Cool flames in micro channel
Recent progress of weak flames in a micro flow reactor
with a controlled temperature profile
The 1st International Workshop on Near Limit Flames, July 29-30, 2017, Boston, MA, USA
Kaoru Maruta1,2, Ryota Tatsumi1, Hisashi Nakamaura1, Takuya Tezuka1
1: Tohoku Univ., Japan
2: Far Eastern Federal Univ., Vladivostok, Russia
A. Yamamoto, M. Hori, P. Grajetzki, T. Okuno, A.K. Dubey, T. Onda, Y. Kizaki
Contents
・ Outline of a micro flow reactor with
a controlled temperature gradient
・ Separated cool flame
・ Recent progress of weak flame studies
For larger fuel diversity, understandings of ignition and
combustion characteristics of various practical fuels are
important. Ex. bio-, synthesized-, reformed-fuels, etc.
A micro flow reactor with a controlled temperature profile
Background
Shock tube, RCM, burners have
been extensively used but
simple and compact method for
examining reaction dynamics is
required. TimeHe
at re
lea
se
ra
te
Cool flame
Hot flame
Two-stage
ignition
Microcombustion projects
Swiss roll
microcombustors
for heat sourcesMicro flow reactor with
controlled temp. profile
Microcombustion for fundamental combustion studies
4
Micro flow reactor with a controlled temperature profile
5
Stationary wall-temperature profile
Diameter of tube < conventional quenching diameter
Gas phase temperature dominated by wall temperature profile
Laminar flow and constant pressure
T
xoTest section xo
TmaxT
xoTest section xo
Tmax
dd
T
Mixture
Room temp
Tmax = 1300 K
d < Quenching
diameter
External heat source
Mixture
Flame behaviors in a micro flow reactor with a controlled temperature profile
Normal flame
Oscillatory flame
High velocity region
Weak flame
Low velocity region
Intermediate velocity region
6
Three kinds of flame responses
(1) Normal flame
(3) Weak flame
(2) Oscillatory flame
V=50cm/s
V=20cm/s
V=0.2cm/s-12 -10 -8 -6 -4 -2 0 2 4 6 8 10
0
20
40
60
80
100
120
= 1.0
Normal flame
FREI
Estimated points
of ignition
Location (mm)
Mean f
low
velo
city (
cm
/s)
400
600
800
1000
1200
1400
1600
Wa
ll Te
mp
era
ture
, Tw (K
)Ignition
Extinction
Tw
Flow
Normal flame, oscillatory flame, weak flame
CH4/air mixture
7Maruta et al., PCI 30
Theoretical S-shaped solutions
V
v
Stable
xStable
v
Unstable
Flow
Two stable and one unstable solutions predicted theoretically
Weak flame corresponds to ignition branch of S-curve
8Minaev et al., CTM 2007
experiments
9
Stable branch
Stable branch: weak flame
Unstable branch
timeChemical
time residense Flowa D
• Normal flame:
preheated premixed
flame
• Oscillatory flame:
flames with
repetitive extinction
and ignition, FREI
• Weak flame: stable
weak flame which
represent ignition
Analogy between S-curve and conventional Findell-curve
Da: Damköhler number
Max
imu
m t
emp
erat
ure
Ign
itio
n
Exti
nct
ion
Three flames are utilized for:
(1) Normal flame
(3) Weak flame
(2) Oscillatory flame
10
Measurements of laminar burning
velocity of highly preheated mixtures
Investigations of non-linear dynamics of
given fuels
Investigations of ignition relevant
properties of various fuels, validation
and modifications of kinetics,
particularly at low temperature
Triple weak flames, n-heptane
U = 3 cm/s
Triple stationary weak flames observed
Weak flame location insensitive to flow velocity
= 1
11
Yamamoto et al., PCI 33
Code PREMIX-based
Reaction scheme: n-heptane, reduced mechanism from LLNL
(159 species, 1540 steps)
Conditions: Stoichiometric gaseous n-heptane/air mixture
Experimental wall temperature profile was provided as Tw(x)
Gas-phase energy equation
21 1
1 4( ) 0
K K
k k pk k k k w g
k kp p p p
dT d dT A dT A A NuM A Y V c h W T T
dx c dx dx c dx c c d
Heat transfer with the wall
Flame position: Peaks of heat-release-rate (HRR) [W/cm3] profile
Computations (one-dimensional)
Seiser et al., PCI 28 (2000)
12
Triple weak flames, n-heptane
U = 3 cm/s
Three-stage heat releases13
Computational
U = 3 cm/s
Three-stage heat releases14
Computational
Triple weak flames, n-heptane
300
400
500
600
700
800
900
1000
1100
1200
1300
0
5
10
15
20
25
3.5 4 4.5 5 5.5 6
Wa
ll t
em
pera
ture
[K
]
Mo
le f
ra
cti
on
[%
]
x [cm]
O2
CO2
CH2O×10 CO
CH4×20
n-C7H16×10
H2O2×10
Tw
U = 2.0 cm/s
Computational species profiles
15Three peaks of heat release rate
Yamamoto et al., PCI 33
300
400
500
600
700
800
900
1000
1100
1200
1300
0
5
10
15
20
25
3.5 4 4.5 5 5.5 6
Wa
ll t
em
pera
ture
[K
]
Mo
le f
ra
cti
on
[%
]
x [cm]
O2
CO2
CH2O×10 CO
CH4×20
n-C7H16×10
H2O2×10
Tw
LTO: CH2O, H2O2, CO, CH4 produced
U = 2.0 cm/s
Computational species profiles
First HRR peak
16
300
400
500
600
700
800
900
1000
1100
1200
1300
0
5
10
15
20
25
3.5 4 4.5 5 5.5 6
Wa
ll t
em
pera
ture
[K
]
Mo
le f
ra
cti
on
[%
]
x [cm]
O2
CO2
CH2O×10 CO
CH4×20
n-C7H16×10
H2O2×10
Tw
U = 2.0 cm/s
CH2O + OH ⇒ HCO + H2O
H2O2 (+M) ⇒ 2OH (+M) HCO + O2⇒ CO + HO2
Computational species profiles
Partial oxidations:
Second HRR peak
17
300
400
500
600
700
800
900
1000
1100
1200
1300
0
5
10
15
20
25
3.5 4 4.5 5 5.5 6
Wa
ll t
em
pera
ture
[K
]
Mo
le f
ra
cti
on
[%
]
x [cm]
O2
CO2
CH2O×10 CO
CH4×20
n-C7H16×10
H2O2×10
Tw
U = 2.0 cm/s
Full oxidations: CO + OH ⇒ CO2 + H
Computational species profiles
Third HRR peak
18
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1
0
5
10
15
20
25
300 500 700 900 1100 1300
Ma
ss c
on
cen
tra
tio
n o
f C
H2O
[%
]
Vo
lum
etr
ic c
on
cen
tra
tio
n [
%]
Wall temperature [K]
O2
CO2
CH2O
CO
CH4×20
300
400
500
600
700
800
900
1000
1100
1200
1300
0
5
10
15
20
25
3.5 4 4.5 5 5.5 6
Wa
ll t
em
pera
ture
[K
]
Mo
le f
ra
cti
on
[%
]
x [cm]
O2
CO2
CH2O×10 CO
CH4×20
n-C7H16×10
H2O2×10
Tw
Measurement by GC
Measurements and computations
Three-stage oxidation process was
confirmed by gas sampling experiment
(U = 2.0 cm/s)
Computations
19
Yamamoto et al., PCI 33
Interpretation of triple weak flames
Conventional two-stage ignition converted into
steady, three-stage, spatially-separated weak
flames in MFR
Cool flame
Hot flame
Hea
t re
leas
e ra
te
time
Flow direction
PRF0
PRF20
PRF50
PRF100
700 800 900 1000 1100 1200Wall temperature (experiment) [K]
-1.5 -1.0 -0.5 0 0.5Location [cm]
Flow direction
PRF0
PRF20
PRF50
PRF100
700 800 900 1000 1100 1200Wall temperature (experiment) [K]
-1.5 -1.0 -0.5 0 0.5Location [cm]
n-heptane/air = 1
1 mm
Ignition in
RCM and ST
(transient)
Cool flame Separated hot flames
20
MFR applied for gasoline PRF
n-heptane + iso-octane (PRF)
Appearances of multiple weak flames
represent Research Octane Number21
Flow direction
PRF0
PRF20
PRF50
PRF100
700 800 900 1000 1100 1200Wall temperature (experiment) [K]
-1.5 -1.0 -0.5 0 0.5Location [cm]
= 1
U0 = 2 cm/s
Flow direction
PRF0
PRF20
PRF50
PRF100
700 800 900 1000 1100 1200Wall temperature (experiment) [K]
-1.5 -1.0 -0.5 0 0.5Location [cm]
1 mm
* Hori, et al., CNF 2012
U=1.2 cm/sec
Weak flames at different RON
Significant LTO in smaller RON
Weak flame behaviors reproduced22
Computation
* Hori, et al., CNF 2012
Flow direction
PRF0
PRF20
PRF50
PRF100
700 800 900 1000 1100 1200
Wall temperature (experiment) [K]
-1.5 -1.0 -0.5 0 0.5Location [cm]
(a)
(d)
Flow direction
700 800 900 1000 1100 1200Wall temperature (K)
2 mm
120
Flow direction
PRF0
PRF20
PRF50
PRF100
700 800 900 1000 1100 1200
Wall temperature (experiment) [K]
-1.5 -1.0 -0.5 0 0.5Location [cm]
methanetoluene
propane
ethane
n-heptane, PRF 0
iso-octane, PRF 100n-butane
PRF 50
PRF 20
120
112
108
100
94
50
20
0
RON
Weak flame appearances for various RON
Weak flame locations: monotonic function of RON
Second weak flame: visible when RON < 100
First weak flame: visible when RON < 20
P = 1 atm
23
Structure of a separated cool flame in MFR
24
0
0.1
0.2
0.3
0.4
0
0.05
0.1
0.15
0.2
0.25
5 6 7 8 9 10 11 12 13
Mole
fra
ctio
n (
-)
HR
R (
J/s/
cm3)
1st flame
nC7H16×10
O2
H2O2×10
CH2O×10
CO×10
CO2×10
(cool flame)
KUCRS mech. Location (cm)
600 700650 600550
Wall Temperature [K]
Tatsumi et al., Heat Transfer Symposium 2017.
Also, will be presented at ICDERS 2017.
Vertical MFR
Small temp. grad.
Sym. & Higher res.
KUCRS mech.
25
Structure of a cool flame, computation
Tatsumi et al., Heat Transfer Symposium 2017.
Also, will be presented at ICDERS 2017.
600 700650 600550
Conditions : nC7H16/air, = 1, P = 1 (atm), V = 2 (cm/s), Exposure time 30 (minutes)
Wall temperature (K)
0
0.2
0.4
0.6
0
0.01
0.02
5 6 7 8 9 10 11 12 13 HR
R (
J/s/
cm3)
0
0.2
0.4
0.6
0
0.01
0.02
5 6 7 8 9 10 11 12 13
Mo
le f
ract
ion (
-)
Location (cm)
00.6
00.02
KUCRS mech.
LLNL mech.
nC7H16
H2O2CH2O
nC7H16 H2O2
CH2O
Reaction paths in a cool flame
RH
R
ROO
QOOH
OOQOOH
OQ-HOOH Cyclic ethers
Aldehydes + ester or ketone group
Alkenes + ester or ketone group
CH2O, H2O2
R : C7H15
Q : C7H14
Q-H : C7H13
(Ketohydroperoxide)
OQ-HO
← OH
← O2
→ OH
→ OH← OH
← O2
→ OH
Radical branching path
Radical propagation path
CO, CO2
27
Reactions in radical branching path
O C·C
C
C
C
CCO O C·C
C
C
C
CCO
O C·C
C
C CCCO
O C·C
C
C
C
C
CO
O C·C CC CCCO
C CC
O
C
O·
CCC C CC
O
C
O·
CCC C CC
O
C
O·
CCC C CC
O
C
O·
CCC
72 %
C2H5CHO + C2H5COCH2
C3H7CHO + CH3COCH2
CH3CHO + C4H9CO
C4H9CHO + CH2CHO
72 %
OQ-HO (C7H13O2)
OOQOOH
Aldehydes
+ ester or ketone group
QOOH (C7H14OOH)
OQ-HOOH
Conditions : nC7H16/air, = 1, P = 1 (atm), V = 2 (cm/s), Tw,max = 700 (K) KUCRS mech.
28
Reactions in radical propagation path
O C·C
C
C
C
CCO
O CC
C
C CCC·O
O CC
C
C
C
CC·OO CC
C
C
C
C·O
C
O CC
C
C CC·O C
C
C
C
C CC C
OCC
CO
C CCC
C
C
C CC C
O
C
93 %
C3H6 + C2H5COCH2
or
C4H8 + CH3COCH2
C3H6 + nC3H7CO
or
C5H10 + CH3CO
C4H8 + C2H5CO
80 %
QOOH (C7H14OOH)
Cyclic ethers
Alkene
+ ester or ketone group
Conditions : nC7H16/air, = 1, P = 1 (atm), V = 2 (cm/s), Tw,max = 700 (K) KUCRS mech.
33
PRFs, ultra lean, higher resolution
Vertical-type reactorGrajetzki, et al., 36 Symp, WIP.
34
CH4, ultra lean and Xe dilution
Okuno et al., PCI 36.
Conducted for analyzing
microgravity experiments
on interaction between
flameball and
counterflow flames
Fuel: CH4; O2/Inert = 0.141; = 0.3
35
Syngas, elevated pressures
H2:CO
H HO2
H2O2 OH
O2(+M)
+H
+O2
+M
Relevant to H2-O2 explosion peninsula
Tanaka et al., 36symp. WIPP.
Soot and PAH formations at temperature below 1400 K
38
CH4 CH3
CH2O
HCO
CO
CO2
CH3O
C2H6
C2H3
C2H2
C2H4
C2H5
CH2CHO
CH2COHCCO
+ CH3(+M)
+H
+OH
+ CH3
+ HO2
+O
+OH
+H
+O
+ HO2
+O2
+O2
+(M)
+H2O
+OH
+HO2
+O2
+(M)
+OH
+H
+ CH3
+O
+OH
+H
+CH3
+O
+(M)
+O2
+(M)
+O2
+O2
+H
+OH
+(M)+O
+O
+ HO2
* Dubey et al., CNF 2017.
Soot and PAH formations at temperature below 1400 K
39* Dubey et al., CNF 2017.
40
Effect of radical quenching at wall
0
0.005
0.01
0.015
0.02
0.025
0.03
0.035
0.04
0.045
0.05
0.7 0.8 0.9 1 1.1 1.2 1.3
Mo
le f
ract
ion
[-]
φ
Inert wall
Quench wall
Experiment
Inert wall (computation)
Quench wall (computation)
Experiment
• CO mole fraction of burned gas:
Quench case > Inert case
• Experimental results agree with Inert case
→ radical quenching effect negligible
Inert: GRI3.0
Quench: GRI 3.0 + Raimondeau
CH4/air; = 1; d = 1.5 mm, 1 atm
Kizaki, et al., PCI35 (2015)
Conclusions
42
Micro flow reactor with prescribed temperature profile
can be a new platform for study on chemical kinetics
particularly for low temperature conditions
Acknowledgements
IHI, IIC, HONDA R&D, MAZDA
IHI, NEDO, HITACHI, TOKYO GAS, JSPS, JST