Airflow and nanoparticle deposition in a 16-generation tracheobronchial airway model

In order to achieve both manageable simulation and local accuracy of airflow and nanoparticle deposition in a representative human tracheobronchial (TB) region, the complex airway network was decomposed into adjustable triple-bifurcation units, spreading axially and laterally. Given Q(in) = 15 and 30 L/min and a realistic inlet velocity profile, the experimentally validated computer simulation model provided some interesting 3-D airflow patterns, i.e., for each TB-unit they depend on the upstream condition, local geometry and local Reynolds number. Directly coupled to the local airflow fields are the convective-diffusive transport and deposition of nanoparticles, i.e., 1 nm < or = d(p) < or = 100 nm. The CFD modeling predictions were compared to experimental observations as well as analytical modeling results. The CFD-simulated TB deposition values agree astonishingly well with analytical modeling results. However, measurable differences can be observed for bifurcation-by-bifurcation deposition fractions obtained with these two different approaches due to the effects of more realistic inlet conditions and geometric features incorporated in the CFD model. Specifically, while the difference between the total TB deposition fraction (DF) is less than 16%, it may be up to 70% for bifurcation-by-bifurcation DFs. In addition, it was found that fully developed flow and uniform nanoparticle concentrations can be assumed beyond generation G12. For nanoparticles with d(p) > 10 nm, the geometric effects, including daughter-branch rotation, are minor. Furthermore, the deposition efficiencies at each individual bifurcation in the TB region can be well correlated as a function of an effective diffusion parameter.