UPCommons. Portal del coneixement obert de la UPC
Publicación
Contribución de congreso (ConferenceOutput)
Numerical insights into pattern formation in binary convection in cylinders
Binary mixture convection exhibits a wide range of dynamical behaviours driven by the interplay of thermal, solutal, and shear gradients. These depend on both the mixture properties and the geometric configuration and boundary conditions. The onset of convection is strongly influenced by the Soret coefficient. Mixtures with a negative Soret coefficient often display oscillatory instabilities and time-dependent patterns, including dispersive chaos near criticality. Primary and secondary bifurcations can be subcritical, leading to temporal complexity. In contrast, mixtures with a positive Soret coefficient typically develop stationary patterns near onset, as observed in early experiments in cylindrical cells (Le Gal et al., 1985; Moses et al. 1986). However, numerical studies suggest that oscillatory instabilities can also arise under specific conditions (Weggler et al., 2010; Hu et al., 2021). We use in-house numerical tools to solve the Navier-Stokes equations for binary convection in cylindrical geometries. Most results come from a three-dimensional time-stepping spectral code (Mercadet et al., 2010) that captures stable patterns. Additionally, a Newton-Krylov solver enables to compute stable and unstable steady states, while continuation techniques and Arnoldi iteration allow tracking solution branches and analyzing their stability. Although we have extensively applied the Newton-Krylov solver to two-dimensional cases, solving the corresponding three-dimensional systems remained computationally prohibitive until now., This research is supported by the Ministerio de Ciencia, Innovación y Universidades (Agencia Estatal de Investigación, projects ns. PID2020-114043GB-I00 (MCIN/AEI/10.13039/501100011033) and PID2023-150029NB-I00 (MCIN/AEI/10.13039/501100011033/FEDER,UE)., Peer Reviewed