Fuel 225 (2018) 47–53

Contents lists available at ScienceDirect

Fuel
journal homepage: www.elsevier.com/locate/fuel

Full Length Article

An experimental and computational study of OH formation in laminar T

coflow syngas diffusion flames
⁎ ⁎
Huanhuan Xua,b, , Shaozeng Suna, Fengshan Liub, Yijun Zhaoa, Yuying Liuc, , Lei Chena,
Shun Mengd
a
Combustion Engineering Research Institute, School of Energy Science and Engineering, Harbin Institute of Technology, PR China
b
Measurement Science and Standards, National Research Council Canada, Building M-9, 1200 Montreal Road, Ottawa, ON K1A 0R6, Canada
c
School of Energy and Power Engineering, Beihang University, 37 Xueyuan Road, Haidian District, Beijing, PR China
d
School of Automotive and Transportation Engineering, Hefei University of Technology, Hefei, PR China

A R T I C LE I N FO A B S T R A C T

Keywords: This paper presents an experimental and numerical study to elucidate the suppressing influence of H2 content in
H2 content in syngas syngas on OH formation in laminar syngas diffusion flames burning in coflow oxidizer of 25%O2/75%H2O. Such
OH an influence was found different from the effect of H2 content in syngas on OH formation when the oxidizer
O2/H2O atmosphere stream consists of O2/N2. OH∗-chemiluminescence images were captured using an intensified CCD (ICCD)
Laminar coflow diffusion flame
camera, the centerline temperatures were measured by a Pt-6%Rh/Pt-30%Rh thermocouple, while a two-di-
mensional flame code was adopted to simulate these laminar coflow diffusion flames using a detailed reaction
mechanism. The different effects of H2O including chemical, thermal, transport, and radiative were examined
and detailed kinetics analyses were conducted to gain insights into how H2O in the oxidizer stream alters the
influence of H2 content in syngas on OH formation.

1. Introduction waste through gasification. In this steam/water oxy-fuel combustion

system, N2 in the conventional oxidizer (for example air) is completely
Syngas is a promising fuel for both civilian and industrial applica- replaced by H2O. The replacement of N2 by H2O affects the combustion
tions, especially under the circumstances of energy shortage and in- process through the following mechanisms: (1) H2O changes several
creasingly stringent emission regulations. H2 and CO are the major elementary reactions as a reactant, production or even through third-
components of syngas [1,2]. So far, many efforts have been made to body collisions; (2) the higher diffusivity of H2O, since H2O has a
study the effect of H2/CO ratio on the combustion characteristics of lighter molecular weight than N2; (3) the high specific heat of H2O than
syngas since the syngas composition varies significantly as a result of N2; and (4) the radiative property of H2O, which plays an important
different fuel feedstocks and gasification techniques [2]. It has been role in decreasing the combustion temperature. These mechanisms
found that H2 enlarges the flammability limit [3], enhances the flame correspond to the chemical, transport, thermal, and radiative effects of
stability [4], decreases the ignition delay time [5], and increases the H2O, respectively. It is consequently anticipated that the characteristics
laminar flame speed of syngas [6]. These findings indicate that the H2 of syngas/(O2/H2O) flames differ significantly from those of syngas
content is beneficial to improve the syngas combustion characteristics. flames burning in air. However, the existing studies were limited to
Among all the ways to utilize syngas in practice, a novel oxy-fuel syngas flames burning in air [1,4] or diluted air [3,5,6] by an inert gas,
combustion system [7,8] has been proposed in which syngas is oxidized such as CO2, N2 and He. For example, Santner et al. [10] observed that
by oxygen instead of air and the flame temperature is moderated using the mass burning rate of H2 and H2/CO (1:1) mixtures decreases
H2O, making the combustion occur in the atmosphere of O2/H2O. Such monotonically with water addition at pressure above 1 atm, and
application not only results in the merit that CO2 can be economically pointed out that the water addition influences the radical pool through
separated for capture and storage by simply condensing the steam in H2O + O = 2OH. Das et al. [11] found that the laminar flame speed of
the flue gas, but also provides an effective and environment-friendly syngas/air mixtures first increases then decreases with water addition
way to utilize the abundant low-quality fuels such as coal, biomass and for the carbon monoxide rich mixtures, and explained this trend as the

⁎
Corresponding authors at: School of Energy Science and Engineering, Harbin Institute of Technology, 92 West Dazhi Street, Harbin 150001, China (H. Xu). School of Energy and
Power Engineering, Beihang University, 37 Xueyuan Road, Haidian District, Beijing 100191, China (Y. Liu).
E-mail addresses: hhxuhit@gmail.com (H. Xu), yyliu@buaa.edu.cn (Y. Liu).

https://doi.org/10.1016/j.fuel.2018.03.144
Received 7 January 2018; Received in revised form 15 March 2018; Accepted 21 March 2018
Available online 02 April 2018
0016-2361/ © 2018 Elsevier Ltd. All rights reserved.
H. Xu et al. Fuel 225 (2018) 47–53

competition between the chemical and thermal effects of water addi- governing equations of continuity, momentum, energy, and species in
tion. Later, Singh et al. [12] observed the non-monotonic effect of H2O the low-Mach number limit formulation using the standard finite vo-
on the laminar flame speed of syngas/air mixture with a H2/CO ratio of lume method. The pressure and velocity coupling was dealt with the
5/95. It is important to point out that most studies about H2O addition SIMPLE algorithm. The thermal radiation transfer was considered and
conducted so far were performed in planar premixed flames [11–13] the radiative transfer equation was solved by using the discrete-ordi-
and counterflow diffusion flames [14,15], and effects of H2O addition nates method in axisymmetric cylindrical coordinates. The statistical
(including chemical, thermal, transport and radiative effects) on com- narrow-band correlated-K (SNBCK) radiation model was used to obtain
bustion characteristics, especially on multi-dimensional flames that are the absorption coefficients of CO, CO2, and H2O. The Soret effect was
more relevant to practical combustion systems, are still not well un- considered to take the preferential diffusion of the lightest species (H,
derstood. Moreover, the effects of H2O addition on syngas combustion H2, and He) into account. Further details of the flame code can be found
in O2/H2O atmosphere have not been studied. in [9]. Chemical reactions were modelled using the Davis et al. me-
Hydroxyl radical (OH) is usually considered as a kinetic indicator chanism [23], and the thermal and transport properties for gaseous
for the flame front and the concentration of OH has been found to species were calculated using CHEMKIN subroutines and the thermal
strongly relate to the combustion intensity, for example, in terms of the and transport databases of Davis et al. mechanism.
laminar flame speed [13]. Khan et al. [16] studied OH formation in The computational domain was 29.62 mm (z) × 23.08 mm (r). The
laminar coflow syngas/air diffusion flames and found that the max- inlet boundaries were placed at 5.08 mm below the burner exit to ac-
imum OH mass fraction increases with increasing the H2 percentage in count for the preheating effects. The presence of burner wall, i.e., the
syngas. Schefer et al. [17,18] pointed out that the addition of hydrogen fuel tube, was handled as a conjugate heat transfer problem to calculate
to the methane/air mixture leads to a significant increase in OH con- the heat transfer between the solid burner wall and the surrounding
centration and consequently extends the lean stability limits for a swirl- fluids. Symmetric boundary condition was used at the centerline. Slip
stabilized burner. The enhancing effect of H2 addition on OH con- condition was specified at the outer radial boundary, and zero gradient
centration was later confirmed by Khalil et al. [19] in both premixed conditions were stipulated at the outlet boundary. Non-slip and non-
and non-premixed flames of methane/air. Fu et al. [20] numerically catalytic conditions were imposed on the surfaces of the solid burner
found that the production of OH radical in laminar one-dimensional wall (fuel tube). Non-uniform meshes used in this study were the same
premixed syngas flames is promoted as the H2 fraction in the fresh as those in Ref. [9]. Very fine meshes were imposed near the burner exit
mixture is increased. It is apparent that H2 has an improving effect on with the smallest resolution of 0.04 mm (z) × 0.03825 mm (r). The
OH formation in both syngas and methane flames when air is the oxi- computational meshes were found sufficiently fine to produce grid-size
dizer. However, the influence of H2 content in fuel on OH formation in independent results.
syngas flames burning in an O2/H2O mixture has not been studied. Three syngas compositions were considered to cover a wide range of
In this study, we investigated OH formation in laminar syngas dif- H2/CO mole ratio as shown in Table 1, and the compositions of the
fusion flames of different H2-to-CO ratios burning in the coflow oxidizer oxidizer streams investigated in this study are also listed in Table 1.
containing O2/H2O or O2/N2 mixtures using OH∗-chemiluminescence Unless otherwise explicitly indicated, the oxidizer stream consists of a
imaging, thermocouple temperature measurements along the flame mixture of 25%O2 and 75%H2O by volume. All the investigations in this
centerline, and numerical simulations. 2D numerical simulations were work were performed at 1 atm with the inlet temperature of 400 K for
carried out using a detailed reaction mechanism to gain insights into both the fuel and oxidizer streams. The fuel flow rate is fixed at 200
how H2 content in syngas suppresses OH formation in laminar coflow SCCM. The corresponding mean velocity of the fuel stream at the
syngas diffusion flames when the oxidizer stream consists of an O2/H2O burner exit varies from 118.67 to 118.49 cm/s when the syngas com-
mixture. position changes, leading to a variation of the Re number from 13.32 to
76.04. The mean velocity of the oxidizer stream at the burner exit is
2. Experimental and simulation methods 24.87 cm/s and the Re number is 219.14.

Details of the experimental set-up used in this study have been de-
scribed in our recent publication [9]. The syngas flames were estab- 3. Results and discussion
lished on a coflow diffusion flame burner with inner diameters of
3.06 mm and 25.4 mm for the fuel tube and co-annular oxidizer tube, 3.1. OH formation in laminar coflow syngas diffusion flames
respectively. Bottled gases were used in the experiments with purities of
99.999%, 99.99%, 99.999% for CO, H2 and O2, respectively. The gas Fig. 1 shows the radial distributions of relative OH∗-chemilumi-
flow rates were controlled by mass flowmeters. A piston pump was used nescence intensities of laminar coflow flames of F1, F2 and F3 burning
to meter the liquid water to be evaporated in an electric furnace. The in O1. It is noticed that the high levels of noise along the flame cen-
resultant water vapor was first mixed with oxygen at the prescribed O2- terline are caused by the Abel inversion and should be neglected. It is
to-H2O ratio before being delivered to the burner. An intensified CCD interesting to observe that the OH concentration decreases with in-
(ICCD) camera (Andor Technology, iStar, 16 bits, 1024 × 1024 pixels, creasing the H2 content in syngas when the oxidizer is O1 (25%O2/75%
spectral response: 180–850 nm, delay: 130 ns, gain: 255, temporal gate: H2O). The suppressing effect of H2 on OH formation in this case was
40 ns; resolution: 0.050 mm/pixel) equipped with a bandpass filter also predicted numerically as shown in Fig. 2. Good qualitative
centered at 308 nm with a 10 nm bandwidth was applied to capture the
OH∗-chemiluminescence (averaged by one hundred shots). An Abel Table 1
inversion algorithm [21] was employed to convert the line-of-sight- Compositions of the fuel and oxidizer streams in the laminar coflow syngas diffusion
flames.
integrated OH∗-chemiluminescence image intensity to radially resolved
one. A Pt-6%Rh/Pt-30%Rh thermocouple (wire diameter: 0.1 mm, bead Name Composition (by volume)
diameter: ∼0.2 mm) was used to measure the centerline temperature of
the flames burning in the oxidizer of O2/H2O mixture. Thermal radia- Syngas stream F1 5%H2/95%CO,
F2 50%H2/50%CO
tion loss from the thermocouple bead was corrected for the measured
F3 95%H2/5%CO
temperatures, and the convective heat transfer coefficient was calcu-
Oxidizer stream O1 25%O2/75%H2O
lated by the Nusselt-Reynolds correlations [22] in the correction.
O2 25%O2/75%N2
Numerical simulations were conducted using a two-dimensional O3 21%O2/79%N2
axisymmetric cylindrical laminar flame code which solves the

48
H. Xu et al. Fuel 225 (2018) 47–53

(a) F1 (b) F2 (c) F3

Fig. 1. The radial distributions of relative OH -chemiluminescence intensities in laminar syngas flames burning in coflow O1.
*

2400
Max: 0.0098 Max: 0.0089 Max: 0.0086
Tmax: 2077.1 K Tmax: 1953.5 K Tmax: 1912.0 K 2200
OH
2000
0.0095
0.009 1800
0.0085
Temperature (K)
0.008 1600
0.0075
0.007 1400
0.0065
0.006
1200
Simulations Measurements
0.0055 1000
0.005 F1/O1 flame F1/O1 flame
0.0045 800 F2/O1 flame F2/O1 flame
0.004
F3/O1 flame F3/O1 flame
0.0035 600
0.003 F2/O2 flame
0.0025 400
0.002
0.0015 200
0.001 0 2 4 6 8 10 12 14 16
0.0005 HAB (mm)
Fig. 3. Comparison between the measured and computed temperature distributions along
the centerlines of syngas/O1 flames. Also plotted in this figure is the computed centerline
temperature in the F2/O2 flame.
(a) F1 (b) F2 (c) F3
Fig. 2. The computed OH mole fraction distribution of syngas flames burning in coflow increasing the H2 percentage in the fuel stream, in agreement with the
O1. The grey shade indicates the tapered fuel nozzle. effect of H2 content in syngas on Tmax shown in Fig. 2. In addition, the
numerical results show that the temperatures in syngas/O1 flame are
agreements are found between the measurements and the simulations much lower than those in syngas/O2 flame, which is attributed to the
that the OH concentration in the 5%H2/95%CO (F1) syngas flame is the higher specific heat of H2O than N2. Therefore, the thermal effect of
highest with the predicted maximum OH mole fraction (OHmax) being H2O decreases the flame temperature, consistent with the findings from
0.0098. The OH concentration is significantly reduced as the H2 content previous studies and the conclusion of our recent study [9]. For the sake
in syngas increases to 50% (F2), and the computed OHmax drops to of clarity, only the computed centerline temperature of F2/O2 flame is
0.0089. When the H2 content in syngas is further raised to 95% (F3), presented in Fig. 3. The thermal effect of H2O on the flame temperature
OH formation is only slightly more inhibited and the simulated OHmax and OH mole fraction will be quantified in Section 3.2.
is decreased to 0.0086 from 0.0089. In addition, both the measurements It is evident from Fig. 3 that although the numerical results correctly
and the simulations indicate a decrease in the flame height and an in- capture the variation of centerline temperatures with the fuel and
crease in the flame radius with increasing the H2 content. oxidizer compositions, at heights below the flame tip the measured
The computed maximum flame temperature, Tmax, in laminar co- temperatures are unexpectedly higher than the simulations. These dis-
flow flames of F1, F2 and F3 burning in O1 are also indicated in Fig. 2. crepancies between the measurement and simulation is likely due to the
It can be seen that Tmax decreases with increasing the H2 content in uncorrected effect of heat conduction along the thermocouple wire
syngas, and the decreasing trend of Tmax matches that of the maximum between the outer high-temperature reaction zone and the flame cen-
OH mole fractions, implying that the thermal condition is critical to OH terline at lower temperatures in the measurements. Despite these dis-
formation in syngas/O1 diffusion flames, which will be discussed in crepancies, both the measurements and simulations show that the
detail later. To further demonstrate the effect of H2 content in syngas on centerline temperature below the flame tip increases with H2 content in
the temperature in F1, F2 and F3 flames, Fig. 3 compares the experi- syngas. At flame heights above the flame tip, very good agreements
mental and computed temperature distributions along the centerlines of between the measurements and the simulations are observed with the
syngas/O1 diffusion flames. Both experimental and numerical results differences being less than 20 K. Based on observations that the flame
show that the maximum centerline temperature decreases with model is able to qualitatively predict the correct trends of OH and

49
H. Xu et al. Fuel 225 (2018) 47–53

0.010
0.014
0.008 Chemical effect
0.012 Thermal effect
Transport effect
0.006
Radiation effect
0.010
0.004

ΔOHmax
OHmax

0.008
0.002

0.006
O1 0.000

0.004 O2
O3 -0.002

0.002
-0.004

0 20 40 60 80 100
0 20 40 60 80 100
H 2 content in syngas (%) H2 content in syngas (%)
Fig. 4. The computed maximum OH mole fractions in laminar coflow syngas diffusion Fig. 5. The isolated effects of H2O substitution of N2 in the oxidizer on the maximum OH
flames burning in oxidizers of O1, O2 and O3. mole fraction in laminar coflow syngas diffusion flames burning in O1.

temperature variations below the flame tip with increasing the H2 of FH2O, expect that the thermal properties (such as the specific heat
content in syngas, Figs. 1–3, and the computed centerline temperatures capacity, Cp, enthalpy, h, etc.) of TH2O were specified as those of N2.
above the flame tip agree very well with the measurements, it is The differences between the results calculated for the oxidizer of 25%
therefore reasonable to state that the flame model can be used to in- O2/75%TH2O and those for the oxidizer of 25%O2/75%FH2O are at-
vestigate the mechanisms and pathways of the effects of H2 content in tributed to the thermal effect. The transport effect of H2O can be
syngas and the oxidizer compositions on OH formation in syngas dif- identified in the similar way by introducing a numerical species XH2O,
fusion flames under the current conditions. which has the same transport properties (such as the diffusion coeffi-
The maximum OH mole fractions (OHmax) in the three investigated cient, D0, thermal conductivity, λ, viscosity, μ, etc.) as N2 and the same
syngas flames burning in oxidizers of O1, O2 and O3 are compared in thermal and radiative properties as FH2O; hence the differences be-
Fig. 4. When the syngas burns in the oxidizer of O2/N2, i.e., O2 and O3, tween the results calculated for the oxidizer of 25%O2/75%XH2O and
OHmax increases with increasing the H2 content, regardless of the O2 those for the oxidizer of 25%O2/75%FH2O are attributed to the trans-
fraction in the oxidizer. This behavior agrees with that observed in port effect of H2O. The fourth artificial species, RH2O, was introduced
previous publications as discussed in Introduction. Besides, the syngas to identify the radiative effect of H2O. RH2O has the identical properties
flames burning in O2 produce much more OH radicals than those in O3, to those of FH2O, except that it is treated as a radiatively transparent
consistent with the finding of Mazas et al. [24] that the OH mole gas, i.e., RH2O does not participate in radiative heat transfer. The dif-
fraction in a one-dimensional stoichiometric methane/oxygen flame is ference between results calculated for the oxidizer of 25%O2/75%RH2O
higher than that in the methane/air flame. However, in the oxidizer of and those for the oxidizer of 25%O2/75%FH2O is due to the radiative
O2/H2O mixture, i.e., O1, OHmax decreases, though insignificantly, with effect of H2O.
increasing the H2 content, in contrast to the increasing trend of OHmax Fig. 5 shows that both the chemical and thermal effects of H2O
in syngas flames burning in an O2/N2 mixture. In addition, when the substitution of N2 in the oxidizer on OH formation are significant while
oxidizer is O1 (25%O2/75%H2O), F1produces a much higher OHmax the transport and radiative effects are much weaker and can be ne-
than when the oxidizer is O2 or O3 (mixtures of O2/N2). The opposite glected in comparison to the chemical and thermal effects, especially
trends of OH mole fraction variation with the H2 content in syngas the transport effect. The chemical effect of H2O enhances the OH for-
burning in oxidizers of O2/N2 and O2/H2O suggest that replacing N2 mation. The significant chemical effect of H2O in the F1 flame explains
with H2O alters the effect of H2 on OH formation. To understand the the much higher OHmax when the oxidizer is O1 shown in Fig. 4.
relative importance of different effects of N2 replacement by H2O in the However, the importance of H2O chemical effect decreases with in-
oxidizer, the isolated effects of H2O are examined next to identify the creasing the H2 content in syngas. On the other hand, all the other three
critical processes for the suppressing effect of H2 on OH formation in effects (thermal, transport, and radiative) of H2O inhibit OH formation
syngas flames burning in O2/H2O. and this suppression influence is stronger with increasing the H2 con-
tent in syngas for all the three effects, especially the thermal effect.
3.2. Different effects of H2O on OH formation Therefore, the weakened chemical effect of H2O substitution of N2 in
the oxidizer on promoting OH formation together with the enhanced
To clearly demonstrate the different effects of H2O substitution of thermal effect of H2O on suppressing OH formation with increasing the
N2 in the oxidizer of 25%O2/75%N2 on OH formation in syngas flames H2 content in syngas are responsible for the overall suppressing influ-
of different H2 contents, the chemical, thermal, transport and radiative ence of H2 in syngas on OH formation when the oxidizer is O1.
effects of H2O were evaluated, as compared in Fig. 5. The method used
to identify the isolated effects of H2O substitution of N2 in the oxidizer
(namely chemical, transport, thermal, and radiative) is the same as that 3.2.1. The thermal effect of H2O
used in Ref. [9]. To isolate the chemical effect of H2O, an artificial The thermal effect of H2O substitution of N2 in the oxidizer affects
species, FH2O, was introduced which does not participate in any che- the combustion process by reducing the temperature as a result of the
mical reactions but has the same thermal, transport and radiative greater specific heat of H2O than N2. To make this point clear, the
properties and third-body collision coefficients as those of H2O. The maximum temperatures (Tmax) in syngas flames of different H2 contents
chemical effect can be quantified as the differences between the results calculated using FH2O and TH2O in the oxidizer are shown in Fig. 6,
calculated for the oxidizer of 25%O2/75%H2O and those for the oxi- and the thermal effect of H2O on the maximum flame temperature
dizer of 25%O2/75%FH2O. To identify the thermal effect of H2O, all the (Tmax,TH2O – Tmax,FH2O) is also plotted. It can be seen that the thermal
properties of the second artificial species, TH2O, were the same as those effect lowers the flame temperature, as expected. As the H2 content in

50
H. Xu et al. Fuel 225 (2018) 47–53

2500 0.0020

2400
186 (a) Radii of OHmax
0.0015

OH production rate (mole/(cm s))

2300
184

Tmax,TH2O - Tmax,FH2O (K)

H+O2=O+OH (R1)

3
0.0010 O+H2=H+OH (R2)
2200
OH+H2=H+H2O (R3)
182
2100 0.0005 OH+OH=O+H2O (R4)
Tmax (K)

2000
180 0.0000
1900
178 -0.0005
1800 H+OH+M=H2O+M (R9)
FH2O
TH2O OH+OH(+M)=H2O2(+M) (R14)
1700 -0.0010
176 CO+OH=CO2+H (R28)
1600 HAB=1.57 mm CO+OH=CO2+H (R29)
0 20 40 60 80 100 -0.0015
H2 content in syngas (%) 0 1 2 3 4 5
R (mm)
0.0015
Fig. 6. The maximum flame temperatures (Tmax) and the thermal effect of H2O sub-
Radii of OHmax
stitution of N2 in the oxidizer, (Tmax,TH2O – Tmax,FH2O), in laminar coflow syngas diffusion (b)
flames burning in O1 for different H2 mole fractions in syngas. 0.0010

OH production rate (mole/(cm3s))

H+O2=O+OH (R1)
O+H2=H+OH (R2)
syngas increases from 5% to 95%, the reduction in the peak flame 0.0005 OH+H2=H+H2O (R3)
temperature by the thermal effect of H2Obecomes more pronounced OH+OH=O+H2O (R4)

from 176.6 K to 186.0 K. Consequently, the thermal effect of H2O sub- 0.0000
stitution of N2 in the oxidizer on the maximum OH mole fraction is
enhanced as the H2 content increases as a result of the reduced flame
-0.0005 H+OH+M=H2O+M (R9)
temperature, as shown in Fig. 5.
OH+OH(+M)=H2O2(+M) (R14)
CO+OH=CO2+H (R28)
-0.0010
CO+OH=CO2+H (R29)
3.2.2. The chemical effect of H2O
HAB=1.46 mm
A kinetics analysis of the OH production rates was performed for F1, -0.0015
F2 and F3 flames burning in coflow stream of O1 to identify the key 1 2 3 4 5
R (mm)
reactions contributing to OH formation. The radial distributions of OH
production rates of the most important reactions at a height above 0.0015 Radii of OHmax
(c)
flame base kernel (HAFBK) of 1.5 mm for the three investigated flames
0.0010 H+O2=O+OH (R1)
OH production rate (mole/(cm3s))

calculated using H2O and FH2O are shown in Fig. 7. It should be noted
O+H2=H+OH (R2)
that the flame base kernel is where the heat release rate maximizes. OH+H2=H+H2O (R3)
0.0005
Therefore, HAFBK = 1.5 mm corresponds to the actual height above OH+OH=O+H2O (R4)
burner (HAB) exit surface HAB = 1.57 mm, 1.46 mm and 1.03 mm in 0.0000
F1, F2 and F3 flames, respectively. The relevant reaction steps discussed
in this work are listed in Table 2. In Fig. 7, the contributions of all the -0.0005

key reactions to OH formation remain either positive or negative over H+OH+M=H2O+M (R9)
-0.0010 OH+OH(+M)=H2O2(+M) (R14)
the entire radial position range except R3 (OH + H2 = H + H2O),
CO+OH=CO2+H (R28)
which reacts backward to produce OH at small radii and then forward -0.0015 CO+OH=CO2+H (R29)
at larger radial locations to consume OH. The chemical effects of H2O
-0.0020 HAB=1.03 mm
on the peak OH production rates of one-direction proceeding reactions,
i.e., R1, R2, R4, R9, R14, R28 and R29, are summarized in Table 3, 1 2 3 4 5
which are calculated as the difference in the results of H2O and FH2O. R (mm)
The positive and negative values in Table 3 indicate that the chemical
Fig. 7. The radial distributions of OH production rates at HAFBK = 1.5 mm in flames of
effect of H2O enhances and suppresses OH formation, respectively. With (a) F1, (b) F2, and (c) F3 burning in O1. The solid lines are calculated using H2O, and the
increasing the H2 content in syngas, the chemical effect of H2O is en- dashed lines are for FH2O.
hanced through reactions R9, R14, R28 and R29, but inhibited by re-
actions R1, R2 and R4, Table 3. For example, when the H2 content in
clear that the chemical effect of H2O significantly promotes OH for-
syngas is increased from 5% to 95%, the chemical effect of H2O is en-
mation through the backward reaction of R3 (OH + H2 = H + H2O) on
hanced from −1.23E−3 to −3.68E−5 mol/(cm3s) by reactions of R9,
the fuel side in the F1 flame, Fig. 7(a), and this influence is weakened as
R14, R28 and R29, but decreased from 1.07E−3 to −8.59E−5 mol/
the H2 content in syngas increases, Fig. 7(b), and eventually becomes
(cm3s) by R1, R2 and R4; since the former enhancement
negligible in the F3 flame, Fig. 7(c). On the other hand, the chemical
(1.19E−3 = −3.68E−5 − −1.23E−3 mol/(cm3s)) is stronger than
effect of H2O through the forward reaction of R3 near the flame front
the latter inhibition (−1.16E−3 mol/(cm3s)), it can therefore be con-
(the location of the maximum OH mole fraction) noticeably promotes
cluded that the chemical effect of H2O through reactions of R1, R2, R4,
OH consumption in the F1 flame, Fig. 7(a), but shifts to inhibit OH
R9, R14, R28 and R29 is enhanced with increasing the H2 content in
consumption in the F3 flame, Fig. 7(c). Therefore, the chemical effect of
syngas. As such, these one-direction proceeding reactions are not re-
H2O on OH consumption through the forward reaction of R3 weakens
sponsible for the reduced OH formation with increasing the H2 content,
with increasing the H2 content in syngas. As a result, the weakened
making reaction R3, OH + H2 = H + H2O, the potentially dominant
chemical effect of H2O substitution on OH formation with increasing
cause for the decrease of OHmax.
the H2 content in syngas is only attributed to the suppressed backward
Fig. 7 shows that as the H2 content in syngas increases, the back-
reaction of R3. In fact, R3 proceeds predominately in the forward di-
ward and forward reaction rate of R3 at small radii and larger radial
rection to consume OH as the H2 content in syngas is increased.
regions is respectively suppressed and enhanced. On one hand, it is

51
H. Xu et al. Fuel 225 (2018) 47–53

Table 2
Summary of the reaction steps.

Reaction No. Reaction steps Reaction No. Reaction steps

R1 H + O2 = O + OH R2 O + H2 = H + OH
R3 OH + H2 = H + H2O R4 OH + OH = O + H2O
R5 H + H + M = H2 + M R6 H + H + H2 = H2 + H2
R7 H + H + H2O = H2 + H2O R8 H + H + CO2 = H2 + CO2
R9 H + OH + M = H2O + M R10 O + H + M = OH + M
R11 O + O + M = O2 + M R12 H + O2(+M) = HO2(+M)
R13 H2 +O2 = HO2 + H R14 OH + OH(+M) = H2O2(+M)
R15 HO2 + H = O + H2O R16 HO2 +H = OH + OH
R17 HO2 + O = OH + O2 R18 HO2 + OH = O2 + H2O
R19 HO2 + OH = O2 + H2O R20 HO2 + HO2 = O2 + H2O2
R21 HO2 + HO2 = O2 + H2O2 R22 H2O2 + H = HO2 + H2
R23 H2O2 + H = OH + H2O R24 H2O2 + O = OH + HO2
R25 H2O2 + OH = HO2 + H2O R26 H2O2 + OH = HO2 + H2O
R27 CO + O(+M) = CO2(+M) R28/R29 CO + OH = CO2 + H

Table 3 H2O.
Chemical effects of H2O substitution of N2 in oxidizer (difference between the rates cal-
culated for H2O and FH2O) on the peak OH production rates (mole/(cm3s)) at
Acknowledgments
HAFBK = 1.5 mm in syngas/O1 flames.

F1 F2 F3 This work was supported by the National Natural Science

Foundation of China (Grant No. 51476048), the Foundation for
R1 2.22E−4 −3.10E−6 −1.40E−4 Innovative Research Groups of the National Natural Science Foundation
R2 −4.70E−5 −2.00E−4 −2.60E−4
R4 8.94E−4 5.18E−4 3.10E−4
of China (Grant No. 51421063) and the Collaborative Innovation
R1 + R2 + R4 1.07E−3 3.11E−4 −8.59E−5 Center of Clean Coal Power Plant with Poly-generation.
R9 −2.30E−4 −2.10E−5 5.38E−5
R14 −2.10E−4 −1.50E−4 −1.10E−4 Reference
R28 −4.90E−4 2.05E−5 2.12E−5
R29 −2.90E−4 −4.90E−5 2.40E−6
R9 + R14 + R28 + R29 −1.23E−3 −1.99E−4 −3.68E−5 [1] Zhang Y. Laminar flame speed of lean premixed H2/CO/air flames. Combust Flame
2014;161:2492–5.
[2] Thakre P, Ribert G, Wang Z, Yetter RA, Yang V. Fundamental Combustion
Characteristics of Syngas, 2010.
4. Conclusions [3] Shih HY, Hsu JR, Lin YH. Computed flammability limits of opposed-jet H2/CO
syngas diffusion flames. Int J Hydrogen Energy 2014;39:3459–68.
[4] Hwang J, Bouvet N, Sohn K, Yoon Y. Stability characteristics of non-premixed
In this study, OH formation characteristics in laminar coflow syngas turbulent jet flames of hydrogen and syngas blends with coaxial air. Int J Hydrogen
flames of 5%H2/95%CO, 50%H2/50%CO, and 95%H2/5%CO burning Energy 2013;38:5139–49.
in different oxidizers containing O2/N2 and O2/H2O mixtures were in- [5] Lee HC, Jiang LY, Mohamad AA. A review on the laminar flame speed and ignition
vestigated. Experimentally, OH∗-chemiluminescence images were cap- delay time of syngas mixtures. Int J Hydrogen Energy 2014;39:1105–21.
[6] Wang J, Huang Z, Kobayashi H, Ogami Y. Laminar burning velocities and flame
tured using an intensified CCD (ICCD) camera. A Pt-6%Rh/Pt-30%Rh characteristics of CO–H2–CO2–O2 mixtures. Int J Hydrogen Energy
thermocouple was used to measure temperatures along the flame cen- 2012;37:19158–67.
[7] Anderson R, Viteri F, Hollis R, Keating A, Shipper J, Merrill G, et al. Power for land.
terline. Numerically, a two-dimensional flame code incorporating de-
Sea Air 2010;2010:733–43.
tailed reaction mechanism, the preheating effect, non-grey radiation [8] Sun S, Meng S, Zhao Y, Xu H, Guo Y, Qin Y. Experimental and theoretical studies of
heat transfer, and complex thermal and transport properties was em- laminar flame speed of CO/H2 in O2/H2O atmosphere. Int J Hydrogen Energy
ployed. Using this flame code, the four different effects of H2O sub- 2016;41:3272–83.
[9] Xu H, Liu F, Sun S, Zhao Y, Meng S, Tang W. Effects of H2O and CO2 diluted oxidizer
stitution of N2 in the O2/N2 oxidizer stream, namely chemical, thermal, on the structure and shape of laminar coflow syngas diffusion flames. Combust
transport, and radiative, were isolated and quantified. Based on the Flame 2017;177:67–78.
experimental and numerical studies, the main conclusions can be drawn [10] Santner J, Dryer F, Ju Y. Effect of water content on syngas combustion at elevated
pressure. In: 50th AIAA Aerospace Science Meeting, Nashville, TN, January 2012. p.
as follows: 9–12.
[11] Das AK, Kumar K, Sung CJ. Laminar flame speeds of moist syngas mixtures.
(1) Both measurements and simulations demonstrated that the OH Combust Flame 2011;158(2):345–53.
[12] Singh D, Nishiie T, Tanvir S, Li Q. An experimental and kinetic study of syngas/air
mole fraction decreases with increasing the H2 content in syngas combustion at elevated temperatures and the effect of water addition. Fuel
when the flames burn in the oxidizer consisting of 25%O2/75% 2012;94(1):448–56.
H2O. This behavior is in contrast to the increasing OH mole fraction [13] Yamamoto K, Ozeki M, Hayashi N, Yamashita H. Burning velocity and OH con-
centration in premixed combustion. Proc Combust Inst 2009;32:1227–35.
with increasing the H2 content when the syngas fuel burns in the
[14] Lee CE, Lee SR, Han JW, Park J. Numerical study on effect of CO2 addition in flame
oxidizer consisting of N2/O2. structure and NOx formation of CH4-air counterflow diffusion flames. Int J Energy
(2) The chemical and thermal effects of H2O substitution of N2 in the Res 2001;25(4):343–54.
oxidizer on OH formation are significant while the transport and [15] Park J, Kim SG, Lee KM, Kim TK. Chemical effect of diluents on flame structure and
NO emission characteristic in methane-air counterflow diffusion flame. Int J Energy
radiative effects are much weaker and can be neglected under the Res 2002;26(13):1141–60.
conditions of this study. [16] Khan N, Raghavan V. Structure and reaction zones of hydrogen–carbon-monoxide
(3) As the H2 content in syngas increases, both the weakened chemical laminar jet diffusion flames. Int J Hydrogen Energy 2014;39:19832–45.
[17] Schefer RW, Wicksall DM, Agrawal AK. Combustion of hydrogen-enriched methane
effect of H2O through the backward reaction of in a lean premixed swirl-stabilized burner. Proc Combust Inst 2002;29:843–51.
OH + H2 = H + H2O on promoting OH formation and the en- [18] Schefer RW. Hydrogen enrichment for improved lean flame stability. Int J
hanced thermal effect of H2O on suppressing OH formation are Hydrogen Energy 2003;28:1131–41.
[19] Gupta AK, Khalil AEE, Yu KH. Hydrogen addition effects on high intensity dis-
responsible for the decrease of OH mole fraction with the H2 con- tributed combustion. Appl Energy 2013;104:71–8.
tent in syngas when the oxidizer stream consists of 25%O2/75% [20] Fu J, Tang C, Jin W, Thi LD, Huang Z, Zhang Y. Study on laminar flame speed and

52
H. Xu et al. Fuel 225 (2018) 47–53

flame structure of syngas with varied compositions using OH-PLIF and spectro- [23] Davis SG, Joshi AV, Wang H, Egolfopoulos F. An optimized kinetic model of H2/CO
graph. Int J Hydrogen Energy 2013;38:1636–43. combustion. Proc Combust Inst 2005;30:1283–92.
[21] Pretzier G. A new method for numerical abel-inversion. Z Naturforsch, A [24] Mazas AN, Fiorina B, Lacoste DA, Schuller T. Effects of water vapor addition on the
2014;46(7):639–41. laminar burning velocity of oxygen-enriched methane flames. Combust Flame
[22] Drysdale D. An introduction to fire dynamics. John Wiley & Sons; 2011. 2011;158:2428–40.

53