Facile autoxidation of 2-(4-hydroxyphenyl)-3,3-dimethylmethylenecyclopropane. The radical stabilizing ability of the phenoxide group

Article abstract of DOI:10.1021/ol9902576

(matrix presented). 2-(4-Hydroxyphenyl)-3,3-dimethylmethylenecyclopropane undergoes rapid reaction with O₂ at room temperature to give a dioxolane. A chain mechanism involving ring opening of a phenoxy radical is proposed. Conversion of the tit

Full text of DOI:10.1021/ol9902576

ORGANIC
LETTERS
1999
Vol. 1, No. 10
1615-1618
Facile Autoxidation of 2-(4-Hydroxy-
phenyl)-3,3-dimethylmethylenecyclopropane.
The Radical Stabilizing Ability of the
Phenoxide Group
Xavier Creary,* Allison Wolf, and Kevin Miller
Department of Chemistry and Biochemistry, UniVersity of Notre Dame,
Notre Dame, Indiana 46556
creary.1@nd.edu
Received August 30, 1999
ABSTRACT
2-(4-Hydroxyphenyl)-3,3-dimethylmethylenecyclopropane undergoes rapid reaction with O at room temperature to give a dioxolane. A chain
2
mechanism involving ring opening of a phenoxy radical is proposed. Conversion of the title compound to the phenoxide results in a remarkably
accelerated methylenecyclopropane rearrangement. Computational studies suggest that the intermediate biradical is greatly stabilized by the
phenoxide substituent.
The quantitative ability of substituents to stabilize benzylic
radicals has been a topic of interest. A number of free radical
stabilization scales have been developed over the years, and
these have been recently discussed.1 A scale developed in
our laboratory is based on the thermal rearrangement of
substituted methylenecyclopropanes 1 to 3, which proceeds
readily at 80 °C through a biradical (Scheme 1).2 Absent
from our scale is the σ• value for the p-OH group and the
corresponding deprotonated form. The first report on these
groups occurred when Adam used his EPR-based method
to evaluate the radical stabilizing effect of a number of
substituents, including the p-OH and p-O- substituents.3 We
now wish to report on the synthesis of the phenol derivative
1 (p-OH) as well as the thermal rearrangement of the
phenoxide 1 (p-O-).
Scheme 1
Our initial attempt to prepare the phenol derivative 1 (p-
OH) involved alkaline hydrogen peroxide oxidation of the
boronic acid 4 which we had available from a previous
study.2 However, this led to formation of the peroxide 5 as
the major product isolated after an aqueous workup. We next
prepared the tert-butyldimethylsilyl-protected methylenecy-
clopropane 6 by carbenoid addition to 1,1-dimethylallene.5
However, attempted desilylation of 6 with tetrabutylammo-
(1) Creary, X.; Engel, P. S.; Kavaluskas, N.; Pan, L.; Wolf, A. J. Org.
Chem. 1999, 64, 5634-5643.
(2) Creary, X.; Mehrsheikh-Mohammadi, M. E.; McDonald, S. J. Org.
Chem. 1987, 52, 3254-3263.
(3) Adam, W.; Harrer, H. M.; Kita, F.; Korth, H.-G.; Nau, W. M. J.
Org. Chem. 1997, 62, 1419-1426.
10.1021/ol9902576 CCC: $18.00 © 1999 American Chemical Society
Published on Web 10/09/1999

nium fluoride gave, after an aqueous workup, significant
amounts of the peroxide 5, along with the desired 1 (p-OH).
Careful exclusion of air during the workup is necessary in
order to isolate 1 (p-OH) (Scheme 2). Further experiments
Scheme 3
Scheme 2
showed that the phenol 1 (p-OH) readily absorbs oxygen
from air at room temperature, giving the peroxide 5. This
reaction can be easily monitored by NMR in CDCl3 and is
complete within an hour when the reaction is carried out
under an atmosphere of oxygen.
We felt that this remarkably facile autoxidation warranted
further investigation.6,7 The following mechanism is pro-
posed. The reaction is initiated by hydrogen atom abstraction
of the phenolic hydrogen. Ring opening of the strained
cyclopropane bond of 7 gives the allylic radical/quinone
methide 8. Capture of molecular oxygen at the center of
highest spin density (the tertiary carbon) leads to 9, which
can cyclize in a 5-exo fashion to give the radical 10.
Hydrogen atom abstraction by 10 from the starting material
1 (p-OH) generates the observed product and propagates the
cycle by regenerating 7 (Scheme 3).
To support this mechanism, methylencyclopropane 6 was
thermally rearranged to isopropylidenecyclopropane 11 and
then desilylated. The resultant phenolic product 12 was also
found to react rapidly with oxygen at room temperature. The
product formed was the identical peroxide 5 that was
produced when 1 (p-OH) was oxidized (Scheme 4). The
proposed common intermediate in the oxidation of both 1
(p-OH) and 12 is the radical 8. This study supports our
suggestion that this intermediate captures oxygen at the center
of highest spin density, i.e., the tertiary center. No product
13 that would have resulted from reaction of oxygen at the
primary end of the allylic radical 8 is formed. The intermedi-
ate peroxy radical 9 has two potential cyclization modes.
The lack of formation of product 15 suggests that the 5-exo
cyclization process in 9 to form 10 is significantly faster
that the 5-endo cyclization that would have led to 14 (and
ultimately 15).
Further insight into this facile autoxidation of 1 (p-OH)
can be gained by examining the substrates 16-18 (Scheme
5). Neither 16 or 17 reacts readily with oxygen at room
temperature or even at 80 °C. In the case of the meta
analogue 16, this lack of reactivity is attributed to the
relatively slow fragmentation of the cyclopropane bond of
the m-phenoxy radical due to its inability to form a quinone
methide structure. In the case of 17, the fragmentation of
the cyclopropane bond of phenoxy radical 20 is relatively
slow due to smaller ring strain in 20 (relative to 7) as well
as the decreased spin delocalization in radical 22 relative to
allylic radical 8. Hence, radical chain processes are not
readily established for 16 and 17. By way of contrast, phenol
18 is readily oxidized by molecular oxygen at room tem-
perature to peroxide 19. This is attributed to facile cyclo-
propane bond cleavage in phenoxy radical 21 due to the
tertiary benzylic nature of radical 23. Hence a radical chain
is readily established.
(4) Olofson, R. A.; Dougherty, C. M. J. Am. Chem. Soc. 1973, 95, 582-
584.
(5) Creary, X. J. Org. Chem. 1978, 43, 1777-1783.
(6) Formation of a dioxolane by reaction of singlet oxygen with
tetranisylcyclopropane has been reported. See: Akasaka, T.; Fukuoka, K.;
Ando, W. Tetrahedron Lett. 1991, 32, 7695-76.
(7) Photosensitized addition of O2 to cyclopropanes to form dioxolanes
has also been reported. See: Schaap, A. P.; Siddiqui, S.; Prasad, G.;
Palomino, E.; Lopwz, L. J. Photochem. 1984, 25, 167-181.
Attention was next turned to the thermal rearrangement
of phenoxide 1 (p-O-). This substrate was generated by
1616
Org. Lett., Vol. 1, No. 10, 1999

Scheme 4
Scheme 6
air-sensitive phenol 1 (p-OH). The thermal rearrangement
of anion 1 (p-O-) to isopropylidenecyclopropane 24 occurred
readily at room temperature (Scheme 6). By way of contrast,
rearrangement of protected material 6 or the parent substrate
1 (Ar ) C6H5) required 80 °C for convenient rates. Relative
rate data are shown in Figure 1. The rate enhancement of 1
Figure 1. Effect of aryl groups on the relative rearrangement rates
of 1 in DMSO-d6.
(p-O-) relative to the unsubstituted (p-H) derivative is a
factor of 2500 at 25 °C. This is, by far, the largest rate
enhancement that we have observed in the methylenecyclo-
propane rearrangement. The previous record holder was the
4-pyridyl-N-oxide substituted methylenecyclopropane, where
the rate enhancement was a factor of 75.8 In this case the
intermediate biradical was proposed to derive significant
stabilization from spin delocalization involving nitroxyl
radical forms.
Scheme 5
The facile rearrangement of 1 (p-O-) is attributed to
significant stabilization in biradical intermediate 25. Forms
such as 25a and 25b are recognizable as ketyl radicals, which
possess unusual stability.9
Computational studies have also been carried out in order
to gain further insight into the apparently large stabilization
reaction of silyl-protected derivative 6 with KOCH3 in
DMSO-d6. This procedure bypassed the need to isolate the
(8) Creary, X.; Mehrsheikh-Mohammadi, M. E.; McDonald, S. J. Org.
Chem. 1989, 54, 2904-2910.
Org. Lett., Vol. 1, No. 10, 1999
1617

of benzylic type radical 25 by the p-O- substituent. Density
functional calculations10 (B3LYP/6-31+G*) have been car-
ried out on simple benzylic radical 26 as well as compounds
27-29. The isodesmic reaction energy is 10.0 kcal/mol (9.8
p-methyl benzyl radical with toluene is only 0.4 kcal/mol at
the B3LYP/6-31+G* level and the reaction energy of the
highly stabilized 4-pyridyl-N-oxide radical [•CH2-C5H4NO]
with toluene is 5.5 kcal/mol. Calculated spin densities also
support the ketyl nature of radical 26, as reflected by a large
amount of positive spin density on the oxygen atom of 26.
Radical anion 26 is isoelectronic with the semiquinone radical
-
anion, [•O-C6H4O ]. However, the isodesmic reaction of
the semiquinone radical anion with toluene gives a calculated
reaction energy of 40.1 kcal/mol. This suggests that 26
(which is 10 kcal/mol more stable than the benzyl radical)
is substantially less stable than the semiquinone radical anion.
The stability order is therefore [•O-C6H4O -] > [•CH2-
-
C6H4O ] > [•CH2-C6H5].
Our findings are in contrast to those of Adam,3 who
reported no unusual radical stabilization by the phenoxide
group. We are uncertain of the reason for this large
discrepancy, but both our experimental and computational
studies indicate that the p-O- substituent is the best radical
stabilizer examined to date.
In summary, certain phenols substituted in the p-position
with cyclopropyl groups undergo room-temperature oxidation
with molecular oxygen to give dioxolanes. The reaction is
believed to involve fragmentation of the strained cyclopro-
pane bond of the intermediate phenoxy radical. In a related
study, the anionic phenoxy substituent provides an enormous
rate enhancement in the methylenecyclopropane rearrange-
ment and this has been attributed to large stabilization of
the radical-like transition state.
kcal/mol at B3LYP/6-31G*) and is indicative of a large
radical stabilization by the p-O- substituent. By way of
contrast, the analogous isodesmic reaction energy of the
(9) For a discussion, see: Hirota, N. In Radical Ions; Kaiser, E. T., Kevan,
L., Eds.; Interscience: New York, 1968; pp 35-85.
(10) Frisch, M, J.; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.;
Johnson, B. G.; Robb, M. A.; Cheeseman, J. R.; Keith, T.; Petersson, G.
A.; Montgomery, J. A.; Raghavachari, K.; Al-Laham, M. A.; Zakrzewski,
V. G.; Ortiz, J. V.; Foresman, J. B.; Peng, C. Y.; Ayala, P. Y.; Chen, W.;
Wong, M. W.; Andres, J. L.; Replogle, E. S.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Binkley, J. S.; Defrees, D. J.; Baker, J.; Stewart, J. P.; Head-
Gordon, M.; Gonzalez, C.; Pople, J. A. Gaussian 94, Revision B.3, Gaussian,
Inc., Pittsburgh, PA, 1995.
Acknowledgment. This research was supported by a
grant from the National Science Foundation.
OL9902576
1618
Org. Lett., Vol. 1, No. 10, 1999