This Thesis considers peculiarities of nanoparticle formation from the gas in different systems. The main role of the surface condensation in the nanoparticle growth in metal flames was established through a series of experiments and was described by the developed model. The stagnation of the post-nucleation nanoparticle growth was experimentally revealed and theoretically explained. The influence of generation conditions on the post-processing nanoparticle properties was examined. The non-isothermal approach to correct the homogeneous nucleation theory was developed. The results of this work can be summarized in 3 categories: (1) Nanoparticle formation in metal flames. In this work, it was demonstrated that the surface condensation is a main process responsible for nanooxides growth during metal combustion. It was shown that the rate of this condensation growth is consistent with the exponential law, which could lead to the formation of the lognormal particle size distribution in the system, where the Brownian coagulation is suppressed. The post-nucleation stagnation of the nanoparticle growth was found. The particle overheating was suggested as a cause of the growth stagnation. The found stagnation leads to the accumulation of the supercritical clusters in the system generating nanoparticles. The role of these supercritical clusters in the nanoparticle agglomeration was considered. (2) Study of properties of nanoparticles generated in different metal flames. The light absorption, photoluminescence and magnetic properties of nanoparticles produced in different metal flames were examined. The significant broadening of the absorption edge was found in nanooxides produced by direct metal combustion. This broadening allowed one to excite the unforeseen photoluminescence from these nanoparticles. The significant light absorption in the visible light found in the titania particles produced by metal combustion allows one to consider these particles as a prospective photocatalyst. The unusual optical properties revealed were related to the extreme conditions of the nanoparticle formation, namely, to high energy release (about 5 eV per condensing molecule). The stabilization of spinel structure was found in iron oxide particles synthesized by iron combustion. It allowed one to produce nanoparticles with magnetization close to the high-limit value of the bulk. (3) Approach to correct the homogeneous nucleation theory. The existing homogeneous nucleation theory implies that nucleation occurs at isothermal conditions, i.e. subcritical clusters have the same temperature as the ambient gas does. However, the theory overestimates the actual nucleation rate and underestimates the critical cluster size. It is understandable that due to release of the latent heat of condensation, the cluster temperature in the nucleating system should be higher than the environment temperature. In this work, it was suggested the method to account for the cluster overheating during nucleation. It was demonstrated that the consistent description of the detailed balance in the nucleating system may allow one to evaluate magnitudes of overestimation of the actual nucleation rate and underestimation of the number of molecules in the critical cluster, usually obtained by the isothermal nucleation theory. The numerical estimates are in good agreement with the wellknown experimental results. The implications of the results to generation of atmospheric aerosols were discussed.

Au cours de la dernière décennie, les nanoparticules métalliques ont trouvé de nombreuses applications dans divers domaines tels que l'optique, la photonique, la catalyse, la fabrication de matériaux, les énergies renouvelables, l'électronique, la médecine et même les cosmétiques. Les nouveaux développements de ces applications nécessitent des méthodes de synthèse de nanoparticules fiables donnant une grande quantité de nanoparticules aux propriétés spécifiques. Les méthodes à base de plasma, tels que des décharges d'étincelles et d’arcs sont parmi les plus prometteuses car elles permettent une augmentation considérable de la vitesse de production et une diminution des coûts. Le contrôle de ces processus est cependant toujours difficile et nécessite de nombreuses études détaillées, à la fois expérimentales et théoriques. Dans cette thèse, les décharges d'étincelles sont étudiées numériquement. L'objectif principal est de mieux comprendre les principaux mécanismes impliqués dans la décharge d'étincelle avec un faible écartement d’électrodes et sous pression atmosphérique. Ensuite, sur la base de la modélisation détaillée proposée, la quantité de nanoparticules produites ainsi que leur distribution en taille est prédite et est comparée avec les résultats expérimentaux correspondants. Dans le modèle proposé, seules les conditions initiales, la géométrie du système et les propriétés du matériau sont utilisés comme paramètres d'entrée. Une décharge d’étincelle unique est divisée en plusieurs unités selon les échelles spatiales et temporelles des processus physiques comme suit: modèles de (i) flux plasma, (ii) décharge, (iii) hydrodynamique, (iv) couche cathodique, (v) érosion d’électrode et (vi) formation de nanoparticules. Les résultats du modèle combiné sont ensuite comparés à la fois avec d'autres résultats théoriques et à des résultats expérimentaux. Enfin, les possibilités d'optimisation de la production de nanoparticules par décharge d'étincelles sont proposées sur la base de la variation des paramètres expérimentaux et sur l'analyse de la quantité de particules produites et de leur taille moyenne / During last decade, metal nanoparticles have found many applications in various areas, such as optics, photonics, catalysis, material manufacturing, renewable energy, electronics, medicine and even cosmetics. Further development of these applications requires reliable nanoparticle synthesis methods providing a large amount of nanoparticle with required properties. Plasma-based methods, such as spark and arc discharges are among the most promising since they allow a considerable increase in the production rate and a decrease in costs. The control over these processes is, however, still challenging and requires many detailed studies, both experimental and theoretical. In this thesis, spark discharge is investigated numerically. The main objective is to better understand main mechanisms involved in spark discharge with a short gap under atmospheric pressure. Then, based on the proposed detailed modeling, the amount of the produced nanoparticles, their size distribution should be predicted and compared with the corresponding experimental results. In the proposed model, only initial conditions, geometry of the system and material properties are used as input parameters. A single spark event is divided into several units according to localization and time scales of physical processes as follows: (i) streamer model, (ii) discharging model, (iii) hydrodynamic model, (iv) cathode layer model, (v) electrode erosion model and (vi) nanoparticle formation model. The results of the combined model are then compared both with other theoretical and experimental results. Finally, possibilities of optimization the nanoparticle production by spark discharge are proposed based on the variation of the experimental parameters and on the analysis of the resulted particle yield and mean size

Nanoparticles have important applications in ceramics, metal catalysts, semiconductors etc. They are normally required to be of small size (~ nm) and monodisperse. The aim of the present work is to model the formation of nanoparticles, obtained by precipitation in reverse micellar microreactors. These are dispersions of tiny water drops in a surfactant laden oil medium. Two systems were investigated: (i) Reverse micelles, having nanometer sized spherical water droplets in the micellar core and (ii) Water-in-oil emulsions, having micron-sized aqueous drops. Two modes of precipitation, namely, gas-liquid (g-1) and liquid-liquid (1-1) were studied. In each case, the models could predict the number, average size and size distribution of the particles reported in literature. Two groups have obtained widely divergent number and size of CaCO3 nanoparticles, formed by g-1 precipitation in reverse micelles. These particles are used as a fine suspension in lube-oil additives, where they serve to neutralize acid produced during combustion in engines. Kandori et al. (J. Colloid Interface Sci, 122,1988, 78) obtained particles of about 100 nm size, by passing CO2 through a reverse micellar solution, containing dissolved Ca(OH)2 in the micellar core. Roman et al. (J. Colloid Interface Sci., 144,1991, 324), instead of using lime solution; added micron-sized solid lime particles in the oil and generated the reverse micelles by in situ reaction. This is a commercial process known as overbasing. It led to a higher amount of lime in the micelles as well as unreacted lime particles in oil, at the beginning of the experiment Upon passing CO2, they got particles of only 6 nm in size, compared to 100 nm reported by Kandori et al.. Furthermore, while Kandori et al. found that one particle formed from 108 micelles, Roman et al. got one particle out of only ten micelles. We have modelled the two processes in a common framework to explain the reported disparity in particle characteristics. A time scale analysis of CO2 mass transfer, reaction, collision-fusion of micelles, nucleation, and growth of particles was carried out It showed that, in the experiments of Kandori et al., the rate limiting steps are nucleation and fusion. The analysis also indicates that the contents of a particular micelle are well mixed and reaction of lime and incoming CO2 can be treated as instantaneous. In the process of Kandori et al., the amount of lime taken initially being very small, the average number of product molecules in a micelle is well below one. Rapid Brownian coalescence and exchange of micellar contents leads to Poisson distribution of CaCO3(l) molecules formed by reaction. The low occupancy therefore suggests that most of the micelles are empty. Nucleation in a particular micelle is much slow and occurs when it has a critical number of molecules. Thus only very few micelles can nucleate. Comparison of nucleation and growth time scales - both intrinsic growth in a micelle and growth during fusion of nucleated and non-nucleated micelles - show that growth is much faster than both nucleation and collision. Hence a micelle can have only one nucleus, with subsequent growth during collisions. A population balance equation (PBE) is written involving the above steps. Solution of the moments of the distribution yields the number of CaCO3 particles, its size, coefficient of variance (COV) etc. The model not only predicts the ratio of number of micelles to particles, obtained experimentally as 108, but also captures the maxima in this quantity with increasing micellar size. The increase in average particle size with micellar size is also predicted well. The process of of Roman et ai, in addition, involves the time scale of solubilization of solid lime into micelles. Its comparison with other time scales demarcates their experiments into two distinct phases. Phase I consists of reaction of lime initially present in micelles. Time scale analysis also suggests that, as the lime content in the micelles is large, a high degree of supersaturation is rapidly generated. This results in a burst of nuclei. The other conclusions, like, well-mixed micelle, Poisson distribution of CaCO3(l) molecules, instantaneous growth and mono-nucleated micelles are found to hold good. Once the pre-existing lime is finished, relative time scales indicate that, further precipitation is controlled entirely by fresh solubilization of lime. This marks the beginning of phase II. However, solubilization being the slowest step, CaCO3(l) in micelles never builds up for any further nucleation. Phase II thus consists of pure growth of the particles formed in phase I. On developing more general PBEs and with solution of resulting moment equations - written separately for the two phases - the experimental data on number of particles and temporal evolution to the final particle size of 6 nm could be predicted very well. The model also captures the qualitative trend in COV of particle radius with time. Thus within the same framework we could successfully predict both the results, differing by seven orders of magnitude. The above analysis indicates that relative rates of nucleation, fusion-growth and mass transfer of gas controls the carbonation process. We further simplify the process and obtain an analytical solution in the limit of instantaneous mass transfer. The solution gives close first estimates for both the experiments and also indicates the smallest panicle size that could be obtained for a given experimental condition. In contrast to g-1 mode, precipitation in 1-1 mode - using two reverse micellar solutions having two reactants- occurs only on coalescence of two micelles. To obviate the solution of multivariate PBEs, we have developed a general Monte Carlo (MC) simulation scheme for nanoparticle formation, using the interval of quiescence technique (IQ). Starting with a fixed number of micelles, we conduct each coalescence-redispersion and nucleation events in this population, in the ratio of their relative frequencies. Our simulation code is much more general and realistic than the scheme of Li and Park (Langmuir, 15,1999, 952). Poisson distribution with realistic micellar occupancies of reactants, binomial redispersion of solutes after fission, a nucleation rate with critical number of molecules and Brownian collision-fusion rates were used. These considerations are based on our earlier findings in g-1 precipitation and those known in the literature too. The simulation of Li and Park then becomes a special case of our code. Our simulation code was then used to predict experimental data on two systems. The results of Lianos and Thomas (Chem. Phys. Lett. 125, 1986, 299 and /. Colloid Interface 5c/., 117, 1987, 505), on number of molecules per CdS particle, as a function of micelle size and reactant concentrations have been predicted very well. For the Fe(OH)3 nanoparticles, our simulation provides a better prediction of the experimental particle size range, than that of Li and Park. Finally, 1-1 precipitation on mixing two emulsions, having respectively the two reactants, has been simulated. Here, large reactant amount leads to multiple nucleation in a single drop and renders growth rate to be finite. This requires solving a PBE for particle population in each drop. Moreover, emulsions have a drop size distribution due to independent coalescence and breakage. The IQ technique was used for handling these events. Thus a composite model of PBE and MC for a drop population was developed. Simulation of particle size distribution in MgCO3 precipitation shows that nearly monodisperse nanoparticles can be produced in emulsions. Furthermore, average particle size can be controlled by changing reactant concentration in a drop. The findings of the thesis have provided new issues to be addressed in modelling nanoparticle formation. It points out the importance of finding models for coalescence efficiency and critical nuclear size in micelles. Extension of our model and simulation to precipitation in other organized surfactant assemblies can be done by starting from appropriate time scale analysis.

Nanoparticles represent a very exciting new area of research. Their small size, ranging from several nanometers to tens of nanometers, is responsible for many changes in the structural, thermal, electromagnetic, optical and mechanical properties in comparison with the bulk solid of the same materials. However, promoting the use of such material requires well-controlled synthesis techniques to be developed. Inductively coupled thermal plasma (ICTP) reactors have been shown to offer unique advantages over other synthesis methods. The purpose of this thesis is to develop a numerical model to assist the design of an ICTP reactor for the efficient and controlled production of nanoparticles at industrial scale. The complete model describes the evaporation of the micron-sized precursor particles in the plasma flow and the subsequent formation of the nanoparticles in the quenching reactor. The plasma flow is described by a coupled system of the fluid mechanics equations of continuity, momentum, and energy with the vector potential formulation of Maxwell's equations. The solid particles precursors are treated following a Lagrangian approach, taking into account the vapor production field in the plasma flow. An Eulerian model based on the method of moments with interpolative closure is used to describe the formation of nanoparticles by simultaneous nucleation and growth by condensation and coagulation. The coupled plasma torch, particle evaporation and nanoparticle formation models are implemented in 2D and 3D configurations, using the OpenFoam source code. The results show that the effects of the particle evaporation on the temperature field are substantial, even for low particle mass loading. The associated vapor concentration which enters in the reactor has then a direct influence on the formation of nanoparticles. The effects of the plasma torch parameters and the quenching configuration (quench type, position, injection angle and cooling rate) on the contribution of the different formation mechanisms and on the generated particle's size and distribution are studied in both 2D axi-symmetric and 3D geometries. The quench mechanism strongly affects the temperature and the vapor concentration in the reactor, and consequently has an impact on the final particle size distribution. It is shown that the size of the nanoparticles obtained for different quenching conditions is not only a consequence of the cooling rate but also of the trajectories of the vapor and the generated particles imposed by quenching gas. The results have also demonstrated that the predicted particle are smaller and more sensitive to the modifications of the quenching condition when quenching at high temperature. The sensitivity of the complete model to the physical properties of the vapor (vapor pressure and surface tension) is also investigated, in order to identify their effect on the final particle size. The results obtained provide an insight into the phenomena involved during the production of nanoparticles and enable the improvement of ICTP rectors design and nanoparticles synthesis process. / Doctorat en Sciences de l'ingénieur et technologie / info:eu-repo/semantics/nonPublished