Regioselective carbon-carbon bond cleavage in the oxidation of cyclopropenylcarbinols

Article abstract of DOI:10.1021/ol201581c

The strained double bond of cyclopropenylcarbinols undergoes a facile oxidation reaction to lead to unsaturated carbonyl derivatives. The distribution of the formed products depends on the relative stability of carbon-centered radical species, and the Sha

Full text of DOI:10.1021/ol201581c

ORGANIC
LETTERS
2011
Vol. 13, No. 15
4076–4079
Regioselective CarbonÀCarbon Bond
Cleavage in the Oxidation of
Cyclopropenylcarbinols
Ahmad Basheer,† Masaaki Mishima,‡ and Ilan Marek*,†
Mallat Family Laboratory of Organic Chemistry, Schulich Faculty of Chemistry and the
Lise Meitner-Minerva Center for Computational Quantum Chemistry, Technion-Israel
Institute of Technology, Technion City, Haifa 32000 Israel, and Institute for Materials
Chemistry and Engineering, Kyushu University, 6-10-1 Hakozaki, Higashi-ku, Fukuoka
812-8581, Japan
chilanm@tx.technion.ac.il
Received June 14, 2011
ABSTRACT
The strained double bond of cyclopropenylcarbinols undergoes a facile oxidation reaction to lead to unsaturated carbonyl derivatives. The
distribution of the formed products depends on the relative stability of carbon-centered radical species, and the Sharpless kinetic resolution leads
to enantiomerically pure BaylisÀHillman enal adducts.
Following his landmark discovery on the enantioselec-
tive epoxidation reaction of allylic alcohols,1 Sharpless
reported the application of the Ti(OPr-i)4/diisopropyl
tartrate catalyst system for the kinetic resolution of sec-
ondary allylic alcohols2 (SKR). The broad scope of the
reaction combined with the readily accessible catalyst
components led to a rapid adoption of the SKR by organic
chemists, and to date, it has certainly been one of the most
often used kinetic resolution reactions involving synthetic
catalysts.3 Therefore, when we had to prepare enantio-
merically enriched cyclopropenylcarbinol species 1, we just
considered them as particularly strained allylic alcohols
and the SKR was logically the method of choice. We were
indeed pleased to observe that, despite the very reactive
nature of the strained double bond, a very efficient
kinetic resolution occurs at À20 °C and the non-oxidized
cyclopropenylcarbinols 1 were obtained with very high
enantiomeric excess and yields (97À99% ee and 40À47%
yield of isolated products, Scheme 1).4 The latter could
then be used as starting materials for the preparation
of enantiomerically enriched alkylidenecyclopropanes5,6
and beyond that to a large variety of synthetically impor-
tant fragments possessing challenging acyclic all-carbon
quaternary stereogenic centers.7 The oxidized product,
(4) Simaan, S.; Masarwa, A.; Bertus, P.; Marek, I. Angew. Chem., Int.
Ed. 2006, 45, 3963.
(5) (a) Masarwa, A.; Stanger, A.; Marek, I. Angew. Chem., Int. Ed.
2007, 46, 8039. (b) Simaan, S.; Marek, I. Org. Lett. 2007, 9, 2569. (c)
Simaan, S.; Marek, I. Chem. Commun. 2009, 292.
(6) For reviews on the chemistry of cyclopropenes and alkylidenecy-
clopropanes, see: (a) Brandi, I. A.; Cicchi, S.; Cordero, F. M.; Goti, A.
Chem. Rev. 2003, 103, 1213. (b) Rubin, M.; Rubina, M.; Gevorgyan, V.
Chem. Rev. 2007, 107, 3117. (c) Goti, A.; Cordero, F. M.; Brandi, A.
Top. Curr. Chem. 1996, 178, 1. (d) Lautens, M.; Klute, W.; Tam, W.
Chem. Rev. 1996, 96, 49. (e) de Meijere, A.; Kozhushkov, S. I.; Khlebnikov,
A. F. Top. Curr. Chem. 2000, 207, 89. (f) Binger, P.; Schuchardt, U. Chem.
Ber. 1981, 114, 3313. (g) Marek, I.; Simaan, S.; Masarwa, A. Angew. Chem.,
Int. Ed. 2007, 46, 7364. (h) Audran, G.; Pelissier, H. Adv. Synth. Catal. 2010,
352, 575.
(7) (a) Simaan, S.; Goldberg, A. F. G.; Rosset, S.; Marek, I. Chem.;
Eur. J. 2010, 16, 774. (b) Simaan, S.; Marek, I. J. Am. Chem. Soc. 2010,
132, 4066. (c) Masarwa, A.; Marek, I. Chem.;Eur. J. 2010, 16, 9712. (d)
Masarwa, A.; Furstner, A.; Marek, I. Chem. Commun. 2009, 5760. (e)
Simaan, S.; Masarwa, A.; Zohar, E.; Stanger, A.; Bertus, P.; Marek, I.
Chem.;Eur. J. 2009, 15, 8449.
† Technion-Israel Institute of Technology.
‡ Kyushu University.
(1) Katsuki, T.; Sharpless, B. K. J. Am. Chem. Soc. 1980, 102, 5974.
(2) Martin, V. S.; Woodard, S. S.; Katsuki, T.; Yamada, Y.; Ikeda,
M.; Sharpless, B. K. J. Am. Chem. Soc. 1981, 103, 6237.
(3) Keith, J. M.; Larrow, J. F.; Jacobsen, E. N. Adv. Synth. Catal.
2001, 343, 5.
r
10.1021/ol201581c
Published on Web 07/06/2011
2011 American Chemical Society

To understand better the mechanism of formation of the
two enones, we first checked that this process was a general
one for a large variety of alkyl-substituted cyclopropenyl-
carbinol species 1 using either the classical Sharpless
epoxidation (method A) or the simplest peracid epoxida-
tion reaction (mCPBA, method B) as summarized in
Table 1. Whatever the method used (Sharpless epoxidation,
method A, or peracid oxidation, method B, Table 1,
entries 1À4), cyclopropenylcarbinols 1a,b always gave the
two enones 3 and 4 in equimolar amount and in good
isolated yield. As similar results were obtainedwiththe two
methods, we initially concentrated on the peracid oxida-
tion reaction for a rapid screening, and we found that, for
R1 = alkyl, whatever the substitutions of the cycloprope-
nylcarbinols (R2, R3 = alkyl, hydrogen, aryl; and R4 =
hydrogen or alkyl), the two isomeric enones were always
formed in a 1:1 ratio (Table 1, entries 1À8). We have also
performed DFT calculations for the transformation of
Scheme 1. Sharpless Kinetic Resolution (SKR) of Cycloprope-
nyl Alcohol: Expected Formation of BaylisÀHillman Adduct 4
namely, the putative chiral 2-oxabicyclo[1.1.0]butane (OBB)
2, leads through a hypothetical cleavage of the two
peripheral σ-bonds8 to the two enantiomerically enriched
R,β-unsaturated ketols 3 and 4 in equal amounts (each in
22À25% yield of isolated product).4 Oxabicyclo[1.1.0]-
butane species 2 have been postulated for more than
45 years as intermediates in various oxidation,9 thermal,10
or photochemical reactions,11 but they have never been
isolated nor spectroscopically detected. Although we were
alsonot ableto isolatesuchintermediates, the formation of
enantiomerically enriched products 1 led us to suggest that
the corresponding oxabicyclobutanes 2 were formed as
reactive intermediates through the SKR reaction.4 How-
ever, high-level ab initio calculations have shown that high
activation barriers characterize unimolecular as well as
acid-catalyzed fragmentation of oxabicyclobutanes, and
these should therefore be thermodynamically stable
molecules.12 Moreover, we were intrigued that the two
enones 3 and 4 were obtained in an exact equimolar
amount in the rearrangement process; this result was
particularly striking as titanium alcoholate in 2 should
have promoted a selective ring-opening reaction into 4
through an initial Payne rearrangement and subsequent
skeletal reorganization (Scheme 1).13
Table 1. Epoxidation Reaction of Cyclopropenylcarbinols 1
entry
R1
R2
R3 R4 method dr a 3:4 yield (%)b
1 (1a) Me
2 (1a) Me
3 (1b) Me
4 (1b) Me
5 (1c) Me
6 (1d) Me
7 (1e) Me
8 (1f) Me
9 (1g) Ph
10 (1g) Ph
11 (1h) Ph
12d(1i) Ph
13 (1j) Ph
14 (1k) Ph
15 (1l) Ph
16(1m) Ph
Ph
Ph
Ph
Ph
Ph
Me
H
H
A
Bc
A
B
B
B
B
B
B
A
B
B
B
B
B
B
B
B
B
B
B
A
B
B
A
1:1
1:1
1:1
1:1
1:1
1:1
1:1
1:1
1:0
1:0
1:0
1:0
1:0
1:0
1:0
1:0
1:0
1:0
1:0
0:1
0:1
0:1
0:1
0:1
0:1
98
96
70
65
88
80
89
65
84
80
82
90
79
72
77
72
94
91
81
65
66
65
60
61
67
H
H
H
Me
Me
H
H
Ph
Me H
(CH2)2Ph H
H
(CH2)2Ph Me Me
Me
Me
Ph
Ph
Me H
Me H
Ph
H
H
H
(CH2)2Ph H
H
Bu
H
H
H
H
H
H
H
H
H
H
H
iPr
H
iBu
H
17 (1n) Me3Si
pBrPh
Me
Me
Me
Me
Me
Me
Me
18 (1o) PhMe2Si pBrPh
19 (1p) Ph2MeSi pBrPh
(8) Kocienski, P. J.; Ciabattoni, J. J. Org. Chem. 1974, 39, 388.
(9) (a) Branan, B. M.; Wang, X.; Jankowski, P.; Wicha, J.; Paquette,
L. A. J. Org. Chem. 1994, 59, 6874. (b) Crandall, J. K.; Conover, W. W.
J. Org. Chem. 1978, 43, 1323. (c) Friedrich, L. E.; Leckonby, R. A.;
Stout, D. M.; Lam, Y.-S. P. J. Org. Chem. 1978, 43, 604. (d) Friedrich,
L. E.; Fiato, R. A. J. Org. Chem. 1974, 39, 416. (e) Friedrich, L. E.; Fiato,
R. A. J. Am. Chem. Soc. 1974, 96, 5783. (f) Ciabattoni, J.; Kocienski,
J. P. J. Am. Chem. Soc. 1971, 93, 4902. (g) Friedrich, L. E.; Cormier,
R. A. J. Org. Chem. 1970, 35, 451. (h) Ciabattoni, J.; Kocienski, J. P. J.
Am. Chem. Soc. 1969, 91, 6534.
(10) (a) Friedrich, L. E.; de Vera, N. L.; Lam, Y.-S. P. J. Org. Chem.
1978, 43, 34. (b) Marmor, S.; Thomas, M. T. J. Org. Chem. 1967, 32, 252.
(11) Corey, E. J.; Bass, J. D.; Le Mahieu, R.; Mitra, R. B. A. J. Am.
Chem. Soc. 1964, 86, 5570.
20 (1q) H
Ph
21 (1r)
22 (1r)
23 (1s)
24 (1t)
25 (1t)
H
H
H
H
H
pMeOPh
pMeOPh
iPr
(CH2)2Ph H Me
(CH2)2Ph H Me
a Diastereomeric ratio determined by 1H NMR. b Combined isolated
yields for all isomers after column chromatography (based on 1).
c Same results were obtained in either THF or toluene at À40 °C or rt.
d As aryl-substituted (R1 = Ph) secondary cyclopropenylcarbinol deri-
vatives (R3 = H) are unstable when neat, the oxidation reaction was
performed only with mCPBA directly on the crude reaction mixture in
solution.
(12) (a) Okovytyy, S.; Gorb, L.; Leszczynski, J. Tetrahedron 2001, 57,
1509. (b) Okovytyy, S.; Gorb, L.; Leszczynski, J. Tetrahedron 2002,
58, 8751.
(13) Hanson, R. M. Org. React. 2002, 60, 3.
Org. Lett., Vol. 13, No. 15, 2011
4077

oxabicyclo[1.1.0]butane 2 into enones and found, for all
substrates, that the barrier for enone 3 is lower than that
for enone 4 (>3 kcal/mol).14 These results suggest that the
two enones should notbe formedin equimolecular amount
in the reaction, which is opposite to our experimental
results. Oxidation of the double bond could also proceed
through an unsymmetrical transition state that could lead
to the formation of a biradical intermediate.15 If formation
of a biradical species was occurring, the equimolar forma-
tion of the two-biradical species A and B would conse-
quently lead to the equimolar formation of the two enones
3 and 4, respectively (Scheme 2). If correct, if one could
design a system leading to a single biradical entity, the
reaction should now be selective and lead to a single enone.
tertiary carbenium ion is pronounced. To distinguish
between these two possibilities (biradical vs zwitterionic),
silyl-substituted cyclopropenyl carbinols 1nÀ1p were pre-
pared and tested in our experimental conditions. Indeed,
the β-silicon effect (silicon hyperconjugation) leads to a
stabilization of carbocation in a β-position,16 and there-
fore, if a carbocationic species is involved, adduct 4 would
be obtained. However, if a radical is involved, silicon
stabilizes the radical in the R-position and 3 should be
formed. When 1nÀ1p were oxidized, a single isomer of
vinyl silanes 3nÀp was obtained in good isolated yields
(Table 1, entries 17À19), suggesting that no transient
carbocationic species were involved but rather biradical
species.17 Vinyl silanes are very useful intermediates for
further synthetic manipulations.18 The formation of car-
bocationic species seems to be excluded, but we still cannot
rule out the initial formation of OBB 2 which would
undergo a subsequent heterolytic ring opening into A
(orB). However, DFT calculations show that the heterolytic
cleavage of oxabicyclobutane into a biradical species is
much higher in energy (>8 kcal/mol) than its direct
transformation into an enone, which is not consistent with
the ratio of enones 3and 4obtained experimentally.14 On the
other hand, a very small energy difference (1À2 kcal/mol)
between biradical Aand Bwas found for 1a (Table 1, entry 1,
R1 = Me, R2 = Ph, R3 =R4 =H) and 1d (Table 1, entry 6,
R1 = R2 = R3 = Me, R4 = H), which is in a good
agreement with the 1:1 ratio of the two enones. Addition-
ally, biradical A of 1g is more stable by almost 12 kcal/mol
than biradical B (Table 1, entry 9, R1 = Ph, R2 = R3 =
Me, R4 = H).14
To increase structural complexity of these enones, a
fundamental strategy is to utilize substrate stereochemistry
to control the introduction of new stereogenic centers
through the Felkin-Anh or Cram chelation models.19
Particularly interesting is the formation of quaternary
stereogenic centers in acyclic systems20,21 and the addition
of cyclopropenyllithium species22 to R-substituted alde-
hydes and ketones, which led to the formation of diaster-
eoisomerically pure cyclopropenylcarbinols 1uÀz in
excellent yields (Scheme 3) as single diastereoisomers.23
The simple addition of mCPBA to these substituted
cyclopropenylcarbinols gave the corresponding enones
3uÀz as unique isomers in good yield. The stereochem-
istry was confirmed by X-ray analysis of 1z and 3u; the
Scheme 2. Possible Explanations for the Formation of Enones 3
and 4
Therefore, by using the conjugative resonance stabiliza-
tion of radicals (i.e., formation of benzyl radicals), the
oxidation reactionofcyclopropenylcarbinol should lead to
a complete and selective reaction.
To check this hypothesis, cyclopropenylcarbinol 1g7e
was prepared (R1 = Ph, R2 = R3 = Me, R4 = H) and
oxidized with mCPBA, and to our delight, a single isomer
3g was obtained in good yield (Table 1, entry 9). The same
isomer was formed when the Sharpless epoxidation con-
ditions were used (Table 1, entry 10), illustrating that,
in both cases, the exclusive formation of intermediate
A (R1 = Ph) leads quantitatively to enone 3. This trans-
formation has a broadscopeand is insensitivetothe nature
of substituents R2 and R3 (aryl, primary and secondary
alkyl groups, Table 1, entries 11À16). However, the exclusive
formation of 3 does not preclude the initially formed
oxabicyclobutane 2, which would undergo heterolytic
opening of the epoxide moiety either (1) to the biradical
intermediate A or (2) to a 1,3-zwitterionic intermediate
with a cationic center on the cyclopropane ring (not shown
in Scheme 2). By a subsequent ring opening of this
zwitterionic species intothe corresponding oxy-substituted
allyl cation, the carbonyl compound could be formed.
The intermediate zwitterion would also explain the observed
regioselectivity for this particular substitution pattern since
the stabilizing effect of an aryl substituent (R1 = Ph) on a
(16) Colvin, E. Silicon in Organic Synthesis; Butterworth: London,
1981.
(17) All of our attempts to get more insight from CIDNP experiments
failed.
(18) Denmark, S. E.; Ober, M. H. Aldrichimica Acta 2003, 36, 75.
(19) (a) Stanton, G. R.; Koz, G.; Walsh, P. J. J. Am. Chem. Soc. 2011,
131, 7969 and references cited therein. (b) Mengel, A.; Reiser, O. Chem.
Rev. 1999, 99, 1191.
(20) (a) Das, J. P.; Chechik, H.; Marek, I. Nat. Chem. 2009, 1, 128. (b)
Sklute, G.; Marek, I. J. Am. Chem. Soc. 2006, 128, 4642. (c) Marek, I.;
Sklute, G. Chem. Commun. 2007, 1683. (d) Kolodney, G.; Sklute, G.;
Perrone, S.; Knochel, P.; Marek, I. Angew. Chem., Int. Ed. 2007, 46,
9291.
(21) For a recent review, see: Das, J. P.; Marek, I. Chem. Commun.
2011, 47, 4593.
(14) See Supporting Information.
(15) Dmitrenko, O.; Bach, R. D. J. Phys. Chem. A 2004, 108, 6886.
(22) Al-Dulayymi, J. R. M.; Baird, S. Tetrahedron 1990, 46, 5703.
(23) Miege, F.; Meyer, C.; Cossy, J. Org. Lett. 2010, 12, 4144.
4078
Org. Lett., Vol. 13, No. 15, 2011

configurations of other reaction products were assigned
by analogy.
ofunreactedcyclopropenylcarbinol. Epoxyalcohols5 were
obtained as single diastereoisomers and enantiomers.25
Scheme 3. Oxidation of Functionalized Cyclopropenylcarbinol
Derivatives 1u,z
Scheme 4. SKR of Cyclopropenylcarbinol Leading to Enan-
tiomerically Enriched BaylisÀHillman Adducts
In conclusion, by a judicious choice of substituent R1 on
the cyclopropenyl core, both isomers 3 and 4 could be
selectively obtained, suggesting the formation of biradical
intermediates A and B. Although no intermediates could
be detected (the ring opening of such entities is much faster
than any possible trapping experiments), the selective
formation of such isomers depends on the relative stability
of carbon-centered radicals and is strongly supported by
high-level ab initio calculations. This mild and selective
oxidation reaction was then successfully used for the pre-
paration of complex conjugated carbonyl compounds pos-
sessing two stereogenic centers. Alternatively, the Sharpless
kinetic resolution (SKR) of monosubstituted cycloprope-
nylcarbinol derivatives (R1 = H) leads to enantiomerically
enriched R-hydroxy enal derivatives (BaylisÀHillman
adducts) with outstanding enantiomeric ratios.
Having access to enones of general structure 3 via the
biradical species A, we then wondered if enones 4 could be
selectively prepared by driving the reaction through the
formation of the biradical intermediate B. As tertiary
radicals are more stable than secondary, we hypothesized
that cyclopropenylcarbinol 1qÀt (R1 = H) would only
lead to intermediates B possessing a tertiary centered
radical. We were delighted to see that our prediction was
correct and only single isomers corresponding to the
BaylisÀHillman enals24 4 were formed (Table 1, entries
20À25) whatever the method (A or B) used. In these cases,
the biradical B for 1q (Table 1, entry 20, R1 = H, R2 = Ph,
R3 = H, R4 = Me) is more stable by 17 kcal/mol than
biradical A.14 The Sharpless kinetic resolution conditions
were then used to oxidize cyclopropenylcarbinols 1r, 1t,
and 1aa, and the corresponding enals were obtainedin very
good chemical yields (45, 40, and 41% yields, respectively)
and with outstanding enantioselectivities25 along the re-
maining cyclopropenylcarbinols 1, as described in Scheme 4.
Even more appealing was the kinetic resolution with an
excess of t-BuOOH (1.5 equiv, Scheme 4). In this case,
epoxidation of the resulting enal was faster that the oxidation
Acknowledgment. This research was supported by the
Israel Science Foundation administrated by the Israel
Academy of Sciences and Humanities (70/08), by the State
of Lower-Saxony and the Volkswagen Foundation, Hann-
over, Germany, and by the Fund for promotion of Research
atthe Technion. The authors thank Dr. ShmuelCohen and
Dr. Mark Botoshansky for X-ray analysis. I.M. is holder
of the Sir Michael and Lady Sobell Academic Chair.
Supporting Information Available. Experimental pro-
1
cedures with a description of H NMR, 13C NMR, and
(24) Basavaiah, D.; Reddy, B. S.; Badsara, S. S. Chem. Rev. 2010,
110, 5447.
crystallographic data (CIF) and all calculated energies.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(25) Enantiomeric excesses were determined on crude 19F NMR after
transformation into Mosher esters; see experimental part.
Org. Lett., Vol. 13, No. 15, 2011
4079