Numerical simulation of aerosol deposition in a 3-D human nasal cavity using RANS, RANS/EIM, and LES

Reynolds Averaged Navier–Stokes (RANS) simulations using a shear stress transport (SST) k-ωk-ω-based turbulence model with both mean flow tracking and random-walk/Lagrangian stochastic eddy interaction-models (EIMs) and large eddy simulation (LES) with Lagrangian particle tracking were used to model deposition of particles in a human nasal cavity geometry. Deposition of monodisperse particles (0.354–16μm diameter) was studied for steady inhalation flow rates, Q, ranging from 7.5 to 60 L/min (through a single nasal passage) and various inlet turbulence intensities. Results show that the RANS/EIM approach tends to significantly over-predict particle deposition for small impaction parameters (da2Q<200) whereas “mean flow particle tracking” tends to under-predict deposition over a wide range of da2Q. LES particle deposition results show better agreement with in vivo data and detailed analysis shows that regional particle deposition efficiency is strongly affected by particle size. For larger particle sizes, inlet geometry of the nasal passage is thought to be important in calculating local deposition. Caution is needed when a standard RANS/EIM is used to predict deposition of particles in a human nasal cavity.