See

discussions, stats, and author profiles for this publication at: http://www.researchgate.net/publication/282408465

Combustion characteristic and heating

performance of stoichiometric biogas
hydrogenair flame
ARTICLE in INTERNATIONAL JOURNAL OF HEAT AND MASS TRANSFER JANUARY 2016
Impact Factor: 2.38 DOI: 10.1016/j.ijheatmasstransfer.2015.09.040

READS

4 AUTHORS, INCLUDING:
H.s. Zhen

C.s. Cheung

The Hong Kong Polytechnic University

The Hong Kong Polytechnic University

20 PUBLICATIONS 81 CITATIONS

114 PUBLICATIONS 1,917 CITATIONS

SEE PROFILE

SEE PROFILE

Zuohua Huang
Xi'an Jiaotong University
357 PUBLICATIONS 3,997 CITATIONS
SEE PROFILE

Available from: Zuohua Huang

Retrieved on: 07 October 2015

International Journal of Heat and Mass Transfer 92 (2016) 807814

Contents lists available at ScienceDirect

International Journal of Heat and Mass Transfer

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

Combustion characteristic and heating performance of stoichiometric

biogashydrogenair flame
H.S. Zhen a,, C.W. Leung a, C.S. Cheung a, Z.H. Huang b
a
b

Department of Mechanical Engineering, The Hong Kong Polytechnic University, Hong Kong, China
State Key Laboratory of Multiphase Flows in Power Engineering, Xian Jiaotong University, Xian 710049, China

a r t i c l e

i n f o

Article history:
Received 20 April 2015
Received in revised form 15 September
2015
Accepted 15 September 2015

Keywords:
BiogasH2 fuel
Flame stability
CO emission
Impingement heating

a b s t r a c t
An experimental study has been conducted to investigate stoichiometric biogashydrogen mixed fuel
flames and the combustion characteristic and heating performance of three biogases enriched with
hydrogen were examined. Both similar flame behavior and new finding are reported with reference to
the literature.
The dependence of flame stability on the fraction of hydrogen and the CO2 concentration was analyzed
in the aspects of fuel property alteration and flow aerodynamics variation. The experiments showed a
favorable effect of hydrogen and detrimental effect of CO2 on flame stability. Examination of flame cone
height showed that under constant Re and U, the cone height is longer at either higher hydrogen fraction
or higher CO2 concentration. In addition, at higher fraction of hydrogen in the mixed fuel, flame flammability strengthens, flame temperature enhances and CO emission reduces. Further, polyhedral reaction
cone was detected for stoichiometric flames under high hydrogen concentration conditions.
Great care was paid to the heat transfer behavior of stoichiometric biogashydrogen flames. First, when
the cone tip touches the copper plate, the stagnation point heat flux is not the maximum peak when compared to heat flux peaks at other nozzle-to-plate distance. Comparison between two flames of identical
cone height indicated that higher flame temperature is a dominant factor influencing heat transfer.
Further, longer flame height tends favor heat transfer only at large nozzle-to-plate distances.
2015 Elsevier Ltd. All rights reserved.

1. Introduction
Biogas is a future alternative to traditional fossil fuels. As it is
derived from landfills, agricultural wastes and other sources of biomass, biogas is environmentally-friendly and renewable. With
increasing interest in utilization of biogas, especially in rural areas
[1], better exploration of biogas is under investigation of many
researchers.
The constituents of biogas are mainly methane and carbon
dioxide, with smaller amounts of hydrogen sulfide, moisture and
siloxanes. The molar fraction of carbon dioxide (CO2) is high, which
ranges from 40% to 60% depending on the source of biogas, so biogas is a low-calorific-value fuel. Due to the high content of diluting
gas CO2, the combustion characteristics of biogas are inferior to
natural gas.
Many attempts were made to upgrade the combustion characteristics of biogas [25]. Lee and Hwang [3,4] proposed the method
of adding a higher-grade fuel to biogas to raise its heating value.
Corresponding author. Fax: +852 23654703.
E-mail address: ban18@126.com (H.S. Zhen).
http://dx.doi.org/10.1016/j.ijheatmasstransfer.2015.09.040
0017-9310/ 2015 Elsevier Ltd. All rights reserved.

The authors devoted to biogasliquefied petroleum gas (LPG)

mixed fuel, and observed higher heating value as well as higher
burning velocity of the mixed fuel in comparison with raw biogas.
They found that the biogasLPG mixed fuel can be used interchangeably with liquefied natural gas. Cardona and Amell [5]
added both propane and hydrogen to biogas, and observed that
laminar burning velocity of the fuel mixture is higher than pure
biogas due to introduced propane and hydrogen.
Interest in upgrading biogas by addition of hydrogen to improve
the fuel quality has grown in recent years. Research efforts were
firstly devoted to flame stabilization by using hydrogen addition.
Leung and Wierzba [6] experimentally tested the stability of a biogashydrogen diffusion flame, and observed a substantially wider
stable operational range even though the fraction of hydrogen in
the biogashydrogen mixture is low. Zhen et al. [7,8] observed
improved stability for both diffusion and premixed flames caused
by hydrogen addition to biogas.
Meanwhile, research attention was paid to the variations in
flame temperature and pollutant emission that caused by addition
of hydrogen to biogas. Most researchers reported higher flame
temperature, higher reaction rate and lower carbon monoxide

808

H.S. Zhen et al. / International Journal of Heat and Mass Transfer 92 (2016) 807814

Nomenclature
Re

qmix
Vexit
d
U
Vfuel
Vair

a
lmix
li

Reynolds number of fuel/air mixture jet

density of fuel/air mixture, kg/m3
flow velocity at burner exit, m/s
inner diameter of the nozzle, m
mixture equivalence ratio
volume flow rate of fuel, m3/s
volume flow rate of air, m3/s

Yi
Mi
EICO

(CO) emission for biogashydrogen mixtures in comparison to raw

biogas [5,7,8]. Furthermore, some researchers [7,8] investigated
the heating performance of biogashydrogen flames. Zhen et al.
[7] examined a biogashydrogen diffusion flame and pointed out
that its low sooting characteristic is favorable for domestic heating
applications. The heat transfer rate from the diffusion flame to the
target plate decreased when more hydrogen was added to the biogas. While in the study of a biogashydrogen premixed flame, the
total heat transfer rate was found to be enhanced at higher fraction
of hydrogen addition [8].
The literature on the combustion characteristics of biogashydrogen flames [58] is relatively scarce in comparison to the
numerous studies conducted on methanehydrogen flames. To
name a few, the methanehydrogen fuel was found easier to ignite
and accelerate the relative slow reaction rate of the menthe fuel
[911], thereby improving flame stability and reducing pollutant
emissions [1214]. Compared to the methanehydrogen fuel, the
biogashydrogen fuel has a presence of CO2 which deteriorates
the thermal and emission performance of the flame [15].
The afore-review indicates that only a few studies were previously conducted on the thermal and emission characteristics of
biogashydrogen flames, with less attention put on the heating
performance. As biogas is in wide use for cooking purpose in rural
areas, the heat transfer behavior of the impinging flame is also an
important parameter to be investigated. In the current work, three
biogas/air flames with different levels of hydrogen addition will be
experimentally investigated, with both the combustion characteristics of open flames and the heating performance of impinging
flames to be tested. It is expected that the findings drawn from this
study will provide direct guidance for better utilization of biogas
for practical applications.
2. Experimental setup and method
Fig. 1 schematically shows the experimental setup for this
study. It mainly includes thermocouples, gas emission analyzers
and a test rig for measuring heat flux from impinging flames to a
target copper plate. In this study, biogas was prepared by mixing
pure gases of CO2 and methane (CH4). The volumetric ratios of
CO2 and CH4 were changed from 4:6 to 5:5 and 6:4, and these three
biogases were referred to as BG60, BG50 and BG40, respectively. For
each biogas, pure hydrogen was introduced in and the volumetric
ratio of hydrogen to the biogas is:

VH2
 100%
VCH4 VCO2

All three gases were of over 99% purity to minimize variation in

the composition of the mixtures prepared. Compressed air is used
as the oxidizer. Prior to the burner, both fuel and air flow rates
were monitored by appropriate flow meters, and then were mixed
together by a cylindrical mixing chamber. The burner used is a copper tube, having 9 mm inner diameter and 11 mm outer diameter.
The tube is 400 mm long such that fully developed flow can be

molar fraction of hydrogen in the fuel mixture, %

dynamic viscosity of fuel/air mixture, N  s/m2
dynamic viscosity of component i, N  s/m2
mole fraction of component i, %
molecular weight of component i, kg/mol
emission index of CO, g/kg

established at the burner exit. For each flame established, the Reynolds number (Re) was adopted to represent the aerodynamic
characteristics of the jet flow:

Re

qmix V exit d
lmix

where dynamic viscosity of gas mixture, lmix, was calculated

according to Ikoku [16]:

P

p 

li Y i Mi

lmix P  p
Y i Mi

The equivalence ratio (U) of the fuel/air mixture is:

fuel-to-air-ratio
V fuel =V air

fuel-to-air-ratiost V fuel =V air st

where the suffix st stands for stoichiometry. The current study

focuses on stoichiometric flames to bring out distinct flame behavior at U = 1.0, in contrast to the previous study of a biogashydrogen
flame at U = 1.2 [8].
Experiments were conducted to test the flame stability. For
each biogas, i.e. BG60, BG50 and BG40, ignition was made near the
burner exit to check whether the flame can be stably operating
or not. The operational conditions tested are Re = 400, 600 and
800, and U = 1.0. Then, hydrogen was added to the biogas and such
ignition testing was repeated for biogashydrogen flames. For the
latter flames, the fraction of hydrogen added in the biogashydrogen mixtures increased from 10% to 50%, at an interval of 10%.
To capture luminous photos of the biogashydrogen flames, a
high-resolution CCD camera with a shutter speed of 1/60 s was
used. The flame height, defined as the vertical distance from the
burner exit to the flame tip, was measured by using a ruler of
0.5 mm precision. Flame temperature was measured by an
uncoated B-type thermocouple which has a wire diameter of
0.25 mm. The registered temperatures were then corrected for
radiation and conduction loss.
Carbon monoxide emitted by the flames was examined and
emission index, which is expressed in grams of CO emitted per
kilogram of fuel (biogas plus hydrogen) burned, was used to represent the overall carbon monoxide emission. A tapered quartz probe
was placed 70 mm above the flame tip to collect flue gases. Next,
the sampled gases were condensed to remove water vapor. Then,
the volumetric CO/CO2 concentrations in the sampled gases were
measured by a CO/CO2 analyzer (California Instruments Corp.,
Model 300, NDIR).
The test rig for measuring impinging heat flux consists of a copper plate which acts as a target surface, and a heat flux sensor
which was inserted to the copper plate, with the sensing face flush
with the plate surface on the flame side. It is a small ceramic heat
flux sensor (Vatell Corporation, Model HFM-6D/H) with an effective sensing area of 2  2 mm2, being able to simultaneously measure conduction, convection and radiation heat transfer from the
flame to the plate. The sensor is supplied with a NIST traceable

H.S. Zhen et al. / International Journal of Heat and Mass Transfer 92 (2016) 807814

809

Fig. 1. Experimental setup.

3. Results and discussions

50

3.1. Flame stability

Stable Flames:
BG60

30

BG50

(%)

40

BG40
20
Instable Flames:
Flash-back
Blow-off

10
0
400 600 800 400

600 800 400 600 800

Re

Fig. 2. Stability map for stoichiometric biogasH2/air flames.

calibration certificate that provides all information necessary to

convert the registered signals to heat flux.
For the experiments conducted, each measurement was
repeated three times and the averaged values were reported in this
paper. According to Ref. [17], an uncertainty analysis was performed. Using a 95% confidence level, the uncertainties are 0.8%
in flame temperature, 2.1% in CO concentration and 0.9% in heat
flux.

For the biogas of BG40, BG50 or BG60, both with and without
hydrogen enrichment, the flame stability was tested under the
operational conditions of 400 6 Re 6 800, U = 1.0. The stability
map obtained is given in Fig. 2.
Ignition testing results showed that pure biogas flames burning
BG40, BG50 or BG60, i.e. flames without hydrogen addition are not
stably established. For all experiments tested, burner-stabilized
flames occur only when a certain amount of hydrogen is introduced in and the fraction of hydrogen in the biogashydrogen
(H2) mixture exceeds a certain threshold value. For an example,
the BG60H2 flame at Re = 400 is stable when 10% hydrogen or over
10% hydrogen is added to the biogas. In contrast, the threshold
fraction of hydrogen for a stable BG50H2 flame is 20%, under the
same Reynolds number of Re = 400. Apparently, the biogas BG50
is of higher CO2 content in comparison to BG60, and thus it requires
a larger fraction of hydrogen to be added for flame stabilization.
Chao et al. [15] pointed out that as the fraction of diluting CO2
increases in the fuel mixture, the flame is more affected by preferential diffusion instability, and thus leading the flame closer to the
transition limits, i.e. from stable to unstable flame.
The stability map shown in Fig. 2 further reveals that for the
biogas of higher CO2 content, the number of stable flame operations decreases under the present experimental conditions tested.
The BG60H2 flame has 13 stable operations. In contrast, there are
only 9 and 6 stable operations for the BG50H2 and BG40H2 flames,

810

H.S. Zhen et al. / International Journal of Heat and Mass Transfer 92 (2016) 807814

respectively. For current experimental condition, BG60 has the lowest CO2 content and thus has the widest range of stable operation,
followed by BG50 and then by BG40.
Fig. 2 shows that the BG60H2 flame at Re = 600 is stable as 10%
hydrogen is added to the biogas. However, the 90%BG6010%H2
mixture cannot operate a stable flame at Re = 800, and the flame
is only stable with 20% hydrogen in the BG60H2 mixture. This is
because a nozzle-stabilized flame generally demands that the nozzle flow speed and the flame burning velocity match with each
other. It can be supposed that the flow speed at Re = 800 is higher
than what the laminar burning velocity of the 10%H290%BG60
mixture can match, resulting in instable flame. As more hydrogen
is introduced in, namely the fraction of hydrogen in the BG60H2
mixture increases to 20%, the laminar burning velocity of the 20%
H280%BG60 mixture which is higher than that of the 10%H290%
BG60 mixture, can match the flow velocity at Re = 800 and thus
achieve flame stabilization.
On the top of the stability map given in Fig. 2, flame flash-backs
were observed during experimental testing. For example, the
BG60H2 flame at a = 50% and Re = 400 flashed back and combustion occurs inside the burner nozzle. For the same mixture with
a = 50% but at higher Reynolds number of Re = 600 (or Re = 800),
stable flames are achieved, indicating that the flow speed at
Re = 600 (or Re = 800) has been sufficiently increased to match
the laminar burning velocity of the BG60H2 mixture.
The flame stability behavior of the biogashydrogen mixed fuel
can be summarized in two aspects. First, as compared to biogas,
hydrogen has wider flammability range and hence due to hydrogen
enrichment, the biogashydrogen mixture has extended flammability limits and better flame stability. This aspect of improved
flame stability originates from improved fuel properties. Similarly,
as the concentration of carbon dioxide in the fuel increases, the
flammability range of the biogashydrogen mixture narrows, leading to worse flame stability. On the other hand, flame stability is
subject to the aerodynamics of the jet flow, and a moderate range
of flow speed which matches the burning velocity of the flame supports better flame stability.

3.2. Flame appearance

Typical Bunsen-type conical flames are observed in this study
for various biogashydrogen mixed fuels. Fig. 3 presents luminous
photos of the BG60H2, BG50H2 and BG40H2 flames at Re = 600
and U = 1.0. It can be seen that for each biogasH2 mixture, as
the hydrogen fraction increases, the flame cone shortens and the
flow speed of the fuel mixture increases. For instance, at Re = 600
and U = 1.0, the BG60H2 flame shortens steadily from 24 mm at
a = 10% to 10 mm at a = 50%, and meanwhile, the mean flow speed
of the BG60H2/air mixture monotonically increases from 1.13 m/s
at a = 10% to 1.19 m/s at a = 50%. This coupling between decreasing
flame height and increasing flow speed predicts an increasing laminar burning velocity. The trend of increasing laminar burning
velocity at higher hydrogen fraction was also reported in the literature for methanehydrogen flames and natural gashydrogen
flames [9,10,1820].
A direct comparison of the cone heights of the BG60H2, BG50
H2 and BG40H2 flames under a = 40% or a = 50% in Fig. 3, a reveals
that the BG40H2 flame is the longest while the BG60H2 flame is
the shortest. This is a fact that at the same operational conditions
of Re, U and a, the biogashydrogen mixed fuel with higher CO2
concentration would have a longer flame height. On the one hand,
at constant Re, U and a, the flow speed of the fuel/air mixture is
higher for the mixed fuel with higher CO2 concentration, as shown
in Table 1. And on the other hand, the laminar burning velocity for
the mixed fuel of higher CO2 content is lower. The overall effect is

that the higher the CO2 concentration in the biogashydrogen

mixed fuel, the longer the flame height.
Within 0% 6 a 6 50%, the color of the flames is mainly blue and
becomes brighter at higher hydrogen fraction. Theoretically, pure
hydrogen combustion is colorless and the blue color of the biogashydrogen flame is due to CH radicals derived from the pyrolysis of biogas. As the fraction of hydrogen increases, hydrogen
enrichment in biogas enhances the reaction H + O2 = O + OH, which
is the main chain branching step and the overall reaction rate indicator [21]. Then, the OH and H radicals produced accelerate the
breakdown of CH4. The faster pyrolysis of CH4 might produce more
CH radicals and thus increase the flame luminosity. H2O gradually
becomes the main emitter of light, leading to a pale blue color [22].
Cozzi and Coghe [23] ascribed the increased brightness to an
increase in soot volume fraction and C2 chemiluminescence emission, because at higher hydrogen fraction, there are higher OH and
O concentrations which contribute to the pyrolysis of carbon species and thus result in higher soot concentration. In Fig. 3, a closer
inspection of the photos at a = 50% reveals a polyhedral front for
the BG50H2 and BG40H2 flames. Polyhedral front is a phenomenon of flame instability, where the burning intensity at the
flame tip is particularly weak, resulting in low temperature and
low luminosity thereunto [24].
3.3. Flame temperature
In this section, the flame temperature was measured. Due to the
difficulty in measuring the flame front temperature, in which case
the flame front deforms near the thermocouple wire, the flame tip
temperature was measured for Re = 600 and U = 1.0. For reference
purpose, the adiabatic flame temperature was calculated by adopting an iterating technique [25].
Both calculated and measured temperatures are given in Fig. 4,
which shows that both are enhanced as the hydrogen fraction in
the flame increases, simply because pure hydrogen combustion
has a much higher temperature than raw biogas. Further, due to
increasing hydrogen fraction, the CO2 concentration in the flame
drops, which also contributes to the increase in the flame temperature. It is seen in the figure that under the same a, either the adiabatic or measured temperature attains a highest value for the
BG60H2 flame, followed by the BG50H2 flame and then the
BG40H2 flame. This is because the component of biogas, CO2, acts
as an inert gas, diluting and absorbing the heat released from the
combustion process.
It is also straightforward from Fig. 4 that the measured temperature is always lower than the adiabatic one, due to incomplete
combustion. Additionally, cold ambient air dilutes the hot combustion gas and there exists flame radiation heat loss to ambient,
which further lower the flame temperature.
3.4. CO emission
The emission index of CO for the biogashydrogen flames at
Re = 600, U = 1.0 is given in Fig. 5. It shows that EICO monotonically
decreases with increasing a. The reduced CO emission caused by
higher hydrogen fraction was also reported in the literature for
methanehydrogen flames [2628], where it is ascribed to three
reasons. First, an increase in hydrogen fraction means a decrease
in the fraction of biogas, which leads to lower carbon input to
the flame and thus lower concentrations of carbon-containing species/products. Second, the increased hydrogen fraction results in
higher OH concentration, which promotes CO oxidation into CO2.
Third, the flame temperature is higher at higher hydrogen fraction,
thus providing a high-temperature environment for more efficient
CO conversion into CO2. Fig. 5 also shows that at the same Re, U and
a, the BG40H2 flame has the highest EICO due to both its highest

H.S. Zhen et al. / International Journal of Heat and Mass Transfer 92 (2016) 807814

811

Fig. 3. Photos of BG60H2/air flames (upper row), BG50H2/air flames (middle row) and BG40H2/air flames (lower row).

CO2 content and its lowest flame temperature. Among three biogases, BG60 has the lowest CO2 content and the flame burning
BG60 has highest flame temperature. Thus, the BG60H2 flame has
the least CO emission, among the three flames.
A comparison of EICO can be made between the biogasH2
flame in this study and the CH4H2 flame in Ref. [29]. The EICO
for the CH4H2 flame is below 3 g/kg in the range of 0% 6
a 6 50%, being much lower than the biogasH2 flame in the same
range of a. The reason is twofold. First, the CH4H2 flame has much
higher temperature, i.e. over 1800 C, while the biogasH2 flame
temperature is low, i.e. below 1500 C. Second, in comparison with
CH4, biogas contains a large fraction of CO2, which depresses the
conversion rate from CO into CO2 in the combustion process due
to chemical inhibition [28].
3.5. Heat transfer
In the literature, numerous studies have been performed on the
heat transfer characteristics of Bunsen-type flames burning hydrocarbon fuels [30]. While only a few attention was paid to hydrocar-

bonhydrogen mixed-fuel flames [25,29]. The literature indicates

that impinging heat transfer is influenced by many factors including the fuel properties, flame thermal characteristics as well as the
operational parameters such as Re, U and H. In view of the complexity of the phenomenon, great care was taken in this study to
analyze the heat transfer of impinging biogashydrogen flame.
For the impinging flames under certain nozzle-to-plate distance, the heat flux was measured along the radial direction starting from the stagnation point. The obtained radial heat flux
distributions for the BG60H2 flame at Re = 600, U = 1.0 and
a = 10% are given in Fig. 6. It illustrates typical heating features
of Bunsen-type premixed flame: (1) when the flame is truncated,
heat flux is low around the stagnation point; (2) when the plate
is above the flame, heat flux peak occurs at the stagnation point.
In comparison to the literature [29,30] where pure hydrocarbon
was considered as fuel, the heat flux peak resides at the stagnation
point as the nozzle-to-plate distance equals to the flame cone
height, the biogashydrogen flame does not exhibit such behavior.
The flame shown in Fig. 6 is 24 mm in height. Thus, as the nozzleto-plate distance increases from H = 20 mm to H = 24 mm, the

812

H.S. Zhen et al. / International Journal of Heat and Mass Transfer 92 (2016) 807814

Table 1
Operational conditions for premixed flames at Re = 600 and U = 1.0.

Air volume flow

rate
(105 m3/s)

CO2 volume flow

rate
(106 m3/s)

CH4 volume flow

rate
(106 m3/s)

H2 volume flow
rate
(106 m3/s)

Fuel/air mixture volume flow

rate
(105 m3/s)

Mean exit flow

velocity
(m/s)

Fuel

10%
20%
30%

6.060
6.070
6.078
6.325
6.085
6.327
6.847
6.088
6.320
6.811

4.057
3.851
3.613
5.472
3.335
4.984
7.613
3.010
4.426
6.602

6.086
5.776
5.419
5.472
5.003
4.984
5.075
4.515
4.426
4.401

1.127
2.406
3.871
4.690
5.559
6.646
8.459
7.525
8.852
11.004

7.18
7.27
7.36
7.88
7.47
7.98
8.96
7.59
8.09
9.01

1.13
1.14
1.16
1.24
1.17
1.26
1.41
1.19
1.27
1.42

BG60H2
BG60H2
BG60H2
BG50H2
BG60H2
BG50H2
BG40H2
BG60H2
BG50H2
BG40H2

40%

50%

2000

220
H = 8 mm
H = 12 mm
H = 16 mm
H = 20 mm
H = 24 mm
H = 28 mm
H = 32 mm

200

1800

180
2

1700

Local heat flux (kW/m )

Flame temperature ( C)

1900

1600
1500

A, BG60

M, BG60

A, BG50

M, BG50

A, BG40

M, BG40

1400
1300
1200
0

10

20

30

40

50

160
140
120
100
80
60
40
20

Fig. 4. Adiabatic temperature (A, hollow symbols) and flame tip temperature (M,
solid symbols).

0
0

10

15

20

25

30

35

40

45

50

Radial distance from stagnation point (mm)

Fig. 6. Radial heat flux profiles of BG60H2 flames at Re = 600 and a = 10%.

9
BG60

EICO (g/kg)

BG50
BG40

7
6
5
4
10%

20%

30%

40%

50%

Fig. 5. EICO of biogashydrogen/air flames at Re = 600 and U = 1.0.

flame height is approached. However, the heat flux peak at

H = 24 mm is prominently lower than the peak at H = 20 mm. This
is because for the hydrocarbon flame, the temperature at the cone
tip is the hottest part of the flame [29,30], causing higher heat flux
to occur at the stagnation point when the flame tip just touches the
copper plate. While for the biogashydrogen flame, the burning
intensity at the flame tip tends to be weakened due to the presence
of hydrogen [24], which is also the reason resulting in polyhedral
flame front at high hydrogen concentration. Therefore, the
maximum heat flux peak does not occur at the stagnation point
as the nozzle-to-plate distance approaches the height of the
biogashydrogen flame.

The effect of hydrogen concentration on impinging flame heat

transfer is examined in Fig. 7, which shows the radial heat flux profiles for the BG40 flame at Re = 600, U = 1.0, and a = 40% and
a = 50%, respectively. At the nozzle-to-plate distances of
H 6 20 mm, the flame at a = 50% has higher heat flux peak and
meantime the peak location is closer to the stagnation point. The
higher peak is due to the higher flame temperature at higher
hydrogen fraction, while the closer peak location is the result of
shorter flame height. Fig. 3 has indicated that the flame at
a = 50% (24 mm) is shorter than the flame at a = 40% (30 mm). At
the nozzle-to-plate distance of H = 24 mm, the flame at a = 40%
(30 mm in height) is truncated by the plate, thus exhibiting low
heat flux at the stagnation region. At H = 24 mm, the tip of the
flame at a = 50% touches the plate, and the heat flux peak resides
at the stagnation point.
It is well known that the total heat transfer rate from a Bunsentype flame to the plate would reach a maximal at a single special
nozzle-to-plate distance where the plate still slightly truncates
the flame [31]. In that circumstance, low heat transfer occurs at
the stagnation region. However, the upper truncated flame develops radially outwards along the plate surface, spreading the high
temperature over a wide region on the plate. Zhen et al. [8] retrofitted such statement by experimentally comparing the heat transfer characteristics of a diffusion flame and a premixed flame. They
proposed that as the total heat transfer is an integration of heat
flux in a circular area starting from the stagnation point, even a
small increase of heat flux in the outer region would incur a signif-

813

H.S. Zhen et al. / International Journal of Heat and Mass Transfer 92 (2016) 807814

180

220
200

Local heat flux (kW/m )

180
160
140
120
100
80
60

175
Total heat transfer rate (W)

H = 8 mm
H = 12 mm
H = 16 mm
H = 20 mm
H = 24 mm

170
165
160
155

BG60, = 10%

150

BG40, = 40%

145

BG40, = 50%

140
8

40
20
0
5

10

15

20

25

30

35

40

45

20

24

50

Radial distance from stagnation point (mm)

Fig. 7. Radial heat flux profiles of BG40H2 flames at a = 40% (open symbols) and
a = 50% (solid symbols) for Re = 600.

220
H = 8 mm
H = 12 mm
H = 16 mm
H = 20 mm
H = 24 mm

200
180
2

16

Fig. 9. Total heat transfer rates of biogasH2 flames at Re = 600 and U = 1.0.

Local heat flux (kW/m )

12

Nozzle-to-plate distance, H (mm)

160
140
120
100
80
60
40
20
0
0

10

15

20

25

30

35

40

45

50

Radial distance from stagnation point (mm)

Fig. 8. Radial heat flux profiles of BG60H2 flames at a = 10% (open symbols) and
BG40H2 flames at a = 50% (solid symbols) for Re = 600.

icant increase in the total heat transfer. Based on their statement,

as the truncated flame just spreads the high heat flux to an outer
region, thus the total heat transfer rate would achieve a maximal.
However, in Ref. [8], how the flame height influences the integration of heat flux is not explored.
This study proceeds to examine the effect of flame height on the
total heat transfer of impinging premixed flames. Fig. 8 shows the
radial heat flux profiles for the BG60H2 flame at a = 10% and the
BG40H2 flame at a = 50%. Both are under Re = 600, U = 1.0 and
have the same height of 24 mm. Direct comparison of the profiles
for two the flames at the same nozzle-to-plate distance in the
range of 8 6 H 6 20 mm shows that the heat flux peaks occur at
identical radial position, which is due to the similar geometric features of the flame cones for the two flames. However, the peak values are different between the two cases, and the BG60H2 flame has
higher heat flux peak than the BG40H2 flame. At H = 24 mm, as the
nozzle-to-plate distance equals to the cone height of both the
BG60H2 and BG40H2 flames, the flame tip just touches the plate.
However, the stagnation point heat flux for the BG60H2 flame is
significantly higher than the BG40H2 flame. This is also because
the BG60H2 flame has higher flame temperature. As shown in

Fig. 4, the tip temperature of the BG60H2 flame at a = 10% is

1432 C, much higher than the BG40H2 flame a = 50% which is
1340 C only. Fig. 9 presents the total heat transfer rates for the
two flames depicted in Fig. 8. At all nozzle-to-plate distances varying from H = 8 mm to H = 24 mm, the total heat transfer rate of the
BG60H2 flame is higher than the BG40H2 flame. As the two flames
are of identical height, this fact indicates a favorable effect of
higher flame temperature in the BG60H2 flame which induces
higher heat transfer. In other words, it indicates an enhancement
effect of higher hydrogen fraction, which agrees with the previous
studies [8,29]. Another supportive factor for the higher heat transfer of the BG60H2 flame is that it has higher energy input than the
BG40H2 flame. The energy or heat input rate, as calculated based
on the mass flow rate and lower heating value of the fuel for the
latter flame is 293.9 W, while it is only 146.1 W for the latter flame.
Therefore, the result is that the BG60 flame at a = 10% delivers a
higher heat transfer rate than the BG40 flame at a = 50%, despite
of identical flame height.
Fig. 9 also shows the total heat transfer rate for the BG40H2
flame at a = 40%, Re = 600 and U = 1.0. Among three flames, the
BG40H2 flame at a = 40% has the longest height and spreads over
a larger area on the plate when compared to the other two flames.
According to the statement made by Zhen et al. [8], the BG40H2
flame at longer height would contribute to the integrated total heat
transfer. However, when compared to the BG40H2 flame a = 50%,
the BG40H2 flame at a = 40% associates with a lower temperature,
which is a more dominant factor influencing the heat transfer.
Hence, the BG40H2 flame at a = 40% tends to have lower total heat
transfer rate at small nozzle-to-plate distances of H < 20 mm, as
shown in Fig. 9. As the nozzle-to-plate distances increases from
H = 20 mm to H = 24 mm, the high-temperature reaction cone is
relatively in less contact with the target plated for the BG40H2
flame a = 50% in comparison to the flame at a = 40%, due to the
longer height of the BG40H2 flame a = 40%. Thus, at H = 20 mm
and H = 24 mm, the BG40H2 flame a = 40% has high total heat
transfer rate, as shown in Fig. 9, indicative of the favorable contribution of longer flame height to higher heat transfer.

4. Conclusions
This paper presents an experimental study focusing on stoichiometric biogashydrogen/air flames to investigate the combustion
and heat transfer characteristics of three biogases enriched with
hydrogen. In each case, biogas was considered as the main fuel
and the molar fraction of hydrogen in the biogasH2 mixture was
varied from 10% to 50%.
Experimental data showed a favorable effect of hydrogen fraction and detrimental effect of CO2 concentration on flame stability.

814

H.S. Zhen et al. / International Journal of Heat and Mass Transfer 92 (2016) 807814

The additive hydrogen in the biogas extended the flammability

limits of the fuel and increases laminar burning velocity of the
flame. While the effect of CO2 concentration is the opposite.
Besides alteration of the fuel property caused by hydrogen, flame
stability also requires match between flow speed and burning
velocity. Experimental results further showed that at higher hydrogen fraction, flame temperature increases, CO emission decreases
and flame height shortens,
In this study, for stoichiometric biogashydrogen flames, polyhedral reaction cone was observed under high hydrogen fraction
conditions. Polyhedral flame is an indicator of weak burning intensity at the flame tip. As a result, the maximum heat flux peak does
not occur at the stagnation point when the nozzle-to-plate distance matches the flame height. Great care was taken to analyze
the roles of flame temperature and cone height for integrated total
heat transfer. It was found that for two flames of identical cone
height, higher flame temperature leads to higher heat transfer.
Longer flame height, as a less dominant factor, only incurs higher
heat transfer at certain large nozzle-to-plate distances.

Conflict of interest
None declared.
Acknowledgment
The Authors thank for the fully financial support from the
Research Grants Council of the Hong Kong SAR (B-Q39F) to the present project.
References
[1] Medium and long term development plan for renewable energy, in: National
Development and Reform Commission of China, Beijing, 2007.
[2] H.S. Zhen, C.W. Leung, C.S. Cheung, Z.H. Huang, Characterization of biogas
hydrogen premixed flames using Bunsen burner, Int. J. Hydrogen Energy 39
(2013) 1329213299.
[3] C.E. Lee, C.H. Hwang, An experimental study on the flame stability of LFG and
LFG-mixed fuels, Fuel 86 (2007) 649655.
[4] C.E. Lee, C.H. Hwang, H.Y. Lee, A study on the interchangeability of LFGLPG
mixed fuels with LFG quality in domestic combustion appliances, Fuel 87
(2008) 297303.
[5] C. Cardona, A.A. Amell, Laminar burning velocity and interchangeability
analysis of biogas/C3H8/H2 with normal and oxygen-enriched air, Int. J.
Hydrogen Energy 39 (2013) 79948001.
[6] T. Leung, I. Wierzba, The effect of hydrogen addition on biogas non-premixed
jet flame stability in a co-flowing air stream, Int. J. Hydrogen Energy 33 (2008)
38563862.
[7] H.S. Zhen, C.W. Leung, C.S. Cheung, Effects of hydrogen addition on the
characteristics of a biogas diffusion flame, Int. J. Hydrogen Energy 38 (2013)
68746881.

[8] H.S. Zhen, C.W. Leung, C.S. Cheung, A comparison of the heat transfer behaviors
of biogasH2 diffusion and premixed flames, Int. J. Hydrogen Energy 39 (2014)
11371144.
[9] Z.H. Huang, Y. Zhang, K. Zeng, B. Liu, Q. Wang, D.M. Jiang, Measurements of
laminar burning velocities for natural gashydrogenair mixtures, Combust.
Flame 146 (2006) 302311.
[10] M. Fairweather, M.P. Ormsby, C.G.W. Sheppard, R. Woolley, Turbulent burning
rates of methane and methanehydrogen mixtures, Combust. Flame 156
(2009) 780790.
[11] G. Hu, S. Zhang, Q.F. Li, X.B. Pan, S.Y. Liao, H.Q. Wang, C. Yang, S. Wei,
Experimental investigation on the effects of hydrogen addition on thermal
characteristics of methane/air premixed flames, Fuel 115 (2014) 232240.
[12] K.A. Francis, R. Sreenivasan, V. Raghavan, Investigation of structures and
reactions zones of methanehydrogen laminar jet diffusion flames, Int. J.
Hydrogen Energy 36 (2011) 1118311194.
[13] S.A.A. El-Ghafour, A.H.E. El-dein, A.A.R. Aref, Combustion characteristics of
natural gashydrogen hybrid fuel turbulent diffusion flame, Int. J. Hydrogen
Energy 35 (2010) 25562565.
[14] H.J. Burbano, A.A. Amell, J.M. Garcia, Effects of hydrogen addition to methane
on the flame structure and CO emissions in atmospheric burners, Int. J.
Hydrogen Energy 33 (2008) 34103415.
[15] Y.C. Chao, C.Y. Wu, K.Y. Lee, Y.H. Li, R.H. Chen, T.S. Cheng, Effects of dilution on
blowout limits of turbulent jet flames, Combust. Sci. Technol. 176 (2004)
17351753.
[16] C.U. Ikoku, Natural Gas Production Engineering, Wiley, New York, 1984.
[17] S.J. Kline, F.A. McClintock, Describing uncertainties in single sample
experiments, Mech. Eng. 75 (1953) 38.
[18] G. Yu, C.K. Law, C.K. Wu, Laminar flame speeds of hydrocarbon + air mixtures
with hydrogen addition, Combust. Flame 63 (1986) 339347.
[19] F. Halter, C. Chauveau, I. Gkalp, Characterization of the effects of hydrogen
addition in premixed methane/air flames, Int. J. Hydrogen Energy 32 (2007)
25852592.
[20] H.J. Burbano, A.A. Amell, J.M. Garcia, Effects of hydrogen addition to methane
on the flame structure and CO emissions in atmospheric burners, Int. J.
Hydrogen Energy 33 (2008) 34103415.
[21] D.J. Hwang, J.W. Choi, J. Park, S.I. Keel, C.B. Oh, D.S. Noh, Numerical study on
NO formation in CH4O2N2 diffusion flames diluted with H2O, Int. J. Energy
Res. 28 (2004) 12551267.
[22] J. Fu, C.L. Tang, W. Jin, L.D. Thi, Z.H. Huang, Y. Zhang, Study on laminar flame
speed and flame structure of syngas with varied compositions using OH-PLIF
and spectrograph, Int. J. Hydrogen Energy 38 (2013) 16361643.
[23] F. Cozzi, A. Coghe, Behavior of hydrogen-enriched non-premixed swirled
natural gas flames, Int. J. Hydrogen Energy 31 (2006) 669677.
[24] C.K. Law, Combustion Physics, 2006, pp. 456459.
[25] H.S. Zhen, C.S. Cheung, C.W. Leung, Y.S. Choy, Effects of hydrogen
concentration on the emission and heat transfer of a premixed LPG
hydrogen flame, Int. J. Hydrogen Energy 37 (2012) 60976105.
[26] C.K. Law, S. Ishizuka, P. Cho, On the opening of premixed Bunsen flame tips,
Combust. Sci. Technol. 28 (1982) 8996.
[27] R.W. Schefer, D.M. Wicksall, A.K. Agrawal, Combustion of hydrogen-enriched
methane in a lean premixed swirl-stabilized burner, Proc. Combust. Inst. 29
(2002) 843851.
[28] G.J. Rortveit, K. Zepter, O. Skreiberg, M. Fossum, J.E. Hustad, A comparison of
low-NOx burners for combustion of methane and hydrogen mixtures, Proc.
Combust. Inst. 29 (2003) 11231129.
[29] H.S. Zhen, C.W. Leung, C.S. Cheung, Y.S. Choy, A comparison of the emission
and impingement heat transfer of LPGH2 and CH4H2 premixed flames, Int. J.
Hydrogen Energy 37 (2012) 1094710955.
[30] S. Chander, A. Ray, Flame impingement heat transfer: A review, Energy
Convers. Manage. 46 (2005) 28032837.
[31] S.S. Hou, Y.C. Ko, Effects of heating height on flame appearance, temperature
field and efficiency of an impinging laminar jet flame used in domestic gas
stoves, Energy Convers. Manage. 45 (2004) 15831595.