Comparison of computational fluid dynamic predictions and experimental results for local particle deposition patterns in idealized human airways

Risk assessments of potentially toxic inhaled particulate matter (PM ₁₀, PM_2.5, etc.) and delivery of aerosolized medication (antibiotics, bronchodilators, etc.) are based upon predictions of particle deposition in the respiratory tract. Usually, these predictions are for the entire lung, however, some models predict particle deposition for groups of lung airways. Recently, computational fluid dynamic (CFD) modeling has been utilized to predict detailed local particle deposition patterns within an airway generation. Several authors have made predictions of local particle deposition in human lung airways using CFD modeling and compared their predictions with experimental studies in a structurally-similar airway model.; This project tested the hypothesis that a CFD modeling approach will accurately predict total and local particle deposition patterns for particles whose deposition is dominated by impaction in an idealized airway model. This work compared total and local particle deposition predictions using CFD modeling with results from bench-top particle deposition studies in the identical airway anatomy. This comparison used monodisperse 1 μm, 3 μm and 10 μm aerodynamic diameter particles drawn through the models at flow rates representing rest and heavy exertion. CFD predictions were preformed using one of the leading commercially available finite element software packages, FIDAP by Fluent Inc. (Lebanon, NJ).; Results of the bench-top particle deposition studies were consistent with those reported in the literature for both total and local deposition patterns. CFD predictions for airflow field were also consistent with measured airflow fields published in the literature, considering the differences in airway geometry and starting parameters used in this research. CFD predictions for total deposition were significantly different from the results of the bench-top particle deposition studies. Better agreement was found between the bench-top particle deposition studies and CFD predictions of local deposition patterns. However, Chi-Square analysis indicated that only for the 10 μm diameter particles at resting ventilation were the results of CFD predictions and experimental studies from the same population. CFD predictions of the ratio of particles depositing on the top and bottom of the hollow model generally agreed with the experimental studies. These results suggest that CFD predicted particle deposition requires further experimental validation before use in risk assessment.