Microgravity Droplet Combustion: Space-Based Experiments and Detailed Numerical Modeling

busyicicleΜηχανική

22 Φεβ 2014 (πριν από 3 χρόνια και 5 μήνες)

69 εμφανίσεις

Advanced Combustion Theory and
Modeling

April 8, 2011

Microgravity Droplet Combustion:

Space
-
Based Experiments and Detailed
Numerical Modeling



q
R


i
i
H

Anthony J. Marchese, Ph.D.

Associate Professor

Dept. of Mechanical Engineering

Colorado State University

Overview

Microgravity Combustion and Heat Transfer

Why study liquid fuel combustion?

Why study liquid fuel combustion in
microgravity?

Spherically symmetric, time
-
dependent
numerical modeling

Ground based microgravity experiments

2.2 second drop tower

5 second Zero Gravity Facility

Space based microgravity experiments

FSDC
-
1, DCE and FSDC
-
2

Continuing Research

Microgravity flame spread through layered
gas mixtures

Microgravity boiling heat transfer


Why are we still studying fossil fuel
combustion?

Reliability on fossil fuels continues...

85%

of all energy consumed in the U.S. is
derived from the combustion of fossil fuels.

39%

of all energy consumed in the U.S. is
derived from the combustion of
liquid fossil
fuels.


97%

of all energy consumed in the
transportation sector is derived from the
combustion of
liquid fossil fuels.


Meanwhile, emissions standards continue to
tighten…

California NO
x

Standards for Gasoline
-
powered*

light duty vehicles:



1971

4.0 g/mile


1993

0.4 g/mile


2003

0.2 g/mile

* Note: Diesel
-
powered light
-
duty vehicles no longer for sale in
California.

Developing accurate models of the combustion
process is the key to designing
more efficient,
cleaner burning engines
...

The
physical phenomena

occurring in an internal
combustion engine includes:

vaporization,

mass transfer,

heat transfer,

turbulent fluid mechanics, and

complex chemical kinetics (~ 100 species).

The problem is
three dimensional

and
time
-
dependent
…and

impossible to solve
with even the
most powerful computers!

Why are we still studying fossil fuel
combustion?

The
spherically symmetric

combustion of a
single liquid droplet in an infinite oxidizing
medium can be solved numerically in
full detail:



Real vapor/liquid equilibrium,



multi
-
component gas
-
phase transport,



liquid
-
phase heat and mass transport,



radiative heat transfer,



detailed gas
-
phase chemistry (~100 species),



time
-
dependent.

r

Multi
-
component

liquid fuel droplet

Flame

What fossil fuel combustion problem

can

we solve?


n
-
C
7
H
16


O
2


C
8
H
18

C
2
H
5

CH
3

H

OH

Spherically Symmetric

Droplet Combustion

By creating and igniting a single liquid droplet in
microgravity

it is possible to achieve
spherically
symmetric

combustion ...

...
of droplets large enough to permit accurate
photographic analysis.

The experimental results are compared
directly

with detailed numerical modeling.

r

1 g

10
-
6

g

Time
-
Dependent, Spherically Symmetric,

Bi
-
component Droplet Combustion Model


m
v



T
r
q
R

m
p
Droplet Interior:

dr
dt
s



l
T
r
D
Y
r
l
i
l
i
,
,








Y
t
r
r
r
D
Y
r
l
i
l
i
l
i
,
,
,







1
2
2








l
p
l
l
l
l
C
T
t
r
r
r
T
r
,







1
2
2






t
r
r
r
v
g
g
r
(
)
(
)


1
0
2
2










g
g
i
g
r
g
i
g
g
i
r
i
g
i
Y
t
v
Y
r
r
r
r
Y
V
,
,
,
,
,
(
)




1
2
2















g
p
g
g
g
p.
g
r
g
g
g
R
g
g
i
i
n
r
i
p
g
i
g
g
i
i
n
g
i
C
T
t
C
v
T
r
r
r
r
T
r
q
Y
V
C
T
r
H
,
,
,
,
,
,
,
(
)
(
)









1
2
2
1
1
Mass Conservation:

Species Equations:

Energy Conservation:

Droplet Surface:



Surface regression



Evaporation of fuel



Condensation of products



Radiative heat addition

Gas Phase:



Multicomponent molecular diffusion



Complex chemical kinetics


(e.g. 50 species, 250 reactions)



Non
-
luminous thermal radiation



UV flame emission

Net Radiative Heat Flux

(Cho,
et al
., 1992; Marchese and Dryer, 1996)

Gas Phase Chemical Kinetic Mechanism

N
-
Alkane Droplet Combustion

Goals
:



Generate test matrix
, and
analyze results

of
DCE

n
-
heptane experiments using detailed,
transient numerical model.

Existing Chemical Kinetic Mechanisms:



Too large:

Chakir (1992)
-

72 species

Lindstedt (1995)
-

109 species




or too empirical:

Warnatz (1984)
-

32 species,

96 reactions

for detailed, time
-
dependent, one
-
dimensional
diffusion flame

modeling.

Result

A new compact semi
-
empirical n
-
heptane
mechanism* has been developed that includes:

Fuel thermal decomposition

Site
-
specific H
-
atom abstraction

37 species, 241 reactions



*
Held, Marchese and Dryer (1997).

N
-
Heptane Droplet Combustion

Fuel Consumption Path

C
7
H
16
2-
C
7
H
15
1-
C
7
H
15
3
-C
7
H
15
4-
C
7
H
15
+
H
+
H
12%
14%
24%
10%
+ H
+ H
7 %
17%
16%
9 %
CH
3
1-
C
6
H
13
C
2
H
5
1-
C
5
H
11
C
3
H
7
C
4
H
9
19%
7%
C
2
H
4
C
3
H
6
1-
C
4
H
8
1-
C
6
H
12
1-
C
5
H
10
22%
100 %
90 %
43 %
15 %
93 %


For typical DCE conditions (He/O
2
)during quasi
-
steady combustion:

C
7
H
16



灲潤畣瑳u⡾‵〥)

C
7
H
16
+ H


灲潤pc瑳

(~ 49%)



Decomposition generally dominates over
isomerization for n
-
alkyl radicals.

How do we perform experiments in

microgravity?

Parabolic Flight Aircraft

“The Vomit Comet”

Drop Towers

Orbiting Spacecraft

Earth
-
Based Microgravity Facilities

NASA 5 Second Zero Gravity Facility

Earth
-
Based Microgravity Facilities

NASA Lewis 2.2 second drop tower

Oxidizer/Inert Inlet Ports

Test Chamber

Power Supplies

Back Light

2.2 Second Drop Tower Experiments

Experimental Apparatus

Optical Access Ports

Video Cameras

High Speed Camera

Microprocessor

User Interface

Earth
-
Based Microgravity Facilities

Rowan 1.1 Second Drop Tower

Deceleration System


100 ft
3

welded steel cage


22
-
oz nylon coated polyester
airbag (100 ft
3
)


12
-
inch polyurethane foam mat


Four 6
-
inch PVC Check Valves

1.5 HP, 127 CFM radial blower

Earth
-
Based Microgravity Facilities

Rowan 1.1 Second Drop Tower

2.2 Second Data Analysis System



Back
-
lit, high
-
speed movie camera



Video “set
-
up” camera



Xybion ISG
-
250 CCD video camera

-

UV Transmissive Lens

-

Narrow band interference filter centered


at 310 nm; full
-
width, half
-
max = 10nm

-

Data acquired at 30 fps

Flame
Xybion Camera
Igniters
Droplet
SVHS
Video Cassette Recorders
SVHS
Set Up Camera
07:28:90
07:28:90
Milliken Camera
Needles
Drop Tower Experiments

Methanol Droplet Combustion

Visual Video Image

Ultraviolet Flame Image

Spherically

Symmetric

Diameter
-
Squared History

Pure Methanol Droplets

time / d
o
2
[sec / mm
2
]
0.0
0.4
0.8
1.2
1.6
d
2
/ d
o
2
0.0
0.2
0.4
0.6
0.8
1.0
18% O
2
21% O
2
24% O
2
30% O
2
For 1 mm droplets, the numerical model accurately
reproduces the measured burning rate for pure
methanol droplets in various O
2
/N
2

oxidizing
environments.

OH* Chemiluminescence

Data Analysis

Relationship Between Measured

Signal and Actual OH* Emission Intensity

Recover actual OH* intensity field, F( r), using

the
Inverse Abel Transform
(Dasch, 1992):

F
r
dP
d
r
d
r
(
)
/
(
)
/





1
2
2
1
2




F(r) : Actual OH* emission
P(r) : Line-of-sight integral
x
y
r
r
intensity field
r
by Xybion Camera
projection as measured
P( r): Line of sight integral

projection as measured by

the Xybion Camera

OH* Chemiluminescence

Numerical Modeling (Marchese,
et al
., 1996)

Possible P
roduction
Routes of
Electroni
cally
Excited OH
:
Collisiona
l De
-
excitation
:
CH
O
CO
OH
H
OH
OH
H
O
OH
OH
M
OH
M
Emission
OH
OH
h
k
k
i
k
i
k
d
em

























2
2
2
2
2
2
2
2
1
2
*(
)
*(
)
*(
)
(
)
:
*(
)
(
)







Numerical Modeling Technique:


Calculated OH* Emission [W/cm
3
]:


Incorporate OH* submechanism into gas phase

chemical kinetic mechanism.

i
r
t
N
k
h
C
r
t
OH
a
em
OH
*
*
(
,
)
(
,
)


OH* Chemiluminescence

Methanol Flame Results

0
3
6
9
12
r / r
s
0
3
6
9
12
Mass Fraction O
2
, CH
3
OH, CO, CO
2
Normalized OH* Emission Intensity
0.0
0.2
0.4
0.6
0.8
1.0
0
3
6
9
12
Mass Fraction OH
0.000
0.004
0.008
0.012
0.016
O
2
CH
3
OH
OH
Temperature
CO
CO
2
i
OH*
Flame Structure, t = 0.90 sec

Methanol/35% O
2
/65% N
2
, 1.0 Atm

Instantaneous Flame Position

Pure Methanol Droplets

time / d
o
2
[sec/mm
2
]
0.0
0.1
0.2
0.3
0.4
0.5
0.6
Calculated Flame Position
Diameter [d
OH*
/ d
s
]
0
1
2
3
4
5
6
21% O
2

30% O
2
35% O
2
21% O
2
- Exp.
30% O
2

- Exp.
35% O
2
- Exp.
Predicted location of maximum OH* emission
agrees with experiment to within 1 normalized radii

Space Shuttle Experiments

Fiber Supported Droplet Combustion

Investigation
-

1 (FSDC
-
1)



Completed experiment aboard Space Shuttle


Columbia flight STS
-
73, November 1995.



Droplet diameters: 3 to 5 mm



Fuels:

Methanol

Methanol/Water

Heptane

Heptane/Hexadecane



Droplet Combustion Experiment (DCE)



Isolated

droplet experiments,

up to 5 mm



First flew aboard Columbia

flight STS
-
83 and STS
-

94 in


April and July 1997.



Heptane in O
2
/He environments


Fiber Supported Droplet Combustion

Investigation
-

2 (FSDC
-
2)



Also flew aboard STS
-
83 and STS
-
94



First ever

space
-
based droplet combustion
experiment:


Single and multicomponent droplets


2 to 5 mm initial diameter


suspended on silicon carbide fiber.



Conducted aboard Space Shuttle Columbia as part
of the Second United States Microgravity Laboratory
(USML
-
2), October 1995.

Fiber Supported Droplet Combustion

FSDC
-
1



q
R


i
i
H

time / d
o
2
[sec/mm
2
]
0.0
0.4
0.8
1.2
1.6
2.0
2.4
d
2
/ d
o
2
0.00
0.25
0.50
0.75
1.00
d
o
= 4.6 mm
d
o
= 2.9 mm
d
o
= 1.3 mm
Numerical Model - No Radiation
K
b
= 0.42 mm
2
/s
K
b
= 0.59 mm
2
/s
K
b
= 0.56 mm
2
/s


Measured burning rate
decreases

with
increasing
initial diameter.



Neglecting radiation, numerical modeling
does not

reproduce this phenomenon.

FSDC
-
1 Results

Pure Methanol Droplets (Dietrich,
et al.
, 1996)

Extinction Diameter vs. Initial Diameter
Initial Diameter [mm]
0
2
4
6
8
Extinction Diameter [mm]
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Numerical Model - No Radiation
FSDC-1 Data

Neglecting radiation, the numerical modeling predicts a

linear increase

in extinction diameter with increasing
initial diameter.


Modeling
under
-
predicts

extinction diameter
measurements.



Measured extinction diameter appears to increase
non
-
linearly

with increasing initial diameter.

FSDC
-
1 Results

Pure Methanol Droplets (Dietrich,
et al.
, 1996)

At increased initial droplet diameters, gas phase

radiative heat loss can no longer be ignored!

The Effect of Radiative Heat Loss

in Microgravity Droplet Combustion





Q
m
H
d
dt
r
H
r
d
dt
d
H
r
c
f
c
s
l
c
l
s
s
c
s












4
3
1
2
3
2




Q
A
T
r
r
r
T
r
r
R
f
g
B
f
f
s
s
g
B
f
s
g
s















4
2
2
4
2
3

g
g
g
s
k
PL
k
PL
r





1
exp(
)
In droplet combustion, the vaporization rate is limited by
the rates of diffusion of heat and mass, resulting in:

Meanwhile, the
radiative heat loss

varies as the
radius
cubed
:

d
dt
d
Cons
t
s
(
)
tan
2

Thus, the mass burning rate and overall instantaneous
heat release
rate in the flame is directly
proportional to the
droplet radius
:

Non
-
Luminous Gas Phase Radiation

Model Results

Calculated gas phase species and temperature for 1, 3,
and 5 mm methanol droplets at t = 0.4

FSDC
-
1 Results

Comparison with Radiation Model *

time / d
o
2
[sec/mm
2
]
0.0
0.5
1.0
1.5
2.0
2.5
d
2
/ d
o
2
0.00
0.25
0.50
0.75
1.00
d
o
= 4.6 mm
d
o
= 2.9 mm
d
o
= 1.3 mm
Diameter
-
squared vs. time

for 1, 3, and 5 mm
droplets in air.

*
Marchese and Dryer, 1997

Comparison with FSDC
-
1 Experiments

The Effect of Initial Water Addition

t / d
o
2
[sec/mm
2
]
0.0
0.5
1.0
1.5
2.0
2.5
d
2
/d
o
2
0.0
0.2
0.4
0.6
0.8
1.0
20% H
2
O - d
o
= 4.6 mm
10% H
2
O - d
o
= 4.8 mm
0% H
2
O - d
o
= 4.6 mm
Radiation
No Radiation
Diameter
-
squared vs. time

for methanol/water

mixtures with 0, 10 and 20% initial water content.

Instantaneous burning rate

for methanol/water

mixtures with 0, 10 and 20% initial water content.

t / d
o
2
[sec/mm
2
]
0.0
0.5
1.0
1.5
2.0
2.5
Gasification Rate [mm
2
/s]
0.0
0.2
0.4
0.6
0.8
20% H
2
O - d
o
= 4.6 mm
10% H
2
O - d
o
= 4.8 mm
0% H
2
O - d
o
= 4.6 mm
No Radiation
Radiation
Comparison with FSDC
-
1 Experiments

The Effect of Initial Water Addition

Comparison with FSDC
-
1 Results

Extinction Diameter vs. Initial Diameter

Initial Diameter [mm]
0
1
2
3
4
5
6
Extinction Diameter [mm]
0
1
2
3
4
0% Water
10% Water
20% Water
0% Water
10% Water
20% Water


Model quantitatively predicts
radiative extinction

predicted asymptotically by Chao,
et al.

(1990).



For methanol in air, flames surrounding droplets
greater than about
6 mm

rapidly self
-
extinguish.



Results may have potential impact on
spacecraft fire
safety.



First
isolated

space
-
based droplet combustion
experiment:



n
-
heptane in O
2
/He mixtures



2 to 5 mm initial diameter



no suspension fiber.



Conducted aboard Space Shuttle Columbia as part
of the Microgravity Science Laboratory (MSL
-
1), April
and July, 1997.


Entire range of droplet combustion phenomena have
been observed:


Radiative flame extinction


Diffusive flame extinction


Complete burn out.

Droplet Combustion Experiment

DCE

Droplet Combustion Experiment

DCE

Space Shuttle Experiments

Results

DCE Space Shuttle Experiments

Results

Complete burnout of droplet (d
o

= 3.5
mm)

Radiative Extinction of flame (d
o

= 5 mm)

Model Predictions

Quasi
-
steady Burning Rate

For n
-
heptane/air:



Model accurately reproduces measured burning
rate and variation with initial diameter.

For n
-
heptane/O
2
/He:



Model appears to over
-
predict the burning rate.



Gas
-
phase transport properties?

Ongoing Work

Combustion of Mars
-
Based Metallized Rocket Propellants

The combination of spherically symmetric combustion
modeling and microgravity experiments can be applied to a
host of problems, such as…




T
r
q
R
Magnesium Particle Surface:

Surface regression


Evaporation of fuel

Condensation of products


Radiative
heat transfer

Surface chemistry:
Mg
(s)
+ CO ->
MgO
+ C
Gas Phase:

Mass, Species, Energy Conservation


Multicomponent
molecular diffusion

Complex chemical kinetics:
Mg
(g)
+ CO
2
->
MgO
+ CO

Thermal radiation

Condensed phase species agglomeration
and
thermophoresis
.
Particle Interior:

Energy transport

Species transport
MgO
(s)
Mg
(g)
Mg
(L)
CO
MgO
(s)
C
(s)
CO
2
CO
MgO
(g)
MgO
(g)
Ongoing Work

Microgravity Boiling Heat Transfer

As computers become faster, they
generate more heat. Is it possible
to use boiling heat transfer to cool
computer chips in space
-
based
applications?

T < 300ºF

Experimental Apparatus


36V/30A Power
Supply
Laptop
HP Function
Generator
HP Data
Acquisition Unit
High
-
Speed
Digital
Camera
Lexan Fluid
Box
Immersion
Heater
Thermocouples
Speaker
Controls
½ Inch Thick
Lexan Fluid
Box
Speaker
Copper
Electrodes
(w/ backup)
Platinum Wire
Ongoing Work

Microgravity Boiling Heat Transfer

Rowan Students Conducting Experiment
on NASA KC
-
135



Boiling Heat Transfer Curve (.008" Platinum in water)
0
200000
400000
600000
800000
1000000
1200000
1
10
100
1000
T - T
sat
(C)
Heat Flux (W/m
2
)
Normal Gravity
KC-135
Summary and Conclusions


Experimental techniques

have been developed to
generate spherically symmetric combustion of large
droplets.


Data analysis techniques

have been developed to
accurately determine burning rates and flame position.


Numerical model

accurately reproduces measured
burning rates and flame position for 1 mm size droplets
neglecting radiation.

For larger droplets, gas phase radiation loss can not be
neglected.


Radiation model

predicts that methanol droplets of > 6
mm will radiatively extinguish.


Result has now been verified in FSDC
-
2 and DCE.


Potential significance for spacecraft fire safety
issues.

Transport, chemistry, vapor/liquid equilibrium and
radiation (non
-
luminous and UV emission)sub
-
models
are applicable to more detailed flow situations.

Ongoing work in microgravity heat transfer and
combustion in support of future manned space
activities.