University of Southern California Dissertations and Theses

Competitive and sequential absorption and dissociation mechanisms in the infrared fragmentation of polyatomics

(USC Thesis Other)

Competitive and sequential absorption and dissociation mechanisms in the infrared fragmentation of polyatomics

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content COMPETITIVE AND SEQUENTIAL ABSORPTION AND DISSOCIATION MECHANISMS IN THE INFRARED FRAGMENTATION OF POLYATOMICS by Julio Faustino Caballero A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Chemistry) May 1983 UMI Number: DP21892 All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. Dissertation Publishing UMI DP21892 Published by ProQuest LLC (2014). Copyright in the Dissertation held by the Author. Microform Edition © ProQuest LLC. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106 -1 3 4 6 UNIVERSITY OF SOUTHERN CALIFORNIA T H E G R A D U A TE S C HO O L U N IV E R S IT Y PARK LOS A N G ELES. C A L IF O R N IA 9 0 0 0 7 Ph.D . C - >8i This dissertation, written by Julio Faustino Caballero under the direction of A . . 1 . ? . . . . Dissertation Com mittee, and approved by all its members, has been presented to and accepted by The Graduate School, in partial fulfillm ent of requirements of the degree of D O C T O R O F P H I L O S O P H Y Si w Dean ........ DISSERTATION COMMITTEE Chairm an .1 lU J ^ To my family i i ACKNOWLEDGMENTS I want to show in these pages my deep appreciation to a number of persons who have encouraged and helped me in my training as a professional scientist. I am indebted to Professors Rafael Calvo, Luis Masperi and Jose Abriata who showed me the human side of science in my early years as a graduate student and to my classmates, now Drs. Juan Carlos Bolcich, Horacio Caceres and Fernando Grinstein for their companionship during the Graduate School years at "Jose A. Balseiro," Physics Institute in San Carlos de Bariloche, Argentina. I appreciate the interaction with Mr. Osvaldo Vilar and Dr. Carlos Rosito which showed me in my early years of laboratory training at CITEFA (Argentina) criteria to face and solve technical problems. At the same time the intellectual and practical contact with my labmate Miss Veronica Slezak, showed me that with patience and cooperation it is possible to succeed despite a negative laboratory atmosphere. i i i Finally, for my last period of training at USC I am indebted to Dr. Curt Wittig that through his own communication method has showed me the professional, sometimes dark, side of science. I had the opportunity to share the experience of a high level academic and technical group through constant discussions and talks with past and present members. I am mainly grateful to my labmate and only American friend Dr. Thomas A. Fischer that taught me that only with an open mind can one understand other people's attitudes, admit one's errors and change them for the group's benefits. Innumerable academic and technical discussions with Drs. Hanna Reisler, Robert Quick, Jr., Metin Mangir, Martin Levy, Anita Renlund, Yehuda Haas, Michael Stuke, Joe Tiee, Michael Addison and Fanao Kong were invaluable to this thesis work. Thanks to J. Steinfeld and particularly to J. Bott for the loan of special equipment. To the Group's Graduate students that share with me the nice feelings of mid term exams and finals. To F. Shokoohi that helped me in desperation times, to Tom Watson for his work in overcoming USC bureaucracy, language barriers and grammatical shortcomings and G. Radhakrishnan for her thesis proofreading. To Henry Helvajian, M.H. Yu and iv again to T.A. Fischer that as senior graduate students gave me some advice upon my arrival on the American way. of life. To Jim Emerson for his repair of glass systems, to Herb Lloyd for his machining and understanding of my lack of conversion factors between inches and millimeters, to Ray Schmidt and Doyle Howland for their photographic and drawing services, to Kristine Roche and Trisha Woitena for various manuscript typing, to Rosalinda Senaha and Rebecca Gronvold for thesis typing. Special thanks to Dr. David Dows for his help and ability to handle some of my paperwork for admission and during my stay at USC, Finally, to my parents without whom it would have been impossible to finish my Graduate Studies in Argentina. Financial support for this research was provided by the United States Department of Energy and the National Science Foundation. This thesis was completed during a leave of absence from CITEFA-CEILAP (Argentina) from January 1979 to March 1982 and partially supported through an Organization of American States Doctoral Fellowship from February 1980 to January 1982. ________________________________________________________________________ v TABLE OF CONTENTS DEDICATION Page ii ACKNOWLEDGMENTS iii LIST OF FIGURES ix LIST OF TABLES xiii ABSTRACT xiv PREFACE xvi I. INTRODUCTION 1 General Experimental Technique 9 References 15 II. PHOTODISSOCIATION OF SUBSTITUTED METHANE, ETHYLENE AND ETHANE: VIBRATIONAL EMISSIONS FROM PARENT AND PRODUCT SPECIES 18 Introduction 18 Experimental 21 Results 21 Discrimination of Infrared Signals 29 c2H3F 36 C2h5F 40 i,ic2h4f2 46 vi Discussion 59 Conclusions 63 References 64 III. IR MULTIPLE PHOTON DISSOCIATION OF C2HC13: MOLECULAR ELIMINATION vs. BOND FISSION, AND EFFICIENT DISSOCIATION OF THE C2C12 PRODUCT 68 Introduction 68 Experimental 73 Results 74 Nascent HC1 74 The production of Cl 81 Fluence dependences 85 Discussion 86 Dissociation of C2C1 93 Conclusions 95 References 97 IV. PRODUCT VIBRATIONAL ENERGY CONTENT IN IRMPD: THREE AND FOUR CENTER ELIMINATION OF DF AND HF FROM CHjCDF 104 Introduction 104 Experimental 106 Results 107 Discussion 111 vii Conclusions 128 References 129 Selected Bibliography 134 v i i i LIST OF FIGURES Figure Page 1-1 Schematic diagram of a molecule which can decompose by two pathways 8 1-2 Schematic drawing of the experimental arrangement 12 Il-la Infrared absorption spectra of 100 Torr each of C2H3F and C2H2 in the 2000-4000 cm”^ region. 24 Il-lb Spectrophotometer traces showing absorption spectra of C2H5F along with C2H4 26 II-lc Spectrophotometer traces of 1 ,1C2H( jF2 along with c2H3F 28 II-2 Typical fluorescence signals obtained with 55 mTorr of 1 ,1C2H^F2 and 5.7 J cm“^ 33 11-3 Peak fluorescence signals vs. pressure of the studied species 35 II-4 Peak fluorescence signals vs. fluence for C2H3F 38 i x II-5a Various emission signals vs. fluence for C2H5F 42 II-5b Dependence on laser fluence of the HF^ peak fluorescence signal from excitation of C2H5F/He mixtures 45 II-6a Dependence on laser fluence of the peak fluorescence signals from different spectral regions corresponding to HF^f 1 ,1C2H4F2 C-H stretches and C2H2 C-H stretches 49 II-6b,c Dependence on laser fluence of the peak fluorescence signals from excitation of 1 ,1C2H4F2/He mixtures. (b) and (c) refer to HF^ and 1 ,1C2H4F2 C-H stretch respectively 52-53 II-6d Dependence on laser fluence — O (5-35 J cm ) fluorescence signal for different He buffer pressures on the dissociation of lflC2H4F2 56 x III-l Spectrophotometer traces showing absorption spectra of trichloroethylene in the regions 800-1100cm*"1 and 2700-3300 cm-1 76 III-2 Identification of the infrared emission at 2900 cm*”1 79 III-3 HCl^ fluorescence following the IRMPD of trichloroethylene 83 III-4 RRKM calculations for molecular elimination and fission from trichloroethylene 89 IV-1 Infrared absorption spectrum of gaseous VFd^ in a) 1100-850 cm""1 region and b) 3600-2200 cm” 1 region 109 IV-2 Typical fluorescence signal in the region 2700-3050 cm” 1 obtained with 53 mTorr of CH2CDF and 30 J cm” 2 113 IV-3 Relevant ground electronic singlet and transition states expected in direct three and four center HF elimination from VF 116 xi IV-4 IV-5 Calculated RRKM rate constant for direct HF and DF elimination from VFdx k (three center)/k (four center) ratio varying the energy of activation of each process 120 122 x i i Table II-l III-l IV-1 LIST OF TABLES Page Summary of parent infrared emissions from different substituted ethane, ethylene and methane, together with its main molecular parameters 59 Parameters used in RRKM calculations 91 Product energy content in VF decomposition 125 xm ABSTRACT The unimolecular reactions of substituted ethanes and ethenes have been studied, using collision free excitation to prepare reactants with energies in excess of reaction threshold. IR multiple photon excitation (IRMPE) insures that reaction occurs from the ground electronic potential surface, so that the observed product excitations are nascent in the degrees of freedom which are pertinent to our measurements. By detecting spontaneous emission from the vibrationally excited hydrogen halide product, we are able to discern sequential and competitive reaction pathways, as well as 3 and 4-center concerted processes. Such experimental results are compared to RRKM calculations and are in sensible agreement with the thermochemistry and known properties of such systems. In the case of tricholorethylene, we are able to follow the decomposition steps all the way to the production of C2 (X1Sg and &3 t t u). The very efficient dissociation of the C2CI2 intermediate is a consequence of its formation via a concerted process, and this __________________________________________________________ xiy accounts for the efficient production of Cl. The long standing uncertainty concerning the dissociation of the C2CI intermediate is also explained in terms of the strong coupling that exists between the ground and low- lying £ A" states. This is a quite general result, and clears up many misconceptions which have appeared in the literature during the past decade. With vinyl fluoride d^r the 3 and 4-center pathways are scrutinized by monitoring DFt and HF^ respectively, and we are able to distinguish rather small energy differences in a very interesting and important part of the potential surface. Finally, at excitations below reaction threshold, we are able to monitor the accumulation of vibrational energy by parent molecules, in order to determine the extent of parent excitation, with some degree of quantitative certainty. We are also able to detect a uniquely rapid flow of energy for the heavier species and this phenomena was related to the state density of the parent molecule. xv PREFACE This thesis uses an experimental technique wherein the infrared spontaneous emission from product and parent molecules is monitored, in order to study the infrared multiple absorption and dissociation (IRMPA and IRMPD) of polyatomics. The results are germane to furthering our understanding of several of the physical and chemical processes which are associated with the excitation and reaction(s) of such parent species. Chapter I is an introduction to the experiments and stresses the utility of IRMPA as a means of energizing molecules, whose subsequent behavior can then be scrutinized in detail. Several basic questions are presented regarding this new phenomenon, along with the specific experimental techniques that are used to address these questions. Chapter II presents results that involve (mainly) the physical absorption in different chlorinated or fluorinated methane, ethane and ethylene. Chapter III shows that the technique is appropriate to studying dissociation mechanisms in the xv i IRMPD of trichloroethylene, and shows how sequential and competitive primary processes can be distinquished. Chapter IV deals with competitive primary pathways in the IRMPD of vinyl fluoride d^ (CH2CDF). References for the chapters are found at the end of each chapter. x v i i CHAPTER I INTRODUCTION Studies of the interaction of light with matter (photochemistry) date back to nearly a century, and consequently the development of photochemistry has been closely coupled to the evolution of different photon sources. Broad band light sources were gradually replaced by mercury, sodium and xenon lamps which were sources of specific emission lines, and which, when used in conjunction with filters and monochromators, made it possible to perform more sophisticated frequency- specific experiments. The results and interpretation of photochemical measurements has been treated in innumerable books and the study of photochemistry is now an important part of the training of new chemists.^ At the same time the development of new light sources for use in photochemical studies is still in a growing stage. A major breakthrough came in the early 60's with the advent of the laser. This has been the most dramatic 1 and far reaching of these developments, and continues even today with the discovery of new laser transitions. Why are laser photons so special? Of the many differences between lamp and laser sources the following points were chosen to give a brief explanation: 1. Due to the processes involved in the laser light generation it is highly monochromatic and coherent, while lamp radiation is not. 2. It is possible to obtain high intensity laser light at almost any wavelength while lamp sources are typically much less intense. 3. Laser radiation may be obtained in continuous or pulsed form and nowadays it is possible to get pulses as — 1 ^ short as 10 sec. 4. The high directionality of laser sources makes possible their handling in a very easy way over long path lengths without loosing power. This discovery had a very important effect on the photochemical and thermochemical research area due to the fact that it was possible to perform experiments which previously could hardly have been conceived. Besides the obvious goal of comparing laser chemistry with the conventional one from the point of view of product analysis in a chemical reaction (laser induced chemistry), ultraviolet (UV) and visible laser photons fulfilled the possibility of efficiently exciting electronic transitions directly (highly excited state chemistry), and infrared laser photons the possibility of exciting a particular vibrational mode. Although quite a few experiments were done in the 60's with continuous radiation, the next major development was that of pulsed laser sources, which provide much higher power than CW lasers and also made it possible to perform more accurate time resolved experiments. The following is a brief overview of the work done on the interaction of pulsed infrared laser radiation with polyatomic molecules in the gas phase. One of the first published works was that of Isenor and Richardson in 1971, followed a couple of years later by works of Isenor^ and Letokhov^, all of which suggested that following the absorption of multiple photons of infrared radiation from pulsed lasers, molecules could be dissociated under collisionless conditions. Ambartzumian® and Lyman^ then showed that this process was highly selective with respect to isotopes and that this property could be exploited for their separation. Their results gave a tremendous impetus to research in 3 an exciting new field which is now commonly referred to as Infrared Multiple Photon Absorption and Dissociation (IRMPA and IRMPD).8 Besides the previously mentioned application of isotope separation owing to the selectivity of the absorption of IR photons, a lot of excitement was generated with the idea of using IR photons to energize molecules. When compared to UV and visible photons, IR photons are relatively low in energy (~ 3 Kcal/mole) and therefore it seemed an interesting prospect to use them on the one hand to excite a molecule to vibrationally excited states and then study the reaction of such species without causing fragmentation and on the other hand to promote molecules to levels above dissociation with comparatively low excess energies. This technique has led one to gain a deeper insight into a number of physical and chemical problems relating to the nature of the absorption process, the effects of collisions on the dissociation event, and the number of photons that a molecule could absorb prior to dissociation. Other interesting phenomena which have been unravelled are the differences in final products of molecules when dissociated photochemically rather than thermally, and the effects of depositing energy in 4 specific vibration modes of the molecule on its dissociation (the Holy Grail of selective mode chemistry).®0 The method may provide a new way to understand potential diagrams that govern chemical reactions as well as information on the relationship between the amount of energy provided to the reactant and what is found in the products. Finally, it may be used to diagnose the primary processes involved in the dissociation mechanisms. Results obtained in various experiments can now be interpreted physically using a model of the Q absorption process suggested by Mukamel and Jortner. This model divides the multiphoton absorption process into three regions. At low excitation energies the absorption involves resonant transitions. However at higher energies, vibrational levels have an increasing anharmonicity due to which the transitions are no longer resonant and in order for the excitation process to continue the anharmonic effect has to be compensated. This is achieved in a second region at medium/high energies, called the "quasicontinum" where the absorption is completely assured due to the high density of vibrational states. The third region lies above the dissociation limit and in this region the rate of 5 absorption of photons competes with the dissociation rate. The model of Jortner et al. provides some physical basis for the absorption of multiple photons. However it does not offer an explanation for the chemical problem of being able to differentiate between competing dissociation channels that may be available to the parent molecule at a certain level of excitation.10 Fig. (1-1) shows in a very schematic form this problem taken from Ref. 11. When a proper chemical species is energized by powerful coherent IR light it is possible to achieve different levels of excitation and this could make more than one dissociation channel accessible. Due to differences in unimolecular rate constants, it is then possible to have a preference for one channel more than another (not necessarily the lowest energy one). One can obtain secondary photodissociation of photoproducts from each channel making it difficult to differentiate competitive from sequential processes. So the understanding of primary processes of decomposition is the key to the actual fragmentation mechanisms. The previously quoted experiments, ° as well as others,10 were in the beginning not very well understood 6 FIGURE (1-1): Schematic diagram of a molecule which can decompose by two pathways. Unimolecular rate constants for the low energy pathway increase slowly with energy above E^ while the high energy pathway increase rapidly over E2. Varying the laser intensity different reaction products can be made to predominate. 2(E) 1(E) E reaction threshold 2 reaction threshold 1 E 1 k(E) 8 due to those problems. New probing methods have been developed and are still being developed to overcome them. These advancements are the driving force for further expansion in this research field and conversely the new technology involved in the methods has often led to the discovery of new phenomena (the dialectic and materialistic way to understand the progress of science) This thesis will provide and explain some experimental results that give more insight into those two problems: the physics of the absorption process and competitive vs. sequential dissociation mechanisms. GENERAL EXPERIMENTAL TECHNIQUE To answer the above questions concerning IRMPA and IRMPD, it is necesary to have the proper diagnostic methods. The analysis techniques that were used in early experiments were quickly changed to time resolved ones, TO O e.g., molecular beams or laser induced fluorescence0 methods to avoid the long time that transpires between sample irradiation and product analysis. 9 This thesis has used the experimental set up and the technique arranged at our laboratory by Charles Robert Quick, Jr., during his thesis work14 and the present one is a natural continuation of it. The experimental technique consists of monitoring the IR spontaneous emission from parent molecules and/or photoproducts that occurs following IRMPA and IRMPD processes in polyatomics. Fig. (1-2) shows the experimental arrangement used in all the experiments. The output of a CO2 TEA laser (Lumonics 101, line tunable, multimode, 1.5 J maximum energy) is directed through a 2000 cc stainless steel chamber and IR spontaneous emission is observed at right angles to the laser beam by an LN2 cooled InSb IR detector (Photovoltaic Spectronics, 1.6 cm area, sensitive between 1 and 5pm, l-2psec rise time). Under typical experimental conditions the laser pulse consists of an initial spike of approximately 100 nsec (FWHM) followed by a tail of -0.8 psec with a 3/2 ratio of energy between them. Energies were measured using a Joule- meter (Gen Tec ED200) and the attenuation of the beam was achieved by either an absorbing gas (such as CF2Br2 or CF3I) or stacking plastic films. 10. FIGURE (1-2): Schematic drawing of the experimental arrangement. 11 SIGNAL PROCESSING AND RECORDING EQUIPMENT INTERFERENCE FILTERS, OPTICS GAS COLD FILTERS InSb SCREENED ENCLOSURE VACUUM SYSTEM Long side arms (13 cm) make it possible to focus the beam with a long focal length lens keeping a low energy density on the entrance NaCl window. Tight focusing can be achieved by mounting lenses inside the cell. Depending on the type of lens used in a particular experiment, the beam focus was estimated through circular aperture transmission measurements to be between 3 and 1 mm^. Interference narrow band pass filters (OCLI) and cold gas cells were used to identify the emitting species and the signals from the detector are amplified and photographed directly from an oscilloscope or digitized/processed by signal averaging techniques as necessary. Briefly, it provides a time resolved probe of the IRMPA and IRMPD processes, has sufficient sensitivity to allow experiments to be carried out at low pressures, makes possible the collection of data on a single shot basis and allows for facile detection of vibrationally excited diatomics (HF, HC1, etc.). Disadvantages include poor sensitivity and rise time (as compared with photomultipliers), inherently long lifetimes and consequently low photon fluxs characteristic of vibrational transitions, and occasional overlapping of IR emissions. 13 In previous works in the LKL,*^ vibrationally excited HF was detected and monitored, being the most suitable diatomic for the chosen detection system. In the present thesis we have extended the search for IR emission to the parent molecule and other diatomic dissociation products such as HC1, DC1, HBr, DBr and DF. Particular experimental details are going to be outlined in each chapter. REFERENCES 1. J.G. Calvert and J.N. Pitts, "Photochemistry," J. Wiley and Sons, N.Y., 1966. 2a. R.T. Hall and G.C. Pimentel, J. Chem. Phys., 38, 1889, (1963). b. Borde, C., Henry, Annie, and Henry, Lucien, C.R. Acad. Sc. Paris, Ser. B, 263, 619 (1966). c. O.R. Wood and S.E. Schwartz, App. Phys. Lett., 11, 88 (1967). d. Cohen, C., Borde, C., Henry, L., C.R. Acad. Sc. Paris, Ser. B, 265, 267 (1967). e. H. Brunet and F. Voignier, C.R. Acad. Sc., Paris, Ser. C, 266, 1206 (1968). f. Quel, E., and X. De Hemptimme, Xavier, Ann. Soc. Sci. Bruxelles, Ser. 1, 83, 262 (1969). g. V.V. Losev, V.F. Papulovskii, V.P. Tychinskii and T.A. Fedina, Khim. Vys. En., 3, 331 (1969). h. T.Y. Chang, C.H. Wang and P.K. Cheo, App. Phys. Lett., 15, 157 (1969) . 3. N.R. Isenor and M.C. Richardson, App. Phys. Lett., 18, 224 (1971). 15 4. N.R. Isenor, V. Merchant, R.S. Hallsworth and M.C. Richardson, Can. J. Phys., 51, 1281 (1973). 5. R.V. Ambartzumian, V.S. Letokhov, E.A. Ryabov and N.V. Chekalin, JETP Lett., 20, 273 (1974). 6 . R.V. Ambartzumian, Yu. A. Gorokhov, V.S. Letokhov and G.N. Makarov, JETP Lett., 21, 171 (1975). 7. J.L. Lyman, R.J. Jensen, J. Rink, C.P. Robinson and S.D. Rockwood, App. Phys. Lett., 27, 87 (1975). 8. This research field has become so important that several books have been devoted to it. Among them we can mention: a. A.B. Shaul, Y. Haas. K.L. Kompa and R.D. Levine, "Lasers and Chemical Change," Springer Verlag Series in Chemical Physics, V10, Springer-Verlag Ed., 1981. b. "Multiple Photon Excitation and Dissociation of Polyatomic Molecules," edited by C.D. Cantrell, Springer-Verlag, Heilderberg, Berlin, 1981. c. "Photoselective Chemistry," VI and 2, edited by J. Jortner, Wiley, N.Y., 1981. 9. S. Mukamel and J. Jortner, Chem. Phys. Lett., 40, 150 (1976). 10. For more detailed references of this problem see Chapters III and IV Introductions* 16 11. R. Duperrex and H. Van der Bergh, Chem. Phys., 40, 275 (1979). 12. Friedrich Engels, "Herr Eugen Duhring's Revolution in Science." 13. I.W. Smith, "Kinetics and Dynamics of Elementary Gas Reactions," Butterworths, London, 1980. 14. C.R. Quick, Jr., Ph.D. Thesis, University of Southern California, January 1979. » 15. C.R. Quick, Jr. and C. Wittig, J. Chem. Phys., 72, 1694 (1980). CHAPTER II PHOTODISSOCIATION OF SUBSTITUTED METHANE, ETHYLENE AND ETHANE: VIBRATIONAL EMISSIONS FROM PARENT AND PRODUCT SPECIES INTRODUCTION The vibrational excitation of polyatomic species with powerful and not so powerful IR lasers has been the focus of considerable experimental and theoretical research during the past few years.One of the more persistent and elusive issues relating to this work is that of the accumulation of vibrational energy during the excitation process, as there have been few data offering a clear understanding of the process involved. In this chapter we present experimental results which address this issue. Experiments are described wherein, among others, 1 ,1C2H4F2, C2H5F and C2H3F are excited and in some cases dissociated, with the mildly focused output from a C02 TEA laser. Vibrational 18 emissions from parent molecules as well as photofragments are monitored with a sensitive IR detector in order to monitor the acumulation of vibrational energy in the molecules being pumped. We chose 1 ,1022^ 2 * C2H5F and C2H3F mainly because we know the HF(v) product distributions for all cases/ C2H3F 4* with some vibrational excitation (C2H3F ) is a product of 1 ,1C2H4F2 photodissociation, there is a reasonable difference between the vibrational state densities of those species and important emission features are spectrally resolved (see below). Other experimental goals are to measure densities of vibrationally excited parent molecules at low IR fluences where there is a small amount of dissociation, to have some insight in the randomization of vibrational energy upon absorption and see if it is possible to distinguish between x molecules each having absorbed y quanta and nx molecules each having absorbed y/n quanta. We show that, in the case of C2H3F that at laser fluences which are too low to produce HF^ emission there is also no C2H3F+ emission except via hot band absorption. For the case of C2H5F we detect C-H stretch emission that appears concomitantly with HF^ emmission while for the largest studied molecule, 1,10213^2, we 19 show that there exists a definite range of laser fluence where C-H stretch emission from the parent occurs, but j, emission from HF does not. In addition to these experiments bearing strongly on our understanding of multiple photon absorption (MPA), they are also germane to optically pumped molecular lasers. Such devices have important techonological applications (e.g., isotope separation) and are presently the topic of serious experimental research.5 When pumping polyatomic molecules in order to achieve population inversions, the traditional wisdom, that pumping harder leads to larger population inversion, needs to be reconsidered in light of the possible onset of MPA at modest fluences. For example, in the work reported below, we show that MPA is evident at -0.3 J cm“ 2 for the case of 1 ,1C2H4F2. Since MPA degrades the specificity of excitation that one normally associates with optical pumping, it is possible that optical pumping at fluences capable of MPA could lead to less rather than more population inversion. Finally, we show what effect the addition of buffer gas has on the observed emissions of C2 H5F and 1 ,1 0 2 1 * ^ 2 and we discuss our experimental results in terms of a simple physical model. 20: EXPERIMENTAL Apart from the general experimental set-up shown in the Introduction chaptert all the low fluence signals were digitized (Biomation, 10 ns minimum gate) and averaged (Tracor Northern 570A) while at high fluence signals were photographed directly. C2H3P and lrlC2H4F2 (Matheson, 99.99%) and C2H5F (CPL, 99.99%) were degassed at 77° K and used with no further purification. RESULTS C2H3F, C2H5F and 1,1C2H4F2 are easily excited with a C02 laser and dissociate to form stable molecular products via the reactions:0 AHfkJmol”1) Ea(kJmol 1) (1) mhv C2h3f C2H2+ + Hpt 71 293 (2) nhv c2H5F h- C2H4+ + HF+ 68 242 (3) lflC2H4F2 1 h v 4 * 4 * -V c2h3ft+ hft 100 259 21 The polyatomics fragments from reactions (1) to (3) are formed with considerable vibrational excitation due to the 2.160 kj mol"'*1 barrier and with a sufficiently intense IR field, this species can absorb radiation and dissociate. Although those species can absorb laser radiation more easily than the "cold" ones, its photodissociation is still a secondary process and will not be important at dissociation threshold. Above it the extent of the fragmentation of vibrationally excited products will reflect the relative dissociation cross sections of the "hot” and "cold" species. As a consequence of absorbing C02 laser photons, the excited polyatomics will fluoresce at frequencies which are characteristic of the particular species. Absorption spectra of the three species are shown in Fig. (Il-la, b and c), where the transmission regions (between points of 5% transmission) of the different interference filters are also shown. Emission spectra will very nearly mimic the absorption spectra for transitions terminating near the ground state. For transitions which occur at higher excitation levels, the emission will be broadened and shifted some to lower frequency, making quantitative analyses of our data very difficult. When undissociated parent molecules 22 FIGURE (Il-la): Infrared absorption spectra (10 cm cell, Beckman 4221 spectrophotometer) of 100 Torr each of C2H3F (solid line) and C2H2 (heavy solid line) in the 2000-4000 cm- 1 region. C2H2 has no significant absorption features in this region except the one shown. Also shown (near the bottom) are the different filters used to monitor the fluorescing species. The dashed line near the top indicates the region of the HF 1 emission. The most prominent features are the G-H stretches of C2H2 (3300 cm”1) and C2H3F (3100 cm”1). 23 PERCENT TRANSMISSION WAVELENGTH ( Mm) 3 3.5 90 80 60 50 40 30 20 2000 2500 i % 3500 3000 FREQUENCY (CM'1) 4000 FIGURE (Il-lb)s Spectrophotometer traces (10 cm cell) showing absorption spectra of C2H5F along with C2H4 (a product of C2H5F unimolecular reaction). Note the concentrations of C2H5F and C2H4. The transmission of the filter used to detect C2HtjF C-H stretch emission (dashed line) is also shown, as is the 4* region of HF: emission (horizontal dashed line) and the 5% transmission points of several interference filters (horizontal solid lines). 25 NJ c\ 0.8 c 0 .6 o C O ( / > E ( / > P 0.4 0.2 0 61 torr 14 torr CoHcF filter transmission (b ) HF emission H I . I — I J i 4 0 0 0 3500 3 0 0 0 frequency (cm 2 500 - 1 2000 ) FIGURE (II-lc): Spectrophotometer traces (10 cm cell, Beckman 4221 spectrophotometer) of 1 ,1C2H4F2 along with C2H3F (a product of 1,1C2H4F2 unimolecular reaction). Note the concentrations of each species. There are no prominent absorptions in the 2000-4000 cm"'*' region other than those shown. 100 torr C0H ,F (c) 2000 3 0 0 0 frequency (cm- 1 ) ro oo are excited via optical pumping and/or intramolecular vibrational relaxation (IVR) to levels which fluoresce, we can detect these emissions, and this is the basis of our measurements. These emissions can be compared to 4 * 4 * those from photofragments such as HF and C2H2 in order to compare the appearance of dissociation to the appearance of fluorescence from undissociated parent species.' Discrimination q£ Infrared Signals In the case of C2H3F, Fig. (Il-la) shows that we + + can detect separately emissions from HF , C2H2 and •j* different C2H3F emissions with minimum overlapping. For C2H5F and from Fig. (Il-lb) and since 4 * 4 * reaction (2) produces both C2H4 and HF , we can estimate the size of the C2H4? C-H stretch signals 4 * relative to those from HF . Using the known HF(v) distributions,^ Einstein coefficients.® and assuming a C2H4 filter transmission which covers the entire C-H 4 * stretch, region, we compute that HF produces a signal 50 times larger than from the O H stretch fundamentals, 4 * assuming each C2H4 molecule contains one quantum of C-H + stretch excitation. Thus we expect the C2H4 C-H 29 stretch signal to be very small compared to the HF+ signal following reaction (2). With the filter transmission shown in Fig. (Il-lb), we can discriminate against C2H4 emission even further m order to measure C” 2h5f^ emission. In the case of 1 ,1C2H4F2 (Pig. (II-lc)), we cah separately detect emissions from HF and the C-H stretches of 1 ,1C2H4F2. A calculation similar to the one above indicates that HF^ produces a signal > 40 4* times longer than that from the C2H3F reaction product, again assuming perfect filter transmission at the C-H stretch frequencies and that each C2H3F+contains one C-H stretch quantum. Realistics filter transmission further discriminates against C2H3F^ emission, so we can detect 1 ,1C2H4F2 C-H stretch emission in the presence of reaction (3). In separate experiments, we ensure that C-H stretch emission from C2H3F+ does not interfere with that from 1 ,1C2H4F2 by photolyzing C^H-^F at high fluence while measuring emission in the 1 ,1C2H4F2 C-H stretch region. Such checks are carried out routinely during the course of the experiments. With one exception (described below), emissions in the 2000-4000 cm“^ region other than HF and C-H stretch are very weak 30 compared to HF^ and C-H stretch emissions and will not be discussed. Typical signals are shown in Fig. (II-2). The rise of the signal is due to the bandwidth of the detector/electronics and the duration of the C02 laser output. The fall of the signal is due to diffusion and energy transfer processes. In the work reported here, only the peak signal is retained, as this is proportional to the number of fluorescing species produced, irrespective of their subsequent fate.9 It is our intention to measure processes which are not influenced by collisions, and this is a sensitive issue since IR detection sensitivity does not permit true collision-free conditions. With 50 mTorr pressure, molecules will experience one collision on the average in 2 ys, the time resolution of our experiments. Thus, we need experimental evidence that collisions do not influence our results. If collisions do not influence our results, then the signals we observe will vary linearly with pressure. Data showing emission signals vs. pressure are given in Fig, (II-3). The variations are linear at low pressure and deviate from linearity at pressures which reflect the particular collisional process(es). In our experiments, we 31 FIGURE (11—2): Typical fluorescence signals obtained with 55 mTorr 1 ,1C2H4F2 an^ 5.7 J cm"^. The solid line is the HF^ signal and the dashed line is the 1 ,1 0 2* 3 ^ 2 C-H stretch signal. The signals are normalized for convenience, and the risetimes are limited by the detection system. 32 £Z FLUORESCENCE SIGNAL (ARB. UNITS) m ro to FIGURE (II-3): Peak fluorescence signals vs. pressure of the studied species: A indicates the HF^ signal from C2H3F dissociation, • indicates the C-H stretch signal from 1 ,1C2H4F2, O indicates the HF^ signal from 1 ,1C2H4F2 dissociation, A indicates the HF^ signal from C2H5F dissociation. 34 S£ PEAK FLUORESCENCE SIGNAL (ARB. UNITS) typically use pressures of 50 mTorr for 1 ,1C2H4F2* 90 mTorr for C2H5F and 150 mTorr for C2H3F, as a reasonable compromise between adequate S/N and avoiding effects due to collisions. The essential experimental results are shown in Figs. (II-4-6). Signal amplitudes should not be confused with concentrations since a species may go undetected because of low oscillator strength or low concentration. £2% f This specie has an absorption band centered at 944 cm”^ (Vg) which coincides with the C02 laser line of the same frequency (P(20) 00° 1-10° 0).11 As the laser fluence is raised, emissions from different vibrations can be detected, and these data are shown in Fig. (II—4). As mentioned above, the overall shapes of the variations shown in Fig. (II-4) are not as important as the intercepts and threshold regions. Reasons for this are (a) the broadening and shift to lower frequency of the emission from vibrationally excited species, (b) the increase in spontaneous emission rate due to the excitation of more than one quantum of a particular mode, (c) the decrease in spontaneous emission rate due 36 FIGURE (II-4): Peak fluorescence signals vs. fluence for C2H3F. The various emissions are identified according to the absorption features shown in Fig. (Il-la). The pressure is 160 mTorr. S£ PEAK FLUORESCENCE SIGNALS (ARB. UNITS) o <jn O m o ro to intramolecular relaxation into the available vibrational degrees of freedom, and (d) the different physical bases for vibrational excitation in the various species (e.g. HF^ derives its vibrational excitation 4* almost exclusively from reaction ( 1), while C2H3F derives its excitation from optical pumping). The CC>2 laser pumps the Vg vibrational mode, and we excite the \ > 3+vg vibration via the hot band absorption (V3=l, Avg=l). Since less than 4x10“^ of the C2H3F molecules have V3=l at room temperature, it is clear that only a small fraction of the C2H3F molecules either fluoresce at v> 2000 cm-- * - or dissociate. There is no fluency threshold for this hot band absorption. We are unable to detect hot band excitation of Vg-Kv9 since the number of molecules with v9=l is an order of magnitude less than for V3=l, and consequently, excitation of V6+V 9 occurs only via MPA. Leaving aside the issue of hot band absorptions, the data in Fig. (II-4) show that the threshold for the 4* observed C2H3F emissions is roughly the same as for dissociation. No 0 2!!^^ emission ( >2200 cm"-*) is detected at fluences less than those required for photodissociation. Since the signals which we detect are proportional to the product of the spontaneous 3 9 emission rate and the number density, we can only conclude from our data that, below 2 J cm“^, the C2H3 F^ density is very small. Thus, in this case the accumulation of vibrational energy may be hindered at levels where C2H3F^ emission is not possible. Our data cannot say more about the precise location of these levels. %>%£ This specie has an absorption band centered at 972 cm-- 1 - (C-C stretch) ^ that coincides with the (R(16) 00° 1-10° 0) C02 laser transition. There are five C-H stretches quite close to one another which center is at 3000 cm“^ (see Fig, (Il-lb)). Data for C2H5F are shown in Fig. (II-5a) and indicate that C-H stretch emission is detected only at fluences where HF^ emission is also seen. Based on the known HF(v) distribution,4 Einstein coefficients®, relative D values, and filter transmissions, we estimate that the HF^ emission signal from a single dissociation event is 2.4 times larger than the C-H stretch signal from a single C2H5F molecule with one quantum of C-H stretch excitation. A C2H5F molecule with (randomly 40 FIGURE (II-5a): Various emission signals vs. fluence for C2H5F. The listed fluence values are obtained by dividing the laser energy by the 3 mm^ aperture through which 80-85% of the laser energy can pass. The sample consists of 95 mTorr C2H5F. There was no detectable signal from the C-H stretch of C2H4. 41 ZP PEAK FLUORESCENCE SIGNAL (ARB. UNITS) O CTJ ~n o t r > n> o m ro o m n> < / > o distributed) vibrational energy equal to Ea on the average contains -4000 cm" 1 of C-H stretch excitation. Thus, the signal amplitudes and the roughly equivalent detection thresholds shown in Fig. (II-5a) mean that there is not a large amount of vibrationally excited parent at fluences where we begin to detect HF . It follows from Fig. (II-5a) that if C-H stretch emission is due to molecules containing randomly distributed vibrational energy, then there are comparable concentrations of dissociation products and parent molecules with energy 2 . 0.5 Ea, and the relative concentrations of these species do not change markedly as the fluence is varied. Our measurements do not identify the precise nature of the vibrational excitation of emitting parent molecules, but there is certainly not a sizeable reservoir of molecules containing randomly distributed vibrational energy at fluences where dissociation is first detected. Fig. (II-5b) shows the effect on the HF^ signal near threshold with He addition. The trend is quite similar with the one obtained previously with C2H3F^k showing the importance of rotational filling at low •I* fluences through the enhancement of HF signal with a few Torr of buffer gas. 43 FIGURE (II-5b): Dependence on laser fluence of the HF peak fluorescence signals from excitation of C2H5F, Conditions are: O 100 mTorr of C2H5F, • with 4 Torr of the added, A with 8 Torr of He added. Also shown is the enhancement vs. He pressure at low fluence (near threshold). 44 Sfr. h f t peak f l u o r e s c e n c e s ig n a l enhancement foctor ‘ ro cr o o o o P o o o r o X X w u> c* r o C D X n> XrlQ2_-&-l This specie has an absorption band centered at 944 cm-1 which has been assigned to the first overtone of the fundamental CF2 rock at 472 cm-- * - , and which coincides with the (P(20)00°1-10°0) CO2 laser transition.1^ There are four C-H stretch frequencies and their absorptions are quite close to one another. The center of these C-H stretches is at 2980 cm-1, and the other absorptions with appreciable oscillator strength in the region v>> 2 0 0 0 cm-1 are easily confused with emission from C2H3F+ an<^ not be dealt with here. The unimolecular reaction of 1 ,1C2H4F2 proceeds via reaction (3), with the subsequent photolysis of the C2H3F+ photoproduct proceeding via reaction (1). The nascent distribution of HF**" formed via reaction (1) is almost identical to that formed via reaction (3),^ and therefore we will make no attempt to distinguish between these species. The extent to which reaction (1) contributes to the HF emission from the IRMPD of 1 ,1C2H4F2 is not known precisely, although we can measure this indirectly by monitoring C2H2 (described below). At low fluence, reaction (3) dominates over reaction (1), and therefore the threshold regions are 46 the most straightforward as far as interpretation of results is concerned. Figure (II-6a) shows the dependence on laser fluence of the peak fluorescence signals which correspond to the different spectral regions under consideration for 1 ,1C2H4F2. The data show clearly the threshold regions for three important emissions and here we can see striking effects. Emissions from the C-H stretches of 1rIC2H4F2 can be seen in the absence of HF emission, and with an appearance threshold of approximately 0.3 J cm“2. Thus, relative to the case of C2H5F, a large number of parent molecules are excited prior to dissociation. In comparing Figs. (Il-lb) and (c), it is noteworthy that the C-H stretch integrated absorption is twice as large for C2H5F as for C2H4F2. Despite this, C-H stretch signals from 1 ,1C2H4F2 ^ are detected at lower fluences than those required to detect emission from HF- * * rather than appearing at 2 the fluences required to detect HF^ as in the case of C2H5F. For both C2H5F and 1 ,1C2H4F2 , we have taken rough emission spectra using interference filters, and near threshold (1-3 J cra"^) we find rather small signals except in the regions of HF and C-H stretch. For example. Fig. (II- 6a) shows how the emission drops in the region spanned 47 FIGURE (II-6a): Dependence of laser fluence of the peak fluorescence signals from different 4* spectral regions corresponding to HF (A), 1,1C2H4F2 C-H stretches (•), and C2H2 C-H stretches (O). The data were taken with 56 mTorr 1,1C2H4F2 and no buffer gas. The signals are corrected for the frequency dependence of the detector and the filter transmissions. Signals in the frequency range 2000- 2600 cm-1 are small compared to those shown. Note that below 1.5 J cm"^ there is no detectable signal in the region between HF^ and the C-H stretch. 48 PEAK FLUORESCENCE SIGNAL (ARB. UNITS) 4* by HF and the C-H stretch of 1 ,1C2H4F2. Here emission in the region 3250-3400 cm"^ is detected only at > 1.5 J cm~^. This emission could be due to the small amount of HF emission which occurs in this region, or to C2H2 + produced by photolysis of the C2H3F photoproduct of reaction (3). Although the latter choice is consistent • f * . with the change in slope of the HF signal at 1.5 — 9 J cm , signal versus fluence variations such as those shown in Fig. (II-6a) cannot be used to prove such hypotheses since the variations can derive from a number of sources. This is particularly true at low fluence and with low parent concentrations where "bottleneck" effects may be present. Under these conditions, only a small subset of the molecular ensemble interacts with the laser field. The emissions which are observed are a consequence of various molecular states interacting with the laser field, but these emissions cannot be used to determine the states initially involved in the excitation process. In the case of l,lC2H4F2f the addition of 10 Torr • f He does not affect either the HF or C-H stretch peak fluorescence signals, as shown in Fig. (Il-6b) and (c). Previously,^*3 we found that of all the fluorinated ethanes and ethenes which we studied, 1 ,1C2H4F2 showed 50 Dependence on laser fluence of the peak fluorescence signals from excitation of 1 ,1C2H4F2* (b) and (c) refer to HF^ and 1 ,1C2H4F2 C-H stretch respectively* In (b), conditions are: • 59 mTorr lrlC2H4F2 r O 56 mTorr 1 ,1C2H4F2 and 10.7 Torr He. In (c), conditions are: # 58 mTorr l,lC2H4F2r O 54 mTorr lrlC2H4F2 and 10 Torr He. 51 ZG P E A K H F + F LU O R E S C E N C E S IG N A L (A R B . U N IT S ) O O ? £S DEAK ~ n m o m o ro 1 ro 1 , 1 - c 2h4 f 2 f l u o r e s c e n c e s i g n a l (CH S TR E TC H ) !N ARB. UNITS O O © + cn cn 00 0 3 3 0 0 0 s X CD — i b - 1 1 O O ro ro X X -n ro rT 1 o O o o o o o j ? the least enhancement with He addition, and we tentatively ascribed this to the ease with which 1 ,1C2H4F2 was excited and the unimportance of rotational occupancy. Our present results underscore this hypothesis, and Fig* (II-6c) shows that C-H stretch emission from lflC2H4F2 is unaffected by the addition of 10 Torr He. This indicates that rotational occupancy is not very important in the excitation of l,lC2H4F2f presumably due to the high density of states of 1,1C2H4P2 .14 4 * This trend for the HF signal is verified up to 5 J cm , when the signal starts to show some enhancement depending on buffer pressure. Figure (II-6d) shows that behaviour for different He pressures and over a wide range of energies. The high fluence results reproduce previous ones^*3 and the data shows that contrary to the other studied species the "hole burning" and "rotational filling" effect tends to become important at E-^E^hres. and that there is an optimum buffer pressure for the latter effect and that an increase of it works relatively against the dissociation process.-*-® We have obtained data similar to those presented above for a number of different molecules-^' and are 54 FIGURE (II-6d): Dependence on laser fluence (5-35 J cm"2) of the HF ^ peak fluorescence signal for different He buffer pressures on the dissociation of 1,1C2H4F2. The maximum enhancement is * found at -5 Torr of He at any fluence over 5 J cm"2. 55 H F t PEAK FLUORESCENCE SIGNAL 55 mtorr 1,1C2H4 F2 O 55 mtorr 1,1C2H4F2 + 5 torr He A 55 mtorr 1, t C2H4 F2 + 3 0 torr He A 59 mtorr 1,1 C2H4 F2 + 6 0 torr He c r > 20 FLUENCE (Jem"2) summarized in Table (II-l). The general trends are the same as for lflC2H^F2 f C2H5F and C2H3F, but offer little new physical insight, so these results will not be discussed in detail. With CF2HC1, where HC1 elimination is the lowest energy dissociation channel, we see *4 * considerable HC1 emission but no C-H stretch emission. This is not surprising given the sparse density of states of this molecule. The same was observed for HF and C-H stretch emissions in 1 ,1021*^ 2. DISCUSSION Our results show important differences in the IRMPA excitation of C2H3F, C2H5F and 1 ,1C2H4F2. With 4* C2H3F, the threshold for observing HF emission is the same, within experimental uncertainty, as the threshold for observing any emission in the 2000-3200 c i t i " " * - region, except for the hot band contribution (see Fig, (11—4)). The relative intensities shown in Fig. (11—4) are of no particular significance, since they are influenced by the spectral width of the emitter, the interference filter bandwidth, spatial effects in the laser beam, excitation level of the emitting species, Einstein _____________________ TABLE (11—1): Summary of parent infrared emissions from different substituted ethane, ethylene and methane, together with its main molecular parameters. 58 M o l e c u l eP r o m p t S i g n a l S t e t c h R e g i o n O p { c m 2 ) a t P u m p i n g M o d e 3 E x p e r i m e n t a l E v i d e n c e o f C - H S t r e t c h E m i s s i o n s O s c i l l a t o r S t r e n g t h C - H S t r e t c h ( x l O ' ) “ r > 1 4 V - R C - H S t r e t c h E m i s s i o n B e l o w D i s s o c i a t i o n T h r e s h o l d E f f e c t , o f B u f f e r o n D i s s o c i a t i o n C h a n n e l i S i g n a l 1 - 1 c 2 h 4 f 2 Y e s 9 x l 0 “ 2 0 Y e s 6 1 5 0 0 Y e s I n c r e a s e a t E > 4 E t n r e s c 2 h 5 p Y e s 1 3 x l 0 ~ 1 9 Y e s \ 1 3 4 1 8 0 N o I n c r e a s e a t E ■ E t h r e s 1 * 1 C 2 H 2 P 2 N o N o 6 . 5 1 7 U N o I n c r e a s e a t E • E t h r e s C 2 » 3 P Y e s 6 . 8 x l 0 “ 1 9 N O 8 . 6 4 0 N o I n c r e a s e a t E ■ E t h r e s C F 2 H C 1 N o N o 5 7 b N o I n c r e a s e a t E * E t n r e s c h 3 p Y e s Y e s 5 1 < 1 0 Y e s N o D i s s o c i a t i o n o b s e r v e d o s c i l l a t o r B t r e n g f c h 8 B P 4 0 0 ‘ o b t a i n e d b y l o w s i g n a l a b s o r p t i o n c o e f f i c i e n t s ( a 0 ) u s i n g d a t a a s i n P i g s . I I - l a , b , c . “ o b t a i n e d b y o w n c a l c u l a t i o n s o f i n t e g r a l a b s o r p t i o n c o e f f i c i e n t s u s i n g i n f r a r e d s p e c t r a a s i n F i g s . I l - l a , b , c . coefficients, etc. The intercepts, however, are more significant since they indicate the onset of the observed emissions, and detection sensitivity does not vary so much that the intercept is markedly affected. In contrast to the above, emission from the ^3 + ^ 5 vibration shows no threshold. Emission here is obtained by pumping the hot band v3"> ' v3+v6' an<^ therefore this result is not surprising. Since < 0.4% of the molecules have one quantum of V3 excitation at room temperature, it follows that 1 0“^ of the C2H3F molecules dissociate at fluences < 5 J cm- 2 ,18 The case of C2H5F is more clear than C2H3F due to the fact that the only signals arose from HF and C-H stretch spectral regions. Following the same arguments as above, with C^gF it is clear that excitation is encumbered at low fluence where fluorescing levels (2000-3200 cm-*) cannot be accessed by optical pumping and/or IVT- Following passage through these levels excitation is facile and continues until the molecule dissociates, even at fluences ~ 1 J cm-2. In the case of 1,1C2H^F2, C-H stretch emission occurs at lower fluences than those required to effect dissociations, and it was not possible to monitor combination band emissions due to the overlap of 60 lflC2H4E*2 vibrations with those of the 0 2^?^ dissociation fragment. The C2H3F+ fragment contains considerable vibrational excitation, and should therefore interact readily with the laser field. In fact, Fig. (II-6a) shows that emission characteristic of the C-H stretches of C2H2 begins at roughly the same fluence as where the HF^ emission signal takes a sharp turn upward. This may be mere coincidence, but it is worth noting nevertheless. For resonant transitions below the quasi-continuum, optical pumping is very — 9 —1 rapid. At 1 J cm , power broadening is ~ 0.1 cm x and molecules present in the laser beam quickly gain a 1 9 quantum of vibrational excitation. Subsequent excitation is affected by the interplay between anharmonicity and power broadening, and it is here that the bottleneck behavior will be manifest. At high levels of molecular excitation, where there is continuum absorption, the absorption cross section will diminish at any particular frequency. Nevertheless, our data indicate that for C2H3F and more clearly for C2H5F this does not result in a serious barrier to further excitation, but rather that the bottlenecks occur at lower levels. Separate experiments in our laboratory confirm that molecules such as C2H3F, C2HJ5F and 61 1/1C2H^F2 are absorbing > 1 photon ns""^ when they dissociate. For 1,^1C2H^F2 it is clear that significant internal excitation can occur without dissociation, and the dissociation threshold occurs at lower fluence than with C2H3F. We have calculated state densities for C2H3F, C2HgF and 1 ,1C2H4F2, and find that the density of states for 1 ,1C2H4F2 exceeds that for C2H3F by roughly an order of magnitude when each species has absorbed one C02 laser photon (see Table (II-l)). Although precise state densities are poorly defined at such low levels of excitation, this is in qualitative accord with our experimental findings. C-H stretch emission requires that the emitter contains at least 3000 cm"-1 - of vibrational excitation and our measurements show that 1 ,1C2H4F2 accumulates this excitation somewhat more easily than does C2H5F and in like manner C2H5F accumulates this energy more readily than does C2H3F under similar experimental conditions. If we related this trend with the size (or density of states) of the species (see Table (11—1)) there seems to be a relation between the increase in molecular size with the ease in the accumulation of vibrational energy measured through our C-H stretch emission "thermometer." 62 The same issue was pointed out in very recent data where it was found some localization of vibrational 22 23 energy as the size of the molecule is decreased. f CONCLUSIONS We have shown that following IRMPA on 1 r1C2H^F2 there is a measurable density of vibrationally excited parent molecules below dissociation levels as compared with C2H5F and C2H3F. The ability to excite these species below the quasi continuum varies with its molecular size being more facile for 1 ,1021*^ 2. Finally we showed that the rotational occupancy tends to be more important near threshold for the smaller species. 63 REFERENCES la. V.S. Letokhov and C.B. Moore, Sov. J. Quantum Electron. 6, 129 (1976); ibid, p. 259. b. S. Kimel and S. Speiser, Chem. Rev. 77, 437 (1977). c. V.S. Letokhov, Sov. Phys. Usp. 21, 405 (1978) - d. P. Schultz, A. Sudbo, D. Krajnovich, H. Kwok, Y. Shen and Y.T. Lee, Ann. Rev. Phys. Chem. 30, 379 (1979). 2a. R.L. Woodin, D.S. Bomse and J.L. Beauchamp, Chem. Phys. Lett., 63, 630 (1979). b. D.M. Cox, R.B. Hall, J.A. Horsley. G.M. Kramer, P. Rabinowitz and A. Kaldor, Science, 205, 390 (1979). 3a. M. Quack, J. Chem. Phys. 69, 1282 (1978). b. J.C. Stephenson, D. King, M. Goodman and J. Stone, J. Chem. Phys. 70, 4496 (1979)- 4. C.R. Quick, Jr., and C. Wittig, J. Chem. Phys. 72, 1694 (1980) . 5a. J.J. Tiee, T.A. Fischer and C. Wittig, Rev. Sci. Instrum. 50, 958 (1979). b. R.S. McDowell, C.W. Patterson, C.R. Jones, M.I. Buchwald and J.M. Telle, Opt. Lett. 4, 274 (1979). 64 c. V.V. Lobko, Sov. J. Quantum Electron. 9, 498 (1979). d. A. Stein, P. Rabinowitz and A. Kaldor, Opt. Lett. 3, 97 (1978). 6a. E. Tschuikow-Roux, W.J. Quiring and J.M. Simmie, J. Chem. Phys. 74, 2449 (1970). b. J.M. Simmie, W.J. Quiring and E. Tschuikow-Roux, J. Chem. Phys. 74, 992 (1970). c. M. Day and A.F. Trotman-Dickenson, J. Chem. Soc. A, 233, (1969). d. S.W. Benson, "Thermochemical Kinetics," 2nd Ed., Wiley, New York, 1976. Tables A8 and All. 7a. C.R. Quick, Jr. and C. Wittig, Chem. Phys. 32, 75 (1978)* b. C.R. Quick, Jr. and C. Wittig, J. Chem* Phys. 69, 4201 (1978). 8a. J.M. Herbelin and G. Emanuel, J. Chem. Phys. 60, 689 (1974). b. A.M. Thorndike, A.J. Wells and E.B. Wilson, J. Chem. Phys. 15, 217 (1947). 9. In general, a curve with exponential rise and fall must be deconvoluted to obtain xrise, x fall, and the pre-exponential factor which is proportional to species concentration. If x rise << % fall, or t rise and Tfall remain constant as some parameter is varied, then the peak of the temporal curve remains proportional to species concentration as the parameter is varied. A complete discussion is given in Ref. 10. 10. A. Hariri and C. Wittig, J. Chem. Phys. 67, 4454 (1977). 11. B. Bak and D. Christensen, Spectroch. Acta 12, 355 (1958). 12. L.Y. Liang, L.W. Daasch and J.R. Nielsen, J. Chem. Phys. 22, 1293 (1954). 13. A. Smith, B. Saunders, C. Nielsen and D, Ferguson, J. Chem. Phys. 20, 847 (1952). 14. Using the Whitten-Rabinovitch approximation15 to evaluate the V,R density of states, we find 590 per cm” 1 for 1 ,1C2H4F2 and 40 per cm- 1 for C2H3F at E = <Evib>300°K + hv(944 cm”1). 15. P.J. Robinson and K.A. Holbrook, "Unimolecular Reactions," Wiley, NY, 1972. 16. We search for the same effect looking at the C-H stretch signal. While up to 5 Torr of He and 5 J mm O * f * cm it follows the HF trend, over that pressure and fluence there is a sharp decrease of its decay time while the HF^ one remains unaffected making it 66 very difficult to draw more conclusions. The same effect was seen in C2 H5F but at threshold and with 1-2 Torr of buffer. 17. J. Caballero and C. Wittig, "Location and Character of Bottlenecks in the IR Multiple-Photon Excitation of Polyatomics Molecules." Presentation D3 at the 11th International Quantum Electronics Conference, Boston, Massachusetts, June 1980. 18. This is estimated by assuming that the number of molecules which interact with the laser field is given roughly by the ratio of the power broadening at threshold (~0.1 cm"^® and the width of the — 1 91 rotational manifold of these molecules (10 cm ). x 19. Here, we assume a laser pulse width of 0.8us, and use oscillator strengths from Ref. 20. 20. R.T. Rung, Chem. Phys. Lett. 57, 273 (1978) , 21a. R. Elst and A. Oskam, J. Mol. Spect. 39, 357 (1971). b. R.T.V. Kung and H.W. Friedman, J. Chem. Phys. 72, 337 (1980). 22. J.S. Francisco, W.D. Lawrence, J.I. Steinfeld and R.G. Gilbert, J. Phys. Chem. 8 6, 724 (1982). 23. D.M. Brenner, J. Phys. Chem. 8 6 , 41 (1982). 67 CHAPTER III IR MULTIPLE PHOTON DISSOCIATION OF C2HC13: MOLECULAR ELIMINATION vs. BOND FISSION, AND EFFICIENT DISSOCIATION OF THE C2C12 PRODUCT INTRODUCTION The unimolecular reactions of halogenated ethanes and ethenes have been studied with both experiments and calculations for many years. These species are important prototypes and building blocks for a significant part of chemistry, and the elementary processes involved in the making and breaking of their bonds are of more than simply academic concern. Reactions which lead to molecular and/or radical fragments have been perused using different methods to prepare reactants with energies in excess of reaction threshold.-*- One of the central issues in this work is the competition between simple bond fission and molecular elimination, as the latter concerns a part of 68 the potential energy hypersurface which is far from the equilibrium configuration, but which is nevertheless accessible to experiments as well as calculations. By measuring the quantum state distributions of nascent reaction products, such as the hydrogen halide diatom, and by comparing these distributions to those which would derive from different potential surfaces, it should be possible to ascertain topological features in regions which are chemically interesting. Typically, these experiments are done in the gas phase on a time scale which is sufficiently short that relaxation processes do not degrade the nascent excitations of interest. It would be valuable to extend such measurements to the condensed phase, but it is unlikely that the present level of experimental detail could be preserved in such environments, at least in the near future. In the cases of fluorinated ethanes and ethenes. reaction has been shown to proceed primarily by HF elimination, through a , a and/or a , $ type processes. Often the HF thus produced contains vibrational excitation, hereafter indicated by , which can be detected by its spontaneous emission, thereby allowing V,R energy distributions to be determined for 6 9 9 — 4 vibrationally excited products. To date, there has been no documented case of bond fission competing successfully with HF elimination in the region near reaction threshold, although the issue of «, a;vs. a, 3 pathways remains unresolved.^ In the cases of chlorinated ethane and ethene, a number of research groups have used IRMPA to energize reactants under conditions which either minimize or encourage collisions prior to reaction. ^ IRMPA offers the advantages of excitation via the ground electronic state and the opportunity for collision free excitation. The main disadvantage is the preparation of a sample with a distribution of energies from which reactions occur. Using product analyses, which is well suited for cases in which unimolecular reaction produces stable Q molecular products, Reiser et al. scrutinized the HC1 elimination pathways in several systems, and found that * both a,a and a, 3 eliminations occurred under their experimental conditions. Their results were supported by the calculations of Kato et al.,^® which showed almost identical activation energies, and comparable dissociation rates, for 3 and 4 center eliminations from vinyl fluoride. Sudbo et al.,^ using molecular beam techniques, were able to measure translational energy 70 distributions of the HC1 product following the IRMPA of halogenated hydrocarbons, thus confirming this reaction pathway, and allowing estimates to be made of the reaction rates and the energies in excess of reaction threshold, Morrison et al.,*^ in a complete study of the IRMPD of methyl freons, showed the importance of molecular CI2 elimination in competition with simple bond fission. Using conventional UV photochemical methods, Holmes and Setser^^ obtained results whose similarity to those of Reiser et al.^ suggests that the ground electronic state may be formed following UV excitation, Berry-^ has discussed the importance of Cl production following UV excitation of chlorinated ethane and ethylene. Because of the high excitations in his work, it is not surprising that bond fission would compete with molecular elimination. As the amount of chlorine substitution is increased, it is reasonable that the molecular elimination channel would benefit, since the chlorine atoms are more accessible to the H atoms, while the energy required for Cl bond fission is not affected significantly. Thus, using trichloroethylene, hereafter referred to as TCE, separate groups have found evidence for only molecular elimination of HC1 following the 71 O Q IRMPA of TCE, using product analyses as a diagnostic. On the other hand, molecular beam experiments using IRMPA,6 and shock tube studies14 which access molecular excitations similar to those of the molecular beam IRMPA experiments, open the possibility of Cl atom bond fission as one of the primary dissociation processes. In order that competitive reaction channels be properly identified, it is necessary that primary reactions be distinguished from secondary ones, which occur subsequent to the fragmentation of parent molecules. For example, Wurzberg et al.16 and Horwitz 1 et al.° were able to show that consecutive IRMPA driven unimolecular reactions were responsible for the signals they obtained when using substituted methane parent molecules. In this paper, we present experimental results in which both HC1 and Cl are nascent products of the irradiation of TCE with the focused output from a pulsed CC>2 laser. In estimating [C1]/[HC1], changing the energy in excess of reaction threshold, and comparing our results to RRKM calculations for a wide range of acceptable parameters, it is clear that the Cl is produced by the efficient IRMPA driven dissociation of the C2Cl2 which derives from reaction (2a), and not from reaction (2b), as per the reaction sequence shown 72 below C2HC13 + nhv+C2HCl3 f (1) C2HC13+ C2C12++ HClf (2a) C2HCl3f -> C2HCl2f+ Cl (2b) C2C12++ mhv - » ■ C2Cl2t (3) C2C121: C2C1+ + Cl (4) where 1* denotes vibrational energy in excess of reaction threshold, and nun can assume values consistent with the approximate equality of the rates of unimolecular reaction and optical pumping to higher levels. EXPERIMENTAL IRMPA and IRMPD are effected using the output from a pulsed C02 laser (Lumonics), which is focused with Ge lenses of either 4 or 30 cm focal length into the chamber, which is connected to a conventional vacuum system. In the focal region, 80% of the energy can pass through 3 and 1.2 mm4 circular apertures, for the 30 and 4 cm lenses respectively and fluence is estimated in this way. The detector output is always amplified, digitized (Biomation, 10 nsec minimum gate), and 73 i averaged (Tracor Northern 570A) until a suitable signal to noise ratio (S/N) is obtained. TCE (Mallinckrodt, AR) was purified by trap to trap distillation, and its purity was > 99.9% as determined by gas chromatography. HBr (Matheson, 99.8%) was distilled to remove Br2, degassed at 77°K, and used with no further purification. DBr and DC1 (Stohler, 98%) were used without purification. RESULTS Nascent HC1 Absorption spectra of TCE in the regions 900-1000 and 2700-3300 cm-1 are shown in Fig. (III-l), The frequency of the (001)-(100)P(20) CC>2 laser transition coincides nicely with those of the absorption feature of TCE (CC12 asymmetric stretch),17 and IRMPA transpires via this absorption system. The transmission ■ f band of the interference filter used to monitor HC1 emission is also shown, and no significant interference from parent or photofragments is expected in the region , j, 2700-3000 cm-1. Spontaneous emission from HC1 is readily detected following the IRMPA of TCE with the 74 FIGURE (III-l): Spectrophotometer traces (10 cm cell) showing absorption spectra of trichloroethylene in the regions 800- 1100 cm-* and 2700-3300 cm-*. Also shown is the transmission of the narrow bandpass interference filter used to monitor IR emission near 2900 cm-*, the HCl emission region. This filter transmits 60, 50, 30, and 15% of the vibrational emission emanating from v = 1, 2, 3, and 4 respectively, for the case of thermalized (300®K) rotations. 75 CO CO CO 2 <t DC i — LlJ u DC UJ a 8 0 7 0 6 0 5 0 4 0 3 0 - 20 « 0 ' /v yi ? ' V '-£V 6 0 Torr CJHCL 1 A I u \ A-i- CH stretch - NARROW BANDPASS FILTER 1 A HCI (v = t) emission 3 2 0 0 J L 2800 cm -1 1 I 11 1 3 .0 3.5 1 0 A 9 Torr C2HCi3 v CCI 10v.u»2 asymmetric stretch 1 1000 I ___ 9 0 0 cm J -i 1 1 12 M nr 76 focused CC>2 laser output. HC1+ is easy to monitor 1 p because of its high spontaneous emission rate, and the uncluttered spectral region where the emission occurs. Even without signal averaging, HC1 could be detected with TCE pressures as low as 10 mTorr. With gas samples containing TCE, either with or without a buffer, HC1 is a nascent product whose production occurs in the presence of the IR laser radiation. Figure (III-2) shows the effect due to the introduction of a cell, containing room temperature HC1, between the irradiated volume and the detector. This gas filter absorbs spontaneous emission from HCl(v=l) except for high J, and transmits most of the radiation from HCl(v^.2) as well as HCl(v=l, high J). Thus, most of the emission is due to HC1 (v=l) . as is common for molecular eliminations of this kind. In the experiments reported here, it is important to verify that collisions do not introduce artifacts into the measurements. Although the signal decay rate is not of concern in our measurements, the risetime is long enough (1.5 ys) that collisions do occur during this time, even at TCE pressures of 10’s of mTorr. Certainly, there is significant rotational relaxation in several of our experiments with respect to both parent 77 FIGURE (III-2): Identification of the IR emission at 2900 cm-1. Trace (a) shows the signal obtained when photolyzing 45 mTorr of TCE with the focused CC>2 laser output (~ 35 J cm"2, (001) - (100)P(20)). Trace (b) shows the effect of an,HC1 gas filter (650 Torr) in front of the detector. The rapid fall of the signal in (b) at short times is due to rotational relaxation of the nascent HC1- Allowing for rotational relaxation, the respective signal amplitudes of (a) and (b) are consistent with the HCl(v) distribution associated with molecular elimination. 78 < o CO (a) EMPTY GAS FILTER (b) 650 Torr OF HCI IN GAS FILTER LU U e UJ u CO UJ a: O 3 Ul U X \ o and product species. However, we are not concerned with rotational degrees of freedom in either parent or products, and it is HC1 vibrational excitation which requires scrutiny. We find that the HC1+(t=0) spontaneous emission signals (t=0 refers to the peak of the signal under collision free conditions) vary linearly with TCE pressures, even with high laser energies, until approximately 200 mTorr, providing an 1 Q ample safety margin in our measurements. There was no measurable effect on the HCl^(t=0) signals with the addition of up to 10 Torr of an inert buffer (either with or without the HC1 gas filter) as has often been reported for other systems, possibly because the high vibrational state density of TCE provides many transitions at a particular C02 laser frequency, thereby lessening the opportunity to enhance the excitation process vxa rotational relaxation. ' The dependence of the HC1+(t=0) signals with laser energies shows a phenomenological threshold at 75 mJ, corresponding to ~2.5 J cm over a 3 mm^ area. We have looked for IR 4* emissions other than those from HC1 , throughout the region 2000-4000 cm-^, and have found none whose intensity was significant (within a factor of 30 of that from HC1 ). 80 The Production of Cl In order to detect Cl atoms, samples were photolyzed which contained both TCE and HBr. Chlorine atoms react rapidly ((7,5+0.7) xlO" 12 cn^molec"1^"1) with HBr:23 Cl + HBr -* HC1 + Br, A H= -15.5 kcal mol"1, (5) and the HC1 thus produced is vibrationally excited and detected by its spontaneous emission. Reaction (5) is not sufficiently exothermic to produce HCl(v^.2), and greater than 80% of the HC1 is produced in v=l.2^ Figure (III-3) shows a typical signal obtained when photolyzing a mixture containing both TCE and HBr. The signal contains contributions from the molecular elimination of HC1 from TCE, and the production of HC1(v=l) via reaction (5). The contribution due to reaction (5) is obtained by subtracting the molecular elimination component from the total signal, and the value of k5 obtained by analyzing curves such as trace (d) in Fig. (III-3), for HBr pressures in the range 10- 400 mTorr, is (6.8±1.6)xlO" 12 cm3molec"1s"1. The large uncertainty derives from the subtraction of trace (a) from trace (b) to obtain trace (d). Nevertheless, the agreement with the more accurate, independently 81 FIGURE (III-3): HCl^ fluorescence following the IRMPD of TCE. Trace (a) is for the case of 40 mTorr of TCE and is similar to trace (a) in Fig. (III-2). Trace {b) is for the case of 40 mTorr TCE and 40 Torr of an HBr/DBr mix. It is clear 4* that HC1 emission from the Cl+HBr HC1 +Br reaction is superimposed on the emission seen in trace (a). DCl^ emission is observed at 2400 cm-1 but is not shown in the figure. Trace (c) is the signal transmitted through an HC1 gas filter. Notice that there is no component due to the Cl+HBr reaction, as expected from the enthalpy change for this reaction. Trace (d) is due to subtracting trace (a) from trace (b), and is due to the Cl+HBr reaction. 82 es HCI FLUORESCENCE SIGNAL Q CL cr 04 measured value of k5 is quite reasonable. That these signals derive from HCl(v=l) was verified with an HC1 gas filter (trace (c)). At low pressures, the spectral + lineshapes of the HC1 spontaneous emission are primarily Doppler broadened, and we found that the transmission through the HC1 gas filter was unaffected by the amounts of HBr used in our experiments. Thus, we were able to verify that the HCl^ formed via reaction (5) is almost entirely HCl(v=l), in agreement with the enthalpy change for the reaction. An alternate experimental approach to determine the separate contributions from reactions (2a) and (5) is to use DBr in place of HBr, thereby allowing DC1 to 4* be detected without interference from the HC1 emission which derives from reaction (2a). We did this, and results obtained thusly were the same as those obtained using the methodology described above. A complication associated with this technique is isotope exchange on the walls of the chamber. Careful and exhaustive seasoning of metal surfaces with D2O is essential, and we were never able to completely eliminate isotope 1* exchange in our measurements. Although a DC1 spontaneous emission signal could be monitored 4* 4* independent of HC1 , the HC1 signal always contained a 84 contribution from reaction (5). Thus, although the use of DBr is aesthetically attractive, it is not clear to us that it leads to more reliable results than those obtained by subtracting the contribution of reaction (2a) from the total signal, and the data presented in this chapter were obtained using HBr rather than DBr. Fluence dependences It is now rather widely recognized that little valuable information is contained in the variation of a particular signal amplitude with CO2 laser fluence in experiments of the type reported here. More germane, are comparisons between "fluence dependences" in instances where the differences, or lack thereof, provide information. Following the method discussed above, we were able to determine [C1]q/[HC1]q from the chemiluminescence traces. The subscripts refer to the t=0 values, and we make use of the HCl(v) distributions associated with reactions (2a) and (5), thus allowing us to convert the relative chemiluminescence signal intensities to relative number densities. When varying the fluence by an order of magnitude (10-100 J cm £), there was no pronounced or.systematic change in the 85 [Cl] q / [HC1] q ratio. Both 30 and 4 cm lenses were used in these experiments, and [C1]q/[HC1]q was (0.6±0,2) throughout the range 10-100 J cm-2. The lack of a strong dependence of this ratio on laser fluence is significant, and this is discussed in the next section. DISCUSSION Our experiments confirm the presence of significant quantities of both Cl and HC1 following the 4* IRMPD of TCE. The nascent HC1 derives from reaction (2a), and the Cl can derive from either the subsequent IRMPD of C2C12 via reaction (4), or the reaction of TCE via reaction (2b): C2HC13 ■ + ■ C2C12 + HC1. AH = 20 kcal mol” 1 26 (2a) Ea = 56 kcal mol-1 2® C2HC13 -► C2HC12 + Cl . AH = 85 kcal mol" 1 14'33 (2b) C2C12 +C2C1 + Cl AH = 100 kcal mol" 1 34 (4) where the AHfs refer to ground state reagents and products. Despite the favorable frequency factor of reaction ( 2b) relative to (2a), the large difference in the activation energies for these reactions should cause reaction ( 2a) to be favored except at very high levels 86 of excitation. This is evident in Fig. (III-4), which shows the results of RRKM calculations for reactions (2a) and (2b).Nevertheless, it has been suggested that Cl may derive from reaction (2b).6 In the calculations shown in Fig. (III-4), transition state parameters were obtained as follows. Firstly, following the work of Kato et al.,*® we noted the changes of the normal mode frequencies in going from the equilibrium configuration to the transition state for the case of HF elimination from vinyl fluoride. Having identified the corresponding normal modes of TCE, we made the fractional changes of the normal mode frequencies the same in both systems. Secondly, following the work of Q Reiser et al. for the case of HC1 elimination from vinyl chloride, we again made the fractional changes of the normal mode frequencies the same for vinyl chloride and TCE. Finally, for the case of simple bond fission, we estimated transition state parameters following the usual prescriptions.^ The various molecular parameters used in the calculations are listed in Table (III-l). The RRKM rates span a wide range, and we believe that for the tight transition complexes this range reflects the extremes. Figure (III-4) also shows the ratios of the rates for chlorine atom bond fission and HC1 87 FIGURE (III-4): RRKM calculations for molecular elimination and bond fission (reactions (2a) and (2b) respectively). For molecular elimination, several distinct transition states were considered. Curves A, B, and ,C were computed as per refs. 10, 9, and 37 respectively (see text for details). The ratios (dashed curves) of the elimination and bond fission rate coefficients are read from the right hand vertical axis, and the parameters used in the calculations are given in Table (III-l). I / ) 2 o: q: 12 11 1 0 <D 2 9 8 c 2hci 3 C2CI 2 + HCI c 2ci2+hci c 2hci 2 + c i I I 1 c 2hci 2 + c i J I 1 - - 1 I I 1 C O ! £ > 80 100 120 140 160 ENERGY ( k c a l mol’1 ) 10 0.1 0.01 u i + _t\J U o' o + _ J \ i o <M o 0.001 TABLE (III-l): Parameters used in RRKM calculations. 9 0 . Table (III-l). Parameters used in RRKM calculations. c2hci3 PARENT17 TRANSITION STATES Cl bond fission t HCI elimination After ref. 9 After ref. 10 1 (a1) 3080 Vibrational frequencies (cm-1) 3080 2526 3173 2 (a') 1590 1590 1780 1572 3 (a1) 1250 1250 1250 757 4 (a*) 850 638 816 893 5 (a') 630 R.C. 621 364 6 <a'> 452 452 727 358 7 (a*) 381 285 R.C. 169 8«a') o Ca1) 274 206 205 548 165 106 165 113 iota'1) 933 933 765 416 U< a“) 783 783 783 606 l2 (a") 2 1 2 206 2 1 2 R.C. $ 163,320,471 Moments of inertia 212,1074,1387 (amu A^) 273,336,531 molecular elimination. These curves show that only at quite high levels of parent excitation can these two reaction channels become competitive, regardless of the choice of transition state parameters. The calculations shown in Fig. (III-4) suggest that energies > 1 2 0 kcal mol” 1 are required in order to make reactions (2a) and (2b) competitive. This is 64 kcal mol” 1 above the 56 kcal mol” 1 activation energy for the molecular elimination channel, and corresponds to dissociation rates of ~ 1011 s”1. The calculations also suggest that changes in the dissociation rates brought about by changing the laser intensity, and therefore the optical pumping rate, will cause a measurable change in [Cl]q/[HCI]q. For example, a change in the dissociation rate of a factor of 10 corresponds to a change of approximately two orders of magnitude in [Cl]q/[HCI]g, and this should be easy to detect. In fact, we detect no systematic change in that ratio as the laser fluence, and therefore average intensity, is changed by a factor of 10 ([Cl] 0/[HCl] o = (0,6±0.2) , 10-100 J cm”2). Although laser pulse shape effects will lessen the dependence of [C1]q/[HC1]q on fluence, it would be quite surprising to see no dependence at all. Thus, we conclude that the chlorine atoms which we detect are not 92 produced by reaction (2b), but by reaction (4) instead, with a contribution from the subsequent photolysis of C2C1 (see below). The very efficient dissociation of C2CI2 derives from the excellent coincidence between its ^3 vibrational mode and the C02 laser frequency,^® and the high vibrational energy content of the nascent C2C12 formed via reaction ( 2a), regardless of whether a, a and/or a , 6 elimination is involved. In the case of a, a elimination, the dichlorovinylidene product lies ~ 40 kcal mol“* above ground state dichloroacetylene, and subsequent isomerization guarantees a very high * vibrational energy content indeed.40 Dissociation of CgCl It is well documented that the IRMPD of TCE proceeds sequentially until C2 molecules are formed in the X^-Eg and low lying a ^ u states, under collision free conditions.4^ - In fact, ethylene and many of its substituted derivatives yield C2 molecules via collision free IRMPD, and the phenomenon seems quite general for A O this class of precursors. A This is a bit curious, since often the immediate C2 precursor must be triatomic if the low energy reaction pathways are to be followed. 93 and dissociation of a triatomic molecule via IRMPD is considered a rather unlikely process. Recent experiments by Carrick et al.^ have confirmed calculations by Shih et al.^ in which the A^A" state of C2H was predicted to be only ~ 0.5 eV above the X^E* ground state. Carrick et al.^ found y that the A^A" state was roughly 3770 cm“^ above the ground state, and this amount of product excitation is available via IRMPD. As predicted, the spectrum is perturbed strongly, indicating a large amount of 9 9 4 * coupling between A A" and X Eg. Also, in our own work, we have monitored reactions of C2H(A A") produced by the IRMPD of selected precursors, thereby establishing its production via this excitation process.45,46 Tjje strong 9 4 * 9 coupling of the X e and A A" states means that the system has degrees of freedom in addition to those of'a triatomic molecule on a single electronic potential surface. The state density can increase significantly compared to the case of a single electronic state, and this may explain the dissociation of C2H during the sequential IRMPD of C2H4 and other similar C2 precursors. Since the increase in state density only 9 occurs at energies > the A A" origin, only those C2H molecules which acquire this energy, either via optical 94 pumping or their nascent excitation, are viable candidates for dissociation via IRMPD. Although the statistics of the situation prevents most of the C2H molecules from being dissociated, a very measurable quantity of C2 molecules can be produced in this manner. Since the A^A" state derives from the promotion of a f electron to a o ' orbital localized near the terminal carbon atom, it follows that species of the general form C2X will have low lying electronic states analogous to the A^A" state of C2H. Thus, the mechanism described above is general, and it follows that C2 can be produced via the IRMPD of species such as C2CI, C2CN, C2Br, etc. In general, dissociation of these heavier species will be more efficient than C2H, but as before, we expect that the majority of these species will not be dissociated via IRMPD, even in the presence of a strong IR electromagnetic field. CONCLUSIONS The first step in the IRMPD driven unimolecular decomposition of C2HC1 3 is the molecular elimination of HCI via concerted 3 and/or 4 center processes. Subsequent IRMPD of the nascent vibrationally excited C2CI2 product produces atomic chlorine and is the main source of this species under the present experimental conditions. A low lying electronic state of C2Clf of symmetry ^A", facilitates the dissociation of this species via IRMPD, thereby accounting for the production of C2 molecules. However, the amount of atomic chlorine produced thusly is expected to be small compared to the atomic chlorine which derives from C2CI2 dissociation. 96 REFERENCES 1. A. Ben-Shaulr Y. Haas, K.L. Kompa, and R.D. Levine, in "Lasers and Chemical Change," Springer Verlag Series in Chemical Physics 10, (1981), and references cited therein, 2. C.R. Quick, Jr. and C, Wittig, J, Chem, Phys, 69, 4201 (1978). 3. G.A. West, R.E. Weston. Jr., and G.W. Flynn, Chem, Phys. 35, 275 (1978). 4. C.R. Quick, Jr. and C. Wittig, J. Chem. Phys. 72, 1694 (1980). 5. A.J. Colussi, S.W. Benson, R.J. Hwang, and J.J. Tiee, Chem. Phys. Lett. 52, 349 (1977) . 6 . A.S. Sudbo. P.A. Schulz, E.R. Grant, Y.R. Shen, and Y.T. Lee, J. Chem. Phys. 6 8 , 1306 (1978). 7. J.B. Marling, I.P. Herman, and S.J. Thomas, J. Chem. Phys. 72, 5603 (1980). 8 . K. Nagai and M. Katayama, Bull. Chem. Soc. Japan 51, 1269 (1978). 9. C. Reiser, F.M. Lussier, C.C. Jensen, and J.I. Steinfeld, J. Am. Chem. Soc. 101, 350 (1979), 10. S. Kato and K. Morokuma, J. Chem. Phys. 74, 6285 (1981). 11. R.J.S. Morrison, R.F. Loring, R.L. Farley, and E.R. Grant, J. Chem. Phys. 75, 148 (1981). 12. B.E. Holmes and D*W. Setser, J. Phys. Chem. 82, 2461 (1978). 13. M.J. Berry, J. Chem. Phys. 61, 3114 (1974). 14. F. Zabel, Ber, Bunsenges. 78, 232 (1974). 15. E. Wurzberg, L.J. Kovalenko, and P.L. Houston, Chem. Phys. Lett. 35, 317 (1978). 16. A.B. Horwitz, J.M* Preses, R.E. Weston, Jr., and G.W. Flynn, J. Chem. Phys. 74, 5003 (1981). 17. H.J. Berstein, Can. J. Res. 28B, 132 (1950). 18. J.M. Herbelin and G. Emannuel, J. Chem. Phys. 60, 689 (1974). 19. In these measurements, we retain only the peak fluorescence signal intensity, which occurs at short times and which is referred to as HCl^ (t=0). Since x rjLSe << Tfall' the Peak of the temporal curve remains proportional to the species concentration as the experimental parameters are varied. 20. Rotational excitation of the diatomic photoproduct was noted by monitoring the effect of He buffer gas _____________________________________________________________ 98 on the IR signal transmitted through the HCI gas filter. There was no measurable effect with up to 10 Torr of He. The use of a gas filter allows us to conveniently measure the signal intensities, I, with and without HCI in the filter, and it is possible to relate those intensities to spectroscopic parameters of the diatomic:^ vibrational level, v v^v-l *s t^ie ^requency of the v->v-l transition, and Rv^v_i is the corresponding transition moment. With TCE pressures of 40 - 100 that the left hand side of the above expression was in the range of 0.9 - 1.5. Using the obtained for the right hand side. Thus, we conclude that HCI rotational excitation is not very important in our measurements. O Q ^Cl in cell E yv[HCl(v)] 2 YI[HCl(v=l)] ? ■'■empty cell IHC1 in cell v-*v-l 2 where [HCl(v)] is the population of the. vfc^ HCI — 9 mTorr and fluences of ~ 40 J cm , we found I - 3 populations given by Berry, a value of 1.8 is 99 21. The quenching of HCl(v=l) by He is not very effective (6x1 0“^ cm^ molec“1s“- 1 -) .22 22. R.V. Steele, Jr. and C.B, Moore, J. Chem. Phys. 60, 2794 (1974). 23. K. Bergmann and C.B. Moore, J. Chem. Phys. , 63, 643 (1975). 24. The vibrational energy of the HCI produced by the IRMPD of TCE is quite different from that produced 1 ^ by reaction (5). Berry reports that only 0.2-0.3 of the barrier energy appears as HCI vibration, while > 80% of the HCI produced by reaction (5) 1 25 is m v=l. J 25. T. Carrington and J.C. Polanyi, "Chemical Kinetics," in MTP International Review of Science, Physical Chemistry Series I. Vol- 9, Ch. 5 (A.D. Buckingham and J.C. Polanyi, Eds.). 26. Obtained using "AHf(C2Cl2 ) = 40 kcal mol-1 from Ref. 27. 27. A.I. Vitiitskii, Zh. Org. Kh. 3, 1354 (1967). 28. To estimate the activation energy of reaction (2a), we used the thermochemical data available for the four center addition of hydrogen chloride to 2 9 acetylene and trichloroethylene. Benson et al. , give 36.4 kcal mol-' * ' for the former while ^ 0 — 1 Tschuikow-Roux et al. give 36 kcal mol for the latter. Thus, we take 36 kcal mol-- * - for the addition to dichloroacetylene. and taking into account the endothermicity of reaction (2a), the energy of activation for that process is approximately 56 kcal mol-- * - . This value is higher 31 than the one deduced from the pyrolysis of TCE, but far below the value obtained from shock tube experiments for Cl elimination,3^ and close to the one reported for HCI production from vinyl chloride.32 29. S.W. Benson and G.R. Haugen, J. Phys. Chem 70, 3336 (1966). 30. E. Tschuikow-Roux and K.R. Maltman, Int. J. Chem. Kin. 7, 363 (1975). 31. A.M. Goodall and K.E. Hoowlett, J. Chem. Soc., 2599 (1954). 32. F. Zabel, Int. J. Chem, Kin. 9, 651 (1977). 33. W.R.J. Tyerman, Trans. Far. Soc. 65, 2948 (1969). 34* K. Evans. R. Scheps. S.A. Rice, and D. Heller, J. Chem. Soc. Faraday II 69, 856 (1973). 35. Rate constants for the molecular elimination and bond fission reactions of C2HC13 were calculated using the Bunker/Hase computer 10 1 program.36 All rotations were treated as inactiver vibrations were treated as harmonic, and no centrifugal corrections were made. 36. W.L. Hase and D.L. Bunker, Program No. 234, Quantum Chemistry Program Exchange, Univesity of Indiana, Bloomington, Indiana, 47401, USA. 37. We have chosen the transition state geometry by extending the C-Cl bond to three times its normal O O length, ° and changing the frequencies according to the ones most affected by replacing the Cl atom with a Br or I atom. 38. S.W. Benson, "Thermochemical Kinetics" (Wiley Interscience New York, 1976) . 39. E. Kloster-Jensen, J. Am. Chem. Soc. 91, 5674 (1969). 40a. M.J. Frisch, R. Krishanan. J.A. Pople, and P.V.R. Schleyer, Chem. Phys. Lett. 81, 421 (1981). b. C. Reiser and J.I. Steinfeld, J. Phys. Chem. 84, 680 (1980). 41. M.S. Mangir, H. Reisler, and C. Wittig, J. Chem. Phys. 73, 2280 (1980). 42. H. Reisler, M. Mangir, and C. Wittig, in Chemical and Biochemical Applications of Lasers, Vol. V, C.B. Moore, ed., Academic Press, 1980. 10 2 43. P.G. Carrick, J. Pfeiffer, R.F. Curl, E. Koester, F.K. Tittel, and J.V.V. Kasper, J. Chem. Phys. 76, 3336 1982. 44. S.K. Shih, S.D. Peyerimhoff, and R.J. Buenker, J. Mol. Spect. 74, 124 (1979). 45. A.M. Renlund, F. Shokoohi, H. Reisler, and C. Wittig, J. Phys. Chem. 8 6 , 4165 (1982). 46. F. Shokoohi, T. Watson, A.M. Renlund, H. Reisler, and C. Wittig, unpublished. 10 3 CHAPTER IV PRODUCT VIBRATIONAL ENERGY CONTENT IN IRMPD: THREE AND FOUR CENTER ELIMINATION OF DF AND HF FROM CH2CDF INTRODUCTION In the past decade, there has been sustained and significant interest in the unimolecular processes available to olefins and halogenated counterparts, when energized using different methods of excitation. These elementary processes include isomerization,1 molecular elimination,2 and simple bond fission,3 and the excitation techniques include chemical activation,^ internal conversion, and infrared multiple photon excitation (IRMPE). Part of the motivation for these studies is a desire to understand dissociation pathways and mechanisms, as well as energy partitioning and flow in the product and parent species, and one such elementary process is hydrogen halid elimination from 10 4 different halogenated ethanes and ethylenes. While molecular elimination of HX(X=C1, F, Br) is well established in the infrared laser photodissociation of different halogenated ethylenes,5 in several cases the dissociation pathways are not clear.5_^ In particular, very little information has been published concerning the transition states (TSfs) that lead to molecular elimination (e.g. does reaction involve inserted three or four center TS's, or different intermediate steps). O For the molecule CH2CDCI, Reisler et al. used infrared and GC/MS final product analyses to show that the main dissociation channel following IRMPE is three center elimination of deuterium chloride, accounting for ~ 70% of the dissociation. Recently, Kato et al.9 performed ab initio calculations to obtain the potential energy surface germane to the unimolecular reaction of CH2CHF (VF), and concluded that the three center elimination is favored slightly over the four center reaction (60% vs. 40%). These results are contrary to those from previous studies using ab initio calculations10 or chemical activation experimental 11 17 techniques r m which the four center channel was the more popular culprit. 10 5 This chapter is devoted to an experimental study of HF and DF elimination from CE^CDF (VFd^) upon the irradiation of static gaseous samples with the mildly focused output from a CO2 TEA laser. By observing the time resolved infrared emission from the vibrationally excited photofragments, we are able to gain insight into the three and four center elimination processes involved in the IRMPE of VFd1# EXPERIMENTAL The system was pumped to 10”^ Torr before introducing static samples. The laser was located within a screened enclosure to avoid electrical transients and the signal was amplified, digitized, and averaged to obtain reasonable S/N. VFd^ (PCR, 98.6%) was degassed at 77° K and used with no other purification. A 30 cm focal length lens was used in all experiments. 106 RESULTS Fig. (IV-la) shows the absorption spectrum of VFdj in the region 850-1100 cm”' * ' . The vibrational mode most accessible to the CO2 laser has been assigned to the C-D deformation,1 3 '14 and the R branch of this , vibration overlaps with the (001) - (100) R(20) R(14) lines of the C02 laser. The results are not sensitive to changes between different C02 laser lines, and all measurements were made using R(16), since this provided the maximum output energy and fluence (1.2 J, 27 J cm"^). Figure (IV-lb) shows the infrared absorption spectrum of VFd^ in the region 3600-2200 cm**1, together with the 5% transmission points of the interference filters used in the experiments. As with VF^, spontaneous emission is detected from vibrationally excited HF molecules, hereafter referred as HF+, upon irradiation of VFdj with the focused output from the C02 TEA laser. We monitored the region 3700-3500 cm""1, since this region is not overlapped by emissions from either the parent or fragments (see Fig. (IV-lb)). 10 7 Figure (IV-1). a) Infrared absorption spectrum (6 Torr, 10 cm) of gaseous VFd^ in the 1100-850 cm“l region. The arrow indicates the position of the (OOl)-(lOO) R(14) C0 2 laser line. b) Infrared absorption spectrum of VFd-^ (120 Torr, 10 cm) in the 3600-2200 cm" 1 region. Also shown (near the bottom) are the 5% transmission points of the filters used to monitor the fluorescing species. The vertical arrows, from left to right, show the location of the band centers of prominent features of anticipated photoproducts: C-H stretch of C2HD, C-H stretch of C2H2, and C-D stretch of c2hd. 10 3 100 90 5 2 BO 60 ° 50 cc bJ HF IR EMISSION DF IR EMISSION ! “ <. - * — ~ ->H REGION 4 0 REGION 1100 1000 9 0 0 3 4 0 0 3 0 0 0 ENERGY (cm"1) 2600 It is important in all experiments that deal with unimolecular reactions induced by IRMPE, that the results are not influenced by collisions. As the time resolution of our experiments is approximately 2 us, and the pressure around 50 mTorr, we need experimental evidence concerning the role of collisions. If collisions are not important, a plot of signal vs. pressure should be linear. We have checked this, and find that the variation of the HF^ signal with VFd^ pressure is quite linear below 50 mTorr. We thus used 50 - 70 mTorr of parent as a compromise between reasonable S/N and avoiding collisional effects. In many respects, the behavior of VFd^ is similar ' C to its non-deuterated analog. The addition of He enhances the HF^ signal by a factor of ~ 4, and this enhancement remains constant above 20 Torr. The dependence of the HF^ emission signal on laser fluence — 9 shows an apparent threshold of 6 J cm , and the dependence is linear up to the maximum fluence available (~ 30 J cm”2). This higher threshold relative to VF may be due to differences in the pumped vibration and its absorption coefficients.13-15 We searched for infrared emission in that part of the spectral region where DF^ emission occurs (see 110 Fig. (IV-lb)), and Fig. (IV-2) shows a typical signal. This signal behaves in the same way as the HF^ signal: linear with parent pressure up to 50-80 mTorr, signal enhancement with He pressure up to 20 Torr, and a threshold of ~ 7 J cm . In an attempt to assign this signal to DF’ * ’ , we used DF gas filters, with pressures ranging from 200-600 Torr, located between the IR detector and the viewing window. Signals such as the one shown in Fig. (IV-2) were unaffected by the gas filter, using either neat VFd^ or a mixture containing several Torr of He--*-® Since our technique is not sensitive to DF in the ground vibrational state, we can only infer that if DF is produced in the decomposition of VFd^, it must be in the ground vibrational state. No strong emissions other than those discussed above were detected in the spectral range of our infrared detector. DISCUSSION Our results show the presence of HF>t and an unknown emitter at the DF ^ emission region upon VFdi dissociation. The following primary steps has been 111 Figure (IV-2). Typical fluorescence signal in the region 2700-3050 cm-- * - obtained with 53 mTorr of CH2CDF and 30 J cm"^, The risetime is detector limited. 112 IR FLUORESCENCE INTENSITY (nrh. united proposed for the decomposition of the non-deuterated compound:® CH2CHF + CH2CHF (isomerization) (1) CH^CF (H migration) (2) CHCH2F ->CHCH + HF (H migration + a, a elimination) (3) - * CHCH + HF (ot, 3 elimination) (4) CH2C + HF CHCH + HF (a, a elimination + H migration) (5) In an effort to determine the mechanism of the IRMPD of VFdlf it is useful to peruse a partial energy level diagram showing several ground electronic state energies for the non-deuterated compound (Fig. (IV-3)). Isotope substitution is believed to have only a slight 9 9 effect on the thermochemistry and therefore we will use information obtained for the more well characterized VF system, in order to discuss possible reaction pathways in the case of VFd^. It is easily seen that the direct three or four center diatomic elimination are the lowest thermodynamical decomposition channels and that they are likely to proceed in a competitive manner.^ We feel that the only process that may obscure a clear cut to assign a direct elimination via a three or 114 Figure (IV-3). Relevant ground electronic singlet and transition states expected in direct three and four center HF elimination from VF taken from Ref. 9 (full line). Also shown is the transition state for the migration/elimination process (------ ). For completion (see text) are shown: a) the experimental activation energy of HF elimination from VF after Simmie et al.^® (.....) and Cadman et al.l® (OCKHJ-), b) calculated singlet vinylidene structure energy states and the barrier for rearrangement after Krishnan et al.^® (— — — —) , c) calculated singled and triplet vinylidene structure energy states and the barrier for rearrangement in the 9 1 singlet case after Osamura et al. ( ) . 115 E (kcal mol 100 50 0 C2H2 + HF FOUR CENTER ELIMINATION CH2 CHF H H THREE CENTER ELIMINATION four center of the hydrogen halide is the possible deuterium migration followed by the elimination step. We believe that this migration can be ruled out due to the height of the barrier for that process compared with the direct elimination ones. Even where our particular excitation technique can provide between 15-20 kcal mol”1 of excess energy over the barrier for elimination,24 the migration/elimination process will still need ~40 kcal mol”1 to be feasible. Dissociation from triplet transition states into singlet products states are spin-forbidden, and early results by Tsumashima et al.25 have been re-evaluated using similar and different techniques2® indicating that dissociation of VP proceeds through ground electronic states. Thus, we may conclude that the main mechanism that leads to the photodissociation of VFdj is the direct three center (DF) or four center (HF) molecular elimination from the singlet ground state in a competitive way. To see the importance of each channel in the overall dissociation process, we performed RRKM calculations of VFd-^ for the direct DF or HF elimination.2^ 117 Figure (IV-4) shows that each channel accounts for roughly half of the total dissociation rates as in the case of VF. Despite the fact that these calculated rates agree reasonably well with the work of Reisler et al.^ on VCld^, it is worthy to point out that these absolute rate calculations are sensitive to the chosen TS parameters for the elimination process. To show that, we have calculated the relation between the rates for DF and HF production while changing the energy of activation of those processes within the limits error quoted in Ref. 9. Fig. (IV-5) depicts those results clearly indicating that it is possible to obtain a wide range of relations, i.e., it is possible to obtain that any process can become more O Q important than the other. Several s t u d i e s ^ ® SUggest, based on decomposition of vibrationally excited fluorinated ethanes by radical recombination, that direct three center elimination comprised of 30% of the total Q decomposition while the previous quoted work on VCld^ showed that the channel gives ^70% of the total dissociation, giving grounds to our previous calculated results of Fig. (IV-4). 118 FIGURE (IV-4): Calculated RRKM rate constant for direct HF (lower trace) and DF elimination (upper trace) from VFd1# 119 DF elimination HF elimination 8 0 110 9 0 100 FIGURE (IV-5): k(three center)/k(four center) ratio varying the energy of activation of each process. Full line: Ea f0ur center = 84 kcal mol"1 and Ea three center = 70 to 90 kcal mol"1; dotted lines: Ea four center = 77 kcal mol"1 and Ea three center = 70 to 90 kcal mol"1. The upper arrow on the left indicates the experimental value obtained in Ref. 8, the middle one is calculated from Ref. 9 and the lower one was obtained experimentally in Ref. 12. Ul O + CM X, CM O f Uu O U CM X (J Ix. X 4* Q X, CM C J 8 0 70 9 0 100 .120— 130 E (kcal mor1 ) CM X u K) K> Besides confirming the presence of photoproducts, our experimental data is closely related to the vibrational energy content of these photoproducts. This information is interesting to analyse under the point of view of how the absorbed energy is disposed into the parent and photofragments. Table (IV-1) summarizes the expected energy content of the different products following direct three and four center elimination of HF from VF according to Ref. 9. The main distinction between the two processes lies in the fact that in the three center one the major proportion of the reverse barrier is going to be located in the acetylene due to the rearrangement of the vinylidene structure (see Fig. (IV-3)). The latter process is believed to take place in less than 10”12 s21,29,31 before ft has the chance of undergoing other chemical process. That leaves the diatomic with an average vibrational energy cntent four times less thatn the one expected for the four center mechanism. The strong HF^ emission and the lack of DF signal that we have seen are in qualitative agreement with that calculation. The DF may be produced vibrationally cold, at least to the extent of our technique sensibility opposite to the HF, that is vibrationally hot. The HF 123 TABLE (IV-1): Product energy content in VF decomposition, after Ref. 9 ( in kcal mol-1). 124 FOUR CENTER THREE CENTER HF Vibration 11.5 3.0 CHCH Vibration 24.8 41.0 Translation 21.0 12.1 125 vibrational energy content was experimentally obtained previously^ and agrees quantitatively with the one calculated for the four center process. Setser et al.^f30 in a systematic work of three vs, four center dehydrohalogenation of a series of substituted ethane quoted the importance of 1,1 dihalogen substitution for the three center process to be in competition with the four center one. Our results and the experimental and calculated ones of others on 09 in 7 f t i n some halogen ethanesJ‘ 6,AW and ethylenes' could say that those processes are more a general rule than an exception in any single or poly substituted methane, ethene, and ethane. It is interesting to study in particular the three center reaction as the DF is expected vibrationally cold and subsequently the acetylene upon rearrangement from the vinylidene structure may be vibrationally very hot (see Table (IV-1) and Fig. (IV-3)). We are going to discuss the rearrangement (vinylidene-acetylene) in an attempt to give an interpretation to the 3.5 11m IR signal, that is not due to DF+. Fig. (IV-3) shows the low-lying states for the diradical structure. It is readily seen that the 12 6 and ^B2 are the possible states reachables by our excitation method.24 The rearranges very fast to give acetylene *5 O O but the B2 has a longer decay time that could lead to its observation via its infrared emission spectrum. In either case the infrared features of a very hot acetylene obtained from the singlet, and the triplet vinylidene may give ground to our 3.Slum IR signal. A vibrationally excited acetylene may have some of this energy in its C-H stretches that in emission could show some red-shift from its emission region at 3.1ym (see Fig. (IV-lb)). For vinylidene the same features were calculated to be in the region 3-3.lym.2^-'22 As those calculations are believed to give results 10-15% over Q Of] Ol the experimental ones ' ' we can expect emission at 3.3-3.5ym if those species are born vibrationally excited. A very rough estimation"^4 of the energy disposal in the C-H stretches of a hot acetylene shows that £10 kcal mol-1 are within them, i.e., £15% of' the composite total energy to be located in acetylene vibrations (see Table (IV-1)). Given a statistical partition of the vibrational energy on the products, 127 this result agrees with the energy content of C-H stretches of acetylene expected in the decomposition of vinyl fluoride.® CONCLUSIONS We have induced the unimolecular decomposition of VFdj by means of the focused output of a C02 TEA laser. Analysis of the infrared fluorescence of photoproducts, thermochemical information and calculations using an RRKM model, show that the main dissociation channel is the direct diatomic elimination. Direct three center dissoication (DF) and four center dissociation (HF) are going to proceed in a competitive way. The diatomic is produced in a collisionless manner and vibrationally excited from the four center as compared to the three center. We quoted a rough estimation for an upper limit for the C-H stretch energy content of the different olefin photoproducts. These conclusions are in qualitative agreement O with similar deuterated compounds together with calculations on the normal compound.9 128 REFERENCES 1. z. Karny and R.N. Zare, Chem. Phys. 23, 321 (1977). 2. A.S. Sudbor P.A. Schultz, Y.R. Shen, and Y.T. Lee, J. Chem. Phys. 69, 2312 (1978). 3. A.S. Sudbo, P.A. Schultz, E.R. Grant, Y.R. Shen, and Y.T. Lee, J. Chem. Phys. 70, 912 (1979). 4. E. Tshchuikow-Roux and S. Kodama, J. Chem. Phys. 50, 5297 (1969). 5. C.R. Quick, Jr. and C. Wittig, J. Chem. Phys. 72, 1694 (1980). 6. C.R. Quick, Jr. and C. Wittig, Chem. Phys. 32, 75 (1978). 7. F.M* Lussier, J.I. Steinfeld, and T.F. Deutsch, Chem. Phys. Lett. 58, 277 (1978). 8. C. Reiser, F.M. Lussier, C.C. Jensen, and J.I. Steinfeld, J. Am. Chem. Soc, 101, 350 (1979). 9. S. Kato and K. Morokuma, J. Chem. Phys. 74, 6298 (1981) . 10. S. Kato and K. Morokuma, J. Phys. Chem. 73, 3901 (1980) . 12 9 11. K.C. Kim and D.W. Setser, J. Phys. Chem. 78, 2166 (1974). 12. K.C. Kim, D.W. Setser, and B.E. Holmes, J.'Phys. Chem. 77, .725 (1973). 13. P. Torkington and H.W. Thompson, Trans. Faraday Soc. 41, 236 (1945). 14. B.Bak and D. Christensen, Spect. Acta 12, 355 (1958). 15. R. Elst and A. Oskam, J. Mol. Spectros. 39, 371 (1971) . 16. These results are at odds with the ones recently published for VF,1^ but working near dissociation threshold. 17. J.F. Caballero and C. Wittig, Chem. Phys. Lett. 82, 63 (1981). 18. J.M. Simmie, W.J. Quiring and E. Tschuikow-Roux, J. Phys. Chem. 74, 992 (1970). 19. P. Cadman and W.J. Engelbrecht, Chem. Comm. 453 (1970). 20. R. Krishnan, M,J. Frisch, J.A. Pople, and P.V.R. Schleyer, Chem. Phys. Lett. 79, 408 (1981). 21. Yoshimiro Osamura, H.F. Schaefer III, S.K. Gray, and W.H, Miller, J. Am. Chem. Soc. 103, 1904 (1981). 22. 23. 24. 25. 26 a. b. 27. S.W. Benson in "Thermochemical Kinetics," 2nd edition, Wiley, New York, (1976), Product analysis of IR laser photodissociation of r * i Q VF agrees with the ones obtained m thermal and I Q shock-tube decomposition that proceed through the lowest thermochemical channel, diatomic elimination. C.R. Quick, Jr. and C. Wittig, J. Chem. Phys. 69, 4201 (1978). S. Tsumashina, H.E. Gunning, and O.P. Strausz, J. Am. Chem. Soc. 98, 1690 (1976). E. Tschuikow-Roux and S. Kodama, J. Chem. Phys. 50, 5297 (1969). H.F. Hunzicker, J. Chem. Phys. 50, 1288 (1969). We performed standard RRKM calculations on VFdj O O using the Bunker-Hase method. We have used the relative changes as in Ref. 9 to find the vibrational frequencies of the transition state from the molecular frequencies of VFd-^. Molecular frequencies of VFdj (in cm_1):^f1^ 3115, 2310, 3150, 1628, 985, 1362, 791, 477, 683, 867, 902, 1165. Vibrational frequencies for three center transition state (in cm-- 1): 3445, 2512, 2491, 131 1776, 1101, 1036, 661, 289, 1031, 978, 507. Ea = 80 kcal/mol. Vibrational frequencies for four center transition state (in cm”^): 3635, 2655, 1949, 1845, 814, 869, 669, 467, 1127 . 897, 762. Ea = 82 kcal mol-- * - 28. W.L. Hase and D.L. Bunker, Program 23 4 Quantum Chemistry Program Exchange, University of Indiana, Bloomington, Indiana, 47401. 29. Similar results were obtained changing the transition states vibrational frequencies while keeping constant the energies of activation. 30. B.E. Holmes, D.W. Setser, and G.O. Pritchard, Int. J. Chem. Kinet. 8, 215 (1976). 31. C. Reiser and J. Steinfeld, J. Phys. Chem. 84, 681 (1980). 32. G.E. Millward and E. Tschuikow-Roux, Int. J. Chem. Kinet. 5, 363 (1973). 33. Y. Osamura and H.F* Schaefer III, Chem. Phys. Lett. 79, 412 (1981). 34. This calculation was performed by a comparison of the infrared signals of HF+ and the 3.45 p m signal assuming that the latter is due to the C-H stretches of C2H2 and C2HD. Taking into account D 132 vs. \ of the detector, filters transmissions and Einstein coefficients of the emitters,35 we estimate the C-H stretch energy content with the ,1, known energy of the HF product. 35. L. Gribov and V.S. Smirnov, Sov. Phys. Usp. 4, 919 (1962). 133 SELECTED BIBLIOGRAPHY 1. Benson, S.W. , Thermochemical Kinetics (Wiley, New York, 1976). 2. Calvert, J.G. and Pitts, J.N., Photochemistry, (John Wiley 8s Sons, Inc., New York, 1966). 3. Cantrell, C.D., edi, Multiple Photon Excitation and Dissociation of Polyatomic Molecules (Springer- Verlag, Heilderberg, Berlin, 1981). 4. Jortner, J., Photoselective Chemistry (Wiley, New York, 1981). 5. Moore, C.B., ed., Chemical and Biochemical Applications of Lasers (Academic Press, New York, 1980). 6. Robinson, P.J. and Holbrook, K.A., Unimolecular Reactions (Wiley, New York, 1972). 7. Shaul, A.B., Haas, Y., Kompa, K.L., and Levine, R.D., Lasers and Chemical Change (Springer-Verlag,Ed., New York, 1981). 8. Smith, I.W., Kinetics and Dynamics of Elementary Gas Reactions (Butterworths, London, 1980). 134

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

CO2 Laser-induced chemistry of C23CN

PDF

Application of nonlinear spectroscopic techniques to the studies of molecular electronic states and reaction dynamics.

PDF

Configuration interaction calculations on the resonance states of HF- the excited states of HF, and the Feschbach states of HF-

Configuration interaction calculations on the resonance states of HCl<super>-</super> and the excited states of cyclopropane

PDF

Configuration interaction calculations on the resonance states of HCl- and the excited states of cyclopropane

PDF

Brillouin scattering in molecular crystals: Sym-trichlorobenzene

PDF

Ab initio calculations of vibrational absorption and vibrational circular dichroism

PDF

A study of the double sulphates of some rare-earth elements with sodium, potassium, ammonium and thallium

PDF

A classical mechanical method of analysis and periodic orbit search: Applications to HCN/HNC, C1HC1, FH2, and their deuterates

PDF

Collision-induced dissociation of Nitrogen dioxide and photofragmentation of Nitrogen dioxide and cyanic acid

PDF

Carbocations, carboxonium dications and their role in superacid catalyzed reactions

$An investigation of the system sodium stearate-water by means of x-ray diffraction and differential thermal analysis$

PDF

An investigation of the system sodium stearate-water by means of x-ray diffraction and differential thermal analysis

PDF

A study of the interactions of polar gases with solid proteins and some simple organic compounds

PDF

Basic studies in the analytical chemistry of beryllium

PDF

A highly efficient and general method for the chemical synthesis of beta,gamma-fluoro- and difluoromethylene analogs of purine and pyrimidine 5'-ribo- and deoxyribonucleotides

PDF

A study of the kinetics of the reaction of sulfenyl halides with olefins

PDF

A photophysical study of pyrene adsorbed onto silicas of variable surface area and porosity

PDF

A critical study of the hydrogenation of talloel

PDF

Crystallographic studies of metal hydrides and other transition metal complexes

PDF

Computer modeling and free energy calculatons of biological processes

PDF

Computer simulation of polar solvents with the SCAAS model

Asset Metadata

Creator Caballero, Julio Faustino (author)

Core Title Competitive and sequential absorption and dissociation mechanisms in the infrared fragmentation of polyatomics

Degree Doctor of Philosophy

Degree Program Chemistry

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag chemistry (physical chemistry),OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Witting, Curt (committee chair), [illegible] (committee member), Dows, D.A. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c17-637583

Unique identifier UC11354753

Identifier DP21892.pdf (filename),usctheses-c17-637583 (legacy record id)

Legacy Identifier DP21892.pdf

Dmrecord 637583

Document Type Dissertation

Rights Caballero, Julio Faustino

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA