CdSe semiconductor nanocrystals (quantum dots--QDs) with diameters ranging between 1.5 and 8 nm exhibit strong, tunable luminescence [1-5]. They have been widely investigated for their size-dependent optoelectronic properties [6], and for their potential use in optical devices [7], biological labels [8] and sensors [9]. Luminescent quantum dots (QDs) show higher photostability and narrower emission peaks compared to organic fluorophores [8]. The objective of my project was to apply QDs magnetic/luminescent nanoparticle as biological labels in cells. Luminescent CdSe QDs emit bright visible light with high quantum yield and sharp emission peak. The CdSe QDs were capped with a ZnS layer. This increased their emission efficiency and photostability due to the larger band gap of ZnS. The QDs were transferred from organic solvent (e.g. chloroform, hexane) to water by exchanging the capping group (Trioctylphosphine Oxide—TOPO) with mercaptoacetic acid. To develop a separation and detection tool for cells, we combined γ-Fe2O3 magnetic particles with CdSe/ZnS QDs in core-shell composite. The composite nanoparticles showed strong fluorescence emission and high water solubility. Different antibodies were attached to the particles through EDAC coupling. The antibody-coated particles were used to successfully separate and detect breast cancer cells in blood cells.

This work encompasses the study of a novel thermal driven desalination system to accomplish full separation of water and solute. This process advantageous over other process because it involves zero recirculation and zero liquid discharge, thus having minimum environmental impact. Since this system provides full separation, salts and other valuable products can be obtained in addition to pure water. This system can operate at high energy efficiencies using medium temperature heat source like industrial reject or solar cells. This plant consists of two technologies, the full separation and multi effect distillation which when integrated together 8ive us water and salt separately. Three different configuration of the FS-MED system have been presented, naming concurrent feed, variable feed, and counter current feed. They vary depending on their flow and feed distribution. Numerical procedure has been developed to solve the energy and mass balance equation for steady state condition has been presented.

Mixed-integer programming provides a natural framework for modeling optimization problems which require discrete decisions. Valid inequalities, used as cutting-planes and cuttingsurfaces in integer programming solvers, are an essential part of today’s integer programming technology. They enable the solution of mixed-integer programs of greater scale and complexity by providing tighter mathematical descriptions of the feasible solution set. This dissertation presents new structural results on general-purpose valid inequalities for mixedinteger linear and mixed-integer conic programs. Cut-generating functions are a priori formulas for generating a cutting-plane from the data of a mixed-integer linear program. This concept has its roots in the work of Balas, Gomory, and Johnson from the 1970s. It has received renewed attention in the past few years. Gomory and Johnson studied cut-generating functions for the corner relaxation of a mixedinteger linear program, which ignores the nonnegativity constraints on the basic variables in a tableau formulation. We consider models where these constraints are not ignored. In our first contribution, we generalize a classical result of Gomory and Johnson characterizing minimal cut-generating functions in terms of subadditivity, symmetry, and periodicity. Our analysis also exposes shortcomings in the usual definition of minimality in our general setting. To remedy this, we consider stronger notions of minimality and show that these impose additional structure on cut-generating functions. A stronger notion than the minimality of a cut-generating function is its extremality. While extreme cut-generating functions produce powerful cutting-planes, their structure can be very complicated. For the corner relaxation of a one-row integer linear program, Gomory and Johnson identified continuous, piecewise linear, minimal cut-generating functions with only two distinct slope values as a “simple” class of extreme cut-generating functions. In our second contribution, we establish a similar result for a one-row problem which takes the nonnegativity constraint on the basic variable into account. In our third contribution, we consider a multi-row model where only continuous nonbasic variables are present. Conforti, Cornuéjols, Daniilidis, Lemaréchal, and Malick recently showed that not all cutting-planes can be obtained from cut-generating functions in this framework. They also conjectured a natural condition under which cut-generating functions might be sufficient. In our third contribution, we prove that this conjecture is true. This justifies the recent research interest in cut-generating functions for this model. Despite the power of mixed-integer linear programming, many optimization problems of practical and theoretical interest cannot be modeled using a linear objective function and constraints alone. Next, we turn to a natural generalization of mixed-integer linear programming which allows nonlinear convex constraints: mixed-integer conic programming. Disjunctive inequalities, introduced by Balas in the context of mixed-integer linear programming in the 1970s, have been a principal ingredient in the practical success of mixed-integer programming in the last two decades. In order to extend our understanding of disjunctive inequalities to mixed-integer conic programming, we pursue a principled study of two-term disjunctions on conic sets. In our fourth contribution, we consider two-term disjunctions on a general regular cone. A result of Kılınç-Karzan indicates that conic minimal valid linear inequalities are all that is needed for a closed convex hull description of such sets. First we characterize the structure of conic minimal and tight valid linear inequalities for the disjunction. Then we develop structured nonlinear valid inequalities for the disjunction by grouping subsets of valid linear inequalities. We analyze the structure of these inequalities and identify conditions which guarantee that a single such inequality characterizes the closed convex hull of the disjunction. In our fifth and sixth contributions, we extend our earlier results to the cases where the regular cone under consideration is a direct product of second order cones and nonnegative rays and where it is the positive semidefinite cone. Disjunctions on these cones deserve special attention because they provide fundamental relaxations for mixed-integer second-order cone and mixed-integer semidefinite programs. We identify conditions under which our valid convex inequalities can be expressed in computationally tractable forms and present techniques to generate low-complexity relaxations when these conditions are not satisfied. In our final contribution, we provide closed convex hull descriptions for homogeneous two-term disjunctions on the second-order cone and general two-term disjunctions on affine cross-sections of the second-order cone. Our results yield strong convex disjunctive inequalities which can be used as cutting-surfaces in generic mixed-integer conic programming solvers.

Syftet med studien är att öka kunskap om socialsekreterarnas motiveringar bakom gällande beslut om barns återförening med sina föräldrar, som tidigare bedömts ha en bristande föräldraförmåga. Min frågeställning består av tre frågor: Vilka faktorer kan påverka ett beslut om ett återkallande av placering av barn enligt SoL eller LVU 2 §? Hur förhåller sig socialsekreterarna till möjligheten om utveckling av föräldraförmågan efter placering? Hur stor andel av föräldrar, vars barn blivit placerade enligt SOL eller LVU 2§, har fått omhändertagandet återkallat? Frågorna besvaras med den kvalitativa forskningsintervjun. Halvstrukturerade intervjuer och dokument valdes som huvudsakliga datainsamlingstekniker. Sju socialsekreterare, var av en enhetschef deltog i undersökningen. Två av informanterna intervjuades tillsammans. För analys av empirin har studien använt sig av: anknytnings teori, lagar och riktlinjer, tidigare forskning om placerade barn och deras föräldrar. Resultatet visade en tvetydighet gällande socialsekreterarnas tillämpning av lagar; ett övervägande fokus på vårdnadsöverflyttningar och bristfälliga insatserna kring återförening. Det är väldigt få barn som återförenas med sina föräldrar. Det visade sig att det skiljer sig i bedömningar från socialkontor och socialsekreterare. Ett barns återförening möjlighet bedöms utifrån barnets anknytning i främsta rum, och i andra hand bedöms förälderns utveckling. Föräldern har ofta tre år på sig att bättra sig och hålla så bra kontakt med barnet så att de ska hålla sin plats som primär vårdnadshavare för att ha en eventuell chans till återförening. Större delen av barnen blir uppväxtplacerade, med tanke på vårdnadsöverflyttning och anknytningen. Går föräldern med på en frivillig placering enligt SoL så har man större möjlighet att påverka en hemtagning, men via tvångsvårdslagstiftningen LVU blir det svårare, samarbete med socialsekreteraren ökar möjligheten till återförening.  Det är tydligt att barnavårdsutredningar är under komplex press, man vill göra det bästa för barnet och följa lagstiftning samtidigt som man ska hålla kostnaderna nere.

Nous présentons dans cette thèse une étude du comportement hydrodynamique des nanoparticules en suspension dans un ferrofluide et soumises à un gradient de champ magnétique. Le but est de caractériser la séparation magnétique, d'une suspension colloïdale, dans le cadre d'un projet de conception d'un système de traitement des eaux HGMS . Nous développons un modèle mathématique qui décrit le suivi Lagrangien des nanoparticules dans un fluide porteur. Il est basé sur la description de Fokker-Planck et la description de Langevin tout en tenant compte des interactions hydrodynamiques et magnétiques . Le modèle prend aussi en compte la géométrie des agrégats qui se forment. A partir de la simulation numérique, nous constatons que le temps de séparation dépend fortement de la taille et de la longueur des agrégats qui se forment au cours du processus de séparation. L'étude de la cinétique d'agrégation montre l'existence d'une loi d'échelle dynamique. Dans l'agrégation irréversible, des chaînes linéaires de particules se forment et leur taille moyenne évolue dans le temps avec une loi d'échelle à faible puissance. L'exposant de cette variation avec l'équation de coagulation de Smoluchowski avec un noyau homogène. En utilisant le comportement asymptotique, aux temps longs, d'une solution des équations de Smoluchowski, nous mettons en évidence un temps caractéristique de l'agrégation et nous montrons que ce temps est une variable de similarité dont dépend le temps de séparation. Nous montrons aussi que la loi d'échelle est toujours valide pour des nanoparticules dans un écoulement de Poiseuille et que la taille moyenne suit une loi de puissance en fonction du nombre de Reynolds. / We present in this thesis a study of the hydrodynamic behavior of nanoparticles suspended in a ferrofluid and subjected to a magnetic field gradient. The goal is to characterize the magnetic separation of a colloidal suspension in a project to design a water treatment system HGMS (High Gradient Magnetic Separation) that can also have other applications. We develop a mathematical model that describes the Lagrangian tracking of nanoparticles in a carrier fluid. It is based on the Fokker- Planck and Langevin descriptions while taking account of magnetic and hydrodynamic interactions between particles. The model also takes into account the geometry of the formed aggregates. From the numerical simulation, we find that the separation time depends strongly on the size and length of the aggregates formed during the separation process. The study of the kinetics of aggregation shows the existence of a regime with dynamic scaling. In the irreversible aggregation linear chains of particles are formed and their average size changes over time with a scaling law with a low power. The variation exponent of the average size of chains is consistent with the Smoluchowski coagulation equation with a homogeneous kernel. Using the asymptotic, long-time behavior, of a solution of Smoluchowski equations we highlight a characteristic time of aggregation and we show that this time is a similarity variable on which depend the separation time. We also show that the scaling law is still valid for nanoparticles in a Poiseuille flow and the average size follows a power law as a function of Reynolds number.

Thesis advisor: Rudolph Hon / Thesis advisor: Yvette Kuiper / Chloride-based deicers (NaCl, CaCl₂, MgCl₂), also referred to as road salt, are the most common substances used in maintaining safe roadway surfaces during the winter months. Upon application, road salt reacts with the accumulated snow or ice to form brine equilibrium solutions along the liquidus line in the salt-water system. Dissolved salts dissociate, leading to increased concentrations of the respective ions in nearby soils, surface water, and groundwater. Of the ions present in road salt, chloride has the advantage of tracking all chloride deicers at the same time and since chloride ions are conservative tracers in soils it stays unaffected by ionic exchange interferences. This study explores the mechanisms of chloride return flows by investigating chloride dissolved loads, chloride concentrations in stream waters, seasonal patterns, and changes over the course of four years in two separate watersheds in Massachusetts with differing degrees of urbanization. The chloride tracking technique used in this study is based on calibrated chloride concentrations obtained from specific conductance signals recorded every 15 minutes by automatic recording systems at two locations, one in rural central Massachusetts and the other in urban eastern Massachusetts. These systems are maintained by the USGS, which also provide the simultaneously recorded stream flow datasets. The dissolved chloride load carried by each river is calculated for each single 15-minute interval by multiplying water volume with the corresponding chloride concentration, resulting in a total of over 34,000 data points per annum per site. Hydrograph separation techniques were used to separate dissolved load transported by each river into two separate flow components, event flow resulting from precipitation events, and baseflow resulting from groundwater discharge. Well defined hydrograph baseflow supported periods yield consistent chloride concentrations independent of the season at either urban or rural study sites. Comparison of direct runoff dissolved chloride loads with the total annual dissolved loads suggests that only a small fraction of the deicers actually removed during the overland runoff events and that a minimum of 60% of the total load discharged each year in both urban and rural systems is transported by groundwater. From groundwater recharge by brines rural watersheds are currently retaining as much as 95% of the total chloride applied to roadways each year while urban and suburban watersheds may only retain 75% of the total chloride applied to roadways each year. The increased retention of chloride in rural areas is likely due to the decreased amount of chloride transported during winter seasons as event flow compared to urban watersheds. / Thesis (MS) — Boston College, 2009. / Submitted to: Boston College. Graduate School of Arts and Sciences. / Discipline: Geology and Geophysics.

A research report submitted to the Faculty of Engineering and the Built Environment, University of the Witwatersrand, Johannesburg, in fulfilment of the requirements for the Degree of Master of Science in Engineering. May, 2016 / Particle size and size range are among the characteristics that affect the segregation of particles in a jig hence they affect the separation efficiency. The effects of these variables on segregation of particles are not fully understood. This work aimed at contributing to knowledge in this area. To better understand how particle size and size range influence segregation, tests were conducted in which the effects of the density and shape of the particles on segregation were minimized by using as the feed material spherical glass beads of uniform shape and density. Batch experiments of two components systems of various particle sizes were conducted under the same set of jigging conditions: the jigging frequency and jigging time were respectively maintained at 60 cycles per minute and 999 seconds (16.65 minutes). The effect of these operating conditions on segregation was not investigated. At the end of each test run, the jig bed was split into horizontal slices and the composition of each slice was determined. The experimental results showed that below a particle size ratio of 1.50:1, the driving force for the segregation of particles, i.e. the particle size difference, was small hence a low degree of segregation was obtained. The degree of segregation increased above this ratio. However, above the size ratio of 2.00:1, interstitial trickling occurred. With the smaller particles tested (8, 6 and 4mm) poor segregation was observed when the size ratios were of 1.50:1 or less along with what is believed to have been remixing due to convective currents within the jig chamber. It was found that the particle size range had a more pronounced effect on size segregation than the particle size. From the results, it can be said that above a size ratio of about 1.50:1, size segregation is very pronounced. This suggests that density separations of real ores, where both the density and size of particles vary, would be impaired if the particle size range of the material fed to the jig exceeds this ratio. However, this needs further confirmation by testing multiple component systems.

La nanofluidique est l’étude du transport de molécules au travers de nanostructures filtrantes dont la taille avoisine l'épaisseur de la double-couche diffuse à la surface du verre. A cette échelle de la centaine de nanomètres, la charge de surface qui induit une exclusion des ions négatifs à l'extérieur du "nanofiltre" produit un effet de rétention des biomolécules. Des études menées sur le transport électrocinétique au sein d’un nanocanal unique ont permis de montrer qu’il était théoriquement possible de concentrer des solutions, même fortement diluées, avec des taux élevés (jusqu’à 103) grâce à un effet de concentration de polarisation. Ce phénomène pourrait être exploité dans de nombreuses applications de diagnostic médical (analyses rapides et précoces d’échantillons bruts), de contrôle qualité (agroalimentaire) ou encore de défense (suivi continu de zones à risques pour la menace terroriste biologique).La modélisation de la dynamique des phénomènes d’électropréconcentration (sous champ électrique) et de rétention (sous gradient de pression) d’un nanocanal unique s’avère extrêmement ardue. Une multitude d’observations, souvent contradictoires quant au profil de préconcentration obtenu, ont été par ailleurs rapportées dans la littérature, avec des points focaux de préconcentration observés parfois du côté anodique ou du coté cathodique pour une même protéine. Certaines expériences observent ces points focaux soit très loin en amont dans le microcanal réservoir soit directement à proximité de l’entrée du nanocanal. C’est dans ce contexte qu’a été développé précédemment un modèle unidimensionnel permettant de prédire le profil de concentration de l’analyte en tout point de la structure MNM(Micro/Nano/Microcanal) proposée. Ce travail de modélisation a démontré l’existence de quatre régimes distincts: deux régimes coté anodique et deux régimes coté cathodique, plus ou moins éloignés du nanocanal. Ce modèle a mis en avant la sélectivité de ce processus vis-à-vis de la mobilité électrophorétique et de la valence des analytes préconcentrés, et il a permis d’appréhender un peu mieux la diversité des expériences rapportées. Cependant, le régime de préconcentration obtenu dépend du bioanalyte étudié. Il serait pourtant intéressant de ne plus être tributaire des caractéristiques intrinsèques de la solution analysée et de ne plus subir le régime obtenu mais, au contraire, d’effectuer ce que l’on pourrait appeler une électropréconcentration sélective du dit analyte. Ceci pourrait permettre d’effectuer deux étapes primordiales de tout diagnostic que sont la séparation et la préconcentration d’un mélange. Pour se faire, nous avons introduit un paramètre expérimental, une composante hydrodynamique (ou surpression), en sus du champ électrique, pour moduler la localisation de la préconcentration.A l’aide d’une technologie "tout verre" récemment brevetée, nous élaborons aujourd’hui des puces intégrant une nanofente dans un long microcanal. Ces puces sont parfaitement isolantes, biocompatibles et présentent une tenue exceptionnelle au cours du temps. Elles sont combinées à un banc expérimental "fait maison" complètement automatisé (interfacé avec Matlab), qui permet de contrôler les différents paramètres imposés. Les données recueillies sont ensuite prétraitées par d’autres codes Matlab que nous avons développés. Grâce à ces divers outils, de nombreuses expériences d’électropréconcentration "classique" (champ électrique seul) et assistée en pression ont été réalisées pour deux bioanalytes modèles : la fluorescéine et la BSA (Bovin Serum Albumin). Elles ont permis de déterminer les différents paramètres influant sur la préconcentration de ces deux analytes et de prouver la sélectivité et l’efficacité de la méthode proposée ici. Des régimes de préconcentration inattendus, stables et présentant des taux élevés ont en effet été obtenus au cours de cette thèse. / Nanofluidics is the study of the transport of molecules through filtering nanostructures whose size approximates the thickness of the diffuse double layer at the surface of the glass. At this scale of hundreds of nanometers, the surface charge induces an exclusion of negative species outside the "nanofilter" and a retention effect of biomolecule. Studies on the electrokinetic transport in a single nanochannel have shown that it was theoretically possible to concentrate solutions, even highly diluted, with high rates (up to 103) thanks to a concentration polarization effect. This phenomenon can be exploited in many medical diagnostic applications (early and fast sample analysis), quality control (food, water) or defense (continuous monitoring of risky areas for biological terrorist threat). Modeling the dynamics of electropreconcentration phenomena (under an electric field) and retention phenomena (under a pressure gradient) of a single nanochannel is extremely difficult. A multitude of observations, often contradictory regarding the obtained preconcentration profile, were also reported with focal points observed sometimes in the anodic side and other times in the cathodic side for the same protein . Some experiments observe these focal points either upstream in the microchannel reservoir or directly at the entrance of the nanochannel. In this context, a one-dimensional model was previously developed to predict the concentration profile of the analyte at any point of the proposed MNM (Micro/Nano/Microchannel) structure. This modeling work has demonstrated the existence of four distinct regimes: two regimes in the anodic side and two regimes in the cathodic side, more or less distant from the nanochannel. This model highlighted the selectivity of the process regarding the electrophoretic mobility and the valence of the preconcentrated analytes and allowed to understand a little better the diversity of reported experiments. However, the obtained regime depends on the bioanalyte. Though it would be interesting not to be dependent of the characteristics of the analyzed solution and, on the contrary, to realize a selective electropreconcentration of the analyte. This could allow to perform two important steps in any diagnosis: the separation and the preconcentration of a mixture. To do so, we introduced an experimental setting, a hydrodynamic component (or pressure) in addition to the electric field to modulate the localization of the preconcentration .Using an "all glass" technology patented in LPN, chips perfectly insulating, biocompatible and with an exceptional resistance over time are now manufactured. These chips presenting a nanoslit in a straight microchannel are combined with a "homemade" experimental set-up fully automated (interfaced with Matlab®) which allow a perfect control of the various experimental parameters. The data obtained during experiments are then preprocessed by other Matlab codes that we have developed. Thanks to these various tools, many electropreconcentration experiments were performed for two bioanalytes: fluorescein and BSA (Bovin Serum Albumin). They have identified different parameters affecting the preconcentration of these analytes and they have demonstrated the selectivity and the efficiency of the method proposed in this thesis. Unexpected and stable preconcentration regime have been obtained with high rates of preconcentration.

PhD, Faculty of Engineering and the Built Environment, University of the Witwatersrand, 2010 / Dual Energy X-Ray Transmission (DE-XRT) Imaging is a multi-sensor technique employed to conduct particle-by-particle sorting. The system makes use of a dual energy x-ray line scan sensor, which generates images of the transmitted x-rays, similar to images generated for suitcase inspection in airport security applications. The dual energy x-ray system allows for rapid approximation of atomic number range, which is utilised to evaluate the mineral and maceral content of a variety of minerals, including coal. The process is independent of particle surface condition, and can thus be utilised as a dry process. A unique application of this technology is in the removal of torbanite from a coal deposit located in Mpumalanga, South Africa. The separation of coal from torbanite has been a problem for the coal industry for a long time. The separation of coal and torbanite by conventional gravity separation techniques is difficult, due to the overlapping densities of torbanite and coal. The commercial value of both commodities is significantly compromised if contaminated with the other, thus impacting negatively on the financial viability of mining such a deposit. Preliminary laboratory DE-XRT testwork results on high quality coal and torbanite products were promising. In order to evaluate the separation of typical Run of Mine (ROM) material on pilot scale, a production scale Mikrosort X-Tract Sorter was purchased. This was the first DE-XRT sorter available in South Africa, and was housed at Mintek in Johannesburg. A 150t sample was provided from a box cut adjacent to the coal deposit under investigation in order to conduct bulk and pilot sorting tests, the focus of which was on obtaining coal products of low ash and torbanite content. Clear distinctions between the coal, torbanite and shale fractions were observed using this technique. The sorter feed (-80mm+20mm) could be upgraded from a CV of 22MJ/kg to 28MJ/kg. Ash content could be reduced from 26% to 10%, which meets export quality standards. Petrographic analysis of the coal product indicated that a high purity coal product (in terms of torbanite and ash content) was attainable (91% by volume) at a mass yield of 42.9% to the coal product, with shale and mixed humic/sapropelic coal as contaminants. Under these conditions, torbanite contamination was marginal. It was demonstrated that shale could be removed from the torbanite product via a second sorting stage. This however was not the primary focus of the study, and was not optimised for this investigation. Two major limitations of the sorting process were identified, viz.; poor liberation and limited sorter feed size range. These impacted on the process as follows:- • The effects of poor liberation on coal quality could be counteracted by adjusting the sorting criteria of the algorithm to reject additional material. This would result in a lower coal product mass yield. In addition, interlocked coal/shale particles would report to the torbanite fraction. • A significant proportion of the ROM feed reported to the -20mm size fraction, and therefore did not fall part of the sorter feed. This resulted in a very low coal mass yield as a proportion of the ROM feed. If this process were to be adopted, means of minimizing fines production during mining and crushing would need to be investigated to improve overall yield to coal product. The capability to process coarse materials (-80mm+20mm) allows for throughputs in excess of 40t/hr. Consequently, this technique may be applied in simpler coal upgrading processes, such as coal deshaling in arid regions.

Um suporte catalítico superparamagnético nanoestruturado do tipo \"core-shell\" constituído de núcleos de magnetita revestidos por sílica, obtido por uma microemulsão reversa, foi utilizado como plataforma para o ancoramento de cátions de metais de transição que serviram como precursores na obtenção de nanopartículas de Rh(0), Pt(0) e Ru(0). A superfície do suporte de sílica foi funcionalizada com um aminosilano que permitiu um aumento significativo na quantidade dos íons metálicos sequestrada das soluções aquosas dos sais dos metais estudados. Os nanocatalisadores foram empregados em reações de hidrogenação de alquenos e cetonas em fase líquida e puderam ser facilmente separados pela aplicação de um campo magnético, que foi realizada pelo contato de um imã de neodímio com a parede do reator contendo o catalisador e o produto. A técnica de separação magnética utilizada foi capaz de isolar completamente o sólido da fase líquida, fazendo com que a utilização de outros métodos de separação como filtração e centrifugação, comumente utilizados em sistemas heterogêneos líquidos, fossem completamente dispensados. As reações de hidrogenação foram realizadas utilizando-se hidrogênio molecular como agente redutor e dispensou a utilização de agentes redutores mais drásticos como hidretos metálicos que não atendem aos princípios verdes em demanda na nossa sociedade.Todos os sólidos catalíticos desenvolvidos mostraram uma excelente possibilidade de reutilização que comprovou a estabilidade da fase ativa do catalisador, destaca-se a hidrogenação do benzeno pelo catalisador magnético de Rh(0) que pôde ser utilizado por até 20 vezes (número total de rotação igual a 10.240), sem queda significativa de atividade, com freqüência de rotação chegando a 1.167 h-1. A ocorrência de lixiviação de espécies cataliticamente ativas, que é comum em catálise heterogênea em fase líquida, não foi observada nas reações de hidrogenação estudadas. Este fato pode ser atribuído às condições reacionais brandas utilizadas, à forte aderência das nanopartículas metálicas ao suporte funcionalizado com grupos NH2 e ao eficiente método de separação magnética empregado. O procedimento de separação catalisador-produto, além de não fazer uso de métodos físicos de separação mais agressivos como uma centrifugação, permitiu o isolamento dos componentes dentro do próprio reator, descartando a exposição do catalisador à condições atmosféricas e o uso de solventes extras durante o procedimento de separação. / A core-shell superparamagnetic catalytic support comprised of magnetite nanoparticles recovered by silica was obtained using a reverse microemulsion. The material was used as a framework for anchoring transition metal cations that were used for the fabrication of Rh(0), Pt(0) and Ru(0) nanoparticles. The catalysts were employed in the hydrogenation of alkenes and ketones in liquid phase and they showed to be easily recoverable from liquid systems by placing a small neodymium magnet on the reactor wall. The magnetic separation technique provided a complete isolation of the catalyst from the liquid phase containing the products. It also made the use of other separation techniques, commonly used in achieving product separation in liquid-solid heterogeneous systems, such as filtration and centrifugation completely unnecessary. The surface of the silica support was modified with an aminosilane leading to a substantial increasing in the metal uptake from aqueous solutions of Rh, Pt and Ru salts. The hydrogenation reactions were carried out using molecular hydrogen as the reducing agent and they did not require any drastic reducers such as metal hydrides that would go against the principles of green chemistry. The stability of the catalysts were evidenced by the outstanding recycling properties. A single portion of the Rh0 catalyst for instance could be used for up to 20 times in the hydrogenation of benzene (TON: 10,240), and no significant loss in the catalytic activity was observed giving TOF values of up to 1167 h-1. Leaching of active catalytic species which is commonly encountered in heterogeneous solid-liquid systems was also absent in the hydrogenation reactions studied and this finding can be attributed to the mild reaction conditions used, the adherence of Rh, Pt and Ru nanoparticles to the magnetic support functionalized with NH2- groups and the efficient magnetic separation method used in isolating the product that dismiss the use of extra solvents and more aggressive separation methods such as a centrifugation.