A KINETIC STUDY OF THE ETHYL RADICAL WITH ALLYL COMPOUNDS by GARY ELLIS TROUGHTON B. Sc. (Hon. Chem.), The University of British Columbia, 1962 A thesis submitted iu partial fulfilment of the requirements for the degree of DOCTOR OF PHILOSOPHY in the Department of CHEMISTRY We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA September, 1965 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f t h e r e q u i r e m e n t s f o r an advanced deg ree a t t h e U n i v e r s i t y o f B r i t i s h C o l u m b i a , I a g r e e t h a t t h e 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 r e f e r e n c e and s t u d y . I f u r t h e r a g r e e t h a t p e r -m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y p u r p o s e s may be g r a n t e d by t h e Head o f my Depar tment o r by h i s r e p r e s e n t a t i v e s , , I t i s u n d e r s t o o d t h a t c o p y i n g o r p u b l i -c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l no t be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . Depar tment o f [Jf^fff/j The U n i v e r s i t y o f B r i t i s h C o l u m b i a V a n c o u v e r ^ , Canada % f lUs: Date The Uni v e r s i t y of 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 Bo Sc. (Hons.), The University of B r i t i s h Columbia TUESDAY, NOVEMBER 16. 1965 AT 3:30 P„M„ IN ROOM 261, CHEMISTRY BUILDING COMMITTEE IN CHARGE Chairman: I, McT. Cowan External Examiner: Peter Gray . Department of Physical Chemistry Leeds University Leeds 2» England Research Supervisor: D.G.L. James of GARY ELLIS TROUGHTON 1962 F. Dalby B. R. James C A . McDowell D.E. McGreer Ro E. Pincock G. B. Porter A KINETIC STUDY OF THE ETHYL RADICAL WITH ALLYL COMPOUNDS ABSTRACT The patterns of i n t e r a c t i o n of the e t h y l r a d i c a l with a v a r i e t y of a l l y l . compounds reveals a uniformity towards the addition whereas considerable differences i n r e a c t i v i t y towards the metathesis and dismutation reactions are expected and found, The metathesis re-actions of some a l l y l compounds were too slow to be measured, and the l a b e l l i n g of the functional groups on the a l l y l . compound with deuterium was necessary i n order t o measure the s p e c i f i c metathesis reaction. In the case of a l l y l alcohol. s primary isotope e f f e c t s i n metathesis of the C-H and 0-H bonds could be measured, and were in good agreement with those predicted by the difference i n zero point energies of the appropriate bonds. Considerable d i v e r s i t y i s expected and found among the energies of a c t i v a t i o n for the dismutation of the adduct r a d i c a l s formed i n the ad d i t i o n and metathesis reactions, The energies of a c t i v a t i o n for the dismutation reaction of the adduct r a d i c a l range from 15 *"o 27 Kcal/mole and are re l a t e d to the corresponding enthalpies of dismutation, The dismutation reactions of the addition adduct r a d i c a l was postulated by Gaylord and.Eirich i n 1952 as an important e f f e c t i v e chain transfer process for a l l y l . esters and i n 1964;, Gaylord;, Katz and Mark presented the f i r s t d i r e c t evidence i n the l i q u i d phase for the radical, displacement mechanism of e f f e c t i v e chain transfer for the 3-buten-2-yl propionate system, Evidence i s presented i n this i n v e s t i g a t i o n i n the gas phase demonstrating the importance of dismutation re-actions for several s t r u c t u r a l types of a l l y l compounds. The dismutation reaction has the a t t r a c t i v e feature i n that this reaction can generate free r a d i c a l s at r e l a t i v e l y low temperatures 3 and therefore represents an excellent method for measuring the disproportionation-combination reaction of various free r a d i c a l s . In t h i s ways the a l l y l . r a d i c a l was generated, and i t s i n t e r a c t i o n with the e t h y l r a d i c a l was found, CH3CH2CH2CHCH2CH2CH=CH2 —>CH2=CHCH2CH2CH3 + (CH2-r^ CH--CH2) 5 (CH2—-CH-_^CH2J + CH3CH2' —> CH3CH-CH2 + CH2CH2 11% ( C H 2 — C H ^ _ r G H 2 ) ° + CH 3CH 2 —> CH2=C=CH2 + CH3CH3 4% (CH2—CH=^-=K]H2)° + CH3CH2 —> CH3CH2CH2GH=CH2 85% The rate constants obtained for the i n t e r a c t i o n of the ethy l r a d i c a l with a given a l l y l compound i n the gas phase are extremely h e l p f u l i n the i n t e r p r e t a t i o n of the ki n e t i c s of the l i q u i d phase polymerization of that monomer. The r a t i o of these rate constants i n the gas phase can be rela t e d to the degree of polymerization i n the l i q u i d phase by means of a modified Hammett equation, and a l i n e a r r e l a t i o n s h i p between these quantities was observed i n this study. GRADUATE STUDIES F i e l d of Study: Gas Kine t i c s Topics i n Physical Chemistry Seminar i n Chemistry Quantum Chemistry Topics i n Chemical Physics Topics i n Inorganic Chemistry Spectroscopy and Molecular Structure Chemical Kinetics Topics i n Organic Chemistry Physical Organic Chemistry Organic Reaction Mechanisms J. A. R. Coope A. Bree N. B a r t l e t t J. A, R. Coope C. A. McDowell B. A. Dunell W. R. Cullen N, B a r t l e t t J . T, Kwon H. C„ Clark C. Reid L . W. Reeves E. J. Wells K. B. Harvey D. G. L , James E. A. Ogryzlo A. I. Scott F. McCapra J. P. Kutney D. E..McGreer R. Stewart R, Pincock R e l a t e d S t u d i e s : Modern P h y s i c s Computer S c i e n c e Mo Bloom B» Henderson PUBLICATIONS Do Go L. James and G. E. Troughton "RADICAL DISPLACEMENT IN ALLYL ESTER POLYMERIZATION" J. Polymer S c i , ; Part A.3, 75 (1965). Do G . L. James and G. E. Troughton "DISPROPORTIONATION REACTIONS OF THE ALLYL RADICAL" Chem. Coram,.No. 5, 94 (1965). i i A B S T R A C T The r e a c t i o n s of the e t h y l r a d i c a l w i t h a s e r i e s of a l l y l compounds w e r e e x a m i n e d e x p e r i m e n t a l l y i n the gas phase, and the r e s u l t s w e r e s e e n to c o n f o r m to a g e n e r a l m e c h a n i s m . S p e c i f i c a l l y , the e t h y l r a d i c a l c a n abstract a h y d r o g e n a t o m f r o m the a l l y l compound, o r add to the o l e f i n i c l i n k a g e of the a l l y l compound. D i s m u t a t i o n r e a c t i o n s of the p r o d u c t r a d i c a l s f o r m e d i n the a d d i t i o n o r a b s t r a c t i o n r e a c t i o n s w e r e a l s o o b s e r v e d i n t h i s study. The m e t a t h e s i s r e a c t i o n s of s o me a l l y l compounds w e r e too s l o w to be m e a s u r e d u n l e s s the f u n c t i o n a l group was d e u t e r a t e d . When t h i s was done i n the c a s e of a l l y l a l c o h o l and a l l y l f o r m a t e , p r i m a r y i s o t o p e e f f e c t s i n m e t a t h e s i s c o u l d be m e a s u r e d , and w e r e i n good a g r e e m e n t w i t h the v a l u e s p r e d i c t e d on the b a s i s of z e r o p o i n t e n e r g y d i f f e r e n c e s . The a c t i v a t i o n e n e r g i e s of the d i s m u t a t i o n r e a c t i o n s of the adduct r a d i c a l s r a n g e d f r o m 15 to 27 K c a l / m o l e and w e r e found to depend on the e s t i m a t e d e n t h a l p i e s of t h i s r e a c t i o n . I n c e r t a i n c a s e s , a c h o i c e between two d i f f e r e n t r e a c t i o n r o u t e s of the d i s m u t a t i o n r e a c t i o n c o u l d be made by c o m p a r i n g the a c t i v a t i o n e n e r g y w i t h the e s t i m a t e d enthalpy change. The d i s m u t a t i o n r e a c t i o n o f f e r s a c o n v e n i e n t m e t h o d f o r the g e n e r a t i o n of f r e e r a d i c a l s a t low t e m p e r a t u r e s . In t h i s way, the a l l y l and ethoxy r a d i c a l s w e r e gen e r a t e d , and t h e i r d i s p r o p o r t i o n a t i o n - c o m b i n a t i o n r a t i o s w i t h the e t h y l r a d i c a l w e r e m e a s u r e d . The k i n e t i c s of i n t e r a c t i o n of the e t h y l r a d i c a l w i t h a n a l l y l compound i n the gas phase c a n be c o m p a r e d w i t h the p o l y m e r i z a t i o n of that m o n o m e r i n the l i q u i d phase. S p e c i f i c a l l y the r a t i o of the a d d i t i o n r a t e c o n s t a n t to the m e t a t h e s i s r a t e c o n s t a n t i n the gas p h a s e c a n be r e l a t e d to the d e g r e e of p o l y m e r i z a t i o n and a l i n e a r r e l a t i o n s h i p betwee these q u a n t i t i e s was found i n t h i s study. i i i ACKNOWLEDGEMENTS T wish to thank Dr. D. G. L . James for his guidance and advice throughout the course of my study. I wish to thank Dr. F. Lossing for the mass spectrometer results pertaining to ally 1-1, l - d 2 acetate and allyl acetate-d 3 and I also wish to thank Dr. D. Frost for the mass spectrometer results pertaining to allyl-1, l - d ^ alcohol and allyl alcohol-d. I am very grateful to the National Research Council of Canada for scholarships covering the periods June 1962 to September 1963 and September 1964 to September 1965, and to Canadian Industries Limited for a scholarship covering the period September 1963 to September 1964. iv TABLE OF CONTENTS Page INTRODUCTION • . . . 1 A. Scope of this Investigation 1 B. A General Mechanism ." . . 2 C. Allyl Radical . 4 D. Allyl Polymerization 6 EXPERIMENTAL METHODS 9, A. Apparatus . . . . . . . . i . . . . . . . 9 B. Analysis 9 C. Reagents . 11 D. Photolysis of Diethyl Ketone in the Presence of an Allyl Compound. 12 RESULTS 14 DISCUSSION . . 44 A. Reaction of the Ethyl Radical with Allyl Compounds in General . . 44 B. The Reactions of the Ethyl Radical with Hexadiene-1,5 49 C. The Reactions of the Ethyl Radical with Allyl Ethyl Ether and Allyl Ether 59 D. The Reactions of the Ethyl Radical with Allyl Acetate and its Deuterated Analogues 71 E. The Reactions of the Ethyl Radical with Allyl Alcohol and its Deuterated Analogues 81 F. Reactions of the Ethyl Radical with Allyl Propionate 93 G. Reactions of the Ethyl Radical with Allyl Formate and Allyl Formate-d 100 H. Kinetic Scheme of the Diallyl Carbonate System . . 107 7 . ' v TABLE OF CONTENTS (Con't) Page CONCLUSIONS .. . . 108 APPENDIX A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . H I APPENDIX B . . . . . . • ' 112 BIBLIOGRAPHY • • • • 1 1 5 vi LIST OF i?IGfcK£S Flguro . . . ' • • 1. Addition and eta thesis feetweea th® ethyl radical and . . . 5« 2. ' Dismataticm of th® addition fiddaet radical of hoxadteno -1,8 58 3. Addison cf -the eihyl radical to the allyl others • . • . 68 4. Metathesis between tho eihy! radical and the nllyi ethers 60 5 . Dilatation of the addition and jsietatheats adduct radicals of th® allyl iftthfifs . . . . . . . .,»..<•.•>•*•««.. «..»««••..««. **••.•' 70 . 6. .Addition of t&s ethyl radical to ally! acetate sobstratesi TS V., Metathesis between tho ethyl radical and den«srat«d allyl acetate • substrates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8. • DiarcratatJcts of th© addition. addact radical of allyl acetate . . . . . . . . . 60 9. . Metathesis acd addition between the ethyl radical sod various allyl alcohol substrates? * 8S 10. CoBsbittfid reaction* (5) snd (6) of metatfeeais festween the ethyl radical and undeuterated altyl alcohol. . . . . . . . 80 n . The isotope effect In metathesis* between the ethyl radical end the hydroxy! grota? of ally! alcohol...... w SO 1%. ©Ismatation of th® addition adduct radical of allyl glcahel SS 1.2,. Addition assd m&$&Hh&®l® bttwe^a the ethyl radical and allyl propionate * *. ©3 14. XHsssutaUoo, of th® addition adduct radical e? allyl propionate. ........ 57 15. Addition of the ethyl radical to allyl formate substrates 104 16. The isotope effect in metathesis between the ethyl radical and the formyl group of allyl formate. 105 17. Dismutation of the addition adduct radical of allyl formate 106 vii LIST O F T A B L E S Table Page I. The Reactions of the Ethyl Radical with Hexadiene-1,5 . . . . . . . . . 15 II The Reactions of the Ethyl Radical with Al lyl Ethyl Ether . 17 m. The Reactions of the Ethyl Radical with Diallyl Ether . . . . . . . . . 19 IV. The Reactions of the Ethyl Radical with Al ly l Acetate 21 V. The Reactions of the Ethyl Radical with Allyl-1 , l - d 2 acetate . . . . . 23 VI. The Reactions of the Ethyl Radical with Al lyl Acetate-dg . . . . . . . . 25 V H The Reactions of the Ethyl Radical with Al ly l Alcohol . 27 VHX The Reactions of the Ethyl Radical with Al ly l AlcohoL-d 29 XX. The Reactions of the Ethyl Radical with Allyl-1, l - d 2 Alcohol and Diallyl Carbonate 31 X . The Reactions of the Ethyl Radical with Allyl Propionate 33 XI. The Reactions of the Ethyl Radical with Allyl Formate and Allyl,Formate-d 35 X E . Arrhenius parameters for the metathesis reactions of the Ethyl Radical with Al ly l Compounds 37 XHI Arrhenius Parameters for the Addition Reactions of the Ethyl Radical with Al lyl Compounds 39 XIV Arrhenius Parameters for the Dismutation Reactions of the Addition and Metathesis Adduct Radicals 4 1 X V . A comparison of predicted and observed values of the degree of Polymerization at 80°C . . 43 INTRODUCTION A. Scope of this Investigation; This dissertation describes a kinetic study of the interaction of the ethyl radical with selected allyl compounds, . CH 2=CHCH 2X. Addition and metathesis are observed in each system: C H 3 C H 2 + CH^CHCHgX^CHgCHgCHgCHC^X C H 3 C H 2 + CH 2=CHCH 2X -S>CH3CH3 + (CH^-CH'-CHX)' and the values of the corresponding Arrhenius parameters are consistant with the values 1 2 obtained for the interaction of the ethyl radical with 1-alkenes and with vinyl monomers. The reaction of special interest in the present investigation is the dismutation of the adduct radical: CH 3 CH 2 CH 2 CHCH 2 X CHgCHgCHgCHsCRg + X' The standard enthalpy of this reaction may be estimated, and is sensitive to the nature of X. The energy of activation is closely linked to the enthalpy of dismutation, and a comparison of these quantities for a series of allyl substrates has enabled a distinction to be made between two different reaction paths for a particular dismutation reaction. This distinction is made for the dismutation reactions of the allyl esters, and is described fully in part A of the Discussion section. In certain cases the reactions of the product of dismutation are of extreme interest. The diallyl system yields the allyl radical in this way, arid the pattern of disproportionation and combination between the allyl radical and the ethyl radical as shown below may be studied under favourable circumstances. C H 3 C H 2 + C H 2 - C H - C H 2 -> CHgCHgCK^CH-CB^ (c) C H 3 C H 2 + C H 2-CH1CH 2— ? C H 2 = C= C H 2 + CHgCHg (d) C H 3 C H 2 + C H 2 - C H i : C H 2 - ? C H 2 = CHCH 3 + CRgCHg (e) Throughout this dissertation, any equation labelled (c) represents a combination reaction of two free radicals, and reactions labelled (d) and (e) represent dispro-2 portionation reactions between two free radicals. The ethoxy radical is generated in the allyl ethyl ether system, and its dispro-portionation-combination reactions with the ethyl radical shown below could be studied quantitatively in this investigation. CRgCH 2 + CH 3 CH 2 0 ' - ) C H 3 C H 2 O C H 2 C H 3 (c) C H 3 C H 2 + CH 3 CH 2 0'-? CHgCHO + C H 3 C H 3 (d) C H 3 C H 2 + C H 3 C H 2 0 ^CH 3CH 20W+ C H 2 C H 2 (e) The relative importance of the reactions of metathesis, addition and dismutation for a given monomer in this investigation may be linked with the behaviour of the same monomer in radical polymerization in the liquid phase. The analogous reactions are degradative chain transfer, propagation and effective chain transfer; the validity of a • 3 quantatative comparison of this type has been established previously. The use of allyl compounds selectively deuterated in particular functional groups has revealed the reactivity of such groups separately towards metathesis, adding greatly to the interest of the results obtained. The reactions of the allyl radical and the characteristics of the radical polymeri-zation of allylic monomers are therefore of particular interest in this study, and will be discussed in greater detail in sections C and D of the introduction. B. A General Mechanism The ethyl radical used to study the reactions of the allyl compounds was generated by the photolysis of diethyl ketone using 3130°A radiation. The mechanism for the 4 5 6 photolysis of diethyl ketone has been studied by many workers ' ' , and is shown below: C 2 H 5 C O C 2 H 5 —> 2C2Hg + CO (1) 2 C 2 H 5 - > C 4 H 1 ( ) (2) 2 C 2 H 5 - > C 2 H 4 + C 2 H 6 (3) C 2 H 5 + C 2 H 5 C O C 2 H 5 -> C 2 H 6 + b 2 H 4 C O C 2 H 5 <4> 3 A high light intensity was maintained in all the photolysis runs of this investigation, and the temperature was kept below 250° C in order that the pentanoyl radical was removed exclusively by the following reaction: C 2 H 4 C O C 2 H 5 + C 2 H 5 _> C 4 H 9 C O C 2 H 5 (4c) From the above equations, it is seen that one molecule of carbon monoxide is equivalent to two ethyl radicals, or one molecule of either butane or ethane.. A material balance can then be defined using the relation M= R C 2 H 6 + R C 4 H 10 R C O -3 -1 where R_ denotes the rate of formation of the species Z in molecules cm sec . Values. ZA 5 1 7 for the material balance were found to be 0. 988 + 0.02 , 0. 997 + 0.03 and 1.006 + 0. 01 . These values equal unity within the experimental error and therefore support the mechanism for the photolysis of diethyl ketone. Many values have been found for the ratio of disproportionation to combination of the 1 6 8 ethyl radicals. Several investigators using diethyl ketone as a source of ethyl radicals ' * have found the value of kg/kg to lie in the range of 0.12 to 0.15. Thynne , using ethyl formate as a source of ethyl radicals, and Kerr and Trotman-Dickenson1^ using prop-ionaldehyde as a source also found the kg/kg values to lie in the above range. In the case of diethyl ketone, at higher temperatures kg/k 2 was no longer independent of temperature, and increased to 0.19 at 222°C. Brinton and Steacie6 explained this by accounting for additional ethylene produced by the thermal decomposition of the pentanoyl radical, C H 3 C H C O C 2 H 5 —> C 2 H 4 + CO + This reaction would have the effect of increasing k^/kg, but would not affect the material balance, "M", mentioned earlier. Hence in the study of diethyl ketone in the presence of unsaturated compounds, it is desirable to study the reactions below 200°, since above this temperature the pentanoyl radicals are thermally unstable and the material balance deviates 4 from unity. The reactions of the ethyl radical in the presence of an unsaturated compound, 2 11 12 CH2=CHCH2R, have been fully studied in recent papers ' ' , and the reactions below were shown to adequately describe the kinetic scheme. In all cases, the reaction con-ditions were controlled so that the ethyl radical was always present in great excess over all the other radicals. C 2 H 5 + C H 2 = CHCH 2 R-» C 2 H 6 + (CR"2- C H ^ ' C H R j ' C 2 H g + [ CH 2 - C H - C H R ] - = ) C H 2 = CHCH ( C 2 H 5 ) R c 2Hg + [CH2-CH,^CHR]'~>C2H5CH2CH= CH2R C 2 H ^ + CH 2 = CHCH 2R - 9 ^HgCHgCHCHgR C 2 H 5 + C^CHgCHCHgR ~* C ^ C H g C H ( C ^ ) CHgR It is shown in this dissertation that the kinetic scheme represented by the above reactions is extremely useful in describing the reactions of the ethyl radical with allyl compounds. C. Allyl Radical 13 14 15 The allyl radical is recognized as an important intermediate in the pyrolysis ' ' ' 16 17 18 19 20 21 22 23 24 ' ' ' , photolysis ' ' ' and radiolysis of various olefins and cycloalkanes. 27 It has been identified by mass spectrometry in several pyrolysis systems. The electron spin resonance spectrum of the allyl radical has been observed by many workers, and 28 29 has been correlated with the theory ' . I t s spectrum has been observed in the. irradi-30 31 32 ation of liquid cyclopropane , of solid propene , of polypropylene and of solid allyl 33 bromide containing a small proportion of sodium . These observations can be supple-mented by many more from other workers, and indicate the interest shown in the allyl radical. Since hydrocarbon free radicals exhibit reactions of addition, metathesis, combination and disproportionation, these same reactions might be expected to flourish for the allyl radical. However, no evidence has been found for addition and metathesis reactions of o 34 35 36 the allyl radical below 300 ' * , and the mutual or cross combination reaction with, other radicals predominates below this temperature accompanied by an unknown con-tribution from the disproportionation reaction. The absence of the metathetical and addition reactions below 300° results from the high energy of activation associated with these reactions. This high energy of activation is seen when a comparison is made of the activation energy for metathesis with the same substrate, cyclopentane and the allyl radical or a typical hydrocarbon free radical, the methyl radical, the corresponding 37 34 activation energies being 32 kcal/mole and 9.3 kcal/mole respectively. The difference in these activation energies is approximately equal to the derealization energy of the allyl radical, estimated to lie in the range 15 to 20 k c a l / m o l e 3 8 ' 3 9 ' 4 0 ' 4 1 ' 4 2 , and is an important quantity in this investigation. This magnitude is in keeping with the estimate made of the derealization energy of the allyl radical based upon a value of 32 kcal/mole 43 for the heat of formation of the allyl radical . There is a need for reliable quantitative information upon the reactions of cross-disproportionation between the allyl radical and a representative alkyl radical; the lack of such information complicates the interpretation of the kinetics of many pyrolytic, photolytic and radiolytic systems. The extent and pattern of the disproportionation reactions of the allyl radical with the ethyl radical were determined in this investigation at temperatures below 200°. The method of generation of the allyl radical must conform to certain conditions if the interpretation of its interaction with the ethyl radical is to be unambiguous. The source must yield allyl and ethyl radicals in abundance, and interfering radicals in negligible quantities only. Also, in order to ensure that the products allene and propene are pro-duced exclusively by the disproportionation reactions between the allyl and ethyl radicals and not by the disproportionation reactions between two allyl radicals, the ethyl radical must be formed in a much greater concentration thafi the allyl radical. A reaction temperature below 300° ensures that the allyl radical will not yield significant amounts 37 45 of propene by metathesis or of allene by dismutation . Mercury resonance radiation 27 presents a problem since it may yield allene through the process: , Hg \ + C 3 H 5 -> Hg \ + C 3 H 4 + H ' Such conditions eliminate two common sources of the allyl radical, the thermal de-34 46 composition of the cyclopentyl radical and of the diallyl substrate . Photolysis and radiolysis are the obvious methods for generation below 2 0 0 ° . However, the concomitant 20 generation of other reactive species renders the radiolytic method unattractive. Propene 20 21 butene-1 and cyclopropane each yield allyl radicals in a primary photochemical process, but competing primary processess render them unattractive as sources. The generation of the allyl radical shown by the reactions below, from the photolysis of diethyl ketone in the presence of diallyl represents a system with none of the complicating factors mentioned above: C H 3 C H 2 + C H 2 = C H C H 2 C H 2 C H = C H 2 - ^ C H 3 C H 2 C H 2 C H C H 2 C H 2 C H = C H 2 C H 3 C H 2 C H 2 C H C H 2 C H 2 C H = » C H 2 - > C H 2 = C H C H 2 C H 2 C H 3 + [ C H ^ ' C H H r C H ^ ' Disproportionation between the allyl radical and an alkyl radical is often assumed to be kinetically negligible. This assumption seems to be based upon the mass spectro-25 o metric study of the pyrolysis of allyl iodide at 750 ; the results imply that when two allyl radicals: interact, mutual combination is more than one hundred times as probable as mutual disproportionation. This upper limit of 0.01 assigned to k; /k for the mutual interaction of two allyl radicals could very well be temperature dependent. Recent 47 studies of the interaction of two ethyl radicals in the gas phase have indicated that k^/k c decreases with increase in temperature, and a decrease of approximately 20% in the value of k^/k 'with an increase of 105° in temperature was revealed. Hence a very large temperature change such as 600° would have a considerable effect on this quantity. D. Allyl Polymerization The characteristic feature of allyl polymerization is found in the low degrees of 48 polymerization lying approximately in the range 2 to 20 . This is the opposite of vinyl 49 50 2 51 polymerization where the chain lengths are powers of ten higher ' ' ' The degree of polymerization represents the number of monomer units in the polymer and is given by the expression k^/k^ , the ratio of propagation and transfer constants. The recon-ciliation of the above observation cannot be found in the propagation step, since the new radical produced is vinylic in nature for both the vinyl and the allyl systems shown below: allyl system M.' + CH„ = CHCH 0 X -> M.CH„CHCH0X (propagation) M.' + CEL = CHCELX —> M.H + [CH'-CH^CHX]' (degradative chain transfer) 1 U Ck 1 it vinyl system M.' + CH = CHX-^M.CH„CHX (propagation) M." + CH0=? CHX M.H + CH„ =» CX (effective chain transfer) The distinction is found in the metathetical reaction whereby the new radical gen-erated in the allyl system is resonance stabilized. This metathetical reaction has been 53 termed "degradative chain transfer" , and can be contrasted to the metathetical reaction termed "effective chain transfer", in the vinyl system whereby the new radicals produced have approximately the same reactivity as the radicals produced in the propagation step. 54 Hence, since the allylic radicals are much lower in reactivity than the vinylic radicals , and are therefore unable to propagate the chain, they undergo termination with either themselves or a growing polymer radical resulting in both structural and kinetic chain termination. One reaction not mentioned thus far that plays an important role in the mechanism of allyl polymerization is the radical displacement reaction involving the adduct radical, an example of which is shown below: CH 3 CH 2 CH 2 CHCH 2 OCCH 3 —> C H 2 r=s. C H C H 2 C H 2 C H 3 + C 0 2 + CHg This reaction has the effect of causing structural, but not kinetic, chain termination, and is termed "effective chain transfer" since the new radical produced has approximately the 56 same reactivity as the radical produced in the propagation step. Gaylord and Eirich 57 proposed this reaction in 1952, while Gaylord and Kujawa have interpreted the poly-8 merization of 3-buten-2-yl acetate in this way. Since benzoyl peroxide was used as the initiator in these systems, one of the reaction products, carbon dioxide generated in the dismutation reaction was obscured by the carbon dioxide produced from the initiator. 58 Gaylord, Katz and Mark have reinvestigated the allyl acetate system using an initiator that did not produce carbon dioxide, and found direct evidence for the radical displace-ment of effective chain transfer. They showed that carbon dioxide and ethane were evolved and were consistent with their reaction scheme. Quantitative results for the dismutation reaction have been found for this allyl ester along with several other allylic compounds in this study. 9 EXPERIMENTAL A. Apparatus Kinetic measurements were conducted on a conventional high vacuum apparatus in which mercury cut-offs were used instead of stopcocks. Hence errors arising from absorption of reactants and products in stopcock grease wero avoided. The optical system is essentially' the same as that used by James and Steacie1. A British-Houston ME /D 250 watt mercury arc lamp supplied the U. V. light. The optical filter was a 0.5% solution of potassium hydrogen phthalate contained in a cylindrical quartz cell of 5cm diameter and 2cm length. This filter does not transmit wavelengths of less than 3000 A and transmits about 70% of the incident 59 '* radiation at 3130 A . The fused quartz reaction cell was a cylinder 10 cm long and 5 cm in diameter, 3 with an illuminated volume of 196 cm . This cell was enveloped by an electric resistance furnace and its temperature was measured by means of three copper-constantan thermocouples, which were distributed over the surface. B. Analysis o o The products were separated into four fractions» volatile at -210 , -150 , r -120°, and +252; the amount of the first three fractions was estimated by a calibrated gas burette. The lowest temperatures, -210?, was obtained by means of a solid nitrogen trap and the other three temperatures using a Ward-Le-Roy still . The mass spectrometer showed that the -210 Cfraction was pure CO except in the case of the allyl formate-d and allyl alcohol and its deuterated analogues whereby methane and hydrogen were measured by the mass spectro-meter as shown in the results. In the case of allyl acetate and its deuterated analogues, the -210 fraction contained only carbon monoxide and methane. (Allyl acetate)-dQ and allyl 1, l-dg acetate gave only CH4, and allyl acetated-d3 gave only S H D 3 . The composition of this fraction was determined using the silica gel column J at 80° and the flame ionization detector. The method was calibrated ip using known amounts of methane; the peak areas were directly proportional to the amount of methane present, and the absolute sensitivity of the method was ob-tained. Duplicate analyses were performed upon several product fractions using the mass spectrometer; corresponding estimates for the methane content agreed to within 5% in each case. Whenever carbon dioxide was present, the -150^fraction was separated into its components on the two metre silica gel J column at 5()£ using helium at a pressure of 8 p. s.i. as the carrier gas and using thermal conductivity for detection. Otherwise, the -150^ fraction was separated on J column at 85?, using nitrogen at a pressure of 12 p. s.i. as the carrier gas, and flame ionization for detection. Flame ionization was used for detection for the other two fractions. When the -150^ fraction contained deuterated products, it was divided into two fractions, one fraction stored in a break seal tube and subsequently analyzed by a mass spectrometer, and the other fraction analyzed by gas chromatography described 13 above. A correction for the natural C isotopic content was included whenever the C^ILD/C^EL ratio was calculated. The rates of formation of the gases in the 42 5 <2 6 -150^ fraction were obtained from the corresponding areas of the chromatograph; these areas .were corrected by pre-determined calibration factors. For example, the peak area of ethane had to be multiplied by a factor of 1.30 when comparing it to that of carbon dioxide. Whenever significant, a correction using Van der Waal's constant 6 1 was used. The lighter gases, € j to Cg, did not require this correction, Similarly the -120 enaction was analyzed on J column at 160? using nitrogen at a pressure of 12 p. s. i . as the carrier gas. Again, whenever deuterated products were found in this fraction, gas chromatography was supplemented by mass ; spectrorr-^try. The 25 cfraction was combined with the -120? fraction and analyzed on polyethylene glycol R column at 60c using nitrogen at a pressure of 12 p. s. i . as a carrier gas. Pentene-1 served as an internal standard for the 25* fraction. The retention times of the liquid products were found on the above column and compared with those of the suspected compound. The retention times of the liquid products were also found on diisodecyl phthalate Column A using the same conditions 11 as column R and compared again with the retention times of the known compounds. In this way identity of the liquid products were established. Control experiments were carried out for all the allylic compounds studied in this investigation in order to establish that they did not thermally decompose. Each substrate was expanded into the reaction cell at a pressure representative of that dur-ing a normal photolysis, and maintained at the highest temperature of the photolysis study. The substrate was then kept under these conditions for the normal time of photolysis which was usually 1800 seconds. No products of thermal decomposition were observed in all cases. Two other control experiments were carried out showing that the more reactive products were not consumed during the period of photolysis. In the first control experiment diethyl ketone was photolyzed at 172^ in the presence of allene, propene and butadiene, each in amounts representative of the final composition at that temperature, and the system was analyzed before and after illumination at normal intensity of 7200 sec. The initial composition was: diethyl ketone, 187 micromoles; allene, propene and butadiene, 23.8, 94.2 and 47.6 millimicromoles respectively, and the final composition was: allene, propene and butadiene, 23.2 93.0 and 47.0 millimi-cromoles respectively. Hence the recovery of the hydrocarbon was therefore about 98% under typical conditions at the high temperature region of this study. In the second control experiment diethyl ketone was photolyzed at 150^ in the presence of acrolein and acetaldehyde, each in amounts representative of the final com-position of a normal experiment at that temperature, and the system was analyzed before and after illumination at normal intensity for 1800 sec. Initially 170 micromole of diethyl ketone, 850 millimicromoles of acrolein and 102 millimicromoles of acetaldehyde were present; after illumination 840 millimicromoles of acrolein and 98 millimicromoles of acetaldehyde remained. Therefore, the recovery of acrolein was about 90% and the re-covery of acetaldehyde about 96% under typical conditions in the high temperature region of this study. C. Reagents Diethyl ketone, and the undeuterated allylic compounds used in this study were reagent 12 grade and supplied by Eastman Organic Chemicals. The deuterated compounds, allyl alcohol-d, allyl-1, l - d 2 acetate, allyl acetate-d^ and allyl formate-d were supplied by Merck, Sharp and Dohme. Each deuterated compound was checked by mass specto-metry after further purification described below, and found to be 99% deuterated in the labelled position. The above reagents were purified on a 2 metre polyethylene gylcol R column using helium at a pressure of 12 p. s. i . as a carrier gas, and their purity sub-sequently checked by the same column using flame ionization for means of detection. The column was run at 110° for the allyl acetates, propionate and alcohols and 60° for allyl formate-d, diallyl carbonate, allyl ethyl ether, diallyl ether and hexadiene-1,5. Allyl-1, l - d 2 alcohol was prepared from allyl-1, l - d 2 acetate in the following manner. A saponification was carried out whereby a mixture consisting of 1 ml of allyl-1, l - d 2 acetate, 1.6 g.. of reagent potassium hydroxide dissolved in 2 ml of water, and 6 ml of methanol was refluxed for one-half hour. After refluxing, a fraction distilling off at 78°C was collected and separated by means of gas chromatography. The separation was carried out on diisodecyl phthalate A column run at 70° C using helium as a carrier gas at a pressure of 12 p. s. i . The peak at 15 minutes was collected and gave approxi-mately 0.1 g. of allyl-1, l - d 2 alcohol subsequently identified by n.m. r. The n. m. r. spectra were measured on a Varian A-60 instrument for both allyl alcohol and attentively, allyl-1, l - d 2 alcohol. The spectrum for allyl alcohol was similar to that given in the 62 catalogue of Varian Associates . In the case of the assignment of allyl-1, l - d 2 alcohol, the n. m. r. spectrum was identical to that for allyl alcohol except for the expected absence of the absorption peak resulting from the two allylic hydrogens. D. Photolysis of Diethyl Ketone in the Presence of an Al lyl Compound (1) Experimental Procedure for a Typical Kinetic Run Diethyl ketone was admitted to the preparation line and reaction cell until the required pressure was obtained. The cell temperature was measured by the cell thermocouples, the reaction cell was then isolated by means of a mercury cut-off, and the reactant pres-sure was measured to 0.001 cm using a cathetometer. The excess ketone remaining in the preparation line was pumped off and the measured ketone in the reaction cell was 13 distilled into a temporary storage vessel with liquid nitrogen. The allyl compound was admitted to the cell and the temperature and pressure were measured by the same pro-cedure as diethyl ketone. The excess allyl compound was then pumped out of the pre-paration line after which the measured ketone in the temporary storage vessel was distilled into the cell cold finger with liquid nitrogen. The two reactants were warmed gently with a circulation of hot air supplied by a hair dryer, and were mixed thoroughly by lowering and raising the mercury in a mixing vessel several times. The photolysis was timed by stop watch and the temperature was calculated from the mean of several e. m. f. readings taken at regular Intervals during the run. After the photolysis the contents of the cell were admitted to the analysis line where the bulk of thesseactants were condensed by a dry ice trap and the products were seperated into fractions using a solid nitrogen trap and Ward-Le-Roy still. These fractions were discussed in detail earlier. 14 RESULTS The values of [D] and [B] quoted in the following tables are initial values of diethyl ketone and the allyl compound respectively. In the case of diallyl carbonate, a correction was applied for the changes in these quantities due to their consumption during the reaction. Values in brackets are excluded on statistical grounds. A l l limits of error are calculated at the 5% probability level. The units of k 2 , kg, k 4 > k 5 j ^ , k ^ \ c 3 -1 k 7c' k 8 C k 8d' k 8 « ' k 9c ' k 9d' a n d k9e a r e c m / molecules sec; of k g and k g , sec . The reactions corresponding to the above rate constants are shown below: C H 2 = CHCHgRH represents an allyl compound and R H represents the substituent group. C H 3 C H 2 + C H 2 = C H C H 2 R H ^ C H 3 C H 3 + C H g - C H C H 2 R ' (5) CHgCH^ + C H 2 = C H C H 2 R'icH 2» C H C H 2 R C H 2 C H 3 (5c) C H 3 C H 2 + C H 2 = C H C H 2 R H \ C H 3 C H 3 + C H 2 = CHCHRH (6) C H 3 C H 2 + CH 2« CHCHRH ^ » t C H 3 C H 2 C H 2 C H = CHRH (6c) C H 3 C H 2 + C H 2 » CHCHRH ^ C H 2 = CHCH ( C ^ ) R H (6ci) C H 3 C H 2 + C H 2 = C H C H R H ^ C H 3 C H 2 C H 26HCH 2 R H (7) C H 3 C H 2 + C H 3 C H 2 C H 2 C H C H 2 R H ^ C H 2 C H 2 C H 2 C H ( C 2 H 5 ) CH2RH (7c) CHgCHgC^CHCHgRH !5S > CH 3 CH 2 CH 2 CH= C H 2 + RH' (8) C H 2 = CHCHRH ^products (9) The subscripts 5H and 6H represent the abstraction of a hydrogen atom from the allyl compound while the subscripts 5D and 6D represent the abstraction of a deuterium atom from the allyl compound. u u The subscript'C-represents the combination reaction, of two free radicals. The rate constants k g c> ^8d' k8e r e P r e s e n t ^ e combination-disproportionation reactions between the ethyl radical and the free radical generated by reaction (8). Similarly, the 14a rate constants kg c > k g (j and k g e represent the combination-disproportionation reactions between the ethyl radical and the free radical generated by reaction (9). The rate constants k 2 , kg and k 4 describe the reactions shown on page 2. The derivation of rate equations for the above reactions is shown in Appendix B. T A B L E I Reactions of the ethyl radical with hexadiene-1. 5. temp time -17 10 A' [DjlO- 1 7 fel i o " 1 2 'Rx (molecule/cm 3 sec) <°C) (sec) 3 (molecule/cm ) CO C2 H4 C 2 H 6 C4 H10 C3 H4 C 3 H 6 C4 H6 C 4 H 8 C2 H2 71 < 1800 4.91 4.73 13.0 1.36 2.40 9.46 0.000 0.000 0.000 n.a. n. a. 0.000 91 1800 4.00 5,43 13.1 1.28 2.85 7.87 0.017 n.a. n. a. n.a. n. a. 0.000 103 1800 5,32 4,19 15.2 1.25 3.66 9.10 n. a. n. a. n.a. n. a. n. a. 0.000 134 1800 3.51 4.50 12.7 0.653 3.74 5. 23 0.280 0.007 0.016 n. a. ii. a. 0.000 149 1800 4.07 4. 84 15.2 0.707 5.18 5.01 0.470 0.009 0,~025 n.a. n. a. 0.000 154 1800 4.59 4.74 16.0 0.786 5.86 4.41 0.670 0.015 0.040 0.015 0.010 0.002 161 1800 3.40 3.46 13.3 0.660 4.75 4.22 0.810 0.020 0.060 0.035 0.030 0.004 162 7200 4.18 4.20 13.3 0.554 5.26 3.46 1.01 0.018 0.064 0.035 0.030 0.004 164 7200 3.09 3.71 10.9 0.448 4.01 2.88 1.00 0.020 0.065 0.035 0.030 0.004 175 5400 3.63 3.45 11.4 0.437 4.78 2.25 1.61 ,0.030 0.080 0.080 0.070 0.007 175 7200 4.07 4.16 11.9 0.388 4.90 1.96 1.93 0.035 0.112 0.080 0.070 0.007 TABLE I (Don't) M k3 k8c k8d k8e k2 k8 k8 k8 0.912 0.144 — .. • 0.818 (0.163) — . 0.840 0.137 — . 0.706 0.122 0.848 0.046 .0.106 0.670 0.136 0.865 0.036 0.099 0.642 (0.165) 0.848 0.041 0.110 0.674 0.142 0.820 0.045 0.135 0.C56 0.142 0.850 0.033 0.117 0.632 0.133 0.843 0.037 0.120 0.617 (0.159) 0. 872 0.035 0.093 0.577 0.141 0.858 0.034 0.108 Mean values 0.137 0.851 0.038 0.111 + 0.017 + 0.037 +0.012 + 0.027 10V k 2 / 2 ° 15.0 2.51 1.65 12JL 0.00 0.00 10.8 1.93 1.20 8.49 0.00 0.00 4.86 1.04 0.53 3.46 0.00 0.00 3.76 0.93 0.33 2.43 0.370 0.31 12.7 2.69 1.09 8.86 0.120 0.10 11.3 2.64 1.11 7.24 0.120 0.09 12.9 3.12 1.09 8.09 0.130 0.10 17.7 5.06 1.54 10.1 0.370 0.30 10.1 3.17 0.73 5.29 0.40 0.34 13.6 4.39 1.26 6.72 1.02 0.80 13.3 4.56 1.14 7.07 0.870 0.050 12.5 4.75 1.36 5.16 1.94 1.30 C_H_OH M* 0.00 0.856 0.00 0.963 0.00 0.925 0.050 0.884 0.020 0.907 0.020 0.876 0.030 0.869 0.050 0.856 0.050 0.837 0.200 0.815 0.300 0.873 0.600 0.787 Table H : (Can't.) F l F2 k3 k8c k8d k8e mean values: kg k V k„ k„ *8 8 8 0.19 0.136 0.20 0.141 0.25 0.154 0.97 0.23 0.115 0.84 0.03 0.14 1.00 0.19 0.121 0.83 0.00 0.17 0.92 0.15 0.151 0.75 0.08 0.17 1.00 0.22 0.131 0.77 0.00 0.23 0.95 0.43 0.148 0.81 0.05 0.14 0.98 0.31 0.126 0.85 0.02 0.13 0.98 0.158 0.78 0.02 0.20 0.92 0.118 0.98 0.148 0.97 + 0.06 0.20 0.137 + 0.08 + 0.015 0.80 0. 08 0.03 + 0.02 0.17 ^0.08 10 1 3 (k 6 +k 6,10 1 3k 710-5 k 8 k^ 2 k l / 2 K2 2.8 3.6 6.2 9.2 8.1 6.4 9.4 8.1 14.7 12.2 15.6 20.0 2.4 3.2 4.8 9.8 10.2 10.9 15.9 18.3 19.1 29.3 (20.2) 30.8 2.4 5.4 7.4 17.6 27.9 62.0 TABLE HI Reactions of the ethyl radical with diallyl ether mp time IO"17ID3 io-17fB7 IO"12?^. 3 (molecule/cm sec) 'C) (sec) 3 (molecule/cm ) CO C 2 H 5 C 2 H 4 . ^ O C 2H 3CH0 C5 H10 C 3 H 4 C 3 H 6 °6 H10 C 3 H 5 C 3 1800 5.51 6.46 11.1 2.64 1.00 6. 85 0.000 0.000 0.000 0.000, 0.000 0.000 3 1800 5.47 6.37 11.0 2.59 0.960 6.80 .0.000 . 0.000 0.000 0.000 0.000 0.000 2 1800 4.60 4.98 13.2 3.29 1.89 7.15 0.000 0.000 0.000 0.000 0.000 0.000 14 1800 4.20 2.66 12.9 3.36 1.01 7.16 0.133 \ 0.131 n. a. n. a. 0.000 0.000 J8 1800 4.01 4.85 .'. 14. 3 4.89 0.700 4.49 0.741 . 0. 631 0.030 0.060 0.000 0.032 J8 1800 4.50 4.58 15.0 5.18 0.660 4.03 1.40 1.25 0.050 0.100 0.000 0.052 10 1800 4.08 4.33 14.6 5.09 0. 657 4.10 .1.50 1.31 0.050 0.100 0.000 0.056 53 1800 4.47 3.97 17.0 6.03 0.752 3.76 2.05 1.79 0.120 0.220 0.020 0.158 53 1800 3.80 4. 22 17.8 5.99 0.787 3.88 2.20 1,94 0.130 0.233 0.020 0.170 53 1800 3.78 4.18 18.0 5.95 0. 802 3.90 2.20 1,90 0.130 0.230 0.020 0.160 Table HI: ( Con't. ) M* Q Fc Fc Fi Fe Fe 10"V2 l0"V2/2 7c Mean Values: 0.856 0.145 7.32 9.44 0.855 0.141 7.40 9.60 0.817 0.135 12.1 18.2 0.816 0.99 0.140 17.5 32.5 2.26 0.654 0.92 0.140 0.88 0. 87 0.040 0.044 0.081 0.088 26.4 48.2 (0.169) (7.95) 0.611 0.96 0.136 0.89 0.89 0.036 0.037 0.072 0.075 26.5 63.5 0.223 27.0 0.626 0.93 0.132 0.90 0.89 0.033 0.036 0.067 0.072 27.9 62.3 0.260 31.9 0.567 0.97 0.133 0.83 0.83 0.060 0.061 0.100 0.112 32.9 95.3 0.530 (80.9) 0.547 0.97 0.134 0.83 0.83 0.060 0.062 0.109 0.111 34.7 97.0 0.526 62.6 0.532 0.96 0.139 0.83 0.82 0.060 0.064 0.109 0.113 34.3 100 0.482 67.7 + 0.96 0.06 + 0.138 0.013 + 0.86 0.06 0.86 + 0.06 0.048 + 0.023 0.051 + 0.024 + 0.091 0.036 + 0.095 0.035 to o TABLE IV Reactions of the ethyl radical with allyl acetate temp time io-17fc aio~17£Bj ile/cm3) -1 9 10 R z 3 (molecule/cm sec) <°c> (sec) (molect; CO CH. 4 C 3 H 8 C ° 2 C 2 H 6 W 4 8 75 4000 6.54 6.37 5.65 n.a. h.a. n. a. 1.11 0.530 4.01 0.000 a. a. . 77 1830 5.05 5.33 41.1 0.030 0.290 0.350 5.71. 4.14 33.0 0.000 0.320 88 1800 5.00 4.65 14.7 0.010 0.110 0.120 2.75 1.35 10.4 0.000 0.120 101 1800 4.70 3.82 14.1 0.010 0.100 0.110 2.74 1.15 9.10 0.000 0.110 114 1800 4.27 3.71 17.0 0.030 0.200 0.250 3.42 1.46 10.2 0.020 0.230 115 1800 4.68 4.66 56.8 0.060 0.679 0.771 9.56 5.03 40.3 0.030 0.740 115 1800 4.33 3.99 20.6 0.040 0.275 0.386 4.28 1.68 12.9 0.040 0.339 120 1800 3.44 4.05 45.4 0.100 0.750 0.983 6.96 3. 82 30.2 0.050 0.850 120 1800 4.06 6.34 51.9 0.400 0.857 1.36 7.75 4.21 32.4 0.050 1.30 125 1800 3.76 7.02 39.4 0.300 0.764 1.22 6.18 2.96 22.5 0.100 1.10 125 1800 4.80 4.65 42.8 0.150 0.685 0.900 7.00 3.46 26.1 0.100 0.800 145 1800 3.79 2.96 16.5 0.250 0.788 1.20 3.42 0.989 7.76 0.200 1.00 145 1800 4.23 3. 60 18.1 0.250 0.940 1.30 3.73 0.940 7.96 0.150 1.20 145 1800 4.19 4.40 38.4 1.00 1.82 3.18 6.85 2.79 21.4 0.200 2i 80 145 1800 3.99 4.86 86.1 1.00 1.82 3.18 6.85 2.79 19.5 0.200 2.90 150 1800 8.13 4.35 11.4 0.80 1.61 3.20 2.36 0.678 4.64 0.500 2.40 170 1800 3.68 3.24 16.7 1.21 1.87 5.00 4.23 0.813 6.43 1.50 3.35 178 1800 4.15 4.15 47.1 4.00 6.25 14.4 9.57 2.61 20.5 4.30 10.0 Mz 1.000 1.094 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.001 1.076 1.002 1.000 0.992 1.034 1.007 1.034 1.017 0.958 1.000 0.963 1.000 1.008 0.963 0.993 1.059 1.028 1.025 0.996 1.103 1.088 1.031 0.976 1.025 Mfean Values 4-1.012 ft f>32 1.017 + ft 035 Q M ©.848 0.009 0.934 0.011 0.895 0.011 0.842 0.020 0.803 0.037 0.807 0.021 0.833 0.025 0.818 0.026 0.773 0.034 0.725 0.023 0.777 0.102 0. 677 0.118 0. 648 0.085 0.737 0.094 0.730 0.384 0.613 0.291 0.638 0.304 0.639 TABLE IV (con't) k3 1 0 *7 0.125 8.93 0.130 10.9 0.130 19.3 0.141 28.5 0.125 23.0 0.130 25.0 0.127 36.8 0.130 33.4 0.132 32.6 0.133 36.7 0.127 64.0 0.117 62.2 0.130 49. 8 0.143 47.0 0.146 47.0 0.126 72.9 0.127 90.5 0.131 + 0.005 1 0 ~ V 2 / 2 1.40 2.34 4.57 S. 92 5.90 6.74 5.35 4.68 (5.6) (5.6) 17.6 16.5 25.1 31.9 64.8 TABLE V Reactions of the ethyl radical with allyl-1, l-d„ acetate temp time i o " 1 7 £DJ IO-17[B3 i o ~1 2 3 R (molecule /cm sec) <°C) (sec) (molecule/cm ) CO CH. 4 C 3 H 8 C 0 2 C 2 H 6 W C2 H4 93 1800 4.96 4.27 13.0 0.010 0.073 0.085 2.33 0.079 1.18 93 1800 4. 86 4.20 13.5 0.010 0.085 0.097 2.56 0.087 1.21 120 1800 4.90 3. 66 14.2 0.100 0.410 0,570 2,94 0.110 1.03 120 1800 4.45 4.00 13. 8 0.100 0.340 0.450 2.69 0.100 0.910 150 1800 4.58 3.35 14.2 0.300 . 1.18 1.80 3. 27 0.129 0.680 150 1800 4,64 3.34 14.6 0.300 1.20 1. 85 3.36 0.136 0.610 162 1800 4.06 3.35 15.8 1.20 2.10 4.12 3.46 0.198 0.728 162 1800 . 4.00 3.28 14.5 1.10 1.36 3.49 2.95 0.155 0. 655 162 1800 4.24 3.30 14.7 1.20 1.39 3.70 3.42 0.158 0.704 to CO C4H10 C A D 2 C 5 H 8 D 2 8. 67 0.010 0.075 8.98 0.010 0.085 8.07 0.050 0.490 8,08 0.050 0.400 5.35 0.200 1.55 5,58 0.200 1.70 5.57 0.800 3.28 5.18 0.800 2. 60 5.34 0.800 2.80 Mean Values: + M l M 2 1.014 1.000 0,988 1,021 0.961 1,056 0.909 1.000 1.047 1.029 1.133 0.974 0.994 1.010 1.057 1.026 1.081 1.028 1.020 1.016 0.049 + 0.032 TABLE V (Con't) Q M h k 2 k7c 0.008 0.848 0.136 15.7 0.063 1.16 0,009 0.857 0.135 15.0 0.069 1.38 0.051 0.785 0.129 29.4 0.106 5.44 0.042 0.777 (0.115) 27.1 6.088 4.24 0.221 0.614 0.130 71.0 0.167 (9.11) 0.215 0.620 (0.109) 70.5 0.172 (10.4) 0.377 0.585 0.130 82.9 0.250 (18.4) 0.263 0.560 0.126 85.5 0.208 (14.4) 0.260 0.596 0.132 77.9 0.207 ao.9) 0.131 + 0.003 to TABLE VI Reactions of the ethyl radical with allyl acetate-d, mp time IO"17ID1IO'17£B] A. i o " 1 2 g R (molecule/cm sec) 'c> (sec) g (molecule/cm ) CC CHDg C 3 H 5 D 3 C ° 2 C 2 H 6 C 2 H 5 D C 2 D 6 1800 4.68 3. 80 14.0 0.010 0.089 0.100 2.73 n. a. n.a. 1.17 9.05 0.000 0.100 0 1800 4.59 4.34 60.6 0.110 0.278 0.417 8.37 0.042 0.008 5021 40.3. 0.050 0.360 15 1800 4.78 3.86 15.2 0.100 0.440 0.606 3.56 . 0.018 0.006 0.930 7.66 0.050 0.550 !5 1800 4.05 3.63 15.2 0.100 0.460 0.593 3.13 0.015 0.020 0.970 6.92 0.050 0.550 .'8 1800 3.84 3,38 59.5 0.800 1.70 2.62 10.8 0.046 0.007 4.53 36.4 0.200 2.30 •3 1800 3.52 3.02 15.3 0.300 0.980 1.48 3.67 0.021 0.007 0.780 6.93 0.200 1.25 18 1800 3.65 4.05 14.3 0.500 2.20 2. 86 3.03 0.026 0.004 0.581 4.61 0.250 2.60 >0 1800 3.89 4.13 64.1 1.50 3.41 5.41 12.3 n. a. n. a. 4.11 35.3 0.400 4. 80 i3 1800 5.11 4.38 16.4 1.20 1.11 3.81 4. 85 0.038 n. a. 0.694 4.85 0.700 8.10 .'5 5 1800 1800 4.07 3.60 3.42 2.98 60.5 16.6 3.75 1.50 5.25 1.56 11.2 6.20 12.7 4.17 0.040 0.035 0.11 0.05 3.68 0.760 30.1 6.59 2.00 2.50 t o 9.10 0 1 3.60 M l M 2 1.000 1.083 1.000 1.000 1.000 1.000 1.000 1.000 0.992 1.000 1.010 1.000 0.9278 1.017 1,0185 1.010 0.982 0.988 0.920 1.048 0.977 0.980 0.963 0.997 0.978 1.039 (1.340) 1.003 1.011 1.006 1.180 1.016 lean Values: 0.994 1.015 + 0.069 +0.031 Q M 0.002 0.998 0.004 0.972 0.005 0.965 0.012 0.890 0.010 0.844 0.010 0.841 0 . O i l 0.870 0.057 0.756 0.066 0.709 0.047 0.793 0.141 0.705 0.477 0.535 0.097 0.743 0.227 0.591 0.174 0.707 0.237 0.649 TABLE VI (Con't.) V i o 1 3 V >13w 10~5kg! -172= K2 , 1/2 k 2 0.133 — — ' 0.130 4.55 — 0.131 6.00 — . 0.129 10.8 0.060 — 0.127 18.9 — 1.44 0.130 19.4 1,41 0.129 25.5 (0.150) 2.82 0.122 34.0 0.168 4.80 0.140 40.0 0.157 3.76 0.125 60.4 0.240 13.8 0.112 55.0 0.264 16.1 0.125 77.0 0.300 13.7 0.118 67.2 • — 24.3 0.133 69.8 0.400 (18. 9) 0.122 93.4 (0.213) 57.2 0.115 76.0 0.550 41.8 0.126 0.006 TABLE VE A radical with allyl alcohol. temp time (°C) (sec) l O " 1 ^ 10"r'CBj (molecule/sec) CO —12 3 ,10 (molecule/cm sec) CH. C3 H8 C2 H4 C 2 H 6 86 1800 3.51 3.07 11.6 0.000 0.000 0.000 1.10 2.54 a. a. 120 1800 3.20 2.28 12.6 0.000 0.000 0.000 1.07 3.36 a. a. 133 1800 3.42 3.60 12.2 0.000 0.000 0.000 0.69 3.98 u. a. 141 1800 3.16 2.38 11.0 0.000 0.000 0.000 0.68 3.69 0.200 141 1800 3.53 2.46 11.4 0.000 0.000 0.000 0.72 4.02 0.220 155 1800 3.19 2.69 11.9 0.003 0.004 0.009 0.65 4.65 0.350 168 1800 3.24 3.36 13.1 0.010 0.010 0.020 0.58 5.46 0.520 to -a T A B L E VII (Cotr*t) 10" 1 2R Z (molecule/cm3sec) M* kg 1013(kg + kg) 10 1 3 k 5 H lO 1 3!^ loAgk^ 2 K TW~ k C_H_OH C nH 0CHO . .2 l C 5 10 2 T 2 3 7.90 0.000 n.a. n.a. 0.895 0.139 7.6 (0.5) 14.1 7.58 0.000 c.a. n.a. 0.368 0.136 14.0 2.8 23.5 . 5.27 0.000 n.a. n.a. 0.758 0.131 13.7 4.3 35.7 4.98 0.037 0.030 0.006 0.787 0.127 23.2 5.9 45.8 0.31 4.97 0.038 0.032 0.007 0.788 0.135 24.5 5.9 45.7 0.36 4.00 0.115 0.090 0.100 0.726 0.140 34.8 9.5 60.4 1.12 3.00 0.61 0.40 0.350 0.645 0.133 43.1 13.7 80.0 2.62 Mean value: 0.134 + 0.011 TABLE Vffl Reactions of the ethyl radical with allyl alcohol-d. Temp Time 10" 7£pjl rH rH rH CO in in CO o Cd CO o CO to fc-• * • • • • • • • • • o T-l in in o rH rH CO CO CO CM rH rH r-l T-i T-i CM CO CO in o o CD CO in CM CO CO t- in CO © to col CO co CO CO CO CO CO 03 C0 CO CO • CO CO col CO o r-j r-i rH r-l 1-1 rH rH rH rH rH rH r " l l rH © o O . ° ' O O © O ' © o o © © ©I o © CM CD © 00 O t - CO CM CNJ fc-rH CO CO in rH C0 rH CO CO O in «« a CO CO CO CC CO CO t> fc- . o O © o ' o o o o o o o © o © 1*5 o CM CO © CO rH CO © CO CM CO cn CO co T-i © CO T-i rH 00 CO ©ft © © © © © © © © © © • © © 1 © • ' * • • • • • * • • • I » © o rH © © rH © © rH rH © © ©! © rH CM O • o +1 ra fl) a! > .CS cu temp (G) time (sec) 10" 1 7fDjlO~ 1 7 £ B ] (molecule/cm ) CO ED D2 CH. CH4D C2 H6 S V C2 H4 C 3H 7D C3 H8 C4 H10 C ° 2 C 3 H 5 D 35 7200 7.00 6.37 9.05 - - 1.43 1.07 7.04 0.123 0.094 105 7200 4.24 4.59 . S.51 0.020 . 0.030 1.13 0.150 0.508 0.050 3.57 2.49 1.96 118 1800 4.42 4.02 13.0 0.030 0.040 2.63 0.300 1.04 0.097 7.29 5. 34 4.45 118* 1800 4.35 3.98 . 13.7 0.030 0.80 3.60 0.310 0.978 1.00 7.35 5.40 4.50 141 1800 2.35 4.55 10.5 0.060 0.040 1.98 0.430 0.548 0.09 3.61 14.3 11.9 120 1800 3,24 4.15 13.0 0.030 0.030 2.55 0.310 1.03 0.07 7.15 5.00 4.35 143 1800 2.40 4.67 12.1 0.060 0.050 2.10 0.490 0.560 0.100 3.94 15.3 12.5 temp time - co H 2 CH, 4 C 2 H 6 C2 H4 C3 H8 C4 H10 c o 2 °3 H6 C5 H10 104 1800 2.52 4.20 9.10 0.02 0.02 1.95 0.730 0.05 5.20 3.40 2. 70 0. 750 107 1800 3.50 4.37 9.30 0.02 0.03 2.06 0.810 0.06 5.33 3.57 2.77 0.800 124 1800 3.94 3.70 9.34 0.03 0.04 2. 79 0.621 0.100 4.00 8.00 6.55 1.13 121 1800 3.68 3.58 8.90 0.03 0.03 2.70 0. 616 0.090 3.92 7.75 6.35 1.00 124* 1800 ' 4.10 3.55 10.50 0.03 0.90 3. 81 0.710 1.10 4.15 8.20 6.65 1.18 140 1800 2. 25 4.50 9.20 0.07 0.080 2.80 0.475 0.120 2.70 13.6 10.5 2.30 152 1800 2.26 2. 01 7.05 0.100 0.150 2.43 0.450 0.150 2.56 10.1 8.00 1.65 158 1800 2.00 2. 30 8.20 0.100 0.210 2.80 0.472 0.180 2.69 12.8 9.50 2.35 160 1800 1.38 2. 15 8.10 0.060 0. 200 2.60 0.510 0.200 2. 89 13.0 9.70 2.34 * indicates 0.50 molecule/cm3 of biacetyl added. CO C l TABLE XI (Con't) C 5 H 1 0 C 3 H 4 C 4 H 8 C 6**io M l M K 3 / k 2 1 f .13 , 1/2 10 k 7/k 2' xo>\ 10~Xkl%c Q 0.027 - - 1^02 0.936 0.152 3.4 (1.0) - -0.510 0.0056 - - 0.996 0.754 0.138 18. 5 - H.-76 •1.7 (6.1) 5.0 0.800 0.011 0.05 0.001 0.995 0.785 0.140 25.8 (2.6) 2.66 2.6 6.50 7.8 0.750 0.010 0.200 0.001 - 0.129 - - • 2r87 2.3 - -1.85 0.017 0.280 0.002 0.989 0.574 0.137 51.8 4.1 • 4.97 5.0 . 9.80 8.0 0.66 0.012 0.050 0.001 1.01 0.771 0.132 27.1 2.3 2.82 2.8 4.13 10.8 2.30 0,019 0.400 0.002 0.995 0.542 0.137 59.7 5.2 • 5.29 5.2 10.6 6.5 • Mean Values: 1.00 0.138 + 0. 016 C 3 H 4 C 4 H 8 C 6 H 1 0 M l + 0". 03 M k + 3 / k 2 101 3k 7/k 13 * 1 0 k 5] [T K 2 io- 5k 8i^ / 2/k 7 c Q 0.028 0.050 0.001 1.02 0.785 0.125 20.5 (7.6) 7.34 1.92 17.4 0.030 0.050 0.001 1.03 0.795 0.134 18.8 (6.6) 7.45 2.24 17.3 0.034 0. 290 0.002 1.01 0.727 0.131 34.5 12.1 11.6 4.80 16.0 0.031 0.250 0.002 0.985 0. 744 0.135 32.2 14.7 11.0 3.50 7.6 0.033 0.600 0.002 - - 0.132 - - • - - • 0.050 0.520 0.003 0.990 0.596 0.123 50.1 22.8 17.0 8.20 8.5 0.028 0.520 0.002 1.01 0.706 0.142 64.5 27.8 21.9 16.5 7.6 0.039 0.600 0.003 0.978 0.570 0.132 71.8 33.0 26.1 20.9 6.7 0.040 0.650 0.003 0.980 0.676 0.135 71.7 35.0 • • 2 7 . 5 22.4 6.5 Mean Values: 1.00 0.132 + 0.04 + 0. 012 CO o substrate allyl alcohol-d allyl alcohol allyl alcohol allyl alcohol allyl acetate-dg allyl 1, d 2 acetate allyl propionate hexadiene-1,5 allyl ethyl ether diallyl ether octene-1* no. of expts. 8 6 8 7 8 9 11 11 11 10 12 allyl alcohol* 17 allyl formate-d 5 ally! formate 8 allyl formate 9 * results of Brown and James 3, reaction observed 5D 5H 6H 5H+ 6H 5D 6D 5H + 6H 6H 6H 6H 6H 5H+ 6H 5D 5H 6H 12 g 2 6.178 6.870 4.404 5.576 3.835 3.132 4.142 4.522 3.789 3.964 5.053 5.385 5.412 5.326 3.592 2.493 2.526 1.312 1.738 1.838 1.222 1.279 1.385 1.065 1.034 1.641 1.665 1.951 1.688 1.223 0.0264 0.0095 0.0153 0.0187 0.0184 0.0451 0.0706 0.0288 0.0696 0.0272 0.033 0.027 0.0110 0.0126 0.0316 0.1809 0.1054 0.1052 0.1298 0.1529 0.2191 0.3423 0.0976 0.2442 0.0919 0.133 0.087 0.109 0.0859 0.3993 0.0744 0.0438 0.0432 0.0529 0.0638 0.0887 0.1329 0.0399 0.0901 0.0354 0.050 0.032 0.0466 0.0348 0.1669 6.2+0.4 6.9+0.3 4.4+0.3 5.6+0.3 3.;8 +0.4 3.1+0.5 4.1+0.8 4.2+0.2 3.8+0.5 3.7+0.2 5.05 + 0.3 5.4+0.2 5.4+0.3 5.3+0.2 3.6+0.9 EM- 1 / 2 E2 11.4+0.8 11.6+0.6 6.0+0.3 8.0+0.6 8.4+0.7 5.6+1.0 5.8+1.4 6.3+0.4 4.9+0.8 4.7+0.4 7.5+0.5 7.6+0.3 8.9+0.6 7.7+0.4 5.6+2.0 CO -a TABLE XH (Con't) 13 + log kj^/Mgk^2 at 100°C 1.49+0.06 0.10+0.03 0.89+0.04 0.92+0.05 2.91+0.04 1.36+0.10 0.72+0.16 0.51+0.07 0.93+0.16 0.89+0.06 0.89+0.06 0.931+0.06 0.18+0.03 0.80 + 0.03 0.30+0.07 1/2 3 B and C are the coefficients of the straight lines 13 + log 1 Q • C-10 B/T fitted 1/2 to the data of tables I to XI.^,. a n d % are the standard deviations in 13 + log , C and B respectively. The limits of error are calculated at the 5% probability level. The units of 3 -1 -1 A » *3D' k 6D ' k 5 H a n d k 6 H a r e 0X11 * m o ^ e c u l e s e c > ^ e onits of E are kcal/mole. tTMg" is the number of allyl groups in the substrate molecule. " M " represents the observed metathesis reac t ion . TABLE XIII Substrate allyl alcohol allyl acetate allyl propionate allyl ethyl ether diallyl ether diallyl octene-1* allyl alcohol* allyl formate no. of expts. 15 42 13 11 10 11 11 18 13 * results of Brown and James 5.677 5. 8.12 5.828 5.407 5, 612 5.142 5.485 £ . 7 3 9 6.042 3, ll B 1.670 1.703 1.675 1. 602 1.555 1.452 1. 664 1.695 1.800 0.0133 0.0707 0.0326 0.0723 0.0322 0.0293 0.0116 0.034 0.0326 0.0646 0.1385 0.1368 0.2738 0.1087 0.0991 0.047 0.102 0.163 B 0.0264 0.0541 0.0525 0.0999 0.0419 0.0404 0.0174 0.037 0.0571 1 3 + 1 ° S 1 P P 7 2 5.7+0.1 5.8+0.3 5.8+ 0.3 5 .4+0 . 5 5.3+0 .2 4 . 8 + 0 . 2 5.5+0.1 5 .7+0 .2 6.0+0.3 TABLE XIH log k r tk^ 2/k- , C and B respectively. The limits of error are calculated at the 5% probability level. The units • 9 2 6c ii (6 >d) C H 3 C H 2 + [CH^CHI^CHXJ -» CH 2 - CH 0 + CHg-CHCHgX <6 >;e) CBLCH* + f C H C H r i C H X ] C H » C H , + • C H 0 C H » CHX (6 e») o / J A J 2 o This is in accord with this investigation in which the above reactions for diallyl were measurable and were found to contribute less than 1% in comparison to reaction (6). In any event, the neglect of not including these reactions in the kinetic scheme would be to increase the pre-exponential factor slightly. (2) Addition Reactions The site of addition of the ethyl radical to an allyl compound CHLCH* + CH C •- CHCI-I_X —> CH9CH_CH0CHCH0X; AH«= -23 kcal/mole (7) is remote from the substituent X, and the energetics of the process would be expected to be independent of the nature of X. It is therefore reasonable to expect a characteristic value for the energy of activation for the addition oi! the ethyl radical to the allylic double bond and this is shown to be the case as evidenced by the Arrhenius parameters in Table xm. The sets of Arrhenius parameters agree very closely for allyl alcohol, acetate and propionate, yielding E ? - 1/2 - 7.7 kcal/mole as a characteristic 12 value for these compounds. The reactivity of octei\e~l is very similar. Further-63 69 more, tills value is primarily an estimate of E^ since radical combination processes * 70 have an activation energy lying close to zero kcal. Shepp and Kutschke measured the combination reaction for two ethyl radicals: and found the following expression: . k 2 = 5.06 x IO - 1 0 exp(-2000 + 1000) /RT cm* mo!?1 gee?1 Absolute values of A and E can be calculated using the above Arrhenius parameters of k 2 > but this operation increases the probable errors in the values of A and E considerably. Therefore, for purposes of comparison, the relative values of A and E are preferable. Allyl ethyl ether and diallyl ether have slightly lower sets of values for the . 4 8 Arrhenius parameters; the reductions are not significant within the limits of error however, and are mutually compensating, with the result that the values of the rate constants lie close to the value for allyl alcohol. The Arrhenius parameters for diallyl are both significantly lower than those for the allyl compounds listed in Table Xin, but a decrease in both parameters has a mutually compensating effect, and the rate constants of the other allyl compounds. The lowering of the activation energy coupled with a diminished pre-exponential factor suggests a small mutually activating interaction between the allyl groups. This interaction is explained more fully in the 'Interpretation of the kinetic scheme" section of diallyl. Upon comparing the pre-exponential factors in Tables XII and XIII, it is seen that log A'7/Ag^2 is significantly greater than log A g/A^ 2 for all the allyl compounds studied in this investigation. This difference is reasonable since a greater shielding at a medial methylene group than at a terminal carbon atom, from attack by an ethyl radical, would be expected. The addition adduct radical can recombine with an ethyl radical given by reaction (7), or can undergo a dismutation reaction described in Part 3 of this section. Two ethyl radicals are consumed without the production of either butane or ethane, and the material balance falls below unity. If there is no extra ethane produced from the dismutation reaction, the addition rate constant shown below may be derived; otherwise the addition rate constant must be modified to account for the extra ethane. ' k = R_^ - R „ - R« TT 7 CO cnK„ C 172- 2 « ,;4 ~"> 2 W lVBiJ*V---4 10 The addition rate equations are listed at the end of the "Mechanism section" for each allyl compound. Again radical-radical reactions of the type similar to (6d) and (Ge) would be expected to be negligible also. 47 CH 3CH 2 + CH 3CH 2CH 2CHCH 2X -> CH 3CH 3 + CH 3CH 2CH= CHCHgX (7d)~ . CH 3CH 2 + CH 3CH 2CH 2CHCH 2X -•CHgCHg + CHgCHgCHgCH- CHX (7d') CH 3CH 2 + CH 3CH 2CH 2CHCH 2X -^CH2= CH 2+CH 3CH 2CH 2CH 2CH 2X (7e)7 This is further supported by the constancy of the kg/k2 values and the linearity of the Arrhenius plots. A high concentration of ethyl radicals was maintained in this investigation; otherwise the adduct radicals would recombine. The effect caused by the recombination-of the adduct radicals would be to raise the material balance, and hence give too low an addition rate constant. No lowering of the addition rate constant was observed in this study as seen by the Arrhenius; plots. (3) Dismutation Reactions The enthalpies of dismutation of the adduct radicals and CH 3CH 2CH 2CHCH 2X -* CHgCHgCHgCH = CH 2 + X" (8) CH 2 = CHCHX -» products (9) are governed by the nature of X . Hence considerable diversity is expected and found among the energies of activation, and these energies are related to the corresponding enthalpies listed in Table X I V . The activation energies for the above dismutation reactions are in the form E_ + l/2 E - E_ _ and E - l/2E - E , but since it is most o —• iC y —• be likely for radical combination processes that Eg = Eg c = E_ c = O, the activation energies are represented primarily by E Q and E n. With the exception of allyl alcohol, the values E_ are all slightly higher than the estimates of _JEL and this result suggests that the o o reverse reaction has a very small activation energy. In calculating the AH values for 71 ' the adduct radicals, the following relationship was used: D ( X - Y ) = AH ( X ' ) + _ ^ H ( ( Y ) .- A H ° ( X - Y ) In the case of allyl acetate, the value of (Eg = 15.4 + 1.3) kcal/mole allows a choice between two possible dismutation processes: CH3CH2CH2CHCH2OOCCH3->CH3CH CHgC^CHg + OOCCHg _\H=» 26 kcal/mole 48 — * CHgCHgCHgCH » CH 2 + COg + CHg, AH = 12 kcal/mole /. The activated complex clearly resembles the products of the second alternative much more closely, and hence the second alternative shown above is said to represent reaction (8). The second alternative would have a larger positive entropy change resulting in a lower free energy change, and therefore a greater driving force would favor this reaction. That no ethyl acetate was found in the products of the allyl acetate system is in accord with the above thermodynamic arguments. The resonance energy of the carbon dioxide molecule provides a large contribution to the facility of this dismutation reaction. A similar distinction for two possible dismutation reactions can be-made for allyl propionate where Eg » (18.2+2.6) kcal/mole CHgCHgCHgCHCHgOCCHgCHg —^CHgCHgCHgCKsCHg + OOCCgHg, AH=26 kcal/mole CH 3CH 2CH 2CH - CH 2 + COg + C 2 H G , 4 H - 12 kcal/mole Again, the dismutation reaction with the lowest enthalpy change is chosen to represent reaction (8) for the same reasons given above for allyl acetate. For the most part, reaction (0) is unimportant in the systems of the allyl compounds studied in this investigation. This is probably due to the loss in resonance energy of the allylic adduct radicals produced in reaction (6) thereby giving a relatively high endothermicity for the dismutatioa reaction. This high endothermicity is shown in the reactions below: £CH 2~€H^CHOHJ-^CH 2« C H C H O + H,' AHg - 4S kcal/mole \CH2L^CH^CH0C0CH3^'->CH2 ^ CHCHO + OCOCHg, £ H 0 - 29 kcal/mole However, reaction (9) is very important in the diallyl ether system, and this is hot surprising for this case since the reaction is exothermic. (cHg* CHCHgOCH l^Hi^CHg^'^g'H^ •=> CHCHO + jJcHg^-CH—CHg^' AHg - -4 kcal/mole The resonance energy of the allyl radical provides the driving force for this reaction and, in general for all the dismutation reactions, if there are any resonance stabilized products generated from the dismutation reaction, the reaction proceeds at a relatively fast rate. Thus, the dismutation reaction of the addition adduct radical can be 49 o measured in the allyl acetate system at 35 whereas dismutation is barely perceptible at 140° in the allyl alcohol system, and is not detectable in the octene-l system at 152 . . The rate expressions for the dismutation reactions are shown at the end of the "Mechanism section" for each allyl compound. B. The reactions of the Ethyl Radical with Hexadiene-1,5 (1) Mechanism Reactions (1) to (4c) described earlier in the introduction are supplemented by the following reactions when a mixture of diethyl ketone and hexadiene-1,5 is photolyzed by 3130 A radiation. CHgCHg + CH 2= CHCH 2CH 9CH= CE 2^CH 3CH 3 + {CH^'CH _:CHCH2CH=CH2"j' ( g ) CHgCHg + [cilg-CH^CHCHgCIfeCHgi^CH^CHCH (CH2GHg) CH2CI5=CH2 (6cT) — C H CHgCHgCBsCHCHgCE-CHg (6c") —> CH 0CH 0 + OL o CHCH=CHCE=CH0 (6d) ( cis and trans isomers) cis - CH- = CHCH= CHCH= CH 0 cyclohexadiene-1,3 (6i) CH 3CH 2 + CH 2= CHCH 2CH 2CH= CH 2—>CH 3CH 2CH 2CHCH 2CH 2CH=CH 2 (7) CHgCH^ .+ -CHgCHgCHgCKCHgCHgCHa CHg--> CH 3CH 2CH 2CH(CH 2CH 3)CH 2CH 2CH=CH 2 (7c) Reactions (6d) and (6i) were included to account for the formation in trace amounts of o o two C-HQ isomers at 91 , and of the same two isomers and cyclohexadiene-1,3 at 164 . The cis isomer of hexatriene-1,3,5 is knowato isomerize readily to cyclohexadiene-1,2 72 73 by both thermal and photochemical * processes. At 134°, appreciable quantities of pentene-1 begin to appear among the products, accompanied by much smaller amounts of allene and propene. Such products are con-sistent with the dismutation of the adduct radical formed in reaction (7) to yield the allyl radical and pentene-1; the ally} radical would be consumed exclusively by interaction with the ethyl radical, since the latter is present in great excess: 50 C H 8 C H 2 C H 2 C H C H 2 C H 2 C H = CHg-* CH CHgC^CH- CH 2 +(CH2-CHi:'CH23 ' (8) C H 3CH' 2 + [ C H ^ - CH^: CH2V-> C H 3 C H 2 C H 2 C H = CH 2 (8C) —» CH„CH + CH„= C* CH, (8d) —} CH 2= CH 2+ CH 3CH= CHg (8e) At 154°C, small amounts of butadiene-1,3, butene-1 and traces of acetylene begin to appear. These products are not formed in any other reaction in this system, and are consistent with the dismutation of the ce'-allyl radical formed in reactioa (6): C CH2Ui CH*^CHCH2CH « CH^ CHg = CHCH « CHg + CH = CH 2 (9) The vinyl radical should be consumed exclusively by interaction with the ethyl radical: CHgCH2 + CH = CHg CH 3CH 2CH= CH 2 (9c) —» CH 3CH 3+ CH= CH (9d) —>> 2 CH 2= CH 2 (9e) This mechanism is supported by a material balance among the products of the reactions involving the vinyl radical. Five experiments performed between 161° and 175° yielded the ratio: ( C 4 H G + C ^ / C 4 H g = 0. 97 + 0.01 The ethylene formed by reaction (9e) cannot be estimated directly, and the defect from unity of 0.03 + 0.01 in this ratio would arise either from the neglect of reaction (9e) or loss of the vinyl radical through the addition reaction. Hence the defect of 0.03 + 0.01 is an upper limit for reaction (9e) and may be compared to a value of 0.04 predicted 74 by the equation of Holroyd and Klein . The mutual disproportionationr combination ratio for ethyl radicals, k 3 A 2 , was found to be 0.137 + 0.017 for the diallyl system, in , : excellent agreement with the 12 recent value of 0.137 + 0.010 found in the octene-1 system ; this agreement further supports the above mechanism. It is concluded that the mechanism is adequately represented by the reactions (1) to (9e) given above, and the rate constants calculated from the above reactions are now shown: 51 k„ = i RC 2H 6 + R C 3 H 6 + 0.06 R C 4H 6 ) - f c 3H^ R C 2H^ BC 8H 4 ^ R C 6H ) f o 3 k 4 T72 ' r~l 1 7 5 PI*1'2 The symbols D and B signify the concentration of diethyl ketone and diallyl respectively, and R, r represents the rate of formation of product X. The term 0.06 R_ „ represents X " Q 4 H 6 the ethylene formed by reaction (9e). ^ / 2 - R c A + V 6 ^ V i o ) 1 / 2 k 7 c ^ C 0 + V 2 + V 8 ) + ± I C 3 H 4 ^ V 6 + V i e ! + V * * * W 1/2 R H (R H ) .1/2 _9fc2 = 4^6 ^4 10 * k t»c 1 1/2 < ; \ V J The patterns of disproportionation for the allyl, vinyl and ethyl radicals may be obtained from the equations. k 8 d = 2 R C 3 « 4 , 2 R C 3 " 6 k R TT +Rr H + TT k s R r H + R r H + Rc H ft C5H10 C 3 H 4 C 3 H 6 8 C5H10 C 3 M 4 L 3 n 6 2k- . hsr " R c 4 H s " R c 4 H 8 " R c 2 H 2 k 9 R C 4 H 6 5 V ' V e 52 (2) Interpretation of the Kinetic Scheme The results for the diallyl system are shown in Tables I, XII, xm, XIV and XV, and the corresponding Arrhenius plots are depicted in figures 1 and 2. .As can be seen from Tables XEL and X I I I , the activation energies of metathesis and addition for diallyl are both slightly but significantly lower for diallyl than for octene-1. However, there exist many examples in the literature whereby an inter-action between the double bonds of diallyl compounds in free radical reactions leads to a lowering in the activation energy, and it therefore seems reasonable to propose a similar interaction in the reactions of the ethyl radical with diallyl. This inter-action is seen in the radical polymerization of numerous non-conjugated diolefins which lead to the formation of recurring cyclic units by an alternating intramolecular-75 intermolecular mechanism . For instance, the heptadiene-1, 6 polymer chain com-prises a sequence of cyclohexane rings linked by methylene groups. Pronounced in-teraction between the double bonds of heptadiene-lj 6 has been postulated to explain 76 77 this tendency to form six-membered rings during the propagation process ' . ' A very pronounced lowering of the total activation energy of polymerization was ob-served when a methyl group was replaced by an allyl group in the compound 78 ally Itrime thy lsilane . Measurement of the rate of reaction (9) could be calculated only at the two highest temperature 164 and 175°; the corresponding values of ^ ~ ^ ^ ^ 2 / ^ Q C ARE 1/9 —3 /2 —1 /2 4.4 and 8.2 molecule cm" sec" respectively, and the ratio k Q k g A g k 7 c yielded the values 6.1 and 4.5 respectively, and since k g c would be expected to be approximately equal to k 7 c > the above ratio would show the rising importance of reaction (9) with increasing temperature. It is interesting to compare reaction (8), having an activation energy of 16kcal/mole, 79 with reaction (a) , having an activation energy of 23 kcal/mole. n - C 4 H 9 ' —> C 2H 4 + C 2H£ (a) Hence the difference in activation energies for the two similar dismutation reactions is approximately 7 kcal/mole, and this difference is most likely attributed to the resonance energy of the allyl radical. 23 A similar situation arises with the decomposition of the cyclopentyl radical , reaction (b), whereby the observed activation energy is 9 kcal/mole lower than the predicted activation energy using equation (a). d' — > C 2 H 4 + C 3 H 5 W However, this lowering of 9 kcal/mole in the activation energy is associated with two effects, the resonance energy of the allyl radical, and the strain energy of the cyclopentyl ring. Reaction (8) is very important in the diallyl system since it generates the allyl radical at relatively low temperatures. A low temperature is important since the estimation of the disproportionation processes (8d) and (8e) is based upon the assumption that the following reactions of the allyl radical with diallyl may be neglected; / C 3 H 5 + C 6 H 1 0 - > C 3 H 6 + C 6 H 9 <6a> C 3 H 5 + CS*1Q-* C 9 K I 5 The energies of activation for these reactions should exceed those for reactions (6) and (7) by an amount approximately equal to the delocalization energy of the allyl radical, D. As mentioned earlier in the introduction, P Lies in the range 15 to 20 kcal/mole, and the lower limit, 15 kcal/mole, is assumed for purposes of calculating the ratios kg a/k g and k7&/k_. The corresponding ratios k g a A g and k 7 Q A 7 will be -7 o given approximately by exp(-D/RT), which has the value 10 at 175 , the highest temperature of this investigation, and it is therefore concluded that reactions (6a) and (7a) are negligible ia the temperature range of this investigation. It is therefore assumed that the only source of propene in the system is reaction (8e), that allyl radicals are consumed entirely by reactions (8c), (8d) and (8e), and that the relative rates of these reactions may pcoperly be calculated from the appropriate equations given at the end of the "Mechanism section". 54 The results of Table I reveal the patterns of combination and disproportionation o 44 shown by the interaction of the ethyl radical and the allyl radical between 134 and 175 : fc H ^ C H i i i C H ^ + CH2CHg ~> ^5^10 (85+4)%; 90% (8c) CH 2 = C = CH .+ CH g CHg (4 + 1)%; 3% (8d) —} CHgCH* CH 2 + CH 2= CH 2 (11 +3)%;' 1% (8e) (Limits of error are given at the 5% probability level.) The values in square brackets shown above for the allyl radical and below for the vinyl radical have been estimated 74 by the equation of Holroyd and Klein : log ( k A ) = 0.131 ( S0,, - S° ) -5.47 0 x dis com' N dis com' Good agreement is found between the measured and predicted values. A few results at the higher temperature range, 161, to 175°, for the inter-action of the ethyl and vinyl radicals yeilded the following pattern of combination and disproportionation: CEL CBL +'CH»GHL -) C.H- (87+3)%; 94% ^ - (9c) -4 CH=CH+ CH„CH_ (10+3)%; 2% (9d) h 2 CH = CH_ ( 3 + 1)%.; 4% (9e) ' 2 2 ~" The predomination of the combination reaction is seen from both the predicted and the measured values, but the agreement between the predicted and measured values for the disproportionation reaction is poor in this case. Disproportionation between allyl and ethyl radicals is clearly not negligibly small. Disproportionation between «£allyl allyl radicals and ethyl radicals was also detected, as mentioned earlier, and the reactions are shown below; the primary products were identified as the cis and trans isomers of hexatrienerl, 3,5 by comparing their 8 0 retention times with those of the authentic isomers . Two other readily identifiable compounds, benzene and cyclohexadiene-1,4 were not found amongst the CQ products. CHgCH^ + (CH 2— CH^'CHCH2CH » C H J -) CH^CH^ + trans -CK2=CHCH=CHCK=CH2 CH 0CH 0 + cis -CH, = CHCH=CHCH=CH0 72 73 of these two isomers, only cis form isomerizes to cyclohexadiene-1,3 * 55 cis - CH = CHCH= CHCH= CH0-7cycl6 - C_HQ -1,3 Z Z o o 81 Recently, the isomerization of the substituted diallyl compound, 3-methyl-hexadiene-1, 5 to cycloheptadiene-1,5 has also been observed. The percentage of the <<• -allyl allyl radicals to undergo disproportionation is equal to 100 R^ „ /Rc» which has the value 0.41 at 91° and 0.40 at 164°. Clearly, disproportionation is far less important for the oc-allyl allyl radical than for the allyl radical. This low per-centage provides more support that in general, disproportionation reactions in-volving the product radicals of metathesis and addition can be neglected. This trend was discussed earlier in section A. The interactions of the allyl and ethyl radicals may be compared with the corresponding reactions of the allyl and methyl radicals: CCH2- CH;:,CH2T+ CH3->CH2= CHCH2CH3 (C) ->CH2 = C= CH 2+ CH 4 (cO 36 82 Unfortunately allene was not identified in two recent studies ' of the reactions of the methyl radical with propene thereby producing allyl radicals; however the concentration of allyl radicals in reference 36 was very low. Nevertheless allene has been shown to 83 be a significant product of the mercury photosensitized decomposition of propene , 84 85 butene-2 and pentene-1 . The propene and butene-2 systems have been studied in the presence of a large excess of methyl-d3 radicals, which act as efficient scavengers of the primary radical species. Allene and l-butene-4-dQ were found among the products in the ratio of 0.06 in the propene system and of 0.14 in the butene-2 system. These results show that allyl radicals are formed in the primary process, and the ratios may be taken as an estimate of k^/k, appropriate to the particular conditions of each experiment. Such estimates are high in comparison with the 74 values of 0. 023 predicted by the equation of Holroyd and Klein , but this difference is not unexpected. The interaction of propyl and methyl-dg radicals has been studied 86 under parallel conditions , and the corresponding value 56 of k^A^4estimated as 0.5, which is three times as great as the accepted value of 0 . 1 7 . A propyl radical formed by the transfer of the triplet state energy of the mercury atom to a propene molecule should possess several quanta of vibrational energy initially. Disproportionation may occur before deactivation is complete, leading to the abnormally high values^'®® of k ( jA ( , . If similar arguments are applied to the estimates of k^A given above for the allyl radical, we may conclude that the predicted value of 0 .023 is of a reasonable order of magnitude for the inter-action of thermally equilibrated allyj and methyl radicals. Ey means of equation (a) shown below the degree of polymerization in the liquid phase given by the expression ^ /(k^k^) is related to the corresponding expression k^y^k^k^ in the gas phase discussed previously in section A of the discussion section, are the analogous of those in the liquid phase, shown earlier in the introduction, k , k , and kf respectively. Equation (a) is a modified Hammett O I I 87 ' equation formulated by Bamford, Jenkins and Johnston and is now given: log k = log k. * CC ' • er O f - «i - «t> « ° CHgCHg + CHg = CHCHgOCHgCH (5) CHgCHg + CH 2 = CHCH2OCH9CH2-* CH 2 » CHCHgOCHgC^CHgCHg (5c) CH0CK^+ C H 3 » CHCHgOCHgCHg^CHgCHg+fcH^-CH^HOC^CHgJ (6) CHgCHj +|CH£: CHOCHOCHgCHgYXJHg = CHCH(CH2CH3> OCHgCHg (6c») -?CHgCH9CH0CH= CHOCHgCHg (6c") CHgCH2 + CK 2 = CHCH2OCH2CHg ^^HgC^CHgCHC^OCHgCHg (7) CHgCH^ + CH 3CH 2CH 2CHCH 2OCH 2CHg -9CHgCH2CH2CH(CH2CHg) CH 2OCH 2CH 3 (7c) In the allyl ethyl ether system, appreciable quantities of pentene-1, diethyl ether and ethanol begin to appear among the products at 83°. These products are consistent with the dismutation of the adduct radical formed in reaction (7) to yield pentene-1 and the ethoxy radical. Subsequently, the ethoxy radical undergoes combination and disproportion-ation reactions with the ethylradical which is present in great excess over all the radicals in this'system. CH 3CH 2CH 2CHCH 2OCH 2CH 3 -^CHgCIL^CHgCH = CH 2 + OCHgCHg; AH = 21 kcal (8) 60 CH 3 CH 2 + OCHgCHg C K 3 C H 2 G C H 2 C H 3 (8c) — ? CHgCHg + 'CHgCHG (Sd) -9 C H 2 = C H 2 -:- CHgCHgOH (8e) The above mechanism is given support by a material balance of the products re-sulting from the ethoxy radicals.. Seven experiments performed between the o o temperatures 83 and 132 Cyielded the ratio. F 1 = (CHgCHgOCHgCHg + CHgCH^E) / = 0. 97 + 0. 06 The defect of 0.03 in this ratio from unity might be accounted for by reaction (Sd); however no acetaldehyde was observed in the products. -Another possible fate that'might account for this defect would be an addition reaction, (7m), analogous to reaction (7). •CBLCH„d+ CH = CHCE o 0CH o CH. —> CH 0 CH 0 OCH 0 CHCH 0 OCH 0 CH_ (7m) The results from the above seven experiments also yielded die fraction of.ethoxy radicals to escape combination with ethyl radicals as: V< C5^0 - C VV C ^ C V C5 Bi0"« [ ) ^0t0.08) The following values listed in Table U yield the disproportionation and combination pattern of the ethoxy radical. The values in square brackets are the calculated 74 values using the equation of Holroyd and Klein ; the values in curly brackets re-89 present the results of Wijnen , and the values in round brackets represent the results of this investigation. CH 3 CH 2 + OCI-I2CH3 CHgCH OCHgCHg (80 +• 8%) [67%]{22%] ' ' (8c) —? CHgCHg+ CHgCHO (3 + 2%)[l3%3|8%} (8d) —? C H 2 = C H 2 + CH 3 CH 2 OH (17 + 4%) [20?o! (8e) The predomination of the combination of ethyl and ethoxy radicals shown in this study 74 is iu agreement with the derived values from the equation of Holroyd and Klein . The 89 results of 'v'ijnen suggest a much stronger trend towards disproportionation between the.above radicals. The results obtained for the interaction of the ethoxy radical with the ethyl radical were obtained below 132^ , since above this temperature, F« rises noticeably indicating 61 that the ethoxy radical is undergoing metathesis with allyl ethyl ether and diethyl ketone. The dismutation reactions of the product radicals produced in reactions (5) and (6) are excluded since the products, acrolein and allyl alcohol, produced from these reactions, the latter product being formed in a disproportionation reaction between allyloxy and ethyl radicals, were found to be absent. ^CH2'^ CH'HCHOCH2CH3}->CH2 = CHCHO + CHgCHg; AH= 16 kcal/mole (9) CH 2 = CHCH2OCH2CH2-> CH 2 = CHCH20'+ CHg CH 2 £H = 15 kcal/mole (10) The other adduct radical produced in reaction (5') did not undergo a dismutation reaction since the products pentene-1, propene and allene generated by the interaction of the allyl radical with the ethyl radicals, and acetaldehyde formed directly were also found to be absent. As reaction (10') is considerably exothermic, it is inferred that reaction (5') would be negligible also. Further, since reactions (5) and (5') are similar, reaction (5) might also be expected to be unimportant in comparison to reaction (6). This assumption is confirmed by comparing the Arrhenius parameters for diallyl ether, in Table XII and is discussed more fully after the "Mechanism section" of diallyl ether. The above mechanism is supported by the value of kg/k2 given as 0.137 + 0.015 12 in good agreement with the value of 0.137 + 0.010 in the octene-1 system . Thus reactions (1) to (8e) represent adequately the mechanism for the allyl ethyl ether system, and the rate expressions are now shown. (5') (10') '<2H6+ C 2 H 5 O H J 0.03 It I?] The term 0. 03 R„ „ represents the ethane formed by reaction (8d) C5 H10 , 62 | a t t l t l l - M , C , g 1 0 ' - P c A * \ H l t » ' 1 - M " E C O t » i B C 5 H l a > ^4 10 ° 4 10 where M* = (R + R ) / (R + 0.03 R „ ) ° 2 t l 6 C4M10 C O C5H10 k7c R C O " ( R C H + R C H + 0 - 8 7 ; a C H > • < C C ° C 2 H 6 V ^ O ° 5 H 1 0 The disproportionation-combination ratio for ethyl radicals i3 given by: k 3 / k 2 » ( R C 2 H 4 " ^ HgCHgO^ / R C 4 H 1 0 (lb) Mechanism: " Diallyl Ether" Diallyl ether is represented by reactions (1) to (7c), analogous to the allyl ethyl ether system, up to 83^with the exclusion of reactions (5) and (5c) since ouly allylic hydrogens are present in the case of diallyl ether. The results for diallyl ether are shown in Tables IH, XII, XIII, XIV and XV, and the Arrhenius plots are given in figures 3,4 and 5. At 104^tlie following new products, pentene-1, allyl ethyl ether, acrolein, propene and allene are observed. The appearance of pentene-1 and allyl ethyl ether suggest that the decomposition of the addition adduct radical is taking place at this temperature. The subsequent reactions of the allyloxy radical with the ethyl radical, which is in great excess over all other radicals, yield allyl ethyl ether, acrolein and allyl alcohol as products: CHgCHgCHgCHCHgCCH^H = CHg CHgCHgCHgCH = CHg + OCH2CH=CH2 .AH = 16 kcal/mole CH 3CH 2 + OCH2CH= CHg-^CHgCHgOCHgCH-CH2 (80%) — A CELCH 0 + CH CHCHO ( 3 % ) —5> CH 0= CH 2+ CH 2= CHCHgOH (17%) The disproportionation and combination reactions of the allyloxy radical with the ethyl radical could not be measured in this system for the following reasons. The 63 two products, acrolein and allyl alcohol, identified with the disproportionation reactions (8d) and Se) are obscured since acrolein is produced in large quantities from reaction (9) given below, and the chromatographic peak for ally! alcohol is hidden by the ex-tremely: large peak of one of the reagents, diethyl ketone, Also, the large quantities of pentene-1 produced in reaction (Sc) make it impossible to measure the combination reaction (8c) with any degree of accuracy. The relative rates of reactions (8c), (8d) and (8e) have been assumed to be the same for the allyloxy radical as for the ethoxy radical, and these values are given in brackets in the above reactions. Using these values, reaction (8) may be calculated in the from Rate_ = 1. 25 Rate , . . J 8 allyl ethyl ether In the study of allyl alcohol, the value found for the ratio k 0 ,/k^ compared favourably d u oC with the assumed value involving the ethoxy radical thereby supporting the above assumptions. The presence of acrolein, propene, allene and pentene-1 suggest that the adduct radical of the metathesis reaction is readily decomposing to acrolein; and allyl radicals. The driving force of the resonance stabilized allyl radical enhances reaction (9) C H , * CHCHOCH f tCH« CH„^rCHJ^.CHii .CH s J+ CE =CHCHQ (8) A H = -4 kcal/mole The resonance stabilization in the allyl radical leads to a much lower enthalpy for this reaction than for reaction (8), and it is seen from the experimental results in Table XIV that reaction (S) is approximately one hundred times faster than reaction (8). The allyl radical is then consumed by interaction with the high concentration of ethyl radicals. .CHgCHg +^I - I 2 ^ - * C H i i C H 2l - ^ C H 3 C H 2 C H 2 C H = C E 2 (9c) CH 3C1' 3 + C H 2 = C = C H 2 (9d) —^ CH 2 = C H 2 + C H 3 C H = C H 2 (9e) At the highest temperature of this study, 15o£, a trace of diallyl is formed in-dicating that a small fraction of me allyl radicals have escaped interaction with the ethyl radicals, and have undergone mutual combination. . 64 2[CH2— CH-CH 9"i - ^ C H 2 - C H C H 2CH 9 C H = C H 2 (9m) Thus the equations (1) to (9m) are proposed to constitute the mechanism for the diallyl system over the temperature range 63»15s£. The validity of this mechanism can be tested in several ways. A material balance involving the products of reactions (8) and (9) Is predicted and is now shown: R C 5 H 1 0 - R 8 + R 9 C - R 8 + R 9 - ^ C 3 H 4 + RCgH f i + 2 <*> R C 2 H 3 C H O = R 9 + K 8 d = R 8 + X i9 " { R C 3 H 5 C C 2 H 5 + . ^ H g O H * 1 * 2 1 R C g H OCgBL^ (b) From (a) and (b), the following expression should be equal to unity if the reactions above represent the mechanism of the diallyl systems: The observed value, Q = 0.96 + 0.06, therefore supports the proposed mechanism, The relative rates of (9c), (9d) and (9e) can be expressed by the expressions shown below: Fd = V Sc + k9d + k9e> = R C 3 H / * S9 c Similarly, F = R T /S and F - R „ / S 0 . The value for SQ may be calculated e h i y C U 5 T.0 S J from the equations above in two ways: Sl" R C 2 H 3 C H O - ° - 0 4 R C 3 H 6 O C 2 H 5 " W - C A o . and sn 9 = S\o + \h + " 1 - 2 6 V s 0 C 2 H S i n The two values Sg and SQ should be equal to one another if the proposed mechanism i 3 correct, and this is shov/n to be the case as evidenced by the values of F , F .and F C . CI © calculated with each value of S g listed in Table III in the "Results section", whereby the differences between members of a pair is negligible. The mean values may be 4 4 compared to the values obtained in the diallyl system and the latter values are now given in square brackets: F c = 0.86+ 0.06 0.85+ 0 . 0 4 ; F d = 0.05+ 0.02 0.04 + 0 . 01; and F g = 0. 09 + 0.03 0.11 + 0.03. Reasonable agreement is obtained in each case. From the following 65 equation kg/kg c a u ^ e calculated: K/K =(Rn H -Rn w " 0.21 R „ „)/R„ „ • 3 2 C2 H4 C3 H6 C 3 H 5 ° C 2 H 5 CAo and the mean value Is 0.138 + 0.013, in good agreement with the value of 0.137 + 0.010 obtained in the octene-1 system. The above experimental findings lend good support for the proposed mechanism, and now the rate equations based on the above mechanism are listed: ' k ? . ^ C O + R C 3 H 4 , 0 ^ ^ C 3 H 5 O C 2 H 5 ) " %*J W2 ^ ^ K p l (1 - M«) CR C Q f + 0.04 \n5oc2E) 4^0 where M* . ( R ^ . + P ^ ^ J / / 2 A 7 C » 1 0 ( 9 - 4 - 1 , 7 ) exp (-18.8+3.3) 103/RT = 0.44+ 0.06 at 150°£ 1 0 " 5 k 9 k ^ / 2 A 6 C = 10 < 1 : L* 9i 0- 5 )exp (-19.6+ 1.0) 103/RT= 55+ 7 at 150°C As can be seen from the above values, the rate constants for addition and metathesis are readily interpreted in terms of the number of ally! groups per molecule; thus the corresponding rate constants for diallyl ether and allyl ethyl ether stand in the ratio of approximately 2:1 at 150$. The values for the energy of activation for metathesis are almost identical for the two ethers, indicating that metathesis is occurring predominantly at the activated methylene group in the case of allyl ethyl ether. Thus reaction (5) contributes to a very small extent over the temperature range studied for allyl ethyl ether. The pre-exponential factors are significantly lower for metathesis than for addition, indicating effective shielding of" thee^ -methylene group from attack of the ethyl radical by the other substituent group. In contrast, the pattern of dismutation is quite different for the two ethers, reaction (8) being predominant for allyl ethyl ether whereas reaction (9) is favoured for diallyl ether.. This finding is reflected in the difference of enthalpies for the two reactions of diallyl ether$HQ -#1 = 20 kcal/mole. It is interesting to note that the b y energies of activation for reaction (8) are slightly greater than the estimated en-thalpy change: for allyl ethyl ether,01 = 21 and E = 22 kcal/mole; and for diallyl 67 ether^Ig = 16 and E g = 10 kcal/mole. 48 Reaction (8) was proposed in 1956 for allyl ethyl ether in order to explain the low degree of polymerization at 4.0 at 80°C. The experimental findings in this investigation show that reaction (8) for allyl ethyl ether proceeds at a slow rate at 80°C, but plays an important role in lowering the degree of polymerization at temperatures greater than 120°C. The values for the degrees of polymerization, for the two ethers are shown in Table XV. Thus kg is the controlling factor in the degree of polymerization for the two ethers, and is much faster than the corresponding, values of kg measured for the allyl esters at any given temperatures. In particular, kg could not be measured for allyl ecetate, but was just measurable following the rate of CgHgD produced in the allyl-1 l - d 2 acetate system. This finding is in accord with the low degree of polymerization found for allyl acetate, and is discussed more fully in the next section. 68 O CD CO OJ Z>. O Q) O E O J ro E o O J ^ O J G P _ o i - r ro 2.0 10 V T ° K Figure 3 Addition of the ethyl radical to the allyl ethers. Circles:- diallyl ether Triangles:- allyl ethyl ether 69 Metathesis between the ethyl radical and the allyl ethers. Circles:- diallyl ether Triangles:- allyl ethyl ether 70 CVJ Figure 5 Dismutation of the addition and metathesis adduct r a d i c a l s of the a l l y l ethers. Solid Circles:- diallyl ether Open Circles:- diallyl ether Triangles:- allyl ethyl ether 71 D. The Reactions of the Ethyl Radical with Allyl Acetate and its Deuterated Analogues 1. Mechanism Reactions (1) to (4c) are supplemented by reactions (5) to (7c) in the allyl acetate system: CHgCH2 + CH 2 » CHCH2OCOCH3 CKgCHg^HCHCHgOCOCH (5) CHgCH2 + CH 2 » CHCHgCCOCHg CH 2 = CHC^CCGCH^CRgCRg (5c) CHgCRg + CH 2 = CHCH2OCCCH3 ^CHgCHg +fcH 2- CH^CHQCOCHg)" (6) CHgCHg + tCHjp CH—CHGCGCHg]-^ CB^ « CHCH(C Hg dig) OCOCHg (6c») —? CHgCH2CH2CH= CHOCOCHg (6c") CHgCH2 + CHg = CHCH2OCOCHg -5 CHgC^C^CHCHgOCOCHg (7) CHgCHg + CH3CH2CH2CI-ICH2OCGCHg-fCH3CH2CH2 CH(CH2CHg) C H2OCOCHg (7c) However, reactions (1) to (7c) fail to describe the reaction mechanism adequately. Even at 35°C> pentene-1, carbon dioxide, propane and a trace of methane are present among the products, and up to 100^ they are formed in equivalent amounts: CJ-H^ Q = CG„ = (CH. + C0H_). This equivalence supports the dismutation of the adduct radical of reaction (7): CHgCH2CH2CHCH2GCOCH3—>CH3CH2CH2CH= CH 2 + C0 2 + CIL^AB* 12 kcal/mole (8) followed by the capture of the methyl radicals so formed by the great excess of ethyl radicals; CH 3 + CH 2 CHg -? CHgCH2CH3 (8c) >.CH 4+CH 2=CH 2 (8e) A measure of the excess of the ethyl radical is given by the quotient: Q s 3 R C 3 H 8 / R C 4 H 1 0 I C H ^ H 2 t H 3 ! l Above 100°C butene-1 begins to appear as a minor product in amounts which in-crease rapidly with temperature. A significant proportion of the methyl radicals appear tc escape capture by the ethyl radicals and instead participate in the sequence of addition and dismutation: CH 3 + CH 2 = CHCH2OCOCH3-7> CHgCHgCHCHgOCGCHg (7m) CHgCHgCHCHgCCOCEg ~v CHgCHgCH = CHg + C0 2 + CHg 72<8ra) which are the methyl radical analogues of reactions (7) and (8). The ratio RCH./RC0H0 also rises with temperature, and we may infer that the extra methane 4 o o is formed by the methyl radical analogues of reactions (4), (5) and (6) followed by (4c), (5c) and (6c) respectively. This reaction scheme is supported by the product ratios; M,* R„ w • ' ' « 1.01+ 0.03; R C H 4 + R C 3 H 8 M 2 ^ R C 0 2 where R„ denotes the rate of formation of the product Z, and the limits of eror are given here and elsehwere at the 5% probability level. Furthermore, the values of the corresponding ratios for the deuterated allyl acetates which would undergo analogous reactions as the undeuterated allyl acetate, were found to be: M l = RCH 0CH.CH 0CH= CD 0 = 1.02 + 0.05; R C H 4 + R C 3 H 8 M = R „ ^ =1.02+0.03 2 C Q 2 R C 5 H 8 D 2 + R C H 3 C H 2 C H 2 C H = C D 2 and for allyl acetate dg, M, = R„ „ = 0.99+ 0.07; 5H10 RCD SH+ RCD 3CH 2CH 3 M 2 ^ R C 0 o =1.01+ 0.03 H c ^ + ^ D g C ^ C H ^ C ^ By using the labelled allyl acetates, reactions (5) and (6) can be distinguished and measured. Hence in the allyl acetate-dg system, the corresponding reactions are: 73 CHgCH* + CH 2 = CHCH2OCOCD3-?> CH 3CH 2D + CH 2 = CHCHgOCOCDg (5D) —$ CH gCH 3 +(CH2—CH^'CH CCOCDg")" (6H) The metathetical reactions of the allyl-1, l - d 2 acetate system are: CH 3CH 2 + CH 2 * CHCD2OCOCH3 -^CB^CF^ + CH 9 = CHCDgOCOCHg (5H) —J> CH 3CH 2D + t c i y — CH~CDCCOCH 3 3' (6D) The values of 0.131 + 0.005 for allyl acetate, 0.126 + 0.006 for allyl acetate-d3 and 0.131 + 0.003 for allyl-1. l - d 2 acetate are in good agreement with the value of 12 0.137 + 0.010 obtained in the octene-1 system , and therefore give more justification for the above mechanism. The' rate equations in the allyl acetate system are now given: i , i rv , 0.06 R_ „ ) - R_ „ , k ? , R C O - ( R C 2 H 8 + R C 4 H T 0 > - ^ RCO T72 r-ci/n , T i 7 2 rSZUT k f / 2 [B1 2 C 4 H 1 0 J-A c 4 i i 1 0 k k l / 2 R C H 1 / 2 k 8 R 2 = C5 1 110 C4 M10 k7c R C O - V 6 + - \ H 1 0 + S C 5 H 1 0 ) In the allyl acetaterdg system, the two metathesis rate equations are given by: ^ = RCH 3CH 2D . k6H 2 = ^ C 2 H 6 + 0 ^ 6 R C H 5 D 3 > - R C 2 H 4 ' N 1 ^ ^4^10 4 10 and similarly in the ally 1-1, l - d 2 acetate system, the corresponding rate equations are: hn - ' V ^ 0 - 0 8 ^ ^ ) - ^ '[Di k4 1/2; k6D . RCH 3CH 2D ^ M 1 / 2 T B ] C B H 4H 1 0> 1 / 2 The symbolslpjand^BJ denote the concentration of diethyl ketone and allyl acetate respectively, R_ represents the rate of formation of the product Z , and M P R ^ W + Z ^2 6 R^ „ / E„„. The term 0.06 R „ „ is an estimate of RQ_, based upon the value of C 4H 1 0 CO C 3H g 8c 0.06 for the disproportionation-combiaation ratio k A . 8e 8c . 2. Interpretation of the kinetic scheme. The sets of results for either addition or dismutation for allyl acetate and its two deuterated analogous were.combined and treated as one group since these sets yielded Arrheaius parameters which were statistically indistinguisable between themselves. All limits of error were calculated at the 5% probability level. Thus, the nine values for the addition rate constant of allyl-1^ acetate, the sixteen values of allyl acetate-dg and the eighteen values of allyl acetate were combined to give one group of forty-three values for purposes of the statistical analyses. The results are listed in Tables IV, V, VI, XHS XJH, XIV. and XV and the Arrhenius plots are shown in figures 6, 7, and 8. As stated before under "Addition reactions in general," the addition reactions of allyl acetate gave Arrhenius parameters charac-teristic of the majority of allyl compounds, and no further comment is given. However, .the pattern of metathesis mentioned-earlier under "Metathesis reactions in geaeral" is now discussed more fully. The relative activation of the ex* -methylene group is evident from the difference in activation energies: Eg^ - 2.8 + 1.7 kcal/mole, and from the expression: log ( k ^ A ^ = 0.95 + 0.14, indicating that „ o almost 90% of the metathesis occurs at the oC -methylene group at 100 .C. The difference in activation energy reflects an estimated difference in reaction enthalpy, A H g H -sdHgg » 23 kcal/mole, and is a consequence o f the formation in reaction (6D) of a substituted allyl radical with a delocalisatioa energy of about 20 kcal/mole. It was just shown that kg was the predominant reaction in the metathesis reaction o f allyl acetate, and the same, tread is found in the allyl alcohol system. It is -therefore interesting to relate the relative reactivity of allyl acetate with its saponified analogue, allyl alcohol, towards the metathetical reaction (6), since these monomers show a much different response towards radical polymerisation in the liquid phase. For purposes of comparing the reactivity of reaction (6) for the monomers, the following expression is used: p6 = k s { allyl alcohol) / k g ( allyl acetate) 75 The above rate constants refer to the same temperature. " P 6D " c a u b e c a l c u l a t e d directly, but " P 6 H " must be extrapolated by means of the deuterium primary isotope effect. The deuterium isotope effect has been studied 91 92 93 by many workers, and several review articles. ' ' have been written oa this topic. Both quantum mechanics and classical rate theory have been used to explain the deuterium isotope effect, and the classical theory incorporating zero point energies has proven very successful in the majority of cases. A few cases arise whereby the predicted primary isotope effect given by the equation below is either much lower or greater^' ^ than the obseryed value. The lower values*** could occur in the case of a linear complex A... H. B where hydrogen is more strongly bounded to atom A than to Atom B. This has the effect of producing a difference in Zero point energy in the activated complex and this contribution partially co.neel3 the difference in zero point energy in the reactants. Many workers** 8'* 0 0' 1^ > 1 have used quantum mechanical treatment in which the 102 tunneling effect is used to explain the high values of kg/k^. Recently, Solomon has explained these high values by taking into consideration the vibrational bending modes. 101 Using absolute rate theory , and making the assumption that the potential energy surface will be essentially the same for a hydrogen compound and its deuterium analogue, and that the nonreacting bonds in the molecule are not affected during reaction, the following equation may be derived: kH - eh<>H -n>>/2RT From the above equation it is seen that the deuterium isotope effect is determined by the difference in zero point energies between the bond to hydrogen and the corres-ponding bond the deuterium. It is interesting to note that differences in zero point 93 energies lie in a narrow range of 1.1 to 1.5 kcal/mole for three dissimilar bonds, C-H, N-H and O-H. The above equation is used in order to convert k^ values of allyl-1, l - d 2 acetate 76 to the corresponding k „ values. Two values of P _ were found: P__ = 2.4 at 105°C xl bJL> bJJ and Pgjj =s 2.7 at 138°C; these values represent a significant difference in the re-activity of the^-methylene-dg groups in these two monomers. The value for Pg^ using the concept of zero point energies was determined to be 2.4 at 80°C reflecting the greater reactivity of k g H for allyl alcohol as compared to kgjj allyl acetate. This difference in reactivity does not appear to arise from the activation energies since the energy of activation for allyl alcohol, Egjj=» 6. 0 kcal/mole is slightly higher than the corresponding value for allyl acetate, 4. 7 kcal/mole. In the case of allyl acetate, E g H was extrapolated by subtracting the zero point energy difference betwee the C-H and C-D bonds, approximately equal to 1.1 kcal/mole, from the value ob-tained for E g D ugjng allyl-1, l - d 2 acetate. Hence the greater reactivity of allyl alcohol as compared to allyl acetate towards reaction (6) must be associated with the pre-exponential factors. Therefore a large pre-exponential factor is associated with allyl alcohol as compared to that of allyl acetate since the site of reaction is at the methylene group, adjacent to the substituent group and the smaller substituent group of allyl alcohol would provide a smaller degree of shielding of the reactive methylene group against attack by the ethyl radical. The dismutation reaction was elaborated upon in the section discussing "dismutation reactions in general". Allyl acetate and allyl alcohol differ considerably in their response to attempts to 3 initiate homogeneous free radical homopolymerization in the monomers . The results given for allyl alcohol and allyl acetate in Table XHI imply that the propagation rate constants for these monomers are very similar. An inspection of Table XII shows that for allyl alcohol abstraction occurs about six times as frequently from the methylene group as from the hydroxyl group at 100°C, and in the case of allyl acetate, abstraction occurs about ten times as frequently from the methylene group as from the acetoxy group at 100°C. The striking difference in the response of these two monomers towards polymerization should therefore be Linked with a difference in degradative chain transfer constants, that for allyl alcohol being the larger. The ratio of the 77 transfer constants is given by the ratio of rate constants of reaction (6H) in the gas phase, which was estimated before as 2.4 at 80°C. This ratio is certainly not large, but the kinetics of allyl polymerization characteristically turn upon a delicate balance between the propagation and degradative chain transfer processes. The factor, Fgjj, showed little temperature dependence since the greater reactivity of allyl alcohol as compared to allyl acetate was shown to arise from a higher pre-exponential factor. From the above statements, at 80°C, the most pre-dominant mode of chain transfer for both allyl alcohol and allyl acetate is degradative chain transfer , and it was shown that allyl alcohol was more reactive than allyl acetate towards reaction (6) by a factor of 2.4. This same factor appears in the degrees of polymerization for these two monomers in the liquid phase and is consistent with the results found in the investigation in the gas phase. If the degree of poly-merization for allyl-1, l - d 2 acetate is approximated by k^/kg^, thenthe calculated value is 21 + 5 as compared to a value of 35 6 4 obtained in the liquid phase at 80°C. Using the extrapolated values of k 5 H and kgjj obtained by means of zero point energies, the degree of polymerization for allyl acetate was calculated to be 4.4 compared to a value.of 14 5 3' 6^in the liquid phase at 80°C. 3 . . . The degree of polymerization for allyl alcohol represented by k ?/(kg^ + kgH) has a value of 1. 8 at 80°C. This value may be compared with that found in the liquid phase 66 67 for allyl alcohol where the degree of polymerization equals 5 ' . However, in the 66 latter system, frequent replenishment of the catalyst, hydrogen peroxide , over a 103 long period of time, 116 hours, was necessary so that reactivation of the "dead" polymer by metathesis with the hydroxyl radicals would be expected and the measured 104 degree of polymerization should be lower. It is interesting to note that Staudinger originally believed that no polymer could be obtained from allyl alcohol. Although the values for the degree of polymerization in the gas phase are always lower by approxi-mately the same amount than those in the liquid phase, the trend in; either phase is the same. This linear correlation between the two phases was observed 2.0 I 0 V T ° K Figure 6 Addition of the ethyl radical to allyl acetate substrates. Open Circles:- allyl acetate -d„ O Closed Circles:- allyl acetate Triangles: - allyl -1, l - d 2 a c e t a t e 79 cv! O CD 0 0 "5 OJ + ro 0 . 5 0 . 0 1 . 5 1 . 0 2 . 5 CHCD20C0CH3 CH 2 =CHCH 2 0C0CD 3 2 . 2 2 . 4 2 . 6 I 0 ° / T ° K 2 . 8 Figure 7 Metathesis between the ethyl radical and deuterated allyl acetate substrates 80 2 . 0 CM O C D o o i E o o o ^ CO 00 o I O V T ° K • Figure 8 Dismutation of the addition adduct radical of allyl acetate Open Circles:- allyl acetate -d„ Closed Circles:- allyl acetate 81 for all the allyl substrates studied in this investigation is shown in Table XV. E. The Reactions of the Ethyl Radical with Allyl Alcohol and its Deuterated Analogues 1. Mechanism Reactions (1) to (4c) are supplemented by reactions (5) to (7d) in the allyl alchol system: CH 3CH 2+CH 2= CHCH 2OH-^CHgCH 3+CH 2= CHCH 20 # . (5) CHgCHg + CHg - CHCHgO* CHg » CHCHgOCHgCHg (5c) —> CHgCHg + CH 2 o CHCHO (5d) — ? CHg » CH 2 + GH 2 » CHCHgOH (5e) CHgCHg + CH 2= CHCHgOH --^CHgCHg + f C H ^ CH^CEOH3 (6) CHgCHg + [CH^~ CH^CHDHj'-^CHg = CHCH(CH2CHg)OH (6cf) CHgCH 2CH 2CH«= CHOH (6c") CHgCHgCHgCHgCHO —> CHgCHg + CH 2 = CHCHO (6d) —9 CH 2= CH 9+ CH 2= CHCHgOH (6e) CHICHI + CH = CHCH-OH —p CH nCH 0CH 0CHCH 0OH • (7) CHgCHg * CHgCH2CH2CHCH2Oa —? CHgCHgCHgCH(CHgCHg) CHgOH (7c) -—•> CH 0CH Q + CH CH„CH_CH= CEOH (7d) CHgCH2CH2CH2CHO Reactions (5d), (6d) and (7d) each yield ethane but the rate of formation e£ such ethane is shown to be extremely small and is not included in the calculation of the rate constants for reactions ( 5 ) , (6), (7) and (8). Reactions (5d> and (6d) also yield an equiv-o alent amount of acrolein, whereas below 141 C acrolein did not occur among the products in greater than trace amounts. The disproportionation combination ratio kg^ A g c was determined to be less than 0.04, therefore reactions (5d) and (6d) are regarded as kinetically negligible. The presence of the reagents prevented an effective search for pentanal among the products, and eo no direct estimate of the rate of reactions (7d) is possible. This reaction is treated as negligible as it is unlikely 82 to be appreciably faster than the closely related reactions (5d) and (8d). Again reactions (5) and (S) can be distinguished using the deuterated substrates, allyl-l,l-d 2 alcohol and allyl alcohcl-d. The reactions for allyl alcohol-d are: CH 3CH 2+ C H 2 = CHCH 2OD^CH 3CH 2D+CH 2= C H C H 2 C ' (5) CH 3CH 2 + CH 2 = CHCH2OD-j>CH3CH3 +[CH2^ CH£: CKOD)' (6) and the corresponding reactions for allyl-1.1-d alcohol are-. a CH gCH 2 + CH 2 = CHCD2OH -PCHgCHg + CH 9 = CHCD20' (5) CHgCH* + C H 2 = CHCD2OH -^C^CHgD +[CH2'— CH- CDOHJ' (6) The mechanism above adequately describes the allyl alcohol system up to 133°C at which temperature pentene-1 and ethanol begin to appear in trace amounts among the products. These products are indicative of the reactions shown below: CH3CH2CH2CHCH2OH-tCH3CH2CH2:CH= CH^ + OH £H = 33 kcal/ao/e (8) CHgCHg + OH* —> CHgCHgOH (8c) CH 2=CH 2+HOH (8e) The ratio R_TT „TJr ^ 0 / R „ „ _0.83 at 141°C, indicating that the majority CH 3CH 2OH C 5H 1 0-of the hydroxyl radicals combine with ethyl radicals at this temperature, The re-mainder may form water by the disproportionation reaction (8e), or by metathesis, or may be captured by addition to the double bond of allyl alcohol. The amount of water formed was too small for detection, even as HOD from the CH 2 = CHCHgOD system, and hence the below inequality is inferred: RHOH^ RC 5K i 0^ ^HgCHgOH) AT 155°C there is a sharp increase in the rate of formation of acrolein, accompaniec by the appearance of the new products hydrogen, methane and propane. Such products could result from the participation of the reactions: f C H 0 — C H ~ C H Q H ] — ? C H = CHCHO + H* AH = 43 kcal (9) 2 ii The reactive hydrogen atoms can then undergo the following series of reactions: CH 3CH 2 + H* H> CH3CH^* (a) 83 CH 3CH 3 *-) 2CH 3 (9b) CHgCHg * + R H —> CHgCH3 + RH* (9c) followed by; CHgCHg + CH 3 -> CH 3CH 2CH 3 (9d) C H 3 C H 2 + C H 3 -?CH 4+C 2H 4 (9e) RH+ CH3~> CH 4+ R " (9f) where RH represents one of the two reactants, diethyl ketone or allyl alcohol. The ethane formed in reaction (9a) would have considerable excess vibrational energy, and several workers*^* have shown that in the low pressure region of this study, decomposition, reaction (9b), would predominate over deactivation, reaction (9c). However, a second possible series of reactions that are Mnetically indist-inguishable from the first alternative, and would account for the above products are no\ given: Isomerization, [ CH;—CH— CHOH)-* C H 2 C H 2 C H 0 ? ^H= 14 kcal/mole (10) followed by, CHgCHg + CHgCHgCHO-^ CHgCHg + CH 2 « CHCHO (10a) and finally two dismutation reactions of low endothermicity: CHgCHgCHO —T> CH 2 = CH 2 + CHO; AH = 23 kcal/mole CHO -7 H+ CO;&H = 14 kcal/mole Based on energetic grounds, the second alternative would be favoured since it involves reactions with lower changes in enthalpy. It is noted that acrolein was shown to be absent from the liquid products of the= allyl acetate system thereby indicating that reaction (9) Is negligible in that system. ^CH2--CH'^CHOOCCHgl —? CH 2 = CHCHO + 'OCCHgjAE= 29 kcal/mole (9) The values of 0.134 + 0. Oil for allyl alcohol and 0.138 + 0.004 for allyl alcohol-d measured for fe^Ag agree well with the value of 0.137 + 0.010 obtained for 12 the octene-1 system , and give further support for the above mechanism. , • 84 The rate constants have been calculated on the basis of the mechanism given above and are now given: Allyl Alcohol p ] ( R C H ) i 7 2 (y^p" [ B 3 P C „ ) 1 / 2 °4 10 4 ^4^0 The final term of this equation is treated as negligible in the calculations. The numerator is a mutually compensating difference cf small terms, each independent of temperature; in any event this approximation would lead to a small error In the preexponential factor. The numerator is approximately equal to (R^ + R G E ) -( R „ D + R G D ) , and below 155°C this is given by 0.21 CK 3CII 3 + CK 2 = CIICH20'' (5) CH 3CH 2 + CH 9 = CHCH2CH — ^ CHgCHg + [cH^- CHirCHGIl]' (6) Hence at 100°C, about 90% of the metathesis is ocurring at the CC- methylene group in-dicating a similar pattern of metathesis as found for allyl acetate. Table XH reveals a significant difference between corresponding values of the Arrhenius parameters of reactions (5H) and (6H). Reaction (6H predominates because the favourable energetic term E C T 3. - E.„ = 5.6 + 0.9 kcal/mole outweighs the unfavourable o i l ori ~" steric factor implicit in the term log ( A 5g/ Agjj) = 2.5 + 0.6. This latter ratio indicates that the hydrogen atoms of the methylene group are remarkably well shielded from attack in comparison with the hydrogen atom of the hydroxyl group. This pattern of reactivity is quite differenet from that found for metathesis between 107 the methyl radical and methanol by Shannon and Harrison in 1963. These workers obtained a low pre-exponential factor for the abstraction of hydrogen from the hydroxyl 108 group, but offered no explanation for this result. Recently, Gray, Jones, and Thynne measured hydrogen abstraction reactions from nitrogen, an el ement similar in electo-negativity in comparison to oxygen. From their study of the methyl radical with ammonia, they found an activation energy of 9.8 kcal/mole, and a "normal" preexponential factor. 8 8 Thynne has measured the metathesis between methyl-dg radicals and methanol in his current research, and has found a pattern of reactivity similar to that found in this in-87 vestigation. The synthesis of allyl-1, l - d 2 alcohol from allyl-1, l - d 2 acetate described in the Experimental section enabled the primary isotope effect for metathesis with the « - methylene group of allyl alcohol to be calculated. The calculations yielded 4.5, [4.3] at 105°C, and 4.1, [3.9l at 138** C in good agreement with the predicted isotope effect shown in square brackets whereby the difference in zero point energies of the C-H and C-D bonds of the ©^-methylene groups of CH 9 « CHCH"2OH and CH"2 = CHCD2OH respectively is used to calculate the predicted primary isotope effect. The experimental value of the primary isotope effect for metathesis with the hydroxyl group of allyl alcohol was determined to be, k ^ / k ^ = 4.0 at 100°C; this latter value was somewhat lower than the predicted value calculated using the difference in zero point energies of the O - H and O - D bonds, kr__ / k «= 6.6 at 100°C. Since the differences in fee Arrhenius parameters for reactions (5H) and (5D) are very small, another inter-pretation of the Arrhenius shown in Table XII may be obtained from the reasonable assumption that A g H « A . Then S g D - E g ^ = 2.303 x 1.987 x (373.2 x0.61) = 1040 kcal/mole, which is slightly lower than the zero point energy difference of 1.4 kcal/mole. The revised Arrhenius parameters are then given by: 10 1 3k, T T / k1/2 = 10 6 * 5 5 exp(-ll. 0) 103 /RT = 1.26 cm 3/2 molecule " l / 2 S e c " l / 2 a t 100° 10 1 3k C T,/k^ 2 = 10 6* 5 5 exp(-12.0) 103/RT = 0.33 cm 3 / 2 molecule~ l / 2 sec" l / 2 at 100°C 51) 2. The above values of the Arrhenius parameters expressed in the rate equations fall within the confidence limits of the experimental values, and follow the interpretation of the nature of the isotope effect in terms of zero point energies. The Arrhenius parameters for metathesis with octene-1 are high in comparison with the parameters for reaction (6H) for diallyl or allyl alcohol shown in Table XII. Abstractios of one of the numerous hydrogen atoms of the n - pentyl group of octene-1 would be characterized by the high activation energy of 10.6 = 0.4 kcal/mole observed for metathesis with n - heptane by James and Steacie1. Significant abstraction from this group would raise the apparent values of the Arrhenius parameters above the true values for reaction (61 88 Figure 9 Metathesis and addition between the ethyl radical and various allyl alcohol substrates. Open Circles:- allyl alcohol -d Closed Circles:- allyl alcohol Triangles:- allyl - 1 , l - d 2 alcohol " * -89 Figure 10 Combined reactions (5) and (6) of metathesis between the ethyl radical and undeuterated allyl alcohol. Closed Circles:- results of this investigation Squares:- results of Brown and James 90 Figure 11 The isotope effect in metathesis between the ethyl radical and the hydroxyl group of a l l y l alcohol. Open Circles:- allyl alcohol -d Closed Circles:- allyl alcohol Triangles:- allyl - 1 , l - d 9 alcohol 92 Figure 12 Dismutation of the addition adduct radical of allyl alcohol. Open Circles:- allyl alcohol -d Closed Circles:- allyl alcohol - • • 'i 93 This phenomenon occurs in the allyl alcohol system whereby the overall metathesis reactions, (5) and (6) yield an activation energy of 8.0 + 0.06 kcal/mole which lies between the independent values: Ea„ - 1/2 E 0 = 6.0 = 0.3 kcal/mole and E_„ - 1/2E. = o i l u oil 2 11.6+0.6 kcal/mole. However, a more likely explanation of the higher metathetical activation energy of octene-1 is probably due to a greater degree of activation of the methylene group by the hydroxyl or allyl group than by the n-pentyl group. This type of activation is also seen in the low values of E g associated with the allyl ethers. As can be seen in Table XTV, the high energy of activation of allyl alcohol for reaction (8) indicates that this reaction is negligible at temperatures around 100°C. This is in contrast to the allyl acetate system whereby reaction (8) plays an importar.. role in the effective chain transfer processes. Thus the low degree of polymerization for allyl alcohol must arise from the other modes of chain transfer, namely reactions (5) and (6). This conclusion is supported by the work of Gaylord and Kuyawa(57) on the polymerization of methyl allyl acetate and methyl allyl alcohol. In general, good correlation has been found between the kinetics of polymerization of the allyl compounds in the gas phase as compared to the liquid phase. F. Reactions of the ethyl radical with allyl propionate 1. Mechanism Reactions (1) to (4c) are supplemented by the reactions shown below, CHgCHg + CH2 » CHCHgOOCCHgCHg C H 3 C H g + CH 2 = CHC^COCCKCHg (5) CHgCH2+ CI-I2= CKCHgOOCCHCHgCH 2= CHCH2OOCCH(CH2CHg)CHg (5c) CHgCHg + CH 2= CHCH2OOCCH2CHg ^ CHgCHg + tCH^'CH'^CHGCCC^CRgV (6) CHgCHg + LCH^CH'JiCHOOCCHgCHgl-ifCHg - CKCH(CH2CHg)COCCH2CHg (6c») — ^ C H 3CH 2CH 2CH= CHCOCCH2CHg (6c") CHgCHg + CH 2 = CHCH2OOCCH2CHgH> CHgCHgC^CHCHgCOCCHgCHg (7) CHgCH* + CH 3CH 2CH 2CHCH 2OOCCH 2CH 3-» CHgCEgC^CHtCHgCHgJC^OOCCHgCHg ( Both carbon dioxide and pentene-1 are formed in equivalent quantities between 79° and 141°C as shown by the material balance, M^ . 94 M* = R / R_r. = 1. 00 + 0. 02 1 C5 T.0 O U ' Hence, the dismutation reaction (8) is proposed to complete the mechanism up to 141°C. The dismutation of the radical adduct of reaction (7): •CHgCHgCHgCHCHgOQCCHgCHg-^CHgCHgCHgCH- CH 2 + C0 2 + CHgCH^AH* 12 kcal (8) Shields pentene-1 and carbon dioxide in equal amounts, and the ethyl radicals also formed are indistinguishable from those generated by the photolysis of diethyl ketone. The es-timated endothereiicity of this reaction is small and not greatly different from the ex-perimental activation energy of 18 + 3 kcal/mole. This reaction is discussed more fully in the next section. . At 151°C a small amount of allene appears among the products, and at 162°C the pro-portion of allene is considerably augmented. No propene was found at either temperature, although it could have been detected easily if present. This implies that the aUene is not formed from the allyl radical by disproportionation with the ethyl radical as the interaction 44 between these two species is known to give propene and allene in the approximate ratio, 3 : 1. Therefore the sequence of reactions shown below can be eliminated: CH 3CH 2 + CH2=CHCH2OOCCH2CH3-^CH3CH3 + CH^CHCHgOOCCHgCHg; AH=0 kcal CH^CHCHgOOCCHgCH^[CH^'CH-CH^ + C0 2 + CH2=CH2; AH - -7 kcal CHgCHg +tcH2-^ CH'iiCH^ +*CH3CHg + CH2=OCH2 (4%) CH2=CH2 + CH3CH=CH2 (11%) —> CK 3CH 2CH 2CH»CH 2 (85%) The following sequence is consistent with the experimental results: CHgCi-I* -> CHg-CHCHgOOCCHgCHg-fr CH 3CH 3 + CH^CCHgOOCCHgCHgjAHsS kcal (/O) CH 9=CCH 2OOCCH 2CH 3-^CH 2=C=CH 2+C0 2+CH 2CH 3;AH= 14 kcal. (it) No propene or extra pentene-1 are formed, and so the total carbon dioxide should exceed the pentene-1 by an amount equal to the allene formed. This is confirmed by the average value of thirteen experiments: M. =(Rp w + R r „ V R p n =1.00+0.02 1 V C 5H 1 Q C 3H 4 C0 2 95 Significant dismutation of the radical products of reactions (5) and (6 ) is rendered im-probable by the high endothermicities and by the absence of appropriate products: CH2=CHCH2OOCCHCH3 —^CH^CHCHgO + 0=C=CECH3; <$H• 41 kcal [ CH^-CHdCHOOCCHgCHgl-^CHg-CHCHO + OCCHgCHgj^H = 31 kcal The ratio of mutual disproportionation to combination for the ethyl radical, fcgA2 ~ 0.134 + 0.003, is in good agreement with the value of 0.1S7 +_ 0.010 obtained in the 12 octene-1 system . Thus the values for k 3/k 9 and indicate that reactions (1) to (11) adequately represent the mechanism of this system over the temperature range 79 to 162°C, The rate equations are now given: ( W V • R C 2 H 6 - R C 2 H 4 - M k4 h ^ ] ( R C 4 H I O ) 2 . . k ? ( ECO + l / 2 R C O ^ - ( RC 2H 6 * - < 1 - M ^ C RCO + l / 2 RCQ 2 ) T p " ."" LB](RC H ) 1 / 2 LB ] (R C U ) ^ 2 C4 M10 °4^0 where M* ( R ^ + R^) / ( R c o + l / 2 R C 0 2 ) 1/2 k l / 2 ' R C EL EL > 8 K 2 = S^IO S^IO k7c K o + R C 3 H ) - C R C 2 H 6 + R C 4 H 1 ( ) + 1 / 2 R C O ^ 2. Interpretation of the kinetic scheme. The results are listed in Tables X, XII, XIH, XIV and XV, and the Arrhenius plots are shown in figures 13 and 14. Unlike allyl acetate, the metathesis rate constant for allyl proprionate could be measured directly. The greater reactivity of allyl propionate towards metathesis with the ethyl radical is aided by the greater number of hydrogen atoms, two of which are secondary, on the substituent group, "X". However, kg would be expected to be the most important reaction of the composite quantity k g + kg + kg, especially over the lower temperature range since the estimated enthalpies of reaction are: A H 5 = O, ^Hg » -23 and = +5 kcal/mole It is proposed that the rate constants measured for interaction of the ethyl radical with a monomer molecule in the gas phase may be useful in the interpolation of the radical Figure 13 Addition and metathesis between the ethyl radical and allyl propionate 97 I 0 3 / T ° K : • Figure 14 Dismutation of the addition adduct radical of allyl propionate 98 polymerization of that monomer in the liquid phase. Using the relationship discussed earlier tor hexadiene-1,5, the ratio of the rate constants for addition and metathesis of allyl proportionate is given by the expression: k,?/ (k_ + k 6 + k9) = 1 0 6 - °' 6 ) exp (-1.6 + 1.0) 103 / RT = 3.9 + 1.1 at 80°C However, as reaction (8), the analogue of the effective chain transfer process, occurs at an appreciable rate in this system at low temperatures, this estimate is reduced slightly in order to account for all the transfer reactions. Since reaction (8) is unimolecular, a comparison of rates of products rather than rate constants must be used, and when this is done the following relation, Degree of Polymerization = Rate^/ (Rate5 + Rate6 + RateQ + Rateg). yield a value of 3.4 at 80°C. As seen in Table XV, the values for degree of polymerization found in the gas phase show the same trend as the degree of polymerization found in the liquid phase for all the allyl compounds studied in this investigation. In the higher temperature range, reaction (8) becomes the controlling factor in deioriuinin£ iho degree of polymerization. Reaction (8) at 80°C has been postulated to be a very important factor in determining the low degree of polymerization for the allyl esters, allyl tri-56 57 methylacetate and 3-butene-2-yl acetate . 1/2 Since the value of the dismutation rate expression kgk /k ? c is higher for aUyl propionate than for allyl acetate throughout the temperature range of this study, the corresponding ratio is 1. 7 at 100°C and 2.8 at 150°C, a greater value for k g for allyl propionate is consistent with the pattern of effective chain transfer for the radical polymerization of these esters at 80°C. The percentage of the total chain transfer 48 which is effective is 43?6 for the allyl propionate and 24% for allyl acetate. In order to obtain a rough estimate of the percentage of effective chain transfer arising from dismutation reaction for allyl propionate, kg^ was assumed to be the same as allyl acetate and the value for kj/(kj+ k^ ) = 0.57 was used and related to the corresponding equation in the gas phase R g/(R 5 + &6 + &8 + Rg)- Thus, at 80°C, the percentage of effective chain transfer arising from reaction (8) was found to be 56% for allyl propionate. At this temperature, the 99 dismutation reaction is seen to be the most important factor in the effective chain transfer for allyl propionate and reaction (8) would constitute even a higher percentage of effective chain transfer for allyl acetate than for allyl propionate since in the allyl acetate system, k would be expected to be a much slower reaction. 100 G. The Reactions of the Ethyl Radical with Allyl Formate and Allyl Formate-d. (1) Mechanism The reactions of the allyl formate system are similar to those of the allyl acetate and allyl propionate systems and are shown below: 'C2H5 + CH 2 = CHCH2OCOH ~> CHgCHg + CH 2 = CHCHgOCO' (5) C 2H 5 + CH 2 = CHCHgOCO' -?CH2 = CHCHgOCOCgHg (5c) *C2Hg+CH,2= CHCH2OCOH—> CHgCHg+[CH2-^CH-CHOCOH]' (6) ' C 2 H 5 + [ C H 2 ^ " C H ^ C K O C O H ] ^ C H 2 C H C H (C2H5) OCOH • (6c) 'C 0 H_ +(CHJ^ CH^.CHOCOH]-> C„H„CH 0CHCKOCOH (6c') •C2HG+CH2 = CHCH2OCOH ~^C 2H 5CH 2CH4-r CH2OCOH (7) 'C 2H-+C 2H 5CH 2CHCH 2OCOH-?C 2H 5CH 2CH (C2Hg) CH2OCOH (7c) Again carbon dioxide and pentene-1 are formed, but unlike the other two allyl ester systems, carbon dioxide is formed in much greater amounts than pentene-1. Another product, propene, is also produced in extremely large amounts. The reactions below are postulated to account for these products. CHgCH2CH2CHCH2OCOH -=>CHgCH2CH2CH = CH 2 + C0 2 + H' (8) The H atoms generated by this dismutation reaction can then initiate a chain reaction with allyl formate producing carbon dioxide and propene as products. 'H + CH 0 = CH CH2OCOH -? CHg CH CHgOCOH (7H) CHgCHCH2OCOH —> CHgCH= C0 2+ (8H) The products CgHg, CH^ and C^Hg were formed in small amounts and can be explained by the subsequent reactions of the methyl radical produced by the reaction shown below: ' H+ CHg CH 2 -7 CHgCHg —? 2CHg (9) 'CH 3 +'C 2H 5-?C 3H 8 (9a) C H 3 + C 2 H 5 - 7 C H 4 ± C 2 H 4 (9b) 'CHg + B -? CH 4 + B (9c) 'CH3 + D ~S> CH 4 + D (9d) 'CHg + CH 2 a CHCH2OCOH - 7 CH 3CH 2 CHCHgOCOH -5>C4HQ + C0 2 + H ' (9e Where B represents allyl formate and D represents diethyl ketone. 103 From the above reactions, the following material balance, M, = ( R ^ „ + R + 1 C 5 H 1 0 C 4 H 8 R c H + R o H + 2 R c H ) / R r o a n d M i = ) - R C 5 H 1 0 1 S R C „ H 4 - O j U * C , H A - ° ' 0 3 R C H .-±= 2 4 Xl4 3 4 3 8 kp — R C 4 H 1 0 k 5 H = R C 3 H 4 + 2 R C 6 H 1 0 .1/2- 0 .04 C 4 H 1 0 The rate equations k/k^2 and fcgkj^Nc f o r ^ formate-d system are analogous to the equation above. The rate equations for reactions (5) and (6) are now given for the allyl formate-d system. k 5 » , = ! W 2 H ° 6 H l O « c - „ 1 0 2 p T T o. 04 ^ ^ 5 2 [B]R C 1/2 [ B J R C H l / 2 R J?4 10 4 10 k6 = C 2 H 6 - C2 B4 . 44 The values kg^/kg = 0.04 and kge/kg = 0.11 measured in the hexadiene-1,5 system expressing the amount of disproportionation between the ethyl and allyl radicals, and GO the value 0.06 representing the disproportionation.-combination ratio for methyl and ethyl radicals were used in the above rate equations. 2. Interpretation of the Kinetic Scheme. The results for the allyl formate and allyl formate-d systems are given in Tables XI, XII, XLTI and XXV, and the Arrhenius plots are shown in figures, 15, 16 and 17. Allyl formate and allyl formate-d exhibited the same reactivity toward the addition of the ethyl radical as the other allyl esters studied in this investigation. Thus at 100°C, log kyAg^ 2 - 1.22 + 0.07 in the allyl formate systems as compared to log k^Ag^ 2 -1.25 +• 0.14 in the allyl acetate system. Again reaction (6) was extremely slow and therefore difficult to measure in the allyl formate systems as was the case for the allyl acetate systems. Using some L \ 1 / 2 M ^ extrapolated values (kg«-kJA 2 in the allyl formate system and some values of l/2 kg/kg' obtained directly in the allyl formate-d system, the Arrhenius parameters for reaction (6) could be calculated and are shown in Table XII. Again the pre-exponential factor for reaction (6) was low indicating that the formyl group provides some shielding towards attack from the ethyl radical, and would account for the low reactivity of this reaction. The measured activation energy for reaction (5), Egjj* 3 7 « 7 £ 0.4 Kcal was in 109 good agreement with the value measured by Thynne ' in the ethyl formate system, E r = 7.8 Kcal, and was 1.2 Kcal lower than E n . This value of 1.2 Kcal is in accord 5H with the difference in zero point energies of a C-H and C-D bond given as 1.1 Kcal. The pre-exponential factors for reactions 5H and 5D are identical within the 103 experimental error as shown in Table XH. The measured primary isotope effect, k 5 H / k g D 4.09, [4.42]compared favourably with that calculated using the concept of zero point energies shown in square brackets. This concept of zero point energies has been used by Gray 1 0 8' 1 1 0 * 3 - 1 1 and his colleagues with good success. The products resulting from the allyl formate dismutation reaction were similar to the products resulting from the allyl ester dismutation reactions. C H 3 C H 2 C H 2 C H C H 2 O C O H - > C H 3 C H 2 C H 2 C H « CH 2+CG 2+H' (8) No evidence was found for the existence of the HCO' radical since the product ethyl o formate, formed by the combination of CgH^ and HCO radicals was not found in the o liquid products. An interesting reaction mentioned in the Mechanism section is the chain reaction betwee the H atom and allyl formate, and analogously, the chain reaction between the D atom and ally formate-d. '11+Clig-CHCH 2OCOH -> CHgCHp^OCOH (7E, CH 3CH 2CH 2 OCOH CH 3CH= CH 2+ C0 9+ H' (8 D'+ CH 2 = CHCHpOCOD -> DCH^CH CHgOCOD (7D; 'DCH2'CKCH2OCOD -> DCH2CH= CH 2+C0 2+D . (8D] No evidence for triplet energy transfer shown by the reactions below was found in the allyl formate-d system or the allyl formate system since the production of C0 2 was not 8 112 113 quenched when biacetyl was added. Biacetyl * * has been shown to be very efficient in the quenching of triplet energy transfer. E>+B->B*+D B —? C0 2 + propene F+D*->F*+D where D represent*diethyl ketone, B represents allyl formate-d or allyl formate and F represenfjbiacetyl. Represents the triplet state. An estimate of the kinetic chain length in the allyl formate system could be JR, — 0 11 & made using the following expression: C3HQ 0T0T C3 H4 R_ T T - 0.85 R r „ C5 H1Q 0T04 C3 U4 i ~ i : r* i ~ i I 1 ; I L _ _1 I I 2.2 ; 2.4 2.6 2.8 I 0 3 / T ° K Figure 15 Addition of the ethyl radical to allyl formate substrates Open Circles:- allyl formate-d" Closed Circles:- allyl formate Figure 16 The isotope effect in metathesis between the ethyl radical and the formyl group of allyl formate. Open Circles:- allyl formate-d Closed Circles:- allyl formate 106 _ j i t L_ _! ; I -2.3 2.4 2.5 2.6 I 0 3 / T ° K Figure 17 Dismutation of the addition adduct radical of allyl formate Open Circles:- allyl formate-d Closed Circles:- allyl formate 107 The expression for the allyl formate-d system is analogous: R n „ n V u f f c w R ° 3 H 4 The values of the kinetic chain length, Q are shown ix Table Xi, and lied in the range 5 to 20 for both the allyl formate and allyl formate-d systems. The kinetic chain length decreased with increase in temperature indicating the influence of the higher activation energy of the dismutation reaction as compared to the activation energy of the propagation reaction. H. Kinetic Scheme of the Diallyl Carbonate System. A brief study was made on the kinetics of the diallyl carbonate system. The diallyl carbonate system was characterized by reactions analogous to the diallyl ether system, and the results are shown in Table IX. It was hoped that in the case of the diallyl carbonate system it would be possible to measure the combination-disproportionation reactions of the allyloxy and ethyl radicals. However, reaction (8) which would generate the allyloxy radicals wci5: negligible at 163°C, whereas reaction (£) proceeds at an extremely fast rate. Hence, the same trend as diallyl ether is displayed, in that the decomposition of the adduct radical formed in metathesis is much faster than the decomposition of the adduct radical formed in addition. Thi/s, reaction (S) is only significant in the diallyl ether and diallyl carbonate systems of all the allylic systems studied in this investigation. Unlike the other allylic systems, reaction (S) is aided by the resonance stabilization of the products in the diallyl ether and diallyl carbonate systems. 0 CH CH 2CH CH CH2GJCGCH2CH = CH2-> CHgCH^^CH = CH 2 + 'OCHgCH = CHg + C\ ( [CH^ ' C H r C H O g O C H g C H a C H g C H g « CHCHO + C0 2 +[CH2'-CH- CH^' < The low volatility of diallyl carbonate made it difficult to measure the rate constants with any degree of precision since corrections had to be applied for the consumption of the reagents, diethyl ketone and diallyl carbonate, but the combinatLon-disproportionation reactions of the ethyl and allyl radicals could be measured quantitatively and the values k = 0. 865, k„ A - 0.034 and k Q = 0.102 were in excellent agreement with'those c bd be k8 k8 k8 found in the diallyl ether and hexadiene-1,5 systems. 108 CONCLUSIONS The reactions of the ethyl radical with a series of allyl compounds were studied experimentally in the gas phase, and from the results of these reactions, a general mechanism was proposed. Specifically the ethyl radical can abstract a hydrogen atom from the ^ -methylene group or the substituent group of the allyl compound, or add to the olefinic linkage of the allyl compound. Dismutation reactions of the product radicals formed in the addition or the abstraction reactions were also observed in this study. The general'mechanism is shown below: CH 9 =» CHCEL RH represents an allyl compound. CH 2 = CHCH2RH + C 2H 5 -> C ^ + CH 2 = CHCHgR.' (5) CH 2 = CHCH2R + C 2H 5 CH 2 = CHCH 2 RCgHg (5c) CH 2 = CHCH2RH + C 2H 5 - 7 C 2H 6 + CH 2= CHCHRH (6) CH 2 » CH CH RH + C2H5" -?CH 2 = CHCH (C2H5) RH (6c) CH 2= CH CHRH + C2H5' C 2H 5CH 2CH= CHRH (6c') CH 2 = CHCH2RH + C 2H & CgHgC H 2 CHCHgRH (7) C 2H 5CH 2CHCH 2RH + C 2H 5 -?C 2H 5CH 2 CH (C2H5) C^RH (7c) C 2H 5CH 2CHCH 2RH —>C 2H 5CH 2CH= CH 2+RH' (8) CH 2 = CHCHRH —> Products (9 ) Rate equations were derived for the above reactions, and by studying these over a temperature range, Arrhenius parameters could then be calculated. The Arrhenius parameters calculated for the addition reaction for a series of allyl com-pounds yielded a characteristic value for the activation energy of 7. 7 kcal/mole as shown in Table XHI. This characteristic value might be expected since the reaction center in the addition reaction, reaction (7)^ is remote from the substituent group, and 109 the energetics of this process would be expected to be independent pf the nature of the substituent group RH. However, the Arrhenius parameters calculated for the meta-thesis reactions, reactions (5) and (6);varied a great deal as shown in Table XQ. This variation might be expected since now for reaction (6), the reaction center is adjacent to the substituent group, and the substituent group could offer some shielding from the attack by the ethyl radical at the reaction center. The variation in the Arrhenius parameters for reaction (5) can be attributed to the differences in bond strengths of the hydrogen bond in the substituent group. These bonds vary from a C-H bond to an O-H bond. The activation energies measured for the dismutation reaction of the adduct radical, reaction (8);ranged from 15 to 27 kcal/mole as shown in Table XTV. This wide range of activation energies might be expected since the energetics of this process de-pends on the nature, of the substituent group. The substituent group varied from an acetoxy group to a hydroxyl group. The kinetics of interaction of the ethyl radical with an allyl compound in the gas phase can be used to interpret the polymerization of that monomer in the liquid phase. Specifically the ratio of addition to metathesis rate constants, k^/{k,_ + k 6 ^ n ^ e gas phase can be related to the degree of polymerization by means of a modified Hammett equation. A definite trend between these two quantities was observed for the • allyl compounds studied in this investigation as shown in Table XV. Thus the relatively high degree of polymerization of 35 for allyl-1, l-dg acetate has a corre-spondingly high value for k^ /k,. + kg)of 21 in the gas phase whereas the low degree of polymerization of 5 for allyl alcohol has a correspondingly low value for k^/^g + kg)of 1. 8 in the gas phase. The dismutation reactions, reactions (8) and (9), offer a convenient method for the generation of free radicals at relatively low temperatures. In this way, the allyl and ethoxy radicals were generated, and their disproportLonation-combination reactions with the ethyl radical were studied. The allyl radical was generated in three kinetic systems, the hexadiene-1,5 system, the diallyl ether system and the diallyl carbonate 110 system. The extent of the disproportionation-combination reactions of the ethyl and allyl radicals are shown below: CH 2= CHCH 2+C 2H 5' -) CHgCHgCHgCH^ CH 2 (85%) combination GH 2 = CHCHg + C2H5-> CH 2 = C = CH 2 + CgHg (4%) disproportionation CH 2 = CHCH 2 + C2H5' CH 2 - CHCRg + C 2H 4 (17%) disproportionation The disproportionation- combination reactions of the ethoxy and ethyl radicals measured in the allyl ethyl ether system are shown below: C2H5' + CgHgO'-; C 2H 5OC 2H 5 (80%) combination C2H5' + CgHgO'^, C 2H 4 + C 2H 5OH (17%) disproportionation C2H5' + CgHgO—>C2Hg + CHgCHO (3%) disproportionation Reactions (5) and (6) could be distinguished and therefore measured by labelling the allyl compound with deuterium at a definite position in the molecule. In the case of allyl alcohol and allyl acetate, primary isotope effects in metathesis were measured experimentally and were in close agreement with the theoretical values predicted on the basis of zero point energy differences. I l l . APPENDIX A Calculation of heats of reaction for the dismutation reaction. Example: CHgCH2CHgCHgCHgCHgCH * CHg -i> H + CHgCHgCHgCH CHgCHgCH • CHg 4H f = -18.8 kcal/mole (ref 61) AHf - 52.1 kcal/mole (ref 71) A H f « X Using the relationship mentioned in reference 71, D (C-H) g e ( J • 94 kcal/mole (ref 54) » AHf(X) +AHf (H) - AH f (octene-1) AH f (X) » 23 kcal/mole CHgCHgCHgCH CHgCHgCH » CHg -> CHgCHgCHgCH• CHg •tCHg'-^-CH^Hg3' A H f « 23 kcal/mole ^ H f » -5 kcal/mole (ref 61) AH f =» 32 kcal/mole (ref 43) Hence the heat of reaction for the above dismutation reaction is A H f (pentene-1)+41^ (CHgi: CH^'CHg)^(CHgCHgCHgCHCHgCHgCH-CHg-aHg" 4 kcal/mole. For the above example, AH^ for octene-1 was readily available, but the AHf's for the other allyl compounds had to be extrapolated since the AHf's were not listed. Heats of combustion could be used to find the heat of formation of a particular compound. For example: £H f of n-amyl alcohol. 15 Cg + CHgCHgCHgCHgCHgOH^^JJHgO +j3COg H c « 799 kcal/mole (ref 61) AH f => X^-68.4 kcal/mole (ref 61)4H-94.4.kcal/mole (ref 61) /_VHc » 6 AH f (HgO) + 5 4H f (COg) - ABf (n-amyl alcohol) A H f (n-amyl alcohol)gafl » -72 kcal/mole since A&evg^* 11 kcal/mole (ref 61) The above procedure given in example 1 was then used to calculate the heat of reaction for n-amyl alcohol. Values for ^ H f (CHgOCO) and AHf (CHgCHgOCO)'were listed In reference 114 and used in calculating <£ Hg for the dismutation reactions of allyl acetate and allyl propionate. The heats of reaction were calculated in a similar fashion as shown above for the other dismutation reactions studied in this investigation and are shown in Table XIV. 112 APPENDIX B Derivation of Rate Equations: Example 1: Derivation of the metathesis rate equation in the allyl propionate system: The reactions for the allyl propionate system occurring in the temperature range 79 to 141°C are shown below: C 2 H 5 C O C 2 H 5 -» 2C 0H C + CO Lt 5 (1) 2 C 2 H 5 -? C4 H10 (2) 2 C 2 H 5 • —? C2 H4 + C 2 H 6 (3) C 2 H 5 + C 2 H 5 C O C 2 H 5 C 2 H 6 + d 2 H 4 C O C 2 H 5 (4) C 2H 4COC 2H 5 +. C 2 H 5 - 7C 2H 4(C 2H 5)COC 2H 5 (4c) C 2H 5*+CH 2=» CHCH 2OCOC 2H 5->C 2H 6+CH 2 = CHCH 2OCOC 2H 4 (5) C 2H 5 + CH 2 » CHCH 2OCOC 2H 4 —? CH 2 =» CHCH 2OCOC 2H 4(C 2H 5) (5c) C2Hg + CH 2 =« CHCH 2OCOC 2H 5 _ j C 2H g + CH 2 = CHCHOCOC2H5 (6) C2H5'+ CH 2= CHCHOCOC2H5 —y CH 2 = CHCH(C2H5) OCOCgfL. (6c) —> C 2 H 5 C H 2 C H » CHOCOC 2H 5 (6c)' C 2Hg+CH 2= CHCH 2OCOC 2H 5 ->C 2H 5CH 2CHCH 2OCOC 2H 5 (7). C 2 H 5 + C 2 H 5 C H 2 C H C H 2 O C O C 2 H 5 ~* C 2 H 5 C H 2 C H ( C2 H5 ) l CH2OCOC2H5(7c) C 2H 5CH 2CHCH 2OCOC 2H 5 C2H5CH2CH=*CR"2 + C0 2 + (8) Reactions (3), (4), (5) and (6) shown above consume two ethyl radicals to produce one molecule of ethane, and these are the only reactions in the allyl propionate system to produce ethane as a product. Thus the rates of reactions (3), (4) (5) and (6) are related to the rate of formation of ethane by the following equation: R C 9 H R " V * 4 + * 5 + 3a6 Ll O 113 where R _ „ represents the rate of formation of ethane, and R q represents the rate 2 H6 6 of reaction (3). £ D ] rep resents the concentration of diethyl ketone, and ^ BJrepresents the concentration of allyl propionate. S i n c e R 2 = *C 4H 1 0 " ^ [ W ] 2 ' r C 2 H 5 ] . k 2l/2 Upon substitution of the above expressions into equation (a), the metathesis rate equation can be calculated. a n d k 5 + 6 • " C J . - V . - [DJk. *» [ B > C 4 H 1 0 > X 2 The Arrhenius parameters for k 4 have been calculated by previous workers. 2 Example 2: Derivation of the addition rate equation in the allyl propionate system. One molecule of CO is equivalent to two ethyl radicals while one molecule of C0 2 is equivalent to one ethyl radical inflie allyl propionate system. In the addition reaction, reaction (7), two ethyl radicals are consumed without the formation of butane or ethane. Thus the rate of reaction (7) can be expressed by the equation shown below. also, R ? - k 7CB][c 2H 5] and [ c ^ ] - ( B ^ ) l / 2 TV2 k2 Upon substitution, k 7 ECO + | E C 0 2 - E C 2 H 6 - E C 4 H 1 0 ° 4 n 1 0 114 Example 3: Derivation of the rate equation for the dismutation reaction in the allyl propionate system. The adduct radical, CgRgCHgCHCHgOCOCgHg, formed in reaction (7) can either combine with an ethyl radical, reaction(7c), or undergo a dismutation reaction, reaction 8, and these reactions are related by the equation shown below. *7 = R7c + R8 where R ? is given by equation (b) in example 2, R g * R^o m k g [adduct radicafjand H 7 c = k 7 c[G 2H 5] j" adduct radical|«k7c 1/2 C 4 H 1 0 C °2 k 1 7 2" k 2 8 Upon substitution of the above expressions .unto equation (c), the rate equation for the dismutation reaction is found. 1/2 KCO + | K C O 2 ^ C 2 H 6 - X H 1 0 = k 7 c "CO, + K C 0 2 K8 2 k, 7c R 1 CO - R - R CO -2 2 C 2H 6 C 4H 1 Q The rate: equations Were derived-{or the other allyl compound systems in a similar manner. 115, BIBLIOGRAPHY 1. D. G. L. James and E. W. R. Steacie, Proc. Roy. Soc., A 244. 289 (1958). 2. D. G. L. James and T. Ogawa, Can. J. Chem., 43_, 640 (1065). 3. A. C. R. Brown and D. G. L. James, Can. J. Chem., 40, 796 (1962). 4. L. M. DorfmanandS. D. Sheldon, J. Chem. Phys., 17., 511 (1949). 5. K. O. Kutschke, M. H. J. Wijnen and E. W. R. Steacie, J. Am. Chem. Soc., 74, 714 (1952). 6. R, K. BrintonandE. W. R. Steacie, Can. J. Chem.* 33, 1840 (1955), 7. A. Brown, PhD. Thesis, 53 (1962). 8. D. S. Weir, J. Am. Chem. Soc., 83, 2629 (1961). 9. J. C. J, Thynne, Trans. Faraday Soc. . 58. 676 (1962). 10. J. A. Kerr and A. F. Trotman-Dickenson, J. Chem. Soc. ,1611 (I960} 11. D. G. L. James andD. MacCallum, Can. J. Chem. .43. .633 (1965). 12. A, C. R. Brown andD. G. L. James, Can. JrChem., 43^ 660 (1965). 13. M. Szwarc, J. Chem. Phys., 17,, 284 (1949). 14. M. J. Molera and J. F. Stubbs, J. Chem. Soc., 381 (1952). 15. W. A. BryceandP. Kebarle, Trans. Faraday Soc.. 54, 1660 (1958). 16. D. J. RusickaandW. A. Bryce, Can. J. Chem. , 38, 827 (1960). 17. K. J. Laidler and B. W. Wojciechowski, Proc, Roy. Soc., Series A. 259. 257 (1960). 18. A. Amano andM. Uchiyama, J. Phys. Chem.. 68. 1188 (1964). 19. A. S. Gordon, Can, J. Chem. 43, 570 (1965). 20. N. Harumiya, S. Shida and S. Aral, Bull. Chem. Soc. Japan, 38, 142 (1965). 21. C. L. Currie, S. J. H. Okabe and J. R. McNesby, J. Phys. Chem., 67, 1494 (1963). 22. R. J. Cvetanovic, Progress in Reaction Kinetics, 2 96 (1964). 23. H. £. Gunning and R. L. Stock, Can. J. Chem., 42, 857 (1964). 24. R. A. Holroyd and G. W. Klein, J. Phys. Chem., 69, 194 (1965). 25. E. P. Lossing, K. U. Ingold and I. H. S. Henderson, J. Chem. Phys., 22_, . 621 (1954). 26. X. F. Palmer and F. P. Lossing, Can. J. Chem., 43/565 (1965). 27. F. P. Lossing, D. Q. H. Mardsen and J. B. Farmer, Can. J. Chem., 34. 701 (1956). 28. H. M. McConnell and D. B. Chesnut, J Chem. Phys., 28, 107 (1958). 29. C. Heller and T. Cole, J. Chem. Phys., 37,, 243 (1962). 30. R. W. Fessenden and R. H. Schuler, J. Chem. Phys., 39, 2147 (1963) 31. B. Smaller and M. S. Matheson, J. Chem. Phys., 28_, 1169 (1958). 32. 35i S. Lebedev and Yu D. Tsvelkov, Zh. Strukt. Khim. 2, 607. 9 (1961). 33. J . E. Bennett and A. Thomas, Nature, 195. 995 (1962). 34. J . R. McNesby and A. S. Gordon, J . Am. Chem. Soc., 79, 825 (1957). 35. J . R. McNesby and A. S. Gordon, J. Am. Chem. Soc., 79, 4593 (1957). 36. D. A. Dominey and C. J. Danby, J. Chem. Soc., 4656 (1962). , 37. A. S. Gordon, S. R. Smith, and J. R. McNesby, J. Am. Chem. Soc., 81. 5059 (1959). 38. C. A. Coulson, Proc. Roy. Soc. A, 164. 383 (1938). 89. J. L. Bollandand G. Gee, Trans. Faraday Soc., 42_, 244(1946). 40. S. W. Benson and P. Nangia, J. Am. Chem. Soc., 84, 3411 (1962). 41. S. W. Benson, A. N. Bose and P. Nangia, J. Am. Chem. Soc., 85.. 1388 (1962). 42. K. W. Egger, D. M. Golden and S. W. Benson, J. Am. Chem. Soc. ,86, 5420 (1964). 43. C. A. McDowell, F. P. Lossing, I. H. S. Henderson and J. B. Farmer, Can. J. Chem., 34, 345 (1956). 44. D. G. L. James and G. E. Troughton, Chemical Communications, No. 6 94 (1965). • 117 45. A. AmanoandM. Uchiyama, J. Phys. Chem.,j37, 1242 (1963). 46. W. A. Bryce and M.S. Hardiman, Can. J. Chem., 40, 1031 (1962). 47. P. S. Dixon, A. P. Stefani and M. Szwarc, J. Am. Chem. Soc., 85, 2551 (1963). 48. N. G. Gaylord, J. Polymer Sci., 22, 71 (1956). 49. D. G. L. James andD. MacCallum, Proc. Chem. Soc., 259 (1961). 50. G. Solomon and C. Konongsberger, J. Polymer Sci., 2, 522 (1947). ' 51. M. S. Matheson, E. E. Auer, E. B. Bevilacqua and E.J. Hart, J. Am. Chem. Soc, 71,, 2610 (1949). 52. C. H. Bamford, W. G. Barb, A. D. Jenkins and P. F. Onyon, The Kinetics of vinyl polymerization by radical mechanisms. Butersworth Scientific Publications, London. 15_, 270 (1958). 53. P. D. Bartlett andR. Altschul, J. Am. Chem. Soc. , 67, 816 (1945). 54. N. N. Semenov, Some Problems in Chemical Kinetics and Reactivity. < Vol. 1, Pergamon Press, London, 37 (1958). 55. D. G. L. James and G. E. Troughton, J. Polymer Sci., Part A 75 (1965.) 56. N. G. Gaylord and F. R. Eirich, J. Am. Chem. Soc., 74, 334 (1952). 57. N. G. Gaylord and F. M. Kujawa, J. Polymer Sci., 41, 495 (1959). 58. N. G. Gaylord, M. Katz, and H. F. Mark, J. Polymer Sci., B2,, 151 (1964). 59. M. Kasha, J. Am. Opt. Soc., 38, 929 (1948). 60. D. J . LeRoy, Can. J. Res., B/28, 492 (1950). 61. Handbook of Chemistry and Physics, the Chemical Rubber Publishing Company, Cleveland, Ohio, 2306 (1961). - - , 62. N. M. R. Spectra Catalogue, Varian Associates, spectrum No. 34. 63. J. C. J. Thynne, Proc. Chem. Soc., 18 (1961). 64. P. D. Bartlett and P. A. Tate, J. Am. Chem. Soc., 75, 91 (1953). 65. M. Littand F. R. Eirich, J. Polymer Sic, 45., 379 (1960). 66. D. AdelsonandH. Gray, U. S. Patent No. 2,555,775 (1951). i 1 1 8 67. G. Oster and Y. Mizutani, J. Polymer Sci., 22, 173 (1956). 68. I. Sakurada and G. Takahashi, Chem. High Polymers (Japan), 11, 348 (1954). 69. J. C. J. Thynne, Proc. Chem. Soc., 68 (1961). 70. A. Shepp and K. O. Kutschke, J. Chem. Phys., 26, 1020 (1957). 71. S. W. Benson, The Foundations of Chemical Kinetics, McGraw-Hill Book Company, Inc., New York, 664 (1960). 72. K. E. Lewis and H. Steiner, J. Chem. Soc, 3080 (1964). 73. R. Srinivasan, J. Am. Chem. Soc, 83, 2806 (1961). 74. R. A. Holroyd and G. W. Klein, J. Phys. Chem., 67, 2273 (1963). 75. C. S. Marvel. J. Polymer Sci.. 48, 101 (1960). 76. N. O. Brace, J. Am. Chem. Soc.,^ 86,, 523 (1964). 7,7. G. B. Butler, J. Polymer Sci., 48 , 279 (1960). 7^ . O. Mikulasova and A.,Hvirik, Chem. Zvest,, 11, 641 (1957). 79. S. Bywater and E. W. R. Steacie, J. Chem. Phys., 19., 172 (1951). 80. R. Srinivasan, Personnel Communications, 81. A. AmanoandM. Uchiyama, J. Phys. Chem.. 69, 1278 (1965). 82. M. Miyoshi andR. K. Brinton, J. Chem. Phys., 36_, 3019 (1962). 83. M. Arvahami and P. Kebarle, J. Phys. Chem., 67, 354 (1903). 84. P. Kebarle andM. Avrahami, J. Chem. Phys., 38, 700 (1963). 85. J. R. Majer, K. T. F. Pinkard and J. C. Robb* Trans, Faraday Soc, 60, 1247 (1964). 86. M. Avrahami and P. Kebarle, Can. J. Chem., 41_t 347 (1963). 87. C. H. Bamford, A. D. Jenkins and R. Johnston, Trans. Faraday Soc, 55 , 418 (1959). 88. J. Thynne, Personnel Communication. 89. M. H. J. Wijnen, J. Am. Chem. Soc, 82, 3034 (1960). 90. C. A. Heller, J. Chem. Phys., 28, 1255 (1958). 91. F. H. Westheimer, Chemical Reviews, 61, 265 (1961), 119 93. V. GoldaadD. P. N. Satchel!, Quart Rev., (Loudon), 9, 61 (1055). ©8. K. B. Wiberg, Chem. Rev., 55, 713 (1955). 94. R. Stewart and R. Van Der Linden, Tetrahedron Letters, ,2, 28 (1960). 95. W. F. K. Wynne-Jones, J. Chem. Phys. 2, 381 (1934), 98. K. B. Wiberg, J. Am. Chem. Soc., 70, 5371 (1954) 97. P. D. Bartlett and J. D. McCulium, J. Am. Chem. Soc., 78, 1441 (1956). 93. F. S. Daiaten, £. A, Creak and K. J, Ivin, Trans. Faraday, Soc.. 58. 326 (1862). W. Jackson, J. R. McNesby and B. Darwent, J. Chem. Phys., 37, 1610 (1962). 100. H. S. Johnston and D. Rapp, J. Am. Chem. Soc., 83, 1 (1961). 101. S. Glasstone, K. J. Laidler, and H. Eyriag, The Theory of Rate Processes, McGraw-Hill Book Company Ins., New York, 153 (1941). 102. M. Solomon, Can. J. Chem., 42, 610 (1964). 103. J. C. Bevington, Radical Polymerisation, Academic Press, London and New York, 115 (1981). 104. H. Standings? and T. Fleitman, Ann. 480. 92 (1930). 105. B. de B. DarwentandE. W. R. Steacie, J. Chem. Phys., 16, 331 (1848). 106. C. A. Heller and A. S. Gordon, J. Chem Phys., 36, 2648 (1962). 107. T. W. Shannon and A. G. Harrison,