaces Al|TiB2 incluant une évaluation de leurs énergies, (ii) de préciser les caractéristiques volumiques des phases Al-B-Ti qui peuvent influencer les propriétés interfaciales. La compréhension des mécanismes sous-jacents restant difficile par les techniques expérimentales à l'échelle atomique, nous avons opté pour une approche reposant sur des simulations numériques à cette échelle, essentiellement avec des calculs ab initio.La première partie de ce travail porte sur l'étude des propriétés des interfaces hétérophases Al|TiB2. En se concentrant sur les interfaces basales Al|TiB2, notre analyse a permis de clarifier diverses ambiguïtés présentes dans la bibliographie. Tout d'abord, nous avons pris soin de prendre rigoureusement en compte le rôle de l'élasticité, dont l'effet a été le plus souvent négligé dans les travaux antérieurs sur Al|TiB2. De plus, un autre point critique concerne les incertitudes des énergies d'interfaces ainsi calculées, en raison de la connaissance insuffisante des potentiels chimiques des éléments présents dans les phases ordonnées impliquées. Nous avons également clarifié ce point par une analyse rigoureuse assurant la stabilité de TiB2 et l'absence d'autres phases indésirables. Enfin, la comparaison des divers systèmes utilisés pour la simulation des interfaces hétérophases a permis de conclure que les conditions aux limites périodiques doivent être préférées, puisque les conditions aux limites en surfaces libres, surtout utilisées dans les travaux précédents, ont conduit à des résultats moins fiables. Dans ces conditions, les énergies d'interfaces obtenues à l'aide de notre méthodologie retrouvent un accord satisfaisant avec la gamme de valeurs attendue pour des interfaces cohérentes. Notre approche pourrait être appliquée avec fruit à d'autres systèmessouffrant actuellement des mêmes déficiences (Al|SiO2, Mg|TiB2, Al|Mg5Si6).Dans la deuxième partie de ce mémoire, nous avons étudié les propriétés volumiques des phases Al-B-Ti pertinentes pour les interfaces précédentes, notamment TiB2 et AlB2. À cette fin, nous avons déterminé les propriétés thermodynamiques de ces phases via l'ADPI (approximation des défauts ponctuels indépendants). Nous avons d'abord réalisé une étude détaillée du composé TiB2, incluant les effets de la température, des interstitiels et de l'aluminium. Nous avons montré que les lacunes de Ti et de B sont les défauts majoritaires, que les interstitiels n'ont aucun effet sur les propriétés thermodynamiques de ce composé, et que Al est présent en substitution dans TiB2, cette présence étant limitée aux composés riches en B. Après avoir complété cette étude des borures métalliques par une brève analyse de la solution solide de Ti et B dans Al, nous avons pu construirepartiellement le diagramme de phases ternaire Al-B-Ti à 1000 K, notamment dans le domaine biphasé (Al)+TiB2. Dans ce domaine, nous avons mis en évidence une structure caractéristique de conodes en "double éventail", fournissant ainsi des informations complémentaires à celles obtenues précédemment par des simulations thermodynamiques phénoménologiques (approche CALPHAD).

A semi-automated and fully integrated millifluidic platform has been developed for rapid development and study of Solid-phase extraction (SPE) processes. In this instrumented setup, the fluid coming out of the extraction mini-column is analysed directly using on-line X-ray fluorescence, enabling a real-time, fast and accurate measurement of the elemental concentrations within the flowing liquid. This platform aims at enhancing and facilitating the development of SPEs whose numerous experimental parameters make it time consuming to optimize and upscale. Hence, this approach solves the labour intensity issue and is as accurate as off-line Inductively Coupled Plasma, for instance. This is exemplified with a system, typical of the electronic wastes recycling industry, comprising of a mixture of Rare-Earth Elements (REEs) and Iron, whose separation is investigated as a function of various experimental parameters. It enabled us to study the performance of the newly released METALICAPT® commercial materials and to determine an optimized path towards the almost total separation of Iron from the initial mixture. Overall, the original SPE process reported herein demonstrates the platform's reliability, performance and range of applicability. National Environmental Agency (NEA) National Research Foundation (NRF) Submitted/Accepted version FO, SC and J.C.G. acknowledge financial support from SCARCE project, which is supported by the National Research Foundation, Singapore, and National Environment Agency, Singapore under its Closing the Waste Loop Funding Initiative (Award No. USS-IF-2018-4). J.C.G. acknowledges funding for the platform from the European Research Council under the European Union’s 7th Framework Program (FP/2007-2013)/ERC Grant Agreement N◦ [320915] “REE-CYCLE”: Rare Earth Element reCYCling with Low harmful Emissions. BK ac- knowledges financial support from the European Research Council under the European Union’s 7th Framework Program (FP/2007-2013)/ ERC Proof of Concept Grant Agreement N◦ [768722] “SYNEXTRACT”: All Aqueous, continuous, Solid-Liquid SYNergistic EXTRACTion of rare earth elements. FO thanks CEA’s program “Mat ́eriaux et Proc ́ed ́es” for a PhD grant support (project MICROPRO).

For several decades, the race towards the miniaturization and the complexification of electronic functions have led to the integration of innovative materials to maintain the required performance in terms of response time and power consumption. The development of biosensors which is the subject of an intensive research activity does not make exception to the rule. Indeed, the coupling of chemical functionalization and nanomaterials (Si nanowires, MoS2 for example) open exciting new opportunities for the realization of miniature sensors, ultrasensitive and selective, used for the detection of chemical or biological species. In this context, the project of this thesis aims the development of a bio-detection platform through a multidisciplinary approach. Two applications are targeted: the measurement of pH and the detection of the hybridization of two complementary DNA strands. The demonstration of a first proof of concept based on innovative materials will allow a better understanding of the contribution of new materials (the duo transducer / bio-receiver), with the objective to improve our biosensors features, such as selectivity, sensitivity, stability, response time, and the ease of use.; Depuis plusieurs dizaines d'années, la course vers la miniaturisation et la complexification des fonctions électroniques suscitent l'intégration de matériaux innovants pour conserver les performances requises en termes de temps de réponse et de consommation électrique. Le développement des biocapteurs qui fait l'objet d'une activité de recherche intensive n'échappe pas à la règle. En effet, le couplage de la fonctionnalisation chimique et des nanomatériaux (nanofils de Si, MoS2 par exemple) ouvre de nouvelles perspectives enthousiasmantes pour la réalisation de capteurs miniatures, très sensibles et sélectifs pour la détection d'espèces chimiques ou biologiques. Dans ce contexte, le projet de cette thèse vise, par une approche pluridisciplinaire, le développement d'une plateforme de bio-détection. Deux applications sont ciblées : la mesure de pH et la détection de l'hybridation de deux brins complémentaires d'ADN. La démonstration d'une première preuve de concept à partir de matériaux innovants permettra de mieux comprendre l'apport des nouveaux matériaux (couple transducteur/bio-récepteur) pour améliorer la sélectivité, la sensibilité, la stabilité, la rapidité, et la simplicité d'utilisation des biocapteurs.

Striving towards a circular economy has led to the re-investigation of many existing processes, with the target of developing more sustainable variants. In our present economy, plastics form an important and omnipresent material affecting our daily lives. They are inexpensive, durable, corrosion resistant, and light weight leading to their use in a wide variety of applications.Within the plastic chemical recycling scheme, Fischer-Tropsch synthesis (FTS) could play a key role as the syngas feedstock that is converted in it, can be generated via the gasification of the considered plastics. This syngas is then chemo-catalytically converted into hydrocarbons such as paraffins and light olefins. Typical FTS catalysts are based on supported cobalt or iron species.Among the mechanistic kinetic models, the comprehensive variant based on the Single Event MicroKinetics (SEMK) concept has been widely applied in the field of oligomerization, autoxidative curing, etc. and has proven to be a versatile tool to simulate Fischer-Tropsch synthesis. However, developing mechanistic models for every chemical engineering challenge is not always feasible due to their complexity and the in-depth knowledge required to build such models.A detailed evaluation on the potential of using machine learning approaches to match the performance of results obtained using the Single-Event MicroKinetic model was carried out. Initially, the focus was on a single dominant output scenario (methane selective catalyst). The current work thus shows that more widely applied techniques in data science can now be applied for systematic analysis and interpretation of kinetic data. Similar analysis using experimental data can also help experimenters in their preliminary analysis, to detect hidden trends in the data, and thus to identify importance features. After gaining confidence on the investigated interpretation techniques, for the FTS reaction with single dominant output, a similar investigation on the potential of iron based catalysts with enhanced light olefin selectivity is carried out next.; En vue de promouvoir l’économie circulaire, de nombreux procédés chimiques sont actuellement réexaminés afin de développer des variantes plus durables. Cela a mené à une forte augmentation de production au cours des 60 dernières années, entraînant une production totale de 367 millions de tonnes en 2020.La méthodologie a ensuite été généralisée à l’aide d’apprentissage automatique non supervisé, ce qui a permis de dépasser les trois dimensions et de réduire le besoin d’intervention humaine. L’espace des descripteurs généré à partir de données microcinétiques (catalyseurs virtuels) est exploré en utilisant la méthode systématique du regroupement (clustering) et du classement (labelling) non supervisés. L’espace de la performance du catalyseur est regroupé en clusters et le nombre minimale de clusters est identifié. Chaque catalyseur virtuel (représenté par une certaine combinaison de descripteurs) est identifié du point de vue du cluster auquel il appartient. Il est ainsi possible d’obtenir l’étendue des valeurs des descripteurs dans le cluster ayant le meilleur rendement d’alcènes légers. Il est observé que les valeurs obtenues sont conformes à celles du catalyseur virtuel optimal identifié dans l’inspection visuelle précédente. On peut donc conclure qu’une méthode combinant la microcinétique et l’apprentissage automatique a été présentée pour le développement des catalyseurs et pour l’investigation détaillée de leurs propriétés, tout en diminuant le besoin d’intervention humaine.Finalement, la méthode d’apprentissage automatique a été étendue dans l’intention de pouvoir réaliser des prédictions de plusieurs sélectivités en se concentrant sur la production des alcènes légers à différentes conditions opérationnelles. Afin d’atteindre cet objectif, 4 modèles d’apprentissage automatique alternatifs ont été employés, i.e. la méthode lasso (lasso regression), la méthode des k plus proches voisins (k nearest neighbor regression ou KNN), la méthode de la machine à vecteurs de support (support vector machine regression ou SVR) et le réseau de neurones artificiels (Artificial Neural Network ou ANN). Les capacités de ces techniques sont évaluées par rapport à la reproduction du comportement linéaire de la conversion et la sélectivité en fonction des variables du procédé comme cela a été simulé par le modèle SEMK. Il est constaté que les modèles à base d’un réseau de neurones artificiels correspondent le plus aux résultats de référence du modèle SEMK. Une analyse supplémentaire utilisant la technique d’interprétation de la valeur SHAP a été appliquée aux modèles à base d’un réseau de neurones artificiels ayant la meilleure performance, en vue de mieux expliquer le fonctionnement des modèles.L’ensemble de l’étude a rapporté des connaissances essentielles, telles que les descripteurs de catalyseur optimales : les enthalpies de chimisorption atomique de l’hydrogène (QH ≈ 234 kJ/mol), du carbone (QC ≈ 622 kJ/mol) et de l’oxygène (QO ≈ 575 kJ/mol), pour la conception de catalyseurs ayant une sélectivité en alcènes légers élevée en utilisant un modèle SEMK mécaniste. De plus, l’étendue des conditions opérationnelles menant à la meilleure sélectivité en alcènes légers a été déterminée en adoptant plusieurs stratégies de modélisation (le concept des SEMK et l’apprentissage automatique). Il a été constaté que les effets de la température (580-620K) et la pression (1-2 bar) étaient les plus importants. Ensuite, une investigation est réalisée dans le but d’évaluer à quel point les résultats des modèles d’apprentissage automatique correspondent à ceux du modèle SEMK. Par exemple, une analyse préliminaire a pu être réalisée en utilisant un modèle d’apprentissage automatique pour l’analyse des données obtenues à l’aide d’expérimentation à haut débit. Ensuite, le modèle mécaniste a permis d’acquérir une compréhension chimique approfondie

International audience

Several hybridization strategies of a solar-autothermal biomass gasifier were examined for stable andcontinuous operation under variable solar irradiation. The ultimate objective was to demonstrate thefeasibility of controlled syngas production, through the modification of oxygen, water, and biomassinjection rates. Various hybridization strategies were probed by thermodynamic analysis andexperimentally validated. Thermodynamic equilibrium calculations detailed the impact of both H2Oand O2 injection rates on the produced syngas composition under constant wood feeding. Oxygeninjection decreased the H2:CO molar ratio, while reducing the solar thermal power required to carryout the gasification reaction. Meanwhile, the total H2+CO production dropped by 1.36 mole of H2 and0.64 mole of CO per mole of O2 added, independently of the quantity of water provided. Validationexperiments were then carried out under real concentrated solar flux in a directly-irradiated conicalspouted-bed reactor, following distinct hybridization paths. Maintaining constant the H2:CO ratioabove 1 during hybridization required to provide high amounts of water steam with oxygen, whichpenalized the gasifier efficient heating. In contrast, minimizing the water injection rate throughouthybridizationstrongly altered the H2:CO ratio but decreased the CO2 production and the solar thermal power requirement. Finally, the successful control of the outlet H2+CO volume flow rate with simultaneous oxygen and wood injection was demonstrated (under constant water feeding rate). Solarto- fuel efficiencies were kept around 20%, while hybridization decreased the cold-gas efficiency below 80%.

International audience; High energy consumption during harvesting is one of the main bottlenecks for sustainable microalgae production. Membranes can efficiently separate microalgae from liquids with low energy consumption, but membrane fouling remains an important issue. Flocculation prior to membrane filtration can increase membrane fluxes and decrease fouling, thus offering a low-cost and efficient solution to harvest microalgae. Biobased cationic cellulose nanocrystals were successfully used as flocculants for microalgae and were effective over a wide pH-range and for both freshwater and marine microalgae. Such flocculation was for the first time combined with vibration-assisted filtration using a charged, surface patterned membrane, enabling operation at very high flux (95 L/m2 h) using a vibration frequency of only 1 Hz, and even under sub-optimal flocculation conditions. Intermittent vibration decreased energy consumption further while keeping excellent filtration performance to finally achieve a record-low energy consumption for the membrane filtration of only 6.7 Wh/m3, which is andgt;25-times lower than that of normal membrane filtration. Interaction forces revealed that increasing particle size through flocculation prior to membrane filtration can significantly prevent microalgae attachment on the vibrating membrane surface. This work opens a new direction for sustainable microalgae harvesting with an ultra-low energy consumption, combined with a very high microalgae recovery, reduced use of chemicals, and lower membrane investment cost.

International audience; The main obstacle to the aqueous carbonation of non-serpentinised magnesium silicates is the formation of surface passivation layers, which severely limits the reaction rate and thus the overall efficiency of the process. A technological solution to overcome this problem is to perform the carbonation process inside a stirred bead mill, which aims to continuously remove the surface by-product layers by attrition. In this work, the aqueous carbonation of ferronickel slag, a mineralogically complex mining waste composed of a Mg/Si rich amorphous phase and a crystalline ferrous forsterite, was studied at 150°C and under 10 bar of CO 2 with different operating configurations: carbonation alone (C mode), attrition followed by carbonation (A-C mode) and concomitant attrition and carbonation (AC mode). By careful observation of the mineralogy and the surface of the secondary phases formed using complementary analytical techniques, the article allows a better understanding of the passivation phenomenon inherent to the carbonation of magnesium silicates, and confirms the effectiveness of continuous surface mechanical depassivation for reaching high carbonation rates with this type of material. Comparative analysis of the products obtained with the three operating modes shows that a true synergy takes place between attrition and carbonation due to the combined effect of continuous exfoliation and mechanical activation of particle surface, which goes far beyond the simple increase in surface area due to particle size reduction. While mechanical depassivation is here substantiated by several evidence, the additional mechanochemical activation effect cannot be delineated from experiment; however its beneficial contribution to carbonation is inferred from its observation in A-C mode. The work finds that the synergy between attrition and carbonation also yields very characteristic products. They consist in micrometric agglomerates formed by bound spherical particles a few tens of nanometers in size. These particles themselves contain an entanglement of nanometric grains of carbonates and amorphous silica dispersed inside a magnesium-depleted alumino-siliceous matrix. These results confirm that concomitant attrition and carbonation offers one of the most promising pathways for developing direct aqueous carbonation processes with non-thermally activatable magnesium silicates.

International audience; We report on a new class of metamaterials, which consists of meta-foams, optimized for high performance in enhanced solar steam generation. The design of these meta-foams involves precise dimensional control of a periodic two-dimensional micro-pore array. Proof-of-concept of meta-foams efficiency is demonstrated using two fabrication technologies. The first one involves microscale plasma etching onto a nanostructured black silicon substrate. As an alternative low-cost technique, we also use 3D printing of a graphene-containing polymer material. The experimental validation shows that the best evaporation rate reaches 1.34 kg/(h·m2) under 1 sun illumination, in open-air conditions under normal temperature and relative humidity conditions of 20 °C and 58%, an unmatched value so far under similar conditions, while the theoretical limit is estimated at 1.5 kg/(h·m2). This corresponding to a 89% conversion efficiency. Experimental data concur well with the predicted performance obtained by numerical simulations. The proposed model simultaneously accounts for both heat and mass transfer, including photothermal conversion and water phase change, to guide the optimization towards the maximization of the evaporation rate, an important figure of merit in many applications, including solar heat harvesting and solar water purification.

International audience

Pollutants of pharmaceutical origin present in water are poorly removed by conventional treatments. In this context, membrane processes are efficient and clean alternative techniques for water depollution. The objective of this thesis was to evaluate the performance of different membrane techniques: nanofiltration (NF), reverse osmosis (RO) and forward osmosis (FO) for the removal of pharmaceutical pollutants (Naproxen, Diclofenac). The three membranes used are made of polyamide of different natures: aromatic for RO and FO membranes and semi-aromatic in the case of the NF membrane. Moreover, for the FO membrane, the active layer of the membrane also contained aquaporins. The evaluation of the new membranes performance showed a much lower rejection rate of naproxen for the NF membrane than for the RO and FO membranes. In addition, during their use, polyamide membranes can be accidentally exposed to sodium hypochlorite used during the cleaning/disinfection steps of the installations. The impact of this exposure on the properties of the membrane materials and their removal performance was evaluated in the case of FO and RO membranes. The membranes were exposed to sodium hypochlorite (NaOCl) solutions at different concentrations, different pH and for different contact times. Surface characterization of the membranes (zeta potential, ATR-FTIR, contact angles) showed a modification of their properties after exposure to NaOCl solutions. Thus, it was highlighted that the contact with NaOCl can lead not only to the modification of the chemical nature of the membrane material but also to the modification of the intermolecular hydrogen bonds network within the polyamide, thus impacting the structure of the membrane material and the retention performances. Nevertheless, depending on the type of membrane (and polyamide), different types of variations have been observed. Finally, it was shown by XPS that, in the case of biomimetic FO membranes, the presence of polymersomes incorporating aquaporins within the polyamide was not impacted by the contact with NaOCl. Thus, the FO membrane retains a good rejection capacity for the molecules studied despite an increase in reverse salt flux.; Les polluants d’origines pharmaceutiques présents dans les eaux présentent une élimination limitée par les traitements conventionnels. Dans ce contexte, les procédés membranaires sont des techniques alternatives de dépollution des eaux efficaces et propres. L’objectif de cette thèse était d’évaluer les performances différentes techniques membranaires : la nanofiltration (NF), l’osmose inverse (RO) et l’osmose directe (FO) pour l’élimination de polluants pharmaceutiques (Naproxène, Diclofénac). Les trois membranes utilisées sont constituées de polyamides de différentes natures : aromatique dans le cas de la RO et la FO et semi-aromatique dans le cas de la NF. De plus, dans le cas de la membrane de FO, la couche active des membranes contenait également des aquaporines. La caractérisation des performances des membranes neuves ont mis en évidence un taux de rejet du naproxène par la membrane de NF bien inférieur à celles des membranes de RO et FO. De plus, lors de leur utilisation, les membranes en polyamide peuvent être exposées accidentellement à l’hypochlorite de sodium utilisé lors de phases de nettoyage/désinfection des installations. L’impact de cette exposition sur les propriétés des matériaux membranaires et leurs performances d’élimination a été évalué dans le cas des membranes de FO et de RO. Les membranes ont été exposées à des solutions d’hypochlorite de sodium (NaOCl) à différentes concentrations, différents pH et pour différents temps de contact. Les caractérisations de surface des membranes (potentiel zêta, ATR-FTIR, angles de contact) ont alors montré une modification de leurs propriétés après exposition aux solutions d’NaOCl. Ainsi, il a été mis en évidence que le contact avec NaOCl peut conduire non seulement à la modification de la nature chimique du matériau membranaire mais également à la modification du réseau de liaisons Hydrogène intermoléculaires au sein du polyamide impactant alors la structure du matériau membranaire et les performances de rétention. Néanmoins, selon le type de membranes (et de polyamide), différents types de variations ont été observées. Enfin, il a été montré par XPS que, dans le cas des membranes biomimétiques de FO, la présence des polymersomes incorporant les aquaporines au sein du polyamide n’était pas impactée par le contact avec NaOCl. Ainsi, la membrane de FO conserve une bonne capacité de rejet des molécules étudiées malgré une augmentation du flux de sel inverse.

Theoretical developments of gas detection devices by Mohamed BENSIFIA. This work is carried out as part of the impulsion Isite FUTURE project “CAPTEUR”, which aims to develop experimentally and theoretically a new generation of reliable and selective gas sensors based on functionalized carbon nanotubes (CNTs) and grahene non-covalently. CNTs and graphene have excellent characteristics for gas detection. The theoretical modeling of such a device involves a precise description of the interaction of the molecules with the surface of the material (CNT, graphene), of the electronic structures, of the charge transfers and therefore of the dopings of the material after functionalization and after gas capture. Two calculation codes based on the DFT are used: CRYSTAL and GAUSSIAN. First, the functionalization was tested for the imidazole molecule which has the advantage of having a limited size but which can present a π-stacking type interaction with carbon nanomaterials. Moreover, the imidazole molecule has a strong interaction with CO2. Therefore, the CNT-imidazole and graphene-imidazole systems were chosen as simple prototypes of CO2 sensors. The effect of imidazole as a functionalizing molecule helping in the detection of CO2 was analyzed. This first study made it possible to set up the theoretical methodology that will be followed in the continuation of this thesis work.Experimentally, graphene and semi-conductor CNTs have been non-covalently grafted by a series of metallic porphyrins and phthalocyanines by our collaborators (F. Bouanis). Characterization methods have been employed to obtain precise chemical, structural and morphological information on these systems which must be efficient in detecting CO and NO gases. DFT calculations were undertaken to reproduce the experimental trends observed for the functionalization of CNT(8,0) semiconductor and graphene by these macrocycles, then for the fixation of CO and NO by these devices. Finally, the functionalization of another type of nanomaterial, MoS2, by a cobalt (II) porphyrin is also under study, Développements théoriques de dispositifs de détection de ga zPar Mohamed BENSIFIA. Ce travail est mené dans le cadre du projet impulsion Isite FUTURE “CAPTEUR”, qui vise à développer expérimentalement et théoriquement une nouvelle génération de capteur de gaz fiables et sélectifs à base de n anotubes de carbone (NTCs) et de graphène fonctionnalisés de façon non-covalente. Les NTCs et le graphène présentent des caractéristiques excellentes pour la détection des gaz. La modélisation théorique d'un tel dispositif implique une description précise de l'interaction des molécules avec la surface du matériau (NTC, graphène) , des structures électroniques, des transferts de charge et donc des dopages du matériau après fonctionnalisation et après capture du gaz. Deux codes de calcul basés sur la DFT sont employés : CRYSTAL et GAUSSIAN. Dans un premier temps, la fonctionnalisation a été testée pour la molécule d’imidazole qui a l’avantage d’avoir une taille limitée mais pouvant présenter une interaction de type π-stacking avec les nanomatériaux carbonés. De plus, la molécule d’imidazole a une interaction forte avec le CO2. Donc, les systèmes NTC-imidazole et graphène-imidazole ont été choisis comme de prototypes simples de capteurs du CO2. L 'effet de l'imidazole comme molécule fonctionnalisante aidant à la détection du CO2 a été analysé. Cette première étude a permis de mettre en place la méthodologie théorique qui sera suivie dans le reste de ce travail de thèse.Expérimentalement, le graphène et des NTCs semi-conducteurs ont été greffés de façon non-covalente par une série de porphyrines et de phtalocyanines métalliques par nos collaborateurs (F. Bouanis). Des méthodes de caractérisation ont été employées pour obtenir des informations chimiques, structurelles et morphologiques précises sur ces systèmes qui doivent être performants pour détecter les gaz CO et NO. Des calculs DFT ont été entrepris pour reproduire les tendances expérimentales observées pour la fonctionnalisation du NTC(8,0) semi-conducteur et du graphène par ces macrocycles, puis pour la fixation de CO et NO par ces dispositifs. Enfin, la fonctionnalisation d’un autre type de nanomatériau, MoS 2, par une porphyrine de cobalt (II) est également en cours d’étude

International audience; In this second part of this series of papers, the use of two physics-based models to analyze the discharge performance of a set of high-energy-density electrodes is discussed. The measured set of parameters from the first part is implemented into these models. First, the regular Newman pseudo-2D model shows a large discrepancy against the experimental values. Then, an extension of the Newman model considering the particle agglomeration due to the calendering effects is presented, allowing for the validation of discharge rate capabilities of all studied industry-grade electrodes with different electrolytes. At the agglomerate scale, the model accounts for both the ionic transport in sub-pores and the inter-particle solid diffusion. The simulation results from this work demonstrate that increasing the electrode loading and/or density leads to either a higher fraction of sub-pores (at the expense of that of macropores) or larger porous agglomerate size, resulting in a poor rate performance. The model analysis suggests that a substantial gain in performance at high C-rates is expected if agglomeration effects are mitigated in these high-energy electrodes.

International audience; The investigations of bubble formation dynamics at a cross-junction in a straight milli-channel were reported. The bubble formation process could be divided into the filling and squeezing stages, and their frequencies were compared at various conditions. It was found that the filling and squeezing frequencies were controlled mainly by the gas and liquid superficial velocities, respectively. The bubble formation frequencies could be related to the gas-liquid superficial velocity ratios, the liquid superficial velocities, and the bubble length. The bubble formation process was then analyzed considering the length of the gas finger right after the bubble pinch-off. Two patterns were identified depending on whether the pinch-off occurred inside of the cross-junction or not. The transition between these two patterns was described by a critical liquid Capillary number. Furthermore, the squeezing and dripping patterns were distinguished as the gas finger could fully block the channel or not. The transitions between the bubble formation patterns were determined by the gas Weber number and liquid Capillary number. Finally, the bubble length, the liquid slug length, and the bubble length normalized by the unit cell length could all be predicted based on the gas-liquid superficial velocity ratios eta(0).

Complexation of polymers at liquid interfaces is an emerging technique to produce all-liquid printable and self-healing devices and membranes. It is crucial to control the assembly process, but the mechanisms at play remain unclear. Using two different reflectometric methods, we investigate the spontaneous growth of H-bonded PPO-PMAA (polypropylene oxide-polymetacrylic acid) membranes at a flat liquid-liquid interface. We find that the membrane thickness h grows with time t as h similar to t(1/2), which is reminiscent of a diffusion-limited process. However, counterintuitively, we observe that this process is faster as the PPO molar mass increases. We are able to rationalize these results with a model which considers the diffusion of the PPO chains within the growing membrane. The architecture of the latter is described as a gel-like porous network, with a pore size much smaller than the radius of the diffusing PPO chains, thus inducing entropic barriers that hinder the diffusion process. From the comparison between the experimental data and the result of the model, we extract some key piece of information about the microscopic structure of the membrane. This study opens the route toward the rational design of self-assembled membranes and capsules with optimal properties.

The objective of this thesis is to improve the understanding of the fundamental microscopic mechanisms that govern the phase transformations induced directly in the solid state by mechanical milling. In particular, three key issues have been addressed: (i) We evaluated the influence of the milling temperature on the effective amorphization rate of a crystalline material. (ii) We determined the influence of monotropic and enantiotropic characters on the mechanisms of polymorphic transformations under milling. (iii) We have shown that the polymorphic transformations under milling are not direct but involve, on the contrary, a transient amorphization mechanism. Our strategy consisted in studying these different problems through a detailed analysis of the transformation kinetics which are the direct reflection of the physical mechanisms that govern them. For this purpose, we have studied in detail the transformation kinetics of physical states of several carefully chosen compounds. These include hydrochlorothiazide, sorbitol, glycine and their mixtures. The investigations were mainly carried out by X-ray diffraction and differential scanning calorimetry (DSC). Moreover, some experimentally demonstrated transformation mechanisms could be modelled and validated by numerical simulations. In summary, all our work indicates that, in general, the transformations under milling seem to result systematically from a competition between an amorphization mechanism due to mechanical shocks and a recrystallization mechanism whose nature depends on both the milling temperature and the crystalline specificities of the material., L'objectif de cette thèse, est d'améliorer la compréhension des mécanismes microscopiques fondamentaux qui gouvernent les transformations de phases induites directement à l'état solide par broyage mécanique. Trois problèmes essentiels ont été plus particulièrement abordés : (i) Nous avons évalué l'influence de la température de broyage sur la vitesse d'amorphisation effective d'un matériau cristallin. (ii) Nous avons déterminé l'influence des caractères monotrope et énantiotrope sur les mécanismes de transformations polymorphiques sous broyage. (iii) Nous avons montré que les transformations polymorphiques sous broyage ne sont pas directes mais qu'elles font, au contraire, intervenir un mécanisme d'amorphisation transitoire. Notre stratégie a consisté à étudier ces différents problèmes par une analyse fine des cinétiques de transformations qui sont le reflet direct des mécanismes physiques qui les gouvernent. Pour cela, nous avons étudié en détail les cinétiques transformations d'états physiques de plusieurs composés judicieusement choisis. Il s'agit notamment de l'hydrochlorothiazide, du sorbitol, de la glycine et de leurs mélanges. Les investigations ont principalement été menées par diffraction des rayons X et par analyse enthalpique différentielle (DSC). De plus, certains mécanismes de transformation mis en évidence expérimentalement ont pu être modélisés et validés par simulations numériques. En résumé, l'ensemble de nos travaux indiquent que, d'une manière générale, les transformations sous broyage semblent résulter systématiquement d'une compétition entre un mécanisme d'amorphisation dû aux chocs mécaniques et un mécanisme de recristallisation dont la nature dépend à la fois de la température de broyage et des spécificités cristallines du matériau.

International audience; Women are increasingly present in the field of engineering, and despite a significant female presence, it has been found that the programmes continue to make no reference to women scientists. In chemical engineering, for example, all the names of scientists mentioned in the programmes belong to men only. To test this hypothesis of the overrepresentation of men in the programmes, a series of random opinion surveys were launched among 600 students from 5 universities to find out whether they had noticed this over-representation and what they thought about it. The results showed that the vast majority did not realize that the scientists presented as examples in classes were all men. In fact, 90% of the student panel were unable to identify a woman in the chemical engineering field, and the remaining 10% could cite only one or two-who were among the most recent and had received most attention from the media. The issue of inequalities between girls and boys and between women and men in education remains central to understanding and combating gender inequalities and to enabling people to develop as persons free from the limitations imposed on them by gender stereotypes. However, these inequalities cannot be explained exclusively by the issue of access to education but must also take the type and content of education into account. This article is a call for reflection on the content of university curricula and has a twofold objective: on the one hand, to raise awareness of this imbalance in representation among students, both male and female, and, on the other hand, to launch reflection on this "invisibility of women" and to propose some avenues for debate.

In dairy industries, production costs are highly impacted by the deposition of fouling onto equipment. Promising coatings to prevent fouling adhesion (i.e. anti-fouling) or to ease fouling removal (i.e. fouling-release) were previously developed. Nonetheless, their durability and food-compatibility were limited. Consequently, to overcome these limitations, this work aims at designing coatings based on innovative surface modification techniques and concepts: (i) polymer deposition by atmospheric pressure plasma torch and (ii) bioinspired slippery surfaces by femtosecond laser ablation and oil infusion. The first technique allowed the deposition of hexamethyldisiloxane (HMDSO) which demonstrated good fouling-release performances but did not allow the durability improvement. The alternative deposition of two precursors (HMDSO and 1H,1H,2H,2H perfluorooctyltriethoxysilane) was studied. Superhydrophobic and very stable bilayers were deposited onto stainless steel, showing good but still improvable fouling-release properties (fouling reduction of 72%). The second research axis consisted in designing slippery liquid-infused surfaces (SLIS) following three steps: (i) laser structuration of stainless steel, (ii) chemical modification of structured surface and (iii) lubricant impregnation. An optimization of laser parameters allowed to reach quickly various types of deep microstructures. Food-compatible SLIS were developed by replacing fluorine-based lubricant by coconut oil. Although a poor durability due to a loss of oil, coconut-SLIS exhibited promising fouling-release performances with a fouling deposit reduction of 114%.; Les industries laitières sont extrêmement impactées par l’encrassement des équipements de pasteurisation et de stérilisation. Une des solutions envisagées pour limiter l’adhésion de l’encrassement consiste à modifier la surface de l’acier inoxydable des échangeurs thermiques. Différent types de revêtements, limitant l’adhésion ou facilitant le nettoyage, ont précédemment démontré des résultats prometteurs. Cependant ces revêtements doivent être durables et compatibles avec les aliments. Cette thèse vise donc à développer des surfaces stables, non toxique et durables à haute température et en régime turbulent. Des techniques de modification de surface innovantes ont été utilisées pour concevoir ces revêtements : (i) dépôt de couches minces par plasma à pression atmosphérique et (ii) surfaces glissantes bio-inspirées obtenues par ablation laser et infusion/imprégnation d’un lubrifiant. La première technique a permis de déposer des couches minces d’hexaméthyldisiloxane (HMDSO) facilitant le nettoyage de l’encrassement mais présentant une faible durabilité. Le dépôt en alternance de deux précurseurs (HMDSO et 1H,1H,2H,2H perfluorooctyltriethoxysilane) a également été étudié. Cela a permis de déposer des bicouches originales, stables et superhydrophobes. Les paramètres de dépôt pourront encore être optimisés pour améliorer leur nettoyabilité (réduction de l’encrassement après rinçage : 72%). Les surfaces glissantes bio-inspirées (SLIS) ont été fabriquées suivant trois étapes : (i) structuration de l’acier inoxydable par ablation laser, (ii) modification chimique de la surface structurée et (iii) imprégnation avec un lubrifiant d’huile. Une optimisation des paramètres laser a permis d’atteindre rapidement divers types de microstructures de différentes profondeurs. L’huile fluorée habituellement infusée a été remplacée par de l’huile de coco, rendant la surface compatible au contact alimentaire. Malgré une faible durabilité dûe à la perte d’huile, les surfaces glissantes infusées à l’huile de coco ont permis de faciliter le nettoyage (réduction de l’encrassement après rinçage : 114%).

Physics-based models of the Li-ion battery are promising to decipher and quantify the electrode limitations, thereby providing valuable insights for choosing the optimal electrode design for a specific application. However, to obtain relevant results from the models, a reliable set of input parameters is required. This work presents a combined experimental/modeling approach relying on the Newman pseudo-2D model for a complete characterization of a set of LiNi0.5Mn0.3Co0.2O2 electrodes. Intrinsic properties of the active materials are determined and validated using low-loading electrodes having negligible porous-electrode limitations. Then, high-energy-density electrode properties are characterized using appropriate experimental methods, which are widely reported in the literature. In the second part of this series of papers, parameters obtained from this part serve as input parameters in the Newman pseudo-2D model as well as in its extension in order to simulate the rate capability during discharge of the aforementioned set of high-energy-density electrodes. List of symbols a i m i 2 / m PE 3 interfacial surface area of phase i c s , surf mol m − 3 concentration at the surface of the AM particle c s , max mol m − 3 maximum concentration of intercalated Li in AM particle c s mol m − 3 solid-phase Li concentration within the AM particle c ¯ s mol m − 3 local volume-averaged solid Li concentration of AM phase within the PA c mol m − 3 salt concentration in a binary electrolyte d 50 μ m median diameter of AM particles D m 2 s − 1 bulk diffusion coefficient of the liquid phase D s m 2 s − 1 diffusion coefficient of Li in the AM particles F C mol − 1 Faraday’s constant i coexisting phase presented in the PE i n 0 A m − 2 exchange current density i Li 0 A m − 2 exchange current density at the Li foil I app A / m CC 2 discharge current density j n mol / m AM 2 · s pore-wall flux across the sandwich k 0 mol m 2 · s · mol m − 3 1.5 − 1 reaction rate constant of the AM k 0 , Li mol m 2 · s · mol m − 3 0.5 − 1 reaction rate constant of Li foil L el μ m PE thickness L sep μ m separator thickness Q th Ah kg − 1 electrode theoretical capacity R J mol · K − 1 ideal gas constant r μ m radial dimension along the AM particle T K absolute temperature t s time t + 0 transference number of Li+ in the electrolyte with respect to the solvent velocity U V equilibrium potential of the AM Δ V V voltage drop between the two inner contacts in the μ4-probe experiment x μ m dimension across the sandwich x 0 initial stoichiometry Greek Symbols α thermodynamic factor β charge transfer coefficient ε m elyte 3 / m PE 3 PE porosity ε sep m elyte 3 / m sep 3 separator porosity κ eff S m − 1 effective ionic conductivity of the liquid phase ρ el g cm − 3 electrode density σ eff S m − 1 effective electronic conductivity of the solid phase of the electrode τ Br tortuosity factor by Bruggeman τ e electrode tortuosity factor τ sep tortuosity factor of the separator Φ 1 , Li V electric potential at Li foil Φ i V electric potential of phase i

The present work is a contribution to better understand the influence of casein micelles on the fouling of serum whey protein solutions. In particular, experimental and numerical approaches have been carried out, at laboratory and pilot scales, to describe denaturation phenomena and better understand the role of calcium in fouling mechanisms. First of all, the effect of casein/whey mass ratio on the whey protein fouling performance was investigated in a pilot-scale PHE. The total fouling deposit mass drop significantly with the addition of casein, resulting in a minimum value located at Casein/WPI of 0.2. Exceeding this critical ratio, fouling deposit increased with elevated casein concentrations. The deposit mass drop (Casein/WPI ≤ 0.2) is unlikely to be linked to the thermal denaturation of BLG and is more probably due to the change in mineral interactions introduced by casein. The increased fouling mass (Casein/WPI ≥ 0.2) was attributed to a co-precipitation of BLG-casein complex that enhances the fouling. It is proposed that micellar casein change deeply the calcium balance and the content of CaP nanocluster modifies sharply the interactions which occur between protein species (BLG, caseins) and mineral elements (ionic calcium, Ca-P) thereby affecting the protein denaturation and fouling behavior. A novel kinetic model concerning thermal unfolding and aggregation of BLG was established. This model interprets mathematically the break-slope behavior in the Arrhenius plot and provides detailed thermodynamic information for both unfolding and aggregation processes. Based on this model, it was confirmed that ionic calcium has a protective role on the thermal unfolding of BLG at low temperature. In contrast, at higher temperatures, calcium promotes aggregation and the formation of unfolded BLG species. A bench-scale fouling rig was built to perform whey protein fouling experiments in a laminar regime. A realistic 3D CFD model was achieved to simulate both the bulk and surface reactions. Results showed a linear relationship between the deposition pre-exponential factor and calcium concentration, suggesting the fouling is built in such a pattern that only one calcium ion per BLG molecule is involved. Calcium was confirmed to be essential to fouling growth with significant effects both on the thermal denaturation and deposition processes. Finally, the effect of casein/whey ratio on the whey protein fouling was investigated in the laboratory-scale fouling device. Results revealed a similar effect of casein on fouling mitigation as those found in the pilot plant. However, in this case, the fouling was suppressed and maintained at a low extent even at high Casein/WPI ratios (up to 4). The presence of individual caseins in the serum phase was considered to be responsible for this fouling mitigation probably through their chaperon-like activities. However, when the pH of the fouling solution is set at 6.6, casein is shown to lose its fouling-mitigating effect at higher ratios. This behavior is related to its weak ability of casein micelle to control ionic calcium in the serum phase at lower pH, resulting in higher calcium concentration facilitating BLG denaturation and deposition accumulation. A lower amount of dissociated caseins in the serum phase at pH 6.6 could also explain the increase in fouling mass because they are not in sufficient concentration to perform chaperone-like functions.; Le présent travail est une contribution pour mieux comprendre l’influence des micelles de caseine sur l’encrassement de solutions de protéines sériques. En particulier, des approches expérimentales et numériques ont été réalisées, à des tailles laboratoires et pilotes, pour décrire les phénomènes de dénaturation et mieux cerner le rôle du calcium dans les mécanismes d’encrassement. Tout d'abord, l'effet du ratio massique caséine / lactosérum sur les performances d'encrassement des protéines de lactosérum a été étudié dans un échangeur à Plaques à l'échelle pilote. La masse totale du dépôt d'encrassement chute d’abord de manière significative avec l'augmentation de la concentration en caséine, atteignant un minimum quand le ratio vaut 0,2. Au-delà de cette valeur, la masse de dépôt réaugmente. La chute de la masse du dépôt, pour un ratio ≤ 0,2, ne semble pas être corrélée à la dénaturation thermique du BLG mais plus probablement due à la modification des interactions minérales introduites par la caséine. L'augmentation de la masse de dépôt, pour un ratio ≥ 0,2, semble être liée à une co-précipitation du complexe BLG-caséine qui augmente l'encrassement. Il est suggéré que la présence de caséine micellaire modifie profondément l'équilibre calcique en solution et que la teneur en nanocluster de Ca-P modifie fortement les interactions entre les espèces protéiques et les minéraux (calcium ionique, Ca-P) affectant ainsi la dénaturation des protéines et la précipitation des minéraux. Un nouveau modèle cinétique concernant le dépliement thermique et l'agrégation de BLG a été établi. Ce modèle est en mesure de justifier la rupture de pente dans le diagramme d'Arrhenius et de fournir des informations thermodynamiques détaillées pour les processus de dépliement et d'agrégation. Sur la base de ce modèle, il a été confirmé que le calcium ionique avait un rôle protecteur sur le dépliement thermique du BLG à basse température. En revanche, à des températures plus élevées, le calcium favorise l'agrégation et la formation d'espèces BLG dépliées. Un dispositif d'encrassement à l'échelle laboratoire a été construit et tester avec des protéines de lactosérum en régime laminaire. Un modèle CFD 3D réaliste a été implémenté simulant à la fois les réactions au cœur du fluide et en surface. Les résultats ont montré une relation linéaire entre le facteur pré-exponentiel et la concentration de calcium, ce qui suggère que l'encrassement nécessite qu’une seule molécule de calcium soit associée à une protéine de BLG. Il est confirmé que le calcium est essentiel à l'encrassement avec des effets significatifs à la fois sur les processus de dénaturation thermique et sur la croissance du dépôt. Enfin, l'effet du ratio caséine / lactosérum sur l'encrassement a été étudié avec un dispositif d'encrassement de laboratoire. Les résultats laboratoires montrent que la caséine réduit l’aptitude à l’encrassement comme déterminé précédemment avec l’installation pilote. Cependant, dans ce cas, l'encrassement reste à un niveau faible y compris pour des ratios élevés (jusqu'à 4). La présence de caséines individuelles dans la phase sérique a été considérée comme responsable de cette atténuation de l'encrassement, probablement par leurs activités de type chaperon. Cependant, quand le pH de la solution d'encrassement est fixé à 6,6, il est démontré que la caséine perd son effet d'atténuation de l'encrassement pour des ratios plus élevés. Ce comportement est lié à sa faible capacité de micelle de caséine à contrôler le calcium ionique dans la phase sérique à un pH plus bas, entraînant une concentration plus élevée en calcium facilitant la dénaturation de la BLG et l'accumulation de dépôts. Une quantité plus faible de caséines dissociées dans la phase sérique à pH 6,6 pourrait aussi expliquer l'augmentation de la masse d'encrassement car elles ne sont pas en concentration suffisantes pour remplir des fonctions de type chaperon.

In France, borosilicate glass is used as a matrix to immobilize nuclear fission products resulting from spent fuel reprocessing. In the high-temperature vitrification process (1200 °C), most of elements to be contained react chemically with the vitrification additives to form a homogeneous glass melt. Platinum Group Metal (PGM) particles are not soluble in the melt and therefore are present as suspended particles of few microns. These particles exhibit an intense aggregation tendency and consequently the suspensions may present an anomalously high apparent viscosity. These systems are characterized by a shear-thinning and a thixotropic behaviors. The present study aims to provide important inputs for the overall rheological behavior of this system and its features through the characterization of a simulated nuclear glass melt containing 3.0 wt% (1.02 vol%) of PGM particles. A mathematical modeling of the thixotropic behavior of glass melts containing PGM particles is presented for the first time using a model similar to that proposed by Houska (Houska, 1981). This predictive model allow to describe experimental results obtained both in steady state and transient conditions. The impact of the shear stress and time on PGM aggregation degree and sedimentation kinetics is determined using an imposed-stress rheometer at high temperature and imaging analyses via Scanning Electron Microscopy (SEM). For the first time, the interplay between the rheological behavior of the system and the aggregation degree is provided, as well as the link with the particles settling. Based on the acquired experimental data, a force balance computation is executed to illustrate the different aggregation scenarios. The work provides a new input for the modeling and control of the vitrification process., En France, les produits de fission nucléaire issus du retraitement des combustibles usés sont conditionnés au sein de matrices de verres borosilicatés. Lors du processus de vitrification à haute température (1200 °C), les éléments à confiner réagissent chimiquement avec les additifs de vitrification pour former un verre homogène. Cependant, parmi ces éléments, les platinoïdes ne sont pas solubles dans le bain de verre et restent en suspension sous forme de particules de quelques microns. Ces particules présentent une forte tendance à l'agrégation et les suspensions peuvent alors présenter une viscosité apparente anormalement élevée. Ces systèmes sont caractérisés par un comportement rhéofluidifiant et thixotrope. La présente étude vise à fournir des données importantes sur le comportement rhéologique global de ces systèmes et leurs propriétés, grâce à la caractérisation d'un verre nucléaire simulé contenant 3,0 % massique (1,02 % en volume) de platinoïdes. Une modélisation mathématique du comportement thixotrope de ce verre est présentée pour la première fois en utilisant un modèle similaire à celui proposé par Houska (Houska, 1981). Ce modèle prédictif permet la description des résultats expérimentaux obtenus à partir de mesures en régimes permanent et transitoire. L'impact de la contrainte de cisaillement et du temps sur le degré d'agrégation et la cinétique de sédimentation des particules est déterminé à l'aide d'un rhéomètre à contrainte imposée opéré à haute température et d'analyses d'images obtenues par microscopie électronique à balayage (MEB). Pour la première fois, l'interaction entre le comportement rhéologique du système et le degré d'agrégation est analysé, tout comme le lien avec la sédimentation des particules. Sur la base des données expérimentales acquises, un calcul d'équilibre des forces est réalisé pour illustrer les différents scénarios d'agrégation. Ce travail fournit des nouvelles données pour la modélisation et le contrôle du processus de vitrification.

International audience; Pharmaceutical contaminants have shown widespread presence in water/wastewater threatening environmental and human health. Existing literature shows that adsorption can efficiently bind pharmaceuticals preventing their transfer into the treated water. However, adsorption only diverts these contaminants from one phase to another requiring the treatment of contaminant-laden adsorbents. Fenton oxidation (H2O2-based) has shown good potential for the degradation of pharmaceuticals. But, the practical use of Fenton oxidation is often limited by its higher chemical demands and low contact of oxidizing species with target contaminants. Efforts to overcome these practical obstacles aim to develop efficient heterogeneous catalysts which also benefit from their strong contaminant adsorption capacity. This is the first review that focuses exclusively on dual-functional materials that can effectively adsorb pharmaceutical pollutants (separation-based removal) and promote Fen -ton oxidation simultaneously (degradation-based removal) in aqueous systems. The dual functionality enables the integration of adsorption and Fenton-based processes (heterogeneous Fenton, photo-Fenton and photo-catalysis) to efficiently remove pharmaceuticals. This integration can be achieved using a single material having both functions or by compositing/impregnating different materials. The properties of these materials have important consequences on their treatment efficiency, stability, and regeneration ability. These processes and materials are critically evaluated to illustrate their potential, associated challenges and their key solutions. Some of these challenges originate from the tested materials while others are inherent to the treatment process and the prospects for improvement exist at their intersection. Future research should aim to rationally improve the catalytic and adsorption properties of these materials and explore their practical implementation in wastewater treatment technologies.

The microfluidic production of Janus/Hecate polymer microfibers with well-defined interfaces from miscible phases is reported. The process offers tunability of the width and composition of each part of the fibers by controlling the flow rate and nature of the monomers in a single step. The enhanced performances of the fibers are outlined for the simultaneous removal of dyes of opposite charges using amphoteric Janus fibers.

Pharmaceutical products (PPs) are considered as emerging micropollutans in wastewaters, river and seawaters, and sediments. The biodegradation of PPs, such as ciprofloxacin, amoxicillin, sulfamethoxazole, and tetracycline by enzymes in aqueous solution was investigated. Laccase from Trametes versicolor was immobilized on silica monoliths with hierarchical meso-/macropores. Different methods of enzyme immobilization were experienced. The most efficient process was the enzyme covalent bonding through glutaraldehyde coupling on amino-grafted silica monoliths. Silica monoliths with different macropore and mesopore diameters were studied. The best support was the monolith featuring the largest macropore diameter (20 µm) leading to the highest permeability and the lowest pressure drop and the largest mesopore diameter (20 nm) ensuring high enzyme accessibility. The optimized enzymatic reactor (150 mg) was used for the degradation of a PP mixture (20 ppm each in 30 ml) in a continuous recycling configuration at a flow rate of 1 ml/min. The PP elimination efficiency after 24 h was as high as 100% for amoxicillin, 60% for sulfamethoxazole, 55% for tetracycline, and 30% for ciprofloxacin.