American Institute of Aeronautics and Astronautics
1
Computational Fluid Dynamics Modeling of Nasal Airflow
to understand Drug Delivery Process
Goutham Mylavarapu
1
, Mihai Mihaescu
2
, Ephraim Gutmark
3
Department of Aerospace Engineering, University of Cincinnati, Cincinnati, OH, 45221
Shanmugam Murugappan
4
, Lee Zimmer
5
, Allen Seiden
6
Dept. of Otolaryngology, MSB, University of Cincinnati, Cincinnati, OH, 45267
Computational Fluid Dynamics (CFD) is off late used in multidisciplinary studies like
respiratory flows in understanding the flow mechanisms and in optimizing therapeutic and
surgical treatments in patients with respiratory disorders. This study is one such attempt to
understand nasal drug delivery processes. Three dimensional anatomically accurate nasal
airway model is reconstructed from axial Computed Tomography (CT) scans of a patient
using MIMICS®. Computational volume for the nasal airway model is discretized using
TGRID® and Gambit®. Flow and Particle tracking simulations are carried for a range of
peak inspiratory flow rates 7.5, 15, 30, 45 and 60 lpm and particle diameters in range of 0.5 -
30 μm using commercial CFD package FLUENT®. Flow field is solved using a steady RANS
k-ω SST turbulence closure model. For discrete phase modeling (DPM), a stochastic random
walk model with a random eddy life time is used for turbulent dissipation. Effect of particle
deposition efficiencies in nasal airway with several factors like particle diameter, particle
density, turbulence intensity, injection types, flow rate, spray half-cone angles were studied.
Increase in particle diameter, particle diameter or flow rate or all increases impaction factor
and also chances of total particle deposition. Increase in turbulence intensities shows only
small improvement in particle deposition and that too for smaller particle diameters only
(<10 μm). Modes of particle injection studied in this report have not so significant
improvement in terms of total deposition; however local deposition varies with modes of
injections. In spray injection, half cone-angles appear to have little influence on total particle
depositions.
Nomenclature
CFD = computational fluid dynamics
CT = computed tomography
DPM = discrete phase modeling
RANS = Reynolds averaged navier stokes
EIM = eddy interaction model
Q = volumetric flow rate
IF = impaction factor
TKE = turbulent kinetic energy
Re = Reynolds number
TI = turbulence intensity
τ = turbulence length scale
d
p
= particle diameter
ρ
p
= particle density
1
Research Assistant, 6303 MSB, Department of Aerospace Engineering, University of Cincinnati.
2
Research Assistant Professor, 700 Rhodes Lab, Department of Aerospace Engineering, University of Cincinnati.
3
Ohio Eminent Scholar, 745, Baldwin Hall, Department of Aerospace Engineering, University of Cincinnati.
4
Research Assistant Professor, Dept. of Otolaryngology, University of Cincinnati Medical Center.
5
Surgeon, Dept. of Otolaryngology, University of Cincinnati Medical Center.
6
Surgeon, Dept. of Otolaryngology, University of Cincinnati Medical Center.
40th Fluid Dynamics Conference and Exhibit
28 June - 1 July 2010, Chicago, Illinois
AIAA 2010-4735
Copyright © 2010 by Goutham Mylavarapu. Published by the American Institute of Aeronautics and Astronautics, Inc., with permission.