Biomimetic reactions in organic synthesis: Semi-pinacol rearrangements of some spirocyclic epoxyalcohols derived from Julia-Colonna asymmetric epoxidations

Article abstract of DOI:10.1139/v02-061

Epoxy tert-alcohols have been prepared from (E)-enones in a two-step approach consisting of Julia-Colonna asymmetric epoxidation followed by Grignard alkylation of the epoxyketone. On treatment with sub-stoichiometric amounts of Yb(OTf)₃ these trans-epoxyalcohols underwent efficient stereoselective semi-pinacol rearrangement to afford anti-α-phenyl-β-hydroxy-ketones (aldols). Under the same conditions, spirocyclic epoxyalcohols derived from 1-tetralone and 1-benzosuberone undergo either ring contraction (via semi-pinacol rearrangement) or fragmentation. A mechanistic rationale is presented to explain the formation of the various products.

Full text of DOI:10.1139/v02-061

546
Biomimetic reactions in organic synthesis:
Semi-pinacol rearrangements of some
spirocyclic epoxyalcohols derived from
Juliá–Colonna asymmetric epoxidations
Bernhard Hauer, Jamie F. Bickley, Julien Massue, Paula C. A. Pena,
Stanley M. Roberts, and John Skidmore
Abstract: Epoxy tert-alcohols have been prepared from (E)-enones in a two-step approach consisting of Juliá–Colonna
asymmetric epoxidation followed by Grignard alkylation of the epoxyketone. On treatment with sub-stoichiometric
amounts of Yb(OTf)3 these trans-epoxyalcohols underwent efficient stereoselective semi-pinacol rearrangement to af-
ford anti-α -phenyl-β-hydroxy-ketones (aldols). Under the same conditions, spirocyclic epoxyalcohols derived from 1-
tetralone and 1-benzosuberone undergo either ring contraction (via semi-pinacol rearrangement) or fragmentation. A
mechanistic rationale is presented to explain the formation of the various products.
Key words: Juliá–Colonna reaction, asymmetric epoxidation, epoxy tert-alcohols, semi-pinacol rearrangement.
Résumé : Faisant appel à l’approche en deux étapes de Juliá–Colonna impliquant une époxydation asymétrique suivie
d’une alkylation de Grignard de l’époxycétone, on a préparé des époxy tert-alcools à partir de (E)-énones. Le traite-
ment de ces trans-époxyalcools avec des quantités sous-stoechiométriques d’Yb(OTf)3 provoque un réarrangement se-
mi-pinacolique stéréosélectif et efficace qui conduit aux anti-α -phényl-β-hydroxycétones (aldols). Dans les mêmes
conditions, les époxyalcools spirocycliques obtenus à partir de la 1-tétralone et de la 1-benzosubérone donnent lieu soit
à une contraction de cycle (par le biais d’un réarrangement semi-pinacolique) ou à une fragmentation. On propose un
mécanisme qui permet de rationaliser la formation des divers produits.
Mots clés : réaction de Juliá–Colonna, époxydation asymétrique, époxy tert-alcools, réarrangement semi-pinacolique.
[Traduit par la Rédaction] Hauer et al. 550
Introduction
in a stereoselective manner. Various Lewis acids have been
used, in stoichiometric quantities, to facilitate this rearrange-
ment, including TiC14 (4), BF3·OEt2 (3, 5, 6), Ti(i-PrO)2C12
(5), Ti(i-PrO)3C1 (7), and SnC14 (8). Until recently, cata-
lytic variants of this rearrangement required a silyl-protected
epoxyalcohol, using either TMSI or TMSOTf as the Lewis
acid (9). Our recent report (10) and that of Tu et al. (11)
have shown, however, that rare-earth triflates and ZnBr2 are
capable of catalyzing semi-pinacol rearrangements of unpro-
tected epoxyalcohols.
Some of the epoxy tert-alcohol substrates used in semi-
pinacol rearrangements may be prepared in enantiomerically
enriched form by a five-step Sharpless AE–oxidation–
alkylation–oxidation–alkylation strategy (4, 5). An alterna-
tive approach to prepare such substrates is the use of a
polyleucine-catalyzed asymmetric epoxidation of an enone
Epoxides are among the most widely used functional
groups in modern synthetic chemistry. This reflects the exis-
tence of numerous reliable, highly enantioselective methods
for epoxidation, and the wealth of chemistry that epoxides
can undergo, allowing a wide range of compounds to be pre-
pared in an efficient and stereoselective manner (1).
Amongst the reactions of epoxides, rearrangements have
been widely used as key steps in the synthesis of biologi-
cally active targets and natural products. Several different
modes of epoxide rearrangement have been identified (2); in
substrates where the epoxide is flanked by an alcohol, or a
silyl-protected alcohol, treatment with a suitable Lewis acid
can lead to a semi-pinacol rearrangement (3), in which 1,2-
migration of one of the alcohol substituents opens the
epoxide, generating an α -substituted-β-hydroxyketone (aldol)
(Juliá–Colonna epoxidation) (12–14) followed by
a
Received 20 November 2001. Published on the NRC Research Press Web site at http://canjchem.nrc.ca on 30 April 2002.
Dedicated to Professor J. Bryan Jones on the occassion of his 65th birthday.
B. Hauer. BASF AG, GVF B9, D67056, Ludwigshafen, Germany.
J.F. Bickley, J. Massue, P.C.A. Pena, S.M. Roberts,1 and J. Skidmore. Department of Chemistry, Liverpool University,
Liverpool, U.K., L69 7ZD.
1Corresponding author (e-mail: smrsm@liv.ac.uk).
Can. J. Chem. 80: 546–550 (2002)
DOI: 10.1139/V02-061
© 2002 NRC Canada

Hauer et al.
547
Scheme 1. (a) Asymmetric epoxidation; (b) Grignard alkylation; (c) Yb(OTf)3 catalyzed rearrangement.
OH
O
O
O
3
H
2
OH
(S)
(S)
R
(S)
O
O
a
b
c
3
(S)
2
R
1
R
2
1
2
1
(E)
R
R
R
R
R
R
(R)
(R)
H
1
R
1
3
4
2
Scheme 2. (a) Asymmetric epoxidation; (b) Grignard alkylation; (c) Yb(OTf)3 catalyzed rearrangement.
2
O
O
HO
R
(S)
(S)
O
O
(E)
1
b
1
1
a
R
R
R
(R)
(R)
n
n
n
6
5
7
c
1
n = 0, 1, 2
R
O
HO
2
R
n
8
Scheme 4. Reagents and conditions: (i) R2MgBr, Et2O–THF,
Scheme 3. Reagents and conditions: (i) urea–H2O2, BEMP,
–78°C.
SiO2-poly-L-leucine, THF, room temp.
2
O
OH
O
R
O
(S)
O
(S)
(S)
O
O
(E)
1
(S)
(R)
i
1
1
R
i
R
1
(R)
R
(R)
R
n
n
n
n
7a - f
6a - d
1
6a
6b
6c
6d
5a
5b
5c
5d
n = 1, R = Et
1
n = 0, R = i-Pr
1
n = 2, R = i-Pr
1
n = 1, R = Ph
Results
E-3,4-Dihydro-2-propylidene-2H-naphthalen-1-one (5a)
was converted into the epoxide 6a in 75% yield and 85% ee
employing silica-supported poly-L-leucine as the catalyst
with urea–H2O2 (UHP) as the oxidant and sub-stoichio-
metric quantities of BEMP as the base (Scheme 3). Using
similar reaction conditions,indanone epoxide 6b was pre-
pared from enone 5b with good stereoselectivity (74% yield,
81% ee), whilst the epoxidation of alkylidene–benzosu-
berone 5c under these conditions proved to be very slow and
to generate racemic product 6c (62% yield). This result con-
trasts with our previously reported asymmetric epoxidation
of 5c (under different conditions), which generated the prod-
uct in 74% yield and 59% ee. The synthesis of epoxide 6d
(76% yield and 84% ee) was reported previously, employing
poly-L-leucine as the catalyst, UHP as the oxidant in
isopropyl acetate containing DBU (17). The alternative use
of sub-stoichiometric quantities of BEMP as the base and
silica-supported poly-L-leucine as the catalyst led to a quan-
titative yield of 6d having the same optical purity.
diastereoselective Grignard alkylation of the resultant
epoxyketone (15, 16). This method entails conversion of
achiral enones 1 into optically active epoxides 2, which are
transformed with high stereoselectivity into enantiomerically
enriched diastereomerically pure epoxyalcohols 3. We have
shown that such epoxyalcohols are efficiently rearranged to
afford aldols 4 in reactions catalyzed by rare-earth triflates
(Scheme 1) (10).
Recent work has extended the Juliá–Colonna epoxidation
to cyclic enones possessing an exocyclic olefin unit; thus
epoxidation of ketones 5 using urea–H2O2 and DBU in the
presence of poly-L-leucine affords spirocyclic epoxides 6 in
63–85% yield and 59–94% ee (Scheme 2) (17).
It was anticipated that alkylation of spirocyclic epoxy-
ketones 6 would afford epoxy alcohols 7 (step b) in which
the aromatic ring is the group with the highest migratory ap-
titude in a semi-pinacol rearrangement unless stereochemical
constraints become dominant. Thus, treatment of 1,2,3,4-
tetrahydronaphthalene epoxyalcohols (7, n = 1) with rare-
earth triflates was expected to result in ring contraction to
afford indans (8, n = 1) bearing a quaternary chiral center
(step c); such systems are not readily accessible, in
enantiomerically enriched form, using conventional method-
ology. We report herein the outcome of Lewis acid catalyzed
rearrangements of compounds of type 7.
Epoxyketones 6a–d were alkylated with methyl-
magnesium bromide to afford the epoxyalcohols 7a–d, re-
spectively; compounds 6a and 6d were also reacted with
phenylmagnesium bromide to afford the tertiary alcohols 7e
and 7f (Scheme 4 and Table 1).
All the ketones gave the appropriate product as a single
diastereoisomer. The operation of a Cram-chelate mode of
© 2002 NRC Canada

548
Can. J. Chem. Vol. 80, 2002
Fig. 1. X-ray crystal structure of compound 7d.
Table 1. Reaction of epoxyketones 6a–d with Grignard reagents
R2MgBr.
Epoxyalcohol
(yield %)
Epoxyketone
R1
R2
n
6a
6b
6c
6d
6a
6d
Et
Me
Me
Me
Me
Ph
1
0
2
1
1
1
7a (82)
7b (66)
7c (57)
7d (97)
7e (69)
7f (94)
i-Pr
i-Pr
Ph
Et
Ph
Ph
Scheme 5. Reagents and conditions: (i) Yb(OTf)3, CH2Cl2, room
temp., 1 h.
O
2
OH
OH
R
(S)
2
O
R
(S)
H
1
i
1
(S)
R
(S)
(R)
R
those of the ring-contracted products 8a and 8e. FAB mass
spectrometry displayed a molecular-ion peak corresponding
to compound 8f. This material was unstable (decomposing
readily but cleanly to ketone 9 and benzaldehyde), which
prevented more extensive characterization. With this addi-
tional evidence, however, it seems highly likely that the
well-precedented semi-pinacol rearrangement forms during
the first part of the reaction.
n
n
1
2
8a
8c
8e
7a
7c
7e
n = 1, R = Et, R = Me
1
2
n = 2, R = i-Pr, R = Me
1
2
n = 1, R = Et, R = Ph
addition (16) was confirmed by X-ray analysis of the
epoxyalcohol 7d (Fig. 1).2
Surprisingly, epoxyalcohol 7d did not produce ring-
contracted material, but instead, formed two major products,
1-methyl-2-tetralone (10) (40% yield) and benzaldehyde,
which were identified by H NMR and by comparison with
authentic materials (10).
To effect the desired semi-pinacol rearrangement, tertiary
alcohol 7a was treated with Yb (OTf)3 (20 mol %) in
dichloromethane for 1 h. The formation of the expected
methyl ketone 8a (Scheme 5) was supported by NMR spec-
troscopy (three-proton singlet at δ 2.03 inter alia) and by a
strong signal at 1693 cm–1 in the IR spectrum. Under the
same reaction conditions, indan-epoxide 7b gave a complex
mixture, whereas treatment of the benzosuberan-epoxide 7c
with Yb(OTf)3 gave the ring-contracted product 8c in 60%
yield.
1
Me
O
Similarly, treatment of 7e with Yb(OTf)3 (20 mol %) in
dichloromethane, followed by purification of the crude prod-
uct by chromatography over silica gel led to the isolation of
a rearranged product. The structure of the isolated product
8e (58% yield) was initially suggested by IR spectroscopy
(λmax = 1661 cm–1), corroborated by NMR spectroscopy, and
firmly established by X-ray crystallography (Fig. 2). As ex-
pected, the epoxide ring opening had occurred with inver-
sion of configuration.
10
Discussion
Indans 8a and 8e are clearly formed by the expected ring-
contraction process, whilst the fragmentation product 9 is as-
sumed to be generated from indan 8f. The fact that indan-
epoxide 7b fails to undergo ring contraction is presumably
due to a high-energy transition state caused by ring strain.
The key stereoelectronic requirement for the semi-pinacol
rearrangement is that the migrating group should be anti-
periplanar to the breaking C–O bond. Note that compound
7e undergoes ring contraction rather than phenyl migration.
Although no crystal structure data are available for 7e, in-
spection of the X-ray crystal structure of the analogous
methyl substituted epoxyalcohol 7d (Fig. 1) suggests an ori-
gin for the selectivity. Thus whilst the dihedral angle be-
tween the epoxide C–O and the methyl group C–C [O(2)-
Treatment of alcohol 7f with Yb(OTf)3 (20 mol %) gave a
mixture of products from which the indan-1-yl phenyl
ketone (9) was isolated in 63% yield (Scheme 6). Ketone 9
is most probably formed by retroaldol fragmentation of the
initially formed ring-contracted product 8f. To try to provide
proof for this hypothesis, the reaction was repeated but
quenched after a much shorter period of time. Purification of
the crude material led to the isolation of a major product —
1
the H NMR spectrum of which was closely analogous to
2 Supplementary material may be purchased from the Depository of Unpublished Data, Document Delivery, CISTI, National Research Coun-
cil Canada, Ottawa, ON K1A 0S2, Canada. For information on obtaining material electronically go to
http://www.nrc.ca/cisti/irm/unpub_e.shtml. Crystallographic information has also been deposited with the Cambridge Crystallographic Data
Centre (CCDC Nos. 181123 and 181124). Copies of the data can be obtained free of charge via www.ccdc.cam.ac.uk/conts/retrieving.html
(or from the Cambridge Crystallographic Data Centre, 12, Union Road, Cambridge CB2 1EZ, U.K.; fax: +44 1223 336033; or de-
posit@ccdc.cam.ac.uk).
© 2002 NRC Canada

Hauer et al.
549
Scheme 6. Reagents and conditions: (i) Yb(OTf)3, CH2Cl2, room temp.
O
O
OH
OH
Ph
O
(S)
Ph
Ph
(S)
H
(S)
(R)
i
Ph
(S)
Ph
9
7f
8f
+ PhCHO
Scheme 7.
Fig. 2. X-ray crystal structure of compound 8e.
HO
Yb(OTf)
HO
Ph
3
(S)
O
path a
(S)
(S)
(S)
(R)
Ph
O
Yb(OTf)
3
7d
11
path b
(TfO) Yb
3
H
OH
O
O
Ph
(S)
Ph
(S)
(R)
(S)
O
Yb(OTf)
3
C(2)-C(1)-C(18)] is 61.0°, the dihedral angle between the
epoxide C–O and the migrant C–C [O(2)-C(2)-C(1)-C(10)]
is 178.45°, i.e., anti-periplanar. The required alignment of
the migrating C–C with the C–O s* of the epoxide dictates
the outcome of these semi-pinacol rearrangements.
15
12
The mechanism by which 1-methyl-2-tetralone (10) is
formed from epoxyalcohol 7d is less clear; two possibilities
are outlined in Scheme 7. Coordination of the Yb(OTf)3 to
the epoxide moiety should assist ring opening to reveal
benzylic carbocation 11 (Scheme 7, path a). It is possible
that this cation can rearrange via oxetane 12 to generate ter-
tiary benzylic carbocation 13. This would readily undergo a
fragmentation reaction to generate benzaldehyde and the
enolate (14) of 1-methyl-2-tetralone. An alternative approach
to benzylic carbocation 13 is via a Lewis acid-catalyzed
Payne rearrangement generating epoxyalcohol 15 followed
by epoxide opening (Scheme 7, path b).
OH
(S)
(E)
O
Yb(OTf)
3
Ph
10
(R)
O
+ PhCHO
Yb(OTf)
3
13
14
Acknowledgments
Work is in progress to try to distinguish between these
two possibilities and the outcome of these experiments will
be reported in a full paper.
Financial support was gratefully received from the
Fundação para a Ciência e Tecnologia (FCT) and BASF (for
PCAP). JM received support from the University of Rennes.
We thank Dr. M. J. Porter (U.C., London) for insightful dis-
cussions.
Conclusions
Application of Juliá–Colonna asymmetric epoxidation fol-
lowed by Grignard alkylation has been used to prepare
epoxyalcohols in enantiomerically enriched form, both from
acyclic enones and from cycloalkanones bearing exocyclic
olefin units. The spirocyclic epoxyalcohols prepared from
the latter substrates undergo either ring contraction or frag-
mentation processes, depending on the substituents present
and the size of the ring. Yb(OTf)3 has been shown to be an
excellent catalyst for epoxyalcohol semi-pinacol rearrange-
ments.
References
1. E.N. Jacobsen, A. Pfaltz, and H. Yamamoto. In Comprehen-
sive asymmetric catalysis. Edited by Springer-Verlag, Berlin.
1999.
2. B. Rickborn. In Comprehensive organic synthesis. Vol. 3.
Edited by B.M. Trost and I. Fleming. Pergamon Press, Oxford.
1991. pp. 733–775.
3. C.J. Cheer and C.R. Johnson. J. Am. Chem. Soc. 90, 178
(1968).
© 2002 NRC Canada

550
4. K. Maruoka, M. Hasagavia, and H. Yamamoto. J. Am. Chem.
Soc. 108, 3827 (1986).
5. M. Shimazaki, H. Hara, H.K. Suzuki, and G. Tsuchihashi. Tet-
rahedron Lett. 28, 5891 (1987).
6. K. Suzuki, M. Miyazawa, M. Shimazaki, and G. Tsuchihashi.
Tetrahedron, 44, 4061 (1988).
7. S.W. Baldwin, P. Chen, N. Nikolic, and D.C. Weinselmer. Org.
Lett. 2, 1193 (2000).
8. C.M. Marson, A.J. Walker, J. Pickering, A.D. Hobson, R.
Wrigglesworth, and S.J. Edge. J. Org. Chem. 58, 5944 (1993).
9. K. Suzuki, M. Miyazawa, and G. Tsuchihashi. Tetrahedron
Lett. 28, 3515 (1987).
10. J.F. Bickley, B. Hauer, P.C.A. Pena, S.M. Roberts, and J.
Skidmore. J. Chem. Soc. Perkin Trans. 1, 1253 (2001).
11. Y.Q. Tu, C.A. Fan, S.K. Ren, and A.S.C. Chan. J. Chem. Soc.
Perkin Trans. 1, 3791 (2000).
Can. J. Chem. Vol. 80, 2002
12. S. Juliá, J. Guixer, J. Masana, J. Rocas, S. Colonna, R. Annuziata,
and H. Molinari. J. Chem. Soc. Perkin Trans. 1, 1317 (1982).
13. B.M. Adger, J.V. Barkley, S. Bergeron, M.W. Cappi, B.E.
Flowerdew, M.P. Jackson, R. McCague, T.C. Nugent, and S.M.
Roberts. J. Chem. Soc. Perkin Trans. 1, 3501 (1997).
14. M.J. Porter, S.M. Roberts, and J. Skidmore. Bioorg. Med.
Chem. 7, 2145 (1999).
15. A.T. Gillmore, S.M. Roberts, M.B. Hursthouse, and K.M. Ab-
dul. Tetrahedron Lett. 39, 3315 (1998).
16. J.F. Bickley, A.T. Gillmore, S.M. Roberts, J. Skidmore, and A.
Steiner. J. Chem. Soc. Perkin Trans. 1, 1109 (2001).
17. P.A. Bentley, J.F. Bickley, S.M. Roberts, and A. Steiner. Tetra-
hedron Lett. 42, 3741 (2001).
© 2002 NRC Canada