Burning velocities of hydrogen-methane-air mixtures at
highly steam-diluted conditions
Göckeler Katharina, Eric Albin, Oliver Krüger, Christian Oliver Paschereit
To cite this version:
Göckeler Katharina, Eric Albin, Oliver Krüger, Christian Oliver Paschereit. Burning velocities of hydrogen-methane-air mixtures at highly steam-diluted conditions. 4th International
Conference on Jets, Wakes and Separated Flows - ICJWSF-2013, Sep 2013, Nagoya, Japan.
2013.
HAL Id: hal-00954523
https://hal.archives-ouvertes.fr/hal-00954523
Submitted on 22 Feb 2016
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diffusion de documents
scientifiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
4th International Conference on Jets, Wakes and Separated Flows, ICJWSF-2013
September 1721, 2013, Nagoya, JAPAN
BURNING VELOCITIES OF HYDROGEN-METHANE-AIR
MIXTURES AT HIGHLY STEAM-DILUTED CONDITIONS
1
2
1
1
Katharina Göckeler , Eric Albin , Oliver Krüger , and Christian Oliver Paschereit
1
Chair of Fluid Dynamics, Technische Universität Berlin
Müller-Breslau-Strasse 8, 10623 Berlin, Germany
[email protected]
2
Université de Lyon, CNRS
Université de Lyon 1, F-69622, France
INSA-Lyon, CETHIL, UMR5008, F-69621, Villeurbanne, France
ABSTRACT
or stationary Bunsen ames [7]. The ame propagation
methods require a correction for ame stretch and cur-
Humidied gas turbines using steam generated from excess heat feature increased cycle eciencies.
vature eects, but yield consistent results, if these cor-
Injecting
relations are known [8]. The stationary ame method is
the steam into the combustor reduces NOx emissions,
regarded as the less complex conguration, and is used in
ame temperatures and burning velocities, promising
the present study. It is based on the extraction of ame
a clean and stable combustion of highly reactive fuels,
such as hydrogen or hydrogen-methane blends.
surface areas from ame images. An error source for this
This
method are stretch and curvature eects at the ame tip
study presents laminar burning velocities for methane,
and the ame shoulder near the burner nozzle. Selle et
and hydrogen-enriched methane (10 mol% and 50 mol%)
al.
at steam contents up to 30% of the air mass ow. Experiments were conducted on prismatic Bunsen ames
Simulations, and quantied the inuence of these eects
stabilized on a slot-burner employing OH planar laser-
to only 1%. However, they postulated that the error may
induced uorescence for determining the ame front ar-
increase for mixtures exhibiting Lewis numbers far from
eas. The experimental burning velocities agree well with
unity.
results from one dimensional simulations using the GRI
3.0 mechanism.
For methane-air mixtures several studies on laminar
Burning velocities are increased with
burning velocities are available [10, 11, 12, 13], whereas
hydrogen enrichment, and reduce non-linearly with as-
investigations of steam-diluted mixtures are more scarce,
cending steam molar fractions, showing the potential of
and often limited to comparably low water vapor con-
steam dilution for a stable combustion of these fuels over
tents [2, 9, 14] or to stoichiometric methane-air-steam
a wide ammability range.
1
mixtures [15, 4, 8]. Selle et al. [9] and Boushaki et al.
[14] conducted experiments on a prismatic Bunsen ame
INTRODUCTION
for water vapor contents up to the saturation point in
air.
In gas turbines, excess heat can be eectively recovered
to generate steam.
[9] compared measurements on a prismatic Bunsen
ame similar to the present setup with Direct Numerical
Mazas et al.
[4] studied methane-air-steam mix-
tures for molar steam fractions up to 15 mol%, and used
Injecting the steam into the com-
oxygen-enriched mixtures to achieve higher steam con-
bustor leads to a higher cycle eciency and reduces the
tents of 45 mol%. They found a quasi-linear decrease in
formation of NOx emissions, meeting modern demands
burning velocity with steam molar fraction. Galmiche et
of power generation [1]. For the inuence of steam on the
al. [8] extracted burning velocitites from spherically ex-
combustion process, three mechanisms have been iden-
panding ames for steam molar fractions up to 25 mol%
tied: (1) a dilution eect reducing the molar fraction
of the stoichiometric mixture at preheat temperatures
of total reactive species, (2) a thermal eect due to an
of 470 K. Their results showed some deviations from a
increased specic heat capacity, and (3) a chemical in-
linear decrease, which they related to a small, but negli-
uence on kinetic reactions often related to its high e-
gible chemical eect of steam. Recently, Albin et al. [16]
ciency in third-body reactions [2, 3, 4, 5, 6]. As a result,
measured burning velocities over a wider equivalence ra-
ame temperatures and burning velocities are reduced,
tio range and steam molar fractions up to 25 mol% at
lowering the risk of ashback. Therefore, wet cycles rep-
preheat temperatures of 480 K using a prismatic Bun-
resent a promising technique for the clean and stable
sen ame and a turbulent rod-stabilized V-ame.
combustion of highly reactive fuels, such as hydrogen-
The
reduction in burning velocity with steam molar fraction
enriched fuels and pure hydrogen.
was akin for the laminar and turbulent ame.
The laminar burning velocity is a fundamental pa-
The inuence of hydrogen addition on methane-
rameter for the assessment of the combustor operabil-
air combustion found some interest in the past decade
ity and the validation of kinetic mechanisms. Measure-
[17, 18, 19, 20, 14].
ments are mainly conducted on either propagating ames
1
Wang et al.
[19] showed that the
4th International Conference on Jets, Wakes and Separated Flows, ICJWSF-2013
September 1721, 2013, Nagoya, JAPAN
(a) Schematic of slot-burner
(b) Flame images
Figure 1: Experimental setup: (a) schematic of slot burner including ow supplies, and (b) photos of the ame from
front-top view (left) and side view (right).
promotion of chemical reaction with hydrogen addition
extended in length in order to damp acoustically intro-
is due to the increase of H, O and OH mole fractions in
duced oscillations of the ame [9].
the ame as hydrogen is added. Boushaki et al. [14] con-
The mass ow rates of methane and air were con-
ducted experiments on a slot burner for molar fractions
trolled by Coriolis mass ow meters, and two pneumat-
of hydrogen between 0 mol% and 30 mol% for dierent
ically driven control valves. The air was heated with a
pressures and inlet temperatures. They showed that the
1.5 kW preheater, and an evaporator of 3 kW, equipped
enrichment of fuel with hydrogen increases burning ve-
with a pressure based ow meter (±10%), was used to
locity linearly.
generate steam. The preheated air and steam were pre-
The current study extends the existing database by
mixed with the fuel upstream of the burner. In order to
showing the combined eect of hydrogen enrichment and
reduce heat losses, the supply lines, as well as the burner
steam dilution of methane-air mixtures. Hydrogen con-
were insulated with ceramic bres. The temperature of
tents in fuel of 0 mol%, 10 mol%, and 50 mol% are studied
the mixture was monitored by a thermocouple positioned
for an increasing dilution with steam up to 0.29 mol% of
95 mm upstream of the nozzle outlet without perturbing
the gas mixture. The experiments were conducted on an
the ame. A second thermocouple was used beforehand
atmospheric prismatic Bunsen ame stabilized on a slot-
to check for deviations to the temperature at the actual
burner at a xed preheat temperature of 440 K. Flame
nozzle outlet. For all investigated operating conditions a
areas are extracted from OH planar laser-induced uo-
xed preheat temperature of 440 K was used.
The steam content
rescence (OH-PLIF). Additionally, ow velocities are obtained from Particle Image Velocitmetry (PIV) for some
Ω
steam to air:
of the investigated operating conditions. The experimen-
Ω=
tal results are compared to one dimensional simulations,
is given as mass fraction of
ṁsteam
ṁair
(1)
The enrichment with hydrogen is characterized by
which were performed using the GRI 3.0 mechanism of
the molar fraction of hydrogen in fuel
Smith et al. [21].
In the following, the experimental setup and the in-
XH2 =
vestigated operating conditions are specied, and the
post processing method for the extraction of ame front
areas from the OH-PLIF images illustrated. Thereafter,
with
n
XH2 :
n (H2 )
n (H2
(2)
+ CH4 )
referring to the molar amount of substance.
the ow eld of the Bunsen ame, the ame structure
The equivalence ratio describes the mass fraction of
and burning velocities of the various mixtures are pre-
fuel in oxidizer normalized with the stoichiometric ratio.
sented and discussed.
It is derived from the global one-step reaction for the
oxidation of hydrogen-methane mixtures [20]:
2
EXPERIMENTAL SETUP
CH4
+ H2 + 2.5O2 → CO2 + 3H2 O
Experiments were conducted on an atmospheric pris-
which yields the equivalence ratio
matic Bunsen ame stabilized on a rectangular slot-
tions
Y:
burner. A schematic illustrates both the test-rig and the
supply lines of steam, fuel, and air (Fig. 1). A honeycomb
φ
based on mass frac-
CH4 + 8Y H2
ṁfuel 4Yfuel
fuel
φ=
O2
ṁair Yair
(3)
and a ne grid, placed in the settling chamber upstream
An overview over the investigated operating condi-
of the nozzle, were used to homogenize and laminarize
tions is provided in Tab. 1. The bulk velocity of unburnt
the ow. The slot outlet area sizes
B xH =10 x 100 mm2 .
gases
Uu
was varied in order to maintain a stable ame
Sl .
More details on the nozzle shape are available in Albin
at dierent burning velocities
et al. [16]. For the present study, the nozzle mouth was
richment with hydrogen, higher bulk velocities relative to
2
For an increasing en-
4th International Conference on Jets, Wakes and Separated Flows, ICJWSF-2013
September 1721, 2013, Nagoya, JAPAN
Table 1: Investigated operating conditions characterized
by the molar fraction of hydrogen in fuel
fraction of steam in air
Ω,
bulk velocity of the unburnt gases
laminar burning velocity
XH2 ,
the mass
φ, the
Uu , and its ratio to the
the equivalence ratio
Sl .
The preheat temperature is
kept constant at 440 K.
XH2
Ω
φ
Uu
-
-
-
(m/s)
Sl /Uu
%
0
0 - 0.15
0.8 -1.3
1.3 - 2.4
14 - 34
0.1
0 - 0.20
0.7 -1.3
1.2 - 2.6
14 - 33
0.5
0 - 0.30
0.6 -1.3
1.1 - 6.3
10 - 22
(a) POD image
(b) Filtered image and spline
Figure 2: Dierent stages of data post processing: (a)
the burning velocity were needed due to a more frequent
rst POD mode of OH-PLIF image series, and (b) spa-
occurrence of ashback.
tially ltered POD image with tted spline at ame
Ω = 0.30
The highest steam content of
front.
was reached for the highest enrichment with
hydrogen tested.
Laminar burning velocities are extracted from measurements of OH planar laser-induced uorescence (OH-
ities, which increase ow velocities at the measurement
PLIF). A dye laser operated with Rhodamine 6G and
plane in the slot center. Selle et al. [9] quantied the ow
pumped with a Nd:YAG laser was used at an energy of
acceleration, and suggested to use a correction factor of
8 mJ per pulse and a wavelength of around 283 nm for
1.054 for the present slot dimensions.
triggering the uorescence.
The two dimensional ame length
The laser sheet was posi-
Lf
is determined
An image intensi-
from the measured OH-PLIF images. The OH radical is
ed camera, equipped with a band pass lter centered at
an intermediate combustion product, and well accepted
308 nm, was positioned perpendicular to the laser sheet.
as an indicator for heat release.
For each case, a series of 100 images was recorded at a
gorithm based on the Canny method is used to detect
frequency of 5 Hz and an image resolution of 14 px/mm.
the maximum gradient of the OH signal.
tioned at the slot length center
H/2.
An edge detection alIn order to
Additionally, ow velocities were measured for some
reduce the inuence of noise and to correct for slight
of the investigated cases by employing Particle Image
movements of the ame, the rst mode of the proper or-
Velocitmetry (PIV) with a Nd:YAG laser of 532 nm out-
thogonal decomposition of the time series of OH-PLIF
let wavelength.
The PIV images of 36 px/mm resolu-
images is used. Further improvement of the reliability of
tion were processed with a nal interrogation area of
the edge detection was achieved by means of a moving av-
2
with 50% overlap. A brush-based powder dis-
erage (100 times 3x3 px ), which reduces local gradients
perser provided a homogeneous distribution of titanium
within the unburnt and burnt regions. Figure 2 shows an
dioxide seeding.
example of a POD image (Fig. 2a) and the corresponding
24 x 24 px
2
spatially ltered image (Fig. 2b).
3
FLAME FRONT DETECTION
The identied image points describing the ame front
are tted with a spline
s(x) by means of a least square er-
The burning velocity is dened as the relative velocity
ror method. Figure 2b shows exemplary a ltered POD
of unburnt gases normal to the ame front as they move
image together with the computed spline.
into the ame [7]. For stationary ames, this is equal in
length
amount to the ame propagation velocity into the un-
tween the left and right ame base, according to [16]:
Lf
burnt mixture. Following the continuity equation, burning velocities
Sl
of laminar Bunsen ames are calculated
from the mass ow of unburnt gases
ρu ,
and the ame front area
ṁu ,
Lf =
their density
Af :
(4)
as the OH signal spreads radially creating a gradient near
the nozzle exit between the quenching distance and OH
Af is obtained from the two
dimensional ame length Lf in the measurement plane
multiplied with the slot length H , thus assuming a pris-
detected in the post ame zone.
Therefore, the spline
was computed only between the left and right longitudinal minimum of the OH distribution.
matic ame shape:
This method
might introduce uncertainties into the ame front detec-
Af = Lf H
ame
(6)
base is not clearly detectable on the OH-PLIF images,
The area of the ame
the
Z q
2
1 + (ds(x)/dx) dx
A shortcoming of this technique is that the ame
ṁu
Sl =
ρ u Af
However,
The ame
is then calculated from the line integral be-
shape
is
not
tion process. Nevertheless, it is believed that the ame
(5)
entirely
surface technique is superior to angle based methods,
prismatic
which assume a perfectly triangular ame shape. A com-
(Fig. 1b) due to wall-boundary eects at the slot extrem-
parison of both methods is available in Selle et al. [9].
3
4th International Conference on Jets, Wakes and Separated Flows, ICJWSF-2013
September 1721, 2013, Nagoya, JAPAN
RESULTS
4.1 Time averaged ow eld
4
For some of the investigated cases the ow eld was measured in order to review the level of velocity uctuations
in the Bunsen ame.
A time averaged ow eld is ex-
emplary presented in Fig. 3. The streamlines show the
purely axial direction of the ow for the unburnt gases.
The ame is stabilized at the slot-nozzle exit leading to a
prismatic ame front, which accelerates the ow both in
axial and lateral directions. Additionally, Fig. 3 presents
I ∗ calculated from the time
0
0
averaged axial and lateral velocity uctuations, u and v ,
the local turbulence intensity
(a)
respectively:
Figure 4:
r 1 02
u + v 02 /u
I =
2
∗
eraged axial velocity
Qualitative OH distributions for increasing
steam contents
(7)
Ω (XH2 = 10%, φ = 1).
The signal
resolution is dierent for each image.
The uctuations are normalized with the local, time av-
in length, as the bulk velocity is three times lower.
u.
A
constant bulk velocity would lead to a ame elongation
at diluted conditions. These observations are consistent
20
with results reported in previous works for a turbulent
18
0.25
ame [22].
0.2
4.3 Laminar burning velocities
14
12
0.15
10
0.1
8
Turbulence intensity I*
16
Axial position x (mm)
Ω = 0, Uu = 2.6 m/s (b) Ω = 0.2, Uu = 1.2 m/s
The experimentally determined burning velocities are
compared with one dimensional simulations using the
GRI 3.0 mechanism of Smith et al.
presented in Fig. 5.
6
0.05
till the lean blow out limit.
0
0
5
Lateral position y (mm)
The results are
The experiments are conducted
for equivalence ratios reaching from fuel-rich mixtures
4
−5
[21] including 325
elementary reactions and 53 species.
In the majority of the
cases, a decent agreement between experimental and simulated burning velocities is achieved, suggesting that the
Figure 3: Local turbulence intensity
streamlines of the time averaged ow
I ∗ overlayed with
(XH2 = 10%, Ω =
GRI 3.0 mechanism is well suited to predict hydrogenmethane-air-steam mixtures of varying composition. In-
0.1, φ = 1.1).
dividual outliers are mostly found for high steam mass
ows, which are present for high steam contents
I∗ ≤
Ω,
but also for high bulk velocities needed to stabilize the
For the unburnt gases, the turbulence intensity is
The highest turbulence
50mol%-hydrogen ames. These deviations might be re-
intensity is encountered in the periphery of the ame
lated to some extend to the comparably high uncertainty
front due to some uctuations in ame position. Aero-
of the pressure based mass ow meter installed in the
dynamic shear with the surrounding quiescent air, en-
evaporator. Also, the ame becomes increasingly more
hanced by lateral density gradients, leads to entrainment
distributed for ascending steam contents (Fig. 4), which
of air into the ame, and a slight movement of the ame
might aect the edge detection method. In average, the
front. However, the basic structure of the ame remains
deviation for all investigated cases amounts to 4% only.
consistent for all instantaneous images, and no wrinkling
The reproducibility of systematically repeated operating
of the ame front was observed, which justies an extrac-
conditions is 5% in average.
found to be below
5%.
An enrichment with hydrogen by 50% in molar frac-
tion of laminar burning velocities from the POD based
averaged images.
tion leads to a strong increase in burning velocity by a
4.2 Flame shape
factor of around 1.6 (Fig. 5a-5c), extending the amma-
This section briey addresses the inuence of a dilution
time, the maximum of burning velocity of approximately
with steam on the ame shape.
φ = 1.05
bility range to lower equivalence ratios.
In Fig. 4, the steam
content is increased from left to right between
Ω = 0.2.
Ω = 0 and
for methane-air is slightly shifted towards a
higher equivalence ratio of about
The steam-diluted ame is more distributed
At the same
φ = 1.1.
A dilution
with steam reduces laminar burning velocities for all in-
resulting in a less distinguishable edge at the ame front
vestigated mixtures.
compared to the dry ame potentially stemming from
a dilution with steam of only 10% of the air mass ow
less steep gradients in OH production [6]. Moreover, the
reduces the maximum burning velocity by almost a half.
diluted ame is slightly lifted, and, here, only similar
The highest steam content of
4
For the methane-air-steam ame,
Ω = 0.3
is reached for
4th International Conference on Jets, Wakes and Separated Flows, ICJWSF-2013
September 1721, 2013, Nagoya, JAPAN
Ω = 0, Exp.
Ω = 0.05, Exp.
Ω = 0.1, Exp.
Ω = 0.15, Exp.
Ω = 0.2, Exp.
Ω = 0.25, Exp.
Ω = 0.3, Exp.
0.5
0
0.5
1
Equivalence ratio φ
(a)
1.5
Burning velocity Sl (m/s)
1
1.5
Ω = 0, GRI 3.0
Ω = 0.05, GRI 3.0
Ω = 0.1, GRI 3.0
Ω = 0.15, GRI 3.0
Ω = 0.2, GRI 3.0
Ω = 0.25, GRI 3.0
Ω = 0.3, GRI 3.0
Burning velocity Sl (m/s)
Burning velocity Sl (m/s)
1.5
1
0.5
0
0.5
1.5
1
Equivalence ratio φ
(b)
XH2 = 0
1
0.5
0
0.5
1.5
1
Equivalence ratio φ
(c)
XH2 = 0.1
1.5
XH2 = 0.5
Figure 5: Laminar burning velocities for varying H2-CH4-air-steam mixtures extracted from prismatic Bunsen ame
experiments and simulated using GRI 3.0 mechanism [21]. The preheat temperature was kept at 440 K. The legends
for (b) and (c) are identical to (a).
a hydrogen content of 50% of fuel (Fig. 5c).
Com-
ammability range, which shows the potential of steam
dilution for the stable combustion of hydrogen-enriched
methane.
molar fraction
XH2O
Sl
with the steam
of the stoichiometric gas mixture
l
rable to those of methane-air ames, but feature a wider
YH2= 0.1, GRI 3.0
YH2= 0.5, GRI 3.0
l
for steam-diluted hydrogen-enriched ames are compa-
The decrease in burning velocity
H2
Y = 0, GRI 3.0
H2
The distribution of burning velocities
Norm. burning velocity S /S (Ω=0)
by around 85%.
Y = 0, Galmiche et al. (2011)
1
pared to dry conditions, the burning velocity is reduced
Linear decrease
0.8
0.6
0.4
0.2
is presented in Fig. 6. Here, the steam molar fraction is
used, since it represents the most common quantity for
characterizing diluent contents. Normalization with the
burning velocity at dry conditions
Sl (Ω = 0)
0
0
0.05
allows for
0.1
0.15
0.2
0.25
Steam molar fraction XH2O
0.3
0.35
a comparison with previous experiments conducted at
dierent preheat temperatures.
Figure 6:
For the present study,
Sl /Sl (Ω = 0)
Sl (Ω = 0) refers to the
mixture (φ = 1).
Normalized burning velocities
the simulated data are shown, as the experimental re-
over steam molar fraction, where
sults exhibit some outliers around the stoichiometry. The
burning velocity of the dry
simulated curves agree closely with data presented in
Galmiche et al.
5
[8], and approximately coincide, sug-
CONCLUSIONS
gesting that for the investigated range the reduction in
The combined inuence of steam dilution and hydrogen
burning velocity with steam molar fraction is relatively
enrichment on the laminar burning velocity of methane-
independent of the level of hydrogen enrichment.
air mixtures was studied. Experiments were conducted
For lower steam contents of less than
XH2O ≤ 0.15,
on an atmospheric Bunsen ame stabilized on a slot-
the decrease in burning velocity seems quasi-linear, as
burner.
pointed out by other authors [4, 14]. However, for higher
burning velocities were extracted from OH planar laser-
steam contents, which are reached in the present study
induced uorescence.
by an enrichment with hydrogen, a signicant deviation
to validate one dimensional simulations employing the
from a linear decrease becomes evident, which is shown
GRI 3.0 mechanism.
The ame front areas needed to determine
The experimental data was used
in Fig. 6 by tting a straight line to the burning ve-
An enrichment with hydrogen signicantly increased
locities at lower steam contents. A correction with the
the laminar burning velocity extending the ammable
thermal diusivity as proposed by Koroll and Mulpuru
range towards leaner conditions. A dilution with steam
[3] leads to a similar slope of the curves, which suggests
reduced burning velocities similarly for the tested lev-
that the inuence of steam is not purely thermal by al-
els of hydrogen addition. The increased burning veloc-
tering the thermal diusivity, but that the steam con-
ities for hydrogen-enriched methane enabled a dilution
tributes to the reaction mechanisms. Further simulations
with steam of 29 mol% of the gas mixture. At these high
are needed to assess the inuence on individual reaction
steam contents, the decrease in burning velocity with
steps of methane-hydrogen combustion, and to quantify
steam molar fraction deviates signicantly from the lin-
this chemical inuence depending on steam content.
ear behavior often found for lower steam contents.
5
4th International Conference on Jets, Wakes and Separated Flows, ICJWSF-2013
September 1721, 2013, Nagoya, JAPAN
6
ACKNOWLEDGEMENTS
[12] G. Rozenchan, D. Zhu, C. Law, and S. Tse, Outward propagation, burning velocities, and chemical
eects of methane ames up to 60atm, Combustion
and Flame, vol. 29, no. 2, pp. 14611470, 2002.
The authors would like to thank Simon Kern, Andy
Göhrs, and Holger Nawroth for their valuable contributions during the conduction and preparations of the
[13] K. J. Bosschaart and L. P. H. D. Goey, The laminar
experiments.
burning velocity of ames propagating in mixtures
The research leading to these results has received
of hydrocarbons and air measured with the heat ux
funding from the European Research Council under the
method,
ERC grant agreement no. 247322, GREENEST.
Combustion and Flame, vol. 136, pp. 261
269, 2004.
[14] T. Boushaki, Y. Dhué, L. Selle, B. Ferret, and
References
T. Poinsot, Eects of hydrogen and steam addition on laminar burning velocity of methaneair pre-
[1] M. Jonsson and J. Yan, Humidied gas turbinesa
En-
mixed ame: Experimental and numerical analysis,
[2] V. Babkin and A. V'yun, Eect of water vapor on
[15] T. Le Cong and P. Dagaut, Experimental and De-
the normal burning velocity of a methane-air mix-
tailed Modeling Study of the Eect of Water Va-
ture at high pressures,
por on the Kinetics of Combustion of Hydrogen and
review of proposed and implemented cycles,
ergy, vol. 30, pp. 10131078, June 2005.
International Journal of Hydrogen Energy, vol. 33,
pp. 111, Apr. 2012.
Combustion, Explosion, and
Shock Waves, vol. 7, no. 3, pp. 339341, 1971.
Natural Gas, Impact on NOx,
[3] G. W. Koroll and S. R. Mulpuru, The eect of
dilution with steam on the burning velocity and
structure of premixed hydrogen ames,
[16] E. Albin, H. Nawroth, S. Göke, Y. D'Angelo, and
Sympo-
C. O. Paschereit, Experimental investigation of
sium (International) on Combustion, vol. 21, no. 1,
burning velocities of ultra-wet methaneairsteam
pp. 18111819, 1988.
mixtures,
[4] A. Mazas, B. Fiorina, D. Lacoste, and T. Schuller,
Eects
of
water
vapor
addition
on
the
Energy & Fuels,
vol. 23, pp. 725734, Feb. 2009.
Fuel Processing Technology, Aug. 2012.
[17] F. Halter, C. Chauveau, N. Djebaïli-Chaumeix, and
lami-
I. Gökalp, Characterization of the eects of pres-
nar burning velocity of oxygen-enriched methane
sure and hydrogen concentration on laminar burn-
ames,
ing velocities of methanehydrogenair mixtures,
Combustion and Flame, vol. 158, pp. 2428
Proceedings of the Combustion Institute,
2440, Dec. 2011.
[5] S. Göke and C. O. Paschereit, Inuence of Steam
Dilution on Nitrogen Oxide Formation in Premixed
Methane/Hydrogen Flames,
and Power, Jan. 2012.
[6] O. Krüger,
vol. 30,
pp. 201208, Jan. 2005.
[18] M. Ilbas, A. Crayford, I. Yilmaz, P. Bowen, and
Journal of Propulsion
N. Syred,
Laminar-burning velocities of hydro-
genair and hydrogenmethaneair mixtures:
Large Eddy Simulations of Hydro-
experimental study,
An
International Journal of Hy-
drogen Energy, vol. 31, pp. 17681779, Sept. 2006.
gen Oxidation at Ultra-Wet Conditions in a Model
Gas Turbine Combustor Applying Detailed Chem-
Journal of Engineering for Gas Turbines and
Power, vol. 135, p. 021501, Jan. 2013.
[19] J. Wang, Z. Huang, C. Tang, H. Miao, and X. Wang,
[7] C. Rallis and A. Garforth, The determination of
vol. 34,
istry,
laminar burning velocity,
Combustion Science,
Numerical study of the eect of hydrogen addition on methaneair mixtures combustion,
ternational Journal of Hydrogen Energy,
Progress in Energy and
In-
pp. 10841096, Jan. 2009.
vol. 6, no. 4, pp. 303329,
[20] S. Bougrine, S. Richard, A. Nicolle, and D. Vey-
1980.
nante, Numerical study of laminar ame proper-
[8] B. Galmiche, F. Halter, F. Foucher, and P. Dagaut,
ties of diluted methane-hydrogen-air ames at high
Eects of Dilution on Laminar Burning Velocity of
Energy & Fuels,
pressure and temperature using detailed chemistry,
[9] L. Selle, T. Poinsot, and B. Ferret, Experimental
[21] G. P. Smith, D. M. Golden, M. Frenklach, N. W.
and numerical study of the accuracy of ame-speed
Moriarty, B. Eiteneer, M. Goldenberg, C. T. Bow-
measurements for methane/air combustion in a slot
man,
burner,
diner,
Premixed Methane/Air Flames,
International Journal of Hydrogen Energy, vol. 36,
vol. 25, pp. 948954, Mar. 2011.
pp. 1203512047, Sept. 2011.
Combustion and Flame, vol. 158, pp. 146
154, Jan. 2011.
R.
K.
Hanson,
S.
V. V. Lissianski,
Song,
W.
C.
and Z. Qin,
J.
Gar-
GRI 3.0.
http://www.me.berkeley.edu/gri_mech/, 2000.
[10] C. M. Vagelopoulos and F. N. Egolfopoulos, Di-
[22] S. Göke, S. Terhaar, S. Schimek, K. Göckeler, and
rect experimental determination of laminar ame
C. O. Paschereit, Combustion of natural gas, hy-
speeds,
drogen and bio-fuels at ultra-wet conditions,
Proceedings of the Combustion Institute,
[11] X.
Gu,
M.
Haq,
M.
Lawes,
and
R.
Woolley,
Laminar burning velocity and Markstein lengths
of methane-air mixtures,
in
Proceedings of the ASME Turbo Expo 2011, June
6-10, 2011, Vancouver, Canada, 2011.
vol. 27, no. 1, pp. 513519, 1998.
Combustion and Flame,
vol. 121, no. 1-2, pp. 4158, 2000.
6