Juan C. Lasheras and Alberto Aliseda University of California, San Diego Department of Mechanical and Aerospace Engineering Wittaker Institute for Bio-Medical Engineering La Jolla CA 92093-0411 [email protected]Liquid Atomization and Sprays for Particle Liquid Atomization and Sprays for Particle Coating and Encapsulation Coating and Encapsulation
79
Embed
Liquid Atomization and Sprays for Particle Coating and ...maeresearch.ucsd.edu/lasheras/papers/Pfpres.pdf•INTRODUCTION • Liquid Jet Breakup and Atomization • Motivation • Dissertation
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Juan C. Lasheras and Alberto Aliseda
University of California, San DiegoDepartment of Mechanical and Aerospace
EngineeringWittaker Institute for Bio-Medical Engineering
Al Berchielli. Pfizer, Inc.Groton/New London Laboratories
Process principles
Spray
Droplet impact (collision )
with the particle
Spray/Air flow InteractionDroplet evaporationHeat transfer
Characteristics of the spray
Drop size distributionDrop velocity distributionDrop volume fraction
Loading of particlesOrientation of the particlesParticle/spay interaction timeCycling time
Particles Air Flow
Coat Drying, Re-crystallization , etc.
Spray
Droplet impact (collision )
with the particle
Spray/Air flow InteractionDroplet evaporationHeat transfer
Characteristics of the spray
Drop size distributionDrop velocity distributionDrop volume fraction
Loading of particlesOrientation of the particlesParticle/spay interaction timeCycling time
Particles Air Flow
Coat Drying, Re-crystallization , etc.
LIQUID ATOMIZATIONLIQUID ATOMIZATION
• atomization consists of the disintegration of a liquid mass into a multitude of small droplets
• in general, the disruption of a liquid mass to form smaller fragments occurs when the stabilizing influences of surface tension and viscosity are overcome by distorting forces
a cylindrical liquid column is unstable when its length exceeds its perimeter and when this condition occurs, two drops will form which have less surface energy than the original
columnDdrop ~ Djet
• pressure atomization ⇒
Ddrop << Djet
Vjet ⇑
LIQUID JET BREAKUPLIQUID JET BREAKUPGAS-ASSISTED or AIRBLAST GAS-ASSISTED or AIRBLAST
ATOMIZATIONATOMIZATION
• the breakup and atomization of a liquid jet injected into a high-speed gas stream is fundamentally different from that which occurs for the same jet discharging into a stagnant gas
• breakup ⇒ kinetic energy transfer from the gas to the liquid
⇒ gas-assisted or airblast atomization
GAS
GAS
Coaxial Atomization
• coaxial jets present a simple way to mix two fluid streams
• common arrangement in fuel-injection applications and many other technologies
MULTI-PARAMETER, TWO-PHASE FLOW PROBLEM
GAS
GAS
LIQUID
BREAKUP OF A LIQUID JET BY A HIGH-SPEED GAS STREAM
COAXIAL JET AND DROPLET BREAKUP COAXIAL JET AND DROPLET BREAKUP REGIMES IN HIGH-SPEED GAS FLOWSREGIMES IN HIGH-SPEED GAS FLOWS
COAXIAL JET BREAKUP REGIMES Lasheras and Hopfinger, Ann. Rev.
Fluid Mech. 2000
DROPLET BREAKUP REGIMES Hsiang and Faeth, Int.
J. Multiphase Flow. 1992
10
15
20
25
30
35
0 20 40 60 80
x/D g
Dro
ple
t S
MD
(µ
m)
OBSERVED SPATIAL VARIATION OF OBSERVED SPATIAL VARIATION OF THE MEAN DROPLET DIAMETERTHE MEAN DROPLET DIAMETER
• beyond the initial breakup region, the mean drop size is observed to evolve with downstream distance
• the droplet SMD first decreases, reaching a minimum value, and then subsequently increases slightly with remaining downstream distance
• secondary droplet breakup mechanisms and other effects associated with the convection of the droplets downstream are necessary to explain these observations
Droplet SMD as a function of downstream distance in a coaxial jet spray
SMD=∑N i di3/N id i
2
Ul
Ug
Ug
• identify the mechanism of primary droplet formation
• develop a phenomenological model for the primary droplet size
• characterize the various secondary breakup mechanisms and effects that lead to downstream variations in the mean droplet diameter
EXPERIMENTAL SETUPEXPERIMENTAL SETUP
Air Air
Water
11.2 mm
1.0 mm
48.0
mm
95.0 mm
59.0
mm
Water
Air Air
1.0 mm
95.0 mm
11.2 mm
48.0
mm
59.0
mm
NOZZLE GEOMETRIES
Convergent Gas Nozzle Straight Channel Gas Nozzle
EXPERIMENTAL SETUP EXPERIMENTAL SETUP AND TECHNIQUESAND TECHNIQUES
Reservoir
Fan
Spray
Atomizer Rig
Traverser
x
Figure 3.4: Experimental Flow Facility
High-Speed Video
Reservoir
PDPA Receiver
PDPA Transmitter
Injector Assembly
Strobe Light
Figure 3.5: Top-View of Experimental Measurement Facility
EXPERIMENTAL CONDITIONSEXPERIMENTAL CONDITIONS
Rel = UlDl/νl
Oh = µl/(ρlσDl)1/2
We = ρg(Ug-Ul)2Dl/σ
Reg = Ug(Dg-Dl)/νg
M = ρgUg2/ρlUl
2
m = ρlUlAl/ρgUgAg
Agl = (Dg2-Dl,o
2)/Dl2
AirAir
Water
UgUg
Ul
Exhaust
Solenoid Valve
Dg
Dl
Dl,o
Dg = 11.2 mm, Dl,o = 1.3 mm,
Dl = (1.00 mm, 0.32 mm)
OBSERVATIONSOBSERVATIONS
Ul
Ug
Ug
Liquid Breakup Process
•the bulk of the liquid atomization is completed in the first few gas jet diameters, well within the gas potential cone by a catastrophic breakup process which generates a fine mist of droplets
•segments of the liquid jet surface are exposed to large perpendicular accelerations
Vary Gas Boundary-Layer ⇒ dg - convergent and straight
channel gas nozzles
Vary Liquid Injection Orientation ⇒ parallel and transverse
DROP-SIZE MEASUREMENTSDROP-SIZE MEASUREMENTS
Droplet size is insensitive to the liquid jet diameter
0
5
10
15
20
25
30
35
40
45
50
0 10 20 30 40 50 60 70
x/D g
Dro
ple
t S
MD
(µ
m)
Dl = 1.0 mm
Dl = 0.32 mm
Droplet SMD as a function of x/Dg, Ug = 165 m/s, m = 0.07
Droplet size is sensitive to the gas boundary-layer thickness
DROP-SIZE MEASUREMENTSDROP-SIZE MEASUREMENTS
20
30
40
50
60
70
80
90
40 60 80 100 120 140 160 180
U g (m/s)
Dro
ple
t S
MD
(µ
m)
Straight Nozzle
Convergent Nozzle
Droplet SMD as a function of Ug at x/Dg = 15, Ul = 1.7 m/s, Dl = 1.0 mm
DROP-SIZE MEASUREMENTSDROP-SIZE MEASUREMENTS
0
5
10
15
20
25
30
35
40
0 10 20 30 40 50 60 70
x/D g
Dro
ple
t SM
D (
µm)
Water
Ethanol
0
20
40
60
80
100
120
140
160
180
20 40 60 80 100 120 140
U g (m/s)D
rop
let
SM
D (
µm)
Water
Ethanol
Droplet SMD as a function of Ug and x/Dg for water and ethanol jets
Droplet size is significantly affected by surface tension
DROP-SIZE MEASUREMENTSDROP-SIZE MEASUREMENTS
SMD Water/ SMD Ethanol as a function of x/Dg
Drop-size ratio ≈ σ water/σethanol ⇒ d ∝ σ1/2
1.40
1.50
1.60
1.70
1.80
1.90
2.00
0 5 10 15 20 25
x/D g
SM
D W
ate
r/S
MD
Eth
ano
l
DROP-SIZE MEASUREMENTSDROP-SIZE MEASUREMENTSEthanol droplet SMD as a function of x/Dg for
Ul = 1.7, 15.5 m/s, Ug = 165 m/s
Droplet size is only slightly sensitive to the liquid stream velocity
0
5
10
15
20
25
30
0 10 20 30 40 50 60 70
x/D g
Dro
ple
t S
MD
(m
)
Ul = 1.7 m/s
Ul = 15.5 m/s
DROP-SIZE MEASUREMENTSDROP-SIZE MEASUREMENTS
0
5
10
15
20
25
30
35
40
0 10 20 30 40 50 60 70
x/D g
Dro
ple
t S
MD
(m
)
Lateral Injection
Parallel Injection
Atomization quality is slightly superior for transverse injection compared to parallel liquid injection
Droplet SMD as a function of x/Dg Ug = 165 m/s, m = 0.07, Dl = 0.32 mm
DROP-SIZE SCALINGDROP-SIZE SCALING
1
10
100
1000
1 10 100
We dg1/2
Dro
ple
t S
MD
(µm
)
Ethanol and Water DataD l = (0.32 mm, 1.0 mm)
U l = (1.7 - 16.6 m/s)
U g = (30 - 200 m/s)
Droplet SMD as a function of Wedg1/2 at x/Dg = 15 for
water and ethanol data
Droplet SMD collapses very well with Wedg1/2
~ 1 cm
1 mm
Joseph et al., UMN Drop Breakup
UCSD Coaxial Jet Breakup
INITIAL JET BREAKUPINITIAL JET BREAKUP
• visual observations of the near-field breakup of large-area-ratio coaxial jets have revealed a mode of liquid jet breakup which appears to share common features with the disintegration of liquid droplets in high-speed gas streams
• waves developed at the liquid jet surface by the primary shear instability are drawn out and destabilized by a secondary instability which develops along these wave and ligament surfaces
• droplets appear in images to be ‘stripped’ from the surface and a visible mist is observed to emanate from the waves in a manner similar to the mist production which occurs in the accelerative destabilization of liquid droplets
Breakup Model
~ 1 cm
1 mm
Joseph et al., UMN Drop Breakup
UCSD Coaxial Jet Breakup
INITIAL JET BREAKUPINITIAL JET BREAKUP
• visual observations of the near-field breakup of large-area-ratio coaxial jets have revealed a mode of liquid jet breakup which appears to share common features with the disintegration of liquid droplets in high-speed gas streams
• waves developed at the liquid jet surface by the primary shear instability are drawn out and destabilized by a secondary instability which develops along these wave and ligament surfaces
• droplets appear in images to be ‘stripped’ from the surface and a visible mist is observed to emanate from the waves in a manner similar to the mist production which occurs in the accelerative destabilization of liquid droplets
ATOMIZATION OF A SMALL-DIAMETER LIQUID JET BY A HIGH-SPEED GAS STREAMATOMIZATION OF A SMALL-DIAMETER LIQUID JET BY A HIGH-SPEED GAS STREAM BREAKUP CONCLUSIONS BREAKUP CONCLUSIONS
• the breakup process for large Weber numbers in the present coaxial jet geometry occurs through an accelerative destabilization of the primary wave surfaces and an aerodynamic stripping process which closely resembles that which occurs in the breakup of liquid droplets by high-speed gas streams
• a phenomenological initial breakup model has been developed which proposes that the primary droplet size should scale with the critical wavelength of the Rayleigh-Taylor instability
• predictions from the RT model have been shown to compare very well with droplet-size measurements and instability wavelengths in the aerodynamic breakup of both water and ethanol jets
ATOMIZATION OF A SMALL-DIAMETER LIQUID JET BY A HIGH-SPEED GAS STREAMATOMIZATION OF A SMALL-DIAMETER LIQUID JET BY A HIGH-SPEED GAS STREAM INITIAL BREAKUP CONCLUSIONSINITIAL BREAKUP CONCLUSIONS
• the initial breakup process for large Weber numbers in the present coaxial jet geometry appears to occur through an accelerative destabilization of the primary wave surfaces and an aerodynamic stripping process which closely resembles that which occurs in the breakup of liquid droplets by high-speed gas streams
• a phenomenological initial breakup model has been developed which proposes that the primary droplet size should scale with the critical wavelength of the Rayleigh-Taylor instability
• predictions from the RT model have been shown to compare very well with droplet-size measurements and instability wavelengths in the aerodynamic breakup of both water and ethanol jets
Mean centerline droplet velocity decays• for given initial distributions
of droplet size and droplet velocities, ni(x) may be calculated, provided that information of vi(x) exists
• ni(x) ⇒ SMD(x)
v i
dv i
dx=
34CD i
ρ g
ρ l
v i−Ug x 2
d i
CDi=24 /Rei1Rei
2 /3/6
-25
-15
-5
5
15
25
35
0 10 20 30 40 50 60
x/D g
Ud -
Ug
(m
/s)
5
10
15
20
25
30
Dro
ple
t S
MD
(m
)Ud>50 - Ugas
SMD15
20
25
30
35
0 20 40 60 80x/D g
Dro
ple
t S
MD
(M
ea
su
red
an
d
Ca
lcu
late
d f
rom
Ac
ce
lera
tio
n)
SMD Calculations
SMD experimental
RELATIVE ACCELERATION EFFECTSRELATIVE ACCELERATION EFFECTSDroplet SMD as a function of x/Dg
Ug = 165 m/s, Ul = 1.7 m/s, Dl = 1.0 mm
• the minimum SMD occurs at the same location where the slip velocity between the gas and the largest droplets is a maximum
• calculated droplet SMD evolution due to relative acceleration effects compared with measured SMD evolution
ATOMIZATION OF A SMALL-DIAMETER LIQUID JET BY A HIGH-SPEED GAS STREAMATOMIZATION OF A SMALL-DIAMETER LIQUID JET BY A HIGH-SPEED GAS STREAM SECONDARY BREAKUP CONCLUSIONSSECONDARY BREAKUP CONCLUSIONS
• an analysis of several secondary breakup mechanisms has been carried out to determine the dominant mechanisms which affect the downstream evolution of the drop-size distribution in coaxial-jet sprays
• the creation of a polydisperse size distribution in the initial breakup process leads to a convective shuffling effect wherein the relative acceleration of different drop size classes leads to a downstream variation in the mean drop size
• the current experimental results support the theory that droplet-droplet collisions leading to shattering or fragmentation are the most plausible explanation for the observed reduction in the mean droplet size in the region which follows the initial breakup process
COAXIAL ATOMIZATION AND THE PDECOAXIAL ATOMIZATION AND THE PDE
2
3
4
5
6
7
8
9
140 160 180 200 220 240
Atomizer Air Velocity (m/s)
∆P
(psi
)
15
17
19
21
23
25
SMD
(µm
)
∆ P
SMD
0
5
10
15
20
25
30
140 160 180 200 220 240
Atomizing Air Velocity (m/s)
SMD
(µm
)Exp. Water Data
Water Predicted
JP-10 Predicted
σJP-10 < σwater
• coaxial atomizer pressure drop requirements are lower than the air pressure drop available at altitude and sufficient for optimum droplet atomization
• at air-injection velocities greater than 220 m/s, which are realizable under typical PDE operating conditions, predicted JP-10 SMD droplet diameters are less than 10 µm (SMD values below 10 µm are currently assumed to be necessary in practice)
Ph.D. Dissertation Presentation – February 15th, 2002
0. 000
0. 016
0. 032
0. 048
0. 064
0. 080
0. 096
0 2 4 6 8 10 12
r/D g
LMF
(g/c
m2 /s
) S = 0
S > S cr
0. 000
0. 004
0. 008
0. 012
0. 016
0. 020
0. 024
0. 028
0 2 4 6 8 10 12 14
r/D g
LM
F (
g/c
m2 /s
)S = 0
S > S cr0. 000
0. 002
0. 004
0. 006
0. 008
0. 010
0. 012
0 2 4 6 8 10 12 14
r/D g
LM
F (
g/c
m2/s
)
S = 0
S > S cr
x/Dg = 20
Radial LMF Distributions at Various Downstream Locations
x/Dg = 40
x/Dg = 60
• Radial liquid mass flux distributions are dramatically Radial liquid mass flux distributions are dramatically affected for supercritical swirl numbers, becoming affected for supercritical swirl numbers, becoming significantly more uniform in the radial directionsignificantly more uniform in the radial direction
SWIRL-ENHANCED ATOMIZATION SWIRL-ENHANCED ATOMIZATION AND FUEL DISPERSIONAND FUEL DISPERSION
Ph.D. Dissertation Presentation – February 15th, 2002
• A high degree of A high degree of control of the liquid control of the liquid mass flux mass flux distribution is distribution is possible through possible through variation of the variation of the
swirl numberswirl number
0.000
0.008
0.016
0.024
0.032
0 2 4 6 8 10 12r/D g
LMF
(g/c
m2 /s
)
S = 0
S = 0.18
S = 0.27
S = 0.47
Radial LMF Distributions at x/Dg = 40
CONTROL OF THE RADIAL LMF DISTRIBUTIONS CONTROL OF THE RADIAL LMF DISTRIBUTIONS WITH VARYING SWIRL NUMBERWITH VARYING SWIRL NUMBER
Ph.D. Dissertation Presentation – February 15th, 2002
• Supercritical gas swirl creates a radial size distribution which is nearly uniform across the spray and also reduces mean drop sizes
8
9
10
11
12
13
14
0 1 2 3 4 5 6 7 8 9 10 11
r/D g
d10
(µ
m)
S = 0
S = 0.47
VARIATION OF THE MEAN DROPLET VARIATION OF THE MEAN DROPLET DIAMETER FOR DIAMETER FOR SS = 0 AND = 0 AND SS > > SScrcr
Ph.D. Dissertation Presentation – February 15th, 2002
• Experimental results indicate that from the viewpoint of ‘good’ atomization quality, coaxial injection is indeed a feasible option for meeting the transient fuel-injection needs of the PDE
• Coaxial atomizer pressure drop requirements have been shown to be lower than the air pressure drop available at design Mach numbers and altitudes, and are sufficient for optimum droplet atomization; SMD < 9 µm for JP-10
• The addition of varying amounts of gas swirl has been shown to facilitate the tailoring of the radial liquid mass flux distribution and yield overall reductions in mean droplet sizes
Ph.D. Dissertation Presentation – February 15th, 2002
0
10
20
30
40
50
60
70
80
90
0 20 40 60 80 100
x/D g
SM
D (
m)
Ug = 50 m/s, m =0.23,M =1.0, We =38
Ug = 80 m/s, m =0.14,M =2.6, We =100
Ug = 100 m/s, m =0.12,M =4.1, We =158
Ug = 165 m/s, m =0.07,M =11.1, We =437
Ph.D. Dissertation Presentation – February 15th, 2002
DROP-SIZE SCALINGDROP-SIZE SCALING
Droplet SMD collapse with We1/2 is very good
0
20
40
60
80
100
120
140
160
180
0 10 20 30 40 50We 1/2
Dro
ple
t S
MD
(m
)Dl=0.32mm, Ul=16.6, Water
Dl=1.0mm, Ul=1.7, Water
Dl=0.32mm, Ul=5.0, Water
Dl=0.32mm, Ul=5.0, Ethanol
Dl=1.0mm, Ul=1.7, Ethanol
Droplet SMD as a function of We1/2 at x/Dg = 15 for water and ethanol data