UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Absolute photoabsorption and photoionization studies of the hydrogen halides HC1, HBr, and HI, by dipole.. Dyck, Monica Karen 1997

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
[if-you-see-this-DO-NOT-CLICK]
[if-you-see-this-DO-NOT-CLICK]
ubc_1997-0230.pdf [ 5.32MB ]
Metadata
JSON: 1.0059604.json
JSON-LD: 1.0059604+ld.json
RDF/XML (Pretty): 1.0059604.xml
RDF/JSON: 1.0059604+rdf.json
Turtle: 1.0059604+rdf-turtle.txt
N-Triples: 1.0059604+rdf-ntriples.txt
Original Record: 1.0059604 +original-record.json
Full Text
1.0059604.txt
Citation
1.0059604.ris

Full Text

ABSOLUTE PHOTOABSORPTION AND PHOTOIONIZATION STUDIES OF THE HYDROGEN HALIDES HC1, HBr, AND HI, BY DIPOLE (e,e) AND (e,e+ion) SPECTROSCOPIES IN THE UV, V U V AND SOFT X - R A Y REGIONS by MONICA KAREN D Y C K B.Sc, Trinity Western University, 1994 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE  in  THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF CHEMISTRY  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA April 1997 © Monica Karen Dyck, 1997  In  presenting  degree freely  at  this  the  available  copying  of  publication  in  partial  fulfilment  of  the  University  of  British  Columbia,  I  agree  for  this  department  thesis  thesis  or . by of  reference  this  for  his thesis  and  scholarly  or for  her  of  ^ E H l ^ t R V  T h e U n i v e r s i t y o f British Vancouver, Canada  Date  DE-6  (2788)  Apnl  06,  (W  Columbia  I  further  purposes  gain  shall  that  agree  may  representatives.  financial  permission.  Department  study.  requirements  It not  be is  that  the  Library  permission  granted  by  understood be  for  allowed  an  advanced  shall for  the that  without  make  it  extensive  head  of  my  copying  or  my  written  Abstract The absolute differential oscillator strengths (cross sections) for the photoabsorption spectra of the three hydrogen halide molecules HC1 (5-280 eV), HBr (5-300 eV), and HI (5-120 eV) have been obtained by using dipole (e,e) spectroscopy, at both low (1 eV fwhm) and high (0.05-0.1 eV fwhm) resolution. Data for the valence and selected inner shells (CI 2p, Br 3d, and I 4d) have been obtained at equivalent photon energies from the U V to soft X-ray regions. The absolute oscillator strength scales were determined by use of the VTRK (S(0)) sum rule for HC1, and the dipole polarizability (S(-2)) sum rule for HBr and HI.  The value of the dipole  polarizability derived from the VTRK normalized spectrum of HC1 is in very good agreement with published experimental values. In further experiments, dipole (e,e+ion) spectroscopy has been used to investigate the molecular and dissociative photoionization of HC1 and HBr, in the valence and selected inner shell (CI 2p, Br 3d) regions. Photoion branching ratios for molecular and dissociative photoionization determined from time-of-flight mass spectra have been used together with the absolute photoabsorption  data to obtain partial photoionization differential oscillator strengths.  Photoionization efficiencies have also been obtained. A consideration of the partial differential oscillator strengths for molecular and dissociative photoionization together with the electronic state partial differential oscillator strengths (obtained from earlier dipole (e,2e) studies, Chem. Phys. 83 (1984) 391, Chem. Phys. 98 (1985) 367) provides information concerning the dipoleinduced breakdown pathways of HC1 and HBr.  ii  iii Table of Contents Abstract  ii  List of Tables  vi  List of Figures  viii  Acknowledgments  xii  Chapter 1: Introduction  1  1.1 Background  1  1.2 Conventional Photoabsorption Methods  3  1.3 Photoabsorption and Photoionization by Fast Electron Impact  6  Chapter 2: Theoretical Background  :  10  2.1 The Dipole (e,e) Method and the Bethe-Born approximation  10  2.2 Photoabsorption by Dipole (e,e) Spectroscopy  11  2.3 Photoionization by Dipole (e,e+ion) Spectroscopy  13  2.4 Absolute Photoabsorption Oscillator Strengths (Cross sections)  16  Chapter 3: Experimental Methods  19  3.1 Introduction  19  3.2 Low Resolution Dipole (e,e) and Dipole (e,e+ion) Spectrometer  20  3.3 High Resolution Dipole (e,e) Spectrometer  24  Chapter 4: Photoabsorption and Photoionization Studies of Hydrogen Chloride  28  4.1 Introduction  28  4.2 Results and Discussion  32  4.2.1 Electronic Structure  32  iv  4.2.2 Photoabsorption Oscillator Strengths  33  4.2.2.1 Low-Resolution Photoabsorption Measurements  33  4.2.2.2 Valence Shell High Resolution Photoabsorption Measurements for HC1  38  4.2.2.3 Inner Shell (Chlorine 2p) High-Resolution Photoabsorption Measurements for HC1  43  4.2.3 Photoionization Studies of HC1 in the Valence Regions  46  4.2.4 Dipole-Induced Breakdown of HC1  53  4.2.5 Photoionization Studies of HC1 in the CI 2p Inner Shell  61  Chapter 5: Photoabsorption and Photoionization Studies of Hydrogen Bromide  75  5.1 Introduction  75  5.2 Results and Discussion  77  5.2.1 Electron Structure  77  5.2.2 Photoabsorption Oscillator Strengths  78  5.2.2.1 Low-Resolution Photoabsorption Measurements  78  5.2.2.2 Valence Shell High Resolution Photoabsorption Measurements for HBr 5.2.2.3 Br 3d Inner Shell High Resolution Photoabsorption  82  Measurements for HBr  87  5.2.3 Photoionization Studies of HBr in the Valence Shell  87  5.2.4 Dipole-Induced Breakdown Pathways of HBr  97  5.2.5 Photoionization Studies in the Br 3d Inner Shell of HBr Chapter 6: Photoabsorption Studies of Hydrogen Iodide 6.1 Introduction  104 115 115  6.2 Results and Discussion  116  6.2.1 Electronic Structure  116  6.2.2 Photoabsorption Oscillator Strengths  117  6.2.2.1 Low-Resolution Photoabsorption Measurements of HI. 117 6.2.2.2 High Resolution Valence Shell Photoabsorption Measurements of HI  120  6.2.2.3 High Resolution I 4d Inner Shell Photoabsorption Measurements  123  Chapter 7: Concluding Remarks  128  Appendix: References  131  List of Tables 4.1 Integrated oscillator strengths for selected energy regions of the valence and CI 2p inner shell photoabsorption spectrum of HC1  ...42  4.2 Appearance potentials (in eV) for molecular and fragment ions resulting from photoionization of HC1  47  4.3 Absolute differential photoabsorption oscillator strengths of HC1 for the valence and CI 2p inner shell regions (1 eV fwhm)  66  4.4 Photoion branching ratios for the molecular and dissociative photoionization of HC1, as a percentage of total photoionization  69  4.5 Absolute partial differential oscillator strengths for the molecular and dissociative photoionization of HC1  71  4.6 Partial photoionization differential oscillator strengths for production of the electronic ion states of HC1  73  5.1 Comparison of integrated oscillator strengths (cross sections) for HBr reported in the literature with selected energy ranges from the present work measurements..86 5.2 Integrated oscillator strengths for selected energy ranges in the Br 3d inner shell of HBr, with and without estimated contributions from the valence shell continuum 5.3 Appearance potentials for the molecular and dissociative ions produced from HBr  89 90  5.4 Absolute differential photoabsorption oscillator strengths for the valence shell and Br 3d inner shell regions of HBr (1 eV fwhm) 5.5 Photoion branching ratios for the molecular and dissociative photoionization of HBr,  106  as a percentage of total photoionization  109  5.6 Absolute partial differential oscillator strengths for the molecular and dissociative photoionization of HBr  Ill  5.7 Partial photoionization differential oscillator strengths for production of electronic ion states of HBr  113  6.1 Integrated oscillator strengths (cross sections) for selected energy regions in the valence shell obtained from the present measurements of HI, with comparison to those reported in the literature  122  6.2 Integrated oscillator strengths (cross sections) for selected energy ranges in the I 4d inner shell of HI, with and without estimated contributions from the valence shell continuum  125  6.3 Absolute differential photoabsorption oscillator strengths for the valence shell and 14d inner shell regions of HI (1 eV fwhm)  126  viii  List of Figures 1.1 Virtual photon field induced in a molecular target by distant collision of a fast electron ..7 3.1 Schematic of the low resolution dipole (e,e) and dipole (e,e+ion) spectrometer  21  3.2 Schematic of the high resolution dipole (e,e) spectrometer  26  4.1 Absolute low resolution (1.0 eV fwhm) photoabsorption oscillator strength spectrum for HCI  34  4.2 Comparison of low resolution (1.0 eV fwhm) and high resolution (0.05 eV fwhm, convoluted with 1 eV fwhm) valence shell photoabsorption measurements of HCI obtained in the present work  36  4.3 Comparison of low resolution absolute photoabsorption oscillator strengths for HCI obtained in the present work and by Daviel et al. [10]  37  4.4 Absolute high resolution (0.05 eV fwhm) and low resolution (1.0 eV fwhm) photoabsorption oscillator strengths (cross sections) for the valence shell of HCI... 37 4.5 Comparison of the present absolute cross sections in the valence shell photoabsorption of HCI with digitized data from the literature  41  4.6 Absolute photoabsorption oscillator strengths (cross sections) for the CI 2p inner shell of HCI  44  4.7 Time-of-flight (TOF) mass spectra for valence shell photoinization of HCI at 20.0 and 80 eV energy loss  48  4.8 Photoion branching ratios (1 eV fwhm) for molecular and dissociative ions from photoionization of HCI, as a percentage of total photoionization 4.9 Comparison of photoion branching ratios (1 eV fwhm) obtained in the present work  49  and by Daviel et al. [10]  50  4.10 Absolute partial oscillator strengths for molecular and dissociative photoionization of HC1 from 6 to 265 eV at 1 eV fwhm resolution  52  4.11 Absolute partial photoionization oscillator strengths for production of the XTI, A 2 , 2  and B 2 electronic ion states of HC1 2  +  +  55  4.12 Absolute partial oscillator strengths for molecular and total dissociative photoionization of HC1, with best linear combination fits of contributions from the electronic ion state partial oscillator strengths  56  4.13 Absolute partial oscillator strengths for dissociative photoionization of HC1, with best fit linear combinations of contributions from the electronic ion state partial oscillator strengths  58  4.14 Dipole ionic photofragmentation breakdown pathways of HC1 in the V U V and soft X-ray regions  60  4.15 Time-of-flight (TOF) mass spectra at selected energy losses in the valence and CI 2p inner shell regions of HC1  62  5.1 Absolute low resolution photoabsorption oscillator strengths (cross sections) for the valence and Br 3d inner shell regions of HBr  79  5.2 Comparison of the low resolution (1 eV fwhm) absolute photoabsorption measurements obtained in the present work with: (a) high resolution (0.05 eV fwhm, convoluted with 1 eV fwhm) spectrum obtained in the present work, and (b) measurements reported by Brion et al. [11] 5.3 Absolute total photoabsorption oscillator strengths for the valence shell of HBr with  81  X  high resolution (0.05 eV fwhm)  83  5.4 Comparison of absolute photoabsorption spectra obtained in the present work with reported spectra from the literature  85  5.5 Absolute photoabsorption oscillator strengths for the Br 3d inner shell region of HBr, obtained with high resolution (0.1 eV fwhm)  88  5.6 Typical peak fits used to resolve isotopic HBr , Br peaks in the HBr time-of-flight +  +  mass spectra, at equivalent photon energies of (a) 12.0 eV, and (b) 20.0 eV. 92 5.7 Time-of-flight mass spectra in the valence shell of HBr at energy losses of 20 and 51.5 eV  93  5.8 Photoion branching ratios for the molecular and dissociative photoionization of HBr, as a percentage of total photoionization  94  5.9 Comparison of photoion branching ratios obtained in the present work with those reported in the previous dipole (e,e+ion) study by Brion et al. [11]  96  5.10 Absolute partial photoionization oscillator strengths for the molecular and dissociative ionic states produced by photoionization of HBr  98  5.11 Electronic ion state partial oscillator strengths obtained from photoelectron dipole (e,2e) branching ratios reported by Brion et al [11] and the presently reported photoabsorption oscillator strengths  99  5.12 Absolute partial photoionization oscillator strengths for molecular and total dissociative photoionization of HBr, with best fit linear combinations of contributions from the electronic ion state partial oscillator strengths  101  5.13 Absolute partial photoionization oscillator strengths for the dissociative ions Br and +  H , with best fit linear combinations from electronic ion state partial +  oscillator strengths  102  5.14 Dipole ionic photofragmentation breakdown pathways of HBr in the V U V and soft X-ray regions  103  5.15 Time-of-flight mass spectra of HBr at selected energies in the inner valence and Br 3d inner shell regions (1 eV fwhm)  105  6.1 Absolute photoabsorption differential oscillator strengths (cross sections) for the valence and I 4d inner shell regions of HI  118  6.2 Comparison of low resolution (1 eV fwhm) and high resolution (0.05 eV fwhm, convoluted with 1 eV fwhm) absolute photoabsorption oscillator strengths forHI  119  6.3 Absolute photoabsorption oscillator strengths (cross sections) for the valence shell of HI, obtained with high resolution (0.05 eV fwhm)  121  6.4 Absolute photoabsorption oscillator strength spectrum of the I 4d inner shell region of HI, obtained with high resolution (0.1 eV fwhm)  124  xu  Acknowledgements I would like to sincerely thank my research supervisor, Dr. G.E. Brion, for his assistance, support, expertise, and patience through the course of my research and the writing of this thesis. Special thanks are also due to Dr. Glyn Cooper and Dr. Terry N. Olney who taught me what I needed to know about the technique of dipole (e,e) spectroscopy, as well as giving me a lot of help along the way and answering many questions. I wish to also thank the rest of the research group for their support and advice: J. Au, J. Neville, J. Rolke, and Dr. Y. Zheng. Funding for this research project was provided by the Natural Sciences and Engineering Research Council (NSERC) of Canada.  I also gratefully acknowledge the receipt of a  McdDowell Graduate Fellowship. Finally, I wish to thank my parents for their constant, wholehearted support and encouragement. This thesis is dedicated to them.  Chapter 1 Introduction 1.1 Background A quantitative knowledge of the primary and secondary processes taking place in radiation chemistry, physics, and biology is essential for understanding and predicting the interaction of energetic radiation with matter. The primary step involves absorption of a photon by the molecule, followed by decomposition (i.e. ionization, dissociation, fluorescence, etc.). Above ~20 eV in the vacuum ultraviolet, decomposition is dominated by molecular ionization and ionic photofragmentation processes, in which one or more positively charged ions are formed per photon absorbed. In particular it is necessary to know not only the energies of the electronically excited and ionized states of atoms and molecules, but also the associated transition probabilities [cross sections (a) or oscillator strengths (df/dE), where a (megabarns) = 109.75(df/dE)eV" ]. 1  Such photoabsorption and photoionization data are useful in a variety of applications. Important areas requiring such absolute measurements include radiation-induced decomposition, dosimetry, radiation therapy and health physics, upper atmosphere modelling, space chemistry and physics, environmental protection, fusion, plasmas, electron and X-ray microscopy, the nuclear industry and indeed any area where energetic radiation or particles are to be found [1]. Dipole electron impact studies have provided much useful information (examples of which can be seen in the following references [2, 3, 4, 5, 6, 7]) regarding the electronic processes that occur within atoms or molecules.  These studies give absolute photoabsorption and  photoionization spectra, and thus absolute optical oscillator strengths (transition probabilities) over wide energy ranges, encompassing the valence and inner-shell regions. In the present work results of high impact energy electron energy loss spectroscopy (EELS) experiments in the dipole 1  Chapter 1. Introduction  2  regime are presented for the photoabsorption, photoionization and ionic photofragmentation of the hydrogen halides HCI, HBr, and HI.  Dipole studies of the photoabsorption and  photoionization of HF have been reported earlier [8, 9]. Traditionally spectroscopic investigations have used direct optical techniques to measure optical oscillator strengths for photoabsorption and photoionization processes.  However,  application of direct optical methods has proven to be somewhat limited, especially for studies of absolute transition probabilities in the far UV, vacuum UV, and soft X-ray regions of the spectrum. These regions are important as they cover formation of most of the electronically excited and ionized states of the valence shell as well as all inner shell transitions. The reasons for these limitations will be discussed in sections 1.2 and 1.3.  A more recent advance in the  measurement of absolute photoabsorption and photoionization oscillator strengths has been the use of inelastic collisions of high energy electrons at zero degree scattering angle [1, 2]. In electron impact experiments with sufficiently high impact energy and negligible momentum transfer, dipole selection rules operate and the excitation or ionization cross section is directly related to the optical oscillator strength.  Therefore such dipole electron impact techniques,  utilizing electron energy loss spectroscopy, can be used to determine the optical oscillator strengths (cross sections) for photoabsorption and photoionization of atoms or molecules as a function of electron energy loss, which is equivalent to the photon energy [1]. Why this is possible and how it is accomplished will be described further in sections 1.3, 2.1 and 2.2. The basic function and features of the dipole (e,e) and dipole (e,e+ion) instruments used to obtain this data will be described in chapter 3. Finally, in chapters 4-6 the experimental results for the three hydrogen halide molecules HCI, HBr, and HI will be reported and discussed, comparing the  Chapter 1. Introduction  3  present work with previously published data, in those cases where such information is available. Very little absolute photoabsorption and photoionization data is available for these molecules using direct optical techniques.  Much earlier low resolution dipole (e,e) and (e,e+ion)  measurements for HC1 and HBr [10, 11] over limited energy ranges are possibly subject to appreciable errors due to limitations in the instrumentation and measurement precedures used at the time. Absolute photoabsorption oscillator strength (cross section) data have been obtained for the valence and selected inner shell processes for the three hydrogen halide molecules HC1, HBr, and HI in the UV, vacuum UV, and soft X-ray energy regions at high resolution. Photoionization and ionic photofragmentation oscillator strength (cross section) data have also been obtained over extended energy ranges for HC1 and HBr from the absolute photoabsorption data and photoion branching ratios determined from time-of-flight (TOF) mass spectra obtained in the valence and inner shell regions of each molecule. 1.2 Conventional Photoabsorption Methods Optical methods of spectroscopy, such as Beer-Lambert law photoabsorption, are capable in principle of giving reliable and accurate absolute measurements. However such information is limited and typically available only over short ranges at lower photon energies. Prior to the advent of synchrotron radiation, the principal limitation was the shortage of suitable continuum light sources at energies above 20 eV. Conventional discharge light sources using hydrogen and noble gas continua give useful but relatively weak structured continua up to ~20 eV [12]. In the past two decades the growing availability of tuneable synchrotron radiation, long viewed as an unwanted by-product of particle acceleration, has provided an intense photon source with a continuum stretching from the IR through the U V and vacuum U V and up to the hard X-ray  Chapter 1. Introduction  4  region, in the case of high GeV machines. Synchrotron radiation sources are giving some impressive spectroscopic results, however the access to such sources is still limited, and the technique  presents  other  difficulties,  particularly for  absolute photoabsorption  and  photoionization measurements. In particular, direct optical studies are subject to the problems involved in the dispersion (i.e. energy selection) of short wavelength radiation. For example, mirrors and gratings, which are necessary for the monochromation used in optical techniques, suffer from very low reflectivity at shorter wavelengths (i.e. photon energies above 20-30 eV). This severely attenuates the usable monochromated photon intensity in spite of the high flux from the synchrotron source. However, it should be noted that significant higher photon fluxes have very recently become available with Third Generation sychrotron radiation sources such as the Advanced Light Source at the Lawrence Berkeley Laboratory [13].  Other difficulties in using  V U V and soft X-ray radiation include the presence of higher order radiation (i.e. order overlapping), stray light corrections and light absorption by optical components, particularly in the carbon K region due to surface contamination [14]. Furthermore, in order to compensate for low light fluxes, energy resolution has often been sacrificed for sufficient intensity in optical work at shorter wavelengths [1]. A further difficulty in absolute optical measurements is the need to determine accurately the target density of the molecules and pathlength of the absorption cell, which is usually windowless in the V U V and soft X-ray regions. Optical measurements can suffer from a further extremely serious problem in determining absolute oscillator strengths (cross sections) for discrete transitions.  In particular the  measurements typically require the use of the Beer-Lambert law, which relates absorption to the ratio of the intensities of the incident to transmitted light. J  This involves a logarithmic  Chapter 1. Introduction  5  transformation, and for discrete transitions the resulting absorption cross sections can suffer from the problem of "line saturation" if the spectrometer bandwidth is not significantly narrower than the natural linewidth of the narrowest line in the spectrum [3, 15]. Line saturation errors, which can in some cases be appreciable [3], result in cross-sections which are too small. In terms of the intensity of the incident radiation, I„ , the Beer-Lambert law states that the intensity of the transmitted radiation, I, is given by I=I e~  (1.3.1)  MO/  0  where n is the number of molecules per cm , a is the photoabsorption cross section (cm ) of the 3  2  molecule, and / is the pathlength in cm. Therefore, to obtain a, the following logarithmic transform must be performed: o=-\n^ nl  (1.3.2)  I  where nl is usually referred to as the number density of the target molecules. The actual observed shape and height of any absorption peak, as well as the ratio (I/I ) depends on the bandwidth, or 0  energy resolution of the instrument, and therefore at each energy the natural log of the ratio will vary with resolution. Thus for an absorption peak corresponding to a discrete excitation the true value of a can, in principle, only be obtained if the instrument has infinite resolution, which requires the bandwidth to be equal to zero. Practically, this is obviously impossible, and this situation presents a further limitation of optical spectrometers in some situations. Ideally, in order to obtain a reasonably accurate absolute cross section, the instrumental bandpass should be significantly narrower than the natural linewidth of all discrete excitations in the spectrum of interest.  The largest errors occur for transitions with high cross sections and very narrow  Chapter 1. Introduction  6  linewidths [3, 6]. These problems are particularly likely to occur at shorter wavelengths in the V U V and soft X-ray regions, since here the monochromated light flux can be quite low, and as a result the slits in the spectrometer are often opened up to compensate, resulting in lower resolution and greater bandpass. Care should therefore be taken in obtaining the optical cross sections, especially in the case of very strong, narrow bands, as they are more prone to the so called "line saturation" effects caused by the Beer-Lambert law. It is possible to reduce these errors by making the Beer-Lambert law measurements as a function of pressure and extrapolating to zero pressure [3, 16]. However this procedure is tedious and can still result in appreciable errors since the error bars are largest at the lowest pressures [16]. 1.3 Photoabsorption and Photoionization by Fast Electron Impact Dipole electron impact techniques avoid the difficulties of limited photon energy range, order overlapping, stray light, and low reflectivity at short wavelengths which are problematic in direct optical methods for photoabsorption measurements. In addition, such electron scattering techniques do not require path length or number density determinations for absolute measurements.  The problem of "line saturation" when using the Beer-Lambert law is also  avoided in dipole electron impact experiments using EELS since the measurements involve only the determination of the number of scattered electrons at each energy loss, and no logarithmic transform is involved. In 1935, using the Born approximation, Bethe showed [17] that for inelastic fast electron impact on an atomic or molecular target there is a quantitative relationship (see section 2.1) between the differential electron scattering cross section and the optical oscillator strength, in the  Chapter 1. Introduction  1  E(t)  TIME  I(V)  FREQUENCY  Figure 1.1. Virtual photon field induced in a molecular target by distant collision of a fast electron.  Chapter 1. Introduction  8  limit of zero momentum transfer (i.e. at high impact energy and zero degree scattering angle). Such fast impacting electrons provide an excellent simulation of a tuneable light source due to their ability to induce a virtual photon (dipole) field in the target molecule. As a fast electron passes by, the target experiences a sharp electric field pulse in the time domain (see Figure 1.1). If this function is Fourier transformed into the frequency domain, the result is a continuum composed of all frequencies at equal intensities. In reality, the initial pulse has a finite, though narrow, width, and the "flat" intensity falls off at higher frequencies. However, under appropriate kinematic conditions a sufficiently wide spectral range of equal intensity can be achieved for use in the laboratory. Continuous energy transfer to the target is possible over essentially a very large portion of the electromagnetic spectrum. In dipole electron energy loss spectroscopy (EELS) experiments the continuously variable energy loss is analogous to photon energy (see section 2.2). An electron beam is readily produced by thermionic emission and the energy losses of the inelastically scattered electrons can be continuously selected by means of a linear ramp potential and an electrostatic electron analyzer, thus simulating photoabsorption using a tuneable source of radiation. Since the intensity of the "pseudo photons" produced by passing fast electrons is constant with energy loss E, there is a "flat" virtual photon source, which therefore gives the correct relative spectral photoabsorption shape [2]. Whereas optical methods require several different monochromators to cover sufficiently wide ranges of wavelength, dipole electron impact methods simply require scanning of the energy loss in a single spectrometer to cover a very wide spectral range. The target density, or pressure, is kept constant and sufficiently low so as to avoid multiple scattering processes. No determination of the absolute target density is required since absolute scales are determined using the S(0) or S(-2) sum rules as described in section 2.4. Thus  Chapter 1. Introduction  9  quantitative photoabsorption, photoionization and photoelectron (PES) spectroscopies can be simulated by appropriate dipole electron impact methods using EELS [2]. Limitations to the technique of electron energy loss measurements by dipole (e,e) spectroscopy include the significant drop-off of the signal at higher energies, due to the ~ E  3  dependence of the Bethe-Born conversion factor (see section 2.2). This effectively imposes an upper limit to the energy range accessible by this method. A second difficulty is the presence of stray electrons and background gases in the measured energy-loss spectrum. This problem can be avoided by careful and correct background subtraction. Thus far, the determination of absolute photoabsorption data by dipole (e,e) methods is limited to gas phase experiments. Dipole electron impact techniques, used within these limitations, still provide an excellent method of obtaining data for the photoabsorption and photoionization of a variety of atomic and molecular targets.  Chapter 2 Theoretical Background 2.1 The Dipole (e,e) Method and the Bethe-Born approximation Bethe's work using the Born theory of particle scattering provides the basis for dipole electron scattering studies [17, 18]. An important result of the Bethe-Born theory is the quantified relationship between the electron impact differential inelastic scattering cross section do/dE and the generalized differential oscillator strength df/dE(K). do  2 k  dE  E k  df(K)  1  ^  K  >  (2.1.1)  dE  2  o  where k>, k , and K are the incident, scattered and transferred momenta respectively, and E is n  the energy loss. Bethe further showed that the generalized oscillator strength could be expanded into a series of terms, the first of which is the optical or dipole oscillator strength, and the higher (non-dipole) terms involve higher powers of K, the momentum transfer. + AK +BK*+... dE dE df(K) df , , _ From the cosine rule,  (2.1.2)  2  0  4  k  Q  K = ko + k„ - 2k k cos6 2  2  (2.1.3)  2  0  n  It can be seen from equation 2.1.3 that K approaches zero at high impact energy E (i.e. ko ~ k ) 2  0  n  and zero scattering angle (6 = 0°). Therefore in the limit of zero momentum transfer K, the generalized oscillator strength becomes equal to the optical oscillator strength and dipole selection rules are operative. Under these (dipole) conditions, 10  Chapter 2. Theoretical Background  ^  11  _.K_. 2  ,da_  2 k  dE  E =  K  dE  n  (2.1.4)  The optical oscillator strength is related to the absolute photoabsorption cross section a  ph  abs  by the  relation o- abs= ph  where  a  pha  bs  (2.1.5)  109.75(dfo/dE)  is in megabarns (1 Mb = 10~ cm ) and dfo/dE has units of eV" . Thus the Bethe18  2  1  Born theory shows that dipole transitions are induced in the target by fast electrons, and also that their optical oscillator strengths (i.e. transition probabilities) can be obtained by a simple kinematic conversion of the corresponding differential electron scattering intensities. The kinematic quantity (2/E)(k /k )(l/K ) in equation 2.1.4 is independent of the target species and is known as the 2  n  0  Bethe-Born conversion factor. In practice this factor must be integrated over the small range of momentum transfers within the finite acceptance angle subtended about zero degrees by the spectrometer entrance slit. The Bethe-Born factors for each of the two instruments used in the present work have been detailed earlier [3, 19, 20]. This allows quantitative optical spectroscopy to be carried out using dipole electron impact methods. 2.2 Photoabsorption by Dipole (e,e) Spectroscopy In direct photoabsorption experiments, when a molecule A B absorbs a photon of energy E = hv, the process results in an excited state AB*: AB + hv -* A B * E  (2.1.1)  The probability for these processes (including all discrete and continuum states), as a function of energy, E, is given by the absolute total photoabsorption cross section. The same molecular excitation process can also occur in an inelastic electron scattering (i.e. dipole (e,e)) experiment  Chapter 2. Theoretical Background  12  using an incident electron beam of fixed high energy E . The process can be represented by: 0  e + A B -> A B * E  "f"  (2.1.2)  ^ scatt  ( E - E)  0  0  where the incident electron (with impact energy E ) imparts an energy E to the target during the 0  excitation, and is scattered with a final energy ( E - E ) . Under the conditions used in the present 0  work (i.e. zero mean scattering angle and high incident electron energy) the kinetic energy transferred to the target molecule is small (< 1 meV), and therefore it can be seen from equations 2.1.1 and 2.1.2 that the energy loss, E , of the electron is equivalent to the photon energy, hv, in dipole (e,e) spectroscopy. With this equivalence in mind all energy scales in the present work are labelled as "photon energy." As shown in section 2.1, in the limit of zero momentum transfer to the target molecule, dipole-allowed transitions are excited by fast electron impact. As the momentum transfer is increased, higher multipole transitions can be excited [18]. Under the high impact energy and zero degree mean scattering angle conditions of dipole (e,e) spectroscopy the momentum transfer is extremely small and the contributions from multipole transitions are negligible (< ~1%). Also, as described in section 2.1, the photoabsorption oscillator strength is related to the differential electron scattering cross section by the kinematic Bethe-Born factor (E/2)(ko/k )K , which can be 2  n  shown to be essentially proportional to E at zero degree scattering angle in the hypothetical case 3  of infinitely small acceptance angle [2]. In practice, integration over the range of K in the finite 2  acceptance angle slightly modifies this relationship [3, 19, 20]. It can be seen that the electron scattering signal will fall off more rapidly with energy loss than does the photoabsorption cross section by a factor of ~ E (or typically ~ E ' to E 3  2  5  2 , 7  in the real case of finite acceptance angles).  Therefore great care must be taken to account for any background due to stray electrons,  Chapter 2. Theoretical Background  13  impurities or background gases. The kinematic Bethe-Born factor is a function of the energy loss E, and is the same for all targets, but is spectrometer-dependent in that it depends on the acceptance angle of the particular spectrometer used and on the impact energy E . The Bethe-Born factor can be determined 0  directly from an analysis of the scattering geometry of the spectrometer [3, 20], or it can be derived from the ratio of the Bethe-Born converted energy-loss signal to a known accurate absolute photoabsorption oscillator strength over a wide range of energies in the smooth continuum region [3,19]. Thus the technique of dipole (e,e) spectroscopy can be used to measure photoabsorption oscillator strengths.  The notation "dipole (e,e)" simply indicates that the conditions of the  experiment determine that effectively only dipole-allowed transitions are studied, and that the process involves an incident electron and the release of a scattered electron. The dipole (e,e) technique uses electron energy loss spectroscopy, EELS, for the detection of scattered electrons with specific energy loss. In an EELS spectrometer the energy loss, E, can be continuously varied by changing a linear ramp voltage. A programmable DC power supply may therefore be used as the analogue of a scanning monochromator. A single power supply can scan a very wide energy range, from the IR to the X-ray region, thus avoiding the necessity of using several complex optical monochromators.  In summary, dipole (e,e) spectroscopy simulates tuneable energy  photoabsorption, with the energy loss being the equivalent of the photon energy [1, 2, 3]. 2.3 Photoionization by Dipole (e,e+ion) Spectroscopy If the photon energy hv supplied to the molecule AB is above the first ionization potential, then an electron, e" j, can be produced and a positively charged molecular or fragment ion will be e  Chapter 2. Theoretical Background  14  formed: A B + hv -» A B + e"  Molecular Photoionization  +  ej  -*• A + B + e"j +  e  (2.2.1)  Dissociative Photoionization  E  The same processes can also occur in an inelastic electron scattering (i.e. dipole (e,e+ion)) experiment, where a fast incident electron beam of energy E and energy loss E , can cause the Q  formation of a molecular or fragment ion, an ejected electron e"j , and a fast scattered electron e  C scatt •  e" + A B -» A B + e"j + e" +  e  scatt  -» A + B + e"j +  e  Eo  +e" t scat  Molecular photoionization  (2.2.2)  Dissociative photoionization  ( E - E) 0  Thus the inelastic electron scattering experiment simulates photoionization with the energy loss E of the incident electron being used in the formation of A B + e"j (or A + B + e" j). +  +  e  e  The total photoionization oscillator strength distribution is made up of the sum of the partial photoionization oscillator strengths for the production of the molecular and all dissociative ions. In photoionization mass spectrometry, or in the equivalent electron-impact-based technique of dipole (e,e+ion) spectroscopy used in the present work, mass spectra are measured as a function of the equivalent photon energy (i.e. the energy loss) in order to determine what products of photoionization are formed at the energies absorbed by the target molecule. A useful quantity is the photoion branching ratio, BR'°", which represents the intensity of a given ion relative to the total number of ions formed at any particular photon energy. It is determined by integration of the peaks in the baseline-subtracted mass spectra, and varies as a function of the photon energy. The partial photoionization oscillator strength (PPOS) for the production of a  Chapter 2. Theoretical Background  15  molecular or dissociative ion, x, (df^/dE) is given by the triple product of the photoion branching ratio for the ion, the photoionization efficiency (r\j), and the absolute photoabsorption oscillator strength, determined in the present work from dipole (e,e) spectroscopy: JJ* ion  Jjyabs  ^— = BR dE  ion  x  • ri. •  dE  V  (2.3.3) '  The photoionization efficiency, r|, , is the probability that an ion is formed when a photon is absorbed.  In the present dipole (e,e+ion) work, time-of-flight mass spectroscopy is used to  determine the photoion branching ratios, by coincident detection of the fast scattered energy loss electron and the molecular ion produced in the dipole (e,e+ion) experiment. The photoionization efficiency T|; is also determined.  The total photoionization differential oscillator strength  distribution is equal to:  dE  ^ dE  dE  and is the sum of the PPOS's for the molecular and each dissociative ion, x. The related techniques of photoelectron spectroscopy (PES) and dipole (e,2e) spectroscopy allow the determination of how the production of electronic states, j, of the molecular ion varies with the energy absorbed by the target [1, 2]. The electronic ion state (i.e. photoelectron) branching ratios can be determined from the background subtracted photoelectron (or dipole (e,2e)) spectrum in a manner analogous to that for photoionization branching ratios, and hence partial photoionization oscillator strengths for the production of electronic ion states, j, can be determined: fion  d  d  = BR  wn  dE  1  -ri. '  f  a  b  dE  s  (2.3.5)  Chapter 2. Theoretical Background  16  Thus the total photoionization differential oscillator strength distribution can also be represented by the sum of the partial photoionization oscillator strengths for the formation of each of the electronic ion states j of the molecular ion. ^  dfr  dJ  dE  *f dE  ^ _ u  dE  The dipole-induced breakdown pathways for a molecule can therefore be determined from an examination of these two complementary manners in which the total photoionization oscillator strength distribution can be partitioned (i.e. equations 2.3.4 and 2.3.6), together with a consideration of ionization and appearance potentials and the associated Frank-Condon widths for production of the electronic and ion states. The latter quantities can be estimated from high resolution PES measurements. 2.4 Absolute Photoabsorption Oscillator Strengths (Cross sections) The "flat" virtual photon field induced in the target atom or molecule allows the correct shape of the spectrum to be obtained in dipole (e,e) spectroscopy as discussed above in sections 1.4 and 2.1. Thus, by an appropriate process of normalization of the Bethe-Born converted relative energy loss spectrum, the absolute intensities of the photoabsorption spectrum can be determined. In principle, normalization can be achieved at a single energy loss or photon energy to a known optical absorption cross section. However, this latter method requires sufficiently accurate optical data from the literature, and this is not always available. The normalization is carried out independently in the present work by the use of one or other of two dipole sum-rules [7, 18, 21], which avoids the very difficult measurement of absolute target densities which are required in direct optical absolute photoabsorption determinations.  and photoionization cross section  The Thomas-Reiche-Kuhn (TRK), or S(0), sum rule equates the total  Chapter 2. Theoretical Background  17  photoabsorption oscillator strength, integrated over all discrete and continuum states of the valence and all inner shells, to the number of electrons in the target molecule. Since in the dipole (e,e) method all inner shells cannot in general be measured, the valence shell modification (VTRK) of the S(0) sum rule is used [22]. In the VTRK procedure Bethe-Born converted the valence-shell electron energy loss spectrum is obtained out to high energy and the small remaining intensity (typically 5 - 10%) to infinite energy is estimated by extrapolation of a polynomial curve fitted to the higher energy region of the valence shell measurements (typically from 100 to 200 eV). Assuming shell separation, the total area is then normalized to the number of valence shell electrons, N ;, plus a small correction, \S , for Pauli excluded transitions to the already occupied Vfl  PE  states [22]. Reasonably accurate values of N  P£  are available from Hartree-Slater calculations  [23]. The valence shell, or VTRK sum-rule is given by (2.4.2) where E„ is the first excitation threshold. Using the VTRK procedures an absolute scale can typically be obtained with an accuracy of better than ~5% [2, 7]. The VTRK sum rule normalization procedure requires the valence electron energy loss spectrum to be obtained up to relatively high energies (typically at least 200 eV), well into the valence shell photoabsorption continuum. For molecules such as HBr or HI, the VTRK sum rule becomes difficult to implement because lower-energy inner-shell transitions (eg. Br 3d and I 4d) can interfere [7]. In such cases other methods of absolute scale normalization must be used. The accuracy of the absolute differential oscillator strength scales established using the VTRK sum rule is estimated to be within ±5% [7]. It has recently been shown [7] that the S(-2) sum rule, which makes use of published experimental and theoretical dipole polarizabilities, can also be used  Chapter 2. Theoretical Background  18  to normalize relative differential oscillator strength spectra or, alternatively, as a check of the absolute scales established by other methods, such as the VTRK sum-rule [7]. The S(-2) sumrule is given by -2  (2.4.2) where cu is the static dipole polarizability and E is the Hartree atomic unit of energy. It can be H  seen that due to the 1/E weighting in equation 2.4.2 the dipole polarizability method of 2  normalization depends overwhelmingly on the low energy region (typically <. 100 eV) of the valence shell spectrum, and thus has the advantage of relying on the region with the highest oscillator strength and typically the best signal-to-noise ratio. The S(-2) sum rule is used in the present work for normalization of the photoabsorption spectra of HBr and HI since the low lying transitions from the Br 3d and I 4d inner shells preclude the use of the VTRK sum rule. In addition, the accuracy of the oscillator strength scale in the VTRK sum rule normalized photoabsorption spectrum of HCI is evaluated by comparing the cu value derived using the S(-2) sum rule with directly measured and calculated values.  The accuracy of measured dipole  polarizabilities are generally considered to be accurate to within ±1-2%.  Chapter 3 Experimental Methods 3.1 Introduction Two instruments were used to collect the data: a low resolution (1 eV full-width halfmaximum (fwhm)) dipole (e,e) and (e,e+ion) spectrometer, and a high resolution (0.048 eV fwhm) dipole (e,e) spectrometer. The energy loss data in each case were converted to relative photoabsorption cross sections by use of the Bethe-Born theory (see section 2.1).  Each  instrument has a known energy-loss-dependent Bethe-Born factor [3, 19], so the conversion is a simple matter of multiplying the intensity in the energy loss spectrum by the value of this factor (at each energy loss) to obtain relative photoabsorption data. The relative spectrum is then made absolute by use of sum rules (see section 2.4). In the case of HC1 use of the VTRK S(0) or the S(-2) sum rules gives absolute scales which are in excellent agreement. For HBr and HI, where inner shell processes (Br 3d and I 4d respectively) at lower energy interfere with the curve fitting and extrapolation, use of the TRK sum rule is not possible, and the spectra in these cases were made absolute by using the S(-2) sum rule and known values of the dipole polarizabilities. Partial photoionization oscillator strengths and photofragmentation branching ratios were obtained for HC1 and HBr on the dipole (e,e+ion) instrument for valence and inner shell processes. The branching ratios were obtained from time-of-flight mass spectral studies.  Partial oscillator  strengths for molecular and dissociative photoionization were obtained by multiplying the branching ratios for each singly or doubly charged ionic species by the total photoabsorption oscillator strength and by the photoionization efficiency at each equivalent photon energy.  19  Chapter 3: Experimental Methods  20  3.2 Low Resolution Dipole (e,e) and Dipole (e,e+ion) Spectrometer The low resolution dipole (e,e) spectrometer (figure 3.1) is actually the non-coincident forward scattering portion of a dipole (e,e+ion) spectrometer which was initially built at the FOM institute in Amsterdam [20, 24, 25]. This instrument was moved to UBC in 1980, and further modified and refined [3, 7, 19, 26, 27] for use in absolute photoabsorption and photoionization studies. It has since been used with great success in studying a variety of molecules [2, 3, 4, 5, 7, 19]. The details of the construction and operation of the low resolution instrument can be found in a number of earlier publications [1, 24, 25, 26, 27]. An electron beam is produced by an indirectly heated barium oxide cathode in a black and white television gun (Phillips 6AW59) which is floated at a potential of -4 kV. The electrons are focussed with an einzel electron lens into the collision chamber which is floated at a potential of +4 kV. This provides acceleration of the incident electrons to 8 keV. The electron beam (typically ~2xl0" amps) inelastically scatters 9  off the target molecule in the collision chamber, to excite electronic transitions to both neutral and ionic states, and in so doing loses energy. The forward scattered electrons (solid angle of acceptance of 1.4xl0" sr) are selected by an angular selection aperture and then decelerated and 4  focussed to a constant pass energy (typically 50 eV) using a series of five electron lenses. The electron beam is aligned through the various regions of the spectrometer using X , Y deflectors and apertures at which the current can be monitored. The energy loss is scanned by applying a linear ramp voltage to the last lens element in the deceleration lens, and to the whole analyzer and detector systems. Most of the electron beam which is not of the selected energy, is collected in the primary beam dump. The selected electrons are bent around the analyzer, pass  21  Chapter 3: Experimental Methods  Dl  po 1 e ( e,e+ i on ) S p e c t romet e r  GUN CHAMBER  ra—u • I Po  Elect ron Gunrr'-4h  r  Pi  fc^-Jr.—^—fl  h  f  DoTT D l  02  Sariple  TGas  DIFFERENTIAL  MAIN  CHAMBER  10  D  = Eleotrostatto  P  = Plate  N.B.  15 cn TURBO  PUMP  PUMP 450  l/s  TURBO PUMP 450  l/s  l/s  deflector.  f o r Measuring boa«  T h e TOF t u b e  CHAMBER  TURBO 300  rj  is  aotually  current at  90 degrees  to  the  fooal  plane  of  the  electron  analyzer.  Figure 3.1. Schematic of the low resolution dipole (e,e) and dipole (e,e+ion) spectrometer. For photoabsorption work, this instrument has been used in the dipole (e,e) mode.  For the  photoionization studies, coincident detection of energy loss electrons and time of flight mass selected ions in the dipole (e,e+ion) mode has been employed.  Chapter 3: Experimental Methods  22  through the exit aperture, and are detected by a channel electron multiplier (Mullard B419AL) which is operated in the saturated pulse counting mode. The system is initially tuned on the main unscattered beam at zero energy loss using an electrometer and the input cone of the channel electron multiplier as a Faraday cup. The energy scale is calibrated by comparison of the obtained spectrum with the convoluted (by 1 eV fwhm) high resolution spectrum obtained on the instrument described in section 3.3. The various sections of the spectrometer are differentially pumped by five turbomolecular pumps. There are two stages of differential pumping to protect the electron gun and minimize "poisoning" of the oxide cathode when reactive substances are used [7, 27]. The electron guns must be replaced periodically due to gradual degradation in performance. In addition the collision region, the forward analyzer and the TOF detector region are separately pumped. Positively charged ions that are formed during the electron scattering (ionization) events are extracted by an electrostatic field of 400 Vcm" across the collision region and accelerated to 1  6 keV. The ions travel down a time-of-flight tube, are focussed by ion lenses, and detected with a double chevron microchannel plate (40 mm diameter, model VUW-8920ES, Electro-Optical Sensors (Intevac)) ion detector, used in the saturated pulse counting mode. The TOF mass spectrometer has a mass resolution (m/Am) of ~50. This spectrometer is capable of running two different types of experiments, involving different modes of operation. In the dipole (e,e) mode the forward scattered electron beam is analyzed as a function of energy loss. The relative differential electron scattering cross sections obtained are converted to a relative optical photoabsorption oscillator strength spectrum by multiplying the signal by the known Bethe-Born conversion factor for the spectrometer [3, 19], as described earlier. As mentioned previously, the electron scattering signal decreases rapidly with  Chapter 3: Experimental Methods  23  energy loss - approximately as the inverse third power of the energy loss. Therefore, in order to handle the very wide dynamic range of signal, the energy loss spectra are typically obtained from a series of overlapping data sets covering the wide energy loss range required for VTRK sum rule normalization. During both data acquisition and background subtraction allowance must be made for the different incident electron currents when matching the different overlapping ranges of energy loss. This change in current is necessary because of the very rapid drop-off of cross section with increase in energy loss. Background subtraction is required to eliminate any effects due to non-spectral electrons and to residual gases at the base pressure of ~lxl0" Torr. Typical sample 7  gas pressure, as measured by an ion gauge on the spectrometer housing, was 6xl0~ Torr, and this 6  was reduced by 70% for background subtraction runs. At these pressures double scattering is negligible. The second type of operation, the dipole (e,e+ion) mode, was used to obtain photoionization mass spectra by detection of the fast forward scattered electrons at each particular energy loss (photon energy) in coincidence with the TOF mass analyzed ions. The flight time is proportional to the square root of the mass to charge ratio (m/e) for each ion and is determined from a single-stop, time-to-amplitude converter using the ion signal as the "start" pulse and the delayed electron signal as the "stop" pulse. Attention must be given to the baseline of random coincidences, as this decays exponentially with increasing time. Since low mass ions from dissociative photoionization processes  can carry significant kinetic energies of  fragmentation, the TOF mass spectrometer has been designed with strong draw-out fields to ensure uniform collection of energetic fragment ions with excess kinetic energies of fragmentation up to 20 eV [25]. TOF mass spectra were obtained for HC1 and HBr, at an energy resolution of 1 eV fwhm, from the first ionization threshold up to 260 eV and 90 eV respectively. The branching  Chapter 3: Experimental Methods  24  ratios for the individual ions formed were obtained by integrating the baseline-subtracted TOF mass spectra which had been corrected for the response function of the ion multiplier (MCP) as a function of m/e [28]. Data acquisition in both the dipole (e,e) and dipole (e,e+ion) modes is controlled by a 286 personal computer. Scanning of the energy loss is achieved under computer control using a D/A converter to drive a voltage programmable power supply. Relative energy loss data are collected by the computer and are multiplied by the instrumental Bethe-Born factor. The data, collected in overlapping segments over different ranges of energy loss, are normalized to each other to give a continuous spectrum and then made absolute by use of either the VTRK (S(0)) or the S(-2) sum rule as described earlier (section 2.4). Analysis of the data is achieved by using a suite of programs written in PASCAL by Dr. G. Cooper of this laboratory. 3.3 High Resolution Dipole (e,e) Spectrometer A detailed account of the construction and operation of the high resolution dipole (e,e) spectrometer can be found in earlier publications [3, 19, 29]. A brief description of the main features of this instrument, pictured in figure 3.2, follows. The instrument consists of four chambers, each differentially evacuated by a turbomolecular pump to a base pressure of ~10"  7  Torr. The electron gun, an oscilloscope gun body (Cliftronics CE5AH) equipped with a thoriated tungstenfilament,is used as the electron beam source, and is floated at -3 keV with respect to the collision chamber, which is at ground potential. The electron beam is decelerated (typically to 10 or 30 eV) to pass through the hemispherical monochromator, which serves to reduce the energy spread of the primary beam, thus selecting a narrow range (in the range 0.03-0.07 eV fwhm) of the initial Maxwell-Boltzmann distribution (~0.9 eV fwhm) of electron energies.  The  monochromated electron beam is re-accelerated to 3 keV and scattered off the sample gas in the  Chapter 3: Experimental Methods  25  collision chamber. Typical sample gas pressures were ~ 7xl0" Torr. Each data collection run 6  was followed by a reduced pressure run with pressures of ~ 2xl0~ Torr to be used for 5  background subtraction. After passing through the collision chamber, electrons scattered in the forward direction are retarded before energy-selection in the hemispherical electron analyzer at a constant pass energy. The overall energy resolution used for valence shell studies is 0.048 eV fwhm at 10 eV pass energy. For inner shell studies a resolution of 0.098 eV fwhm (at 30 eV pass energy) is used. Careful fine-tuning of the instrument is necessary to maintain the high resolution. The energy loss is added to the complete analyzer and detector systems using a computer controlled linear ramp voltage to drive a programmable power supply. The energy loss range of this spectrometer is presently limited to the region below 100 eV and this generally precludes use of the VTRK sum rule for normalization of high resolution data. The selected electrons pass through the analyzer exit aperture (with most of the remaining beam going to the beam dump) and are detected by a channel electron multiplier (Mullard B419AL) operating in the pulse counting saturated mode. The energy scale of the high resolution instrument is calibrated by introducing approximately equal partial pressures of helium and the sample gas into the collision chamber and referencing to the 1 *S -» 2 P transition of He which occurs at 21.218 eV [30]. Such corrections X  are normally less than 0.030 eV. The high resolution data is collected by a 386 personal computer, multiplied by the known  Chapter 3: Experimental Methods  MONOCHROMATOR  ANALYSER DETECTOR ELECTRON GUN  REACTION CHAMBER  TURBO RJMP 3 6 0 L/S  SCALE  r0  20 cm  TURBO PUMP 4 5 0 L/S  Figure 3.2. Schematic of the high resolution dipole (e,e) spectrometer  TURBO PUMP 3 6 0 L/S  Chapter 3: Experimental Methods  instrumental Bethe-Born factor, and background subtracted. This is achieved using software written by Dr. G. Cooper. Thus the collected high resolution EELS spectrum is converted to a relative photoabsorption oscillator strength spectrum, which is made absolute either by normalizing it in the smooth continuum region to the absolute low resolution oscillator strength spectrum obtained from the low resolution instrument, or by using the S(-2) sum rule [7, 21, 31] and a known value of the dipole polarizability.  27  Chapter 4 Photoabsorption and Photoionization Studies of Hydrogen Chloride 4.1 Introduction Absolute cross-sections (oscillator strengths) for the photoabsorption and photoionization of hydrogen chloride in the vacuum UV and soft X-ray regions are of interest in many areas of science and technology. Various chloride compounds are involved in etching and growing, and hydrogen chloride is an important molecule in plasma processing [32]. Chloride compounds are also important in interstellar and atmospheric chemistry [33]. HC1 is of interest as a major constituent or by-product of some gas discharge and chemical laser systems [34]. Previously published absolute photoabsorption measurements using direct optical methods cover only limited energies in the valence and CI 2p regions. The present work is the only data set providing photoabsorption and photoionization oscillator strengths (cross sections) over the energy range from 5 to 280 eV, covering the valence shell and CI 2p inner shell regions. Earlier low resolution dipole (e,e) absolute photoabsorption (5 to 60 eV) and photoionization (up to 40 eV) measurements (see below) have been reported [10], but the accuracy of the results is somewhat in doubt due to limitations in instrumentation and measurement procedures used at the time. These limitations have since been overcome [7, 27]. The electronic spectrum of HC1 has been investigated in a variety of experimental studies using direct optical techniques.  Previously published valence shell absolute photoabsorption  oscillator strengths have been obtained in the 170 to 230 nm (5.4 - 7.3 eV) range [35, 36], in the 1050 - 2150 A (5.8 - 11.8 eV) region [37], from 106 to 185 nm (6.7 - 11.7 eV) [33], and in the 980 - 720 A (12.65 - 17.22 eV) range using synchrotron radiation [38]. The CI 2p inner shell of HC1 has also been investigated by Hayes and Brown [39], who used synchrotron radiation in the 28  Chapter 4: Photoabsorption and Photoionization Studies of HCI L3  >2  edge region for HC1 from 196 to 210 eV to obtain absolute cross sections. In other  synchrotron radiation work Ninomiya et al. measured absolute cross sections of H Q in the range between 198 and 280 eV [40]. A number of relative photoabsorption studies of H Q have been previously published, using direct optical techniques in the valence shell region below 20 eV excitation energy [41, 42]. The absorption [43, 44] and emission [45] spectra of HC1 have been the subject of several studies at photon energies below 10 eV. Relative intensities in electron energy-loss spectra below 20 eV have also been published by Wang et al. [34], who measured the electron energy-loss spectrum of HC1 from 5 to 19 eV, with an incident electron energy of 200 eV and scattering angles of 0° to 19°. York and Comer [46] studied both dipole and non-dipole transitions in the energy range from 9 to 16.5 eV using a spectrometer with a range of impact energies from 10 to 120 eV and scattering angles from 0° to 90°. Shaw et al. [47] measured the energy-loss spectrum of HC1 in the outer valence (8 - 18 eV) and CI 2p (198 - 205 eV) regions with an incident energy of 1500 eV and 0° scattering angle. HC1 has been the subject of a previous photoionization study [48] in the energy range below 21 eV, but no absolute measurements were reported. Photoelectron spectroscopy (PES) studies of the binding energy spectrum of HC1, determined using the 584 A resonance line of helium (Hel) have also been reported [49, 50]. More recently the photoelectron spectrum of HC1 up to 40 eV binding energy was reported by Adam [51], by Svensson et al. [52] and by Will et al. [53] using synchrotron radiation. Faegri and Kelly [54] have reported calculations of absolute partial cross sections for photoionization from the 2jt, and 5a orbitals of HC1. In the earlier work on HC1 in 1984 in this laboratory, the low resolution (1 eV fwhm) absolute photoabsorption (7-60 eV) oscillator strengths and partial photoionization oscillator  29  Chapter 4: Photoabsorption and Photoionization Studies ofHCl  strengths for molecular and dissociative photoionization as well as for production of the electronic ion states (up to 40 eV) have been obtained [10], using dipole (e,e), dipole (e,e+ion) and dipole (e,2e) spectroscopy respectively. However, possible problems with a few [7] earlier low resolution dipole (e,e) photoabsorption measurements have since been discovered in the cases of SiH [27], C H [31], C H [55], and freons [56]. In these cases the target gases have been found 4  2  4  2  2  to have had a significant pressure dependent effect on the electron gun emission characteristics, due to a poisoning effect on the oxide cathode used in the low resolution dipole (e,e) measurements. It is now known that when such spurious effects are present, serious errors can occur in the background subtraction procedures (and thus in the shape and magnitude of the resulting oscillator strength spectrum) due to the significant change in operating pressures. These problems have since been overcome by the installation, in 1991, of an additional differential pumping stage between the electron gun chamber and the collision chamber [27]. Similarly the background subtraction procedures have been further refined and improved [3]. Thus there is some doubt as to the reliability of the previous low resolution photoabsorption and photoionization oscillator strength determinations for HC1 [10]. An additional consideration in the earlier H Q work [10] concerns the data analysis procedures and in particular the normalization of the data, in which the VTRK sum rule was used without taking into account the small correction (~ 4%) due to Pauli-excluded transitions (see section 2.4). Furthermore no attempt was made to fit and extrapolate the data to infinite energy. Rather the spectrum was simply integrated up to the rather low energy limit of the data at 60 eV. Extrapolation from 60 eV to infinity and the neglect of the Pauli excluded correction will have opposing effects on the oscillator strength scale estimated by the VTRK sum rule. Thus the resulting absolute scale [10]  30  Chapter 4: Photoabsorption and Photoionization Studies of HCI  31  should have been reasonably accurate, although the magnitude of the Pauli excluded correction is expected to be somewhat smaller than the effect of data truncation at 60 eV. It is important to note that any errors in the total photoabsorption differential oscillator strength would also be reflected in the partial photoionization oscillator strengths (PPOS's) for molecular and dissociative photoionization (see equation 2.3.3) as well as those for producing electronic ion states (see equation 2.3.5). Furthermore the efficiencies of the ion detectors have since been calibrated [28] with respect to different m/e (i.e., mass to charge) values. For the above reasons the previous dipole (e,e) and (e,e+ion) studies of HCI have been reinvestigated in the present work with the modified (i.e. differential pumping and calibrated ion detector) low resolution dipole (e,e) and (e,e+ion) spectrometer and with improved data processing procedures, incorporating extrapolation to infinite photon energy and correction for Pauli excluded transitions. Furthermore the valence shell photoabsorption and photoionization data have been considerably extended to 195 eV and 85 eV respectively.  In addition,  photoabsorption and photoionization studies have also been carried out in the CI 2p inner shell region from 195 to 280 eV. In further new work the absolute photoabsorption oscillator strength spectrum has been obtained in both the valence and CI 2p inner shell regions with the high resolution (0.05 eV fwhm) dipole (e,e) spectrometer (see section 3.3) which was developed some years after the original low resolution studies were reported [10]. The interpretation of the electronic spectrum of HCI has been believed to be relatively simple with most states being assigned to Rydberg transitions [57]. The ground state vibrational structure has been investigated by Huber and Herzberg [58] and by Terwilliger and Smith [41]. The lowest four vibrational levels of the B !? state and their rotational structure have been 1  recorded in emission studies [45] in the 2000A region, i.e., transitions to high vibrational levels of  Chapter 4: Photoabsorption and Photoionization Studies of HCl  32  the X 2 ground state. Tilford et al. [44] have identified the lowest Rydberg states bTI, and CTI x  +  and higher Rydberg species [43], in particular the f A state. Hirst and Guest [59] and Bettendorf 3  et al. [57] have reported ab initio calculations on the excited states of HCl below 12 eV. The present work addresses mainly the question of absolute transition probabilities (oscillator strengths) rather than assignments. The assignments shown in figures 4.4 and 4.6 are taken from references [41, 47]. Further progress on the assignments will require high quality calculations of both transition energies and oscillator strengths. The present work provides the latter data. 4.2 Results and Discussion 4.2.1 Electronic Structure The ground state electron configuration of HCl in the independent particle approximation can be expressed as: (la)  (2a)  2  CI Is  2  (3a) (ln) 2  CI 2s  CI 2p  4  (4a) (5a) (2n) 2  2  Valence  4  (6a)° Virtual Valence  Lempka et al. [49] reported the values of the ionization potentials of the 2jt and 5a outer valence orbitals to be 12.74 ( X  2  n ), 3/2  12.82 ( X  2  n ), 1/2  and 16.23 (A 2 ) eV using Hel high resolution 2  +  photoelectron spectroscopy. Svensson et al. [52] have reported the complete XPS valence shell binding energy of HCl up to 55 eV using monochromated A l Kc radiation at a resolution of ~0.6 eV fwhm. In other work, binary (e,2e) spectroscopy (i.e. electron momentum spectroscopy, or EMS) has been used by Suzuki et al. [60] to investigate the full valence shell binding energy spectrum of HCl, including the inner valence (4a) region at low resolution (2 eV fwhm). Further slightly higher resolution EMS spectra have been reported by Hollebone et al. [61]. Shaw et al. [47] have reported ionization potentials for the CI 2p inner shell to be 207.40 eV (2p / ) and 3 2  209.03 eV (2p ). 1/2  Chapter 4: Photoabsorption and Photoionization Studies of HCI  33  4.2.2 Photoabsorption Oscillator Strengths 4.2.2.1 Low-Resolution Photoabsorption Measurements The low resolution (1 eV fwhm) relative oscillator strength spectrum in the energy range 5 to 280 eV was obtained from the Bethe-Born converted EELS spectrum (see section 2.1) and was put on an absolute scale by using the VTRK sum-rule, as described in section 2.4. The resulting absolute photoabsorption oscillator strength spectrum covering extensive ranges of the valence and CI 2p spectra is presented in figure 4.1. The low resolution oscillator strengths are given numerically in table 4.3. The higher energy region of the valence shell spectrum (85-198 eV) was fitted to a polynomial curve of the form: df/dE = AE" + BE" + CE" 2  3  (4.1)  4  where E is the energy loss, A, B, and C are the best-fit parameters which were determined to be A = 1.6662 x 10 eV 4  1  B = -9.9539 x 10 eV 5  2  C = 1.3954 x 10 eV . 7  3  Thus the contribution to the valence shell region above 198 eV was estimated by extrapolation of the fitted curve to infinite energy. The percentage of the total valence shell oscillator strength in the region from 198 eV to infinity was found to be 8.67%. The absolute oscillator strength scale of the photoabsorption spectrum in the valence shell was obtained by normalizing the area under the Bethe-Born converted EELS spectrum to the number of valence electrons (8) plus a small correction (0.309) for Pauli-excluded transitions, for a total integrated oscillator strength area of  Chapter 4: Photoabsorption and Photoionization Studies of HCl  34  Photon Energy (eV) Figure 4.1. Absolute low resolution (1.0 eV fwhm) photoabsorption oscillator strength spectrum for HCl. (a) Valence shell region; (b) Long range spectrum showing the valence shell and CI 2p inner shell regions. The inset shows the photoionization efficiency (n,).  Chapter 4: Photoabsorption and Photoionization Studies of HCI 8.309. The Pauli excluded correction was obtained by using the value for the atom argon [62], which is isoelectronic with HCI (see section 2.4). The energy scale of the low resolution spectrum was set by comparison with the high resolution spectrum, convoluted with 1 eV fwhm (figure 4.2). It can be seen that the two spectra show very good quantitative agreement over the entire energy range. As described in section 2.4 above, the S(-2) sum rule can be used to determine the absolute scale of the oscillator strength spectrum. Conversely, the S(-2) sum rule can also be used to obtain the dipole polarizability from the VTRK sum rule normalized absolute spectrum. Thus, from the absolute differential oscillator strength spectrum of HCI, a dipole polarizability of 16.97 a.u. was determined using the S(-2) sum rule. For greater accuracy, the high resolution (0.05 eV fwhm) valence shell spectrum, presented in section 4.2.2.2 below, was used for this determination in the discrete region. The value thus obtained is in good agreement (2.5%) with the experimental values for the static dipole polarizability (etc) of 17.4 a.u. [63, 64, 65, 66] and is also in excellent agreement (sl.5%) with values obtained from theory (17.24 a.u. [67], 16.86 a.u. [68]). In a comparison (see figure 4.3 below) with the present work, the earlier photoabsorption data reported by Daviel et al. [10] using dipole (e,e) spectroscopy are ~20% higher than the oscillator strengths obtained in the present work. This discrepancy is consistent with the differing treatments in the two measurements, of integration to infinite energy and consideration of the Pauli excluded transitions (see comments near the beginning of section 4.1). It has been found in the present work that the Pauli excluded transitions contribute less than 4% to the total integrated oscillator strength, compared to the ~26% contribution from the integration from 60 eV (the limit  35  Chapter 4: Photoabsorption and Photoionization Studies of HCl  36  Figure 4.2. Comparison of low resolution (1.0 eV fwhm) and high resolution (0.05 eV fwhm, convoluted with 1 eV fwhm) valence shell photoabsorption measurements obtained in the present work.  Chapter 4: Photoabsorption and Photoionization Studies of HCl  37  Figure 43. Comparison of low resolution absolute photoabsorption oscillator strengths for HCl obtained in the present work and by Daviel et al. [10].  Chapter 4: Photoabsorption and Photoionization Studies of HCI of the earlier data [10]) to infinite energy. Neglect of both these latter contributions would have the net effect of increasing the absolute oscillator strength for the data of reference [10] by ~22% relative to the present work. Due to the low energy resolution (1 eV fwhm), the spectrum presented in figure 4.1(a) exhibits very little discrete structure.  Three broad peaks are observed with maxima at  approximately 9, 13, and 15 eV, respectively. Also, a shoulder is visible ~ 8 eV. The discrete region is shown in much greater detail in the high resolution spectrum discussed in section 4.2.2.2 below. A large percentage (70 %) of the valence shell oscillator strength is located below the Cooper minimum, which is at ~40 eV [69]. Figure 4.1(b) also shows the CI 2p inner shell spectrum in the region above ~ 200 eV. The pre-edge discrete structure is shown in detail in the significantly higher resolution spectrum infigure4.6 below and discussed in section 4.2.2.3. The oscillator strength rises steeply at the CI 2p3 ,i/2 ionization edges and the continuum levels off out /2  to the limit of the data at 280 eV. 4.2.2.2 Valence Shell High Resolution Photoabsorption Measurements for HCI The valence shell absolute photoabsorption oscillator strength spectrum is shown at higher resolution (0.050 eV fwhm) from 5 to 18 eV and from 5 to 40 eV infigures4.4(a) and 4.4(b) respectively. The assignments shown for the discrete transitions on figure 4.4(a) are from reference [41] and the ionization potentials are from high resolution PES measurements [49]. The spectrum was placed on an absolute scale by normalizing the Bethe-Born converted energy-loss data to the low resolution absolute photoabsorption oscillator strength in the smooth ionization continuum region (see figure 4.4(b) and table 4.3). It can be seen from a comparison of the spectra obtained at high and low resolution (figure 4.4(b)), that the shapes of the two spectra are  38  Chapter 4: Photoabsorption and Photoionization Studies of HCl  n=4  c  (a) V 4  i T• •I I' _ Ii •P T l WL  n-nsa  6  5  n-ndal 7T — n d n I  x n 2  r f  39  3/2.1/2  Valence  AE=0.05eV  fwhm a-nsal"" n=4  I"" 5  n=4  A I I 2  a-npal""l""  +  CM  2  5  O C  q CD  00 I  6  8  1  —  -  r  ^  —  10  i  1  12  ,  m m  , —  14  O  18  16  P h o t o n Energy (eV)  CJ  c o  (b) HR Dipole ( e , e ) 0.05eV f w h m LR Dipole ( e , e ) 1 eV f w h m  3  '_p  Q_ O  m 2 -°  O H—'  o  1 CL  20  30  40  P h o t o n Energy (eV) Figure 4.4:  Absolute high resolution (0.05 eV fwhm) and low resolution (1.0 eV fwhm)  photoabsorption oscillator strengths (cross sections) for the valence shell of HCl.  Chapter 4: Photoabsorption and Photoionization Studies of HCl  40  the same over the smooth continuum region from 20 to 40 eV. This is a good consistency check of the results, since the high and low resolution dipole (e,e) spectrometers have different impact energies and scattering geometries, and thus different Bethe-Born factors. As discussed in the introduction, only a few previous studies of the valence shell absolute oscillator strength spectrum have been reported, and these only cover small ranges at low energies. Using a Xe discharge lamp Roxlo and Mandl [35] measured the absolute 2  photoabsorption spectrum of HCl from 5.8 to 7.3 eV, and reported absolute cross sections which are significantly (33%) lower than those in the present work, as shown in figure 4.5(a) and table 4.1. Roxlo and Mandl also showed the work of Romand [36], who had earlier reported very similar results. Myer and Samson [37] used direct optical techniques to measure the absolute o  cross sections of HCl in the region 5.77-11.81 eV, at a resolution of 0.25A (AE ~0.002 eV fwhm). A detailed comparison with the present work is not possible due to the difficulty of integrating the spectrum reported by Myer and Samson [37], which showed a non-linear baseline (due to stray light effects) and truncated peaks as well as large line-saturation errors. The intensity of the broad peak at ~8 eV reported by Myer and Samson [37] is only 5% higher than that found in the present work. However the intensities reported by Myer and Samson [37] in the region of the much narrower Rydberg transitions up to 11.8 eV (the limit of the data in reference [37]) are very different from those in the present work, particularly in the region ~ 11.3 eV where there is clear indication of severely attenuated intensities due to line saturation (i.e. line width/bandwidth interaction) effects in the Beer-Lambert law optical work. The intensities [37] in the region of ~11.3 eV are only a few percent of those obtained in the present work where line saturation effects cannot occur. It is noteworthy in this regard that these large line-saturation  41  Chapter 4: Photoabsorption and Photoionization Studies ofHCI  _Q 34  Ld  HC P r e s e n t Work Roxlo ic Mandl  2H  Nee et a l .  c _o  [35]  [33]  '->->  u CD CO  CO CO  o  6.0  O  6.5  c _o  9.0  7.0  Photon Energy (eV) (b)  ' 80  o  Frohlich and Glass-Maujean [ 3 8 ]  co _Q 60 A O  O  o _c  40  CL  CD  -i—' _3  20 H P r e s e n t Work  O  CO  _Q <  i —  13  14  15  16  17  Photon Energy (eV) Figure 4.5.  Comparison of the present absolute cross sections in the valence shell  photoabsorption of HCI with the digitized data from (a) Roxlo and Mandl [35] and Nee et al. [33] and (b) Frohlich and Glass-Maujean [38].  Chapter 4: Photoabsorption and Photoionization Studies of HCI  42  Table 4.1: Integrated oscillator strengths for selected energy regions of the valence and CI 2p inner shell photoabsorption spectrum of HCI. Energy Region (eV)  Present Work (Mb)  Literature Values (Mb)  Ref.  5.80- 7.27  0.52  0.31  [35]  0.31  [36]  5.11  [33]  6.20- 9.5  5.27  9.25-10.8  20.96  a  0.0026  [33]  16.90  0.133  [33]  9.87-10.00  2.22  0.0173  [33]  10.10-10.35  0.62  10.35-10.80  0.46  9.3- 9.4  0.40  9.5- 9.7  10.90-12.11  b  46.20  10.93-11.05  1.37  0.01  [33]  11.05-11.18  4.59  0.0213  [33]  11.18-11.34  15.37  0.032  [33]  11.44-11.67  5.24  0.0056  [33]  11.77-12.11  ' 17.23  12.11-12.42 12.11-17.2  14.82 208.1  198 - 210  11.8  18.06  [39]  200 - 211  13.7  16.97  [40]  Integrated area excluding small peak at 9.32 eV. This and the following values from Nee et al. [33] are sums of reported oscillator strengths obtained at much higher resolution than in the present work. a  b  Chapter 4: Photoabsorption and Photoionization Studies of HCl errors occur even though the resolution (AE -0.002 eV fwhm) in the optical work [37] is much better than that in the presently reported dipole (e,e) work (0.05 eV fwhm). In 1986, Nee et al. [33] reported the photoabsorption and fluorescence cross sections of HCl in the valence shell region from 6.7 to 11.7 eV using synchrotron radiation, with an optical monochromator resolution of 0.05 nm (AE -0.005 eV fwhm). It is clear from a comparison of integrated peak areas (Mb) at selected energies (table 4.1) that the cross sections reported by Nee et al. are very much lower (typically by two orders of magnitude) than those obtained in the present work despite the further improved resolution.Evidently the cross sections of the peaks observed by Nee et al. are dramatically reduced by Beer-Lambert law line-saturation effects (see section 1.2 above). Comparison of the broad band between 7 and 9 eV shows a 3% difference between the work by Nee et al. [33] and the present work; this is not surprising since linesaturation effects well not perturb the intensity of such a broad band. A more recent study by Frohlich and Glass-Maujean [38] has reported absolute cross sections with high resolution from 12.65 to 17.22 eV using synchrotron radiation. Their measured cross sections are from -3 to 50% higher than those reported in the present work (see figure 4.5(b) and table 4.1 above). 4.2.2.3 Inner Shell (Chlorine 2p) High Resolution Photoabsorption Measurements for HCl The absolute photoabsorption spectrum of the CI 2p inner shell of HCl was obtained from 198 to 213 eV at 0.1 eV fwhm (figure 4.6(a)) using the high resolution dipole (e,e) spectrometer, in order to examine the discrete excitation region in some detail. The high resolution spectrum was normalized to the low resolution spectrum (figure 4.1(b)) at 198 eV below the first CI 2p electronic excitation threshold as shown on figure 4.6(a). The assignments on figure 4.6(a) are taken from the work of Shaw et al. [47].  43  44  Chapter 4: Photoabsorption and Photoionization Studies of HCI  3d 5p 7s 4s 4p  >  CI 2p  CM CD  I  i~i—n . 4p  a*  P r e s e n t w o r k , 0.1 eV f w h m O  Present work,  h  • (2pv0'  5s 6s  4  5 s 6s7s 3d  1 eV f w h m  Hayes a n d B r o w n [ 3 9 ]  Ld  N i n o m i y a et a l . [ 4 0 ]  JD  c  o  cn c  o  CD  i_  Hayes  ~U~)  and  Brown  00  [39]  (f) if)  o  o \- 2 'o cn O  O c  1  H  h 1  c  o  CL  o JD O O  o  Ninomiya  et  al.  [40]  O  a  o o  CL  -*->  _c a. a; 1  H  O C/5 <  198  200  202  204 Photon  206 Energy  208  210  212  (eV)  Figure 4.6. Absolute photoabsorption oscillator strength (cross section) for the CI 2p inner shell of HCI obtained in the (a) present work and digitized from data reported by (b) Hayes and Brown [39], and (c) Ninomiya et al. [40]. Peak assignments in (a) are taken from reference [47].  Chapter 4: Photoabsorption and Photoionization Studies of HCl Two previous studies [39, 40] of the absolute photoabsorption cross-sections in the inner shell region of HCl have been reported. Hayes and Brown [39] obtained absolute cross sections from 198 to 210 eV at 0.16 eV fwhm. The shape of the spectrum [39] agrees quite well with that obtained in the present slightly higher resolution (0.1 eV fwhm) work. However the absolute cross sections reported by Hayes and Brown [39] typically are 35% higher than those measured in the present work, as can be seen from the broader features and continuum region in figure 4.6(a,b) and from the numerical data in table 4.1. The results obtained by Hayes and Brown [39] included corrections for stray light which may have contributed to some of the differences in cross section, as well as differences in shape with respect to the present work. The second study, by Ninomiya et al. [40], using synchrotron radiation, reported absolute photoabsorption crosssections for HCl between 198 and 280 eV. Although the general shape of the spectrum and relative intensities agree reasonably well with the spectrum obtained in the present work, (figure 4.6), this data is typically 10-20% lower than the data of Hayes and Brown [39] but still 20-50% higher in cross section than the present work. Significant differences are also observed in the energy scales between the spectrum of Hayes and Brown [39], that of Ninomiya et al. [40], and that obtained in the present work (figure 4.6). The presently reported energy scale is based on the very accurate valence shell energy calibration (see section 3.3). Ninomiya et al. [40] quote an uncertainty of ±20% in the cross section at the CI L23 edge. It should be noted that the absolute oscillator strength scale of the presently reported CI 2p inner shell spectrum has been established by use of the VTRK (S(0)) sum rule normalization of the valence shell, which gives a very accurate (within 2%) value of the dipole polarizability of HCl (see section 4.2.2.1) via the S(-2) sum rule.  45  Chapter 4: Photoabsorption and Photoionization Studies of HCI  46  4.2.3 Photoionization Studies of HCI in the Valence Regions Time of flight (TOF) mass spectra for HCI were obtained from 12 to 85 eV in the valence shell region (figure 4.7) and from 195 to 250 eV in the CI 2p inner shell region (shown in figure 4.15 below) using the dipole (e,e+ion) spectrometer at 1 eV (fwhm). The relative intensities of chlorine containing peaks are consistent with the natural relative abundances of C1 and C1 35  (~3:1). The doubly charged ion Q  2 +  37  is observed with a (double) ionization potential of 48 eV.  No stable HC1 is detected in the valence shell photoionization spectra. Table 4.2 shows a list of 2+  the positive ions that were detected in the TOF mass spectra (figure 4.7), together with their respective mass to charge ratios and appearance potentials. Photoion branching ratios for the molecular and dissociative ions were determined as percentages of the total photoionization from integration of the background subtracted TOF peaks. Figure 4.8 presents the photoionization branching ratios as a function of photon energy for the valence and inner shell, and table 4.4 gives the photoion branching ratios. As can be seen infigure4.8, the HC1 branching ratio decreases +  steeply initially, and levels off around 50 eV. This steep decrease correlates with steep increases observed in the CT and H branching ratios, with a similar leveling above 50 eV. The doubly +  charged C l  2+  appears at 48 eV and likewise levels off above 50 eV. These trends confirm that in  the valence shell continuum the ionic fragmentation pattern of HCI remains constant above the upper limit of the Franck-Condon region [60, 74] of the B 2 state of HC1 . A large change is 2  +  observed in the branching ratios in the region of CI 2p excitation and ionization, indicating the onset of different ionic photofragmentation processes, which are discussed in section 4.2.5 below. The branching ratios determined (only up to 40 eV) in the earlier dipole (e,e+ion) study reported by Daviel et al. [10] are compared with the present results in figure 4.9. The ion  Chapter 4: Photoabsorption and Photoionization Studies of HCl  Al  Table 4.2. Appearance potentials (in eV) for molecular and fragment ions resulting from photoionization of HCl.  Ion  m/e 37  HC1  +  Cl H Cl  +  HC1  3  (n ) 2  35,36  17.5  17 ± 1  17.44  1  19  18 ± 1  18.071  17,18  197  19  211  +  ++  Appearance Potentials (eV) Dipo e (e,e) Calculated Experiment Present Daviel Ref [70] Work et al. 12.5 12.474 12.748 12.828 16.254° 13 ± 1  ++  a: b: c: d:  3 / 2  (n ) 2  17.34  b  14.5  C  35.5  d  photoelectron spectroscopy [50] photoionization [71] retarding potential difference threshold experiment [72] nth root extrapolation threshold experiment [73]  1 / 2  Chapter 4: Photoabsorption and Photoionization Studies of HCl  100  HCl  A  Valence Shell  48  35 H  +  ci  c CD  H  2+  °  H 0* ;  CD 50 > 0H  H  8 0 . 0 eV CD  37  0  x  35  10 2 0 . 0 eV  0  x  10  2.0  +  CI  CI  +  VJL l  -i—f—i—i—i—i—|—r1.5  CI  37  2.5  3.0  3.5  •Time of Flight (JJS)  Figure 4.7. Time-of-flight (TOF) mass spectra for valence shell photoionization of HCl at 20.0 eV (lower trace) and 80.0 eV (upper trace).  Chapter 4: Photoabsorption and Photoionization Studies of HCI  49  Photon Energy (eV)  Figure 4.8. Photoion branching ratios (leV fwhm) for molecular and dissociative ions from photoionization of HCI, as a percentage of total photoionization.  50  Chapter 4: Photoabsorption and Photoionization Studies of HCI  100 A  •  8 8  -S88  8  8  80 A  H CI * o  o  +  °-4—'  o  20 J  co c o  10  c  P r e s e n t Work  Daviel et al. [ 1 0 ]  • Present Work o Daviel et al. [ 1 0 ]  to  .•  CO  §  0  o • o  +  60 A  o  8  .. e888o ° 8  8  o  o -4—'  H  O 15  • Present Work o Daviel et al. [ 1 0 ]  D_ 10  .88888  8  8 8  o .o. » 8. A 8 8» oS  •988« 1  10  '  20  Photon  e S  0 0  T  30  40  Energy (eV)  Figure 4.9. Comparison of photoion branching ratios (1 eV fwhm) obtained in the present work and by Daviel et al. [10]. The photoion branching ratios obtained by Daviel et al have been corrected for detector response (see text).  Chapter 4: Photoabsorption and Photoionization Studies of HCl  51  detector used in the previous dipole (e,e+ion) TOF MS studies (a Johnston focused mesh multiplier) was calibrated some considerable time after the 1984 study was published and found to exhibit a slow decrease in sensitivity above m/e = ~ 30 [28]. A correction factor was therefore applied to the previously obtained data for purposes of comparison with the branching ratios obtained in the present work. It can be seen from figure 4.9 that the data from the two studies agree reasonably well. The largest difference can be seen in the branching ratio data for the H  +  ion. The ion detector has since been replaced with a calibrated [28] double microchannel plate (MCP) ion detector which shows considerably improved mass resolution. The MCP ion detector was used for the present work. Absolute partial photoionization oscillator strengths (PPOS's) for the molecular and each dissociative ion were obtained by taking the triple product of the photoionization efficiency, the photoion branching ratio and the total photoabsorption oscillator strength (see equation 2.3.3) at each photon energy. These PPOS's are shown in figure 4.10 and the numerical values are given in table 4.5. The relative photoionization efficiency, as obtained from the ratio of the total photoionization to the forward scattered energy loss counts at each photon energy, making the reasonable assumption that the ratio of ionic products becomes unity above 25 eV, is shown in the inset to figure 4.1(b). From the valence shell branching ratios, appearance potentials, and the PPOS's for molecular and dissociative photoionization, information regarding the dipole induced breakdown pathways has been obtained.  Chapter 4: Photoabsorption and Photoionization Studies of HCI  52  Photon Energy (eV)  Figure 4.10. Absolute partial oscillator strengths for molecular and dissociative photoionization of HCI from 6 to 265 eV at 1 eV fwhm resolution.  Chapter 4: Photoabsorption and Photoionization Studies of HCl 4.2.4 Dipole-Induced Breakdown of H C l As discussed in section 2.3 (see equations 2.3.4 and 2.3.6), the total absolute photoionization oscillator strength distribution can be partitioned into contributions from the partial oscillator strengths for molecular and dissociative photoionization or into contributions from the electronic ion state partial photoionization oscillator strengths [8]. The latter quantities can be obtained from tunable energy photoelectron spectroscopy (PES), or the equivalent technique of dipole (e,2e) spectroscopy [2, 75]. It has previously been discussed [10] how given ion partial photoionization oscillator strengths can be described as linear combinations of the electronic ion state partial photoionization oscillator strengths or vice-versa. These procedures operate under the assumption that the fragmentation pattern arising from a given ionic state is constant above the upper limit of the Franck-Condon region for a given electronic ion state [8]. The validity of this assumption has been discussed in earlier publications [8, 76], along with a discussion of the conditions under which the assumption breaks down. Complicating factors can arise from (energy localized) autoionization of neutral states, or in the presence of significant direct multiple photoionization processes [8]. The partial oscillator strengths for production of each particular molecular or dissociative ion involve appropriate contributions from the breakdown of the initially formed electronic states of the molecular ion [2, 8, 77], and hence the dipole-induced breakdown for a molecule can be investigated in the valence shell above ~40 eV (the upper limit of the B 2 Franck-Condon region corresponding to ionization of the most tightly 2  bound valence orbital of HCl). The earlier study [10] in this laboratory investigated the dipole induced breakdown of HCl in the valence shell up to 40 eV by a joint consideration of the absolute photoabsorption and partial photoionization oscillator strengths derived from dipole (e,2e) and (e,e+ion) measurements. For purposes of comparison, in the present work the dipole  53  Chapter 4: Photoabsorption and Photoionization Studies of HCI  54  (e,2e) photoelectron branching ratios obtained by Daviel et al. [10] have been multiplied by the photoionization efficiencies (see figure 4.1(b)) and the absolute total photoabsorption oscillator strengths (figure 4.1(a)) obtained in the present work. The resulting PPOS's for production of the X TI, A Z and B 2 electronic states (figure 4.11 and table 4.6) were then examined together 2  2  +  2  with those for production of HCT, C l and H , and from this quantitative information on the +  +  dipole-induced breakdown pathways of HCI can be deduced. Also shown in the top panel of figure 4.11 is the complete valence shell low resolution binding energy spectrum as obtained by Suzuki et al. [60] using electron momentum spectroscopy (EMS). By considering together the appearance potentials (table 4.2) and electronic ion state PPOS's, an attempt has been made to partition the ion PPOS's into contributions from the decomposition of the various ion electronic states. Figure 4.12 shows the best fit for such contributions to the molecular (figure 4.12(b)) and total dissociative (Cl + H , figure 4.12(c)) +  +  photoionization oscillator strengths. The best fit for the molecular ion PPOS is given by the sum (X n + 0.905 A 2 + 0.55 B Z). Below 16.5 eV, formation of the molecular ion HC1 is due 2  2  +  2  +  solely to the X n state, whereas at higher energies the A 2 2  2  and B 2 states make further  +  2  contributions. The PPOS for the total photoionization (i.e. CT + H ) is well matched by the +  remaining oscillator strength (0.095 A 2 + 0.45 B 2) as shown in figure 4.12(c). There is a 2  +  2  1.5 eV difference in the initial onsets of the ionic and electronic state PPOS's, which is to be expected, as can be seen by referring to the binding energy spectrum of HCI [60] shown in the top panel of figure 4.12(a). The predicted contribution from the A 2 state to the total dissociative 2  +  photoionization cross section is very small (less than 10%), and it can be seen infigure4.12 that  55  Chapter 4: Photoabsorption and Photoionization Studies of HCl  CO  'c Z>  > _o  EMS b i n d i n g  HCl  3  energy  spectrum  A E = 2.1 e V f w h m  n  2  (p  AV  1  = 8°  B I ( m a n y body ion s t a t e s ) 2  en  0 40 (N  * A E =  >  I  20 A  •  •  ••  • ••  •  \  '  •••  •  1 •  AI  1  -  i  i  i 2  •  Q_  2  •  • • • .  • •  -  •  _D o  <  -  •• . . .  o OJ -t-i  +  • •  •  CO  '  •• •  •  o  n  2  1_  CO  o  1  •  -4—'  2  1 ~  en c 10 4 <D  o  x  • • • .  r  —I  JZ  1.4eV f w h m  •• 10  15  20  • 1 25  •  '•  1 30  — 35  1 —  40  Photon Energy (eV)  Figure 4.11. Absolute partial photoionization oscillator strengths for production of the XTT, A Z , and B 2 electronic ion states of HC1 (lower three panels). This data has been obtained from the dipole (e,2e) "photoelectron" branching ratios of Daviel et al. [10] together with the presently reported total photoabsorption oscillator strengths and photoionization efficiencies. The top panel shows the valence shell EMS binding energy spectrum of HCl on the same energy scale [60]. 2  +  2  +  Chapter 4: Photoabsorption and Photoionization Studies of HCI  EMS  binding  AE=2.1eV  energy  spectrum  fwhm  CP = 8  (many body ion states)  > co  40  Molecular Photoionizationl  0.905A  30 A  • HCI  LL)  *8  o X-t-0.905A-r0.55Bl e  20  \9.  .0.55B 10 CO 1_  -t-»  co  o -*-» _o  0 3  'u  A E=  Total Dissociative Photoionization  CO  3  O  _  2H  + H  Cl  o  0.095A+0.45B  +  0)  -t—»  •8  Q  »o  n  •  o o „  °  o o  o Q_  „ o ° °  •  +  • 1 e V fwhm o 1.4eV fwhm  f  1  0.45B  •  S o  JD  o  / •  en X)  <  0.095A  n 10  15  20  25  30  35  40  Photon Energy (eV) Figure 4.12.  Absolute partial oscillator strengths for molecular and total dissociative 2  2 +  photoionization of HCI, with best linear combination fits of contributions from the X n , A 2 , and B 2 electronic state partial oscillator strengths. The top panel shows the EMS binding energy 2  spectrumfromreference [60].  Chapter 4: Photoabsorption and Photoionization Studies of HCl  57  this is consistent with the onset of the dissociative ionization oscillator strength coming from the upper tail end of the Franck-Condon region for production of the A 2 state (figure 4.12(a)). 2  +  The partitioning of the separate CT and H partial oscillator strengths into electronic ion +  state contributions is more difficult, as can be seen in figure 4.13(b,c). For the dissociative ion CT, the best linear combination of electronic states was given by (0.065 A 2 2  + 0.33 B 2). It can  +  2  be seen that there is a small difference (~1 eV) between the onsets of the C l and the A 2 +  2  +  PPOS's. Clearly the onset of the dissociative ion CT is in the upper portion of the A 2 Franck2  +  Condon region, consistent with the major (lower energy) portion going to the molecular ion as discussed above. The sharp rise in CT oscillator strength above 20 eV clearly corresponds with the onset of the B 2 electronic state oscillator strength, shown in the binding energy spectrum in 2  figure 4.13(a). As can be seen from figure 4.13(b), there is a significant difference between the CT PPOS and the B 2 electronic ion state contribution in the 23 to 30 eV region, which could be 2  due to dissociative autoionization of neutral excited states of HCl down to dissociative states leading to H + CT. Above ~30 eV the proposed breakdown linear combination fits very well, as might be expected [8]. The remaining linear combination of electronic ion state contributions, (0.03 A 2 + 0.12 2  +  B 2), can be seen in figure 4.13(c) to fit the H partial oscillator strength quite well above 2  +  ~30 eV. The initial onset of H does not correspond with the onset of the A 2 state of HC1 , +  2  +  +  similar to the situation for CT in figure 4.13(b). Again, this is not unreasonable since the H  +  appearance potential indicates that it comes only from the upper energy portion of the A 2  +  2  Franck-Condon zone. This is consistent with the small fraction (0.03 A 2 ) predicted by the 2  model. Further contributions to H come from the B 2 state of HC1 . +  2  +  +  58  Chapter 4: Photoabsorption and Photoionization Studies of HCI  EMS b i n d i n g e n e r g y  H C I  c  ID  > -*-< o  1H  AE=2.1eV  spectrum  fwhm  <P=  AV B*X  8"  ( m a n y body ion s t a t e s )  a: •  CI  i CD  Cl  o  +  (AE= 1eV f w h m )  0.065A+0.33B 1.4eV f w h m )  (AE=  CM  i  o o  LU  = ° o  ° ° °  8  o  °  o  o  0.33B  o  0 •  cn c  0.065A  CD  oo  0  o 0.8  H  +  0  °„  o  o  •  H  o  0.03A+0.12B  o  w 0.6 O o  "5 o D_  •  0.4 0  0.12B  •  o  cu 0 . 2  0  0.03A  o 3 o.o 15  20  25  40  35  30  Photon Energy (eV)  Figure 4.13. Absolute partial oscillator strengths for dissociative photoionization of HCI, with best fit linear combinations of contributions from the A 2 2  +  and B 2 electronic state partial 2  oscillator strengths. The top panel shows the EMS binding energy spectrum from reference [60].  Chapter 4: Photoabsorption and Photoionization Studies of HCl  59  The sharp peak at ~24 eV is most likely due to dissociative autoionization of neutral excited states of HCl, converging on the B 2 ion state, to give H + CI. The differences in the 23-30 eV region 2  +  are understandable due to the "extra" oscillator strength in the CT PPOS in this region (see figure 4.13(b) and preceding discussion). It is possible that dissociative autoionization processes are the reason for the differences below 30 eV in figures 4.13(b,c). For figure 4.12(c), it can be seen that the overall agreement is much better below 30 eV when the total dissociative photoionization is considered.  However it should be remembered (see figure 4.13) that the 25-30 eV range  corresponds to the Franck-Condon region of the main B 2 peak of HC1 , and the breakdown 2  +  model is therefore not likely to apply strictly in these circumstances [8]. A summary of the dipole breakdown pathways, detailing the electronic ion state contributions to the formation of the molecular and dissociative ions, is shown in figure 4.14. In the earlier work reported by Daviel et al. [10], the dipole breakdown pathways of HCl were examined, with similar results to those reported in the present work. The A 2 and B 2 2  +  2  electronic ion state contributions to the molecular ion (HC1 ) PPOS, as reported by Daviel et al. +  [10], were 3% and 5% higher, respectively, than in the present work, while contributions to the total dissociative (Cl + H ) PPOS were 3% and 5% lower than in the present work. These slight +  +  differences are understandable in light of the possible problems encountered in the earlier work [10], as discussed above in sections 4.2.2.1 and 4.2.3. In particular, the branching ratio for H  +  obtained in the earlier work [10] was reduced due to decreased response of the ion detector (see section 4.2.3 and figure 4.9). Since the branching ratios are used to determine the PPOS's for production of the ions, a lower PPOS for H was obtained, which thus required smaller +  contributions from the A 2 2  +  and B 2 electronic ion states. In the present work, more accurate 2  Chapter 4: Photoabsorption and Photoionization Studies of HCl  60  HCl + hv Valence (above ~40 eV)  CI 2p Inner Shell (above 197 eV)  \  ci  2+  cr  HC1  2+  Cl  2+  (Small yields from V V V Auger and/or direct dissociative double photoionization)  Figure 4.14. Dipole ionic photofragmentation breakdown pathways of HCl in the V U V and soft X-ray regions.  Chapter 4: Photoabsorption and Photoionization Studies of HCI photoionization data were obtained, and thus slightly different fits of electronic ion state contributions to molecular and dissociative ion PPOS's were found. The differences between CT and H oscillator strengths and the electronic ion state contributions (figures 4.13(b,c)) in the 25+  30 eV range, which have been observed in the present work, were also seen in the earlier work by Daviel et al. [10]. 4.2.5 Photoionization Studies of HCI in the CI 2p Region Time-of-flight (TOF) mass spectra at selected energies in the CI 2p inner shell region, and in the valence shell at 197 eV just below the onset of CI 2p excitation, are shown in figure 4.15. In this region (see figure 4.1) the total photoabsorption, and thus the photoionization, is dominated by CI 2p processes, with only ~10% being due to valence shell photoionization. The intensity scale for each panel of figure 4.15 is relative to the tallest (i.e. the height of the most intense) peak in each spectrum. As can be seen in the top panel, at 197.0 eV, below the first 2p excitation edge, the most intense peak is the H CT, consistent with this being due only to the 35  valence shell. As the equivalent photon energy increases, i.e. as the CI 2p a* excitation threshold is reached (201.5 eV), the relative intensity of the HC1 peaks (two isotopes) decreases with +  respect to the CT peaks, indicating an increasing role for dissociative ionization. At 197 eV, in the valence continuum, the doubly charged ion C l  2+  is formed, but there are still no discernible  HC1 ions. As the energy increases further into the CI 2p inner shell Rydberg excitation region 2+  (206 eV), the intensity of the C l  2+  peaks increases, and HC1 decreases markedly. In the 2p +  ionization continuum (220 and 260 eV), HC1 appears while HC1 is dramatically reduced. The 2+  +  H peak increases in relation to the other ion peaks as the energy increases. +  61  62  Chapter 4: Photoabsorption and Photoionization Studies of HCl  100  197.0 eV  (Valence Shell)  100  206.0 eV  (2p —» Rydberg)  CO  c  CD  C CD  50  H  n HCl  .2+  CI  _D i ^-  100 A  ' 'i  i  1 1  i  i  220.0 eV  1  1  v  1  4  i  (2p —»• continuum)  n CI  50  HCI  260.0 eV 50  2+  (2p —» continuum)  A  t  1.2  1.4  1  1.6  r  1. 8  1  r  2.0  2.2  2.4  =r~—r  2.6  2.8  Time of Flight  3.0  3.2  3.4  3.6  3.8  (LIS)  Figure 4.15. Time-of-flight (TOF) mass spectra at selected energy losses in the valence shell and CI 2p inner shell regions for the photoionization of HCl.  Chapter 4: Photoabsorption and Photoionization Studies of HCI  63  The trends outlined above are consistent with what is observed in the photoabsorption studies. As can be seen in figure 4.1(b), above the CI 2p ionization edge, the valence shell cross section represents only a small fraction of the total cross section. Thus the molecular ion, HC1 , is +  formed mostly from valence shell processes and drops off significantly above the CI 2p edge with respect to the other ion peaks. The formation of CA in the valence shell could result from 2+  dissociative Auger decay of inner valence ion states (B S), i.e. V V V Auger processes, or by 2  dissociative double photoionization. Clearly the relative yield of C l increases significantly in the 2+  CI 2p excitation and ionization regions. In the CI 2p a* and Rydberg excitation regions this could be due to dissociative autoionization (i.e. Resonance Auger [56]) of neutral, CI 2p, excited states of HCI down to H + C l  2+  with ejection of two electrons. In the CI 2p continuum, C l  2+  can be  formed by dissociative ( L W ) Auger processes. Furthermore, the increase of H as well as Cl +  +  with respect to HC1 is likely due to Coulomb explosion of doubly charged (HC1 ) ions to give +  2+  H + CT. The increased width of the H peaks (see figure 4.15) at 220 and 260 eV is consistent +  +  with the high kinetic energy of H ions expected from Coulomb explosions. The doubly charged +  HC1 , which appears in the CI 2p continuum, could be the result of a competing Auger process in 2+  which ionization of HCI results in a 2p hole state, which then decays, without dissociation, to form HC1 . 2+  Photoion branching ratios for the CI 2p inner shell region are shown in the higher energy portions of figure 4.8 and given numerically in table 4.4. It should be noted that the branching ratios, obtained from TOF mass spectra, are relative to the peak with the greatest integrated area, as opposed to the height of the most intense peak in the case of the TOF spectra presented in figure 4.7. As discussed in section 4.2.3 above, the branching ratios of all ions present in the  Chapter 4: Photoabsorption and Photoionization Studies of HCl  64  valence shell level off above 50 eV and up to 197 eV, as expected by the ionic photofragmentation model [8]. However, not surprisingly some changes occur above 197 eV in the CI 2p inner shell region. Most noticeably, the amount Of HC1 decreases sharply with respect +  to the other ions, then levels off above 210 eV. This is expected, since as discussed above HC1 is +  formed mostly from valence shell processes. The very small yield of HC1 in the CI 2p region +  indicates that when a 2p hole occurs in the molecule due to excitation or ionization, then the molecular ion has a very high probability of break-up (i.e. dissociation) by non-radiative processes such as autoionization (resonance Auger) and Auger decay. The relative yield of the C l ion +  increases sharply (figure 4.8) near 200 eV, just below the CI 2p ionization edge, then drops and levels off above 210 eV. The resulting spike is apparently due to dissociative autoionization of CI 2p neutral (valence and/or Rydberg) excited states down to H + CT, which would account for both the sharp drop in HC1 as well as the spike in Cl . The photoion branching ratio for C l rises +  +  2+  sharply at the CI 2p excitation edge. Similarly the relative yields of HC1  2+  and H increase +  dramatically at the CI 2p ionization edge. HC1 is formed only from CI 2p ionization with no 2+  detectable contribution from valence shell or CI 2p excitation processes. Partial oscillator strengths for photoionization of HCl in the CI 2p inner shell region (derived from the product of the branching ratios, the photoionization efficiency and the total photoabsorption)  are shown in figure 4.10 and given numerically in table 4.5.  The  photoionization efficiency has been assumed to be unity in the CI 2p inner shell region. The trends observed in the PPOS's in the CI 2p inner shell correspond to those already observed in the TOF mass spectra and in the photoion branching ratios, but in this case (figure 4.10) absolute, rather than relative, yields are shown. The PPOS for HC1 drops dramatically in the CI 2p region, +  corresponding with increases in the yields of the other four (dissociative) ionic species. The  Chapter 4: Photoabsorption and Photoionization Studies of HCI PPOS's conclusively show that in the soft X-ray region above ~198 eV, CI 2p excited HCI molecules or singly charged cations (HCf) with a CI 2p hole are extremely unstable and very rapidly decay to give a range of singly or doubly charged fragment ions as well as neutral fragments. The overall breakdown pathways of HCI in the V U V and soft X-ray regions above ~40 eV (i.e. above the Franck-Condon limits of all valence shell ionization processes) resulting from the present measurements and analysis are summarized in figure 4.14 above. Further information and confirmation of the dipole breakdown pathways described above could be obtained from photoelectron-photoion coincidence experiments as a function of energy.  65  Chapter 4: Photoabsorption and Photoionization Studies of HCI  66  Table 4.3: Absolute differential photoabsorption oscillator strengths of HCI for the valence shell and CI 2p inner shell regions (1 eV fwhm). Total Photoabsorption Oscillator Strength (df/dE) eV  lO^eV  1  eV  lO^eV  1  eV  10 eV 2  x  eV  IOVT  6.0  0.10  18.5  39.66  31.0  8.29  47.0  l.n  6.5  0.33  19.0  38.94  31.5  7.48  48.0  1.13  7.0  1.04  19.5  37.79  32.0  6.90  49.0  1.12  7.5  2.12  20.0  36.73  32.5  5.89  50.0  1.14  8.0  2.87  20.5  35.25  33.0  5.46  51.0  1.13  8.5  2.75  21.0  33.60  33.5  4.85  52.0  1.11  9.0  6.43  21.5  32.24  34.0  4.37  53.0  1.12  9.5  18.76  22.0  31.16  34.5  3.94  54.0  1.15  10.0  12.37  22.5  29.51  35.0  3.36  55.0  1.12  10.5  7.48  23.0  28.04  35.5  3.00  56.0  1.17  11.0  24.16  23.5  26.15  36.0  2.86  57.0  1.13  11.5  33.21  24.0  25.29  36.5  2.63  58.0  1.14  12.0  40.29  24.5  24.02  37.0  2.34  59.0  1.17  12.5  41.50  25.0  22.58  37.5  2.08  60.0  1.13  13.0  33.05  25.5  21.30  38.0  2.02  61.0  1.13  13.5  27.26  26.0  19.50  38.5  1.81  62.0  1.15  14.0  31.36  26.5  18.42  39.0  1.76  63.0  1.15  14.5  39.69  27.0  17.16  39.5  1.66  64.0  1.11  15.0  42.57  27.5  16.30  40.0  1.47  65.0  1.14  15.5  41.77  28.0  14.78  41.0  1.35  66.0  1.10  16.0  41.92  28.5  13.55  42.0  1.27  67.0  1.12  16.5  41.68  29.0  12.37  43.0  1.19  68.0  1.13  17.0  41.39  29.5  11.23  44.0  1.16  69.0  1.12  17.5  41.06  30.0  10.22  45.0  1.12  70.0  1.10  18.0  40.57  30.5  9.35  46.0  1.12  71.0  1.09  1  Chapter 4: Photoabsorption and Photoionization Studies of HCl  67  Table 4.3 (continued): Absolute differential photoabsorption oscillator strengths of HCl for the valence shell and CI 2p inner shell regions (1 eV fwhm). Total Photoabsorption Oscillator Strength (df/dE) eV  IOVV  eV  lO^eV  eV  eV  IOVV  72.0  1.06  114.0  0.70  164.0  10" eV 0.41  73.0  1.08  116.0  0.69  166.0  0.40  204.0  0.59  74.0  1.08  118.0  0.66  168.0  0.39  204.5  0.68  75.0  1.08  120.0  0.65  170.0  0.40  205.0  0.66  76.0  1.07  122.0  0.64  172.0  0.38  205.5  0.77  77.0  1.03  124.0  0.64  174.0  0.37  206.0  0.98  78.0  1.00  126.0  0.61  176.0  0.37  206.5  1.05  79.0  1.02  128.0  0.59  178.0  0.35  207.0  1.18  80.0  1.01  130.0  0.59  180.0  0.36  207.5  1.35  82.0  0.99  132.0  0.57  182.0  0.36  208.0  1.46  84.0  0.97  134.0  0.55  184.0  0.35  208.5  1.56  86.0  0.94  136.0  0.55  186.0  0.35  209.0  1.70  88.0  0.92  138.0  0.54  188.0  0.34  209.5  1.78  90.0  0.91  140.0  0.52  190.0  0.35  210.0  1.83  92.0  0.89  142.0  0.50  192.0  0.31  210.5  1.96  94.0  0.87  144.0  0.51  194.0  0.31  211.0  2.00  96.0  0.85  146.0  0.50  196.0  0.33  211.5  2.08  98.0  0.83  148.0  0.48  198.0  0.30  212.0  2.17  100.0  0.82  150.0  0.48  200.0  0.33  212.5  2.20  102.0  0.79  152.0  0.46  200.5  0.39  213.0  2.24  104.0  0.76  154.0  0.47  201.0  0.47  213.5  2.27  106.0  0.76  156.0  0.43  201.5  0.54  214.0  2.29  108.0  0.74  158.0  0.43  202.0  0.56  214.5  2.38  110.0  0.72  160.0  0.43  202.5  0.57  215.0  2.38  112.0  0.71  162.0  0.41  203.0  0.56  215.5  2.41  1  1  a  x  203.5  0.55  1  Chapter 4: Photoabsorption and Photoionization Studies of HCI  68  Table 4.3 (continued): Absolute total photoabsorption oscillator strengths of HCI for the valence and CI 2p inner shell regions. Total Photoabsorption Oscillator Strength (df/dE) eV 216.0  lO'eV 2.48  eV 232.0  lO^eV 2.35  eV 257.0  lO^eV 2.20  216.5  2.46  233.0  2.36 '  258.0  2.15  217.0  2.48  234.0  2.35  259.0  2.18  217.5  2.47  235.0  2.35  260.0  2.09  218.0  2.50  236.0  2.39  261.0  2.08  218.5  2.53  237.0  2.35  262.0  2.16  219.0  2.53  238.0  2.37  263.0  2.08  219.5  2.50  239.0  2.30  264.0  2.04  220.0  2.47  240.0  2.27  265.0  2.13  220.5  2.56  241.0  2.32  266.0  2.05  221.0  2.51  242.0  2.27  267.0  2.02  221.5  2.52  243.0  2.28  268.0  2.05  222.0  2.56  244.0  2.24  269.0  2.06  222.5  2.52  245.0  2.30  270.0  2.13  223.0  2.55  246.0  2.22  271.0  2.10  223.5  2.56  247.0  2.25  272.0  2.13  224.0  2.54  248.0  2.22  273.0  2.09  224.5  2.46  249.0  2.25  274.0  2.10  225.0  2.55  250.0  2.29  275.0  2.13  226.0  2.49  251.0  2.16  276.0  2.13  227.0  2.42  252.0  2.23  277.0  2.08  228.0  2.45  253.0  2.21  278.0  2.04  229.0  2.42  254.0  2.08  279.0  2.17  230.0  2.34  255.0  2.11  280.0  2.10  231.0  2.44  256.0  2.15  1  1  1  Chapter 4: Photoabsorption and Photoionization Studies of HCl  69  Table 4.4: Photoion branching ratios for the molecular and dissociative photoionization of HCl, as a percentage of total photoionization. Photon Energy (eV)  Photoion Branching Ratio (%) HCI  12.5 13.0 13.5 14.0 14.5 15.0 15.5 16.0 16.5 17.0 17.5 18.0 18.5 19.0 19.5 20.0 20.5 21.0 21.5 22.0 22.5 23.0 23.5 24.0 24.5 25.0 25.5 26.0 26.5 27.0 27.5  100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 98.80 98.26 97.99 97.50 96.74 96.54 96.24 95.59 95.32 94.63 93.94 92.28 90.75 89.73 89.24 89.20 88.59 86.99 86.94 86.79 85.88  +  Cl  +  1.20 1.74 2.01 2.33 2.66 2.79 2.66 3.19 3.25 3.90 4.35 5.20 6.41 7.38 7.90 8.66 9.39 10.78 11.30 11.26 12.11  HC1  2+  Cl  2+  H  +  0.18 0.61 0.67 1.11 1.21 1.43 1.47 1.72 2.51 2.84 2.89 2.87 2.15 2.03 2.23 1.76 1.96 2.02  Chapter 4: Photoabsorption and Photoionization Studies of HCI Table 4.4 (continued): Photoion branching ratios for the photoionization of HCI, as a percentage of total photoionization. Photon Energy (eV)  Photoion Branching Ratio (%) HC1  28.0 28.5 29.0 29.5 30.0 30.0 32.0 34.0 36.0 38.0 40.0 42.0 44.0 46.0 48.0 50.0 50.0 55.0 60.0 65.0 70.0 75.0 80.0 197.0 201.5 206.0 211.0 220.0 240.0 260.0  85.51 84.98 84.39 83.43 82.15 80.92 77.04 71.92 68.39 61.37 57.30 50.88 48.58 46.24 45.80 46.00 45.36 44.75 44.49 45.96 46.11 46.03 45.95 40.05 19.82 8.59 3.37 2.53 3.18 3.03  +  cr 12.37 12.52 12.66 12.88 13.40 14.45 16.11 17.57 19.06 22.78 25.13 27.19 28.29 29.89 29.10 28.91 29.16 28.11 28.15 26.31 25.94 26.33 26.17 27.23 41.43 35.78 21.35 20.23 22.95 21.54  HC1  2+  1.86 2.41 1.76 1.71  CI  2+  3.88 9.33 9.29 9.74 10.22 14.54 15.66  H  +  2.12 2.51 2.95 3.69 4.45 4.62 6.85 10.51 12.55 15.85 17.57 21.94 23.13 23.87 25.10 25.09 25.49 27.15 27.37 27.73 27.94 27.64 27.88 28.84 29.43 46.35 63.68 64.62 57.57 58.07  Chapter 4: Photoabsorption and Photoionization Studies of HCl Table 4.5: Absolute partial differential oscillator strengths for the molecular and dissociative photoionization of HCl. Photon Energy (eV)  Partial Differential Oscillator Strength (lO^eV ) Her  12.5 13.0 13.5 14.0 14.5 15.0 15.5 16.0 16.5 17.0 17.5 18.0 18.5 19.0 19.5 20.0 20.5 21.0 21.5 22.0 22.5 23.0 23.5 24.0 24.5 25.0 25.5 26.0 26.5 27.0  4.63 12.06 16.20 21.13 28.27 32.90 34.20 35.28 36.06 36.86 36.87 36.69 35.81 35.68 34.24 33.45 32.28 30.27 29.44 28.40 27.28 25.40 23.44 22.25 21.20 20.14 18.87 16.96 16.02 14.89  1  cr  0.45 0.65 0.74 0.85 0.94 0.97 0.89 1.01 1.00 1.17 1.26 1.43 1.66 1.83 1.88 1.96 2.00 2.10 2.08 1.93  HCI  2+  ci  2+  H  +  0.07 0.22 0.23 0.37 0.38 0.44 0.44 0.50 0.69 0.73 0.72 0.68 0.49 0.43 0.44 0.33 0.34  Chapter 4: Photoabsorption and Photoionization Studies of HCI  72  Table 4.5 (continued): Absolute partial differential oscillator strengths for molecular and dissociative photoionization of HCI. Photon Energy (eV)  Partial Differential Oscillator Strength (lO'VV" ) HC1  27.5 28.0 28.5 29.0 29.5 30.0 32.0 34.0 36.0 38.0 40.0 42.0 44.0 46.0 48.0 50.0 55.0 60.0 65.0 70.0 75.0 80.0 197.0 201.5 206.0 211.0. 220.0 240.0 260.0  14.00 12.64 11.52 10.44 9.37 8.27 5.32 3.14 1.96 1.24 0.84 0.65 0.57 0.52 0.52 0.52 0.50 0.50 0.52 0.51 0.50 0.46 0.13 0.11 0.08 0.07 0.06 0.07 0.06  1  +  cr 1.97 1.83 1.70 1.57 1.45 1.48 1.11 0.77 0.55 0.46 0.37 0.35 0.33 0.34 0.33 0.33 0.31 0.32 0.30 0.28 0.28 0.26 0.09 0.22 0.35 0.43 0.50 0.52 0.45  HC1  0.04 0.06 0.04 0.04  2+  Cl  0.01 0.05 0.09 0.20 0.25 0.33 0.33  2+  H  +  0.33 0.31 0.34 0.36 0.42 0.47 0.47 0.46 0.36 0.32 0.26 0.28 0.27 0.27 0.28 0.29 0.30 0.31 0.32 0.31 0.30 0.28 0.09 0.16 0.45 1.28 1.60 1.31 1.21  Chapter 4: Photoabsorption and Photoionization Studies ofHCl Table 4.6: Partial photoionization differential oscillator strengths for production of the electronic ion states of HCl. Photon Energy (eV)  Partial Oscillator Strength" (lO^eV ) X n A 2 B2 1  2  2  12.5 13.0 13.5 14.0 14.5 15.0 15.5 16.0 16.5 17.0 17.5 18.0 18.5 19.0 19.5 20.0  4.63 12.06 16.20 21.13 28.27 32.90 34.20 34.93 33.53 30.22 27.24 25.39 24.12 23.42 23.01 21.48  0.35 2.52 6.63 10.08 11.95 12.43 13.18 12.39 13.17  20.5 21.0  20.80 20.26  12.75 11.40  21.5  20.08  10.81  22.0  19.51 18.01 17.07 15.75  10.21 9.58 8.26  22.5 23.0 23.5  6.71  +  2  0.30 1.45 2.20 3.36  a. These values have been obtained from photoelectron branching ratios [10] using the presently reported photoabsorption differential oscillator strengths and photoionization efficiencies.  73  Chapter 4: Photoabsorption and Photoionization Studies of HCI  1A  Table 4.6 (continued): Partial photoionization differential oscillator strengths for production of the electronic ion states of HCI. Photon Energy (eV)  Partial Oscillator Strength (10- eV ) X II A 2 B2 2  2  2  x  +  2  24.0 24.5 25.0 25.5 26.0 26.5 27.0 27.5 28.0 28.5 29.0 29.5 30.0 31.0 32.0  15.87 14.49 14.24 12.55 11.87 10.57 8.86 8.48 7.69 6.91 6.19 4.94 4.29 2.65 2.69  6.69 5.94 5.42 4.81 4.35 4.08 3.58 3.20 2.81 2.57 2.35 1.69 1.74 1.20 1.24  2.23 3.33 2.94 3.56 < 3.56 3.90 4.60 4.16 4.29 4.20 3.84 4.61 4.19 4.71 3.04  33.0  1.92  2.93  34.0  0.87  0.79 0.57  35.0 36.0 37.0 38.0 39.0  0.47 0.49 0.24 0.08 0.16  0.40 0.31 0.24 0.22 0.23  2.93 2.75 2.06 1.95 1.72 1.36  40.0  0.07  0.40  1.00  a. These values have been obtained from photoelectron branching ratios [10] using the presently reported photoabsorption differential oscillator strengths and photoionization efficiencies.  Chapter 5 Photoabsorption and Photoionization Studies of Hydrogen Bromide 5.1 Introduction Hydrogen bromide is of interest in molecular spectroscopy, as well as in other areas of scientific research. Hydrogen halides, including HBr, exist in the Earth's upper atmosphere, and thus quantitative spectroscopic data for HBr are needed for the understanding of atmospheric photochemistry [78]. Since HBr absorbs light at a longer wavelength than HCl or HF, it is more subject to solar photodissociation, which is believed to be an important chemical process in destroying HBr in the upper atmosphere [79]. HBr is also of interest in the chemical etching of semiconductor material, since photodissociation of HBr is an important process for the Br source [80]. In biochemistry, the photodissociation cross section of HBr is needed to determine the concentration of Br atoms produced by photolysis [78]. The present work is, to the best of our knowledge, the only study measuring absolute photoabsorption (5-300 eV) and photoionization (10-91 eV) oscillator strengths (cross sections) over a very wide range of the valence and Br 3d inner shell regions. A much earlier study of HBr [11] in this laboratory reported absolute photoabsorption (7-100 eV) and photoionization (11-40 eV) measurements using low resolution dipole (e,e) and (e,e+ion) spectroscopies at low resolution (1 eV fwhm). However, limitations in the earlier instrumentation and measurement procedures as well as in the calibration of the absolute scale, similar to the situation discussed in chapter 4 for HCl, place some doubt on the reliability of these results. This suggests that new absolute measurements should be performed for HBr. There are few previously published absolute photoabsorption studies of HBr. Nee et al [78] reported the absolute photoabsorption and fluorescence cross sections of HBr in the 105-235 75  Chapter 5: Photoabsorption and Photoionization Studies of HBr nm (5.3-11.8 eV) region, using synchrotron radiation. Earlier works by Roxlo and Mandl [35], by Huebert and Martin [81], by Roman [36], and by Goodeve and Taylor [82] reported absolute photoabsorption cross sections in the pre-edge region below 8 eV. Apart from the earlier low resolution dipole (e,e) study, no absolute photoabsorption studies have been reported for the valence shell in the V U V (11-80 eV) and beyond, nor are there any absolute results for the Br 3d inner shell spectrum of HBr. Qualitative photoabsorption studies (i.e. limited range spectra without absolute intensity scales) of HBr have been reported in a number of papers. Photographic spectrophotometry was used by Ginter and Tilford [83, 84, 85], by Baig et al. [86] below 12 eV, and by Terwilliger and Smith [87] from 12.5 to 15.5 eV. Puttner et al. [88] used synchrotron radiation to investigate ligand-field splitting of the Br 3d inner shell levels in the region 69-79 eV at high resolution. Photoelectron spectroscopy has also been used to study the photoionization of HBr. Carlson et al. [69] measured angular distribution parameters as a function of photon energy from the outer valence photoelectron spectrum of HBr. Morin and Nenner [89] used synchrotron radiation to study the HBr photoelectron spectrum at photon energies on and off the 3d-»cr* resonance in HBr (see section 5.2.2.3). The high resolution valence shell photoelectron spectra of HBr and DBr have been measured by Adam et al [90] using an AlKa X-ray source. Salzmann et al. [91] have obtained spin-resolved photoelectron spectra of HBr in the region 12-15 eV using circularly polarized synchrotron radiation. Liu et al. [92, 93] have investigated resonance Auger spectroscopy of HBr with the photon energy tuned to the HBr Br 3d -* 5p% resonance. Other studies of the electronic excitation spectra of HBr have been reported using electron impact techniques. England et al. [94] obtained energy-loss spectra of HBr and DBr between 8 and 16 eV with a resolution of 30 meV fwhm, using different combinations of scattered electron  76  Chapter 5: Photoabsorption and Photoionization Studies of HBr  energy and angle, from 0.5 eV and 90° to 35 eV and 2°. Electron energy loss spectroscopy has also been used by Shaw et al. [47] to measure the energy loss spectrum of HBr in the valence (715 eV) and Br 3d inner shell (69-79 eV) regions, using an electron impact energy of 1.5 keV and 0° scattering angle. In the present work the photoabsorption and photoionization of HBr have been reinvestigated using dipole (e,e) and dipole (e,e+ion) spectroscopies with the improved instrumentation [7, 27] and the refined data collection and analysis techniques [7,22]. The range of the low resolution photoabsorption data has also been considerably extended to 300 eV, and photoionization data have been extended to 91 eV on the low resolution dipole (e,e), dipole (e,e+ion) spectrometer.  In addition, high resolution (0.05-0.1 eV fwhm) absolute  photoabsorption spectra have been obtained from 5 to 100 eV on the high resolution dipole (e,e) spectrometer (see references [3, 19, 29] and section 3.3) developed after the original low resolution [11] studies of HBr were reported. 5.2 Results and Discussion 5.2.1 Electron Structure The ground state electron configuration of HBr in the independent particle model can be expressed as: core  (3d)  (7a) (8a) (4jt)  10  2  Br 3d  2  (9a)°  4  Valence  Virtual Valence  The electronic states of HBr produced by removal of an electron from each of the valence +  orbitals are X n ,m, 2  3/2  (4JI)" ; A 2 , (8a)" ; and B 2, (7a)" [11, 49]. Ionization potentials for x  2  +  1  2  1  these states have been found by PES and the analogous method of electron momentum spectroscopy. Lempka et al [49] have reported values of 11.67 and 12.00 eV for the spin-orbit  77  Chapter 5: Photoabsorption and Photoionization Studies of HBr  split doublet X n / ,i/2, and a value of 15.27 eV for the A 2 state using Hel high resolution PES. 2  2  3  +  2  These values are in good agreement with other PES studies [95, 96]. Electron momentum spectroscopy has been used at low resolution by Brion et al. [74], who reported values of 11.9 eV (X n i/2), 15.6 eV (A 2 ), and 24.3, 30.3 eV (B 2 manifold). The valence shell photoelectron 2  2  +  2  3/2)  spectrum, including the HBr B 2 inner valence region has been studied in detail by Adam et al. +  2  [90] using AlKa radiation. The Br 3d inner shell ionization potentials have been found by Shaw et al. [47] to be 77.12 eV and 78.23 eV for the 3d and 3d states respectively. 5/2  3/2  5.2.2 Photoabsorption Oscillator Strengths 5.2.2.1 Low-Resolution Photoabsorption Measurements Figures 5.1(a,b) show the low resolution (1 eV fwhm) absolute differential photoabsorption oscillator strength spectrum (5-50 eV and 5-300 eV) obtained in the present work by S(-2) sum-rule normalization of the relative oscillator strength spectrum, as described in section 2.4. The relative spectrum was obtained from the Bethe-Born converted EELS spectrum (see section 2.1) measured on the low resolution dipole (e,e) spectrometer. The absolute total photoabsorption oscillator strengths are given numerically in table 5.4. The ionization potentials shown on figure 5.1 are taken from experimental measurements [49, 74, 90]. The low energy onset (-70 eV) of the Br 3d inner shell spectrum causes the VTRK sum rule normalization procedures to be extremely unreliable. In particular, curve fitting and extrapolation of the higher energy region is not possible. Therefore, the S(-2) sum rule was used to place the spectrum on an  78  79  Chapter 5: Photoabsorption and Photoionization Studies of HBr  _Q  c o ~o CD  if) -,  o  10  15  1  20  25  -I  30  '  35  CO CO  r  4-0  45  o  P h o t o n Energy ( e V )  O  V a l e n c e a n d Br 3 d Inner Shells  60  c o "Q_  50 40 30 20  o CO JD  O O O  _c  Q_  10 0  80  120  160  200  240  280  P h o t o n Energy ( e V )  Figure 5.1: Absolute low resolution photoabsorption oscillator strengths (cross sections) for the valence and Br 3d inner shell regions of HBr.  Chapter 5: Photoabsorption and Photoionization Studies of HBr  80  absolute scale by using an experimental dipole polarizability (a ) value of 23.7 a.u. [65]. The N  energy scale of the low resolution spectrum, as in the case of HCI (see section 4.2.2.1), was set by comparison with the high resolution spectrum convoluted with 1 eV fwhm (see section 3.3 and 5.2.2.2). As a consistency check, a comparison was made between the low resolution spectrum and the high resolution data convoluted with 1 eV fwhm and excellent agreement is obtained, as can be seen infigure5.2(a). In the earlier dipole (e,e) work on HBr [11], the absolute oscillator strength scale was obtained using the VTRK sum rule, without however taking into account the small correction factor due to Pauli-excluded transitions (see sections 2.4 and 4.1).  Furthermore,  the  normalization took into account only the spectrum below 100 eV (i.e. the upper limit of the experimental measurements), with no attempt to fit and extrapolate the spectrum to infinite energy. It should be noted (see also the discussion on HCI in section 4.2.2.1) that these two considerations, i.e. the Pauli-excluded correction and the extrapolation to infinite energy, have opposing (and thus offsetting) effects on the magnitude of the resulting oscillator strength scale. However, as was found for HCI (section 4.1), neglect of the extrapolation to infinite energy is expected to have the major effect (see section 4.1 for the situation in HCI). However, in the case of HBr, the higher energy valence shell portion cannot be estimated due to the presence of the Br 3d continuum region (see figure 5.1(b)).  The contribution of Pauli excluded transitions is  expected to constitute less than 4% of the total valence shell oscillator strength [22]. It is therefore to be expected that the magnitudes of the previously determined oscillator strengths will differ from that obtained in the present work and this can be seen in figure 5.2(b). The differences in shape can be attributed to the improved background subtraction procedures in the present work  Chapter 5: Photoabsorption and Photoionization Studies of HBr  81  P h o t o n Energy (eV) co  P h o t o n Energy (eV)  Figure 5.2:  Comparisons of the low resolution (I.eV fwhm) absolute oscillator strength  spectrum obtained in the present work with: (a) the high resolution (0.05 eV fwhm) spectrum obtained in the present work, convoluted with 1 eV fwhm; and (b) the absolute photoabsorption spectrum reported by Brion et al. [11].  Chapter 5: Photoabsorption and Photoionization Studies of HBr  82  (see discussion in section 4.2.2.1 for HCl). At the low resolution of the spectrum in figure 5.1(a), very little discrete structure is observable in the valence shell region. Three broad peaks with maxima at ~ 8.5, 11.5, and 14.5 eV are evident, with a slight shoulder below 8 eV. The Br 3d inner shell spectrum is also distinguished by a peak with maximum at ~ 71 eV, corresponding to the 3d-*a* transition. The high resolution (0.05-0.1 eV fwhm) spectra, presented in the following sections, show much more detailed structure in the valence and the inner shell photoabsorption spectra of HBr. 5.2.2.2 Valence Shell High Resolution Photoabsorption Measurements for HBr The valence shell high resolution (0.05 eV fwhm) absolute photoabsorption oscillator strength spectrum, shown in figure 5.3(a) from 5 to 16 eV, and in figure 5.3(b) from 5 to 25 eV, was normalized by the S(-2) sum rule using an experimental value of 23.7 a.u. for the static dipole polarizability (a ) [65]. The high resolution valence shell shows much discrete structure. Below N  8 eV, a low broad peak has been assigned to the aT^A !! electronic state [78]. The higher 1  energy structures present above 8 eV can be assigned to Rydberg series, as represented by the assignments given in figure 5.3(a), taken from references [87, 94]. Above the A 2 ionization 2  +  limit at 15.3 eV, no discrete structures are visible in the valence shell photoabsorption continuum (see figure 5.3(b)). As summarized above, very few absolute photoabsorption spectra of HBr have been previously published.  The most recent study is by Nee et al. [78], who measured the  photoabsorption and fluorescence spectra of HBr in the 5.3-11.8 region using synchrotron radiation. The absorption cross sections in the region between 5 and 8 eV reported by Nee et al. are about 16% lower than those reported in the present work, as shown in figure 5.4 and table  83  Chapter 5: Photoabsorption and Photoionization Studies of HBr  r > CD  A (a)  5  LU  A  4  "a cn c CD  i_  6  6  7 89  7T~*nsa FT n= 5 . n-»nscr I 6 , n=5 I 7T - * n d7T n= 5  V a l e n c e Shell 3 4 0 . 0 5 eV f w h m  n  TUX  3/2  TTsTWl , , gj,  _Q  0  r i  x  n  i  /  2  6 7  CN  A £  a-»nsa  2H  n =  6  c  7 8  2  O  1  O CD CO  0  co co  x 3  00  h  i_  O  -4—>  _o  , — — ,  8  "o  ,  . —  10  i  r-  1  12  14  16  P h o t o n Energy (eV)  c  o  4  0.1 eV f w h m  c  CL  •  3 4  1.0 eV f w h m  ^ °  J  CO  o  2  A  2 ^  o  CL  ° CL  o  _Q  O i_  O  CO  o  O  o" —»npa",npTT  CO  _Q O  °  -4-'  o  1 ^  CD JD  o  CO  _Q <  10  20 P h o t o n Energy (eV) 15  25  Figure 53: Absolute total photoabsorption oscillator strengths for the valence shell of HBr with high resolution (0.05eV fwhm): (a) 5-16 eV with peak assignments from references [87, 94]; (b) 5-25 eV showing agreement with low resolution (leV fwhm) measurements.  Chapter 5: Photoabsorption and Photoionization Studies of HBr 5.1. Part of the discrepancy can be accounted for by the ±20% uncertainty in the absorption cross section reported by Nee et al. [78]. In the energy range from 8 to 12 eV (the limit of their data [78]), the spectrum reported by Nee et al. [78] appears to be quite different from the spectrum obtained in the present work. As can be seen in table 5.1, integrated oscillator strengths reported by Nee et al. [78] are significantly (~ 2 orders of magnitude) lower than those obtained in the present work. This large difference is evidence of line-saturation (i.e. bandwidth-linewidth) effects that result from use of the Beer-Lambert law, as discussed in section 1.3. An attempt was made by Nee et al. to minimize this effect by taking measurements at very low sample pressures (below 2 mTorr) [78], but errors still occurred in this discrete region (8-12 eV) of the spectrum. A comparison was also reported [78] with earlier work by Huebert and Martin [81], who measured extinction coefficients for HBr in the region below 7.1 eV. The latter data [81] are ~3% higher [78] than the oscillator strengths obtained in the present work. Roxlo and Mandl [35] have reported absolute photoabsorption cross sections in the energy region 5.4 to 7.3 eV. There is a large difference in their data when compared to the present work, as shown in figure 5.4 and table 5.1. The reported cross sections by Roxlo and Mandl are much lower than those reported in the present work. In earlier studies, Roman [36] and Goodeve and Taylor [82] also measured the absorption spectrum of HBr below 8 eV using photographic spectrophotometry. Both of these studies [36, 82] reported cross section which are considerably lower than those obtained in the present work, as can be seen in figure 5.4 and table 5.1. Agreement in this lower energy range below 7.9 eV is somewhat better in the case of the absolute spectra reported by Nee et al. [78].  84  Chapter 5: Photoabsorption and Photoionization Studies of HBr  5.5  6.0  6.5  7.0  7.5  8.0  P h o t o n Energy (eV)  Figure 5.4: Comparison of absolute photoabsorption spectra obtained in the present work with reported spectra from the literature.  Chapter 5: Photoabsorption and Photoionization Studies of HBr Table 5.1: Comparison of integrated oscillator strengths (cross sections) for HBr reported in the literature with selected energy ranges from the present work measurements.  Energy Range  Present Work (Mb)  Previous Work (Mb)  Ref.  5.40-6.9  2.00  0.305  [82]  5.40-7.3  3.13  1.383  [35]  2.11  [36]  5.30- 7.9  4.41  3.69  [78]  8.25- 8.35  4.31  0.0323  [78]  8.69- 8.79  20.08  0.172  [78]  8.99- 9.08  1.60  0.013  [78]  9.80-10.00  4.94  10.00-10.11  6.91  0.0124  [78]  10.11-10.25  3.06  0.018  [78]  10.25-10.34  3.47  10.34-10.50  13.94  0.010  [78]  10.55-10.98  14.71  0.0335  [78]  10.98-11.19  17.45  0.0512  [78]  11.19-11.31  6.38  11.31-11.52  14.49  11.52-11.65  9.61  11.65-11.79  8.43  86  Chapter 5: Photoabsorption and Photoionization Studies of HBr 5.2.2.3 Br 3d Inner Shell High Resolution Photoabsorption Measurements for HBr The absolute photoabsorption oscillator strength spectrum for the Br 3d inner shell from 68 to 80 eV, obtained on the high resolution dipole (e,e) spectrometer at 0.1 eV fwhm, is shown in figure 5.5. The Br 3d relative oscillator strength spectrum was normalized to the S(-2) sum rule normalized valence shell spectrum (see previous section). The good agreement with the low resolution measurements below the Br 3d excitation region (figure 5.5) provides a consistency check on the accuracy of the absolute scale of the spectrum. A significant portion of the oscillator strength in the inner shell is contained in the broad peak with a maximum at ~71.5 eV, which corresponds to the strong 3d-»a* transition. The peaks at higher energies have been assigned to Rydberg states converging on the (ids^a)' limits. 1  The peak assignments and ionization  potentials in figure 5.5 are taken from reference [47]. Integrated oscillator strengths for selected energy ranges are given in table 5.2. The sharp structures in the 73.5-78 eV photon energy region, particularly the peak at -76 eV, show slight separation, which is most likely due to spin-orbit and ligand-field splitting. Puttner et al. [88] have reported a high resolution (10 meV fwhm) relative photoionization spectrum in the Br 3d excitation and ionization region (69-79 eV), obtained using synchrotron radiation. Their much higher resolution spectrum shows clear evidence of ligand-field splitting of both the Br 3d and Br 3d levels. 5/2  3/2  5.2.3 Photoionization Studies of HBr in the Valence Shell Region Time-of-flight (TOF) mass spectra were obtained using the low resolution dipole (e,e) spectrometer in the dipole (e,e+ion) mode (see section 3.2) from 10.5 to 66 eV in the valence shell. The positively charged ionic species resulting from photoionization of HBr are listed in  87  88  Chapter 5: Photoabsorption and Photoionization Studies of HBr  > UJ T3  0.04 H  5s  c  _b 0.03 H  Br 3 d Inner shell  O  T O 5s .  0 . 0 9 eV f w h m  o _o  5d 6p7p8p 6s 7s  (3d  5/2  )  h 0.04  o-*  a*  CO  o  4d 5p  / !  6s 7s (3d..,)  5p 6p7p8p 4d 5d  - 1  V2  0.03  1.0 eV f w h m  XI  O  C  o  0.02  0.02 '.+-' U d)  c o o  _Q  00  0.01 A  0.01 CO CO  o o  o  CL  £<D 0.00 _3 O 03 XI <  0.00  —l—  68  70  72  74 •  76  78  80  Photon Energy (eV)  Figure 5.5: Absolute photoabsorption oscillator strengths for the Br 3d inner shell region of HBr, obtained with high resolution (0.1 eV fwhm). Peak assignments are from reference [47].  Chapter 5: Photoabsorption and Photoionization Studies ofHBr  89  Table 5.2: Integrated oscillator strengths (cross sections) for selected energy ranges in the Br 3d inner shell of HBr, with and without estimated contributions from the valence shell continuum.  Energy Range  df/dE (Mb) With Without Valence Valence  69.19-73.52  6.38  3.57  73.64-74.31  0.62  0.19  74.31-75.37  1.77  1.07  75.48-76.40  1.54  0.95  76.40-77.44  1.43  0.75  77.44-80.02  3.66  1.99  Chapter 5: Photoabsorption and Photoionization Studies of HBr  90  Table 5.3: Appearance potentials for the molecular and dissociative ions produced from HBr.  Ion HBr Br H Br +  +  2+  +  Appearance Potential Experiment Calculated Present ref. [11] ref. [97] ref. [98] Work 10.0 11.66 10.0 14.0 15.0 15.606 15.669 19.0 17.39 17.37 21.0 36.0 a  Derived from heats of formation of the ionic and neutral species  a  Chapter 5: Photoabsorption and Photoionization Studies of HBr  91  table 5.3, together with their appearance potentials. The two isotopes of bromine (masses 79 and 81) occur with nearly equal natural abundance (49% and 51%, respectively). The close spacing of the two Br and the two HBr peaks (all between masses 79-82) precluded complete resolution +  +  at the low mass resolution of the instrument. However with careful peak fitting procedures it was possible to separate and integrate the individual peaks. Two gaussian peaks (necessary to model the asymmetric TOF peaks) of equal width were fitted to each isotopic peak of HBr and Br . +  +  The resulting peak areas reflected the bromine natural abundances. Typical peak fits are shown in figure 5.6 at equivalent photon energies of 12 eV (figure 5.6(a)), below, and at 20 eV (figure 5.6(b)), above the Br appearance potential respectively. For each Br containing ion the 79 Br +  and 81 Br contributions were summed. The ions detected below 66 eV are H , Br , Br , and +  +  2+  HBr , as can be seen in the TOF mass spectra shown in figure 5.7 at 20 and 51.5 eV. +  Photoion branching ratios, as a percentage of total photoionization, were obtained from integration of the peaks in the baseline subtracted TOF mass spectra. The resulting ratios are shown in figure 5.8 and given numerically in table 5.5. The branching ratios for the molecular ion HBr show (figure 5.7) an initially steep decrease, with some levelling off occurring above 25 eV. +  The opposite is observed for Br , where there is an initially steep increase above the appearance +  potential and a second rise in the region of the B 2 (i.e. (7a)" ) onset. The Br branching ratios 2  1  +  then appear to level off somewhat above 30 eV. The dissociative ions H and Br +  2+  both show  slow increases from their appearance potentials up to the Br 3d threshold. No doubly charged HBr is observed. In marked contrast to the situation for HCl the photoion branching ratios for 2+  HBr do not level off. This is to be expected since the EMS [74] studies clearly show that the B 2 2  ionization manifold extends to at least 50 eV and possibly beyond. In this regard it should be  92  Chapter 5: Photoabsorption and Photoionization Studies of HBr  4.70  4.75  4.80  4.85  4.90  Time —of-flight (LIS)  Figure 5.6: Typical peak fits used to resolve isotopic HBr\ Br peaks in the HBr time-of-flight +  mass spectra, at equivalent photon energies of (a) 12 eV, and (b) 20 eV.  Chapter 5: Photoabsorption and Photoionization Studies of HBr  100  Valence  A  Shell H  79  Br  H  50 Br  5 1 . 5 eV  x 20 0  | i  1.0  i* i"i | i 1.5  x 20  2 0 . 0 eV i  i  r | i  2.0  i i 1  'i | i i 'i •> | i 2.5  2+  3.0  i' 'i  t  T 3.5  4.0  4.5  Time-of-Flight (/xs)  Figure 5.7: Time-of-flight mass spectra at 20 and 51.5 eV energy loss in the valence shell region of HBr  Chapter 5: Photoabsorption and Photoionization Studies of HBr  94  Photon Energy (eV)  Figure 5.8: Photoion branching ratios for the molecular and dissociative photoionization of HBr, as a  percentage of total photoionization. The top panes show the absolute photoabsorption  spectrum.  Chapter 5: Photoabsorption and Photoionization Studies of HBr  95  noted that the breakdown model [8] predicts that the fragmentation ratios will only level off at photon energies above the upper limit of the Franck-Condon region for the most tightly bound orbital. In the previous work on HBr in this laboratory [11], photoion branching ratios were obtained up to 40 eV. As discussed in the case of HCI (section 4.2.3), there are some concerns with the reliability and accuracy of the results from the earlier work, especially with regard to the efficiency of the ion detector as a fucntion of m/e. This detector has since been replaced by a microchannel plate device [28] (see section 4.2.3). The photoion branching ratios obtained earlier [11] have been appropriately corrected for the response of the Johnston multiplier and compared with the photoion branching ratios obtained in the present work. As can be seen in figure 5.9, the two sets of branching ratios for the molecular ion agree reasonably well, while there are significant differences in the dissociative ion branching ratios. For HBr and Br , the large local +  +  differences in the 16-23 eV region and of H at all energies are surprising and not understood. +  Absolute partial photoionization oscillator strengths were obtained above 20 eV (where r|; can reasonbly be assumed to by unity) from the product of the absolute photoabsorption and the photoion branching ratio for the molecular and each dissociative ion.  Difficulties were  encountered in determining the photoionization efficiency during the HBr TOF measurements due to an intermittent computer problem which creates an inconsistency in the time scale over long data collection runs. This problem (only discovered after the measurements were completed) does not however affect the relative peak areas in the TOF mass spectra, and thus does not affect the photoion branching ratios. For these reasons, partial photoionization oscillator strengths have only been obtained above 20 eV (where r\i =1). The absolute partial oscillator strengths thus  Chapter 5: Photoabsorption and Photoionization Studies of HBr  1 0 0 -•  o  80 -  40  HBr o o o O O o  60 -  HBr • o  9  9  s'o-8  °  +  •  •  ° o O o  0  o  O  ft  P r e s e n t Work Brion et a l . [ 1 1 ]  Br  +  o » o ° o o o o  o.o..«o«°.  • o  P r e s e n t Work Brion et al [ 1 1 ] .  +  o o o  H • o  96  o o  P r e s e n t Work B r i o n et a l . [ 1 1 ] o o o o  o #  •  •  o „ o 10  15  20  w  25  Photon Energy  30  35  40  (eV)  Figure 5.9: Comparison of photoion branching ratios obtained in the present work with those reported in the previous dipole (e,e+ion) study by Brion et al. [11].  Chapter 5: Photoabsorption and Photoionization Studies of HBr  97  obtained are presented graphically in figure 5.10, and numerically in table 5.6. As can be seen from figure 5.10, the PPOS's for HBr and Br decrease steeply above 20 eV, while H has a +  +  +  broad peak with a maximum at 29 eV. Only a very small amount of Br  2+  oscillator strength is  present in the valence shell above 35 eV. The Br is formed in the valence shell region by direct 2+  dissociative double photoionization and/or via W V Auger processes. These trends will be discussed further in section 5.2.4 below. 5.2.4 Dipole-Induced Breakdown Pathways of HBr As discussed previously (section 4.2.4), the dipole induced breakdown of a molecule can be investigated by considering together the partial oscillator strengths for the molecular and dissociative photoionization with those for production of the electronic ion states. This is because the total absolute photoionization oscillator strength distribution can be partitioned into the partial oscillator strengths for molecular and dissociative photoionization or into the contributions from the electronic ion state partial oscillator strengths (see equations 2.3.4 and 2.3.6 respectively). Also, the partial oscillator strengths for the molecular or any dissociative ion can be partitioned into contributions from the various electronic ion state PPOS's [2, 8, 77]. The dipole induced breakdown of HBr has therefore been investigated in the present work by attempting to fit linear combinations of the electronic ion state partial oscillator strengths to the partial oscillator strength spectra for the molecular and each dissociative ion. In order to accomplish this, the electronic state PPOS's from the previous work [11] were re-evaluated above 20 eV by taking the triple product of the absolute photoabsorption from the present work, the photoelectron branching ratios from the previous work, and an ionization efficiency equal to unity. The electronic state PPOS s thus obtained are shown in figure 5.11 and numerically in table 5.7.  Chapter 5: Photoabsorption and Photoionization Studies of HBr  Photon Energy  Figure 5.10:  (eV)  Absolute partial photoionization oscillator strengths for the molecular  dissociative ionic states produced by photoionization of HBr. The top panels show the abs photoabsorption spectrum.  98  Chapter 5: Photoabsorption and Photoionization Studies of HBr  EMS  99  binding energy spectrum A E= 2.1 eV fwhm ( m a n y body ion s t a t e s )  eV f w h m  AE= 1.4  cn 10 c  -t •  2 +  A I  CD  i_  -t-i  00  5  4  o -I—'  o ,  °  n  0  6 -|  1  ^  1  *  1  '  '  "  t  T  l  BI 2  CL CD  -t—' _3  2 4  O  CO  ^ -9  o  20  ^  22  ,  24  ,  26  p  28  _  30  T  T  32  34  36  38  40  Photon Energy (eV) Figure 5.11: Electronic ion state partial oscillator strengths obtained from photoelectron dipole (e,2e) branching ratios reported by Brion et al. [11] and the presently reported photoabsorption oscillator strengths. The binding energy spectrum (top panel), in the region of the HBr\ B 2 2  many body manifold, is taken from low resolution EMS measurements [74].  Chapter 5: Photoabsorption and Photoionization Studies of HBr  100  The spectra in figure 5.12 show the best fit linear combinations of the electronic state PPOS's to the molecular (HBr , figure 5.12(b)) and total dissociative (Br + FT, figure 5.12(c)) +  +  ion state PPOS's. For HBr , the best linear combination was given by (l.OXTI + 0.4A 2 + +  2  +  0.4B 2). Thus the molecular ion partial oscillator strength arises mostly from the X n state, with 2  2  small contributions from the A 2 and B 2 states. Considering the fragment ion appearance 2  +  2  potentials (see table 5.3) and the ionization potentials (see section 5.2.1), Br must first arise from +  the A 2 state of HBr . Similarly, Ff can only be formed from the B 2 ionization manifold. 2  +  +  +  2  Given the statistics in the two different measurements, the remaining electronic state PPOS is fairly well matched by the total dissociative (Br + H ) ion state PPOS, as can be seen in figure +  +  5.12(c). The total dissociative photoionization can be partitioned into individual contributions from Br and Ff , as can be seen in figure 5.13(b) and (c). The best fit linear combination for Br is +  +  +  given by (0.6A 2 + 0.5B 2). As can be seen in figure 5.13(b), below 21.5 eV Br arises only 2  +  2  +  from the A 2 electronic state. Again it is observed that the contribution from the B 2 state 2  +  2  corresponds to the B 2 state onset in the PES spectrum [74] shown in figure 5.13(a). For the 2  partial oscillator strength of H , the remaining oscillator strength (10%) from the B 2 electronic +  2  state gives a reasonable fit, considering the small magnitude of the partial oscillator strength and the respective data statistics. The breakdown pathways above 40 eV in the valence shell and in the Br 3d region suggested by the present work are summarized in figure 5.14. Photoelectronphotoion coincidence studies would enable a more detailed and accurate analysis of the processes involved in photoionization in this region.  Chapter 5: Photoabsorption and Photoionization Studies of HBr  Photon Energy (eV) Figure 5.12: Absolute partial photoionization oscillator strengths for the molecular and total dissociative photoionization of HBr, with best fit linear combinations of contributions from the electronic ion state partial oscillator strengths. The top panel shows the EMS spectrum obtained at low resolution by Suzuki et al. [74].  102  Chapter 5: Photoabsorption and Photoionization Studies of HBr  J  co  :  (a)  'c CD  1H  >  l  I  L  EMS b i n d i n g e n e r g y  CD = 5°  I  —1  1  I  L  HBr  spectrum A E = 2.1eV f w h m  B I ( m a n y body ion states)  o CD  cr  •  i  .  t  * ,  r  Br + >  Br  CM CD  I  o  +  ( A E = 1eV f w h m )  0.6A + 0 . 5 B ( A E = 1 . 4 e V fwhm)  4 4  cr> c CD  Lo o  o  -i  1  1  1  1  r  (c)  -l-l  o o 0.4 c/) O  O CL  ~\  r  H  o o  H  +  0.1B  o  0.2 4  CD  O CO  <  0.0 i 20  e-r  22  i  1  1  1  24  26  28  30  :  i  i  32  34  f  36  -  38  40  Photon Energy (eV) Figure 5.13: Absolute partial photoionization oscillator strengths for the dissociative ions Br and H , with best fit linear combinations of contributions from electronic ion state partial +  oscillator strengths. The top panel shows the EMS spectrum obtained by Brion et al. [74].  +  103  Chapter 5: Photoabsorption and Photoionization Studies of HBr  HBr + hv Valence (above ~40 eV)  Br 3d Inner Shell (above 70 eV)  Br (Small yields from V W Auger and/or direct dissociative double photoionization) z  HBr  2+  Br  2+  Figure 5.14. Dipole ionic photofragmentation breakdown pathways of HBr in the VUV and soft X-ray regions.  Chapter 5: Photoabsorption and Photoionization Studies of HBr  104  5.2.5 Photoionization Measurements in the Br 3d Inner Shell Region of HBr Time-of-flight mass spectra were also obtained in the Br 3d inner shell excitation and ionization region of HBr at selected equivalent photon energies at 1 eV fwhm energy resolution. Typical TOF spectra are shown in figure 5.15 at energies of 66 eV (inner valence), 71, 75, and 91 eV (selected Br 3d regions). Note that the ion yield scale is relative to the tallest peak, which is not necessarily the peak with the largest area. At 66 eV, in the valence shell continuum below the Br 3d threshold, HBr (m/e = 80, 82) is the highest peak. There is a large amount of H present, +  +  but very little Br . At 71 eV, in the middle of the strong 3d—>a* transition (see figure 5.1, 1 eV 2+  fwhm, andfigure5.5, 0.1 eV fwhm), the HBr peaks are substantially reduced, while the Br , Br +  +  2+  peaks are increasing in intensity. The width of the H peak is appreciably broadened compared +  with the situation in the valence shell pre-edge region at 66 eV. At 75 eV, in the Rydberg excitation region, the relative intensities of the HBr , Br , and Br * peaks decrease, while H +  +  2  +  increases. At 91 eV, well into the Br 3d ionization continuum, HBr and Br further decrease in +  +  relative intensity and Br increases. In addition, a small amount of the doubly charged HBr is 2+  2+  formed, although no doubly charged molecular ions are observed at any of the lower energy TOF mass spectra. At 91 eV, the H peak shows further broadening, indicating high kinetic energy +  protons probably formed in the Coulomb explosion of doubly charged ions (i.e. HBr *). 2  Photoion branching ratios for the Br 3d inner shell region are given infigure5.8 and table 5.5. At the Br 3d excitation threshold, the branching ratio for HBr decreases, while those for H , +  +  Br , and Br increase. For each of H , Br , Br , there is a peak ~71 eV which corresponds to +  2+  +  +  2+  the 3d—»a* virtual valence transition. This is apparently due to processes involving dissociative autoionization (resonance Auger decay [56]) of neutral excited states of HBr down to H + Br, +  Chapter 5: Photoabsorption and Photoionization Studies of HBr  100  61 eV ( V a l e n c e Shell)  HBr  50  0 100  71 eV (3d — * cr*)  79  + 8  CD  81  50  CD > IP O  79  0  Br  2  +  +  .  0 1  Br  2  +  r-,  1  «^«A» i»w.^«. f--|-f,-f •,Vi'*—~*ft»Hni.»« l  t  ^W<WMw~*Ai«^a»».  0  7 5 eV ( 3 d — > R y d b e r g )  CD  H  79  Br  +  31 79  50  +  Br. 81  +  Br  0 100  50  rv /—i  i  9 0 eV ( 3 d — ^ c o n t i n u u m )  H  1.0  1.2  1.4  3.5  4.0  Time-of-flight  4.5  5.0  5.5  (LIS  Figure 5.15: Time-of-flight mass spectra of HBr at selected energies in the inner valence and 3d inner shell regions (1 eV fwhm).  Chapter 5: Photoabsorption and Photoionization Studies of HBr  106  H + Br , and/or ionization to doubly charged HBr *, which rapidly dissociates to form H + Br +  2  +  +  (Coulomb explosion) or to Br + H. In the 3d ionization continuum (91 eV) the relative yields of 2+  HBr and Br further decrease, while H and Br increase. In addition HBr appears for the first +  +  +  2+  2+  time and in low yield (see figure 5.15). The appearance of HBr is consistent with the onset of 2+  Auger processes expected in the 3d ionization continuum but not lower energies in the 3d excitation region. The Auger processes may lead to doubly charged HBr and Br ions and also 2+  2+  Coulomb explosion of HBr to give H + Br . The slightly increased width of the H peak (figure 2+  +  +  +  5.15) is evidence for the latter process and/or dissociative single ionization (or autoionization) to H + Br. +  Partial photoionization oscillator strengths (PPOS's) for molecular and dissociative photoionization in the Br 3d inner shell region are shown in figure 5.10 above. In this energy region the remaining HBr oscillator strength is formed essentially from valence shell processes +  only. The Br PPOS shows a small peak at around 71 eV, corresponding to the strong 3d-»a* +  transition (figures 5.1 and 5.5), as already discussed above with reference to the photoion branching ratios. The H and Br +  2+  PPOS's increase at the Br 3d excitation threshold. H  +  continues to increase in the Rydberg excitation and ionization continuum regions, as also discussed above with reference to the branching ratios in (figure 5.8). In contrast Br  2+  has a  maximum at 71 eV, which drops in the Rydberg region before increasing again in the 3d ionization continuum where a small amount of HBr (see figure 5.15) is also formed. 2+  Chapter 5: Photoabsorption and Photoionization Studies of HBr  107  Table 5.4: Absolute differential photoabsorption oscillator strengths for valence shell and Br 3d inner shell regions of HBr (1 eV fwhm).  Total Photoabsorption Oscillator Strength (df/dE) eV  IOVV  eV  lO^eV  5.0 5.5  eV  0.60 0.57  19.5 20.0  35.95 33.99  34.0 34.5  6.0  1.01  20.5  32.22  6.5  1.75 2.26  21.0 21.5  8.5  2.46 7.02 20.07  22.0 22.5 23.0  9.0  18.68  9.5 10.0  eV  lO^eV  6.01 5.69  57.0 58.0  0.77 0.74  35.0  5.15  59.0  0.71  30.39 29.10  35.5 36.0  4.77 4.42  60.0 61.0  0.69 0.68  27.28 25.75  36.5 37.0  4.27 3.95  62.0 63.0  0.66 0.65  23.5  25.22 23.28  37.5 38.0  3.51 3.49  64.0 65.0  0.64 0.61  12.54 25.86  24.0  22.16  38.5  3.21  66.0  0.61  24.5  21.03  39.0  3.15  67.0  0.61  10.5  36.97  25.0  39.5  2.90  46.39  25.5  40.0  2.68  68.0 69.0  0.59  11.0 11.5  20.21 18.75  52.35  26.0  17.55  41.0  2.44  70.0  1.18  12.0  44.67  26.5  71.0  36.76  27.0  42.0 43.0  2.17  12.5  16.35 15.61  1.95  72.0  1.93 1.75  13.0  40.48  27.5  14.86  44.0  1.78  73.0  1.05  13.5  50.17  28.0  13.73  45.0  1.59  74.0  1.05  14.0  53.83  28.5  12.72  46.0  1.45  75.0  1.49  14.5  51.53  29.0  11.92  47.0  1.38  76.0  1.51  15.0  50.34  29.5  11.18  48.0  1.25  77.0  1.44  15.5  48.62  30.0  10.57  49.0  1.16  78.0  1.39  16.0  47.43  30.5  9.66  50.0  1.08  79.0  1.37  16.5  46.07  31.0  8.89  51.0  1.00  80.0  1.38  17.0  44.27  31.5  8.58  52.0  0.97  81.0  1.39  17.5  42.65  32.0  7.82  53.0  0.92  82.0  1.41  18.0  41.07  32.5  7.24  54.0  0.88  83.0  1.39  18.5  39.33  33.0  6.92  55.0  0.83  84.0  1.46  19.0  37.31  33.5  6.30  56.0  0.79  85.0  1.49  7.0 7.5 8.0  1  1  luVV  1  0.63  1  Chapter 5: Photoabsorption and Photoionization Studies of HBr  108  Table 5.4 (continued): Absolute differential photoabsorption oscillator strengths for valence shell and Br 3d inner shell regions of HBr (1 eV fwhm).  Total Photoabsorption Oscillator Strength (df/dE) eV 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 102 104  1.53 1.58 1.59 1.66 1.71 1.81 1.88 1.90 1.95 1.97 2.09 2.06 2.19 2.27 2.34 2.46  eV 130 132 134 136 138 140 142 144 146 148 150 152 154 156 158 160  lO^eV 4.23 4.35 4.36 4.36 4.50 4.53 4.56 4.63 4.69 4.67 4.71 4.70 4.75 4.72 4.69 4.75  106  2.60 2.76  162 164  108 110  2.90 3.06  166 168  112  3.19  170  4.77 4.76  114 116 118 120 122  3.33 3.47 3.58 3.69 3.88  172 174 176 178 180  4.67 4.70 4.72 4.69 4.75  124 126  3.94  182  4.71  4.10  184  128  4.13  186  l<r eV 2  x  eV 188 190 192 194 196 198 200 202 204 206 208 210 212 214 216 218  lO^eV 4.93 5.00 5.00 5.00 4.98 4.93 4.96 4.94 4.93 4.99 4.96 4.81 4.89 4.79 4.85 4.74  eV 246 248 250 252 254 256 258 260 262 264 266 268 270 272 274 276  lO^eV 4.43 4.39 4.46 4.43 4.34 4.43 4.43 4.34 4.38 4.32 4.23 4.16 4.15 4.20 4.23 4.24  4.78 4.70  220 222  4.80 4.70  278 280  3.91 4.04  4.80  224 226  4.65  282 284  4.03 4.03  286  3.91  232 234 236 238  4.63 4.57 4.55 4.48 4.51  288 290 292 294 296  3.93 3.82 3.82 3.78 3.89  4.37 4.62  298  4.89  240 242  3.72 3.61  4.99  244  4.48  1  228 230  4.69 4.74  1  300  1  Chapter 5: Photoabsorption and Photoionization Studies of HBr  109  Table 5.5 Photoion branching ratios for the molecular and dissociative photoionization of HBr, as a percentage of total photoionization. Photon Energy eV 10.0 10.5 11.0 11.5 12.0 12.5 13.0 13.5 14.0 14.5 15.0 15.5 16.0 16.5 17.0 17.5 18.0 18.5 19.0 19.5 20.0 20.5 21.0 21.5 22.0 22.5 23.0 23.5  Photoion Branching Ratio (%) HBr 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 95.66 90.71 90.29 86.25 77.03 84.83 83.11 84.67 83.78 83.96 83.08 82.62 82.89 82.33 80.87 82.09 80.26 78.66 76.81 74.51  +  Br  4.34 9.29 9.71 13.75 22.97 15.17 16.89 15.34 16.22 16.04 16.89 17.29 16.88 17.33 18.82 17.48 19.19 20.75 22.45 24.41  +  Br  2+  FT  0.03 0.09 0.23 0.34 0.31 0.43 0.55 0.59 0.74 1.09  Chapter 5: Photoabsorption and Photoionization Studies of HBr Table 5.5. (continued):  110  Photoion branching ratios for the molecular and dissociative  photoionization of HBr, as a percentage of total photoionization. Photon Energy eV  HBr  24.0 24.5 25.0 25.5 26.0 26.5 27.0 27.5 28.0 28.5 29.0 29.5 30.0 32.0 34.0 36.0 38.0 40.0 42.0 44.0 46.0 48.0 50.0 56.0 61.0 66.0 71.0 75.0 76.0 81.0 90.0  75.01 72.97 73.43 72.32 71.57 72.02 69.94 72.75 69.48 69.73 69.71 68.31 69.01 68.22 67.46 65.03 62.71 58.23 61.45 58.32 54.84 51.93 50.68 37.29 35.83 31.34 8.74 6.44 8.40 7.46 5.16  Photoion Branching Ratio (%) +  Br 23.83 25.72 25.20 25.95 26.70 25.95 27.60 24.72 27.82 27.26 26.78 28.32 27.40 27.34 27.35 28.83 29.42 33.57 29.08 30.61 32.76 34.96 33.87 42.90 42.48 43.05 46.86 43.10 41.67 28.37 25.15  +  Br  2+  0.10 0.35 0.51 0.70 1.09 1.47 1.65 2.13 3.43 3.61 4.61 20.58 14.90 17.04 23.10 26.00  H  +  1.16 1.31 1.37 1.72 1.73 2.03 2.46 2.53 2.70 3.01 3.51 3.37 3.59 4.43 5.19 6.04 7.52 7.69 8.77 9.98 10.92 11.47 13.32 16.38 18.08 21.00 23.81 35.57 32.89 41.07 43.70  Chapter 5: Photoabsorption and Photoionization Studies ofHBr  111  Table 5.6: Absolute partial differential oscillator strengths for the molecular and dissociative photoionization of HBr.  Photon Energy (eV)  Partial Differential Oscillator Strength (lO^eV ) HBr Br Br H 1  +  20.0 20.5 21.0 21.5  28.17 26.53 24.57 23.89  22.0 22.5 23.0 23.5 24.0 24.5 25.0 25.5 26.0  21.89 20.26 19.37 17.35 16.62 15.35 14.84 13.56 12.56  26.5 27.0  +  2+  +  0.08 0.11 0.09 0.13  11.77 10.92  5.74 5.58 5.?2 5.09, 5.24 5.35 5.66 5.68 5.28 5.41 5.09 4.87 4.69 4.24 4.31  27.5 28.0 28.5 29.0 29.5 30.0  10.36 9.54 8.87 8.31 7.64  4.12 3.82 3.47 3.19 3.17  0.38 0.37 0.38 0.42 0.38  7.29  2.90  0.38  32.0  5.33  2.14  0.35  34.0 36.0  4.05 2.87  1.64 1.27  0.00 0.00  0.31 0.27  38.0 40.0  2.19 1.56  1.03 0.90  0.01 0.01  0.26 0.21  0.15 0.15 0.19 0.25 0.26 0.28 0.28 0.32 0.30 0.33 0.38  Chapter 5: Photoabsorption and Photoionization Studies of HBr  112  Table 5.6 (continued): Absolute partial differential oscillator strengths for the molecular and dissociative photoionization of HBr.  Photon Energy (eV) 42.0 44.0 46.0 48.0 50.0 56.0 61.0 66.0 71.0 76.0 81.0  Partial Differential Oscillator Strength (lO^eV ) HBr Br Br H 1  +  1.33 1.04 0.80 0.65 0.55 0.30 0.24 0.19 0.17 0.13 0.10  +  0.63 0.55 0.48 0.44 0.37 0.34 0.29 0.26 0.90 0.63 0.39  2+  0.02 0.02 0.02 0.02 0.02 0.03 0.02 0.03 0.40 0.26 0.32  0.19 0.18 0.16 0.14 0.14 0.13 0.12 0.13 0.46 0.50 0.57  +  Chapter 5: Photoabsorption and Photoionization Studies of HBr  113  Table 5.7: Partial photoionization differential oscillator strengths for the production of electronic ion states of HBr.  Photon Energy (eV)  Partial Oscillator Strength (lO^eV )  3  1  Xn  A2  20.0  23.79  10.20  20.5  21.59  10.06  21.0  21.20  8.96  21.5  20.32  8.79  0.00  22.0  17.66  7.83  1.79  22.5  15.96  7.26  2.53  23.0  16.32  6.62  2.28  23.5  14.01  5.52  3.75  24.0  14.77  5.26  2.13  24.5  12.44 .  4.96  3.63  25.0  11.74  4.63  3.84  25.5  11.57  3.64  3.54  26.0  9.36  3.40  4.79  26.5  8.86  3.06  4.43  27.0  8.44  2.53  4.64  27.5  8.44  2.67  3.74  28.0  6.55  1.98  5.20  28.5  6.41  1.43  5.09  29.0  6.47  1.82  3.64  29.5  5.39  2.03  3.76  30.0  5.39  1.35  3.83  2  2  +  B2 2  a. Obtained using the dipole (e,2e) "photoelectron" branching ratios of reference [11] and the presently reported absolute photoabsorption oscillator strengths.  Chapter 5: Photoabsorption and Photoionization Studies ofHBr  114  Table 5.7 (continued): Partial photoionization differential oscillator strengths for the production of the electronic ion states of HBr. Photon Energy (eV)  Partial Oscillator Strength (lO^eV ) X TI A2 B2  3  1  2  2  +  2  30.5  4.13  1.01  4.53  31.0  3.38  0.97  4.54  31.5  3.75  1.28  3.55  32.0  2.83  0.78  4.20  32.5  2.41  0.74  3.90  33.0  3.12  0.88  2.92  33.5  1.65  0.39  4.27  34.0  1.97  0.47  3.57  34.5  1.90  0.40  3.40  35.0  1.65  1.07  2.43  35.5  1.31  0.37  2.99  36.0  1.33  0.53  2.56  36.5  1.12  0.46  2.69  37.0  0.99  0.36  2.61  37.5  1.31  0.08  2.12  38.0  1.93  0.46  1.10  38.5  1.13  0.09  2.00  39.0  2.01  0.70  0.44  39.5  1.00  0.08  1.49  40.0  1.26  0.59  0.84  a. Obtained using the dipole (e,2e) "photoelectron" branching ratios of reference [11] and the presently reported absolute photoabsorption oscillator strengths.  Chapter 6 Photoabsorption Studies of Hydrogen Iodide 6.1 Introduction Very little information is available for the photoabsorption of HI. A possible reason for the scarcity of studies involving HI is the difficulty in obtaining a pure sample because of disproportionation in the cylinder (i.e. 2HI ** H + h )• In the present work, great care had to be 2  taken in order to eliminate a large molecular hydrogen impurity present in the HI gas sample as well as signals due to desorption of HBr and HCl from the surfaces in the gas inlet system and the spectrometer. Repeated purgings and evacuations were necessary to obtain the spectrum of pure HI. Similar problems with impurities occurred in earlier photoabsorption studies of hydrogen fluoride [101]. To the best of our knowledge, very few absolute photoabsorption spectra of HI have been published. Roxlo and Mandl [35] reported absolute photoabsorption cross sections for HI in the region 5.4-7.3 eV. Romand [36] and Goodeve and Taylor [99] also obtained measurements in the same region. Wang et al. [100] measured the absolute photoabsorption spectrum of HI in the very limited energy range 10.4 to 11.1 eV. Other than in these restricted energy ranges the valence shell electronic absorption spectrum of HI is apparently unknown. Several photoelectron spectroscopy studies have been performed on HI. Lempka et al. [49] obtained the Hel high resolution photoelectron spectrum of HI in the outer valence region. More recently, Adam et al. [102] used X-ray (hv=1487 eV), He I (hv=21.22 eV) and He II (hv=40.8 eV) radiation to investigate the inner valence photoelectron spectra of HI. Angle resolved PES studies and partial cross-sections obtained using synchrotron radiation have been reported by Bowering et al. [103], and by Carlson et al. [104]. Karlsson et al. [105] and Cutler et 115  Chapter 6: Photoabsorption Studies ofHI  ~ 116  al. [106] have measured the NVV Auger electron spectrum which involves the I 4d inner shell of HI. Brion et al. [74] have measured binding energy spectra at low resolution over the entire valence shell binding energy range using EMS spectroscopy. Photoionization mass spectroscopy has been used to measure relative photoionization cross sections for HI by Tsai and Baer [107], and by Eland and Berkowitz [108]. Absolute photoionization cross sections have been calculated by Raseev et al. [109]. In the present work, absolute photoabsorption oscillator strengths (cross sections) for the valence and I 4d inner shells of HI were obtained from 5 to 120 eV with low resolution (1.0 eV fwhm), and from 5 to 60 eV with high resolution (0.05-0.1 eV fwhm) using dipole (e,e) spectroscopy.  To the best of our knowledge, no previous absolute photoabsorption  measurements of the I 4d inner shell region of HI have been reported. 6.2 Results and Discussion 6.2.1 Electronic Structure The ground state electronic structure of HI can be represented as: core  (4d)  10  14d  (10a) (lla) (6jt) 2  2  Valence  Ionization produces the following ionic states: X  2  n  3 / 2  ,i/2  4  (12a)° Virtual Valence  from (6JI)"\ A 2 from (11a)" , and B 2 2  +  1  2  from (10a)" . Ionization potentials have been reported by Lempka et al. [49] to be 10.38 eV 1  ( X n ) , 11.05 eV ( X n ) , and 13.85 eV (A 2 ). Adam et al. [102] reported values of 21.2, 2  2  3/2  2  +  1/2  22.5, and 23.6 eV for the B 2 many body states involved in the manifold [74]. Karlsson et al. 2  [105] reported the I 4d inner shell ionization potentials with values of 57.49 and 59.25 eV for the I 4ds/2 and I 4d states respectively. 3/2  Chapter 6: Photoabsorption Studies ofHI  111  6.2.2 Photoabsorption Oscillator Strengths 6.2.2.1 Low-Resolution Photoabsorption Measurements of HI The low resolution (1.0 eV fwhm) absolute photoabsorption oscillator strength spectrum of HI from 5 to 120 eV was obtained by normalizing the Bethe-Born converted relative spectrum obtained on the low resolution dipole (e,e) spectrometer (see section 3.2). The low energy onset of the I 4d inner shell (~50 eV) makes use of the VTRK sum rule unreliable (see discussion for HBr in section 5.2.2.1). Thus the S(-2) sum rule was used, together with an experimental static dipole polarizability (a ) of 35.3 a.u. [64], to normalize the valence shell spectrum. N  The low resolution valence shell spectrum (5-40 eV) and the long range spectrum (5-120 eV), including the valence shell and I 4d inner shell regions are shown in figure 6.1 below. The absolute oscillator strengths are given in table 6.3. In the valence shell spectrum (figure 6.1(a)), two broad peaks are visible at ~10 and 12.5 eV, as well as a pronounced shoulder around 8 eV. The high resolution spectra shown and discussed in sections 6.2.2.2 and 6.2.2.3 below provide more detailed investigations of the discrete valence shell and I 4d inner shell regions respectively. In the long range spectrum (figure 6.1(b)), the I 4d inner shell spectrum can be seen above 48 eV. The peak at ~51 eV corresponds to the strong 4d->a* transition, and there is a very broad continuum resonance at ~ 92 eV. The energy scale of the spectrum was set by comparison with the high resolution spectrum, convoluted with 1 eV fwhm. A comparison (figure 6.2) of the low resolution (1 eV fwhm) and high resolution (0.05 eV fwhm, convoluted with 1 eV fwhm) gives a good consistency check of the absolute scale normalization of the spectra obtained using the two different instruments.  Chapter 6: Photoabsorption Studies of HI  118  P h o t o n Energy (eV) Figure 6.1: Absolute photoabsorption differential oscillator strengths (cross sections) for the valence and I 4d inner shell regions of HI. The ionization energies are from references [49, 102].  Chapter 6: Photoabsorption Studies of HI  119  Photon Energy (eV)  Figure 6.2: Comparison of low resolution (1 eV fwhm) and high resolution (0.05 eV fwhm, convoluted with 1 eV) absolute photoabsorption oscillator strengths for HI.  Chapter 6: Photoabsorption Studies ofHI  120  6.2.2.2 Valence Shell High Resolution Photoabsorption Measurements of HI The high resolution (0.05 eV fwhm) valence shell (5-20 eV) absolute photoabsorption spectrum is shown in figure 6.3, with assignments from reference [110]. This represents the first full valence shell spectrum of HI to be reported. The spectrum was obtained from S(-2) sum rule normalization of the relative Bethe-Born converted energy-loss spectrum obtained on the high resolution dipole (e,e) spectrometer (see section 3.3). In the spectrum in figure 6.3(a), it is interesting to note the lack of discrete structure between 11 eV and 14 eV, i.e. between the X n i and A 2 limits. Integrated oscillator strengths (cross sections) for selected energy regions 2  2  +  / 2  in the valence shell are given in table 6.1. Only very few previously published absolute photoabsorption spectra [35, 36, 99, 100] of HI were found, and these are in very limited energy ranges. The most recent of these is by Roxlo and Mandl [35], who used a direct optical technique with a pulsed Xe discharge lamp to measure 2  the photoabsorption spectrum of HI from 230 to 170 nm (5.4-7.3 eV). A single transition at 176 nm (7.04 eV) was observed in their spectrum [35], with a peak height of ~4.2 Mb and an integrated cross section of 1.76 Mb, which is significantly lower than value of 12.38 Mb reported in the present work, as is shown in table 6.1. This large difference is evidence for large line saturation errors in the optical work of Roxlo and Mandl [35], which caused significant reduction of the sharp transition observed at ~7 eV. Two other absolute photoabsorption studies were reported in the work by Roxlo and Mandl [35]: Romand [36] and Goodeve and Taylor [99], who studied the absorption spectrum of HI in the pre-edge outer valence region (below 8 eV). The spectrum reported by Romand shows a single transition at ~176 nm (~7eV) that is somewhat different in shape from that observed by  121  Chapter 6: Photoabsorption Studies of HI  10  15  20  25  P h o t o n Energy (eV)  Figure 6.3: Absolute photoabsorption oscillator strengths (cross sections) for the valence shell of HI, obtained with high resolution (0.05 eV fwhm).  Chapter 6: Photoabsorption Studies of HI  122  Table 6.1: Integrated oscillator strengths (cross sections) for selected energy regions in the valence shell obtained from the present measurements of HI, with comparison to those reported in the literature.  Energy Range  Present Work (Mb)  5.40- 7.3  12.38  6.90- 7.1  10.98  7.50- 7.9  25.35  8.35- 8.68  14.60  8.68- 8.90  7.46  9.01-9.14  3.61  9.14- 9.31  10.06  9.31-9.59  2.56  9.59- 9.75  7.22  9.75- 9.93  11.58  9.93-10.37  38.52  10.37-10.60  21.07  10.60-10.72  10.63  10.72-10.80  6.26  ref. [35] (Mb) 1.76  Chapter 6: Photoabsorption Studies ofHI  123  Roxlo and Mandl [35]. The cross sections obtained by both Romand [36] and Goodeve and Taylor [99] are lower than those reported by Roxlo and Mandl [35]. Wang et al. [100] measured the absolute photoabsorption spectrum of HI in the region between the X n 2  and X n 2  3 / 2  1 / 2  limits (10.4-11.1 eV) with apparently slightly higher resolution  than in the present work. The cross sections reported by Wang et al. are significantly lower (~90%) than those reported in the present work. 6.2.2.3 High Resolution I 4d Inner Shell Photoabsorption Measurements of HI The absolute photoabsorption oscillator strength (cross section) spectrum of the I 4d inner shell of HI, obtained with high resolution (0.1 eV fwhm) from 48 to 60 eV is shown infigure6.4 below. The two broad peaks with maxima at 51 and 52.8 eV correspond to the strong 4d-*a* transitions, with the spacing between the peaks corresponding to the spin-orbit splitting of the 4d5/2,3/2  states. Above 54 eV, some discrete structure is visible, corresponding to transitions to  Rydberg states converging on the  4d /2,3/2 5  ionization limits [105]. Integrated oscillator strengths  (cross sections) for selected energy regions in the I 4d inner shell of HI are given in table 6.2.  124  Chapter 6: Photoabsorption Studies of HI  48  50  52 54 56 Photon Energy (eV)  58  60  ° <  Figure 6.4: Absolute photoabsorption oscillator strength (cross section) spectrum of the I 4d inner shell region of HI, obtained with high resolution (0.1 eV fwhm).  Chapter 6: Photoabsorption Studies of HI  125  Table 6.2: Integrated oscillator strengths (cross sections) for selected energy ranges in the I 4d inner shell of HI, with and without estimated contributions from the valence shell continuum.  Energy Range  df/dE (Mb) Without With Valence Valence  49.38-53.83  13.06  6.29  53.83-54.38  1.00  0.17  54.38-55.69  3.73  1.74  55.69-57.43  5.13  2.47  57.43-58.20  2.22  1.05  58.20-59.92  5.39  2.81  126  Chapter 6: Photoabsorption Studies of HI  Table 6.3: Absolute differential photoabsorption oscillator strengths for the valence shell and I 4d inner shell regions of HI (1 eV fwhm).  Total Photoabsorption Oscillator Strength (df/dE) IOVV eV lOVV eV luVV eV 1  1  6.0  -0.23  16.0  49.50  26.0  9.09  6.5  3.44  16.5  45.82  26.5  8.07  7.0  12.19  17.0  42.12  27.0  7.25  7.5  24.26  17.5  38.66  27.5  6.54  8.0  25.36  18.0  35.56  28.0  5.94  8.5  26.49  18.5  32.85  28.5  5.32  9.0  38.42  19.0  30.49  29.0  4.68  9.5  54.87  19.5  28.38  29.5  4.44  10.0  67.26  20.0  25.63  30.0  4.00  10.5  72.87  20.5  23.61  30.5  3.70  11.0  59.07  21.0  22.32  31.0  3.43  11.5  57.90  21.5  20.58  31.5  2.97  12.0  73.93  22.0  18.97  32.0  2.86  12.5  80.17  22.5  17.08  32.5  2.74  13.0  72.08  23.0  15.72  33.0  2.52  13.5  68.35  23.5  14.24  33.5  2.39  14.0  64.94  24.0  12.87  34.0  2.24  14.5  60.95  24.5  11.93  34.5  2.08  15.0  57.24  25.0  10.87  35.0  2.12  15.5  52.97  25.5  9.76  35.5  1.88  1  Chapter 6: Photoabsorption Studies of HI  127  Table 6.3 (continued): Absolute differential photoabsorption oscillator strengths for the valence shell and 14d inner shell regions of HI (1 eV fwhm).  Total Photoabsorption Oscillator Strength (df/dE) lO^eV 10 eV eV IOVV eV eV 1  2  x  36.0  1.93  58.0  2.57  92.0  18.34  36.5  1.68  60.0  2.90  94.0  18.43  37.0  1.65  62.0  3.56  96.0  18.42  37.5  1.75  64.0  4.47  98.0  17.89  38.0  1.77  66.0  5.47  100.0  17.36  38.5  1.72  68.0  6.52  102.0  16.63  39.0  1.57  70.0  7.57  104.0  15.93  39.5  1.56  72.0  8.72  106.0  14.92  40.0  1.44  74.0  10.03  108.0  13.72  42.0  1.31  76.0  11.29  110.0  12.70  44.0  1.24  78.0  12.57  112.0  11.54  46.0  1.18  80.0  13.72  114.0  10.45  48.0  1.15  82.0  14.89  116.0  9.41  50.0  1.48  84.0  15.87  118.0  8.44  52.0  2.52  86.0  16.81  120.0  7.39  54.0  1.91  88.0  17.58  56.0  2.18  90.0  18.06  -  1  Chapter 7 Concluding Remarks Dipole (e,e) spectroscopy has been used to obtain relative energy-loss spectra of HCl, HBr, and HI at high impact energies (8000 eV or 3000 eV) and at zero degree mean scattering angle. The Bethe-Born converted relative spectra were put onto an absolute photoabsorption scale by sum-rule normalization procedures. In the case of HCl, the VTRK sum rule (S(0)) was used, with extrapolation of the measured spectrum to infinite energy, and taking into account the correction factor for Pauli-excluded transitions.  The accuracy of the absolute scale was  confirmed by the very good value of the dipole polarizability of HCl obtained from S(-2) sum rule analysis of the absolute oscillator strength spectrum. The spectra of HBr and HI were normalized using the S(-2) sum rule, using experimental values of the dipole polarizability, due to the lowenergy onsets of the Br 3d and I 4d inner shells, which preclude use of the VTRK sum rule. Thus absolute photoabsorption oscillator strengths (cross sections) were obtained for HCl, HBr, and HI. Previously published optical oscillator strength measurements using the Beer-Lambert law, have been found to be subject to serious "line-saturation" errors (i.e. linewidth-bandwidth interactions) in the discrete excitation regions of the photoabsorption spectra (see sections 1.3, 4.2.2.2, 5.2.2.2, 6.2.2.2). Previous absolute photoabsorption data in the literature for the valence and inner shells of the halogen halides have been found to be quite limited. Where such data are available they are generally over rather small energy regions. Few studies have focused on HCl, and even fewer on HBr and HI. The present work has added significantly to scientific information in this area, and is the first study in which absolute photoabsorption data was obtained for all three molecules over  128  Chapter 7. Concluding Remarks  129  the UV, vacuum UV and soft X-ray regions, covering the valence as well as selected inner shells of the molecules (CI 2p, Br 3d, and I 4d for HCI, HBr, and HI respectively). In the cases of HCI and HBr, the present work gives absolute photoabsorption spectra for the entire range from 5 to 280 eV for HCI, and 5 to 300 eV for HBr. The present work has served to test the reliability of earlier, more limited energy range dipole (e,e) and dipole (e,e+ion) studies of HCI and HBr in this lab [10, 11]. The results from the earlier and present work have been compared, and certain problems anticipated from instrumental and data analysis methods used in the earlier work were confirmed. Comparisons have also been made with other published direct optical studies. In general these earlier Beer-Lambert law measurements have much lower cross sections in the discrete region than those measured in the present work due to line saturation errors in the optical work. In the case of HI, absolute photoabsorption studies prior to the present work were limited to the regions below 8 eV, and between 10 and 11 eV. The present work is therefore the first absolute photoabsorption study to cover the entire valence and the I 4d inner shell regions, from 5 to 120 eV with low resolution (1.0 eV fwhm), and from 5 to 60 eV with high resolution (0.05-0.1 eV fwhm). The previously published absolute photoabsorption data were found to be significantly higher than the cross sections measured in the present work. Dipole (e,e+ion) measurements for HCI and HBr were used to obtain photoion branching ratios and partial photoionization oscillator strengths (cross sections) for production of the molecular and dissociative ions. A consideration of these partial oscillator strengths for molecular and dissociative photoionization, together with partial oscillator strengths for the electronic states of the molecular ion (from dipole (e,2e) spectroscopy) has made possible some investigation of the dipole-induced breakdown pathways of HCI and HBr at V U V and soft X-ray energies.  Chapter 7. Concluding Remarks  130  Contributions from the various electronic ion state partial oscillator strengths to the partial oscillator strengths for the molecular and each dissociative ion were obtained. The present work could be expanded in the future to include photoionization studies of HI, as well as photoion-photoion or photoelectron-photoion coincidence studies of HCl, HBr, and HI to give a more detailed analysis of the dipole breakdown pathways of each molecule. The numerical data for the photoabsorption of HCl, HBr, and HI at both low and high resolution can be obtained as ascii data files at the web location ftp://chem.ubc.ca/pub/cooper.' Further details can be found in reference [7].  REFERENCES [I]  C.E. Brion, Comm. At. Mol. Phys., 16(5), (1985) 249.  [2]  J.W. Gallagher, C.E. Brion, J.A.R. Samson, P.W. Langhoff, JILA Data Center Rep. no. 32, (1987).  [3]  W.F. Chan, G. Cooper, and C.E. Brion, Phys .Rev. A, 44 (1991) 186.  [4]  W.F. Chan, G. Cooper, and C.E. Brion, Phys. Rev. A, 45 (1992) 1420.  [5]  W.F. Chan, G. Cooper, and C.E. Brion, Chem. Phys. 170 (1993) 123.  [6]  W.F. Chan, G. Cooper, R.N.S. Sodhi and C.E. Brion, Chem. Phys. 170 (1993) 81.  [7]  T.N. Olney, N.M. Cann, G. Cooper, and C.E. Brion, to be published (1997).  [8]  F. Carnovale and C.E. Brion, Chem. Phys. 74 (1983) 253.  [9]  F. Carnovale, R.Tseng, and C.E. Brion, / . Phys. B: At. Mol. Opt. Phys. 14 (1981) 4771.  [10]  S. Daviel, Y. Iida, F. Carnovale, and C.E. Brion, Chem. Phys. 83 (1984) 319.  [II]  C.E. Brion, Y. Iida, F. Carnovale and J.P. Thomson, Chem. Phys. 98 (1985) 327.  [12]  J. A. R. Samson, Techniques of Vacuum Ultraviolet Spectroscopy, John Wiley & Sons, Inc., New York, 1967.  [13]  M.S. Lubell, Can. J. Phys. 74 (1996) 713.  [14]  C.E. Brion, S.Daviel, R. Sodhi, and A.P. Hitchcock, AIP Conf. Proc, 94 (1982) 429.  [15]  R.D. Hudson, Rev. Geophys. Space Phys. 9 (1971) 305.  [16] . G.M. Lawrence, D.L. Mickey and K. Dressier, / . Chem. Phys. 48 (1968) 1989. [17]  H. Bethe, Awz. Phys. 5 (1930) 325.  [18]  M. Inokuti, Rev. Mod. Phys. 43 (1971) 297.  [19]  W.F. Chan, G. Cooper, X. Guo and C.E. Brion, Phys. Rev. A 45 (1992) 1420.  [20]  M.J. Van der Wiel, Physica 49 (1970) 411.  131  Appendix:References  132  [21]  J. Berkowitz, Photoabsorption, photoionization and photoelectron spectroscopy, Academic Press, New York, 1979.  [22]  G.R. Burton, W.F. Chan, G. Cooper, C.E. Brion, A. Kumar and W.J. Meath, Can J Chem, 71 (1993) 341.  [23]  J.L. Dehmer, M. Inokuti, and R.P. Saxon, Phys. Rev. A 12 (1975) 102.  [24]  C. Backx, R.R. Tol, G.R. Wight, and M.J. Van der Wiel, J Phys B: At Mol Phys 9 (1975) 2050.  [25]  C. Backx and M.J. Van der Wiel, J Phys B: At Mol Phys, 8 (1975) 3020.  [26]  Y. Iida, F. Carnovale, S. Daviel, and C.E. Brion, Chem. Phys., 105 (1986) 211.  [27]  X. Guo, G. Cooper, W.F. Chan, G.R. Burton and C.E. Brion, Chem. Phys. 161 (1992) 453.  [28]  G. Cooper, Y. Zheng, G.R. Burton, and C.E. Brion, Rev. Sci. Instrum., 64 (1993) 1140.  [29]  S. Daviel, C.E. Brion, and A.P. Hitchcock, Rev. Sci. Instrum. 55 (1984) 182.  [30]  W.L. Wiese, M.W. Smith and B.M. Glenon, Atomic Transition Probabilities, Vol 1. Hydrogen through Neon, NSRDS-NBS Circular No. 4 (U.S. GPO, Washington, D.C., 1966).  [31]  G. Cooper, T.N. Olney, and C.E. Brion, Chem. Phys. 194 (1995) 175.  [32]  T. Ogawa, T. Tsuboi, and K. Nakashima, J. Chem. Phys. 101 (1994) 3696.  [33]  J.B. Nee, M. Suto, and L.C. Lee, / . Chem. Phys., 85 (1986) 719.  [34]  R.G. Wang, M.A. Dillon, and D. Spence, J. Chem. Phys. 80 (1984) 63.  [35]  C. Roxlo and A. Mandl, / . AppL Phys. 51 (1980) 2969.  [36]  J. Romand, Ann. Phys. 4 (1949) 35.  [37]  J. Myer and J. Samson, / . Chem. Phys. 52 (1970) 266.  [38]  H. Frohlich and M. Glass-Maujean, Phys. Rev. A 42 (1990) 1396.  [39]  W. Hayes and F.C. Brown, Phys. Rev. A 6 (1972) 21.  Appendix.-References  133  K. Ninomiya, I. Ishiguro, S. Iwata, A.Mikuni, and T. Sasaki, / . Phys. B: At. Mol. Opt. Phys. 14 (1981) 1777. D.T. Terwilliger and A.LSmith,/. Mol. Spec. 45 (1973) 366. D.S. Ginter and M.L. Ginter, / . Mol. Spec. 90 (1981) 177. S.G. Tilford and M.L. Ginter/ Mol Spec 40 (1971) 568. S.G. Tilford, M.L. Ginter, and J.T Vanderslice, / . Mol. Spec. 33 (1970) 505. J.K. Jacques and R.F.Barrow, Proc. Phys. Soc. London 73 (1958) 538. T.A York and J. Comer, / . Phys. B: At. Mol. Phys. 17(1984) 2563. D.A. Shaw, D. Cvejanovic, G.C. King, and F.H Read, / . Phys. B: At. Mol. Phys. 17 (1984) 1173. A.J.C. Nicholson,/. Chem. Phys. 43 (1965) 1171. H.J. Lempka, T.R. Passmore and W.C. Price, Proc. Roy. Soc. A304 (1968) 53. M.J. Weiss, G.M. Lawrence, and R. A. Young,/. Chem. Phys. 52 (1970) 2867. M.Y. Adam, Chem. Phys. Lett. 128 (1986) 280. S. Svensson, L. Karisson, P. Baltzer, B. Wannberg, U. Gelius, M.Y. Adam, / . Chem. Phys. 89 (1988) 7192. A.A. Wills, D. Cubric, M. Ukai, F. Currell, B.J. Goodwin, T. Reddish, and J. Comer, / . Phys. B: At. Mol. Opt. Spec. 26 (1993) 2601. K. Faegri Jr. and H.P. Kelly, Chem. Phys. Lett. 85 (1982) 472. G. Cooper, G.R. Burton and C.E. Brion,/. El. Spec. 73 (1995) 139. J.W. Au, G.R. Burton and C.E. Brion, Chem. Phys. submitted (1997). M. Bettendorf, S.D. Peyerimhoff and R.J. Buenker, Chem. Phys. 66 (1982) 261. K.P. Huber and G. Herzberg, Molecular spectra and molecular structure, Vol 4, Constants of diatomic molecules, Van Nostrand, Princeton, 1979. D.M. Hirst and M.F. Guest, Mol. Phys. 41 (1980) 1483.  Appendix.-References  134  I.H. Suzuki, C.E. Brion, E. Weigold, and G.R.J. Williams, Intl. J. Quant. Chem. 18 (1980) 275. B. P. Hollebone, C.E. Brion, E.R. Davidson, and C. Boyle, Chem. Phys. 173 (1993) 193. M. Inokuti, private communication Version (2) June 4,1994. C. Cuthbertson and M. Cuthbertson, Phil. Trans R. Soc. Lond. A, 213 (1913) 1. H.A. Stuart, Landolt-Bornstein, Atom und Molekularphysik, (Springer-Verlag, Berlin, Vol 1, Part 3, pp 509,1951). A.A. Maryott and F. Buckley, U.S. National Bureau of Standards Circular No. 537, (US GPO, Washington, DC, 1953). U. Hohm and U. Trumper, Chem. Phys. 189 (1994) 443. A. Kumar and W.J. Meath, Can. J. Chem. 63 (1985) 1616. J. Dougherty and M.A. Spackman, Mol. Phys. 82 (1994) 193. T.A. Carlson, M.O. Krause, F.A. Grimm, P.R. Keller and J.W. Taylor, Chem. Phys. Lett. 87 (1982) 552. M.W. Chase, Jr, C.A. Davies, J.R. Downey, Jr, D J . Frurip, R.A. McDonald, and A.N. Syverud,/. Phys. Chem. Ref Data, 14 (1985). M. Krauss, J.A. Walker, and V.H. Dibeler, / . Res. Nat. Bur. Stand. 72A (1968) 281. R.E. Fox, / . Chem. Phys. 26 (1957) 1281. F.H. Dorman, J.D. Morrison,/. Chem. Phys. 35 (1961) 575. C.E. Brion, I.E. McCarthy, I.H. Suzuki, E. Weigold, G.R.J. Williams, K.L Bedford, A.B. Kunz, and R. Weidman, / . El. Spec. Rel. Phen. 27 (1982) 83. M.J. Van der Wiel and C.E. Brion, J. El. Spec. 1 (1972/1973) 309. F. Carnovale, A.P. Hitchcock, J.P.D. Cook, and C.E. Brion, Chem. Phys. 66 (1982) 249. K.H. Tan, C.E. Brion, P.E. Van der Leeuw, and M.J. Van der Wiel, Chem. Phys. 29 (1978) 299. J.B. Nee, M. Suto, and L.C. Lee, / . Chem. Phys., 85 (1986) 4919.  Appendix:References  135  [79]  G. Brasseur and S. Solomon, Aeronomy of the Middle Atmosphere, Reidel, Amsterdam (1984).  [80]  P.D. Brewer, S. Halle, and R.M. Osgood, Jr., Appl. Phys. Lett. 45 (1984) 475.  [81]  J. Huebert and R.M. Martin, J. Phys. Chem. 72 (1968) 3046.  [82]  A. Goodeve and W.C. Taylor, Proc. R. Soc. London. Ser. A 152 (1935) 221.  [83]  M.L. Ginter and S.G. Tilford, / . Mol. Spec. 34 (1970) 206.  [84]  M.L. Ginter and S.G. Tilford, / . Mol Spec. 37 (1971) 159.  [85]  D.S. Ginter, M.L. Ginter and S.G. Tilford, J. Mol. Spec. 90 (1981) 152.  [86]  M.A. Baig, J. Hormes, J.P. Connerade and W.R.S. Garton, / . Phys. B: At. Mol. Phys. 14 (1981) L147.  [87]  D.T. Terwilliger and A.L Smith, J. Mol. Spec. 50 (1974) 30.  [88]  R. Puttner, M. Domke, K. Schulz, A. Gutierrez and G. Kaindl, / . Phys. B: At. Mol. Opt. Phys. 28(1995)2425.  [89]  P. Morin and I. Nenner, Phys. Rev. Lett. 56 (1986) 1913.  [90]  M.Y. Adam, M.P. Keane, A. Naves de Brito, N. Correia, P. Baltzer, B. Wannberg, L. Karlsson and S. Svensson, / . El. Spec. Rel. Phen. 58 (1992) 185.  [91]  M. Salzmann, N. Bowerin, H-W. Klausing, R. Kuntze and U. Heinzmann, / . Phys. B: At. Mol. Opt. Phys. 27 (1994) 1981.  [92]  Z.F. Liu, G.M. Bancroft, K.H. Tan and M. Schachter, Phys. Rev. Lett. 72 (1994) 621.  [93]  Z.F. Liu, G.M. Bancroft, K.H. Tan and M. Schachter, J. El. Spec. Rel. Phen. 67 (1994) 299.  [94]  K. England, T. Reddish, and J. Comer, J. Phys. Lett. 23 (1990) 2151.  [95]  D.W. Turner, C. Baker, A.D. Baker and C R . Brundle, Molecular Photoelectron Spectroscopy, Wiley-Interscience, London (1970).  [96]  J. Delwiche, P. Natalis, J. Monigny and J.E. Collin, J. El. Spec. Rel. Phen. 1 (1973) 219.  [97]  J.G. Lias, J.E. Bartness, J.F. Liebman, J.L. Holmes, R.D. Levin, and W.G. Mallard, / . Phys. Chem. Ref. Data, 17 (1988) Suppl No. 1, 1.  Appendix.-References  136  [98]  H.M. Rosenstock, K. Draxl, B.W. Steiner and J.T. Herron, / . Phys. Chem. Ref. Data 6 (1977) Suppl. No. 1  [99]  A. Goodeve and W.C. Taylor, Proc. R. Soc. London Ser. A 154 (1936) 181.  [100] H.T. Wang, W.S. Felps, G.L. Findley, A.R.P. Rau, and S. P. McGlynn, J. Chem. Phys. 67 (1977)3940. [101] C.E. Brion and A.P. Hitchcock, J. Phys. B 13 (1980) L677. [102] M.Y. Adam, M.P. Keane, A. Naves de Brito, N. Correia, B. Wannberg, P. Baltzer, L. Karlsson and S. Svensson, Chem. Phys. 164 (1992) 123. [103] N. Bowering, M. Muller, M. Salzmann, and U. Heinzmann, J. Phys. B: At. Mol. Opt. Phys. 24 (1991) 4793. [104] T.A. Carlson, A.Fahlman, M.O. Krause, P.R. Keller, J.W. Taylor, T. Whitley, and F.A. Grimm,/. Chem. Phys. 80 (1984) 3521. [105] L. Karlsson, S Svensson, P. Baltzer, M. Carlsson-Gothe, M.P. Keane, A. Naves de Brito, N. Correia, and B. Wannberg, / . Phys. B: At. Mol. Opt. Phys. 22 (1989) 3001. [106] J.N. Cutler, G.M Bancroft, and K.H. Tan, / . Phys B: At. Mol. Opt. Phys. 24 (1991) 4897. [107] B.P. Tsai and T. Baer, / . Chem. Phys. 61 (1974) 2047. [108] J.H.D. Eland and J. Berkowitz, / . Chem. Phys. 67 (1977) 5034. [109] G. Raseev, F. Keller, and H. Lefebvre-Brion, Phys. Rev. A 36 (1987) 4759. [110] J.N. Cutler, G.M. Bancroft and K.H. Tan, / . Chem. Phys. 97 (1992) 7932.  

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
United States 10 1
Russia 2 0
China 1 2
City Views Downloads
Ashburn 6 0
Saint Petersburg 2 0
Unknown 2 1
Los Angeles 1 0
Sunnyvale 1 0
Beijing 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}
Download Stats

Share

Embed

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"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0059604/manifest

Comment

Related Items