Chemical Engineering Science 131 (2015) 235242

Contents lists available at ScienceDirect

Chemical Engineering Science

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

Combustion characteristics and performance evaluation of premixed

methane/air with hydrogen addition in a micro-planar combustor
Aikun Tang a,b, Yiming Xu a, Jianfeng Pan a,n, Wenming Yang b, Dongyue Jiang b, Qingbo Lu a
a
b

School of Energy and Power Engineering, Jiangsu University, Zhenjiang 212013, China
Department of Mechanical Engineering, National University of Singapore, Singapore 117576, Singapore

H I G H L I G H T S

Hydrogen addition has a dramatic function on ame stabilizing of methane combustion.

The reaction rate of methane will be apparently raised by burned hydrogen.
The effects on the growth of wall temperature and radiant energy are investigated.
Hydrogen addition ratio should exceed 10% so as to ensure a good operation effect.

art ic l e i nf o

a b s t r a c t

Article history:
Received 27 November 2014
Received in revised form
18 February 2015
Accepted 14 March 2015
Available online 28 March 2015

Combustion performance of premixed methane/air with hydrogen addition in a micro-planar combustor

is numerically investigated, aiming to extend the application of various fuels in a microthermophotovoltaic system. The simulation results show that hydrogen addition has signicant effect
on raising methane reaction rate and ame stability. With the increase of hydrogen mass fraction, the
location of the ame shifts toward the combustor inlet gradually, and the temperature gets a steady
growth. For the micro-combustor with a channel height of 3 mm, it is found that the ame of pure
methane is far away from inlet, which does not meet the basic working requirement of the microthermophotovoltaic system. However, the combustor wall temperature will be signicantly improved
when a small amount of hydrogen is added into the mixture, which is mainly due to the extension of
ammability range. Along with the increase of H2 addition ratio, both the total radiant energy and the
proportion of usable radiant energy to total input chemical energy increase dramatically, which brings a
signicant improving of electricity output and system efciency.
& 2015 Elsevier Ltd. All rights reserved.

Keywords:
Micro-combustion
Hydrogen addition
Numerical simulation
Flame position
Radiant energy

1. Introduction
With the development of micro-electro-mechanical systems
(MEMS), great attention has been paid to the design of its energy
supply system. Recently, several combustion-based micro-power
generators have aroused the research interest of scholars around
the world, which have the advantages of high energy density,
small volume as well as long working time (Ju and Maruta, 2011;
Chia and Feng, 2007; Chou et al., 2011). Apart from serving as
electric source for MEMS such as micro-pumps and micro-robots,
the micro-power generators can also provide adequate electricity
or power for portable electronics, wireless communication equipments, vehicles, and military devices (Chia and Feng, 2007).
However, due to the very short history, the technology is still far

Corresponding author. Tel.: 86 511 88780210; fax: 86 511 88780216.

E-mail address: mike@ujs.edu.cn (J. Pan).

http://dx.doi.org/10.1016/j.ces.2015.03.030
0009-2509/& 2015 Elsevier Ltd. All rights reserved.

away from mature. So, it is essential to further improve the

performance of these micro-power devices.
The micro-thermophotovoltaic (MTPV) system is one of the
typical micro-power generators, which consists of three major
components: micro-combustor, PV cells and optical lter. Due to
its high energy density, no moving parts and convenient to
fabricate, it is superior to other portable power generation systems
(Chou et al., 2011). During operation, rstly, the chemical energy
from hydrocarbon fuel is released during the combustion process.
The outer wall of combustor can reach a high-temperature state
and emits photons. Then, the short-wave radiation will transmit
the lter, and arrive at the surface of PV cells. Finally, free electron
can be generated due to photoelectric effect. Therefore, the output
power density and energy conversion efciency of the MTPV
system are directly inuenced by the micro-combustion process
and the outer wall temperature.
For this reason, various research works on the structure
optimization of micro-combustor have been carried out in the

A. Tang et al. / Chemical Engineering Science 131 (2015) 235242

past decade. Yang et al. (2007) tested a micro-cylindrical combustor with backward facing step, and found that the step is very
useful in controlling ame position and widening the operational
range of mixture ow rate. Pan et al. (2010) presented a design
concept of modular TPV power generator based on the submillimeter planar combustor. The combustion characteristics of
premixed hydrogenoxygen were investigated through both
experimental and computational methods. Yang et al. (2011,
2012) also conducted similar studies about premixed hydrogen
air in the micro-planar micro-combustor. The SiC porous media
foam was employed in the combustion channel to enhance heat
transfer between the hot gas and the wall (Yang et al., 2011). A
micro-combustor with a percolated platinum tube was proposed
by Li et al. (2013), which served as catalyst, emitter, and ame
stabilizer to overcome the critical heat loss and improve the ame
instability. Wan et al. (2012) developed a micro-combustor with a
bluff body which can extend the blow-off limit by 35 times.
Besides, several kinds of micro-heat-recirculating combustors for
the MTPV system were reported by Park et al. (2011, 2012), Yang
et al. (2014) and Jiang et al. (2013a, 2013b), which all have positive
effects on improving the temperature of emitters. In addition,
studies of ames propagating in micro-channels were also
reported (e.g. Veeraragavan and Cadou (2011), Kurdyumov and
Jimenez (2014)), which would provide some promotional effects
on the improving of working performance of micro-combustor.
For the MTPV system, the scholars prefer to choose hydrogen as
the fuel of the combustion process. This is typically attributed to
the higher heating value and ignition propensity of hydrogen. Li
et al. (2009) pointed out that the blowout phenomenon occurs
when the velocity of premixed methane and air exceed 1 m/s,
while the blowout limit of hydrogen case can reach 8 m/s in the
same combustion channel. Similar problem of propane combustion in a single channel micro-combustor was also found by
Federici and Vlachos (2008), and they employed a heat recirculation combustor to increase the blowout limit of propane air
mixture. In the study of conventional scale combustion, it was
found that the combustion of hydrocarbon fuel could be greatly
improved by adding a small amount of hydrogen into the mixture,
so a great deal of research on this subject have been conducted
(Khalil and Gupta, 2013; Wang et al., 2009; Sabia et al., 2007;
Titova et al., 2014; Cuoci et al., 2013). When it comes to microscale combustion, Norton and Vlachos (2005) realized the selfignition of propane/air mixtures with the assistance of hydrogen
addition in a catalytic micro-burner. Yan et al. (2014) studied the
hydrogen assisted catalytic combustion of methane on platinum,
and concluded that hydrogen addition has a great inuence on
lowering the methane ignition temperature and shortening the
ignition time. Seshadri and Kaisare (2010) also investigated the
self-ignition property of hydrogen, in which two different modes
of hydrogen-assisted propane ignition were compared. A novel
catalytic micro-reactor which combined the concept of catalyst
segmentation and cavities was designed by Li et al. (2012), and the
enhancement effects on H2/CO/CH4 multi-fuel combustion were
analyzed through numerical simulation. Zarvandi et al. (2012)
conducted a study of CH4/air combustion with hydrogen addition
in a micro-stepped tube, and found that adding hydrogen could
intensify the production of some crucial species which are very
vital for establishing a stable combustion. Tacchina et al. (2010)
investigated the combustion performance of CH4/H2/air mixtures
using different catalysts, and concluded that the presence of H2 in
the fuel will improve the conversion of CH4 due to an increase in
the production of OH radicals.
At the condition of micro-scale, the blending fuel combustion
process may present very different characteristic compared with
conventional scale combustion, which is subjected to the problems
of short residence time and large heat loss. In contrast, the

research on blending fuel combustion in micro-combustor is very

limited, and most of the studies mainly focus on the aspect of
catalytic combustion. However, very few related research for the
micro-thermophotovoltaic system using blending fuel has been
reported. In this paper, the combustion process of premixed
methane/air with hydrogen addition in a micro-planar combustor
for the MTPV system is studied by a simulation method. The
positive effects of hydrogen addition on the growth of wall
temperature distribution and radiant energy output are also
discussed, so as to give a reference for further improvement of
the system working performance.

2. Construction and verication of a computational model

2.1. Geometric model and grid generation
In this work, a micro-planar combustor is designed, as shown
in Fig. 1. The dimensions of the micro-combustor are 18 mm in
length, 9 mm in width and 4 mm in height. The wall thickness is
0.5 mm, so the height of straight combustion channel is 3 mm.
During the operation process, premixed fuel (methane and hydrogen) and air ow into the straight parallel channel through a
rectangle inlet which is located at one end of the combustor. The
burned gas will be expelled out from an outlet at the other end of
combustor. The material of 316 stainless steel is chosen as
combustor wall, which can withstand a temperature of 2000 1C
and has a large emissivity at high-temperature state.
A 3D uniform grid is developed to predict the combustion and the
heat transport in the micro-combustor. Half of the combustor space is
employed as computational domain to save the calculation time.
Centerline gas temperature proles at three different mesh densities
have been checked so as to determine the grid size, as seen in Fig. 2.
After comparison, the mesh with 0.1 mm density in all three directions
and 324,000 total cells number is nally chosen.
Heat loss : convection & radiation
Combustor wall
0.5 mm
Fuel/air inlet
3 mm

Uniform velocity profile

Product outlet

18 mm

Fig. 1. Schematic diagram of micro-planar combustor.

2100
Centerline temperature (K)

236

1800
1500
1200
900
162000 cells (0.20,0.10,0.10)
324000 cells (0.10,0.10,0.10)
648000 cells (0.10,0.10,0.05)

600
300

6
8
10
12
14
Distance from inlet (mm)

16

18

Fig. 2. Centerline temperature proles of combustor at different mesh densities

(10% hydrogen addition, inlet velocity: 2 m/s).

A. Tang et al. / Chemical Engineering Science 131 (2015) 235242

237

Table 1
Important parameters of the computational model.
Parameters

Methods

Flow
Reaction
Discretization
Pressurevelocity coupling
Solver

Laminar
Finite-rate (mechanism with16 species and 25 reversible reactions ) (Li et al., 2009; Zarvandi et al., 2012)
First-order upwind
SIMPLE algorithm
Segregate/implicit (under-relaxation method) (Li et al., 2009)

Software
Mixture physical properties

Fluent
Density: incompressible-ideal-gas law (Norton and Vlachos, 2003)
Specic heat: mixing-law (Zarvandi et al., 2012)
Thermal conductivity: mass-weighted-mixing law (Jiang et al., 2013a; Li et al., 2009; Norton and Vlachos, 2003)
Viscosity: mass-weighted-mixing law (Jiang et al., 2013a; Li et al., 2009; Norton and Vlachos, 2003)
Mass diffusivity: kinetic-theory

Species physical properties

Specic heat: a piecewise polynomial tting of temperature (Jiang et al., 2013a; Li et al., 2009)
Thermal conductivity: kinetic-theory (Zarvandi et al., 2012)
Viscosity: kinetic-theory (Zarvandi et al., 2012)

Boundary conditions

Inlet: velocity-inlet
Exit face: pressure-outlet
External wall: mixed thermal conditions

2.2. Methods of model building

Important parameters of the computational model are shown
in Table 1. A skeletal mechanism for methaneair combustion is
employed in the computational model, which has 16 species and
25 reversible reactions (Li et al., 2009). The mechanism has been
proved to be applicable in the simulation of micro-combustion.
Due to the low inlet ow velocity, laminar model and laminar
nite-rate model are chosen as ow and chemistry reaction
models, respectively. The basic governing equations of gas phase
include the mass conservation equation, the momentum conservation equation, the chemical components transport equation and
the energy conservation equation, which are expressed as follows.
Continuity:

ui 0
xi

where is the density, ui is the velocity components of xi direction

(i 1, 2, 3).
Momentum:

p
ui uj  ij 
xj
xi

where p is the pressure, ij is the stress tensor.

Species:

uj ml J l;j Rl
xj

2.3. Validation of the computational model

3

where ml is the mass fraction of species l, J l;j is the diffusion ux of

species l in the xj direction, Rl is the production rate of species l by
chemical reaction.
Energy:

p
u
uj h F h;j uj ij i
xj
xj
xj

chemistry solver and under-relaxation method are used to overcome

this question.
In our simulation, the specic heat of each component is
calculated using a piecewise polynomial tting of temperature,
while both of thermal conductivity and viscosity are calculated by
kinetic-theory. The density and specic heat of the fuel and air
mixture are calculated by incompressible-ideal-gas law and mixing-law, respectively. The mixture gas thermal conductivity and
viscosity are calculated as a mass fraction-weighted average of all
species (Norton and Vlachos, 2003), and the kinetic-theory is
selected for mass diffusivity calculation of mixture (Zarvandi
et al., 2012).
The boundary conditions in the computational model are set as
follows. At the inlet, velocity-inlet boundary condition is chosen. A
uniform at velocity prole is assumed, and the mass fraction of
each component is specied. Pressure-outlet boundary condition
is employed at the exit face, the constant pressure and temperature of mixture are given, and the mass fraction of each product is
estimated. No slip and zero diffusive ux species boundary
conditions are employed at the gassolid interfaces. For each
outer wall, both convection and radiation heat transfer with
environment should be considered, so mixed thermal conditions
are set for them, the convective heat transfer coefcient and wall
emissivity are taken to be 15 W/(m2 K) and 0.65, respectively.

where h is the total enthalpy, F h;j is the energy ux of the xj

direction.
All of the governing equations are discretized by a rst-order
upwind scheme, and SIMPLE algorithm is used to deal with the
pressurevelocity coupling. Fluent is chosen to solve the governing
equations implicitly and the convergence criterions of all residuals
are set to be 10  6. Compared to case of one-step reaction mechanism, the calculation is more difcult to get converged, so stiff

The experimental validation has been conducted on the microplanar combustor with the fuel of pure methane. The experimental setup is given in the literature (Pan et al., 2010), and an infrared
thermographer (Camera model: ThermovisionTM A40) is adopted
to measure the outer wall temperature distribution of combustor.
It is found that the premixed methane and air cannot be ignited
when the channel height is less than 3 mm. This is the reason why
the 3 mm inner height combustor is tested and simulated in this
paper. Fig. 3(a) shows the comparison on combustor wall temperature distribution patterns of experiment and simulation at
inlet velocity of 0.6 m/s. Fig. 3(b) illustrates the centerline temperature proles of experimental and simulation result at two
different velocity conditions. The comparisons demonstrate that
an excellent agreement between simulation results and experimental tests is achieved. The maximum deviations of temperature
in Fig. 3(b) is 2.3% (inlet velocity 0.4 m/s) and 3.7% ( inlet velocity
0.6 m/s), respectively.

238

A. Tang et al. / Chemical Engineering Science 131 (2015) 235242

T/K
1800
1700
1600
1500
1400
1300
1200
1100
1000
900
800
700
600
500
400

1150

experimental data (0.4 m/s)

simulation result (0.4 m/s)
experimental data (0.6 m/s)
simulation result (0.6 m/s)

Wall temperature (K)

1100
1050
1000
950
900
850

6
8
10
12
14
Distance from inlet (mm)

16

18

Fig. 3. Experimental validation of computational model (equivalence ratio 1.0).

(a) Comparison of temperature distribution patterns on the combustor outer wall,
(left: experiment; right: simulation; inlet velocity: 0.6 m/s), (b) Comparison of
centerline temperature proles along the combustor outer wall.

3. Results and discussions

3.1. Combustion characteristics
To obtain the effect of hydrogen addition on methane combustion in the micro-channel, the cross section temperature distributions of the case with pure methane and the case with 10%
hydrogen addition are compared. Here, the addition ratio means
10% of methane is replaced by hydrogen with the same mass
fraction. Thus, the total fuel mass will remain constant. Comparing
to the methane combustion, more oxygen will be consumed in
hydrogen combustion of the same mass fraction. Therefore, the
original methaneair equivalence ratio is chosen to be 0.8 so as to
ensure full combustion of mixed fuel. As shown in Fig. 4(a), the
methane/air ame location is greatly inuenced by inlet velocity,
which is very similar to the 2D simulation results by Li et al. (2009).
At each case, the high temperature ame zone is attached to the
inner wall of combustor, and will be a little backward at the central
region of channel. The location of ame apparently moves toward
the outlet while mixture ow rate is increasing. The position of
maximum temperature ame is about 6 mm far from the inlet
when the inlet velocity is 0.4 m/s. But the high temperature ame
zone is very close to the outlet when the inlet velocity is 0.6 m/s. If
the inlet velocity is further increased, not only incomplete combustion but even total blow-off would happen. The result implies the
short ammability range of methane/air micro-combustion.
When 10% of hydrogen is added to the mixture, a great change
will take place in the micro-combustion process. As shown in

0.3 m/s

0.4 m/s

0.5 m/s

0.6 m/s

T/K
1900
1800
1700
1600
1500
1400
1300
1200
1100
1000
900
800
700
600
500
400

0.6 m/s

1.0 m/s

2.0 m/s

3.0 m/s

Fig. 4. Comparison of temperature distributions on combustor cross section at

different inlet velocities. (a) Pure methane (b) 10% hydrogen addition.

Fig. 4(b), the position of high temperature zone is obviously shifted

to the inlet when velocity is 0.6 m/s, and actually there is little
difference while the inlet velocity is less than 0.6 m/s. When it
comes to the case of 1 m/s, the high temperature ame zone
begins at the inner wall attachment point and is still close to the
inlet. And the position of ame front at the centerline is only 2 mm
away from the inlet. This phenomenon is very similar to the pure
hydrogen combustion (Pan et al., 2010; Yang et al., 2012). Therefore, it is concluded that the mixture will be ignited very rapidly
due to the highly combustible characteristic of hydrogen. And the
mixed fuel exhibits more combustion characteristics of hydrogen
when inlet velocity is relatively low.
As a result, the inlet velocity is further increased so as to examine
the ammability range of methane with hydrogen addition. It can be
seen from Fig. 4(b) that the shortening of residence time by
increasing inlet velocity has arose some difference in ame shape.
Although the reaction positions near the inner wall do not change
much with an increase in ow rate, there is an obvious phenomenon
that the ame of the central zone moves backward, and the ame
front becomes more and more sharp. This is due to the ame stretch
which is caused by longer warm-up time. The chemical reaction
takes place 4 mm away from inlet along the centerline for the 2 m/s
case, and 7 mm away from inlet for the 3 m/s case. It is very close to
the exit when the velocity reaches 4 m/s. These calculation data
illustrate that the ammability range has been expanded from
0.6 m/s to 4 m/s, which indicates a positive effect of stabilizing
ame location by 10% hydrogen addition.
A comparison on different addition ratios is conducted so as to
further analyze the effect of hydrogen addition. The ve addition

239

2400

5% hydrogen addition
10% hydrogen addition
15% hydrogen addition
20% hydrogen addition
25% hydrogen addition

Centerline temperature (K)

7.5
7.0
6.5
6.0
5.5
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0

2100
1800
1500

5% hydrogen addition
10% hydrogen addition
15% hydrogen addition
20% hydrogen addition
25% hydrogen addition

1200
900
600
300

6
8
10
12
14
Distance from inlet (mm)

16

18

Fig. 5. Kinetic rate of reaction-19 at different addition ratios (inlet velocity: 1.5 m/s).

ratios are 5%, 10%, 15%, 20% and 25%, respectively. It should be
pointed out that the mass fraction of mixed fuel and oxygen are
very close to the stoichiometric balance when hydrogen addition
ratio reaches 25%, so incomplete combustion would be caused if
more methane is replaced by hydrogen.
First, we compared the centreline chemical reaction rate of
different addition ratios (as shown in Fig. 5). Here we adopt the
reaction-19 (CH4 OH-CH3 H 2 ) for comparison which can provide some useful reference for methane reaction rate. When the
addition ratio is 5% and inlet velocity is 1.5 m/s, the mixture cannot
be ignited until arriving at the position of 10 mm away from inlet
along the centerline, and the reaction rate will reach the highest
value at 11 mm. However, the reaction will be closer to the inlet
zone obviously while the hydrogen addition ratio exceeds 10%, and
the locations of maximum reaction rate are only 4 mm (10%),
3 mm (15%), 1 mm (20%) and 0.5 mm (25%) away from the inlet,
respectively. In fact, the maximum velocity for the occurrence of
combustion is 2 m/s when the hydrogen addition ratio is 5%,
which is about only half as much as the 10% case. As a result, more
hydrogen should be added into the mixture so as to achieve an
effective improvement of ammability. Furthermore, the maximum rate of reaction-19 will be raised gradually along with the
increase of hydrogen fraction. The case with 25% hydrogen addition can reach 2.1 times as much as that of 5% case. The result
demonstrates that the addition of hydrogen also has an effective
impact on improving reaction rate of methane.
Fig. 6 shows the centerline temperature proles at different
addition ratios. It can be seen that the high temperature zone is a
little behind with the reaction position in the Fig. 5. Apart from the
5% case, all of the mixture can climb from normal temperature to
the highest temperature within a 5 mm distance, which is due to
the rapid chemical reaction. Then, the mixture temperature will
decline slowly alone the ow direction as a result of heat
absorption by the combustor wall. Meanwhile, it should be noted
that the maximum value of ame temperature has get a steady
growth with the increase of addition ratio. The adiabatic ame
temperature of methane and hydrogen with air are 2226 K and
2400 K, respectively (Turns, 2011), so this is an inevitable trend. In
addition, the growth of methane reaction rate also has an
important promoting effect on the ame temperature increasing.
At the ve hydrogen addition cases, the highest ame temperature
can be increased by about 94 K when 5% of the methane is
replaced by hydrogen. Thus, the maximum value of ame temperature can reach 2276 K at 25% case, which has already exceeded
the adiabatic ame temperature of methane.

6
8
10
12
14
Distance from inlet (mm)

16

18

Fig. 6. Centerline temperature proles at different addition ratios (inlet velocity:

1.5 m/s).

1400

Outer wall mean temperature (K)

Kinetic rate of reaction-19 (kgmol/m .s-1)

A. Tang et al. / Chemical Engineering Science 131 (2015) 235242

1300
1200
1100
1000
inlet velocity 0.6 m/s
inlet velocity 1.5 m/s
inlet velocity 2.0 m/s
inlet velocity 3.0 m/s

900
800

10
15
20
Hydrogen addition ratio (%)

25

Fig. 7. Outer wall mean temperature at different addition ratios and inlet
velocities.

3.2. Working performance of micro-combustor

For the micro-combustor of the MTPV system, besides the
acquisition of a steady combustion process, much more concern
should be paid to the outer wall temperature distribution and
radiant energy output. A comparison of the outer wall mean
temperature at different inlet velocities and addition ratios is
shown in Fig. 7. First, it should be noted that the outer wall
temperature of pure methane combustion case is only 868 K at
inlet velocity of 0.6 m/s. Obviously, the further thermophotovoltaic
conversion could hardly carry out due to such low temperature.
When 5% and 10% of methane are replaced by hydrogen, the mean
wall temperature can reach 980 K and 1039 K, which have shown
an apparent promoting effect. However, the growth trend will
become very slowly while more hydrogen is added into the
mixture, and the increment form 10% to 25% case is only 22 K.
As mentioned before, hydrogen will occupy a predominant position in the mixed fuel when inlet velocity is relatively low and
hydrogen addition ratio is larger than 10%. The fuel is ignited
immediately once entering the combustion channel, which lead to
little variation in the mean temperature of mixture. As a result, the
assistance of wall temperature increasing will become quite
limited. However, the characteristic of methane will be revealed
apparently when the mixture ow rate is further increased. The

240

A. Tang et al. / Chemical Engineering Science 131 (2015) 235242

T/K
1350
1300
1250
1200
1150
1100
1050
1000
950
900
850

Fig. 8. Comparison of temperature distributions on outer wall at different addition ratios (inlet velocity: 1.5 m/s). (a) 5% (b) 10% (c) 15% (d) 20% (e) 25%.

Total chemical energy

Radiant energy

200
175

Energy (W)

150
125
100
75
50
25
0

10

15
20
Hydrogen addition ratio (%)

25

Fig. 9. Total chemical energy and radiant energy at different addition ratios (inlet
velocity: 2 m/s).

40

Radiant energy / chemical energy (%)

reaction position gradually shifts forward according to the increasing addition ratio, and the mixture temperature also increases
continuously (as seen from Fig. 6). As a result, a linear growth
tendency of wall temperature can be found at each high velocity
case. The outer wall temperature of 10% addition case is higher
than 1200 K at inlet velocity of 2 m/s, which has reached the basic
working requirement of the MTPV system. When increasing the
hydrogen mass fraction from 10% to 25%, another 87 K (at 2 m/s
case) and 102 K (at 3 m/s case) temperature increments will be
gained. From this point of view, the positive effect of hydrogen
addition is very remarkable.
Fig. 8 shows the outer wall temperature distributions at different
addition ratios. The high temperature zone (T41200 K) of the four
cases are all situated at the front part of outer wall except for 5% case,
which keep in step with the high ame temperature locations in
Fig. 6. It can be predicted that the location of high temperature zone
at 10% case will also move backward when inlet velocity is further
increased. Meanwhile, as the hydrogen mass fraction increased, the
area of high temperature zone has presented an apparent growing
trend. At the 25% hydrogen addition case, 62% part of the outer wall
is higher than 1200 K, and the region above 1300 K occupies 1/3 of
the total wall area, which brings an average temperature of 1243 K to
the outer wall. However, the maximum temperature of 5% case is
only 1186 K, which also indicates that the hydrogen addition ratio
higher than 10% is favorable. Besides, the largest temperature
differences at the ve addition ratios are 339 K, 211 K, 251 K, 273 K
and 290 K respectively, so the minimum value is at the 10% case.
Thus, we can conclude that there is a close relationship between the
uniformity of wall temperature distribution and the mixture ignition
location, and the uniformity can be controlled by proper setting of
hydrogen addition ratio and inlet velocity. In fact, the promotion of
temperature distribution uniformity also has an important effect on
reducing thermal stress of surface, which is absolutely critical to
prolong the service life of micro-combustor.
According to the hydrogen addition plan, the total mass ow
rate will remain constant when inlet velocity is xed. However,
the total input chemical energy of the mixed fuel is bound to
change because of the heating value difference between methane
and hydrogen. Fig. 9 gives a comparison of input chemical energy
and the output radiant energy at four selected hydrogen addition
ratios. The high heating values of methane and hydrogen are
55.5 MJ/kg and 141.9 MJ/kg, respectively (Turns, 2011). So the total
chemical energy increases gradually with the increase of hydrogen
mass fraction. The total radiant energy has also got a steady
growth due to the increasing of wall mean temperature.
Fig. 10 shows the specic proportions of these two energies at
different inlet velocities and addition ratios. First, the proportion
presents a steady growth with the increase of hydrogen mass
fraction at the 2 m/s case. This indicates that the growth rate
of outer wall radiant energy is larger than that of input chemical energy. Therefore, it is suggested that the mass fraction of

35
30
25
20
15

inlet velocity 0.6 m/s

inlet velocity 1.5 m/s
inlet velocity 2.0 m/s
inlet velocity 3.0 m/s

10
5

10
15
20
Hydrogen addition ratio (%)

25

Fig. 10. Proportion of radiant energy to total chemical energy at different addition
ratios and inlet velocities.

hydrogen should be improved as much as possible if the oxidant is

enough. Good agreements have been shown at both 1.5 m/s and
3 m/s cases in Fig. 10, however, the variation trend is rather
different at a low inlet velocity case (e.g. for 0.6 m/s case). It can
be seen that the proportion of radiant energy to total chemical
energy will get a big increment from 0% to 10% hydrogen addition
ratio, which are 20.8% and 36.7% respectively. When the hydrogen
mass fraction exceeds 10%, the proportion goes into a slow decline

A. Tang et al. / Chemical Engineering Science 131 (2015) 235242

3.75

20
2 m/s

19

3.50

18

3.25

17

3.00
2.75

3 m/s

16

2.50

15

2.25

3 m/s

14

2.00

13

2 m/s

1.75

12

1.50

11

1.25

10

10

15

20

25

1.00

Usable radiant energy / chemical energy (%)

Usable radiant energy / radiant energy (%)

Hydrogen addition ratio (%)

Fig. 11. Proportions of usable radiant energy to total radiant energy and chemical
energy at different addition ratios.

0.50

1.75

1.50
0.40
1.25
0.35
1.00

0.75

System efficiency (%)

0.45
Electricity output (W)

rather than a further increasing. This phenomenon is mainly

caused by the low growth rate of wall temperature, which is
clearly shown in Fig. 7. Thus, like hydrogen addition ratio, the
mixture ow rate also should not be too low so as to ensure a good
operation effect. Besides, it should be noted that the proportion of
radiant energy to total chemical energy decreases evidently with
the increase of inlet velocity, while hydrogen ratio is at the same
condition. It follows that although the values of mean temperature
and output radiant energy will get a continuous increment by
rising mixture ow rate. However, the growing trend is signicantly lower than that of input chemical energy which presents a
linear relationship with the inlet velocity.
As shown in Fig. 9, the total radiant energy from the outer wall
reaches 35 W when the addition ratio is 10%. Actually only a small
portion of photons could be converted to electricity, whose
wavelength is shorter than the cut-off wavelength of the PV cells
(e.g. for GaSb cell, 1.78 m). To further study the improving effect
on working performance by hydrogen addition, detailed proportions of usable radiant energy to total radiant energy and chemical
energy are displayed in Fig. 11. First, these two proportions are also
going up with the increase of hydrogen addition ratio, which
shows similar trend as mean wall temperature, total radiant
energy as well as its proportion in chemical energy. Take the case
of 3 m/s for example, the values of usable radiant energy have
reached 5.0 W and 9.4 W at the 10% and 25% addition ratios, and
the proportion of usable radiant energy to total chemical energy
will rise from 1.95% to 3.05%. As a result, it is concluded that the
overall system efciency can be improved by 50% if other system
working conditions remain unchanged. Therefore, from the perspectives of both power output and system efciency, the positive
effects of hydrogen addition are very signicant. Meanwhile, it is
found that the two proportions present an opposite tendency with
the increase of inlet velocity. The proportion of usable radiant
energy to total radiant energy at each addition ratio can be raised
by about 2% from 2 m/s to 3 m/s, and the specic value has
reached 16% at 25% addition case. However, this growth can only
make a limited compensation to the adverse effect brought by the
slow increasing rate of total radiant energy. Thus, proportions of
usable radiant energy to total chemical energy at inlet velocity of
2 m/s are 2.2%, 2.7%, 3.0% and 3.3% respectively, which are 0.2%
higher than that of each 3 m/s calculation result on average.
Consequently, when burning mixed fuel such as methane and
hydrogen, although the merely increasing of mixture ow rate can
contribute to the boost of output power, it will not contribute to
the improvement of the system efciency.

241

0.30

10

15

20

25

0.25

Hydrogen addition ratio (%)

Fig. 12. Electricity output and system efciency at different addition ratios (inlet
velocity: 3 m/s).

Finally, based on the temperature predictions, the performance

of the MTPV system is estimated. It is assumed that two GaSb PV
cells are arranged parallel to the external wall of micro-combustor,
which has the same surface area. The distance between the PV cell
and external wall of micro-combustor is 1 mm. Meanwhile, it is
supposed that there is a good cooling system behind the PV cell.
That is, the temperature of the PV cell is kept 300 K and its
efciency is not changed. During the calculation process, view
factor between each elementary face of external wall and PV cell is
considered, and the detailed computational model can be found in
our previous work (Pan et al., 2010). Electricity output and
efciency of the MTPV system at four addition ratios are calculated, as shown in Fig. 12. It can be seen that both the electricity
output and system efciency improve signicantly with the
increase of hydrogen addition ratio. Under the same condition of
inlet velocity 3 m/s, the electricity output exceeds 1.48 W when
hydrogen addition ratio is 25%, which is nearly twice as much as
the 10% case. And due to the increase of wall temperature and
usable radiant energy, the system efciency reaches 0.48% at
25% case.
Furthermore, it should be noted that a great amount of radiant
energy from the outer wall cannot be used to photo-electric
conversion due to the cell bandgap. It not only creates great
wastage of energy, but also increases the cooling load of cells.
Therefore, besides intensication of micro-combustion process,
the successful development of lower bangap cell is also very
essential to the working performance improving of the MTPV
system.

4. Conclusions
In this paper, the effect of hydrogen addition on methane/air
combustion in a micro-planar combustor is investigated through
3D simulation method. Also, the radiation performance of the
micro-combustor at different addition ratios and ow rates are
analyzed and compared, which are absolutely crucial to the MTPV
system. Main conclusions are summarized as follows:
(1) Hydrogen addition has a signicant function on ame stabilizing, and the ammability range can be extended from 0.6 m/s
to 4 m/s when 10% of methane is replaced by hydrogen.
(2) The reaction rate of methane will be increased signicantly by
the burned hydrogen. Along with the increase of addition
ratio, the ame location apparently shifts toward the inlet and
the ame temperature gets a steady growth.

242

A. Tang et al. / Chemical Engineering Science 131 (2015) 235242

(3) The mean wall temperature is raised with the increase of

hydrogen mass fraction, resulting in an increment of radiant
energy and the proportion of usable radiant energy to total
chemical energy, as well as the electricity output and system
efciency.
(4) It is found that the hydrogen addition ratio should be higher
than 10% so as to ensure a good operation effect.

Acknowledgments
This work is supported by National Natural Science Foundation
of China (Nos. 51206066 and 51376082), China Postdoctoral
Science Foundation (No. 2014M551514), Natural Science Foundation of Jiangsu Province (No. BK20131253), Priority Academic
Program Development of Jiangsu Higher Education Institutions
(PAPD), and Scientic Research Starting Foundation for Advanced
Talents of Jiangsu University (No. 11JDG139).
Appendix A. Supporting information
Supplementary data associated with this article can be found in
the online version at http://dx.doi.org/10.1016/j.ces.2015.03.030.
References
Chia, L.C., Feng, B., 2007. The development of a micropower (micro-thermophotovoltaic) device. J. Power Sour. 165, 455480.
Chou, S.K., Yang, W.M., Chua, K.J., Li, J., Zhang, K.L., 2011. Development of micro
power generators: a review. Appl. Energy 88, 116.
Cuoci, A., Frassoldati, A., Faravelli, T., Ranzi, E., 2013. Extinction of laminar,
premixed, counter-ow methane/air ames under unsteady conditions: effect
of H2 addition. Chem. Eng. Sci. 93, 266276.
Federici, J.A., Vlachos, D.G., 2008. A computational uid dynamics study of propane/
air microame stability in a heat recirculation reactor. Combust. Flame 153,
258269.
Jiang, D.Y., Yang, W.M., Chua, K.J., Ouyang, J.Y., 2013a. Thermal performance of
micro-combustors with bafes for thermophotovoltaic system. Appl. Therm.
Eng. 61, 670677.
Jiang, D.Y., Yang, W.M., Chua, K.J., 2013b. Entropy generation analysis of H2/air
premixed ame in micro-combustors with heat recuperation. Chem. Eng. Sci.
98, 265272.
Ju, Y.G., Maruta, K., 2011. Microscale combustion: technology development and
fundamental research. Prog. Energy Combust. 37, 669715.
Khalil, A.E.E., Gupta, A.K., 2013. Hydrogen addition effects on high intensity
distributed combustion. Appl. Energy 104, 7178.
Kurdyumov, V.N., Jimenez, C., 2014. Propagation of symmetric and non-symmetric
premixed ames in narrow channels: inuence of conductive heat-losses.
Combust. Flame 161, 927936.
Li, J., Chou, S.K., Yang, W.M., Li, Z.W., 2009. A numerical study on premixed microcombustion of CH4-air mixture: effects of combustor size, geometry and
boundary conditions on ame temperature. Chem. Eng. J. 150, 213222.

Li, Y.H., Chen, G.B., Wu, F.H., Cheng, T.S., Chao, Y.C., 2012. Effects of catalyst
segmentation with cavities on combustion enhancement of blended fuels in a
micro-channel. Combust. Flame 159, 16441651.
Li, Y.H., Chen, G.B., Cheng, T.S., Yeh, Y.L., Chao, Y.C., 2013. Combustion characteristics
of a small-scale combustor with a percolated platinum emitter tube for
thermophotovoltaics. Energy 61, 150157.
Norton, D.G., Vlachos, D.G., 2003. Combustion characteristics and ame stability at
the microscale: a CFD study of premixed methane/air mixtures. Chem. Eng. Sci.
58, 48714882.
Norton, D.G., Vlachos, D.G., 2005. Hydrogen assisted self-ignition of propane/air
mixtures in catalytic microburners. Proc. Combust. Inst. 30, 24732480.
Pan, J.F., Yang, W.M., Tang, A.K., Chou, S.K., Duan, L., Li, X.C., Xue, H., 2010. Microcombustion in sub-millimeter channels for novel modular thermophotovoltaic
power generators. J. Micromech. Microeng. 20, 125021.
Park, J.H., So, J.S., Moon, H.J., Kwon, O.C., 2011. Measured and predicted performance of a micro-thermophotovoltaic device with a heat-recirculating microemitter. Int. J. Heat Mass Trans. 54, 10461054.
Park, J.H., Lee, S.I., Wua, H., Kwon, O.C., 2012. Thermophotovoltaic power conversion from a heat-recirculating micro-emitter. Int. J. Heat Mass Trans. 55,
48784885.
Sabia, P., Joannon, M.D., Fierro, S., Tregrossi, A., Cavaliere, A., 2007. Hydrogenenriched methane Mild Combustion in a well stirred reactor. Exp. Therm. Fluid
Sci. 31, 469475.
Seshadri, V., Kaisare, N.S., 2010. Simulation of hydrogen and hydrogen-assisted
propane ignition in pt catalyzed microchannel. Combust. Flame 157, 20512062.
Tacchina, S., Vella, L., Specchia, S., 2010. Catalytic combustion of CH4 and H2 into
micro-monoliths. Catal. Today 157, 440445.
Titova, N.S., Kuleshov, P.S., Favorskii, O.N., Starik, A.M., 2014. The features of ignition
and combustion of composite propane-hydrogen fuel: modeling study. Int. J.
Hydrog. Energy 39, 67646773.
Turns, S.R., 2011. An introduction to combustion: concepts and applications, third
ed. McGraw-Hill, New York.
Veeraragavan, A., Cadou, C.P., 2011. Flame speed predictions in planar micro-/
mesoscale combustors with conjugate heat transfer. Combust. Flame 158,
21782187.
Wan, J.L., Fan, A.W., Maruta, K., Yao, H., Liu, W., 2012. Experimental and numerical
investigation on combustion characteristics of premixed hydrogen/air ame in
a micro-combustor with a bluff body. Int. J. Hydrog. Energy 37, 1919019197.
Wang, J.H., Huang, Z.H., Tang, C.L., Miao, H.Y., Wang, X.B., 2009. Numerical study of
the effect of hydrogen addition on methaneair mixtures combustion. Int. J.
Hydrog. Energy 34, 10841096.
Yan, Y.F., Tang, W.M., Zhang, L., Pan, W.L., Yang, Z.Q., Chen, Y.R., Lin, J.Y., 2014.
Numerical simulation of the effect of hydrogen addition fraction on catalytic
micro-combustion characteristics of methane-air. Int. J. Hydrog. Energy 39,
18641873.
Yang, W.M., Chou, S.K., Shu, C., Li, Z.W., Xue, H., 2007. Experimental study of micro
thermophotovoltaic systems with different combustor congurations. Energy
Convers. Manag. 48, 12381244.
Yang, W.M., Chou, S.K., Chua, K.J., Li, J., Zhao, X., 2011. Research on modular microcombustor-radiator with and without porous media. Chem. Eng. J. 168, 799802.
Yang, W.M., Jiang, D.Y., Chou, S.K., Chua, K.J., Karthikeyan, K., An, H., 2012. Experimental study on micro modular combustor for micro-thermophotovoltaic system
application. Int. J. Hydrog. Energy 37, 95769583.
Yang, W.M., Chua, K.J., Pan, J.F., Jiang, D.Y., An, H., 2014. Development of microthermophotovoltaic power generator with heat recuperation. Energy Convers.
Manag. 78, 8187.
Zarvandi, J., Tabejamaat, S., Baigmohammadi, M., 2012. Numerical study of the
effects of heat transfer methods on CH4/(CH4 H2)-AIR pre-mixed ames in a
micro-stepped tube. Energy 44, 396409.