DOI: 10.

13009/EUCASS2019-79

8TH EUROPEAN CONFERENCE FOR AERONAUTICS AND AEROSPACE SCIENCES (EUCASS)

DOI: ADD DOINUMBER HERE

Combustion Modeling in Solid Rocket Motor Plumes

Tobias Ecker?† , Sebastian Karl? and Klaus Hannemann?
?
German Aerospace Center (DLR)
Institute of Aerodynamics and Flow Technology
Bunsenstr. 10, Göttingen, Germany
tobias.ecker@dlr.de · sebastian.karl@dlr.de · klaus.hannemann@dlr.de
†
Corresponding author

Abstract
The thermo-chemistry of plumes emanating from solid rocket motors (SRM) is strongly inuenced by
hydrogen/oxygen/chlorine chemistry. These after-burning reactions have a signicant impact onto the
infrared signature, as well as on chemical and thermal loads within the propulsion system. This study
provides an overview on available combustion mechanism and evaluate their applicability to CFD. Based
on the results an improved skeletal kinetic model is proposed and validated against detailed mechanisms.
The results of this study oer an evaluation of the performance of current nite rate chemistry models and
the impact on oweld characteristics of SRM plumes.

1. Introduction

Plumes emanating from solid rocket motors (SRM) exhibit ow statistics strongly inuenced by complex hydro-
gen/oxygen/chlorine chemistry. These after burning reactions within the plume have a signicant impact onto the
infrared (IR) irradiance signature, as well as on the chemical erosion of any active or passive mechanical steering sys-
tem exposed to the reactive motor plume. As SRM combustion chambers operate at oxidizer-fuel ratios considerably
less than stoichiometric, afterburning within the plume shear layer occurs. Asides from the intermediate thermochemi-
cal loads of the plume on the launch vehicle, the resulting plume exhaust gases of Ammonium perchlorate (AP) based
SRM may also have an impact on ozone layer depletion1,2 in the atmosphere and the biosphere of the launch site.3
Within this study we provide an overview on the currently available combustion mechanism and evaluate their ap-
plicability to SRM plume modeling. For a preliminary evaluation, the performance of the considered mechanisms
is evaluated using a constant volume reactor. Based on the results an improved 28 reaction skeletal kinetic model is
proposed and validated against detailed mechanisms in constant volume reactor test cases and a counter-owing diu-
sion ame. Subsequently, selected mechanisms are applied in Reynolds-averaged Navier-Stokes CFD calculations of
a small scale AP/HTPB SRM plume test case. From the evaluation of the plume thermochemistry it can be shown that
the proposed model oers improved performance at lower computational costs. The results of this study oer a relevant
evaluation of the performance of current SRM nite rate chemistry models and their impact on oweld characteristics
and are helpful for future scale-resolved simulations of multispecies, reactive solid-rocket motor plumes.

2. Combustion kinetics modeling for solid rocket motor plumes

Modeling combustion is a challenging process, modeling solid rocket combustion even more so due to the wide range
of physical phenomena involved. Challenges include the chemical complexity of the various single and double base
propellants, the inuence of the heterogen composition of the fuel grain, as well as wall heatux on the pyrolysis of
the fuel on the grain surface.4 SRM boosters for large launch vehicles like the Space Shuttle Solid Rocket Boosters
(SRBs), Ariane 5 solid boosters (P241/P238), but also launchers with complete SRM based stages like the Vega C,
some sounding rockets and tactical missiles (e.g. M51) are often based on mixtures of AP/HTPB/aluminum com-
positions. Therefore several kinetic models which include chlorine pathways have been proposed for modeling the
thermochemical properties of SRM plumes.2,5,6 Numerical studies of SRM plume ows using computational uid
dynamics (CFD)2,5–9 and Direct Simulation Monte Carlo (DSMC)10 - for plumes in hight altitudes - have been con-
ducted in the past with a focus on base plate heat ux, plume thermochemistry and infrared (IR) signatures. Many of
these CFD studies have combined reduced nite rate chemistry models of the SRM afterburning with a model for the
particle-gas interactions,2,5,8 which can have an additional heterogeneous eects on the processes in the plume.

Copyright © 2019 by Ecker et al. Published by the EUCASS association with permission.
 DOI: 10.13009/EUCASS2019-79

COMBUSTION MODELING IN SOLID ROCKET MOTOR PLUMES

Modeling the reaction kinetics in solid rocket motor plumes is a dicult task due to the vast number of species
and reaction pathways involved. A kinetic model for this application would include subsets of hydrogen/oxygen/chlorine/carbon
oxides and possible nitrogen/nitrogen oxide reactions. There are several nite rate chemistry mechanisms2,5,6,11,12
which may be suitable for the modeling or haven been used in the past for the modeling of SRM plumes. An overview
of the kinetic models in the order of their complexity is given in table 1.

Table 1: Available reaction mechanisms

Mechanism name No. of reactions Reference

Troyes (2006) 17 [5]
currenta 28
Niu (2017) 30 [9]
Poubeau (2016) 37 [6]
Denison (1994) 46 [2]
Roessler (1992) 56 [11]
Pelucchi (2015) 102 [12]
a will be discussed further in section 3

Of these six previously published kinetic models, the models by Denison et al.,2 Troyes et al.,5 Poubeau et al.6
and Niu et al.9 have been used in past works for modeling the thermo-chemistry of SRM plume ow elds. To the
author’s knowledge there are currently no published studies in the open literature available which can validate these
models against each other or more importantly against experimental results of SRM plumes.
Denison et al.2 presented a study on the interaction of chlorine based SRM exhaust with the atmosphere and the
associated ozone depletion. In their study the authors present several reduced kinetic models for the SRM afterburning
chemistry and dispersion in the hot plume and the chemistry and the successive diusion in the cool plume. The
presented hot plume mechanism is based on 46 reactions. For a hypothetical solid rocket motor based on AP, binder
and aluminum loading the authors calculated the resulting hot plume with nite rate chemistry and then evaluated the
ozone depletion. Their results indicated local ozone depletion to be only present over a very short periods of time.
Troyes et al.5 presented a skeletal mechanism choosen as a computational ecient compromise based on avail-
able models. In their study the authors focus on the numerical simulation of the reactive two-phase ow from the
end of the motor combustion chamber to the exhaust plume using both nite rate plume chemistry, as well as models
for the aluminium oxide content. In their study Troyes et al. demonstrate the application in a RANS calculation of
a SRM plume ow and show the signicant afterburning eect due to the chemical reactions between exhaust and
air in the jet’s shear layer. This kinetic model has been used for several studies8 on SRM plumes by ONERA since.
Poubeau et al.6 recently published a computational study on performing large eddy simulation (LES) of a SRM plume
in conjunction with an oine chemistry model in order to evaluate the impact of SRM plumes on ozone in the earth’s
atmosphere. The kinetic model used for the oine chemistry is based on a reduced GRI 3.013 with additional chlorine
pathways. The chemistry is solved based on a constant pressure reactor modelled in CANTERA along streamlines of
a time-averaged oweld of a single inert species in order to determine temperature and chemical species distribution.
It should be noted due the nature of the frozen chemistry it is likely that eects of species composition and associated
temperature change on the plume ow eld itself are not fully captured with this approach. Another kinetic model
consisting of 30 reactions has been presented by Niu et al..9 The model is based on a hdyrogen/oxygen combustion
mechanism by Jensen and Jones14 and extended with pathways for nitrogen and chlorine reactions. In contrast to all
other mechanisms presented here nitrogen reaction pathways are considered by the Denison et al.2 and Niu et al.9
kinetic model. This is mostly relevant at elevated temperature within the jet core region. The study by Niu et al. details
an approach to modeling the afterburning in a SRM plume and successive determination of spectral intensities in the
IR bands. The authors found afterburning eects to signicantly increase the size and shape of radiance images and
good agreement with IR experimental data.
Asides from these kinetic models which have been used for numerical studies on SRM plumes, two other much
more detailed models containing the described pathways are available. Roessler et al.11 presented a study on moist CO
oxidation inhibited by trace quantities of HCl. The 56 reactions step mechanism is validated by data from an atmo-
spheric pressure ow reactor. The study is most applicable to SRM plume ows as the combustion chamber products
from a typical SRM based on AP/HTPB consist of large quantities of HCl, CO and H2 O.15 The author’s kinetic model
seems to be a main inspiration for the skeletal mechanisms previously described. A more detailed model consisting of
102 reactions was recently presented by Pelucchi et al..12 The authors give a very detailed account on how they have
updated on the high temperature chlorine chemistry for an improved H2 /Cl2 /HCl/CO/O2 kinetic model. The presented
model is validated against selected experimental data and is the currently most detailed mechanism to describe the

2
 DOI: 10.13009/EUCASS2019-79

COMBUSTION MODELING IN SOLID ROCKET MOTOR PLUMES

chemistry of chlorine species/fuel interactions – a process directly applicable to SRM plume environments.

2.1 Reactor studies

In order to verify and validate the dierent reaction mechanisms with their vastly diering amounts of kinetic pathways,
several numerical studies on a constant volume reactor were conducted. For the processes in a reacting ow, the rate
change of species can be modeled using a nite rate chemistry approach. The rate at which each species changes during
a chemical process is determined from the chemical reactions as:16
 
∑  f ∏ ∏ 
r  βrs 
ωs = Ms r 
β s − α s kr αrs b
(n s ) − kr (n s ) 
r s s

where αrs and βrs are the stoichiometric coecients for the educt and product for reaction r respectively, krf and
krbthe forward and backward reaction rate for reaction r respectively, n s is the molar density and M s is the molar mass
of species s. krf and is the forward reaction rate according to the Arrhenius law
( E )
a
k = A T n exp −
RT
where A is the pre-exponential factor, n is the temperature exponent and Ea is the activation energy. The rate
constants kbf for the reverse reactions are evaluated from the equilibrium constants.
The constant pressure reactor problem is solved by using the CANTERA17 open-source suite of tools. An
overview of the 3 dierent cases, which include a moist CO/HCl case - similar to actual SRM plume conditions, a
H2 /Cl2 case and a classic stoichiometric hydrogen/oxygen combustion case is given in table 2.

Table 2: Overview reactor test cases.

Case name type T start [K] p [atm] X start

CO: 0.0086, O2 : 0.0053, H2 O:
Roessler [11] nozzle ow in N2 1000.0 1.0
0.0057, HCl: 0.00019
Lifshitz [18] shock tube - 1.0 Ar: 0.792, H2 : 0.104, Cl2 : 0.104
Slack [19] shock tube - 1.0 stoichiometric H2 and air

2.2 Reactive moist CO/HCl mixture

The time histories of the mole fractions of selected educts and products within the reactor case are shown in gure 1. For
the two most detailed mechanisms (see gure 1a), the mechanism by Roessler et al.11 and the mechanism by Pelucchi
et al.,12 results are in good agreement with the experimental data by Roessler et al..11 Most notable is the oxidation
of CO to CO2 , which is expected to play a role in most SRM plumes due to the high amount of carbon monoxide in
the SRM combustion chamber. From the results a strong inuence of species thermodynamic database used for the
computations is noticable. Three dierent publicly available databases were tested, the original NASA CEA database
based on NASA 9 polynomials (N9), the thermodynamics database available from the GRI 3.0 mechanism (N7 - GRI)
and the thermodynamics database used by Pelucchi et al. (N7 - PEL)12 which generally gives similar results to the
NASA 9 CEA database but is presented in NASA 7 polynomials. For this test case, the original numerical data by
Roessler et al.11 is best reproduced by using the GRI 3.013 NASA polynomials. For this comparison the temporal
axis of each computational dataset is anchored with an oset on the point where 50% of the nal CO2 concentration
is reached, therefore ignoring dierences in ignition delay times. The results using the simplied kinetic models by
Troyes et al.5 and Denison et al.2 are displayed in gure 1b. The changing molar fractions exhibit a general behavior
similar to the more detailed model. However strong dierences arise on the reactor mixture fractions during the reaction
process and the reaction time scales. In terms of species composition the strongest dierences are visible in the HCl
mole fraction development which exhibits a much stronger drop in comparison to the more detailed models. The
kinetic mechanism by Poubeau et al.6 was also tested but displayed 2-3 longer reaction timescales and was therefore
not included in the gure. In order to exclude an transcription error, the original GRI 3.013 was reduced to only include
the pathways reported by Poubeau. This approach allowed exact reproduction of the behaviour of the Poubeau model.
Similar issues were present with the kinetic model by Niu et al.,9 which is also not included for clarity.
The development of the reactor temperature as predicted by all investigated models is shown in gure 2. As is to
be expected the models by Roessler et al.11 and Pelucchi et al.12 give the closest agreement to the experimental data.

3
DOI: 10.13009/EUCASS2019-79
DOI: 10.13009/EUCASS2019-79
DOI: 10.13009/EUCASS2019-79
DOI: 10.13009/EUCASS2019-79
 DOI: 10.13009/EUCASS2019-79

COMBUSTION MODELING IN SOLID ROCKET MOTOR PLUMES

Table 3: Current improved skeletal kinetic model. Units in cm, mol, cal, s.

Rate constants
No. Reaction A n Ea Reference
1 H2 + O2  HO2 + H 1.0000e+14 0.00 5.6000e+04 [22]
2 H + O2  OH + O 2.6000e+14 0.00 1.6800e+04 [22]
3 O + H2  OH + H 1.8000e+10 1.00 8.9000e+03 [22]
4 OH + H2  H2 O + H 2.2000e+13 0.00 5.1500e+03 [22]
5 2 HO  H2 O + O 6.3000e+12 0.00 1.0900e+03 [22]
6a H + OH + M  H2 O + M 2.2000e+22 -2.00 0.0000e+00 [22]
7b 2 H + M  H2 + M 6.4000e+17 -1.00 0.0000e+00 [22]
8c H + O + M  OH + M 6.0000e+16 -0.60 0.0000e+00 [22]
9d H + O2 + M  HO2 + M 2.1000e+15 0.00 -1.0000e+03 [22]
10 HO2 + H  2 HO 1.4000e+14 0.00 1.0800e+03 [22]
11 HO2 + H  H2 O + O 1.0000e+13 0.00 1.0800e+03 [22]
12 HO2 + O  O2 + OH 1.5000e+13 0.00 9.5000e+02 [22]
13 HO2 + OH  H2 O + O2 8.0000e+12 0.00 0.0000e+00 [22]
14 2 HO2  H2 O2 + O2 2.0000e+12 0.00 0.0000e+00 [22]
15 H + H2 O2  H2 + HO2 1.4000e+12 0.00 3.6000e+03 [22]
16 O + H2 O2  OH + HO2 1.4000e+13 0.00 6.4000e+03 [22]
17 OH + H2 O2  H2 O + HO2 6.1000e+12 0.00 1.4300e+03 [22]
18e H2 O2 + M  2 HO + M 1.2000e+17 0.00 4.5500e+04 [22]
19 2 O + M  O2 + M 6.0000e+13 0.00 -1.8000e+03 [22]
20 CO + OH  CO2 + H 1.5000e+07 1.30 -7.6500e+02 [5]
21 CO + O2  CO2 + O 2.5300e+12 0.00 4.7700e+04 [5]
22f CO + O + M  CO2 + M 2.5100e+13 0.00 -4.5400e+03 [5]
23 HCl + H  H2 + Cl 2.3000e+13 0.00 3.5000e+03 [5]
24 Cl2 + H  HCl + Cl 8.5900e+13 0.00 1.1700e+03 [5]
25 HCl + OH  H2 O + Cl 2.4500e+12 0.00 1.1000e+03 [5]
26 HCl + O  OH + Cl 5.2400e+12 0.00 6.4000e+03 [5]
27 Cl + Cl + M  Cl2 + M 2.3400e+14 0.00 -1.8000e+03 [5]
28 H + Cl + M  HCl + M 7.2000e+21 -2.00 0.0000e+00 [5]
a H O = 6.0
2
b H = 2.0, H O = 6.0
2 2
c H O = 5.0
2
d H = 2.0, H O = 16.0
2 2
e H O = 15.0
2
f CO = 3.8, CO = 1.9, H = 2.5, H2 O = 12.0
2 2

3.2 Counter owing diusion ame

A counter owing diusion ame, also called laminar diusion is generally referred to as a pure diusion ame. Usu-
ally it describes a setup where two opposing stream of fuel and oxidant create a stagnation point at which combustion
occurs (for setup and coordinate denition see gure 10). At the stagnation plane between the two streams where
combustion occurs, the reaction rate is generally much higher than the molecular diusion rate which is therefore rate
governing.

Table 4: Case parameters

Variable Value species mass fraction fuel mass fraction oxidizer

pressure 0.9 atm H2 O 0.2375 -
mass ux fuel 0.3 kg/s/m2 HCl 0.2600 -
mass ux oxidizer 0.7 kg/s/m2 CO2 0.2579 -
gas temperature fuel 2400.0 K CO2 0.1427 -
gas temperature oxidizer 260.0 K N2 0.1020 0.7646
distance between inlets 0.02 m O2 - 0.2354

8
DOI: 10.13009/EUCASS2019-79
DOI: 10.13009/EUCASS2019-79
DOI: 10.13009/EUCASS2019-79
 DOI: 10.13009/EUCASS2019-79

COMBUSTION MODELING IN SOLID ROCKET MOTOR PLUMES

Table 5: Case parameters

Variable Value Variable Value

chamber pressure 48.25 bar co-ow Mach number 0.8
mass ux (fuel) 13.8295 kg/s/m2 co-ow density 1.225 kg/m3
gas temperature 2674.28 K co-ow pressure 1.0 bar
nozzle wall temperature 1000.0 K outside wall temperature 1000.0 K

Table 6: Input species mass fraction.

species mass fraction species mass fraction

HCl 2.5555e-01 CO2 1.4279e-01
CO 2.5307e-01 N2 9.9524e-02
H2 O 2.3307e-01 H2 1.0908e-02

The external and internal motor geometries are based on dimensions given by Sabnis et al..23 The geometry was
extended to include the external domain for the plume ow studies and is shown in gure 13. For the present case,
the motor and freestream boundary conditions are given in table 5. The base ow conditions were estimated based on
equilibrium conditions for the propellant formulation (AP/HTPB at 83.53%/16.47%) without aluminum contribution
using the NASA CEA code (see table 6 for species distribution). The fuel wall temperature is set to 2674.28 K, which
represents the ame temperature of the AP/HTPB mixture at 48 kPa chamber pressure. All other internal and external
walls are set to 1000.0 K surface temperature. The external M = 0.8 co-ow is modeled as a freestream boundary
condition with air properties at sea level conditions.
Several of the kinetic models previously presented are used to calculate nite rate chemistry for the SRM plume
test case. The Troyes model is used as it is the computationally most aordable, representing the low side of complexity,
the Roessler model is used instead of the Pelucchi model to obtain results using detailed chemistry. Both models
are compared to results from the proposed improved skeletal model. For comparison a non-reactive multi-species
calculation is also performed.
The oweld temperature for all cases are shown in gure 14. The temperature contours show a ow structure
representative for a classical under-expanded supersonic jet with nozzle exit pressures much higher than the environ-
mental static pressure. The ow expands and turns outwards and is then compressed through an oblique shock wave at
x/D of about 6. A series of expansion and shock waves follows until the jet potential core breaks down at around x/D
> 40. The comparison with the non-reactive case shows the large role post-combustion or afterburning plays for SRM
plumes. For this ctive non aluminized AP/HTPB temperature of up 2200 K in the shear layer and up to 2400 K in the
core breakdown region are reached. While all models show a generally similar prediction of the plume temperature,
some dierences exist. The more detailed model by Roessler shows a distinctively higher shear layer temperature by
about 75 – 100 K with peak spots located above the rst shock and in the break down region. The proposed skeletal
model gives a prediction of these hot ow features closer to the detailed model.
The temperature and species concentration on the centerline are displayed in gure 15. For both the species
distribution and the temperature the general trends between the dierent models do not dier dramatically and are in
line with previously reported CFD studies on SRM plumes.2,5,8 However, as there is only very little oxygen in this core
region large dierences in this zone were not expected. At x/D > 30 when more and more oxygen mixes into the core
region, combustion within this region is started. It is within this region where dierences between the models are most
obvious. Again, the proposed model predicts the kinetics very closely to the more detailed Roessler model.
The distribution of the Hydroxyl radical (OH) and HCl, both which are representative of partially intermittent
products of the SRM chemistry, are shown for the reactive ow in gures 16 and 17 respectively. The OH concentration
peaks in the growing shear layer starting from slightly above the nozzle lip line (r/D > 1.5) until jet potential core
breakdown. The simpler Troyes model over predicts OH production both in the shear layer and especially close
to the core breakdown region (x/D ∼ 34–38) when compared to the more detailed Roessler model. The proposed
kinetic model is very good in capturing most of the features present in the detailed model and also predicts lower OH
concentration in the core breakdown region.
From the reactor studies it can be shown that the HCl mass fraction distribution is another indicator to evaluate
these SRM plume kinetics. In dierence to OH, which is only available in trace amounts within the SRM combustion
chamber, HCl is one of the main products. Eventually, HCl is transferred trough several reactions to Cl and Cl2 which
can play a role in atmospheric ozone depletion.2 A rst peak of HCl is near the nozzle exit slightly above the lipline
(r/D > 1.5). This peak is predicted by all three mechanism. However, the Roessler model shows a second region of
high concentrations starting at x/D of 10, spanning up to x/D of 24. These regions are of relatively low temperature

12
DOI: 10.13009/EUCASS2019-79
 DOI: 10.13009/EUCASS2019-79

COMBUSTION MODELING IN SOLID ROCKET MOTOR PLUMES

(a) Troyes kinetic model.

(b) Current improved skeletal kinetic model.

(c) Roessler kinetic model.

Figure 16: Plume OH mass fractions.

(a) Troyes kinetic model.

(b) Current improved skeletal kinetic model.

(c) Roessler kinetic model.

Figure 17: Plume OH mass fractions.

values up to 4% higher compared to the Troyes model. The dierence between the models in temperature and CO2
concentration on the centerline for this region is shown in gure 19. As CO2 has a much higher IR absorption than CO
this dierence could inuence the prediction of IR signatures of SRM plumes. Due to higher temperature, this eect
on IR emissions would be enhanced. The proposed model reduces the dierence in temperature and CO2 prediction to
around 1.5% of the absolute value.

14
DOI: 10.13009/EUCASS2019-79
 DOI: 10.13009/EUCASS2019-79

COMBUSTION MODELING IN SOLID ROCKET MOTOR PLUMES

Upstream of the core breakdown region, as more and more oxygen mixes into the core region, combustion
within this region is started. It is within this region where dierences between the models are most obvious. The OH
concentration, which is an intermittent product in SRM combustion, peaks in the growing shear layer starting from
slightly above the nozzle lip line until jet potential core breakdown. The simpler models over-predict OH production
both in the shear layer and especially close to the core breakdown region when compared to the more detailed model.
The proposed kinetic model is very good in capturing most of the features present in the detailed model and also
predicts lower OH concentration in the core breakdown region. A rst peak of HCl production appears near the nozzle
exit at the binning of the shear layer. This peak is predicted by all investigated mechanism. However, the more
detailed model shows a second region of high concentration starting behind the rst expansion wave in the jet core,
which are regions of relatively low temperature. Comparison between kinetic models indicates an important interaction
between the chlorine reaction subset and the hydrogen/oxygen reaction subset. Lastly the carbon-dioxide shows peak
concentration close to the core breakdown region, which is also where the highest plume temperatures are reached.
The dierence between the models in temperature and CO2 concentration in this region are as much as 4.5 % which
could impact IR emissions.
The proposed skeletal model gives a prediction of these hot ow features much closer to the detailed model at
low computational costs and could be used for future scale-resolved simulations of multispecies, reactive solid-rocket
motor plumes in order to predict plume thermochemistry, IR signature and environmental impact at launch sites.

References

[1] M. N. Ross, J. O. Ballenthin, R. B. Gosselin, R. F. Meads, P. F. Zittel, J. R. Benbrook, and W. R. Sheldon. “In-
situ measurement of Cl2 and O3 in a stratospheric solid rocket motor exhaust plume”. In: Geophysical Research
Letters 24.14 (July 1997), pp. 1755–1758. doi: 10.1029/97GL01592.
[2] M. R. Denison, John J. Lamb, William D. Bjorndahl, Eric Y. Wong, and Peter D. Lohn. “Solid rocket exhaust
in the stratosphere - Plume diusion and chemical reactions”. In: Journal of Spacecraft and Rockets 31.3 (May
1994), pp. 435–442. doi: 10.2514/3.26457.
[3] Bernard Jerey Anderson and Vernon W Keller. Space shuttle exhaust cloud properties. NASA Technical Paper
NAS 1.60:2258. NASA, 1983.
[4] Thomas L. Jackson. “Modeling of Heterogeneous Propellant Combustion: A Survey”. In: AIAA Journal 50.5
(May 2012), pp. 993–1006. doi: 10.2514/1.J051585.
[5] Julien Troyes, Isabelle Dubois, Vincent Borie, and Annie Boischot. “Multi-Phase Reactive Numerical Simula-
tions of a Model Solid Rocket Exhaust Jet”. In: 42nd AIAA/ASME/SAE/ASEE Joint Propulsion Conference &
Exhibit. American Institute of Aeronautics and Astronautics, July 2006. doi: 10.2514/6.2006-4414.
[6] Adèle Poubeau, Roberto Paoli, and Daniel Cariolle. “Evaluation of Afterburning Chemistry in Solid-Rocket
Motor Jets Using an O-Line Model”. In: Journal of Spacecraft and Rockets 53.2 (Feb. 2016), pp. 380–388. doi:
10.2514/1.A33311.
[7] June Woo Lee, Jae Gwan Kim, and Kyu Hong Kim. “Eects of Al2O3 Particle on Convective and Radiative
Heat Flux to Rocket Base Surface”. In: Seventh International Conference on Computational Fluid Dynamics
(ICCFD7), Big Island, Hawaii. July 2012.
[8] V Rialland, A Guy, D Gueyer, P Perez, A Roblin, and T Smithson. “Infrared signature modelling of a rocket
jet plume - comparison with ight measurements”. In: Journal of Physics: Conference Series 676.1 (2016),
p. 012020.
[9] Qinglin NIU, Zhihong HE, and Shikui DONG. “IR radiation characteristics of rocket exhaust plumes under
varying motor operating conditions”. In: Chinese Journal of Aeronautics 30.3 (2017), pp. 1101–1114. doi: 10.
1016/j.cja.2017.04.003.
[10] Jonathan M. Burt and Iain D. Boyd. “High Altitude Plume Simulations for a Solid Propellant Rocket”. In: AIAA
Journal 45.12 (2007), pp. 2872–2884. doi: 10.2514/1.30129.
[11] J. F. Roesler, R. A. Yetter, and F. L Dryer. “Detailed Kinetic Modeling of Moist CO Oxidation Inhibited by
Trace Quantities of HCl”. In: Combustion Science and Technology 85.1-6 (Sept. 1992), pp. 1–22. doi: 10 .
1080/00102209208947157.
[12] Matteo Pelucchi, Alessio Frassoldati, Tiziano Faravelli, Branko Ruscic, and Peter Glarborg. “High-temperature
chemistry of HCl and Cl2”. In: Combustion and Flame 162.6 (2015), pp. 2693–2704. doi: 10 . 1016 / j .
combustflame.2015.04.002.

16
 DOI: 10.13009/EUCASS2019-79

COMBUSTION MODELING IN SOLID ROCKET MOTOR PLUMES

[13] GP Smith, DM Golden, M Frenklach, NW Moriarty, B Eiteneer, M Goldenberg, CT Bowman, RK Hanson, S

Song, WC Gardiner, et al. GRI-Mech version 3.0. 1999.
[14] D.E. Jensen and G.A. Jones. “Reaction rate coecients for ame calculations”. In: Combustion and Flame 32
(1978), pp. 1–34. doi: 10.1016/0010-2180(78)90078-0.
[15] Tobias Ecker, Sebastian Karl, and Klaus Hannemann. “Modeling of Aluminum Particle Combustion in Solid
Rocket Combustion Chambers”. In: 53rd AIAA/SAE/ASEE Joint Propulsion Conference, AIAA Propulsion and
Energy Forum, (AIAA 2017-4781). July 2017. doi: 10.2514/6.2017-4781.
[16] Volker Hannemann. Numerical simulation of shock-shock-interactions considering chemical and thermal nonequi-
librium. DLR-Forschungsbericht 97-07, 1997.
[17] David G. Goodwin, Harry K. Moat, and Raymond L. Speth. Cantera: An Object-oriented Software Toolkit for
Chemical Kinetics, Thermodynamics, and Transport Processes. Version 2.3.0. 2017. doi: 10.5281/zenodo.
170284.
[18] Assa Lifshitz and Pinchas Schechner. “The mechanism of the H2 + CL2 reaction: Ignition behind reected
shocks”. In: International Journal of Chemical Kinetics 7.1 (Jan. 1975), pp. 125–142. doi: 10 . 1002 / kin .
550070113.
[19] M. Slack and A. Grillo. “Investigation of hydrogen-air ignition sensitized by nitric oxide and by nitrogen dioxide.
Final report”. In: N-78-10233; NASA-CR-2896 (Oct. 1977).
[20] Victor P. Zhukov. “Validation, and Testing of Kinetic Mechanisms of Hydrogen Combustion in Fluid-Dynamic
Computations”. In: ISRN Mechanical Engineering 2012 (Feb. 2012). doi: 10.5402/2012/475607.
[21] Casimir J. Jachimowski. An analytical study of the hydrogen-air reaction mechanism with application to scram-
jet combustion [microform] / Casimir J. Jachimowski. National Aeronautics, Space Administration, Scientic,
and Technical Information Division, 1988.
[22] P Gerlinger, H. Moebius, and D. Brueggmann. “An Implicit Multigrid Method for Turbulent Combustion”. In:
Journal of Computational Physics 167.2 (2001), pp. 247–276. doi: 10.1006/jcph.2000.6671.
[23] Jayant S. Sabnis. “Numerical Simulation of Distributed Combustion in Solid Rocket Motors with Metalized
Propellant”. In: Journal of Propulsion and Power 19.1 (2003), pp. 48–55. doi: 10.2514/2.6101.
[24] S. Langer, A. Schwöppe, and N. Kroll. “The DLR Flow Solver TAU – Status and Recent Algorithmic Develop-
ments”. In: AIAA Paper AIAA-2014-0080. 52nd Aerospace Sciences Meeting (2014). doi: 10.2514/6.2014-
0080.
[25] F. R Menter. “Two-Equation Eddy-Viscosity Turbulence Models for Engineering Applications”. In: AIAA Jour-
nal 32.8 (1994), pp. 1598–1605. doi: 10.2514/3.12149.
[26] Weichen Wang, Shipeng Li, Qiao Zhang, and Ningfei Wang. “Infrared radiation signature of exhaust plume
from solid propellants with dierent energy characteristics”. In: Chinese Journal of Aeronautics 26.3 (2013),
pp. 594–600. doi: 10.1016/j.cja.2013.04.019.

17