Abstract

Particle deposition is utilized to determine exposure limits, design inhaled medications, and study pulmonary disorders that are a result of airborne pathogens. This information can be obtained from analytical equations, experimental studies, empirical relationships, and numerical analysis.

This research utilizes analytical equations, empirical relationships, and experimental data to investigate the accuracy of three Computational Fluid Dynamics (CFD) software packages; Fluent Discrete Phase Model (DPM), Fluent Fine Particle Model (FPM) and CFX. The sedimentation, molecular diffusion and impaction deposition mechanisms are investigated in a straight tube, bifurcating tube, and three generation lung geometry.

Sedimentation is evaluated in the straight tube for s values between 1E-5 and 1. CFD predictions for parabolic flow are compared to Pich (1972), Wang (1975) and Yeh and Schum (1980), while predictions for uniform flow are compared to Yu, et al. (1977) and Yeh and Schum (1980). Diffusion is evaluated in the straight tube for A values between 1E-6 and 1E-1. CFD predictions for parabolic and uniform flow conditions are compared to Ingham's (1975) analytical equation for parabolic and uniform flow conditions, respectively. Impaction is evaluated in the bifurcating tube for Stokes number between 0.017 and 0.27. CFD predictions in a bifurcating tube are compared to Kim and Iglesias' (1989) experimental data for nearly the same geometry and theoretical predictions from Zhang, et al. (1997), Cai and Yu (1988), and Yeh and Schum (1980) for parabolic and uniform flow conditions. Finally, CFD predictions in the three generation lung geometry are compared to experimental data gather by Dr. Oldham at the University of California, Irvine for 3 um, and 10 um in particles at 1.5 1pm (121pm tracheal flow rate) and 1 um, 3 um, and 10 um at 7.5 1pm (60 lpm tracheal flow rate).

Fluent FPM aligns almost exactly with predictions from Pich (1972) and Wang (1975) for sedimentation from parabolic flow at all 8 values in the straight tube. Fluent DPM and CFX agree well with predictions from Yu, et al. (1977) for sedimentation from uniform flow at all s values investigated in the straight tube. Fluent FPM is the only software package able to accurately predict deposition by diffusion in the straight tube at the A value investigated when compared to predictions from Ingham (1975). There is substantial variation in the analytical equations and CFD predictions for deposition by impaction in the bifurcating tube geometry.

In the three generation lung geometry, CFD predictions from CFX are able to accurately predict experimental data at the 7.5 lpm flow rate for the 10 um particle size with combined impaction and sedimentation and the parabolic velocity profile. Fair to poor correlation was obtained at all other particle sizes at both the 1.5 and 7.5 lpm flow rates. Discrepancies between CFD and theoretical predictions in straight tube and three generation lung geometry are consistent for the sedimentation and diffusion deposition mechanism; therefore, simple straight tube geometry predictions can be used to ascertain the uncertainty in CFD predictions for more complicated geometries.

Library of Congress Subject Headings

Particles--Measurement; Fluid dynamics--Data processing--Evaluation; Atmospheric deposition--Mathematical models; Lungs--Mathematical models; Lungs--Foreign bodies; Respiration--Mathematical models

Publication Date

2005

Document Type

Thesis

Student Type

Graduate

Degree Name

Mechanical Engineering (MS)

Department, Program, or Center

Mechanical Engineering (KGCOE)

Advisor

Risa J. Robinson

Advisor/Committee Member

Steven W. Day

Advisor/Committee Member

Amitabha Ghosh

Comments

Physical copy available from RIT's Wallace Library at TD156.P3 S69 2005

Campus

RIT – Main Campus

Share

COinS