UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Negative ion photoelectron spectroscopy of nineteen electron organometallic complexes Martel, Arthur Andrew 1995

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1995-0164.pdf [ 2.53MB ]
JSON: 831-1.0061744.json
JSON-LD: 831-1.0061744-ld.json
RDF/XML (Pretty): 831-1.0061744-rdf.xml
RDF/JSON: 831-1.0061744-rdf.json
Turtle: 831-1.0061744-turtle.txt
N-Triples: 831-1.0061744-rdf-ntriples.txt
Original Record: 831-1.0061744-source.json
Full Text

Full Text

N E G A T I V E ION P H O T O E L E C T R O N SPECTROSCOPY OF N I N E T E E N ELECTRON ORGANOMETALLIC COMPLEXES by ARTHUR ANDREW M A R T E L B.Sc, The University of Adelaide, 1989 B.Sc.(Hons)., The University of Adelaide, 1990  A THESIS SUBMITTED IN PARTIAL F U L F I L L M E N T OF T H E REQUIREMENTS FOR T H E D E G R E E OF M A S T E R OF SCIENCE in T H E F A C U L T Y OF G R A D U A T E STUDIES (Department of Chemistry) We accept this thesis as conforming to the required standard  T H E UNIVERSITY OF BRITISH C O L U M B I A April 1995 © Arthur Andrew Martel  In  presenting  degree freely  at  this  the  available  copying  of  publication  or of  in  partial  fulfilment  of  the  University  of  British  Columbia,  I  agree  for  this  department  thesis  reference  thesis by  this  for  his  and. s t u d y . scholarly  or  thesis  for  her  Department  of  T h e U n i v e r s i t y o f British Vancouver, Canada  Date  DE-6  (2/88)  vb  Apn'f  Columbia  iqqs  further  purposes  gain  shall  that  agree  may  representatives.  financial  permission.  I  requirements  It not  be is  that  the  Library  an  granted  by  allowed  advanced  shall  permission  understood be  for  the that  without  for head  make  it  extensive of  my  copying  or  my  written  ABSTRACT  The photoelectron spectra of the organometallic complexes CpCo(CO)2, MeCpMn(CO)3~ and Co2(CO) (NO) ", where Cp is the cyclopentadienyl ligand (C5H5) 4  2  and MeCp the singly methylated analogue (CH3-C5H4), have been obtained using Negative Ion Photoelectron Spectroscopy (NTPES) techniques. The experiments on CpCo(CO)2~ and MeCpMn(CO)3~ were performed at two separate laser photon wavelengths, 355 and 532 run. The cobalt dimer complex was examined only at 355 nm. Minimum vertical detachment energies for the three complexes were found to be 0.9 ± 0.1, 0.5 ± 0.1 and 2.0 ± 0.2 eV, respectively. The bond dissociation energy, D((CO)3MnCpCH ~-H), of one of the methyl hydrogens in the 2  MeCpMn(CO) " anion was found to be 2.2 ± 0.9 eV. 3  ii  TABLE OF CONTENTS Abstract  ii  Table of Contents  iii  List of Tables  v  i List of Figures  vi  Acknowledgment  vii  Chapter 1  Introduction  1  1.1  General Introduction  1  1.2  Selection Rules  6  1.3  Spectroscopic Data obtained using Negative  1.4 Chapter 2  Ion Photoelectron Spectroscopy  9  1.3.1  Electron Affinity Measurements  9  1.3.2  Vibrational Structure  9  1.3.3  Excited Electronic States  10  Aim of the Experiments  11  Experimental  13  2.1  General Description  13  2.2  Negative Ion Formation  15  2.3  Mass Selection of Ions  17  2.4  Laser Photodetachment of the Negative Ions  21  2.5  Data Collection and Timing  24  iii  Chapter 3  Results  26  3.1  Resolution of the Mass Spectrometer  26  3.2  Resolution of the Electron Spectrometer  29  3.3  Calibration of the Electron Spectrometer  33  3.4  Calibration at 355 nm. The Photoelectron Spectrum of Q f  34  3.5  Calibration at 532 nm. The Photoelectron Spectrum of NO"  3.6  3.7  3.8  Chapter 4  Chapter 5  38  Photoelectron Measurements on CpCo(CO)2  43  -  3.6.1  Measurement on CpCo(CO) ~ in He at 355 nm  43  3.6.2  Measurement on CpCo(CO) ~ in Ar at 355 nm  43  3.6.3  Measurement on CpCo(CO) " in He at 532 nm  44  2  2  2  Photoelectron Measurements on MeCpMn(CO)3  _  50  3.7.1  Measurement on MeCpMn(CO)3~ at 355 nm  50  3.7.2  Measurement on MeCpMn(CO) " at 532 nm  50  3  Photoelectron Measurements on Co2(CO) (NO) "  55  3.8.1  55  4  2  Measurement on Co (CO) (NO) " at 355 nm 2  4  2  Discussion  58  4.1  Cyclopentadienyl cobaltdicarbonyl  58  4.2  Methylcyclopentadienyl manganesetricarbonyl  64  4.3  Cobalt dicarbonylnitrosyl Dimer  68  Conclusion  70  References  71  iv  LIST OF TABLES  Table 1  Magnetic Field Strengths in the Electron Drift Tube  23  Table 2  Mass Resolution for Oxygen Negative Ions  Table 3  0 Calibration of the Electron Spectrometer  37  Table 4  NO Calibration of the Electron Spectrometer  42  Table 5  Vertical Detachment Energy Measurements on CpCo(CO)2~  49  Table 6  Vertical Detachment Energy Measurements on MeCpMn(CO)3  54  Table 7  Vertical Detachment Energy Measurement on Co (CO) (NO)2~  55  Table 8  CO and Ring Stretching Frequencies of CpCo(CO)  58  Table 9  CO Stretching Frequencies of Manganese Complexes with Cp and MeCp ligands  65  29  2  _  2  4  2  v  LIST OF FIGURES Figure 1  Schematic Diagram of how Negative Ion Photoelectron Spectra are Obtained  7  Figure 2  Schematic Diagram of the Spectrometer  14  Figure 3  Beam Modulation Technique  19  Figure 4  Electron Drift Tube  23  Figure 5  Mass Spectrum of Oxygen Species  28  Figure 6  Photoelectron Spectrum of 0~ at 355 nm  32  Figure 7  Photoelectron Spectrum of 0 ~  35  Figure 8  Mass Spectrum of Anions from N 0  40  Figure 9  Photoelectron Spectrum of NO"  41  Figure 10  Mass Spectrum of CpCo(CO) ~  45  Figure 11  Photoelectron Spectrum of CpCo(CO) ~ in He at 355 nm  46  Figure 12  Photoelectron Spectrum of CpCo(CO) ~ in Ar at 355 nm  47  Figure 13  Photoelectron Spectrum of CpCo(CO) ~ in He at 532 nm  48  Figure 14  Mass Spectrum of MeCpMn(CO) "  51  Figure 15  Photoelectron Spectrum of MeCpMn(CO)3~ at 355 nm  52  Figure 16  Photoelectron Spectrum of MeCpMn(CO)3~ at 532 nm  53  Figure 17  Mass Spectrum of Co (CO) (NO) ~  56  Figure 18  Photoelectron Spectrum of Co (CO) (NO) ~ at 355 nm  57  Figure 19  Fenske-Hall Molecular Orbital Diagram for CpCo(CO)  60  Figure 20  Thermodynamic Cycle involving MeCpMn(CO)3~  2  2  2  2  2  2  3  2  4  2  2  vi  4  2  2  67  ACKNOWLEDGMENTS I would like to express my gratitude for services rendered by the support staff of the Chemistry department. The electronic, mechanical and glass-blowing workshops, as well as the staff of the first year teaching labs. To the members, past and present, of my group; Nick, Jenn, Shu-Ping and Dr. Waller, I extend my deep appreciation for all that you have done. Finally, although I feel at times that I have actually lived out the 'Rime of the Ancient Mariner' during my stay here, many people have helped to make the journey worthwhile. So to Nick, Mel, Conine, Bruce, Lui, Lisa (both of you), Angelo and Jamie, thank-you.  For A C , F.V., and D.  5  vii  CHAPTER 1 INTRODUCTION  1.1 General Introduction  Transition metal centred complexes play an important role in many redox reactions, chain mechanisms, homolytic bond cleavages and in the catalysis of C - C bond formation, as well as acting as catalysts in industrial processes, such as the Wacker reaction . 1  Most isolable organometallic complex reactions conform to the 18 electron rule . 2  This rule broadly states that organometallic complexes with a total of 18 valence electrons will be thermodynamically stable, the number of valence electrons being the sum of the metal d electrons plus the electrons conventionally regarded as being supplied by the ligands. Complexes of this type often undergo ligand transfer reactions via dissociative mechanisms. Yet, while reaction pathways involving 16/18 valence electron species are generally preferred, many examples of reactions involving 15/17 valence electron intermediates have been reported " . 3  7  The first reported example of a 19 electron complex , and hence the possibility that 8  some reactions may proceed via a 17/19 valence electron pathway, was cobaltocene,  CP2C0. It has been proposed that the reaction pathways of some metal-centred organometallic radicals are governed by a 17/19 electron rule in much the same way as the 16/18 electron rule governs the reactivity of closed shell organometallic complexes . 9  The definition of a 19 electron complex is one that has 19 valence electrons with a predominantly metal character for the 19th electron in the highest occupied molecular orbital (HOMO). However, many complexes that atfirstappear to qualify as 19 electron species may, in fact, not be. Astruc therefore suggests that nineteen electron complexes 10  may be classified as belonging to one of the following six groups: (1)  Metal-centred radicals that have 19 valence electrons (the HOMO has a high metal character).  (2)  Species with a high degree of covalency of the HOMO level. ("18.5" valence electron species such as Ni sandwiches.)  (3)  37 electron average valence bi-metallic complexes, where each metal centre has 18.5 valence electrons.  (4)  Metal-carbonyl species resulting from one electron reduction of 18 electron complexes (the 19th resides in an a* or  (5)  K*  orbital, for example  Fe(CO)5*~).  Base adducts of 17 electron complexes where the 19th electron resides in a o* orbital such as [CpFe(CH )(CO)(PPh )(py)] , where P is the phosphine ligand and +  3  3  py is pyridine. (6)  18 electron complexes with a monoreduced ligand (the radical centre resides on the ligand, such as Mo(CO) (bpy)~), where bpy is bipyridine. 4  So there is some ambiguity about the naming of some compounds as 19 electron complexes, as the. extra electron might be primarily ligand based, ("19 electron" implies  2  that the extra electron is in a predominantly metal orbital). Many 19 electron complexes couple via the ligands, suggesting that the extra electron is indeed ligand based. However, Astruc has shown that the location of the spin density in the transition state of a coupling reaction can be very different from that of a ground state, therefore spectroscopic measurements are needed to further elucidate electronic structures, as determinations based solely on the analysis of reactant products can be misleading  11  Indirect evidence suggests that some 19 electron adducts can avoid the occupation of a high energy Metal-Ligand orbital by changing the geometry of a ligand, i.e. the linear Metal-N-0 group can bend, or flat Cp rings can slip into bent configurations . Evidence 12  for this is primarily from reactivity studies, the tendency for molecules to undergo ligand substitution via either associative or dissociative mechanisms may also help indicate how the 'extra' electron is accommodated within the complex. It has been recommended that 19 electron complexes be referred to as 18 + 8 complexes , where 6 is the fraction of unpaired electronic charge that is delocalised on 10  the metal (i.e. an authentic 19 electron compound has 6 = 0.5, or if the electron was located totally on a ligand, 8 = 0). The value of 8 is difficult to determine, although ESR studies have reported 8 = 0.015 for Co(CO)3L,2 where L is 2,32  bis(diphenylphosphino)maleic anhydride . 13  Studies using Fourier Transfer Ion Cyclotron Resonance (FT-ICR) techniques have been used to determine bond dissociation energies and gas phase acidities of compounds of this nature. The gaseous sample is ionized in the centre of a trapped ion analyzer cell by electrons which are accelerated through the cell. Ions that form are then  3  trapped within the cell for approximately one second; this is done using a shallow electric well. The ions are then accelerated by a radio-frequency signal that increases linearly with time. After this sweep is completed, the current produced by the various ion packets is collected. The time-domain signal is then transformed into the mass-domain, using a Fourier transform. With the addition of reagent gases into the cell, ion molecule reactions may be studied . 14  Infra-red (IR) spectroscopic analysis can offer insight into the electronic nature of this type of molecule by monitoring the effects the 19th electron has on the stretching frequencies of substituent ligand, such as the CO, N O or cyclopentadienyl groups. Transient IR spectroscopy can monitor the products formed by flash photolysis of organometallic molecules . The parent molecule is photolysed by ultra-violet radiation 15  produced in the flash. A pulse from a continuum source is initiated and passes down the cell to a spectrometer so the products may be monitored. The process of photolysis flash followed by source flash is repeated many times to build up a spectrum. Gaseous reactant species can be introduced into the cell and unstable species formed with short life times, may be investigated . 16  Many transition metal organometallic molecules contain an uneven number of electrons, and as such, they may be studied using Electron Spin Resonance (ESR) techniques. The spin of an unpaired electron can couple with the spin of the nuclei in these species to give splitting analogous to those observed for nuclear spin-spin coupling. When the electron interacts with n equivalent nuclei, its resonance peak is split into (2«I +1)  4  peaks, where I is the spin quantum number of the nuclei. This is known as hyperfine splitting . 14  Microwave radiation, usually generated by a Klystron tube, is transmitted to a sample held in a quartz tube between the poles of a permanent magnet. A Helmholtz coil is used to vary the field over the small range where resonance occurs. ESR studies can be applied to chemical, photochemical and electrochemical reactions that proceed via a free radical as well as to obtain structural information about transition metals and their complexes. Flowing Afterglow methods may also be employed to study molecules of this type . This technique involves forming negative ions from neutral precursor complexes by 17  electron impact ionization. Reagent gases are then introduced to these ions as they travel downstream in an inert carrier gas. The products of any reactions are then monitored using a quadrupole mass spectrometer. By using Negative Ion Photoelectron Spectroscopy (NTPES) techniques we hope to utilize the unique advantages of the method to further increase the data available to assist in the determination of the electronic properties of organometallic compounds having '18 + 8' valence electrons. The basis of these experiments consists of the photodetachment of an electron from the negative ion analogues of the molecules of interest. From a sample of a parent compound, a variety of negatively charged anions are formed, and these are then separated by a time-of-flight mass spectrometer. This separation serves two purposes: the species formed tend to be very reactive and so it is desirable to keep them away from possible  5  reactants; and secondly, by unambiguously selecting which ion is irradiated, the assignment of the resulting electron energy spectrum is simplified as overlap from two or more different molecules is eliminated. Figure 1 shows schematically how the spectra are obtained. The transition from the negative ion to the neutral is induced using afixed-frequencylaser. The difference between the photon energy of the output from this laser and the kinetic energy of the ejected electron is equal to the electron binding energy of the neutral species. It is by analysing the kinetic energy of these ejected electrons that information about states in the neutral molecule of interest may be gathered.  1.2 Selection Rules  The selection rules that govern NIPES stem from the dipole approximation; Pabs = [ I M>f* li Vi dx ] , 2  where P b represents the transition probability, u. is the dipole operator and a  S  , vj/f are the  initial and final states, respectively, of the molecule. In order for a transition to occur, the condition P bs * 0 must be satisfied. Therefore, by evaluating Equation 1.1 below, the a  selection rules for. a transition may be determined;  I i|/* f  1.1  [i \)/i dx = 0.  Taking advantage of the Born-Oppenheimer approximation, the wavefunction, \\i, can be separated into its electronic and nuclear components;  W = NWnu-  6  Photoelectron Spectrum  laser photon]  v-3  e" kinetic energy. v-0  Electron Affinity e* counts  AB"  AB  Figure 1. Negative Ion Photoelectron Spectroscopy. Schematic diagram of how NTPES spectra are obtained.  7  Equation 1.1 then becomes; I V|/f,el*V|/f, * n u  => 1  Vf,el*  P-  U , \ | / i , e i V | / y m dl  J  V|/i,el d X e l .  =  0,  V|/f, * V | / n u  i ; n u  dx = 0.  1.2  nu  The electronic component of 1.2 provides the selection rules for electronic transitions. If vj/i,ei  is an even function, then \ | / f ,  e  must be an odd function for a transition to occur (as p.  i*  is an odd function). Similarly, if \\i  \  ue  i|/f,ei*  is an odd function then the selection rules dictate that  needs to be even. In the case of negative ion photoelectron spectrometry however, Vf,el  —  U'continuum-  The final electronic wavefunction corresponds to the continuum wavefunction, and so the selection rule conditions are satisfied. That is, the integral / v|/ i* \i f;e  v|/i, i e  dx is non-zero.  The second part of Equation 1.2, the nuclear motion component, contains the Franck-Condon overlap of the vibrational wave-functions, which determines the intensities of the transitions. The ground state vibrational level of the anion is always symmetric, and for totally symmetric modes in the neutral molecule all of the vibrational levels are also symmetric. This usually results in substantial Franck-Condon overlap between these levels, if the change in geometry between the anion and the neutral molecules is not too large. Reasonable overlap is also usually obtained between the even levels in anti-symmetric modes of the neutral and the ground state of the anion, but there is little Franck-Condon overlap between that state and the odd levels of anti-symmetric neutral molecule modes.  8  1.3 Spectroscopic Data Obtained using NIPES  1.3.1 Electron Affinity Measurements. The electron affinity of an atom or molecule is defined as the minimum energy required to remove an electron from a negative ion in the ground electronic, vibrational and rotational state and produce the corresponding neutral species. A"(v = 0, j = 0) + hv -> A (v = 0, j = 0) + e" . The AE of the above reaction is equal to the electron affinity of A. Trends in electron affinity can indicate the relative stability of the negative ions. They can also be indicators of such factors as electron correlation, the extent of electron delocalisation, n back-bonding, the degree of conjugation and ligand polarisability . 18  Molecular electron affinities of iso-electronic species generally reflect the electron attaching power of the central atom in the molecule, an increase in electron affinity may also indicate increases in conjugation in a certain molecule.  1.3.2 Vibrational Spectra. The photoelectron detachment of an electron from a negative ion is a vertical process and therefore is governed by the Franck-Condon principle. Geometrical rearrangements between the negative ion and neutral structures should produce extensive vibrational progressions. For example, if a ligand such as cyclopentadienyl (CsH )"that is originally afiveelectron donating species, r| , is forced to 5  5  change its hapticity to r| to accommodate the extra electron in the negative ion, then a 3  9  sizable rearrangement would be expected upon electron detachment. If the geometry of one is known, then possible geometries of the other may be inferred. In organometallic complexes, coordinated carbonyl and nitrosyl ligand vibrational frequencies are sensitive to their immediate environment. Shifts in these frequencies may indicate differences in the back-bonding from neighboring ligands, and so the relative back-bonding abilities of various ligand species may be compared. Carbonyl vibrational stretches are typically in the range between 1800 and 2100 cm" , while the nitrosyl stretch 1  is usually seen between 1700 and 1900 cm" . They are therefore both able to be resolved 1  by our spectrometer. The observation of'hot' bands, transitions from states other than v = 0, j = 0, can also provide spectroscopic information about the species of interest.  1.3.3 Excited Electronic States. It is possible to access excited electronic states of transition metal complexes using negative ion photoelectron spectroscopy. This may then assist in providing information on their geometries and vibrational structures. The photoelectron spectrum of FeCO" has been reported and the vibrational structure of the excited Z~ state, as well as the ground 19  3  5  E ~ state, resolved.  10  1.4 Aim of the Experiments  Initial investigations by this group ' have centred on the molecule Co(CO)3NO 20 21  and its daughter complex Co(CO)2NO. This was the first instance where an organometallic complex containing a nitrosyl ligand had been studied using NTPES techniques. Most previous efforts centred around species that had solely carbonyl groups as ligands ' . The minimum vertical detachment energies of both these compounds were 22 26  reported ' , and possible structural differences between the neutral and anion forms were 20 21  discussed. The formation of the anion Co(CO)3NO~ meant the creation of a "nineteen electron" organometallic radical. It was proposed that the NO ligand altered its hapticity, from r| to r\ , in order to accommodate the extra electron and hence form a complex that 3  l  corresponds to a type (6) (see section 1.1). The current project was designed to further investigate the formation of nineteen electron organometallic radicals by utilizing ligands that show a tendency to be able to alter their hapticities in response to added electron density. In this case, the cyclopentadienyl ligand (C5H5, Cp) and methyl-cyclopentadienyl ligand  (C5H4-CH3,  MeCp) were used. The compounds involved were cyclopentadienyl cobaltdicarbonyl (CpCo(CO)2) and methylcyclopentadienyl manganesetricarbonyl (MeCpMn(CO)3). CpCo(CO) is related to Co(CO)3NO and it was hoped that a comparison of the minimum 2  vertical detachment energies of the two complexes would allow the relative propensity of the NO and Cp ligands to accommodate the added electron density to be determined.  ll  In addition, a cluster complex formed during the ionization of Co(CO)3NO was also investigated. This molecule [Co(CO)2NO]2~ seems to correspond to the 35 electron dimer type. Any effect due to electron delocalisation over the cluster may be observed, again, by comparison of the detachment energies of Co(CO)2NO and the dimer.  12  CHAPTER 2 EXPERIMENTAL 2.1 General Description  A series of experiments were performed on the three organometallic molecules Co(CO) Cp, Mn(CO) (MeCp) and (Co) (CO) (NO) . The first two of these were 2  3  2  4  2  purchased from Strem Chemicals (CAS #'s 12078-25-0 and 12108-13-3 respectively) in liquid form, while the third, a dimer, was formed in the spectrometer from the liquid precursor Co(CO)3NO (Strem Chemicals CAS # 14096-82-3). Spectroscopic information about the neutral organometallic molecules was obtained in the way described briefly below and in more detail in the rest of this chapter. The Negative Ion Photoelectron Spectrometer used in these experiments combines an ion formation capability with an electron detection region. Figure 2 shows a schematic diagram of the spectrometer's construction. A variety of negative ions may be formed from a neutral precursor which are then separated according to their relative masses by a time-of-flight mass spectrometer. Once mass selected, an electron from a particular ion of interest is selectively photodetached using the output from a fixed frequency laser and energy analyzed by way of a second time-of-flight spectrometer.  13  . electron |i-metal shielding  pulsed-molecular beam valve  acceleration plates gate valve  Figure 2. Schematic Diagram of the Negative Ion Photospectrometer.  14  2.2 Negative Ion Formation  The ion source region of the spectrometer takes advantage of the properties of supercooled gases to produce negative ions with low rotational and vibrational temperatures. The construction is based on the design by Johnson et al and consists of a 27  pulsed molecular beam that is crossed with an electron beam. The volatile compound of interest is placed in a glass tube. This tube may be heated either by a 50 ml Glas-Col heating mantle or by using heating tape in order to increase the vapour pressure. An inert carrier gas, usually helium although argon was also used, passes through the tube. With a typical backing pressure of 25 psig, a 10 % mixture of Co(CO)3NO in the backing gas is created. The mixture then enters the source chamber via a pulsed valve (General Valve Corporation). A 0.8 mm diameter nozzle is used, at a 30 Hz repetition rate. Inside the source chamber the operating pressures are of the order of 5 xlO Torr. -5  Negative ions are created by directing a 1 keV beam of electrons produced from an electron gun (Tektronix), into the continuum flow region of the free jet expansion. The fast moving electrons ionize the carrier gas producing slower moving electrons which may attach to the molecule of interest. As the relative density of the gas is high in the region close to the valve nozzle, third body collisions will help facilitate this attachment. It is found that the comparative abundance of different ions varied with the position of the electron beam relative to the nozzle, the greater density of molecules and increased  15  likelihood of collisions near the nozzle tip compared to further downstream, influencing the formation of certain ions over others. The collisions that occur in this region play an important part in the cooling of the ions in the molecular beam . Close to the nozzle exit the relative pressure of the gas is 28  high. In this region continual collisions will take place between the small amount of the molecular gas of experimental 'interest' and the larger amount of monoatomic carrier gas. These collisions allow the energy from the rotational and vibrational motion in the larger molecules to be transferred to the translationally cold bath of monoatomic gas, thereby cooling these molecules. As they travel downstream from the nozzle, differences in translational velocity (the component of velocity perpendicular to the direction of the molecular beam flow) will cause the beam to spread out and so collisions will become less frequent. Molecules occupying a small area downstream will necessarily have similar translational velocity components and so remain cold as temperature is independent of the velocity along the direction of the molecular flow. Using this method molecules may be produced at much lower temperatures, and still remain in the gas phase, than with usual refrigeration techniques. In effect, the molecular gas is rapidly cooled below it's condensation temperature and then allowed to expand until it's low density inhibits formation of clusters on which other molecules could condense. The practical limit is defined by the pumping capacity to handle the gas discharged by one pulse before the next in order to keep the background pressure low . 29  In our experiments this chamber was pumped out by a 2000 1/sec Edward's Diffstak diffusion pump. Measurements conducted by the Neumark group at U . C. Berkeley, using  16  a very similar source of ions found rotational temperatures for SH~ of 50-75 K and vibrational temperatures for IHT of 100 K ' . 30  31  The inside of the chamber is held at negative 1000 volts, this allows the ions in the small section of the molecular beam that passes out of the source chamber through a 5 mm aperture, to be accelerated to ground as they enter the mass selection region of the spectrometer.  2.3 Mass Selection of Ions  The resolution of the mass spectrometer utilized in this apparatus is independent of both the mass of the ions and the location where the ions are initially formed, and is proportional to the square of the length of the drift chamber. It is based on a Beam Modulation technique developed by Bakker  32,33  .  As the ions leave the source chamber they pass through an acceleration region. A stack of 10 circular electrode plates connected by a resistor chain raise the voltage from -1 kV to ground. The monoenergetic (but not homogeneous in mass) ion pulse then travels into the mass spectrometer proper. The mass spectrometer consists of a pair of Beam Modulation Plates (BMP) and a defining slit at the end of a 1 metre drift tube. The ions are influenced by the electric field between the plates, which are separated by 1 cm and whose polarities may be switched. One plate is held at V/2 while the other is alternated between 0 and V (typically V is between 25 and 50 volts with a 10 ns fall time). Ions that pass through the plates before  17  the electric field changes will experience an overall downward force, those that pass through after the field has changed will experience an upward force. The motion of ions that are inside the field region when the polarity is switched will be dependent on the exact position when the switch took place. Ions that happen to be exactly halfway along the plates at the time of the switch, T, will experience a downward force for 1/2 T and an equal but upward force also of 1/2 T. There will be no net change in the velocity of these ions, and they will continue to move parallel to their initial direction. However, they will be displaced slightly relative to their position when entering the plates. Initially the ions have no up or down component of velocity, yet at time T when they experience the electric field change, the ions have acquired a downward velocity component. So they will emerge from the plates slightly below the position that they entered. Only a small slice of the ion pulse remains undeflected enough to pass through the 5 mm defining slit at the end of the drift tube. The net result is shown in Figure 3. The timing of the B M P voltage switch may be controlled to maximize the number of ions that exit the mass spectrometer. Three focusing components have been added along the ion flight path in the drift tube in order to increase the proportion of the ion bunch that reaches the detector. After exiting the beam modulation plates, the ions pass through an Einzel lens, a three element lens consisting of three concentric stainless steel pieces of tubing 55 mm in diameter, separated from each other by two 5.5 mm thick rings of Delrin (a polyacetal resin). The voltage of the first and third sections are the same (Vi/V = 1). This produces focusing 3  without an overall change in the energy of the transmitted particle. The Einzel lens used here is in an 'accelerating mode', as the voltage on the second section is kept such that  18  Figure 3. Beam Modulation Technique.  T+ = Time with top plate positive. T_ = Time with top plate negative. T> = Ion bunch pathway.  19  V2/V1 > 1. Typical operating voltages during the experiments were V i = V 3 = 0 volts and V2 = 750 volts. This acts to focus off axis ions onto the slit consequently decreasing the resolution but allowing more ions to exit the spectrometer. The ions pass through another pair of metal plates between which an electric field may be created to deflect the ion bunches either upward or downward relative to the axis of the ion path. These ion deflection plates were envisioned to assist in aligning the ion bunches with the laser pulse encounted as the ions leave the mass spectrometer. In practice though, the ion deflector was seldom used. As the ions continue to separate according to their masses along the tube, ions of a specific chosen mass may be further bunched along the axis of the ion path. The ion focus consists of a pair of metal discs 12.5 mm apart through which the ions travel. On the first disc is a negative voltage usually ranging from -50 to -100 volts, while the farther one is kept at ground. This voltage applied to the first plate is pulsed and coordinated with the beam modulation pulse, so the ion of interest alone is focused in this way. The mass resolution of the signal is improved for that ion but as we are now introducing a velocity spread for molecules with identical mass, the resolution of the electron spectrometer in the apparatus will be affected. The two regions of the mass spectrometer (see Figure 2) were kept at operating pressures of about 1 x 10" and 1 x 10" Torr using a pair of Edward's Diffstak diffusion 6  7  pumps (with pumping speeds of 160 and 100 litres per second respectively). The Bakker designed beam modulated type of mass spectrometer was chosen over the more commonly used Wiley-MacLaren and Wien Velocity Filter designs for two 34  35  20  main reasons. This set-up was found to produce ions with lower temperatures, and so would be beneficial in the study of larger molecules and clusters , and secondly, with the 30  focusing elements removed the Bakker design does not introduce velocity differences in molecules with the same mass (unlike the Wiley-MacLaren design) allowing increased electron resolution and permitting the current machine to incorporate a threshold detachment capability at a later date. The primary disadvantage is the coaxial alignment of the pulse valve and the entrance to the mass spectrometer, it allows unwanted neutral molecules to travel down the length of the drift tube and enter the laser interaction area where they may be ionized by the laser output and contribute to background noise.  2.4 Laser Photodetachment of the Negative Ions  The separated ions pass out of the mass selection area into the ultra high vacuum (UHV) region. This is kept at an operating pressure of 1 x 10" Torr using a 400 litre per 8  second Leybold Turbovac 340M turbo-molecular pump. The ions are detected by a chevron mounted pair of microchannel plates (Galileo MCP-040N, diameter 42 mm). As the ions travel across the space between the mass spectrometer slit and the detection plates, the ion mass of choice is intersected at 90 degrees with the output of a pulsed Nd: Y A G laser (Spectra Physics GCR Series 3) running at 30 Hz. In the experiments conducted here the 355 nm (3.492 eV) and 532 nm (2.335 eV) wavelengths were utilized. The output of the laser enters the UHV region of the machine through a 2.5 cm diameter fused silica (UVFS) window and exits via a similar window on the other side of  21  the chamber in order to minimize the problems caused by scattered light. In some of the experiments the laser remained unfocussed, the diameter of the beam being 10 mm. In others it was telescoped down to a diameter of 5 mm in an attempt to reduce background noise. As previously stated, this beam intersects the ion path and may be temporally aligned to strike a particular ion mass. The radiation from the laser is used to photodetach an electron from an ion, the resulting neutral molecule will continue along its path towards the detector. An ion block, consisting of a grid on the front of the detector held at -1 kV, is used to repel non-detached molecules (i.e. the remaining negative ions) and allow the neutral molecules to be detected. As the mass of the neutral molecules are essentially the same as their negative ion precursor's, the time of arrival at the detector corresponds to those seen for the precursor, allowing easy confirmation of the particular ion mass being irradiated. A fraction of the detached electrons now enter the second time-of-flight spectrometer orientated perpendicular to both the axes of the ion and laser output paths. The flight tube is 90 cm in length and is shielded using two concentric tubes of Conetic A A magnetic shielding (0.76 mm thick), the magnetic properties of this material help minimize the effects of outside magnetic fields on the detached electrons. The magnetic field strength was found to vary considerably in different locations within this section of the spectrometer (see Figure 4 and Table 1). In order to negate any possible distortions of the electron path due to the large field strength at the top edge of the magnetic shielding tubes, the detector was lowered by means of four 10 cm long struts further into the  22  Figure 4. Electron Drift Tube.  Section of Tube  Magnetic Field Strength (mG)  Top Edge  3000  Detector  20  Centre  5  Laser Interaction Area  10  Bottom Edge  60  Table 1. Magnetic Field Strengths in the Electron Drift Tube.  electron flight tube. In earlier experiments  the detector was situated 2.5 cm above the  top rim. The detector used for the collection of the electrons is of identical design to that used in the mass spectrometer.  As only electrons ejected at right angles to the ion  and laser paths are collected, degradation of the spectrometer resolution due to the kinematic effect of electrons with the same centre-of-mass energy but differing laboratory frame energies is reduced. Electrons from ions in different areas of the ion bunch will travel slightly divergent paths depending on where they are detached. However the most significant parameter in defining the electron spectrometer's resolution is the temporal pulse width of the laser (4-6 ns). The energy resolution degrades as E  3 / 2  and therefore  higher energy electrons are not as well resolved as comparatively lower energy ones . 36  2.5 Data Collection and Timing  The ion signal from the microchannel plates is processed by a 200 M H z transient digitiser (LeCroy TR8828D) and then stored in a computer, while the production of neutrals may be monitored on a 300 MHz oscilloscope (Tektronix). The signal from the electrons detected by the pair of microchannel plates is further enhanced by passing through an external lOx pre-amplifier (Ortec 9301) then a 300 MHz discriminator (Phillips 6904), before being processed by the LeCroy transient digitiser. The digitiser is read after every laser shot by a signal averager, and data is transferred after each 100 laser shots to a computer. A reflection from the laser output is used to trigger the transient digitiser.  24  The timing of the various components in the spectrometer is coordinated using a Stanford Research Systems DG535 pulse generator. There are five steps in the process: firstly the laser Q switch is sent a high voltage pre-trigger, T . The pulse valve is then 0  fired, A, and next the beam modulation plates, B (the oscilloscope is also triggered at this time to monitor the ion signal). Following this the ion focus is pulsed, C , andfinallythe Q switch is triggered to produce the laser pulse, D. The relationships and typical values can be expressed as:  A = T + 2.7 milliseconds 0  B = A + 500 microseconds C = B + 2.0 microseconds D = To + 3.2 milliseconds.  25  CHAPTER 3 RESULTS  In this chapter the negative ion photoelectron spectrometer will be characterized and the results of the experiments shown. The first part describes the resolution of both the mass selection and electron detachment components of the spectrometer, as well as the method used to calibrate the final results. The second part gives the experimental conditions and spectra obtained from the investigations of the three organometallic molecules: Cyclopentadienylcobalt dicarbonyl, methylcyclopentadienylmanganese tricarbonyl and the cobalt dicarbonylnitrosyl dimer Co2(CO)4(NO)2, from which we may determine the minimum vertical detachment energies for these compounds.  3.1 Resolution of the Mass Spectrometer  Resolution may be defined as the mass of the anion divided by the uncertainty in its mass: R = m/Am.  3.1  As the ions are measured as a function of time of arrival, it is necessary to express Equation 3.1 in the time domain. Using: E = l/2mv = l/2mL /t , 2  2  2  where E is the energy of the anion, v denotes the velocity of the anion, t the flight time of the anion and L is the length of the spectrometer.  26  Then: 3.2  t = Lm /(2E) 1/2  For a slightly larger mass (Am) the difference in arrival times is given by:  At = [L/(2E) ].((m + Am) ' - m "),  3.3  At = [Lm /(2E) ].((l + Am/m) - 1).  3.4  1/2  =>  1 2  1/2  1/2  1  1/2  Since Am/m is small, (1 + Am/m) = 1 +Am/2m,  3.5  1/2  =>  At = [Lm /(2E) ].(Am/2m). 1/2  3.6  1/2  From Equations 3.1 and 3.2:  At = t/2R  hence,  3.7  R = t/2At.  3.8  Using a mass spectrum of anions formed in the source region from a parent gas of O2, the resolution, or mass resolving power, of the spectrometer used in these experiments was determined. Table 2 gives the values for ions of different masses, where At is the full width half maximum (FWHM) of each peak. The spectrum is shown in Figure 5.  27  6000  +  Figure 5. Mass Spectrum of Oxygen Species. The figure shows the negative ions formed from Oxygen. The peaks shown are 0~ (a.m.u. 16), O2" (a.m.u. 32) and at 50 a.m.u., the O2-H2O anion. Experimental data points are shown as +. The solid line represents a fast fourier transform (FFT)filtersmoothing curve (n = 2). The smoothing function removes components with frequencies higher than l/n*At. Where n is the number of data points considered at a time, and At is the time spacing between adjacent data points.  28  Negative Ion  Flight Time ( L I S )  FWHM (LIS)  Resolution  0"  13.080  0.120  54.9  o-  18.500  0.200  46.2  02-H20"  23.100  0.240  48.2  2  Table 2. Mass Resolution for Oxygen Negative Ions.  By using the ion focus the spectrometer resolution for a specific ion may be increased. For example the resolution of the O2" peak in Figure 5 increases threefold by applying 100 volts to the ion focus.  3.2 Resolution of the Electron Spectrometer  For a time-of-flight spectrometer, the drift time T, of an electron in aflighttube of length L , is related to the electron energy by:  E = l/2m,L /T , 2  3.9  2  where me is the mass of an electron. The resolution of the spectrometer is limited by uncertainties in both theflighttime and the length of the drift tube. These lead to uncertainties in the measurement of the electron energy, AE. Since both T and L are 36  29  squared relative to mass, and energy is linear with respect to mass, Equation 3.9 can be expanded to include these inaccuracies: [ A E / E ] = [2AT/T] + [2AL/L] . 2  2  3.10  2  Since only the time-of-flight of the electrons is measured experimentally, uncertainties in the drift tube length may be expressed as additional errors in the timing (AT'). These are related by: AT' = A L / V ,  3.11  where: V = (2E/me) .  3.12  1/z  The total time width, AT , has components of AT and AT' and Equation 3.10 becomes: tot  [AE/E] = [2AT /T] . 2  3.13  2  tot  From Equation 3.9: T = (l/meE).L , 2  3.14  2  and substituting into Equation 3.13: A E = [(2AT ot)2m /L ].E.E 2  2  t  AE a E  e  2  and thus,  3.15  3.16  3 / 2  30  The resolution of the electron spectrometer is thus proportional to the kinetic energy of the electrons to the power of three-halves^ The resolution therefore increases with electron flight time. The photoelectron spectrum of O" was examined to demonstrate the resolution of the electron spectrometer used in these experiments. The spectrum shown in Figure 6 consists of the spectra induced by 500,000 laser shots added together with the wavelength of the laser output 355 nm (3.492 eV). From the measurement of the F W H M of the peak at 2.031 eV a resolution of 60 meV is derived. As the resolution of the spectrometer degrades as E  m  , the F W H M of the peak  may be adjusted to correspond to any value of the electron kinetic energy. For example, Posey et al, using a different incident photon energy, have reported a F W H M resolution 38  of 40 meV. By utilizing Equation 3.15 and the results from Figure 6, we may compare their values with those obtained by this spectrometer. The equivalent calculated resolution is 32 meV. A single peak is observed for the O" + hv —» O + e" reaction but this consists, however, of six separate transitions that are not resolved by our spectrometer. These derive from transitions from either the P /2 or ¥\n level in the anion to the P2,i,o levels in 2  2  3  3  the neutral atom. The transition corresponding to the electron affinity for oxygen, P 2  3  3 / 2  —>  P2, occurs in the centre of the peak . The electron affinity has been accurately measured 39  by Lineberger et al as 1.461 122 eV. Therefore the upper bound for our resolution is 60 40  meV at 2.031 e V e K E .  31  30000  20000  co c 3  hv  O  O 10000  1.0  1.5  2.0  2.5  3.0  3.5  4.0  Electron Energy (eV)  Figure 6. Photoelectron Spectrum of 0~ at 355 nm. The peak maximum occurs at 2.03 eV. The experimental data points are shown as + and the solid line is a FFT smoothing curve (n = 2). The photon energy corresponding to the output of the laser is marked by hv.  32  3.3 Calibration of the Electron Spectrometer  In conjunction with each of the experiments conducted, spectra of'standard' molecules with well known and characterized transitions were also taken in order to calibrate the electron spectrometer. The standard anions used were, ( V when the laser output was 355 nm and N O when the laser was operated at 532 nm. -  Known relative energies E i of the electron detached from calibrant anions are re  converted to their laboratory energies Eub, by correcting for the centre of mass factor:  3.17  Elab = E l - (mjWfy.E, r e  where E  c m  is the centre of mass energy of the anion, me is the mass of an electron and M  the mass of the anion. The kinetic energy of the electrons in the laboratory frame can be expressed by:  E  = l/2meL /T 2  l a b  3.18  2  where L is the length of the flight tube and T is the flight time of the electron. This becomes:  E  3.19  = 1/2 me L /(T - To) , 2  l a b  2  33  where T is the finite time delay between the pulsing of the laser and the triggering of the 0  transient recorder, via a signal from a photodiode initiated by a back reflection from the laser. By rearranging Equation 3.18 we arrive at a time offlightfor the electron as:  T = (1/2 mo/Ei,b) .L + T .  3.20  1/2  0  As E i  a  b  is calculated from 3.17 and T is measured experimentally, a least squares fit  to Equation 3.20 will provide values for both L and the time offset, To. Typical values for L and T are 93.90 cm and 10.11 nsec. respectively. 0  The spectrum then undergoes a further transformation to account for the change in resolution with electron energy (dT oc dE/E , see section 3.2) to ensure the peak areas 3/2  remain constant.  3.4 Calibration at 355 nm. The Photoelectron Spectrum of 0 ~ 2  Oxygen was pulsed directly into the spectrometer, with typical conditions being a backing pressure of 30 psig, pulse valve repetition rate of 30 Hz. After optimization, 250350 mV of O2"" ions were detected. The various ionic species produced in the mass spectrometer are shown in Figure 5. This ion beam was crossed with the 355 nm output from the laser (laser power 0.8 W, horizontally (H) polarized, beam diameter 5 mm, repetition 30 Hz). In the spectrum shown, Figure 7, 60,000 laser shots were used and the spectra cumulatively added, with an equivalent number of background counts subtracted.  34  8000  j i h g f e d c b a 6000  1.5  2.0  2.5  3.0  3.5  4.0  Electron Energy (eV)  Figure 7. Photoelectron spectrum of O2".  Photoelectron spectrum taken at 355 nm. The peaks a to e correspond to transitions to the first five vibrational levels in the neutral O2 (XS") from the ground state of the 0" anion (XL"g). The peak labeled a at 3.0 eV corresponds to the electron affinity transition of O2 at 0.45 leV. Experimental data points are shown as +, while the solid line represents a FFT smoothing curve (n = 2). 3  g  2  35  2  The background spectra are taken under identical conditions but with the electron gun filament turned off, prohibiting the formation of ions. The O2 molecule has been investigated several times previously using similar techniques to those used here ' ' . Travers reports an electron affinity for diatomic 37 38 41  oxygen as 0.451 ± 0.007 eV. A representative spectrum obtained from our spectrometer is shown in Figure 7. Thefivepeaks labeled a through/ correspond to the transitions from the ion ground state (X E' ) to the neutral ground state (X I" {v}, where v = 0 to 4). The 2  3  g  g  peaks marked g-i represent the transitions from the O2" (X I" ) to the neutral molecules' 2  g  excited electronic state a'Agfv}. The peak labeled f probably has contributionsfromboth the X E' {v = 5} and n A {\ = 0} states. The peak a shows the electron affinity of 2  l  g  g  diatomic oxygen as it represents the transitionfromthe ground electronic, vibrational state of the anion to the corresponding level in the neutral 0 molecule. 2  These peaks are used to calibrate the electron spectrometer. Vibrational spacings (e ) between the v = 0 ground state and subsequent levels are determined using the v  equation:  e = (v + l/2)co - (v + l/2) (o Xe + (v + l/2) (o y ... 2  v  e  3.21  3  e  e  where co is the oscillation frequency for oxygen and © e  e  Xe  e  and co y are thefirstand second e  e  anharmonicity constants respectively . 42  As the electron affinity for O2 is known to be 0.451 ± 0.007 eV and the laser photon energy at the 355 nm wavelength is 3.492 eV, the vibrational spacings can be used  36  to predict the different energy levels corresponding to the different vibrational states of the neutral molecule. These energy differences can then be matched to the electron flight times found in the spectrum, see Table 3. Utilizing Equation 3.20 both L and T may be 0  ascertained for a particular experiment.  Electron Kinetic Energy (eV)  Typical eArrival Times (ns)  3.041  920  0.1932  2.845  950  3876.30  0.3830  2.658  980  3 <-0  5385.71  0.5701  2.471  1020  4<-0  6872.12  0.7544  2.287  1060  Peak Assignment  Transition v' <- v"  8 (cm" )  a  0 <- 0  787.18  b  1 <-0  2343.56  c  2<-0  d e  1  V  AE (eV)  —  Table 3. 0 Calibration of the Electron Spectrometer. 2  Calculation of energy differences and ejected electron energies. co = 1580.361 cm" , © X e 12.073 cm" and co y = 0.0546 cm" (ref42). E.A. ( 0 ) = 0.451 ± 0.007 eV (ref41). 1 eV = 8065.48 cm" . 1  e  1  1  e  e  2  1  37  e  =  3.5 Calibration at 532 nm. The Photoelectron Spectrum of NO  NO" was produced inside the source chamber of the spectrometer along with a number of other anions from the parent gas of N 0. The nitrous oxide was pulsed directly 2  into the machine at a backing pressure of 25 psig, a pulse valve rate of 30 Hz, resulting in a source pressure of 3.5 x 10" Torr. Under optimal conditions, 150 mV of NO' were 5  detected. Figure 8 shows the mass spectrum of the species observed. The electron energy spectrum was collected by photodetaching electrons from the NO' ions using 532 nm frequency photons (laser power 1.375 W, H polarized, beam diameter 5 mm, 30 Hz repetition rate). For the spectrum shown in Figure 9, 100,000 laser shots were used and the spectra cumulatively added. The same number of counts were taken with the electron gunfilamentturned off, these were subtracted from the original spectrum to minimize the background noise. The electron affinity of NO has been reported as 24 ± 14 meV  43,44  . The spectrum  shows peaks corresponding to the transitions; NO ( X n , v' = 0 to 4) <- NO' (X E", v" = 2  3  0). The peaks labeled a to e are slightly asymmetric, appearing to have a shoulder on the left hand side. The NO n ground state is split due to spin-orbit coupling of the electrons' 2  spin with it's angular orbital momentum into Hm and n 2  2  3 / 2  states, separatedfromeach  other by 121.1 cm' (15 meV) . The presence of these two states will mean that the 1  43  electron affinity cannot be measuredfromthe centre of the peak corresponding to the (0,0) transition (peak a). Temperature dependent rotational effects must also be considered. Seigal et al estimated that for a source temperature of 630 K, a correction 43  38  factor of 12.5 meV was required. Coe et al, approximating their source temperature at 44  200 K, used a correction of 13 meV for the combined spin-orbit and rotational effects. For the purposes of the calibration here, the maximum of each peak has been taken as the energy difference between the ground state of the NO" anion and the successive vibrational levels in the N O molecule. As was the case with the oxygen calibration of the 355 nm work, using known values for the electron affinity and the spectroscopic constants of NO, along with the knowledge of the laser output photon energy (532 nm —» 2.334 eV), arrival times of ejected electrons can be matched to the corresponding Eub to yield the values of L and T needed to calibrate that day's experiments (see Table 4).  39  0  8000 7000 6000 5000 h §  4000 h  o O 3000 2000 1000 h  10  20  30  40  50  Mass (a.m.u.)  Figure 8. Mass spectrum of anions obtained from N 0 . The figure shows the species formed inside the source chamberfromthe parent gas N 0. The peak at 16 a.m.u. corresponds to 0\ at 46 a.m.u. N0 ' and the peak at 30 a.m.u. to NO". The experimental data points are shown as +, the solid line is the FFT (n = 2) smoothing curve. 2  2  2  40  4000  1.0  1.5  2.0  2.5  Electron Energy (eV)  Figure 9. Photoelectron Spectrum of NO". Photoelectron spectrum taken at 532 nm. The peaks a toe correspond to the transitions to thefirstfivevibrational levels in the neutral NO (XTT) from the NO' (X E') anion. The maxima of peak a is taken to be electron affinity of NO at 0.024 eV. Experimental data points are expressed as +, the solid line is a FFT (n = 2) smoothing curve. 3  41  e (cm" )  AE (eV)  Electron kinetic energy (eV)  Typical earrival time (ns)  0 <- 0  948.52  -  2.307  1060  b  1 <-0  2824.61  0.2326  2.074  1125  c  2 <- 0  4672.74  0.4617  1.845  1185  d  3 <-0  6492.92  0.6874  1.619  1265  e  4<-0  8285.13  0.9096  1.397  1370  Peak assignment  Transition  a  1  v  v' <- v"  Table 4. NO Calibration of the Electron Spectrometer. Calculation of energy differences and ejected electron energies. co = 1904.03 cm" , co Xe = 13.97 cm" and ca y = - 0.0012 cm" (ref 42). E . A (NO) = 0.024 ± 0.01 eV (ref 43). 1 eV = 8065.48 cm" . 1  e  1  1  e  e  1  42  e  3.6 Photoelectron measurements of CpCo(CO)2  Cyclopentadienylcobalt dicarbonyl was introduced into the spectrometer by flowing a carrier gas of helium or argon (with a backing pressure of 25 psig) through a tube containing liquid CpCo(CO)2. The mass spectrum obtained (Figure 10) shows a prominent peak at 180 a.m.u. corresponding to the anion of the parent molecule. Under optimal conditions a signal of 60 mV (into 50 Q) is recorded.  3.6.1. Measurement of CpCo(CO)2~ in He using 355 nm laser photons. The electron spectrum of cyclopentadienylcobalt dicarbonyl anion using helium as the backing gas is shown in Figure 11. The spectrum shows two broad peaks that overlap one another. The first begins at 2.6 ± 0.1 eV eKE, while the second larger peak originates at 1.7 ± 0.1 eV eKE. In the spectrum shown, the spectra resulting from 490,000 laser shots were added together (laser power 0.98 W, 30 Hz, H polarized, beam diameter 6 mm), with an equal number of background runs subtracted.  3.6.2. Measurement of CpCo(CO)2~ in Ar using 355 nm laser photons. Equivalent experiments to those mentioned above but using argon as the carrier gas were conducted, the results are shown in Figure 12. Again the spectrum consists of a steep slope rising from 1.7 ± 0.1 eV to 1.0 eV eKE and an overlapping smaller peak beginning at 2.6 ± 0.1 eV eKE. Only a third as many ions were produced in the mass spectrometer region (~ 20 mV) compared to the previous experiment conducted in helium  43  and the spectrum shown consists of 270,000 laser shots (laser power 1.25 W, 30 Hz, H polarized, beam diameter 6 mm), resulting in a decreased signal-to-noise ratio to that seen in Figure 11. As argon tends to assist in the cooling of anions formed in the supersonic expansion to a greater extent than helium, as far as it is possible to test with the means available to our spectrometer, the fact that the peaks appear at the same energies as those in Figure 11 suggests that the peak originating at 2.6 eV eKE is due to ground state transitions and not to the presence of excited 'hot' vibrational bands in the anion.  3.6.3. Measurement of CpCo(CO)2~ in He using 532 nm laser photons. The laser output wavelength was changed to 532 nm (corresponding to the photons possessing 2.334 eV of energy, laser power 1.25 W, 30 Hz, H polarized, beam diameter 8 mm). This should reduce the energy of the ejected electrons and hence increase the resolution in the electron spectrometer. Figure 13 shows the result of adding spectra from 410,000 laser counts. The spectrum has a peak beginning at 1.5 ± 0.1 eV eKE. There appears to be another sharp increase at 0.6 ± 0.1 eV eKE, which may correspond to the second increase seen in the 355 nm data (Figure 11). However data collected in this region are close to the limit of the spectrometer's range and so a definite assignment is difficult. Table 5 shows a summary of the lower limits for vertical detachment energies (E vD) due to the reaction CpCo(CO) ~ + hv -> CpCo(CO) + e" under various m  2  2  conditions.  44  4000  3000 h  Mass (a.m.u.)  Figure 10. Mass Spectrum of CpCo(CO) ~. The major peak at 180 a.m.u. corresponds to the CpCo(CO) " anion. The peaks seen further along the spectrum at 198, 218 and 236 a.m.u. are thought to be anions CpCo(CO)2.H2Cr, MeCpMn(CO) " and MeCpMn(CO) .H 0" respectively. The experimental data are shown as +, the solid line represents a FFT (n = 2) smoothing curve. 2  2  3  3  45  2  150000  0.5  1.0  1.5  2.0  2.5  3.0  3.5  4.0  Electron Energy (eV) Figure 11. Photoelectron Spectrum of CpCo(CO) " in He at 355 nm. The experimental data are shown as +, the solid line represents a FFT (n = 10) smoothing curve. The laser photon energy of 3.492 eV is indicated by hv. 2  46  60000  Figure 12. Photoelectron Spectrum of CpCo(CO)2~ in A r at 355 nm. The experimental data points are shown as +, the solid line is a FFT (n = 5) smoothing curve. The laser photon energy of 3.492 eV is indicated by hv.  47  Figure 13. Photoelectron Spectrum of CpCo(CO) " in He at 532 nm. The experimental data points are shown as +, the solid line represents a FFT (n = 10) smoothing curve. The laser photon energy of2.334 eV is indicated by hv. 2  48  Electron Energy (eV)  E VD  He  1.5 ± 0 . 1  0.8 ± 0 . 1  355  He  2.6 + 0.1  0.9 + 0.1  355  Ar  2.6 + 0.1  0.9 + 0.1  hv (nm)  Carrier Gas  532  M  (eV)  Table 5. Vertical Detachment Energy Measurements of C p C o ( C O ) 2  49  3.7 Photoelectron measurements for MeCpMn(CO)3  Helium gasflowingover liquid methylcyclopentadienylmanganese tricarbonyl placed in the bottom of a tube, was used to introduce MeCpMn(CO)3 into the spectrometer. The vapour pressure at room temperature of this compound is less than that for the CpCo(CO)2 molecule, consequently the number of ions formed was also less. After optimization, an approximately 20 mV signal (into 50 Q) of the parent anion was typically detected. Figure 14 shows the mass spectrum of the species formed in the source region. The major peak seen at 218 a.m.u. is due to the MeCpMn(CO)3~ parent anion.  3.7.1. Measurement of MeCpMn(CO)3~ at 355 nm. The photoelectron spectrum of MeCpMn(CO)3~ in a helium backing gas, using a laser output wavelength of 355 nm (laser power 1.4 W, 30 Hz, H polarized, beam diameter 7 mm), is shown in Figure 15. A peak may be seen originating at 3.0 ± 0.1 eV eKE which increases to a maximum near 2.5 eV and then begins to decrease until at 1.9 + 0.1 eV eKE the signal increases once again. The spectrum shown here represents the cumulative spectra from 200,000 laser shots.  3.7.2. Measurement of MeCpMn(CO) " at 532 nm. 3  The above experiment was conducted again with a different laser output wavelength. Figure 16 shows the spectrum at 532 nm (laser power 1.25 W, 30 Hz, H polarized, beam diameter 7 mm). Here the first peak rises at 1.9 + 0.1 eV eKE, an energy  50  2500  2000  -  1500  •  •  &  •  1000  500  ft?  175  /TVI  •+  ++^  -  +  200  1  1  225  250  275  Mass (a.m.u) Figure 14. Mass Spectrum of MeCpMn(CO) ~. The spectrum shows peaks at 218 a.m.u., corresponding to the parent anion MeCpMn(CO)3~, as well as a peak before and after (at 180 and 236 a.m.u.). These peaks probably represent the anions CpCo(CO)2~ and MeCpMn(CO) .H 0" respectively. Experimental data points are shown as +, the solid line is a FFT (n = 2) smoothing curve. 3  3  51  2  10000  2.0  2.5  3.0  3.5  4.0  Electron Energy (eV)  Figure 15. Photoelectron Spectrum of MeCpMn(CO) ~ at 355 nm. The experimental data points are represented as +, the solid line is a FFT (n = 10) smoothing curve. The laser photon energy of 3.492 eV is indicated by hv. 3  52  8000  ^  1.0  +  1.5  2.0  2.5  3.  Electron Energy (eV)  Figure 16. Photoelectron Spectrum of MeCpMn(CO) " at 532 nm. The experimental data points are shown as +, the solid line represents a FFT (n smoothing curve. The laser photon energy of 2.334 eV is indicated by hv. 3  53  difference from the laser photon energy of 0.4 ± 0.1 eV. In this spectrum it is not possible to distinguish if the peak reaches a maximum and then further on begins to climb once more as suggested in the experiment at 355 nm. The spectra from a total of 470,000 laser shots were collected in this run. Table 6 gives a summary of the results for the reaction  MeCpMn(CO) ~ + hv -> MeCpMn(CO) + e~ 3  (3.7.1 and 2).  3  E  (eV)  hv (nm)  Electron Energy (eV)  355  3.0±0.1  0.5 ± 0 . 1  532  1.9 + 0.1  0.4 ± 0 . 1  m V D  Table 6. Vertical Detachment Energy Measurements for MeCpMn(CO)  54  3  3.8 Photoelectron measurements of Co2(CO) (NO)2 4  Cobalt tricarbonylnitrosyl in a carrier gas of helium, was introduced into the spectrometer using the methods previously described. A number of anions were formed situ, including the cluster Co2(CO)4(NO)2\ at 290 a.m.u. Figure 17 shows the mass spectrum of the ions formed in the gas expansion. The parent anion can be seen at 173 a.m.u., as well as the daughter anion, Co(CO)2NO" at 145 a.m.u. Under optimal conditions, a 50 mV signal (into 50 Q) was observed for the cobalt dimer ions.  3.8.1 Measurement of Co (CO) (NO) ~ at 355 nm. 2  4  2  Figure 18 shows the photoelectron spectrum obtained after adding spectra collected during 320,000 laser shots (laser power 0.72 W, 30 Hz, H polarized, beam diameter 5 mm), and subtracting an equivalent number of background counts. This increases the uncertainty in the determination of the E  m  V  D  for the reaction;  (Co) (CO) (NO) - + hv -»(Co) (CO) (NO) + e". 2  4  2  2  4  hv (nm)  Electron Energy (eV)  355  1.5 ± 0 . 2  2  E  m  V  D  (eV)  2.0 ± 0 . 2  Table 7 Vertical Detachment Energy Measurement for Co (CO) (NO)2 . 2  55  4  4000 3500 3000 2500 C 2000 r -  100  200  300  Mass (a.m.u.)  Figure 17. Mass Spectrum of Co (CO) (NO) ~. 2  4  2  The four main peaks in this spectrum correspond to (a) Co(CO) NO" at 145 a.m.u., (b) Co(CO) NO" at 173 a.m.u. (c) Co (CO) (NO) ~ at 262 a.m.u. and (d) Co (CO) (NO) ~ at 290 a.m.u.. The experimental data points are shown as +, the solid line is a FFT (n = 2) smoothing curve. 2  3  2  3  2  56  2  4  2  120000  Figure 18. Photoelectron Spectrum of Co (CO)4(NO) ~ at 355 nm. The experimental data points are shown as +, the solid curve represents a FFT (n = 10) smoothing curve. The laser photon energy of 3.492 eV is indicated by hv. 2  57  2  CHAPTER 4 DISCUSSION  4.1 Cyclopentadienyl Cobaltdicarbonyl CpCo(CO)  2  The negative ion radical of the parent molecule cyclopentadienylcobalt dicarbonyl was generated in the source region of the spectrometer. Although formally a nineteen electron system with respect to the cobalt metal centre, this anion has been reported on previous occasions. For example, in ICR experiments by Beauchamp and Taylor , and 45  46  by McDonald and Schell using a FA designed spectrometer, ESR techniques have also 47  been utilised . 48  The neutral molecule, CpCo(CO) , has eighteen valence electrons around the 2  cobalt metal centre, with the cyclopentadienyl ligand acting as a r\ five electron donor. s  Crichton et al have reported the IR v(CO) stretching bands for this compound, as shown 49  in the table below.  v-(CO) v-(ring)  CO matrix  Ar matrix  2032.1 1971.9 817.3  2032.8 1971.0 816.5  Table 8. CO and Cp ring stretching frequencies for CpCo(CO) at 12 K . (Ref. 50). 2  58  Using the relative intensities of the symmetric and asymmetric CO stretches and the relationship I  = cot [0/2] (for M(CO)2 type complexes), the OC-Co-CO angle 2  /I  sym  asym  was found to be 94 degrees, with C v local symmetry . 49  2  The Fenske-Hall molecular orbital diagram for CpCo(CO) has been determined 2  most recently by Lichtenberger and previous to that by Albright and Hoffmann . A 50  51  schematic is shown in Figure 19. The four occupied 'predominantly metal' orbitals of the Co(CO) component are the lai, la , lbi orbitals, stabilized due to back-bonding by the +  2  2  carbonyl ligands, and the 2ai orbital which is also stabilized by back-bonding but sits slightly higher than the other three due to some a anti-bonding character. The most significant interaction between the ring and the metal occurs via electron donation from the ring ef orbital to the metal lb orbital. There is also an interaction between the ei ring +  2  orbital and the metal lbi. Many reactions in which CpCo(CO) is involved proceed via an associative 2  mechanism. For example, the photochemical reaction; r| -CpCo(CO) + CO 5  2  n -CpCo(CO) . 3  3  The reversibility of this reaction may be interpreted as confirming the presence of an expanded coordination number species as an intermediate . Crichton also found that CO 49  exchange in CpCo(CO) showed a first order dependence on the CO concentration, which 2  suggests a bimolecular SN2 associative mechanism. Similar CO exchange in Co (CO)g 2  proceeds through a dissociative mechanism. As bonding between M-CO in Cp(CO)„M (M = Metal) complexes is essentially the same as that for simple metal carbonyls such as Co (CO)g, thefirstorder dependence on CO concentration is consistent 2  59  Co(CO>2  +  CpCo(CO)2  ^"s  Figure 19. Fenske-Hall Molecular Orbital Diagram for CpCo(CO) . The electrons in the HOMO are shown. From reference 50. 2  60  with the proposal that the cyclopentadienylringligand plays a direct role in the exchange. It has been suggested that the presence of the Cpringallows a transfer of electron density from the metal centre to the ring . 52  Susceptibility toward nucleophilic attack by negative ions may be ascribed to either coordination saturation of the metal centre, the presence of a formal positive charge on the metal centre or the presence of a ligand which can accommodate an increase in charge of the substrate. Nucleophilic attack on CpCo(CO)2 proceeds via reaction at a CO ligand with subsequent elimination of the other CO ligand and migration to the metal centre, or through a r| to r) change in the bonding of the C5H5 ligand prior to the bond formation 5  3  between the nucleophile and the cobalt atom, followed by elimination of the CO ligand . 53  The change in hapticity of the cyclopentadienyl ligand requires a structural rearrangement. The crystal structures of molecules containing the r^-CsHs ligand show it to be bent, with an angle of 20° between the allylic and olefinic systems ' . This removes 49 54  part of the 7t system from the coordination sphere of the metal. In effect, the cobalt metal centre behaves as a 17 electron system. Byers and Dahl investigating the substituted Cp 55  form of the molecule (Cp*Co(CO)2 where Cp* = C (CH ) ) found in their crystallographic 5  3  5  analysis a 'statistically significant' distortion of the Cp*ringdue to its noncylindrical bonding interaction with the M(CO)2 fragment. This structural alteration on hapticity change would be expected to be seen in the electron energy spectrum we observe in our experiments. However, the size of the molecules involved makes determination of any progression due to structural realignment difficult. Due to the relatively large number of atoms in the molecules being investigated here, compared to most published studies using  61  the NTPES technique, broad spectra many times wider than the instrument resolution are observed. This broadening is probably due to overlapping of the various vibrational (and rotational) frequencies in the complex. McDonald and Schell also found that reactions of the negative ion proceeded via 47  an associative mechanism that necessitated a hapticity change. Using a FA apparatus they observed that CpCo(CO) ~ underwent substitution reactions with NO, PF and O2, 2  3  amongst others. They found that the CpCo(CO)2 ~ anion could participate in electron transfer reactions as well. Using the examples of the reactions below, upper and lower bounds for the electron affinity of CpCO(CO) were proposed. 2  CpCo(CO) + S0 -> CpCo(CO) + S0 " _  2  2  2  2  CpCo(CO) •" + [CH CO] -> CpCo(CO) + [CH CO] " 2  3  2  2  3  2  As the electron affinity of S0 is known to be 1.097 ± 0.036 eV , and that of [CH CO] 56  2  3  2  to be 0.69 eV , an upper limit of 0.69 eV was suggested. The reaction involving 57  [CH CO] was significantly slower than that for S0 hinting that the actual electron 3  2  2  affinity may be quite close to the 0.69 eV limit. A lower boundary was determined using the reaction below; CpCo(CO) " + CS -> CpCo(CO) (CS r 2  2  2  2  CS ~ was not observed, no electron transfer took place, and so the lower limit to the EA 2  of CpCo(CO)2 was placed at 0.51 eV (the EA of CS . ). Combining the boundaries 58  2  McDonald and Schell proposed EA(CpCo(CO) ) = 0.60 ± 0.1 eV. 2  In the experiments conducted here, a minimum vertical detachment energy of 0.9 ± 0.1 eV was found for the reaction;  62  CpCo(CO) •" + hv -> CpCo(CO) + e 2  2  This value conformed with a recently calculated theoretical value for the electron affinity 59  of CpCo(CO) of 0.84 eV, and is in reasonable agreement with the range published by 2  McDonald et al. The minimum vertical detachment energy observed for Co(CO)3NCT is 1.72 ± 0.03 eV . This value is significantly larger than the one obtained for Co(CO) Cp~. More energy 21  2  is required to remove the electron from the cobalt nitrosyl anion. This is counter intuitive, however, as the Cp ligand is larger than the NO, and we would expect tofindthe added electron delocalised to a greater extent. In addition, the expected geometry change due to the Cp ligands proposed hapticity change, rj -> rj , is much larger than the bend 5  3  associated with the NO ligand reducing its symmetry from C 3 to C . Both of these effects V  would likely result in a larger value for the E  m V  D  s  of CpCo(CO) ~. 2  It may be that the relative ease of the hapticity change for the NO ligand results in the Co(CO) NO~ anion resembling closely a 17 electron species (it's E vo is reported to 3  m  be very similar to the E D of the 17 electron Co(CO) NO~), whereas the CpCo(CO) ~ mV  2  anion seems to show characteristics of an 18 + 8 species.  63  2  4.2 Methylcyclopentadienyl Manganesetricarbonyl (MeCp)Mn(CO)  3  In contrast to a number of proposed reaction mechanisms discussed for the cobalt complex above, the majority of known photochemistry involving CpMn(CO)3 is consistent with dissociative loss of a CO ligand as the principal result of electronic excitation . The 61  quantum yields obtained for reactions of the type described below, the substitution of CO by a nucleophile, are reportedly near 1.0  62  (ri -Cp)Mn(CO) + X  (Ti -Cp)Mn(CO) X + CO,  5  5  3  2  where X is CH3COCH3 or PhC=CPh. The chemistry of (r| -MeCp)Mn (CO) displays many similarities to that of (r| 5  5  3  Cp)Mn(CO)3. However, a number of changes in the nature of the molecule occur with the addition of a methyl group to the cyclopentadienyl ligand. The effect is illustrated in the comparison of spectroscopic data for the two complexes. Absorption maxima for (r| -MeCp)Mn(CO)3 are red-shifted a small amount 5  compared to (r| -Cp)Mn(CO) . This is consistent with the electron releasing nature of 5  3  CH , which places more negative charge on the Mn metal centre. The IR v(CO) stretch 3  frequencies are also slightly shifted upon methylation (see Table 9). Using the relationship between carbonyl stretching frequencies and oxidation potential, Connelly and Kitchen estimated values for E , the potential at which oxidation 66  p  occurs, of 1.33 and 1.30 V for the Cp and MeCp complexes respectively, citing the carbonyl stretching force constants as k(CO) =15.57 and 15.52 mdynA" . 1  64  Complex  v(CO) cm"  1  2025, 1942  CpMn(CO)  3  (MeCp)Mn(CO)  3  2024, 1938  Table 9. v(CO) stretching frequencies for Manganese carbonyl complexes with the Cp and MeCp ligand.  (Note: There are some differences in the reported stretching frequencies for CO in (MeCp)Mn(CO) , see references 63, 64 and 65. However, all studies that directly 3  compare the Cp and MeCp analogues show a shift to lower frequency on methylation. The values used in Table 9 are from reference 61). 13  C NMR studies have found a downfield shift in the 8 values for C(MeCp) with  respect to the value for C(Cp), Lyatifov et al report a AS of 19.62 ppm. The proposed 67  explanation suggests that in MeCp ligand complexes of transition metals, additional deshielding may be due to a reduction of the influence of 7i-electron current in the central Cp-Metal bond. The shielding effects ofrc-electronsare commonly regarded as a primary reason for upfield shifts in ligand nuclei signals uponrc-complexformation. Introduction of a methyl group results in lower symmetry due to increased distortion in the cyclopentadienyl ligand and the removal of the degeneracy in the e-orbitals ( e i and e ^f*. g  2  Valence ionization's are also found to decrease in energy with ring methylation. This conforms to the proposed greater electron releasing (or inductive) effect of the methyl groups that increase the electron charge in the ring, increasing its 'basicity'. However, Lichtenberger has suggested that the electron releasing nature of the methyl  65  groups plays only a small role ' . Using the Fenske-Hall notation, the degenerate e" 69  70  orbitals in the free C5H5" anion are split upon coordination with the Mn(CO)3 cation, and +  that the splitting in these levels is even greater for CsfLJVle". A distortionfromthe fivefold symmetry offreeCp' was proposed to account for this splitting. It was suggested that the perturbation of the electronic structure upon methylation could be viewed as being due to two primary factors; (i) the charge redistribution (induction) through the molecule and (ii) the different direct orbital overlap interactions in the valence shell. However, as mentioned earlier, there are several reported examples of the (MeCp)Mn(CO)3 molecule undergoing CO substitution via a dissociative pathway, similar to that of the non-methylated compound ' ' ' ' ' . In all of these examples, the cyclopentadienyl ligand (whether methylated or not) is assumed to be occupying three coordination sites in the pseudo octahedral symmetry of a M L species, Cp" is proposed to 6  be isolobal with (CO)3, and t) electron donating . 5  74  In our experiments the anion (MeCp)Mn(CO)3~ is formed. It may be that the manganese complex acts more like a "true" 19 electron compound (i.e. it has a greater than 18 electron density around the metal centre, corresponding to a type (1) in section 1.1) than the cobalt complex considered earlier. That molecule seems to show a greater tendency for ligand hapticity alteration to lower the electron density around the central atom to a level more akin to a 17 electron compound (more akin to a type (6)). If this were the case, then the structural alteration upon loss of the extra electron ( AB" —> AB + e") would be less for the manganese complex. The broad nature of our obtained spectra makes speculation of different degrees of structural change difficult.  66  MeCp (r| - » rj ) change is still a possibility. Cooper et al have reported evidence s  3  65  for an r| -MeCp ligand intermediate in the reaction of (MeCp)Mn(CO) with [K(18-C3  3  6)]K and r f to form the diene (C6H )Mn(CO) " (where [K(18-C-6)]K is a crown ether ). 76  8  3  A minimum vertical detachment energy of 0.4 ± 0.1 eV is reported in this work. Using FT-ICR techniques, Van Orden and Buckner have bracketed the AH^d for 76  (MeCp)Mn(CO) . The manganese complex reacts with nucleophiles NH2* and F' via 3  deprotonation with no nucleophilic activation at the carbonyl. (MeCp)Mn(CO) + (F\ NH ) -> (C6^)Mn(CO) ' + (HF, NH ), 3  2  3  3  where AH d(HF) = 371.5 kcalmol' , 1  aci  and  AHacid (NH3) = 403.6 kcalmol' . 1  It is found that CI" (AH  (HC1) = 333 kcalmol' ) is unreactive with (MeCp)Mn(CO) due 1  acid  3  to its lower proton affinity. This then allows us to bracket the gas phase acidity of the manganese complex, yielding; 333 > [AHacid (Mn(MeCp)(CO) )] > 372 kcalmol' . 1  3  Utilizing these values, a thermodynamic cycle may be constructed;  AHacid AH  EA(AH)  A+H+  IP(H)  67  where A - (CO) MnCpCH , and the Ionization Potential (IP) of Hydrogen is 13.595 eV . 77  3  2  From this, the value for the bond dissociation energy of the methyl-hydrogen bond, D(A~—H), may be determined as: E(*) = AH ac id(AH) + EA(AH) - IP(H)  Therefore * may be bracketed between 1.365 and 3.065 eV. Consequently, a value for D((CO) MnCpCH "^I) of 2.2 ± 0.9 eV is proposed. 3  2  4.3 Cobalt dicarbonylnitrosyl Dimer [Co(CO) NO] 2  2  The parent molecule of this dimer was found to undergo both CO ligand loss as well as clustering in our experiments. This opens up the possibility of determining CO—Metal bond strengths and elucidation of some structural properties at a later date, as not all of the relevant information needed is available at this time. The minimum vertical detachment energy of Co(CO) NO has been reported previously , and was found to be 20  2  1.72 eV, the corresponding detachment energy of the parent Co(CO) NO has been 3  tentatively placed at 1.73 eV . A value for the bond dissociation energy for the CO—Co 21  bond in the anion is required in order to estimate CO—Co bond strength in the neutral species. In this set of experiments, the minimum detachment energy of the dimer [Co(CO) NO] is estimated at 2.0 ± 0.2 eV. A value for the CO—Co bond dissociation 2  2  energy in the corresponding anion has been reported at 1.12 + 0.15 eV . The peak 78  belonging to the ion Co2(CO) (NO) ~ can be seen in the mass spectrum reported in the 3  2  68  Results Chapter (Figure 17), if the electron affinity of the neutral molecule can be determined then a bond strength for the Co (CO) (NO)2—CO bond can be estimated. 2  3  The comparison of the CO—Co bond strengths in Co(CO) NO and 3  Co2(CO)4(NO)2 may assist in determining whether or not the carbonyl ligands play any role in the Co—Co bond of the dimer, that is, if they bridge the metal-metal bond. The minimum vertical detachment energies of Co(CO)2NO" and [Co(CO) NO] ~ 2  2  are 1.72 ± 0.03 and 2.0 ± 0.2 eV eKE respectively. The larger value corresponding to the dimer probably represents the greater delocalisation of the extra electron in this molecule.  69  SUMMARY  The minimum vertical detachment energies (E d) of CpCo(CO) ", mV  2  MeCpMn(CO) " and Co2(CO) (NO)2~ were determined using NTPES techniques. The 3  E vD m  4  of CpCo(CO)2 at 0.9 ± 0.1 eV was found to be significantly lower than that -  determined for the related Co(CO)3NO~ anion, reported as 1.72 ± 0.03 eV. The  E  m V  D  of the cobalt dimer Co (CO) (NO)2~ was determined at 2.0 + 0.2 eV, a 2  4  value slightly higher than that reported for the Co(CO) NO" anion, at 1.73 ± 0.03 eV, 2  obtained using the same technique. If the E  m V  D  of the anion Co2(CO)3(NO)2~ can be  determined, some insight into the nature of the Co-Co bond in the dimer may be gained. The  EmVD  for MeCpMn(CO)3~ was determined as 0.4 ± 0.1 eV. Utilizing a  thermodynamic cycle, containing the known values for AH id(MeCpMn(CO) ~) and the ac  3  IP(H), the bond dissociation energy of the H—CH CpMn(CO)3~ bond was found to be 2  D(H—CH CpMn(CO) l = 2.2 ± 0.9 eV. 2  3  70  REFERENCES. 1.  J. Schmit, R. Hafner, T. Jira, R. Sedlmeir, R. Seiber, R. Ruttinger and H . Kojer Angew. Chem. 71 176 (1959).  2.  C A . Tolman Chem. Soc. Rev. 1 337 (1972).  3.  T.L. BrownAnn. N.Y. Acad Sci. 333 80 (1980).  4.  J.K. Kochi Organometallic Mechanisms and Catalysts Academic Press (1978).  5.  N G . Connelly and W.E. Geiger^^v. Organomet. Chem. 23 1 (1984).  6.  W.E. Geiger and N G . Connelly Adv. Organomet. Chem. 24 87 (1985).  7.  J. Halpern Pure Appl. Chem. 51 2171 (1979).  8.  G. Wilkinson J.Am. Chem. Soc. 74 6148 (1952).  9.  W . C Trogler Int. J. Chem. Kinetics 19 1025 (1987).  10.  D. Astruc Chem. Rev. 88 1189 (1988).  11.  J.R. Hamon, D. Astruc and P. Michaud J. Am. Chem. Soc 103 758 (1981).  12.  J.M. O'Conner and C P . Casey Chem. Rev. 87 307 (9187).  13.  D R . TylerAcc. Chem. Res. 24 325 (1991).  14.  D. A. Skoog Principles of Instrumental Analysis Saunders College 3rd ed. (1985).  15.  E. WeitzJ. Phys. Chem. 91 3945 (1987).  16.  J.M. Hollas Modern Spectroscopy Wiley (1987).  17.  D . Smith and N . G . Adams Gas Phase Ion Chemistry Vol. 2 M . Bowers ed. Academic Press (1979).  18.  B . K . Janousek and J.I. Brauman Gas Phase Ion Chemistry Vol. 2 M . Bowers ed. Academic Press (1979).  19.  P.W. Villalta and D.G. Leopold J. Chem. Phys. 98 7730 (1993).  20.  N.J. Turner, A. A. Martel and L M . Waller J. Phys. Chem. 98 474 (1994).  71  21.  N.J. Turner Masters Thesis University of British Columbia (1993).  22.  A.E. Stevens, C.S. Feigerle and W.C. Lineberger J. Chem. Phys JS 5420 (1983).  23.  A.E. Stevens-Miller, C.S. Feigerle and W.C. Linberger J. Chem. Phys. 87 1549 (1987).  24.  A.E. Stevens-Miller, C.S. Feigerle and W.C. Lineberger J. Chem. Phys. 84 4127 (1986).  25.  P.C. Engelking and W.C. Lineberger J. Am. Chem. Soc. 101 5569 (1979).  26.  A.E. Stevens-Miller, C.S. Feigerle and W.C. Lineberger J. Am. Chem. Soc. 104 5026(1982).  27.  M.A. Johnson, M.L. Alexander and W.C. Lineberger Chem. Phys. Lett. 112 285 (1984).  28.  D.H. Levy Sci. Amer. 250 96 (1984).  29.  D.H. "Levy Arm. Rev. Phys. Chem. 31 197 (1980).  30.  T.N. Kitsopolous, IM. Waller, J.G. Loeser and D M . Neumark Chem. Phys. Lett. 159 300 (1989).  31.  I.M. Waller, T.N. Kitsopolous and D M . Neumark J. Phys. Chem. 94 2240 (1990).  32.  J.M.B. Bakkery. Phys. E. 6 785 (1973).  33.  J.M.B. Bakker J. Phys. E. 7 364 (1974).  34.  W.C. Wiley and I.H. McLaren Rev. Sci. Instrum. 26 150 (1955).  35.  L. Wahlin Nucl. Instrum. Methods. 27 55 (1964).  36.  D. Wilden, P. Hicks and J. Comer J. Phys. B. 9 1959 (1976).  37.  R.J. Celotta, R. A. Bennett, J.L. Hall, M.W. Seigal and J. Levine Phys. Rev. A. 6 631 (1972).  38.  L.A. Posey, M.J. Deluca and M.A. Johnson Chem. Phys. Lett. 131 170 (1986).  39.  F. Breyer, P. Frey, H. Hotop Z Phys. A. 286 133 (1978).  72  40.  D.M. Neumark, K.R. Lykke, T. Anderson and W.C. Lineberger Phys. Rev. A. 32 1890 (1985).  41.  M.J. Travers, D.C. Cowles and G.B. Ellison Chem. Phys. Lett. 164 449 (1986).  42.  G. Herzberg Spectra of Diatomic Molecules Van Nostrand Reinhold (1950).  43.  M.W. Seigal, R.J. Celotta, J L . Hall, J. Devine and R A . Bennett Phys. Rev. A. 6 607 (1972).  44.  J.V. Coe, J.T. Snodgrass, C B . Freidhoff, K M . McHugh andK.H. Bowen J. Chem. Phys. 87 4302 (1987).  45.  RR. Corderman and J.L. BeauchampInorg. Chem. 16 3135 (1977).  46.  S.Taylor Ph. D. Thesis University of British Columbia 1992.  47.  R.N. McDonald and P L . Schell Organometallics7 1806 (1988).  48.  M.CR. Symons and SW. Bratt J. Chem. Soc. Dalton Trans. 1739 (1979).  49.  O. Crichton, A.J. Rest and D.J. Taylor J. Chem. Soc. Dalton Trans. 167 (1980).  50.  D.L. Lichtenberger, D.C. Calabro and G.E. Kellogg Organometallics 3 1623 (1984).  51.  T A . Albright and R. Hoffmann Chem. Ber. I l l 1578 (1978).  52.  F. Basolo J. Am. Chem. Soc. 88 1659 (1966).  53.  K.J. Van Der Berg, S. Ingemann and N.M.M. Nibbering Organometallics 11 2389 (1992).  54.  G. Huttner, H.H. Brintzinger and D. Neugebauer J. Organometallic Chem. 145 329 (1978).  55.  L.R. Byers and L.F. Dahl Inorg. Chem. 19 277 (1980).  56.  R.J. Celotta, R.J. Bennett and R A . Hall J. Chem. Phys. 60 1740 (1974).  57.  E.P. Grimsrud, G. Caldwell, S. Chowdbury and P. Kebarle J. Am. Chem. Soc. 107 4627 (1985).  58.  P. Kebarle and S. Chowdhury Chem. Rev. 87 513 (1987).  73  59.  J.M. Campbell Private Communication.  60.  W. Evans and J. Zink J. Am. Chem. Soc. 103 2635 (1981).  61.  P.J. Giordano and M.S. Wrighton Inorg. Chem. 16 160 (1977).  62.  G. Teixeira, T. Aviles, A.R. Dias and F. Pina J. Organometallic Chem. 353 83 (1988).  63.  J.W. Hershberger and.J.K. Kochi Polyhedron 2 929 (1983).  64.  M. Akita, N. Ishii, A. Takabuchi, M. Tanaka and Y. Moro-oka Organometallics 13 258(1994).  65.  S. Lee and N.J. Cooper J. Am. Chem. Soc. 113 716 (1991).  66.  N G . Connelly and M.D. Kitchen J. Chem. Soc. Dalton Trans. 931 (1977).  67.  I.R. Lyatifov, G.I. Gulieva, V.N. Babin, R B . Materikova, P.V. Petrovskii and E.I. Fedin J. Organometallic Chem. 326 93 (1987).  68.  Rettig, Drago J. Am. Chem. Soc. 91 3432 (1969).  69.  D C . Calabro, J.L. Hubbard, C.H. Blevans, A.C. Campbell and D.L. Lichtenberger J. Am. Chem. Soc. 103 6841 (1981).  70.  D C . Calabro and D.L. Lichtenberger J. Am. Chem. Soc. 103 6847 (1981).  71.  H. Handwerker, H. Beruda, M. Kleine and C. Zybill Organometallics 11 3542 (1992) .  72.  R.J.P. Corriu, G.F. Lanneau and B.P.S. Chauhan Organometallics 12 2001 (1993) .  73.  R.H. Fong, C-H. Lin, H. Idmoumaz and W.H. Hersh Organometallics 12 503 (1993).  74.  B.E.R. Schilling, R. Hoffmann and D.L. Lichtenberger J. Am. Chem. Soc. 101 585 (1979).  75.  J. Dye J. Phys. Chem. 88 3842 (1984).  76.  S.L. VanOrden and S.W. Buckner Polyhedron 10 2487 (1991).  77.  CRC Handbook. 56th edition. Weast (1975-76).  74  T.B. McMahon Private Communication.  75  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items