Mechanism of oxygen transfer in the epoxidation of an olefin by molecular oxygen in the presence of an aldehyde

Article abstract of DOI:10.1021/ol991304x

(formula presented) The reaction pathway for peroxide-initiated aldehyde-mediated oxidation of olefins to epoxides by molecular oxygen has been studied. The pathways of reaction via a peroxy acid or an acyl peroxy radical have been differentiated by investigation of the reaction of 4 with oxygen to provide 6 via 8.

Full text of DOI:10.1021/ol991304x

ORGANIC
LETTERS
2000
Vol. 2, No. 3
357-360
Mechanism of Oxygen Transfer in the
Epoxidation of an Olefin by Molecular
Oxygen in the Presence of an Aldehyde
Stephen G. Jarboe and Peter Beak*
Department of Chemistry, UniVersity of Illinois at Urbana-Champaign,
Urbana, Illinois 61801
beak@aries.scs.uiuc.edu
Received December 2, 1999
ABSTRACT
The reaction pathway for peroxide-initiated aldehyde-mediated oxidation of olefins to epoxides by molecular oxygen has been studied. The
pathways of reaction via a peroxy acid or an acyl peroxy radical have been differentiated by investigation of the reaction of 4 with oxygen to
provide 6 via 8.
Development of oxidation processes using molecular oxygen
as a safe, inexpensive, and readily available oxidant is of
considerable interest. While a number of metal catalysts have
been used to achieve this end, Kaneda and co-workers
recently reported an uncatalyzed, stereospecific epoxidation
of olefins by pretreatment of an aldehyde with molecular
oxygen.1,2 Valentine and Mizuno each report a similar, but
in situ, epoxidation protocol that provides epoxides nonste-
reospecifically.3 The present work identifies the active
oxygen transfer agents, as well as the mechanism, of the
epoxidation reactions.
were all initially present, the yields were highly variable.
We found the yields to be reproducible if 10-15 mol % of
tert-butyl hydroperoxide was added to the aldehyde and
olefin immediately following the addition of oxygen. Control
experiments established that epoxidation did not proceed in
the absence of aldehyde with or without the peroxide present.
Our studies have been carried out with 10 mol % of the
peroxide.4
The requirement of the aldehyde and an initiator suggests
that the reaction proceeds via formation of acyl radical 1
which reacts with oxygen to afford acyl peroxy radical 2.
This species could then lead to the corresponding peroxy
acid 3. Kaneda favors 3, while Valentine and Mizuno favor
2, as the active oxidizing agent (Scheme 1).
Reaction of pivaldehyde with oxygen for 30 min, followed
by addition of trans-stilbene, provided the trans-epoxide in
good yields. However, when the aldehyde, olefin, and oxygen
We have used the reaction of substrate 4 to differentiate
between the two possible oxidants for an in situ reaction. If
the reaction proceeds via the peracid as the oxidant, the
expected product would be the epoxide 5, formed by an
(1) (a) Catalytic ActiVation of Dioxygen by Metal Complexes; Simandi,
L. I., Ed.; Kluwer Academic Publishers: Dordrect, The Netherlands, 1992;
pp 318-331. (b) Yamada, T.; Takai, T.; Rhode, O.; Mukaiyama, T. Bull.
Chem. Soc. Jpn. 1991, 64, 2109. (c) Haber, J.; Mlodnicka, T.; Poltowicz,
J. J. Mol. Catal. 1989, 54, 451.
(2) Kaneda, K.; Haruna, S.; Imanaka, T.; Hamamoto, M.; Nishiyama,
Y.; Ishii, Y. Tetrahedron Lett. 1992, 33, 6827.
(3) (a) Nam, W.; Kim, H. J.; Kim, S. H.; Ho, R. Y. N.; Valentine, J. S.
Inorg. Chem. 1996, 35, 1045. (b) Mizuno, N.; Weiner, H.; Finke, R. G. J.
Mol. Catal. A 1996, 114, 15.
(4) We believe that the reactions without added initiator are being initiated
by adventitious initiators and propagated by a radical chain mechanism (vide
infra).
10.1021/ol991304x CCC: $19.00 © 2000 American Chemical Society
Published on Web 01/12/2000

Scheme 1
Scheme 3
intermolecular reaction.5 If the oxidant is the acyl peroxy
radical, the epoxy acid 5 is a possible product, and it could
arise by either an intramolecular or intermolecular pathway.
The enal 4 was synthesized by reaction of the potassium
enolate of isobutyraldehyde with trans-cinnamyl bromide.6
We find reaction of 4 with molecular oxygen in the presence
of 10 mol % of tert-butyl hydroperoxide to provide 3-hy-
droxy-5,5-dimethyl-2-phenyltetrahydrofuran 6 in 80% yield
as a single diastereomer, after extractive workup with
aqueous sodium bisulfite (Scheme 2).7
NMR spectrum of this compound is consistent with the
structure assigned as 8.8 Treatment of 8 with triphenylphos-
phine provides epoxy alcohol 9. This alcohol cyclizes
quantitatively to 6 upon exposure to an aqueous bisulfite
wash.
We propose the mechanism in Scheme 4 to account for
the formation of 6 via 8 from 4. This mechanism begins
with an aldehyde autoxidation, which is followed by in-
tramolecular olefin epoxidation of the acyl peroxy radical
11.3,9,10 The oxygen transfer occurs via 12 to provide 13,
which undergoes decarboxylation to 14. Reaction of 14 with
molecular oxygen affords the radical chain propagating
species 15 and subsequently the peroxide 8.11,12 Reaction of
8 with NaHSO3 to 9 followed by a 6-endo-tet ring opening
of the epoxide then provides 6.13 A number of experiments
were performed to evaluate this reaction pathway.
Scheme 2
Scheme 4. Proposed Reaction Mechanism
The structure of 6 was confirmed by conversion to
p-bromobenzoyl ester 7, which was unambiguously identified
by X-ray crystallographic analysis of crystals grown by slow
evaporation of Et2O. The X-ray crystal structure reveals that
7 has a trans relationship between the phenyl and hydroxyl
groups.
Examination of the reaction by TLC prior to extractive
workup reveals a single product, different from the cyclic
ether 6, along with starting material. This product affords 6
1
upon exposure to aqueous NaHSO3 (Scheme 3). The H
(5) For evidence establishing an intramolecular peracid epoxidation
cannot occur within a small endocyclic ring, see: Woods, K. W.; Beak, P.
J. Am. Chem. Soc. 1991, 113, 6281. For a review of the endocyclic restriction
test, see: Beak, P. Acc. Chem. Res. 1992, 25, 215.
(6) Groenewegen, P.; Kallenberg, H.; van der Gen, A. Tetrahedron Lett.
1978, 5, 491.
(7) Control experiments were performed in the absence of either the
aldehyde functionality or the oxygen atmosphere to test for direct oxidation
by the initiator. No oxidation was detected.
To determine the molecularity of the epoxidation, a
mixture of 4 and the methyl ester 16 was employed in an
358
Org. Lett., Vol. 2, No. 3, 2000

application of the endocyclic restriction test.5 The potential
crossover experiment was performed with a 1.00:1.05 molar
ratio of 4 to 16, at a total olefin concentration of 60 mM, as
shown in Scheme 5. Normal reaction conditions provided a
Scheme 6
Scheme 5. Endocyclic Restriction Test
reaction time of 40 h, triphenylphosphine was added as the
reductant. Both the epoxy alcohol 9 and triphenylphosphine
oxide 18 were analyzed by FI/MS and FD/MS, respectively,
for isotopic oxygen incorporation. The isotopic enrichments
of 9 and 18 are given in Table 1.
1
mixture of 4, 6, and 16, as determined by both GC and H
NMR examination of the reaction mixture. A 40% conversion
of 4 to 6 was observed, while no oxidation product of 16
was detected. This result is consistent with the intramolecular
transfer of oxygen within 11, as shown in Scheme 4. If the
olefin of 16 is assumed to be as reactive as the olefin of 4,
an intermolecular oxygen transfer would have afforded
oxidation of both substrates. In a control experiment 16 was
shown to be oxidized to 17 by intermolecular oxygen transfer
with pivaldehyde, oxygen, and tert-butyl hydroperoxide.14
This evidence of intramolecular reaction establishes that
the oxygen transfer from the acyl peroxy radical may proceed
through a six-membered ring. A transition state geometry
for oxygen transfer of less than 180° is required for this
reaction, thus excluding the peroxy acid as the active oxidant
in this case.5
Table 1. Isotopic Distribution of Oxygen Transfer Reaction
exptla
calcd (1 equiv)b
calcd (2 equiv)b
9-16O16
9-16O16
9-18O18
O
O
O
27 ((5%)
50 ((5)
23 ((5%)
50
25
50
25
50
exptlc
calcdd
calcde
18-16
18-18
O
O
64 ((5%)
35 ((5%)
50
50
70
30
a Determined by FI/MS, assuming a 5% error in the analysis. b Calculated
assuming a statistical insertion of 16O-16O and 18O-18O and an intramo-
lecular transfer (vide supra). Number of equivalents refers to the number
of separate O2 molecules incorporated into the product. c Determined by
FD/MS, assuming a 5% error in the analysis. d Calculation assumes a
statistical insertion of 16O and 18O. e Calculation takes into account the
excess 18 formed relative to the recovered mass of 9.
To establish the stoichiometry of the reaction with respect
to oxygen, and to further evaluate the identity of 8, the
oxygen transfer reaction was performed using a 50/50
mixture of 18O-18O and 16O-16O (Scheme 6). After a
(8) Out of concern for the potential hazards of peroxides, this compound
was isolated only once in a small quantity.
(9) (a) Maslov, S. A.; Blyumberg, E. A. Russ. Chem. ReV. 1976, 45 (2),
155. (b) McNesby, J. R.; Heller, C. A. Chem. ReV. 1954, 54, 325.
(10) Filippova, T. V.; Blyumberg, E. A. Rus. Chem. ReV. 1982, 51 (2),
582.
(11) Martin, J. C.; Dombchik, S. A. AdV. Chem. Ser. 1968, 75, 269.
(12) The formation and decarboxylation of an acyl radical, followed by
oxygen trapping and hydroperoxide reduction, has been reported as a
synthetic method of converting a carboxylic acid to an alcohol, less a single
carbon unit. Barton, D. H. R.; Ge´ro, S. D.; Holliday, P.; Quiclet-Sire, B.;
Zard, S. Z. Tetrahedron 1998, 54, 6751.
(13) This cyclization is, according to Baldwin’s rules for ring closures,
a formally disfavored process. However, Hevko and co-workers have
reported a similar epoxide-opening reaction under basic conditions, where
the product of the 6-endo-cyclization is preferred over that from a 4-exo-
process. Baldwin, J. E. J. Chem. Soc., Chem. Commun. 1976, 734. Hevko,
J. M.; Dua, S.; Talyor, M. S.; Bowie, J. H. J. Chem. Soc., Perkin Trans. 2
1998, 1629.
From the data in Table 1, it is clear that two molecules of
oxygen are incorporated upon the intramolecular reaction of
4 to 9, consistent with the mechanism shown in Scheme 4.
The first equivalent donates the epoxide oxygen atom, while
the other oxygen atom is lost in the decarboxylation of 13.
Following the trapping of the second equivalent of oxygen
by radical 14, both atoms are incorporated within the
hydroperoxide functionality of 8. Upon reduction, 9 then
contains two atoms of oxygen, each from separate molecules.
The statistical distribution of non-, mono-, and di-labeled
epoxy alcohol 9 shows that these two atoms are incorporated
by the consumption of two separate oxygen molecules. This
result also rules out a pathway in which the initial step is
decarbonylation of the acyl radical 10, followed by formation
of a peroxy radical and intramolecular transfer to afford 9.
Triphenylphosphine oxide 18 is also produced in this
reaction. On the basis of the isotope ratio of the starting
(14) Product 17 is presumed to be formed by olefin epoxidation, followed
by ester carbonyl oxygen-assisted ring-opening of the epoxide and hydroly-
sis. This compound was provided in 94% conversion (1H NMR), while a
single recrystallization provided 12% of 17 suitable for X-ray crystal-
lographic analysis.
Org. Lett., Vol. 2, No. 3, 2000
359

oxygen atmosphere, it is be predicted that 50% of 18 isolated
would contain the 18O label. From the data in Table 1, it is
shown that there is considerable isotopic incorporation into
18, but it is in a 65:35 ratio. However, the recovered
quantities of 9 and 18 (0.23 and 0.39 mmol, respectively)
suggest that the oxidation of triphenylphosphine may be
occurring not only by 8 but by a secondary oxidant as well.
If the excess phosphine oxide is assumed to have been
formed during workup by an oxygen source with a natural
abundance of 18O, the calculated ratio of 18 to 18-18O is
∼70:30, in closer accordance with the experimental results.15
A stereochemical test has been used to probe the nature
of 12. cis-Olefin 19 was prepared using cis-cinnamyl
bromide.16 Reaction of 19 with initiator under an O2
atmosphere gave 6 as the major product (Scheme 7).17 This
sufficient lifetime for rotational equilibration and must
undergo preferential ring closure to 13, with the trans
arrangement of substituents.
The present experimental results provide support for the
formation of 6 from 4 by the path depicted in Scheme 4.
The intramolecular transfer of oxygen excludes the peroxy
acid as the active oxidant. Additionally, the loss of a carbon,
presumably as carbon dioxide, is consistent with the forma-
tion of the carboxyl radical 13 resulting from oxygen transfer
within 11. The formation of 6 from both 4 and 19 is further
support for radical 12 as an intermediate.
These results are consistent with the acyl peroxy radical
as the oxidizing agent under the in situ conditions of
Valentine and Mizuno. The epoxidation of olefins by
molecular oxygen in the presence of aldehydes may be
considered to proceed by formation of an acyl peroxy radical
followed by an addition-elimination mechanism of epoxi-
dation of the olefin. However, under the sequential conditions
of Kaneda, the acyl peroxy radical may give the peroxy acid,
thereby affording a different oxidizing agent.
Scheme 7
Acknowledgment. We are grateful for support of this
work by a grant from the National Science Foundation (98-
19422).
Supporting Information Available: Experimental pro-
cedures for the preparation of all compounds and intermedi-
ates. This material is available free of charge via the Internet
at http://pubs.acs.org. The crystallographic data for the
structures reported in this paper have been deposited with
the Cambridge Crystallographic Data Centre as supplemen-
tary publications CCDC-137494 and CCDC-137495. Copies
of the data can be obtained free of charge on application to
CCDC, 12 Union Road, Cambridge CB21EZ, U.K. (fax
(+44)1223-336-033; email deposit@ccdc.cam.ac.uk).
observed change in geometry of the olefin with respect to
the epoxide upon reaction with oxygen is consistent with
the pathway in Scheme 4. Intermediate 12 must have a
(15) A control experiment exposing triphenylphosphine to workup
conditions afforded only about 3% the phosphine oxide. This control,
however, does not include the possible oxidation by residual tert-butyl
hydroperoxide, nor does it exclude the formation of a phoshine complex in
the presence of the reaction mixture that is more easily oxidized during
workup. Thus, at this time, we are unable to provide a definitive source of
the excess phosphine oxide.
(16) To test for an initiator-influenced preisomerization path to the
E-olefin prior to epoxidation, tert-butyl hydroperoxide was stirred with 19
in the absence of O2. This control experiment resulted in no detectable
isomerization, indicating that the formation of 6 from 19 presumably
proceeds via 12.
OL991304X
1
(17) By H NMR, 6 is the major product (>70% by integration) of the
reaction of 19. Although the cis-diastereomer was not identified, we believe
that, if present, it would have constituted <10% of the total reaction mixture.
360
Org. Lett., Vol. 2, No. 3, 2000