CHAPTER 18

ROCKET EXHAUST PLUMES

The behavior of rocket exhaust plumes is included in this book because it has
gained importance in recent years. In this chapter we provide an introduction
to the subject, general background, a description of various plume phenomena
and their effects, and references for further study.
The plume is the moving formation of hot rocket exhaust gases (and some-
times also entrained small particles) outside the rocket nozzle. This gas forma-
tion is not uniform in structure, velocity, or composition. It contains several
different flow regions and supersonic shock waves. It is usually visible as a
brilliant flame, emits intense radiation energy in the infrared, visible, and ultra-
violet segments of the spectrum, and is a strong source of noise. Many plumes
leave a trail of smoke or vapor or toxic exhaust gases. At higher altitudes some
of the plume gases can flow backward around the nozzle and reach compo-
nents of the flight vehicle.
The plume characteristics (size, shape, structure, emission intensity of
photons or sound pressure waves, visibility, electrical interference, or smoki-
ness) depend not only on the characteristics of the particular rocket propulsion
system or its propellants, but also on the flight path, flight velocity, altitude,
weather conditions, such as winds, humidity, or clouds, and the particular
vehicle configuration. Progress has been steady in recent decades in gaining
understanding of the complex, interacting physical, chemical, optical, aerody-
namic, and combustion phenomena within plumes by means of laboratory
experiments, computer simulation, measurements on plumes during static fir-
ing tests, flight tests, or simulated altitude tests in vacuum test chambers. Yet
much is not fully understood or predictable. As shown in Table 18-1, there are
639
640 ROCKET EXHAUST PLUMES
TABLE 18-1. Applications of Plume Technol ogy
Design~develop~operate Flight Vehicles, their Propulsion
Systems, and Launch Stands or Launch Equipment
For a given propul si on system and operat i ng conditions (altitudes, weather, speed,
afterburning, with at mospheri c oxygen, etc.) determine or predict the plume
dimensions, t emperat ure profiles, emissions, or ot her plume parameters.
Det ermi ne likely heat transfer to component s of vehicle, test facility, propul si on
system or launcher, and prevent damage by design changes. Include aft erburni ng
and recirculation.
Estimate the ability of vehicle and test facilities to withstand intensive plume noise.
Det ermi ne the aerodynami c interaction of the plume with the airflow ar ound the
vehicle, which can cause changes in drag.
Reduce i mpi ngement on vehicle component s (e.g., plumes from attitude control
thrusters hitting a solar panel); this can cause excessive heating or i mpi ngement
forces that may t urn the vehicle.
Estimate and minimize erosion effects on vehicle or launcher component s.
Prevent deposits of condensed species on spacecraft windows, optical surfaces, solar
panels, or radiating heat emission surfaces.
Det ermi ne the backscatter of sunlight by plume particulates or condensed species,
and minimize the scattered radi at i on that can reach into optical instruments on the
vehicle, because this can give erroneous signals.
Protect personnel using a shoulder-fired rocket launcher from heat, blast, noise,
smoke, and toxic gas.
Detect and Track Flight of Vehicles
Analysis and/ or measurement of plume emission spectrum or signature.
Identify plumes of launch vehicles from a distance when observing from spacecraft,
aircraft, or ground stations, using IR, UV, or visible radiations and/ or radar
reflections.
Distinguish their emissions from background signals.
Detect and identify smoke and vapor trails.
Track and predict the flight path.
Alter the propel l ant or the nozzle to minimize the radiation, radar signature, or
smoke emissions.
Estimate weather conditions for appearance of secondary smoke.
Develop Sensors f or Measuring Plume Phenomena
Improve calibration and dat a interpretation.
Devel op i mproved and novel instruments for plume measurements, for bot h remot e
and close by locations.
TABLE 18-1. (Continued)
18.1. PLUME APPEARANCE AND FLOW BEHAVIOR 641
Improve Understanding of Plume Behavior
Improve theoretical approaches to plume phenomena.
Improve or create novel computer simulations.
Provide further validation of theory by experimental results from flight tests,
laboratory investigations, static tests, or tests in simulated altitude facilities.
Understand and minimize the generation of high-energy noise.
Understand the mechanisms of smoke, soot, or vapor formation, thus learning how
to control them.
Provide a better understanding of emission, absorption, and scatter within plume.
Provide a better prediction of chemiluminescence.
Understand the effect of shock waves, combustion vibration, or flight maneuvers on
plume phenomena.
Understand the effects of plume remains on the stratosphere or ozone layer.
Develop a better algorithm for simulating turbulence in different parts of the plume.
Minimize Radio-Frequen O' Interference
Determine the plume attenuation for specific antennas and antenna locations on the
vehicle.
Reduce the attenuation of radio signals that have to pass through the plume,
typically between an antenna on the vehicle and an antenna on the ground or on
another vehicle.
Reduce radar reflections from plumes.
Reduce the electron density and electron collision frequency in the plume; for
example, by reducing certain impurities in the gas, such as sodium.
many appl i cat i ons or si t uat i ons where a predi ct i on or a quant i t at i ve under-
st andi ng of pl ume behavi or is needed.
18.1. PLUME APPEARANCE AND FLOW BEHAVI OR
The size, shape, and i nt ernal st ruct ure of a pl ume changes dramat i cal l y with
altitude. Fi gure 18-1 shows the const r uct i on of a l ow-al t i t ude pl ume at hei ght s
typically bet ween 3 and 10 km. The pl ume di amet er and l engt h are oft en
several times l arger t han the vehicle di amet er and length. In the near field
t here is an inviscid i nner core (exhaust gases t hat have not yet mi xed wi t h
air) and a relatively t hi n out er mi xi ng layer where oxygen f r om the air burns
t urbul ent l y with the fuel-rich species in the exhaust gases. In the far field the
ambi ent air and exhaust gases are well mi xed t hr oughout a cross section of the
pl ume, and the local pressure is essentially t hat of the ambi ent air. In the
i nt er medi at e field the shock wave intensities di mi ni sh and mor e of the mass
flow is mi xed with ambi ent air. The r adi at i on emi ssi ons come f r om all part s of
6 4 2 ROCKET EXHAUST PLUMES
~ - Near field
I n n e r /
supersonic
core with
shock waves
Plume bow shock
PrandtI-Meyer
expansion fans
Thickness of mixing or
afterburning layer increases
with length
/ /
Transition region , ~ / Far f i el d- - - - - - - - - - ~-
~ji~:~-. --. ............ ................... ' ~ i ~ i
Velocity profile
-----Plume mixing layer
Plume slipstream
....... ;::~"~ . . . . . . . . . . . . . . ! ~ i ;
Mach disk / \ Mach disk
Nozzle exit plane ~__ or normal shock mixing layer
I nviscid supersonic region
FI GURE 18--1. Half sections of schematic diagrams of a rocket exhaust plume at low
altitude. Upper sketch shows full plume and lower sketch is an enlargement of the near
field. (Reprinted with permission from Ref. 18-1.)
the plume, whereas the interactions with the vehicle occur only as a result of
near-field phenomena.
Figure 18-2 shows sketches of the variation of the plume configuration with
altitude. When the nozzle exit pressure is approximately equal to the ambient
pressure (condition for opt i mum nozzle expansion), the plume has a long,
nearly cylindrical shape. With increasing altitude the plume shape becomes
more of a cone and the plume length and diameter increase. The core of the
plume emerges supersonically from the nozzle exit and goes t hrough an oblique
compression shock wave, known as the barrel wave, which originates near the
nozzle exit lip and has the approxi mat e shape of an inverted but somewhat
curved cone. The central part of the plume then goes t hrough the Mach disk,
which is a strong normal compression shock wave; here the gases suddenly
slow down in velocity and are raised to a higher pressure and temperature. The
flow immediately behind the Mach disk is subsonic for a short distance, but
downst ream again becomes supersonic. This pat t ern of normal shock waves
and short subsonic zones is repeated several times in the core of the plume, but
the strength of the shock and the rises in temperature or pressure are reduced in
each sequence.
,~
Pl ume
c o n f i g u r a t i o n
I nner
superson ic
core
Mi x i n g layer
( af t er bur ni ng)
Nozzle exi t
pressure P2
and a mb i e n t
pressure P3
Fl i ght
vel oci t y
A l t i t u d e ,
km
Rh~ kcet --__ , , ' I ] . ~ ~
~ e ~ . .::."::'~
~ ....,..:.~::.:.:~ ..... :':!5.7:
" : i i ~ ! " i ; ! ! : ' i : ; i ~ S h o c k w a v e s ~ : ; . : ? : - " . . ; .
. : . : . ; . ~ - . , : ; . " . . . : " . "
- . . . .'::v ...:..'." . . ' "
: : < _ _ - .:
a f t e r b u r n i n g ~ . v . - : : : ? ...:..-.-. . : - . . . -
r e g i o n .. . . . . . . .
. , -
: : . : : i : : - " i i : . : ! . . "
- ' - , - . . -
. . . . .
- - . . : . ~ . . , .
Bow shock
w a v e ( a i r ) ~ ? ~ ~
." ; i-'" ." ":" '"::!:-'" . -" . . ' aust gas
.. .-" . - : ' shock wave
. . . . :
. .
. . .
. .
. . , . . -
. .
Narrow, can see sever al shock d i a mo n d s
Narrow
P 2 -- P 3
Very low, subsoni c
Ot o 5
Larger di amet er , some shockwave pat t er n,
f ewer v i s i bl e shock waves
Wider, unsteady, t u r b u l e n t
P 2 > P 3
Subsoni c, t r a n s o n i c and s l i g h t y super soni c
10 t o 2 5
Only one or t wo sets of shock waves are v i s i bl e
Very wi de, i r r egul ar
P2 >> P3
Super soni c
Above 3 5
F I GURE 18-2. The visible pl ume grows in l engt h and di amet er as t he r ocket vehicle gains al t i t ude. The af t er bur ni ng of t he fuel-
rich combus t i on pr oduct s with t he oxygen f r om t he air occurs in t he mi xi ng layer. At very hi gh al t i t ude, above per haps 200 km,
t here is no air and t herefore no af t er bur ni ng.
644 ROCKET EXHAUST PLUMES
The ambi ent air mixes with the hot exhaust gases and secondary combust i on
or aft erburni ng occurs in the mixing layer. It is a t urbul ent layer sur r oundi ng
the core and its thickness increases with distance from the nozzle as well as with
altitude. The i ncompl et el y oxidized fuel species in the exhaust gases, such as
H2, CO, NO, or CH2, react chemically with the oxygen from the at mospher e
and are largely burned to H20, CO2, or NO2, and the heat of this secondary
combust i on raises the t emper at ur e and the specific vol ume in this aft erburni ng
layer. As explained in Chapt er 5, most propel l ant s are fuel rich to achieve
opt i mum specific impulse or opt i mum flight performance, so addi t i onal oxida-
tive heat release is possible.
As the altitude increases, the ambi ent local air pressure decreases by several
orders of magni t ude and the pressure ratio in the gases between the nozzle exit
and the local ambi ent pressure is increased greatly, appr oachi ng infinity when
the rocket operat es in a vacuum in space. Wi t h higher altitudes, furt her expan-
sion (increase in specific volume) occurs and this causes a furt her reduct i on of
gas t emper at ur es and an expansi on in bot h di amet er and length; for the prin-
cipal propul si on systems these usually exceed the dimensions of the vehicle.
Some species in the pl ume will condense and become liquid; they will freeze as
the t emper at ur e drops and gases like H20 or CO2 will form clouds or a vapor
trail.
As the vehicle at t ai ns supersonic velocity (relative to the ambi ent air) two
shock waves form. One is an oblique compressi on shock wave in the air ahead
of the vehicle and the ot her is a trailing wave ori gi nat i ng at the vehicle' s tail,
where the air meets the exhaust pl ume gases. These wave fronts are usually
l umi nescent and highly visible and can reach di amet ers of several kilometers.
As the supersonic exhaust gas flow emerges from the nozzle, it experiences
Pr andt l - Meyer - t ype expansi on waves, which at t ach themselves to the nozzle
lip. This expansi on allows the out er streamlines just outside the nozzle to be
bent and an increase in the Mach number of the gases in the out er layers of the
plume. This expansi on can, at higher altitudes, cause some por t i on of the
supersoni c pl ume to be bent by more t han 90 from the nozzle axis. The
t heoret i cal limit of a Pr andt l - Meyer expansi on is about 129 for gases with k =
1.4 (air) and about 160 for gases with k = 1.3 (typical for a rocket exhaust
mixture; see Ref. 18-2). This backwar d flow needs to be anal yzed to est i mat e
the heat and i mpi ngement effects and possible cont ami nat i on of vehicle com-
ponent s (see Ref. 18-3).
The boundar y layer next to the nozzle wall is a region of viscous flow, and
the flow velocity is lower t han in the mai n nozzle inviscid flow. The velocity
decreases to zero right next to the wall. For large nozzles this boundar y layer
can be quite thick, say 2 cm or more. Fi gure 3-16 shows a subsonic and a
supersoni c region within the boundar y layer inside the nozzle divergent section;
it also shows a t emper at ur e and a velocity profile. While the supersonic flow
layer is restricted in the angle t hr ough which it can be deflected, the subsonic
boundar y layer flow at the nozzle lip is in a cont i nuum regime and may be
deflected up to 180 . Al t hough the subsonic boundar y layer represents only a
18.1. PLUME APPEARANCE AND FLOW BEHAVIOR 645
small por t i on of the mass flow, it nevertheless lets its exhaust gases flow back-
ward on the outside of the nozzle. This backflow has caused heating of and
sometimes chemical damage to the vehicle and propul si on system parts.
The mass di st ri but i on or relative density is not uni form, as can be seen in
Fig. 18-3, which is based on a calculated set of dat a for a high-altitude plume.
Here 90% of the flow is within -t-44 of the nozzle axis and only one hundr ed
t housandt h or 10 -5 of the t ot al mass flow is bent by more t han 90 . The flow
near the center cont ai ns most of the heavier molecules, such as CO2, NO2, or
CO, and the out er regions, which are deflected the most, consist largely of the
lighter species, such as most l y H 2 and perhaps some H20.
Figure 18-4 shows the drastic change (log scale) in the overall radi at i on
emission intensity as a funct i on of altitude for a typical three-stage satellite
l aunch to a 300- to 500-km orbi t or a l ong-range ballistic missile with a boost er
stage, a sustainer stage, and a payl oad velocity adj ust ment stage. The boost er-
stage rocket propul si on system gives the largest intensity because it has the
Mass fraction = 10 -10 10 -7 10 -6 10 -5 10 -4 10-3 10-2
\ \ I I I I /
1 0
6
cD
r-
5
._m
.02_
4
o
\
\
\
X~
\
\
\
\
\
\
\
\
\
\
- 5 - 4 - 3 - 2 -1 0 1 2 3 4 5
Axial distance x/Re
I 0 - I
/
FIGURE 18-3. Density profile for vacuum plume expansion using a one-dimensional
flow model for a small storable bipropellant thruster. The axial distance x and the plume
radius R have been normalized with the nozzle exit radius Re. Here k = 1.25, the Mach
number of the nozzle exit is 4.0, and the nozzle cone half angle is 19
6 4 6 R O C K E T E X H A U S T P L U ME S
Vac uum
A f t e r b u r n i n g l i mi t
f l a me d i a me t e r 0 . 0 7 - 1 0 m di a.
is - 1 0 - 1 0 0 m Co n t i n u u m Mo l e c u l a r
f l o w r egi me f l o w r egi me l i ne
~ 0 . 1 - 1 km di a. - 1 - 1 0 km di a.
/
o r
.~_ t A ttitude.
~0 c ont r ol or
0
" , - _ / o r b i t
I I I I I I I i 4, I I I I a d j u s t me n t
t "
0 2 0 4 0 6 0 8 0 1 0 0 1 2 0 1 4 0 1 6 0 2 0 0 3 0 0 4 0 0 5 0 0
Al t i t u d e , km
FIGURE 18-4. For a multistage ascending vehicle the plume radiation intensity will
vary with the altitude, thrust or mass flow, propellant combination, and plume tem-
perature. The four sketches describing the plume are not drawn to the same scale.
highest rocket gas mass flow or the highest thrust, a relatively dense plume, and
its radiation is enhanced by afterburning of the fuel-rich gas with oxygen from
the air. The rise in the intensity of the sustainer stage is due to the large increase
in plume volume caused by the expansion of the exhaust gases. Both operate in
that part of the atmosphere where continuum flow prevails; that is, the mean
free paths of the molecular mot i ons are relatively small, frequent collisions
between molecules occur, the gases follow the basic gas laws, and they can
experience compression or expansion waves.
As higher altitude is reached the cont i nuum regime changes into a free mole-
cularflow regime, where there are fewer molecules per unit volume and the mean
free pat h of the molecules between collisions becomes larger than the key
dimension of the vehicle (e.g., length). Here the plume spreads out even more,
reaching diameters in excess of 10 km. Only the exhaust gases close to the nozzle
exit experience cont i nuum flow conditions, which allows the streamlines in the
flow to spread out by means of successive Pr andt l - Meyer expansion waves;
once the gas reaches the boundary shown by the elliptical dashed line in the
last sketch on the right in Fig. 18-4, the flow will be in the free molecular flow
regime and molecules will continue to spread out in straight lines. The regions of
free molecular flow and the transition from cont i nuum flow can be analyzed as
shown as Ref. 18-4. The third or upper stage, which operates at very high
altitudes, has very low emission intensity, because it has a relatively very low
gas flow or thrust and because only the inviscid port i on of the exhaust gas flow
near the nozzle is hot enough to radiate significant energy. This makes it difficult
to detect and identify from a distance. The phenomenol ogy of rocket exhaust
plumes as seen from a space-based surveillance system is described in Ref. 18-5.
18.1. PLUME APPEARANCE AND FLOW BEHAVIOR 647
Spectral Distribution of Radiation
The primary radiation emissions from most of the plume gases are usually in
the infrared spectrum, to a lesser extent in the ultraviolet spectrum, with rela-
tively little energy in the visible spectrum. The emissions depend on the parti-
cular propellants and their respective exhaust gas compositions. For example,
the exhaust from the liquid hydrogen-liquid oxygen propellant combination
contains mostly water vapor and hydrogen, and with a minor percentage of
oxygen and dissociated species. Its radiation is strong in specific wavelength
bands characteristic of the emissions from these hot gases (such as 2.7 and 6.3
~m, wat er--i nfrared region) and 122 nanometers (hydrogen--ultraviolet
region). As shown in Fig. 18-5, the hydrogen-oxygen plume is essentially
transparent or colorless, since there are no strong emissions in the visible
segment of the spectrum. The propellant combination of nitrogen tetroxide
with methylhydrazine fuel gives strong emissions in the infrared region; in
addition to the strong emissions for H20 and hydrogen mentioned previously,
there are strong emissions for CO 2 at 4.7 ~tm, CO at 4.3 ~tm, and weaker
FIGURE 18-5. Visible plume created by the oxygen-hydrogen propellants of the
Vulcain 60 thrust chamber, with a specific impulse of 439 sec at altitude, a nozzle
expansion area ratio of 45, and a mixture ratio of 5.6. Multiple shockwave patterns
are visible in the core of the plume because of emissions from luminescent minor species.
(Courtesy of ESA/CNES/SEP/Daimler-Benz, Europe.)
648 ROCKET EXHAUST PLUMES
emission in the ultraviolet (UV) and visible ranges (due to bands of CN, CO,
N2, NH3, and other intermediate and final gaseous reaction products). This
gives it a pink orange-yellow color, but the plume is still partly transparent.
The exhausts of many solid propellants and some liquid propellants contain
also solid particles. In Tables 5-8 and 5-9 examples of solid propellant were
given that had about 10% of small particles as aluminum oxide (A1203) in their
incandescent white exhaust plumes; some kerosene-burning liquid propellants
and most solid propellants have a small percentage of soot or small carbon
particles in their exhaust. The radiation spectrum from hot solids is a contin-
uous one, which peaks usually in the infrared (IR) region, but it also has strong
emissions in the visible or UV region; this continuous spectrum is usually
stronger in the visible range than the narrow-band emissions from the gaseous
species in the plume. Afterburning increases the temperature of the particles by
several hundred degrees and intensifies their radiation emissions. With 2 to 5%
solid particles, the plumes radiate brilliantly and are therefore very visible to
the eye. However, these particles in the outer layers of the plume obscure the
central core and the shock wave patterns can no longer be observed.
The visible radiation of plumes from double-base propellant can be reduced
or suppressed by adding a small amount (1 to 3%) of potassium compound.
With composite propellants the control of visible emissions by additives has
not been as effective.
The radiation (which is a function of the absolute temperature to the fourth
power) cools the plume gases, but it also heats adjacent vehicle or propulsion
system components. The prediction of radiative emission requires an under-
standing of the plume composition, the temperature and density distribution in
the plume and the absorption of radiation by intervening atmospheric or plume
gases (see Refs. 18-5 to 18-7). The heat transferred from the plume to vehicle
components will depend on the propellant combination, the nozzle configura-
tion, the vehicle geometry, the number of nozzles, the trajectory, altitude, and
the secondary turbulent flow around the nozzles and the tail section of the
vehicle.
The observed or measured values of the radiation emissions have to be
corrected. The signal strength diminishes as the square of the distance between
the plume and the observation station, and its observed magnitude can change
tremendously during a flight. The attenuation is a function of wavelength, rain,
fog or clouds, the mass of air and plume gas between the hot part of the plume
and the observing location, and depends on the flight vector direction relative
to the line of sight. The total emission is a maximum when seen at right angles
to the plume (see Refs. 18-5 to 18-7). Radiation measurements can be biased
by background radiations (important in satellite observation) or Doppler shift.
Multiple Nozzles
It is c o mmo n t o ha v e mo r e t ha n o ne p r o p u l s i o n s y s t e m o p e r a t i n g at t he s a me
t i me , or mo r e t ha n o n e n o z z l e s e n d i n g o ut ho t e x h a u s t gas pl ume s . F o r e x a m-
18.1. PLUME APPEARANCE AND FLOW BEHAVIOR 649
ple, the Space Shuttle has three mai n engines and two solid rocket boost ers
runni ng simultaneously. The interference and i mpi ngement of these plumes
with one a not he r will cause regions of high t emper at ur e in the combi ned
plumes and t herefore larger emissions, but the emissions will no longer be
axisymmetrical. Also, the multiple nozzles can cause di st ort i ons in the airflow
near the rear end of the vehicle and influence the vehicle drag and augment the
hot backflow from the pl ume locally.
Plume Signature
This is the t erm used for all the characteristics of the pl ume in the i nfrared,
visible, and ul t ravi ol et spect rum, its electron density, smoke or vapor, for a
part i cul ar vehicle, mission, rocket propul si on system, and propel l ant (see Refs.
18-8 to 18-10). In many mi l i t ary appl i cat i ons it is desirable to reduce the
pl ume si gnat ure in order to minimize being detected or t racked. The initial
st agnat i on t emper at ur e of the nozzle exit gas is perhaps the most significant
fact or influencing pl ume signature. As pl ume t emperat ures increase, higher
levels of radi at i on and radi o-frequency i nt eract i on will occur. Emissions can
be reduced if a propel l ant combi nat i on or mi xt ure ratio with a lower combus-
tion t emper at ur e is selected; unfort unat el y, this usually gives a lower perfor-
mance. One way to reduce smoke is to choose a reduced-smoke or mi ni mum-
smoke solid propel l ant ; they are described in Chapt er 12. The pl ume si gnat ure
is t oday often specified as a requi rement for a new or modified rocket -propel l ed
vehicle, and it imposes limits on spectral emissions in certain regions of the
spect rum and on the amount of accept abl e smoke.
The at mospher e absorbs energy in certain regions of the spectrum. For
example, the air cont ai ns some car bon dioxide and wat er vapor. These mole-
cules absorb and at t enuat e the radi at i on in the frequency bands pecul i ar to
these two species. Since many pl ume gases cont ai n a lot of CO2 or H20, the
at t enuat i on within the pl ume itself can be significant. The pl ume energy or
intensity, as measur ed by spect rographi c i nst rument s, has to be correct ed for
the at t enuat i on of the i nt erveni ng air or pl ume gas.
Vehicle Base Geometry and Recirculation
The geomet ry of the nozzle exit(s) and the flight vehicle' s tail or aft base have
an influence on the plume. Fi gure 18-6 shows a single nozzle exit whose dia-
met er is al most the same as the base or tail di amet er of the vehicle body. I f
these two di amet ers are not close to each other, then a flat doughnut - shaped
base or tail surface will exist. In this region the high-speed combust i on gas
velocity is larger t han the air speed of the vehicle (which is about the same as
the local air velocity relative to the vehicle) and an unst eady flow vort ex type
recirculation will occur. This greatly augment s the aft erburni ng, the heat
release to the base, and usually creates a low pressure on this base. This
lower pressure in effect increases the vehicle' s drag.
6 5 0 ROCKET EXHAUST PLUMES
Ai r f l ow
Rocket ~ M i x i n g layer
propulsion
system
flow
Aft end !
of vehicle
Ai r f l ow
Mi x i ng l a y e r
. . . . . . . upersoni__c_c com___b_usti____on gas flo__w
A n n u l a r r ec i r c ul at i on z o n e
FIGURE 18--6. Diagrams of flow patterns around two different boat tails or vehicle aft
configurations, with and without hot gas recirculation.
The air flow pat t ern at the vehicle tail can be different with different tail
geometries, such as cylinder (straight), a diminishing diameter, or an increasing
diameter conical shape, which helps to maintain the vehicle's aerodynami c
stability. The air flow pattern and the mixing layer change dramatically with
angle of attack, causing an unsymmetrical plume shape. Flow separation of the
air flow can also occur. In some cases the recirculation of fuel-rich exhaust gas
mixed with air will ignite and burn; this dramatically increases the heat transfer
to the base surfaces and causes some changes in plume characteristics.
Compression and Expansion Waves
A shock wave is a surface of discontinuity in a supersonic flow. In rocket
plumes it is the very rapid change of kinetic energy to potential and thermal
energy within that very thin wave surface. Fluid crossing a stationary shock
wave rises suddenly and irreversibly in pressure and decreases in velocity.
When it passes t hrough a shock wave surface that is perpendicular or normal
to the incoming supersonic flow, then there is no change in flow direction. Such
a normal shock produces the largest increase in pressure (and local down-
stream temperature) and is known as a normal shock wave. The flow velocity
18.1. PLUME APPEARANCE AND FLOW BEHAVIOR 651
behind a normal shock wave is subsonic. When the incoming flow is at an angle
less t han 90 to the shock wave surface, it is known as a weak compression wave
or as an oblique shock wave. Figure 18-7 illustrates the flow relationships and
shows the angle of flow change. The t emperat ure of the gas i mmedi at el y behind
a normal shock wave approaches the st agnat i on t emperat ure. Here the radia-
tion increases greatly. Also, here (and in other hot plume regions) dissociation
of gas species and chemical luminescence (emission of visible light) can occur,
as can be seen (downst ream of strong shock waves) in Fig. 18-5.
The behavior of gas expansion in the supersonic flow has a fairly similar
geometric relationship. It occurs at a surface where the flow undergoes a
Pr andt l - Meyer expansion wave, which is a surface where pressure is reduced
and velocities are increased. Often there is a series of weak expansion waves
next to each other; this occurs outside the lip of the diverging nozzle exit
section when the nozzle exit gas pressure is higher t han the ambi ent pressure,
as shown in Fig. 18-1.
The plume from hydrogen-oxygen liquid propel l ant combust i on consists
most l y of superheated water vapor and hydrogen gas and is basically invisible.
However, it is faintly visible because of the chemically generated luminescence
t hat is believed to be responsible for the pale pink orange and white skeletal
wave pattern, particularly in its hot regions. The pat t erns are shown in Figs.
18-2 and 18-5.
The supersonic gas flow out of the nozzle exit is undi st urbed until it changes
direction in a wave front or goes t hrough a normal shock. Di amond- shaped
Oblique shock wave
(compression)
r
Normal shock wave
(compression)
v
Expansion wave Multiple expansion waves
FIGURE 18-7. Simplified diagrams of oblique shock wave or compression wave, nor-
mal shock wave, and expansion wave. The change in the length of the arrows is an
indication of the change in gas velocity as the flow crosses the wave front.
652 ROCKET EXHAUST PLUMES
patterns are formed by compression and expansion wave surfaces. These pat-
terns (shown in Figs. 18-2 and 18-5) then repeat themselves and are clearly
visible in largely transparent plumes, such as those from hydrogen-oxygen or
alcohol-oxygen propellant combinations. The pattern becomes weaker with
each succeeding wave. The mixing layer acts as a reflector, because an expan-
sion wave is reflected as a compression wave.
The inter-face surface between the rocket exhaust plume gas and the air
flowing over the vehicle (or the air aspirated by the high velocity plume) acts
as a free boundary. Oblique shock waves are reflected at a free boundary as an
opposite wave. For example, an oblique compression wave is reflected as an
oblique expansion wave. This boundary is not usually a simple surface of
revolution, but an annular layer, sometimes called a slip stream or mixing
layer. See Figs. 8-1, 8-2, and 8-5.
18.2. PLUME EFFECTS
Smoke and Vapor Trails
Smoke is objectionable in a number of military missiles. It interferes with the
transmission of optical signals, such as with a line-of-sight or electro-optical
guidance system. Smoke would also hamper the vision of a soldier who guides
a wire-controlled anti-tank weapon. Smoke, or a vapor trail, allows easy and
rapid detection of a missile being fired, and visually tracking the flight path can
reveal a covert launch site. Smoke is produced not only during rocket opera-
tion, but also by chuffing, the irregular combustion of propellant remainders
after rocket cutoff. Chuffing, described in Chapter 13, produces small puffs of
flame and smoke at frequencies of perhaps 10 to 150 Hz.
Primary smoke is a suspension of many very small solid particles in a gas,
whereas secondary smoke is a set of condensed small liquid droplets suspended
in a gas, such as condensed moisture-forming clouds, fog, or mist. Many
propellants leave visible trails of smoke and/or vapor from their plumes during
powered rocket flight (see Refs. 18-8 to 18-10). These trails are shifted by local
winds after the vehicle has passed. They are most visible in the daytime,
because they depend on reflected or scattered sunlight. The solid particles
that form the primary smoke are mainly aluminum oxide (A1203, typically
0.1 to 3 lam diameter) in composite propellants. Other solid particles in the
exhaust of solid propellant are unburned aluminum, zirconium or zirconium
oxide (from combustion stabilizer), or iron or lead oxides (in burn-rate cata-
lyst). Carbon particles or soot can be formed from various solid propellants
and liquid propellants using hydrocarbon fuels.
During the external expansion of rocket exhaust plume gases the gas mix-
ture is cooled by radiation, gas expansion, and convection with colder ambient
air to below its dew-point temperature, where the water vapor condenses. Of
course, this depends on local weather conditions. If the ambient temperature is
18.2. PLUME EFFECTS 653
low (e.g., at high altitude) and/ or if the gas expands to low t emperat ures, the
wat er dropl et s can freeze to form small ice crystals or snow. At high altitude,
CO2, HC1, and ot her gases can also condense. Many rocket exhaust gases
cont ai n between 5 and 35% water, but the exhaust from the liquid hydro-
gen/liquid oxygen propel l ant combi nat i on can cont ai n as much as 80%. If
the exhaust cont ai ns tiny solid particles, these then act as nuclei upon which
wat er vapor can condense, thus increasing the amount of nongaseous mat eri al
in the plume, maki ng the pl ume even more visible.
I f reduci ng smoke in the pl ume is i mpor t ant to the mission, then a reduced-
smoke solid propel l ant or a mi ni mum- s moke propel l ant is often used. They are
described in Chapt er 12. Even then, a secondary smoke trail can be formed
under col d-weat her and hi gh-humi di t y conditions. However, under most
weat her condi t i ons it will be difficult to see a trail cont ai ni ng vapor only.
Toxi ci t y
The exhaust gases from many rocket propul si on systems cont ai n toxic and/ or
corrosive gas species, which can cause severe health hazards and pot ent i al
envi r onment al damage near l aunch or test sites. Acci dent al spills of some
liquid oxidizers, such as ni t rogen t et roxi de or red fumi ng nitric acid, can create
toxic, corrosive gas clouds, which have higher density t han air and will stay
close to the ground. Exhaust gases such as CO or CO2 present a heal t h hazar d
if i nhal ed in concent r at ed doses. Gases such as hydr ogen chloride (HC1) from
solid propel l ant s using a perchl orat e oxidizer (see Ref. 18-11), ni t rogen dioxide
(NO2) , ni t rogen t et roxi de (N204) , or vapors of nitric acid (HNO3) have rela-
tively low levels of allowable i nhal at i on concent r at i on before heal t h damage
can occur. Chapt er 7 lists some of the safe exposure limits. The pot ent i al
damage increases with the concent r at i on of the toxic species in the exhaust ,
the mass flow or t hrust level, and the dur at i on of the rocket firing at or near the
t est / l aunch site.
Di spersi on by wind and diffusion and dilution with air can reduce the con-
cent rat i ons of toxic mat eri al s to tolerable levels within a few minutes, but this
depends on local weat her condi t i ons, as explained in Chapt er 20. Careful
at t ent i on is therefore given to scheduling the l aunch or test operat i ons at
times when the wi nd will carry these gases to nearby uni nhabi t ed areas. For
very highly toxic exhaust gases (e.g., those cont ai ni ng beryl l i um oxide or cer-
tain fluorine compounds), and usually for relatively small t hrust levels, the
exhaust gases in static test facilities are capt ured, chemically t reat ed, and pur-
ified before release into the at mosphere.
Noise
Acoust i cal noise is an unavoi dabl e by- pr oduct of thrust; it is part i cul arl y
i mpor t ant in large l aunch vehicles and is a pr i mar y design consi derat i on in
the vehicle and in much of the gr ound- suppor t equi pment , part i cul arl y elec-
654 ROCKET EXHAUST PLUMES
tronic components. Besides being an operational hazard to personnel in and
around rocket-propelled vehicles, it can be a severe annoyance to communities
near rocket test sites. The acoustic power emitted by the Saturn V vehicle at
launch is about 2 x 108 W, enough to light up about 200,000 average homes if
it were available as electricity.
Acoustic energy emission is mainly a function of exhaust velocity, mass of
gas flow, exhaust gas density, and the velocity of sound in the quiescent med-
ium. In these terms, the chemical rocket is the noisiest of all aircraft and missile
propulsion devices. Sound intensity is highest near the nozzle exit and
diminishes with the square of the distance from the source. Analytical models
of noise from a rocket exhaust usually divide the plume into two primary areas,
one being upstream of the shock waves and one being downstream (subsonic),
with high-frequency sound coming from the first and low-frequency from the
second. The shock wave itself is a generator of sound, as is the highly turbulent
mixing of the high-velocity exhaust with its reltively quiescent surroundings.
Sound emission is normally measured in terms of microbars (gbar) of sound
pressure, but is also expressed as sound power (W), sound intensity (W/ft2), or
sound level (decibels, dB). The relationship that exists among a decibel scale,
the power, and intensity scales is difficult to estimate intuitively since the
decibel is a logarithm of a ratio of two power quantities or two intensity
quantities. Further, expression of a decibel quantity must also be accompanied
by a decibel scale reference, for example, the quantity of watts corresponding
to 0 dB. In the United States, the most common decibel scale references 10 -13
W power, whereas the European scale references 10 -12 W.
A large rocket can generate a sound level of about 200 dB (reference 10 -13
W), corresponding to 107 W of sound power, compared to 140 dB for a 75-
piece orchestra generating 10 W. Reducing the sound power by 50% reduces
the value by only about 3 dB. In terms of human sensitivity, a 10-dB change
usually doubles or halves the noise for the average person. Sound levels above
140 dB frequently introduce pain to the ear and levels above 160 become
intolerable (see Ref. 18-12).
Spacecraft Surface Contamination
Contamination of sensitive surfaces of a spacecraft by rocket exhaust products
can be a problem to vehicle designers and users. It can degrade functional
surfaces, such as solar cells, optical lenses, radiators, windows, thermal-control
coatings, and mirrored surfaces. Propellants that have condensed liquids or
solid particles in their exhaust appear to be more troublesome than propellants
with mostly gaseous products, such as oxygen-hydrogen. Plumes from most
solid propellant contain contaminating species. Practically all the investigative
work has been concerned with small storable liquid propellant attitude control
pulse motors in the thrust range 1.0 to 500 N, the type commonly used for
controlling vehicle attitude and orbit positioning over long periods of time.
Deposits of hydrazinium nitrate and other material have been found. The
18.2. PLUME EFFECTS 655
accumulation of exhaust products on surfaces in the vicinity is a function of
many variables, including the propellants, combustion efficiency, combustion
pressure, nozzle expansion ratio, surface temperature, and rocket-vehicle inter-
face geometries. The prediction of exhaust contamination of spacecraft sur-
faces is only partly possible through analytical calculations. Reference 18-13
provides a comprehensive analytical model and computer program for liquid
bipropellant rockets.
Another effect of clouds of condensed species (either tiny liquid droplets or
solid particles) is to scatter sunlight and cause solar radiation to be diverted to
optical instruments on the spacecraft, such as cameras, telescopes, IR trackers,
or star trackers; this effect can cause erroneous instrument measurements. The
scatter depends on the plume location relative to the instruments, the propel-
lant, the density and size of particulates, the sensitive optical frequency, and the
surface temperature of the instrument.
Radio Signal Attenuation
All rocket exhaust plumes generally interfere with the transmission of radio-
frequency signals that must pass through the plume in the process of guiding
the vehicle, radar detecting, or communicating with it. Solid propellant exhaust
plumes usually cause more interference than liquid rocket engine plumes.
Signal attenuation is a function of free electron density and electron collision
frequency. Given these two parameters for the entire plume, the amount of
attenuation a signal experiences in passing through the plume can be calcu-
lated. Figure 18-8 shows the minimum plume model sufficient for predicting
signal attenuation that contains contours of constant electron density and
electron collision frequency for moment um transfer. Free-electron density
Electron density
contours, electrons/cm 3
Exhaust nozzle
FIGURE 18-8. Exhaust plume model for predicting attenuation of radio communica-
tions signals. The contours shown are for either equal electron density or electron
collision frequency; the highest values are near the nozzle exit.
656 ROCKET EXHAUST PLUMES
and activity in the exhaust plume are influenced by many factors, including the
propellant formulation, propellant alkali impurities, exhaust temperature,
mot or size, chamber pressure, flight speed and altitude, and the distance down-
stream from the nozzle exit. Methods have been developed for analyzing (with
computers) the physical and chemical composition, including electron density,
and the attenuation characteristics of exhaust plumes (Refs. 18-14 and 18-15).
The relationship between several influential mot or and vehicle design factors
can be summarized from experience with typical solid propellant rockets as
follows:
1. The presence of alkali metal impurities increases attenuation; changing
the impurity level of potassium from 10 to 100 ppm increases the relative
attenuation some 10-fold at low altitude. Both potassium (~ 150 ppm)
and sodium (~ 50 ppm) are impurities in commercial grade nitrocellulose
and ammoni um perchlorate.
2. The percentage of aluminum fuel is a major influence; increasing the
percentage from 10 to 20% increases the attenuation fivefold at sea
level and three- to fourfold at 7500 m altitude.
3. Increasing the chamber pressure for a given aluminized propellant from
100 to 2000 psi reduces the relative attenuation by about 50%.
4. Attenuation varies with the distance downstream from the nozzle exit
plane and can be four to five times as great as at the nozzle exit plane,
depending on the flight altitude, nozzle geometry, oxidizer-to-fuel ratio,
flight velocity, altitude, and other parameters.
For many solid rocket applications, attenuation of radio or radar signal
strength by the exhaust plume is no problem. When it is, attenuation can
usually be kept at acceptable levels by controlling the level of alkali impurities
in propellant ingredients and by using nonmetal fuels or a low percentage
( < 5%) of aluminum. Motors with high expansion ratio nozzles help, since
electrons combine with the positive ions as the exhaust temperature falls.
The electrons in the plume greatly increase its radar cross section, and hot
plumes can readily be picked up with radar. The plume is usually a stronger
radar reflector than the flight vehicle. A radar homing missile seeker would
focus on the plume and not the vehicle. A reduction of the plume cross section
is desirable (lower gas temperature, less sodium impurities).
Plume Impingement
In most reaction control systems there are many small thrusters and they are
pointed in different directions. There have been cases where the plumes of some
of these thrusters have impinged upon a space vehicle surface, such as extended
solar cell panels, radiation heat rejection surfaces, or aerodynamic control
surfaces. This is more likely to happen at high altitude, where the plume
18.3. ANALYSIS AND MATHEMATICAL SIMULATION 657
diameters are large. This can lead to the overheating of these surfaces and to
unexpected turning moments.
Also, during stage separation, there have been occasions where the plume of
the upper stage impinges on the lower vehicle stage. This can cause overheating
and impact damage not only to the lower stage (being separated), but by
reflection also to the upper stage. Other examples are docking maneuvers or
the launching of multiple rockets (nearly simultaneously) from a military bar-
rage launcher. The plume of one of the rocket missiles impinges on another
flying missile and causes it to experience a change in flight path, often not
hitting the intended target.
18.3. ANALYSI S AND MATHEMATI CAL SI MULATI ON
The complicated structure, the behavior, and many of the physical phenomena
of plumes have been simulated by mathematical algorithms, and a number of
relatively complex computer programs exist (see Refs. 18-16 to 18-20).
Although there has been remarkable progress in using mathematical simula-
tions of plume phenomena, the results of such computer analyses are not
always reliable or useful for making predictions of many of the plume char-
acteristics; however, the models help in understanding the plumes and in ex-
trapolating test results to different conditions within narrow limits. There are
some physical phenomena in plumes that are not yet fully understood.
All simulations are really approximations, to various degrees; they require
simplifying assumptions to make a reasonable mathematical solution possible,
and their field of application is usually limited. They are aimed at predicting
different plume parameters, such as temperature or velocity or pressure pro-
files, radar cross section, heat transfer, radiations, condensation, deposits on
optical surfaces, impact forces, or chemical species. The analyses are usually
limited to separate spatial segments of the plume (e.g., core, mantle, supersonic
versus subsonic regions, continuum versus free molecular flow, near or far
field), and many have different assumptions about the dynamics or steadiness
of the flow (many neglect turbulence effects or the interaction between bound-
ary layers and shock waves). The algorithms are also different in the treatment
of chemical reactions, solids content, energy releases, composition changes
within the plume, different flight and altitude regimes, the interactions with
the airflow and the vehicle, or selected regions of the spectrum (e.g., IR only).
Many require assumptions about particle sizes, their amounts, spatial and size
distribution, gas velocity distribution, the geometry and boundaries of the
mixing layer, or the turbulence behavior. The mathematical models are com-
plex and can use one-, two-, or three-dimensional mesh models. The analysis of
a plume often requires using more than one model to solve for different pre-
dictions. Many solutions are based in part on extrapolating measured data
from actual plumes to guide the analyses. The specific analytical approaches
658 ROCKET EXHAUST PLUMES
are beyond the scope of this book and their mat hemat i cal resolutions are the
domai n of experts in this field.
The act ual measur ement s on pl umes duri ng static and flight tests are used to
verify the theories and they require highly specialized i nst rument at i on, careful
cal i brat i ons and charact eri zat i on, skilled personnel, and an intelligent applica-
tion of vari ous correct i on factors. Ext r apol at i ng the comput er pr ogr ams to
regions or par amet er s t hat have not been val i dat ed has often given poor
results.
PROBLEMS
1. List at least two parameters that are likely to increase total radiation emission from
plumes, and explain how they accomplish this. For example, increasing the thrust
increases the radiating mass of the plume.
Look up the term chemiluminescence in a technical dictionary or chemical encyclo-
pedia; provide a definition and explain how it can affect plume radiation.
If a high-altitude plume is seen from a high-altitude balloon, its apparent radiation
intensity diminishes with the square of the distance between the plume and the
observation platform and as the cosine of the angle of the flight path tangent with
the line to the observation station. Establish your own trajectory and its relative
location to the observation station. For a plume of an ascending launch vehicle,
make a rough estimate of the change in the relative intensity received by the ob-
serving sensor during flight. Neglect atmospheric absorption of plume radiation and
assume that the intensity of emitted radiation stays constant.
REFERENCES
18-1.
18-2.
18-3.
18-4.
18-5.
S. M. Dash, "Analysis of Exhaust Plumes and their Interaction with Missile
Airframes," in M. J. Hemsch and J. N. Nielson (Eds.), Tactical Missile
Aerodynamics, Progress in Astronautics and Aeronautics, Vol. 104, AIAA,
Washington, DC, 1986.
S. M. Yahya, Fundamentals of Compressible Flow, 2nd revised printing, Wiley
Eastern Limited, New Delhi, 1986.
R. D. McGregor, P. D. Lohn, and D. E. Haflinger, "Plume Impingement Study
for Reaction Control Systems of the Orbital Maneuvering Vehicle," AIAA
Paper 90-1708, June 1990.
P. D. Lohn, D. E. Halfinger, R. D. McGregor, and H. W. Behrens, "Modeling
of Near-Continuum Flows using Direct Simulation Monte Carlo Method,"
AIAA Paper 90-1663, June 1990.
F. S. Simmons, Rocket Exhaust Plume Phenomenology, Aerospace Press, The
Aerospace Corporation, 2000.
REFERENCES 659
18-6.
18-7.
18-8.
18-9.
18-10.
18-11.
18-12.
18-13.
18-14.
18-15.
18-16.
18-17.
18-18.
18-19.
18-20.
A. V. Rodionov, Yu A. Plastinin, J. A. Drakes, M. A. Simmons, and R. S. Hiers
III, "Modeling of Multiphase Alumina-Loaded Jet Flow Fields," AIAA Paper
98-3462, July 1998.
R. B. Lyons, J. Wormhoudt, and C. E. Kolb, "Calculation of Visible Radiations
from Missile Plumes," in Spacecraft Radiative Heat Transfer and Temperature
Control, Progress in Astronautics and Aeronautics, Vol. 83, AIAA, Washington,
DC, June 1981, pp. 128-148.
A. C. Victor, "Solid Rocket Plumes," Chapter 8 of: G. E. Jensen and D. W.
Netzer (Eds.), Tactical Missile Propulsion, Progress in Astronautics and
Aeronautics, Vol. 170, AIAA, 1996.
Rocket Motor Plume Technology, AGARD Lecture Series 188, NATO, June
1993.
Terminology and Assessment Methods of Solid Propellant Rocket Exhaust
Signatures, AGARD Advisory Report 287, NATO, February 1993.
D. I. Sebacher, R. J. Bendura, and G. L. Gregory, "Hydrogen Chloride
Measurements in the Space Shuttle Exhaust Cloud," Journal of Spacecraft
and Rockets, Vol. 19, No. 4, July-August 1982.
J. M. Seiner, S. M. Dash, and D. E. Wolf, "Analysis of Turbulent
Underexpanded Jets, Part II: Shock Noise Features Using SCIPVIS," AIAA
Journal, Vol. 23, No. 5, May 1985, pp. 669-677.
R. J. Hoffman, W. D. English, R. G. Oeding, and W. T. Webber, "Plume
Contamination Effects Prediction," Air Force Rocket Propulsion Laboratory,
December 1971.
L. D. Smoot and D. L. Underwood, "Prediction of Microwave Attenuation
Characteristics of Rocket Exhausts," Journal of Spacecraft and Rockets, Vol.
3, No. 3, March 1966, pp. 302-309.
W. A. Wood and J. R. De More, "Microwave Attenuation Characteristics of
Solid Propellant Exhaust Products," AIAA Paper 65-183, February 1965.
I. Boyd, "Modeling of Satellite Control Thruster Plumes," PhD thesis,
Southampton University, England, 1988.
S. M. Dash and D. E. Wolf, "Interactive Phenomena in Supersonic Jet Mixing
Plumes, Part I: Phenomenology and Numerical Modeling Technique," AIAA
Journal, Vol. 22, No. 7, July 1984, pp. 905-913.
S. M. Dash, D. E. Wolf, R. A. Beddini, and H. S. Pergament, "Analysis of Two
Phase Flow Processes in Rocket Exhaust Plumes," Journal of Spacecraft, Vol.
22, No. 3, May-June 1985, pp. 367-380.
C. B. Ludwig, W. Malkmus, G. N. Freemen, M. Slack, and R. Reed, "A
Theoretical Model for Absorbing, Emitting and Scattering Plume
Radiations," in Spacecraft Radiative Transfer and Temperature Control,
Progress in Astronautics and Aeronautics, Vol. 83, AIAA, Washington, DC,
1981, pp. 111-127.
S. M. Dash, "Recent Developments in the Modeling of High Speed Jets, Plumes
and Wakes," AIAA Paper 85-1616, presented at AIAA 18th Fluid Dynamics
Plasma-Dynamics and Laser Conference, July 1985.