Le biodiesel, provenant de sources renouvelables, devient de plus en plus intéressant pour des raisons environnementales et à cause de la perspective de la pénurie des carburants fossiles. La méthode principale pour la production de biodiesel est la réaction de transestérification de l'huile végétale catalysée par des catalyseurs acides, basiques ou enzymatiques en phase homogène ou hétérogène. Le but de ce travail est d'étudier une nouvelle voie de production de biodiesel à partir de l'huile de colza en utilisant un liquide ionique comme catalyseur et le CO2 supercritique comme milieu réactionnel. Les liquides ioniques fonctionnalisés sulfoniques sont des liquides ioniques acides possédant de bonnes propriétés catalytiques pour la transestérification de l'huile. Le caractère lipophile du liquide ionique ainsi que l'alcool influence la performance de la réaction. L'éthanol et le 1-propanol sont plus réactifs que le méthanol à cause de leur meilleure solubilité dans l'huile. Le liquide ionique peut être recyclé et la séparation des esters d'acides gras a lieu par simple décantation. L'effet du CO2 sur la transestérification de l'huile de colza catalysée par les liquides ioniques n'est pas clairement positif à cause de l'extraction préférentielle de l'éthanol. Cependant, en rajoutant l'éthanol en continu, l'influence du CO2 devrait être plus favorable. Le CO2 pourra être également utilisé en fin de réaction comme solvant permettant de séparer des esters, du glycérol et du catalyseur. / Biodiesel, resulting from renewable sources, becomes more and more interesting for environmental reasons and because of the perspective of the shortage of fossil fuels. The main method for the production of biodiesel is the reaction of transesterification of vegetable oil catalyzed by acid, basic or enzymatic catalysts in homogeneous or heterogeneous phase. The purpose of this work is to study a new way of biodiesel production from the rapeseed oil by using an ionic liquid as catalyst and supercritical CO2 as reaction media. The HSO3 functionalized ionic liquids are acid ionic liquid possessing good catalytic properties for the oil transestefication. The lipophilic character of ionic liquid as well as alcohol influences the performance of the reaction. Ethanol and 1-propanol are more reactive than methanol because of their best solubility in the oil. The ionic liquid can be recycled and the separation of the esters of fatty acids takes place by simple settling. The effect of CO2 on the transesterification of rapeseed oil catalyzed by ionic liquids is not clearly positive because of the preferential extraction of the ethanol. However, by adding the ethanol continuously, the influence of CO2 should be more favorable. Supercritical CO2 can be also used to separate esters, glycerol and catalyst at the end of reaction.

L'idée à la base de ce projet est de mettre à profit le confinement nanométrique unidimensionnel pour décupler la conduction ionique des électrolytes et donc la puissance des accumulateurs au lithium. Nous avons concentré nos efforts sur une classe particulière d'électrolytes qui, en raison de leurs stabilités physique et électrochimique, ont été identifiés comme prometteurs: les Liquides Ioniques (LI). Nous avons confiné les LI dans deux systèmes poreux présentant une topologie commune (pores cylindriques macroscopiquement orientés) mais aux propriétés physico-chimiques complémentaires: i) des alumines poreuses (AAO, interface hydrophile, diamètre des pores de 25 à 160 nm) et ii) des membranes de NanoTubes de Carbone (NTC, interface hydrophobe, diamètre des pores 4 nm).Nous avons développé un modèle microscopique multi-échelle original, prenant en compte la dynamique complexe des cations des LI: combinaison i) de la dynamique de réorientation rapide des chaînes latérales alkyle, ii) de la diffusion de la molécule au sein des agrégats nanométriques spontanément formés par les LI, puis iii) de la diffusion entre ces agrégats. Ce modèle reproduit de façon remarquablement robuste les données de diffusion quasi-élastique de neutrons sur une gamme étendue de vecteurs de diffusion (0,1 à 2,5 Å-1) et de temps (10-1 à 2.103 ps). A cette échelle locale, nous ne détectons pas d'influence du confinement sur la dynamique du LI confiné au sein des AAO et des CNT. Nous montrons cependant qu'à l'échelle microscopique (PFG-NMR) et macroscopique (spectroscopie d'impédance) le confinement des LI au sein des NTC permet d'obtenir un gain de conductivité d'un facteur 3. Un brevet est déposé. / The idea behind this project is to exploit the 1D nanometric confinement in order to increase the electrolytes ionic conductivity and, thus, the power of the lithium accumulators. We have focus on a specific class of electrolytes, which, owing to their physical and electrochemical stabilities, have been identified as very promising: the Ionic Liquids (ILs). We have confined the ILs in porous systems having a common topology (cylindrical pores macroscopically oriented), but with complementary physico-chemical properties: i) the porous alumina (AAO, hydrophilic interface, pores diameter between 25 and 160 nm) and ii) Carbon NanoTubes based membranes (CNT, hydrophobic interface, pores diameter of 4 nm).We have developed an original microscopic multiscale model, taking into account the complex dynamics of ILs cations: combination of i) fast reorientation dynamics of side alkyl-chains, ii) molecule diffusion within nanometric aggregates spontaneously formed in the ILs and iii) diffusion between the aggregates. This model reproduces in a very robust way the quasi-elastic neutrons scattering data on an extent interval of wave vector (0.1 à 2.5 Å-1) and time (10-1 à 2.103 ps). At this local scale, we do not observe any influence due to the confinement on the dynamics of the ILs confined in the AAO and CNTs. We show however that at microscopic (PFG-NMR) and macroscopic (impedance spectroscopy) scale the ILs confinement within the NTCs allows to obtain a conductivity gain of factor 3. A patent is filed.

L’arthrose affecte une dizaine de millions de français. La présence de microcristaux calciques notamment ceux de pyrophosphates de calcium (CPP) dihydratés (Ca2P2O72H2O ; mono- et/ou triclinique : respectivement m- et/ou t-CPPD) au sein de l’articulation est l’un des facteurs aggravant identifié de cette maladie. Cependant, les conditions et mécanismes de formation de ces cristaux ainsi que l’ordre d’apparition de ces deux phases de CPPD restent méconnus ou n’ont pas encore été entièrement décrits et les traitements existants sont seulement symptomatiques. Dans ce mémoire nous nous sommes intéressés à l’étude de l’effet d’additifs ioniques sur la formation et les cinétiques de croissance cristalline de cristaux de CPP qui constitue une étape clé pour la mise au point à moyen ou long terme d’une stratégie thérapeutique visant à empêcher la formation, transformer ou dissoudre ces cristaux. La synthèse in vitro par précipitation des phases pures de CPP hydratés d’intérêt biologique (m-CPPD, t-CPPD et CPP monoclinique tétrahydraté ) a été réalisée afin d’identifier les paramètres opératoires conduisant à leur apparition et d’analyser l’effet d’additifs ioniques (Mg2+, Cu2+, Zn2+, Fe3+ ou S2O32-) lors de leur précipitation en conditions maîtrisées. La méthode de synthèse établie par Gras et al. a été transposée en réacteur agité. Il s’agit d’une précipitation avec ajout simultané de deux solutions mères respectivement de nitrate de calcium et de pyrophosphate de potassium dans une solution tampon d’acétate d’ammonium. Les composés obtenus ont été caractérisés par diffraction des rayons X, résonance magnétique nucléaire du solide, spectroscopies FTIR et RAMAN, microscopie électronique à balayage (MEB) et spectrométrie à plasma à couplage inductif. Grâce à une analyse semi-quantitative par diffraction des RX, nous avons montré que la présence d’additifs ioniques (Mg2+, Zn2+, Fe3+ ou Cu2+) lors de la précipitation change significativement les proportions des phases de CPP du produit final. Ainsi nous montrons que les ions Mg2+, Cu2+ ou Zn2+ orientent la formation de cristaux de m-CPPD lors de la précipitation en conditions maîtrisées de t-CPPD, et selon les conditions l’ion Fe3+ peut totalement inhiber la cristallisation de t-CPPD. Des analyses fines par RMN du solide 31P ont permis de déterminer les modes d’action de ces additifs. Les ions Cu2+ et Mg2+ ont notamment agi par substitution au Ca2+ dans la maille de CPP. Nous avons aussi mis en œuvre la méthode de croissance cristalline à composition constante pour étudier la germination et la croissance cristalline à température, pH, sursaturation et force ionique constants de t-CPPD sur une semence de t-CPPD en présence ou non d’additif ionique. La semence est introduite dans une solution sursaturée métastable de CPP à pH 6,2 incluant l’additif ionique et placée dans un réacteur agité et thermostaté à 37°C. La sursaturation du milieu est calculée à l’aide du logiciel Phreeqc et les temps de germination et vitesses initiales de croissance cristalline ont été déterminés. Les cristaux ont été analysés par spectroscopie FTIR et MEB. Nous avons mis en évidence l’influence d’additifs ioniques (Mg2+, Zn2+ ou Cu2+) sur les paramètres cinétiques de germination et croissance cristalline de t-CPPD. Ainsi les ions Cu2+ et Mg2+ ont respectivement inhibé partiellement et totalement la croissance cristalline de t-CPPD. Les résultats obtenus dans cette étude pourraient aider à expliquer la formation in vivo de cristaux de m-CPPD au sein des articulations de patients arthritiques alors que cette phase pure est difficile à obtenir in vitro par précipitation sans ions autres que Ca2+ et P2O74-. Ces résultats cinétiques pourront donner une indication sur l’évolution du ratio des deux phases (m-CPPD et tCPPD) formées in vivo dans les articulations arthrosiques et des pistes de traitement pour éviter la formation de ces cristaux inflammatoires ou orienter la formation de phases non inflammatoires. / Osteoarthritis (OA) affects about ten million people in France. The presence of microcrystals including those of calcium pyrophosphate (CPP) dihydrates (Ca2P2O7•2H2O; mono- and/or triclinic: m-CPPD and/or t-CPPD respectively) in the joints has been identified as an aggravating factor of OA. However the conditions of formation of both CPPD phases and their order of appearance remain unknown or have not been totally described and the existing treatments are only symptomatic. In the present work we were interested in the effect of ionic additives on the formation and the kinetics of the crystallization of CPP which is a key step in the medium to long term development of a therapeutic strategy to prevent the formation of these crystals, their transformation or their dissolution. At first the in vitro synthesis by precipitation of hydrated CPP pure phases of biological interest (mCPPD, t-CPPD and monoclinic CPP tetrahydrated ) have been realized to identify the operating parameters (concentration, pH, temperature) leading to their formation and to analyze the effect of ionic additives (Mg2+, Cu2+, Zn2+, Fe3+ or S2O32-) on their precipitation under controlled conditions. To achieve this objective, the synthesis method developed by Gras et al. was transposed in stirred reactor. It consists in a precipitation by simultaneous addition of two reagent solutions of calcium nitrate and potassium pyrophosphate in an ammonium acetate buffer solution. Powders synthesized were analyzed by X-ray diffraction, solid state nuclear magnetic resonance, FTIR and RAMAN spectroscopies, scanning electron microscopy and inductively coupled plasma atomic emission spectroscopy. Thanks to a semiquantitative analysis by X-ray diffraction, we showed that the presence of ionic additives (Mg2+, Zn2+, Fe3+ or Cu2+) during the precipitation significantly modified the proportions of CPP phases constituting the final product. We showed that Mg2+, Cu2+ or Zn2+ promote the formation of m-CPPD crystals during the precipitation under controlled conditions for t-CPPD phase synthesis and depending on the conditions Fe3+ ion can completely inhibit t-CPPD phase crystallization. 31P solid state NMR fine analyzes have allowed determining the mode of action of these ionic additives. Thus, Cu2+ or Mg2+ ions have substituted for Ca2+ in the CPP lattice. Additionally, we have implemented the method of constant composition crystal growth to study the nucleation and crystal growth at constant temperature, pH, supersaturation and ionic strength of tCPPD on seeds of t-CPPD crystals in the presence or not of ionic additive. The seed was introduced into a supersaturated metastable solution of calcium pyrophosphate at pH 6.2 including the ionic additive and placed in a stirred reactor maintained at 37°C. Supersaturation of the medium was calculated using Phreeqc software and nucleation time and the initial rate of crystal growth were determined. The resulting crystals were analyzed by FTIR spectroscopy and scanning electron microscopy. We were able to highlight the influence of ionic additives (Mg2+, Zn2+ or Cu2+) on the kinetic parameters of nucleation and crystal growth of t-CPPD. Thereby, Cu2+ and Mg2+ ions have respectively inhibited partially and totally the growth of the tCPPD crystals. These results may help to explain why m-CPPD phase is identified in vivo in the joint of arthritic patients while this pure phase is difficult to obtain by in vitro precipitation (very narrow temperature-pH domain in a medium free of any ionic additives). These results could also provide indications on the order of appearance and potential evolutions of m-CPPD and tCPPD phases formed in vivo in osteoarthritic joints and provide some clues to avoid the formation of CPPD crystals with inflammatory potential or orient the formation of noninflammatory CPP phases

In organic-inorganic hybrid compounds an organic and an inorganic component are combined to form either a coordination or an ionic material. Relevant to the current study are hybrid materials composed of an organic part that contains one or more functional groups, for example amine, amide or carboxylic acid functional groups, and a metal halide inorganic portion. These functional materials display a range of interesting and desired properties, as evidenced from numerous literature reports on their properties. In order to utilise these properties in applications, a detailed understanding of the way that the crystal structure influences the properties of a material is required. However, before this step can be achieved, it is necessary to obtain information on the structural trends of the materials, and to use the approach of crystal engineering to identify robust supramolecular synthons that may afford structural control and prediction. The aim of the current study was to investigate the synthesis and crystal structures of hybrid materials, both ionic and coordination, composed of divalent transition metal halides and the organic components 4-aminobenzoic acid, 4-aminobenzamide and isonicotinic acid, and to identify the structural trends and crystal engineering synthons displayed by these materials. A secondary objective was the preliminary identification of properties exhibited by selected materials, in order to decide on the suitability of the materials for detailed future property investigations. Part of the work describes the investigation of the structural characteristics of coordination materials prepared by the combination of the organic and inorganic components. Five novel crystal structures of coordination materials were determined, and these are compared with six related coordination structures reported in the literature. Two of the novel structures display interesting one-dimensional coordination polymers, one of which has never been reported previously in the literature. The molecular and structural characteristics of both the novel and the literature coordination structures are presented in detail, and this discussion includes a description of the coordination geometry, the molecular geometry, packing trends, hydrogen bonding interactions and aromatic interactions. A comparison study across the three families of organic components in which the structural trends, hydrogen bonding interactions, aromatic interactions, ligand geometry and coordination modes are compared, is included. The results of the synthesis of the coordination materials by means of a mechanochemical method are presented, and the products afforded by this method are compared with those prepared via solution crystallisation. Finally, the results of preliminary studies of the thermal and electronic propertries of the materials are presented and interpreted. The combination of the hybrid components as cations and anions to form ionic materials yielded nine novel structures, and these were compared with five related ionic structures reported in the literature. The novel structures include three polar structures that contain the 4-ammoniumbenzamide cation, and to our knowledge no structures containing this cation have ever been reported in the literature, hence a significant contribution to the structural knowledge of perhalometallate salts of 4-ammoniumbenzamide is made by this study. In addition two novel structures display interesting one-dimensional and two-dimensional polymeric anions, respectively, are reported. The discussion of the novel and literature ionic structures includes a description of the molecular geometry of each of the components, the identification of packing trends, and an analysis of the hydrogen bonding and aromatic interactions occurring in the structures. The structures of all three families of organic components are compared, and trends in structural type, anion geometry, water inclusion, hydrogen bonding and functional group recognition are presented. In addition, a detailed analysis of robust crystal engineering synthons occurring in these structures is presented. Lastly the results of preliminary property investigations of the thermal and electronic properties of the materials are presented and discussed / Dissertation (MSc)--University of Pretoria, 2011. / Chemistry / unrestricted

Lithium Air (Li/O2) batteries are energy conversion devices that produce electricity from the oxidation of lithium metal at the anode and the reduction of molecular oxygen at the cathode. These batteries are considered as promising rechargeable cells for high power applications due to their high power density ranging from 1000 to 2000 Wh/kg. However, one of the most significant challenges is the need to separate the metallic lithium anode from any oxygen or water-containing environment while at the same time allowing fast and efficient lithium ion transport through the electrolyte. Therefore, lithium ion conducting materials that are water and CO2 resistant are a prerequisite. Common materials used as anode protective films and/or Li+ conducting electrolytes for lithium air batteries are perovskite-type oxides (formula: ABO3). Perovskites are good candidates for this application because of their versatility, particularly in regards to ionic conductivity. In the present work, a low cost perovskite family such as SFO (SmFeO3) is developed as a lithium ion conducting material by the introduction of Li+ into its lattice. The perovskites have been synthesized using a solid-state reaction method (SSR) and characterized using different techniques such as powder X-ray diffraction (PXRD), scanning electron microscopy (SEM), energy dispersive X-ray Spectroscopy (EDS) and electrochemical impedance spectroscopy (EIS). The synthesized perovskites are based on samarium lithium ferrite and divided into two groups depending on the formal presence of vacancies in the stoichiometric formula. The first group (SLFO) with no formal vacancies has the stoichiometric formula of SmxLi1-xFeO2+x (where x = 0.1, 0.2, 0.3, 0.5 and 0.7). While the second group (SLFO*) was generated with less metal atoms than specified in the perovskite structure, thereby generating a structure with intrinsic vacancies and with the formula, Sm(x)Li([1-x] – [0.1] or [0.2]) FeO3-δ (where x = 0.3, 0.4, 0.5 and 0.6). Finally, the effect of varying Li and Sm concentrations in both groups and vacancies created in the lattice for the second group, on the ionic conductivity is explored.

Ces travaux portent sur l'élaboration de catalyseurs hétérogènes en milieu liquide ionique (LI). Ces sels liquides à température ambiante permettent d'ajuster les propriétés du solvant suivant la composition chimique du LI mis en jeu (sels d'imidazolium ou Deep Eutectic Solvent à base de chlorure de choline et urée). Ce type de solvant a permis la synthèse et le dépôt de nanoparticules (NPs) d'Au et Au-Pd sur TiO2. Le type de LI utilisé influence la stabilité des NPs en solution, la force de l'interaction métal/support et la nanostructuration des particules bimétalliques. Les performances catalytiques en hydrogénation sélective du butadiène ont montré une amélioration de l'activité des catalyseurs monométalliques à base d'Au en présence de Pd, une promotion de l'activité des catalyseurs (mono-et bimétalliques) par des espèces phosphorées résiduelles issues du LI et une inhibition de l'activité par des espèces soufrées. Des supports TiO2 ont également été préparés en milieu DES. L'utilisation de ce LI et d'un précurseur de Ti spécifique a permis de contrôler la texture et la structure du polymorphe obtenu (anatase, rutile ou mélange anatase-rutile). L'influence de la nature du support a été étudiée en oxydation du CO après dépôt d'Au par dépôt-précipitation à l'urée. Les catalyseurs Au/TiO2 les plus actifs ont été obtenus pour des mélanges anatase-rutile, la proximité entre phases anatase et rutile menant à un optimum de réactivité et de stabilité. / This work deals with the elaboration of heterogeneous catalysts in ionic liquids (ILs). These salts, liquid at room temperature, were chosen because they permit to adjust the solvent properties depending on their chemical composition (imidazolium salts or Deep Eutectic Solvent based on choline chloride and urea). These solvents allowed the synthesis and deposition of Au and Au-Pd nanoparticles (NPs) on TiO2. The IL nature controls the NPs stability in solution, the strength of the metal/support interaction and the nanostructuration of bimetallic particles. The catalysts performances, evaluated by selective hydrogenation, showed an increase in activity of the Au monometallic catalysts after addition of Pd, a promotion of the catalysts activity due to the presence of P residues from the ILs and an inhibition of the activity caused by S species. TiO2 supports were also prepared in DES. The use of this IL, in addition to a specific Ti precursor, led to a textural and structural control of the obtained polymorphs (anatase, rutile or anatase-rutile mixture). The influence of the support type was studied in CO oxidation after Au deposition by urea deposition-precipitation. The most active Au/TiO2 catalysts were obtained with anatase-rutile mixtures, the vicinity between anatase and rutile phases leading to an optimum activity and stability.This work deals with the elaboration of heterogeneous catalysts in ionic liquids (ILs). These salts, liquid at room temperature, were chosen because they permit to adjust the solvent properties depending on their chemical composition (imidazolium salts or Deep Eutectic Solvent based on choline chloride and urea). These solvents allowed the synthesis and deposition of Au and Au-Pd nanoparticles (NPs) on TiO2. The IL nature controls the NPs stability in solution, the strength of the metal/support interaction and the nanostructuration of bimetallic particles. The catalysts performances, evaluated by selective hydrogenation, showed an increase in activity of the Au monometallic catalysts after addition of Pd, a promotion of the catalysts activity due to the presence of P residues from the ILs and an inhibition of the activity caused by S species. TiO2 supports were also prepared in DES. The use of this IL, in addition to a specific Ti precursor, led to a textural and structural control of the obtained polymorphs (anatase, rutile or anatase-rutile mixture). The influence of the support type was studied in CO oxidation after Au deposition by urea deposition-precipitation. The most active Au/TiO2 catalysts were obtained with anatase-rutile mixtures, the vicinity between anatase and rutile phases leading to an optimum activity and stability.

Hydrogels based on self-assembling peptides are believed to have potential for use in a wide range of biomedical and biodiagnostic applications. For many of these, control over various properties of the gels is essential for tuning the gels to fit certain constraints or requirements in terms physical properties such as diffusive properties and swelling. One important property to control for applications such as cell culture and drug delivery is its mechanical strength, and this study investigates three different strategies by which the individual peptide monomers can be modified in order to effect a change in the macromolecular self-assembled structure and therefore a bulk change in the mechanical stiffness. In chapter 4, two ionic-complementary octapeptides, FEFKFKFK and FEFQFKFK are described, with monomer charges of +2 and +1, respectively at physiological pH. FEFKFKFK was observed to form largely discrete fibrils, characteristic of similar systems, while FEFQFKFK formed fibril bundles – believed to be a limited form of an aggregation effect frequently seen in similar peptides with neutral charge. As a result of this structural change, FEFQFKFK was found to have values for the elastic and viscous moduli (which are often used to measure the ‘strength’ of a gel) between 5 and 10 times larger than those of FEFKFKFK at the same concentration. The same behaviour was seen in FEFKFKFK when the monomer charges were reversed by adjusting pH, suggesting that the monomer charge is indeed responsible for the bundling effect. In chapter 5, two branched peptides were designed and synthesized: KG17, with two arms consisting of self-assembling FKFEFKFK-motifs, and KG28 which had three such arms. Each branched peptide was doped into pure FKFEFKFK and the resulting gels investigated. While no obvious structural changes were observed for either dopant (save for a potential fibril parallelisation effect with KG17 observed in Small-Angle Neutron Scattering (SANS)), both were observed to increase the elastic and viscous moduli of the gels at overall peptide concentrations of 30 and 50 mg mL-1 (gels), but not at 10 mg mL-1 (viscous liquid). The most dramatic change was observed in the 50 mg mL-1 gels, suggesting that higher concentrations could enhance the effect of the dopants. In chapter 6, three thermo-responsive polymers (pTEGMA), of Degrees of polymerisation (DPs) 17, 47 and 142 were conjugated to CGFKFEFKFK and incorporated into a peptide hydrogel. Gels containing the non-conjugated versions of each polymer were also tested. While no changes in morphology were observed at the fibillar level, the polymer Lower Critical Solution Temperature (LCST) behaviour could be observed in SANS in all samples apart from the DP17 conjugate. However, in rheological tests gels doped with this conjugate appeared to show the strongest the elastic and viscous moduli. In general the conjugates appeared to increase the elastic and viscous moduli, particularly at temperatures above ca. 50°C. Rather than this being LCST behaviour, it was suggested that the polymers can act to enhance a natural thermo-response that was observed in the peptide, with the shortest polymer (DP17) experiencing the least steric hindrance and therefore having the strongest effect. It was postulated that this interaction could involve the screening of charge on the peptide fibril. Non-conjugated polymer appeared to have little effect on the mechanical properties, with elastic modulus values correlating strongly to the overall peptide concentration.

Atualmente, pelo menos para fins de engenharia, os modelos preditivos de maior sucesso para coeficientes de atividade são os baseados em grupos funcionais, tais como UNIFAC e suas variantes. Enquanto esses modelos requerem grandes quantidades de dados experimentais, os baseados em COSMO-RS (COnductor-like Screening MOdel - for Real Solvents) requerem a calibração de um pequeno conjunto de parâmetros universais. No entanto, a precisão requerida por tarefas de engenharia, tais como a otimização de sistemas de separação, é maior do que a obtida por esta última categoria de modelos. Assim, um novo modelo é proposto neste trabalho, aqui chamado de F-SAC (Functional-Segment Activity Coefficient). Este novo modelo também é baseado no conceito de grupos funcionais, mas a energia de interação entre os grupos vem da teoria COSMO-RS. No presente trabalho, foram consideradas apenas misturas em que não há formação de ligação de hidrogênio ou quando esta pôde ser assumida negligenciável. Assim, foram necessários apenas três parâmetros para descrever cada grupo funcional. A princípio, uma vez ajustados os parâmetros de cada grupo, estes funcionariam para descrever a interação para qualquer par de grupos. Esta é a principal vantagem do modelo proposto. O número de parâmetros do modelo cresce proporcionalmente ao número de grupos funcionais, enquanto que no UNIFAC o número de parâmetros de interação cresce proporcionalmente ao quadrado do número de grupos. Para o banco de dados experimentais de coeficientes de atividade em diluição infinita considerado, a correlação do F-SAC apresentou um erro médio absoluto de 0,07 unidades de ln, enquanto que os modelos UNIFAC (Do) e COSMO-SAC apresentaram, respectivamente, erros de 0,12 e 0,21. O F-SAC foi também avaliado para mais de 1000 misturas binárias de um soluto dissolvido em líquido iônico com dados disponíveis na literatura. O modelo apresentou uma boa correlação aos dados experimentais, com erro médio absoluto de 0,17 unidades de ln, similar ao apresentado pelas misturas orgânicas. O poder de predição do novo modelo foi avaliado utilizando dados de equilíbrio líquido-vapor não considerados no procedimento de ajuste do modelo. Uma ótima concordância com os dados experimentais foi possível em toda a faixa de composição, bem como na predição de azeótropos. Esses resultados demonstram o potencial do modelo proposto. / At present, at least for engineering purposes, the most successful predictive models for activity coefficients are those based on functional groups, such as UNIFAC and its variants. While these models require large amounts of experimental data, the ones based on COSMO-RS require the calibration of a small set of universal parameters. However, the resolution required by engineering tasks, such as the optimization of separation systems, is higher than that obtained by COSMO-RS models. Thus, in this work a novel Functional-Segment Activity Coefficient (F-SAC) model is proposed. This new model is also based on the concept of functional groups, but the interaction energy between groups comes from the COSMO-RS theory. In this study, we considered only mixtures where there is no formation of hydrogen bonds or when they could be assumed negligible. Then, only three parameters were required to describe each functional group. In principle, once the parameters for each functional group are properly calibrated, they should work to describe the interactions with any other group. This is the main advantage of the proposed model, the number of model parameters grows linearly with the number of functional groups, whereas in UNIFAC the number of interaction parameters exhibit quadratic growth with respect to the number of groups. For the experimental database of infinite dilution activity coefficients considered, the correlation of F-SAC have shown a mean absolute error of 0.07 ln-unit. The UNIFAC (Do) and COSMO-SAC models presented errors of 0.12 and 0.21, respectively. F-SAC was also evaluated for more than 1000 binary mixtures of solute in ionic liquid with data available in the literature. Again, the model have shown good correlation to the experimental data, with mean absolute error of 0.17 ln-unit, similar to the performance with the organic mixtures. The predictive strength of the model was assessed by using vaporliquid equilibrium data not considered in the model fitting process. Very good agreement with experimental data was possible over the entire composition range, as well as in the prediction of azeotropes. These results demonstrate the potential of the proposed method.

Orientador: Carlos Raimundo Ferreira Grosso / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia de Alimentos / Made available in DSpace on 2018-08-22T05:50:23Z (GMT). No. of bitstreams: 1 Almeida_TalitaEmanuelaMeloe_M.pdf: 2060488 bytes, checksum: 4adc0b3a48b5ba83103d990622a1286a (MD5) Previous issue date: 2013 / Resumo: Este estudo teve como objetivo a produção de micropartículas por técnicas combinadas de gelificação iônica e interação eletrostática utilizando alginato de sódio, concentrado proteico do soro do leite (WPC) e manteiga fundida para imobilização do micro-organismo Erwinia sp. D12 e avaliação da conversão de sacarose em isomaltulose por células íntegras livres e imobilizadas em processo descontínuo durante quatro dias. No estudo preliminar, os biopolímeros alginato de sódio e WPC foram caracterizados para produção de micropartículas com boas propriedades. A análise da carga elétrica disponível de alginato de sódio em solução ou emulsão com diferentes concentrações de manteiga fundida e das proteínas do WPC em solução indicou que em uma faixa de pH de 3,0 - 4,5 pode ocorrer uma interação de natureza eletrostática entre os biopolímeros. Misturas formadas em diferentes proporções entre emulsão de alginato de sódio com diferentes concentrações de manteiga fundida e solução proteica de WPC foram caracterizadas com relação à carga elétrica e tamanho médio dos coacervados insolúveis formados. Maiores tamanhos de coacervados, variando entre 29,12 µm a 33,72 µm, foram obtidos na proporção volumétrica 1:6 (emulsão de alginato e manteiga fundida:WPC) sendo que a adição de diferentes concentrações de manteiga fundida não modificou o tamanho dos coacervados formados. As micropartículas de alginato contendo manteiga fundida foram preparadas pela técnica de gelificação iônica associado com recobrimento, por interação eletrostática, com solução de WPC em diferentes concentrações e em seguida foi avaliada a adsorção proteica sobre as micropartículas em sistemas diluídos e concentrados. Um aumento da adsorção proteica pôde ser observado com o aumento da concentração de proteína em solução e diminuição da concentração de manteiga adicionada. Utilizando-se solução de WPC na concentração de 4% (m/m) foram obtidas maiores quantidades de proteínas adsorvidas sobre as micropartículas com taxas de adsorção variando de 44,35% para micropartículas com baixo nível de manteiga fundida, de 37,58% para micropartículas contendo nível intermediário de manteiga e de 30,61% para micropartículas contendo alto nível de manteiga. Micropartículas com elevado teor de umidade, conteúdo proteico acima de 30% e tamanho variando entre 130,0 µm a 161,5 µm puderam ser produzidas pela técnica de gelificação iônica com posterior recobrimento, por interação eletrostática, com solução 4% (m/m) de WPC. A partir do estudo preliminar, as micropartículas com os diferentes tratamentos foram estudadas para imobilização celular do micro-organismo Erwinia sp. D12 com uma contagem inicial de 108-107 UFC/mL e atividade enzimática de 10,7 U/mL. A incorporação de diferentes concentrações de manteiga fundida modificou a atividade enzimática do micro-organismo encapsulado sendo que uma maior adição de manteiga fundida (2%, m/m) acarretou em uma maior atividade enzimática (17,68 U/mL). O tamanho médio das micropartículas sem recobrimento variou entre 91,54 ¿ 106,52 µm sendo observado um aumento no tamanho, 118,34 ¿ 143,12 µm, após o recobrimento. Células íntegras na forma livre e imobilizadas com os diferentes tratamentos foram avaliadas com relação à conversão de sacarose em isomaltulose em processo descontínuo durante quatro dias. Maiores taxas de conversão foram obtidas no primeiro dia de análise empregando células livres ou imobilizadas com os diferentes tratamentos. A taxa de conversão diminuiu no decorrer dos dias, em todos os tratamentos, sendo observada uma queda mais acentuada na conversão utilizando células íntegras livres. As maiores médias aritméticas de conversão de sacarose em isomaltulose foram alcançadas por micropartículas de gelificação iônica sem adição de manteiga fundida e micropartículas de gelificação iônica adicionadas de 2% (m/m) de manteiga fundida, 33,77% e 35,12%, respectivamente, sem diferenças estatísticas entre as mesmas. O recobrimento adicional das micropartículas acarretou em uma diminuição na taxa de conversão devido provavelmente à diminuição no tamanho dos poros ou recobrimento total dos mesmos. Outra hipótese levantada está no pH utilizado para produção das micropartículas (pH 3,75) que encontra-se fora da faixa ótima de crescimento do micro-organismo. Os resultados indicam o potencial de emprego de células íntegras de Erwinia sp. D12 imobilizadas em hidrogel de alginato, sem recobrimento adicional das micropartículas, para conversão de sacarose em isomaltulose em processos descontínuos / Abstract: This study aimed to produce microparticles by combined techniques of electrostatic interaction and ionic gelation using sodium alginate, whey protein concentrated (WPC) and butter for immobilization of micro-organism Erwinia sp. D12 and evaluating the conversion of sucrose to isomaltulose by immobilized whole cells and free cells in a batch process during four days. In a preliminary study, biopolymers sodium alginate and WPC were characterized for production of microparticles having good properties. The analysis of electric charge available sodium alginate in solution or emulsion with different concentrations of butter and WPC solution indicated that in a pH range from 3.0 to 4.5 may occur electrostatic interactions between the biopolymers. Mixtures formed in different ratios emulsion of sodium alginate with different concentrations of butter and WPC solution were characterized with respect to size and load presented average coacervates insoluble formed. Larger coacervates sizes, ranging from 29.12 µm to 33.72 µm, were obtained in ratio 1:6 (emulsion alginate and butter: WPC) being that addition of different concentrations of butter not statistically change the size of coacervates formed. Alginate microparticles containing butter were prepared by ionic gelation associated with coating by electrostatic interaction with WPC at different concentrations and then the protein adsorption on the microparticles in dilute and concentrated was assessed. An increase in protein adsorption could be observed even in concentrated systems, with increasing concentration of protein in solution and decreased concentration of butter added. Using WPC solution at a concentration of 4% (w/w) were obtained larger quantities of proteins adsorbed on the microparticles with varying rates of adsorption of 44.35% for microparticles containing low level of butter , of 37.58% for microparticles containing intermediate level of butter and 30.61% for microparticles containing high level of butter. Microparticles content high moisture, protein content above 30% and size ranging from 130.0 µm to 161.5 µm could be produced by ionic gelation technique with subsequent coating by electrostatic interaction with solution 4% (w/w) WPC. From the preliminary study, the microparticles with different treatments were studied for cell immobilization of micro-organism Erwinia sp. D12 with an initial count of 108-107 CFU/mL and enzyme activity of 10.7 U/mL. The incorporation of different concentrations of butter modify the enzymatic activity of the encapsulated micro-organism being greater than the addition of butter (2%, w/w) resulted in an increased enzymatic activity (17.68 U/mL). The average size of uncoated microparticles ranged from 91.54 µm to 106.52 µm an increase in size was observed, 118.34 - 143.12 µm, after coating. The addition of different concentrations of butter not influence statistically the average size of microparticles. Whole cells in free form and immobilized with different treatments were evaluated for conversion of sucrose to isomaltulose in a batch process during four days. Higher rates of conversion of sucrose to isomaltulose were obtained on the first day of analysis using free cells or immobilized with the different treatments. The conversion rate decreased during the days, in all treatments, with a more pronounced decrease observed in conversion using free cells. The greatest average conversion of sucrose into isomaltulose were reached by microparticle ionic gelling without addition of butter and microparticles ionic gelling added 2% (w/w) of butter, 33.77% and 35.12% respectively, no statistical differences between them. The additional coating of microparticles by electrostatic interaction resulted in a decrease in conversion rate probably due to a reduction in pore size or full coating thereof. Another hypothesis is the pH used for production of microparticles, at pH 3.75, which is outside the optimum range for growth of the micro-organism. The results indicate the potential use of whole cells of Erwinia sp. D12 immobilized in alginate hydrogel without additional coating of microparticles for conversion of sucrose to isomaltulose in batch processes / Mestrado / Nutrição Experimental e Aplicada à Tecnologia de Alimentos / Mestra em Alimentos e Nutrição

Orientadores: Maria Regina Wolf Maciel, Carlos Eduardo Vaz Rossell / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Química / Made available in DSpace on 2018-08-26T10:10:41Z (GMT). No. of bitstreams: 1 Alves_Moises_D.pdf: 2341942 bytes, checksum: 15b4471c1a44c4b51fad94e70bd4ebc4 (MD5) Previous issue date: 2014 / Resumo: Neste trabalho, realizou-se o estudo da produção de etanol anidro por meio do processo de destilação extrativa, tomando por base uma usina sucroalcooleira que processa 15000 toneladas de cana por dia. Utilizou-se como ferramenta computacional o software Aspen Plus®, o qual possui recursos computacionais que permitem compatibilizar as simulações com processos reais. Nas simulações deste trabalho foram utilizados os solventes monoetilenoglicol e o líquido iônico cloreto de 1- metilimidazólio, assim como os modelos de estágios de equilíbrio e de não equilíbrio. Os resultados da análise de sensibilidade para utilização do solvente monoetilenoglicol na coluna de destilação extrativa, em alta temperatura, apresentou economia de até 0,04 kg de vapor/litro de etanol anidro e de 0,37 kg de vapor/litro de etanol anidro ao alimentar o etanol hidratado produzido em colunas de destilação diretamente em colunas extrativas, sem estocagem e sem necessidade de aquecimento deste etanol hidratado. As colunas de destilação extrativa e de recuperação, do processo de destilação extrativa, apresentaram eficiências globais de 71,68% e 64,25%, respectivamente, segundo a correlação de O¿Connell. Foi realizado o estudo do consumo energético em colunas que operavam com pressões de 1,0 atm, 0,7 atm, 0,5 atm e 0,3 atm produzindo etanol anidro com fração mássica de 0,997. Os resultados mostraram que as colunas que operam em pressões menores que a atmosférica apresentaram redução na razão de refluxo em mais de 4,5 vezes e redução no consumo de vapor de 39,62%. Ao comparar o consumo de vapor em sistema que utiliza coluna extrativa e de recuperação, com pressão de 1,0 atm ou com pressão de 0,3 atm, foi constatado que a redução na pressão de operação das colunas gerou um excedente de vapor de 1,845 MWh. Os resultados também mostraram que o consumo de energia de uma bomba de vácuo necessária para gerar baixa pressão em colunas extrativa e de recuperação é menor que 1,5% para os diferentes níveis de pressão. O estudo com o solvente líquido iônico cloreto de 1-metilimidazólio apresentou redução mássica de utilização do solvente igual a 41,16%, demonstrado por meio da razão S/F igual a 0,353 para o líquido iônico e de 0,60 para o monoetilenoglicol. O estudo com o cloreto de 1-metilimidazólio mostrou a redução do consumo de vapor no processo em 27,3%, em relação ao solvente monoetilenoglicol, somadas as colunas extrativa e de recuperação. Foi realizada a simulação do modelo de não equilíbrio, levando em consideração parâmetros específicos das bandejas. Os resultados mostraram que a energia consumida no modelo de estágios de não equilíbrio apresentou aumento de 92,5% em relação ao modelo estágio de equilíbrio / Abstract: In this work, the study of the production of anhydrous ethanol by means of the extractive distillation process was carried out, based on a sugar and alcohol plant that processes 15,000 tons of sugar cane per day. The software Aspen Plus® was used as a computational tool, which has computational resources that adapt the simulations with real processes. In this study¿s simulations monoethylene glycol and the ionic liquid 1- methylimidazolium were used as solvents, as well as models of equilibrium and nonequilibrium. The results of the sensitivity analyses when using monoethylene glycol as the solvent, in the extractive distillation column, at high temperature, showed savings of up to 0.04 kg of steam / liter of anhydrous ethanol, and 0.37 kg of steam / liter of anhydrous ethanol when feeding the hydrous ethanol produced in the distillation columns directly into the extractive columns without the need for storage and of heating that hydrous ethanol. The extractive distillation and recovery columns, of the extractive distillation process, have shown overall efficiencies of 71.68% and 64.25%, respectively, according to the O'Connell correlation. We have conducted the study of energy consumption in columns operating at pressures of 1.0 atm, 0.7 atm, 0.5 atm and 0.3 atm producing anhydrous ethanol with a mass fraction of 0.997. The results have shown that the columns operating at pressures lesser than the atmospheric one decreased the reflux ratio more than 4.5 times, and the steam consumption by 39.62%. When comparing the steam consumption in the system using an extraction and recovery column, with a pressure of 1.0 atm or a pressure of 0.3 atm, it was found that the reduction in the column operating pressure generated 1.845 MWh of steam surplus. The results have also shown that the power consumption of a vacuum pump needed to generate a low-pressure in the extraction and recovery columns is lesser than 1.5% for different levels of pressure. The study of the ionic liquid solvent 1-methylimidazolium chloride had a mass reduction of solvent usage equal to 41.16%, shown by the S / F ratio equal to 0.353 for the ionic liquid and 0.60 for MEG. The study with 1-methylimidazolium chloride revealed that the steam consumption reduction in the process is 27.3%, compared to the monoethylene glycol solvent, combining the extraction and recovery columns. The simulation model of the non-equilibrium has been carried out taking into account specific parameters of the trays. The results showed that the power consumption in the nonequilibrium stage model displayed an increase of 92.5% compared to the equilibrium stage model. / Doutorado / Engenharia de Processos / Doutor em Engenharia Química