Epoxide formation by ring closure of the cinnamyloxy radical

Article abstract of DOI:10.1021/ol005749t

(equation presented) Ar = 4-nitrobenzene The cinnamyloxy and oxiranyl benzyl radicals were generated by photolysis of alkyl 4-nitrobenzenesulfenates. The yet unprecedented epoxide ring formation from a primary alkoxy radical was observed. Experimental evidence supports the fact that the mode of ring opening of the oxiranyl carbinyl radical system is thermodynamically driven. B3LYP/6-31G* calculations indicate that the closed form of the radical is ～5 kcal/mol more stable than the open one.

Full text of DOI:10.1021/ol005749t

ORGANIC
LETTERS
2000
Vol. 2, No. 9
1251-1254
Epoxide Formation by Ring Closure of
the Cinnamyloxy Radical
Je´roˆme Amaudrut and Olaf Wiest*
Department of Chemistry and Biochemistry, UniVersity of Notre Dame,
Notre Dame, Indiana 46556-5670
owiest@nd.edu
Received March 2, 2000
ABSTRACT
The cinnamyloxy and oxiranyl benzyl radicals were generated by photolysis of alkyl 4-nitrobenzenesulfenates. The yet unprecedented epoxide
ring formation from a primary alkoxy radical was observed. Experimental evidence supports the fact that the mode of ring opening of the
oxiranyl carbinyl radical system is thermodynamically driven. B3LYP/6-31G* calculations indicate that the closed form of the radical is ∼5
kcal/mol more stable than the open one.
The oxiranyl carbinyl radical 1, shown in Scheme 1, usually
rearranges by a fast carbon-oxygen bond cleavage to form
the allyloxy radical 2,1 except for the rearrangement with
R3 or R4 ) phenyl or vinyl. In that case, a carbon-carbon
bond cleavage to form 3 has been reported.2 C-O bond
fragmentation forms a highly reactive oxygen-centered
radical from a carbon-centered radical due to the relief of
∼27.5 kcal/mol ring strain.
It was also incorporated in tandem sequences involving
[1,5] hydrogen abstractions4 or â-scissions2b,5 followed by
radical cyclizations. The mechanistic aspect of this rear-
rangement has attracted considerable interest, as the C-O
bond cleavage in an epoxide caused by a radical is a very
fast process, and the oxiranyl carbinyl radical 1 is diffi-
cult to detect or trap.6 Recently, it has been trapped for the
case where R1,R2 ) cyclohexyl, and an approximate rate
constant of k ) 3.2 × 1010 s-1 for the ring opening to the
(2) (a) Strogryn, E.; Gianni, M. H. Tetrahedron Lett. 1970, 3025-3028.
(b) Corser, D. A.; Marples, B. A.; Dart, R. K. Synlett 1992, 987-989. (c)
Murphy, J. A.; Patterson, C. W.; Wooster, N. F. J. Chem. Soc., Perkin
Trans. 1 1993, 405-410. (d) Breen, A. P.; Murphy, J. A.; Patterson, C.
W.; Wooster, N. F. Tetrahedron 1993, 49, 10643-10654.
Scheme 1. Modes of Ring Opening of the Oxiranyl Carbinyl
Radical
(3) (a) Barton, D. H. R.; Motherwell, R. S. H.; Motherwell, W. B. J.
Chem. Soc., Perkin Trans. 1 1981, 2363-2367. (b) Johns, A.; Murphy, J.
A. Tetrahedron Lett. 1988, 29, 837-840. (c) Gash, R. C.; MacCorquodale,
F.; Walton, J. C. Tetrahedron 1989, 45, 5531-5538.
(4) (a) Rawal, V. H.; Newton, R. C.; Krishnamurthy, V. J. Org. Chem.
1990, 55, 5181-5183. (b) Rawal, V. H.; Krishnamurthy, V. Tetrahedron
Lett. 1992, 33, 3439-3442. (c) Rawal, V. H.; Iwasa, S. Tetrahedron Lett.
1992, 33, 4687-4690. (d) Rawal, V. H.; Zhong, H. M. Tetrahedron Lett.
1992, 33, 5197-5200. (e) Rawal, V. H.; Krishnamurthy, V.; Favre, A.
Tetrahedron Lett. 1993, 34, 2899-2902.
(5) (a) Bowman, W. R.; Marples, B. A.; Zaidi, N. A. Tetrahedron Lett.
1989, 30, 3343-3344. (b) Kim, S.; Lee, S. Tetrahedron Lett. 1991, 32,
6575-6578. (c) Galatsis, P.; Millan, S. D.; Faber, T. J. Org. Chem. 1993,
58, 1215-1220.
The formation of an allyloxy radical by this process has
been used in synthesis for further radical reactions, such as
5-exo cyclization to a remote double bond.3
(1) This rearrangement was first reported in the following: Sabatino, E.
C.; Gritter, R. J. J. Org. Chem. 1963, 28, 3437-3440.
10.1021/ol005749t CCC: $19.00 © 2000 American Chemical Society
Published on Web 04/12/2000

allyloxy radical has been determined.7 This observed rate is
consistent with the results of high-level computational
studies, which predict a low barrier, <4 kcal/mol, for the
C-O bond cleavage of 1.8 Evidence for the reversibility of
this reaction, i.e., the cyclization of allyloxy radical 2 to
oxiranyl carbinyl radical 1 has also been reported for cases
when alkoxy radical 2 is either secondary9 or tertiary4b,5c,10
(R3 ) R4 ) alkyl). The rate of this cyclization was measured
experimentally at 70 °C (k ) 2 × 108 s-1)11 for R3 ) Ph
and R1 ) R2 ) R3 ) H.
Scheme 3. Synthesis of the Radical Precursor 4
Here, we report a study of the rearrangement of 1 with R1
) Ph and R2 )R3 ) R4 ) H. Radical 1 was trapped with an
arylthiyl radical when either 1 or 2 was generated, which
proved that cinnamyloxy radical 2 undergoes ring closure
to give epoxide 1. To the best of our knowledge, this is the
first example of a primary alkoxy radical undergoing epoxide
ring closure. This experimental evidence supports the
hypothesis that the ring opening of oxiranyl carbinyl radical
1 is not kinetically but thermodynamically governed, as
suggested by Ziegler.10d,11 Generation of radicals 1 and 2
was achieved by photolysis of alkyl 4-nitrobenzenesulfenates
4 and 5, as shown in Scheme 2. The p-nitrobenzene group
a ratio ∼2.5:1, determined by integration in 1H NMR
analysis. The major diastereoisomer was isolated and reacted
with triethylamine and 4-nitrobenzenesulfenyl chloride in
methylene chloride. Sulfenate 4 was obtained as a single
diastereoisomer in a 39% yield after purification.
Photolysis of sulfenates 4 was performed in a Rayonet
UV reactor equipped with 350 nm wavelength light bulbs
in deuteriobenzene as the solvent. The reaction was moni-
1
tored by H NMR analysis. The major products of the
reaction were identified by comparing their 1H NMR spectra
with the one obtained from an independently prepared
material, except for 1,2-epoxy-3-phenyl-3-propyl-4-nitroben-
zene sulfide 8, which was isolated by flash chromatography
and fully characterized. Cinnamaldehyde, 4-nitrobenzene
disulfide, and a 1:1 ratio of two diastereoisomers of sulfide
8 are the major products of the reaction. Our proposed
mechanism to account for their formation is shown in
Scheme 4. Photolysis of sulfenate 4 cleaves the O-S bond
Scheme 2. Sulfenates Are Versatile Precursors for the
Generation of Alkoxy and Alkyl Radicals
Scheme 4. Mechanism of Photolysis of 4 and 5
confers to the sulfenate both stability, owing to its electron-
withdrawing properties, and UV absorbance at ∼350 nm.
Light of that wavelength cleaves the oxygen-sulfur bond
homolytically, forming an alkoxy radical as well as a sulfenyl
radical. When the alkoxy radical is tertiary, it will undergo
a â-scission, losing a molecule of acetone, and give an alkyl
radical. This versatile method for the generation of both
alkoxy and alkyl radicals has been used to study the mode
of ring opening of the oxyranyl methyl radicals.12
Cinnamyl 4-nitrobenzenesulfenate 5 was prepared by the
reaction of cinnamyl alcohol with 4-nitrobenzenesulfenyl
chloride in the presence of triethylamine.13 Sulfenate 4 was
prepared via the synthesis shown in Scheme 3. Cinnamyl
chloride was reacted with acetone in the presence of zinc in
an aqueous solution of ammonium chloride to yield 2-methyl-
3-phenylpent-4-en-2-ol 6 in 91% yield. Epoxidation of the
terminal alkene was achieved using m-chloroperbenzoic acid
in methylene chloride. The corresponding epoxy alcohol 7
was obtained in 82% as a mixture of diastereoisomers with
homolytically, forming the 4-nitrobenzenethiyl radical as well
as the corresponding tertiary alkoxy radical. The later
undergoes a â-scission, forming oxiranyl benzyl radical 1.
In this process, a molecule of acetone is liberated and its
1
formation was detected by H NMR analysis. Coupling of
1252
Org. Lett., Vol. 2, No. 9, 2000

the oxiranyl benzyl radical and the arylthiyl radical forms
sulfide 8. Epoxide ring opening also occurs via a C-O bond
cleavage to form cinnamyloxy radical 2. Coupling between
this radical and the arylthiyl radical produces sulfenate 5.
During the course of the reaction, 5 appears as an intermedi-
ate that is slowly consumed, as can be observed by the
disappearance of the doublet at δ ) 4.14 ppm in Figure 1.
tionation. The disulfide is formed by self-coupling of the
arylthiyl radical.
Photolysis of sulfenate 5 was performed under similar
conditions and afforded very similar products. This implies
that cinnamyloxy radical 2 undergoes ring closure to oxiranyl
benzyl radical 1, which is subsequently coupled with the
arylthiyl radical to form sulfide 8.
This observation is consistent with the previously observed
reversibility of the oxiranyl carbinyl rearrangement. It
supports the fact that the ring opening of the oxiranyl carbinyl
radical is not kinetically but thermodynamically governed.10d,11
Therefore, in the photolysis of 4, sulfide 8 is the product of
a coupling reaction rather than a “trapping” reaction, which
suggests a kinetically controlled mechanism.
To further investigate the formation of 8, the relative
energies of the cinnamyloxy- and oxiranylbenzyl radicals
were calculated, as well as their relative coupling products,
i.e., cinnamyl benzylsulfenate and 1,2-epoxy-3-phenyl-3-
propyl phenyl sulfide. All geometries were optimized at the
B3LYP/6-31G*14 level of calculation15 using Gaussian 9816
and are shown in Figure 2. Interestingly, these theoretical
Figure 1. Partial 1H NMR spectra, in C6D6, of the products of the
photolysis of sulfenate 4 as a function of time. δ 3.37 (m, sulfenate
4, methine H of epoxide), 3.9 (d, 1st diastereoisomer of sulfide 8,
H R to epoxide), 4.02 (d, 2nd diastereoisomer of sulfide 8, H R to
epoxide), 4.14 (d, sulfenate 5, 2H on C R to O).
Figure 2. Relative energies in kcal/mol of the open and closed
forms of the oxiranyl benzyl radical as well as their respecting
coupling products. Geometries optimized at the B3LYP/6-31G*
level.
Under photolytic conditions, the O-S bond of sulfenate 5
is homolytically cleaved, forming back cinnamyloxy radical
2 as well as the arylthiyl radical. Formation of cinnamalde-
hyde can be explained by hydrogen abstraction on the
cinnamyloxy radical by the arylthiyl radical or dispropor-
considerations show that the closed form of the cinnamyloxy
radical is more stable by ∼5 kcal/mol than the open form.
(6) (a) Krosley, K. W.; Gleicher, G. J.; Clapp, G. E. J. Org. Chem. 1992,
57, 840-844. (b) Laurie, D.; Nonhebel, D. C.; Suckling, C. J.; Walton, J.
C. Tetrahedron 1993, 49, 5869-5872. (c) Chatgilialoglu, C.; Ingold, K.
U.; Scaiano, J. C. J. Am. Chem. Soc. 1981, 103, 7739-7742. (d) Newcomb,
M.; Glenn, A. G. J. Am. Chem. Soc. 1989, 111, 275-277. (e) Bowry, V.
W.; Lusztyk, J.; Ingold, K. U. J. Am. Chem. Soc. 1991, 113, 5687-5698.
(f) Johns, A.; Murphy, J. A.; Patterson, C. W.; Wooster, N. F. J. Chem.
Soc., Chem. Commun. 1987, 1238-1240. (g) Dickinson, J. M.; Murphy, J.
A.; Patterson, C. W.; Wooster, N. F. J. Chem. Soc., Perkin Trans. 1 1990,
1179-1184. (h) Ayral-Kaloustian, S.; Agosta, W. C. J. Org. Chem. 1983,
48, 1718-1725. (i) Krosley, K. W.; Gleicher, G. J. J. Phys. Org. Chem.
1993, 6, 228-232.
(7) Krishnamurthy, V.; Rawal, V. H. J. Org. Chem. 1997, 62, 1572-
1573.
(8) (a) Pasto, D. J. J. Org. Chem. 1996, 61, 252-256. (b) Smith, D. M.;
Nicolaides, A.; Golding, B. T,; Radom, L. J. Am. Chem. Soc. 1998, 120,
10223-10233.
(9) Nussbaum, A. L.; Wayne, R.; Yuan, E.; Zagneetko, O.; Oliveto, E.
P. J. Am. Chem. Soc. 1962, 84, 1070-1071.
(10) (a) Weinberg, J. S.; Miller, A. J. Org. Chem. 1979, 44, 4722-
4725. (b) Suginome, H.; Wang, J. B. J. Chem. Soc., Chem. Commun. 1990,
1629-1631. (c) Galatsis, P.; Millan, S. D. Tetrahedron Lett. 1991, 32,
7493-7496. (d) Ziegler, F. E.; Petersen, A. K. J. Org. Chem. 1994, 59,
2707-2714.
Org. Lett., Vol. 2, No. 9, 2000
1253

The epoxide ring-containing radical is the thermodynamically
preferred product. This result is consistent with previous
calculations on the oxiranyl methyl and allyloxy radicals,
which predict the later to be more stable by ∼1 kcal/mol.8b
Introduction of a phenyl stabilizes preferentially the oxiranyl
carbinyl radical by approximately 5-7 kcal, consistent with
our calculated value.
Photolysis of sulfenate 4 yields first sulfenate 5 as an
intermediate that is observed by 1H NMR analysis while no
sulfide 8 is detected, as shown in Figure 1. The different
rates of coupling of 1 and 2 with the arylthiyl radical are
probably at the origin of this observation. Even thought 1 is
more stable than 2, a rapid equilibrium exists between these
two forms. The arylthiyl radical seems to couple fast enough
to 2 rather than to 1, forming preferentially sulfenate 5 over
sulfide 8. Under photolytic conditions, sulfenate 5 undergoes
O-S bond homolysis, the exact reverse reaction of the
coupling reaction of 2 and the arylthiyl radical. However,
sulfide 8 does not react under photolytic conditions and once
formed accumulates over the course of the experiment.
Therefore, formation of sulfide 8 is suggested to be the result
of a photodynamic equilibrium involving the radicals 1, 2,
and arylthiyl as well as the light-reactive sulfenate 5 and
the light-unreactive sulfide 8.
In conclusion, experimental evidence suggests that the
rearrangement of the oxiranyl carbinyl radical is thermody-
namically driven. In light of these results, the assumption
that the ring closure of the allyloxy radical is negligible in
the estimation of the rate of ring opening of the oxiranyl
methyl radical seems to be incorrect.7 Thus, the proposed
rate constant k ) 3.2 × 1010 s-1 is likely to be a lower limit
of the actual one.
(11) Ziegler, F. E. J. Org. Chem. 1995, 60, 2666-2667.
(12) Pasto, D. J.; Cottard, F.; Picconatto, C. J. Org. Chem. 1994, 59,
7172-7177.
Acknowledgment. We thank the late Dr. Daniel J. Pasto,
who initiated the use of sulfenates as radical precursors and
this work. We gratefully acknowledge support of this
research by the Department of Chemistry and Biochemistry
of the University of Notre Dame and the donors of the
Petroleum Research Fund, administered by the American
Chemical Society (Grant 27724-AC4).
(13) Amaudrut, J.; Wiest, O. J. Am. Chem. Soc. In press.
(14) (a) Lee, C.; Yang, W. Parr, R. G. Phys. ReV. B 1988, 37, 785-789.
(b) Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652.
(15) This method was found to be appropriate when compared with highly
accurate levels of calculations and experimental results. See reference in
footnote 13 and Amaudrut, J.; Pasto, D. J.; Wiest, O. J. Org. Chem. 1998,
63, 6061-6064. Differing results on the accuracy of the B3LYP method
were also published. See: Gregory, D. D.; Jenks, W. S. J. Org. Chem.
1998, 63, 3859-3865.
(16) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.;
Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A.
D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi,
M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.;
Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick,
D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.;
Ortiz, J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi,
I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M.
W.; Johnson, B. G.; Chen, W.; Wong, M. W.; Andres, J. L.; Head-Gordon,
M.; Replogle, E. S.; Pople, J. A. Gaussian 98, revision A.7; Gaussian, Inc.:
Pittsburgh, PA, 1998.
Supporting Information Available: Experimental pro-
cedures for the preparation and photolysis of radical precur-
sors 4 and 5. Cartesian coordinates of the structures presented
1
in Figure 2 with the corresponding energies and ZPEs. H
NMR, 13C NMR, and IR spectra of compounds 4, 5, and 8.
1H NMR spectra of the products of photolysis of 4 and 5.
This material is available free of charge via the Internet at
http://pubs.acs.org.
OL005749T
1254
Org. Lett., Vol. 2, No. 9, 2000