UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Threshold electron impact excitation Olsen, Lyle Allen Roger 1971

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

Item Metadata

Download

Media
831-UBC_1971_A1 O48.pdf [ 4.24MB ]
Metadata
JSON: 831-1.0060045.json
JSON-LD: 831-1.0060045-ld.json
RDF/XML (Pretty): 831-1.0060045-rdf.xml
RDF/JSON: 831-1.0060045-rdf.json
Turtle: 831-1.0060045-turtle.txt
N-Triples: 831-1.0060045-rdf-ntriples.txt
Original Record: 831-1.0060045-source.json
Full Text
831-1.0060045-fulltext.txt
Citation
831-1.0060045.ris

Full Text

THRESHOLD ELECTRON IMPACT EXCITATION BY L.A.R. OLSEN B.Sc. University of B r i t i s h Columbia, 1968 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n the Department of CHEMISTRY We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA June, 1971 In presenting th i s thes i s in pa r t i a l f u l f i lment of the requirements fo r an advanced degree at the Univers i ty. of B r i t i s h Columbia, I. agree that the L ib ra ry sha l l make i t f r ee l y ava i l ab le for reference and study. I fu r ther agree that permission for extens ive copying of th i s thes i s fo r s cho la r l y purposes may be granted by the Head of my Department or by his representat ives . It is understood that copying or pub l i ca t i on of th i s thes is fo r f i nanc i a l gain sha l l not be allowed without my wr i t ten permiss ion. Department of The Un ivers i ty o f B r i t i s h Columbia Vancouver 8, Canada ABSTRACT Threshold electron impact spectra have been obtained for helium, neon, argon, krypton, xenon and carbon monoxide using a v e l o c i t y -selected electron beam. Zero energy electrons ejected at e x c i t a t i o n thresholds are scavenged by sul f u r hexafluoride and the r e s u l t i n g SFg ions are detected using a mass f i l t e r . O p t i c a l l y forbidden t r a n s i t i o n s are observed to make a large contribution to the t o t a l threshold e x c i t a t i o n . S i n g l e t - t r i p l e t t r a n s i t i o n s are especially prominent i n helium and carbon monoxide. Ex c i t a t i o n of outer s and p electrons i s observed for neon, argon, krypton and xenon. - i i i -TABLE OF CONTENTS Page CHAPTER ONE: INTRODUCTION 1 CHAPTER TWO: ELECTRON-ATOM AND ELECTRON-MOLECULE INTER-ACTIONS 4 2.1 Selection Rules 5 2.2 Autoionization 8 2.3 Temporary Negative Ions 14 CHAPTER THREE: INSTRUMENTAL METHODS IN ELECTRON SPECTROSCOPY 18 3.1 Electron Mono chroma tors 20 3.1.1 The Retarding P o t e n t i a l Difference Method.. 20 3.1.2 Deflection Analyzer Employing E l e c t r o s t a t i c Fields 21 3.1.3 Deflection Analyzers Employing Magnetic Fields 23 3.1.4 Deflection Analyzers Employing Crossed E l e c t r o s t a t i c and Magnetic Fields 24 3.2 Electron Scattering Experiments 24 3.2.1 Transmission Experiments 24 3.2.2 Energy-Loss Experiments 26 3.2.3 Threshold Ex c i t a t i o n Methods 27 CHAPTER FOUR: EXPERIMENT 32 4.1 The Instrument 32 4.1.1 The 127° Electron Velocity Selector 34 4.1.2 The Monopole 36 4.1.3 The Detection and Data Retrieval Systems... 40 - i v -Page 4.2 The SF 6 Scavenging Method 42 4.3 Calibration of Energy Scales 51 CHAPTER FIVE: RESULTS AND DISCUSSION 52 5.1 Helium 52 5.2 Neon 59 5.3 Argon 62 5.4 Krypton 7 1 5.5 Xenon '•• 75 5.6 Carbon Monoxide 80 5.7 Summary and Conclusions 87 BIBLIOGRAPHY 8 9 - v -LIST.OF FIGURES Figure Page 1 The Instrument 33 2 Residual Mass Spectrum 38 3 Electron Impact Spectrometer and Data R e t r i e v a l System 41 4 SFg Ion Current as a Function "of Electron. Energy .. 43 5 The Negative Ions of Tetracyanoethylene 49 6 The Tetracyanoethylene Scavenging Curve of Krypton.. ' 50 7 The Threshold E l e c t r o n Impact E x c i t a t i o n Spectrum of Helium (High Resolution) 53 8 The Threshold E l e c t r o n Impact E x c i t a t i o n Spectrum of Helium (Low Resolution) 54 9 The Threshold Electron Impact Spectrum for the E x c i t a t i o n of the 2p E l e c t r o n i n Neon . 60 10 The Threshold E l e c t r o n Impact Spectrum for the E x c i t a t i o n of the 3p E l e c t r o n i n Argon '. 63 11 Comparison of Argon E x c i t a t i o n Spectra at Three D i f f e r e n t Electron Energies 64 12 The Threshold E l e c t r o n Impact Spectrum of Argon i n the Region of the 3s El e c t r o n E x c i t a t i o n 67 13 The Threshold E l e c t r o n Impact Spectrum for the E x c i t a t i o n of the 4p El e c t r o n i n Krypton .. 72 14 The Threshold E l e c t r o n Impact Spectrum of Krypton i n the Region of the 4s El e c t r o n E x c i t a t i o n 73 15 The Threshold Electron Impact Spectrum for the E x c i t a t i o n of the. 5p E l e c t r o n i n Xenon 76 16 The Threshold E l e c t r o n Impact Spectrum of Xenon i n the Region of the 5s E l e c t r o n E x c i t a t i o n 77 17 The Threshold E l e c t r o n Impact E x c i t a t i o n Spectrum of Carbon Monoxide 81 18 The Threshold E l e c t r o n Impact E x c i t a t i o n Spectrum of the a^n State of Carbon Monoxide 84 - v i -LIST OF TABLES Table Page 1 Energy Levels (in eV) for E x c i t a t i o n of the 3s El e c t r o n i n Argon 68 2 Energy Levels ( i n ev) for E x c i t a t i o n of the 4s El e c t r o n i n Krypton ; 74 3 Energy Levels ( i n eV) for E x c i t a t i o n of the 5s El e c t r o n i n Xenon 78 4 Approximate Rela t i v e I n t e n s i t i e s of V i b r a t i o n a l 3 Levels of the a II State of Carbon Monoxide , 85 - v i i -ACKNOWLEDGEMENT This work was carried out under the supervision of Dr. CE. Brion, to whom I am indebted for his u n f a i l i n g encouragement and invaluable advice. I would also l i k e to thank my colleagues i n the laboratory, G.E. Thomas and W.B. Stewart, for t h e i r u nselfish assistance on many occasions, and the National Research Council for i t s much appreciated f i n a n c i a l support. - 1 -CHAPTER ONE INTRODUCTION It has been recognized for many years that effects such as ex c i t a t i o n , d i s s o c i a t i o n and i o n i z a t i o n can be produced i n a substance by i n e l a s t i c c o l l i s i o n s with energetic electrons. The nature of these interactions i s important to the understanding of radiation-induced decompositions and polymerizations, astrophysical phenomena, gas discharges, and controlled thermonuclear reactions. These c o l l i s i o n s and t h e i r products are also of significance i n topics related to atmospheric behaviour, including meteorology, medicine and radio-t e l e v i s i o n communication. At high electron energies, the results of i n e l a s t i c electron-atom and electron-molecule c o l l i s i o n s are very s i m i l a r to the results produced by photons. But at lower energies, electrons may undertake negative ion formation and e x c i t a t i o n of o p t i c a l l y inaccessible states i n addition to a l l the electronic transitions possible with photons. Another major difference i s that a photon must have exactly the appropriate energy (hv) to i n i t i a t e a t r a n s i t i o n between discrete states, while for electron impact e x c i t a t i o n the requirement i s only that the 1 2 electron energy (-^  mv ) be equal to or greater than the required energy (since the excess can be carried away as k i n e t i c energy of the scattered - 2 -electron and other products). The point at which the electron energy i s just equal to the e x c i t a t i o n energy (where the scattered electron would have zero energy) i s known as the "threshold of e x c i t a t i o n " . The electron c o l l i s i o n may be studied by observing such processes as i o n i z a t i o n or photon emission from the target or by observing the scattered electrons (electron spectroscopy). The data obtained from electron spectroscopy experiments w i l l be sensitive to the method of detection, however, for the electrons may be scattered a n i s o t r o p i c a l l y . The results would then depend at which angle, and over which s o l i d angle, the scattered electrons were detected. This thesis describes a study of the electron impact e x c i t a t i o n of some simple gaseous species with electrons of threshold energies. This was done using a mass spectrometer equipped with an electron source that provides electron beams nossessing narrow energy spreads. The detection of scattered zero energy electrons i s accomplished by mixing small amounts of sulfur hexafluoride with the sample gas to be studied. Zero or thermal energy electrons that are produced from i n e l a s t i c c o l l i s i o n s with the target gas are "scavenged" by SF, to form o SF^ , which i s detected i n the mass spectrometer. SF^ i s a useful scavenger because i t captures electrons of very low or thermal energies with very high p r o b a b i l i t y . By recording the SF^ ion current as a function of the incident electron beam energy, a l l i n e l a s t i c processes re s u l t i n g i n scattered electrons of thermal energies w i l l be observed. This may appropriately be called a threshold electron e x c i t a t i o n spectrum. The usefulness of the e x c i t a t i o n spectrum w i l l of course depend on the s i g n a l to noise l e v e l and the energy resolution. The main problem - 3 -i s i n obtaining a "monoenergetic" electron source of s u f f i c i e n t i n t e n s i t y . The usual approach to this problem i s to use a hot filament, which generates an intense f l u x of electrons but with a wide d i s t r i b u t i o n of energies, and a monochromator device to f i l t e r out a l l but a narrow s l i c e of the electron energy d i s t r i b u t i o n . The various kinds of mono-chromators that have been used w i l l be covered i n Section 3.1. An electron beam that has passed through a monochromator w i l l be considered "monoenergetic" i n this thesis, even though the energy d i s t r i b u t i o n of such a beam w i l l have a ful l - w i d t h at half-maximum (FWHM) of 0.01 to 0.1 eV. The term resolution w i l l be used synonymously with ful l - w i d t h at half-maximum and "half-width" i n this thesis. The electron c o l l i s i o n processes that are important from the point of view of th i s work, namely electronic and v i b r a t i o n a l excitation (including excitation of o p t i c a l l y forbidden and autoionizing states) and temporary negative ion formation, w i l l be considered i n Chapter 2. Chapter 3 i s a summary of the main types of electron spectroscopy experiments, and of the various kinds of electron monochromators that have been used. Further details of the electron scavenger method may be found i n Chapters 3 and 4. - 4 -CHAPTER TWO ELECTRON-ATOM AND ELECTRON-MOLECULE INTERACTIONS A theory which would give electron impact cross sections obviously would be of great value. Unfortunately, even for electron-hydrogen atom scattering the quantum-mechanical equations are too d i f f i c u l t to solve exactly, and approximations must be made i n order that any solutions are obtained at a l l . One of the simplest and best known of these i s the Born approximation, which assumes a weak i n t e r -action between the target atom and the p r o j e c t i l e electron. This i s a reasonable assumption only i f the k i n e t i c energy of the p r o j e c t i l e electron i s large compared with the energy of the l e v e l being excited. Thus the Born approximation i s generally found to be an adequate description at electron energies of 100 to 200 eV and higher, but i s completely i n v a l i d at threshold energies. Refinements to allow for such effects as exchange, d i s t o r t i o n and p o l a r i z a t i o n improve the agreement somewhat at lower energies, but the calculations soon become p r o h i b i t i v e l y d i f f i c u l t for a l l but the very simplest systems. Numerous other approaches have been used i n attempts to calculate electron impact scattering cross sections, including the Born-Oppenheimer approximation, e f f e c t i v e range theory, close-coupling methods, the impulse approximation, as w e l l as c l a s s i c a l and semi-classical methods, but the only methods that have so far been successful at threshold energies are the close-coupling method and one of i t s v a r i a t i o n s , the polarized o r b i t a l approximation. Even these methods are so complex, however, that t h e i r usefulness has been l i m i t e d to hydrogen, helium and the a l k a l i metals. A comprehensive review of the theory of electron-atom c o l l i s i o n s has been given by Peterkop and Veldre (1966). Numerous other review a r t i c l e s include those of Mott and Massey (1965), Gerjuoy (1965), McDaniel (1964), and Moiseiwitsch and Smith-(1968). 2.1 Selection Rules One concept that i s frequently mentioned throughout t h i s thesis i s that of o p t i c a l l y allowed ( o p t i c a l l y accessible, dipole allowed) versus o p t i c a l l y forbidden t r a n s i t i o n s . This has i t s o r i g i n i n that, according to wave mechanics, the t r a n s i t i o n p r o b a b i l i t i e s for e l e c t r i c dipole r a d i a t i o n can be calculated from the eigenfunctions corresponding to the i n i t i a l and f i n a l states. The t r a n s i t i o n p r o b a b i l i t y vanishes for some combinations of states, so that t r a n s i t i o n s between these states are o p t i c a l l y forbidden. This leads to a set of s e l e c t i o n rules which are summarized for the case of atomic e x c i t a t i o n by Herzberg (1944) as follows:• 1. The se l e c t i o n rule for the t o t a l angular momentum quantum number J i s AJ = 0, ±1 (J = 0 «~H- J = 0) . 2. Even terms can combine only with odd terms and odd only with even. Even terms are those r e s u l t i n g from electron configurations which have the sum of o r b i t a l angular momenta i n the atom equal to an even number = 0, 2, 4...). This i s known as the Laporte Rule. - 6 -3. In the case of! Russell-Saunders coupling (assumes weak coupling between the spin and o r b i t a l angular momenta of the i n d i v i d u a l electron), additional selection rules are AS = 0 for the spin angular momentum quantum number S, and AL = 0, ±1 for the o r b i t a l angular momentum quantum number L. At the same time, A£ = i l for the electron making the quantum jump. 4. Transitions i n which two or more electrons jump at the same time are not forbidden by any s t r i c t selection r u l e , but w i l l be considerably weaker than one electron quantum jumps. For two electrons i and k, the Laporte Rule becomes: when A£^ = ±1, AJl^ must be 0 or ±2 and vice versa. 5. For ( j , j ) coupling, A j ^ = 0, ±1 for the electron performing the quantum jump. Herzberg points out that these selection rules are true only to a f i r s t approximation - tra n s i t i o n s i n v i o l a t i o n of these se l e c t i o n rules do occur even for e l e c t r i c dipole radiation (and c e r t a i n l y do for magnetic dipole or e l e c t r i c quadrupole r a d i a t i o n ) , but these are very weak. Thus " o p t i c a l l y forbidden" should be considered to be a r e l a t i v e term. The selection rules depend on the nature of the coupling-Russell-Saunders coupling holds for the ground and excited states of helium, but for the other rare gases t h i s type of coupling applies only to the ground states. For elements of higher atomic number and for more highly excited states, ( j , l ) or ( j , j ) coupling become important. Thus the selection rule prohibiting intercombination of states of diffe r e n t m u l t i p l i c i t i e s w i l l not apply i n these cases as S w i l l no longer be a good quantum number. Because of the greater number of available couplings of the angular momenta for diatomic molecules (Herzberg, 1950), the selection rules for diatomics are more complex than for atoms. However, the s e l e c t i o n rule for changes i n the t o t a l angular momentum remains the same: AJ = 0, ±1 (J = 0 «-+-*• J = 0). Also, i f spin-orbit i n t e r -actions are small, then S i s defined and AS = 0 applies as f o r atoms. Again, the assumption of n e g l i g i b l e spin-orbit i n t e r a c t i o n holds less rigorously for n u c l e i of higher atomic number. These selection rules are upheld i n excitations caused by high energy electrons to the same degree as they are for photons. The most probable types of forbidden t r a n s i t i o n seen i n u l t r a v i o l e t absorption spectra are those involving changes of m u l t i p l i c i t y , and to a lesser extent, those allowed as magnetic dipole or e l e c t r i c quadrupole r a d i a t i o n , but forbidden as e l e c t r i c dipole r a d i a t i o n . These tr a n s i t i o n s appear only weakly, i f at a l l , i n the u l t r a v i o l e t spectrum, but t r a n s i t i o n s involving changes i n m u l t i p l i c i t y or changes of 2 i n the t o t a l quantum number J (allowed as e l e c t r i c quadrupole, where the selection rule i s AJ = 0, ±1, ±2) are often very prominent i n low energy electron impact e x c i t a t i o n spectra, especially at threshold (Lassettre et a l . , 1968). This i s because the p r o j e c t i l e electron carries with i t one unit of spin angular momentum, and a p r o j e c t i l e electron with spin up may be exchanged for one of the target electrons with spin down so that AS would be equal to one. O p t i c a l l y forbidden tra n s i t i o n s may be weakly observed with low energy electrons because the wavefunction of the incident electron i s perturbed by the valence electrons of the target, and vice versa. - 8 -2.2 Autoionization ' Direct i o n i z a t i o n i s generally regarded as a one step process, with the necessary energy being supplied through the c o l l i s i o n of an energetic electron (or an ion, neutral, etc.) or by a photon. Since energy i n excess of the minimum required (the I.P.) can be carried away as k i n e t i c energy of the scattered electron(s), a continuous range of energies above the minimum i s possible for both photoionization and electron impact i o n i z a t i o n . Autoionization i s a two step process. The f i r s t step i s the e x c i t a t i o n of a discrete state which has an e x c i t a t i o n energy greater than the f i r s t i o n i z a t i o n p o t e n t i a l . Then, after a short period of -13 -15 time ( t y p i c a l l y 10 to 10 seconds), a radiationless t r a n s i t i o n occurs from the highly excited state into the i o n i z a t i o n continuum, and an electron i s ejected with a k i n e t i c energy equal to the e x c i t a t i o n energy of the autoionizing state less the i o n i z a t i o n energy of the electron. The autoionizing state may arise as a result of the e x c i t a t i o n of an inner s h e l l or valence s h e l l electron to an u n f i l l e d outer o r b i t a l , or of two or more electrons to outer o r b i t a l s . In wave-mechanical terms, the autoionization occurs when the wave-function of the highly excited state becomes coupled with wave-functions of the neighboring i o n i z a t i o n continuum, and as a consequence the highly excited state assumes something of the character of the states of the continuum. Continuum states are unstable i n that one of the electrons i s moving i n an i n f i n i t e o r b i t , and thus the discrete l e v e l acquires the property of spontaneous i o n i z a t i o n . - 9 -Autoionizing states have a l i f e t i m e t which i s r e l a t e d to the width -14 T though the uncertainty p r i n c i p l e : "ft = rt. A l i f e t i m e of 10 seconds corresponds to a width of 0.1 eV; longer l i f e t i m e s are r e f l e c t e d i n narrower widths, and shorter l i f e t i m e s by broader structure. In photoionization, electron impact or ion impact spectra, autoioniza-t i o n may be observed as peaks, dips, or an asymmetric combination. Fano (1961) and Fano and Cooper (1965) have developed a theory based on configuration i n t e r a c t i o n which explains the observed shapes'and widths of a u t o i o n i z a t i o n resonances. The asymmetry ar i s e s because the r e l a t i v e contributions of the states of various configurations, which mix to form the stationary state of energy E, vary r a p i d l y as E passes through an aut o i o n i z i n g l e v e l . The e a r l i e s t experimental evidence of auto i o n i z a t i o n was found i n Auger's cloud chamber studies of the absorption of X-rays performed i n the e a r l y 1920's. In 1931, Shenstone and Majorana both demonstrated the importance of a u t o i o n i z a t i o n i n the vacuum UV region of atomic spectra, and since that time there have been numerous observations of the phenomenon. The rare gases have been p a r t i c u l a r l y w e l l investigated i n t h i s respect by u l t r a v i o l e t absorption as w e l l as by e l e c t r o n impact and ion impact spectroscopy, and the r e s u l t s of the most important of these studies from the standpoint of the present work w i l l be summarized below. Madden and Codling (1966), using the l i g h t from an e l e c t r o n synchrotron as a background source of r a d i a t i o n , observed autoionizing structure i n the rare gases He through Xe corresponding to both s i n g l e and double e l e c t r o n e x c i t a t i o n s . The r e s o l u t i o n of t h e i r .instrument - 10 -was 0.01 to 0.02 eV so that a great many o p t i c a l l y allowed levels were observed. For helium, four Rydberg series were c l a s s i f i e d i n the region above 60 eV which correspond to levels where both of the Is ground state electrons are excited. In neon, the e x c i t a t i o n of the 2s 2 6 electron to 3p, 4p, etc., o r b i t a l s leads to a series denoted: 2s 2p -> 6 6 4 2s2p np. Also observed were the two electron excitations 2p ->- 2p mlm'l' 2 6 5 and 2s 2p ->- 2s2p m l ' l ' . However, the i n t e r p r e t a t i o n of the neon spectrum i s hindered by the fact that the 2s single electron excitations occur i n the same region as the 2p double electron excitations. The s i t u a t i o n for Ar, Kr, and Xe i s s i m i l a r : the s electron e x c i t a t i o n 2 6 6 l e v e l s ns np -* nsnp mp overlap with the p subshell double electron 2 6 2 A e x c i t a t i o n levels ns np -»• ns np mlm'l'. Madden, Ederer, and Codling (1969) emphasize that t h e i r interpretation of the two electron e x c i t a t i o n structure i n argon i s tentative and somewhat fragmentary because of the complexity of the spectrum due to the large number of l i n e s and the lack of extensive t h e o r e t i c a l calculations. The spectra of Kr and Xe are even more complex as e x c i t a t i o n of inner d s h e l l electrons i s also i n evidence: (n-l)d ns np ->• (n-l)d ns np mp and (n-l)d ns np mlm'l'. Structure i n Xe i s observed which i s related to the promotion of an inner s h e l l p electron to outer s o r b i t a l s : (n-l)p^(n-1)d^ns^np^ r , v 5, , s,10 2 6 (n-l/p (n-l)d ns np ms. Samson (1966) reported vacuum u l t r a v i o l e t absorption spectra of Ne, Ar, Kr, and Xe i n the region up to the i o n i z a t i o n of the outermost f i l l e d s electron subshell, that i s , up to the edge of Ne, M^ edge of Ar, N^ edge of Kr, and 0^ edge of Xe. He was able to assign the 2 6 autoionizing levels observed to the s electron excitations ns np -»• nsnp^mp on the basis of some simple calculations. These calculations - 11 -are based on the s i m i l a r i t y between the electron configurations of the rare gases and the a l k a l i metals. The method i s i l l u s t r a t e d by considering the 5s -> np transitions i n Xe and the 6s -»• np transitions i n Cs. "In both cases the electrons excited into the vacant p she l l s see an atomic core of charge Z = 1 surrounded by a completed subshell of 6 p electrons. I t i s assumed that the excited electrons i n the np s h e l l s of Cs are bound to t h e i r core with the same binding energies as required for electrons i n the np s h e l l s of Xe. The binding energies for the np electrons i n Cs are r e a d i l y found from known spectroscopic term values" (Samson, 1966). Equating these binding energies to the np s h e l l i n Xe and subtracting from them the i o n i z a t i o n potentials of the s electron of Xe, the position of the 5s -*• np absorption series can be found. S i m i l a r l y , the predicted values of the 4s -> np series i n krypton can be obtained from the spectroscopic binding energies of the np electrons i n rubidium and the i o n i z a t i o n p o t e n t i a l of the 4s electron i n krypton, and so on for Ar and Ne. The agreement between the levels predicted i n t h i s way and the levels observed by Samson i s very good, (see Tables 1, 2, and 3). Several u n c l a s s i f i e d "window" type resonances were also observed (decreases i n the continuous absorption which forms the background) which may be due to double electron excitations. Bergmark et a l . (1969) studied the energy d i s t r i b u t i o n s of electrons emitted from autoionizing levels i n the rare gases after e x c i t a t i o n by high energy electrons. They were able to a t t a i n a t o t a l resolution of 0.013 eV (measured at 10 eV) using an incident electron beam energy of 3 to 5 keV. At such a high energy, i t might be expected that - 12 -Bergmark et al.'s autoionization spectra would be s i m i l a r to those observed by Samson or Madden and Codling since i t i s usually assumed that the selection rules for e x c i t a t i o n by electron impact become es s e n t i a l l y o p t i c a l above electron energies of a few hundred eV. Weak structure corresponding to o p t i c a l l y forbidden levels was observed, however. In He, the (2sns)^S, (sp,2n+)"S?, and (2pnp)"4) levels were i d e n t i f i e d . Transitions to (2pnp)"^D levels are Laporte forbidden and are not observed i n Madden and Codling's o p t i c a l spectra. For neon, the 6 1 o p t i c a l l y allowed tr a n s i t i o n s to (2s2p np) P le v e l s were observed, as were the forbidden (2s2p^np) 1S l e v e l s . For Ar, Kr, and Xe, trans i t i o n s 6 6 6 to nsnp mp ( o p t i c a l l y allowed) as wel l as nsnp ms and nsnp md levels (both o p t i c a l l y forbidden and thus not observed by Samson or Madden and Codling) were evident. For Ne, Kr, Ar, and Xe, the resolution was 2 s u f f i c i e n t to observe the s p l i t t i n g a r i s i n g from decay to a n c* 2 T?l/2 states of the ion. The spectra of Bergmark et a l . also exhibit much other structure which they do not attempt to explain. I n e l a s t i c resonances i n the forward scattering energy-loss spectra of He, Ne and Ar were observed by Simpson, Chamberlain and Mielczarek (1965) with a resolution of 0.1 eV. They were able to vary the r e l a t i v e i n t e n s i t i e s of the o p t i c a l l y allowed and forbidden tr a n s i t i o n s as a function of the incident electron energy. O p t i c a l l y forbidden t r a n s i t i o n s were s t i l l observed to be -making a contribution 2 1 2 1 at the highest energy used, 400 eV. In He, the states (2s ) S, (2p ) D, (2s2p)"4> and (sp, 23+) were observed between 57 and 64 eV. For Ne, 2 6 6 the most prominent structure was assigned to the series 2s 2p ->• 2s2p np, 6 2 A- 2 with possible evidence of 2s2p 3s and 2s 3p 3s states. In argon, the - 13 -6 6 o p t i c a l l y allowed levels' 3s3p 4p and 3s3p 5p appeared as dips i n the scattered current (and as decreases i n the o p t i c a l absorption by Madden et a l . ) - Also observed was a downward step assigned to 3s3p 5s and a peak believed to be 3s3p 3d. Numerous autoionizing t r a n s i t i o n s have also been observed i n He, Ne, and Ar by means of ion impact studies. Rudd and Lang (1965) measured the energy d i s t r i b u t i o n s of electrons emitted from these gases after c o l l i s i o n s with 45-75 keV beams of H + and I^ "*". Although a number of peaks appeared i n the autoionizing region of Ar, no attempt was made to assign the structure. However, large numbers of l i n e s were i d e n t i f i e d i n the neon autoionizing spectrum by Edwards and Rudd (1968) i n c o l l i s i o n s of Ne atoms with high energy H +, He + and Ne + ions. 13 Rydberg series were i d e n t i f i e d , including o p t i c a l l y forbidden states. Gerber, Morgenstern and Niehaus (1969) found that Ar autoionizing 6 2 4 levels 3s3p ns,np and 3s 3p 4sns i n evidence from the electrons emitted i n c o l l i s i o n s of A r + with Ar at energies of 0.2 to 3 keV. Besides the present work, threshold electron impact studies of the rare gases i n the autoionizing regions have been confined to trapped electron studies of He (Burrow and Schulz, 1969, and Grissom, Compton and Garrett, 1969) and Ne (Grissom, Garrett and Compton, 1969). In the 57 to 60 eV region of He, Burrow and Schulz i d e n t i f y four doubly excited states of helium as w e l l as two temporary negative ion resonances. They calibrated t h e i r energy scale by considering that two of the observed peaks were due to e x c i t a t i o n of autoionizing states of known energies. This places t h e i r i n t e r p r e t a t i o n i n some doubt. Grissom et a l . , who c a l i b r a t e t h e i r energy scales more d i r e c t l y and by two 14 -d i f f e r e n t methods, f i n d only one peak i n both He and Ne that corresponds to an autoionizing l e v e l . They suggest that a l l the structure i n the trapped electron spectra of He and Ne may be due to interference effects of negative 'ion resonances. 2.3 Temporary Negative Ions Resonances due to an electron being temporarily retained by an atom or molecule are observed, l i k e autoionization, as peaks and dips i n the electron scattering experiments at low energies. Like autoionizing states, these resonances have widths that are inversely related to t h e i r l i f e t i m e s , have l i n e p r o f i l e s that appear to follow Fano's configuration i n t e r a c t i o n theory, and may turn up i n close coupling calculations of such simple systems as the hydrogen and helium atoms. These resonances may be c l a s s i f i e d into two groups according to the way i n which the trapping of the p r o j e c t i l e electron may be r a t i o n a l i z e d . One way i s for the incident electron to excite the target and then f i n d i t s e l f with i n s u f f i c i e n t energy to escape the a t t r a c t i v e p o t e n t i a l generated by the excited target. This type i s variously known as a Feshbach, core-excited or closed-channel resonance. The s t a b i l i t y of such a negative ion w i l l n a t u r a l l y depend on the strength of the a t t r a c t i v e f i e l d , which w i l l be strongest when a number of degenerate or closely, spaced states are involved. Feshbach resonances generally occur at an energy s l i g h t l y less than that of the excited target state with which they are associated, with the energy differences accounted for as the "electron a f f i n i t y " of the temporarily held electron. These resonances are generally observed to have narrow widths and long l i f e t i m e s , although t h i s i s not always the case. - 15 -A second trapping mechanism i s that of a potential b a r r i e r which arises from a combination of a t t r a c t i v e e l e c t r o s t a t i c forces and repulsive centrifugal forces. Such resonances are known as shape, single-p a r t i c l e or open-channel resonances. These generally require the incoming electron to be bound i n an o r b i t a l of non-zero angular momentum because the centrifugal term i s given by ^ — • I n contrast to the r Feshbach resonance, the shape resonance usually l i e s higher i n energy than the associated target state, which may be the ground state or an excited state. Decay of the negative ion occurs when an electron tunnels through the pot e n t i a l b a r r i e r . The wider the b a r r i e r , the narrower the resonance w i l l be, and th i s also means that negative ion states l y i n g deeper i n the a t t r a c t i v e w e l l w i l l be narrow because the bar r i e r becomes wider. In general, shape resonances are wider and have shorter l i f e t i m e s than Feshbach resonances. Theoretical attempts to calculate the energies and widths of resonances have been of two types: scattering theory methods and quasistationary methods. Scattering theory approaches are more exact, but are much more d i f f i c u l t to compute.. For example, the close-coupling approximation of Burke and Schey (1962), which finds resonances as part of the t o t a l scattering cross section, i s mathematically i n t r a c t -able for molecules and for atoms containing more than two electrons. Quasistationary theories, of which there are many, calculate energies and widths of resonances from i n t u i t i v e l y selected t r i a l wave-functions without attempting to calculate scattering cross sections. Taylor, Nazaroff and Golebiewski (1966) discuss the various quasistationary techniques and show that they are a l l approximations which should be - 16 -considered as methods f6r r e f i n i n g good guesses. Negative ion resonances i n atoms have been reviewed by Burke (1966) and Smith (1966), and i n molecules by the extensive reviews of Bardsley and Mandl (1968) and of Chen (1969). Negative ions associated with singly excited electronic states have been observed i n He, Ne, Ar, Kr, and Xe, and negative ions associated with doubly excited states have been observed i n He and Ne. Kuyatt, Simpson and Mielczarek (1965) observed eleven transmission resonances i n He, s i x i n Ne, two i n Ar, two i n Kr, and f i v e i n Xe. They attempted to assign configurations to the observed structure on the basis of the predicted l i n e shapes of Fano (1961) and Fano and Cooper (1965). However, t h i s could only be done with certainty i n a 2 2 few cases. In the case of helium, the (ls2s ) ^^_/2 c o n ^ i g u r a t i o n was 2 assigned to the 19.31 eV resonance and (Is2s2p) P to the 19.43 eV peak. For Ne, the major structure appears as a pair of sharp resonant 2 2 decreases i n transmission assigned to the ^^/2 a n ^ ^1/2 s t a t e s °^ 2 5 2 the 2s 2p 3s negative ion at 16.04 and 16.135 eV, which are lower i n energy than the f i r s t two electronic states of neon at 16.62 and 16.715 eV. Four other resonances were observed, which Kuyatt et a l . believe are Ne resonances related to the 2p^3p states of neutral neon. For Ar, Kr, and Xe, only one pair of resonances i n each was 2 2 5 2 i d e n t i f i e d : the a n c* ^1/2 s t a t e s °^ t* i e (n-l)p ns configuration each about half an eV below the corresponding f i r s t excited states i n each gas. Other unidentified structure was observed at 11.235 eV i n Ar, and 9.02, 9.33, 10.71, and 10.81 eV i n Xe. Resonances associated with doubly excited l e v e l s of He have been observed i n the 57 to 60 eV region (Burrow and Schulz, 1969; Kuyatt, Simpson and Mielczarek, 1965; and Grissom, Garrett, and Compton, 1969). As mentioned e a r l i e r , the differences i n c a l i b r a t i o n of the energy scales make a great deal of difference to the i n t e r p r e t a t i o n of the r e s u l t s . Grissom, Garrett and Compton (1969) observe structure i n the Ne trapped electron spectrum i n the 40 to 50 eV region, but because of the complexity they could not make d e f i n i t e assignments to a l l of the structure. They suggest the v i s i b i l i t y of the autoionizing structure may be very low i n the trapped electron spectrum and that a l l of the v i s i b l e structure may be due to negative ion resonances. Resonances ascribed to temporary negative ion formation have also been observed for a large number of molecules, and many of these occur at energies below the f i r s t e lectronic state. For example, the N£ resonance between 1.8 and 3.5 eV has been observed i n electron swarm experiments, electron transmission experiments, energy loss spectra, and trapped electron and SF^ scavenging spectra. Negative ion resonances may be observed i n threshold electron e x c i t a t i o n spectra i f an excited state of the neutral matches the energy of the resonant state. For nitrogen, t h i s means a v i b r a t i o n a l l e v e l of N_ (^£ + ) l i e s 2 at the same energy as one of the ( TT) v i b r a t i o n s , so that the ejected electron has zero energy. A s i m i l a r structure has been observed i n the 1 to 3 eV region of carbon monoxide (Schulz, 1959, 1964; Rempt, 1969; Hubin-Franskin and C o l l i n , 1970). - 18 -CHAPTER THREE EXPERIMENTAL METHODS IN ELECTRON SPECTROSCOPY Three main classes of electron scattering experiments have evolved: t o t a l scattering or transmission-type experiments; energy-loss methods; and threshold methods. In a l l three methods, a monoenergetic electron beam i s directed into a target gas and the scattered (or unscattered) beam i s analyzed. In transmission type experiments the energy of the electron beam i s varied while the electrons l o s t from the beam show up as decreases i n transmission. A v a r i a t i o n of this i s to monitor the i n e l a s t i c a l l y scattered electrons at some angle with respect to the incident electron beam. In contrast, the energy of the primary beam i s fi x e d i n energy-loss experiments and the energy d i s t r i b u t i o n s of the i n e l a s t i c a l l y scattered electrons are studied at various angles. Most threshold experiments use an electron beam of variable energy and detect i n e l a s t i c a l l y scattered electrons at a single energy (zero or thermal), over a narrow range of energies (usually less than 0.01 eV) and over a l l angles (4TT) simultaneously. A l l three methods require an electron source that i s both intense enough that sample gas pressures can be low (for single c o l l i s i o n conditions to hold) and monoenergetic enough so that the f i n e r d e t a i l s of the spectra are not l o s t . Unfortunately the inherently monoenergetic - 19 -sources discovered so far are very low i n i n t e n s i t y , and the sources producing intense fluxes of electrons such as a hot surface (thermionic source) have very inhomogeneous energy d i s t r i b u t i o n s . Nearly, a l l electron scattering experiments at present employ a thermionic source i n conjunction with an electron monochromator, a f i l t e r which reduces both the i n t e n s i t y of the beam and the width of i t s energy d i s t r i b u t i o n . These monochromators employ e l e c t r i c and/or magnetic f i e l d s and may also be used to analyze the energy d i s t r i b u t i o n produced by another monochromator, or to determine the energies of the scattered electrons (energy-loss experiments). The t o t a l resolution obtained i n a scattering experiment w i l l depend on the resolution of the p o s t - c o l l i s i o n detector as w e l l as on the energy d i s t r i b u t i o n of the incident electron beam. The usefulness of the information obtained i n electron spectroscopy experiments w i l l often depend upon the resolution of the spectra and so the major experi-mental e f f o r t i s generally directed towards the design and operation of the electron monochromators. However, the t o t a l resolution i s generally twice as much as the width of the electron energy d i s t r i b u t i o n from the monochromator. Part of th i s decrease i n resolution i s due to Doppler broadening which arises because of the thermal motion of the sample gas atom or molecules. For l i g h t species such as He or H2, where the effect i s maximized, the broadening at room temperature i s less than 0.04 eV. The thermionic source i s usually a d i r e c t l y heated filament of tungsten, iridium or rhenium, or an i n d i r e c t l y heated oxide cathode. An e l e c t r i c a l l y heated filament generates electrons with a so-called - 20 -half-Maxwellian energy d i s t r i b u t i o n (Lindsay, 1960) a r i s i n g from the thermal motion of the electrons, i n addition to the spread i n energies caused by the voltage drop across the filament. An i n d i r e c t l y heated oxide cathode operates at somewhat lower temperatures and avoids the voltage drop problem and so narrower energy d i s t r i b u t i o n s r e s u l t . But the electron energy d i s t r i b u t i o n i s s t i l l some tenths of an eV wide, and the i n t e n s i t i e s obtained are lower than for a d i r e c t l y heated filament. 3.1 Electron Monochromators Two main types of instruments have been developed to improve the energy (velocity) homogeneity of an electron beam. These are the Retarding P o t e n t i a l Difference technique, which produces a "pseudo-monoenergetic" beam using retarding electrodes i n the path of the beam, and a number of deflection-type analyzers which use e l e c t r o s t a t i c or magnetic f i e l d s (or a combination) to deflect and s p a t i a l l y separate the electrons according to t h e i r i n i t i a l energies. 3.1.1 The Retarding Potential Difference Method In the R.P.D. method, electrons are beamed through a hole i n an electrode which i s held at a negative po t e n t i a l so as to create an energy d i s t r i b u t i o n with a low energy cut-off. Then the retarding potential i s changed by a small amount and the difference current produced i s due to a narrow s l i c e of the.energy d i s t r i b u t i o n between the two potentials. The electron beam i s usually confined and collimated by a magnetic f i e l d . This method was introduced by Fox et a l . (1955) and was subsequently used by many others, with energy d i s t r i b u t i o n s - 21 -of 0.1 to 0.2 eV (FWHM) at difference currents of 10 to 10 amps. Recently Golden and Zecca (1970) appear to have substantially refined the technique for they report an RPD gun capable of producing an energy d i s t r i b u t i o n as narrow as 0.008 eV, using small beam currents, no confining magnetic f i e l d , and a very sensitive detection system. One disadvantage of the RPD method i s that of r e l a t i v e l y high noise levels because the noise on the difference signal results from the t o t a l transmitted current rather than from the difference current i t s e l f . The method has also been c r i t i c i z e d as causing d i s t o r t i o n i n exc i t a t i o n or i o n i z a t i o n curve shapes, both because of space charge effects (Marmet, 1964) and because of the magnetic f i e l d used to collimate the electron beam (Anderson, Eggleton and Keesing, 1967). I t appears that these effects can be minimized with careful design and operating conditions. The most serious problem affecting the resolution, however, i s that the RPD method selects electrons according to t h e i r l i n e a r momentum only, and the t o t a l energy d i s t r i b u t i o n of the resultant beam i s broadened by the r o t a t i o n a l momentum of the electrons (Hall et a l . , 1970). 3.1.2 Deflection Analyzers Employing E l e c t r o s t a t i c Fields A simple e l e c t r o s t a t i c f i e l d can be used to deflect a collimated, inhomogeneous beam of electrons so that fast electrons are deflected least and the slow electrons most. An ex i t s l i t placed i n the dispersal region w i l l only allow through those electrons possessing a narrow range of v e l o c i t i e s . I f the beam i s not collimated, however (and thi s i s the usual s i t u a t i o n ) , electrons entering the f i e l d at the same - 22 -point at different angles and di f f e r e n t v e l o c i t i e s would be able to pass through the same ex i t s l i t . Thus a successful monochromator must have angle-focussing as wel l as velocity-dispersive properties. That i s , electrons entering the deflecting f i e l d at the same v e l o c i t y but at d i f f e r e n t angles with respect to an equipotential of the f i e l d , should be refocussed at the ex i t s l i t . Angle-focussing also substantially increases the resultant i n t e n s i t y of the velocity-selected beam. The three types of e l e c t r o s t a t i c deflection analyzers that have been developed a l l have focussing properties, although with varying effectiveness. Harrower (1955) developed a p a r a l l e l plate condenser, which produces a uniform e l e c t r i c f i e l d , to v e l o c i t y - s e l e c t a thermionic source. Electrons enter the e l e c t r i c f i e l d region between the plates at 45° to one of the plates and t r a v e l i n a parabolic path to be refocussed at a s l i t i n the same plate. This type of analyzer has limi t e d focussing capacity, however, as only those equal energy electrons entering i n a small range around 45° w i l l be passed. Foner and H a l l 0.961) were able to pass 3 x 10 ^ " amps at a half-width of 0.1 eV through such an analyzer. This type of analyzer remains one of the least popular types. Two other types of deflection analyzers which have been more extensively used i n recent years are the 127° c y l i n d r i c a l selector and the 180° hemispherical selector. These p a r t i c u l a r geometries are chosen because i t can be calculated that maximum focussing and dispersion properties can be expected over a 127° sector of an inverse f i r s t power f i e l d such as generated between two concentric cylinders. - 23 -The same i s true for an inverse second power f i e l d such as i s generated between two concentric spheres, where focussing occurs at 180°. The 180° hemispherical selector i s superior i n performance to both the p a r a l l e l plate analyzer and the 127° c y l i n d r i c a l analyzer. This i s because the hemispherical analyzer focusses over two dimensions rather than only one as for the other analyzers. This means less space charge effects and thus better resolution for a given i n t e n s i t y . An energy-loss spectrometer employing hemispherical selectors was developed by Simpson (1964). The electron beam i s collimated before entering the selector to reduce r e f l e c t i o n and space-charge problems, and the use of v i r t u a l s l i t s prevents electrons from r e f l e c t i n g back into the analyzer. This instrument i s capable of a resolution of -14 -7 0.005 eV at 10 amp or 0.1 eV at 10 amp. Lassettre (1969) has also used a si m i l a r type of analyzer i n an energy-loss spectrometer with good r e s u l t s . The 127° c y l i n d r i c a l analyzer w i l l be considered i n greater d e t a i l i n Section 4.1, as i t i s t h i s type of monochromator which i s —8 employed i n this study. A beam current of 10 amps at 0.04 eV FWHM was.possible with the analyzer used although the t o t a l resolution of the instrument for scattered electrons was never better than about 0.15 eV (see Section 4.6). 3.1.3 Deflection Analyzers Employing Magnetic Fields The magnetic analyzer b u i l t by Ramsauer (1921) to study the t o t a l c o l l i s i o n cross section of the rare gases was perhaps the f i r s t use of an electron monochromator. The active p r i n c i p l e here i s that - 24 -an electron moving with a v e l o c i t y component that i s perpendicular to a magnetic f i e l d , w i l l move i n a c i r c l e with a radius proportional to the perpendicular v e l o c i t y component. The greater the i n i t i a l v e l o c i t y , the larger the c i r c l e , and so dispersion i s accomplished. Golden and Bandel (1965) b u i l t a more sophisticated version of Ramsauer's analyzer and were able to obtain a resolution of 0.10 eV FWHM at an electron energy of 2 eV. Magnetic analyzers suffer the major drawback that the resolution i s proportional to the electron energy, so the performance i s much less s a t i s f a c t o r y at higher electron energies. 3.1.4 Deflection Analyzers Employing Crossed E l e c t r o s t a t i c and Magnetic Fields Electrons ejected into a crossed e l e c t r i c and magnetic f i e l d exhibit c y c l o i d a l motion, and undergo a net deflection i n the d i r e c t i o n perpendicular to both the e l e c t r i c and magnetic f i e l d s . This dispersive effect has been employed i n the trochoidal monochromator developed by Stamatovich and Schulz (1968). A half-width of 0.02 eV at a beam current of 10 amps i s claimed for t h i s monochromator. Crossed e l e c t r i c and magnetic f i e l d s are also used i n the Wien f i l t e r employed by Boersch (1965) and Boersch et a l . (1962) at high resolution and high electron energies. 3.2 Electron Scattering Experiments 3.2.1 Transmission Experiments In transmission, or t o t a l scattering experiments a monoenergetic electron beam i s passed through a gas and the transmitted electron - 25 -current i s measured as a' function of electron energy. In the Ramsauer (1921) method the apparatus i s designed so that electrons which are e l a s t i c a l l y scattered through more than a small angle, or those that undergo i n e l a s t i c scattering at any angle, are l o s t to the beam and are not detected. Kuyatt, Simpson and Mielczarek (1965) used this method to study He, Ne, Ar, Kr and Hg. I t i s found that the observed anomalies i n the transmitted current can be attributed to i n e l a s t i c scattering r e s u l t i n g from the ex c i t a t i o n of atomic energy l e v e l s , and to e l a s t i c resonances which are interpreted as temporary negative ion formation. The structure corresponding to i n e l a s t i c scattering i s j u s t barely v i s i b l e , while the e l a s t i c resonances are much more i n evidence, although s t i l l a small part of the t o t a l current. A v a r i a t i o n of the transmission experiment i s to measure the e l a s t i c a l l y scattered electrons at some angle to the incident beam. Schulz (1963), measuring the e l a s t i c current i n He at 72°, was the f i r s t to observe the He negative ion resonance at 19.3 eV, approximately half an eV below the f i r s t excited state. A si m i l a r experiment with Ne also showed a resonance at 16 eV which i s 0.5 eV below the f i r s t excited state of Ne. In a subsequent t o t a l scattering experiment these resonances i n He and Ne were again observed i n addition to s i m i l a r structure below the f i r s t excited states of Kr and Xe. These types of experiments appear to be p a r t i c u l a r l y w e l l suited to the observation of temporary negative ion formation. Because the resonances are a small part of the t o t a l current, however, good signal to noise c h a r a c t e r i s t i c s are required. Also, the resonances may be very narrow (estimates range to widths of 0.001 eV and less) - 26 -for long-lived states, arid very wide for states of short l i f e t i m e s , so good resolution i s ess e n t i a l . 3.2.2 Energy-Loss Experiments In t h i s method a monochromatic beam of fixed energy i s passed into a gas and the i n e l a s t i c a l l y scattered electrons are energy analyzed at some angle with respect to the incident beam. The resultant energy-loss spectra are sensitive to both the energy of the incident beam and the angle at which the scattered electrons are detected. Energy-loss spectra have proven to be a r i c h source of information on discrete l e v e l s , including o p t i c a l l y forbidden and autoionizing l e v e l s , as w e l l as temporary negative ion resonances. Two groups which have used the technique extensively are those working with Lassettre and with Simpson. One of the advantages of the energy-loss type of experiment i s that observing the v a r i a t i o n i n r e l a t i v e i n t e n s i t y with d i f f e r e n t incident electron energies may be an aid i n id e n t i f y i n g a t r a n s i t i o n . This was done i n the forward scattering energy-loss studies of He, Ne and Ar by Simpson, Chamberlain and Mielczarek (1965). Rice, Kuppermann and Trajmar (1968) were able to observe s i g n i f i c a n t variations i n the r e l a t i v e i n t e n s i t i e s of spin and symmetry forbidden tra n s i t i o n s i n He as a function of electron energy and scattering angle. Recently, McPherson Instruments has introduced a commercial version of Simpson and Kuyatt's (1967) hemispherical analyzer energy-loss spectrometer. I t i s claimed (Rendina and Grojean, 1970) that the - 27 -instrument i s capable of 0.010 eV resolution. The introduction of a commercial high resolution electron impact spectrometer undoubtedly means t h i s type of spectroscopy w i l l soon become another p r a c t i c a l t o o l of the chemical analyst. 3.2.3 Threshold Excitation Methods In these types of experiments a monoenergetic beam i s passed into the sample and only those electrons that lose nearly a l l t h e i r momentum in i n e l a s t i c c o l l i s i o n s are detected. These thermal energy electrons res u l t when the energy of the beam electrons i s marginally greater than the e x c i t a t i o n energy of a l e v e l i n the sample gas. The threshold electron e xcitation spectra which i s produced by monitoring the slow electron current as a function of the electron beam energy i s expected to be especially r i c h i n structure r e s u l t i n g from o p t i c a l l y forbidden t r a n s i t i o n s . Two methods of detecting zero energy electrons are used. The f i r s t method developed was the trapped electron method, by Schulz (1958). Zero energy electrons are retained by a small e l e c t r o s t a t i c p o t ential w e l l i n the c o l l i s i o n chamber and are collected, while more energetic electrons are able to escape. The c o l l i s i o n chamber i s actually a gr i d of p a r a l l e l wires, and a p o s i t i v e l y biased outer cylinder produces the pot e n t i a l w e l l by penetration through the gri d . Well depths of 0.1 eV are generally used, although smaller w e l l depths may be desirable for improved resolution. Schulz's o r i g i n a l spectro-meter was not capable of very high resolution (0.3 eV), but the technique has been used by H a l l et a l . (1970) with a resolution of - 28 -0.10 eV and this represents about the best resolution obtained so f a r i n a threshold spectrum. An RPD gun i s v i r t u a l l y always employed as the monochromator for the trapped electron technique. Negative ions of low energy may reach the c o l l e c t o r and contaminate the spectra, as the method does not dist i n g u i s h between negative ions and electrons. The second threshold method, and the one that i s used i n t h i s study, i s the electron scavenger method, which u t i l i z e s a gas that has a sharply resonant negative ion capture cross section near zero energy as the detector of zero energy electrons. The f i r s t demonstration of t h i s was by Curran (1963) who mixed SF^ and N^ i n the c o l l i s i o n chamber of a mass spectrometer, and by monitoring the SF^ ion current as a function of the incident electron energy, was able to produce a threshold excitation spectrum of N£ s i m i l a r to that from the trapped electron method. The scheme i s as follows: e (E = E*) + X > e (E = 0) + X* +SF, where E i s the 6 exc i t a t i o n energy Sulfur hexafluoride i s i d e a l l y suited as an electron scavenger because -15 2 of i t s very high electron capture cross section (10 cm ) and the fact that i t captures electrons only over a very narrow range of energies. Stamatovic and Schulz (1968) found the width of the resonance to be less than 0.020 eV. As the scattered zero energy electrons are captured by SF^ over a l l angles with respect to the incident electron beam, the scavenging method, l i k e the trapped electron method, i s a t o t a l c o l l e c t i o n technique. - 29 -Curran's scavenging- was done at a high ion source pressure (0.2 torr) using an RPD gun. Jacobs and Henglein (1964) also used the scavenging method for a number of atoms and small molecules, but used an electron source without benefit of a monochromator, so t h e i r curves reveal only the most general features. The technique was refined by Compton et a l . (1968, 1969) using an RPD gun i n conjunction with a time of f l i g h t mass spectrometer, attaining a t o t a l resolution of 0.2 to 0.3 eV. The SFg scavenging spectra revealed temporary negative ion resonances i n benzene and benzene derivatives (Compton et a l . , 1966) and i n pyridine (Huebner et a l . , 1968). Scavenging spectra were also reported for He, HC1, ^ 0 , benzene, n i t r o -benzene and naphthalene (Compton et a l . , 1968): NH^ and ND^ (Compton et a l . , 1969); XeF^ and XeFfi (Begun and Compton, 1966); and HBr, HI and 1-chloronaphthalene (Huebner et a l . , 1968). L i f s h i t z and Grajower (1970), using a MAT CH4 mass spectrometer and an RPD gun, looked for temporary negative ions i n carbon t e t r a -f l u o r i d e and tetrafluoroethylene without success, but such states were detected i n perfluoropropylene and perfluoropropane. Stamatovich and Schulz (1969) used the i r trochoidal monochromator and the scavenging technique to study the e x c i t a t i o n of v i b r a t i o n a l modes of the ground electronic states of CO2 and ^ 0 at high resolution. Their instrument has no provision for mass se l e c t i o n , however, so that the r e s u l t i n g scavenging curve may be contaminated with SF,. , F and other negative ions. Three other scavenging studies have been reported recently, which resemble the e a r l i e r experiment of Jacobs and Henglein (1964) i n - 30 -that no electron monochromator i s used to reduce the thermionic energy spread. Hubin-Franskin and C o l l i n (1970) reported SF^ scavenging spectra of N^, CO, benzene and ethylene with a resolution of about 0.5 eV. In an e f f o r t to improve t h i s resolution, smoothing and deconvolution techniques were applied to the data. However, th i s procedure i s not s u f f i c i e n t to produce any hint of v i b r a t i o n a l structure. O'Malley and Jennings (1969) and Ridge and Beauchamp (1969) have used thermionic electron beams i n ion cyclotron mass spectrometers to produce some poorly resolved scavenging spectra. Ridge and Beauchamp employed CI /CCl^ rather than SFg as the zero energy electron detector. In t h i s laboratory scavenging experiments have been carried out with an instrument incorporating a radio frequency mass spectrometer and a 127° electron v e l o c i t y selector as a monochromator. This instrument had been previously used to study i o n i z a t i o n e f f i c i e n c y curves of both pos i t i v e and negative ions (Brion and Thomas, 1968). Preliminary scavenging results for helium (Brion and Eaton, 1968) and f i v e other substances (Thomas, 1969) were quite promising, although there remained room for considerable improvement i n both resolution and i n t e n s i t y . I n i t i a l scavenging studies of xenon (Brion, Eaton, Olsen, and Thomas, 1969) revealed structure below the I.P., due to transitions of the xenon 5p electrons to 6s, 6p, 6d, 7s, etc. o r b i t a l s , as w e l l as structure of low v i s i b i l i t y at electron energies above the I.P. Subsequently, i t was decided that a comprehensive investigation of the rare gases should be undertaken. I t was hoped that the structure i n the io n i z a t i o n continuum of xenon could be examined i n d e t a i l and i d e n t i f i e d , and a search was planned for analogous structure i n a l l of the rare gases. A second objective was to determine - 31 -the r e l a t i v e contributions from the various types of t r a n s i t i o n s both above and below the I.P., i n the hope of providing information concerning the " s e l e c t i o n r u l e s " for low energy electron e x c i t a t i o n . The r e s u l t s of these i n v e s t i g a t i o n s , which have now been published (Brion and Olsen, 1969, 1970b), are given i n Chapter 5. The scavenging curve showing the 3 v i b r a t i o n a l l e v e l s of the a II i e v e l of carbon monoxide (figure 18) has also been published (Brion.and Olsen, 1970a). - 32 -CHAPTER FOUR EXPERIMENTAL 4.1 The Instrument Figure 1 i s a schematic diagram of the instrument. The 127° electron v e l o c i t y selector l a b e l l e d "Selector" i s the monochromator which reduces the energy spread of the electron beam before i t enters the c o l l i s i o n chamber. A second 127° selector l a b e l l e d "Analyzer" i s used to determine the energy d i s t r i b u t i o n of the "monoenergetic" electron beam crossing the c o l l i s i o n chamber. These energy d i s t r i b u t i o n measurements are carried out by varying the potential of the Analyzer with respect to the Selector. Negative ions which d r i f t out of the c o l l i s i o n chamber are accelerated (by approximately 35 vol t s ) and focussed i n the ion gun before entering the monopole mass f i l t e r . An outer box surrounds the c o l l i s i o n chamber, shielding i t from stray electrons. The outer box i s held at a constant p o s i t i v e p o t e n t i a l with respect to the c o l l i s i o n chamber i n order to provide a constant fr i n g i n g f i e l d at the entrance and exit' s l i t s of the c o l l i s i o n chamber while the electron energy i s varied. The electron energy i s given by the accelerating p o t e n t i a l between a filament centre tap and the c o l l i s i o n chamber, plus a contact p o t e n t i a l . The contact po t e n t i a l depends upon the p a r t i c u l a r gas which i s introduced and upon ANALYSER U+ Vcoswt SCHEMATIC OF ELECTRON IMPACT SPECTROMETER Figure 1. The Instrument. - 34 -how much gas i s introduced. This i s because the contact p o t e n t i a l varies according to the amount and type of gas molecules which are adsorbed on, or dissolved i n , the electrode surfaces. Changes i n the contact potential w i l l s h i f t the electron energy scale, and because of t h i s , p a r t i c u l a r care must be taken that the gas pressures do not change during the course of running a spectrum. The filament i s heated by a d.c. power supply and i s located i n a soft i r o n housing (to minimize the magnetic f i e l d due to the filament current). Rhenium and tungsten were used as filament materials both i n ribbon and wire form, and a l l of these types performed about equally w e l l . A l l parts i n the Selector, Analyzer, c o l l i s i o n chamber, ion lens and monopole are constructed of gold plated brass, with the exception of the filament housing and the entrance s l i t to the selector (soft i r o n ) , the wire grids of the Selector and Analyzer (gold plated tungsten) and of course, the insulators (boron n i t r i d e , t e f l o n , or nylon). Other d e t a i l s concerning the construction, dimensions and operation of the instrument may be found i n the paper by Brion and Thomas (1968). The electronics and a n c i l l a r y equipment have been described by Thomas (1969), including a c i r c u i t diagram of the controls for the Selector Analyzer and c o l l i s i o n chamber. 4.1.1 The 127° .Electron Velocity Selector ' The p r i n c i p l e of this electron v e l o c i t y selector (or f i l t e r ) was f i r s t proposed by Hughes and Rojansky (1929) and was experimentally v e r i f i e d soon after by Hughes and McMillen (1929). They showed that charged p a r t i c l e s entering an inverse e l e c t r o s t a t i c f i e l d (such as i s - 35 -generated between concentric cylinders) would be refocussed according to t h e i r i n i t i a l energies, after being deflected through a segment of the f i e l d equal to — (127° 17'). The f i r s t working monochromator /2 of t h i s type was b u i l t by Clark (1954) who used two 127° coaxial stainless s t e e l electrodes and entry and e x i t s l i t s to define the electron beam. The "best" resolution obtained was about 0.5 eV. Marmet and Kerwin (1960) pointed out that low energy electrons are more readily reflected than absorbed, and that electrons deflected from the 127° segments would increase space charge effects and could also reduce the resolution by being d i r e c t l y reflected through the e x i t s l i t . They b u i l t a selector which used highly transparent tungsten grids rather than s o l i d 127° segments. P o s i t i v e l y biased "catcher" electrodes beyond the grids were used to remove the electrons which pass through the grids. A resolution of 0.05 eV was achieved with this design. Recently Salop, Golden and Nakano (1969) have reported a non-gridded selector that i s capable of energy resolutions of the same order as a Marmet-Kerwin selector, although of lower i n t e n s i t y . This i s accomplished by using lenses to focus and l i m i t the electron beam before i t enters the monochromator. With appropriate lens potentials a v i r t u a l image of the lens entrance s l i t i s produced at some distance from the entrance plane of the selector. In t h i s way the maximum' divergence of the electrons entering the selectbr i s held to a small angle and the deflection problem i s largely overcome. The Selector and Analyzer depicted i n Figure 1 are b a s i c a l l y of the Marmet-Kerwin type. The focussing grids are constructed by winding gold-plated tungsten wire on pre-grooved frames. The r e s u l t i n g double g r i d helps to prevent f i e l d penetration from the catcher - 36 -electrodes. I t was found necessary to wrap the frames for the Selector as loosely as possible, since the heat generated by the filament causes the brass frames to expand to a much greater degree than the tungsten wires. This causes the frames to warp, which d i s t o r t s the f i e l d and often produces shorts. An attempt was made to compensate for possible stray magnetic f i e l d s by i n s t a l l i n g s i x f i e l d c o i l s outside the brass vacuum housing. However, only a s l i g h t improvement i n resolution and i n t e n s i t y was achieved by optimum adjustment of the currents through the c o i l s . 4.1.2 The Monopole The monopole mass analyzer was f i r s t developed by Von Zahn (1963) and i s a v a r i a t i o n of the quadrupole analyzer of Paul and Raether (1955). As shown i n Figure 1, the monopole consists of a c i r c u l a r rod and a 90° V-shaped electrode held p a r a l l e l to and insulated from each other. The V-shaped electrode i s grounded while a combined r . f . (Vcostot) and d.c. (U) p o t e n t i a l i s applied to the c i r c u l a r rod. This creates a p o t e n t i a l i n the space between the electrodes which i s given by: 2 2 * = ^ ~2 ^ ( - U + Vcoswt) (1) r o where r i s the distance between the rod and the apex of the V-o electrode, and the y d i r e c t i o n i s up and the x d i r e c t i o n i s horizontal looking at the electrodes from the end view. To transmit p o s i t i v e ions, -U i s changed to +U. The equation of motion of the ions i n both - 37 -the x and y directions is'given by the Mathieu equation: ,2 ^-y + [a + 2qcos2(T - C )] u = 0 (2) dT U where u i s either the x or y coordinate, T = 1/2 ut, to i s the angular r . f . frequency, t i s the time, 8eU a = -a = — r — T — x y 2 2 mo) r o and 4eV q x = - q y = "TIT mai r o The equations of motion have two types of solutions, depending on the values of a and q. In one type, the amplitude becomes very large a f t e r a short time; these are the unstable solutions. In the other, the motion remains l i m i t e d i n amplitude so these o s c i l l a t i o n s are stable. Since a and q are mass dependent, only ions of a certain mass w i l l be able to pass through the monopole for given values of the r . f . and d.c. f i e l d s . A l l others w i l l s t r i k e the rod or the V-electrode and be l o s t . A mass spectrum of the residual gas i n the instrument, as produced by the monopole, i s shown i n Figure 2. The unit resolution obtained i s more than s u f f i c i e n t for the purposes of this study, namely to separate SF^ from SF,. . I f substances possessing a negative ion i n the 140 to 150 mass range were to be studied, interference problems R E S I D U A L M A S S S P E C T R U M - 39 -could be expected; however, this was not the case for any of the species of t h i s study. There are three main advantages i n using a radio frequency mass analyzer i n this type of study. F i r s t , the comparative s i m p l i c i t y Cwhere high resolution i s not required) means construction and operation are r e l a t i v e l y simple. Second, the fact that e l e c t r i c rather than -magnetic f i e l d s are used for the mass analysis means there are no complications from magnetic f i e l d s penetrating into the c o l l i s i o n chamber or i n t e r f e r i n g with the operation of the selector. And t h i r d ; since high i n i t i a l ion energies are not required for th i s type of mass analyzer, the ion gun can be operated at close to ground p o t e n t i a l . This reduces the effect of e l e c t r i c f i e l d s penetrating into the c o l l i s i o n chamber. One drawback of the monopole, however, i s that 50 percent of the ions entering at the wrong phase of the r . f . cycle are driven into the V-electrode and lo s t to the beam (Dawson and Whetten, 1968a). This i s undesirable for t h i s type of experiment, since the ion current i s already limited by the necessity to keep the sample gas pressures low enough so that few electrons w i l l undergo more than one c o l l i s i o n . Monopoles s t i l l transmit ions more e f f i c i e n t l y than do magnetic mass spectrometers, however. For a more detailed discussion of the monopole see Dawson and Whetten (1968b). Thomas (1969) has i l l u s t r a t e d the c i r c u i t diagram of the o s c i l l a t o r power supply for t h i s instrument. - 40 -4.1.3 The Detection'and Data Retrieval Systems Ions which pass through the monopole are detected at a 16 stage venetian-blind copper-beryllium electron m u l t i p l i e r which provides a 4 6 gain of approximately 10 to 10 , depending on i t s condition. The signal i s then further amplified by a Cary Model 31 vi b r a t i n g reed electrometer. In order to detect negative ions the entire detection system i s floated 1500 volts p o s i t i v e with respect to ground. The output from the vib r a t i n g reed electrometer i s then reduced to ground pot e n t i a l by an insulated d.c. to d.c. converter (see Figure 3). This signal may then be fed to a s t r i p chart recorder, but i n many cases the noise l e v e l i s s t i l l f a i r l y high, so signal averaging techniques are used to improve the signal to noise l e v e l . This was usually done with a 400 channel Nuclear Chicago Multichannel Analyzer, but occasionally a Fabri Tek Model 1064 unit of 1024 channels was used. Generally between f i v e and twenty-five cumulative scans were made (each scan was eight minutes i n duration), t h i s number being l i m i t e d by the pre v a i l i n g pressure (and thus, contact potential) s t a b i l i t y . In the early part of th i s work, the st a r t of the multichannel analyzer scan was manually synchronized with the electron energy scan, but l a t e r an electronic device was i n s t a l l e d to automate the st a r t i n g and accumulation procedure. A paper-tape output from the multichannel analyzer was available so that the data could be transfered to the university's main computer f a c i l i t y . However, a l l spectra i n this thesis are direct plots from the multichannel analyzer to an X-Y recorder. ELECTRON IMPACT SPECTROMETER OUTPUT AND DATA RETRIEVAL SYSTEM ELECTRON MONOCHROMATOR AND ION SOURCE r MASS FILTER ELECTRON MULTIPLIER EMI 9603 CARY M00EL 31 VIBRATING REED ELECTROMETER DC TO DC CONVERTER CHART RECORDER 1-1 PLOTTER I VIDAR 241 VOLTAGE TO FREQUENCY CONVERTER OSCILLOSCOPE COMPUTER TAPE PUNCH MULTICHANNEL ANALYSER NUCLEAR CHICAGO 34-27 TYPEWRITER I PLOTTER F i g u r e 3 - 42 -4.2 The SFg Scavenging/Method The SF, and sample gas mixture are introduced into the c o l l i s i o n o chamber through a dual i n l e t system with separate leak valves, permitting fine control over the r e l a t i v e and t o t a l pressures. Generally an SF^/sample r a t i o of 1/10 i s used, with the t o t a l pressure -4 of the order of 5 x 10 torr as measured by an i o n i z a t i o n gauge attached to the Selector region of the vacuum housing. Because the entrance and ex i t s l i t s for the electron beam and the ex i t aperture for the ions i n the c o l l i s i o n chamber are f a i r l y large, i t i s not expected that the actual pressure i n the c o l l i s i o n chamber would be more than an order of magnitude greater than that indicated by the io n i z a t i o n gauge. Figure 4 shows the SF^ ion current which i s observed between 6 and 14 eV when SF^ alone i s introduced into the c o l l i s i o n chamber, at a pressure of 5 x 10 ^  t o r r . The absence of any features other than the general increase i n current indicates that no interference effects from the electronic l e v e l s of SF^ i t s e l f need be expected i n the scavenging spectra, at least at t h i s pressure. At higher pressures, -4 approximately 3 x 10 torr and above, some structure does begin to -4 appear. The background current of Figure 4 i s only about 2 x 10 times as intense as the SF^ primary peak shown i n the inset. The general increase of the background ion current i s probably due to the scattering of electrons from the metal surfaces i n the c o l l i s i o n chamber, especially at the s l i t s . The observed half-width of the SF^ primary peak was never less than 0.120 eV i n t h i s study. This width i s presumably instrumental i n o r i g i n , rather than the inherent width of the resonance. Compton - 44 -et a l . (1966) showed the' cross section for electron capture by SF,, u (mainly to form SF^ ) to be inversely proportional to the electron v e l o c i t y . Stamatovich and Schulz (1968), using t h e i r high resolution trochoidal monochromator, found the half-width to be less than or equal to 0.020 eV. This result has been questioned by H a l l (1971) who claims that differences i n r e l a t i v e i n t e n s i t i e s of the peaks i n the He spectrum, as determined by the scavenging and trapped electron methods, would indicate an SF,. width of 0.1 eV or more. The results o of Section 5.1 do not support Hall's arguments i n th i s regard however. I f the 0.020 half-width reported by Stamatovich and Schulz i s v a l i d , then the SFg half-width of 0.120 observed i n t h i s study i s c l e a r l y incompatible with the electron beam half-width of 0.04 eV as determined by the Analyzer. This discrepancy could be explained i n at least two ways. One p o s s i b i l i t y i s that the electron energy d i s t r i b u t i o n measured by the Analyzer i s not the actual energy d i s t r i b u t i o n of the entire electron beam entering the c o l l i s i o n chamber. That i s , i f the electrons entering at a few degrees to the perpendicular are of diffe r e n t energy than those entering perpendicular to the wal l of the c o l l i s i o n chamber, then t h i s would not be detected by the Analyzer, but would be reflected i n the broadness of the SF^ peak. McGowan and Clark (1968) have also considered the p o s s i b i l i t y of t h i s type of behaviour from a 127° selector. The second possible explanation i s that the -assumption that the electron attachment to SF^ i s taking place i n a f i e l d - f r e e region i s not j u s t i f i e d . Small f i e l d s i n the c o l l i s i o n chamber, res u l t i n g from non-uniform surface conditions or from f i e l d penetration - 45 -(especially from the ion'gun), could retard electrons which o r i g i n a l l y had too much energy to be scavenged. A l t e r n a t i v e l y , zero energy electrons could be accelerated by the f i e l d to energies where electron attachment would not occur. This type of effect would be p a r t i c u l a r l y d i f f i c u l t to eliminate. F i e l d penetration effects had been considered i n the o r i g i n a l design. One of the c h a r a c t e r i s t i c differences of scavenging spectra from trapped electron spectra i s the persistent " r i s i n g background" which i s present i n a l l scavenging spectra. This background i s much higher than that of SF^ i t s e l f (Figure 4) and cannot be eliminated by sub-tr a c t i n g the SFg background from the scavenging spectrum. The most l i k e l y explanation for t h i s i s the increasing number of scattered electrons with increasing electron energy (because of a greater number of possible scattering channels) which are subsequently reduced to zero energy i n c o l l i s i o n s with the walls. This could presumably be reduced by coating the c o l l i s i o n chamber with a non-reflective material. Such coatings(benzene soot, Aquadag) were found to be unsuitable, however, perhaps because of charging and the formation of non-uniform surface potentials. Another possible cause of the r i s i n g background and large SF^ D width i s that excited neutral species are being formed i n the ion gun or monopole regions and are reaching the electron m u l t i p l i e r . This p o s s i b i l i t y was p a r t i a l l y tested by mounting the whole Selector-Analyzer and ion gun assembly at angles of 10 and 15 degrees to the longitudinal axis of the monopole. Any energetic neutral species should then be directed into the V-block while the negative ions are - 46 -brought back on course by the monopole f i e l d . This configuration did not decrease the ion beam i n t e n s i t y ; i n f a c t , a s l i g h t increase was noted. But the r i s i n g background and SF^ width were unchanged from before, r u l i n g out the ion gun (but not the monopole) as a source of such energetic species. Yet another p o s s i b i l i t y i s that, because of the nature of the ion gun potentials necessary to accelerate and focus the ion beam, SF^ ions are being formed i n the ion gun or i n the monopole regions. The electrons taking part i n t h i s ion formation would come from the i n e l a s t i c a l l y and e l a s t i c a l l y scattered electrons and negative ions from the c o l l i s i o n chamber. These spurious ions would have a f i n i t e p r o b a b i l i t y of being transmitted to the m u l t i p l i e r since the monopole i s e s s e n t i a l l y a mass rather than momentum analyzer. This could be a contributing factor to the r i s i n g background or the SF^ width. I t should be possible to eliminate this effect by using a momentum analyzer such as a magnetic or time of f l i g h t mass spectrometer. In fa c t , the scavenging curves of Hubin-Franskin (1970) (using a magnetic f i e l d ) and those of Compton et a l . (1968, 1969) (time of f l i g h t ) exhibit backgrounds that are not as steep as those seen i n th i s study. This supports the hypothesis of the r i s i n g background being due to the formation of excited neutrals or ions i n the monopole, but i t cannot be taken as conclusive evidence because the difference i n background might w e l l be due to greater scattering from the walls and s l i t s i n the instrument of t h i s study. For an ion energy of 35 v o l t s i t w i l l take an SF^ ion approximately f i v e microseconds to t r a v e l from the ion gun through the monopole to - 47 -the electron m u l t i p l i e r . / Assuming an autodetachment l i f e t i m e of 25 microseconds for the SF, ion (Compton et a l . , 1966), t h i s leaves only D 20 microseconds for these ions to d r i f t out of the c o l l i s i o n chamber and into the ion gun after they are formed. No drawout f i e l d s can be applied as t h i s would preclude the p o s s i b i l i t y of zero energy electrons. The fact that f a i r l y large SF^ ion currents are observed despite the short l i f e t i m e of SF^ may be because the zero energy electron which autodetaches from one ion can be "rescavenged" by another SF, molecule. 6 Nevertheless, i t i s probably safe to assume that a scavenger with an equivalent capture cross section and a longer l i f e t i m e would be a more sensitive detector. With t h i s i n mind, three other gases were investigated for t h e i r possible usefulness as electron scavengers. CH^I, which gives I at zero energy, has been used by Jacobs and Henglein (1964) to produced scavenging curves for SO2 and CS2. ^ n the present study, t h i s species was found to be an unsuitable electron scavenger for three reasons. The I /CH^I peak i s s i g n i f i c a n t l y wider than the SF^ peak, so that scavenging spectra with CH^I are less well, resolved. In addition, I /CH^I i s not as intense as SF^ at equal gas pressures, and i t has a much higher ion current at energies above 1 eV than does SF ~. 6 The Cl /CCl^ process, which also occurs at close to zero energy, has been used as an electron scavenger i n an ion- cyclotron mass spectro-meter by Ridge and Beauchamp (1969). In the monopole instrument, the Cl /CCl^ peak was found to be as narrow as the SF^ peak, but only about one-tenth as intense for the same pressure. There i s also considerable structure i n the Cl /CCl^ i o n i z a t i o n e f f i c i e n c y curve at higher electron energies. - 48 -Tetracyanoethylene''(TCE) was also t r i e d as an electron scavenger. The parent and fragment negative ions observed i n this substance as a function of energy are shown i n Figure 5. The parent molecular ion TCE (a) has a greater i n t e n s i t y for a given pressure than does SF^ , making i t a good prospect i n spite of a high energy shoulder. However, the vapor pressure of TCE (a so l i d ) i s quite low so that even though the leak and i n l e t system was by-passed i n order to introduce the vapor d i r e c t l y , only a small pressure of TCE could be introduced into the —6 c o l l i s i o n chamber (1 x 10 t o r r ) . Because of t h i s , signal to noise r a t i o s are less favorable than for SF,. scavenging. Figure 6 shows o a TCE scavenging curve for krypton, representing a t o t a l accumulation time of 200 minutes i n the multichannel analyzer. The r i s e at 14 eV corresponds to the i o n i z a t i o n p o t e n t i a l of krypton. A portion of the curve with the y-axis expanded 100 times (inset) shows a small peak at 2 6 2 5 10 eV corresponding to the 4s 4p -»- 4s 5p 5s e x c i t a t i o n i n krypton (compare with Figure 13), but no other structure i s v i s i b l e . The TCE scavenging curve of Xe exhibits no structure below the i o n i z a t i o n p o t e n t i a l . Hence TCE i s also far i n f e r i o r to SF^ as an electron scavenger. Substances which would form a sharply resonant negative ion at non-zero energy could presumably be used to investigate e x c i t a t i o n functions above threshold. However, the only such negative ion which has yet been p a r t i a l l y successful for this purpose i s SF,. /SFg> which peaks at about 0.5 eV. This resonance i s so much wider than SF^ that i t i s not r e a l l y very useful either. (a)CC(CN) 4 J I L 1.0 ZD \ (b)CC(CN) 3 x l O O 10 15 (c) CC (CN) 2 x 3 0 o * . * 10 15 (d)CC(CN)" x 5 0 0 .>; O o < (e) CN •: * x IOO 4>-IO ELECTRON ENERGY (eV) Figure 5. The Negative Ions of Tetracyanoethylene. Figure 6. The Tetracyanoethylene Scavenging Curve of Krypton. (0 x l O O # « • • • • * • • < » * < w ' . A ' J «/ * 0 0 * K r / O 9 I O 11 1 2 1 3 ELECTRON ENERGY (eV) 14 - 51 -No e f f o r t s were made to purify the sample gases used. Manufacturer's stated p u r i t i e s were generally 99.8% or better. In the case of every substance studied, samples taken from at least two dif f e r e n t lecture bottles produced no noticeable differences i n the scavenging spectra. 4.3 Energy Scale C a l i b r a t i o n Electron energies were measured with a d i g i t a l voltmeter. The voltmeter used i n the case of Ne, Ar, Kr, and Xe had a resolution of 0.02 v o l t s , but for CO and the He spectrum of Figure 7, a much superior meter (accurate to 0.001 v o l t ) was used. Ca l i b r a t i o n of the energy scales could be carried out (as for CO) by introducing helium as w e l l as SF^ and sample and referencing the spectrum to the w e l l -known spectroscopic l e v e l s of helium. However, i n the case of the rare gases, the energy scales can be calibrated by the unambiguous structure due to the exc i t a t i o n of the outer p electrons. Energy scales established i n th i s fashion should be accurate to 0.1 eV or better'. I t i s found that c a l i b r a t i o n of energy scales using the zero energy SF^ peak i s inaccurate. This i s because of f i e l d penetrations which a l t e r the pot e n t i a l along the path of the electron beam i n the c o l l i s i o n chamber. This effect has been discussed by Schulz (1960). Compton et a l . (1968) used the SF^ peak to establ i s h the energy scale of the scavenging spectrum of nitrogen, r e s u l t i n g i n an incorrect assignment of the observed structure (but the use of CI /HC1 appears to give s a t i s f a c t o r y r e s u l t s ) . Similar problems can be expected for any zero energy calibrant. - 52 -CHAPTER FIVE EXPERIMENTAL RESULTS AND DISCUSSION 5.1 Helium The threshold e x c i t a t i o n spectrum of helium i s shown i n Figure 3 7. The energy scale i s calibrated by setting the 2 S peak to 19.82 eV. The positions of the atomic energy levels (Moore, 1949) are indicated by the v e r t i c a l l i n e s of the diagram. I t can be seen that the o p t i c a l l y forbidden transitions to states (symmetry forbidden: 3 3 AL = 0 and J = 0 <—>- J = 0) , P states (spin forbidden) , and S states (symmetry and spin forbidden) are w e l l represented, while the o p t i c a l l y allowed tr a n s i t i o n s are making very l i t t l e contribution. The t r a n s i t i o n to the 2"h? l e v e l i s only just v i s i b l e . Figure 8 shows another helium scavenging spectrum over approximately the same energy range but at decreased resolution. The r e l a t i v e i n t e n s i t i e s of the peaks of Figure 8 are quite di f f e r e n t from those 3 1 3 of Figure 7, especially for the 2 S, 2 S and 2 S l e v e l s . The r e l a t i v e i n t e n s i t i e s are independent of pressure and are quite reproducible under conditions of i d e n t i c a l resolution, however. 3 Comparison of Figures 7 and 8 shows that at lower resolution the 2 S ex-3 c i t a t i o n i s r e l a t i v e l y favored while the 3 S i s not. This trend i s supported by the low resolution scavenging spectrum of Compton et a l . He/SF, 23S 2'S 33S 3'S r 4 3S 4'S 33P ' f n = 5 / A./ • o3p 21p - . •V.4« A/ 9 20 21 22 23 24 ELECTRON ENERGY (eV) Figure 7. The Threshold Electron Impact Excitation Spectrum of Helium (High Resolution). SF6~ ion current (arbitrary units) r o l O • •. r o CO m ro o «—.• -1 o ro ZD - i t o <-< r o r o r o r o r o •«S . r o r o (JT 04 LO II -rj ll o N a r-t-O V r o - +?g -- 55 -(1968) where the 2 S/2 S" r a t i o i s 1.5, by the medium resolution trapped electron spectrum of Brongersma et a l . (1969) where this r a t i o i s 1.2, and by the high resolution trapped electron spectrum of H a l l (1971) where the r a t i o i s 0.93. H a l l also found that the height of 3 1 the 2 S peak r e l a t i v e to the 2 S was increased by increasing the w e l l -depth of the trapped electron spectrometer. And, as here, he found 3 that the 3 S peak becomes r e l a t i v e l y higher with increasing resolution (decreasing well-depth). In the case of the trapped electron spectra, t h i s change of r e l a t i v e i n t e n s i t i e s can be explained as a difference i n the slopes or a non-linearity i n the ex c i t a t i o n cross sections near threshold. This type of behaviour i s to be expected, as the work of Chamberlain (1968) and Erhardt et a l . (1967, 1968) shows a resonance i n the 2^ S cross 3 section just above threshold, while the 2 S cross section i s more lin e a r i n t h i s region. As the well-depth i s increased, the trapped electron spectrum represents the exc i t a t i o n cross section integrated over a greater range of energies above threshold. Variations i n the half-width of the electron energy d i s t r i b u t i o n could not account for such a change i n the r e l a t i v e i n t e n s i t i e s (unless one peak was inherently sharper than another, i n which case the r e l a t i v e heights, but not the peak areas would vary). If i t i s assumed that the resolution i n scavenging experiments i s limi t e d only by the half-width of the electron beam and by the inherent half-width of the SF^ peak, then no such f l u c t u a t i o n i s expected since the width of the SF^ peak represents a constant "well-depth". I t was with t h i s assumption that H a l l questioned the 0.020 eV - 56 -result of Snamatovic and Schulz (1968) for the half-width of the SF^ peak, since the scavenging spectra of Compton et a l . (1968) and Brion and Olsen (1970b) (gee Figure 8) have r e l a t i v e i n t e n s i t i e s that correspond to those observed for the lower resolution trapped electron work. But as the spectra of Figures 7 and 8 show, this assumption of constant well-depth i n scavenging experiments i s not j u s t i f i e d . In f a c t , the spectrum of Figure 8 i s quite s i m i l a r to Hall's i n r e l a t i v e i n t e n s i t y . This suggests that the major contribution to the SFg half-width i s coming from a p o s t - c o l l i s i o n effect (such as the f i e l d penetration, non-uniform surface p o t e n t i a l s , or spurious ion formation, as discussed i n Section 4.2) rather than from the electron beam or the inherent SF, width. D The only He threshold e x c i t a t i o n study that disagrees with the i n t e n s i t y - r e s o l u t i o n behaviour outlined above i s the low resolution (by present standards) trapped electron spectrum of Schulz (1958). 3 1 This spectrum shows 2 S and 2 S peak heights which are approximately equal.' This could be because of further changes i n slope of the e x c i t a t i o n cross sections at a greater energy above threshold or i t could be an instrumental e f f e c t . The He threshold studies mentioned above (with the exception of the work of Chamberlain and Erhardt et al.) detect the scattered zero-energy electrons over a l l angles with respect to the incident electron beam. If the angular d i s t r i b u t i o n s of the scattered electrons were different for the various excitations at threshold, and i f an experiment did not c o l l e c t over a l l angles with equal e f f i c i e n c y , then the r e l a t i v e i n t e n s i t i e s would again be affected. The 36 eV energy loss studies of Kuppermann et a l . (1968) - 57 -1 3 3 show that the r e l a t i v e i n t e n s i t i e s of the 2 S, 2 P and 2 S peaks are indeed dependent on the scattering angle at this energy and that forbidden transitions are i n general more highly represented at larger scattering angles. Thus any anisotropy i n the method of detection could also be expected to affect the results at threshold, and t h i s should be kept i n mind when comparing the results of d i f f e r e n t threshold experiments. And, of course, at higher incident electron energies, as i n the 50 eV forward scattering energy-loss spectra of Lassettre et a l . (1968), the o p t i c a l l y allowed t r a n s i t i o n s begin to dominate the spectrum while the forbidden transitions become much less prominent. Reference to Figures 7 and 8 shows that the scavenging current begins to r i s e steeply at approximately 0.4 eV before the i o n i z a t i o n p o t e n t i a l . This i s probably at least p a r t i a l l y due to the large density of states (n = 6,7,8... °°) i n t h i s region. I t may also be due to reactions of the type: He* + SF 6 > He + + SY~ (1) Quantum mechanical treatments of the hydrogen atom (Flu'gge, 1957) have 4. 5 shown that the l i f e t i m e of Rydberg states increases as n so that a p r i n c i p a l quantum number (n) of 10 corresponds' to a l i f e t i m e of 10 ~* seconds. Cermak and Herman (1964) demonstrated experimentally that highly excited states of atoms (with e x c i t a t i o n energies close to the i o n i z a t i o n potential) can have long l i f e t i m e s . Hotop and Niehaus (1967) found that such reactions (1) do occur i n c o l l i s i o n s of highly excited He, Ne, and Ar atoms with SF^ molecules. The cross sections were -12 2 determined to be of the order of 10 cm . I t i s not known, however, how much of a contribution t h i s type of reaction i s making to scavenging spectra either below or above the i o n i z a t i o n p o t e n t i a l . Above the I.P. the primary contribution to the SFg ion current presumably would be coming from the electrons ejected i n the direct i o n i z a t i o n process: He + e y H £ + 2e (2) Brion and Olsen (1970b) have discussed the shape of the scavenging curve above the I.P. with respect to the threshold laws for the production of zero energy electrons i n t h i s process. However, because of the unknown effect of reactions of the type (1) and because of the uncertain influence of a possible space-charge effect due to build-up of p o s i t i v e ions i n the c o l l i s i o n chamber, i t i s d i f f i c u l t to draw any quantatitive conclusions from the shape of the curve. An unsuccessful search was made for possible structure due to autoionization or temporary negative ion processes i n the i o n i z a t i o n continuum. Structure has been observed i n the trapped electron spectrum i n the 56 to 60 eV region by Grissom et a l . (1969) and by Burrow and Schulz (1969). The search was hampered by the fact that -4 i t was not possible to introduce more than 2 x 10 t o r r of helium into the instrument, for above this value the pressure became w i l d l y unstable. This was possibly due to a leak valve or pumping problem. Two d i f f e r e n t leak valves made no difference, however. This phenomenon - 59 -recurred repeatedly for'helium, but not for any other gas. 5.2 Neon 2 2 6 1 The neon ground state configuration i s Is 2s 2p , S. Although Russell-Saunders coupling properly describes the ground state, the excited levels are more appropriately designated according to the ( j , j ) coupling scheme. The positions of the energy levels from Moore (1949) are shown i n the neon threshold e x c i t a t i o n spectrum i n Figure 9. These l e v e l s arise from the e x c i t a t i o n of a 2p electron to 3s, 3p, 3d, 4s, etc. o r b i t a l s . For example, four closely spaced levels may be expected for excited states involving 3s o r b i t a l s : ( i n order of increasing energy) J = 2 and J = 1 states of the 2p^(^P°^)3s configura-t i o n , and J = 0 and J = 1 states of the 2p"* (^P°^) 3s configuration. 2 o 2 o Note that the two series of l i n e s are based on ^T,/! a n c* P l / 2 s t a t e s of the ion core and that the l a t t e r are denoted by a prime ('). The l i n e spacing for the 3s,3s' levels i s s t i l l c h a r a c t e r i s t i c of Russell-Saunders coupling while for the more highly excited states the s p l i t t i n g becomes more l i k e that expected for ( j , j ) coupling. The most prominent feature i n the Ne scavenging spectrum i s the peak due to transitions to the o p t i c a l l y allowed 3s,3s' states. The i n d i v i d u a l contributions are not resolved. Also represented are the excitations to 3p,3p' l e v e l s , which are p a r i t y forbidden. The J = 0 l e v e l of the 3p' at 18.97 eV i s resolved. This corresponds to a J = 0 J = 0 t r a n s i t i o n . The contributions from the higher states appear smaller and are not w e l l resolved. The steep r i s e i n ion current begins approximately 0.5 eV before the i o n i z a t i o n p o t ential which i s at C0 I C O UL 0) Ne/SR 3s 3s' IONIZATION 3 p , 3 p ' 4 s 4 p 5s 6s l\ • /\ A A . * .' II v*« 3 d 14 16 18 20 22 o ELECTRON ENERGY (eV) Figure 9. The Threshold Electron Impact Spectrum for the Excitation of the 2p Electron i n Neon. - 61 -21.7 eV. This i s s i m i l a r to the s i t u a t i o n for helium, and i t i s again possible that highly excited neon atoms are c o l l i d i n g with SF^ to give SF, ions i n t h i s region, b This scavenging spectrum i s si m i l a r to the trapped electron spectrum of neon obtained by Grissom, Compton and Carrett (1970) and i s of about equal resolution. The structure from n = 4 and n = 5 states i s s l i g h t l y more d i s t i n c t i n the trapped electron curve. Also, the scavenging curve has more of a r i s i n g background at the higher electron energies, and the I.P. i s marked by a much greater increase than i s the case for the trapped electron spectrum. The sharp increase i n ion current 0.5 eV before the I.P. i n the scavenging spectrum has no counterpart i n the trapped electron spectrum. This supports the suggestion of contributions from excited atom-induced i o n i z a t i o n of SFg as i n reaction (1). No other threshold spectra of Ne are available for comparison with the scavenging r e s u l t s . Although much e f f o r t was devoted to a search for f i n e structure i n the i o n i z a t i o n continuum of Ne, none could be found except for a f a i n t hint of a peak at 43.7 eV. This could be the 2s2p^3s l e v e l which has been observed by Bergmark et a l . (1969) to be at 43.65 eV. Unfortunately the scavenging signals for neon are quite weak, both above and below the I.P. As mentioned i n Chapter 2, Grissom, Compton and Garrett (1970) have reported fine structure i n the trapped electron spectrum i n the 40 to 50 eV region. 5.3 Argon The threshold e x c i t a t i o n of the 3p electron i n argon (ground 2 2 6 2 6 1 state configuration: Is 2s 2p 3s 3p ,. S) i s shown i n Figure 10. Again, the v e r t i c a l l i n e s show the energies of the excited states from Moore (1949). This spectrum i s s i m i l a r to that observed for neon i n that the o p t i c a l l y allowed 4s and the o p t i c a l l y forbidden 4p excitations are both w e l l represented, with a certain amount of contribution from 3d, 5s, 5p and higher l e v e l s . Note from the increased separation of the 4s and 4s' l e v e l s , and the 4p and 4p' l e v e l s , that the s p l i t t i n g i s becoming more ch a r a c t e r i s t i c of ( j , j ) coupling. Again, as for He and Ne, the SF^ ion current increases rapidly i n the region of the highly excited Rydberg l e v e l s . The trapped electron spectrum of Grissom et a l . (1970) for Ar does not show t h i s behaviour. The trapped electron spectrum i s not s u f f i c i e n t l y resolved to show the s p l i t t i n g i n the 4s,4s' l e v e l s . I t w i l l be useful to compare the threshold e x c i t a t i o n curve of Figure 10 with the results obtained i n electron impact e x c i t a t i o n studies at higher incident electron energies. In Figure 11, the scavenging spectrum i s drawn to the same energy scale as the 40.7 eV forward scattering energy-loss spectrum published by Chamberlain et a l . (1965) and the e s s e n t i a l l y o p t i c a l energy-loss spectrum of Boersch et a l . (1962) at 50 keV. The o p t i c a l l y allowed 4s tr a n s i t i o n s show up i n a l l three spectra, although for (a) and (b) i t i s only the two J = 1 levels which appear. High resolution zero-angle energy-loss spectra at 50 eV by Lassettre et a l . (1968) also found that only the J = 1 levels were excited. However, i t i s probable that e x c i t a t i o n of the (0 > < O O I CD Ar/SF 4 s 4s ' A A IONIZATION V V 4 d 4d» 3 d 3 d ' ON Figure 10. ll 12 13 14 15 ELECTRON ENERGY (eV) The Threshold Electron Impact Spectrum f o r the E x c i t a t i o n of the 3p Electron i n Argon. - 64 -00 H Z D > CC < cc H QQ CC < l-z UJ cc cc D o z o (a) ENERGY LOSS E = 5 0 keV (b) ENERGY LOSS E = 40.7 eV 12 13 14 15 ELECTRON ENERGY (eV) Figure 11. Comparison of Argon Ex c i t a t i o n Spectra at Three Different Electron Energies. - 65 -o p t i c a l l y forbidden, meta'stable J = 2 (4s) and J = 0 (4s') states are also contributing to the threshold spectrum (c) , since the e x c i t a t i o n of these states would be more e a s i l y detected at threshold energies and large scattering angles. Some of the p a r i t y forbidden p states are observed i n the 40.7 energy loss spectrum (b) but these are completely absent i n the high energy spectrum (a). Note that the lowest energy 4p l e v e l , the J = 1 l e v e l at 12.91 eV, i s observed i n the threshold spectrum but i s absent i n (b). This l e v e l i s also absent from the zero-angle 50 eV energy loss spectrum of Lassettre et a l . (1968), where the 4p,4p' excitations are seen to exhibit strong angular scattering dependencies. Lassettre et a l . assigned the t r a n s i t i o n s to the f i r s t three levels as e l e c t r i c dipole forbidden, but e l e c t r i c quadrupole allowed, while the highest energy l e v e l has the same term symbol as the ground state. Transitions of the l a t t e r type are also disallowed o p t i c a l l y , but are allowed i n electron impact. Lassettre et a l . observed quite d i f f e r e n t angular dependencies for the scattering from these two kinds of forbidden t r a n s i t i o n s . This would suggest that scattering r e s u l t i n g from transitions to metastable states occurs mostly at large scattering angles. In the region just above 14 eV, the 3d and 5s levels (both o p t i c a l l y allowed) overlap to some extent and the resolution i s not s u f f i c i e n t to determine the i n d i v i d u a l contributions. Both Lassettre et a l . (1968), at 60 eV, and Boersch (1965) at 25 keV, were able to achieve high enough resolution to resolve the s and d excitations i n this region. Both the threshold spectrum and the 40.7 eV energy loss spectrum show a small contribution from 5p states, while t h i s i s absent i n the high energy spectrum (a). - 66 -The search for fine'structure i n the i o n i z a t i o n continuum was more successful for argon than for helium or neon. Figure 12 i l l u s t r a t e s the scavenging spectrum i n the 24 to 35 eV region. The zero l e v e l i s greatly suppressed because the fine structure represents less than 1% of the t o t a l ion current i n t h i s region. The height (and to some extent, the shape) of the underlying continuum i s sensitive to the pressures, however, so the exact r a t i o w i l l vary. The region from 25 eV to the edge i s where the excited states of the 3s electron are found. Table 1 summarizes the observations of Samson ( u l t r a v i o l e t absorption) and Bergmark et a l . (electron spectro-scopy studies of the electron impact induced autoionization). Samson only observed the o p t i c a l l y allowed 3s np t r a n s i t i o n s , of course. I t can be seen that the results of his simple calculations (Section 2.2) agree quite w e l l with the observed l e v e l s . Similar calculations were carried out to predict the 3s ->• ns and 3s -*• nd t r a n s i t i o n energies, and these are also shown i n Table 1. Where available the results of Bergmark et a l . seem to agree with these values. The positions of the np states observed by Samson and the calculated ns and nd values are indicated i n Figure 12. The value for the 4s l e v e l , which cannot be calculated by t h i s method, i s taken from Bergmark et a l . (25.22 eV). The parity-forbidden 4s and 5s l e v e l s , and the o p t i c a l l y allowed 4p and 5p levels appear to correlate with peaks i n the scavenging curve. For.states of higher n, the density of states Is too great and the resolution too low to expect to resolve any structure. I t i s i n t e r e s t i n g that the edge (io n i z a t i o n of the 3s electron) i s marked by a small increase i n SF, ion current, m o 52 • CD n=4 M, E D G E I I I II MM 3 s 2 3 p 6->s - 3 s 3 p 6 n s 5 6 r I III V//////, 3 s 2 3 p 6 - s ~ 3 s 3 p 6 n p n = 4 5 6 | I Ml VM 3 s 2 3 p 6 - * - 3 s 3 p 6 n d n=3 4 5 r V V . * " Ar/SF P n = 4 n=4\6,75 V < M M [nil 3 s 2 3 p 6 — > - 3 s 2 5 p 4 3 d nf 4 I 3 s 2 3 p 6 — s - 3 s 2 3 p 4 4 s nf n=4 5 6 n=4 LL 4 5 5 6 5 8910 3 s 2 3 p 6 — i 3 ~ 3 s 2 3 p 4 3 d np , 3 s 2 3 p 6 — ^ 3 s 2 3 p 4 4 s np 2 4 2 6 28 3 0 32 3 4 • ELECTRON ENERGY (eV} Figure 12. The Threshold Electron Impact Spectrum of Argon i n the Region of the 3s Electron Excitation. 68 -TABLE 1 Energy Levels (in eV) for E x c i t a t i o n of the 3s Electron i n Argon 3s -»• ns 3s -> np 3s -> nd n calc u b obs calc obs° i b obs . c a l c3 , b obs 3 - - - - 27.57 27.51 4 25.22 26.51 26.63 26.63 28.30 28.27 5 27.51 27.58 27.96 28.00 28.00 28.64 28.62 6 28.30 - 28.49 28.51 28.51 28.83 28.84 7 28.65 - 28.75 28.76 28.75 28.94 28.94 8 28.84 - 28.89 28.90 28.89 29.01 29.01 9 28.95 _ 28.98 28.99 28.97 29.06 — Edge 29.24 calculated from Samson's (1966) method k from Bergmark et a l . (1969) c observed i n o p t i c a l absorption by Samson (1966) d from Moore (1949) - 69 -while the f i r s t i o n i z a t i o n p o t e n t i a l (the 3p electron) i s enormous by comparison. Madden et a l . (1969) have reported doubly excited states of 2 A argon of the type 3s 3p n l n ' l ' , beginning i n the region of the M^  edge. The positions of these states are also shown i n Figure 12; however these should be regarded as only approximate, as Madden et a l . point out that the exact positions are uncertain due to the shape and structure of the observed resonances. I t i s possible that the states 2 6 corresponding to the 3s 3p 4s4p configuration at 29.03 and 29.21 eV are contributing to the r i s e i n the curve i n the region of the M^ edge. There also appears to be some cor r e l a t i o n between the observed structure and some of the lower members of the series observed by Madden et a l . However, the density of states i s quite high i n this region. Three factors which complicate the inte r p r e t a t i o n of th i s kind of spectrum are: (a) the positions of the o p t i c a l l y forbidden double electron excitations are not known. These w i l l not be observed i n the' work of Madden et a l . (b) i t i s not known whether to expect these excitations as peaks or dips i n the threshold spectrum. The 3s -*• np excitations are observed by Samson as windows (decreases) i n the o p t i c a l absorption, and as asymmetric combinations of increases and decreases by Madden and Codling (1963). Bergmark et a l . see the auto-i o n i z a t i o n of the same np states mainly as dips i n the electron emission, while the ns and nd le v e l s show up as peaks. (c) the influence of temporary negative ion formation i s uncertain. For example, the large dip at 30 eV could be due to such a state. For that matter, much of the structure might be better explained as interference of temporary negative ion resonances, as i s claimed for the trapped electron results - 70 -for He and Ne (Grissom et a l . , 1969). This would have to await more complete information on the doubly excited autoionizing l e v e l s , however. A fourth p o s s i b i l i t y , that of double c o l l i s i o n energy-losses, i s ruled out. This type of behaviour could be expected at high pressures, where a beam electron would lose a l l i t s k i n e t i c energy i n two successive i n e l a s t i c c o l l i s i o n s . For example, the peaks at 25.2 and 26.6 eV could be explained as 4s (4s'), 4p (4p') and 4p (4p'),4p (4p') double c o l l i s i o n s , respectively, at low resolution. A peak (or series of peaks, at a resolution comparable to that of Figure 10) would then also be expected i n the 23.2 to 23.6 eV region from 4s (4s'), 4s(4s') double c o l l i s i o n s ; none i s observed however. In p r i n c i p l e , this could be further checked by measuring the pressure dependencies of the i n t e n s i t i e s of these peaks. Single c o l l i s i o n peaks would vary l i n e a r l y with pressure while double c o l l i s i o n peaks would be expected to vary with the square of the pressure. In practice t h i s i s d i f f i c u l t to do because of the low v i s i b i l i t y of the structure (Figure 12 represents approximately 5 hours of signal accummulation i n the multichannel analyzer) and because the -magnitude of the underlying continuum i s also sensitive to r e l a t i v e and t o t a l pressures. The structure of Figure 12 i s e n t i r e l y reproducible at d i f f e r e n t pressures, although the o v e r a l l shape of the underlying continuum i s not. Unfortunately, no other threshold results -are available with which to compare t h i s spectrum. Grissom et a l . (1970) did observe some small structure i n t h i s region of the Ar trapped electron spectrum, but did not attempt an analysis. - 71 -5.4 Krypton . ' Figure 13 shows the threshold spectrum representing the excitation of the 4p electron of krypton. As with neon and argon, 2 2 there are two series of l i n e s a r i s i n g from ^^/2 a n <^ ^1/2 s t a t e s °^ the ion core but the s p l i t t i n g i s greater for krypton. The contribution from the 5s levels (J = 2,1] i s much greater than from the 5s' CJ = 0,1) l e v e l s , but the resolution i s not great enough to resolve the c o n t r i -butions from the d i f f e r e n t J states. Again, the J = 1 levels are o p t i c a l l y accessible while the J = 2 (5s) and J = 0 (5s') are o p t i c a l l y inaccessible and w i l l be metastable. The high r e l a t i v e i n t e n s i t y of the 5s e x c i t a t i o n i n krypton i s quite unlike the s i t u a t i o n for neon, argon and xenon, where the ns and ns' are much more nearly equal i n i n t e n s i t y . The p a r i t y forbidden np levels appear to be w e l l represented again here, but the contributions from nd states i s less certain. As before, an abrupt r i s e takes place about 0.5 eV below the I.P. The spectrum of krypton i n the region of the edge i s shown i n Figure 14. The positions of the np levels are from Samson (1966) while the ns and nd energies are calculated from the i o n i z a t i o n energy of the 4s electron i n Kr and the spectroscopic values of the binding energies of electrons i n excited s and d states of rubidium. Table 2 shows these values along with the results of Bergmark et a l . whose 23.73 eV value for the 5s l e v e l i s marked i n Figure 14. The corre l a t i o n of the observed peaks with ns and np states i s quite good i n t h i s spectrum. The edge i s not marked by any large feature. The shoulder preceding the 5s l e v e l may be due to a t r i p l e t state associated with th i s l e v e l (J = 0). The nd states, p a r t i c u l a r l y the n = 4 l e v e l , do P >-OC < CC < UJ CC CC 3 O O L> 0) 5s IONIZATION Kr/SR 5s' II : 5P 5p' A 6s I I I 6p6s* 7p 6p': K \ 7 s A 8 ^ ; * ll 4d 5d 6d I 4d ' 10 II 12 13 14 ELECTRON ENERGY (eV) •-J Figure 13. The Threshold Electron Impact Spectrum for the E x c i t a t i o n of the 4p Electron i n Krypton. CO > < H cm < H Si o o I (0 0) Kr/SF 6 • . ••• v n=5 4s 2 4p6-6 78 oo 4s 4p°ns I.l I': I .1 I ill N,EDGE i i •% ••••• 2£ t— •• n = 5 4s 2 4p6-6 7 0 0 4s4p6np •V n=4 5,6 oo 4 s 2 4 p 6 — • 4s4p6nd 22 24 26 28 30 ELECTRON ENERGY (eV) 32 Figure 14. The Threshold Electron Impact Spectrum of Krypton i n the Region of the 4s Electron Excitation. - 74 -TABLE 2 Energy Levels ( i n eV) for Exc i t a t i o n of the 4s Electron i n Krypton n 4s i a calc -*• ns u b obs calc 4s ->• np obs° . b obs 4s -> -t a calc nd . b obs 4 - - - - - 25.73 25.59 5 - 23.73 24.90 24.93 24.74 24.92 24.95 25.00 24.94 25.01 26.52 26.47 6 25.83 25.89 26.28 26.29 26.31 26.89 26.87 7 26.59 26.61 26.79 26.80 26.79 27.07 27.06 8 26.93 26.95 27.04 27.08 27.02 27.20 27.19 9 27.11 27.12 27.17 27.18 27.19 27.28 27.27 10 27.22 27.23 27.26 27.27 27.25 27.33 -11 27.29 27.27 27.32 27.33 27.32 27.36 -N l Edge 25. ,71d a calculated from Samson's (1966) method b from Bergmark et a l . (1969) c observed i n o p t i c a l absorption by Samson (1966) d from Moore (1952) -75 -not appear to be making much of a contribution, but i t i s d i f f i c u l t to say exactly because of overlap with ns and np states. Madden et a l . (1964) observed double electron excitations above the edge, but the positions of the o p t i c a l l y forbidden double electron excitations are not known. The possible influence of temporary negative ion states i s also uncertain. The large dip at 30 eV i s s i m i l a r to structure ' found i n argon and might be due to a negative ion resonance associated with a doubly excited state. This structure does not correspond to what would be expected for double c o l l i s i o n energy-losses. No structure of any degree of v i s i b i l i t y appears i n the regions immediately above or below the region shown i n Figure 14. 5.5 Xenon The threshold electron impact e x c i t a t i o n spectrum of xenon i n the region below the I.P. i s shown i n Figure 15 along with the spectroscopic term values. The general features of t h i s spectrum are s i m i l a r to those of the analogous regions i n neon, argon and krypton, but with a larger spin-orbit s p l i t t i n g of the ion core. For the 6s l e v e l s , the J = 1 state appears at 8.44 eV while the J = 2 state appears as a shoulder at 8.31 eV. As for the 6s' e x c i t a t i o n s , the lower energy J = 0 state i s only very weakly observed and the J = 1 overlaps with the f i r s t 6p state. Although the lower 5d members are overlapped with the highest energy 6p l e v e l , the higher 5d states appear to be making some contribution. The threshold spectrum of xenon i n the region of the 5s electron (0 O Xe/SR 6s' .^ v. 6s I I / V M / X IONIZATION 6 P ' 8p 9s 9p 7 p o 6p linn 8s 7s —v-v—' 5d 6d 7d i<o 8 9 IO II 12 (0 ELECTRON ENERGY (eV) Figure 15. The Threshold Electron Impact Spectrum for the Exc i t a t i o n of the 5p Electron in Xenon. ON r /> = 6 5s 2 5p 6 -^>5s5p 6 .os C A ! j l l I I 'I ! I i i 3 5 8 5 8 9 10 oo Oj edge /> = 6 7 5s 2 5p 6 —*-5s5p 6 />p 8 .9 II co f I I IH VTJ7TJ n=S 6 7 8 5 3 2 5 p 6 — * 5 s 5 p 6 / 7 d CO 18 19 22 23 Figure 16. 20 21 Electron energy CeV) The Threshold Electron Impact Spectrum of Xenon i n the Region of the 5s Electron Excitation. - 78 -TABLE 3 Energy Levels (in eV) for E x c i t a t i o n of the 5s Electron i n Xenon 5s + n c a l c 3 ns obs , a calc 5s -> np obs° i b obs 5s ->• calc nd obs 5 - - - - 21.30 21.44 6 20.08 20.89 20.96 20.67 20.81 20.95 21.03 20.94 21.03 22.30 -7 21.80 21.86 22.20 22.22 22.23 22.70 22.76 8 22.52 22.57 22.69 22.70 22.70 22.95 22.99 9 22.85 22.87 22.93 22.93 22.94 23.07 -10 23.01 22.99 23.06 23.07 - . 23.16 -11 23.11 23.15 23.15 23.21 0 Edge 23.43 d calculated from Samson's (1966) method from Bergmark et a l . (1969) observed i n o p t i c a l absorption by Samson (1966) from Moore (1958). 1 - 79 -e x c i t a t i o n (Figure 16) exhibits r e l a t i v e l y intense structure as compared to the s electron excitation i n argon and krypton. This i s a r e f l e c t i o n of the increasing cross sections with increasing atomic number that are observed both above and below the I.P. for the rare gases. The ns and nd levels indicated i n the diagram are calculated from the spectroscopic values for the excited states of caesium and the I.P. of the 5s electron i n xenon, and the np levels are again those of Samson (see Table 3). The 6s value indicated i s from Bergmark et a l . I t appears that the forbidden t r a n s i t i o n s to ns states are making the strongest contribution to t h i s spectrum. However, there i s also evidence of 6p and 7p l e v e l s . The arrows at X, Y, and Z denote the energies at which Samson (1964) observed u n c l a s s i f i e d o p t i c a l t r a n s i t i o n s , possibly due to double electron excitations. The shoulder at 21.4 eV could be the 5d l e v e l which Bergmark et a l . put at 21.44 eV. The small structure at 22.3 eV appears to be from the 7p but i s also within the c a l i b r a t i o n uncertainty (± 0.1 eV) of the 6d l e v e l . Bergmark et a l . are able to resolve the nd contributions at high energy, but these are uncertain here. As for the peak at 23 eV, i t does not seem plausible that t h i s i s due to the higher Rydberg members of the ns, np, or nd s e r i e s , for then the cross section would not be expected to f a l l off abruptly u n t i l after the 0£ edge, whereas there appears too l i t t l e contribution i n the region before the edge. Above the 0^ edge there i s a peak at 23.57 eV, and another (not shown i n Figure 16) of comparable height at 24.54 eV. No further structure i s observed at electron energies up to 60 eV. - 80 -The peak a t 19.80 eV may be t h e t r i p l e t s t a t e ( J = 1) a s s o c i a t e d w i t h t h e s i n g l e t l e v e l ( J = 0) of t h e 6s l e v e l a t 20.08 eV. The 0.28 eV d i f f e r e n c e i s o f the r i g h t o r d e r of magnitude when compared w i t h t h e known s i n g l e t - t r i p l e t s p l i t t i n g s i n ms, ns s t a t e s i n Be ( 0 . 3 2 ) , Mg ( 0 . 2 8 ) , Ca ( 0 . 2 1 ) , Sr (0.19) and Ba ( 0 . 2 5 ) . The s m a l l peak a t Y c o u l d a l s o be a s i m i l a r t r i p l e t l e v e l a s s o c i a t e d w i t h t h e 7s s t a t e . Once a g a i n , t h e p o s s i b l e i n f l u e n c e f r o m d o u b l e e l e c t r o n e x c i t a -t i o n s and t e m p o r a r y n e g a t i v e i o n r e s o n a n c e s i s unknown. Double c o l l i s i o n e n e r g y - l o s s e s a r e d i s c o u n t e d . 5.6 Carbon Monoxide The e l e c t r o n i c e x c i t a t i o n o f c a r b o n monoxide has been s t u d i e d e x t e n s i v e l y , by u l t r a v i o l e t s p e c t r o s c o p y and by e l e c t r o n impact a t a v a r i e t y o f e n e r g i e s and s c a t t e r i n g a n g l e s . C o l l i n (1970) has c o m p i l e d t h e U.V. and e l e c t r o n i m p a c t d a t a . Kuppermann, R i c e and T r a j m a r (1968) o b s e r v e d t h e e n e r g y l o s s s p e c t r u m a t i n c i d e n t e n e r g i e s o f 25 t o 60 eV and a t a n g l e s f r o m 0° t o 80°. L a s s e t t r e and c o - w o r k e r s have a l s o r e p o r t e d e n e r g y - l o s s s t u d i e s a t e n e r g i e s from 48 t o 500 eV and a t a v a r i e t y o f s c a t t e r i n g a n g l e s ( L a s s e t t r e and S i l v e r m a n , 1964; Meyer e t a l . 1965; S k e r b e l e e t a l . , 1967; L a s s e t t r e e t a l . , 1968; L a s s e t t r e and S k e r b e l e , 1971). The t h r e s h o l d s p e c t r u m has been o b s e r v e d i n t h e t r a p p e d e l e c t r o n work o f Brongersma and O o s t e r h o f f (1967) and Rempt ( 1 9 6 7 ) , and as SF^ s c a v e n g i n g by H u b i n - F r a n s k i n and C o l l i n ( 1 9 7 0 ) . F i g u r e 17 shows t h e s c a v e n g i n g s p e c t r u m of CO as o b t a i n e d i n t h i s 3 work, and F i g u r e 18 shows t h e v i b r a t i o n s of t h e a n s t a t e a t h i g h e r 1 + 3 1 + 3 + r e s o l u t i o n . The s p i n f o r b i d d e n e x c i t a t i o n s X £ -> a n and X £ -»- b £ a r e s e e n as t h e most p r o m i n e n t peaks i n t h e s p e c t r u m , w h i l e t h e z D > cc < cc K DO, CC <, H Z LU cc CC; D U O Li_ a 3 7 r C O / S B IONIZATION 6 8 10 12 ELECTRON ENERGY (eV) 14 Figure 17. The Threshold Electron Impact Excitation Spectrum of Carbon Monoxide. - 82 v i b r a t i o n a l levels of ttie o p t i c a l l y allowed A^-n state are making only a small contribution between 8 and 9 eV. In contrast the 50 eV energy-loss spectrum by Skerbele et a l . (1967) finds the A^ir e x c i t a t i o n 3 to be approximately 1500 times as intense as the a TT and 250 times as 3 + 3 + 1 + 1 intense as the b £ . The assignment of the b £ , B £ , and F TT levels i s reasonably certain, since these are the only known states i n the respective areas. The broad peak centered at 11.3 eV i s probably due 3 + to the j E state (11.28 from T i l f o r d and Vanderslice, 1967, and 11.27 from Brongersma, 1968) , with possible contribution from C^"E+ and 3 + C £ at 11.39 and 11.41 eV respectively (Herzberg, 1950). The peak at 13 eV i s probably the G^TT l e v e l observed by Tanaka et a l . at 13.05. The small peak at 9.9 eV does not correspond to any known electronic l e v e l of carbon monoxide. I t i s also observed i n the trapped electron spectrum of Brongersma and Oosterhoff (1967) and of Rempt (1969) and i n the SF^ scavenging spectrum of Hubin-Franskin and C o l l i n (1970) but i s not seen i n the energy loss spectra. Brongersma and Oosterhoff explain the peak as negative ion contamination from the di s s o c i a t i v e capture process: CO + e(E = 10 eV). + C + o" Various studies (Craggs and Tozer, 1958; Schulz,' 1962; and Rapp and B r i g l i a , 1965) put the peak maximum for th i s process at 9.9 to 10.1 eV. Such negative ion contamination i s not expected i n SF^ scavenging because of the mass f i l t e r . Hubin-Franskin and C o l l i n suggest the peak may be due to a new unknown CO excited state or to an excited - 83 -transient negative molec'ular ion CO • Another p o s s i b i l i t y i s that a charge transfer reaction i s taking place upon c o l l i s i o n s of SF^ molecules and 0 ions. Such a reaction could occur i n the c o l l i s i o n chamber, ion gun or monopole regions. I f the oxygen atom was l e f t i n 2 2 3 3 the excited state configuration Is 2s 2p 3s, S, which has an e x c i t a t i o n energy of 9.52 eV, then the small amount of excess energy would be absorbed as k i n e t i c energy of the products: 0~ + SF 6 • 0 ( 3S) + SFg" I t i s not known what role such reactions play i n scavenging experiments, however. 3 The energy scale i n Figure 18 was calibrated using the 2 S state of helium at 19.82 eV, as observed i n a mixture of He, CO and SF^. The 3 energies of the a TT v i b r a t i o n a l levels (+ 0.05 eV) are then: 5.98 (v = 0); 6.20 (v = 1); 6.41 (v = 2); 6.62 (v = 3); and 6.80 (v = 4). This i s i n substantial agreement with other r e s u l t s . Brongersma and Oosterhoff find the f i r s t t r a n s i t i o n at 5.96 eV and the energy s p l i t t i n g to be 0.20 eV. Tanaka et a l . (1957) give the f i r s t three levels as 6.01, 6.22 and 6.43 eV, and Herzberg l i s t s the v = 0 l e v e l as 6.01 and the s p l i t t i n g as 0.21 eV. The r e l a t i v e i n t e n s i t i e s of these v i b r a t i o n a l levels did not appear to be a function of resolution, as was the case of the electronic levels of He. A comparison of the observed i n t e n s i t i e s with those from other electron impact results i s shown i n Table 4. I t V = 0 1 2 3 4 • *. •••• • • • . • _ • • • • » • • I l _ 6 .0 ELECTRON 7 . 0 ERGY (eV) a o ure 18. The Threshold Electron Impact Excitation Spectrum of the a t State of Carbon Monoxide. - 85 -TABLE 4 3 Approximate Relative I n t e n s i t i e s of V i b r a t i o n a l Levels of the a n State of Carbon Monoxide v = 0 1 2 3 This work 0.76 • 1.00 0.75 0.47 Calculated^ 0.84 1.00 0.69 0.36 c 50 eV energy loss 0.76 1.00 0.76 0.34 Trapped e l e c t r o n (high resolution) . 0.86 1.00 0.72 0.44 ir.apped e l e c t r o n (medium resolution) 1.00 0.96 0.68 0.38. with " r i s i n g background" cont r i b u t i o n subtracted. c a l c u l a t e d Franck-Condon f a c t o r s ( N i c h o l l s , 1962) measured from spectrum by Lassettre (1969) measured from spectrum by H a l l and Reinhardt (1970) measured from Brongersma and Oosterhoff (1967) and Rempt (1969) - 86 -can be seen that these are i n reasonable agreement, including the t h e o r e t i c a l c a l c u l a t i o n s of Franck-Condon factors by N i c h o l l s (1962), with the exception of the trapped electron r e s u l t s of Brongersma and Oosterhoff (1967) and Rempt (1969). It i s probable that some apparatus function i s a f f e c t i n g these trapped e l e c t r o n r e s u l t s . As pointed out for helium, one such apparatus function that can be expected to a f f e c t the r e l a t i v e i n t e n s i t i e s i s a detection system that does not c o l l e c t over a l l angles equally. Lasse'ttre et a l . (1968) 3 1 found that the r e l a t i v e i n t e n s i t i e s of the a II and A It states changed very l i t t l e with s c a t t e r i n g angle or e l e c t r o n energy at incident energies of 48 and 98 eV. Kuppermann et a l . found the same to be true at incident energies of 25 and 35 eV. However, at energies closer to threshold the s i t u a t i o n appears to be d i f f e r e n t , f o r Trajmar (1970) finds the band i n t e n s i t i e s at.10 ev do indeed depend on angle and d i f f e r from Franck-Condon pr e d i c t i o n s at a l l angles. This may be because the e x c i t a t i o n proceeds through a temporary negative ion at threshold. In any case, where such angular v a r i a t i o n s do occur, threshold experiments may w e l l give non-equivalent r e s u l t s i f the zero energy electrons are not c o l l e c t e d over a complete 4TT geometry. Scavenging experiments detect over a l l angles equally, but probably most trapped electron spectrometers do not. The transient molecular CO ion at approximately 2 eV (Schulz, 1959, 1964; Rempt, 1969; Hubin-Franskin and C o l l i n , 1970) was not observed i n t h i s study. This may have been because of the large scattered background i n t h i s region, which could obscure any possible structure. - 87 -5.7 Summary and Conclusions The half-width of the scavenging peaks i s about 0.2 eV. This compares favorably to the resolution of most other threshold studies, although recently somewhat higher resolutions have been achieved (Schulz, 1969; H a l l et a l . , 1970). A half-width of 0.2 eV i s s u f f i c i e n t to resolve the v i b r a t i o n a l levels of small molecules l i k e carbon monoxide. For helium the r e l a t i v e i n t e n s i t i e s of the peaks i n the threshold spectrum are a function of the resolution, which appears to be due to rapid changes i n the cross sections of the i n d i v i d u a l excitations near threshold. As expected, the threshold electron impact spectra show many excited states which are not observed i n o p t i c a l e x c i t a t i o n or i n high energy electron impact ex c i t a t i o n . For helium and carbon monoxide, s i n g l e t - t r i p l e t t r a n s i t i o n s dominate the spectrum. For Ne, Ar, Kr, and Xe the m u l t i p l i c i t y i s less w e l l defined and i t was generally not possible to resolve the i n d i v i d u a l contributions from states of different J values (except i n one case, the 5p 6s e x c i t a t i o n i n Xe) . For the rare gases Ne-Xe, ex c i t a t i o n to s and p o r b i t a l s appears to be most favored, although i n most cases the contribution from d levels cannot be accurately assessed because of overlap with s and p l e v e l s . In a l l cases, the spectra show a large r i s e beginning about 0.5 eV before the f i r s t i o n i z a t i o n p o t e n t i a l . I t i s suggested that reactions of highly excited atoms with SF^ are contributing to the SF^ ion current i n t h i s region. The attempts to correlate the structure i n the i o n i z a t i o n continuum with the s electron excitation i n Ar, Kr, and Xe appear to be at least p a r t i a l l y successful. Other structure may be due to double electron - 88 -excitations. The possible interference from temporary negative ion states makes this structure d i f f i c u l t to assign. For Ar, Kr and Xe, the io n i z a t i o n of the outer s electron i s not marked by any large feature i n the scavenging spectrum. - 89 -BIBLIOGRAPHY Anderson, N., P.R. Eggleton and R.G.W. Keesing, Rev. S c i . I n s t r . , 38, 924 (1967); i b i d . , 38, 1524 (1967). Bardsley, J.M., and F. Mandl, "The Resonant Scattering of Electrons by Molecules", Progress i n Physics, _31, 471 (1968). Begun, G.M., and R.N. Compton, J. Chem. Phys., 51, 2367 (1969). Bergmark, T., R. Spohr, N. Magnusson, L. Werme, C. Nordling and K. Siegbahn, Uppsala University I n s t i t u t e of Physics Report No. ' 589 (1969). Boersch, H. , J. Geiger and H. Hellwig, Phys. Letters, 3^, 64 (1962). Boersch, H., Abstr. 4th Int. Conf. on the Physics of Electronic and Atomic C o l l i s i o n s , Quebec (New York: Science Bookcrafters), 351 (1965). Brion, C.E., and G.E. Thomas, J. Mass Spectrometry and Ion Physics, 1, 25 (1968). Brion, C.E., and C.R. Eaton, J. Mass Spectrometry and Ion Physics, JL, 102 (1968), Brion, C.E., C.R. Eaton, L.A.R. Olsen, and G.E. Thomas, Chem. Phys. Letters, J3> 6 0 0 (1969). Brion, C.E., and L.A.R. Olsen, Phys. Rev., 187, 111 (1969). Brion, C E . , and L.A.R. Olsen, J. Chem. Phys., J52_, 2163 (1970a). Brion, CE. , and L.A.R. Olsen, J. Phys. B: Atom Molec. Phys., _3» 1 0 2 0 (1970b). Brongersma, H.H., and E.J. Oosterhoff, Chem. Phys. Le t t e r s , 1, 169 (1967). Burke, P.G., "Close Coupling Calculations and Results" i n "Autoionization", A. Temkin, Ed., (Baltimore, Mono Book Corp.) (1966). Burke, P.G., and M.M. Schey, Phys. Rev., 126, 147 (1962). Burrow, P.D., and G.J. Schulz, Abstr. 6th Int. Conf. on the Physics of Electronic and Atomic C o l l i s i o n s (Boston: MIT Press), p. 561 (1969). Cermak, V., and Z. Herman, C o l l . Czech. Chem. Comm., 29_, 953 (1964). Chamberlain, G.E., Phys. Rev., 155, 46 (1967). - 90 -Chamberlain, G.E., H.G.M. Heideman, J.A. Simpson, and CE. Kuyatt, Abstr. 4th Int. Conf. on the Physics of Electronic and Atomic C o l l i s i o n s , Quebec (New York: Science Bookcrafters), p. 378 (1965). Chen., J.C.Y., "Theory of Transient Negative Ions of Simple Molecules" i n Advances i n Radiation Chemistry, M. Burton and J.M. Magee, Ed., (New York: John Wiley) 1969; Clark., E.M., Can. J. Phys., 32, 764 (1954). Compton, R.N., L.G. Christophorou, and R.H. Huebner, Phys. Letters, 23, 656 (1966). Compton, R.N., R.H. Huebner, P.W. Reinhardt, and L.G. Christophorou, J. Chem. Phys., 48, 901 (1968). Compton, R.N., J.A. Stockdale and P.W. Reinhardt, Phys. Rev., 180, 111 (1969). Craggs, J.D., and B.A. Tozer, Proc. Roy. Soc. (London), A247, 337 (1958). Curran, R.K. , J. Chem. Phys., JI8, 780 (1963). Dawson, P.H., and N.R. Whetten, Research and Development (Vacuum Technology) 19_, 2 (1968a). Dawson, P.H. , and N.R. Whetten, Rev. S c i . In s t r . 39_, 1417 (1968b). Edwards, A.F., and M.E. Rudd, Phys. Rev., 170, 140 (1968). Erhardt, H., and K. Willmann, Z. Phys., 203,1(1967). Erhardt, H., L. Langhans and F. Linder, Z. Phys., 214, 179 (1968). Fano, II., Phys. Rev., 124, 1866 (1961). Fano, U. , and J.W. Cooper, Phys. Rev., 138>, A400 (1965). Flugge, S., Ed., Handbuch der Phys. 15, Atoms 1 (B e r l i n : Springer-Verlag) (1957). Foner, S.N., and B.H. H a l l , Phys. Rev., 122, 512 (1961). Fox, R.E., W.M. Hickam, D.J. Grove and T. Kjeldass, Rev. S c i . Instr. 26, 1101 (1955). Gerber, G., R. Morgenstern, and A. Niehaus, Phys. Rev. Letters, 23, 511 (1969). Gerjuoy, E., Physics Today, p. 24 (May 1965). Golden, D.E., and H.W. Bandel, Phys. Rev. Letters, 14, 1010 (1965), - 91 -Golden, D.E., and A. Zecca, Rev. S c i . Instr. 5j42, 210 (1970). Grissom, J.T., W.R. Garrett, and R.N. Compton, Phys. Letters, 30A, 117 (1969). Grissom, J.T., R.N. Compton, and W.R. Garrett, Phys. Rev. Letters, 23, 1011 (1969). Grissom, J.T., R.N. Compt on, and W.R. Garrett, Oak Ridge N a t l . Lab. Report No. 0RNL-TM-2618 (1970). H a l l , R.J., J. Mazeau, J. Reinhardt, and C. Schermann, J. Phys. B., J?, 991 (1970). H a l l , R.I., private communication (1970). Harrower, G.A., Rev. Sc i . Instr.,26, 850 (1955). Herzberg, G., "Atomic Spectra and Atomic Structure" (New York: Dover Publications) (1944). Herzberg, G., "Spectra of Diatomic Molecules", 2nd E d i t i o n , (New York: Van Nostrand) (1950). Hotop, H., and A. Niehaus, J . Chem. Phys., 47, 2506 (1967). Hubin-Franskin, M.-J. and J.E. C o l l i n , J . Mass Spectrometry and Ion Phys., 4, 451 (1970); i b i d . , 5, 163 (1970); i b i d , 5, 255 (1970). Huebner, R.H., R.N. Compton, and H.C. Schweinler, Chem. Phys. Letters, 2, 407 (1968). Huebner, R.H., R.N. Compton, L.G. Christophorou, and P.W. Reinhardt, Oak Ridge Natl. Lab. Report No. ORNL-TM-2156 (1968). Hughes, A.L. and V. Rojansky, Phys. Rev., _34, 284 (1929). Hughes, A.L. and J.M. McMillen, Phys. Rev., 34, 291 (1929). Jacobs, G., and A. Henglein, Z. Naturf., 239, 1383 (1964). Kuppermann, A., J.K. Rice, and S. Trajmar, J. Phys. Chem. > 7_2, 3894 (1968). Kuyatt, C.E., J.A. Simpson and S.R. Mielczarek, Phys. Rev., 138, A385 (1965). Lassettre, E.N., and S.M. Silverman, J . Chem. Phys., 40, 1265 (1964). Lassettre, E.N., A. Skerbele, M.A. D i l l o n and K.J. Ross, J. Chem. Phys. , j48, 5066 (1968). -.92 -Lassettre, E.N., Can. J.- Chem., 47, 1733 (1969). Lassettre, E.N. and A. Skerbele, J. Chem. Phys., 5_4, 1597 (1971). L i f s h i t z , C., and R. Grajower, J . Mass Spectrometry and Ion Phys., 4-, 92 (1970). Lindsay, P.A., i n "Advances i n Electronic and Electron Physics", L. Marton, Ed., 13, (New York: Academic Press Inc.) p. 202 (1960). McDaniel, E.W. " C o l l i s i o n Phenomena i n Ionized Gases", (New York: Wiley), p. 305 (1964). McGowan, J.W., and E.M. Clarke, Phys. Rev. 16_7, 43 (1968). Madden, R.P., and K. Codling, Phys. Rev. Letters, 10, 516 (1963); J. Opt. Soc. Am. , 54-, 268 (1964). Madden, R.P., and K. Codling, i n "Autoionization" A. Temkin, Ed., (Baltimore, Mono Book Corp.) (1966). Madden, R.P., D.L. Ederer, and K. Codling, Phys. Rev., 177^ , 136 (1969). Marmet, P., and L. Kerwin, Can. J . Phys.,_38, 787 (1960). Marmet, P., Can. J. Phys., 42, 2120 (1964). Meyer, V.D., A. Skerbele and E.N. Lassettre, J. Chem. Phys., 43, 805 (1965). Moisewitsch, B.L. , and S.J. Smith, Rev. Mod. Phys. 4i0, 238 (1968). Moore,. CE. , "Atomic Energy Levels", 1, 2, and _3, National Bureau of Standards Ci r c u l a r No. 467 (Washington: U.S. Gov't P r i n t i n g Office) (1949, 1952, 1958). Mott, N.F., and H.S.W. Massey "The Theory of Atomic C o l l i s i o n s " , 2nd Edi t i o n , (Oxford Press), p. 136 (1949). N i c h o l l s , R.W., J. Quant. Spectry. Radiative Transfer, 2, 433 (1962). O'Malley, R.M. and K.R. Jennings, J. Mass Spectrometry and Ion Physics, 2, App. 1 (1969). Paul, W., and M. Raether, Z. Physik, 140, 262 (1955). Peterkop, R., and Veldre, V., "The Theory of Electron-Atom C o l l i s i o n s " , i n "Advances i n Atomic and Molecular Physics" D.R. Bates and I. Estermann, Ed. (New York: Academic Press) (1966). Ramsauer, C , Annal. Phys.,64_, 513 (1921). - 93 -Rapp, D., and D.D. B r i g l i a , J. Chem. Phys., 43, 1480 (1965). Rempt., R.D., Phys. Rev. Letters, _22, 1034 (1969). Rendina, J.F. and R.E. Grojean, Applied Optics, 9_, 2148 (1970). Rice, J.K., A. Kuppermann and S. Trajmar, J . Chem. Phys., 4_8, 945 (1968) . Ridge, D.P., and J.L. Beauchamp, J. Chem. Phys., 51, 470 (1969). Rudd, M.E., and D".V. Lang, Abstr. 4th Int. Conf. on the Physics of Electronic and Atomic C o l l i s i o n s , Quebec (New York: Science Bookcrafters), p. 153 (1965). Salop, A., D.E. Golden and H. Nakano, Rev. S c i . I n s t r . , 40, 733 (1969) . Samson, J.A.R. , Phys. Letters, _8, 107 (1964). Samson, J.A.R., Advances i n Atomic and Molecular Physics. Vol. 2, (New York: Academic Press), p. 177 (1966). Schulz, G.J.', Phys. Rev., 112, 150 (1958). Schulz, G.J., Phys. Rev., 116, 5 (1959). Schulz, G.J., J. Appl. Phys.,31, 1134 (1960). Schulz, G.J., Phys. Rev. 128, 178 (1962). Schulz, G.J., Phys. Rev.letters,10, 104 (1963). Schulz, G.J., Phys. Rev., 136, A650 (1964). Skerbele, A., M.A., D i l l o n and E.N. Lassettre, J. Chem. Phys. 46, 4162 (1967). Simpson, J.A., Rev. S c i . Instr., 35, 1698 (1964). Simpson, J.A.,G.C. Chamberlain and S.R. Mielczarek, Phys. Rev., 139, A1039 (1965). Simpson, J.A. , CE. Kuyatt, and S.R. Mielczarek,' J. Chem. Phys., 44, 4403 (1966). Simpson, J.A., and CE. Kuyatt, Rev. S c i . Ins t r . y 38, 103 (1967). Stamatovic, A., and G.J. Schulz, Rev. S c i . I n s t r . , 3£, 1752 (1968). Stamatovic, A., and G.J. Schulz, Phys. Rev., 188, 213 (1969). Takanka, Y. , A.S. Jursa and F.J. Leblanc, J. Chem. Phys., 26^ , 862 (1957). - 94 -Taylor, H.S., G.V. Nazaroff and A. Golebiewski, J. Chem. Phys. 45, 2872 (1966). T i l f o r d , S.G., and J.T. Vanderslice, J. Mol. Spectrosc. 26_, 419 (1968). Thomas, G.E., Ph.D. Thesis, University of B r i t i s h Columbia (1969). Trajmar, S., private communication (1970). Von Zahn, U., Rev. S c i . I n s t r . , 34, 1 (1963). 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

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.831.1-0060045/manifest

Comment

Related Items