Eulerian modeling of inertial and diffusional aerosol deposition in bent pipes. (15th December 2017)

• Record Type: Journal Article
• Authors: Frederix, E.M.A.; Kuczaj, A.K.; Nordlund, M.; Veldman, A.E.P.; Geurts, B.J.
• Is Part Of: Computers & fluids. Volume 159(2017)
• Page Start: 217
• Page End: 231
• Publication Date: 2017-12-15
• Subjects: Aerosol -- Deposition -- Drift -- Diffusion -- Transport -- Bent pipe -- Droplet size distribution; Fluid dynamics -- Data processing -- Periodicals; 532.050285
• DOI: 10.1016/j.compfluid.2017.09.018
• Languages: English
• ISSNs: 0045-7930

Abstract

Highlights: An Eulerian sectional aerosol model is applied to simulation of aerosol deposition. Arbitrary polydispersity is accounted for by the sectional formulation. Both diffusional and inertial aerosol deposition are modeled. Good agreement with literature is found for deposition inside a bent pipe. Dependence of deposition on the Reynolds number and curvature ratio is shown. Abstract: This paper presents a sectional Eulerian aerosol model for size-dependent droplet deposition at walls of the domain, driven by both diffusion and inertia. The model is based on the internally mixed assumption and employs the formulation for compressible aerosols. It is validated in a bent pipe geometry against models and experimental and numerical data from literature. Good agreement is found in both the diffusion and inertial deposition regimes. To improve the overprediction of inertial deposition by a boundary treatment that adopts zero-gradient droplet wall velocity, we use a corrected wall velocity, based on an analytical solution of the droplet motion near the wall. In the bent pipe setting the corrected wall velocity is found to reduce the overprediction of deposition and is less sensitive to grid refinement. We also show that refining the computational mesh near the pipe wall improves the predicted deposition efficiency, significantly. Finally, we present a parameter study varying the Reynolds number and the bend curvature. The deposition efficiency curve is recorded for droplet diameters ranging from the nanometer scale to beyond the micrometer scale, which is a unique contribution of this paper. The complete size range is simulated in only one simulation, due to the sectional approach. In the diffusion-dominated regime an increase in Reynolds number leads to a gradual enhancement of deposition. In the inertial regime, where droplet drift dominates deposition, a much stronger dependence on the Reynolds number is found. The dependence of the deposition on the bend curvature is less pronounced. The results shown in this paper establish the role of Eulerian simulation in predicting deposition of aerosol droplets and are useful for understanding size-dependent aerosol deposition in other more complex confined geometries.