Automatic learning for the classification of chemical reactions and in statistical thermodynamics

This Thesis describes the application of automatic learning methods for a) the classification of organic and metabolic reactions, and b) the mapping of Potential Energy Surfaces(PES). The classification of reactions was approached with two distinct methodologies: a representation of chemical reactions based on NMR data, and a representation of chemical reactions from the reaction equation based on the physico-chemical and topological features of chemical bonds. NMR-based classification of photochemical and enzymatic reactions. Photochemical and metabolic reactions were classified by Kohonen Self-Organizing Maps (Kohonen SOMs) and Random Forests (RFs) taking as input the difference between the 1H NMR spectra of the products and the reactants. The development of such a representation can be applied in automatic analysis of changes in the 1H NMR spectrum of a mixture and their interpretation in terms of the chemical reactions taking place. Examples of possible applications are the monitoring of reaction processes, evaluation of the stability of chemicals, or even the interpretation of metabonomic data. A Kohonen SOM trained with a data set of metabolic reactions catalysed by transferases was able to correctly classify 75% of an independent test set in terms of the EC number subclass. Random Forests improved the correct predictions to 79%. With photochemical reactions classified into 7 groups, an independent test set was classified with 86-93% accuracy. The data set of photochemical reactions was also used to simulate mixtures with two reactions occurring simultaneously. Kohonen SOMs and Feed-Forward Neural Networks (FFNNs) were trained to classify the reactions occurring in a mixture based on the 1H NMR spectra of the products and reactants. Kohonen SOMs allowed the correct assignment of 53-63% of the mixtures (in a test set). Counter-Propagation Neural Networks (CPNNs) gave origin to similar results. The use of supervised learning techniques allowed an improvement in the results. They were improved to 77% of correct assignments when an ensemble of ten FFNNs were used and to 80% when Random Forests were used. This study was performed with NMR data simulated from the molecular structure by the SPINUS program. In the design of one test set, simulated data was combined with experimental data. The results support the proposal of linking databases of chemical reactions to experimental or simulated NMR data for automatic classification of reactions and mixtures of reactions. Genome-scale classification of enzymatic reactions from their reaction equation. The MOLMAP descriptor relies on a Kohonen SOM that defines types of bonds on the basis of their physico-chemical and topological properties. The MOLMAP descriptor of a molecule represents the types of bonds available in that molecule. The MOLMAP descriptor of a reaction is defined as the difference between the MOLMAPs of the products and the reactants, and numerically encodes the pattern of bonds that are broken, changed, and made during a chemical reaction. The automatic perception of chemical similarities between metabolic reactions is required for a variety of applications ranging from the computer validation of classification systems, genome-scale reconstruction (or comparison) of metabolic pathways, to the classification of enzymatic mechanisms. Catalytic functions of proteins are generally described by the EC numbers that are simultaneously employed as identifiers of reactions, enzymes, and enzyme genes, thus linking metabolic and genomic information. Different methods should be available to automatically compare metabolic reactions and for the automatic assignment of EC numbers to reactions still not officially classified. In this study, the genome-scale data set of enzymatic reactions available in the KEGG database was encoded by the MOLMAP descriptors, and was submitted to Kohonen SOMs to compare the resulting map with the official EC number classification, to explore the possibility of predicting EC numbers from the reaction equation, and to assess the internal consistency of the EC classification at the class level. A general agreement with the EC classification was observed, i.e. a relationship between the similarity of MOLMAPs and the similarity of EC numbers. At the same time, MOLMAPs were able to discriminate between EC sub-subclasses. EC numbers could be assigned at the class, subclass, and sub-subclass levels with accuracies up to 92%, 80%, and 70% for independent test sets. The correspondence between chemical similarity of metabolic reactions and their MOLMAP descriptors was applied to the identification of a number of reactions mapped into the same neuron but belonging to different EC classes, which demonstrated the ability of the MOLMAP/SOM approach to verify the internal consistency of classifications in databases of metabolic reactions. RFs were also used to assign the four levels of the EC hierarchy from the reaction equation. EC numbers were correctly assigned in 95%, 90%, 85% and 86% of the cases (for independent test sets) at the class, subclass, sub-subclass and full EC number level,respectively. Experiments for the classification of reactions from the main reactants and products were performed with RFs - EC numbers were assigned at the class, subclass and sub-subclass level with accuracies of 78%, 74% and 63%, respectively. In the course of the experiments with metabolic reactions we suggested that the MOLMAP / SOM concept could be extended to the representation of other levels of metabolic information such as metabolic pathways. Following the MOLMAP idea, the pattern of neurons activated by the reactions of a metabolic pathway is a representation of the reactions involved in that pathway - a descriptor of the metabolic pathway. This reasoning enabled the comparison of different pathways, the automatic classification of pathways, and a classification of organisms based on their biochemical machinery. The three levels of classification (from bonds to metabolic pathways) allowed to map and perceive chemical similarities between metabolic pathways even for pathways of different types of metabolism and pathways that do not share similarities in terms of EC numbers. Mapping of PES by neural networks (NNs). In a first series of experiments, ensembles of Feed-Forward NNs (EnsFFNNs) and Associative Neural Networks (ASNNs) were trained to reproduce PES represented by the Lennard-Jones (LJ) analytical potential function. The accuracy of the method was assessed by comparing the results of molecular dynamics simulations (thermal, structural, and dynamic properties) obtained from the NNs-PES and from the LJ function. The results indicated that for LJ-type potentials, NNs can be trained to generate accurate PES to be used in molecular simulations. EnsFFNNs and ASNNs gave better results than single FFNNs. A remarkable ability of the NNs models to interpolate between distant curves and accurately reproduce potentials to be used in molecular simulations is shown. The purpose of the first study was to systematically analyse the accuracy of different NNs. Our main motivation, however, is reflected in the next study: the mapping of multidimensional PES by NNs to simulate, by Molecular Dynamics or Monte Carlo, the adsorption and self-assembly of solvated organic molecules on noble-metal electrodes. Indeed, for such complex and heterogeneous systems the development of suitable analytical functions that fit quantum mechanical interaction energies is a non-trivial or even impossible task. The data consisted of energy values, from Density Functional Theory (DFT) calculations, at different distances, for several molecular orientations and three electrode adsorption sites. The results indicate that NNs require a data set large enough to cover well the diversity of possible interaction sites, distances, and orientations. NNs trained with such data sets can perform equally well or even better than analytical functions. Therefore, they can be used in molecular simulations, particularly for the ethanol/Au (111) interface which is the case studied in the present Thesis. Once properly trained, the networks are able to produce, as output, any required number of energy points for accurate interpolations.

Development of biocompatible and “smart” porous structures using CO2-assisted processes

Dissertação apresentada para a obtenção do grau de Doutor em Engenharia Química, especialidade Engenharia da Reacção Química, pela Universidade Nova de Lisboa, Faculdade de Ciências e Tecnologia

The expression in saccharomyces cerevisiae of a glucose/xylose symporter from Candida intermedia is affected by the presence of a glucose/xylose facilitator

Microbiology, 154

2,4,6- Ttrichloroanisol:a consumer panel evaluation

“Cork taint” is a major problem in wine industry and is caused by contamination of wines. This contamination is usually attributed to wine cork stoppers and 2,4,6-trichloroanisole (2,4,6-TCA) is one of the compounds mostly associated to this off-flavour. In this work, a consumer panel performed “forced choice” triangular tests in order to measure Odour Detection Thresholds (ODT) and Taste Detection Thresholds (TDT) of 2,4,6-TCA in water, hydro-alcoholic solutions (11.5% and 18% ethanol) and white and red wines. A paired preference test was also performed by the panel in order to measure Odour Rejection Threshold (ORT) in white and red wine spiked with 2,4,6-TCA. Results obtained show that the ODT and the TDT for 2,4,6-TCA in water were 0.2 and 0.3 ng/L, respectively. In hydro-alcoholic solutions with 11.5% and 18% ethanol the ODT were 4 and 10 ng/L respectively. In red wine the ODT and the TDT were 0.9 and 1.7 ng/L and in white wine were 1.5 and 1.0 ng/L respectively. ORT for white was 10.4 ng/L and for red wines 16.0 ng/L. These results suggest that although this group of consumers detected very low concentrations of 2,4,6-TCA in wines, they did not reject the wine at these low concentration values.

Synthesis of biodiesel from wastes of food industry via direct transesterification with methanol/carbon dioxide mixtures

Dissertation presented at Faculdade de Ciências e Tecnologia from Universidade Nova de Lisboa to obtain the degree of Master in Chemical and Biochemical Engineering

Modelling and optimization of the ion exchange membrane bioreactor for removal of anionic pollutants from drinking water streams

Dissertação para obtenção do Grau de Doutor em Engenharia Química, especialidade de Engenharia Bioquímica

Biogenic amines in wine: gene transcription as a tool for selection of Oenococcus oeni starter strains

Dissertation presented to obtain the Ph.D degree in Biology

Application of OSN in membrane cascade for purification of the API Amoxicillin

Dissertação para obtenção do Grau de Mestre em Engenharia Química e Bioquímica

Valorization of pumpkin residue by supercritical extraction of added-value compounds

Dissertação para a obtenção do Grau de Mestre em Engenharia Química e Bioquímica

Development of a "green process" for the isolation of natural functional extracts with anti-cancer activity - Application of high-pressure technology

Dissertação para obtenção do Grau de Mestre em Engenharia Química e Bioquímica

Malnutrição em idosos na admissão hospital actuação perante a sua elevado prevalência

RESUMO: Ao longo das últimas décadas a redistribuição etária da população mundial tem vindo a apresentar um aumento do número de pessoas com 65 ou mais anos, integrando um grupo populacional comummente designado por população idosa. Importa aprofundar mecanismos fisiológicos que conduzem ao envelhecimento e de que forma podem condicionar não só aspetos clínicos, como também nutricionais, entre outros, com a perspetiva da sua origem no aparecimento de doenças crónicas. Com esse enfoque, a desnutrição na pessoa idosa é hoje considerada pela European Nutrition for Health Alliance(ENHA) um problema de saúde pública. Está descrito que a sua prevalência ronda os 60% a nível de instituições hospitalares, 40% em unidades residenciais e 5 a 10% na pessoa idosa a residir em domicílio próprio ou de familiares, e na sua maioria permanece por diagnosticar e tratar. Assim, foi objetivo deste estudo caracterizar e estimar a prevalência da desnutrição e do risco de desnutrição na pessoa idosa, nas primeiras 72 horas de admissão hospitalar. Aplicou-se um estudo observacional, analítico, transversal, quantitativo e correlacional, cujos dados foram recolhidos por entrevista ao próprio e por observação. O estudo desenvolveu-se em duas vertentes de investigação, uma focada na caracterização da desnutrição em pessoas idosas institucionalizadas em hospitais portugueses da zona centro e sul do Continente e Madeira, nos períodos de julho/agosto de 2009, abril/junho de 2010, maio/julho de 2011, através do MNA®. A outra, uma avaliação nutricional detalhada, efetuada no Centro Hospitalar de Lisboa Central – Hospital de Santa Marta, EPE, entre o início de janeiro de 2009 e o fim de janeiro de 2010, sendo a amostra recrutada de entre os idosos de ambos os sexos, internados. Foram estudados dados sociodemográficos, de saúde e feita uma avaliação nutricional extensa. A avaliação nutricional constou de colheita de parâmetros laboratoriais (hematológicos e bioquímicos) e antropométricos (índice de massa corporal (IMC), prega cutânea tricipital(PCT), prega cutânea subescapular (PCSE), perímetro braquial (PB), adequação do perímetro braquial (APB), área muscular braquial (AMB) e perímetro Geminal (PG), análise da composição corporal (Massa Gorda Corporal (MGC), Massa Isenta de Gordura (MIG)),caracterização de um dia alimentar tipo e questionário Mini Nutritional Assessment Long Form®– MNA LF®. Dos dados obtidos em hospitais portugueses, destaca-se que dos 402 idosos avaliados, 53% eram do sexo masculino, tinham uma idade média de 75,8 + 6,52 (65 – 100) e segundo o MNA® 57,5% encontravam-se Desnutridos ou em Risco de Desnutrição.Na amostra, dos dados obtidos, a nível sociodemográfico salienta-se que 50% dos doentes eram do sexo masculino, a idade média rondava os 75,5 + 7,22 (65 – 100) anos, 55% eram naturais de Lisboa e 80% residiam em Lisboa e Vale do Tejo, 38% não tiveram estudos formais e 43% fizeram-no apenas até ao 4º ano de escolaridade. Em relação aos dados de saúde, a maioria dos doentes foi admitida através do Serviço de Urgência do Centro Hospitalar de Lisboa Central – Hospital de São José e foram internados no Serviço de Medicina (38%) e no Serviço de Cardiologia (30%), por patologia médica (38%) e patologia do sistema circulatório (56%). Nos hábitos de vida, quanto à mobilidade, um terço dos doentes estavam acamados e os restantes deambulavam ou tinham uma mobilidade normal, 74% não apresentaram hábitos etanólicos regulares, 19% apresentavam um consumo elevado (> 30g de etanol/dia); 95% dos doentes não apresentavam hábitos tabágicos. Relativamente à caracterização nutricional, os valores médios encontrados em relação aos parâmetros laboratoriais revelaram-se inferiores aos valores padrão para a idade e sexo e eram inferiores no sexo feminino. Na caracterização antropométrica verificaram-se os seguintes achados: o cálculo do IMC mostrou-se pouco sensível na identificação de doentes desnutridos; a PCT e a PCSE revelaram valores de massa gorda dentro do intervalo considerado normal;segundo o PB, 88% não apresentavam valor indicativo de desnutrição e 8% estavam desnutridos; a APB identificou 50% de doentes desnutridos; a AMB, revelou que 97% dos homens e 95% das mulheres apresentavam deficit da massa magra e segundo o PG, 18% apresentavam um valor inferior a 31cm descritor de desnutrição. Na análise da composição corporal verificou-se que ambos os sexos apresentavam uma percentagem de MGC classificada como demasiado alta e que esta era superior nas mulheres em relação aos homens. Ao analisar a ingestão nutricional verificou-se que esta era inferior às Dietary Reference Intakes (DRIs) para a ingestão hídrica (p=0,00), energética (p=0,00), proteica (p=0,00), lipídica (p=0,01), MUFA (p=0,00), PUFA (p=0,00), e glícidos (p=0,00), fibra (p=0,02), potássio (p=0,00), cálcio (p=0,00), magnésio (p=0,00), fósforo (p=0,00), zinco (p=0,00), vitamina D (p=0,00), vitamina E (p=0,00) e folato (p=0,00). No que diz respeito ao MNA®, a sua aplicação permitiu identificar 62% de situações de risco nutricional ou de desnutrição já instalada. Valores de MNA® indicativos de desnutrição ou risco estavam associados a níveis de escolaridade mais baixos (r=0,32; p=0,00). Verificou-se correlação entre o MNA® e a PCT (r=0,30;p=0,00), PCSE (r=0,19;p=0,03) e PG (r=0,27;p=0,00). Na análise da amostra por sexo e escalão etário, apenas se distinguiram as mulheres mais velhas, que apresentaram situação de IMC indicador de risco de desnutrição (IMC <23,5 + 2,9, (r=0,42;p=0,02)), e de valores médios de PB de 25,6+3,84cm (r=0,42;p=0,01), em situações de menor mobilidade caraterizados pelo MNA®. Os homens maisvelhos apresentaram correlação entre o MNA® e PCSE (r=0,41;p=0,02), APMB (r=0,57;p=0,00)e PG (r=0,55;p=0,00), e as mulheres mais velhas apenas com a PCT (r=0,39;p=0,02). A análise multivariada do MNA® em função do sexo e do escalão etário, revelou que estes são independentes. Os homens apresentaram valores médios de MNA® superiores às mulheres e à medida que a idade aumenta, os valores de MNA® em ambos os sexos diminuem, sendo indicativos de risco de desnutrição. Consideramos que, tendo em conta a natureza e objetivos do presente estudo, foi possível caracterizar e estimar a prevalência da desnutrição e do risco de desnutrição em pessoas idosas nas primeiras 72 horas de admissão hospitalar. Os resultados obtidos sinalizam a sua elevada prevalência e alertam para a necessidade de procedimentos protocolados de avaliação e intervenção nutricional da população idosa na admissão hospitalar. Para este efeito a aplicação do MNA® provou a sua aplicabilidade, assim como a medição e cálculo da AMB, que poderão ser muito precocemente aplicados e contribuir para potenciar melhorias do estado de saúde e diminuir o tempo de internamento, nomeadamente de pessoas idosas. Em relação ao padrão alimentar, este estudo contribuiu para uma chamada de atenção dos profissionais de saúde que a população idosa pode apresentar carências nutricionais na admissão, e que estas se não forem devidamente sinalizadas e colmatadas tendem a agravar-se durante o internamento podendo contribuir para o aumento da morbilidade.-------------ABSTRACT:Over the last decades the age redistribution group of the population worldwide has been presenting an increasing number of people aged 65 years or more, incorporating a population group commonly referred to as the elderly population. It´s important to further analyze the physiological mechanisms that lead to aging and how they might influence not only clinical aspects, but also nutritional, among others, with the perspective of their origin in the onset of chronic diseases. With this approach, malnutrition in the elderly is now considered by the European Nutrition for Health Alliance (ENHA) a public health problem. It is reported that its prevalence is around 60% at the level of hospital units, 40% in residential units and 5 to 10% in the elderly living in their own home or family's, and mostly remains to diagnose and treat. The aim of this study was to characterize and estimate the prevalence of malnutrition and risk of malnutrition in the elderly, in the first 72 hours of hospital admission. We applied an observational, analytical, cross-sectional and correlacional quantitative type of study and data were collected by interview and observation itself. The study was developed in two lines of research: one focused on the characterization of malnutrition in elderly institutionalized in Portuguese hospitals, in the central and southern mainland and Madeira, in the periods between July - August 2009, April - June 2010, May - July 2011, through the MNA®; and the other: a detailed nutritional assessment, conducted in Hospital Lisbon Center - Hospital de Santa Marta, EPE, between early January 2009 and late January 2010, and the sample was recruited from among the elderly of both sexes at hospital admission. We studied intensively sociodemographic, health and nutritional assessment done extensive. Nutritional evaluation consisted of harvesting different parameters: hematological, biochemical and anthropometric (body mass index (BMI), triceps skinfold (TSF), sub-scapular skinfold (SSF), arm circumference (AC), arm muscle area (AMA), geminal perimeter (GP), analysis of body composition (Fat Mass (FM), Fat Free Mass (FFM)), characterization of a daily food type and Mini Nutritional Assessment Long Form® questionnaire - MNA LF®. Form the data obtained in Portuguese hospitals, it is noteworthy that of the 402 patients included, 53% were male, had a mean age of 75,8 + 6,52 (65 - 100) and, according to the MNA®, 57,5% were malnourished or at risk of malnutrition. In the sample, from the sociodemographic data obtained, we saw that 50% of patients were male, the average age was around 75,5 + 7,22 years (65-100), 55% were from Lisbon and 80 %lived in Lisbon, 38% had no formal education and 43% did so only until the 4th grade. Regarding health data, the majorities of patients were admitted through the ER of Hospital Lisbon Center - S. José Hospital - and were admitted to the Medicine Unit (38%) and to the Cardiology Unit (30%), by medical pathology (38%) and circulatory system disease (56%). In regard to lifestyle, and considering mobility, one third of patients were bedridden and the rest were ambulating or had a normal mobility. 74% had no regular ethanol habits, 19% had a high intake (> 30 g ethanol / day); 95% of the patients had no smoking habits. Regarding nutritional assessment, the mean values for laboratory parameters proved inferior to standard values for age and sex and were lower in females. In anthropometric assessment these were the findings: BMI calculation showed to be scarcely sensitive in the identification of undernourished patients; the TSF and SSF revealed values of fat mass within the normal range; in AC, 88% did not have an indicative value of malnutrition and 8% were malnourished; in AMA, 97% of men and 95% women had a deficit of lean mass and in GP, 18% had a value of less than the 31cm malnutrition descriptor. In body composition analysis found that both sexes showed a percentage of FM ranked too high and this was higher in women compared to men. By analyzing the nutritional intake was found that this was less than the Dietary Reference Intakes (DRIs) for water intake (p=0,00), energy (p=0,00), protein (p=0,00), lipid (p=0,01), MUFA (p=0,00), PUFA (p=0,00), carbohydrates (p=0,00), fiber (p=0,02), potassium (p=0,00), calcium (p=0,00), magnesium (p=0,00), phosphorus (p=0,00), zinc (p=0,00), vitamin D (p=0,00), vitamin E (p=0,00) and folate (p=0,00). Regarding MNA®, its application identified 62% of cases of nutritional risk or malnutrition already installed. MNA® values indicative of malnutrition or risk were associated with lower levels of education (r=0,32; p=0,00). There was a correlation between the MNA ® and TSF (r =0,30, p = 0,00), SFF (r = 0,19, p = 0,03) and GP (r=0,27, p = 0,00). In the analysis of the sample by gender and age group, the highlight was in older women who had BMI status indicator of malnutrition risk (BMI <23,5 + 2,9 (r=0,42;p=0,02)) and mean values of AC 25,6 +3,84cm (r=0,42; p=0,01), in situations characterized by low mobility MNA®. Older men showed a correlation between the MNA® and SFF (r = 0,41; p = 0,02), AMA (r = 0,57; p = 0,00) and GP (r=0,55;p=0,00), and in older women only TSF showed a correlation(r = 0,39; p =0,02). Multivariate analysis of the MNA® by gender and age group, revealed that they are independent. The men had MNA® mean superior to women, and as the age increases, the values of MNA® in both sexes declined, being indicative of risk of malnutrition. We believe that, given the nature and objectives of the present study, it allowed us to characterize and estimate the prevalence of risk of malnutrition and malnutrition in older people during the first 72 hours of hospital admission. The results indicate a high prevalence and point to the need for protocol procedures of nutritional assessment and intervention in the elderly population at hospital admission. For this purpose the application of MNA® has proved its applicability, as well as measuring and calculating AMA, which may be applied in early stages thus contributing to enhance health state improvements and to shorten the time of hospitalization, particularly in elderly people. In relation to dietary pattern, this study contributed to call of attention from health professionals that the elderly may have nutritional deficiencies on admission, and that these are not properly marked and addressed tend to worsen during hospitalization may contribute to increased morbidity.

Fermentation of fruit juices by the osmotolerant yeast Candida magnoliae

This study focuses on the assessment of the fermentation conditions required to modulate the metabolic flux in the osmotolerant yeast Candida magnoliae and evaluate its potential to produce low-alcoholic and low-caloric fermented beverages. For that purpose, two strains, PYCC 2903 and PYCC 3191, were used and fermentation conditions as oxygenation, sugar concentration and the ratio of glucose to fructose were studied using synthetic culture media. Candida magnoliae PYCC 2903 was subsequently used to ferment real industrial fructose-rich substrates such as fruit juices. Sugar consumption profiles for C.magnoliae PYCC 2903 incubated aerobically in the presence of high fructose and glucose concentrations (15%, 10% and 5%) showed a selective utilization of fructose, denoting a preference for this sugar over glucose. The lower ratio between ethanol and sugar alcohols yield was obtained for both strains incubated under oxygen limitation simulating industrial fructose-rich substrates, confirming the ability of this yeast to direct fermentation towards alternative products. Enzymatic assays for hexokinase activity in terms of capacity and affinity for glucose and fructose were performed, aiming to elucidate its contribution to the fructophilic behaviour of this yeast. Enzymatic assays for both strains showed that the Vmax is two to threefold higher for fructose than for glucose but Km is also 10-20-fold higher for this sugar than for glucose. Hence, hexokinase kinetic properties do not explain fructophily in C.magnoliae. This indicates that fructose transport is probably determining in this respect, as observed for other fructophilic yeasts. Fruit juice fermentations with C.magnoliae PYCC 2903 revealed a potential for the production of beverages with interesting sensorial properties. Pear and peach fermentations exhibited the best results with the lowest ratio between ethanol and sugar alcohols yield and the most pleasant organoleptic features.

Development and application of NMR techniques to study water-in-CO2 microemulsions

Self-assembly is a phenomenon that occurs frequently throughout the universe. In this work, two self-assembling systems were studied: the formation of reverse micelles in isooctane and in supercritical CO2 (scCO2), and the formation of gels in organic solvents. The goal was the physicochemical study of these systems and the development of an NMR methodology to study them. In this work, AOT was used as a model molecule both to comprehensively study a widely researched system water/AOT/isooctane at different water concentrations and to assess its aggregation in supercritical carbon dioxide at different pressures. In order to do so an NMR methodology was devised, in which it was possible to accurately determine hydrodynamic radius of the micelle (in agreement with DLS measurements) using diffusion ordered spectroscopy (DOSY), the micellar stability and its dynamics. This was mostly assessed by 1H NMR relaxation studies, which allowed to determine correlation times and size of correlating water molecules, which are in agreement with the size of the shell that interacts with the micellar layer. The encapsulation of differently-sized carbohydrates was also studied and allowed to understand the dynamics and stability of the aggregates in such conditions. A W/CO2 microemulsion was prepared using AOT and water in scCO2, with ethanol as cosurfactant. The behaviour of the components of the system at different pressures was assessed and it is likely that above 130 bar reverse microemulsions were achieved. The homogeneity of the system was also determined by NMR. The formation of the gel network by two small molecular organogelators in toluene-d8 was studied by DOSY. A methodology using One-shot DOSY to perform the spectra was designed and applied with success. This yielded an understanding about the role of the solvent and gelator in the aggregation process, as an estimation of the time of gelation.

Production and characterization of electrospun composite fibers: confinement of thermosensitive microgels

Materials engineering focuses on the assembly of materials´ properties to design new products with the best performance. By using sub-micrometer size materials in the production of composites, it is possible to obtain objects with properties that none of their compounds show individually. Once three-dimensional materials can be easily customized to obtain desired properties, much interest has been paid to nanostructured poly-mers in order to build biocompatible devices. Over the past years, the thermosensitive microgels have become more common in the framework of bio-materials with potential applicability in therapy and/or diagnostics. In addition, high aspect ratio biopolymers fibers have been produced using the cost-effective method called electrospinning. Taking advantage of both microgels and electrospun fibers, surfaces with enhanced functionalities can be obtained and, therefore employed in a wide range of applications. This dissertation reports on the confinement of stimuli-responsive microgels through the colloidal electro-spinning process. The process mainly depends on the composition, properties and patterning of the precur-sor materials within the polymer jet. Microgels as well as the electrospun non-woven mats were investigated to correlate the starting materials with the final morphology of the composite fibers. PNIPAAm and PNIPAAm/Chitosan thermosensitive microgels with different compositions were obtained via surfactant free emulsion polymerization (SFEP) and characterized in terms of chemical structure, morphology, thermal sta-bility, swelling properties and thermosensitivity. Finally, the colloidal electrospinning method was carried out from spinning solutions composed of the stable microgel dispersions (up to a concentration of about 35 wt. % microgels) and a polymer solution of PEO/water/ethanol mixture acting as fiber template solution. The confinement of microgels was confirmed by Scanning Electron Microscopy (SEM). The electrospinning process was statistically analysed providing the optimum set of parameters aimed to minimize the fiber diameter, which give rise to electrospun nanofibers of PNIPAAm microgels/PEO with a mean fiber diameter of 63 ± 25 nm.

The perfect paint in modern art conservation: a comparative study of 21st century vinyl emulsions

Contemporary painting places, and will continue to place, several questions about its meaning, its chemical nature, its durability and the best way to preserve it. This research aims at putting together comprehensive data on vinyl based paints, including their components, their properties, their aging behavior and their response to selected cleaning products. In this project degradation mechanisms of vinyl binders and formulations used in the 20th and 21st century were studied. Stability over time of selected vinyl polymers was assessed through natural indoor and artificially aging. The objective was to enhance knowledge and understanding of vinyl emulsion formulations and their performance over time. Overall conservation state of pictorial layers namely, adhesion, cohesion and discoloration of selected case studies from the Portuguese artist Julião Sarmento (b.1948) was correlated with the observed molecular level changes studied in laboratory experiments. Sarmento’s paintings were chosen due to conservation concerns (discoloration) on some of his works from the 90’s. Besides, research was carried out to start increasing the knowledge of what can be expected of PVAc based paints in terms of response to conservation treatments namely, surface cleaning. Artificial aging showed that the most recent formulations which are based on a poly(vinyl acetate), poly(vinyl chloride) and polyethylene terpolymer are less stable when compared to some homopolymer formulations. From the four pigments studied, titanium dioxide rutile and a carbon based black proved to be stabilizers for both types of polymer. The mixture lithopone plus calcium carbonate has showed to have a photocatalytic effect on the binders. The studied paintings showed to be in an overall good state of conservation except for the paintings created in the 90’s with white glue and a mixture of white lithoponeand calcium carbonate. Discoloration of this white paint seems to be irreversible and ongoing and is still a major concern. The disapearance of the plasticizer was the only change detected. The current works created by Sarmento are expected to be more stable as they were painted using the rutile titanium dioxide. Immersion/cleaning tests showed that vinyl based paints can be susceptible to water and organic solvents like ethanol as some evidences point to the removal/diffusion of additives from the paint. The observations made point to the need to further proceed in this research field.