Clean Combustion Research Center, King Abdullah University of Science & Technology (KAUST)
The Effect of Dimethyl Ether Addition on
Sooting Limits in Counterflow Diffusion
Flames at Elevated Pressures
Zepeng Li, Peng Liu, Hafiz M. F. Amin, Yu Wang, Suk Ho Chung, William L. Roberts
Clean Combustion Research Center, (CCRC)
King Abdullah University of Science and Technology (KAUST)
Saudi Arabia
Clean Combustion Research Center, King Abdullah University of Science & Technology (KAUST)2
CONTENT
Introduction
Methodology
Results and Discussion
Conclusions
Clean Combustion Research Center, King Abdullah University of Science & Technology (KAUST)3
Motivation and background
Soot: carbonaceous particles
resulting from incomplete
combustion of hydrocarbon
fuels.
Incomplete Combustion: Efficiency
Deposition : Burner Lifetime / Performance
Health: Carcinogenic and Mutagenic
Climate: Global Warming & Regional
precipitation
Dimethyl Ether
(DME)
AdvantagesHigh oxygen content & the
absence of C–C bonds: smokeless
combustion, low formation and high
oxidation rates of particulates.
High cetane number
Low boiling point
Disadvantages Low energy density
High requirements on sealing
materials
Clean Combustion Research Center, King Abdullah University of Science & Technology (KAUST)4
Sooting tendency
Quantitative sooting metrics were pronounced to evaluate sooting
tendency among a wide range of hydrocarbon fuels, e.g., TSI, YSI.
[1] Yang Y, Boehman AL, Santoro RJ. Combust Flame
2007;149:191–205.
[2] McEnally CS, Pfefferle LD. Combust Flame 2007;148:210–22.
Threshold soot index (TSI) [1] Yield soot index (YSI) [2]
Clean Combustion Research Center, King Abdullah University of Science & Technology (KAUST)5
CONTENT
Introduction
Methodology
Results and Discussion
Conclusions
Clean Combustion Research Center, King Abdullah University of Science & Technology (KAUST)6
Experimental setup
Schematics of experimental apparatus: Light scattering setup (a), and atmospheric
counterflow burner (b)
Light scattering technique + Counterflow diffusion flames
Clean Combustion Research Center, King Abdullah University of Science & Technology (KAUST)7
Numerical Simulation
Flame temperatures and species concentrations of
experimental flames were computed using Chemkin Pro. A
soot model is not included in the simulation due to the
negligible impact of soot on flame temperature at the sooting
limit.
• Module: Opposed-flow Flame package
• Reaction kinetic model: KAUST-Aramco PAH Mech 1
(KAM1)
[1] Park, Sungwoo, et al. "Compositional effects on PAH and soot formation in counterflow diffusion flames of gasoline
surrogate fuels." Combustion and Flame 178 (2017): 46-60.
Clean Combustion Research Center, King Abdullah University of Science & Technology (KAUST)8
CONTENT
Introduction
Methodology
Results and Discussion
Conclusions
Clean Combustion Research Center, King Abdullah University of Science & Technology (KAUST)9
How to determine sooting limits
2.1 2.4 2.7 3.00
2
4
6
Xo=0.16
Xo=0.17
Sca
tter
ing
in
ten
sity
(A
. U
.)
Distance from fuel nozzle [mm]
P=1 atm
Xf=1
Xo,cr
=0.17
Xo=0.18
The sooting limit map defined as the critical oxygen mole fraction
(Xo) related to the fuel mole fraction (Xf) at soot inception point is
detailed discussed in [1,2].
[1] Y. Wang, S.H. Chung. Combust. Flame 161.5 (2014): 1224-1234.
[2] P.H. Joo, Y. Wang, A. Raj, S.H. Chung, Proc. Combust. Inst. 34 (2013) 1803–1809.
Notation
β: DME mixing ratio
β≡[𝐶
2𝐻6𝑂]
𝐶2𝐻6𝑂 +[𝐶
2𝐻4]
Clean Combustion Research Center, King Abdullah University of Science & Technology (KAUST)10
Repeatability
[1] Y. Wang, S.H. Chung. Combust. Flame 161.5 (2014): 1224-1234.
[2] P.H. Joo, Y. Wang, A. Raj, S.H. Chung, Proc. Combust. Inst. 34 (2013) 1803–1809.
0.0 0.2 0.4 0.6 0.8 1.00.0
0.2
0.4
0.6
0.8
1.0
SFO SF
F F
O O
SFO
Non-sooting region
Fuel mole fraction, XF
Ox
yg
en m
ole
fra
ctio
n, X
O
Ref. [8]
Setup 1
Setup 2
Ethylene Flame
1470
1644
1819
1993
2168
2342
2516
2691
2865
Max.
Fla
me
Tem
per
atu
re (
K)
Sooting region
SF
OF
Flame zone
Soot zone
Oxidizer
Fuel mixture
Clean Combustion Research Center, King Abdullah University of Science & Technology (KAUST)11
0.0 0.2 0.4 0.6 0.8 1.00.0
0.2
0.4
0.6
0.8
1.0
=0
SFO
Fuel mole fraction, XF
Oxyg
en m
ole
fra
ctio
n, X
O
Sooting region
Non-sooting region
SF
(a)
SFO
SF
=0.5
∆Xo
1. Sooting Limit Map
Highest sooting propensity:
0~ 10% DME addition
∆Xf 1. The addition of DME
reduces sooting tendency
among β=0.1-0.5.
2. ∆Xf > ∆Xo : small thermal
effect in SF flames, while
high thermal effect in SFO
flames.
3. Sooting limit curves are
overlapped when β<0.1.
Clean Combustion Research Center, King Abdullah University of Science & Technology (KAUST)12
2. Highest sooting tendency
0 2 4 6 8 10
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Norm
aliz
ed m
ax. sc
atte
ring
inte
nsi
ty (
A. U
.)
DME mixing ratio (%)
=5.7%
R2=0.787
Maximum scattering intensity as a function of DME mixing ratios (β) with 95%
confidence band.
Highest sooting
propensity: 5.7%
DME addition
Critical DME mixing
ratio = 5.7%
Clean Combustion Research Center, King Abdullah University of Science & Technology (KAUST)13
0.0 0.1 0.2 0.3 0.4 0.50.16
0.20
0.24
0.28
0.4 0.6 0.8 1.0
Quadratic term
coefficient
Fuel mole fraction, xf
Xf=1
Xf=0.85
Xf=0.7
Xf=0.55
Xf=0.4
DME mixing ratio,
Cri
tica
l o
xy
gen
mo
le f
ract
ion
, X
o,c
r
0.05
3. Dilution effect
When the fuel
(or carbon flow)
is diluted, the
dependence of
dilution on
sooting
tendencies
increases
Critical oxygen mole fraction as a function of DME mixing ratios (β)
Clean Combustion Research Center, King Abdullah University of Science & Technology (KAUST)14
0.0
0.2
0.4
0.6
0.8
1.0
P (atm)
1
2
3.5
0.0 0.2 0.4 0.6 0.8 1.00.0
0.2
0.4
0.6
0.8
1.0
Fuel mole fraction, XF
Ox
yg
en
mo
le f
racti
on
, X
O
0.0 0.2 0.4 0.6 0.8 1.0
4. Pressure effect
As pressure
increases,
sooting tendency
increases
Sooting limit maps with respect to DME mixing ratios (β) at P=1, 2, 3.5 atm
Clean Combustion Research Center, King Abdullah University of Science & Technology (KAUST)15
0.00 0.05 0.10 0.15 0.20 0.250.0
0.2
0.4
0.6
0.8
1.0
P=1 atm
cr=0.057
R2=0.787
P (atm) 1
2
3.5
P=3.5 atm
cr=0.144
R2=0.739
P=2 atm
cr=0.106
R2=0.806
No
rmali
zed
max
. sc
att
erin
g i
nte
nsi
ty (
A. U
.)
DME mixing ratio,
4. Pressure effect
The range of β in
increasing soot
propensity expands
with pressure.
The effort to
reduce soot by
doping DME may
not be effective at
high pressures.
Sooting propensity with DME mixing ratio at several pressures
Clean Combustion Research Center, King Abdullah University of Science & Technology (KAUST)16
5. Kinetic analysis
0.0 0.1 0.2 0.3 0.4 0.50.0
1.5
3.0
4.5
6.0
7.5M
ole
fra
ctio
n o
f b
enze
ne
(1E
-4)
DME (or FDME) mixing ratio
P=1atm, Thermal contribution
P=1atm, Chemical contribution
P=3atm, Thermal contribution
P=3atm, Chemical contribution
Thermal and chemical contributions to benzene production at P = 1, 3 atm.
Clean Combustion Research Center, King Abdullah University of Science & Technology (KAUST)17
CONTENT
Introduction
Methodology
Results and Discussion
Conclusions
Clean Combustion Research Center, King Abdullah University of Science & Technology (KAUST)18
Conclusions
The sooting limits (Xf and XO) are significantly sensitive to pressure, and the sooting
tendency increases at higher pressure. The range of β in increasing soot propensity
with DME addition expands with pressure, thus the effort to reduce soot by doping DME
may not be effective at high pressures.
When the fuel (or carbon flow) is diluted, the dependence of dilution on sooting
tendencies increases. That is, as Xf decreases, the increase of Xo,cr with respect to DME
mixing ratio becomes more sensitive, meaning that the effect of DME addition on sooting
tendency increases as the fuel is diluted.
The behaviors of the DME effect on sooting limits in SF flame and SFO flame are
slightly different. Sooting tendency is more sensitive in SFO flames than in SF flames
due to the thermal effect.
Kinetical analysis indicate that soot formation with DME addition is dominantly
determined by the synergistic chemical effect, which is likely realized by the pathway
of DME→CH3(H)→C2H2(C2H3&C3H3)→C6H6→soot.
Clean Combustion Research Center, King Abdullah University of Science & Technology (KAUST)19
Acknowledgements
This work is supported by Clean Combustion
Research Center (CCRC), King Abdullah University
of Science and Technology (KAUST), Saudi Arabia.