Combustion Data for Jet-A, its Constituent Components, and Surrogate Mixtures Kamal Kumar* and Chih-Jen Sung†

* Department of Mechanical and Aerospace EngineeringCase Western Reserve University

† Department of Mechanical EngineeringUniversity of Connecticut

Multi-Agency Coordination Committee for Combustion Research – 2009 Fuels SummitGeneration of Comprehensive Surrogate Kinetic Models and

Validation Databases for Simulating Large Molecular WeightHydrocarbon Fuels

University of Southern CaliforniaLos Angeles, CA 90089

September 17, 2009

Key Tasks

• Acquire comprehensive experimental validation database for jet fuels, their constituents, and surrogate mixtures.– Autoignition delays and pre-ignition species evolutions at

elevated pressures and low-to-intermediate temperatures

– Fundamental flame properties, including laminar flame speeds and extinction stretch rates

• Establish the pattern of ignition response for the neat components vis-à-vis fully blended jet fuels.

• Explore the feasibility of reproducing combustion trends of Jet-A using simple binary/ternary surrogates.

Progress to Date• Autoignition of real jet fuels, including Jet-A, JP-8, and S-8.• Autoignition of neat hydrocarbon components under high

pressure conditions: n-decane, iso-octane, toluene, benzene, DME, DIB-1, MCH, 1,2,4-TMB, and xylenes.

• Chemical kinetic interactions in binary fuel blends: toluene+iso-octane, toluene+DIB-1, n-decane+iso-octane.

• Autoignition of Aachen surrogate and MURI mixture model.• Laminar flame speeds of preheated fuel/air mixtures:

MCH, toluene, iso-octane, n-heptane, n-decane, n-dodecane, Jet-A, and S-8.

• Extinction limits of preheated fuel/O2/N2 mixtures: n-decane, n-dodecane, Jet-A, and S-8.

Outline of Presentation• Autoignition Response

– Comprehensive ignition response for

a key paraffin component: n-decane

– Real jet fuels: Jet-A, JP-8, and S-8

– Jet-A constituents: aromatics, cycloalkanes, branched alkanes

– Binary blends and surrogates

• Flame Response– Laminar flame speeds and extinction stretch rates

– Neat surrogate components and real jet fuels

• Future Work

Autoignition Responseunder High Pressure Conditions

Test Matrix: n-DecaneMolar Ratio Molar composition (%)

n-C10H22 O2 N2 XFuel XO2 XN2

0.500 15.5 58.28 0.673 20.87 78.46

0.800 15.5 58.28 1.073 20.78 78.14

0.800 15.5 88.03 0.767 14.86 84.38

1.200 15.5 140.00 0.766 9.89 89.34

1.687 15.5 140.00 1.073 9.86 89.07

2.218 15.5 189.00 1.073 7.50 91.43

2.090 15.5 292.80 0.673 4.99 94.33

• Compressed Temperature: TC= 630 − 770 K

• Compressed Pressure: PC= 7 − 30 bar

n-Decane Autoignition

10-4

10-3

10-2

10-1

1.25 1.3 1.35 1.4 1.45 1.5 1.55 1.6

ExperimentCorrelation

e

xp (

T a/TC )

1000/TC (K-1)

1 = A PCn [XFuel]a [XO2]b exp( Ta /TC )

A = 3.21x10-11

a = -1.93

Ta = 12198

Overall Ignition Delay

b = -1.25n = -1.14

10-4

10-3

10-2

10-1

100

101

102

1.3 1.35 1.4 1.45 1.5 1.55 1.6

ExperimentCorrelation

e

xp (

T a/TC )

1000/TC (K-1)

1= A PCn [XFuel]a [XO2]b exp( Ta /TC )

A = 2.69x10-16

a = -0.83

Ta = 22370

First-Stage Ignition Delay

b = -0.36n = -0.40

First stage delay has higher T sensitivity

Data limited to two-stage activity region

Autoignition Comparison (1)

10-7

10-6

10-5

10-4

10-3

10-2

10-1

100

0.6 0.8 1 1.2 1.4 1.6

Zhao et al.Bikas & PetersUtah SurrogateDagaut & Cathonnet

Glaude et al.Current Work: = 0.5, 14.3 barPfahl et al.: = 0.5, 13 barWestbrook et al.

Igni

tion

Del

ay (s

)

1000/TC (K-1)

Comparative Mechanism PerformanceSimulation Results

n-Decane / Air =0.5; PC=15 bar

Range ofAll CurrentExperiments

Autoignition Comparison (2)

N2/(N2+O2)=0.790

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35

Computational

Experimental

Pres

sure

(bar

)

Time (ms)

Nonreactive Trace

End of Compression

n-Decane/Oxidizer, =0.8, PC=14.3 bar, TC=659 K

Westbrook et al.(2009)

Bikas & Peters(2001)

Ignition Delay Sensitivity

-40 -20 0 20 40QOOH+O2=>O2QOOH

Ketohydroperoxide Production

Ketohydroperoxide Decomposition

R-H+X=>R+X-H

RO2 =>QOOH

ROOH=>RO+OH

RO2+RO2=>RO+RO+O2

QOOH=>cyclic ether+OH

QOOH=>Q+HO2

Sensitivity of Ignition Delay to Different Reaction ClassesMechanism of Westbrook et al. (2008)

Percent Sensitivity

n-Decane/Air=0.8PC=14.3 barTC=659 K

• In general, the sensitivity of a particular class of reactions tends be in the same direction, for both 1 and 2.

• A noticeable exception is the ketohydroperoxide decomposition step, which is also the most endothermic of all the reactions pertinent to the low temperature chemistry.

• Increasing rate constants for the ketohydroperoxide decomposition class would decrease 1and increase 2, while minimally affect the total ignition delay.

Jet Fuels Comparison (1)

4

6

8

10

12

14

16

18

20

-5 0 5 10 15 20 25 30 35

(a) Jet Fuel/Air, O/F=13, TC673 K, PC=7 bar

Pres

sure

(bar

)

Time(ms)

Jet-A JP-8S-8

15

20

25

30

35

40

-5 0 5 10 15 20

(b) Jet Fuel/Air, O/F=13, TC664 K, PC=15 bar

Pres

sure

(bar

)

Time(ms)

Jet-A JP-8S-8

• Two-stage activity for all three fuels with similar first-stage ∆P.

• Nearly similar first stage ignition delay for conventional jet fuels.– Longer second stage ignition delay for JP-8.

• Shortest ignition delay for S-8.

Jet Fuels Comparison (2)

• Two-stage activity suppressed at this higher temperature.

• Shortest ignition delay for S-8, followed by Jet-A and JP-8.

2

4

6

8

10

12

14

16

18

-10 0 10 20 30 40

(c) Jet Fuel/Air, O/F=13, TC714 K, PC=7 bar

Pres

sure

(bar

)

Time(ms)

Jet-A JP-8S-8

Jet Fuels Comparison (3)

100

101

1.4 1.44 1.48 1.52 1.56 1.6 1.64

JP-8S-8Jet-A

Firs

t Sta

ge Ig

nitio

n D

elay

(ms)

1000/TC (K-1)

(a) Jet Fuel/Air, O/F=13, PC=7 bar

101

102

1.08 1.16 1.24 1.32 1.4 1.48 1.56 1.64

JP-8S-8Jet-AO

vera

ll Ig

nitio

n D

elay

(ms)

1000/TC (K-1)

(b) Jet Fuel/Air, O/F=13, PC=7 bar

• First stage ignition delays for conventional jet fuels are similar.• The alternative jet fuel has a shorter first stage ignition delay.• Overall ignition delays for all three jet fuels differ: S-8<Jet-A<JP-8.

Jet Fuels Comparison (4)

100

101

1.4 1.44 1.48 1.52 1.56 1.6 1.64

JP-8S-8Jet-A

Firs

t Sta

ge Ig

nitio

n D

elay

(ms)

1000/TC (K-1)

(a) Jet Fuel/Air, O/F=13, PC=15 bar

100

101

1.28 1.32 1.36 1.4 1.44 1.48 1.52 1.56 1.6

JP-8S-8Jet-A

Ove

rall

Igni

tion

Del

ay (m

s)

1000/TC (K-1)

(b) Jet Fuel/Air, O/F=13, PC=15 bar

• Similar first stage delay for Jet-A and JP-8.

• S-8 shows highest reactivity.

Ignition Delay Correlations: Jet-A

10-4

10-3

10-2

10-1

1.3 1.35 1.4 1.45 1.5 1.55 1.6

Experiment

Correlation/ [(

P C)n

[fu

el] a

[O

2] b

]

1000/TC (K-1)

= 1+2 = A (PC)n [fuel]a [O2]b exp(Ta /TC)

A = 2.57x10-10

a = -2.28; b=-1.08; n=-1.39Ta = 11275

(a) Overall Ignition Delay

635 K < TC < 750 K

0.0076 < fuel < 0.0148

0.1020 < O2 < 0.316710-4

10-3

10-2

10-1

100

101

102

1.3 1.35 1.4 1.45 1.5 1.55 1.6

Experiment

Correlation 1/ [

(PC

)n [

fuel

] a [

O2]

b)

1000/TC (K-1)

1 = A (PC)n [fuel]a [O2]b exp(Ta /TC)

A = 1.37x10-12

a = -1.11; b= -0.39; n= -0.74Ta = 17132

(b) First Stage Ignition Delay

635 K < TC < 750 K

0.0076 < fuel < 0.0148

0.1020 < O2 < 0.3167

10-2

10-1

100

101

0.9 1 1.1 1.2 1.3 1.4

Experiment

Correlation/ [(

P C)n

[fu

el] a

]

1000/TC (K-1)

= 1+2 = A (PC)n [fuel]a exp(Ta /TC)

A = 4.00x10-5

a = -1.46; n= -2.13Ta = 9090

(c) Overall Ignition Delay

750 K < TC < 1060 K0.0027 < fuel < 0.0146

XO2 ~10.25%; XAr ~50%; Balance N 2

Autoignition Comparison: Jet-A vs. Neat Components (1)

Test Conditions

Fuel A/F Mass Ratio Diluent

iso-Octane 13.0 1.16 N2

n-Decane 13.0 1.15 N2

Methylcyclohexane 12.7 1.16 N2

Toluene 15.4 1.16 Ar

1,2,4-Trimethyl Benzene 15.8 1.16 Ar

m-, o-, p-Xylene 15.6 1.16 Ar

MURI Mixture Model (49.5/28.2/22.3 liq. vol% n-decane/iso-octane/toluene ) 13.0 1.12 N2

Aachen Surrogate(20/80 mass% 1,2,4-TMB/n-decane) 13.0 1.13 N2

Jet-A (04POSF4658) 13.0 1.12 N2

Diluent : Oxygen = 3.76 : 1

1

10

100

0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7

Total Ignition Delay for Neat Components (PC=7 bar)

Igni

tion

Del

ay (m

s)

1000/TC (1/K)

i-C8

1,2,4-TMB

n-C10

Toluene

MCH

1

10

100

0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7

Ignition Delay Bounds for Neat Components (PC=7 bar)

Igni

tion

Del

ay (m

s)

1000/TC (1/K)

Toluene

MCH

Jet-A

n-C10

Autoignition Comparison: Jet-A vs. Neat Components (2)

1

10

100

0.85 0.9 0.95 1 1.05 1.1 1.15 1.2 1.25

Ignition Delay for Xylene Isomers (PC=7 bar)

o-Xylene

p-xylene

m-xylene

Igni

tion

Del

ay (m

s)

1000/TC (1/K)

Autoignition Comparison: Surrogate Mixtures vs. Neat Components

1

10

100

0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7

Ignition Delays for MURI Mixture Model and its Constituents (PC = 7 bar)

Igni

tion

Del

ay (m

s)

1000/TC (1/K)

Toluene

n-C10

MURIMixture Model

i-C8

1

10

100

0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7

Ignition Delays for Aachen Surrogate and its Constituents (PC=7 bar)

Igni

tion

Del

ay (m

s)

1000/TC (1/K)

n-C10

1,2,4-TMBAachen Surrogate

• Surrogate mixtures inherit the traits of the neat constituents

– NTC trend from i-C8 in the current range for MURI mixture model

Autoignition Comparison: Jet-A vs. Binary Mixtures

• Fuel mixtures with similar CN, but different constituents, can display markedly different ignition delay times in the low-to-intermediate temperature range.

• While CN is a parameter, further refinement is required to reproduce the ignition characteristics of fully blended fuels.

100

101

102

0.9 1 1.1 1.2 1.3 1.4 1.5 1.6

JP-8, A/F=13, DCN~40Jet-A, A/F=13, DCN~45

Ove

rall

Igni

tion

Del

ay (m

s)

1000/TC (K-1)

PC=7 bar

=1.16C6H6:O2:Ar1.16:7.5:28.2

CN=0

=1.16n-Decane/Air

A/F=13CN=77

=1.15n-C10H22+iso-C8H18/Air

A/F=13CN=48

100

101

102

1.3 1.4 1.5 1.6 1.7

iso-Octaneiso-Octane +n-Decane (50.79 :49.21)n-Decane

Ove

rall

Igni

tion

Del

ay (m

s)

1000/TC (K-1)

PC=7 bar, ~1.16

n-C10H22

iso-C8H18

iso-C8H18

+ n-C10H22

Autoignition Comparison: Jet-A vs. Surrogate Mixtures

1

10

100

0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7

Ignition Delays of Jet-A and Surrogates (PC=7 bar)

Jet-AAachen SurrogateMURI Mixture Model

Igni

tion

Del

ay (m

s)

1000/TC (1/K)

• Both surrogates perform well in the pre-NTC low-T range• MURI mixture model appears to capture NTC, while the Aachen

surrogate does not.

Laminar Flame Speeds and Extinction Limitsof Preheated Fuel/Air Mixtures

Laminar Flame Speed Comparison

30

40

50

60

70

80

90

100

110

10 12 14 16 18 20 22

Comparative Laminar Flame SpeedsJet-A (400 K)Jet-A (470 K)S-8 (400 K)S-8 (470 K)n-Decane (400 K)n-Decane (470 K)

Lam

inar

Fla

me

Spee

d (c

m/s

)

Air-to-Fuel Mass Ratio

Laminar Flame Speeds: Jet-A vs. Aachen Surrogate

• Aachen kerosene surrogate performs well at stoichiometric conditions.

30

40

50

60

70

80

90

100

110

10 12 14 16 18 20 22

Jet-A/Air Mixtures

Tu=400K

Tu=450K

Tu=470K

Lam

inar

Fla

me

Spee

d (c

m/s

)

Air-to-Fuel Mass Ratio

---- Computational at 400 and 470K, Achen Kerosene Surrogates 30

40

50

60

70

80

90

100

110

0.6 0.8 1 1.2 1.4

Jet-A/Air Mixtures

Tu=400K

Tu=470K

Tu=450K

Equivalence Ratio,

Lam

inar

Fla

me

Spee

d (c

m/s)

----- Computational at 400 and 470K, Aachen Kerosene Surrogate

Extinction Limit Comparison

0

200

400

600

800

0.8 1 1.2 1.4 1.6

Experimental Extinction Stretch Rates

Jet-AS-8 n-Decanen-DodecaneEx

tinct

ion

Stre

tch

Rate

(s-1

)

Equivalence Ratio,

N2/(N2+O2) = 0.84Tu=400 K

Extinction Stretch Rates: Jet-A vs. Aachen Surrogate

• Qualitative trend is well captured• Significantly higher extinction stretch rates predicted

– Fair agreement at lowest equivalence ratio

1500

1550

1600

1650

1700

1750

1800

1850

200 400 600 800 1000Max

imum

Fla

meT

empe

ratu

re (K

)

Stretch Rate (s-1)

0.89

N2/(N2+O2)=0.84

Aachen Kerosene Surrogate/O2/N2 Mixtures, Tu=400 K

0

200

400

600

800

1000

1200

0.8 1 1.2 1.4 1.6

Jet Fuel/O2/N2, Tu=400 K

Jet-A, Experimental

Computational, Aachen Surrogate

Extin

ctio

n St

retc

h R

ate

(s-1

)

Equivalence Ratio,

N2/(N2+O2) = 0.84

Laminar Flame Speed: Neat Component (1)

Laminar Flame Speed: Neat Component (2)

Flame Speed Comparison:Linear Alkane > Cycloalkane > Branched Alkane ≈Aromatic

Extinction Stretch Rate: Neat Component

Future Work

• Acquire autoignition data of neat surrogate components.– n-dodecane, n-propylbenzene, 1,3,5-TMB, xylenes, etc.– Coordinated with Fred Dryer and Ken Brezinsky.

• Acquire autoignition data of surrogate mixtures and MURI mixture models.

– Comparison with Jet-A.– Coordinated with Fred Dryer.

• Conduct high-pressure flame measurements for neat surrogate components, binary fuel blends, and real jet fuels.

– Coordinated with Yiguang Ju.

Acknowledgements

• Tim Edwards of AFRL for jet fuel supplies.

• Fred Dryer and Marcos Chaos of Princeton,

Hai Wang and Xiaoqing You of USC, as well

as Ken Brezinsky and Soumya Gudiyella of

UIC for thermodynamic property calculations.

• Graduate Students: Xin Hui and Bryan Weber.