In silico screening of venturi designs and operational conditions for gas-liquid mass transfer applications

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaperJournal articleResearchpeer-review

Venturi ejectors are commonly used to facilitate gas-liquid mass transfer, but their mass transfer characteristics depend on various operational and ejector design parameters, which make experimental identification of optimal conditions cumbersome. A CFD model using a two-phase Eulerian approach with a free surface model to describe local venturi hydrodynamics was developed and used to characterize the influence of gas-liquid flow ratio and venturi design on venturi-mediated gas-liquid mass transfer in bulk liquid. When standardized to gas injection rate, computed venturi-local gas-liquid interfacial area showed same dependency on gas injection rate as experimentally determined k La during aeration of a 1 m 3 tank. The model consequently enables in silico characterization of the venturi flow regime, which is critical to the venturi's mass transfer performance. An experimental factorial design study investigated the influence of venturi geometry on gas-liquid mass transfer and found that the venturi throat diameter is a critical design parameter. The computational model revealed that this was due to its significant impact on generated gas-liquid interfacial area. The ability to characterize venturi flow regimes and qualitatively predict the relative effect of process parameters on k La by modelling local venturi dynamics rather than the complete reactor system greatly reduces computational complexity, suggesting the model's usage as an efficient screening methodology to increase venturi gas-liquid mass transfer performance.

Original languageEnglish
Article number123119
JournalChemical Engineering Journal
Publication statusPublished - 1 Mar 2020

    Research areas

  • Aeration, CFD, Ejector design, Flow regime, Gas-liquid mass transfer, Venturi

See relations at Aarhus University Citationformats

ID: 169164244