9/26/2013

1

Aerosol Transport and Deposition

Computational Modeling Particle Removal Environmental Applications

Biomedical Applications

Conclusions

Direct Numerical Simulation Large Eddy Simulation Stress Transport Model Two-Equation Models

Direct Numerical Simulation

Subgrid Scale Simulation

Gaussian Models- Filtered White Noise- Eddy Life Time

Pdf – Based Model

9/26/2013

2

)t(T

2

Tdt

d

L

______2

i

L

ii

uuuu i )t(T

2

Tdt

d

L

______2

i

L

ii

uuuu i

duu

)t(u)t(uT

0pp

____________________pp

L

duu

)t(u)t(uT

0pp

____________________pp

L

Instantaneous Velocity

Lagrangian Time Macro-Scale

Thompson (1987)

forceBrownian

i

forceElectric

Ei

forcenalGravitatio

i

forceLift

Li

forceDrag

pii

PDpi

)t(nfg

fuu1

24

ReC

dt

du

forceBrownian

i

forceElectric

Ei

forcenalGravitatio

i

forceLift

Li

forceDrag

pii

PDpi

)t(nfg

fuu1

24

ReC

dt

du

Re

]Re15.01[24C

687.0

D

Re

]Re15.01[24C

687.0

D

1000 Re 1 1000 Re 1

Assumptions: Dilute Flows, One-Way Interaction,Neglect Particle Collisions

Spectral Intensity

Particle Relaxation Time

White Noise Model

kT2Snn

kT2

Snn

1818

22cc

p CSdCd

1818

22cc

p CSdCd

Ounis, Ahmadi and McLaughlin (1991)Ounis, Ahmadi and McLaughlin (1991)

9/26/2013

3

CRCD Web-Based Course Module

Dt2)t(x 2 Simulations

g

He and Ahmadi (1999)10

-410

-210

010

210

410

-6

10-5

10-4

10-3

10-2

10-1

100

+

ud+

Wood Eq.(34)Fan & Ahmadi Eq.(38)Fan & Ahmadi Eq.(38) withgravity in flow directionPapavergos & HedleyLi & AhmadiHe & AhmadiRSM & "Two-layer"k- & "Two-layer"k-, "Two-layer" & Eq.(23)RSM,"Two-layer" & Eq.(23)

g

Tian and Ahmadi (2007)

9/26/2013

4

Hinze, 1975 22' yv 0y

Li and Ahmadi, 1993

2Ayv 4y

*yuy

*

2'

u

vv 008.0A

Ounis, et al. 1993 (DNS)

Quadratic Variation Near Wall Quadratic Variation Near Wall Particle and fiber depositionin human lung and noseParticle and fiber depositionin human lung and nose

EPA

NIOSH

The schematics of multi-level bifurcation model

At interface aa’:

(a) velocity magnitude (b) x- velocity(c) y- velocity(d) turbulence kinetic energy(e) turbulence dissipation rate(f) Reynolds stress uu(g) Reynolds stress vv

a

a’

a a’

(a)

(b)

(d) (e)

(c)

(g)(f)

9/26/2013

5

1 µm

Translational Motion

Rotational Motion

),()( tmmdt

dm BLhfpp fffg

v

),()(ˆˆˆˆˆˆ

ˆ

ˆtTTII

dt

dI B

x

h

xzyzy

x

x

),()(ˆˆˆˆˆˆ

ˆ

ˆtTTII

dt

dI B

y

h

yxzxz

y

y

),()(ˆˆˆˆˆˆ

ˆ

ˆtTTII

dt

dI B

z

h

zyxyx

z

z

Translational Brownian Force

Rotational Brownian Torque

)'()'(ˆ)(ˆ,0)(ˆ tStttt tBBB ff f

)'()'(ˆ)(ˆ,0)(ˆ tStttt rBBB TT T

/2ˆˆˆˆ

t

ii

t

iiKTRS

/2ˆˆˆˆ

r

ii

r

iiKTRS

Ellipsoidal Fibers Duct Flows

DNS

9/26/2013

6

d = 0.1 μm, aspect ratio = 12, length = 1.2 μm

10-1

100

101

102

10-3

10-2

10-1

100

Trachea and Bifurcation 1

+

Dep

osi

tion E

ffic

iency

Zhou et al. (2007): Cast A, 40-60L/minZhou et al. (2007): Cast B, 40-60L/minZhou et al. (2007): Cast A, 15L/minZhou et al. (2007): Cast B, 15L/minWang & Hopke (2008): 15-22L/minSim. 60L/min, u* = 0.86m/sSim. 60L/min, u

* = 0.53m/s

Sim. 60L/min, u* = 0.20m/s

Sim. 40L/min, u* = 0.53m/s

Sim. 15L/min, u* = 0.53m/s

Sim. 15L/min, u* = 0.20m/s

Sim. 15L/min, Laminar

Velocity magnitude contours in the nose.

NIOSH

9/26/2013

7

Particle size (nm)

Ca

ptu

reE

ffic

ien

cy%

0 20 40 60 80 1000

20

40

60

80

Simulation

Swift et al. - Cast B (1992)

Swift et al. - Cast C (1992)

Cheng et al. - Subject A (1996)

Cheng et al. - Subject B (1996)

Cheng et al. - Subject C (1996)

Cheng et al. - Subject D (1996)

Particle size (nm)

Ca

ptu

reE

ffic

ien

cy%

0 20 40 60 80 1000

20

40

60

80

100 Simulation

Swift et al. - Cast B (1992)

Swift et al. - Cast C (1992)

Cheng et al. - Subject A (1996)

Cheng et al. - Subject B (1996)

Cheng et al .- Subject C (1996)

Cheng et al. - Subject D (1996)

Kelly et al. - SLA (2004)

Kelly et al. - Viper (2004)

4 L/min 10 L/min

Ultra fine Particles

Pe-1

Ca

ptu

reE

ffic

ien

cy%

10-6 10-5 10-4 10-3

20

40

60

80

SimulationEmperical Equation

DL

QPe

DL

QPe

)85.01(1008.0280 Pee )85.01(1008.0280 Pee

Ultra fine Particles

Diameter in Micron

De

po

sitio

nin

Pe

rce

nta

ge

0 2 4 6 8 100

10

20

30

40

50

60

70

80

90

100 3.75 L/min7.5 L/min15 L/min

*

De

po

sitio

nin

Pe

rce

nta

ge

0.1 0.2 0.3 0.410

20

30

40

50

60

70

80

90

100

9/26/2013

8

• Deposition was influenced byanatomy

• Fibers deposit less frequentlythan spheres

7th International Conference on Multiphase Flow ICMF 2010, Tampa, FL, USA

Case Study

Before Septoplasty

After Septoplasty

CT Scan Images

Abouali et al. (2012)

Before Septoplasty After Septoplasty

9/26/2013

9

Before Operation After Operation

Abouali et al. (2012)

Q=7 L/min

The stream lines of the airflow (colored by velocity magnitude

Effect of Thermal Plume and facial

features

Effect of Thermal Plume and facial

features

Air temperature =20oC

Mannequin surface temperature = 34oC

9/26/2013

10

Sajadi et al. (2013)

9/26/2013

11

125h

630

630

x

y

z

Upper Wall

Center Line

Mean Flow

Lower Wall

y

z

One-way coupling

Four-way coupling, M.L.=20%

Four-way coupling, M.L.=40%

Mean Flow Direction

9/26/2013

12

du

Effect of Mass Loading

Two-Way Coupling

Particle Collisions

Four-Way Coupling

Using fundamentals of multiphase flows provides a better understanding ofenvironmental and biomedical systems

Computational modeling helps inoptimizing industrial processes

● Dr. P. Zamankhan ● Prof. J. Tu● Dr. L. Tian ● Dr. Kiato● Dr. Kevin Shanley ● Prof. Bohl● Dr. Mazyar Salmanzadeh ● Dr. M. Shams● Dr. F-G Fan ● Prof. McLaughlin● Dr. C. He ● Prof. Saidi● Dr. K. Nazridoust ● Dr. A. Li● Dr. M. Soltani ● Dr. O. Abouali● Dr. A. Mazaheri ● Dr. W. Kvasnak● Dr. H. Zhang ● Dr. X. Zhang ● Dr. H. Nasr

9/26/2013

13