CFD study on particle grouping under an oscillatory flow in a wavy duct

Anoop K. Gupta, Shing Bor Chen, Liya E. Yu, Ammu Prhashanna, David Katoshevski

    Research output: Contribution to journalArticlepeer-review

    1 Scopus citations


    Particle grouping that arises from temporal and spatial variation of air flow is a promising method to facilitate agglomeration and removal of aerosol. We conduct a computational investigation on this phenomenon under an oscillatory flow in a wavy duct using a discrete phase method. This setup is simple and practically feasible. The grouping behavior is affected by the flow Reynolds number, duct wavelength, angular frequency of oscillation, ratio of minimum to maximum duct diameter and velocity ratio. Three different regimes are identified: no grouping, stable/strong grouping and unstable/weak grouping. It is found that a moderate delay of flow response to the imposed pulsations is critical for grouping. Among the investigated values for the periodicity parameters, the strongest grouping takes place when the duct wavelength is four times the maximum duct diameter, and the angular frequency of oscillation is five times the reciprocal of characteristic flow time. Moreover, an increase in the velocity ratio and a decrease in the diameter ratio further enhance the grouping behavior under otherwise identical conditions. Due to the flow complexity in this realistic setup, the criterion for particle grouping is different from that based on an ideal 1D standing wave flow field in the literature.

    Original languageEnglish
    Pages (from-to)303-313
    Number of pages11
    JournalSeparation and Purification Technology
    StatePublished - 15 Apr 2019


    • Discrete phase modelling
    • Oscillatory flow
    • Particle grouping
    • Reynolds number
    • Wavy duct

    ASJC Scopus subject areas

    • Analytical Chemistry
    • Filtration and Separation


    Dive into the research topics of 'CFD study on particle grouping under an oscillatory flow in a wavy duct'. Together they form a unique fingerprint.

    Cite this