• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

Pyrolysis of organic molecules relevant to combustion as monitored by photoionization time-of-flight mass spectrometry

Dissertation
Author: Kevin Howard Weber
Abstract:
Flash pyrolysis coupled to molecular beam extraction and single photon ionization time-of-flight mass spectrometry along with quantum chemical calculations are employed to study the pyrolysis of organic molecules with relevance to combustion processes. Branching ratios for the molecular elimination and bond fission pathways was achieved for ethyl and propyl iodides providing information about the nature of the mechanism with relation to chemical structure. Similarly, the decompositions of a series of alkyl methyl ethers, whose anti-knock ability in commercial fuels is attributed to the molecular elimination pathway, was investigated to examine the competition of primary and subsequent homolyses at higher temperatures, which tend to promote "knock". The isomerization/decomposition of isoprene was looked at in detail with special attention to the formation of soot precursor, most notably the "first ring" benzene. Cyclopentadiene and methylcyclopentadiene are known to be highly sooting fuels and were pyrolyzed to study the initial steps in aromatization. The pyrolysis of cyclohexene, cyclopentene, and 1,4-cyclohexadiene were conducted to facilitate interpretation of aromatic formation and to compare experimental results with well established mechanisms. Finally, the pyrolysis of methylcyclohexane, an important component some jet fuels, was studied.

TABLE OF CONTENTS

ANOWLEDGEMENTS CHAPTER I. INTRODUCTION ………………………………………………………………1 II. EXPERIMENTAL APPARATUS ……………………………………………..9 III. PYROLYSIS OF ETHYL AND PROPYL IODIDES ………………………..22 IV. PYROLYSIS OF TERT-AMYL METHYL ETHER (TAME) ……………….64 V. PYROLYSIS OF 2-METHOXY TRIMETHYL BUTANE-d 6 (MTMB-d 6 ) …………………………………………………………………………….103 VI. PYROLYSIS OF 2-METHYL 1,3-BUTADIENE (ISOPRENE) AND ISOMERS …………………………………………………………………………….138 VII. PYROLYSIS OF CYCLOPENTADIENE & METHYLCYCLOPENTADIENE …………………………………………………………………………….162 VIII. PYROLYSIS OF METHYLCYCLOHEXANE …………………………193 IX. CONCLUSIONS …………………………………………………………..…214 X. FUTURE WORK ………………………………………………………….....216 XI. APPENDIX A …(alignment of the laser)………………………………..…218 XII. APPENDIX B …(pulse driver upkeep)……………………………..……223

ix

LIST OF TABLES

TABLE 4.1 Theoretical and experimental energy barriers for molecular eliminations and simple bond energy thresholds for TAME using geometries optimized at B3LYP/6- 31+G(2df,p). …………………………………………………………………………....84 TABLE 5.1 Solubilities of tert-alkyl methyl ethers in water at 20ºC. ……………..….110 TABLE 5.2 DFT electronic energies (a.u.), unscaled zero point energies (kJ mol -1 ), G3X energies (a.u.), and CCSD energies (a.u.) of HME, MTMB, MTBE, and selected homolysis fragments using geometries optimized at B3LYP/6-31G(2df,p). ..............122

x

LIST OF FIGURES

FIGURE 2.1 Schematic of the photoionization time-of-flight mass spectrometer. ……10

FIGURE 3.1 (a) Stack plot of mass spectra for pyrolysis of CH 3 CH 2 I (2%) in argon with internal nozzle temperatures from room temperature (295K) to 840K. ………………..29 FIGURE 3.1 (b) Stack plot of mass spectra for pyrolysis of CH 3 CH 2 I (12%) in helium with internal nozzle temperatures from room temperature (295K) to 860K. …………. 30 FIGURE 3.2 (a) Stack plot of mass spectra for pyrolysis of CH 3 CH 2 I (2%) in argon with internal nozzle temperatures from 865 K to 1115 K. …………………………………..32 FIGURE 3.2 (b) Stack plot of mass spectra for prolysis of CH 3 CH 2 I (12%) in helium with internal nozzle temperatures from 930 K to 1210 K. …………………………………..34 FIGURE 3.3 (a) Mass spectra for the pyrolysis of CH 3 CD 2 I (4%) in helium with heater temperatures of 300, 710, 815, 910, and 990 K. ……………………………………….36 FIGURE 3.3 (b) Mass spectra for the pyrolysis of CD 3 CH 2 I (2%) in argon with heater temperatures of 295, 700, 850, 1000, and 1200 K. …………………………………….37 FIGURE 3.3 (c) Mass spectra for the pyrolysis of CD 3 CD 2 I with heater temperatures from 295 K to 990 K. …………………………………………………………………..39 FIGURE 3.4 (a) Mass spectra for the pyrolysis of n-propyl iodide (1% in Ar) with heater temperatures from 295 K to 1065 K. …………………………………………………..41 FIGURE 3.4 (b) Mass spectra for pyrolysis of the isotopomer CD 3 CD 2 CH 2 I. ……….43 FIGURE 3.5 (a) Mass spectra for the pyrolysis of iso-propyl iodide with heater temperatures from 295 K to 1065 K. …………………………………………………..44 Figure 3.5 (b) Mass spectra for the pyrolysis of isotopomer CD 3 CHICD 3 with heater temperatures from 295 K to 1075. ……………………………………………………..46

xi

Figure 3.6 (a) C-I bond fission branching fractions for ethyl iodide. …………………..55 Figure 3.6 (b) C-I bond fission branching fractions for n-propyl iodide and for iso-propyl iodide. …………………………………………………………………………………..56 Figure 4.1 (a) Stack plot of mass spectra for pyrolysis of TAME in argon with internal nozzle temperatures from room temperature (~295 K) to 1000 K. …………………….70 Figure 4.1 (b) Stack plot of mass spectra for pyrolysis of TAME in sulfur hexafluoride with internal nozzle temperatures from 575 K to 1050 K. ……………………………..73 Figure 4.2 Stack plot of mass spectra for pyrolysis of TAME in argon with internal nozzle temperatures from 1100 K to 1240 K. …………………………………………..76 Figure 4.3 (a)Stack plot of mass spectra for pyrolysis of 2-methyl-1-butene in argon with internal nozzle temperatures from 1090 K to 1240 K. ………………………………….80 Figure 4.3 (b) Stack plot of mass spectra for prolysis of 2-methyl-2-butene in argon with internal nozzle temperatures from 1110 K to 1245 K. ………………………………….81 Figure 4.4 Stack plot of mass spectra for the pyrolysis of 2-butanone with unheated nozzle (RT) and over the temperatures from 890 – 1215 K in helium carrier gas. …….90 Figure 4.5 Stack plot of mass spectra for the pyrolysis of 2-butanone with unheated nozzle (RT) and over the temperatures from 950 – 1225 K in argon carrier gas. ………91 Figure 4.6 Stack plot of mass spectra for the pyrolysis of acetone-d 6 in helium carrier gas with nozzle temperatures up to 1275 K. ………………………………………………. 95 Figure 4.7 Stack plot of mass spectra for the pyrolysis of acetone-d 6 in argon in argon with nozzle temperatures up to 1250 K. ………………………………………………. 97 Figure 5.1 Stack plot of pyrolysis/supersonic jet expansion/118.2 nm photoionization TOF mass spectra of MTMB-d 6 seeded in argon (with a small amount of 2,3,3-trimethyl- 2-butene as internal standard) as a function of nozzle temperature. …………………..113 Figure 5.2 Stack plot of pyrolysis/supersonic jet expansion/118.2 nm photoionization TOF mass spectra of MTMB-d 6 seeded in helium (with a small amount of MTBE as internal standard) as a function of nozzle temperature. ……………………………….114 Figure 5.3 Stack plots of pyrolysis/supersonic jet expansion/118.2 nm photoionization TOF mass spectra of MTBE-d 3 seeded in helium with inset showing low mass fragments at 950K, intensity x40). ………………………………………………………………..117 Figure 5.3 Stack plots of pyrolysis/supersonic jet expansion/118.2 nm photoionization TOF mass spectra of MTBE-d 3 seeded in argon as functions of nozzle temperature. ..118

xii

Figure 5.4 Stackplot of mass spectra for the pyrolysis of isobutylene (3%) in argon. ...128 Figure 5.5 Stack plot of mass spectra for the pyrolysis of t-butyl cellosolve with unheated nozzle (RT) and over the temperatures from 850 – 1060 K. …………………………..130 Figure 5.6 Stack plot of mass spectra for the pyrolysis of t-butyl cellosolve with nozzle temperatures from 1130 – 1300 K. …………………………………………………….132 Figure 6.1 Relative energies of the species involved in the pyrolysis of isoprene. ……142 Figure 6.2 Energetics of the isomerization and dissociation pathways of isoprene. …..143 Figure 6.3 (a) Stack plot of mass spectra for pyrolysis of isoprene (15% in Ar) with internal nozzle temperatures from room temperature (295 K) to 1140 K. …………….144 Figure 6.3 (b) Stack plot of mass spectra for pyrolysis of isoprene (15% in Ar) with internal nozzle temperatures from 1215 K to 1400. …………………………………...146 Figure 6.3 (c) Stack plot of mass spectra for pyrolysis of isoprene (1.5% in Ar) with internal nozzle temperatures from 1200 K to 1390 K. …………………………………147 Figure 6.4 Stack plot of mass spectra from pyrolysis of allene and isobutylene. ……..156 Figure 7.1 (a) Stack plot of mass spectra for the pyrolysis of 1,4-cyclohexadiene diluted in argon. ………………………………………………………………………………..166

Figure 7.1 (b) Intensities of select m/e signals relative to the molecular ion. …………167 Figure 7.2 (a) Stack plot of mass spectra for the pyrolysis of cyclopentene diluted in argon. …………………………………………………………………………………..168

Figure 7.2 (b) Intensities of select m/e signals relative to the molecular ion. …………169

Figure 7.3 (a) Stack plot of mass spectra for the pyrolysis of cyclohexene with nozzle temperatures up to 1060 K. …………………………………………………………….170

Figure 7.3 (b) Stack plot of mass spectra for pyrolysis of cyclohexene with nozzle temperatures ranging from 1100 – 1420 K. ……………………………………………171

Figure 7.4 Stack plot of mass spectra for the pyrolysis of cyclopentadiene up to 1350 K. ……………………………………………177

Figure 7.5 Stack plot of mass spectra for the pyrolysis of methyl-cyclopentadiene. ….173

Figure 7.6 (a) Relative intensities of select ions for the cyclopentadiene pyrolysis. …..175

Figure 7.6 (b) Select relative intensities for the pyrolysis of methylcyclopentadiene. ...179

xiii

Figure 7.6 (c) Select relative intensities for the pyrolysis of methylcyclopentadiene. ...180

Figure 7.6 (d) Relative intensities of select ions in the pyrolysis of methycyclopentadiene. …………………………………………..181

Figure 7.7 Comparison of the aromatic growth region of the mass spectra for the pyrolysis of (a) cyclopentadiene and (b) methyl-cyclopentadiene. ……………………184

Figure 7.8 Stack plot of mass spectra for the pyrolysis of propargyl bromide. ………..185 Figure 8.1 (a) Room temperature trace for MCH. ……..………………………………198 Figure 8.1 (b) Pyrolysis of MCH spectral trace at 1090 K. ……………………………199 Figure 8.1 (c) Pyrolysis of MCH spectral trace at 1155 K. ……………………………200 Figure 8.1 (d) Pyrolysis of MCH spectral trace at 1220 K. ……………………………201 Figure 8.2 (a) Pyrolysis of MCH spectral trace at 1290 K. ……………………………203 Figure 8.2 (b) Pyrolysis of MCH spectral trace at 1355 K. ……………………………204 Figure 8.2 (c) Pyrolysis of MCH spectral trace at 1425 K. ……………………………205 Figure 8.2 (d) Pyrolysis of MCH spectral trace at 1450 K. ……………………………206 Figure 8.3 (a) Stackplot of mass spectra for the pyrolysis of methylyclohexene. ….…207 Figure 8.3 (b) Stack plot of mass spectra for the pyrolysis of methylcyclohexene. …...209 Figure A.1 Rough alignment with He/Ne laser. …………………………………….…218 Figure A.2 Alignment of the 355 nm light. ……………………………………….…...220 Figure A.3 Depiction of the timing elements used in the experiments. ……….……….221 Figure B.1 Diagram of the pulse driver assembly. ……………………………….…....225 Figure B.2 Diagram of the nozzle heater resistances. ………………………………….226

xiv

LIST OF SCHEMES

Scheme 3.1 Initiation events in the pyrolysis of alkyl halides. …...................................23 Scheme 4.1 The vicinal molecular elimination of methanol from TAME leads to 2- methy-1-butene (2m1b) or 2-methyl-2-butene (2m2b). …………………………………67 Scheme 4.2 Photoionization fragments in the ionization of TAME. …………………………………………...…………………………………..…………...71 Scheme 4.3 The disrotary 1,4 elimination of molecular hydrogen from 2-methyl-2-butene to produce 2-methyl-1,3-butadiene (isoprene). ………………………………………….78 Scheme 4.4 Bond fission and radical decomposition in the pyrolysis of TAME. ………83 Scheme 5.1 Molecular elimination and bond homolysis pathways expected in the pyrolysis of MTBE-d 6 . ……………………………………………………………..…..106 Scheme 5.2 Molecular elimination and bond homolysis pathways expected in the pyrolysis of MTMB-d 6 . ………………………………………………………………...111

1

CHAPTER 1 INTRODUCTION Radicals play an important role in variety of important processes such as pyroylsis/combustion, atmospheric chemistry, polymerization, and biochemistry. Unfortunately, they are typically highly reactive making their study difficult to achieve. Traditional experimental methods have made great advances in uncovering knowledge about such processes. Nevertheless, these methods are largely based on detection of stable end products and interpretation of intermediate steps of the mechanism with complex kinetics modeling. 1-3 Typical experimental methods include flow reactors and shock tubes, coupled with chromatographic and/or mass spectrometric detectors. Direct observation of intermediates using time-resolved spectroscopic methods has been the focus of more recent investigations. 4 Interpretation of the results from these methods is difficult due to the interference of other species in the spectral range of interest and the complexity of the reaction mechanisms. Computational methods have greatly aided the interpretation of these processes. Multi-reaction models can calculate concentration profiles if the rates of each reaction are known or can be estimated. Sensitivities of the concentration profiles to each reaction can be calculated and in this manner the most important reactions involved in the overall model can be elucidated. Accurate molecular geometries, heats of formation, and activation barriers can also be obtained from theory assisting identification of the lowest energy pathways available (thermal decompositions generally follow the lowest energy pathways). Ideally, direct and unambiguous detection

2

of all reaction intermediates and products and their concentration distributions in time would verify mechanistic details of the chemical reaction.

Characterization of the early stages of complex chemical reactions, particularly of the elusive “free” radical intermediates, can provide important information for verification of current models and discovery of previously unconsidered reaction mechanisms, thus providing a deeper understanding of the chemistry. In this work, the limitations of traditional methods are overcome with the employment of a high- temperature flash pyrolysis micro-reactor coupled with supersonic cooling and vacuum ultraviolet (VUV) photoionization mass spectrometry (TOFMS) 5,6 which enables the direct identification of the initial reaction intermediates, including the radical species. In order to understand the initial steps of a reaction it must be halted in a short amount of time with unequivocal detection of reactive intermediates and products. The approach described herein quenches the reaction by utilizing a fast-flow microreactor with a 20- 100 s residence time, followed by a “freezing” out of the reactive intermediates and products in a molecular beam. This results in the effectual isolation of the initial intermediates in a supersonic jet without any subsequent reaction and effects the decoupling of the key intermediates in the early time window from the rest of the reaction.

The technique was originally developed by Chen and co-workers 5 and has been demonstrated as a successful experimental approach. Briefly, the chemical moiety of

3

interest is “seeded” in an inert carrier gas which travels through a hot SiC tube where the thermal decomposition occurs. The heated mixture is then cooled upon expansion into a vacuum chamber from which a molecular beam is extracted where the intermediates, products and unreacted parent molecules can then be detected. With sufficient dilution in inert gas and short contact times, surface reactions can be minimized. A simple model indicates that under typical conditions <10% of the molecules (which are primarily carrier gas) in the heater would suffer surface collisions while the gas-phase collisions (~10 4 -10 6 in the residence time) can efficiently transfer thermal energy inducing unimolecular dissociation of the precursors and limited bimolecular reactions, if the residence time is sufficiently long. Further studies at higher sample concentrations can be carried out to examine the bimolecular reactions.

The detection scheme in this work employs the usage of a time-of-flight mass spectrometer (TOFMS) that affords the determination of the molecular mass of each species produced by ionizing a chemical species and determining its mass/charge ratio. Most often mass spectrometers use a high-energy ionization source to maximize efficiency with broad applicability. The large amounts of internal (vibrational and electronic) energy installed in many resulting ions result in a significant amount of fragmentaion of the parent ion, thus complicating the spectra (while creating a “fingerprint”) and often the molecular ion in not the base peak or is not obvserved at all. In this work a vacuum ultra-violet (VUV) photoionization (10.5 eV) ionization source

4

imparts significantly less internal energy to the resulting parent ion (which is often the base peak or only significant peak observed) and allows the identification of free radicals produced upon thermolysis. This photon energy is sufficient to detect most of the polyatomic free radicals and all C3 and larger stable organic species with the exception of propane and butane. 7 This approach is powerful for the direct observation of reactive intermediates. This adventitious technique is herein employed to investigate the mechanisms for a variety of hydrocarbon pyrolyses.

Many chemical reactions are known to proceed to a significant extent by both radical and molecular processes. Such is the case in the thermal decomposition of alkyl iodides and alkyl methyl ethers. It is known that the nature of the alkyl group in these compounds can have tremendous influence on the degree of competition (branching ratios) for these processes. The true extent of one pathway versus another often remains unclear due the transient radicals. In the first portion of this dissertation the decomposition of ethyl and the propyl iodides are investigated to characterize the affect on branching ratios as a consequence of possessing a primary versus secondary alkyl group. Next, the pyrolysis of tert-amyl methyl ethers (TAME) and 2-methoxy trimethylbutane (MTMB) is explored to produce additional information on electronic versus field effects in thermolysis of tertiary methyl ethers. These experiments contribute valuable insight to the fundamental knowledge of how these reactions proceed as a function of the chemical structure and establish the reliability and utility of this technique.

5

The remainder of this work studies the production of aromatic compounds (particularly formation of the “first ring”), which are precursors of soot, from the pyrolysis of small fuel-like hydrocarbons, specifically isoprene (2-methyl-1,3,-butadiene), cyclohexene, 1,4-cyclohexadiene, cyclopentene, cyclopentadiene, and methyl- cyclopentadiene. Combustion has been a great ally for humans since we started kindling fires providing for heat in the cold, light in the dark, cooking of food, and more recently electricity. For a fuel to be desirable it must be high intensity, transportable, and controllable. Without question, the fossil fuels (hydrocarbons) best satisfy these descriptors. Today, with the incredible energy demands of the modern lifestyle, fossil fuels are the dominant source of energy to fill our need. In 2004 hydrocarbon combustion accounted for >80% of the total global energy budget and combustion will undoubtedly continue to be an important source of energy in the future. Unfortunately, the combustion of fossil fuels (hydrocarbons) produces pollution such as the greenhouse gas carbon dioxide, nitrogen and sulfur oxides (leading to acid rain, smog, and ozone), VOCs (volatile organic carbon compounds), and soot. The production of soot (and particulate matter) is of particular concern for several reasons. Although soot can be comprised of a complicated gallimaufry of compounds, at its core it is made up of polyaromatic hydrocarbons (PAHs) which can be carcinogenic (e.g. benzo (a) pyrene). The role of high exposure to soot as a cause of cancer in adolescent chimney sweeps was first noted by Percival Potts in London in 1775. 8 Recently, studies in the US 9-11 and UK 12

have found a range of carcinogenic PAH originating primarily from vehicle emissions.

6

Additionally, soot can strongly absorb solar radiation. 13 It was proposed in 2000 that the fastest way to fight global warming is the reduction of black carbon, methane, and other warming pollutants which can be more easily controlled than CO 2 . 14 A recent assessment of the contribution to global warming from black carbon estimates the forcing to be 0.9 watts per square meter, which is a larger contribution than methane and about 55% of that from CO 2 . 15 Jacobson contends that up to 30% of global warming could be controlled if you could control soot. 16 Lastly, soot formation represents an inefficient use of the fuel.

Despite extensive study, the fundamental mechanisms and basic chemistry of hydrocarbon combustion, even in the early stages of pyrolysis and oxidations, needs further improvement. 1-4,17-19 In this work the direct observation of initial reactive intermediates is accomplished by the powerful experimental approach of coupling flash pyrolysis of the compound of interest with subsequent cooling and isolation in a supersonic beam and analysis by VUV-MS, which can detect radicals and limits fragmentation.

REFERENCES [1] W. C. Gardiner Jr. (Ed.), Combustion Chemistry, Springer-Verlag, New York, 1984. [2] W.C. Gardiner Jr. (Ed.), Gas-phase Combustion Chemistry, Springer-Verlag, New York, 2000.

7

[3] J. Warnatz, U. Maas, R. W. Dibble, Combustion, Springer-Verlag, Berlin, 1996. [4] J. A. Miller, G. A. Fisk, Chem. Eng. News 65 (1987) 22. [5] D. W. Kohn, H. Clauberg, P. Chen, Rev. Sci. Instrum. 63 (1992) 4003. [6] J. Boyle, L. Pfefferle, J. Lobue, S. Colson, Combust. Sci. and Tech. 70 (1990) 187. [7] S. G. Lias, J. E. Bartmess, J. F. Liebman, J. L. Holmes, R. D. Levin, W. G. Mallard, Gas-Phase Ion and Neutral Thermochemistry, American Chemical Society, New York, 1988. [8] R. F. Sawyer, Eighteenth Symposium (International) on Combustion (1981) 1. [9] B. A. Benner Jr., G. E. Gordon, S. S. Wise, Environ. Sci. Technol. 23 (1989) 1269. [10] L. M. Hildemann, G. R. Markowski, G. R. Cass, Environ. Sci. Technol. 35 (1991) 744. [11] W. F. Rogge, L. M. Hildemann, M. A. Mazurek, G. R. Cass, B. R. T. Simoneit, Environ. Sci. Technol. 27 (1993) 636. [12] R. M. Harrison, D. J. T. Smith, L. Luhana, Environ. Sci. Technol. 30 (1996) 825. [13] J. Quaas, Science 32 (2009) 153. [14] J. Hanson, M. Sato, R. Ruedy, A. Lacis, V. Oinas, Proc. Natl. Acad. Sci. USA 97 (2000) 9875. [15] V. Ramanathan, G. Carmichael, Nature Geosci. 1 (2008) 221.

8

[16] M. Z. Jacobson, Science 409 (2001) 695. [17] R. W. Walker, 22nd Int. Symposium on Combustion (Combustion Institute, Pittsburgh) (1989) 883. [18] R. W. Walker, Sci. Prog. 74 (1990) 163. [19] H. Brockhorn (Ed.), Soot Formation in Combustion, Springer-Verlag, New York, 1994.

9

CHAPTER 2 EXPERIMENTAL Thermal decomposition experiments were accomplished in an apparatus that is schematically depicted in Figure 2.1. It makes use of a Wiley-MacLaren type linear time-of-flight mass spectrometer (TOFMS, model D-651, R. M. Jordan Company 1 to monitor pyrolysis products by means of 118.2 nm (10.48 eV) photoionization. A typical resolution (m/m) obtained with this instrument has been assessed to be = 200 at m/z 150. 2 The main reactor chamber is pumped by a cryobaffled Varian VHS-6 diffusion pump, and the mass spectrometer is differentially pumped by means of a turbomolecular pump. Pyrolyses were carried out by expanding the parent molecules, seeded in inert carrier gas (typically helium or argon), via a heated silicon carbide tube (1 mm i.d., 2 mm o.d., Carborundum Corp) through a pulsed valve into the photoionization region of the mass spectrometer, similar to the apparatus described by Chen and coworkers. 3 The silicon carbide tubing was attached to a machinable piece of alumina by use of a high temperature ceramic adhesive (Cotronics Corporation) that had a 2 mm i.d. channel to allow the gas flow from the pulsed valve to the silicon carbide tube and was mounted to the faceplate of a General Valve series 9 pulsed valve operating at 10 Hz. The alumina isolates the silicon carbide microreactor both thermally and electronically from the pulsed valve and was sandwiched between the faceplate and a MACOR disk to provide additional stability. Two graphite electrodes (Poco Graphite) separated by ~1 cm were press-fitted onto the silicon carbide tubing and were heated resistively. The electrical

10

current was controlled by a Variac transformer and light bulbs hooked in parallel served as current limiters.

11

only a best approximation, and nominal temperatures reported are believed to be accurate to within 50 K. With a near sonic velocity of the sample within the nozzle, the residence time in the heater has been estimated to be approximately 20s when helium is utilized as the carrier gas. 3

The possibility of heterogeneous catalysis occurring on the SiC surface is a complication with flash pyrolysis nozzles of this type. The issue was addressed by Peter Chen with the pyrolysis of t-butyl nitrite, 4 (CH 3 ) 3 CONO, which under homogeneous pyrolysis conditions decomposes to t-butoxy radical and NO by cleavage of an O-N bond. Heterogeneous decompositions can cleave the C-O bond producing t-butyl radical and NO 2 , which were not observed. It was therefore concluded that the flash pyrolysis in the system proceeds primarily as a homogeneous process. Nevertheless, the contribution of heterogeneous processes cannot be neglected. For example, it was discovered that kinetic parameters are not reproduced accurately with this method. 2 Pyrolysis reactions with well-defined activation energies were investigated with this experimental apparatus and Arrhenius plots constructed in order to compare the experimental activation energies to literature values. The result of this investigation revealed that the experimentally obtained activation energies were ~60-70% of literature values. This result is a clear indication that heterogeneous catalysis on the SiC surface is attenuating the activation energies. Due to these considerations, the flash pyrolysis technique described in this body of work is best relied on for qualitative mechanistic insight.

12

Short residence times in the microreactor are achieved by maintainging a near sonic flow velocity throught the SiC tube. This occurs by supersonic injection of the gas pulse into the tube. This expansion occurs upon increase in the cross-sectional area between the pulsed valve faceplate orifice (0.75 mm) and the SiC tube (1.0 mm I. D.), similar to the nozzle and test section of a supersonic wind tunnel. 5 Supersonic flow cannot be maintained throughtout the SiC tube, but a series of expansions and compressions within the tube maintains an average flow velocity which is approximately sonic. With a high enough initial stagnation pressure, the pressure at the end of the tube is sufficient to expand into the vacuum chamber as a supersonic jet. The characteristics of this expansion resemble those occurring from a capillary tube, with a large length to diameter ratio, which has been found to be similar to a sharp edged orifice. 6 Although no cross-sectional change occurs at the tube exit, wall friction or heat transfer can achieve flow choking in a capillary nozzle, and sonic velocity is reached just before the tube exit.

Chen and coworkers 3 tested the difference in pulse width and velocity exiting both the bare pulsed valve faceplate and with a pyrolysis nozzle attached and found little or no difference between the two above 1.0 atmosphere stagnation pressure. From these results, they conclude that the flow velocity within the tube is near sonic, and the contact time within the tube can be easily estimated. The residence time within the pyrolysis region of our source was estimated to be ~ 20s with helium as the carrier gas.

13

After exiting from the heated zone the unreacted precursor and its products are cooled by supersonic expansion. The degree of cooling increases with the mass of the carrier gas, and the rotational temperature of molecules in the jet after expansion has been ascertained to be ≤ 50 K by Chen and coworkers 7,8 by means of multiphoton ionization (the vibrational temperature tends to be higher). The present studies primarily utilize argon as the carrier gas providing the best balance of cooling and signal intensity/quality.

The use of a supersonic molecular beam coupled to flash pyrolysis in the apparatus used in these experiments allows for the isolation of the molecules under study. A molecular beam is extracted from an underexpanded, supersonic expansion from a high-pressure reservoir to a low-pressure region. The same type of expansion occurs from converging-diverging supersonic rocket nozzles. It is now possible to extract molecular beams from the core of expanding gas flow where the internal energy content of the molecules have been reduced to extremely low values without massive condensation. The molecular populations are then concentrated into one or a few internal states preparing coherent states excellent for characterization.

The unique properties of molecular beams are due to the supersonic nature of the gas resulting in an underexpansion described below. The gas starts from a negligible

14

small velocity, referred to as the stagnation state (P 0 ,T 0 ). The pressure gradient from the initial pressure (P 0 ) to the background pressure (P b ) causes acceleration of the gas flow towards the source exit. This flow can be approximated as isentropic, neglecting viscosity and heat conduction effects. The gas can reach the speed of sound at the source exit when the ratio of the reservoir pressure to background pressure, P 0 /P b , becomes greater than G ≡ [(γ+1)/2]^(γ/(γ-1)) where γ is the ratio of heat capacities at constant pressure versus constant volume, C p /C v . Alternatively, if the pressure ratio is less than this critical value, the flow will exit subsonically, with an exit pressure nearly equal to P b , without any further expansion. As the ratio P 0 /P b increases to be greater than G, the exit pressure becomes independent of the background pressure, P b , and is therefore considered to be “underexpanded”.

There are two characteristic features of supersonic flow. First, in a supersonic expansion beyond the source exit as the flow area increases so does the velocity of the flow, in contrast to subsonic flow. Second, supersonic flow cannot “sense” downstream conditions due to the supersonic nature. Essentially, particles traveling in the same direction with the same velocities do not interact with each other. Thus, the flow does not know the boundary conditions, yet it must adjust. Eventually, collisions with the background gas in the expansion chamber will reduce the molecular beam flow to subsonic velocity where in the flow can adjust to boundary conditions. This results in

Full document contains 241 pages
Abstract: Flash pyrolysis coupled to molecular beam extraction and single photon ionization time-of-flight mass spectrometry along with quantum chemical calculations are employed to study the pyrolysis of organic molecules with relevance to combustion processes. Branching ratios for the molecular elimination and bond fission pathways was achieved for ethyl and propyl iodides providing information about the nature of the mechanism with relation to chemical structure. Similarly, the decompositions of a series of alkyl methyl ethers, whose anti-knock ability in commercial fuels is attributed to the molecular elimination pathway, was investigated to examine the competition of primary and subsequent homolyses at higher temperatures, which tend to promote "knock". The isomerization/decomposition of isoprene was looked at in detail with special attention to the formation of soot precursor, most notably the "first ring" benzene. Cyclopentadiene and methylcyclopentadiene are known to be highly sooting fuels and were pyrolyzed to study the initial steps in aromatization. The pyrolysis of cyclohexene, cyclopentene, and 1,4-cyclohexadiene were conducted to facilitate interpretation of aromatic formation and to compare experimental results with well established mechanisms. Finally, the pyrolysis of methylcyclohexane, an important component some jet fuels, was studied.