CHEMICAL ENGINEERINGTRANSACTIONS 
 

VOL. 61, 2017 

A publication of 

 

The Italian Association 
of Chemical Engineering 
Online at www.aidic.it/cet 

Guest Editors:Petar SVarbanov, Rongxin Su, Hon Loong Lam, Xia Liu, Jiří J Klemeš 
Copyright © 2017, AIDIC Servizi S.r.l. 

ISBN978-88-95608-51-8; ISSN 2283-9216 

Experimental and Numerical Study of Oxygen Diluted Partially 

Premixed Dimethyl Ether/ Methane Counter Flow Flame 

Jing Luo*, Weibo Li, Liang Tian, Liansheng Liu 

School of Energy and Environmental Engineering, Hebei University of Technology, Tianjin, 300401, China   

luojing@hebut.edu.cn 

The oxygen diluted partially premixed dimethyl ether/methane counter flow flames were discussed in this paper. 

Flame images and flame structures were obtained through experiments and numerical simulation. Flame 

propagation velocity of premixed fuel and emissions were investigated over wide range of DME blending ratios. 

The methane / dimethyl ether ratio is varied from 1 to 0, while the premixed equivalence ratios of fuel-side and 

oxygen concentrations are unchanged. In order to exclude the influence of strain rate, the velocities of both 

sides keep invariable in the experiments and simulation processes and the strain rate are fixed at 49.2 s-1. The 

results show that the premixed flames are much brighter and flame surfaces are becoming thicker with the 

increase of DME blending ratios, in addition, the flame surface gradually moves to the premixed side and the 

spacing between the double flames is increased. With the increase of the proportion of DME, the flame 

propagation velocities gradually increase, but they are not linearly related, that is the growth rate of flame 

propagation velocity gradually slowed down. For the oxygen diluted partially premixed flame, OH radicals have 

two peaks in the whole combustion zone, and they are consistent with the two temperature peaks. About two 

thirds OH radicals are produced in the premixed combustion zone. With the increase of DME blending ratio, 

temperature distributions are higher and the raised flame temperatures accelerate the produce of thermal 

nitrogen oxide, resulting in increase of NO emissions. 

1. Introduction 

In the last two decades years, a lot of worldwide researchers investigated the combustion characteristics and 

performance of alternative and renewable fuels. Among these, Dimethyl ether (DME) has been regarded as a 

promising substitute not only because it is the simplest ether, but also because it has no carbon-carbon bond 

and high oxygen content, making it soot-free combustion (Yu et al., 2014). 

Kaiser et al. (2000) investigated the species profiles of two different equivalence ratios DME–air flat flames 

experimentally and numerically using the mechanism of Fischeret et al. (2000). Results showed that the CH2O 

mole fractions were 5 – 10 times larger in DME flames when compared to methane flames of similar equivalence 

ratios. Zheng et al. (2005) studied the ignition of nitrogen-diluted DME by counter flowing it against heated air 

at pressures from 1.5 to 3 atm. 

DME is also a good additive. Experiments and simulations were performed to investigate partially premixed 

combustion using different types of fuels with DME, including methane (Lowry et al., 2011), methanol (Chen et 

al., 2009), n-butane and propane (Lee et al., 2011). Among these, the most widely used blending fuel is 

DME/methane binary fuel because of potential low emissions. Chen et al. (2007) experimentally and numerically 

studied the effect of DME addition in methane/air mixtures. He discovered that DME significantly decreased the 

high temperature ignition delay and increased the flame speed. Lowry et al. (2011) investigated the laminar 

burning characteristics of DME/methane binary fuel at 20 % and 40 % DME mixing ratios and various initial 

pressures. Their results indicated that the flame speed was increased with DME addition. 

There are many ways to study the flame structure and characteristics of laminar flame, including standard tube, 

constant volume, counter flow flame, plane flame and Bunsen lamp method. The counter flow flame method is 

used most widely because of its stability of flame surface, the same degree of stretch rate, easy to post-

measurement and analysis. In this paper, the oxygen diluted partially premixed dimethyl ether/methane counter 

                               
 
 

 

 
   

                                                  
DOI: 10.3303/CET1761067

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Please cite this article as: Luo J., Li W., Tian L., Liu L., 2017, Experimental and numerical study of oxygen diluted partially premixed dimethyl 
ether/ methane counter flow flame, Chemical Engineering Transactions, 61, 415-420  DOI:10.3303/CET1761067  

415



flow flames were used to investigate flame propagation velocity and emissions over wide range of DME blending 

ratios.  

2. Experimental and computational approaches 

2.1  Experimental apparatus 

In the counter flow experiments, the flow equalization of nozzle outlet flow is the basic of the experiments, and 

the nozzle design has a direct impact on the flow equalization effect. Therefore, the nozzle design and selection 

is a very important part. Figura and Gomez (2014) used the shrink round nozzle, as the counter flow device, 

finding that flow equalization of shrink nozzle is better in the center of the flow, but it is easy to cause the flame 

overflow in the outside of the nozzle. Lockett et al. (1999) found that, with the circular nozzle, the centre of 

premixed flame would be raised when the equivalence ratio was low, and the flame surface would not be one-

dimensional. Kim et al. (2014) changed the conventional circular nozzle structure, with using a rectangular 

nozzle to build a linear counter flow experimental device. He found that the flames of various conditions can 

show satisfied quasi-one-dimensional characteristics. Compared with the circular nozzle, the rectangular one 

can be used in the wider equivalence ratio range, so the rectangular nozzles are used as the counter flow device 

in this paper. 

The line-oriented counter flow flame experimental device is shown in Figure 1, the nozzle outlet width 30 mm, 

height 10 mm, relative distance 30 mm. In order to achieve a better flow equalization effect, a filter and a 

honeycomb ceramic are placed in the concave of the inlet downstream. The device was covered with quartz 

glass, using of dimethyl silicone oil to produce adsorption, to ensure the sealing requirements. 

 

  

Figure 1: Schematic diagram of the experimental apparatus 

2.2 Experimental methodology and operational parameter 

The stretch rate is the velocity gradient in the vertical direction of the flame surface, and it has a large effect on 

the flame pattern. Therefore, the velocity of both sides is fixed in this paper to ensure that the stretch rate is 

constant, which is expressed by the Eq(1). Here Uf and Uo are the velocity on each side, and d is the separation 

distance between the nozzles. In the experiment, the methane / dimethyl ether ratio is varied from 1 to 0, while 

the premixed equivalence ratios of fuel-side and oxygen concentrations are unchanged. The experimental 

conditions are shown in Table 1, where subscript o indicates the oxidizer-side, and subscript f indicates the fuel-

side. 

Table 1: Experimental conditions 

case Methane/ 

dimethylether 

Equivalence 

ratioΦf 

O2f 

(%) 

Uf 

(m/s) 

O2O 

(%) 

UO 

(m/s） 
Nozzle spacing 

d (mm) 

Stretch rate 

K (s-1) 

1 1:0  1.1   18 1.068    26 0.408         30        49.2 

2 

3 

4 

3:2 

2:3 

0:1 

1.1 

1.1 

1.1 

  18 

  18 

  18 

1.068 

1.068 

1.068 

   26 

   26 

   26 

0.408 

0.408 

0.408 

        30 

        30 

        30 

       49.2 

       49.2 

       49.2 

416

http://dict.cn/flow%20equalization
http://dict.cn/flow%20equalization
http://dict.cn/flow%20equalization


f
K U U d

o
= ( + ) /  (1) 

2.3 Experimental images 

The flame images, shown in Table 2, are measured with Canon EOS 5D Mark II DSLR camera. The camera 

lens is 1 meter away from the flame. The exposure time is set to 1/5 seconds, and the position of the camera 

bracket is fixed at each case. The flame temperature profile is measured with B type thermocouple. The 

thermocouple probe diameter is 0.4 mm and it can be used under 2,200 K with measurement accuracy at 0.5 %t. 

It is shown clearly that all flames have distinct double flame structures. It is worth noting that with different DME 

blending ratios, their flame structures have obvious difference. With the increase of DME blending ratios, the 

upper part flame (fuel- side) is much brighter and flame surfaces are becoming thicker, in addition, the flame 

surface gradually moves to the premixed side and the spacing between the double flames is increased. From 

the CH radical distributions, it can be seen that, the methane pyrolyzes completely in the nonpremixed 

combustion zone when pure methane as a fuel, and CH radicals increased gradually with the increase of DME 

blending ratios. 

Table 2: Flame images and CH distribution at different fuel ratios 

case 
Methane/ 

dimethylether 
Flame images CH distribution 

1 1:0 

  

2 3:2 

  

3 2:3 

  

4 0:1 

  

 

2.4 Experimental methodology and operational parameter 

Computations were performed with the opposed flow flame code, OPPDIF, using Zheng mechanism with the 

recommended multi component transport and thermo chemical database. The boundary conditions at the inlets 

included measured inlet temperature, velocity, and the species flux, shown in Table1. Figure 2 shows predicted 

temperature distributions and comparisons of predicted temperature with the corrected measured data. Since 

the deviation caused by radiative losses should be considered, temperature corrections can be calculated 

according to Eq(2), which is deduced from the heat balance equation of thermocouple junction. Here Tmc, Tm, 

Tw are corrected measured, measured, and ambient temperature. ε is the thermocouple surface emissivity. σ is 

Stefan-Boltzmann constant. Dc is the probe diameter. Kf is the thermal conductivity of flue gas. 

 
 4 4

0.5 0.33

e
2 0.6 R Pr

c

mc m m w

f f

D
T T T T

K


  

   
 

 

(2) 

417

javascript:void(0);
javascript:void(0);


It is shown that the simulation results predict the trend of combustion temperatures well. However, the measured 

highest temperature is a little lower than the simulation results. The reason is that amount of heat dissipation 

was carried away by nitrogen and quartz glass, which is not considered in the simulation.  

 

 

Figure 2: Temperature distributions 

3.  Results and discussions 

3.1 Flame propagation velocity  

In the case of pure methane combustion, for example, as shown in Figure 3(a), when the gas flows out of the 

nozzle, the gas flow rate decreases continuously due to the expansion of the flow field, and the axial velocity 

gradually increases when the flow rate decreases to a certain value.  

From this minimum velocity, the methane and oxygen concentrations begin to decrease, demonstrating that the 

premixed gas enters the premixed combustion zone where it begins to burn, at which time the airflow velocity 

is increased. At this point the air velocity and the flame propagation velocity are balanced, so the minimum 

velocity of the airflow is the flame propagation velocity here SL, and the local stretch rate at the upstream is Ki. 

The flame propagation velocity at different methane / dimethyl ether blending ratios is shown in Figure 3(b). It 

can be found that with the increase of the proportion of DME, the flame propagation velocity gradually increases, 

but they are not linearly related. With the increase of DME blending ratio, the growth rate of flame propagation 

velocity gradually slowed down. 

 
 (a) distribution of axial velocity  (b) flame propagation velocities at different fuel 

Figure 3: Flame propagation velocities 

3.2 Flame structure 

Figure 4 shows the simulated fuel consumption and some important chemical species profiles (O2, CO2, CO, 

H2O, OH, O) with different DME blending ratios. They have distinct double flame structures. Some intermediate 

products (e.g. CO) are generated in the fuel-rich side and the position of CO concentration peak is consistent 

 

0

500

1000

1500

2000

2500

30 25 20 15 10 5 0 

T
e

m
p

e
ra

tu
re

 K

Distance mm 

predicted case1

predicted case2

predicted case3

predicted case4

 masured case1

 

 

 

0 5 10 15 20 25 30

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

(case1)Distance (mm)

A
x
ia

l 
v
e

lo
c
it
y
 U

 (
m

/s
)

 Axial velocity

 O
2

 CH
4

0.00

0.05

0.10

0.15

0.20

0.25

c
o

n
c
e

n
ra

ti
o

n

 

S
L

 

1:0 4:1 3:2 2:3 1:4 0:1

26

28

30

32

34

36

38

F
la

m
e

 p
ro

p
a

g
a

ti
o

n
 v

e
lo

c
it
y
 S

L
 (

m
/s

)

CH4： DME

 

418



with that of temperature peak in the premixed combustion zone. OH radicals have two peaks in the whole 

combustion zone, and they are consistent with the two temperature peaks. About two thirds OH radicals are 

produced in the premixed combustion zone.  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

                                                     

Figure 4: Flame structures 

3.3 NO emission 

The distribution of NO concentration at different methane / dimethyl ether blending ratios is shown in Figure 

5(a). It can be found that with the increase of the proportion of DME, the peak of NO concentration increases 

gradually, and the range of NO formation field increases too. The NO concentration peak appears at the peak 

of the temperature. In addition, the NO concentration in the premixed combustion zone is much lower than in 

the nonpremixed combustion zone, the reason is that the oxygen concentration of the oxidizer-side is 26 %, 

much more than 21 %. 

In this paper, in order to ensure the same flow rate and equivalence ratio, the fuel and calorific value are different 

with different methane / dimethylether blending ratios. The NO concentrations per calorific value are compared 

in Figure 5(b), and it can be found that with the increase of the proportion of DME, NO emissions are still 

increased. 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 5: NO concentration at different fuel ratios 

4. Conclusions 

Experimental and simulation work had been conducted to study one-dimensional oxygen diluted partially 

premixed dimethyl ether/methane counter flow flames in this paper. Flame propagation velocity of premixed fuel 

and emissions were investigated over wide range of DME blending ratios. In order to exclude the influence of 

strain rate, the velocities of both sides keep invariable in the experiments and simulation processes. The main 

results can be summarized as follows. 

0.0

5.0E-6

1.0E-5

1.5E-5

2.0E-5

2.5E-5

302520151050

N
O

 C
o

n
c
e

n
tr

a
ti
o
n

Distance (mm)

 case1

 case2

 case3

 case4

 

 

 

(a) NO concentration distributions (b)  Total NO concentration per calorific value 

0.00E+000

5.00E-010

1.00E-009

1.50E-009

2.00E-009

2.50E-009

3.00E-009

case4case3case2case1

N
O

 C
o

n
c
e

n
tr

a
ti
o
n

 p
e

r 
c
a

lo
ri
fi
c
 v

a
lu

e
  
 k

J
-1

 

 

0.00

0.05

0.10

0.15

0.20

0.25

0.30

302520151050
Distance (mm)

C
o

n
c
e

n
tr

a
ti
o

n

(b) case3

 CH
4

 CH
3
OCH

3

 O
2

 CO
2

 H
2
O

OH*10

 O*50

 

 

0.00

0.05

0.10

0.15

0.20

0.25

0.30

 CH
4

 CH
3
OCH

3

 O
2

 CO
2

 H
2
O

OH*10

 O*50

302520151050

Distance (mm)

(a) case 2

C
o

n
c
e

n
tr

a
ti
o

n

 

419



(1) With the increase of DME blending ratios, the premixed flames are much brighter and flame surfaces are 

becoming thicker, in addition, the flame surface gradually moves to the premixed side and the spacing 

between the double flames is increased.  

(2) With the increase of DME blending ratios, the flame propagation velocity increases, while the growth rate 

of flame propagation velocity gradually slowed down. 

(3) OH radicals have two peaks in the whole combustion zone, and they are consistent with the two 

temperature peaks. About two thirds OH radicals are produced in the premixed combustion zone.  

(4) With the increase of the proportion of DME, NO emission per calorific value increases gradually, and the 

range of NO formation field increases too. 

Acknowledgments  

This work was funded by Natural fund of Hebei Province (E2015202321) and Science and technology research 

project of Hebei Province (QN2015063). 

References 

Chen Z., Qin X., Ju Y., Zhao Z., Chaos M., Dryer F.L., 2007, High temperature ignition and combustion 

enhancement by dimethyl ether addition to methane–airmixtures, Proceedings of the Combustion Institute, 

31, 1215–1222. 

Chen Z., Yao M., Zheng Z., Zhang Q., 2009, Experimental and numerical study of methanol/dimethyl ether dual-

fuel compound combustion, Energy and Fuels, 23, 2719–2730. 

Figura L., Gomez A., 2014, Structure of incipiently sooting ethylene–nitrogen counterflow diffusion flames at 

high pressures. Combustion and Flame, 161(6), 1587-1603. 

Fischer, S.L., Dryer, F.L., Curran, H.J., 2000, The reaction kinetics of dimethyletherI: high temperature pyrolysis 

and oxidation in flow reactors, International Journal of Chemical Kinetics, 32, 713–740. 

Kaiser F.W., Wallington T.J., Hurley M.D., 2000, Experimental and modeling study of premixed atmospheric-

pressure dimethyl ether−air Flames, Journal of Physical Chemistry A, 104, 8194–8206. 

Lee S., Oh S., Choi Y., Kang K., 2011, Effect of n-butane and propane on performance and emission 

characteristics of an SI engine operated with DME-blended LPG fuel, Fuel, 90, 1674–1680. 

Lockett R. D., Boulanger B., Harding S. C., et al., 1999, The structure and stability of the laminar counter-flow 

partially premixed methane/air triple flame, Combustion and Flame, 119(1–2), 109-120. 

Lowry W., Serinyel Z., Krejci M., Curran H., Bourque G, Petersen E., 2011, Effect of methane-dimethyl ether 

fuel blends on flame stability, laminar flame speed, and Markstein length, Proceedings of the Combustion 

Institute, 33, 929–937. 

Min L., Cho M. S., Kim N. I., 2014, Characteristics of opposed flow partially premixed flames in mesoscale 

channels at low strain rates, Proceedings of the Combustion Institute, 35, 3439–3446. 

Yu H., Hu E., Cheng Y., Zhang X., Huang Z., 2014, Experimental and numerical study of laminar premixed 

dimethylether/methane–air flame, Fuel, 136, 37–45. 

Zheng X., Lu T., Law C.K., Westbrook C.K., Curran H.J., 2005, Experimental and computational study of 

nonpremixed ignition of dimethyl ether in counterflow, Proceedings of the Combustion Institute, 30, 1101-

1109. 

420