Post on 16-Oct-2021
1
Development of Detailed and Reduced Kinetic Mechanisms for Surrogates of Petroleum-Derived and Synthetic Jet Fuels:
Drexel Flow Reactor Studies at Low and Intermediate Temperatures
David L. Miller and Nicholas P. Cernansky
Matthew Kurman and Robert Natelson
Department of Mechanical Engineering and MechanicsDrexel University, Philadelphia, PA 19104-2875
MACCCR Fuels Research ReviewGaithersburg, MD
8-10 September 2008
Research supported by the Air Force Office of Scientific Research
2
Motivation
• Detailed kinetic models will aid in the development and optimization of the next series of advanced air-breathing propulsion systems and their use with alternative fuels
– Model development and verification require quantitative characterization of combustion properties
• Oxidation at low and intermediate temperatures (600-1100 K) is important in some engine designs
– High-quality, reproducible data quantifying the combustion of jet fuel reference components at these conditions is scarce
3
Alkane Oxidation Mechanism
• CO is the major indicator of reactivity at low temperatures
HOORROOHRORRRH RHMOH +→+→ →←→ +++−2
2
2OH +22OH +ROOHOH +2
ROOHO +2
COROCROHCHOROR MO +′→′+→′+′′ ++ 2
2
2' OOROR ++
OROR '+
2'OR +
RH+
CORORCOHRCHOOHROOHQ MO +→+→+→ ++ 2
2'β
β RROH ′=+2
HOCHORRR +′′+′=
RRHO ′−+O
OQOOHOHOOQOOHR RH ←+ +
CHORORHOOROOHHOOOHRHOO ′′+′+→+→
RRR ′′=′+
β
MO ++ 2
RROH ′=+2
βROH +22
HO2 2O+
4
Program Overview
• Objectives: 1) Explore the preignition oxidation behavior of petroleum and alternative jet fuel surrogate components
2) Quantify the combustion properties of the surrogate components
3) In Year 1, n-dodecane studies
• Approach:– React the fuel/oxidizer/diluent systems under well-controlled conditions in
our Pressurized Flow Reactor (PFR) » Perform bench scale tests on n-dodecane
• Coordinate with Stanford flow reactor experiments at higher temperatures
» Monitor reactivity and collect gas samples as a function of experimental and reactant conditions
» Perform detailed chemical analysis of extracted gas samples
– Mechanistic analysis and development– Provide data for USC chemical kinetic model development
5
PFR Facility
Nitrogen OxygenNitrogenDiluent
Exhaust
PressureTransducers
PressureRegulating
Valve
On-line AnalyzersCO/CO2/O2 NDIR
Exhaust
Off-line Analyzers GC/MS & GC/FID
SampleStorage
Cart
Fuel
3kWHeater
10kWHeater
QuartzReactor
Gas SamplingProbe & TC
MixingNozzle Computer Controlled
Probe Positioning TableProbe Cooling
System
Temperature: 600 – 850 KPressure: 2 – 16 atmQuartz Reactor ID: 2.2 cmQuartz Reactor Length: 40 cm
Data Control &
Acquisition System
6
Representative Reactivity Map Profile
CO
pro
duct
ion
[ppm
]
Temperature [K]
NTC RegionNTC start
NTC endStart ofReactivity
7
GC/MS/FID Facility
• Samples from PFR collected in 10-ml loops in sample storage cart heated to 180°C
• Samples injected into GC/MS/FID for analysis
• Separation aided by sub-ambient initial oven temperature
• Identification performed by mass spectra compared to NIST MS Version 2.0
• Supelco Petrocol DH column in GC– 100 m, 0.5 µm film thickness – 0.25 mm OD, 1250 Phase Ratio (β)
• MS parameters– Ion source 200°C– Electron ionization -70 eV– Multiplier voltage 1456 V– Emission current 100 µΑ
GC temperature ramp
8
Other Measurements
• Measure CO/CO2/O2 with on-line analysis– Errors are 50 ppm for CO and CO2, and 1250 ppm for O2
• Measure F/O2 ratios with Total Hydrocarbon Analyzer combined with O2 Analyzer to calibrate equivalence ratio before each experiment
• Calibration of lighter hydrocarbons is achieved with FID calibration using purchased gas-phase standards at 15, 100, and 1000 ppm
• Calibration of heavier species is achieved with correction factors of FID signals that account for differences in carbon, hydrogen, and oxygen numbers between different molecules of similar structures (Schofield, 2008)
• In current results, uncertainties are 1 standard deviation of two experiments (Exp 1+2) with similar initial conditions– Uncertainty calculations will improve with additional experiments
Schofield, K. (2008). Prog Energ Combust 34:330-350.
9
Experimental Conditions
Exp 1 (5/14/08) “M Phi #1”
Exp 2 (5/28/08) “M Phi #2”
Exp 3 (6/23/08) “H Phi”
Exp 4 (7/28/08) “L Phi”
n-Dodecane (ppm) 1297 1529 1667 519
O2 (ppm) 37700 37500 35200 42000
φ 0.64 0.75 0.88 0.23
N2 (ppm) 961000 961000 963100 957500
• Equivalence ratios for Exp 1 and Exp 2 are essentially identical, considering uncertainties in O2 and n-dodecane
• Exp 4 was conducted to reduce hydrocarbon loading on experiment
• For all experiments, pressure maintained at 8.000 0.025 atm and residence time maintained at 120 10 ms
10
CO
• Peak CO production of Moderate Phi #1 and #2 are within uncertainty– 50 ppm
uncertainty in CO
• High Phi produced only slightly more CO than Moderate Phi’s
• Low Phi had much lower initial n-dodecane molar fraction– NTC region
started at lower temperature
0
500
1000
1500
2000
2500
3000
3500
4000
600 650 700 750 800
Temperature (K)
Mo
lar
fra
cti
on
(p
pm
)
M Phi #1
M Phi #2
H Phi
L Phi
0
200
400
600
800
1000
1200
1400
600 650 700 750 800
Temperature (K)
Mol
ar fr
actio
n (p
pm)
11
CO2
50 ppm uncertainty in CO2
M Phi #1
M Phi #2
H Phi
L Phi
25000
27000
2900031000
33000
35000
3700039000
41000
43000
600 650 700 750 800
Temperature (K)
Mol
ar fr
actio
n (p
pm)
12
O2
1250 ppm uncertainty in O2
M Phi #1
M Phi #2
H Phi
L Phi
13
n-Dodecane
• All experiments exhibited similar n-dodecane molar fractions at 670-745 K
• Low Phi had significantly lower initial molar fraction
M Phi #1
M Phi #2
H Phi
L Phi
0
100
200
300
400
500
600
600 650 700 750 800
Temperature (K)
Mo
lar
fra
cti
on
(p
pm
)
14
Alkane Oxidation Mechanism – Alkene Formation
• In general, alkenes produced from β-scission of alkyl (R.) and alkylhydroperoxy (QOOH.) radicals, and from O2 addition to alkyl radicals
HOORROOHRORRRH RHMOH +→+→ →←→ +++−2
2
2OH +22OH +ROOHOH +2
ROOHO +2
COROCROHCHOROR MO +′→′+→′+′′ ++ 2
2
2' OOROR ++
OROR '+
2'OR +
RH+
CORORCOHRCHOOHROOHQ MO +→+→+→ ++ 2
2'β
β RROH ′=+2
HOCHORRR +′′+′=
RRHO ′−+O
OQOOHOHOOQOOHR RH ←+ +
CHORORHOOROOHHOOOHRHOO ′′+′+→+→
RRR ′′=′+
β
MO ++ 2
RROH ′=+2
βROH +22
HO2 2O+
Ethene
050
100150200250300
600 650 700 750 800 850
Temperature (K)
Mo
lar
fra
cti
on
(p
pm
)
15
Alkenes
• Decrease in light alkenes at 805-820 K temperatures appears in Low Phi experiment
M Phi #1M Phi #2
H PhiL Phi
Propene
0
50
100
150
600 650 700 750 800 850
Temperature (K)
Mo
lar
fra
cti
on
(p
pm
)
1-Butene
0
20
40
60
80
100
600 650 700 750 800 850
Temperature (K)
Mo
lar
fra
cti
on
(p
pm
)
16
Alkenes
M Phi #1M Phi #2
H PhiL Phi
1-Pentene
010203040506070
600 650 700 750 800 850
Temperature (K)
Mo
lar
fra
cti
on
(p
pm
)
1-Hexene
0102030405060
600 650 700 750 800 850
Temperature (K)
Mo
lar
fra
cti
on
(p
pm
)
17
Alkenes
M Phi #1M Phi #2
H PhiL Phi
1-Heptene
0
10
20
30
40
600 650 700 750 800 850
Temperature (K)
Mo
lar
fra
cti
on
(p
pm
)
1-Octene
05
101520253035
600 650 700 750 800 850
Temperature (K)
Mo
lar
fra
cti
on
(p
pm
)
18
Alkenes
M Phi #1
M Phi #2
H Phi
L Phi
1-Nonene
05
101520253035
600 700 800
Temperature (K)
Mo
lar
fra
cti
on
(p
pm
)
1-Decene
05
1015202530
600 650 700 750 800 850
Temperature (K)
Mo
lar
fra
cti
on
(p
pm
)
19
Alkenes
M Phi #1
M Phi #2
H Phi
L Phi
• Several dodecene isomers are shown lumped as overlapping of GC peaks makes positive identification difficult
Dodecenes
01020304050
600 650 700 750 800 850
Temperature (K)
Mo
lar
fra
cti
on
(p
pm
)
20
Alkane Oxidation Mechanism – Aldehyde Paths
• In general, aldehydes produced from decomposition of ketohydroperoxides (OROOH) , β-scission of alkylhydroperoxy (QOOH.) radicals, and destruction of alkylperoxy (RO2
.) radicals
HOORROOHRORRRH RHMOH +→+→ →←→ +++−2
2
2OH +22OH +ROOHOH +2
ROOHO +2
COROCROHCHOROR MO +′→′+→′+′′ ++ 2
2
2' OOROR ++
OROR '+
2'OR +
RH+
CORORCOHRCHOOHROOHQ MO +→+→+→ ++ 2
2'β
β RROH ′=+2
HOCHORRR +′′+′=
RRHO ′−+O
OQOOHOHOOQOOHR RH ←+ +
CHORORHOOROOHHOOOHRHOO ′′+′+→+→
RRR ′′=′+
β
MO ++ 2
RROH ′=+2
βROH +22
HO2 2O+
Propanal
0
20
40
60
80
100
600 650 700 750 800 850
Temperature (K)
Mo
lar
fracti
on
(p
pm
)
21
Aldehydes
M Phi #1
M Phi #2
H Phi
L Phi
Butanal
0
20
40
60
80
600 650 700 750 800 850
Temperature (K)
Mo
lar
fra
cti
on
(p
pm
)
Pentanal
0
10
20
30
40
50
600 650 700 750 800 850
Temperature (K)
Mo
lar
fra
cti
on
(p
pm
)
22
Aldehydes
M Phi #1M Phi #2
H PhiL Phi
Hexanal
05
101520253035
600 650 700 750 800 850
Temperature (K)
Mo
lar
fracti
on
(p
pm
)
23
Aldehydes
M Phi #1
M Phi #2
H Phi
L Phi
Heptanal
05
1015202530
600 650 700 750 800 850
Temperature (K)
Mol
ar fr
actio
n (p
pm)
2-Nonanone
0
1
2
3
4
5
600 650 700 750 800 850
Temperature (K)
Mo
lar
fra
cti
on
(p
pm
)
24
Ketones
M Phi #1
M Phi #2
H Phi
L Phi
2-Decanone
0
1
2
3
4
5
600 650 700 750 800 850Temperature (K)
Mo
lar
fra
cti
on
(p
pm
)
25
Ketone-substituted iso-Dienes
M Phi #1
M Phi #2H PhiL Phi
5,6,7,7-tetramethyl-octa-3,5-dien-2-one (C12H20O)
02468
101214
600 650 700 750 800 850
Temperature (K)
Mo
lar
frac
tion
(p
pm
)
2,4,4,7-tetramethyl-octa-5,7-dien-3-one (C12H20O)
02468
10
600 650 700 750 800 850
Temperature (K)M
ola
r fr
actio
n
(pp
m)
26
Enones
M Phi #1M Phi #2
H PhiL Phi
Methyl vinyl ketone
010
2030
4050
60
600 650 700 750 800 850
Temperature (K)
Mol
ar fr
actio
n (p
pm)
Dihydro-2(3H)-furanone (C4H6O2)
0102030405060
600 650 700 750 800 850
Temperature (K)
Mo
lar
fra
cti
on
(p
pm
)
27
Lactones (Cyclic Esters)
M Phi #1
M Phi #2
H Phi
L Phi
• Must be cautious about sampling
– dihydro-2(3H)-Furanone (γ-butyrolactone) is easily produced from oxidation of tetrahydrofuran via catalysis
– Lactone can be produced from condensation of molecule with alcohol group and carboxylic acid group (such as alkanoic acids with hydroxyl groups)
– Further investigation underway
tetrahydrofuran
Dihydro-5-methyl-2(3H)-furanone (C5H8O2)
05
1015202530
600 650 700 750 800 850
Temperature (K)
Mo
lar
fracti
on
(p
pm
)
28
Alkyl-substituted Lactones
M Phi #1
M Phi #2
H Phi
L Phi
Dihydro-5-ethyl-2(3H)-furanone (C6H10O2)
0
5
10
15
20
600 650 700 750 800 850
Temperature (K)
Mo
lar
fra
cti
on
(p
pm
)
Dihydro-5-pentyl-2(3H)-furanone (C9H16O2)
02468
10
600 650 700 750 800 850
Temperature (K)
Mo
lar
fra
cti
on
(p
pm
)
29
Alkyl-substituted Lactones
M Phi #1
M Phi #2
H Phi
L Phi
30
Alkane Oxidation Mechanism – Cyclic Ether Paths
• In general, cyclic ethers produced from decomposition of alkylhydroperoxy (QOOH.) radicals
HOORROOHRORRRH RHMOH +→+→ →←→ +++−2
2
2OH +22OH +ROOHOH +2
ROOHO +2
COROCROHCHOROR MO +′→′+→′+′′ ++ 2
2
2' OOROR ++
OROR '+
2'OR +
RH+
CORORCOHRCHOOHROOHQ MO +→+→+→ ++ 2
2'β
β RROH ′=+2
HOCHORRR +′′+′=
RRHO ′−+O
OQOOHOHOOQOOHR RH ←+ +
CHORORHOOROOHHOOOHRHOO ′′+′+→+→
RRR ′′=′+
β
MO ++ 2
RROH ′=+2
βROH +22
HO2 2O+
Tetrahydropyran-Z10-dodecenoate (C17H30O3)
0
3
6
9
600 650 700 750 800 850
Temperature (K)
Mo
lar
fra
cti
on
(p
pm
)
31
Ester-substituted Cyclic Ethers
M Phi #1
M Phi #2
H Phi
L Phi
• Low GC peak made identification questionable– Experiments at higher fuel
loadings may increase identification capabilities
2-Ethyl-5-methyl-furan (C7H10O)
0369
121518
600 700 800
Temperature (K)
Mo
lar
fra
cti
on
(p
pm
)
32
Alkyl-substituted Furans
M Phi #1M Phi #2
H PhiL Phi
• Possible dehydrogenation of 2-ethyl-5-methyl-tetrahydrofuran (a cyclic ether)
2-ethyl-5-methyl-tetrahydrofuran
33
Highlights and Ongoing / Future Work
• Four detailed experiments on the oxidation of n-dodecane in our PFR with NDIR and GC/MS/FID analysis have been conducted
• Over 30 species have been identified and quantified in the oxidation of n-dodecane at low and intermediate temperatures
• Further studies at similar initial conditions will be conducted to test for reproducibility so that experimental uncertainties can be characterized with high accuracy
• Collaboration with Hai Wang is underway so that our data are used in the development of the USC chemical kinetic model
• Coordination with Tom Bowman is underway so that our low and intermediate temperature flow reactor experiments mesh with the Stanford intermediate and high temperature flow reactor experiments
• Next hydrocarbon will be selected in collaboration with the consortium