Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Kinetic studies of O(¹S) formation from atomic oxygen recombination Wassell, Peter Thomas 1981

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-UBC_1981_A1 W37_7.pdf [ 7.51MB ]
Metadata
JSON: 831-1.0060743.json
JSON-LD: 831-1.0060743-ld.json
RDF/XML (Pretty): 831-1.0060743-rdf.xml
RDF/JSON: 831-1.0060743-rdf.json
Turtle: 831-1.0060743-turtle.txt
N-Triples: 831-1.0060743-rdf-ntriples.txt
Original Record: 831-1.0060743-source.json
Full Text
831-1.0060743-fulltext.txt
Citation
831-1.0060743.ris

Full Text

KINETIC STUDIES OF 0( S) FORMATION FROM ATOMIC OXYGEN RECOMBINATION by PETER THOMAS WASSELL B.Sc, U n i v e r s i t y of Exeter, 1973 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE -OF DOCTOR OF PHILOSOPHY in the Department of CHEMISTRY We accept t h i s thesis as comforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA @ June 1981 In presenting t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f the r e q u i r e -ments f o r an advanced degree at the U n i v e r s i t y of B r i t i s h Columbia, I agree that the L i b r a r y w i l l make i t f r e e l y a v a i l a b l e f o r reference and study. I f u r t h e r agree that permission f o r extensive copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the Head of my Department or by h i s r e p r e s e n t a t i v e s . I t i s understood that copy-ing or p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l gain w i l l not be allowed without my w r i t t e n permission. Department of Chemistry The U n i v e r s i t y of B r i t i s h Columbia Vancouver 8, Canada Date: IS yJuU^, / W ( i i ) ABSTRACT The molecular and atomic dependencies of the O^S) emission have been studied i n the laboratory by generating oxygen atoms in a discharge flow system. The i n t e n s i t y of the emission i s found to have the dependence k T[0] 2[M] 1(557.7 nm) = — = where M i s the t o t a l p a r t i c l e concentration i n the system. From absolute measurements of the 0( 1S) atomic l i n e , the o v e r a l l rate constant i s found to be equal to 2.7 ± 0.3 x 10 cm s at 300K. The observed dependence i s shown to be inconsistent with the proposed Chapman mechanism f o r the e x c i t a t i o n of 0( 1S) k 0 + 0 + 0 i.0(*S) + o 2 1 . rv (JJ 1 r" 1 1 0( S) ——*» 0 ( D ) + hv(557.7 nm) However the observations are found to be i n agreement with a * "Barth-type" mechanism, where a metastable oxygen molecule ( 0 2 ) i s formed i n the recombination of two oxygen atoms i n the presence of a t h i r d body, 0 + 0 + M 0 2 + M (2) in ( i i i ) followed by energy t r a n s f e r to a t h i r d oxygen atom to form 0(*S) 0 2* + 0 — L o ^ S ) + 0 2 (3) and the r a d i a t i v e emission of Of^S) at 557.7 nm. * The possible e l e c t r o n i c states of 0^ corresponding to 0 2 are d i s -cussed. Although the i d e n t i t y of the 0^ remains to be found, the major loss process of t h i s intermediate i s found to be quenching by atomic oxygen k4 0 2 +0 — • » quenched products (4) rather than by molecular oxygen or argon * k ^2 + ^2 *» quenched products (5) k 6 0 2 + Ar *» quenched products (6) U n t i l r e c e n t l y the major quencher of 0( 1S) was thought to be the ground-state oxygen atom k7 0( 1S) + 0( 3P) *» quenched products (7) However, i n t h i s system, the dependence of the 557.7 nm emission found to be consistent with the predominate quenching of 0(^S) i s by 0 2 ( a 1 A g ) : k n^S) + o (a A ) ^.quenched products (8) ^ 2 The rate constant k g i s estimated to be 7 ± 3 x 10 1 cm s by comparison for quenching of 0( 1S) by 0 2-(iv) 1 k 9 0( S) + 0 2 i» quenched products (9) 0 ?(a 1A ) i s expected to be present i n a l l laboratory systems involving atomic oxygen due to i t s formation by e i t h e r heterogeneous 3 or homogeneous recombination 6'f (K P) . Due to the magnitude of kg, O^Ca^A ) i s expected to be the major quencher of 0( 1S) i n most of these systems. Using currently accepted values for the concentration of (^(a^A ) i n the t e r r e s t r i a l nightglow layer, the quenching of 0(*S) by 0„(a*A ) i s found to be unimportant i n t h i s region compared to i t s r a d i a t i v e decay. Other atmospheric implications of t h i s inves-t i g a t i o n ' s r e s u l t s are discussed. (v) TABLE OF CONTENTS PAGE ABSTRACT TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES ACKNOWLEDGEMENT PREFACE CHAPTER 1: INTRODUCTION 1.1 E l e c t r o n i c States and Radiative T r a n s i -tions of Atomic and Molecular Oxygen 1.1.1 Atomic Oxygen Energy Levels 1.1.2 Atomic Oxygen Tra n s i t i o n s 1.1.3 Energy Levels of Molecular Oxygen 1.1.4 Radiative t r a n s i t i o n s of Molecular Oxygen 1.2 Atmospheric Emission of 0( 1S) 1.2.1 Aurora 1.2.2 Dayglow 1.2.3 Nightglow 1.3 Laboratory Studies of Oxygen Atom Recombina-t i o n 1.3.1 Generation of Atomic Oxygen 1.3.2 Measurements of the Atomic Oxygen Concentration 1.3.3 Detection of Metastable Atoms and Mole-cules i n the Gas Phase 1.3.4 • Absolute Intensity Measurements ( i i ) (v) ( v i i i ) (x) ( x i i i ) (xiv) 1 5 5 6 7 13 19 19 23 27 46 47 50 61 64 (vi) PAGE 2.3.2 Method 2 -- Isothermal Calorimetric Detection 2.3.3 Method 3 -- Measurement of the Intensity of N0 2 Emissions from the 0-NO Reaction 2.4 Absolute Emission Intensity Measurements 2.4.1 The 557.7 nm Emission from 0( 1S) 2.4.2 The 1.27 ym Emission from 0_(a 1A ) CHAPTER 3: RESULTS AND ANALYSIS (PART 1) 3.1 Molecular Oxygen Dependence of the 557.7 nm Emission 3.2 Atomic Oxygen dependence of the 0(*S) Emission 3.3 Pressure or M Dependence of the 557.7 nm Emission 70 70 CHAPTER 2: EXPERIMENTAL 2.1 Materials 2.1.1 Gas Handling and Storage 72 2.2 Apparatus 73 2.2.1 Flow System 2.2.2 Generation of Atomic Oxygen 2.2.3 Flow and Pressure Measurement 79 2.2.4 Optical Detection System 82 2.3 Atomic Oxygen Measurement 85 2.3.1 Method 1 Chemical T i t r a t i o n 87 74 78 89 95 96 98 103 105 106 113 121 3.3.1 Collision-Induced Emission 1 2 6 3.4 Quenching by Other Gases 130 ( v i i ) 3.5 Temperature Dependence of the Quenching of the 557.7 nm Emission by 0^ 3.6 Summary of the Preliminary Results CHAPTER 4: RESULTS AND ANALYSIS (PART 2) 4.1 V a r i a t i o n of [ 0 o ( a 1 A ) ] 4.2 0 7 ( a 1 A ) Quenching of 0( 1S) 4.3 Atomic Oxygen Dependence of the 557.7 nm Emission 4.4 Pressure or M Dependence of the Of^S) Emission Intensity 4.5 The Complete Rate Equation 4.6 Determination of the Absolute Emission Constant 4.7 Determination of k o CHAPTER 5: DISCUSSION 5.1 The Quenching of 0( 1S) by 0 2 ( a 1 A g ) 5.2 E f f e c t of Quenching of 0( 1S) by 0 2 ( a 1 A J on E a r l i e r Investigations 5.3 Properties of the Precursor 5.4 Atmospheric Implications of t h i s Investigat CHAPTER 6: CONCLUSIONS APPENDIX REFERENCES ( v i i i ) LIST OF TABLES PAGE Table 1.1 Theoretical and Experimental Results for^the Radiative T r a n s i t i o n P r o b a b i l i t i e s of 0( S) Table 1.2 Theoretical and Experimental Results for^the Radiative T r a n s i t i o n P r o b a b i l i t i e s of 0( D) Table 1.3 D i s s o c i a t i o n Threshold Wavelengths i n the Photolysis of 0 2 Table 1.4 Quenching Rate Constants of 0( 1S) Table 1.5 Reported Rate Constants for the E x c i t a t i o n of 0(*S) Table 1.6 Rate Constants for the Quenching of 0 2(A 3E^) Table 1.7 Surface Recombination C o e f f i c i e n t y of Oxygen-Atoms at 20°C Table 2.1 Excited Species Detection by the Isothermal Probe Table 3.1 Intercept/Slope Values of 0« Quenching Plots at D i f f e r e n t [0] Table 3.2 Oxygen Atom Dependence of the 557.7 nm Emission Obtained by Measuring [0] with the Isothermal Probe Technique Table 3.3 Oxygen Atom Dependencies Determined i n the 2 and 5 - l i t r e Observation C e l l s Table 3.4 Results of the [0] 2/I(557.7 nm) Versus [0] Plots Table 3.5 [M] Dependence at Constant [0] Using the Iso-thermal Probe to Measure the Relative 0-atom Flow Rate Table 3.6 Stern-Volmer Quenching Slopes for Several Gases Table 3.7 Temperature Dependence of 0 2 Quenching of the 557.7 nm Emission 25 33 35 39 51 94 110 116 119 122 127 133 139 (ix) PAGE Table 4.1 Atomic Oxygen Dependence of^the 557.7 nm Emission at D i f f e r e n t [0„(a A )] Table 4.2 Evaluation of the Absolute Emission Intensity Constant Table 5.1 A l t i t u d e Dependence of [ 0 2 ] , [M], [0] and 557.7 nm Emission Table 5.2 Calculated k k A 4 Values Table 5.3 Relative M E f f e c t i n 3-body Reactions 150 155 182 184 186 (x) LIST OF FIGURES PAGE 10 Figure 1.1 The Spectrum of the T e r r e s t r i a l Nightglow from 400 to 700 nm Figure 1.2 The Energy Levels of the Three Lowest Elec-t r o n i c States of the Oxygen Atom Figure 1.3 P o t e n t i a l Energy Curves for the Six Lowest States of Molecular Oxygen 12 Figure 1.4 The Major Emission Band Systems of Molecular Oxygen 18 Figure 1.5 The A l t i t u d e Dependence of the 0(*S) E x c i t a -t i o n Mechanisms and Measured Emission Inten-s i t i e s i n the Dayglow 26 Figure 1.6 Rocker-borne Photometer Measurements of the A l t i t u d e P r o f i l e of the 557.7 nm Emission Intensity i n the Nightglow 42 Figure 1.7 The A l t i t u d e P r o f i l e of the Atomic Oxygen Density Measured with Rocket-borne Resonance Lamps 43 Figure 1.8 Simultaneous A l t i t u d e P r o f i l e Measurements of the 557.7 nm, Herzberg I Emissions and Atomic Oxygen Density P r o f i l e 45 Figure 1.9 Normalized Spectrum of the NO-0 Chemiluminescence Obtained by Several Workers Figure 1.10 Absolute Rate Constants f o r the Chemiluminescence Reaction Between NO and 0 as a Function of Wave-length Figure 2.1 Schematic of the Apparatus used i n t h i s I n v e s t i -gation Figure 2.2 The Dependence of the flow Rate on the Total Pressure of Argon i n the Discharge Flow System 58 68 75 (xi) PAGE Figure 2.3 Spectral Responce of the V i s i b l e Detection System . 84 Figure 2.4 The Observed Spectrum of the 1.27 ym Emission from 0 o(a 1A ) n 2 V g v=0 Figure 2.5 Wheatstone Bridge C i r c u i t of the Isothermal Calorimetric Probe to the NO* Emission 86 91 Figure 2.6 Intensity of the Emission at 556.4 nm as a Function of NO added to a Gas Stream of Nitrogen Atoms 97 Figure 2.7 Schematic Representation of the Spectral Dis-t r i b u t i o n of the 0( S) Emission with Respect 99 108 111 Figure 3.1 The j ^ ] Dependence of the 557.7 nm Emission Intensity Figure 3.2 The Oxygen Atom Dependence of the Intercept/Slope Ratio of the 0^ Quenching Plots Figure 3.3 The Oxygen Atom Dependence of the 557.7 nm Emission Intensity Plotted as log 1(557.7 nm) [0 ] versus log [0] 115 Figure 3.4 The Pressure Dependence of the 557.7 nm Emission Intensity 123 Figure 3.5 The Pressure Dependence of the 557.7 nm Emission Intensity using the Isothermal Probe Technique to Measure [0] i n 19557.7 nm)/[0] 2 125 Figure 3.6 Spectrum of the Collision-induced Emission from Ar - 0( 1S) Figure 3.7 Plot of the Inverse of the Quenching Slope Minus I n i t i a l 0 ? Concentration versus the Atom Oxygen Concentration 135 Figure 3.8 The Dependence of the Inverse of the Stern-Volmer Quenching Slope M u l t i p l i e d by the I n i t i a l 0^ Con-centration on [O]/!^] for Several Gases 136 129 ( x i i ) PAGE Figure 3.9 Arrhenius Plot of the Temperature Dependence of the 0 2 Quenching Plots 140 Figure 4.1 The 0 ?(a 1A ) Dependence of the 557.7 nm Emission Intensity g 146 Figure 4.2 The Atomic Oxygen Dependence of the 1(557.7 nm) Emission Intensity i n a l n - l n Plot 148 Figure 4.3 The Pressure Dependence of the 557.7 nm Emission Intensity over the Range of 2-7 t o r r Argon 151 Figure 4.4 Plot of 1/(557.7 nm)[0 2(a 1A )] versus [0] 2[M] 156 Figure 4.5 Plot of 1/1(557.7 nm)[0 ?(a 1A )] versus [0-]/ [ 0 2 ( a 1 A g ) ] g 160 Figure 4.6 I l l u s t r a t i o n of the E f f e c t of D i f f e r e n t Values of k i n the Interpretation of the 0 2 Quenching of tne 557.7 nm Emission 162 Figure 5.1 The A l t i t u d e Dependence of [0 2(a*A )] i n the Dayglow and Nightglow g Figure 5.2 The A l t i t u d e P r o f i l e s of the Major Constituents i n the Region of the T e r r e s t r i a l Nightglow 175 Figure 5.3 Loss Rates of 0( XS) i n the Nightglow 179 lc-\ -; „ + V . ^ M i n V i + n l n u 181 ( x i i i ) ACKNOWLEDGEMENTS My sincerest appreciation and thanks to Dr. E.A. Ogryzlo for his guidance and support throughout the course of t h i s work. My thanks also to Dr. R. Kenner, my co-worker, f o r h i s suggestions and h i s assistance, and for proof-reading t h i s manu-s c r i p t . I would l i k e to thank the capable technicians i n the E l e c t r i c a l and Mechanical Workshops for t h e i r assistance, with s p e c i a l thanks to the glassblower, Mr. S. Rak, f o r the development and construc-t i o n of the glassware used i n t h i s work. F i n a l l y , I would l i k e to o f f e r very special thanks to my wife, Nancy Wiggs, for typing t h i s manuscript. (xiv) PREFACE Since the u n i t s of l i g h t i n t e n s i t y , p a r t i c l e concentrations, and rate constants used i n t h i s thesis may appear unorthodox to chemists, the following comments may be useful to those unfamiliar with aeronomy. Upper atmosphere r a d i a t i o n emission i n t e n s i t i e s are normally reported in " r a y l e i g h s " where 1 r a y l e i g h = IR = apparent emission rate of 1 megaphoton -2 , , ^-1 -1 cm (column) s The u n i t i s convenient because for a t y p i c a l 10 km thick emitting layer 1 r a y l e i g h i s equivalent to an emission of 1 photon per cubic centimetre per second. It i s t r a d i t i o n a l to omit the word photon -3-1 and write the units as "cm s ". S i m i l a r l y , concentrations are _3 expressed as p a r t i c l e s per cubic centimetre and written "cm " where " p a r t i c l e " i s omitted. Second and t h i r d order rate constants have un i t s of 3-1 -1 cm s (where " p a r t i c l e s " i s omitted) and 6-1 - 2 cm s (where " p a r t i c l e s " i s omitted) re s p e c t i v e l y . (XV ) Unless otherwise s p e c i f i e d , a molecule or an atom i s denoted i n i t s ground state, f o r example o2 = o2cxV) In a recent designation o f the 0^ states by K.P. Huber and G. Herzberg ("Constants of Diatomic Molecules", Van Nostrand Rheinhold, 3 3 New York (1979)), the 0_(C A ) state has been renamed 0.(A'A ) , v J J 2 u 2 u 7 however the former has been used throughout t h i s t h e s i s . To aid the reader, those reactions to which most frequent reference i s made are l i s t e d i n an appendix at the end of the thesis. CHAPTER 1: INTRODUCTION The earth's upper atmosphere has been the subject of intense i n t e r e s t i n the l a s t few years due to the p o t e n t i a l harm of n i t r o -gen oxides and fluorocarbons to the ozone layer. Fundamental to an understanding of such perturbations i s a knowledge of the i d e n t i t y and concentrations of species present i n the upper atmosphere. Oxygen i s important i n the chemistry of the upper atmosphere because of i t s abundance and the c r u c i a l r o l e of molecular oxygen and i t s al l o t r o p e ozone in the absorption of u l t r a - v i o l e t r a d i a t i o n . Excited states of molecular and atomic oxygen have been i d e n t i f i e d i n the upper atmosphere from the spectrum of the nightglow (Figure 1.1). In d i r e c t i o n s away from the Milky Way, most of the v i s i b l e l i g h t at night comes from a glowing layer of gas in the earth's atmosphere. The most prominent feature of t h i s glow, as can be seen from the spectrum, i s the 557.7 nm emission from the *S excited state of atomic oxygen. The maximum i n t e n s i t y of t h i s emission has been shown [1] to be from the 95-100 km a l t i t u d e region of the upper atmosphere, which i s also the region of maximum atomic oxygen density [2]. The concentration of atomic oxygen i n the upper atmosphere i s extremely v a r i a b l e . Although the a l t i t u d e dependence of the atomic oxygen density shows the largest v a r i a t i o n , s i g n i f i c a n t differences have been observed according to the time of day, l a t i t u d e , season - 2 -Figure 1.1 The spectrum of the t e r r e s t r i a l nightglow from 400 to 700 nm. (Reproduced from the spectrum of A.L. Broadfoot and K.R. Kendall, J. Geophys. Res., 7_3 (1968) 426.) - 2a -400 500 600 W A V E L E N G T H (nm) 700 - 3 -and even i n l o c a l i s e d areas of j u s t a few square miles. A s i m i l a r v a r i a t i o n has been observed i n studies of the 0(*S) 557.7 nm emission [ 3 ] . The use of the 557.7 nm emission, observed from the ground, as a monitor of atomic oxygen d e n s i t i e s i n the upper atmosphere has been a major goal of aeronomers i n the past two decades. To achieve t h i s goal, a knowledge of the formation and loss processes of the 0(^S) atom i s required. The "green l i n e of the night sky" has been the subject of numerous inv e s t i g a t i o n s and consi-derable controversy ever since i t was f i r s t observed i n the nine-teenth century [4]. Following the i d e n t i f i c a t i o n [5] of the t r a n s i -t i o n , the e x c i t a t i o n of Of^S) was proposed i n 1931 [6] to be due to the reac t i o n . 0 + 0 + 0 — 0 ( 1 S ) + o 2 (1) which has come to be c a l l e d the "Chapman mechanism". This r e a c t i o n was generally accepted u n t i l 1961 when Barth [7] suggested a two step mechanism, i n order to explain the absence of the "green l i n e " i n laboratory systems containing oxygen atoms. Barth's proposed * scheme involved the formation of an intermediate 0^ molecule i n the recombination of two oxygen atoms. 0 + o + M — 0 2 * + M (2) (where M i s any " t h i r d body" present) followed by energy t r a n s f e r to a t h i r d oxygen atom forming 0("^S): - 4 0* + 0 — - * 0 ( 1 S ) + 0 2 (3) The 557.7 nm emission i s due to the r a d i a t i v e t r a n s i t i o n of 0( 1S) to the 0( 1D) state 0( 1S) •»0( 1D) + hv(557.7 nm) . (10) Both mechanisms can y i e l d rate equations consistent with labora-tory and e a r l i e r atmospheric observations. However, i n the Barth mechanism, the 0 2 precursor (with i t s undetermined properties) could lead to very d i f f e r e n t calculated oxygen atom d e n s i t i e s , since laboratory measurements are obtained under conditions that are very d i f f e r e n t from those of the upper atmosphere. Both mechanisms have been accepted and rejected by various workers i n the decades follow-ing Barth's proposal. To date, no laboratory studies have been able to provide con-vi n c i n g evidence for e i t h e r mechanism. To some extent t h i s can be at t r i b u t e d t o the i n a b i l i t y of anyone to observe the 0(*S) emission i n a conventional oxygen discharge flow system. Preliminary experiments i n the laboratory showed that the emission at 557.7 nm does indeed occur when oxygen atoms are pre-pared in a gas stream that i s p r i n c i p a l l y argon. Such an observa-t i o n made i t possible to undertake a k i n e t i c study of t h i s species, i n order to d i s t i n g u i s h between the proposed e x c i t a t i o n mechanisms i n both the laboratory flow system and i n the upper atmosphere. - 5 -Before reviewing the emissions of the 0(^S) atom, i t i s useful to consider the relevant e l e c t r o n i c energy l e v e l s of the oxygen atom and molecule. 1.1 ELECTRONIC STATES AND RADIATIVE TRANSITIONS  OF ATOMIC AND MOLECULAR OXYGEN Due mainly to i t s abundance and importance i n many chemical and biochemical systems, oxygen has been the subject of extensive research. The study of i t s spectroscopy alone has been an active area of i n v e s t i g a t i o n as exemplified by the large amount of l i t e r a -ture on the subject. A d e t a i l e d review w i l l not be attempted, but a discussion of the lower e l e c t r o n i c states of oxygen, and some of the t r a n s i t i o n s i n which these are involved, w i l l be presented. 1.1.1 Atomic Oxygen Energy Levels The ground state e l e c t r o n i c configuration of atomic oxygen can be represented by 2 2 4 Is 2s 2p The energy l e v e l s of the ground state atom can be calculated using the Russell-Saunders coupling scheme [8] to f i n d the i n d i v i d u a l contributions of the 4 p-electrons to the t o t a l o r b i t a l and spin angular momentum. For the ground state oxygen atom, t h i s r e s u l t s 1 1 3 3 in 3 d i f f e r e n t energy states, D^, S Q and P2 1 0 w i t n t h e P state having three close l e v e l s of d i f f e r e n t t o t a l angular momen-- 6 -turn. Hund's Rule [9] for equivalent electrons predicts that the state of lowest energy f o r a given electron configuration i s that state having the greatest m u l t i p l i c i t y . I f more than one state has the same m u l t i p l i c i t y , then the lowest of these states i s the one with the greatest o r b i t a l angular momentum. These r u l e s predict that for 3 an oxygen atom the ground state w i l l be the ^ 1 0 s t a t e » t n e highest state the *SQ, and the ^ 2 state l y i n g somewhere between them. Observations of t r a n s i t i o n s between these states confirmed t h i s assignment, as w i l l be shown i n the next section. 1.1.2 Atomic Oxygen Transitions In general the s e l e c t i o n rules [8] f o r t r a n s i t i o n s between e l e c t r o n i c states are given by: AS = 0 AL = ±1 AJ = 0,±1 This r e s u l t s i n the t r a n s i t i o n s between the 3 low-lying states of atomic oxygen being "forbidden", i . e . : there i s a low proba-b i l i t y of a spontaneous emission from the higher states to a lower one. It has been shown [10] that such t r a n s i t i o n s , when they occur, can be att r i b u t e d to multipole as opposed to dipole allowed tran-s i t i o n s . McLennan [11] c o r r e c t l y assigned the 557.7 nm emission of the airglow to the (*Sn - '*'D7) t r a n s i t i o n i n atomic oxygen. The - 7 e l e c t r i c quadropole nature of t h i s t r a n s i t i o n was demonstrated by observations of the Zeeman s p l i t t i n g of the emitting state i n a magnetic f i e l d [12, 13]. 1 1 The 0( S) and 0( D) states are known as "metastable states" due to the forbidden r e l a x a t i o n of these states by r a d i a t i v e tran-s i t i o n s . This means that, i n the absence of other deactivation processes (such as c o l l i s i o n s with other molecules), these states may have long l i f e t i m e s r e l a t i v e to excited states that have "allowed" (and hence fast) r a d i a t i v e t r a n s i t i o n s to lower energy l e v e l s . The t r a n s i t i o n p r o b i l i t i e s of the 0(*S) and 0( 1D) states have been calculated t h e o r e t i c a l l y and measured experimentally many times and the values obtained are l i s t e d i n Tables 1.1 and 1.2. Figure 1.2 i l l u s t r a t e s the energy le v e l s of the ground state oxygen atom and i t s observed t r a n s i t i o n s . The r a d i a t i v e l i f e t i m e s shown are the most re c e n t l y published experimental values. 1.1.3 Energy Levels of Molecular Oxygen Oxygen i s one of the few molecules, with an even number of electrons, which i s paramagnetic i n the ground state. This r e s u l t s from a degenerate p a i r of H 2p o r b i t a l s which are s i n g l y occupied by electrons with p a r a l l e l spins. The order of molecular o r b i t a l s f o r 0^ was f i r s t given by TABLE 1.1 THEORETICAL AND EXPERIMENTAL RESULTS FOR THE  RADIATIVE TRANSITION PROBALITIES OF O^S), ACs" 1) O^S) A(5S7-72. O^D) + hv(557.7 nm) 0( 1 S) A 1 : 2 9 7 ,22T 0( 3P) + hv(297.2 nm) A(557.7) A(297.7) A(557.7) A(297.2) A(557.7) + A(297.2) Ref. Theoretical 2.0 0.18 11.1 2. 18 [14] 2.2 0.09 24.4 2. 29 [15] 2.04 0.067 30.4 2. 11 [16] 1.28 0.078 16.4 1. 36 [17] 1.25 0.071 17.6 1 32 [18] 1.18 [19] Experimental 1.43 ± 0.2 [20] _ 22 ± 2 - [21] 18.6 ± 3.7 - [22] 1.0 _ _ - [23] _ 1.31 ± 0.05 [24] 1.06 ± 0.32 0.045 ± 0.014 23.7 ± 2.4 1.11 ± 0 . 3 4 [25] - 9 -TABLE 1.2 THEORETICAL AND EXPERIMENTAL RESULTS FOR THE RADIATIVE TRANSITION PROBALITIES OF 0( 1D), A f s " 1 x 1Q~3) Q ( 1 D ) A^O-O^ 0C 3P 2) + hv(630.0 nm) 0 ( 1 D ) A(636.4^ 0 C 3 p ) + h v ( - 6 3 6 > 4 n m ) A(630.0) A(636.4) A(636.4) A(630.0) A(636.4) + A(636.0) Ref. Theoretical 7.5 2.5 0.33 10.0 [14] 7.8 2.6 0.33 10.4 [15] 6.9 2.2 0.32 9.1 [17] 5.69 1.85 0.325 7.54 [16] Experimental 0.33 _ [26] _ _ - 5.3 [20] _ _ 8.3 [27] _ 0.33 - [22] 5.15 (±1.25) 1.66 (±0.42) 0.32 (±0.03) 6.81 (±1.32) [25] - 10 -Figure 1.2 The energy l e v e l s of the three lowest elec-t r o n i c states of the oxygen atom, i n d i c a t i n g the wavelengths and r a d i a t i v e l i f e t i m e s for each t r a n s i t i o n . - 10a -417 E c L O L O l/> o II J Term 0 1 S 1.96 E c o ci ro ID in I? E c ro I/) O O II 2 1 D E c ao LT) cn CM Mulliken i n 1932 [28]: o Is < a Is < a 2s < o 2 s < a 2p < I 2p < II 2p < o 2p g g g u g 1 y ^ g ^ u 1 k k b a b b a a The bonding character of each o r b i t a l i s designated as "k" (inner s h e l l ) , "b" (bonding) or "a" (anti-bonding). The three 3 - 1 lowest states of molecular oxygen, designated X I , a A and b^Z + a r i s e from the electron configuration [29]: 0„{KK(a 2s)„(a 2 s ) 2 ( a 2p ) 2 ( n 2p ) 4 ( n 2p) 2} 2 g 2 u g r u 1 g Three other bond states of oxygen have been found and observed spectroscopically from between 4.06 and 4.35 eV above the ground state. These are the c^Z , c 3A and A 3 E + states which r e s u l t from u u u the electron configuration: 0„{KK(a 2 s ) 2 ( a 2 s ) 2 ( a 2p ) 2 ( n 2p ) 3 ( n 2p) 3} 2 g u ' ^ g r - ^ ^ u g A p o t e n t i a l energy diagram of the six lowest states i s shown in Figure 1.3 which has been drawn from the spectroscopic data c o l l e c t e d by Krupenie [30] . It i s important to note that a l l s i x 3 lowest states of oxygen d i s s o c i a t e to give two ground state 0( P) atoms. Above the 1st d i s s o c i a t i o n l i m i t , the only i d e n t i f i e d bound state of 0^ that has been characterised i n any d e t a i l , i s the 3 - 3 -B E^ state l y i n g at 6.14 eV above the ground state. The B E^ state d i s s o c i a t e s to give an O^D) and an 0( 3P) atom. - 12 -Figure 1.3 Potential Energy curves for the s i x lowest states of molecular oxygen. (Drawn from the spectroscopic data c o l l e c t e d by Krupenie [30]). - 12a -- 13 -1.1.4 Radiative Transitions of Molecular Oxygen 3 - 3 -With the exception of the f u l l y allowed B Z X Z v u g Schumann-Runge system, a l l other t r a n s i t i o n s i n the oxygen molecule are e l e c t r i c dipole forbidden. This i s r e f l e c t e d i n t h e i r long r a d i a t i v e l i f e t i m e s , which means that each state can be considered "metastable". 1 3 -(a) 0„(a A ««—»• X Z ) Infra-Red Atmospheric System g g. 1 3 -The a A to X Z t r a n s i t i o n was f i r s t observed i n the absorp-g g t i o n spectrum of l i q u i d oxygen by E l l i s and Kneser i n 1933 [31]. Herzberg obtained a spectrum [23] showing the f i n e structure of t h i s system, and assigned the t r a n s i t i o n s of these bands from photo-1 3 -graphs of the so l a r spectrum. The (^(a A ^ ^ X Z ) band system l i e s between 924 nm and 1580 nm with the (0,0) band at 1270 nm. Gush et a l . [33] observed the (0,0) band i n both the day and nightglows of the atmosphere, although the emission i s weaker than expected because of reabsorption by the lower atmosphere. The (0,1) band at 1580 nm has also been observed i n the emission of the night sky [34]. Badger et a l . [35] have measured the r a d i a t i v e l i f e t i m e of O^a^A ) to be 3880 seconds, which r e f l e c t s t h i s r a d i a t i v e tran-s i t i o n ' s v i o l a t i o n of the s e l e c t i o n rules f o r an e l e c t r i c dipole t r a n s i t i o n . The t r a n s i t i o n occurs by a magnetic dipole mechanism. - 14 -1 + 3 -Cb) 0 2(b Z «*—»»X Z ) Red Atmospheric System S g. Observations of these weak bands date back to the work of Wollaston i n 1802 and Fraunhofer i n 1817. The most extensive measurements of the b^E+««—•» X 3Z~ t r a n s i t i o n were made by Babcock g g and Herzberg [36] i n absorption by 0^ i n both the laboratory and atmosphere. The (0,0) and (1,0) bands at 761.9 nm and 688.2 nm were among the f i r s t molecular bands observed and were designated as the Fraunhofer A and B bands. In 1948 Meinel [37] f i r s t iden-t i f i e d the (0,1) emission band at 864.5 nm i n the airglow, t h i s and other b^I + *» X 3Z emission bands have also been observed i n g g aurora [38] and the nightglow [39]. The i n t e n s i t y of t h i s t r a n s i -1 3 -t i o n i s about 400 times greater than the a A *» X Z bands 6 g g observed in the atmosphere. (c) 0 2 ( A V J * - » » X 5Z~) Herzberg I System This e l e c t r i c dipole forbidden system was f i r s t observed as weak absorption bands by Herzberg [40]. The absorption bands l i e i n the range of 243 nm to 488 nm, with the highest (predisociat-ing) v i b r a t i o n a l l e v e l observed being v = 11. Dufay [41] suggested that part of the u l t r a - v i o l e t nightglow was due to t h i s system. Analysis of the f i n e structure of the nightglow spectrum [42, 43] confirmed Dufay's assignment. Several workers [44, 45, 46, 47] have detected the Herzberg I bands i n emissions from laboratory afterglow systems. - 15 -There have been no d i r e c t measurements of the r a d i a t i v e l i f e -times (T) of the three upper states of molecular oxygen. Estimates of T f o r the 0_(A3E+) state have been varied by several orders of 2 u 3 magnitude (<1 to 10 s) [44, 48, 49, 50]. Hasson et a l . [51] have measured the absolute absorption c o e f f i c i e n t of the (7,0) band, which implies a T of about 0.2 s. Slanger [52] has used Hasson's band strengths to obtain values of T between 0.16 and 0.25 s, depending upon the v i b r a t i o n a l l e v e l . The r a d i a t i v e decay rates maximize at v = 5 - 7 and f a l l sharply at higher v i b r a t i o n a l l e v e l s due to Franck-Condon overlap factors with the ground state. (d) 0 2(c 1 E " * - * . X3E") Herzberg II System Absorption bands of t h i s very weak and forbidden system were f i r s t observed by Herzberg i n 1953 [53]. They l i e i n the range from 254 to 700 nm (v' = 0 - 13). Degen [54] observed a weak emission i n the afterglow of an 0 2 - Ar discharge ( 0 2 ~ 3%) that 1 - 3 -he a t t r i b u t e d to the c E — ^ X E band system. u g In 1977, Lawrence et a l . [55] generated strong emission from the c^E^ (v = 0) state i n the laboratory by discharging an 02/He mixture and adding C0 2 to the gas stream. This enabled these workers to i d e n t i f y the Herzberg II system as being the strongest spectral feature i n the nightglow of Venus between 300 and 800 nm, measured by the Venera 9 and 10 probes [56]. (C0 2 being a major component of the Venus atmosphere.) - 16 -Recently, Slanger [52] reported emission from v' = 0 to v" = 6 - 13 in another He - 0 2 - CO^ discharge system. Reassessing Herzberg's e a r l i e r absorption work, Slanger has estimated a value of 25 - 50 s for the r a d i a t i v e l i f e t i m e of the c^l state. u (e) 0 2 ( C 3 A u « » - * » X 3 E " ) Herzberg III System The C 3 A state i s the least observed of the 6 lowest states u of 0,,. Herzberg reported several weak bands i n 0 2 absorption i n 1953 [53] which he at t r i b u t e d to t h i s system. Herzberg III bands have also been i d e n t i f i e d i n m a t r i x - i s o l a t i o n experiments [57, 58] and gas phase afterglows [52] of the 02~He mixtures i n the range 470 nm to 630 nm f o r v' = 0 to v" = 8 - 12. The spin components of t h i s state exhibit d i f f e r e n t r a d i a t i v e t r a n s i t i o n pro-b a b i l i t i e s , and Slanger [52] has estimated a value of x = 5 - 50 s for C 3 A (fi = 1, v = 0 - 6) and x = 10 - 1000 s for C 3 A (n = 2 , v = 6). u u ' (f) 0 2 ( C 3 A u — — a 1 A g ) Chamberlain Band System In 1958 Chamberlain [59] t e n t a t i v e l y assigned 27 weak bands i n 3 1 the airglow to the C A •» a A system, but generation of these bands i n the laboratory was not accomplished u n t i l 1979 [52] when 18 16 is o t o p i c s u b s t i t u t i o n of 0 for 0 allowed Slanger to confirm Chamberlain's assignment. Following t h i s , Slanger demonstrated that the bands were present i n laboratory spectra from as early as 1954 [44]. D i f f i c u l t i e s i n assigning these bands were i n part - 17 -3 due to only the lowest spin component of C A u r a d i a t i n g to the a^A state. The Chamberlain bands were observed for v' - 0 to v" = 3 - 9 g between 455.3 nm and 717.2 nm i n Slanger's afterglow experiments, (g) Other Oxygen Band Systems In a low pressure discharge of 0^ - He, Noxon [60] reported an emission band at 1908 nm which he i d e n t i f i e d as the Q-branch of the b*E + a*A system. However, t h i s t r a n s i t i o n i s more g g strongly forbidden than even the b 1 E + ^ X 3E~ t r a n s i t i o n and hence does not s i g n i f i c a n t l y a f f e c t the r a d i a t i v e l i f e t i m e of the b E + state. 3 - 3 -The B E •*-»» X I Schumann-Runge system i s probably the most extensively studied t r a n s i t i o n i n molecular oxygen spectra, being the only f u l l y allowed e l e c t r i c dipole t r a n s i t i o n . The r a d i a t i v e l i f e t i m e i s short enough to make c o l l i s i o n a l deactivation unimpor-tant in both the upper atmosphere and normal discharge flow systems. A f u l l discussion of t h i s state i s given by Krupenie [30]. Since t h i s state l i e s above the 0^ d i s s o c i a t i o n l i m i t and cannot be formed by the recombination of ground state oxygen atoms, t h i s state w i l l not be considered any further. Figure 1.4 i l l u s t r a t e s the major band systems of molecular oxygen that are relevant to t h i s study. The following sections w i l l expand upon the occurrence of atomic and molecular oxygen r a d i a -t i v e emissions and discuss possible sources of e x c i t a t i o n . - 18 -Figure 1.4 The major emission band systems of mole-cular oxygen. 19 -1.2 ATMOSPHERIC EMISSION OF O^S) As mentioned i n section 1.1, many of the atomic and molecular r a d i a t i v e t r a n s i t i o n s of oxygen have been i d e n t i f i e d i n spectra of the l i g h t emitted from the earth's atmosphere. This i s not sur p r i s i n g since oxygen i s one of the major species present i n the atmosphere. The following sections w i l l examine the major processes thought to give r i s e to 0(*S) emissions i n aurora, the dayglow and the nightglow. No sharp boundaries may be applied i n separating these phenomena, although they show d i s t i n c t a l t i t u d e dependencies. Many e x c i t a t i o n processes are common to a l l three atmospheric emissions to a greater or lesser extent. However, i t i s generally accepted that each one shows a d i f f e r e n t major e x c i t a t i o n process which accounts f o r most, but not a l l , of the emission. 1.2.1 Aurora Aurora are one of the most d i s t i n c t i v e types of atmospheric emissions and the only type discernable to the naked eye. They are caused by the i n t e r a c t i o n of the solar wind with the atmosphere [61]. The solar wind consists of a f l u x of high energy protons and electrons whose c o l l i s i o n s with molecules i n the atmosphere can form i o n i c species. This occurs mainly at a l t i t u d e s greater than 120 km. Well-known examples of t h i s phenomenon are the Aurora Borealis and A u s t r a l i s i n the a r c t i c and a n t a r c t i c regions respec-- 20 -t i v e l y , which are caused by the magnetic a t t r a c t i o n between the earth's ignetic poles and the charged p a r t i c l e s of the solar wind. mat One of the strongest features i n the v i s i b l e spectra of aurora i s the 557.7 nm l i n e from 0(*S). It was f i r s t detected by Angstrom [4] i n 1868 and confirmed by Struve [62] and many others. The suggested mechanisms f o r 0(*S) formation i n aurora f a l l into two major categories, d i r e c t and i n d i r e c t formation. (a) Direct Formation Mechanism Electron Impact 3 It has been suggested [63] that electron impact on the 0( P) ground state atom can excite i t to i t s higher e l e c t r o n i c states: 0( 3P) + e ~ — ^ 0 ( 1 S ) + e~ (12) A l t e r n a t i v e l y an electron could cause the d i s s o c i a t i o n of molecular oxygen with one of the r e s u l t i n g atoms excited to the 0( 1S) state [63] , i . e . : 0 2(X 3 E") + e"—»»0( 1S)'+ 0 + e~ (13) Dalgarno et a l . [64] have also suggested the d i s s o c i a t i v e recombination of 0 2 + , 0 2 + + e" •»0( 1S) + 0 (14) as a source of the 0(^S) e x c i t a t i o n . - 21 -The rates of reactions (12 - 14) depend upon the c o l l i s i o n cross section f o r each species and the electron density i n the auroral region. Z i p f et a l . [65] have found that the c o l l i s i o n cross sec-t i o n i n reaction (13) i s too small to account f o r the observed 0(*S) i n t e n s i t y i n a known electron f l u x . Reactions (12) and (14) were also considered to be inadequate mechanisms by Rees et a l . [66] who analyzed the i n - s i t u measurements of Donahue et a l . [67]. Chemical Reactions Henricksen [68] has suggested that the reaction of atomic nitrogen with n i t r i c oxide may be important i n the e x c i t a t i o n of 0(^S) i n aurora: Although t h i s reaction w i l l be endothermic f or the production of 0( 1S) i f the i n i t i a l c o l l i s i o n i s between ground state N and NO, 2 the reaction of e l e c t r o n i c a l l y excited N( D) atoms would be exo-thermic : N + NO — ^ N ( X V ) + 0( S) (15) N ( X V ) + O^S) Ionic reactions of 0 +( P) with N were suggested as a s i g n i -f i c a n t source of 0( 1S) by Torr et a l . [69]. 0 +( 2P) + N( 4S) O^S) + N +( 3P) - 22 -(b) Indirect Formation Mechanisms Electron impacts may lead to a v a r i e t y of atoms and molecules. One or more of these products may be formed i n an e l e c t r o n i c a l l y excited state. I f the r a d i a t i v e t r a n s i t i o n s from these states are s u f f i c i e n t l y forbidden, r e s u l t i n g i n the species being long l i v e d (metastable), then they may undergo c o l l i s i o n s with other atmospheric species before r a d i a t i v e l y decaying. In section 1.1, i t has been shown that almost a l l of the bound excited states of oxygen are metastable. Other atmospheric species such as N 2 form long-lived states analogous to those of 0 2- Parkinso et a l . [70] suggested that the energy trans f e r from N 2 ( A 3 E * ) may be the e x c i t a t i o n source of 0(^S) i n the aurora, i . e . : N 2 ( A V ) + 0( 3P) _-».N 2 ( x V ) + O^S) (16) Henricksen [71] estimated that 80% of the 557.7 nm emission i n t e n s i t y i n aurora could be accounted for by reaction (16). The electron impact cross sections of 0 2 leading to higher states have been found to be quite large [72]. Yau and Shepherd [73] pointed out that the Herzberg bands are not strongly enhanced i n the aurora, even though the electron impact cross sections for 0 2 3 + 3 1 -e x c i t a t i o n to the A I , C A and c E states are quite large. They u u u o ^ suggested that t h i s may be due to the process 0 O ( A 3 E + , C 3A , cV) + 0( 3P) — « * 0„(X 3E") + 0( JS) (17) 2 u u u 2 g - 23 -These workers concluded that reaction (16) was the predomi-nant production mechanism above the a l t i t u d e of peak emission (~105 km) while below t h i s a l t i t u d e reaction (17) was important. Solheim and Llewellyn [74] came to the same conclusion by a d i f f e r e n t route and t e n t a t i v e l y i d e n t i f i e d 0„(c"*E ) as the i n t e r -J 2 u mediate f o r reaction (17). 1.2.2 Dayglow The dayglow i s defined as the airglow emission at a time when sunlight i s shining on the emitting region of the atmosphere from above. S i m i l a r l y , the t w i l i g h t airglow occurs when sunlight i s shining on the emitting layer from below. A sharp d i s t i n c t i o n between the two airglows i s not convenient since some emission may ar i s e from the same e x c i t a t i o n process. In the following discussion they w i l l be treated as dayglow. This type of atmospheric emission i s the most d i f f i c u l t to detect and study because of the extremely large background of solar r a d i a t i o n scattered by the atmosphere. Observations of the day-glow require the detection of a few k i l o r a y l e i g h s of emission against a background of several thousand k i l o r a y l e i g h s . The f i r s t success-f u l study of the dayglow was made by Noxon and Goody [75] who u t i l i z e d the differ e n c e i n p o l a r i z a t i o n between the background solar r a d i a t i o n and r a d i a t i o n emitted from the atmospheric layers. Rocket borne photometers allowed measurements of the dayglow at r i g h t angles - 24 -to the zenith and f a c i l i t a t e d measurements of the a l t i t u d e dependence of the dayglow components. Wallace and McElroy [76] measured the a l t i t u d e dependence of the 557.7 nm atomic l i n e i n the dayglow and found two maxima at approximately 200 km and 100 km. Above 170 km the 557.7 nm emission p r o f i l e c l o s e l y follows that of the [77] , which suggests that d i s s o c i a t i v e recombination (reaction (14)) i s the major e x c i t a t i o n source of 0(^S) in t h i s region, with a small contribution from elec-tron impact e x c i t a t i o n (reactions (12) and (13)). Below t h i s region, the photodissociation of molecular oxygen i s thought to be important, i . e . : hv 0 (xV) »-20( 3P, V 1S) (18) 8 The d i s s o c i a t i o n threshold wavelengths required to form oxygen atoms i n several states are l i s t e d i n Table 1.3. The p r i n c i p a l uncertainty i n the importance of reaction (18) in the dayglow l i e s i n a lack of knowledge about the d e t a i l e d processes responsible for 0^ absorption below 134.4 nm. Chemical reactions that may form Of^S) w i l l be discussed i n d e t a i l i n the next section but they are thought to be equally important i n the dayglow and the nightglow. Figure 1.5 i l l u s t r a t e s the a l t i t u d e dependence of the 557.7 nm emission suggested f or various e x c i t a t i o n mechanisms together with the measured p r o f i l e 25 TABLE 1.3 DISSOCIATION THRESHOLD WAVELENGTHS IN THE PHOTOLYSIS OF 0, 0 r x V ) 2 g Product Atoms Dis s o c i a t i o n Threshold (hv) 0( 3P) + 0( 3P) 242.1 nm 0( 3P) + 0( 1D) 175.1 nm 0( 1D) + 0( 1D) 137.1 nm 0( 3P) + 0( 1S) 133.4 nm 0( 1D) + 0( 1S) 110.2 nm - 26 -Figure 1.5 The a l t i t u d e dependence of the OC'S) exc i t a -t i o n mechanisms and measured emission inten-s i t y i n the dayglow. (Reproduced from Schaeffer et a l . [77].) - 26a -- 27 -of Schaeffer et a l . [77]. 1.2.3. Nightglow Nights with an unusually large brightness over the whole sky have been recorded as e a r l y as 1788. Towards the end of the nine-teenth century, several astronomers began to recognize the exis-tence of a t e r r e s t r i a l component to the l i g h t of the night sky. Scattered s t a r l i g h t was found [78] to be i n s u f f i c i e n t to explain the increase in night sky i n t e n s i t y away from the zenith. Other e v i -dence f o r the existence of a nightime-airglow included the obser-vation that the brightness of the l i g h t of the night sky was not concentrated toward the Milky Way to the extent that would be expected based upon star counts. One of the most convincing arguments for the existence of a night-airglow was the suggestion that the 557.7 nm auroral l i n e was present at a l l times over the e n t i r e sky. Angstrom [4] had pro-posed t h i s idea i n 1868 when he was studying aurora, and Campbell [79] and others confirmed t h i s suggestion i n 1895. Over the years the nightime airglow has been r e f e r r e d to as "permanent aurora" [78], "non-polar aurora" [80] and "nightglow" [81], with the l a t t e r being the most frequently used since the 1950's In the following discussion, the nightime airglow w i l l be referred • to as the nightglow. - 28 -(a) O r i g i n of the Green Line Several suggestions were made about the o r i g i n of the green l i n e before the work of McLennan and Shrum [5] i n 1925. It had been suggested that the emission came from Krypton or an unknown li g h t gas geokoronium. Following h i s bombardment of s o l i d nitrogen with cathode rays, Vegard [82] suggested that s o l i d nitrogen dust p a r t i c l e s in the upper atmosphere may be the source of the 557.7 nm emission. He at t r i b u t e d the green luminescence observed from the s o l i d nitrogen to the auroral green l i n e . However, McLennan and Shrum [83] demonstrated that the s o l i d nitrogen luminescence was due to a set of bands at 555.6, 561.7 and 565.4 nm. Atkinson [84] showed that the temperature could not be low enough f o r s o l i d nitrogen to exist i n the upper atmosphere. In 1925, McLennan and Shrum [5] obtained the nightglow green l i n e i n an uncondensed discharge of helium with a trace of a i r . A f a i n t 557.7 nm emission was also observed i n a discharge of pure oxygen, and McLennan and Shrum c o r r e c t l y assigned the t r a n s i t i o n to an excited oxygen atom. (b) E x c i t a t i o n Mechanisms of 0(^S) i n the Nightglow In 1931, Chapman [6] suggested that the energy required to excite the ground state atom of oxygen to i t s *S l e v e l could be provided by the vast energy r e s e r v o i r formed by d i s s o c i a t e d oxygen during the daytime. Although he did not require that the reaction - 29 occur i n a termolecular c o l l i s i o n , the simple t h i r d order re a c t i o n 0( 3P) + 0( 3P) + 0( 3P) »»0( 1S) + 0 2 (1) has come to be c a l l e d the Chapman reaction . This mechanism was widely accepted as being responsible f o r the green l i n e nightglow f o r the following 30 years, during which time many measurements of the green l i n e i n t e n s i t y and atomic oxygen density were made. The rate constant for reaction (1) was measured i n 1960 by Young and Clark [85]. They produced oxygen atoms by the addition of NO to a stream of nitrogen atoms: N + N O — • » 0 + N 2 (19) The 557.7 nm emission was observed and the rate constant c a l -- 36 6 -1 culated to be < 10 cm s . This was considered to be too small to explain the nightglow emission of approximately 200 ra y l e i g h + 12 -3 using an upper bound of 2.5 x 10 cm f o r atomic oxygen density [86, 87]. It was also found that the i n t e n s i t y at 557.7 nm did not vary with the t h i r d power of the atomic oxygen concentration, as would be expected from re a c t i o n (1). Barth [88] found that when oxygen atoms were added to an active nitrogen stream the green l i n e was observed. Addition of NO to t h i s mixture resulted i n the 557.7 nm emission going through a maximum and then decreasing r a p i d l y as the nitrogen atom density approached zero (and hence atomic oxygen density approached a maximum - 30 -from reaction (19)). This was interpreted as the reaction N(4S) + N(4S) + 0 ( 3 P ) — ^ 0 ( 1 S ) + N 2 (20) being much more efficient than the Chapman mechanism (reaction (1)) at producing 0( 1S) in the laboratory. Although no inferences were drawn for the nightglow, the atomic nitrogen concentration is too low in the upper atmosphere for this process to explain the night-glow intensity of the 557.7 nm line. Shortly after this experiment, Barth and Hildebrandt [89] dis-charged pure 0 2 and.did not observe the 557.7 nm emission in the afterglow. From the sensitivity of their detection system, they placed an upper limit of 8 x 10 cm s for the rate constant of reaction (1), even smaller than Young and Clark's value. In 1961, Barth [7, 90] suggested that 0( 1S) is excited via the two step process: 0( 3P) + 0( 3P) + M — ^ 02* + M (2) followed by 0 * + 0( 3P) * O^S) + 0 (3) * where M refers to a third body and 0 2 is an excited oxygen molecule having enough energy to raise a ground state atom to the *S state. The identity of 0 2 w i l l be discussed in later sections, but the debate over the 0( 1S) formation mechanism in the nightglow was far - 31 -from over. In 1966 Young and Black [50] re-measured k^ (the rate constant for reaction (1)) and obtained a value of 1.5 x 10 3 4 cm^s * which was interpreted as being almost large enough to explain the night-glow 557.7 nm emission with OC^S) being formed by the Chapman mech-anism. In 1969 Gadsenand Marovich [91] interpreted the rocket measurements of Packer [1] as being consistent with a mixture of both the Barth and Chapman mechanisms. The analysis of the data obtained i n the experiments before 1966 [50] had only considered r a d i a t i v e decay as the loss process for the 0(*S) atom. C o l l i s i o n a l d e activation was assumed negligable, p a r t l y because the e a r l i e r experiments of K v i f t e and Vegard [92] had yielded very small rate constants for the quenching of 0(*S) by 0 2 and Ne. Young and Black [50] were the f i r s t to measure these rate constants and apply the values obtained i n a determination of k^. These new quenching constants indicated that below 110 km, where the density of the atmosphere i s appreciable, c o l l i s i o n a l deactivation (or quenching) must be considered. (c) C o l l i s i o n a l Deactivation of 0(^S) Due to the uncertainty of the production mechanism of 0(*S) from atomic oxygen recombination, most of the quenching rate cons-tants f o r 0(*S) have been obtained from experiments i n which O^S) was formed d i r e c t l y by photodissociation. The usual method - 32 -employed for d i r e c t formation of Of^S) has been ei t h e r the u l t r a -v i o l e t photolysis of N 20 [93, 94, 95, 96, 97]: N 20 + hv (< 130 nm) ^ N 2 ( A 3 E * ) + o( 1S) (21) or C0 2 [98, 99, 100]: C0 2 + hv (< 128.6 nm) ••COtX 1!) + 0( 1S) (22) The rate of decay of the 557.7 nm signal following the photo-l y s i s was measured in the presence and absence of any p a r t i c u l a r species. Many quenching rate constants have been measured. Table 1.4 shows the values and ranges of quenching rate constants obtained by various workers. Of major importance to t h i s study are the quenching rate constants of the atmospheric constituents at the a l t i t u d e s of the 557.7 nm emitting layer. Young and Black [102] reported that oxygen atoms were the major quenching species of 0(^S) i n the upper atmosphere. The rate constant for the quenching of 0(^S) by atomic oxygen 0( 1S) + 0( 3P) - 7-*. quenched products (7) i s of c r i t i c a l importance to the processes involved i n both the upper atmosphere and the laboratory. The value of k^ reported by -13 3 -1 Young and Black [102] of 1.8 x 10 cm s was subsequently r a i s e d -12 3 -1 by Felder and Young [95] to 7.5 x 10 cm s and f i n a l l y , i n 1976, Slanger and Black [92] reported an ever larger k^ of 1.8 x 10 ^ cm 3s Because laboratory measurements of the 0(^S) forma-33 -TABLE 1.4 QUENCHING RATE CONSTANTS OF O^S) Quenching Species 0 CO N 20 NO Rate Constant Reference 3 -1 @300k(cm s ) 1.8 x IO' 1 3 [101] 7.5 x l O " ! 2 [95] 1.8 x IO' 1 1 [93] 3.6 x I0']l [99] 2 2.8 x IO'} 3 [101] 2.2 x I O - 1 3 [94] 3.6 x IO"! 3 [99] 2 4.1 x I O - 1 3 [101] ,u 1.1 x I0~]l [99] 9.3 x l O " 1 ^ [97] 8.0 x io'::" [99] , 5.0 x io"r!r [ I O O ] H O 1 7.0 x I O " 1 ; . [99] 0 5.8 x 10":" [98] A r 3.9 x 1 0 " J ; [99] < 5 x 10',' [101] 2.0 x 10',, [99] He 2 x 10 'JS [99] < 5 x i o " 1 7 [ioi] 1. Chemical Reactions £ Physical Quenching 2. Possible Reaction 34 tion rate constant were always determined as k^/k^, the effect of this increasing value of k^ was to increase the magnitude of k^ for the Chapman mechanism, as indicated in Table 1.5. As the apparent magnitude of k^ i n i t i a l l y increased, the Chapman mechanism appeared to be in agreement with the observed 557.7 nm intensity from the in-situ atomic density profile measure-ments. But as the reported value continued to increase, Slanger and Black [104] calculated that the Chapman mechanism would yield peak 557.7 nm intensities of 1000 - 2300 rayleighs using a value for — 3 2 6 —1 of 5.4 x 10 cm s @ 200k and the atomic oxygen density pro-f i l e s of Scholtz and Offermann [105] and Dickinson et a l . [2]. These calculated intensities are an order of magnitude greater than the typical range of 75 - 200 rayleigh measured in the night-glow. Aeronomers have used both the Chapman and Barth mechanisms in attempting to model the nightglow, but there has been no conclu-sive evidence to dismiss or support either. Donahue et a l . [106], in 1972, used the formation rate constant obtained by Felder and Young [95] in an attempt to match their 557.7 nm nightglow altitude profile with the atomic oxygen concentration. Donahue et a l . found the matching unsatisfactory i f they assumed either the Chapman or Barth mechanisms. In 1977 Slanger and Black [107] showed that the Chapman mech-anism requires the rate constant k^ to be of the order of 1.5 x 10 35 TABLE 1.5 REPORTED RATE CONSTANTS FOR THE EXCITATION OF 0( 1S) Year k or k ^ / k g Reference 'cmV1 @ 300k) 1960 < 10" 3 6 [85] 1961 < 8 x 10" 3 8 [7] 1966 1.5 x 10" 3 4 [50] 1972 6.3 x 10" 3 3 [95] 1976 1.7 x 10" 3 2 [93] 1.55 x 10" 3 1 [104] * dominant quenching of precursor assumed to be by M - 36 -exp (+ 3200/RT) cm 6s _ 1 (or 3.2 x I O - 3 4 cn^s" 1 @ 300K) to explain the -30 nightglow. This value i s very d i f f e r e n t from the value of 1.4 x 10 exp (-1300/RT) cm 6s _ 1 (or 1.55 x 10~ 3 1 cm 6s _ 1 @ 300K) measured by the same workers i n the laboratory. Hence, either t h e i r rate constant i s wrong or the Chapman mechanism i s i n c o r r e c t . In order to evaluate the p o s s i b i l i t y of a Barth mechanism, i t i s * necessary to consider possible i d e n t i t i e s for the 0^ metastable intermediary. (d) Oxygen Metastables i n the Nightglow For the energy transfer process: 0( 3P) + 0 2* •» O^S) + 0 2 (3) * to be e n e r g e t i c a l l y v i a b l e , the 0 2 must l i e at least 4.17 eV above 3 - 1 the ground state. The three low l y i n g states of 0 2 (X E , a A , b^E +) would require the trans f e r of large amounts of v i b r a t i o n a l 3 1 energy into e l e c t r o n i c energy i n order to excite 0( P) to 0( S) . The minimum v i b r a t i o n a l levels that would be required are v = 27 for 0 o(X 3E"), v = 20 for 0-(a 1A ) and v = 17 for 0 o ( b 1 E + ) . Even under 2' g 2' gJ 2 V g the low pressure conditions of the upper atmosphere ( i . e . infrequent c o l l i s i o n s with other molecules), i t i s u n l i k e l y that such v i b r a t i o n a l l y excited species would be present i n s u f f i c i e n t quantities to excite O^S). Nor has such multiquanta v i b r a t i o n a l to e l e c t r o n i c energy transfer even been observed. On the other hand, both 0_(A 3E +) and 0-,(C3A ) have s u f f i c i e n t 2 g 2 u - 37 -energy to excite 0( 1S) even in t h e i r lowest v i b r a t i o n a l l e v e l s . Any l e v e l of ^ ^ ^ i P a ^ o v e v = 1 n a s s u f f i c i e n t energy f or pro-cess (3) . The u l t r a v i o l e t nightglow [41, 42], which i s predominantly composed of the Herzberg I bands, indicates the presence of the 02(A3E*) state i n the nightglow. Measurements [1] of the Herzberg I height p r o f i l e have shown that the maximum of the emission layer i s located at 95 km, at approximately the same a l t i t u d e as the maximum atomic oxygen density [2] . Barth [90] a t t r i b u t e d the formation of the 02(A3£*) state to the reaction 0( 3P) + 0( 3P) + M—»-02(AV) + M (23) which i s very s i m i l a r to the f i r s t step i n his mechanism to explain the 0(^S) e x c i t a t i o n , and Barth hence proposed ^ ( A 3 ^ * ) as the pre-cursor. This suggestion was supported by Vlasov [108] who estimated the density p r o f i l e of ^ (A3!^ ) ^ N T*IE UPPER atmosphere using the Barth Mechanism i n h i s c a l c u l a t i o n . The Herzberg I bands have been observed in the laboratory i n discharge flow systems of oxygen atoms [48, 50, 109] and the inten-s i t y of the emission was found to vary as the second power of the atomic oxygen concentration: I (Herzberg I) = k Q b s [ 0 ] 2 - 38 --21 3 -1 -20 3 -1 where k , = 2.5 x 10 cm s i n N„ and 0 o and 2.5 x 10 cm s obs 2 2 in Ar and He (a pressure dependence was also observed and a t t r i b u t e d to the reduced quenching of Ar and He [48]). Because of the l i m i t e d pressure range of a conventional discharge flow system, i t i s d i f f i -c u l t to d i f f e r e n t i a t e between the e f f e c t of an added gas on the formation step (23) and quenching step": 0 o ( A 3 Z + ) + M—•»0_ + M (24) 2 u 2 Harteck and co-workers [110, 111] have observed the formation of 0 2(A 3E*) on a n i c k e l surface due to the recombination of oxygen atoms. This method allows the separation of r e a c t i o n (24) from the formation reaction (23) i n the measurement of quenching rates. Kenner and Ogryzlo [112] employed t h i s method i n measurements of the deactivation of 0 o ( A 3 £ + ) by 0„, 0 and Ar. Table 1.6 l i s t s the 2 u 2 published measurements of the quenching rate constants obtained by various workers for the ^ ^ " ^ u ^ m e t a s 1 : a f r l e state of oxygen. The 0 2(C A ) state i s observable i n the nightglow as the Chamberlain Bands [59]. In the laboratory, emission from t h i s state has been generated by laser e x c i t a t i o n of 0^ i n rare gas and nitrogen matrices [113, 114]. Slanger [52] produced emission from t h i s state i n an He - 0 2 mixture discharge system and has shown [115] that previously observed (but u n i d e n t i f i e d ) bands i n the 300 -3 1 500 nm region of 0 2 afterglow spectra belong to the 0 2(C A^—•*> a A ) 3 1 -system. No k i n e t i c data for e i t h e r the 0„(C A ) or 0„(c Z ) have J 2 u 2 u - 39 -TABLE 1.6 RATE CONSTANTS FOR THE QUENCHING OF 02CAV.) in : 0o(AV) + M ^ « - 0 9 + M (24) 2 u ^ k 24 M T(K) k24 Reference r 3 - i \ cm s J N2 300 -13 3.3 x 10 1 - 3 [108] N20 300 -12 4.7 x 10 [108] No 200 -13 8.3 x 10 [107] 2 0 200 ~5 x 10" 1 2 [107] N2 200 -12* =1.5 x 10 [91] ° 9 300 3 x 10" 1 3 [109] 2 N 300 <6 x 10"1 4 [109] co2 300 kco2 > ko2 [109] 0 300 9 x 10" 1 2 [112] ° 9 300 -13 2.9 x 10 [112] I Ar 300 8.6 x 10" 1 6 [112] * Derived values from the nightglow intensity profi les - 40 -been reported, although they are expected to be formed i n a s i m i l a r mechanism to reaction (23), i . e . : 0 + 0 + M — » » 0 „ ( C 3 A , c V ) + M (25) 2^ u u 1 - 3 -The Herzberg II band system ( 0 2 ( c T.^—*» X E )) was found to be the predominant emission i n the nightglow of Venus [55], but was not i d e n t i f i e d i n the t e r r e s t r i a l nightglow u n t i l 1980, when Slanger and Huestis [116] matched synthetic spectra incorporating three oxygen systems with the nightglow spectra of Broadfoot and Kendall [117]. Recently, Llewellyn et a l . [118] have proposed O^c^Z ) as the precursor to 0(*S) a f t e r considering the apparent absence of the Herzberg I bands i n aurora and the experimental r e s u l t s of Kenner et a l . [119] concerning the r a t i o of Herzberg I to Herzberg II emission i n t e n s i t i e s under various conditions i n a discharge flow system. * 1 The i d e n t i t y of the 0 2 intermediary, i f the 0( S) i s excited by a Barth Mechanism, remains speculative. In order to equate laboratory studies with atmospheric systems, i t i s necessary to have a deta i l e d knowledge of the conditions i n the upper atmosphere, including the density p r o f i l e s for each constituent. The following section w i l l describe the atmospheric measurements pertinent to t h i s study. (e) Atmospheric Measurements of 0(^S) and 0 Many methods have been used to determine the a l t i t u d e p r o f i l e of - 41 -the 0( 1S - 1D) emission. These include the Van Rhijn [120] method and t r i a n g u l a t i o n [121, 122] of patchy areas of the 557.7 nm emission i n the night sky. The ground-based methods have indicated varying a l t i t u d e s f or the peak emission i n t e n s i t y because of the circum-stances described below. The use of rocket-borne photometers [1, 123, 124] has yielded a f a i r l y well-defined emission maximum at 97 ± 2 km with an emitting layer approximately 30 km thick (Figure 1.6). Reed and Chandra [3] have used continuous s a t e l l i t e monitoring of the 557.7 nm l i n e over several months to show that there are considerable f l u c t u a -tions of the emission (of over a factor of 10) depending upon both season and l a t i t u d e . Apart from these global v a r i a t i o n s , there i s also a patchy nature to the emission layer within several square miles. The wide range of values reported for the emission inten-s i t y and a l t i t u d e dependence are not s u r p r i s i n g under these circum-stances . One of the main goals of aeronomers i s to use the 557.7 nm emission to continuously monitor the atomic oxygen concentration i n the atmosphere. Atomic oxygen p r o f i l e s such as the one shown in Figure 1.7 (obtained by Dickinson et a l . [2]) have been mea-sured by a v a r i e t y of i n - s i t u methods. These include the oxida-t i o n of s i l v e r f o i l [125], mass spectrometry [126] and resonance-fluorescence [2, 127]. In view of the wide v a r i a t i o n of 0 and 0(*S) with time, i t i s c l e a r that measurements of these species - 42 -Figure 1.6 Rocket-borne photometer measurement of the a l t i t u d e p r o f i l e of the 557.7 nm emission i n t e n s i t y i n the nightglow on 6 November 1959 at 00.25 M.S.T., White Sands, New Mexico. (Reproduced from Packer [1].) - 4 2 a -- 43 -Figure 1.7 The a l t i t u d e p r o f i l e of the atomic oxygen density measured with rocket-borne resonance lamps by Dickinson et a l . [2], (The error bar indicates the calculated accuracy of the measurement at 110 km.) - 43a -[ 0 ] (atoms cm"3) - 44 -should be made simultaneously. This has not always been the case. In 1979 Thomas et a l . [127] observed the 557.7 nm emission, Herzberg I bands and atomic oxygen concentration using two rockets launched within 30 minutes of each other. The a l t i t u d e p r o f i l e s obtained i n the 80 - .115 km region are shown i n Figure 1.8. A thorough analysis of t h e i r data, using the rate constants of Slanger and Black [93, 104] f o r both the formation rate constant ( k p and 3 OC P) quenching rate (k,,), showed that the emission p r o f i l e was incompatible with a Chapman Mechanism unless the rate constant k^ for formation was k = 8.0 x 10" 3 7 exp (2800/RT) cm 6s _ 1. This rate constant d i f f e r s considerably from Slanger's value: k 2 k 3 -30 ft i K or (- 1r—) = 1.4 x 10 exp (-1300/RT) cm s 6 i n both i t s temperature dependence and the si z e of the pre-exponential f a c t o r . Thomas et a l . [127] concluded that t h e i r observations were not inconsistent with the Barth Mechanism for 0(*S) formation, possibly involving 02(A 3I*) as the intermediate. However, they pointed out that t h e i r derived expression for k^, which could explain the observed 557.7 nm p r o f i l e using the Chapman Mechanism, was c r i t i c a l l y dependent upon the magnitude of 1 3 the quenching rate constant of 0( S) by 0C P) i n reaction (7). This quenching constant i s extremely important i n any nightglow - 45 -Figure 1.8 Simultaneous a l t i t u d e p r o f i l e measurements of the 557.7 nm, Herzberg I emissions and atomic oxygen density p r o f i l e according to Thomas et a l . [127]. (Error bars indicated at 4 km i n t e r v a l s . ) - 45a -0( P) c o n c e n t r a t i o n , cm"-3 10I i £ f _ 1— i - 46 mechanism and w i l l play a cen t r a l r o l e i n t h i s i n v e s t i g a t i o n . Although the Barth Mechanism appears to be the most favoured e x c i t a t i o n i n the past 10 years, the Chapman Mechanism was resurrected recently by P e t t i t d i d i e r and Teitelbaum [128] i n the study of the mean diurnal v a r i a t i o n of the green l i n e emission with respect to atmospheric t i d e s . P e t t i t d i d i e r and Teitelbaum found that the Barth Mechanism was not consistent with the semi-diurnal t i d e when the rate constants used are those proposed by Slanger and Black [93, 104] . J * * * * * * In order to study the 0(*S) emission i n the laboratory, we need to be able to generate oxygen atoms, measure t h e i r concen-t r a t i o n and observe the 557.7 nm emission. The following section w i l l describe laboratory processes and procedures that have been previously used i n studies of the recombination of atomic species. 1.3 LABORATORY STUDIES OF OXYGEN ATOM RECOMBINATION In an attempt to simulate atmospheric reactions i n the labora-tory, one must take into consideration the atmospheric conditions at the a l t i t u d e s where most of the chemical reactions occur. At 100 km the pressure i s about 10 t o r r , which r e s u l t s i n a mean free path of approximately 6 cm for p a r t i c l e s at t h i s a l t i t u d e . Whereas the chemosphere behaves l i k e a huge wa l l - l e s s reaction vessel, at pressures of 10 t o r r or l e s s , i n observation vessels of fe a s i b l e s i z e , wall c o l l i s i o n s become important. Higher pressure - 47 -are used i n the laboratory because the working range of the d i s -charge flow system l i e s between 0.3 and 30 t o r r . In the atmosphere, the average l i f e t i m e of atoms that recombine i s of the order of days, but at pressures greater than 1 t o r r t h i s time i s reduced to seconds or l e s s . The increased recombination rate at these pres-sures allows these reactions to be studied. However extrapola-t i o n of the r e s u l t s obtained i n the laboratory to the atmosphere can be uncertain, since the mechanism may be d i f f e r e n t i n these two s i t u a t i o n s . The three major pr e r e q u i s i t e s f or the study of the luminescence from atomic oxygen recombination i n the laboratory are (1) a method of generating atomic oxygen, (2) a method of measuring the concen-t r a t i o n of atomic oxygen and (3) a method of measuring the l i g h t emitted. 1.31 Generation of Atomic Oxygen The d i s s o c i a t i o n of molecular gases i n t o t h e i r component atoms has been attempted by several d i f f e r e n t methods with varying degrees of success. Thermal d i s s o c i a t i o n on a heated tungsten wire was found to y i e l d up to 1% d i s s o c i a t i o n of molecular [129]. Lundell et a l . [130] have reported up to 2% decomposition of ^ 0 and 0^ upon passage of these gases through a Nernst glower. The high tempera-ture decomposition of 0^ has also been reported to be a "clean" - 48 -source of 0( P) atoms [131, 132]. Photochemical decomposition of either oxygen or an oxygen containing molecule i n the gas phase requires a high f l u x of u l t r a -v i o l e t r a d i a t i o n which i s d i f f i c u l t to generate i n the laboratory. Kistiakowski [133] has reported up to 1% d i s s o c i a t i o n of 0 2 following i r r a d i a t i o n at 186 nm and 172 nm. Electrode discharge methods have been used s u c c e s s f u l l y with H 2 and N 2 [134, 135], but due to the corrosive e f f e c t of 0 2 on the metal electrodes at high temperatures, the atomic oxygen concentra-tions obtained were found to deteriorate with time. The electrodeless discharge has been found to be the most con-venient method of producing oxygen (and many other) atoms i n a flow system. Radiofrequency ( t y p i c a l l y ~ 20 MHz, <_ 100 IV) or microwave ( t y p i c a l l y 2450 MHz, 25 - 200 W) r a d i a t i o n can be con-nected to a discharge tube v i a external c o i l s or c a v i t i e s [136, 137]. Both types of electrodeless discharge are e f f i c i e n t i n atomic oxygen production,but the microwave discharge has been favoured due to i t s ease of operation [137] . (a) Di s s o c i a t i o n E f f i c i e n c i e s of Electrodeless Discharges The degree of d i s s o c i a t i o n of molecular gases that are passed through microwave discharges depends upon such factors as the power of the discharge, flow rate and pressure of the gas stream. Also, - 49 -the condition of the walls of the discharge tube and the impurity l e v e l of the gas play important rol e s i n determining the degree of d i s s o c i a t i o n [138, 139]. For example, i t has been found that u l t r a -pure gases do not d i s s o c i a t e appreciably [140, 141]. Unfortunately, the gas stream leaving the microwave discharge i s not comprised of ground state atoms and parent molecules alone. In discharges of N 2 or ^ - i n e r t gas mixtures, the main active species produced i s N( 4S) although appreciable concentrations of N( 2D) (~ 10%), N( 2P) (- 4%), and N 2(A"V) have been detected near the discharge plasma [142, 143]. These species are expected to be destroyed within a few milliseconds, but v i b r a t i o n a l l y excited N 2 remains an active species present i n the flow system [144] . In 0 2 or 0 2 - Ar discharges s i g n i f i c a n t amounts of O^a^A^) and, to a lesser extent, O-Cb^E"1") have been i d e n t i f i e d [145]. However, there i s some evidence that metastable 0 2 i s not important when very d i l u t e mixtures of highly p u r i f i e d 0 2 i n Ar are discharged [146]. (b) Heterogeneous Recombination of Atoms As mentioned i n the introduction to t h i s section, wall reac-tions can be important when using a rea c t i o n vessel of f i n i t e s i z e . The e f f i c i e n c y of a surface i n the recombination of atomic oxygen i s extremely dependent upon the nature of that surface. The wall reaction rate constant k i n : w 0 + w a l l — w ^ i / 2 0 2 + wall [26) - 50 -i s d i r e c t l y proportional to the surface recombination c o e f f i c i e n t y [147], which i s the f r a c t i o n of the wall c o l l i s i o n s which r e s u l t i n recombination. A p a r t i a l l i s t of measured recombination c o e f f i -cients f o r various surfaces i s given i n Table 1.7. In p a r t i c u l a r , oxyacids such as H^PO^, H^SO^, andHClO^ coated on glass surfaces, have been found to be very e f f e c t i v e i n the reduction of surface recombination of atomic oxygen [147, 148]. Fluorocarbon coatings such as Teflon [154] and "Fluoro-Kote" [155] have been found to be p a r t i c u l a r l y e f f e c t i v e i n the reduction of halogen-atom recombination. In view of Harteck et a l . ' s observation [110, 111] that excited states of 0^ can form i n some surface reactions, i t i s clear that the e f f e c t of surfaces w i l l have to receive c a r e f u l attention i n the present study. 1.3.2 Measurements of the Atomic Oxygen Concentration Many techniques have been used to measure atomic concentrations. A thorough review of these methods w i l l not be attempted, but the major techniques used i n the determination of atomic oxygen con-centrations w i l l be discussed, p a r t i c u l a r l y as they apply to the methods used i n t h i s i n v e s t i g a t i o n . The methods of measuring atomic oxygen concentration f a l l into two major categories, those that are non-specific to oxygen atoms and those that are s p e c i f i c . The Wrede-Harteck gauge and c a t a l y t i c - 51 -TABLE 1.7 SURFACE RECOMBINATION COEFFICIENT y OF  OXYGEN ATOMS AT 20°C SURFACE. Y REFERENCE Metals: Mg 2. ,6 X i o ' 3 [149] Au 5. ,2 X i o " 3 I! Ni 2. .8 X i o " 2 It Fe 3, .6 X i o " 2 tl Cu 1, .7 X i o " 1 tl Ag 2 .4 X i o " 1 11 Halides : KCl 7.8 X i o ' 4 1! NaCl 9.2 X i o " 4 1! KF 9.4 X i o " 4 H KBr 1.3 X i o " 3 II L i C l 1.9 X i o " 3 II Oxides: Pyrex 1.2 x I O - 4 [141][149] 3.1 - 4.5 x 10' 5 [150] 2 x IO" 5 [151] S i l i c a 7.1 x 10~ 4 [149] 1.6 x 10" 4 [150] H 3P0 4 coated Pyrex: I O - 6 - I O - 7 [152] 3 x 10" 6 (0 2/Ar) [153] 5 x IO" 5 ( 0 2 ) [153] - 52 -probe are examples of non-specific detection methods, (a) Wrede-Harteck Gauge The p r i n c i p l e behind the Wrede-Harteck gauge i s the pressure dif f e r e n c e between the discharged gas (containing the atomic species to be measured) and an enclosed volume connected to the former by one or several small holes or s l i t s (also f r i t t e d d i s c s ) . Both molecules and atoms d i f f u s e i n t o the closed volume where the atoms recombine due to an e f f i c i e n t c a t a l y s t (usually s i l v e r ) and hence only molecules d i f f u s e out of the enclosed volume. The resultant pressure increase i n the enclosed space i s given by Ap = a p ( l - 1/2 VI ) where p i s the pressure of the discharged gas and a = f r a c t i o n of atomic species. This method, developed by Wrede [156] and Harteck [157] has several advantages: 1. There i s no interference by excited species. 2. Inert gases contribute equally to the mass flow through both sides of the o r i f i c e and hence,do not a f f e c t Ap. The disadvantages include: 1. This method requires the accurate measurement of small pressure - 53 -differences, although the development of accurate capacitance manometers has lessened t h i s problem. 2. Steady-state conditions are d i f f i c u l t to achieve within a reasonable amount of time, depending upon the pressure of the flow system [147]. The technique has not been used i n any recent studies of atom reactions. (b) C a t a l y t i c Probe Technique The r i s e i n temperature of a metal surface due to the recombina-t i o n of hydrogen atoms was f i r s t measured by Bonhoeffer [158], although T o l l e f s o n and Leroy [159] were the f i r s t to use t h i s e f f e c t q u a n t i t a t i v e l y . The technique was f i r s t applied to oxygen atoms by Ogryzlo [160], and i t was one of the techniques used in the present study. The main disadvantage of t h i s method i s the n o n - s p e c i f i c i t y of the probe. Excited species present in the gas stream can recombine on the surface and cause a temperature r i s e i n addition to that due to atomic oxygen. This e f f e c t was demonstrated by E l i a s et a l . [162] by comparison of atomic concentrations measured by the i s o -thermal probe and Wrede-Harteck gauge. An estimate of [0] that was 10% higher than the value obtained by a Wrede-Harteck gauge and 25% higher than that from a chemical t i t r a t i o n was obtained with 54 the c a t a l y t i c probe. It has been reported [147] that the d i s c r e -pancy between probe measurements and t i t r a t i o n techniques disappears in the absence of metastable oxygen molecules (reaction (19)), however, data supporting t h i s claim have not been published. * * * * * The p r i n c i p a l detection methods that have been used i n atomic oxygen studies and that are s p e c i f i c to atomic oxygen are o p t i c a l spectrophotometry, mass spectrometry, electron spin resonance and chemiluminescence. In the following sections each technique w i l l be discussed, but only the chemiluminescence detection method w i l l be reviewed i n d e t a i l because of i t s importance to t h i s i n v e s t i g a t i o n . (c) Optical Spectrophotometry in c l u d i n g Resonance  Fluorescence and Absorption Although o p t i c a l spectrophotometry i s extremely s e l e c t i v e i n the detection of atomic species, t h i s method has severe disadvan-tages i n a discharge flow system. The major experimental require-ments are a su i t a b l e l i g h t source and detection equipment. The lowest energy allowed t r a n s i t i o n of atomic oxygen i s i n the vacuum u l t r a - v i o l e t at 130.2 nm. The t y p i c a l short path lengths of discharge flow systems and low absorption cross section of atoms - 55 -present d i f f i c u l t i e s i n the choice of a l i g h t source. Multiple t r a v e r s a l s across the flow tube have been used to increase the path length but the number of t r a v e r s a l s i n the u l t r a v i o l e t wavelength region seldom exceeds 12 before alignment becomes d i f f i c u l t and the reflectance e f f i c i e n c y of the mirror surfaces reduces the i n t e n s i t y of the r a d i a t i o n . The development of resonance l i n e sources has g r e a t l y increased the s e n s i t i v i t y of o p t i c a l absorption, due to close matching of the source l i n e o p t i c a l p r o f i l e and absorption l i n e p r o f i l e . Con-venient lamps emitting resonance r a d i a t i o n are discharges through helium or argon with a trace of 0^. Both resonance absorption [163] and fluorescence [164] have been used to study the reactions of atomic oxygen. (d) Mass Spectrometry Mass spectrometric detection requires that a representative sample of the gas stream be removed from the reaction zone for analysis. However, due to the low operating pressure of the mass spectrometer (~ 10 3 torr) only small amounts of gas can be sampled, u s u a l l y by a small o r i f i c e or "leak" connecting the discharge flow system to the i o n i z a t i o n chamber of the mass spectrometer. Very short distances are required between the leak and i o n i z a t i o n chamber or the atoms are l o s t on the chamber walls. The oxygen atoms would be detected at m/z = 16. Unfortunately, there i s normally a large - 56 -contribution to t h i s peak i n the mass spectrum due to the fragmen-t a t i o n of molecular oxygen. Kle i n and Herron [165] have measured 0 atoms by mass spectrometry in t h e i r studies of the reactions of 0 with NO and N0 2 > (e) Electron Paramagnetic Resonance Electron Paramagnetic resonance i s a powerful technique for the detection of atoms and r a d i c a l s . E.P.R. has the advantage of being able to follow several species, although t h i s i s somewhat o f f s e t by the poor s e n s i t i v i t y of t h i s technique. The lower l i m i t s of detection are at least one to two orders of magnitude poorer than those for resonance fluorescence. Cupitt and Glass [166] 12 -3 have found that the detection l i m i t for [0] was ~ 1 x 10 cm i n t h e i r E.P.R. experiment. Absolute values for [0] are determined 3 _ by c a l i b r a t i o n of the E.P.R. instrument with 0~(X E ) under 2 g i d e n t i c a l c a v i t y conditions[167] . (f) Chemiluminescence Techniques The atom detecting techniques of resonance fluorescence, mass spectrometry and often absorption spectroscopy do not provide absolute atom concentrations and therefore must be c a l i b r a t e d by another method. The most commonly used method of c a l i b r a t i o n i n the case of oxygen atoms i s a "chemical t i t r a t i o n " using the " a i r -afterglow emission". - 57 -The chemiluminescent reaction: 0 + N O — ^ N 0 2 + hv (397.5 - 1400 nm) (28) occurs when NO i s added to a stream of 0 atoms. The normalized spectrum of t h i s emission i s shown i n Figure 1.9. Kaufman [151, 168] has shown that the i n t e n s i t y of t h i s chemiluminescence i s pro-p o r t i o n a l to the concentration of 0 and NO, I = k 2 g[0][NO] (29) independent of any other species present. It i s also pressure independent i n the range 1 - 1 0 t o r r [169], due to c a n c e l l a t i o n of pressure dependent terms. The NO concentration i n the flow stream remains constant because any N0 2 formed i n reaction (28) or the fa s t e r termolecular recombination: 0 + NO + M—»• N0 2 + M (30) i s r a p i d l y destroyed by the very fast step 0 + N0 2 »«N0 + 0 2 (31) which regenerates NO. There are several methods of using these reactions to determine * [0]. I f N0 2 i s added to the gas stream, the i n t e n s i t y of the N0 2 continuum increases as long as 0 atoms are present i n excess. The maximum i n t e n s i t y i s reached when - 58 -Figure 1.9 Normalized spectrum of the NO-0 chemilumi-nescence obtained by several workers. (Reproduced from Sutoh et a l . [178].) W A V E L E N G T H - p m - 59 -[N02] = l/2[0] and decreases as more N0 2 i s added u n t i l i t i s sharply extinguished when the N0 2 flow rate i s equal to the o r i g i n a l concentration of atomic oxygen. The N0 2 flow rate may be determined by several methods, namely by measuring (i) the pressure drop of N0 2 (in equilibrium with N,^) in a known volume, ( i i ) the weight loss of N0 2 generated from N 20^ i n an ice bath, or ( i i i ) by the use of ca l i b r a t e d flowmeters, with the back pressure kept constant by immersing the N0 2/N 20^ source i n a constant temperature bath. Because of i t s lower b o i l i n g point, NO i s more convenient to measure. One method that was used i n t h i s i n v e s t i g a t i o n was that of Reeves et a l . [170] who, a f t e r obtaining the maximum afterglow emission using N0 2 > shut o f f the N0 2 flow (without measuring i t ) and added NO u n t i l the same afterglow i n t e n s i t y was achieved. At t h i s point the NO flow rate i s equal to 1/4 of the o r i g i n a l 0 atom flow and may be conveniently measured by the pressure drop of NO i n a known volume. A v a r i a t i o n of t h i s technique has been discussed by Clyne et a l . [171]. In the expression - 60 -I o b s = K e f £ [0][NO] (32) * where I , = the observed i n t e n s i t y of the afterglow from N0„ , obs  J 6 2 ' [0] may be obtained by adding known amounts of NO to the gas stream i f i s known. K may be obtained by adding small, known increments of NO to a stream of atomic nitrogen produced i n a discharge of N 2 or an N 2 - Ar mixture. The chemiluminescence observed before the end point of t h i s " t i t r a t i o n " i s dominated by the NO a-, g- and y- bands (known as the " n i t r i c oxide afterglow") * emitted from NO [48] formed i n : N + 0 + M »-N0* + M (33) The i n t e n s i t y of the NO bands decreases as the end point i s approached, at which point the amount of NO added C[N0]encj p 0i n-j) i s equal to the o r i g i n a l atomic nitrogen concentration, C[N]Q) and the chemiluminescence vanishes completely. A f t e r the end point ( i . e . : with an excess of NO) the nitrogen atom concentration i s zero and the atomic oxygen concentration ([0]) i s equal to the o r i g i n a l nitrogen atom concentration ( [ N ] q ) . The chemiluminescence i n the flow system i s then dominated by the a i r afterglow of reaction (28). K ~_ i s then obtained from the rate of increase of t h i s a i r e f f afterglow i n t e n s i t y as further NO i s added. Once K £ £ i s deter-mined, the atom ( i . e . , oxygen) concentration can be evaluated by measuring the i n t e n s i t y of the a i r afterglow when a small known amount of NO i s added to the gas stream. Wavelength s e l e c t i o n f o r the monitoring of the a i r afterglow i s not u s u a l l y necessary [172] unless other chemiluminescent reactions i n the gas stream are pre-sent as interferences to the measurement of I , . The detection obs 11 -3 l i m i t s of t h i s method can be as low as 1 x 10 oxygen atoms cm 1.3.3 Detection of Metastable Atoms and Molecules  i n the Gas Phase Metastable atoms and molecules are u s u a l l y detected by t h e i r forbidden r a d i a t i v e t r a n s i t i o n s to lower states. In t h i s i n v e s t i -gation, the major species of i n t e r e s t i s the Of^S) atom. As shown i n section 1.1.2, t h i s atom has a forbidden t r a n s i t i o n to Of^D) which r e s u l t s i n the emission of 557.7 nm r a d i a t i o n . Observations of the i n t e n s i t y of t h i s emission, together with a knowledge of the r a d i a t i v e l i f e t i m e of the state allows the monitoring of the O^S) concentration from [0( 1S)] = T 1(557.7 nm) (34) where x = ( t r a n s i t i o n p r o b a b i l i t y of Of^S) ^Of^D)) * or approxi-mately the r a d i a t i v e lifetime,and 1(557.7 nm) = absolute emission -3 -1 i n t e n s i t y ({photons} cm s ). The major d i f f i c u l t i e s with these measurements are i n (i) obtaining separation of the 0(*S) emission at 557.7 nm from other chemiluminescent emissions i n t h i s wavelength region and - 62 -( i i ) measuring the absolute emission i n t e n s i t y i n the gas phase With respect to ( i ) , the d i f f e r e n t i a t i o n of l i g h t of one wave-length from another i s normally accomplished by the use of e i t h e r monochromators or interference f i l t e r s . Because the r a d i a t i o n i s very weak, most investigations of O^S) emission at 557.7 nm have been ca r r i e d out using interference f i l t e r s [93, 95, 96, 99, 101] which can transmit more r a d i a t i o n than a monochromator. T y p i c a l l y the f i l t e r has a F.W.H.M. ( f u l l band width of h a l f the maximum transmission) of 2 nm, centered near the atomic l i n e , e.g. 557.3 nm [93]. The disadvantage of t h i s method i s the i n a b i l i t y of such a f i l t e r to discriminate between the desired r a d i a t i o n and other radia-tions at that wavelength. In many discharge flow systems, the " a i r -afterglow" due to impurities i n the discharged gases i s the major source of background r a d i a t i o n and l i e s i n the region of i n t e r e s t . Bingham et a l . [103] are the only workers who have reported the study of 0(^S) emission using a monochromator (Czerny-Turner) to i s o l a t e the 557.7 nm atomic l i n e (in t h e i r measurements of O^S) quenching by H^O). The other metastable species of i n t e r e s t i n t h i s i n v e s t i -gation i s O^ia^hg)• This metastable oxygen molecule has been detected using the isothermal probe [162] described i n section 1.3.2. However the n o n - s p e c i f i c i t y of the probe, e s p e c i a l l y i n the presence of atomic oxygen, precludes i t s use as a s e n s i t i v e - 63 -monitor of [0-(a^A )] i n t h i s i n v e s t i g a t i o n . CCa^A ) concentrations have been determined using both con-ventional [173] and photoionization mass spectrometers [174]. Bader and Ogryzlo [175] found two unique emission bands i n a stream of excited oxygen molecules at 634 nm and 703 nm. Notic-ing that the energy of the 634 nm band i s equivalent to twice the e x c i t a t i o n energy of C-fa^A ) above i t s ground state, they proposed that the bands were a r e s u l t of the following processes: 20 2(a 1A g)«i5=!!S0 4*^ »• 20 2(X 3Zg) + hv (634 nm) (35) and 20„(a 1A )^=^0,* *»0 o(X 3I~) + 0 o(X 3E~) n + hv (703 nm) (36) 2 V gJ 4 2K g v=l 2^ g'v=0 Although the i n t e n s i t i e s of these dimol emissions are more convenient to observe than the 1.27 ymband, they depend upon the square of the [O^a^A^)] and the emission i n t e n s i t y disappears very quickly at low O^a^A^) concentrations. The 1.27 ym emission from O - C a 1 * ) n * « 0 „ ( x V ) n + hv (1.27 um) (37) 2K g v=0 2 g v=0 occurs i n a d i f f i c u l t spectroscopic region where se n s i t i v e detectors (such as the i n t r i n s i c germanium detector) are very expensive. How-ever, t h i s emission has been used to monitor [0 9(a^A^)] [176]. - 64 -1.3.4 Absolute Intensity Measurements The normal procedure for measuring the absolute i n t e n s i t y of a r a d i a t i n g body i s to c a l i b r a t e the detection equipment using a stan-dard l i g h t source. However, because of the d i f f u s e nature of gas phase chemiluminescence, a d i f f e r e n t c a l i b r a t i o n method i s r e q u i r e d / Fo n t i j n , Meyer and S c h i f f [177] suggested that the l i g h t emission from the reaction of atomic oxygen with n i t r i c oxide could be used as a standard f o r the c a l i b r a t i o n of gas phase emissions. This r e a c t i o n (equation (28)) has already been described i n d e t a i l i n section 1.3.2. * As long as the NO^ emission and the uncalibrated emission are measured i n the same vessel with the same detection system, a l l geometric and s e n s i t i v i t y factors i n the c a l i b r a t i o n w i l l cancel out. This method consists of adding a known amount of NO to a stream of oxygen atoms (which have been previously t i t r a t e d to mea-sure t h e i r concentration). The chemiluminescence emitted from t h i s i n t e r a c t i o n i s then measured i n the range of 0 . 4 — ^ 1 . 4 ym at wavelengths or areas of wavelengths that contain the uncalibrated emission. In reaction (28) k s NO + 0 — » • N0 2 + hv (28) the absolute i n t e n s i t y of the emission i s equal to: - 65 -1.4 um / I (X)d A = k g[NO] [0] (38) 0.4 um or f o r a p a r t i c u l a r wavelength A^ I (X ) = k s[N0][0]-A f- L- (39) s where F (A,) i s the spe c t r a l flux of the emission at A, and T s 1 r I s the t o t a l f l u x , i . e . : T = / F (A)dA (40) A=0 where T g and F g(A) have the same but a r b i t r a r y u n i t s . S i m i l a r l y , for an emission from an uncalibrated reaction k U ^ p r o d u c t s + hv (A.^  »• A^) the emission i n t e n s i t y w i l l be given by I u(A)dA = k u[u] (41) I f we define k g f o r the region of emission i n reaction (28) from A^ to A^ as A2 • k s ; A F s ( A ) d A k = s A l 5 (42) S T s and observe both emissions i n the same vessel ( i . e . : r a d i a t i n g volume), then we can r e l a t e equations (38) to (41) by d i v i s i o n to give: - 66 -= M M / 2 I (X)dX f . (43) k [u] X u v / X / 'x'VX)DX The absolute emission i n t e n s i t y i s not required as long as the same detection system i s used f o r both emissions. i An observed emission I from the NO-0 glow i n reaction (28) i s r e l a t e d to the absolute emission I by the expression VV = G(VVV (44) where g ( X p i s the r e l a t i v e s e n s i t i v i t y of the detection system as a function of X. S i m i l a r l y , f o r the emission from U using the same detection system VV = GCVVV (45) therefore I'Cx.) i ' ( x . ) u 1 _ s*- V , T T x T " T~Tx~T ( 4 6 ) u 1 s 1 Substituting equation (46) into (43) we obtain the expression k u fNOUO] A 2 V ^ V ^ / X2 V= ™ X1 V™ / X . V ^ (47) i i The [NO], [0] and r a t i o I ( X)/I s ( X ) can be obtained i n the laboratory and k g and I s Q 0 may be calculated using the wavelength dependent rate constants f o r reaction (28) that have been measured - 67 -by several workers [177, 178, 179, 180, 181, 182]. Hence k y may be obtained i f [U] i s known or vice-versa. It can be seen from Figure 1.10 that the quantum y i e l d of the NO-0 glow i n the i n f r a red region (> 0.9' ym) i s not as well defined as i n the v i s i b l e region. At the wavelength of i n t e r e s t f or [O^a^A )] determination (1.27 ym), the rate constant of Sutoh et a l . [178] i s e s s e n t i a l l y the same as that of Woolsey et a l . [182] but somewhat larger than the value reported in the e a r l i e r work by F o n t i j n et a l . [177]. Vanpee et a l . [179] and Golde et a l . [180] obtained a rate constant several times larger than the others f o r the 1.27 ym region. The difference may be p a r t l y explained by contributions from the chemiluminescent reaction: NO + 0 3 *-N0* + 0 2 (48) Clough and Thrush have reported [183] that 7% of r e a c t i o n * (48) y i e l d s e l e c t r o n i c a l l y excited N0 2 from which the continuum emission emanates. Vanpee et a l . and Golde et a l . mixed NO with discharged 0 2, which may contain 0^ whereas Sutoh et a l . and Woolsey et a l . obtained the NO-0 glow from the t i t r a t i o n of nitrogen atoms with NO, which diminishes the possible presence of ozone. As the exact value of the quantum e f f i c i e n c y of reaction (28) i s not accurately known i n the region of ~ 1.27 ym, the O^a^A^) concentrations deduced from t h i s emission may be inaccurate by as - 68 -Figure 1.10 Absolute rate constants for the chemilumi-nescent reaction: NO + 0 — N 0 2 + hv as a function of wavelength. (Reproduced from Sutoh et a l . [178].) ( x 1 0 " 2 0 c m 3 s - W 1 ) 2 0 , 1 1 V A N P E E etd.11 79] W A V E L E N G T H ( j jm) - 69 -much as a factor of 2. The next chapter will describe in detail the experimental methods used in this investigation. - 70 -CHAPTER 2: EXPERIMENTAL 2.1 MATERIALS The gases used in this investigation were obtained from the Canada Liquid Air and Matheson companies. Nitrogen was the most objectionable impurity in any gas that was passed through the discharge because i t results in the formation of NO which reacts with oxygen atoms to produce an interfering green emission from reaction (28) . Various attempts were made to remove nitrogen from the argon carrier gas, since this constituted the greatest source of N 2 in the discharge. In the most successful purification pro-cedure, argon was passed through a quartz tube containing titanium metal maintained at approximately 950°C. The metal was used in the form of "sponge" or strips which were obtained from the Alfa Division of the Ventron Corporation. The strips were made by cutting 0.25 mm thick sheets of Ti into 2 x 30 cm pieces, whereas the "sponge" (4 and 40 mesh size) was used without modification. Both metal forms had a manufacturer's specified purity of 99.7 - 99.8%. * By monitoring the N02 emission while varying the heater current, the optimum temperature of the furnace surrounding the quartz tube was determined. A furnace temperature (measured by an iron-constantan thermocouple) of 950 ± 25°C was found to give the lowest N02 intensity in the flow system. Although both the sponge and strips of Ti appeared to have similar efficiencies, the strips were considered to be safer in the event of an accidental air leak while - 71 -the furnace was hot. While using the Ti sponge, the furnace was destroyed, when an airleak occurred, due to a vigorous exothermic reaction between the atmospheric oxygen and T i , resulting in the sintering of both the quartz tube and quartz liner of the furnace. Canada Liquid Air argon (< 40 ppm N2) and Matheson Ultra High Purity (< 10 ppm N2) argon were found to give similar after-glow intensities after passage through the furnace and the less expensive C.L.A. argon was used for the major part of this study. Further passage of the argon through a liquid N 2 trap was found to have no additional effect on minimizing the intensity of the back-ground, and i t s use was discontinued. Removal of trace amounts of N 2 from 0 2 is d i f f i c u l t due to both the inertness of the N 2 and the similarity between the physical properties of the two gases, e.g.: boiling points - N,, -195.8°C, 0 2 - 183.0°C. In the i n i t i a l part of this study "zero-gas" oxygen, with a typical analysis of 99.65% 0 2 and 0.05% N 2, was used. Matheson ultra high purity oxygen (typical analysis 99.986% 0 2, 25 ppm N2) was used in later experiments in an attempt to reduce the afterglow background. The U.H.P. 0 2 was found to give only a slight reduction of the afterglow background. An attempt to decrease the N 2 content of the 0 2 even more was made by trapping the 0 2 with liquid N2, d i s t i l l i n g off 80% and then using the remainder. The resultant intensities of the background were too variable and the procedure was abandoned. - 72 -N i t r i c Oxide (99% minimum purity) was obtained from the Matheson Company and c o l l e c t e d i n a c a l i b r a t e d 12 l i t r e bulb following trap to trap d i s t i l l a t i o n to remove the main impurities of N0 2 and H 20. The i n i t i a l and f i n a l f i f t h s of the d i s t i l l a t e were discarded following condensation of the NO i n l i q u i d N 2 cooled traps. Nitrogen Dioxide (99.9% min. purity) was obtained from the Matheson Company with the main impurity being H 20. The N0 2 was trapped with a dry ice/acetone slush bath, maintained at high vacuum to remove non-condensable gases and the f i n a l f i f t h of the d i s -t i l l a t e was discarded to remove water. It was stored i n a 500 ml bulb as the l i q u i d , and the vapour above the l i q u i d was i n t r o -duced into the flow system when required. Since the N0 2 was not measured q u a n t i t a t i v e l y , the N0 2/N 20 4 equilibrium did not a f f e c t the r e s u l t s . A l l other gases were used d i r e c t l y from the cyl i n d e r without further p u r i f i c a t i o n . In the quenching experiments, H 20 was the most undesirable impurity, but the use of cold traps on the de l i v e r y l i n e s did not appear to a f f e c t the r e s u l t s and t h e i r use was d i s -continued. 2.1.1 Gas Handling and Storage The gas cyl i n d e r s of argon and oxygen were connected to the flow system with e i t h e r copper or s t a i n l e s s s t e e l tubing of 1/4" - 73 -diameter. Connections to and i n t h i s tubing were made with Swage-R R lok or Cajon U l t r a - t o r r f i t t i n g s . A constant "over-pressure" of approximately 5 p . s . i . was maintained i n the d e l i v e r y system using Matheson two-stage High-purity regulators with s t a i n l e s s -steel diaphrams. The d e l i v e r y system was tested frequently f o r possible leaks by evacuation and the use of a commercial soap solution on the pressurized l i n e . A l l gas d e l i v e r y l i n e s were evacuated through the flow system and purged with t h e i r respec-t i v e gases before each set of experiments to minimize and s t a b i -l i z e the afterglow background i n t e n s i t y detected by the photon counting system. The NO and NO^ were stored i n bulbs connected to a separate vacuum system. The glass tubing connecting the bulbs to the flow system could be evacuated i n between the use of each gas to pre-vent contamination. When not in use t h i s glass tubing was main-tained under high vacuum to minimize the absorption of the nitrogen oxides by Apiezon N grease i n the stopcocks and j o i n t s . Frequent re-greasing of t h i s system was s t i l l required due to the degenera-t i o n of Apiezon N i n the presence of NO^. 2.2 APPARATUS The apparatus was designed to obtain a steady state oxygen atom concentration i n an observation bulb. Using a fast rotary pump, oxygen atoms were produced, by passing 0 ^ ( d i l u t e d by Ar) - 74 through a microwave discharge, and flowed into an observation c e l l i n which the excited states present were observed by t h e i r emissi on spectra. The major components of the apparatus are i l l u s t r a t e d somewhat schematically i n Figure 2.1 and w i l l be described i n d e t a i l below. 2.2.1 Flow System The major part of the flow system consisted of Pyrex tubing connecting the discharge tube, observation c e l l , rotary pump and manometers. Connections between sections of glassware were made using Ace Glass 0-ring j o i n t s of 7, 15, 25 and 40 mm i . d . Metal-glass connections were made with e i t h e r Kovar seals or 1/4" Cajon U l t r a - t o r r 0-ring f i t t i n g s which were found to be extremely convenient for taking apart and re-assembling the system. The vacuum pump was connected to the flow system with a short length of 65 mm o.d. heavy wall tubing, which helped prevent v i b r a t i o n s from the pump being transferred to the glassware. A Sargent-Welch two-stage Duo-seal rotary vacuum pump with a pumping speed of 500 l i t r e s min * was used throughout t h i s study. Small c o n s t r i c t i o n s i n the pump i n l e t were used to " t h r o t t l e down" the pump to produce stable pressures i n the 1 to 10 t o r r range without the consumption of excessive amounts of gas. The amounts of 0 2 and Ar entering the flow system were con-t r o l l e d by Edwards f i n e - c o n t r o l needle valves. Due to the corro sive - 75 -Figure 2.1 Schematic of the apparatus used in t h i s i n v e s t i g a t i o n . - 75a -<: IPbS iCell RECORDER LOCK-IN AMf? McPHERSON 0.3M' Monochrom-ator STB Microwave Discharge RECORDER PHOTON COUNTER > SPEX 1M onochrom-ator P.M.T. ISOTHERMAL PROBE VACUUM PUMP - 76 -e f f e c t o f on the metal components of these valves, a glass and t e f l o n valve was used to control the NO^ flow. In the l a t t e r stages of t h i s work, control of extremely small flows of NO was made possible by the use of a Nupro S-type needle valve with micro-meter handle. The discharge region was constructed of 11 mm diameter quartz tubing to withstand the high temperature of the discharge. Graded seals connected the quartz discharge area to the Pyrex tubing. Approximately 12 cm below the discharge region were two l i g h t traps set at 90° to each other to stop l i g h t from the d i s -charge from entering the observation c e l l . Also, the discharge tube and l i g h t traps were sprayed with matt-black paint and covered with black c l o t h to keep extraneous l i g h t to a minimum. Entrance A (shown i n Figure 2.1) was situated approximately 14 cm from the observation c e l l and used to add quenching gases to the flow stream. By adding part of the 0^ flow through A, the r a t i o of 0 2 before and a f t e r the discharge could be c o n t r o l l e d . Entrance B was connected to the storage system containing the NO and N0 2 bulbs. A compro-mise was required i n the placement of entrance B because i t needed to be far enough away from the observation c e l l to ensure adequate mixing of the added NO or N0 2 with the atom stream and yet not f a r enough away to give unacceptable atomic oxygen decay due to reactions with NO or NO,,. In an attempt to achieve good mixing, the gas i n l e t s A and B were constructed as perforated Pyrex b a l l s - 77 -approximately 7mm in diameter and attached to the entrance by ring seals. Entrance B was situated approximately 4 cm from the entrance to the observation cell. Several different observation cells were used in the course of this investigation: (1) A pyrex tube, 28 mm diameter and 15 cm length, which was viewed along its major axis. (2) A similar design to (1) but encapsulated in a jacket through which a liquid could be passed to control the temperature of the gas stream. In this case, the view-ing window was separated from the cell by a sealed eva-cuated space to prevent condensation on the window when the liquid was below room temperature. (3) A 2-litre bulb which was viewed across its diameter and encased in aluminum foil to increase the light collec-tion efficiency. (4) Several 5-litre bulbs of various shapes. The shape of this bulb was changed from spherical to elliptical in an attempt to obtain a homogeneous afterglow in the bulb when NO was admitted through entrance B. A spherical bulb with 25 mm diameter entrance and exit holes was found to produce a streaming of the N0 2 chemiluminescence, espe-- 78 -c i a l l y when large flows of NO were admitted to the gas stream. The eventual observation c e l l i l l u s t r a t e d i n Figure 2.1 had 40 mm diameter entrance and e x i t holes and was smoothly tapered to i t s o r i g i n a l diameter. This shape, combined with the small amounts of NO required i n the second method of t i t r a t i o n r esulted i n a v i s i b l e homogeneity of the a i r afterglow i n the observation c e l l . The i n t e r n a l surfaces of the discharge tube and the tubing connecting i t to the observation c e l l were "poisoned" to reduce atom recombination on the walls. F i r s t the i n t e r n a l surfaces were thoroughly cleaned using hot NaOH and then rinsed with d i s t i l l e d water and inspected to ensure that the surface "wetted" uniformly. A 5% sol u t i o n of phosphoric acid was then poured through the tubing and i t was allowed to drain before the apparatus was re-assembled and evacuated. Excess F^O was removed by passing argon through the system for several hours. At one stage i n the i n v e s t i g a t i o n , the surfaces were coated with a teflon-based lubricant Fluoroglide i n an attempt to increase the atomic oxygen concentration i n the observation c e l l . The atomic oxygen concentrations were not found to be s i g n i f i c a n t l y changed using t h i s surface, and due to the d i f f i c u l t y of applying a smooth, even coat of t h i s l u b r i c a n t , i t s use was abandoned. 2.2.2 Generation of Atomic Oxygen The i n i t i a l experiments were performed using a Raytheon 2450 MHz - 79 (25 - 100VV) microwave generator, but t h i s was replaced with a more powerful E.M.I. Microtron 200 (50 - 20CW) unit i n an attempt to increase the range of atomic oxygen concentrations obtainable for a given Ar - 0^ mixture. The microwave power was transmitted to the discharge region by means of a cable terminating i n a tunable 1/4 wave c a v i t y that has been described by Broida and co-workers [137]. I n i t i a t i o n of the discharge was accomplished by means of a spark from a Tesla c o i l , and the discharge could be operated i n the pressure range of 1 - 20 t o r r i n the flow system. The power supplied by the discharge was u s u a l l y varied using the adjustment on the microwave generator, but f o r very small atom concentrations the c a v i t y could be "de-tuned" f o r short periods of time. Compressed a i r was used to cool the cavity and the quartz tubing. 2.2.3 Flow and Pressure Measurement The pressure of the flow system was measured at port C on Figure 2.1. In the i n i t i a l stages of the i n v e s t i g a t i o n , a Texas Instruments P r e c i s i o n Pressure Gauge was used to monitor the pressure in the flow system and the pressure drop i n the NO storage bulb for t i t r a t i o n s or absolute emission i n t e n s i t y measurements. The pressure gauge incorporated a Bourbon quartz s p i r a l capsule with a range of 0 - 760 t o r r and r e s o l u t i o n of 0.008 t o r r . In the l a t t e r stages of the i n v e s t i g a t i o n , the Bourbon gauge was replaced with two Baratron capacitance manometers. The t o t a l system pressure was measured with a "0 - 100 t o r r absolute" mano-meter and a "0 - 10 t o r r d i f f e r e n t i a l " manometer was incorporated - 80 -into the NO storage system. Each manometer provided a 0-10 V analog signal i n proportion to the pressure being measured. The "absolute model" was connected to a d i g i t a l voltmeter which displayed the pressure d i r e c t l y . In the measurements of small pressure drops i n the NO bulb, the signal from the " d i f f e r e n t i a l manometer" was ampli-f i e d and recorded on a s t r i p chart recorder. The resultant slope of the l i n e on the recorder was proportional (and calibrated) to the pressure drop. The s t a b i l i t y of the NO flow was indicated by the l i n e a r i t y of the l i n e . The flow rates of "non-condensable" gases were monitored with Matheson 150 mm b a l l flowmeters. C a l i b r a t i o n of these flowmeters was accomplished by c o l l e c t i n g the gas throughput of the vacuum pump at several flowmeter readings and measuring the time taken to displace a known volume of water i n inverted volumetric f l a s k s . The gas was passed through the water for approximately one h a l f hour p r i o r to the c a l i b r a t i o n so that the pump o i l and water became saturated with the gas. The volume c o l l e c t e d was corrected for the vapour pressure of water, atmospheric pressure and temperature. The flow rate of argon was found to be d i r e c t l y proportional to the pressure of argon i n the system i n the range of 1 - 8 t o r r as i l l u s t r a t e d i n Figure 2.2. However, caution was required when adding other gases to the flow stream because the addition of a si m i l a r amount of a d i f f e r e n t gas could produce a d i f f e r e n t pressure increase i f the molecular weights of the two gases are very d i f f e r e n t . - 81 -Figure 2.2 The dependence of the flow r a t e on the t o t a l pressure of argon i n the discharge flow, system. - 81a -- 82 -Hence the r i s e i n pressure of the system may not simply be equal to the p a r t i a l pressure of the added gas. The concentration of added gas [X] was therefore calculated from the equation r v , „ 9.65 x 1 0 ^ -3 [ X ] = X P T X TOO C m ( 4 9 ) where f = flow rate of X from flowmeter x £^ - t o t a l flow rate = flow rate of Ar (from o r i g i n a l system pressure) + flow rate of X = f i n a l system pressure i n t o r r T = temperature i n k e l v i n . When the added gas was oxygen (M.Wt. = 32) i t was found that the p a r t i a l pressure of 0^ calculated from equation (48) was essen-t i a l l y the same as the pressure r i s e i n the flow system (Ar -M.Wt.= 40). Since the Baratron manometer was much more se n s i t i v e to changes i n the amount of added gas than the flow meter, the pressure increase i n the system was used to c a l c u l a t e [0^] using equation (49). 2.2.4 Optical Detection System The observation c e l l s had two synthetic sapphire windows with diameters of about one inch and a thickness of 0.06 inches. To detect v i s i b l e r a d i a t i o n ,a Spex 1 meter scanning spectrometer of - 83 -Czerny-Turner configuration was positioned with i t s entrance s l i t s centred on one of the sapphire windows. The d i f f r a c t i o n grating of the spectrometer was blazed at 750 nm and ruled with 600 lines/mm. The detector at the exit s l i t of the monochromator consisted of a E.M.I. 9783A photomultiplier tube housed i n a Brookdeal 5032 detector head. Signals from t h i s detector were processed by a Brookdeal 9511 Quantum Photometer and displayed on a s t r i p chart recorder. The r e l a t i v e spectral response of the spectrometer and detection system from 420 nm to 600 nm i s shown i n Figure 2.3. Although the e f f i c i e n c y of the detection system i s not maximized i n the wavelength region of the 557.7 nm emission, the high reso-l u t i o n of the spectrometer and s e n s i t i v i t y of the photon counting equipment made the s e n s i t i v i t y of the detection system at 557.7 nm quite acceptable. The i n t e n s i t y of the 557.7 nm emission was recorded on a Hewlett-Packard s t r i p chart recorder by repeated scans from 556.0 to 558.5 nm and the heights of these peaks above the background afterglow were measured and averaged to obtain each "observation point". A 0.3 m GCA/McPherson 218 spectrometer was situated on the opposite side of the observation c e l l to the v i s i b l e detection system in order to observe the i n f r a red emission at 1.27 ym. This emission was monitored i n second order using a 3.0 ym blazed d i f f r a c t i o n grating ruled with 300 lines/mm. A 1/2" sapphire lens, positioned just a f t e r the e x i t s l i t of the I.R. spectrometer, focused the - 84 -gure 2.3 Spectral Response of the Vis ible detection system. W A V E L E N G T H - nm - 85 -emission onto a 1 x 4 mm lead sulphide detector encased i n a dewar allowing i t to be cooled to -78°C with a dry ice-acetone mixture. The c r i t i c a l p o s i t i o n i n g of the PbS c e l l was accomplished using an X-Y-Z t r a n s l a t o r . Because of the high background current inherent i n the use of a PbS c e l l , the l i g h t entering the monochromator was chopped at 200 Hz by a Bulova tuning fork. A Princeton Applied Research 5101 Lock-in amp l i f i e r removed the background current from the modulated signal which was then displayed on a s t r i p chart recorder. A t y p i c a l spectrum of the 1.27 ym emission from 0 2(a 1 A ^ ) i s reproduced i n Figure 2.4. In the experiments involving 0 2(a 1 A g ) detection, i t was found that monitoring of the 1.27 ym peak inten-s i t y was preferable to the continuous scanning of the emission spec-trum because of the amount of noise associated with t h i s weak si g n a l . Before or during the experiment, the base l i n e i n t e n s i t y was monitored at both 1.24 ym and 1.30 ym and subtracted from the peak i n t e n s i t y . This background i n t e n s i t y was found to be indepen-dent of [0], [^(a ^ A )] and the pressure of the system. 2.3 ATOMIC OXYGEN MEASUREMENT Determination of the atomic oxygen concentration proved to be the most troublesome aspect of t h i s i n v e s t i g a t i o n . Because of t h i s , several d i f f e r e n t techniques were t r i e d . The theory of the - 86 -Figure 2.4 The observed spectrum of the 1.27 um emission from 0„(a 1A ) 2 g v=0 - 86a -1 1 1 1 r 0 2 ( 1 A g ) -K ) 2 feg ) ,0O Emission J I I I I L 1.21 123 125 127 129 131 133 WAVELENGTH - jum - 87 -" c h e m i c a l - t i t r a t i o n " method (and i t s vari a t i o n s ) and isothermal probe technique have been described i n section 1.3.2 and t h e i r adaption to t h i s i n v e s t i g a t i o n w i l l be described below. 2.3.1 Method 1 -- Chemical T i t r a t i o n In the i n i t i a l stages of t h i s work, the method described by Reeves et a l . [170] was adopted. A f t e r the i n t e n s i t y of the 557.7 nm emission was recorded for any "observation point", the Spex * monochromator was set at 555 nm or 559 nm where the N0 2 emission could be measured independently. N0 2 was then added to the oxygen atom stream u n t i l the maximum emission of the "afterglow" was observed on the chart recorder. This was repeated by adding more N0 2 and then reducing i t slowly to get back to a consistent maximum in t e n s i t y . Care was required to avoid an erroneous maximum inten-s i t y due to sudden "bursts" of N0 2 into the gas stream. The N0 2 bulb was then i s o l a t e d and the storage l i n e s evacuated before NO was connected to the flow system. NO was then added u n t i l the emission i n t e n s i t y observed by the photomultiplier was the same as the maximum obtained with N0 2. Although the NO flow was easier to control than the N0 2 flow, i n experiments with small atomic oxygen concentrations the NO was d i l u t e d with Ar to increase both the control of the gas flow and the s i z e of the measured pressure drop on the bulb. The NO flow rate was calculated from the equation: - 88 -where An = flow rate of NO (moles/sec) AP = rate of pressure drop of NO system (torr/sec) V = volume of NO system (l i tres) R = Gas constant = (62.4 moles ^ 1 ^ torr * K) T = Temperature in kelvin . The concentration of NO was then calculated using equation (49) and was related to the atomic oxygen concentration as: [0] = 4 [NO] The major d i f f i c u l t i e s with this procedure were: (a) An excessive length of time was required to complete one t i t r a t i o n because of the exchange of the delivery system between N02 and NO and the wait for the residual after-glow intensity to decrease enough to continue observations of the 557.7 nm emission. (b) Extremely fine control of the N02 flow rate was required to reproduce and maintain the maximum afterglow emission intensity. At one stage double fine needle valves were used to control the N02 flow but this procedure was s t i l l found to be d i f f i c u l t . (c) Although the Ar di lut ion allowed small NO flow rates to be - 89 -measured the pressure increase in the system was greater. If the pressure increase was large, the conditions in the discharge could change and the "measured atomic oxygen concentration" could be different from that in the absence of NO. In search of a more effective atom measuring technique, the isothermal calorimetric probe was investigated. Even though i t is d i f f i c u l t to use this technique to obtain absolute atom concentra-tions, i t was hoped that a calibration would be possible using the above tit r a t i o n procedure. 2.3.2 Method 2 -- Isothermal Calorimetric Detection The catalytic probe used in this investigation consisted of a helically wound spiral of platinum wire electrochemically plated with silver. It can be seen from Table 1.7 that silver i s the most active catalytic surface for atomic oxygen. When exposed to oxygen atoms, the surface of the silver becomes blackened due to the formation of silver peroxide [161]. This new surface retains a high catalytic efficiency without further oxidation of the sur-face. It was placed in the flow system just below the observation c e l l as shown in Figure 2.1. The current to operate the probe isothermally was provided by a commercial 6 volt battery which was continually charged between experiments. The probe forms one arm of the Wheatstone Bridge/ - 90 -galvanometer c i r c u i t drawn i n Figure 2.5. Before the microwave discharge was started, the bridge was balanced by adjusting the current and i Q w a s measured. As the temperature of the probe rose due to atomic oxygen recombination, the bridge went out of balance (indicated by the d e f l e c t i o n of the galvanometer). By reducing the current through the c o i l , the bridge was rebalanced to obtain i . The current passing through the detector was measured as the voltage across the standard in r e s i s t o r shown in Figure 2.5. In order to use the equation which r e l a t e s the atom flow to probe current [160] : .2 .2 0 flow = ^ ^ o ^ ^ moles/sec 4.18 x AH x 10 where R = resistance of the probe (fi) AH = heat of recombination of atomic oxygen = 58.6 Kcals/mol the probe must be 100% e f f i c i e n t i n recombining the atoms present in the gas stream. This requires a very t i g h t l y c o i l e d wire (with no c o i l s touching) or the use of two or more detectors i n s e r i e s . However, f o r the i n v e s t i g a t i o n of the atom dependency of the 0(*S) emission, only r e l a t i v e values of the atomic oxygen concentration were required (under any p a r t i c u l a r set of conditions). In a t y p i c a l atomic oxygen dependence 557.7 nm emission inten-s i t y and the probe resistance were measured at various microwave - 91 -Figure 2.5 Wheatstone bridge c i r c u i t of the isothermal c a l o r i m e t r i c probe. - 91a -- 92 -discharge powers keeping a l l other parameters constant. The power level was increased and decreased randomly to minimize any hysteresis e f f e c t s . At the f i n a l "experimental p o i n t " the oxygen flow was t i t -rated in order to convert the r e l a t i v e atom concentrations into absolute values for that set of points, i . e . : where i s obtained from the t i t r a t i o n . It was found that a f t e r a t i t r a t i o n was completed, i t took approximately 15 minutes for the system to recover to i t s o r i g i n a l background afterglow i n t e n s i t y and isothermal probe balance point. This was most l i k e l y due to residual NO i n the flow system and t i t r a t i o n i n l e t . Reverse • evacuation through the i n l e t shortened the "recovery" time but did not i r r a d i c a t e the problem. Because of t h i s , the t i t r a t i o n was only performed at the f i n a l point of the experiment and not before and a f t e r as would have been desirable. The greatest disadvantage of the isothermal probe i s i t s non-s e l e c t i v i t y . To test for the p o s s i b i l i t y of other excited species heating the probe, a series of experiments was performed. F i r s t the c a t a l y t i c probe was used to determine the apparent oxygen atom concentration,and then the atoms were removed by flowing N0 2 into the gas stream u n t i l the afterglow i n t e n s i t y was extinguished. At t h i s point a l l of the oxygen atoms were removed by r e a c t i o n (31) 0 flow (absolute) = ^ ( i 2 - i 2 ) (51) (31) - 93 -The Wheatstone Bridge was then rebalanced to obtain the current (i^Q ). When the discharge was turned o f f , the current required to balance the bridge was higher ( i ), i n d i c a t i n g another species was being deactivated by the probe. Turning o f f the Nf^ flow at t h i s point did not unbalance the bridge, i n d i c a t i n g that the flow of past the probe did not a f f e c t i t s thermal or c a t a l y t i c proper-t i e s . The f r a c t i o n of heat generated by other species i n the gas stream w i l l be proportional to ( i o " ^  " V n l 2 - i2) where i i s the current measured when the discharge was on (but before NO^ was added). The f r a c t i o n of heat generated by other species(compared to the heat generated by atomic oxygen recombination) was found to be f a i r l y con-stant as the microwave power was changed at a p a r t i c u l a r pressure but extremely v a r i a b l e as the pressure changed (as shown i n Table 2 .1) . Since the atomic oxygen dependencies of the 557.7 nm emission were each measured at constant pressure,and the probe was only used f o r r e l a t i v e oxygen atom concentration determinations, i t appeared that the oxygen atom dependence experiments would s t i l l be v a l i d . However the large concentrations of other excited species r e l a t i v e to the atomic oxygen concentrations was cause for further - 94 -TABLE 2.1 EXCITED SPECIES DETECTION BY THE ISOTHERMAL PROBE " 0 - r e a l " "O-apparent" 0 r e a l 0 apparent 3 TORR ARGON 0.2 .362' 1.165 1.197 1.226 1.302 0.94 .603 1.157 1.182 0.975 1.034 0.94 4 TORR ARGON 0.1 .834 1.122 1.141 0.563 0.606 0.93 .442 1.102 1.141 1.019 01.107 0.92 " .108 1.098 1.140 1.194 1.287 0.93 .223 1.223 1.270 1.446 1.563 0.92 6 TORR ARGON trace 655 ,845 ,888 0.1 1. 037 1. .288 1. .340 0.1 0. .873 1. .252 1. .340 0.1 0. .617 1, .226 1, .351 0.1 0. .464 1, .210 1 .340 0.5 0, .010 1 .539 1 .668 0.5 0, .391 1 .452 1 .630 .285 .359 0.79 .584 .720 0.81 .877 1.105 0.79 1.122 1.444 0.78 1.748 1.580 0.79 1.348 1.762 0.77 1.955 2.504 0.78 8 TORR ARGON fl.l 1. 226 1. .359 1. ,405 11 1. .006 1. .303 1. .399 I t 0. .637 1. .223 1, .400 I t 0, .297 1, .169 1 .387 1.0 1, .609 1 .704 1 .764 I I 1 .425 1 .640 1 .767 t t 0 .293 1 .405 1 .735 .344 .471 0.73 .686 .945 0.73 1.089 1.544 0.70 1.278 1.835 0.70 .315 .523 0.60 .659 1.092 0.61 1.888 2.924 0.65 12 TORR ARGON 0.1 0.276 0.773 1.075 .521 1.079 0.48 - 95 -in v e s t i g a t i o n into t h e i r i d e n t i t y and e f f e c t on the 557.7 nm emission, 2.3.3 Method 3 --Measurement of the Intensity of  NO., emissions from the 0 + NO Reaction The t h i r d method of atomic oxygen measurement that was used i n t h i s i n v e s t i g a t i o n involves the emission i n t e n s i t y of the afterglow from the reaction 0 + NO ^ N 0 2 + hv (0.4 - 1.4 ym) (28) which i s proportional to only the amount of NO and atomic oxygen concentration [151, 168], i . e . : I o b s « [0] [NO] Hence the r e l a t i v e atomic oxygen concentration may be determined by adding a f i x e d , small amount of NO to the gas stream and moni-to r i n g the afterglow i n t e n s i t y . To obtain absolute values for [0] the emission i n t e n s i t y was c a l i b r a t e d by the t i t r a t i o n of nitrogen atoms with NO. A stream of nitrogen atoms was prepared by discharg-ing a N 2 - Ar ( t y p i c a l l y 0.1 t o r r - 4 t o r r r a t i o ) mixture and monitor-ing the chemiluminscence i n the bulb at 556.4 nm while adding small increments of NO to the atom stream. The chemiluminescence was observed to diminish as more NO was added u n t i l a point was reached where the i n t e n s i t y was at a minimum. Addition of more NO resulted i n an increase of the i n t e n s i t y . A plot of the emission i n t e n s i t y during a t i t r a t i o n of nitrogen - 96 -atoms i s shown i n Figure 2.6. By extrapolating from the points i n the v i c i n i t y of the end point, a value of [NO] was found which was equal to the o r i g i n a l nitrogen atom concentration [ N ] q . At t h i s point the oxygen atom concentration i s equal to [N] Q. After the end point,the i n t e n s i t y increase has a slope equal to K £ ^ [ 0 ] i n the r e l a t i o n s h i p (from equation (32)): I , — - = K f f [0] (52) [NO] E T T and as [ 0 ] = [N] or [NO] , . ^, K was obtained for the 1 J L Jo 1 Jendpomt e f f system. Once the system was calibrated,the oxygen atom concentration i n the flow system was determined by adding a small, known amount of NO and measuring the i n t e n s i t y I , at 556.4 nm. The atom concentration 6 J obs i s given by *obs [0> • w n r ^ (53> where the I , i s i n the same units as the i n t e n s i t y measured i n obs J the c a l i b r a t i o n . This technique had the advantage of perturbing the system l e s s , since far less NO was added. Other advantages included ease of operation and f a s t e r "recovery" times from the afterglow. 2.4 ABSOLUTE EMISSION INTENSITY MEASUREMENTS The absolute emission i n t e n s i t y measurements were made using - 97 -Figure 2.6 Intensity of the emission at 556.4 nm as a function of N O added to a stream of n i t r o -gen atoms. - 97a -[NO] (molecule cm"3) x10" - 98 -the method of Fontijn et a l . [177]. 2.4.1 The 557.7 nm Emission from 0( 1S) Because the atomic line emission of 0(^S) extends over a very small wavelength range, the determination of the absolute emission intensity is greatly simplified. A schematic representation of the wavelength region for 0(^S) emission is shown in Figure 2.7. For the 557.7 nm line, where the - X^ difference i s of'the order of 2 nm, the spectral distribution of the NO - 0 glow (Figure 1.9) and spectral sensitivity of the system (Figure 2.3) are essentially constant, i.e.: I (X) = I s s and From equation (47) (page 66) the expression ^.INgUgl/2 V s ; ' " ^ ' / f \ d x t 5 4 ) is obtained where k is the radiative rate constant for the 557.7 nm r transition of 0( 1S) = 1.06 s" 1 [25]. Equation (54) can be rewritten as 4 = JNQHO] f W r ( x ) A ( 5 5 ) k s [ o ^ s n y x . - x p x i 0 ( S ] - 99 -Figure 2.7 Schematic representation of the spectral d i s t r i b u t i o n of the 0( 1S) emission with respect to the NCL emission. - 99a -- 100 -For the atomic l i n e of 0(*S), the i n t e g r a l i n equation (55) i s equal to the area under the t r i a n g l e between X^ and i n Figure 2. which i s equal to the product of the peak height and f u l l - w i d t h half-peak maximum or e f f e c t i v e band-pass of the monochromator at a given s l i t width, i . e . : fx 1 ^ 1 ^ (A)dX = I 0 ( 1 S ) (557.7 nm) x band-pass (nm) (56) In the region X^  to X^  the spectral flux F^(^) i s e s s e n t i a l l y con-stant, i . e . : F ( X) = F s^ •* s so from equation (42) we obtain: k' = k s V » 2 - V (57) s j s and the o v e r a l l expression,combining equations (55) to (57): i , p I'/ T = f^Oj [0] x 0( S) x band-pass s V ' s [O^S)] I s The N O - 0 glow was measured at 556.4 nm^which from Figure 1.1 y i e l d s a value f o r the absolute rate constant i n t h i s region of: k s F s . _ ..-19 3 -1 -1 — — = 1.2 x 10 cm s nm s At a band-pass of 0.25 nm (0.25 mm s l i t width), the absolute i n t e n s i t y of the 557.7 nm l i n e was calculated from - 101 -I a b s ( 0 ( 1 S ) ) = 3.0 x 10" 2 ° C T n " 3 [ 0 ] [NO] °[ S ) (58) *s In the i n i t i a l part of t h i s i n v e s t i g a t i o n , the [0] i n equation (58) was determined by NO^ /^NO t i t r a t i o n and then the i n t e n s i t y of the NO-0 glow, I , was monitored for a known amount of NO. However the absolute emission i n t e n s i t y c a l i b r a t i o n s were found to have an NO dependence even a f t e r correction for NO i n equation (58). One p o s s i b i l i t y considered was that the [0] measured by the t i t r a t i o n was not the [0] value needed i n expression (58) because of the decay of oxygen atoms i n the observation c e l l . The atomic oxygen concentration used i n evaluating (58) should be the "average" [0] i n the observation c e l l . However, the t i t r a t i o n of oxygen atoms must be c a r r i e d out at the entrance of the bulb as the t i t r a t i o n i n l e t i s added approximately 4 cm above the entrance. When NO i s added to the flow system, oxygen atoms are consumed p r i n c i p a l l y by the reaction k30 0 + NO + M ^ N 0 2 + M (30) To compensate for t h i s loss of 0, an average value of the atom concentration i n the c e l l , [0] , can be calculated from the 1 av expression [ 0 ] a y = [0] oexp(-k 3 Q[M][N0]t) where [0] = o r i g i n a l atomic concentration at the t i t r a t i o n i n l e t k„„ = rate constant for r e a c t i o n (30) - 102 -= 6 x 10" 3 2cm 6s' 1 [177] [M] = t o t a l p a r t i c l e concentration t = flow time between the t i t r a t i o n i n l e t and centre of the observation c e l l = volume between t i t r a t i o n i n l e t and centre of observation c e l l (cm"3)  •2 I flow rate of gas stream (cm s" 1) In the case of the c y l i n d r i c a l observation c e l l , the volume was e a s i l y calculated from the i n t e r n a l diameter of the tubing, but the s p e r i c a l and e l l i p t i c a l observation c e l l s posed a greater problem due to possible non-homogeneity of the gas stream and uncertainty i n the l i g h t i n t e g r a t i n g e f f i c i e n c y of the v e s s e l . Corrections of [0] i n expression (58) f o r decay by reaction (30) did not account for the observed [NO] dependency of the sensi-t i v i t y of the photon counting system. The smaller the amount of NO used i n the c a l i b r a t i o n , the greater the "apparent s e n s i t i v i t y " calculated from equation (58) became. Whether the discrepancy lay i n the t i t r a t i o n method for determining [0] or the i n t e n s i t y c a l i -b ration method i t s e l f was uncertain. Fortunately one can use the NO t i t r a t i o n of nitrogen atoms described i n section 2.3.3 to obtain the absolute emission i n t e n s i t y c a l i b r a t i o n . The slope of the emission i n t e n s i t y versus NO i n Figure 2.6 i s equal to I g/[N0] and [0] i s equal to the [ N O ] e n d p o i n t • - 103 -Substituting into equation (58) one obtains -2f) 3 ' , n , l e , , 3.0 x 10 cm [NO] , . x I n , l e . I a b s ( 0 ( S)) = 1 Jendpoint 0( S) slope of N-NO t i t r a t i o n where I ^ l ^ , the measured peak height, i s measured i n the same units as the i n t e n s i t y measurements of the nitrogen atom t i t r a t i o n with NO. One advantage of t h i s method i s that the NO concentrations f o r which the slope of Figure 2.6 i s measured are small and the contribution of reaction (30) should be minimal. Removal of 0 by reaction (30) would lead to a curvature of the slope, which + 13 -3 was not observed when [NO] < 3 x 10 cm . The amount of NO added for the atomic oxygen determination by method-3 was u s u a l l y 1 ^ 1 2 -3 < 6 x 10 cm . The o p t i c a l s e n s i t i v i t i e s that were measured by the technique described above were found to be s e l f - c o n s i s t e n t . 2.4.2 The 1.27 ym Emission from O ^ a ^ ) The absolute emission i n t e n s i t y at 1.27 ym was required i n order to c a l c u l a t e the 0_(a*A ) concentration i n the observation c e l l . Using a technique s i m i l a r to that used for the 0(*S) emission, the 0 2 ( 3 ^ ^ ) band was scanned with the G.C.A./McPherson monochromator to obtain the 1.27 ym emission i n t e n s i t y - The monochro-mator wavelength was then changed to about 1.24 ym where the emission from the reaction - 104 -i k s 0 + NO *"N02 + hv(1.24 ym) -20 could be measured. From Figure 1.10 one obtains a value of 2 x 10 3 s 1nm"1 for k' and hence the absolute 0„((a 1A ) emission i n t e n s i t y s g can be calculated from the r e l a t i o n s h i p i I a b s ( 0 2 ( a 1 A g ) ) = 2 x 10~ 2°cm~ 3[0][NO] )'21 m x B.P. (nm) (59) T1.24 ym i i where I (1.27 ym) and I (1.24 ym) are the r e l a t i v e i n t e n s i t i e s 1 * of the 0 (a A ) and NO- emissions, B.P. = Band-pass of the mono-Z. g £. chromator (10.6 nm) and [0] and [NO] are obtained using methods previously described i n t h i s section. Using the r a d i a t i v e t r a n s i t i o n p r o b a b i l i t y f o r the 1.27 ym t r a n s i t i o n of 2.57 x 10" 4s 1 [35], the concentration of 0 2(a 1A ) in the bulb was calculated from the expression [0 2(a 1A )] = I a b s ( 1 . 2 7 ym) x 2.57 x 1 0 ' V 1 . - 105 -CHAPTER 5: RESULTS AND ANALYSIS (PART 1)  PRELIMINARY STUDIES In previous major studies of Of^S) formation from oxygen atom recombination, Young and Black [50] and Slanger and Black [93] observed a 557.7 nm emission i n t e n s i t y which was second order i n atomic oxygen and independent of pressure. Young and Black i n t e r -preted t h i s r e s u l t i n terms of Of^S) formation by a Chapman reac-1 3 t i o n (reaction (1)) with predominant 0( S) quenching by 0( P), although Slanger and Black noted that formation of Of^S) by a 1 3 Barth mechanism, with predominant removal of 0( S) by 0( P) and the * precursor 0^ by M, would r e s u l t i n the same observed dependence, i . e . : 1 (557.7 nm) <* [ 0 ( 3 P ) ] 2 In the present i n v e s t i g a t i o n , the atomic oxygen stream was obtained from a discharge of 0 2 i n excess Ar. Since the oxygen molecules are not f u l l y d i s s o c i a t e d , the mixture that enters the observation c e l l was composed of 0, 0^ and Ar. However, both Young and Black, and Slanger and Black produced atomic oxygen i n t h e i r study of Of^S) by the t i t r a t i o n of atomic nitrogen (in an i n e r t gas) with NO: N + NO *>0 + N 2 (19) At the t i t r a t i o n n u l l , t h e i r gas mixture consisted of 0, N 2 - 106 and the i n e r t gas. They reported that i d e n t i c a l r e s u l t s were obtained f o r d i f f e r e n t N 2 / i n e r t gas r a t i o s , and- concluded that the N 2 a n d i n e r t gas ef f e c t s f o r Of^S) formation were equivalent. Comparing the present study of 0(^S) to those of previous workers; Gas Mixture Reference 0, 0^ (+ i n e r t gas) This work 0, (N 2 + i n e r t gas) Young and Black [50] " " " " Slanger and Black [93] the major diffe r e n c e i s 0 2, which was always present i n t h i s system. Because of t h i s , the f i r s t experiments attempted were to determine the molecular oxygen dependence of the 557.7 nm i n t e n s i t y . 3.1 MOLECULAR OXYGEN DEPENDENCE OF THE 557.7 nm EMISSION The 0 2 concentration could be changed by adding 0 2 before or af t e r the discharge. However, i f 0 2 was added to the flow stream before the microwave discharge region, the concentration of atomic oxygen changed, and the e f f e c t of 0 2 on the 557.7 nm emission could have been obscured or complicated by the e f f e c t of changing [0]. By adding 0 2 a f t e r the discharge, i t was determined (by N0/N02 t i t r a t i o n ) that the p a r t i a l pressure of atomic oxygen was not notice-ably affected as long as the t o t a l pressure r i s e i n the system was - 107 -A t y p i c a l p l o t of the inverse of the 557.7 nm emission i n t e n s i t y versus added [0^] i s shown i n Figure 3.1. The i n t e n s i t y was observed to decrease f a i r l y l i n e a r l y as 0^ was added, u n t i l a c e r t a i n [0 2] was reached where the i n t e n s i t y decreased more r a p i d l y with added 0^. Curvatures of the 1/(1(557.7 nm) versus [0 2] p l o t s , s i m i l a r to Figure 3.1, were found under a l l conditions of pressure and atomic oxygen concentration, although the onset of n o n - l i n e a r i t y varied from one experiment to another. In the i n i -t i a l a n a l y s i s , only the l i n e a r slope and extrapolated intercept of the plo t s were used. If the oxygen quenching data i s analysed i n terms of the Chap-man Reaction: k l 1 0 + 0 + 0 »»0( S) + o 2 (1) 1 3 with quenching of 0( S) by both 0( P) and 0 2 i n : 1 k7 0( S) + 0 ^quenched products (7) 1 k 9 0( S) + 0 2 ^ q u e n c h e d products (9) and the r a d i a t i v e t r a n s i t i o n , 1 t " 1 1 OrS) I — » • 0( D) + 557.7 nm (10) then, using standard steady state assumptions, the r e l a t i o n s h i p : - 108 -Figure 3.1 The [0 2] dependence of the 557.7 nm emission i n t e n s i t y , I. - 108a -- 109 --1 3 k ! T [°] (60) K557.7 nm) = y o j + y o ^ i s obtained. Re-writing equation (60), one obtains , vo] + yo2] 1(557.7 nm) ^ " V f -(61) and a p l o t of 1/1(557.7 nm) versus [0 2] (e.g. Figure 3.1) should y i e l d : M O ] Intercept _ _7_ S.lope kg The r e s u l t s of these pl o t s are summarized i n Table 3.1 and pl o t t e d i n Figure 3.2. It can be seen that there i s a c o r r e l a t i o n between the observed intercept/slope values and the atomic oxygen concentration, although there i s considerable scatter i n the data. The slope of Figure 3.2 gives the r a t i o of 0 to 0 2 quenching rate constants, which from the above Chapman reaction would be: k y/k g = 5 ± 5 However, t h i s i s not consistent with the l i t e r a t u r e value of k_/kn = 65 [93, 101]. Hence, i t was concluded that either the l i t erature quenching rate constants of 0(^S) were in c o r r e c t or that the Chapman mechanism was not responsible for the e x c i t a t i o n of 0(*S) i n t h i s system. I f the data are now analyzed i n terms of the Barth type mechan 110 -TABLE 3.1 INTERCEPT/SLOPE VALUES OF 0 2 QUENCHING PLOTS AT DIFFERENT [0] Intercept/Slope ( x l 0 1 6 c m _ 3 ) [0] (xlO cm 0.4 1.1 1.0 1.3 2.4 1.6 0.95 0.77 2.33 1.95 1.7 3.4 1.4 0.78 - I l l -Figure 3.2 The oxygen atom dependence of the intercept/ slope r a t i o of the 0 2 quenching p l o t s . (1/1(557.7 nm) vs. [ 0 2 ] ) . 112 i . e . 0 + 0 + M—Lo * + M (2) k 0 * + 0—L t)( 1S) + 0? (3) -1 0( 1S) -^-». 0( 1D) + 557.7 nm (10) with predominant removal of 0(^S) by 0( 3P): 1 3 k 7 0( S) + 0( P)—^quenched products (7) 3 and quenching of the precursor by both 0( P) and 0 2: * 3 k4 0„ + 0( P) »»quenched products (4) 5 0^ + quenched products (5) then, using standard steady state assumptions, one obtains the expression: , _ k 2k 3 T _ l [ o r[M] ^ 5 5 7 - 7 n m> = k 7(k 4[o] + k5[o2]) ( 6 2 ) or 1/1 (557.7 nm) = k ^ f O ] + k 5 [ 0 2 ] ) A ^ T - 1 [0] 2 [M] (63) From equation (63) i t can be seen that a plot of the inverse of the i n t e n s i t y versus [0 2] should y i e l d : Intercept _ k4 ^ Slope k 5 As was observed e a r l i e r , Figure 3.2 y i e l d s a value of 5[0] for t h i s r a t i o . However, whereas the Chapman Mechanism required that the r a t i o k^/kg = 5 (which i s unacceptable), the Barth mechanism equates k^/k^ to 5. Since the r a t i o k^/k^ has not been determined independently the mechanism i s at least acceptable.. The curvature of the 0^ quenching pl o t s at high [O^] could have been due to e i t h e r ( i ) a reduction of the atomic oxygen concentration when [0^] becomes large enough to e f f e c t the discharge con-d i t i o n s , or 2 1 ( i i ) a 1/ [^2] dependency of the 0( S) emission. Although such a curvature could be due to a second order quenching rate a r i s i n g from quenching of both the precursor 0^ and O^S) by 0 2 (equations (5) and (9)), p l o t s of 1(557.7 nm)(k^[0] + k ^ ] ) ' v e r s u s ^ ( 6 4 ) using the l i t e r a t u r e values of k_, and k^, were also non-linear. Hence i t was assumed that at high [0,,] , a reduction in atomic oxygen concentration resulted i n a larger decrease i n 557.7 nm i n t e n s i t y than would be expected from 0 2 quenching. The e f f e c t was not investigated further since the "curved region" was not used i n the analysis. 3.2 ATOMIC OXYGEN DEPENDENCE OF THE 0( 1S) EMISSION The dependence of the 0( XS) emission on the atomic oxygen c centration i s of prime importance i n an understanding of the mech-anisms of both the formation and removal processes f o r 0(^S) i n the laboratory. In the i n i t i a l experiments both the amount of 0 2 through the discharge and the microwave power were varied to increase the range of [0] obtainable. The i n t e n s i t y was found to increase as the microwave power to the discharge was increased but to decrease when more 0^ was added before the discharge (even though the atomic oxygen concentration increased). This observa-t i o n i s consistent with the r e s u l t s presented i n section 3.1, where 0 2 appeared to be the major quencher of the precursor as k 4 / k 5 = 5±5 and the r a t i o of [0]/[0 2] < 0.2 ( i . e . : ^[OoJ > k 4[0]). Figure 3.3 i l l u s t r a t e s the observed atomic oxygen dependence of the 557.7 nm emission as a log (1(557.7 nm)[0 2]) versus log [0] p l o t . The slope of t h i s p l o t (~ 2) shows that the i n t e n s i t y of the 557.7 nm emission i s dependent upon the square of the atomic oxygen concentration, i n agreement with e a r l i e r work [50, 93]. The atomic oxygen concentration at each data point of Figure 3.3 was measured using the N02/N0 t i t r a t i o n , and the 0(*S) inten-s i t y was observed i n the c y l i n d r i c a l observation c e l l . In an attempt to obtain an [0] dependence of the 557.7 nm emission with less s c a t t e r (than Figure 3.3), these measurements were repeated using the isothermal c a l o r i m e t r i c probe. Table 3.2 l i s t s the r e s u l t s obtained f o r the [0] dependence at several d i f f e r e n t pres-sures (using the probe to measure [0]). - 115 -Figure 3.3 The oxygen atom dependence of the 557.7 nm emission i n t e n s i t y p l o t t e d as log 1(557.7 nm)[C^] versus log [0] (Indicated slope = 2. 05) . - 115a -1 I 1 1 1 1 >-1.1 1.7 2.3 L o g [0] - 116 -TABLE 3.2 OXYGEN ATOM DEPENDENCE OF THE 557.7 nm EMISSION  OBTAINED BY MEASURING [0] WITH ISOTHERMAL PROBE TECHNIQUE IN THE CYLINDRICAL OBSERVATION Argon (torr) Oxygen (torr) Slope of In I vs. In plot 18 0.1 2.0 18 0.1 2.1 12 0.1 2.2 12* 0.1 2.0 10** 0.1 2.2 8 0.6 2.4 8 0.1 2.25 8 0.1 2.1 4 0.1 2.7 4 0.1 2.9 4 0.1 2.6 3 0.1 3.2 [0] * I n c l . 8 t o r r Ar added a f t e r the discharge ** I n c l . 5 t o r r Ar added a f t e r the discharge - 117 -It can be seen from Table 3.2 that the order of the atomic oxygen dependence was observed to decrease as the pressure increased In order for the atom dependence to decrease, the quenching rate 1 * of the 0( S) in reaction (7), or the precursor 0^ i n reaction (4), 3 by 0( P) must be becoming more important as the pressure increases. 1 * Simple M (Ar) quenching of 0( S) or 0^ would r e s u l t i n the opposite e f f e c t . However, the observations could be explained i f d i f f u s i o n c o n t r o l l e d wall quenching were important, i . e . : * 1 wall CL or 0( S) ». quenched products (65) k w then the i n t e n s i t y of the 557.7 nm emission would show the r e l a -tionship : 1/1(557.7 nm) « k (66) w The theory of wall deactivation has been treated by a number of workers [142, 184, 185]. In d i f f u s i o n c o n t r o l l e d systems, the rate constant k i y reduces [112] to the equation: k = Y 2 D / r 2 (67) w o o o where y = 2.405 which i s the f i r s t root of J , the Bessel func-o o t i o n of order zero, r i s the flow tube radius, and D i s the o o d i f f u s i o n c o e f f i c i e n t . Since D can be written as D/[M], where D i s a pressure inde-o pendent d i f f u s i o n c o e f f i c i e n t , and M i s the t o t a l pressure or con-- 118 -centration i n u n i t s consistent with those of D, one can write W r > l This would r e s u l t i n surface deactivation of Of/s) or 0 2 (equation (65)) becoming.more important r e l a t i v e to atomic oxygen quenching (reaction (7)) as the pressure decreased. In order to assess the importance of wall deactivations i n t h i s i n v e s t i g a t i o n , the c y l i n d r i c a l observation c e l l was replaced with a 2 - l i t r e and then a 5 - l i t r e bulb. Since the magnitude of 2 depends on the square of the radius of the flow tube ( r Q i n equation (67)), the magnitude of k^ should have decreased by the r a t i o of approximately 1 ': 0.032 : 0.017 ( C y l i n d r i c a l vessel) : (2-1 bulb) : (5-1 bulb) as the observation'cells were exchanged for larger ones. Table 3.3 shows the r e s u l t s obtained i n oxygen atom dependence measurements performed i n the 2 and 5 - l i t r e observation c e l l s . In general the atomic oxygen dependencies obtained i n the spherical observation vessels were smaller than those observed i n the c y l i n -d r i c a l vessel (Table 3.2). * 1 This indicated that the surface deactivation of 0 2 or 0( S) - 119 -TABLE 3.3 OXYGEN ATOM DEPENDENCIES DETERMINED IN  THE 2 AND 5-LITRE OBSERVATION CELLS, USING THE ISOTHERMAL PROBE TO MEASURE [0] Argon (torr) Oxygen (torr) Observation c e l l (1) Slope of In I vs. In [0] 8 0.1 2 1.9 8 0.5 it 1.6 4 0.1 tt 2.0 4 0.1 5 1.9 2 it ti 2.3 2 tt I I 2.4 3 it it 2.5 4 tt ti 1^8 4 11 it 2.1 8 it I I 1.8 6 11 it 1.8 6 0.2 n 1.8 6 0.4 ti 2.0 6 0.6 11 1.9 6 0.3 tt 1.8 6 0.1 ti 1.9 - 120 -in reaction (65) may have been an important loss process i n the smaller v e s s e l . However, no discernable di f f e r e n c e s were observed i n the atomic oxygen dependencies obtained i n the 2 and 5 - l i t r e bulbs. Hence, surface deactivation was considered unimportant i n studies performed i n these larger observations vessels. Also, from Table 3.3 i t can be seen that many of the oxygen atom dependence measurements resulted i n an emission i n t e n s i t y order with respect to 0 of less than 2. The smallest atomic oxygen order that would be consistent with a Chapman mechanism i s 2, since the formation step i s 3rd order i n [0] while the loss processes are eit h e r zero or 1st order i n [0]. However, i n a Barth mechanism an order of less than 2 could be explained by 0 quenching of both the 1 * 0( S) and the precursor 0^ i - n reactions (7) and (4) . k4 0^ +0 ^ quenched products (4) 1 k 7 0( S) + 0—^quenched products (7) Rearranging equation (63) one can obtain the expression k [k [0] + k [0 ]) [ 0 ] z J-—z — - — 1(557.7 nm) = - T ^ k ^ M ] ( 6 8 ) Under conditions of constant [0 2] and fM] the 557.7 nm emission i n t e n s i t i e s measured for d i f f e r e n t [0] can be pl o t t e d as 2 . , c l f f i l IT- versus [0] 1(557.7 nm) J and from equation (68) t h i s y i e l d s : - 121 -intercept k f , „ slope V U 2 J / k 4 2 The p l o t s of [0] /I versus [0] were found to be rather scattered., e s p e c i a l l y as the 0 dependence approached second order. Table 3.4 shows the values of intercept/slope for a number of experi-ments performed at d i f f e r e n t pressures and molecular oxygen concen-t r a t i o n s . The r a t i o k^/k^ obtained from Table 3.4 ranges from 0.22 to 0.94, but i s at least consistent with the r a t i o ^ 4 / ^ 5 = 5 ± 5 obtained i n the oxygen quenching experiments described i n section 3.1. Within experimental error, the r e s u l t s presented so far i n d i -cate a second order dependence on 0. In the next set of experiments the assumption was made that any v a r i a t i o n i n [0], which occurred when [M] was changed, could be corrected for i n the 557.7 nm 2 emission i n t e n s i t y measurement by measuring 1(557.7 nm)/[0] . 3.3 PRESSURE OR M DEPENDENCE OF THE 557.7 nm EMISSION The dependence of the 557.7 nm emission on [M] could provide a means of d i s t i n g u i s h i n g between the Barth and Chapman mechanisms because the Chapman mechanism does not include M i n i t s formation step (reaction (1)), whereas the Barth mechanism does (reaction (2)) Figure 3.4 shows the f i r s t determination of the M dependence, correcting the 557.7 nm i n t e n s i t y f or the [0] and [0,,] dependencies found i n the previous experiments (sections 3.1 and 3.2) by p l o t t i n g . - 122 -TABLE 3.4 RESULTS OF r o l 2/I(557 .7 nm) vs. [0] PLOTS Argon (torr) Oxygen (torr) [o 2] (cm xlO ) Intercept .. Slope ( x l 0 1 5 c m - 3 ) I/S [0 2] = k 5 A 4 4 0.1 3.2 2.2 0.69 6 0.1 3.2 3.0 0.94 8 0.1 3.2 2.4 0.75 6 0.2 6.4 3.1 0.48 6 0.3 9.6 2.6 0.27 6 0.4 12.8 4.0 0.31 8 0.5 16.0 3.7 0.22 6 0.6 19.2 4.8 0.25 - 123 -Figure 3.4 The pressure dependence of the 557.7 nm emission i n t e n s i t y (corrected f or [0] and [0 2] by p l o t t i n g 1(557.7 nm)[0 2]/[0] 2) . - 123a -- 124 -.1(557.7 nm)[0 2] The atomic oxygen concentration at each data point i n Figure 3.4 was determined by the N02/N0 t i t r a t i o n and the [0 2] was corrected for the amount of d i s s o c i a t i o n . In order to ensure that the M dependence observed was not due to a pressure dependence of the t i t r a t i o n technique, the experiment was repeated i n the 5 - l i t r e observation c e l l using the isothermal probe to measure [0]. The v a r i a t i o n i n [0] was corrected f o r by 2 p l o t t i n g the 1(557.7 nm)/[0] as a function of M. The r e s u l t s of t h i s experiment are shown i n Figure 3.5, and they indicate that the apparent M dependence of the emission i s independent of the method of atomic oxygen measurement. Also, to decrease the influence of other v a r i a b l e s , an attempt was made to keep a l l atomic and mole-cular concentrations (with the exception of [M]) constant. This was accomplished by adjusting the microwave power of the discharge to obtain constant [0] at each pressure. Because the isothermal probe i s only a measure of the flow of atomic oxygen, the current at each point was normalized to obtain a constant p a r t i a l pressure of 0 i n the system. By t h i s point i n the i n v e s t i g a t i o n we had found the large d i f f e r e n c e between "apparent 0-flow" and "actual 0-flow" due to the detection of another excited species. Therefore the current without a flow of atomic oxygen i w n i n the r e l a t i o n s h i p : - 125 -Figure 3.5 The pressure dependence of the 557.7 nm emission i n t e n s i t y using the isothermal probe technique to measure [0] i n 1(557.7 nm)/[0] 2. -125a -- 126 2 2 0 flow ex. ( i N Q - i D R was determined at the e x t i n c t i o n point when NO^ was added to the flow system. Table 3.5 l i s t s the values obtained i n t h i s experiment. Each point was extremely d i f f i c u l t to obtain due to the requirement that both i and i ^ g had to be measured separately to obtain the desired 2 2 2 ( i ^ - i ) value corresponding to a constant [0]. However, Table 3.5 shows that there i s a d e f i n i t e [M] dependence to the 557.7 nm emission i n t e n s i t y when [0] and [O^] are kept constant. A l l of the M dependence experiments are consistent with a single 1st order M dependence for the i n t e n s i t y of the 557.7 nm emission. This r e s u l t can be explained by the M i n the formation step of the Barth mechanism (reaction (2)) i f there i s no s i g n i f i -cant quenching of either the precursor or 0(^S) i n * + M ^quenched products 0(^S) + M — ^ q u e n c h e d products However, there i s an a l t e r n a t i v e explanation which needs to be considered. 3.3.2 Collision-Induced Emission The r a d i a t i v e decay of Of^S) has been shown to proceed at a rate depending l i n e a r l y on the concentration of i n e r t gases present - 127 -TABLE 3.5 fM! DEPENDENCE AT CONSTANT fO-1 USING THE  ISOTHERMAL PROBE TO MEASURE THE RELATIVE 0-ATOM FLOW 1(557.7 nm) arb. units Argon t o r r "0-flow" arb. units r.2 .2, flo w - p a r t i a l pressure normalizing f a c t o r [0] arb. units 46 8 1.055 1 1.055 40 6 1.028 0.976 1.053 26 4 1.001 0.9371 1.068 - 128 -i n the gas phase [186, 187, 188, 189]. This e f f e c t i s known as c o l l i s i o n - i n d u c e d emission, which (in argon) can be represented by the equations. Ar + 0( 1S) a i — f * - A r Q ( 1 S ) — ^ Ar + 0 + hv (green system) (69) The green system, i n reaction (69), observed by Cunningham and Clark [188] i n 70 t o r r of argon i s reproduced i n Figure 3.6. Although the rate constant for the t o t a l emission band from reaction (69) i s of the order of 3.0 to 4.7 x 10" 1 8cm" 3s _ 1 [188, 189], only a small proportion of the induced band l i e s below the atomic l i n e i t s e l f . Cunningham et a l . [188] estimated an upper l i m i t of o 2% of the induced emission l y i n g within t h e i r 1.6A bandwidth mea-surement of the atomic l i n e . The observations i n t h i s study were done with either 0.25 or 0.55 mm s l i t widths on the v i s i b l e mono-o chromator, which res u l t e d i n bandwidths of 2.5 or 6.OA. The 557.7 nm emission dependence on [M] was found to be the same for e i t h e r s l i t width. If i t i s assumed that 10% of the induced emission system was -19 -3 -1 recorded as the regular atomic l i n e , then k 6 g ~ 4 x 10 cm s and Total Emission P r o b a b i l i t y (Induced Emission + -19 -1 Spontaneous emission) = ~ 4 x 10 [M] + x where x i s the r a d i a t i v e l i f e t i m e f o r 0( 1S) = 0.94 seconds [25]. - 129 -Figure 3.6 Spectrum of the c o l l i s i o n - i n d u c e d emission from Ar - 0( 1S) at 70 t o r r t o t a l pressure (0.01% 0^) reproduced from Cunningham and Clark [188]. - 129a -0I(557.7nm) - 130 -If the above estimates are cor r e c t , then between 2 and 10 to r r the t o t a l emission p r o b a b i l i t y i s predicted to change from 1.06s"1 to 1.19s" 1. Even cor r e c t i n g the emission i n t e n s i t i e s by t h i s upper l i m i t f o r the induced emission does not appreciably a l t e r the observed [M] dependence since t h i s changes by a factor of 5 when the induced emission changes only 10%. From the above evidence i t was concluded that the M dependence of the O^S) emission i n t e n s i t y i s not a " t r i v i a l " e f f e c t and can be taken as strong evidence that the Chapman mechanism i s inadequate and a Barth type mechanism i s required to explain O^S) formation in our system. 3.4 QUENCHING BY OTHER GASES If the 557.7 nm signal i s diminished when another gas, Q, i s added to the flow system, the gas can be quenching either the pre-* 1 cursor 0^ or 0( S) d i r e c t l y , i . e . : k Q 0 2 + Q — q u e n c h e d products (70) 1 k ' 0( S) + Q — q u e n c h e d products (71) The e f f e c t of reactions (70) and (71) on the i n t e n s i t y of the 0( 1S) emission may be represented by k ^ T - V l V ] ( 7 2 ) 1(557.7 nm) = ( ^ [ Q ^ + ^ [ Q ] ) ^ [ 0 ] + k ^ [ Q ] ) - 131 -I f Q i s an e f f e c t i v e quencher i n both (70) and (71) with res-pect to the predominant quenching processes (7) and (5), then one 2 should observe a [Q] dependence f o r 1/1(557.7 nm). The 557.7 nm i n t e n s i t y was found to be f a i r l y l i n e a r f o r a l l of the gases added with the exception of 0 2 (see section 3.1). Hence, for most systems only one of reactions (70) or (71) contributes to the quenching of the 0(*S) emission. The e f f e c t of reactions (70) or (71) w i l l depend upon the size of k^ or k , compared to k^ [0,,] or k^[0]. Although we have no knowledge of k^ or k , the rate constants k^,, for the quenching of 0(^S) d i r e c t l y , have been reported (see Table 1.4) f o r several gases. The r a t i o k q I /ky i s very small f or 0 2, C0 2 and SF 6 (0.015, 0.020, 0.0027). With the usual [0] of about 15 -3 1 x 10 cm , i t would require 2 t o r r of added 0 2 to equal the quenching by 0. In these experiments much smaller concentrations of the quencher were used, and hence i t was assumed that i n a l l cases the added gases were quenching the precursor 0 2 . For the assumed mechanism, given by equations (2), (3), (5) and (7), and with standard steady state assumptions, the i n t e n s i t y , I , of the 557.7 nm emission before the addition of the quenching gas, Q, i s given by: x " 1 k 2 k 3 [0] 2 [M] I o(557.7 nm) = k (73) Assuming that the quenching of 0( 1S) by Q i s n e g l i g i b l e , the i n t e n s i t y , I, observed upon addition of the quenching gas Q (equa-- 132 -tion (70)) is given by T _ 1 k 2 k 3 [ 0 ] 2 [ M ] (74) 1(557.7 nm) = ^ [ c ^ ] + k [Q]) Dividing equation (73) by (74), a "Stern-Volmer" type of equation is obtained: k [0 ] + k [Q] k [Q] I / i = _ 2 — f 9 = 1 + _3 (75) k 5[o 2] k 5[o 2] and a plot of I Q / I versus [Q] should have a slope equal to kq/k,-[02] The slopes of the plots of I /I versus [Q] listed in Table 3.6 were found to be not very constant at fixed [0 2], and in fact show a definite 0 dependence. If the contribution of 0 quenching (reaction (4)) of the precursor is included in equation (75), one obtains the expression (k4[0] + k 5[0 2] + kq[Q]) V 1 = (k 4[0] + k 5[0 2]) ( 7 6 ) k [Q] 1 + - a i H . k 4[0] + k 5[0 2] (77) The slope of I /I versus [Q] should now yield an [0] dependent o value, i.e.: k _9 s l ° P e = k 4[0] + k 5[o 2] - 133 -TABLE 3.6 Quenching Gas Q [ ° 2 ] 0 (torr) Stern-Volmer Quenching Slope (x 10" 1 5cm 3) 1 Slope (x 10 1 5cm 3) [0] 14 -3 (x 10 cm ') Oxygen 0. 1 0. 432 2.3 4.4 0. 1 0. 16 0.86 3.9 0. 1 0. 172 5.8 10.1 0. 1 0. 082- 12.0 23.2 0. 1 0. 066 15.0 18.4 0. 1 0. 158 6.3 9.87 0. 1 0. 098 10 21.9 0. 1 0. 285 3.51 4.4 0. 1 0. 216 4.6 7.1 0. 02 0. 60 1.6 1.8 0. 02 1. 6 0.62 1.13 0. 05 0 58 1.7 2.06 Carbon 0. 1 0 122 8.2 8.95 Dioxide 0. 1 0 188 5.3 4.47 0 1 0 200 5.0 4.40 0 1 0 139 7.2 7.10 0 02 0 313 3.2 1.80 Sulphur 0 1 0 178 5.6 5.8 Hexafluoride 0 1 0 260 3.8 2.8 0 05 0 179 5.6 2.6 0 08 0 .182 5.5 3.75 Carbon 0 .1 0 .15 6.6 4.52 Tet r a f l u o r i d e 0 .1 0 .15 6.6 6.79 0 .02 0 .26 3.8 1.13 0 .05 0 .22 4.5 2.06 - 134 In the case where Q = 0 2 e q u a t i o n (77) y i e l d s the expression k 4[0] [0,1 (78) slope ( I Q / I vs. [0 2]) k 5 + '"2Jo Where [021 i s the o r i g i n a l concentration of oxygen i n the measure-ment of I . o A pl o t of the inverse of the Stern-Volmer quenching slope minus [0 2] versus the oxygen atom concentration i s shown i n Figure 3.7. The slope of Figure 3.7 should (from equation (78)) be equal to k 4 / k 5 and the value of 4 i s i n good agreement with the values of ^ A ^ . obtained i n sections 3.1 and 3.2. In the case of Q / 0 2 one can obtain, from equation (77) , the r e l a t i o n s h i p 1 V ° J . V ° 2 L o S k k q q (79) or k4[o] + k 5 (80) S [ 0 J " k [ 0 J ^ k 1 2 Jo q 2 Jo q where S = slope of I /I versus [Q] p l o t . The r e s u l t s of a p l o t of the l e f t hand side of equation (80) versus [ 0 ] / [ 0 2 ] o for quenching experiments where Q = C0 2, SF^ and CF 4 i s shown i n Figure 3.8. It can be seen from t h i s f i g u r e that - 135 -Figure 3.7 Plot of the inverse of the quenching slope (of the 0 2 Stern-Volmer plots) minus i n i t i a l CL concentration, [CL] , versus the atomic 2 L 2 Jo oxygen concentration. - 136 -Figure 3.8 The dependence of the inverse of the Stern-Volmer quenching slope m u l t i p l i e d by the i n i t i a l CL concentration, [CL] , on -TTP~I 2 2 o [ 0 2 J o for several added gases. A= SF o = c o 2 X = CF 4 ([0 2] q i s indicated f o r each data point.) - 136a -X-0.02T o r i 6 9 o:o?r rM o a I [opel in c JE o X-°-0 5 T •OO-IT c cu 13 O \ 2 A0.08T Q0.1T X - a n i—< 0 A-0.1T I . _i.. [0 ] [ 0 2 ] o - 137 -although the points obtained for [ 0 2 ] Q = 0.1 t o r r are f a i r l y con-s i s t e n t , when [ 0 2 l o i s less than 0.1 t o r r the Stern-Volmer quench-ing slopes do not agree with the proposed mechanism. The p r i n c i p a l conclusion drawn from the above experiments was that there was an uncontrolled v a r i a b l e i n these systems. Ei t h e r Of^S) was being formed by some other reaction, or i t was being quenched by an unknown species. 3.5 TEMPERATURE DEPENDENCE OF THE QUENCHING OF THE 557.7 nm EMISSION BY 0 2 In order to be able to apply the r e s u l t s of laboratory studies to the upper atmosphere, the temperature dependence of the reaction rate constants for the 557.7 nm emission must be known. The tempera-ture of the atmosphere at the maximum emission a l t i t u d e of the 557.7 nm atomic l i n e i s approximately 200K. However, temperature dependent studies of t h i s system pose great experimental problems, the greatest being the determination of the atomic oxygen concentra-t i o n by N02/N0 t i t r a t i o n . Because of the p o s s i b i l i t y of the N0 2 condensing i n a low temperature system, the temperature dependence of the 0 2 quenching e f f e c t was investigated at fixed argon and oxygen atom concentrations. The atomic oxygen flow rate was determined at room temperature (- 300K) and i t s concentration at other tempera-tures, T, was calculated using the equation: 138 -.23 r r i 1 _ 0 flow pressure (atm.) x 6.023 x 1 0 i J x 273K f o . [ 0 J = t o t a l flow X ~ ~ A 4~^3 ~ ( 1 } 2.24 x 10 cm x T(K) Table 3.7 shows the r e s u l t s of oxygen quenching experiments performed between -60°C and +20°C. From the adopted emission inten-s i t y r e l a t i o n s h i p (equation (62)) p l o t s of 1/1(557.7 nm) versus [0 2] should give Intercept _ , ,, Slope " V U J / K 5 at each temperature, or , Intercept 4 / K5 " slope [0] using the Arrhenius equation: k = A exp(-E /RT) - v a where E„ i s the a c t i v a t i o n energy and T the temperature i n k e l v i n , cL one can obtain an expression for the temperature dependence of k 4/k 5: k 4/k g = A 4/A 5 exp(-E a(4) + E &(5)/RT) or l n k /k = ln A /A - ( E a ( 4 ) " E a C 5 ) ) = ln f I n t e r c e P t ^ i K 4 / 5 i 4 / 5 R T in ( g l o p e [ Q ] ) A p l o t of the far l e f t hand side of t h i s expression versus 1/T i s shown i n Figure 3.9. The resultant r e l a t i o n s h i p obtained from the intercept and slope of t h i s p l o t i s - 139 TABLE 3.7 TEMPERATURE DEPENDENCE OF 0 2 QUENCHING OF THE 557.7 nm EMISSION Temperature (K) [0] (x 10 1 4cm" 3) Intercept/Slope of l/l vs. [0 2] (x 10 1 6cm 3) 298 263 253 233 213 6.02 6.71 6.98 7.57 8.29 1.49 1.08 0.94 0.49 0.33 - 140 -Figure 3.9 Arrhenius pl o t of the temperature dependence of the 0 2 quenching p l o t s . (Error bars indicate estimates of the v a r i a t i o n of the intercept/slope at each temperature.) - 140a -( 3 0 0 K ) 1 / TempretureCK~ 1) - 141 -k 4 / k 5 = 3 x IO 3 exp(-1400/T(K)) (82) At 298 K (the temperature at which the previous values of k^/kj. were determined) the k^/k^ r a t i o i s approximately 27, almost an order of magnitude greater than the values obtained previously (~ 5) i n sections 3.1 and 3.2. However equation (82) does con-t a i n large u n c e r t a i n t i e s due to the curvature i n the quenching plots which were discussed i n section 3.1. Nevertheless, the r e s u l t s do indicate that the r a t i o of k^ A ^ becomes smaller at lower temper-atures (upper atmosphere temperatures), i . e . : the 0^ quenching * rate of 0^ becomes greater r e l a t i v e to the quenching by 0. The "temperature dependency" experiments were performed using the jacketed c y l i n d r i c a l observation c e l l of 28 mm i n t e r n a l d i a -meter. As previously mentioned, wall deactivation may be s i g n i f i -cant i n these narrow vessels. Compounding t h i s problem i s the fact that the e f f i c i e n c y of the wall surface (for deactivation of the excited species) has been found to be temperature dependent [147]. It was not possible to vary the temperature of the 5 - l i t r e bulb over a s u f f i c i e n t l y large range to obtain a reasonable temperature dependence. 3.6 SUMMARY OF THE PRELIMINARY RESULTS It appeared from the preliminary i n v e s t i g a t i o n described i n the previous sections that the O^S) emission i n t e n s i t y at 557.7 nm showed the dependency: - 142 -1(557.7 nm) « [0]2[M] [0 9] Such a rate equation i s consistent with the formation of the 0( 1S) state i n a Barth type mechanism: 0 + 0 + M — 0 2 + M (2) 0* + 0 — ^ O ^ S ) + 0 2 (3) •k with 0„ predominantly quenched by 0,,: 0* + 0 2 ^quenched products (5) and OC^S) quenched by oxygen atoms: 0(*S) + 0 — ^ q u e n c h e d products (7) assuming the l i t e r a t u r e values for 0(*S) quenching by 0^ and 0 (Table 1.4). However, inconsistencies and scattered data points were found in the preliminary work. The quenching slopes of 0 2 (and other gases) were found to show appreciable [0] dependencies. Although attempts to explain t h i s e f f e c t as a combination of 0 and 0^ * quenching of the precursor, 0 2 , i n 0 * + 0 ^quenched products (4) 0* + 0 2 — ^ quenched products (5) - 143 -appeared to be successful and fairly consistent, the quenching slopes obtained when different amounts of were passed through the discharge showed large inconsistencies. The reasons for these inconsistencies were not apparent until some new experiments by Slanger were brought to our attention (private communication). These observations and our subsequent experiments are described in the next section. - 144 -CHAPTER 4: RESULTS AND ANALYSIS (PART 2) In Chapter 3, the analysis of the r e s u l t s assumed dominant quenching of 0( 1S) by 0 ( 3 P ) . Slanger's value [93] of 1.8 x 10" 1 1cm" 3s*' 1 f o r k y at 298 K indicated that i n the present experimental system, quenching of Of^S) by Ar and 0^ would be i n s i g n i f i c a n t . In a private communication, Slanger indicated that h i s measurement of k_, was erroneous due to the pre-sence of another quenching species. The only metastable species present i n these systems, i n s u f f i c i e n t l y high concentrations to quench 0(*S) at the required rate^ i s O^fa^A ) [162]. As w i l l be shown i n t h i s chapter, 0 (a A ) was found to be the major quencher of 0(^S) i n t h i s system. Although the dependencies of the 557.7 nm emission on [0] and [Ar] (with constant [O-fa^A )]) were found to be the same as those described i n Chapter 3, these experiments were repeated i n order to obtain quantitative r e s u l t s . The 5 - l i t r e observation vessel was used for the experiments reported i n t h i s chapter, and the apparatus described i n section 2.2.4 was assembled in order to observe the 1.27 ym emission from O^fa^A^). 4.1 VARIATION OF THE [ 0 2 ( a 1 A p ) ] The i n t e n s i t y of the 0_(a 1A ) emission was found to be f a i r l y inde-J 2 V gJ pendent of both the pressure of the system and the microwave discharge power. Increasing the amount of 0^ through the discharge was found to r e s u l t i n a f a i r l y l i n e a r increase i n [0 (a'^A ) ] , but addition of 0 af t e r the discharge resulted i n a decrease of the 0 2(a A ) emission. These observations allowed control of the - 145 -amount of O^a^A^) i n the system, and hence a study of the depen-dence of the 557.7 nm emission on t h i s species. 4.2 0 2 ( a 1 A ) QUENCHING OF 0( 1S) In order to obtain unambiguous observations of the 557.7 nm i n t e n s i t y as a function of [CLCa^A ) ] , the flows of 0 2, Ar and atomic oxygen were kept constant as the concentration of CLCa^A ) was varied. This was accomplished by adding 0 2 both before and af t e r the discharge i n varying amounts, but keeping the t o t a l [0 2] constant. The atomic oxygen concentration was kept constant * by varying the power of the microwave discharge, using the NC>2 emission i n t e n s i t y (when a small constant flow of NO was added to the gas stream) to monitor the oxygen atom concentration. The 1 13 -3 [0 2(a A )] could be adjusted between 1 and 4 x 10 cm . However, even within t h i s limited range, the dependence of the 557.7 nm emission i n t e n s i t y on [0 2(a*A )] i s c l e a r , as can be seen i n Figure 4.1. It i s c l e a r that the predominant quencher of 0(*S) i n t h i s system i s 0 o(a^A ), f o r which one can write the reaction: 1 1 ^ 8 0( S) + 0~(a A ) ^quenched products (8) ^ S This experiment was repeated with several d i f f e r e n t atomic oxygen concentrations and pressures, and the r e s u l t s showed that O^fa^A^) was c l e a r l y the only s i g n i f i c a n t quencher under a l l condi-- 146 -Figure 4.1 The ° 2 ^ g ^ d e P e n d e n c e °f t h e 5 5 7 • 7 n m emission i n t e n s i t y (I) at two oxygen atom 17 -3 concentrations [Ar] = 1.3 x 10 cm - 146a -- 147 -tions for which the 557.7 nm emission was observable i n t h i s d i s -charge flow system. Inspection of e a r l i e r data, aimed at determining the order of the process with respect to oxygen atoms, reveals that there was probably not much v a r i a t i o n of [O^Ca^A )] during the experiments. Hence the observed dependencies appeared to be v a l i d . Nevertheless, the measurements were repeated to confirm the r e s u l t s . 4.3 ATOMIC OXYGEN DEPENDENCE OF THE 557.7 nm EMISSION As mentioned i n section 4.1, the microwave discharge power had l i t t l e e f f e c t on the [O^Ca^A )] and therefore, the atomic oxygen could be varied by changing the power, us u a l l y without a f f e c t i n g 1 the concentration of 02(a A ). At very low and high microwave powers the [0„(a*A )] was adjusted using the discharge by-pass system described i n section 4.2. The atomic oxygen concentration was determined by the NO t i t r a t i o n method described i n section 2.3.3. Figure 4.2 shows a i n I [557.7 nm) versus ln [0] p l o t that i s t y p i c a l of the atomic oxygen dependency r e s u l t s . The slope of the p l o t i s 2 }which indicates that the 0(^S) emission i s depen-dent upon the square of 0 (as was found i n the preliminary r e s u l t s where 02(a^A ) was not monitored). The atomic oxygen dependency 13 -3 was found to be second order at both low (1.1 x 10 cm ) and 13 -3 1 high (3.4 x 10 cm ) concentrations of 0 2 ( a A ). These concentra-- 148 -Figure 4.2 The atomic oxygen dependence of the 1(557.7 nm) emission i n t e n s i t y i l l u s t r a t e d i n a l n - l n p l o t . - 148a -4.4 A 3.8H 3.2H - 149 -t r a t i o n s were at the l i m i t s of the experimental conditions that permitted the v a r i a t i o n of [0] by a f a c t o r of 2 while keeping [0 9(a^A )] constant. The r e s u l t s of these two experiments are given i n Table 4.1. The second order dependence of the 557.7 nm emission at both f a i r l y high and low [(^(a^A )] shows that, within t h i s range of CLfa^A ), the quenching of Of /s) by 0 (reaction (7)) i s unimportant. Also, the second order dependence suggested that the precursor i s predominantly quenched by oxygen atoms i n * C>2 + 0 — ^ q u e n c h e d products (4) otherwise a t h i r d order dependence on [0] would be observed. 4.4 PRESSURE OR M DEPENDENCE OF THE 0( 1S) EMISSION INTENSITY The i n t e n s i t y of the Of^S) emission was measured with 2, 4 and 7 t o r r of argon i n the discharge flow system. Both [(^(a^A )] 13 -3 14 -3 and [0] were kept constant at 2 x 10 cm and 7.3 x 10 cm r e s p e c t i v e l y by adjusting both the oxygen r a t i o (added before and a f t e r the discharge) and the microwave power of the discharge. It was not possible to extend t h i s pressure range and maintain the [O^is^A^)] and [0] at the values indicated above. Figure 4.3 shows that the pressure dependence of the O^S) emission i s l i n e a r with respect to [M] (argon i n t h i s case). This r e s u l t confirms the proposed mechanism (Barth). - 150 -TABLE 4.1 ATOMIC OXYGEN DEPENDENCE OF THE 557.7 nm EMISSION AT DIFFERENT [0 2(a A )] [ 0 2 ( a 1 A g ) ] = 3 i n 1 3 " 3 .4 x 10 cm f [Ar] = 1.3 x < m 1 7 " 3 10 cm .4 x 10 cm 4 I r e l a t i v e (557.7 nm) I absolute (x 10 7cm - 3) NO* i n t e n s i t y (arb. units) [NO] (x 10 1 2cm" 3) [0] (x 10 cm 190 3.0 60 2.1 1.8 170 2.7 58 2.3 1.6 120 1.9 54 2.3 1.5 78 1.2 46 2.7 1.1 54 0.9 40.5 2.6 1.0 [ 0 2 ( a 1 A g ) ] = 13 -3 1.15 x 10 cm f [Ar] = 1.3 x 10 cm m 1 5 " 3 3.6x10 cm 4 t 0 2 l t o t a l = ' 105 1.7 34 2.6 0.83 63 1.0 31 2.6 0.76 54 0.85 27.5 2.6 0.64 31 0.49 22.5 2.7 0.51 24 0.38 19.5 2.7 0.45 - 151 -Figure 4.3 The pressure dependence of the 557.7 nm emission i n t e n s i t y (I) over the range of 2 - 7 t o r r Argon. [ O ^ A ) ] = 2 x 1 0 1 3 -3 14 -3 cm , [0] = 7 x 10 cm - 151a -8 16 [Ar] (cm-3) x lO 1 6 - 152 -4.5 THE COMPLETE RATE EQUATION The preceding sections have shown that the emission i n t e n s i t y of 557.7 nm l i n e from Of^S), formed i n the recombination of atomic oxygen, displays an inverse dependence on the 0 ?(a 1A ) concentra-t i o n , a f i r s t order dependence on the t o t a l p a r t i c l e concentration (or pressure) and a second order dependence on the atomic oxygen concentration. These r e s u l t s may be summarized i n the rate equa-t i o n k [ 0 ] 2 [ M ] 1(557.7 nm) = — (83) [ O ^ a ^ ) ] which i s not consistent with a reaction scheme based on a simple Chapman mechanism. For example, i f the 0(*S) were formed i n reac-t i o n (1) 0 + 0 + 0 — » • 0 ( 1 S) + o 2 (1) and quenched by O^a^A^) : 0( 1S) + 0„(a 1A ) ^quenched products (8) the emission i n t e n s i t y from 0( 1S) ^ 0 ( 1 D ) + hv(557.7 nm) (10) would y i e l d a rate law of k T [ 0 ] 3 1(557.7 nm) = — = [ 0 2 ( a X A g ) ] - 153 -which i s not observed. However, the experimental rate equation given by equation (79) i s consistent with a more complex mechanism i n which an excited oxygen molecule, 0 2 , i s formed i n the recombination of two oxygen atoms with a t h i r d body 0 + 0 + M — ^ 0 2 * + M (2) and transfers i t s energy to another oxygen atom to form the 0(*S) state. 0 2* + 0 »-0(1S) + 0 2 (3) I f the 0 2 i s p r i m a r i l y removed i n the laboratory by c o l l i s i o n s with oxygen atoms k * 4 0 2 + 0 fc. quenched products (4) and the Of^S) i s predominantly quenched by O^a^A ): 0( S) + 0 (a A ) — q u e n c h e d products (8) and r a d i a t i v e decay to the 0(*D) state r e s u l t s i n the emission of the 557.7 nm l i n e , k = x"1 0 ( 1 S ) — ••O^D) + hv(557.7 nm) (10) * 1 then with standard steady state assumptions f o r 0 2 and 0( S) the constant k T i n equation (83) w i l l be equal to a composite of - 154 -the rate constants of reactions (2, 3, 4, 8, 10) i n the form k k T " 1  1 k 4 k 8 In order to evaluate k^ ,, the absolute values of emission i n t e n s i t y and concentrations were measured. 4.6 DETERMINATION OF THE ABSOLUTE EMISSION CONSTANT k T The constant k^ i n equation (83) k [0] 2[M] 1 (557.7 nm) = — (83) [ O ^ a ^ ) ] can be obtained from any or a l l of the experiments described so f a r i n section 4, as long as the absolute values of i n t e n s i t y and con-centration are known. Table 4.2 gives the values of k^ obtained from the data of the experiments which were p r i m a r i l y performed to determine the molecular dependencies of the 0(^S) emission. To average these 1 2 values, 1(557.7 nm) [0 2(a A )] was p l o t t e d versus [0] [M] as shown in Figure 4.4. The slope of t h i s l i n e y i e l d s k T = 2.8 (± 0.3) x 10" 2 7cm 3s _ 1 where the uncertainty i s expressed to 90% confidency l i m i t s . From the proposed mechanism: - 155 -TABLE 4.2 EVALUATION OF THE ABSOLUTE EMISSION INTENSITY CONSTANT k ^ x " 1 l[02(ahg)] k4 k8 [0] 2[M] [0] " 3 i n 1 4 cm x 10 [M] -3 .-17 cm x 10 [ o 2 ( \ ) ] -3 *13 cm x 10 1(557.7 nm) -3 -1 i n 6 cm s x 10 3 -1 ,„-27 cm s x 10 6.4 1.3 1.15 9.4 2.0 I I it ti 6.1 1.8 it it 1.-85 4.4 1.5 t i I I 1.90 4.1 ti it n 2.15 3.8 i i 2.4 1.9 2.35 2.0 4.3 I I I I 2.05 2.4 4.5 I I I I 1.45 4.3 5.7 it I I 1.20 5.7 6.2 17.6 1.3 3.4 30.0 2.5 15.9 tt 11 26.8 2.8 14.8 it it 19.0 2.3 10.6 I I n 12.3 2.9 9.6 I I ti 8.5 2.4 8.3 it 1.15 16.6 2.1 7.6 tt tt 9.9 1.5 6.4 it 11 8.5 1.8 5.1 ti I I 4.9 1.7 4.5 it it 3.8 n 7.3 1.3 2.0 12.3 3.6 tt 0.65 tt 6.2 tt 11 2.3 it 19.4 3.2 5.3 1.3 3.05 2.1 1.8 tt it 2.50 2.4 1.6 it ti 2.10 2.8 i i 11 tt 1.20 6.0 2.0 11 it 1.50 4.4 1.8 tt it 2.15 3.0 it 6.9 tt 1.2 16.6 3.2 11.0 I I 4.55 9.5 2.7 13.7 tt 4.75 21.3 4.1 4.9 ti 2.75 2.8 2.5 11.4 ti 3.0 17.5 3.1 4.3 tt 0.35 10.8 1.6 - 156 -Figure 4.4 Plot of 1(557.7 nm) [0 2(aA )] versus [0] [M] s - 156a -157 T ^A 27 +3 _ l k = f - i = 2.8 x 10 1 cm+ s k 4 k 8 x" 1 = 1.06 s" 1 [25] so that 2 3 n , i n-27 +3 . v = 2 . 6 x 1 0 cm V s of the remaining constants, only k. i s accessible since the other constants concern the formation and removal of a precursor whose i d e n t i t y i s uncertain. In t h i s steady state system kg can only be determined r e l a t i v e to another quenching constant. I f that quench-ing constant i s known then kg can be ca l c u l a t e d . 0^ 1 S s u c n a quencher for which the rate constant kg i n 1 k 9 0( S) + 0 2 ^quenched products (9) i s known [101]. 4.7 DETERMINATION OF k g When 0 2 was added to the gas stream of atomic oxygen, the i n t e n s i t y of the 557.7 nm emission was reduced even though the primary 0(^S) quencher, 0 ?(a*A ) was diminished. It was not possible to change the 0 2 concentration while keeping both the [0] and [0 9(a 1A )] constant. The analysis of these observations was therefore complex. - 158 -1 * Ground state 0 2 could be quenching either 0( S) or the C>2 precursor, i . e . : 0( 1S) + 0 2 quenched products (9) or * 0 2 + 0 2 ^quenched products (5) and the rate equation takes the form 1(557.7 nm) = ( k g [ 0 2 ( a 1 A g ) ] + k 9 [ 0 2 ] ) ( k 4 [ 0 ] + k 5 [ 0 2 ] ) Analysis of the data i n terms of t h i s equation i s very d i f f i c u l t and i t was therefore decided to evaluate the two p o s s i b i l i t i e s (reac-tions (5) and (9)) independently: 1 * 1. If 0 2 i s quenching 0( S) and not 0 2 , the rate equation for the proposed mechanism may be written as T r * 7 > = ^ — 5 tk [0 ( a ^ J ] + k [0 ]} (86) 1(557.7 nm) ^ ^ - 1 [Q] 2 [ M ] 8 2 g 9 1 and a p l o t of 1 [°21 = versus- , 1(557.7 n m ) [ 0 2 ( a 1 A ) ] [ 0 2 ( a A g ) ] should r e s u l t i n a 159 -intercept = ^ ~ k 2 k 3 T _ i [ 0 ] Z [ M ] and slope and hence k 4 k 9 k 2 k 3 T _ 1 [ 0 ] 2 [ M ] intercept , slope " K 8 / K 9 Figure 4.5 shows an example of t h i s p l o t , and the experiment was repeated three times to obtain an average value of slope = kg/kg = 2.6(± 0.3) x IO 3 Since k g = 2.8(± 0.8) x 10 cm s 1 [101], k g = 7.3(± 3.1) x -13 3 -1  10 ^cm 3s (with the error shown as 90% confidence l i m i t s ) . In view of t h i s very large rate constant f o r the quenching of 0(*S) by 0 2(a^A ), i t i s desirable to consider the p o s s i -* b i l i t y that 0 2 i s quenching the precursor 0 2 . 2. The k_ evaluated from Figure 4.5 assumes that the reduction i n i n t e n s i t y of the 557.7 nm emission i s e n t i r e l y due to 0(^S) quenching by 0,,. However, i f there i s a combination of both precursor and 0(^S) quenching by 0 2 then the data y i e l d s a value f o r k c which i s even la r g e r . In view of the fact that o k_ i s already at the upper l i m i t determined by the gas k i n e t i c o c o l l i s i o n r a te, t h i s i s u n l i k e l y . Since i t i s not immediately obvious that a smaller value of kg would not r e s u l t i f 0 2 - 160 -Figure 4.5 Plot of 1/1(557.7 nm)[0 2(a A g )] versus [ 0 2 ] / [ O ^ A )]• - 160a -1 1 1 1 1 1 o 4 e 12 [cy/Kyfcg)] xio-2 - 161 -quenched the precursor, some ca l c u l a t i o n s were performed to i l l u s t r a t e t h i s conclusion. In order to evaluate the possible contribution of reaction ( 5 ) t the expression l k 4[0] + k [o 2 ] 1(557.7 nm){k g(0 2(a 1A )] + k q [ 0 2 ] } k ^ x " 1 [0] 3[M] (87) from equation (85) can be evaluated i f values of k and k n are o y known. Figure 4.6 shows a series of p l o t s of the l e f t hand side of equation (87) versus 0o for d i f f e r e n t values of k Q. The l i t e r a -t i o -13 3 -1 ture value of = 2.8 x 10 cm s i s assumed for 0 2 quenching of 0(*S) d i r e c t l y , and the experimental data i s the same as that used i n Figure 4.5. It can be seen from Figure 4.6 that the slope of the p l o t s where k < 7 x 10 ^ c m + 3 s ^ o -1 3 are negative. Since the slope should be equal to k ^ / k ^ ^ t ^ [0] [M] , and t h i s cannot be negative, the quenching of the precursor by 0 2 must be n e g l i g i b l e i n t h i s system. - 162 -Figure 4.6 I l l u s t r a t i o n of the e f f e c t of d i f f e r e n t values of k c i n the i n t e r p r e t a t i o n of the o 0 quenching of the 557.7 nm emission. - 162a -[0 2] (crrf3) - 163 -CHAPTER 5: DISCUSSION 5.1 THE QUENCHING OF 0( 1S) BY O ^ a 1 ^ ) In Chapter 4, i t was shown that the major loss process of Of^S) i n t h i s study was i t s quenching by 02(a*A ) : 1 1 k8 0( S) + 0_(a A ) m-quenched products (8) with k = 7.3 x 10~ 1 0cm 3s _ 1 at 300 K. This rate constant i s extremely o large compared with most quenching rate constants of Ot^S) and the reaction rate constants of 0~(a^A ) [190]. However, k n i s of the same order 2 g L 8 as the rate constants reported f o r 0(*S) quenching by NO (5 x 10 ^ cm 3s _ 1 [100], 8 x 10" 1 0cm 3s _ : L [99]) and 0 3(5.8 x 10" 1 0cm 3s _ 1 [98]). A l l of these reactions are occuring at rates f a s t e r than the c l a s s i -c a l l y predicted c o l l i s i o n frequency of the two p a r t i c l e s . The extremely e f f i c i e n t quenching of OC^S) by O^fa^A ) can, in a l l l i k e l i h o o d , be att r i b u t e d to an energy transfer from OC^S) 1 1 3 to 0-(a A ). In t h i s t r a n s f e r , the 0( S) relaxes to.the 0( P) ground state, and the O^a^Ag) i s excited to the repulsive portion of one of the three upper Herzberg states (see Figure 1.3). This process can be written as 0(1S) + 0 (a*A ) » 0 ( 3 P ) + 0 o ( A 3 E + , C 3A , cl') ' 2 g 2K u u u followed by 0 o ( A V , C 3A , c V ) ^ 2 0( 3P) 2 V u u u - 164 -The energy between CLCa^A ) „ and the f i r s t d i s s o c i a t i o n l i m i t 6 / 2 g v=0 1 3 of 0 2 i s a near perfect match of the 0( S) 0( P) energy gap. The energy transfer i s therefore a resonance process with very l i t t l e e l e c t r o n i c energy having to be degraded into r o t a t i o n a l and t r a n s l a t i o n a l energy. Furthermore, the Frank-Condon factors f o r t h i s t r a n s i t i o n are excellent. Reaction (81) i s very s i m i l a r to the energy pooling process c a l l e d "energy-disproportionation". An example of t h i s process i s the reaction i n which two 0 7(a 1A ) molecules pool t h e i r energy, r a i s i n g one molecule to the 0 2 ( b 1 Z g ) state and the other relaxes to the ground state [191], i . e . : 0-(a 1A ) + 0 o(a 1A ) » 0 „ ( b 1 E + ) + 0-(X 3Z~) 2 g 2' gJ 2' gJ 2' gJ 5.2 EFFECT OF QUENCHING OF 0( 1S) BY 0 2 ( a 1 A g ) ON  EARLIER INVESTIGATIONS The O-fa^A ) concentration i n the flow system was found to 2K gJ be f a i r l y independent of [0] but strongly dependent on the amount of 0„ entering the discharge. This suggests that 0„(a*A ) i s formed ^ ^ g pr i m a r i l y i n the discharge, rather than from the gas phase recombi-3 nation of 0( P), i . e . : 0 + 0 + M — • • C L ^ A ) + M (88) 2K gJ or on the walls: 165 0 ( 3 p ) w a l l + O C ^ — ^ O ^ a ^ ) (89) 1 3 In e a r l i e r studies [50, 93] of the 0( S) formation from 0( P) recombination , oxygen atoms were formed i n the t i t r a t i o n of atomic nitrogen with NO: N + N O — 0 + N 2 (19) This technique precludes the formation of O^a^Ag) d i r e c t l y i n the discharge but 0^ (a"*"A ) may be formed from the recombination of 3 1 0( P) atoms i n reactions (88) or (89). Since the 0 2 ( a A g) concen-t r a t i o n was not monitored i n the studies of Young and Black [50] and Slanger and Black [93] (and there was no reason at that time why i t should have been), i t i s not possible to say how much O^a^A ) was present i n t h e i r systems. S i m i l a r l y , i n Slanger and Black's [93] studies of the quench-1 3 ing of 0( S) by 0( P) the oxygen atoms were produced i n a N-NO t i t r a t i o n . These workers were aware of the caution required i n thi s experiment to be sure that 0( P) atoms alone were causing the observed increase i n the O^S) quenching rate. To avoid the p o s s i -b i l i t y of varying the concentration of whatever metastable species i n the discharge afterglow might contribute to Of^S) quenching, Slanger and Black varied [0( P)] without changing the discharge c h a r a c t e r i s t i c s , unlike Felder and Young [95] i n 1972. However, 3 Slanger and Black observed that the 0( P) i n t h e i r observation c e l l was only 30 - 50% of the atomic oxygen formed i n the t i t r a t i o n - 166 -(reaction (19)) upstream. The large f r a c t i o n of atomic oxygen that was l o s t before the observation c e l l may have been recombining on the walls or i n the gas phase. A substantial f r a c t i o n of 0^ recombined on surfaces are i n the 0 ?(a 1A ) state [192]. If 30% of the wall recombinations of atomic oxygen lead to the formation of O-fa^A ) i n Slanger and 3 Black's quenching experiment, then for a t y p i c a l value of [0( P)] = 1 3 - 3 3 4 x 10 cm i n the observation c e l l , and a 50% reduction of 0( P) (formed i n the N_N0 t i t r a t i o n ) by wall recombination, the O^fa^A ) +12 -3 concentrations would be about 6 x 10 cm The r a t i o of 1 3 0 o ( a A ): 0( P) would be of the order of 1:7 which i s far greater than the t y p i c a l r a t i o of 1:40 in t h i s study, where 02(a^A g) quenching i s found to be dominant. 3 Even where the wall recombination of 0( P) i s unimportant, the homogeneous formation of O-fa^A ) (reaction (88)) has been estimated [193] to be formed i n 20% of the t o t a l gas phase recom-bination of atomic oxygen. It appears that O^fa^Ag) dominates 0(^S) quenching processes, 1 3 and previous studies of the 0( S) - 0( P) i n t e r a c t i o n need to be rein t e r p r e t e d . There i s no evidence to suggest that the rate con-3 1 stant, k ?, for 0( P) quenching of 0( S) i s f a s t e r than the theore--14 3 -1 t i c a l value of Krauss and Neuman [194], i . e . : 2 x 10 cm s The absence of the 0(*S) atomic l i n e i n spectra of the a f t e r -167 glows of pure 0^ discharges [89] can now be explained. A discharge i n 0 2 would r e s u l t i n very large concentrations of 0 2 ( a 1 A g ) being formed i n the discharge, eliminating Of/s) from the flow system. Since i t would be d i f f i c u l t to design a system containing 3 1 0( P) atoms i n the absence of 0 2 ( a A ), further studies of the Of/s) atom reactions w i l l need to monitor the [0 2(a*A ) ] . 5.3 PROPERTIES OF THE PRECURSOR reaction: A knowledge of the reactions and molecular dependencies of * 1 recursor, 0^ which leads to the formation of 0( S) in the 0 2* + 0 »0( 1S) + 0 2 (3) i s e s s e n t i a l i n the adaptation of the proposed 0(^S) formation mechanism to the upper atmosphere. Since a square dependence of 1(557.7 nm) on [0] and predominant quenching of 0(*S) by 0 2(a^A ) was found in t h i s i n v e s t i g a t i o n . * =f" the 0 2 precursor must be predominantly quenched by atomic oxygen, k4 0 2 +0 — ^ quenched products (4) T The 0( S) formation step (reaction (3)) i s included i n reaction (4), but r e f e r s to s p e c i f i c products, the rate con-stant r a t i o k 3 / k 4 i s unknown but can not be greater than unity. - 168 -to cancel the e f f e c t of one of the three oxygen atoms involved i n the formation mechanism. In previous studies of O^S) formation from the recombination of atomic oxygen [50, 93] the i n t e n s i t y of the 557.7 nm emission was reported to be pressure independent. This r e s u l t was interpreted for a "Barth type" mechanism as being due to predominant removal of 0* by k 6 0 2 + M ^quenched products (6) rather than removal by 0 (reaction (4)) or r a d i a t i v e decay * k l l o 2 i i . o 2 + hv (11) i . e . : k 6[M] » k 4[0] + k n -The r a t i o of 0 to M i n the experiments of Young and Black [50] and Slanger and Black [93] were approximately 1: 6400 - 16000, an order of magnitude greater than the present study (0:M = 1:300). Under these conditions>M quenching of the precursor (reaction (6)) may be predominant. However, t h i s assumption c o n f l i c t s with t h e i r observed square dependence of the 557.7 nm i n t e n s i t y on atomic oxygen atoms i f 0(*S) i s p r i m a r i l y removed by 0 ?(a*A ) i n reaction - 169 -On the other hand, i f the O.-fa^A ) i s being formed i n the homo-geneous recombination of atomic oxygen ( t a c t i o n (88)) then the con-centration of 0„(a^A ) i n these systems would be increasing with 2 g pressure, i . e . : [ O ^ a ^ g ) ] - [ 0] 2[M] This would tend to cancel out any pressure dependence of the emission i n t e n s i t y , and would r e s u l t i n the observed dependencies of Young and Slanger, i . e . : 1(557.7 nm) *- [ 0 ] 2 * I f , as t h i s study p r e d i c t s , the 0 2 metastable i s predominantly 3 quenched by 0( P) i n reaction (4) and i s formed i n the recombma-3 t i o n of two 0( P) atoms i n the presence of a t h i r d body i n reac-t i o n (2), then the emission i n t e n s i t y f o r t h i s precursor, 0 2* » - 0 2 + hv (11) should show the following dependence: I(0 2*) <* [ 0][M] (91) The search for an 0(^S) precursor then reduces to one with (a) s u f f i c i e n t energy and (b) an emission i n t e n s i t y which i s f i r s t order i n 0 and f i r s t order i n M. * The 0 2 formed i n reaction (2) could i n p r i n c i p l e be any of - 170 -the s i x states of 0^ with a t t r a c t i v e p o t e n t i a l s . However, i n the case of 0-(X 3E , a 1A and b 1 Z + ) very large amounts of v i b r a t i o n a l 2 g g g energy would need to be converted into e l e c t r o n i c energy i n reac-ti o n (3) to produce 0(^S). Such conversions are possible i f they are favoured by Frank-Condon fac t o r s , however the Frank-Condon factors are extremely poor f o r a t r a n s i t i o n from those states which could t r a n s f e r s u f f i c i e n t energy to excite Of^S). Also, no e v i -dence has been found for s i g n i f i c a n t concentrations of these states i n l e v e l s above v = 1 i n these systems, and hence the mole-cular dependencies of these l e v e l s have not been measured. 3 + 3 1 -The three higher bound states of 0„(A E , C A and x E ) are more 6 2^ g 5 u uJ l i k e l y candidates for the precursor, because of the more favourable Frank-Condon factors i n a t r a n s i t i o n that could excite Of^S). A l l v i b r a t i o n a l l e v e l s of these upper states with the exceptions of O^c^E ) v = 0 and 1 have enough energy to excite Of /s) i n a t r a n s i -3 _ t i o n to ground state 02(X E ) v = 0. However, t r a n s i t i o n s from higher v i b r a t i o n a l l e v e l s (v' > 3) have more favourable Frank-Condon fa c t o r s . Unfortunately, the gas phase emission dependencies of these states have only been p a r t i a l l y investigated. The Herzberg I emission system from 02(A 3E*) has been reported [119] to have the i n t e n s i t y dependence: 2 I (Herzberg I) oL [ 0 ] [ Q [ f (92) 171 -,under s i m i l a r conditions of pressure and oxygen atom concentrations as t h i s study. This i s i n agreement with the work of McNeal and Durana [109] who found I (Herzberg I) t*- [ 0 ] 2 i n a pure 0^ system ( i . e . : M = 0 2 ) . These r e s u l t s are consistent with the predominant quenching of 0 2(A 3Z*) by 02-. .3„+ 0„(A Z ) + 0N fc. quenched products (93) 2 u I 0 2(A 3Z*) has also been studied [112] following i t s formation on a n i c k e l surface, and the rate constants f o r the quenching of 3 + -12 3 -1 0 2(A £ ) by 0, 0 2 and Ar reported as 9 x 10 cm s , 2.9 x 10 1 3cm 3s 1 and 8.6 x 10 ^cm 3s 1 r e s p e c t i v e l y . These values indicate that under the gas phase conditions of Kenner et a l . [119] oxygen atom quenching rather than the reported molecular oxygen quenching of 0 ? ( A 3 Z + ) should be the dominant process. There i s no obvious explanation of the discrepencies between the gas phase and surface catalyzed formation studies of 0 2(A 3Z*). There have been even fewer k i n e t i c i nvestigations of O-t^Z ) to 2 u ^ Z ^ u' for these states although Kenner et a l . [119] have reported the dependence of the r a t i o of the Herzberg I to Herzberg II band and 0„(C 3A ) than 0 o ( A 3 + ) . No rate constants have been published 2 u £  1 - 3 emissions. The Herzberg II emission from 0 2 ( c Z u X Z ) was - 172 -was found to increase with respect to Herzberg I as the pressure increased and oxygen atom concentration decreased, i . e . : I (Herzberg II) ^ _[M]_ f q 4 1 I(Herzberg I) [0] J From the dependence o f I(Herzberg I) found i n the same study (equa-t i o n (89)) one obtains: 2 1 (Herzberg II) oC (95) Kenner et a l . [119] interpreted t h i s dependence of the 0 2 ( c ' E ^ , v = 0) emission as i n d i c a t i n g a precursor being formed i n the o r i g i n a l recombination of atomic oxygen followed by deactiva-t i o n to t h i s state by c o l l i s i o n with a t h i r d body (M), i . e . : 0 + o + M H » 0 o ( C V , v = 0) (96) 2 2 u and the precursor 0^ predominantly removed by either 0 or 0 2 quenching 0 2 + 0 — ^ q u e n c h e d products (97) 0 2 + 0 2 —^quenched products (98) leaving the 0 2 ( C * Z J v = 0) having to be quenched by 0 2 or 0: 0 2 ( c 1 I u , v = 0)+ 0 2 — ^ q u e n c h e d products (99) 0 2 ( c l l f > v = 0)+ 0 — ^ q u e n c h e d products (100) to s a t i s f y the observed dependence. It i s possible that the 0 2 - 173 -i n r e a c t i o n (96) i s the precursor to the O^S) i f quenching by 0 in reaction (97) i s i t s predominant loss process. 0^ therefore could be a higher v i b r a t i o n a l l e v e l of the O-fc^E ) state. & 2 u As previously mentioned, the (^(c^E" , v = 0)that was studied by Kenner et a l . does not have enough energy to excite O^S). However, the Frank-Condon factors appear to increase f o r higher v i b r a t i o n a l levels of 0 o ( c 1 E ) i n t r a n s i t i o n s to the 6 2 u ground state. There have been no reported observations of O-fc^Z ) emissions from above the f i r s t v i b r a t i o n a l l e v e l (v' = 1). 2 u 3 Slanger (private communication) has indicated that the 0^(0 A ) emissions show the same molecular dependencies as those of 0^ (A 3E*), however there have been no published r e s u l t s of how emission 3 from 0 2(C A ) varies with [0] or [M]. At present there i s no experimental evidence a v a i l a b l e to * 1 3 + p o s i t i v e l y i d e n t i f y theQ precursor to 0( S), although 0 9(A E ) 3 (and possibly O^CC A u )3 does not appear to be consistent with the predominant 0 quenching of the precursor observed i n t h i s study. * The f i n a l i d e n t i f i c a t i o n of 0n must await the i n v e s t i g a t i o n of the complex energy t r a n s f e r and quenching processes of oxygen molecules f o r each state and v i b r a t i o n a l l e v e l . - 174 -5.4 ATMOSPHERIC IMPLICATIONS OF THIS INVESTIGATION 5.4.1 0^(a.^L\ ) Quenching of 0(^S) in the Upper Atmosphere The 02(a^A g) density a l t i t u d e p r o f i l e s have been measured, using the 1.27 ym emission of the (0 - 0) band, i n the dayglow [195, 196], aurora [197] and nightglow [39, 196]. The p r o f i l e of ^ ( a ^ A )] i n the dayglow and nightglow i s shown i n Figure 5.1. In t h i s i n v e s t i g a t i o n the rate constant k was measured at 300K, whereas the temperature of the upper atmosphere i n the region of maximum O^S) emission i s approximately 200K. Since the 02(a^A ) quenching rate constant was found to be of the order of c o l l i s i o n frequency, i t must have an i n s i g n i f i c a n t a c t i v a t i o n energy and there-fore i t i s u n l i k e l y to be s i g n i f i c a n t l y slower at the lower upper atmosphere temperatures. Using t h i s assumption, the quenching of 0(^S) by 02(a''"A ) becomes competitive with i t s t o t a l r a d i a t i v e -1 1 9 - 3 decay (1.11 s [25]) when the [Op^ Ca A )] i s greater than ~ 10 cm Although Op (a'* A ) d e n s i t i e s of up to 4 x lO^cm 3 are observed in the dayglow [196], at the 97 km a l t i t u d e of maximum 0(^S) 9 - 3 1 emission, the density i s ~ 1 0 cm . This suggests that 0„(a A ) quenching of 0( S) i n dayglow i s of the same magnitude as i t s rad i a -t i v e decay, and should be considered as a s i g n i f i c a n t loss process of 0(*S) in the dayglow. Enhancement of the Opta^Ag) (0 - 0) 1.27 ym emission i n - 175 -Figure 5.1 The a l t i t u d e dependence of [0" 2(aA )] (mea-sured at 1.27 um) i n the dayglow [196] and nightglow [39]. - 175a -1201 E 110-[ 0 2 ( a 1 A g ) ] ( c m " 3 ) - 176 -aurora has been reported by Noxon [197], who recorded i n t e n s i t i e s 1 of up to 1 MR. This suggests that concentrations of Opfa^A ) may 9 - 3 1 be s i g n i f i c a n t l y greater than 10 cm in some aurora. Op(a A ) may play an important r o l e i n the quenching of O^S) i n these phenomena and the 1.27 ym emission should be monitored concurrently when the 557.7 nm emission i s measured i n aurora. In the nightglow, the 0p(a*A ) density i s approximately 1.5 x 8 - 3 10 cm [39] at 95 km. Even with the quenching rate constant mea-sured at 300K, the Of^S) would only be removed at a rate of 0.075 s ^ (compared to 1.11 s * for r a d i a t i v e decay). Hence 0p(a^A ) i s not a major quencher of 0(^S) i n the nightglow. 5.3.2 0(^S) Formation and Loss Processes i n the Nightglow In previous studies of the e x c i t a t i o n of the 557.7 nm emission the model adopted for c o r r e l a t i n g the observed emission and mea-sured atomic density p r o f i l e has been by either the one step Chapman process or the two step Barth mechanism. Since there i s no laboratory or atmospheric evidence for the e x c i t a t i o n of 0(^S) in a Chapman reaction, only the Barth mechanism (reactions (2) and (3)) w i l l be considered for the formation of 0(^S) in the upper atmosphere i n the same way i t i s found to be responsible for 0(^S) e x c i t a t i o n i n t h i s i n v e s t i g a t i o n . The d e a c t i -* 1 vation processes of both Op and 0( S) must be included in an analysis of 0(^S) emissions i n the upper atmosphere. - 177 -At the low pressures of the upper atmosphere the r a d i a t i v e deactivation of metastable species may be comparable or greater than the c o l l i s i o n a l deactivation frequency (depending upon the l i f e t i m e of the metastable). In the modelling of these atmos-pheric emissions one must be c a r e f u l to include a l l deactivation processes, some of which may be minor loss processes under labora-tory conditions. * In the case of 0, we include the loss processes: 0 2 + O ^ P ) — » » quenched products (4) * k5 O2 + 0n ^ quenched products (5) V O2 + N2 quenched products ( 6 ' ) k - x"1 * 11 2 0 2 — 0 2 + hv (11) and for Of^S) loss by 1 k7 0( S) + 0 ^ quenched products (7) 1 i k o 0( S) + 0 (a A ) 2 — ^ quenched products (8) 1 k 9 0( S) + 0 2 quenched products (9) 1 T " 1 1 0( S) — • — ^ 0 ( D) + hv(557.7 nm) (10) The i n t e n s i t y of the Of^S) emission w i l l be given by the equation 178 • k k [0] 3[M] 1(557.7 nm) = -=-= r-{1 + T [k ?[0] + k 8 [ 0 2 ( a x A g ) ] + k q [ 0 2 ] ) } x ( i " 1 + k 4 [ 0 ] + k 5 [ 0 2 ] + k^ [N 2]} (101) In the c a l c u l a t i o n s of Slanger and Black [107] and Thomas et a l . [127] the major loss process of 0(*S) was assumed to be by 0( P) quenching i n reaction (7). Also, the temperature dependent rate constant k k k or = 1.4 x 10" 3 0exp(-130/RT) C m 6 s - 1 lc ' that was used i n the atmospheric modelling was calculated from the 3 r e s u l t s of Slanger [93] using the erroneous 0( P) quenching rate constant for r e a c t i o n (7) of k ? = 5.0 x 10" exp(-610/RT) cm s" Since O-fa^A ) was not measured i n the experimental determina-2 g t i o n of K 2k^/k , the value obtained i s probably inc o r r e c t and cannot be used in atmospheric modelling of 0(*S). 3 1 With the assumption of dominant 0( P) quenching of 0( S) removed, the other species present i n the nightglow should be con-sidered. Figure 5.2 shows the height p r o f i l e s of the major atmos-pheric constituents of the nightglow. Using the accepted rate con-stants for 0(*S) quenching by these molecules' at 200K [198] (with 3 1 the exception of 0( P) quenching) the loss rates of 0( S) with - 179 -V Figure 5.2 The a l t i t u d e p r o f i l e s of the major c o n s t i -tuents i n the region of the t e r r e s t r i a l nightglow. (Modified from a drawing by L. Thomas and M.R. Bowman, J. Atmos. Terr. Phys., 34_ (1972) 1843 and incorporating the data of Evans et a l . [39] ( O p ^ A ) ) , Dickinson et a l . [2] (0), and the C.I.R.A. Mean Reference Atmosphere 1972 (M).) - 179a -- 180 -respect to a l t i t u d e may be calcu l a t e d . Figure 5.3 shows that above 85 km, r a d i a t i v e deactivation i s the predominant removal process of 0(*S) in the nightglow, with 0 2 quenching becoming more impor-tant at lower a l t i t u d e s . Equation (101) i s now s i m p l i f i e d to k k [0] 3[M] 1(557.7 nm) = {1 + x k Q[0 2]}{T2 + k 4 [ 0 ] + k 5 [ ° 2 ] + V [ N 2 ] } ^ 1 0 2 ^ The only information that i s known (from t h i s study) about the * quenching of 0 2 , i s that i t i s p r i m a r i l y quenched by atomic oxygen (reaction (4)). The 0: 0 2: M r a t i o i n t h i s study was usually about 1: 30: 300, which i s a much smaller r e l a t i v e concentration of atomic oxygen than i n the nightglow 5where i t i s for example: 0: 0 2: M = 1: 0.2: 2 at 120 km 0: 0 2: M = 1: 17: 80 at 90 km * Unless the r a d i a t i v e l i f e t i m e of the 0 2 metastable i s short, or the r e l a t i v e quenching e f f i c i e n c i e s of 0, 0 2 and N 2 i n reactions (4) (5) and (6') have a very large temperature dependence, the 3 major quencher of the precursor would be 0( P) i n t h i s region of the nightglow. Equation (102) may now be s i m p l i f i e d to k 2 k 3 [ 0 ] 2 [ M ] I ( 5 5 7 ' 7 ^ = k 4 ( l + T k 9 [ 0 2 ] ) • Table 5.1 shows the atomic oxygen, molecular oxygen and t o t a l - 181 -gure 5.3 Loss rates of 0( S) i n the n i g h t g i (modified from a drawing by Thomas a l . [125]). - 181a -r - 182 -TABLE 5.1 A l t i t u d e 1 Temp [o 2]* [M]* * * [0] 1(557.7 nm) ** (km) CK) (cm-3) (cm ^ (cm ) _3 (photons cm ) 110 245 2. 5 (11* 2.2 (12) 3.5 (11) 0.75 105 216 6 6 (11) 4.8 (12) 5.0 (11) 12 100 199 2 0 (12) 1.1 (13) 7.5 (11) 50 95 190 5 5 (12) 2.8 (13) 6.2 (11) 70 90 183 1 4 (13) 7.1 (13) 4.0 (11) 10 85 185 3 .5 (13) 1.7 (14) 5 (10) < 1 * from C.I. R.A. 1972 Mean Reference Atmosphere ** from Thomas et a l . [127] * 2.5 (11) = 2.5 x 1 0 U - 183 -p a r t i c l e concentrations as a function of a l t i t u d e and temperature in the region of 85 - 110 km. Using these quantities i n equation (103) and k g = 4.8 x 10" 1 2exp(-850/T) for On quenching of Of^S) [198], T = 0.94 s [25], one can calculate the values of k^k^/k^ that would be required to obtain the observed i n t e n s i t i e s of the 557.7 nm emission (Table 5.2). These rate con-stants are not consistent with a normal Arrhenius temperature depen-dence. This could be due to the occurrence of other loss processes 1 * for e i t h e r 0( S) or the 0 2 precursor i n the upper atmosphere. A d e t a i l e d model equating the 557.7 nm emission to the number den s i t i e s of the constituents of the nightglow must await the iden-t i f i c a t i o n and ch a r a c t e r i z a t i o n of the precursor. In t h i s study we have found k k -^-3- = 1.9(± 1.0) x 10" 3 6cm 6s _ 1  k4 at 298K, t h i s value i s not inconsistent with the required atmospheric values (Table 5.2). However, i t cannot be used in upper atmospheric models because: 1. In the laboratory the t h i r d body M in reaction (2) i s argon, whereas i n the upper atmosphere the major constituents are - 184 -TABLE 5.2 A l t i t u d e Temp (K) [0] 2[M] (1 + T k 9 [ 0 2 ] ) W k 4 110 245 2.7 (35)* 1.04 2.9 (-36) 105 216 1.2 (36) 1.06 1.1 (-35) 100 199 6.2 (36) 1.13 9.1 (-36) 95 190 1.1 (37) 1.30 8.3 (-36) 90 183 1.1 (37) 1.65 1.5 (-36) 85 185 4.2 (35) 2.70 < 6.4 (-36) 2.7 (35 ) = 2.7 x 10 - 185 -Np, Op and 0. The rate of a three-body reaction has been shown [199] to be dependent upon the nature of the t h i r d body. Table 5.3, which i s a modification of the table presented by Kaufman [199], shows the r e l a t i v e "M e f f e c t " of several molecules in three-body reactions s i m i l a r to reaction (2). It i s possible that the t h i r d body i n reaction (2) i s s p e c i f i c to the forma-* 1 t i o n of the Op precursor required to excite 0( S). 2. The temperature dependence of kpk^/k^ has not been determined i n t h i s study and t h i s i s required to evaluate the observed emission i n t e n s i t y at 557.7 nm i n the nightglow with respect to the atomic oxygen concentration. 186 -TABLE 5.3 RELATIVE M EFFECT IN 3-BODY REACTIONS REACTION THIRD-BODY [M] Ar : He : Hp : Np : COp : NpO : HpO 0 + 0 + M *-0p + M 1.0 : 0.8 : 2.9 : 1.9 N + 0 + M ^NO + M 1.0: 0.5: 3 . 0 : 1 . 3 : 4.0: 3.0: N + N + M »»Np + M 1.0: 1.9: 1.8: 1.0: 1.0: 0.9: 2.7 Ar : He : Np : Op : COp : NpO : SF 6 : H 20 0 + Op + M ^ 0 3 + M 1:0 : 0.8 : - : 1.1 : 3.4 1.0 : 1.0 : 1.4 : 1.6 : 3.7 : 3.7 : 8.5 : 1.5 1.0 : 0.8 : — : - : 5.0 : 4.2 : 1.0 : 2.5 : 2.2 Modified from Kaufman [199] - 187 -CHAPTER 6: CONCLUSIONS The major conclusions are: 1. In the laboratory, OC^S) i s formed i n an energy trans f e r mechanism. The i n i t i a l r e action involves the recombination of two oxygen atoms in the presence of a t h i r d body forming a metastable oxygen molecule. This i s followed by energy transf e r to another oxygen atom r e s u l t i n g i n the formation of OC^S). Although the metastable oxygen intermediate remains to be i d e n t i f i e d , under the conditions used i n t h i s study, i t i s p r i m a r i l y quenched by atomic oxygen. 2. No evidence i s found for the formation of 0(*S) i n the d i r e c t termolecular reaction of three oxygen atoms. 3. Opfa^Ag) i s the p r i n c i p a l quencher of 0(^S) i n the laboratory. The rate constant f o r t h i s process i s equal to 7 ± 3 x 10 ^ cm 3s - 1 at 300 K. 4. O-Ca^A ) quenching i s not a major loss process f o r Of^S) i n the t e r r e s t r i a l nightglow compared with i t ' s r a d i a t i v e decay. 5. The rate equation governing the formation of 0(^S) i n the laboratory i s given by <LL2lkl= k [ 0 ] 2 [ A r ] dt where k = 1 . 9 ± l x l 0 3^cm^s 188 -APPENDIX 0( 3P) + 0( 3P) + 0( 3P) 0( 3P) + 0( 3P) + M * 7 o 2 + orp) o * + 0( 3P) °2 + 0 2 °2 + M oc1s) + O( 3P) O ^ S ) + 0 2 ( a 1 A g ) OC S) + o 2 or^s) 0(1S) + o. 0 + M 2 oc s) + o 2 CD (2) (3) quenched products (4) quenched products (5) quenched products (6) quenched products (7) quenched products (8) quenched products (9) 0C1D) + hvf557.7 nm) (10) 0 2 + hv c m - 189 -REFERENCES 1. D.M. Packer, Ann. Geophys., 17_ (1961) 145. 2. P.H. Dickinson, R.C. Bolden and R.A. Young, Nature, 252 (1974) 291. 3. E.I. Reed and S. Chandra, J . Geophys. Res., 80 (1975) 3053. 4. A.J. Angstrom, Reserches sur le Spectre S o l a i r e , W. Shultz, Upsalla, (1868). 5. J.C. McLennan and G.M. Shrum, Proc. R. S o c , A108 (1925) 501. 6. S. Chapman, Proc. R. S o c , A132 (1931) 353. 7. C. Barth, J . Geophys. Res., 67_ (1962) 1628. 8. C.N. Banwell, Fundamentals of Molecular Spectroscopy, McGraw-H i l l , (1966) p. 166. 9. F. Hund, Z. Phys., 42_ (1927) 93. 10. A. Rabinowitz, Z. Phys., 53_ (1929) 267. 11. J.C. McLennan, Proc. R. S o c , A120 (1929) 327. 12. A. Rabinowitz, Z. Phys., 61 (1930) 338. 13. R. Frerichs and J.S. Campbell, Phys. Rev., 36 (1930) 1460. 14. E.U. Condon, Astrophys. J . , 79 (1934) 217. 15. S. Pasternack, Astrophys. J . , 9_2 (1940) 126. 16. T. Yamanouchi and H. Horie, J . Phys. Soc. Jpn., 7_ (1952) 52. 17. R.H. Garstang, Mon. Not. R. Astron. S o c , 111_ (1951) 115. 18. R.H. Garstang. In E.B. Armstrong and A. Dalgarno (eds.), "The Airglow and Aurora", Pergamon Press, New York, (1956). 19. C. Nicholaides, 0. Siranoglu and P. Westhaus, Phys. Rev. A4 (1971) 1400. 20. A. Omhold, i n "The Airglow and Aurora" (reference 18). - 190 -21. F.S. Le Blanc, 0. Oldenberg and N.P. Carleton, J . Chem. Phys. 45 (1966) 2200. 22. J.W. McConkey, D.J. Burns, K.A. Moran and K.G. Emeleus, Phys. Lett- 22 (1966) 416. 23. J.W. McConkey and J.A. Kernahan, Planet. Space S c i . , 17 (1969) 1972. 24. A. Corney and O.M. Williams, J . Phys. B., 5_ (1972) 686. 25. J.A. Kernahan and P.H. Pang, Can. J . Phys., 5_3 (1975) 455. 26. G. Kvi f t e and L. Vegard, Geophys. Norv. (Oslo), 17 (1947) 3. 27. W. Stoffregen and H. Derblom, Nature, 185 (1960) 28. 28. R.S. Mulliken, Rev. Mod. Phys., 4 (1932) 1. 29. CA. Coulson, Valence, Oxford Univ. Press, (1961). 30. P.H. Krupenie, J . Phys. Chem. Ref. Data, 1_ (1972) 423. 31. J.W. E l l i s and H.O. Kneser, Z. Phys., 86 (1933) 583. 32. L. Herzberg and G. Herzberg, Astrophys. J . , 105 (1947) 353. 33. H.P. Gush and H.L. Buij s , Can. J . Phys., 42: (1964) 1037. 34. A. Vallance Jones and A.W. Harrison, J . Atmos. Terr. Phys., 1_3 (1958) 45. 35. R.M. Badger, A.C. Wright and R.F. Whitlock, J . Chem. Phys., 43 (1965) 4345. 36. H.D. Babcock and G. Herzberg, Astrophys. J . , 108_ (1948) 167. 37. A.B. Meinel, Astrophys. J . , 112 (1950) 464. 38. , Astrophys. J . , 113 (1951) 583. 39. W.F.J. Evans, E.J. Llewellyn and A. Vallance Jones, J . Geophys. Rev., 77_ (1972) 4899. 40. G. Herzberg, Naturwissenschaften, 20_ (1932) 577. 41. J . Dufay, CR. Acad. S c i . ( P a r i s ) , 21_3 (1941) 284. 42. G. Herzberg, Can. J . Phys., 30 (1952) 185. - 191 -43. J.W. Chamberlain, Astrophys. J . , 121_ (1955) 277. 44. H.P. Broida and A.G. Gaydon, Proc. R. Soc. (London), A222 (1954) 181. 45. CA. Barth, Ph.D. Thesis, Dept. of Physics, U.C.L.A., (1958). 46. CA. Barth and J . Kaplan, J . Chem. Phys., 26^  (1957) 506. 47. , J . Mol. S p e c t r o s c , 3_ (1959) 583. 48. R.A. Young and R.L. Sharpless, J . Chem. Phys., 39 (1963) 1071. 49. V.I. Krassovsky, N.N. Shefov and V.I. Yarin, Planet. Space S c i . , 9 (1962) 883. 50. R.A. Young and G. Black, J . Chem. Phys., 44 (1966) 3741. 51. V. Hasson, R.W. N i c h o l l s and V. Degen, J . Phys. B., 3_ (1970) 1192. 52. T.G. Slanger, J . Chem. Phys., 69_ (1978) 4779. 53. G. Herzberg, Can. J . Phys., 31_ (1953) 657. 54. V. Degen, Can. J . Phys. 46 (1968) 783. 55. G.M. Lawrence, CA. Barth and V. Argabright, Science, 195 (1977) 573. 56. V.A. Krasnopolsky, A.A. Krysko, V.N. Rogachev and V.A. Parshev, Kosm. Is s l e d . , 14 (1976) 789. 57. I.M. Campbell and B.A. Thrush, Proc. R. S o c , A296 (1970) 201. 58. R.P. Saxon and B. L i n , J . Chem. Phys., 67_ (1977) 5432. 59. J.W. Chamberlain, Astrophys. J . , 128_ (1958) 713. 60. J.F. Noxon, Can. J . Phys., 39 (1961) 1110. 61. Aurora and Airglow, M. McCormac (ed.), Proc. N.A.T.O. Adv. St. Inst. (1966). 62. 0. Struve, B u l l . Acad. St. Petersbourg, 1_3 (1869) 49. 63. T.M. Donahue, E.C. Zipf and T.D. Parkinson, Planet. Space S c i . , 18 (1970) 171. - 192 -64. A. Dalgarno and S.P. Khare, Planet. Space S c i . , 15_ (1967) 938. 65. E.C. Zipf and A. Vallance Jones, i n "Aurora" (Reidel Dordrecht, Holland), (1974). 66. M.H. Rees, A.I. Stewart and J.C.G. Walker, Planet. Space S c i . , 17_ (1969) 1997. 67. T.M. Donahue, T.D. Parkinson, E.C. Zi p f , J.P. Doering, W.G. Faster and R.E. Muler, Planet. Space S c i . , 1_6 (1968) 737. 68. W. Henricksen, J . Atmos. Terr. Phys., 37 (1978) 1173. 69. M.R. Torr, D.G. Torr, R.A. Hoffman, W.B. Hanson, J.H. Hoffman, W.K. Peterson and J.C.G. Walker, Planet. Space S c i . , 23 (1975) 1669. 70. T.D. Parkinson and E.C. Zi p f , Planet. Space S c i . , 1_8 (1970) 895. 71. K. Henricksen, Planet. Space S c i . , 21_ (1973) 863. 72. C E . Watson, V.A. Dulock, R.S. S t o l a r s k i and A.E.S. Green, J. Geophys. Res., 72 (1967) 3961. 73. A.W. Yau and G.G. Shepherd, Planet. Space S c i . , 27. (1979) 481. 74. B.H. Solheim and E.J. Llewellyn, Planet. Space S c i . , 27_ (1979) 473. 75. J.F. Noxon and R.M. Goody, J . Atmos. S c i . , 10 (1962) 342. 76. L. Wallace and M.B. McElroy, Planet. Space S c i . , 14_ (1966) 677. 77. R.C Shaeffer, P.D. Feldman and E.C. Z i p f , J . Geophys. Res., 77 (1972) 6828. 78. L. Yntema, Publ. Ast. Groningen, 22 (1909) 1. 79. W.W. Campbell, Astrophys. J . , 2_ (1895) 161. 80. Lord Rayleigh (R.J. S t r u t t ) , Proc. R. Soc. (London), A106 (1924) 117. 81. F.E. Roach and M.B. P e t t i t , J . Geophys. Res., 56 (1951) 325. 82. L. Vegard, P h i l . Mag., 46 (1923) 193, 577. - 193 -83. J.C. McLennan and G.M. Shrum, Proc. R. S o c , A106 (1924) 138. 84. R. d'E. Atkinson, Proc. R. S o c , A106 (1924) 429. 85. R.A. Young and K.C. Clark, Phys. Rev. Lett., 5_ (1960) 320. 86. F.E. Roach, J.W. McCaulley, E. Marovich and CM. Purdy, J . Geophys. Res., 65_ (1960) 1503. 87. E. Tandberg-Hassen and F.E. Roach, J . Res. Natl. Bur. Stand., 630 (1959) 319. 88. CA. Barth, Science, 134_ (1961) 1426. 89. CA. Barth and A.F. Hildebrandt, J . Geophys. Res., 66 (1961) 985. 90. CA. Barth, Ann. Geophys., 20 (1964) 182. 91. M. Gadsen and E. Morovich, J . Atmos. Terr. Phys., 31_ (1969) 817. 92. G. Kvi f t e and L. Vegard, Geophys. Norv., 17_ (1947) 1. 93. T.C Slanger and G. Black, J . Chem. Phys., 64_ (1976) 3763, 3767 94. 100. T.C Slanger, G. Black and B.J. Wood, J . Phys. Lett. , 17_ (1972) 401. 95. W. Felder and R.A. Young, J . Chem. Phys., 56 (1972) 6028. 96. R.A. Young, G. Black and T.C Slanger, J . Chem. Phys., 5£ (1969) 309. 97. T.C Slanger and G. Black, J . Chem. Phys., 65 (1976) 2025. 98. G. London, R. G i l p i n , H.I. S c h i f f and K.H. Welge, J . Chem, Phys., 54 (1971) 4512. 99. S.V. F i l s e t h , F. Stuhl and K.H. Welge, J . Chem. Phys., 52 (1970) 239. R. Atkinson and K.H. Welge, J . Photochem., 1_ (1972/3) 341, 101. , J . Chem. Phys., 57_ (1972) 3689. 102. R.A. Young and G. Black, Planet. Space S c i . , 1_4 (1966) 113. 103. F.W. Bingham, A.W. Johnson and T.K. Rice, J . Chem. Phys., 65_ (1976) 1663. - 194 -104. T.G. Slanger and G. Black, Planet. Space S c i . , 25 (1976) 79. 105. T.G. Scholtz and D. Offermann, J . Geophys. Res., 79 (1974) 307. 106. T.M. Donahue, B. Guenther and R.J. Thomas, J . Geophys. Res., 78 (1972) 6662. 107. T.G. Slanger and G. Black, Planet. Space S c i . , 25 (1977) 79. 108. M.N. Vlasov. Geomagn. Aeron. (Engl. Transl.) 9 (1969) 719. 109. R.J. McNeal and S.C. Durana, J . Chem Phys., 51 (1969) 2955. 110. G. Manilla and P. Harteck, J . Chem. Phys., 34 (1961) 2177. 111. P. Harteck and R.R. Reeves J r . , Discuss. Faraday S o c , 37 (1964) 82. 112. R.D. Kenner and E.A. Ogryzlo, Int. J . Chem. Kinet., 12_ (1980) 501. 113. J.L. Richards and P.M. Johnson, J . Chem. Phys., 65 (1976) 3948. 114. J. Goodman and L.E. Brus, J . Chem. Phys., 67 (1977) 1482. 115. T.G. Slanger, Chem. Phys. Lett. , 66 (1979) 344. 116. T.G. Slanger and D.L. Huestis, i n press (1980/1). 117. A.L. Broadfoot and K.R. Kendall, J . Geophys. Res.,'73 (1968) 426. 118. E.J.Llewellyn, B.H. Solheim, G. Witt and R.G.H. Greer, J . Photochem., 12 (1980) 179. 119. R.D. Kenner, E.A. Ogryzlo and S. Turley, J . Photochem., 10_ (1979) 199. 120. F.E. Roach, L.R. M e g i l l , M.H. Rees and E. Morovich, J . Atmos. Terr. Phys., 12 (1958) 171. 121. E.R. Manring and H.B. P e t t i t , J . Geophys. Res., 63 (1958) 39. 122. J.H. Brandy, Can. J . Phys., 41_ (1963) 2003. 123. J.P. Heppner and L.H. Meredith, J . Geophys. Res., 63 (1958) 51. 124. G. Witt, J . Stegman, B.H. Solheim and E.J. Llewellyn, Planet. Space S c i . , 27 (1979) 341. - 195 -125. W.R. Henderson and H.I. S c h i f f , Planet. Space Sci.,'18_ (1970) 1527. 126. D. Offermann and A. Drascher, J . Geophys. Res., 78_ (1973) 6690. 127. L. Thomas, R.G.H. Greer and P.H.G. Dickinson, Planet. Space S c i . , 27 (1979) 925. 128. M. P e t t i t d i d i e r and H. Teitelbaum, Planet. Space S c i . , 27 (1979) 1409. 129. L. Langmuir, J . Am. Chem. S o c , 37 (1915) 417. 130. O.R. Lundell, R.D. Ketcheson and H.I. S c h i f f , 12th Symp. Int. Combus. P r o c , _12_ (1969) 307. 131. F. Kaufman and J.R. Kelso, Faraday Discuss. Chem. S o c , 37_ (1964) 26. 132. P.P. Bernard, M.A.A. Clyne and R.T. Watson, J . Chem. Soc. Faraday Trans. 2., 70 (1974) 564. 133. G.B.K. Kistiakowski, J . Am. Chem. S o c , 5_2 (1930) 1868. 134. R.W. Wood, Proc. R. S o c , A97 (1920) 455. 135. , P r o c R. S o c , A102 (1923) 1. 136. M.A.A. Clyne and B.A. Thrush, Proc. R. S o c , A261 (1961) 259. 137. F.C. Fehsenfeld, K.M. Evenson and H.P. Broida, Rev. S c i . Instrum., 36 (1969) 294. 138. I.M. Campbell and B.A. Thrush, Ann. Rep. Chem. S o c , 6J_ (1965) 17. 139. M.A.A. Clyne, i n B.P. L e v i t t (ed.), Physical Chemistry of fa s t Reactions, Plenum, New York, 1_ (1973) 245. 140. F. Kaufman, Adv. Chem. Ser., 80 (1969) 29. 141. B. Brocklehurst and J.R. Jennings, Prog. React. Kinet., 4_ (1967) 1. 142. C L . Lin and F. Kaufman, J . Chem. Phys., 55 (1971) 3760. 143. D.W. Setser and D.H. Stedman, J . Phys. Chem., 74 (1970) 2238. - 196 -144. F. Kaufman and J.R. Kelso, J . Chem. Phys., 28 (1958) 510. 145. R.P. Wayne, Adv. Photochem., 1_ (1969) 311. 146. M.A.A. Clyne, D.J. McKenney and B.A. Thrush, Proc. R. Soc. London Ser. A, 295 (1965) 355. 147. F. Kaufman, Prog. React. Kinet., 1_ (1961) 1. 148. E.A. Ogryzlo, Can. J . Chem., 39 (1961) 2556. 149. J.W. Linnett and D.G.H. Marsden, Proc. R. S o c , A234 (1956) 489, 504. 150. J.C. Greaves and J.W. Linnett, Trans Faraday Soc., 5_5 (1959) 1346. 151. F. Kaufman, Proc. R. S o c , A247 (1958) 123. 152. P. Harteck, R.R. Reeves J r . and G. Mannella, J . Chem. Phys., 29 (1958) 1333. 153. C.B. Kretschmer, A.F.O.S.R.-TR-59-62, A.S.T.I.A. #217-008, AEROJET Report #1611. 154. P.E. Debeley, Rev. S c i . Instrum., 39 (1961) 2556. 155. R.B. Badachhape, P. Kamarchik, A.P. Conroy, G.P. Glass and J.L. Margrave, Int. J . Chem. Kinet., 8^  (1976) 23. 156. E. Wrede, Z. Instrumentenkd. , 48_ (1928) 201. 157. P. Harteck, Z. Phys. Chem., A139 (1928) 98. 158. Bonhoeffer, Z. Phys. Chem., 113_ (1924) 199, 492. 159. E.E. To l l e f s o n and D.J. Leroy, J . Chem. Phys., 16 (1948) 1057. 160. E.A. Ogryzlo, Ph.D. Thesis, McGill U n i v e r s i t y , (1958). 161. D.G.H. Marsden and J.W. Linnett, 5th Combus. Symp., Rheinhold, New York, (1955) 685. 162. L. E l i a s , E.A. Ogryzlo and H.I. S c h i f f , Can. J . Chem., 37_ (1959) 1680. 163. F.A. Morse and F. Kaufman, J . Chem. Phys., 42 (1965) 1785. 164. P.P. Bernhard, M.A.A. Clyne and R.T. Watson, J . Chem. Soc. Faraday Trans., 69_ (1973) 1356. - 197 -165. F.S. Klei n and J.T. Herron, J . Chem. Phys., 41 (1964) 1285. 166. L.T. Cuppitt and G.P. Glass, Int. J . Chem. Kinet. Symps., 1_ (1975) 39. 167. A.A. Westenberg and N. DeHass, J . Chem. Phys., 43_ (1965) 1550. 168. F. Kaufman, J . Chem. Phys., 28 (1958) 352. 169. , i n Chemiluminescence and Bioluminescence, Plenum, New York, (1973). 170. R.R. Reeves J r . , G. Mannella and P. Harteck, J . Chem. Phys., 32 (1960) 632. 171. M.A.A. Clyne, H.W. Cruse and R.T. Watson, J . Chem. Soc. Faraday Trans., 68 (1972) 153. 172. CM. Owens and J.M. Roscoe, Can. J . Chem., 5_4 (1976) 984. 173. S.N. Foner and R.L. Hudson, J . Chem. Phys., 25 (1956) 601. 174. I.N. Jones and K.D. Bayes, J . Chem. Phys., 59 (1973) 3119. 175. L.W. Bader and E.A. Ogryzlo, Faraday Disc. Chem. S o c , 37_ (1964) 46. 176. D.J. Benhard and N.R. Pchelkin, Rev. S c i . Instrum., 49 (1978) 794. 177. A. Fon t i j n , C B . Meyer and H.I. S c h i f f , J . Chem. Phys., 40 (1964) 46. 178. M. Sutoh, M. Morioka and M. Nakamura, J . Chem. Phys., 72_ (1980) 1. 179. M. Vanpee, K.D. H i l l and W.R. Kineyko, AIAA. J . , 9 (1971) 135. 180. M.F. Golde. A.E. Roche and F. Kaufman, J . Chem. Phys., 59_ (1973) 3953. 181. K.M. Becker, W. Groth and D. Thran, Chem. Phys. L e t t . , 6^  (1970) 583 and 15_ (1972) 215. 182. G.A. Woolsey, P.H. Lee and W.D. Sl a t e r , J . Chem. Phys., 67 (1977) 1220. 183. R.N. Clough and B.A. Thrush, Trans. Faraday S o c , 63 (1967) 915. - 198 -184. E.E. Ferguson, F.C. Fehsenfeld and A.L. Scheltekopf, Adv. At. Mol. Phys., 5_ (1969) 1. 185. M. Cher and C S . Hollingsworth, Adv. Chem. Ser., 80 (1969) 118. 186. C. Dewey Cooper and M. Lichtenstein, Phys. Rev., 109 (1958) 2026. 187. R.F. Hampson J r . and H. Okabe, J . Chem. Phys., 52 (1970) 1930. 188. D.L. Cunningham and C. Clark, J . Chem. Phys., 61_ (1974) 1118. 189. G. Black, R.L. Sharpless and T.G. Slanger, J . Chem. Phys., 63_ (1975) 4546. 190. A. Leiss, U. Schurath, K.M. Becker and E.H. Fink, J . Photo-chem., 8 (1978) 211. 191. R.A. Young and G. Black, J . Chem. Phys., 42_ (1965) 3240. 192. T.G. Slanger and G. Black, submitted to J . Chem. Phys. f o r pu b l i c a t i o n . 193. J.V. Evans, Planet. Space S c i . , 15_ (1967) 1557. 194. M. Krauss and D. Neumann, Chem. Phys. L e t t . , 36 (1975) 372. 195. W.F.J. Evans, D.M. Hunten, E.J. Llewellyn and A. Vallance Jones, J . Geophys. Res., 73 (1968) 2885. 196. W.F.J. Evans and E.J. Llewellyn, Ann. Geophys., 26 (1970) 167. 197. J . Noxon, J . Geophys. Res., 75_ (1970) 1879. 198. K. Schofield, J . Photochem., 9 (1978) 55. 199. F. Kaufman, Ann. Rev. Phys. Chem., 20 (1969) 45. 

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}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0060743/manifest

Comment

Related Items