Available online at www.sciencedirect.

com
Proceedings
of the
Combustion
Institute
Proceedings of the Combustion Institute 34 (2013) 727–733
www.elsevier.com/locate/proci

Flame propagation of mixtures of air

with high molecular weight neat hydrocarbons
and practical jet and diesel fuels
Bo Li ⇑, Ning Liu, Runhua Zhao, Hai Zhang, Fokion N. Egolfopoulos
Department of Aerospace and Mechanical Engineering, University of Southern California, Los Angeles,
CA 90089-1453, USA

Available online 25 September 2012

Abstract

Laminar flame speeds of mixtures of air with n-C14H30, n-C16H34, a petroleum-derived JP-5 jet fuel, a
camelina-derived hydrotreated renewable JP-5 jet fuel, a petroleum-derived F-76 diesel fuel, and an algae-
derived hydrotreated renewable F-76 diesel fuel, were measured in the counterflow configuration at atmo-
spheric pressure and elevated unburned mixture temperatures. Digital particle image velocimetry was used
to measure the axial flow velocities along the stagnation streamline. The experiments for n-C14H30/air and
n-C16H34/air mixtures were modeled using recently developed kinetic models, and the experimental data
were predicted satisfactorily. Both experiments and simulations revealed that the laminar flame speeds
of n-C14H30/air and n-C16H34/air mixtures are very close to each other, as expected. On the other hand,
the laminar flame speeds for the four practical fuels were found to be lower than n-C14H30 and n-
C16H34, due to the presence of aromatics and branched hydrocarbons. Similarly, the laminar flame speeds
for the alternative fuels were found to be higher than the petroleum-derived ones, again due to the presence
of aromatic compounds in the latter. Further insight into the effects of kinetics and molecular transport
was obtained through sensitivity analysis.
Ó 2012 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

Keywords: n-Tetradecane; n-Hexadecane; Jet fuels; Diesel fuels; Flame propagation

1. Introduction from first principle and understood. The first step

Jet and diesel fuels are used extensively in both
the civilian and military sectors worldwide. These
fuels contain hundreds of chemical components,
which are largely n-alkanes, iso-alkanes, cyclo-
alkanes, and aromatics. Due to the chemical com-
plexity of these compounds, the combustion char-
acteristics of practical fuels cannot be modeled
⇑ Corresponding author. Fax: +1 213 740 8071.
E-mail address: bl_116@usc.edu (B. Li).
towards that goal is to perform fundamental com-
bustion studies for both the practical fuels and the
attendant surrogate compounds and their blends,
so that insight is achieved into the rate-controlling
processes during the conversion of chemical to
thermal energy in engines.
In the past, extensive studies on global flame
properties for neat liquid n-alkanes have been
conducted. For example, laminar flame speeds,
S ou , of n-C7H16/air mixtures has been measured
using different methodologies (e.g., [1–5]). Zhao
et al. [6] studied experimentally and numerically

1540-7489/$ - see front matter Ó 2012 The Combustion Institute. Published by Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.proci.2012.05.063
728 B. Li et al. / Proceedings of the Combustion Institute 34 (2013) 727–733
S ou ’s of n-C10H22/air mixtures. Kumar and Sung
[7] measured S ou ’s and extinction strain rates of
n-C10H22 and n-C12H26 flames. Ji et al. [8] investi-
gated the propagation and extinction of C5–C12 n-
alkane flames, and close agreements were found
between the experimental data and computed
results using a recently developed kinetic model.
Humer et al. [9] studied the ignition and extinction
characteristics of n-C10H22 and n-C12H26 flames.
Liu et al. [10] studied the ignition of non-premixed
C3–C12 n-alkane flames. Although there is notable
amount of work on flames of n-alkanes up to C12,
flame studies on heavier fuels such as, for exam-
ple, in the C14–C16 range, are scarce or non-exist-
ing, and these heavier compounds are of direct
relevance to practical fuels such as JP-5 and diesel.
It should be noted, that autoignition or speciation
studies of heavy fuels could be done more readily
in homogeneous reactors, as ppm levels of fuels
are required typically. In flames on the other
hand, significant amounts of fuel need to be pres-
ent to make the mixture flammable, and thus the
unburned mixture temperature needs to be higher
in order to achieve high enough fuel vapor pres-
sure to exceed its partial pressure and be sustained
thus in the gaseous phase.
Furthermore, compared to neat component
liquid fuels, flame propagation studies of practical
fuels are fewer. Various experimental methodolo-
gies have been used to investigate combustion
characteristics of conventional and alternative
practical fuels, such as flow reactors (e.g., [11]),
shock tubes (e.g., [12]), and counterflow flames
(e.g., [13]).
In terms of flame studies, Holley et al. [13]
studied the ignition and extinction characteristics
of non-premixed flames of various jet fuels and
their surrogates. Kumar et al. [14] investigated
the propagation and extinction of flames of petro-
leum and synthetic jet fuels. Chong and Hochgreb
[15] reported S ou ’s of flames of Jet-A, diesel, palm
methyl esters, and their blends. Ji et al. [16] stud-
ied the propagation and extinction of flames of
conventional and alternative jet fuels. Dooley
et al. [17] developed a method for the formulation
of fuel-specific jet fuel surrogates via an empirical
correlation that was tested successfully against
autoignition data. It is apparent that the existing
flames studies of practical fuels are limited to fuels
with average carbon numbers up to 12, for rea-
sons that are similar to those discussed earlier
for neat hydrocarbons. It is realized also, that
there is need to extend the flame investigations
to higher molecular weight fuels such as JP-5 that
is used in naval aviation and diesel, which could
be petroleum-derived and/or bio-derived.
In view of these considerations, the main goal
of this study was two-fold. First, to extend the
current experimental and computational capabili-
ties so that fuels with carbon numbers greater
than C12 could be investigated in flame environ-
ments with confidence. Second, to use those capa-
bilities and study the flame propagation for
mixtures of air with n-C14H30, n-C16H34, as well
as four practical naval fuels, namely a petro-
leum-derived JP-5 jet fuel (hereafter referred to
as JP-5), a camelina-derived hydrotreated renew-
able JP-5 jet fuel (hereafter referred to as HRJ),
a petroleum-derived F-76 diesel fuel (hereafter
referred to as F-76), and an algae-derived hydro-
treated renewable F-76 diesel fuel (hereafter
referred to as HRD). As it will be discussed later,
the carbon number of the aforementioned fuels
ranges from C10 to C17.
2. Experimental approach
The experiments were carried out in the coun-
terflow configuration (e.g., [3,8,10,13,16,18–20]) at
atmospheric pressure. In all measurements, the
burner nozzle diameters and the burner separa-
tion distance were 14 mm. The gaseous flow rates
were metered using sonic nozzles, which were cal-
ibrated using a wet-test meter.
The key component of the experimental config-
uration is the fuel vaporization and subsequent
handling throughout the heating path, as it has
been reported in previous investigations by the
authors for molecules up to C12 (e.g.,
[3,8,10,13,16]). In order to study heavier com-
pounds, the fuel vaporization and heating path
had to be upgraded notably.
A high-precision syringe pump was used to
control the liquid fuel flow rates. The pump has
a reported accuracy of ±0.5%, resulting in reac-
tant concentration uncertainty of about ±0.5%.
A glass nebulizer was used to produce micron-size
liquid fuel droplets, which were introduced then
into a preheated air flowing in a 90o orientation
with respect to the fuel spray. This cross-flow
injection allows for very efficient mixing and
vaporization at temperatures lower than the fuel
boiling point. Additionally, it was determined
through trial and error, that the cross-flow injec-
tion minimizes the unsteadiness associated with
the vaporization process, which is more profound
for large mass fuel loadings. Unsteady vaporiza-
tion can result in notable flame unsteadiness and
compromise thus the quality of the experimental
data. In order to avoid fuel cracking, the temper-
ature of the fuel/air mixture was maintained
below 490 K throughout the heating path, and
its control was achieved through an array of ther-
mocouples to assure that there are no hot spots.
The burners were heated with ceramic heating
jackets.
A K-type thermocouple was used to monitor
the temperature of the unburned mixture, Tu, at
the center of the burner exit. The majority of the
measurements were carried out for Tu = 443 K,
while for n-C14H30/air flames one additional set
 B. Li et al. / Proceedings of the Combustion Institute 34 (2013) 727–733
of measurements was taken at Tu = 423 K. Flow
velocity measurements were made by seeding the
flow with submicron size silicon oil droplets, and
by using digital particle image velocimetry (e.g.,
[8,10,13,16]).
The minimum point upstream of the flame was
defined as the reference flame speed, Su,ref, corre-
sponding to the imposed strain rate, K, which
was defined as the maximum absolute value of
the axial velocity gradient in the hydrodynamic
zone (e.g., [8,18–20]). Monitoring the Su,ref vs. K
variation, S ou was determined through non-linear
extrapolations to K = 0 using a recently developed
computationally-assisted method [16,21,22].The
velocity measurements were determined to be
accurate to within ±1%. The sampling errors in
Su,ref and K were considered also, and the 2r stan-
dard deviations of S ou were determined and are
reported in the appropriate figures.
Measurements were performed for n-C14H30/
air, n-C16H34/air, JP-5/air, HRJ/air, F-76/air,
HRD/air mixtures. n-C14H30 and n-C16H34were
acquired from TCI America. Water content was
less than 0.01% and most of the impurities
(<1%) were isomers of similar molecular weight.
The U.S. Navy provided the four practical fuels,
and their average carbon numbers range and
chemical-classification compositions are summa-
rized in Table 1 (Personal Communication: R.E.
Morris, N.J. Begue, Chemometrics and Fuel
Technology, Code 6181, Navy Technology Center
for Safety & Survivability). Based on the data
shown in Table 1, it is of interest to note the
higher molecular weights of HRJ and the two die-
sel fuels, and the lack of aromatics in the bio-
derived HRJ and HRD fuels. The studies of n-
C14H30/air and n-C16H34/air mixtures were
included in this investigation as their molecular
weights are similar to those of the practical fuels
considered in this study, and because detailed
models are available for their pyrolysis and oxida-
tion kinetics. To the author’s knowledge, reliable
data for laminar flame speeds for n-C14H30 and
n-C16H34 and the heavy practical fuels considered
in this investigation are not available in the
literature.
Table 1
Fuel specific properties.
Petroleum JP-5a
c
Saturates 75.7%
Olefins 3.0%
Aromatics 11.6%
Heteroatomicd 8.9%
a
Average carbon range C10–C13.
b
Average carbon range C14–C17.
c
n-and iso-alkanes.
d
Sulfur-bound, nitrogen-bound, oxygen-bound.
729
3. Numerical approach
S ou ’s were computed using the PREMIX code
[23]. The counterflow flames were simulated using
the opposed-jet code [24,25] in order to produce
the non-linear extrapolation curves. Both codes
were integrated with the CHEMKIN [26] and
Sandia Transport [27] subroutine libraries. A skel-
etal mechanism based on a detailed C8–C16 n-
alkane high-temperature kinetic model developed
by Westbrook et al. [28] for n-alkanes up to n-
C16H34 was used, consisting of 157 species and
1161 reactions, hereafter referred to as Model I.
The n-C14H30 and n-C16H34 sub-models of Model
I were superimposed on to JetSurF 2.0 [29] that
describes the hydrocarbon kinetics up to C12,
and which has been tested successfully against
flame propagation data for n- and cyclo-alkanes.
This combined model, hereafter referred to as
Model II, consists of 241 species and 1841
reactions.
The non-linear extrapolations curves com-
puted for both n-C14H30 and n-C16H34 flames
were found to be indistinguishable under the same
thermodynamic and boundary conditions, and
were used also for the flames of all practical fuels,
similarly to a recent study on JP-8 type jet fuels
[16].
4. Results and discussion
4.1. Propagation of n-C14H30/air and n-C16H34/air
flames
Figure 1 depicts the experimental and com-
puted S ou ’s of n-C14H30/air flames at Tu = 443 K
for a wide range of equivalence ratios, /. The
experimental peak value, (S ou )peak, is 75 cm/s at
(/)peak = 1.05. The computed S ou ’s using Model I
are slightly higher than the data for / > 1.0 and
notably lower for / < 1.0, with a maximum dis-
crepancy of 6.0 cm/s at / = 0.9. Furthermore,
using Model I (/)peak = 1.1. The predicted S ou ’s
using Model II are in better agreement with the
data, with a maximum discrepancy of 3.8 cm/s
Camelina HRJb Petroleum F-76b Algae HRDb
99.3% 80.0% 95.43%
0.1% 0.8% <0.8%
0.0% 16.6% <0.1%
0.6% 2.5% 3.2%
730 B. Li et al. / Proceedings of the Combustion Institute 34 (2013) 727–733

Fig. 3. Ranked logarithmic sensitivity coefficients of

laminar flame speed with respect to binary diffusion
Fig. 1. Experimental and computed laminar flame
coefficients computed using Model II, for / = 0.7, 1.0
speeds of n-C14H30/air mixtures at Tu = 443 K.
and 1.4 n-C14H30/air flames at Tu = 443 K.

around / = 1.0, and (/)peak = 1.05 in agreement
with the experimental (/)peak.
In order to obtain further insight into the
effects of kinetics and molecular transport, sensi-
tivity analyses were performed using Models I
and II. Figure 2 depicts the logarithmic sensitivity
coefficients (LSC) of S ou on kinetics for n-C14H30/
air flames at / = 0.7, 1.0, and 1.4,respectively.
Results show that the high-temperature oxidation
of n-C14H30 is dominated by H2, CO, and small
hydrocarbon kinetics only. This is reasonable, as
reactions involving heavy hydrocarbons and their
intermediate products are too fast to be rate lim-
iting, in agreement with previous observations
(e.g., [8]).
LSC’s of S ou on binary diffusion coefficients
(BDC) are shown in Fig. 3. The results were
obtained using Model II, and show that S ou is
mostly sensitive to O2–N2 BDC and not to n-
C14H30–N2 BDC. Meanwhile S ou was found to be
somehow sensitive to BDC’s of the major radical
species H, O, and OH to N2. Nearly identical sen-
sitivity results were obtained for n-C16H34 flames
as well, which will not be discussed further.
Additional measurements were carried for n-
C14H30/air mixtures at Tu = 423 K, in order to
assess Tu effects, especially in terms of experimen-
tal consistency for these heavier fuels at those ele-
vated Tu’s. The results are shown in Fig. 4 along
with numerical predictions using Model II. It
can be seen that the experimentally determined
(S ou )peak = 69 cm/s and (/)peak = 1.05. In addition,
the predicted S ou ’s of n-C14H30/air mixtures at
Tu = 423 K using Model I, which are not shown
here, compare to the data very similarly to
what was reported earlier for the Tu = 443 K

Fig. 2. Ranked logarithmic sensitivity coefficients of laminar flame speed with respect to kinetics computed using (a)
Model I and (b) Model II, for / = 0.7, 1.0 and 1.4 n-C14H30/air flames at Tu = 443 K.
 B. Li et al. / Proceedings of the Combustion Institute 34 (2013) 727–733
Fig. 4. Experimental and computed laminar flame
speeds of n-C14H30/air mixtures at Tu = 423 and
443 K. Model II was used in the simulations.
n-C14H30/air flames, with the predicted S ou ’s using
Model I being “shifted” towards slightly higher /
’s compared to S ou ’s using Model II. Furthermore,
the experimental Tu = 423 and 443 K data appear
to be consistent with each other and with the com-
puted results using Model II, which provides con-
fidence in both the experimental and modeling
approaches that were taken. Additionally, it was
determined that the S ou differences are caused by
the density variation rather than kinetics, due to
the small differences between the two Tu’s.
Figure 5 depicts the experimental and com-
puted S ou ’s of n-C16H34/air flames at Tu = 443 K.
Experimentally, (S ou )peak = 74 cm/s and (/)peak =
1.05. Similarly to n-C14H30/air flames, the pre-
dicted S ou ’s using Model I under-predict the data
for / < 1.0 and slightly over-predict the data for
/ > 1.0, with a maximum discrepancy of 2.5 cm/
Fig. 5. Experimental and computed laminar flame
speeds of n-C16H34/air mixtures at Tu = 443 K.
731
s around / = 1.0. On the other hand, the pre-
dicted S ou ’s using Model II are more consistent
with the data and result in (/)peak = 1.05, which
coincides with the experimental value. It was
determined also that both the experimental and
computed S ou ’s for n-C14H30/air and n-C16H34/air
flames are very close to each other for the same
Tu. This is reasonable, as flame propagation is
sensitive to H2, CO, and small hydrocarbon kinet-
ics, and not to fuel-related kinetics, as discussed
earlier.
4.2. Propagation of flames of conventional and
alternative jet and diesel fuels
Figure 6 depicts the experimental S ou ’s of JP-5/
air and HRJ/air flames at Tu = 443 K, which are
compared with the experimental and computed
S ou ’s of n-C16H34/air flames at Tu = 443 K; Model
II was used in the simulations. The exact average
CxHy chemical formulas of JP-5 and HRJ were
not available, and the results are presented thus
in terms of fuel to air mass ratio, (F/A)mass, that
can be computed readily in both the experiments
and simulations. The (S ou )peak values or JP-5/air
and HRJ/air mixtures were determined to be 70
and 74 cm/s respectively at (F/A)mass,peak = 0.07
that is close to(F/A)mass,peak of n-C16H34 flames.
Furthermore, the JP-5 data are notably lower
compared to n-C16H34, except for very fuel-rich
conditions for which S ou ’s are within experimental
uncertainty. It is of interest to note that the exper-
imental S ou ’s of HRJ flames are much closer to S ou ’s
of n-C16H34 flames, and thus higher than S ou ’s of
JP-5 flames.
Similarly to Figs. 6 and 7 depicts the experi-
mental S ou ’s of F-76/air and HRD/air flames at
Tu = 443 K, which are compared with the experi-
mental and computed S ou ’s of n-C16H34/air flames
at Tu = 443 K; Model II was used in the simula-
tions. The (S ou )peak values of F-76/air and HRD/
air mixtures were determined to be 71 and
74 cm/s respectively at (F/A)mass,peak = 0.071 that
is close to (F/A)mass,peak of n-C16H34 flames. Simi-
larly to JP-5 and HRJ, the F-76 flames are notably
slower compared to n-C16H34 flames except for
very rich mixtures, while the HRD flames are
slightly slower than the n-C16H34 flames.
The relative values of flame propagation rates
between the jet and diesel fuels considered in this
investigation, can be explained based on the
chemical classifications of the various components
being present in the fuels. Davis et al. [30] showed
for example, that aromatic fuels exhibit lower S ou ’s
compared to n-alkanes. It has been shown also in
recent studies [31,32], that the initial reactions of
aromatic compounds are sensitive not only to
small hydrocarbon kinetics but also to the kinetics
of the products of initial consumption of fuels.
The intermediate products of alkyl-benzenes are
relative stable resulting thus in lower S ou ’s. Ji and
732 B. Li et al. / Proceedings of the Combustion Institute 34 (2013) 727–733

Fig. 6. Experimental laminar flame speeds of (a) JP-5/air and (b) HRJ/air mixtures at Tu = 443 K. The data are
compared against experimental and computed laminar flame speeds of n-C16H34/air mixtures at Tu = 443 K.

Fig. 7. Experimental laminar flame speeds of (a)F-76/air and (b) HRD/air mixtures at Tu = 443 K. The data are
compared against experimental and computed laminar flame speeds of n-C16H34/air mixtures at Tu = 443 K.

Egolfopoulos [31] measured S ou ’s of flames of 80%
n-C12H26 + 20% toluene a blending ratio that is
representative in practical fuels, and it was shown
that the presence of toluene results in notably
lower S ou values compared to n-C12H26 flames. In
Table 1, it can be seen that there are notable
amounts of aromatics present in the petroleum-
derived JP-5 and F-76 fuels, while the composi-
tions of the bio-derived HRJ and HRD are
dominated by saturated n- and iso-alkanes. On
the other hand, the lower, compared to n-alkanes,
S ou ’s of the bio-derived fuel flames, are caused by
the presence of iso-alkanes in the bio-derived fuels
that are known to inhibit also the overall reactiv-
ity (e.g., [33]).
5. Concluding remarks
Laminar flame speeds of mixtures of air with n-
C14H30, n-C16H34, JP-5, a bio-derived jet fuel, F-
76, and a bio-derived diesel fuel were determined
in the counterflow configuration at atmospheric
pressure and elevated reactant temperatures. Mea-
surements for such heavy fuels became possible
after upgrading the liquid fuel injection and heat-
 B. Li et al. / Proceedings of the Combustion Institute 34 (2013) 727–733
ing path so that vaporization unsteadiness as well
as fuel decomposition and condensation are elim-
inated. The flow velocities were measured using
digital particle image velocimetry. The experi-
ments were modeled using recently developed
kinetic models that describe the pyrolysis and oxi-
dation kinetics of H2, CO, and C1–C16 hydrocar-
bons. Insight into the controlling mechanisms
was obtained through sensitivity analysis on both
the kinetics and molecular transport.
The results showed that the laminar flame
speeds of n-C14H30/air and n-C16H34/air mixtures
are indistinguishable, stemming from the fact that
for both flames fuel-related kinetics are not rate-
controlling and that flame propagation is sensitive
largely to H2, CO, and small hydrocarbon chemis-
try. The simulations reproduced the experimental
data for both n-C14H30/air and n-C16H34/air mix-
tures satisfactorily. Compared to n-C14H30/air
and n-C16H34/air mixtures, the laminar flame
speeds for the petroleum-derived fuels are lower
due to the presence of aromatic compounds. On
the other hand, the compositions of the bio-
derived fuels are dominated by n- and iso-alkanes
and as a result they exhibit laminar speeds that are
higher that the petroleum-derived fuels but lower
compared to n-alkanes.
It is emphasized in closing, that the measured
laminar flame speeds for those higher molecular
weight fuels considered in this study are the first
ones to be reported, and are expected to contrib-
ute towards the fundamental understanding of
the combustion of low vapor-pressure practical
fuels.
Acknowledgements
The work was supported by the United Tech-
nologies Research Center (Grant No. N00014-
09-C-0211) under the technical supervision of
Dr. Meredith Colket, and the Office of Naval Re-
search (Grant No. N00014-11-1-0924) under the
technical supervision of Dr. Sharon Beermann-
Curtin.
References
[1] S.G. Davis, C.K. Law, Combust. Sci. Technol. 140
(1998) 427–499.
[2] J.T. Farrell, R.J. Johnston, I.P. Androulakis, SAE
Paper NO.2004-01-2936, 2004.
[3] A.T. Holley, Y. Dong, M.G. Andac, F.N. Egolfo-
poulos, Combust. Flame 144 (2006) 448–460.
[4] Y. Huang, C.J. Sung, J.A. Eng, Combust. Flame
139 (2004) 239–251.
[5] O.C. Kwon, M.I. Hassan, G.M. Faeth, J. Propul.
Power 16 (2004) 513–522.
[6] Z. Zhao, J. Li, A. Kazakov, F.L. Dryer, S.P.
Zeppieri, Combust. Sci. Technol. 177 (2004) 89–106.
733
[7] K. Kumar, C.J. Sung, Combust. Flame 151 (2007)
209–224.
[8] C. Ji, E. Dames, Y.L. Wang, H. Wang, F.N.
Egolfopoulos, Combust. Flame 157 (2010) 277–287.
[9] S. Humer, A. Frassoldati, S. Granata, et al., Proc.
Combust. Inst. 31 (2007) 393–400.
[10] N. Liu, C. Ji, F.N. Egolfopoulos, Combust. Flame
159 (2012) 465–475.
[11] D.B. Lenhert, D.L. Miller, N.P. Cernansky, Com-
bust. Sci. Technol. 179 (2007) 845–861.
[12] D.F. Davidson, D.R. Haylett, R.K. Hanson, Com-
bust. Flame 155 (2008) 108–117.
[13] A.T. Holley, Y. Dong, M.G. Andac, F.N. Egolfo-
poulos, T. Edwards, Proc. Combust. Inst. 31 (2007)
1205–1213.
[14] K. Kumar, C.J. Sung, X. Hui, Fuel 90 (2011) 1004–
1011.
[15] C.T. Chong, S. Hochgreb, Proc. Combust. Inst. 33
(2011) 979–986.
[16] C. Ji, Y.L. Wang, F.N. Egolfopoulos, J. Propul.
Power 27 (2011) 856–863.
[17] S. Dooley, S.H. Won, M. Chaos, et al., Combust.
Flame 157 (2010) 2333–2339.
[18] C.K. Wu, C.K. Law, Proc. Combust. Inst. 20 (1984)
1941–1949.
[19] C.K. Law, D.L. Zhu, G. Yu, Proc. Combust. Inst.
21 (1986) 1419–1426.
[20] D.L. Zhu, F.N. Egolfopoulos, C.K. Law, Proc.
Combust. Inst. 22 (1988) 1537–1545.
[21] Y.L. Wang, A.T. Holley, C. Ji, F.N. Egolfopoulos,
Proc. Combust. Inst. 32 (2009) 1035–1042.
[22] P.S. Veloo, Y.L. Wang, F.N. Egolfopoulos, C.K.
Westbrook, Combust. Flame 157 (2010) 1989–2004.
[23] R.J. Kee, J.F. Grcar, M.D. Smooke, J.A. Miller,
Report No. SAND85-8240, Sandia National Lab-
oratories, 1985.
[24] F.N. Egolfopoulos, C.S. Campbell, J. Fluid Mech.
318 (1996) 1–29.
[25] R.J. Kee, J.A. Miller, G.H. Evans, G.D. Lewis,
Proc. Combust. Inst. 22 (1988) 1479–1494.
[26] R.J. Kee, F.M. Rupley, J.A. Miller, Report No.
SAND89-8009, Sandia National Laboratories,
1989.
[27] R.J. Kee, J. Warnatz, J.A. Miller, Report No.
SAND83-8209, Sandia National Laboratories,
1983.
[28] C.K. Westbrook, W.J. Pitz, O. Herbinet, H.J.
Curran, E.J. Sillke, Combust. Flame 156 (2009)
181–199.
[29] H. Wang, E. Dames, B. Sirjean, et.al., A high-
temperature chemical kinetic model of n-alkane (up
to n-dodecane), cyclohexane, and methyl-, ethyl-, n-
propyl and n-butyl-cyclohexane oxidation at high
temperatures, JetSurF version 2.0, September 19,
2010 available at (http://melchior.usc.edu/JetSurF/
JetSurF2.0).
[30] S.G. Davis, H. Wang, K. Brezinsky, C.K. Law,
Proc. Combust. Inst. 26 (1996) 1025–1033.
[31] C. Ji, F.N. Egolfopoulos, Proc. Combust. Inst. 33
(2011) 955–961.
[32] C. Ji, E. Dames, H. Wang, F.N. Egolfopoulos,
Combust. Flame 159 (2012) 1070–1081.
[33] C. Ji, S.M. Sarathy, P.S. Veloo, C.K. Westbrook,
F.N. Egolfopoulos, Combust. Flame 159 (2012)
1426–1436.