Four-way coupled Eulerian–Lagrangian Direct Numerical Simulations in a vertical laminar channel flow

Abstract

International audienceDirect Numerical Simulations of a laminar two-phase flow into a vertical channel are investigated. An Eulerian–Lagrangian approach allows tracking each bubble position with a four-way coupling strategy, i.e. taking into account bubble-fluid and bubble-bubble interactions. The flow configuration has been chosen to highlight the buoyancy effects due to significant values of void fraction (high numbers of bubbles); hence the bubbles collisions and wall effects are the critical parameters to ensure the dispersion of the bubble plume. The DNS approach is self-consistent and does not rely on closure relations or empirical correlations for describing the collective bubble dynamics. It is found that the DNS predicts well the behavior of the bubble plume and its back effect on the liquid phase when compared with a mixture model and experimental data. The elastic nature of collisions, the sensitivity of the mean and RMS values of velocities and void fraction to the mesh quality are explored

Water electrolysis: from textbook knowledge to the latest scientific strategies and industrial developments

“…Secondly, because the separator is porous, intermixing between H 2 and O 2 bubbles is possible if the mass-transport is not wellbalanced, which has adverse consequences in terms of safety of operation, but also of gas purity. 237,238 In practice, AWE cells need several hours to reach their steady-state, in terms of electrolyte flow, temperature and current density (hence of bubbles generated), 57,239 which means that AWE can usually not be operated in transient regime, making their coupling to renewable sources of solar/wind electricity awkward (although this coupling is being studied). 240 For the same reasons, operation under pressure is awkward.…”

“…55 The reactions are not severely mass-transport limited in a well-designed cell, except if bubbles are poorly managed, in particular in AWE. 56,57 Besides compensation of activation barriers at the anode and cathode side caused by charge-transfer limitations, the overall overpotential results from solution and contact resistances. Activation barriers can be reduced by exploiting improved electrocatalysts suitable for OER and HER, whereas a clever cell design can substantially reduce Ohmic and mass-transport resistances.…”

Oil/water flow in a horizontal pipe—dispersed flow regime

“…Numerical simulations have been performed using methods based on the solution of the transport equations for the fluids through the direct simulation of local and instantaneous equations, such as level-set method [16][17][18], phase-field method [19,20] or volume of fluids method [20,21], or through the solution of the average equations. In the latter case, the systems can be modelled using a homogeneous model, where the phases are treated as pseudofluids with average properties, and in this approach, the flow pattern is treated in a less detailed way [22], through a Lagrangian-Eulerian approach, where one of the phases is treated from an Eulerian perspective (as in the single phase flow) and the other phase receives a Lagrangian treatment [23,24], or using an Eulerian-Eulerian approach, where each phase is treated continuously and the coupling between the phases occurs through interfacial terms [25][26][27].…”

Numerical investigation of nanofluid particle migration and convective heat transfer in microchannels using an Eulerian–Lagrangian approach

“…2014), direct numerical simulation (Schillings et al. 2017), large eddy simulation (de Souza, de Vasconcelos Salvo & de Moro Martins 2012), and Reynolds-averaged Navier–Stokes (El-Askary et al. 2016).…”

“…In Eulerian-Lagrangian modelling, the carrier phase is treated in an Eulerian reference frame based on a variety of formulations, which largely depends on whether the flow is laminar or turbulent and whether the fluid is gaseous or liquid. This includes finite difference (Mirzaei, Saffar-Avval & Naderan 2014), finite volume (Deepak Selvakumar & Dhinakaran 2017), finite element (Dong et al 2014), direct numerical simulation (Schillings et al 2017), large eddy simulation (de Souza, de Vasconcelos Salvo & de Moro Martins 2012), and Reynolds-averaged Navier-Stokes (El-Askary et al 2016). On the other hand, large numbers of particles are individually tracked as point masses (i.e.…”