Cfd application cases involving particle-laden flows

PHOENICS-FLAIR is equipped with an Eulerian-based multi-phase model for simulating dispersion and deposition of aerosol particles in indoor environments. Typical applications include studying indoor air quality and designing ventilation systems to deal with: human exposure to biological or radiological aerosols in healthcare or laboratory environments; health hazards from industrial aerosols; protective environments and isolated clean rooms; and surface contamination of artworks, electronic equipment, etc. The aerosol model assumes a very dilute particle phase (one-way coupling) with no collisions or coalescence, and Eulerian drift-flux modelling is used to represent slip between the particle and gas phases due to gravitational effects. In practice, aerosols can be deposited on surfaces by various mechanisms, including particle inertia, gravitational settling, Brownian diffusion (where particles are transported towards the surface as a result of their collision with fluid molecules), turbulent diffusion (where particles are transported towards the surface by turbulent flow eddies), turbophoresis (where particles migrate down decreasing turbulence levels as a result of interactions between particle inertia and inhomogeneities in the turbulence field) and thermophoresis (where temperature gradients drive particles towards or away from surfaces). The PHOENICS model considers all of these mechanisms apart from thermophoresis, which is planned for a future release. The surface-deposition fluxes themselves are calculated by using semi-empirical wall models as a function of particle size, density and friction velocity, and the deposition rates are reported automatically for all surfaces by the CFD solver. The new aerosol model has been validated successfully for particle deposition from fully-developed turbulent air streams in both horizontal and vertical ventilation ducts. For vertical ducts, inertial impaction and gravitational settling are absent, so this case provides a test of deposition influenced by molecular and turbulent processes. The PHOENICS results agree well with the measured data, and the “S-shaped” curve of dimensionless deposition velocity versus particle relaxation time is well simulated, as can be seen in Fig.5. The particle relaxation time provides a measure of a particle’s ability to adjust to new conditions. Small particles “relax” quickly, whereas larger particles are more stubborn, and tend to follow their original trajectory.
casestudy
Figure 5: Comparison of measured and predicted particle-deposition rates on smooth vertical duct walls.
The aerosol model has also been validated for the case of steady, isothermal airflow with aerosol transport and deposition of 10 μm particles in a laboratory-scale room environment. Surface deposition is computed using the 3-layer deposition model of Chen & Lai (2004), which accounts for the deposition mechanisms of gravity, Brownian & turbulent diffusion. Experimental and numerically results have been reported for this case by Chen et al (2006), Lai & Chen (2006, 2007), Zhao & Wu (2007), Gao & Niu (2007), Zhao et al (2008) and Xu & Wang (2017). Particles mainly deposit on the floor, and the model predicts a floor deposition fraction of 69%, which compares favourably with the range of values (60 to 80%) reported by Lai & Chen (2006, 2007) using both Eulerian and Lagrangian CFD models.
For this case, one-half symmetry has been exploited in the simulations, and Figure 6 shows velocity vectors superimposed on contours of the particle mass fraction normalized by its inlet value (C6=C/Cin). Vertical profiles of the predicted particle concentration at three different axial stations show a good match with the measurements of Chen et al (2006), as seen in Figure 7.
casestudy
Figure 6: Ventilated Room: Contour plot of particle concentration.
casestudy
Figure 7: Ventilated Room: Vertical profiles of particle concentration compared with experimental results.

REFERENCES

  • Chen, F.Z. & Lai, A.C.K, "An Eulerian model for particle deposition under electrostatic and turbulent conditions", J.Aerosol Science, Vol.35, p47-62, (2004).
  • Chen, F.Z., Yu, S.C.M., Lai, A.C.K.,"Modeling particle distribution and deposition in indoor environments with a new drift-flux model", Atmospheric Environment 40, 357–367, (2006).
  • Lai, A.C.K., Chen, F.Z.,"Modeling particle deposition and distribution in a chamber with a two-equation Reynolds-averaged Navier–Stokes model", Aerosol Science 37, 1770–1780, (2006).
  • Lai, A.C.K., Chen, F.Z., "Comparison of a new Eulerian model with a modified Lagrangian approach for particle distribution and deposition indoors", Atmospheric Environment 41, 5249–5256, (2007).
  • Zhao, B., Wu, J. "Particle deposition in indoor environments: Analysis of influencing factors", Journal of Hazardous Materials, Vol. 147, Issues 1–2, page 439-448, (2007).
  • Gao, N.P., Niu, J.L., "Modeling particle dispersion and deposition in indoor environments", Atmospheric Environment 41, 3862-3876, (2007)
  • Zhao, B., C Yang, C., Yang, X.,Liu, S.,"Particle dispersion and deposition in ventilated rooms: testing and evaluation of different Eulerian and Lagrangian models", Building and Environment 43 (4), 388-397, (2008).
  • Xu, G., Wang, J.,"CFD modeling of particle dispersion and deposition coupled with particle dynamical models in a ventilated room", Atmospheric Environment 166, 300-314, (2017).