International Journal of Hydrogen Energy 31 (2006) 1768 – 1779

www.elsevier.com/locate/ijhydene

Laminar-burning velocities of hydrogen–air and

hydrogen–methane–air mixtures: An experimental study
M. Ilbasa,∗ , A.P. Crayfordb , İ. Yılmaza , P.J. Bowenb , N. Syredb
a Department of Mechanical Engineering, Erciyes University, 38039 Kayseri, Turkey
b School of Engineering, Cardiff University, CF24 OYF, Cardiff, UK

Received 15 November 2005

Available online 14 February 2006

Abstract
The laminar burning velocities of hydrogen–air and hydrogen–methane–air mixtures are very important in designing and
predicting the progress of combustion and performance of combustion systems where hydrogen is used as fuel. In this work,
laminar flame velocities of hydrogen–air and different composition of hydrogen–methane–air mixtures (from 100% hydrogen
to 100% methane) have been measured at ambient temperatures for variable equivalence ratios (ER = 0.8.3.2). A modified test
rig has been developed from the former Cardiff University ‘Cloud Chamber’ for this experimental study. The rig comprises of
a 250 mm length cylindrical stainless steel explosion bomb enclosed at one end with a stainless steel plug which houses an
internal stirrer to allow mixing. The other end is sealed with a 120 mm diameter round quartz window. Optical access for filming
flame propagation is afforded via two diametrically opposed quartz windows in both sides. Flame speeds are determined within
the bomb using a high-speed Schlieren photographic technique. This method is an accurate way to determine the flame–speed
and the burning velocities were then derived using a CHEMKIN computer model to provide the expansion ratio. The design
of the test facility ensures the flame is laminar which results in a spherical flame which is not affected by buoyancy. The
experimental study demonstrated that increasing the hydrogen percentage in the hydrogen–methane mixture brought about an
increase in the resultant burning velocity and caused a widening of the flammability limits. This experiments also suggest that a
hydrogen–methane mixture (i.e. 30% hydrogen+70% methane) could be a competitive alternative fuel for existing combustion
plants.
䉷 2006 International Association for Hydrogen Energy. Published by Elsevier Ltd. All rights reserved.

Keywords: Burning velocity; Hydrogen–methane composite fuel; Schlieren photography; Cardiff explosion vessel

1. Introduction exploding the different composition of hydrogen–

methane–air mixtures in the developed explosion vessel.
A number of experimental studies on hydrogen–air Crayford [7] undertook a benchmarking study, com-
and methane–air mixtures can be found in literature paring values for methane–air explosions obtained
on spherical flames under a variety of conditions (i.e., from the Cardiff Cloud Chamber, and comparing them
[1–6]). The present study extends these works to in- against values quoted in literature. However, as seen
clude hydrogen–methane composite fuel combustion in Fig. 1 what was found was that the values quoted
by Crayford were much lower than those quoted by
∗ Corresponding author. Tel.: +90 352 4374901; others, and it was suggested that perhaps the relatively
fax: +90 352 3302740. small size of the rig did not allow the flame to accel-
E-mail address: [email protected] (M. Ilbas). erate freely for long enough to achieve a maximum

0360-3199/$30.00 䉷 2006 International Association for Hydrogen Energy. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijhydene.2005.12.007
 M. Ilbas et al. / International Journal of Hydrogen Energy 31 (2006) 1768 – 1779 1769

Nomenclature Rair gas constant air, j/kg K

A flame area, m2 Sn flame speed, m/s
ER expansion ratio, dimensionless Su burning velocity, m/s
c capacitance, f t time from ignition, s
Cp heat capacity, j/kg/K T temperature, K
e energy, j u velocity, m/s
f focal length, mm V volume, m3
M mass, kg V voltage, v
P pressure, atm
density, kg/m3
rb flame radius, mm  wavelength of light, nm
R0 universal gas constant, j/kg K ⭋ equivalence ratio, dimensionless

45
40
Burning velocity (cm/s)

Spark effected regine

35 2.75
30
2.5
25
20 2.25
Sn (m/s)

15
Majority of published data 25°C
10 2
Cardiff predicted data 25°C
5 1.75
0 Gu et al.
0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.5
Cardiff Cloud chamber
Equivalence ratio, Φ
1.25
0 10 20 30 40 50 60
Fig. 1. Results of a methane–air benchmarking study, undertaken
by Crayford [7], for the Cardiff Cloud Chamber. rb (mm)

Fig. 2. Comparison of measured flame speeds at different flame

radius in Leeds Bomb and Cardiff Cloud Chamber for methane/air
mixture, ⭋ = 1.0, T0 = 300 K.
steady-state value as found in the work of Gu et al. [1]
and others [2–4,8]. When comparing the raw data for a
stoichiometric case for both work undertaken by Cray- mixtures [1–3,7], many studies were also performed
ford [7] and Gu [1] in Fig. 2 there is a very close match about burning velocities of hydrogen–air mixtures
in flame speed, Sn until the flame radius, rb reaches [5,6,9–15]. Because of the high reactivity of hy-
15 mm. Beyond this point, the flame speed within the drogen, there is substantial interest in the use of
Leeds bomb carries on rising to a plateau, whereas the hydrogen–hydrocarbon mixtures as an alternative fuel
Cardiff results reach a premature peak and fall away. It for improved performance, extended flammability lim-
is thought that this may be due to two reasons: the first its and reduced pollutant emissions [16,17] compared
reason is that the Cardiff vessel is much smaller than with pure hydrocarbon fuels.
the chamber used by Gu [1], and pressure effects may
begin to have an effect before the flame reaches its max-
imum stabilised flame speed, the second reason may be 2. Experimental facilities and programme
the effect of a local pressure effect. At present, although
the vessel has a diameter of 60 mm, where the elec- A specially adapted laminar explosion bomb was
trodes are currently located, the distance between the developed for this study within the Cardiff School of
electrode centre and the front window is only 35 mm, Engineering. The rig was developed from the former
and this may affect the progression of the flame. Cardiff University ‘Cloud Chamber’ previously used to
In addition to the experimental and computational study aerosol clouds and methane interactions with
studies on burning velocities, Su of methane–air water droplets [7].
1770 M. Ilbas et al. / International Journal of Hydrogen Energy 31 (2006) 1768 – 1779

Fig. 3. Schematic representation of modified Cardiff explosion vessel.

The modified chamber is depicted by Fig. 3 below. As A variable energy capacitive-discharge ignition sys-
can be seen the rig comprises of a cylindrical stainless tem is used to initiate a propagating flame kernel. The
steel explosion bomb plugged at one end with a remov- system which was all built into a screened aluminium
able stainless steel plug which houses an internal stirrer die cast box consisted of three 0.47 f rapid-discharge
to allow mixing. The other end is sealed with a 120 mm capacitors positioned in series that were charged by a
diameter round quartz glass window. The chamber was variable high voltage (DC) power supply (0–350 V). Us-
of 120 mm diameter. Optical access for filming flame ing the energy equation (1), a theoretical discharge of
propagation is afforded via two diametrically opposed between 0 and 85 mJ was afforded.
quartz windows.
A vacuum can be formed in the bomb by opening a e = 21 cV 2 . (1)
valve connected to a rotary vane pump (BOC Edwards).
The extent of the vacuum is measured using a strain The rapid discharge of the stored energy to the primary
gauge-type pressure transducer connected to a digital coil of a tightly bound automobile ignition coil (AIC)
display (BOC Edwards). This gives an accuracy in mea- was initiated by the supply of a 5 V TTL pulse to a
surement to 0.1 mbar (accuracy 0.01% of total volume). controlling thyristor. The AIC then acted in a similar
Gases can be introduced into the chamber by three manner to that of a standard (AC) transformer, boosting
separate supply lines each of which can be indepen- the energy on the secondary side by the ratio of turns in
dently controlled using a series of pneumatic and the primary to secondary turns (100:1), thus providing
manual valves positioned in line for safety reasons. a theoretical peak voltage of between 0 and 35 000 V.
 M. Ilbas et al. / International Journal of Hydrogen Energy 31 (2006) 1768 – 1779 1771

Fig. 4. Schematic of Schlieren setup employed for flame front imaging.

Fig. 5. Demonstration of flame front measurement .

The ‘boosted’ voltage, which can be termed the 60

‘breakdown’ voltage, on discharge between the spark 50
Flame front (mm)

gap, is capable of bringing about ionisation of the air al-

40
lowing a spark to propagate between the two electrodes.
The electrodes used for this study are constructed from 30
1 mm diameter stainless steel ground to a 60◦ point. 20
The use of a fine electrode minimised the effects on the
10
propagating flame, and the ground tip provided a more
repeatable spark gap and thus spark to be attained. 0
On testing the circuit it was found that sparks could 0 1 2 3 4
only be reliably created if the supply voltage was set Time (ms)
at 40 V. This meant that the approximate energy levels
Fig. 6. Graphical representation of flame front versus time showing
at the primary coil could be accurately controlled from linear line of best fit.
approximately 1.5–85 mJ.
The calculation of flame speed within the combustion
chamber is reliant on the recording of the flame front the focal length of the first mirror (M1 ). This causes the
position with respect to time, and from this flame speed working section to be fully illuminated by a collimated
may be derived. beam of parallel light.
In order that flame structure of the ‘fast’ hydrogen and The collimated beam of light is then passed onto a
hydrogen–methane flames could be visualised a film- second concave mirror (M2 ), which can be positioned
ing rate of 6000 fps was deemed necessary. Due to the at a convenient distance after the working section. This
low luminosity of hydrogen flames it was necessary to causes an image of the source to be formed at the focal
use a Schlieren photographic technique which records point of the second mirror, at the point depicted by the
changes in density rather than light (Fig. 4). tip of the knife edge (k).
A Schlieren system as demonstrated in Fig. 4 is con- The light source used was built from a 100 W pro-
structed by positioning a light source at a distance of jector bulb with a straight filament, collimated and
1772 M. Ilbas et al. / International Journal of Hydrogen Energy 31 (2006) 1768 – 1779

Fig. 7. Schlieren photographs for different fuel compositions, ⭋ = 1.0. (a) 100% CH4 ; (b) 30% H2 + 70% CH4 ; (c) 70% H2 + 30% CH4 ;
(d) 100% H2 .

focused using two 4 in diameter concave mirrors with ideal for measurement within the specialist graphic
focal lengths of 36.8 in. software.
In this setup, a large portion of the undisturbed light Pure hydrogen, pure methane and differing composi-
source (solid line) from the second/focusing mirror is tions of hydrogen and methane blends in air at ambient
stopped using an adjustable camera aperture as a knife temperature ≈ 25◦ C and atmospheric pressure (1000
edge. This enables an image of the diffracted light to mbar) are studied at equivalence ratios of 0.8 to 3.2.
be captured.
A Photron APX-rs high-speed video recording 3. Data analysis
system, filming at a rate of 6000 fps with a shutter
speed of 1/60 000 second allowed a sharp image to The measurement of the flame front position with
be recorded as a series of compressed digital images time was conducted within specialist software. This
 M. Ilbas et al. / International Journal of Hydrogen Energy 31 (2006) 1768 – 1779 1773

60

50
Flame front (mm)
40

30

20
1
10 2

0
0 1 2 3 4 5
(a) Time (ms)

16

14

12
Flame speed (m/s)

10

4 1
2
2

0
0 10 20 30 40 50 60
(b) Distance (mm)

Fig. 8. 100% H2 (a) flame front versus time; (b) flame speed versus distance, ⭋ = 1.0.

was achieved by positioning guide lines at the point used to model gas phase kinetics. It was conceived at
of ignition (the first line) and the transient position Sandia National Laboratories as a tool for use in re-
of the flame front (the second line) respectively action simulations requiring detailed chemistry. Burn-
as shown in Fig. 5 and recording the relevant ing velocities, Su were calculated from the following
coordinates. relation;
Using a known scaling which is calculated by film-
ing a stationary Vernier calliper set at 40 mm it then Su = Sn /ER. (2)
becomes possible to plot the position of the flame front
with respect to time (Fig. 6).
Microsoft Excel was then used to generate a sixth- 4. Results and discussions
order polynomial fit which was then used to derive a
flame speed for the hydrogen–methane flames. After The experimental procedure consisted of evacuating
obtaining flame speeds, Sn for all conditions (differ- the chamber to a zero bar pressure and charging the
ent fuel compositions and different equivalence ratios), chamber up to the atmospheric pressure (1 bar) with the
ChemKin [18] computational package was used to ob- required amounts of hydrogen–methane and air. The
tain expansion ratios for all the experimental conditions. gases were mixed thoroughly by operating the fan for
ChemKin is a package of about 100 Fortran subroutines a few minutes. After the mixing was complete, the
1774 M. Ilbas et al. / International Journal of Hydrogen Energy 31 (2006) 1768 – 1779

60

50

Flame front (mm)

40

30

20 1
2
10

0
0 2 4 6 8 10 12
(a) Time (ms)

8
7
Flame speed (m/s)

6
5
4
3
2
1
1 2
0
0 10 20 30 40 50 60
(b) Distance (mm)

Fig. 9. 70% H2 + 30% CH4 (a) flame front versus time; (b) flame speed versus distance, ⭋ = 1.0.

mixture was ignited and the combustion event was pho- the flame front reaches the end of the chamber (z =
tographed with the high-speed camera. Wide range of 60 mm) in 28.1 ms for 100% CH4 , while it reaches in
experimental matrix was used in this study (i.e. differ- 11.3 ms for 70% H2 + 30% CH4 , and in only 4.6 ms for
ent fuel compositions, different equivalence ratios). 100% H2 .
In this study, many experiments have been undertaken Graphs of flame front versus time and flame speed,
for different fuel compositions (from pure hydrogen to Sn versus distance for all experimental conditions were
the pure methane). The influence of hydrogen content obtained and as examples some of them are presented
on the burning velocity of methane–air mixtures was in Figs. 8–12. It can be seen that the compositions of
also investigated. Markstein lengths and the influence hydrogen and methane are crucial in the resultant flame
of stretch on flame speed will be discussed in another speed witnessed with the fasted mixtures containing
paper. Flame speeds were recorded for each data point. approximately 100% hydrogen. It is clearly seen that
For accuracy of the measurements, each experiment was higher the hydrogen content, higher the flame speed.
repeated two times for all conditions. The raw data of At these conditions flame speeds in excess of 18 ms−1
flame images and flame speed derivation are given in are witnessed. As hydrogen levels are reduced so does
the photographs (Fig. 7) and summarised for ease of the resultant flame speed observed with the lowest
viewing in Figs. 8–16. The photographs (Fig. 7) clearly recorded speed of 1.8–2.5 ms−1 being witnessed at
show the flame growth with time from the ignition to conditions of low hydrogen percentage (0.0–30% H2 ).
the end of the explosions for different fuel compositions It is clearly seen in the figures that flame speed distribu-
(100% CH4 , 30% H2 +70% CH4 , 70% H2 +30% CH4 , tions vary with fuel compositions; at z = 0.1, generally
and 100% H2 ) at stoichiometric combustion conditions. flame speed was higher than that at further distances
From the Schlieren photographs, it is clearly seen that for pure hydrogen and high hydrogen content mixtures
 M. Ilbas et al. / International Journal of Hydrogen Energy 31 (2006) 1768 – 1779 1775

60

50

40
Flame front (mm)

30

20

1
10
2
0
0 2 4 6 8 10 12 14 16 18

(a) Time (ms)

5
Flame speed (m/s)

1
1
2
0
0 10 20 30 40 50 60

(b) Distance (mm)

Fig. 10. 50% H2 + 50% CH4 (a) flame front versus time; (b) flame speed versus distance, ⭋ = 1.0.

(see Figs. 8–10), while it was lower than that at further methane combustion are not far away from the burning
distances for pure methane and low hydrogen content velocities of hydrogen–air mixtures and methane–air
mixtures. This was because of lower ignition energy mixtures measured by Gu et al. [1], by Scholte and
needed for combustion of hydrogen compared with Vaags [2], by Bradley et al. [3], and by Shebeko et al.
methane combustion. Flame speeds after the middle [4]. All authors found the maximum velocity of about
positions, z = 30 mm towards the end of the chamber, 3.0 ms−1 for hydrogen–air mixtures and 0.3 ms−1 for
z = 60 mm, were decreasing for all conditions. methane–air mixtures for similar conditions.
Figs. 13 and 14 show the variation of flame speed with The effect of equivalence ratio on burning velocities
equivalence ratio and hydrogen percentage in methane, of hydrogen–air, methane–air and hydrogen–methane–
respectively, whilst the variation of burning velocity air mixtures is clearly seen in Fig. 15. As seen in the
with equivalence ratio and with hydrogen percentage in figure, pure hydrogen combustions were performed at
methane are given in Figs. 15 and 16, respectively. The very wide range of equivalence ratios (from 0.8 to 3.2),
maximum burning velocity was of about 3.2 ms−1 for whilst the pure methane combustions were performed at
100% H2 at ⭋ = 1.8, whilst the minimum burning ve- relatively narrow range of equivalence ratios (from 0.8
locity was about 0.25 ms−1 for 100% CH4 at ⭋ = 0.8. to 1.2). The burning velocity of pure hydrogen combus-
The obtained maximum burning velocities of 3.2 ms−1 tion was about 2.2 ms−1 for ⭋ = 0.8, it was 2.6 ms−1
for pure hydrogen combustion and 0.25 ms−1 for pure for ⭋ = 1.0, 3.2 ms−1 for ⭋ = 1.8 and 2.1 ms−1 for
1776 M. Ilbas et al. / International Journal of Hydrogen Energy 31 (2006) 1768 – 1779

60

50

Flame front (mm)

40

30

20
1
10
2
0
0 5 10 15 20 25
(a) Time (ms)

3.5

3
Flame speed (m/s)

2.5

1.5

1
1
0.5 2

0
0 10 30 50 40 50 60
(b)
Distance (mm)

Fig. 11. 30% H2 + 70% CH4 (a) flame front versus time; (b) flame speed versus distance, ⭋ = 1.0.

⭋ = 3.2. The measured maximum burning velocity was each experiment. It is obtained that increasing the
about 3.2 ms−1 and was obtained at ⭋ = 1.8. As seen hydrogen content in the mixture increased the
in the figure, the burning velocity is increasing, when burning velocity dramatically (i.e. 0.25 ms−1 for 0%
increasing the equivalence ratio from low equivalence H2 , 0.9 ms−1 for 50% H2 + 50% CH4 , 1.8 ms−1 for
ratio to over stoichiometric up to ⭋ = 1.8 and then it is 80% H2 + 20% CH4 , 2.9 ms−1 for 100% H2 ).
decreasing, when increasing the equivalence ratio from
⭋ = 1.8 to 3.2. The burning velocity of pure methane
combustion is increasing, when increasing the equiva- 5. Conclusions
lence ratio from ⭋ = 0.8 to 1.1 and it is decreasing,
when increasing the equivalence ratio from ⭋ = 1.1 to 1. Flame speeds and burning velocities of hydrogen–air
1.2. The addition of 50% of hydrogen to the methane and hydrogen–methane–air mixtures have been mea-
increased the burning velocity from 0.25 up to 0.8 ms−1 sured at room temperature and atmospheric pressure
and widened the flammable regions (i.e. from ⭋ = 1.2 using Schlieren photography technique.
to 1.8). 2. Ignition and thus combustion could not be performed
The effect of hydrogen content on burning veloc- at ⭋ = 1.4 and over equivalence ratios for pure
ities of hydrogen–methane–air mixtures is clearly methane–air mixtures, while ignition and thus com-
seen in Fig. 16. The measurements were performed bustion was easily performed at very high equiva-
for very wide range of fuel compositions from pure lence ratios up to ⭋ = 3.2. This was because of the
methane–air mixtures to pure hydrogen–air mix- lower ignition energy needed for hydrogen combus-
tures increasing the hydrogen content 5–10% in tion compared with methane combustion.
 M. Ilbas et al. / International Journal of Hydrogen Energy 31 (2006) 1768 – 1779 1777

60

50

Flame front (mm)

40

30

20
1
10
2

0
0 5 10 15 20 25 30 35
(a) Time (ms)

2.5

2
Flame speed (m/s)

1.5

0.5 2
1
0
0 10 20 30 40 50 60
(b) Distance (mm)
Fig. 12. 100% CH4 (a) flame front versus time; (b) flame speed versus distance, ⭋ = 1.0.

20 14
18
12
16

14 10
Flame speed (m/s)
Flame speed (m/s)

12
8
10
6
8

6 Hydrogen 4
4 Methane
50% H2+ 50% CH4 2
2

0 0
0 1 2 3 4 0 20 40 60 80 100
Equivalence Ratio % Hydrogen
Fig. 13. Flame speeds of different fuels for different equivalence Fig. 14. Flame speed for different percentage of hydrogen in
ratios. methane, (⭋ = 1.0).
1778 M. Ilbas et al. / International Journal of Hydrogen Energy 31 (2006) 1768 – 1779

4 flame stability, wide flammable regions and rela-

tively higher burning velocity.
3.5

3
Burning Velocity (m/s)

Acknowledgments
2.5
This study was supported by the Cardiff Univer-
2 sity, School of Engineering, United Kingdom, the
Erciyes University, Turkey, and UNIDO-ICHET-
1.5 Istanbul, Turkey. The authors gratefully acknowledge
the support of these foundations.
1 Hydrogen
Methane
0.5 50% H2 50% CH4 References
0 [1] Gu XJ, Haq MZ, Lawes M, Wooley R. Laminar burning velocity
0 1 2 3 4
and Markstein lengths of methane–air mixtures. Combust Flame
Equivalence Ratio 2000;121:41–58.
[2] Scholte TG, Vaags PB. The burning velocity of hydrogen–air
Fig. 15. Burning velocities for different fuel compositions and equiv- mixtures and mixtures of some hydrocarbons with air. Combust
alence ratios. Flame 1959;3:495–501.
[3] Bradley D, Gaskell PH, Gu XJ. Burning velocities, Markstein
lengths, and flame quenching for spherical methane–air flames:
3 a computational study. Combust Flame 1996;104:176–98.
[4] Shebeko YuN, Tsarichenko SG, Korolchenko AYa, Trunev AV,
Navzenya VYu et al. Burning velocities and flammability limits
2.5 of gaseous mixtures at elevated temperatures and pressures.
Combust Flame 1995;102:427–37.
Burning Velocity (m/s)

[5] Edmondson H, Heap MP. The burning velocity of hydrogen–air

2 flames. Combust Flame 1971;16:161–5.
[6] Andrews GE, Bradley D. Determination of burning velocity by
double ignition in a closed vessel. Combust Flame 1973;20:
1.5 77–89.
[7] Crayford AP. Suppression of methane–air explosions with water
in the form of ‘Fine’ mists. PhD thesis, University of Wales,
1 Cardiff, UK; 2004.
[8] Pink IP. Investigation into the capabilities of the Cardiff cloud
chamber. MPhil thesis, Cardiff University, Cardiff, United
0.5
Kingdom; 2005.
[9] Liu DDS, MacFarlane R. Laminar burning velocities of
0 hydrogen–air and hydrogen–air-steam flames. Combust Flame
0 20 40 60 80 100 1983;49:59–71.
[10] Milton BE. Laminar burning velocities in stoichiometric
% Hydrogen hydrogen and hydrogen–hydrocarbon gas mixtures. Combust
Flame 1984;58:13–22.
Fig. 16. Burning velocities for different percentage of hydrogen, [11] Aung KT, Tseng LK, Ismail MA, Faeth GM. Brief
⭋ = 1.0. communication, response to comment by S.C. Taylor and D.B.
Smith on Laminar burning velocities and Markstein numbers
of hydrocarbon/air flames. Combust Flame 1995;102:526–30.
3. It has been shown that, with hydrogen content in- [12] Law CK, Kwon OC. Effects of hydrocarbon substitution on
creases in the hydrogen–methane mixtures, the flame atmospheric hydrogen–air flame propagation. Int J Hydrogen
Energy 2004;29:867–79.
speeds and thus the burning velocities also increased [13] Wierzba I, Kilchyk V. Flammability limits of hydrogen–carbon
dramatically. monoxide mixtures at moderately elevated temperatures. Int J
4. The influence of methane addition on flammabil- Hydrogen Energy 2001;26:639–43.
ity of hydrogen–air mixtures was also investigated. [14] Dahoe AE. Laminar burning velocities of hydrogen–air
The flammable regions were widened with hydrogen mixtures from closed vessel gas explosions. J Loss Prevention
Process Industries 2005;18:152–66.
content increased in the mixtures. [15] Kwon OC, Tseng L-K, Faeth GM. Laminar burning
5. The addition of hydrogen to methane gives a good velocities and transition to unstable flames in H2 /O2 /N2 and
alternative fuel to hydrocarbon fuels as giving good C2 H8 /O2 /N2 mixtures. Combust Flame 1992;90:230–46.
 M. Ilbas et al. / International Journal of Hydrogen Energy 31 (2006) 1768 – 1779 1779

[16] Ilbas M, Yilmaz I, Kaplan Y. Investigations of hydrogen and Further reading

hydrogen–hydrocarbon composite fuel combustion and NOx
emission characteristics in a model combustor. Int J Hydrogen [19] Koroll GW, Kumar RK, Bowles EM. Burning velocities of
Energy 2005;30:1139–47. hydrogen–air mixtures. Combust Flame 1993;94:330–40.
[17] Ilbas M, Yilmaz I, Veziroglu TN, Kaplan Y. Hydrogen as [20] Wierzba I, Ale BB. Rich flammability limits of fuel mixtures
burner fuel: modelling of hydrogen–hydrocarbon composite fuel involving hydrogen at elevated temperatures. Int J Hydrogen
combustion and NOx formation in a small burner. Int J Energy Energy 2000;25:75–80.
Res 2005;29:973–90.
[18] Kee RJ, Rupley FM, Miller JA. Technical Report SAND89-
8009. Sandia National Laboratories, Livermore, CA; 1989.