- Email: [email protected]
Free radical pathway for the baeyer-villiger reaction
TetrahedronLettersNo.30, Pp. 2871-2874,1967. Pergmon Press Ltd. Printed in Great Britain.
F'FEERADICAL PATHWAY FOB THE BAEYER-VILLIGER FEACTIOB J.C. Robertson and A. Swelim (1) Department of Chemistry, Colorado State University, Fort Collins, Colorado, U.S.A. (Received 12 April 1967; in revised form 17 May 1967) There is a great deal of interest in the stereochemistry of the Baeyer-Villiger reaction and the nature of the factors which influence the preferential migration of one group over another (2,3).
This communication presents evidence for a free radical
pathway for this reaction in addition to the accepted cationic mechanism (4,5,6). We have determined the migratory aptitudes of substituted phenyl in the BaeyerVilliger reaction for a series of p,p' -unsymmetrically substituted benzophenones. peroxyacetic acid and trifluoroperoxyacetic
Both
acid in refluxing methylene chloride were
investigated as the oxidizing agents in the reaction.
Trifluoroacetic acid was used es
the acid catalyst (5) in all the reactions studied to facilitate the nucleophilic attack of the peroxyacid on the carbonyl carbon.
The ester products from each reaction were
hydrolyzed, and the relative yields of the two benzoic acids determined spectrophotometricsllg.
These data were used to obtain the migratory aptitudes shown in Tables I and II.
When trifluoropero=Jacetic in Table I lere obtained.
acid was used as the oxidizing agent the results shown
The typical order for activation (to>rard electrophiles) of
the ring is found, that is, activating groups favor migration and electron-rrithdralring groups retard migration.
A Hammett plot of the data gives a good line :,ritha pvalue
of -0.77 (the migration step).
This value is in agreement with the results of Hawthorne
and Emmons (5), rrho obtained n pvalue
of -1.10 for the total reaction in ethylene chloride.*
Part of the discrepancy between the tT,ro values is due to the different values of oused to obtain p.
The value of -0.77 was obtained from a (I+plot and the value of -1.10
was obtained from a uplot.
The fact that both values are small and negative adds support
to the accepted mechanism.
The migratory aptitudes are in agreement with what would be
eqpected from a cationic transition state. *
Hauthorne and Emmons obtained their value by measuring the overall rate of the reaction a.?ldsuggest that for&he rearrangement step is a major contributor to the reported p value.
2871
No.30
2872
TABLE I RelativeIiigratory Aptitude of p-SubstitutedPhenyl Groups Using Trifluoroperoxyacetic Acid
p-Substituent RQ2 t-Butyl Cl
Migratory Aptitude
p-Substituent
Migratory Aptitude
1.00
Br
13.94
7.93
CH
26.05
3
12.15
CH30
251.00
When peroxyaceticaci( was used as the oxidizingagent, the results shown in Table II were obtained. These data do not supportan ionic mechanism. A Hamnett plot of the data, in contrastto the earlier resultswith trifluoroperoxyacetic acid, shows no correlation. Examinationof Table II re\.eals that groups which are deactivatingrelativeto hydrogen in aromaticelectrophilicEubstitutionactuallyenhancemigrationof the phenyl ring to which they are attached. This order of migrationis observedin known free radical migrations. Bartlett and Cotman (7) studiedthe thermal decompositionof p-nitrotriphenylmethylhydroperoxideand found the ratio of phenyl migrationto p-nitrophenyl migrationwas from 1:3 to 1:4 and suggestedthat enhancedp-nitrophenylmigrationis a good test for a free radicalmechanism. TABLE II RelativeYigratoq Aptitude of p-SubstitutedPhenyl Groups Using PeroxyeceticAcid
p-Substituent
Migratory Aptitude
p-Substituent
Kigratory Aptitude
H
1.00
CR
8.63
Cl
1.22
OH
26.80
Br
2.14
CH
152.50
B02
3.20
3
3
Augood, Hey and Williams (8) determinedthe rates at which phenyl radicals,generated by the decompositionof benzoyl peroxide,attack benzene, cblorobenzene and nitrobenzene. The observedrate ratios are 1 (benzene):l.T(chlorobenzene):4.0 (nitrobenzene) which are in good agreementwith the relativerates of migrationwhich we-have observedfor these
2873
No.30
three groups in the Baeyer-VilligerReaction;1 (benzene):1.2(chlorobenzene):3.2 (nitrobenzene). The chsnge in mechanisnwith the change in oxidizingagent can be explainedby the followingrationale: The Criegee intermediate,formedby the attack of the peracid on the carbonylcarbon of the ketone-acidcomplex,may decomposeby either heterolyticor homolyticcleavageof the oxygen-oxygenbond. The acceptedcationicmechanismcalls for heterolyticcleavageof this bond, a process which is enhancedby the strong inductive effect of the trifluoroacetyl group, however,when peroxyaceticacid is used as the oxidantthe much smaller inductiveeffect of tineacetyl group favors homolyticcleavage It is possiblethat the more active aryl and leads to a free radical rearranCe;:ent. Croups still nigrateby a cationicmechanismor perhapsboth mechenisnsare open to these groups. :!efeel slat in as much as all substituentswill stabilizea free radical relativeto hydrogenthat the enhancedmigrationof both +R and -R groups relativeto hydrogenis not "a priori" evidenceof a change in nechonisnwith a change in the nature of tinesubstituentgroup. Other factorsthat undoubtedlyaffect the mechenisn are the ionizingpower of the solvent,-thenature of the acid catalystand the temperature of the reaction. It might be expectedthat a chenge in mechanismwould lead to a change in products. ..rylmigrationto oxygen radicalsis well documented(7,9) so the ester product resulting from such a migrationin the Baeyer-Villigerreactionshouldbe no surprise. One would e,xpect to observe the evolutionof carbon dioxide from aceto:ri radicalsproducedin the homolyticcleavage. We have not observeda gros8 evolutionof Gas from the system end are current& ,unning experimentsto determinethe yield of carbon dioxide. A lo7 yield of carbon dioxiderrouldnot be uneqected. It is easy to visualizethe free ace-toxy radical abstractinga hydrogen from the grotonatedoxygen of the incipientester while within the solventcage.* It is in an excellentpositionto do so, the carbonyloxygen of the scetoxy Croup is very likely hydrogenbonded to this proton in the CrieCee intermediate, end this conformationis likely to continueeven during the homolytic cleavage. The free radicalmechanismexplainsthe anomalousmigrationsof p-chlorophenylsnd c
Xartin, Taylor and Drew (10) have observedthe reactionof acetoxy radicalswith cyclohexenewithin the solventcage to form the ester and report that this reaction is com@itive yrithdecarboxylation.
No.30
p-bromophenylduring the Bayer-Vllliger oxidation(usq
peroxyacetlcacid) of the
p-halobenzophenones in the absence of sulfuricacid as observedby Doering and Speers (ll). A free radicalmechanism#isconsistent-dth the data for our systemunder the conditionsemployed;substitutedbenzophenonesoxidizedw&th peroxyaceticacid in nethylene.chlorlde solventusing trifluoroacetic acid as the acid catalyst.
REFERENCES 1. Taken from the Ph.D. Dissertationof A. Swelim. 2. P.A.S. Smith, Molecula:Rearrangements, Paul de Mayo, ed., Part I, p. 577, Interscience Publishers,New York (~~63). 3.
LG.
Israel,Ph.D. Dissertation,Universityof New Hampshire,1962.
4. M.F. Hawthorne,W.D. Eimnonsand K.S. McCsllum,J. Am. Chem. Sot., 8C, 6393 (1958). 5. M.F. Hawthorneand W.D. Emmons, ibid.,&,
6398 (1958).
6. E. Hedaya and S. Wlnstzin,TetrahedronLetters,No. 13, 563 (1962). 7. C.D. Barlett and J.D. l:otman, Jr., J. Am. Chem. Sot., 72, 3095 (1950). 8. D.R. Augood, 1C.H.Hey 8LndG,H, Williams,J. Chem. Sot., 1952, 3412. J. Org. Chem., 29, 921 (1964). 9. J.W. Wilt and J.A. Lunliquist, 10. J.C. Martin, J.W. Taylor and E.H. Drew, J. Am. Chem. Sot., 5, 11. W. von E. Doering and :,.Spsers, ibid.,72, 5515 (1950).
129 (1967).
F'FEERADICAL PATHWAY FOB THE BAEYER-VILLIGER FEACTIOB J.C. Robertson and A. Swelim (1) Department of Chemistry, Colorado State University, Fort Collins, Colorado, U.S.A. (Received 12 April 1967; in revised form 17 May 1967) There is a great deal of interest in the stereochemistry of the Baeyer-Villiger reaction and the nature of the factors which influence the preferential migration of one group over another (2,3).
This communication presents evidence for a free radical
pathway for this reaction in addition to the accepted cationic mechanism (4,5,6). We have determined the migratory aptitudes of substituted phenyl in the BaeyerVilliger reaction for a series of p,p' -unsymmetrically substituted benzophenones. peroxyacetic acid and trifluoroperoxyacetic
Both
acid in refluxing methylene chloride were
investigated as the oxidizing agents in the reaction.
Trifluoroacetic acid was used es
the acid catalyst (5) in all the reactions studied to facilitate the nucleophilic attack of the peroxyacid on the carbonyl carbon.
The ester products from each reaction were
hydrolyzed, and the relative yields of the two benzoic acids determined spectrophotometricsllg.
These data were used to obtain the migratory aptitudes shown in Tables I and II.
When trifluoropero=Jacetic in Table I lere obtained.
acid was used as the oxidizing agent the results shown
The typical order for activation (to>rard electrophiles) of
the ring is found, that is, activating groups favor migration and electron-rrithdralring groups retard migration.
A Hammett plot of the data gives a good line :,ritha pvalue
of -0.77 (the migration step).
This value is in agreement with the results of Hawthorne
and Emmons (5), rrho obtained n pvalue
of -1.10 for the total reaction in ethylene chloride.*
Part of the discrepancy between the tT,ro values is due to the different values of oused to obtain p.
The value of -0.77 was obtained from a (I+plot and the value of -1.10
was obtained from a uplot.
The fact that both values are small and negative adds support
to the accepted mechanism.
The migratory aptitudes are in agreement with what would be
eqpected from a cationic transition state. *
Hauthorne and Emmons obtained their value by measuring the overall rate of the reaction a.?ldsuggest that for&he rearrangement step is a major contributor to the reported p value.
2871
No.30
2872
TABLE I RelativeIiigratory Aptitude of p-SubstitutedPhenyl Groups Using Trifluoroperoxyacetic Acid
p-Substituent RQ2 t-Butyl Cl
Migratory Aptitude
p-Substituent
Migratory Aptitude
1.00
Br
13.94
7.93
CH
26.05
3
12.15
CH30
251.00
When peroxyaceticaci( was used as the oxidizingagent, the results shown in Table II were obtained. These data do not supportan ionic mechanism. A Hamnett plot of the data, in contrastto the earlier resultswith trifluoroperoxyacetic acid, shows no correlation. Examinationof Table II re\.eals that groups which are deactivatingrelativeto hydrogen in aromaticelectrophilicEubstitutionactuallyenhancemigrationof the phenyl ring to which they are attached. This order of migrationis observedin known free radical migrations. Bartlett and Cotman (7) studiedthe thermal decompositionof p-nitrotriphenylmethylhydroperoxideand found the ratio of phenyl migrationto p-nitrophenyl migrationwas from 1:3 to 1:4 and suggestedthat enhancedp-nitrophenylmigrationis a good test for a free radicalmechanism. TABLE II RelativeYigratoq Aptitude of p-SubstitutedPhenyl Groups Using PeroxyeceticAcid
p-Substituent
Migratory Aptitude
p-Substituent
Kigratory Aptitude
H
1.00
CR
8.63
Cl
1.22
OH
26.80
Br
2.14
CH
152.50
B02
3.20
3
3
Augood, Hey and Williams (8) determinedthe rates at which phenyl radicals,generated by the decompositionof benzoyl peroxide,attack benzene, cblorobenzene and nitrobenzene. The observedrate ratios are 1 (benzene):l.T(chlorobenzene):4.0 (nitrobenzene) which are in good agreementwith the relativerates of migrationwhich we-have observedfor these
2873
No.30
three groups in the Baeyer-VilligerReaction;1 (benzene):1.2(chlorobenzene):3.2 (nitrobenzene). The chsnge in mechanisnwith the change in oxidizingagent can be explainedby the followingrationale: The Criegee intermediate,formedby the attack of the peracid on the carbonylcarbon of the ketone-acidcomplex,may decomposeby either heterolyticor homolyticcleavageof the oxygen-oxygenbond. The acceptedcationicmechanismcalls for heterolyticcleavageof this bond, a process which is enhancedby the strong inductive effect of the trifluoroacetyl group, however,when peroxyaceticacid is used as the oxidantthe much smaller inductiveeffect of tineacetyl group favors homolyticcleavage It is possiblethat the more active aryl and leads to a free radical rearranCe;:ent. Croups still nigrateby a cationicmechanismor perhapsboth mechenisnsare open to these groups. :!efeel slat in as much as all substituentswill stabilizea free radical relativeto hydrogenthat the enhancedmigrationof both +R and -R groups relativeto hydrogenis not "a priori" evidenceof a change in nechonisnwith a change in the nature of tinesubstituentgroup. Other factorsthat undoubtedlyaffect the mechenisn are the ionizingpower of the solvent,-thenature of the acid catalystand the temperature of the reaction. It might be expectedthat a chenge in mechanismwould lead to a change in products. ..rylmigrationto oxygen radicalsis well documented(7,9) so the ester product resulting from such a migrationin the Baeyer-Villigerreactionshouldbe no surprise. One would e,xpect to observe the evolutionof carbon dioxide from aceto:ri radicalsproducedin the homolyticcleavage. We have not observeda gros8 evolutionof Gas from the system end are current& ,unning experimentsto determinethe yield of carbon dioxide. A lo7 yield of carbon dioxiderrouldnot be uneqected. It is easy to visualizethe free ace-toxy radical abstractinga hydrogen from the grotonatedoxygen of the incipientester while within the solventcage.* It is in an excellentpositionto do so, the carbonyloxygen of the scetoxy Croup is very likely hydrogenbonded to this proton in the CrieCee intermediate, end this conformationis likely to continueeven during the homolytic cleavage. The free radicalmechanismexplainsthe anomalousmigrationsof p-chlorophenylsnd c
Xartin, Taylor and Drew (10) have observedthe reactionof acetoxy radicalswith cyclohexenewithin the solventcage to form the ester and report that this reaction is com@itive yrithdecarboxylation.
No.30
p-bromophenylduring the Bayer-Vllliger oxidation(usq
peroxyacetlcacid) of the
p-halobenzophenones in the absence of sulfuricacid as observedby Doering and Speers (ll). A free radicalmechanism#isconsistent-dth the data for our systemunder the conditionsemployed;substitutedbenzophenonesoxidizedw&th peroxyaceticacid in nethylene.chlorlde solventusing trifluoroacetic acid as the acid catalyst.
REFERENCES 1. Taken from the Ph.D. Dissertationof A. Swelim. 2. P.A.S. Smith, Molecula:Rearrangements, Paul de Mayo, ed., Part I, p. 577, Interscience Publishers,New York (~~63). 3.
LG.
Israel,Ph.D. Dissertation,Universityof New Hampshire,1962.
4. M.F. Hawthorne,W.D. Eimnonsand K.S. McCsllum,J. Am. Chem. Sot., 8C, 6393 (1958). 5. M.F. Hawthorneand W.D. Emmons, ibid.,&,
6398 (1958).
6. E. Hedaya and S. Wlnstzin,TetrahedronLetters,No. 13, 563 (1962). 7. C.D. Barlett and J.D. l:otman, Jr., J. Am. Chem. Sot., 72, 3095 (1950). 8. D.R. Augood, 1C.H.Hey 8LndG,H, Williams,J. Chem. Sot., 1952, 3412. J. Org. Chem., 29, 921 (1964). 9. J.W. Wilt and J.A. Lunliquist, 10. J.C. Martin, J.W. Taylor and E.H. Drew, J. Am. Chem. Sot., 5, 11. W. von E. Doering and :,.Spsers, ibid.,72, 5515 (1950).
129 (1967).