Our reference: HE 10552 P-authorquery-v9

AUTHOR QUERY FORM

Journal: HE

Article Number: 10552

Please e-mail or fax your responses and any corrections to:

E-mail: corrections.essd@elsevier.tnq.co.in

Fax: +31 2048 52789

Dear Author,

Please check your proof carefully and mark all corrections at the appropriate place in the proof (e.g., by using on-screen

annotation in the PDF file) or compile them in a separate list. Note: if you opt to annotate the file with software other than

Adobe Reader then please also highlight the appropriate place in the PDF file. To ensure fast publication of your paper please

return your corrections within 48 hours.

For correction or revision of any artwork, please consult http://www.elsevier.com/artworkinstructions.

Any queries or remarks that have arisen during the processing of your manuscript are listed below and highlighted by flags in

the proof.

Location

in article

Query / Remark: Click on the Q link to find the query’s location in text

Please insert your reply or correction at the corresponding line in the proof

Q1 Please confirm that given names and surnames have been identified correctly.

Please check this box if you have no

corrections to make to the PDF file ,

Thank you for your assistance.

Highlights

< Three ignition regimes are identified according to hydrogen fraction. < Simulated ignition delays using four models were

compared to experimental data. < At high temperature, ignition delay mainly is governed by chain branching reaction. < At

middle-low temperature, HO2 and H2O2 chemistries affect ignition delay.

Available online at www.sciencedirect.com

journal homepage: www.elsevier .com/locate/he

123456789

1011121314151617181920

2122232425262728293031323334353637383940

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y x x x ( 2 0 1 2 ) 1

HE10552_grabs ■ 27 September 2012 ■ 1/1

Please cite this article in press as: ZhangY, et al., Experimental andmodeling study on auto-ignition characteristics ofmethane/hydrogen blends under engine relevant pressure, International Journal of Hydrogen Energy (2012), http://dx.doi.org/10.1016/j.ijhydene.2012.09.056

0360-3199/$ e see front matter Copyright ª 2012, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.ijhydene.2012.09.056

Experimental and modeling study on auto-ignitioncharacteristics of methane/hydrogen blends under enginerelevant pressure

Q1 Yingjia Zhang, Xue Jiang, Liangjie Wei, Jiaxiang Zhang, Chenglong Tang, Zuohua Huang*

State Key Laboratory of Multiphase Flow in Power Engineering, Xi’an Jiaotong University, Xi’an 710049, China

a r t i c l e i n f o

Article history:

Received 24 April 2012

Received in revised form

25 August 2012

Accepted 9 September 2012

Available online xxx

Keywords:

Shock tube

Methane

Hydrogen

Ignition delay

Sensitivity analysis

a b s t r a c t

Auto-ignition characteristics of methane/hydrogen mixtures with hydrogen mole fraction

varying from 0 to 100% were experimentally studied using a shock tube. Test pressure is

kept 1.8 MPa and temperatures behind reflected shock waves are in the range of 900

e1750 K and equivalence ratios from 0.5 to 2.0. Three ignition regimes are identified

according to hydrogen fraction. They are, methane chemistry dominating ignition ðXH2 �40%Þ, combined chemistry of methane and hydrogen dominating ignition ðXH2 ¼ 60%Þ, andhydrogen chemistry dominating ignition ðXH2 � 80%Þ. Simulated ignition delays using four

models including USC Mech 2.0, GRI Mech 3.0, UBC Mech 2.1 and NUI Galway Mech were

compared to the experimental data. Results show that USC Mech 2.0 gives the best

prediction on ignition delays and it was selected to conduct sensitivity analysis for three

typical methane/hydrogen mixtures at different temperatures. The results suggest that at

high temperature, ignition delay mainly is governed by chain branching reaction

H þ O2 5 OH þ O, and thus increasing equivalence ratio inhibits ignition of methane/

hydrogen mixtures. At middle-low temperature, contribution of equivalence ratio on

ignition delay of methane/hydrogen mixtures is mainly due to chemistries of HO2 and H2O2

radicals.

Copyright ª 2012, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights

reserved.

1. Introduction

With increased demand on energy and further strengthened

vehicle emission regulation adoption, the mixtures of

methane, the primary composition of natural gas, and

hydrogenwill become one of most promising alternative fuels

due to large reserve in natural gas, and clean combustion

through hydrogen enrichment [1,2]. Many researches have

been conducted in the practical combustion facilities, like IC

engines [3,4] and aerospace [5]. Meanwhile, the fundamental

combustion researches were made, like flame speed

measurements [6e8] and species concentration profile [9,10].

The ratios of methane and hydrogen sometimes vary due

to different supplying and use in different combustion facili-

ties. Different fractions of hydrogen can influence the ignition

and combustion chemistry of methane-hydrogen fuel blend,

and influence the safety and performance in the particular

combustion device. Thus, understanding the ignition kinetics

of methane/hydrogen mixtures will optimize the operation of

the combustion device and avoid the possible faults occurring,

such as knocking, flashback and unstable combustion.

* Corresponding author. Tel.: þ86 29 82665075; fax: þ86 29 82668789.E-mail address: zhhuang@mail.xjtu.edu.cn (Z. Huang).

Available online at www.sciencedirect.com

journal homepage: www.elsevier .com/locate/he

123456789

1011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465

66676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130

i n t e rn a t i o n a l j o u rn a l o f h y d r o g e n en e r g y x x x ( 2 0 1 2 ) 1e9

HE10552_proof ■ 27 September 2012 ■ 1/9

Please cite this article in press as: ZhangY, et al., Experimental andmodeling study on auto-ignition characteristics ofmethane/hydrogen blends under engine relevant pressure, International Journal of Hydrogen Energy (2012), http://dx.doi.org/10.1016/j.ijhydene.2012.09.056

0360-3199/$ e see front matter Copyright ª 2012, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.ijhydene.2012.09.056

As the fundamental combustion data are hard to obtain in

the practical combustion device, thus, these fundamental

data are usually measured using the standard facilities, such

as shock tube. Lifshitz et al. [11] first experimentally investi-

gated the ignition of methane/hydrogen mixtures using

a shock tube. Cheng and Oppenheim [12] investigated ignition

characteristics of methane/hydrogen mixtures. Recently,

some scholars also reported the ignition characteristics of

methane/hydrogen mixtures [13e19].

In reviewing the literature, it is clear that only limited

studies are available for the ignition delays of methane/

hydrogen at engine relevant pressure, and the fractions of

hydrogen in fuel are also limited in these studies. Meanwhile,

providing sufficient ignition delays can develop and modify

the mechanisms of methane/hydrogen mixtures. For this

purpose, the detailed ignition delays of methane/hydrogen

mixtures were measured in a high-pressure shock tube at

engine relevant pressure, and the effects of equivalence ratio

on ignition of methane/hydrogen mixtures were analyzed. In

addition, normalized sensitivity analysis was made using the

USC Mech 2.0 mechanism [20] to identify key elementary

reactions in the ignition chemistry of methane/hydrogen

mixtures.

2. Experimental setup and procedure

2.1. Experimental setup

Experiments were made in a high-pressure stainless shock

tubewitha115mminnerdiameter. This shock tube facilityhas

been fully described in the previous studies [18,19]. A double

diaphragm machine separates the shock tube into a 4-m long

driver section and a 5.3-m long driven section, as shown in

Fig. 1. In order to measure longer ignition delays at relatively

low temperatures, the tailored interface method was used

through tuning the thermodynamics properties of the driver

gases using different mixtures of helium and nitrogen. Four

fast-response piezoelectric pressure transducers are installed

at fixed intervals along the driven section and three time-

interval counters are used to measure incident shock velocity

with typical attenuation value from 1.8 to 3.2%/m. The test

temperature behind the reflected shock wave is determined

fromthe incident shockvelocitybyusing chemical equilibrium

software GASEQ [21]. Uncertainty of reflected shock tempera-

ture is calculated using a standard error analysis procedure

[22]. In this study, at the reflected shock temperature of 1450 K

and950K, theestimatederrorsare approximately20Kand10K

respectively. The test mixtures were prepared in a 128 L

stainless steel tank according to Dalton’s law of partial pres-

sure andwere kept over 12 h to ensure a sufficientmixing. The

purities of methane, hydrogen, oxygen, nitrogen and helium

are 99.995%, 99.999%, 99.995%, 99.995%, and 99.999%, respec-

tively. When the driven section is evacuated to an ultimate

pressure below 1.0� 10�7 MPa, the test mixtures are filled into

it, and the double diaphragmsare brokenwithin a fewminutes

afterfilling.ThermodynamicdatabaseofUSCMech2.0 [20]was

used in the study. Vibrational relaxation is neglected due to

small amounts of diatomic and polyatomic molecules in this

work.

The reflected shock pressure was traced by a piezoelectric

pressure transducer with acceleration compensation located

at 20 mm from the end-wall of shock tube. Chem-

iluminescence emission from OH* radicals near 307 nm was

detected by a photomultiplier with a narrow band pass filter.

Ignition of the mixtures was monitored by the reflected shock

pressure and OH* emission. The definition of ignition delay is

consistent with Ref. [18]. Typical pressure rise rate (dp/

dt ¼ 4%/ms) from boundary layer effect (i.e. BL effect) is

observed in this study.

2.2. Numerical simulation

Calculated ignition delays and sensitivity analysis of the

methane/hydrogenmixtures were made using the CHEMKINⅡ

[23] program with extensively incorporating the SENKIN [24]

package. In general, for high-temperature (s < 1.5 ms), the

ideal assumption of constant volume (constantU, V) and zero-

Fig. 1 e Schematic of the high-pressure shock tube.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y x x x ( 2 0 1 2 ) 1e92

131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195

196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260

HE10552_proof ■ 27 September 2012 ■ 2/9

Please cite this article in press as: Zhang Y, et al., Experimental andmodeling study on auto-ignition characteristics ofmethane/hydrogen blends under engine relevant pressure, International Journal of Hydrogen Energy (2012), http://dx.doi.org/10.1016/j.ijhydene.2012.09.056

dimensional model (ZOAM) is reasonable for calculating

ignition delay. For the middle-low temperature (s > 1.5 ms),

however the BL effect becomes important and the dp/dt needs

to be considered in simulation. Therefore, an improved

assumption of constant U, V is used to calculate the ignition

delay of the methane/hydrogen mixtures. Fig. 2 gives

a comparison on the ignition delays between the calculated by

using both methods (i.e. ideal constant U, V and the improved

constant U, V) and measured. Whether methane or hydrogen,

both assumptions are highly consistent with the current data

when s < 1.5 ms. However, an obvious discrepancy is pre-

sented when s > 1.5 ms. Compared to improved constant U, V,

the ideal constant U, V over-predicts the ignition delay in

a factor of 3 for methane at T ¼ 1250 K and a factor of 3 for

hydrogen at T ¼ 1000 K. The related research was reported by

Pang et al. [25].

3. Experimental results and discussion

18 kinds of methane/hydrogen mixtures were tested and

detailed compositions are given in Table 1. The measured

ignition delays of methane/hydrogen are given in Figs. 3e5. de Vries and Petersen [13] and Huang et al. [14] suggested that

certain hydrogen addition ðXH2 < 40%Þ to methane did not

change dominant kinetic of the methane oxidation. The

ignition characteristics of the methane/hydrogen mixtures

were classified into three ignition regimes based on hydrogen

fractions proposed in Ref. [18], i.e. methane chemistry domi-

nating ignition (MCDI), combined chemistry of methane and

hydrogen dominating ignition (CCMHDI) and hydrogen

chemistry dominating ignition (HCDI). Furthermore, two

ignition temperature regions were identified from changing in

global activation energy using a dotted separating line.

3.1. Reaction system in MCDI mixtures

For XH2 � 40% in fuel, the effects from equivalence ratio on

ignition delays are shown in Fig. 3. At high-temperature

region, the ignition delays increase with the increase in

equivalence ratio. Here, ignition kinetic depends on oxygen

concentration in fuel blends. Oxygen concentration increases

with the decrease in equivalence ratio, which increases the

reaction rate of chain branching reaction H þ O2 5 OH þ O

which plays a dominant role at high temperature and

promotes the ignition. Note that the ignition delays increase

slightly for the leanmixtures and increase significantly for the

rich mixture with the increase in equivalence ratio when

XH2 ¼ 40%. The non-linear relationship between logarithmic

of ignition delay and equivalence ratio appears and becomes

obvious with increasing hydrogen fraction.

At middle-low temperature region, the ignition delays still

increase with increasing the equivalence ratio. But the reason

is different to that at high-temperature region. The ignition

kinetics of methane/hydrogen mixtures depends largely on

fuel concentration rather than oxygen concentration. Fuel

concentration increases with the increase in equivalence

ratio, which increases production rate of CH3 and promotes

reaction rate of the chain termination reaction

CH3 þ CH3(þM) 5 C2H6(þM), and then increases the ignition

delays.

Fig. 2 e Comparison of ignition delay calculated by using

both ideal U, V and improved U, V to experiments. (a)

Ignition delay for methane, (b) Ignition delay for hydrogen.

Table 1 e Composition of the test mixtures in the study( p [ 1.8 MPa for all mixtures).

Fuel blends XCH4=ð%Þ XH2=ð%Þ XO2=ð%Þ XAr/(%) f

100%CH4 0.998 0 3.990 95.012 0.5

1.900 0 3.801 94.299 1.0

3.471 0 3.471 93.058 2.0

80%CH4/20%H2 0.931 0.233 3.956 94.880 0.5

1.759 0.440 3.738 94.063 1.0

3.170 0.792 3.368 92.670 2.0

60%CH4/40%H2 0.837 0.558 3.907 94.698 0.5

1.565 1.043 3.652 93.740 1.0

2.769 1.846 3.231 92.154 2.0

40%CH4/60%H2 0.697 1.046 3.834 94.423 0.5

1.282 1.924 3.527 93.267 1.0

2.211 3.316 3.309 91.434 2.0

20%CH4/80%H2 0.464 1.856 3.713 93.967 0.5

0.832 3.327 3.327 92.515 1.0

1.377 5.508 2.754 90.361 2.0

0%CH4 0 3.471 3.471 93.058 0.5

0 5.915 2.958 91.127 1.0

0 9.130 2.283 88.587 2.0

i n t e rn a t i o n a l j o u rn a l o f h y d r o g e n en e r g y x x x ( 2 0 1 2 ) 1e9 3

261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325

326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390

HE10552_proof ■ 27 September 2012 ■ 3/9

Please cite this article in press as: ZhangY, et al., Experimental andmodeling study on auto-ignition characteristics ofmethane/hydrogen blends under engine relevant pressure, International Journal of Hydrogen Energy (2012), http://dx.doi.org/10.1016/j.ijhydene.2012.09.056

3.2. Reaction system in CCMHDI mixture

For the 40%CH4/60%H2 mixture, the effect of equivalence ratio

on the ignition delays is shown in Fig. 4. At high-temperature

region, the ignition delays of the mixture increase with the

increase in equivalence ratio like that of MCDI mixture. The

non-linear relationship between logarithmic of the ignition

delay and equivalence ratio becomes more obvious.

At middle-low temperature region, the ignition delays are

insensitive to the variation of equivalence ratio. Here, it is

Fig. 3 e Effects of equivalence ratio on ignition for methane

dominating ignition. (a) 100%CH4, (b) 80%CH4/20%H2, (c)

60%CH4/40H2.

Fig. 4 e Effects of equivalence ratio on ignition delay of

hydrogen and methane dominating mixture.

Fig. 5 e Effects of equivalence ratio on ignition for hydrogen

dominating ignition. (a) 20%CH4/80%H2, (b) 100%H2.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y x x x ( 2 0 1 2 ) 1e94

391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455

456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520

HE10552_proof ■ 27 September 2012 ■ 4/9

Please cite this article in press as: Zhang Y, et al., Experimental andmodeling study on auto-ignition characteristics ofmethane/hydrogen blends under engine relevant pressure, International Journal of Hydrogen Energy (2012), http://dx.doi.org/10.1016/j.ijhydene.2012.09.056

hard to know whether a single chemistry is dominant or both

chemistries are the balance competition in the ignition kinetic

of CCMHDI mixture. Furthermore, the chemistries of HO2 and

H2O2 become very important for ignition of the methane/

hydrogen mixture at low temperature. A detailed kinetic

analysis will be discussed in Section 4.2.2.

3.3. Reaction system in HCDI mixtures

Fig. 5 shows influence of equivalence ratio on the ignition

delays for hydrogen chemistry dominating. For the 20%CH4/

80%H2 mixture as shown in Fig. 5(a), at high-temperature

region, the results show that a negligible influence from

equivalence ratio is presented at T > 1350 K, and the shortest

ignition delay is given for f¼ 0.5 and the longest ignition delay

is appeared for f ¼ 2.0 at 1220 K < T < 1350 K, and the ignition

delays are close to each other at 1160 K < T < 1220 K. This

reason is the discrepancy from inhibited effect of methane on

ignition at different conditions. At middle-low temperature

region, the ignition delay is insensitive to equivalence ratio.

For pure hydrogen, at high-temperature region, a complex

dependence of ignition delay on equivalence ratio is pre-

sented in Fig. 5(b). The longest ignition delay is given at

f ¼ 0.5, and the shortest one at f ¼ 1.0. At f � 1.0, the

hydrogen concentration increases with the increase in

equivalence ratio which increases the H and OH radicals

through the reaction O þ H2 5 H þ OH, and then promotes

the rate of reaction H þ O2 5 OH þ O. Furthermore, enthalpy

of the mixture increases with increasing hydrogen concen-

tration which increases reaction system temperature. Thus,

the rates of above two endothermic reactions are further

promoted and decrease the ignition delays. At f � 1.0, the

oxygen concentration decreases with the increase in equiv-

alence ratio which decreases the rate of reaction

H þ O2 5 OH þ O, and decreases rate of total reaction and

ignition delays. At middle-low temperature region, the effect

of equivalence ratio on ignition delays is consistent to that at

high-temperature region, and the stoichiometric mixture

gives the shortest ignition delay. Here, the chain termination

reaction H þ O2(þM) 5 HO2(þM) dominates the ignition

kinetic instead of the reaction H þ O2 5 OH þ O. At f � 1.0,

the oxygen concentration decreases with the increase in

equivalence ratio which decreases the rate of reaction

H þ O2(þM) 5 HO2(þM), and then promotes rate of total

reaction and reduces ignition delays. At f � 1.0, the rates of

reaction O þ H2 5 H þ OH and H þ O2(þM) 5 HO2(þM) are

increased as increasing hydrogen concentration and

decreasing oxygen concentration with the increase in equiv-

alence ratio. This inhibits rate of total reaction and increases

the ignition delay.

4. Numerical predictions and comparisonwith experiments

The calculated ignition delays of the methane/hydrogen

mixtures have large discrepancy using different kinetic

models due to different species and rates of elementary reac-

tions in these models. Therefore, the comparison on ignition

delay predicted among different models is necessary for

selecting a reasonable kinetic model in predicting the ignition

delays of the methane/hydrogen mixtures and making sensi-

tivity analysis under the experimental conditions.

4.1. Mechanism selection

Gersen et al. [15] suggested that if a model could well predict

the ignition delays for pure hydrogen and pure methane, this

mechanism could also give reasonable prediction in ignition

delays for methane/hydrogen fuel blends. The view had been

verified by Huang et al. [14] and Zhang et al. [18]. Comparisons

of the calculated ignition delays by using four mechanisms

USCMech 2.0 [20], GRIMech 3.0 [26], UBCMech 2.1 [14] andNUI

Galway Mech [27] with measured ignition delays of methane

and hydrogen are shown in Figs. 6 and 7. For pure methane

mixture, the four models can well predict the ignition delays.

For pure hydrogen mixture, however the models can well

predict the “S” structure except the UBC Mech 2.1 [12]. The

result is not surprising because the hydrogen submodel in

UBC Mech 2.1 [14] is mainly based on GRI Mech 1.2 [28]. It is

well known that the hydrogen submodel in GRI Mech 1.2 [28]

has been only validated by the experimental data at low

pressure and high temperature, but not is optimized by

those data at high pressure and middle-low temperature.

Huang et al. [14] modified the rate coefficients of several

reactions key to hydrogen ignition including the reactions

H þ O2 þ M ¼ HO2 þ M, OH þ H2 ¼ O þ H2O and

H þ O2 þ N2 ¼ HO2 þ N2, nevertheless the equally important

elementary reactions such as HO2 þ HO2 5 O2 þ H2O2 and

OH þ HO2 5 H2O þ O2 are not improved in UBC Mech 2.1 [14].

It is noteworthy that all kinetic models over-predict the igni-

tion delays at middle-low temperature (T < 1000 K), especially

for stoichiometric mixture. This suggests that the current

hydrogen model needs further to be modified at higher pres-

sure and middle-low temperature.

Through the above comparative analyses, the USC Mech

2.0 [20] was selected to calculate the ignition delays of the

methane/hydrogen mixtures and make normalized sensi-

tivity analysis. This model was developed on the basis of GRI

Fig. 6 e Measured and calculated ignition delays for

methane using different kinetic models.

i n t e rn a t i o n a l j o u rn a l o f h y d r o g e n en e r g y x x x ( 2 0 1 2 ) 1e9 5

521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585

586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627628629630631632633634635636637638639640641642643644645646647648649650

HE10552_proof ■ 27 September 2012 ■ 5/9

Please cite this article in press as: ZhangY, et al., Experimental andmodeling study on auto-ignition characteristics ofmethane/hydrogen blends under engine relevant pressure, International Journal of Hydrogen Energy (2012), http://dx.doi.org/10.1016/j.ijhydene.2012.09.056

Mech 1.2 [28] and GRI Mech 3.0 [26], including the optimized

reaction model of H2/CO combustion of Davis et al. [29], the

detailed C2 fuel combustionmodel of Wang and Laskin [30] on

the basis of C2H4 and C2H2 chemistries, the detailed C3 fuel

combustionmodel of Davis et al. [31] on the basis of chemistry

of C3H6 decomposition and oxidation, and the C4 combustion

model of Laskin et al. [32] on the basis of 1, 3-butadiene

oxidation chemistry. This detailed kinetic model includes

111 species and 784 reactions, and was validated by lots of

experimental data, like ignition delays, species concentration

profiles and laminar flame speeds. Furthermore, the rates of

following reactions

HO2 þ HO2 5 O2 þ H2O2 (R18eR19)

OH þ HO2 5 H2O þ O2 (R20eR24)

CO þ OH 5 CO2 þ H (R31eR32)

CO þ HO2 5 CO2 þ OH (R34)

weremodified in their studies, and rates of several C1 and C2

reactionswere re-evaluated. The oxidationmodels of benzene

and toluene are added to the USC Mech 2.0 [20].

4.2. Sensitivity analysis

Sensitivity analysis is used to identify key promotion and/or

inhibition reactions in the ignition process of combustible

mixture, and it will help to further understand the ignition

chemical kinetic. The normalized sensitivity analysis was

conducted for three typical methane/hydrogen mixtures to

study the effects of equivalence ratio on ignition delay at

different temperatures (1400 K and 930 K) and 1.8 MPa using

the USCMech 2.0 [20] in this study. The normalized sensitivity

coefficient is defined as follows,

S ¼ sð2:0kiÞ � sð0:5kiÞ1:5sðkiÞ (1)

where s is ignition delay and ki is species rate coefficient.

Negative value of sensitivity coefficient indicates a promotion

effect on the total reaction rate while positive value of sensi-

tivity coefficient indicates an inhibition effect on the total

reaction rate.

4.2.1. Reaction system in MCDI mixturesThe sensitivity analysis of the 80%CH4/20%H2 mixture was

made, as shown in Fig. 8. At T ¼ 1400 K, the main ignition

promotion reactions R1, R93 and R96 are identified. Moreover,

the key ignition inhibition reactions are the chain termination

reaction R104 and the consumption reactions of methane

R123, R124 and R125. The sensitivity coefficients of ignition

promotion reactions R1 decreases with increase in equiva-

lence ratio, thus the chain branching efficiency is inhibited

through reaction R1, and decreases the accelerated rate of

total reaction. The sensitivity coefficients of reactions R124

and R125 decrease with increase in equivalence ratio, thus the

inhibition effect on rate of total reaction is weakened.

Furthermore, the reactions R93 and R96 which are the reac-

tions in production of CH3O radical become importantly at

high temperature and promote the ignition since CH3 radical

is consumed and free radicals O and OH are produced through

Fig. 7 e Measured and calculated ignition delays for

hydrogen using different kinetic models. (a) f [ 0.5, (b)

f [ 1.0, (c) f [ 2.0.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y x x x ( 2 0 1 2 ) 1e96

651652653654655656657658659660661662663664665666667668669670671672673674675676677678679680681682683684685686687688689690691692693694695696697698699700701702703704705706707708709710711712713714715

716717718719720721722723724725726727728729730731732733734735736737738739740741742743744745746747748749750751752753754755756757758759760761762763764765766767768769770771772773774775776777778779780

HE10552_proof ■ 27 September 2012 ■ 6/9

Please cite this article in press as: Zhang Y, et al., Experimental andmodeling study on auto-ignition characteristics ofmethane/hydrogen blends under engine relevant pressure, International Journal of Hydrogen Energy (2012), http://dx.doi.org/10.1016/j.ijhydene.2012.09.056

R93 and R96 respectively. The sensitivity coefficients of R93

and R96 decreasewith increase in equivalence ratio, leading to

the weakening on accelerated ignition tendency. It is note-

worthy that both reactions R104 which is recombination

reaction of two CH3 radicals and R123 which is H-scavenging

reaction are insensitive to the variation of equivalence ratio.

Thus, overall the ignition of the mixtures is inhibited with

increase in equivalence ratio and this is consistent to the

experimental results in Fig. 3(b).

At T¼ 930 K, the chemistries of HO2 and H2O2 becomemore

important such as reactions R14, R19, R22, R86, R95 and R96. It

is clearly identified that R96 plays the most important ignition

promotion reaction instead of R1. The sensitivity coefficient of

R96 decreases with increase in equivalence ratio, and the

ignition promotion effect is weakened. Furthermore, the

sensitivity coefficient of R104 increases and that of R86 and

R94 decrease with increase in equivalence ratio, as the result

the ignition inhibition effect is promoted. Therefore, for low

temperature, the ignition is inhibited and ignition delay is

increased with increase in equivalence ratio.

4.2.2. Reaction system in CCMHDI mixtureFig. 9 gives the sensitivity analysis of the 60%CH4/40%H2

mixture. At T ¼ 1400 K, the result indicates that almost all of

the key elementary reactions are frommethane chemistry for

this mixture. Therefore, the dependence of ignition delay on

equivalence ratio is similar to the reaction system inmethane

chemistry dominating ignition, and related analysis and

discussion are not included in this part.

At T ¼ 930 K, however, the hydrogen chemistry becomes

also very important except the methane chemistry. The

sensitivity coefficient of the reaction R14 increases with

increase in equivalence ratio, and the H2O2 radical produced

by R14 will provide the contributions toward two different

aspects. One is that the rates of ignition promotion reactions

R97 and R25 are promoted and the sensitivity coefficient

increases with increase in equivalence ratio. Thus the ignition

promotion effect is enhanced. The other is that the rate of

ignition inhibition reaction R19 is also promoted by HO2

radical produced through R25 and the sensitivity coefficient

Fig. 8 e Normalized sensitivity analysis of ignition delays

for the 80%CH4/20%H2 mixture using USC Mech 2.0. (a)

T [ 1400 K, (b) T [ 930 K.

Fig. 9 e Normalized sensitivity analysis of ignition delays

for 40%CH4/60%H2 mixture using USC Mech 2.0. (a)

T [ 1400 K, (b) T [ 930 K.

i n t e rn a t i o n a l j o u rn a l o f h y d r o g e n en e r g y x x x ( 2 0 1 2 ) 1e9 7

781782783784785786787788789790791792793794795796797798799800801802803804805806807808809810811812813814815816817818819820821822823824825826827828829830831832833834835836837838839840841842843844845

846847848849850851852853854855856857858859860861862863864865866867868869870871872873874875876877878879880881882883884885886887888889890891892893894895896897898899900901902903904905906907908909910

HE10552_proof ■ 27 September 2012 ■ 7/9

Please cite this article in press as: ZhangY, et al., Experimental andmodeling study on auto-ignition characteristics ofmethane/hydrogen blends under engine relevant pressure, International Journal of Hydrogen Energy (2012), http://dx.doi.org/10.1016/j.ijhydene.2012.09.056

increases with increase in equivalence ratio. Thus the ignition

promotion effect is weakened. Furthermore, it is identified

that the reaction R96 competes the CH3 radical to that of R97,

and these two reactions result in an opposite influence on

ignition delay with variation of equivalence ratio. Thus, the

ignition delay is insensitive to equivalence ratio at low

temperature region for the 60%CH4/40%H2 mixture, and this is

consistent to the experimental results in Fig. 4.

4.2.3. Reaction system in HCDI mixturesThe sensitivity analysis of the 20%CH4/80%H2 mixture was

made to study the effects of equivalence ratio on ignition

delay, as shown in Fig. 10. At T ¼ 1400 K, the reactions R2 and

R3 become more important in addition to reaction R1. The

sensitivity coefficients of the reactions R2 and R3 decrease

with increase in equivalence ratio, and the ignition promotion

is weakened. Moreover, the sensitivity coefficient of H-scav-

enging reaction R88 increases with increase in equivalence

ratio. Thus the rate of total reaction is decreased and ignition

delay is increased. It is noteworthy that the maximum values

of sensitivity coefficients of competition reactions R1 and R123

occur at f ¼ 1.0, which suggests the strongest inhibition effect

on ignition at the stoichiometric equivalence ratio. It is

because of these complex dependencies of the key elementary

reactions to equivalence ratio, a complex effect of equivalence

ratio on ignition delays is demonstrated in Fig. 5(a).

At T¼ 930 K, the chemistries of HO2 and H2O2 dominate the

ignition of fuel mixture. The chain termination reactions R18,

R19 and R22 become key ignition inhibition reactions instead

of the consumption reactions of methane R123, R124 and

R125. Analysis indicates that the sensitivity coefficients of

reactions R14 and R25 increase with increase in equivalence

ratio, thus the ignition promotion effect is enhanced through

these two reactions. However, the sensitivity coefficient of

reaction R1 decreases with increase in equivalence ratio,

which weakens the ignition promotion effect. It is identified

that R1 is main competitor to R25 for H radical. If reaction R1

dominates the ignition chemistry, the ignition delays will

increase with increase in equivalence ratio, and this case is

more likely appearing under the leanmixture condition due to

sufficient oxygen molecular. However, if the reaction R25

dominates the ignition chemistry, the ignition delays will

decrease with increase in equivalence ratio and this case is

more likely appearing under the richmixture condition. Thus,

the insensitive effect of equivalence ratio on ignition as

shown in Fig. 5(a) may be attributed to the competition

between R1 and R25. Furthermore, the sensitivity coefficient

of the ignition promotion reaction R97 increases with

increasing the equivalence ratio.

It is noted that the ignition characteristics of HCDI

mixtures at middle-low temperature are similar to those of

CCMHDI mixtures, although ignition mechanisms are

different as discussed by above.

5. Conclusions

Experimental and modeling studies on ignition delays of the

methane/hydrogen mixtures with various hydrogen volu-

metric contents (0%, 20%, 40%, 60%, 80% and 100%) were

conducted at engine relevant pressure using a shock tube.

Main conclusions are summarized as follows:

1. For MCDI mixtures ðXH2 � 40%Þ, at high temperature, the

ignition promotion is weakened with increasing equiva-

lence ratio by reaction R1 because ignition kinetic mainly

depends on oxygen concentration. At middle-low temper-

ature, the ignition inhibition is enhanced with increasing

equivalence ratio by the chemistries of CH3 and HO2

because ignition kinetic more depends on fuel concentra-

tion. Therefore, increasing equivalence ratio reduces the

ignition delay in this study. The non-linear relationship

between logarithmic of ignition delay and equivalence ratio

becomes increasingly apparent with increasing hydrogen

fraction.

2. For CCMHDI mixture ðXH2 ¼ 60%Þ, at high temperature, the

dependence of ignition delay on equivalence ratio is similar

to that of methane chemistry dominating reaction system.

At middle-low temperature, the ignition delays are insen-

sitive to the variation of equivalence ratio.When increasing

Fig. 10 e Normalized sensitivity analysis of ignition delays

for 20%CH4/80%H2 mixture using USC Mech 2.0. (a)

T [ 1400 K, (b) T [ 930 K.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y x x x ( 2 0 1 2 ) 1e98

911912913914915916917918919920921922923924925926927928929930931932933934935936937938939940941942943944945946947948949950951952953954955956957958959960961962963964965966967968969970971972973974975

976977978979980981982983984985986987988989990991992993994995996997998999

10001001100210031004100510061007100810091010101110121013101410151016101710181019102010211022102310241025102610271028102910301031103210331034103510361037103810391040

HE10552_proof ■ 27 September 2012 ■ 8/9

Please cite this article in press as: Zhang Y, et al., Experimental andmodeling study on auto-ignition characteristics ofmethane/hydrogen blends under engine relevant pressure, International Journal of Hydrogen Energy (2012), http://dx.doi.org/10.1016/j.ijhydene.2012.09.056

the equivalence ratio, the H2O2 produced through reaction

R14 promotes the rates of ignition promotion reaction R97

and ignition inhibition reaction R19, and this leads to

balanced influence of equivalence ratio on ignition delay.

3. For HCDI mixtures ðXH2 � 80%Þ, ignition delays give

a complex dependence on equivalence ratio. At high

temperature, the maximum values of the sensitivity coef-

ficient of the key elementary reactions appear at stoichio-

metric ratio, and this leads to the shortest ignition delay at

f ¼ 1.0. At middle-low temperature, the hydrogen chem-

istry becomes more obvious, and the effect of equivalence

ratio on ignition delay ismainly due to contribution of the H

competition reactions (R1 and R25) and CH3 competition

reactions (R96 and R97) on ignition kinetic.

Acknowledgments

This work is supported by the National Natural Science

Foundation of China (Grant No. 51136005, 51121092).

r e f e r e n c e s

[1] Karim GA, Wierzba I, AL-Alousi Y. Methaneehydrogenmixtures as fuels. Int J Hydrogen Energy 1996;21:625e31.

[2] Klell M, Eichlseder H, Sartory M. Mixtures of hydrogen andmethane in the internal combustion engine-synergies,potential and regulations. Int J Hydrogen Energy 2012;37:11531e40.

[3] Verhelst S, Wallner T. Hydrogen-fueled internal combustionengines. Prog Energy Combust Sci 2009;35:490e527.

[4] Moreno F, Munoz M, Arroyo J, Magen O, Monne C, Suelves I.Efficiency and emissions in a vehicle spark ignition enginefueled with hydrogen and methane blends. Int J HydrogenEnergy 2012;37:11495e503.

[5] Ju Y, Niioka T. Ignition simulation of methane/hydrogenmixtures in a supersonic mixing layer. Combust Flame 1995;102:462e70.

[6] Huang Z, Zhang Y, Zeng K, Liu B, Wang Q, Jiang D.Measurements of laminar burning velocities for naturalgasehydrogeneair mixtures. Combust Flame 2006;146:302e11.

[7] Konnov AA, Riemeijer R, de Goey L. Adiabatic laminarburning velocities of CH4 þ H2 þ air flames at low pressures.2010; 89: 1392e6.

[8] Yu G, Law CK, Wu CK. Laminar flame speeds ofhydrocarbon þ air mixtures with hydrogen addition.Combust Flame 1986;63:339e47.

[9] Dagaut P, Nicolle A. Experimental and detailed kineticmodeling study of hydrogen-enriched natural gas blendoxidation over extended temperature and equivalence ratioranges. Proc Combust Inst 2005;30:2631e8.

[10] Dagaut P, Dayma G. Hydrogen-enriched natural gas blendoxidation under high-pressure conditions: experimental anddetailed chemical kinetic modeling. Int J Hydrogen Energy2006;31:505e15.

[11] Lifshitz A, Scheller K, Burcat A, Skinner GB. Shock-tubeinvestigation of ignition in methaneeoxygeneargonmixtures. Combust Flame 1971;16:311e21.

[12] Cheng RK, Oppenheim AK. Autoignition inmethaneehydrogen mixtures. Combust Flame 1984;58:125e39.

[13] de Vries J, Petersen EL. Autoignition of methane-based fuelblends under gas turbine conditions. Proc Combust Inst 2007;31:3163e71.

[14] Huang J, Bushe WK, Hill PG, Munshi SR. Experimental andkinetic study of shock initiated ignition in homogeneousmethaneehydrogeneair mixtures at engine-relevantconditions. Int J Chem Kinet 2006;38:221e33.

[15] Gersen S, Anikin NB, Mokhov AV, Levinsky HB. Ignitionproperties of methane/hydrogen mixtures in a rapidcompression machine. Int J Hydrogen Energy 2008;33:1957e64.

[16] Chaumeix N, Pichon S, Lafosse F, Paillard CE. Role ofchemical kinetics on the detonation properties of hydrogen/natural gas/air mixtures. Int J Hydrogen Energy 2007;32:2216e26.

[17] Herzler J, Naumann C. Shock-tube study of the ignition ofmethane/ethane/hydrogen mixtures with hydrogencontents from 0% to 100% at different pressures. ProcCombust Inst 2009;32:213e20.

[18] Zhang Y, Huang Z, Wei L, Zhang X, Law CK. Experimentaland modeling study on ignition delays of lean mixtures ofmethane, hydrogen, oxygen, and argon at elevatedpressures. Combust Flame 2012;159:918e31.

[19] Zhang Y, Huang Z, Wei L, Niu S. Experimental and kineticstudy on ignition delay times of methane/hydrogen/oxygen/nitrogen mixtures by shock tube. Chin Sci Bull 2011;56:2853e61.

[20] Wang H, You X, Joshi AV, Davis SG, Laskin A, Egolfopoulos F,et al. USC 2.0 Mech. Available from: http://ignis.usc.edu/USC_Mech_II.htm/; 2007.

[21] Morley C. Gaseq: a chemical equilibrium program forWindows. Available from: http://www.%20gaseq.%20co.%20uk/; 2007.

[22] Zienkiewicz OC, Zhu J. A simple error estimator and adaptiveprocedure for practical engineerng analysis. Int J NumerMethods Eng 1987;24:337e57.

[23] Kee RJ, Rupley FM, Miller JA. Chemkin II. Report no. 89e8009.Sandia National Laboratories; 1989.

[24] Lutz AE, Kee RJ, Miller JA. Senkin: a Fortran program forpredicting homogeneous gas phase chemical kinetics withsensitivity analysis. Report no. SAND87e8248. SandiaNational Laboratories; 1987.

[25] Pang GA, Davidson DF, Hanson RK. Experimental study andmodeling of shock tube ignition delay times forhydrogeneoxygeneargon mixtures at low temperatures.Proc Combust Inst 2009;32:181e8.

[26] Smith GP, Golden DM, Frenklach M, Moriarty NW, Eiteneer B,Goldenberg M, et al. GRI-Mech 3.0. Available from: http://www.me.%20berkeley.%20edu/gri_mech/; 2000.

[27] Petersen EL, Kalitan DM, Simmons S, Bourque G, Curran HJ,Simmie JM. Methane/propane oxidation at high pressures:experimental and detailed chemical kinetic modeling. ProcCombust Inst 2007;31:447e54.

[28] Frenklanch M, Wang H, Yu C, Goldenberg M, Bowman CT,Hanson RK, et al. GRI-Mech 1.2. Available from: http://www.me.berkeley.edu/gri_mech/; 1995.

[29] Davis SG, Joshi AV, Wang H, Egolfopoulos F. An optimizedkinetic model of H2/CO combustion. Proc Combust Inst 2005;30:1283e92.

[30] Wang H, Laskin A. A comprehensive kinetic model ofethylene and acetylene oxidation at high temperatures.Internal report. Univ. of Delaware; 1998.

[31] Davis SG, Law CK, Wang H. Propene pyrolysis and oxidationkinetics in a flow reactor and laminar flames. CombustFlame 1999;119:375e99.

[32] Laskin A, Wang H, Law CK. Detailed kinetic modeling of 1, 3-butadiene oxidation at high temperatures. Int J Chem Kinet2000;32:589e614.

i n t e rn a t i o n a l j o u rn a l o f h y d r o g e n en e r g y x x x ( 2 0 1 2 ) 1e9 9

10411042104310441045104610471048104910501051105210531054105510561057105810591060106110621063106410651066106710681069107010711072107310741075107610771078107910801081108210831084108510861087108810891090109110921093109410951096109710981099110011011102110311041105110611071108

11091110111111121113111411151116111711181119112011211122112311241125112611271128112911301131113211331134113511361137113811391140114111421143114411451146114711481149115011511152115311541155115611571158115911601161116211631164116511661167116811691170117111721173117411751176

HE10552_proof ■ 27 September 2012 ■ 9/9

Please cite this article in press as: ZhangY, et al., Experimental andmodeling study on auto-ignition characteristics ofmethane/hydrogen blends under engine relevant pressure, International Journal of Hydrogen Energy (2012), http://dx.doi.org/10.1016/j.ijhydene.2012.09.056