Rozprawy doktorskie na temat „Atomization and Sprays”

This thesis presents an extensive study of turbulent air-blasted sprays aimed at advancing the current understanding of the atomization and the turbulent combustion of dense sprays. The burner employed controls the spray quality by recessing the liquid-injecting needle inside the air-blast tube to transition the spray from dilute to dense. A pilot is used to stabilize the flame to the burner which is sitting in a co-flowing stream of secondary air. Three fuels, acetone, ethanol, and biodiesel, are used to generate several sprays that cover a broad range of non-dimensional numbers. Probability distributions of wavelength and amplitude of instabilities forming on coaxial air-blast atomizers are measured directly using high-speed shadowgraphs (or back-lit microscopic imaging), in a range of cases that investigate the independent effects of a suite of parameters. The influence of jet velocity and gas velocity on the initiation and growth of jet instabilities is discussed. The range of mechanisms governing the formation of liquid fragments and their relation to surface instabilities is discussed. Previous work suggested that the mean wavelength scales with the boundary layer thickness. This is confirmed here and extended to demonstrate that the wavelength probability distribution correlates well with the ligament length probability distribution. This establishes a direct link between interfacial instabilities and ligament formation in air-assisted primary atomization. The complete structure of atomizing liquid fragments is analyzed by employing multi-dimensional visualization techniques and advanced image processing, where objects from multiple views are matched to extract three-dimensional information. An in-house MATLAB script is developed to extract the spray volume, which employs the principle of image discretization, where each image is divided into a number of slices, and the individual slice from each camera is matched to compute the liquid volume fraction in each image. The volume of individual objects is calculated based on their planar area and orientation. An error analysis is performed using dozens of three-dimensional virtual models of fragment-like shapes with a known volume. Local characteristics of atomizing fragments are discussed by using the information obtained through the slicing method. A detailed account of fragment statistics are provided for the atomizing sprays. The LIF-OH-CH2O technique is used to measure the product of OH and CH2O ([OH]*[CH2O]) and hence the heat release zones in turbulent, moderately dense spray flames of ethanol and biodiesel fuels. A combination of several filters is used to remove interference from droplet luminosity. Mie scattering is measured jointly on a separate camera to locate the droplets with respect to the reaction zones. It is found that while the overall flame structure is similar to that of a diffusion flame. Structures referred to as burning rings of different sizes are observed, and these are ignited by heat source before they grow, propagate, and burnout. Statistics about the occurrence of these rings, with and without clouds of fragments within them, are presented. Occurrences of single-droplet combustion are also noted. Overall this thesis expands the current understanding of the primary and secondary atomization zones and provides new insights into the turbulent combustion characteristics of denser sprays. An extensive and novel data set in generated, and this will be made available to modelers with the aim of changing the predictive capability of both atomization and combustion of such flows.

This thesis presents an extensive study of the atomization and combustion of dielectric liquids using a hybrid air-blast electrostatic atomizer. While airblast atomization relies on the shear stresses generated at the liquid-air interface, electrostatic atomization introduces an electric charge into the bulk liquid, and the resulted Coulombic repulsive force facilitates the fragmentation process. The atomizer introduced in this contribution is specifically designed to operate in either a single (air-blast or electrostatic) or hybrid mode to enable the delivery of a charged and/or air-assisted spray for combustion applications. The aim is to understand the effect of adding electric charge to a liquid jet which is subject to break up in a co-flowing air stream. In addition to analysis of atomization processes, the influence of charge on flame structure is also analyzed. Laser diagnostics are utilized for measurements and the results obtained for the atomizer in hybrid mode (air-blast + electrostatic) are compared with the pure air-blast mode. Firstly, a high-speed microscopic shadowgraphy technique is implemented to examine near-field spray structure. Diesel is used as a dielectric liquid to create various sprays that cover a range of non-dimensional numbers. The effect of charge on liquid jet unsteadiness and on the probability distribution of wavelength and amplitude of instabilities is discussed. The influence of charge on droplet and ligament size and their population is also analyzed. The findings show that the application of charge makes the liquid jet more unstable and the instabilities forming on the liquid core exhibit a shift to a shorter wavelength with a broadening in the probability distribution of wave amplitude. In addition, a droplet and ligament size reduction along with an increase in droplet count is observed with the addition of charge. The thesis then progresses to discussing results from reacting sprays stabilized on a pilot where kerosene is chosen as the liquid. A premixed pilot flame is used to provide a steady heat source for stabilizing the hybrid atomized sprays. Flame stability characteristics, in terms of blow-off velocity, are presented as a function of controlling parameters, without and with charge. Downstream droplet statistics and flow field for both non-reacting and reacting sprays are shown using laser Doppler velocimetry/phase Doppler anemometry (LDV/PDA) revealing key features in the droplet fields from this burner. Due to relatively low spray specific charge for the aerodynamic Weber numbers investigated, the droplet size and velocity remained largely unaffected through the addition of charge, however, a rise in particle concentration at the center of the spray was noted. Finally, high-speed hydroxyl planar laser induced fluorescence (OH-PLIF) is used to locate reaction zones and comment on the morphology of the reaction zone structures. In a hybrid mode, the charge was seen to push reaction zones radially outward and assisted in stabilizing the flame by keeping OH islands more connected when compared to a pure air-blast mode. This observation was also consistent with a slight improvement in flame stability with the addition of charge.

Understanding the physics governing primary atomization of high pressure fuel sprays is of paramount importance to accurately model combustion in direct injection engines. The small length and time scales of features that characterize this process falls below the resolution power of typical grids in CFD simulations, which necessitates the inclusion of physical models (sub-models) to account for unresolved physics. Unfortunately current physical models for fuel spray atomization used in engine CFD simulations are based on significant empirical scaling because there is a lack of experimental data to understand the governing physics. The most widely employed atomization sub-model used in current CFD simulations assumes the spray atomization process to be dominated by aerodynamically-driven surface instabilities, but there has been no quantitative experimental validation of this theory to date. The lack of experimental validation is due to the high spatial and temporal resolutions required to simultaneously to image these instabilities, which is difficult to achieve. The present work entails the development of a diagnostic technique to obtain high spatial and temporal resolution images of jet breakup and atomization in the near nozzle region of Gasoline Direct Injection (GDI) sprays. It focuses on the optical setup required to achieve maximum illumination, image contrast, sharp feature detection, and temporal tracking of interface instabilities for long-range microscopic imaging with a high-speed camera. The resolution and performance of the imaging system is characterized by evaluating its modulation transfer function (MTF). The setup enabled imaging of GDI sprays for the entire duration of an injection event (several milliseconds) at significantly improved spatial and temporal resolutions compared to historical spray atomization imaging data. The images show that low to moderate injection pressure sprays can be visualized with a high level of detail and also enable the tracking of features across frames within the field of view (FOV)

[ES] La comprensión de los fenómenos físicos que acontecen en la región densa (también conocida como campo cercano) durante la atomización de los sprays ha sido una de las mayores incógnitas a la hora de estudiar sus aplicaciones. En el sector industrial, el rango de interés abarca desde toberas en aplicaciones propulsivas a sprays en aplicaciones médicas, agrícolas o culinarias. Esta evidente falta de conocimiento obliga a realizar simplificaciones en la modelización, provocando resultados poco precisos y la necesidad de grandes caracterizaciones experimentales en la fase de diseño. De esta manera, los procesos de rotura del spray y atomización primaria se consideran problemas físicos fundamentales, cuya complejidad viene dada como resultado de un flujo multifásico en un régimen altamente turbulento, originando escenarios caóticos. El análisis de este problema es extremadamente complejo debido a la ausencia sustancial de teorías validadas referentes a los fenómenos físicos involucrados como son la turbulencia y la atomización. Además, la combinación de la naturaleza multifásica del flujo y su comportamiento turbulento resultan en una gran dificultad para afrontar el problema. Durante los últimos 10 años, las técnicas experimentales han sido finalmente capaces de visualizar la región densa, pero la confianza, análisis y efectividad de dichos experimentos en esta región del spray todavía requiere de mejoras sustanciales. En este contexto, esta tesis trata de contribuir al entendimiento de estos procesos físicos y de proporcionar herramientas de análisis para estos flujos tan complejos. Para ello, mediante Direct Numerical Simulations se ha afrontado el problema resolviendo las escalas de movimiento más pequeñas, y capturando todas las escalas de turbulencia y eventos de rotura. Uno de los objetivos de la tesis ha sido evaluar la influencia de las condiciones de contorno del flujo entrante en la atomización primaria y en el comportamiento turbulento del spray. Para ello, se han empleado dos condiciones de contorno diferentes. En primer lugar se ha empleado una condición de contorno sintética para producir turbulencia homogenea a la entrada, simulando el comporamiento de la tobera. Una de las características más interesantes de este método es la posibilidad de retocar los parámetros dentro del algoritmo. En particular, la escala de longitud integral se ha variado para evaluar la influencia de las estructuras mas grandes de la tobera en la atomización primaria. El análisis de la condición de contorno sintética también ha permitido el diseño óptimo de simulaciones de las cuales se han derivado estadísticas turbulentas significativas. En este escenario, se han llevado a cabo estudios más profundos sobre la influencia de propiedades de las estructuras turbulentas como la homogeneidad y la anisotropía tanto en el espectro de los flujos como en las estadísticas de las gotas. Para tal fin, se han desarrollado metodologías novedosas para computar el análisis espectral y la estadística de las gotas Entre los resultados de este análisis destaca la independencia de la condición de contorno de entrada en las estadísticas de las gotas, mientras que por otra parte, recalca que las características turbulentas desarrolladas en el interior de la tobera afectan a la cantidad total de masa atomizada. Estas consideraciones se encuentran respaldadas por el análisis espectral realizado, mediante el cuál se concluye que la turbulencia multifásica comparte el comportamiento universal descrito por las teorías de Kolmogorov.
[CAT] La comprensió dels fenòmens físics que succeïxen en la regió densa (també coneguda com a camp pròxim) durant l'atomització dels sprays ha sigut una de les majors incògnites a l'hora d'estudiar les seues aplicacions. En el sector industrial, el rang d'interés comprén des de toveres en aplicacions propulsives a sprays en aplicacions mèdiques, agrícoles o culinàries. Esta evident falta de coneixement obliga a realitzar simplificacions en la modelització, provocant resultats poc precisos i la necessitat de grans caracteritzacions experimentals en la fase de disseny. D'esta manera, els processos de ruptura del spray i atomització primària es consideren problemes físics fonamentals, la complexitat dels quals ve donada com resultat d'un flux multifàsic en un règim altament turbulent, originant escenaris caòtics. L'anàlisi d'este problema és extremadament complex a causa de l'absència substancial de teories validades dels fenòmens físics involucrats com són la turbulència i l'atomització. A més, la combinació de la naturalesa multifàsica del flux i el seu comportament turbulent resulten en una gran dificultat per a afrontar el problema. Durant els últims 10 anys les tècniques experimentals han sigut finalment capaces de visualitzar la regió densa, però la confiança, anàlisi i efectivitat dels experiments en esta regió del spray encara requerix de millores substancials. En este context, esta tesi tracta de contribuir en l'enteniment d'estos processos físics i de proporcionar ferramentes d'anàlisi per a estos fluxos tan complexos. Per a això, per mitjà de Direct Numerical Simulations s'ha afrontat el problema resolent les escales de moviment més menudes, al mateix temps que es capturen totes les escales de turbulència i esdeveniments de ruptura. Un dels objectius de la tesi ha sigut avaluar la influència que les condicions de contorn del flux entrant tenen en l'atomització primària i en el comportament turbulent del spray. Per a això, s'han empleat dos condicions de contorn diferents. En primer lloc s'ha empleat una condició de contorn sintètica per a produir turbulència homogènia a l'entrada, simulant el comportament de la tovera. Una de les característiques més interessants d'este mètod és la possibilitat de retocar els paràmetres dins de l'algoritme. En particular, l'escala de longitud integral s'ha variat per a avaluar la influència de les estructures mes grans de la tovera en l'atomització primària. L'anàlisi de la condició de contorn sintètica també ha permés el disseny òptim de simulacions de les quals s'han derivat estadístiques turbulentes significatives. En este escenari, s'han dut a terme estudis més profunds sobre la influència de propietats de les estructures turbulentes com l'homogeneïtat i l'anisotropia tant en l'espectre dels fluxos com en les estadístiques de les gotes. Per a tal fi, s'han desenrotllat metodologies noves per a computar l'anàlisi espectral i l'estadística de les gotes. Entre els resultats d'esta anàlisi destaca la independència de la condició de contorn d'entrada en les estadístiques de les gotes, mentres que d'altra banda, es recalca que les característiques turbulentes desenrotllades en l'interior de la tovera afecten a la quantitat total de massa atomitzada. Estes consideracions es troben recolzades per l'anàlisi espectral realitzat, per mitjà del qual es conclou que la turbulència multifásica compartix el comportament universal descrit per les teories de Kolmogorov.
[EN] The understanding of the physical phenomena occurring in the dense region (also known as near field) of atomizing sprays has been long seen as one of the biggest unknown when studying sprays applications. The industrial range of interest goes from nozzles in combustion and propulsion applications to medical sprays, agricultural and food process applications. This substantial lack of knowledge is responsible for some important simplification in modeling, that often result to be inaccurate or simply partial, leading to the evident need of large experimental characterization during the design phase. In fact, the spray breakup and primary atomization processes are indeed fundamental problems of physics, which complexity results from the combination of a multiphase flow in a highly turbulent regime that leads to chaotic scenarios. The analysis of this problem is extremely problematic, due to a substantial lack of definitive theories about the physical phenomena involved, namely turbulence and atomization. Furthermore, the combination of the multiphase nature of the flow and its turbulent behavior makes substantially difficult to address the problem. Only within the last 10 years, experimental techniques have been capable of visualizing the dense region, but the experiments reliability, analysis and effectiveness in this region still requires vast improvements. In this scenario, this thesis aims to contribute in the understanding of these physical process and to provide analysis tools for these complex flows. In order to do so, Direct Numerical Simulations have been used for addressing the problem at its smallest scale of motion, while reliably capturing all turbulence scales and breakup events. The multiphase nature of the flow is accounted for by using the Volume of Fluid method. One of the goal of the thesis was to assess the influence of the inflow boundary conditions on the primary atomization and on the spray's turbulence behavior. In order to do so, two different boundary conditions were used. In a first place, a synthetic inflow boundary condition was used in order to produce a homogeneous turbulence inflow, simulating the nozzle behavior. One of the interesting features of this method was the possibility of tweaking the parameters within the algorithm. In particular, the integral length scale was varied in order to assess the influence of nozzle larger turbulent structures on the primary atomization. The analysis on the synthetic boundary condition also allowed to optimally design simulations from which derive meaningful turbulence statistics. On this framework, further studies were carried over on the influence of turbulent structures properties, namely homogeneity and anisotropy, on both the flows spectra and droplets statistics. In order to achieve this goal, novel procedures for both computing the flow spectra and analyzing droplets were developed and are carefully addressed in the thesis. The results of the analysis highlight the independence of droplets statistics from the inflow boundary condition, while, on the other hand, remarking how the total quantity of atomized mass is significantly affected by the turbulence features developed within the nozzle. This considerations are supported by the spectrum analysis performed, which also highlighted how multiphase turbulence shares the universal features described in Kolmogorov theories.
Crialesi Esposito, M. (2019). Analysis of primary atomization in sprays using Direct Numerical Simulation [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/133975
TESIS

Fuel sprays play a major role in order to achieve the required combustion characteristics and pollutant emissions reduction on internal combustion engines, and thus, an accurate prediction of its behavior is required to perform reliable engine combustion and pollutant simulations. A great effort both on experimental and theoretical studies of spray atomization and dispersion has been performed in the latest years. As a result, Computational Fluid Dynamics (CFD) calculations have become a standard tool not only for spray physics understanding but also for design and optimization of engine spray systems. However, spray modeling in its different uses in the Internal Combustion Engine (ICE) context is still nowadays a challenging task due to the complex interrelated phenomena taking place, some of them still not fully understood. Primary atomization and secondary breakup, droplet collision, coalescence and vaporization, turbulent interactions between phases have to be solved under high Reynolds (so they are turbulent) and Weber numbers conditions due to the high speed (~500 m/s) and small nozzle diameter (~100 µm) imposed by current engine injection systems technologies. Moreover, Taylor numbers cover a wide range, according to the composition of the injected liquid. Those conditions make experimental observations quite challenging and probably insufficient, especially in the very near nozzle region, where primary atomization takes place. Most of the CFD spray models are currently based on the Discrete Droplet Method. The continuous liquid jet is discretized into 'blobs' or 'parcels', which consists in a number of droplets with the same characteristics. A Lagrangian method is applied to track the liquid phase parcels, which are subject to breakup according to atomization models mainly based on the linear instability theory proposed by Reitz and later extended by Huh and Gosman for liquid turbulence effects to be considered. This approach has been successfully applied b
Khuong ., AD. (2012). The Eulerian-Lagrangian Spray Atomization (ELSA) Model of the Jet Atomization in CFD Simulations: Evaluation and Validation [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/17237
Palancia

The current study aims to investigate and develop numerical tools for the simulation of interfacial flows, including those with turbulence and primary spray atomization. In the past, many approaches were developed to track the interface of a multi-phase/immiscible flow system. Two of them, the volume of fluid (VoF) and the level set (LS), have shown promising findings and have been considered a benchmark for further research. VoF methods have proved good mass conservation behaviour during the advection of solution. However, LS methods have shown better interface representation when used for the calculations of interfacial flows. Most of the recent development in this context either originated from VoF, LS or a combination of both. In the first part of this thesis importance of the current study is highlighted in detail, and an overview of the recent developments for the numerical solutions of immiscible flows is presented. A detailed description of interface-capturing methods and their development over time is presented in a systematic way, leading to the issues addressed in this study. A stochastic field PDF-LES solution of the VoF approach is analyzed against the Sydney needle spray measurements. Solution behaviour for different combinations of stochastic fields and LES grid cells are examined in detail. In the second part of this work, three different LES treatment for unresolved sub-grid fluctuations, TFVoF, TF-AC and newly developed EVD, has been presented. The performance analysis of two already existing solutions, TFVoF and TF-AC, is compared with the EVD approach. Sydney needle spray measurements are used as a benchmark for this comparison. The suitability of the EVD approach is also tested for different flow conditions. Simulation results confirm that sub-grid surface tension plays an important role in the realistic prediction of the jet decay rate, and proper closure models to account for sub-grid surface tension are very necessary. An already existing sub-grid surface tension closure model is incorporated into OpenFOAM and tested for all three LES filtered solutions. The findings of this work suggest that the newly developed EVD approach has shown promising improvements to produce more reliable and realistic solutions for multi-phase flow applications. Sub-volume surface tension closure model, following the EVD formulation, is helpful for more accurate jet decay rate predictions.

This thesis presents the development of GPU accelerated solvers for use in simulation of the primary atomization phenomenon. By using the open source continuum mechanics library, OpenFOAM, as a basis along with the NVidia CUDA API linear system solvers have been developed so that the multiphase solver runs in part on GPUs. This aims to reduce the enormous computational cost associated with modelling primary atomization. The modelling of such is vital to understanding the mechanisms that make combustion efficient. Firstly, the OpenFOAM code is benchmarked to assess both its suitability for atomization problems and to establish efficient operating parameters for comparison to GPU accelerations. This benchmarking then culminates in a comparison to an experimental test case, from the literature, dominated by surface tension, in 3D. Finally, a comparison is made with a primary atomizing liquid sheet as published in the literature. A geometric multigrid method is employed to solve the pressure Poisson equations, the first use of a geometric multigrid method in 3D GPU accelerated VOF simulation. Detailed investigations are made into the compute efficiency of the GPU accelerated solver, comparing memory bandwidth usage to hardware maximums as well as GPU idling time. In addition, the components of the multigrid method are also investigated, including the effect of residual scaling. While the GPU based multigrid method shows some improvement over the equivalent CPU implementation, the costs associated with running on GPU cause this to not be significantly greater.

[EN] The main objective of this work is the modeling of diesel sprays under engine conditions, including the atomization, transport and evaporation processes pivotal in the diesel spray formation and its development. For this purpose, an Eulerian single fluid model, embedded in a RANS environment, is implemented in the CFD platform OpenFOAM. The modeling approach implemented here is based on the ⅀-Y model. The model is founded on the assumption of flow scales separation. In actual injection systems, it can be assumed that the flow exiting the nozzle is operating at large Reynolds and Weber numbers and thus, it is possible to assume a separation of features such as mass transport (large scales) from the atomization process occurring at smaller scales. The liquid/gas mixture is treated as a pseudo-fluid with variable density and which flows with a single velocity field. Moreover, the mean geometry of the liquid structures can be characterized by modeling the mean surface area of the liquid-gas interphase per unit of volume. Additionally, an evaporation model has been developed around the particular characteristics of the current engine technologies. This means that vaporization process is limited by fuel-air mixing rate and fuel droplets evaporate as long as there is enough air for them to heat up and vaporize. Consequently, the evaporation model is based on the Locally Homogeneous Flow (LHF) approach. Under the assumption of an adiabatic mixing, in the liquid/vapor region, the spray is supposed to have a trend towards adiabatic saturation conditions and to determine this equilibrium between phases Raoult's ideal law is considered. Finally, the spray model is coupled with an advanced combustion model based on approximated diffusion flames (ADF), which reduces the computational effort especially for complex fuels and is a natural step for modeling diesel sprays. First, the model is applied to a basic external flow case under non-vaporizing conditions, extremely convenient due to both the experimental database available and the symmetric layout which allows important simplification of the modeling effort. Good agreement between computational results and experimental data is observed, which encourages its application to a more complex configuration. Secondly, the model is applied to the "Spray A" from the Engine Combustion Network (ECN), under non-vaporizing conditions, in order to reproduce the internal structure of diesel sprays as well as to produce accurate predictions of SMD droplets sizes. Finally, vaporizing "Spray A" studies are conducted together with the baseline reacting condition of this database. The calculated spray penetration, liquid length, spray velocities, ignition delay and lift-off length are compared with experimental data and analysed in detail.
[ES] El objetivo principal de este trabajo es el modelado de chorros diésel en condiciones de motor, incluyendo los fenómenos de atomización, transporte y evaporación fundamentales en la formación y desarrollo del chorro. Para este fin, se implementa un modelo de spray euleriano de tipo monofluido en un entorno RANS en la plataforma CFD OpenFOAM. El enfoque de modelado aplicado aquí sigue la idea de un modelo del tipo ⅀-Y. El modelo se fundamenta en la hipótesis de separación de escalas del flujo. En los sistemas de inyección actuales, es posible asumir que el flujo que sale de la tobera opera a altos números de Reynolds y Webber y por tanto, es posible considerar la independencia de fenómenos como el transporte de masa (grandes escalas del flujo) de los procesos de atomización que ocurren a escalas menores. La mezcla líquido/gas se trata como un pseudo-fluido con densidad variable y que fluye según un único campo de velocidad. Además, la geometría promedio de las estructuras de líquido se puede caracterizar mediante el modelado de la superficie de la interfase líquido/gas por unidad de volumen. Completando el modelo de chorro, se ha desarrollado un modelo de evaporación alrededor de las características particulares de las tecnologías actuales de los motores. Esto supone que el proceso de evaporación está controlado por mezcla aire-combustible y las gotas de combustible se evaporan siempre que exista suficiente aire para calentarlas y evaporarlas. Debido a esto, el modelo de evaporación implementado está basado en el enfoque de Flujos Localmente Homogéneos (LHF). Considerando una mezcla adiabática, en la región líquido/vapor, se supone que el chorro tiende a las condiciones adiabáticas de saturación y para determinar este equilibrio entre fases, se utiliza la ley ideal de Raoult. Finalmente, el modelo de chorro se acopla con un modelo avanzado de combustión basado en llamas de difusión aproximadas (ADF), que reduce el coste computacional especialmente para combustibles complejos y supone el paso lógico en el desarrollo del modelo para simular chorros diesel. En primer lugar, el modelo se aplica al cálculo de un caso básico de flujo externo no evaporativo, muy adecuado tanto por la extensa base de datos experimentales disponible como por la simetría geométrica que presenta, permitiendo una importante simplificación de la simulación. Los resultados obtenidos presentan un buen acuerdo con los experimentos, lo cual estimula su aplicación en configuraciones más complejas. En segundo lugar, el modelo se aplica al cálculo del "Spray A" del Engine Combustion Network (ECN), no evaporativo, para reproducir la estructura interna del chorro diesel así como predecir tamaños de gota (SMD) de forma precisa. Finalmente, se realizan estudios evaporativos del "Spray A" junto con la condición nominal reactiva de esta base de datos. La penetración de vapor, la longitud líquida, velocidad, el tiempo de retraso y la longitud de despegue de llama calculados se comparan con los datos experimentales y se analizan en detalle.
[CAT] L'objectiu principal d'aquest treball és el modelatge de dolls dièsel en condicions de motor, incloent els fenòmens d'atomització, transport i evaporació fonamentals en la formació i desenvolupament del doll. Amb aquesta finalitat, s'implementa un model de doll eulerià de tipus monofluid en un entorn RANS a la plataforma CFD OpenFOAM. L'enfocament de modelatge aplicat ací segueix la idea d'un model del tipus ⅀-Y. El model es fonamenta en la hipòtesi de separació d'escales del flux. En els sistemes d'injecció actuals, és possible assumir que el flux que surt de la tovera opera a alts nombres de Reynolds i Webber, i per tant és possible considerar la independència de fenòmens com el transport de massa (grans escales del flux) dels processos d'atomització que ocorren a escales menors. La mescla líquid / gas es tracta com un pseudo-fluid amb densitat variable i que flueix segons un únic camp de velocitat. A més, la geometria mitjana de les estructures de líquid es pot caracteritzar mitjançant el modelatge de la superfície de la interfase líquid / gas per unitat de volum. Completant el model, s'ha desenvolupat un model d'evaporació al voltant de les característiques particulars de les tecnologies actuals dels motors. Això suposa que el procés d'evaporació està controlat per la mescla aire-combustible i les gotes de combustible s'evaporen sempre que hi hagi suficient aire per escalfar i evaporar. A causa d'això, el model d'evaporació implementat està basat en el plantejament de fluxos Localment Homogenis (LHF). Considerant una mescla adiabàtica, a la regió líquid / vapor, se suposa que el doll tendeix a les condicions adiabàtiques de saturació i per determinar aquest equilibri entre fases, s'utilitza la llei ideal de Raoult. Finalment, el model de doll s'acobla amb un model avançat de combustió basat en flamelets de difusió aproximades (ADF), que redueix el cost computacional especialment per a combustibles complexos i suposa el pas lògic en el desenvolupament del model per simular dolls dièsel. En primer lloc, el model s'aplica al càlcul d'un cas bàsic de flux extern no evaporatiu, molt adequat tant per l'extensa base de dades experimentals disponible com per la simetria geomètrica que presenta, permetent una important simplificació de la simulació. Els resultats obtinguts presenten un bon acord amb els experiments, la qual cosa estimula la seva aplicació en configuracions més complexes. En segon lloc, el model s'aplica al càlcul del "Spray A" no evaporatiu de la xarxa Engine Combustion Network (ECN), per reproduir l'estructura interna del doll dièsel així com predir mides de gota (SMD) de forma precisa. Finalment, es realitzen estudis evaporatius del "Spray A" juntament amb la condició nominal reactiva d'aquesta base de dades. La penetració de vapor, la longitud líquida, velocitat, el temps de retard i la longitud d'enlairament de flama calculats es comparen amb les dades experimentals i s'analitzen en detall.
Pandal Blanco, A. (2016). Implementation and Development of an Eulerian Spray Model for CFD simulations of diesel Sprays [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/68490
TESIS

Thesis (Ph. D.)--University of California, Riverside, 2010.
Includes abstract. Available via ProQuest Digital Dissertations. Title from first page of PDF file (viewed May 18, 2010). Includes bibliographical references. Also issued in print.

This paper investigated the effects of swirl number and momentum ratio on the atomization and mixing performance of Swirl-Venturi Lean Direct Injection technology. Mie scattering of liquid water, was used to identify the location of water droplets in a cross section of the injector spray. Experiments were performed with three air swirlers with vane angles of 45, 52 and 60 degrees. The swirl number varied from 0.58 to 1.0 and air-to-liquid ratios from 15.8 to 35.6. A transition was observed in the liquid spray distribution for the 52 degree case, which unexpectedly produced twice as much signal than the 45 and 60 degree cases. The main cause of this increased signal may be due to instabilities in the flow when transitioning from low to high swirl states. The results from investigation of swirl number it was found that the spray pattern for is sensitive to swirl intensity. Two flow states were observed for a lower and higher swirl flow as well as a transition state that occurred with the lower swirl state. This work may aid in the specific inquiry of physical mechanisms relating to the effect of flow states on spray distribution. It is found that improved atomization and mixing performance are a result of increase in swirl number.

Master thesis focused on the research of the effervescent atomizers. Effervescent atomizers belong to a group of two-phase atomizers, which are often used in combustion applications. Right there in combustion applications, the degree of the stability sprays has a significant impact on combustion efficiency and exhaust gas emissions. The main aim of this work was to asses the level of spray unsteadiness depending on the atomizer design and its operating mode. The effect of construction was studied on the diameter and length of mixing chamber, and then on the size, number and location of aeration holes. Seventeen specific variants of the atomizer were constructed by different combinations of these design parameters. Each of these variants was measured in three operating modes, which were represented by a liquid pressure at the inlet to the atomizer and gas-to-liquid mass flow ratio (GLR). To evaluate the level of spray unsteadiness was used a methodology, which compares the ideal element´s distribution into the interparticle time bin, defined for the ideal (stable) spray, with the experimentally observed distribution. The laser measurement system P/DPA (Phase Doppler Particle Analyzer) was used to determine the experimental interparticle distribution. The result of the comparison of the ideal and the experimental distribution was the parameter , which expresses the level of spray unsteadiness for a specific atomizer and operating mode. With that parameter it was possible to compare the individual atomizers and determinate to the benefit of various construction´s correction of the atomizer. The results showed the recommendation for the modifications of the atomizer, creating a spray with a minimum level of spray´s unsteadiness. For the surveyed atomizer and his individual costruction´s variations the drawing was made.

Dans le cadre de la simulation à grandes échelles (LES), la thèse aborde la simulation des sprays dans les conditions d’un moteur à injection directe. La vitesse de l’injection des sprays dans ces conditions est très élevée. En conséquence, des structures énergétiques intermittentes aux petites échelles turbulentes peuvent se manifester dans l’écoulement produit. C’est pourquoi l’accent est mis sur la simulation stochastique des effets turbulents aux échelles non-résolues par LES dans les conditions d’un moteur à injection directe. L’impact de ces effets sur l’atomisation primaire et secondaire, la dispersion des gouttelettes et leur vaporisation représente l’élément essentiel de cette thèse. Dans le but de modéliser ces effets d’intermittence aux échelles non-résolues, deux différentes approches ont été proposées récemment dans la littérature. Dans la thèse, l’accent est mis sur leur application et une éventuelle amélioration pour les conditions d’un moteur à injection directe. La première approche est LES-SSAM (Stochastic Subgrid Acceleration Model). Contrairement aux LES classiques, la LES-SSAM modélise l’accélération turbulente non résolue par le forçage de sous-maille des équations de Navier-Stokes. Ce forçage représente un processus stochastique de type Ornstein-Uhlenbeck construit de telle façon que les propriétés stochastiques de l’accélération, observées par les expériences et les simulations directes, sont représentées. Une telle LES-SSAM, où l’expression de la norme de l’accélération de sous-maille est modifiée, a été appliquée et testée pour la modélisation de l’écoulement interne de l’injecteur d’une simple configuration. Les résultats ont démontré l’efficacité de cette approche malgré la résolution grossière du maillage. Une autre application de LES-SSAM, dans la thèse, concerne sa combinaison avec la méthode VoF pour la simulation de l’écoulement à l’interface au voisinage de l’injecteur. Ici aussi, l’efficacité de cette combinaison a été démontrée en comparaison avec l’expérience et les méthodes numériques actuellement employées pour la simulation de l’atomisation primaire. La deuxième approche abordée dans la majeure partie de la thèse, et qui vise aussi à représenter les effets de l’intermittence aux échelles non-résolues, se base sur la formulation stochastique de la dynamique des gouttes en pulvérisation et en vaporisation, tout en couplage two-way avec l’écoulement turbulent. Les travaux contribuent à la vérification et l’amélioration de cette formulation stochastique. Ainsi le modèle stochastique d’atomisation secondaire est contrôlé par le processus stochastique log-normal pour la dissipation visqueuse. La même variable est la variable-clé pour le modèle de dispersion de gouttes, ces dernières étant soit inférieures soit supérieures à l’échelle de Kolmogorov. La dernière situation a été décrite par la modification de l’équation de mouvement d’une goutte. Enfin, un nouveau modèle stochastique de vaporisation des gouttes, dont le mélange turbulent fait partie du modèle, a été proposé et testé. Tous ces modèles stochastiques ont été implantés dans le code OpenFoam puis testés en comparaison avec d’autres modèles et avec les données expérimentales présentées par le réseau Engine Combustion Network (ECN). L’avantage de l’application de ces modèles sur les maillages à la résolution grossière a été clairement démontré
This thesis concerns with the Large Eddy Simulations (LES) of fuel sprays in direct-injection engines. Given the high injection velocities of sprays, the resulting turbulent flow may be characterized by energetic intermittent structures at small spatial scales. Therefore, the emphasis in this thesis is put on stochastic simulation of turbulent effects on unresolved scales in the engine relevant conditions. The impact of this effect on spray primary and secondary atomization, on droplets dispersion and evaporation represents the main focus in this thesis. The further assessment and modification of two different approaches, developed recently, was the main objective in this thesis. The first one is addressed to LES-SSAM (stochastic sub-grid acceleration model) approach, in which the Navier-Stokes equations are forced on residual scales. This forcing is given by the Ornstein-Uhlenbeck stochastic process constructed in a way to represent the stochastic properties of the subgrid acceleration, known from the experiment and DNS. In the framework of this approach, with the expression of the acceleration norm modified for the wall-bounded conditions, the first step concerned the simulation of the nozzle internal flow on the coarse grid. The results showed the efficiency of this approach. Another step in this part was to combine LES-SSAM with the interface tracking VOF method in the simulation of the near-field of the spray. The performed assessment of this approach in comparison with measurements and with alternative approaches known from the literature demonstrated a potential of such combination of two methods. The second approach in this thesis, in which the intermittency effects on residual scales are also on target, concerned the stochastic modeling of the secondary breakup, dispersion and evaporation of droplets; introducing the two-way coupling between droplets and a highly turbulent flow. Here, the assessment and further development of stochastic models of droplets represent the main contribution in this thesis. So, the model of the secondary breakup is controlled by the stochastic log-normal process for the viscous dissipation rate. The same stochastic variable is the key variable for the dispersion model of droplets below and above the Kolmogrov scale. The droplet equation of motion for the latter case was modified addressing the significant role to simulation the stochastic direction of the droplet acceleration. Finally, the new stochastic model of the turbulent evaporation, in which the stochastic mixing process is a part of the evaporation model, is also represented in this thesis. The different stochastic models outlined above are assessed in comparison to the state-of-art models available in literature and the experiments of Engine Combustion Network (ECN). The results have shown that stochastic models give a good representation of both macroscopic and microscopic spray characteristics on relatively coarse grids

Expansionen komprimierter Fluide finden breite Anwendung und werden hier mit Hilfe der Schwingungsspektroskopie untersucht. Dabei ist der Aggregatzustand der bei der Zerstäubung entstehenden Partikel von besonderem Interesse. In Abhängigkeit von Stagnationsdruck, Düsentemperatur, und Düsenabstand werden neben Lachgas tert-Butylalkohol und n-Pentan hinsichtlich ihrer Partikelbildung charakterisiert. Die Ergebnisse bilden eine Grundlage für die Expansion nahkritischer Lösungen.

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.

Air, nitrogen, argon and carbon dioxide were used as the atomizing gas in an 'air-assist' spray nozzle to determine the effect of these gases on mean droplet size, number density, velocity and their distributions in kerosene fuel spays and spray flames using a two dimensional phase Doppler interferometer. Data have been obtained with these atomizing gases using a base, air assisted case as a reference, since this is the most commonly used atomizing fluid in almost all applications. Comparisons were made between the gases on a mass and momentum flux basis. Both burning and nonburning sprays were investigated. The results show significant differences in atomization characteristics from the atomizer with different gases and under conditions of constant mass and momentum flux of the gas. The results also show that the presence of oxygen in the air atomized sprays assists in the combustion process, since it produces smaller and faster moving droplets, especially at locations near to the nozzle exit. In nonburning sprays, droplets had similar size and velocity. Lighter gases such as nitrogen more effectively atomized the fuel in comparison to the denser gases. Argon and carbon dioxide produced larger, slower moving droplets than air and nitrogen assisted cases in both the burning and nonburning sprays. Flame photographs revealed the argon and carbon dioxide atomized flames to have greater luminosity than air or nitrogen atomized flames.

Master of Science

Droplet collisions and coalescence play an important role in the droplet size evolution in several processes involving sprays. These processes strongly affect the spray characteristics, making it difficult to control and predict the final droplet size. This thesis has experimentally quantified the effect of the operating conditions and multi-nozzle atomisation on droplet coalescence in a full-cone spray. The coalescence mechanisms that lead to droplet collisions and coalescence in the spray have also been investigated. This experimental investigation was carried using two-fluid atomising nozzles with external mixing for low flowrates, model number Mt” XAOOSROSOA from BETE, which produce a poly-dispersed full-cone spray with an angle of 17-22°. In this work, measurements of the integrated Sauter mean diameter, across the spray at various positions downstream of the nozzle, were obtained by means of a laser diffraction instrument from Malvern Instruments (Malvern 2600C). The measurements were carried out at five different operating conditions for one-nozzle atomisation, where the principal manipulated variables were the water flowrates, from 16.7 to 40 ml/min, and air flowrates, from 11.5 to 14.6 l/min. For the twonozzle atomisation case, two nozzles were pointed towards each other, and the measurements were carried out at the most favourable operating conditions for coalescence, found in the single nozzle experiments. The coalescence mechanisms involved in droplet collisions and coalescence were investigated from the local values of the droplet size, and droplet mean and fluctuating velocities, which were measured using a two component phase—Doppler anemometer RSA 3100 from Aerometrics, Inc. The measurements were carried out for one and two nozzle atomisation, and the droplet trajectories were studied for the two—nozzle atomisation case. This work quantitatively confirms that droplet coalescence depends on the operating conditions used. The most favourable operating conditions for coalescence were found when the droplets generated were smaller, which corresponded to relatively low water flowrates (16.7 ml/min) and high air flowrates (14.6 l/min) for this type of nozzle. At these conditions the greatest size increase (52%) was obtained, the initial droplet size was the smallest (12.6 um), and the estimated droplet number concentration was the highest (6.7x107 m'3). By contrast, a lower size increase (27%) was obtained when larger droplets were generated, at a lower air flowrate of 13.1 l/min and a constant water flowrate of 16.7 ml/min. At these conditions, the initial droplet size was approximately 18 am, and the estimated droplet number concentration was just half (3.5x107 m’3) that for the most favourable operating conditions for coalescence. The results for the two-nozzle atomisation case have quantitatively demonstrated that droplet coalescence is further increased by pointing two nozzles towards each other. The size increase observed for the two nozzle atomisation was significantly greater (86%) compared with that for the single nozzle case (52%). This is the result of an increase in estimated droplet number concentration, from 6.7x107 m"3 to l.3><108 m‘3, which was achieved using two nozzles. Coalescence among droplets interacting can be induced by different mechanisms, such as Brownian motion, shearing flow and turbulence, differential motion by external forces (gravity or centrifugal forces), aerodynamic and electrical forces, and acoustic fields. The mechanism of coalescence was found to be mainly the differential velocity between droplets of different sizes induced by the differential inertia between them, and the differential velocity associated with the eddying fluid motion or turbulence. This differential velocity allows droplets to approach each other, collide and coalescence. Droplet number concentration (inter particle distance) is also an important parameter, as shown by the results for the two-nozzle case compared with those for one nozzle. Finally, a more detailed study at the edge of the spray at two different operating conditions, generating relatively small and large droplets, quantitatively suggested that the greatest size increase observed in this case for the smaller droplets case was because of both a 50% increase in the velocities and a 20% increase in droplet number concentration when generating smaller droplets.

The impact of adjuvants on atomization and patternation of spray mixtures was evaluated. The data showed that certain adjuvants, in particular drift control agents, could potentially detrimentally affect the distribution of herbicide dose across the sprayed swath. The present research sets out to evaluate the impact of this distribution and to seek ways of improving the way researchers and users characterize and possibly mitigate these effects with a view to minimizing the potential detriments and maximizing the efficiency of herbicide active ingredient (AI) utilization. Different formulations of glyphosate with and without a novel polymeric drift control agent (AgRho DR 2000) applied to contrasting broad-leaved and grass weeds were used to evaluate several effects of polymer use. Variables included nozzle type (XR TeeJet extended range flat spray tips, TT Turbo TeeJet wide angle flat spray tips, and TurboDrop air induction nozzles) sampling position (principally under the nozzle centers and under the overlap between two adjacent nozzles), boom height (30, 45, and 60 cm above the target), spray delivery (the volume of spray arriving at the target), spray retention ( the volume of spray actually retained by the target foliage), and herbicide efficacy ( the response of the target weeds to the herbicide dose applied). The data showed that when the polymer was included in the spray mixture, the nozzle used, boom height, presence of the adjuvant, sampling position and certain interactions between these variables were all significant. Spray retention was affected by plant type and retention of coarse sprays was improved by the inclusion of DR 2000. Very coarse sprays reduced glyphosate efficacy on both grasses and broad-leaved weeds although that effect was reduced by use of DR. Addition of drift control agents always ii resulted in increased variability in spray distribution with concomitant increases in both retention and efficacy variability. Variability was shown to decrease with decreasing boom height. There was little correlation between spray delivery and herbicide efficacy. Deposit structure was shown to be a highly important factor in understanding herbicide dose transfer. A novel methodology utilizing digital imaging technology and diversity statistics was developed and evaluated to improve the way we measure and characterize deposit structures. Separation of qualitatively different treatments with similar volumetric distributions was possible. This methodology will be of use to both biologists and fOnTIulation chemists for prediction or explanation of biological results relating to deposit structure. Use of Scanning electron microscopy, and epi-fluorescence microscopy was used to characterize deposit morphology. Differences in deposit morphology were observed and documented leading to a possible explanation for the enhanced glyphosate activity in the presence of DR 2000 iii

Despite the staggering volume of work in the open literature on primary and secondary atomization, there is nothing known that addresses the mechanisms for, and injector geometry implications for, primary atomization within a self-sustained pulsating transonic three-stream injector. Thus, a computational effort involving 86 simulations, including multiple validation exercises, has been executed in order to develop a numerical foundation and then study the effects of nozzle geometry, numerical methodology, grid resolution, modeled domain extent, feed rates, feed flow modulation, feed flow swirl, feed materials, and operating conditions. This is the first undertaking ever reported to disclose the intense details of transonic pulsating flows within the three-stream injector. Metrics for assessment of acoustics and temporal spray character were numerous. Frequency responses among those metrics implied a common pulsation-driving mechanism. It has been discovered that liquid bridging with the production of a liquid fountain and shocklet-like structures in the retracted (pre-filming) zone, along with localized gas-liquid normal pressure gradients, are responsible for bulk pulsations. These findings were never reported in the literature, thus represent an important contribution of this study. Unexpectedly, a new trend for temporal mean droplet size, when normalized by distance from the nozzle, versus distance from the nozzle has been found, which took a common form among all geometries and feed materials tested. Therefore, there is some value to simulate air-water flows, first, to scope general parameters and characteristics, before modeling more computationally challenging slurry flows. This represents an additional contribution of this work not previously reported in the literature. Newly unveiled strong interactions between feed materials, geometry, and feed rate were discovered. Various combinations of inner nozzle retraction and slurry annular thickness were shown to be advantageous, depending on the goals of the injection system. The importance of either geometry variable for three-stream injectors has not been quantified until now. The predictive power of various modeling frameworks has been assessed for the first time. Axi-symmetric (AS) simulations can successfully predict absolute acoustic details; remarkably and surprisingly, AS simulations can also be used for directional indicators of bulk droplet size. This is an especially powerful revelation given the massive reduction in computational requirements for AS models. Reduced order 3-D models are required for better droplet size estimates. A relatively simple eddy-viscosity turbulent model seems to be adequate for predicting droplet sizes for three-stream injectors, in which the primary energy source is bulk pulsations. For larger two-stream systems (atomization energy is sourced in local shear layer instability development), however, a state-of-the-art hybrid model (newly implemented for this effort) appeared to be necessary to capture the resulting droplet scales. Lastly, droplet size and characteristic flow length scale predictions for two open literature non-Newtonian liquid atomizers were made available.
Ph. D.

Dans les moteurs à combustion interne, l'injection de carburant est une phase essentielle pour la préparation du mélange et la combustion. En effet, la structure du jet liquide joue un rôle essentiel pour la qualité du mélange du combustible avec le gaz. Le présent travail porte sur les phénomènes d'atomisation de jet liquides dans les conditions opératoires des moteurs diesel. Dans ces conditions, la morphologie du jet liquide comprend une phase liquide séparée (c'est à dire un noyau liquide) et une phase liquide dispersée (c'est à dire un spray). Ce manuscrit décrit les étapes de développement d'un nouveau modèle d'atomisation, pour un jet liquide à grande vitesse, basée sur une approche eulérienne diphasique. Le phénomène d'atomisation est modélisée par des équations définissant une densité de surface pour le noyau liquide en plus de celle des gouttelettes du spray. Ce nouveau modèle a été couplé avec un système d'équations diphasique et turbulent de type Baer-Nunziato. Le processus de rupture des ligaments et son éclatement subséquent en gouttelettes sont modélisés en utilisant des connaissances rassemblées à partir des expériences disponibles et des simulations numériques précises. Dans la région dense du jet de liquide, l'atomisation primaire est modélisée comme un processus de dispersion en raison de l'étirement turbulent de l'interface, à partir du côté du liquide en plus du côté du gaz. Différents cas tests académiques ont été effectués afin de vérifier la mise en œuvre numérique du modèle dans le code IFP-C3D. Enfin, le modèle est validé avec les résultats DNS récemment publiés dans des conditions typiques de moteurs Diesel à injection directe
In internal combustion engines, the liquid fuel injection is an essential step for the air/fuel mixture preparation and the combustion process. Indeed, the structure of the liquid jet coming out from the injector plays a key role in the proper mixing of the fuel with the gas in the combustion chamber. The present work focuses on the liquid jet atomization phenomena under Diesel engine conditions. Under these conditions, liquid jet morphology includes a separate liquid phase (i.e. a liquid core) and a dispersed liquid phase (i.e. a spray). This manuscript describes the development stages of a new atomization model, for a high speed liquid jet, based on an eulerian two-phase approach. The atomization phenomenon is modeled by defining different surface density equations, for the liquid core and the spray droplets. This new model has been coupled with a turbulent two-phase system of equations of Baer-Nunziato type. The process of ligament breakup and its subsequent breakup into droplets are handled with respect to available experiments and high fidelity numerical simulations. In the dense region of the liquid jet, the atomization is modeled as a dispersion process due to the turbulent stretching of the interface, from the side of liquid in addition to the gas side. Different academic test cases have been performed in order to verify the numerical implementation of the model in the IFP-C3D software. Finally, the model is validated with the recently published DNS results under typical conditions of direct injection Diesel engines

Increasing concerns over the condition of our environment and its long term health have led to the development of greener combustion techniques for use in turbomachinery applications. Lean Direct Injection is an active area of research for how fuel is introduced and burned in the combustor section of a jet engine or land based liquid fuel turbine. Overall lean combustion results in lower NOx emmisions while direct injection insures shorter combustor lengths. Lean Direct Injection and other lean burning combustor designs are susceptible to thermoacoustic instabilities. The SARA or Synchronously Actuated Response Atomizer is a liquid fuel atomizer and supply system designed to allow for the active control of droplet size, cone angle, and mass flow rate. These three parameters have been shown to be important in controlling combustion quality and heat release. This research investigates the capabilities of the SARA design in a series of non-reacting tests. Static and Dynamic tests were performed on the SARA nozzle with a maximum actuation of 400 Hz. Also, a novel use of hot-film anemometry was developed to measure the dynamic flow rate fluctuations.
Master of Science

The present research is an experimental investigation of the atomization of a superheated pressurized liquid jet that is exposed to the ambient pressure due to a sudden depressurization. This phenomena is called flashing and occurs in several industrial environments.

Liquid flashing phenomena holds an interest in many areas of science and engineering. Typical examples one can mention: a) the accidental release of flammable and toxic pressure-liquefied gases in chemical and nuclear industry; the failure of a vessel or pipe in the form of a small hole results in the formation of a two-phase jet containing a mixture of liquid droplets and vapor, b) atomisation improvement in the fuel injector technology, c) flashing mechanism occurrence in expansion devices of refrigerator cycles etc. The interest in flashing events is especially true in the safety field where any unexpected event is undesirable. In case of an accident, flammable or toxic gas clouds are anticipated in close regions of the release because of the sudden phase change .Due to the non-equilibrium nature of the flow in these near field regions, conducting accurate data measurements for droplet size and velocity is a challenging task resulting in scarce data in the very close area.

This research has been carried out at the von Karman Institute (VKI) within the 5th framework of European Commission to fulfill the goal of understanding of source processes in flashing liquids in accidental releases. The program is carried out under name of FLIE (Flashing Liquids in Industrial Environments)(Contract no: EVG1-CT-2000-00025). The specific issues that are presented in this thesis study are the following:a) a comprehensive state of art of the jet break up patterns, spray characteristics and studies related to flashing phenomena; b)flashing jet breakup patterns and accurate characterization of the atomized jet such as droplet diameter size, velocity and temperature evolution through carefully designed laboratory-scale experiments; c) the influence of the initial storage conditions on the final atomized jet; d) a physical model on the droplet transformation and rapid evaporation in aerosol jets.

In order to characterize the atomization of the superheated liquid jet, laser-based optical techniques like Particle Image Velocimetry (PIV), Phase Doppler Anemometry (PDA) are used to obtain information for particle diameter and velocity evolution at various axial and radial distances. Moreover, a high-speed video photography presents the possibility to understand the break-up pattern changes of the simulating liquid namely R-134A jet in function of driving pressure, superheat and discharge nozzle characteristics. Global temperature measurements with an intrusive technique such as thermocouples, non-intrusive measurements with Infrared Thermography are performed. Cases for different initial pressures, temperatures, orifice diameters and length-to-diameter ratios are studied. The break-up patterns, the evolution of the mean droplet size, velocity, RMS, turbulence

intensity and temperature along the radial and axial directions are presented in function of initial parameters. Highly populated drop size and velocity count distributions are provided. Among the initial storage conditions, superheat effect is found to be very important in providing small droplets. A 1-D analytical rapid evaporation model is developed in order to explain the strong temperature decrease during the measurements. A sensitivity analysis of this model is provided.

Doctorat en sciences appliquées
info:eu-repo/semantics/nonPublished

Liquid fragmentation phenomenon is explored from both a fundamental (fully resolved) and an engineering (modeled) perspective. The dual objectives compliment each other by providing an avenue to gain further understanding into fundamental processes of atomization as well as to use the newly acquired knowledge to address practical concerns. A compressible five-equation interface model based on a Roe-type scheme for the simulation of material boundaries between immiscible fluids with arbitrary equation of state is developed and validated. The detailed simulation model accounts for surface-tension, viscous, and body-force effects, in addition to acoustic and convective transport. The material interfaces are considered as diffused zones and a mixture model is given for this transition region. The simulation methodology combines a high-resolution discontinuity capturing method with a low-dissipation central scheme resulting in a hybrid approach for the solution of time- and space-accurate interface problems. Several multi-dimensional test cases are considered over a wide range of physical situations involving capillary, viscosity, and gravity effects with simultaneous presence of large viscosity and density ratios. The model is shown to accurately capture interface dynamics as well as to deal with dynamic appearance and disappearance of material boundaries. Simulation of atomization processes and its interaction with the flow field in practical devices is the secondary objective of this study. Three modeling requirements are identified to perform Large-Eddy Simulation (LES) of spray combustion in engineering devices. In concurrence with these requirements, LES of an experimental liquid-fueled Lean Direct Injection (LDI) combustor is performed using a subgrid mixing and combustion model. This approach has no adjustable parameters and the entire flow-path through the inlet swirl vanes is resolved. The inclusion of the atomization aspects within LES eliminates the need to specify dispersed-phase size-velocity correlations at the inflow boundary. Kelvin-Helmholtz (or aerodynamic) breakup model by Reitz is adopted for the combustor simulation. Two simulations (with and without breakup) are performed and compared with measurements of Cai et al. Time-averaged velocity prediction comparison for both gas- and liquid-phase with available data show reasonable agreement. The major impact of breakup is on the fuel evaporation in the vicinity of the injector. Further downstream, a wide range of drop sizes are recovered by the breakup simulation and produces similar spray quality as in the no-breakup case.

The droplet size distribution produced by the atomization system within a spray d1yer is vital in defining the output particle size distribution, which in turn can strongly influence the subsequent flow, dissolution, ingestion or bioavailability of the final product. This thesis explores the capability of a simple atomizing system for the production of narrow droplet size distributions within an industrially-relevant spray dryer. The combination of constrained droplet size distribution, defined solute concentration and selected processing conditions has been evaluated to determine the influence upon the physicochemical properties of a model active pharmaceutical ingredient, and the dosing characteristics of formulated products. The flow focusing atomizer was demonstrated to be capable of generating distinct droplet size distributions, and able to produce similar performance in the dynamic environment within a spray dryer. Experiments conducted using the inhaled corticosteroid mometasone furoate indicate that droplet size, concentration and drying conditions are important in determining the particle size, texture and thermodynamic properties of the output powder. Analysis of formulations containing the flow-focused mometasone furoate particles defined a clear link between the impaction profile in a cascade impactor and droplet size and spray d1ying conditions.

Os objetivos do estudo foram elaborar iogurtes com a cultura tradicional e a cultura probiótica de Bifidobacterium animalis subsp. lactis, desidratar os produtos em spray dryer utilizando maltodextrina como carreador e caracterizar os pós obtidos, bem como determinar a resistência dos probióticos ao processo de atomização. O presente estudo avaliou três temperaturas de entrada do ar de secagem (130, 150 e 170°C) em iogurtes com duas diferentes concentrações de maltodextrina (10 e 20%), totalizando 6 tratamentos: T1 (10%malto/130°C), T2 (20%malto/130°C), T3 (10%malto/150°C), T4 (20%malto/150°C), T5 (10%malto/170°C) e T6 (20%malto/170°C). A secagem do iogurte foi realizada em spray dryer piloto e a enumeração das células viáveis de Bifidobacterium animalis subsp. lactis foi realizada através de plaqueamento em profundidade. Os pós apresentaram baixos valores de umidade e elevada higroscopicidade. A atividade de água (Aw) dos pós variou de 0,09 a 0,19 e aumentou após 30 dias de armazenamento, comprovando o caráter higroscópico dos pós obtidos. Verificou-se que após a desidratação dos iogurtes, apesar deles apresentarem contagens inferiores que os produtos integrais, ainda apresentaram contagens acima de 106 UFC/g, ou seja, ainda estavam dentro do limite estabelecido pela legislação para um produto ser considerado probiótico. Os tratamentos que passaram por maiores temperaturas durante o processamento de secagem (T5 e T6) foram os que tiveram maiores perdas de micro-organismos probióticos, sugerindo que as altas temperaturas exerceram forte influência na viabilidade dos probióticos. O T1 obteve maiores contagens do micro-organismo analisado, com contagens acima de 106 UFC/g, com até 60 dias de armazenamento, indicando ser o melhor tratamento entre os estudados, em relação à obtenção de um iogurte caprino probiótico em pó com maior período de vida de prateleira. De maneira geral, conclui-se que o processo de atomização possibilitou a obtenção de iogurte de leite de cabra em pó estável do ponto de vista microbiológico. Além disso, obteve-se um produto que pode ser uma alternativa para incrementar o consumo de leite de cabra, bem como o de probióticos.
The aim of this study was to develop yogurts with the traditional culture and Bifidobacterium animalis subsp. lactis probiotic culture, dehydrate products in spray drying using maltodextrin as a carrier and characterize the powders, as well as determining the resistance of probiotics to atomization process. The present study evaluated three different inlet air temperatures of spray dryer (130, 150 and 170°C) in yoghurts with two different maltodextrin concentrations (10 and 20%), totaling six treatments: T1 (10%malto/130°C), T2 (20%malto/130°C), T3 (10%malto/150°C), T4 (20%malto/150°C), T5 (10%malto/170°C) e T6 (20%malto/170°C). The yogurt drying was performed in a pilot spray dryer and the viable cells of Bifidobacterium animalis subsp. lactis enumeration was performed by pour plate. The powders showed low levels of humidity and high hygroscopicity. The water activity (Aw) of the powders ranged from 0.09 to 0.19 and increased after 30 days of storage, showing the hygroscopic powders character. It was found that after yogurt dehydration, despite their counts were less than integral products, still had counts above 106 CFU/g, therefore were still within regulation limits for a product to be considered as probiotic. The treatments that have undergone higher temperatures during the drying process (T5 and T6) were those who had higher losses of probiotic microorganisms, suggesting that high temperatures had a strong influence on the viability of probiotics. The T1 (130°C/10%) obtained higher counts of the microorganism analyzed, with counts above 106 CFU/g, during 60 days of storage, indicating that is the best treatment among those studied in relation to obtaining a goat probiotic yoghurt powder with longer shelf life. In general, it is concluded that the atomization process allows the obtention of stable goat milk yogurt powder, in a microbiological point of view. Furthermore, it was obtained a product that can be an alternative for increasing the consumption of goat milk as well as probiotics.

The disintegration of a liquid jet emerging from a nozzle has been under investigation for several decades. A direct consequence of the liquid jet disintegration process is droplet formation. The breakup of a liquid jet into discrete droplets can be brought about by the use of a diverse forcing mechanism. Cavitation has been thought to assist the atomization process. Previous experimental studies, however, have dealt with cavitation as a secondary phenomenon assisting the primary atomization mechanism. In this dissertation, the role of the energy created by the collapse of cavitation bubbles, together with the liquid pressure perturbation is explicitly configured as a principal mechanism for the disintegration of the liquid jet. A prototype of an atomizer that uses this concept as a primary atomization mechanism was developed and experimentally tested using water as working fluid. The atomizer fabrication process and the experimental characterization results are presented. The parameters tested include liquid injection pressure, ultrasonic horn tip frequency, and the liquid flow rate. The experimental results obtained demonstrate improvement in the atomization of water. To fully characterize the new atomizer, a novel infrared thermography-based technique for the characterization and visualization of liquid sprays was developed. The technique was tested on the new atomizer and two automotive paint applicators. The technique uses an infrared thermography-based measurement in which a uniformly heated background acts as a thermal radiation source, and an infrared camera as the receiver. The infrared energy emitted by the source in traveling through the spray is attenuated by the presence of the droplets. The infrared intensity is captured by the receiver showing the attenuation in the image as a result of the presence of the spray. The captured thermal image is used to study detailed macroscopic features of the spray flow field and the evolution of the droplets as they are transferred from the applicator to the target surface. In addition, the thermal image is post-processed using theoretical and empirical equations to extract information from which the liquid volume fraction and number density within the spray are estimated.

A azitromicina é um antibiótico da classe dos macrolídeos utilizado principalmente no tratamento de infecções de pele e do trato respiratório. Suas propriedades farmacocinéticas lhe conferem ampla distribuição e acumulação tecidual. Devido à sua baixa biodisponibilidade (37%) e instabilidade ao pH ácido, foram investigadas alternativas para o desenvolvimento de microparticulados com propriedade gastro-resistente para aplicação em diferentes formas farmacêuticas. As micropartículas de azitromicina foram obtidas pela técnica de secagem por pulverização (spray-drying) com o polímero pH-dependente Eudragit® L30-D55 e trietilcitrato como plastificante. O desenvolvimento experimental seguiu um planejamento fatorial tipo Box-Behnken de quatro variáveis em três níveis. As micropartículas foram caracterizadas pela densidade aparente, densidade compactada, fator de Hausner, índice de Carr, ângulo de repouso e cromatografia líquida. O processo de secagem foi avaliado pelo rendimento e recuperação da azitromicina. A partir das características físicas do microparticulado e o rendimento do processo foram escolhidas três micropartículas para avaliação mais detalhada, contendo 15, 30 e 45% de polímero de revestimento sobre a concentração de azitromicina dihidrato. Estas foram avaliadas por difração de raios-X, calorimetria, infra-vermelho, microscopia eletrônica de varredura e teste de dissolução. Os fatores que mais influenciaram o processo de secagem e obtenção das micropartículas foram a concentração de Eudragit® L30D-55 e a temperatura do ar de secagem, sendo que, quanto menor a concentração de polímero e maior a temperatura de secagem melhores foram os rendimentos do processo. A recuperação da azitromicina para todas as micropartículas foram superiores a 80%. As micropartículas apresentaram baixa densidade aparente, porém com propriedades de fluxo que variaram de bom a excelente na maioria das condições de preparação. As avaliações por calorimetria e difração de raios-X sugerem que a azitromicina está encapsulada. Isto pode ser confirmado por meio das imagens da microscopia eletrônica de varredura que revelaram estruturas esféricas e uniformes. As micropartículas com maiores concentrações de polímero apresentaram as melhores características com proteção gástrica, acima de 50%, e são candidatas promissoras para o desenvolvimento de formas farmacêuticas sólidas orais gastro-resistentes contendo azitromicina.
The azithromycin is an antibiotic of the macrolide class used in cutaneous and respiratory infections treatments. Its pharmacokinetics properties confer a large cell distribution and accumulation in tissues. Due to its small bioavailability (37%) and acid pH instability, alternatives to the development of gastric-resistance microparticles to be uses in different pharmaceuticals forms were investigated. Azithromycin microparticles were obtained through spray-drying technique with the pH-dependent polymer Eudragit® L30-D55 and triethyl citrate as plasticizer. The experimental development followed a factorial planning Box-Behnken with four factors in three levels. The microparticles were characterized by bulk density, tapped density, Hausner ratio, Carr index, repose angle and HPLC. The drying process was evaluated by azithromycin yield and recover. From the microparticles physics parameters and the process yield, three samples were chosen to be better evaluated. The samples contained 15, 30 and 45% of coat polymer over azithromycin dehydrate. They were evaluated by X-ray powder diffraction (XRPD), calorimetry, infrared, surface electronic microscopy (SEM) and dissolution assay. The factors that most affected the drying process and the microparticles development were Eudragit® L30D-55 concentration and drying air temperature, and lower was the polymer amount and higher was the drying air temperature, better was the process yield. The azithromycin recover for all microparticles were superior to 80%. The microparticles presented low bulk density, but good flow properties for the great part of the samples. The calorimetry and the XRPD evaluations suggest that azithromycin is encapsulated, what can be confirmed through SEM pictures, which showed uniform and spherical structures. The microparticles with higher polymers concentrations showed the best characteristics with gastric protection above 50% and they are promising candidates to the development of oral solid pharmaceutical forms containing gastro-resistant azithromycin.

Este trabalho investigou a deposição dos filmes finos de ferritas de lantânio dopado com estrôncio e cobalto La1-xSrxCo1-yFeyO3-δ (LSCF) através da técnica de spray pirólise empregando precursores inorgânicos e silício como substrato. Foram utilizados água e etanol na proporção de (3:1) como solvente na preparação da solução a ser aspergida. As deposições foram realizadas em diferentes temperaturas: 130ºC, 150ºC, 170ºC e 200ºC. Os filmes de LSCF antes e após o tratamento térmico foram analisados por difração de raios X (DRX) e microscopia eletrônica de varredura (MEV). O spray de deposição dos filmes em função das soluções com diferentes proporções de etanol (0%, 50%, 75% e 100%) foi investigado utilizado uma câmera rápida de alta resolução e um laser com comprimento de onda 532 nm. Os resultados evidenciaram que a temperatura de ebulição do solvente e de deposição são parâmetros importantes na determinação da morfologia dos filmes. Maiores teores de etanol promovem um aumento nos valores do ângulo de cone e que a propriedade da tensão superficial exerce significativamente maior influência nos aspectos que envolvem a estrutura do spray, a atomização do líquido, formação de gotas e no ponto de ruptura das soluções Os resultados de DRX apresentaram a fase cristalina majoritária de LSCF após o tratamento térmico a 750°C por 2 horas e a fase secundária de La2O3. As imagens de MEV evidenciaram a formação de filmes para as distâncias de deposição de 120 mm, embora apenas na temperatura de 200°C apresentasse uma estrutura porosa antes e após o tratamento térmico. A presença de interdifusão e microporosidade foi observada para esse filme. Para as diferentes composições de LSCF nas condições de 120 mm e a 200°C, apenas as composições de La1-xSrxCo0,2Fe0,8 com x = 0,6, 0,9 e 0,7 apresentaram estrutura porosa após o tratamento térmico. O filme de La0,8Sr0,2Co0,2Fe0,8 apresentou maior condutividade iônica/elétrica e energia de ativação. A presença de fases secundárias (La2O3, SrO2 e La2CoO3) contribuiu para a redução da condutividade elétrica e iônica e a energia de ativação.
This work investigated the deposition conditions of thin lanthanum ferrite films doped with strontium and cobalt – La1-xSrxCo1-yFeyO3-δ (LSCF) through the spray pyrolysis technique using inorganic precursors and silicon as a substrate. Water and ethanol were used in the proportion (3:1) as solvent in the preparation of the solution to be sprinkled. The depositions were carried out at different temperatures: 130°C, 150°C, 170°C and 200°C. The films of LSCF were analyzed by X-ray diffraction and scanning electron microscopy before and after thermal treatment. The XRD results showed the obtainment of crystalline phase of LSCF after thermal treatment at 750°C/2hours and secondary phase of La2O3. The sprays of solutions with different proportions of ethanol (0%, 50%, 75% and 100%) were analyzed using a fast high-resolution camera and a laser with a wavelength of 532nm.The results showed those deposition temperature and solvente boiling points are important parameters to determine the thin morphology Higher ethanol levels promote an increase in the cone angle values and that the property of the surface tension exerts significantly greater influence on issues involving the structure of the spray, the atomization of the liquid, the droplets formation and the breaking point of the solutions. The XRD results showed the obtainment of majority crystalline phase of LSCF after thermal treatment at 750°C/2 hours with the presence of secondary phase of La2O3. The SEM images showed formation of films for the distance of 120 mm, although only at the temperature of 200°C it presents porous morphology before and after thermal treatment. The presence of interdiffusion and microporosity was observed for this film. For different LSCF compositions under the conditions of 120 mm and 200°C, the films of La1-xSrx Co0,2Fe0,8 with x = 0,6; 0,9 and 0,7 presented porous morphology after thermal treatment. Otherwise, the film of La0,8Sr0,2Co0,2Fe0,8 showed the highest ionic/electrical conductivity. The presence of secondary phase (La2O3, SrO2 and La2CoO3) evidenced by XRD promoted a decrease in ionic/electrical conductivity and in the activation energy.

Un moyen d'accroître l'efficacité et de réduire la pollution dans les domaines du transport et de l'énergie consiste à concevoir des injecteurs de carburant produisant une meilleure atomisation. Au cours de cette thèse, des expériences ont été effectuées sur un injecteur airblast souvent utilisé dans les turbines à gaz. Pour réaliser ces expérimentations, un dispositif modèle en configuration annulaire a été créé afin d'étudier le cisaillement d'un film d'eau soumis à un écoulement d'air interne à forte vitesse. La technique d'imagerie rapide par ombroscopie a été utilisée pour analyser le développement du film liquide (fréquence et célérité des ondes) et l'atomisation de la nappe en sortie d'injecteur (modes de rupture). La modification des paramètres d'injection (vitesse des écoulements) a révélé un lien entre la topologie du film liquide et le régime d'atomisation primaire. Finalement, à titre exploratoire, l'influence de la géométrie de l'injecteur (longueur de préfilm) sur le mode d'atomisation primaire a également été mise en évidence
One way to increase efficiency and reduce pollution in the transportation and energy domain is designing fuel injectors with better atomization. In this thesis, experiments were performed on a prefilming airblast atomizer often used in gas turbines. For this purpose, a model device with a cylindrical configuration was created to study the shearing of a film of water subjected to an internal high speed air flow. High speed shadowgraphy technique was used to analyse the development of the liquid film (frequency and wave celerity) and the atomization of the sheet at the injector outlet (breakup mode). The modification of the injection parameters (velocity of flows) revealed a link between the topology of the liquid film and the primary atomization regime. Finally,the influence of the geometry of the injector (prefilming length) about the mode of primary atomization was also highlighted with an exploratory study

O Brasil é um grande produtor e consumidor de café. A bebida é a segunda mais consumida no país. O café solúvel é uma alternativa para o consumo diário visto que dispensa todo aparato de preparo da bebida, sendo necessária apenas água quente. O tamanho de partícula seca está diretamente ligado à densidade aparente do produto. Foram realizados testes de secagem de extrato de café em spray-dryer com atomização por disco rotativo e fluxo co-corrente variando-se a temperatura do ar de entrada (140 - 170 °C) e rotação do disco atomizador (27000 - 33000 rpm). A análise de tamanho de partícula mostrou tendência de partículas maiores quanto menor a temperatura de secagem (26 - 36 µm). Notou-se a influência tanto da temperatura quanto da rotação do disco atomizador na análise da atividade de água. A densidade aparente não se mostrou significativamente alterada pelos parâmetros estudados nos experimentos.
Brazil is a big producer and consumer of coffee. The drink is the second most consumed in the country. Instant coffee is an alternative for daily consumption since it dispenses with all the apparatuses for preparing the drink, being necessary only hot water. The dry particle size is directly related to the bulk density of the product. Coffee extract drying tests were made in a co-current flow spray-dryer with atomization by disc with a varying inlet air temperature (140 - 170 °C) and disc velocity (27000 - 33000 rpm). The particle size analysis has shown the tendency for bigger particles to form at lower drying temperatures (26 - 36 µm). The influence of inlet air temperature and the atomizer disc spin velocity were noted in water activity analysis. The bulk density was not significantly modified by the parameters studied in the experiments.

La projection plasma de suspensions (SPS) est un procédé de revêtement de surface qui consiste à injecter une suspension (particules solides d’environ 1 μm ou moins, dispersées dans une phase liquide) dans un jet de plasma énergétique. Les particules sont chauffées, accélérées en direction d’un substrat, écrasées et soumises à une solidification très rapide (de l’ordre de 106 K.s-1). Couche après couche, un dépôt se forme en surface du substrat et lui apporte de nouvelles propriétés fonctionnelles. Cette variante de la projection plasma conventionnelle permet la fabrication de revêtements avec des épaisseurs plus fines de quelques dizaines de μm et une échelle microstructurale réduite, pouvant conduire à améliorer, par exemple, les performances de dureté ou de conductivité thermique des dépôts. Bien que ce procédé soit étudié depuis le milieu des années 1990 et connaisse un intérêt grandissant, les applications industrielles ne sont pas finalisées et leur développement nécessite d’être poursuivi. En effet, l’injection d’une suspension dans un jet thermique conduit à des phénomènes complexes tels que la fragmentation des gouttes de suspension ou encore l’évaporation de la phase liquide. A ce jour, ces mécanismes ne sont pas parfaitement compris et maîtrisés et méritent d’être étudiés pour comprendre les interactions de ces fines particules avec le plasma. Les travaux décrits dans ce mémoire s’intéressent au cas de la projection SPS de céramiques avec un atomiseur bi-fluide comme système d’injection. Deux matériaux ont été choisis : l’alumine, connue pour sa difficulté à être projetée conventionnellement et dont la formation de phases cristallines particulières constitue une source d’informations sur l’histoire thermique des particules, ainsi que l’yttrine, qui permet de confirmer les tendances observées pour l’alumine. Dans un premier temps, l’optimisation de l’injection de la suspension a été effectuée en travaillant sur deux axes. Le premier axe concerne la formulation des suspensions, qui a conduit à l’obtention, avec différentes phases liquides, de suspensions stables et dispersées, de propriétés parfaitement connues. De telles suspensions assurent une reproductibilité du procédé à ce niveau et limitent le bouchage du système d’injection. Le deuxième axe porte sur la conception mécanique en trois étapes d’un atomiseur pneumatique approprié au procédé SPS. Cette étude a commencé par la caractérisation d’une buse commerciale notamment par des tests d’injection de suspension dans le plasma. Les tests étant peu concluants, l’étude s’est poursuivie par la mise au point d’une nouvelle géométrie d’atomiseur inspirée du modèle commercial. Les essais ont conduit à la réalisation de cordons et de dépôts satisfaisants. Cette étude s’est terminée enfin par l’optimisation de sa géométrie grâce à la mise en évidence de l’influence de plusieurs paramètres-clé sur les caractéristiques du jet atomisé. Dans un second temps, des outils de diagnostic ont été mis en oeuvre pour mesurer la qualité de l’injection. Le jet de suspension a été caractérisé en termes de géométrie et de tailles de gouttes, respectivement par ombroscopie et diffraction laser. L’ombroscopie a été réutilisée pour l’optimisation de l’injection de la suspension dans le plasma en permettant le réglage en temps réel des pressions d’entrée de l’atomiseur. Les propriétés des particules en vol ont ensuite été étudiées grâce à des collectes de particules sur substrat et à la vélocimétrie par images de particules (PIV). Cet outil a apporté des informations complémentaires sur l’injection de la suspension. Enfin, les revêtements obtenus ont été caractérisés en termes de morphologie (MEB), taux de porosité (analyse d’images MEB et USAXS) et de phases cristallines (DRX et EBSD). Le couplage des informations obtenues entre ces différentes techniques a permis de faire ressortir le rôle de la phase liquide et de la charge massique sur la microstructure
Suspension plasma spray (SPS) is a surface coating process that consists in injecting a suspension (solid particles of about 1 μm or less, dispersed in a liquid phase) in a high-energy plasma flow. Particles are heated, accelerated towards a substrate, flattened and submitted to a rapid solidification (order of 106 K.s-1). Layer after layer, a coating is formed on the substrate surface and brings new functional properties. This variation of the conventional plasma spray process allows the manufacturing of coatings with finer thickness of few tens of μm and a reduced structural scale that can lead to improved coating properties, like hardness or thermal conductivity. Even though this process has been studied since the middle of the 1990’S and known a fast-growing interest, industrial applications are not finalized and their development needs to be pursued. Indeed, the suspension injection in a thermal jet leads to complex phenomena such as suspension droplet fragmentation or liquid phase evaporation. Up to now, these mechanisms are not perfectly understood and controlled and deserve to be further studied to understand interactions between these fine particles and the plasma. This thesis focuses on the SPS process with ceramic suspensions and a twin-fluid nozzle as injection system. Two materials were chosen: alumina, known for its difficulty to be conventionally sprayed and whose crystalline phase formation represents a source of information about particle thermal history, and also yttria, in order to confirm the tendencies observed for alumina. Firstly, the suspension injection was optimized by working on two areas. The first area concerns suspension formulation. This led to obtain, with different liquid phases, stable and dispersed suspensions, whose properties are perfectly known. Such suspensions ensure reproducibility of the process at this level and limit the risk of injection system clogging. The second area is about the three-step mechanical conception of a pneumatic atomizer, adapted to the SPS process. This study began with the characterization of a commercial nozzle, in particular by testing the suspension injection into a plasma flow. Tests being little convincing, the study was carried on with the development of a new atomizer geometry, inspired from the commercial model. Trials drove to the manufacturing of satisfying spray beads and coatings. This study was finally completed with the optimization of this new geometry by highlighting the influence of several key parameters on the atomized jet features. Secondly, diagnostic tools were implemented to qualify the injection. Suspension jet was characterized in terms of geometry and droplet sizes, using respectively shadowgraphy and laser diffraction. Shadowgraphy was used again for optimizing the suspension injection into plasma by allowing the adjustment in real time of inlet atomizer pressures. In-flight particle properties were then studied thanks to particle collection onto a substrate and particle image velocimetry (PIV). This tool also provided additional information on the suspension injection. Finally, the resulting coatings were characterized in terms of morphology (SEM), porosity rate (SEM image analysis and USAXS) and crystalline phases (DRX and EBSD). The cross-checking of the information obtained with all these techniques brought out the role of the suspension liquid phase and of the mass load on the coating microstructure. These works contributed to enhance the knowledge about the SPS process and justified the use of a twin-fluid nozzle to obtain specific microstructures of coatings, whose functional characterizations have still to be done

El proceso de atomización desde una vena o lámina líquida hasta multitud de gotas dispersas en un medio gaseoso ha sido un fenómeno de interés desde hace varias décadas, especialmente en el campo de los motores de combustión interna alternativos. Multitud de estudios experimentales han sido publicados al respecto, pues una buena mezcla de aire-combustible asegura una evaporación y combustión mucho más eficientes, aumentando la potencia del motor y reduciendo la cantidad de contaminantes emitidos. Con el auge de las técnicas computacionales, muchos modelos han sido desarrollados para estudiar este proceso de atomización y mezcla. Uno de los últimos modelos que han aparecido es el llamado ELSA (Eulerian-Lagrangian Spray Atomization), que utiliza un modelo Euleriano para la parte densa del chorro y cambia a un modelo Lagrangiano cuando la concentración de líquido es suficientemente pequeña, aprovechando de esta manera las ventajas de ambos. En el presente trabajo se ha desarrollado un modelo puramente Euleriano para estudiar la influencia de la geometría interna de la tobera de inyección en el proceso de atomización y mezcla. Se ha estudiado únicamente el proceso de inyección diésel. Este modelo permite resolver en un único dominio el flujo interno y el externo, evitando así las comunes simplificaciones y limitaciones de la interpolación entre ambos dominios resueltos por separado. Los resultados actuales son prometedores, el modelo predice con un error aceptable la penetración del chorro, el flujo másico y de cantidad de movimiento, los perfiles de velocidad y concentración, así como otros parámetros característicos del chorro.
Martí Gómez-Aldaraví, P. (2014). DEVELOPMENT OF A COMPUTATIONAL MODEL FOR A SIMULTANEOUS SIMULATION OF INTERNAL FLOW AND SPRAY BREAK-UP OF THE DIESEL INJECTION PROCESS [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/43719
TESIS
Premiado

L’atomisation d’un jet ou d’une nappe liquide assistée par un courant gazeux rapide est couramment utilisée dans l’industrie ainsi qu’en propulsion aéronautique (turboréacteur) et spatiale (moteur-fusée cryotechnique). Plusieurs processus permettent la fragmentation de la structure cohérente liquide en gouttes. L’épluchage, qui intervient à courte distance en aval de l’injection, a été assez largement étudié (Marmottant et Villermaux 2004, Hong et al 2004) et les mécanismes sont assez bien décrits. En revanche, l’origine des instabilités large échelle – ou « flapping » - intervenant plus loin en aval, instabilités qui sont à l’origine de la production de large gouttes, reste mal comprise. Ceci est particulièrement vrai pour des jets cylindriques qui, contrairement au cas de nappes, ont fait l’objet de très peu d’études. Nous nous sommes donc attachés à comprendre l’origine du « flapping », à analyser ses liens avec les instabilités interfaciales de cisaillement, et à quantifier son impact sur la structure du jet ainsi que sur les gouttes produites. Pour cela, des expériences ont été menées en eau/air sur de larges plages de paramètres, aussi bien en termes de vitesses phasiques que des dimensions des veines gaz et liquide. Un soin particulier a été apporté au contrôle des écoulements internes.Pour l’ensemble des géométries, nous avons montré que la longueur du dard liquide est pilotée par le battement large échelle et non par le processus d’épluchage. Par ailleurs, la longueur de brisure présente une décroissance marquée avec la vitesse gaz, puis reste constante au delà d’une vitesse gaz critique. Un modèle a été proposé pour ce comportement asymptotique dans lequel la longueur de brisure est pilotée par le rapport de la vitesse liquide d’injection à une vitesse capillaire construite sur le diamètre liquide.La technique de mesure de la fréquence du battement large échelle mise en œuvre à partir d’images acquises par ombroscopie s’est avérée opérationnelle sur toute la plage de vitesses gaz considérées. Cette fréquence, qui ne varie pas spatialement, présente deux comportements : un premier où elle augmente avec la vitesse gaz, et un second où elle reste indépendante de la vitesse gaz. Ce second régime n’est pas mentionné dans la littérature. Pour le premier régime, le lien entre flapping et instabilité de cisaillement a été démontré en s’appuyant notamment sur des analyses de stabilité. Le nombre de Strouhal associé est piloté par le cisaillement côté gaz. La dépendance de la fréquence de battement à l’épaisseur de vorticité côté gaz est ainsi établie lorsque l’instabilité de cisaillement est pilotée par un mécanisme inviscide. Pour le second régime, le caractère opportuniste du flapping a été démontré l’aide d’une expérience de forçage : le flapping amplifie des structures liquides de longueur d’onde plus grande que celle associée à l’instabilité de cisaillement. Un nombre de Strouhal construit sur le diamètre liquide du jet et la vitesse du jet liquide à la distance de brisure a été proposé. Enfin, le rapport du diamètre du jet liquide à la longueur d’onde de l’instabilité de cisaillement semble pertinent pour définir la frontière entre ces deux régimes.Les tailles des gouttes produites sur l’axe de symétrie ont été mesurées à l’aide d’une sonde optique. Il apparaît que la distribution granulométrique évolue fortement avec la vitesse gaz, et qu’elle est multi-modale, ce qui traduit la présence de plusieurs mécanismes de brisure. La taille moyenne des gouttes décroit globalement comme UG-2, dans la limite de forts nombres de Weber aérodynamique. Cette taille moyenne s’avère aussi très sensible à la géométrie : elle diminue lorsque l’épaisseur gaz augmente jusqu’à atteindre une valeur plancher, et elle croît avec le diamètre liquide
Jet or sheet atomized by a fast coaxial gas jet is currently used in industry, like aeronautical propulsion (turbofan) or spatial propulsion (cryotechnic rocket engine). Many physical processes allows liquid coherent structure fragmentation into drops. Stripping, which appears downstream near injector, has been largely studied (Marmottant et Villermaux 2004, Hong & al 2004), mecanisms has been correctly described.However, the origin of large scale - or 'flapping' instabilities - intervening further downstream, instabilities that are causing the production of large drops, remains poorly understood. This is particularly true for cylindrical jets which, unlike the case of sheets, have been the subject of very few studies. We are therefore committed to understand the origin of the "flapping", to analyze its relationship with interfacial shear instabilities, and to quantify its impact on the structure of the jet as well as on the drops produced. For this, experiments were carried out in water/air on wide set of parameters, both in terms of phasic speed than the dimensions of the gas gap and liquid diameter. Special care were made to the internal flow control.For all the geometries, we showed that the length of the liquid cone is driven by the large scale displacements and not by the stripping process. Furthermore, the length of brokenness jet presents a decline marked with the gas speed, then remains constant beyond a critical gas speed. A model was proposed for this asymptotic behavior in which the break-up length is driven by the report of the liquid injection speed to a capillary speed built on the liquid diameter.Measurement of the frequency of large scale displacement technology has been implemented from images acquired by shadowgraphy proved operational over the gas velocity range considered. This frequency, which varies not spatially, present two behaviors: a first where it increases with the speed of the gas, and a second where it remains independent of the gas speed. This second scheme is not mentioned in the literature. For the original plan, the link between flapping and shear instability has been demonstrated based on analyses of stability. The associated Strouhal number is controlled by the shear gas side. The dependence of the frequency of heartbeat to the thickness of vorticity gas side is thus established when shear instability is driven by an inviscide mechanism. For the second scheme, the opportunistic nature of the flapping has been demonstrated using forcing experience: the flapping amplifies liquid structures of wavelength greater than those associated with shear instability. A Strouhal number built on liquid jet diameter and the speed of the liquid jet at break distance has been proposed. Finally, the ratio of the diameter of the liquid jet at the wavelength of the shear instability seems relevant to define the border between these two regimes.Sizes drops produced on the symmetry axis were measured using an optical probe. It appears that granulometric distribution is evolving strongly with speed gas, and it is multi-modal, reflecting the presence of several mechanisms of brokenness. The average size of the drops decreases overall as UG - 2, in the limit of strong numbers of aerodynamic Weber. This medium size is also very sensitive to geometry: it decreases when the thickness of the gas increases until it reaches a floor value, and it grows with the liquid diameter. Finally, by forcing large amplitude lateral displacement, the average radial distribution of sizes of drops has been made much more homogeneous, and the average size of the drops on the axis has been reduced by a factor of 2. These results therefore open opportunities in terms of control of atomization

La réglementation plus stricte sur les émissions des moteurs aéronautiques posée par l’OACI nécessite des outils de conception toujours plus prédictifs. La distribution des diamètres des gouttelettes produites pendant le processus de combustion est un paramètre clé pour prédire les émissions de polluants libérées pendant le processus de combustion. L’étude du phénomène d’atomisation avec son caractère multi-échelles est donc un défi pertinent et important. Pour cette raison, les objectifs de cette étude sont: tout d’abord de passer en revue les modèles existants dans la littérature, pour comprendre leurs caractéristiques clés, afin de définir une classification qui donne des lignes directrices sur les choix de modélisation; deuxièmement, d’appliquer des approches plus compatibles avec les problématiques industrielles du secteur sur une configuration aéronautique, afin de proposer une amélioration des outils de conception disponibles. Une classification systématique des modèles est effectuée par rapport à l’échelle de longueur considérée pour représenter les caractéristiques de l’interface. De ce point de vue, il est possible de distinguer deux types d’approches: la représentation des phases séparées et la représentation des phases mixtes. Les approches diffuse interface appartiennent à la seconde catégorie avec de nombreuses autres approches, compressibles et incompressibles, qui partagent la même caractéristique: elles considèrent un mélange qui contient les deux phases. Une configuration de nappe de liquide assistée par air a été conçue pour tester différents modèles afin de définir une mesure de comparaison. Deux représentations différentes de l’approche sharp interface (ARCHER et InterFoam ), deux représentations de l’approche diffuse interface (CEDRE et ELSA ) et un modèle hybride (ICMelsa ) ont été considérés dans cette étude. Une comparaison en deux parties basée sur des quantités statistiques a été proposée. Une première partie appelée "étude classique", compare les statistiques du premier ordre montrant que toutes les approches conduisent à des résultats très similaires dès qu’un certain niveau de résolution de maillage est atteint. Au contraire, les secondes statistiques présentent des différences notables. Ces résultats motivent une deuxième partie appelée «analyse de phase» pour étudier le lien entre la représentation à petite échelle de l’interface et la statique du second ordre. En particulier, la variance du marqueur de phase et le niveau de ségrégation associé se révèlent être des indicateurs sensibles de la description del’interface. Une analyse de signal 1D montre qu’ils peuvent être utilisés pour détecter toutécart par rapport à la représentation des phases séparées, puis l’importance de la variance de l’indicateur de phase est démontrée sur d’autres statistiques de second ordre: les composantes de contrainte de Reynolds et le flux de liquide turbulent. Ainsi, les statistiques du second ordre sont partiellement décrites avec des représentations directes de phases mixtes et nécessitent un modèle complémentaire pour être entièrement récupérées. Une première tentative, basée sur une approche linéaire, est proposée pour modéliser le niveau de ségrégation de la représentation des phases mixtes, construit en filtrant un signal totalement ségrégé à une échelle donnée
The more stringent regulation about aeronautical engines emission posed by ICAO requires always more predictive design tools. The droplets diameter distribution produced during the atomization process is a key parameter in order to predict the pollutant emission released during the combustion process. Thus the study of the atomization phenomenon with itsmulti-scale nature is a relevant and an important challenge. For this reason the objective of this work are: first to review the existent models in the literature to understand their key features in order to define a classification that gives guidelines on the modeling choices; second to apply industrial oriented approaches on an aeronautical configuration, in order to propose an improvement of the available design tools. A systematic classification of the models is done with respect to the length-scale considered to represent the interface characteristics. From this point of view, it is possible to distinguish two kinds of approaches: the separated phases representation and the mixed phases representation. The diffuse interface approaches belongs on the second category together with many other approaches, compressible and incompressible, that share the same characteristic: they considers a mixture that contains both phases. An air-assisted liquid sheet configuration has been built to test different models in order to define a metric of comparison. Two different models using the sharp interface approach (ARCHER and InterFoam ), two models using the diffuse interface approach (CEDRE and ELSA ) and an hybrid model (ICMelsa ) have been considered on this test case. A comparison on two parts, based on statistical quantities, has been proposed. A fist part called "classical study", compare the first order statistics showing that all approaches lead to very similar results, as soon as certain level of mesh resolution is achieved. At the contrary the second order statics present noticeable differences. These results motivate a second part called "phase analysis" to study the link between the small scale representation of the interface and the second-order statics. In particular, the phase marker variance and the associated segregation level are found to be sensible indicators of the interface description. A 1D signal analysis shows that they can be used to detect any departure from the separated phases representation.Then the importance of the phase indicator variance is demonstrated on other second-order statistics: Reynolds stress components and turbulent liquid flux. Thus, second-order statistics are partly described with direct mixed phases representations and require complementary model to be fully recovered. A first attempt, based on a linear approach, is proposed to model the level of segregation of mixed phases representation. It is based on the filtering of a fully segregated signal at a given scale. In a second part of this thesis, an industrial test case (a pressure swirling injector) proposed by SAFRAN Aircraft Engines is studied. Three industrial oriented models, among those studied in the first part, have been applied to simulate this injector flow (InterFoam , ELSA , ICMelsa). Their present numerical approaches are able to work with complex geometries, with a computational effort representative of the industrial current standards. The results of the three models (liquid film thickness, breakup length and Sauter Mean Diameter) have been compared with respect to the available experimental data. Eventually, a proposal to improve the ICMelsa model multi-scale have been successfully tested on the liquid sheet configuration and implemented to further improve the results of the SAFRAN Aircraft Engines industrial case. These results have shown that we are very close to predict the characteristics of a spray produced by a real aeronautical injection system
Le normative più rigorose sulle emissioni dei motori aeronautici poste dall’ICAO, richiedono strumenti di progettazione sempre più predittivi. La distribuzione dei diametri delle droplets prodotte durante il processo di atomizzazione è un parametro chiave per predire l’emissione di inquinanti rilasciati durante il processo di combustione. Lo studio del fenomeno di atomizzazione con la sua natura multi scala diventa una sfida rilevante. Per questo motivo, gli obiettivi di questo lavoro sono: in primo luogo una revisione dei modelli esistenti in letteratura per comprenderne le caratteristiche chiave al fine di definire una classificazione che dia delle linee guida sulle scelte di modellistica; in secondo luogo applicare approcci orientati all’industria ad una configurazione aeronautica, al fine di proporre un miglioramento degli strumenti di progettazione disponibili al giorno d’oggi. Nella prima parte del PhD proponiamo una classificazione sistematica dei modelli rispetto alla scala di lunghezza considerata per rappresentare l’interfaccia. Da questo punto di vista, è possibile distinguere due tipi di approccio: una rappresentazione a fasi separate e una rappresentazione delle fasi miste. Gli approcci a interfaccia diffusa appartengono alla seconda categoria insieme a molti altri approcci (che ricorrano ad un approccio comprimibile o incomprimibile) che condividono la stessa caratteristica: considerano una miscela che contiene entrambe le fasi. È stata realizzata una configurazione air-assisted liquid sheet per testare diversi modelli al fine di definire una metrica di confronto. In questo caso per i test sono stati considerati due diversi modelli che utilizzano l’approccio sharp interface (ARCHER e InterFoam ), due modelli che utilizzano l’approccio a interfaccia diffusa (CEDRE e ELSA ) e un modello ibrido (ICMelsa ). È stato proposto un confronto su due parti, basato su statistiche di diversa natura. Una prima parte chiamata "studio classico", confronta le statistiche del primo ordine, le quali mostrano che tutti gli approcci portano a risultati molto simili non appena viene raggiunto un certo livello di risoluzione della mesh. Al contrario, le statistiche di secondo ordine presentano notevoli differenze. Questi risultati motivano una seconda parte dello studio, chiamata "analisi di fase" sviluppata per studiare il legame tra le piccole scale dell’interfaccia e le statiche di secondo ordine. In particolare, la varianza del marker di fase e il livello di segregazione associato si trovano ad essere indicatori sensibili alla descrizione dell’interfaccia. Un’analisi di un segnale 1D mostra che questi due indicatori possono essere utilizzati per rilevare qualsiasi deviazione dalla rappresentazione a fasi separate. successivamente l’importanza della varianza del marker di fase è dimostrata su altre statistiche di secondo ordine: componenti del tensore degli sforzi di Reynolds e flusso liquido turbolento. Pertanto, le statistiche di secondo ordine che sono descritte con una rappresentazione a fasi miste richiedono un modello complementare per essere completamente recuperate. Un primo tentativo, basato su un approccio lineare, è proposto per modellare il livello di segregazione della rappresentazione a fasi miste. Si basa sul filtraggio di un segnale completamente separato su una data scala. In una seconda parte della tesi viene studiato un caso industriale (un iniettore di tipo swirling) proposto dall’azienda SAFRAN Aircraft Engines . Per simulare il flusso prodotto da questo iniettore sono stati applicati tre modelli (con un approccio industriale) tra quelli studiati nella prima parte (InterFoam , ELSA , ICMelsa )

L’atomisation texturelle désigne le mécanisme d’arrachage de gouttes à l’interface d’un écoulement liquide libre. Cemécanisme est contrôlé par les caractéristiques de l’écoulement au sortir de l’injecteur et se manifeste dans son champproche. Peu étudiée, l’atomisation texturelle est un phénomène rapide, impliquant de très petites structures ligamentaireset produisant un brouillard de très fines gouttes. Le travail de cette thèse est une étude expérimentale d’un processusd’atomisation texturelle observé sur des écoulements produits par des injecteurs cavitants. Trois atomiseurs académiquestransparents sont utilisés et des diagnostics optiques sont mis en oeuvre : la LDV (vélocimétrie doppler laser) et le PDPA(Phase Doppler Particle Analyzer) pour décrire l’écoulement interne et le spray, respectivement, et l’imagerie fixe à forterésolution spatiale ou à haute-cadence pour les écoulements interne et externe. Une première observation montre un lienimportant entre le régime de cavitation et le processus d’atomisation texturelle. Une analyse impliquant la mesure devariabilité de l’écoulement et du processus d’atomisation texturelle quantifie ce lien. Ici, le processus d’atomisation estdécrit par la mesure de sa distribution d’échelle. Associée au concept de système équivalent, cette analyse multi-échellepermet de produire une écriture mathématique du processus étudié. Ce résultat est sans précédent. Complété par unedescription mathématique de la distribution de taille des gouttes produites, il offre un appui nouveau pour construire unmodèle d’atomisation ligamentaires présenté dans ce travail et qui relie taille et forme des ligaments aux populations degouttes formées. Ces analyses fines amènent une meilleure connaissance du mécanisme d’atomisation étudié. Par exemple, on apprend qu’à débit fixé, la hauteur du canal d’alimentation de l’orifice de décharge n’influence pas les processus d’atomisation texturelle. Par ailleurs, un critère est établi pour identifier la plus petite échelle de déformation des ligaments impliquée dans la production des gouttes
Textural atomization designates the mechanism of drop peeling from the interface of a free liquid flow. This mechanismis controlled by the characteristics of the flow issuing from the injector and manifests at its vicinity. Almost uninvestigated,textural atomization is a rapid phenomenon, implies very small ligamentary structures and produces a mist of fine droplets.The work of this thesis is an experimental investigation of a textural atomization process observed on flows issuing fromcavitating injector. Three academic transparent atomizers are used and optical diagnostics are implemented: LDV (LaserDoppler Velocimetry) and PDPA (Phase Doppler Particle Analyzer) to describe the internal flow and the spray, respectively,and still imaging at high spatial resolution or high-speed imaging for the internal and external flows. A first observation revealsa strong link between the cavitation regime and the textural atomization process. An analysis implying the measurementof the variability of the internal flow and of the atomization process quantifies this link. Here, the atomization process isdescribed by the measurement of its scale distribution. Associated with the concept of equivalent system, this multi-scaleanalysis returns a mathematical expression for the investigated atomization process. This result is unprecedented. Completedby a mathematical description of the spray drop-diameter distribution, it offers a new support to build a model of ligamentaryatomization processes presented in this work and that connects ligament size and deformation to the drop populations. Thesefine analyses provide a better knowledge of the investigated atomization process. For instance, we learn that, at fixed flowrate, the height of the inlet pipe feeding the orifice has no influence on the atomization process. Furthermore, a criterion hasbeen established to identify the smallest ligament deformation scale implied in the drop production

Émissions polluantes, les motoristes souhaitent contrôler au mieux l’atomisation du carburant, injecté généralement sous forme de jets ou de nappes liquides. Les essais étant long et coûteux, leur remplacement par un outil numérique capable de simuler le processus d’atomisation permettrait non seulement une réduction des coûts importante mais faciliterait également la phase de conception. Toutefois, en raison du caractère multi-échelle du phénomène, il est difficile de le décrire dans son ensemble avec les approches habituellement utilisées en mécanique des fluides numérique.L’objectif de cette thèse est de concevoir une nouvelle approche qui permettra à terme de simuler l’atomisation pour une configuration industrielle complète. Celle-ci consiste à coupler deux types de modèles. Le premier, dit modèle bifluide, est un modèle à deux fluides compressibles basé sur les équations de Navier-Stokes diphasiques. Celui-ci permet de décrire les grandes échelles du phénomène d’atomisation correspondant à la formation de ligaments et d’amas liquides dans la zone proche de l’injecteur. Le second, dit modèle de spray, est basé sur une équation cinétique. Dans la zone située en aval de l’injecteur, ce dernier permet de décrire de manière statistique l’évolution du brouillard de gouttelettes issues de la fragmentation primaire du jet de carburant. Le point délicat, à la fois sur le plan de la modélisation et sur celui de l’algorithmique, réside dans le couplage des deux modèles. Celui ci a été réalisé grâce à l’introduction de deux modèles auxiliaires permettant de traiter le transfert de liquide entre le modèle bifluide et le modèle de spray par atomisation ou ré-impact.L’approche proposée a été appliquée à la simulation numérique de nappes liquides cisaillées. Les comparaisons entre les résultats numériques et des résultats expérimentaux montrent que le modèle bifluide permet de prévoir l’influence de la géométrie et des conditions d’injection sur l’atomisation primaire de la nappe liquide. Le modèle d’atomisation permet quant à lui, de reproduire le caractère instationnaire des mécanismes de production de gouttes lors du transfert de la phase liquide depuis le modèle bifluide vers celui de spray. Des cas de ré-impact valident également la robustesse et la généralité de la méthodologie de couplage
In order to improve efficiency of aircraft combustion chambers and reduce polluting emissions,engine manufacturers try to achieve a better control on fuel atomization, which is usually injectedas a jet or liquid sheet. As experiments are expensive and time consuming, a numerical tool able to simulate atomization would be a powerful asset in engine conception design. However, simulation ofthe whole atomization process with commonly used approach in computational fluid dynamics is still prohibitive due to the multi-scale nature of the phenomenon.The objective of this thesis is to develop a new approach allowing the simulation of the spray formation for a industrial configuration in the near future. This involves coupling of two types of models.The first one, called two-fluid model, is based on the Navier-Stokes equations for two immiscible compressible fluids. This one is used to describe the large scales of the atomization mechanism corresponding to the formation of ligaments and liquids blobs in the near-injector area. The second one,called spray model, is based on a kinetic equation. Further downstream from the injector, this model describes statistically the evolution of the droplet cloud produced by the primary fragmentation of liquid jet. The main difficulty, in terms of both modeling and algorithmic, is the coupling of these twomodels.This has been achieved by introducing an atomization and an impact models which ensure liquid transfer between the two-fluid model and the spray model.This new approach was applied to the numerical simulation of sheared liquid sheets. Comparisons between numerical and experimental results show how the two-fluid model predicts the influence of injector geometry and injection conditions on the primary atomization of the liquid sheet. Concerning droplets production, the atomization model is able to reproduce the unsteady nature of this mechanism when transferring liquid phase from the two-fluid model to the spray model. Test cases for the impact model also validate the robustness and generality of the coupling approach

The thesis deals with experimental research of mechanism of liquid breakup at twin-fluid atomizers. Four different atomizers were examined at the beginning of the research. Two of them were of standard design (Y-jet and effervescent nozzles), and the rest two atomizers were developed as a part of the thesis (so called CFT and inversed effervescent atomizers). Results show that only the inversed effervescent atomizer was capable of generating stable spray under examined conditions due to the specific breakup mechanism. This mechanism is similar to what was observed in effervescent atomizers. However, the mixing process inside the inversed effervescent atomizer is very different. The specific breakup mechanism was then defined as the main scope of the thesis. It was investigated by the high-speed imaging. The images were then processed by proper orthogonal decomposition and by fast Fourier transformation. Spray spatial development was examined using phase Doppler anemometer. The data was analyzed to describe the dynamics of the spray. A detailed description of the breakup mechanism is made from the combination of the experimental and post-processing techniques. The thesis brings new insight into the understanding of the liquid breakup mechanism and shows a potential for a further development of the inversed effervescent atomizer.

Cette thèse présente une extension de l'approche stochastique de l'atomisation primaire de type air assisté près d'un injecteur. Cette approche avait déjà été introduite dans les publications de Gorokhovski et al. Dans le cadre de la simulation des grandes échelles, la zone d'atomisation primaire est simulée comme un corps immergé avec une structure stochastique. Ce dernier est défini par la simulation stochastique de la position et de la courbure de l'interface entre le liquide et le gaz. La simulation de la position de l'interface est basée sur l'hypothèse de symétrie d'échelle pour la fragmentation. La normale extérieure à l'interface est modélisée en supposant une relaxation statistique vers l'isotropie. Les statistiques de la force du corps immergé servent de conditions aux limites pour le champ de vitesse issu de la LES ainsi que pour la production des gouttes de l'atomisation primaire. Celles-ci sont ensuite transportées par une approche lagrangienne. Les collisions entre les gouttes dans la zone d'atomisation primaire sont prises en compte par analogie avec l'approche standard de la théorie cinétique des gaz. Une fermeture est proposée pour la température statistique des gouttelettes. Cette approche est validée par des comparaisons avec les mesures expérimentales de la thèse de Hong. Les résultats numériques pour la vitesse et de la taille des gouttes dans le spray à différentes distances du centre du jet et de l'orifice de la buse sont relativement proches des résultats expérimentaux. Différentes conditions d'entrée pour la vitesse sont testées et comparées aux résultats expérimentaux. Par ailleurs, le rôle spécifique de la zone de recirculation devant le dard liquide est soulignée par le battement du dard liquide et la production de gouttelettes
This thesis introduced an extension to stochastic approach for simulation of air-blast atomization closely to injector. This approach was previously proposed in publications of Gorokhovski with his PHD students. Our extension of this approach is as follows. In the framework of LES approach, the contribution of primary atomization zone is simulated as an immersed solid body with stochastic structure. The last one is defined by stochastic simulation of position-and-curvature of interface between the liquid and the gas. As it was done previously in this approach, the simulation of the interface position was based on statistical universalities of fragmentation under scaling symmetry. Additionally to this, we simulate the outwards normal to the interface, assuming its stochastic relaxation to isotropy along with propagation of spray in the down-stream direction. In this approach, the statistics of immersed body force plays role of boundary condition for LES velocity field, as well as for production of primary blobs, which are then tracked in the Lagrangian way. In this thesis, the inter-particle collisions in the primary atomisation zone are accounted also by analogy with standard kinetic approach for the ideal gas. The closure is proposed for the statistical temperature of droplets. The approach was assessed by comparison with measurements of Hong in his PHD. The results of computation showed that predicted statistics of the velocity and of the size in the spray at different distances from the center plane, at different distances from the nozzle orifice, at different inlet conditions (different gas velocity at constant gas-to-liquid momentum ratio, different gas-to-liquid momentum ratio) are relatively close to measurements. Besides, the specific role of recirculation zone in front of the liquid core was emphasized in the flapping of the liquid core and in the droplets production

L’application de cette thèse est le contrôle actif de la combustion dans les brûleurs industriels à combustible liquide. Il s’agit d’explorer les possibilités de contrôle d’un spray par des jets gazeux auxiliaires. Deux familles d’actionneurs utilisant ce procédé ont été testées sur un atomiseur coaxial assisté par air. Le premier dispositif est appelé (Dev). Composé d’un unique jet actionneur, il vise à dévier le spray. La seconde configuration, appelée (Sw), est équipée de 4 jets auxiliaires tangents au spray afin de lui conférer un effet de swirl et d’en augmenter le taux d’expansion. Les mesures de granulométrie par PDA et les visualisations du spray par strioscopie démontrent un effet important du contrôle sur l’atomisation et la forme du spray. On observe en outre une déviation pouvant atteindre 30°avec l’actionneur (Dev) et une augmentation du taux d’expansion de 80% dans le cas (Sw). Des simulations du banc expérimental ont de plus été menées avec le code AVBP. L’écoulement de gaz est calculé par simulation aux grandes échelles (SGE ou LES en Anglais). L’approche lagrangienne est utilisée pour simuler la phase dispersée. Une attention particulière a été portée aux conditions d’injection du gaz et des gouttes dans le calcul. Ceci a abouti au développement d’une nouvelle condition limite caractéristique non réfléchissante (VFCBC) destinée à l’injection d’écoulements turbulents en LES compressible. Les résultats de LES présentent un bon accord avec les mesures expérimentales. Les effets du contrôle sur la dynamique des gouttes et sur la topologie du spray (forme, déviation, expansion) sont correctement décrits
The present work focuses on active control of two-phase combustion in industrial burners. The generic method explored in this thesis consists in controlling the injected fuel spray with transverse air jets. Two families of these jet actuators are tested on a coaxial airblast atomizer. The first system (Dev) is used to modify the trajectory of the spray, while the second one (Sw) introduces swirl into the spray to modify its spreading rate and mixing with the surrounding air. Experimental characterisations of the controlled flow with Schlieren visualisations and Phase Doppler Anemometry (PDA) show that actuators induce important effect on the spray. The deviation angle reaches 30° for the actuator (Dev) and the expansion rate increases of 80 % in the swirl case (Sw). Simulations of the experiment are then performed with the CFD code AVBP. The gas flow is computed with Large Eddy Simulation (LES). A Lagrangian formulation is used to simulate droplets trajectories. A particular attention is given to the injection of the gas flow and the droplets in the calculations. Therefore, a new non-reflecting characteristic boundary condition (VFCBC) has been derived to inject turbulent flows in compressible LES. A good agreement is observed between simulation and experiment. Control effects on the spray topology ( features, deviation, spread rate) and on the droplets velocities and diameters are correctly described by the Lagrangian LES

Afin de se rapprocher des conditions du mélange homogène du moteur essence, plusieurs fluides sont injectés dans l'atmosphère à une pression amont et une température variées. Cinq prototypes d'injecteurs réels trois-trous de Continental ont été utilisés. En augmentant la pression d'injection, l'écoulement passe par quatre régimes où le niveau de développement de cavitation varie. Le coefficient de décharge Cd dépend essentiellement du nombre de cavitation. Au point critique de cavitation, deux corrélations ont été obtenues reliant respectivement Cd et le nombre de cavitation critique au nombre de Reynolds correspondant. Le jet en champ proche est gouverné par trois nombres sans dimensions : celui de Weber, de Reynolds et de cavitation. L'effet de chacun d'eux sur l'angle du jet à la sortie a été obtenu. La comparaison des résultats entre deux injecteurs a montré que le rapport entre la longueur et le diamètre de l'orifice est d'une influence d'ordre 1 sur l'angle du jet
In order to get closer to the homogeneous mixture conditions of a gasoline engine, different fluids are injected into the atmosphere at varying upstream pressure and temperature. Five three-hole real injector prototypes from Continental were used. When injection pressure is increased, the internal flow goes through four regimes where the cavitation development level varies from one to another. The discharge coefficient Cd was found mainly dependent on the cavitation number. At the cavitation critical point, two correlations between Cd and the critical cavitation number on one side respectively, and the correspondent Reynolds number on the other side were found. The near field jet is ruled by three dimensionless numbers : Weber, Reynolds and cavitation. The effect of each one of them on the jet angle at the orifice outlet was obtained. By comparing the results of two injectors, it was found that the length over diameter ratio has a first order influence on the jet angle