[ES] El proyecto ESTiMatE nace con el objetivo de mejorar la precisión de las predicciones de hollín para la reducción de emisiones y una mayor sostenibilidad ambiental en la aviación. Para ello es crucial comprender a fondo fenómenos que influyen en la formación de hollín, incluyendo atomización del combustible, mezcla, combustión y formación de emisiones. La presente tesis, como parte del proyecto, se centra en el primer paso para caracterizar las spray flames, el proceso de atomización del líquido. Existen modelos empíricos para ello, pero el objetivo de este trabajo es desarrollar un nuevo modelo fenomenológico a través de simulaciones de alta fidelidad. Se propone acoplar este modelo a los códigos PRECISE-UNS y Alya para inyectar gotas en simulaciones Lagrangianas, reproduciendo de forma efectiva las características clave de atomización primaria en la formación del spray. Realizar estudios sobre la configuración anular de atomizadores prefilming airblast es un gran desafío, por lo que los esfuerzos se han enfocado en la configuración plana como alternativa que proporciona información valiosa. El KIT ha hecho contribuciones notables en este sentido al desarrollar un banco de pruebas de configuración plana, generando una base de datos extensa con resultados que abarcan diversos fluidos de trabajo y condiciones de operación para validar los cálculos. Esta tesis investiga la atomización primaria en estos dispositivos con un enfoque computacional a través de simulaciones eDNS con el código PARIS Simulator. Se utilizan simulaciones VOF-DNS del borde del atomizador para obtener información, y simulaciones LES monofásicas precursoras que permiten tener en cuenta la turbulencia del gas en la entrada. No obstante, se han propuesto dos metodologías derivadas a partir de eDNS, extendiéndola con simulaciones LES bifásicas adicionales. La primera de ellas impone un espesor y velocidad constantes para la película líquida en la entrada, pero sus valores medios se obtienen procesando las nuevas simulaciones LES bifásicas. Esta metodología ha sido validada para una condición de operación de referencia, representativa del reencendido en altitud. La segunda, introduce un espesor y velocidad variables de la película líquida, constituyendo una condición de contorno más realista. Se han realizado dos estudios diferentes. Por un lado, utilizando la primera metodología, se realizó un estudio paramétrico para investigar el impacto de la velocidad media del gas y las propiedades del fluido en la formación de estructuras líquidas. Al variar estos parámetros se observan diferencias en los mecanismos de ruptura basadas en M como en la literatura. La base para un modelo fenomenológico se ha establecido utilizando estos resultados y puede extenderse con más condiciones de operación siguiendo la metodología descrita. Por otra parte, se ha realizado una comparación entre ambas metodologías para considerar la influencia de la historia de la película de combustible en la atomización primaria. En este estudio, al mejorar las condiciones de contorno a la entrada, se ha reproducido con éxito la fenomenología presente en los experimentos, al contrario que con la eDNS estándar. Se ha propuesto una nueva técnica de postprocesado para caracterizar las estructuras líquidas. Los métodos habituales detectan gotas en un dominio 3D, pero solamente detectan ligamentos en una proyección 2D utilizando optimización de funciones algebraicas. Sin embargo, este nuevo método es capaz de encontrar gotas y ligamentos en 3D mediante OpenCV. También ha sido validado con resultados de la literatura. Con todo, la tesis ha sentado las bases para el desarrollo de modelos específicos de atomización airblast mediante simulaciones de alta fidelidad, contribuyendo a una mejor predicción de los fenómenos de inyección, atomización, evaporación, mezcla y combustión de aeromotores que permita realizar un diseño más eficiente de éstos hacia una movilidad aérea sostenible. / [CA] El projecte ESTiMatE naix amb l'objectiu de millorar la precisió de les prediccions de sutge per a la reducció d'emissions i una major sostenibilitat ambiental en l'aviació. Per això, és crucial comprendre a fons fenòmens que influeixen en la formació de sutge, incloent atomització del combustible, mescla, combustió i formació d'emissions. La present tesi, com a part del projecte, se centra en el primer pas per caracteritzar les spray flames, el procés d'atomització del líquid. Existeixen alguns models empírics per això, però l'objectiu d'aquest treball és desenvolupar un nou model fenomenològic a través de simulacions d'alta fildelitat. Es proposa acoblar aquest model als codis PRECISE-UNS i Alya per injectar gotes en simulacions Lagrangianes, reproduint de manera efectiva les característiques clau d'atomització primària en la formació de l'spray. Realitzar estudis sobre la configuració anul·lar d'atomitzadors prefilming airblast és un gran desafiament, per la qual cosa els investigadors han enfocat els seus esforços en la configuració plana com alternativa que proporciona informació valuosa. El KIT ha fet contribucions notables en aquest sentit en desenvolupar un banc de proves de configuració plana, generant una base de dades extensa amb resultats que abasten diversos fluids de treball i condicions d'operació per a validar els càlculs. Esta tesi investiga l'atomització primària en aquests dispositius amb un enfocament computacional a través de simulacions eDNS emprant el codi PARIS Simulator. S'utilitzen simulacions VOF-DNS de la vora de l'atomitzador per a obtindre informació, i simulacions LES monofàsiques precursores que permeten tindre en compte la turbulència del gas a l'entrada. No obstant això, s'han proposat dues noves metodologies derivades a partir de eDNS, estenent-la amb simulacions LES bifàsiques addicionals. La primera d'elles imposa una grossària i velocitat constants per a la pel·lícula líquida en l'entrada, però els seus valors mitjans s'obtenen processant les noves simulacions LES bifàsiques. Aquesta metodologia ha sigut validada per a una condició d'operació de referència, representativa de la reencesa en altitud. La segona, introduïx una grossària i velocitat variables de la pel·lícula líquida, constituint una condició de contorn més realista. S'han realitzat dos estudis diferents. D'una banda, utilitzant la primera metodologia, es va realitzar un estudi paramètric per a investigar l'impacte de la velocitat mitjana del gas i les propietats del fluid en la formació d'estructures líquides. En variar estos paràmetres s'observen diferènciesen els mecanismes de ruptura basades en M com a la literatura. La base per a un model fenomenològic s'ha establit utilitzant aquests resultats i pot estendre's amb més condicions d'operació seguint la metodologia descrita. D'altra banda, s'ha realitzat en paral·lel una comparació entre totes dues metodologies per a considerar la influència de la història de la pel·lícula de combustible en l'atomització primària. En aquest estudi, en millorar les condicionsde contorn a l'entrada, s'ha reproduït amb èxit la fenomenologia present en els experiments, al contrari que amb la eDNS estàndard. S'ha proposat una nova tècnica de postprocessament per a caracteritzar les estructures líquides. Els mètodes habituals detecten gotes en un domini 3D, però solament detecten lligaments en una projecció 2D utilitzant optimització de funcions algebraiques. No obstant això, aquest nou mètode és capaç de trobar gotes i lligaments en 3D mitjançant la llibrería OpenCV. També ha sigut validat amb resultats de la literatura. Amb tot, la tesi ha establit les bases per al desenvolupament de models específics d'atomització airblast mitjançant simulacions d'alta fidelitat, contribuïnt a una millor predicció dels fenòmens d'injecció, atomització, evaporació, mescla i combustió dels aeromotors que permeta realitzar un disseny més eficient dels mateixos cap a una mobilitat aèria sostenible. / [EN] The project ESTiMatE emerges aiming to enhance the precision of soot predictions leading to reduced emissions and improved environmental sustainability in aviation. To achieve this, it is crucial to thoroughly understand various phenomena that influence soot formation, including fuel atomization, mixing, combustion, and subsequent emissions formation. The present thesis, as part of the project, focuses in the first step to characterize spray flames, the liquid atomization process. There exist some empirical models to characterize atomization, but the objective of this specific work package is to develop a new phenomenological model through high-fidelity simulations instead of empirical results. This model is proposed to be coupled to the PRECISE-UNS and Alya codes to inject droplets in Lagrangian simulations, effectively reproducing the key features of the primary atomization process on the spray formation. Conducting fundamental studies on the annular configuration of prefilming airblast atomizers is challenging, so researchers have directed their attention towards planar configurations as an alternative that can provide valuable insights. The KIT has made notable contributions in this regard by developing a dedicated planar test rig, generating a substantial database of results encompassing various working fluids and operating conditions for validation. This thesis investigates primary atomization in these devices with a computational approach using eDNS within the PARIS Simulator code. VOF-DNS simulations of the atomizing Edge are performed to obtain useful data, and precursor one-phase LES simulations in order to account for gas inflow turbulence at the DNS inlet. Nevertheless, two methodologies have been derived from eDNS, extending it with additional precursor two-phase LES simulations. The first one, imposes constant liquid film thickness and velocity at the DNS inlet, but their mean values are obtained processing the new two-phase LES simulations. It has been validated for a reference operating condition, representative of altitude relight. The second one, introduces variable liquid film thickness and velocity instead, constituting a more realistic boundary condition for the DNS. Two different studies have been conducted in this regard. On the one hand, using the former methodology, a parametric study was performed to investigate the impact of mean gas velocity and fluid properties on the formation of liquid structures. By varying these parameters differences on the breakup mechanisms are observed based on M as found in the literature. The basis for a phenomenological model has been established using these results, and it can be extended with more operating conditions following the described methodology. On the other hand, a comparison between both methodologies has been carried out in parallel, in order to account for the influence of film history on the primary atomization process. In this study, by improving the inlet boundary conditions, the phenomenology observed in the experiments has been succesfully reproduced, unlike with the standard eDNS. A novel post-processing technique for characterizing liquid structures has been proposed. Usual post-processing methods detect droplets in a 3D domain, but only detect ligaments in a top view 2D projection of it using algebraic functions optimization. However, this new method is able to find both droplets and ligaments as well as its 3D properties using the Open Computer Vision Library (OpenCV) instead. It has been also validated with the results in the literature. Overall, the thesis has laid the foundations for the development of specific airblast atomization models through the use of high-fidelity simulations,contributing to a better prediction of the injection, atomization, evaporation, mixing and combustion phenomena in aero engines, enabling a more efficient design towards sustainable air mobility. / Moreno Montagud, C. (2024). Computational Modelling of the Primary Atomization in Aero Engine Airblast Atomizers through a DNS Approach [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/211359

Um estudo sistemático a respeito da utilização de padrão interno em determinações multielementares por espectrometria de absorção atômica (ETAAS) foi desenvolvido. O objetivo principal do presente trabalho foi verificar a possibilidade de melhorar a precisão e a exatidão dos resultados analíticos, que são obtidos na análise de fluidos biológicos. O pré-tratamento dessas amostras foi simplificado e reduzido a uma única etapa de diluição com surfactante (Triton X-100) e ácido (HNO3). Conseqüentemente, a complexidade da solução diluída de amostra, a ser introduzida no tubo de grafite, apresenta uma elevada quantidade de concomitantes que podem provocar interferências químicas. A seleção preliminar dos elementos a serem testados como padrão interno considerou a semelhança de parâmetros físico-químicos relacionados com o processo de atomização. Desta forma, Ag, Bi, In e Tl foram testados como padrão interno para a determinação simultânea de Cd/Pb em sangue e urina, enquanto Bi, Ge, In, Sb, Sn e Te foram os elementos selecionados para a determinação de Mn/Ni/Se em soro sangüíneo. A melhoria da qualidade dos resultados analíticos obtidos na determinação simultânea de Cd e Pb em sangue foi observada quando Ag foi utilizada como padrão interno, na presença de NH4H2PO4 como modificador químico. Verificou-se uma melhoria na exatidão dos resultados obtidos para Cd e Pb, após a correção com padrão interno. Por outro lado, os resultados obtidos na análise de urina não foram corrigidos por nenhum dos elementos testados. Os melhores resultados para a determinação simultânea de Mn, Ni e Se foram obtidos com a utilização de Bi, Sn e Te como padrão interno. Entretanto, verificou-se que a correção de todos os resultados não seria viável com o uso de um único padrão interno. O melhor desempenho nos testes realizados na presença de soro sangüíneo foi obtido com Bi, que melhorou discretamente a precisão dos resultados obtidos para Se. Desta forma, a padronização interna visando a determinação simultânea de Mn, Ni e Se não foi eficiente. A padronização interna em ETAAS, com a finalidade de melhorar a precisão e a exatidão dos resultados analíticos, é uma estratégia tão complexa, quanto os efeitos interferentes que se pretende corrigir: são necessários mais estudos para compreender melhor como a utilização de uma condição de compromisso afeta os processos de atomização, bem como mais informações a respeito das interferências físicas e químicas causadas por amostras complexas, analisadas por ETAAS após uma simples etapa de diluição. Deve-se considerar com especial atenção o modificador químico e as temperaturas das etapas de pirólise e de atomização empregadas, que são parâmetros críticos para o desempenho de um elemento como padrão interno. / A systematic study involving the use of internal standard for multielement determinations by electrothermal atomic absorption spectrometry was developed. The main objective of this work was evaluate the possibility of improving precision and accuracy of the analytical results for biological fluids. The sample pre-treatment was reduced to a single dilution step with surfactant (Triton X-100) and acid (HNO3), increasing the amount of concomitant introduced into the atomizer. The preliminary selection of the elements to be tested as internal standard considered the resemblance of physico-chemical parameters related with the atomization process. Thus, Ag, Bi, In and Tl were tested as internal standard for the simultaneous determination of Cd/Pb in blood and urine, and Bi, Ge, In, Sb, Sn and Te were the selected elements for the determination of Mn/Ni/Se in blood serum. The correction of the results obtained for the simultaneous determination of Cd and Pb in blood was achieved when Ag was used as internal standard, in presence of NH4H2PO4 as chemical modifier. An improvement for the accuracy of the results was observed for both analytes after their correction with the internal standard. On the other hand, the results obtained for the urine analysis were not corrected by using the tested elements. The best results for the simultaneous determination of Mn, Ni and Se were observed when Bi, Sn and Te were used as internal standard. However, the correction for the results for all analytes was not possible by using only one internal standard. The best performance in presence of the serum was obtained for Bi, which improves slightly the precision for the Se results. Thus, the internal standardization for the simultaneous determination of Mn, Ni and Se was not efficient. The internal standardization in ETAAS, aiming the improvement of precision and accuracy of the analytical results, is a strategy as complex as the interference effects to be corrected: more studies are required in order to better understand how the adoption of a compromised condition disturbs the atomization processes, as well as to get more information about the physical and chemical interference caused by complex samples, analyzed by ETAAS after a single dilution step. The chemical modifier and the selected temperatures for the pyrolysis and atomization steps are critical parameters for the performace of an internal standard and they should be carefully considered.

Les verres métalliques ont commencé à être produit dans les années 1960 et sous forme massive dans les années 1980. De nombreuses études se sont intéressées à ces matériaux sous leur forme amorphe et ont conclu qu’ils avaient une forte résistance mécanique mais présentaient un comportement très fragile. Dans le cadre du projet EDDAM débuté en 2011, ces matériaux ont été introduits sous forme de petites sphères dans une matrice d’aluminium. Le premier objectif de notre étude est de voir si le verre métallique sous cette forme permet de le rendre peu fragile. Le second objectif est de trouver une alternative aux renforts céramique dans les composites à matrice métallique qui présentent une faible cohésion à l’interface matrice/inclusion. Dans le but de caractériser l’endommagement dans des nouveaux composites amorphe-cristallins métalliques, la tomographie aux rayons X a été utilisée. Cette technique permet de caractériser de manière non destructive l’endommagement des matériaux et de le visualiser en 3D. Cela apporte une contribution à l’étude des matériaux composites par rapport aux techniques classiques utilisées. L’objectif général de cette thèse a été d’étudier l’endommagement en termes d’amorçage, de croissance et de coales- cence des matériaux composites amorphe-cristallins métallique par tomographie aux rayons X lors d’essais de traction monotone in situ. Les matériaux sélectionnés sont constitués d’une matrice aluminium ("molle" de type 1070A ou "dure" de type 5083) et de renforts en verre métallique Zr57Cu20Al10Ni8Ti5 de taille peu dispersée et répartis de manière homogène, avec différentes fractions volumiques (1%, 4% et 10%). Les matériaux composites ont été élaborés par la voie de la métallurgie des poudres au Spark Plasma Sintering (SPS) suivi d’une étape d’extrusion à chaud. Une attention particulière a été portée sur la caractérisation microstructurale des constituants de base. L’analyse qualitative a permis de comparer l’ensemble des composites fabriqués au SPS et ceux extrudés à chaud après SPS. Les différents modes d’amorçage de l’endommagement ont été observés ainsi que la croissance et la coa- lescence amenant la rupture des composites. L’analyse quantitative a été essentiellement consacrée au premier stade de l’endommagement. La croissance et la coalescence étant très rapide, il a été difficile de les suivre lors des essais interrompus. La modélisation d’un composite amorphe-cristallin métallique à matrice molle a été introduite dans le but de reproduire l’endommagement observé lors des analyses expérimentales. Cette première approche nécessite d’être approfondie dans le but de prédire, compte tenu des propriétés mécaniques des différentes phases et de la fraction volumique des renforts, le mode d’endommagement préférentiel apparaissant dans les composites étudiés. Elle montre cependant les prémices d’une modélisation innovante basée sur la microstructure expérimentale. / Metallic glasses have been produced in the 1960s and bulk metallic glasses in the 1980s. Many studies, focused on these materials in their amorphous state, concluded that they had high mechanical strength but shown low ductility. As part of EDDAM project that started in 2011, these materials were introduced as small particles in an aluminum matrix. The first objective of this study is to see if the metallic glass is less brittle in this form. The second objective is to find an alternative of ceramic reinforcements in metal matrix composites. These materials have low cohesion at the matrix/inclusion interface. In order to characterize the damage in new amorphous-crystalline composite, X-ray tomography was used. This allows to characterize damage in materials and to obtain a 3D viewing. The main objective of this thesis was to study damage (nucleation, growth and coalescence) in composite materials using X-ray tomography during tensile tests. Selected materials are constituted of an aluminum matrix and small metallic glass reinforcements (Zr57Cu20Al_10Ni8Ti5). Composites with different volume fractions (from 1vol.% to 10vol.%) were prepared by Spark Plasma Sintering (SPS) and hot extrusion. A particular attention was paid to the microstructural characterization of the basic constituents. Qualitative analysis was used to compare SPS composites with SPS plus hot extrusion composites. Damage nucleation, growth and coalescence were observed. Quantitative analysis was mainly devoted to the first damage step. Growth and coalescence were difficult to follow due to fast rupture and interrupted tensile tests. The modeling of an amorphous-crystalline composite has been introduced in order to reproduce experimental damage analyses. The first approach requires further investigation to predict damage with different volume fractions. However, this part shows the beginning of an innovative model based on the experimental microstructure.

Kraft recovery boiler smelt is shattered into small droplets by an impinging steam jet to prevent smelt-water explosions in the dissolving tank. Inadequate shattering increases the likelihood of dissolving tank explosions. While industry has not dedicated much effort to smelt shattering, the safety implications require smelt shattering to be studied in detail. An experimental set-up was constructed to simulate the shattering operation using a water-glycerine solution and air instead of smelt and steam respectively. The objective was to examine how physical properties and flow characteristics affect shattering. It was found that increasing shatter jet velocity greatly reduced droplet mean diameter. Increasing the liquid flow rate greatly increased droplet size, as expected. Shattering was not significantly affected by viscosity, unless a weak shatter jet was used on a highly viscous fluid. Increasing the proximity of the shatter jet nozzle decreased droplet size.

An experimental and numerical investigation has been conducted to examine steady, internal, nozzle-generated, gas/liquid mist cooling in vertical channels with ultra-thin, evaporating subcooled liquid films. Interest in this research has been motivated by the need for a highly efficient cooling mechanism in high-power lasers for inertial fusion reactor applications. The aim is to quantify the effects of various operating and design parameters, viz. liquid atomization nozzle design (i.e. spray geometry, droplet size distribution, etc.), heat flux, liquid mass fraction, film thickness, carrier gas velocity, temperature, and humidity, injected liquid temperature, gas/liquid combinations, channel geometry, length, and wettability, and flow direction, on mist cooling effectiveness. A fully-instrumented experimental test facility has been designed and constructed. The facility includes three cylindrical and two rectangular electrically-heated test sections with different unheated entry lengths. Water is used as the mist liquid with air, or helium, as the carrier gas. Three types of mist generating nozzles with significantly different spray characteristics are used. Numerous experiments have been conducted; local heat transfer coefficients along the channels are obtained for a wide range of operating conditions. The data indicate that mist cooling can increase the heat transfer coefficient by more than an order of magnitude compared to forced convection using only the carrier gas. The data obtained in this investigation will allow designers of mist-cooled high heat flux engineering systems to predict their performance over a wide range of design and operating parameters. Comparison has been made between the data and predictions of a modified version of the KIVA-3V code, a mechanistic, three-dimensional computer program for internal, transient, dispersed two-phase flow applications. Good agreement has been obtained for downward mist flow at moderate heat fluxes; at high heat fluxes, the code underpredicts the local heat transfer coefficients and does not predict the onset of film rupture. For upward mist flow, the code underpredicts the local heat transfer coefficients and, contrary to experimental observations, predicts early dryout at the test section exit.

Kraft recovery boiler smelt is shattered into small droplets by an impinging steam jet to prevent smelt-water explosions in the dissolving tank. Inadequate shattering increases the likelihood of dissolving tank explosions. While industry has not dedicated much effort to smelt shattering, the safety implications require smelt shattering to be studied in detail. An experimental set-up was constructed to simulate the shattering operation using a water-glycerine solution and air instead of smelt and steam respectively. The objective was to examine how physical properties and flow characteristics affect shattering. It was found that increasing shatter jet velocity greatly reduced droplet mean diameter. Increasing the liquid flow rate greatly increased droplet size, as expected. Shattering was not significantly affected by viscosity, unless a weak shatter jet was used on a highly viscous fluid. Increasing the proximity of the shatter jet nozzle decreased droplet size.

Um estudo sistemático a respeito da utilização de padrão interno em determinações multielementares por espectrometria de absorção atômica (ETAAS) foi desenvolvido. O objetivo principal do presente trabalho foi verificar a possibilidade de melhorar a precisão e a exatidão dos resultados analíticos, que são obtidos na análise de fluidos biológicos. O pré-tratamento dessas amostras foi simplificado e reduzido a uma única etapa de diluição com surfactante (Triton X-100) e ácido (HNO3). Conseqüentemente, a complexidade da solução diluída de amostra, a ser introduzida no tubo de grafite, apresenta uma elevada quantidade de concomitantes que podem provocar interferências químicas. A seleção preliminar dos elementos a serem testados como padrão interno considerou a semelhança de parâmetros físico-químicos relacionados com o processo de atomização. Desta forma, Ag, Bi, In e Tl foram testados como padrão interno para a determinação simultânea de Cd/Pb em sangue e urina, enquanto Bi, Ge, In, Sb, Sn e Te foram os elementos selecionados para a determinação de Mn/Ni/Se em soro sangüíneo. A melhoria da qualidade dos resultados analíticos obtidos na determinação simultânea de Cd e Pb em sangue foi observada quando Ag foi utilizada como padrão interno, na presença de NH4H2PO4 como modificador químico. Verificou-se uma melhoria na exatidão dos resultados obtidos para Cd e Pb, após a correção com padrão interno. Por outro lado, os resultados obtidos na análise de urina não foram corrigidos por nenhum dos elementos testados. Os melhores resultados para a determinação simultânea de Mn, Ni e Se foram obtidos com a utilização de Bi, Sn e Te como padrão interno. Entretanto, verificou-se que a correção de todos os resultados não seria viável com o uso de um único padrão interno. O melhor desempenho nos testes realizados na presença de soro sangüíneo foi obtido com Bi, que melhorou discretamente a precisão dos resultados obtidos para Se. Desta forma, a padronização interna visando a determinação simultânea de Mn, Ni e Se não foi eficiente. A padronização interna em ETAAS, com a finalidade de melhorar a precisão e a exatidão dos resultados analíticos, é uma estratégia tão complexa, quanto os efeitos interferentes que se pretende corrigir: são necessários mais estudos para compreender melhor como a utilização de uma condição de compromisso afeta os processos de atomização, bem como mais informações a respeito das interferências físicas e químicas causadas por amostras complexas, analisadas por ETAAS após uma simples etapa de diluição. Deve-se considerar com especial atenção o modificador químico e as temperaturas das etapas de pirólise e de atomização empregadas, que são parâmetros críticos para o desempenho de um elemento como padrão interno. / A systematic study involving the use of internal standard for multielement determinations by electrothermal atomic absorption spectrometry was developed. The main objective of this work was evaluate the possibility of improving precision and accuracy of the analytical results for biological fluids. The sample pre-treatment was reduced to a single dilution step with surfactant (Triton X-100) and acid (HNO3), increasing the amount of concomitant introduced into the atomizer. The preliminary selection of the elements to be tested as internal standard considered the resemblance of physico-chemical parameters related with the atomization process. Thus, Ag, Bi, In and Tl were tested as internal standard for the simultaneous determination of Cd/Pb in blood and urine, and Bi, Ge, In, Sb, Sn and Te were the selected elements for the determination of Mn/Ni/Se in blood serum. The correction of the results obtained for the simultaneous determination of Cd and Pb in blood was achieved when Ag was used as internal standard, in presence of NH4H2PO4 as chemical modifier. An improvement for the accuracy of the results was observed for both analytes after their correction with the internal standard. On the other hand, the results obtained for the urine analysis were not corrected by using the tested elements. The best results for the simultaneous determination of Mn, Ni and Se were observed when Bi, Sn and Te were used as internal standard. However, the correction for the results for all analytes was not possible by using only one internal standard. The best performance in presence of the serum was obtained for Bi, which improves slightly the precision for the Se results. Thus, the internal standardization for the simultaneous determination of Mn, Ni and Se was not efficient. The internal standardization in ETAAS, aiming the improvement of precision and accuracy of the analytical results, is a strategy as complex as the interference effects to be corrected: more studies are required in order to better understand how the adoption of a compromised condition disturbs the atomization processes, as well as to get more information about the physical and chemical interference caused by complex samples, analyzed by ETAAS after a single dilution step. The chemical modifier and the selected temperatures for the pyrolysis and atomization steps are critical parameters for the performace of an internal standard and they should be carefully considered.

Initiated with the purpose of assigning the Fraunhofer lines in the solar spectrum to atomic transitions in the 18th century, the collaboration between spectroscopists and astrophysicists has remained fruitful, successful and ever fascinating. This collaboration has resulted in the unique detection of over 200 different molecular species in the interstellar medium (ISM). These interstellar molecular species play significant roles in diverse fields such as atmospheric chemistry, astrochemistry, prebiotic chemistry, astrophysics, astronomy, astrobiology, etc, and in our understanding of the solar system ''the world around us''. This Thesis work focuses on understanding of the different aspects of the chemistry of the various classes of these molecular species. Chapter one starts with an historical perspective of what is now regarded as Molecular Astrophysics or Astrochemistry and discusses the interstellar medium and its properties; interstellar molecular species and their importance; molecular spectroscopy as an indispensible tool in interstellar chemistry and the different formation routes of these molecular species. It also discusses hydrogen bonding which is one of the most important of all the intermolecular interactions. The chapter ends by setting the stage for the present investigations. The chapter two of the Thesis saddled with the task of describing the methodology employed in this Thesis begins by setting the stage on the importance of computational chemistry in interstellar chemistry. It discusses the Gaussian 09 suite of programs and the various theoretical methods used in all the quantum chemical calculations reported in this Thesis. The chapter ends with a brief summary on the homebuilt Pulsed Nozzle Fourier Transform Microwave (PN-FTMW) spectrometer used for the preliminary studies on Isoprene...Argon weakly bound complex reported in the appendix. After the introductory chapters, chapter three begins with what is unarguably one of the most important classes of interstellar molecular species - 'interstellar isomers'. In this chapter, the Energy, Stability and Abundance (ESA) relationship existing among interstellar molecular species has been firmly established using accurate thermochemical parameters obtained with the composite models and reported observational data. From the relationship, “Interstellar abundances of related species are directly proportional to their stabilities in the absence of the effect of interstellar hydrogen bonding”. The immediate consequences of the relationship in addressing some of the questions in interstellar chemistry such as: Where are Cyclic Interstellar Molecules? What are the possible candidates for astronomical observation? Why are more Interstellar Cyanides than isocyanides? among others are briefly discussed. Following the ESA relationship, other studies addressing some of the whys and wherefores in interstellar chemistry are discussed in details. From ESA relationship, though there has not been any successful astronomical observation of any heterocycle, the ones so far searched remain the best candidates for astronomical observation in their respective isomeric groups. The observation of the first branched chain molecule in ISM is in agreement with the ESA relationship and the C5H9N isomers have been shown to contain potential branched chain interstellar molecules. That molecules with the C-C-O backbone have less potential of formation in ISM as compared to their counterparts with the C-O-C backbone has been demonstrated not to be true following the ESA relationship. A detailed investigation on the relationship between molecular partition function and astronomical detection of isomeric species (or related molecules) shows that there is no direct correlation between the two rather there is a direct link between the thermodynamic stability of the isomeric species (or related molecules) and their interstellar abundances which influences the astronomical observation of some isomers at the expense of others. Chapter four presents an interesting and a fascinating phenomenon among the interstellar molecular species as it discusses for the first time, the existence and effects of Interstellar Hydrogen Bonding. This interstellar hydrogen bonding is shown to be responsible for the deviations from thermodynamically controlled processes, delayed observation of the most stable isomers, unsuccessful observations of amino acids among other happenings in interstellar chemistry and related areas. On the prediction that ketenes are the right candidates for astronomical searches among their respective isomers, a ketenyl radical; HCCO has recently been detected in line with this prediction. The deviation from the rule that the ratio of an interstellar sulphur molecule to its oxygen analogue is close to the cosmic S/O ratio is well accounted for on the basis of hydrogen bonding on the surface of the dust grains. Detecting weakly bound complexes in ISM has not been a major interest in the field so far but the detectability of weakly bound complexes in ISM is very possible as discussed in this chapter. Following the conditions in which these complexes are observed in the terrestrial laboratory as compared to the ISM conditions; it suffices to say that weakly bound complexes are present and are detectable in ISM. They could even account for some of the 'U' lines. Chapter five of the Thesis discusses the Linear Interstellar Carbon Chains which are the dominant theme in interstellar chemistry accounting for over 20% of all the known interstellar and circumstellar molecular species. Accurate spectroscopic parameters within experimental accuracy of few kHz which are the indispensable tools for the astronomical observation of these molecular species; are obtained for over 200 different species from the various chains using an inexpensive combined experimental and theoretical approach. With the availability of the spectroscopic parameters; thermodynamics is utilized in accounting for the known systems and in examining the right candidates for astronomical searches. These molecular species are shown to also obey the ESA relationship observed for the isomeric species discussed in chapter three of this work. The effect of kinetics on the formation processes of these molecular species is well controlled by thermodynamics as discussed in this chapter. Finally, the application of these studies in reducing the 'U' lines and probing new molecular species has been briefly summarized. Chapter six discusses Interstellar Ions and Isotopologues which are two unique classes of interstellar molecular species. Different studies on interstellar ions and isotopologues are presented. From the studies on interstellar protonated species with over 100 molecular species; protonated species resulting from a high proton affinity prefers to remain protonated rather than transferring a proton and returning to its neutral form as compared to its analogue that gives rise to a lower proton affinity from the same neutral species. The studies on detectable interstellar anions account for the known interstellar anions and predict members of the C2nO-, C2nS-, C2n-1Si-, HC2nN-, CnP-, and C2n chains as outstanding candidates for astronomical observation including the higher members of the C2nH- and C2n-1N- groups whose lower members have been observed. From high level ab initio quantum chemical calculations; ZPE and Boltzmann factor have been used to explain the observed deuterium enhancement and the possibility of detecting more deuterated species in ISM. Though all the heterocycles that have so far been searched for in ISM have been shown to be the right candidates for astronomical observation as discussed in the ESA relationship, they have also been shown to be strongly bonded to the surface of the interstellar dust grains thereby reducing their abundances, thus, contributing to their unsuccessful detection except for furan which is less affected by hydrogen bonding. The D-analogues of the heterocycles are shown from the computed Boltzmann factor to be formed under the dense molecular cloud conditions where major deuterium fractionation dominates implying very high D/H ratio above the cosmic D/H ratio which suggests the detectability of these deuterated species. Chapter seven examines the isomerization of the most stable isomer (which is probably the most abundant) to the less stable isomer(s) as one of the plausible formation routes for interstellar molecular species. An extensive investigation on the isomerization enthalpies of 243 molecular species from 64 isomeric groups is reported. From the results, the high abundances of the most stable isomers coupled with the energy sources in interstellar medium drive the isomerization process even for relative enthalpy difference as high as 67.4 kcal/mol. Specifically, the cyanides and their corresponding isocyanides pairs appear to be effectively synthesized via this process. The following potential interstellar molecules; CNC, NCCP, c-C5H, methylene ketene, methyl Ketene, CH3SCH3, C5O, 1,1-ethanediol, propanoic acid, propan-2-ol and propanol are identified and discussed. In all the isomeric groups, isomerization appears to be an effective route for the formation of the less stable isomers (which are probably less abundant) from the most stable ones. Chapter eight summarizes the conclusions drawn from the different studies presented in this Thesis and also highlights some of the future directions of these studies. The first appendix presents the preliminary study on Isoprene...Ar weakly bound complex while the second appendix contains a study on interstellar C3S describing the importance of accurate dipole moment in calculating interstellar abundances of molecular species and in astrophysical and astronomical models.

Les écoulements diphasiques jouent un rôle prépondérant dans les moteurs-fusées à ergols liquides cryogéniques, équipant par exemple les lanceurs de la famille Ariane. L'étude expérimentale de tels engins propulsifs étant complexe et onéreuse, disposer d'outils numériques à même de simuler fidèlement leur fonctionnement se révèle être un objectif aussi important qu'ambitieux. La difficulté majeure réside dans le caractère fortement multi-échelles du problème, si bien qu’aucune approche numérique existante n'est capable à elle seule de décrire parfaitement l'ensemble des échelles liquides. Partant de ce constat, les travaux présentés dans cette thèse visent à mettre en place une stratégie de couplage entre des modèles bien adaptés aux différentes topologies d'écoulement diphasique, et ce dans le cadre de la plateforme logicielle multi-physique CEDRE développée par l'ONERA. La démarche adoptée consiste précisément à coupler un modèle à interface diffuse de type ``4 équations'' pour les zones à phases séparées, et un modèle cinétique eulérien pour la phase dispersée, rendant ainsi possible la description de l’atomisation primaire. Par ailleurs, les conditions sévères qui règnent dans les moteurs cryotechniques, où de forts gradients de température, vitesse et densité sont rencontrés, mettent à l'épreuve la robustesse des méthodes numériques. Une nouvelle méthode MUSCL multipente pour maillages non structurés généraux a ainsi été développée, permettant d’améliorer la robustesse et la précision des schémas de discrétisation spatiale. L’ensemble de la stratégie de couplage est finalement appliquée à la simulation du banc Mascotte de l'ONERA pour la combustion cryotechnique. / Two-phase flows play a significant role for the proper functioning of cryogenic liquid-propellant rocketengines, such as those that equip the launchers of the Ariane family. Since the experimental investigationof such propulsion devices is complex and expensive, developing numerical tools able to accuratelysimulate their functioning, is a crucial but nonetheless ambitious objective. The major difficulty is due tothe multiscale nature of the problem, as a result of which there is currently no numerical approach ableto perfectly describe all the liquid scales on its own. Based on this observation the work presented in thisthesis aims at setting up a coupling strategy between models well-adapted to each two-phase flowtopology, in the framework of the ONERA’s multiphysics CEDRE software. The approach adoptedprecisely consists in coupling a 4-equation diffuse interface model for the separated phases and aeulerian kinetic model for the dispersed phase, thus making it possible to describe primary atomization.Besides, the harsh conditions within cryogenic rocket engines, where large temperature, velocity anddensity gradients are encountered, severely challenge the robustness of numerical methods. A newmultislope MUSCL method for general unstructured meshes is thus developed in order to improve therobustness and accuracy of space discretization schemes. The whole coupling strategy is finally appliedto the numerical simulation of the ONERA’s Mascotte test bench for cryogenic combustion research.

Práce komplexně popisuje vodní a vodovzdušné chlazení pomocí metod CFD (Computational Fluid Dynamics) konkrétně s využitím software ANSYS FLUENT. Skládá se ze dvou hlavních částí, z nichž první se zabývá numerickým popisem jediné vodní kapky a druhá popisem směsí kapek představující paprsek válcové a ploché trysky. Je založena převážně na vícefázových modelech proudění a vlastních uživatelsky definovaných funkcí (User Defined Functions, UDF) představujících stěžejní část práce. Uvedené výpočtové modely jsou ve většině případů verifikovány pomocí experimentálních dat nebo jiných numerických modelů. V první části práce jsou teoreticky postupně rozebrány všechny tři použité vícefázové modely proudění. První z nich, Volume Of Fluid model (VOF), byl použit pro modelování jediné kapky (mikromodel). Zatímco zbývající dva, Euler-Euler model a Euler-Lagrange model, byly aplikovány v modelu celého paprsku trysky (makromodel). Mikromodel popisuje dynamiku volného pádu vodní kapky. Pro malé průměry kapek (~100µm) standardní model povrchového napětí (Continuum Surface Force, CSF) způsoboval tzv. parazitní proudy. Z toho důvodu je v práci rozebrána problematika výpočtu normál, křivostí volných povrchů a povrchového napětí jako zdroje objemových sil v pohybových rovnicích. Makromodel se zabývá studiem dynamiky celého paprsku tj. oblastí od ústí trysky po dopad na horký povrch, bere v úvahu kompletní geometrii, tzn. např. podpůrné válečky, bramu, spodní část krystalizátoru apod. V práci je rozebrána 2D simulace dopadu paprsku válcové trysky pomocí VOF modelu Euler-Lagrange modelu na horký povrch. Pro případ s VOF modelem byl navržen model blánového varu. Euler-Euler model a Euler-Lagrange model byly využity pro simulaci paprsku ploché trysky horizontálně ostřikující horkou bramu přímo pod krystalizátorem nad první řadou válečků. Pro Euler-Euler model byl navržen model sekundárního rozpadu paprsku založený na teorii nejstabilnější vlnové délky (Blob jet model). Jelikož diskrétní Lagrangeovy částice tvořily v určitých místech spíše kontinuální fázi, byl navržen a otestován model pro konverzi těchto částic do VOF.