EPA-600/2-75-021-Q
August 1975
Environmental Protection Tfch.-io'oqv Se
EVALUATION OF SYSTEMS
FOR CONTROL OF EMISSIONS
FROM ROCKET MOTORS -
PHASE i
33
ro
UJ
0
U.S. Environmental Protsction Agency
Office of Research and Development
Washington, D.C. 20460
-------
EPA-600/2-75-021-a
EVALUATION OF SYSTEMS
FOR CONTROL OF EMISSIONS
FROM ROCKET MOTORS -
PHASE I
by
Seymour Calvt-rt and Samuel Staibrrg
A. P. T. , Inc.
4901 Morena Boulevard, Suite 402
San Diego, California 92117
Contract No. 68-02-1328, Task 8, and
Interagency Agreement No. EPA-IAG-R5-0644
ROAP No. 21ADL-010
Program Element No. 1AB012
EPA Project Officer: Leslie E. Sparks
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, North Carolin 27711
Prepared for
U.S . AIR FORCE ROCKET PROPULSION LABORATORY
Edwards Air Force Base, California 93523
and
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, D.C. 20460
August 1975
-------
EPA REVIEW NOTICE
This re-port has been reviewed by the National Environmental Research
Center - Research Triangle Park, Office of Research and Development.
EPA, and approved for publication. Approval does not signify that the
contents necessarily reflect the views and policies of the Environmental
Protection Agency, nor does mention of trade names or commercial
products constitute endorsement or recommendation for use.
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environ-
mental Protection Agency, have been grouped into series. These broad
categories were established to facilitate further development and applica-
tion of environmental technology Elimination of traditional grouping was
consciously planned to foster technology transfer and maximum interface
in related fields. These series are:
1. ENVIRONMENTAL HEALTH EFFECTS RESEARCH
2. ENVIRONMENTAL PROTECTION TECHNOLOGY
3. ECOLOGICAL RESEARCH
4. ENVIRONMENTAL MONITORING
5. SOCIOECONOMIC ENVIRONMENTAL STUDIES
6. SCIENTIFIC AND TECHNICAL ASSESSMENT REPORTS
9. MISCELLANEOUS
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed to
develop and demonstrate instrumentation, equipment and methodology
to repair or prevent environmental degradation from point and non-
point sources of pollution. This work provides the new or improved
technology required for the control and treatment of pollution sources
to meet environmental quality standards.
This document is available to the public for sale through the National
Technical Information Service, Springfield, Virginia 22161.
Publication No. EPA-600/2-75-021-a
11
-------
ACKNOWLEDGEMENT
A.P.T., Inc. wishes to express its appreciation
for excellent technical coordination and for
very helpful assistance in support of our effort
to Mr. Jack Hewes, A.F.R.P.L. Program Manager and
Dr. Leslie Sparks, E.P.A. Project Officer.
111
-------
TABLE OF CONTENTS
PAGE
Acknowledgement iii
List of Figures v
List of Tables vi
Nomenclature vii
Abstract ix
Sections
1. Introduction 1
2. Summary and Conclusions 3
3. Survey of Existing Systems 7
4. General Momentum and Energy Balances 17
5. Momentum Extraction and Reinjection 29
Appendix 43
References 47
IV
-------
LIST OF FIGURES
No. Paj
3-1 Diagram of Pilot Scrubber at
A.F.R.P.L 9
3-2 Predicted Effect of Water Loading On Per-
formance of Scrubber for 2.22x10' Newton
(5,000 Ib.) Solid Rocket. Duct
Diameter is 0.91 m(3 ft), (after
Garrett, 1972 data) 10
4-1 System for Cooling and Slowing
Rocket Exhaust 21
4-2 Water Requirements for Reducing Exhaust
Velocity 24
5-1 Flow Around a Scoop 31
5-2 "k" vs. "x" for a Scoop, With Outlet
Angle "" as a Parameter 34
5-3 Final Stream Velocity vs. Water Input,
With Number of Scoop Banks as Para-
meter. For Each Scoop Bank, x =0.20
and = 20° 37
5-4 Location of Pressure Taps Along Scrubber
System. Numbers Refer to the A.F.R.P.L.
Designation for the Pressure Taps .... 40
5-5 Supporting Scoops on I-beam 41
v
-------
LIST OF TABLES
No. Page
4-1 Solid Rocket Motor Characteristics . . ... 19
4-2 Chemical Composition of Exhaust 20
4-3 Minimum Water Evaporation Rates 27
5-1 Scoop Experiment - Required Data 42
VI
-------
NOMENCLATURE
Latin
2
A = area, m
A 2
exit = area of nozzle exit, m
2
AT = area of nozzle throat, m
a = number of scoop banks
c = velocity of sound, m/sec
F = rocket thrust, newtons (N)
H = specific enthalpy, kcal/kg
k = velocity reduction factor
k = individual 'k' for a scoop bank
£1
kT = overall velocity reduction factor
Ma = Mach number
MW = molecular weight, g/g mole
m.. = mass flow rate of water in stream before
scoop, kg/sec
m? = total mass flow rate of water, sum of
incoming mass flow rate and reversed mass
flow rate, kg/sec
m, = mass flow rate of water after scoop, kg/sec
P = pressure, atm
Q. = actual volumetric flow rate, m /sec
QN = normal volumetric flow rate, ni /sec
R = gas constant = 1.987 —— (Chapter 4)
g mole °K
= o.082 * " atm (Appendix)
g mole °K
VII
-------
T = temperature, °K
v = velocity, m/sec
V-. = droplet velocity before scoop, m/sec
V? = final droplet velocity after mixing with
reversed stream, m/sec
w = mass flow rate, kg/sec
Aw0 = rate at which water is evaporated, kg/sec
/C
x = area ratio of scoop inlet to duct cross-
section, or mass fraction of water reversed
y = fraction of "m1 " being extracted at steady
state x
SUBSCRIPTS
0 = rocket chamber
1 = plane 1 on Figure 4-1
2 = plane 2 on Figure 4-1
g = gas
£ - liquid
ns = gas other than steam
r = rocket exhaust
st = steam
GREEK
= specific heat ratio
= angle between scoop inlet and main stream,
degrees
= density, g/cm
= angle between scoop outlet and main stream,
degrees
Vlll
-------
ABSTRACT
Design studies related to the control of emissions
from solid rocket test firings are presented in this report.
Literature in the subject area and contact with those
installations treating rocket exhausts are summarized.
4
A pilot scale scrubber currently in operation on a 2.22 x 10
newton (5,000 Ib) motor at the Air Force Rocket Propulsion
Laboratory in Edwards, California is examined.
If a similar scrubber is usJed on a large engine (e.g.
2.00 x 10 newton or 450,000 Ibs) , the amount of liquid required
to lower the exhaust velocity is quite high. As a remedy,
the installation of scoops is proposed to recycle the
liquid. Theoretical performance characteristics for the
scoops are derived, and an experiment to test their
utility is proposed.
Work is still in progress on designing such an
experiment and in devising other methods, involving both
wet and dry systems, for reducing the exhaust velocity
and temperature.
IX
-------
Section 1
INTRODUCTION
Executive Order 11507 and Air Force Regulation
19-1 direct the Air Force to reduce air pollution
resulting from its rocket propulsion research and
development activities. In addition, the San
Bernardino (California) Air Pollution Control District
has ruled that rocket tests of less than five minutes
duration will be exempt from emission standards
with the exception of the release of halogenated
compounds.
In order to comply with these orders, the
Air Force Rocket Propulsion Laboratory (A.F.R.P.L.)
at Edwards, California has contracted with the
Environmental Protection Agency (E.P.A.) to "define
the technology required for reducing the gaseous
and particulate emissions from rocket motor firings."
The primary emissions to be controlled are partic-
ulates (Al?0,) and the gases CO, HC1 and HF, which
u J
result from solid rocket motor firings. The E.P.A.,
in turn, has contracted with A.P.T. (Air Pollution
Technology), Inc. to perform Contract No. 68-02-1328,
Task No. 8, concerning "Evaluation of Systems for
Control of Emissions from Rocket Motors."
This report was prepared by A.P.T., Inc. as
required in E.P.A. Contract 68-02-1328, Task No. 8.
-------
The object of the task effort was to conduct an
engineering study of existing systems and equipment used
to remove gaseous and particulate matter from effluent
gas streams to determine what systems might be applic-
able to cleaning the exhaust from test firing of rocket
motors. The primary emissions requiring control are
CO, HC1, and HF gases and AL-O. particles. Contact
was to be made with the Air Force Rocket Propulsion
Laboratory at Edwards, California for additional in-
formation on the emissions from rocket motors.
Work on the development of scrubber technology
for the control of rocket exhaust is continuing under
Task No. 9 of Contract E.P.A. 68-02-1328.
-------
Section 2
SUMMARY AND CONCLUSIONS
The evaluation of systems for control of emissions
from rocket motors has been partially completed at
this time. The study indicates that the type of
scrubber now in operation at A.F.R.P.L. as a pilot
system requires an excessive amount of water if
scaled up for a rocket of larger thrust. Therefore,
if water or some other scrubbing liquid is to be
utilized for the purposes of cooling, slowing and
scrubbing the exhaust, then some method for reducing
the liquid used must be devised.
SURVEY OF EXISTING SYSTEMS
A literature search was carried out to
evaluate the present technology of treating exhausts
from rockets. Utilization of the resources of the
Air Pollution Technical Information Center and the
Defense Documentation Center uncovered just one report
directly applicable to this study. The system described
therein, which served as the design basis for the pilot
scrubber at A.F.R.P.L^ was discussed and evaluated.
Governmental installations and private organizations
which have had experience in handling exhausts from
rockets were contacted, and information gained from
them summarized.
GENERAL MOMENTUM AND ENERGY BALANCES
Momentum and energy balances were performed on
a straight-through scrubbing system using liquid
-------
injection, similar to the pilot scrubber. Initial
conditions were obtained primarily from data
supplied by A.F.R.P.L., and calculations were
4
carried out for three thrust levels, 2.22 x 10 N
(5,000 Ibs), 3.78 x 105 N (85,000 Ibs) and
2.00 x 106 N (450,000 Ibs) .
MOMENTUM EXTRACTION AND REINJECTION
In order to reduce the quantity of liquid
needed to lower the exhaust velocity to some specified
level, it was proposed that scoops be placed inside
the scrubber duct. These devices would remove momentum
from the stream in the form of water droplets, which
would then be turned around and reinjected at an
angle opposing the gas flow. The theoretical effec-
tiveness of the scoops was modeled and found to depend
on the outlet angle and the area ratio of scoop inlet
to duct cross-section. An experiment is to be carried
out on the pilot scrubber to test the theoretical
expressions.
CONCLUSIONS
1. A straight-through scrubber requires an excessive
amount of water for larger solid motors. Therefore,
a method of reducing that quantity must be devised.
2. Scoops provide one simple means of accomplishing
this objective.
3. The scoop idea should be tested experimentally on
on the pilot system, and a program to achieve this
should be planned.
-------
The proper placement of scoops or any other
obstruction in the flow path depends heavily
on the extent of the supersonic core of gas.
In this region the exhaust maintains its extreme
velocity and temperature conditions , whereas
gas outside the core is considerably slower and
cooler.
Other methods, involving both wet and dry
systems, for lowering the extreme velocity and
temperature of the exhaust must be studied.
-------
Section 3
SURVEY OF EXISTING SYSTEMS
The resources of the Air Force Rocket Propulsion
Laboratory (A.F.R.P.L.)» the Air Pollution Technical
Information Center (A.P.T.I.C.) and the Defense Documen-
tation Center (D. D. C.) were utilized to survey the
literature in this field and to contact those government
installations and private companies which have had exper-
ience in treating rocket test firing emissions, or in
handling high temperature, high velocity gases. The in-
formation thus gained is summarized in the remaining
material of this section.
The only publication which is directly applicable
to the present study is by Garrett (Garrett et. al., 1972).
This report served as the design basis for the pilot scrub-
ber presently installed on the 2.22 x 10 N (5,000 Ib)
test engine at A.F.R.P.L. Thus, it may be considered as
the state of the art in the field of rocket exhaust scrub-
bing.
As can be seen in Figure 3-1, the rocket exhaust is
collected by a diffuser which ejects it into the spray
chamber. An aqueous solution of potassium hydroxide
(KOH) is introduced at this point to cool the exhaust
by evaporation and to reduce its velocity by absorp-
tion of momentum. These processes continue through the
length of the mixing chamber, and the intimate contact
between the liquid and exhaust provides the opportunity
to scrub out the pollutants. As the gas velocity
decreases, more liquid settles to the bottom of the duct.
What does not is removed by a massive packing entrainment
-------
separator. Wastewater drains from the system to an
evaporation pond. The exit gas and wastewater are
sampled to determine the effectiveness of the unit.
A number of pressure, temperature and flow sensors
are used to monitor the fluid mechanics of the system.
Garrett set up general momentum, energy and material
balances which he solved by computer for a number of
different propellants , thrust levels, scrubber duct
diameters and mass ratios of scrubbing liquid to
rocket propellant. The results of these calculations
formed the basis for the selection of a 91.5 cm (3 ft)
diameter mixing duct for a 2.22 x 10 N thrust solid motor.
Figure 3-2 shows Garrett's predictions for pressure
rise through the scrubber, velocity and temperature at
the scrubber exit and the concentration of potassium
salts in the scrubber effluent, all as functions of
the liquid-to-propellant mass ratio. It can be seen
that the addition of scrubbing fluid has a marked effect
on the exit velocity and on a portion of the potassium
salts curve. After about eight grams of water per gram
of propellant has been injected, additional water has
a much smaller effect on the potassium salts concentra-
tion.
In actual practice at A.F.R.P.L., a high velocity
core of gas existed throughout the entire 20 meter (65.5
ft) length of the scrubber. This gas was harmful to the
entrainment separator and also passed through untreated,
because liquid droplets were unable to penetrate to it.
The problem was solved by installing a core buster - in
this case, a piece of angle iron. This device broke
up the core so that it underwent the same treatment
as the remainder of the exhaust.
-------
ENTRAINMENT
SEPARATOR
LOCATION
OF CORE
DIFFUSER BUSTER
ROCKET
MOTOR
SPRAY
CHAMBER
DRAIN
FIGURE 3-1 - Diagram of pilot scrubber at A.F.R.P.L.
-------
100
o
U 0
o to
«* ^^
P 6
•H
X •»
0 -p
•H
0 0
3 0
fH ,Q
O (~i
w 3
JH
P U
nl t/i
•P -P
to rt
3
rt P
X 3
0 rt
111 ^>
o 0
0 t+H
•P X
rt +-»
0 o
P* O
£ ^
0 0
H >
II II
H >
w
H
o
e »-«
•P U
rt io
•> ti
tH -H
(1)
f-i oj
O V)
0 3
,£ -H
-p (/}
*£^ rt 0
bo -P -P
3 O ccl
O PL, 3:
POO
(U fc -P
»H -P C
3 C 0)
-------
One method which can be used to reduce the
exhaust velocity to subsonic levels is by means of
a diffuser. Basically, if the diffuser is long enough
(L/D > 8) and the back pressure is high enough, the
diffuser should produce the same effect as a normal
shock. Most supersonic diffusers are used to produce
high altitude conditions at the rocket nozzle exit.
The pressure of one atmosphere is then high enough
to allow the diffuser to decelerate the exhaust to
subsonic values. Roschke et. al (1962) made theoretical
and experimental studies of diffusers for simulation
of high altitude flight conditions. One of the
conclusions reached was that predicted phenomena are
not necessarily verified by experiments.
Roschke ran his test on a 2.67 x 10 N (6,000 Ib)
thrust motor. The diffuser was cooled by a water
jacket; yet temperatures in the metal reached as high
as 920° K.
Information regarding the experiences of other
installations treating rocket exhaust is sketchy because
no reports on them have been published. A description
of each of these systems follows:
1. The Naval Ordnance Station, Indian Head, Md. tested a
rocket whose nozzle pointed upward and which released its
exhaust at 2,200-2,300° K. The gas entered a holding
tank 7.62 m (25 ft) long and 1.83 m (6 ft) in diameter,
with nozzles located along the center. The holding
tank volume equaled the exhaust volume produced
by one second of burn. A spray type scrubber
followed. The design was faulty in that it did not
11
-------
account for all of the gas coming out of the
rocket at once.
On larger motors it was found necessary to place
baffles 9-12 m (30-40 ft) from the horizontally
fired rockets. An open system was used, with a
fan between the pre-cooler and the scrubber because
the pressure build-up in a small closed system was
too great to handle.
Arnold Engineering Developement Center, Arnold AFB,
Tn. has worked with closed systems and found there
will be back pressure. It has been observed that the
AL?0_ particles emitted from solid motors are largely
M O
in the sub-micron range. A scanning electron micro-
scope reveals that many of the particles are
spheres (some are hollow) and some are plates.
Growth of HC1 does not occur by self-nucleation.
Instead, the solid particles and water combine,
followed by condensation of HC1 onto the aggregate.
Aerojet Liquid Rocket Company, Sacramento, Ca.
has operated a scrubber system for six to seven
years, handling thrusts from 4,400 N to 1.58 x 105 N
(1,000 to 35,000 Ibs). Burn times for the fluorine-
hydrazine liquid propellant used lasted anywhere from
20 to 2,000 seconds.
The exhaust gas enters a supersonic diffuser 10.7-12
meters (35-40 ft) long at Mach number 3-4 and at
a temperature of 1,900-3,300°K. The diffuser
12
-------
is surrounded by a water-cooled jacket, but no
water is injected into the diffuser.
Ductwork 1.83 meters (6 feet) in diameter follows,
with expansions to 2.44 meters (8 ft) and then to
3.66 meters (12 ft). These sections extend for several
hundred meters. Water is introduced here to cool,
slow and scrub the exhaust. A series of screens
is positioned at the end of this section to act as
an entrainment separator. Aerojet claimed complete
removal of fluorine by means of this system.
4. Atlantic Research Corporation, Alexandria, Va. has
studied the use of a magnesium laser to react with
toxic chemicals. The products of the reaction are
then dissolved in water.
Experience at Atlantic Research has shown that:
A. The temperature and pressure changes caused
by a diffuser also cause a shift in the
equilibrium concentrations of the exhaust
components.
B. Thus far, stainless steel equipment has been
used. At high temperatures HF (which is much
more toxic than HC1) reacts with the metal at
the walls to produce a fluoride coating.
However, this coating is removed by the
Al-O- particles by abrasion, thus exposing
more clean surface. Hence, an erosion problem
exists..
13
-------
Concern was expressed that oxygen added to the system
would cause an explosive reaction with carbon monoxide
and hydrogen at the elevated temperature. It was
also stated that proper diffuser design involves
problems with cooling, materials and flow rates
which require the aid of a diffuser design expert.
Hercules, Inc., Magna, Utah operated an exhaust treatment
system eight to ten years ago in connection with a sub-
scale motor, whose mass flow rate was 4,500 g/sec.
It ran at a simulated altitude of 21,000 meters,
where the ambient pressure is 0.10-0.14 atm. The
main interest of the study was in removing beryllium
oxide particles of an unknown size distribution.
The motor had a 10:1 expansion ratio, the temperature
at the nozzle exit was 2,000-2,500°K, and the velocity
at that point was about 2,600 m/sec.
A straight walled supersonic diffuser 137 cm long
and 25.4 cm in diameter followed. A series of
water jets impinging on the outside kept the dif-
fuser cooled. However, a number of problems which
were never completely satisfied were encountered.
The most prominent of these was sizing the diffuser
to match the mass flow rate from the motor. During
the shutdown transient the diffuser broke down as
the mass, flow rate decreased. A nitrogen purge was
introduced to help counteract this effect. The
pressure rise across the diffuser was a maximum
of about 0.3 atm; no water was injected into the
diffuser.
14
-------
A long quench duct with spray heads to cool the gas
from about 980°K was followed by a vertical cyclone
scrubber whose inlet diameter was 35.2 cm. A removal
efficiency of beryllium oxides of 99.9% by weight was
claimed. The gas then entered a cooling tower through
a 38 cm diameter duct at a temperature of 370-420°K.
A total of 102 fc/sec of water was sprayed in the tower,
cooling the exhaust to 322°K and lowering its velocity
to 58 m/sec at the tower exit.
15
-------
Section 4
GENERAL MOMENTUM AND ENERGY BALANCES
The firing of a rocket motor produces an exhaust
stream of extremely high velocity and temperature.
Momentum and energy must therefore be transferred
from the exhaust in order to protect the treatment
equipment from severe conditions and to facilitate
the scrubbing process.
The gas properties at the nozzle exit plane give
the initial state of the exhaust, which then must be
altered to some specified final state. Data supplied
by A.F.R.P.L. for solid motors of three different
thrust levels are sufficient to calculate nozzle
exit conditions. The data, for thrusts of 2.22 x
104 N ( 5,000 Ib) , 3.78 x 105 N (85,000 Ib) and
2.00 x 106 N (450,000 Ib) appear in Tables 4-1 and
4-2.Equations used to find the calculated values
of Table 4-1 are presented in the Appendix.
Consider a scrubbing system such as the one at
A.F.R.P.L. The water or scrubbing liquid which is
injected in the spray chamber lowers the gas velocity
and temperature. Application of a momentum balance
provides information concerning the amount of water
needed to absorb the momentum of the gas, while an
enthalpy balance tells how much of the injected water
must evaporate to bring down the temperature.
A schematic diagram of an A.F.R.P.L.-type system
is shown in Figure 4-1. Assumptions are made that:
17
-------
1. Liquid enters between planes 1 and 2 in a
radial direction, thus making no contribution to
x- moment urn.
2. There are no frictional losses.
3. The gas and liquid velocities are equal at plane 2.
4. There are no radial variations in stream properties.
5. Inbled air at plane 1 is neglected.
The applicable form of the momentum balance can
thus be written as :
F + P1 (A2 - Ax) = P2A2 + w2v2 (4-1)
where F = rocket thrust, newtons
P = pressure, atm
A = area, m
w = mass flow rate, kg/sec
v = velocity, m/sec
subscript 1 = plane 1
subscript 2 = plane 2
If P, = P2, Equation (4-1) reduces to:
F = P^ + w2v2 (4-2)
If P-jA is negligibly small when compared to w-v- ,
Equation (4-2) can further be reduced to:
F = wv (4-3)
The term "w2" consists of flow from both the
rocket and the liquid. •
w2 = wr + w£ (4-4)
18
-------
TABLE 4-1: SOLID ROCKET MOTOR CHARACTERISTICS
GIVEN VALUES
Chamber Pressure (atm)
Atmospheric Pressure(atm)
Burning Duration(sec)
Nozzle Exit Temp.(°K)
Nozzle Throat Dia.(m)
Nozzle Exit Dia.(m)
Exhaust Flow Rate(kg/sec)
Specific Impulse (m/sec)
Molecular Wt. (g/g mole)
Specific Heat Ratio
CALCULATED VALUES
Gas Density (g/cm3)
Exit Static Pressure (atm)
Exit Velocity(m/sec)
Sonic Vel. at Exit(m/sec)
Exit Mach Number
Exit Flow Rate
(actual m3/sec)
Exit Flow Rate
(normal
.22x10 N
(5,000 Ibs)
3.78x10° N 2.00x10°N
(85.000 Ibs) (450.000 Ibs)
68
1 0
30
2112
-
-
1 8
2600
29
1
1.5.
.0
.898
.55*
.08
.20
xlO"4
n) 0.898
2493
) 850
2
56
7
.93
.6
.32
68.0-204
0.898
75
1460
0.206
0.711
142
2655*
31.79
1.18
2.38xlO"4
.70-2.10
2372
670
3.54
597
112
95.2-163
0.898
60
1460
0.381
1.524
748
2674*
31.79
1.18
2.38xlO"4
.67-1.15
2245
670
3.35
3140
587
* = Calculated (from: Thrust = Mass Flow Rate
X Specific Impulse
19
-------
Table 4-2: CHEMICAL COMPOSITION OF EXHAUST
2.22xl(T N (5,000 Ibs)
3.78x10° N and 2.00x10 N
(85,000 and 450,000 Ibs)
CO
co2
Cl
HC1
H+
HO
H2
H2°
N2
A.l203(s)
Others
MOLES/lOOg
0.749
0.131
0.003
0.619
0.006
0.001
0.867
0.750
0.313
0.260
-
MOLE 1
20.2
3.5
0.1
16.7
0.2
0.1
23.4
20.3
8.5
7.0
-
WEIGHT %
15.12
6.75
0.37
22.86
0.01
0.05
1.15
16.27
9.03
28.32
0.07
MOLE 1
15.8
4.5
0.3
18.3
0.3
0.09
16.8
26.4
9.4
8.1
0.01
20
-------
t tt t
\
A,
Water Injection
FIGURE 4-1 System for cooling and slowing rocket
exhaust.
21
-------
where subscripts "r" and "A" refer to the rocket
and liquid, respectively
Figure 4-2 presents "v2" as a function of "w^" for the
three thrust levels being examined. Water requirements
can be seen to rise sharply as the final velocity
reaches 6 m/sec, a fairly high velocity through the en-
trainment separator (Calvert, et.al., 1974). This
velocity is designated in Figure 4-2 by the dashed
line.
The scrubbing liquid performs its second function,
temperature reduction, primarily by evaporation upon
contact with the hot exhaust. For the purposes of
this discussion, the liquid is assumed to be water.
An enthalpy balance can be written to describe the
cooling process as follows:
H_w-.+ H w = H w + H w (4-5)
r r x,1 £ i g2 *2 *2
where H = specific total enthalpy, kcal/kg
subscript g = gas
Equation (4-5) states that the enthalpy entering the
system, that of the rocket (H w ) plus that of the injected
liquid (H w0), must equal the enthalpy leaving the
J6 -i Af
system, that of the gas phase (H2w , where "w " includes
the rocket exhaust and steam) plus that of the unevap-
orated liquid (H w ). If H = H = H. then Equation (4-5)
X* f) X* *% A/-! O
22
-------
may be written as
Vr
where AW. = w.- w. = rate at which water is
^2 evaporated, kg/sec
The gas phase at the exit consists of evapor-
ated water and the rocket exhaust, assumed to undergo
no mass transfer with or scrubbing by the injected
liquid. Therefore the term MH?w " may be written as:
87
T0C w *
2 pns? ns?
H w = £ L- + H (w + Aw ) (4-7)
L g2 MW S 1
ns2
where T = temperature, °K
C = heat capacity, cal/g mole °K
MW = molecular weight, g/ g mole
subscript ns = gas other than steam
subscript st = steam
Equations (4-6) and (4-7) may be combined to yield:
T r w
12 pns? ns?
H w + H Aw = L- £• + H (w +Aw ) (4-8)
r r I. I st stj_ i
ns2
Equation (4-8) may be solved to find the "Aw "
required to achieve a given temperature "T-.11 The
velocity contribution to energy at "2" has been ne-
glected. The following values have been assigned in
carrying out the calculation:
23
-------
10'
o
o
10
bO
,X
ex
(—(
,-J
PH
O
w
H
O
_4
ttl
CO
CO
10'
10'
10'
0 6
25
50 75 100
• , FINAL VELOCITY; m/sec
125
150
FIGURE 4-2 -
Water requirements for reducing exhaust
velocity.
24
-------
Cpns~ - 4R (4-9)
2
where R = 1.987 cal/g mole °K
w -w .
r st,
\ V^" (4-10)
MW MW ^
r st
2(4184)
(4-11)
where C = 11.91 cal/g mole °K
pl
Other approximate conditions pertaining to plane 1
are taken from Tables 1 and 2. Values for "H " and
X/
"Hst" may be found in a set of steam tables, eg. Schmidt
and List(1962).
The minimum water required to cool the rocket
exhaust is that amount of water that would satisfy the enthal
py balance and saturate the gas with steam. All of the
water is assumed to be converted to steam. The results
for the three rocket sizes are presented in Table 4-3.
It can be seen that the mass rate of liquid evaporation
is three to four times the mass flow rate form the rocket.
However, comparison with Figure 4-2 reveals that both of
these rates are over an order of magnitude below the mass
flow rate of liquid required for velocity reduction. Hence,
water droplets constitute practically the entire mass of
25
-------
the flow stream.
A further examination of the curves in Figure 4-2
indicates that the amount of liquid involved is very large,
particularly in the case of the higher-thrust motors.
Therefore, it is necessary to provide a different method
for removing the momentum from the gas, one which utilizes
less water to perform its function.
26
-------
Table 4-3: MINIMUM WATER EVAPORATION RATES
Rocket Thrust
Newtons
2.22 x 104
3.78 x 105
2.00 x 106
Temperature at "2"
T °r
2'
94
92
92
Water Vapor
Pressure *
atm
0.79
0.77
0.75
Min. Water
Evap. Rate
Aw^, kg/sec
36.4
481.
2250.
* Water vapor partial pressure is calculated from
st.
18 (wr-.'Aw£)
(MWr=20)
Pst + Pr * °'94 atm
P . must then equal the saturation pressure at T9
Sr
27
-------
Section 5
MOMENTUM EXTRACTION AND REINJECTION
As stated in the previous section, a problem
exists in that the amount of liquid required to
reduce the gas velocity in the larger systems is excessive.
The configuration under consideration can be characterized
as a straight-through system; the liquid is injected
and carried through the duct until falling out or
encountering the entrainment separator. If this
fluid could be recycled in order to work more than once,
then the need to introduce more could be eliminated,
and the liquid requirements therefore would be reduced.
It is most desirable to accomplish this objective
with as little additional expense and equipment as
possible. One simple way would be to install scoop-
ing mechanisms, as shown in Figure 5-1 into the mixing
duct. The injection of the liquid provides an opportu-
nity for momentum to be transferred from the exhaust
to the liquid.
This liquid will be collected by a scoop, extract-
ing momentum from the main stream, and then reinjected
at an upstream point. Thus a drop of liquid can work
a second time by absorbing extra exhaust momentum.
Reinjection may be accomplished either radially or at
some angle opposite the gas flow direction. In this
manner a recycling system is set up within the duct
which should lead to an ultimate reduction in the
quantity of liquid injected.
The effect which scoops have on the flow and,
conversely, the effect of the flow on scoops, must
29
-------
be determined on a theoretical basis. As a result of
these calculations, the following scoop characteristics
will be determined:
1. Number at each axial location
2. Number of axial locations
3. Inlet angle, "9"
4. Outlet angle, ""
5. Width
6. Height, penetration to center
7. Shape
8. Material of construction
9. Thickness
The configuration which is finally selected will
be the one which best combines maximum action on the
exhaust stream with minimum structural deformation.
The desired results are opposed to each other; therefore,
compromises in the scoop design parameters will likely
be necessary.
Figure 5-1 is a diagram of a scoop and the flow
around it. Its effect on gas velocity can be obtained
by performing a material and momentum balance around
it.
The following assumptions are made in the analysis:
1. The scoop inlet angle "e" is sufficiently large
that all water striking the scoop is reversed.
2. Only water is reversed. No gas is entrained
in the extracted flow.
3. The water looses no momentum while being reversed,
4. Water leaving the scoop mixes instantly with
the main stream, reducing velocity instantan-
eously.
30
-------
mainstream flow direction
Vi
.route of reversed
water flow
scoop
FIGURE 5-1
Flow around a scoop
31
-------
5. Because of the exhaust's high velocity and
turbulence, velocity profiles are taken to
be flat, and water droplets are distributed
evenly across the duct.
Let "m," be the mass flow rate of water in the
stream before the scoop and "m," be the mass flow
rate of water after the scoop. At steady state, these
two flow rates should be equal to each other, except for
evaporation.
ml = m3 (5-D
Define "y" as the fraction of "m, " being extracted
at steady state and "x" as the area ratio of scoop
inlet to duct cross-section. Since water droplets are
assumed to be evenly distributed, "x" can also be re-
garded as the mass fraction of water reversed.
A material and momentum balance on the scoop
outlet yields :
m, + xm2 = m2 (5-2)
,Vi ~ xm27 cos< = m?2 (5-3)
where V-, = droplet velocity before the scoop
V- = final velocity of the droplet after
mixing with the reversed stream
m- - total mass flow rate, which is equal
to the sum of the incoming mass flow
rate and the reversed mass flow rate
= angle between scoop outlet and main
stream.
32
-------
Under steady state conditions, the following
equation also holds:
ym, = xm~ (5-4)
Combine Equation (5*2) and (5-4) to yield
y = — (5-5)
1-x
From Equation (5-2) one obtains
m, = m?(l-x) (5-6)
JL fc<
By substituting Equation (5-6) into Equation (5-3) ,
the following equation results:
Yl = 1-x (5-7)
Vl 1 + XCOSff)
or V2 = kV (5-8)
where k = — (5-9)
1 + XCOStj)
"1-k" has the physical meaning of velocity re-
duction due to the momentum reversal action of the scoop.
"k" is a function of "(j>" and "x". Once a scoop is in-
stalled in the duct "k" becomes a constant. Figure 5-2
presents "k" as a function of "x" for a scoop, with "<|>"
as parameter.
33
-------
1.0
0.6
0.05
0.10
0.15
0.20
FIGURE 5-2 -"k" vs. "x" for a scoop, with outlet
angle as parameter.
34
-------
Values of "x" greater than 0.20 are not shown
because at any axial position, no more than 201
of the cross-sectional area should be covered by scoops,
in order to avoid a great flow restriction. In reality,
an "x" value of 0.20 may signify one scoop occupying
20% of the duct area or four scoops at the same axial
point, each occupying 5% of the duct area, etc.
Figure 5-2 reveals that the momentum.reducing
effectiveness of a scoop increases with increasing "x"
and with decreasing "". However, the outlet angle also
has a practical lower limit, about 20°.
The next step involves the placement of scoops
at more than one location along the duct. The total
reduction in velocity, from the final velocity when
no scoops are used to that when a certain scoop con-
figuration is employed, is
a
k = II k (5-10)
r a=l a
where a = the number of axial points at which
scoops are located, or number of scoop
banks
k = individual k factor for each scoop bank
k. = overall velocity reduction factor
If each "k " is the same, then
a '
(5-11)
35
-------
A diagram showing final stream velocity as a funct:.on
of water loading rate for a 2.22 x 104 N (5,000 Ib)
was presented in Figure 4-2. This curve is the basis
for Figure 5-3, which shows how the water loading re-
quired to attain a given velocity decreases as more SCOODS
banks are added. Each scoop bank is identical, with
x = 0.20 and 4> = 20°. Therefore, kt can be calculated
by Equation (5-11). The final velocity we wish to attain
is 6 m/sec, the maximum for which reentrainment of drop-
lets from an entrainment separator can be avoided. The
water requirement for reaching that velocity is seen in
Figure 5-3, to become more modest as scoop banks are
added.
In order to examine the validity of these predic-
tions and as a basis for the final design selection,
it is proposed that experiments be run on the pilot
scrubber at A.F.R.P.L. In the course of these ex-
periments, certain data must be examined and observations
made as to how quantities change as the scoop parameters
are altered. Velocity, pressure and temperature measure-
ments should be made at various axial locations in the
duct. The temperature data can serve as a check that
sufficient cooling water is being evaporated. Because
of the great amount of liquid present in the flow stream,
a conventional pitot tube cannot be utilized to find
the velocity. However, that quantity may be arrived at w
indirectly if the pressure, temperature and mass flow
rates of gas and liquid are known. It would also be
advantageous to measure the magnitude of the forces
acting on each scoop, as an aid in determining the
36
-------
150
125 T
100
0>
"e
0
o
c
•H
m
75
50
25
NUMBER
OF
SCOOP
BANKS
102
FIGURE 5-3
Water loading, kg/sec
10-
Final stream velocity vs. water input, with number of scoop banks
as parameter. For each scoop bank, x=0.2 and =20° . Thrust = 2. 2xl04N
-------
rate of liquid flow to the scoop and the structural
requirements. In this way it can be determined how
much of the flow area consists of a supersonic core,
unpenetrated by droplets of liquid.
Load cells can be used to make the measurements
of force on the scoops. In order to guard against the pos
sibility of the scoops cocking and giving misleading
readings, three load cells could be arranged at each
scoop to account for that possibility.
At the present time there are a number of pressure
and temperature taps installed on the pilot system.
In addition, near the duct outlet there are two rakes
designed to give a radial pressure distribution.
One measures the total pressure, while the other detects
static pressure, so that velocities can be determined
here. Figure 5-4 gives the location of the pressure
sensors in the scrubber system.
Two methods have been proposed for mounting the
scoops that would permit the forces acting on them to
be ascertained.
1. The scoops could be welded to the duct wall
or to a sleeve fitting snugly inside the duct.
2. I-beams could be welded to the wall, forming
a rail on which scoops could roll by means of
bearings. A short section could be cut and
welded for each scoop, or several scoops
could be placed on one long beam. Portions
of the support can be cut to allow removal
of scoops at positions other than the end of
the beam.
38
-------
A diagram of this configuration appears in Figure 5-5
The length of the mixing duct currently on the
pilot scrubber, 7.62 m (25 ft), limits the number of
scoop banks which may be installed. Five evenly
spaced banks seems to be the greatest number wherein
one bank does not block flow to the one behind it.
Each bank consists of four scoops, each of which inter-
cepts a total of 5% of the cross-sectional area of the
duct and is positioned 90° from each other around the duct.
The scoops of successive banks are shifted 45° to provide
for a more complete coverage of the flow.
The presence of a core of supersonic gas which
should dissipate as it flows through the mixing duct also
plays a role in scoop design. Near the spray section,
where the core is expected to be greatest in extent,
the scoops themselves should not extend very high; however,
the outlet angle should be large to permit the reinjected
water to penetrate and break up the core as much as
possible. As the core weakens, scoops can be made taller
and their outlet angle decreased in an effort to absorb
more momentum. Penetration to the center becomes less
important as the distance downstream increases.
Therefore, the scoops must serve a number of purposes,
depending on their axial location in the system. In
order to do this, each scoop or bank of scoops may have
a different value for "x" and "". Inlet angle "6" can
always be 180°. The end product is thus likely to
consist of a variety of scoop designs.
Table 5-1 summarizes the data required in the scoop
experiments, which will be used to determine the final
velocity in the mixing duct.
39
-------
1939
1940
1908
ROCKET NOZZLE
1937
1909 1910 1911 1912
1935 1936
RAKE: 1920-1929 TOTAL PRESSURE
1930-1932 STATIC PRESSURE
FIGURE 5-4 - Location of pressure taps along scrubber system. Numbers
refer to the AFRPL designation for the pressure taps.
-------
1\\\\ \ \\
\\\\\\yi
Plate Containing
Scoop
Bearings
I - B e am
\\NA\\M
Axial View
Cut in I-Beam
Top View
FIGURE 5-5 - Supporting scoops on I-beam,
41
-------
Table 5-1. SCOOP EXPERIMENT - REQUIRED DATA
(Refer to Figure 5-4 for locations)
PRESSURE: Static wall along mixing duct.
Total and dynamic at rake.
TEMPERATURE: Static wall along mixing duct.
MASS FLOW RATE: Measure flow rate of liquids
into scrubber.
FORCE: Use load cells to measure axial force
on one scoop in each bank.
42
-------
APPENDIX
The calculated values for exhaust at the nozzle exit,
which appear in Table 4-19 are determined according to
the equations presented in this section.
Gas density is calculated by assuming the exhaust
is ideal and applying the ideal gas law in the form
PCMW)
RT
p --S5— x 10'3 (A-l)
where p = density, g/cm
R = 0.082 £-atm/g mole °k
The exit velocity for the 2.22 x 104 N (5,000 Ib)
motor was figured from the following expression (Bennett
and Myers, 1962) :
(Y-D/Y
v
2
P
o
(A-2)
where y = specific heat ratio
P = pressure
subscript 0 = rocket chamber
Po = 6.81 x 10"3 g/cm3
The speed of sound may be calculated as a function
of temperature according to
/(I.01 x 10a)y RT (A_3)
where c = speed of sound, m/sec
43
-------
Mach number is defined as the ratio
velocity to sonic velocity.
Ma = - (A-4)
where Ma = Mach number
In the case of the 3.78 x 105 N (85,000 Ib) and
2.00 x 10 N (450,000 Ib) motors, chamber density is
not given but the nozzle geometry is known. Mach number
may therefore be calculated from the following equation
(Bennett and Myers, 1962):
y+1
'A. ^ ' - —
1
+XlL
(A-5)
/ I •V*II - II
Ma'
2
where A - = area of nozzle exit, m
2
AT = area of nozzle throat, cm
For these rockets "Ma" is calculated from Equation (A-5)
and "c" is calculated from Equation (A-3) , leaving "v"
to be determined from Equation (A-4) .
In all cases, the actual volumetric flow rate of
exhaust at the nozzle can be figured by:
QA = ? X 10~3 CA-6)
where QA = actual volumetric flow rate, m /sec
44
-------
-------
REFERENCES
Bennett, C.O. and J.E. Myers. Momentum, Heat and Mass
Transfer. McGraw-Hill, Inc. New York, 1962.
Calvert, S., I.L. Jashnani, S. Yung and S. Stalberg.
Entrainment Separators for Scrubbers - Initial Report.
A.P.T., Inc. E.P.A. Contract 68-02-0637, October
1974, E.P.A.-650/2-74-119a.
Garrett, J.W. et. al. A Design Study for Toxic
Rocket Exhaust Gas Cleaning. ARO, Inc. AEDC-TR-72-97,
August, 1972.
Roschke, E.J., P. F. Massier and H.L. Gier. Experi-
mental Investigation of Exhaust Diffusers for Rocket
Engines. Jet Propulsion Laboratory. NASA Contract
NAS 7 - 100, March 1962, TR 32-210.
Schmidt, A. and H. List. Material and Energy Balances.
Prentice-Hall, Inc. Englewood Cliffs, N.J. 1962.
47
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-75-021-a
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Evaluation of Systems for Control of Emissions from
Rocket Motors--Phase I
S. REPORT DATE
August 1975
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Seymour Calvert and Samuel Stalberg
8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING OR9ANIZATION NAME AND ADDRESS
A.P.T. , Inc.
4901 Morena Boulevard
San Diego, CA 92117
10. PROGRAM ELEMENT NO.
1AB012; ROAP 21ADL-010
11. CONTRACT/GRANT NO.
68-02-1328, Task 8
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research-and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Phase I Interim:12/74-6/7
14. SPONSORING AGENCY CODE
is.SUPPLEMENTARY NOTES Report prepared for EPA and for U.S. Air Force Rocket Propul-
sion Laboratory (EPA-IAG-R5-0644), Edwards Air Force Base, CA 93523.
APTtC. 4*
16. ABSTRACT
The report gives results of design studies related to the control of emissions from
solid rocket test firings. It summarizes literature in the subject area and contact
with those installations treating rocket exhausts. It gives results of an examination
of a pilot scale scrubber currently in operation on a 22,200 N (5,000 Ib) motor at
the Air Force Rocket Propulsion Laboratory, Edwards Air Force Base, California.
If a similar scrubber is used on a large engine (e. g. , 2 million.N or 450,000 Ib), the
amount of liquid required to lower the exhaust velocity is quite high. As a remedy,
the installation of scoops is proposed, to recycle the liquid. Theoretical performance
characteristics for the scoops are derived, and an experiment to test their utility is
proposed.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Air Pollution
Exhaust Gases
Solid Propellant
Rocket Engines
Test Facilities
Scrubbers
Liquid Cooling
Exhaust Velocity
Air Pollution Control
13 B
21B
21H
14 B
13A
8. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
58
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
48
------- |