• Citer

Chemical Control of Hydrodynamic Interfacial Instabilities

par Oyero, Johnson Olubori
Président du jury Rongy, Laurence
Promoteur De Wit, Anne
Publication Non publié, 2025-12-11

Thèse de doctorat

• Accès en ligne
• Détails
• Contenu
• Statistiques

Résumé :

Mixing fluids involves bringing fluids into close contact until the composition is uniform in space and time. In labs and industry, stirring, blending, and shearing from mechanical agitation are standard ways to accelerate the mixing of fluids. When fluids mix without external forcing, for instance in groundwater, the ocean, and Earth’s mantle, gradients in density, viscosity, or surface tension typically trigger interfacial instabilities that drive self-organized convective fluid mixing. If the contacting layers contain reactive or non-reactive chemical species, chemistry can then be leveraged to modulate the ensuing mixing. In that regard, this thesis investigates how composition and reactions control hydrodynamic interfacial instabilities and expresses the controls as practical scaling laws. We first consider buoyancy-driven mixing in binary miscible layers of gaseous-phase bulk fluids. Unstable density stratifications trigger Rayleigh–Taylor (RT) convection, while nominally stable ones can destabilize via double diffusion (DD) or diffusive-layer convection (DLC) when solute diffusivities differ. Under suitable conditions, mixed-modes (RT–DD, RT–DLC) fluid mixing also occur. Using Direct Numerical Simulations of the Navier–Stokes equation with species transport, we quantify mixing by mixing-layer growth rate and the instability onset time. We found that differential diffusion creates a dynamic density contrast that makes convective mixing start earlier and intensify relative to single-solute cases. The central result is that onset is governed by the density contrast generated before convection, not the initial jump, with large dynamic density difference generated by mixed-mode instabilities found to speed up growth and advances onset of convective mixing. We then increased the Schmidt number to represent liquid-liquid mixing in stratified fluids. For liquid-liquid mixing, sharper concentration fronts persist to small dissipative scales, increasing intermittency and slowing complete homogenization. In stable stratifications, onset becomes more sensitive to dynamic density contrast while in RT-dominated cases, background forcing largely sets the timing . Dynamic density contrast generated by mixed-modes facilitates mixing the most. Next, we couple simple bimolecular reactions with buoyancy-driven flow to obtain chemo-hydrodynamic patterns. We showed that reactions without differential diffusion can create or erase in-situ density extrema, influencing onset times and front motion as well as the product yields. Finally, we used the area under the evolving density profile to predict the onset of instabilities. Across reactive and non-reactive cases, the onset time of instabilities follows power-law function of the identified area under the density profile at the onset of instabilities. This approach enables prediction from measurable profiles without a full stability analysis. The resulting design rules emphasize tuning the relevant density contrast at the interface of stratified fluids, accounting for the Schmidt number, and using reaction selectively to trigger, suppress, or steer convection and product distribution. These insights pave way for adaptable control for mixing fluids in stratified bulk fluid layers.