Systems trained on linguistically annotated data achieve strong performance for many language processing tasks. This encourages the idea that annotations can improve any language processing task if applied in the right way. However, despite widespread acceptance and availability of highly accurate parsing software, it is not clear that ad hoc information retrieval (IR) techniques using annotated documents and requests consistently improve search performance compared to techniques that use no linguistic knowledge. In many cases, retrieval gains made using language processing components, such as part-of-speech tagging and head-dependent relations, are offset by significant negative effects. This results in a minimal positive, or even negative, overall impact for linguistically motivated approaches compared to approaches that do not use any syntactic or domain knowledge. In some cases, it may be that syntax does not reveal anything of practical importance about document relevance. Yet without a convincing explanation for why linguistic annotations fail in IR, the intuitive appeal of search systems that ‘understand’ text can result in the repeated application, and mis-application, of language processing to enhance search performance. This dissertation investigates whether linguistics can improve the selection of query terms by better modelling the alignment process between natural language requests and search queries. It is the most comprehensive work on the utility of linguistic methods in IR to date. Term selection in this work focuses on identification of informative query terms of 1-3 words that both represent the semantics of a request and discriminate between relevant and non-relevant documents. Approaches to word association are discussed with respect to linguistic principles, and evaluated with respect to semantic characterization and discriminative ability. Analysis is organised around three theories of language that emphasize different structures for the identification of terms: phrase structure theory, dependency theory and lexicalism. The structures identified by these theories play distinctive roles in the organisation of language. Evidence is presented regarding the value of different methods of word association based on these structures, and the effect of method and term combinations. Two highly effective, novel methods for the selection of terms from verbose queries are also proposed and evaluated. The first method focuses on the semantic phenomenon of ellipsis with a discriminative filter that leverages diverse text features. The second method exploits a term ranking algorithm, PhRank, that uses no linguistic information and relies on a network model of query context. The latter focuses queries so that 1-5 terms in an unweighted model achieve better retrieval effectiveness than weighted IR models that use up to 30 terms. In addition, unlike models that use a weighted distribution of terms or subqueries, the concise terms identified by PhRank are interpretable by users. Evaluation with newswire and web collections demonstrates that PhRank-based query reformulation significantly improves performance of verbose queries up to 14% compared to highly competitive IR models, and is at least as good for short, keyword queries with the same models. Results illustrate that linguistic processing may help with the selection of word associations but does not necessarily translate into improved IR performance. Statistical methods are necessary to overcome the limits of syntactic parsing and word adjacency measures for ad hoc IR. As a result, probabilistic frameworks that discover, and make use of, many forms of linguistic evidence may deliver small improvements in IR effectiveness, but methods that use simple features can be substantially more efficient and equally, or more, effective. Various explanations for this finding are suggested, including the probabilistic nature of grammatical categories, a lack of homomorphism between syntax and semantics, the impact of lexical relations, variability in collection data, and systemic effects in language systems.

Orientador: Ana Maria Canesqui / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Ciências Médicas / Made available in DSpace on 2018-08-16T19:59:33Z (GMT). No. of bitstreams: 1 DeSordi_GeorgiaSoares_M.pdf: 944066 bytes, checksum: 430549e702dc15a95cf4ee97a486d9cb (MD5) Previous issue date: 2010 / Resumo: O município de Campinas possui uma rede assistencial densa e complexa na área de saúde mental, fruto de investimentos técnico-assistenciais e políticos, ancorados nas reformas psiquiátrica e sanitária. Desde a década de 1990, o município realiza um movimento de desospitalização e desinstitucionalização, visando a criação de uma rede assistencial articulada aos princípios do SUS e substitutiva ao manicômio. Dentro de um panorama histórico ampliado, no período de 2001 a 2004, foram formuladas novas políticas de saúde mental, havendo um incremento considerável na oferta da rede de cuidado. O objetivo desta pesquisa é analisar a formulação da política de saúde mental no período acima citado, enfocando os principais agentes formuladores, assim como reconstruir o contexto das políticas nacionais de saúde mental como substrato à compreensão da política local. A formulação é uma etapa da política e os agentes, gestores da política local, peças fundamentais, na medida em que decidem qual é o modelo político/assistencial a ser implementado no município. Nesta pesquisa, de cunho qualitativo, os dados foram coletados através de entrevistas semi-estruturadas com os agentes decisores da formulação da política, tendo como fonte secundária documentos arquivados no Cedoc - Centro de Documentação, e alguns documentos cedidos pelos agentes decisores. A análise de formulação contribuirá para a compreensão do modelo técnico-assistencial proposto pelos gestores, reconstruindo aspectos históricos inerentes ao processo de formulação/implementação das políticas e sua importância frente aos resultados alcançados pelo modelo assistencial empregado / Abstract: The city of Campinas has a dense and complex public mental health care network as a result of technical-assistance and political investments, anchored in the psychiatric and sanitary reform. Since the decade of 1990 the city has started a movement for the attention of patients outside hospitals and institutions, aiming to create an assistance network articulated to the principles of SUS and replacing the mental hospitals. In a panoramic history context, in the period from 2001 to 2004, new policies for mental health assistance have been formulated, with a considerable increase in the mental health care network. The scope of this research is to analyze the public mental health care policy formulation during this period, focusing on the main actors, as well as to reconstruct the context of national public mental health care policies as a substrate for the comprehension of local policies. Formulation is one of the phases of policy implementation and the actors and administrators of local policies are fundamental parts of this process, since they are the ones who decide which choice of political/assistance model would be implemented in the city. In this research, with a qualitative scope, data was collected through semi-structured interviews with the deciding agents of policies formulation, having as secondary source the documents from City Health Secretary (Secretaria Municipal de Saúde) archives of CEDOC (Center for Documentation of City Health Secretary), as well as some documents provided by the deciding agents. The analysis of the policy formulation may help the comprehension of the mental health care model proposed by the administrators, reviewing historical aspects inherent to the process of formulation/implementation of policies and their importance to the results attained by the implemented assistance model / Mestrado / Saude Coletiva / Mestre em Saude Coletiva

Orientador: Ronei Jesus Poppi / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Quimica / Made available in DSpace on 2018-08-13T00:09:21Z (GMT). No. of bitstreams: 1 Alves_JulioCesarLaurentino_M.pdf: 1206213 bytes, checksum: 33bf289d85e04218ef31b314138a5d44 (MD5) Previous issue date: 2009 / Resumo: Neste trabalho foi desenvolvida uma metodologia para determinação simultânea da mistura farmacêutica de ácido acetilsalicílico (AAS), paracetamol e cafeína, através de fluorescência de excitação-emissão em fase sólida. Esta metodologia é aplicável mesmo na presença de interferentes e com sobreposição espectral dos componentes da mistura, sem prejuízo da boa reprodutibilidade e exatidão, obtendo erros menores que 5%. Para tanto utilizou-se os métodos quimiométricos de calibração de segunda ordem PARAFAC e UPLS, além do método de calibração de primeira ordem PLS2. Vantagens observadas em relação aos métodos de referência como o menor custo, a não necessidade de longo preparo da amostra e análise simples e rápida, além de não haver geração de resíduos, tornam esse método bastante atrativo, permitindo a determinação simultânea dos compostos na mistura estudada. Ainda, a fim de propor uma solução para o problema que foi observado durante o desenvolvimento dos modelos, relativo à mudança da matriz da amostra, utilizou-se transferência de calibração. Nesse estudo, a padronização de espectros através do método da Padronização Direta por Partes (PDS) foi aplicada a modelos UPLS construídos a partir dos espectros de fluorescência de excitação-emissão de amostras de misturas farmacêuticas com excipiente amido/celulose, e amostras utilizando como excipiente a lactose. Excelentes resultados de previsão foram obtidos, com erros relativos abaixo de 5%. Também, comprovou-se que os resultados de previsão obtidos a partir dos dados de transferência de calibração aplicados ao modelo UPLS, não apresentam diferença em relação a previsão de amostras idênticas às usadas no modelo de calibração / Abstract: In this work it was developed a methodology for determination of acetylsalicylic acid (ASA), paracetamol and caffeine in pharmaceutical formulations using solidphase molecular fluorescence and second order calibration methods. This methodology is usefull even in the presence of unknown interferences and with spectral overlap of the components in the mixture. Parallel Factor Analysis (PARAFAC) and Unfolded Partial Least Squares (UPLS) were used for second order calibration models development. Errors below to 5% were obtained for all compounds using an external validation set. Advantages not included in the reference methods such as low cost, no need of sample preparation, simple and fast analysis and no generation of waste, make this method very attractive, allowing to the simultaneous determination of compounds with good reproducibility and accuracy. Also, to propose a solution for the problem observed during the models development, due to change of sample matrix, the transfer of calibration was used. In this study it was used the method of Piecewise Direct Standardization applied to UPLS models constructed from samples of pharmaceutical mixtures of acetilsalicilic acid (ASA), paracetamol and caffeine with starch/cellulose 1:1 as excipient and samples with lactose as excipient, for accomplishment of the calibration transfer procedures. The transfer of calibration through Piecewise Direct Standardization method (PDS) applied to the UPLS calibration models provided excellent prediction results for pharmaceutical mixture in diferent excipient with relative errors less than 5%. From the values of RMSEP obtained for the results of the models of reference and the models of calibration transference, it was evaluated through a F test, that there is no difference, with 99% of significance level, in the results obtained by two models / Mestrado / Quimica Analitica / Mestre em Química

Conselho Nacional de Desenvolvimento CientÃfico e TecnolÃgico / Cetorolaco de trometamina (Cetorolaco) à um antiinflamatÃrio nÃo esteroidal (AINE) usado no tratamento da dor. O uso da formulaÃÃo sublingual tem diversas vantagens, como por exemplo, aumento da biodisponibilidade e fÃcil administraÃÃo, especialmente em pacientes que tem dificuldades para engolir. Assim, o objetivo deste estudo foi avaliar o perfil farmacocinÃtico da formulaÃÃo sublingual do Cetorolaco em voluntÃrios brasileiros do sexo masculino. Este foi um estudo aberto, nÃo randomizado, 01 perÃodo, 01 tratamento, com dose Ãnica de rÃpida absorÃÃo, em que foi administrado um comprimido de 30 mg de Cetorolaco por via sublingual. Foram incluÃdos apenas voluntÃrios brasileiros sadios. ApÃs uma noite internados, os voluntÃrios receberam uma dose Ãnica da formulaÃÃo sublingual. Amostras de plasma foram obtidas em um perÃodo de 24 horas apÃs a administraÃÃo. A concentraÃÃo plasmÃtica de Cetorolaco foi analisada por Cromatrografia LÃquida de Alta EficiÃncia acoplada a um EspectrÃmetro de Massa para anÃlise dos parÃmetros farmacocinÃticos, incluindo o Cmax, ASC0-24, e ASC0-∞. A tolerabilidade foi avaliada pelo monitoramento dos sinais vitais, resultados das anÃlises laboratoriais, anamnese e exame fÃsico com os voluntÃrios. 14 voluntÃrios do sexo masculino foram incluÃdos e completaram o estudo. Os valores dos parÃmetros farmacocinÃticos (mÃdia  desvio padrÃo, exceto mediana para o tempo do Tmax) calculados para a formulaÃÃo foram os seguintes: ASC0-∞ (9682  1908 ng*h/mL); ASC0-24 (9346  1789 ng*h/mL); Cmax (2605  465 ng/mL); Tmax (0,58  0,22 h); t1/2 (5,76  0,69 h) and Ke (0,12  0,02 1/h). A formulaÃÃo do cetorolaco foi bem tolerada na dose administrada e nenhuma reaÃÃo adversa foi observada. A vantagem do uso da formulaÃÃo sublingual de Cetorolaco à sua administraÃÃo prÃtica. Essa pode ser uma escolha para dor moderada e grave, especialmente em pacientes em que a via parenteral à indesejada ou impraticÃvel, ou para aqueles que tenham dificuldade de engolir. Os mÃtodos foram aplicados com sucesso, sendo possÃvel analisar os parÃmetros farmacocinÃticos e avaliar a seguranÃa da formulaÃÃo do comprimido de 30mg do cetorolaco de trometamina administrado por via sublingual em voluntÃrios saudÃveis. / Ketorolac tromethamine (ketorolac) is a non-steroidal anti-inflammatory drug (NSAID) used in the treatment of pain. The use of a sublingual analgesic has several advantages such as increased bioavailability and ease of administration, especially in patients who have difficulty swallowing. Thus, the aim of this study was to evaluate the pharmacokinetic profile of a formulation of ketorolac in Brazilian male volunteers. This was an open, non-randomized, 01 period, 01 treatment, single dose under fasting conditions study where was administered 30 mg sublingual tablet formulation of ketorolac. Healthy Brazilians male were eligible for inclusion. After an overnight fast, subjects received a single sublingual dose of the formulation. Plasma samples were obtained over a 24-hour period after administration. Plasma ketorolac concentrations were analyzed by High Performance Liquid Chromatography coupled Mass Spectrometry for analysis of pharmacokinetic properties, including Cmax, AUC0-24, and AUC0-∞. Tolerability was assessed by vital sign monitoring, laboratory analysis results, anamnesis and physical examination. A total of 14 male subjects were enrolled and completed the study. The pharmacokinetics parameters values (mean  standard deviation, except median for time to Tmax) calculated for formulation were as follows: AUC0-∞ (9682  1908 ng*h/mL); AUC0-24 (9346  1789 ng*h/mL); Cmax (2605  465 ng/mL); Tmax (0,58  0,22 h); t1/2 (5,76  0,69 h) and Ke (0,12  0,02 1/h). The ketorolac formulation was well tolerated at the administered dose and no adverse reactions were observed. The advantage of the use of the sublingual formulation of ketorolac is its practical administration. It can be the treatment of choice for moderate and severe pain, especially in patients where parenteral route is undesirable or impracticable, or those who have difficulty of swallowing. The methods were successfully applied, it is possible to analyze the pharmacokinetic parameters and assess the safety of the tablet formulation of 30 mg of ketorolac tromethamine administered sublingually in healthy volunteers.

Submitted by Biblioteca da Faculdade de Farmácia (bff@ndc.uff.br) on 2017-03-30T17:15:27Z No. of bitstreams: 1 Miura, Daniele Yuri [Dissertação, 2012].pdf: 3475068 bytes, checksum: 14b2a219ff5a9c48bcab7248e3b57e35 (MD5) / Made available in DSpace on 2017-03-30T17:15:27Z (GMT). No. of bitstreams: 1 Miura, Daniele Yuri [Dissertação, 2012].pdf: 3475068 bytes, checksum: 14b2a219ff5a9c48bcab7248e3b57e35 (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro / A papaína é uma enzima proteolítica, extraída do látex da espécie Carica papaya Linne. Ela é utilizada no tratamento tópico de feridas como agente desbridante, podendo ser aplicada em concentrações de 2 a 10%, dependendo da fase do processo de cicatrização. Porém, sua baixa estabilidade é um fator limitante para aplicação em formulações. O presente estudo analisou a influência dos adjuvantes técnicos EDTA dissódico, cloridrato de cisteína e propilenoglicol no aumento da atividade proteolítica de géis de Carbopol® 940 contendo papaína a 2 e a 4% (p/p). No estudo de formulação dos géis, as formulações foram definidas por um desenho fatorial 33 e o efeito dos adjuvantes foi avaliado através de um estudo de superfície de resposta. Para cada concentração de papaína, foram selecionadas duas formulações para o estudo de estabilidade acelerada: a formulação com melhor resultado de atividade proteolítica e aquela sem adição de adjuvantes. O estudo de estabilidade acelerada foi realizado por sessenta dias a 5ºC ± 2ºC, 26ºC ± 2ºC e 45ºC ± 2ºC. As quatro formulações apresentaram grande perda de atividade enzimática durante o tempo do estudo, mesmo aquelas armazenadas a 5ºC ± 2ºC. Contudo, o armazenamento em temperaturas mais altas intensificou a perda de atividade. Para os géis de papaína a 2%, o uso de adjuvantes não foi eficiente na manutenção da atividade proteolítica. Já para os géis de papaína a 4%, houve um intenso aumento na atividade com o uso de adjuvantes, porém este efeito não foi mantido durante o tempo. Apesar destes resultados, o estudo indicou a possibilidade de aprimoramento das formulações de gel de papaína com o uso de adjuvantes técnicos, podendo gerar uma melhora no desempenho da atividade proteolítica dos produtos / Papain is a proteolytic enzyme extracted from latex of Carica papaya Linne. It is used in the topical wounds treatment as a debriding agent and can be applied at concentrations of 2-10%, depending on the phase of the healing process. However, its low stability is a limiting factor for use in formulations. The present study examined the influence of technical adjuvants disodium EDTA, cysteine hydrochloride and propylene glycol on increasing of proteolytic activity of Carbopol® 940 gels containing 2 and 4% (w/w) of papain. In the gels preformulation study, the formulations were defined by a 33 factorial design and the effect of adjuvants was evaluated using a response surface study. For each concentration of papain, two formulations were selected for the accelerated stability study: a formulation with best result of proteolytic activity and that without adjuvants. The accelerated stability study was carried out for sixty days at 5°C ± 2°C, 26ºC ± 2°C and 45ºC ± 2°C. The four formulations showed a large loss of enzyme activity during the time of the study, even those stored at 5°C ± 2°C. However, storage at higher temperatures increased the loss of activity. For gels with 2% papain, the use of adjuvants was not efficient in maintaining the proteolytic activity. As for the gels with 4% papain, there was an intense increase in activity with the use of adjuvants, but this effect was not maintained over time. Despite these results, the study indicated the possibility of improvement of papain gel formulations with the use of technical adjuvants, wich can provide a better performance of the proteolytic activity of the products.

[pt] A formulação híbrida de raios e modos é apresentada e aplicada ao cálculo da intensidade de campo em dutos troposféricos. Essa formulação foi desenvolvida para suprir deficiência das técnicas tradicionais, que utilizam apenas modos ou apenas raios nos cálculos de campo. Partindo de um modelo cilíndrico da troposfera, é mostrada a equivalência entre este problema, e o de Terra plana, com variação linear da refratividade. Isso permite a aplicação imediata das formulações geradas para uma geometria cilíndrica a alguns tipos de perfis reais obtidos de medidas. Apresenta-se, então, comparações entre os resultados teóricos e as medidas. / [en] The hybrid formulation of rays and modes is presented and applied to evaluate field intesities in tropospheric ducts. This formulation was developed to evercome some inefficiencies in the traditional techniques that used only modes or only rays in field calculations. Starting from a cylindrical model of the troposphere, the equivalence between this situation and a flat-earth model with linear M-profile is then shown. This allows the direct application of the formulations developed for the cylindrical geometry to some measured profiles. Finally, some comparisons are shown between theoretical and experimental results.

When an intense laser beam interacts with matter, the Kerr nonlinearity results in self-focusing. Above the critical intensity, self-focusing dominates pulse spreading through diffraction leading to continuous pulse narrowing and thus an increase of the laser peak intensity. Collapse is prevented through the fact that peak intensities ultimately reach a level where ionization occurs. The profile of ionized electrons represents a negative lens which balances Kerr nonlinear self-focusing and causes the formation of stable filaments. From filaments radiation is emitted in a cone around the filament which has been termed conical emission. Filament formation happens at non-perturbative intensities where the formalism of perturbative nonlinear optics loses its validity. This opens the question of how the Kerr nonlinearity behaves in the non-perturbative limit and how large the Kerr nonlinear coefficient is. The expression for the Kerr nonlinearity is derived by perturbation theory; the validity of this expression in the non-perturbative limit is questionable. Further, experimentally the Kerr nonlinear index is extracted from measurements of the self-focusing distance as a function of beam intensity which is called the Z-scan method. This method fails at non-perturbative intensities due to the presence of the negative lens coming from the ionized electrons. The effects of the positive focusing and negative self-defocusing lens cannot be separated by the Z-scan method. As a result, not much is known about the Kerr nonlinearity in the regime of non-perturbative nonlinear optics. The purpose of this thesis is twofold. First, recently it has been discovered that conical emission can be utilized as a broadband and very efficient amplification mechanism in the far infrared. The process has been dubbed Kerr instability amplification. The difference between conical emission and Kerr instability amplification is that they take place in two different regimes of the nonlinear interaction. Whereas conical emission grows out of noise and therewith only takes place once the pump pulse has been substantially restructured due to filamentation, Kerr instability amplification is seeded with a second pulse and therewith occurs long before filamentation happens. The theory developed for Kerr instability amplification has been developed based on a stability analysis of the scalar wave equation. This analysis has shown that with pump lasers in the 1-2 μm range amplification of infrared radiation up to the 10’s of μm can be achieved. For amplification over such a wide range it is not adhoc clear to which exent vectorial wave effects can be neglected. The first part of the thesis closes this gap by developing the vectorial theory of vector instability amplification. The second part uses the results derived for Kerr instability amplification to answer the question of how to measure the Kerr nonlinear index in the nonperturbative laser intensity limit. The idea rests on the fact that Kerr instability amplification is maximum for a specific angle between pump and seed beam which varies as a function of laser pump intensity. A relation is derived that connects this angle with the Kerr nonlinear refractive index. As a result, from the maximum angle measured as a function of pump intensity, both magnitude and functional form of the Kerr nonlinear index as a function of laser intensity can be determined.

Masters in Public Administration - MPA / This study focuses on the contradictory ways officials frame the problem of providing water services in Kampala with specific emphasis on the debates about framing of key factors that constrain delivery. Perceptions about and framing of policy issues implicitly suggest the causes of problems. Policy scholars acknowledge that formulating policy is “messy” and that it is about values, and not as “evidence-based” as is often assumed. Using a social constructionist perspective that stresses that perceptions tell one how people define situations, this study demonstrates the often contradictory ways that state officials see public problems. Drawing on debates about rational and muddling-through approaches, the study uses thirty structured questionnaires and several semi-structured in-depth interviews to explore the narratives of government officials regarding policy formulation in water delivery. The thesis found that almost all officials see rapid growth in population density in informal areas coupled with unplanned settlements as the single biggest combined issue inhibiting progress. Officials favoured market-related solutions even though their diagnosis of the problem suggested more concerted and comprehensive public planning of urban development.

Magister Pharmaceuticae - MPharm / Co-crystals are a solid phase phenomena that could enhance the physicochemical properties of an active pharmaceutical ingredient. A co-crystal has never been incorporated into a liquid dosage form with the assurance of maintaining its co-crystal state until absorption under defined conditions. This study aims to develop a liquid formulation with a nevirapine co-crystal. A protocol was developed to investigate all the five co-formers that were used to make the nevirapine co-crystals to-date. The most appropriate co-former was selected for a liquid dosage form to study the integrity and the scaling up of the co-crystal in a suspension formulation. Co-formers used were viz. saccharin, glutaric acid, salicylic acid, rac-tartaric acid and maleic acid. These were characterized according to their physical, chemical, pharmacological and pharmaceutical properties. A grading scale was used to select the most appropriate co-former for a suspension formulation. Comparatively, saccharin produced the best combination of physical, chemical, pharmacological and pharmaceutical properties, especially with regard to the particle size and the specific gravity which proved to be very useful as optimal criteria for suspension formulation. Upon selection of the ideal co-former, scale-up of the nevirapine saccharin co-crystal was performed from a small scale of 350 mg to a large scale of 5 g. Nevirapine-saccharin (NVSC) co-crystals were prepared utilizing the slow evaporation technique, using methanol as the solvent and the percentage yield of the co-crystals were > 80 %. The identity of co-crystals was confirmed using hot stage microscopy (HSM), differential scanning calorimetry (DSC), fourier transform infra- red (FTIR) and thermogravimetric analysis (TGA). Three co-crystal suspension formulations were prepared using the excipients identified in the branded, Viramune® suspension, with each formulation containing viscosity enhancers such as aerosil 200, carbopol 971G and carbopol 974P. To ascertain the co- crystal integrity in the suspension, it was filtered and the filtrate was identified with DSC and FTIR while the filtered solution was identified with ultraviolet spectroscopy (UV). The co-crystal suspension formulation with optimal pH, viscosity and assurance of co-crystal integrity was the carbopol 974P formulation. The UV and DSC of the filtrate of the suspension revealed that the co-crystal had not separated into its individual components and remained intact while in suspension form irrespective of the excipients added. This formulation proceeded to the quality control stage. It was assessed for its pH, viscosity and dissolution according to the USP 32 standards and compared to the branded nevirapine suspension, Viramune ®, presently on the market. The suspension was characterized for particle size, zeta potential and polydispersity index. The dissolution results assayed by High Performance Liquid Chromatography (HPLC) revealed a drug release of 86 % in the Viramune® suspension while the NVSC co- crystal suspension achieved a drug release of 94% within 30 minutes of dissolution. / National Research Foundation (NRF)

Cette thèse concerne l’étude de l’influence de la formulation sur le comportement rhéologique des émulsions inverses très concentrées. Nous avons utilisé une démarche originale consistant à positionner le module élastique (G’) de ces émulions sur une carte bidimensionnelle formulation-composition, afin d’étudier, en particulier, l’effet de la formulation dite optimale. Deux effets nets sont liés à celle-ci: les émulsions préparées à HLB=10, c’est-à-dire à proximité de la valeur optimale (HLB=10,5), présentent des valeurs de G’ nettement plus faibles que les autres formulations, tandis qu’à une distance particulière (HLB=7,7), les valeurs de G’ sont toujours plus élevées comparativement aux autres formulations. Cette valeur particulière de HLB correspond aussi à une stabilité maximale de la dispersion et à une taille moyenne des gouttes minimale. Les émulsions très concentrées étant généralement opaques et fragiles, l’analyse du transport stationnaire incohérent de lumière polarisée, qui ne nécessite pas de dilution, a été utilisée comme méthode alternative aux techniques classiques pour déterminer l’épaisseur moyenne du film interfacial (h). La taille moyenne des gouttes (R) a pu ensuite être déterminée à partir d’un modèle géométrique qui relie R à l’épaisseur moyenne du film interfacial (h), à la fraction volumique de phase dispersée (f) et à un facteur géométrique moyen (f9) tenant compte du caractère polyédrique des gouttes. Ainsi, une expression modifiée du modèle proposé par Mougel et al. (2006) a été établie pour modéliser le module élastique (G’) en le reliant à la taille moyenne de gouttes (R), la tension interfaciale (s) et la fraction volumique de phase dispersée (f) / This thesis deals with the study of the influence of formulation on the rheological behavior of highly concentrated reverse emulsions. We used an original approach consisting in representing the elastic modulus (G’) of these emulsions on a formulation-composition bidimensional map to study, in particular, the effect of the so-called optimum formulation. Two major effects are observed in relation to this one: emulsions prepared at HLB=10, i.e. at the proximity of optimum value (HLB=10.5) present values of G’ remarkably lower than those at other formulations, while at a certain distance (HLB=7.7) the values of G’ are higher in comparison to others formulations. This last particular HLB matches with the occurrence of a maximum stability and a minimum drop size of dispersion. Highly concentrated emulsions being generally opaque and fragile, the analysis of incoherent polarized steady light transport, that does not imply any dilution, has been used as an alternative method to classical techniques to determine the average film thickness (h). The average drop size (R) has then to be determined from a geometrical model linking it to the average film thickness (h), to the dispersed phase volume fraction (f) and to a geometrical factor (f9) that takes into account the polyhedral shape of the drops. Hence, a modified expression of the model proposed by Mougel et al. (2006) has been established to model the elastic modulus (G’) by relating it to the average drop size (R), the interfacial tension (s) and the dispersed phase volume fraction (f)