UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A kinetic study of the termination and decomposition reactions of the cyclohexadienyl radical in the… Suart, Robert David 1966

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

Item Metadata

Download

Media
831-UBC_1966_A1 S85.pdf [ 12.52MB ]
Metadata
JSON: 831-1.0062136.json
JSON-LD: 831-1.0062136-ld.json
RDF/XML (Pretty): 831-1.0062136-rdf.xml
RDF/JSON: 831-1.0062136-rdf.json
Turtle: 831-1.0062136-turtle.txt
N-Triples: 831-1.0062136-rdf-ntriples.txt
Original Record: 831-1.0062136-source.json
Full Text
831-1.0062136-fulltext.txt
Citation
831-1.0062136.ris

Full Text

A KINETIC STUDY OF THE TERMINATION AND DECOMPOSITION REACTIONS OF THE CYCLOHEXADIENYL RADICAL IN THE GAS PHASE by ROBERT DAVID SUART B.Sc., University of B r i t i s h Columbia, 1962 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n the Department of CHEMISTRY We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA A p r i l , 1966 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 of the requirements 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 s h a 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-for 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 representatives„ I t i s understood that copying 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 s h a 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 The U n i v e r s i t y o f B r i t i s h Columbia FACULTY OF GRADUATE STUDIES PROGRAMME OF THE FINAL ORAL EXAMINATION FOR THE DEGREE OF DOCTOR OF PHILOSOPHY o f ROBERT DAVID SUART B. Sc. (Hons.)s The U n i v e r s i t y of B r i t i s h Columbia. 1962 WEDNESDAY. OCTOBER 5TH, 1966 AT 3;30 P.M. IN ROOM 2 6 l s CHEMISTRY BUILDING COMMITTEE IN CHARGE External Examiner: S. W. Benson Department of Thermochemistry and Chemical K i n e t i c s Stanford Research I n s t i t u t e Menlo Park, C a l i f o r n i a Research Supervisor: D. G. L. James Chairman: I- McT. Cowan G. G. S . Dutton D. C, Frost B. James D. G. L, James C. A. McDowell E. Peters R. C. Thompson D. C, Walker ABSTRACT The p h o t o l y s i s o f azomethane s d i - i s o p r o p y l k e t o n e and a z o i s o b u t a n e has been examined b r i e f l y i n the gas phase and t h e s e compounds have been found t o be c o n v e n i e n t s o u r c e s o f the meth y l , . i s o p r o p y l and t e r t - b u t y l r a d i c a l s respectively„ The p h o t o l y s i s o f the mixed v a p o u r s o f c y c l o h e x a d i e n e = 1 s4 w i t h each o f azomethane,, d i - i s o p r o p y l k e t o n e and a z o i s o b u t a n e has been examined o v e r a s e r i e s o f t e m p e r a t u r e s . These s t u d i e s a f f o r d e d t he A r r h e n i u s parameters f o r t h e a b s t r a c t i o n o f t h e m e t h y l e n i c h y d r o g e n atom from c y c l o h e x a d i e n e - l s 4 by the m e t h y l 3 i s o p r o p y l and t e r t - b u t y l r a d i c a l s . There was found no s i g n i f i c a n t d i f f e r e n c e i n the r e a c t i v i t i e s o f t h e s e r a d i c a l s towards the c y c l o h e x a d i e n e = l s 4 s u b s t r a t e . The r a t e c o n s t a n t s measured were k = U T 1 2 • 2 ± °° 2exp (-5.5 ± 0 . 4 ) , k. = m e t h y l r i s o p r o p y l .--11.9 + 0.7 (-6.4 ± 1 . 1 ) . . !0 -• exp RTT a n d k t e r t - b u t y l = . -12.3 + 0.5 ( = 5.3 + 0 . 8 ) , „ . -3. , -10 — exp =- 3 a l l i n cm. /molecule s e c . RT • The c y c l o h e x a d i e n y l ^ r a d i c a l i s g e n e r a t e d i n t h i s m e t a t h e t i c a l r e a c t i o n , and the i n t e r a c t i o n of the c y c l o h e x a d i e n y l r a d i c a l with the v a r i o u s m e n t i o n e d i n i t i a t o r r a d i c a l s was examined k i n e t i c a l l y . I t was found that the i n t e r a c t i o n of a l k y l r a d i c a l s with t h e c y c l o h e x a d i e n y l r a d i c a l p r o d u c e d e i t h e r benzene and the h y d r o c a r b o n , RH, ( d i s p r o p o r t i o n a t i o n ) or l = a l k y l e y c l o -hexadiene=2,4 ( c o m b i n a t i o n , I ) or l - a l k y l c y e l o h e x a d i e n e ~ 2,5 ( c o m b i n a t i o n , I I ) . .The r a t i o of- the r a t e s of f o r m a t i o n of t h e two c o m b i n a t i o n p r o d u c t s ( I / I I ) has been found t o have t h e c o n s t a n t v a l u e 0.77 + 0.17 w i t h i n t h e e x p e r i m e n t a l e r r o r f o r a l l t h e a l k y l r a d i c a l s s t u d i e d . The r a t i o o f t h e r a t e o f d i s p r o p o r t i o n a t i o n t o t h e combined r a t e s o f c o m b i n a t i o n was found t o v a r y s y s t e m a t i c a l l y o v e r the v a l u e s 0.27 + 0.07 s 0.52 + 0.09 3 and 1.33 + 0.24 f o r the m e t h y l , i s o p r o p y l and t e r t - b u t y l r a d i c a l s r e s p e c t i v e l y . A p r e v i o u s s t u d y o f t h e e t h y l r a d i c a l ' s r e a c t i o n s w i t h t h e c y c l o h e x a d i e n y l r a d i c a l i n t h i s l a b o r a t o r y had shown t h a t t h i s r a t i o f o r the e t h y l r a d i c a l was 0.38 + 0.03. The measured v a l u e s o f t h e t e r m i n a t i o n r a t e r a t i o s f o r t h e v a r i o u s systems a r e c o n s i s t e n t w i t h t h e e x p e c t a t i o n t h a t t h e p r o d u c t o f g r e a t e r e n t r o p y s h o u l d form p r e f e r e n t i a l l y and t h e r e s u l t s a r e c o n s i d e r e d t o s u p p o r t t h e d i s p r o p o r t i o n a t i o n t r a n s i t i o n s t a t e model o f B r a d l e y and R a b i n o v i t c h . D u r i n g .the p h o t o l y s i s o f a z o m e t h a n e - e y e l o h e x a d i e n e -l s 4 m i x t u r e s a t lower i n t e n s i s t i e s , t h e r e was o b s e r v e d the f o r m a t i o n o f c y c l o h e x e n e and g r e a t e r t h a n t h e e x p e c t e d amount o f benzene. T h i s was c o n s i d e r e d t o a r i s e f rom the d e c o m p o s i t i o n r e a c t i o n C^Hy—XJ^Hg + H° . K i n e t i c a n a l y s i s o f t h i s s y s t e m has a f f o r d e d an e s t i m a t e o f the h e a t o f f o r m a t i o n o f t h e c y c l o h e x a d i e n y l r a d i c a l , , (45 + 5 k c a l . / m o l e ) , and c o n s e q u e n t l y o f i t s r e s o n a n c e anergy, (24 + 5 k c a l . / m o l e ) . T h i s has been c o n s i d e r e d t o be e v i d e n c e t h a t t h e r e i s an i n t e r a c t i o n o f the d e l o c a l i z e d s y s t e m a c r o s s the m e t h y l e n i c c a r b o n bridge,, s i n c e the r e s o n a n c e energy i s s u b s t a n t i a l l y g r e a t e r t h a n t h a t measured i n a n o t h e r l a b o r a t o r y f o r t h e s t r a i g h t c h a i n p e n t a ' d i e n y l r a d i c a l (15.5 k c a l . / m o l e ) . The r e a c t i o n s o f t h e i s o p r o p y l r a d i c a l w i t h t h e c y c l o h e x a d i e n e ~ l s 3 m o l e c u l e have been s t u d i e d i n t h e gas phase. The i s o p r o p y l r a d i c a l adds t o t h e u n s a t u r a t e d - 1 1 9 + O 2 l i n k a g e s w i t h a r a t e c o n s t a n t k = 10 ~ exp (-5.8 ± 0.4% c 3/molec.sec. a b s t r a c t i o n o f a RT m e t h y l e n i c hydrogen atom p r o c e e d s w i t h a r a t e g o v e r n e d by the r a t e c o n s t a n t k = 1 0 " i l = 7 i °°^exp ( ° 7 o l p ^ ° ' 7 ) o Thus Ri t h e - m e t a t h e t i c a l r e a c t i o n p r o c e e d s more s l o w l y f o r the c o n j u g a t e d c y c l o h e x a d i e n e t h a n f o r t h e u n c o n j u g a t e d s y s t e m by a f a c t o r o f 1.9 a t 100°. T h i s b e h a v i o r was a l s o found f o r the r e a c t i o n s o f the e t h y l r a d i c a l w i t h the c y c l o h e x a d i e n e i s o m e r s i n a p r e v i o u s s t u d y , and p r o b a b l y a r i s e s from a s l i g h t l y l o wer f r e e e n e r g y o f the c y c l o h e x a d i e n e - l s 3 m o l e c u l e . The r a t i o o f a d d i t i o n t o a b s t r a c t i o n between i s o p r o p y l and c y c l o h e x a d i e n e - l s 3 i s low, (4.3 a t 100°) and d e g r a d a t i v e c h a i n t r a n s f e r has been s u g g e s t e d as the r e a s o n f o r the v e r y low tendency o f the c y c l o h e x a d i e n e - 1 , 3 m o l e c u l e t o p o l y m e r i z e under homogeneous, f r e e r a d i c a l c o n d i t i o n s . G R A D U A T E S T U D I E S F i e l d of Study: Chemistry Physical Chemistry Seminar Quantum Chemistry Chemical Physics Inorganic Chemistry Molecular Spectroscopy Chemical Ki n e t i c s Organic Chemistry Dr„ Coope Dr. Bree Dr. B a r t l e t t Dr. Coope Dr. L i n V Dr„ Frost'- 9 Dr. Dunne11 Dr. Cullen Dr. Rwon Dr. Clark Dr. B a r t l e t t Dr. Harvey Dr., Reeves Dr. Reid Dr. Wells Dr. Do Go L . James Dr. Ogryzlo Dr. Kutney Dr» Scott Dr. McCapra Physical Organic Chemistry Dr. Stewart Organic Reaction Mechanisms Dr. Pincock Related Studies: Abstract Algebra Dr. Divinsky PUBLICATIONS D.G.L. James and R.D. Suart - A K i n e t i c Study of the Cyclohexadienyl Radical. I. Disproportionation and Combination with the Isopropyl Radical, J.Am.Chem. Soc„ = 86, 5424 (1964). D.G.L. James and R.D. Suart - The R e a c t i v i t y of the Cycl Polyenes towards Free Radicals. IV. Cyclohexadiene-1. 3 and the Isopropyl Radical. J.Phys.Chem..69^2362(19' D.G.L. James and R.D. Suart - C h a r a c t e r i s t i c Reactions c the Cyclohexadienyl Radical Below 200° C. Chem, Commun. 484, (1966). ROBERT DAVID SUART. A KINETIC STUDY OF THE TERMINATION AND DECOMPOSITION REACTIONS OF THE CYCLOHEXADIENYL RADICAL IN THE GAS PHASE. S u p e r v i s o r : D. G. L. James. 1 1 ABSTRACT The photolysis of azomethane, di-isopropyl ketone and azoisobutane has been examined b r i e f l y i n the gas phase and these compounds have been found to be convenient sources of the methyl, isopropyl and t e r t - b u t y l r a d i c a l s respectively. The photolysis of the mixed vapours of cyclohexadiene-1,4 with each of azomethane, di-i s o p r o p y l ketone and azoisobutane has been examined over a series of temperatures. These studies af-forded the Arrhenius parameters f o r the abstraction of the methyl-enic hydrogen atom from cyclohexadiene-1,4 by the methyl, i s o -propyl and t e r t - b u t y l r a d i c a l s . There was found no s i g n i f i c a n t difference i n the r e a c t i v i t i e s of these r a d i c a l s towards the cyclohexadiene-1,4 substrate. The rate constants measured were . a 1 0-12.2 +. 0.2 (-5.5 ± 0.4) , „ i c T 1 1 - 9 ± °' 1 Kmethyl 1 U e x p R T ' ^isopropyl " 1 U (-6.4 * 1.1) v i 1 0-12.3 + 0.5 e x p ( - 5 . 3 ± 0.8) P R T k t e r t - b u t y l i U e x p R T a l l i n cm.^/molecule sec. The cyclohexadienyl r a d i c a l i s generated i n t h i s metathet-i c a l reaction, and the i n t e r a c t i o n of the cyclohexadienyl r a d i c a l with the various mentioned i n i t i a t o r r a d i c a l s was examined kinet-" ^ i c a l l y . I t was found that the i n t e r a c t i o n of a l k y l r a d i c a l s with the cyclohexadienyl r a d i c a l produced either benzene and the hydrocarbon, RH, (disproportionation) or 1-alkylcyclo-hexadiene-2,4 (combination, I) or l-alkylcyclohexadiene-2,5 (combination, I I ) . The r a t i o of the rates of formation of the two combination products ( I / l l ) has been found to have the constant value 0.77 + 0.17 within the experimental error f o r a l l the a l k y l r a d i c a l s studied. The r a t i o of the rate of disproportionation to the combined rates of combination was found to vary systematically over the values 0.27 + 0.07, 0.52 + 0.09, and 1.33 + 0.24 f o r the methyl, isopropyl and t e r t -butyl r a d i c a l s respectively. A previous study of the ethyl rad-i c a l ' s reactions with the cyclohexadienyl r a d i c a l i n t h i s lab-oratory had shown that t h i s r a t i o f o r the ethyl r a d i c a l was 0.38 + 0.03. The measured values of the termination rate r a t i o s fo r the various systems are consistent with the expectation that the product of greater entropy should form p r e f e r r e n t i a l l y and the r e s u l t s are considered to support the disproportionation t r a n s i t i o n state model of Bradley and Rabinovitch. During the photolysis of azomethane-cyclohexadiene-1,4 mixtures at lower i n t e n s i s t i e s , there was observed the formation of cyclohexene and greater than the expected amount of benzene. This was considered to arise from the decomposition reaction C 6 H 7 ^ C 6 H 6 + H** K i n e t i c analysis of t h i s system has af f o r d -ed an estimate of the heat of formation of the cyclohexadienyl r a d i c a l , (45 + 5 kcal./mole), and consequently of i t s resonance energy, (24 + 5 kcal./mole). This has been considered to be evidence that there i s an i n t e r a c t i o n of the delocalized system across the methylenic carbon bridge, since the resonance energy i s substantially greater than that measured i n another laboratory f o r the straight chain pentadienyl r a d i c a l (15.5 kcal./mole). The reactions of the isopropyl r a d i c a l with the cyclohexa-diene-1, 3 molecule have been studied i n the gas phase. The isopropyl r a d i c a l adds to the unsaturated linkages with a rate constant k = 1 0 " 1 1 * 9 0 , 2 e x p ( ~ 5 ' 8 R f °' 4) cm?/molec.sec. The abstraction of a methylenic hydrogen atom procedes with a rate governed by the rate constant k = 10 ~11'7 - 0 , 4 e x p ( ~ 7 , 1 R ^ °' 7^. Thus the metathetical reaction procedes more slowly f o r the conjugated cyclohexadiene than f o r the unconjugated system by a factor of 1.9 at 100°. This behavior was also found f o r the re-actions of the ethyl r a d i c a l with the cyclohexadiene isomers i n a previous study, and probably a r i s e s from a s l i g h t l y lower free energy of the cyclohexadiene-1,3 molecule. The r a t i o of addition to abstraction between isopropyl and cyclohexadiene-1,3 i s low, (4.3 at 100°) and degradative chain transfer has been suggested as the reason f o r the very low tendency of the cyclohexadiene-1,3 molecule to polymerize under homogeneous, free r a d i c a l conditions. - V -TABLE OF CONTENTS Page INTRODUCTION A. The cyclohexadienyl r a d i c a l and i t s importance i n kin e t i c systems 1 B. Techniques . 3 C. Termination reactions of free r a d i c a l s 9 D. Aims and scope of t h i s i n v e s t i g a t i o n 25 EXPERIMENTAL METHODS A. Description of the apparatus 27 B. Reagents 36 PHOTOLYTIC SYSTEMS A. Azomethane as i n i t i a t o r . 40 B. . Di-isopropyl ketone as i n i t i a t o r 104 C. 2,2*-Azoisobutane as i n i t i a t o r 144 D. Diethyl ketone as i n i t i a t o r 162 E. Rate constants, Arrhenius parameters, and s t a t i s t i c a l data 167 GENERAL DISCUSSION A. Disproportionation and combination reactions of the cyclohexadienyl r a d i c a l 170 B. Metathesis between a l k y l r a d i c a l s and cyclohexadiene-1,4 190 CONCLUSIONS 196 BIBLIOGRAPHY 199 - v i -LIST OF FIGURES Figure Page 1. Schematic diagram of the vacuum apparatus 29 2. Schematic diagram of the o p t i c a l system 30 3. Addition of the methyl r a d i c a l to azomethane 47 4. Metathesis between the methyl r a d i c a l and azomethane .. 48 5. Addition of the methyl r a d i c a l to azomethane; compari-son with previous determinations 51 6. Packed column analysis of the products of the high i n t e n s i t y photolysis of azomethane-cyclohexadiene-1,4 mixtures • •• 63 7. C a p i l l a r y column analysis of the products of the high i n t e n s i t y photolysis of azomethane-cyclohexadiene-1,4 mixtures • • 65 8. Metathesis between the methyl r a d i c a l and cyclohexa-diene-1,4 68 9. Combination and disproportionation of the methyl r a d i -c a l with the cyclohexadienyl r a d i c a l 72 10. Temperature dependence of the yie l d s of the products .. 83 11. K i n e t i c order of cyclohexadiene-1,3 and benzene pro-duction 88 12. K i n e t i c order of cyclohexene production 89 13. Temperature dependence of the disproportionation r a t i o s 91 14. Unimolecular decomposition of the cyclohexadienyl rad-i c a l 96 15. Metathesis between the isopropyl r a d i c a l and d i - i s o -propyl ketone 110 16. Metathesis between the isopropyl r a d i c a l and cyclohexa-diene-1,4 119 17. Combination and disproportionation of the isopropyl r a d i c a l with the cyclohexadienyl r a d i c a l 123 18. Addition and metathesis between the isopropyl r a d i c a l and cyclohexadiene-1,3 137 19. Metathesis between the t e r t - b u t y l r a d i c a l and cyclohexa-diene-1,4 160 - v i i -LIST OF FIGURES (continued) Figure Page 20. Disproportionation and combination of the t e r t - b u t y l r a d i c a l with i t s e l f and with the cyclohexadienyl rad-i c a l 163 21. Disproportionation and combination of the ethyl rad-i c a l with the cyclohexadienyl r a d i c a l 166 22. Comparison of Z\(R, C5H7) values to similar r e s u l t s for a l k y l r a d i c a l s 173 23. The dependence of E, on the bond d i s s o c i a t i o n energy of the attacking r a d i c a l 195 - v i i i -LIST OF TABLES TABLE PAGE I. Absolute rate constants of r a d i c a l recombination reactions 12 I I . Disproportionation to combination r a t i o s of a l k y l r a d i c a l s 17 I I I . Addition and metathesis between the methyl r a d i c a l and azomethane 44 IV. Metathesis between the methyl r a d i c a l and cyclohexa-diene-1,4 60 V. The unimolecular decomposition of the cyclohexadienyl r a d i c a l , 77 VI. Photolysis of pure di-isopropyl ketone 107 VII. Metathesis between the isopropyl r a d i c a l and cyclo-hexadiene-1,4 116 VIII. Upper accessible l i m i t s f o r the addition of the i s o -propyl r a d i c a l to cyclohexadiene-1,4 125 IX. Addition and metathesis between the isopropyl rad-i c a l and cyclohexadiene-1,3 128 X. Rate constants for reaction of the isopropyl r a d i c a l with cyclohexadiene-1,4 and cyclohexadiene-1,3 138 XI. Comparison of the ethyl and the isopropyl r a d i c a l i n r e a c t i v i t y towards the cyclohexadiene molecules .... 143 XII. Photolysis of pure 2,2'-azoisobutane 148 XIII. Metathesis between the t e r t - b u t y l r a d i c a l and cyclo-hexadiene-1,4 156 XIV. Reactions of the ethyl r a d i c a l with the cyclohexa-dienyl r a d i c a l 165 XV. Reaction Arrhenius parameters and s t a t i s t i c a l data . 168 XVI. Absolute rate constants of the addition and meta-t h e t i c a l reactions 169 XVII. Termination reactions of cyclohexadienyl and phenoxy r a d i c a l s 171 XVIII. Thermochemistry of the i n t e r a c t i o n of r a d i c a l s , R, with the cyclohexadienyl r a d i c a l 175 - i x -LIST OF TABLES (continued) TABLE PAGE XIX. P r o b a b i l i t i e s for ortho and para attack of a l k y l r a d i c a l s on the cyclohexadienyl r a d i c a l calculated on the basis of the Kerr and Trotman-Dickenson model 181 XX. D i s t r i b u t i o n of the products of combination and d i s -proportionation of a l k y l r a d i c a l s with the cyclo-hexadienyl r a d i c a l - dependence on the standard entropy 187 ACKNOWLEDGEMENT Sincerest thanks are due to Dr. D. G. L. James f o r en-couragement and guidance throughout the course of t h i s i n -vestigation. I t i s with the deepest appreciation that I acknowledge the f i n a n c i a l support of the University of B r i t i s h Columbia f o r a fellowship i n 1962-63 and of the National Research Council of Canada f o r Studentships covering the period from October, 1963 to A p r i l 1966. - 1 -INTRODUCTION A. The Cyclohexadienyl Radical and I t s Importance i n K i n e t i c  Systems Previous work i n t h i s laboratory"^ had indicated that the cyclohexadienyl r a d i c a l could be generated i n the gas phase under conditions where i t s reactions with various a l k y l r a d i c a l s could be studied. Very l i t t l e attention has been paid to the p o s s i b i l i t y of a comprehensive study of the pattern of i n t e r -action of a reference r a d i c a l with a complete sequence of s t r u c t u r a l l y related r a d i c a l s . Such a study i s possible and has been conducted i n t h i s work. Furthermore, there i s a great paucity of data about the combination and disproportionation reactions of r a d i c a l s whose free valence i s extensively de-l o c a l i z e d . Certainly termination reactions are very poorly understood t h e o r e t i c a l l y , and i t could be hoped that the r e s u l t s of a study of the termination reactions of a representative series of r a d i c a l s would o f f e r insight into the fundamental nature of these processes. F i n a l l y , the cyclohexadienyl r a d i c a l i s a species that i s an important intermediate i n a large variety of k i n e t i c systems. This i s so since the extensive de-l o c a l i z a t i o n of the free valence renders the r a d i c a l unreactive towards the majority of molecules because of resonance s t a b i l -i z a t i o n . The role of the cyclohexadienyl r a d i c a l i n such sys-tems i s therefore, to take part i n r a d i c a l termination reactions. Quantitative information on the termination reactions of the cyclohexadienyl r a d i c a l would be of help i n the i n t e r p r e t a t i o n of these systems. Many systems produce the cyclohexadienyl r a d i c a l as an intermediate. The mechanism of homolytic a r y l a t i o n involves derivatives of cyclohexadienyl r a d i c a l s as intermediates. In t h i s technique, r a d i c a l s are generated i n a solution of the aromatic substrate, but the precise mechanism through which these o are converted to the ultimate products i s unknown . A si m i l a r mechanism presumably holds f o r the methylation of benzene as 3 studied recently by Corbett and Williams . I r r a d i a t i o n of aromatic compounds t y p i c a l l y generates sub-- 4 . s t i t u t e d cyclohexadienyl r a d i c a l s . Benzene has been extensive-ly studied; recent experiments by Gaumann^ and by Eberhardt^ are i n essential agreement, and suggest that the majority of products arise through r a d i c a l reactions involving the phenyl and cyclohexadienyl r a d i c a l s . Phenyl and cyclohexadienyl rad-i c a l s are generated e f f i c i e n t l y when benzene vapour i s subjected 7 to electrodeless discharge . Recoil t r i t i u m atoms apparently generate the cyclohexadienyl r a d i c a l i n gaseous and l i q u i d 8 9 benzene . Chexniak et a l . have shown that hydrogen atoms generally add r a p i d l y to benzene forming cyclohexadienyl r a d i c a l s . The r a d i o l y s i s of cyclohexadiene -1,4^° also generates cyclohexadienyl r a d i c a l s , and the d i s t r i b u t i o n of products i s generally i n agreement with the benzene r a d i o l y s i s experiments. The above examples indicate that the cyclohexadienyl r a d i c a l i s of considerable importance to a variety of processes. As indicated above, the r o l e played by the cyclohexadienyl r a d i c a l i s generally accepted to be to remove r a d i c a l s from the system i n r a d i c a l termination reactions. Very l i t t l e attention i s given to the p o s s i b i l i t y that the cyclohexadienyl r a d i c a l may be thermally unstable and expell a hydrogen atom because of the gain of s t a b i l i z a t i o n energy attending aromatization. + H* AH = 72 - AH|(C 6H}) The heat required f o r t h i s process i s 72 kcal./mole less the heat of formation of the cyclohexadienyl r a d i c a l . Benson x x has quoted the heat of formation of the cyclohexadienyl r a d i c a l as 49 kcal./mole, therefore the decomposition of the cyclohexadienyl r a d i c a l would be only 23 kcal./mole endothermic and could e a s i l y be important i n some systems where the r a d i c a l s had a reasonably long l i f e t i m e . Such conditions are possible i n t h i s work and indeed, evidence for thermal i n s t a b i l i t y of the cyclohexadienyl r a d i c a l has been obtained. K i n e t i c study of t h i s reaction o f f e r s the means to estimate the resonance s t a b i l i z a t i o n energy of the cyclohexadienyl r a d i c a l , a quantity of considerable t h e o r e t i c a l i n t e r e s t . At present, the resonance energy has been calculated by Fisher to be around 29 - 30 kcal./mole, a value quite high when compared to the very large value of about 37 kcal./mole for 13 the benzene molecule. Benson has recently suggested that the cyclohexadienyl resonance energy w i l l l i e between 23 and 25 k c a l . / mole. An e a r l i e r publication l i s t e d the bond energy of the methylenic C-H bonds i n the cyclohexadiene-1,4 molecule as 74 kcal./mole; t h i s r e s u l t i s consistent with s l i g h t l y lower resonance energy of 20.5 kcal./mole assuming that without res-onance s t a b i l i z a t i o n of the r a d i c a l , the bond d i s s o c i a t i o n energy would be 94.5 kcal./mole, the value f o r the secondary 14 C-H bond i n propane. B. Techniques Brown-*", working i n t h i s laboratory, has previously examined the reactions of the ethyl r a d i c a l with cyclohexadiene-1,4, and - 4 = with the cyclohexadienyl r a d i c a l thus produced.As t h i s study t y p i f i e s much of the work reported here, i t w i l l be reviewed to provide a convenient background. Ethyl r a d i c a l s were generated i n the gas phase by the o photolysis of d i e t h y l ketone with 3130 A. r a d i a t i o n . In the photolysis of the pure d i e t h y l ketone the ethyl r a d i c a l s thus generated react by combination or disproportionation, or they can attack the ketone. The mechanism of photolysis of the pure ketone has been s3 from 50 to 250°), hown^""^ to be ( i n the range of temperatures 0 , h? + C 2H 5CC 2H 5 > 2 G2Hg + CO 2 C 2 H 5 " " ^ C4 H10 ^ C 2 H 4 + C 2 H 6 C 2H' + (C 2H 5) 2C0 C 2H 6 + C2H4Ec2H5 C 2 H 5 + C 2 H 4 § C 2 H 5 " > C4H98c2H5 At the higher i n t e n s i t i e s of l i g h t and below 250°, reaction (5) accounts f o r a l l of the pentanonyl r a d i c a l , C^i^C^i^, and a material balance, M, may be defined as M = (EL, u + FL> u )/Rrri> U 2 H 6 ^4"10 • where R^ i s the rate of formation of compound X obtained by d i v i d i n g the t o t a l y i e l d of X by the duration of the photolysis, assuming a steady-state regime. Under the experimental condi-tions, M has the value 0.988 + 0.02 (Kutschke et al,.) and 0.997 + 0.03 (James and Steacie) in d i c a t i n g that a l l r a d i c a l s are accounted f o r by the above mechanism. When t h i s compound i s used as a source of ethyl r a d i c a l s f o r experimental study of t h e i r metathetical reactions with com-pounds RH containing l a b i l e hydrogen atoms, the ketone can be photo lyzed i n a gaseous mixture w i t h RH. I n t h i s case , ethane i s formed i n the meta thes is from RH as w e l l as from r e a c t i o n s (3) and ( 4 ) . The mechanism may be expanded to i n c l u d e t h i s r e a c t i o n (6) under the assumption tha t r a d i c a l s R* are l o s t i n recombin-a t i o n p rocesses . Thus? W + C 2 H 5 § C 2 H 5 — > 2 C 2 H£ + CO 2 C 2 H * ? > C 4 H 1 Q -> C 2 H 4 + C 2 H 6 G 2^5 G 2^5^ G 2^5 ^ ^ G 2 ^ 6 G 2^4^ G 2^% ,Hrt§C«H= C 2 H 5 + C 2 H 4 C C 2 H 5 > C 4 H 9 C 2 5 C 2 H 5 + R H > C 2 H 6 + R * 6c Now, C 2 H J + R" — > C 2 H 5 - R (c = combinat ion) [Et] k 6 [RH] - R 6 = R ^ ^ - R G ^ - R 4 = R r „ - R r u • 77 Rr u _ M HI C 2 H 6 C 2 H 4 ,y2 C 4 H 1 0 2 where [D] = c o n c e n t r a t i o n of d i e t h y l ketone and [Etl = [p2H5^ = R^ / k ^ 4^10 The re fo re , R R k 6 G 2 H 6 ° G 2 H 4 k 4 k | Rp n M k f ^ U 4 H 1 0 z 1/ Since k 4 / k | i s r e a d i l y a v a i l a b l e from study of the pure ketone, 1/ values of the r a t e constant r a t i o k ^ / k ^ are a v a i l a b l e over a range of temperatures from s tud i e s of the p h o t o l y s i s of mix tu res of d i e t h y l ketone w i t h v a r i o u s hydrocarbons. I n s o f a r as measure-19 ment of k 2 has been made by Kutschke and Shepp, k 2 = 5.06 x 10' = ' : L 0 exp(-2000 + 1000)/RT (cm?/molec. s ec . ) - 6 -the values of k^/k^ measured can be used as a source of absolute rate constants k^. In the mechanism above, i t has been assumed that a l l rad-i c a l s FT are l o s t i n recombination with ethyl r a d i c a l s . I t i s well known, however, that r a d i c a l - r a d i c a l termination processes frequently involve disproportionation, as well. Thus, i n add-i t i o n to reaction (6c), we may imagine such processes as C 2 H 5 + R ' ~ > C 2 H 6 + R ( ~ H ) £ o r s o m e o l e f i n R(-H) or C 2H^ + R* — > G 2H 4 + RH may account f o r some of the r a d i c a l s R' formed i n reaction (6). Now study of the quantity Ro u /Rr u = k 0/k 0 f o r pure d i e t h y l ^2"4 °4 H10 3 z Q ketone has shown that kg/k,^ = 0.136 independent of temperature 1 20 I t has been found furthermore ' that when d i e t h y l ketone i s photolyzed i n mixtures of a variety of hydrocarbons, the quantity R^ . u /Rn „ does not increase s i g n i f i c a n t l y which t> 2H 4 O 4 H 1 0 shows that reaction (6e) i s not important. This i s true of mixtures of d i e t h y l ketone with cyclohexadiene-"l,4. No such simple test e x i s t s i n the case of reaction (6d) however, save actually searching f o r the o l e f i n R(-H). However, since the main purpose of previous studies with ketone-hydrocarbon mix-tures was to measure the a c t i v a t i o n energies f o r the abstraction of hydrogen atoms by ethyl r a d i c a l s , the possible occurrence of reaction (6d) was not important since i t s e f f e c t was probably re-s t r i c t e d to a change i n the measured A fa c t o r . This would be so since probably k 6 c A 6 d ± s i n d e p e n d e n t ( o r n e a r l y so) of temper-ature, i n common with most other combination-disproportionation r a t i o s of free r a d i c a l s . Therefore R ^ i s a constant propor-t i o n of R^ or R^d =;^)R^ and the error incurred i n neglecting reaction (6d) i s just a constant, temperature independent factor 1/(1 + 0 ) which w i l l a f f e c t only the value of A^/Aj. In the case when RH = cyclohexadiene-1,4, however, R ( - H ) i s the re a d i l y measured compound benzene and a means of study of the dispro-portionation to combination r a t i o of the ethyl r a d i c a l with cyclohexadienyl (C^H^) r a d i c a l presents i t s e l f . The complete mechanism i n t h i s case i s ( C 2 H 5 ) 2 C O +.. h? ^ > 2 C 2H* + C O 2 C 2 H ' 2 > C 4 H 1 Q 2 C 2H' 3 - > C 2 H 4 + C 2 H 6 C 2H' t ( C 2 H 5 ) 2 C 0 * > C 2 H 6 + C 2 H 4 C 0 C 2 H 5 C 2 H 5 + C 2 H 4 C 0 C 2 H 5 5 > C 4 H 9 C 0 C 2 H 5 C 2 H 5 + SV1'4 " > C 2 H 6 + C 6 H 7 C 2 H ^ C 6 H - Q f ^ o r QfC*H I II C 2H' + C 6H' ^ — » C 2 H 6 * C 6 H 6 The reactions (6e), C 2H* + C 6H^ ^ — - > C 2 H 4 + C 6 H g - l , 3 or C^Hg-1,4 had been quite rigourously ruled out of the mechanism because (a) R ( C 2 H 4 ) / R ( C 4 H ^ 0 ) was not s i g n i f i c a n t l y enhanced over the normal value 0.136, and (b) no trace of cyclohexadiene-1,3 was produced. Furthermore, i n the analoguous study of the photol-y s i s of die t h y l ketone-cyclohexadiene-1,3 mixtures, no trace of the 1,4 isomer was produced. From the mechanism R 6 " R 6 c + R 6 d " " C J H J " \ H 4 • <V kM 4H 1 0 t e6 H 8]- \ • < k6 / k2 ) « C 4 H 1 0 W - 8 -k6 R C 2 H 6 " - R C 6 H 6 k4 K whence — ^ ^ 6 6 and 4~ kI t C6 H8 ] ' k c % H 6 " V 4 - 2 V 6 - ( k 4 / k 2 ) [ D ] ^ 4 « 1 0 In t h i s way, Brown-*" was able to measure 13 + log -^- = (5.7 + 0.1) - 5 , 8 * °' 1 (units of ^ , ^ ) 2.3 RT J4 molec. sec. K2 2. and — T = 0.38 + 0.03 (independent of temperature), over the k c temperature range from 50 to 120 . In view of the successful application of t h i s analysis to the photolysis of d i e t h y l ketone with cyclohexadiene-1,4, i t was proposed that similar r e s u l t s would be obtainable from suitable "clean": sources of other free r a d i c a l s . Thus, from the photol-ys i s of mixtures of cyclohexadiene-1,4 with azomethane, d i -isopropyl ketone and 2,2'-azoisobutane, i t may be possible to measure k 6 d A 6 c values f o r C^Hj and CH3, (CH 3) 2CH, and (CH 3) 3C" r a d i c a l s as well as CH^CH^ r a d i c a l s as described above. Also, c careful study of the heavy products of the photolyses may allow i s o l a t i o n , i d e n t i f i c a t i o n , and measurement of the extent of formation of the combination products from reaction (6c). In the case of ethyl + C^Hy these products are expected to be and I II Brown attempted to i s o l a t e these products but f a i l e d to separ-ate them using the packed gas chromatography columns available - 9 -to him then. He was able to obtain an u l t r a v i o l e t spectrum of a mixture of the two, however, and reported t h i s to be si m i l a r to that of cyclohexadiene-1,3. This i s what would be expected from a mixture of I and I I . As w i l l be described l a t e r , the use of c a p i l l a r y columns has permitted the measurement of I T T k_/k f o r the compounds formed from the methyl, ethyl and i s o -propyl r a d i c a l s studied i n t h i s work. In t h i s study, the i n i t i a t o r systems employed have been azomethane, di-i s o p r o p y l ketone, 2,2'-azoisobutane as well as d i e t h y l ketone. Preliminary photochemical experiments performed on each established t h e i r s u i t a b i l i t y as r a d i c a l sources and t h i s w i l l be discussed i n appropriate sections. In the ex-periments with mixed vapours of i n i t i a t o r s and cyclohexadiene, the k i n e t i c analysis follows that given f o r d i e t h y l ketone with appropriate variations as w i l l be indicated. C. Termination Reactions of Free Radicals Free r a d i c a l s are the reactive intermediates of a large number of chemical conversions, p a r t i c u l a r l y those occuring i n the gaseous phase. They are able to attack stable molecules i n addition, metathetical and displacement reactions of very low ac t i v a t i o n energy. P a r t i c u l a r l y important to a l l such proces-ses i s the fact that free r a d i c a l s are removed from the system i n an extremely e f f i c i e n t termination process that i s bimolecular i n free r a d i c a l s . Since free r a d i c a l s are so reactive towards other r a d i c a l s and are eliminated i n the reaction with free r a d i c a l s , the termination process i s extremely important i n determining the concentration of r a d i c a l s i n the system. 1. Pressure e f f e c t s Intermittent i l l u m i n a t i o n experiments of Roberts and 10 Kistiakowsky have established the absolute rate constant f o r methyl recombination. These workers have observed that t h i s rate constant i s dependent on pressure i n the region below about 10 mm. Hg of acetone. Such an e f f e c t i s understood i n terms of the "hot intermediate" mechanism. where T i s the natural l i f e t i m e of CgH^ with respect to uni-molecular decomposition. Only at high pressures i s the assoc-i a t i o n rate constant, k^, measured. In accordance with the simple Kassel expression, the l i f e t i m e , , w i l l r a p i d l y lengthen as the degree of complexity of the recombining r a d i c a l s increases. Thus, This requires that the ex c i t a t i o n energy, E, be r a p i d l y d i s -tributed amongst the o s c i l l a t o r s , a pos i t i o n that has received 22 experimental comfirmation recently. Butler has generated excited methylcyclopropane by two routes which both ultimately 2 y i e l d the same d i s t r i b u t i o n of butene products. Harrington et a l . has formed excited sec-butyl r a d i c a l s by addition of hydrogen and deuterium atoms to 2-butene producing the same product 24 d i s t r i b u t i o n . Lee and Rowland f i n d that even i n the l i q u i d phase, the product of r e c o i l t r i t i u m atom addition to hexene-2 decomposes to ethyl r a d i c a l s and radio butene. Even the for a molecule of s o s c i l l a t o r s , energized by an energy E greater than E , the minimum energy required f o r decomposition. - 11 -e f f i c i e n t quenching of the l i q u i d i s not rapid enough to halt energy migration to a neighbouring bond to r e s u l t i n decomposition. 17 Brinton and Steacie have found the termolecular region f o r ethyl r a d i c a l recombination i s much lower than for methyl re-combination. In general, the higher r a d i c a l s should not display termolecular e f f e c t s above one or two mm. Hg pressure. 25 Benson has shown that, i n the termolecular region, a temperature dependence may be observed for recombination rate constants of the form kOC T ( l " s ) This arises from thermal energy, (s-l)kT, resident i n the bonds of the r a d i c a l s before reaction. 2. Absolute rates of recombination of a l k y l r a d i c a l s i n  the high pressure l i m i t Using the well known rotating sector or intermittent i l -lumination technique, the absolute rate constants for recombin-ation of certain simple a l k y l r a d i c a l s have been measured. The re s u l t s of these experiments have been compiled i n Table I. Certai n l y , these measurements are subject to error and probably are, at best, accurate to one s i g n i f i c a n t f i g u r e . I t i s evident, however, that a l l the a l k y l r a d i c a l recombination rate constants are extremely large and r e f l e c t c o l l i s i o n theory s t e r i c factors i n the neighbourhood of 0.1. The rate of recombination of isopropyl radicals has been measured at a series of temperatures and the r e s u l t s do not 19 indicate an a c t i v a t i o n energy. Kutschke and Shepp claim that t h e i r studies of the ethyl r a d i c a l show an ac t i v a t i o n energy for recombination and measure the Arrhenius parameters - 12 -TABLE I Absolute Rate Constants of Radical Recombination Reactions o 3 ( C) (cm./mole sec.) reagents T re f . ^2 CH' + CH' 125 a 1 013.34 135 b 165 b 1 013.90 1 Q13.5 CD' + CD3 135 b I 013.61 165 b 1 Q13.41 ^Q13.5 CF' + CF* f. 127 d J^ Q13 • 36 C2 H5 + C2H- 100 c 1 Q13.30 CH' + C2H- 125 e J^ Q13 .62 iso-C : 3H ? + iso-C 3H 7 115 f 1 013.88 *NC2 + *N02 127 g 1 0 1 1 , 7 CH' + 'NO 25 h IO 1 1- 8 480 i 1 011.1 900 • l 1 011.1 (adopted value) independent of •-temperature may have been measured i n termolecular pressure region a. A. Shepp, J . Chem. Phys., 24, 939 (1956). b. E. K. Roberts and G. B. Kistiakowsky, i b i d . . 21, 1637 (1953). c. K. 0. Kutschke and A. Shepp, i b i d . , 26, 1020 TT957). d. P. Ayscough, i b i d . , 24, ,944 (1956). e. C. A. Heller, i b i d . , 28, 1255 (1958). f. E. L. Metcalfe and A. F. Trotman-Dickenson, J . Chem. S o c , 1962, 4620. g. A. F. Trotman-Dickenson, "Gas Ki n e t i c s " , Butterworths, London, 1955, pp. 36, 125. h. W. C. Sleppy and J . G. Calvert, J . Am. Chem. S o c , 81, 769 (1959). i . W. A. Bryce and K. U. Ingold, J . Chem. Phys., 23, 1968 (1955). - 13 -A = 10 * cmr/mole s e c , E = 2 + 1 kcal./mole. Their a c t i v a t i o n energy depends on only three measurements, the lower one being subject to very large error, and i s probably not inconsistent with a value of zero for the a c t i v a t i o n energy. Since Trotman-26 Dickenson found a zero a c t i v a t i o n energy f o r isopropyl r a d i c a l recombination, i t i s unlikely that ethyl should have a b a r r i e r to recombination. I t i s evident from the table of recombination rates that the pre-exponential factors are extremely high. Thus isopropyl reacts either by recombination or disproportionation on almost every c o l l i s i o n i n the gas phase . Other r a d i c a l s seem to have s t e r i c factors of about 0.1. Such very high frequency factors are consistent only with "loose" t r a n s i t i o n states where reacting r a d i c a l s e i t h e r r e t a i n t h e i r r o t a t i o n a l freedom i n the t r a n s i t i o n state or convert t h e i r rotations into very "s o f t " , low frequency vibrations with large p a r t i t i o n functions. 27 The t r a n s i t i o n state theory of reaction rates shows that the s t e r i c factors of bimolecular reactions between A + B should be given by the expression, f & r o t ' n ) q ^ v i b ' n ) ^ q A ( r o t , n ) * q B ( r o t ' n ) q A(vib'n)*q B(vib'n) where ^denotes the activated complex, q(vib'n), q(rot'n) are molecular vibrat i o n and r o t a t i o n p a r t i t i o n functions respectively, and f r e f e r s to the sum over states of a single degree of free-dom. Since, for normal vibrations and rotations f ( v i b ' n ) / f ( r o t ' n ) r^j 0.1, the s t e r i c factor f o r the association -5 of two, non-linear free r a d i c a l s w i l l be on the order of 10 from application of equation ( l ) , r e a l i z i n g that the t o t a l - 14 -number of degrees of freedom must remain constant and hence rotations w i l l become vibrations. Since measured s t e r i c factors 3 4 are 10 to 10 times greater than expected, c l e a r l y the trans-i t i o n states of r a d i c a l recombinations are of an exceptional nature. I t has been suggested that the recombining r a d i c a l s do not lose t h e i r r o t a t i o n a l freedom i n the t r a n s i t i o n state and r o t a t i o n a l entropy i s not l o s t . The s t e r i c factor therefore 2 8 need not f a l l below unity. Eyring and coworkers have recently calculated the entropies of the hypothetical t r a n s i t i o n state of hydrocarbon pyrolyses. Their model fo r the t r a n s i t i o n state assumed f r e e l y rotating r a d i c a l fragments held together by 2 9 p o l a r i z a t i o n forces at rather large separation. The entropies thus calculated agreed closely to measured values obtained from intermittent i l l u m i n a t i o n experiments on the recombination of the appropriate r a d i c a l s . Further, they observe that even small r e s t r i c t i o n of the rotation would seriously depress the entropies of the t r a n s i t i o n states. For larger r a d i c a l s , i t i s d i f f i c u l t to imagine a f r e e l y rotating t r a n s i t i o n state consistent with the necessity of maintaining the centers of free valence close to-30 gether ; i n such cases i t seems more reasonable to suggest that the rotations of the free r a d i c a l s are converted into very low frequency rocking and bending vibrations possessing large 30 p a r t i t i o n functions. Recently, Benson has discussed several types of reaction, possessing high pre-exponential factors and suggests that a l l of these may have loose vibrations i n t h e i r t r a n s i t i o n states caused by a strong representation of i o n i c resonance forms i n the electronic description of the t r a n s i t i o n states. Examining p a r t i c u l a r ca e of the methyl r a d i c a l - 15 -recombination, he calculates from i o n i z a t i o n potentials, electron a f f i n i t i e s , and p o l a r i z a b i l i t i e s of methyl r a d i c a l s that the transfer of charge from one methyl group to another may be of s u f f i c i e n t l y low energy, even i n the gas phase that charged structures may be important i n the electronic description of recomining methyl r a d i c a l s . A crude approximation of t h i s energy 30 i s given by the equation , - A H 0 = Q c + EA - IP - Q d where = 83 kcal., the C-C bond energy i n ethane, IP = ion-i z a t i o n potential of methyl r a d i c a l s , EA = the electron a f f i n i t y of methyl r a d i c a l s and Q i s the energy of a t t r a c t i o n f o r two ions separated by a distance r i n vacuum. Q C = f ^ o +-1-25 In t h i s expression £ i s the electron's charge, 4.8 x 10~ X U esu., N Q i s Avogadro's number, and OC i s the r a d i c a l ' s p o l a r i z a b i l i t y , i n cm? For r between 2.5 and 2 A\ , the i o n i c energy i s suf-f i c i e n t l y low that i o n i c structures may become important. According to the theory, i o n i c bonding would be important i n lowering the force constants opposing rocking and bending motions of the methyl groups since i o n i c forces are non-d i r e c t i o n a l . The v i b r a t i o n a l p a r t i t i o n function r a t i o of equation ( l ) would have a large numerator and the loss of rotations would be largely compensated. 31 Rice has suggested that three-center, hydrogen bonds may provide s t a b i l i t y to the t r a n s i t i o n state even though the rad-i c a l centers rotate away from each other. - 16 H H H ©C-H rotation H H-C-H H © O H H In t h i s way the necessary freedom would remain i n the structures while s t i l l being f r a c t i o n a l l y bound together. An examination of the rate of disruption of the methyl-3 2 a l l y l bond i n chemically activated butene has shown that t h i s process, the reverse of the recombination of methyl and a l l y l r a d i c a l s , has a t i g h t e r t r a n s i t i o n state than i s observed f o r 33 C-C bond rupture i n alkanes . From th© data, a s t e r i c factor, -3 p = 5 x 10 , can be calculated f o r methyl plus a l l y l recombin-34 ation. Rabinovitch believed that such behavior may b© general for recombination reactions of delocallzed free r a d i c a l s . 3. Disproportionation The above discussion of free r a d i c a l recombination In the high pressure l i m i t makes no mention of disproportionation, an alternative termination process that competes e f f i c i e n t l y with recombination when hydrocarbon r a d i c a l s possess hydrogen atoms oC to the free valence. In f a c t , i n the case of t e r t - b u t y l rad-i c a l s , disproportionation i s much more probable than combination. In Table I I , a few of the more important Z\ values have been tabulated, where ) i s defined as kj/k where? Since we have been required to allow much freedom of re-combining r a d i c a l s i n the t r a n s i t i o n state to account f o r the high e f f i c i e n c y of t h i s process, we are forced to allow a great deal of freedom i n the t r a n s i t i o n states of disproportionating systems as well. Otherwise the loss of r o t a t i o n a l entropy would R + R' R + R' - 17 -TABLE II Disproportionation to Combination Ratios of A l k y l Radicals R A(Me,R) A(Et,R) A(iso-Pr,R) A(tert-Bu,R) Et 0.04 c 0.14 3 0.19 C 0.31 c 0.055 e 0.13^ c 0.065 d 1.7 1 0.06 9 0.1 1 0.3 1 iso-Pr 0.21 c 0.43 C 0.58 b 0.67 c 0.195 f 0.21 k 0.65 C 1.2 1 , 1 1 1 -0.17 j 0.43 d 0.2 0.53 h 0.5 1 tert-Bu 0.70 c 0.48 C 0.70 c 2.32 b 0.54 k 0.5 i 3.19 C 0.3 1 2.2 1 (Preferred values are underlined) 4.6 1 a. D. G. L. James and E. W. R. Steacie, Proc. Roy. S o c , A, 244, 289 (1958). b. This work. c. J . A. G. Dominguez, J . A. Kerr and A. F. Trotman-Dickenson, J . Chem. S o c , 3357 (1962). d. J . C. J . Thynne, Proc. Chem. S o c , 68 (1961). e. J . C. J . Thynne, Trans. Farad. S o c , 58, 676 (1962). f. J . C. J . Thynne, Trans. Farad. S o c , 58, 1394 (1962). g. P. Ausloos and E. Vtf. R. Steacie, Can. J . Chem., 33, 1062 (1955) h. R. H. Riem and K. 0. Kutschke, Can. J . Chem., 38, 2332 (1960). i . P. J . Boddy and J . C. Robb, P r o c Roy. S o c , A, 249, 547 (1959). j . C. A. He l l e r , J . Chem. Phys., 35, 1711 (1961). k. R. A. Holroyd and G. W. K l e i n , J . Phys. Chem., 67, 2273 (1963). 1.. K. W. Kraus and J . G. Calvert, J . Am. Chem. S o c , 79, 5921 (1957). A m. The greater value has been preferred here because £-± = 0.67 violat e s the pattern of the other r e s u l t s . - 18 -cause a drop i n i t s pre-exponential factor and the dispropor-tionation process would be unobservable i n comparison with the recombination process. We s h a l l discuss the current ideas about such t r a n s i t i o n states i n an appropriate paragraph. A number of pertinent experiments have been performed that shed l i g h t on the mechanism of these two termination processes. The products of the photolysis of d i e t h y l ketone-d^ have 35 been analyzed by Wijnen and Steacie. CH 3CD 2CCD 2CH 3 - — — > 2 CH 3CD 2 + CO 2 CH 3CD 2 > CH 2CD 2 + CH 3CD 2H > CH 3CD 2CD 2CH 3 These workers found that a l l hydrogen atoms transferred i n the disproportionation process arose from the |3> p o s i t i o n . That i s , no deuterium from the OC po s i t i o n was tranferred and the pro-cess CD0----CD0 ^ 1 ^ CD 2CD 2 + CH 3CH 3 CFU----CH, *3 ~"3 did' not occur with any p r o b a b i l i t y . They suggested, on t h i s evidence, that the t r a n s i t i o n state f o r disproportionation was l i k e l y of the "head-to-tail" variety t y p i c a l of hydrogen ab-str a c t i o n t r a n s i t i o n states CH3CD2---H-CH2CD2 Kraus and Calvert have studied the mutual i n t e r a c t i o n of the Isomeric butyl r a d i c a l s and found /\(sec-Bu, sec-Bu). A ( i s o ~ B u , iso-Bu) and A ( t e r t - B u , tert-Bu) to be 2.3, 0.42 and 4.59, respectively. These r e s u l t s were cl o s e l y i n accord with the number of reactive hydrogen atoms available i n each system - 19 -and tended to substantiate the head-to-tail t r a n s i t i o n state picture. The recent measurement of A(n~Bu, n-Bu) = 0.14, however, does not f i t t h i s pattern. Furthermore, Trotman-Dickenson^ points out that the A values vary systematically when various r a d i c a l s remove a hydrogen atom from a common donor r a d i c a l . In a great many k i n e t i c studies where considerable concen-tr a t i o n s of free r a d i c a l s i n t e r a c t homogeneously i n the gas phase, the A values f o r the r a d i c a l s have been observed to be sensibly constant with temperature. Considering the experimental accuracy of such studies, the a c t i v a t i o n energy difference be-tween disproportionation and combination i s less than a large 39 c a l o r i e . The one instance where marked temperature dependence had been suspected i n the A(n-Bu, n-Bu) values i s probably i n -correct because the values measured, around 0.7, are larger than are expected f o r n-alkyl r a d i c a l s . Indeed, the most recent 37 redetermination of t h i s r a t i o has the more reasonable value of 0.14 independent of temperature. At any rate, no reason ex i s t s at present to consider i t to be temperature dependent. Much more precise measurements by Szwarc 4^ , 4^"have revealed very small but apparently r e a l temperature e f f e c t s i n A ( E t , Et) measurements i n the photolysis of azoethane i n the gaseous phase over a wide temperature range from -65° to 40°. The temperature e f f e c t observed could be attributed to an a c t i v a t i o n energy difference E(disp) - E(comb) = -0.3 k c a l . A l t e r n a t i v e l y , the ef f e c t could be assigned to the pre-exponential factors, i f —O 7 k(disp)/k(comb) o c T~ * . Similar r e s u l t s have been found by 42 K l e i n and co-workers i n azoisopropane photolysis where, f o r - 20 -isopropyl r a d i c a l s E(disp) - E(comb) = -0.26 kcal. The r e s u l t s of these workers extend over the s o l i d as well as the gaseous phase, but from t h e i r r e s u l t s , i t would appear that the A E ob-served i s r e a l also i n the gas phase. Interestingly enough, the larger a c t i v a t i o n energy ( i f i t i s correct to assign the temp-erature e f f e c t to energy differences) belongs to the combination process. 41 In the work of Dixon et a_l., the ef f e c t s of the solvent cage and changes i n phase of the condensed reaction medium have been observed on measurements of /SiCrfl^, C^H^). Such phase change e f f e c t s are not observed, however, f o r CHg + ^2^0 9 e n ~ 43 erated i n the solvent cage. In the solvent cage, a greater temperature dependence of the A values was observed than i n the gas phase. I t was suggested that s p e c i f i c orientating forces due to the cage may be responsible f o r such e f f e c t s , but no concrete conclusions were possible. Furthermore, Watkins and Moser 4 4 observed that A ( E t , Et) i s unusually high at 63° f o r r a d i c a l s that were required to d i f f u s e into the cage from the bulk solvent, and hence the e f f e c t i s not due to r a d i c a l s being formed by azoethane photolysis i n the cage bearing an orientation predisposed towards disproportionation. The e f f e c t could be caused by difference i n the a c t i v a t i o n volume of the two pro-cesses, which t h e y 4 1 calculate to be V ^ d i s p ) - V^(comb) = 2.4 cc./mole, based on i n t e r n a l pressures of the solvents. 40 In an important study, azoethane has been photolyzed i n the gas phase and the e f f e c t s of added C X ^ J up to a t o t a l pressure of one atmosphere has been examined. Surprisingly, A ( E t , Et) was found to be e n t i r e l y i n s e n s i t i v e to t h i s ad-= 2 1 -mixture, both at -65°, and at 0°. S i m i l a r l y , H 20 was not ef-f e c t i v e i n a l t e r i n g the e f f i c i e n c y of disproportionation. Such an observation suggests that disproportionation i s not a product of unimolecular decomposition of the hot "quasi-molecule" formed on ethyl r a d i c a l recombination, competing with c o l l i s i o n a l deact-i v a t i o n . In t h i s regard, c e r t a i n processes have been observed bearing a s u p e r f i c i a l resemblance to disproportionation reactions, but which are e f f e c t i v e l y quenched by added, i n e r t gases. For 45 46 example, G i l e s and Whittle and e a r l i e r P r itchard and others have reported that f l u o r i n a t e d methyl r a d i c a l s may exhibit an unusual form of elimination following termination. Thus when 45 CH^ and CF^ recombine, there may be formed HF and C H 2 C F 2 as products, as well as the more normal CHgCF^. The y i e l d of C H 2 = C F 2 i s strongly depressed by pressure and surely r e s u l t s from elimination from the excited CH^CF^ quasi-molecule. Such r e a c t i v i t y i s analoguous to rupture of the C-C bond following 47 recombination of hydrogen atoms and C 2H£ r a d i c a l s which i s also pressure dependent. Many examples of excited r a d i c a l de-compositions are known, p a r t i c u l a r l y from the laboratory of 33 B. S. Rabinovitch, and the r e s u l t s are highly pertinent to unimolecular reaction rate theory. The pressure studies of 40 Szwarc and co-workers demonstrates that disproportionation i s a fundamentally d i f f e r e n t type of reaction. Primary isotope e f f e c t s have been measured f o r the dispro-portionation to combination r a t i o s of the ethyl r a d i c a l s . Boddy and S t e a c i e 4 8 have found A ( G 2 H 5 , G 2H 5 ) / A(C 2D 5, C 2D 5) = 1.4, independent of temperature from 50° to 300°. S i m i l a r l y , James 49 and Steacie observed the p r e f e r e n t i a l transfer of hydrogen - 22 - • atoms i n the disproportionation of p a r t i a l l y deuterated ethyl 44 r a d i c a l s of the type CD.Hg .CDA. Watkins and Moser found t r i t i u m atoms 2.3 times and deuterium atoms 1.5 times slower to transfer than hydrogen atoms. These l a t t e r measurements are most in t e r e s t i n g since they were obtained at 63°K., where isotope e f f e c t s a r i s i n g from zero point energy differences between the ground and the t r a n s i t i o n states would be expected to be enormous. 50 Salomon has recently formulated an isotope e f f e c t theory based on t r a n s i t i o n state theory which treats hydrogen abstraction reactions. Based on a t r a n s i t i o n state involving a l i n e a r com-plex where bending and symmetrical stretching C--H--C frequencies may or may not be involved, he find s exponential expressions of the form k H/k D = G exp ( - A ( zero point energy)/kT) f o r the isotope e f f e c t s . From such an expression, the d i f f e r -ences i n k^j/kp observable at room temperature would become very large at 63°K. Since Moser's deuterium isotope e f f e c t i s quite si m i l a r to Boddy and Steacie's value at high temperature, the t r a n s i t i o n state i s probably not of the sort used by Salomon i n his treatment of isotope e f f e c t s . Substantial C-H bond breaking apparently has not occurred i n the t r a n s i t i o n state. 4. The T r a n s i t i o n State f o r Disproportionation and Combin- ation The combination of free r a d i c a l s had been considered to 29 proceed through a f r e e l y rotating t r a n s i t i o n state, or at l e a s t one with very loose, low frequency l i b r a t i o n a l motions about the 30 i n c i p i e n t bond. Since the e f f i c i e n c y of disproportionation i s comparable to that of combination (Table II) s i m i l a r freedom i s necessary i n the t r a n s i t i o n state governing t h i s process. The - 23 -t r a n s i t i o n state proposed by Wijnen and Steacie (a head-to-tail structure similar to that of the hydrogen abstraction reactions) has been c r i t i c i z e d by Bradley. According to th e o r e t i c a l 51 52 methods to estimate the p a r t i t i o n function of such a structure, ' i t would not be of s u f f i c i e n t l y high entropy to explain the large pre-exponential factors of the disproportionation process. Kerr and Trotman-Dickenson proposed a mechanism where both combination and disproportionation proceed through the same, fr e e l y rotating t r a n s i t i o n state forming an excited molecule intermediate. For example, the ethyl r a d i c a l termination reac-tions are viewed as passing through the following sequences common, CH3CH2.+ CH 3GH 2 e n t g o p y > G H 3 C H 2 C H 2 C H 3 ^ M * + C H ^ C H ^ >CH 3CH 3 + CH 2CH 2 I t would be expected, therefore, that combination would be enhanced at high pressures; t h i s i n not observed i n the ethyl termination c a s e . 4 0 Benson^ 0 has further C r i t i c i z e d t h i s model as not explaining the is o t o p i c d i s t r i b u t i o n i n the pyr o l y s i s of deuterium l a b e l l e d butane under t o t a l i n h i b i t i o n conditions. • A similar, but d i s t i n c t l y d i f f e r e n t t r a n s i t i o n state model 53 has been advanced by Bradley and Rabinovitch. These authors consider that both disproportionation and combination have very loosely associated, f r e e l y moving t r a n s i t i o n states, although not necessarily i d e n t i c a l f o r both processes. The system be-comes committed to either combination or disproportionation while passing through t h i s state or states, and not as a competitive reaction system of an excited molecule intermediate. This i s a very much less d e t a i l e d view of the process than that of Kerr - 24 -and Trotman-Dickenson, but avoids the d i f f i c u l t i e s encountered i n the excited molecule mechanism. One prediction of t h i s pro-posal i s that the ultimate d i s t r i b u t i o n of products w i l l be governed by the entropy differences between them, and that the energy w i l l not be of importance. The system, situated at the t r a n s i t i o n state, w i l l f a l l into the most probable, or highest 54 entropy, configurations. Holroyd and K l e i n have successfully correlated a large number of disproportionation to combination r a t i o s f o r a l k y l r a d i c a l s by the equation l o g i 0 ( k d / k c ) = 0.131 ^ S ? ( d i s p ) - S°(comb) - 5.47 55 James and Troughton have found that t h i s r e l a t i o n s h i p success-f u l l y predicts the / \ ( a l l y l , ethyl) and / \ ( e t h y l , a l l y l ) values measured i n t h e i r experiments. Just as he has argued that extensive representation of charge transfer structures i n the combination t r a n s i t i o n state 30 would e f f e c t i v e l y increase i t s entropy, Benson suggests that such a description could be applied to the disproportionation t r a n s i t i o n state. According to t h i s view, the l i n e a r head-to-t a i l t r a n s i t i o n state geometry of Wijnen and Steacie may be acceptable i f charge transfer resonance forms are important enough to weaken the force constants opposing l i b r a t i o n a l motions. There would be two d i s t i n c t t r a n s i t i o n states, one f o r each termination process, and the product d i s t r i b u t i o n would be determined by the r e l a t i v e ease of forming them. The i o n i z a t i o n potentials, electron a f f i n i t i e s , and s t e r i c c h a r a c t e r i s t i c s of the reacting r a d i c a l s a l l could be important i n determining the ultimate product d i s t r i b u t i o n . In t h i s connection, Benson has observed that i n the mutual termination of the ethyl, isopropyl, - 25 -a n d t e r t - b u t y l r a d i c a l s , the tendency to disproportionate varies i n the opposite sense as the i o n i z a t i o n potentials of the reac-tants, which show a strong, uniform decrease from ethyl to t e r t -butyl. Indeed, Table II suggests that such trends are quite general for these methyl r a d i c a l homologues. Unfortunately, s t e r i c e f f e c t s and electron density at the CC -carbon atom also vary i n a uniform manner over these reactants and firm con-clusions are not possible. D. Aims and Scope of t h i s Investigation As there i s s t i l l no d e f i n i t e understanding of what factors influence the p r o b a b i l i t y of disproportionation during the termination reaction of free r a d i c a l s , i t should prove valuable to conduct a study of the d i s t r i b u t i o n of the combination and disproportionation products during the i n t e r a c t i o n of various r a d i c a l p a i r s . The procedure most l i k e l y to be i n s t r u c t i v e i s to examine the disproportionation to combination r a t i o s obtain-ed when a reference r a d i c a l i s allowed to react with a represent-ative series of r a d i c a l s . The work of Dominguez, Kerr and 57 Trotman-Dickenson has provided such series i n the case of the interactions of the various members of the homologous series represented by the formulae * C H . ( C H o ) o i (see Table I I ) . A simi l a r study of the disproportionation r a t i o s of these homologues with the cyclohexadienyl r a d i c a l i s a t t r a c t i v e f o r the following reasons. F i r s t , i t i s of i n t e r e s t to learn i f the tendency, ob-57 served by Dominguez et a_l., f o r the higher homology of the series C H ^ C H ^ ) ^ . . to be more e f f e c t i v e i n the removal of a hydrogen atom from a given substrate i s reproduced f o r the cyclohexadienyl r a d i c a l as donor, a system markedly d i f f e r e n t - 26 -from the a l k y l donors e a r l i e r studied. Secondly, the cyclohexa-dienyl r a d i c a l possesses a delocalized free valence and t h i s may be e f f e c t i v e In determining i t s r e a c t i v i t y . The p o s s i b i l i t y of the release of the very great benzene resonance energy i n the formation of the disproportionation products may also influence the r e a c t i v i t y of the system; the cyclohexadienyl system o f f e r s a unique opportunity to test f o r the influence of energetics i n these reactions. F i n a l l y , there i s at present a considerable body of information available on the properties and the structure of the cyclohexadienyl. r a d i c a l . For example, Fessenden and 58 Shuler have prepared the cyclohexadienyl r a d i c a l i n l i q u i d cyclohexadiene-1,4 and measured i t s electron spin resonance spectrum. This spectrum showed that the r a d i c a l was structur-a l l y planar, at least i n t h i s environment, and also afforded estimation of the electron spin densities at the ortho, meta and para positions to be 0.35, -0.10, and 0.51 respectively. 59 Harrison et, a l . have measured the heat of formation of the cyclohexadienyl cation to 232 + 3 kcal./mole. From the present estimate of the heat of formation of the cyclohexadienyl r a d i c a l (45 kcal./mole), the i o n i z a t i o n p o t e n t i a l of the cyclohexadienyl r a d i c a l can be estimated as 8.0 + 0.2 ev. Beyond the examination of the termination reactions of the cyclohexadienyl r a d i c a l , we s h a l l want to search f o r evidence of the thermal i n s t a b i l i t y of the species and to obtain thermo-chemical data on the decomposition reaction. Under the procedure used to generate the cyclohexadienyl r a d i c a l , we s h a l l be able to lower the rate of t h e i r formation and therefore to lower the the rate at which they are l o s t i n mutual termination. These conditions should favour the decomposition.-- 27 -EXPERIMENTAL METHODS A. Description of Apparatus Measurements of the k i n e t i c s of the systems studied i n t h i s work were made using conventional high vacuum techniques. As a precaution against the interference of oxygen, and against ab-sorption of materials i n grease, mercury cut-offs were employed i n place of grease taps i n the connecting tubing around the c e l l , the storage vessels and the a n a l y t i c a l l i n e . The general features of the apparatus are i l l u s t r a t e d schematically i n Figures 1 and 2. The apparatus consists of three e s s e n t i a l sections; the pre-parative l i n e , the reaction c e l l and o p t i c a l system, and the a n a l y t i c a l t r a i n . These w i l l be b r i e f l y described. 1* Preparative l i n e This consists of a manifold connected i;o the photolysis c e l l through a mercury cut-off on one end and a large-bore mercury manometer (A, Figure l ) at the other. Materials were stored behind mercury cut-offs i n vessels that were connected to t h i s manifold. The c e l l could be pumped out through the manometer by lowering the mercury thus connecting i t to the mercury d i f -fusion pump. 2. C e l l and o p t i c a l system Photolyses were carried out i n a c y l i n d r i c a l , s i l i c a c e l l 3 of 253 cm. illuminated volume. This c e l l was p r e c i s e l y f i l l e d with the beam of l i g h t from a B r i t i s h Thomson-Houston ME/D 250 W. medium pressure mercury arc lamp which provides intense l i g h t from almost a point source. This l i g h t was focussed by a quartz lens of f o c a l length 7 cm. into a p a r a l l e l beam. A front si l v e r e d mirror was placed j u s t beyond the c e l l (N, Figure 2) and served to return the beam of l i g h t back through the c e l l , - 28 -increasing the e f f e c t i v e i n t e n s i t y of the il l u m i n a t i o n , and also reducing the gradient of l i g h t i n t e n s i t y within the c e l l due to absorption. The beam was then f i l t e r e d using standard f i l t e r AO solutions, depending on the i n i t i a t o r being used, to i s o l a t e o o either the 3130 A. radiati o n (ketones) or 3660 A. radiati o n (azo compounds). The lamp was run at 65 volts and 3.6 amps. developing 234 Watts. No r a d i a t i o n i s produced between 2482 and o o 2752 A. owing to the reversal of the 2537 A. resonance l i n e eliminating the p o s s i b i l i t y of mercury photosensitization. For.the photolysis of d i e t h y l ketone and di-isopropyl ketone, an o p t i c a l f i l t e r of 0.5 % potassium hydrogen phthalate 2 cm. thick was employed. As the ketones do not absorb at longer o wavelengths than 3200 A. the e f f e c t i v e r a d i a t i o n was limited to o the region from 3000 to 3200 A. Due to photolysis of the potassium hydrogen phthalate, t h i s solution was renewed f o r each photolysis. For the photolysis of azomethane and 2,2'-azoisobutane the l i g h t from the mercury lamp was f i l t e r e d through a Pyrex #7380 f i l t e r and a 2 cm. thickness of a solution of 250 g. CuSO^Sr^O o per l i t r e of water. This combination passed 65 % of 3660 A. o rad i a t i o n and was opaque below 3400 A. The CuS0 4 f i l t e r was e f f e c t i v e to remove red and i n f r a - r e d r a d i a t i o n that may have contributed to temperature-instability i n the c e l l . Neutral density f i l t e r s of aluminum l i g h t l y deposited on s i l i c a were useful f o r attenuating the i n t e n s i t y of i l l u m i n a t i o n . The c e l l was housed i n an aluminum block e l e c t r i c a l r e s i s -tance furnace. I t s temperature was measured by meant of.three copper-constantan thermocouples taped to the surface of the c e l l . This arrangement was s u f f i c i e n t to maintain the c e l l temperature Figure 1. Schematic Diagram of the Vacuum Apparatus - 31 -Key to the Labelling of Figures 1 and 2. A. Wide bore mercury manometer. B. Standard volume bulb f o r measuring out gas samples. C. Liquid reagent storage c e l l s . D. Toepler pump and three-way grease tap to evacuate the gas burette and gas sample withdrawal. E. Toepler pump and gas burette. F. Second variable temperature trap. G. S p i r a l trap arranged for s o l i d nitrogen cooling. H. F i r s t variable temperature trap. I. Five l i t r e gas storage bulb. J . Cold finger system f o r removing l i q u i d photolysis residues. K. Vapour mixer. L. Gold finger. M. S i l i c a c e l l . N. Front s i l v e r e d mirror. 0. E l e c t r i c a l l y heated aluminum block furnace. P. Corning #774 f i l t e r . Q. Pyrex #7380 f i l t e r (for 3660 A. work only). R. 2 cm. s i l i c a c e l l for solution f i l t e r s . S. Neutral density f i l t e r s ( p a r t i a l l y s i l v e r e d mirrors). T. S i l i c a lens of 7 cm. f o c a l length. U. S i l i c a window. V. B r i t i s h Thompson-Houston ME/D 250 W. medium pressure mercury arc lamp:. - 32 -constant to within + 1° during the course of a k i n e t i c run. 3. The a n a l y t i c a l t r a i n The work required that an a n a l y t i c a l system of considerable f l e x i b i l i t y be employed as a variety of products would be pro-duced from the various i n i t i a t o r systems used. Accordingly, a low temperature f r a c t i o n a t i n g system was used containing two variable temperature traps which were modified Ward s t i l l s as proposed by Le Roy^ "*- capable of maintaining temperatures between l i q u i d nitrogen temperature and room temperature. Also a s o l i d nitrogen s p i r a l trap was used capable of maintaining a temperature of -215° by pumping on l i q u i d nitrogen i n i t i a l l y contained i n a Dewar f l a s k . The arrangement of these traps i s shown i n Figure 1. The temperatures used f o r analysis of each of the compounds w i l l be indicated i n an appropriate paragraph. B r i e f l y , the Le Roy s t i l l s consist of a v e r t i c a l l y arranged trap fashioned from two concentric tubes about 30 cm. long joined by a ring seal at the top. The outside of the trap i s equipped with thermocouples at three places along i t s length to monitor the temperature. I t i s then wrapped with lead f o i l and glass tape over which a heating wire of about 100 ohms i s wrapped. The entire trap i s placed i n a Pyrex jacket capable of being evacuated and surrounded with l i q u i d nitrogen. Temperatures above that of l i q u i d nitrogen are e a s i l y arranged by passing a small current (less than one half ampere) through the wire. Fractions of gaseous products v o l a t i l e f o r any p a r t i c u l a r set of temperatures i n the a n a l y t i c a l l i n e traps were pumped into a gas burette by a Toepler pump (E, Figure l ) assisted by a mer-cury d i f f u s i o n pump. The amount of gas coll e c t e d was estimated by measuring the pressure required to confine i t within a certa i n • - 33 -measured volume of the gas burette at the ambient temperature of the room. Provision was made to remove the product f r a c t i o n by means df another Toepler pump (D, Figure 1) into a bulb to be analyzed by gas chromatography i n the case that a f r a c t i o n may be complex. As the anaysis of heavier products such as benzene, d i -isopropyl and others was necessary f o r the k i n e t i c studies done, provision was made to remove the residual material a f t e r the l i g h t products had been removed. The residues were condensed into a small tube and removed from the vacuum system where they were di l u t e d with a convenient solvent and injected immediately onto gas chromatography columns f o r analysis. A l i g h t solvent such as pentane or pentene-1 was used as t h i s material passed quickly and cleanly through the column and did not i n t e r f e r e with analysis of the k i n e t i c products which are retained more strongly. This procedure was checked with synthetic mixtures of benzene, heptane and cyclohexadiene-1,4 and found to give r e a d i l y reproducible r e s u l t s as long as the sample tube was kept cool by suspending i t i n a Dewar with l i q u i d N 2 below i t , and as long as p r e - c h i l l e d syringes were used. Mixing was effected by drawing the mixture rapidly up and down i n the syringe used f o r analysis. Di-isopropyl was not very accurately measured by t h i s tech-nique due to i t s considerable v o l a t i l i t y and alternative procedures were employed i n estimation of t h i s compound. This w i l l be i n -dicated i n the appropriate place. 4. Chromatographic Analysis Two types of gas chromatographic analysis were used. For the majority of products the Perkin Elmer Vapour Fractometer 154C - 34 -instrument using 2 or 4 meter by V4 inch I.D. packed columns was preferred. This system, with a Flame Ionization detector and Leeds Northrup Model G, 1 mV. recorder offered superior accuracy and convenience. Gas samples were injected by means of ;a c a r r i e r gas bypass sampling loop of standard type. For the l i g h t e s t gases, a sampling loop with grease taps was adequate, but f o r analysis of butane and butene mixtures encountered i n the photolysis of azo-isobutane, a loop was constructed using greaseless taps. Liquid samples were introduced by a hypodermic syringe through a rubber septum. Packed column analysis was used f o r a l l gaseous samples and f o r analysis of the heavy products benzene, cyclohexadiene, cyclohexene and d i - i s o p r o p y l . In a l l cases, products were meas-ured r e l a t i v e to an i n t e r n a l standard added to the product mixture p r i o r to removal from the vacuum system. Such i n t e r n a l standards were either n-heptane or 3-methyl pentane measured out manometric-a l l y as vapours i n a standard volume (B, Figure 1) bulb on the vacuum system's preparative l i n e . For a l l analyses, n-heptane was used except f o r those pertaining to the experiments where azoisobutane was used as i n i t i a t o r as, i n t h i s case, n-heptane was not separated from products. In these experiments 3-methyl-pentane was used. In measuring out the vapours i n the standard volume bulb, pressures were kept below 2 cm. Hg so that nonideal-i t y corrections were unnecessary. Nitrogen was used as c a r r i e r gas. The flame i o n i z a t i o n detector was operated on H 2 gas flowing at 12 psig. from a tank through a porous sintered plug into the detector. S i m i l a r l y , pure a i r was passed at 30 psig. through a sintered plug into the detector. The response of the detector to each component of an - 35 -analytic mixture was calibrated by running synthetic mixtures of the compounds and comparing peak areas to the known mole r a t i o s . The measurement of area r a t i o s from chromatographic analysis per-formed was done using a planimeter. In cases where small peaks required measurement, the average on ten independent determina-tions of area was used. In t h i s way, the error from measurement of peak areas was maintained on the order of 1 %. During photolyses at high i n t e n s i t i e s of cyclohexadiene-1,4 with the various i n i t i a t o r s , two products arose from recombination of the cyclohexadienyl r a d i c a l with the i n i t i a t o r r a d i c a l . These products consistently elute a f t e r the passage of the cyclohexa-diene-1,4 peak i n the v.p.c. analyses on polyethylene g l y c o l column. For convenience, these have been assigned the labels products I and I I , I being the f i r s t to elute. Experiments to be described have shown that these products have the structures? f o r R = CH3-, ^2*%-, iso-CgH-y- or tert-C^HQ- depending on the p a r t i c u l a r i n i t i a t i n g r a d i c a l , R. The packed columns as des-cribed above were not capable of separating these peaks but the superior resolving power of Golay c a p i l l a r y columns permitted t h e i r separation. Accordingly, two analyses of each k i n e t i c run were performed on the heavy r e s i d u a l products; one on the packed columns using the Perkin Elmer 154G Vapour Fractometer, and a second on a 150 foot, 0.01 inch I.D. c a p i l l a r y column with polyethylene g l y c o l substrate. This second analysis was done on a Perkin Elmer 226 Gas Chromatograph having a temperature pro-gram feature of great u t i l i t y when analyzing products from the - 36 -in t e r a c t i o n of t e r t - b u t y l and isopropyl r a d i c a l s with the cyclo-hexadienyl radical.. As these compounds were quite i n v o l a t i l e , sampling on the Perkin Elmer 226 was by means of i n j e c t i o n of solutions of sample through a rubber septum as f o r the packed column analyses above. Detection was also by flame i o n i z a t i o n using H 2 gas and pure a i r introduced through sintered disks under pressure. Nitrogen was used as c a r r i e r gas at 12 psig. pressure. The column was kept at room temperature u n t i l the cyclohexadiene reagent has passed (about 10 minutes) then the temperature of the column was rapid l y raised to 120° where the heavier products eluted as sharp peaks. In the case of azomethane-cyclohexa-diene-1,4 experiments, the heating of the column was not neces-sary as these products were quite v o l a t i l e , and i n t h i s case alone the entire c a p i l l a r y column analysis was carried out at ambient temperature. B. Reagents In general the reagents, a f t e r p u r i f i c a t i o n as described below, were admitted immediately to the vacuum system and sub-jected to several degassing cycles of freezing down and pumping. In the case of i n i t i a t o r s , samples of these were photolyzed i n the c e l l and the products discarded i n order to condition the c e l l . Otherwise, i t was observed that unusual r e s u l t s may be obtained f o r i n i t i a l photolyses. A f t e r the p u r i f i c a t i o n , a sample of the p u r i f i e d material was analyzed on several chrom-atographic columns and spectra were taken. These r e s u l t s are indicated below. Azomethane, obtained from Merke, Sharp and Dohme (Canada) - 37 -Ltd. was p u r i f i e d by d i s t i l l a t i o n under vacuum from a Le Roy s t i l l set at -70°. The vapours passed through a second Le Roy s t i l l at -90°. A middle f r a c t i o n was reserved f o r use. This compound was stored behind a manometric mercury cut-off i n a blackened bulb and kept frozen down with l i q u i d nitrogen. Be-fore use, each sample was degassed by pumping on i t when i t was held down by l i q u i d nitrogen. This was necessary to remove nitrogen and methane that accumulated regardless of the spec i f i e d precautions i n storage. Diethyl ketone was used from a sample p u r i f i e d by e a r l i e r work 1 i n t h i s laboratory. This material was Eastman Kodak White Label grade that had been p u r i f i e d on a Ucon Polar column at 105° using a Beckman Megachrom Preparative Gas Chromatograph. Analysis showed impurities less.than 0.1 % and these residual impurities were not affected by photolysis of the sample. Di-isopropyl ketone was obtained from K and K Laboratories. Substantial impurities were completely removed by passage of the ketone through a 12 foot Apiezon-J column of the Beckman Mega-chrom at 95° using nitrogen c a r r i e r gas at 10 psig. The n.m.r. spectrum of the r e s u l t i n g material was i n complete agreement with that expected from di-isopropyl ketone - a heptuplet and a doublet with area r a t i o l s 6 . The heptuplet was centered at 6.72X 5 the doublet was centered at 8.98 ^  • 2,2-Azoisobutane was obtained from Merke, Sharp and Dohme-(Canada) Ltd., and found to be v i r t u a l l y pure by analysis on c a p i l l a r y columns using the Perkin Elmer 226 Gas Chromatograph. One impurity could be found which was less than 0.1 % and was not affected by photolysis. The n.m.r. spectrum showed only one, strong peak at 8.89 X > even under high amplification. I t was - 38 -degassed and used as i s . Cyclohexadiene-1,4 was obtained from A l d r i c h Chemical Company. I t had an impurity of about 5 % benzene and smaller amounts of cyclohexadiene-1,3 and cyclohexene. This crude material was pas-sed through a 12 foot Apiezon-J column on the Beckman Megachrom Preparative Gas Chromatograph at 75° which e f f e c t i v e l y removed a l l umpurities except benzene, which remained to the extent of about 0.1 %. The area r a t i o of impurity benzene to cyclohexa-diene-1,4 was determined on the polyethylene g l y c o l packed column under the same conditions as were used f o r the analysis of benzene during k i n e t i c runs. This data was used to correct the rate of formation of benzene as measured i n these runs f o r the impurity benzene. This was always a small correction (less than 10 % of the benzene found a f t e r photolysis corresponded to impurity benzene, generally). The 60 megacycle n.m.r. spectrum revealed two absorptions only. One, due to the methylene protons, was centered at 8.40T, and the other due to the o l e f i n i c protons, was found at 4.40T. Each peak was j u s t perceptibly s p l i t into a t r i p l e t with a coup-l i n g constant of about 0.2 c.p.s1. This unusually small s p l i t t i n g may be compared to the lack of any s p l i t t i n g between the two 62 classes of protons i n cyclobutene. A further, i n t e r e s t i n g point i s the apparently single ab-sorption of methylene protons. Any great fol d i n g of the molecule into a boat shaped structure of C 2 v symmetry would presumably cause the two protons within a methylene group to be non-equivalent. They then would absorb at d i f f e r e n t frequencies unless the i n -version of the molecule (through the D O K form) occured very - 39 rapidly at room temperature, which would seem l i k e l y . H H H D 2h H A l t e r n a t i v e l y , the molecular structure must have very nearly symmetry, a position adopted by Monostori and Weber i n recent r o t a t i o n a l Raman spectroscopic studies, and by Stidham i n v i -brational spectroscopic studies. Cyclohexadiene-1,3 was obtained from the Columbia Organic Chemical Co. I t was p u r i f i e d by passage through the 12 foot Apiezon-J column of the Beckman Megachrom Preparative Gas Chromatograph at 75°. Once again benzene was not completely re-moved and was present i n the f i n a l material to the extent of about 1 %. This did not i n t e r f e r e with studies done using t h i s compound, as benzene formation could be determined by an a l t e r -native method, and the benzene was not a chemically reactive substance. - 40 -PHOTOLYTIC SYSTEMS In t h i s chapter are presented the detailed experimental pro-cedures, r e s u l t s and some of the discussion of the photolysis of the various i n i t i a t o r systems used and of t h e i r photolysis i n the presence of substrate vapours. The purpose of the studies of the photolysis of the mixed vapours of i n i t i a t o r s and cyclohexadiene-1,4 has been to measure the r e l a t i v e rate constants and Arrhenius parameters fo r the attack of methyl, isopropyl and t e r t - b u t y l r a d i c a l s on the diene molecule. Also, more importantly, the d i s t r i b u t i o n of products of the i n t e r a c t i o n of these r a d i c a l s and of the ethyl r a d i c a l with the cyclohexadienyl r a d i c a l has been sought. These r e s u l t s are best discussed together, and therefore have been considered i n the General Discussion section. A. Azomethane as I n i t i a t o r 1. Pure Azomethane Photolysis Azomethane i s finding increasing use i n recent years as a source of methyl r a d i c a l s whose reactions with various substrate molecules may then be studied. I t was f i r s t examined c a r e f u l l y 65 i n gas phase photolysis by Jones and Steacie and since then, 66—6Q many workers have studied the photolysis. The quantum y i e l d of nitrogen has been measured as u n i t y ^ and the primary process i s almost e n t i r e l y a s p l i t into methyl r a d i c a l s and nitrogen gas. No evidence of excited molecule intermediates has been obtained. The methyl r a d i c a l s liberated attack the azo-methane molecule either abstracting a hydrogen atom^ 5'^'^ 9 ,or attacking the nitrogen-nitrogen double bond i n an addition r e a c t i o n . ^ * 7 0 The rate of t h i s l a t t e r reaction has not been accurately determined. - 41 -The gas phase photolysis of azomethane has been re-examinsd here with special attention to the evaluation of the rate data for methyl r a d i c a l reactions with the azomethane molecule. Experimental The photolysis of azomethane has been studied i n sixteen experiments between 24° and 167°. Azomethane reactant was passed from the storage bulb into the preparative l i n e and reaction c e l l u n t i l the desired pressure was obtained. This material was degassed at l i q u i d nitrogen temperature f o r at least f i f t e e n minutes, then allowed to vapoux ize and f i l l the c e l l . I t s pressure was measured using the wide bore manometer (A, Figure 1) with an accurate cathetorneter. At the same time the c e l l temperature was read from the copper-constantan thermocouples with a m i l l i v o l t potentiometer. The o il l u m i n a t i o n was the 3660 A.radiation i s o l a t e d from the B.T.H. medium pressure lamp with the f i l t e r combination described above. When the photolysis was started, the temperature i n the c e l l was observed to increase slowly. The mean temperature from a number of readings throughout the run was used. Reactant concentration was, of course, calculated from the i n i t i a l reading. At the pressure of azomethane used, considering the rather low r e a c t i v -i t y of methyl r a d i c a l s towards azomethane, the reactions could not be measured unless the l i g h t was attenuated by a neutral density f i l t e r of o p t i c a l density 1.40. This necessitated a reaction time of about one hour to accumulate s u f f i c i e n t pro-ducts f o r unequivocal analysis. A f t e r the photolysis was completed the entire reaction mixture was admitted to the analysis l i n e , where the various - 42 -traps were maintained at the following temperatures? l s ^ Variable temperature trap -135° Spiral trap -215° 2 n d Variable temperature trap -175° The f irst trap removes the bulk of the azomethane and heavier products. Nitrogen and methane pass through a l l traps and ethane is held back on the spiral trap. Pumping the volatile fraction into the Toepler pump-gas burette then gives the total yield of nitrogen and methane manometrically. Since the methane alone and not the nitrogen wil l give a signal on the flame ionization de-tector, i t is possible to estimate the relative concentrations of these two gases in a sample only i f a known internal standard can be introduced against which the methane can be measured. The ethane yield from the reaction has served as this standard in the following way. It was demonstrated by chromatographic examination that the ethane pumped into the gas burette when the spiral trap was warmed to room temperature was pure. It could therefore be independently measured manometrically and added thereafter to the N2-CH4 fraction. This mixture was then chromatographed on the 4 meter packed si l ica gel column at 50° with 8 psig. nitrogen as carrier gas. The ratio of CH^ to C 2H^ could then be obtained by multiplying by 1.87 the ratio of peak areas. That is P(CHJ Area (CH.) — — = 1.87- — P(C 2H 6) Area (C 2H 6) The calibration factor had been obtained in the usual way by comparison of the peak areas of synthetic mixtures of methane and ethane. The methane yield is known then from the ethane yield and the nitrogen can be determined by difference. The - 43 -values of the rates of formation of these products have been l i s t -ed i n Table III along with certain derived quantities to be ex-plained below. Treatment of Data According to current understanding of the azomethane photolysis the system can be described i n terms of the following reactions at high i n t e n s i t i e s and between 20-200°. CH3NNCH3 + hj>- ->2 CH 3 + N 2 m-a ^ C 2 H 6 + N 2 2 CH 3 - > C 2 H 6 CH 3 + CH3NNCH3 — >GH 4 + CH2NNCH3 A CH 3 + CH3NNCH3 — >(CH 3) 2NNCH 3 CH 3 + CH2NNCH3 — > C 2H 5NNCH 3 CH 3 + CH 3NN(CH 3) 2 — > (CH 3) 2NN(CH 3) 2 , AS The quantum y i e l d of nitrogen, 0 ^ , i s unity. Therefore 0 X + 0 f f l = 1. The alte r n a t i v e , molecular-ethane, mode of de-6T Y l composition has been established by'Rebbert ahd Ausloos. ' I t occurs to a very s l i g h t extent (0m - 0.012) and represents a negl i g i b l e source of ethane i n high i n t e n s i t y photolyses where reaction (2) has a high rate. During the f l a s h photodecomposition of azomethane, Sleppy 72 and Calvert did not detect the presence of the possible i n t e r -» mediate species, CH3NN, and no interference from t h i s source i s anticipated. I t i s under t h i s assumed mechanism, A, that the system has been analyzed. At the higher i n t e n s i t i e s of l i g h t , i t i s as-sumed that a l l r a d i c a l s produced from the i n t e r a c t i o n of methyl with azomethane are removed by recombination with another methyl - 44 -TABLE III Addition and Metathesis between the Methyl Radical and Azomethane :°K) T (sec. ) t 10 1 2R(X) [A] 3 N 2 (molec./ i CH 4 cm?sec. C 2H 6 ) ( M cm 3/molec.sec. 1 0 1 3 y 1 0 1 3  k4rr/ k2 k4a /' k' 297 300 3.06 90.1 0.302 88.3 0.983 - -306 3600 3*85 3.65 0.061 3.67 0.936 0.87 3.16 318 3600 2.81 .3211 0.0589 2.86 0.939 1.25 4.01 331 3600 3.58 4.11 0.143 3.53 0.893 2.14 6.59 339 3600 3,57 33.99 0.161 3.32 0.873 2.49 7.81 351 3600 3.66 3.92 0.260 3.11 0.860 4.03 8.50 359 1800 2.02 11.01 0.428 9.90 0.938 6.76 10.7 374 1800 2.70 16.97 0.821 14.20 0.885 8.12 19.1 380 1800 2.90 17.09 0.993 13.84 0.868 9.24 21.0 381 1800 2.44 14.67 0.850 11.96 0.873 10.2 22.2 386 1800 3.39 20.05 1.13 15.86 0.347 (8.37) b 22.7 391 1800 1.83 11.89 0.964 9.57 0.886 17.1 24.1 411 1800 3.34 21.70 2.97 13.67 0.767 24.1 40.9 423 300 2.54 19.32 3.28 12.70 0.826 36.3 37^3 430 1800 2.03 13.61 2.86 8.06 0.802 50.2 48.1 440 1800 1.72 11.91 2.64 6.45 0.763 61.0 65.4 a 10 [A] = Concentration of azomethane (molecules/cm.). D Value excluded on s t a t i s t i c a l grounds. I - 45 -r a d i c a l s . If t h i s mechanism holds true, one/molecule of nitrogen should represent eit h e r the formation of one molecule of CH^, one molecule of C 2H^ or the consumption of two CH^ r a d i c a l s i n an addition reaction followed by termination of the adduct r a d i c a l . We may define the following "material balance quotient", M, where M = ( R C H 4 +. a C 2H 6>/ RN 2 <2> When the value of M f a l l s below unity, i t indicates the occurence of the addition reaction (4a). In order to assess the accuracy of the a n a l y t i c a l technique, a photolysis was done using extremely high l i g h t i n t e n s i t y , where r a d i c a l recombinations are favoured, and l i t t l e addition, comparatively, would be anticipated. The f i r s t run entered into Table III corresponds to t h i s experiment which was done at room temperature to further discourage addition to azomethane. As can be seen, the measured material balance i s very close to unity, confirming the a n a l y t i c a l accuracy. I t i s e a s i l y shown, from the above mechanism, using the usual steady-state treatment f o r methyl r a d i c a l s , that R N o g - M) y - £ -_ {3) ko Azomethane R~ u r C 2 H 6 where R(X) means the rate of formation of compound X. In t h i s expression, R~ „r ref e r s actually to the ethane formed i n b i -C 2 H 6 molecular recombination of methyl r a d i c a l s , and not to that form-ed i n the molecular elimination process. R r u r i s equal to U 2 H 6 R- u - 0.012 R M where the factor 0.012 has been measured i n the C 2 H 6 N 2 low i n t e n s i t y photolysis experiments to be considered i n another section. Since molecular elimination as measured i n t h i s work, and also by Thynne, Szwarc et, a_l. and Rebbert and Ausloos ' - 46 -i s approximately one percent of R M , the difference i s i n s i g n i f -2 icant. S i m i l a r l y , the steady-state treatment yi e l d s the expression: k4m RGH^ Vo ~ r n l i ( 4 ) k£ Azomethane „r C 2 H 6 These rate r a t i o s along with computed values of the material balance have been tabulated f o r each run i n Table I I I . Arrhenius plots of the rate r a t i o s are given i n Figure 3 and Figure 4. From a least mean squares treatment of the Arrhenius data, we obtain the following values f o r the Arrhenius parameters? 1 Q 1 3 k4m _ 1 Q ( 6 . 0 + 0.3) / 8.5 ± O . D Y 4 \ R T / / cm? Y Imolec. s e c J 1 0 1 3 ^4a = 1 Q ( 4 . 7 + 0 . 2 ) e x p / 5.9 ± 0.4^ 4 \ R T The errors have been expressed at the 5 % pr o b a b i l i t y l e v e l . 75 Toby and Weiss have obtained evidence that the recombination of methyl r a d i c a l s during azomethane photolysis may be pressure IT 3 dependent at azomethane concentrations below 4 x 10 molec./cmT r i V Increases i n the function R O L I / A R i „r of 10 - 20 % were de-CH 4 L J C 2H 6 tected as the azomethane concentration f e l l to 1.7 x 10 , the lowest concentration used i n the present study. I t i s suspected, therefore, that the Arrhenius parameters measured here are s l i g h t l y i n error because of t h i s pressure e f f e c t . We may replace the rate equation (2) of Scheme A by the more detailed sequence 2 CH 3 k2" ^ C 2 H * C 2 H 6 k b ^ 2 C H 3 C 2 H 6 + C H 3 N 2 C H 3 ^ > C 2 H 6 + C H 3 N 2 C H 3 • Experiments done with the variable temperature trap at -160° - 48 -Figure 4. Metathesis between the Methyl Radical and Azomethane - 49 -Applying a steady-state treatment to jC^Hg], we r e a d i l y show that k4m R C H 4  k2 R C 2 H £ M k 4 a R N U - M ) -y2 :2 K C 2 H | [ A ] v --c where [A] = [Azomethane] . Therefore, each r e s u l t of Table III can be corrected f o r pressure e f f e c t s I f we can evaluate the function (1 + k k / k c [ A ] ) . Unfortunately, Toby and Weiss' r e s u l t s do not provide an accurate assessment of k ^ A ^ . However, fo r purposes of making a small correction, t h i s r a t i o has been estimated as l o g l 0 k j ; ) / k c = 18.2 - 0.44 10 /T i n molecules/cmf, which i s con-sis t e n t with t h e i r r e s u l t s . Recalculation.of k 4 m / k 2 and of k 4 g / k 2 using t h i s correction provides the Arrhenius expressions ' V= l O * 5 ' 7 ± °-2>exp -(Q'2±- °'4) k* RT , y2 ( cm.J V t Vmolec. sec./ -4$- = i o ( 4 ' 4 ± °- 3 )exp -(5.6 ± 0.4) kf RT This s l i g h t change i s barely s i g n i f i c a n t at the 5 % l e v e l of error. In view of the tentative nature of the values of the pressure dependent rate constants f o r methyl recombination, we s h a l l prefer the uncorrected r e s u l t s ; i f more precise pressure dependence studies are available i n the future, these can be recorrected i n the manner indicated above. An i n t e r e s t i n g e f f e c t was noticed i n the photolysis exper-iments that may perhaps deserve further i n v e s t i g a t i o n . Results quoted above were measured with the second variable temperature - 50 -trap set at -175° when measuring ethane. Preliminary work was done on t h i s system using -160° for the setting of t h i s variable temperature trap. At t h i s higher temperature i t was observed that the ethane f r a c t i o n did not come over pure, but contained small but k i n e t i c a l l y important amounts of another unidentified product, the e f f e c t of which was to cause a depression of the measured value of k^ a. This i s undoubtedly due to the fact that M values measured would be very sensitive to such spurious mater-i a l s that would a r t i f i c i a l l y enhance the value of R r u . The C 2H 6 e f f e c t dies out at about 85 . The e a r l i e r measurements are i n -dicated on Figure 3. The presence of such an e f f e c t implies that the simple mechanism presented above for azomethane photoysis may not hold exactly. Discussion Of chief i n t e r e s t i n the r e s u l t s of the azo-methane photolysis i s the value of the addition rate constants. This has been measured previously by Jones and Steacie and more 70 recently by Kerr and Calvert and the r e s u l t s of these workers have been shown on Figure 5 along with values from t h i s work. Cl e a r l y , agreement i s poor among the three determinations, a l -though the points of Jones and Steacie are i n reasonable agreement with t h i s work, except f o r a persistent lowering by about 20 % above 50°. The Arrhenius parameters reported i n each study are l i s t e d i n the following Present work Jones and Steacie (1952) Kerr and Calvert (1965) 3 K (kcal.) (cm./molec. s e c ) 2 E4a'" ^ E2 log(A 4 a/A!|) + 13 5.9 + 0.4 4.7 + 0.2 6.4 not quoted 7.1 5.7 - 5 1 -qure 5. Addition of the Methyl Radical to Azomethane; Comparison with Previous Determinations. O Experimental points of Jones and Steacie. - 52 -The r e s u l t s of Jones and Steacie have been calculated using the same assumptions as i n t h i s work, that i s , that the r a d i c a l s formed during the addition reaction are a l l removed from the system by termination with a methyl r a d i c a l . However, c a l c u l a t i o n of M values from t h e i r data according to the corresponding ex-pression (2) shows that t h e i r methyl r a d i c a l flux was not suf-f i c i e n t l y high to support t h e i r assumption. Thus, above 90° t h e i r M values are consistently less than 0.5, i n d i c a t i n g such a high rate of formation of adduct r a d i c a l s that mutual termination must be e f f e c t i v e . Furthermore, the scatter of t h e i r experimental r e s u l t s indicates i n s u f f i c i e n c i e s i n t h e i r assumptions. They have also measured the addition a c t i v a t i o n energy by mass spectrometric determination of the product tetramethylhydrazine, (CHg^NNlCHgJg, and found es s e n t i a l agreement with t h e i r a c t i v a t i o n energy of 6.4 kcal. However, as other fates must e x i s t f o r the adduct r a d i c a l than to form tetramethylhydrazine, the same ob-jections apply to t h i s r e s u l t as apply to t h e i r material balance technique. The present experiments have been carried out with a rate of nitrogen generation (or methyl generation) about 5-10 times greater than that of Jones and Steacie, r e s u l t i n g i n M values never less than 0.75. Under these conditions, the mat-e r i a l balance technique i s much more secure. The l i m i t i n g case, when the rate of i n i t i a t i o n f a l l s very low would r e s u l t i n conditions where the material balance quotient i s c o r r e c t l y formulated ass M- - (* R c % • R C 2 H 6 ) / R N 2 (5) Such conditions probably exist at the highest temperatures studied by Jones and Steacie where t h e i r M values are less than - 5 3 -0.5. I f we calculate t h e i r high temperature points according to t h i s assumption of mutual termination of adduct r a d i c a l s , we fi n d that t h e i r r e s u l t s are very close to those of t h i s work. A further d i f f i c u l t y with Jones and Steacie's r e s u l t s i s that they have measured t h e i r ethane y i e l d using a trap at -160°. As i t i s now found that these conditions r e s u l t i n an a r t i f i c i a l f a l l off of k^ a/k2 below 80°, and as t h e i r lowest temperature r e s u l t s (24°) seem to be rather depressed (Figure 5), i t i s pos-si b l e that t h e i r measurements of E 4 g - J £ E 2 are increased by t h i s source of error. I t i s also e a s i l y shown that the mutual term-ination of adduct r a d i c a l s would cause a compensating decrease i n the measured ac t i v a t i o n energy, although probably not a very great decrease. 70 The values measured by Kerr and Calvert d i f f e r more ser-t iously from those of t h i s work, having a larger a c t i v a t i o n energy and pre-exponential factor. They have measured the addition reaction i n a rather complex system. Azomethane was photolyzed i n the presence of acetaldehyde and the product, (CH 3) 2NNHCH 3 was observed. This was imagined to be formed i n the following steps: CH3NNCH3 + h? —0rIa-> 2 *CH3 + N 2 *CH3 + CH3NNCH3 k4a -> (CH 3) 2NNCH 3 B CH3CH0 + (CH 3) 2NNCH 3 k s > CH3C0 + (CH 3) 2NNHCH 3 This product, (CH 3) 2NNHCH 3 was measured and taken as an index of the rate of addition of methyl to azomethane. Such a pro-cedure can only be v a l i d i f a l l (CHg^NNCrLy are e f f i c i e n t l y scavenged by CHgCHO, otherwise the acti v a t i o n energy of the scavenging step w i l l be a component of the addition a c t i v a t i o n energy measured. As they report the presence of tetramethyl-- 54 -hydrazine, t h e i r a c t i v a t i o n energy can be considered only an upper l i m i t . The a c t i v a t i o n energy of 5.9 k c a l . measured i n t h i s work for methyl addition to azomethane may be compared with 6.0 kcal. "7f> measured by Cerfontain and Kutschke f o r the addition of ethyl r a d i c a l s to azoethane. I t i s quite unlikely that the addition of methyl r a d i c a l s should have a higher ac t i v a t i o n energy than add-i t i o n of ethyl r a d i c a l s , therefore the 5.9 k c a l . r e s u l t i s prefer-rable to that of Kerr and Calvert. The a c t i v a t i o n energy determined here i s lower than Kerr and _. Calvert's value and goes i n hand with a low s t e r i c f actor. I f we assume the diameters of methyl and of azomethane to be 3.5 and o _R 5.5 A. respectively, then the s t e r i c factor becomes 7.1 x 10 i f kg = 10 * litre/mole sec. From these assumptions, Kerr -4 and Calvert's s t e r i c factor becomes 6.7 x 10 , ten times greater 77 and i n accord with s t e r i c factors observed f o r ethyl r a d i c a l addition to terminal o l e f i n s . As we s h a l l see, there i s reason to believe that non-terminal o l e f i n s have much lower s t e r i c factors and hence the smaller value may be i n accord with experience. I t i s d i f f i c u l t to f i n d methyl addition rate constants i n the l i t e r a t u r e with which to compare the azomethane r e s u l t s . Of p a r t i c u l a r i n t e r e s t would be the rate of methyl attack on 2-butene i n the gaseous phase. V i r t u a l l y a l l work has been done in the laboratory of M. Szwarc i n iso-octane solution and solvent e f f e c t s may well cloud the picture. However, accurate r e s u l t s 77 are available f o r ethyl r a d i c a l addition to terminal and non-terminal o l e f i n s and may be used f o r comparison. Also available 78 i s work by Mandelcorn and Steacie concerning the addition of, methyl r a d i c a l s to ethylene and propylene. James and Steacie - 55 -have found that ethyl addition to non-terminal o l e f i n s , eg. trans-octene-2 and cyclohexene i s at least ten times lower than that f o r terminal mono-olefins and were not able to measure addition to these compounds. Results f o r terminal o l e f i n s indicated that substitution of a l k y l groups lowers the ac t i v a t i o n energy f o r 78 r a d i c a l addition. S i m i l a r l y , Mandelcorn and Steacie found a lower a c t i v a t i o n energy (6 kcal./mole) f o r methyl r a d i c a l add-i t i o n to propylene than to ethylene (7 kcal./mole). I t i s l i k e l y that the low r e a c t i v i t y of non-terminal mono-olefins i s the r e -sul t of a s t e r i c e f f e c t . Since azomethane i s a non-terminal s i t e of unsaturation, a low s t e r i c factor may well be -5 expected and, indeed the value 7.1 x 10 i s rather small and would seem more reasonable than the value of Kerr and Calvert. The a c t i v a t i o n energy found f o r methyl addition to azomethane i s lower than that found by Kerr and Calvert and lower than f o r t y p i c a l a l k y l r a d i c a l addition to the C=C double bond, although the disubstituted mono-olefins appear to show a comparable act-i v a t i o n energy towards ethyl attack. In view of the facts that (a) The N=N double bond i s l i k e l y to be weak compared to the C=C double bond because of the lower catenating power of nitrogen. (b) The C-N bond i s only s l i g h t l y weaker than the C-C bond, and (c) The azomethane molecule may be somewhat activated by the methyl groups attached to i t , a low a c t i v a t i o n energy may be expected f o r methyl attack on the N=N double bond. The r e s u l t s f o r the abstraction reaction are more i n accord with other work. The ac t i v a t i o n energy of 8.6+0.5 kcal. i s i n 79 good agreement with that measured by Kutschke and Toby (E =?= 80 8.4 kcal.) and by Gray et, a l . (E = 8.2 kcal.) who have measured the reaction over the same temperature range as i n t h i s work. - 56 -6 6 Toby has pointed out that the various values measured f o r the act i v a t i o n energy of t h i s process tend to d r i f t with temperature, the lower values being measured at lower temperatures. His ex-periments suggested a component of the reaction may occur on the vessel walls with a lower a c t i v a t i o n energy. The sensitive anal-y t i c a l method employed i n t h i s i n v e s t i g a t i o n f o r the determination of methane has allowed the measurement of the reaction at very high i n t e n s i t i e s where the quantum y i e l d of methane i s very low, p a r t i c -u l a r l y at lower temperatures. This higher methyl r a d i c a l con-centration would tend to disfavour any heterogeneous reaction, and therefore the measured temperature dependence of the rate constant l i e s at the upper bcurid of those measured previously, which may have suffered from t h i s interference. 75 On the other hand, a recent redetermination of E 4 m has given the lower r e s u l t , E 4 m = 7.5 + 0.3 kcal./mole, where a l l the mechanistic d e t a i l s have been taken into account. Inspection of th e i r data shows that the low ac t i v a t i o n energy depends strongly on one experimental r e s u l t at 25°, and that, neglecting t h i s , the remaining four points are not inconsistent with E 4 m = 8.5, i f reasonable errors are assigned. Another complication arises when lower r a d i c a l fluxes are used i n measurements of the metathetical reaction. As mentioned under Treatment of the Data, the rate constants f o r each run are determined from experimental data from r e l a t i o n (4), where the ethane rate has been corrected f o r ethane a r i s i n g from molecular elimination. Now at the lowest i n t e n s i t y and highest temperatures, the o v e r a l l quantum y i e l d of ethane can f a l l quite low because many r a d i c a l s are consumed i n reactions with azomethane. - 57 -Neglect of the correction for the ethane generated by molecular elimination i n azomethane photolysis may produce substantial errors when the quantum y i e l d of ethane f a l l s below 0.1. This can occur when low i n t e n s i t i e s are employed or when the azomethane concen-t r a t i o n i s great, and nearly a l l methyl r a d i c a l s form methane, or are l o s t i n addition. In the present work, the ethane quantum y i e l d i s always greater than 0.54 and the molecular elimination ethane i s a n e g l i g i b l y small correction. The pressure dependence 65 of rate constants measured by Jones and Steacie derives from t h i s cause. I t i s i n t e r e s t i n g to compare the r e s u l t s f o r hydrogen ab-stra c t i o n from azomethane by methyl r a d i c a l s to si m i l a r r e s u l t s obtained f o r the i s o e l e c t r o n i c m< compiled i n the following table. 81 82 olecules butene-2 and acetone (kcal.) abstraction 10 p a CH; + CHoC0CHo a 1.2 9.7 '3 ~ " 3 ^ " 3 I 3 + CH3N=NCH3 I 3 + CH3CH=CHCh3 CHA  CH0N=NCH„ b 1.4 8.6 CHA 0 H0 c 1.71 7.7 k Trotman-Dickenson and Steacie. This work. pp H Knox and Trotman-Dickenson. o a Using O^(CHJ) = 3.5 X.; 6^(substrate) = 5.5 A. The a c t i v a t i o n energy f o r abstraction from acetone i s t y p i c a l of the values found f o r primary hydrogen atom abstraction from a l -kanes, which l i e around 10 k c a l . The lower value f o r abstrac-t i o n from butene-2 surely arises from the fac t that the a l l y l resonance energy i s p a r t i a l l y released i n the t r a n s i t i o n state, thus lowering the ac t i v a t i o n energy. The measured ac t i v a t i o n - 5 8 -energy f o r a b s t r a c t i o n from azomethane l i e s between the other two, and i t i s tempting t o suppose t h a t the extent of a c t i v a t i o n of the v a r i o u s hydrogen a b s t r a c t i o n r e a c t i o n s l i e s i n the same order as the degree of resonance s t a b i l i z a t i o n of the i n c i p i e n t a l l y l i c m o i e t i e s . Thus i t i s proposed t h a t the resonance e n e r g i e s of these groups l i e i n the orders CH^CH-CH^CH-^ > CH 2 - N - N(CH 3) > CH^-C-X) CH 3 2. Azomethane-cyclohexadiene-1,4 mixtures- h i g h i n t e n s i t y  p h o t o l y s i s As azomethane appears to be a good source of methyl r a d i c a l s when used under a p p r o p r i a t e c o n d i t i o n s , i t has been photolyzed i n the presence of cyclohexadiene-1,4. The o b j e c t i v e s were (a) to generate c y c l o h e x a d i e n y l r a d i c a l s under c o n d i t i o n s where t h e i r r e a c t i o n s w i t h methyl r a d i c a l s c o u l d be examined, (b) to gather data on the a b s t r a c t i o n of hydrogen atoms from cyclohexadiene-1,4 by methyl r a d i c a l s , and (c) to determine i f any a d d i t i o n of methyl r a d i c a l s to the double bonds of cyclohexadiene-1,4 c o u l d be found. Because of the very g r e a t r e a c t i v i t y of cyclohexadiene-1,4 towards methyl r a d i c a l s , a very h i g h r a t e of p h o t o l y s i s was necessary to ensure consumption of c y c l o h e x a d i e n y l r a d i c a l s by r e a c t i o n s w i t h methyl r a d i c a l s . To secure t h i s , the f u l l i n t e n s i t y of the B.T.H. o mercury lamp a t 3660 A. was employed. Experimental The system has been examined i n s i x t e e n exper-iments between 24° and 117°. Azomethane was measured out i n t o the r e a c t i o n c e l l i n the same way as d e s c r i b e d f o r the pure azomethane experiments, t h o r -oughly degassing each sample p r i o r t o measurement. T h i s measured - 59 -sample was then temporarily kept i n a storage c e l l while a sample of cyclohexadiene-1,4 was s i m i l a r l y measured into the photolysis c e l l . Both vapours were then frozen into a cold finger attached to the c e l l and the mercury cut-off closed. The mixed vapours were then allowed to v o l a t i l i z e and were thoroughly mixed by r a i s i n g and lowering the mercury i n the mixer bulb (K, Figure l ) twenty times. A f t e r mixing, the sample was i r r a d i a t e d with the maximum o in t e n s i t y of il l u m i n a t i o n available at 3660 A. from the B.T.H. medium pressure lamp using the same f i l t e r s as for pure azomethane photolysis. With the concentrations of reactants and temperatures used as tabulated i n Table IV, t h i s l i g h t was s u f f i c i e n t to gen-erate nitrogen at approximately 1 0 1 4 molecules/cm? s e c , or about 10-20 times f a s t e r than f o r pure azomethane. As cyclohexadiene-1,4 i s very reactive towards methyl, rather small concentrations were employed and even at the short reaction times employed ( t y p i c a l l y 300 seconds, see Table IV) a s i g n i f i c a n t amount of the cyclohexa-diene (up to 25 %) was consumed. A correction was applied as explained under Treatment of the Data and values of [B] l i s t e d i n Table IV are corrected values, where B i s the e f f e c t i v e concentration of cyclohexadiene-1,4. A f t e r photolysis was completed, the reagents were refrozen into the cold finger, and then allowed to vapourize into the a n a l y t i c a l l i n e where the three traps (Figure l ) were set at the following temperatures: 1 s t Variable temperature trap -135° S p i r a l trap -215° 2nc* Variable temperature trap -175° TABLE IV Metathesis between the Methyl Radical and Cyclohexadiene-1,4 °K) (sec.) 10" 1 2R(X) (mol ec./cm? sec.) / cm? V* unol. sec J T t N 2 CH 4 C 2 H 6 C 6 H 6 M* 10i 3k 6/k| A ( D A(2) 296 120 6.24 0.98 131.2 8.25 118.3 1.66 0.970 59 0.358 0.302 0.698 304 600 3.14 1.35 74.1 10.57 64.6 2.84 0.976 70 (0.594) (0.432) 0.804 305 1200 2.91 1.28 56.0 9.86 48.3 1.20 1.020 96 (0.162) 0.274 0.719 321 120 7.46 1.49 182.4 31.87 155.3 5.05 0.999 138. 0.233 (0.379) 0.659 321 300 2.91 1.49 74.6 19.87 66.5 3.24 1.160 135 0.248 0.239 0.631 323 600 4.01 0.90 98.7 14.55 83.4 2.00 0.971 146 0.201 0.230 0.780 325 300 5.11 0.78 119.2 13.94 102.9 2.47 0.960 142 0.302 0.239 0.846 334 300 3.25 0.80 83.3 17.33 69.6 2.86 1.010 208 0.261 0.295 0.722 347 300 6.98 0.51 191.2 23.95 163.9 4.10 0.961 251 0.265 0.251 0.778 355 120 9.18 0.94 219.7 47.67 164.4 7.70 0.930 284 0.292 0.248 0.697 356 300 10.46 0.27 262.1 22.59 217.5 2.65 0.918 298 0.292 0.242 0.800 357 300 5.77 1.65 158.8 71.28 98.0 12.18 0.989 344 0.278 0.305 0.841 TABLE IV (continued) 3 -12 3 ( K) (sec.) 10° R(X) (molec./cm. sec.) mol.sec T t [ A ] 3 [ B ] 3 N 2 CH 4 C 2 H 6 C 6 H 6 M* 10 A ( D A(2) 363 600 8.48 0.74 187.8 44.33 131.1 7.19 0.896 364 0.306 0.319 0.880 368 120 9.98 0.96 240.0 62,62 155.5 .11.06 0.863 354 0.356 (0.404) 0.683 376 300 8.45 0.76 223.8 64.45 144.4 10.07 0.888 492 0.293 0.311 0.800 390 300 8.65 0.74 223.0 79.32 124.9 15.10 0.848 610 (0.435) (0.413) 0.824 Mean value , excluding entries i n parentheses 0.283 0.271 0.871 i Limits of error a t 5$ probability • l e v e l +0.097 +0.076 +0.159 »-a 10, 7 [A] = Concentration of azomethane (molecules/cm.) „ 10 \B] = Concentration of cyclohexadiene-1,4 (molecules/cm.) k ° A U ) = R c H / ( R C H - 2 R c „ — - f ^ A l R ^ r) . C 6 H 6 C H 4 C 6 H 6 k| L J C 2 H 6 C A ( 2 ) = B-/:j/R/:r Calculated as described i n text, page 70. - 62 -These are the same temperatures as used f o r pure azomethane s t u d i e s . The f i r s t trap stops the bulk of the azomethane, a l l of the c y c l o -hexadiene and heavy products. The methane, ethane and n i t r o g e n y i e l d s were determined using e x a c t l y the same procedure as was employed f o r pure azomethane. The p u r i t y of the gaseous product f r a c t i o n s was checked f o r some runs by a n a l y s i s on an A s s o c i a t e d E l e c t r i c a l I n d u s t r i e s M.S.9 mass spectrometer. The mass peaks observed were p r e c i s e l y what would have been expected from a mix-tur e of ethane, methane and n i t r o g e n and confirmed the r e s u l t s of the r o u t i n e l y used v.p.c. a n a l y s i s f o r the ethane t o methane r a t i o . The a n a l y t i c a l method f o r gaseous products was judged, t h e r e f o r e , t o be s u f f i c i e n t . A f t e r the gaseous products had been removed, azomethane was s t r i p p e d from the r e s i d u e s by pumping them through a trap at -78° and the remaining material.was mixed w i t h a measured,amount of n-heptane as a chromatographic i n t e r n a l standard and then f r o z e n i n t o a sample tube and removed from the vacuum l i n e . Two analyses were performed, each on a separate p o r t i o n of the sample. (1) About 5 j j l . of the mixture was i n j e c t e d onto a 2 meter, packed, polyethylene g l y c o l column "RM, using a P e r k i n Elmer Vapour Fractometer 154C w i t h flame i o n i z a t i o n d e t e c t i o n . Care was taken to ensure t h a t the i n j e c t i o n port temperature never rose above 100° to guard against p o s s i b l e p y r o l y s i s of the sam-p l e s . When n i t r o g e n gas at 12 p s i g . was passed through the column, the m a t e r i a l s were e l u t e d i n the orders pentene-1 ( s o l v e n t ) , n-heptane ( i n t e r n a l standard), benzene, cyclohexadiene-1,4, and a f i n a l peak due to product I I . A t y p i c a l chromatographic record i s reproduced i n F i g u r e 6. No other products were observed even when a l t e r n a t i v e columns were employed to search f o r them. The Figure 6. Packed Column Analysis of the Products of the High Intensity Photolysis of Azomethane-Cyclohexadiene-1,4 Mixtures. O o 2m. polyethylene glycol column °^ run at 52° with nitrogen c a r r i e r gas at 12 psi gauge pressure. time (min) - 64 -r a t i o of areas, benzene to cyclohexadiene-1,4 was used to make a small correction to the benzene y i e l d due to the presence of a small amount of benzene i n the cyclohexadiene-1,4 sta r t i n g mater-i a l . This correction was always less than 10 % of the t o t a l ben-zene peak area. The residual area was converted to moles of ben-zene by comparison with the area of the n-heptane peak using the calibrations moles (benzene) = ^ Area (benzene) moles (n-heptane) Area (n-heptane) The c a l i b r a t i o n factor 1.095 was determined from measurements of synthetic benzene-n-heptane mixtures. (2) A second analysis was done on the residues i n order to measure the long retention time peaks I and II (see Figure 7A) which are better resolved on a c a p i l l a r y column than on the pack-ed columns used f o r benzene analysis. (I i s not separated from the t a i l of the cyclohexadiene-1,4 peak on the packed column.) About 5 jj.1. of the product mixture was injected i n t o the stream s p l i t t i n g i n j e c t i o n block on a Perkin Elmer Model 226 gas chrom-atograph using a c a p i l l a r y polyethylene g l y c o l column "R", as described under Experimental Methods. When the column was main-tained at room temperature the products were eluted as shown schematically i n Figure 7A. Peak areas f o r compounds I and I I , benzene and cylohexadiene-1,4, were measured with a planimeter. Special care was taken while running t h i s analysis to check for the presence of cyclohexadiene-1,3 and other products. None were observed. S p e c i f i c a l l y , R~ u , ~ <fo.01 R~ u C 6H 8-1,3 C 6H 6 Treatment of the Data A n a l y t i c a l examination of the photol-'. y s i s of azomethane-cyclohexadiene-1,4 mixtures at high i n t e n s i t i e s has revealed the following products? nitrogen, methane, ethane, - 65 -Figure 7. C a p i l l a r y Column Analysis of the Products of the High Intensity Photolysis of Azomethane-Cyclohexadiene-1,4 Mixtures. Q O O o-Reproduction of a t y p i c a l analysis. C o u o c o > o o o CM o o O JL O "O c o -»-» a c L_ O I V O CM C9 C CJ N C CJ n « o c O X o u >^  u O CM o CM 0 B, time (min) Reproduction of the analysis of the products of the hydrogenation experiment. O o m I c o •+-> u CJ o c o s o u u o c o sz -So &«> /U > JZ E o c N c CJ -Q 10 time (min) 10 - 66 -benzene and two heavy products designated I and II which w i l l be shown to be the two products of recombination expected i n the reaction of methyl and cyclohexadienyl r a d i c a l s . C H 3 ^ / C H I I I The accessible a n a l y t i c a l data and certain other derived quantities be to^discussed have been presented i n Table IV. The reaction scheme C has been proposed to account f o r the experimental observations. This.scheme i s very si m i l a r to that found to be adequate i n many other investigations of t h i s type. At the high i n t e n s i t i e s employed i t i s supposed that a l l second-ary r a d i c a l s r e s u l t i n g from methyl attack on a substrate molecule react with another methyl r a d i c a l . At the moderate temperatures of the photolysis, such an assumption i n reasonable. The dispro-portionation step, 6d^ i s necessary to account f o r the s i g n i f i c a n t benzene y i e l d s found. CH 3N 2CH 3 + h? — I a ^ r > 2 CH 3 + N 2 — Ia^m-> C 2H 6 + N 2 2 CH 3 ^ ^2^6 CH 3 .+ CH3NNCH3 — > CH 4 + CH2NNCH3 GH 3 + CH2NNCH3 — > C 2H 5NNCH 3 CH 3 + CH3NNCH3 — > .(CH3)2NNCH3 Q CHJ + (CH 3) 2NNCH 3 — > (CH 3) 2NN(CH 3) 2 CHJ + C 6H 8-1,4 ^ > GH 4 + C 6H ? CHI + G.H4 6 c — > I 43 ' ~6ll7 • 6 ^ II 6d "> C H 4 + C 6 H 6 - 67 -The above scheme leads to the f o l l o w i n g r e l a t i o n s h i p s when analyzed a c c o r d i n g to the usual s t e a d y - s t a t e treatments k 6 R C H 4 ™ R G 6 H 6 M k4m T_ [ B ] ' M 4 where [A] = c o n c e n t r a t i o n of azomethane; [B] = c o n c e n t r a t i o n of cyclohexadiene-1,4. As CH^ r e a c t s very r a p i d l y w i t h c y c l o h e x a -diene-1,4, small r e a c t a n t p r e s s u r e s and h i g h i n t e n s i t i e s of i l -l u m i n a t i o n are r e q u i r e d . These c o n d i t i o n s cause a s i g n i f i c a n t consumption of cyclohexadiene i n the course of the run by the time enough products have accumulated f o r unequivocal a n a l y s i s . The v a l u e s of [B] i n T a b l e IV have been c o r r e c t e d f o r t h i s com-sumption by the f o l l o w i n g methods - [B] _t i n " 2 Vr R C 6 H 6 - 4{a]r4»6 where t = the r e a c t i o n time. Once again, the c o r r e c t i o n f o r ethane formed i n the molecular e l i m i n a t i o n p r o c e s s a t the r a t e 0mRN = 0*°12 rN L S N O T s i 9 i f i 0 3 1 " 1 ^ clue t o the l a r g e o v e r a l l y i e l d s of ethane. I n these experiments 0^ H ^ 0 . 5 5 . 13 y '2 6' Values of 10 k ^ / k j have been compiled f o r each experiment i n T a b l e IV and t h e i r l o g a r i t h m s are p l o t t e d as a f u n c t i o n of o 10 /I i n F i g u r e 8. The l e a s t mean squares treatment of the data y i e l d s the A r r h e n i u s expressions 1 0 1 3 ^ 6 = 1 Q ( 5 . 9 + 0 . 2 ) e x p _ (5.52 ± 0.35) k| RT As i n the case of the p h o t o l y s i s of pure azomethane, a t e n t -a t i v e c o r r e c t i o n f o r the pressure dependence of the recombination - 68 -- 69 -75 of methyl r a d i c a l s can be made using the data of Toby and Wiess. Replacing r e a c t i o n (2) w i t h the system of r e a c t i o n s 2 C H ' 2 > C 2 H * C 2 H 6 " > 2 C H 3 C 2 H ^ + C H 3 N N C H 3 Q 1> C H 3 N N C H 3 + C 2 H 6 i t i s r e a d i l y shown th a t each run i n Table IV must be m u l t i p l i e d by a c o r r e c t i o n f a c t o r , k 1 + 2 wh k c ( [A]+[B] )J ere we estimate k j 3 / k c from Toby and Weiss' data using the ap-proximate r e l a t i o n l o g ( k b A c ) = 18.2 - 0.44(10 3/T). When the runs are cor r e c t e d i n t h i s f a s h i o n there i s obtained the new Arrhenius expression 1 0 1 3 j ^ 6 _ = 1 Q ( 5 . 9 + 0 . 2 ) e x p (5.55 ± 0.32) 4 RT Thi s r e s u l t i s not d i f f e r e n t from the uncorrected one. The d i s p r o p o r t i o n a t i o n to combination r a t i o of methyl and cyclohexadienyl r a d i c a l s can be c a l c u l a t e d from the r e l a t i o n s h i p s A ( C H ' , C 6 H 7 ) = j. ^ - - = — k 6 c + k 6 c R 6 - R C 6 H 6 ( ? ) R C 6 H 6 ( R C H - ~w W R r H R - 2 R c H ) C H 4 k | L J C 2 H 6 C 6 H 6 Values of t h i s expression have been presented i n Table IV under the heading f o r each k i n e t i c run. T h i s method of determin-a t i o n of / \ ( C H 3 . C^Hy) does not make use of the experimentally a v a i l a b l e chromatographic peak areas of compounds I and I I obtained from the c a p i l l a r y column a n a l y s i s . Presumably, the r a t i o of the - 70 -areas of the benzene peak to the sum of the areas of peaks I and I I , m u l t i p l i e d by the appropriate c a l i b r a t i o n factor would give A ( C H 3 , C^Hy) d i r e c t l y . The lack of authentic samples of the combination products makes c a l i b r a t i o n of the detector impossible, so an independent determination of / \ ( C H 3 , C^Hy) by t h i s method i s precluded. The area r a t i o s from the c a p i l l a r y column analysis have a lower s t a t i s t i c a l standard deviation than the values from equation (7) and c l e a r l y arise from more d i r e c t experimental procedures. Accordingly, values of /\(2) are presented i n Table IV which have been found from these area r a t i o s using a c a l i b r a t i o n factor calculated from / \ ( l ) values i n the following way. The values of the area quotient Q, where Area(l) + Area (II) Q = Area(C 6H 6) were found from the c a p i l l a r y column analyses f o r each run. Q was found to have the value 2.92 + 0.81, independent of temperature. Each value of / \ ( l ) from equation (7) was then m u l t i p l i e d by the Q value f o r each experiment to give the quotient P = 0.793 + 0.172 for sixteen runs whose median value l i e s between 0.762 and 0.834. The mean value, P, was used as a c a l i b r a t i o n factor to recalculate values f o r /\(2) = P/Q f o r each run. In t h i s way we are able to u t i l i z e the more accurate c a p i l l a r y column a n a l y t i c a l data, but, of course, the actual measured value of /\(2) r e f l e c t s the re-sults obtained by use of equation (7). The mean value of / \ ( l ) d i f f e r s s l i g h t l y from the mean value of /\(2) because d i f f e r e n t data have been ommitted on s t a t i s t i c a l grounds from the f i n a l averaging but t h i s difference i s not s i g n i f i c a n t i n view of the ove r a l l errors. - 71 -The r a t i o of the areas of peaks I and I I has been used d i r -e c t l y to calculate the values f o r k^/k 1 1 l i s t e d i n Table IV. This i s reasonable insofar as i t i s unlikely that the response of the detector would be much d i f f e r e n t towards such si m i l a r compounds as these two isomeric recombination products. The logarithms of the values of /\(2) and of k^/k 1 1 have c o been presented as funtions of 10 /T i n Figure 9. As can be seen. no significant temperature trend i s evident i n these plots 0. 10 The value k^/k 1 1 = 0.76 + 16 can be compared to the sim i l a r C c ratio-found by Eberhardt et al_." L W i n r a d i o l y s i s experiments of cyclohexadiene-l,4- 1 4CH 3I solutions. In these l i q u i d phase ex-periments k^/k* 1 = 0.59 + 0.13 at room temperature i n reasonable L> w """"" agreement with present measurements, and substantiating the ob-served preference for the non-conjugated combination product. Since f a i r l y rapid addition of methyl r a d i c a l s to azomethane occurs, and causes a depression of the material balance, M = ( f W , + R~ „ - R~ u )/R.T , i t i s d i f f i c u l t to check f o r methyl CH 4 C 2H 6 C 6H 6 N 2 addition to cyclohexadiene-1,4 on a material balance basis. We can rule t h i s reaction out however, on the following grounds, ( l ) The reaction sequences CH- + CH 3 + :7c —> :7d—> + CH, (a) would cause an increase i n k^/k 1 1. In fa c t , t h i s r a t i o i s lower for the methyl r a d i c a l than for other r a d i c a l s studied 72 -Figure 9. Combination and Disproportionation of the Methyl Radical with the Cyclohexadienyl Radical. 0 O ® k 6 c / 6 c - 73 -here where addition does not occur. (See General Discussion) (b) would c e r t a i n l y lead to the product 4,5-dimethylcyclohexene which i s readi l y detectable on the c a p i l l a r y chromatography column, but was not found. (2) Dominguez and Trotman-Dickenson have examined the methyl, ethyl, isopropyl and t e r t - b u t y l addition reactions to acetylene, and found the methyl r a d i c a l to be no more reactive than the isopropyl r a d i c a l , which has been found not to add to cyclohexadiene-1,4. The detection of minute amounts of cyclohexadiene-1,3 i n the reaction products could be e a s i l y accomplished from the c a p i l l a r y column analysis, but was not, i n f a c t , observed. This indicates that there i s no mutual termination of the cyclohexadienyl rad-i c a l s since l a t e r work (See Low Intensity Photolysis of Azomethane-Cyclohexadiene-1,4 Mixtures) has shown that the 1,3-diene i s a product of such reactions, i e : 2 C6H* d i s P - > |Q) + ( ^ ) or comb. s. p „ ' ^ 12 14 l s o m e r s This knowledge adds support to the use of equation (7). Another in d i c a t i o n from the low in t e n s i t y work i s that C^ HA- r a d i c a l s decompose into hydrogen atoms and benzene above _ca. 136°. As the work done here does not extend beyond 117°, and as e f f i c i e n t removal of C^Hj r a d i c a l s by methyl termination w i l l cause t h i s species to have very short l i f e t i m e s i n t h i s system, interference from decomposition i s not anticipated. Special experiments made on Azomethane-Cyclohexadiene-1,4  system to i d e n t i f y Products I and II In a l l the experiments made at high r a d i c a l concentration on the abstraction of hydrogen atoms from cyclohexadiene-1,4, two heavy products appear i n the - 74 -c a p i l l a r y column chromatograms for a l l i n i t i a t o r r a d i c a l s except t e r t - b u t y l . In the case of methyl r a d i c a l s , these elute r a p i d l y , and are widely separated. As one proceeds towards t e r t - b u t y l rad-i c a l s the two products are retained to longer retention times and are less widely separated, u n t i l , f o r t e r t - b u t y l experiments, they are no longer resolved on the column. Of course, we suspect these compounds to be the representatives of the homologous series? ^ \ ^ R r ^ r ^ R i t I J and r I f o r R = CH 3-, C 2H 5~, C3Hy-, and C^Hg-I II In the methyl case, a hydrogenation experiment has shown that these products have the skelatal structure of methylcyclohexane. ( l ) Hydrogenation of the heavy products. 16 mm. of azo-methane and 4 mm. of cyclohexadiene were photolyzed.. at room temperature (23°) f o r 30 minutes. Following t h i s treatment the l i g h t products and azomethane were pumped off through a dry i c e -acetone trap and the remaining materials were frozen into 1 ml. n-pentane and removed from the vacuum system. Chromatographic analysis of t h i s mixture showed the t y p i c a l d i s t r i b u t i o n of pro-ducts, plus cyclohexadiene-1,4 as shown i n Figure 7A (no heptane, of course, was present i n t h i s sample.) The remaining solution was shaken f o r 30 minutes under a pressure of one atmosphere of hydrogen i n the presence of 10 mg. of Adam's hydrogenation catalyst, (Pt0 2) at 0°. 100 LU. of the product of t h i s treatment injected onto the c a p i l l a r y column gave the chromatogram shown schematically i n Figure 7B where the cyclohexadiene-1,4 has given r i s e to the strong cyclohexane peak and the previous benzene peak has been much enlarged. Run-ning the sample again using a c a p i l l a r y column with a hydrocarbon 75 -grease substrate demonstrated that t h i s l a t t e r peak contained both benzene and methylcyclohexane. Thus the compounds I and II were observed to be hydrogenated to methylcyclohexane. I t was concluded, therefore that I and II are unsaturated derivatives of methylcyclo-hexane. (2) Chromatographic retention times of I and II on the cap-i l l a r y column are 9.5 and 10.9 min. respectively (Figure 7A). This represents a s h i f t of 1.8min. between cyclohexadiene-1,3 and I and 2.0 min. between cyclohexadiene-1,4 and I I . Since both products I and II have the methylcyclohexane carbon skeleton and t h e i r v.p.c. retention times are consistent with methylated cyclohexadiene-1,3 and cyclohexadiene-1,4 respectively, there i s l i t t l e doubt that they have the structures I t w i l l be assumed that the higher homologues are analogously formed i n the cases of more complex i n i t i a t i n g r a d i c a l s . 3. Azomethane-cyclohexadiene-1,4 mixtures - low i n t e n s i t y  photolysis The very high r e a c t i v i t y of cyclohexadiene-1,4 towards the methyl r a d i c a l q u a l i f i e s i t as an e f f i c i e n t scavenger i f i t i s present i n s u f f i c i e n t concentration and i f the rate of photolysis of azomethane i s s u f f i c i e n t l y slow. Under these conditions a l l r a d i c a l s i n the system w i l l be replaced by the cyclohexadienyl r a d i c a l . This has been used here as a means of producing the cyclohexadienyl r a d i c a l to study i t s mutual interactions and to seek evidence f o r i t s unimolecular decomposition at higherAtmpCt^ I II - 76 -Some evidence f o r the reaction, CAHA > H* +. C AH A, has been found, and Arrhenius parameters, have been estimated f o r t h i s reaction, although i t i s clear that the mechanism of the process taking place i s quite complex. Experimental Photolysis of azomethane-cyclohexadiene-1,4 mixtures was done i n the same way as f o r the previous high i n -tensity studies, : except that much larger pressures of the diene and lower pressures of azomethane were used. These conditions are indicated i n Table V. Neutral density f i l t e r s of o p t i c a l d ensities 1.88 and 1.4 were used to attenuate the 3660 A. i l -lumination i s o l a t e d from the mercury arc lamp. Af t e r photolysis, analysis preceded generally as f o r the high i n t e n s i t y experiments. A l l products v o l a t i l e at -175° were pumped into the gas burette. These comprised methane, nitrogen and ethane. Also, minute traces of ethylene (quantum y i e l d , found. This i s of unknown o r i g i n and i s so small that i t has not been considered i n the mechanism. Ethylene l i k e l y a r i ses as a low p r o b a b i l i t y decomposition of photo-excited azomethane. The ethane-methane r a t i o was determined by chromatographic analysis, then the t o t a l product d i s t r i b u t i o n could be calculated from the r e l a t i o n R. , = ^ R r u + Rp „ . This r e l a t i o n uses the observation of the high i n t e n s i t y studies that methyl r a d i c a l s do not add s i g n i f i c a n t l y to the cyclohexadiene-1,4 molecule. Three mass spectrometric analyses (Table V) had s u f f i c i e n t -ly strong N 2 peaks to give accurate r e s u l t s i n spite of the natural nitrogen background. The r a t i o of the (corrected) peak heights at m/e = 16 and at m/e = 28 i s representative of the o about 0.0001 to 0.0005) were consistently TABLE V The Unimolecular Decomposition of the Cyclohexadienyl Radical (°C) (min.) r R x x 10 — 1 ? **4 3 V (molecules/cmfsec.) ^ 10 (mol/cm. sec.) 2 T t [e] b N 2 CH 4 R 6 C 2H 6 C6 H10 C 6Hg-l,3 C 6 H 6 k8 / / k7d' 63 1080 0.380 10.1 0.107 0.210 0.210 0.0016 0 0.0116 0.0306 - -73 1020 0.528 10.5 0.160 0.316 0.315 0.0019 0 0.00671 0.0493 - -80 1200 0.435 7.49 0.120 0.238 0.238 0.0012 0 0.0135 0.0403 -89 420 0.229 7.36 0.0974 0.193 0.193 0.0009 0 0.0101 0.0290 - -101 1110 0.363 7.025 0.111 0.219 0.218 0.0019 0 0.0122 0.04 05 - -118 240 0.470 5.86 0.388 0.751 0.749 0.0122 0 0.0499 0.150 - -123 180 1.13 3.79 0.847 1.658 1.646 0.0185 0 0.0664 0.358 - -136 a 1200 0.444 8.55 0.145 0.289 0.288 0.0006 0.0093 0.0174 0.0983 1.73 4.98 145 a 1080 0.340 8.25 0.105 0.208 0.208 0.0012 0.0248 0.0159 0.0946 5.45 13.9 151 a 1080 0.618 7.90 0.172 0.338 0.337 0.0028 0.0384 0.0244 0.179 6.60 17.4 157 1170 0.408 5.78 0.154 0.305 0.305 0.0012 0.0576 0.0226 0.178 10.5 27.2 162 420 0.514 8.93 0.380 0.752 0.751 0.0041 0.250 0.0861 0.575 28.8 60.2 -0 TABLE V (continued) (°C) (min.) r R x x 10 (molecules/cm.sec.) —4 y 10 (mol/cmfsec} 2 T t M b N 2 CH 4 R 6 C 2 H 6 C6 H10 C 6H 8-1, 3 C 6 H 6 k g/k| kg/k| d, 163 1200 1.258 8.93 0.383 0.761 0.757 0.0027 0.216 0.066 0.550 24.8 59.5 164 210 0.281 6.79 0.230 0.455 0.454 0.0022 0.134 0.0366 0.242 .19.9 49.5 165 1260 0.357 2.19 0.121 0.241 0.239 0.0007 0.108 0.0146 0.207 22.1 63.3 166 1140 0.434 2.97 0.160 0.316 0.315 0.0018 0.145 0.0245 0.259 25.8 65.0 166 a 1020 0.429 7.88 0.0972 0.189 0.189 0.0028 0.179 0.0247 0.275 41.1 80.0 166 180 1.046 7.806 0.780 1.537 1.529 0.0134 0.295 0.131 1.005 23.9 57.7 172 a 1170 0.446 7.40 0.139 0.277 0.276 0.0010 0.268 0.0298 0.366 51.0 110. 186 1080 0.355 6.30 0.173 0.341 0.34.0 0.0020 0.580 0.0201 0.600 99.5 290. Product analysis checked by mass spectrometry. b n = 10 = 10 ~, 7 Concentration of azomethane (molecules/cm.) 3 Concentration of cyclohexadiene-1,4 (molecules/cm.) - 79 -r a t i o of the pressures of methane and nitrogen. The mass spectro-meter was calibrated by examination of mass spectra taken during the high in t e n s i t y photolysis experiments of azomethane and cyclo-hexadiene-1,4, where the r e l a t i v e pressures of methane and n i t r o -gen were accurately determined a n a l y t i c a l l y . This indicated that the r a t i o of peak heights, h(l6)/h(28), could be m u l t i p l i e d by 1.528 to obtain r e l a t i v e pressures. The three determinations gave the following results? T R C H 4 / R N 2 (°C) Mass Spectrometry Normal Analysis 136 1.94 1.99 151 2.00 1.97 172 2.04 1.99 Independent analysis confirms, therefore, the adopted a n a l y t i c a l method up to the high temperature region. The l i q u i d residues were analyzed on the packed, polyethylene g l y c o l column at 50°, i n the usual way, providing measurements for cyclohexene, cyclohexadiene-1,3, benzene and f i n a l l y the star t i n g material, cyclohexadiene-1,4. Yields of these products have been presented i n Table V. I t i s possible to monitor the e f f i c i e n c y of the scavenging of the methyl r a d i c a l by measuring the rate of formation of l-methylcyclohexadiene-2,5 (II) whose chromatographic peak i s not too seriously i n t e r f e r e d with by the t a i l of the cyclohexadiene-1,4 peak i n the c a p i l l a r y column analysis. I t can be calculated from the disproportionation r a t i o s measured i n t h i s section that i f R C H -1 3 l s -,-we-'-ve times greater than RJJ, then the rate of loss of methyl r a d i c a l s i n recombination with the cyclohexadienyl r a d i c a l i s less than one percent of the t o t a l methyl r a d i c a l s - 80 -i n t r o d u c e d . T r a c e s of I I are found i n most runs; those i n which i t s chromatographic area i s g r e a t e r than one t w e l f t h t h a t of cyclohexadiene-1,3 have been r e j e c t e d . Treatment of the data A t 123° and below, there i s produced no cyclohexene, and the r a t i o of benzene t o cyclohexadiene-1,3 p r o d u c t i o n becomes constant, Under these c o n d i t i o n s , the system i s l i k e l y d e s c r i b e d by Scheme D, where e f f i c i e n t scavenging of methyl r a d i c a l s and u l t i m a t e t e r m i n a t i o n by c y c l o h e x a d i e n y l r a d -i c a l s i s co n s i d e r e d . A t the r e a c t a n t p r e s s u r e s used, a t t a c k of CH 3 on azomethane i s q u i t e n e g l i g i b l e , amounting t o w e l l under 1 % of a l l m e t h y l - ' r a d i c a l s produced. Values of R^ i n T a b l e V have been c a l c u l a t e d from Rg^ by s u b t r a c t i n g the methane t h a t a r i s e s from a t t a c k of methyl r a d i c a l s upon the azomethane molecule. T h i s i s r e a d i l y y y done from the measured val u e s of k 4 m / k | and k^/k^ l i s t e d i n Table XV. CH 3N=NCH 3 + h ^ — 2 ^ * a — > 2 CHJ + N 2 — 0m1 a — > C 2 H 6 + N 2 CH* + C 6H 8-1,4 2 C^Hy 6 -> GH 4 + C 6 H ? _Zd! > (11 ±2%) D 7d" + [I I) (20 + 8 %) 7c -> C 1 2 H 1 4 i s o m e r s (69 + 6 ^) The ethane produced a r i s e s from the mol e c u l a r e l i m i n a t i o n p r ocess p r e v i o u s l y d e s c r i b e d , and a q u a n t i t a t i v e estimate ( o r at l e a s t an upper l i m i t , i f b i m o l e c u l a r recombination of CH 3 i s not completely quenched) i s a v a i l a b l e . I f the o v e r a l l quantum y i e l d of n i t r o g e n i s u n i t y , then the quantum y i e l d of mol e c u l a r - 81 -elimination i s given by the equation: l o g 0 m = l o g^ RC 2H 6 /' RN 2 ) =2.08 + 0.4 independent of temperature, where the estimate i s obtained by the s t a t i s t i c a l analysis of 20 experiments at reactions temperatures from 63 to 186°. These r e s u l t s are given as an Arrhenius p l o t i n Figure 10B, and are i n good agreement with independent e s t i -71 73 mates based upon the use of oxygen, sulphur dioxide and cyclo-74 hexadiene-1,4 as scavengers, and upon the k i n e t i c analysis of 67 the photolysis of a mixture of azomethane and azomethane-d^. The mean value of 0.012 i s c l e a r l y i n general agreement with the re s u l t s of the other studies, and the greater scope of t h i s i n -vestigation increases the weight of the evidence that the quantum y i e l d of the molecular elimination process i s independent of temp-erature between 25 and 186°. Since temperature apparently neither quenches nor enhances the molecular elimination process, i t pro-' bably arises through a highly excited, upper ele c t r o n i c state where thermal energy i s of l i t t l e s i g n i f i c a n c e . This i s consis-tent with the fa c t that the o v e r a l l quantum y i e l d , 0(N2), i s unity over a range of temperatures i n the vapour. Since Rr „ /R, i s le s s than one-half, i t i s apparent that ° 6 M6 ° recombination occurs and accounts f o r an appreciable f r a c t i o n of the C^Hy r a d i c a l s . S t a t i s t i c a l treatment of the low temperature r e s u l t s y i e l d s ( k 7 d i + k 7 d n ) A 7 c = 0.45 + 0.12 and k 7 d , / ( k 7 d , + k 7 d „ ) = 0.36 + 0.08 at the 5 % p r o b a b i l i t y l e v e l of error. With the data at the head of the following page, and from the r e l a t i o n , ( k ? d , + k 7 d „ ) 2 ( R c H /R.6) we k 7 c 1 " 2 <t . H 6 6 f i n d (k 7 d„ + k 7 d „ ) / k 7 c - 0.45 + 0.12. This r e s u l t has not been - 82 -previously reported. T(°C) R C 6 H 6 / R 6 RC 6H 8-1,3 / R6 63.1 0.1445 0.0552 72.5 0.1562 (0.0214) 79.8 0.1691 0.0567 89.2 0.1507 0.0525 100.9 (0.1853) 0.0558 Avg. = 0.155 + 0.029 0.0551 + 0.0078 k7d' ( RC 6H 8-1,3 / / R6) We can evaluate .= = 0.356 k7d' + k7d" ^ C ^ ^ ^ Assessment of the error proceedes as follows: ^ ^ 3 - 1 , 3 ^ 6 d ( ^ C 6 H 6 \ d ( R C 6 H 8 - l , 3 / / ' R 6 ) ^ R G 6 H 6 / R 6 ) 2 \ R 6 / R c 6 H 6 / R 6 = 0.166 at 0.25 % l e v e l = 0.075 at 5 % l e v e l . The r e s u l t , 0.36 + 0.08, can be compared to Eberhardt's re-sult of 0.27 + 0.04 from r a d i o l y s i s of l i q u i d cyclohexadiene-1,4. 1 0 Agreement i s reasonable, although the condensed phase apparently yi e l d s less of the 1,3-diene. As the temperature r i s e s above 120° (Figure 10A), the y i e l d of benzene increases with respect to methane, and we begin to de-tect cyclohexene. As the cyclohexadienyl r a d i c a l i s highly act-ivated towards thermal decomposition, the observed behavior may be caused by the extra reactions given i n Scheme E. This se-quence i s part of the mechanism proposed f o r the r a d i o l y s i s of l i q u i d cyclohexadiene-1,4, 1 0 and reactions (9) and (10) have been shown to be rapid i n that system, where hydrogen atoms are generated 83 -Figure 10. Temperature dependence of the yiel d s of the products 1.6 CD 1.0 cr 0.4-_i_ CD I OJ cr + CM I S cD I CM L) CD o a 2.1 — r 1 1 A x =CeH10 0 X = C 6 H 6 a X=C§HQ-1,3 T © _ _ . © _ . ©.0 — — — J Q-—•• g • J I I mBm _L 3D j_ i " — 1 1 r — A Rebbert and Ausloos, J. Phys. Chem., 67, 1925 (1963 V Rebbert and Ausloos, J. Phys. Chem., 6 6 , 2253 (1962) a Thynne (unpublished work) • Herk, Feld and Szwarc, J . Ara. Chem. S o c , 83, 2998 2.6h © 2.2^- ° o 38K 34LJ (1961L o © G 2.2 2.6 3.0 3.4 103/T r a d i o l y t i c a l l y . C 6 H 7 ^ H ° + C 6 H 6 AH = 2 7 k c a l . 8 9 9m \ 1 0 -> 84 H* + || II 9 > [ II AH = - 3 8 k c a l . rr AH = - 3 3 kcal . ~ 6 " 7 + C 6 H ? AH = - 2 3 'kcal. The thermochemical values quoted have been calculated from stan-dard values and the following data. X H°(X) Source 0 4 5 Measured i n t h i s work. 2 6 Cox, Tetrahedron, 1 9 , 1 1 7 5 ( 1 9 6 3 ) . 2 7 AH^(cyclohexane) - 2 H^y^fcyclohexene) Harrison et a l . , J . ACS, 8 7 , 5 0 9 9 ( 1 9 6 5 ) . (^j) 4 1 Assumes H + ( M ) — ^ ' ( ^ j ) ^° ke 3 8 kcal./mole exothermic. , Examination of the gaseous products of several representative runs (indicated i n Table V) by mass spectrometry showed that the rate of molecular hydrogen production was l e s s than one per-cent of the methane, rate. Reaction (9m) does not appear to be very important, therefore, and i t s h a l l be assumed that i t does not occur at a l l . We ultimately wish to calculate Arrhenius para-meters f o r r e a c t i o n . ( 8 ) . I t w i l l then be shown that the neglect of (9m) causes no s i g n i f i c a n t error as long as hydrogen production i s below one percent of the methane production. The o l e f i n i c linkages of cyclohexadiene-1,4 are apparently much more reactive towards hydrogen addition than the methylenic - 85 -hydrogen atoms are towards metathesis. In view of the very low 84 activation energies of hydrogen atom addition to o l e f i n s , t h i s conclusion i s not unreasonable. A similar pattern of r e a c t i v i t y i s shown by the in t e r a c t i o n of the hydrogen atom with propene; the rate constants f o r metathesis and addition stand i n the r a t i o of 0.04 at 25° and less than 0.08 at 2 4 0 ° . 8 5 - 8 7 The predominance of the addition process with propylene i s probably due to the 84 favourably low ac t i v a t i o n energy of 2.2 kcal./mole. The pre-dominance of the addition process with cyclohexadiene-1,4 may be due to a favourable difference i n the ac t i v a t i o n energies, Egm - Eg = 3 kcal./mole. The ac t i v a t i o n energies f o r the metathe-s i s of the methyl, ethy l , isopropyl and t e r t - b u t y l r a d i c a l s with cyclohexadiene-1,4 are indistinguishable at 5.5 + 0.3, 5.8 + 0.1, 6.5 + 1.0 and 5.3 + 0.8 kcal./mole respectively (See Table XV), ind i c a t i n g that the strength of the i n c i p i e n t bond has l i t t l e influence on the value of the ac t i v a t i o n energy. Let us tentative-l y assign the values: Eg = 2 kcal./mole, Egm = 5 kcal./mole and A 9 = A 9 m » w e then obtain the r a t i o k^/k^ = 0.03 at 164°, i n qu a l i t a t i v e agreement with the observed behavior. Reaction (9), therefore, accounts f o r a l l of the hydrogen atoms released, form-88 ing a secondary r a d i c a l , unstabilized by resonance. Slaugh has found t h i s species not susceptible to structural isomerization. The cyclohexen-4-yl r a d i c a l i s s t r u c t u r a l l y related to the i s o -propyl r a d i c a l which has been found (Table XV) to have similar r e a c t i v i t y towards C^Hg-1,4 as the methyl r a d i c a l . Inspection of the l i q u i d products by c a p i l l a r y column chromatography shows that the products, CH^-C^Hy, (both isomers I and II) are i n h i b i t e d strongly by the C,Hp=l»4 scavenger, therefore no reason e x i s t s - 86 -to believe the cyclonexen-4-yl r a d i c a l should i n t e r a c t with the cyclohexadienyl r a d i c a l i n t h i s system. Thus a l l termination i s by reactions (7c) and (7d), as i s expected i n the presence of an e f f i c i e n t scavenger. The reactions of Scheme E are the propagation steps of a chain reaction. That a chain i s involved i s indicated by the fa c t that FL, „ /Ft, becomes greater than unity at 186°. C 6 H 1 0 6 Some possible, alternative reactions can be shown to be un-l i k e l y . I t i s not easy to vary the pressure of cyclohexadiene-1,4 over a very wide range, therefore experimental evidence i s lack-ing to disprove the bimolecular process, if-l + i f ^ j ] 8 b — ^ \ A H = - i i k c a i . which yie l d s the same products as reaction(8). Hydrogen atoms are not generated i n t h i s process, so that the formation of molec-ular hydrogen need not a r i s e . The value of Eg^ obtained by a k i n e t i c analysis based upon the complete suppression of reaction (8) by reaction (8b) would not d i f f e r s i g n i f i c a n t l y from the value f o r Eg of 31+5 kcal./mole that we s h a l l see characterizes 17 the formation of cyclohexene. Using a mean value of 5 x 10 molec./cm?; fo r the concentration of cyclohexadiene-1,4, we can calculate that an absurdly high value of 1G litre/mole sec. would be necessary f o r Ag^ i n order f o r t h i s reaction to have the observed rate. Should reaction (8b) compete at a l l , i t must be at most a minor process. The reaction i « ] + [ M l ——> f ^ l + f ^ l AH = - i k c a i . - 87 -may contribute to the observed increase i n PL, H , -/R^,(Figure 10A) 6 8 ' I t i s unlikely that i s less than the value of about 5.5 .. . kcal./mole observed f o r the reactions R* + (Q) > RH + C^Hy where R* i s an a l k y l r a d i c a l ; methyl, ethyl, isopropyl or tert-butyl.and may be much greater. We would expect a much sharper r i s e i n the r a t i o Rn „ , 0/R~ u than i s observed, hence 6 8" ' 4 reaction (13) i s not a l i k e l y explanation f o r t h i s e f f e c t . The d i r e c t p y r o l y s i s of cyclohexadiene-1,4 has been reported 89 by Frey. The reaction, C^Hg-1,4 > C^H^ + H 2 was found to have the rate constant k = 10 J- 2* 0 2exp(-42,690/RT) sec" 1. At 500°K. and assuming C^Hg-1,4 = 5 x 1 0 1 7 moleo./cm?, the rate 12 3 of t h i s reaction would be 0.017 x 10 molec//cmTsec., about one hundred times less than the benzene y i e l d from other sources. Py r o l y s i s experiments were done i n t h i s work to determine thermal s e n s i t i v i t y of the reactant, and none was found. The molecular hydrogen elimination — > H 2 + ft ] - AH = 26 kcal. i s not important as H 2 i s not detected i n the gaseous products. Several runs have been performed at 164 + 1.7°, where the rate of i n i t i a t i o n has been varied over as wide a range as i s convenient witnout breaking down the assumption of t o t a l scaveng-ing. Figure 11 shows.that the benzene a r i s i n g from dispropor-tionation, i e . , RQ u ~ R C H ' a n c* ^ e c y c l ° n e x a d i e n e - l , 3 are 6 6 6 10 generated at a rate proportional to the, rate of i n i t i a t i o n , R^. This i s consistent with the assumption that they form i n a reac-t i o n bimolecular i n the cyclohexadienyl r a d i c a l . We conclude that RR u i Q and RR U - R R u are r e l i a b l e measures of the square C 6H 8-1,3 C 6 H 6 ^ 6 H I 0 of the concentration of O ^ H j . Figure 12 shows that cyclohexene - 88 -ure 11. K i n e t i c Order of Cyclohexadiene-1,3 and Benzene Production. Q 8 i - 89 -Figure 1 2 . K i n e t i c Order of Cyclohexene Production. - 90 -i s produced at a r a t e p r o p o r t i o n a l to the square r o o t of the r a t e s of p r o d u c t i o n of cyclohexadiene-1,3, of the benzene from d i s p r o -p o r t i o n a t i o n and the r a t e of i n i t i a t i o n . T h i s i n d i c a t e s t h a t the r a t e c o n t r o l l i n g p rocess i n cyclohexene p r o d u c t i o n i s f i r s t o rder i n the c y c l o h e x a d i e n y l r a d i c a l c o n c e n t r a t i o n . To t h i s ex-t e n t , the experiments at 164° s u b s t a n t i a t e the mechanism of Schemes D and E. I n s p e c t i o n of the slope of the benzene p l o t of F i g u r e 11 i n d i c a t e s t h a t F^/R^ = (^7^? + ky^,, ) / ( k y d , + k y d „ + k y c ) = 0.90, which i s c o n s i s t e n t with ( k y d , + k y d n ) / k y c = 9.1. Below 101°, t h i s r a t i o was found to be ( k y d , + k y d „ ) / k y c = 0.45 + 0.12; i t i s not l i k e l y t h a t such a strong temperature c o e f f i c i e n t e x i s t s f o r t h i s d i s p r o p o r t i o n a t i o n r a t i o . I n F i g u r e 13A and 13B the l o g a r i t h m s of the d i s p r o p o r t i o n a t i o n r a t i o s assuming Schemes D and E have been p l o t t e d a g a i n s t 10 /T. There i s a strong i n -crease i n the d i s p r o p o r t i o n a t i o n products w i t h r i s i n g temperature. E v i d e n t l y , Schemes D and E do not cover a l l of the processes of the mechanism. A number of m o d i f i c a t i o n s c o u l d be made to account f o r the i n c r e a s i n g k y d / k y c r a t i o s . The combination p r o d u c t s ^^.2^14 might be u n s t a b l e , l e a d i n g back to benzene and c y c l o h e x a d i e n e s . Two p o s s i b i l i t i e s f o r t h i s i n c l u d e a u n i m o l e c u l a r c o n v e r s i o n , w i t h s i m i l a r r e a c t i o n s f o r other C J ^ H J ^ isomers. A simpler scheme would r e q u i r e only the r e v e r s a l of ( 7 c ) . C 1 2 H 1 4 : > 2 C 6 H 7 Such a r e v e r s e r e a c t i o n would cause the observed r i s e i n benzene with concomitant r i s e i n CgHg-1,3. - 91 -• - 92 -To have a h a l f - l i f e of 10 hours at 434°K., the a c t i v a t i o n energy of r e a c t i o n (-7c) must be 35 kcal./mole, assuming A_ 7 c= 1 3 - 1 10 sec. T h i s would correspond t o a weakening of the i n t e r -annular C-C bond by about 45 k c a l . due t o the development of the cyclohexadienyl resonance energy i n each of the C^Hy fragments. As we s h a l l see, t h i s i s c o n s i s t e n t w i t h the estimate of 24 kcal./mole resonance energy obtained from measurement of the a c t i v a t i o n energy of r e a c t i o n ( 8 ) . P y r o l y s i s of the dimers at 160° was confirmed by a simple 17 experimento A gaseous mixture of azomethane (0.824 x 10 molecules/cmf) and cyclohexadiene-1,4 (7.50 x 10 molecules/cmf) were p h o t o l y z e d r at 59° f o r 18.0 hours t o accumulate the dimers; experience i n d i c a t e d t h a t no cyclohexene should form under these c o n d i t i o n s . The v o l a t i l e products and azomethane were removed by f r a c t i o n a t i o n at -78° and methane, ethane and n i t r o g e n were measured. The r e s i d u e , i n c l u d i n g dimers, was heated i n the dark at 160° f o r 23 hours. Cyclohexene was subsequently detected i n the products and the r a t i o s of t o t a l y i e l d s of products were C 6 H I 0 / C H 4 » 0.27, ( P 6 H 6 - C 6 H 1 0 ) / C H 4 = 0.15. T h i s i s c o n s i s t e n t w i t h thermal decompositioh of dimers t o cycl o h e x a d i e n y l r a d i c a l which, being at low co n c e n t r a t i o n , y i e l d e d cyclohexene and ben-zene by the sequence of r e a c t i o n s ( 8 ) , ( 9 ) , (10). A s i m i l a r experiment w i t h p o s t - i r r a d i a t i v e heating at 174° gave C ^ H ^ Q / C H 4 » 0.125, and ( C 6 H 6 - C 6 H 1 0 ) / C H 4 *» 0.31, suggesting t h a t the de-composition may be more complex at higher temperatures. Of course, under these c o n d i t i o n s , w a l l r e a c t i o n s and s i m i l a r com-p l i c a t i o n s render the q u a n t i t a t i v e i n t e r p r e t a t i o n of the e x p e r i -ments u n c e r t a i n , but the p y r o l y s i s , («-7c), does seem to be im-p l i c a t e d . - 9 3 -To eliminate side e f f e c t s caused by t h i s p y r o l y s i s , a r e c i r -culatory system might be useful, where the accumulating £±2^14 dimers could be trapped out of the system before they caused interference. A l t e r n a t i v e l y , an independent study of the p y r o l y t i c reaction, (-7c), could provide data needed for the k i n e t i c analysis, Indeed, pyroly s i s of these compounds i s in t e r e s t i n g i n i t s own ri g h t , since the bond d i s s o c i a t i o n energy of the interannular C-C bond can be related to the cyclohexadienyl resonance energy. 9 0 A possible synthetic route to the dimers has been found. Without the r e s u l t s of a study of t h i s p y r o l y s i s , quantitative treatment of the system i s very d i f f i c u l t because py r o l y s i s of the dimeric products makes the steady-state assumption i n v a l i d . As recombination products accumulate, they become an increasingly important source of C^Hy r a d i c a l s u n t i l f i n a l l y , a new steady-state exists when Ry c = R-_yc° The i n i t i a l and steady-state concentrations of C^Hy are then, respectively -WJin = R6/< k7c + k7d' + k 7 d » ^ "Wlss = R6/< k7d- + k 7 d " ^ Therefore, since (ky d, + k 7 d M ^ k 7 c = °* 4 5> the C^Hy concentration increases by about 8 0 % under the worst conditions. i n V k7d' + k7d" [ C6 H7 ss = / k7c * k7d' * k7d " Y = ± Q The fact that, at 164°,(Figure 11), the disproportionation reactions (7d) account f o r 9 0 % of the injected C^Hy r a d i c a l s indicates that the equilibrium between O^Hj and Cj_2^i4 i s n e a r l y established, and a new steady-state regime e x i s t s . If so, RC H -1 3 i s a 9 a i n a r e l i a b l e index of jC^Hy'j2. The values presented i n Table V for R^, the rates of form-- 94 -a t i o n of the various products, are a c t u a l l y c a l c u l a t e d by d i v i d -i n g the t o t a l product y i e l d by the r e a c t i o n time. As such, then, they represent mean r a t e s averaged over the r e a c t i o n p e r i o d . I n t h i s way, the r a t e of cyclohexene production as measured i s properly w r i t t e n RQ ^  = kg ^ [C^HyJ^ , and the r a t e r a t i o kg/ky^, becomes k ^ , - ( R c ^ / * ^ F o r a slowly changing r a d i c a l c o n c e n t r a t i o n such as we have here, the c o r r e c t i o n r a t i o of the mean r a d i c a l c o n c e n t r a t i o n to the r o o t -mean-square r a d i c a l concentration i s not l i k e l y to be of impor-tance. Neglecting f o r the moment the p y r o l y s i s r e a c t i o n , -7c, and analyzi n g the system of Schemes D and E according to a steady-s t a t e treatment leads t o the f o l l o w i n g simple r e l a t i o n s h i p s ? k 8/k| d, = R c ^ / ^ ^ H g - l . S ( 8 ) k 8 / k | = R c 6H 1 ( / R 6 ( 9 ) v4 • - W ^ C 6 H 6 - ( 1 0 ) where k 7 d = k 7 d , + ky d„ and k 7 = k y d , + k7d,, + kyG» I t can be shown tha t equations (8) and (10) should be m u l t i p l i e d on the r i g h t by a f a c t o r ^[c^Hy] " ^ / ^ ^ T ^ to c o r r e c t f o r the p y r o l y s i s of the dimers. Equation (9) r e q u i r e s the more complex c o r r e c t i o n f a c t o r , C 6 H 7 7 7 \ -7c where t i s the r e a c t i o n time, t o compensate f o r the d r i f t i n the r a d i c a l c o n c e n t r a t i o n . Since the r a t e constant, k„y c i s not known, such c o r r e c t i o n s are not p o s s i b l e to c a r r y out; however, we suspect that they are not of larg e numerical magnitude. A - 95 -rough check of t h i s p osition would be afforded i f the measured acti v a t i o n energy obtained from equation (9) were to agree with the other two when these expressions are plotted i n the Arrhenius form. The experimental points f o r equations (8) and (9) have been plotted i n Figure 14. Least mean squares analysis of these r e s u l t s gives the Arrhenius rate constants? k 8 / k 7 d « = 1 0 ( 2 i ° ° 1 i , 5 ) e x p -(30.4 + 2.9)AT moled? kg/k^ o l O * 2 1 ' 0 ±. 2' 4>exp -(31.2 + 4.7)/RT c c* s e <# Results for equation (10) are inaccurate i n three cases because the benzene y i e l d approaches the cyclohexene y i e l d . However similar treatment gives k 8 / ( k 7 d , +• k7d,,) = 1 0 ( 2 2 " 9 -± 2 - 4 ) e x p -(34.8 + 4.9)/RT The reasonable agreement between these three determinations supports the mechanistic i n t e r p r e t a t i o n . Analysis of representative runs indicates that hydrogen gas production i s less than one percent of methane production. Using t h i s as an estimate f o r the maximum perturbation caused by un-analyzed hydrogen, we derive M m a x ) 1 R c 6 H 1 0 * 0 , 0 1 & G H 4 = 1 Q(20.2 + 1.3) /(28.9 ± 2.6)\ /2 & ^2 \ Rl J K7d' KC 6H 8-1,3 x 7 M m a x ) _ R c 6 H 1 0 + Q ' 0 1 RCH 4 = i Q(20.1 + 2 . 2 ) e x p /(29.5 ± 4.2)\ " R\ " \ Kl J Thus the error i n neglecting hydrogen i s not s i g n i f i c a n t at the 5 % l e v e l . Discussion Within the l i m i t a t i o n s discussed above, the rate constant f o r the decomposition of the cyclohexadienyl r a d i c a l has been measured with respect to the disproportionation of the cyclo-- 96 -qure 14. Unimolecular Decomposition of the Cyclohexadienyl Radical - 97 -hexad i e n y l r a d i c a l . We s h a l l accept the r e s u l t , kg/k^ = 1 0 ^ 2 1 ' 0 i 2 ° 4 ^ e x p -(31.2 + 4.7)/RT (cm?/molecule s e c . ) ^ 84 as r e p r e s e n t a t i v e . Yang's r e s u l t f o r the back r e a c t i o n k_ g = i o ~ 1 2 - 2 exp~3.5/RT (cm 3/molecule sec.) w i l l be of use i n i n t e r p r e t a t i o n of the measurement. T h i s r a t e constant can be w r i t t e n i n the A r r h e n i u s form k_ 8 = Q2 1 0 ~ 1 2 ' 2 < T ) ^ exp(-3.5 - ^R<T> )/RT (cm 3/molecule sec.) = 2.3 x 10""11 exp(-3.9/RT) (cm 3/molecule sec.) i f <T> = 434°K. From the a c t i v a t i o n energy, Eg = 31.2 + 4.7 kcal./mole f o r c y c l o h e x a d i e n y l r a d i c a l decomposition, and E_g = 3.9 kcal./mole 84 f o r the a d d i t i o n of hydrogen atoms to benzene, we c a l c u l a t e the heat of the o v e r a l l r e a c t i o n , 8 to be /SHQ = 27.3 + 4.7 kcal./mole. The resonance energy, Q, of the c y c l o h e x a d i e n y l r a d i c a l can be c a l c u l a t e d from the f o l l o w i n g , thermochemical c y c l e . H" + k 94.5 26 3.5 .+• Q 98 2 H* + 2x52.1 19.8 A 14 ( a l l heats expressed i n k i l o c a l o r i e s / m o l e . ) Here we have adopted D°(secondary C-H) = 94.5 kcal./mole A H j(C 6 H g-l,3) = 26 k c a l . / m o l e 9 1 and other standard v a l u e s . 9 2 The r e l a t i o n 3.5 + Q = D ° ( C 6 H 6 ~ H ) = 31.2 + 4.7 - 3.9, g i v e s Q = 23.8 + 4.7 kcal./mole. T h i s value i s i n complete agreement 13 w i t h Benson's r e c e n t suggestion t h a t the resonance energy should l i e between 23 and 25 kcal./mole. Furthermore, 23.8 - 98 -kcal./mole weakening of the methylenic C-H bond of cyclohexadiene-1,4 would suggest a bond energy of 94.5 - 23.8 = 70.7 kcal./mole. This quantity has been reported to be 74 kcal./mole. We can reconcile the present r e s u l t s for the decomposition reaction with the Arrhenius parameters of the back reaction. To make the c a l -culations for t h i s comparison, we must have information about k 7 and about the o v e r a l l entropy change of the reaction. Recent 32 experiments of Rabinovitch indicate lower s t e r i c factors f o r recombination of delocalized r a d i c a l s than are common for simple a l k y l r a d i c a l s . Thus the s t e r i c factor for methyl plus a l l y l -3 -1 recombination i s 5 x 10 compared with ca_. 10 f o r a l k y l rad-i c a l s . Trotman-Dickenson has calculated a s t e r i c factor of -3 3 x 10 f o r methyl plus benzyl recombination and recent exper-93 iments substantiate a low value. The phenomenon may be more 34 pronounced f o r recombination of two delocalized species. Cer-V 21 3 14 t a i n l y the value of kg/k.2 = 10 (cmf/molec. sec.) 2 i s very high —10 33 and i f p 7 (, =0.1 and therefore <k7 = p 7 c Z ( k 7 / k 7 c ) = 10" cm./molec.sec. at 434 K., then the pre-exponential factor Ag has the improbably high value of 1 0 1 5 * 8 4 sec." 1. Such a high pre-exponential factor i s unlikely f o r a decomposition where an atom i s liberated which cannot a t t a i n extra, r o t a t i o n a l freedom i n the t r a n s i t i o n state. Indeed, as we s h a l l see below, calcu-l a t i o n of Ag from A mg measured i n the study of hydrogen atom addition to benzene indicates a value of A g ^ l O 1 3 ' 9 sec." 1 The sit u a t i o n i s improved i f the decomposition of C^2^14 13 -1 isomers has a "normal" pre-exponential factor, around 10 sec. . -3 I f t h i s i s the case, then p 7 > w i l l be around 10 and Ag w i l l 14 84 -1 become 10 * sec. ; s t i l l about two powers of ten above the 13 -1 +o 4 -1 "normal" 10 sec. , but well within the error of 10-^" sec. - 99 -I t i s possible to calculate the value of A R from the measure-ment 8 4of A_ 8 = i o " 1 0 , 6 4 cm?/molecules sec. = 1 0 8 , 5 9 (atm s e c . ) " 1 at 434°K. Reaction of an atom with a molecule r e s u l t s i n con-version of two r o t a t i o n a l motions into vibrations and a s t e r i c —2 94 factor of about 10 i s expected. Assuming a mean c o l l i s i o n diameter of 3.5 A*., the c o l l i s i o n y i e l d of hydrogen atoms with benzene at 434 K. i s 1.2 x 10™ cm./molec. s e c ; Yang's temp-erature dependent pre-exponential factor, I O 1 1 * ^ ^ = 1.4 x io™ 1 1 cm./molec s e c , i s therefore i n agreement with a s t e r i c factor of -2 10 . We may use t h i s r e s u l t i n conjunction with an estimate of the o v e r a l l entropy change of the equilibrium, C.HA - 8 > H* + C,H A 6 7 -< _g 6 6 at 434°K., to calculate a value of Ag that i s compatible with A_ 8 = i o " " 1 0 * 6 4 cm3/molec. sec. 5 8 9 5 Since the C^ HA- r a d i c a l i s planar, ' i t s t r u c t u r a l l y re-sembles benzene. Correcting the standard entropy of benzene (64.3 eu./mole) from the symmetry number twelve to two, and add-ing a spin degeneracy contribution R ln2 gives an estimate of the cyclohexadienyl r a d i c a l ' s entropy. We obtain S°(C^Hy) = 69.2 eu./mole at 298°K. The entropy gain i n the forward d i r e c t i o n i s then 64.3 + 27.4 - 69.2 + 1.9 = 24.4 eu./mole a f t e r correcting to 434°K., and assuming equal s p e c i f i c heat capacities f o r benzene and f o r the cyclohexadienyl r a d i c a l . Using t h i s value and the equation, A 8/A_ 8 = i o ^ S / 4 - 5 7 5 (atmosphere s ) " 1 , 13 9 -1 we obtain the r e s u l t A Q = 10 sec. . This i s probably an upper l i m i t since most l i k e l y we underestimate the entropy of the cyclohexadienyl r a d i c a l by r e l a t i n g i t to the entropy of the s t r u c t u r a l l y very r i g i d benzene molecule. Examination of the 100 -addition reaction, therefore, does not indicate a high pre-exponential factor. This calculated pre-exponential factor i s in reasonable agreement with the measured value. The value of 24 kcal./mole f o r the resonance energy of the cyclohexadienyl r a d i c a l i s much lower than that predicted by 12 Fisher using molecular o r b i t a l and valence bond calculations, which give 29.3 and 30.0 kcal./mole respectively. Electron spin resonance investigations indicate that the unpaired electron of 96 97 the phenoxy and of the benzyl r a d i c a l resides within the ring to a substantial degree, thus resonance forms with the cyclohexadienyl structure cannot be of extremely high energy with respect to the benzene structure. From a p r a c t i c a l viewpoint, the 24 kcal. of resonance energy may be compared with the a l l y l resonance energy of 13 kcal./mole? Reactive C-H bonds are thus weakened about twice as e f f e c t i v e l y with two adjacent o l e f i n i c linkages than^one, the 2 kcal. d i s -crepancy being explained by s t r a i n energy i n the CAij system. This i s at least an i n t u i t i v e l y appealing r e s u l t and may be com-pared with the o b s e r v a t i o n 1 1 that the benzene resonance (37 kcal./mole) i s approximately three times the a l l y l r a d i c a l s resonance energy. The resonance energy of 24 kcal./mole f o r the cyclohexadienyl r a d i c a l , however, i s substantially greater than the 15.5 kcal./mole resonance energy that has been measured f o r the open chain, penta-13 dienyl r a d i c a l . The difference may r e s u l t from the extra i n t e r -bond weakened by 13 kcal./mole H bond weakened by 24 kcal./mole - 101 -action involving the methylenic carbon atom that i s not possible fo r the open chain system. The electron spin resonance spectrum shows that the coupling constant to the methylenic hydrogen atoms i s 50 Oe., whereas only 30 Oe. would normally be observed f o r such protons i f they were not s p e c i f i c a l l y involved i n the 12 99 system. Hanazaki et a l . have assumed a hyperconjugative i n t e r a c t i o n between the TT system and the methylenic hydrogen atoms during quantum mechanical treatment of the cyclohexadienyl 12 cation. Fisher has treated the bonding of the cyclohexadienyl r a d i c a l as a superposition of the valence bond structures: K> HO HHO- W> * I II I I I • *IV V •H2Q • -H 2 D VI VII and finds that the structures VI and VII have significant repre-sentation, and that the unpaired electron i s to be found upon the methylenic hydrogen system accounting for the large observed coupling constants. I f the above in t e r p r e t a t i o n i s correct, the resonance energy of the cyclohexadienyl r a d i c a l i s not t y p i c a l of a simple 5 electron system and the observation that the 24 k c a l . resonance energy i s approximately twice the a l l y l resonance energy and two-thirds the benzene resonance energy w i l l bear no t h e o r e t i c a l significance. As we s h a l l see, i n t e r a c t i o n of the cyclohexadienyl r a d i c a l with an a l k y l r a d i c a l , eg. the isopropyl r a d i c a l , does not produce either of the cyclohexadienes as a product; the disproportionation - 102 -always involves the transfer of a methylenic hydrogen atom from the cyclohexadienyl r a d i c a l to produce benzene. I f the cyclo-hexadienyl free valence i s strongly delocalized onto the methyenic hydrogen atoms, then these may well be abnormally activated and the product d i s t r i b u t i o n w i l l contain no cyclohexadienes. T r o u g h t o n 1 0 0 has recently measured several decompositions of free r a d i c a l s under similar conditions to those of t h i s work. The systems studied can be generally represented as CH3CH2CH2CHCH2-X > CHgCHjCHgCH-CHg + X' where X = -OH, -GCCHg, -0CC2H5, -0C 2H &, - O A l l y l , - A l l y l , and -8cH. Arrhenius parameters measured showed that, i n general, reaction heats were close to the measured a c t i v a t i o n energies. In contrast to the r e s u l t s f o r the cyclohexadienyl r a d i c a l de-composition, the pre-exponential factors were uniformly low, about 10 1 1 sec." 1, f o r each substituent, except f o r X =-OH or -OC2H5. In these cases alone, the pre-exponential factor was 10 sec. and 10 sec. respectively, values which are similar to the p r e f e r r e d pre-exponential factor f o r cyclohexa-dienyl decomposition. I t i s probable that resonance s t a b i l i z e s the decomposition t r a n s i t i o n state f o r 'those examples with the low pre-exponential factor. E f f e c t i v e resonance implies a part-i c u l a r molecular configuration with attendant low entropy or low pre-exponential factor f o r decomposition. As resonance or s p e c i f i c orientation i s not possible f o r the hydrogen atom, the higher pre-exponential factor, 10 1 4* 8 sec.""1 f i t s into t h i s pattern. The decomposition of the cyclohexadienyl r a d i c a l has been observed i n the mechanism of in t e r a c t i o n of hot t r i t i u m atoms - 103 -o with benzene. In the gas phase experiments, C^FL/T* r a d i c a l s generated by addition of r e c o i l t r i t i u m atoms to benzene decom-pose to benzene-t and hydrogen atoms. This decomposition was substantially reduced i n the l i q u i d phase, in d i c a t i n g the reaction of excited cyclohexadienyl-t r a d i c a l s . The disproportionation of cyclohexadiene-1,4 to cyclohexene and benzene has been observed i n other work. When the diene i s exposed to sunlight i n the presence of dissolved, molecular i o d i n e , 1 0 1 hydrogen iodide i s generated which i n i t i a t e s the following chain disproportionation process, o I ~= ^ II I + l o HI + + I + H I r The cyclohexefl-4-yl r a d i c a l assumes the same role as has been postulated here f o r the gas phase, thermal reaction. S i m i l a r l y , i n r a d i o l y s i s of l i q u i d cyclohexadiene-1,4, cyclohexene i s believed to arise through the processes 10 + H-+ E.s.r. experiments did not detect the C^ HA, r a d i c a l i n d i c a t i n g i t s rapid removal by hydrogen abstraction from C^Hg-1,4. The C^Hy r a d i c a l was r e a d i l y detectable, however. An analogous process i s believed to cause disproportionation i n the action of t e r t -butoxy anions on cyclohexadienes. 102 Thus the n e u t r a l i z a t i o n - 104 reaction or \ I + t-BuO© > tert-BuOH + \ © produces the cyclohexadienyl .anion Which can reduce the diene by hydride transfer, giving benzene and the strong, conjugate base of cyclohexene. or 0 tert-Bu-OH^ tert-BuO® + The disproportionation i s therefore a generally favourable reac-tion and i s i n i t i a t e d i n widely d i f f e r e n t systems. B. Di-isopropyl Ketone as I n i t i a t o r 1. Pure di-isopropyl ketone photolysis In recent years, di-isopropyl ketone has occasionally been used as source of isopropyl r a d i c a l s for k i n e t i c studies. I t 103 104 was extensively studied by H e l l e r and Gordon ' who photol-yzed the k e t o n e 1 0 3 and the ketone deuterated i n the OC-H p o s i t i o n 1 0 4 This Work was done at temperatures mainly above those.of inter e s t f o r use of the ketone as a source of isopropyl r a d i c a l s , but i n -dicated that below 200° and above 70° the ketone would provide a convenient source of isopropyl r a d i c a l s , at least at high i n t e n s i t i e s of il l u m i n a t i o n . Previous work by Whiteway and 105 ° Masson had shown that i t was decomposed by 3130 A. radiati o n with a quantum y i e l d of unity from 50 to at least 150°. Accord-ingly, experiments were done on pure di-isopropyl ketone to assess , i t s usefulness and confirm the r e s u l t s of H e l l e r and Gordon i n the temperature range of in t e r e s t from 70 to 200°. - 105 -o Experimental Di-isopropyl ketone was photolyzed with 3130 A. radiation using pressures from 10 to 14 mm. of the ketone. Most experiments were done using the f u l l i n t e n s i t y of ill u m i n a t i o n o available at 3130 A. from the B.T.Ha Hg. arc lamp affording about 13 / 3 10 molec./cm. sec. CO gas production at the ketone pressures used. Some experiments were done with a reduced i n t e n s i t y of l i g h t f o r greater p r e c i s i o n i n measurement of the abstraction reaction. In these experiments, the f u l l i n t e n s i t y of the lamp was attenuated with a neutral density f i l t e r of o p t i c a l density 1.3, affording a 20sl reduction i n i n t e n s i t y . Measurements were attempted i n 18 experiments between 71° and 193°, although at the lower l i m i t of the temperature range, the metathetical reaction proceeds too slowly to pearmit meaningful rates to be determined at convenient photolysis rates. In i n d i v i d u a l experiments, di-isopropyl ketone was measured out into the photolysis c e l l i n the usual way. Generally, the f u l l vapour pressure at room temperature was used as higher con-centrations of ketone would have been desirable. This would have required heating of the connecting tubing that services the c e l l and preparative l i n e which was not convenient with the use of mercury cut-offs, therefore pressures approaching the 14 mm. vapour pressure of di-isopropyl ketone had to be accepted as the maximum. Af t e r photolysis, the reaction mixture was allowed to d i s t i l into the a n a l y t i c a l l i n e where the traps were set at the following temperatures* I s * Variable temperature trap - 78° S p i r a l trap -215° 2 n < 1 Variable temperature trap -140° Under these conditions, the CO was collected into the gas burette - 106 -and measured. . Analysis of t h i s f r a c t i o n using a s i l i c a gel column at room temperature and flame i o n i z a t i o n detection showed i t to be pure and free from traces of methane. Following CO analysis, the s p i r a l trap was allowed to warm to room temperature allowing the Cg f r a c t i o n to be collected into the gas burette, measured, and then analyzed i n each experiment on a 2 meter s i l i c a gel column at 110°. This f r a c t i o n was found to consist e n t i r e l y of propane and propylene. Analysis f o r the r e l a -t i v e pressures of these components could be made i f the chromato-graphic area of the propane peak were multiplied by 0.913. Coupled with the t o t a l Cg y i e l d measured manometrically, t h i s information — permitted c a l c u l a t i o n of the rates of formation of CgH^ and CgHg i n d i v i d u a l l y . A f t e r removal of the Cg f r a c t i o n , the 2 variable temper-ature trap was Warmed to -94° where pure di-isopropyl was c o l l e c t -ed into the gas burette. When appropriate, a correction f o r non-id e a l behavior was made from Van der Waals constants tabulated i n the "Handbook of Chemistry and Physics". Values of the experimental conditions, rates of formation of products and certa i n other derived quantities to be discussed below have been compiled i n Table VI. Although not routinely required f o r determination of rate data, analyses were occasionally run on the l i q u i d residues using the c a p i l l a r y polyethylene g l y c o l column. P a r t i c u l a r attention was paid to the possible presence of isobutyraldehyde suspectedd to form i n the low temperature experiments. Although i t was demonstrated that traces of t h i s material could pass through t h i s column, none was ever detected i n the photolysis products. No peak attributable to dimethyl ketene was observed either, which - 107 -TABLE VI Photolysis of Pure Dl-isopropyl Ketone (°K) (sec.) 10~ 1 2R x (molec/cml sec.) cc. mol.sec, T t M" CO G 3 H 8 C 3 H 6 C6 H14 M k3/k2 345 3600 3.36 10.61 3.86 3.81 6.59 0.985 0.58 0 346 3600 3.20 10.35 3.75 3.79 6.53 0.994 0.58 0 353 3600 3.08 11.25 4.06 4.09 7.06 0.998 0.58 0 356 b 18000 2.63 1.16 0.488 0.406 0.675 (0.973) (0.60) 1.94 369 b 18000 2.88 1.16 0.459 0.417 0.647 (0.958) (0.66) 1.81 373 3600 2.98 11.17 4.13 3.93 6.74 1.03 0.58 2.58 396 b 14400 3.28 0.448 0.185 0.124 0.285 (1.05) (0.44) 3.5 398 3600 2.91 10.48 4.13 3.50 6.31 0.996 0.56 8.6 410 3600 2.47 8.89 3.43 3.05 5.26 0.978 0.58 6.7 418 1800 2.81 18.43 7.16 6.43 12.3 1.05 0.52 7.4 425 3600 2.47 7.99 3.22 2.66 4.43 0.958 0.60 10.8 425 b 14400 2.82 0.560 0.350 0.196 0.356 (1.27) (0.56) 9.1 433 b 10800 1.86 0.646 0.318 0.168 0.258 (0.891) (0.65) 15.9 435 b 14400 2.22 0.609 0.378 0.177 0.246 (1.02) (0.72) 18.2 437 3600 2.61 9.92 4.30 3.12 5.45 0.983 0.57 19.4 452 3600 2.38 8.92 4.20 2.69 4.75 1.00 0.57 29.1 452 3600 2.71 9.37 4.37 2.88 4.90 0.989 0.59 24.8 466 3600 2.70 10.44 5.20 3.07 5.21 0.995 0.59 34.6 Mean .value, excluding values i n parentheses 0.99 0.58 Limit of error at 5% p r o b a b i l i t y l e v e l +0.05 +0.04 a 10 [r/] = Concentration of di-isopropyl ketone (molecule s/cm?) b Measured at reduced l i g h t i n t e n s i t y . (See page 108) - 108 -is reasonable as Heller and Gordon 1 0 4 report that formation of this compound dies out below the lowest temperature of this work. Treatment of the data In view of the previous work report-103 104 ed ' on di-isopropyl ketone photolysis, and of the range of temperatures studied here, the present results were examined for consistency with the simple Scheme F. WP + C3H7C0C3H7 - 1 -> 2 C 3 H 7 + CO 2 C 3 H 7 2 -> C6H14 3 C3H8 + C3H6 4 y € 3 H 7 • C 3 H 7 C0C 3 H 7 —> G3^8 *G3^6G0G3^ G3H7 + C3 H6 G O C3 H7 5 —> C6H13 C 0 C3H7 The adequacy of this mechanism to explain the measurements made, at least at the higher light intensities, is discussed below. From the steady-state equations, k 3 / k 2 = R C 3 H 6 / R C 6 H 1 4 V k 2 <\HQ ' R C 3 H ^ M R C 6 H i 4 ^ where [D] is the concentration of di-isopropyl ketone. Further-more, from the above mechanism, we may define a "material balance M where M = ^ R C 3 H 8 + R C 6 H 1 4 ^ C 0 <13* This follows since one molecule of CO is equivalent to two iso-propyl radicals, or to one molecule of either CgHg or C^H^ 4, pro-viding the above mechanism holds. Values of these three quantiti are listed in Table VI for each kinetic run. Experiments using the lower intensities of light are indicated and have not been included in the statisticalyanalysis bf k3/k 2 values, or of M values because deviation from the simple mechanism apparently is - 109 -evident at the lower i n t e n s i t i e s , as may be expected when r a d i c a l t e r m i n a t i o n i s not e f f i c i e n t at elevated temperatures. The low V i n t e n s i t y experiments give s a t i s f a c t o r y measurements of k 4 / k 2 however, and t h i s i s expected even i n the presence of the compli-103 c a t i o n s i n d i c a t e d by H e l l e r and Gordon f o r t h i s temperature range. Values of 13 + l o g ^ Q k 4 / k 2 have been p l o t t e d i n Figure 15 as a f u n c t i o n of 1G / T , along w i t h the values c a l c u l a t e d from 103 the paper of H e l l e r and Gordon. Reasonable agreement i s e v i -dent. D i s c u s s i o n For the high i n t e n s i t y experiments, s t a t i s t i c a l a n a l y s i s shows M = 0.99 + 0.05 and k 3 / k 2 = 0.58 + 0.04 at the 5 % p r o b a b i l i t y l e v e l . The very small d e v i a t i o n of M from u n i t y con-f i r m s the simple mechanism under the high i n t e n s i t y c o n d i t i o n s . The a n a l y t i c a l technique i s a l s o confimmed. The presence of i s o -b u t y r y l r a d i c a l s , C-jHySo, i n k i n e t i e a l l y important q u a n t i t i e s i s i n c o n s i s t e n t w i t h such values of M. As r e a c t i o n s of such r a d i c a l s would be expected to remove CO from the system i n a way not l e a d -i n g t o the production of the f r e e gas, we would expect M to r i s e above u n i t y , whereas the measured values are very s l i g h t l y below, over the e n t i r e temperature range. Furthermore, the absence of isobutyraldehyde i n d i c a t e s e f f i c i e n t d i s r u p t i o n of the i s o b u t y r y l r a d i c a l , (CH3)2\CHC=0 > C 3H y + CO above 71°, the lowest temperature s t u d i e d . The measured value of k 3 / k 2 = 0.58 l i e s between values pre-v i o u s l y measured using d i - i s o p r o p y l ketone as a source of the l cvx ^ 7 i s o p r o p y l r a d i c a l ( H e l l e r and Gordon, 0.6; Dominguez et a l . , 1 OA 0.65) and using azoisopropane as a source (Reim and Kutschke, 0.53). The high i n t e n s i t y experiments are expected to provide - 110 -Figure 15+ Metathesis between the Isopropyl Radical and D i -isopropyl Ketone 103/T # Not included i n the s t a t i s t i c a l analysis extent of metathesis i s too low to permit accurate analysis. A Points measured by H e l l e r and Gordon. - 1121 -the most r e l i a b l e determination of t h i s quantity. In the low i n t e n s i t y experiments, general agreement with the high i n t e n s i t y measurements of M were obtained, but greater scat-t e r was evident. Values of k^/k^ were also scattered and may d r i f t upwards at higher temperatures (433, 435°K.) due to the chain reaction: CgHy + CgHyCOCgHy - *^ C^Hg +• 'CgH^COCgHy *C 3H 6C0C 3H 7 > C 3H 6 + CO + C 3H| 103 which was reported by H e l l e r and Gordon. This complication i s apparently quenched at higher i n t e n s i t i e s by e f f i c i e n t termin-ation of the ketonyl r a d i c a l s . The abstraction of hydrogen atoms (reaction 4) has been calculated by equation(12) f o r each run and the r e s u l t s are shown i n Figure 1 5 A s can be seen, these values are i n excellent numerical agreement with those calculated from H e l l e r and Gordon's re s u l t s (entered as t r i a n g l e s on the f i g u r e ) . The high and low i n t e n s i t y measurements are i n agreement in d i c a t i n g that the only complexity at the lower i n t e n s i t i e s i s the chain decomposition of ketonyl r a d i c a l s , which would not i n t e r f e r e with k 4/k2 measure-ments. The actual values of the abstraction constants are very small i n t h i s temperature range, and quite unmeasureable below 110°, under the conditions of these experiments. As t h i s rate of abstraction i s much lower than the rate of attack of isopropyl r a d i c a l s on cyclohexadiene-1,4 and cyclohexadiene-1,3, i t would appear that di-isopropyl ketone i s a good source of isopropyl r a d i c a l s at high i n t e n s i t y of i l l u m i n a t i o n i n the temperature range from 70-200°. No significance can be attached to Arrhenius parameters - 112 -calculated from the IC4A2 data, as examination of the r e s u l t s shows that the precise slope of the l i n e i s strongly dependent on the c a l i b r a t i o n of the a n a l y t i c a l gas chromatograph. This i s so since k 4 A 2 depends c r i t i c a l l y upon the value of RQ H - R Q H 3 8 3 6 where, except at the highest temperatures, RR „ i s only s l i g h t l y u3 t t8 larger than R^ . ^  . H e l l e r and Gordon's measurements over a wide 3 6 o range of temperatures to 4GG are much superior f o r t h i s purpose than the present data, and no reason ex i s t s to doubt t h e i r r e s u l t s . However, for purposes of making a correction (always very small) to the propane y i e l d i n ketone-cyclohexadiene mixture experiments to be considered below, a value of the abstraction rate constant r a t i o i s needed. Without attaching any significance to the coef-f i c i e n t s as Arrhenius parameters, l e a s t mean squares analysis of the points of Figure 15 with temperatures at or above 125°, gives the following r e s u l t : 13 + l o g i n — 7 . 5 0 6 - 2.753-^°-1 0 k f T This has been used to estimate the extent of isopropyl r a d i c a l attack upon di-isopropyl ketone i n subsequent experiments. 2. The high i n t e n s i t y photolysis of di-isopropyl ketone- cyclohexadiene-1,4 mixtures To extend the data on /S(RF C^H^) and abstraction rate data to the isopropyl r a d i c a l , the high i n t e n s i t y photolysis of d i r isopropyl ketone-cyclohexadiene-1,4 mixtures was investigated. The method i s conceptually i d e n t i c a l to that already discussed f o r azomethane as the i n i t i a t o r system and only the technical variations w i l l be emphasized here. Experimental, The system was examined i n 14 experiments car-r i e d out between the temperatures 74 and 136°. Very nearly the - 113 -f u l l vapour pressure of the ketone at room temperature was em-ployed i n each run i n order to obtain maximum r a d i c a l concentra-tions when i r r a d i a t e d . The unattenuated l i g h t from the B.T.H. o Hg arc lamp at 313G A. was employed f o r photolysis. Smaller pressures, from 1-5 mm., of cyclohexadiene-1,4 were employed depending on the reaction temperature. These reagents were measured out, mixed and photolyzed i n the usual way. A f t e r photolysis, the reaction mixture was allowed to d i s t i l into the "analytical l i n e where the various traps were set at the following temperatures: Under these conditions, the carbon monoxide and C 3 products were coll e c t e d and measured i n exactly the same way as was done i n the experiments with pure di-isopropyl ketone. Unfortunately, i t was not possible to measure C^H^ manometric-a l l y i n the presence of cyclohexadiene-1,4. Therefore, a f t e r completion of the CO and Cg analyses, the remaining reaction mix-ture was mixed with the n-heptane i n t e r n a l standard, removed from the vacuum l i n e and subjected to the usual gas chromatographic analysis using packed and c a p i l l a r y polyethylene g l y c o l columns. The packed column analysis yielded peaks f o r d i - i s o p r o p y l , n-heptane, benzene, and cyclohexadiene-1,4, the l a t t e r being used to correct the benzene peak f o r benzene present i n the s t a r t i n g material. The molar r a t i o of di-isopropyl:n-heptane was obtained from the peak areas i f the di-isopropyl peak was m u l t i p l i e d by a c a l i b r a t i o n factor, 1.165. The molar r a t i o of benzene to n-1 Variable temperature trap S p i r a l trap 2 Variable temperature trap - 114 -heptane could be found i f the benzene peak area was multiplied by 1.095, The c a p i l l a r y column analysis was run f o r 10 minutes at room temperature, which allowed passage of the pentene-1 solvent, column was r a p i d l y heated to 120° and held there while the ketone peak followed by the f i n a l p a i r of peaks due to compounds I and II were eluted. No cyclohexadiene-1,3 was ever observed, although i t could have been s e n s i t i v e l y detected i n the c a p i l l a r y column analysis. The conditions of the experiments, concentrations of reac-tants and rates of formation of the detected products have been compiled i n Table VII along with other derived data to be d i s -cussed below. Treatment of the data' Along with the reactions of the i s o -propyl r a d i c a l with di-isopropyl ketone discussed under "The Photolysis of Pure Di-isopropyl Ketone" we must consider the pos-s i b l e reactions i n the photolysis of mixtures of the ketone with cyclohexadiene-1,4, which are l a b e l l e d 6 i n Scheme G. We s h a l l see that the l a s t reactions (6e) do not contribute to a measureable degree to the fates of the cyclohexadienyl r a d i c a l s i n the sys-tem, therefore we are able to write the rate equations analogous-to those of the methyl case. From the steady-state treatment (and using k, = 0 ) , we obtain: [Dj and [BJare the concentrations of di-isopropyl ketone and cyclohexadiene-1,4 respectively. Also k./k« r e f e r s to the ab-n-heptane, benzene, and cyclohexadiene-1,4 peaks a f t e r which the (14) - 115 C 3H 78c 3H 7 + h? -> 2 G 3H ? + CO 2 C 3 H 7 "> C 6 H 1 4 -> C 3Hg + C 3 H 6 C 3 H 7 + C 3 H 7 C ^ C 3 H 7 > C 3 H 8 • .C 3H 6CC 3H 7 0 C 3 H 7 * * C3 H6 C' C3 H7 ~ > C6 H13 <" C3 H7 C 3 H 7 • -2s C 3Hg + i • 6c or II 6d -> C 3 H 8 + TABLE VII " M e t a t h e s i s between the Isopropyl Radical and Cyclohexadiene-1,4 (°K) (sec.) l O " 1 ^ (molecules/cm?sec.) Vmol^sec? T t M a [ B ] 3 CO C 3 H 8 C 3 H 6 C 6 H 6 C6 H14 M* k 3 /k 2 l O 1 3 ^ k 6 A 2 b A D A(2) C k I / k I I 6c / l c6c 348 3600 2.57 1.99 11.80 8.63 2.76 1.40 4.88 1.03 0.57 102 0.46 0.47 0.86 348 3600 2.54 1.65 10.12 6.14 2.34 0.81 2.66 (0.79) (0.88) 112 0.35 (0.36) 0.88 353 36*00 2.80 1.53 14.91 8.81 3.77 1.46 6.03 (0.90) 0.63 95 0.69 (0.90) (0.60) 355 3600 2.88 2.04 17.13 12.80 3.84 2.21 6.25 0.98 0.61 132 0.49 0.58 0.88 359 3600 2.42 1.49 8.96 6.68 1.91 1.29 3.91 1.04 (0.49) 118 0.59 0.47 0.81 361 1800 • 2.69 1.63 5.91 5.51 1.22 0.78 1.50 1.05 (0.81) 175 0.29 (0.36) (1.00) 368 3600 2.28 1.23 7.50 5.98 1.62 1.26 2.55 0.97 0.64 158 0.69 0.56 0.80 370 1800 2.94 1.08 10.57 8.21 2.32 2.00 4.00 0.97 0.58 177 (1.09) (0.74) 0.82 376 3600 2.61 1.21 9.69 7.85 2.08 1.70 3.71 1.02 0.56 171 0.75 0.54 0.79 381 3600 2.57 1.65 10.70 9.87 1.80 1.92 3.11 1.01 0.62 189 0.46 0.47 0.90 394 3600 2.71 1.94 9.27 9.96 1.03 2.04 1.69 1.04 0.61 268 0.43 0.52 0.84 ON I TABLE VII (continued) ( cc^ (°K) T (sec.) t M a 10 R^ (molecules/cmfsec.) [ B ] a CO C 3 H 8 C 3 H 6 C 6 H 6 C 6 H 1 4 M* \ k 3/k 2 -molrsecr k 6/k| J b A D A 2 ) ° k6c / / k6c 396 1800 1.97 0.95 7.14 7.14 1.36 1.17 2.00 (1.12) (0.68) 337 0.35 0.51 (1.00) 403 1800 2.60 0.69 14.64 11.59 3.03 2.09 5.15 1.00 0.59 398 0.51 0.50 0.81 409 1800 2.57 0.57 17.73 13.14 4.10 2.31 6.64 0.99 0.62 437 0.58 0.58 0.93 Mean value, excluding entries i n parentheses 1.01 0.60 0.51 0.52 0.85 Limits of -error at S% p r o b a b i l i t y l e v e l +0.05 +0.05 +0.30 +0.09 +0.09 a l r t17 r, 1 0 1 7 IO 1' = Concentration of di- i s o p r o p y l ketone (molecules/cnu = Concentration of cyclohexadiene-1,4 (molecules/cm. B ( 1 ) = RC,H/(RCM0 " R C „ H „ " 2 " t R<1H,„ D 5 '6"6 °3"8 °3 n6 ^ 6 n 6 °6 n14 K2 c (2) calculated as described i n the text, page 121. - 118 -s t r a c t i o n of hydrogen atoms from di-isopropyl ketone by isopropyl r a d i c a l s . The extent of t h i s reaction was measured and reported under "Pure di-isopropyl ketone photolysis". The correction f o r the abstraction from the ketone was always extremely small, and completely n e g l i g i b l e below 90° due to the "inertness of the ke-tone towards isopropyl r a d i c a l attack. Values of k^/k^ calculated from equation (6) have been calculated f o r each k i n e t i c run and entered into Table VII. These have been plotted as a function of 10 /T i n Figure 16. S t a t i s t i c a l l e a s t mean squares analysis shows that the data f i t the following Arrhenius expression: )i V RT / \molec. sec./ "2 where the errors are expressed at the 5 % p r o b a b i l i t y l e v e l . The value of M , where M * _ ^3 H8 C6 H14 C 6 H 6  RC0 should be unity i f no addition of isopropyl r a d i c a l s takes place to the double bonds of cyclohexadiene-1,4, and providing the a n a l y t i c a l method i s s u f f i c i e n t . M* has been calculated and presented i n Table VII f o r each run. I t i s indeed very close to unity f o r a l l the experiments and we conclude that addition i s unimportant f o r isopropyl r a d i c a l attack on cyclohexadiene-1,4. T h i s point w i l l be further considered below. Values of k 3 / k 2 , where k 3 / k 2 = R c H ^ RC H N A V E B E E N C A L" 3 8 6 14 culated f o r each run and entered into Table VII. The close agree-ment of the mean value, k 3 A 2 = 0.60 + 0.05, to that measured i n experiments with the pure ketone, k 3 / k 2 = 0.58 + 0104, indicates, though not as s e n s i t i v e l y as i n another t e s t , the absence of 120 '-r e a c t i o n s ( 6 e ) . Reactions (6e) can a l s o be e l i m i n a t e d by the f o l l o w i n g t e s t . F i r s t , no cyclohexadiene-1,3 i s found amongst the p r o d u c t s of t h i s s e r i e s of experiments, showing t h a t , i f 6e o c c u r s , i t must gener-ate cyclohexadiene-1,4 which cannot be seen. We s h a l l d i s c u s s i n the next s e c t i o n the p h o t o l y s i s of d i - i s o p r o p y l ketone-cyclohexa-diene-1,3 mixtures where the c y c l o h e x a d i e n y l r a d i c a l i s generated by the s i m i l a r r e a c t i o n In t h i s system, no cyclohexadiene-1,4 was generated, although i t c o u l d have been e a s i l y d e t e c t e d . As a p r a c t i c a l matter, the form-a t i o n of the cyclohexadienes may be s a i d t o be l e s s than one p e r -cent of the f o r m a t i o n of benzene. T h e r e f o r e we see t h a t i s o p r o p y l may recombine w i t h c y c l o h e x a d i e n y l r a d i c a l s , or d i s p r o p o r t i o n a t e w i t h them to form benzene and propane, but not t o form c y c l o h e x a -diene and propylene. A p p a r e n t l y the f o r m a t i o n of benzene i s able to a c t i v a t e the d i s p r o p o r t i o n a t i o n step which otherwise does not occur s i g n i f i c a n t l y . Brown 1 a l s o found the f o r m a t i o n of ben-zene but not of e i t h e r of the cyclohexadienes i n a study of the e t h y l r a d i c a l w i t h both the 1,3- and 1,4-cyclohexadienes. T TT The v a l u e s of k £ C A g c » l i s t e d i n T a b l e VII were c a l c u l a t e d d i r e c t l y from the peak areas of compounds I and I I i n the c a p i l -l a r y column a n a l y s i s , assuming equal s e n s i t i v i t y of the gas chromatograph's d e t e c t o r towards each of the i s o m e r i c substances. Thus k ^ c / k ^ = Area ( l ) / A r e a ( l l ) = 0.85 + 0.09, independent of temperature. The columns headed Z ^ l ) and Z ^2) i n T a b l e VII r e f e r t o the r e s u l t s of the two d i f f e r e n t methods of c a l c u l a t i o n of k 6 c / ( k 6 c + k6c^ = A ( C 3 H 7 , C 6 H ? ) . These two calculations have been set up i n the same manner as was done i n the treatment of the methyl experiments. Thus ZMl) r e f e r s to values of the expression: A ( l ) o (15) R C „ F U " R C „ H , " 2 RC,H, ' " T l ' W ^ H 34 8 3^6 6' 6 k£ ~6ll14 = 0.51 + 0.30 (median value = 0.50) from R 6 = R £ C • R £ + R 6 D and R 6 - R C 3 H 8 " R C 3 H 6 - \ * 6 " ( k 4 / k 2 ^ D J R C 6 H 1 4 The values of A 2 ) l i s t e d i n Table VII were found from the c a p i l l a r y column peak areas of benzene, I, and II using the Z ^ l ) values to c a l i b r a t e the detector just as was done fo r methyl plus cyclohexadienyl experiments. In t h i s case, a considerable increase i n p r e c i s i o n was r e a l i z e d since scatter i s r e a d i l y induced into values calculated from equation (15) because of the differences taken i n the denominator of quantities of comparable magnitude. No systematic error i s anticipated, however. This i s substant-iated by the agreement of the median value with the mean. To obtain the values of A 2 ) , the quotient n - Area (I) + Area ( I i )  y AreaTC^H^) was found f o r each experiment, having the value Q = 2.3 + 0.4 independent of temperature. Each Q value was mul t i p l i e d by the corresponding value of Z\(l) .calculated from equation (15) giving P = 1.18 + 0.20 independent of temperature with a median value of 1.17. The mean value was used as a c a l i b r a t i o n factor and i n d i v i d u a l values of ZM2) were calculated from i n d i v i d u a l Q - 122 -values by the relations =—^—= A(C3H7, C,H7) =^M. k I + u l l ^ / 6 / Q K6c + K6c which yie l d s Z^2) = 0.52 + 0.09. Individual values of the log-arithm of t h i s quantity, and also of k ^ c / k ^ have been plotted i n Figure 17 as a function of 10 /T. No i n d i c a t i o n of temperature dependence i s evident i n view of the experimental errors. Discussion The significance of the metathetical rate con-y stants, k^/k^ a n d °f the disproportionation to combination r a t i o s f o r isopropyl and cyclohexadienyl r a d i c a l s w i l l be discussed i n the "General Discussion" section of t h i s thesis along with similar r e s u l t s determined with other i n i t i a t o r systems. The lack of addition of isopropyl r a d i c a l s to cyclohexadiene-1,4 i s evidenced by the fact that M determined f o r each experiment i s never s i g n i f i c a n t l y below unity. I t i s of i n t e r e s t to determine the maximum possible value i / that the addition rate constant r a t i o k 7/k2, r e f e r r i n g to the reactions . ^ iso-C 0I-U k 7 \„ ^ \ / o / iso-C 3H 7 + 2 iso-C 3H 7 k 2 — > C6H14 could have, and s t i l l remain undetectable. An estimate of t h i s maximum rate constant r a t i o can be obtained f o r each experiment i f we suppose that addition i s undetectable u n t i l i t proceeds at a rate f a s t enough to cause the value of M to f a l l below 0.98. Such a value of M would correspond to a rate constant r a t i o of k?(max) 0.02 R c o 2 r - ~6"14 - 123 -Figure 17. Combination and Disproportionation of the Isopropyl Radical with the Cyclohexadienyl Radical. 1.Ch CD O 0 ® r—®'—©-2.5 T T ®_ T T • 2.7 1 0 3 A 2.9 r s ® k / k s " k6d/<*6c + k " > - 124 -Accordingly, values of t h i s quantity, and of ky(max)/kft have been presented i n Table VIII for each k i n e t i c run. C l e a r l y the rate of addition i s le s s than 5 % of the rate of metathesis; i t may well be very much l e s s , indeed. Brown 1 found that addition of ethyl r a d i c a l s to cyclohexa-diene-1,4 was unmeasurable i n h i s systems and had accessible upper l i m i t s comparable i n magnitude to those found here f o r isopropyl r a d i c a l s . 3. Di-isopropyl ketone-Cyclohexadiene-1,3 mixtures - high  i n t e n s i t y photolysis The examination of the reactions of the isopropyl r a d i c a l with cyclohexadiene-1,3 may c l a r i f y questions concerning the i n t e r -actions of cyclohexadienyl r a d i c a l s with Isopropyl r a d i c a l s . For example the product cyclohexadiene-1,4 cannot be detected i n the system di-isopropyl ketone-cyclohexadiene-l,4yy In the cyclohexadiene-1,3 system, however, i t may be possible to discount the process iso-C^uy « 7 + ( v i ^ 2 - ^ C 3 H 6 + more c e r t a i n l y than merely by observing that R^ H /RQ ^ does 3 6 6 14 not increase. This has indeed been the case. Furthermore, the rate constant of metathesis from the 1,3-diene by isopropyl i s of i n t r i n s i c i n t e r e s t i n comparison to that already measured f o r metathesis from cyclohexadiene-1,4. Recent 107 experiments performed on equilibrium mixtures of the 1,3- and 1,4-cyclohexadienes show that the standard free energy of the conjugated isomer l i e s only 0.58 kcal./mole below that of the unconjugated one. Furthermore, early hydrogenation experiments by Kistiakowsky and c o w o r k e r s 1 0 8 * 1 0 9 show that the heat of hydro-genation of the double bonds of cyclohexadiene-1,3 i s only 1.8 k c a l . - 125 -TABLE VIII Upper Accessible Limits for the Addition of the Isopropyl Radical to Cyclohexadiene-1,4 °K) (cm?/molec. sec.)^ T 1 0 1 3 0 , 0 2 ^ 0 3 M R C H C6 H14 k7(max k6 348 5.36 0.053 348 7.51 0.067 353 7.94 0.084 355 6.72 0.051 359 10.6 0.098 361 5.94 0.034 368 7.64 0.048 370 9.76 0.055 376 8.30 0.049 381 7.35 0.039 394 7.36 0.027 396 10.6 0.031 403 18.7 0.047 409 24.2 0.055 [B] = Concentration of cyclohexadiene-1,4 (molec./cm.) - 126 -less than twice the heat of hydrogenation of cyclohexene. Both molecules, therefore, stand to gain very nearly the same amount of resonance energy i n forming the cyclohexadienyl structure. The metathetical rate constant f o r ethyl attack 1 on .cyclohexadiene-1,3 has been found to be d i s t i n c t l y lower than that f o r attack on cyclohexadiene-1,4, so further study i s indicated. The conjugated double bond structure of the 1,3-diene activates t h i s molecule towards addition of free r a d i c a l s . A comparison of the r e a c t i v i t y of the isopropyl r a d i c a l ' s r e a c t i v -i t y towards addition with that of the ethyl r a d i c a l promised to be i n s t r u c t i v e . Insofar as the 1,4-diene does not display any r e a c t i v i t y towards the addition of isopropyl r a d i c a l s , and i t s k i n e t i c s had been previously studied with the isopropyl r a d i c a l , the analysis of t h i s more complex system was rendered more t r a c t -able. Addition of r a d i c a l s to cyclohexadiene-1,3 has been studied previously f o r the m e t h y l , 1 1 0 e t h y l 1 1 1 and p o l y a c r y l o n i t r i l e 1 1 2 r a d i c a l s . Homopolymerization 1 1 3" 1 1^ and copolymerization 1 1 2 , 1 1° of cyclohexadiene-1,3 have also been described. In each of these systems, the addition of the r a d i c a l to the 1,3-diene was shown to occur much more r a p i d l y than the corresponding addition to cyclohexadiene-1,4. Cyclohexadiene-1,3 i s also the more reactive isomer towards the formation of a 1:1 adduct with ethyl azodi-117 carboxylate, although t h i s process i s unaffected by hydro-quinone, a free r a d i c a l scavenger, and hence i s not l i k e l y a free r a d i c a l process. In the discussion of the addition reaction, we s h a l l assume that the addition occurs at a terminal unsaturated carbon atom; i n t h i s way the a l l y l i c resonance energy i s released and w i l l - 127 -cause the terminal addition process to be the more exothermic one. Terminal addition has been observed both f o r 2,3-dimethylbuta-d i e n e - 1 , 3 1 1 ^ * 1 1 9 and fo r h e x a d i e n e - 2 , 4 , 1 1 9 ' 1 2 0 although s t e r i c conditions vary greatly at the terminal atoms of these two sys-tems. This pattern has accordingly been adopted i n reaction?(7) below. Experimental The photolysis of di-isopropyl ketone-cyclo-hexadiene-1,3 mixtures was examined i n eleven experiments, from 42 to 133°. A s u f f i c i e n t l y high rate of photolysis of ketone was maintained to ensure that a l l secondary r a d i c a l s were terminated by isopropyl r a d i c a l s . The reactants were measured into the c e l l i n the usual way, o mixed, and thenophotolyzed with 313G A. radiatio n from the med-ium pressure Hg lamp. The reactant pressures used have been tab-ulated i n Table IX along with temperatures, the rates of formation of the products determined as described below and certain derived quantities. The reaction time was 300 seconds i n each instance. A f t e r photolysis, the reaction mixture was allowed to d i s t i l i nto the a n a l y t i c a l l i n e where the various traps were set at the following temperatures: l s ^ Variable temperature trap - 78° S p i r a l trap -215° 2 Variable temperature trap -140 The same d i s t r i b u t i o n of temperatures was used f o r the di-isopropyl ketone-cyclohexadiene-1,4 experiments. Under these conditions, the GO was removed pure, into the gas burette and measured. Warm-ing of the s p i r a l trap permitted c o l l e c t i o n of the propane and propylene into the gas burette, and these gases were analyzed just as previously described. No attempt was made to analyze TABLE IX Addition and Metathesis between the Isopropyl Radical and Cyclohexadiene-1,3 °K) 10~ 1 2R X (mol./ cm?sec.) —(cm./molecule sec. J -M a io 1 3 1 0 1 3 1 0 1 3 „ 1 0 1 3 1 0 1 3 T CO C 3 H 8 C 3 H 6 Mf C "Yo U k 6/k| V o V c +0.342U ky/k2 315 3.14 1.53 33.9 11.2 10.00 0.841 17 13 ( 7 ) d 85 89 95 323 2.97 0.867 30.5 10.9 9.22 0.879 49 37 (29) d 107 120 127 330 3.08 1.47 33.0 11.4 8.62 0.795 51 38 30 119 132 141 340 2.71 1.01 34.6 11.9 9.37 0.811 62 46 35 162 178 190 349 2.46 1.07 28.8 9.9 7.13 0.771 74 55 43 176 195 207 363 3.59 1.08 38.1 12.7 8.68 0.725 128 95 77 251 283 302 374 2.84 0.628 39.4 13.9 9.25 0.757 180 134 107 382 428 456 384 2.84 0.427 34.6 12.6 8.60 0.794 226 168 136 434 491 522 395 2.54 0.277 38.7 14.3 10.70 0.846 271 202 166 502 571 608 TABLE IX (continued) ( K) 10" (mol./cmTsec) ( (cmf/molecule s e c . ) 2 ^ b T. M a M a CO G 3 H 8 C 3 H 6 Mf C 1 0 1 3  U o 1 0 1 3 U 1 0 1 3 u k 6 / k | 1 0 1 3  V o 1 0 1 3 V Q+0.342U ^\ ky/kg 400 2.49 0.260 35.4 12.9 9.45 0.825 290 216 173 590 664 708 406 2.48 0.292 35.6 13.2 8.93 0.804 330 246 202 609 693 738 = C o n c e n t r a t i o n of d i - i s o p r o p y l ketone (molecule s/cmf) = C o n c e n t r a t i o n of cyclohexadiene-1,3 (molecules/cm.) ro R e a c t i o n time = 300 sec. o^ • c d M ' - V H « " R C ^ + RC QH/°' 5 8 R C 0 ' w h e r e R 7 - R 7 d " t 1 - M'^ RC0 3 8 6 6 3 6 The values i n parentheses were excluded on s t a t i s t i c a l grounds. - 130 -the di-isopropyl y i e l d manometrically. The remaining material was mixed with a small amount of n-heptane as an i n t e r n a l chromato-graphic standard and removed from the vacuum system into a cold f i n g e r . Pentene-1 solvent was added as the sample was opened, the whole was mixed and about 5JULl. was immediately injected into the Perkin Elmer 154 Vapour fractometer using a 2 meter polyethylene g l y c o l packed column which resolved peaks f o r pentene-1 (solvent), d i - i s o p r o p y l , n-heptane(internal standard), cyclohexadiene-1,3 v starting material, benzene, and cyclohexadiene-1,4. A second 5^1. sample was analyzed on the Perkin Elmer model 226 gas chromato-graph using the c a p i l l a r y polyethylene gly c o l column to analyze for the heavy products of the reaction. At room temperature pentene-1 (solvent), n-heptane ( i n t e r n a l standard), cyclohexa-diene-1,3, and benzene eluted. A f t e r heating the column to 120°, di-isopropyl ketone eluted followed by three peaks close together, the f i r s t quite small, the second and t h i r d corresponding to the products I and II observed i n the di-isopropyl ketone-cyclohexa-diene-1,4 experiments. In these experiments, however, the peak I was consistently larger than the peak I I . A f t e r 20 minutes of heating at 120°, a group of three peaks of about equal area were eluted. These were not present i n the cyclohexadiene-1,4 experi-ments and probably correspond to isomeric addition products of formula ci2 H22» l e * P r°ducts VI and VII (Scheme H). Treatment of the data The products of t h i s photolytic system have been considered as a r i s i n g from the reactions ( l ) to (7e). (Scheme H) The previous study of the behavior of cyclohexadienyl r a d i -cals had shown that no cyclohexadiene-1,3 was formed, although small amounts of i t could have been detected. This furnished - 131 -conclusive evidence that no mutual termination of cyclohexadienyl r a d i c a l s occured under the experimental conditions, and also that reactions (6e) leading to cyclohexadiene-1,3 did not proceed. In the present system, no cyclohexadiene-1,4 was formed thereby con-firming that the cyclohexadienyl r a d i c a l did not react s i g n i f i c a n t -l y with any r a d i c a l other than isopropyl, and that reactions (6e) were n e g l i g i b l e . The combination and disproportionation reactions (6c) and (6d) therefore remain as the only s i g n i f i c a n t reactions of the cyclohexadienyl r a d i c a l . The products I and II from combination, and the benzene from disproportionation were separated and meas-ured by gas chromatography i n the study of isopropyl and cyclo-hexadiene-1,4, giving the r a t i o s k^ d/k^ c = 0.52 + 0.09 and k^ c/k^ G = 0.85 +. 0.09. In the present system the products I and II could be i d e n t i f i e d and estimateddby gas chromatography, but the amount of benzene formed i n reaction (6d) was l e s s than the residual benzene i n the reactant cyclohexadiene-1,3 which could not conveniently be further p u r i f i e d . For greater predision, therefore, the r e s u l t s of the previous study were used to estimate the benzene y i e l d from the i d e n t i t y R G 6 H 6 " R6 k6d/< k6c + k6d> " ° ' 3 4 2 R 6 ( 1 6 ) where R^ i s the rate of metathesis from cyclohexadiene-1,3, and can be calculated from accessible data using equation (19). Addition b£ the isopropyl r a d i c a l to cyclohexadiene-1,3 yi e l d s the r a d i c a l C^H^5; we have made the reasonable assumption that the r a d i c a l w i l l resemble the cyclohexadienyl r a d i c a l by reacting exclusively with isopropyl r a d i c a l s . The relevant reactions are1 then (7c-e). 132 -C 3H 7C0C 3H 7 + hf — 1 —> 2 C 3H ? + CO 2 G 3H ? 2 — > C6 H14 3 -> C 3 H 6 + C 3 H 8 C 3H ? • C 3H 7COC 3H 7 4 -> C 3H R + C 3H 6COC 3H C 3 H 7 + G 3 H 6 C 0 G 3 H 7 5 -> C 6 H 1 3 C 0 C 3 H 7 C 3H 7 + C3H- + C 3H 7 + C 3H ? + C 3 H 7 6c C 3 H 7 or 6d •> C 3 H 8 + 6e "> C 3 H 6 + or 7c C 3 H 7 C 3H 7 C 3 H 7 or H ?C 3 7d C 3H ? II H C 3H ? VII 7e -> C 3 H 6 + or or C 3 H 7 III C 3H ? V ^ IV The disproportionation reaction (7d) yi e l d s the product I, which i s also formed by the combination reaction (6c). Appropriate-l y , the corresponding r a t i o of peak areas A ( l ) / A ( l l ) exceeds the value of 0.85 + 0.09 found i n the isopropyl-cyclohexadiene-1,4 system. Comparison of the observed value of A ( l ) / A ( l l ) with 0.85 allows the estimation of the rate of formation of product I by - 133 -reaction (7d). This rate has been equated to the rate of propane formation by reaction (7d); t h i s i s completely accurate only i f product III gives a chromatographic peak which augments the area of the peak due to product I, or i f product III i s formed i n a neg l i g i b l e amount. A very small peak appears j u s t before the peaks of compounds I and I I ; i t cannot be d e f i n i t e l y assigned and may be due to one or more of the products III-V. The formation of product III may be discriminated against since i t involves the transfer of a hydrogen atom shielded by the isopropyl group. Despite uncertainties, the rate of (7d) has been estimated from the areas of the peaks due to I and II by means of the expression: R 7 d = k6c A(I) - ( k * c / k * * ) A ( l l ) R 6 k6c + k6d L 1 + (klc'kllh{ll) (17) = A(I) - 0.85 A(II) = H 2.81 A(II) When the propane formed due to the reaction (7d) i s estimated by t h i s procedure, corrections can be applied to values of k 7/k^ and k^/k^ determined i n the usual way. The corrections due to t h i s source of error are not s i g n i f i c a n t at the 5 % p r o b a b i l i t y l i m i t s of e r r o r s hence i n s u f f i c i e n c i e s i n the a n a l y t i c a l technique described above are not overly important. Reaction (6e) has been shown to be negli g i b l e f o r the cyclo-ne) hexadienyl r a d i c a l , and very probably^is n e g l i g i b l e f o r the CgHj^ r a d i c a l . The only chromatographic peak which could be reasonably assigned to products IV or V i s the p a r t i c u l a r l y small one men-tioned above. This leaves reaction (3) as the only s i g n i f i c a n t source of propylene. The rate of formation of di-isopropyl c; ;, be; can be estimated by the equation: Rn „ = (k 0/k 0)EU „ = C6 H14 2 3 ^3 H8 - 134 -0.58 R- u . This method gave r e s u l t s which were consistent with, C 3 H 8 but more precise than, those obtained from the d i r e c t estimation of the di-isopropyl i n the chromatographic analysis of the Gft f r a c t i o n , which i s considerably l e s s precise than the manometric analysis of propylene. Since the estimation of the addition rate constant requires evaluation of a material balance involving d i -isopropyl rate, the alternative method of analysis indicated above i s preferable. The recombination reaction(7c) should y i e l d four products, as VI and VII each represent a p a i r of geometrical isomers. Three peaks of about equal area were found on the chromatogram at appropriate retention times? perhaps one of the stereoisomers was not formed 9 or the column f a i l e d to resolve one of the p a i r s . The repective f r a c t i o n s , F and G of the C^H* and G^H£ 5 r a d i c a l s suffering disproportionation ares F .» ^ = 0.342 § = ^ -k 6 c + k6d k7d + k7c The k i n e t i c analysis requires a procedure f o r estimation of R^, R 7 and G from accessible data. Propylene Is formed only i n reaction (3), propane only i n reactions (3), ( 4 ) s (6), (6d) and (7d). Moreover R^ ~ FR^, R 7 d .« GR 7 = HR^j and Rfi = k^ EKR^ . H A 2)^ where [B] i s the 6 14 concentration of cyclohexadiene-1,3. We estimate R r „ = °6 H14 R C H /® #^®* Probably H i s temperature dependent whereas F and G 3 6 are not. \/ \/ The evaluation of k^A 2 a n d k 7 A 2 i s based upon the equations? R C 3 H 8 = R 3 + R 4 * R 6 + R 6 d + R 7 d (18) 135 -"C 3H 8 - RC 3"6 ' M R J 6H 1 4 ( k4 / k t ) (1 + F + H) R 6 = R 7 = ( R C Q + R 6 d + R ? d) - ( Rc 3H 8 + ^ H ^ 5 = ( R ^ + (F + H)R 6) - ( RC 3H 8 + RG 6H 1 4) (19) (20) where equation (20) assumes a material balance i n r a d i c a l s . Be-cause the rate of metathesis, R^, i s required to f i n d the addition rate, R 7, some complexity i s introduced into the usual steady-state treatment, leading to simultaneous equations that must be solved fo r the i n d i v i d u a l rate constants. R^ and R 7 can be evaluated i f H i s known. The value of H i s determined by reaction conditions, and i s known f o r a p a r t i c u l a r experiment only i f A ( l ) and A ( l l ) have been measured. Then we can make use of the expressions; k6 _ C 3 H 8 R C 3 H 6 k| (1 + F + H) [[B](Rc h /0.58)^ [B]k| 3 6 (21) R, 7 i G 3 H 6 0.58 R r „ + (F + H)R r „ C 3 H 8 °3 H6 (1 + F + H) [ B ] ( R C /0.58) o o < f + h > " ( 2 2 ) (1 + F + H)(Y|k| A l t e r n a t i v e l y , these rate constants may be calculated using the parameter G, which i s independent of reaction conditions, as long as the isopropyl r a d i c a l concentration i s kept high enough. This parameter can be evaluated from those runs where unambiguous measurement of H can be achieved. Rearrangement of the expression fo r R 7 gives the equation: /0.58 — (23) 1 - G RC0 + F R 6 R C 3 H 8 - R C 3 H 6 HR, 136 Applying equation (23) y i e l d s the value G = 0.08 + 0.03 indepen-dent of temperature, as expected. Let U and V Q be defined by the equations U = 1 + F C 3 H 8 C 3 H 6 [ l ] ( R c H / 0 . 5 8 ) ^ 3 6 [ D > . [B]k« R. V o = CO " R C 3 H Q " R C 3 H 6 / 0 * 5 8 [ B ] ( R C H / 0 . 5 8 ) 3 6 I t i s e a s i l y shown that: k. (1 + F ) ( l - G)U - GV 0 —9- = 2 = 0.979 U - 0.064 V, 1 + F - G (24) k 7 (1 + F)(V Q - FU) = 1.064 V Q + 0.346 U 1 + F - G (25) Discussion The k i n e t i c r e s u l t s obtained f o r addition and metathesis between the isopropyl r a d i c a l and cyclohexadiene-1,3 are given i n Figure 18 and Tables IX:and X. The importance of reactions (6d) and (7d) i n the reaction scheme can be judged by comparison of the l a s t six columns of. the tables. The e f f e c t of neglecting propane from reaction(7d) can be found by placing G == 0 i n equations (24) and (25) when k 6/k| — > U and k 7 / k | — > V Q + 0.342 U; the corresponding values are given i n Table IX. Stat-i s t i c a l analysis reveals that neglect of reaction(7d) leads to no s i g n i f i c a n t change i n the values of the rate constants at 100° or i n the Arrhenius parameters; the p a i r s of values are compared i n Table X. The e f f e c t of neglecting both reactions (6d) and (7d) can be found by placing F = G = 0 i n equations (24) and (25) when k 6/k| >U Q and k ?/k| >VQ, where U Q = ( l + F)U = 1.342 U. Table IX shows that t h i s leads to much larger errors i n the rate TABLE X Rate Constants f o r Reaction of the Isopropyl Radical with Cyclohexadiene-1,4 and Cyclohexadiene-1,3. (kcal./mole) substrate cyclohexadiene-1,3 cyclohexadiene-1,3 no. of reaction expt's. Addition Addition 11 11 r 2 C B 13 + logAr/A: 6.024 1.276 6.0 + 0.2 6.016 1.283 6.0 + 0.2 E r - *2 5.8 + 0.4 5.9 + 0.3 13 + logk^/klf b 2.61 + 0.06 2.58 + 0.06 cyclohexadiene-1,3 Metathesis 9 6.156 1.557 6.2 + 0.4 7.1 + 0.7 cyclohexadiene-1,3 Metathesis 9 6.136 1.514 6.1 + 0.4 6.9 + 0.6 1.98 + 0.09 2.08 + 0.08 cyclohexadiene-1,4 Metathesis 14 6.047 1.408 6.0 + 0.6 6.4 + 1.0 2.27 + 0.13 The underlined values were calculated without correcting for the propane formed i n reaction 7d; B and G are c o e f f i c i e n t s of the straight l i n e 13 + log, 0k / k i = C - lO^B/T, f i t t e d to the experimental r e s u l t s ; k r represents k^ and k7, respective?y, for metathesis and addi-t i o n ; a l l l i m i t s of error are expressed at the 5 % pr o b a b i l i t y l e v e l . b At 100°. - 139 -constants, and thericorresponding Arrhenius parameters deviate s i g -n i f i c a n t l y from the more accurate values of Table X. I t has been concluded that equations (24) and (25) are adequate to y i e l d ac-curate values of rate constants and Arrhenius parameters from the experimental data of t h i s i n v e s t i g a t i o n . Cyclohexadiene-1,3 i s much more reactive towards the isopro-pyl r a d i c a l i n addition than i n metathesis; the r a t i o of the rate constants i s 4.3 + 1.5 at 100°. This difference i n a c t i v i t y i s due to a difference i n the a c t i v a t i o n energies of 1.3 + 1.1 k c a l . / mole. High i n t r i n s i c r e a c t i v i t y f o r cyclohexadiene-1,3 i n r a d i c a l polymerization i s i m p l i c i t i n the value of.1.76 f o r the A l f r e y -112 Price parameter Q. I t i s not surprising therefore, that high polymer i s formed both by r a d i c a l homopolymerization of cyclo-hexadiene-1,3 i n the s o l i d state as a canal complex with thiourea 121 at room temperature and by the r a d i c a l copolymerization of cyclohexadiene-1,3 with a c r y l o n i t r i l e i n the l i q u i d state at 60 . In contrast, attempts to homopolymerize cyclohexadiene-1,3 under homogeneous, f r e e - r a d i c a l conditions have not succeeded i n form-li f t ing polymer with a higher degree of polymerization than 12. C l e a r l y , some process of structural chain termination i s able to compete with the process of propagation f a r more e f f i c i e n t l y i n the t h i r d system than i n the other two. During publication of 122 t h i s work, a referee had suggested that t h i s competing process i s mutual combination, and i t s unusual e f f i c i e n c y i n the t h i r d system i s due to an exceptional slowness of homopropagation assoc-iated with the delocalized cyclohexenyl structure of the propagat-ing r a d i c a l species. This explanation must be reconciled with formation of high copolymer with a very nearly l s l molar compo-- 140 -s i t i o n from mixtures of cyclohexadiene-1,3 and a c r y l o n i t r i l e cover-ing a wide range of composition. In such systems, the correspond-ing copropagation process between a r a d i c a l chain terminated by a cyclohexenyl unit and a molecule of a c r y l o n i t r i l e cannot be ex-ceptionally slow, or i n h i b i t i o n would r e s u l t . No doubt a favour-able polar ef f e c t promotes addition i n t h i s instance, but c l e a r l y d e r e a l i z a t i o n of the free valence alone i s not enough to confer upon a r a d i c a l the property of sluggish addition to an o l e f i n i c linkage. Degradative, chain transfer to monomer may be considered as an alternative or complementary process f o r e f f i c i e n t structural termination i n the homogeneous homopolymerization of cyclohexa-diene-1,3. The cyclohexadienyl r a d i c a l formed i s very much stab-i l i z e d and unl i k e l y to r e i n i t i a t e by addition to cyclohexadiene-1,3. Degradative chain transfer competes with propagation for the same reactant species, and these processes are therefore analogues of reactions (6) and (7) i n which a substituted cyclohexenyl r a d i c a l has replaced the isopropyl r a d i c a l . As the difference - E 7 i s small, we may reasonably predict that the corresponding difference E^ - Ep i s also small, and that the r a t i o ky/k^ w i l l provide a measure of the order of magnitude of kp/k^, the r a t i o of the rate constants of propagation and degradative chain transfer. The low degree of polymerization i s then consistent with the low value of kp/k^ given above. Mutual termination of polymer r a d i c a l s i s eliminated from the mechanism of the emulsion homopolymerization of cyclohexa-diene-1,3, and therefore the competition between the propagation and degradative chain transfer processes f o r polymer r a d i c a l s can be assessed d i r e c t l y i n an emulsion system. - 141 -In emulsion polymerization, the chain growth occurs i n a micelle into which the diene feed d i f f u s e s , and which, at any time, contains only one active r a d i c a l centre u n t i l i t becomes termin-ated by acquiring another r a d i c a l centre from the aqueous phase. Thus, termination i s controlled, not by bimolecular k i n e t i c s as i n bulk phase polymerization, but by the rate of supply of i n i t -i a t o r r a d i c a l s from the aqueous phase, which may be very. low. In t h i s way, very long chain polymers are synthesized. An emulsion polymerization experiment was carried out on a mixture of cyclohexadiene-^1,3 (4.3 g.) i n water (7.7 g.) contain-ing sodium stearate (0.21 g.) and potassium persulfate (0.0064 g.) f o r 200 minutes at 50°, under constant, vigorous s t i r r i n g . The reactants had been previously thoroughly degassed and the reaction was run i n an atmosphere of nitrogen. A f t e r the reaction time had elapsed, the emulsion was broken by adding a concentrated solution of calcium chloride i n water. The o i l y layer was ex-tracted into chloroform and the calcium stearate f i l t e r e d o f f . The v o l a t i l e organic materials were removed under vacuum and the residue was dissolved i n benzene which was again pumped off at ice temperature (freeze drying). A very low y i e l d of polymeric material (0.007 g.) was i s o l a t e d by t h i s procedure. The n.m.r. spectrum of t h i s material i n deuterochloroform showed very broad, poorly resolved peaks centred around 4.1X and from 8-9T« Similar spectra have been observed f o r styrene-cyclohexadiene-1,3 copolymers i n samples where the mole f r a c t i o n of styrene becomes very low. A degree of polymerization of 6 was estimated f o r the homo-polymer using a Mechrolab Model 301A osmometer with benzene as solvent. This value i s of the same order of magnitude as those - 142 -obtained for homogeneous, homopolymerizations under comparable 116 conditions, and therefore, degradative chain transfer i s the p r i n c i p a l cause of low degree of polymerization of cyclohexadiene-1,3 under these homogeneous conditions. The very low y i e l d of polymer i n the emulsion system i s consistent with low r e a c t i v i t y of the cyclohexadienyl r a d i c a l produced i n the degradative chain transfer. The rate constant of metathesis with the isopropyl r a d i c a l at 100° i s s i g n i f i c a n t l y higher f o r cyclohexadiene-1,4 than f o r cyclohexadiene-1,3, but l i m i t s of error do not allow the ident-i f i c a t i o n of any s i g n i f i c a n t v a r i a t i o n among the Arrhenius para-meters. Energetic and s t e r i c factors may be expected to combine to make cyclohexadiene-1,3 the les s reactive isomer. F i r s t , i t has the lower standard free energy of f o r m a t i o n ; 1 0 ^ ' 1 2 4 secondly, i t s methylene groups are adjacent, so that an unfavourable s t e r i c f actor may be imposed by mutual crowding of the methylene groups. A comparison of rate constants f o r corresponding reactions measured for the ethyl r a d i c a l 1 to those measured here with the isopropyl r a d i c a l i s given i n Table XI. An in t e r e s t i n g feature of these r e s u l t s i s the difference i n s t e r i c factors f o r ethyl and isopropyl r a d i c a l s . There seems to be about a factor of ten between them i n a l l cases. Of course, t h i s r e s u l t depends upon the accuracy of determinations of the r a d i c a l recombination rate constants f o r e t h y l 1 9 and i s o p r o p y l ^ 0 r a d i c a l s , but would seem to be s i g n i f i c a n t i n view of the magnitude of the e f f e c t . No apparent reason f o r the high e f f i c i e n c y of isopropyl attack e x i s t s . The abstraction from cyclohexadiene-1,3 has s i g n i f i c a n t l y higher ac t i v a t i o n energy f o r isopropyl attack than f o r ethyl TABLE XI Comparison of the Ethyl and the Isopropyl Radical i n r e a c t i v i t y towards the Cyclohexadiene Molecules (kcal./mole) reaction reagent E r - ^ E 2 13 + log^^/A;? l o 9 i o A r 1 D addition to C^g-1,3 C2H* a 5.2 + 0.3 5.5 + 0.2 -12.7 0.40 C 3 H ? 5.8 + 0.4 6.0 + 0.2 -12.0 2.2 abstraction from C 6 H g - l s 3 C 2Hg 5.4 + 0.5 5.0 + 0.3 -13.3 0.11 C 3H 7' 7.1+0.7 6.2+0.4 -11.8 3.5 abstraction from C 6Hg-l,4 C 2Hg 5.8+0.1 5.7+0.1 -12.5 0.58 C 3H y 6.5 + 1.0 ' 6.0 + 0.6 -12.0 2.4 a Values f o r ethyl r a d i c a l s are from the Ph.D. thesis of A. C. R. Brown, University of B r i t i s h Columbia, (1962). Values f o r isopropyl are from t h i s work. b 3 A fac t o r s are i n (cm./molecule s e c ) . c ° Assumes c o l l i s i o n diameters of 4.0, 4.75 and 6.5 A. fo r ethyl, isopropyl and cyclohexadiene re s p e c t i v e l y . Also, A 2 has been taken as 10-10.48 and 10-9.9 cm?/molec.sec. f o r ethyj and isopropyl respectively. - 144 -attack at the 5 % p r o b a b i l i t y l e v e l . This higher act i v a t i o n en-ergy i s just what Is anticipated on the basis of the C-H bond di s s o c i a t i o n energy of the attacking r a d i c a l , 1 2 5 ' 1 2 ^ which i s about 3.5 k c a l . l e s s f o r isopropyl than f o r ethyl , according to recent d e t e r m i n a t i o n s ^ 4 s 1 2 7 The difference i s not s i g n i f i c a n t f o r the abstraction reactions from cyclohexadiene-1,4 and, i n f a c t , consideration of the r e s u l t s from a l l of the attacking r a d i c a l s studied here indicates that the difference i s very small indeed, because t e r t - b u t y l attack has a low a c t i v a t i o n energy. I t i s strange that a difference i n a c t i v a t i o n energies i s observed f o r the 1,3-diene, and not f o r the 1,4-diene. C 2,2*-Azoisobutane as I n i t i a t o r 1» Preliminary k i n e t i c examination of 2,2'azoisobutane  photolysis As the photolysis of 2,2'-azoisobutane was anticipated to be a clean source of t e r t - b u t y l r a d i c a l s , i n common with other azo-alkanes, but has not previously been examined k i n e t i c a l l y , ex-periments were undertaken to determine i t s s u i t a b i l i t y as an i n i t i a t o r system. The p y r o l y s i s of 2,2'-azoisobutane has been examined i n two 128 previous studies by Levy and Cdpeland and by Blackham and Eatough. 1 2 9 128 Levy and Copeland have pyrolyzed 2,2'eazoisobutane i n a s t a t i c system between 180 and 220°. They found the reaction to proceed homogeneously forming nitrogen gas and isobutane. Con-siderable polymeric material was also formed on the glass walls of the vessel. I n t e r e s t i n g l y enough, no isobutylene was found that would arise from t e r t - b u t y l disproportionation; indeed, they found that isobutylene was consumed when i t was added to the - -145 -py r o l y s i s system. They suggested that the t e r t - b u t y l r a d i c a l s presumably released on pyroly s i s e f f i c i e n t l y attack the primary product, isobutylene, and ultimately t h i s process leads to poly-meric residues. The added isobutylene did not suppress the rate of decomposition so they concluded that chain decomposition of the azo compound was unimportant i n t h i s system. This follows from the fact that isobutylene would be expected to be a more ef f e c t i v e chain breaking agent than propylene, which i s known to be an active i n h i b i t o r of r a d i c a l chains at high temperatures because of i t s l a b i l e hydrogen atoms available f o r metathesis. From pressure measurements, Levy and Copeland f i n d that the decomposition, (CH 3) 3C-N=N-C(GH 3) 3 - > N 2 + 2 (CHgJgC* has the rate constant k = 10 1 6 o 3 4exp(-42.8/RT) sec." 1. 129 Blackham and Eatough have examined the p y r o l y s i s from 250 to 290° i n a flow system using hydrogen and helium as c a r r i e r gases. They used the nitrogen evolved as index of the amount of decomposition and found Arrhenius parameters i n ess e n t i a l agree-ment with those of Levy and Copeland. They measured k = 10 1 7 , 1 1exp(-43/RT) sec.""1 corresponding to a p o s i t i v e entropy of a c t i v a t i o n of 16 eu/mole. Using a flow system which protects products from secondary reaction, they were able to recover i s o -butylene as well as isobutane i n equimolar amounts. The r e s u l t s of these two authors are consistent with the i n i t i a l decomposition of azoisobutane into nitrogen and t e r t -butyl radicals, which do not attack the starting material to an appreciable extent. The photo-oxidation of 2,2'-azoisobutane i n oxygen r i c h atmospheres at 25° has been examined by Thomas and C a l v e r t . 1 3 0 - 146 -Their observations are consistent with an excited molecule decom-posi t i o n i n the primary process, since they have measured a pres-sure dependence on the quantum y i e l d of nitrogen produced at 25°. 65 This i s i n contrast to the azomethane photolysis which exhibits a nitrogen quantum y i e l d of unity, independent of pressure. A pressure dependence has also been observed f o r the photo-decompo-37 s i t i o n of l,l"-azo-n-butane where a more complete study has shown the decomposition of the photo-excited molecule to require 76 3,800 c a l . of a c t i v a t i o n energy f o r decomposition. Also azoethane 106 and azoisopropane have been found to have pressure and temper-ature sensitive nitrogen quantum y i e l d s . The same si t u a t i o n i s l i k e l y to p r e v a i l for the 2,2'-azoisobutane case, and may be gen-e r a l f o r the larger, more complex azoalkane systems. 130 Thomas and Calvert suggest the sequence of processes I to explain the photo-decomposition of 2,2'-azoisobutane. AIB + \0> — > AIB* AI.B* + M — > AIB + M* I AIB* > 2 (CH 3) 3C* + N 2 where AIB » 2,2'-azoisobutane In t h e i r system, the t e r t - b u t y l r a d i c a l s thus produced react ex-c l u s i v e l y with oxygen molecules and i n i t i a t e a photo-oxidation sequence. The measurement of /\(tert-Bu, tert-Bu) has been reported several times and, i n t h i s case, no p a r t i c u l a r agreement has been reached. Thus Kraus and Calvert reported /\(tert-Bu, tert-Bu) = 57 4.59 from photolysis of d i - t e r t - b u t y l ketone; Dominguez et a l . 131 reported 3.19 from the same source; B i r e l l and Trotman-Dickenson 132 reported 4.38 from the photolysis of pivaldehyde; Boddy and Robb reported 2.2 from study of the mercury photo-sensitized addition - 147 -of hydrogen atoms to isobutylene. Furthermore, Blackham and . 129 Eatough found t h e i r data consistent with a value of 1.4, a l -though t h i s must be viewed with reserve because of the high tem-peratures and attendant mechanistic complexities not f u l l y con-sidered. In no case were the a n a l y t i c a l r e s u l t s consistent with a material balance between the r a d i c a l s generated and the products removed from the system. In any event, the A r a t i o , f o r t e r t - b u t y l r a d i c a l s i s i n need of c l a r i f i c a t i o n ; since azoalkane photolysis i s frequently found to be a clean source of a l k y l r a d i c a l s , free of troublesome side e f f e c t s , the 2,2'-azoisobutane system promised to be conven-ie n t . Experimental Due c h i e f l y to d i f f i c u l t i e s i n analysis, d i f -ferent techniques have been employed i n t h i s system than have been used i n the case of other i n i t i a t o r s . The photolysis of 2,2'-azoisobutane has been examined f o r the pure compound and f o r the photolysis i n the presence of a large excess of cyclohexadiene-1,4 which functions as an e f f i c i e n t scavenger. These photolyses comprise eight experiments, between 23 and 153°. The r e s u l t s and conditions of these experiments have been collected i n Table XII. The reactants were measured out as usual from storage c e l l s o and photolyzed with 3660 A. a c t i n i c l i g h t . A f t e r photolysis was complete, the reaction mixture was allowed to d i s t i l into the a n a l y t i c a l l i n e where the various traps were maintained at the following temperatures. 1 Variable temperature trap S p i r a l trap 2 n d Variable temperature trap -112 o - 148 -TABLE X I I P h o t o l y s i s of pure 2,2'-Azoisobutane cc. (°K) (sec.) 1 0 ^ 1 2 R X (molec./cm.sec.) T t M a C 4 H 8 G 4 H 1 0 b k 3 / k 2 297 900 2.58 69.1 49.7 49.7 2.56 360 72000 0.478 11.8 0.219 0.0 0.431 -362 7200 4.07 4.64 3.27 3.27 2.39 362 7200 4.21 - 1.80 1.19 1.24 2.22 362 75600 0.580 11.8 0.0703 0.0 0.142 -366 300 4.46 127. 97.6 97.6 (3.28) 400 7200 1.90 1.15 0.685 0.845 2.24 426 7200 1.36 1.44 0.448 1.65 c mol.sec. 1013y2 k 4 / k 3 1.09 10.1 132. Mean value, excluding e n t r i e s i n parentheses 2.35 L i m i t s of e r r o r at 5 % p r o b a b i l i t y l e v e l +0.68 a 1 0 1 7 10 l O t l lA = Concentration of 2,2'-azoisobutane (molecules/cm.-LB] = Concentration of cyclohexadiene-1,4 (molecules/cm., k 3 / k 2 = RC 4H 8^N 2 " RC 4H 1 Q) C RC 4H 1 0> RN 2 - 149 -Nitrogen gas was col l e c t e d pure into the gas burette, under these conditions and measured. After pumping out the gas burette, the s p i r a l trap was warmed to room temperature and a mixture of i s o -butane and isobutylene was collected,into the gas burette and measured. This f r a c t i o n was found to be free of any appreciable impurities by examination by gas chromatography. In routine runs, t h i s f r a c t i o n was analyzed on a s i l i c a gel packed column at 180°. A greaseless sampling valve was used to avoid absorption of the butane or butene into the tap lubricant. The r a t i o of pressures of isobutane to isobutylene could be found by m u l t i p l i -cation of the peak area r a t i o by 0.97. This c a l i b r a t i o n factor was found i n the usual way by running chromatograms of synthetic mixtures of isobutane and isobutylene. This, together with the t o t a l amount of both gases from the gas burette measurement gave the separate amounts of each component. No convenient analysis f o r the d i - t e r t - b u t y l product was established. On a l l packed columns available, t h i s compound was not separated from the azoisobutane peak i n a s u f f i c i e n t amount to y i e l d reasonable analysis. I t appears that t h i s compound i s amazingly similar to i t s azoisobutane precursor, perhaps due to the f a c t that the bulky t e r t - b u t y l groups e f f e c t i v e l y shield the nitrogeneous portion of the azoisobutane molecule. Analysis of the residues from photolysis on the c a p i l l a r y polyethylene g l y c o l column did succeed i n s p l i t t i n g the azoisobutane peak, showing the presence of the d i - t e r t - b u t y l product, but s t i l l was not convenient for quantitative purposes. No other compounds were evident i n the c a p i l l a r y column analysis, even a f t e r the column was baked at 120° (the maximum operating temperature) for extend-ed periods. Treatment of the data There i s no evidence i n the analyt-i c a l data f o r any products other than isobutane, isobutylene and d i - t e r t - b u t y l . Furthermore, as Table XII reveals, i n those runs where no cyclohexadiene-l 94 has been added, and below 127°, the isobutane y i e l d i s seen to be equal to the isobutylene y i e l d . Under these conditions, i t i s proposed that Scheme J below com-p l e t e l y accounts f o r the products. AIB + - > AIB* (la) AIB* + M > AIB + M * (lb) AIB* > 2 (GH 3) 3C- + IM2 (Ic) ^ 2 (GH 3) 3C- > (CH 3) 3C-C(CH 3) 3 (2) 2 ( C H 3 ) 3 0 > (CH 3) 3CH + (CH 3) 2C=CH 2 (3) where AIB = 2,2'-azoisobutane In t h i s regime, k3 R c 4 H 8 R ° 4 H 8 k 2 \ H 1 8 \ - \ H 1 0 (26) and calculated values have been presented i n Table XII. This r e l a t i o n depends upon the assumption that t e r t - b u t y l r a d i c a l s do not add to the double bond of the azo compound i n the way methyl r a d i c a l s add to azomethane. In view of the s t e r i c conditions p r e v a i l i n g around the N=N double bond, t h i s assumption seems j u s t i f i e d . Even i n the azomethane case, the addition i s suffucient-l y slow that i t would not be observed at the high rate of photol-y s i s used during experiments with azoisobutane-cyclohexadiene-1,4 mixtures, which are l i s t e d i n Table XIII, and which provide sub-s t a n t i a l l y the same r e s u l t as found here f o r /\(tert-Bu, tertrBu). The value f o r ZMtert-Bu, tert-Bu) measured i n t h i s way i s 2.35 + 0.68. This i s v e r i f i e d , and considerably improved i n accuracy In - 151 -runs with added cyclohexadiene-1,4, to be described i n the next section of t h i s thesis. The excellent agreement i n the rates of isobutane and of i s o -butylene below 127° Indicates that the metathetical reaction, (CH 3) 3C + (CH 3) 3CN=NC(CH 3) 3-^>(CH 3) 3CH + (CH 3) 2(CH 2)CNNC(CH 3) 3 does not proceed under these conditions. An experiment at 89° does show a s l i g h t excess of isobutane over isobutene and indicates Jy -13 3 V a value of k 4 / k 2 - 1.09 x 10" (cmf/molec.sec.) 2. I f t h i s i s ac-cepted as accurate, then the metathetical reaction with t h i s com-pound i s 3.2 times slower than the comparable process measured for azomethane. Kraus and Calvert found d i - t e r t - b u t y l ketone i n e r t towards the t e r t - b u t y l r a d i c a l up to 322° under similar conditions of rapid photolysis. The experiments at 127 and 153° show les s isobutylene than isobutane; t h i s may indicate the onset of reaction (4), or, a l t e r n a t i v e l y , may be caused by the consum-sumption of Isobutylene by t e r t - b u t y l r a d i c a l s as observed by 128 Levy arid Copeland f o r t h e i r p y r o l y s i s system. I f i t i s the abstraction reaction (4) that i s responsible, then equation (26) should s t i l l apply assuming termination of secondary r a d i c a l s by t e r t - b u t y l r a d i c a l s . However, the r e s u l t at 153° has FU u FL, , hence application of equation (26) f a i l s . On the other hand, consumption of the isobutylene seems unl i k e l y at these low con-versions. The e f f e c t may, aft e r a l l , be a separate thermal re-action y i e l d i n g isobutane. At any rate, below 100°, the system seems safe from interference from t h i s cause, and use of the 2,2 f-azoisobutane system as an i n i t i a t o r has been l i m i t e d to t h i s low temperature range. Two experiments have been done with a twentyfold excess of - 152 -lcyclohexadiene-1,4 at 87 and 89°. Low l i g h t i n t e n s i t i e s have been employed to ensure scavenging by the reactions •C(CH 3) 3 + (MJ £ -> + (CH 3) 3CH The r a t i o FL, u /RKT then takes the values of 1.97 and 2.02 re-^4 H10 N 2 spectively at 87 and 89°, i n close agreement with the predicted value of 2.00. The molecular elimination reactions (CH 3) 3C-N=N-C(CH 3) 3 + h ^ -> N 2 + (CH 3) 3C-C(CH 3) 3 i s therefore u n l i k e l y to account f o r more than 1 % of the nitrogen formed; i t i s therefore not more important than i t i s f o r azo-methane. K l e i n et a _ l . 4 2 have suggested that large values of 01^ may ex i s t f o r azoisopropane photolysis i n order to explain cer-t a i n of t h e i r r e s u l t s . In view of t h i s t e r t - b u t y l r e s u l t , a large <0m f o r the isopropyl system seems unlikely. Discussion These experiments on azoisobutane photolysis have yielded a value of / \ t e r t - B u , tert-Bu) = 2.35 + 0.68 based upon t the assumption that the t e r t - b u t y l r a d i c a l does not add to the c' i double bond of the azo compound. Results from.; the experiments on cyclohexadiene-l,4-2,2'-azoisobutane mixtures agree with t h i s measurement and provide greater pr e c i s i o n . These r e s u l t s are l i s t -ed i n Table XIII and w i l l be described i n d e t a i l . A l l of the ex-periments taken together y i e l d the value A (tert-Bu. tert-Bu) = 2.33 + 0.28, Independent of temperature from 23 to 127°; the error expressed at the 5 % p r o b a b i l i t y l e v e l . An Arrhenius plot of these t o t a l r e s u l t s has been included on Figure 20. This value i s con-siderably lower than that of previous measurements save f o r that of Boddy and Robb ( ZX= 2.2) and the rather uncertain measure-ment of Blackham and Eatough, 1 2 9 ( A = 1.4). - 153 -36 A re-evaluation of the r e s u l t s of Kraus and Calvert i s in? st r u c t i v e . They have measured the r a t i o , k 3 A 2 where k 3 A 2 = (Rr; „ + FL, u ) / 2 R p U a 4.4 + 0.7 (102 to 178°). However, U4 H8 % H 1 0 U8 H18 an alternative i n t e r p r e t a t i o n of t h e i r r e s u l t s y i e l d s the value: = 2.3 + 1.1 k 3 A 2 = R c 4 H 8 / / R c o - ^ ( R ^ + RQ^) over the same temperature range i n close agreement with the pre-sent r e s u l t s . The ambiguity i n the i n t e r p r e t a t i o n of t h e i r re-sults arises from a f a i l u r e to achieve a material balance, ex-ressed by the r a t i o : Rr „ / Rpn - ^(FU „ + Rr „ ) = 0.50 + 0.16 U8 H18 ^ U C4 H8 b4 H10 whereas t h e i r mechanism predicts the value 1.00. A consideration of the previous measurements of A values for the various n-alkyl r a d i c a l s suggests that the azo compounds may provide the most r e l i a b l e r e s u l t s . Determinations of A ( E t , Et) = 0.14, 7 7 A(n-Pr, n-Pr) = 0.16, 1 3 3 A(Me, n-Bu) = 0.15 lead one to expect values of ca_. 0.15 f o r disproportionation r a t i o of n-alkyl r a d i c a l s . However, A(n-Bu, n-Bu) has been found to 134 have ,the value 0.94, much higher than f o r similar n-alkyl sys-tems, during chemically sensitized decomposition of n-butyl form-39 ate. Trotman-Dickenson found the r a t i o to be temperature de*.. pendent during photolysis of n-valeraldehyde. However, the re-cent redetermination of A(n-Bu, n-Bu) using the photolysis of 1,1'-azo-n-butane yielded the value 0.14; t h i s was observed i n both the f l a s h photodecomposition of the azobutane and i n the stationary state photolytic system. This r e s u l t i s i n accord with expectations based on the behavior of the lower homologues. Furthermore, r e s u l t s from azoalkane photolysis f o r the lower - 154 -homologues are also lower than those f o r ketone photolysis, but the e f f e c t i s much smaller. Thus A ( E t , Et) = 0.12 from azo-ethane"7^ and A('i'so-Pr, iso-Pr) = 0.54 from azoisopropane 1 0 0 vs. 0.14 and 0.58 respectively from ketone systems. From the above considerations, i t i s concluded that the "low" value of the disproportionation-combination r a t i o of t e r t - b u t y l r a d i c a l s observed during 2,2'-azoisobutane photolysis may be the most representative one. Furthermore, redetermination of A values fo r a l l the higher molecular weight a l k y l r a d i c a l s using azo-alkane photolysis as the r a d i c a l source seems advisable. In the meantime, Kraus and Calvert's suggestion 3^ that A values f o r the butyl r a d i c a l isomers are i n the same r a t i o as the number of hy-drogen atoms available f o r disproportionation i s uncertain. 2. The photolysis of 2,2'-azoisobutane-cyclohexadiene-1,4  mixtures In order to extend the information on a l k y l r a d i c a l r e a c t i v -i t y with cyclohexadiene-1,4 and cyclohexadienyl r a d i c a l s , the photolysis of the mixed vapours of 2,2'-azoisobutane and cyclo-hexadiene-1,4 has been examined. The nature of the experiments was the same as that already discussed f o r azomethane and d i -isopropyl ketone photolysis i n the presence of cyclohexadiene-1,4. Experimental The system has been examined i n eight experi-ments from 26 to 100°. The mixed vapours have been photolyzed with o 3660 A. radiati o n using the f u l l l i g h t i n t e n s i t y available at that wavelength from the B.T.H. medium pressure mercury lamp. After photolysis, the reaction mixture was allowed to d i s t i l into the a n a l y t i c a l l i n e where the various traps were maintained at the following temperatures? I s Variable temperature trap - 90° S p i r a l trap -198° 2nc* Variable temperature trap -112° Nitrogen was removed pure into the gas burette, and measured. : o Following t h i s , the s p i r a l trap was warmed and butane and butyl-ene were collected and measured i n the same way as f o r the analy-s i s of the pure azoisobutane photolyses. Following the analysis of the gases, the residues i n the a n a l y t i c a l l i n e were allowed to vapourize, were mixed with a manometrically measured amount of 3-methylpentane as an i n t e r n a l gas chromatographic standard, and removed from the vacuum system. Since the usual i n t e r n a l standard, n-heptane, was not separated s u f f i c i e n t l y from the azoisobutane peak on the preferred a n a l y t i -c a l column, the isomeric hexane, 3-methylpentane, was used i n -stead. This sample was mixed with a small amount of pentene-1 s o l -vent and injected onto a s p e c i a l l y prepared 1 meter, packed 20 % hexadecane on f i r e b r i c k column run at 54°. This column was e f f e c t -ive i n separating the reaction mixture into the following com-ponents i n order of elutions pentene-1 (solvent), 3-methylpentane (i n t e r n a l standard), benzene, cyclohexadiene-1,4 and f i n a l l y azo-isobutane and d i - t e r t - b u t y l together. The benzene y i e l d could be found from the known amount of 3-methylpentane by comparison of t h e i r peak areas i f the benzene peak area were mult i p l i e d by 0.971. I t was not possible to drive off the expected compounds I and II tert-Bu tert-Bu 0 TABLE XIII Metathesis between the tert - B u t y l Radical and Cyclohexadiene-1,4 (°K) (sec.) 10"12RX (molec./cm? sec.) (sec."* 1) ( c£ ' \mol.2sec.2 V k 3 T t W A N2 C4H8 °4 H 10 C6H6 ( W + t B > RN/W k 3/k 2 A B 300 1800 7.00 3.22 108.2 66.6- 88.5 8.50 10.22 15.5 2.33 (1.74) 51.0 303 1800 6.14 3.14 104.8 61.6 88.4 9.78 9.28 17 a 2.36 1.36 69.0 320 1800 5.05 1.36 98.7 62.4 78.4 5.65 6.41 19.3 2.31 1.20 96.6 326 1800 6.51 1.37 115.3 73.8 92.2 6.64 7.88 17.7 2.49 1.30 100. 337 1800 5.65 1.18 97.9 61.0 79.0 6.50 6.83 17.3 2.40 1.30 125. 357 1800 5.16 1.03 93.9 54.9 78.3 8.00 6.19 18.2 2.33 (1.08) 202. 370 900 3.35 0.816 115.6 64.0 96.2 11.70 6.16 21.6 2.13 1.33 324. 373 300 5.51 0.668 126.7 73.6 103.8 11.25 6.18 23.0 2.16 1.46 330. Mean value, excluding entries i n parentheses 2.32 1.33 Limits of error at 5 % p r o b a b i l i t y l e v e l +0.29 +0.24 (Tl ON a 1 Q 1 7 17 10 1' = Concentration of 2,2'-azoisobutane (molecules/cm^) BJ -Concentration of cyclohexadiene-1,4 (molecules/cm.) A - R c 6 H 6^ G 4 H "10 6He) - 157 -at the maximum temperature (85-90°) that the column w i l l safely withstand. Supplementary analysis of a portion of the reaction mixture on the c a p i l l a r y polyethylene g l y c o l column at room temperature separated the reaction mixture into the following components i n order of e l u t i o n : pentene-1 (solvent), 3-methylpentane ( i n t e r n a l standard), azoisobutane and benzene together, preceded by an a l -most resolved d i - t e r t - b u t y l shoulder, and cyclohexadiene-1,4. Heating the column gradually to 120° caused the elu t i o n of a-nother, single peak judged to be the two, unresolved compounds, I and I I , i n analogy with the other systems studied. In t h i s case, these two compounds are so similar i n vapour pressure and chemical properties that the c a p i l l a r y column i s no longer e f f e c t -T T T ive i n resolving them. No analysis f o r the r a t i o k ^ c ^ c c was possible, therefore, i n the case of the reactions of t e r t - b u t y l r a d i c a l s with the cyclohexadienyl r a d i c a l . The c a p i l l a r y column analysis would have shown the presence of any cyclohexadiene-1,3 present i n the reaction products. None was observed during any run. The experimental conditions, and rates of formation of the various products have been collected i n Table XIII along with r e s u l t s to be explained below. Treatment of the data In view of the previous study of the photolysis of 2,2'-azoisobutane, and by analogy with the other high i n t e n s i t y studies of i n i t i a t o r systems with cyclohexadiene-1,4, the r e s u l t s have been considered to arise from the mechanism of Scheme K. Although no d i r e c t experimental evidence has been found to disprove the reactions (6e)s - 158 -(CH 3) 3C- + C6H* ^ > (CH 3) 2C=CH 2 + (C^Hg-1,3 or CgHg-1,4) t h i s has not been considered i n the mechanism because (a) i t does not occur f o r either ethyl or isopropyl r a d i c a l s and (b) by the material balance method of analysis f o r reaction (6c) used i n the k i n e t i c analysis, whence kg d/k^ c = ^,5^/(^5 ~ R5<j)» neglect of (6e) would lower k ^ / k ^ values, where i n f a c t we s h a l l see that i t i s su r p r i s i n g l y high. S i m i l a r l y , k 3 / k 2 values would be abnormally high, e s p e c i a l l y at the higher temperatures i f reaction (6e) were important, whereas the measured value, 2.33, i s lower than previous determinations. AIB + — > AIB* .: AIB* + M — > AIB + M* AIB* — > 2 (CH 3) 3t> + N 2 2 (CH 3) 3C. 2 > ( C H 3 ) 3 C - C ( e H 3 ) 3 K > (CH 3) 3CH + (CH 3) 2C=CH 2 C 6H 8 -i,4 + (CH 3) 3C* > ( C H 3 ) 3 C H + C 6 H 7 (CH 3) 3C* + C 6H ? ^ > 6d •> (CH 3) 3GH + C 6 H 6 No metathetical reaction from the azo compound has been con-sidered as the preliminary i n v e s t i g a t i o n of 2,2'-azoisobutane photolysis did not reveal such r e a c t i v i t y below 127°. In t h i s study, a l l experiments were done below 100°. S i m i l a r l y , no add-i t i o n reaction has been considered; such a reaction would not a l t e r conclusions about metathesis from the diene or subsequent r e a c t i v i t y of the cyclohexadienyllradical, even i f i t were to occur. Assuming the simple mechanism above, Scheme K, the following - 159 -steady-state equations are r e a d i l y derived. = R c 4 H 1 0 R c 4 H 8 R c 6 H 6 k3 _kj3 _ R g 4 H 8  k2 R N 2 " R C 4 H 1 0 * RG 6H 6 _ k6d , R g 6 H 6  k6c R 6 ' R C 6 H 6 R C 6 H 6 (27) (28) (29) Ftp I T — FL-> j. — 2 R/— T I U4 H10 C 4 H 8 C 6 H 6 Here, the r a t i o k^/k^ has been calculated i n preference to the more usual k^/krf since the rate of formation of the dimeric product, d i - t e r t - b u t y l , was impossible i n t h i s series of experi-ments. As both reactions are probably without a c t i v a t i o n energy, t h i s i s of no serious consequence. Because of the lack of analy-s i s f o r d i - t e r t - b u t y l . the rate of i t s formation has been assumed to be R Q ,_J = R N " R 3 ~ R 6 + R 6 d ' t h l s ^ e a d s ' t o form of equation 8 18 2 (28). Values f o r k^/k|, k^/k^ and ZA= k^ d/k^ c, have been l i s t e d i n Table XIII. Thesresuits f o r abstraction of hydrogen from cyclohexadiene-1,4 by t e r t - b u t y l r a d i c a l s are plotted on Figure 19. S t a t i s t i c a l analysis of the r e s u l t s shows that 1 0 1 3 k6 _ 1 0 ( 5 . 6 + 0 . 5 ) ( r 5 . 3 ± 0.8) / cm? ^ ^2 RT \molec. sec.. Also, for the disproportionation ratios 1,, s t a t i s t i c a l analysis - 161 -shows kg 7/^ = 2.3 + 0.3 and l c ^ / k ^ c = 1.3 + 0.2 Independent of temperature i n both cases within experimental error. Also i n Table XIII are values of F L , / | X I B] and of [AIB] + source remains constant from run to run, one expects /[_AIBj to be nearly constant i f the quantum y i e l d of nitrogen i s invar-iant with temperature and t o t a l pressure. As t h i s r a t i o increases steadily from the lowest to the highest temperature, i t i s a qual-i t a t i v e i n d i c a t i o n that either pressure or temperature or both i n -fluence the primary photochemical process. Since both temperature and pressure vary, i t i s not possible to specify the d e t a i l s , but the scheme given i n the proposed mechanism seems to be reasonably well established f o r similar azoalkane systems.76>106 Discussion As has already been mentioned, the value of /\(tert-Bu, tert-Bu) =2.3 + 0.3 measured here agrees with the measurement made on the pure 2,2'-azoisobutane system and appears to be representative. In t h i s system, the value of k^ d/k^ c = Ai tert-Bu. C^ Hy.) has not been calculated from the c a p i l l a r y column peak areas f o r com-pounds I, II and C^H^ as was done f o r azomethane and di-isopropyl ketone experiments because the benzene peak was not resolved from azoisobutane. However, s u f f i c i e n t p r e c i s i o n was obtained from the material balance technique ( i e . , by equation (29)) to obtain a r e l i a b l e value. The two methods are adjusted to give the same value, anyway, and the c a p i l l a r y column technique i s used only to improve the errors, which are not severe i n these experiments i n measurement of k ^ / k ^ . The numerical value of the disproportionation to combination r a t i o i s s u r p r i s i n g l y high, i n the case of t e r t - b u t y l r a d i c a l . I t To the approximation that the i n t e n s i t y of the l i g h t - 162 -i s consistent with 57 % of tert-butyl-cyclohexadienyl interactions leading to disproportionation and 43 %, the smaller f r a c t i o n , re-combining. The surprising tendency f o r the higher molecular weight, branched r a d i c a l s to disproportionate when the number of hydrogen atoms available f o r disproportionation remains constant i s contrary to the usual expectations. Thus i n t u i t i v e l y one may expect t e r t - b u t y l to be a sluggish hydrogen atom remover as i t has the weaker C-H bond. These factors are considered more f u l l y i n r e l a t i o n to the whole series of r a d i c a l s i n the "General Discus-sion" section of t h i s thesis. No temperature dependence has been found f o r the Aitert-Bu»C^Hy values, as i s evident from Figure 20. D. Diethyl Ketone as I n i t i a t o r Diethyl ketone has been examined and used extensively i n t h i s laboratory as an ethyl r a d i c a l source i n photochemical systems. No complications have arisen to impair i t s usefullness. The d i e t h y l ketone-cyclohexadiene-1,4 system has been studied i e a r l i e r by Brown and t h i s work has been described i n the ."Intro-duction" section as background fo r the present research. His ex-periments yielded the r e s u l t s : 10 1 3-!^ = 1 0 ( 5 - 7 ± 0.1) c x p-(5.8 ± O.D / cm3 \ ^ ^6. R T \molec.sec.y and A E t , C^H-p = k$c/ k6c " ° * 3 8 - 0 , 0 3 i n d e P e n d e n t o f temper-ature f o r the reactions C 2 H 5 + C 6 H 8 ~ X ' 4 " > C 2 H 6 + C 6 H 7 C 2 H 5 + C 6 H 7 ~ > C 2 H 6 + C 6 H 6 6c \ ^ 92H5 „ / \ j?2H5 - 163 •-Figure 20. Disproportionation and Combination of the ter t - B u t y l Radical with I t s e l f and with the Cyclohexadienyl i ^ r 1 » i • i » i • i • i 1.3 t1 5 ® — ® c/) ^ Q9I f a * No measurement of k ^ c / k ^ i s possible i n t h i s case, Q5h — i - A _ A ± A A—A—-A J 1 I i I . I i l . l . l 2.7 2.9 3.1 3,3 10 3 A ® k * / k s = 1 0 k6d/( k6c + k6c> See Scheme J x S <d o - 164 -Since he did not have the use of temperature-programmed c a p i l l a r y columns when his invest i g a t i o n was carried out, no analysis of the r e l a t i v e y i e l d s of I and II was obtained. In order to complete the inves t i g a t i o n of the ethyl-cyclo-hexadienyl r a d i c a l interactions, some experiments have been car-r i e d out using the more complete a n a l y t i c a l procedure f o r the heavy reaction products. Experimental Six experiments between 34 and 84° have been performed on gaseous mixtures of d i e t h y l ketone and cyclohexa-diene-1,4. No detailed product analysis was attempted, as the aim of t h i s study was only to obtain data on the dimeric products from , , o reaction (6c). A f t e r photolysis of the mixture with 3130 A. r a d i a t i o n under.the conditions l i s t e d i n Table XIV, the gaseous products were pumped off through a dry ice - acetone trap and the l i q u i d residues were removed from the vacuum l i n e and mixed with pentene-1 solvent. This mixture was then immediately analyzed on the polyethylene g l y c o l c a p i l l a r y gas chromatography column. The column was held at room temperature f o r 10 minutes while pentene-1, benzene, cyclohexadiene-1,4 and d i e t h y l ketone were eluted. A f t e r heating the column to 120°, the expected p a i r of peaks due to compounds I and II were observed. The area r a t i o of these peaks was measured and t h i s was assumed to be equal to the mole r a t i o of IsII i n the mixture? thus Area(l)/Area(II) = k^ c/k^. This r a t i o has been entered f o r each run i n Table XIV T T T and the function.! + l ° 9 i o ^ 6 c A 6 c ^ a s keen Plotted as a function of 10 /T i n Figure 21. These data are consistent with k^ c/k^ c = 0.811 + 0.088, independent of temperature. - 165 -TABLE XIV Reactions of the Ethyl Radical with the Cyclohexadienyl Radical (°K) T (sec.) t M a M a 307 3600 10.0 1.55 0.812 307 1200 8.76 1.05 0.834 322 1200 7.80 1.58 0.780 334 1200 6.89 1.32 0.870 336 1200 7.64 1.67 0.792 358 1200 Mean value 7.23 1.37 0.776 0.811 Limits of error at 5 % p r o b a b i l i t y l e v e l +0.088 a 10, 7 (bl = Concentration of d i e t h y l ketone (molecules/cnu) 10 [B] = Concentration of cyclohexadiene-1,4 (molecules/cm.) - 166 -Figure 21. Disproportionation and Combination of the Ethyl Radical with the Cyclohexadienyl Radical 1.0 CD 3 Q 5 t as 1 + log k 6 d/(k£ c + k**) (A. C. R. Brown) 3.0 I D 3 / ! 3.2 ® k / k s = k6c' k6c T T T 1 + log k 6 d / ( k 6 c + k££) measured by A. C. R. Brown. - 167 -E. Rate Constants, Arrhenius Parameters and S t a t i s t i c a l Data A summary of a l l the rate constants measured i n t h i s work has been compiled i n Table XV. Also l i s t e d are the constants B and C obtained by f i t t i n g the experimental data to the straight l i n e : k 3 13 + l o g 1 0 - ^ - = (C + 6£) (B + 0g) In the table, (5g, 0^ and (3^ . are standard deviations of C, B and of the i n d i v i d u a l measurements of 13 .+ l o g ^ ^ / k ^ . The con-stants and standard deviations were calculated using the lea s t mean squares technique. Calculations were carried out using an IBM 7040 computer and a l i b r a r y program available through the University Computer Centre. Errors calculated from the standard deviations have been expressed at the 5 % p r o b a b i l i t y l e v e l . In Table XVI, the s t e r i c f a c t o r , p, has been calculated f o r each reaction studied assuming values f o r the c o l l i s i o n diameter as l i s t e d i n the table. The C o l l i s i o n y i e l d has been determined at the mean temperature (T ) over which the rate measurements were made. The values used f o r the rates of the pertinent r a d i -c a l recombination reactions are tabulated. In view of the uncer-t a i n t i e s i n these rates and i n the selection of c o l l i s i o n diameters, these s t e r i c factors are only crude approximations at best. TABLE XV Reaction Arrhenius Parameters and S t a t i s t i c a l Data (kcal./mole) reaction No. of runs C b B Sk ^B 13 + 1 0 9A| E r • - y2E2 C H 3 + CH3NNCH3 (M) a 14 6.009 1.883 0.0565 0.137 0.0506 6.0 + 0.3 8.6 + 0.5 CH' + CH3NNCH3 (A) 13 4.717 1.295 0.0387 0.094 0.035 4.7 + 0.2 5.92 + 0.35 C 3 H 7 + C gHy.COC 3Hy ( M ) 11 7.506 2.753 0.155 0.526 0.222 7.5 + 1.2 12.6 + 2.3 C 3H' + C^Hg=l 93 ( M ) 9 6.156 1.557 0.0388 01182 0.067 6.2 + 0.4 7.1 + 0.7 C 3 H 7 + C 6H 8-1,3 (A) 11 6.024 1.276 0.0285 0.100 0.036 6.0 + 0.2 5.8 + 0.4 CH' + C 6H 8-1 ,4 (M) 16 5.882 1.207 0.0341 0.1072 0.0361 5.9 + 0.2 5.52 + 0.35 C 2 H 5 + C 6H 8-1 94 d ( M ) 7 5.724 1.270 0.0074 0.038 0.0137 5.7 ;•+ 0.1 5.8 J + 0.1 C 3 H 7 + C 6H 8-1 ,4 ( M ) 14 6.047 1.408 0.0596 0.302 0.112 6.05 + 0.66 6.4 + 1.1 C 4 H 9 + C 6H 8-1 S4 ( M ) 8 5.618 1.166 0.0492 0.220 0.073 5.6 + 0.5C 5.3 + 0.8 a The l e t t e r s i n parentheses r e f e r to the metathetical (M) and addition (A) reactions. b B and C are the c o e f f i c i e n t s of the straight l i n e s 13 + l o g 1 0 ( k r / k | ) = +C - 10 3 B A f i t t e d to the experimental data, m,, 6Q and <3g are the standard deviations i n 13 + lpg^o^r/^^* ^ a n d . B respectively.I^The l i m i t s of error are calculated at the 5 % probability l e v e l . c • v-> J y> For t h i s case, t h i s r e s u l t represents 13 + l°9^oA6/''A3' r a ' , : n e r than 13 + log^QA^/Ag. d A. C. R. Brown.1 TABLE XVI Absolute Rate Constants of the Addition and Metathetical Reactions a y + l ° 9 i 0 A r / / A 2 (°K) ( cm?/molec. s< reaction 13 l o g 1 0 A 2 l o 9 l O A r m l o g i 0 Z 10 3p l o g 1 0 k r ( 6 0 CH" + CH3NNCH3 (M) b 6.0 -10.3 -12.14 373 4.5 -9.29 1.41 -17.8 CH' + CH3NNCH3 (A) 4.7 -10.3 -13.44 373 4.5 -9.29 0.0707 -17.3 C 3 H 7 + C 6H 8-1,3 (M) - 6.2 - 9.9 -11.74 368 5.6 -9.28 3.47 -16.4 C 3 H 7 + C 6Hg-l s3 (A) 6.0 - 9.9 -11.94 361 5.6 -9.28 2.19 -15.7 CH' + € 6 H 8 - 1 , 4 (M) 5.9 -10.3 -12.24 343 5.0 -9.20 0.912 -15.8 C 2 H 5 + C 6H 8-1,4 (M) f 5.7 -10.5 -12.52 358 5.3 -9.28 0.575 -16.3 C 3 H 7 + C 6H 8-1,4 (M) 6.05 - 9.9 -11.89 379 5.6 -9.27 2.40 -16.1 C 4 H 9 + C 6H 8=1 S4 (M) . 5.6 C - 9.8d -12.29 337 6.0 -9.28 0.967 -15.8 vO a "3 A l l pre-exponential factors and c o l l i s i o n frequencies expressed i n (cmf/molec.sec.) b The l e t t e r s i n parentheses r e f e r to the metathetical (M) and addition (A) reactions. c % V For t h i s case, t h i s r e s u l t represents 13 + log^A^/A^, rather than 13 + log l 0A^/A|. d This entry i s l o g i o A 3 where A 3 has been assumed to be 10 1 J"(l./mole sec.) - 10° 9 , 3(cm?/mol.sec.) i n the absence of a measured value. e T m = mean temperature over which A^ was measured. f A. C. R. Brown. 1 - 170 -GENERAL DISCUSSION  A * Disproportionation and Combination Reactions of the Cyclohexa- dienyl Radical I• Summary of the experimental observations. The numerical r e s u l t s for the disproportionation reactions observed i n the experiments reported above have been collected i n Table XVII. A l s o 9 i n t h i s t a b l e 9 the r e s u l t s of a similar invest-i g a t i o n o£ the combination reactions of the methyl r a d i c a l with 135 the phenoxy and ortho-methyIphenoxy r a d i c a l s have been compil-135 ed. The reaction scheme used by Mulcahy and Williams to ex-p l a i n the product d i s t r i b u t i o n i s A similar scheme describes the r e s u l t s of the o-methylphenoxy case. As the reacting species i s a cyclohexadienyl type r a d i c a l i n at least some of i t s resonance forms, the obvious s i m i l a r i t y of t h i s study to the work reported here w i l l provide a useful comparison i n subsequent discussion. The isomerization of the unsaturated ketone intermediates i s assumed to occur without any disturbance of the i n i t i a l l y formed ortho-para d i s t r i b u t i o n . TABLE XVII Termination Reactions of Cyclohexadienyl and Phenoxy Radicals reactants para comb* ortho comb, other a Z^R? k^/k^ 1 CH' + b phenoxy 45 % 49 % 6 % CH' + o-methylphenoxy b 66 26 8 -CH 3 + cyclohexadienyl c 45 34 21 0.27 0.07 0.76 ± c 2 H i + cyclohexadienyl 40 C 33 C 27 d 0.38 ± 0.03d 0.81 + iso=C 3H° + cyclohexadienyl c 36 30 34 0.52 ± 0.09 0.85 ± ter't-C^H^ + cyclohexadienyl c 43 57 1.33 + 0.24 -;lohexadienyl + cyclohexadienyl c 69 31 0.45 + 0.12 For phenoxy r a d i c a l s , t h i s column refe r s to oxygen attack; for cyclohexadienyl, t h i s column r e f e r s to disproportionation, where C^ HA- donates a hydrogen atom . m. F. R. Mulcahy and D. J . Williams, Nature, 199, 761 (1963). This work. A. C. R. Brown.1 - 1 7 2 -As we can see from Table XVII, the r a t i o of ortho to para combination generally favours the para p o s i t i o n to an extent that exceeds the expected s t a t i s t i c a l d i s t r i b u t i o n that requires twice as much ortho combination as para combination, except for the 6-methyiphenoxy and methyl system. A l s o 9 the combination r a t i o k^/k^ 1 f p r the compounds studied presently does not vary s i g n i f i c a n t l y from compound to compound, i n view of the experimental error- In view of the constancy of I II the k^/ic^ r a t i o s , i t i s quite probable that the corresponding r a t i o f o r t e r t - b u t y l r a d i c a l i s i n the region of 0.77 + 0.17 as f o r the other r a d i c a l s . On the contrary, the C^Hy) values vary smoothly from the lowest, ZM^e, C^Hy), through the ethyl and isopropyl values to the highest, /jjtert-Bu, C^Hy). This trend i s quite r e a l at the 5 % p r o b a b i l i t y l e v e l of experimental error, and spans a f a c -t o r of f i v e from methyl to t e r t - b u t y l r a d i c a l s . In Figure 2 2 the A ( R S ^ 6 H 7 ^ r e s u l t s measured here are pre-sented i n a "log-log p l o t " i n the form 1 + logZ\(R, ^6^7) P i t -ted against 1 + logZ\(Ftj R) f o r the ethyl r a d i c a l and i t s higher homologues. Also plotted are similar r e s u l t s measured i n other laboratories f o r hydrogen donor r a d i c a l s other than C^Hy • The values adopted have been l i s t e d i n Table II where the preferred r e s u l t has been underlined. In t h i s f i g u r e , i t can be seen that the tendency to dispro-portionate increases with considerable r e g u l a r i t y f o r a given constant hydrogen donor as the conjugate reactant changes from ethyl to t e r t - b u t y l . This behavior i s i n contrast to the pre-di c t i o n s of Kraus and Calvert who suggested that the Z \ values of the r a d i c a l s would be i n proportion to the number of available - 173 -Figure 2 2 . Comparison of A(R, C6H*) values to similar r e s u l t s f o r A l k y l Radicals r»—i—i—i—r toh of < 0 o •Q5h J I I i i C 2H£ 0.5 C 6 H - C ^ H - i to 1 + L O G A(R,R) ^ R= tex-t-Butyl = Isopropyl A J I I I L^i = Cyclohexadienyl ® = Ethyl - 174 -hydrogen atoms f o r disproportionation. In Figure 22, a l l connect-ed points would then l i e on a horizontal straight l i n e . The num-ber of available hydrogen atoms i s c l e a r l y not s u f f i c i e n t to ex-p l a i n Z \ values, even i n related systems. The value f o r AjC^Hj, C^Hj) has been measured i n the low i n t e n s i t y photolysis of azomethane-cyclohexadiene-1,4 mixtures. I t s p o s i t i o n on Figure 22 i s reasonably close to the l i n e defined by the other r a d i c a l s ; however, as w i l l be discussed i n a l a t e r paragraph, the cyclohexadienyl r a d i c a l does not behave t y p i c a l l y as a hydrogen atom acceptor i n i t s reactions with the a l k y l r a d i -c a l s . I t i s l i k e l y that the point f o r A(C^H 7» C^Hj)xs not to be expected to correlate with the C^Hy) r e s u l t s . The r e s u l t s of the measurements on the a l k y l plus cyclohexa-dienyl systems studied here lead to the following observations^ 1. /X(R-, C^Hy) values increase by a factor of 5 from R = CH^ to R = t e r t - b u t y l . 2. The r a t i o k^/k 1 1 i s reasonably constant; the value 0.77 + 0.17 i s c h a r a c t e r i s t i c of a l l the a l k y l r a d i c a l s studied. I t seems l i k e l y that the t e r t - b u t y l r a d i c a l would reveal a similar r e s u l t , were i t measureable. 3. The recombination consistently favours the unconjugated dierie structure, not only f o r the r e s u l t s of t h i s study, but also f o r an independent but related study of the recombination of methyl r a d i c a l s with phenoxy r a d i c a l s . 4. Neither ^ ^ c ^ n o r ^ r a t i o s show any s i g n i f i c a n t temper-ature dependence. (Figures 9, 17, 20 and 21) 5. Although formally there e x i s t s the p o s s i b i l i t y that the r a d i c a l s methyl, isopropyl and t e r t - b u t y l may disproportionate with the cyclohexadienyl r a d i c a l to form one of the cyclohexa-- 175 -dienes and the corresponding o l e f i n s (reactions (6e)), t h i s mode of r e a c t i v i t y i s never observed. A l l disproportionation yi e l d s the benzene product. Thermochemistry Crude estimates of the energy li b e r a t e d i n various termination reactions involving the cyclohexadienyl r a d i c a l have been made and the r e s u l t s tabulated i n Table XVIII. Table XVIIIg Thermochemistry of the Interactions of r a d i c a l s , R, with the cyclohexadienyl r a d i c a l . Reactants Heat l i b e r a t e d i n forming the products (kcal./mole) R + C 6 H 7 RH + C 6H 6 R ~ C 6 H 7 R(-H) + C 6H 8 CH' C 6 H 7 77 60 = 84 ~ Q C 2 H 5 + C 6 H 7 71 57 = 84 - Q - 3 32.5 C 3 H - 1 + C 6 H 7 67.5 56 sr. 8 4 - Q - 4 32 + C 6 H 7 64 52 a 84 - Q - 8 25 C 6 H 7 + C 6 H 7 44 32 = 80 - 2Q 44 In preparing t h i s compilation the following values have been adopted? Q = resonance energy of C^H^ ~ 24 kcal./mole; D(CH 3-C 6H ?) = 84 - Q where 84 ™ D(CH 3-CH(CH 3) 2) j 1 1 the decrements i n bond d i s s o c i a t i o n energies of the alkyi-=C^H7 compounds are the same as i n the series CHg-CHg, CHg-^H^, CH 3~CH(CH 3) 2 and CH 3-C(CH 3) 3, i e . 3, 4, and 8 kcal. ; L I A H ° , ( C 6 H 7 ) = 45 kcal./mole and A H £ ( C 6 H 8 - 1 , 4 ) (C 6H 8-1 93) = 26 kcal./mole 9 1 (the s l i g h t difference between the enthalpies of these two isomers w i l l be of no consequence here and has been neglected). From the table i t i s evident that the formation of the disproportionation products, RH and C^H^, i s energet i c a l l y more favourable than the recombin-ation products. The disproportionation to R(-H) + C^Hg l i b e r a t e s - 176 -only about one-half as much energy as benzene formation. Some question exists as to whether the observed d i s t r i b u t i o n of the products I and II i s c h a r a c t e r i s t i c of the i n i t i a l attack on the ring p o s i t i o n by the a l k y l r a d i c a l , or whether the f i n a l d i s t r i b u t i o n r e s u l t s from unimolecular isomerization of the hot recombination products before they can be c o l l i s i o n a l l y s t a b i l -ized. Under the experimental conditions, an excited CH^-C^Hy 8 9 molecule can be expected to undergo from 10 to 10 c o l l i s i o n s per second with surrounding molecules. Let us t e n t a t i v e l y assume that the hydrogen transfer rate constant can be expressed as /E - E Xs"1 , k(isomerization) - A j sec. V E / where fo r C^H^-CH^, s ~ 3 x 7 - 6 = 1 5 , using only skeletal v i b r -ations, and E = 60 kcal./mole from the c a l c u l a t i o n outlined above (Table XVIIl). For E Q we assume a minimum value of 30 kcal . based 1 0*7 on the observation of ter Borg et a_l. " that 31.5 kcal./mole are required to move the hydrogen atom around the ring of the cyclo-heptatriene molecule i n solution. Also Frey finds a b a r r i e r r o f 32.8 kcal./mole opposing hydrogen migration i n 2-methylpentadiene-1 3 8 1 3 9 1,3 and 31.2 kcal./mole i n cis-1-methy1-2-vi nyl cyclopropane. These examples have activated complexes with f i v e or six membered rings and may be markedly more favourable- both e n e r g e t i c a l l y and s t e r i c a l l y than the isomerization, 1^=^11, which must pass through a four centre complex. Using E Q = 30 kcal./mole, the isomeriza-- 5 '-I ti o n rate constant i s 6.1 x 10 ' A sec. , so i f A has the usual 13 -1 value of 10 sec. ", then the c o l l i s i o n rate i s about equal to the estimated isomerization rate. However, the high pressure pre-exponential factor should be equal to A, according to the _ 177 -Kassel theory that i s the basis f o r the expression used f o r the c a l c u l a t i o n of k(isoraerization). The pre-exponential factor 10 2 measured by t e r Borg i n the cycloheptatriene case i s only 10 sec." 1, therefore probably we overestimate the isomerization rate using the simple Kassel expression. The consistency of the r e s u l t s , and the f a c t that the unconjugated s t r u c t u r e , 1 1 , seems to be favoured over the very s l i g h t l y more stable structure, I, i n a l l cases, suggests that isomerization i s unimportant under the experimental conditions. The higher homologues should be less sensitive to hot molecule reactions than the methyl case. T T T Since the r a t i o k^/k^ i s i n s e n s i t i v e to the a l k y l r a d i c a l involved, the ortho-para combination r a t i o i s determined by pro-per t i e s of the cyclohexadienyl system. One possible explanation i s that the density of the unpaired electron i n the ring e f f e c t -i v e l y controls the d i s t r i b u t i o n of attack of the reacting r a d i -c a l . Electron spin resonance experiments have been performed on 58 the cyclohexadienyl r a d i c a l and on the phenoxy and o-methyl-96 phenoxy r a d i c a l s . In these studies the electron spin densities have been determined assuming that they are proportional to meas-ured coupling constants to the appropriate ring proton. These electron spin densities have been put on the following diagrams r e l a t i v e to the para spin density taken as unity. 0 >0.26 0 0.23 H H 0.63 T l 0.52 For methyl attacks Ft (ortho recomb.) R(para recomb.) Ratio expected s t a t i s t i c a l l y - 178 -In a l l cases, the para spin density substantially exceeds that measured at the ortho p o s i t i o n . S i m i l a r l y para recombination i s much favoured over recombination at the ortho s i t e . The a v a i l -a b i l i t y of the valence electron therefore o f f e r s a convenient explanation f o r the d i s t r i b u t i o n of the recombination products. Unfortunately, t h i s simple picture cannot be generally suf-f i c i e n t to explain observed Z \ values. In a recent publication, Mulcahy has reported no recombination of CH^ r a d i c a l s to the r i n g positions of the benzyl r a d i c a l , even though a s i g n i f i c a n t elec--tron density exists there; a l l termination i n t h i s system ap-parently occurs by recombination at the - C H 2 p o s i t i o n . 2 . T r a n s i t i o n state of the disproportionation reaction Bradley has pointed out that the conventional l i n e a r t r a n s i t i o n state t y p i c a l of the metathesis reactions would lead to f a r too great a loss of entropy and hence to pre-exponential factors that are too low to be applicable to the disproportion-ation reaction. Furthermore, we make the experimental observation that the power of an a l k y l r a d i c a l to remove a hydrogen atom from another r a d i c a l (eg. C^H^) i n a disproportionation reaction l i e s i n the order tert-Bu >^ iso-Pr Et ]>Me. This order i s the reverse of that experienced i n normal hydrogen abstraction reactions where strong C-H bond formers are expected to be the best hydrogen atom abstractors. Ideas about the t r a n s i t i o n states of the termination reac-1 3 6 5 3 txons have been advanced by Bradley and Rabinovitch, ' 9 Kerr and Trotman-Dickenson, 3^ and by Benson. 3 0 The disproportionation model of Kerr and Trotman-Dickenson 3 8 Kerr and Trotman-Dickenson's proposal i s the most detailed and 179 s p e c i f i c . These authors envisioned the disproportionation reac-t i o n as a r i s i n g from a tights, highly energized molecule formed by the recombination of two r a d i c a l s . The energy released i n t h i s recombination process i s available, i n t h e i r view, to cause the subsequent rearrangement and f i s s i o n of the hot molecule into the products of disproportionation. High entropy i n the trans-i t i o n state would be assured because the recombination step lead-ing to the energized molecule i s rate c o n t r o l l i n g , and i s govern-ed by a f r e e l y rotating t r a n s i t i o n state. The formulation of the Kerr-Trotman-Dickenson proposal con-siders hydrogen transfer to occur between reacting r a d i c a l s by way of a "four centre complex". For example, the ethyl r a d i c a l disproportionation would pass through the configuration A. M 2 C^ H: 2 W5 '• ^ G 4 H 1 0 G4^10 (successful recombination) H—CH, L>CH 3-C—CH 2 > C 2H 6 + C 2H 4 H 2 . (A) The application of t h i s idea to the present system leads us to configurations such as B H H R + -> RC 6H ? > •> RH + C 6H 6 H R : • > H > < ^ > (C) r e s u l t i n g i n the disproportionation products RH and benzene. The configuration C re s u l t i n g from para attack on the cyclohexa-- 180 -dienyl r a d i c a l cannot conveniently form a "four cetre" structure which embodies a reactive hydrogen atom. If we i n s i s t on four centre configurations leading to d i s -proportionation, we may suppose that the preference observed for para recombination arises from t h i s cause. Thus ortho attack leads to both disproportionation and combination, while para at-tack must e n t i r e l y end up as a combination product. Taken toget-her, t h i s would cause a net favouring of para recombination, i f the o v e r a l l p r o b a b i l i t y of i n t e r a c t i o n with an ortho s i t e i s c l o s e l y comparable to that of i n t e r a c t i o n with a para s i t e . Table T T T XIX shows that t h i s explanation of observed k^/k^ values i s en-t i r e l y compatible with the Z\(R, measurements. In t h i s table, the p r o b a b i l i t y of ortho attack co, and the p r o b a b i l i t y that ortho attack w i l l lead to a disproportionation, x, have been calculated to be consistent with the measured values of /\[R, C^Hj) and of k ^ / l c ^ f o r a p a r t i c u l a r a l k y l r a d i c a l , R. The p r o b a b i l i t y of para attack,TT, i s 1 - 201 A s l i g h t change i n the p r o b a b i l i t y of ortho ((D) or para (TT) attack i s s u f f i c i e n t to cause the observed product r a t i o s , accord-ing to the four centre complex proposal. I t i s not l i k e l y that t h i s can be the correct explanation of the unusually large extent of para recombination. The r e s u l t s of Mulcahy and W i l l i a m s , 1 3 5 l i s t e d i n Table XVII, show that the attack of the methyl r a d i c a l on the phenoxy and o-methylphenoxy r a d i c a l s also favours the para attack to a greater extent than that anticipated s t a t i s t i c a l l y . In t h i s example, there i s no mechanism where the ortho recombination can be lowered by four centre complexes. Some attack on the oxygen atom does occur, but probably t h i s i s fundamentally no d i f f e r e n t in,mechanism - 181 -than recombination to one of the r i n g positions. Furthermore, the v a r i a t i o n of the p r o b a b i l i t y , x, i s probably the reverse to that expected on the basis of the rather t i g h t , well defined tran-s i t i o n state of Kerr and Trotman-Dickenson. S t e r i c conditions around the reactive carbon atom i n the more complex a l k y l r a d i c a l s , p a r t i c u l a r l y the t e r t - b u t y l r a d i c a l should cause a decrease i n the pr o b a b i l i t y , x, rather than the increase seen i n Table XIX. Table XIX P r o b a b i l i t i e s f o r Ortho and Para Attack of A l k y l Radicals on the Cyclohexadienyl Radical calculated on the basis of the Kerr and Trotman-Dickenson Model R A(R, Cyl) X : d CH' 0.27 0.76 0.276 0.448 0.385 C 2 H 5 # 0.38 0.81 0.300 0.400 0.460 C 3 H 7 1 0.52 0.85 0.322 0.356 0.530 1.35 0.0 a 0.288 0.424 1.0 0.8 a 0.382 0.236 . 0.753 Assumed values f o r C^H^ where ^ c 1 m e a surement impossible CO i s the p r o b a b i l i t y that R* w i l l i n i t i a l l y associate with ortho p o s i t i o n . TT i s the p r o b a b i l i t y t h a t R* w i l l i n i t i a l l y a s s o c i a t e w i t h para p o s i t i o n * x i s the p r o b a b i l i t y t h a t o r t h o a t t a c k w i l l l e a d t o d i s p r o p o r -t i o n a t i o n . The evidence seems to favour separate, d i s t i n c t associations of the a l k y l r a d i c a l with either the hydrogen atoms of the cyclo-hexadienyl system or with one of the rin g positions leading to recombination. The favouring of the para p o s i t i o n must be due to more fundamental causes than the necessity of passing through - 182 -four centre configurations i n the disproportionation step. The Kerr-Trotman-Dickenson model of the t r a n s i t i o n state has 30 been c r i t i c i z e d by Benson, e s p e c i a l l y on the grounds that the expected pressure dependence of the Z\(Et, Et) values was not observed i n the experiments of Matsuoko et a l _ . 4 0 where the pres-sure was varied over a wide range. In a l a t e r publication, Dixon 41 et a l . suggested that the pressure dependence requirement may be avoided i f the d i s s i p a t i o n of the energy i s an intramolecular process i n nature, and that when the energy i s dispersed through-out the newly formed molecule, the system has a n e g l i g i b l e pro-b a b i l i t y of disproportionation. Only i n the f i r s t phase of the reaction, when the l i b e r a t e d energy i s concentrated i n the newly formed C-C bond i s there any l i k e l i h o o d of the tranfer of a hy-drogen atom taking place, according to t h i s modified picture. If t h i s i s the case, then one would i n t u i t i v e l y expect that the p r o b a b i l i t y of disproportionation would be dependent upon the amount of e x c i t i n g energy l i b e r a t e d into the system by recombin-ation. Also, the number of o s c i l l a t o r s available to accept t h i s energy leading to a successful'.', combination process should be important i n determing the Z \ r a t i o of a given system. However, we see from Tables XVII and XIX that the reverse i s true. The methyl-cyclohexadienyl system has the most e-xcitation energy and the fewest v i b r a t i o n a l o s c i l l a t o r s , yet displays the least tend-ency towards disproportionation. In view of the available e v i -dence, the model of Kerr and Trotman-Dickenson, whether or not modified to account f o r the lack of pressure e f f e c t s , seems un-tenable. Separate t r a n s i t i o n state models While Kerr and Trotman- . Dickenson suggest that a single t r a n s i t i o n state governs the - 183 -formation of both combination and disproportionation products, separate t r a n s i t i o n states have been proposed by Bradley and Rabinovitch, ' and by Benson. Neither of these proposals includes an excited molecule intermediate, and d i f f e r s fundamen-t a l l y from the Kerr-Trotman-Dickenson model. The Bradley-Rabinovitch disproportionation t r a n s i t i o n state, i s intermediate between a completely loose, f r e e l y rotating struc-ture (that only constrains the reactants from independent trans-l a t i o n ) and the r i g i d , low entropy structure t y p i c a l of hydrogen atom metathesis from molecules. These authors do not specify the precise structure of the t r a n s i t i o n states of either combination or disproportionation, but say only that the bonding i s "loose", 31 thus characterized by l i t t l e lossr of entropy. Rice has suggest-ed the presence of three centre hydrogen bonding i n the t r a n s i t i o n state complex of methyl r a d i c a l s and Bradley and Rabinovitch suggest such structures may be general. The terms "head-to-head" and "head-to-tail" are often used to describe the configuration of r a d i c a l i n t e r a c t i o n systems; these terms lose t h e i r meaning i f applied to the models of the t r a n s i t i o n state proposed by Bradley, Rabinovitch and Rice. 30 Benson agrees that loose binding i n separate t r a n s i t i o n states i s e s s e n t i a l to explain the observed disproportionation and combination rate constants, but does not believe that the very low v i b r a t i o n a l frequencies required to explain the high entropy of the t r a n s i t i o n state complexes are consistent with normal covalent bonding. On the basis of estimates of the ener-getics of charge transfer between a l k y l r a d i c a l s separated by distances t y p i c a l of t r a n s i t i o n states of recombining r a d i c a l s , he concludes that i o n i c forms may make s i g n i f i c a n t contribution - 184 -to the Valence Bond d e s c r i p t i o n of the bonding. Such forms, r e -p r e s e n t i n g n o n - d i r e c t i o n a l bonding, would be expected to lower f o r c e constants opposing r o c k i n g and bending motions of the r e a c -t a n t s w i t h r e s p e c t to one another e x p l a i n i n g the h i g h entropy of the t r a n s i t i o n s t a t e s . As the i o n i z a t i o n p o t e n t i a l of the p a r t i c i p a t i n g r a d i c a l i s the l a r g e s t energy term opposing charge t r a n s f e r , and i s q u i t e s e n s i t i v e to the s t r u c t u r e of the r a d i c a l , the r a t i o of d i s p r o -p o r t i o n a t i o n to combination may be i n f l u e n c e d p r o f o u n d l y by i o n -i z a t i o n p o t e n t i a l changes. Such i n f l u e n c e i s apparent i n the mutual t e r m i n a t i o n behavior of e t h y l , i s o - p r o p y l and t e r t - b u t y l r a d i c a l s which f a v o u r d i s p r o p o r t i o n a t i o n very much more f o r the h i g h e r homologues. The i o n i z a t i o n p o t e n t i a l s have a wide v a r i -a t i o n w i t h v a l u e s 5 6 ' 1 4 0 10.0, 8.8, 7.9 and 7.4 eV. r e s p e c t i v e l y f o r methyl, e t h y l , i s o p r o p y l and t e r t - b u t y l r a d i c a l s . The e l e c t -ron a f f i n i t i e s are p robably constant w i t h i n the range 1.1 + 0.1 e V . 1 4 1 The a p p l i c a t i o n of t h i s i d e a to the present r e s u l t s would seem a t t r a c t i v e . The c y c l o h e x a d i e n y l r a d i c a l has the low 5 9 i o n i z a t i o n p o t e n t i a l of 8.0 + 0.2 eV. i n the gas phase, and l i k e l y has an e l e c t r o n a f f i n i t y l a r g e r than i s common f o r a l k y l r a d i c a l s s i n c e the e x t r a e l e c t r o n can be accomodated i n a d e l o c -a l i z e d o r b i t a l . S i m i l a r l y , the p o l a r i z a b i l i t y of t h i s r a d i c a l would exceed t h a t of an a l k y l r a d i c a l , c o n t r i b u t i n g to the stab-i l i t y of charge t r a n s f e r s t r u c t u r e s . One would imagine t h a t i f t r a n s f e r of charge should occur at a l l , i t would occur i n t h i s system, and t h a t the change i n ZMR> ^6^7^ l s d u e ^° P r o ~ 30 nounced v a r i a t i o n i n the i o n i z a t i o n p o t e n t i a l of Ft*. Benson suggests t h a t the i o n i z a t i o n p o t e n t i a l v a r i a t i o n may be g e n e r a l l y important i n c o n t r o l l i n g the extent of d i s p r o p o r t i o n a t i o n . - 1 8 5 -On the other hand, application of the model of Bradley and 1 2 9 Rabinovitch leads to the suggestion that entropy differences between the products of disproportionation and of combination 5 4 determines t h e i r d i s t r i b u t i o n . Holroyd and K l e i n have success-f u l l y tested t h i s suggestion. They f i n d that /\ values of a wide variety of simple a l k y l r a d i c a l s are s a t i f a c t o r ^ l y correlated by the equation log k d/k c = 0.13l(2s°, - S°) - 5 . 4 7 This equation also predicted the pattern of i n t e r a c t i o n between 5 5 the a l l y l and the ethyl r a d i c a l s , a system su b s t a n t i a l l y d i f -ferent from those studied by Holroyd and K l e i n . We s h a l l see below that the trends i n the AR*, C^H^) r e s u l t s are also con-sistent with t h i s entropy c o r r e l a t i o n . Since the Holroyd-Klein r e l a t i o n and the i o n i z a t i o n potential patterns of a l k y l r a d i c a l s predict similar r e s u l t s , no test of the models of Benson and of Bradley and Rabinovitch i s conclusive within a l k y l systems. 3 2 Recently, Dorer and Rabinovitch have studied the chemically activated decomposition of alkenes according to the equation RCH2CH=CH* > R • + CH 2CHCH 2 for a l k y l r a d i c a l s R = methyl, ethyl and n-propyl. Their studies reveal that activated complexes fo r the recombination of a l k y l r a d i c a l s and a l l y l r a d i c a l s are s i g n i f i c a n t l y t i g h t e r than those common for a l k y l r a d i c a l mutual combination. Thus the s t e r i c - 3 - 1 factor f o r m e t h y l - a l l y l combination i s 5 x 1 0 whereas 1 0 i s common for a l k y l systems. Rabinovitch believes t h i s behavior i s 3 4 t y p i c a l of delocalized systems. Since i o n i z a t i o n potentials, electron a f f i n i t i e s and p o l a r i z a b i l i t i e s of delocalized r a d i c a l s are more favourable to charge tranfer than they are i n a l k y l - 1 8 6 -systems, we might have anticipated a high e f f i c i e n c y f o r a l k y l -a l l y l recombination. I f Rabinovitch's suggestion i s correct and low recombination e f f i c i e n c i e s p r e v a i l , i o n i c forms are apparently unimportant i n delocalized systems. The trend i n values of A(R. ^5^7) 1 S similar to that observed f o r a l k y l r a d i c a l s as R i s varied, (Figure 22); the same factors apparently control both systems. We can e a s i l y show that the present alkyl-cyclohexadienyl r e s u l t s are consistent with the entropy c o r r e l a t i o n within the bounds of certain assumptions. Holroyd and Klein's equation can be written 7.63 log/\(R, G 6H 7) = (S° H - SRC^H^ + 2 2 * 6 ( 3 0 ) i f SQ J_J = 6 4 . 3 4 cal./mole.deg. We assume the -C^H^ group con-6 6 t r i b u t e s a constant amount of entropy to the compound R-C^Hy. This i s supported by the constant values of k^/k^ 1, regardless of the attacking r a d i c a l . Any s t a t i s t i c a l correction for the m u l t i p l i c i t y of s i t e s f o r recombination i n the cyclohexadienyl r a d i c a l should be constant f o r our systems. Equation (30) be-comes . 7 . 6 3 0 log A(R-, C 6H 7) = OSj 0^ - 6 S ° C H 6 7 where 6logA(R, P 6H ?) = logA^y C 6H ?) - logA(CH 3, C 6H ?) RH ~ RH bCH 4 <S So = S° - S° °RG6H7 RC 6H ? CH 3C 6H ? Table XX shows that the values predicted for & S 0^ H compare favourably with those for &S^p rn, ^^Rpr^- a n c* ^^ RC H c a l c u l a t -55 6 5 ed from the appropriate standard values. This means that the values of ^ R , C^H-j) must l i e close to a l i n e having the gradient TABLE XX D i s t r i b u t i o n of the Products of Combination and Disproportionation of Al k y l Radicals with the Cyclohexadienyl Radical - Dependence on the Standard Entropy R 7.63 log A &; ° b Ss° RC^Hy RH Ss° R C 6 H 7 Ss° RC.Hj 6 Me Et iso-Pr tert-Bu 44.50 44.85 64.51 70.42 0.27 0.38 0.52 1.33 - 4.34 - 3.20 - 2.17 -.0.95 0 1.14 2.17. 5.29::: 0 10.35 . 20.01 :. .:. 25.95 0 9.21 17.84 20.63 0 9.28 16.83 19.73 0 11.70 17.00 21.16 0 9.73 16.45 a Entropy values taken from American Petroleum Institute Res. Proj. #44 (Carnegie Press, Pittsburgh, 1952) pp. 466 et seq. b T h e symbol Ss£x = s£x -- 188 -predicted by Holroyd and K l e i n , and to t h i s extent the r e s u l t s are consistent with t h e i r equation! Complete i d e n t i f i c a t i o n i s not possible as the lack of appropriate standard entropies pre-vents evaluation of the corresponding intercept. 107 Bates et aJL. have found the free energy of cyclohexadiene-1,4 to be only 0.58 kcal./mole greater than that of cyclohexadiene-1,3 at 95°, i n solution. An estimate of the enthalpy difference can be made as follows. Kistiakowsky found the heat of hydrogen-ation of cyclohexene to be 28.59 kcal./mole at 8 2 ° . 1 0 8 The heat of hydrogenation of cyclohexadiene-1,4 w i l l probably be at least twice as great, since some ring s t r a i n w i l l l i k e l y e x i s t i n the diene molecule. The heat of hydrogenation of cyclohexadiene-1,4 at 82° w i l l be at l e a s t 57.18 kcal./mole, which i s 1.81 kcal./mole larger than the si m i l a r value found f o r the heat of hydrogenation of cyclohexadiene-1,3. Assuming that the same difference exists at 95° i n solution, the entropy of cyclohexadiene-1,4 must exceed that of cyclohexadiene-1,3 by 3.34 eu./mole. Now Holroyd and Klein's r e l a t i o n s h i p for the disproportionation to combination r a t i o can be applied to the alternative combinations kj and k^ 1; we r e a d i l y derive l 0 ° - = ° ' 1 3 1 ^ c 1 " Sc> If we approximate S* 1 - S* as S(C 6Hg-l,4) - S(C 6Hg-l,3) = 3.34, T T T then lg c /k^ = 2.74. Correcting f o r the s t a t i s t i c a l factor of T T T two which favours the compound I, we f i n d that k£ /k_£ = 1.37, p r e c i s e l y what i s experimentally observed here, and very nearly equal to the r e s u l t s of E b e r h a r d t 1 0 (^"Vkji = 1.6) f o r the i n t e r -action of;the methyl and the cyclohexadienyl r a d i c a l i n l i q u i d cyclohexadiene-1,4, substantiating the use of the l i q u i d phase - 189 -107 r e s u l t s of Bates et a l . f o r the free energy estimate. A l l the aspects of disproportionation and combination of R* and O^Hj f a l l , into place i n the entropy c o r r e l a t i o n of Holroyd and K l e i n . Sim-i l a r l y , the favouring of the process 2 C^H^ >C^Hg-l,4 + C^H^ over the process 2 C ^ i j > C^Hg-1,3 + C^H^ i s reasonable i f S°(C 6Hg-l,4) i s greater than S°(C 6Hg-l,3). The Bradley-Rabinovitch model i s compatible with most experi-mental studies. The low primary isotope e f f e c t s measured for a l k y l disproportionation reactions at l i q u i d nitrogen temperatures Indicate very l i t t l e G-H bond breaking i n the t r a n s i t i o n state; kj_j/kj has been measured f o r disproportionation of ethyl-t (kj^/kj= 2 at 6 3 ° K . ) 4 4 and f o r isopropyl-t (kjj/lcj. = 1.7 at 77°K. ) 1 4 2 r a d i c a l s . Careful study of the temperature dependence of the values of A C 2 H ' , C 2 H £ ) 4 0 ' 4 1 , Z^CHg, C 2 H * ) 4 3 and Aiso-CgH^, Iso-CgH}) 4 2 has indicated that combination becomes more important than d i s -proportionation as temperature r i s e s ; a l l cases are compatible with an ac t i v a t i o n energy difference E c - E^ = 0.3 kcal./mole. Such a difference i s d i f f i c u l t to reconcile with a single t r a n s i -t i o n state model. The difference i n energy, E c - E^, i s small, however. Such a c t i v a t i o n b a r r i e r s provide a plausible explanation f o r the f a i l u r e of the cyclohexadienyl r a d i c a l to remove a hydrogen atom from a l k y l r a d i c a l s which otherwise donate them e f f i c i e n t l y . Removal of hydrogen from the a l k y l r a d i c a l without an a c t i v a t i o n energy means strong bonding must occur between the cyclohexa-dienyl r a d i c a l and the reactive hydrogen to compensate f o r the energy required to break the o r i g i n a l C-H bond. A s u f f i c i e n t l y strong C^H^-H bond may not be possible u n t i l a substantial degree - 190 -of l o c a l i z a t i o n of the free valence i n the r i n g occurs. This would impose a b a r r i e r not present f o r the recombination process where concerted bond formation-electron l o c a l i z a t i o n could more e a s i l y occur. Furthermore, we have seen that the free valence may be extensively located on the methylenic hydrogen atoms of the cyclohexadienyl r a d i c a l by the operation of exceptionally e f f i c i e n t hyperconjugation. This may render these hydrogens more reactive than those of the a l k y l r a d i c a l s and the transfer of hydrogen from the a l k y l system to the cyclohexadienyl system becomes most u n l i k e l y r e l a t i v e to the transfer to the a l k y l rad-i c a l from the cyclohexadienyl r a d i c a l . In the case where the cyclohexadienyl r a d i c a l donates the hydrogen atom, dispropor-tionation i s assisted by the release of additional, benzene resonance energy. In f a c t , when the reactive hydrogen atom i s weakened i n t h i s way the cyclohexadienyl r a d i c a l i s able to re-move i t ; the value AjC^Hj, = 0.45 i s s i m i l a r to A(iso-Pr«, C^H7) and indicates that any b a r r i e r s opposing d i s -proportionation are overcome. B. Metathesis between A l k y l Radicals and Cyclohexadiene-1,4 The experimental systems studied i n t h i s work have provided rate constants f o r the metathesis reaction between the methyl, isopropyl and t e r t - b u t y l r a d i c a l s and the cyclohexadiene-1,4 mol-ecule. Results f o r the ethyl r a d i c a l are available from previous work. 1 The following Arrhenius parameters r e s u l t f o r the reac-t i o n of a l k y l r a d i c a l s , R*, with the cyclohexadiene-1,4 molecule: R- + C 6H 8-1,4 ^ -> RH + C 6H ? k 6 = p.Z6exp(-E6/RT) - 191 -(kcal.) • - xi I - (cm./molecule sec.) R E 6 , 103p l o g l 0 A 6 log 1 Gk 6(60°) GH 3 5.52 + G.35 0.9 - 1 2 . 2 + 0 . 2 -15.8 C 2 H 5 5.7 + 0.1 0.6 - 1 2 . 5 +. 0 . 1 -16.3 i s o - C 3 H 7 6.4 + 1.1 2.4 - 1 2 . 0 + 0 . 6 -16.1 tert-C 4Hg 5.3 + 0.8 0.9 - 1 2 . 3 ± 0 . 5 -15.8 -16.0 + 0.2 Errors quoted for, I O Q ^ Q A ^ are those measured f o r l o g ^ A ^ / A ^ , and do not account f o r the error of measurement of A 2* which i s much greater, but i s not known. The great s i m i l a r i t y of the var-ious r a d i c a l s i n t h e i r r e a c t i v i t y towards the cyclohexadiene-1,4 molecule i s indicated i n the constancy of the ac t i v a t i o n energies and the fac t that, at 60°, the metathetical rate constant f o r each reactant r a d i c a l l i e s i n the range k^ = 1 0 " 1 6 * 0 — 0 # 2 cm?/molec,sec. Measurements of metathetical rate constants have been made for a single r a d i c a l over a series of substrates, but studies of a variety of r a d i c a l s with a common substrate are rare. Unfor-tunately, t h i s p a r t i c u l a r substrate, while necessary to provide the cyclohexadienyl r a d i c a l , i s a poor choice for comparative study of the attacking r a d i c a l s . The molecule i s so very highly activated towards metathesis that the i n d i v i d u a l r a d i c a l s are ndt selective i n attacking i t . The acti v a t i o n energies measured are a l l about the same, considering the experimental error. The s i m i l a r i t y of the s t e r i c factors i s expected at le a s t within the accuracy to which i t can be determined. 38 Trotman-Dickenson , i n a study of the rates of the reactions R* + RCHO > RH + RCO f o r a l k y l r a d i c a l s , R-, found a c t i v a t i o n energies that dropped from E(R* = CH^) = 7.6 to E(R* = tert-Bu) = - 192 -4.3. Rate constants at 182° were a l l about the same however. As the RCO-H bond i s weak, he suggested a comparison of these r e s u l t s to those measured f o r the systems R. + C^H^-CH^; how-ever, r e s u l t s for such a study are not, at present, available. The present r e s u l t s with C^Hg-1,4 are f o r a s i m i l a r l y activated system. The constancy of the measured ac t i v a t i o n energies con-t r a d i c t s the aldehyde system r e s u l t s , and would seem to be more expected. 1 1 OA A simple rel a t i o n s h i p ' has been used to correlate a c t i v a t i o n energies with bond energies, the so c a l l e d Evans-Polanyi relationships 2 AE a c t |=0CAn (3D 2 Z\H i s the difference i n the heats of two related reactions, f o r example, two of the hydrogen abstraction reactions studied here. / A E a c ^ i s the difference i n the a c t i v a t i o n energies of these 126 reactions, and OC i s a constant; t y p i c a l l y OC = 0.25. Applying the Evans-Polanyi r e l a t i o n s h i p to the reactions of t h i s work, where a l k y l r a d i c a l s R* remove hydrogen atoms from C^Hg-1,4, consider two r a d i c a l s R*and R!s R* + C 6H 8-1,4 E 6 > RH + C 6H ? R* .+ G H -1,4 E 6 > R'H + C 6 H ? In t h i s case, &i = D°(R-H) - D°(R'-H). For R = CHg, R'- = t e r t -butyl, Z§H = 104 - 91 = 13 k c a l . / m o l e .1 4 3 So we may have expect-ed that an a c t i v a t i o n energy difference Z\E = 13/4^-3 kcal./mole would have existed between the methyl and t e r t - b u t y l cases, on the basis of the Evans-Polanyi r e l a t i o n s h i p . C e r t a i n l y the re-sults are not consistent with such a conclusion. This i s r e a l l y quite surprising since, assuming 71 k c a l . f o r the C^Hy-H bond energy, the exothermicities of the reactions vary from 20 kcal/mole - 193 -(tert-butyl) to 33 kcal./mole (methyl). A 65 % increase i n the reaction exothermicities would normally be r e f l e c t e d i n the act-i v a t i o n energies. Inspection of Figure 23 shows that the maximum value of consistent with the errors of measurement at the 5 % p r o b a b i l i t y l e v e l i s OC = 0.017. The t e r t - b u t y l and methyl r e s u l t s strongly suggest that OC i s , i n f a c t , zero. 143 Johnson and Parr have recognized that the Evans-Polanyi relationship (31) i s not s u f f i c i e n t when ZAH i s large and have proposed a more complex c o r r e l a t i o n of a c t i v a t i o n energies to bond energies. They obtain considerable success i n predicting a c t i v a t i o n energies by treating the reacting system as having three separable energy terms; the a t t r a c t i o n between R^  and H , and between R 2 and H and repulsion between R^  and R2» R ..«H-•'R2 Applying t h i s c a l c u l a t i o n to the a l k y l radical-cyclohexa-diene-1,4 systems leads to a c t i v a t i o n energies that are too low (lying around 3.5 kcal./mole rather than the observed 5.5 kcal./mole) but which do show a d e f i n i t e increase of 40 % from the methyl case to the t e r t - b u t y l case. Figure 23 indicates that such a trend i s not present i n t h i s system; the lack of t h i s expected trend i n -dicates that the energetics of the t r a n s i t i o n state are deter-mined e n t i r e l y by the cyclohexadiene-1,4 species. The extensive resonance energy of the cyclohexadienyl r a d i c a l i s probably responsible for t h i s e f f e c t . At a c e r t a i n point dur-ing stretching of the C^Hy-H bond by the attacking a l k y l r a d i c a l a s u f f i c i e n t change i n the hybridization of the methylenic car-bon atom occurs to allow e f f e c t i v e conjugation to occur to the TT' system. A f t e r t h i s , very l i t t l e further increase i n energy i s - 194 -required. The a c t i v a t i o n energy i s therefore set by conditions at the methylenic carbon atom and Is not dependent on the attack-ing r a d i c a l . I t may very well be that constant a c t i v a t i o n ener-gies are c h a r a c t e r i s t i c of a l l hydrocarbon substrates with a c t i -vated hydrogen atoms. These ideas can be i l l u s t r a t e d with the simple diagram at the head of the next page. Curves A, B, and C represent attack of t e r t i a r y , secondary and primary r a d i c a l s respectively on a normal, non-activated secondary hydrogen atom, showing the usual dispersion of a c t i v a t i o n energies. Curve D i s the value of the cyclohexadienyl resonance energy as a function of the extension of the methylene C-H bond. Curves E, F and G represent the sums of A, B and C with D. In t h i s case, the dispersion of a c t i v a t i o n energies i s n e g l i g i b l e . - 195 -Figure 23. The Dependence of E^ on the Bond Diss o c i a t i o n Energy of the Attacking Radical. R . + (Q) 6 >RH+0 - . maximum slope within e r r o r s = 0.017 90 95 100 •D(R-H) (kcal.) - 196 -CONCLUSIONS Examination of the d i s t r i b u t i o n of the products of the i n t e r -actions of the cyclohexadienyl r a d i c a l with the a l k y l r a d i c a l s *CHj(CHg)^.. and with another cyclohexadienyl r a d i c a l has i n d i c a t -ed that generally these products are di s t r i b u t e d p r e f e r r e n t i a l l y amongst those of maximum entropy. The pattern of the r a t i o s A(CHj (CH 3) 3 - =_j, C^Hy) i s similar to that of the r a t i o s A C H . ( C H o K ., R*), f o r a l k y l r a d i c a l s , R's i n a l l systems the effectiveness of CHj(CH 3) 3_j as a hydrogen acceptor decreases as j increases. Such behavior i s predicted by the t r a n s i t i o n state proposals of Bradley and Rabinovitch and suggests that reacting r a d i c a l s pass through d i s t i n c t configurations f o r each mode of r e a c t i v i t y which are weakly bound, resembling the reac-tants. The p r o b a b i l i t y of attaining configurations leading to a p a r t i c u l a r product i s related to the entropy of the product when no energy b a r r i e r s e x i s t to influence the reaction course. Confirmation has been obtained f o r the suggestion that a c t i v a t i o n b a r r i e r s may exis t to oppose termination reactions that involve the disruption of conjugated TT electron systems. Thus the transfer of a hydrogen atom to the cyclohexadienyl rad-i c a l does not occur unless the donor r a d i c a l i s also a cyclo-hexadienyl r a d i c a l ; here the gain of the benzene resonance energy compensates the loss of the cyclohexadienyl resonance energy. The d i s t r i b u t i o n of the cyclohexadienes produced i n the reaction-of two Cyclohexadienyl r a d i c a l s favours the unconjugated, 1,4-diene. As evidence suggests that t h i s has the greater standard entropy, t h i s r e s u l t accords with the maximum entropy p r i n c i p l e . The magnitude of the cyclohexadienyl resonance energy has - 197 -been estimated as 24 kcal./mole, subs t a n t i a l l y larger than the value of 15.5 kcal./mole measured elsewhere f o r the straight chain pentadienyl resonance energy. A simple explanation f o r such a discrepancy assumes a pronounced, hyperconjugative i n t e r -action i n the c y c l i c system across the methylenic carbon atom. Since the benzene resonance energy (^37 kcal./mole) i s s t i l l greater than the cyclohexadienyl resonance energy, the expulsion of a hydrogen atom from the cyclohexadienyl r a d i c a l i s more favourable, energetically, than i t i s from a l k y l r a d i c a l s . The considerable resonance energy of the cyclohexadienyl system makes i t sluggish i n reactions with molecules, but makes vigorous those reactions which generate i t . This combination of circumstances provides that the free r a d i c a l homopolymerization of cyclohexadiene-1,3 s h a l l be severely retarded by degradative chain transfer. The rate of generation of the cyclohexadienyl r a d i c a l by metathesis from cyclohexadiene-1,3 i s slower than by metathesis from cyclohexadiene-1,4; t h i s supports the contention that the free energy of the 1,4-diene exceeds that of the 1„3-diene i n the gas phase, just as i t has been found to do i n l i q u i d amyl alcohol. The rate of hydrogen abstraction by various a l k y l r a d i c a l s from cyclohexadiene-1,4 i s independent of the attacking r a d i c a l , although a considerable va r i a t i o n i n the reaction heat i s experienced. Features of the cyclohexadiene-1,4 molecule are responsible for control of the reaction rate; the release of the resonance energy probably sets a l i m i t on the energy re-quired to at t a i n the t r a n s i t i o n state, about 5.5 kcal./mole. The reactions of the methyl r a d i c a l with azomethane have been studied. Addition to the nitrogen-nitrogen double bond - 198 -occurs with a low acti v a t i o n energy (5.9 kcal.) and a low s t e r i c factor (0.7 x IO" 4), but at an o v e r a l l rate comparable to the rate of attack of methyl r a d i c a l s upon mono-olefins. The a c t i v -ation energy for the abstraction of a hydrogen atom from azo-methane, when compared to that measured for abstraction from acetone and from 2-butene, suggests that the resonance energy varies i n the order CH2-C==CH2^> GH2~C=0^> CH2-N=N- for each of the a l l y l resonance s t a b i l i z e d residues. - 199 -BIBLIOGRAPHY 1. A. C. R. Brown, Ph.D. Thesis, The University of B r i t i s h Columbia, 1962. 2. E. L. E l i e l , S. Meyerson, Z. Welvard and S. H. Wilen, J . Am. Chem. S o c , 82, 2963 (i960). 3. G. E. Corbett and G. H. Williams, J . Chem. S o c , 3437 (1964). 4. J . W. T. Spinks and R. J . Woods, "An^Introduction to Rad-i a t i o n Chemistry", John Wiley and Sons, Inc., New York, 1964, Chapter 10, pp. 365-384.-5. T. Gaumann, Helv. Chim. Acta, 46, 2873 (1963). 6. M. K. Eberhardt, J . Phys. Chem., 67, 2856 (1963). 7. J . K. S t i t l e , R. L. Sung and J . V. Kooi, J . Org. Chem., 30, 3116 (1965). 8. J . K. Garland and F. S. Rowland, Sonderdruck aus Radio-chimica Acta, 4, 115 (1965). 9. E.,A. Cherniak, E. Collinson and F. S. Dainton, Trans. Far. S o c , 60, 1408 (1964). 10. M. K. Eberhardt, G. W. K l e i n and T. G. Krivak, J . Am. Chem. S o c , 87, 696 (1965). 11. S. W. Benson, J . Chem. Educ., 42, 502 (1965). 12. H. Fisher, J . Chem. Phys., 37, 1094 (1962). 13. K. W. Egger and S. W. Benson, J . Am. Chem. S o c , 88, 2*H (1966). 14. P. S. Nangia and S. W. Benson, J . Am. Chem. S o c , 86, 2773 (1964). 15. L. M. Dorfman and Z. D. Sheldon, J . Chem. Phys., 17, 511 (1949). 16. K. 0. Kutschke, M. H. J. Wijnen and E. W. R. Steacie, J . Am. Chem. S o c , 74, 714 (1952). 17. R. K. Brinton and E. W. R. Steacie, Can. J . Chem., 33, 1840 (1955). 18. D. S. Weir, J . Am. Chem. S o c , 83, 2629 (1961). 19. K. 0. Kutschke and A. Shepp, J . Chem. Phys., 26, 1020 (1957). 20. D. G. L. James and E. W. R. Steacie, Proc. Roy. S o c , A, 244, 289 (1958). - 200 -21. G. B. Kistiakowsky and E. K. Roberts, J . Chem. Phys., 21, 1637 (1953). 22. J . N. Butler and G. B. Kistiakowsky, J . Am. Chem. S o c , 82, 759 (1960). 23. R. E. Harrington, B. S. Rabinovitch and H. M. Frey, J . Chem. Phys., 33, 1271 (i960). 24. E. K. C. Lee and F. S. Rowland, J . Chem. Phys., 36, 554 (1962). 25. S. W. Benson, J . Chem. Phys., 20, 1064 (1952). 26. E. L. Metcalfe and A. F. Trotman-Dickenson, J . Chem. S o c , 4620 (1962). 27. S. W. Benson, "Foundations of Chemical K i n e t i c s " , McGraw-H i l l Co., Inc., New York, 1961, p. 281. 28. T. S. Ree, T. Ree, H. Eyring, and T. Fueno, J . Chem. Phys., 36, 281 "(1962). 29. S. Glasstone, K. L a i d l e r and H. Eyring, "The Theory of Rate Processes", McGraw-Hill Co., Inc., New York, 1941, p. 260. 30. S. W. Benson, Advances i n Photochemistry 3., 1, p. 12. (1964); 31. 0. K. Rice, J . Phys. Chem., 65, 1588 (1961). 32. F. H. Dorer and B. S. Rabinovitch, J . Phys. Chem., 69, 1952 (1965). 33. B. S. Rabinovitch and D. W. Setser, Adv. Photochemistry, 3, 1 (1964). 34. Prof. B. S. Rabinovitch, private communication. 35. M. H. J . Wijnen and E. W. R. Steacie, Can. J . Chem., 29, 1092 (1951). 36. K. W. Kraus and J . G. Calvert, J . Am. Chem. S o c , 79, 5921 (1957). 37. W. E. Morganroth and J . G. Calvert, Diss. Abs., 26, 1369 (1965). 38. A. F. Trotman-Dickenson and J . A. Kerr, "Progress i n Chem-i c a l K i n e t i c s " , Pergamon Press, Ner York, 1961, vol. 1, p.107. 39. J . A. Kerr and A. F. Trotman-Dickenson, J . Chem. S o c , 1602 (1960). 40. M. Matsuoka, P. S. Dixon, A. P. Stefani and M. Szwarc, Proc. Chem. S o c , 302 (1962). 41. P. S. Dixon, A. P. Stefani and M. Szwarc, J . Am. Chem. S o c , 85, 2551 (1963). - 201 -42. R. K l e i n , M. D. Scheer, and R. Kelley, J . Phys. Chem., 68, 598 (1964). 43. P. S. Dixon, A. P. Stefani and M. Szwarc, J . Am. Chem. S o c , 85, 3344 (1963). 44. K. W. Watkins and H. C. Moser, J. Phys. Chem., 69, 1040 (1965). 45. R. D. G i l e s and E. Whittle, Trans. Far. S o c , 61, 1425 (1965). 46. G. 0. Pritchard, M. Venugopalan and T. F. Graham, J . Phys. Chem., 68, 1786 (1964). 47. C. A. Hel l e r and A. S. Gordon, J . Chem. Phys., 36, 2648 (1961). 48. P. J . Boddy and E. W. R. Steacie, Can. J . Chem., 38, 1576 (1960). 49. D. G. L. James and E. W. R. Steacie, Proc. Roy. S o c , A, 245, . 470 (1958). 50. M. Salomon, Can. J . Chem., 42, 610 (1964). 51. 0. Sinonoglu and K. S. P i t z e r , J . Chem. Phys., 30, 422 (1958). 52. H. S. Johnston, D. R. Herschback and D. Rapp, J . Chem. Phys., , 31, 1652 (1959). 53. J . N. Bradley and B. S. Rabinovitch, J . Chem. Phys., 36, 3498 (1962). 54. R. A. Holroyd and G. W. K l e i n , J . Phys. Chem., 67, 2273 (1963). 55. D. G. L. James and G. E. Troughton, Trans. Farad. S o c , 62, 145 (1966). 56. "Mass Spectrometry", C. A. McDowell ( E d i t o r ) , McGraw-Hill Co., Inc., New York, 1963, p. 501. 57. J . A. G. Dominguez, J . A. Kerr, and A. F. Tronman-Dickenson, J. Chem. S o c , 3357 (1962). 58. R. W. Fessenden and R. H. Schuler, J . Chem. Phys. 38, 773 (1963). • "~ 59. A. G. Harrison, P. Haynes, S. McLean, and F. Meyer, J . Am. Chem. S o c , 87, 5099 (1965). 60. M. Kasha, J. Opt. Soc. Amer.,. 38, 929, (1948). 61. D. J . Le Roy, Can. J . Research, B., 28, 492 (1950). - 202 -62. L. M. Jackman, "Applications of Nuclear Magnetic Resonance Spectroscopy i n Organic Chemistry", Pergamon Press, New York, 1962, p. 86. 63. B. J . Monostori and A. Weber, J . Mol. S p e c , 12, 129 (1964). 64. H. D. Stidham, Spectrochimica Acta, 21, 23 (1965). 65. M. H. Jones and E. W. R. Steacie, J . Chem. Phys., 21, 1018 (1953) . 66. S. Toby, J. Am. Chem. S o c , 82, 3822 (I960). 67. R. E. Rebbert and P. Ausloos, J . Phys. Chem., 67, 1925 (1963). 68. P. Ausloos and E. W. R. Steacie, Can. J . Chem., 32, 593 (1954) . 69. S.Toby and B. H. Weiss, J . Phys. Chem., 66, 2681 (1962). 70. J . A. Kerr and J . G. Calvert, J . Phys. Chem., 69, 1022 (1965). 71. R. E. Rebbert and P. Ausloos, J . Phys. Chem., 66, 2253 (1962). 72. W. Sleppy and J . G. Calvert, J . Am. Chem. S o c , 81, 769 (1959). 73. J . C. J . Thynne, private communication. 74. L. Herk, M. Feld, M. Szwarc, J . Am. Chem. S o c , 83, 2998 (1961). 75. S. Toby and B. H. Weiss, J . Phys. Chem., 68, 2492 (1964). 76. H. Cerfontaine and K. 0. Kutschke, Can. J . Chem., 36, 344 (1958). 77. 0. G. L. James and E. W. R. Steacie, Proc. Roy. S o c , A, 244 297 (1957). 78. L. Mandelcorn and E. W. R. Steacie, Can. J . Chem., 32, 474 (1954). 79. S. Toby and K. 0. Kutschke, Can. J . Chem., 37, 672 (1959). 80. P. Gray, A. Jones, and J . C. J . Thynne, Trans. Farad. S o c , 61, 474 (1965). 81. A. F. Trotman-Dickenson and E. W. R. Steacie, J . Chem. Phys., 19, 329 (1951). 82. J . H. Knox and A. F. Trotman-Dickenson, Chemical Society (London) Special Publications No. 9, pp. 35-45, (1957). - 203 -83. G. Dominguez and A. F. Trotman-Dickenson, J . Chem. S o c , 940 (1962). 84. K. Yang, J . Am. Chem. S o c , 84, 3795 (1962). 85. W. F. Falconer, B. S. Rabinovitch and R. J . Cvetanovic, J. Chem. Phys., 39, 40 (1963). 86. K. "Yang, J . Am. Chem. S o c , 84, 719 (1962). 87. M. Takahasi and R. J . Cvetanovic, Can. J . Chem, 40, 1037 (1962) . 88. L. H. Slaugh, J . Am. Chem. S o c , 87, 1522 (1965). 89. R. J . E l l i s and H. M. Frey, J . Chem. S o c , (1966), i n press. 90. W. Huckel and R. Schwen, Ber., 89, 150 (1956). 91. J . D. Cox, Tetrahedron, 19, 1175 (1963). 92. American Petroleum Inst., Research Pro j . #44, Carnegie Press, Pittsburgh, Pa., 1953. 93. M. F. R. Mulcahy, D. J . Williams, and J . R. Wilmhurst, Aust. J . Chem., 17, 1329 (1964). 94. S. W. Benson, "Foundations of Chemical K i n e t i c s " , McGraw-H i l l Co., Inc., New York, I960, pp. 277, ^ 80. 95. R. W. Fessenden and R. H. Schuler, J . Chem. Phys., 39, 2147 (1963) . 96. T. J . Stone and W. A. Waters, Proc. Chem. S o c , 253 (1962). 97. G. Vincow and P. M. Johnson, J . Ghem. Phys., 39, 1143 (1963). 98. D. M. Golden, K. W. Egger, and S. W. Benson, J . Am. Chem. S o c , 86, 5420 (1964). 99. I. Hanazaki and S. Nagakura, B u l l . Chem. Soc. Japan, 38, 1298 (1965). 100. G. E. Troughton, Ph.D.. Thesis, The University of B r i t i s h Columbia, 1965. 101. M. K, Eberhardt, Tetrahedron, 21, 1^83 (1965). 102. J..E. Hofman, P. A. Argabright, A. Schriesheim, Tetra-hedron l e t t . No. 17, 1005 (1964). 103. C. A. He l l e r and A. S. Gordon, J . Phys. Chem., 60, 1315 (1956). 104. C. A. He l l e r and A. S. Gordon, J . Phys. Chem., 62, 709 (1958). • - 204 -105. S. G. Whiteway and C. R. Masson, J . Am. Chem. S o c , 74, 4731 (1955). 106. R. H. Riem and K. 0. Kutschke, Can. J . Chem., 38, 2332 (1960). "~" 107. R. B. Bates,.R. H. Carnigham and C. E. Staples, J. Am. Chem. S o c , 85, 3030 (1963). 108. G. B . Kistiakowsky, J . R. Ruhoff, H. A. Smith and W. E. Vaughan, J . Am..Chem. S o c , 58, 146 (1936). 109. G. B. Kistiakowsky, J . R. Ruhoff, H. A. Smith and W. E. Vaughan, J . Am. Chem. S o c , 58, 137 {1936). 110. J . Gresser, A. Rajenbach and M. Szwarc, J . Am. Chem. S o c , 83, 3005 (1961). 111. A. C. R. Brown and D. G. L. James, P r o c Chem. S o c , 81 (1962). 112. E. S. Ferdinandi, W. P. Garby and D. G. L. James, Can. J . Chem., 42, 2568 (1964). 113. S. F. Naumova and L. G. Tsykalo, Vysokomolekul. Soedin, 3, 1031 (1961). 114. P. S. Shantarovitch and E. A. Shlyapnikova, Vysokomolekul. Soedin, 3, 1364 (1961). 115. B. V. Erofeev, S. F. Naumova and L. G. Tsykalo, Dokl. Akad. Nauk, SSSR, 147, 106 (1962). 116. G. Lefebvre and F. Dawans, J . Polymer S c i . , A, 2, 3277 (1964) . 117. (a) B. Franzus and J . H. Surridge, J . Org. Chem., 27, 1951 (1962). (b) B. Franzus, J . Org, Chem., 28, 2954 (l"9~63). 118. C. S. Hsia Chen and E. F. Hosterman, J . Org. Chem., 28, 1585 (1963). 119. A. A. Oswald, K. Griesbaum, and B. E. Hudson, J r . , J . Org. Chem., 28, 1262 (1963). 120. A. A. Oswald, H. E. Hudson, J r . , G. Rodgers, and F. Noel, J. Org. Chem., 27, 2439 (1962). 121. J . F. Brown, J r . , and D. M. White, J . Am. Chem. S o c , 82, 5671 (1960). 122. D . G . L. James and R. D . Suart, J. Phys. Chem., 69, 2362 (1965) . 123. D . L. Jones and D. G. L. James, unpublished r e s u l t s . - 205 -124. Ft. B. Bates, Ft. H. Carnigham and C. E. Staples, J . Am. Chem. S o c , 85, 3032 (1963). 125. M . G. Evans and M. Polanyi, Trans. Farad. Soc. 34, 11 (1938). 126. N. N. Semenov, "Some Problems i n Chemical K i n e t i c s " , Princeton University Press, Princeton, New Jersey, 1958 Volume 1, p. 29. 127. S. W. Benson and D. B. Hartley, J. Chem. Phys., 39, 132 . (1963). 128. J . B. Levy and B. K. W. Copeland, J . Am. Chem. S o c , 82, 5314 (1960). 129. A. U. Blackham and N. L. Eatough, J. Am. Chem. S o c , 84, 2922 (1962). 130. S. S. Thomas and J . G. Calvert, J . Am. Chem. S o c , 84, 4207 (1962) . 131. Ft. N. B i r r e l l and A. F. Trotman-Dickenson, J. Chem. S o c , 3357 (1962). 132. P. J . Boddy and J . C. Robb, P r o c Roy. S o c , A, 249, 518 (1959). 133. J . A. Kerr and J . G. Calvert, J . Am. Chem. S o c , 83, 3391 (1961). 134. J . C. J . Thynne, P r o c Chem. S o c , 18 (1961). 135. M . F. R. Mulcahy and D. J . Williams, Nature, 199, 761(1963) 136. J . N. Bradley, J . Chem. Phys., 35, 748 (1961). 137. A. P. ter Borg, H. Kloosterzid, and N. Van Moors, Rec. Trav. Chim., 82, 717 (1963). 138. H. M. Frey, J . Chem. S o c , 4770 (1965). 139. R. J . E l l i s and H. M. Frey, J . Chem. S o c , 5578 (1964). 140. A. Streitwieser, J r . , Prog. Phys. Org. Chem., 1_, 1 (1963). 141. J . J . Kaufman, J. Phys. Chem., 68, 3155 (1964). 142. H. B. Yun and H. C. Moser, J. Phys. Chem., 69, 1059 (1965). 143. H. S. Johnston and C. Parr, J . Am. Chem. S o c , 85, 2544 (1963) . ~ 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0062136/manifest

Comment

Related Items