A quantitative and generalized assessment of bubble-induced turbulence models for gas-liquid systems. (May 2019)

• Record Type: Journal Article
• Authors: Magolan, Ben; Lubchenko, Nazar; Baglietto, Emilio
• Is Part Of: Chemical engineering science. Volume 2(2019)
• Publication Date: 2019-05
• Subjects: Bubble-induced turbulence -- k-ε models -- Bubbly flows -- Multiphase CFD; Chemical engineering; Periodicals; 660.05
• DOI: 10.1016/j.cesx.2019.100009
• Languages: English
• ISSNs: 2590-1400

Abstract

Highlights: Bubble-Induced Turbulence (BIT) models poorly predict liquid turbulence. Existing BIT models often worsen liquid velocity predictions. Models that best predict turbulence yield worst velocity predictions. BIT turbulent viscosity, time-scale, and source term formulations are incomplete. BIT models improvement requires additional physics in their formulation. Abstract: In gas-liquid systems, bubble motion and interaction with the surrounding liquid medium serves to dramatically modify the liquid turbulent kinetic energy profile. While several two-equation bubble-induced turbulence (BIT) models have been advanced to predict this phenomenon, the intrinsic non-linearities that accompany the solution of the governing equations, interfacial forces, and turbulence models complicate their assessment. This hinders understanding of model performance and obstructs necessary model improvements. Here, the mathematical formulation of existing BIT models is investigated, and selected models are quantitatively assessed through simulation of the entire Liu (1989) air/water pipe flow experimental database in OpenFOAM. Critical to this work is the approach adopted to decouple the connection between turbulence and momentum closures, which ensures physically consistent volume fraction profiles and enables fair comparison between models. The assessment reveals that existing closures struggle with reliably predicting the turbulent kinetic energy profile as well as routinely worsen mean flow predictions. These observations are used to propose a pathway for the assembly of new BIT model formulations.