University of Southern California Dissertations and Theses

Photoelectron and ion imaging investigations of spectroscopy, photoionization, and photodissociation dynamics of diazomethane and diazirine

(USC Thesis Other)

Photoelectron and ion imaging investigations of spectroscopy, photoionization, and photodissociation dynamics of diazomethane and diazirine

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content PHOTOELECTRON AND ION IMAGING INVESTIGATIONS OF SPECTROSCOPY, PHOTOIONIZATION, AND PHOTODISSOCIATION DYNAMICS OF DIAZOMETHANE AND DIAZIRINE by Igor Fedorov ___________________________________________________________________ A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (CHEMISTRY) August 2009 Copyright 2009 Igor Fedorov ii Acknowledgements During my years at USC I have been fortunate to interact with a number of great people: faculty, postdoctoral fellows, graduate students, and staff members. First of all, I would like to thank my research advisor Professor Hanna Reisler, for whom I have the greatest admiration and respect. I am proud to have been a member of her research group, and I thank her for allowing me the opportunity to do my thesis work with her. I am grateful to her for the years of guidance throughout my time in graduate school, contagious enthusiasm, and ability to inspire me to do my best. The work presented in this dissertation is a result of teamwork, with contributions from many individuals. I wish to thank my former and current colleagues, members of our “imaging team”: Dr. Vladimir Dribinski, Dr. Aaron Potter, Dr. Andrei Demyanenko, Dr. Guosheng Li, Dr. Jessica Parr, Miss Blithe Casterline, Dr. Andrew Mollner, and Ms. Lee Ch’ng. I want to acknowledge Dr. Vladimir Dribinski for showing me the ropes of experimentation and for constant, fruitful interactions. Also, I wish to thank the former and current members of our “radical team”: Dr. Lin Feng, Dr. Jei Wei, Mr. Boris Karpichev, Ms. Laura Edwards, and Mr. Mikhail Ryazanov. I thank Mr. Mikhail Ryazanov for helping with problems in the lab and for useful scientific discussions. It was my pleasure to work in collaboration with Dr. Lucas Koziol and Prof. Anna Krylov who provided significant theoretical support. Dr. CJ Bigler Jones, Prof. Karl O. Christe and colleagues helped enormously on synthesis of the systems iii described in this dissertation. Without these productive collaborations the achievements of these research projects would have been far fewer. I am very grateful to Prof. Karl O. Christe for his helpful advice and thoughtful discussions about academic and everyday life. He has shown me that I am capable of more than I ever would have thought. I would also like to thank several other professors who have enlightened me academically with their constructive criticism. They are Professors Curt Wittig, Stephen Bradforth, Andrei Vilesov, Anna Krylov, and Robert Bau. I would like to mention a number of former and current postdocs and graduate students in Prof. Curt Wittig, Andrey Vilesov, and Anna Krylov’s groups with whom I worked side by side throughout these years: Dr. Daniil Stolyarov, Dr. Elena Polyakova, Dr. Joelle Underwood, Dr. Sergey Malyk, Dr. Samantha Hawkins, Dr. George Kumi, Dr. Lee-Ann Smith-Freeman, Mr. Anton Zadorozhnyy, Mr. Oscar Rebolledo-Mayoral, Mr. Chris Nemirow, Mr. Jordan Fine, Mr. William Schroeder, Dr. Zhou Lu, Dr. Dmitry Skvortsov, Dr. Kirill Kuyanov, Dr. Mikhail Slipchenko, Mr. Russell Sliter, Dr. Sergey Levchenko, Dr. Lyudmila Slipchenko, Dr. Piotr Pieniazek, Dr. Lucas Koziol, Mr. Vadim Mozhayskiy, and Dr. Kadir Diri. I would also like to thank my friends Dr. Filipp Baron, Dr. Askat Jailaubekov, and Mr. Sergey Zakharov. Thank you for the constant fruitful interactions, maintaining a friendly working atmosphere, and making USC and Los Angeles an exciting place to work and live. iv I would especially like to thank Dr. CJ Bigler Jones for helping me acclimate to American culture and whom I will always remember as my first American friend. Thanks to Dr. Vladimir Dribinski and Dr. Sergey Levchenko for teaching me experimental and theoretical techniques applicable in gas-phase molecular dynamics and to Dr. Andrew Mollner, Mr. Mikhail Ryazanov, and Ms. Laura Edwards for proofreading some parts of this dissertation. Additionally, I would like to acknowledge our support staff: Ms. Michele Dea, Mrs. Yuki Yabuta, and Mrs. Valerie Childress, who are our Administrative Assistants, for providing much needed assistance on paperwork and administrative aspects of my projects, and Mrs. Heather Meunier-Connor and Mrs. Danielle Hayes, Student Services Advisors, for help in meeting requirements for the graduate school. I would also like to thank the teams of Mr. Victor Jordan and Mr. Don Wiggins of the USC Machine Shop for technical support; Mr. Ross Lewis, the electronic technician, for fixing our electronics; Mr. Frank Neirtit, the Director of Chemical Instrumentation, for maintaining our computers; Mr. James Merritt, the former director of the Glassblowing Shop, for crafting my synthesis line; Mr. Phillip Sliwoski, the current director of the Glassblowing Shop, for other glassblowing support; Mr. Allan Kershaw, the Director of the Instrument Facilities, for teaching me about using IR spectrometer; and Dr. Elizabeth Erickson, the General Chemistry Course Administrator, from whom I learned many invaluable lessons and obtained skill in teaching undergraduates. v I am grateful to the faculty and staff of the Moscow Institute of Electronic Technology (Technical University). In particular I am thankful to Prof. Vasiliy Shevyakov, my scientific adviser, and to Prof. Vladimir Roschin and Prof. Sergey Gavrilov, my scientific consultants who set a foundation for my academic career and have helped me grow both as a scientist and as an individual. Last, but not least I must thank my family and old friends that have been an inspiration for me throughout my life. Their endless patience and great love and fascination helped me incalculably. I would not be where I am today without them. vi Table of Contents Acknowledgements………………………………………………………… ii List of Tables………………………………………………………………. ix List of Figures……………………………………………………………… xi Abstract…………………………………………………………………….. xviii Chapter 1. Introduction……………………………………………… 1 1.1 CH 2 N 2 ……………………………………………………... 1 1.2 Diazomethane……………………………………………... 1 1.2.1 Importance………………………………… 1 1.2.2 History…………………………………….. 3 1.2.3 Structural and Spectroscopic Properties of Ground State………………………………. 4 1.2.4 Electronic Structure……………………….. 6 1.2.5 Preparation………………………………... 9 1.3 Diazirine…………………………………………………... 9 1.3.1 Importance………………………………… 9 1.3.2 Structural and Spectroscopic Properties of Ground State………………………………. 11 1.3.3 Electronic Structure……………………….. 13 1.4 Photolysis of Diazomethane/Diazirine…………………… 14 1.5 Thermochemistry of Diazomethane and Diazirine……….. 17 1.6 Diazomethane and Diazirine as Precursors of Methylene Diradical…………………………………………………... 20 1.7 Major Goals and Specific Aims…….……………………... 21 1.8 Chapter 1 References……………………………………… 23 Chapter 2. Experimental Methods and Details…………………….. 30 2.1 Supersonic Molecular Beam Systems…………………….. 30 2.2 Preparation of Radicals in Molecular Beam Systems…….. 31 2.3 Photofragment Ion and Photoelectron Imaging…………... 33 2.4 Resonance Enhanced Multiphoton Ionization…………….. 38 2.5 Vacuum system, experimental arrangement and data analysis……………………………………………………. 40 2.6 Chemical Preparation of Diazomethane and Diazirine….... 46 2.7 Production of CH 2 radical………………………………… 56 2.8 Chapter 2 References……………………………………… 66 vii Chapter 3. Theoretical and Experimental Investigations of the Electronic Rydberg Ttates of Diazomethane: Assignments and State Interactions…………………….. 70 3.1 Introduction………………………………………………... 70 3.2 Experimental Details……………………………………… 74 3.3 Computational Details…………………………………….. 78 3.4 Computational Results…………………………………….. 79 3.4.1 Equilibrium geometries and ionization energies……………………………………. 79 3.4.2 Excited electronic states…………………... 80 3.5 Experimental Results……………………………………… 86 3.6 Discussion…………………………………………………. 94 3.6.1 The 2 1 A 2 (3p y ! !) and 2 1 B 1 (3p z ! !) Rydberg states and their interaction………. 94 3.6.2 The 3 1 A 1 (3p x ! !) state…………………... 101 3.7 Conclusions………………………………………………... 103 3.8 Chapter 3 References……………...………………………. 105 Chapter 4. Vibronic Structure and Ion Core Interactions In Rydberg States of Diazomethane: An Experimental and Theoretical Investigation……………..…………….. 108 4.1 Introduction………………………………………………... 108 4.2 Experimental Details……………………………………… 111 4.3 Experimental Results and Analysis……………………….. 114 4.3.1 REMPI spectra and eKE distributions…….. 114 4.3.2 Spectroscopic Analysis: Band origins, K-structure and state interactions…………. 123 4.4 Computational Details…………………………………….. 127 4.5 Discussion…………………………………………………. 133 4.5.1 Vibrational assignments for the 2 1 A 2 (3p y ) Rydberg state……………………………… 133 4.5.2 Vibrational assignments for the 1 2 B 1 ground-state cation……….……………….. 136 4.5.3 Vibrational assignments for the 1 2 B 1 ground-state cation………………………... 137 4.6 Summary……………………….………………………….. 142 4.7 Chapter 4 References……………………………………… 144 Chapter 5. Multiphoton Ionization and Dissociation of Diazirine: A Theoretical and Experimental Study………………… 148 5.1 Introduction………………………………………………... 148 5.2 Experimental Details……………………………………… 151 viii 5.3 Computational Studies of the Electronically Excited and Ionized States of Diazirine………………………………… 153 5.4 Experimental Results……………………………………… 160 5.5 Discussion…………………………………………………. 168 5.5.1 Excited States and Photoionization of Diazirine…………………………………... 168 5.5.2 Detection of Ionization Products……….…. 169 5.5.3 Pathways leading to CH(X) fragments……. 170 5.6 Summary…………………………………………………... 179 5.7 Chapter 5 References………………...……………………. 181 Chapter 6. Imaging Studies of Photochemistry and Spectroscopy of the Triplet Methylene………………………………… 187 6.1 Introduction………………………………………………... 187 6.2 Experimental Details……………………………………… 190 6.3 Experimental Results……………………………………… 191 6.4 Future Experiments………………………………………... 193 6.4.1 Detection of the 1 3 A 1 and 2 3 B 1 3s Rydberg States of CH 2 ……………………………… 193 6.4.2 Dissociation on the CH 2 1 3 A 1 (3s) Surface... 194 6.4.3 Dissociation on the 2 3 B 1 /1 3 A 2 Coupled Surfaces…………………………………… 195 6.5 Chapter 6 References……………………………………… 197 Bibliography …………………………………………………………….. 201 Appendix A. Technical Drawings of the Pulsed Pyrolysis Source…… 220 Appendix B. Deperturbation of Energy Levels of the 9 1 0 Transition to the 2 1 A 2 (3p y ) State and the Band Origin of the 2 1 B 1 (3p z ) Transition Using Two-Level Approximation…………………………………………… 235 ix List of Tables Table 1.1 Experimental vibrational frequencies of neutral ground state (1 1 A 1 ) of diazomethane. a …………………………………… 5 Table 1.2 Experimental vibrational frequencies of the ground state (1 1 A 1 ) of diazirine. a …………………………………………. 12 Table 1.3 Experimental and theoretical values of "H f o of diazomethane and diazirine (kcal/mol) at 0 and 298 K.......... 18 Table 2.1 Typical voltages applied to the electrostatic lens system, MCP detector and phosphor screen in different modes………………………………………………………... 45 Table 2.2 Values of heats of formation of molecules at 0 K……........... 57 Table 2.3 Values of heats of reaction at 0 K. a …………………………. 58 Table 3.1 Vertical excitation energies ("E vert , eV), oscillator strengths (f L ), dipole strengths (! 2 tr , a. u.), and changes in second dipole moment of charge distributions ("<R 2 >, (a.u.) 2 ) for the excited states of CH 2 N 2 at EOM-CCSD/ 6-311(3+,+)G*.…………………………………………..….. 82 Table 3.2 Calculated vertical ("E vert ) and adiabatic ("E ad ) excitation eneries and quantum defects (") and the corresponding experimental values…………………………………………. 85 Table 3.3 CCSD(T)/cc-pVTZ harmonic vibrational frequencies for cation ground state (1 2 B 1 )…………………………………… 99 Table 4.1 Transition energies and vibrational assignments a for the 2 1 A 2 (3p y ) ! 1 1 A 1 transition of CH 2 N 2 . b …………………… 116 Table 4.2 Transition energies and vibrational assignments a for the 2 1 A 2 (3p y ) ! 1 1 A 1 transition of CD 2 N 2 . b ………………….… 117 Table 4.3 Transition energies and vibrational assignments a for the 2 1 A 2 (3p y ) ! 1 1 A 1 transition of CHDN 2 . b ………………….. 118 x Table 4.4 The calculated equilibrium structures and nuclear repulsion energies for the ground state of the neutral and cation and the 3p Rydberg states of CH 2 N 2 . a ……................................... 129 Table 4.5 Transitions energies and vibrational frequencies of neutral ground state, 3p Rydberg states, and cation of CH 2 N 2 . a ……. 130 Table 4.6 Transitions energies and vibrational frequencies of neutral ground state, 3p Rydberg states, and cation of CD 2 N 2 . a ……. 131 Table 4.7 Transitions energies and vibrational frequencies of neutral ground state, 3p Rydberg states, and cation of CHDN 2 . a .….. 132 Table 5.1 Calculated equilibrium structures for the ground, 1 1 B 2 and 1 1 A 2 valence states of the neutral and the ground state of the cation………………………………………………………... 155 Table 5.2 Vertical excitation energies ("E vert , eV), oscillator strengths (f L )/ (oscillator strengths from 1 1 B 2 state), dipole strengths (! 2 tr , atomic units), and changes in second dipole moment of charge distributions ("<R 2 >, squared atomic units) for the excited states of c-CH 2 N 2 at EOM-CCSD/6-311(3+,+)G*. a ... 158 Table 5.3 Calculated values of "H f o of diazomethane and diazirine (kcal/mol)…………………………………………………… 172 xi List of Figures Figure 1.1 Compounds with the formula CH 2 N 2 ………………............ 2 Figure 1.2 Interatomic distances and bond angles in diazomethane [1,2]....................................................................................... 4 Figure 1.3 Two Lewis structures of diazomethane……………………. 6 Figure 1.4 Bond length and bond angles in diazirine [3,4]. The HCH and NCN planes are mutually orthogonal…..……………... 11 Figure 2.1 Two typical REMPI applications and the type of spectra that might result: (a) typical one color REMPI at the two- photon energy resonance (2 + 1 REMPI scheme); an ion signal is observed whenever a resonance is reached; (b) the wavelength is fixed to excite the two-photon transition and the kinetic energies of the resulting photoelectrons are analyzed……………………………………………………. 39 Figure 2.2 Schematic diagram of the experimental apparatus: 1 – piezo-electric nozzle; 2 – skimmers; 3 – electrostatic lens assembly; 4 – #-metal shielding; 5 – MCP/phosphor screen assembly..……………………... 41 Figure 2.3 Schematic diagram of electrostatic lens system in the interaction chamber, field free drift tube of TOF in detection chamber, and the detector assembly…………….. 42 Figure 2.4 Schematic representation of the all-glass vacuum line used to prepare diazomethane and diazirine which is comprised of three sections: a reactor fitted with a pressure-equalized addition funnel, two traps and a vacuum manifold………... 47 Figure 2.5 Typical IR spectrum of diazomethane in the 500$3500 cm -1 region, partial pressure ~ 35 Torr……….... 48 Figure 2.6 Typical IR spectrum of deuterated diazomethane in the 500$3500 cm -1 region, partial pressure ~ 35 Torr................ 49 Figure 2.7 Typical IR spectrum of partially deuterated diazomethane in the 500$3500 cm -1 region, partial pressure ~ 35 Torr…... 51 xii Figure 2.8 Typical UV spectra of diazomethane in (a) 190–270 nm region, partial pressure < 1 Torr; (b) 200–600 nm region, partial pressure ~ 35 Torr………………………………….. 52 Figure 2.9 Typical IR spectrum of diazirine in the 500–4000 cm -1 region, partial pressure ~ 15 Torr………………………….. 55 Figure 2.10 Typical UV spectrum of diazirine in 200–400 nm region, partial pressure ~ 75 Torr………………………………….. 55 Figure 2.11 (a) Schematic cross-section of the pulsed pyrolysis source: 1 – SiC tube; 2 – electrodes; 3 – SiC-tube holder; 4 – water-cooled block; 5 – plunger; 6 – sample inlet chamber; 7 – face plate of the sample inlet chamber; 8 – main chamber; 9 – face plate of the main chamber; 10 – piezoelectric disk translator; (b) Electrical circuit for the pulsed pyrolysis source……………………... 63 Figure 2.12 The diazomethane pyrolysis efficiency curve based on CH 2 N 2 + signal……………………………………………… 64 Figure 3.1 Molecular orbitals relevant to ground and excited electronic states of CH 2 N 2 …………………………………. 72 Figure 3.2 Left panel: Ground state equilibrium structures (Å and deg) of CH 2 N 2 for: the neutral (1 1 A 1 ) at CCSD(T)/cc-pVTZ (regular print) and B3LYP/6-311G(2df,p) (italics), and for the cation (1 2 B 1 (" ! !)) at CCSD/6-311G** (underlined). The corresponding nuclear repulsion energies are: 61.280112, 61.514227, and 61.118198 hartrees, respectively. Right panel: Excited state equilibrium structures for the 2 1 A 2 (3p y ! !), 2 1 B 1 (3p z ! !), and 3 1 A 1 (3p x ! !) states shown in normal, italics, and underlined print, respectively. CNN and HCNN refer to respective angle and dihedral angle for the C S 3 1 A 1 (3p x ! !) state. The corresponding nuclear repulsion energies are: 60.705257, 59.502297, and 60.715012 hartrees. Experimental parameters of ground state: CN length: 1.300 Å; NN length: 1.139 Å; CH length: 1.075 Å; HCH angle: 126.0 o …............................................. 79 xiii Figure 3.3 The bars show calculated vertical excitation energies of CH 2 N 2 . Allowed and forbidden transitions are indicated by filled and hollow bars, respectively……………………….. 81 Figure 3.4 Survey 2 + 1 REMPI spectrum in the 6.32!7.30 eV (51,000!58,900 cm -1 ) region. The empty bars indicate the calculated values of vertical excitation energies…………... 84 Figure 3.5 2 + 1 REMPI spectrum of diazomethane in the region of excitations to the 1 A 2 (3p y ) and 1 B 1 (3p z ) states obtained by measuring m/e = 42 as a function of excitation energy. The laser wavelength increment was 0.005 nm. See the text for details of the C!F bands……………..……………………. 88 Figure 3.6 Photoelectron image and the corresponding eKE distributions obtained at excitation wavelength # = 382.69 nm (2h$ = 52,262 cm -1 ; middle peak of band C)…………………………………………………….. 89 Figure 3.7 Photoelectron image and the corresponding eKE distributions obtained at excitation wavelength # = 380.75 nm (52,528 cm -1 ; middle peak of band D)…….. 90 Figure 3.8 Photoelectron image and the corresponding eKE distributions obtained at excitation wavelength # = 380.39 nm (52,577 cm -1 ; band E)……...……………… 91 Figure 3.9 Photoelectron image and the corresponding eKE distributions obtained at excitation wavelength # = 379.50 nm (52,700 cm -1 ; middle peak of band F)…...... 92 Figure 3.10 Photoelectron image and the corresponding eKE distributions obtained at excitation wavelength # = 351.09 nm (56,965 cm -1 )………………………………. 93 Figure 4.1 The two Lewis structures for diazomethane. The z axis is along CNN, the y axis is in the plane, perpendicular to CNN, and the x axis is out of plane………………………... 108 xiv Figure 4.2 2 + 1 REMPI spectra for (a) CH 2 N 2 (b) CD 2 N 2 (c) CHDN 2 following two-photon laser excitation at 51,750!54,900 cm -1 . An inset in (a) for CH 2 N 2 shows a 54,900!56,700 cm -1 spectrum magnified ten times, whereas an inset in (b) diplays the 54,5000!55,000 cm -1 range in x10 magnification…................................................ 115 Figure 4.3 The eKE distribution for CD 2 N 2 obtained from the photoelectron image at excitation wavelength # = 381.14 nm (2h$ = 52,477.7 cm -1 )……………………... 120 Figure 4.4 The eKE distributions for CD 2 N 2 obtained from photoelectron images at excitation wavelength (a) # = 380.93 nm (52,507 cm -1 ) and (b) # = 379.51 nm (2h$ = 52,700 cm -1 )………………………………………... 121 Figure 4.5 The eKE distributions for CHDN 2 obtained from photoelectron images at excitation wavelength (a) # = 380.81 nm (52,520 cm -1 ) and (b) # = 379.53 nm (52,697 cm -1 )……………………………………………… 122 Figure 4.6 Harmonic frequencies of the neutral and cation ground state of CH 2 N 2 compared to those of the 2 1 A 2 (3p y ! !) and 2 1 B 1 (3p z ! !) Rydberg excited states……………….. 140 Figure 5.1 Molecular orbitals relevant to ground and excited electronic states of c-CH 2 N 2 . The three-membered ring lies in the yz plane, with the z-axis coinciding with the C 2 symmetry axis....................................................................... 157 xv Figure 5.2 Left panel: Ground state equilibrium structures (Å and deg) of diazirine for: the neutral (1 1 A 1 ) at CCSD(T)/cc-pVTZ (normal print) and B3LYP/6-311G(2df,p) (italics) and for the cation, 1 2 B 1 (" ! n), at B3LYP/6-311G(2df,p) (underlined). The corresponding nuclear repulsion energies are: 64.158275, 64.295975, and 62.797366 hartrees, respectively. Right panel: Excited state equilibrium structures for the 1 1 B 2 (%* ! n) and 1 1 A 2 (%* ! % NN ) excited states at CCSD(T)/ 6-311G** shown in normal and italics, respectively. The corresponding nuclear repulsion energies are: 62.313876 and 61.730487 hartrees. Experimental parameters of the neutral ground state are: r C-N : 1.482 ± 0.003 Å; r N-N : 1.228 ± 0.003 Å; r C-H : 1.09 ± 0.02 Å; HCH: 117 ± 2 o [3]…………………. 159 Figure 5.3 Survey REMPI spectra of diazirine obtained by monitoring (a) m/e = 14 (CH 2 + ) and (b) m/e = 13 (CH + ) at wavelengths 305–327 nm with 0.8 and 0.3 mJ energies (40-cm f.l. lens) for (a) and (b), respectively. The spectrum in (b) is about x10 higher in intensity than the one in (a)………………..... 162 Figure 5.4 2 + 1 REMPI spectrum of the CH(X) fragment (m/e = 13). Assignments of rotational CH(X, N”) levels for the D 2 & (v’ = 2) ! ! X 2 & (v’’ = 0) transition are marked on top of the corresponding peaks. Lines marked “a” belong to a different band system; the line marked “b” points to the 2p3p 1 S 0 ! ! 2p 2 1 D 2 atomic carbon transition. Arrows mark peak wavelengths at which ion images of CH + have been taken………………………………………………….. 163 Figure 5.5 Fluence dependence of the CH + (m/e = 13) ion signal resulting from 2 + 1 REMPI through the CH D 2 & (v’ = 2, N” & 9) ! X 2 & (v’’ = 0, N” = 9) transition. The dissociation wavelength was 312.10 nm and the radiation was focused by a 40-cm f.l. lens. The signal depends on the n = 2.15 power of laser energy…………………………….. 164 Figure 5.6 In the top panel the image obtained in dissociation at 312.10 nm by monitoring CH(X, v" = 0, N” = 9) is shown. In the bottom panel, the c.m. translational energy distribution, P(E T ), of the CH(X) fragments is shown (right axis) as well as the recoil anisotropy parameters ' i (E T ) (left axis)………………………………………………………... 167 xvi Figure 5.7 In the top panel the image obtained in dissociation at 310.91 nm by monitoring CH(X, v" = 0, N” = 6 and 9), which is in the R-bandhead region, is shown. In the bottom panel, the c.m. translational energy distribution, (P(E T ), of the CH(X) fragments is shown (right axis) as well as the recoil anisotropy parameters ' i (E T ) (left axis)…………….. 167 Figure 6.1 Geometry and carbon atom orbital occupation in ground state CH 2 (1 3 B 1 ) [5,6]………………………………………. 189 Figure 6.2 2 + 1 REMPI spectrum of CH 2 (1 3 B 1 ) radicals in the region of excitations to the 3p state obtained by measuring m/e = 14 as a function of excitation energy at 305–317 nm (65,574–63,091 cm -1 ) and using a) diazomethane; b) diazirine as a precursors. The sharp peak at 313.48 nm belongs to to the intense 2p3p 1 S 0 ! ! 2p 2 1 D 2 transition of atomic carbon…………………………………………… 192 Figure A.1 Electrode. The dimensions are in mm [inches]..…………... 220 Figure A.2 Graphite split disk. The dimensions are in mm [inches].….. 220 Figure A.3 SiC-tube holder. The dimensions are in mm [inches]……... 221 Figure A.4 Water-cooled block. The dimensions are in mm [inches].… 222 Figure A.5 Sample inlet chamber. The dimensions are in mm [inches]. 224 Figure A.6 Face plate of the sample inlet chamber. The dimensions are in mm [inches].…….………………………………………. 225 Figure A.7 Main chamber. The dimensions are in mm [inches].…...…. 226 Figure A.8 Face plate of the main chamber. The dimensions are in mm [inches].……………..……………………………………... 228 Figure A.9 Piezoelectric disk translator. The dimensions are in mm [inches].………………….………………………………… 229 Figure A.10 Ultem cylinder. The dimensions are in mm [inches].……... 230 Figure A.11 Plunger. The dimensions are in mm [inches].……………... 231 xvii Figure A.12 Fasteners. The dimensions are in mm [inches].…………… 232 Figure A.13 ASA Flange. The dimensions are in mm [inches].………... 234 xviii Abstract The photophysics and photochemistry of two isomers of CH 2 N 2 , diazomethane and diazirine, was studied in molecular beam. Their pyrolysis was used to produce molecular beam of methylene radicals in the ground electronic state. The spectroscopy, photoionization, and photodissociation pathways of diazomethane and diazirine were investigated experimentally using a combination of 2 + 1 REMPI and velocity map imaging techniques. For three isotopologs of diazomethane, the 2 1 A 2 (3p y ! !) and 2 1 B 1 (3p z ! !) Rydberg states are of mostly pure Rydberg character, whereas the 3 1 A 1 (3p x ! !) state is mixed with the valence 2 1 A 1 (!*! !) state. Normal mode vibrational frequencies were obtained for the 2 1 A 2 (3p y ) Rydberg state and cation, and mixed levels of the 2 1 A 2 (3p y ) and 2 1 B 1 (3p z ) states of the three isotopologs were identified. The agreement between experiment and theory was very good allowing a full analysis of trends in structure and vibrational frequencies going from the neutral species to the excited Rydberg state, and the cation. In the experiments with diazirine, the weak one-photon absorption to the 1 1 B 2 state is immediately followed by more efficient absorption of another photon to reach the 1 1 A 2 state from which competition between ionization and fast dissociation takes place. Strong signals of CH + ions are also detected and assigned to 2 + 1 REMPI of CH fragments. Velocity map CH + images show that CH (X, v"=0, N") fragments are born with substantial translational energy indicating that they arise from absorption of two photons in diazirine. It is argued that two photon processes via the 1 1 B 2 intermediate state are very efficient in 304–325 nm xix wavelength range, leading predominantly to dissociation of diazirine from the 1 1 A 2 state. The most likely route to CH(X) formation is isomerization to isodiazirine followed by dissociation to CH + HN 2 . We recommend revisions of the heats of formation of diazomethane and diazirine to 67 ± 3 and 77 ± 3 kcal/mol, respectively. CH 2 radicals were produced at high concentration with low contamination in a molecular beam by pyrolysis of diazomethane and diazirine in a home-built pulsed pyrolysis source. The 3 CH 2 radicals were detected by 2 + 1 REMPI via the 3p Rydberg state. 1 Chapter 1. Introduction 1.1 CH 2 N 2 Consider first the series of compounds with the formula CH 2 N 2 ; diazomethane, 3H-diazirine (henceforth the term diazirine will be used without the 3H-prefix), cyanamide, isocyanamide, nitrilimine, carbodiimide, isodiazirine, and the carbene cyclodiiminomethylene, which may be a singlet or triplet (Figure 1.1) [1-5]. The first three of these isomers are well known as isolated compounds. Only the derivatives of next four coumpounds (isocyanamide, nitrilimine, carbodiimide, and isodiazirine) are known but the parent compounds are not. The last compound, a carbene, has not been observed. Theoretical studies of the geometries and relative stabilities of compounds with the formula CH 2 N 2 have been carried out by several investigators [6-8]. Diazomethane and diazirine are the most important isomers because they serve as major sources of methylene, which occupies a significant place in all branches of chemistry [1-5,9]. 1.2 Diazomethane 1.2.1 Importance Diazomethane, CH 2 N 2 , is the simplest diazoalkane and it plays an important role in various fields of chemistry [1,2,4,9-16]. It is widely used in organic synthesis for the conversion of acids into esters, alcohols and phenols into ethers, alkenes into cyclopropanes, etc. [1,2,4,9-16]. Diazomethane is used on a large scale in the pharmaceutical industry [17], and its derivatives have been used as 2 antimalarial [18,19], anticancer [20], and antibacterial agents [21]. Diazomethane plays an important role in many branches of chemistry as a convenient methylene (:CH 2 ) precursor [1,2,4,9-11]. It is also a good model for understanding the photochemistry of diazoalkanes. Its photochemistry is relevant to understanding the chemistry in atmospheres rich in N 2 and methane, such as in Titan, Triton, and Pluto [22]. Figure 1.1 Compounds with the formula CH 2 N 2 . In addition, diazomethane attracted the attention of organic, physical, and theoretical chemists because of interesting structure, bonding, and reactivity characteristics and also because this smallest organic diazo compound is isoelectronic with several other molecules and other compounds with the formula CH 2 N 2 . Diazomethane possesses 16 valence electrons and is isoelectronic with 3 ketene (H 2 CCO) and dinitrogen oxide (N 2 O), carbon dioxide (CO 2 ), hydrogen azide (HN 3 ), isocyanic acid (HNCO), etc., and of course, its structural isomers. 1.2.2 History Diazo compounds all contain the neutral (zwitteronic) group =N 2 attached to one atom of carbon. The beginning of diazo compound chemistry dates back to 1858 when Peter Griess discovered and identified the first aromatic diazo compound [23]. The first aliphatic diazo compound and first diazoalkene, ethyl diazoacetate, was isolated in 1883 by Theodor Curtius [24]. The first report of diazomethane was published in 1890 when Franchimont [25] observed a yellow gas after treating nitrosomethyl carbamates with alkai, and between 1894 and 1895 von Pechmann [26,27] established the formula CH 2 N 2 for diazomethane after investigating this reaction carefully. von Pechmann adopted a cyclic structure for diazomethane that was originally proposed by Curtius (1889) [28] for ethyl diazoacetate and other aliphatic diazo compounds. Angeli [29,30] and Thiele [31] advocated open structures and Langmuir [32,33], who pointed out the steric similarities between diazoalkenes, ketenes, and azides, preferred the open form. It took almost 52 years to answer the question whether the structure of diazomethane was linear or cyclic, until Boersch’s X-ray diffraction study in 1935 [34] defined the linear arrangement of the carbon atom and the two nitrogen atoms in diazomethane and related compounds such as ethyl diazoacetate [24]. In 1949, Ramsay recorded the infrared 4 (IR) spectrum of diazomethane and concluded that it can only be interpreted on the basis of linear formula [35]. 1.2.3 Structural and Spectroscopic Properties of the Ground State The microwave studies of Cox et al. [36] in 1958 and IR spectroscopic measurements of Moore and Pimentel [37,38] in 1964 determined atomic distances and angles in the ground state (Figure 1.2). Diazomethane belongs to the C 2v symmetry group, where the C 2 axis is chosen along the z axis, the y axis is in the plane of the molecule, and the x axis lies perpendicularly to the plane of the molecule. The molecule is an asymmetric top (A = 9.112, B = 0.377109, and C = 0.361759 cm -1 ). Figure 1.2 Interatomic distances and bond angles in diazomethane [36,37]. The IR spectra of diazomethane recorded by Moore and Pimentel, were characterized by a strong NN-stretching band (2,102 cm -1 in the gas phase, 2,075 cm -1 , and 2,096 cm -1 in Ar or N 2 matrices [37-39] (and references cited therein) typical for diazoalkenes. Table 1.1 summarizes the experimental vibrational frequencies of ground state (1 1 A 1 ) of diazomethane. 5 Table 1.1 Experimental vibrational frequencies of neutral ground state (1 1 A 1 ) of diazomethane. a Mode Assignment Symmetry Frequency, cm -1 ! " CH 2 symmetric stretching a 1 3077 ! # NN stretching a 1 2102 ! $ CH 2 symmetric bending a 1 1414 ! % CN stretching a 1 1170 ! & CNN bending (out-of-plane) b 1 564 ! ' CH 2 wagging b 1 406 ! ( CH 2 asymmetric stretching b 2 3185 ! ) CH 2 rocking b 2 1109 ! * CNN bending (in-plane) b 2 421 a Data from Ref. [37,39] (frequency accuracy is ± 2 cm -1 ). They also measured spectra of the three isotopologs of diazomethane: CHDN 2 , CD 2 N 2 , and CH 2 15 N 14 N. Fadini et al. [40] measured the CN and NN stretching bands of three isotopic analogs of diazomethane CH 2 14 N 2 , CH 2 14 N 15 N, and CH 2 15 N 14 N and found the following values: CN-stretching: 1,136, 1,112, and 1,170 cm -1 ; NN-stretching: 2,097, 2,075, and 2,073 cm -1 , respectively. More recently, Vogt et al. [41,42] recorded the NN stretching band (! # ) of gaseous diazomethane by means of an interferometer and a tunable diode laser spectrometer, with a resolution of 0.07 cm -1 . Khlifi et al. [43] measured the IR spectrum of diazomethane to determine or reassess the absolute intensities of the vibrational bands. Theoretical studies of the geometry and vibrational frequencies of 6 diazomethane show relatively good agreement with the experimental values [12,44- 50]. The bonding in diazomethane is usually represented by the two resonance structures shown in Figure 1.3. Moore and Pimentel [51] deduced that the structure in Figure 1.3a form predominates over the structure in Figure 1.3b on the basis of the estimated dipole moments for both structures 5.46 D and ! 6.24 D, respectively. The measured dipole moment of diazomethane is 1.50 ± 0.01 D [36,52], much smaller than the dipole moments implied by either structure in Figure 1.3. The measured value is explained with the fact that there is comparable contribution of both structures with oppositely oriented dipole moments which tend to cancel on mixing. Figure 1.3 Two Lewis structures of diazomethane. 1.2.4 Electronic Structure The absorption spectrum of diazomethane has been studied in the visible and near-UV regions [53-55], and with high resolution in the vacuum UV region [56,57]. At room temperature, diazomethane is a yellow gas with a boiling point of ! 23 o C. The color is caused by weak absorption below 475 nm first studied by Kirkbride and Norrish [53] and later by Brinton and Volman [54]. In the gas-phase 7 UV spectrum, the first absorption system (475!320 nm) starting at ~ 475 nm has a number of very diffuse bands which merge into a continuous spectrum at about 420 nm with a maximum at 395 nm (! <10 L·mol -1 ·cm -1 , where ! is the molar extinction coefficient). A second, much stronger, absorption system (265!200 nm) starts at 265 nm with a maximum at 215!217 nm (! > 10 4 L·mol -1 ·cm -1 ) and has diffuse structures at 230, 218, and 214 nm [53,54,58-61]. The absorption spectrum below 210 nm was first studied by Herzberg [56] and in much greater detail by Merer [57] (200"135 nm) for gaseous diazomethane and deuterated diazomethane. Merer observed in the absorption a series of perpendicular bands near 190 nm, which he characterized as having 1 B 1 and 1 B 2 symmetry [57]. Some had a resolved K structure, and for CD 2 N 2 , the J structure was also resolved. He assigned the bands in the region near 190 nm to transitions to three Rydberg states, most probably 3p, which he denoted as D( 1 B 1 ), E( 1 B 2 ), and F( 1 B 1 ). Merer also identified perturbations among some of these states, and attributed them to Coriolis coupling. A very clear Rydberg nd # $ series begins with a narrow group of bands at 167 nm and leads to a limit at 138 nm corresponding to the ionization energy (IE) of 8.999 eV. In addition, there are two intense and diffuse band systems, centered at 175 and 140 nm, which do not fit any Rydberg series. In general, the spectrum of diazomethane is similar to that of ketene (H 2 CCO) because these molecules are isoelectronic (16 valence electrons) and have similar geometrical structure. However, the diazomethane spectrum is shifted to longer wavelengths since the IE is lower compared to that of ketene (9.00 eV for diazomethane [57,61,62] vs. 8 9.67 eV for ketene [61]). Because the ionization energy (IE) of diazomethane is low, the 3s, 3p and 3d Rydberg states are located in the same region as the valence states. The ion’s low-lying excited states are at 14.13 and 15.13 eV, and the parent ion is quite stable [62]. Theoretical electronic structure studies show that the first absorption system (475!320 nm) involves excitation from the ground state, 1 1 A 1 , to the valence 1 1 A 2 ($ * % # $) state [63,64]. The second, stronger absorption system (265!200 nm) is assigned to the intense valence transition 2 1 A 1 #1 1 A 1 ("*# ") (5.53 eV vertically) [54,65]. The antibonding character of the target molecular orbital makes this state unbound, dissociating adiabatically to N 2 ( 1 # g + ) + CH 2 ( 1 A 1 ), in agreement with Herzberg’s observations [56,61]. As of yet, the Rydberg states have not been fully characterized either experimentally or theoretically. An earlier theoretical study estimated them to lie vertically at 5.89 (3s), 6.65!6.87 (3p) and 7.48–7.68 (3d) eV [63]. However, the proximity of valence states can lead to Rydberg-valence interactions, with core electrons, giving rise to valence-Rydberg character and affecting quantum defects. A more recent study suggested that all states at 5.31!7.29 eV contain some Rydberg character, and that the 2 1 A 1 ("* # ") state, found at 5.53 eV vertically, is mostly valence and has the largest oscillator strength [64]. In the spectrum obtained by Merer [57], the groups of diffuse bands near 175 and 140 nm, which do not belong to any Rydberg series, may involve mixed states containing large contributions of valence "* # " excitations [57,64]. 9 1.2.5 Preparation The most common and convenient approach of diazomethane synthesis is by base-catalyzed decomposition of N-methyl-N-nitroso amines. Currently, the majority of chemists use one of two commercially available precursors, N-methyl- N-nitro-N-nitrosoguanidine (MNNG) and N-methyl-N-nitroso-p- toluenesulfonamide (Diazald) [37-39,43,51,57,62,66,67]. Specifically labeled Diazalds for the generation of CD 2 N 2 , 13 CH 2 N 2 , and 13 CD 2 N 2 are also commercially available. Diazomethane must be prepared very carefully for two reasons. First, it is extremely toxic. The main danger is due to the fact that one may work with it for some time without noticeable effect, but later symptoms similar to asthma develop, followed by allergic oversensitivity. Second, diazomethane is explosive. Sharp edges facilitate the explosive decomposiotion of diazomethane, so the ends of glass tubes should be rounded in a flame, no ground-glass joints must be used, etc. 1.3 Diazirine 1.3.1 Importance of Diazirine As mentioned above, a cyclic structure for diazomethane was originally adopted by von Pechmann in 1894 [26]. 3H-diazirine, which is a three-membered ring compound containing two equivalent nitrogen nuclei, i.e. the cyclic structure, was first prepared only in 1960 by Paulsen [68] and in the following year independently by Schmitz and Ohme [69-71]. Subsequently in 1962 Graham 10 reported a different method of synthesis of diazirine [72]. All methods are completely different from diazomethane synthesis. Diazirine is the most interesting isomer of diazomethane, and its chemistry has been the subject of increasing interest from several branches of chemistry. The most outstanding property of diazirines is their ability, compared to diazomethanes, in forming carbenes by photolysis or pyrolysis. Unlike their isomeric linear diazo compounds, diazirines are remarkably stable towards organic and inorganic reagents (uncreative with alkalines and strong acids) although they are thermally unstable. Therefore, diazirines are the preferred precursors compared to diazo compounds. Diazirines also find applications in biological and synthetic macromolecular systems as photoaffinity reagents to label receptors (biological systems which bind ligands) and as reagents for investigating the organization of biological membranes [73,74]. The structural uniqueness, bonding, and reactivity of diazirines have also attracted the attention of physical and theoretical chemists. Rearrangement of diazoalkanes into diazirines and vice versa can be initiated by UV radiation, although it is always accompanied by N 2 extrusion [75]. Diazirine is also a good model for understanding the photochemistry of substituted diazirines. For synthesis, most researchers follow the procedure of Shmitz and Ohme [69-71]. There are several reports of the explosive nature of diazirines. Shmitz and Ohme reported that it exploded at ! 40 o C but Graham reported handling it routinely as a gas and neat liquid, although it exploded once upon warming in a liquid nitrogen trap. He also 11 reported several explosions when air was admitted to the gaseous sample. Beacause diazirine is explosive it should always be handled in small quantities and with caution. To date more than 250 diazirines have been prepared and their chemistry has been the subject of several reviews [73,74]. 1.3.2 Structural and Spectroscopic Properties The structure of diazirine was first elucidated in 1962 by Pierce and Dobyns by using microwave spectroscopy in the frequency region 7.8!41 GHz [76]. The bond distances and angles in diazirine are shown in Figure 1.4. Diazirine, like diazomethane, belongs to the C 2v symmetry group, where the C 2 axis is chosen along the z axis and the N=N bond is parallel to the y axis. The CH 2 group lies in the xz plane. The molecule is an asymmetric top (A = 1.366, B = 0.789, and C = 0.558 cm -1 ). 1 H NMR [78] and IR [79-86] results are consistent with this structure. The measured dipole moment of diazirine is 1.56 ± 0.06 D [76]. Graham [72] reported the first gas-phase IR spectrum in 1962. In 1964 Ettinger [87] published IR spectra of diazirine, d 1 -diazirine and d 2 -diazirine recorded with Figure 1.4 Bond length and bond angles in diazirine [76,77]. The HCH and NCN planes are mutually orthogonal. 12 higher resolution, as well as a nearly complete assignment of the fundamental vibrations together with the first normal mode calculation. The high-resolution IR spectrum was assigned by Winnewisser’s group [81,82]. The experimental vibrational frequencies of the neutral ground state (1 1 A 1 ) of diazirine are summarized in Table 1.2. The calculated equilibrium geometries and vibrational frequencies of diazirine are in a good agreement with the experimental values [88]. Table 1.2 Experimental vibrational frequencies of the ground state (1 1 A 1 ) of diazirine. a Mode Assignment Symmetry Frequency, cm -1 ! " CH 2 symmetric stretching a 1 3023 ! # NN stretching a 1 1623 ! $ CH 2 symmetric bending a 1 1459 ! % CNN symmetric stretching a 1 992 ! & CH 2 twisting a 2 964 ! ' CH 2 asymmetric stretching b 1 3132 ! ( CH 2 rocking b 1 1125 ! ) CH 2 wagging b 2 967 ! * CNN asymmetric stretching b 2 807 a Data from Ref. [5,81,82] (frequency accuracy is ± 2 cm -1 ). 1.2.3 Electronic Structure Diazirine is a colorless gas with a boiling point of ! 14 o C. All diazirines display characteristic absorption with the longest wavelength band at 350"310 nm (! ~ 250 L·mol -1 ·cm -1 ) and a short wavelength band with a tail starting at about 13 250 nm and a peak below 200 nm. This absorption is far removed from the absorption maxima of proteins (280 nm) and nucleic acids (260 nm), and in this regard diazirines are excellent labeling reagents. The absorption spectrum of gaseous diazirine reported by Graham absorption at ~ 324.5 nm and shows many sharp regularly spaced peaks [72]. The molar extinction coefficient at 308.5 nm is 176 L·mol -1 ·cm -1 . Robertson and Merritt recorded the electronic absorption spectra of diazirine, 15 N 1 -diazirine, d 1 -diazirine and d 2 -diazirine and assigned this band to the 1 B 2 # 1 A 1 ($* # n) system with a band origin at 31,187 cm -1 [89-91]. The sharp peaks were characterized by a strong CN symmetric stretch (! % ) progression with 797 cm -1 intervals and another progression as a combination of CH 2 twisting (! & ) and CNN asymmetric stretching (! * ) with 848 cm -1 intervals. The short wavelength band (in the VUV) shows an intense, structureless band at ~ 145"185 nm with a maximum at ~ 157 nm, and a second band at ~ 20"143 nm with a maximum at 127 nm [92]. The excited electronic states of diazirine have not been fully characterized either experimentally or theoretically. In an earlier theoretical study, Han et al. calculated the singlet and triplet excited electronic states but did not fully characterize them [93]. These states have also been calculated by Arenas et al. [94] The lowest ones are of valence character. The S 1 state is of B 2 symmetry (in the axis chosen here) and the observed transition whose origin is at ~ 323 nm was assigned as $* # n. The second and third states, S 2 (B 1 ) and S 3 (A 2 ), correspond to $* # $ and $* # % transitions, respectively, with absorption at ~ 200 nm. In these calculations, only S 3 was characterized as a bound 14 state. The vertical (adiabatic) ionization energy (IE) of diazirine was determined experimentally at 10.75 eV (10.3 eV) and the ion’s low-lying excited states at 13.25 eV (12.8 eV) and 14.15 eV (14.15 eV) [95]. 1.4 Photolysis of Diazomethane/Diazirine The photolysis of diazirines was investigated intensively in recent years [5,9,96,97]. In general, the excited electronic state of the precursor can decay by at least four competing pathways: i) fluorescence; ii) intersystem crossing via a conical intersection with production of triplet carbene; iii) production of singlet carbene; and iv) diazirine & diazoalkane isomerization. In 1980 Erni and Khorana [98] in their elegant work showed that the photochemical isomerization of 6,6,8,8- tetrafluoro-7-diazotridecane (diazoalkane) into the 3,3-bis(1,1-difluorohexyl) diazirine required light of wavelength ' = 410 nm, whereas the reverse isomerization is possible only with light of ' = 310 nm. Moore and Pimentel [67] photolysed matrix isolated diazirine, d 1 -diazirine and d 2 -diazirine in solid N 2 . Their experiments showed that photolysis of diazirine in solid N 2 produces CH 2 radicals, which subsequently react with neighboring N 2 molecules to form the isomer diazomethane. The photolysis of diazirine in Xe matrices at 5 K has been monitored using ESR spectrometry and the formation of 3 CH 2 was detected for photolysis wavelengths between 380 and 220 nm [99]. For irradiation at 185 nm, the major dissociation path involves the formation of the diazirinyl radical, c-HCN 2 [99]. 15 Another fundamental question concerns the formation of carbene and pathways of the carbenic photoreaction. Diazo compounds and diazirines are the major precursors for carbene generation. It has been observed that there is a difference between photolysis or pyrolysis products and/or their relative yields [3,5]. For example, Frey et al. [100] observed different relative product yields for methylethyldiazirine depending on the reaction method. The reactive intermediate, methylethylcarbene, undergoes hydrogen migration to reach the final products. To explain this observation, Platz et al. [97,101] proposed the rearrangement in the excited states (RIES) mechanism, which has been supported by other groups. The RIES mechanism suggests the possibility of formation of electronically excited carbene. Although significant experimental work has been carried out since 1960, the level of theoretical studies during this early period was too low to provide reliable and accurate data. With the availability of high-level electronic structure computer codes, a renewed effort to elucidate the electronic structure of diazirines/diazo compounds and their photolysis and pyrolysis was initiated in the 1990’s [75,94,102]. Several theory papers describe the excited state dynamics of diazomethane/diazirine [75,94,102,103]. Arenas et al. [94] suggested that carbene is mostly formed in the singlet ground state by S 1 /S 0 conical intersection (CI). They also concluded that the S 1 state of diazirine does not isomerize to the S 1 state of diazomethane, because vibrational coupling to other modes prevents the molecule 16 from proceeding along this route, and the most likely pathway is S 1 /S 0 conical intersection followed by dissociation on S 0 . Isomerization also cannot take place on the S 2 surface. On S 2 , most of the calculated trajectories lead to an S 2 /S 1 conical intersection, propelling the system toward methylene and nitrogen in their lowest singlet states. The earlier work of Yamamoto et al. [75] reached the conclusion that the S 1 /S 0 CI pathway for electronically excited diazomethane is dominant compared to the RIES mechanism. The authors also suggest that isomerization to diazomethane from both S 1 and S 2 of diazirine can occur, and that direct dissociation from S 2 takes place and is more efficient at higher excitation energies. In all the theoretical studies it was concluded that participation of the triplet states is minimal. Recently, Lee et al. studied theoretically the formation of electronically excited methylene from photoexcited diazomethane [102]. Contrary to previous theoretical work, their calculations showed direct formation of excited 1 CH 2 (1 1 B 1 ) that supported the RIES mechanism rather than the S 1 /S 0 CI pathway. The quantum yield of excited carbene formation is dependent on the excitation energy. It is surprising that despite the importance of diazirine there are no experimental results on its dissociation dynamics that test these predictions. Whereas several aspects of diazirine's photochemistry on the lowest excited state 1 1 B 2 have been discussed [5,75,94,103,104], little is known from experiment or theory about photodissociation on higher electronic states, and in particular about multiphoton dissociation pathways. In the only experimental study of its photodissociation dynamics, expansion-cooled diazirine was excited to the S 1 # S 0 17 origin (322.96 nm) and fluorescence centered at wavelengths > 575 nm that decayed with a lifetime > 16 µs was reported [104]. This fluorescence was ascribed to high vibrational levels of the 1 1 B 1 state of singlet methylene, whose origin lies ~ 1 eV above the 1 1 A 1 state. 1.5 Thermochemistry of Diazomethane and Diazirine As mentioned above, only three isomers of CH 2 N 2 (diazomethane, cyanamide, and diazirine), have been isolated and characterized. The thermochemical stability of these isomers was estimated by experimental determination and theoretical calculation of their heats of formation. Heats of formation for stable molecules are typically obtained from calorimetric determinations. However, because of the explosive properties of the diazirines and diazo compounds, direct calorimetric measurements are not possible. The NIST Chemistry Webbook [105] gives values of the heats of formation of diazomethane and diazirine, which were derived from electron impact experiments by Ettinger and Paulett [87] and photoionization studies by Laufer and Okabe [92] over 35 years ago. These values are 49 [106-108] and > 51 kcal/mol [109] for diazomethane and 79.3 [106-108] and 60.6"66 kcal/mol [92], for diazirine at 298 K. Ettinger and Paulett measured the appearance potential of the CH 2 + ion in the ground state [106-108], whereas Laufer and Okabe determined the appearance threshold of CH(A ( X) emission following VUV photolysis [92,109]. Other experimental values obtained for the heat of formation of diazomethane and 18 diazirine are summarized in Table 1.3. The experimental values of )H f o for diazomethane and diazirine show large variations, and each determination is associated with experimental difficulties. Accurate thermochemical values are crucial for analysis of chemical processes. Thus, further independent measurements of the heat of formation are highly desirable. Table 1.3. Experimental and theoretical values of )H f o of diazomethane and diazirine (kcal/mol) at 0 and 298 K. Diazomethane Diazirine Experiment Experiment Ref. 0 K 298 K 0 K 298 K - 49 - 79 [106,107] - 51+ - 60.6"66 [92,109] - - - 55.6 ± 9.5 [110] - 71 - - [111] - 67+ - - [112] - 64"77 - - [113] - 64"73 - - [114] Theory Theory 0 K 298 K 0 K 298 K 65.42, 65.68 63.18, 64.15 77.9, 74.10 76.11, 74.10 [115] 66.7 65.3 - - [47] 68.0 - 77.7 - [116] 64.3 63.1 74.5 73.0 [117,118] In the past 15 years, the ability of quantum chemistry to predict thermochemical properties for chemical systems has been rapidly improving. Thermodynamic properties of chemical systems composed of first and second row elements in most cases can be calculated with chemical accuracy ( ± 1 kcal/mol) [47,119-121]. The heats of formation of diazomethane and its structural isomer diazirine (whose heat 19 of formation is just as controversial) were calculated using high-level electronic structure methods, and a re-evaluation of the accepted values was called for [122]. Table 1.3 shows the calculated heats of formation, $H f ° 0 , which are much more convergent than the experimental ones. Several authors have calculated the heats of formation of diazomethane and diazirine through various pathways. Gordon and Kass [117,118] employed the atomization reaction and two isodesmic reactions using G2 theory, giving average values of 64.3 and 74.5 kcal/mol for $H f ° 0 of diazomethane and diazirine at 0 K, respectively. Catoire [115] reported CBS-Q and G2 heats of formation for several species produced by the decomposition reactions of monomethylhydrazine. Ab initio atomization reactions and atomic heats of formation at 0 K (gas phase) were calculated, giving values of 65.4 and 65.7 kcal/mol for CBS-Q and G2 methods, respectively, for diazomethane, and 78.0 and 76.0 kcal/mol for diazirine. Walch [116] reported heats of formation using CASSCF methods with double- zeta Dunning basis sets for geometries, and internally contracted configuration interaction (ICCI) for energetics with double-, triple-, and quadruple-zeta Dunning bases. He recommended the value of $H f ° 0 = 77.7 kcal/mole for diazirine. Dixon et al. reported 65.3 and 66.7 kcal/mol for the heat of formation of diazomethane at 298 and 0 K, respectively, computed using the CCSD(T) method and CBS extrapolation, which is the most accurate theoretical estimate. All calculations included zero point corrections via harmonic frequency calculations. 20 For diazomethane, both G2 calculations [115,117,118] are within 1 kcal/mol of the result of Dixon et al. [47], suggesting similar accuracy for the respective diazirine values. Thus, for diazomethane, the preferred theoretical value is $H f ° 0 = 67 ± 3 kcal/mole, much closer to the experimental value of 67 kcal/mol recommended by Setser and Rabinovitch [112] than to values given in the NIST Chemistry Webbook [105]. Moreover, in all the calculations it was found that diazirine lies 10 ± 1 kcal/mole above the ground state of diazomethane, allowing us to adapt the thermochemistry of diazomethane reactions to the case of diazirine (see below) [122]. This leads to theoretical heats of formation for diazirine of 75 ± 2 and 77 ± 2 kcal/mol at 298 and 0 K, respectively. These values are much closer to those recommended by Paulett and Ettinger [106-108] than to those of Laufer and Okabe [92]. 1.6 Diazomethane and Diazirine as Precursors of Methylene Diradical As mentioned above, aliphatic diazo compounds and diazirines serve as important sources of carbenes, which occupy a significant place in all branches of chemistry [1-5,9]. Diazirines are the preferred precursors compared to aliphatic diazo compounds because although thermally unstable, chemically they are inert. The high reactivity of diradicals like carbenes and difficulties in their preparation under clean conditions has hindered experimental physical chemistry investigation; of their spectroscopy and photochemistry [9,123-130]. The photochemistry of the simplest carbene, methylene (CH 2 ), provides a fine example of a case in which 21 theory is well ahead of experiment. Although much experimental work has been done on its ground 1 3 B 1 state and the lowest-lying excited singlet state 1 1 A 1 , only a few spectroscopic and dynamics investigations examined photodissociation on triplet surfaces, and no work has been published to date on the spectroscopy and photophysics of the electronically excited states of this diradical. For methylene diradical generation, the photolysis or pyrolysis of diazirine and diazomethane can be employed, but the initially produced singlet CH 2 (1 1 A 1 ) needs to be relaxed to the triplet ground state. 1.7 Major Goals and Specific Aims The goals of this dissertation are: (i) understanding the spectroscopy, photoinization, and photodissociation dynamics of diazomethane (CH 2 N 2 ) and diazirine (c-CH 2 N 2 ), and (ii) developing a clean and intense source of the triplet methylene diradical, specifically for its spectroscopic and photochemical/ photophysical studies in the gas phase. The specific aims of this investigation were to: (i) adapt the traditional preparation methods of diazomethane and diazirine for work in molecular beams; (ii) use a combination of time-of-flight, REMPI and VMI techniques and ab initio calculations to characterize the nature and properties of their electronic and ionic states; (iii) find heats of formation of diazomethane and diazirine and study their photoionization and dissociation; (iv) produce a molecular beam of triplet 22 methylene diradical by means of a pyrolysis nozzle using diazomethane and diazirine as precursors and use the REMPI technique for its diagnostic. Theoretical support for the project described in this dissertation was provided by Lucas Koziol and Prof. Krylov who perfomed extensive ab initio calculations of diazomethane and diazirine. These calculations were crucial to the understanding of the electronic structure, photoionization, and photodissociation dynamic of these chemical systems. 23 1.8 Chapter 1 References 1. H. Zollinger, Azo and Diazo Chemistry; Aliphatic and Aromatic Compounds. 1961: Interscience Publishers, New York. 444 pp. 2. S. Patai, ed. The Chemistry of Diazonium and Diazo Groups. 1978, John Wiley & Sons. 3. M.T.H. Liu, Chem. Soc. Rev., 11, 127, (1982). 4. M. Regitz and G. Maas, Diazo Compounds : Properties and Synthesis 1986: Academic Press, Inc. 596 pp. 5. M.T.H. Liu, ed. Chemistry of Diazirines. 1987, CRC Press: Boca Raton, FL. 6. C. Thomson and C. Glidewell, J. Comput. Chem., 4, 1, (1983). 7. C. Guimon, S. Khayar, F. Gracian, M. Begtrup, and G. Pfisterguillouzo, Chem. Phys., 138, 157, (1989). 8. J.B. Moffat, J. Mol. Struct., 52, 275, (1979). 9. J. Michl and V. Bonacic-Koutecky, Electronic Aspects of Organic Photochemistry. 1990: John Wiley & Sons, Inc. 475 pp. 10. P.G. Stecher, ed. The Merck Index; an Encyclopedia of Chemicals and Drugs. 8th ed. 1968, Rahway, New Jersey: Merck & Co., Inc. 1713 pp. . 11. M.P. Doyle, M.A. McKervey, and T. Ye, Modern Catalytic Methods for Organic Synthesis with Diazo Compounds : from Cyclopropanes to Ylides 1998: John Wiley & Sonc, Inc. 652 pp. 12. V. Branchadell, E. Muray, A. Oliva, R.M. Ortuno, and C. Rodriguez- Garcia, J. Phys. Chem. A, 102, 10106, (1998). 13. M.D. Su, H.L. Liao, W.S. Chung, and S.Y. Chu, J. Org. Chem., 64, 6710, (1999). 14. C. Rodriguez-Garcia, O. Gonzalez-Blanco, A. Oliva, R.M. Ortuno, and V. Branchadell, European Journal of Inorganic Chemistry, 1073, (2000). 15. C. Rodriguez-Garcia, A. Oliva, R.M. Ortuno, and V. Branchadell, J. Am. Chem. Soc., 123, 6157, (2001). 24 16. X. Lu, X. Xu, N. Wang, and Q. Zhang, J. Org. Chem., 67, (2002). 17. T.G. Archibald, Chim. Oggi, 18, 34, (2000). 18. S.-U. Kim, Antimalarial Artemisinin Derivatives. 1998. 19. S.U. Kim, J.H. Choi, and S.S. Kim, J. Nat. Prod., 56, 857, (1993). 20. E. Bombardelli, B. Gabetta, and A. Pontiroli, Process for the Preparation of Taxane Carbonate Derivatives from 10-deacetylbaccatin III for Use as Anticancer Drugs, P.I. Appl., Editor. 2001. p. 31. 21. H. Hotoda, M. Kaneko, M. Inukai, Y. Muramatsu, Y. Utsui, and S. Ohya, Preparation of A-500359 Derivatives as Antibacterial Agents. 2001. p. 459. 22. F. Raulin, M. Khlifi, M. Dang-Nhu, and D. Gautier, Adv. Space Res., 12, 11/181, (1992). 23. P. Griefs, Justus Liebigs Ann. Chem., 106, 123, (1858). 24. T. Curtius, Ber. Dtsch. Chem. Ges., 16, 2230, (1883). 25. A.P.N. Franchimont, Recl. Trav. Chim. Pays-Bas, 9, (1890). 26. H. von Pecmann, Ber. Dtsch. Chem. Ges., 27, 1888, (1894). 27. H. von Pecmann, Ber. Dtsch. Chem. Ges., 28, 855, (1895). 28. T. Curtius, Ber. Dtsch. Chem. Ges., 22, 2161, (1889). 29. A. Angeli, Accademia Nazionale dei Lincei, Classe di Scienze Fisiche, Matematiche e Naturali, Rendiconti, 16, 790, (1907). 30. A. Angeli, Accademia Nazionale dei Lincei, Classe di Scienze Fisiche, Matematiche e Naturali, Rendiconti, 20, 625, (1911). 31. J. Thiele, Ber. Dtsch. Chem. Ges., 44, 2522, (1911). 32. I. Langmuir, J. Am. Chem. Soc., 41, 1543, (1919). 33. I. Langmuir, J. Am. Chem. Soc., 42, 2190, (1920). 34. Boersch, Monatshefte fuer Chemie, 65, 311, (1935). 35. D.A. Ramsay, J. Chem. Phys., 17, 666, (1949). 25 36. A.P.T. Cox, L. F.; Sheridan, J., Nature, 181, 1000, (1958). 37. C. Moore and G. Pimentel, J. Chem. Phys., 40, 329, (1964). 38. C.B. Moore and G.C. Pimentel, J. Chem. Phys., 40, 342, (1964). 39. C.B. Moore, J. Chem. Phys., 39, 1884, (1963). 40. A. Fadini, E. Glozbach, P. Krommes, and J. Lorberth, J. Organomet. Chem., 149, 297, (1978). 41. J. Vogt and M. Winnewisser, Z. Naturforsch., A: Phys. Sci., 38, 1138, (1983). 42. J. Vogt, M. Winnewisser, K. Yamada, and G. Winnewisser, Chem. Phys., 83, 309, (1984). 43. M. Khlifi, P. Paillous, P. Bruston, F. Raulin, and J.C. Guillemin, Icarus, 124, 318, (1996). 44. A. Papakondylis and A. Mavridis, J. Phys. Chem. A, 103, 1255, (1999). 45. M.K. Georgieva, Bulgarian Chem. Commun., 35, 195, (2003). 46. D.W. Ball, THEOCHEM, 722, 213, (2005). 47. D.A. Dixon, W.A. de Jong, K.A. Peterson, and T.B. McMahon, J. Phys. Chem. A, 109, 4073, (2005). 48. I.S.K. Kerkines, P. Carsky, and A. Mavridis, J. Phys. Chem. A, 109, 10148, (2005). 49. I. Fedorov, L. Koziol, G.S. Li, J.A. Parr, A.I. Krylov, and H. Reisler, J. Phys. Chem. A, 111, 4557, (2007). 50. I. Fedorov, L. Koziol, G.S. Li, H. Reisler, and A.I. Krylov, J. Phys. Chem. A, 111, 13347, (2007). 51. C.B. Moore and G.C. Pimentel, J. Chem. Phys., 40, 1529, (1964). 52. J. Sheridan, Proc. Intern. Meeting Mol. Spectry., 4th, 139, (1962). 53. F.W.N. Kirkbride, Ronald G. W., J. Chem. Soc., 119, (1933). 54. R. Brinton and D. Volman, J. Chem. Phys., 19, 1394, (1951). 26 55. J.F. Ogilvie, Photochem. Photobiol., 9, 65, (1969). 56. G. Herzberg, Proc. R. Soc. London, Ser. A, 262, 291, (1961). 57. A.J. Merer, Can. J. Phys., 42, 1242, (1964). 58. D.W. Adamson and J. Kenner, J. Chem. Soc., 1551, (1937). 59. G. Kortum, Z. Phys. Chem. Abt. B:, 50, 361, (1941). 60. J.N. Bradley, A. Ledwith, and G.W. Cowell, J. Chem. Soc., 353, (1964). 61. G. Herzberg, Electronic Spectra and Electronic Structure of Polyatomic Molecules (Molecular Spectra and Molecular Structure). Vol. III. 1966: Van Nostrand, New York. 875 pp. 62. J. Bastide and J.P. Maier, Chem. Phys., 12, 177, (1976). 63. S.P.G. Walch, William A., III., J. Am. Chem. Soc., 97, 5319, (1975). 64. M.P. Habas and A. Dargelos, Chem. Phys., 199, 177, (1995). 65. R.J. Wolf and W.L. Hase, J. Phys. Chem., 82, 1850, (1978). 66. W. Braun, A.M. Bass, and M. Pilling, J. Chem. Phys., 52, 5131, (1970). 67. C.B. Moore and G.C. Pimentel, J. Chem. Phys., 41, 3504, (1964). 68. S.R. Paulsen, Angew. Chem. Int. Ed., 72, 781, (1960). 69. E. Schmitz and R. Ohme, Angew. Chem., 73, 115, (1961). 70. E. Schmitz and R. Ohme, Chem. Ber. Recl., 94, 2166, (1961). 71. E. Schmitz and R. Ohme, Tetrahedron Lett., 612, (1961). 72. W.H. Graham, J. Am. Chem. Soc., 84, 1063, (1962). 73. A. Klip and C. Gitler, Biochem. Biophys. Res. Commun., 60, 1155, (1974). 74. P. Chakrabarti and H.G. Khorana, Biochemistry, 14, 5021, (1975). 75. N. Yamamoto, F. Bernardi, A. Bottoni, M. Olivucci, M.A. Robb, and S. Wilsey, J. Am. Chem. Soc., 116, 2064, (1994). 76. L. Pierce and V. Dobyns, J. Am. Chem. Soc., 84, 2651, (1962). 27 77. E. Schmitz, Comprehensive Heterocyclic Chemistry, Small and Large Rings. COMPREHENSIVE HETEROCYCLIC CHEMISTRY ed. A.R. Katritzky and C.W. Rees. Vol. 7. 1984: Pergamon, Oxford. 994. 78. J. Mason and J.G. Vinter, J. Chem. Soc., Dalton Trans., 2522, (1975). 79. A. Gambi, M. Winnewisser, and J.J. Christiansen, J. Mol. Spectrosc., 98, 413, (1983). 80. M. Bogey, M. Winnewisser, and J.J. Christiansen, Can. J. Phys., 62, 1198, (1984). 81. A. Gambi, J. Mol. Spectrosc., 103, 321, (1984). 82. A. Gambi, U.P. Verma, M. Winnewisser, and G. Guelachvili, J. Mol. Spectrosc., 108, 264, (1984). 83. K. Moller, J. Vogt, M. Winnewisser, and J.J. Christiansen, Can. J. Phys., 62, 1217, (1984). 84. J. Vogt, M. Winnewisser, and J.J. Christiansen, J. Mol. Spectrosc., 103, 95, (1984). 85. A. Gambi, M. Pedrali, M. Winnewisser, and G. Guelachvili, J. Mol. Spectrosc., 113, 250, (1985). 86. U.P. Verma, K. Moller, J. Vogt, M. Winnewisser, and J.J. Christiansen, Can. J. Phys., 63, 1173, (1985). 87. R. Ettinger, J. Chem. Phys., 40, 1693, (1964). 88. C. Puzzarini, A. Gambi, and G. Cazzoli, J. Mol. Struct., 695, 203, (2004). 89. J.A. Merritt, Can. J. Phys., 40, 1683, (1962). 90. L.C. Robertson and J.A. Merritt, J. Mol. Spectrosc., 19, 372, (1966). 91. T.S. Kim, S.K. Kim, Y.S. Choi, and I. Kwak, J. Chem. Phys., 107, 8719, (1997). 92. A.H. Laufer and H. Okabe, J. Phys. Chem., 76, 3504, (1972). 93. M.S. Han, H.-G. Cho, and B.-S. Cheong, Bull. Korean Chem. Soc., 20, 1281, (1999). 28 94. J.F. Arenas, I. Lopez-Tocon, J.C. Otero, and J. Soto, J. Am. Chem. Soc., 124, 1728, (2002). 95. M.B. Robin, K.B. Wiberg, G.B. Ellison, C.R. Brundle, and N.A. Kuebler, J. Chem. Phys., 57, 1758, (1972). 96. J.E. Ogara and W.P. Dailey, J. Am. Chem. Soc., 114, 3581, (1992). 97. D.A. Modarelli and M.S. Platz, J. Am. Chem. Soc., 115, 470, (1993). 98. B. Erni and H.G. Khorana, J. Am. Chem. Soc., 102, 3888, (1980). 99. D. Bhattacharya and J.E. Willard, J. Phys. Chem., 86, 967, (1982). 100. H.M. Frey and I.D.R. Stevens, J. Am. Chem. Soc., 84, 2647, (1962). 101. D.A. Modarelli and M.S. Platz, J. Am. Chem. Soc., 113, 8985, (1991). 102. H. Lee, Y. Miyamoto, and Y. Tateyama, J. Org. Chem., 74, 562, (2009). 103. B. Bigot, R. Ponec, A. Sevin, and A. Devaquet, J. Am. Chem. Soc., 100, 6575, (1978). 104. S.M. Lim, T.S. Kim, G.I. Lim, S.K. Kim, and Y.S. Choi, Chem. Phys. Lett., 288, 828, (1998). 105. H.Y. Afeefy, J.F. Liebman, and S.E. Stein, Neutral Thermochemical Data, in NIST Chemistry WebBook, NIST Standard Reference Database Number 69,. 2003, edited by Linstrom, P. J. and Mallard, W. G. (NIST, Gaithersburg, MD, 2003) <http://webbook.nist.gov/>. 106. G.S. Paulett and R. Ettinger, J. Chem. Phys., 39, 3534, (1963). 107. G.S. Paulett and R. Ettinger, J. Chem. Phys., 39, 825, (1963). 108. G.S. Paulett and R. Ettinger, J. Chem. Phys., 41, 2557, (1964). 109. A.H. Laufer and H. Okabe, J. Am. Chem. Soc., 93, 4137, (1971). 110. M. Jones, Jr. and R.A. Moss, Carbenes. Vol. 1. 1973: A Wiley-Interscience publication, New York. 111. S.W. Benson, Cruicksh.Fr, D.M. Golden, G.R. Haugen, H.E. Oneal, A.S. Rodgers, R. Shaw, and R. Walsh, Chem. Rev., 69, 279, (1969). 112. D.W. Setser and B.S. Rabinovitch, Can. J. Chem., 40, 1425, (1962). 29 113. J.C. Hassler and D.W. Setser, J. Am. Chem. Soc., 87, 3793, (1965). 114. M.S. Foster and Beaucham.Jl, J. Am. Chem. Soc., 94, 2425, (1972). 115. L. Catoire and M.T. Swihart, J. Propul. Power, 18, 1242, (2002). 116. S.P. Walch, J. Chem. Phys., 103, 4930, (1995). 117. M.S. Gordon and S.R. Kass, J. Phys. Chem., 99, 6548, (1995). 118. M.S. Gordon and S.R. Kass, J. Phys. Chem. A, 101, 7922, (1997). 119. B. Ruscic, M. Litorja, and R.L. Asher, J. Phys. Chem. A, 103, 8625, (1999). 120. B. Ruscic, J.E. Boggs, A. Burcat, A.G. Csaszar, J. Demaison, R. Janoschek, J.M.L. Martin, M.L. Morton, M.J. Rossi, J.F. Stanton, P.G. Szalay, P.R. Westmoreland, F. Zabel, and T. Berces, J. Phys. Chem. Ref. Data, 34, 573, (2005). 121. M.H. Matus, A.J. Arduengo, and D.A. Dixon, J. Phys. Chem. A, 110, 10116, (2006). 122. I. Fedorov, L. Koziol, A.K. Mollner, A.I. Krylov, and H. Reisler, J. Phys. Chem. A, (2009). 123. W.T. Borden, Diradicals. 1982: New York : Wiley. 343 pp. 124. V. Bonacickoutecky, J. Koutecky, and J. Michl, Angew. Chem. Int. Ed., 26, 170, (1987). 125. M.S. Platz, Kinetics and Spectroscopy of Carbenes and Biradicals. 1990: New York : Plenum Press. 372 pp. 126. K.B. Eisenthal, R.A. Moss, and N.J. Turro, Science, 225, 1439, (1984). 127. L. Salem and C. Rowland, Angew. Chem. Int. Ed., 11, 92, (1972). 128. W. Sander, Angew. Chem. Int. Ed., 29, 344, (1990). 129. B.J. Finlayson-Pitts and J.N.P. Jr., Chemistry of the Upper and Lower Atmosphere : Theory, Experiments and Applications. 2000: San Diego, California; Oval Road, London : Academic Press. 969 pp. 130. P.D. Lightfoot, R.A. Cox, J.N. Crowley, M. Destriau, G.D. Hayman, M.E. Jenkin, G.K. Moortgat, and F. Zabel, Atmos. Environ. Part A, 26, 1805, (1992). 30 Chapter 2. Experimental Methods and Details 2.1 Supersonic Molecular Beam Systems Since the early work of Kantrowitz and Grey [1], Kistiakowsky and Slichter [2], and others, supersonic molecular beams and their applications have been extensively studied. In 1977 Smalley, Wharton, and Levy [3] reported laser experiments with supersonic beams of molecular systems and demonstrated their usefulness in molecular optical spectroscopy. Since that time laser spectroscopy has become powerful tool for investigating of the electronic structure and photodissociation dynamics of samples in supersonic beams and provides an ideal comparison to theory. Theses studies, in addition to their pure scientific value, have a broad range of applications in studies of air pollution, ozone hole, combustion, interstellar chemistry, etc. [4-6]! In these experiments the molecule of interest is seeded in low concentration into a high-pressure rare gas and supersonically expanded in a continuous or pulsed mode into a vacuum system held at low- pressure and then excited by laser radiation. The cooling effect occurs because collisions cause energy transfer from the internal degrees of freedom of the molecules to translational energy of atoms as they enter the region of free, collisionless flow. This approach provides an almost ideal spectroscopic environment because nearly all molecules are in the lowest quantum state(s) and they travel in free space 31 with narrow velocity distribution and at such a density that intermolecular interactions are unimportant. This simplifies spectroscopic assignments and the well resolved lines allow observation of natural linewidths. Thus, it is possible to obtain vibrationally and rotationally cold molecules in the gas-phase under collision-free conditions and at sufficient concentrations for detection. Also, it is possible to prepare short-lived transients such as free radicals and metastable species and stabilize highly reactive species that decompose at higher temperatures. Collision-free molecular beam techniques have an advantage compared to matrix isolation methods because the line broadening induced by interaction of the molecule with the matrix can be avoided. 2.2 Preparation of Radicals in Molecular Beam Systems The major complications in the study of radicals are their high reactivity and low number density. Some of the most successful approaches to radical production involve the dissociation of precursors that form the desired radicals. This fragmentation can be accomplished photolytically, pyrolytically or through electric discharge [7,8]. While the photolysis technique can be successful, there are some difficulties. The intense light sources required (usually lasers) not only induce photodissociation of the precursors but often induce secondary photochemical reactions yielding undesirable products. Also, radicals produced in the first step can subsequently fragment or isomerize. All these processes result in lowering the 32 concentration and cleanliness of the desired reactive species. The electric discharge source has similar problems. Although production of radicals by pyrolysis in molecular beams was known, the design by Chen and coworkers [9,10] of a pyrolysis pulsed nozzle for the production of cooled, high-concentration, low-contamination reactive intermediates in a supersonic jet expansion made the technique popular. In their method, the organic radicals are produced by pulsing a suitable precursor at low partial pressure (~ 1 Torr) seeded in a high total pressure (1!3 atm.) of inert carrier gas, usually He, through a resistively-heated SiC tube (1-mm inner diameter, about 2-cm long, temperature up to 1800 K) and then expanding them into a high vacuum chamber (~ 10 -5 Torr). Number densities of radicals were estimated to be ~10 13 !10 14 cm -3 at the nozzle exit. The expansion of the gas pulses into the tube is supersonic, which has been achieved by an increase in cross-sectional area upon going from the orifice of the pulsed valve (0.1!0.5 mm diameter) into the SiC tube. The supersonic gas flow velocity (~ 5·10 4 cm·s -1 ) through the tube allows for shortening the residence time in the pyrolysis tube to ~ 10 µs. This time is much shorter than the self-reaction half-life time of the radicals (>160 µs), which is essential to avoid radical-radical recombination [9,10]. The temperature of the nozzle is usually adjusted experimentally to balance the increasing efficiency of pyrolysis with increasing temperature while at the same time minimizing all other reactions including further fragmentation of the desired species, recombination etc. It should be kept in mind that the temperature for decomposition of the precursors is defined 33 by the internal temperatures of the molecules in the gas stream. The temperature of the mixture cannot exceed the temperature of the walls of the heated tube but it might be much less dependent on the dynamics of the heat transfer. 2.3 Photofragment Ion and Photoelectron Imaging The ion imaging technique, pioneered by Chandler and Houston in 1987 [11], has become a powerful tool in the study of energy disposal in elementary chemical reaction dynamics. The method consists of a combination of the state-selective REMPI technique (see Section 2.4) and two-dimensional imaging of photodissociation products. It provides the velocity (speed and angular) distribution of the photofragments. In 1997, Eppink and Parker [12] improved the resolution of the ion imaging technique by replacing the conventional grid electrodes with an electrostatic ion lens with open electrodes. This electrostatic lens system is designed in such a way that particles with the same initial velocity vector but originating in different initial distance from the ion lens axis arrive at the same point on the detector. This eliminates the effect of the finite size of interaction volume of the laser/molecular beam cross-section. This technique, commonly called velocity map imaging (VMI), is very convenient and is used now by many research groups. From the velocity distributions of photofragments measured by the imaging technique much information about the dissociation process can be revealed: speed and angular distributions of the products, product branching ratios, recoil 34 anisotropy parameters ("), orientation and alignment, and pair-correlated energy distributions of “dark” products. Also, this method is useful for the measurement of dissociation energies of the studied species. This can improve measurements of heats of reaction and give insight into izomerization pathways and fragmentation mechanisms. Photofragment imaging of positive ions formed by REMPI detection is only one of the areas where charged particle imaging has become useful. Another area is in the detection of photoelectrons. Photoelectron VMI and variations of this technique have been shown to be powerful methods for the study of cation vibrational structure and they can give insight into the nature of the intermediate excited states. For example, conical intersections, resonances of rovibronic states, nonadiabatic transitions and/or isomerization can affect photoelectron images, reflecting the underlying dynamics [13,14]. In VMI the pulsed laser radiation induces photodissociation of molecules in the molecular beam and then radiation from the same or different laser ionizes the resultant photofragments by REMPI. The generated fragment ions or photoelectrons are extracted and accelerated by the ion optics into the time-of-flight (TOF) mass spectrometer and detected with a two-dimensional position-sensitive detector, usually a microchannel-plate (MCP) detector coupled to a phosphor screen. A charge-coupled device (CCD) camera records the signal from the detector for each laser shot. In the original approach, the image is collected over many thousands laser shots by integrating the signal intensity of the CCD chip either 35 directly on the chip or in the computer, or a combination. Systems which employ the original ion-imaging approach possess cylindrical symmetry with the symmetry axis parallel to the surface of the detector and to the polarization direction of the dissociating light. The image is a two-dimensional projection of the three- dimensional velocity distribution. The three-dimensional velocity distribution can be recovered from the two-dimensional projection by the use of the inverse Abel transform or other related transformations. The speed distribution is obtained from the velocity distribution by integration over all angles: , (2.1) where is the speed of the detected fragment and is the angle between the symmetry axis (axis of polarization of the light) and velocity vector. The angular distribution of the photodissociation fragments is obtained by fitting to the following expression [15,16]: , (2.2) where are the recoil anisotropy parameters (for one-photon dissociation only is non-zero, for two-photon and are non-zero etc.); and is the Legendre Polynomial of order n. For a one-photon process the anisotropy parameter ranges from ! 1 (for an ideal perpendicular transition) to + 2 (for an ideal parallel transition). 36 Under fixed voltages in the VMI arrangement, the speed is proportional to the radius of the image and inversely proportional to the square root of the mass of the charged particle. However, the center-of-mass (c.m.) translational energy of the charged particle does not depend on and is proportional to the square of the radius: , (2.3) where is the magnification factor, which depends on the geometrical parameters of the setup, is directly proportional to voltages applied to the electrostatic ion lens and also depends and does not depends on the mass. In order to calibrate the imaging system, is determined at experimental conditions from measuring the radii of features with well-defined kinetic energies. Thus, the c.m. translational energy distribution of the charged particle can be plotted using Eq. 2.3 and the corresponding Jacobian as: . (2.4) Additionally, the total c.m. translational energy distribution of the two products can be plotted using the equation: , (2.5) where is the mass of the parent molecule. 37 The corresponding transformation is: . (2.6) The speed can be calculated from the experimental radius using the following relation between and : . (2.7) Using the Jacobian , the speed distribution in velocity units can be obtained from the speed distribution in pixel units: . (2.8) The use of event counting and centroiding methods [17,18] in VMI has improved its resolution dramatically and also reduced the detection noise. All imaging data reported in Chapters 3!5 was obtained using VMI with event counting. Whereas the ion-imaging approach has been applied successfully in many chemical dynamical and spectroscopic studies, this technique has some limitations. First, the requirement that the axis of cylindrical symmetry is parallel to the detector plane imposes a limitation on the direction of laser polarization. For example, two-color experiments with two laser beams with two different polarizations needed for studies of vector correlations, such as orientation and 38 alignment of the dissociation fragments, are not possible. Also, inversion methods for reconstruction of velocity distributions from VMI images introduce artificial noise into the reconstructed image that can lead to lowering of the resolution. Currently many variations of the original method exist and scientists are trying to discover new variants to overcome the limitations of the original ion-imaging approach and increase the image resolution. For example, the groups of Kitsopoulos [19,20] and Suits [21] recently developed “slice” and “dc slice ” imaging methods, respectively. In their methods, the detector is gated with pulse much shorter than the temporal spread of the ion cloud along TOF axis. Thus, only the signal from the central section of the ionic fragment cloud is collected. Beacause the slice contains the full angular and translational energy information, their approaches remove the need for image reconstruction and thus lead to higher signal-to-noise ratio and, in the case of Suits’s approach, to enhanced resolution. Images presented in Chapter 5 were obtained using “dc slice” imaging. 2.4 Resonance Enhanced Multiphoton Ionization Resonance enhanced multiphoton ionization (REMPI) is a simple, highly- sensitive, and state-selective ionization technique applied to the spectroscopy of atoms and small molecules. REMPI is a specific case of multiphoton ionization (MPI) when the energy of the first, second, … photon is in resonance with an excited state (intermediate) of the molecule and additional photon(s) ionize the excited state. This technique can be used for characterization of excited states, 39 preparation of state-selected ions, and detection of fragments. Also, this technique allows selected molecules to be ionized while other components remain transparent to the ionizing radiation. Figure 2.1 Two typical REMPI applications and the type of spectra that might result: (a) typical one color REMPI at the two-photon energy resonance (2 + 1 REMPI scheme); an ion signal is observed whenever a resonance is reached; (b) the wavelength is fixed to excite the two-photon transition and the kinetic energies of the resulting photoelectrons are analyzed. 40 The REMPI technique typically involves a resonant one- or two-photon excitation to an electronically excited intermediate state followed by absorption of another photon, of the same or different wavelength, which ionizes the atom or molecule. Figure 2.1 shows two of the simplest (and most practiced) types of REMPI spectroscopy. It is possible to gain additional information about the excited state by determining the kinetic energy and angular distributions of the ejected photoelectrons in ionization of the ground or excited state. The selection rules for two-photon or other multiphoton excitations are different than those for one-photon transitions. As a result, multiphoton REMPI can provide spectroscopic information about excited states which are dark in one-photon excitations. 2.5 Vacuum System, Experimental Arrangement and Data Analysis The molecular beam apparatus used in all experiments has been described previously in detail [22-25]. Briefly, it consisted of three differentially pumped chambers: source chamber, main (ionization) chamber, and detection chamber (see Figure 2.2). The source chamber was pumped by a Leybold TMP 1000C (1100 L/s N 2 ) turbomolecular pump backed by a mechanical pump producing an operating pressure of 10 -5 Torr with a pulsed nozzle running at 10 Hz. The interaction and detection chambers were evacuated by Leybold TMP 361 (345 L/s N 2 ) and TMP 450 (500 L/s N 2 ), respectively. The operating pressure in the detection chamber was 2!4·10 -7 Torr. 41 Figure 2.2 Schematic diagram of the experimental apparatus: 1 - piezo-electric nozzle; 2 - skimmers; 3 - electrostatic lens assembly; 4 - #-metal shielding; 5 – MCP/phosphor screen assembly. (a) and (e): refs. 10 and 11; (b): ref. 11; (c): refs. 6 and 12; (d): refs. 10, 11 and 35 42 Figure 2.3 Schematic diagram of electrostatic lens system in the interaction chamber, field free drift tube of TOF in detection chamber, and the detector assembly. (a) and (e): refs. 10 and 11; (b): ref. 11; (c): refs. 6 and 12; (d): refs. 10, 11 and 35 43 The molecular beam consisted of the species of interest seeded in a carrier gas (He or Ar) and expanded into the source chamber through a home-built pulsed piezoelectric nozzle utilizing a commercial piezoelectric disk translator (Physik Instrumente, P-286.23, Germany). Since diazomethane decomposes on metallic surfaces, metal parts of the pulsed nozzle making contact with the gas mixture were replaced with fiberglass-reinforced polyetherimide (Ultem® 3040). Black nylon tubing was used to make connections. The nozzle, operating at 10 Hz repetition rate, had an orifice of 0.5 mm diameter and was driven by negative ~ 500 V square pulses of ~ 200 µs duration. The nozzle opening time was ~ 150 µs. The molecular beam cooling conditions were adjusted by optimizing carrier gas mixture, backing pressure, nozzle voltage, and time delay between laser and molecular pulses. After expansion into the source chamber the molecular beam expanded supersonically into the main chamber, forming a cold molecular beam. In order to reduce the transverse velocity spread and separate the coldest part of the molecular beam, it was collimated by two skimmers (1.29 and 0.78 mm diameter; Beam Dynamics, Inc) separated by 4 cm. Following expansion through the skimmers the molecular beam passed into the ionization region (see Figure 2.3). The photodissociation and/or ionization of the molecular species took place in the region of four-electrode lens system where the molecular beam was crossed by counter-propagating laser beam, approximately 5 cm from the orifice of the second skimmer (see Figure 2.2). The electrostatic lens system has been designed by 44 Ashfold and co-workers [26] and adopted in our experimental setup in 2003. The products (ions or photoelectrons) produced in the ionization chamber were extracted and accelerated by ion optics into a linear time-of-flight (TOF) mass spectrometer (60 cm long) in the detection chamber. After passing through the TOF region, the products impact a microchannel-plate (MCP) position-sensitive detector (42 mm diameter) with a coupled phosphor screen (Burle Electro-optics, Inc.; APD 3040FM assembly with P-47 phosphor). The ion optics was shielded from magnetic fields by a µ-metal alloy tube (nickel-iron-copper-molybdenum; AD- Vance Magnetic Inc.; AD-MU-80). In order to eliminate the effect of stray field on the trajectories of charged particles, the TOF region was also shielded by a µ-metal tube. The apparatus could be operated in two different modes: TOF and imaging. In TOF mode the current collected by the phosphor screen was measured by oscilloscope (Tektronix, TDS 3054) through a 100x amplifier with 50 $ input impedance (Phillips Scientific; DC-100 MHz bipolar amplifier, model 6931), and transferred to a PC for analysis. The typical voltages applied to the electrostatic lens system, MCP detector and phosphor screen in different modes are shown in Table 2.1. The optimal voltage ratio for the electrostatic lens in our apparatus was found to be: V(Repeller):V(Extractor):V(Lens) = 2.49:2:1 [24]. The measured REMPI spectra were recorded in TOF mode by monitoring the signal from the parent or product ion as a function of the laser excitation wavelength. In the imaging mode a charge-coupled device (CCD) camera 45 (LaVision, Imager 3, 12 bit, 1280x1024 pixel array) recorded the image of the phosphor screen for each laser shot and exported them to the PC for analysis using the DaVis software package (LaVision) that included event counting. In ion imaging mode the ability to selectively detect a single mass was achieved by gating the detector electronically using pulse generator (DEI, PVX-4041, ~ 60 ns minimum pulse). In dc slice images where only the central slice of the single mass ion cloud is recorded, a home-built pulse generator (high voltage (~ 2 kV) pulse of ~ 5 ns FWHM) was used to gate the MCP detector. Table 2.1 Typical voltages applied to the electrostatic lens system, MCP detector and phosphor screen in different modes. MCP Mode Repeller Extractor Lens Plate 1 Plate 2 Phosphor Screen TOF + 4980 + 4000 + 2000 ! 1500 ! 100 0 Ion Imaging + 4980 + 4000 + 2000 0 + 1750 + 5000 Photoelectron Imaging ! 4980 ! 4000 ! 2000 0 + 1750 + 5000 In experiments described in Chapters 3 and 4, photoelectron speed distributions were determined from the images obtained in photoelectron imaging mode, using event counting and centroiding [17,18], and the basis set expansion (BASEX) Abel transform method [27]. The photoelectron speed distributions were converted to photoelectron kinetic energy distributions by applying Eq. 2.3 and using 46 = 0.181 at V(Repeller) = 4980 V for and corresponding Jacobian for (Eq. 2.4). was determined by NO ionization via the A 2 % + state [28]. The relative width of the most narrow peak was &E/E = 3.0 ± 0.2 %, limited by instrument resolution [29]. In Chapter 5, the speed distributions of photofragment ions were determined from images obtained in dc sliced ion imaging mode, using event counting and centroiding [17,18], and then integrating the experimental images over all angles. The total c.m. translational energy distributions of photofragment ions were plotted by applying Eq. 2.3 and 2.5 using = 0.181 at V(Repeller) = 4980 V for and corresponding Jacobian for (Eq. 2.6). 2.6 Chemical Preparation of Diazomethane and Diazirine Prior to currently mentioned study, no molecular beam studies of diazomethane and diazirine have been reported because of difficulties in preparing and delivering them intact into the interaction chamber. In an attempt to develop efficient sources of carbenes, we have adapted the traditional preparation methods of diazomethane and diazirine for work in molecular beams. We also modified our inlet and pulsed- nozzle systems for stable and safe delivery (see Section 2.7). 47 Figure 2.4 Schematic representation of the all-glass vacuum line used to prepare diazomethane and diazirine which is comprised of three sections: a reactor fitted with a pressure-equalized addition funnel, two traps and a vacuum manifold. (a) and (e): refs. 10 and 11; (b): ref. 11; (c): refs. 6 and 12; (d): refs. 10, 11 and 35 48 Figure 2.5 Typical IR spectrum of diazomethane in the 500–3500 cm -1 region, partial pressure ~ 35 Torr. For synthesis of diazomethane a glass vacuum line, constructed of greaseless Teflon ® high vacuum valves and o-ring joints, was used (Figure 2.4). This line consisted of three sections: a reactor, two traps and a vacuum manifold. The reactor was fitted with a pressure-equalized addition funnel. First the reactor was charged with the reactant and degassed. CH 2 N 2 was prepared under a static vacuum by the dropwise addition of an excess of of 2.5M NaOH (aqueous) of ~ 15 ml to 2.6 g of N-methyl-N'-nitro-N-nitrosoguanidine (MNNG; CAS: 70-25-7; FW:147.09 g·mol -1 ; TCI America ® ). The resulting CH 2 N 2 was purified by passing the gas though two 49 traps held at ! 78 o C (195 K) with a dry ice/ethanol slush, and collected in a 12-L glass flask housed in a steel mesh box and protected from exposure to light. The pressure of CH 2 N 2 is kept less than 30 Torr; otherwise it will begin to condense in the dry ice/ethanol traps where it is likely to explode. Figure 2.6 Typical IR spectrum of deuterated diazomethane in the 500!3500 cm -1 region, partial pressure ~ 35 Torr. The method used to produce CD 2 N 2 and CHDN 2 is based on the one for production of CH 2 N 2 . This method for simultaneously producing CH 2 N 2 , CD 2 N 2 , and CHDN 2 has the advantage that isotopologs can be prepared using protonated precursors and solvents, and only the aqueous NaOD in D 2 O needs to be deuterated. The same glass vacuum line was used for the synthesis. In the modified procedure, CH 2 N 2 , CD 2 N 2 , and CHDN 2 were generated simultaneously under vacuum in a 50 closed reactor by the reaction of 2.6 g of N-methyl-N'-nitro-N-nitrosoguanidine (TCI America ® ) dissolved in 30 mL of Tetra (ethylene glycol) dimethyl ether, 99 % (CAS: 143-24-8; FW: 222.28 g·mol -1 ;Aldrich ® ) with an excess of ~ 7.5 ml aqueous solution of NaOD (2.5M) and 7.5 ml of NaOH (2.5M) mixture. The Tetra (ethylene glycol) dimethyl ether was used in this preparation to give a solvent where CH 2 N 2 can dissolve giving it time to exchange H with D in the aqueous solution. The commercially available precursor, Diazald (CAS: 80-11-5; FW: 214.24 g·mol -1 ) also can be used. The solution was stirred for ~ 10 minutes at 0 o C (273 K) and expanded through two traps held at ! 78 o C (195 K) with a dry ice/ethanol slush to a 12-L glass flask that was evacuated, protected from exposure to light, and housed in a steel mesh box. A mixture of approximately 0.5 % CH 2 N 2 , 0.5 % CD 2 N 2 , and 0.5 % CHDN 2 in He at 2 atm total pressure was prepared in this flask. When using only ~ 15 ml of aqueous of NaOD (2.5M) solution, CD 2 N 2 in isotopic purity of up to 94% and high overall yield [30] is prepared. The IR spectra of CH 2 N 2 , CD 2 N 2 , and CHDN 2 prepared by this procedure are shown in Figures 2.5–2.7, respectively and UV spectrum of CH 2 N 2 in Figure 2.8. All spectra were in good agreement with published spectra [31-36] and contained negligible amounts of impurities. Based on the IR spectra, sample of CH 2 N 2 :CD 2 N 2 :CHDN 2 ~ 1:1:1 is simultaneously generated; all samples survive for several days until depleted by use. 51 Figure 2.7 Typical IR spectrum of partially deuterated diazomethane in the 500!3500 cm -1 region, partial pressure ~ 35 Torr. We wish to emphasize that CH 2 N 2 is a toxic and hazardous gas, which can decompose explosively and spontaneously, and thus appropriate safety precautions must be taken. CH 2 N 2 should be handled only at low pressures and on a small scale. At no time should the gas be allowed to condense into the liquid phase. Throughout our experiments a pressure of 25 Torr was never exceeded. Safety equipment (safety shields, safety glasses, face shields, leather gloves and protective clothing, such as leather suits, Kevlar® sleeves and earplugs) must be used at all times. Care must be taken to avoid known triggers of CH 2 N 2 decomposition such as intense light, exposure to rough, metallic, or acidic surfaces, and abrupt changes in temperature, pressure, and phase. 52 (a) (b) Figure 2.8 Typical UV spectra of diazomethane in (a) 190–270 nm region, partial pressure < 1 Torr; (b) 200–600 nm region, partial pressure ~ 35 Torr. 53 Diazirine was prepared by means of a modification of the method of Ohme and Schmitz [37]. A mixture of 120 g (4 mol) paraformaldehyde (CAS: 30525-89-4; FW: (30.03) n g·mol -1 ) and 360 g (8 mol) formamide (CAS: 75-12-7; FW: 45.04 g·mol -1 ) were heated to 120!135 o C with stirring for 36 hours. After allowing 2!3 days for crystallization at ambient temperature, the mixture was filtered and the obtained crystals washed with cold methanol. 25 g Methylene-bis-formamide were cautiously added to 162 g 50% sulfuric acid at 0!10 o C. 35.9 g of methylenediamine-sulfate crystallized from the clear solution over the course of 2 days in refrigerator at ~ 4 o C. The sulfate was washed with methanol and dried under dynamic vacuum until it reaches room temperature. The composition of the salt is 3CH 2 (NH 2 ) 2 ·4H 2 SO 4 . Diazirine was produced in the all-glass vacuum line described above except modified to increase the sizes of the reactor and a pressure-equalized addition funnel were increased to ~ 600 and 250 ml, respectively. The glass reactor was charged with 15.0 g 3CH 2 (NH 2 ) 2 ·4H 2 SO 4 , in powder form, and a NaCl/ice mixture 54 at ! 15 to ! 10 o C. (The commercially available precursor, Methylenediamine dihydrochloride (CAS: 57166-92-4; FW: 118.99 g·mol -1 ) also can be used.) The reactor was submerged in an external salt/ice bath at the same temperature. A pressure-equalized addition funnel isolated from the reactor by greaseless Teflon ® /glass high vacuum valves was charged with a mixture of 150 mL NaOCl solution (10-14% by weight; CAS: 7681-52-9; FW: 74.44 g·mol -1 ), and 30 mL NaOH (~25 M). The reactor and the addition funnel were evacuated and the solutions degassed. The NaOCl/NaOH solution was then added drop-wise over the course of about 30 minutes to the reactor. The resulting c-CH 2 N 2 was purified by passing the gas though a bubbler with a gas dispersion tube and the trap filled with concentrated NaOH solution in water, to remove any CO 2 . The gas then passes through the two traps held at ! 78 o C with a dry-ice/ethanol slush, and was collected in a 12-L glass flask housed in a steel mesh box and protected from exposure to light as normal. The pressure of c-CH 2 N 2 was kept at less than 30 Torr. The purity of the diazirine sample was assessed from its IR and UV spectra (Figures 2.9 and 2.10, respectively), which were in good agreement with published spectra [38-40] and contained negligible amounts of impurities. 55 Figure 2.9 Typical IR spectrum of diazirine in the 500–4000 cm -1 region, partial pressure ~ 15 Torr. Figure 2.10 Typical UV spectrum of diazirine in 200–400 nm region, partial pressure ~ 75 Torr. 56 We wish to emphasize that diazirine is a toxic and hazardous gas, which can decompose explosively and spontaneously, and thus appropriate safety precautions must be taken. c-CH 2 N 2 should be handled only at low pressures and on a small scale. At no time should the gas be allowed to condense into the liquid phase. Throughout our experiments a pressure of 30 Torr was never exceeded. Safety equipment (safety shields, safety glasses, face shields, leather gloves and protective clothing, such as leather suits, Kevlar® sleeves and earplugs) must be used at all times. Care must be taken to avoid known triggers of c-CH 2 N 2 decomposition such as intense light and abrupt changes in temperature, pressure, and phase. 2.7 Production of CH 2 Radical As was mentioned above, both diazomethane and diazirine are the prime sources of methylene and a pulsed pyrolysis source is an excellent source of jet- cooled, high-concentration, low-contamination radicals. In the experiments described in Chapter 6, CH 2 radicals were produced by thermal decomposition of diazomethane or diazirine in a pulsed pyrolysis source. According to the thermochemical data (see Tables 2.2 and 2.3), the lowest spin- allowed dissociation channel is to the CH 2 (1 1 A 1 ) + N 2 (X 1 " g + ) products. Altough the CH 2 (1 3 B 1 ) + N 2 (X 1 " g + ) channel is spin-forbiden, it is known that various inert additives collisionally quench the singlet methylene to the triplet state [41-46]. From flash photolysis studies of ketene and diazomethane precursors, the quenching rate constant ( ) of the 1 CH 2 + He ' 3 CH 2 + He reaction has been 57 determined to be (3.0 ± 0.7)·10 -13 cm 3 ·molecule -1 ·s -1 at 298 K [41]. No evidence has been found for the transition from triplet to singlet methylene. Triplet methylene was monitored by means of its absorption at 141.5 nm and singlet methylene was not observed directly. Table 2.2 Values of heats of formation of molecules at 0 K. Molecule State !H f o 0 , kcal/mol References H 2 S 51.63 [47] CH 1 2 # 141.61 ± 0.14 [48] N 2 1 1 " g + 0 ! H 2 1 1 " g + 0 ! NH 1 3 " - 85.4 ± 0.3 [49,50] NH 1 1 & 121 ± 1 [49-51] CNN 1 3 " - 138.2 ± 0.7 [52] CNN 1 1 & 157 ± 1 [52,53] CH 2 1 3 B 1 93.377 ± 0.382 [48] CH 2 1 1 A 1 102.364 ± 0.382 [48] HCN 1 1 " + 30.9 ± 0.7 [52] HN 2 1 2 A’’ 60.8 ± 1 [54] HCNN 1 2 A’’ 111.7 ± 0.7 [52] c-CHN 2 1 2 A 2 117.8 ± 0.7 [52] CH 2 N 2 1 1 A 1 67 ± 3 [54-58] c-CH 2 N 2 1 1 A 1 77 ± 3 [54-57] 58 According to the thermochemical data (see Tables 2.2 and 2.3), the lowest spin- allowed dissociation channel is to the CH 2 (1 1 A 1 ) + N 2 (X 1 " g + ) products. Although the CH 2 (1 3 B 1 ) + N 2 (X 1 " g + ) channel is spin-forbiden, it is known that various inert additives collisionally quench the singlet methylene to the triplet state [41-46]. From flash photolysis studies of ketene and diazomethane precursors, the quenching rate constant ( ) of the 1 CH 2 + He ' 3 CH 2 + He reaction has been determined to be (3.0±0.7)·10 -13 cm 3 ·molecule -1 ·s -1 at 298 K [41]. No evidence has been found for the transition from triplet to singlet methylene. Triplet methylene was monitored by means of its absorption at 141.5 nm and singlet methylene was not observed directly. Table 2.3 Values of heats of reaction at 0 K. a Products &H r o 0 , kcal/mol (CH 2 N 2 ) &H r o 0 , kcal/mol (c-CH 2 N 2 ) CH 2 (1 3 B 1 ) + N 2 (X 1 " g + ) 26 16 CH 2 (1 1 A 1 ) + N 2 (X 1 " g + ) 35 25 HCN(1 1 " + ) + NH (1 1 &) 85 75 CNN(1 1 &)+ H 2 (1 1 " g + ) 90 80 HCNN(1 2 A’’) + H( 2 S) 97 86 c-CHN 2 (1 2 A 2 ) + H( 2 S) ! 92 CH(1 2 #) + HN 2 (1 2 A’’) 135 125 a Ucnertanity ± 4 kcal/mol. 59 Ketley et al. [45], directly measured of 1 1 A 1 and 1 1 B 1 states CH 2 , for different additives using laser-induced IR multiple photon dissociation to prepare the radicals, and time-resolved laser induced fluorescence to observe them. The authors concluded that the previously accepted values of for singlet methylene removal were underestimated. For example the measured using He was 3.1·10 -12 cm 3 ·molecule -1 ·s -1 . Taking into account that the concentration of He in the molecular beam is ~ 5·10 19 molecule·cm -3 (2 atm), the quenching time for this reaction is 6 ns ( ). The collisions with precursor molecules and self- reactions can be neglected because of their low concentration. This quenching time is much shorter than the residence time in the pyrolysis tube. No data is available for the quenching rate constants at pyrolysis temperatures (up to 1800 K) but it is not likely that the difference in temperature would reduce . Using molecular collision theory the total number of collisions of molecules of gas 1 with molecules of gas 2 per unit time per unit volume (both molecules moving) can be expressed by: , (2.9) where is the collision frequency (cm -3 ·s -1 ); and are concentrations of gases 1 and 2 (cm -3 ); is the collision cross-section and are the radii of molecules of gas 1 and 2, respectively; is the mean speed of the Maxwell speed distribution (cm·s -1 ): 60 , (2.10) where is the Boltzmann constant (J·K -1 ); is temperature of the molecules (K); is the reduced mass of two molecules (kg). And the number of efficient collisions ( ) per unit time per unit volume can be expressed by: . (2.11) Thus, the average number of collisions ( ) required for one quenching event is the relation of to : (2.12) and the reverse ratio describe the efficiency of the one collision and is called collision efficiency: . (2.13) Using the 1.40 Å and 1.75 Å values for the radii of He and CH 2 and = 3.1·10 -12 cm 3 molecule -1 ·s -1 , at 298 K. Hall et al. [59], using Doppler-resolved transient frequency modulation absorption spectroscopy, studied the role of mixed states in the collision-induced intersystem crossing in CH 2 . Direct monitoring of the time-dependent populations of rotational levels containing mixed singlet and triplet character has revealed a rapid interconversion between the two components. Thus, it should be possible to 61 produce a convenient source of methylene diradicals in the triplet ground state by using diazomethane and diazirine as precursors and employing the molecular beam technique with a pulsed pyrolysis nozzle. Table 2.3 shows that for both precursors, the second and third channels, HCN(1 1 " + ) + NH(1 1 &) and CNN(1 1 &) + H 2 (1 1 " g + ), are 50 and 55 kcal/mol higher compare to CH 2 (1 1 A 1 ) + N 2 (X 1 " g + ) channel and in addition will have high activation barriers. Therefore, it should be possible to optimize the conditions in the pulsed pyrolysis source to maximize production of CH 2 . The experiments were performed using a pulsed molecular beam containing 0.5!1.0 % diazomethane or diazirine in 2 atm He. The procedures for preparation of the precursors are described in detail in Section 2.6. The sample was expanded through a pulsed pyrolysis source (10 Hz repetition rate), constructed based on the design described by Chen and co-workers [9,10] and utilizing the arrangement of nozzle attachment by Lester and co-workers [60]. Detailed technical drawings of this nozzle are given in Appendix A. The nozzle consists of a 1 mm inner diameter silicon carbide (SiC) tube (1) (Saint-Gobain Ceramics; product number: 45155) as depicted in Figure 2.11. The SiC tube was resistively heated (temperature 300!1800 K) using variable voltages (Variac) via two copper electrodes (2) attached to the SiC tube. In order to provide better electrical contact and compensate for the difference in thermal expansion coefficients of Cu (17·10 -6 1/ o C) and SiC (4·10 -6 1/ o C), the electrodes were attached to the tube through home-built graphite split disks. The length of the heated region of the SiC tube could be varied 62 between 1 and 2 cm by adjusting the position of the carbon electrodes and was optimized for maximum generation efficiency of the CH 2 radicals. The SiC posseses negative temperature coefficient (resistance decreasing with temperature increasing) resulting in a sharp increase of the current with temperature. Therefore, to maintain constant current, three plug-in 100 W incandescent light bulbs (which have positive temperature coefficient) were added in series to the electric circuit. The SiC carbide tube is mounted in a stainless-steal water-cooled block (4) through a holder made of low thermal conductivity machineable glass-ceramic (MACOR ® ) (3). Since CH 2 N 2 decomposes on metallic surfaces, parts of the pulsed nozzle making contact with the gas mixture: the plunger (5), sample inlet chamber (6) and its face plate (7) were made from fiberglass-reinforced polyetherimide (Ultem® 3040). The o-rings on the plunger are made of Kalrez 4079. The faceplate with a 0.5 mm orifice of the pulsed valve is mounted to the back side of the water-cooled block. Approximately the same pressure of buffer gas was maintained inside the main chamber of the piezoelectric nozzle to prevent leakage across the o-ring seal and to facilitate normal plunger movement. In order to keep the piezoelectric translator from overheating, water-cooling of the main chamber of the nozzle was also provided. The assembly was cooled with circulated water at 0 o C. Under these conditions during experiments on pyrolysis of diazomethane or diazirine the nozzle operated reliably for more than 24 hours. A circular stainless steel flange with electrical feedthroughs and cooling water connectors was used to introduce the pyrolysis source into a source chamber of the vacuum system. 63 (a) (b) Figure 2.11 (a) Schematic cross-section of the pulsed pyrolysis source: 1 – SiC tube; 2 – electrodes; 3 – SiC-tube holder; 4 – water-cooled block; 5 – plunger; 6 – sample inlet chamber; 7 – face plate of the sample inlet chamber; 8 – main chamber; 9 – face plate of the main chamber; 10 – piezoelectric disk translator; (b) Electrical circuit for the pulsed pyrolysis source. 64 The opening time of the nozzle was optimized to be not too short in order to provide high enough concentration of the precursor, but not too long to not reduce the conversion efficiency of the precursor. The latter effect is due to the fact that a higher gas flux through the SiC tube results in a poorer heat transfer and lower average temperature of the gas. Figure 2.12 The diazomethane pyrolysis efficiency curve based on CH 2 N 2 + signal. In order to test the pulsed pyrolysis nozzle we recorded the efficiency curve (Figure 2.12) for diazomethane pyrolysis versus applied electrical power (0.6 % diazomethane in 2 atm of He mixture). The diazomethane was detected by 2 + 1 REMPI through the band origin of the 2 1 A 2 (3p y ( )) Rydberg state at 382.94 nm (52227 cm -1 ). Resistive heating of the SiC tube nozzle by applying 65 powers of 8!9 W resulted in decomposition of ~ 90 % of the precursor to CH 2 and byproducts. The diazirine ion was not observed under any conditions; therefore it was not possible to obtain the diazirine pyrolysis efficiency curve. However, in light of the similar dissociation energies of diazirine and diazomethane, the two curves should not be to different. To detect and optimize the signal from CH 2 (1 3 B 1 ), 2 + 1 REMPI detection via the manifold of 3p states at 311.6 nm whose vertical excitation energies are at 7.6!7.9 eV was employed [61]. At a certain high powers (temperatures) the efficiency of production of CH 2 started to decrease presumably as a result of both secondary reactions and increasing contribution of higher-threshold dissociation pathways in diazomethane (see Table 2.3). After the supersonic expansion, the radicals were skimmed twice and introduced into the reactor chamber. 66 2.8 Chapter 2 References 1. A. Kantrowitz and J. Grey, Rev. Sci. Instrum., 22, 328, (1951). 2. G.B. Kistiakowsky and W.P. Slichter, Rev. Sci. Instrum., 22, 333, (1951). 3. R.E. Smalley, L. Wharton, and D.H. Levy, Acc. Chem. Res., 10, 139, (1977). 4. B.J. Finlayson-Pitts and J.N.P. Jr., Chemistry of the Upper and Lower Atmosphere : Theory, Experiments and Applications. 2000: San Diego, California; Oval Road, London : Academic Press. 969 pp. 5. M.T. Bowers, ed. Ion Chemistry of the Earth's Atmosphere. Gas Phase Ion Chemistry. Vol. 1. 1979, Academic: New York. 6. W.W. Duley and D.A. Williams, Interstellar Chemistry 1984: Academic Press: New York. 251 pp. 7. S. Davis, D.T. Anderson, G. Duxbury, and D.J. Nesbitt, J. Chem. Phys., 107, 5661, (1997). 8. P.C. Engelking, Chem. Rev., 91, 399, (1991). 9. P. Chen, S.D. Colson, W.A. Chupka, and J.A. Berson, J. Phys. Chem., 90, 2319, (1986). 10. D.W. Kohn, H. Clauberg, and P. Chen, Rev. Sci. Instrum., 63, 4003, (1992). 11. D.W. Chandler and P.L. Houston, J. Chem. Phys., 87, 1445, (1987). 12. A.T.J.B. Eppink and D.H. Parker, Rev. Sci. Instrum., 68, 3477, (1997). 13. A.G. Suits and R.E. Continetti, eds. Imaging in Chemical Dynamics. 2001, American Chemical Society: Washington, DC. 412 pp. 14. B.J. Whitaker, ed. Imaging in Molecular Dynamics. Technology and Applications. 2003, Cambridge University Press. 249 pp. 15. R.N. Zare, Mol. Photochem., 4, 1, (1972). 16. R.N. Zare, ed. Angular Momentum : Understanding Spatial Aspects in Chemistry and Physics. (the George Fisher Baker Non-Resident Lectureship in Chemistry at Cornell University). 1988, New York : Wiley. 349. 67 17. B.Y. Chang, R.C. Hoetzlein, J.A. Mueller, J.D. Geiser, and P.L. Houston, Rev. Sci. Instrum., 69, 1665, (1998). 18. Y. Tanaka, M. Kawasaki, Y. Matsumi, H. Fujiwara, T. Ishiwata, L.J. Rogers, R.N. Dixon, and M.N.R. Ashfold, J. Chem. Phys., 109, 1315, (1998). 19. T.N. Kitsopoulos, C.R. Gebhardt, and T.P. Rakitzis, J. Chem. Phys., 115, 9727, (2001). 20. T.P. Rakitzis and T.N. Kitsopoulos, J. Chem. Phys., 116, 9228, (2002). 21. D. Townsend, M.P. Minitti, and A.G. Suits, Rev. Sci. Instrum., 74, 2530, (2003). 22. A. Sanov, T. Droz-Georget, M. Zyrianov, and H. Reisler, J. Chem. Phys., 106, 7013, (1997). 23. M. Zyrianov, Photoinitiated Decomposition of Hnco, Chemistry, University of Southern California Los Angeles, 1998. 24. V. Dribinski, Photoelectron and Ion Imaging Studies of the Mixed Valence/Rydberg Excited States of the Chloromethyl Radical, Ch2cl, and the Nitric Oxide Dimer, (No)2, Dissertation, Chemistry, University of Southern California, Los Angeles, 2004. 25. Potter, Ion Imaging Studies of the Spectroscopy and Photodissociation Dynamics of Chloromethyl Radical and Nitric Oxide Dimer. , Chemistry, Universuty of Southern California, Los Angeles, 2005. 26. E. Wrede, S. Laubach, S. Schulenburg, A. Brown, E.R. Wouters, A.J. Orr- Ewing, and M.N.R. Ashfold, J. Chem. Phys., 114, 2629, (2001). 27. V. Dribinski, A. Ossadtchi, V.A. Mandelshtam, and H. Reisler, Rev. Sci. Instrum., 73, 2634, (2002). 28. V. Dribinski, A.B. Potter, I. Fedorov, and H. Reisler, Chem. Phys. Lett., 385, 233, (2004). 29. I. Fedorov, L. Koziol, G.S. Li, J.A. Parr, A.I. Krylov, and H. Reisler, J. Phys. Chem. A, 111, 4557, (2007). 30. S.M. Hecht and J.W. Kozarich, Tetrahedron Lett., 1501, (1972). 31. C. Moore and G. Pimentel, J. Chem. Phys., 40, 329, (1964). 68 32. C.B. Moore, J. Chem. Phys., 39, 1884, (1963). 33. R.H. Pierson, A.N. Fletcher, and E.S.C. Gantz, Anal. Chem., 28 1218, (1956). 34. M. Khlifi, P. Paillous, P. Bruston, F. Raulin, and J.C. Guillemin, Icarus, 124, 318, (1996). 35. R. Brinton and D. Volman, J. Chem. Phys., 19, 1394, (1951). 36. R.J. Wolf and W.L. Hase, J. Phys. Chem., 82, 1850, (1978). 37. E. Schmitz and R. Ohme, Chem. Ber. Recl., 94, 2166, (1961). 38. W.H. Graham, J. Am. Chem. Soc., 84, 1063, (1962). 39. R. Ettinger, J. Chem. Phys., 40, 1693, (1964). 40. A. Gambi, M. Winnewisser, and J.J. Christiansen, J. Mol. Spectrosc., 98, 413, (1983). 41. W. Braun, A.M. Bass, and M. Pilling, J. Chem. Phys., 52, 5131, (1970). 42. M.Y. Chu and J.S. Dahler, Mol. Phys., 27, 1045, (1974). 43. K.C. Kulander and J.S. Dahler, J. Phys. Chem., 80, 2881, (1976). 44. K.C. Kulander and J.S. Dahler, Chem. Phys. Lett., 41, 125, (1976). 45. M.N.R. Ashfold, M.A. Fullstone, G. Hancock, and G.W. Ketley, Chem. Phys., 55, 245, (1981). 46. U. Bley and F. Temps, J. Chem. Phys., 98, 1058, (1993). 47. S.G. Lias, J.E. Bartmess, J.F. Liebman, J.L. Holmes, R.D. Levin, and W.G. Mallard, J. Phys. Chem. Ref. Data, 17, 1, (1988). 48. B. Ruscic, J.E. Boggs, A. Burcat, A.G. Csaszar, J. Demaison, R. Janoschek, J.M.L. Martin, M.L. Morton, M.J. Rossi, J.F. Stanton, P.G. Szalay, P.R. Westmoreland, F. Zabel, and T. Berces, J. Phys. Chem. Ref. Data, 34, 573, (2005). 49. W.R. Anderson, J. Phys. Chem., 93, 530, (1989). 50. R. Tarroni, P. Palmieri, A. Mitrushenkov, P. Tosi, and D. Bassi, J. Chem. Phys., 106, 10265, (1997). 69 51. P. Pd and P. Chandra, J. Chem. Phys., 114, 7450, (2001). 52. E.P. Clifford, P.G. Wenthold, W.C. Lineberger, G.A. Petersson, K.M. Broadus, S.R. Kass, S. Kato, C.H. DePuy, V.M. Bierbaum, and G.B. Ellison, J. Phys. Chem. A, 102, 7100, (1998). 53. R. Pd and P. Chandra, J. Chem. Phys., 114, 1589, (2001). 54. D.A. Dixon, W.A. de Jong, K.A. Peterson, and T.B. McMahon, J. Phys. Chem. A, 109, 4073, (2005). 55. M.S. Gordon and S.R. Kass, J. Phys. Chem., 99, 6548, (1995). 56. M.S. Gordon and S.R. Kass, J. Phys. Chem. A, 101, 7922, (1997). 57. L. Catoire and M.T. Swihart, J. Propul. Power, 18, 1242, (2002). 58. S.P. Walch, J. Chem. Phys., 103, 4930, (1995). 59. A.V. Komissarov, A. Lin, T.J. Sears, and G.E. Hall, J. Chem. Phys., 125, (2006). 60. M.T. Berry, M.R. Brustein, and M.I. Lester, Chemical Physics Letters, 153, 17, (1988). 61. K.K. Irikura, R.D. Johnson, and J.W. Hudgens, Journal of Physical Chemistry, 96, 6131, (1992). 70 Chapter 3. Theoretical and Experimental Investigations of the Electronic Rydberg States of Diazomethane: Assignments and State Interactions 3.1 Introduction Diazo compounds serve as sources of carbenes in organic synthesis and participate in alkylation reactions [1]. Diazomethane is an important source of methylene and its photochemistry is relevant to understanding the chemistry in atmospheres rich in N 2 and methane, such as in Titan, Triton, and Pluto [2]. Not surprisingly, elucidating its excitation and fragmentation mechanisms has attracted interest since 1933 [3]. In spite of this interest, its excited electronic states have not been fully assigned. The reason is, of course, the instability of diazomethane and its explosive nature. The C-N bond is weak: the dissociation energy to triplet methylene is < 38 kcal/mol and to singlet methylene it is < 47 kcal/mol [1,4], and the molecule is unstable. The goal of the work presented in this Chapter is to characterize the electronic structure and photophysics of diazomethane in the region of its lowest Rydberg states by using a combination of high-level electronic structure calculations and photoelectron and photofragment ion imaging experiments. This would aid, among others, in developing sensitive spectroscopic diagnostics for diazomethane via its Rydberg states. Therefore, our focus is also on identifying electronic state interactions and estimating dissociation rates. 71 The electronic configuration of ground state diazomethane (1 1 A 1 ) is: [core + low-lying] 16 (7a 1 ) 2 (2b 2 ) 2 (2b 1 ) 2 (3b 2 ) 0 (8a 1 ) 0 (3b 1 ) 0 = [core + low-lying] 16 (!) 2 (" ! ) 2 (") 2 (" ! * ) 0 (! * ) 0 (" * ) 0 , where “low-lying” refers to the 1s core orbitals, two ! CN orbitals, as well as one extended sigma-bonding and two "-type molecular orbitals (which are not involved in excitations). Relevant molecular orbitals are shown in Figure 3.1. CH 2 N 2 lies in the yz plane with the z axis coinciding with the C 2 symmetry axis (CNN bond). The lowest valence excitations promote an electron from the " orbital to the "* or " ! * orbital [5]. Because the ionization energy (IE) of diazomethane is low (experimentally determined at 9.00 eV) [6], the 3s, 3p and 3d Rydberg states are located in the same region as the valence states [6]. The ion’s excited states are at 14.13 and 15.13 eV, and the parent ion is quite stable [6]. The most intense electronic transition of diazomethane is 2 1 A 1 !1 1 A 1 ("* ! "). The antibonding character of the target molecular orbital makes this state unbound, dissociating adiabatically to N 2 ( 1 # g + ) + CH 2 ( 1 A 1 ), in agreement with Herzberg’s observations [7,8]. As of yet, the Rydberg states have not been fully characterized. An earlier theoretical study estimated them to lie vertically at 5.89 (3s), 6.65$6.87 (3p) and 7.48!7.68 (3d) eV [5]. However, the proximity of valence states can lead to interactions with core electrons, giving rise to valence-Rydberg character and affecting quantum defects. A more recent study suggested that all states at 5.31!7.29 eV contain some Rydberg character, and that the 2 1 A 1 ("* ! ") 72 state, found at 5.53 eV vertically is mostly valence and has the largest oscillator strength [9]. Figure 3.1 Molecular orbitals relevant to ground and excited electronic states of CH 2 N 2 . Absorption spectra identify the strongest absorption at 260!190 nm (4.77!6.53 eV), a region that includes the intense 2 1 A 1 !1 1 A 1 ("*! ") transition [10,11]. The absorption exhibits diffuse structures at 230, 218, and 214 nm, which may belong to several mixed valence/Rydberg states. At yet shorter wavelengths (154!193 nm), Merer observed in absorption a series of perpendicular bands, which he characterized as having 1 B 1 and 1 B 2 symmetry [12]. Some had a resolved 73 K structure, whereas others were more diffuse. Specifically, in the region near 190 nm he assigned the bands to transitions to three Rydberg states, most probably 3p, which he denoted D( 1 B 1 ), E( 1 B 2 ), and F( 1 B 1 ). He also identified perturbations among some of these states, and attributed them to Coriolis coupling. In addition, two intense and diffuse band systems, which do not fit any Rydberg series, are centered at 175 and 140 nm and may indicate mixed states containing large contribution of valence "* ! " excitations [9,12]. To date, no molecular beam studies of diazomethane have been reported because of difficulties in preparing and delivering it intact into the interaction chamber. In an attempt to develop efficient sources of carbenes, we have recently adapted the traditional MNNG (N-methyl-N’nitro-N-nitrosoguanidine) + KOH preparation method of diazomethane [13,14] for work in molecular beams. We also modified our inlet and pulsed-nozzle systems for stable and safe delivery. We have used 2 + 1 resonance enhanced multiphoton ionization (REMPI) complemented by photoelectron imaging of the excited states to characterize Rydberg states in the region around 190 nm previously studied by Merer [12] We have obtained REMPI and photoelectron spectra and identified intense, K-resolved transitions not seen in the one-photon absorption experiments of Merer. We assign them to the 2 1 A 2 ! 1 1 A 1 (3p y ! ") Rydberg transition allowed only in two-photon absorption. Moreover, we obtain for the first time photoelectron spectra from excited states of diazomethane. These spectra reveal interactions between electronic states. The accompanying high-level electronic structure calculations reported here 74 determine ground, Rydberg, and ionic state geometries and vertical transition energies. Combined with experiments, they enable us to characterize the nature of the excited electronic states and their interactions. Our results show that the observed transitions in the region 51,000!58,900 cm -1 (6.32-7.30 eV) can be assigned to three electronic Rydberg states: the 2 1 A 2 (3p y ! ") state, which can be excited by a two-photon transition, and the 2 1 B 1 (3p z ! ") and 3 1 A 1 (3p x ! ") states, which can be accessed by both one- and two-photon transitions. We show that specific vibronic levels of the Rydberg 2 1 A 2 (3p y ) and 2 1 B 1 (3p z ) states interact due to accidental resonances. In addition, we find that the Rydberg 3 1 A 1 (3p x ) wavefunction is strongly mixed with a dissociative valence 2 1 A 1 ("*! ") state. We propose that the one-photon absorptions seen by Merer in this wavelength region are due primarily to vibrational progressions in the 2 1 B 1 (3p z ) state, and that some levels are mixed with the 2 1 A 2 (3p y ) state, which is dark in one-photon excitation. In Chapter 4, we report assignments of other vibronic levels of the excited Rydberg states of diazomethane and its isotopologs and compare them to vibrational states of the ground state ion. The complete vibronic analysis confirms the assignments proposed in this Chapter. 3.2 Experimental Details The experimental arrangement has been described before in Chapter 2 and in Ref. [15,16] and only relevant experimental details are presented here. In particular, 75 we describe the synthesis of CH 2 N 2 , a reactive and explosive compound, and its transport to the molecular beam without decomposition. For CH 2 N 2 synthesis an all glass vacuum line, constructed of greaseless Teflon® high vacuum valves and o-ring joints, was used. This line consisted of three sections: a reactor, two traps and a vacuum manifold. The reactor was fitted with a pressure-equalized addition funnel. CH 2 N 2 was prepared under vacuum by the reaction of 2.6 g of N-methyl-N'-nitro-N-nitrosoguanidine (TCI America) with an excess of aqueous ~ 15 ml of NaOH (2.5M). The resulting CH 2 N 2 was purified by passing the gas though two traps held at $ 78 o C (195 K) with a dry ice/ethanol slush, and collected in a 12 L glass flask housed in a steel mesh box and protected from exposure to light. The pressure of CH 2 N 2 is kept less than 30 Torr; otherwise it will begin to condense in the dry ice/ethanol traps where it is likely to explode. The IR and UV spectra of CH 2 N 2 prepared by this procedure were in good agreement with published spectra [10,11,17,18], and contained negligible amounts of impurities. The sample survived for several days until it was depleted by use. A molecular beam of CH 2 N 2 was obtained by supersonic expansion of a gaseous mixture containing ~ 1.5% CH 2 N 2 in helium at a backing pressure of 2 atm into the source chamber. This mixture was introduced into the source chamber through a pulsed piezoelectric nozzle (10 Hz, diameter 0.5 mm) and the molecular beam was skimmed before entering the ionization chamber. Since CH 2 N 2 decomposes on metallic surfaces, metal parts of the pulsed nozzle making contact with the gas mixture were replaced with fiberglass-reinforced polyetherimide 76 (Ultem® 3040). Black nylon tubing was used to make connections. The rotational temperature of the skimmed molecular beam was not determined directly, but based on expansion conditions and previous results it is estimated to be ~ 10 K. We wish to emphasize that CH 2 N 2 is a toxic and hazardous gas, which can decompose explosively and spontaneously, and thus appropriate safety precautions must be taken. CH 2 N 2 should be handled only at low pressures and on a small scale. At no time should the gas be allowed to condense into the liquid phase. Throughout our experiments a pressure of 25 torr was never exceeded. Safety equipment (safety shields, safety glasses, face shields, leather gloves and protective clothing, such as leather suits, Kevlar® sleeves and earplugs) must be used at all times. Care must be taken to avoid known triggers of CH 2 N 2 decomposition such as intense light, exposure to rough, metallic, or acidic surfaces, and abrupt changes in temperature, pressure, and phase. In order to characterize the excited states of CH 2 N 2 in the region 51,000!58,900 cm -1 (6.32!7.30 eV), two types of experiments were carried out. In the first set, CH 2 N 2 was excited by a resonant two-photon transition at 51,000!58,900 cm -1 (6.32!7.30 eV) and ionized by absorbing non-selectively another photon of the same wavelength. A 2 + 1 REMPI spectrum was recorded by monitoring the signal from the parent ion (m/e = 42) as a function of laser excitation wavelength. The tunable UV laser beam (0.7!1.0 mJ focused by a 45 cm focal length lens) was generated by frequency doubling (Inrad Autotracker III) the linearly polarized output of a Nd:YAG (Spectra Physics GCR230) laser pumped 77 dye laser system (Continuum ND6000, LDS 751, 25!30 mJ). No attempt was made to normalize the spectrum (the laser power varied by ± 30 %). Ions produced in the ionization chamber were extracted and accelerated by three-stage ion optics into a linear time-of-flight (TOF) mass spectrometer. Only ions of m/e = 42 (CH 2 N + 2 ) were detected using a 40 mm diameter microchannel-plate (MCP) detector with a coupled phosphor screen (Burle Electro-optics, Inc.). The ion optics and TOF region were shielded from magnetic fields by a µ-metal tube. Signals from the MCP were amplified, collected with a digital oscilloscope (Tektronix TDS 3054), and transferred to a PC for analysis. In the second set of experiments, CH 2 N 2 was again ionized by 2 + 1 REMPI and images of the ejected photoelectrons were obtained at selected wavelengths in the 51,000!58,900 cm -1 (6.32!7.30 eV) region. Using the velocity map imaging technique [19], event counting and centroiding [20,21], and the basis set expansion (BASEX) Abel transform method [22], photoelectron kinetic energy (eKE) distributions were determined from the images. Electrons produced in the ionization chamber were extracted and accelerated by the ion optics into the TOF mass spectrometer and detected with the same MCP detector. Signals from the detector were monitored with a CCD camera (LaVision, Imager 3, 12 bit, 1280x1024 pixel array), and transferred to a PC for analysis. 78 3.3 Computational Details The equilibrium geometry and vibrational frequencies of neutral CH 2 N 2 were calculated by CCSD(T) [23,24], using the cc-pVTZ basis [25], and by B3LYP [26], using the 6-311G(2df,p) basis. Geometry and frequencies of the ground state of the cation were calculated by CCSD(T)/cc-pVTZ, using the unrestricted (UHF) reference. Vertical excitation energies from the ground state were calculated using EOM-CCSD[27,28]/6-311(3+,+)G* at the B3LYP optimized geometry. The basis sets were derived from the polarized split-valence 6-311G(d,p) basis by adding additional polarization and diffuse functions [29,30]. The error bars [31] for equilibrium geometries for CCSD(T)/cc-pVTZ are much lower than the geometrical changes observed in this study. Nonsystematic errors for bond lengths are on the order of about 0.002 Å. CCSD(T) underestimates bond angles by about 0.5°, and has a nonsystematic error of about 1°. EOM-CCSD excitation energies are accurate to within 0.1!0.3 eV [32]. A basis set with adequate diffuse functions is necessary for states with Rydberg character. As in our previous work [33], the assignment of valence and Rydberg character to the excited states was based on: (i) the symmetry of the transition, (ii) leading EOM-CCSD amplitudes and the character of the corresponding HF orbitals, and (iii) the second moments <X 2 >, <Y 2 >, and <Z 2 > of the EOM-CCSD electron density. The character of the HF orbitals was determined using the Molden interface [34]. All optimizations, frequencies, and excited state calculations were performed using 79 the Q-Chem [35] and ACES II [36] electronic structure programs. Franck-Condon factors were modeled using the PES4 program.[37] 3.4 Computational Results 3.4.1 Equilibrium geometries and ionization energies The equilibrium geometries of neutral diazomethane in its ground (1 1 A 1 ) and Rydberg 3p excited states, as well as of the cation in its ground (1 2 B 2 ) state, are shown in Figure 3.2. Both the CCSD(T) and B3LYP optimized geometries agree with experimentally determined values [38], which are also summarized in this figure. The first ionization removes an electron from the highest-occupied, out-of- plane " orbital (Figure 3.1) with vertical IE of 8.95 eV, as calculated by EOM-IP- CCSD [39-41] with 6-311G(2df,p) basis set. The experimental vertical IE is 9.00 eV [6], and the adiabatic IE calculated at CCSD(T)/cc-pVTZ geometries and corrected by zero point energy is 8.84 eV. The second ionization energy, which corresponds to removing an electron from the " * ! orbital (Figure 3.1) in the plane of the molecule, is much higher and equals 14.00 eV vertically, at the EOM-IP- CCSD level. Calculated structural changes resulting from the first ionization are consistent with removing an electron from a bonding " orbital. The CN bond length increases substantially (by 0.064 Å) due to a decrease in bond order. At the same time, the NN bond length slightly contracts (by 0.024 Å) due to the antibonding nature of the " orbital along the NN bond. The increase in HCH angle relative to the neutral (by 2.8°) is due to depleted electron density along the CN bond, which 80 allows the two CH bonds to move away from each other to minimize electrostatic repulsion. The molecule has C 2v symmetry in both the neutral and the cation ground states. Figure 3.2 Left panel: Ground state equilibrium structures (Å and deg) of CH 2 N 2 for: the neutral (1 1 A 1 ) at CCSD(T)/cc-pVTZ (regular print) and B3LYP/6- 311G(2df,p) (italics), and for the cation (1 2 B 1 (% ! ")) at CCSD/6-311G** (underlined). The corresponding nuclear repulsion energies are: 61.280112, 61.514227, and 61.118198 hartrees, respectively. Right panel: Excited state equilibrium structures for the 2 1 A 2 (3p y ! "), 2 1 B 1 (3p z ! "), and 3 1 A 1 (3p x ! ") states shown in normal, italics, and underlined print, respectively. CNN and HCNN refer to respective angle and dihedral angle for the C s 3 1 A 1 (3p x ! ") state. The corresponding nuclear repulsion energies are: 60.705257, 59.502297, and 60.715012 hartrees. Experimental parameters of ground state: CN length: 1.300 Å; NN length: 1.139 Å; CH length: 1.075 Å; HCH angle: 126.0 o . 3.4.2 Excited electronic states The vertical excitation energies, oscillator strengths, transition dipole moments, and second moments of the excited states of CH 2 N 2 are listed in Table 3.1. The excitation energies and oscillator strengths are also plotted as a stick spectrum in Figure 3.3. In the C 2v point group, states of A 2 symmetry have zero oscillator strength in one-photon excitation; these are depicted by hollow bars. The electronic 81 spectrum is fairly dense in the 5.00!8.00 eV region and dominated by Rydberg transitions. All the calculated singlet states derive from excitations from the highest occupied " orbital. The lowest singlet excited state, which is well separated from the others and lies at 3.21 eV, corresponds to the valence 1 1 A 2 ! 1 1 A 1 ("*! ") transition. In one- photon absorption, this state was observed as a weak, broad band centered at 3.14 eV. This one-photon forbidden transition becomes weakly allowed by vibronic couplings with other electronic states. Figure 3.3 The bars show calculated vertical excitation energies of CH 2 N 2 . Allowed and forbidden transitions are indicated by filled and hollow bars, respectively. 82 Table 3.1 Vertical excitation energies ("E vert , eV), oscillator strengths (f L ), dipole strengths (! 2 tr , a. u.), and changes in second dipole moment of charge distributions ("<R 2 >, (a.u.) 2 ) for the excited states of CH 2 N 2 at EOM-CCSD/6-311(3+,+)G*. a State "E vert f L ! 2 tr "<X 2 > "<Y 2 > "<Z 2 > 1 1 A 2 (# * $ ! #) 3.21 0 0 1 -2 1 1 1 B 1 (3s ! #) 5.33 0.0200 0.1186 15 10 6 2 1 A 1 (#* ! #) 5.85 0.2100 1.4904 5 16 14 2 1 A 2 (3p y ! #) 6.35 0 0 39 10 7 2 1 B 1 (3p z ! #) 6.39 0.0100 0.0644 13 11 41 3 1 B 1 (3d 2 z ! #) 7.08 0.0100 0.0431 24 22 74 3 1 A 1 (3p x ! #) 7.15 0.0700 0.3821 23 69 16 3 1 A 2 (3d yz ! #) 7.23 0 0 53 15 48 4 1 B 1 (4s ! #) 7.34 0.0010 0.0071 75 76 33 4 1 A 1 (3d xz ! #) 7.36 0.0001 0.0005 22 68 63 5 1 A 1 (4p x ! #) 7.77 0.0100 0.0582 69 210 82 a At the B3LYP/6-311G(2df,p) optimized geometry; E CCSD = $ 148.426979 hartree. In the 5.00!6.00 eV region, we found two transitions: the Rydberg 1 1 B 1 ! 1 1 A 1 (3s ! ") transition at 5.33 eV, and the valence transition 2 1 A 1 ! 1 1 A 1 ("* ! ") at 5.85 eV. In agreement with the antibonding valence character of the upper state, the latter transition has oscillator strength an order of magnitude higher than the Rydberg’s (oscillator strength 0.21 compared to 0.02) due to high spatial overlap between the " and "* orbitals. In this region (5.50!7.00 eV) Rabalais assigned a single intense, broad band centered at 5.70 eV 83 based on known data [42], but diffuse structures have also been reported [10,11]. The neighboring Rydberg state is likely hidden under the intense valence band. Between 6.00 and 7.00 eV our calculations predict only two electronic transitions: the 2 1 A 2 ! 1 1 A 1 (3p y ! ") and 2 1 B 1 ! 1 1 A 1 (3p z ! ") at vertical excitation energies 6.34 and 6.38 eV, respectively. In the region 7.00!7.25 eV, three electronic transitions have been calculated (Table 3.1): 3 1 B 1 ! 1 1 A 1 (3d 2 z ! ") (vertical excitation energy 7.08 eV), 3 1 A 1 ! 1 1 A 1 (3p x ! ") (7.15 eV), and 3 1 A 2 ! 1 1 A 1 (3d yz ! ") (7.23 eV). The latter has zero oscillator strength due to symmetry and could only borrow intensity. The 3 1 B 1 ! 1 1 A 1 (3d 2 z ! ") transition has very low oscillator strength; also, its pure Rydberg character does not imply any broadening. The calculated vertical excitation energies, superimposed on the 2 + 1 REMPI spectrum, are shown in Figure 3.4. The 3 1 A 1 ! 1 1 A 1 (3p x ! ") Rydberg transition has oscillator strength about one order of magnitude greater than transitions to the other 3p states due to mixing with the valence 1 1 A 2 ! 1 1 A 1 ("*! ") transition of the same symmetry. This is confirmed by the composition of its EOM-CCSD wavefunction, which contains a large contribution from the "* ! " configuration. Both the intensity and diffuse nature of the two peaks observed in the 56,860 – 58,900 cm -1 region (see below) suggest assignment of this band as the 3 1 A 1 ! 1 1 A 1 (3p x ! ") Rydberg transition. The vertical and adiabatic excitation energies for the Rydberg 3p states and the 84 cation ground state (1 2 B 1 ) are compared with the experimental result in Table 3.2. The 3 1 A 1 ! 1 1 A 1 (3p x ! ") transition shows the largest difference between its vertical and adiabatic excitation energies (0.24 eV), whereas the 2 1 A 2 ! 1 1 A 1 (3p y ! ") and 2 1 B 1 ! 1 1 A 1 (3p z ! ") transitions exhibit much smaller differences (~ 0.1 eV) and are within 0.02 eV of each other. The magnitude of the vertical- adiabatic relaxation in these states is close to that in the cation, supporting their description as pure, cation-like Rydberg states. The 3 1 A 1 ! 1 1 A 1 (3p x ! ") transition’s large difference indicates strong Rydberg/valence mixing. Table 3.2 Calculated vertical (!E vert ) and adiabatic (!E ad ) excitation eneries and quantum defects (!) and the corresponding experimental values. State "E vert "E ad "E vert-ad E exp " calc " exp 2 1 A 2 (3p y ! #) 6.35 6.24 0.11 6.48 a 0.73 0.68 2 1 B 1 (3p z ! #) 6.39 6.30 0.09 6.52 b 0.72 0.66 3 1 A 1 (3p x ! #) 7.15 6.91 0.24 7.05 c 0.29 0.36 1 2 B 1 (% ! #) 8.95 8.85 0.10 9.00 d - - a Measured at 52,227 cm -1 (K’ = 0 # K” = 0, and K’ = 1 # K” = 1). b Measured at 52,679 cm -1 (K’ = 0 #K” = 1). c Measured at 56,898 cm -1 . d Ref [6]. Referring to the equilibrium geometries of the states shown in Figure 3.2, we find that the 2 1 A 2 (3p y ) and 2 1 B 1 (3p z ) states are more similar in structure to the 1 2 B 1 cation than to the neutral. The 3 1 A 1 (3p x ) state, in contrast, is different than both the cation and the other two 3p states. Whereas the 1 2 B 1 cation and the 2 1 A 2 , and 2 1 B 1 neutral excited states retain the C 2v symmetry of the neutral ground state, the 3 1 A 1 (3p x ! ") state has C s equilibrium structure. The CN bond length increases 85 and the CNN bond is no longer linear (161°, Figure 3.2). The plane of the CH 2 sp 2 center is also broken by about 3°. This twisted geometry reflects the partial population of the strongly antibonding "* orbital. Figure 3.4 Survey 2 + 1 REMPI spectrum in the 6.32!7.30 eV (51,000!58,900 cm -1 ) region. The empty bars indicate the calculated values of vertical excitation energies. The 3p y and 3p z orbitals in diazomethane are oriented to maximize overlap with the positive centers of the nuclear charge distribution. According to natural bond order (NBO) analysis [43], about half of the + 1 charge of the nuclear core is accommodated by the hydrogens. The lobes of the 3p y orbital have highest density 86 on top of the hydrogen atoms (Figure 3.1). Donation into the electron-deficient CH bonds leads to a contracted CH bond length relative to the cation, and a larger HCH bond angle (Figure 3.2). A similar effect along the CNN bonds is prevented by symmetry. In contrast, the 3p z orbital has significant density along the CNN axis. The resulting electronic donation leads to contracted CN and NN bond lengths relative to the cation and a smaller HCH angle to minimize electrostatic repulsion. Similar effects have been seen in the electronic structure of several vinyl radicals [33]. 3.5 Experimental Results The 2 + 1 REMPI spectrum of CH 2 N 2 recorded by monitoring the parent mass peak (m/e = 42) as a function of laser excitation at 51,000!58,900 cm -1 (6.32!7.30 eV) is shown in Figure 3.4, whereas Figure 3.5 shows an expanded region (52,000!55,000 cm -1 ; 6.45!6.82 eV) recorded at higher resolution (smaller step-size). The dependence of the parent ion signal on laser pulse energy, determined at several excitation wavelengths, is nearly quadratic. This suggests that the observed spectral features derive from two-photon resonances. The laser frequency was calibrated to published lines of diazomethane in the region ~ 52,512!52,721 cm -1 [12]. In order to characterize the C!F bands near the observed onset of the absorption in this region, photoelectron kinetic energy (eKE) distributions were obtained by imaging following ionization of selected levels of the excited state(s). 87 Figures 3.6!3.9 show photoelectron images obtained at excitation wavelengths & = 382.69 nm (2h' = 52,262 cm -1 ; middle peak of band C), 380.75 nm (52,528 cm -1 ; middle peak of band D), 380.39 nm (52,577 cm -1 ; band E), and 379.50 nm (52,700 cm -1 ; middle peak of band F), respectively, and the corresponding eKE distributions. In all images, the electric field vector of the laser is set parallel to the vertical direction of the image plane. The relative width of the peaks is "E/E = 3.0 ± 0.2 %, limited by instrument resolution. There is one strong, narrow band in each of the eKE distributions obtained at & = 382.69 nm (at 5,742 cm -1 ; Figure 3.6) and & = 380.39 nm (at 5,889 cm -1 ; Figure 3.8). Both photoelectron images have an isotropic angular distribution. The eKE distributions obtained at & = 380.75 nm (52,528 cm -1 ) and at 379.50 nm (52,700 cm -1 ) are also similar to each other (Figures 3.7 and 3.9) but different from those shown in Figures 3.6 and 3.8. First, each has two strong narrow bands (at 5,754 cm -1 and 6,172 cm -1 in Figure 3.7 and 5,994 cm -1 and 6,414 cm -1 in Figure 3.9), and in each, the two peaks are separated by ~ 420 ± 20 cm -1 . Second, both distributions exhibit a weak progression of three peaks with ~ 400 ± 20 cm -1 separation. Last, the strong outermost ring in both images has an anisotropic angular distribution, whereas the strong inner rings are isotropic. It is difficult to characterize the angular distributions of the photoelectrons corresponding to peaks of the weak vibrational progression because of their low intensities. The main difference between the two eKE distributions shown in Figures 3.7 and 3.9 is in the relative intensities of the two intense features. The band at ~ 7.05!7.30 eV ( ~ 56,860!58,900 cm -1 ) shown 88 in Figure 3.4 is much more diffuse than the other bands in this region. Figure 3.10 shows a photoelectron image and the corresponding photoelectron KE distribution obtained at & = 351.09 nm (56,965 cm -1 ). It has an intense band at 8,098 cm -1 , as well as a progression of three peaks at 11,020, 11,992, and 12,800 cm -1 , and is very different from the distributions shown in Figures 3.6!3.9. The relative width of the peaks "E/E = 8.7 ± 0.2 %, is broader than the instrumental resolution. Figure 3.5 2 + 1 REMPI spectrum of diazomethane in the region of excitations to the 1 A 2 (3p y ) and 1 B 1 (3p z ) states obtained by measuring m/e = 42 as a function of excitation energy. The laser wavelength increment was 0.005 nm. See the text for details of the C!F bands. 89 Figure 3.6 Photoelectron image and the corresponding eKE distributions obtained at excitation wavelength & = 382.69 nm (2h' = 52,262 cm -1 ; middle peak of band C). 90 Figure 3.7 Photoelectron image and the corresponding eKE distributions obtained at excitation wavelength & = 380.75 nm (52,528 cm -1 ; middle peak of band D). 91 Figure 3.8 Photoelectron image and the corresponding eKE distributions obtained at excitation wavelength & = 380.39 nm (52,577 cm -1 ; band E). 92 Figure 3.9 Photoelectron image and the corresponding eKE distributions obtained at excitation wavelength & = 379.50 nm (52,700 cm -1 ; middle peak of band F). 93 Figure 3.10 Photoelectron image and the corresponding eKE distributions obtained at excitation wavelength & = 351.09 nm (56,965 cm -1 ). 94 3.6 Discussion Promotion of an electron from the highest occupied " molecular orbital of CH 2 N 2 (Figure 3.1) into 3p atomic-like orbitals generates the 3 1 A 1 (3p x ! "), 2 1 A 2 (3p y ! "), and 2 1 B 1 (3p z ! ") Rydberg states. These states correlate with the ground state of the ion, 1 2 B 1 (% ! "). For small molecules, the excitation energies of the Rydberg states can often be approximated by the Rydberg formula [8]: E Ryd = IE $109,737.3/(n - !) 2 cm -1 , (3.1) where E Ryd is the excitation energy of the Rydberg state (in cm -1 ), IE is the adiabatic ionization energy (in cm -1 ), n is the principal quantum number, and ! is the quantum defect. The calculated and experimental quantum defects are listed in Table 3.2. Typical values for n = 3 Rydberg states are ~ 0.9%1.2, 0.4%0.7, and ~ 0.1 for s, p, and d states, respectively. The calculated and experimental quantum defects for the 3 1 A 1 (3p x ) state fall outside this region, and differ from the values for the other two Rydberg states, in agreement with its partial valence composition. Below we discuss first the 2 1 A 2 (3p y ! ") and 2 1 B 1 (3p z ! ") Rydberg states, followed by a discussion of the 3 1 A 1 (3p x ! ") state. 3.6.1 The 2 1 A 2 (3p y " ") and 2 1 B 1 (3p z " ") Rydberg States and Their Interaction The internal energy (E int ) of the ion corresponding to each measured photoelectron peak is obtained from the peak position of the eKE distribution: E int = 3h'!$ eKE – IE. (3.2) 95 In the 2 + 1 REMPI spectrum of CH 2 N 2 shown in Figure 3.5, the first group of three peaks (band C; 52,227!52,295 cm -1 ) was not observed in the one-photon absorption spectrum reported by Merer. We assign this group to the origin band of the 2 1 A 2 ! 1 1 A 1 transition for the following reasons. According to electric-dipole transition selection rules in C 2v symmetry, the A 2 ! A 1 transition is optically forbidden in one-photon but allowed in two-photon excitation. The onset of this transition is close to the calculated one and its quantum defect (! = 0.68) is typical of a Rydberg p state. The Rydberg nature of the state is also confirmed by the photoelectron image (Figure 3.6) obtained at & = 382.69 nm (52,262 cm -1 ). From the correpsonding eKE distribution shown in Figure 3.6, we conclude that the CH 2 N 2 + ions are generated in the vibrational ground state. This suggests a similarity in geometries of the neutral excited state and ion ground state, which results in a propensity for ionization via the diagonal 1 2 B 1 ! 2 1 A 2 (% ! 3p y ) 0 0 0 transition. The internal energy of the ion obtained by excitation through the 0 0 0 band is ~ 60 cm -1 , indicating little rotational excitation in the neutral parent. Similar photoelectron kinetic energy distributions are obtained for the other two peaks in this triad. Additional support for this assignment is obtained from the ab initio calculations, which show the presence of only two electronic states in this region (Table 3.1): the 2 1 A 2 (3p y ! ") state at 6.34 eV (51,213 cm -1 ) and the 2 1 B 1 (3p z ! ") state at 6.38 eV (51,535 cm -1 ). The 1 B 1 ! 1 A 1 transition is optically allowed in one- photon excitation but the 1 A 2 ! 1 A 1 is not, and therefore the observed band 96 located at 52,227 - 52,295 cm -1 , which has not been seen in one-photon absorption, is assigned as the band origin of 2 1 A 2 ! 1 1 A 1 (3p y ! ") transition. The three components of the band are attributed to resolved K structure. CH 2 N 2 in its neutral ground state (1 1 A 1 ) is a near-symmetric prolate top (A’’ = 9.112 cm -1 , B’’ = 0.377 cm -1 , and C’’ = 0.362 cm -1 ) [8], and according to calculations its 3p Rydberg states also have C 2v symmetry (except 3 1 A 1 (3p x )). In C 2v transitions to the vibronic bands of A 1 and A 2 symmetry are expected to be governed by "K = 0, ± 2 selection rules with a 4A” spacing between K bands [44]. Transitions to B 1 and B 2 vibronic bands are governed by "K = ± 1 selection rules, with a 2A” spacing. The separation of ~ 35 cm -1 between the three peaks in this groups, which is approximately equal to 4A’’, indicates that the rotational transition is governed by "K= 0, ± 2 selection rules and that the excited state is of A symmetry. All the observed transitions originate in K = 0 and 1, the only ones significantly populated in the supersonic expansion. The first peak, in order of increasing energy, arises from the K’ = 0 ! K” = 0 and K’ = 1 ! K” = 1 transitions. The next two components of this triad correspond to the K’ = 2 ! K” = 0, and K’ = 3 ! K” = 1 transitions, respectively. In this triad, the outermost bands are more intense because they include transitions from K” = 1, which according to nuclear-spin statistics are x3 more intense than those from K” = 0. The next three groups of bands at 52,513!52,541 cm -1 , 52,574 cm -1 , and 52,679!52,722 cm -1 were observed in one-photon absorption and assigned by Merer as transitions to three different electronic states denoted as D( 1 B 1 ), E( 1 B 2 ), 97 and F( 1 B 1 ), respectively. According to the ab initio calculations, there exist only two electronic states (2 1 B 1 (3p z ) and 2 1 A 2 (3p y )) in this energy region. The absence of other states within the error bars of the method make this assignment conclusive. The photoelectron spectra can aid in assigning the spectra. The bands with partially resolved rotational structure at 52,513!52,541 cm -1 (D) and 52,679!52,722 cm -1 (F) are assigned as mixed bands composed of the 9 1 0 transition to the 2 1 A 2 (3p y ) state and the band origin of the 2 1 B 1 (3p z ) ! 1 1 A 1 transition. This assignment is based on the two-peak structure in the eKE distributions shown in Figures 3.7 and 3.9. Comparing the two eKE distributions, which have equal energy separations but different peak heights, suggests that two electronic states are coupled. The difference in anisotropy of the strong inner and outer rings also suggests that the photoelectrons corresponding to these rings may be ejected from different electronic states. In Merer's absorption experiments, the optically forbidden 9 1 0 band of the 2 1 A 2 (3p y ) ! 1 1 A 1 transition becomes allowed by vibronic coupling to the 2 1 B 1 (3p z ) state mediated by the non-totally symmetric " 9 # (b 2 ) vibration. B 1 symmetry in the 2 1 A 2 state is obtained for A 2 (electronic symmetry) & b 2 (vibrational symmetry) vibronic bands. Because the eKE distributions in Figures 3.7 and 3.9 display only two major peaks, we use a two state approximation to obtain the energy of the deperturbed states and the coupling matrix elements. The two molecular eigenstates and are expressed as linear combinations: 98 ! " # =#$% 1 0 0 0 (a 1 ), 1 B 1 (3p z ) { } +&% 2 9 0 1 (b 2 ), 1 A 2 (3p y ) { } , (3.3) ! " + =#$ 1 0 0 0 (a 1 ), 1 B 1 (3p z ) { } +%$ 2 9 0 1 (b 2 ), 1 A 2 (3p y ) { } . (3.4) We obtain ! " 2 and ! " 2 from the relative peak heights in the two photoelectron distributions, giving an average ratio of . Using , we obtain and , and putting these values into the two-state coupled equations (see Appendix B), we finally obtain: E 1 = 52,638 cm -1 ; E 2 = 52,590 cm -1 ; and V 12 ~ 83 cm -1 , where E 1,2 are the deperturbed energies of the two states, and V 12 is the coupling matrix element. The eKE distributions show that the F band (52,679!52,722 cm -1 ) has a larger contribution from the 2 1 B 1 (3p z ) state, and therefore E 1 = 52,638 cm -1 is assigned as the adiabatic origin of this state. Its associated quantum defect, ( = 0.65, is typical of a Rydberg p state. The peaks in the eKE distributions shown in Figures 3.7 and 3.9 correspond to vibrational levels " + 0 , " + 9 , and " + 4 !+ n" + 9 (n=0-2) in the CH 2 N + 2 ion (where " + 9 is the CNN in-plane bend and " + 4 is the CN stretch). The frequency of " + 9 is ~ 420 ± 10 cm -1 , and the frequencies of " + 4 !+ n" + 9 combination modes are 1,002 ± 20, 1,404 ± 30, and 1,808 ± 40 cm -1 for n = 0, 1 and 2, respectively. The existence of a vibrational progression suggests that there is a small geometry change between the Rydberg and the cation states. Calculated vibrational frequencies for the ground state of the 1 2 B 2 cation are given in Table 3.3 for 99 comparison. The " + 4 !fundamental band was observed before in the He I PES with a frequency of 970 ± 80 cm -1 [6]. Table 3.3 CCSD(T)/cc-pVTZ harmonic vibrational frequencies for cation ground state (1 2 B 1 ). Mode Assignment Symmetry Frequency, cm -1 " + 1 CH 2 symmetric stretching a 1 3164 " + 2 NN stretching a 1 2199 " + 3 CH 2 symmetric bending a 1 1432 " + 4 CN stretching a 1 1001 " + 5 CNN bending (out-of-plane) b 1 440 " + 6 CH 2 wagging b 1 712 " + 7 CH 2 asymmetric stretching b 2 3311 " + 8 CH 2 rocking b 2 1133 " + 9 CNN bending (in-plane) b 2 377 The three K bands of each of the mixed transitions of B 1 symmetry are separated by ~ 14 cm -1 and ~ 21 cm -1 (approximately 2A’’). This indicates that these transitions are governed by "K = ± 1 selection rules and confirms that the excited states have B symmetry. The contributions to the three components, in order of increasing energy, are from the K’ = 0 ! K” = 1, K’ = 1 ! K” = 0, and K’ = 2 ! K” = 1 transitions. The band at 52,574 cm -1 [E( 1 B 2 ), Figure 3.8] is assigned as the 5 1 0 transition to 2 1 A 2 (3p y ). The single peak in the corresponding eKE is assigned as " + 5 (b 1 ; CNN 100 out-of-plane bend) of CH 2 N + 2 whose frequency is 318 ± 10 cm -1 . The slight difference between the value reported here and that given by Merer [33] (52,598.7 cm -1 and 52,668.6 cm -1 for the K’ = 2 ! K” = 3, and K’ = 2 ! K” = 1 transitions, respectively) reflects the lower rotational temperature in our experiment, where only K” = 0, 1 are significantly populated. All the other bands in the region 52,000!55,000 cm -1 are fairly sharp and have a typical triad K structure. Their photoelectron spectra display single peaks and they are assigned to unperturbed transitions to the 2 1 A 2 (3p y ! ") state. A detailed analysis of the vibronic spectrum will be given in a separate publication. Here we note only that from the widths of the bands, we conclude that the 2 1 A 2 (3p y ! ") state is bound and only weakly predissociative. Finally we compare briefly our interpretation of the coupled states to the one offered by Merer. As discussed above, we see evidence only for coupling of two bands, D and F, of B 1 vibronic symmetry, and no evidence for coupling in the E( 1 B 2 ) band. In contrast, Merer concluded that all three bands interact via Coriolis coupling. This conclusion was reached in analogy with the interaction of the ' " 5 , ' " 6 , and ' " 9 vibrational modes of the neutral ground state of diazomethane (1 1 A 1 ) [45], which couple the b 2 level via Coriolis interaction with two b 1 levels. The difference may derive from the rotational excitation of the parent. Merer carried out his measurements at 300 and 196 K, where K’ levels up to K’ = 10 are populated, whereas in our molecular beam measurements (T rot ~ 10 K) only K’ ) 2 are 101 significantly populated. Using the Coriolis constants given by Merer, we obtain that the values of the *V FE * and *V ED * Coriolis coupling matrix elements for K’ = 10 are ~ 82 and ~ 142 cm -1 , respectively, whereas for K’ = 2 the corresponding terms are 5 times smaller ( ~ 16 and ~ 28 cm -1 , respectively). The direct coupling matrix element between the F and D state obtained in our study, *V FD * ~ 83 cm -1 , is much higher than the *V FE * and *V ED * Coriolis coupling matrix elements for K’ = 2. These considerations and the photoelectron images support our interpretation that at low rotational temperatures there is significant interaction only between the two states of 1 B 1 symmetry. 3.6.2 The 3 1 A 1 (3p x " ") State In the 57,000 cm -1 region of the one-photon absorption spectrum two strong diffuse bands spaced by ~ 430 ± 20 cm -1 were observed by Merer [33]. These bands did not fit the previous progressions and it was suggested that they represent more than one electronic state. As discussed above, our EOM-CCSD study of the excited states of CH 2 N 2 using the 6-311(3+,+)G* basis set reveals several states in this region (Figure 3.4): 3 1 B 1 (3d 2 z ! "), 3 1 A 1 (3p x ! "), and 3 1 A 2 (3d yz ! "). The latter cannot be observed in one-photon absorption due to symmetry considerations. The 3 1 B 1 (3d 2 z ! ") state has very low oscillator strength, and its pure Rydberg character should not result in the significant broadening observed in the measurements. The transition to the 102 1 A 1 (3p x ) state has the largest oscillator strength among the Rydberg states in this region due to its mixing with the valence 2 1 A 1 ("*! ") state. Therefore the 3 1 A 1 ! 1 1 A 1 (3p x ! ") transition is the best candidate for these bands. Calculations of the geometries of the excited states show that the geometry of the 3 1 A 1 (3p x ) state differs considerably from that of 2 1 A 2 (3p y ! "), 2 1 B 1 (3p z ! "), and the cation 1 2 B 1 (% ! "). The 3 1 A 1 (3p x ! ") state has an in-plane bent geometry corresponding to C s symmetry, as discussed in Section 3.4.2. The different geometries of the neutral 3 1 A 1 (3p x ) and the cation 1 2 B 1 states should give rise to a propensity for ionization via off-diagonal transitions, resulting in multiple peaks in the photoelectron spectrum as indeed seen in the experiment. The eKE distribution obtained at & = 351.09 nm (56,965 cm -1 ) has an intense band at 8,098 cm -1 , and a progression of three peaks at 11,020, 11,992, and 12,800 cm -1 (Figure 3.10). This progression can be assigned to excitation of the" + 6 (CH 2 wag) mode in the cation. The experimental frequencies for " + 6 and 2" + 6 are 808 ± 30 and 1,780 ± 70 cm -1 , respectively, suggesting that these modes are rather anharmonic. The theoretical harmonic frequency for the " + 6 mode is 712 ± 20 cm -1 . The strong peak at 8,089 cm -1 can be assigned to excitation of either " + 1 !+ " + 3 or " + 7 !+ " + 3 vibrations (4,765 ± 190 cm -1 ) in the cation (where" + 1 and " + 3 are the CH 2 symmetric stretch and bend, respectively and " + 7 is the CH 2 asymmetric stretch). The eKE distributions at & = 351.51 nm (56,898 cm -1 ) and & = 349.28 nm 103 (57,261 cm -1 ) have similar character. The diffuse nature of the bands is attributed to their strong coupling to the 2 1 A 1 ("*! ") valence state, which lends them oscillator strength and is repulsive. The observed dissociation products, N 2 ( 1 # g + ) + CH 2 ( 1 A 1 ), indeed correlate with a molecular state of A 1 symmetry. 3.7 Conclusions The electronic states of diazomethane in the region 3.00!8.00 eV have been obtained by ab initio calculations, and transitions in the region 6.32!7.30 eV have been characterized experimentally using a combination of 2 + 1 REMPI spectroscopy and photoelectron imaging in a molecular beam. Specifically, in the region where experiments have been carried out we find that only three Rydberg 3p states are excited, the 2 1 A 2 (3p y ! "), 2 1 B 1 (3p z ! "), and 3 1 A 1 (3p x ! "). The former two states are of mostly pure Rydberg character, whereas the 3 1 A 1 (3p x ! ") state is mixed with the valence 2 1 A 1 ("* ! ") state and is strongly predissociative. We find that the ground vibrational level of the 2 1 B 1 (3p z ) state is mixed with the 2 1 A 2 (3p y ) ' 9 level, which is of B 1 vibronic symmetry. The other 2 1 A 2 (3p y ) vibronic states exhibit pure Rydberg character and generate ions in single vibrational states. They also predissociate rather slowly. Thus, this state can serve both as a sensitive and state-specific two-photon diagnostic for diazomethane and as a gateway state for preparation of diazomethane ions in specific vibrational levels. The photoelectron spectra of the 3 1 A 1 (3p x ! ") state, on the other hand, give rise to many states of the ion. The two-photon excitation scheme used here excited 104 efficiently vibronic levels in the 2 1 A 2 (3p y ) state, whereas vibronic levels of the 2 1 B 1 (3p z ) state are best reached by one-photon absorption, as done by Merer [12]. The experimental and theoretical results agree very well both in regard to excitation energies and to vibrational modes of the ion. Analysis of the photoelectron spectra of the excited states elucidates the interactions between the Rydberg electron with the ion core and reveals mixings with valence states. 105 3.8 Chapter 3 References 1. W. Kirmse, Carbene Chemistry. 2nd ed. 1971: Academic: New York. 622 pp. 2. F. Raulin, M. Khlifi, M. Dang-Nhu, and D. Gautier, Adv. Space Res., 12, 11/181, (1992). 3. F.W.N. Kirkbride, Ronald G. W., J. Chem. Soc., 119, (1933). 4. H. Okabe, Photochemistry of Small Molecules. 1978: (Wiley, New York, N. Y.). 432 pp. 5. S.P.G. Walch, William A., III., J. Am. Chem. Soc., 97, 5319, (1975). 6. J. Bastide and J.P. Maier, Chem. Phys., 12, 177, (1976). 7. G. Herzberg, Proc. R. Soc. London, Ser. A, 262, 291, (1961). 8. G. Herzberg, Electronic Spectra and Electronic Structure of Polyatomic Molecules (Molecular Spectra and Molecular Structure). Vol. III. 1966: Van Nostrand, New York. 875 pp. 9. M.P. Habas and A. Dargelos, Chem. Phys., 199, 177, (1995). 10. R. Brinton and D. Volman, J. Chem. Phys., 19, 1394, (1951). 11. R.J. Wolf and W.L. Hase, J. Phys. Chem., 82, 1850, (1978). 12. A.J. Merer, Can. J. Phys., 42, 1242, (1964). 13. A.F.W. McKay, George F., J. Am. Chem. Soc., 69, 3028, (1947). 14. A.F. McKay, J. Am. Chem. Soc., 70, 1974, (1948). 15. A. Sanov, T. Droz-Georget, M. Zyrianov, and H. Reisler, J. Chem. Phys., 106, 7013, (1997). 16. V. Dribinski, Photoelectron and Ion Imaging Studies of the Mixed Valence/Rydberg Excited States of the Chloromethyl Radical, CH2Cl, and the Nitric Oxide Dimer, (NO)2, Dissertation, Chemistry, University of Southern California, Los Angeles, 2004. 17. C. Moore and G. Pimentel, J. Chem. Phys., 40, 329, (1964). 106 18. M. Khlifi, P. Paillous, P. Bruston, F. Raulin, and J.C. Guillemin, Icarus, 124, 318, (1996). 19. A.T.J.B. Eppink and D.H. Parker, Rev. Sci. Instrum., 68, 3477, (1997). 20. Y. Tanaka, M. Kawasaki, Y. Matsumi, H. Fujiwara, T. Ishiwata, L.J. Rogers, R.N. Dixon, and M.N.R. Ashfold, J. Chem. Phys., 109, 1315, (1998). 21. B.Y. Chang, R.C. Hoetzlein, J.A. Mueller, J.D. Geiser, and P.L. Houston, Rev. Sci. Instrum., 69, 1665, (1998). 22. V. Dribinski, A. Ossadtchi, V.A. Mandelshtam, and H. Reisler, Rev. Sci. Instrum., 73, 2634, (2002). 23. K. Raghavachari, G.W. Trucks, J.A. Pople, and M. Headgordon, Chem. Phys. Lett., 157, 479, (1989). 24. J.D. Watts, J. Gauss, and R.J. Bartlett, J. Chem. Phys., 98, 8718, (1993). 25. T.H. Dunning Jr., J. Chem. Phys., 90, 1007, (1989). 26. A.D. Becke, J. Chem. Phys., 98, 5648, (1993). 27. H. Koch, H.J.A. Jensen, P. Jorgensen, and T. Helgaker, Journal of Chemical Physics, 93, 3345, (1990). 28. J.F. Stanton and R.J. Bartlett, J. Chem. Phys., 98, 7029, (1993). 29. R. Krishnan, J.S. Binkley, R. Seeger, and J.A. Pople, J. Chem. Phys., 72, 650, (1980). 30. A.D. Mclean and G.S. Chandler, J. Chem. Phys., 72, 5639, (1980). 31. T. Helgaker, P. Jorgensen, and J. Olsen, Molecular Electronic-Structure Theory. 2000. 908 pp. 32. S. Hirata, M. Nooijen, and R.J. Bartlett, Chem. Phys. Lett., 326, 255, (2000). 33. L. Koziol, S.V. Levchenko, and A.I. Krylov, J. Phys. Chem. A, 110, 2746, (2006). 34. G. Schaftenaar and J.H. Noordik, J. Comput. Aided Mater. Des., 14, 123, (2000). 107 35. J. Kong, C.A. White, A.I. Krylov, D. Sherrill, R.D. Adamson, T.R. Furlani, M.S. Lee, A.M. Lee, S.R. Gwaltney, T.R. Adams, C. Ochsenfeld, A.T.B. Gilbert, G.S. Kedziora, V.A. Rassolov, D.R. Maurice, N. Nair, Y.H. Shao, N.A. Besley, P.E. Maslen, J.P. Dombroski, H. Daschel, W.M. Zhang, P.P. Korambath, J. Baker, E.F.C. Byrd, T. Van Voorhis, M. Oumi, S. Hirata, C.P. Hsu, N. Ishikawa, J. Florian, A. Warshel, B.G. Johnson, P.M.W. Gill, M. Head-Gordon, and J.A. Pople, J. Comput. Chem., 21, 1532, (2000). 36. J.F. Stanton, J. Gauss, J.D. Watts, W.J. Lauderdale, and R.J. Bartlett, ACES II, 1993. The package also contains modified versions of the MOLECULE Gaussian integral program of J. Almlöf and P. R. Taylor, the ABACUS integral derivative program written by T. U. Helgaker, H. J. Aa. Jensen, P. Jørgensen, and T. P. Taylor, and the PROPS property evaluation integral code of P. R. Taylor. 37. A.D. Wesley, Study of Radicals, Clusters and Transition State Species by Anion Photoelectron Spectroscopy, Chemistry, University of California Berkley, Berkley, 1994. 38. J.a.P.J.H. T.H. Dunning, in Modern Theoretical Chemistry, ed. e.b.H.F. Schaefer. Vol. 3. 1977: Plenum Press, New York. 1. 39. D. Sinha, S. Mukhopadhyay, and D. Mukherjee, Chem. Phys. Lett., 129, 369, (1986). 40. R. Chaudhuri, D. Mukhopadhyay, and D. Mukherjee, Chem. Phys. Lett., 162, 393, (1989). 41. D. Sinha, S.K. Mukhopadhyay, R. Chaudhuri, and D. Mukherjee, Chem. Phys. Lett., 154, 544, (1989). 42. J.W. Rabalais, J.M. Mcdonald, V. Scherr, and S.P. Mcglynn, Chemical Reviews, 71, 73, (1971). 43. E.D.B. Glendening, J. K.; Reed, A. E.; Carpenter, J. and J.A.M. E.; Bohmann, C. M.; Weinhold, F., NBO. 2001: Theoretical Chemistry Institute, University of Wisconsin: Madison, WI,. 44. E.C.E. Lim, Excited States. Vol. 3. 1977: (Academic, New York, N. Y.). 351 pp. 45. C.B. Moore, J. Chem. Phys., 39, 1884, (1963). 108 Chapter 4. Vibronic Structure and Ion Core Interactions in Rydberg States of Diazomethane: An Experimental and Theoretical Investigation 4.1 Introduction Diazomethane (CH 2 N 2 ) has been the subject of considerable interest, because its photolysis and pyrolysis provide efficient sources of singlet and triplet methylene [1], and its spectroscopy is important for the formation of molecules in N 2 -rich media such as the atmospheres of Titan, Triton, and Pluto [2-4]. In a previous publication we characterized the three 3p Rydberg states of diazomethane by using resonance enhanced multiphoton ionization (REMPI) spectroscopy, photoelectron velocity map imaging (VMI) and high-level electronic structure calculations [5]. Emphasis was placed on the effect of electronic state interactions on the spectroscopy, and both Rydberg-Rydberg and Rydberg-valence interactions were identified and analyzed. The two main resonant structures of ground state diazomethane are shown in Figure 4.1. Both structures are ionic and have positively charged central nitrogen, with the negative charge localized on either the terminal nitrogen or the carbon atom. The two structures differ by the hybridization of the carbon: sp 2 versus sp 3 . Figure 4.1 The two Lewis structures for diazomethane. The z axis is along CNN, the y axis is in the plane, perpendicular to CNN, and the x axis is out of plane.! 109 The present work is centered on the characterization of the normal modes of the 3p Rydberg states of diazomethane and its isotopologs. By comparing the normal mode frequencies of the 3p Rydberg states to those of the ground state of the neutral (1 1 A 1 ) and the cation (1 2 B 1 ), we analyze the influence of the unpaired electron in each of the 3p orbitals on the structure and vibrational motions in the Rydberg states. The strategy adopted in this work is the following. Using high level theory we calculate the normal mode harmonic frequencies of the target states for CH 2 N 2 , CD 2 N 2 , and CHDN 2 , and compare them to available experimental results. Some experimental frequencies for the neutral and ion ground states and the 2 1 B 1 (3p z ) Rydberg state are available in the literature [6-17], and we complement those with new experimental data on the normal modes of the 2 1 A 2 (3p y ) Rydberg state and the ground-state cation. In our previous work [5], 2 + 1 REMPI spectra and photoelectron VMI of the excited states of CH 2 N 2 have been used for the first time to characterize the spectrum of diazomethane in a molecular beam, and these studies are extended here to the isotopologs of diazomethane and to higher excitation energies (51,750!58,500 cm -1 ). The excellent agreement between theory and experiment allows us to present a full discussion of the influence of the 3p Rydberg electron on the vibrational frequencies of the corresponding excited states as compared to those of the ground states of the neutral and the cation. In the 2 + 1 REMPI spectrum of the 2 1 A 2 (3p y ! !) transition of CH 2 N 2 obtained before [5], strong K-resolved transitions not seen in one-photon 110 absorption[16] were observed. Using a combination of experiment and theory, the upper states of the observed transitions were assigned, in order of increasing energy, to the 2 1 A 2 (3p y ! !), 2 1 B 1 (3p z ! !), and 3 1 A 1 (3p x ! !) Rydberg states. Although the out-of-plane 3p x Rydberg orbital is usually the least perturbed by the molecular core, the spectrum associated with this state is found to be more perturbed than those associated with the 3p y and 3p z states whose unpaired electrons occupy in- plane orbitals. The 2 + 1 REMPI signal for the 3 1 A 1 (3p x ! !) state is broader and several times lower in peak intensity than that for transitions to the 2 1 A 2 (3p y ) or 2 1 B 1 (3p z ) states. This broadening is shown by ab initio calculations to result from mixing of the 3 1 A 1 (3p x ) Rydberg state with the dissociative valence 2 1 A 1 (!*! !) state [5], which shortens the lifetime of this state and reduced its ionization efficiency. Also, the geometry of the 3 1 A 1 (3p x ) state is quite different from the cation (C 2v ) having C s symmetry [5], in contrast to the 2 1 A 2 (3p y ) and 2 1 B 1 (3p z ) states, which like the cation have C 2v symmetry. In addition, analyses of photoelectron kinetic energy (eKE) distributions of CH 2 N 2 indicate that the band origin of the 2 1 B 1 (3p z ) state is mixed with the 2 1 A 2 (3p y ) " 9 level, which is of B 1 vibronic symmetry [5]. However, most of the other bands in its 2 + 1 REMPI spectrum can be assigned as pure transitions to the 2 1 A 2 (3p y ) state. The Chapter 4 is organized as follows. In Section 4.2 we describe the 2 + 1 REMPI and VMI techniques used here and the procedures employed for recording REMPI and photoelectron spectra. Section 4.3 presents experimental 111 results for the three isotopologs of diazomethane, and describes rotational analyses, the assignment of the band origins of the transitions to the 2 1 A 2 (3p y ) and 2 1 B 1 (3p z ) states, and the mixings of vibronic levels of these Rydberg states. Section 4.4 describes the electronic structure models, and the results of calculations of geometries and vibrational frequencies of the neutral and cation ground states and the three 3p Rydberg states for the isotopologs of diazomethane. In Section 4.5, after discussing the proposed assignments, we present a detailed analysis of the structure and normal mode frequencies of Rydberg states of diazomethane and their dependence on the Rydberg electron. The main results and conclusions are summarized in Section 4.6. 4.2 Experimental Details The experimental setup, techniques, and CH 2 N 2 synthesis have been described in detail in Chapter 2 and elsewhere [5,18,19], and only changes and modifications are elaborated upon here. The method used to produce CD 2 N 2 and CHDN 2 is based on the earlier one for production of CH 2 N 2 . This method for simultaneously producing CH 2 N 2 , CD 2 N 2 , and CHDN 2 has the advantage that isotopologs can be prepared using protonated precursors and solvents, and only the aqueous NaOD in D 2 O needs to be deuterated. The same glass vacuum line was used for the synthesis [5]. In the modified procedure, CH 2 N 2 , CD 2 N 2 , and CHDN 2 were generated under vacuum in a closed reactor by the reaction of 2.6 g of N-methyl-N'-nitro-N- nitrosoguanidine (TCI America ® ) dissolved in 30 mL of Tetra (ethylene glycol) 112 dimethyl ether, 99 % (Aldrich ® ) with an excess of ~ 7.5 ml aqueous solution of NaOD (2.5 M) and 7.5 ml of NaOH (2.5 M) mixture. When using only ~ 15 ml of aqueous of NaOD (2.5M) solution, CD 2 N 2 in isotopic purity of up to 94 % and high overall yield [20] is prepared. The solution was stirred for ~ 10 minutes at 0 o C (273 K) and expanded through two traps held at # 78 o C (195 K) with a dry ice/ethanol slush in an 12 L glass flask that was evacuated, protected from exposure to light, and housed in a steel mesh box. A mixture of approximately 0.5 % CH 2 N 2 , 0.5 % CD 2 N 2 , and 0.5 % CHDN 2 in He at 2 atm total pressure was prepared in this flask. This mixture was introduced into the source chamber of the differentially pumped vacuum system. The IR and UV spectra of CH 2 N 2 , CD 2 N 2 , and CHDN 2 prepared by this procedure were in good agreement with published spectra, [6-9,21,22] and contained negligible amounts of impurities. Based on the IR spectra, samples of CH 2 N 2 :CD 2 N 2 :CHDN 2 ~ 1:1:1 are generated; they survive for several days until depleted by use. 2 + 1 REMPI spectra of CH 2 N 2 , CD 2 N 2 , and CHDN 2 were recorded simultaneously by integrating parent ion peaks of m/e = 42, and 44, 43, respectively, as a function of laser excitation wavelength. The UV laser radiation (0.4!1.1 mJ focused by a 45 cm focal length lens) was generated by frequency doubling (Inrad Autotracker III) the linearly polarized output of a Nd:YAG (Spectra Physics GCR230) pumped dye laser system (Continuum ND6000, LDS 751 and LDS 698, 25!35 mJ). No attempt was made to normalize the spectrum. 113 Using VMI [23], we recorded photoelectron images at wavelengths corresponding to state-selected rovibronic levels in the excited 3p Rydberg states, as described before [5,18,19]. For energy calibration, via the A 2 " + state was used [19]. The peaks in the photoelectron kinetic energy (eKE) distributions allowed us to determine vibrational frequencies in the resulting cation. The photoelectrons were extracted and accelerated by the ion optics toward the MCP detector. Signals from the detector were monitored with a CCD camera, transferred to a PC and accumulated. The eKE distributions were determined from the recorded images, by using event counting and centroiding [24,25] and the basis set expansion (BASEX) Abel transform method [26]. Caution! We wish to emphasize that CH 2 N 2 is a toxic, hazardous, and potentially explosive gas, which can decompose violently and spontaneously, and thus appropriate safety precautions must be taken. CH 2 N 2 should be handled only at low pressures and on a small scale, and at no time should the gas be allowed to condense into the liquid phase. The pressure must be kept at < 25 torr. Safety equipment (blast shields, safety glasses, face shields, leather gloves and protective clothing such as leather suits, Kevlar® sleeves and earplugs) must be used, and care must be taken to avoid known triggers of CH 2 N 2 decomposition such as intense light, exposure to rough, metallic, or acidic surfaces, and abrupt changes in temperature, pressure, or phase. 114 4.3 Experimental Results and Analysis 4.3.1 REMPI spectra and eKE distributions Figure 4.2 displays the 2 + 1 REMPI spectra of (a) CH 2 N 2 , (b) CD 2 N 2 , and (c) CHDN 2 obtained by recording parent ion masses of m/e = 42, 44, and 43, respectively, as a function of two-photon laser excitation from 51,750 to 54,900 cm -1 . Figure 4.2a also includes an inset of the 54,900 to 56,700 cm -1 spectrum that shows the CH stretch region. The observed band positions together with the proposed assignments are listed in Tables 4.1!4.3 (see also Section 4.5). Bands characterized as pure vibronic transitions are assigned by their positions in the 2 + 1 REMPI spectra and the corresponding eKE distributions obtained by VMI. The latter typically have a single, narrow band corresponding to the diagonal transition between the vibronic level in the Rydberg state and the corresponding vibrational level of the cation. An example is shown in Figure 4.3, which displays the eKE distribution for CD 2 N 2 obtained at excitation wavelength $ = 381.14 nm (2h" = 52,478 cm -1 ), and which has one strong, narrow band at 5,857 cm -1 assigned as " 5 + (see Section 4.5). Most of the vibronic bands of the three isotopologs of diazomethane exhibit similar single peaks in their eKE distributions. 115 Figure 4.2 2 + 1 REMPI spectra for (a) CH 2 N 2 (b) CD 2 N 2 (c) CHDN 2 following two-photon laser excitation at 51,750!54,900 cm -1 . An inset in (a) for CH 2 N 2 shows a 54,900!56,700 cm -1 spectrum magnified ten times, whereas an inset in (b) diplays the 54,5000!55,000 cm -1 range in x10 magnification. 116 Table 4.1 Transition energies and vibrational assignments a for the 2 1 A 2 (3p y ) ! 1 1 A 1 transition of CH 2 N 2 . b Two – photon energies #E c Intensity Assignments K’ ! K” 52227 0 0 0 0 0 ! 0, 1 ! 1 52262 35 strong 0 0 0 2 ! 0 52295 68 0 0 0 3 ! 1 52513(52575) d 286(348) 9 1 0 0 ! 1 52528(52590) 301(363) strong 9 1 0 1 ! 0 52541(52603) 314(376) 9 1 0 2 ! 1 52574 347 strong 5 1 0 52850 623 weak (6 1 0 ) 52947 720 weak (5 2 0 ;6 1 0 ;9 2 0 ;) 53016 789 weak (5 2 0 ;6 1 0 ;9 2 0 ;) 53196 969 4 1 0 0 ! 0, 1 ! 1 53230 1003 strong 4 1 0 2 ! 0 53265 1038 4 1 0 3 ! 1 53296 1069 weak (9 3 0 ) 53359 1132 weak (5 3 0 ) 53459 1232 5 1 0 6 1 0 53489 1262 weak 5 1 0 6 1 0 53522 1295 5 1 0 6 1 0 53585 1358 (6 2 0 ) 53619 1392 medium (6 2 0 ) 53654 1427 (6 2 0 ) 53690 1463 3 1 0 0 ! 0, 1 ! 1 53723 1496 strong 3 1 0 2 ! 0 53756 1529 3 1 0 3 ! 1 54289 2062 2 1 0 0 ! 0, 1 ! 1 54324 2097 strong 2 1 0 2 ! 0 54358 2131 2 1 0 3 ! 1 54627 2400 3 1 0 4 1 0 54662 2435 weak 3 1 0 4 1 0 54697 2470 3 1 0 4 1 0 54986 2759 2 1 0 6 1 0 55006 2779 weak 2 1 0 6 1 0 55027 2800 2 1 0 6 1 0 55286 3059 1 1 0 0 ! 0, 1 ! 1 117 Table 4.1 Continued. 55318 3091 weak 1 1 0 2 ! 0 55353 3126 1 1 0 3 ! 1 55597 3370 weak 7 1 0 a Where known, the K assignment is also given. Tentatively assigned frequencies are enclosed in parentheses. b All values in cm -1 (uncertaity ± 5 cm -1 ). c #E are energies of transitions determined with respect to the band origin. d Deperturbed band positions are shown in parentheses. Table 4.2 Transition energies and vibrational assignments a for the 2 1 A 2 (3p y ) ! 1 1 A 1 transition of CD 2 N 2 . b Two – photon energies #E c Intensity Assignments K’ ! K” 52196 # 18 0 0 0 0 ! 2 52214 0 0 0 0 0 ! 0, 1 ! 1, 2 ! 2 52231 17 strong 0 0 0 2 ! 0 52249 35 0 0 0 3 ! 1 52266 52 0 0 0 4 ! 2 52461 247 5 1 0 52478 264 strong 5 1 0 52495(52526) d 281(312) 9 1 0 0 ! 1 52507(52538) 293(324) strong 9 1 0 1 ! 0 52804 590 (6 1 0 ) 52818 604 medium (6 1 0 ) 52832 618 (6 1 0 ) 52908 694 9 2 0 52925 711 weak 9 2 0 52943 729 9 2 0 53115 901 4 1 0 0 ! 2 53133 919 4 1 0 0 ! 0, 1 ! 1, 2 ! 2 53150 936 strong 4 1 0 2 ! 0 53168 954 4 1 0 3 ! 1 53191 977 4 1 0 4 ! 2 53241 1027 3 1 0 0 ! 2 53258 1044 3 1 0 0 ! 0, 1 ! 1, 2 ! 2 53276 1062 strong 3 1 0 2 ! 0 53293 1079 3 1 0 3 ! 1 53310 1096 3 1 0 4 ! 2 118 Table 4.2 Continued. 53334 1120 6 2 0 0 ! 2 53352 1138 6 2 0 0 ! 0, 1 ! 1, 2 ! 2 53369 1155 strong 6 2 0 2 ! 0 53386 1172 6 2 0 3 ! 1 53404 1190 6 2 0 4 ! 2 54265 2051 medium 2 1 0 0 ! 0, 1 ! 1, 2 ! 2 54282 2068 2 1 0 2 ! 0 54299 2085 2 1 0 3 ! 1 54316 2102 medium 3 2 0 2 ! 0 54333 2119 3 2 0 3 ! 1 54403 2189 1 1 0 0 ! 0, 1 ! 1, 2 ! 2 54420 2206 medium 1 1 0 2 ! 0 54437 2223 1 1 0 3 ! 1 54558 2344 weak 7 1 0 a Where known, the K assignment is also given. Tentatively assigned frequencies are enclosed in parentheses. b All values in cm -1 (uncertaity ± 5 cm -1 ). c #E are energies of transitions determined with respect to the band origin. d Deperturbed band positions are shown in parentheses. Table 4.3 Transition energies and vibrational assignments a for the 2 1 A 2 (3p y ) ! 1 1 A 1 transition of CHDN 2 . b Two – photon energies #E c Intensity Assignments K’ ! K” 52197 # 24 0 0 0 0 ! 2 52221 0 0 0 0 0 ! 0, 1 ! 1, 2 ! 2 52245 24 strong 0 0 0 2 ! 0 52268 47 0 0 0 3 ! 1 52291 70 0 0 0 4 ! 2 52500(52543) d 279(322) 9 1 0 52507(52550) 286(329) strong 9 1 0 52520(52563) 299(342) 9 1 0 52546(52589) 325(368) 9 1 0 52845 624 (6 1 0 ) 52865 644 weak (6 1 0 ) 52887 666 (6 1 0 ) 52910 689 9 2 0 119 Table 4.3 Continued. 52931 710 weak 9 2 0 52953 732 9 2 0 53180 959 4 1 0 0 ! 2 53204 983 4 1 0 0 ! 0, 1 ! 1, 2 ! 2 53229 1008 strong 4 1 0 2 ! 0 53251 1030 4 1 0 3 ! 1 53275 1054 4 1 0 4 ! 2 53526.32 1305.32 3 1 0 0 % 0, 1 ! 1, 2 ! 2 53548.67 1327.67 strong 3 1 0 2 ! 0 53569.7 1348.7 3 1 0 3 ! 1 53835 1614 (3 1 0 9 1 0 ) 53858 1637 weak (3 1 0 9 1 0 ) 53878 1657 (3 1 0 9 1 0 ) 54281 2060 2 1 0 0 ! 0, 1 ! 1, 2 ! 2 54305 2084 strong 2 1 0 2 ! 0 54328 2107 2 1 0 3 ! 1 54458 2237 1 1 0 0 ! 0, 1 ! 1, 2 ! 2 54472 2251 medium 1 1 0 2 ! 0 54487 2266 1 1 0 3 ! 1 55197 2976 weak 7 1 0 a Where known, the K assignment is also given. Tentatively assigned frequencies are enclosed in parentheses. b All values in cm -1 (uncertainty ± 5 cm -1 ). c #E are energies of transitions determined with respect to the band origin. d Deperturbed band positions are shown in parentheses. Mixed levels involving the band origin of the 2 1 B 1 (3p z ) Rydberg state and the " 9 level of the 2 1 A 2 (3p y ) Rydberg state in CD 2 N 2 and CHDN 2 were characterized as described before for CH 2 N 2 [5]. Briefly, the eKE distributions of CD 2 N 2 and CHDN 2 were similar in shape to those of CH 2 N 2 at spectral positions where the mixed levels were expected [5], and Figures 4.4 and 4.5 show representative examples. The distributions for CD 2 N 2 obtained at $ = 380.93 nm (52,507 cm -1 ) and at $ = 379.51 nm (52,700 cm -1 ) are similar to one another (Figure 4.4). Each 120 has two prominent, narrow bands (at 5,799 and 6,136 cm -1 and at 6,089 and 6,429 cm -1 in Figures 4.4a and 4.4b, respectively) separated by ~ 340 ± 20 cm -1 . The eKE distribution in Figure 4.4b has an additional peak (at 5,514 cm -1 ), which does not appear in Figure 4.4a. In the case of CHDN 2 , the images and the corresponding eKE distributions depicted in Figures 4.5a ($ = 380.81 nm (52,520 cm -1 )) and 4.5b ($ = 379.53 nm (52,697 cm -1 )) are similar to those shown in Figure 4.4. The separation between the two most intense bands (at 5,807 and 6,182 cm -1 and at 6,078 and 6,453 cm -1 in Figures 4.5a and 4.5b, respectively) is ~ 375 ± 20 cm -1 . The eKE distribution in Figure 4.5b has additional peaks at 5,493 and 5,157 cm -1 . Similar to CH 2 N 2 [5], the strong outermost ring in the CD 2 N 2 and CHDN 2 images has an anisotropic angular distribution, whereas the strong inner rings are isotropic. The relative width of the peaks is #E/E = 3.0 ± 0.2 %, close to the instrument resolution. Figure 4.3 The eKE distribution for CD 2 N 2 obtained from the photoelectron image at excitation wavelength $ = 381.14 nm (2h" = 52,477.7 cm -1 ). 121 Figure 4.4 The eKE distributions for CD 2 N 2 obtained from photoelectron images at excitation wavelength (a) $ = 380.93 nm (52,507 cm -1 ) and (b) $ = 379.51 nm (2h" = 52,700 cm -1 ). 122 Figure 4.5 The eKE distributions for CHDN 2 obtained from photoelectron images at excitation wavelength (a) $ = 380.81 nm (52,520 cm -1 ) and (b) $ = 379.53 nm (52,697 cm -1 ). 123 4.3.2 Spectroscopic Analysis: Band Origins, K-Structure and State Interactions CH 2 N 2 , CD 2 N 2 and CHDN 2 in their neutral ground states are near-symmetric prolate tops [6,27,28]. According to electronic structure calculations, the 3p y and 3p z Rydberg states as well as the ground-state cation are planar, similarly to the neutral ground state. Thus, in accordance with the Franck!Condon principle, the origin bands in the 2 + 1 REMPI spectrum and bands associated with the totally symmetric vibrational (a 1 ) levels should have the highest intensities, as indeed observed. The K-structure can aid in assigning the symmetry of the transitions. For CH 2 N 2 and CD 2 N 2 , vibronic transitions from the vibrational ground state to vibronic bands of A 1 and A 2 symmetry are governed by #K = 0, ± 2 selection rules with a 4A’ spacing between K bands, whereas transitions to B 1 and B 2 vibronic bands are governed by a #K = ± 1 selection rule with a 2A’ spacing [29]. The electronic wavefunctions of the 3p y and 3p z Rydberg states of CHDN 2 have C 2v symmetry and, therefore, follow the same selecton rules. For CH 2 N 2 we found before that all the observed transitions originate from K’’ = 0 and 1, and those originating from K’’ = 1 are the most intense, in accordance with nuclear-spin statistics [5]. Because of the smaller rotational constants of CD 2 N 2 and CHDN 2 , rotational bands originating from K’’ = 2 are also observed. For CD 2 N 2 , rotational transitions from K’’ = 0 and 2 have twice the intensity of those from K’’ = 1 because of nuclear-spin statistics. These intensity 124 patterns explain the differences in relative intensities of vibronic bands associated with different K transitions, and we use them, whenever possible, in assigning the vibronic transitions (see Section 4.5). In experiments employing VMI [5,18,19,23], the internal energy (E int ) of the ion corresponding to each measured photoelectron peak is calculated from the peak position in the eKE distribution: E int = 3h"!# eKE – IE, (4.1) where the adiabatic ionization energy IE was taken as the published value for (9.00 ± 0.02 eV; 72,585 ± 160 cm -1 ) [17]. This value was confirmed by our photoelectron spectra, and was the same, within the error bar, for all isotopologs (see below). The 2 + 1 REMPI spectrum of CH 2 N 2 in the 51,750!56,700 cm -1 energy region is shown in Figure 4.2a, and transitions in the 52,227!52,722 cm -1 region have been assigned before [5]. The most intense group of bands, at 52,227!52,295 cm -1 , belongs to the origin band of the 2 1 A 2 ! 1 1 A 1 (3p y ! !) transition. The three components of this band, which are spaced by ~ 4A”, correspond to K rotational structure (see Table 4.1). The strong bands with partially resolved rotational structure at 52,513!52,541 cm -1 and 52,679!52,722 cm -1 were assigned before as mixed bands comprised of the 9 1 0 transition to the 2 1 A 2 (3p y ) state and the band origin of the 2 1 B 1 ! 1 1 A 1 (3p z ! !) transition [5]. These bands were deperturbed by using a two-level approximation. The deperturbed origin band of the 2 1 B 1 ! 1 1 A 1 125 (3p y ! !) transition was determined at 52,638 cm -1 and the coupling matrix element was found to be 83 cm -1 [5]. The 2 + 1 REMPI spectra of CD 2 N 2 and CHDN 2 in the 51,750 to 54,900 cm -1 energy region are shown in Figures 4.2b and 4.2c, respectively, and the band assignments follow those for CH 2 N 2 . For CD 2 N 2 , the origin band of the 2 1 A 2 ! 1 1 A 1 (3p y ! !) transition is at 52,196!52,266 cm -1 . The five components of this band, which are spaced by ~ 4A”, correspond to the K rotational structure. The lower energy peak arises from the K’ = 0 ! K” = 2 transitions. The next one arises mostly from the K’ = 0 ! K” = 0 and K’ = 1 ! K” = 1 transitions with some contribution from K’ = 2 ! K” = 2. The most intense peak corresponds to the K’ = 2 ! K” = 0 transition, and the last two peaks are from the K’ = 3 ! K” = 1 and K’ = 4 ! K” = 2 transitions, respectively. Analyses of the two-peak structures in the eKE distributions for CD 2 N 2 obtained at $ = 380.93 nm (52,507 cm -1 ) and at $ = 379.51 nm (52,700 cm -1 ) (Figure 4.4) show that the mixing of the 2 1 A 2 (3p y ) 9 1 0 transition and the band origin of 2 1 B 1 ! 1 1 A 1 (3p z ! !) is only slightly weaker than in CH 2 N 2 . This is because the " ' 9 mode (CNN in-plane bend) does not involve distortion of the CH 2 group and therefore the mixing does not change significantly with H/D substitution. A deperturbation analysis (using a two-level approximation) places the 2 1 B 1 origin of CD 2 N 2 at 52,664 cm -1 with a coupling matrix element of 71 cm -1 . 126 The peaks in the eKE distributions shown in Figure 4.4b correspond to vibrational levels " + 0 and " + 9 (340 ± 20 cm -1 ) of the cation. The peaks in the eKE distributions obtained at 52,495 cm -1 and 52,507 cm -1 (Figure 4.4a) have similar relative intensities and peak separations (340 ± 20 cm -1 ). The peak at 5,514 cm -1 in Figure 4.4b is assigned to the " + 4 mode (915 ± 20 cm -1 ). The three peaks in the REMPI spectrum at 52,690!52,710 cm -1 are assigned, in order of increasing energy, to the K’ = 0 ! K” = 1, K’ = 1 ! K” = 0, and K’ = 2 ! K” = 1 transitions, respectively. No transitions from K” = 2 were detected. For CHDN 2 the origin band of the 2 1 A 2 ! 1 1 A 1 (3p y ! !) transition is at 52,197!52,291 cm -1 and similarly to CD 2 N 2 , transitions originating from K” = 2 were observed. The spacing between the five components of the K rotational structure is ~ 4A” and the proposed K assignments are given in Table 4.3. The eKE distributions for CHDN 2 obtained at $ = 380.81 nm (52,520 cm -1 ) and $ = 379.53 nm (52,697 cm -1 ) (Figure 4.5) are similar to those shown in Figure 4.4. Analysis of the structure in the eKE distributions indicates that the interaction between the 2 1 A 2 (3p y ) 9 1 0 transition and the band origin of the 2 1 B 1 ! 1 1 A 1 (3p z ! !) transition is of intermediate strength between CH 2 N 2 and CD 2 N 2 . A deperturbation analysis (using a two-level approximation) places the 2 1 B 1 origin at 52,648 cm -1 with a coupling matrix element of 76 cm -1 . The peaks in the eKE distributions shown in Figure 4.5 correspond to vibrational levels " + 0 and " + 9 (375 ± 20 cm -1 ). The peaks at 5,493 and 5,157 cm -1 127 in Figure 4.5b are assigned to " + 4 and " + 4 + " + 9 (where " + 9 is the CNN in-plane bend and " + 4 is the CN stretch) in the CHDN + 2 ion, which have frequencies ~ 960 ± 20and 1,296 ± 30 cm -1 , respectively. The three peaks in the REMPI spectrum at 52,684 – 52,710 cm -1 are assigned, in order of increasing energy, to the K’ = 0 ! K” = 1, K’ = 1 ! K” = 0, and K’ = 2 ! K” = 1 transitions to the perturbed 2 1 B 1 (3p z ! !) state, respectively. No transitions originating from K” = 2 were detected for this state. 4.4 Computational Details The equilibrium geometry and vibrational frequencies of neutral diazomethane were calculated by CCSD(T) [30,31]/cc-pVTZ [32] and B3LYP [33]/6- 311G(2df,p). Geometry and frequencies of the ground state of the cation were calculated by CCSD(T)/cc-pVTZ using the unrestricted (UHF) reference. The basis sets were derived from the polarized split-valence 6-311G(d,p) basis by adding additional polarization and diffuse functions [34,35]. Isotope shifts for the ground-state neutral and the cation were calculated at the B3LYP/6-311G(2df,p) level. The equilibrium structures, frequencies, and isotope shifts of the 2 1 A 2 (3p y ) and 2 1 B 1 (3p z ) Rydberg states were calculated by EOM-EE- CCSD [36-38] with 6-311(3+,+)G(2df). While equilibrium structures, excitation energies, and most of the skeletal frequencies are reasonably converged with just a single polarization set, additional polarization functions are crucial for out-of-plane 128 vibrations, probably due to re-hybridization induced by those motions. For instance, the CH 2 wagging frequency in the 2 1 A 2 (3p y ) and 2 1 B 1 (3p z ) states increases by 93 and 74 cm -1 , respectively, upon increasing polarization from 6-311(3+,+)G(d) to 6-311(3+,+)G(2df). A smaller 6-31(2+)G(d) basis was employed for (lower- symmetry) 3 1 A 1 (3p x ) state to reduce computational costs. Vertical excitation energies and adiabatic IE’s were calculated using EOM- CCSD/6-311(3+,+)G(d) and EOM-IP-CCSD [39-41]/6-311G(2df,p), respectively, at the B3LYP/6-311G(2df,p) optimized geometries. All optimizations, frequencies, and excited state calculations were performed using the Q-Chem [42] and ACES II [43] electronic structure programs. The Natural Bond Orbital (NBO) program [44] was employed to analyze bonding in neutral, electronically excited and ionized diazomethane. All equilibrium geometries are summarized in Table 4.4. Note that within the Born-Oppenheimer approximation, the equilibrium geometries of all isotopologs are identical. Moreover, since the symmetry of the nuclear Coulomb potential is also the same, the C 2v point group may be used for the electronic wave functions in all cases. Tables 4.5!4.7 present calculated (harmonic) and experimental (fundamental) vibrational frequencies for CH 2 N 2 , CD 2 N 2 and CHDN 2 , respectively. The B3LYP and CCSD(T) results for the neutral are in good agreement, which validates the B3LYP results for isotope shifts. The comparison between the theoretical and 129 experimental results, as well as the changes in structures and frequencies induced by ionization and electronic excitation are discussed in Section 4.5. Table 4.4 The calculated equilibrium structures and nuclear repulsion energies for the ground state of the neutral and cation and the 3p Rydberg states of CH 2 N 2 . a 1 1 A 1 b 3 1 A 1 (3p x ) c 2 1 A 2 (3p y ) d 2 1 B 1 (3p z ) d 1 2 B 1 e E nuc (hartree) 61.514227 61.280112 59.502297 61.132745 61.137542 61.118198 r C-N (Å) 1.298 1.292 1.398 1.352 1.343 1.362 r N-N (Å) 1.139 1.132 1.130 1.105 1.107 1.115 r C-H (Å) 1.071 1.077 1.084 1.079 1.084 1.086 HCH (deg) 125.00 124.07 128.64 129.96 124.76 127.80 NNC (deg) 180 180 160.55 180 180 180 NNHC (deg) 0 0 92.7 0 0 0 a All values remain unchanged for CHDN 2 and CD 2 N 2 . b Upper values calculated using B3LYP/6-311G(2df,p), lower values # CCSD(T)/cc-pVTZ, respectively. Experimental geometrical values: r C-N : 1.300 Å; r N-N : 1.139 Å; r C-H : 1.077 Å; HCH: 126.1 o , NNC: 180 o , NNHC: 0 o [28]. c EOM-CCSD/6-31(2+)G(d). d EOM-CCSD/6-311(3+,+)G(2df). e CCSD(T)/cc-pVTZ. 130 Table 4.5 Transitions energies and vibrational frequencies of neutral ground state, 3p Rydberg states, and cation of CH 2 N 2 . a 1 1 A 1 3 1 A 1 (3p x ) 2 1 A 2 (3p y ) 2 1 B 1 (3p z ) 1 2 B 1 Exp. b Calc. c Exp. d Calc. Exp. e Calc. Exp. f Calc. Exp. g Calc. 56898 52227 52628 72620, 72585 h Band origin, IE 77665 51213 51535 71375 3077 … 3059 2980 3015 " 1 #a 1 $! CH 2 sym. stretch 3185, 3230 3202 3182 3104 3164 2102 … 2062 2142 2110, 2180h " 2 #a 1 $! NN stretch 2203, 2173 2093 2225 2208 2199 1414 … 1463 1246 1420 " 3 #a 1 $! CH 2 sym. bend 1448, 1462 1367 1424 1370 1432 1170 … 969 864 985, 970 h " 4 #a 1 $! CN stretch 1214, 1196 768 1007 1007 1001 564 … 347 … 320 " 5 #b 1 $! CNN o.p. bend 586, 578 571 448 460 440 406 … (623) 819 810 " 6 #b 1 $! CH 2 wag 413, 420 526 594 767 712 3185 … 3370 … … " 7 #b 2 $! CH 2 asym. stretch 3305, 3347 3368 3311 3290 3311 1109 … … … … " 8 #b 2 $! CH 2 rock 1125, 1129 1134 1136 1088 1133 421 … 356 … 420 " 9 #b 2 $! CNN i.p. bend 432, 419 431 352 446 377 a All values in cm -1 ; data from this work, unless otherwise stated. Tentatively assigned frequencies are enclosed in parentheses. b Data from Ref. [6,7] (frequencies rounded to integer cm -1 ; accuracy = ± 2 cm -1 ). c Harmonic frequencies, see text. For the 1 1 A 1 state, the left values calculated using CCSD(T)/cc-pVTZ, and the right values by B3LYP/6-311G(2df,p). For the 3 1 A 1 (3p x ) state, symmetry is lowered to C s , b 1 modes are of a’’ symmetry, while all others are a’. d Uncertainty ± 15 cm -1 . e Uncertanity ± 5 cm -1 . f Fundamental frequencies are from Ref. [16] relative to the determined deperturbed value of the 2 1 B 1 (3p z ) band origin. The perturbed value for the band origin (52,690 cm -1 ) was defined as the average value for the 1 ! 0 and 0 ! 1 transitions. g Frequency accuracy defined the spacing between the rotational lines of cation is ± 50 cm -1 . h Data from Ref. [17]. Frequency accuracy is ± 80 cm -1 . 131 Table 4.6 Transitions energies and vibrational frequencies of neutral ground state, 3p Rydberg states, and cation of CD 2 N 2 . a 1 1 A 1 3 1 A 1 (3p x ) 2 1 A 2 (3p y ) 2 1 B 1 (3p z ) 1 2 B 1 Exp. b Calc. c Exp. d Calc. Exp. e Calc. Exp. f Calc. Exp. g Calc. 56871 52214 52664 72620, 72585 h Band origin, IE … … … … 2245 … 2189 2183 … " 1 #a 1 $! CD 2 sym. stretch 2305 2313 2302 2244 2246 2096 … 2051 2081 2180 h " 2 #a 1 $! NN stretch 2198 2060 2213 2197 2145 1213 … 1044 907 … " 3 #a 1 $! CD 2 sym. bend 1267 1034 1076 1054 1108 970 … 919 766 915, 970 h " 4 #a 1 $! CN stretch 984 734 949 932 957 … … (256) … (225, 275) " 5 #b 1 $! CNN o.p. bend 571 564 446 444 425 318 … (590) 606 … " 6 #b 1 $! CD 2 wag 327 417 457 611 592 2414 … 2344 … … " 7 #b 2 $! CD 2 asym. stretch 2470 2527 2484 2471 2440 903 … … … … " 8 #b 2 $! CD 2 rock 919 914 899 891 903 … … (318) … 340 " 9 #b 2 $! CNN i.p. bend 392 376 328 401 357 a All values in cm -1 ; data from this work, unless otherwise stated. Tentatively assigned frequencies are enclosed in parentheses. b Data from Ref. [6,7] (frequencies rounded to integer cm -1 ; accuracy = ± 2 cm -1 ). c Harmonic frequencies, see text. For the 1 1 A 1 state, the values were calculated using B3LYP/6-311G(2df,p). For the 3 1 A 1 (3p x ) state, symmetry is lowered to C s , b 1 modes are of a’’ symmetry, while all others are a’. d Uncertainty ± 15 cm -1 . e Uncertainty ± 5 cm -1 . f Fundamental frequencies are from Ref. [16] relative to the determined deperturbed value of the 2 1 B 1 (3p z ) band origin. The perturbed value for the band origin (52,695 cm -1 ) was defined as the average value for the 1 ! 0 and 0 ! 1 transitions. g Frequency accuracy defined the spacing between the rotational lines of cation is ± 50 cm -1 . h Data from Ref. [17]. Frequency accuracy is ± 80 cm -1 . 132 Table 4.7 Transitions energies and vibrational frequencies of neutral ground state, 3p Rydberg states, and cation of CHDN 2 . a 1 1 A 1 3 1 A 1 (3p x ) 2 1 A 2 (3p y ) 2 1 B 1 (3p z ) 1 2 B 1 Exp. b Calc. c Exp. d Calc. Exp. e Calc. Exp. f Calc. Exp. g Calc. 56936 52221 52648 72620 Band origin, IE … … … … 2331 … 2237 … … " 1 #a 1 $! CD stretch 2382 2410 2385 2343 2335 2097 … 2060 … … " 2 #a 1 $! NN stretch 2201 2076 2221 2205 2149 1310 … 1305 … … " 3 #a 1 $! CHD bend 1351 1273 1309 1256 1297 1157 … 983 … (960) " 4 #a 1 $! CN stretch 1196 751 1017 1012 1063 549 … … … … " 5 #b 1 $! CNN o.p. bend 578 565 448 455 434 368 … (624) … … " 6 #b 1 $! CHD wag 375 482 529 691 673 3133 … 2976 … … " 7 #b 2 $! CH stretch 3262 3307 3253 3209 3203 … … … … … " 8 #b 2 $! CHD rock 942 963 923 908 925 … … 326 … (375) " 9 #b 2 $! CNN i.p. bend 409 395 339 420 371 a All values in cm -1 ; data from this work, unless otherwise stated. Tentatively assigned frequencies are enclosed in parentheses. b Data from Ref. [6,7] (frequency accuracy is ± 2 cm -1 ). c Harmonic frequencies, see text. For the 1 1 A 1 state, the values were calculated using B3LYP/6-311G(2df,p). For the 3 1 A 1 (3p x ) state, symmetry is lowered to C 1 . d Uncertaity ± 15 cm -1 . e Uncertaity ± 5 cm -1 . f Fundamental frequencies from Ref. [16] relative to the determined deperturbed value of the 2 1 B 1 (3p z ) band origin. The perturbed value for the band origin (52,691 cm -1 ) was defined as the average value for the 1 ! 0 and 0 ! 1 transitions. g Frequency accuracy defined the spacing between the rotational lines of cation is ± 50 cm -1 . h Data from Ref. [17] (frequency accuracy is ± 80 cm -1 ). 133 4.5 Discussion 4.5.1 Vibrational assignments for the 2 1 A 2 (3p y ) Rydberg state As stated above, one of the goals of the present work is to determine experimentally the fundamental vibrational modes of the 2 1 A 2 (3p y ) state of diazomethane and its cation in order to compare them with theoretical calculations. In the REMPI spectra, the strongest transitions to the Rydberg 2 1 A 2 (3p y ) state are those of a 1 vibrational symmetry and their assignment are robust. Bands of b 1 and b 2 symmetry are much weaker and often do not show a discernible K-structure. Their assignments, which rely mainly on calculations, are tentative. The proposed assignments for CH 2 N 2 , CD 2 N 2 , and CHDN 2 are shown in Figures 4.2a!4.2c and the fundamental frequencies are listed in Tables 4.5!4.7. In assigning fundamental frequencies in the 2 1 A 2 (3p y ) state we relied on: (i) the measured positions of the REMPI vibronic bands; (ii) the K-structure of the vibronic bands; (iii) the energy positions of the diagonal peaks in the eKE distributions; (iv) changes observed for H/D isotopologs; and (v) results of ab initio calculations. The observed eKE distributions of all the unperturbed peaks appeared isotropic. As discussed previously [5], 2 + 1 REMPI excites mostly vibronic levels in the 2 1 A 2 (3p y ) state, and transitions of a 1 symmetry to " ' 1 !" ' 4 exhibit the highest intensity. For example, in the 2 + 1 REMPI spectrum of CH 2 N 2 (Figure 4.2a), the strong bands at 53,196!53,265, 53,690!53,756, 54,289!54,358, and 134 55,286!55,353 cm -1 are assigned, respectively, as the 4 1 0 , 3 1 0 , 2 1 0 , and 1 1 0 transitions to the 2 1 A 2 (3p y ! !) state. The corresponding frequencies of the totally symmetric (a 1 ) " ' 4 , " ' 3 , " ' 2 % and " ' 1 fundamentals are 969, 1,463, 2,062, and 3,059 cm -1 , respectively, in good agreement with the calculated values (Table 4.5). These assignments are confirmed by the positions of combination bands at 54,627!54,679 and 55,783!55,846 cm -1 , which are assigned as 3 1 0 4 1 0 and 2 1 0 3 1 0 , respectively. CD 2 N 2 and CHDN 2 transitions involving the totally symmetric modes are also intense, and it is easy to identify the 4 1 0 , 3 1 0 , 2 1 0 , and 1 1 0 transitions. For CD 2 N 2 (Figure 4.2b), the frequencies of " ' 4 , " ' 3 , " ' 2 % and " ' 1 are 919, 1,044, 2,051, and 2,189 cm -1 , respectively, and for CHDN 2 (4.2b), they are 983, 1,305, 2,060, and 2,237 cm -1 . A further test of the reliability of the assignments is that bands involving CH or CD motions change their frequency as expected for isotopic substitution. For example, the CH and CD stretch fundamentals in CHDN 2 have a frequency ratio CH:CD ~ 1.4. All the experimentally determined values of the fundamental vibrational frequencies of a 1 symmetry are in good agreement with the calculated harmonic frequencies (Tables 4.5!4.7). While it is fairly easy to assign the totally symmetric fundamentals, this is not the case for the weak bands of b 1 and b 2 symmetry. The transitions to the " 9 (b 2 ) fundamentals are mixed with the 2 1 B 1 origin bands, and this mixing lends them intensity. The unperturbed energies of the upper states were determined by using a 135 two-state approximation, as described before, and the deperturbed values are listed in the tables. In Chapter 3 we assigned the strong band of CH 2 N 2 located at 52,574 cm -1 as the 5 1 0 transition to 2 1 A 2 (3p y ) [5]. The separation between the triad of bands in CD 2 N 2 at 52,461!52,495 cm -1 is ~ 17 cm -1 , which is typical of the transitions of A 1 or A 2 rovibronic levels. However, these bands could not be assigned to any of the a 1 modes or their combinations. They are closest to the calculated frequency of the 5 1 0 transition, resulting in a " ' 5 (b 1 ; CNN out-of-plane bend) frequency of 264 cm -1 . The single peak in the corresponding eKE distribution, whose frequency is 275 ± 10 cm -1 is assigned as " + 5 . A similar eKE distribution was observed in ionization through the 52,461 cm -1 transition for which the internal energy of the cation was calculated to be 225 cm -1 . We therefore tentatively assign the upper state of this REMPI transition as " ' 5 . The " 6 ’ and " 7 ’ normal modes of CH 2 N 2 , CD 2 N 2 , and CHDN 2 , are assigned based primarily on the closeness of observed (weak) REMPI band to the calculated vibrational frequencies and their isotopic variations. Tentative assignments are shown in parentheses in Tables 4.1!4.3 and 4.5!4.7. All the assigned fundamental frequencies are summarized and compared with calculations in Tables 4.5!4.7. 136 4.5.2 Vibrational Assignments for the 1 2 B 1 Ground!State Cation The He(II) photoelectron spectrum of diazomethane was reported before [17], and the adiabatic IE of the ion and the frequencies of several of its vibrational levels were determined (Table 4.5 and 4.6). We obtained these and additional vibrational frequencies from the images by using Eq. 4.1. In assigning the ion’s vibrational modes and frequencies we used mainly those eKE distributions that had a single peak resulting from the diagonal Franck-Condon transition; i.e. the vibrational frequencies obtained for the excited Rydberg state and the cation were rather similar. As discussed before, in the case of the mixed levels described above, we obtained ionic vibrational frequencies from the peak separations in the eKE distributions [5]. We note that only strong transitions, whose signal was high above background, could be used reliably because we detect all photoelectrons produced by ionization disregarding of their origin. The eKE for the origin band places the adiabatic IE of the cation at IE = 72,620 ± 100 cm -1 , in excellent agreement with the published value of 72,585 ± 160 cm -1 [17]. Other isotopologs had the same IE. The uncertainty in our values reflects mainly uncertainty in K of about one unit in the ionization step. As with the values for neutral diazomethane, the experimental and theoretical vibrational frequencies for the ions agree very well. 4.5.3 Structure and Vibrational Motions in Neutral and Ionic Diazomethane The observed changes in structure and frequencies induced by ionization and electronic excitation (Tables 4.5!4.8) can be explained by simple molecular orbital 137 considerations in combination with NBO analysis. As expected from the wave function analysis, 5 the structures and vibrational frequencies of the 2 1 A 2 (3p y ) and 2 1 B 1 (3p z ) Rydberg states are similar to those of the cation and they both retain C 2v structure. The 3 1 A 1 (3p x ) state, however, differs considerably from both the cation and the other two 3p states due to its mixing with the valence 2 1 A 1 (!* ! !) state, and it has C s equilibrium structure [5]. Below we first compare the calculated and experimental values to validate the theoretical results and the assignments, and then proceed to analyze differences in structures and frequencies between the Rydberg states and the cation in order to understand the structural and spectroscopic signatures of Rydberg-valence, Rydberg-Rydberg and Rydberg-ion core interactions. As far as structures are concerned, the calculated bond lengths of the neutral are within 0.002 Å of the experimental values [28], as expected for CCSD(T)/cc-pVTZ level of theory. The B3LYP values are also very close. The maximum discrepancy between the calculated and experimental frequencies for all three isotopologes is about 5 %, which is a typical value for anharmonicities. For the cation, the three lowest frequencies exhibit larger deviations, i.e., 10!12 % for the CH 2 wag and CNN in-plane bend, and 37!80 % for the CNN out-of-plane bend. The out-of-plane vibrations involving the carbon atom hosting the unpaired electron are similar to the out-of-plane mode in substituted methyl radicals, which has been found to be extremely anharmonic [45]. A similar trend is observed for the two Rydberg states as well – most of the calculated frequencies are within 12 % from the experimental 138 ones, except for the same out-of-plane modes, CNN out-of-plane bend and CH 2 wag. Overall, the observed changes in structure and vibrational frequencies are consistent with removing an electron from the bonding ! CN -orbital, which also has an antibonding character with respect to NN. To explain the differences in structures and frequencies between the cation and the Rydberg states, we analyze the interactions of the Rydberg electron with the ion core. For example, the 3p y Rydberg orbital is localized in the plane of the molecule perpendicular to the principal rotation axis (see Figure 4.1 in Ref. 5). Its electron density is greatest on top of the hydrogen atoms and the C and middle N (directly bonded to C) atoms. The A 2 symmetry imposes a nodal plane along this axis. The 2 1 A 2 (3p y ) state differs from the cation mostly in the HCH angle (129.91 o relative to 127.80 o in the cation). The NBO analysis of the electron density of both states reveals that about half of the + 1 charge of the nuclear core is accommodated by the hydrogens. The lobes of the 3p y orbital, located directly on the hydrogen atoms in space, can interact with the positively-charged hydrogen atoms. The larger HCH angle in the 2 1 A 2 (3p y ) state is thus attributed to increased electron density along the CH bonds. The 3p y orbital does not affect the CN bond in a similar way due to symmetry restrictions, so the net effect is to increase repulsion between the hydrogens. A similar argument explains the decrease in the HCH angle in the 2 1 B 1 (3p z ) state relative to the cation (124.74 o compared to 127.80 o ). The occupied 3p z orbital 139 has electron density centered along the CNN axis, with one lobe centered directly in the space between the two hydrogens, while the other is located on the terminal nitrogen, which appropriates almost all of the remaining total nuclear positive charge. Thus, the orientation of the Rydberg orbital allows its electron density to overlap with the centers of positive charge in the nuclear core. Similar examples of Rydberg orbital orientation and the anisotropy of the cation core have been observed in a series of unsaturated hydrocarbon radicals [46]. In the 2 1 B 1 (3p z ) state of diazomethane, the HCH angle decreases to maximize this interaction. The 3p z orbital, which has a node on the central nitrogen, can donate density along both the CN and NN bonds; hence the observed contraction of these bonds with respect to the cation. The variations in the calculated vibrational frequencies for the ground state neutral and cation, and 3p y and 3p z Rydberg states of CH 2 N 2 are depicted in Figure 4.6. Only modes below 3,000 cm $1 are shown – the frequencies of the symmetric and asymmetric CH stretches do not vary significantly with electronic excitation/ionization and are therefore omitted. To explain the observed trends in vibrational frequencies, we divide the vibrational modes into three groups: (i) those that involve displacements mainly along the CNN framework (CN and NN stretches, and CNN bends); (ii) those with displacements primarily in the CH 2 moiety (CH 2 wag, rock, and bend); and (iii) the CH stretching vibrations, which are not affected by the excitation/ionization. For the different electronic states, trends in the first group are due mostly to the effect 140 of lower CN and NN bond orders, while those in the second are due to the interaction between the positively-charged hydrogens and the Rydberg electron density, and the hybridization of the carbon. Within each group, we also observe marked differences between the in-plane and out-of-plane modes. Figure 4.6 Harmonic frequencies of the neutral and cation ground state of CH 2 N 2 compared to those of the 2 1 A 2 (3p y ! !) and 2 1 B 1 (3p z ! !) Rydberg excited states. The four modes that comprise the first group are the CN and NN stretches (both a 1 ), and the b 1 and b 2 CNN bends. As shown in Figure 4.6, the CN stretch is strongly affected by the removal of an electron from the HOMO ! orbital; whether this electron is ionized or placed in a Rydberg orbital has almost no effect on the frequency. Thus, ionization/electronic excitation results in elongation of the CN bond, and a slight contraction of the NN bond. The changes in vibrational frequencies involving CNN motions are consistent with these changes in bond order. 141 Referring to the out-of-plane b 1 CNN bend (" 5 ’) mode, reducing the order of the ! bond leads to a strong decrease in frequency in the Rydberg states as well as the cation; i.e., the in-plane Rydberg orbitals provide no additional contribution relative to the cation. In contrast, the in-plane b 2 CNN bend (" 6 ’) shows a strong frequency change between the two Rydberg states (Figure 4.6). The trend in this mode is complementary to that in the analogous mode in the second group – the b 2 CH 2 bend (rock). For the 3p y state, the CNN bending frequency drops significantly with respect to the neutral (by 67 cm $1 ) whereas the CH 2 bend mode increases slightly (by 7 cm $1 ). For the 3p z state, the CNN bend frequency increases relative to its value in the 2 1 A 2 (3p y ) state by 94 cm $1 , to above the frequency of the neutral, whereas the CH 2 bend decreases by 48 cm $1 to below that of the neutral. For the cation, the CNN mode drops by 69 cm $1 relative to the 2 1 B 1 (3p z ) state, falling again below the neutral value, while the CH 2 bend increases by 45 cm $1 and is, within error, the same as in the neutral. The largest difference within both modes occurs between the 2 1 A 2 (3p y ) and 2 1 B 1 (3p z ) states. For the CNN bend, the displacement moves the CNN framework off the nodal plane and into the electron density of the 3p y orbital in the yz-plane. However, this displacement moves the atoms out of the density of the 3p z orbital, which is hindered by the donation of electron density into the CN and NN bonds. Consequently, the frequency of this vibration is significantly higher in the 2 1 B 1 (3p z ) state than in the 2 1 A 2 (3p y ) state. 142 Finally and quite surprisingly, the frequency of the CH 2 out-of-plane wag (" 6 ’) increases significantly upon excitation/ionization. The reason for this is the competition between the two resonance forms in the ground-state wave function (Figure 4.1) and the change in hybridization of the carbon induced by ionization/electronic excitation. The NBO analysis confirms the competition between the two resonance structures in the ground-state wave function, which gives rise to sp 2 and sp 3 hybridized carbon for the left and right structures of Figure 4.1, respectively. Removing an electron from either of these structures results in sp 2 hybridized carbon and, therefore, a reduction in the sp 3 contribution, as confirmed also by NBO analysis. The increased sp 2 character leads to a stiffer out-of-plane vibration, which is exactly what is obtained in the calculations. 4.6 Summary The joint experimental and theoretical investigation discusses the structure and normal mode frequencies of the ground and excited Rydberg states of diazomethane and its isotopologs and of the corresponding cations. The experimental measurements exploit REMPI spectroscopy and velocity map imaging of photoelectrons from excited vibronic levels of the 2 1 A 2 (3p y ) state to obtain vibronic assignments in the 2 1 A 2 (3p y ) and 2 1 B 1 (3p z ) Rydberg states, and vibrational states of the cation. The accompanying high-level ab initio calculations determine structures and vibrational states in the ground states of the neutral and cations as well as the three Rydberg 3p states. The good agreement between the 143 electronic structure results and the current experimental results on the 2 1 A 2 (3p y ) state and the cation, as well as previous studies on other states, allows a full analysis of Rydberg-ion core interactions and trends in vibrational frequencies. Although the 2 1 A 2 (3p y ) and 2 1 B 1 (3p z ) Rydberg states have planar C 2v symmetry like the ion, they exhibit differences in geometry due to specific interactions of the electron in the 3p y or 3p z orbital with the nuclei charge distributions of the ion core. Trends in vibrational frequencies in the ground states of the neutral and ion and the 2 1 A 2 (3p y ) and 2 1 B 1 (3p z ) states are consistent with removing an electron from the bonding ! CN -orbital, which nevertheless has an antibonding character with respect to NN. In explaining the observed trends, the vibrational modes are divided into two groups, which involve displacements mainly (i) along the CNN framework, and (ii) in the CH 2 moiety. Trends in the first group are due mostly to effects of the lower CN and NN bond orders, whereas those in the second group are due to the interaction between the positively-charged hydrogens and the Rydberg electron density, and the hybridization of the carbon. Within each group, marked differences in behavior between the in-plane and out-of-plane modes are observed. The largest changes in frequencies upon ionization are observed in the CN stretch, CH 2 wag and the two CNN bending modes. Differences in vibrational frequencies between the 2 1 A 2 (3p y ) and 2 1 B 1 (3p z ) Rydberg states reflect state-specific interactions of the charge density of the electron in the Rydberg 3p orbital with the nuclei charge density in the ion core. 144 4.7 Chapter 4 References 1. W. Kirmse, Carbene Chemistry. 2nd ed. 1971: Academic: New York. 622 pp. 2. R.B. Bohn, S.A. Sandford, L.J. Allamandola, and D.P. Cruikshank, Icarus, 111, 151, (1994). 3. F. Raulin, M. Khlifi, M. Dang-Nhu, and D. Gautier, Adv. Space Res., 12, 11/181, (1992). 4. M.H. Moore and R.L. Hudson, Icarus, 161, 486, (2003). 5. I. Fedorov, L. Koziol, G.S. Li, J.A. Parr, A.I. Krylov, and H. Reisler, J. Phys. Chem. A, 111, 4557, (2007). 6. C. Moore and G. Pimentel, J. Chem. Phys., 40, 329, (1964). 7. C.B. Moore, J. Chem. Phys., 39, 1884, (1963). 8. R.H. Pierson, A.N. Fletcher, and E.S.C. Gantz, Anal. Chem., 28 1218, (1956). 9. M. Khlifi, P. Paillous, P. Bruston, F. Raulin, and J.C. Guillemin, Icarus, 124, 318, (1996). 10. B.L. Crawford, Jr., W.H. Fletcher, and D.A. Ramsay, J. Chem. Phys., 19 406, (1951). 11. H.W.T. I.M. Mills, Transactions of the Faraday Society, 50, 1270, (1954). 12. W.H. Fletcher and T.P. Garrett, J. Chem. Phys., 25, 50, (1956). 13. J. Vogt and M. Winnewisser, Z. Naturforsch., A: Phys. Sci., 38, 1138, (1983). 14. J. Vogt, M. Winnewisser, K. Yamada, and G. Winnewisser, Chem. Phys., 83, 309, (1984). 15. L. Nemes, J. Vogt, and M. Winnewisser, J. Mol. Struct., 218, 219, (1990). 16. A.J. Merer, Can. J. Phys., 42, 1242, (1964). 17. J. Bastide and J.P. Maier, Chem. Phys., 12, 177, (1976). 145 18. A. Sanov, T. Droz-Georget, M. Zyrianov, and H. Reisler, J. Chem. Phys., 106, 7013, (1997). 19. V. Dribinski, A.B. Potter, I. Fedorov, and H. Reisler, Chem. Phys. Lett., 385, 233, (2004). 20. S.M. Hecht and J.W. Kozarich, Tetrahedron Lett., 1501, (1972). 21. R. Brinton and D. Volman, J. Chem. Phys., 19, 1394, (1951). 22. R.J. Wolf and W.L. Hase, J. Phys. Chem., 82, 1850, (1978). 23. A.T.J.B. Eppink and D.H. Parker, Rev. Sci. Instrum., 68, 3477, (1997). 24. Y. Tanaka, M. Kawasaki, Y. Matsumi, H. Fujiwara, T. Ishiwata, L.J. Rogers, R.N. Dixon, and M.N.R. Ashfold, J. Chem. Phys., 109, 1315, (1998). 25. B.Y. Chang, R.C. Hoetzlein, J.A. Mueller, J.D. Geiser, and P.L. Houston, Rev. Sci. Instrum., 69, 1665, (1998). 26. V. Dribinski, A. Ossadtchi, V.A. Mandelshtam, and H. Reisler, Rev. Sci. Instrum., 73, 2634, (2002). 27. A.P.T. Cox, L. F.; Sheridan, J., Nature, 181, 1000, (1958). 28. J. Sheridan, Proc. Intern. Meeting Mol. Spectry., 4th, 139, (1962). 29. E.C.E. Lim, Excited States. Vol. 3. 1977: (Academic, New York, N. Y.). 351. 30. K. Raghavachari, G.W. Trucks, J.A. Pople, and M. Headgordon, Chem. Phys. Lett., 157, 479, (1989). 31. J.D. Watts, J. Gauss, and R.J. Bartlett, J. Chem. Phys., 98, 8718, (1993). 32. T.H. Dunning Jr., J. Chem. Phys., 90, 1007, (1989). 33. A.D. Becke, J. Chem. Phys., 98, 5648, (1993). 34. R. Krishnan, J.S. Binkley, R. Seeger, and J.A. Pople, J. Chem. Phys., 72, 650, (1980). 35. A.D. Mclean and G.S. Chandler, J. Chem. Phys., 72, 5639, (1980). 146 36. H. Koch, H.J.A. Jensen, P. Jorgensen, and T. Helgaker, J. Chem. Phys., 93, 3345, (1990). 37. J.F. Stanton and R.J. Bartlett, J. Chem. Phys., 98, 7029, (1993). 38. S.V. Levchenko and A.I. Krylov, J. Chem. Phys., 120, 175, (2004). 39. D. Sinha, S. Mukhopadhyay, and D. Mukherjee, Chem. Phys. Lett., 129, 369, (1986). 40. S. Pal, M. Rittby, R.J. Bartlett, D. Sinha, and D. Mukherjee, Chem. Phys. Lett., 137, 273, (1987). 41. J.F. Stanton and J. Gauss, J. Chem. Phys., 101, 8938, (1994). 42. J. Kong, C.A. White, A.I. Krylov, D. Sherrill, R.D. Adamson, T.R. Furlani, M.S. Lee, A.M. Lee, S.R. Gwaltney, T.R. Adams, C. Ochsenfeld, A.T.B. Gilbert, G.S. Kedziora, V.A. Rassolov, D.R. Maurice, N. Nair, Y.H. Shao, N.A. Besley, P.E. Maslen, J.P. Dombroski, H. Daschel, W.M. Zhang, P.P. Korambath, J. Baker, E.F.C. Byrd, T. Van Voorhis, M. Oumi, S. Hirata, C.P. Hsu, N. Ishikawa, J. Florian, A. Warshel, B.G. Johnson, P.M.W. Gill, M. Head-Gordon, and J.A. Pople, J. Comput. Chem., 21, 1532, (2000). 43. J.F. Stanton, J. Gauss, J.D. Watts, W.J. Lauderdale, and R.J. Bartlett, ACES II, 1993. The package also contains modified versions of the MOLECULE Gaussian integral program of J. Almlöf and P. R. Taylor, the ABACUS integral derivative program written by T. U. Helgaker, H. J. Aa. Jensen, P. Jørgensen, and T. P. Taylor, and the PROPS property evaluation integral code of P. R. Taylor. 44. E.D.B. Glendening, J. K.; Reed, A. E.; Carpenter, J. and J.A.M. E.; Bohmann, C. M.; Weinhold, F., NBO. 2001: Theoretical Chemistry Institute, University of Wisconsin: Madison, WI,. 45. S.V. Levchenko, A.V. Demyanenko, V.L. Dribinski, A.B. Potter, H. Reisler, and A.I. Krylov, J. Chem. Phys., 118, 9233, (2003). 46. L. Koziol, S.V. Levchenko, and A.I. Krylov, J. Phys. Chem. A, 110, 2746, (2006). 148 Chapter 5. Multiphoton Ionization and Dissociation of Diazirine: A Theoretical and Experimental Study 5.1 Introduction Diazirine (c-CH 2 N 2 ) belongs to the family of isoelectronic molecules known as "16-electron molecules", which have attracted considerable attention for decades because of the inherent complexity of their photodissociation dynamics and certain similarities in their properties [1,2]. In addition, HNCO, H 2 CNN and H 2 CCO, all members of this family, are known to have structural isomers [1-16]. Of these molecules, the least studied is the H 2 CN 2 group, mainly due to the instability of its prototype member, diazomethane, and its cyclic counterpart diazirine. Recently it has become possible to produce stable molecular beams of these species, and studies of their photophysics and photochemistry in highly excited states have begun, aided by high-level electronic structure calculations. In previous papers we described the two-photon excitation of diazomethane to 3p Rydberg states, and discussed its electronic structure and couplings among its excited states [17,18]. In the present paper, we report first results on the two-photon dissociation of diazirine, specifically the formation of CH products, as well as electronic structure calculations on its excited and ionized states. Diazirines contain a three-membered ring composed of one carbon atom and two double-bonded nitrogen atoms. Although diazomethane had been known since the 1920's, diazirines were first synthesized only in the 1960's [19-22]. However, 149 due to their structural uniqueness and their roles as precursors of carbenes, much has been learned since then about their spectroscopy, photochemistry and thermal decomposition [22]. For the prototype diazirine (c- CH 2 N 2 ), the simplest member of the group, the lowest lying UV absorption bands were assigned to the structured 1 1 B 2 ! 1 1 A 1 (!* ! n) system with a band origin at 31,187 cm -1 (320.65 nm) [21,23]. Its VUV absorption spectrum shows an intense, structureless band centered at 145–185 nm as well as diffuse structures at ~ 120–143 nm [24]. The vertical (adiabatic) ionization energy (IE) of diazirine was determined experimentally at 10.75 eV (10.3 eV), and the ion’s low-lying excited states are at 13.25 eV (12.8 eV) and 14.15 eV (14.15 eV) [25]. Paulett and Ettinger reported the 298 K heat of formation of diazirine "H f o 298 = 79.3 kcal/mol [26,27], whereas Laufer and Okabe obtained 60.6 < "H f o 298 < 66 kcal/mol [24]. Although significant experimental work has been carried out since the 1960’s, the level of theoretical studies during this early period was too low to provide reliable information on excited states. With the availability of high-level electronic structure computer codes, a renewed effort to elucidate the electronic structure of diazirines/diazo compounds and their decomposition mechanisms was initiated in the 1990’s [10-13,28-30]. The molecular structure of c-CH 2 N 2 , its related cation, and its ionization energy were the subject of several investigations [31,32]. Electronically excited states and energy differences among isomers were also calculated [13,28]. 150 Whereas several aspects of diazirine's photochemistry on the lowest excited state 1 1 B 2 have been discussed [10,13,33], little is known from experiment and theory about photodissociation on higher electronic states, and in particular about multiphoton dissociation pathways. By combining high-level electronic structure calculations and photofragment laser spectroscopic experiments in molecular beams, we were able to investigate the photodissociation of c-CH 2 N 2 following two-photon absorption. In the only previous molecular beam study, diazirine was excited to the 1 1 B 2 ! 1 1 A 1 origin band and fluorescence, tentatively assigned to emission from high vibronic levels of the excited 1 1 B 2 state of the singlet methylene product, was observed [34]. We have adapted the traditional preparation method of diazirine [20,35] to molecular beam studies, and used resonance enhanced multiphoton ionization (REMPI) complemented by dc slice velocity map imaging (VMI) [36] to detect CH (X 2 ") photodissociation fragments. Specifically, CH products from two-photon dissociation were detected by 2 + 1 REMPI at the dissociation wavelength (one- color experiment) using the D 2 " (v’ = 2) ! ! X 2 " (v’’ = 0) intermediate transition (where the double arrow indicates two-photon transition). The dc sliced images of CH + ions were anisotropic and typical of those of a perpendicular transition and fast dissociation. Several channels are discussed as possible sources of CH(X 2 ") fragments and the results suggest that the predominant pathway is two-photon absorption by the parent molecule followed by isomerization to isodiazirine and subsequent dissociation to CH + HN 2 . 151 5.2 Experimental Details The experimental arrangement has been described before in Chapter 2 and in Ref. [17,18,37,38], and therefore only procedures that have changed are discussed in detail. Diazirine (c-CH 2 N 2 ) was synthesized in the same glass vacuum line that was used in the synthesis of diazomethane [20,39]. A mixture containing ~ 1.5 % c-CH 2 N 2 in helium at a backing pressure of ~ 2 atm was introduced into the source chamber of the differentially pumped vacuum apparatus through a pulsed piezoelectric nozzle (10 Hz, 0.5 mm diameter). The rotational temperature of the skimmed molecular beam was estimated at ~ 10 K on the basis of expansion conditions and previous results. The diazirine/He sample survived for several days until depleted by use. REMPI spectra were recorded by integrating separately ion peaks of appropriate masses as a function of laser excitation wavelength. The UV laser beam was the linearly polarized radiation from a Continuum Nd:YAG - pumped dye laser (Surelite III/ ND6000) using rhodamine 610 dye (Exciton). The dye laser output was frequency-doubled (Inrad Autotracker III), producing UV pulses with maximum energy of 3.0 mJ at a repetition rate of 10 Hz. The laser radiation was focused with a 40-cm focal length (f.l.) lens at the center of the ionization region. Wavelength calibration was accomplished by using atomic carbon lines observed in 2 + 1 REMPI [40]. Ion images of CH + were obtained by using the dc slice variant of VMI [36]. CH + ions produced by 2 + 1 REMPI via the D 2 " (v’ = 2) ! ! X 2 " (v’’ = 0) 152 transition were extracted and accelerated by the ion optics towards a 42-mm MCP detector coupled to a P47 phosphor screen (Burle Electro-optics, Inc.). Using a home-built pulser, a high voltage (~ 2 kV) pulse of ~ 5 ns FWHM was generated and applied to the MCP detector to select only the central slice of the CH + ion cloud. Images on the detector were recorded with a CCD camera (Imager 3, 12 bit, LaVision, 1280x1024 pixel array), averaged and processed using the DAVIS package (LaVision) that included event counting. CH + ion velocity distributions were obtained directly from the recorded sliced images as described elsewhere [36]. CH (X 2 " (v"=0, N”)) photofragment translational energy distributions were obtained by integrating the experimental images over all angles and converting velocity to energy distributions. For energy calibration, photoelectrons obtained by NO ionization via the A 2 # + state were used [37]. Diazirine was produced in a glass reactor that was charged with 15.0 g 3CH 2 (NH 2 ) 2 ·4H 2 SO 4 , in powder form, and a NaCl/ice mixture maintained at # 15 to # 10 o C. 3CH 2 (NH 2 ) 2 ·4H 2 SO 4 was prepared according to the method of Ohme and Schmitz [20,35]. The reactor was submerged in an external salt/ice bath at the same temperature. A pressure-equalized addition funnel isolated from the reactor by greaseless Teflon ® /glass high vacuum valves was charged with 150 mL NaOCl solution (10#14% by weight), and 30 mL NaOH (~ 25 M). The reactor and the addition funnel were evacuated and the solutions degassed. The NaOCl/NaOH solution was then added drop-wise over the course of 10 minutes to the reactor. The resulting c-CH 2 N 2 was purified by passing the gas though two traps held at # 78 o C 153 with a dry-ice/ethanol slush, and collected in a 12-L glass flask housed in a steel mesh box and protected from exposure to light. The pressure of c-CH 2 N 2 was kept at less than 30 Torr. The purity of the diazirine sample was assessed from its IR and UV spectra, which were in good agreement with published spectra [21,39,41] and contained negligible amounts of impurities. We wish to emphasize that c-CH 2 N 2 is a toxic and hazardous gas, which can decompose explosively and spontaneously, and thus appropriate safety precautions must be taken. c-CH 2 N 2 should be handled only at low pressures and on a small scale. At no time should the gas be allowed to condense into the liquid phase. Throughout our experiments a pressure of 30 Torr was never exceeded. Safety equipment (safety shields, safety glasses, face shields, leather gloves and protective clothing, such as leather suits, Kevlar® sleeves and earplugs) must be used at all times. Care must be taken to avoid known triggers of c-CH 2 N 2 decomposition such as intense light and abrupt changes in temperature, pressure, and phase. 5.3 Computational Studies of the Electronically Excited and Ionized States of Diazirine The equilibrium geometry and vibrational frequencies of neutral diazirine ground state were calculated by CCSD(T) [42-44] using the cc-pVTZ basis [45] and by B3LYP [46] using the 6-311G(2df,p) basis. The equilibrium geometry of the cation was calculated using B3LYP/6-311G(2df,p). Vertical excitation energies were calculated using EOM-CCSD[47-51]/6-311(3+,+)G* at the B3LYP optimized 154 geometry. This basis set was derived from 6-311G* by adding three diffuse sp functions to heavy atoms and one diffuse s function to hydrogen. The assignment of valence and Rydberg character to the excited states was based on: (i) the symmetry of the transition, (ii) leading EOM-CCSD amplitudes and character of corresponding HF orbitals, and (iii) the second moments <X 2 >, <Y 2 >, and <Z 2 > of the EOM-CCSD electron density. The character of the HF orbitals was determined using the Molden interface [52]. All EOM-CCSD excited states were dominated by single excitations. The excited valence states 1 1 B 2 (!* ! n) and 1 1 A 2 (!* ! $ NN ) were optimized by EOM-CCSD/6-311G** under C 2v constraint using analytic gradients [53]. For the lowest state, 1 1 B 2 (!* ! n), the optimized structure is a true minimum as confirmed by vibrational frequency calculations. For the 1 1 A 2 (!* ! $ NN ) state, we were not able to perform frequency calculations in this basis because of the limitations of the finite differences code and the large density of states in this energy range. To validate the structure of this state, we optimized the geometry starting from the C 1 distorted structure and computed frequencies using the smaller 6-31G* basis. This calculation produced a similar C 2v structure and no negative frequencies. All optimizations, frequencies, and excited state calculations were performed using the Q-Chem [54] and ACES II [55] electronic structure programs. 155 The electronic configuration of the ground state (1 1 A 1 ) of diazirine is: [core+low-lying] 16 (6a 1 ) 2 (2b 2 ) 2 (3b 1 ) 2 (1a 2 ) 0 (4b 1 ) 0 = [core+low-lying] 16 ($ NN ) 2 (% NN ) 2 (n) 2 (%* NN ) 0 ($* CN ) 0 , where “core + low-lying” refers to the 1s core orbitals and combinations of $ CN and % NN orbitals that are not involved in excitations. Relevant molecular orbitals are shown in Figure 5.1. The CNN ring lies in the yz plane with the z axis coinciding with the C 2 symmetry axis. The equilibrium geometries of diazirine in its ground neutral and cation states are shown in Figure 5.2 and the geometries of the excited states are given in Table 5.1. The neutral ground state geometry agrees well with experimental values [56]. Table 5.1 Calculated equilibrium structures for the ground, 1 1 B 2 and 1 1 A 2 valence states of the neutral and the ground state of the cation. 1 1 A 1 a 1 1 B 2 (%* ! n) c 1 1 A 2 (%* ! $ NN ) b 1 2 B 1 c E nuc (hartree) 64.158275 64.295975 62.313876 61.730487 62.797366 r C-N (Å) 1.477 1.479 1.534 1.453 1.617 r N-N (Å) 1.229 1.216 1.260 1.422 1.144 r C-H (Å) 1.075 1.081 1.083 1.093 1.082 HCH (deg) 119.6 119.5 120.3 117.1 129.5 NNC (deg) 65.4 65.7 65.8 60.7 69.3 NCN (deg) 49.2 48.5 48.5 58.6 41.4 a Upper values calculated using CCSD(T)/cc-pVTZ, lower values – B3LYP/6-311G(2df,p), respectively. Experimental geometrical values: r C-N : 1.482 ± 0.003 Å; r N-N : 1.228 ± 0.003 Å; r C-H : 1.09 ± 0.02 Å; HCH: 117 ± 2 o [56]. b Values calculated using EOM-CCSD/6-311G**. c B3LYP/6-311G(2df,p). 156 The first ionization removes an electron from the highest-occupied molecular orbital (HOMO), which is the 3b 1 orbital denoted as n in Figure 5.1. This orbital features sigma-type bonding between an atomic p orbital on carbon and out-of- phase s orbitals on the nitrogens (see Figure 5.1). The geometry change upon ionization from this orbital is consistent with a weakening bonding interaction in the CN bonds; consequently, the cation resembles a weakly bound N 2 ...CH 2 + as described before [31]. Indeed, at the CCSD/6-311G(2df,p) level of theory, the cation lies only 0.73 eV (16.9 kcal/mol) below the N 2 + CH 2 + asymptote. The vertical IE calculated at the B3LYP neutral geometry with the EOM-IP-CCSD/6- 311G(2df,p) method is 10.71 eV. This agrees well with the experimental value [25]; triples corrections are not required for this system. The Rydberg states are expected to have similar structures to the cation, and thus should be weakly-bound. The vertical excitation energies, their leading configuration state functions, and one-photon oscillator strengths are summarized in Table 5.2, along with oscillator strengths for transitions from the 1 1 B 2 state to higher electronic states. One-photon vertical excitations from the ground state show that the strongest excitation is to the 3p x Rydberg state. 157 Figure 5.1 Molecular orbitals relevant to ground and excited electronic states of c-CH 2 N 2 . The three-membered ring lies in the yz plane, with the z-axis coinciding with the C 2 symmetry axis. Transition dipole moments were also calculated between excited states in a single EOM-CCSD calculation. This allowed us to obtain oscillator strengths specifically from the 1 1 B 2 (!* ! n) state to the other EOM states. The lowest excited state is a valence 1 1 B 2 (!* ! n) excitation at 4.27 eV (293 nm). Well separated from this state, there is a cluster of states between 7.2 and 158 8.0 eV (174–156 nm); these are the n = 3 Rydberg states and the valence 1 1 A 2 (!* ! $ NN ) state. Absorption to these states agree reasonably well with the measured absorption spectrum of diazirine at 165–145 nm [24]. Table 5.2 Vertical excitation energies (!E vert , eV), oscillator strengths (f L )/ (oscillator strengths from 1 1 B 2 state), dipole strengths (µ 2 tr , atomic units), and changes in second dipole moment of charge distributions (!<R 2 >, squared atomic units) for the excited states of c-CH 2 N 2 at EOM-CCSD/6-311(3+,+)G*. a State !E vert f L " 2 tr "<X 2 > "<Y 2 > "<Z 2 > 1 1 B 2 (%* ! n) 4.27 0.004/(-) 0.037 0 1 -2 1 1 B 1 (3s ! n) 7.32 0.0006/(0) 0.003 9 17 12 1 1 A 2 (%* ! $ NN ) 7.61 0/(0.059) 0 -1 1 -1 2 1 A 1 (3p x ! n) 7.86 0.099/(0.013) 0.516 25 8 7 2 1 B 1 (3p z ! n) 7.88 0.0210/(0) 0.021 6 11 36 2 1 A 2 (3p y ! n) 7.97 0/(0.015) 0 8 33 9 a At the B3LYP/6-311G(2df,p) optimized geometry; E CCSD = # 148.417293 hartree. The oscillator strengths in excitation from the 1 B 2 state show that the dominant transition is to the second valence state, 1 1 A 2 , rather than to the Rydberg states. The oscillator strength for the 1 1 A 2 ! 1 1 B 2 transition is about an order of magnitude greater than that for 1 1 B 2 ! 1 1 A 1 excitation, implying high probability for a two- photon transition via the intermediate 1 1 B 2 state. Excitation to 1 A 2 is forbidden in one-photon transition from the ground state, but is allowed in two-photon excitation via the 1 B 2 state. Excitation to the 1 1 A 2 (!* ! $ NN ) state removes an electron from a bonding orbital along NN (Figure 5.1), and places it into the !* orbital, which is 159 antibonding along NN. Both orbitals have electron density localized along the NN bond, and thus the excitation doubly weakens this bond. Figure 5.2 Left panel: Ground state equilibrium structures (Å and deg) of diazirine for: the neutral (1 1 A 1 ) at CCSD(T)/cc-pVTZ (normal print) and B3LYP/6-311G(2df,p) (italics) and for the cation, 1 2 B 1 (% ! n), at B3LYP/6-311G(2df,p) (underlined). The corresponding nuclear repulsion energies are: 64.158275, 64.295975, and 62.797366 hartrees, respectively. Right panel: Excited state equilibrium structures for the 1 1 B 2 (%* ! n) and 1 1 A 2 (%* ! $ NN ) excited states at CCSD(T)/ 6-311G** shown in normal and italics, respectively. The corresponding nuclear repulsion energies are: 62.313876 and 61.730487 hartrees. Experimental parameters of the neutral ground state are: r C-N : 1.482 ± 0.003 Å; r N-N : 1.228 ± 0.003 Å; r C-H : 1.09 ± 0.02 Å; HCH: 117 ± 2 o [56]. Comparing geometries of the 1 A 2 and ground state reveals elongation of the NN bond by 0.22 Å, which is very close to a single N-N bond length (1.45 Å), and opening of the NCN angle by approximately 9°. The CN bond length decreases only slightly, by 0.024 Å. Even though the geometry optimization produces a stable structure, the considerable reduction of the formal bonding character of this state suggests a small barrier to dissociation and nearly-dissociative character. Moreover, the elongated NN bond may facilitate isomerization to isodiazirine in which the NN bond is a single bond (see below). The limitations of the employed single-reference 160 methods do not allow us to characterize the potential energy surface (PES) too far from the Franck-Condon region. This state was also found to be bound by the CASSCF calculations of Arenas et al. [13], who performed surface-hopping calculations to model photoinduced dynamics on this state, referred to by them as S 2 . They reported fast (~ 40 fs) radiationless relaxation to the 1 1 B 2 state (S 1 in their notation) via a conical intersection followed by dissociation to methylene and N 2 . However, the low level of theory employed in their work (i.e., small basis, considering only valence states, no dynamical correlation included) can adversely affect their results. For example, Ref. [13] reported unbound equilibrium structure for the 1 1 B 2 (!* ! n) state, in disagreement with our calculations and the experimental Franck-Condon progressions [21,23]. 5.4 Experimental Results Our first experiments were aimed at observing the REMPI spectrum of the parent diazirine by 1 + 2 REMPI via the bound 1 B 2 state. Figure 3a presents an action spectrum recorded at m/e = 14 (CH 2 + ) with 1.0 mJ of 305#327 nm laser radiation focused by a 40-cm f.l. lens. The spectrum consists mainly of a structureless continuum with absorption starting at ~ 324.0 nm (30,860 cm -1 ). Superimposed on this continuum are broad peaks at 321.4 nm (31,114 cm -1 ), 318.9 nm (31,358 cm -1 ), 317.8 nm (31,466 cm -1 ), 315.1 nm (31,736 cm -1 ), and 307.3 nm (32,541 cm -1 ). 161 In our detection system, strong ion signals induce electrical ringing for a few hundred ns after the large ion peak reaches the detector. This results in some leakage from the intense signals of m/e = 12 and 13 into the m/e = 14 region (which is 6 times less intense), and this leakage can be seen in Figure 5.3a. The two sharp peaks at 313.48 and 320.33 nm were assigned to the intense 2p3p 1 S 0 ! ! 2p 2 1 D 2 and 2p3p 1 D 2 ! ! 2p 2 1 D 2 transitions of atomic carbon (m/e = 12 ) [40]. The group of sharp peaks at 311–312.5 nm belongs to the D 2 " (v’ = 2) ! ! X 2 " (v’’ = 0) transition of CH detected by 2 + 1 REMPI (see below). The CH 2 + action spectra were seen only at fairly high laser fluences (> 0.5 mJ) and they exhibited qualitatively a greater than linear dependence on fluence. However, because of the small CH 2 + signal, no accurate fluence dependence could be measured. No signal at m/e = 42 from the parent diazirine ion has been detected suggesting that the generated c-CH 2 N 2 + cations are unstable. Because CH 2 + ions are observed only at 305–324 nm laser wavelengths, which is the region of one-photon 1 1 B 2 ! 1 1 A 1 absorption, we suggest that the CH 2 + ion fragment is formed by first excitation to 1 1 B 2 followed by excitation to higher excited states leading to dissociative ionization or autoionization of a superexcited state. 162 Figure 5.3 Survey REMPI spectra of diazirine obtained by monitoring (a) m/e = 14 (CH 2 + ) and (b) m/e = 13 (CH + ) at wavelengths 305–327 nm with 0.8 and 0.3 mJ energies (40-cm f.l. lens) for (a) and (b), respectively. The spectrum in (b) is about x10 higher in intensity than the one in (a). Simultaneous recordings of m/e = 12 (C + ), 13 (CH + ), and 14 (CH 2 + ) signals at 0.8 mJ laser fluence (40-cm f.l. lens) show that the m/e = 14 (CH 2 + ) signal is about 6 times smaller than the m/e = 13 (CH + ) signal. The m/e = 12 and 13 spectral 163 shapes are similar (see below). The c-CH 2 N 2 ion mass (m/e = 42) was not observed under any conditions. Figure 5.4 2 + 1 REMPI spectrum of the CH(X) fragment (m/e = 13). Assignments of rotational CH(X, N”) levels for the D 2 " (v’ = 2) ! ! X 2 " (v’’ = 0) transition are marked on top of the corresponding peaks. Lines marked “a” belong to a different band system; the line marked “b” points to the 2p3p 1 S 0 ! ! 2p 2 1 D 2 atomic carbon transition. Arrows mark peak wavelengths at which ion images of CH + have been taken. At lower laser fluences, the predominant peak is CH + at m/e = 13 and its action spectrum, shown in Figure 5.3b, is quite different from that of CH 2 + . Figure 5.4 shows part of the structured REMPI spectrum of CH + observed at 310.5-314.2 nm with 0.3 mJ laser fluence. It is noteworthy that similar spectra were observed in the multiphoton dissociation of ketene (CH 2 CO) [57,58], bromoform [59,60], and t-butylnitrite [58], and were assigned to the CH 2 + 1 REMPI spectrum via the 164 CH D 2 " (v’ = 2) ! ! X 2 " (v’’ = 0) transition [58]. Rotational assignments for this transition are marked in Figure 5.4. We use N” for the rotational states because CH(X) reaches Hund’s case (b) at low rotational levels. The dependence of the CH + ion signal on laser pulse energy, determined at several excitation wavelengths, is slightly higher than quadratic (Figure 5). We note that a similar but weaker REMPI spectrum was obtained by monitoring C + at m/e = 12; C + is known to be a product of predissociation of CH + [61,62]. Figure 5.5 Fluence dependence of the CH + (m/e = 13) ion signal resulting from 2 + 1 REMPI through the CH D 2 " (v’ = 2, N” & 9) ! X 2 " (v’’ = 0, N” = 9) transition. The dissociation wavelength was 312.10 nm and the radiation was focused by a 40-cm f.l. lens. The signal depends on the n = 2.15 power of laser energy. 165 Dc sliced images of CH + ions resulting from 2 + 1 REMPI of CH(X 2 ") fragments generated in diazirine dissociation at 312.10 nm (64,082 cm -1 ; Q 2 (N” = 9)) and 310.91 nm (64,328 cm -1 ; R branch near bandhead (N” = 6 and 9)) and the corresponding center-of-mass (c.m.) translational energy distributions (assuming dissociation to CH + HN 2 ) are displayed in Figures 5.6 and 5.7, respectively, along with their recoil anisotropy parameters & 2 and & 4 . The derived translational energy distribution is broad and structureless, peaking at about 8,000 cm -1 and extending to 20,500 ± 1,000 cm -1 . The angular distributions were fit by the formula, I(') = C[1+ & 2 P 2 (cos ') + & 4 P 4 (cos ')], where P n (x) is the n th Legendre Polynomial, & n is the recoil anisotropy parameter, and C is a normalization constant [63,64]. Both images have anisotropic angular distributions, which are well described for E T = 5,000#15,000 cm -1 by & 2 = # 0.61 ± 0.1 and & 4 = # 0.10 at ( ph = 312.10 nm, and & 2 = # 0.72 ± 0.1 and & 4 ' 0 at ( ph = 310.91 nm. The translational energy distribution does not change when the laser fluence is varied. 166 Figure 5.6 In the top panel the image obtained in dissociation at 312.10 nm by monitoring CH(X, v" = 0, N” = 9) is shown. In the bottom panel, the c.m. translational energy distribution, P(E T ), of the CH(X) fragments is shown (right axis) as well as the recoil anisotropy parameters & i (E T ) (left axis). 167 Figure 5.7 In the top panel the image obtained in dissociation at 310.91 nm by monitoring CH(X, v" = 0, N” = 6 and 9), which is in the R-bandhead region, is shown. In the bottom panel, the c.m. translational energy distribution, (P(E T ), of the CH(X) fragments is shown (right axis) as well as the recoil anisotropy parameters & i (E T ) (left axis). 168 5.5 Discussion 5.5.1 Excited States and Photoionization of Diazirine The electronic structure calculations of diazirine's excited states show that in addition to the low-lying 1 1 B 2 ( !* ! n) state common to all diazirines [21- 23,28,65], there is a cluster of states in diazirine at 7.6–8.3 eV, which consists of the n = 3 Rydberg states (s and p) and the 1 1 A 2 (!* ! $ NN ) valence state. The weakly-bound Rydberg states appear to be pure and do not interact appreciably with nearby valence states. The most efficient one-photon absorption is expected to be to the 2 1 A 1 (3p x ) Rydberg state. The situation is quite different in two-photon excitation via the 1 1 B 2 intermediate state. The oscillator strength calculations from the 1 1 B 2 state suggest that the most efficient excitation is to the 1 1 A 2 (!* ! $ NN ) valence state, and this oscillator strength is greater by a factor of > 4 than to the 2 1 A 1 (3p x ) and 2 1 A 2 (3p y ) Rydberg states. Thus, we believe that the main excitation process responsible for ionization and photodissociation of diazirine in our experiments is a sequential two-photon process 1 1 A 2 ! 1 1 B 2 ! 1 1 A 1 , which accesses the bound but dissociative 1 1 A 2 valence state (see Section 5.3). Accepting this as the main excitation process explains the high propensity for dissociative photoionization observed in the REMPI spectrum of diazirine. Both the 1 1 B 2 and 1 1 A 2 states have geometries that differ significantly from the ground-state ion and this would lead to internally excited and predissociative diazirine cations. Three photons are required for ionization of diazirine, giving rise to ions that can have > 1 eV of internal energy above the 10.3 eV adiabatic ionization energy [25]. 169 The dissociation energy of the ion is calculated at 0.73 eV above its ground state, and its geometry resembles that of the loosely bound CH 2 + ...N 2 complex (see Table 5.1 and Figure 5.2); thus, the observation that CH 2 + fragment ions are the main photoionization products is not surprising. 5.5.2 Detection of Ionization Products As discussed above, following 324!305 nm laser irradiation multiphoton processes lead to ionization and photodissociation. The wavelength range at which CH 2 + ions are detected coincides with the 1 1 B 2 ! 1 1 A 1 structured absorption system of diazirine, but the vibronic features observed in the CH 2 + action spectrum are much broader than those in the corresponding one-photon absorption spectrum [21,22]. This difference is rationalized by realizing that the CH 2 + photoionization spectrum is a result of multiple photon excitation via the dissociative 1 1 A 2 (!* ! $ NN ) intermediate state. Thus, there is competition between fast dissociation and ionization and the effective lifetime for ionization via 1 + 1 + 1 REMPI is determined by dissociation in the second intermediate state. This explains both the significant lifetime broadening and the low yield of CH 2 + ions, which are detected only at higher laser fluences. We also note that no REMPI signal from neutral N 2 and CH 2 photofragments is detected. The CH(X) 2 + 1 REMPI spectrum observed via the D 2 " (v’ = 2) ! ! X 2 " (v’’ = 0) transition is well known, but its analysis is complicated by severe predissociation in the upper D state [57,59]. As discussed above, similar REMPI 170 spectra of CH(X) have been observed in multiphoton dissociation of several precursors [57-61,66,67]. In studies of photoion and photoelectron spectroscopy, it has been established that the predissociation rate in the D 2 " i (v’ = 2) state increases greatly for rotational levels N’ (11 as a result of curve crossing with repulsive states [57-59], and rotational transitions from CH(X) that terminate in N’ ( 11 in the D state cannot be observed. As a result of this predissociation, it is impossible to determine the effective rotational line strengths and infer populations. However, it is clear from our spectra that the decrease in the observed line intensities starts at N” < 11, before fast predissociation sets in, indicating that the maximum in the rotational distribution is lower than the predissociation limit. It is also evident that the lowest rotational levels have small populations, indicating that the rotational distribution is shifted to higher rotational states and is probably nonstatistical. Particularly striking is the similarity between the CH(X) REMPI spectra obtained in this work and the corresponding REMPI and LIF spectra of CH(X) obtained in two-photon dissociation of the isoelectronic ketene at comparable levels of parent excitation [57-59,68]. In the next section we expand on this similarity and discuss possible dissociation mechanisms. 5.5.3 Pathways Leading to CH(X) Fragments The most intriguing experimental finding of this work is the intense REMPI spectrum assigned to CH(X) fragments. Its anisotropic angular distribution indicates that fast dissociation via a perpendicular transition is responsible for its 171 formation. From the observation that all the rotational branches in the CH(X) spectrum in the region 310–316 nm can be detected, we infer that the absorption by the CH(X) precursor is broad and not state specific. Although we cannot offer a definitive mechanism for the production of CH(X), we describe below several possible pathways and discuss our strong preference for one of them. From the high CH(X) translational energies determined from the photofragment images, we conclude that this fragment must be generated by absorption of at least two photons. One-photon production of CH(X) from ground state diazirine requires wavelengths < 207 nm (48,300 cm -1 ; 138 kcal/mol), whereas CH(X) production is observed with 314 nm excitation (31,800 cm -1 ) and the fragments are born with substantial translational energies. Before discussing possible reaction pathways, an assessment of the dissociation energy of diazirine to produce CH fragments is needed. The largest uncertainty derives from the value of the heat of formation of diazirine. The two experimental values (obtained over 35 years ago) differ greatly from each other, 60.6–66 [24] and 79.3 kcal/mol [26,27], and each determination is associated with experimental difficulties. The NIST Chemistry Webbook [69] gives both values and the issue has remained the subject of debate [22,24,70-73]. In the past 15 years, with theoretical methods achieving chemical accuracy, the heats of formation of diazirine and its structural isomer diazomethane (whose heat of formation is just as controversial) were calculated using high-level electronic structure methods, and a re-evaluation of the accepted values was called for [72,73]. In Table 5.3 we summarize the 172 calculated heats of formation, )H f ° 0 , which are much more convergent than the experimental ones. Several authors have calculated heats of formation of diazomethane and diazirine through various pathways. Gordon and Kass [11,12] employed the atomization reaction and two isodesmic reactions using the G2 model chemistry, giving average values of 64.3 and 74.5 kcal/mol for )H f ° 0 of diazomethane and diazirine at 0 K, respectively. Table 5.3. Calculated values of !H f o of diazomethane and diazirine (kcal/mol). Diazomethane Diazirine 0 K 298 K 0 K 298 K References 65.42, 65.68 63.18, 64.15 77.9, 74.10 76.11, 74.10 Ref.[74] 66.7 65.3 - - Ref.[73] 68.0 - 77.7 - Ref.[72] 64.3 63.1 74.5 73.0 Ref.[11,12] Catoire [74] reported CBS-Q and G2 heats of formation for several species produced by the decomposition reactions of monomethylhydrazine. Ab initio atomization reactions and atomic heats of formation at 0 K (gas phase) were calculated, giving values of 65.4 and 65.7 kcal/mol for CBS-Q and G2 methods for diazomethane, respectively, and 78.0 and 76.0 for diazirine. Walch [72] reported heats of formation using CASSCF methods with double- zeta Dunning basis sets for geometries, and internally contracted configuration interaction (ICCI) for energetics with double-, triple-, and quadruple-zeta Dunning bases. He recommended the value of )H f ° 0 = 77.7 kcal/mole for diazirine. 173 Dixon et al. [73] reported 65.3 and 66.7 kcal/mol for the heat of formation of diazomethane at 298 and 0 K, respectively, computed using the CCSD(T) method and CBS extrapolation, which is the most accurate theoretical estimate. All calculations included zero point corrections via harmonic frequency calculations. For diazomethane, both G2 calculations [11,74] are within 1 kcal/mol of the result of Dixon et al. [73], suggesting similar accuracy for the respective diazirine values. Thus, for diazomethane the preferred theoretical value is )H f ° 0 = 67 ± 3 kcal/mole, much closer to the experimental value of 67 kcal/mol recommended by Setser and Rabinovitch [75], than to values given in the NIST Chemistry Webbook [69]. Moreover, in all the calculations diazirine is to found to lie 10 ± 1 kcal/mole above the ground state of diazomethane, allowing us to adapt the thermochemistry of diazomethane reactions to the case of diazirine (see below) [11-13,72,74]. We conclude that theoretical heats of formation for diazirine are 75 ± 2 and 77 ± 2 kcal/mol at 298 and 0 K, respectively. These values are much closer to those recommended by Paulett and Ettinger,[26] than to those of Laufer and Okabe [24]. In agreement with the calculations, in what follows we have adopted the value of 77 ± 3 kcal/mol as the heat of formation of diazirine at 0 K. We use the calculated value )H f ° 0 = 60.8 kcal/mole for the heat of formation of HN 2 [73] and the accepted values of )H f ° 0 for H, CH, and CH 2 (1 1 A 1 ) of 51.63 [76], 141.61 ± 0.14 [77], and 102.37 ± 0.38 kcal/mol [77], respectively. 174 The most direct dissociation process leading to CH(X) formation is: c-CH 2 N 2 (1 1 A 1 ) + 2h* ) CH(X 2 ") + HN 2 (1 2 A’’); "H r o 0 = 125 ± 4 kcal/mol I The HN 2 product is metastable and is calculated to lie 9 kcal/mole above the thermochemical threshold for H + N 2 ;[78] thus the overall dissociation process is: c-CH 2 N 2 (1 1 A 1 ) + 2h* ) CH(X 2 ") + H( 2 S) + N 2 (X 1 + g + ); I(a). which likely evolves in sequential steps. In order to assess the feasibility of reaction (I), the maximum allowed translational energy release in the CH(X) product needs to be determined. The best estimate is obtained from Figure 6, since this image was obtained for state-selected CH(X, N” = 9). For this image 2h* = 64,082 cm -1 , and for N” = 9 of CH(X) E rot = 1,300 cm -1 (B(CH) = 14.457 cm -1 ) [79]. Thus, the energy required for reaction (I) terminating in CH(X, N” = 9) is 43,900 ± 1,400 cm -1 (125 ± 4 kcal/mol) plus 1,300 cm -1 . Subtracting this energy from the photon energy, we obtain that the maximum allowed c.m. translational energy is E T = 18,900 ± 1,400 cm -1 , in good agreement with the observed value of 20,500 ± 1,000 cm -1 . A different route to assess the maximum allowed c.m. translational energy associated with CH(X, N” = 9) in reaction (I) is to start with the calculated value for the dissociation of diazomethane [80]: CH 2 N 2 (1 1 A 1 ) ) CH 2 (1 1 A 1 ) + N 2 (X 1 + g + ); "H r o = 32.6 kcal/mol, add to it the energy required to dissociate CH 2 (1 1 A 1 ) to CH(X, N” = 9) + H and subtract 9 kcal/mole for the formation of the HN 2 product. Taking into account the 10 kcal/mol difference between the heats of formation of diazirine and 175 diazomethane, we obtain that reaction (I) should require 126 kcal/mole, which corresponds to a c.m. translational energy of < 19,900 cm -1 , again in good agreement with the experimental result. We conclude, therefore, that our results are well explained by reaction (I). Moreover, on the basis of our results and the recent theoretical calculations, we argue that the heat of formation of diazirine at 0 K should be revised upward to 77 ± 3 kcal/mol and that the diazomethane value should be lower by 10 kcal/mol. We note that using the calculated value for the heat of formation of the diazirinyl radical of 117 ± 1 kcal/mol [74,81,82] and the calculated D 0 = 93 kcal/mol for c-CH 2 N 2 (1 1 A 1 ) ) c-CHN 2 (1 2 A 2 ) + H( 2 S),[31] we obtain that the heat of formation of diazirine is 76 kcal/mol. Reaction I is the only one that can generate CH(X) via absorption of two photons, and thus should be the favored dissociation pathway. Inspection of the recoil anisotropy parameters shows that in the region where the intensity of CH(X) is substantial, the & 2 parameter is fairly constant and is typical of one-photon excitation via a perpendicular transition. This suggests that the anisotropy is determined largely in the second step, i.e. excitation from the long-lived intermediate 1 1 B 2 state to the dissociative 1 1 A 2 state. The fluence dependence of CH(X) production in these one-color experiments is not very revealing. At the fluence levels where reasonable signals are obtained (> 0.15 mJ, 40 cm f.l. lens) the intensity dependence is slightly higher than 176 quadratic; however, this dependence reflects mainly the two-photon nature of the CH D ! X excitation rather than the 1 + 1 photon excitation process in diazirine. The most likely mechanism of reaction (I) involves initial isomerization to isodiazirine, with transfer of a hydrogen atom from carbon to nitrogen: I(b) Such isomerization is in general inefficient; however, it is known that in the family of the so-called 16-electron molecules, atom shifts are unusually facile [1,3,4,6,16,83]. In particular, in the photodissociation of isotopically labelled ketene (H 2 12 C 13 CO), scrambling between the two carbon isotopes in the CO product has been ascribed to the formation of the cyclic intermediate oxirane, c-HCOCH ( ) [16,83], accompanied by a hydrogen shift. Moreover, in the two-photon excitation of ketene at comparable wavelengths (279.3 and 308 nm) [68], Ball et al. observed efficient production of CH(X) fragments and obtained their rotational distributions by LIF. The rotational distributions are bell-shaped, with widths of 10 # 15 N and maxima that shift to higher N values at higher excitation energies. The authors favor a mechanism in which isomerization of ketene to the formylmethylene isomer (HCCHO) via the cyclic oxirane precedes dissociation to CH(X). 177 In contrast, in the one-photon dissociation of ketene at 157.6 nm, the predominant product channel is CH 2 + CO, while the H + HCCO and CH + HCO channels account for less than 5% of the products [84]. One-photon dissociation, in this case, is assumed to proceed via excitation to the 3d Rydberg state, and the mechanism is different than in 1 + 1 photon excitation at about the same level of energy. It appears, therefore, that one- and two-photon dissociation processes in ketene at comparable excitation energies proceed via different mechanisms. In conclusion, a pathway that is initiated by two-photon absorption in diazirine and evolves via isomerization and H-shift to dissociation (either simultaneously or sequentially) agrees well with our experimental observations and can also explain the high internal energies deposited in the co-fragment. We point out that in the 1 1 A 2 state of diazirine, the calculated N-N bond length is 1.422 Å, close to the single bond length of isodiazirine, calculated at 1.56–1.63 Å [85-87]. This should facilitate isomerization either directly on the excited state, or via a conical intersection with the ground state. Below we discuss briefly two other possible routes to CH(X) and assess their feasibility. The first is a sequential pathway initiated by breaking one C-H bond by two-photon absorption followed by one-photon dissociation of the diazirinyl radical, c-HCNN: c-CH 2 N 2 (1 1 A 1 ) +2h* ) c-HCNN(1 2 A’’) + H( 2 S); "H r o = 93 kcal/mol II(a) c-HCNN(1 2 A’’) + h* ) CH(X 2 ") + N 2 (X 1 + g + ); "H r o = 26 kcal/mol II(b) 178 Assuming that the maximum CH(X) translational energy allowed by the thermochemisty in each step is achieved, translational energies higher than 21,000 cm -1 may be observed. In step II(a), both the cyclic and the open-chain HCNN may be generated. To date, there is no experimental information on the diazirinyl radical, but calculations show that it is a stable species, close in its heat of formation to the open-chain HCNN, HNCN, and CNNH structural isomers [81,82,88-91]. The HCNN system has recently attracted attention because of its relevance to the CH + N 2 reaction mechanism, and the facile isomerization among different structural isomers has been discussed [81,82,88-91]. Experimentally, while no studies of diazirinyl photodissociation are available, Neumark and coworkers have studied the photodissociation of jet-cooled open-chain HCNN radicals via a parallel transition at wavelengths that coincide with the photon energies used in the current experiments [92]. HCNN exhibits broad absorption features at 26,000–40,000 cm -1 , with an onset of dissociation at about 25,400 cm -1 , resulting in a broad translational energy distribution in the CH(X) radical. While this mechanism can explain the observed high translational energies, it does not explain the observed anisotropy parameter. Also, the requirement for three-photon absorption plus two photon for detection makes this pathway less likely than mechanism (I), and the importance of reaction II(a) needs to be established independently. Another reaction sequence that may, in principle, lead to formation of CH(X) with high translational energies is initial one-photon dissociation on the 1 1 B 2 state 179 of diazirine to generate CH 2 fragments in one of the three lowest singlet states, followed by two-photon absorption in CH 2 to generate CH(X). We feel that such a sequence is unlikely, because (i) the one-photon 1 B 2 ! 1 A 1 absorption in diazirine is highly structured; and (ii) the subsequent photon absorption by the small radical CH 2 should also be structured and not necessarily coincide with the CH D ! X absorption at all wavelengths. Thus, the broad and even nature of the absorption spectrum in our studies leads us to believe that the first step involves 1 + 1 excitation to the 1 A 2 state followed by dissociation to CH(X). 5.6 Summary Multiphoton ionization and dissociation processes in diazirine have been studied both experimentally (via 304–325 nm two-photon absorption) and theoretically by using the EOM-CCSD and B3LYP methods. The electronic structure calculations identified two valence states and four Rydberg states in the region 4.0–8.5 eV. In one-photon excitation, the calculated strongest absorption is to the 2 1 A 1 (3p x ) Rydberg state, whereas in two-photon absorption at comparable energies via the low-lying 1 1 B 2 valence state, the strongest absorption is predicted to reach the dissociative valence 1 1 A 2 state. The diazirine ion should be rather unstable, with a binding energy of only 0.73 eV and a geometry that resembles a weakly bound CH 2 + ...N 2 complex. On the basis of the electronic structure calculations, we conclude that two- photon absorption in diazirine is very efficient. Weak absorption to the 1 1 B 2 state is 180 immediately followed by more efficient absorption of another photon to reach the 1 1 A 2 state from which competition between ionization and fast dissociation takes place. Absorption of a third photon leads to dissociative photoionization with the formation of CH 2 + fragment ions. No parent diazirine ions are detected. Two-photon dissociation on the 1 1 A 2 state leads to efficient detection of CH(X) fragments. We propose that the most likely route to CH(X) formation is isomerization to isodiazirine followed by dissociation. This mechanism agrees well with the measured maximum in the c.m. translational energy, and is similar to the proposed mechanism for formation of CH(X) in two-photon dissociation of the isoelectronic ketene [68]. Comparison between one- and two-photon dissociation at 150–170 nm would be enlightening. In closing, and in agreement with previous theoretical papers [11,72-75], we call for a revision of the heats of formation of diazirine and diazomethane to 77 ± 3 and 67 ± 3 kcal/mol, respectively. 181 5.7 Chapter 5 References 1. J.W. Rabalais, J.M. Mcdonald, V. Scherr, and S.P. McGlynn, Chem. Rev., 71, 73, (1971). 2. W.M. Jackson and H. Okabe, Adv. Photochem., 13, 1, (1986). 3. A.D. Mclean, G.H. Loew, and D.S. Berkowitz, J. Mol. Spectrosc., 64, 184, (1977). 4. N. Pinnavaia, M.J. Bramley, M.D. Su, W.H. Green, and N.C. Handy, Mol. Phys., 78, 319, (1993). 5. C.B. Moore, Faraday Discuss., 102, 1, (1995). 6. W.A. Shapley and G.B. Bacskay, J. Phys. Chem. A, 103, 6624, (1999). 7. M.S. Schuurman, S.R. Muir, W.D. Allen, and H.F. Schaefer, J. Chem. Phys., 120, 11586, (2004). 8. M. Mladenovic and M. Lewerenz, Chem. Phys., 343, 129, (2008). 9. N. Goldberg, A. Fiedler, and H. Schwarz, Helv. Chim. Acta, 77, 2354 (1994). 10. N. Yamamoto, F. Bernardi, A. Bottoni, M. Olivucci, M.A. Robb, and S. Wilsey, J. Am. Chem. Soc., 116, 2064, (1994). 11. M.S. Gordon and S.R. Kass, J. Phys. Chem., 99, 6548, (1995). 12. M.S. Gordon and S.R. Kass, J. Phys. Chem. A, 101, 7922, (1997). 13. J.F. Arenas, I. Lopez-Tocon, J.C. Otero, and J. Soto, J. Am. Chem. Soc., 124, 1728, (2002). 14. E.R. Lovejoy, S.K. Kim, R.A. Alvarez, and C.B. Moore, J. Chem. Phys., 95, 4081, (1991). 15. E.R. Lovejoy and C.B. Moore, J. Chem. Phys., 98, 7846, (1993). 16. J.D. Gezelter and W.H. Miller, J. Chem. Phys., 103, 7868, (1995). 17. I. Fedorov, L. Koziol, G.S. Li, J.A. Parr, A.I. Krylov, and H. Reisler, J. Phys. Chem. A, 111, 4557, (2007). 182 18. I. Fedorov, L. Koziol, G.S. Li, H. Reisler, and A.I. Krylov, J. Phys. Chem. A, 111, 13347, (2007). 19. S.R. Paulsen, Angew. Chem. Int. Ed., 72, 781, (1960). 20. E. Schmitz and R. Ohme, Chem. Ber. Recl., 94, 2166, (1961). 21. W.H. Graham, J. Am. Chem. Soc., 84, 1063, (1962). 22. M.T.H. Liu, ed. Chemistry of Diazirines. 1987, CRC Press: Boca Raton, FL. 23. L.C. Robertson and J.A. Merritt, J. Mol. Spectrosc., 19, 372, (1966). 24. A.H. Laufer and H. Okabe, J. Phys. Chem., 76, 3504, (1972). 25. M.B. Robin, K.B. Wiberg, G.B. Ellison, C.R. Brundle, and N.A. Kuebler, J. Chem. Phys., 57, 1758, (1972). 26. G.S. Paulett and R. Ettinger, J. Chem. Phys., 39, 825, (1963). 27. G.S. Paulett and R. Ettinger, J. Chem. Phys., 39, 3534, (1963). 28. M.S. Han, H.-G. Cho, and B.-S. Cheong, Bull. Korean Chem. Soc., 20, 1281, (1999). 29. D.W. Ball, THEOCHEM, 722, 213, (2005). 30. A.I. Boldyrev, P.V. Schleyer, D. Higgins, C. Thomson, and S.S. Kramarenko, J. Comput. Chem., 13, 1066, (1992). 31. C. Puzzarini and A. Gambi, Chem. Phys., 306, 131, (2004). 32. C. Puzzarini, A. Gambi, and G. Cazzoli, J. Mol. Struct., 695, 203, (2004). 33. T.S. Kim, S.K. Kim, Y.S. Choi, and I. Kwak, J. Chem. Phys., 107, 8719, (1997). 34. S.M. Lim, T.S. Kim, G.I. Lim, S.K. Kim, and Y.S. Choi, Chemical Physics Letters, 288, 828, (1998). 35. J. Vogt, M. Winnewisser, and J.J. Christiansen, J. Mol. Spectrosc., 103, 95, (1984). 36. D. Townsend, M.P. Minitti, and A.G. Suits, Rev. Sci. Instrum., 74, 2530, (2003). 183 37. V. Dribinski, A.B. Potter, I. Fedorov, and H. Reisler, Chem. Phys. Lett., 385, 233, (2004). 38. A. Sanov, T. Droz-Georget, M. Zyrianov, and H. Reisler, J. Chem. Phys., 106, 7013, (1997). 39. A. Gambi, M. Winnewisser, and J.J. Christiansen, J. Mol. Spectrosc., 98, 413, (1983). 40. C.E. Moore, Atomic Energy Levels. (Nsrds-Nbs 3 (Sect. 4)), A7 Iv-1-B7 Vii-3. Vol. 1. 1971: Official Standards Reference Data Series, National Bureau of Standards 35; U.S. Goverment: Washington, D.C. 41. R. Ettinger, J. Chem. Phys., 40, 1693, (1964). 42. G.D. Purvis and R.J. Bartlett, J. Chem. Phys., 76, 1910, (1982). 43. K. Raghavachari, G.W. Trucks, J.A. Pople, and M. Headgordon, Chem. Phys. Lett., 157, 479, (1989). 44. J.D. Watts, J. Gauss, and R.J. Bartlett, J. Chem. Phys., 98, 8718, (1993). 45. T.H. Dunning Jr., J. Chem. Phys., 90, 1007, (1989). 46. A.D. Becke, J. Chem. Phys., 98, 5648, (1993). 47. H. Sekino and R.J. Bartlett, Int. J. Quantum Chem. Symp., 18, 255, (1984). 48. H. Koch, H.J.A. Jensen, P. Jorgensen, and T. Helgaker, J. Chem. Phys., 93, 3345, (1990). 49. J.F. Stanton and R.J. Bartlett, J. Chem. Phys., 98, 7029, (1993). 50. S.V. Levchenko and A.I. Krylov, J. Chem. Phys., 120, 175, (2004). 51. A.I. Krylov, Annu. Rev. Phys. Chem., 59, 433, (2008). 52. G. Schaftenaar and J.H. Noordik, J. Comput. Aided Mater. Des., 14, 123, (2000). 53. S.V. Levchenko, T. Wang, and A.I. Krylov, J. Chem. Phys., 122, (2005). 184 54. Y. Shao, L.F. Molnar, Y. Jung, J. Kussmann, C. Ochsenfeld, S.T. Brown, A.T.B. Gilbert, L.V. Slipchenko, S.V. Levchenko, D.P. O'Neill, R.A. DiStasio, R.C. Lochan, T. Wang, G.J.O. Beran, N.A. Besley, J.M. Herbert, C.Y. Lin, T. Van Voorhis, S.H. Chien, A. Sodt, R.P. Steele, V.A. Rassolov, P.E. Maslen, P.P. Korambath, R.D. Adamson, B. Austin, J. Baker, E.F.C. Byrd, H. Dachsel, R.J. Doerksen, A. Dreuw, B.D. Dunietz, A.D. Dutoi, T.R. Furlani, S.R. Gwaltney, A. Heyden, S. Hirata, C.P. Hsu, G. Kedziora, R.Z. Khalliulin, P. Klunzinger, A.M. Lee, M.S. Lee, W. Liang, I. Lotan, N. Nair, B. Peters, E.I. Proynov, P.A. Pieniazek, Y.M. Rhee, J. Ritchie, E. Rosta, C.D. Sherrill, A.C. Simmonett, J.E. Subotnik, H.L. Woodcock, W. Zhang, A.T. Bell, A.K. Chakraborty, D.M. Chipman, F.J. Keil, A. Warshel, W.J. Hehre, H.F. Schaefer, J. Kong, A.I. Krylov, P.M.W. Gill, and M. Head- Gordon, Phys. Chem. Chem. Phys., 8, 3172, (2006). 55. J.F. Stanton, J. Gauss, J.D. Watts, W.J. Lauderdale, and R.J. Bartlett, Aces Ii, 1993. The Package Also Contains Modified Versions of the Molecule Gaussian Integral Program of J. Almlöf and P. R. Taylor, the Abacus Integral Derivative Program Written by T. U. Helgaker, H. J. Aa. Jensen, P. Jørgensen, and T. P. Taylor, and the Props Property Evaluation Integral Code of P. R. Taylor. 56. L. Pierce and V. Dobyns, Journal of the American Chemical Society, 84, 2651, (1962). 57. P. Chen, J.B. Pallix, W.A. Chupka, and S.D. Colson, J. Chem. Phys., 86, 516, (1987). 58. P. Chen, W.A. Chupka, and S.D. Colson, Chem. Phys. Lett., 121, 405, (1985). 59. Y.M. Wang, L.P. Li, and W.A. Chupka, Chem. Phys. Lett., 192, 348, (1992). 60. J.W. Hudgens, C.S. Dulcey, G.R. Long, and D.J. Bogan, J. Chem. Phys., 87, 4546, (1987). 61. Y.M. Wang, L.P. Li, and W.A. Chupka, Chem. Phys. Lett., 185, 478, (1991). 62. U. Hechtfischer, C.J. Williams, M. Lange, J. Linkemann, D. Schwalm, R. Wester, A. Wolf, and D. Zajfman, J. Chem. Phys., 117, 8754, (2002). 63. R.N. Zare, Mol. Photochem., 4, 1, (1972). 185 64. R.N. Zare, ed. Angular Momentum : Understanding Spatial Aspects in Chemistry and Physics. (the George Fisher Baker Non-Resident Lectureship in Chemistry at Cornell University). 1988, New York : Wiley. 349. 65. L.C. Robertson and J.A. Merritt, J. Mol. Spectrosc., 24, 44, (1967). 66. J.B. Pallix, P. Chen, W.A. Chupka, and S.D. Colson, J. Chem. Phys., 84, 5208, (1986). 67. P.J.H. Tjossem and K.C. Smyth, Chem. Phys. Lett., 144, 51, (1988). 68. S.M. Ball, G. Hancock, and M.R. Heal, J. Chem. Soc., Faraday Trans., 90, 523, (1994). 69. H.Y. Afeefy, J.F. Liebman, and S.E. Stein, Neutral Thermochemical Data, in Nist Chemistry Webbook, Nist Standard Reference Database Number 69,. 2003, edited by Linstrom, P. J. and Mallard, W. G. (NIST, Gaithersburg, MD, 2003) <http://webbook.nist.gov/>. 70. J.A. Bell, J. Chem. Phys., 41, 2556, (1964). 71. G.S. Paulett and R. Ettinger, J. Chem. Phys., 41, 2557, (1964). 72. S.P. Walch, J. Chem. Phys., 103, 4930, (1995). 73. D.A. Dixon, W.A. de Jong, K.A. Peterson, and T.B. McMahon, J. Phys. Chem. A, 109, 4073, (2005). 74. L. Catoire and M.T. Swihart, Journal of Propulsion and Power, 18, 1242, (2002). 75. D.W. Setser and B.S. Rabinovitch, Can. J. Chem., 40, 1425, (1962). 76. S.G. Lias, J.E. Bartmess, J.F. Liebman, J.L. Holmes, R.D. Levin, and W.G. Mallard, J. Phys. Chem. Ref. Data, 17, 1, (1988). 77. B. Ruscic, J.E. Boggs, A. Burcat, A.G. Csaszar, J. Demaison, R. Janoschek, J.M.L. Martin, M.L. Morton, M.J. Rossi, J.F. Stanton, P.G. Szalay, P.R. Westmoreland, F. Zabel, and T. Berces, J. Phys. Chem. Ref. Data, 34, 573, (2005). 78. H. Koizumi, G.C. Schatz, and S.P. Walch, J. Chem. Phys., 95, 4130, (1991). 186 79. K.P. Huber and G. Herzberg, Molecular Spectra and Molecular Structure. Constants of Diatomic Molecules. Vol. IV. 1979: Van Nostrand Reinhold Company: New York. 80. A. Papakondylis and A. Mavridis, J. Phys. Chem. A, 103, 1255, (1999). 81. M.R. Berman, T. Tsuchiya, A. Gregusova, S.A. Perera, and R.J. Bartlett, J. Phys. Chem. A, 111, 6894, (2007). 82. L.B. Harding, S.J. Klippenstein, and J.A. Miller, J. Phys. Chem. A, 112, 522, (2008). 83. W.J. Bouma, R.H. Nobes, L. Radom, and C.E. Woodward, J. Org. Chem., 47, 1869, (1982). 84. I.C. Lu, S.H. Lee, Y.T. Lee, and X.M. Yang, J. Chem. Phys., 124, (2006). 85. J.B. Moffat, J. Mol. Struct., 52, 275, (1979). 86. M. Alcami, J.L.G. Depaz, and M. Yanez, J. Comput. Chem., 10, 468, (1989). 87. C. Guimon, S. Khayar, F. Gracian, M. Begtrup, and G. Pfister-Guillouzo, Chem. Phys., 138, 157, (1989). 88. M.R. Manaa and D.R. Yarkony, J. Chem. Phys., 95, 1808, (1991). 89. M.R. Manaa and D.R. Yarkony, Chem. Phys. Lett., 188, 352, (1992). 90. L.V. Moskaleva and M.C. Lin, Proc. Combust. Inst., 28, 2393, (2000). 91. L.V. Moskaleva, W.S. Xia, and M.C. Lin, Chem. Phys. Lett., 331, 269, (2000). 92. A.E. Faulhaber, J.R. Gascooke, A.A. Hoops, and D.M. Neumark, J. Chem. Phys., 124, (2006). 187 Chapter 6. Imaging Studies of Photochemistry and Spectroscopy of the Triplet Methylene 6.1 Introduction Diradicals are open-shell molecules with two unpaired electrons that are important in synthetic chemistry and photochemistry, and play a role in atmospheric, stratospheric and interstellar chemistry [1-9]. The existence of two nearly-degenerate orbitals results in a large number of excited valence and Rydberg electronic states with different multiplicities. The spectra and reactivity of these species have intrigued chemists for decades [1-7]. In contrast to extensive work in organic chemistry, the high reactivity of diradicals and difficulties in their preparation under clean conditions have hindered experimental physical chemistry investigations, e.g. spectroscopy and photochemistry [10-17]. This is particularly striking for small diradicals, for which high-level theoretical investigations of excited states have been publeshed [18-30]. In fact, this is one of the few areas of photochemistry in which theory is ahead of experiment. Consider for example the simplest carbene ! the methylene radical (CH 2 ), which is one of the most theoretically studied carbenes. However, experimental studies are far behind calculations. The main reason is the methylene radical’s high reactivity, which leads to difficulties in its preparation under clean conditions. Although much experimental work has been done on its ground 1 3 B 1 state and low- 188 lying singlet states (detected by LIF), only a few spectroscopic and dynamical investigations have examined its dissociative triplet surfaces [31-40]. The ground state of the methylene radical, 1 3 B 1 , has the following electronic configuration [18-28]: [core](2a 1 ) 2 (1b 2 ) 2 (3a 1 ) 1 (1b 1 ) 1 . 6.1 The geometry of the ground state and its orbital occupation is shown in Figure 6.1. It consists of two C-H bonds and singly-occupied sp-hybridized " and p-like # orbitals. Promotion of an electron from either the 1b 1 or the 3a 1 orbitals localized on carbon to ns, np, nd orbitals gives rise to two series of Rydberg states converging to the 1 2 A 1 (ground) and 1 2 B 1 states of CH 2 + , respectively [41]. The experimentally-measured first ionization potential is 10.3962 ± 0.0036 eV [42,43], whereas the calculated vertical value for the second one is 11.29 eV [26,41]. The strongest transitions (f ~ 0.05) are to the 3s states: 1 3 A 1 (3s) ! 1 3 B 1 (4a 1 ! 1b 1 ), and 2 3 B 1 (3s) ! 1 3 B 1 (4a 1 ! 3a 1 ), whose calculated vertical energies are around 6.25 and 7.42 eV, respectively [20,26]. According to the computational work, there is a valence state 1 3 A 2 arises from 3a 1 ! 1b 2 excitation. This state is interesting because unlike the ground state geometry, it has extraordinarily bent geometry ( HCH = 41.61 o , r CH = 1.293 Å, and r HH = 0.9 Å) resembling C + H 2 . If excited directly, it will dissociate preferably to C( 3 P) + H 2 without a significant barrier [23]. However, theory also predicts that in the Franck-Condon region ( HCH = 134 o ), there is a conical intersection between 2 3 B 1 and 1 3 A 2 [20,23,26,29,30], which are both A! in C s symmetry. Since 189 the 2 3 B 1 state has a linear geometry, passage through the conical intersection is predicted to access the near-linear part of the 1 3 A 2 surface, propelling the system towards CH(a 4 $ - ) + H channel rather than C( 3 P) + H 2 . Figure 6.1 Geometry and carbon atom orbital occupation in ground state CH 2 (1 3 B 1 ) [18,19]. Although the lowest-lying singlet states have been studied in detail [35,44-52], there are only a few experimental studies on the lowest-lying triplet excited states. In 1959 the first absorption bands of CH 2 were obtained in flash photolysis of diazomethane near 141.5 nm by Herzberg and Shoosmith [31] and in 1971 Herzberg and Johns assigned these bands to the 3 A 2 ! 3 B 1 transition with a Rydberg d character [36]. They also determined the first ionization potential of CH 2 to be 83,851 cm -1 [34]. Rydberg states have also been detected via mass-resolved resonance-enhanced multiphoton ionization (REMPI) spectroscopy by Irikura and Hudgens in 1992 [37]. They concluded that the spectra arose from three-photon resonances with 3d and 4d Rydberg states between 78,950 and 68,200 cm -1 . The methylene radicals were produced by the reaction of fluorine atoms and methyl radicals. The same group also detected, by 2 + 1 REMPI, the 3p and 4p Rydberg 190 states at 311.80 nm (64,126 cm -1 ) and 269.7 nm (74,254 cm -1 ), respectively [38]. They concluded that 2 + 1 REMPI through the 3p state is an excellent detection scheme for triplet methylene, since it has a prominent peak. No REMPI scheme for the detection of the first excited singlet 1 1 A 1 state of CH 2 is known, but it can be detected by LIF. Diazomethane and diazirine are major sources of methylene [2,53-56]. The traditional preparation method of diazomethane and diazirine has been recently adapted for work in molecular beams (see Chapter 2). As mentioned above, diazirines are more stable towards organic and inorganic reagents although they are thermally less stable. This fact makes diazirine a preferred precursor compared to diazomethane, and also secondary reactions with unpyrolyzed precursor are less important. In this Chapter, the production and characterization of CH 2 radicals in a molecular beam produced by thermal decomposition of diazomethane or diazirine in a pulsed pyrolysis source is described. The thermochemical thresholds for several dissociation channels are listed in Chapter 2. 6.2 Experimental details The experimental setup, techniques and method for the production of the methylene radicals has been described in Chapter 2, and only some details are elaborated here. 2 + 1 REMPI spectra of CH 2 (1 3 B 1 ) radicals were recorded by integrating parent ion peaks of m/e = 14 as a function of laser excitation wavelength. The radiation was generated by frequency-doubling (Inrad 191 Autotracker III) the linearly polarized output of a Nd:YAG (neodymium: yttrium aluminum garnet) laser pumped dye laser system (Continuum, Surelite III / ND6000, DCM (Exciton); 0.8–2.0 mJ; 40 cm focal lengths lens, " pulse = 5–10 ns). The laser wavelength was calibrated by using published lines of CH 2 and atomic carbon lines in the region 305–317 nm detected in 2 + 1 REMPI [38]. 6.3 Experimental Results Figure 6.2 displays the 2 + 1 REMPI spectra of CH 2 obtained by recording parent ion mass m/e = 14 as a function of two-photon laser excitation at 305–317 nm (65,574–63,091 cm -1 ) following pyrolysis of a) diazomethane and b) diazirine. The spectral features (positions of prominent peaks, spectral shapes, and intensities) are similar to the spectrum recorded by Irikura et al. [38] The strongest band, at 311.80 nm, is identified as two-photon transition to the band origin of the 3p Rydberg state. No vibrational progression or a well-resolved rotational structure was observed. The broad rotational structure could be the result of dissociation of 3 CH 2 to the predissociative 3p Rydberg state. The laser linewidth #$ is ~ 0.05–0.09 cm -1 and not likely to affect the rotational structure. The rotational temperature of the skimmed CH 2 molecular beam was not measured. However, it might be possible to determine the rotational temperature of CH 2 by measuring how changing the pyrolysis heatup conditions affect the spectra (intensity and width) and then modeling the spectra using the program PGOPHER developed by 192 (a) (b) Figure 6.2 2 + 1 REMPI spectrum of CH 2 (1 3 B 1 ) radicals in the region of excitations to the 3p state obtained by measuring m/e = 14 as a function of excitation energy at 305–317 nm (65,574–63,091 cm -1 ) and using a) diazomethane; b) diazirine as a precursors. The sharp peak at 313.48 nm belongs to to the intense 2p3p 1 S 0 ! ! 2p 2 1 D 2 transition of atomic carbon. 193 Western [57]. The sharp peak at 313.48 nm was assigned to the intense 2p3p 1 S 0 ! ! 2p 2 1 D 2 transition of atomic carbon (m/e = 12 ) [58]. 6.4 Future Experiments 6.4.1 Detection of the 1 3 A 1 and 2 3 B 1 3s Rydberg States of CH 2 Since clean molecular beams of methylene radicals have been prepared, the spectroscopy, photoionization, photoelectron spectroscopy, and photodissociation dynamics of low-lying Rydberg states of the triplet methylene can now be studied by using TOF techniques combined with velocity map imaging (VMI). According to calculations, the strongest transitions (f ~ 0.05) are to the 1 3 A 1 and 2 3 B 1 3s states, whose calculated vertical energies are at 6.25 and 7.42 eV, respectively [20,26]. However, after ~ 30 years these states have not yet been characterized experimentally. It might be possible to excite CH 2 in the electronic ground sate to these states with two photons at 330–350 nm and then ionize them with additional photon(s) of the same or different wavelength. Since the ground state geometries of the neutral (1 3 B 1 ) and ion (1 2 A 1 ) are similar [19,26,41,59,60], it is expected that Rydberg states converging to the ground state ion will also have similar geometries. This means that ionization from state-selected vibronic levels of Rydberg states will produce predominantly the same level in the ion. Therefore, and in the absence of state mixing, the photoelectrons will have a single narrow peak in the eKE distributions derived from photoelectron images. Photoelectron VMI will also be useful for revealing ionization mechanisms and possible 194 interactions between rovibronic states of the two Rydberg manifolds that converge to the 1 2 A 1 and 1 2 B 1 ionic states. 6.4.2 Dissociation on the CH 2 1 3 A 1 (3s) Surface Calculations predict that the 1 3 A 1 (3s) state is nearly repulsive [20-22,26]; it dissociates in about 15 fs (only 1 % of the wavepacket survives to 80 fs) [21,22]. The calculated spectrum has a maximum at 6.2 eV that can be accessed by one photon excitation at % < 210 nm. Therefore, it will be interesting to study the photodissociation dynamics and surface crossing of 3 CH 2 following excitation to the 3s Rydberg states. High-level electronic structure and dynamical calculations reveal that in dissociation from the 1 3 A 1 state, only one channel is possible [20-22,26]: CH 2 (1 3 A 1 ) & CH(X 2 %) + H( 2 S), (I) CH vibrational excitation increase at higher photolysis energy and the rotational level distribution has a Gaussian-like shape peaking at N” = 14 for CH(X, v” = 0) with a width of N” ~ 10. The only experimental work shows evidence of production of v” = 0 and 1 of CH(X) at 193 nm (300 K sample) but the rotational level distribution extends to the lowest N” in contrast to theory [39]. Velocity map images of H photofragments following photodissociation of 3 CH 2 using one or two photon excitation schemes will allow for extracting information regarding the CH(X, v”, N”) rovibrational energy distribution. The well-known 195 2 + 1 REMPI scheme via the D 2 % (v’ = 2) state can be used for detection of CH(X, v” = 0, N”) fragments [61-64]. 6.4.3 Dissociation on the 2 3 B 1 /1 3 A 2 Coupled Surfaces Computational calculations predict that there is a valence state 1 3 A 2 arising from 3a 1 ! 1b 2 excitation which has extraordinarily bent geometry ( HCH = 41.61 o , r CH = 1.293 Å, and r HH = 0.9 Å) resembling C + H 2 . Theory also predicts that in the Franck-Condon region ( HCH = 134 o ), there is a conical intersection between valence state 1 3 A 2 and Rydberg state 2 3 B 1 (3s ! 3a 1 ) [20,23,26,29,30], which are both A! in C s symmetry and can be denoted as 2 3 A! and 3 3 A!, respectively [23-25,29,30]. In dissociation from the coupled 2 3 B 1 /1 3 A 2 states, the two possible channels are: CH(1 4 $ - ) + H( 2 S), (II) CH 2 (2 3 B 1 /1 3 A 2 ) & C( 3 P) + H 2 (1 1 $ g + ), (III) Theory predicts that: (i) dissociation can occur only on the 1 3 A 2 surface; (ii) channels (II) and (III) constitute 93 % and 4 %, respectively, of the products; and (iii) the CH(X 2 %) and CH(A 2 &) channels are unimportant. When the 1 3 A 2 state is accessed via the linear 2 3 B 1 state, channel (III) should be favored. However, in linear geometry, the 2 3 B 1 and 1 3 A 1 states, which constitute a Renner-Teller pair, coalesce to the 3 % u state, and some population can be transferred in this way to the CH(X 2 %) channel, a situation not yet addressed by theory. Thus, measurements of the ratios CH(X 2 %): CH(a 4 $ - ): C( 3 P) for channels 196 (II)–(III) will serve as sensitive probes of surface crossing. In particular, it may answer questions such as whether channel (III), so favored by the highly bent geometry of the 1 3 A 2 state, would be suppressed in accessing this surface via 2 3 B 1 state. I propose to monitor H and C( 3 P 2,1,0 ) photofragments by REMPI following two photon excitation at 360–330 nm (7.0–7.5 eV), and determine the change in channel (II) vs. (III) branching ratio as a function of excitation energy. CH(X 2 %) fragments can be detected using 2 + 1 REMPI scheme via the D 2 % (v’ = 2) state [61-64]. No dynamical calculations are available for energy disposal into these two channels, and this information will be extracted from the photofragment images obtained in this study. 197 6.5 Chapter 6 References 1. W.T. Borden, Diradicals. 1982: New York : Wiley. 343 pp. 2. J. Michl and V. Bonacic-Koutecky, Electronic Aspects of Organic Photochemistry. 1990: John Wiley & Sons, Inc. 475 pp. 3. V. Bonacickoutecky, J. Koutecky, and J. Michl, Angew. Chem. Int. Ed., 26, 170, (1987). 4. M.S. Platz, Kinetics and Spectroscopy of Carbenes and Biradicals. 1990: New York : Plenum Press. 372 pp. 5. K.B. Eisenthal, R.A. Moss, and N.J. Turro, Science, 225, 1439, (1984). 6. L. Salem and C. Rowland, Angew. Chem. Int. Ed., 11, 92, (1972). 7. W. Sander, Angew. Chem. Int. Ed., 29, 344, (1990). 8. B.J. Finlayson-Pitts and J.N.P. Jr., Chemistry of the Upper and Lower Atmosphere : Theory, Experiments and Applications. 2000: San Diego, California; Oval Road, London : Academic Press. 969 pp. 9. P.D. Lightfoot, R.A. Cox, J.N. Crowley, M. Destriau, G.D. Hayman, M.E. Jenkin, G.K. Moortgat, and F. Zabel, Atmos. Environ. Part A, 26, 1805, (1992). 10. K.N. Houk, J. Gonzalez, and Y. Li, Acc. Chem. Res., 28, 81, (1995). 11. S. Wilsey, K.N. Houk, and A.H. Zewail, J. Am. Chem. Soc., 121, 5772, (1999). 12. A.A. Scala, E.W.G. Diau, Z.H. Kim, and A.H. Zewail, J. Chem. Phys., 108, 7933, (1998). 13. S. De Feyter, E.W.G. Diau, A.A. Scala, and A.H. Zewail, Chem. Phys. Lett., 303, 249, (1999). 14. S. Pedersen, J.L. Herek, and A.H. Zewail, Science, 266, 1359, (1994). 15. R.L. Schwartz, G.E. Davico, T.M. Ramond, and W.C. Lineberger, J. Phys. Chem. A, 103, 8213, (1999). 16. P.G. Wenthold, R.R. Squires, and W.C. Lineberger, J. Am. Chem. Soc., 120, 5279, (1998). 198 17. P.G. Wenthold, J. Hu, R.R. Squires, and W.C. Lineberger, J. Am. Chem. Soc., 118, 475, (1996). 18. D.C. Comeau, I. Shavitt, P. Jensen, and P.R. Bunker, J. Chem. Phys., 90, 6491, (1989). 19. P. Jensen and P.R. Bunker, J. Chem. Phys., 89, 1327, (1988). 20. R.A. Bearda, M.C. Vanhemert, and E.F. Vandishoeck, J. Chem. Phys., 97, 8240, (1992). 21. R.A. Bearda, G.J. Kroes, M.C. Vanhemert, B. Heumann, R. Schinke, and E.F. Vandishoeck, J. Chem. Phys., 100, 1113, (1994). 22. G.J. Kroes, E.F. Vandishoeck, R.A. Bearda, and M.C. Vanhemert, J. Chem. Phys., 99, 228, (1993). 23. G.J. Kroes, M.C. vanHemert, G.D. Billing, and D. Neuhauser, J. Chem. Phys., 107, 5757, (1997). 24. R.A. Bearda, M.C. Vanhemert, and E.F. Vandishoeck, J. Chem. Phys., 102, 8930, (1995). 25. E.F. vanDishoeck, R.A. Bearda, and M.C. vanHemert, Astron. Astrophys., 307, 645, (1996). 26. J. Romelt, S.D. Peyerimhoff, and R.J. Buenker, Chem. Phys., 54, 147, (1981). 27. Y. Yamaguchi and H.F. Schaefer, J. Chem. Phys., 106, 1819, (1997). 28. Y. Yamaguchi and H.F. Schaefer, J. Chem. Phys., 106, 8753, (1997). 29. D.R. Yarkony, J. Chem. Phys., 104, 2932, (1996). 30. N. Matsunaga and D.R. Yarkony, J. Chem. Phys., 107, 7825, (1997). 31. G. Herzberg and J. Shoosmith, Nature, 183, 1801, (1959). 32. G. Herzberg, Proc. R. Soc. London, Ser. A, 262, 291, (1961). 33. G. Herzberg, Proc. R. Soc. London, Ser. A, 262, 291, (1961). 34. G. Herzberg, Can. J. Phys., 39, 1511, (1961). 199 35. G. Herzberg, Electronic Spectra and Electronic Structure of Polyatomic Molecules (Molecular Spectra and Molecular Structure). Vol. III. 1966: Van Nostrand, New York. 875 pp. 36. G. Herzberg and J.W.C. Johns, J. Chem. Phys., 54, 2276, (1971). 37. K.K. Irikura and J.W. Hudgens, J. Phys. Chem., 96, 518, (1992). 38. K.K. Irikura, R.D. Johnson, and J.W. Hudgens, J. Phys. Chem., 96, 6131, (1992). 39. B. Bohn and F. Stuhl, J. Chem. Phys., 102, 8842, (1995). 40. C. Kassner, F. Stuhl, M. Luo, M. Lehner, R. Fink, and M. Jungen, J. Chem. Phys., 105, 4605, (1996). 41. W. Reuter and S.D. Peyerimhoff, Chem. Phys., 160, 11, (1992). 42. M. Litorja and B. Ruscic, J. Chem. Phys., 108, 6748, (1998). 43. B. Ruscic, M. Litorja, and R.L. Asher, J. Phys. Chem. A, 103, 8625, (1999). 44. G. Herzberg and J.W.C. Johns, Proc. R. Soc. London, Ser. A, 295, 107, (1966). 45. M.E. Jacox, J. Phys. Chem. Ref. Data, 1, (1994). 46. P.R. Bunker, P. Jensen, Y. Yamaguchi, and H.F. Schaefer, J. Mol. Spectrosc., 179, 263, (1996). 47. P.R. Bunker, P. Jensen, Y. Yamaguchi, and H.F. Schaefer, J. Phys. Chem., 100, 18088, (1996). 48. Y. Yamaguchi, C.D. Sherrill, and H.F. Schaefer, J. Phys. Chem., 100, 7911, (1996). 49. A.V. Komissarov, A. Lin, T.J. Sears, and G.E. Hall, J. Chem. Phys., 125, (2006). 50. Y. Kim, G.E. Hall, and T.J. Sears, Phys. Chem. Chem. Phys., 8, 2823, (2006). 51. K. Kobayashi, G.E. Hall, and T.J. Sears, J. Chem. Phys., 124, (2006). 52. G.E. Hall, T.J. Sears, and H.G. Yu, J. Mol. Spectrosc., 235, 125, (2006). 200 53. H. Zollinger, Azo and Diazo Chemistry; Aliphatic and Aromatic Compounds. 1961: Interscience Publishers, New York. 444 pp. 54. M. Regitz and G. Maas, Diazo Compounds : Properties and Synthesis 1986: Academic Press, Inc. 596 pp. 55. M.T.H. Liu, ed. Chemistry of Diazirines. 1987, CRC Press: Boca Raton, FL. 56. M.P. Doyle, M.A. McKervey, and T. Ye, Modern Catalytic Methods for Organic Synthesis with Diazo Compounds : from Cyclopropanes to Ylides 1998: John Wiley & Sonc, Inc. 652 pp. 57. C.M. Western, PGOPHER, A Program for Simulating Rotational Structure, University of Bristol, http://pgopher.chm.bris.ac.uk, 58. C.E. Moore, Atomic energy levels. (NSRDS-NBS 3 (Sect. 4)), A7 IV-1-B7 VII-3. Vol. 1. 1971: Official Standards Reference Data Series, National Bureau of Standards 35; U.S. Goverment: Washington, D.C. 59. S. Willitsch and F. Merkt, Journal of Chemical Physics, 118, 2235, (2003). 60. R. Schinke, Photodissociation dynamics : spectroscopy and fragmentation of small polyatomic molecules. Cambridge monographs on atomic, molecular, and chemical physics ; 1. 1993: Cambridge ; New York : Cambridge University Press. 417. 61. P. Chen, W.A. Chupka, and S.D. Colson, Chem. Phys. Lett., 121, 405, (1985). 62. P. Chen, J.B. Pallix, W.A. Chupka, and S.D. Colson, J. Chem. Phys., 86, 516, (1987). 63. J.W. Hudgens, C.S. Dulcey, G.R. Long, and D.J. Bogan, J. Chem. Phys., 87, 4546, (1987). 64. Y.M. Wang, L.P. Li, and W.A. Chupka, Chem. Phys. Lett., 192, 348, (1992). 201 Bibliography D.W. Adamson and J. Kenner, J. Chem. Soc., 1551, (1937). H.Y. Afeefy, J.F. Liebman, and S.E. Stein, Neutral Thermochemical Data, in Nist Chemistry Webbook, Nist Standard Reference Database Number 69,. 2003, edited by Linstrom, P. J. and Mallard, W. G. (NIST, Gaithersburg, MD, 2003) <http://webbook.nist.gov/>. M. Alcami, J.L.G. Depaz, and M. Yanez, J. Comput. Chem., 10, 468, (1989). W.R. Anderson, J. Phys. Chem., 93, 530, (1989). A. Angeli, Accademia Nazionale dei Lincei, Classe di Scienze Fisiche, Matematiche e Naturali, Rendiconti, 16, 790, (1907). A. Angeli, Accademia Nazionale dei Lincei, Classe di Scienze Fisiche, Matematiche e Naturali, Rendiconti, 20, 625, (1911). T.G. Archibald, Chim. Oggi, 18, 34, (2000). J.F. Arenas, I. Lopez-Tocon, J.C. Otero, and J. Soto, J. Am. Chem. Soc., 124, 1728, (2002). M.N.R. Ashfold, M.A. Fullstone, G. Hancock, and G.W. Ketley, Chem. Phys., 55, 245, (1981). D.W. Ball, THEOCHEM, 722, 213, (2005). S.M. Ball, G. Hancock, and M.R. Heal, J. Chem. Soc., Faraday Trans., 90, 523, (1994). J. Bastide and J.P. Maier, Chem. Phys., 12, 177, (1976). R.A. Bearda, G.J. Kroes, M.C. Vanhemert, B. Heumann, R. Schinke, and E.F. Vandishoeck, J. Chem. Phys., 100, 1113, (1994). R.A. Bearda, M.C. Vanhemert, and E.F. Vandishoeck, J. Chem. Phys., 97, 8240, (1992). R.A. Bearda, M.C. Vanhemert, and E.F. Vandishoeck, J. Chem. Phys., 102, 8930, (1995). A.D. Becke, J. Chem. Phys., 98, 5648, (1993). 202 J.A. Bell, J. Chem. Phys., 41, 2556, (1964). S.W. Benson, Cruicksh.Fr, D.M. Golden, G.R. Haugen, H.E. Oneal, A.S. Rodgers, R. Shaw, and R. Walsh, Chem. Rev., 69, 279, (1969). M.R. Berman, T. Tsuchiya, A. Gregusova, S.A. Perera, and R.J. Bartlett, J. Phys. Chem. A, 111, 6894, (2007). M.T. Berry, M.R. Brustein, and M.I. Lester, Chem. Phys. Lett., 153, 17, (1988). D. Bhattacharya and J.E. Willard, J. Phys. Chem., 86, 967, (1982). B. Bigot, R. Ponec, A. Sevin, and A. Devaquet, J. Am. Chem. Soc., 100, 6575, (1978). H. Bitto, I.C. Chen, and C.B. Moore, J. Chem. Phys., 85, 5101, (1986). U. Bley and F. Temps, J. Chem. Phys., 98, 1058, (1993). Boersch, Monatshefte fuer Chemie, 65, 311, (1935). M. Bogey, M. Winnewisser, and J.J. Christiansen, Can. J. Phys., 62, 1198, (1984). B. Bohn and F. Stuhl, J. Chem. Phys., 102, 8842, (1995). R.B. Bohn, S.A. Sandford, L.J. Allamandola, and D.P. Cruikshank, Icarus, 111, 151, (1994). A.I. Boldyrev, P.V. Schleyer, D. Higgins, C. Thomson, and S.S. Kramarenko, J. Comput. Chem., 13, 1066, (1992). E. Bombardelli, B. Gabetta, and A. Pontiroli, Process for the Preparation of Taxane Carbonate Derivatives from 10-Deacetylbaccatin Iii for Use as Anticancer Drugs, P.I. Appl., Editor. 2001. p. 31. V. Bonacickoutecky, J. Koutecky, and J. Michl, Angew. Chem. Int. Ed., 26, 170, (1987). W.T. Borden, Diradicals. 1982: New York : Wiley. 343 pp. W.J. Bouma, R.H. Nobes, L. Radom, and C.E. Woodward, J. Org. Chem., 47, 1869, (1982). J.N. Bradley, A. Ledwith, and G.W. Cowell, J. Chem. Soc., 353, (1964). 203 V. Branchadell, E. Muray, A. Oliva, R.M. Ortuno, and C. Rodriguez-Garcia, J. Phys. Chem. A, 102, 10106, (1998). W. Braun, A.M. Bass, and M. Pilling, J. Chem. Phys., 52, 5131, (1970). R. Brinton and D. Volman, J. Chem. Phys., 19, 1394, (1951). P.R. Bunker, P. Jensen, Y. Yamaguchi, and H.F. Schaefer, J. Mol. Spectrosc., 179, 263, (1996). P.R. Bunker, P. Jensen, Y. Yamaguchi, and H.F. Schaefer, J. Phys. Chem., 100, 18088, (1996). P.R.J. Bunker, Per., Molecular Symmetry and Spectroscopy. 1998: NRC Research, Ottawa. 747 pp. J.E. Butler, L.P. Goss, M.C. Lin, and J.W. Hudgens, Chem. Phys. Lett., 63, 104, (1979). L. Catoire and M.T. Swihart, J. Propul. Power, 18, 1242, (2002). P. Chakrabarti and H.G. Khorana, Biochemistry, 14, 5021, (1975). D.W. Chandler and P.L. Houston, J. Chem. Phys., 87, 1445, (1987). B.Y. Chang, R.C. Hoetzlein, J.A. Mueller, J.D. Geiser, and P.L. Houston, Rev. Sci. Instrum., 69, 1665, (1998). R. Chaudhuri, D. Mukhopadhyay, and D. Mukherjee, Chem. Phys. Lett., 162, 393, (1989). I.C. Chen, W.H. Green, and C.B. Moore, J. Chem. Phys., 89, 314, (1988). P. Chen, W.A. Chupka, and S.D. Colson, Chem. Phys. Lett., 121, 405, (1985). P. Chen, S.D. Colson, W.A. Chupka, and J.A. Berson, J. Phys. Chem., 90, 2319, (1986). P. Chen, J.B. Pallix, W.A. Chupka, and S.D. Colson, J. Chem. Phys., 86, 516, (1987). Y. Chen, J. Jin, L.S. Pei, X.X. Ma, and C.X. Chen, J. Electron. Spectrosc. Relat. Phenom., 108, 221, (2000). J.J. Chu, P. Marcus, and P.J. Dagdigian, J. Chem. Phys., 93, 257, (1990). 204 M.Y. Chu and J.S. Dahler, Mol. Phys., 27, 1045, (1974). E.P. Clifford, P.G. Wenthold, W.C. Lineberger, G.A. Petersson, K.M. Broadus, S.R. Kass, S. Kato, C.H. DePuy, V.M. Bierbaum, and G.B. Ellison, J. Phys. Chem. A, 102, 7100, (1998). D.C. Comeau, I. Shavitt, P. Jensen, and P.R. Bunker, J. Chem. Phys., 90, 6491, (1989). A.P.T. Cox, L. F.; Sheridan, J., Nature, 181, 1000, (1958). B.L. Crawford, Jr., W.H. Fletcher, and D.A. Ramsay, J. Chem. Phys., 19 406, (1951). A.G. Csazar, P.G. Szalay, and M.L. Leininger, Mol. Phys., 100, 3879, (2002). T. Curtius, Ber. Dtsch. Chem. Ges., 16, 2230, (1883). T. Curtius, Ber. Dtsch. Chem. Ges., 22, 2161, (1889). S. Davis, D.T. Anderson, G. Duxbury, and D.J. Nesbitt, J. Chem. Phys., 107, 5661, (1997). S. De Feyter, E.W.G. Diau, A.A. Scala, and A.H. Zewail, Chem. Phys. Lett., 303, 249, (1999). D.A. Dixon, W.A. de Jong, K.A. Peterson, and T.B. McMahon, J. Phys. Chem. A, 109, 4073, (2005). M.P. Doyle, M.A. McKervey, and T. Ye, Modern Catalytic Methods for Organic Synthesis with Diazo Compounds : From Cyclopropanes to Ylides 1998: John Wiley & Sonc, Inc. 652 pp. V. Dribinski, Photoelectron and Ion Imaging Studies of the Mixed Valence/Rydberg Excited States of the Chloromethyl Radical, CH 2 Cl, and the Nitric Oxide Dimer, (NO)2, Dissertation, Chemistry, University of Southern California, Los Angeles, 2004. V. Dribinski, A. Ossadtchi, V.A. Mandelshtam, and H. Reisler, Rev. Sci. Instrum., 73, 2634, (2002). V. Dribinski, A.B. Potter, I. Fedorov, and H. Reisler, Chem. Phys. Lett., 385, 233, (2004). 205 W.W. Duley and D.A. Williams, Interstellar Chemistry 1984: Academic Press: New York. 251 pp. T.H. Dunning Jr., J. Chem. Phys., 90, 1007, (1989). T.H. Dunning Jr. and P.J. Hay, In Modern Theoretical Chemistry, H.F. Schaefer, Editor. 1977, Plenum Press: New York. p. 1. K.B. Eisenthal, R.A. Moss, and N.J. Turro, Science, 225, 1439, (1984). P.C. Engelking, Chem. Rev., 91, 399, (1991). A.T.J.B. Eppink and D.H. Parker, Rev. Sci. Instrum., 68, 3477, (1997). B. Erni and H.G. Khorana, J. Am. Chem. Soc., 102, 3888, (1980). R. Ettinger, J. Chem. Phys., 38, 2427, (1963). R. Ettinger, J. Chem. Phys., 40, 1693, (1964). A. Fadini, E. Glozbach, P. Krommes, and J. Lorberth, J. Organomet. Chem., 149, 297, (1978). A.E. Faulhaber, J.R. Gascooke, A.A. Hoops, and D.M. Neumark, J. Chem. Phys., 124, (2006). I. Fedorov, L. Koziol, G.S. Li, J.A. Parr, A.I. Krylov, and H. Reisler, J. Phys. Chem. A, 111, 4557, (2007). I. Fedorov, L. Koziol, G.S. Li, H. Reisler, and A.I. Krylov, J. Phys. Chem. A, 111, 13347, (2007). I. Fedorov, L. Koziol, A.K. Mollner, A.I. Krylov, and H. Reisler, J. Phys. Chem. A, (2009). E.J. Feltham, R.H. Qadiri, E.E.H. Cottrill, P.A. Cook, J.P. Cole, G.G. Balint-Kurti, and M.N.R. Ashfold, J. Chem. Phys., 119, 6017, (2003). E.E. Ferguson, F.C. Fehsenfeld, and D.L. Albritton, Ion Chemistry of the Earth's Atmosphere. Gas Phase Ion Chemistry, ed. M.T. Bowers. Vol. 1. 1979: Academic: New York. B.J. Finlayson-Pitts and J.N.P. Jr., Chemistry of the Upper and Lower Atmosphere : Theory, Experiments and Applications. 2000: San Diego, California; Oval Road, London : Academic Press. 969 pp. 206 W.H. Fletcher and T.P. Garrett, J. Chem. Phys., 25, 50, (1956). M.S. Foster and Beaucham.Jl, J. Am. Chem. Soc., 94, 2425, (1972). A.P.N. Franchimont, Recl. Trav. Chim. Pays-Bas, 9, (1890). H.M. Frey and I.D.R. Stevens, J. Am. Chem. Soc., 84, 2647, (1962). A. Gambi, J. Mol. Spectrosc., 103, 321, (1984). A. Gambi, M. Pedrali, M. Winnewisser, and G. Guelachvili, J. Mol. Spectrosc., 113, 250, (1985). A. Gambi, U.P. Verma, M. Winnewisser, and G. Guelachvili, J. Mol. Spectrosc., 108, 264, (1984). A. Gambi, M. Winnewisser, and J.J. Christiansen, J. Mol. Spectrosc., 98, 413, (1983). I. Garciamoreno, E.R. Lovejoy, and C.B. Moore, J. Chem. Phys., 100, 8902, (1994). M.K. Georgieva, Bulgarian Chem. Commun., 35, 195, (2003). J.D. Gezelter and W.H. Miller, J. Chem. Phys., 103, 7868, (1995). E.D.B. Glendening, J. K.; Reed, A. E.; Carpenter, J. and J.A.M. E.; Bohmann, C. M.; Weinhold, F., Nbo. 2001: Theoretical Chemistry Institute, University of Wisconsin: Madison, WI,. N. Goldberg, A. Fiedler, and H. Schwarz, Helv. Chim. Acta, 77, 2354 (1994). M.S. Gordon and S.R. Kass, J. Phys. Chem., 99, 6548, (1995). M.S. Gordon and S.R. Kass, J. Phys. Chem. A, 101, 7922, (1997). W.H. Graham, J. Am. Chem. Soc., 84, 1063, (1962). P. Griefs, Justus Liebigs Ann. Chem., 106, 123, (1858). C. Guimon, S. Khayar, F. Gracian, M. Begtrup, and G. Pfister-Guillouzo, Chem. Phys., 138, 157, (1989). Y. Guo and D.L. Thompson, J. Chem. Phys., 116, 3670, (2002). M.P. Habas and A. Dargelos, Chem. Phys., 199, 177, (1995). 207 O.F. Hagena and W. Obert, J. Chem. Phys., 56, 1793, (1972). G.E. Hall, T.J. Sears, and H.G. Yu, J. Mol. Spectrosc., 235, 125, (2006). M.S. Han, H.-G. Cho, and B.-S. Cheong, Bull. Korean Chem. Soc., 20, 1281, (1999). L.B. Harding, S.J. Klippenstein, and J.A. Miller, J. Phys. Chem. A, 112, 522, (2008). J.C. Hassler and D.W. Setser, J. Am. Chem. Soc., 87, 3793, (1965). S.M. Hecht and J.W. Kozarich, Tetrahedron Lett., 1501, (1972). U. Hechtfischer, C.J. Williams, M. Lange, J. Linkemann, D. Schwalm, R. Wester, A. Wolf, and D. Zajfman, J. Chem. Phys., 117, 8754, (2002). T. Helgaker, P. Jorgensen, and J. Olsen, Molecular Electronic-Structure Theory. 2000. 908 pp. G. Herzberg, Proc. R. Soc. London, Ser. A, 262, 291, (1961). G. Herzberg, Can. J. Phys., 39, 1511, (1961). G. Herzberg, Electronic Spectra and Electronic Structure of Polyatomic Molecules (Molecular Spectra and Molecular Structure). Vol. III. 1966: Van Nostrand, New York. 875 pp. G. Herzberg, ed. Spectra of Diatomic Molecules (Molecular Spectra and Molecular Structure). Vol. I. 1966, New York, Van Nostrand. G. Herzberg and J.W.C. Johns, Proc. R. Soc. London, Ser. A, 295, 107, (1966). G. Herzberg and J.W.C. Johns, Astrophys. J., 158, 399, (1969). G. Herzberg and J.W.C. Johns, J. Chem. Phys., 54, 2276, (1971). G. Herzberg and J. Shoosmith, Nature, 183, 1801, (1959). S. Hirata, M. Nooijen, and R.J. Bartlett, Chem. Phys. Lett., 326, 255, (2000). H. Hotoda, M. Kaneko, M. Inukai, Y. Muramatsu, Y. Utsui, and S. Ohya, Preparation of a-500359 Derivatives as Antibacterial Agents. 2001. p. 459. K.N. Houk, J. Gonzalez, and Y. Li, Acc. Chem. Res., 28, 81, (1995). 208 K.P. Huber and G. Herzberg, Molecular Spectra and Molecular Structure. Constants of Diatomic Molecules. Vol. IV. 1979: Van Nostrand Reinhold Company: New York. J.W. Hudgens, C.S. Dulcey, G.R. Long, and D.J. Bogan, J. Chem. Phys., 87, 4546, (1987). K.K. Irikura and J.W. Hudgens, J. Phys. Chem., 96, 518, (1992). K.K. Irikura, R.D. Johnson, and J.W. Hudgens, J. Phys. Chem., 96, 6131, (1992). W.M. Jackson and H. Okabe, Adv. Photochem., 13, 1, (1986). M.E. Jacox, J. Phys. Chem. Ref. Data, 1, (1994). P. Jensen and P.R. Bunker, J. Chem. Phys., 89, 1327, (1988). M. Jones, Jr. and R.A. Moss, Carbenes. Vol. 1. 1973: A Wiley-Interscience publication, New York. A. Kalemos, A. Mavridis, and A. Metropoulos, J. Chem. Phys., 111, 9536, (1999). A. Kantrowitz and J. Grey, Rev. Sci. Instrum., 22, 328, (1951). C. Kassner, F. Stuhl, M. Luo, M. Lehner, R. Fink, and M. Jungen, J. Chem. Phys., 105, 4605, (1996). I.S.K. Kerkines, P. Carsky, and A. Mavridis, J. Phys. Chem. A, 109, 10148, (2005). M. Khlifi, P. Paillous, P. Bruston, F. Raulin, and J.C. Guillemin, Icarus, 124, 318, (1996). S.-U. Kim, Antimalarial Artemisinin Derivatives. 1998. S.U. Kim, J.H. Choi, and S.S. Kim, J. Nat. Prod., 56, 857, (1993). T.S. Kim, S.K. Kim, Y.S. Choi, and I. Kwak, J. Chem. Phys., 107, 8719, (1997). Y. Kim, G.E. Hall, and T.J. Sears, Phys. Chem. Chem. Phys., 8, 2823, (2006). F.W.N. Kirkbride, Ronald G. W., J. Chem. Soc., 119, (1933). W. Kirmse, Carbene Chemistry. 2nd ed. 1971: Academic: New York. 622 pp. G.B. Kistiakowsky and W.P. Slichter, Rev. Sci. Instrum., 22, 333, (1951). 209 T.N. Kitsopoulos, C.R. Gebhardt, and T.P. Rakitzis, J. Chem. Phys., 115, 9727, (2001). A. Klip and C. Gitler, Biochem. Biophys. Res. Commun., 60, 1155, (1974). K. Kobayashi, G.E. Hall, and T.J. Sears, J. Chem. Phys., 124, (2006). H. Koch, H.J.A. Jensen, P. Jorgensen, and T. Helgaker, J. Chem. Phys., 93, 3345, (1990). D.W. Kohn, H. Clauberg, and P. Chen, Rev. Sci. Instrum., 63, 4003, (1992). H. Koizumi, G.C. Schatz, and J.M. Bowman, ACS Symp. Ser., 502, 37, (1992). H. Koizumi, G.C. Schatz, and S.P. Walch, J. Chem. Phys., 95, 4130, (1991). A.V. Komissarov, A. Lin, T.J. Sears, and G.E. Hall, J. Chem. Phys., 125, (2006). A.V. Komissarov, M.P. Minitti, A.G. Suits, and G.E. Hall, J. Chem. Phys., 124, (2006). J. Kong, C.A. White, A.I. Krylov, D. Sherrill, R.D. Adamson, T.R. Furlani, M.S. Lee, A.M. Lee, S.R. Gwaltney, T.R. Adams, C. Ochsenfeld, A.T.B. Gilbert, G.S. Kedziora, V.A. Rassolov, D.R. Maurice, N. Nair, Y.H. Shao, N.A. Besley, P.E. Maslen, J.P. Dombroski, H. Daschel, W.M. Zhang, P.P. Korambath, J. Baker, E.F.C. Byrd, T. Van Voorhis, M. Oumi, S. Hirata, C.P. Hsu, N. Ishikawa, J. Florian, A. Warshel, B.G. Johnson, P.M.W. Gill, M. Head-Gordon, and J.A. Pople, J. Comput. Chem., 21, 1532, (2000). G. Kortum, Z. Phys. Chem. Abt. B:, 50, 361, (1941). L. Koziol, S.V. Levchenko, and A.I. Krylov, J. Phys. Chem. A, 110, 2746, (2006). R. Krishnan, J.S. Binkley, R. Seeger, and J.A. Pople, J. Chem. Phys., 72, 650, (1980). G.J. Kroes, E.F. Vandishoeck, R.A. Bearda, and M.C. Vanhemert, J. Chem. Phys., 99, 228, (1993). G.J. Kroes, M.C. vanHemert, G.D. Billing, and D. Neuhauser, J. Chem. Phys., 107, 5757, (1997). A.I. Krylov, Annu. Rev. Phys. Chem., 59, 433, (2008). K.C. Kulander and J.S. Dahler, J. Phys. Chem., 80, 2881, (1976). 210 K.C. Kulander and J.S. Dahler, Chem. Phys. Lett., 41, 125, (1976). I. Langmuir, J. Am. Chem. Soc., 41, 1543, (1919). I. Langmuir, J. Am. Chem. Soc., 42, 2190, (1920). A.H. Laufer and H. Okabe, J. Am. Chem. Soc., 93, 4137, (1971). A.H. Laufer and H. Okabe, J. Phys. Chem., 76, 3504, (1972). H. Lee, Y. Miyamoto, and Y. Tateyama, J. Org. Chem., 74, 562, (2009). W.Y. Lee, W.B. Lee, H. Fu, C.C. Pan, and K.C. Lin, J. Chem. Phys., 115, 7429, (2001). S.V. Levchenko, A.V. Demyanenko, V.L. Dribinski, A.B. Potter, H. Reisler, and A.I. Krylov, J. Chem. Phys., 118, 9233, (2003). S.V. Levchenko and A.I. Krylov, J. Chem. Phys., 120, 175, (2004). S.V. Levchenko, T. Wang, and A.I. Krylov, J. Chem. Phys., 122, (2005). S.G. Lias, J.E. Bartmess, J.F. Liebman, J.L. Holmes, R.D. Levin, and W.G. Mallard, J. Phys. Chem. Ref. Data, 17, 1, (1988). J. Lievin and G. Verhaegen, Theor. Chim. Acta, 42, 47, (1976). P.D. Lightfoot, R.A. Cox, J.N. Crowley, M. Destriau, G.D. Hayman, M.E. Jenkin, G.K. Moortgat, and F. Zabel, Atmos. Environ. Part A, 26, 1805, (1992). E.C.E. Lim, Excited States. Vol. 3. 1977: (Academic, New York, N. Y.). 351 pp. S.M. Lim, T.S. Kim, G.I. Lim, S.K. Kim, and Y.S. Choi, Chem. Phys. Lett., 288, 828, (1998). M. Litorja and B. Ruscic, J. Chem. Phys., 108, 6748, (1998). J. Liu, F.Y. Wang, H. Wang, B. Jiang, and X.M. Yang, J. Chem. Phys., 122, (2005). M.T.H. Liu, Chem. Soc. Rev., 11, 127, (1982). M.T.H. Liu, ed. Chemistry of Diazirines. 1987, CRC Press: Boca Raton, FL. E.R. Lovejoy, S.K. Kim, R.A. Alvarez, and C.B. Moore, J. Chem. Phys., 95, 4081, (1991). 211 E.R. Lovejoy and C.B. Moore, J. Chem. Phys., 98, 7846, (1993). I.C. Lu, S.H. Lee, Y.T. Lee, and X.M. Yang, J. Chem. Phys., 124, (2006). X. Lu, X. Xu, N. Wang, and Q. Zhang, J. Org. Chem., 67, (2002). M.R. Manaa and D.R. Yarkony, J. Chem. Phys., 95, 1808, (1991). M.R. Manaa and D.R. Yarkony, Chem. Phys. Lett., 188, 352, (1992). J. Mason and J.G. Vinter, J. Chem. Soc., Dalton Trans., 2522, (1975). N. Matsunaga and D.R. Yarkony, J. Chem. Phys., 107, 7825, (1997). M.H. Matus, A.J. Arduengo, and D.A. Dixon, J. Phys. Chem. A, 110, 10116, (2006). J.R. Mcdonald, J.W. Rabalais, and S.P. Mcglynn, J. Chem. Phys., 52, 1332, (1970). A.F. McKay, J. Am. Chem. Soc., 70, 1974, (1948). A.F.W. McKay, George F., J. Am. Chem. Soc., 69, 3028, (1947). A.D. Mclean and G.S. Chandler, J. Chem. Phys., 72, 5639, (1980). A.D. Mclean, G.H. Loew, and D.S. Berkowitz, J. Mol. Spectrosc., 64, 184, (1977). A.J. Merer, Can. J. Phys., 42, 1242, (1964). J.A. Merritt, Can. J. Phys., 40, 1683, (1962). J. Michl and V. Bonacic-Koutecky, Electronic Aspects of Organic Photochemistry. 1990: John Wiley & Sons, Inc. 475 pp. I.M. Mills and H.W. Thompson, Trans. Faraday Soc., 50, 1270, (1954). M. Mladenovic and M. Lewerenz, Chem. Phys., 343, 129, (2008). D.A. Modarelli and M.S. Platz, J. Am. Chem. Soc., 113, 8985, (1991). D.A. Modarelli and M.S. Platz, J. Am. Chem. Soc., 115, 470, (1993). J.B. Moffat, J. Mol. Struct., 52, 275, (1979). K. Moller, J. Vogt, M. Winnewisser, and J.J. Christiansen, Can. J. Phys., 62, 1217, (1984). 212 C. Moore and G. Pimentel, J. Chem. Phys., 40, 329, (1964). C.B. Moore, J. Chem. Phys., 39, 1884, (1963). C.B. Moore, Faraday Discuss., 102, 1, (1995). C.B. Moore and G.C. Pimentel, J. Chem. Phys., 40, 342, (1964). C.B. Moore and G.C. Pimentel, J. Chem. Phys., 40, 1529, (1964). C.B. Moore and G.C. Pimentel, J. Chem. Phys., 41, 3504, (1964). C.E. Moore, Atomic Energy Levels. (Nsrds-Nbs 3 (Sect. 4)), A7 Iv-1-B7 Vii-3. Vol. 1. 1971: Official Standards Reference Data Series, National Bureau of Standards 35; U.S. Goverment: Washington, D.C. M.H. Moore and R.L. Hudson, Icarus, 161, 486, (2003). D.H. Mordaunt, D.L. Osborn, H. Choi, R.T. Bise, and D.M. Neumark, J. Chem. Phys., 105, 6078, (1996). L.V. Moskaleva and M.C. Lin, Proc. Combust. Inst., 28, 2393, (2000). L.V. Moskaleva and M.C. Lin, Z. Phys. Chem., 215, 1043, (2001). L.V. Moskaleva, W.S. Xia, and M.C. Lin, Chem. Phys. Lett., 331, 269, (2000). V.C. Mota and A.J.C. Varandas, J. Phys. Chem. A, 111, 10191, (2007). E. Muller and W. Kreutzmann, Justus Liebigs Ann. Chem., 512, 264, (1934). E. Muller and D. Ludsteck, Chem. Ber. Recl., 87, 1887, (1954). F. Muller and W. Rundel, Chem. Ber. Recl., 88, 917, (1955). L. Nemes, J. Vogt, and M. Winnewisser, J. Mol. Struct., 218, 219, (1990). D.J. Nesbitt, H. Petek, M.F. Foltz, S.V. Filseth, D.J. Bamford, and C.B. Moore, J. Chem. Phys., 83, 223, (1985). J.E. Ogara and W.P. Dailey, J. Am. Chem. Soc., 114, 3581, (1992). J.F. Ogilvie, Photochem. Photobiol., 9, 65, (1969). H. Okabe, Photochemistry of Small Molecules. 1978: (Wiley, New York, N. Y.). 432 pp. 213 D.L. Osborn, D.H. Mordaunt, H. Choi, R.T. Bise, D.M. Neumark, and C.M. Rohlfing, J. Chem. Phys., 106, 10087, (1997). S. Pal, M. Rittby, R.J. Bartlett, D. Sinha, and D. Mukherjee, Chem. Phys. Lett., 137, 273, (1987). J.B. Pallix, P. Chen, W.A. Chupka, and S.D. Colson, J. Chem. Phys., 84, 5208, (1986). A. Papakondylis and A. Mavridis, J. Phys. Chem. A, 103, 1255, (1999). S. Patai, ed. The Chemistry of Diazonium and Diazo Groups. 1978, John Wiley & Sons. G.S. Paulett and R. Ettinger, J. Chem. Phys., 39, 3534, (1963). G.S. Paulett and R. Ettinger, J. Chem. Phys., 39, 825, (1963). G.S. Paulett and R. Ettinger, J. Chem. Phys., 41, 2557, (1964). S.R. Paulsen, Angew. Chem. Int. Ed., 72, 781, (1960). P. Pd and P. Chandra, J. Chem. Phys., 114, 7450, (2001). R. Pd and P. Chandra, J. Chem. Phys., 114, 1589, (2001). S. Pedersen, J.L. Herek, and A.H. Zewail, Science, 266, 1359, (1994). L. Pierce and V. Dobyns, J. Am. Chem. Soc., 84, 2651, (1962). R.H. Pierson, A.N. Fletcher, and E.S.C. Gantz, Anal. Chem., 28 1218, (1956). N. Pinnavaia, M.J. Bramley, M.D. Su, W.H. Green, and N.C. Handy, Mol. Phys., 78, 319, (1993). M.S. Platz, Kinetics and Spectroscopy of Carbenes and Biradicals. 1990: New York : Plenum Press. 372 pp. Potter, Ion Imaging Studies of the Spectroscopy and Photodissociation Dynamics of Chloromethyl Radical and Nitric Oxide Dimer , Chemistry, Universuty of Southern California, Los Angeles, 2005. G.D. Purvis and R.J. Bartlett, J. Chem. Phys., 76, 1910, (1982). C. Puzzarini and A. Gambi, Chem. Phys., 306, 131, (2004). 214 C. Puzzarini, A. Gambi, and G. Cazzoli, J. Mol. Struct., 695, 203, (2004). J.W. Rabalais, J.M. Mcdonald, V. Scherr, and S.P. McGlynn, Chem. Rev., 71, 73, (1971). J.W. Rabalais, J.R. Mcdonald, and S.P. Mcglynn, J. Chem. Phys., 51, 5095, (1969). J.W. Rabalais, J.R. Mcdonald, and S.P. Mcglynn, J. Chem. Phys., 51, 5103, (1969). K. Raghavachari, G.W. Trucks, J.A. Pople, and M. Headgordon, Chem. Phys. Lett., 157, 479, (1989). T.P. Rakitzis and T.N. Kitsopoulos, J. Chem. Phys., 116, 9228, (2002). D.A. Ramsay, J. Chem. Phys., 17, 666, (1949). F. Raulin, M. Khlifi, M. Dang-Nhu, and D. Gautier, Adv. Space Res., 12, 11/181, (1992). M. Regitz and G. Maas, Diazo Compounds : Properties and Synthesis 1986: Academic Press, Inc. 596 pp. W. Reuter and S.D. Peyerimhoff, Chem. Phys., 160, 11, (1992). L.C. Robertson and J.A. Merritt, J. Mol. Spectrosc., 19, 372, (1966). L.C. Robertson and J.A. Merritt, J. Mol. Spectrosc., 24, 44, (1967). L.C. Robertson and J.A. Merritt, J. Chem. Phys., 57, 941, (1972). L.C. Robertson and J.A. Merritt, J. Chem. Phys., 56, 2919, (1972). M.B. Robin, K.B. Wiberg, G.B. Ellison, C.R. Brundle, and N.A. Kuebler, J. Chem. Phys., 57, 1758, (1972). C. Rodriguez-Garcia, O. Gonzalez-Blanco, A. Oliva, R.M. Ortuno, and V. Branchadell, European Journal of Inorganic Chemistry, 1073, (2000). C. Rodriguez-Garcia, A. Oliva, R.M. Ortuno, and V. Branchadell, J. Am. Chem. Soc., 123, 6157, (2001). J. Romelt, S.D. Peyerimhoff, and R.J. Buenker, Chem. Phys., 54, 147, (1981). B. Ruscic, J.E. Boggs, A. Burcat, A.G. Csaszar, J. Demaison, R. Janoschek, J.M.L. Martin, M.L. Morton, M.J. Rossi, J.F. Stanton, P.G. Szalay, P.R. Westmoreland, F. Zabel, and T. Berces, J. Phys. Chem. Ref. Data, 34, 573, (2005). 215 B. Ruscic, M. Litorja, and R.L. Asher, J. Phys. Chem. A, 103, 8625, (1999). L. Salem and C. Rowland, Angew. Chem. Int. Ed., 11, 92, (1972). W. Sander, Angew. Chem. Int. Ed., 29, 344, (1990). A. Sanov, T. Droz-Georget, M. Zyrianov, and H. Reisler, J. Chem. Phys., 106, 7013, (1997). A.A. Scala, E.W.G. Diau, Z.H. Kim, and A.H. Zewail, J. Chem. Phys., 108, 7933, (1998). E. Schafer, M. Winnewisser, and J.J. Christiansen, Chem. Phys. Lett., 81, 380, (1981). G. Schaftenaar and J.H. Noordik, J. Comput. Aided Mater. Des., 14, 123, (2000). R. Schinke, Photodissociation Dynamics : Spectroscopy and Fragmentation of Small Polyatomic Molecules. Cambridge Monographs on Atomic, Molecular, and Chemical Physics ; 1. 1993: Cambridge ; New York : Cambridge University Press. 417. E. Schmitz, Comprehensive Heterocyclic Chemistry, Small and Large Rings. Comprehensive Heterocyclic Chemistry ed. A.R. Katritzky and C.W. Rees. Vol. 7. 1984: Pergamon, Oxford. 994. E. Schmitz and R. Ohme, Chem. Ber. Recl., 94, 2166, (1961). E. Schmitz and R. Ohme, Tetrahedron Lett., 612, (1961). E. Schmitz and R. Ohme, Angew. Chem., 73, 115, (1961). M.S. Schuurman, S.R. Muir, W.D. Allen, and H.F. Schaefer, J. Chem. Phys., 120, 11586, (2004). R.L. Schwartz, G.E. Davico, T.M. Ramond, and W.C. Lineberger, J. Phys. Chem. A, 103, 8213, (1999). H. Sekino and R.J. Bartlett, Int. J. Quantum Chem. Symp., 18, 255, (1984). D.W. Setser and B.S. Rabinovitch, Can. J. Chem., 40, 1425, (1962). 216 Y. Shao, L.F. Molnar, Y. Jung, J. Kussmann, C. Ochsenfeld, S.T. Brown, A.T.B. Gilbert, L.V. Slipchenko, S.V. Levchenko, D.P. O'Neill, R.A. DiStasio, R.C. Lochan, T. Wang, G.J.O. Beran, N.A. Besley, J.M. Herbert, C.Y. Lin, T. Van Voorhis, S.H. Chien, A. Sodt, R.P. Steele, V.A. Rassolov, P.E. Maslen, P.P. Korambath, R.D. Adamson, B. Austin, J. Baker, E.F.C. Byrd, H. Dachsel, R.J. Doerksen, A. Dreuw, B.D. Dunietz, A.D. Dutoi, T.R. Furlani, S.R. Gwaltney, A. Heyden, S. Hirata, C.P. Hsu, G. Kedziora, R.Z. Khalliulin, P. Klunzinger, A.M. Lee, M.S. Lee, W. Liang, I. Lotan, N. Nair, B. Peters, E.I. Proynov, P.A. Pieniazek, Y.M. Rhee, J. Ritchie, E. Rosta, C.D. Sherrill, A.C. Simmonett, J.E. Subotnik, H.L. Woodcock, W. Zhang, A.T. Bell, A.K. Chakraborty, D.M. Chipman, F.J. Keil, A. Warshel, W.J. Hehre, H.F. Schaefer, J. Kong, A.I. Krylov, P.M.W. Gill, and M. Head-Gordon, Phys. Chem. Chem. Phys., 8, 3172, (2006). W.A. Shapley and G.B. Bacskay, J. Phys. Chem. A, 103, 6624, (1999). J. Sheridan, Proc. Intern. Meeting Mol. Spectry., 4th, 139, (1962). D. Sinha, S. Mukhopadhyay, and D. Mukherjee, Chem. Phys. Lett., 129, 369, (1986). D. Sinha, S.K. Mukhopadhyay, R. Chaudhuri, and D. Mukherjee, Chem. Phys. Lett., 154, 544, (1989). R.E. Smalley, L. Wharton, and D.H. Levy, Acc. Chem. Res., 10, 139, (1977). J.F. Stanton and R.J. Bartlett, J. Chem. Phys., 98, 7029, (1993). J.F. Stanton and J. Gauss, J. Chem. Phys., 101, 8938, (1994). J.F. Stanton, J. Gauss, J.D. Watts, W.J. Lauderdale, and R.J. Bartlett, Aces Ii, 1993. The Package Also Contains Modified Versions of the Molecule Gaussian Integral Program of J. Almlöf and P. R. Taylor, the Abacus Integral Derivative Program Written by T. U. Helgaker, H. J. Aa. Jensen, P. Jørgensen, and T. P. Taylor, and the Props Property Evaluation Integral Code of P. R. Taylor. P.G. Stecher, ed. The Merck Index; an Encyclopedia of Chemicals and Drugs. 8th ed. 1968, Rahway, New Jersey: Merck & Co., Inc. 1713 pp. . M.D. Su, H.L. Liao, W.S. Chung, and S.Y. Chu, J. Org. Chem., 64, 6710, (1999). A.G. Suits and R.E. Continetti, eds. Imaging in Chemical Dynamics. 2001, American Chemical Society: Washington, DC. 412 pp. R. Sumathi and M.T. Nguyen, J. Phys. Chem. A, 102, 8013, (1998). 217 Y. Tanaka, M. Kawasaki, Y. Matsumi, H. Fujiwara, T. Ishiwata, L.J. Rogers, R.N. Dixon, and M.N.R. Ashfold, J. Chem. Phys., 109, 1315, (1998). R. Tarroni, P. Palmieri, A. Mitrushenkov, P. Tosi, and D. Bassi, J. Chem. Phys., 106, 10265, (1997). J. Thiele, Ber. Dtsch. Chem. Ges., 44, 2522, (1911). C. Thomson and C. Glidewell, J. Comput. Chem., 4, 1, (1983). P.J.H. Tjossem and K.C. Smyth, Chem. Phys. Lett., 144, 51, (1988). D. Townsend, M.P. Minitti, and A.G. Suits, Rev. Sci. Instrum., 74, 2530, (2003). E.F. van Dishoeck, R.A. Bearda, and M.C. van Hemert, Astron. Astrophys., 307, 645, (1996). G.J. Vazquez, J.M. Amero, H.P. Liebermann, R.J. Buenker, and H. Lefebvre-Brion, J. Chem. Phys., 126, (2007). U.P. Verma, K. Moller, J. Vogt, M. Winnewisser, and J.J. Christiansen, Can. J. Phys., 63, 1173, (1985). J. Vogt and M. Winnewisser, Z. Naturforsch., A: Phys. Sci., 38, 1138, (1983). J. Vogt, M. Winnewisser, and J.J. Christiansen, J. Mol. Spectrosc., 103, 95, (1984). J. Vogt, M. Winnewisser, K. Yamada, and G. Winnewisser, Chem. Phys., 83, 309, (1984). H. von Pecmann, Ber. Dtsch. Chem. Ges., 27, 1888, (1894). H. von Pecmann, Ber. Dtsch. Chem. Ges., 28, 855, (1895). D.D. Wagman, W.H. Evans, V.B. Parker, R.H. Schumm, I. Halow, S.M. Bailey, K.L. Churney, and R.L. Nuttall, J. Phys. Chem. Ref. Data, 11, 1, (1982). S.P. Walch, J. Chem. Phys., 103, 4930, (1995). S.P. Walch, R.J. Duchovic, and C.M. Rohlfing, J. Chem. Phys., 90, 3230, (1989). S.P.G. Walch, William A., III., J. Am. Chem. Soc., 97, 5319, (1975). Y.M. Wang, L.P. Li, and W.A. Chupka, Chem. Phys. Lett., 185, 478, (1991). Y.M. Wang, L.P. Li, and W.A. Chupka, Chem. Phys. Lett., 192, 348, (1992). 218 J.D. Watts, J. Gauss, and R.J. Bartlett, J. Chem. Phys., 98, 8718, (1993). P.G. Wenthold, J. Hu, R.R. Squires, and W.C. Lineberger, J. Am. Chem. Soc., 118, 475, (1996). P.G. Wenthold and R.R. Squires, Int. J. Mass spectrom., 175, 215, (1998). P.G. Wenthold, R.R. Squires, and W.C. Lineberger, J. Am. Chem. Soc., 120, 5279, (1998). A.D. Wesley, Study of Radicals, Clusters and Transition State Species by Anion Photoelectron Spectroscopy, Chemistry, University of California Berkley, Berkley, 1994. C.M. Western, PGOPHER, A Program for Simulating Rotational Structure, University of Bristol, http://pgopher.chm.bris.ac.uk B.J. Whitaker, ed. Imaging in Molecular Dynamics. Technology and Applications. 2003, Cambridge University Press. 249 pp. S. Willitsch and F. Merkt, J. Chem. Phys., 118, 2235, (2003). S. Wilsey, K.N. Houk, and A.H. Zewail, J. Am. Chem. Soc., 121, 5772, (1999). R.J. Wolf and W.L. Hase, J. Phys. Chem., 82, 1850, (1978). E. Wrede, S. Laubach, S. Schulenburg, A. Brown, E.R. Wouters, A.J. Orr-Ewing, and M.N.R. Ashfold, J. Chem. Phys., 114, 2629, (2001). Y. Yamaguchi and H.F. Schaefer, J. Chem. Phys., 106, 1819, (1997). Y. Yamaguchi and H.F. Schaefer, J. Chem. Phys., 106, 8753, (1997). Y. Yamaguchi, C.D. Sherrill, and H.F. Schaefer, J. Phys. Chem., 100, 7911, (1996). N. Yamamoto, F. Bernardi, A. Bottoni, M. Olivucci, M.A. Robb, and S. Wilsey, J. Am. Chem. Soc., 116, 2064, (1994). D.R. Yarkony, J. Chem. Phys., 104, 2932, (1996). R.N. Zare, Mol. Photochem., 4, 1, (1972). R.N. Zare, ed. Angular Momentum : Understanding Spatial Aspects in Chemistry and Physics. (the George Fisher Baker Non-Resident Lectureship in Chemistry at Cornell University). 1988, New York : Wiley. 349. 219 H. Zollinger, Azo and Diazo Chemistry; Aliphatic and Aromatic Compounds. 1961: Interscience Publishers, New York. 444 pp. M. Zyrianov, Photoinitiated Decomposition of HNCO, Chemistry, University of Southern California Los Angeles, 1998. 220 Appendix A. Technical Drawings of the Pulsed Pyrolysis Source The technical drawings for the pulsed pyrolysis source are given below. Figure A.1 Electrode. The dimensions are in mm [inches]. Figure A.2 Graphite split disk. The dimensions are in mm [inches]. 221 Figure A.3 SiC-tube holder. The dimensions are in mm [inches]. 222 Figure A.4 Water-cooled block. The dimensions are in mm [inches]. 223 Figure A.4 Continued. 224 Figure A.5 Sample inlet chamber. The dimensions are in mm [inches]. 225 Figure A.6 Face plate of the sample inlet chamber. The dimensions are in mm [inches]. 226 Figure A.7 Main chamber. The dimensions are in mm [inches]. 227 Figure A.7 Continued. 228 Figure A.8 Face plate of the main chamber. The dimensions are in mm [inches]. 229 Figure A.9 Piezoelectric disk translator. The dimensions are in mm [inches]. 230 The piezoelectric disk translator is mounted to the main chamber on four ultem cylinder. Figure A.10 Ultem cylinder. The dimensions are in mm [inches]. 231 Figure A.11 Plunger. The dimensions are in mm [inches]. 232 Figure A.12 Fasteners. The dimensions are in mm [inches]. The plunger is mounted to piezoelectric disk translator by fasteners. 233 Figure A.12 Continued. 234 Figure A.13 ASA flange. The dimensions are in mm [inches]. 235 APPENDIX B. Deperturbation of Energy Levels of the 9 1 0 Transition to the 2 1 A 2 (3p y ) State and the Band Origin of the 2 1 B 1 (3p z ) Transition Using Two-Level Approximation The wave functions and of the perturbed states are represented as linear combinations of the wave functions ! " 1 0 0 0 (a 1 ), 1 B 1 (3p z ) { } and ! " 2 9 0 1 (b 2 ), 1 A 2 (3p y ) { } of the unperturbed states: ! " # =#$% 1 0 0 0 (a 1 ), 1 B 1 (3p z ) { } +&% 2 9 0 1 (b 2 ), 1 A 2 (3p y ) { } , B1 ! " + =#$ 1 0 0 0 (a 1 ), 1 B 1 (3p z ) { } +%$ 2 9 0 1 (b 2 ), 1 A 2 (3p y ) { } , B2 where and are numerical coefficients. and give the probability of finding the states described by and in the ! " 1 0 0 0 (a 1 ), 1 B 1 (3p z ) { } and ! " 2 9 0 1 (b 2 ), 1 A 2 (3p y ) { } components. The band intensity, , is proportional to the square of the matrix element between the initial and final wave functions: B3 where is the electronic transition dipole moment from the initial to the final vibronic state, and and are the nuclear wave functions of the initial and final states, respectively. Thus, the intensities of transitions from the state to and 1 states of the cation are: 236 , B4 , B5 respectively. and are the electronic transition dipole moments from the 2 1 A 2 (3p y ! !) and 2 1 B 1 (3p z ! !) Rydberg states to the 1 2 B 1 (" ! !) cation state, and is the nuclear wave function of the cation. Likewise, the intensities of transitions from the state to " = 0 and 1 states of the cation are: , B6 , B7 respectively. The similarity in the geometries (Figure 3.2) and electronic core structures of the 2 1 A 2 (3p y ! !) and 2 1 B 1 (3p z ! !) Rydberg states and the 1 2 B 1 (" ! !) cation state (Table 3.2) result in a strong propensity for ionization via diagonal (#" = 0) transitions. In addition, the quantum defects of the 3p z and 3p y electronic states are similar and typical of a Rydberg states. As no state-specific selection rules are expected in the ionization from these states to the cation continuum, we assume that the electronic transition dipole moments for the two transitions are very similar. 237 Therefore, from the eKE distributions at # = 380.75 nm (52,528 cm -1 ) and assuming that we obtain: , B8 and from the eKE distributions at # = 379.50 nm (52,700 cm -1 ): . B9 giving an average value of = 1.82. Using , we obtain and . In order to find the unperturbed energy levels the equations below have been solved for E - = 52,528 cm -1 (9 1 0 ,2 1 A 2 (3p y ! !)); E + = 52,700 cm -1 (0 0 0 ,2 1 B 1 (3p z ! !)): , B10 , B11 . B12 The solution is: E 1 = 52,638 cm -1 ; E 2 = 52,590 cm -1 ; V 12 ~ 83 cm -1 .

Abstract (if available)

Abstract The photophysics and photochemistry of two isomers of CH(subscript 2)N(subscript 2), diazomethane and diazirine, was studied in molecular beam. Their pyrolysis was used to produce molecular beam of methylene radicals in the ground electronic state. The spectroscopy, photoionization, and photodissociation pathways of diazomethane and diazirine were investigated experimentally using a combination of 2 + 1 REMPI and velocity map imaging techniques. For three isotopologs of diazomethane, the 2(superscript 1)A(subscript 2)(3p(subscript y) ← π) and 2(superscript 1)B(subscript 1)(3p(subscript z) ← π) Rydberg states are of mostly pure Rydberg character, whereas the 3(superscript 1)A(subscript 1)(3p(subscript x) ← π) state is mixed with the valence 2(superscript 1)A(subscript 1)(π*← π) state. Normal mode vibrational frequencies were obtained for the 2(superscript 1)A(subscript 2)(3p(subscript y)) Rydberg state and cation, and mixed levels of the 2(superscript 1)A(subscript 2)(3p(subscript y)) and 2(superscript 1)B(subscript 1)(3p(subscript z)) states of the three isotopologs were identified. The agreement between experiment and theory was very good allowing a full analysis of trends in structure and vibrational frequencies going from the neutral species to the excited Rydberg state, and the cation. In the experiments with diazirine, the weak one-photon absorption to the 1(superscript 1)B(subscript 2) state is immediately followed by more efficient absorption of another photon to reach the 1(superscript 1)A(subscript 2) state from which competition between ionization and fast dissociation takes place. Strong signals of CH+ ions are also detected and assigned to 2 + 1 REMPI of CH fragments. Velocity map CH(superscript +) images show that CH (X, v"=0, N") fragments are born with substantial translational energy indicating that they arise from absorption of two photons in diazirine.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Photodissociation dynamics of atmospherically relevant small molecules in molecular beams

PDF

Spectroscopy and photodissociation dynamics of hydroxyethyl radicals

PDF

Dissociation energy and dynamics of HCl and aromatic-water clusters

PDF

Photodissociation dynamics of group V hydrides

PDF

Dissociation energy and dynamics of water clusters

PDF

A multifaceted investigation of tris(2-phenylpyridine)iridium

PDF

Unraveling photodissociation pathways in pyruvic acid and the role of methylhydroxycarbene

PDF

Infrared spectroscopy of carbocations in helium droplets

PDF

Temperature-dependent photoionization and electron pairing in metal nanoclusters

PDF

Development and implementation of methods for sliced velocity map imaging. Studies of overtone-induced dissociation and isomerization dynamics of hydroxymethyl radical (CH₂OH and CD₂OH)

PDF

A technique for assessing water cluster sizes by pickup momentum transfer and a new source with loading system for lithium clusters

$Study of rotation and phase separation in ³He, ⁴He, and mixed ³He/⁴He droplets by X-ray diffraction$

PDF

Study of rotation and phase separation in ³He, ⁴He, and mixed ³He/⁴He droplets by X-ray diffraction

Asset Metadata

Creator Fedorov, Igor (author)

Core Title Photoelectron and ion imaging investigations of spectroscopy, photoionization, and photodissociation dynamics of diazomethane and diazirine

School College of Letters, Arts and Sciences

Degree Doctor of Pharmacy / Doctor of Philosophy

Degree Program Chemistry (Chemical Physics)

Publication Date 07/02/2009

Defense Date 06/05/2009

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

Tag diazirine,diazomethane,imaging,methylene,OAI-PMH Harvest,photodissociation dynamics,photoionization,pyrolysis,resonance enhanced multiphoton ionization,spectroscopy

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Reisler, Hanna (committee chair), Gruntman, Michael A. (committee member), Krylov, Anna I. (committee member)

Creator Email fedorov.rus@gmail.com,fedorov@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m2322

Unique identifier UC1448929

Identifier etd-Fedorov-3025 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-571777 (legacy record id),usctheses-m2322 (legacy record id)

Legacy Identifier etd-Fedorov-3025.pdf

Dmrecord 571777

Document Type Dissertation

Rights Fedorov, Igor

Type texts

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

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu