Experimental and Modeling Study on the Thermal Decomposition of

Jul 2, 2014 - *Telephone: +32-0-9-264-5677. ... decomposition chemistry of JP-10 and can be used to validate future kinetic models of JP-10 pyrolysis...
0 downloads 0 Views 851KB Size
Subscriber access provided by UNIV LAVAL

Article

Experimental and modeling study on the thermal decomposition of Jet Propellant-10 Nick M. Vandewiele, Gregory R. Magoon, Kevin Marcel Van Geem, Marie-Françoise Reyniers, William H. Green, and Guy B Marin Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/ef500936m • Publication Date (Web): 02 Jul 2014 Downloaded from http://pubs.acs.org on July 9, 2014

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Energy & Fuels is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Experimental and modeling study on the thermal decomposition of Jet Propellant-10 Nick M. Vandewielea, Gregory R. Magoonb, Kevin M. Van Geema, MarieFrançoise Reyniersa*, William H. Greenb, Guy B. Marina a

Laboratory for Chemical Technology, Universiteit Gent, Technologiepark 914 B-9052 Gent, Belgium b Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge, MA 02139, USA *

Corresponding

author:

Technologiepark

914

B-9052

Gent

Belgium,

e-mail:

[email protected], Phone: +32 (0)9 264 5677

Abstract JP-10 pyrolysis is performed in a continuous flow tubular reactor near atmospheric pressure in the temperature range of 930 – 1080K, a conversion range of 4-94% and two dilution levels of 7 and 10 mol % JP-10 in nitrogen. Identification and quantification of the pyrolysis products of JP10 is based on on-line two-dimensional gas chromatography with a time-of-flight mass spectrometer and a flame ionization detector. JP-10 starts to react at 930K and is fully converted at 1080K. Among the more than seventy species up to C14H10 that were identified and quantified, tricyclo[5.2.1.02,6]dec-4-ene was identified for the first time, indicating the importance of bimolecular H-abstraction reactions in the consumption of JP-10. Critical assessment of the experimental data with the JP-10 combustion model of Magoon et al.

1

showed that the model

predictions of JP-10 agree reasonably well. The newly acquired and highly detailed experimental 1 ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 30

data helps in understanding the thermal decomposition chemistry of JP-10 and can be used to validate future kinetic models of JP-10 pyrolysis. Keywords: JP-10; pyrolysis; thermal decomposition; PAH formation; GC×GC

1 Introduction Twenty-first century advanced aviation applications impose ever increasing demands on fuel performance 2-5. At present, Jet Propellant-10 is the only air-breathing missile fuel in operational use 4. This single-component fuel consists almost entirely of a tricyclic hydrocarbon, exotricyclo[5.2.1.02,6]decane (exo-TCD) that has superior qualities to other aviation fuels in terms of volumetric energy density 6-8, freezing point 4 and thermal stability [7, 8]. Next to the utilization in combustion devices, the fuel can also be used as a heat sink, i.e. the endothermic, chemical decomposition prior to combustion, in applications where severe demands are placed on the structural integrity and the thermal management of the propulsion technology 9-13. For the design and optimization of technologies using this fuel, a thorough understanding of the thermal decomposition of JP-10 is very helpful.

Although a lot of information is available about the thermochemical and thermophysical properties, less is known about the thermal decomposition chemistry of exo-TCD

14-16

. Many

authors identified components such as hydrogen, methane, ethene, propene, 1,3-cyclopentadiene (CPD), cyclopentene, benzene, and toluene as important pyrolysis products of JP-10 over a wide range of operating conditions

17-22

, while higher molecular weight intermediates were less

frequently observed. The latter could provide more insights into the primary decomposition routes of exo-TCD. Herbinet et al.

21

identified 3-cyclopentylcyclopentene as an important

primary product using mass spectrometry in experiments at atmospheric pressure between 848K 2 ACS Paragon Plus Environment

Page 3 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

and 933K. Under supercritical conditions at a pressure of 34 105 Pa and temperatures ranging from 373 to 923K, Wohlwend

18

and Striebich

19

detected substituted cyclopentenes, and

cyclopentadienes and a significant amount of aromatics in addition to benzene and toluene such as alkylbenzenes and naphthalene. Xing et al. 23 investigated the thermal cracking of exo-TCD at sub- and supercritical conditions and observed that the product spectrum under both regimes was very similar.

The agreement among the experimental studies with regards to the identification of the most important pyrolysis products is interesting since the operating conditions such as temperature, pressure, dilution and space time varied greatly. The shock tube of Davidson et al.

17

was

operated at temperatures between 1100K and 1700K and atmospheric pressure under very diluted conditions. The micro flow tube used by Nakra et al. 22 was used to investigate JP-10 pyrolysis on the millisecond timescale at temperatures between 900K and 1700K and pressures between 300 and 400 Pa. Wohlwend 18 and Striebich 19 studied the thermal stability and decomposition of exoTCD in a flow reactor under supercritical conditions at a pressure of 34 105 Pa and temperatures ranging from 373 to 923K. The jet-stirred reactor of Herbinet et al.

21

operated at atmospheric

pressure between temperatures of 848K and 933K in which 0.7 - 4 mol% JP-10 was fed. The batch reactor used by Xing et al. 23 contained undiluted JP-10 while pressure was increased from 1 105 – 37 105 Pa.

The lack of a fundamental understanding is not only limited to the major decomposition pathways, but also applies to the initial decomposition chemistry. The initial ring opening reactions of exo-TCD are hard to determine accurately, even using advanced, first-principles calculations 1. Moreover, it was shown

16

that the JP-10 feedstock contains small amounts of 3

ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 30

impurities which are often not reported or not taken into account in experimental studies. It is possible that these impurities may affect the initial decomposition of the main component of JP10, exo-TCD. The combination of the complexity of the chemical process, the lack of fundamental understanding of the chemistry, and the absence of accurate, chemical data for many of the model parameters resulted in a very limited set of kinetic models

1, 21, 24

, some of them originally

intended for combustion applications, rather than pyrolysis.

This work presents a comprehensive experimental dataset for the thermal decomposition of exoTCD. The JP-10 feedstock is characterized and the importance of impurities on the initial decomposition of exo-TCD is assessed. A diluent was added and operating conditions were chosen to complement previous experimental studies. The entire conversion range of exo-TCD is covered to gain more insights in the primary decomposition chemistry of exo-TCD, but also in the importance of polyaromatic hydrocarbons at higher conversions. Finally, the experimental data is used to critically assess the presently available thermal decomposition models that were proposed for JP-10.

2 Materials and Methods 2.1

Materials and experimental setup

JP-10 was provided by Sichuan Zhongbang Technical Developing Limited and was used without further purification. Analytical gases (N2, He, CO2) were obtained from Air Liquide at a minimum purity of 99.999%.

4 ACS Paragon Plus Environment

Page 5 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

JP-10 pyrolysis experiments were conducted in a continuous flow, tubular reactor. The reactor is a 1.475-m long, 6-mm internal diameter tube made of Incoloy 800HT (Ni, 30–35; Cr, 19 – 23; and Fe > 39.5 wt %). This bench-scale reactor set-up has been discussed extensively 25-27; a brief description is given in Section S1 of the Supplementary Information. The stream exiting the reactor (623K) is sent to a C5+ analysis section consisting of a GC×GC (Thermo Scientific Interscience Belgium) equipped with a Time of Flight Mass Spectrometer (ToF-MS) for the online identification of components in the effluent, and equipped with a Flame Ionization Detector (FID) for the quantification of peaks on the GC×GC chromatogram. The analysis section for C4- hydrocarbons consists of a dedicated GC equipped with two thermal conductivity detectors for the quantification of hydrogen and other permanent gases respectively, and one FID for the quantification of hydrocarbons up to C4.

The current experiments were carried out in presence of a diluent of nitrogen gas. The inert stream of nitrogen not only serves as a diluent but also as the primary internal standard to determine the absolute flow rates of the other components in the reactor effluent25. The flow rates of hydrogen, methane, and hydrocarbons up to C2 were calculated using the peak areas measured using the GC thermal conductivity detectors, the experimentally determined relative response factors for each of the detectors, and the known flow rate of the primary internal standard. Quantification of the peaks of the hydrocarbons with higher molecular masses on the GC-FID and GC × GC-FID was accomplished by considering ethene as a secondary internal standard for the detectors. A schematic of the experimental apparatus, and settings and characteristics of the analysis section such as column dimensions, temperature programs, component response factors

5 ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 30

and signal quantification are described in more detail in Section S1 of the Supplementary Information.

Experiments were carried out to cover the entire conversion range, by varying the temperature inside the reactor, while maintaining the mass flow rate of the JP-10 feedstock and diluent constant, and operating at constant pressure. Table 1 shows an overview of the applied conditions during the experiments. Table 1: Range of experimental conditions. Temperature range (K) Pressure (105 Pa) JP-10 mass flow rate (10-2 g s-1) Dilution levels of JP-10 in N2 (mol %) Conversion range (%)

936 – 1083 1.70 2.33 7 and 10 4–94

Two dilution levels were set by increasing the nitrogen flow rate while keeping the JP-10 flow rate constant. This results in dilution levels of 7 and 10 mol % JP-10 in nitrogen.

Overall mass balances taking into account all identified products attained on average 100 ± 5 wt%. Elemental balances for carbon closed within 1% and were then normalized. A rigorous derivation of the uncertainties associated with reported mole fractions is based on the uncertainties of the different variables that affect the calculation of the eventual mole fractions. The repeatability of the experiments is excellent, i.e. within 1% rel. for almost all the components. However, the uncertainty intervals for reported mole fractions is considered to be higher and is determined to be on average 5% rel. because the uncertainty on the measured mole fractions should account for among others uncertainties of the mass flow rates of JP-10 or the internal standard, uncertainties of the response factors, and uncertainties of the reported temperature measurements with Type K thermocouples. Thermocouple manufacturers adhere to the American Society for Testing Materials (ASTM) specifications for calibration accuracy (“limits of error”) 6 ACS Paragon Plus Environment

Page 7 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

for Type K thermocouples is: 0-1250°C: ±2.2°C or 0.75% of reading in °C, whichever is greater 28

. Response factors are experimentally determined for the main components. For the remainder

of the components the FID response factors are obtained using the effective carbon number concept29. In addition, even small errors for the response factor of the secondary internal standard (ethene) will affect the uncertainty for components detected on the GC × GC-FID as the secondary internal standard is used to determine the corresponding mass fractions. The formation of coke deposits was monitored by the pressure drop along the reactor. The pressure drop over the reactor was negligible under normal operating conditions. Typically less than 0.01 wt% of the feed is converted into cokes during the pyrolysis of hydrocarbons30.

2.2

Kinetic and reactor modeling

The obtained experimental data set was used to critically assess available kinetic models of the thermal decomposition of JP-10 (Table 2). Table 2: Kinetic models for thermal decomposition of JP-10. Model

San Diego 24

Magoon et al. 1

Primary application domain

Combustion

Combustion

Number of species

36

320

Number of reactions

174

Mechanism generator

Manual

7740 24

RMG 31

The first kinetic model for JP-10 combustion was developed by Li et al.

24

, hereafter referred to

as the “San Diego model”. The manually assembled model included 174 reactions among 36 species, primarily consisting of an earlier developed C1-C3 combustion model 32 consisting of 147 reactions between 33 species, while the remaining reactions comprised global pathways from exo-TCD to cyclopentene, 1,3-butadiene and lighter species. Herbinet et al.

21

assembled a

pyrolysis model containing 898 species and 2623 reactions, but this model is not publicly available. The most recent kinetic model for JP-10 was built by Magoon et al.

1

using the 7

ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Reaction Mechanism Generator code (RMG)

31

Page 8 of 30

, and hereafter referred to as the “RMG model”.

The latter comprised over 300 species and more than 7000 reactions.

The tubular reactor is modeled as an ideal plug flow reactor. Previous studies of this reactor 25, 26 showed that radial temperature and concentration gradients are negligible. The temperature and pressure along the axial coordinate of the reactor were set to the values measured during the experiments. Experimental conditions for each of the experiments can be found in Section S2 of the Supplementary Information. Numerical integration of each reactor simulation was carried out using the Chemkin-Pro package33.

3

Results and Discussion

3.1

JP-10 purity and conversion

JP-10 feedstock was analyzed off-line using a GC×GC–FID/ToF-MS. In the present study the JP10 feed contains three major components (Fig. 1), besides exo-TCD. Exo-TCD

Endo-TCD

a

448 419a 410a H H H

436a H

H 413a

H

H 412a H

H e

429b H

421c H

Adamantane

391d H 400 H

400e H

H Cis-Decalin

Fig. 1: Left: GC×GC chromatogram of JP-10 feedstock. Right: structure of exo-TCD and the major impurities in the JP-10 feedstock. Bond dissociation energies (BDE) of C-H bonds from literature. a) Ref 34 b)

8 ACS Paragon Plus Environment

Page 9 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Ref. 35 c) Ref. 36 d) Ref. 37 e) taken equal to BDE in cyclohexane.

Based on GC×GC–ToF-MS, the components were identified as endo-tricyclo[5.2.1.02,6]decane (endo-TCD), adamantane and cis-decalin. GC×GC/FID analysis reveals that the JP-10 sample consists of 98.43 mol% exo-TCD (in line with fuel specifications38), 0.66 mol% endo-TCD, 0.14 mol% adamantane, and 0.06 mol% decalin. Many components with significantly lower mole fractions were also observed with m/z ranging from 124 to 136 Da and collectively comprise 0.70 mol % of the feedstock. They are formed as by-products in the production of exo-TCD during the acid-catalyzed isomerization of endo-TCD to exo-TCD 39. Bruno et al.

16

reported similar levels

of these components and also provided tentative assignments of them.

Fig. 2 shows the conversion as a function of the average reactor temperature for exo-TCD, endoTCD and adamantane.

Fig. 2: Conversion as a function of average reactor temperature of ●: exo-TCD, ■: endo-TCD, and ♦: adamantane. Conditions: 10 mol % JP-10 in N2 at reactor inlet. P = 1.7 105 Pa. ± 5% error bars are indicated. Lines are spline interpolations.

The very small mole fractions of cis-decalin were not reported because of the level of uncertainty associated with them. Essentially full exo-TCD conversion for the experiments with 10 mol %

9 ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 30

exo-TCD at the reactor inlet occurs at 1080K. Conversion of endo-TCD is slightly higher than for exo-TCD, while adamantane becomes reactive at temperatures above 1000K. Since information on the C–C bond strengths is lacking, the global ring strain energy in the feedstock components was used to assess the importance of the initiation of the feedstock impurities. Only endo-TCD has a higher global ring strain energy than exo-TCD

34

while the other components are more

stable, cf. Table 3. Table 3: Standard enthalpies of formation ∆fH° at 298K, and component ring strain energies of exo-TCD and the major impurities in the JP-10 feedstock. Units are kJ mol-1. Exo-TCD Endo-TCD ∆fH°(298K) -81.6a -63.6a c Ring Strain Energy 95 113c a 34 b 40 c 41 d 42 Ref. , Ref. , Ref. , Ref.

Adamantane -134b 29c

Cis-Decalin -169 b 17d

It is therefore believed that adamantane, and cis-decalin, have a negligible impact on the initiation of exo-TCD. Endo-TCD may be important for the initiation of exo-TCD, despite the low mole fractions found in the feedstock. Next to the C–C bond strengths, carbon-hydrogen can be analyzed as well, to assess which C–H would be preferentially attacked by radicals. Bond dissociation energies (BDE) in decalin are the lowest of all C-H bonds in the four components (Fig. 1). As a consequence, radical H-abstraction reactions will therefore preferentially attack decalin C-H bonds but given the low amounts of this component its impact is believed to be limited.

3.2

Product distribution

Fig. 3 shows the GC×GC-FID chromatograms of two experiments at low and high exo-TCD conversion. 10 ACS Paragon Plus Environment

Page 11 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Fig. 3: GC×GC-FID chromatogram of pyrolysis products at average reactor temperatures 933K (upper) and 1083K (lower), corresponding to 4% and 90% exo-TCD conversion respectively.

The comprehensive GC×GC allows distinguishing between components based on their carbon numbers using the first dimension retention time of the chromatogram. Since sufficient resolution was available for the adequate quantification of lower molecular mass components, modulation was only initiated after the elution of the C4- components. The chromatogram at low conversion (933K) shows the peak of exo-TCD along with feedstock components such as adamantane and endo-TCD (Fig. 3). A cloud of small peaks (C9-C10) at the lower left of exo-TCD is visible. In this region both feedstock components and minor primary decomposition products were found with mole fractions of less than 1 mol %. Lower molecular mass products such as ethene, propene, CPD, cyclopentene, benzene, and toluene are also detected.

11 ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 30

The group of C9-C10 species observed at low conversion experiments completely disappeared on the chromatogram at 1083K and instead substituted monocyclic aromatic hydrocarbons such as ethylbenzene, styrene, propylbenzene, and indene were observed. The separation power of the second dimension column makes that heavier aromatics elute in the band indicated. To the right of exo-TCD, fused ring aromatic components such as methylindenes, naphthalene, methylnaphthalenes, and fluorene were observed. Hydrocarbons containing up to fourteen carbons, such as anthracene, were found in non-negligible quantities at conversions close to 100%.

3.2.1 Major pyrolysis products Mole fractions of all quantified components per experiment can be found in Section S3 of the Supplementary Information. Fig. 4 shows the mole fractions of the components with the highest mole fractions throughout the experiments as a function of the average temperature of the reactor.

12 ACS Paragon Plus Environment

Page 13 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Fig. 4: Normalized (excluding N2) mole fractions as a function of average reactor temperature of a) ○: exoTCD (left axis), ▲: ethene, ●: hydrogen, ■: methane, , ♦: 1,3-cyclopentadiene b) ●: propene, ■: benzene, ♦: toluene, ▲: cyclopentene. Conditions: 10 mol % exo-TCD in N2 at reactor inlet. P = 1.7 105 Pa. Lines are spline interpolations.

Hydrogen, methane, ethene, propene, CPD, cyclopentene, benzene and toluene are identified as important pyrolysis products, in agreement with other studies

18, 20-22

. The overall selectivity

towards these eight components is typically 75 wt%, starting from conversions higher than 40%. Mole fractions of hydrogen, ethene, CPD, and propene are significantly higher at low conversions than mole fractions of methane, benzene and toluene and may indicate that the former components are more prominent during the initial decomposition of exo-TCD than the latter components. Mole fractions of propene, CPD and cyclopentene stagnate as the exo-TCD conversion increases and eventually decrease, in contrast to the other major components. Maximum mole fractions of 8%, 12%, and 2% were observed for propene, CPD and cyclopentene at temperatures of 1045 K, 1040 K, and 990K respectively, and may indicate that they are further converted to secondary products.

3.2.2 Primary decomposition products Although components such as hydrogen and ethene prominently appear as exo-TCD decomposition products, their presence offers little mechanistic insights in the primary 13 ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 30

decomposition of exo-TCD. To elucidate the primary decomposition mechanism of exo-TCD, and to assess the importance of the many parallel decomposition pathways that are possible, higher molecular weight products should be identified that can only be formed through a limited number of possible pathways. Exo-TCD may be consumed via unimolecular scission reactions of one of the carbon-carbon bonds of exo-TCD that ruptures one of the rings. The detection of 3cyclopentylcyclopentene by Herbinet et al.

21

as a major primary decomposition product

corroborates the importance of these unimolecular scission reactions for the consumption of exoTCD. This should indicate the importance of the C–C scission leading to the biradical containing two non-fused five-membered rings (R 1, R 2). H

R1 H

BR1

R2

R3

Recent quantum-chemical calculations suggest that the rearrangement of exo-TCD into 3cyclopentylcyclopentene may also proceed as one concerted step (R 3) parallel to the reaction via a distinct biradical intermediate

1

(R 1, R 2). In the current study, 3-cyclopentylcyclopentene,

with retention time in between exo-TCD and endo-TCD 21, is not detected. Since experiments in the current study are performed at higher mole fractions of exo-TCD as compared to the experiments by Herbinet et al., bimolecular pathways such as hydrogen abstractions from exoTCD gain in importance and could explain the reduced formation of unimolecular decomposition products such as 3-cyclopentylcyclopentene. Adamantane is the only observed component with retention time between exo-TCD and endo-TCD (Fig. 3). An alternative to the unimolecular C-C 14 ACS Paragon Plus Environment

Page 15 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

scission reactions for the consumption of exo-TCD are H-abstraction reactions by radicals. Through these reactions, exo-TCD is converted into tricyclodecyl monoradicals (C10H15) with an intact tricyclo[5.2.1.02,6]decane ring structure. An experimental observation that suggests the importance of these exo-TCD decomposition channels is the detection of significant quantities of a component with molecular formula C10H14 with a maximum mole fraction of 0.35 mol % at 983K (Fig. 5). This component is tentatively identified as tricyclo[5.2.1.02,6]dec-4-ene. The tentative identification of tricyclo[5.2.1.02,6]dec-4-ene and other components was based on the comparison of measured and database (NIST Mass Spectral Library v2.6) mass spectra as illustrated in Section S4 of Supplementary Information. The C-H bond in the non-bridged cyclopentane ring in α-position to the bridgehead atom (Fig. 1) as the weakest C-H bond of exoTCD is identified as the weakest C-H bond of exo-TCD

34

. The formation of

tricyclo[5.2.1.02,6]dec-4-ene through TCDR5 by hydrogen abstraction reactions followed by a βscission reaction of an adjacent C-H bond (R 4) thus seems plausible.

R4

15 ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 30

Fig. 5: Normalized (excluding N2) mole fractions as a function of average reactor temperature of ●: 3-ethenylcyclopentene, ■: tricyclo[5.2.1.02,6]-dec-4-ene, ▲: 2-norbornene. Conditions: 10 mol % exo-TCD in N2 at reactor inlet. P = 1.7 105 Pa. Lines are spline interpolations.

A peak to the right of toluene on the low conversion chromatogram (Fig. 3) was tentatively identified as 3-ethenylcyclopentene and attains mole fractions of 1 mol % under the operating conditions of the experiments (Fig. 5). It can be formed from exo-TCD as a primary product through consecutive β-scission reactions starting from TCDR5 (R 5).

R5

2-Norbornene was detected in very small quantities of 0.05 mol % with a maximum at 980K (Fig. 5). 2-Norbornene may be formed via TCDR4 which lacks C–C β-scission reaction pathways that destroy the strained norbornane ring system (R 6). The formed radical subsequently yields 2norbornene and allyl via a β-scission reaction. It is known that 2-norbornene further decomposes to ethene and CPD via a concerted retro-Diels-Alder mechanism 43.

16 ACS Paragon Plus Environment

Page 17 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

H

H

H

H

H

H

TCD

TCDR4

+R

R6 2-norbornene

3.2.3 Secondary products The 1083K chromatogram (Fig. 3) shows both the appearance of new peaks with first dimension retention times between CPD and exo-TCD and an additional trail of higher molecular mass components with elevated second dimension retention times that elute after exo-TCD. Fig. 6 shows the mole fractions for some of the most important aromatics besides benzene and toluene.

Fig. 6: Normalized (excluding N2) mole fractions as a function of average reactor temperature of a) ●: propylbenzene, ■: styrene, ▲: ethylbenzene, b) ●: indene ■: naphthalene. Conditions: 10 mol % exoTCD in N2 at reactor inlet. P = 1.7 105 Pa. Lines are spline interpolations.

The sum of all components containing at least one aromatic ring amounts to 18 mol% at 1080K. Of the entire aromatics lump, mono-aromatic hydrocarbons comprise at least 70 mol % throughout the entire conversion range. A pronounced selectivity towards benzene, toluene, and styrene was observed with mole fractions at the highest temperature of 10%, 3% and 0.8% respectively. Ethylbenzene exhibits a maximum mole fraction at 1040K of 0.15 mol % (Fig. 6a). More than 20 individual polycyclic aromatic hydrocarbons (PAHs) with up to fourteen carbons were detected at temperatures starting from 980K and quantified. A total cumulative mole fraction of less than 3 mol % at 1090K is measured, with indene, and naphthalene the most prominently present (Fig. 6b). Indene and naphthalene attain effluent mole fractions (excluding 17 ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 30

nitrogen) of 1 mol % at 1090K. Other bicyclic PAHs such as methyl-substituted indenes, methylsubstituted naphthalenes, and biphenyl were found with mole fractions not exceeding 0.5 mol%. Indene-derived structures are formed at slightly lower temperatures than naphthalene.

The CPD derived 1,3-cyclopentadienyl radical is known to play a significant role in the formation of polycyclic aromatic hydrocarbons (PAHs) and soot

44-49

. Butler and Glassmann found that

naphthalene comprises almost 80% of the product spectrum during CPD pyrolysis at 1200K and atmospheric pressure

50

. Computational studies indicate the importance of 1,3-cyclopentadienyl

self-recombination 51 and 1,3-cyclopentadienyl addition to CPD 52, 53 as pathways to naphthalene. Indene was also observed as a major product of CPD pyrolysis

50

. Many routes from lower

molecular mass moieties were suggested such as the 1,3-cyclopentadienyl addition to CPD with methyl elimination (e.g.

52, 54

), and successive ethyne additions to CPD through seven-membered

cyclic moieties 55-57, or via 1,3-cyclopentadienyl self-recombination 58. The significant decline in CPD mole fractions occurs at a temperature where the PAHs are formed, suggesting that the formation of the observed PAHs is governed by successive growth chemistry initiated from abundant five-membered rings.

3.2.4 Effect of dilution The effect of dilution can be observed from Fig. 7.

18 ACS Paragon Plus Environment

Page 19 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Fig. 7: a) exo-TCD conversion as a function of average reactor temperature b) Normalized (excluding N2) mole fractions of ethane as a function of exo-TCD conversion. Conditions: 10 mol % exo-TCD (―, full symbols) and 7 mol % exo-TCD (- -, hollow symbols) in N2 at reactor inlet. P = 1.7 105 Pa. Lines are spline interpolations.

Higher exo-TCD conversions are observed as a function of the average reactor temperature (Fig. 7a) at lower levels of dilution, at a given space time. Mole fractions of ethane (Fig. 7b), for which the effect of dilution is most pronounced, as a function of conversion are higher for lower levels of dilution. A plausible explanation for the increased exo-TCD conversion at higher hydrocarbon partial pressures is that a large fraction of exo-TCD is consumed via bimolecular reactions, as these reactions are favored over unimolecular reactions by higher hydrocarbon partial pressures. Similar explanations can be drawn for ethane, which is primarily formed through bimolecular reactions, such as H-abstraction reactions by ethyl.

3.3

Kinetic model validation

In this section, the predictions of the San Diego model and the RMG model are compared against the experimental data. The San Diego model, which was built with the purpose of modeling JP10 ignition, was previously validated against a wide range of experimentally measured ignition delays

59, 60

and showed remarkable agreement with the data despite the limited detail of the

kinetic model. However, comparison of the model predictions with the current experimental data

19 ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 30

shows that the exo-TCD conversion is systematically underpredicted (Fig. 8), and makes the model less suitable to model the thermal decomposition of JP-10 in absence of oxygen, at temperatures of less than 1100K.

Fig. 8: JP-10 conversion as a function of temperature for two dilution levels of JP-10 in N2. Experimental data: ●/○: 10 / 7 mol% JP-10 in N2. Predictions of San Diego model 24: ― and- -: 10 and 7 mol% JP-10 in N2. P = 1.7 105 Pa.

Comparison of the mole fractions of components as a function of JP-10 conversion (Fig. 9) shows that the selectivity of some of the major products predicted by the San Diego model also shows significant discrepancies.

20 ACS Paragon Plus Environment

Page 21 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Fig. 9: Mole fractions as a function of JP-10 conversion for a) experimental: ●: hydrogen, ▲: ethene, ■: propene, predictions: ―: hydrogen, – -: ethene, – –: propene, b) experimental: ♦: ethyne, ▼: 1,3-butadiene, predictions: ―: ethyne, – –: 1,3-butadiene, c) experimental: ◄: cyclopentene, predictions: ―: cyclopentene. Conditions: 10 mol % JP-10 in N2 at reactor inlet. P = 1.7 105 Pa. Predictions of San Diego model 24.

In the San Diego model, a pronounced selectivity is observed towards ethene, ethyne, and 1,3butadiene while mole fractions of hydrogen, propene and cyclopentene are significantly underpredicted. These observations can be explained by analyzing the relative importance of the global decomposition pathways of JP-10. Reaction R 7, which represents a direct pathway to ethyne, ethene, and 1,3-butadiene, appears to be very dominant under the investigated conditions, and explains the pronounced selectivity for these components versus competing pathways to hydrogen or cyclopentene (R 8, R 9). R7 Exo-TCD → C2H2 + 2 C2H4 + C4H6 Exo-TCD → H + C2H4 + C3H3 + C5H8 R8 R9 Exo-TCD → H + C2H2 + C3H5 + C5H8 Rate parameters for the global decomposition pathways of JP-10 in the San Diego model were derived from earlier studies of decane and heptane

61-63

. The observed discrepancies between

model predictions and experiments in this study indicate that this approach is not sufficient for obtaining accurate, quantitative predictions of the distribution of the products of JP-10 pyrolysis.

21 ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 30

Although the simplified San Diego model has its merits as an engineering tool to provide accurate predictions for a single observable such as ignition delays, it lacks fundamental, molecular detail, which should make it more robust. In this regard, the RMG combustion model is a step forward towards a more fundamental understanding of JP-10 thermal decomposition as it consists of detailed, elementary steps for the decomposition chemistry of JP-10. Fig. 10 shows the JP-10 conversion as a function of temperature of the current experimental data versus the predictions of the RMG model 1.

Fig. 10: JP-10 conversion as a function of temperature for two dilution levels of JP-10 in N2. Experimental data: ●: 10 mol%, and ○: 7 mol % JP-10 in N2. Predictions of the RMG model 1: ―: 10 mol%, and - -: 7 mol % JP-10 in N2. P = 1.7 105 Pa.

The RMG model results in better model predictions for the exo-TCD conversion in the current pyrolysis experiments than the San Diego model, but still slightly underestimates the conversion. Fig. 11 shows the mole fractions of major products as a function of exo-TCD conversion. Predictions of hydrogen and methane (Fig. 11a) are slightly higher than experimentally observed. Ethene and propene are significantly underpredicted across the entire exo-TCD conversion range (Fig. 11b). On the other hand, predicted mole fractions of CPD and cyclopentene agree with experiments at lower conversions (Fig. 11c). While the experimentally observed decrease of

22 ACS Paragon Plus Environment

Page 23 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

cyclopentene mole fractions is present in the RMG model predictions too, this is not the case for CPD.

Fig. 11: Mole fractions as a function of conversion for a) experimental: ●: hydrogen, ■: methane, predictions: ―: hydrogen, – -: methane, b) experimental: ▼: ethene, ♦: propene, predictions: ―: ethene,– -:propene, c) experimental: ►: 1,3-cyclopentadiene, *: cyclopentene, predictions: ―: 1,3-cyclopentadiene, – -: cyclopentene. Conditions: 10 mol % JP-10 in N2 at reactor inlet. P = 1.7 105 Pa. Predictions of RMG model 1.

At 17% conversion, the fraction of exo-TCD converted through unimolecular initiation routes via biradicals BR1, BR2, and BR3 amounts to 18% (Fig. 12).

23 ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 30

Fig. 12: Reaction pathway analysis for the decomposition of exo-TCD with the RMG model 1. The reported percentages represent the reaction rate relative to the total exo-TCD decomposition rate. Conditions: T=1100K, P=1.7 105 Pa, conversionexo-TCD = 17%, 10 mol% exo-TCD in N2 at the reactor inlet. Only major exo-TCD decomposition pathways are shown.

Exo-TCD is primarily consumed via three H-abstraction reactions. TCDR5 further decomposes to a cyclopentylcyclopentene radical isomer (C10H15), while a channel to the experimentally observed tricyclo[5.2.1.02,6]dec-4-ene is lacking. Many primary decomposition pathways lead to fused, bicyclic stable products rather than small components such as ethene and propene and may explain the underpredicted mole fractions of these components. For CPD, the maximum in mole fractions at 50% exo-TCD conversion is not present in the model calculations because CPD decomposition reactions are lacking. Important aromatics such as benzene, indene and naphthalene are also absent in the model. Furthermore, the reactant in the San Diego and RMG model was represented as 100% exo-TCD, thus neglecting the 5% endo-TCD, because the endo isomer was not present in the model. Since the endo-isomer is less stable compared to the exo-

24 ACS Paragon Plus Environment

Page 25 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

isomer 34, it is possible that the initiation of exo-TCD is influenced by radicals originating from endo-TCD. Future kinetic modeling efforts need to investigate this is in more detail.

4 Conclusions New experimental data was presented in this work for exo-TCD pyrolysis under mildly diluted conditions. The data shows that exo-TCD starts to decompose at 930K and attains full conversion at 1080K at the investigated conditions. Hydrogen, ethene, cyclopentadiene and benzene were among the major decomposition products of exo-TCD in line with literature. The presence of tricyclo[5.2.1.02,6]dec-4ene as a primary product suggested that bimolecular exo-TCD conversion pathways become important in contrast to earlier findings.

Important pathways to PAHs were identified to start from CPD. These reactions result in the formation of naphthalene, indene, and substituted derivatives of these bicyclic aromatic components. Indene derived moieties are further converted into tricyclic PAHs such as fluorenes and confirmed the propensity of strained-ring hydrocarbons to soot, and similar pyrolytic deposition in hypersonic engine designs.

The agreement between experimental data and predictions of the RMG model compared to the San Diego model suggest that detailed, microkinetic models built by automatic model construction algorithms are a promising tool for detailed kinetic modeling of the thermal decomposition of complex fuels such as JP-10. The agreement between the RMG model predictions and the experiments is altogether remarkable since none of the model parameters were fitted to the experimental data. The insights for the thermal decomposition of JP-10 gained

25 ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 30

through the experiments of this study will help in constructing new kinetic models for JP-10 under pyrolysis conditions.

5 Acknowledgements Nick M. Vandewiele acknowledges financial support for a doctoral fellowship from the Fund for Scientific Research – Flanders. Financial support from the Long Term Structural Methusalem Funding by the Flemish Government – grant number BOF09/01M00409 is acknowledged. Gregory R. Magoon and William H. Green acknowledge financial support from the Naval Air Warfare Center under contract number N68335-10-C-0534. The authors acknowledge Maarten Eestermans for analyzing the JP-10 pyrolysis experiments and thank Henning Richter for useful discussions.

6 Supplementary Information This information is available free of charge via the Internet at http://pubs.acs.org/.

7 References 1. Magoon, G. R.; Aguilera-Iparraguirre, J.; Green, W. H.; Lutz, J. J.; Piecuch, P.; Wong, H. W.; Oluwole, O. O., Detailed chemical kinetic modeling of JP-10 (exotetrahydrodicyclopentadiene) high-temperature oxidation: Exploring the role of biradical species in initial decomposition steps. International Journal of Chemical Kinetics 2012, 44, (3), 179-193. 2. Maurice, L. Q.; Lander, H.; Edwards, T.; Harrison, W. E., Advanced aviation fuels: a look ahead via a historical perspective. Fuel 2001, 80, (5), 747-756. 3. Edwards, T., USAF Supercritical Hydrocarbon Fuels Interests. In: 31 st Aerospace Sciences Meeting & Exhibit, Reno, Nevada, USA, 1993. 4. Edwards, T., Advancements in Gas Turbine Fuels From 1943 to 2005. Journal of Engineering for Gas Turbines and Power 2007, 129, (1), 13-20. 5. Chung, H. S.; Chen, C. S. H.; Kremer, R. A.; Boulton, J. R.; Burdette, G. W., Recent developments in high-energy density liquid hydrocarbon fuels. Energy & Fuels 1999, 13, (3), 641-649. 26 ACS Paragon Plus Environment

Page 27 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

6. Van Devener, B.; Anderson, S. L., Breakdown and combustion of JP-10 fuel catalyzed by nanoparticulate CeO2 and Fe2O3. Energy & Fuels 2006, 20, (5), 1886-1894. 7. Osmont, A.; Gokalp, I.; Catoire, L., Evaluating missile fuels. Propellants Explosives Pyrotechnics 2006, 31, (5), 343-354. 8. Meylemans, H. A.; Quintana, R. L.; Harvey, B. G., Efficient conversion of pure and mixed terpene feedstocks to high density fuels. Fuel 2012, 97, (0), 560-568. 9. Cooper, M.; Shepherd, J. E. Thermal and Catalytic Cracking of JP-10 for Pulse Detonation Engine Applications; Graduate Aeronautical Laboratories, California Institute of Technology: Pasadena, California, USA, 2002. 10. Heneghan, S. P.; Harrison, W. E. In JP-8+100: The Development of High Thermal Stability Jet Fuel, 6th International Conference on Stability and Handling of Liquid Fuels, Vancouver, Canada, 1997; Vancouver, Canada, 1997. 11. Huang, H.; Sobel, D. R.; Spadaccini, L. J., Endothermic Heat-Sink of Jet Fuels for Scramjet Cooling. In 38th AIAA/ASME/SAE/ASEE Joint Propulsion Conference and Exhibit, AIAA Journal: Indianapolis, Indiana, USA, 2002. 12. Xing, Y.; Li, D.; Xie, W. J.; Fang, W. J.; Guo, Y. S.; Lin, R. S., Catalytic cracking of tricyclo [5.2.1.0(2.6)] decane over HZSM-5 molecular sieves. Fuel 2010, 89, (7), 1422-1428. 13. Hou, L.-Y.; Dong, N.; Sun, D.-P., Heat transfer and thermal cracking behavior of hydrocarbon fuel. Fuel 2013, 103, 1132-1137. 14. Meylemans, H. A.; Baldwin, L. C.; Harvey, B. G., Low-Temperature Properties of Renewable High-Density Fuel Blends. Energy & Fuels 2013, 27, (2), 883-888. 15. Xie, W.; Fang, W.; Li, D.; Xing, Y.; Guo, Y.; Lin, R., Coking of Model Hydrocarbon Fuels under Supercritical Condition. Energy & Fuels 2009, 23, 2997-3001. 16. Bruno, T. J.; Huber, M. L.; Laesecke, A.; Lemmon, E. W.; Perkins, R. A. Thermochemical and Thermophysical Properties of JP-10; 2006. 17. Davidson, D. F.; Horning, D. C.; Oehlschlaeger, M. A.; Hanson, R. K., The Decomposition Products of JP-10. American Institute of Aeronautics and Astronautics Journal 2001, 3707, (1). 18. Wohlwend, K.; Maurice, L. Q.; Edwards, T.; Striebich, R. C.; Vangsness, M.; Hill, A. S., Thermal stability of energetic hydrocarbon fuels for use in combined cycle engines. J. Propul. Power 2001, 17, (6), 1258-1262. 19. Striebich, R. C.; Lawrence, J., Thermal decomposition of high-energy density materials at high pressure and temperature. Journal of Analytical and Applied Pyrolysis 2003, 70, (2), 339352. 20. Rao, P. N.; Kunzru, D., Thermal cracking of JP-10: Kinetics and product distribution. Journal of Analytical and Applied Pyrolysis 2006, 76, (1-2), 154-160. 21. Herbinet, O.; Sirjean, B.; Bounaceur, R.; Fournet, R.; Battin-Leclerc, F.; Scacchi, G.; Marquaire, P. M., Primary mechanism of the thermal decomposition of tricyclodecane. Journal of Physical Chemistry A 2006, 110, (39), 11298-11314. 22. Nakra, S.; Green, R. J.; Anderson, S. L., Thermal decomposition of JP-10 studied by micro-flowtube pyrolysis-mass spectrometry. Combustion and Flame 2006, 144, (4), 662-674. 23. Xing, Y.; Fang, W. J.; Xie, W. J.; Guo, Y. S.; Lin, R. S., Thermal Cracking of JP-10 under Pressure. Industrial & Engineering Chemistry Research 2008, 47, (24), 10034-10040. 24. Li, S. C.; Varatharajan, B.; Williams, F. A., Chemistry of JP-10 ignition. American Institute of Aeronautics and Astronautics J. 2001, 39, (12), 2351-2356.

27 ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 30

25. Harper, M. R.; Van Geem, K. M.; Pyl, S. P.; Marin, G. B.; Green, W. H., Comprehensive reaction mechanism for n-butanol pyrolysis and combustion. Combustion and Flame 2011, 158, (1), 16-41. 26. Van Geem, K. M.; Pyl, S. P.; Marin, G. B.; Harper, M. R.; Green, W. H., Accurate HighTemperature Reaction Networks for Alternative Fuels: Butanol Isomers. Industrial & Engineering Chemistry Research 2010, 49, (21), 10399-10420. 27. Van Geem, K. M.; Cuoci, A.; Frassoldati, A.; Pyl, S. P.; Marin, G. B.; Ranzi, E., An Experimental and Kinetic Modeling Study of Pyrolysis and Combustion of Acetone–Butanol– Ethanol (ABE) Mixtures. Combustion Science and Technology 2012, 184, (7-8), 942-955. 28. Nakos, J. T., Uncertainty analysis of thermocouple measurements used in normal and abnormal thermal environment experiments at Sandia's Radiant Heat Facility and Lurance Canyon Burn Site. United States. Department of Energy: 2004. 29. Beens, J.; Boelens, H.; Tijssen, R.; Blomberg, J., Quantitative aspects of comprehensive two-dimensional gas chromatography (GC x GC). Journal of High Resolution Chromatography 1998, 21, (1), 47-54. 30. Dhuyvetter, I.; Reyniers, M. F.; Froment, G. F.; Marin, G. B.; Viennet, D., The influence of dimethyl disulfide on naphtha steam cracking. Industrial & Engineering Chemistry Research 2001, 40, (20), 4353-4362. 31. Green, W. H.; Allen, J. W.; Ashcraft, R. W.; Beran, G. J.; Class, C. A.; Gao, C.; Goldsmith, C. F.; Harper, M. R.; Jalan, A.; Magoon, G. R.; Matheu, D. M.; Merchant, S. S.; Mo, J. D.; Petway, S.; Raman, S.; Sharma, S.; Song, J.; Van Geem, K. M.; Wen, J.; West , R. H.; Wong, A.; Wong, H. S.; Yelvington, P. E.; Yu, J., RMG - Reaction Mechanism Generator v4.0. 2013. 32. Varatharajan, B.; Williams, F. A., Ethylene ignition and detonation chemistry, Part 1: Detailed modeling and experimental comparison. J. Propul. Power 2002, 18, (2), 344-351. 33. CHEMKIN-PRO, 150101; Reaction Design: San Diego, 2009. 34. Hudzik, J. M.; Asatryan, R.; Bozzelli, J. W., Thermochemical Properties of exo-Tricyclo 5.2.1.0(2,6) decane (JP-10 Jet Fuel) and Derived Tricyclodecyl Radicals. Journal of Physical Chemistry A 2010, 114, (35), 9545-9553. 35. Fattahi, A.; Lis, L.; Tehrani, Z. A.; Marimanikkuppam, S. S.; Kass, S. R., Experimental and Computational Bridgehead C-H Bond Dissociation Enthalpies. Journal of Organic Chemistry 2012, 77, (4), 1909-1914. 36. Abboud, J. L. M.; Castano, O.; Davalos, J. Z.; Gomperts, R., The standard enthalpies of formation of 1-and 2-Adamantyl cations and radicals. An ab initio study. Chem. Phys. Lett. 2001, 337, (4-6), 327-330. 37. Luo, Y.-R., Handbook of Bond Dissociation Energies in Organic Compounds. Taylor & Francis: 2002. 38. Council, C. R. Handbook of Aviation Fuel Properties; Atlanta, Georgia, 1983; p 122. 39. Huang, M.-Y.; Wu, J.-C.; Shieu, F.-S.; Lin, J.-J., Preparation of high energy fuel JP-10 by acidity-adjustable chloroaluminate ionic liquid catalyst. Fuel 2011, 90, (3), 1012-1017. 40. Boyd, R. H.; Sanwal, S. N.; Sharyteh.S; McNally, D., Thermochemistry, Thermodynamic Functions, and Molecular Structures of some Cyclic Hydrocarbons. Journal of Physical Chemistry 1971, 75, (9), 1264-1271. 41. Gasteiger, J.; Dammer, O., Automatic Estimation of Ring Strain Energies. Tetrahedron 1978, 34, (19), 2939-2945. 42. Chang, S.; Boyd, R. H.; McNally, D.; Sharyteh.S; Hickey, M. J., Heats of Combustion and Strain Energies of Bicyclo[N.M.0]-Alkanes. J. Am. Chem. Soc. 1970, 92, (10), 3109-3122. 28 ACS Paragon Plus Environment

Page 29 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

43. Kiefer, J. H.; Kumaran, S. S.; Sundaram, S., Vibrational Relaxation, Dissociation, and Dissociation Incubation Times in Norbornene. J. Chem. Phys. 1993, 99, (5), 3531-3541. 44. D'Anna, A.; Violi, A., Detailed modeling of the molecular growth process in aromatic and aliphatic premixed flames. Energy & Fuels 2005, 19, (1), 79-86. 45. Skjoth-Rasmussen, M. S.; Glarborg, P.; Ostberg, M.; Johannessen, J. T.; Livbjerg, H.; Jensen, A. D.; Christensen, T. S., Formation of polycyclic aromatic hydrocarbons and soot in fuel-rich oxidation of methane in a laminar flow reactor. Combustion and Flame 2004, 136, (1-2), 91-128. 46. Richter, H.; Howard, J. B., Formation of polycyclic aromatic hydrocarbons and their growth to soot - a review of chemical reaction pathways. Progress in Energy and Combustion Science 2000, 26, (4-6), 565-608. 47. Lindstedt, P.; Maurice, L.; Meyer, M., Thermodynamic and kinetic issues in the formation and oxidation of aromatic species. Faraday Discussions 2001, 119, 409-432. 48. Marinov, N. M.; Pitz, W. J.; Westbrook, C. K.; Vincitore, A. M.; Castaldi, M. J.; Senkan, S. M.; Melius, C. F., Aromatic and polycyclic aromatic hydrocarbon formation in a laminar premixed n-butane flame. Combustion and Flame 1998, 114, (1-2), 192-213. 49. Lamprecht, A.; Atakan, B.; Kohse-Hoinghaus, K., Fuel-rich flame chemistry in lowpressure cyclopentene flames. Proceedings of the Combustion Institute 2000, 28, 1817-1824. 50. Butler, R. G.; Glassman, I., Cyclopentadiene combustion in a plug flow reactor near 1150 K. Proc. Combust. Inst. 2009, 32, 395-402. 51. Melius, C. F.; Colvin, M. E.; Marinov, N. M.; Pitz, W. J.; Senkan, S. M., Reaction mechanisms in aromatic hydrocarbon formation involving the C5H5 cyclopentadienyl moiety. 26th International Symposium on Combustion: Naples, Italy, 1996; p 685-692. 52. Wang, D.; Violi, A.; Kim, D. H.; Mulholland, J. A., Formation of naphthalene, indene, and benzene from cyclopentadiene pyrolysis: A DFT study. Journal of Physical Chemistry A 2006, 110, (14), 4719-4725. 53. Alder, R. W.; East, S. P.; Harvey, J. N.; Oakley, M. T., The azulene-to-naphthalene rearrangement revisited: a DFT study of intramolecular and radical-promoted mechanisms. J. Am. Chem. Soc. 2003, 125, (18), 5375-5387. 54. Kislov, V. V.; Mebel, A. M., An ab initio G3-type/statistical theory study of the formation of indene in combustion flames. II. The pathways originating from reactions of cyclic C-5 species - Cyclopentadiene and cyclopentadienyl radicals. Journal of Physical Chemistry A 2008, 112, (4), 700-716. 55. Cavallotti, C.; Mancarella, S.; Rota, R.; Carra , S., Conversion of C5 into C6 Cyclic Species through the Formation of C7 Intermediates The Journal of Physical Chemistry A 2007, 111, (19), 3959-3969. 56. Fascella, S.; Cavallotti, C.; Rota, R.; Carra, S., The peculiar kinetics of the reaction between acetylene and the cyclopentadienyl radical. Journal of Physical Chemistry A 2005, 109, (33), 7546-7557. 57. Fascella, S.; Cavallotti, C.; Rota, R.; Carra , S., Quantum Chemistry Investigation of Key Reactions Involved in the Formation of Naphthalene and Indene. The Journal of Physical Chemistry A 2004, 108, (17), 3829-3843. 58. Robinson, R. K.; Lindstedt, R. P., On the chemical kinetics of cyclopentadiene oxidation. Combustion and Flame 2011, 158, (4), 666-686. 59. Davidson, D. F.; Horning, D. C.; Herbon, J. T.; Hanson, R. K., Shock tube measurements of JP-10 ignition. Proceedings of the Combustion Institute 2000, 28, 1687-1692. 29 ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 30

60. Colket, M. B.; Spadaccini, L. J., Scramjet fuels autoignition study. J. Propul. Power 2001, 17, (2), 315-323. 61. Held, T. J.; Marchese, A. J.; Dryer, F. L., A semi-empirical reaction mechanism for nheptane oxidation and pyrolysis. Combustion Science and Technology 1997, 123, (1-6), 107-146. 62. Douté, C.; Delfau, J.-L.; Vovelle, C., Modeling of the structure of a premixed n-decane flame. Combustion science and technology 1997, 130, (1-6), 269-313. 63. Marchese, A. J.; Dryer, F. L.; Nayagam, V., Numerical modeling of isolated n-alkane droplet flames: Initial comparisons with ground and space-based microgravity experiments. Combustion and Flame 1999, 116, (3), 432-459.

30 ACS Paragon Plus Environment