HAL (Le Centre pour la Communication Scientifique Directe), Sep 1, 2003
L'hypothèse quasi-stationnaire appliquée au processus d'évaporation d'une goutte de liquide pur c... more L'hypothèse quasi-stationnaire appliquée au processus d'évaporation d'une goutte de liquide pur critique dans sa vapeurà la pression critique conduità des résultats qui diffèrent assez significativement des résultats numériques. Deux sources d'instationnarités peuventêtre invoquées pour expliquer cesécarts : la présence d'une zone instationnaire loin de la goutte et la relaxation des conditions initiales. Des corrections aux résultats quasi-stationnaires prenant en compte ces deux effets sont déterminées. La comparaison avec les résultats numériques s'avère satisfaisante.
Il a ete recemment montre durant les experiences spatiales et par des considerations theoriques q... more Il a ete recemment montre durant les experiences spatiales et par des considerations theoriques que le transfert de chaleur a proximite du point critique liquide vapeur se faisait par le mecanisme de l'effet piston. En effet, la tres forte expansion de la fine couche limite initiale entraine, par un processus adiabatique, une augmentation rapide de la temperature dans le reste du fluide par le biais du va-et-vient des ondes acoustiques. Dans ce present travail sont explores les mecanismes de transport de masse et de chaleur dans une cavite rectangulaire chauffee par le cote, remplie d'un fluide supercritique, et la competition entre la convection naturelle et l'effet piston (ou convection theroacoustique). Les equations de navier-stokes pour un fluide proche de son point critique, suivant une loi d'etat de type van der waals, sont resolues numeriquement par la methode des volumes finis avec la procedure de filtrage acoustique. Les resultats ont ete surprenants compares a ceux d'un fluide parfait: i) l'equilibration en temperature est toujours aussi rapide que dans le cas de conditions de gravite nulle (simulations 1d) par effet piston (avant que la convection ne demarre): ii) l'equilibration de la masse se fait sur une echelle de temps beaucoup plus longue par la convection naturelle quasi-isotherme iii) a cause de la tres forte compressibilite, un effet de point d'arret similaire a celui que l'on rencontre dans les ecoulements a forte vitesse, provoque une surchauffe de mur superieur
Using iodine as transport species and octofluorocyclobutane (C4F8) as inert background gas, physi... more Using iodine as transport species and octofluorocyclobutane (C4F8) as inert background gas, physical vapor transport rates in horizontal cylindrical ampoules were accurately measured over a wide range of well-defined ampoule filling and thermal boundary conditions. Nonlinear axial temperature distributions were applied that minimized multiple crystal nucleation. Ampoule filling conditions were chosen that resulted in the dominance of either compositionally or thermally driven convection. Three-dimensional numerical modeling of the transport process yielded quantitative agreement with measured transport rates. Analysis of the calculated flow and concentration fields reveals the physical reasons for the failure of the various two-dimensional models to reproduce the experimental results.
The influence of latent heat and natural convection in the melt and the shape of the melt-crystal... more The influence of latent heat and natural convection in the melt and the shape of the melt-crystal interface are analyzed for a vertical Bridgman crystal growth system by direct numerical simulation. The temperature and the velocity field in the melt and in the crystal are computed using an homogenization technique. Initially the ampoule contains poly-crystal material, these container is translated inside a furnace which includes two heating zones characterized by two temperatures which are respectively greater and lower than the melted temperature of the material. The two heating zones in the furnace are separated by an adiabatic zone when the solid-liquid phase change occur. The control of the solidification conditions (phase change front velocity, gravity ...) permits to produce with these technique very high quality single crystals. The results presented in this paper show the effects of various gravitational conditions upon the flow in the melted part of the material inside the ampoule. We have also analyzed the effect of the latent heat upon the shape of the melt-crystal interface and the convective cells in the melted pool.
The aim of the numerical study described in this paper is to understand how the unsteadiness of t... more The aim of the numerical study described in this paper is to understand how the unsteadiness of the wake of an outer layer device in laminar boundary layer could reduce the skin friction. In order to do this the authors have solved the full Navier-Stokes equations for a laminar transient flow using a finite volume technique. The "device" chosen is of rectangular section and has been studied numerically for a symmetrical configuration in the absence of a wall.
A numerical study was performed in order to understand the mechanisms leading to skin friction re... more A numerical study was performed in order to understand the mechanisms leading to skin friction reduction appearance due to large eddy break-up (LEBU). Specifically, the skin friction reduction device constituted of either one or two thick rectangular plates. The chord-to-thickness of a plate is 3, and the value of the Reynolds number based on the plate's chord is constant and equal to 1C)4. The problem is numerically treated by the resolution of Navier-Stokes equations for a laminar unsteady flow using a finite volume technique [1].
The problem of determining the sensitivity of Bridgman-Stockbarger directional solidification exp... more The problem of determining the sensitivity of Bridgman-Stockbarger directional solidification experiments to residual accelerations of the type associated with spacecraft in low earth orbit is analyzed numerically using a pseudo-spectral collocation method. The approach employs a novel iterative scheme combining the method of artificial compressibility and a generalized ADI method. The results emphasize the importance of the consideration of residual accelerations and careful selection of the operating conditions in order to take full advantages of the low gravity conditions.
HAL (Le Centre pour la Communication Scientifique Directe), Aug 31, 2018
Two-dimensional and three-dimensional numerical simulations were carried out to simulate the natu... more Two-dimensional and three-dimensional numerical simulations were carried out to simulate the natural convection in a cylindrical envelope with an internal concentric cylinder with slots. For the case of steady flow, the numerical solutions in 2D and 3D simulations are in good agreement with the experimental results. When the convection in experiment becomes unsteady convection at the larger slot degree, the oscillated solutions in 3D simulation differ essentially from steady solutions in 2D simulation. The critical Rayleigh numbers from steady to unsteady flow in the 3D simulations are lower than those in the 2D simulations.
HAL (Le Centre pour la Communication Scientifique Directe), Jun 20, 2019
Two-dimensional and three-dimensional numerical simulations were carried out to simulate the natu... more Two-dimensional and three-dimensional numerical simulations were carried out to simulate the natural convection in a cylindrical envelope with an internal concentric cylinder with slots. For the case of steady flow, the numerical solutions in 2D and 3D simulations are in good agreement with the experimental results. When the convection in experiment becomes unsteady convection at the larger slot degree, the oscillated solutions in 3D simulation differ essentially from steady solutions in 2D simulation. The critical Rayleigh numbers from steady to unsteady flow in the 3D simulations are lower than those in the 2D simulations.
Abstract A Fourier-Chebyshev pseudospectral method for solving the steady 3-D Navier-Stokes and e... more Abstract A Fourier-Chebyshev pseudospectral method for solving the steady 3-D Navier-Stokes and energy equations in cylindrical cavities is presented and discussed. The general method is pseudo-unsteady and uses a semi-implicit finite-difference scheme for the time integration. The generalized ADI procedure is then applied to reduce the problem to successive solutions of 1-D problems. The spatial discretization uses a Fourier-Galerkin approximation in the periodic direction and a Chebyshev-collocation approximation in the other directions. Difficulties related to the pressure are surmounted by using the artificial compressibility method. A suitable variable change has been chosen to avoid the problem of singularity at the axis generated by cylindrical coordinates. The method is first tested on an advection-diffusion equation and then on the Navier-Stokes and heat equations. Finally, the method is illustrated by the problem of convection in a differentially heated fluid.
The paper deals with the continuation of the development of high performance computer codes and t... more The paper deals with the continuation of the development of high performance computer codes and their application to modelling of crystal growth processes started by the authors’ research groups and reported in [1]. The mathematical model is based on the continuity equation and the conservation equations for momentum and heat transfer combined with mass transfer including chemical reactions. The thermal radiation analysis assumes a non-participating medium and semi-transparent walls. The radiation heat transfer is coupled with convection and conduction. The heat conduction includes thermal solid/fluid interactions between the gas and solid parts of the computational domain. The results of thermal calculations are used for the analysis of thermal stresses. The models are implemented in finite volume (both, block-structured and unstructured on non-orthogonal grids), and spectral and coupled finite volume/spectral numerical solution procedures.
Dew water is water vapour that passively condenses from air. Once properly collected, it can prov... more Dew water is water vapour that passively condenses from air. Once properly collected, it can provide a useful supplementary water resource for plants and humans. I ts production can be significantly improved by using specific materials and particular geometry. In this context, new shapes for dew collectors are presented and their water yields are compared with those of a 1 m 2 , 30° , inclined planar condenser used as a standard. The experiments were carried out in Pess ac (SW France), situated about 45 km from the Atlantic Ocean, during summer and fall 2009. In addition to conical shapes, which have 30 % larger yields than the planar reference condenser and whose functioning was simulated numerically, two new families of forms are considered: egg - box and origami types. The egg - box shape yields 9 % more water as compared to the reference planar condenser, a result nearly independent of the dew yield. In contrast, the origami shape gives yields 150 % larger than the referen ce pl...
The European Physical Journal Special Topics, 2017
Motivated by the space project EFILE, a 2D axisymmetric numerical model in the framework of ALE m... more Motivated by the space project EFILE, a 2D axisymmetric numerical model in the framework of ALE method is developed to investigate the coupled physical mechanism during the evaporation of a pinned drop that partially wets on a heated substrate. The model accounts for mass transport in surrounding air, Marangoni convection inside the drop and heat conduction in the substrate as well as moving interface. Numerical results predict simple scaling laws for the evaporation rate which scales linearly with drop radius but follows a power-law with substrate temperature. It is highlighted that thermal effect of the substrate has a great impact on the temperature profile at the drop surface, which leads to a multicellular thermocapillary flow pattern. In particular, the structure of the multicellular flow behavior induced within a heated drop is mainly controlled by a geometric parameter (aspect ratio). A relationship between the number of thermal cells and the aspect ratio is proposed.
HAL (Le Centre pour la Communication Scientifique Directe), Sep 1, 2003
L'hypothèse quasi-stationnaire appliquée au processus d'évaporation d'une goutte de liquide pur c... more L'hypothèse quasi-stationnaire appliquée au processus d'évaporation d'une goutte de liquide pur critique dans sa vapeurà la pression critique conduità des résultats qui diffèrent assez significativement des résultats numériques. Deux sources d'instationnarités peuventêtre invoquées pour expliquer cesécarts : la présence d'une zone instationnaire loin de la goutte et la relaxation des conditions initiales. Des corrections aux résultats quasi-stationnaires prenant en compte ces deux effets sont déterminées. La comparaison avec les résultats numériques s'avère satisfaisante.
Il a ete recemment montre durant les experiences spatiales et par des considerations theoriques q... more Il a ete recemment montre durant les experiences spatiales et par des considerations theoriques que le transfert de chaleur a proximite du point critique liquide vapeur se faisait par le mecanisme de l'effet piston. En effet, la tres forte expansion de la fine couche limite initiale entraine, par un processus adiabatique, une augmentation rapide de la temperature dans le reste du fluide par le biais du va-et-vient des ondes acoustiques. Dans ce present travail sont explores les mecanismes de transport de masse et de chaleur dans une cavite rectangulaire chauffee par le cote, remplie d'un fluide supercritique, et la competition entre la convection naturelle et l'effet piston (ou convection theroacoustique). Les equations de navier-stokes pour un fluide proche de son point critique, suivant une loi d'etat de type van der waals, sont resolues numeriquement par la methode des volumes finis avec la procedure de filtrage acoustique. Les resultats ont ete surprenants compares a ceux d'un fluide parfait: i) l'equilibration en temperature est toujours aussi rapide que dans le cas de conditions de gravite nulle (simulations 1d) par effet piston (avant que la convection ne demarre): ii) l'equilibration de la masse se fait sur une echelle de temps beaucoup plus longue par la convection naturelle quasi-isotherme iii) a cause de la tres forte compressibilite, un effet de point d'arret similaire a celui que l'on rencontre dans les ecoulements a forte vitesse, provoque une surchauffe de mur superieur
Using iodine as transport species and octofluorocyclobutane (C4F8) as inert background gas, physi... more Using iodine as transport species and octofluorocyclobutane (C4F8) as inert background gas, physical vapor transport rates in horizontal cylindrical ampoules were accurately measured over a wide range of well-defined ampoule filling and thermal boundary conditions. Nonlinear axial temperature distributions were applied that minimized multiple crystal nucleation. Ampoule filling conditions were chosen that resulted in the dominance of either compositionally or thermally driven convection. Three-dimensional numerical modeling of the transport process yielded quantitative agreement with measured transport rates. Analysis of the calculated flow and concentration fields reveals the physical reasons for the failure of the various two-dimensional models to reproduce the experimental results.
The influence of latent heat and natural convection in the melt and the shape of the melt-crystal... more The influence of latent heat and natural convection in the melt and the shape of the melt-crystal interface are analyzed for a vertical Bridgman crystal growth system by direct numerical simulation. The temperature and the velocity field in the melt and in the crystal are computed using an homogenization technique. Initially the ampoule contains poly-crystal material, these container is translated inside a furnace which includes two heating zones characterized by two temperatures which are respectively greater and lower than the melted temperature of the material. The two heating zones in the furnace are separated by an adiabatic zone when the solid-liquid phase change occur. The control of the solidification conditions (phase change front velocity, gravity ...) permits to produce with these technique very high quality single crystals. The results presented in this paper show the effects of various gravitational conditions upon the flow in the melted part of the material inside the ampoule. We have also analyzed the effect of the latent heat upon the shape of the melt-crystal interface and the convective cells in the melted pool.
The aim of the numerical study described in this paper is to understand how the unsteadiness of t... more The aim of the numerical study described in this paper is to understand how the unsteadiness of the wake of an outer layer device in laminar boundary layer could reduce the skin friction. In order to do this the authors have solved the full Navier-Stokes equations for a laminar transient flow using a finite volume technique. The "device" chosen is of rectangular section and has been studied numerically for a symmetrical configuration in the absence of a wall.
A numerical study was performed in order to understand the mechanisms leading to skin friction re... more A numerical study was performed in order to understand the mechanisms leading to skin friction reduction appearance due to large eddy break-up (LEBU). Specifically, the skin friction reduction device constituted of either one or two thick rectangular plates. The chord-to-thickness of a plate is 3, and the value of the Reynolds number based on the plate's chord is constant and equal to 1C)4. The problem is numerically treated by the resolution of Navier-Stokes equations for a laminar unsteady flow using a finite volume technique [1].
The problem of determining the sensitivity of Bridgman-Stockbarger directional solidification exp... more The problem of determining the sensitivity of Bridgman-Stockbarger directional solidification experiments to residual accelerations of the type associated with spacecraft in low earth orbit is analyzed numerically using a pseudo-spectral collocation method. The approach employs a novel iterative scheme combining the method of artificial compressibility and a generalized ADI method. The results emphasize the importance of the consideration of residual accelerations and careful selection of the operating conditions in order to take full advantages of the low gravity conditions.
HAL (Le Centre pour la Communication Scientifique Directe), Aug 31, 2018
Two-dimensional and three-dimensional numerical simulations were carried out to simulate the natu... more Two-dimensional and three-dimensional numerical simulations were carried out to simulate the natural convection in a cylindrical envelope with an internal concentric cylinder with slots. For the case of steady flow, the numerical solutions in 2D and 3D simulations are in good agreement with the experimental results. When the convection in experiment becomes unsteady convection at the larger slot degree, the oscillated solutions in 3D simulation differ essentially from steady solutions in 2D simulation. The critical Rayleigh numbers from steady to unsteady flow in the 3D simulations are lower than those in the 2D simulations.
HAL (Le Centre pour la Communication Scientifique Directe), Jun 20, 2019
Two-dimensional and three-dimensional numerical simulations were carried out to simulate the natu... more Two-dimensional and three-dimensional numerical simulations were carried out to simulate the natural convection in a cylindrical envelope with an internal concentric cylinder with slots. For the case of steady flow, the numerical solutions in 2D and 3D simulations are in good agreement with the experimental results. When the convection in experiment becomes unsteady convection at the larger slot degree, the oscillated solutions in 3D simulation differ essentially from steady solutions in 2D simulation. The critical Rayleigh numbers from steady to unsteady flow in the 3D simulations are lower than those in the 2D simulations.
Abstract A Fourier-Chebyshev pseudospectral method for solving the steady 3-D Navier-Stokes and e... more Abstract A Fourier-Chebyshev pseudospectral method for solving the steady 3-D Navier-Stokes and energy equations in cylindrical cavities is presented and discussed. The general method is pseudo-unsteady and uses a semi-implicit finite-difference scheme for the time integration. The generalized ADI procedure is then applied to reduce the problem to successive solutions of 1-D problems. The spatial discretization uses a Fourier-Galerkin approximation in the periodic direction and a Chebyshev-collocation approximation in the other directions. Difficulties related to the pressure are surmounted by using the artificial compressibility method. A suitable variable change has been chosen to avoid the problem of singularity at the axis generated by cylindrical coordinates. The method is first tested on an advection-diffusion equation and then on the Navier-Stokes and heat equations. Finally, the method is illustrated by the problem of convection in a differentially heated fluid.
The paper deals with the continuation of the development of high performance computer codes and t... more The paper deals with the continuation of the development of high performance computer codes and their application to modelling of crystal growth processes started by the authors’ research groups and reported in [1]. The mathematical model is based on the continuity equation and the conservation equations for momentum and heat transfer combined with mass transfer including chemical reactions. The thermal radiation analysis assumes a non-participating medium and semi-transparent walls. The radiation heat transfer is coupled with convection and conduction. The heat conduction includes thermal solid/fluid interactions between the gas and solid parts of the computational domain. The results of thermal calculations are used for the analysis of thermal stresses. The models are implemented in finite volume (both, block-structured and unstructured on non-orthogonal grids), and spectral and coupled finite volume/spectral numerical solution procedures.
Dew water is water vapour that passively condenses from air. Once properly collected, it can prov... more Dew water is water vapour that passively condenses from air. Once properly collected, it can provide a useful supplementary water resource for plants and humans. I ts production can be significantly improved by using specific materials and particular geometry. In this context, new shapes for dew collectors are presented and their water yields are compared with those of a 1 m 2 , 30° , inclined planar condenser used as a standard. The experiments were carried out in Pess ac (SW France), situated about 45 km from the Atlantic Ocean, during summer and fall 2009. In addition to conical shapes, which have 30 % larger yields than the planar reference condenser and whose functioning was simulated numerically, two new families of forms are considered: egg - box and origami types. The egg - box shape yields 9 % more water as compared to the reference planar condenser, a result nearly independent of the dew yield. In contrast, the origami shape gives yields 150 % larger than the referen ce pl...
The European Physical Journal Special Topics, 2017
Motivated by the space project EFILE, a 2D axisymmetric numerical model in the framework of ALE m... more Motivated by the space project EFILE, a 2D axisymmetric numerical model in the framework of ALE method is developed to investigate the coupled physical mechanism during the evaporation of a pinned drop that partially wets on a heated substrate. The model accounts for mass transport in surrounding air, Marangoni convection inside the drop and heat conduction in the substrate as well as moving interface. Numerical results predict simple scaling laws for the evaporation rate which scales linearly with drop radius but follows a power-law with substrate temperature. It is highlighted that thermal effect of the substrate has a great impact on the temperature profile at the drop surface, which leads to a multicellular thermocapillary flow pattern. In particular, the structure of the multicellular flow behavior induced within a heated drop is mainly controlled by a geometric parameter (aspect ratio). A relationship between the number of thermal cells and the aspect ratio is proposed.
Uploads
Papers by Jalil Ouazzani