ABSTRACT VIBRATIONAL SPECTRA OF DIAZOPROPYNE AND ITS PHOTOLYSIS PRODUCT: THE PROPARGYLENE RADICAL BY Frank K. Chi H Diazopropyne (H-CEC-C=N=N) and a partially deuterated isotopic mixture (H-CEC-CH=N=N, H-CEC-CD=N=N, D-CEC-CH=N=N and D-CEC-CD=N=N) were synthesized. Vibrational spectra have been obtained for diaZOprOpyne and its mixed deuterated analogs in the gaseous and solid states, and in matrix iso- lation. A valence force field normal coordinate analysis has been performed on the basis of CS symmetry, using molecular geometric parameters and initial force constants from related molecules, and refined to fit the infrared and Raman spectra of the isotopic Species. Information re— garding the electronic charge distribution of diazopropyne was provided by the final calculated force constants which suggest that diazopropyne has possible resonance structures, such as (H-CEC-CH=N=N <—> H—czc-EH-NEN <—> H-<'2°=C=CH-N_=.N). Photolysis of matrix-isolated diazoprOpyne gives rise to several new Spectral features, some attributable to the propargylene free radical. The existence of triplet H-é=C=é-H in the matrix is confirmed by ESR measurements. The normal coordinate analysis of C3H2, C3HD, and C3D2 has Frank K. Chi been carried out on the basis of Dooh symmetry (C00v for C3HD). The final calculated force constants give fre- quencies in excellent agreement with the matrix infrared spectra, and suggest that the skeletal C=C=C bonding situ- ation in C3H2 is similar to that of C3 and allene (C3H4). C3H2 has possible resonance structures (H—C=C=é-H ‘—> H-CEC-C;H <-> H-CLCEC-H) with outer orbital electron density distributed primarily at the end carbons of the radical. The C=C=C bending force constant of C3H2 is intermediate between those of C3 and C02, supporting the correlation between the occupancy of the fig orbital and the value of the central carbon bending force constant in molecules of this type. VIBRATIONAL SPECTRA OF DIAZOPROPYNE AND ITS PHOTOLYSIS PRODUCT: THE PROPARGYLENE RADICAL BY Frank K5 Chi A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1972 Q“) m To My Parents ii ACKNOWLEDGMENTS I wish to thank Professor George E. Leroi for his advice, interest and patience which have contributed greatly to this research and to my education. I also wish to thank Professor R. H. Schwendeman for his help in the microwave study and Professor J. F. Harrison for his helpful discussions regarding the propargylene radical. I would like to thank the members of the Molecular Spectroscopy group, especially, Glenn R. Elliott, and also Mrs. Naomi Hack, for their friendship. Financial support from Michigan State University and the Office of Naval Research is appreciated. Finally, I wish to thank my wife, May, for her patience and understanding. iii TABLE OF CONTENTS Page INTRODUCTION 0 o o o o o o o o o o o o o o o o o o o 1 CHAPTER I. THE INFRARED AND RAMAN SPECTRA OF DIAZOPROPYNE 4 Experimental . . . . . . . . . . . . . . . 5 Results and Discussion . . . . . . . . . . 20 Assignment of the Observed Fundamentals 42 Other Bands . . . . . . . . . . . . . . 59 An Outline of Normal Coordinate Analysis . 62 Vibrational Potential Function and Related Properties of Diazopropyne . . . . . . . 64 Normal Coordinate Analysis . . . . . . 64 Vibrational Potential Function of Diazopropyne (A') . . . . . . . . . . 75 Vibrational Potential Function of Diazopropyne (A") . . . . . . . . . . 79 Conclusion . . . . . . . . . . . . . . . . 82 II. INFRARED SPECTRUM OF THE PROPARGYLENE RADICAL 83 Experimental . . . . . . . . . . . . . . . 84 Results and Discussion . . . . . . . . . . 86 Assignment of C3H2 Fundamentals . . . . 92 Other Bands . . . . . . . . . . . . . . 101 Vibrational Potential Function and Related Pr0perties of Propargylene . . . . . . . 102 Normal Coordinate Analysis . . . . . . 102 Vibrational Potential Function of Propargylene . . . . . . . . . . . . . 107 Conclusion . . . . . . . . . . . . . . . . 110 REFERENCES 0 O O O C O O O O O O O O O O O O O O O O 1 1 1 LIST OF TABLES TABLE Page I. Observed infrared absorption bands of diazoprOpyne vapor (2980K) . . . . . . . 26 II. Observed infrared absorption bands of a mix- ture of gaseous diazopropyne and its deuterated molecules . . . . . . . . . . 27 III. Observed IR frequencies of matrix-isolated diazopropyne and its mixed deuterated molecules . . . . . . . . . . . . . . . . 43 IV. Observed Raman shift frequencies of solid C3H2N2 and its mixed deuterated molecules 45 V. Observed infrared frequencies of solid C3H2N2 and its mixed deuterated molecules . . . VI. Fundamental frequencies of diazopropyne . 60 VII. Geometry and symmetry coordinates for diazopropyne . . . . . . . . . . . . . . 66 VIII. Normal coordinate analysis of diazoprOpyne (C3H2N2) o o o o o o o o O o o o o o o o 70 IX. Normal coordinate analysis of diazopropyne (C3HDN2 ) o o o o o o o o o o o o o o o o 71 X. Normal coordinate analysis of diazopropyne (C3DHN2 ) o o o o o o o o o o o o o o o o 72 XI. Normal coordinate analysis of diazopropyne (C3D2N2) o o o o o o o o o o o o o o o o 73 XII. Normal coordinate analysis of diazopropyne (C3H2N15N) o o o o o o o o o o o u o o o 74 XIII. Valence force constants for diazopropyne . 80 LIST OF TABLES (Continued) TABLE Page XIV. Bands occurring upon photolysis of diazo- propyne which disappear upon annealing . 90 XV. Symmetry coordinates for the propargylene radical O O O O O O O O O O O O O O O O O 105 XVI. Normal coordinates analysis of the propargy- lene radical . . . . . . . . . . . . . . 106 vi 10. 11. 12. 13. 14. LIST OF FIGURES Andonian Associates Model D-307 liquid helium dewar . . . . . . . . . . . . . . Low temperature Raman cell . . . . . . . . Spray-on infrared cell for Malaker refrig- eratOI o o o o o o o o o o o o o o o o o Spray-on Raman cell for Malaker refrigerator Infrared Spectrum of gaseous diazoprOpyne . Infrared spectrum of a mixture of gaseous diazopropyne and its deuterated molecules Principle inertial axes of diazopropyne . . Infrared spectrum of matrix (Kr)-isolated diazopropyne . . . . . . . . . . . . . . Page 10 13 15 17 23 25 29 35 Infrared Spectrum of a mixture of matrix (Kr)- isolated diazopropyne and its deuterated molecules . . . . . . . . . . . . . . . . Raman Spectrum of solid diazopropyne . . . Raman spectrum of a mixture of solid diazo- propyne and its deuterated molecules . . Infrared spectrum of a mixture of matrix- isolated diazopropyne and its deuterated molecules (expanded scale in the region 600-250 cm-l) o o o o o o o o o o o o o 0 Internal coordinates for diazopropyne . . . Infrared Spectrum of photolyzed matrix- isolated diazopropyne . . . . . . . . . . vii 37 39 41 55 68 89 LIST OF FIGURES (Continued) FIGURE Page 15. Infrared spectra of photolyzed diazopropyne in krypton matrix (2150 - 2000 cm— ) . . 95 16. Infrared spectra of photolyzed mixed diazo- propyne and deuterated diazo ropyne in kryp- ton matrix (2150 - 2000 cm'1 . . . . . . 97 17. Internal coordinates for the propargylene radical O O O O I O O O O C O O O O O O O 104 viii INTRODUCTION Free radicals have played an important role in the de- velopment of chemistry in this century. In 1918 Nernst suggested (1) that the individual atoms H and Cl were involved as intermediates in the photochemical reaction of hydrogen and chlorine molecules. In 1929, Paneth and Hofeditz (2) postulated that alkyl free radicals were formed upon thermal decomposition of metal alkyls, as ac- counted for by the removal of a metal film in the metal mirror experiments. Free radicals were subsequently pro- posed as reaction intermediates by many prominent Chemists (3). The understanding of the nature of free radicals in- volved in reaction schemes can improve the validity of a proposed mechanism in a related chemical process. Free radicals are defined as Species which have one or more unpaired electrons. Thus they provide unique tests for the theory of chemical bonding, since the unpaired electrons of the free radical provide a great opportunity for electronic interaction with neighboring atoms and groups. Knowledge of the vibrational Spectrum of Simple free radicals can provide information regarding the structure and bonding of these interesting chemical Species. Due to the extremely reactive nature of many free radicals, the direct detection 1 2 of free radicals by means of conventional Spectrosc0py was not possible. However, the use of rigid, inert host matrix materials such as argon and nitrogen for the isolation of guest molecules providing sources of free radical upon in §i£3_photolysis was pioneered in the laboratory of Pimentel (Shittle g£_al,, 1954) (4). Subsequent application of this technique by Milligan and Jacox (5) and many other workers has demonstrated its power for preparing and stabilizing free radicals and other reactive species for direct Spectro- scopic observation. The purpose of this work is twofold: to elucidate the vibrational Spectrum and structure of the prOpargylene free radical C3H2 (H-C=C=C-H), and concomitantly to study the vibrational spectrum of its parent molecule, diazopropyne, C3H2N2 (H-CEC-CH=N=N). Few diazo-compounds have been studied by vibrational spectrosc0py, the only Complete work being diazomethane, CH2N2 (6), which provides some helpful informa- tion for the vibrational study of diazoprOpyne. The propargylene free radical has been observed by Skell (7), who has demonstrated by means of electron Spin resonance spectroscopy that diazopr0pyne yields the propar- gylene radical when photolyzed in various organic solvents at liquid nitrogen temperature (770K). In addition the electronic absorption Spectra of the propargylene radical and C3 radical from the flash photolysis of diazopropyne were obtained by Merer (8), however, no ground state vibra- tional information was Obtained in this study. The diazo- propyne molecule thus offered the possibility of producing 3 the propargylene radical (gig photolysis in an inert gas matrix) which could be detected by infrared and/or Raman spectroscopy. The vibrational study of the prOpargylene radical would add information on the correlation of elec- tronic structures and vibrational force constants in some linear molecules with cumulated double bonds, such as C3 (9)! C302 (10): C352 (11): C3N2 (12), C3H4 (13), etc. CHAPTER I THE INFRARED AND RAMAN SPECTRA OF DIAZOPROPYNE Diazopropyne, C3H2N2, the molecule Chosen in this work as the potential parent of the propargylene radical, has not been previously studied spectroscopically. There have been a number of studies of related molecules, such as diazo- methane CH2N2 (6), dicyanodiazomethane (CN)2CN2 (12), and the propargyl halides H-CEC-CHZX (14), which provide cor- relation in the assignment of fundamental frequencies for diazopropyne. A valence force field can be constructed using the force constants adapted from these related molecules and used as a starting point for the normal coordinate analysis of diazopropyne. The relative positions of the vibrational fundamentals, such as the Characteristic N=N stretch and C=N stretch, and the values of the associated force constants in the diazo-compound should provide some indication of the elec- tronic distribution, for example the degree to which possible resonance structures are important in this molecule. The thorough understanding of the diazopropyne parent molecule will also provide assistance in the understanding of the vibrational spectra of the photolysis products, in particular the propargylene radical. 5 In this Chapter, the author presents infrared spectral data concerning gaseous, solid and matrix-isolated diazo— propyne C3H2N2 (H-CEC-CH:N=N), and its isotopic molecules C3HDN2 (D-CEC-CH=N=N), C3DHN2 (H-Csc-CD=N=N), C3D2N2 (D-CEC-CD=N=N), and C3H2N15N (H-CEC-CH=N=15N), and Raman Spectral data regarding solid diazopropyne and its deuterated isotOpic molecules C3HDN2, C3DHN2, and C3D2N2. A complete vibrational analysis has been performed on diazopropyne; thus the vibrational assignment of the fundamentals is achieved. Experimental Preparation of diazopropyne: As most diazo-compounds, diazopropyne is a very unstable material with a relatively high vapor pressure at room tem- perature. It decomposes quickly in the presence of light. Diazopr0pyne was prepared according to the general method for synthesizing diazo-compounds (15). The starting material used was prOpargyl amine; the entire process of preparation can be Shown in the following reaction equations: HCEC-CH2NH2 + HCl > HCEC-CHZNHZ'HCl H SO KQEQ—> HCEC-CHz-NH-C-NHz -3——3¢ HCEQ-CHZ-N —-c—NH2 " NaN02 | H O NO 0 Propargyl urea nitroso-propargyl urea NaOH > NazHPO4-7H20 HCEC-CH=N=N + NH3 + co2 . diazopropyne 6 The final product diazopropyne was carried from aqueous solution by dry nitrogen gas into a trap which was Connected through a second trap to a vacuum line. These two traps were immersed in liquid nitrogen contained in dewarS. The resulting mixture was vacuum distilled at 800K to remove the nitrogen gas, followed by vacuum distillation from 193°K (dry ice-acetone bath) to 800K (liquid nitrogen) and from 273°K (salt-ice bath) to 193°K (dry ice-acetone bath), which gave essentially pure diazoprOpyne, confirmed by the absence of significant impurity absorption in the infrared spectrum. The final product was a bright yellow solid at liquid nitrogen temperature and a Slightly darker yellow solid at 1930K. This indicates that solid diazoprOpyne might exist in two different solid phases at the two different temperatures. Deuterated diaZOprOpyne was also prepared, the pro- cedure being simply the reaction of nitrosopropargyl urea, sodium hydroxide and sodium hydrogen phOSphate in heavy water (D20) solution instead of normal water (H20) solution at the last step of the synthesis. There were four products formed at the end of this reaction, namely, C3H2N2 (H-CEC-CH=N=N), CSHDNZ (D-CEC-CH=N=N), C3DHN2 (H-CEC-CD=N=N) and C3D2N2 (D-CEC-CD=N=N). Attempts to prepare pure C3D2N2 by using NaOD and NazDPO4 instead of NaOH and NazHPO4 respectively in the last step of the reaction sequence were not successful; this might be due to the slow exchange rate between hydrogen and deuterium under the conitions of the synthesis. 7 The C3H2N15N was prepared by using Na15N02 instead of NaNOz in the nitrosation of prOpargyl urea, the remaining reaction sequence being the same. Since only a small quan- tity of Na15N02 was available, it was necessary to mix the labeled material with some normal NaNOz in this step in order to produce a satisfactory yield of the diazopropyne product for detection by infrared and Raman spectroscopy. The decomposition of diaZOpropyne, especially in the gaseous phase, occurred when it was exposed to heat or radiation. Solid diazopropyne eXploded readily at liquid nitrogen temperature under any kind of disturbance. Thus it is important that diazopropyne, especially in solid form, should be handled with extreme care. It is advisable to use no more than 1.5 grams of nitrosopropargyl urea for the reaction in base solution to produce diazopropyne, because the greater the quantity of diazopropyne present the harder it is to handle and to control. At the start of this work the author experienced several unpleasant explosions, but gradually it was learned how to handle diazopropyne with reasonable confidence. Spectra of DiazoprOpyne: (a) Gaseous diaZOprOpyne was placed in a 10 cm long gas cell fitted with CSI windows. The gas phase infrared Spectrum had to be taken in a short period of time Since diazopropyne decomposed under exposure to the source radi— ation, as well as thermally. 8 (b) Pure liquid diazopropyne could not be obtained, and there were also no suitable solvents available. Ether and pentane are good solvents for diazopropyne, but their absorption bands overlap with diazopropyne's in the certain regions of the infrared Spectrum. Therefore it was not possible to obtain the complete spectrum of diazopropyne in the liquid phase or in solution. (c) The infrared spectra of solid diazopropyne at 800K were obtained by spraying the sample onto a cooled CsI window which was installed in an Andonian Associates Model D-307 liquid helium dewar (Fig. 1). The coolant used in the inner jacket was liquid nitrogen instead of liquid helium. The temperature of the solid was in the neighborhood of 800K. The Andonian liquid helium dewar consists of two parts: (1) the outer jacket holding liquid nitrogen which also cools a radiation shield, and (2) the inner jacket which normally holds liquid helium. A stainless steel tube in which a variable pressure of helium gas which can be maintained in good thermal contact withthe window holder passes through the coolant in the inner jacket. The pres- sure of the helium gas and an auxilliary heater can be used to control the temperature at the window holder. Helium gas is liquified in the bottom of this tube if the gas pressure is greater than one atmosphere when liquid helium is contained in the inner coolant reservoir. The space between the outer jacket and the inner jacket can be evacu- ated. The lower portion of the jacket can be rotated, Figure 1. Andonian Associates Model D-307 liquid helium dewar. Exchange gas port Nitrogen fill Helium fill and vent Nitrogen vent Dewar body outer shell Nitrogen reservoir Helium well Vacuum insulation Exchange gas Chamber Extended area heat exchanger Outer tail Nitrogen temperature radiation shield Cold finger - sample mounting platform Demountable window OZKbNQl-I’JEO'TJMUOUJIP Sample temperature radiation shield 11 allowing deposition and photolysis to be performed sequenti- ally. For taking Raman Spectra of solid diazopropyne at 800K, an "L"-shaped Pyrex glass cryostat (Fig. 2) was used, which consists of two parts: (1) the outer portion for insulation, and (2) the concentric inner jacket holding liquid nitrogen. The copper deposition head is connected to the inner jacket with a Kovar metal-to-glass seal. For the solid diazopropyne spectra at around 300K, a Malaker Corp. Cryomite Mark VII-C closed-cycle helium re- frigerator was used. A platinum resistor and a Zener diode were installed at the cold head of the Cryomite for moni- toring and controlling the temperature. For taking solid and matrix-isolated diazopropyne infrared Spectra a copper window holder was fitted on the Cryomite, and an inverted “T"-shaped brass outer jacket was used for insulation, spray—on and photolysis purposes (see details in Fig. 3). The windows used were 051 for the normal infrared region and polyethylene for the far infrared region. The substrate window upon which the sample was deposited was mounted in good thermal contact with the copper window holder to in- sure rapid dissipation of the thermal energy of the sample upon condensation. Indium spacers were used flnrthis purpose. For obtaining Raman Spectra, a sharp-angle Spray-on copper sample head was fitted on the Cryomite (Fig. 4), and a glass outer jacket was used to maintain vacuum. For matrix isolation experiments, noble gases such as neon, krypton and xenon are the best host materials, since 12 .ucfloh HHMQ Ccsonm :N\H suaB xooomoum mauuoc nonmmnmm 0mm: cofluflmommw Hommou amen Hm>ox :N\H ucflon Hana chonm :N cmmouuac Uflzoaa mcfloaon now umxumn Mecca oauucmocou coaumasmc« How umeMn Hmuso .Hamu cmemm eunumummEmu 304 .N musmflm 13 \ \ _______________1 0') — .N madman Tlln Figure 3. 14 Spray-on infrared cell for Malaker refrigerator. O-ring CsI window for infrared spectrum taking. Spray-on nozzle is soldered on a circular copper disk, which is pressed onto the cell with an aluminum flange, Needle valve for controlling the flow of the gas mixture. To place quartz window for photolysis purpose, CSI window used for deposition purpose is placed in the COpper window holder which is attached to the cryomite, Connection to the cryomite, which enables rotation of the outer jacket for spray-on and photolysis purposes. 1 5 SIDE I FRONT T r-W L d l: l a : Sit/é ((6% j «R; ‘f 1": J TOP BACK 16 Figure 4. A. Spray-on cell for Malaker refrigerator. B. Closed sample head for Malaker refrigerator. a. Both sample cells were contained inside a 2-inch outer vacuum jacket b. Attached to Malaker cold head c. 1" Kovar seal d. Spray-on nozzle e. Remainder same as design B. f. Silver soldered g. 1/2" Kovar seal h. 1/8" Kovar seal i. 1/16" thin-walled stainless steel tubing coil j. 1/2" outside vacuum jacket Spacer with 1/16" tubing feed-thru 18 they are free from absorption in the vibrational frequency region. Nitrogen is also a good matrix material which is often used. Since the Cryomite in our laboratory cannot provide temperatures lower than 250K, neon, argon and nitrogen could not be used for matrix material in our experi- ments. Experience from various experiments demonstrated that krypton was the best matrix material for diazopropyne at 250K. Xenon was also suitable, but due to its large Size it scattered quite strongly. Thus the matrix-isolated diazopropyne Spectra in a krypton host were better than those obtained using xenon matrices. (Carbon tetrachloride, carbon monoxide and carbon dioxide were not good matrix materials for diazopropyne at 250K). Matrix samples are the collection of randomly oriented micro-crystals formed by rapid quenching from the gas. It is desired to produce a matrix with the maximum possible sample isolation within the host matrix, yet which is as transparent as possible in order to transmit the maximum amount of source radiation. The sample deposition rate is an important factor because it controls the degree of iso- lation of the sample within the host lattice and the trans- parency of the matrix. A relatively slow deposition rate of 1-3 m mole per hour was preferred, and the host-to- sample ratio was about 200 to 1 in the diazopropyne matrix isolation experiments. Several reviews of the matrix iso- lation technique have been published (16), which are most useful for planning and carrying through such eXperiments. 19 Infrared Spectra were obtained using a Perkin-Elmer Model 225 grating spectrophotometer in the region 4000 cm-1 1 to 200 cm- , with resolution better than 1 cm"1 above 450 cm.1 and 1 to 2 cm"1 below 450 cm_1. The Spectrophotometer was calibrated against water vapor and carbon dioxide, and the reported frequencies Should be accurate within i2 cm_1. An RIIC - Beckman Model FS—720 Fourier Transform Spectrometer (interferometer) was used for obtaining infra— red Spectra below 200 cm-1. AS mentioned earlier, the final product, diazoprOpyne, was collected from aqueous solution. Drying agents such as Drierite and Silica gel were not used in the preparation since they adsorbed gaseous diazopropyne strongly. Since water vapor absorbs quite strongly below 300 cm-1 and the amount of diazopropyne was quite small, no gas phase diazopropyne Spectra could be obtained in this range. Spectra of solid diaZOpropyne were obtained below 200 cm"1 with 2 cm-1 resolution. It required a large amount of sample in order to obtain a decent spectrum. It was not possible to obtain infrared spectra of matrix isolated diazopropyne below 200 cm-1. One potential cause of this problem might be the relatively poor thermal conductivity between the COpper window holder and the poly— ethylene window, which could lead to the formation of a non- rigid matrix, with subsequent aggregation of the guest molecules. Solid Raman Spectra were obtained using a Spectra— Physics Model 125 He-Ne laser or a Coherent Radiation 20 Laboratory Model 528 krypton laser as excitation sources, and a Spex Model 1400 double monochromator to disperse the scattered radiation. The Raman spectrum above 3000 cm-1 was not obtainable since the sensitivity of the photo- multipliers used (ITT-FW-130 and RCA C31034) decrease rapidly in this range when the red lines of the He - Ne or krypton laser are used as excitation sources. An argon ion laser could not be used Since diazopropyne absorbs in the green (5145 X) and blue (4880 A), and may also decompose under the influence of the radiation from this laser. Reso- lution and frequency accuracy were approximately :4 cm_1 in the Raman studies. Results and Discussion From comparison with the related molecule diazoaceto- nitrile, NEC-CH=N=N (17), which has CS symmetry, the diazopropyne is assumed to have the symmetry CS: Pvib = 11 A' + 4 A". It follows that there will be fifteen dis- tinct fundamental vibrational frequencies, of which eleven will belong to the A' symmetry Class and four to the A" symmetry Class. All normal vibrations are both infrared and Raman active. The gas phase Spectrum of diaZOpropyne is preferred for the normal coordinate analysis. Since the molecules vibrate freely in the vapor, the observed gas phase fre- quencies represent the true vibrational frequencies, con— sidering vibration-rotation interactions and other small 21 perturbations to be negligible. The Spectra of gaseous diazopropyne and its mixed deuterated molecules C3H2N2, C3HDN2, C3DHN2, and C3D2N2 are Shown in Figure 5 and Figure 6 respectively. The frequencies and their relative intensi- ties are listed in Tables I and II. In assigning the gas phase spectrum of diazopropyne, the following information was utilized: (1) type of rota- tion—vibration band envelope, (2) the frequencies of related molecules, (3) an initial normal coordinate calculation uti- lizing force constants transferred from Similar bonding situ- ations, (4) frequencies from Spectra of solid and matrix-iso- lated diaZOpropyne, and (5) frequencies shifts upon isotOpic substitution. Since there are only two symmetry classes in diazopropyne, the infromation from the type of band envelope is very helpful. Diazopropyne is an asymmetric top molecule, and thus all the vibrations in the molecular plane will give A-type or B-type bands (with higher intensity in the P, R branches or 01, 02 branch), whereas vibrations out of the molecular plane will give C-type bands (with higher intensity in the Q branch) (18). On this basis, one should be able to assign all the absorption bands to their proper symmetry species, A' or A". The principal inertial axes of diazo- prOpyne, which are calculated from bond angles and lengths (adapted from the related molecules, diazoacetonitrile and diazomethane) are depicted in Figure 7. Microwave spectra of diaZOpropyne were taken in the regions, 26500-31000 MHz and 36380-39600 MHz. The Spectra in the above ranges were quite decent. The Spectrum in the range 18000-215000 MHz was also taken, but was not 22 .mcxmoumoumflp msommmm mo Eduuuwmm CommuwcH .m musmam 23 o. a. .m madmam 7.5 No — v— 0— .— . _ _ [21916331. 0« on 0v .o~ oO‘l .oo .8 log 24 .mmasomaoa Umumumusmp mua Cam mcwmoumoumap msommmm mo musuxfls m mo Eduuommm CmumumcH .m madmam 25 .m musmflm TEu No— v o m o. «_ z o. a. on nu on an _ _ _ _ _ . . . _ _ _ . on Io .ON .ov .00 lo? we: 1 26 Table I. Observed infrared absorption bands of diazopropyne vapor (2980K). Frequency Frequency (cm-1) Band Type (cm-1) Band Type 3338 Q (s) B 832 R w) B 3325 Q (s 818 P w) 3107 R (w) B 689 R (m) B 3093 P w 674 P (m) 2335 R (w) B 540 R m) C 2328 P (w 531 Q s) 3:23 8 (S) B 504 (w) ? S 2079 Q 476 (w) C vs 2060 Q (vs) B 355 Q s) B 1364 346 Q S) R m 1353 p (m) B 1180 R w 1168 Q w B 1163) Q w 1150 P w 1064 R m B 1053 P m Symbols: R, Q, P — branch designation. vs — very strong, s - strong, m - medium, w - weak, ( ) - uncertain band. 27 Table II. Observed infrared absorption bands of a mixture of gaseous diaZOpropyne and its deuterated molecules. FrequenCies Band Type FrequSTCIes Band Type (cm ) (cm ) 3335 Q S B 1172 vw 3320 Q s 1152 vw 3110 R w) B 1062 Em 3100 p w 1050 m B 2605 Q (s) B 816 R (w) B 2594 Q s 804 p (m) 2270 R w B 688 R w) B 2260 P w 676 P m 2090 R vs) 540 R m) C 2075 Q vs A 529 Q S 2060 P vs 475 Q (m) C 1985 R m B 1975 P m 442 w) 410 w) 1360 R w) B 404 w 1346 P w) 356 Sh) 346 S 332 m 1284 R w) B 288 vw) 1270 P w 284 w) Symbols: R, Q, P - branch designation. vs - very strong, S - strong, m — medium, w — weak vw - very weak, Sh - shoulder, ( ) - uncertain band. 28 .TCAQOHQONMHU mo mmxm HMHDHTSH mamflocflnm .h musmflm 29 .h mudmfim 30 satisfactory. Satellite bands of some transitions were ob- served. The satellite band has the same band shape as its parent transition (rotational transition in the ground vibra- tional state), but with less intensity, and is due to the rotational transition in the first excited vibrational state. Such satellite bands only occur when the first ex- cited vibrational state lies not too far above the ground vibrational state. If the molecule has a low bending fre- quency (around 200 cm"1 or below), satellite bands will pos- sibly be present in the microwave Spectrum. From the in- tensities of the parent and satellite bands, a vibrational frequency for the transition from the vibrational ground state to first excited state can be approximately calculated by the Boltzmann equation. By using bond lengths and bond angles adapted from the related molecule diazoacetonitrile (17), and Pro- fessor R. H. Schwendeman's computer programs STRUCT and EIGVALS, a rotational analysis can be carried out to fit observed frequencies, and refined rotational constants of diaZOpropyne can be calculated. A reasonable fit was obtained in the 26500-31000 MHz and 36380-39600 MHz regions. The fit below 26500 MHz was not satisfactory; this is perhaps due to the uncertainty of the observed bands below 26500 MHz. Thus a harder search has to be made in the region below 26500 MHz. However, with the aid of the rotational calculation, a preliminary A-type assignment was achieved, Which is not shown in detail in this dissertation. The 31 tentative rotational constants for diazopropyne are obtained as follows: A = 28968 MHz, B = 2785 MHz, and C = 2653 MHz. Due to the unstable nature of the molecule, the amount of diazoprOpyne in the gas phase Should decrease with time because of decomposition. The intensities of the vibrational absorption bands due to diazopropyne should be decreased proportionally as time elapses. For those bands which are not due to diazopropyne, the intensities Should either grow or remain the same with respect to the time. Thus it is not too difficult to identify the vibrational absorption bands which are due to diazopropyne. Much higher resolution than is available in our laboratory would be required for an attempt to perform a rotational analysis of the vibra- tion-rotation bands of diazopropyne. Only the envelopes of the bands were obtained for the gas phase Spectra. Based on the shapes of the band enve10pes, the fre- quencies from related molecules, the results of an initial normal coordinate calculation, the frequencies of matrix isolated diaZOpIOpyne, and the deuterium - substitution study, most of the absorption bands have been assigned with certainty, except those bands below 600 cm-1. There are four or five overlapping bands between 570 cm—1 and 440 cm-1, with total band width of approximately 80 cm.1 (see Fig. 5). The normal A— or B-type band width is about 1 20 cm”1 for diaZOpropyne. The bands at 531 cm- and 476 cm.1 are the only two distinct C-type bands, which are due 32 to two of the out—of-plane bending motions of the diazo- propyne molecule. The band at 540 cm-1 looked like an A- or B-type band overlapping with the 531 cm-1 band, but in the matrix Spectrum, there is only one band at 525 cm-1, corresponding to the 531 cm-1 band in the gas phase spectrum. The band at 540 cm.1 in the matrix spectrum is due to di- azopropyne dimer or polymer since its intensity varies with different host—to-sample ratio. Thus, there is no band in the matrix Spectrum corresponding to the 540 cm-1 band in the gas phase Spectrum. Therefore, the band at 540 cm.1 is possibly the R branch of the 531 cm.1 band. The two bands between 531 cm-1 and 476 cm.1 are around 504 and 490 cm.1 respectively; the structures of these two bands are not distinctive. The former loses intensity, the lat- ter gains intensity with time, so it is logical to eliminate the 490 cm.1 band from the list. Now the bands from 570 to 440 cm- are narrowed down to three. The frequency of the C-CEC out-of-plane bending motion in the propargyl halides (20) is about 300 cm-1, but no bands are observed around 300 cm.1 in the gaseous diazopropyne Spectrum. The gas phase infrared Spectra of deuterated mixtures of diazopropyne, namely C3H2N2. C3HDN2, C3DHN2, and C3D2N2, are relatively weak and the positions and Shapes of many of the isotopic bands are not clearly evident. AS mentioned in the experimental section, a most unfortunate problem is the difficulty in separating CstNz, C3HDN2. C3DHN2, and CaDaNa. Due to these problems with the gas phase infrared Spectra of diazopropyne and its deuterated mixtures, it was 33 necessary to obtain solid phase Raman spectra as well as solid phase and matrix-isolated infrared Spectra of diazo- propyne and its deuterated mixtures in order to assign the fandamental vibrations. The advantages of matrix isolation spectra are the Similarity of the matrix isolated frequencies to the gas phase frequencies and the sharpness of the matrix isolated absorption bands, which permit us to identify some bands which cannot be resolved in the gas and solid phase Spectra. In addition, the matrix isolated diazopropyne was used to prepare the prepargylene free radical by photolysis with ultraviolet light; this work will constitute the second chapter of this dissertation. Raman spectra of solid diazopropyne help identify those bands which have weak intensities in infrared spectra, but strong intensities in Raman Spectra. For example, the 1165 cm.1 band is very weak in the infrared, but very strong in the Raman spectrum. In Spectra of solid and matrix isolated diazopropyne, the identification of absorption bands due to diazopropyne was Simply carried out by ultra- violet light photolysis. The bands due to diazopropyne should decrease in intensity, and those due to other mole- cules should increase in intensity or remain unchanged. The infrared spectra of matrix-isolated diazopropyne and the deuterated isotopic mixture are shown in Figures 8 and 9. The solid Raman spectra of diazoprOpyne and its deuterated mixture are shown in Figures 10 and 11. The 34 .mcwmoumoumac Umumaomfllfiuxv xauume mo Eduuummm CmnmnmcH .w musmflm 35 .w mnsmflm 1.5 .o. m .V W a 6.. N._— J. 0.. O_— 0." JR 0.0 0." lo . ON ‘ ‘ - 0V , _ 1 _ __. _ _ u 00 r o I _ _. , it I H I ., I f .133 i k t L. _m . 1%.... -8 c 5... f 5L4 . .1 r a I. a???) F... ,4. 1). . arc \isf ff 6].! 115F164 . lfil’lX be. .12).»)... 1/ (11>; e) * 1 36 .wchmonmonmac 00 05C uoc mum x £ua3 pmxume mncmn one .mmasomaoe Cmumumudmw mud can mammoumoumfip CTDMHOmfllfiuxv xauumfi mo musuxwe m mo Eduuommm UmumumcH .m mudmflm 37 .m musmflm 7.5.2 0.. a.— J— 0.. 0.. o.« JN JN N.n 0..” I0 I -2 I). _ .s. ‘1 . fl _ . 1 _ . 00 c, C r a .<. ck {/3}? [09 omuoq 38 .mmcfla mocmommuosHm mum X LDAB cmmeE mccmn one .mcwmoHQONMHC cflaom mo Eouuommm cmemm .oa onsoem 39 Vv«~ n—.N moo“ moo“ «.0 .OH mudmflm .783 xucmagoi 2 «amp u «an. «on Nov :2. x no.— «u— 40 Figure 11. Raman spectrum of a mixture of solid diazo- propyne and its deuterated molecules. The bands marked with X are fluorescence lines. 41 ‘68 156 1961 I274 2”? 1162 990 2295 206i .3“ 977 ,f/‘// m»? ' ,._, Frequency -1 Figure 11. 42 frequencies and intensities of the vibrational bands ob- served in solid and matrix infrared and solid Raman spec- tra of diazopropyne and its deuterated mixture are given in Tables III - V. Assignment of the Observed Fundamentals A' symmetry class: (The bands in this class are due to in-plane vibrations with A— or B-type band enve10pes in the gas phase spectrum.) v1 and v2 The gas phase bands of C3H2N2 at 3330 cm.1 and 3100 cm.1 shift to 2599 cm"1 and 2263 cm.1 respectively in deuterated diaZOpropyne. According to the isotope shift calculation, and by comparing the frequencies with the isotopic shift of C-H stretching frequencies of related molecules (3335 to 2607 cm'1 from H-CEC-CHzBr to D-CEC-CHzBr [20,21] and 3077 to 2244 cm"1 from H2CN2 to D2CN2 [6]), the v1 and v2 bands at 3330 and 3100 cm"1 are assigned re- spectively to the stretching motion of the acetylenic C-H bond and the apex C-H bond in diazopropyne. v3 The band at 2118 cm_1 in the gas Shifts to 1980 cm-1 in the deuterated diazopropyne mixture. The matrix band at 2118 cm.1 shifts to 1982 cm.1 in the deuterated mixture, but there remain a couple of bands around 2118 cm_1; these are due to the presence of C3H2N2 and C3DHN2 in the mixture. Table III. 43 Observed IR frequencies of matrix isolated diazo- propyne and its mixed deuterated molecules. C3H2N2 in Kr C3H2N2 and Its Deuterated C3H2 N2 and Isotopes in Kr C3H2N15N in Kr 3321(S 3109(w) 2320(w) 2166 w) 2141 m 2118 S 2069 vs) 1361 m),1358(sh) 1352 w) 1050 m),1047(m) 1175 vw) 1040 w,sh) 691(w) 680 vs) 540 w 525 s 510 w 494 m 485 vs) 476 w,sh) 359(3) 357 s,sh) 309(vw) 3321 s 3321(s) 3109w 2599 ,2587(m) 2280(:; sh) ,2272(m ) 2265 w Sh) 2141,2138Ew,)sh) 2141 m) 2120w M2118 2118 s ,2112(m) 2085 ,2069( vs) 2068 vs),2052(vs) 1985 w, Ssh) ),1982(S) 1972 vw, sh) H1960(VW,Sh 1355 1357(m) 1352 ,1348( (m, sh) 1291 w) )1285(m) 1282 m, vwsh) ,1276(w, Sh) 1175 1045m1047(w) 1038 Sh) 1040 w,sh) 812w ,809 w, E,Shg 803 ,800( m, Sh 691w 692 m) 680 680 S 540w ”532 541 m),532(w,sh) 525 W523 525 m) 510 494 494(m,sh) 485 485 s) 476 m, Ssh) 460 ),443(m) 405 m), )402(s) 359 s.sh) 359 s) 357 3, ;sh) 358 m,sh 352 354 m,sh 346 350 w,sh 340 m, Sash) 303 297 289 w, sgsh) Continued 44 Table III. (continued) C3H2N2 in Xe C3H2N2 and Its Deuterated Molecules in Xe 3309 gs; 3309 m; 3090 w 2591 s , 2580 gm) 2268 m), 2249 w 2135 (w) 2133 vw) 2114 gvsg 2112 m), 2110 (m) 2063 vs 2080 vs), 2063 Evs) 1978 Sh , 1975 s), 1973 (m) 1356 (m) 1357 m), 1355 Em; 1285 m , 1279 m 1044 (w) 1044 w: 1038 w 807 w , 798 (m) 685 w,sh) 686 w,sh) 679 s 679 s 535 w 531 S 523 s 523 m , 520 m) 492 Em,sh) 493 m , 490 w,sh) 483 s) 484 s , 483 s) 478 m,sh) 445 w), 441 (m) 402 w,sh), 399 (S), 394 (w,sh) 359 ES; 358 s 354 m 354 s 350 sh) 351 s 346 s 335 w 303 s 297 s 289 m Symbols: vs - very strong; 5 - strong; m - medium; w — weak; sh - shoulder. 45 Table IV. Observed Raman shift frequencies of solid C3H2N2 and its mixed deuterated molecules (300K) C3H2N2 C3H2N2 and its mixed deuterated molecules (2295) (w) 2115 (vs) 2112 (m) 2095 (m) 2079 (w) 2063 (s) 2061 (s) 1961 (s) 1352 (S) 1344 (m) 1274 (vs) 1168 (vs) 1162 (s) 695 (w) 674 (m) 613 (m) 601 (w) 492 (m) 491 (w) 443 (S) 418 (m) 362 (m) 352 (w) 342 (w) (244) (vw) 310 (w) 172 (vs) 168 (VS) 156 (vs) Symbols: vs - very strong; S - strong; m - medium; w - weak; vw - very weak; ( ) - uncertain band 46 Table V. Observed infrared frequencies of solid C3H2N2 and its mixed deuterated molecules. C H N C3H2N2 and its mixgdzdguterated (850K) molecules (300K) 3315 s 3290 (s; 3285 s 3073 m 3072 m 2581 s 2575 s 2265 mg 2250 m 2130 (m) 2140 sh) 2112 vs) 2116 S) 2088 VS) 2062 (vs) 2068 vs 1971 s) 1950 w,sh) 1350 (m) 1348 s) 1276 s 1165 (vw) 1166 vw 1160 vw 1050 (w) 1048 w 836 w 803 w 695 (m) 695 m,sh) 681 S 613 (w) 611 m 554 (m 532 S 495 s 487 (s) 490 s 445 s 403 S 360 (m,sh) 361 s,sh) 358 s) 355 s 349 s 305 s 300 5 Symbols: vs - very strong; 5 - strong; m - medium; w - weak; vw — very weak; sh - Shoulder. 47 The CEC stretching frequency at 2138 cm-1 in propargyl bromide shifts to 2006 cm.1 in D—CEC-CHzBr (21). From this comparison it is quite obvious that v3 the gas phase band at 2118 cm.1 is mainly due to the CEC stretch in diazoprOpyne. V4 The gas phase band at 2069 cm.1 in diazoprOpyne shifts to around 2080 cm.1 in the deuterated mixture. The 2080 cm.1 band is difficult to locate with precision because of the presence of C3H2N2, C3HDN2, C3DHN2, and C3D2N2 in the gas mixture. In the matrix Spectrum of the deuterated di- azopropyne mixture, the 2069 cm"1 and 2086 cm.1 bands are quite distinct. In the matrix spectrum of the C3H2N2 and C3H2N15N mixture, bands at 2068 and 2052 cm.1 are observed; the latter is due to participation of the N=15N stretch in the normal vibration. The observed deuterium and nitro- gen-15 isotopic shifts are in good agreement with the normal coordinate calculation result, including the upward shift of this normal mode upon deuteration. Since the deutera- tion of the acetylenic hydrogen atom of diazopropyne can change the composition of this normal mode, this result is not completely unexpected. According to the isotOpic shift results and from comparison with the frequencies of related molecules (2102 and 2147 cm.1 for N=N stretches in diazo- methane (6) and dicyanodiazomethane (12) respectively), the V4 band at 2069 cm.1 is mainly due to the N=N stretching ‘vibration. 48 V5 The band at 1358 cm-1 in gaseous diazoprOpyne shifts to 816 cm.1 in the deuterated diazopropyne mixture. The shift caused by replacing one or two hydrogen atoms with deuterium is quite large, so it is clear that hydrogen motion must contribute strongly to the 1358 cm-1 band. In acetaldehyde the C-H wagging frequency Shifts from 1400 cm.1 to 849 cm- upon replacement of the hydrogen atom on the carbonyl group (22). It is evident that the C-H wagging motion in diazopropyne is the primary contributor to v5. V6 The 1165 cm-1 band is quite weak in gas, solid and matrix isolated infrared Spectra of diaZOpropyne, but it is strong in the Raman Spectrum of the solid, so it is designated as a fundamental. The c=N stretching frequency in diazomethane is 1170 cm.1 and it shifts to 1157 cm-1 in CHDN2 and 1213 cm.1 in CD2N2 (6). There is one band ob- served in the mixed deuterated diaz0propyne gas phase spec- trum at 1272 cm-1 and one band with strong intensity at 1274 cm“1 in the solid Raman Spectrum of mixed deuterated diazopropyne. According to the normal coordinate calcula- tion the c=N stretch contributes about 50 percent to this normal mode. At this point, the author would like to stress that a given normal mode is often not representative of only one particular group vibration; sometimes it corresponds to the 49 combined vibrations of several groups, each contributing in different weight. The percentage of the contribution of each vibration to the normal mode is calculated in the nor- mal coordinate analysis and is listed in the output of the computer program under the title of potential energy distri- tion (PED). For instance, v1 and v2 of diazopropyne are due almost solely to the acetylenic C-H and apex C-H stretches respectively; this is because C-H stretching motions normally do not couple with other vibrations due to their high frequency nature. (However, a C-H motion will couple with other C-H stretching vibrations if they are geometrically close to each other in the molecule.) However, if some groups are close to one another structurally in the molecule and their individual frequencies are not too far apart then they will have a tendency to couple with one another. The normal mode represents the resultant frequency of a particular coupling of internal coordinates. V7 According to the normal coordinate calculation, the gas phase band at 1058 cm.1 is due to a complex vibration of several groups, including the C-H wag, C-C stretch, and C-C=N bend, each contributing less than 50 percent to the potential energy distribution. It is thus not appropri- ate to assign the 1058 cm-1 band (v7) to any one of the group vibrations mentioned above. A Similar situation is encountered in the mixed deuterated diazopropyne gas phase 50 spectrum. As mentioned above, a band at 1272 cm.1 is ob- served, which is not Observed in the gaseous diazoprOpyne spectrum. It is obviously due to the isotopic shift, but it is difficult to tell precisely to which band in diazo- propyne it corresponds. According to the normal coordinate calculation, the 1272 cm-1 band is due to the complex contri- butions of the c=N stretch, C-C stretch, C-D wag, N=N stretch, and CEC stretch, and it is assigned to v5 in C3D2N2. In fact in the Raman spectrum of the solid deuterated diazopropyne mixture, the 1274 cm_1 band is quite strong and the 1162 cm"1 band, which is due mainly to the c=N stretch and is very strong in diazopropyne, is relatively less strong than the 1274 cm-1 band, so it is 1 not surprising that 1274 cm- band has some contribution from the c=N stretch. V8 In the gas phase spectrum, the deuterium isotope shift of the 681 cm.1 band of C3H2N2 is difficult to Observe, because the deuterium-shifted frequency overlaps the 531 cm.1 (C-type) band of diazoprOpyne. However, in the mixed deuterated diazopropyne matrix spectrum, bands at 691, 680, and 532 cm-1 are Observed. The 691 cm-1 band is probably due to the dimer or a polymer of diazopropyne, since it is present in the matrix diazopropyne spectrum and its in- tensity varies with different matrix-to—sample ratios. The 680 cm-1 band is due to C3H2N2 and/or C3DHN2 (deuterium 51 1 substitution at the apex hydrogen atom). The 532 cm- ab- sorption is due to C3D2N2 and/or C3HDN2 (deuterium substi— tution at the acetylenic hydrogen atom), since it is not present in the matrix diazopropyne spectrum. This assign- ment also receives support from the deuterium isotope study of prOpargyl bromide (21), in which the CEC-H bending frequency shifts from 637 to 501 cm-1 upon replacement of the acetylenic hydrogen atom with deuterium. The frequencies from the normal coordinate analysis are 678 and 537 cm"1 for the CEC-H and CEC-D bends respectively, which are in good agreement with observed frequencies at 681 and 532 cm-1 for diazopropyne and its deuterated molecule. Thus the v8 band at 681 cm.1 is due to the CEC-H in-plane bend and is assigned to v8. 1 The author assigns the 680 cm- band to the CEC-H 1 band to the bends in C3H2N2 and C3DHN2, and the 532 cm- CEC-D bends in C3D2N2 and C3HDN2. This is based on the similarity of the structures. The replacement of the non- acetylenic hydrogen atom with a deuterium atom on C3H2N2 does not affect the CEC-H bending motion, and the same is true for the CEC-D bending motion upon the replacement of the non-acetylenic deuterium atom with an hydrogen atom on C3D2N2. This assignment receives support from the normal coordinate analysis. 52 v9 The band at 494 cm_1 for diazoprOpyne in a Kr matrix has medium intensity. From comparisons with acetaldehyde (22) and fumaronitrile (23), the former having the C-c=o bending frequency at 509 cm.1 and the latter the C-C=C bending frequency at 538 cm—1, the band at 494 cm.1 in diazopropyne is probably related to the C-C=N bend. The 509 cm-1 band shifts to 500 cm"1 in monodeutero acetaldehyde (CH3CDO) (22), and it is also indicated in the normal coordinate analysis that this isotopic substitution of acetaldehyde does not appreciably affect the C-C=O bending frequency. In the gas phase diazopropyne spectrum, there is a rather indistinct band around 504 cm-1, which corresponds to the 494 cm-1 band in the matrix. The normal coordinate analysis shows that the C-C=N bend, C-C stretch and c=N=N bend are jointly responsible for this normal mode at 494 cm-1, and it also predicts that deuteration should not affect the frequency appreciably. Indeed, there are no bands observed between 494 and 460 cm.1 in the mixed deuter- ated diazopropyne matrix Spectrum. v10 In the gas phase spectrum of diazoprOpyne, a strong distinct B-type band at 351 cm.1 is observed. In the diazo- propyne matrix (Kr) spectrum a strong band at 359 cm.1 is Observed with shoulder at 357 cm-1. The shoulder at 357 cm 1 is not distinct. In the corresponding mixed deuterated 53 diazoprOpyne matrix spectrum there are five bands observed, namely: 359, 357, 352, 345, and 340 cm'1 (Fig. 12). They are present with good intensities except that 340 cm-1 band is only a weak shoulder. The three lower frequency bands are absent in the diazopropyne matrix spectrum and the 357 cm"1 band increase in intensity with respect to 359 cm-1 band in the mixed deuterated diazopropyne matrix spectrum. According to a relative intensity study of these bands in matrix infrared spectra with different isotopic distributions, the 359, 357, 352, and 346 cm"1 bands are assigned to C3H2N2, C3DHN2, C3HDN2, and C3D2N2 respectively. The observed deuterium isotope shifts are in good agreement with the normal coordinate calculation result. In the C3H2N2/C3H2N15N mixed matrix (Kr) spectrum bands at 360, 358, 354, and 350 cm-1 are observed, the last three being shoulders. The 354 cm”1 feature is due to the nitro- gen-15 isotope shift. Thus the 351 cm.1 band in the gas phase spectrum of diaZOpropyne is due to some vibration in- volving the terminal nitrogen atom. Since the frequency of this band is rather low, it is likely this vibration is the c=N=N bend. The c=N=N bending frequencies are between 380 and 420 cm-1 for diazomethane and its various deuterated ana- logs (6). For example, the c=N=N in-plane bending fre- quency of diazomethane shifts from 421 to 393 cm”1 upon replacing one hydrogen atom with deuterium. Therefore, the gas phase band v10 at 351 cm.1 is assigned to the c=N=N in- plane bend. 54 .A H .xauume H& CH Auznomo + uzmamu + azamno + «Zammov mo Esuuommm UmumumcH .m .xfluumE ux CH mzmmmu mo Esuuommm OmumecH .¢ 1E0 omNIoom cowmmu mnu cfl mamom Umpcmmxmv mmasomaoe Omumumusmo mu“ Ucm mammoumoumfln mumaomfl xfluume mo wmduxfle M No Esuuowmm UmumnmcH .NH musmflm 55 0&n 3n .NH musmflm own can one 9% Rn 2.4 . own om... '10 < < - ea . _ _ - ow . i, -oe 3+ 1 r. a / 5‘ fl 1” u , > . x. r . _ a -s <1 . r.F IS 1 ( .. ownumluoq 56 V11 Bands at 168 and 170 cm-1 are observed in the Raman and infrared spectra of solid diazopropyne respectively. In addition to the 168 cm"1 band, a band at 156 cm.1 is observed in the Raman spectrum of mixed solid deuterated diazopropyne. In the related molecule propargyl bromide (20,21), the CEC-C in-plane bending frequency shifts from 186 to 165 cm.1 when the acetylenic hydrogen atom is re- placed with deuterium. Thus V11: the last band in the A' symmetry class, is assigned as the C-CEC bend at 168 cm‘l. A" symmetry class: (The bands in this class are due to out-of-plane vibrations with C-type band enve10pes in the gas phase spectrum.) V12 and V13 Two distinct C-type bands are observed, at 530 cm-1 and 475 cm-1 respectively, in the gaseous diazoprOpyne spectrum. The corresponding isotopic shift bands in the mixed deuterated diaZOprOpyne gas spectrum are at 442 and 404 cm-1. However, their intensities are quite weak and their band shapes are rather structureless. In the mixed deuterated diazopropyne matrix spectrum, bands at 525, 523, 485, 443, 405, and 402 cm‘1 are observed. The 525, 485 cm“1 bands correspond reSpectively to the 530 and 475 cm”1 bands in the gas phase spectrum, and are also present in the di- azopropyne matrix spectrum. Thus, the remaining absorptions 57 must be due to deuterated diazopropyne molecules. In re- lated molecules, the c=N'N out-of—plane bending frequency shifts from 564 to 528 cm-1, from HZCNZ to D2CN2 (6) and the C-H out-of-plane bending frequency shifts from 637 to 501 cm-1, from prOpargyl bromide to acetylenic deuterated propargyl bromide (21‘. According to the normal coordinate calculation result, the observed matrix frequencies 523 (C3DHN2), 405 (C3HDN2), and 402 cm-1 (canznz) are the deu- terium-shifted frequencies corresponding to the 525 cm-1 band (C3H2N2), and 443 cm_1 is the deuterium-shifted fre- quency for the 485 cm.1 band of C3H2N2. In addition to the above information, the author also has observed a C-type band around 480 cm.1 in a preliminary gas phase infrared Spectrum of diazoacetonitrile.* There should be only two out-of-plane bending frequencies in the 1 region of 500 to 300 cm- for diazoacetonitrile, namely the C-CEN and c=N=N out-of-plane bends. The former should have a frequency around 350 cm.1 (24), so the 480 cm-1 band is probably due to the c=N=N out-of-plane bend. This supports the assignment from the normal coordinate calcu- lation and frequencies from related molecules of v12 and v13, the gas bands at 530 and 476 cm"1 in diazoprOpyne, as the H-CEC and c=N=N out-of-plane bends respectively. *The author attempted to obtain an infrared spectrum for diazoacetonitrile (NEC-CH=N=N) which would be a strong sup- port to the assignment of the vibrational spectrum of di- azopropyne, since their structures are so similar. Unfor- tunately, the complete spectrum of diazoacetonitrile was not obtainable because its vapor pressure is not adequate for infrared spectrosc0pic study. 58 V14 In the infrared spectrum of the mixed deuterated diazo- 1 1 propyne matrix (Kr), distinct bands at 303 cm- , 297 cm- and a weak shoulder at 290 cm_1, are observed. They are absent in the diazopropyne matrix (Kr) spectrum. According to the normal coordination calculation prediction, four bands at 309, 308, 295, and 291 cm"1 are eXpected, corre- sponding to the out-of-plane CEC-C bends in C3H2N2, C3DHN2, C3HDN2, and C3D2N2, respectively. Indeed, there is a band observed at around 309 cm.1 in the C3H2N2 matrix; however, it is quite weak in comparison with its counterpart in the mixed deuterated matrix (Kr) diazopropyne spectrum. The CEO Ammo mm; mno .Ammv zzo «.mH m.mnw o.msv MH> Ana's mww .Amv ouo .lHoHo omoum m.» v.sHm o.mmm* a.) .4 Aomuv zww Amfiv ouo .Aomv zzo .Amvv zoo .Amvv 000 In H.HoH on.moHV fl.) fines zzo .Aomo ooo o.uu o.Hom o.mmm* o~> lame zww .Afimv zzo .Amvv zoo .Aomv uuo o.m- o.mmv o.vmv* a. Amauv mww .Asofiv muomo ¢.m m.w>m o.Hwo m> Amfiv zww .Aufiv zoo .Amflv ouo .onv mm; muo o.m o.mmoH o.mmofi b> lame Zuz .Asvc zno o.m o.omHH o.mmHH m> Amfiuv mm; “MW .Anfiv Zuo .Aomv ono .Ammv mm: :10 o.n v.mmmH o.mmm~ a» Aofiv omu .Ammv zuo .immv zmz n.mu m.fisom o.mmom «> AnHV ouo .imov omo m.nu m.mmHm o.mHHN a. lame mac H.mH m.owom o.oon a. Ammo mum m.o- m.Hmmm o.Hmmm H) .< Aa.m.mv muousnwuucoo mumfiflum >4 .UHMU .on > Eu mmaocmsqmum HI .ANZNanV mammoumoumwo mo mflmwamcm mumcfiouooo HmEHoz .HHH> mHQMB .coaumHsono ucmumcou mUHOM CH pom: pom .UHHOm mo mocmsvmum amamm MA v .mocosvmum xflnum2* 71 Aofiv zzo .Ammv ooo .Ammv om3 muo u- m.Hvu nun nu» “flat on; z-o .Aomv zzo .Avmv ooo o.H w.mmm o.smm* vH> Asfiumuono .AmHv zzo .lmfiv ooo .Afimv anomo s.mu >.wov o.mov* a.» AoHV ouomwow,ofiv ooo .Aomv omz muo .Ammv zzo m.o m.wm¢ o.msv NH> =4 Amfiuv WWW .Amfiv ouo .Amfiv zzo .Afiov zoo .Aomv ooo u- o.omfl on.omHV a.) Amos ooo .Avvv zzo o.o o.mvm o.mmm* o.> Ronny WWW .immv zzo .lmmo zoo .Awov ouo m.H m.Hmv o.vmv* .. Avmuv WWW .AROHV anomo m.m- m.smn o.mmm* a. AmHv WWW .AmHv Zoo .AmHv UIU .Abmv mm3 mIU m.ml m.mv0H o.mvoH* N.> Aofiv mm; mo .Avmv zuz .Auov zuo m.m >.mnHH o.mmfia o» ANHIV mm3 MW .ANHV Ono .AmHv ZHU .Ammv mmB mIU v.m m.vvm~ o.vaH* n> Avfiv ono .Ammv ouom .Ammo omo m.m m.osmfl o.momfi* o. Aomv zuo .Asov zuz >.o m.omom o.mmom* a) homo omo .Awov ouom o.m o.omnm o.mmnm a) Ammo zuo o.mH o.Hmom o.oon .> .< fl.o.m.mv mucosnfluocoo mumsflum >< .oamo .mpo > A H Eov mmflocmsqmum .Aaznmnov mcxmoumonHU mo mflmhamcm mumcflpuooo HMEHOZ .NH magma 72 .coflumasoawu ucmumcoo mouom CH poms uoc .UHHom mo hocwsqwum swamm AV m mocmswmum xanumz* ANHV zzo .Amsv mm; ouo u: m.omfi nun nu> Amfiv zzo .Amwo ooo m.vu m.uom o.mom* «H> Aoflc mm; auo .Aoov zzo «.mu v.mm¢ o.mvv* ”H. Amfiuc zuwww .Amv ooo .Amoflv zuomo s.m m.>Hm o.mmm* ~.> Amfiuv WWW .Amfiv ouo .Aomv zzo .Amvv zoo .xwov ooo In v.ooH on.moHv H.) Avvv zzo .Aomv 000 m.HI H.mmm o.bmm* oH> Asmuv ”mm .Ammv zzo .Aflov zoo .Aomv ouo m.H- m.mmv o.vmw* m> Aflfiuv mww .Asoflv mnomo m.m u.sno o.Hwo m> loos mmz o.o m.HH- m.>mm o.on a. Amflv zuz .AHNV zoo .Ammv zuo u- o.omoH nu- 0) Aoauv mm3 oww .AVHV zuz .Amfiv mas ouo .Asmv ouo .Ammv zuo m.Hu m.mmmH o.mmmfl* n> hofio omo .Awflv zno .Aomv zuz nu m.mvom .1. «> Awe zno .ANHV zuz .xmfic ouo .Amov omo v.H- ¢.HmHm o.omHN* m> Ammo ouo s.ovu s.mfimm o.msmu* fl) Ammo zuom m.ou m.Hmmm o.Hmmm .> .« A.o.m.mv muonsnfluoaoo mumsflum >< .ono .mp0 > AHIEOV mmflocmsvmum .Amzmomov mammoumoumflo mo mfimwamcm mumcwouooo HmEHoz .x magma .coflumHoono ucmumcoo mouow CH pom: no: .UHHom mo mocwsqmnw cmemm V Mr .mocosvoum xfluumz Aofiy ooo .AHHV zzo .Ansv mm3 ouo nu m.mmH 1.. a.) Acav mmB 010 .,mHV zzo .Aumv ooo N.HI N.Hmm o.omN* on» flofiuv ouWWW .ANHV ooo .ARHV zzo .Amwv ouo o u.nu m.soo o.mov* a.» Amflv anomo .Amfiv mm; auo .Aoflv ooo .Asmv zzo «.mHu «.mow o.mov* a.» =< Amfluv WWW .Amfiv ouo .xwflv zzo .Afivv zoo .Aomv ooo u: m.mnH on.omav “.1 Amos ooo .xmvv zzo m.m s.mvm o.ovm* o.» “onus WWW .Awmv zzo .Ammo zoo .Amvv ouo s.m m.Hmv o.omv* a» mm Ammuv WWW .lsoflv anomo s.mu >.mmm o.mmm* a» lows mm: auo o.oH- o.omm o.oHo b) AOHV WWW .ANHV zuz .immv zoo .Afimv zro u: m.owo~ nun o. mmB 010 I Aofiuv zo .AoHV ouo .Amflv znz .Amfiv on; ono .Anmv ouo .Avmv zuo o.mu o.ommH o.mmmfi* a. “may ono .Aumo onom .lfimc omo a.» o.oom~ o.msmfi* v. Ammo zno .Amov zrz m.o- m.moom o.moom a. homo nuo «.mon o.oomu o.momm* «. Ammo omo .Amoc ouom o.m c.5mmm o.mmmu .. .< A.Q.m.mv muousafluucou mnmawum >< .UHMU .mno > «aleoflwmaocmswmum .Am2namov wcwm0um0NMHo mo mammamcm wumaflouooo HMEHOZ .Hx magma 74 .mocmsqmum xflnumz * AvHv zmazu .Amfiv 000 .Ame mmB mun II o.>vN III na> Ammo zofizo .Amsv ooo u: v.mom nu- vd> Ammo mmz muo .Afiov znfizo m.oH H.wmo o.msv ”H. AHoHV zuomo v.5 m.sum o.mmm* NH> :4 loans zWW .Amfiv ouo .Aomv znfizo .Amvv zoo .Asvo ooo u- m.mmH nun H.> Amos anzo .Aflno ooo o.n o.mmm o.¢mm* o.» Amm-c zWW .iomv zmflzo .Amvv zoo .Aomv ono o.o o.vmv o.vmo* a. Amauv WWW .Amofiv mnomo v.m o.mso o.Hmm a. Amflc zWW zoo .Amfiv ono .onv mm3 mo H.m m.vnofl o.wmoH b. “was zofiz .Anvv zo .. s.moafi 1.. .. Amfinv mm; MW .Amfiv zo .Aomv ono .,vmv mmz mo 5.» m.ommH o.man n> Ammo zo .Amov znfiz m.Hu m.mmom o.mmom* .. Amfiv ouo .Amnv o o H.oH- H.mmHm o.mHHm a» lame muo H.mH m.omom o.oon a. Ammo muo m.o- m.Hmmm o.Hmmm .> .a A.Q.m.mv muousnfluucou humeflum >Q .on0 .mp0 > AHIEUV mmflocmsvmnm .AZnHzmmnov mahmoumonmap mo mammamcm mumcflpuoou HmEuoz .HHX magma 75 potential energy distribution among the symmetry coordinates for each normal mode. (Only those symmetry coordinates con- tributing more than 10% to each normal mode are listed in the Tables.) Introduction of additional interaction force constants brought about no significant improvement in the agreement between the calculated and observed frequencies. The main source of the frequency error shown in Tables VIII- XII is probably anharmonicity, most of it Occurring in vibrations where hydrogen atoms are involved. For instance, the calculated deuterium shifts of the apex C-H stretching 1 to 2308 and 2313 cm“1 from c3112»:2 to frequency (3081 cm- C3D2N2 and C3DHN2 respectively) do not agree with the ob- served frequencies (3100 cm.1 to 2265 and 2272 cm-1 from C3H2N2 to C3D2N2 and C3DHN2 respectively). These frequency errors seem too big to be accounted for solely by anharmon- icity. The same discrepancy occurs in the normal coordinate calculation of CH3CHO and CH3CDO (22). In the calculated 2308 and 2313 cm.1 normal modes, about 5% c=N and N=N stretches are contributed respectively to the potential energy distribution for each normal mode. Introduction of CH, CN and CH, NN stretching interaction force constants fails to improve the calculated frequencies. Vibrational Potential Function of DiaZOpIOpyne (A') The C-N bond force constant K11 = 8.92 x 105 dyn/cm is between a single and a double bond (28). The C-N force constant is close to the corresponding C-N force constant, 76 8.34 x 105 dyn/cm, in diazomethane (6). The interaction force constant between the C-N stretch and the in-plane C-H wag, F19 = 0.402 x 10-3 dyn/rad, is fairly large; it makes about a 12% potential energy contribution of v5. This interaction force constant is significant in the valence force field for diazopropyne and it has the same sign and is comparable in magnitude to that of HN3 (0.466 x 10-3 dyn/rad) (29). In order to rationalize the final calculated force constants and the importance of some interaction force constants, three valence bond resonance structures for diazopropyne are drawn below: H-CEC~CH=N=N <-—> H-czc-éH-NEN <—> H-&=c=CH-NEN A B C It is well established that relatively large stretch-bend interaction force constants are required to fit the ob- served vibrational spectra for molecules where intermediate hybridization is implied by such resonance forms (6,12,29,32). The C-C stretching force constant, K22 = 5.00 x 105 dyn/cm, is greater than a normal C-C single bond. The bond length of 1.424 R is shorter than an ordinary single bond. The interaction force constant between the C-C=N deformation and the C-C stretch is fairly large (F27 = -0.876 x 10’3 dyn/rad); it makes about 15%, 38% and 20% potential energy contributions to v7, v9, and VII respec- tively. 77 The N-N force constant, K33 = 14.13 x 105 dyn/cm is not quite analogous to that of N20: 18.3 x 105 dyn/cm (34) or that of CHzNz: 16.9 x 105 dyn/cm (6). Addition of an interaction force constant between the N-N and C-N stretches failed to improve the calculated frequencies and contributed insignificantly to the potential energy. However, the N-N force constant is greater than that of diazirine (CHZNZ): 11.18 x 105 dyn/cm (30) and that of CF2N2: 11.18 x 105 dyn/cm (31), and the N-N bond length is shorter than that of these two molecules: both = 1.228 R (31). The N-N bond force constant and bond length in diazirine CH2N2 are characteristic of a typical N-N double bond. Thus the final calculated force constants K11 and K33 support the im- portance of the suggested resonance structures. The contri- bution of resonance structures (B) and (C) tends to weaken the C-N bond and strengthen the N-N bond relative to double bonds. The apex C-H force constant, K44 = 5.14 x 105 dyn/cm, and the bond length, 1.075 R, are close to those of ketene (32) and diazomethane (6). The CEC force constant, K55 = 15.00 x 105 dyn/cm, is smaller than that of methylacetylene (15.58 x 105 dyn/cm (19)) and that of acetylene (15.80 x 105 dyn/cm (33)). The somewhat low value of the CEC stretching force constant and high value of the C-C stretch- ing force constant may be rationalized by considering the resonance structure (C), which tends to weaken the CEC bond and strengthen the C-C bond. 78 The acetylenic C-H force constant, K66 = 5.89 x 105 dyn/cm, is close to that of methylacetylene (5.85 x 105 dyn/cm (19)) and that of the propargyl halides (5.96 x 105 dyn/cm (14)). The acetylenic C-H bond length is a normal value 1.06 R. The C-C=N bending force constant H77 = 0.905 x 10"11 erg/rad2 and the C-H wagging force constant H88 3 0.456 x 10-11 erg/rad2 are close to the C-C=O bending force constant (1.0 x 10-11 erg/radz) and the C-H wagging force constant (0.43 x 10-11 erg/radz) of acetaldehyde (22), re- spectively. The in-plane c=N=N bending force constant, H99 = 0.418 x 10.11 erg/radz, is close to that of diazomethane (0.48 x 10-11 erg/radz). The CEC-C and H-CEC in-plane skeletal bending force constants, H1o,1o = 0.291 x 10-11 erg/radz; H11,11 = 0.245 x 10-11 erg/rad2 respectively, are close to those of the prOpargyl halides (CEC-C bending -11 erg/rad2 and CEC-H bending force constant 0.26 x 10 force constant 0.19 x 10-11 erg/rad2 (14)). The inter- action force constant between the CEC-C bend and the CEC-H bend, F10,11 = 0.086 x 10-11 erg/radz, contributes about 12% and 9% in potential energy to v8 and VII, re— Spectively. This interaction force constant also supports the existence of resonance structure (C). The remaining interaction force constants do improve the calculated frequencies slightly, but they contribute 79 insignificantly to the potential energy. The final calcu- lated force constants are listed in Table XIII. A" symmetry class: The A" symmetry class is composed of four out-of-plane internal coordinates. Thirteen frequencies for the isotopes C3H2N2, C3D2N2, C3HDN2, C3DHN2, and C3H2N15N were used as input data for the force constant calculation. These 13 frequencies were fitted by 9 force constants: 4 diagonal and 5 off-diagonal force constants. The frequency fit was not quite as satisfactory, with an average error of 8 cm-1 or 1.8%. The normal coordinate calculation should be quite simple since only 4 x 4 matrices are involved in the calcu- lation. However, the potential energy distribution in the normal coordinate calculation indicates that three internal coordinates namely, CEC-C, c=N=N out-of-plane bend and C-H out-of-plane wag are involved in v13, v14, and v15. Due to the uncertainty in v14 and the absence of v15, coupled with the high mixing of internal coordinates in the normal modes, the normal coordinate calculation result is not completely satisfactory in the A" symmetry class. Vibrational Potential Function of Diazopropyne (A“) The OUt-Of-plane c=N=N bending force constant, Hi1 = 0.324 x 10-11 erg/rad2,js somewhat smaller than that of di— azomethane (0.53 x 10-11 erg/rad2 (6)) and that of ketene (0.57 x 10’11 erg/radz (32)). 80 paw .mcHocwn How «p00\m00 .mc00uomumuc0 ocmnwnoumnum 00m omn\cmo 0100 OH .mucmumcoo mouom mGHsUpmuum How EU\cmo acnumuHcD 00: muomo 000.0: zzmm mm3 mo muomu mmS mo 000 000.0- oomm 000.0 oowm 000.0 zzwm 000.0 zzmm 000.0 muombm 000.0 003 0&0 000.0 oowm 000.0 zsz AmGOHumunw> 0cmHm|wonusov =< mm3 m0 mlo 000.0 oom 000.0- omom UMUIE UWUIm 000 200 000.0 ooom 000.0 zzom 000.0 zzom 000.0- oom mmB mo 000.0 zom 000.0 moumm 000.0 ooom 000.0 zzom 000.0 003 mom 000.0 zoom 00.0 muomz 00.00 omoz 00.0 muoz 00.00 zzz 00.0 ouoz 00.0 zoz Amcoflumunw> mamHmlsHv .4 .mcwmoumoumfin How mucmumcoo mouom wocwam> .HHHN OHQMB 81 The out-of-plane CEC-C and CEC-H bending force constants, H52 = 0.239 x 10_11 erg/rad2 and Héa = 0.140 x 10-11 erg/rad2 respectively, also do not correlate closely with those of the propargyl halides (14): 0.33 x 10-11 erg/rad2 and 0.19 x 10-11 erg/rad2 for the CEC-C out- of—plane bending force constants, reSpectively. The out-of-plane C-H wagging force constant cannot be calculated, since no observed frequency is assigned to v15. However, due to the similarity between diazomethane and diazopropyne, the author used the out-of-plane CH2 wagging force constant of diazomethane, 0.042 x 10"11 erg/radz, in the normal coordinate analysis. Further work should be done on the out-of-plane C-H wag of diaZOpropyne, which may well lie below the most efficient region of the Perkin-Elmer 225 infrared spectrophotometer. The interaction force constant between the CEC-C and CEC-H out-of-plane bends, Féa = 0.045 x 10.11 erg/radz, contributes about 15% to the potential energy distribution for v12. The remaining interaction force constants are relatively small in value, and contribute no more than 5% to the potential energy distribution of any normal mode, but introduction of these interaction force constants does improve the calculated frequencies in A" symmetry class. The final calculated force constants of A" symmetry class are also listed in Table XIII. 82 Conclusion From a comparison of the force constants of diazopropyne with those of diazomethane, the propargyl halides, and with some typical force constants such as those for ordinary N=N double bonds, CEC triple bonds, C-C single bonds, and c=N double bonds, it is concluded that the electronic structure of diazopropyne is slightly different from that of diazomethane and of the propargyl halides. However, diazo- propyne has possible resonance structures, such as (A), (B), and (C) which cannot be written for these prototype molecules. Although prOperly a molecular orbital population analysis should be carried out to better describe the elec— tronic structure of diazopropyne, many of the force con- stants derived from the normal coordinate analysis can be rationalized in terms of contributions from these valence bond structures. M CHAPTER II INFRARED SPECTRUM OF THE PROPARGYLENE FREE RADICAL Having established a background spectrum of diazo- propyne, the author proceeds to seek, identify and charac- terize the photolysis products of this fairly unstable molecule. Of special interest is the propargylene radical, C3H2, which has been observed by Skell and coworkers (7) who demonstrated by means of electron spin resonance spec- trOSCOpy that diazopropyne yields the propargylene radical when photolyzed in organic solvents at 770K. Analysis of the ESR spectrum showed the C3H2 radical to be linear in the organic solvent (7). The author has reproduced the ESR spectrum of the propargylene radical in ether and poly- chlorotrifluoroethylene at 770K. The infrared spectra of matrix isolated propargylene radical and of isotopic prop- argylene produced by photolysis of matrix-isolated mixed deuterated diazOpr0pyne were Obtained, and provided informa- tion regarding the symmetry of this radical, which is estab— lished as D Attempts to obtain a Raman spectrum of a>h' matrix-isolated diaZOpropyne were made, but due to the weak scattering activity of the dilute guest molecules the ex- periment was not successful. Thus the Raman spectrum of propargylene radical could not be obtained. 83 84 Experimental Photolysis studies using matrix isolation can be per- formed in two ways: photolysis after deposition (deposition followed by in §i£u_photolysis) and photolysis during deposition. If ig_§itu_photolysis is successful, it gives the most clear-cut results. A matrix containing parent molecules is deposited and the intensity of the parent mole- cules absorption monitored. After a period of photolysis ranging from five minutes to one hour the spectrum is studied. A decrease in the intensity of the absorption bands of the parent molecule, and the appearance of new absorptions clearly indicate the formation of new molecules or radicals. Unfortunately, in situ photolysis is often inefficient. The "cage effect" is used to explain this in- efficiency. In the matrix the parent molecule is "locked" in the host lattice. If the molecule is dissociated by the radiation one of the fragments must have sufficient energy to penetrate the wall of the site it occupies and attain a different site. If this does not occur, generally the fragments will recombine and no result will have been achieved by the photolysis of the parent molecules. This is so-called "cage effect". One way to overcome the "cage effect" is to photolyze during deposition. In this case, fragments from molecules photolyzed in the gas phase separ— ate before they are condensed on the sample window, and under these circumstances the "cage effect" operates to prevent recombination. If molecules are photolyzed as they 85 condense they can also separate on the surface in the few instants before the matrix becomes truly rigid. In the diazopropyne case, since the fragment N2 is quite stable, ig_§itg_photolysis will be relatively efficient. As mentioned in the Experimental section of Chapter I, a Malaker Corp. Cryomite VII-C closed cycle helium refriger— ator was used, which was fitted with a spray-on copper window holder and an inverted "T"-shaped brass outer jacket (see detail on Fig. 3, Chapter I) for the present infrared matrix isolation and photolysis study. In the diazopropyne matrix isolation experiment the matrix-to-sample ratio M/S was about 200 to 1, and the deposition rate was 1-3 m mole per hour. This deposition rate was controlled with a needle valve and was chosen to give minimum scattering and maximum transparency. The deposition period was about 2-3 hours for each experiment. The entire sample was deposited at 30°K before initiating photolysis. The i2_§itg photolysis was carried out using a Bausch and Lomb 33-86—20 xenon light source with 150 W output, and in some experiments a Dow Corning CS051 filter was placed in front of the xenon lamp to cut off wave lengths of less than 3500 A. The purpose of applying a Pyrex glass filter was to distinguish the various photolysis products which were induced by photons with different energies. After taking a few spectra, the Pyrex glass filter was removed and a few more spectra were taken after unfiltered photolysis. 86 An infrared study was performed over the range of 200 to 4000 cm.1 with a purged Perkin Elmer 225 grating infra- red spectrophotometer which has resolution of better than 1 cm.1 above 450 cm"1 and better than 2 cm”1 below 450 cm-1. The reported frequencies should be accurate to within :2 cm-l. The matrix gases used were research grade krypton and xenon from the Matheson Gas Company. The propargylene radical was also produced for ESR study by trapping gaseous dia20pr0pyne in polychlorotrifluoroethylene at liquid nitro- gen temperature and then photolyzing with the xenon discharge lamp. The ESR spectrum was obtained on a Varian V-4500-108 X-band spectrometer. Results and Discussion Based on the conclusion of Skell's ESR study (7), that the propargylene radical is linear, the author assumes D03h symmetry for C3H2. Thus it should have ten normal modes, of which three are degenerate, so there should be seven distinct fundamental frequencies for C3H2. Two belong to f" 4; symmetry, two to 2: symmetry, one to fig symmetry and two to ”u symmetry. The 2: and Wu vibrations are infrared active and the 2; and Hg vibrations are Raman active. However, for the monodeutero propargylene radical, C3HD, the symmetry is lowered to Coov’ In this case four fundamentals belong to 2+ symmetry and three to w sym- metry: all showing both infrared and Raman activity. The infrared spectra of diazopropyne and of mixed deuterated 87 diazopropyne in krypton matrices after photolysis with light from a xenon discharge lamp are shown in Figure 14. The frequencies of new absorption bands occurring upon photolysis and disappearing upon annealing in these matrix spectra are shown in Table XIV. The identification of the absorption bands due to photolysis products, in particular free radicals is achi- eved by observing the growth of the new absorption bands upon continued photolysis and the decrease or disappearance of these new absorption bands upon controlled annealing. The annealing softens the matrix and permits the diffusion of the isolated reactive Species. The absorption bands due to the same reactive Species should behave in the same fashion (change of band intensity) upon photolysis and annealing. Furthermore, photolysis through various filters may help distinguish different photolysis products. Bands at 3285, 3279, 2140, 408, 402, 259, and 249 00" are suggested as being due to the C3H2 radical. The growth rate of these bands is effectively unchanged when the Pyrex glass filter is placed in front of the xenon photolysis lamp. The species responsible for these absorptions is thus formed by radiation with wavelength longer than 3500 A. This is in agreement with the probable mechanism of the formation of the C3H2 radical, since the formation of C3H2 requires only enough energy to break the very labile N2 group away from C3H3N3. The formation of other possible free radicals would probably require higher energy radiation. 88 .0000 0:0 ammo .mm000c0 0000000500 000 0:0 0000000 0c0awm000000 00 050 000 #0050 0 £003 00x00e 00:00 059 .000000 mmmam X00mm £003 mflm>000onm 00000 0z0aao 0:0 .0zmoao .0zomno .02000o mo 0000x0s 0 .000000 mm0Hm x00>m 050:003 mwm>000osm 000mm «Zumau .0cmmoumou000 00000000|x00008 00N>HO00£Q mo Edu0o0mm 0000uwcH .m .4 .00 0Hsm0m 89 .00 00:00m Tano— a. J— o— _ m— .__. 3))??{0 {J .. ON v« on an _ _ _ S x. _ i. . . . < _ x . . \.(\.\L (3755‘ ‘0 _ _ . x a <0 x‘ \ 0n .on -ON .00 IOQ ownuwwm 1 031100010on 90 Table XIV. Bands occurring upon photolysis of diazopropyne which disappear upon annealing. C3H2N2 hv > C3H2N2. C3HDN2 hv > C3DHN2, canznz Frequencies (cm-1) Frequencies (cm-1) . *3291 m *3285 (s) *3285 (s) *3279 (w,sh) *3270 w *3260 (m) *3260 m *2482 m *2471 s *2469 m a *2458 m a 2145 m 2145 m *2140 (w) *2140 vw *2115 vw 1747 (m) 1747 w), 1745 (vw) 1264 vw) 1150 w,sh) 1150 w,sh) 1146 m) 1146 w 754 w 754 w 738 w 738 w 581 m 581 m 550 m 550 m * 416 w * 408 (s) * 408 s * 402 sh) * 402 s * 392 s * 386 w,sh) * 375 w; * 259 (s) * 259 s * 257 sh) * 257 sh; * 254 sh * 249 m * 249 s; * 247 (m) * 247 s *The bands which behave in same fashion are due to C3H2 or its deuterium isouxes C3D2 and C3HD. 3Due to CO monomer. ‘1' II')‘ 'I‘ ll" I'll 91 These absorption bands show corresponding intensity behavior upon photolysis and annealing, and the suggested assignment receives support from the deuterium isotopic shift study and the normal coordinate analysis. Assignment of C3H2 Fundamentals Assuming that propargylene is a linear, symmetric molecule, the infrared spectrum should exhibit four funda- mentals: two each in the z: and Wu symmetry classes. These are designated v3, v4, v6, and v7 according to the accepted numbering scheme (18). As in Chapter I, only fre- quencies observed in krypton matrices will be discussed in the text. V3 Bands at 3285, 3279, and 3260 cm.1 are observed in the photolyzed diazopropyne matrix spectrum Figure 14. The 3285 cm.1 band is the strongest among them, 3279 cm-1 is a very weak shoulder, and the 3260 cm”1 band is rela- tively weak. The weak shoulder at 3279 cm“1 is possibly due to a "matrix site effect" (the prOpargylene radicals occupying two different kinds of sites in the host matrix lattice), since its frequency is quite close to 3285 cm-1 and its intensity is so weak. The band at 3260 cm-1 may be due to the perturbation of the prOpargylene radical by a single N2, also a photolysis product, located on a site -1 adjacent to the propargylene radical (35). The 3285 cm 92 band is within the range of frequencies for an acetylenic C-H stretching motion, and this motion is expected for C3H2 since the outer carbon atoms have sp hybridization char— acter. In the photolyzed mixed deuterated diazopropyne matrix spectrum (Fig. 14), two sets of related bands are observed: 3291, 3285, 3270, 3260 cm“1 and 2482, 2471, 2469, 2458 cm-1. In the first set, the four bands appear as two pairs, at 3285 and 3260 cm"1 and at 3291 and 3270 cm-1, the lower frequency in each pair may be due to the perturbation by N2. The first pair is obviously due to C3H2, and the second doublet represents the correSponding C-H motion in C3HD. Again, the stronger component (3291 cm—1) was chosen as the vibrational frequency for the normal coordinate analysis. The same situation is encountered in the C-D stretching region. The pair of bands at 2471 and 2458 cm-1 grows together and is assigned to C3HD. The 2471 cm"1 band is normally the strongest in this set. One would expect more C3HD formed than C3D2 in the photolysis of the deuterated mixture, and relative intensities of par- ent molecule absorptions support this expectation. The pair of bands at 2482 and 2469 cm“1 grow in the same fashion, and are due to C3D2. Again, the 2469 and 2458 cm.1 bands may be due to the perturbation by N2. Thus, the bands at 3291 and 3285 cm"1 are respectively assigned to the C-H stretch (VI) in C3HD and the antisymmetric C-H stretch (v3) in C3H2. The bands at 2482 and 2471 cm—1 are assigned re- spectively to the antisymmetric C-D stretch (v3) in C3D2 93 and the C-D stretch (v2) in C3HD. These assignments re- ceive support from the normal coordinate calculation. V4 In a previous experimental investigation of carbon vapor condensed in an argon matrix at 40K, a band at 2040 cm.1 was observed, and assigned to the antisymmetric stretching mode in the C3 radical (9). Also bands due to the c=c antisymmetric stretch were observed at 1957, 1940 and 1921 cm”1 in the spectra of allene (C3H4) and the iso- topic allenes C3H2D2 and C3D4 respectively (13). On the basis of this information, one would expect that the c=C antisymmetric stretching vibrations should fall in the neighborhood of 2000 cm.1 for the propargylene radical and its deuterated isotopes, since the carbon-carbon bonding situation is similar in all these cases. When matrix-isolated CaHzNa is photolyzed with near ultraviolet radiation (3500 R) a band is observed at 2140 cm-1 in the infrared spectrum, Figure 15. The intensity of this band is difficult to follow, since this region is complicated by absorption from monomeric and aggregated CO (36). A weak band at 2115 cm-1, Figure 16, remains in the matrix infrared spectrum of the deuterated diazopropyne mixture after complete photolysis of the parent molecules. A weak shoulder at about 2065 cm"1 is also sometimes ob- served on a strong parent molecule absorption in the latter experiments. These bands appear to grow upon continued Figure 15. A. B. C. D. 94 Infrared spectra of photolyzed diazopropyne in krypton matrix (2150 - 2000 cm‘l). Before photolysis After photolysis After photolysis and partial annealing at 600K EXpanded scale of B Expanded scale of C \A 0f if I I I I I I | | I l-l-I-l-l-l-l-I-I-l-I-I-I-I-I-I-I V E I I | I I I I I I I I 2I50 2I20 2150 2120 -I-I-I-l-I-I-I-I-I-I-I-II. -I-I-I-I-I-l-l-I-l-I-I-I! 'Il .ll-I Ill! 2150 2000 Figure 15. 96 Figure 16. Infrared spectra of photolyzed mixed diazo— propyne and deuterated diazopropyne in krypton matrix (2150 - 2000 cm’l). A. Before photolysis B. After photolysis C. After photolysis and partial annealing at 60°K 2150 I 2000 l 2150 97 I I 2000 Figure 16. 98 photolysis, and disappear upon annealing, and they are therefore attributed to the isotOpic propargylene radicals. The intensities of these absorptions are extremely weak in comparison to the C-H and C-D stretching vibrations. On the basis of the normal coordinate calculation, the bands at 2140 (v4) and 2115 cm-1 (v3) are c=C antisymmetric stretching vibrations in C3H2 and C3HD. The absorption near 2065 cm.1 is less well established. However, the calculated frequency for the antisymmetric c=c stretch of C3D2 is 2064 cm-1, so one may tentatively assign the shoulder at 2065 cm—1 to v4 of perdeuteropropargylene. A very weak band at 1264 cm-1 was also observed in the infrared Spectrum of mixed deuterated propargylene radical. It grows and disappears respectively upon photolysis and annealing. On the basis of normal coordinate calculation the band at 1264 cm.1 is assigned to v4 of C3HD, the "symmetric stretching" vibration. Wu symmetry class V6 A strong band at 408 cm.1 and a medium intensity band at 402 cm"1 are observed in the photolyzed diazopropyne matrix spectrum, the latter is perhaps due to the perturba- tion of C3H2 by N2. After photolysis of mixed deuterated diazopropyne matrices bands at 416, 408, 402, 392, 386, and 375 cm.1 are observed. It should be noted that Since C3HD 99 has CCIDV symmetry, the fig mode in Dooh changes into a W mode and becomes infrared active. Thus one more band is expected to be observed in C3HD than in C3H2 or C3D2 in the 0 symmetry class. The 416 and 375 cm"1 bands grow and disappear respec— tively upon photolysis and annealing in the same fashion as the 3291 and 2471 cm"1 bands (v1 and v2 in C3HD), and the behavior of the 392 and 386 cm”1 bands parallel that of the 2482 cm.1 band (v3 in C3D2) with respect to varying isotopic distributions in all experiments. According to the normal coordinate calculation, the deuterium isotopic shift, the relative intensity behavior, and a comparison of frequencies with those of related mole- cules [C382 (11) and C3N2 (12) have v6 frequencies of 462 and 392 cm.1 respectively], the 408 cm"1 band in the photo- lyzed diaZOpropyne matrix spectrum is assigned to v6 for the C3H2 radical and the 416, 375, and 392 cm.1 bands in the photolyzed mixed deuterated diazopropyne matrix Spec- trum are assigned to v5 and v6 for the C3HD radical and v6 for the C3D2 radical respectively. Because of the change in symmetry, and the consequent different contribu- tion of the internal modes to the normal modes of vibration, the C3HD frequencies (v5 and v6) span tflua values of v6 for C3H2 and C3D2. This complication also affects the rela- tive intensities of these motions. 100 v7 There are few bands observed below 300 cm.1 in the photolyzed diazopropyne matrix spectrum. A strong, poorly resolved doublet is noted at 259 and 257 cm-1, along with medium intensity feature at 249 and 247 cm-1. (The latter pair may be due to the perturbation of C3H2 by N2.) The intensities of the bands at 259 and 247 cm"1 are approxi- mately equal in the infrared Spectrum of the photolyzed deuterated mixture, in contrast with their counterparts in photolyzed diazopropyne matrix spectrum. The increase in intensity at 247 cm-1 in the deuterated spectrum is probably due to the presence of some other photolysis product since the 247 cm.1 band does not disappear proportionately with 259 cm.1 band upon annealing. The 259 cm.1 band behaves in the same fashion upon photolysis and annealing as the 3285 and 408 cm.1 bands. Thus the 259 cm.1 band is assigned to v7 for C3H2. The deuterium shifts of this absorption in C3HD and C3D2 were not observed. According to the normal coordinate calculation, v7 for C3D2 is around 200 cm-1 and v7 for C3HD is around 220 cm_1; the former is probably beyond the effective low-frequency limit of the infrared instrument, and the latter is perhaps too weak to be ob- served. (The other bending modes of C3HD are quite weak.) According to the normal coordinate analysis, the v6 and v7 normal modes of C3H2 correspond to a mixture of CCH and CCC bends. It is expected that the electron 101 density around c=C=C and c=C-H in C3H2 will change when the C3H2 radical is bent. In other words, there will be interactions between c-C-H and c=c=c bends. Therefore, the mixing of these two bending vibrations in the normal modes is expected, and it is not appropriate to assign indi- vidual group frequencies to v6 and v7. Other Bands Bands at 1747, 1146, 754, 738, 581, and 550 cm“1 are also observed both in infrared spectra of photolyzed matrix isolated diaZOprOpyne and mixed deuterated diaZOpropyne. All of these bands behave like reactive Species. The in— tensities of the 1747, 1146, 754, and 738 cm"1 bands are quite strong when the Pyrex glass filter is not used in the photolysis, but weak in filtered photolysis. The intensi- ties of the 581 and 550 cm-1 bands do not grow and disap- pear, in photolysis and annealing experiment, proportionately with bands assigned to C3H2. Furthermore, there are no cor- responding deuterium isotope shifts observed for any of the bands discussed in this section. Thus one must conclude that these bands are not due to C3H2. Thus a further study of these bands is suggested. 102 Vibrational Potential Function and Related Properties of Pr0pargylene Normal Coordinate Analysis Due to the absence of the six Raman active fundamentals of the C3H2 and C3D2 radicals, a complete normal coordinate analysis cannot be carried out for propargylene. However, there are still enough frequencies from infrared active fundamentals of C3H2, C3D2, and C3HD to carry out the normal coordinate calculation and obtain meaningful valence force field potential constants. The calculation is carried out by the same method used for diazopropyne. The internal and symmetry coordinates for the prOpargylene radical are shown respectively in Figure 17 and Table XV. The skeletal structure parameters rCH = 1.07 X and rCC = 1.30 R are taken from the related molecule allene (13). The initial force constants were transferred from the related molecules allene (13) and C3 (9). Twelve frequencies for the isotopes C3H2, C3D2, and C3HD radicals were used as input data for the force constant calculation. These twelve frequencies were fit by eight independent force constants, of which four interaction and F ) cc,CH' cc,cc' CCH,CCH' CCC,CCH were important in the analysis. The calculated results are force constants (F F F compared to experiment in Table XVI, together with the po- tential energy distribution among the internal coordinates 103 .0000000 0c0amm00moum 050 How m000¢00uooo 002000CH .00 005m0m 104 .00 00500m 105 Table XV. Symmetry coordinates for the propargylene radical. 2" 9 SI = ARI + ARZ 82 : Arl + Arz Z'I' 11 S4 = Arl - Arz 7T 9 S5 = A51 ‘ A52 77' 11 36 = A0 106 Table XVI. Normal coordinate analysis of the propargylene radical. Frequencies (cm-1) Primary Contributors v Obs. Calc. Av (PED) C3H2 2; v1 -- 3288.0 -- C-H (100) v2 -- 1297.2 -- c=c 92, g:g (7) 2: v3 3295.0 3286.7 -1.7 C-H (100) v4 2140.0 2149.0 -9.0 c=c 106, g:g] (-7) 0 v5 -- 430.9 -- 5 + 86 (100) 03 v6 408.0 409.2 -1.2 s - as (52), n-Bn (48) v7 259.0 257.7 1.3 6-66 (48), n +Eq (52) 0102 29 v1 -- 2445.8 -- C-D (96) - 02 —- 1233.3 -- c=c 87, g;g (16) 2: v3 2482.0 2483.2 -1.1 C-D (81), c=c (26) v4 (2065.0) 2063.6 -- c=c (82), C-D (19) 0 v5 —— 336.2 -- 5+55 (100) F3 v6 392.0 390.8 1.2 q-8n(67), B-BB (33) C3HD 2* v1 3291.0 3287.7 3.3 C-H (100) v2 2471.0 2466.3 4.7 C—D (87), c=c (18) v3 2115.0 2105.3 9.7 c=c (92), C-H (11) v, 1264.0 1264.3 -0.3 c=c (90), c c C C1 (7) 0 v5 416.0 416.0 0.0 6-66 (75), q-fin (24) v6 375.0 375.1 -0.1 B+BB (61), n-Bn (38) V7 -- 219.5 -— 6-66 (63). n+fin (36) Kc-H ch KCCH Hccc FCH Fcc FCCH Fccc cc CC CCH CCH 5.93 12.05 0.078 0.246 0.564 0.868 -0.020 0.006 ( ) uncertain band not used in force constant calculation. Unitszlo5 dyn/cm for stretching force constants, 10-ll erg/rad2 for bending, and 10 3 dyn/rad for bend-stretch interactions. 107 for each normal mode (only those contributing more than 10% to each normal mode are listed), and the final calculated force constants. The fit of the calculated frequencies with those observed is excellent, with average error 2.7 cm"1 or 0.11%. Vibrational Potential Function of Propargylene The C-H stretching force constant, KC-H = 5.93 x 105 dyne/cm, is close to that of propargyl halides (5.96 x 105 dyne/cm) (14) and that of methylacetylene (5.85 x 105 dyne/cm) (19) a prototype acetylenic C-H stretching force constant. This indicates that the C-H bond in the propargylene radical has acetylenic character. The C-C stretching force constant, KC-C = 12.05 x 105 dyne/cm, of C3H2 is Slightly higher than that of C3H4 (10.08 x 105 dyne/cm) (13), Since in the latter case the outer carbon atoms are Sp2 hybridized,but in the former case the outer carbon atoms are Sp hybridized. The H-C=C bending force constant, HHCC = 0.078 x 10-11 erg/radz, is somewhat smaller than the H-CEC bending force '11 erg/radz) (14). constant of propargyl halides (0.14 x 10 However, the bonding situation is rather different in the present case. The unpaired spin density on the terminal carbon atoms would tend to more readily permit the rehybridi- zation which accompanies CCH bending, and thus lead to a lower bending force constant. 108 The C=C=C bending force constant, HCCC = 0.246 x 10-11 erg/radz, falls between that of allene (0.40 x 10.11 erg/radz) (13) and that of the first excited state of the C3 radical (0.11 x 10-11 erg/radz) (37). The C=C=C bending force constant is worth special discussion, Since it can be correlated with the electronic structures of some simple linear molecules with cummulated double bonds, such as C3, C3H2, C3H4, etc. Considering first the C=C=C bending force constant of the ground state of the C3 radical, and extremely low value (around 0.005 ‘11 x 10 erg/radz) is found (37). However, the same bending 11 erg/rad2 in the force constant increases to 0.11 x 10— first excited state of C3 radical. This may be understood in light of the suggested correlation between central car- bon atom electron density and ease of bending. The electron density in the Wu orbital of ground state C3 radical (03 0:), as calculated by Clementi and McLean (38), is primarily located on the central carbon atom and therefore would remain relatively unaffected by a change in geometry such as that of the bending motion. Thus an extremely small bending force constant is expected. The low bending force constant of C3 also receives support from a recent ab- initio calculation (39). The bending potential for this normal mode is quite anharmonic (near to a Square well with minimum at a central carbon angle around 180°), which would lead to a low bending frequency. 109 However a similar calculation for €02 (a: v: 4 W9) (40) Shows that the fig orbital has a "dumbell" shape. For the case of first excited state of the C3 radical (a: "3 ng), the promoted electron density is distributed primarily on the ends of the molecule. Undoubtedly, bending either molecule would increase electronic repulsion considerably, with a much greater effect for C02. Thus a relatively large bending force constant is expected. Following the same argument, the c=C:C bending force constant of the C3H2 radical (a: v; #2) would be between the C02 bending force constant (0.57 x 10.11 erg/radz) (18) and that of the first excited state of the C3 radical (0.11 x 10-11 erg/radz). Indeed, this prediction is borne out by the present normal coordinate analysis; the C3H2 radical -11 erg/radz. has a CCC bending force constant of 0.246 x 10 Although the stretching interaction force constants, FCH'CC = 0.564 x 105 dyne/cm and FCC'CC = 0.869 x 105 dyne/cm, contribute no more than 10% to the potential energy distribution of the related normal modes, the intro- duction of these two interaction force constants certainly improves the fit between the observed and calculated fre- quencies. The relatively large FCC,CC interaction force constant suggests the possible resonance structures H-C=c=é-H <—> H-E-CEc-H <—> H-CEc-E-H), similar to those which may be drawn for co2 (0=c:.0 <—> O-CEO <—> osc-o) (41). Introduction of the FCH CC interaction force constant I 110 improves the calculated frequencies in the 2: symmetry block a great deal. The bending interaction force constants, FCCH,CCH = 0.020 x 10.11 erg/rad2 and FCCC,CCH = 0.006 x 10_11 erg/radz, contribute significantly to the potential energy of v6, v7 (C3H2 and C3D2) and v5, v6, v7 (C3HD) respectively. Thus they too must be included in the normal coordinate calculation. Conclusion Only one C-H stretching frequency is observed in the infrared Spectrum of the C3H2 radical, which implies that the two hydrogen atoms in propargylene are equivalent. The acetylenic C-H stretching force constant and large C=C=C bending force constant for the molecule indicate that the electron density in the outermost valence orbital is located primarily on the terminal carbon atoms of the radical. The value of the calculated bending force constant, HCCC' is in line with the expected occupancy of this "dumbbell Shape" orbital, intermediate between that of C3 and C02. All of the observations in the infrared Spectra of the matrix- isolated isotopic prOpargylenes support the expected Da3h symmetry for this radical. REFERENCES 10. 11. 12. 13. 14. 15. REFERENCES W. Nernst, Z. Elektrochem. 24, 335 (1918). F. Paneth and W. Hofeditz, Chem. Ber. 62, 1335 (1929). E. W. R. Steacie, Atomic and Free Radical Reactions (Reinhold, New York) 2nd ed. (1954). E. Whittle, D. A. Dows, and G. C. Pimentel, J. Chem. Phys. 22, 1943 (1954). D. E. Milligan and M. E. Jacox, In Physical Chemistr , An Adganced Treatise, 4, Chap. 4 (Academic, New York , 1969 . C. B. Moore and G. C. Pimentel, J. Chem. Phys. 40, 342 (1964). R. A. Bernheim, R. J. Kempf, J. V. Gramas and P. S. Skell, J. Chem. Phys. 43, 196 (1965). A. J. Merer, Can. J. Phys. 43, 1795 (1965). W. Weltner, Jr. and D. McLeod, Jr., J. Chem. Phys. éé, 3096 (1966). W. H. Smith and G. E. Leroi, J. Chem. Phys. 45, 1767 (1966). W. H. Smith and G. E. Leroi, J. Chem. Phys. 45! 1778 (1966). W. H. Smith and G. E. Leroi, Spectrochim. Acta 25A, 1917 (1969). W. H. Fletcher and W. T. Thompson, J. Mol. Spectry. 22, 240 (1968). G. Zerbi and M. Gussoni, J. Chem. Phys. ‘41, 456 (1964). Organic Syntheses Collective Vol. 2, 165, 461 (John Wiley and Sons Inc., New York) (1944). 111 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 112 L. Andrews, In Annual Review of Physical Chemistry, 22, Chap. 5 (1971). C. C. Costain, J. Chem. Phys. 42, 1961 (1966). G. Herzberg, Molecular Spectra and Molecular Structure, Infrared and Raman Spectra of Polyatomic Molecules. Vol. 2, (D. Van Nostrand Co., Inc., Princeton, N.J.) (1966). B. L. Crawford, Jr., J. Chem. Phys. 2, 526 (1940). J. C. Evans and R. A. Nyquist, Spectrochim. Acta 12, 1153 (1963). R. A. Nyquist, T. L. Reder, F. F. Stee and G. J. Kallos, Spectrochim. Acta 27A, 897 (1971). P. Cossee and J. H. Schachtschneider, J. Chem. Phys. 44, 97 (1966). F. A. Miller, 0. Sala, P. Devlin, J. Overend, E. Lippert, W. Luder, H. Moser and J. Varchmin, Spectrochim. Acta 22, 1233 (1964). T. Fujiyama and T. Shimanouchi, Spectrochim. Acta 29, 1233 (1964). E. B. Wilson, J. C. Decius and P. C. Cross, Molecular Vibration (McGraw-Hill Book Co., Inc., New York). J. H. Schachtschneider, Shell Development Co., Tech. Rep. NO. 231-64 0 J. H. Schachtschneider, Shell Development Co., Tech. Repto NO. 57.65. W. J. Orville-Thomas, Chem. Rev. 21, 1179 (1957). W. T. Thompson and W. H. Fletcher, Spectrochim. Acta 22, 1907 (1967). R. Ettinger, J. Chem. Phys. 42, 1693 (1964). C. W. Bjork, N. C. Crai , R. A. Mitsch and J. Overend, J. Am. Chem. Soc. 21(6 , 1186 (1965). C. B. Moore and G. C. Pimentel, J. Chem. Phys. 22, 2816 (1963). F. Stitt, J. Chem. Phys. 2, 56 (1940). 34. 35. 36. 37. 38. 39. 40. 41. 113 J. L. Griggs, Jr., K. N. Rao, L. H. Jones, and R. M. Potter, J. Mol. Spectry. 12, 212 (1965). D. E. Milligan and M. E. Jacox, J. Chem. Phys. 41, 278 (1967). G. E. Leroi, G. E. Ewing and G. C. Pimentel, J. Chem. Phys. 49, 2298 (1964). L. Gausset, G. Herzberg, A. Lagerqvist and B. Rosen, Astrophys. J. 142, 45 (1965). E. Clementi and A. D. McLean, J. Chem. Phys. ‘22, 45 (1962). D. H. Liskow, C. F. Bender and H. F. Schaefer III, J. Chem. Phys. 22, 5075 (1972). A. D. McLean, J. Chem. Phys. ‘22, 1347 (1963). H. W. Thompson and J. W. Linnett, J. Chem. Soc. 138 (1937).