Tesi sul tema "Physicochemical model"

Increased and combined knowledge of food processing, molecular biology, health and nutrition has triggered production of many different types of functional foods and pharmaceutics recently. The efficacy and safety of such products are being assessed prior to marketing by in vivo and/or in vitro studies. Traditional in vivo studies require excessive time, cost and labour, as well as ethical approvals with subject to humans or animals in some instances. Therefore excessive number of runs may be avoided if reliable in vitro system is available. During the course of this study, an in vitro physicochemical upper gastrointestinal tract system (IPUGS), the first of its kind in literature, has been developed to simulate the relevant conditions of the gastrointestinal tract (GIT) as closely as possible to the human physiology with multi-disciplinary approach, combining biology, physiology, gastroenterology, process technology, chemical engineering and automation. The IPUGS is aimed at having a high predictive capability towards the real digestion processes occurring in the human upper GIT which allows for examining of the bioavailability of nutrients and drugs, drug-nutrient interactions, viability of probiotics and case studies of gastrointestinal disorders. Digestion of rice and baby foods have been studied with the IPUGS by UV-spectrophotometer, HPLC, light microscope and pH meter under the conditions of normal state and common gastric disorders, such as gastroparesis, dumping syndrome, Zollinger-Ellison syndrome and hypochlorhydria. By comparing the data from many physiological and clinical sources in the literature, it would seem that the IPUGS was able to generate more reliable data compared to the existing in vitro digestion (mechanical) models in the literature. In future, computer-controlled and computer-recorded data by possibly designing a new software or equations would be desirable to implicate a better understanding of the digestive processes.

Ce travail s'inscrit dans un but de caractérisation et de quantification des propriétés physiques et chimiques de la matière vivante et de ses constituants. Les objectifs recherches sont de deux ordres: une meilleure compréhension des mécanismes d'interaction des ondes ultrasonores avec les milieux biologiques, qui sont parmi les plus complexes; le développement de méthodes ultrasonores permettant de mesurer des propriétés physico-chimiques des milieux biologiques. Le choix des glucides comme modèle moléculaire pour étudier l'interaction ultrasons-milieux biologiques réside dans le fait que ces molécules sont d'une importance fondamentale dans la composition des structures vivantes au même titre que les protéines et les lipides. De plus, elles forment un groupe comprenant des isomères et des polymères plus ou moins complexes. L’influence de la structure moléculaire est accessible en examinant l'atténuation, la vitesse et les effets non linéaires. L’analyse de ces divers paramètres est menée dans une gamme de fréquence ultrasonore située entre 50 et 200 mhz, par une technique acousto-optique.

Doctor of Philosophy<br>Department of Grain Science and Industry<br>Praveen V. Vadlani<br>The purpose of this study was to investigate micro- and macro-scale aspects of solid state fermentation (SSF) for production of cellulolytic enzymes using fungal cultures. Included in the objectives were investigation of effect of physicochemical characteristics of substrate on enzymes production at micro-scale, and design, fabrication and analysis of solid-state bioreactor at macro-scale. In the initial studies response surface optimization of SSF of soybeans hulls using mixed culture of Trichoderma reesei and Aspergillus oryzae was carried out to standardize the process. Optimum temperature, moisture and pH of 30ºC, 70% and 5 were determined following optimization. Using optimized parameters laboratory scale-up in static tray fermenter was performed that resulted in production of complete and balanced cellulolytic enzyme system. The balanced enzyme system had required 1:1 ratio of filter paper and beta-glucosidase units. This complete and balanced enzyme system was shown to be effective in the hydrolysis of wheat straw to sugars. Mild pretreatments– steam, acid and alkali were performed to vary physicochemical characteristics of soybean hulls – bed porosity, crystallinity and volumetric specific surface. Mild nature of pretreatments minimized the compositional changes of substrate. It was explicitly shown that more porous and crystalline steam pretreated soybean hulls significantly improved cellulolytic enzymes production in T. reesei culture, with no effect on xylanase. In A. oryzae and mixed culture this improvement, though, was not seen. Further studies using standard crystalline substrates and substrates with varying bed porosity confirmed that effect of physicochemical characteristics was selective with respect to fungal species and cellulolytic activity. A novel deep bed bioreactor was designed and fabricated to address scale-up issues. Bioreactor’s unique design of outer wire mesh frame with internal air distribution and a near saturation environment within cabinet resulted in enhanced heat transfer with minimum moisture loss. Enzyme production was faster and leveled within 48 h of operation compared to 96 h required in static tray. A two phase heat and mass transfer model was written that accurately predicted the experimental temperature profile. Simulations also showed that bioreactor operation was more sensitive to changes in cabinet temperature and mass flow rate of distributor air than air temperature.

In the present thesis, the correlation between the physicochemical and antigenic properties of different recombinant Streptococcus gordonii vaccine vectors was studied. Evaluation of vaccine efficacy, antigenicity and immugenicity is a crucial step in developing vaccines, thus investigating a simple method to analyze vaccine efficacy besides other methods could be a major part of developing bacterial vaccine vectors. To approach this, isoelectric point measurements and zeta-potential titration as well as antigenicity and immunogenicity of S.gordonii vaccine vectors (with fbpA, RPS, gtfg genes mutations expressing H1 antigen) were used. These data showed that strains with more positive surface charge had higher heterologous antigen recognition and lower antibody responses in the serum of immunized mice. This correlation between surface charge and antigenicity and immunogenicity revealed the importance of using simple methods such as zeta potential titration and isoelectric point measurements to predict engineered vaccine vectors antigenicity and possible efficacy. In the second part of the thesis the immune recall in the days following Streptococcus pneumoniae lung infection by transcriptomic analysis was studied. S. pneumoniae is the most common bacterial cause of community-acquired pneumonia. Host-pathogen interaction is poorly understood, and factors that drive a more severe phenotype are unknown. One way to study host response to pathogen is using the stimulation of immune system cells with live or killed bacteria. We combined transcriptomic and cytokine level analysis on stimulated mouse splenocytes revealing the presence of a recall immune response involving both innate and adaptive immunity, stronger from the fourth day after infection. This model could analyze immune responses involved in pneumococcal infection as well as vaccine and experimental therapies efficacy in future studies. Finally, the development of a S. pneumoniae mouse model of pneumonia by intra-tracheal infection was set up. The nasopharynx of humans is the only natural reservoir for the pneumococci. To mimic human pneumonia, mice models are widely used. Bacteria can be administered to mice intranasally, intratracheally or as aerosols. Pneumococcal pneumonia was induced in mice by intra-tracheal inoculation with different doses of S. pneumoniae TIGR4. Data showed high colonization of bacteria in lung, liver and spleen starting 24 hours post-infection. Pneumonia mortalities were observed in all mice infected by 108 within 24 hours of infection. Further analysis should be done to investigate the host-pathogen interaction as well as vaccine and experimental therapies efficacy by using this model.

A pharmaceutical dosage form is an entity that is administered to patients so that they receive an effective dose of an active pharmaceutical ingredient (API). The proper design and formulation of a transdermal dosage form require a thorough understanding of the physiological factors affecting percutaneous penetration and physicochemical characteristics of the API, as well as that of the pharmaceutical exipients that are used during formulation. The API and pharmaceutical excipients must be compatible with one another to produce a formulation that is stable, efficacious, attractive, easy to administer, and safe (Mahato, 2007:11). Amongst others, the physicochemical properties indicate the suitability of the type of dosage form, as well as any potential problems associated with instability, poor permeation and the target site to be reached (Wells & Aulton, 2002:337). Therefore, when developing new or improved dosage forms, it is of utmost importance to evaluate the factors influencing design and formulation to provide the best possible dosage form and formulation for the API in question. Delivery of an API through the skin has long been a promising concept due to its large surface area, ease of access, vast exposure to the circulatory and lymphatic networks, and non-invasive nature of the therapy. This is true whether a local or systemic pharmacological effect is desired (Aukunuru et al., 2007:856). However, most APIs are administered orally as this route is considered to be the simplest, most convenient and safest route of API administration. Since ibuprofen is highly metabolised in the liver and gastrointestinal tract, oral administration thereof results in decreased bioavailability. Furthermore, it also causes gastric mucosal damage, bleeding and ulceration. Another obstacle associated with oral API delivery is that some APIs require continuous delivery which is difficult to achieve (Bouwstra et al., 2003:3). Therefore, there is significant interest to develop topical dosage forms for ibuprofen to avoid side effects associated with oral delivery and to provide relatively consistent API levels at the application site for prolonged periods (Rhee et al., 2003:14). The aim of this study was to determine the influence of selected formulation factors on the transdermal delivery of ibuprofen. In order to achieve this aim, the physicochemical properties of ibuprofen had to be evaluated. The aqueous solubility, pH-solubility profile, octanol-water partition coefficient (log P-value) and octanol-buffer distribution coefficient (log D-values, pH 5 and 7.4) of ibuprofen were determined. According to Naik et al., (2000:319) the ideal aqueous solubility of APIs for transdermal delivery should be more than 1 mg.ml-1. However, results showed that ibuprofen depicted an aqueous solubility of 0.096 mg.ml-1 ± 25.483, which indicated poor water solubility and would therefore be rendered less favourable for transdermal delivery if only considering the aqueous solubility. The pH-solubility profile depicted that ibuprofen was less soluble at low pH-values and more soluble at higher pH-values. Previous research indicated that the ideal log Pvalues for transdermal API permeation of non steroid anti-inflammatory drugs (NSAIDs) are between 2 and 3 (Swart et al., 2005:72). Results obtained during this study indicated a log P-value of 4.238 for ibuprofen. This value was not included in the ideal range, which is an indication that the lipophilic/hydrophilic properties are not ideal, and this might therefore; contribute to poor ibuprofen penetration through the skin. Furthermore, the obtained log D-values at pH 5 and 7.4 were 3.105 and 0.386, respectively. Therefore, it would be expected that ibuprofen incorporated into a formulation prepared at a pH of 5 would more readily permeate the skin compared to ibuprofen incorporated into a formulation prepared at a pH of 7.4. A gel, an emulgel and a Pheroid™ emulgel were formulated at pH 5 and 7.4, in order to examine which dosage form formulated at which pH would deliver enhanced transdermal delivery. Obtained diffusion results of the different semi-solid formulations were furthermore compared to a South African marketed commercial product (Nurofen® gel) in order to establish if a comparable formulation could be obtained. An artificial membrane was used to conduct the membrane permeation studies over a period of 6 h, in order to determine whether ibuprofen was in fact released from the formulations through the membrane. Skin permeation studies were conducted using Franz diffusion cells over a period of 12 h where samples were withdrawn at specified time intervals. All the formulations exhibited an increase in the average cumulative amount of ibuprofen released from the formulations and that permeated the membrane when compared to Nurofen® gel. This increase was statistically significant (p<0.05) for the gel, emulgel and Pheroid™ emulgel at pH 7.4. The gel at pH 7.4 exhibited the highest cumulative amount of ibuprofen that permeated the membrane. Preparations formulated at a pH of 5, did not differ significantly from Nurofen® when the average cumulative amount of ibuprofen that permeated the membrane were compared. The following rank order for the average cumulative amount released from the formulations could be established: Gel (pH 7.4) >>>> Pheroid™ emulgel (pH 7.4) > Emulgel (pH 7.4) >>> Gel (pH 5)> Pheroid™ emulgel (pH 5) ≈ Emulgel (pH 5) > Nurofen® gel. On the other hand, all the formulations exhibited an increase in the average cumulative amount of ibuprofen that permeated the skin when compared to Nurofen® gel. This increase was statistically significant (p < 0.05) for the gel, emulgel and Pheroid™ emulgel at pH 5, as well as the emulgel and Pheroid™ emulgel at pH 7.4. The emulgel at pH 5 exhibited the highest cumulative amount of ibuprofen that permeated the skin. The following rank order for the average cumulative amount released from the formulations and that permeated the skin could be established: Emulgel (pH 5) >> Pheroid™ emulgel (pH 5) > Gel (pH 5) > Emulgel (pH 7.4)> Pheroid™ emulgel (pH 7.4) ≈ Emulgel (pH 7.4) >> Nurofen® gel > Gel (pH 7.4). From this rank order it was clear that a trend was followed where the pH of formulation also played a role in ibuprofen permeation. All the formulations exhibited a higher release rate and flux when compared to Nurofen® gel. This was statistically significant for the emulgel, gel and Pheroid™ emulgel at pH 7.4. The gel at pH 7.4 exhibited the highest release rate and flux. This was observed for the membrane and skin permeation studies. All the formulations (including Nurofen® gel) presented a correlation coefficient (r2) of 0.972 – 0.995 for membrane permeation studies, and 0.950 – 0.978 for skin permeation studies; indicating that the release of ibuprofen from each of the formulations could be described by the Higuchi model. Furthermore, all the formulations exhibited a prolonged lag time compared to Nurofen® gel which indicated that the ibuprofen was retained for a longer time by the base. This was statistically significant (p < 0.05) for the emulgel at pH 7.4, the gel and Pheroid™ emulgel at pH 5. The gel at pH 7.4 exhibited a lag time closest to that of Nurofen® gel and this difference could not be classified as statistically significant (p > 0.286). This was observed for the membrane and skin permeation studies. Nurofen® gel exhibited the highest ibuprofen concentration in the stratum corneum as well as in the epidermis followed by the gel at pH 7.4. However, results obtained for all the formulations indicated that topical as well as transdermal delivery of ibuprofen was achieved. The pH of a formulation plays an important role with respect to API permeation. Ibuprofen is reported to have a pKa value 4.4 (Dollery, 1999:I1); and by application of the Henderson-Hasselbach equation, at pH 5, 20.08% of ibuprofen will be present in its unionised form and at pH 7.4, 0.1% ibuprofen will exist in its unionised form. Since the unionised form of APIs is more lipid soluble than the ionised form, unionised forms of APIs permeate more readily across the lipid membranes (Surber & Smith, 2000:27). Therefore, it would be expected that ibuprofen formulated at pH 5 would be more permeable than formulations at pH 7.4. However, this did not correspond to the results (membrane studies) obtained in this study. It may be attributed to the solubility of ibuprofen in the different formulations. According to the pH-solubility profile of ibuprofen obtained in this study, it was more soluble at pH 7.4 than at pH 5. This was due to the fact that ibuprofen is a weak acidic compound, and for every 3 units away from the pKa-value, the solubility changes 10-fold (Mahato, 2007:14). However, with regard to the skin permeation studies, enhanced permeation was obtained with the formulations prepared at pH 5. This was in accordance with Corrigan et al., (2003:148) who stated that NSAIDs are less soluble and more permeable at low pH values, and more soluble and less permeable at high pH values. This was most probably due to the fact that unionised species, although possessing a lower aqueous solubility than the ionised species, resulted in enhanced skin permeation due to being more lipid-soluble. Finally, stability tests on the different semi-solid formulations for a period of three months at different temperature and humidity conditions were conducted to determine product stability. The formulations were stored at 25 °C/60% RH (relative humidity), 30 °C/60% RH and 40 °C/75% RH. Stability tests included: mass variation, pH, zeta potential, droplet size, visual appearance, assay, and viscosity. No significant change was observed for mass variation, pH, zeta potential and droplet size over the three months for any of the different formulations stored at the different storage conditions. In addition, no significant change in colour was observed for the gel and emulgel formulations at pH 5 and 7.4 over the three months at all the storage conditions. However, it was observed that the formulations containing Pheroid™ showed a drastic change in colour at all the storage conditions. This might have been due to oxidation of certain components present in the Pheroid™ system. Consequently, further investigation is necessary to find the cause of the discolouration and a method to prevent it. The gel formulated at pH 5 depicted the formation of crystals. This might have been due to the fact that the solubility of ibuprofen was exceeded, leading to it precipitating from the formulation. A possible contributing factor to the varying assay values obtained during the study might have been due to non-homogenous sample withdrawal. On the other hand, no significant change was observed for the emulgel and Pheroid™ emulgel formulated at pH 5 and 7.4. The emulgel and Pheroid™ emulgel formulated at pH 5 depicted relative instability (according to the International Conference on Harmonisation of Technical Requirements For Registration of Pharmaceuticals for Human Use, ICH) only at 40 °C/75% RH with a change in ibuprofen content of more than 5% (6.78 and 6.46%, respectively). The gel, emulgel and Pheroid™ emulgel at pH 7.4 exhibited the least variation in ibuprofen concentration at all of the storage conditions. This might indicate that the pH at which a semi-solid formulation is produced will have a direct influence on the stability of the product. No significant changes in viscosity (%RSD < 5) was observed for the gel and emulgel formulated at pH 7.4 and stored at 25 °C/60% RH. The remaining formulations at all of the specified storage conditions exhibited a significant change in viscosity (%RSD > 5) with a decrease in viscosity being more pronounced at the higher temperature and humidity storage conditions. A possible contributing factor to the change in viscosity over three months at the specified storage conditions might have been due to the use of Pluronic® F-127 (viscosity enhancer). This viscosity enhancer possesses a melting point of approximately 56 °C (BAST Corporation. s.a). The problem with this might have been the temperature (70 °C) at which the formulations were prepared. The higher preparation temperature might have caused the Pluronic® F-127 to degrade, thereby losing its ability to function appropriately. A balance must be maintained between optimum solubility and maximum stability (Pefile & Smith, 1997:148). Despite the lower skin permeation of the gel formulated at pH 7.4, this formulation performed the best, as it was considered stable (least variation during the 3 month stability test) and the obtained tape stripping results showed that this formulation depicted the highest ibuprofen concentrations in the stratum corneum and epidermis. Thus, topical as well as transdermal delivery were obtained.<br>Thesis (MSc (Pharmaceutics))--North-West University, Potchefstroom Campus, 2013.

This research offers a specific strategy for the management of a global health problem associated with smoking addiction. Novel controlled release nicotine-loaded chitosan nanoparticles have been developed as a potential therapy. In vitro and in vivo evaluation of these nanoparticles indicate that they are suitable as dry powder inhaler formulations for pulmonary delivery. Results from a mouse model should translate to humans to provide a safe and effective approach to treat smoking dependence.

L’acide phosphorique et les sels de phosphates sont utilisés dans le monde entier dans la production de fertilisants, d’additifs alimentaires, mais également comme agent de gravure dans l’industrie électronique et pour diverses applications de l’industrie pharmaceutique. Selon l’application visée, un degré de pureté plus ou moins élevé peut être requis vis-à-vis d’un certain nombre d’impuretés dissoutes conjointement avec l’acide phosphorique lors de la lixiviation des roches phosphatées par l’acide sulfurique (voie humide, constituant aujourd’hui le procédé quasi exclusif de production de H₃PO₄). Il est donc nécessaire d’éliminer ces impuretés (espèces métalliques, sulfate, chlorure, etc.). A l’échelle industrielle, la purification de l’acide phosphorique est réalisée principalement par extraction liquide-liquide. L’opération consiste à extraire le plus sélectivement possible les molécules d’acide phosphorique initialement contenues dans le jus de lixiviation dans une phase organique appropriée. Cependant une telle extraction n’est pas totalement sélective et certaines impuretés sont co-extraites. Par la suite, des étapes complémentaires doivent être ajoutées, ce qui complexifie l’ensemble du procédé et le rend plus coûteux. La connaissance fine de la spéciation des impuretés gênantes présentes dans l’acide phosphorique concentré (8-14,5 mol.L⁻¹) est donc une donnée importante pour comprendre les raisons physicochimiques de leur co-extraction et, in fine, pour modifier le procédé afin de s’affranchir de cette dernière. Par exemple, les solvants actuellement utilisés ne sont pas suffisamment sélectifs vis-à-vis du titane(IV). Ce travail de thèse est donc centré sur l’étude de la spéciation du titane(IV) sur une large gamme de concentration en acide phosphorique dans le but de pouvoir identifier ultérieurement les équilibres responsables de sa co-extraction avec l’acide phosphorique. Cette thèse repose sur une approche originale combinant des méthodes d’analyses spectroscopiques et des calculs de modélisation moléculaire. Des solutions synthétiques d’acide phosphorique contenant du titane(IV) ont été caractérisées par spectroscopies UV-Visible, de Résonnance Magnétique Nucléaire (³¹P RMN) et par spectroscopie d’absorption des rayons X. Ainsi, la structure des complexes formés dans l’acide phosphorique a pu être étudiée par la comparaison des spectres UV-Visible et de calculs implémentant la théorie de la fonctionnelle de la densité dépendante du temps (TD-DFT). La nucléarité de ces complexes a également pu être évaluée en comparant les valeurs des coefficients d’autodiffusion déterminées expérimentalement par RMN ³¹P aux valeurs calculées par dynamique moléculaire pour différentes espèces de titane(IV) potentiellement présentes en solution. La coordination du titane(IV) a pu aussi être étudiée en analysant les spectres EXAFS (Extended X-Ray Absorption Fine Structure) à l’aide de simulations de dynamique moléculaire ab-initio prenant explicitement en compte la solvatation des complexes. Enfin, l’exploitation des spectres UV-Visible par un outil chimiométrique fondé sur une analyse en composants principaux a permis d’extraire des informations quantitatives sur la répartition des complexes de titane(IV) présents dans l’acide phosphorique concentré. A partir de l’ensemble de ces résultats, il a été possible de proposer pour la première fois un diagramme de spéciation du titane(IV) dans l’acide phosphorique pour une gamme de concentrations comprises entre 6 et 13 mol.L⁻¹, soulignant la présence évolutive de trois complexes de titane(IV) mono- et poly-nucléaires dont l’espèce prédominante est [Ti(OH)(H₃PO₄)₂(H₂PO₄)]²⁺<br>The conventional wet-process of production of phosphoric acid consists of a leaching of phosphate ores with sulfuric acid during which several impurities (metallic, sulfate, chloride are dissolved concomitantly. Phosphoric acid and phosphate salts are used in various applications such as fertilizers, food additives, electronic etching agent or pharmaceutical excipients and must therefore meet appropriate of specifications regarding their purity. As a consequence, the concentration of these impurities must be reduced by performing purification steps. At the industrial scale, the purification of phosphoric acid is performed mainly by liquid-liquid extraction. The operation consists in extracting as selectively as possible the phosphoric acid molecules initially contained in the leaching juice into an appropriate organic phase. However this process is not selective enough and some of the impurities are co-extracted. This leads to the necessity of performing additional purification steps to meet the requested specifications, which increases both the complexity of the global treatment and its cost. The knowledge of the speciation of impurities in concentrated phosphoric acid is essential to understand the physicochemical reasons for their co-extraction and, in fine, to design more selective extraction solvents. For example, the solvents presently used for the purification of H₃PO₄ are not selective enough against titanium(IV). Thus, this PhD thesis work aims at characterizing the speciation of this metal in a large range of phosphoric acid concentration, in order to identify subsequently the equilibria responsible for its co-extraction with H₃PO₄. This thesis is based on an original approach combining the use of spectroscopic and molecular modeling techniques. Synthetic solutions containing both titanium(IV) and phosphoric acid have been characterized using different spectroscopic techniques including UV-Visible, Nuclear Magnetic Resonance (³¹P NMR) and X-Ray Absorption (XAS) spectroscopies. Thus, the structure of the complexes formed in phosphoric acid has been studied by comparing UV-Visible spectra and calculations implementing time-dependent density functional theory (TD-DFT). The nuclearity of these complexes has also been estimated by comparing the values of the self-diffusion coefficients determined experimentally by ³¹P NMR with the values calculated by molecular dynamics for different species of titanium (IV) potentially present in solution. The coordination of titanium (IV) was also studied by analyzing the EXAFS (Extended X-Ray Absorption Fine Structure) spectra using ab-initio molecular dynamics simulations explicitly taking into account the solvation of the complexes.Finally, UV-Visible spectral data have been analyzed by a chemometric approach, based on a principal component analysis (PCA), allowing us to extract quantitative information about the distribution of the complex species identified in concentrated phosphoric acid. From all these results, it was possible to propose for the first time a diagram of speciation of titanium (IV) in phosphoric acid for a range of concentrations between 6 and 13 mol.L⁻¹, underlining the evolutionary presence of three mono- and poly-nuclear titanium (IV) complexes, the predominant species of which is [Ti(OH)(H₃PO₄)₂(H₂PO₄)]²⁺

Gelification des pectines amidees en presence de calcium. Etude de modeles simplifies (acides pectiques amides). Les pko dependent du parametre de densite de charge et non pas du taux d'amidation/repartition non aleatoire des fonctions amides. La sensibilite au calcium des pectines amidees resulte de l'alternance de blocs de fonctions acides libres et methyles reparties aleatoirement et de fonctions amides, generateurs de liaisons h intermoleculaires, provenant de leur synthese concentre en ammoniac et en pectine

Doutoramento em Química<br>During the last years, the use of high hydrostatic pressure (HHP) as a non-thermal technology for preservation or aging of wine has increased substantially in the academic community. However, HHP treated wine has been only analysed after the pressure treatment, with no knowledge available on the effects of HHP during subsequent storage. The results presented in this thesis showed that HHP treatments influence the chemical and sensorial properties of wine during storage. The application of high hydrostatic pressure treatments in winemaking for wine preservation, as an alternative to sulphur dioxide, was evaluated studying the effect of HHP in the physicochemical and sensorial properties of red and white wines during bottle storage. High pressure treatments with 5 min of processing time and pressures of 425 and 500 MPa were shown to influence on both red and white wine physicochemical and sensorial characteristics. However, the effects were only perceptible after, at least, 6 months of storage. The alterations that occurred on the pressurized red wine characteristics, such as the more orange-red colour and the lower antioxidant activity (15-27% less), total phenolic content (9% less), and anthocyanins content (45–61% less), were due to an increase of condensation reactions of phenolic compounds. The increase of these condensation reactions lead to the formation of compounds with higher degree of polymerisation that became insoluble along storage, increasing consequently the amount of wine deposits in the pressurized wines. In terms of white wines, pressurized wines showed, after one year of storage, a more brownish colour and a lower antioxidant activity (15% less) and total content of phenolic compounds (10% less) when compared to the unpressurized wines. These results, together with the lower content of free amino acids (15-20% less) and higher content of furans (up to 70% more), present in the pressurized wines after nine months of storage, led to propose an effect of HHP treatments in the acceleration of Maillard reactions that occur during the wine storage period. Therefore, contrary to the pressurized sulphur dioxide-free red wine, the pressurized white wines were not considered suitable for commercialization as table wines due to the higher brownish colour and cooked fruit aroma, characteristics of an aged or thermally treated wine.Additionally, the impact of the pressure treatments on the volatile composition of sulphur dioxide-free red and white wines, during bottle storage, was evaluated. More than 160 volatile compounds, distributed by 12 chemical groups, were identified in both wines. At the end of storage, the pressurized wines presented a higher content of furans, aldehydes, ketones, and acetals when compared to the unpressurized wines. These results indicate that pressure influences the white and red wine long term volatile composition, being this particularly evident for longer storage periods. The changes on the volatile composition of the pressurized wines, indicated that the HHP treatments accelerate the Maillard reactions, and the oxidation of alcohols and fatty acids, leading to wines with a volatile composition network approaching the characteristic of faster aged and/or thermally treated wines. The acceleration of Maillard reactions and phenolic compounds condensation by HHP treatments was also studied in model wine solutions (hydro alcoholic solution at acidic pH). The results showed that the high pressure treatment accelerated the Maillard reaction and this effect was quantifiable, mainly, after 6 months of storage. Pressurized model solutions presented higher concentration of 2-furfural, phenylacetaldehyde and benzaldehyde, when compared to the controls. In terms of phenolic compounds condensation reactions, the pressurized model wine solutions showed no relevant differences, when compared to controls. Therefore, it seems that the pressure treatment had a higher impact in terms of kineticks of reactions and in less extent in terms of different compounds formed. Lastly, the application of HHP treatments in winemaking to improve the properties of young wines was evaluated. For this propose, the effect of HHP treatments in the phenolic composition of a red wine was studied and compared with the effect of different oenological practices. Wines pressurized at 500 MPa for 5 min, and 600 MPa for 20 min, at 20 ºC, showed, after 5 months of storage, a lower monomeric anthocyanins (8-14%), phenolic acids (8-11%) and flavonols (14-22%) content, when compared to the unpressurized ones. The wine pressurized at 500 MPa presented a flavanols content and a degree of polymerization very similar to the wines treated by traditional aging processes. In terms of sensorial properties, the pressure treatments increased the cooked fruit aroma and decreased the floral and fruit odours and, in the case of the 600 MPa treatment, increased the bitterness. Therefore, the HHP treatments seem to promote reactions that are similar to those observed in wines treated with wood aging processes. In conclusion, the results presented in this thesis showed that HHP treatments accelerated the Maillard reaction and the polymerization reactions between phenolic compounds present in the wine, influencing the chemical and sensorial properties of wine. HHP can be potentially used to preserve or accelerate the wine aging process, producing wines with pleasant and distinct characteristics.<br>Durante os últimos anos, o uso de alta pressão hidrostática (APH) como tecnologia não-térmica para a preservação ou envelhecimento de vinho tem aumentado substancialmente na comunidade académica. No entanto, os vinhos tratados por APH têm sido analisados após o tratamento de pressão, não havendo referências sobre as suas propriedades durante o armazenamento. Os resultados apresentados nesta tese mostram que a aplicação de tratamentos de APH altera as propriedades químicas e sensoriais de vinhos ao longo do armazenamento. Os tratamentos de alta pressão hidrostática foram aplicados na vinificação para a preservação de vinho, como alternativa ao dióxido de enxofre, sendo o seu efeito avaliado nas propriedades físico-químicas e sensoriais de vinhos tintos e brancos durante o armazenamento em garrafa. Os tratamentos de alta pressão com 5 min de processamento e pressões de 425 e 500 MPa mostraram influenciar as características físico-químicas e sensoriais de vinhos tintos e brancos. No entanto, o efeito foi apenas percetível após pelo menos 6 meses de armazenamento. As alterações que ocorreram nas características do vinho tinto pressurizado, tais como a cor mais laranja-vermelho, menor atividade antioxidante (menos 15 a 27%), menor conteúdo de compostos fenólicos totais (menos 9%) e menor teor de antocianinas (menos 45-61%), foram devidas a um aumento das reacções de condensação de compostos fenólicos. O aumento destas reações de condensação levou à formação de compostos com maior grau de polimerização que se tornaram insolúveis no vinho ao longo do armazenamento, aumentando consequentemente a quantidade de depósito nos vinhos pressurizados. Em relação ao vinho branco, os vinhos pressurizados mostraram, depois de um ano de armazenamento, uma cor mais acastanhada, menor atividade antioxidante (menos 15%) e menor teor de compostos fenólicos totais (menos 10%) comparando com os vinhos não pressurizados. Estes resultados, juntamente com o baixo teor de aminoácidos livres (menos 15 a 20%) e um maior teor de furanos (até 70% mais) para os vinhos pressurizados após nove meses de armazenamento, levam a propor que os tratamentos de APH aceleraram as reações de Maillard que ocorrem durante o período de armazenamento do vinho. No entanto, ao contrário dos vinhos tintos pressurizados, os vinho brancos pressurizados não foram considerados adequados para comercialização como vinhos de mesa, visto que apresentavam uma cor acastanhada e um elevado aroma a fruta cozida, características estas de vinhos envelhecidos ou tratados termicamente.Adicionalmente, foi avaliado o impacto dos tratamentos de APH sobre a composição volátil dos vinhos tintos e brancos sem dióxido de enxofre durante o armazenamento em garrafas. Mais de 160 compostos voláteis, distribuídos por 12 grupos químicos, foram identificados em ambos os vinhos. No final do armazenamento, os vinhos pressurizados apresentaram um teor mais elevado de furanos, aldeídos, cetonas e acetais quando comparados com os vinhos não pressurizados. Estes resultados indicam que os tratamentos de APH influenciam a composição volátil de vinhos brancos e tintos, , sendo mais evidente em longos períodos de armazenamento. As mudanças na composição volátil dos vinhos indicaram que os tratamentos de APH aceleraram as reações de Maillard e também a oxidação de álcoois e ácidos gordos, originando vinhos com uma composição volátil próxima de vinhos com envelhecimento acelerado ou tratados termicamente. A aceleração das reações de Maillard e de polimerização dos compostos fénolicos causada pelos tratamentos de APH foi também estudada em soluções modelo de vinho (solução hidroalcoólica com pH ácido). Os resultados mostraram que o tratamento de APH acelera a reação de Maillard , sendo este efeito quantificado, apenas, após 6 meses de armazenamento. As soluções modelo de vinho pressurizadas apresentaram concentrações mais elevadas de 2-furfuraldeído, fenilacetaldeído e benzaldeído, em comparação com os controlos. Em termos de polimerização dos compostos fénolicos, as soluções modelo pressurizadas não apresentaram diferenças relevantes, em comparação com os controlos. Por conseguinte, os tratamentos de APH aparentem ter mais impacto em termos de modificações nas cineticas de reação do que na formação de novos compostos. Por último, a aplicação de tratamentos de APH foi estudada para melhorar as propriedades de vinhos jovens. Para este propósito, o efeito de tratamentos de APH na composição fenólica de um vinho tinto foi estudado e comparado com o efeito de diferentes práticas enológicas. Vinhos pressurizados a 500 MPa durante 5 min e a 600 MPa durante 20 min, a 20 ºC, mostraram depois de 5 meses de armazenamento um menor teor de antocianinas monoméricas (8-14%), ácidos fenólicos (8-11%) e flavonóis (14 -22%), quando comparados com os vinhos não-pressurizados. O vinho pressurizado a 500 MPa apresentou um teor de flavonóis e um grau de polimerização de taninos muito semelhante aos vinhos tratados por processos de envelhecimento tradicionais. Em termos de propriedades sensoriais, os tratamentos de pressão aumentaram o aroma de fruta cozida e diminuiram os aromas florais e frutados, tendo no caso do tratamento de 600 MPa sido verificado também um aumento da amargura. Assim sendo, os tratamentos de APH parecem promover reações que são semelhantes às observadas em vinhos tratados com processos de envelhecimento em madeira. Em conclusão, os resultados apresentados nesta tese mostram que a aplicação de tratamentos de APH acelera as reações de Maillard e a polimerização dos compostos fenólicos presentes no vinho, ao longo do armazenamento, alterando assim as propriedades químicas e sensoriais dos vinhos. A APH pode ser potencialmente utilizada para preservar ou acelerar o processo de envelhecimento de vinho tinto produzindo vinhos com características agradáveis e distintas.

This thesis describes a toolbox of complementary techniques that together measure and mechanical properties of bone. Three-point bending is used to measure the mechanical properties of bone; micro computed tomography provides cortical geometry and parameters describing trabecular bone.  The material properties, elastic modulus and density, are measured directly using ultrasound and Archimedes’ principle, while composition and bone chemistry are investigated by ashing and Raman microscopy.  These methods are used to characterise bone from the naturally occurring Gunmetal mouse and the engineered neuronal nitric oxide synthase (nNOS) knockout mouse. Comparison was also made between femora and tibiae and cortical and trabecular bone from wild type mice. Gunmetal mice had inferior mechanical properties, but unaffected material and chemical properties.  Cortical area but not second moment of area was also reduced.  nNOS knockouts had superior bone mechanically, due to increased mineralisation and geometrical parameters.  Femora and tibiae had different mechanical and material properties that were not linked to the size or shape of the bones.  Cortical bone  had characteristics of older bone compared to trabecular material, possibly due to the lower turnover rate. These results show the necessity for measuring material properties directly, rather than inferring them from mechanical and geometrical properties.  The differences in femora and tibiae suggest testing only femur or tibia may result in the risk of missing important results.  Application of this toolbox of methods provides a comprehensive description of bone’s overall fitness for purpose and an understanding of the origin of any defect or enhancement in its properties.

Cette these se situe dans le cadre des modelisations chimiques de proteines fer-soufre, plus particulierement les proteines fer-soufre a haut potentiel (hipip). L'objectif est de synthetiser des centres 4fe-4s dans les deux etats d'oxydation physiologiques 4fe-4s#2+ et 4fe-4s#3+ et de les caracteriser d'un point de vue physico-chimique. L'hypothese a partir de laquelle le sujet a ete aborde consiste a supposer qu'il est necessaire d'etablir une zone steriquement encombree et hydrophobe autour de l'agregat afin de proteger le centre 4fe-4s#3+ d'une degradation oxydante. Pratiquement, un tel environnement encombre et hydrophobe a ete realise grace a l'utilisation de ligands thiolates riches en noyaux aromatiques et tres encombres. Pour chaque ligand utilise, le centre 4fe-4s#2+ a ete synthetise et caracterise par les spectroscopies uv-visible, rmn et mossbauer, et l'etude de sa susceptibilite magnetique a aussi ete effectuee. Dans un deuxieme temps, des etudes d'oxydation sont realisees sur ces nouveaux composes modeles 4fe-4s#2+ dans le but d'obtenir des centres 4fe-4s#3+. Des phenythiolates d'une part et des benzylhtiolates d'autre part, tous tres hydrophobes, ont ete utilises comme ligands. Dans chacun des cas, un nouveau modele de hipip a ete obtenu et completement caracterise dans les deux etats redox. Celui avec le ligand de type benzylthiolate est un meilleur modele chimique de ces proteines car il utilise des ligands plus biomimetiques. Par ailleurs, l'influence exercee par l'encombrement sterique a ete etudiee de facon systematique. En utilisant une serie de ligands encombres de facon croissante, il a pu etre montre tres clairement que la presence d'un environnement encombre (et toujours tres hydrophobe) au voisinage immediat de l'agregat assure la protection du centre 4fe-4s#3+ vis-a-vis des degradations ulterieures. L'ensemble des resultats obtenus autorisent a conclure sur la validite de l'hypothese de depart, laquelle preconisait l'emploi de ligands encombres pour l'obtention de tels modeles stables. Ce travail permet alors egalement de confirmer l'hypothese sur les proteines qui consiste a dire que l'hydrophobicite est un critere fondamental pour qu'une proteine a 4fe-4s fonctionne comme une hipip plutot que comme une ferredoxine

L'agent actif considéré a été l'éthoprophos, principe nématicide utilisé en agrochimie, mais dans un premier temps, l'aniline a été utilisé comme agent simulant. Les processus d'absorption et de désorption dans l'eau, ont été modélisés, dans le cas d'un granulé composé d'EVA pur, à l'aide d'une solution analytique de l'équation de Fick. Un modèle mathématique basé sur une méthode numérique a été construit pour décrire les transferts à travers un granulé composé par un noyau et une enveloppe

L'utilisation de pompes à chaleur fonctionnant à très hautes températures permettrait de pallier au déficit de chaleur à haut niveau (500°C) très souvent constaté sur les grandes installations industrielles qui, par contre, présentent généralement un excédent de chaleur à bas niveau (200°C). Les travaux présentés dans ce mémoire s'inscrivent dans le cadre d'une étude des pompes à chaleur à absorption susceptibles d'être utilisées dans le domaine des températures comprises entre 200°C et 1 200°C ; ceci au moyen de fluides de travail constitués d'un mélange de métaux liquides ou de sels fondus et, plus particulièrement, des mélanges mercure-sodium et trichlorure d'aluminium-chlorure de sodium. Pour chaque système, les propriétés physico-chimiques ont été estimées et l'équilibre liquide-vapeur a été décrit à l'aide de modèles thermodynamiques. Dans le cas du système mercure-sodium, ce modèle a été validé à l'aide des mesures de pression de vapeur réalisées sur une installation pilote. Associés aux bilans de matière et d'énergie établis dans le cas d'une pompe à chaleur idéale, ces modèles ont permis de déterminer, en fonction des conditions de fonctionnement (température, pression,. . . ) Les coefficients de performance de tels thermotransformateurs. La désorption du mercure, par ébullition nucléée d'amalgames de sodium, a été étudiée au moyen d'expériences spécifiques et, dans l'optique du dimensionnement de prototypes, les coefficients de transfert de chaleur ont été déterminés. Plusieurs projets de pompes à chaleur, à caractère expérimental ou industriel, sont proposés. Ces derniers utilisent le système mercure-sodium et leurs coefficients de performance théoriques sont compris entre 1,5 et 1,9.

Étude des possibilités offertes par les méthodes semi-empiriques de la chimie quantique, en matière de modélisation moléculaire. Étude du caractère acidobasique d'une molécule (amine, alcool). Étude du phénomène de liaison hydrogène. Modélisation de la surface des alumines amorphes

Au cours du procédé d'électrodialyse avec membrane de filtration (EDMF), les peptides chargés migrent sélectivement à travers des membranes de filtration (MFs) dans les compartiments respectifs de récupération des peptides anioniques (ARC) ou cationiques (C+ RC). Par conséquent, le type d'interaction entre les peptides et l'interface de la MF, en raison de ses propriétés physicochimiques, doit avoir un impact significatif sur la performance globale de l’EDMF (migration et sélectivité des peptides). Donc, l'objectif principal de cette thèse de doctorat était d'étudier les propriétés physicochimiques principales des MFs qui contribuent aux interactions interfaciales peptide-membrane facilitant ou entravant la migration globale et la séparation sélective des peptides pendant l’EDMF, et de comprendre les mécanismes impliqués dans ces interactions. Ainsi, dans cette étude, 16 MFs, caractérisées en termes de propriétés physicochimiques (potentiel zêta, conductivité, nature hydrophile/hydrophobe de la surface et des pores, épaisseur, rugosité, porosité et pourcentage de distribution des macropores dans la couche filtrante), ont été testées lors de l'EDMF pour séparer simultanément les peptides anioniques et cationiques d'un hydrolysat de protéines de lactosérum complexe et bien caractérisé. Dans la première étude, 6 MFs, différentes en termes de matériau, ont été testées incluant une membrane d’ultrafiltration (polyéthersulfone (PES)) comme contrôle et cinq membranes de microfiltration (fluorure de polyvinylidène (PVDF) et chlorure de polyvinyle (PVC-silice, fonctionnalisée (sulfopropyle ou amine quaternaire) ou non)). Les analyses de redondance (RDA) et de régression multivariées ont démontré qu’au moins deux des quatre propriétés suivantes des MF avaient un impact significatif sur la migration de tout peptide chargé ; le potentiel zêta, l’hydrophilie de surface/des pores, la porosité et la rugosité. De plus, l'effet important de la taille des pores sur la sélectivité des peptides a également été rapportée dans cette étude. Enfin, des modèles statistiques prédictifs qui relient la migration des peptides avec les propriétés de MF significatives ont été proposés. Dans la deuxième étude, réalisée sur des membranes de PES avec une large gamme de seuils de coupure (MWCO) de (5 à 300 kDa), une relation linéaire a été observée entre le MWCO et la migration globale des peptides (MGP) pour les deux compartiments de récupération. iii Cependant, la migration sélective des peptides vers ARC ou C+ RC s'est révélée être influencée par le MWCO des MFs ainsi que par les propriétés physicochimiques (charge et poids moléculaire (PM)) des peptides ; la migration d'un peptide ayant un faible PM et une faible charge (positive ou négative) était favorisée lorsqu’une MF ayant un petit MWCO était utilisée, tandis que l’inverse se produisait pour un peptide ayant un PM élevé et une charge élevée. Dans la troisième étude, l'effet de la combinaison du matériau de la membrane (polyacrylonitrile (PAN), PES et PVDF) /MWCO (30 et 50 kDa) sur la migration et la sélectivité des peptides, a tout d’abord été étudié. Les effets simples du matériau membranaire et du MWCO sur la MGP vers C+ RC, de même que l'effet combiné des matériaux membranaires/MWCO sur la MGP vers ARC et la migration sélective des peptides vers les deux compartiments de récupération ont été observés. Deuxièmement, une RDA réalisée sur l’ensemble des données obtenues pour les MFs sélectives testées dans cette recherche doctorale, a démontré l'impact significatif du potentiel zêta, de la conductivité, de la rugosité et du pourcentage de distribution des macropores dans la couche filtrante des MFs sur la MGP. Concernant la migration sélective des peptides, en plus des propriétés des MFs susmentionnées, l'impact significatif de l'angle de contact a été démontré pour au moins la migration d’un peptide anionique et/ou cationique vers leurs compartiments de récupération respectifs. Ces propriétés significatives ont favorisé différentes interactions telles qu’électrostatique, exclusion de taille et hydrophile/hydrophobe entre l’interface de la MF et le peptide, ce qui a eu pour effet de, soit faciliter, soit inhiber la migration de ce peptide. Enfin, des modèles statistiques prédictifs globaux ont été développés pour la MGP et pour la migration de chaque peptide individuel vers ARC et/ou C+ RC en fonction des propriétés importantes de la MF utilisée. Ces modèles permettent ainsi l'estimation du comportement de migration de ces peptides lorsque les MFs, sur une large gamme de propriétés physicochimiques, sont utilisées en EDMF. Les résultats obtenus dans cette thèse ont démontré, pour la première fois, la corrélation significative entre les propriétés physicochimiques des MFs, et la migration et la sélectivité des peptides pendant l'EDMF. Cependant, les modèles prédictifs développés dans cette étude iv peuvent être utilisés pour la gamme de peptides et les propriétés physicochimiques des MFs testées. Par contre, les mécanismes et explications proposés dans cette étude, concernant les interactions MF/peptide, peuvent être généralisés afin de comprendre tous les types d'interactions peptide/membrane. Comme perspectives à ce travail, l’étude de différentes sources d'hydrolysats, d’autres MFs et d’un hydrolysat produit par d’autres enzymes permettra la validation de ces modèles statistiques et leur généralisation.<br>During electrodialysis with filtration membranes (EDFM), charged peptides selectively migrate through filtration membranes (FMs) to their respective anionic (ARC) or cationic (C + RC) peptide recovery compartments. Consequently, the type of interactions occurring between FM and peptide at the interface, due to their physicochemical properties, must have significant impact on overall EDFM performances (peptide migration and selectivity). Therefore, the main objective of this doctoral thesis was to investigate the major FM properties that contribute to peptide-membrane interactions at the interface, which either facilitates or hinders global migration and selective separation of peptides during EDFM, and to understand the mechanisms involved behind those interactions. Thus, in this study, 16 FMs, characterized in terms of their physicochemical properties (zeta potential, conductivity, hydrophilic/hydrophobic nature of the surface and pores, thickness, roughness, porosity and percentage of macropores distribution in filtrating layer) were tested during EDFM to simultaneously separate anionic and cationic peptides from a well-characterized complex whey protein hydrolysate. In the first study, 6 FMs were tested, differing in terms of membrane materials, including one ultrafiltration (polyethersulfone (PES)) as a control and 5 microfiltration ( one polyvinylidene fluoride (PVDF) and four polyvinyl chloride (PVC)-silica: two functionalized (sulfonyl or amino) or two non-functionalized). Redundancy analysis (RDA) and multivariate regression analysis demonstrated that at least two FM properties among zeta potential, pore/surface hydrophilicity, porosity and roughness significantly impacted the migration of any charged peptide. In addition, the important effect of pore size on peptide selectivity was also reported. Finally, predictive statistical models that link each peptide migration with significant FM properties were proposed. In the second study, which was carried out on PES membranes with a wide range of molecular weight cut-offs (MWCOs) (5 kDa to 300 kDa), a linear relation was noticed between MWCO and global peptide migration (GPM) to both recovery compartments. However, the selective peptide migration to A - RC or C + RC was found to be influenced by the vi MWCO of FMs as well as physicochemical properties (charge and molecular weight (MW)) of peptides. For instance, the migration of a peptide having low MW and low charge (positive or negative) was favored when a FM with small MWCO was used, while the opposite was observed for a peptide having high MW and high charge. In the third study, the effect of combination of membrane material (PAN, PES and PVDF)/MWCO (30 and 50 kDa) on peptide migration and selectivity was first studied. The simple effect of membrane material and MWCO on GPM to C+ RC was observed, while the combined effect of membrane materials/MWCO on GPM to A - RC and selective peptide migration to both recovery compartments was observed. Secondly, a RDA was performed on the data obtained for all the selective FMs tested in this doctoral research, which demonstrated the significant impact of zeta potential, conductivity, roughness and percentage of macropores distribution in the filtrating layer of FMs on GPM. Concerning selective peptide migration, in addition to the aforementioned FM properties, the significant impact of contact angle was noticed for at least one anionic and/or cationic peptide migration to their respective recovery compartments. These significant FM properties were found to trigger different interactions such as electrostatic, size exclusion and hydrophilic/hydrophobic between FM and peptide at the interface resulting in either facilitation or inhibition of peptide migration. Finally, global predictive statistical models were developed for GPM and each individual peptide migration to ARC and/or C+ RC based on these significant FM properties, which allow the estimation of their migration behavior when FMs having a wide range of physicochemical properties are used during EDFM. The results obtained in this Ph.D. thesis demonstrated, for the first time, the significant correlation between physicochemical properties of FMs, and peptide migration and selectivity during EDFM. The predictive models developed in this study can be used for the range of peptides and FMs tested. Moreover, the types of interactions occurring between FMs and peptide at the interface, and mechanisms and explanations proposed in this study can be applied to understand all types of peptide/membrane interactions. Validation of such models vii by using different sources of hydrolysates or different FMs or a hydrolysate produced by other enzymes will be the main perspectives of this research work.

Key physicochemical properties for a wide spectrum of chemical pollutants are unknown. This thesis analyses the prospect of assessing the environmental distribution of chemicals directly from supervised learning algorithms using molecular descriptors, rather than from multimedia environmental models (MEMs) using several physicochemical properties estimated from QSARs. Dimensionless compartmental mass ratios of 468 validation chemicals were compared, in logarithmic units, between: a) SimpleBox 3, a Level III MEM, propagating random property values within statistical distributions of widely recommended QSARs; and, b) Support Vector Regressions (SVRs), acting as Quantitative Structure-Fate Relationships (QSFRs), linking mass ratios to molecular weight and constituent counts (atoms, bonds, functional groups and rings) for training chemicals. Best predictions were obtained for test and validation chemicals optimally found to be within the domain of applicability of the QSFRs, evidenced by low MAE and high q2 values (in air, MAE&#8804;0.54 and q2&#8805;0.92; in water, MAE&#8804;0.27 and q2&#8805;0.92).<br>Las propiedades fisicoquímicas de un gran espectro de contaminantes químicos son desconocidas. Esta tesis analiza la posibilidad de evaluar la distribución ambiental de compuestos utilizando algoritmos de aprendizaje supervisados alimentados con descriptores moleculares, en vez de modelos ambientales multimedia alimentados con propiedades estimadas por QSARs. Se han comparado fracciones másicas adimensionales, en unidades logarítmicas, de 468 compuestos entre: a) SimpleBox 3, un modelo de nivel III, propagando valores aleatorios de propiedades dentro de distribuciones estadísticas de QSARs recomendados; y, b) regresiones de vectores soporte (SVRs) actuando como relaciones cuantitativas de estructura y destino (QSFRs), relacionando fracciones másicas con pesos moleculares y cuentas de constituyentes (átomos, enlaces, grupos funcionales y anillos) para compuestos de entrenamiento. Las mejores predicciones resultaron para compuestos de test y validación correctamente localizados dentro del dominio de aplicabilidad de los QSFRs, evidenciado por valores bajos de MAE y valores altos de q2 (en aire, MAE&#8804;0.54 y q2&#8805;0.92; en agua, MAE&#8804;0.27 y q2&#8805;0.92).

Les processus de transfert de pesticide dans les sols sont encore mal connus. Avec l'objectif de mieux comprendre les mécanismes impliqués, des essais dans des colonnes remplies avec différents types de sols ont été effectués sous les conditions de saturation et de saturation partielle en eau. Des expériences avec un traceur ont été utilisées pour la caractérisation hydrodynamique du milieu. Les mécanismes d'interaction physicochimiques d'une molécule pesticide modèle, le pentachlorophenol (pcp) avec la matrice solide, sont observés par la comparaison des courbes de restitution du pesticide avec celles du traceur. Les résultats expérimentaux ont montré que les interactions du pcp avec le solide sont dépendants de la succession chronologique des expériences, de la vitesse moyenne d'écoulement, du profil de la teneur en eau dans la colonne, de la concentration en pcp, du ph et de la présence de sel dans la solution injectée. L'isotherme d'adsorption du pcp sur le sable suit le modèle de freundlich, et la restitution partielle est probablement due à une fixation irréversible du pcp ionisé ou non. Un modèle numérique, résolu à l'aide de la méthode des différences finies a été développé. Les descriptions des courbes de restitution des solutes obtenues sur un sable en condition de saturation en eau se sont avérées satisfaisantes. En revanche, le modèle n'a pas permis de décrire complètement les courbes d'élution résultantes de créneaux injectés dans la colonne de sable saturé en semence avec des bactéries, dans le sable partiellement saturé et dans le sol saturé

У дисертації висвітлюються актуальні питання визначення положення газонафтового і водонафтового контактів на “водоплавних” нафтогазоконденсатних покладах за результатами геофізичних досліджень у відкритому стовбурі свердловини. Такі поклади є об'єктами з дуже складними електричними властивостями присвердловинної зони і потребують підвищення інформативності комплексу методів ГДС. Теоретично обґрунтовано і доведено фактичними дослідженнями в свердловинах переваги методу ВІКІЗ для визначення параметрів геоелектричної моделі порід-колекторів, які виповнюють теригенні відклади юрського комплексу ЗСНГП. Методологічні властивості методу зумовлюють достовірнішу оцінку геоелектричних параметрів зони проникнення і дають можливість виявити облямовуючу зону. На базі геофізичної інформації під час дослідження розрізу свердловин та інформації по визначенню фізико-хімічних властивостей пластового флюїду, створено фізико-геологічну модель присвердловинної зони, яка характеризує залежність електричних параметрів присвердловинної зони від геологічних властивостей порід-колекторів з різним типом пластового флюїду. Свердловинні дослідження методом ВІКІЗ на Кинському і Харампурському нафтогазоконденсатних родовищах показують що зміна параметрів геоелектричного розрізу свердловини відбувається під час проходження свердловиною через ГНК і ВНК. Створена фізико-геологічна модель присвердловинної зони для «водоплавних» нафтогазоконденсатних покладів має можливість не тільки виділяти інтервал продуктивних колекторів, але і розділяти їх на нафтонасичені і газонасичені. Запропонована уніфікована схема використання порівняння газонасичених інтервалів, виділених за методом ВІКІЗ і способом ПНК, яка дає можливість визначити положення ГНК як на етапі оперативного висновку, так і на етапі побудови флюїдальної моделі покладу.<br>В диссертации освещаются актуальные вопросы определения положения газонефтяного и водонефтяного контактов на «водоплавающих» нефтегазоконденсатных залежах по результатам геофизических исследований в открытом стволе скважины. Такие залежи являются объектами с очень сложными электрическими свойствами прискважинной зоны и требуют повышения информативности комплекса методов ГИС. Теоретически обосновано и доказано на практических исследованиях преимущество метода ВИКИЗ при определении параметров геоэлектрической модели пород-коллекторов, которые составляют терригенные отложения юрского комплекса ЗСНГП. Методологические свойства метода обеспечивают более однозначную оценку геоэлектрических параметров зоны проникновения и дают возможность выявлять окаймляющую зону. На базе геофизической информации по исследованиям разрезов скважин и информации по определению физико-химических свойств пластового флюида, создано физикогеологическую модель прискважинной зоны, которая характеризует зависимость электрических параметров прискважинной зоны от геологических свойств пород-коллекторов с разным типом пластового флюида. Доказано, что в течении первых 5-10 часов после раскрытия разреза по электрическим параметрам техногенных неоднородностей с высоким уровнем достоверности можно определить тип насыщающего пластового флюида. Скважинные исследования методом ВИКИЗ на Кынском и Харампурском нефтегазоконденсатных месторождениях показывают, что изменение параметров геоэлектрического разреза скважины происходит при прохождении скважиной через ГНК и ВНК. Предложено унифицированную схему использования соответствия газонасыщенных интервалов, выделенных по методу ВИКИЗ и методике ПИК, которая дает возможность определять положение ПНК как на этапе оперативного заключения, так и на этапе построения флюидальной модели залежи.<br>The Theses illustrates relevant issues of gas-oil and oil-water contacts position location in bottom water-drive oil-gas condensate reservoirs, belonging to laminated deposits of Jurassic bedrock in West-Siberian petroleum province. Such deposits are objects with very complicated electrical properties of the well bore zone, abruptly changing with the deposit’s height. Investigation of deposits with multicomponent reservoir fluid composition requires implementation of new geophysical wells survey methods and increase of useful information volume, extractable from the carried out complex of geographical information system (GIS). As a result of the author’s investigations, a number of scientific conclusions were drawn, which represent great practical importance in the field of geophysical survey of oil and gas wells and geological simulation for deposit fluidic structures with multicomponent composition of reservoir fluid. Advantages of the VIKIZ method for surveying lamellar terrigenous deposits of West-Siberian petroleum province of Jurassic horizon have been theoretically validated and proved in practice. Based on geophysical information and the information on formation fluid property investigations, main points and directions for a physical and geological model were formulated. The model characterizes relation between the penetration zone parameters and geological properties of reservoirs with different types of fluids. A method was developed which allows locating the OWC based on the parameters of the penetration zone and the low-resistivity zone. This can be important information for of specification standard oil-water surface interpretation results. A physical and geological model was created which allows discriminating oil-saturated reservoirs in the pay zone from gas-saturated ones and determines OWC position in bottom water-drive oil and gas condensate reservoirs of West-Siberian petroleum province. A unified scheme for gas-saturated formation comparison usage was suggested. The intervals were distinguished based on the VIKIZ method and PNK technique, which allowed determining GOC position at the operative conclusion stage, as well as at the stage of fluidic model deposit creation.

Etude de petrologie experimentale sur les transformations a hautes pressions et hautes temperatures, de silicates et de germanates a structure olicine et pyroxene. Il a ete determine en fonction de la pression et de la temperature, la distribution du magnesium, du fer et du calcium entre mineraux mantelliques. Un modele thermodynamique coherent est construit et a permis de determiner le champ de stabilite de la perovskite silicatee et de proposer une interpretation de la discontinuite sismique des 670 km. Des mecanismes microscopiques de transformations de phase de l'olivine sont proposes. De quelques composes etudies par spectroscopie raman, a ete tire des parametres thermodynamiques harmoniques et anharmoniques de ces phases

L'acquisition du fer et l'echange de ce metal entre deux siderophores sont des processus biologiques majeurs. Des etudes preliminaires ont montre l'existence de complexes ternaires intermediaires lors de l'echange de ligand. Nous presentons une approche structurale des complexes ternaires par le biais de nouveaux ligands de syntheses. Quatre ligands ont ete prepares, deux comportent des structures mixtes : deux sortes de sous-unites regroupees sur une charpente tripodale. Deux voies de syntheses ont ete envisagees. Dans la premiere, la charpente tripodale etait obtenue par alkylation d'une amine secondaire. La seconde voie selectionnee implique le couplage des sous-unites sur un espaceur tripodal, le tris-(1,2-aminoethylamine) ou tren. Les intermediaires clefs de cette synthese sont des molecule possedant un groupe amine primaire sur l'un des bras de l'espaceur. Les ligands ont ete rendus hydrosolubles par sulfonation controlee du systeme aromatique des sous-unite chelatantes. En absence de cristaux, les etudes des complexes ont ete realisees par spectrometrie de masse electrospray, par spectrophotometrie uv-visible, et par etudes rmn des complexes de gallium iii correspondants. Les etudes physico-chimiques nous ont permis de determiner le comportement de ces modeles de siderophores lors de la complexation du fer (iii) : par les mesures de pka, des courbes de distribution des especes en fonction du ph et des constantes thermodynamiques de complexation. L'activite biologique des ligands a ete evaluee par mesure de leur capacites nutritives sur des cultures cellulaires de plantes. Un de nos complexe c'est avere etre plus efficace que les complexes de l'edta, traditionnellement employes comme source de fer dans les milieux de culture.

Nutritional oil-in-water emulsion products are usually fortified with minerals, which can interact with other ingredients and cause physiochemical instability reducing product shelf-life. The objective of this study was to investigate the impact of mineral-ingredient interactions on properties and stability of protein stabilized emulsion in order to find potential approaches to prevent the emulsion instability. Commercial nutritional beverages had mean particle diameters between 0.5–11 μm and pH values around 7. The monovalent and divalent minerals present in the products at the highest average concentrations were potassium and calcium, respectively. Both calcium and potassium ions induced droplet aggregation in a model emulsion, which had a mean particle size of 0.6–0.7 μm, pH value at 7, 7% oil, and was stabilized by whey protein isolate (WPI). The aggregation led to an increase in emulsion particle size, apparent viscosity, shear-thinning behavior, and creaming instability with increasing mineral concentration. The critical concentration of the aggregation was 3–7 mM for calcium and 200 mM for potassium. The emulsion particle surface charge was decreased by minerals, probably because of electrostatic screening and binding of the mineral ions to negatively charged adsorbed proteins. Calcium binding affinity and enthalpy of EDTA was greater than those of citrate and WPI, respectively. EDTA and citrate bound calcium ions in 1:1 ratio, whereas WPI bound to calcium ions in about 1:3 to 1:4 ratio. WPI that was heated at 70–90°C had an increase in calcium binding affinity and enthalpy. Citrate consisted of two types of binding sites. EDTA, citrate, and non-adsorbed WPI prevented or reduced calcium-induced droplet aggregation with increasing their concentration. EDTA was more effective than citrate and WPI, respectively. The chelating agents lowered free-calcium concentration and magnitude of particle surface charge. The emulsions had less free-calcium ions to induce the instability. Heat treatment of non-adsorbed WPI at 70 or 90°C did not improve the performance of non-adsorbed WPI. EDTA, citrate, and non-adsorbed WPI could not prevent droplet aggregation induced by thermal treatment, when emulsions containing calcium were heated at 80–120°C for 15 min. It was suggested that hydrophobic attraction upon heating induced droplet aggregation.

碩士<br>國立中興大學<br>食品暨應用生物科技學系所<br>99<br>Instant rice flour belongs to the pregelatinized cereal products group and mainly uses the dry methods: puffing, roasting or wet methods: drum drying, steaming or extrution to carry out gelatinization, drying, grinding, sieving to obtain the cereal particles and paste. The puffing gun and drum dryer are often used to produce edible cereal products because of their low costs and easy to use. This study uses puffing and drum drying to prepare instant rice flour and examines the effects of (1) different puffing pressures (0, 3, 5, 7, 9, 11kg/cm2), (2) cold and hot slurry feed, (3) different surface temperatures, (4) amylase content of rice (TCN1, TK9 and TCW70) on the physicochemical properties, rehydration properties and product turbidity stability; and multiple regression analysis was used to set up math models of the processing effect parameter of puffing and drum drying, absorbent index, and solubility. Moreover, modify the quality of instant rice flour by adding gums and puffed TCW70 -11 kg / cm2. The results show that the WAI, WSI, SP, degree of gelatinization, dextrin content, separation ratio and turbidity stability of puffed instant rice flour increase with the puffing pressure rises, but particle size and molecular weight decrease. The products’ hydration preperties and separation ratio can be estimated by three math model: WAI = 8.00100－0.233223×A+ 0.014585×T+ 0.0288663×P (r 2=0.87), SP = 10.70046－0.42053×A+ 0.04572×T+0.40982×P (r 2=0.84), WSI = 7.514986－0.821779×A +0.131696×T +1.730803×P (r 2=0.87) and separation ratio = 0.538692-0.020034×A + 0.002299 T+0.049799×P (r 2=0.89). Multiple-porous structures and completed gelatinization are found in drum dried instant rice flour, and it’s WSI, dextrin, turbidity stability and separation ratio increase with surface temperature increase. The hydration preperties and separation ratio of drum dried instant rice flour also can be estimated by math modle, respectively: cold slurry feeded could be used by WAI = 9.861863－0.334415A +0.087127T+0.039369TS (r2=0.87), SP = 1.101079－0.481397A +0.199154T+0.106375TS (r2=0.95) and separation ratio = 0.562169-0.019469A+0.000952T+0.003206Ts(r2= 0.94); Hot slurry feeded could be used by WSI=53.87873－2.61321A+ 0.21491T+0.09996TS (r2= 0.99) and separation ratio =0.230230- 0.014633A+ 0.001675T+0.005635Ts (r2= 0.88). There are good separation stability, turbidity stability and most popular overall acceptance when puffed instant rice flour is modified by adding 0.2 % guan gum and drum dried instant rice flour is modified by adding 2 % puffed TCW70 -11 kg / cm2 flour.

Yes<br>Lipid membrane-water partitions (e.g., immobilized artificial membrane systems where the lipid membrane is a neutral phospholipid monolayer bound to gel beads) were compared to various organic solvent-water partitions using linear free energy relationships. To this end, we also measured the retention factors of 36 compounds (including neutral and ionic species) from water to liposomes made up of 3-sn-phosphatidylcholine and 3-sn-phosphatidyl-l-serine (80:20, mol/mol), employing liposome electrokinetic chromatography in this work. The results show that lipid membranes exhibit a considerably different chemical environment from those of organic solvents. For both neutral species and ionic species, partitions into the more polar hydroxylic solvents are chemically closer to partition into the lipid membrane as compared to partitions into the less polar hydroxylic solvents and into aprotic solvents. This means that solutes partition into the polar parts of lipid membranes, regardless of whether they are charged or not. In addition, cerasome (i.e., liposome composed mainly of stratum corneum lipids) was compared with regular phospholipid liposomes as a possible model for human stratum corneum in partitions. It was found that the cerasome-water partition exhibits a better chemical similarity to skin permeation. This is probably due to the unique structures of ceramides that occur in cerasome and in the stratum corneum lipid domain. We further show that membranes in membrane-water partitions exhibit very different properties.

No<br>The electrophoretic mobilities of three beta-blocker drug practolol, timolol and propranolol, have been measured in electrolyte systems with mixed binary and ternary water-methanol-ethanol solvents with acetic acid/sodium acetate as buffer using capillary electrophoresis. The highest mobilities for the analytes studied have been observed in pure aqueous. the lowest values in ethanolic buffers The measured electrophoretic mobilities have been used to evaluate the accuracy of a mathematical model based on a mixture response surface method that expresses the mobility as a function of the solvent composition. Mean percentage error (MPE) has been computed considering experimental and calculated mobilities as an accuracy criterion. The obtained MPE for practolol, timolol and propranolol in the binary mixtures are between 0.9 and 2.6%, in the ternary water-methanol-ethanol solvent system the MPE was about 2.7%. The MPE values resulting from the proposed equation lie within the experimental relative standard deviation values and can he considered as an acceptable error.