Vanadium-catalyzed asymmetric epoxidation of allylic alcohols mediated by (+)-norcamphor-derived hydroperoxide

Article abstract of DOI:10.1002/ejoc.200400736

A protocol for the asymmetric epoxidation of allylic alcohols has been established that employs VO(acac)₂ as catalyst, the commercial achiral hydroxamic acid, N-hydroxy-N-phenyl-benzamide, and optically pure (+)-norcamphor-derived hydroperoxide

Full text of DOI:10.1002/ejoc.200400736

FULL PAPER
Vanadium-Catalyzed Asymmetric Epoxidation of Allylic Alcohols Mediated by
(
+)-Norcamphor-Derived Hydroperoxide
[a]
[a]
[a]
Alessandra Lattanzi,* Sandro Piccirillo, and Arrigo Scettri
Keywords: Optically pure hydroperoxides / Asymmetric synthesis / Epoxidation / Vanadium / Kinetic resolution / Dia-
stereoselectivity
A protocol for the asymmetric epoxidation of allylic alcohols
has been established that employs VO(acac) as catalyst, the
commercial achiral hydroxamic acid, N-hydroxy-N-phenyl-
benzamide, and optically pure (+)-norcamphor-derived hy-
droperoxide as oxygen atom donor and chiral source. Varia-
tion of the reaction parameters has a significant effect on the
level of asymmetric induction. Under optimized conditions,
the epoxy alcohols have been isolated in good yields with up
to 67% ee and complete control of the diastereoselectivity.
2
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2005)
Introduction
achieved when using enantiopure, secondary alkyl hydro-
peroxides. In general, the most efficient oxygen atom do-
Asymmetric epoxidation of allylic alcohols catalyzed by
metal complexes in the presence of enantiopure ligands
using alkyl hydroperoxide as oxygen donors is the most fun-
damental process for obtaining epoxy alcohols with a high
level of enantioselectivity[ and impressive results have been
nors are tertiary hydroperoxides, as confirmed by the data
[13–15]
obtained from Sharpless-type oxidations.
Until now,
enantiopure alkyl hydroperoxides have rarely been em-
[16–19]
ployed as stereoselective reagents,
hence this type of
1]
investigation provides a new and complementary insight
into the field of asymmetric oxidations mediated by alkyl
hydroperoxides.
[
2,3]
obtained by using the Sharpless–Katsuki protocol
Ti(OiPr) /l-DET/TBHP].
[
4
A less developed methodology involving the use of
Recently we have become interested in the synthesis of
VO(acac) /optically active hydroxamic acids with TBHP as
2
optically pure, tertiary alkyl hydroperoxides 1 and 2 derived
oxidant was previously reported by Sharpless and co-
[20]
[21]
from (R)-camphor
and (S)-norcamphor
(Figure 2),
[
4,5]
workers
to afford epoxy alcohols in up to 80% ee. Modi-
respectively, in which (unlike TADOOH) the hydroperoxyl
group is directly bound to the stereogenic centre.
fied versions of this protocol have recently been proposed in
which structurally different enantiopure hydroxamic acids
[
6,7]
[8]
derived from binaphthyl,
α-amino acid, [2.2]paracyclo-
[
9]
[10]
phane, (+)-ketopinic acid
and (R)-α-phenethylamine
[
11]
compounds were employed. Only one example involving
the use of an achiral hydroxamic acid and the optically pure
hydroperoxide TADOOH (Figure 1) has so far been re-
[
12]
ported.
Figure 2. (R)-Camphor- and (S)-norcamphor-derived hydroperox-
ides.
These oxidants have been employed as oxygen donors in
[
20]
the Ti(OiPr) -catalyzed epoxidation of allylic alcohols
and in asymmetric sulfoxidation reactions,
4
[
21,22]
achieving in
some examples the best levels of enantioselectivity reported
when using optically pure hydroperoxides.
Figure 1. TADOOH hydroperoxide.
[
23]
Tertiary, sterically demanding TADOOH afforded the
A recent NMR analysis of the enantiopure [2.2]para-
[
9]
epoxy alcohols in up to 72% ee, much better than the levels cyclophane-based hydroxamic acid protocol supported
[
4,5]
the original mechanistic proposal
for the vanadium-cata-
[
a] Dipartimento di Chimica, Università di Salerno,
Via S. Allende, 84081, Baronissi, Italy
Fax: +39-089-965296
[24]
lyzed epoxidation reaction. Sharpless and co-workers
suggested that control of vanadium/ligand complexation is
important in the epoxidation reaction as different species
E-mail: lattanzi@unisa.it
Eur. J. Org. Chem. 2005, 1669–1674
DOI: 10.1002/ejoc.200400736
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1669

A. Lattanzi, S. Piccirillo, A. Scettri
FULL PAPER
can equilibrate (Figure 3). With an enantiopure hydroper- Results and Discussion
oxide and an achiral ligand, enantioselective epoxidation
A preliminary study of the catalytic asymmetric epoxid-
ation reaction was carried out under various conditions
reactions can occur via complexes A and B, unlike the clas-
sical methodology, which involves the use of an enantiopure
ligand and an achiral hydroperoxide, in which the asymmet-
ric epoxidation takes place only via complex B. Up to now,
no investigations into the competitive enantioselective epox-
idation by species A have been reported when using an op-
tically pure hydroperoxide, so it would be of interest to
check the efficiency and the sense of asymmetric induction
provided by this pathway.
(Scheme 1, Table 1). It has been demonstrated that the
VO(OiPr) complex is the most efficient catalyst for the
3
epoxidation reaction and that the nature of the solvent can
[
4–11]
affect the level of enantioselectivity.
Scheme 1. Vanadium-catalyzed epoxidation of allylic alcohols.
Hydroperoxide 1 was found to be less reactive and en-
antioselective than oxidant 2 in the epoxidation of geraniol
under usual reaction conditions [VO(OiPr) in toluene] at
3
0
°C (entries 1 and 2). Taking into account the reaction
pathway proposed for the vanadium-catalyzed epoxidation
[
4,5]
of allylic alcohols
and as recently supported by NMR
[
23]
studies, the lower reactivity of hydroperoxide 1 could be
tentatively ascribed to strong steric interactions in a transi-
tion state involving temporary coordination of the encum-
Figure 3. Vanadium/ligand complexation equilibria.
Herein, we report our results of the catalytic asymmetric bered oxygen atom donor and the allylic alcohol. In any
epoxidation of allylic alcohols using VO(acac) , the achiral case, the possibility of a low reaction rate for the ligand
2
and commercially available N-hydroxy-N-phenylbenzamide exchange, necessary for the epoxidation reaction to occur
and hydroperoxide 2 as chiral oxidant.
under catalytic conditions, cannot be excluded.
Table 1. Vanadium-catalyzed asymmetric epoxidation of allylic alcohols by 1 and 2.
1
[
a] Isolated yields after flash chromatography. [b] Determined by HPLC analysis using a chiral column or by H NMR shift experiments
3
on the acetylated epoxy alcohol using Eu(hfc) . The absolute configurations given in parentheses were determined by comparison with
the optical rotations reported in the literature. [c] The reaction was carried out with 10 mol-% catalyst/15 mol-% hydroxamic ligand. [d]
The reaction was carried out with 5 mol-% catalyst/7.5 mol-% hydroxamic ligand. [e] The reaction was carried out in the absence of
hydroxamic acid and with 10 mol-% catalyst. [f] The reaction was carried out by using 10 mol-% catalyst/30 mol-% hydroxamic ligand.
[g] The reaction was carried out in the absence of hydroxamic acid and with 2 mol-% catalyst.
1670 © 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.eurjoc.org
Eur. J. Org. Chem. 2005, 1669–1674

Vanadium-Catalyzed Asymmetric Epoxidation of Allylic Alcohols
The epoxidation of 3b (entry 3) using 2 under the same Table 2. Asymmetric epoxidation of 3b by VO(acac)
FULL PAPER
2
/2 and dif-
[a]
2 2
ferent hydroxamic acids at –20 °C in CH Cl .
conditions took place to provide the (R,R)-epoxy alcohol
with a better ee. Halving the catalyst loading and lowering
the temperature to –20 °C reduced the reactivity but im-
proved the enantioselectivity of the reaction (entry 4). Al-
though toluene has been reported to be the best solvent,
when the reaction was performed in CH Cl (entry 5), the
2
2
conversion to the epoxide increased while the enantio-
selectivity was maintained. When the reaction was carried
out with 10 mol-% of VO(acac) in toluene at 0 °C, the
2
same conversion and enantioselectivity were achieved but
with a longer reaction time (entry 6). By using 5 mol-% of
catalyst at –20 °C (entry 7) the reactivity was reduced but
the ee of epoxide improved (compare with entry 5). Hydro-
peroxide 1 was employed under these conditions, but with-
out success; in fact after two days no trace of epoxide was
detected, confirming the poor reactivity of this oxidant.
When 10 mol-% of VO(acac) was used at –20 °C (entry 9)
2
the epoxide was obtained in good yield and with an im-
proved enantioselectivity (61% ee), which increased to 67%
ee when the reaction was performed at –40 °C (entry 10).
In the absence of the ligand (entry 11) an almost compar-
able reaction time (compare with entry 9) was required to
furnish the (R,R)-epoxy alcohol in a higher yield but with
a lower enantioselectivity. Predictably, complex A was more
[
a] The reactions were carried out by using a mixture of 3b/
active, although to only a small extent, than complex B VO(acac) /hydroxamic acid/2 in a molar ratio of 1:0.10:0.15:1.2. [b]
2
Isolated yields after flash chromatography. [c] Determined by
(Figure 3). In order to be sure that the competing epoxid-
HPLC analysis using a Chiralcel OD column and detection at
54 nm; eluent: hexane/iPrOH, 97:3. The absolute configurations
ation by species A would be negligible, the equilibria had
to be shifted towards complexes B, C and D. Hence, up to
2
given in parentheses were determined by comparison with the op-
tical rotations reported in the literature. [d] In this case 10 mol-%
3
equivs. of the ligand (entry 12), with respect to the metal,
3
were used to ensure that only complex B was involved in of VO(OiPr) was used as the catalyst.
[
24]
the asymmetric epoxidation pathway. As expected,
the
reactivity was lowered (compare with entry 9) but the ee
was almost comparable to that reported in entry 9; thus
Steric hindrance of the substituent on the nitrogen atom
with a metal/ligand ratio of 1:1.5, we can be assured that significantly reduced the reactivity and the enantio-
species A is not involved in the epoxidation process and selectivity of the reaction (entries 1 and 2). The N-methyl-
that only the most enantioselective complex B is involved substituted ligand slowed the conversion but a moderate
in the asymmetric epoxidation.
level of asymmetric induction was observed in the product
Note that the reaction in entry 11 represents the first (entry 3). The N-unsubstituted hydroxamic acid furnished a
asymmetric example of the well-known system reported by very slow reacting complex with low asymmetric induction
Sharpless and co-workers for the regio- and diastereoselec- (entry 4). The reaction of N-methylnaphthohydroxamic
tive epoxidation of allylic alcohols[
4,25]
[VO(acac) /TBHP/ acid failed completely (entry 5). The benzyl derivative also
2
allylic alcohol]. When the reaction was carried out with afforded an unreactive complex when using VO(acac) as
2
only 2 mol-% of VO(OiPr) and in the absence of a ligand the catalyst, while a modest conversion to the epoxide with
3
(
entry 13), (R,R)-epoxy alcohol was isolated after a short 55% ee was achieved with VO(OiPr) (entry 6). In all cases
3
reaction time in a satisfactory yield and with the same 38% the (R,R)-epoxide was preferentially obtained. Simple alkyl-
ee (compare with entry 11). In the last case, in order to or aryl-substituted hydroxamic acids had, as previously re-
suppress the involvement of complex A in the reaction, the ported, pronounced but not easily accountable effects on
[
26]
use of a hydroxamic acid/vanadium catalyst ratio
the reactivity and enantioselectivity of the epoxidation reac-
Ͼ1.5:1.0 equiv. would seem to be better, although it has the tion, which were inferior to the levels achieved with the
disadvantage of lowering the reaction rate.[
24]
commercial ligand 5a (Table 1, entry 9).
As suggested by the data in Table 1, the optimized reac-
On the basis of these results, N-hydroxy-N-phenylbenz-
tion conditions involve the employment of easy-to-handle amide (5a) was the ligand used to assess the scope of the
and inexpensive VO(acac) in CH Cl . Hence, we screened epoxidation reaction (Table 3, Scheme 2).
2
2
2
various achiral hydroxamic acids, synthesized according to
Geraniol and its (Z)-isomer nerol were epoxidized to give
in the epoxidation of model the epoxy alcohols in good yields and with moderate
alcohol 3b under the best conditions found previously enantioselectivities, showing that the asymmetric induction
Table 2). does not depend on the geometry of the double bond (en-
known procedures,[
27,28]
(
Eur. J. Org. Chem. 2005, 1669–1674
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1671

A. Lattanzi, S. Piccirillo, A. Scettri
FULL PAPER
alcohols as substrates, operating under Ͻ50% conversion of
the alcohol and by using VO(Oi-Pr) , since these substrates
3
are less reactive than primary allylic alcohols.
Compound 3h was epoxidized with complete stereoselec-
tivity to give the syn epoxide but with poor enantio-
/5a and hydro- selectivity (entry 8). Indeed, kinetic resolution took place
with low efficiency as demonstrated by the value of the
Scheme 2. Asymmetric epoxidation with VO(acac)
peroxide 2.
2
[
29]
stereoselectivity factor (S = 2.1).
Alcohol 3i was con-
tries 2 and 3). The smaller pentenol was epoxidized more verted exclusively to the erythro epoxy alcohol although
quickly (entry 4), while the epoxidation of α-phenylcinn- with comparable asymmetric induction and efficiency of
[
30]
amyl alcohol furnished better results with VO(OiPr) in tol- kinetic resolution (entry 9). The level of diastereoselectiv-
3
uene (entry 5). Cyclic alcohol 3f was also epoxidized in ity was better than that observed when using other common
good yield and with a comparable ee (entry 6). β-Phenyl- systems such as Ti(OiPr) /L-DET/TBHP, Ti(OiPr) /TBHP
4
4
[
5,31]
cinnamyl alcohol, even after a prolonged reaction time, was and VO(acac) /TBHP,
which preferentially furnished
2
[
32–35]
found to be completely unreactive (entry 7).
the erythro
isomer of 4i. Surprisingly, experiments per-
Finally, the investigation into the vanadium-catalyzed formed in the absence of ligand afforded, in a shorter reac-
asymmetric epoxidation turned onto secondary allylic tion time, exclusively the erythro epoxy alcohol (entry 10)
Table 3. Asymmetric epoxidation of allylic alcohols by VO(acac)
2
/5a and hydroperoxide 2.[a]
[
a] The reactions were carried out by using a mixture of 3/VO(acac)
2
/5a/2 in a molar ratio of 1:0.10:0.15:1.2. [b] Isolated yields after flash
1
chromatography. [c] Determined by HPLC analysis using chiral columns or by performing H NMR shift experiments on the acetylated
1
epoxy alcohol using Eu(hfc)
determined by comparison with the optical rotations reported in the literature. [d] Determined by H NMR analysis (400 MHz) of the
crude reaction mixture. [e] The reaction was carried out in toluene using VO(OiPr) /5a in a molar ratio of 0.05:0.075. [f] The reaction
was carried out in toluene using VO(OiPr) /5a/2/3h in a molar ratio of 0.30:0.45:0.7:1. [g] Yield of recovered alcohol. [h] Enantiomeric
excess of recovered alcohol. [i] The reaction was carried out in toluene using VO(OiPr) /5a/2/3i in a molar ratio of 0.15:0.23:0.7:1. [l] The
reaction was carried out in the absence of hydroxamic acid and with 15 mol-% of VO(OiPr) in toluene.
3
or by H NMR analysis of the MTPA-derived ester. The absolute configurations given in parentheses were
1
3
3
3
3
1672
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 1669–1674

Vanadium-Catalyzed Asymmetric Epoxidation of Allylic Alcohols
FULL PAPER
with only a marginal effect on the kinetic resolution. The The solvent was evaporated under reduced pressure and the crude
product was purified by flash chromatography (petroleum ether/
ethyl acetate, 98:2) to give epoxy alcohols 4 and alcohol 6.
presence of the ligand seems to be of minor importance and
does not improve the diastereoselectivity, as observed for
the Ti(OiPr) /L-DET/TBHP system compared with
4
[
32]
Ti(OiPr) /TBHP.
This result clearly indicates that the
4
structure of the hydroperoxide can be crucial for controlling
the diastereoselectivity of the metal-catalyzed epoxidation
reactions and improvements can be expected by steric
modifications to the oxidant when employing the same me-
tal catalyst in the absence of any ligand.
Acknowledgments
We thank the Università degli Studi di Salerno and the MIUR for
financial support.
[
[
1] R. A. Johnson, K. B. Sharpless, Catalytic Asymmetric Synthe-
sis (Ed.: I. Ojima), Wiley-VCH, New York, 2000, p. 231.
Conclusions
2] T. Katsuki, K. B. Sharpless, J. Am. Chem. Soc. 1980, 102,
In summary, we have established a simple protocol for
5974–5976.
the asymmetric epoxidation of allylic alcohols using [3] Y. Gao, R. M. Hanson, J. M. Klunder, S. Y. Ko, H. Masamune,
K. B. Sharpless, J. Am. Chem. Soc. 1987, 109, 5765–5780.
[4] R. C. Michaelson, R. E. Palermo, K. B. Sharpless, J. Am.
Chem. Soc. 1977, 99, 1990–1992.
VO(acac) , commercial N-hydroxy-N-phenylbenzamide and
a (+)-norcamphor-derived hydroperoxide as oxidant.
The epoxy alcohols have been obtained with moderate
levels of asymmetric induction. Although the kinetic resolu-
2
[
36]
[
5] K. B. Sharpless, T. R. Verhoeven, Aldrichimica Acta 1979, 12,
63–74.
tion of the secondary allylic alcohols has a low efficiency, [6] N. Murase, Y. Hoshino, M. Oishi, H. Yamamoto, J. Org. Chem.
1
999, 64, 338–339.
the diastereoselectivity observed is excellent even in the ab-
sence of the ligand.
[
7] Y. Hoshino, N. Murase, M. Oishi, H. Yamamoto, Bull. Chem.
Soc. Jpn. 2000, 73, 1653–1658.
[
8] Y. Hoshino, H. Yamamoto, J. Am. Chem. Soc. 2000, 122,
1
0452–10453.
[9] C. Bolm, T. Kühn, Synlett 2000, 6, 899–901.
10] H.-L. Wu, B.-J. Uang, Tetrahedron: Asymmetry 2002, 13, 2625–
628.
Experimental Section
[
General Remarks: All reactions requiring anhydrous conditions
2
were conducted in flame-dried apparatus under argon. CH
2
Cl
2
and
[11] B. Traber, Y.-G. Jung, T. K. Park, J.-I. Hong, Bull. Korean
Chem. Soc. 2001, 22, 547–548.
toluene were distilled from calcium hydride under argon. Petroleum
ether showed a boiling range of 40–60 °C. Standard techniques
were used for handling air-sensitive reagents. All commercially
available reagents were purchased from Aldrich or Fluka. Reac-
tions were monitored by thin-layer chromatography (TLC) on
Merck silica gel plates (0.25 mm) and visualized by UV light or
[12] W. Adam, A. K. Beck, A. Pichota, C. R. Saha-Möller, D. Seeb-
ach, N. Vogl, R. Zhang, Tetrahedron: Asymmetry 2003, 14,
1355–1361.
[
[
[
13] M. G. Finn, K. B. Sharpless, Asymmetric Synthesis Academic
Press, New York, 1985, p. 247.
14] Y. Gao, R. M. Hanson, J. M. Klunder, S. Y. Ko, H. Masamune,
K. B. Sharpless, J. Am. Chem. Soc. 1987, 109, 5765–5780.
15] P. Pitchen, E. Duñach, M. N. Deshmukh, H. B. Kagan, J. Am.
Chem. Soc. 1984, 106, 8188–8193.
[16] H.-J. Hamann, E. Höft, D. Mostowicz, A. Mishnev, Z. Ur-
ba n´ czyk-Lipkowska, M. Chmielewski, Tetrahedron 1997, 53,
185–192.
[17] D. Mostowicz, M. Jurczak, H.-J. Hamann, E. Höft, M. Chmie-
lewski, Eur. J. Org. Chem. 1998, 11, 2617–2621.
18] W. Adam, M. N. Korb, K. J. Roschmann, C. R. Saha-Möller,
J. Org. Chem. 1998, 63, 3423–3428.
19] W. Adam, P. L. Alsters, R. Neumann, C. R. Saha-Möller, D.
Seebach, A. K. Beck, R. Zhang, J. Org. Chem. 2003, 68, 8222–
8231.
[20] A. Lattanzi, P. Iannece, A. Vicinanza, A. Scettri, Chem. Com-
mun. 2003, 1440–1441.
21] A. Lattanzi, P. Iannece, A. Scettri, Tetrahedron: Asymmetry
2004, 15, 1779–1785.
2 4
by a 10% H SO /ethanol spray test. Flash chromatography was
performed on Merck silica gel (60, particle size: 0.040–0.063 mm).
Optical rotations were performed on a Jasco Dip-1000 using a Na
lamp.
Spectroscopic characterizations of hydroperoxide 2 and alcohol 6
[22]
have previously been reported. All the epoxy alcohols are known
[
2,3]
compounds.
The hydroxamic acids in Table 2 are known com-
[
37]
[38]
[27]
[39]
[28]
[40]
pounds (5b, 5c, 5d, 5e, 5f and 5g ) and were synthe-
[
sized according to procedures reported in the literature.[
27,28]
The
enantiomeric excesses of the epoxy alcohols 4a and 4c were deter-
[
1
mined by H NMR spectral analysis of the corresponding acety-
[
2,3]
lated alcohols using Eu(hfc)
4
3
;
the enantiomeric excesses of 4d,
1
f, 4h and 4i were determined by H NMR spectral analysis of the
[
2,3]
corresponding Mosher esters;
the enantiomeric excesses of 4b
[
and 4e were determined by HPLC analysis on Chiralcel OD and
Chiralpak AD chiral columns.[
41]
[
22] A. Lattanzi, P. Iannece, A. Scettri, Tetrahedron: Asymmetry
2004, 15, 413–418.
Epoxidation of Allylic Alcohols 3. General Procedure: Ligand 5a
(
(
0.03 mmol, 6.4 mg) was added to a solution of VO(acac)
0.02 mmol, 5.3 mg) in dry CH Cl (1 mL) under argon at room
2
[23] K. P. Bryliakov, E. P. Talsi, T. Kühn, C. Bolm, New J. Chem.
2003, 27, 609–614.
2
2
[
[
[
24] D. J. Berrisford, C. Bolm, K. B. Sharpless, Angew. Chem. Int.
Ed. Engl. 1995, 34, 1059–1070.
temperature. The mixture was stirred for 20 min, then allylic
alcohol 3 (0.2 mmol) was added and the solution was stirred for a
further 40 min at room temperature. The reaction mixture was then
cooled to –20 °C and 2 (0.24 mmol, 47 mg) dissolved in dry CH
(
of the reaction, as verified by TLC, the crude reaction mixture was
filtered through a small pad of silica gel, eluting with a mixture of
petroleum ether/diethyl ether (2:1, 100 mL), to remove the catalyst.
25] K. B. Sharpless, R. C. Michaelson, J. Am. Chem. Soc. 1973, 95,
6
136–6137.
2
Cl
2
51
26] From V NMR analysis of a benzene solution of VO(OiPr)
3
1.5 mL) was added to the flask by means of a cannula. At the end
[8]
and an α-amino acid based hydroxamic acid, at a V/ligand
ratio of 1:1.5, a B/A ratio of 95:5 was determined (see Fig-
ure 3).
[27] R. M. Coates, S. J. Firsan, J. Org. Chem. 1986, 51, 5198–5209.
Eur. J. Org. Chem. 2005, 1669–1674
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1673

A. Lattanzi, S. Piccirillo, A. Scettri
FULL PAPER
[
28] R. V. Hoffman, N. K. Nayyar, J. Org. Chem. 1994, 59, 3530–
539.
29] H. B. Kagan, J. C. Fiaud, Top. Stereochem. 1988, 18, 249–330.
[36] At the end of the epoxidation reactions, alcohol 6 was isolated
by flash chromatography in 80–90% yield and recycled for the
one-step synthesis of 2.
3
[
[
[
30] S = 1.6 was calculated from the equation reported in ref.[
29]
.
[37] D. E. Robinson, M. D. Holladay, Org. Lett. 2000, 2, 2777–
31] Vanadium catalysts have been shown to be the most erythro-
2779.
selective for the epoxidation of allylic alcohols.[
5]
[38] J. A. Warshaw, D. E. Gallis, B. J. Acken, O. J. Gonzalez, D. R.
Crist, J. Org. Chem. 1989, 54, 1736–1743.
[39] M. Couturier, J. L. Tucker, C. Proulx, G. Boucher, P. Dubé,
B. M. Andresen, A. Ghosh, J. Org. Chem. 2002, 67, 4833–4838.
[40] R. V. Hoffman, N. K. Nayyar, W. Chen, J. Org. Chem. 1992,
57, 5700–5707.
[41] W. Adam, P. L. Alsters, R. Neumann, C. R. Saha-Möller, D.
Seebach, R. Zhang, Org. Lett. 2003, 5, 725–728.
Received: October 15, 2004
[32] The diastereoisomeric (erythro/threo) ratios obtained for 4i
by using Ti(OiPr)
acac)
tively.
4
/L-DET/TBHP, Ti(OiPr)
3/3T] BHP systems were 98:2, 88:12 and 91:9, respec-
4
/TBHP and VO-
(
2
[
[
33] V. S. Martin, S. S. Woodard, T. Katsuki, Y. Yamada, M. Ikeda,
K. B. Sharpless, J. Am. Chem. Soc. 1981, 103, 6237–6240.
34] W. Adam, B. Nestler, Tetrahedron Lett. 1993, 34, 611–614.
35] W. Adam, T. Wirth, Acc. Chem. Res. 1999, 32, 703–710.
[
[
1674
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 1669–1674