Fuel 207 (2017) 389–401

Contents lists available at ScienceDirect

Fuel
journal homepage: www.elsevier.com/locate/fuel

Full Length Article

An ignition delay time and chemical kinetic study of ethane sensitized by

nitrogen dioxide
Fuquan Deng a, Youshun Pan a, Wuchuan Sun a, Feiyu Yang a, Yingjia Zhang a,⇑, Zuohua Huang a
a
State Key Laboratory of Multiphase Flows in Power Engineering, Xi’an Jiaotong University, Xi’an 710049, People’s Republic of China

a r t i c l e i n f o a b s t r a c t

Article history: Nitrogen dioxide (NO2) is a dominant component of NOx pollution in combustion of internal combustion
Received 20 March 2017 engines and gas turbines. Its sensitization on ignition of ethane which is a main component of natural gas
Received in revised form 12 June 2017 has been investigated in this experimental and kinetic study. Ignition delay times of NO2/C2H6/O2/Ar mix-
Accepted 21 June 2017
tures, with blending ratios of NO2:C2H6 of 0.3:1 and 1:1, were measured in a shock tube. Experimental
Available online 29 June 2017
conditions cover a range of pressures (1.2–20 atm), temperatures (950–1700 K) and equivalence ratios
(0.5–2.0). Similarly to our previous work of CH4/NO2 (Deng et al., 2016) [14], NO2 addition promotes
Keywords:
the reactivity of ethane and reduces the global activation energy particularly at higher pressures
Ethane
Nitrogen dioxide
(p > 5.0 atm) and lower temperatures (T < 1175 K), whereas it only presents a limited effect at low pres-
Ignition delay times sure (1.0 atm) and higher temperatures (T > 1175 K). Furthermore, an opposite effect of NO2 addition is
Chemical sensitization observed in both the experiments and the simulations at different temperature regimes. Compared to
fuel-rich mixture, NO2 addition exhibits more significantly promoting effect on the ignition of fuel-
lean mixture under given NO2 concentration. Four literature kinetic mechanisms and an updated mech-
anism proposed in this study have been compared to simulate the new ignition delay time data and the
literature data. Overall, the proposed model is capable of reproducing the experimental results measured
by various facilities over a wide range of conditions. The proposed model is thus used to carry out the
sensitivity and flux analyses to clarify the chemistry interaction between NO2 and ethane. Based on
the kinetic analyses, the impact of NO2 has been expounded at different conditions.
Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction small amounts of nitrogen oxides can cause a dramatic impact

Ethane combustion is of interest of gas-engine applications
blended with natural gas, coal-bed methane and shale gas. There-
fore, proper-understanding of ethane chemistry is of great impor-
tance to hierarchically develop a detail kinetic mechanism due to
its significance as a critical role of intermediate products in the
combustion of hydrocarbon fuels [1,2]. With this in mind,
researchers conducted a great amount of fundamental combustion
studies of ignition and oxidation of ethane, including ignition delay
times [3–6], flame speed[7,8], species concentrations [9,10] and
chemical kinetic mechanisms [11–13].
Nitrogen oxides (NOx) is well known as main gas emissions in
engine combustion, and it can be recirculated into the fresh cycle
particularly at internal combustion engines with exhaust gas recir-
culation. The studies reported in our current work [14,15] and lit-
erature [14,16,21] have demonstrably indicated that even a very
⇑ Corresponding author.
E-mail address: yjzhang_xjtu@xjtu.edu.cn (Y. Zhang).
on combustion characteristics of hydrocarbons due to the fact that
the presence of NOx promotes the establishment of radical pool
and thus accelerates the oxidation of fuels. Unfortunately, there
is only limited work with respect to the chemical interaction of
ethane and NOx. Faravelli et al. [22] studied the combustion char-
acteristics of ethane/NO mixtures at 1.0 atm over a temperature
range of 600–1100 K in a plug-flow reactor (PFR) reactor. They pro-
posed a low-temperature NOx kinetic models and illustrated the
impact of NO on the ethane oxidation. Sivaramakrishnan et al.
[21] investigated the mutual sensitization of NO and natural gas
(CH4/C2H6 = 10:1 in mole fraction) at pressures of 10–50 atm and
equivalence ratios of 0.3–1.5 over a temperature range of 1000–
1500 K in a high-pressure shock tube and a jet-stirred reactor
(JSR). They found that the addition of NO significantly promoted
the oxidation of natural gas at the conditions investigated in their
study and pointed out that the promoting-effect of NO was caused
by the formation of abundant ȮH radicals via the reaction
HȮ2 + NO <=> ȮH + NO2. Herzler and Naumann [17] studied the
promoting effect of NO2 (20–250 ppm) on the ignition of
CH4/C2H6/O2/Ar mixtures at 16 bar and equivalence ratios of

http://dx.doi.org/10.1016/j.fuel.2017.06.099
0016-2361/Ó 2017 Elsevier Ltd. All rights reserved.
390 F. Deng et al. / Fuel 207 (2017) 389–401

0.25–1.0 over a temperature range of 1000–1700 K in a shock tube. 10 0.20

They found that small amount addition of NO2 can dramatically End-wall pressure trace
*
accelerate the ethane ignition, and the promoting-effect was clo- End-wall OH emission
8
sely related to the change in equivalence ratios. Clearly, their 0.15
observation is consensus with the previous work [14,16–21] men-
tioned above. Gersen et al. [19] performed the study of NO2 sensi- 6 dp/dt = 4%
tization on two mixtures of CH4/C2H6/O2/N2/Ar and C2H6/O2/N2/Ar

OH emission
Pressure / atm
0.10
blend with 100–270 ppm NO2 at temperatures of 900–1050 K and
4
at pressures of 25–50 bar in a rapid compression machine. For the
former mixture, they observed that the addition of NO2 showed a

significantly promoting effect on the reactivity, and this effect
increased with decreasing temperatures. However, only a moder-
ately promoting-effect of NO2 addition was observed for the latter
mixture.
Considering the literature review above, it is clearly that the
impact of NO2 on ethane ignition differs from the effect on both
pure methane and methane/ethane mixtures. In order to full in
the lack of experimental data and understanding of the mutual
sensitization of ethane and NOx, we firstly measure ignition delay
times of C2H6/NO2/O2/Ar mixtures over a wide range of conditions.
Based on our measurements, Whereafter, we use the new ignition
data to evaluate the performances of models by replacement of
sub-models of NOx proposed in literature. Finally, we propose a
detailed kinetic model of C2H6/NO2 to explore the chemical inter-
action of NO2 and C2H6. This work is undoubtedly valuable to
design the combustors of gas engines and to develop the NOx con-
trolling strategies.
2. Experimental
All of the ignition delay times were measured in a stainless steel
shock tube, which has been described in detail in our previous
study [14,23]. Briefly, the shock tube with a diameter of 11.5 cm
is divided into a 4.0 m long driver section and a 4.8 m long driven
section by a double-diaphragm machine. Four fast-response pres-
sure transducers (PCB 113B26) connecting with three time-
interval counters (FLUKE PM 6690) are equally distributed
(30.0 cm interval) at the last 1.3 m of the driven section to measure
the incident shock velocities. Before each experiment, the tube was
evacuated to below 1 Pa with a mechanical-roots vacuum pump,
with a leak rate of less than 1 Pa/min. A piezoelectric pressure
transducer with acceleration compensation (PCB 113B03) is
located at the end-wall to monitor the reflected shock pressure.
Typical pressure rise was evaluated to within 4%/ms due to the
physical interaction between the reflected shock waves and
boundary layer caused from the incident shock waves. OH⁄ light
emission during the ignition was detected by a photomultiplier
(PMT: HAMAMASSU CR131) equipped with a 307 ± 10 nm narrow-
band filter.
The NO2/C2H6/O2/Ar mixtures were prepared in a stainless steel
tank which was allowed to be static >12 h to ensure sufficient mix-
ing and diffusion. The tank was also evacuated to below 1 Pa with a
mechanical-roots vacuum pump system before mixture prepara-
tion. The gas purities used in this study were C2H6 (>99.9%), NO2
*
τ = 2616 s 0.05
2
0 0.00
1000 1500 2000 2500 3000 3500 4000 4500
Time / μs
Fig. 1. Typical end-wall profiles of pressure and OH* emission for N30/1.0 at 5.0 atm
and 1044 K.
(>99.9%) diluted with 20% Ar, O2 (>99.99%) and Ar (99.99%). He
(>99.999%) and N2 (purity 99.999%) were used as the driver gases
to achieve a tailored condition. Detailed compositions of the tested
mixtures are listed in Table 1.
Fig. 1 illustrates the typical profiles of pressure and OH⁄ emis-
sion. The ignition delay time in this study was defined as the time
interval between the incident shock wave arrive at the end-wall,
and the onset of ignition detected by both the pressure sensor
and the PMT. Clearly, the ignition event monitored by the pressure
is in good agreement with that by the light emission. Uncertainty
of the ignition delay time was evaluated to within 20% according
to our previous study [14].
3. Modeling
The simulations were carried out by CHEMKIN program [24]
with SENKIN code [25]. To better reproduce the non-ideal facility
effect, the pressure rise (4%/ms) has been considered in all the sim-
ulations using SENKIN/VTIM approach proposed by Chaos et al.
[26] Moreover, to eliminate the influence of hydrocarbon chem-
istry, the state-of-the-art mechanism, Aramco Mech 2.0 [13,27–
32] was chosen as the base model of ethane because it has been
validated by various data at wide range of conditions. For clarifica-
tion, we compared the ignition delay times of model predictions
and experimental measurements for pure ethane reported in our
previous work [1], Fig. 2. It can be seen that the Aramco Mech
2.0 shows an excellent agreement with the experimental data over
the studied conditions. Four recently published NOx models,
Sivaramakrishnan et al. [21], Giménez-López et al. [33], Mathieu
et al. [18] and Faravelli et al. [22] were integrated into the Aramco
Mech 2.0 to evaluate the effects of NOx sub-models. The four
assembled models are thus named Aramco-S, Aramco-G, Aramco-
M and Aramco-F, respectively. In addition, an updated model

Table 1
The detail constitutes of these mixtures studied in current experiment.

No. Mix Mole blending ratio C2H6 (%) NO2 (%) O2 (%) Ar (%) ref
1 N0/1.0 0% NO2/100% C2H6 1.0 0.0 3.5 95.5 [41]
2 N30/1.0 0% NO2/100% C2H6 1.0 0.3 3.5 95.2 This study
3 N100/1.0 0% NO2/100% C2H6 1.0 1.0 3.5 94.5 This study
4 N0/0.5 5% NO2/100% C2H6 1 0.0 7.0 92.0 [41]
5 N30/0.5 15% NO2/100% C2H6 1 0.3 7.0 91.7 This study
6 N0/2.0 50% NO2/100% C2H6 1.0 0.0 1.75 97.25 [41]
7 N30/2.0 5% NO2/100% C2H6 1.0 0.3 1.75 96.95 This study
 F. Deng et al. / Fuel 207 (2017) 389–401 391

For the fuel-rich mixture (N30/2.0), Fig. 3a–c, all the five models
(a) Xethane = 1% are in satisfactory agreement with the ignition delay times partic-
ϕ = 0.5 ularly at low pressures (p < 5 atm). However, the differences of the
p = 1.2 atm models in predicting ignition delay times are obvious at 20 atm
Ignition delay time (μs)

3
10 p = 5.0 atm and at temperature below 1200 K. The Aramco-M shows the slow-
p = 20 atm est ignition times which are clearly beyond the experimental
uncertainty. For the fuel-stoichiometric mixture (N30/1.0), Fig. 3d–f,
the proposed model and Aramco-M present similar predictions
and reproduce well the experimental data at all pressures over
the entire temperatures. However, the Aramco-F dramatically
under-predicts especially at lower temperatures (<1300 K) and
high pressures (>5.0 atm). Specifically, the ignition delay times
2
10 simulated by Aramco-F shows three times lower than the experi-
Simulations: Aramco Mech 2.0 ments at 20 atm and 1050 K. The Aramco-S and the Aramco-G pre-
sent close predictions but still give 60% faster prediction in the
0.65 0.70 0.75 0.80 0.85 0.90 0.95
-1 reactivity at 20 atm and at temperature below 1100 K. For the
1000/T (K ) fuel-lean mixture, N30/0.5, Fig. 3g–i, the differences in the model
predictions become more significant at even low pressure
(b) ϕ = 1.0 (1.0 atm). The Aramco-M shows the slowest ignition and over-
predicts the experimental results by a factor of two at the temper-
atures below 1175 K under all pressures. Moreover, the Aramco-F
Ignition delay time (μs)

3
10
presents three times faster prediction in the ignition delay times
at the temperature below 1110 K, and this deviation becomes more
prominent with increasing pressures and decreasing temperatures.
In contrast, the Aramco-S, the Aramco-G and the proposed model
reproduce well the ignition delay times and the global activation
energy (Ea) at all conditions investigated in this study. Fig. 4a–c
2 shows the comparison of the five model simulations and the
10
current experiments for the N100/1.0 mixture at pressures of
1.2–20 atm. It can be seen that all the models are capable of repro-
ducing the experimental data in consideration of the measurement
errors except the Aramco-F. Furthermore, the results simulated by
0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 the proposed model, the Aramco-S, the Aramco-G and the Aramco-
-1
1000/T (K ) M are similar (within 30%) and are in good agreement with the
experimental measurements. However, the simulations predicted
(c) ϕ = 2.0 by Aramco-F show too fast ignition and this discrepancy increases
with increasing pressures and decreasing temperatures.
In an attempt to further test the model performance, the litera-
Ignition delay time (μs)

10
3
ture data reported by Herzler and Naumann [17] and Sivaramakr-
ishnan et al. [21] have been simulated using the proposed model. It
can be seen that the proposed model shows an excellent agree-
ment with the ignition delay times of C2H6/CH4/O2/Ar mixture
with 50 ppm NO2 at pressure of 16 atm and equivalence ratio of
0.25 reported by Herzler and Naumann [17], Fig. 5. Likewise, the
proposed model is also capable of reproducing the concentrations
2
10 of major species in the oxidation of C2H6/CH4/N2 mixtures with
200 ppm NO2 at pressure of 10 atm reported by Sivaramakrishnan
et al. [21], Figs. 6 and 7.

0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90

-1
1000/T (K ) 4. Results and discussion

Fig. 2. Comparisons of simulated and measured ignition delay times for pure
4.1. Pressure-dependence
ethane mixtures at pressures of 1.2–20 atm and equivalence ratios of 0.5–2.0.
Experiments are taken from Hu et al. [41] and model predictions are simulated by
Aramco Mech 2.0. Similarly to our previous work of CH4/NO2 [14], the trends with
respect to the influence of pressure on the ignition delay times are
in-line for all tested mixtures. Overall, increasing pressure pro-
proposed in this study is also used to compare with other four mod- motes the reactivity of C2H6/NO2 mixtures at the entire tempera-
els and the experiments. Specifically, the proposed model consists tures and this promotion is more prominent with increasing
of reactions involving the H2/CO/O2/NOx sub-mechanism devel- temperatures, Fig. 8. Specifically, the ignition delay times decrease
oped by Zhang et al. [34], the C1/NOx sub-mechanism constructed 58.9% and 43.4% at 1430 K when pressure increases from 1.2 to
by Mathieu et al. [18], the C2/NOx sub-mechanism built by 5.0 atm and from 5.0 to 20.0 atm, respectively. However, it only
Giménez-López et al. [33], and the C2H5NO2 sub-mechanism taken decreases 27.6% and 28.7% at 1150 K. Interestingly, the increase
from Zhang et al. [35,36] The rate constants of reactions C2H6 + of pressure presents an opposite impact on the global activation
NO2 <=> Ċ2H5 + HONO/HNO2 and CH4 + NO2 <=> ĊH3 + HONO/ energy for ethane ignition with NO2 (N100/1.0) and without NO2
HNO2 was updated by the latest value calculated by Chai et al. [37]. (N0/1.0). For the N0/1.0 mixture, Fig. 8(a), the increase of pressure
392 F. Deng et al. / Fuel 207 (2017) 389–401

(a) N30/2.0 ϕ = 2.0 (b) N30/2.0 ϕ = 2.0 (c) N30/2.0 ϕ = 2.0

p = 1.2 atm p = 5.0 atm p = 20 atm

Ignition delay times (μs)

103

Model predictions:
Aramco-S
Aramco-F
102 Aramco-G
Aramco-M
Current model
0.60 0.65 0.70 0.75 0.80 0.60 0.65 0.70 0.75 0.80 0.85 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00
1000/T (K-1) 1000/T (K-1) 1000/T (K-1)

(d) N30/1.0 ϕ = 1.0 (e) N30/1.0 ϕ = 1.0 (f) N30/1.0 ϕ = 1.0

p = 1.2 atm p = 5.0 atm p = 20.0 atm
Ignition delay time (μs)

103

Model predictions:
Aramco-S
102 Aramco-F
Aramco-G
Aramco-M
Current model
0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00 0.70 0.75 0.80 0.85 0.90 0.95 1.00
1000/T (K-1) 1000/T (K-1) 1000/T (K-1)

(g) N30/0.5 ϕ = 0.5 (h) N30/0.5 ϕ = 0.5 (i) N30/0.5 ϕ = 0.5

p = 1.2 atm p = 5.0 atm p = 20.0 atm

Ignition delay time (μs)

3
10

Model predictions:
Aramco-S
Aramco-F
102 Aramco-G
Aramco-M
Current model

0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00 0.70 0.75 0.80 0.85 0.90 0.95 1.00 0.75 0.80 0.85 0.90 0.95 1.00 1.05 1.10
1000/T (K-1) 1000/T (K-1) 1000/T (K-1)

Fig. 3. Comparisons between the experimental data and the model predictions for the N30/0.5-2.0 mixtures at different pressures.

(a) N100/1.0 ϕ = 1.0 (b) p = 5.0 atm (c)

p = 1.2 atm p = 20.0 atm

Ignition delay time (μs)

103

Model predictions
Aramco-S
Aramco-F
102
Aramco-G
Aramco-M
Current model
0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00 0.70 0.75 0.80 0.85 0.90 0.95 1.00 1.05 1.10
1000/T (K-1) 1000/T (K-1) 1000/T (K-1)

Fig. 4. Comparisons between experimental data and model predictions for the N30/1.0 mixture at pressures of 1.2–20 atm.

increases the Ea. For example, when the pressure increases from presents the limited effect and slightly reduces the Ea. The differ-
1.2 to 5.0 atm, the Ea increases from 10 kcal/mol to 15 kcal/mol. ent pressure-dependence of global activation energy for the mix-
However, when the pressure further increases, no obvious change tures with and without NO2 can be attributed to such a fact that
in Ea can be observed in both experimental measurements and reactive Ḣ atoms are eliminated via reaction Ḣ + O2 (+M) = HȮ2
model predictions. For the N100/1.0 mixture, Fig. 8(b), the pressure (+M) to form stable HȮ2 radicals at high pressure and lower
 F. Deng et al. / Fuel 207 (2017) 389–401 393

4
10 220 (a) p = 10 atm ϕ = 1.0
C2H6/CH4/O2/Ar with 50 ppm NO2
200
p = 16 atm ϕ = 0.25 180
160

Mole fraction (ppm)

Ingition delay times (μs)

140
3 120
10
100
NO
80
NO 2
60
Solid lines: Current model
40
2 20
10
Experiments 0
Current model 750 800 850 900 950 1000 1050 1100 1150 1200
Temperature (K)
250
0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 (b)
-1
1000/T (K )
200
Fig. 5. Comparisons between the simulated and experimental ignition delay times

Mole Fraction (ppm)

for NO2/C2H6/CH4/O2/Ar mixture at p = 16 atm and / = 0.25. Experiments are taken 150
from Herzler and Naumann [17].

100
C2H6
temperatures for pure C2H6. This reduces the reactivity of the reac-
C2H4
tion system. For the case of NO2 addition to C2H6, however, the 50
CH2O
influence of the reaction Ḣ + O2 (+M) = HȮ2 (+M) dramatically is
weakened due to the twofold: a) reaction NO2 + Ḣ = NO + ȮH com- 0
petes Ḣ atoms with the reaction Ḣ + O2 (+M) = HȮ2 (+M) and b) the
formed stable HȮ2 radicals readily react with NO to yield reactive 750 800 850 900 950 1000 1050 1100 1150 1200 1250
chain-initiating ȮH radicals via reaction NO + HȮ2 = NO2 + ȮH. As a Temperature (K)
result, the global activation energy exhibits little pressure- 6000
dependence. Moreover, the ignition delay times appear to be more (c)
sensitive to the change in pressures for the N100/1.0 mixture. 5000
CH4
CO
Mole Fraction (ppm)

4.2. The effect of NO2 on ethane reactivity
Fig. 9 illustrates the effect of NO2 addition on ignition of stoi-
chiometric ethane mixtures at pressures of 1.2, 5.0 and 20 atm.
Again, the similar impact of NO2 on CH4 ignition [14] is also
observed in C2H6/NO2 system, namely the NO2 addition increases
the reactivity of C2H6 and reduces the ignition delay times over
the entire conditions investigated here. Moreover, the promoting
effect presents a strong pressure- and temperature-dependence
meaning that the global activation energy is changed with NO2
addition. With increasing pressures and decreasing temperatures,
the effect of NO2 addition becomes more significant. Specifically,
at low pressure (1.2 atm), Fig. 9(a), the NO2 addition only shows
a negligible effect. However, at higher pressures (>5 atm), the
NO2 remains the ignorable impact when the temperatures over
1330 K whereas it shows a greater promoting effect when the tem-
peratures below 1330 K, Fig. 9(b) and (c). In more detail, the reduc-
tions in the ignition delay times are 54.1% and 75.9% at 20 atm and
1100 K for the N30/1.0 and N100/1.0 mixtures, respectively.
Fig. 10 depicts the effect of NO2 addition on ignition of lean
ethane mixtures at pressures of 1.2, 5.0 and 20 atm. Again, the
NO2 addition exhibits the obvious pressure- and temperature-
dependence. It is similar to stoichiometric mixtures, the promoting
effect of NO2 is significant at higher pressures (
5.0 atm) and
lower temperatures (<1250 K). Except for the deference that the
NO2 effect at fuel-lean condition appears to be more noticeable rel-
ative to that at fuel-stoichiometric condition. Taking 20 atm as an
example, for fuel-lean mixture, 30% NO2 addition reduces the igni-
tion delay times by a factor of four at 1110 K, Fig. 10(c). However, it
only decreases the ignition delay times by a factor of two at the
some temperatures for fuel-stoichiometric mixture, Fig. 9(c). Fur-
thermore, a turning point of temperature can be found in Fig. 10
4000
CO2
H2O
3000
2000
1000
0
750 800 850 900 950 1000 1050 1100 1150 1200
Temperature (K)
Fig. 6. Comparisons between the simulated and experimental species concentra-
tions for NO2/C2H6/CH4/O2/N2 mixture at p = 10 atm and u = 1.0. Experiments are
taken from Sivaramakrishnan et al. [21] (a) Concentrations of NOx; (b) concentra-
tions of C2H6, C2H4 and CH2O; (c) concentrations of CH4, CO, CO2 and H2O.
(a) and (b) where the NO2 presents an opposite effect on the ethane
ignition, namely inhibiting impact at high temperatures and
enhancing at lower temperatures. With increasing pressures, the
turning point moves to higher temperature side. Specifically, the
turning point appears at 1190 K under 1.2 atm while it arises to
1260 K under 5.0 atm. Detailed interpretation will be presented
in the next section.
Fig. 11 shows the NO2 effects on ignition of rich ethane mixtures
at three pressures (1.2, 5.0 and 20 atm). Overall, the NO2 addition
presents a similar pressure- and temperature-dependence as
fuel-stoichiometric and fuel-lean conditions, but the effect is
insignificant particularly at pressures of 1.0 and 5.0 atm. Only
when the pressure increases to 20 atm and the temperature
decreases to 1300 K, the NO2 addition pays a role in accelerating
the ethane ignition, Fig. 11(c).
394 F. Deng et al. / Fuel 207 (2017) 389–401

200 (a) N0/1.0

(a)
p = 1.2 atm
p = 5.0 atm

Ignition delay time (μs)

3
150 10 p = 20 atm
Mole fraction (ppm)

15.0 kcal/mol

100 15.0 kcal/mol

NO
NO2
10.1 kcal/mol
p = 10 atm ϕ = 0.5 2
50 10
Solid lines: Current model

Solid lines: Current model

0
0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95
800 900 1000 1100 1200 -1
1000/T (K )
Temperature (K)
250
(b)
(b) N100/1.0
200
9.8 kcal/mol

Ignition delay time (μs)

Mole fraction (ppm)

3
10
150

C2H6 11.0 kcal/mol

100 9.3 kcal/mol
CH2O

50
C2H4
102

800 900 1000 1100 1200 0.6 0.7 0.8 0.9 1.0 1.1
Temperature (K) 1000/T (K-1)

Fig. 8. Effects of pressure on the reactivity of stoichiometric C2H6/NO2 mixtures. (a)

6000
(a) N0/1.0; (b) N100/1.0.

5000 CH4
CO pressure and high temperature (1.2 atm and 1500 K) where the
Mole fraction (ppm)

4000 CO2 NO2 has only weak effect, were selected to identify the controlling
reactions. The normalized sensitivity coefficient is defined as fol-
H2O
3000 lowing [38]:
sð2ki Þ  sð0:5ki Þ
Si ¼ ð1Þ
2000 1:5sðki Þ

1000
0
800 900 1000
Temperature (K)
Fig. 7. Comparisons between the simulated and experimental species concentra-
tions for NO2/C2H6/CH4/O2/N2 mixture at pressure of 10 atm and u = 0.5. Experi-
ments are taken from Sivaramakrishnan et al. [21] (a) Concentrations of NOx; (b)
concentrations of C2H6, C2H4 and CH2O; (c) concentrations of CH4, CO, CO2 and H2O.
5. Kinetic analysis
To chemically clarify the promoting effect of NO2 on ethane
ignition at various conditions, kinetic analyses including brute-
force sensitivity analysis and flux analysis were carried out for
the C2H6 mixtures with and without NO2. Two typical conditions,
high pressure and lower temperature (20 atm and 1050 K) where
the NO2 addition presents significantly promoting effect and low
where Si is sensitivity coefficient, ki is rate constant of ith reaction,
and s is the ignition delay time.
5.1. Controlling reactions at high pressure and lower temperature
1100 1200
Fig. 12 shows the brute force sensitivity analysis for three stoi-
chiometric C2H6/NO2 mixtures (N0/1.0, N30/1.0 and N100/1.0) at
20 atm and 1050 K. Results indicate that for N0/1.0, Fig. 12(a), it is
no exception that chain-branching reaction (R5: O2 + Ḣ <=> Ö
+ ȮH) is the most promoting reaction, while H-atom abstraction
from ethylene by ȮH radical (R248: C2H4 + ȮH <=> Ċ2H3 + H2O) is
the second promoting one. Chain propagation reaction producing
methoxy (CH3Ȯ) and ȮH radical via R99: ĊH3 + HȮ2 <=> CH3Ȯ
+ ȮH is the third promoting reaction. Note that the reactions
involving HȮ2 chemistry pay a role in ethane ignition at this con-
dition. Moreover, chain termination reaction producing methane
(CH4) and molecular oxygen via R49: ĊH3 + HȮ2 <=> CH4 + O2 is
the most inhibiting reaction.
When NO2 adds to the ethane system, the top dominating reac-
tions are changed remarkably, Fig. 12(b). It can be seen clearly that
 F. Deng et al. / Fuel 207 (2017) 389–401 395

promoting reaction due to such the fact that a) formation of active

(a) p = 1.2 atm, ϕ = 1.0
Ḣ atoms via generated CH3Ȯ decomposition R174: CH3Ȯ (+M)
<=> CH2O + Ḣ (+M) and b) consumption of relatively stable CH3
radicals. The former can feed R5 to accelerate the establishment
Ignition delay times (μs)

10
3
of radical pool. Furthermore, the second sensitive reaction R248
in pure ethane system is less important in the C2H6/NO2 system.
Interestingly, the HȮ2 chemistry plays no long dominant role in
the controlling ignition kinetics. The reason is that NO2 is more
favorable to react with fuel relative to molecular oxygen which
N100/1.0 leads to significantly decrease the formation of HȮ2 radicals via
N30/1.0 H-atom abstraction from Ċ2H5 radicals by O2 in the fuel oxidation
2 [14]. Additionally, the small amount of HȮ2 radicals formed in the
10 N0/1.0
system can be readily eliminated via the conversion of NO to NO2
Solid lines: Current model cycle [39–44]. The reactions R3151 and R3139 show the compara-
ble sensitivity coefficients with R5 because those reactions con-
0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 sume the stable species (C2H6 and CH2O) and generate HONO
1000/T (K )
-1 which quickly decomposes to ȮH radicals and NO.
It is well known that chemical kinetic process during initial
ignition period is vital to establish radical pool which controls
(b) p = 5.0 atm, ϕ = 1.0 induced time and reactivity. To explore the contribution of NO2
on the development of radical pool, flux analysis was conducted
for stoichiometric N0/1.0, N30/1.0 and N100/1.0 mixtures at 1050 K
and 20 atm during the initial induced time (reaction time < 10 ls),
Ignition delay times (μs)

3
10
Fig. 13. For N0/1.0, Fig. 13(a), ethane is mainly consumed via two
pathways: a) decomposition R194: C2H6 (+M) <=> ĊH3 + ĊH3 (+M)
producing less reactive ĊH3 radicals followed by O2 addition form-
ing CH3Ȯ2 and b) H-atom abstraction by O2 to produce ethyl (Ċ2H5)
and HȮ2 radicals. Subsequently, the Ċ2H5 radicals either add to O2
forming C2H5Ȯ2 radicals or react with O2 via H-atom abstraction
2 producing C2H4 and HȮ2 radicals. Obviously, the HȮ2 chemistry
10
pays a role in the ignition of pure ethane at current condition,
and this is consent with the brute force analysis above. However,
with increasing NO2, Fig. 13(b) and (c), majority of ethane react
0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00 with NO2 via reactions R3139 rather than direct decomposition.
1000/T (K )
-1
The formed HONO readily decomposes to ȮH radicals via
R2754: HONO (+M) <=> NO (+M)) + ȮH, which feeds R199:
(c) p = 20 atm, ϕ = 1.0 C2H6 + ȮH <=> Ċ2H5 + H2O to further accelerate the oxidation of
fuels. Additionally, the stable ĊH3 and CH2O can also react with
NO2 to generate active CH3Ȯ and HĊO radicals followed by
H-atom elimination. As a result, the radical pool is rapidly
Ignition delay times (μs)

3
10
2
10
0.70 0.75 0.80 0.85
1000/T (K )
Fig. 9. Effects of NO2 mixing ratio on the ethane ignition at p of 1.2–20.0 atm. (a)
1.2 atm; (b) 5.0 atm; (c) 20 atm.
NO2 addition perturbs the ethane ignition process resulting in a
significant role of the reactions involving NO2 interaction chem-
istry in the C2H6/NO2 system, and these reactions become more
prominent with increasing NO2 concentration. As a result, the four
reactions (R3045: ĊH3 + NO2 <=> CH3Ȯ + NO, R3151: CH2O
+ NO2 <=> HCO + HONO, R3139: C2H6 + NO2 <=> Ċ2H5 + HONO and
R3049: CH3O2 + NO <=> CH3Ȯ + NO2) in the top five promoting
reactions are all relative to the interaction chemistry of C2H6
and NOx. Among them, R3045 replacing R5 becomes the most
developed during the ignition induced time. Furthermore, mole
fraction analysis also suggests that the NO2 addition increases
the mole fractions of the radical pool (Ḣ, Ȯ and ȮH) by two orders
of magnitude during the induced time Fig. 14.
In contrast, for N0/1.0, the C2H6 is mainly consumed through
H-atom abstraction by ȮH (72.2%) and HȮ2 (6.5%) radicals and Ḣ
atoms (15.3%) to produce Ċ2H5 radicals at main ignition stage,
Fig. 15. The Ċ2H5 radicals subsequently undergo three pathways:
a) 56.9% of Ċ2H5 radicals react with O2 forming C2H4 and HȮ2 rad-
icals via reaction (R215: Ċ2H5 + O2 <=> C2H4 + HȮ2); b) 24% of Ċ2H5
0.90 0.95 1.00 1.05 1.10 radicals directly decompose to C2H4 and Ḣ atoms via C–H beta scis-
-1 sion (R–207: C2H4 + Ḣ (+M) <=> Ċ2H5 (+M)), and both the species
promote the reactivity of the mixture; and c) 11% of Ċ2H5 radicals
react with HȮ2 producing C2H5Ȯ radicals which finally decompose
to ĊH3 and CH2O which inhibit further chain branching or propaga-
tion. With NO2 addition, the fuel consumption via the H-atom
abstraction by ȮH increases (72.2% for N0/1.0 increases to 92.2%
for N100/1.0) and via the H-atom abstraction by HȮ2 decreases
(6.5% for N0/1.0 decreases to 0% for N100/1.0). The reason can be
attributed to twofold: a) abundant ȮH radicals accumulated in
the initial induced time naturally accelerate the fuel consumption
via R199 and promote the reactivity and b) most of HȮ2 radicals
generated from ethane system can be converted to ȮH radicals
via the internal conversion of NO to NO2 (R2722: NO
+ HȮ2 <=> NO2 + ȮH). Moreover, the Ċ2H5 radicals can react with
NO2 either to produce C2H5Ȯ via chemically activated process or
396 F. Deng et al. / Fuel 207 (2017) 389–401

(a) ϕ = 0.5 (b) (c)

p = 1.2 atm p = 5.0 atm p = 20 atm

Ignition delay time (μs)

Ignition delay time (μs)

103

Ignition delay time (μs)

103 103

N0/0.5 102
102
N30/0.5
102
Solid Lines: Current model

0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00 0.70 0.75 0.80 0.85 0.90 0.95 1.00 0.75 0.80 0.85 0.90 0.95 1.00 1.05 1.10
1000/T (K-1) 1000/T (K-1) 1000/T (K-1)

Fig. 10. Effects of NO2 addition on the ethane ignition at pressures of 1.2–20 atm and u = 0.5. (a) 1.2 atm; (b) 5.0 atm; (c) 20 atm.

(a) ϕ = 2.0 (b) (c)

p = 1.2 atm p = 5.0 atm p = 20 atm

Ignition delay time (μs)

3
Ignition delay time (μs)

103 10
Ignition delay time (μs)

103

N0/2.0
N30/2.0 102
102
Solid Lines: Current model
102
0.60 0.65 0.70 0.75 0.80 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.65 0.70 0.75 0.80 0.85 0.90 0.95
1000/T (K-1) 1000/T (K-1) 1000/T (K-1)

Fig. 11. Effects of NO2 addition on the ethane ignition at pressures of 1.2–20 atm and u = 2.0. (a) 1.2 atm; (b) 5.0 atm; (c) 20 atm.

(a) (b)

N100/1.0
N0/1.0 N300/1.0
p = 20 atm
T = 1050 K

-0.15 -0.10 -0.05 0.00 0.05 0.10 -0.20 -0.15 -0.10 -0.05 0.00 0.05 0.10 0.15
Sensitivity analysis Sensitivity analysis

Fig. 12. Normalized sensitivity analysis for (a) N0/1.0; (b) N15/1.0 and (c) N50/1.0 mixtures at 20 atm and 1050 K using current model.

to form C2H5NO2 via stabilization process. The C2H5NO2 will subse- It is worth noting that the radical concentration stagnates after
quently decompose to C2H4 and HONO which further undergoes the exponential increase for the mixtures with NO2 addition but
unimolecular decomposition to form ȮH radicals and this is sec- with the continuous increase for pure ethane, Fig. 14. To clarify this
ondary promoting route. This analysis is consensus with the liter- phenomenon, the radical pool analysis was conducted at the reac-
ature [14,18] that the NO2 promoting ignition of hydrocarbons is tion time = 200 ls, as shown in Table 2.
attributed to accelerate establishment of free radical pool. For the pure ethane, Ċ2H5 radicals play the dominated role on
In addition, the stable intermediates (ĊH3 and CH2O) generated the formation of Ḣ atoms and HȮ2 radicals. The Ḣ atoms feed the
in the ethane oxidation react with NO2 via R3001: ĊH3 + NO2 (+M) chain-branching reaction R5 to produce Ö atoms, while the most
<=> CH3NO2 (+M), R3045 and R3151. The active CH3Ȯ and HONO of HȮ2 radicals react with C2H6 to form H2O2, followed by directly
generated from the latter two reactions readily decompose to form decomposition to produce ȮH radicals. It can be thus concluded
abundant Ḣ atoms and ȮH radicals to further promote the fuel oxi- that the production and consumption of Ċ2H5 dominate the estab-
dation, and their promoting effects have also been reflected to the lishment of radical pool at this stage. After the initiating period, the
sensitivity analysis, Fig. 12. system accumulates a certain amount of radicals (Ḣ, Ö, ȮH
 F. Deng et al. / Fuel 207 (2017) 389–401 397

C2H 6
M 52.4
CH 3
10.9 31.6 O 2 66.0
O 2 30.7
CH 3 O 2
O2
C2H 5O 2 C2H 5
23.3 Branching ratio of CH 3
M 14.7
O2 Branching ratio of C 2 H 6
60.1
C2H 4
(a)

C2H 6 C2H 6
O 4.50 H 0.00
OH 27.8 O 7.30
NO 2 67.0 (produce: HONO) OH 27.8
NO 2 64.7 (produce: HONO)

C2H 5 29.3
C2H 5O C2H 5 29.7
C2H 5O
NO 2 92.2 NO 2 92.2
NO 2 NO 2
69.0 CH 3 69.7 CH 3
NO 2 24.0 NO 2 NO 2 24.3 NO 2

C 2H 5NO 2 CH 2O(produce: H)CH 3O

99.9
C 2H 5NO 2 CH 2O(produce: H)CH 3O
74.9 99.9 75.4

NO 2
NO 2 97.1
97.2 (produce: HONO)
(produce: HONO)

HCO CH 3NO 2 HCO CH 3NO 2

(b) (c)
Fig. 13. Flux analysis of stoichiometric C2H6/NO2 mixtures at 20 atm and 1050 K during the ignition induction period (reaction time < 10 ls) using current model (10% fuel
consumption). (a) N0/1.0; (b) N30/1.0; (c) N100/1.0.

For the case of NO2 addition to C2H6, the produced Ċ2H5 radicals
1E-3 mainly react with NO2 to produce C2H5NO2 rather than react with
T = 1050 K O2 to form HȮ2 or decompose to Ḣ atoms for one thing. Thus, the
1E-4
p = 20 atm production of radical pool through the consumption of Ċ2H5 radi-
ϕ = 1.0 cals generally reduces. But for another, abundant of ȮH radicals
1E-5 can be produced through the internal cycle between NO and
Mole Fraction

NO2. The opposite contribution to the radical pool by both the

aspects above results in slow increase in the radical concentration.
1E-6

1E-7 5.2. Controlling reactions at low pressure and higher temperature

N0/1.0
N30/1.0 Fig. 16 depicts the brute force sensitivity analysis for three sto-
1E-8
N100/1.0 ichiometric C2H6/NO2 mixtures (N0/1.0, N30/1.0 and N100/1.0) at
1.2 atm and 1500 K to identify the controlling reactions in the
1E-9
C2H6/NO2 ignition at low pressure and high temperature. For
10 100 1000 N0/1.0, Fig. 16(a), again, R5 thoroughly dominates the ethane igni-
Time (μ s) tion while the others show only small sensitivity coefficients. With
increasing NO2, the dominant reactions remains similar order as
Fig. 14. Effects of NO2 on the mole fraction of the radical pool (Ḣ, Ö and ȮH) in
stoichiometric C2H6 ignition at 20 atm and 1050 K using current model.
the pure ethane, and no significant change can be found, Fig. 16
(b). All the top sensitive reactions are not relative to NOx chem-
istry, indicating that the presence of NO2 has quite limited impact
and HȮ2) which accelerate the consumption of ethane to produce on the ignition kinetic of ethane at such the condition.
Ċ2H5 radicals, meaning the radical pool will continuously increase Likewise, the flux analysis of N0/1.0, N30/1.0 and N100/1.0 during
after the initial stage. initial ignition period was performed at this condition, Fig. 17.
398 F. Deng et al. / Fuel 207 (2017) 389–401

NO2 0.00 18.7 26.3

C2H5NO2
M 0.00
95.2
90.5
OH H HO2 NO2 O2 M
C2H6 C2H5 56.9 23.8 C2H4HO215.90.000.00 C2H4O1-2
72.2 15.3 6.50 0.00 6.60 2.00
89.2 5.20 0.00 1.60 0.00 0.00
92.2 0.00 0.00 6.40 NO2 0.00 66.5 71.0 OH 60.5 72.3 0.00

C2H6 HO2 10.8 0.00 0.00

O2 29.9 23.3 0.00
28.2 C2H5O O2 C2H3 C2H3OO
0.00
0.00 10.8 28.3 O2 NO2 NO2
1.90 11.8 1.70 91.4 20.7 0.00
CH3O2 O2
91.6
0.00
35.3
26.7 8.30
0.00
15.5
0.00 0.00 0.00
51.6 25.7 0.00
CH3NO2 NO2 CH3 CH2O CH2CHO C2H3NO2
CH4 0.00 0.0013.2
HO2 NO2 0.00 55.8 45.7
H
OH
20.9 7.30 0.00
43.8 55.0 33.9
HO2 42.5 0.00 0.00 HO2 22.0 0.00 0.00
NO2 0.00 23.7 51.3

CH3O HCO
Fig. 15. Flux analysis for stoichiometric C2H6/NO2 mixtures at 20 atm and 1050 K using current model (10% fuel consumption). N0/1.0: red, N30/1.0: blue and N100/1.0: black. (For
interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Table 2 N30/1.0 and 81.6% for N100/1.0) still dominates the fuel consumption
The main production channels of radicals for pure ethane at p = 20 atm, T = 1050 K meaning that the NO2 only limited participates into the ignition
and reaction time = around 200 ls.
kinetic of ethane during the ignition induced stage. Furthermore,
Radical pool Reactions Ratio only little change in establishment of free radical pool can be
Ö atom Ḣ + O2 = Ö + ȮH 95% observed with the change of NO2 concentration at low pressure
Ḣ atom C2H4 + Ḣ ( + M) = Ċ2H5( + M) 90% and high temperature compared to higher pressure and lower tem-
ȮH radical H2O2 ( + M) = 2ȮH ( + M) 88% perature, Fig. 18.
Other radicals Reactions Ratio Fig. 19 shows the flux analysis of stoichiometric C2H6/NO2 mix-
HȮ2 Ċ2H5 + O2 = C2H4 + HȮ2 80% tures at 1.2 atm and 1500 K during main ignition stage. For N0/1.0,
Ḣ + O2 ( + M) = HȮ2 ( + M) 6% fuel mainly undergoes H-atom abstractions by Ḣ atoms (58.8%),
H2O2 C2H6 + HȮ2 = Ċ2H5 + H2O2 93% ȮH radicals (22.2%) and Ö atoms (11.1%) forming Ċ2H5 radicals fol-
lowed by decomposition to form two active species, C2H4 and Ḣ
atom, via R207. When adding NO2 to the reaction system however,
For N0/1.0, Fig. 17(a), ethane mainly undergoes the decomposition the branching ratio of Ċ2H5 consumption is changed notably. The
reaction R194 (95.2%) producing ĊH3 radicals, which feed R203: direct decomposition of Ċ2H5 reduces from 90.4% for N0/1.0 to
C2H6 + ĊH3 <=> Ċ2H5 + CH4 to form Ċ2H5 radicals. The Ċ2H5 radicals 55.8% for N100/1.0 while the reaction of Ċ2H5 and NO2 increases
subsequently decompose to C2H4 and Ḣ atoms via C–H beta ses- from 0% to 42.1%. As a result, more stable species including
sion. With increasing NO2, Fig. 17(b) and (c), R194 (82.9% for CH2O, ĊH3 and CH3CHO can be formed via C–C beta scission and

(a) (b)

N100/1.0
N0/1.0 N30/1.0
p = 1.2 atm
T = 1500 K

-1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6
Sensitivity analysis Sensitivity analysis
Fig. 16. Normalized sensitivity analysis for C2H6/NO2 mixtures at 1.2 atm and 1500 K using the current model. (a) N0/1.0; (b) N30/1.0 and N100/1.0.
 F. Deng et al. / Fuel 207 (2017) 389–401 399

C2H6
M 95.2 CH3
3.50 89.5
Branching ratio of C2H6 Branching ratio of CH3

CH4
C2H5
M 98.4 Produce: H

C2H4
(a)

C2H6
M 82.9 CH3
75.1
NO2
CH3O C2H6
M 81.6 CH3
87.4
NO2
CH3O
M 88.8 Produce: H
M 87.6
2.70 18.4 O2 10.6 Produce: HO 2
1.2 6.4 Produce: H

Branching ratio of C2H6 Branching ratio of CH3 Branching ratio of C2H6 Branching ratio of CH3

CH4 CH2O CH4 CH2O

OH 6.0 13.6 46.0
C2H5 NO2
C2H5 NO C2H5O
2
M 98.4 Produce: H M 52.3 Produce: H

C2H4 C2H4
(b) (c)

Fig. 17. Flux analysis at 1.2 atm and 1500 K during the ignition induction period (reaction time < 10 ls) using current model. (a) N0/1.0; (b) N30/1.0; (c) N100/1.0.

T = 1500 K CH2CHO
-2 C2H6
10 p = 1.2 atm
ϕ = 1.0 H 58.8 21.0 7.00 0.00
O 11.1 4.50 2.40 2.90
OH 22.2 63.2 76.4 12.1
-3
O
Mole Fraction

10
O2 M
C2H5 C2H4
1.40 90.4
NO2 0.00 81.3
OH 44.0 76.1 76.2
0.00 0.00 55.8 O(+HCO)
-4
10 14.5 H 21.7 4.70 1.30
N0/1.0 42.1 NO2 0.00 0.00 9.90
14.8
N30/1.0 3.60
N100/1.0 0.00
0.00 72.0 71.8 C2H3
C2H5O
10 100
Time (μs)
0.00
Fig. 18. Effects of NO2 on the mole fraction of the radical pool (Ḣ, Ö and ȮH) in
CH2O CH3
stoichiometric C2H6 ignition at 1.2 atm and 1500 K using current model. 28.0
28.2

C–H beta scission in the C2H5Ȯ generated by the reaction of Ċ2H5

with NO2. In contrast, the formation of C2H4 and Ḣ atoms is inhib- CH3CHO
ited by the NO2 addition. The two reasons above cause that adding
NO2 inhibits the reactivity of ethane at low pressure and high tem- Fig. 19. Flux analysis for stoichiometric C2H6/NO2 mixtures at 1.2 atm and 1500 K
using current model. N0/1.0: red, N30/1.0: blue and N100/1.0: black. (For interpretation
perature, and this is consistent with experimental observations,
of the references to colour in this figure legend, the reader is referred to the web
Figs. 10 and 11. version of this article.)

6. Conclusions
performances against the new ignition data. A high temperature
In this study, ignition delay times of NO2/C2H6/O2/Ar mixtures kinetic mechanism was proposed to interpret the chemical interac-
with mole blending ratio of NO2:C2H6 = 0.3/1.0 and 1.0/1.0 were tion of C2H6 and NO2 at different conditions. The main conclusions
measured behind reflected shock wave at pressure ranging from are summarized as follows:
1.2 to 20 atm and at equivalence ratios of 0.5–2.0 for temperature
ranging from 950 to 1700 K. Four recent NOx sub-mechanisms 1. Addition of NO2 generally promotes the reactivity of ethane at
were integrated with Aramco Mech 2.0 to evaluate their high pressures and lower temperatures and this effect becomes
400 F. Deng et al. / Fuel 207 (2017) 389–401
more significant with increasing pressures and decreasing tem-
peratures and equivalence ratios. However, it only plays a lim-
ited impact at lower pressures and higher temperatures.
Moreover, the promotion enhances for the mixtures with high
concentration of NO2.
2. Performances of four assembled models, Aramco-S, Aramco-F,
Aramco-G and Aramco-M, were evaluated against the new
experimental data. Results indicate that these models are con-
sensus and are in acceptable agreement with the ignition delay
times at fuel-rich condition. However, the four models show the
significant deviation at fuel-lean and fuel-stoichiometric condi-
tions, and none of them is capable of reproduce the experimen-
tal observations at even low pressure. In contrast, the proposed
model shows in good agreement with the new measurements
as well as the literature data at all conditions.
3. Sensitivity and flux analyses indicate that addition of NO2
changes the dominant reactions and critical branching ratios
of key reactions during ethane oxidation. At high pressure and
lower temperature, the promoting effect of NO2 can be con-
cluded below:
a) At ignition induction period, the presence of NO2 changes
initial consumption of ethane, and accelerates the develop-
ment of radical pool via the reaction sequence: C2H6 +
NO2 <=> Ċ2H5 + HONO, HONO (+M) = ȮH + NO (+M) and
C2H6 + ȮH = Ċ2H5 + H2O). The accumulated active ȮH
radicals further promote the fuel oxidation and increase
the reactivity of ethane.
b) NO2 can react with stable intermediates ĊH3 and CH2O via
R3045: ĊH3 + NO2 <=> CH3Ȯ + NO and R3151: CH2O
+ NO2 <=> HĊO + HONO, to form active CH3Ȯ HĊO and
HONO, which readily release Ḣ atoms and ȮH radicals via
unimolecular decomposition. The reactivity is thus
enhanced further.
c) Stable HȮ2 radicals convert to active ȮH radicals via R2722
(NO + HȮ2 <=> NO2 + ȮH), and it results in the promotion
of chain branching process.
At low pressure and high temperature however, the addition of
NO2 only limited perturbs the consumption of ethane either during
ignition induction period or during main oxidation. No obvious
effect on the radical pool can be observed.
Acknowledgements
The authors would like to thank the support from the National
Natural Science Foundation of China (No. 91541115, 91441203).
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.fuel.2017.06.099.
References
[1] Held T, Marchese A, Dryer F. A semi-empirical reaction mechanism for n-
heptane oxidation and pyrolysis. Combust Sci Technol 1997;123(1–6):107–46.
[2] Zeppieri SP, Klotz SD, Dryer FL. Modeling concepts for larger carbon number
alkanes: a partially reduced skeletal mechanism for n-decane oxidation and
pyrolysis. Proc Conbust Insit 2000;28(2):1587–95.
[3] Westbrook CK. An analytical study of the shock tube ignition of mixtures of
methane and ethane. Combust Sci Technol 1979;20(1–2):5–17.
[4] Herzler J, Naumann C. Shock-tube study of the ignition of methane/ethane/
hydrogen mixtures with hydrogen contents from 0% to 100% at different
pressures. Proc Conbust Insit 2009;32(1):213–20.
[5] Lamoureux N, Paillard C-E. Natural gas ignition delay times behind reflected
shock waves: application to modelling and safety. Shock Waves 2003;13
(1):57–68.
[6] de Vries J, Hall JM, Simmons SL, Rickard MJ, Kalitan DM, Petersen EL. Ethane
ignition and oxidation behind reflected shock waves. Combust Flame 2007;150
(1):137–50.
[7] Egolfopoulos F, Zhu D, Law C. Experimental and numerical determination of
laminar flame speeds: mixtures of C2-hydrocarbons with oxygen and nitrogen.
In: Symposium (International) on Combustion. Elsevier; 1991. p. 471–8.
[8] Jomaas G, Zheng X, Zhu D, Law C. Experimental determination of counterflow
ignition temperatures and laminar flame speeds of C 2–C 3 hydrocarbons at
atmospheric and elevated pressures. Proc Conbust Insit 2005;30(1):193–200.
[9] Dagaut P, Cathonnet M, Boettner JC. Kinetics of ethane oxidation. Int J Chem
Kinet 1991;23(5):437–55.
[10] Tranter RS, Sivaramakrishnan R, Brezinsky K, Allendorf MD. High pressure,
high temperature shock tube studies of ethane pyrolysis and oxidation. PCCP
2002;4(11):2001–10.
[11] Smith GP, Golden DM, Frenklach M, Moriarty NW, Eiteneer B, Goldenberg M,
Bowman CT, Hanson RK, Song S, Gardiner Jr WC. GRI-Mech 3.0. URL: http://
www. me. berkeley. edu/gri_mech; 1999, 51. p.55.
[12] Wang H, You X, Joshi AV, Davis SG, Laskin A, Egolfopoulos F, Law CK. USC Mech
Version II. High-temperature combustion reaction model of H2/CO/C1-C4
compounds; 2007.
[13] Li Y, Zhou C-W, Somers KP, Zhang K, Curran HJ. The oxidation of 2-butene: a
high pressure ignition delay, kinetic modeling study and reactivity comparison
with isobutene and 1-butene. Proc Conbust Insit 2016.
[14] Deng F, Yang F, Zhang P, Pan Y, Bugler J, Curran HJ, et al. Towards a kinetic
understanding of the NOx promoting-effect on ignition of coalbed methane: a
case study of methane/nitrogen dioxide mixtures. Fuel 2016;181:188–98.
[15] Deng F, Yang F, Zhang P, Pan Y, Zhang Y, Huang Z. Ignition delay time and
chemical kinetic study of methane and nitrous oxide mixtures at high
temperatures. Energy Fuels 2016;30(2):1415–27.
[16] Hori M, Matsunaga N, Marinov N, William P, Charles W. An experimental and
kinetic calculation of the promotion effect of hydrocarbons on the NO-NO2
conversion in a flow reactor. In: Symposium (International) on
combustion. Elsevier; 1998. p. 389–96.
[17] Herzler J, Naumann C. Shock tube study of the influence of NOx on the ignition
delay times of natural gas at high pressure. Combust Sci Technol 2012;184
(10–11):1635–50.
[18] Mathieu O, Pemelton JM, Bourque G, Petersen EL. Shock-induced ignition of
methane sensitized by NO2 and N2O. Combust Flame 2015;162(8):3053–70.
[19] Gersen S, Mokhov A, Darmeveil J, Levinsky H, Glarborg P. Ignition-promoting
effect of NO2 on methane, ethane and methane/ethane mixtures in a rapid
compression machine. Proc Conbust Insit 2011;33(1):433–40.
[20] Rasmussen CL, Rasmussen AE, Glarborg P. Sensitizing effects of NOx on CH4
oxidation at high pressure. Combust Flame 2008;154(3):529–45.
[21] Sivaramakrishnan R, Brezinsky K, Dayma G, Dagaut P. High pressure effects on
the mutual sensitization of the oxidation of NO and CH4–C2H6 blends. PCCP
2007;9(31):4230–44.
[22] Faravelli T, Frassoldati A, Ranzi E. Kinetic modeling of the interactions between
NO and hydrocarbons in the oxidation of hydrocarbons at low temperatures.
Combust Flame 2003;132(1):188–207.
[23] Zhang Y, Huang Z, Wei L, Zhang J, Law CK. Experimental and modeling study
on ignition delays of lean mixtures of methane, hydrogen, oxygen, and argon
at elevated pressures. Combust Flame 2012;159(3):918–31.
[24] Kee RJ, Rupley FM, Miller JA. Chemkin-II: a Fortran chemical kinetics package
for the analysis of gas-phase chemical kinetics. Livermore, CA (USA): Sandia
National Labs; 1989.
[25] Lutz AE, Kee RJ, Miller JA. SENKIN: A FORTRAN program for predicting
homogeneous gas phase chemical kinetics with sensitivity analysis. Livermore,
CA (USA): Sandia National Labs; 1988.
[26] Chaos M, Dryer FL. Chemical-kinetic modeling of ignition delay:
considerations in interpreting shock tube data. Int J Chem Kinet 2010;42
(3):143–50.
[27] Zhou C-W, Li Y, O’Connor E, Somers KP, Thion S, Keesee C, Mathieu O, et al. A
comprehensive experimental and modeling study of isobutene oxidation.
Combust Flame 2016;167:353–79.
[28] Burke U, Metcalfe WK, Burke SM, Heufer KA, Dagaut P, Curran HJ. A detailed
chemical kinetic modeling, ignition delay time and jet-stirred reactor study of
methanol oxidation. Combust Flame 2016;165:125–36.
[29] Burke SM, Burke U, Mathieu O, Osorio I, Keesee C. An experimental and
modeling study of propene oxidation. Part 2: Ignition delay time and flame
speed measurements. Combust Flame 2015;162(2):296–314.
[30] Burke SM, Metcalfe WK, Herbinet O, Battin-Leclerc F, Haas FM, Santner J, Dryer
FL, et al. An experimental and modeling study of propene oxidation. Part 1:
Speciation measurements in jet-stirred and flow reactors. Combust Flame
2014;164(11):2765–84.
[31] Metcalfe WK, Burke SM, Ahmed SS, Curran HJ. A hierarchical and comparative
kinetic modeling study of C1–C2 hydrocarbon and oxygenated fuels. Int J
Chem Kinet 2013;45(10):638–75.
[32] Kéromnès A, Metcafle WK, Heufer KA, Donohoe N, Das AK, Sung CJ. An
experimental and detailed chemical kinetic modelling study of hydrogen and
syngas mixtures at elevated pressures. Combust Flame 2013;160:995–1011.
[33] Giménez-López J, Alzueta M, Rasmussen C, Marshall P, Glarborg P. High
pressure oxidation of C 2 H 4/NO mixtures. Proc Conbust Insit 2011;33
(1):449–57.
[34] YingJia Z, Mathieu O, Petersen EL, Bourque G, Curran HJ. Assessing the
predictions of a NOx kinetic mechanism on recent hydrogen and syngas
experimental data. Combust Flame 2017;182:122–41.
[35] Zhang K, Zhang L, Xie M, Ye L, Zhang F, Glarborg P, et al. An experimental and
kinetic modeling study of premixed nitroethane flames at low pressure. Proc
Conbust Insit 2013;34(1):617–24.
 F. Deng et al. / Fuel 207 (2017) 389–401
[36] Zhang K, Li Y, Yuan T, Cai J, Glarborg P, Qi F. An experimental and kinetic
modeling study of premixed nitromethane flames at low pressure. Proc
Conbust Insit 2011;33(1):407–14.
[37] Chai J, Goldsmith CF. Rate coefficients for fuel+NO2: Predictive kinetics for
HONO and HNO2 formation. Insit: Proc. Conbust; 2016.
[38] Zhang J, Wei L, Man X, Jiang X, Zhang Y, Hu E, et al. Experimental and modeling
study of n-butanol oxidation at high temperature. Energy Fuels 2012;26
(6):3368–80.
[39] Frassoldati A, Faravelli T, Ranzi E. Kinetic modeling of the interactions between
NO and hydrocarbons at high temperature. Combust Flame 2003;135
(1):97–112.
[40] Dayma G, Dagaut P. Effects of air contamination on the combustion of
hydrogen—effect of NO and NO2 addition on hydrogen ignition and oxidation
kinetics. Combust Sci Technol 2006;178(10–11):1999–2024.
401
[41] Hu E, Chen Y, Zhang Z, Li X, Cheng Y, Huang Z. Experimental study on ethane
ignition delay times and evaluation of chemical kinetic models. Energy Fuels
2015;29(7):4557–66.
[42] Mével R, Shepherd J. Ignition delay-time behind reflected shock waves of small
hydrocarbons–nitrous oxide (–oxygen) mixtures. Shock Waves 2015;25
(3):217–29.
[43] Konnov AA, Zhu JN, Bromly JH, Zhang D-K. The effect of NO and NO 2 on the
partial oxidation of methane: experiments and modeling. Proc Conbust Insit
2005;30(1):1093–100.
[44] Dagaut P, Mathieu O, Nicolle A, Dayma G. Experimental study and detailed
kinetic modeling of the mutual sensitization of the oxidation of nitric oxide,
ethylene, and ethane. Combust Sci Technol 2005;177(9):1767–91.