Oxidation of Potential Surrogate Fuel Components of JP-8

Kenneth BrezinskyMechanical Engineering

University of Illinois, Chicago

2008 AFOSR MURI MeetingGeneration of Comprehensive Surrogate Kinetic Models and Validation Databases for

Simulating Large Molecular Weight Hydrocarbon Fuels

September 8, 2008

Simulate combustion behavior of JP-8 for improved combustor designs.

Inadequate thermochemical, kinetic and transport data for JP-8 surrogate components.

Selection of surrogate components, C8-C16 based on TSI, H/C and CN numbers.

Potential surrogate compounds – m-xylene, 1,3,5-trimethylbenzene, n-propylbenzene, 1,2,4-trimethylbenzene, n-decane, n-dodecane and methylcyclohexane

2%

60%

20%18%

Paraffins

NaphthenesAromatics

Olefins

Average Composition of JP-81

Purpose of Study

1. A.Violi, S. Yan, E.G.Eddings and A.F. Sarofim, Experimental Formulation and Kinetic Model for JP‐8 Surrogate Mixtures, Combust Sci and Tech., 174 (11&12): 399‐417, 2002

Technical Approach

Ignition delay studies, pyrolysis and oxidation experiments of individual surrogate fuel components, their mixtures and real fuels in the High Pressure Single Pulse Shock Tube (HPST).

Experimental regime:

Temperature: 800-2500K, pressure: 10-40atm, equivalence ratios: 0.5-4.0, time: 0.5-3ms.

Validate the experimental data against currently available literature models.

Develop chemical kinetic models for our experimental conditions.

Shock front 1

Contact surfa

ce

23

4

5

Reflected shockRarefaction fan

Reflected rarefaction

Distance x

Tim

e t

Driver DrivenDiaphragm

Head Tail

Shock Wave x-t Schematic

HPST Facility

Analytical: GC/MS, GC/FID-TCD

HPST Operating ConditionsTemperatures: 600-2500 KPressures: 5-1000 atmReaction Times: 0.5-3.0 ms

SamplingShock Tube Facility

Schematic of GCxGC apparatus

Analysis of diesel fuel; LECO CORPORATION, Separation Science Technical Note.

DURIP SUBMISSION

Program: CHEMKIN version 3.6.2Subroutine: SENKIN

Time evolution of homogeneous reacting gas mixtures in closed vessel.First order kinetic sensitivity analysis with respect to reaction rates.

Chemical kinetics models:Dagaut et al. (2007)2 model

Validated for jet-stirred reactor data at P = 1atm, Φ = 0.5-1.5, 900-1400K.

Battin-Leclerc et al. (2005)3 modelModeling of ortho-, meta- and para-xylenes.Validated for ignition delay times in shock tube, P = 6.7 to 9 bar , Φ = 0.5-2, 1330-1800K.

m-Xylene Oxidation ModelingP=22 bar, Φ=0.35

2. S.Gail., P.Dagaut., (2007) Oxidation of m‐Xylene in a JSR: Experimental Study and Detailed Chemical Kinetic Modeling. Combust. Sci. and Tech., 179: 813‐844.3. F.Battin~Leclerc, R. Bounaceur, N.Belmekki, P.A. Glaude., (2005) Wiley Interscience.

m-Xylene Oxidation, P=22 bar, Φ=0.35

1000 1100 1200 1300 1400 1500 1600

0

20

40

60

1000 1100 1200 1300 1400 1500 1600500

1000

1500

2000

1000 1100 1200 1300 1400 1500 1600

0

200

400

1000 1100 1200 1300 1400 1500 1600

0

10

20

30

40

0

5

10

15

1000 1100 1200 1300 1400 1500 16000

4

8

12

Mol

efra

ctio

n/pp

m

m-C8H10

Mol

efra

ctio

n/pp

m

O2

Experiment Battin-Leclerc Model Dagaut Model

Mol

efra

ctio

n/pp

m

Reflected Shock Temperature/K

CO

Mol

efra

ctio

n/pp

m

Reflected Shock Temperature/K

C2H2

Mol

efra

ctio

n/pp

m

C6H6

Mol

efra

ctio

n/pp

m C6H5CH3

Preliminary data-subject to revision

UIC m-Xylene Model:A hierarchical structure based on High Pressure Toluene

Oxidation Model4.

Elementary reactions of sequential oxidation and methyl side chains abstraction of m-xylene added to the High Pressure Oxidation Model.

Thermochemistry of the species taken from Dagaut Model.

UIC m-Xylene ModelHigh Pressure Toluene Oxidation Model4

+Sequential oxidation reactions of m-xylene

+Methyl side chain abstraction reactions of m-xylene

4. R.Sivaramakrishnan., R. S. Tranter., K.Brezinsky., (2005) A High Pressure Model for the Oxidation of Toluene, Proc. Combust. Inst., 30,  1165‐1173.

G5

G534

G571

G544

G536

G535

G2

G577

G14

G527

G456

G527

G127

-0.20 -0.15 -0.10 -0.05 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35

Oxidation of m-Xylene, P=22 bar, Φ=0.35Sensitivity Analysis of CO, t = 2.1 ms

Normalized Sensitivity Coefficient

T = 1342K T = 1283K T = 1170K

Sensitivity Analysis

G456. C6H5O=CO+C5H5

G527. C6H5O+O=C6H4O2+H

G534. MXYLYL+H=MXYLENE

G535. MXYLENE+H=C6H5CH3+CH3

G536. MXYLENE+H=MXYLYL+H2

G544. MXYLYL+O=C8H8O+H

G571. OC7H7=CO+OC6H7

G577. OC7H7+H=CH3C6H4OH

G127. CO+OH=CO2+H

G14. O+HO2=O2+OH

G5. H+O2(+M)=HO2(+M)

G2. H+O2=OH+O

Rxn # Reaction in ‘UIC m-Xylene Model’

1000 1100 1200 1300 1400 1500 1600400

800

1200

1600

2000

1000 1100 1200 1300 1400 1500 1600

0

10

20

1000 1100 1200 1300 1400 1500 1600

0

150

300

450

1000 1100 1200 1300 1400 1500 1600

0

20

40

60

1000 1100 1200 1300 1400 1500 16000

2

4

6

8

1000 1100 1200 1300 1400 1500 1600

0

4

8

12

16

Mol

efra

ctio

n/pp

m

O2

Reflected Shock Temperature/K

Mol

efra

ctio

n/pp

m

C2H2

Experiment UIC m-Xylene Model Battin-Leclerc Model

Mol

efra

ctio

n/pp

m

Reflected Shock Temperature/K

CO

Mol

efra

ctio

n/pp

mm-C8H10

Mol

efra

ctio

n/pp

m

C6H5CH3

Mol

efra

ctio

n/pp

m

C6H6

Preliminary data-subject to revision

m-Xylene Oxidation, P=22 bar, Φ=0.35

Chemical kinetics Mechanism:Based on m-xylene oxidation mechanism5.

UIC 135TMB Model:Detailed chemical kinetic model unavailable in literature.

13 species and 41 reactions added to the UIC m-Xylene Model.

Arrhenius parameters taken from analogous toluene reactions from High Pressure Toluene Oxidation Model.

1,3,5-Trimethylbenzene Oxidation Modeling

P=24 bar, Φ=1.26

UIC 1,3,5TMB ModelUIC m-Xylene Model+Sequential oxidation reactions of 1,3,5-trimethylbenzene

+Methyl side chain abstraction reactions of 1,3,5-trimethylbenzene

5. J.L. Emdee., K.Brezinsky., and I.Glassman., (1992) High‐temperature oxidation mechanisms of m‐ and p‐xylene. J. Phys. Chem., 95, 1626‐1635.

Thermochemistry

∆H0f 298K of stable compounds determined using Group Additivity based

Estimates6.

DFT calculation done for radicals using Gaussian 03, Scheme: B3LYP/6-31G(d).

∆H0f 0K of radicals determined using Ring Conserved Isodesmic

Reaction Scheme7 and DFT energies.

ΔHf0

298K estimated using DFT energies and statistical mechanics.

NASA polynomials obtained through FITDAT utility in CHEMKIN 3.7.

6.  http://webbook.nist.gov/chemistry/gpr‐add/ga‐app.shtml.7. R.Sivaramakrishnan., R. S. Tranter., K.Brezinsky., (2005) Ring Conserved Isodesmic Reactions: A New Method for Estimating the Heats of Formation of Aromatics and PAHs. J. Phys. Chem. 109, 1621‐1628.

1200 1400 1600 18000

40

80

120

1000 1200 1400 1600 18000

300

600

900

12001200 1400 1600 1800

0

200

400

600

1200 1400 1600 1800

0

200

400

600

800

1200 1400 1600 18000

2

4

6

1200 1400 1600 18000

4

8

12

Mol

efra

ctio

n/pp

m

C9H12

Mol

efra

ctio

n/pp

m

O2

Mol

efra

ctio

n/pp

m CO

Preliminary data-subject to revision

Reflected Shock Temperature/K

Experiment UIC 135TMB Model

Mol

efra

ctio

n/pp

m

Reflected Shock Temperature/K

CO2

Mol

efra

ctio

n/pp

m

Mol

efra

ctio

n/pp

m C2H4

C2H2

1,3,5-Trimethylbenzene Oxidation

P=24 bar, Φ=1.26

Program: CHEMKIN version 3.6.2Subroutine: SENKINChemical kinetics model:

Dagaut et al. (2002)8 modelValidated for jet-stirred reactor data at P = 1atm, Φ = 0.5-1.5,

900-1250K

n-Propylbenzene Oxidation ModelingP=17 bar, Φ=0.55

8. P.Dagaut, A. Ristori, A. El Bakali, M. Cathonnet., (2002) Experimental and kinetic modeling study of oxidation of   n‐propylbenzene. Fuel 81 173‐184.

n-Propylbenzene Oxidation, P=17 bar, Φ=0.55

■ Experiments ○ Dagaut’s Model

800 1000 1200 1400 1600 1800

0

10

20

30

40

50

60

70

800 1000 1200 1400 1600 1800

200

400

600

800

1000

1200

1400

1600

800 1000 1200 1400 1600 1800

0

100

200

300

400

800 1000 1200 1400 1600 1800

0

10

20

30

40

50

60

800 1000 1200 1400 1600 1800

0

100

200

300

400

500

600

700

800

800 1000 1200 1400 1600 1800

0

5

10

15

20

25

30

Mol

e Fr

actio

n /p

pm

C9H12

Mol

e Fr

actio

n /p

pm

O2

Mol

e Fr

actio

n /p

pm

CO

Mol

e Fr

actio

n /p

pm

C2H4

Mol

e Fr

actio

n /p

pm

Reflected Shock Temperature /K

CO2

Reflected Shock Temperature /K

Mol

e Fr

actio

n /p

pmC7H8

Preliminary data-subject to revision

1,3,5-Trimethylbenzene and n-PropylbenzeneMixture (1.5:1)

1000 1200 1400

0.0

0.2

0.4

0.6

0.8

1.0

1000 1200 1400 1600

0.0

0.2

0.4

0.6

0.8

1.0

D ecay o f 1 ,3 ,5-T rim ethylbenzenew ith and w ithout n-P ropylbenzene

D ecay of n -P ropylbenzenew ith and w ithout 1 ,3 ,5-T rim ethylbenzene

[ npb

] / [

npb

] o

Reflected Shock Tem perature /K

n-P ropylbenzene n-P ropylbenzene w /1,3,5-tm b

Prelim inary data-subject to revision

[ tm

b ] /

[ tm

b ] o

Reflected Shock Tem perature /K

1 ,3,5-T rim ethylbenzene 1,3,5-T rim ethylbenzene & n-propylbenzene

Summary

HPST:

Oxidation experiments of m-xylene, 1,3,5-trimethylbenzene and n-propylbenzene and a mixture of 1,3,5-trimethylbenzene and n-propylbenzene.

Experimental conditions:

Temperature: 924-1587K, pressure: 17-24bar, equivalence ratio: 0.35-1.4, residence time: 0.5-3ms.

Modeling:

Preliminary oxidation models developed for m-xylene and 1,3,5-trimethylbenzene

Validated n-propylbenzene experimental data using Dagaut’smodel.

Oxidation and pyrolytic experiments on surrogate fuel mixtures and real fuels.

Development of m-xylene and 1,3,5-trimethylbenzene models by inclusion of more comprehensive fuel decay pathways.

Future Work

AFOSR MURI FA9550-07-1-0515PI: Professor F.L. Dryer

http://www.princeton.edu/~combust/MURI

HPST Laboratory StudentsDr. R. Sivaramakrishnan

Acknowledgements