Catalytic enantioselective epoxidation of alkenes with a tropinone-derived chiral ketone

Article abstract of DOI:10.1039/a708695d

α-Fluoro-N-ethoxycarbonyltropinone is an efficient catalyst for the epoxidation of alkenes by Oxone with good enantioselectivity.

Full text of DOI:10.1039/a708695d

View Online / Journal Homepage / Table of Contents for this issue
Catalytic enantioselective epoxidation of alkenes with a tropinone-derived
chiral ketone
Alan Armstrong*† and Barry R. Hayter
Department of Chemistry, University of Nottingham, University Park, Nottingham, UK NG7 2RD
a-Fluoro-N-ethoxycarbonyltropinone is an efficient catalyst
for the epoxidation of alkenes by Oxone® with good
enantioselectivity.
from 1 by treatment of the derived trimethylsilyl enol ether with
12
a Selectfluor reagent. § Racemic 2 proved to be an excellent
catalyst for the epoxidation of (E)-stilbene (Table 1): using 10
mol% 2, reaction was complete in less than 2 h (entry 4), while
reasonable conversions were possible at the 1 mol% level over
longer time periods (entry 6).¶ There was no evidence for
The development of catalytic methods for the asymmetric
epoxidation of unfunctionalised alkenes continues to be an
1
1
important goal in organic chemistry. The chiral manganese
Baeyer–Villiger reaction by H NMR spectroscopy, and the
salen complexes developed by Jacobsen and Katsuki provide a
solution to this problem for certain alkene substitution patterns,
but these catalysts generally give poor enantioselectivity for
epoxidation of, for example, trans-disubstituted alkenes.1
catalyst could be recovered (ca. 70% yield) by column
chromatography.
In view of the extremely promising catalytic activity of 2, we
then attempted its synthesis in enantiomerically pure form.
Following the work of the Simpkins group on desymmetrisation
®
Recently, attention has focused on the catalysis of Oxone
2
3–6
7
13
epoxidation of alkenes by chiral ketones (via a dioxirane ‡)
of tropinones using chiral lithium amide bases, treatment of 1
8
14
and by chiral iminium salts (via an oxaziridinium species) as a
with the lithium amide base derived from 3 (1 equiv.) and
n
possible solution to this problem, and some impressive
advances have been recorded. Yang and co-workers described a
catalytic binaphthyl-based ketone which, with appropriate
substitution, can give high selectivity [up to 84% ee for the
Bu Li (2 equiv.) in the presence of Me
3
SiCl (5 equiv.) and LiCl
(1 equiv.) gave the crude silyl enol ether which was reacted
without purification with the Selectfluor reagent. The resulting
sample of 2 (36% yield, unoptimised), ca. 60% ee, was
3
epoxidation of (E)-stilbene]. Shi has reported a d-fructose-
recrystallised once from Et
2
O–light petroleum and then once
derived chiral ketone which gives extremely high enantiose-
from CH Cl –light petroleum to provide ketone 2 in > 98% ee
2
2
®
lectivities, but is destroyed under the Oxone epoxidation
according to chiral HPLC.∑ The relative stereochemistry of 2
was proven by X-ray crystallography;** the absolute ster-
eochemistry is assigned based on the precedent of Simpkins,
who converted the same intermediate silyl enol ether into
4
conditions, presumably by Baeyer–Villiger reaction. It must
therefore be used in relatively large quantities, although
modified pH conditions have improved the catalytic efficiency
of the process.4b The enantiomeric catalyst derived from
l-fructose is less readily available, however. Here we describe
some of our own results in the development of new ketone
catalysts which have led to the discovery of a new class of chiral
ketone which can be used catalytically and recycled if required,
as well as affording high enantioselectivities for alkene
epoxidation.
13
(2)-anatoxin-a. It should be noted that the antipodal form of
14
the chiral amine 3 is also readily available, and so it should be
possible to access the other enantiomer of 2 by this route. We
are also currently exploring resolution methods.
Enantiomerically enriched 2 was used as catalyst for the
®
Oxone epoxidation of a range of alkenes (Table 2) and the
results are extremely promising. Enantioselectivities of up to
83% have been obtained. Interestingly, all our results to date fit
a simple transition state model (Fig. 1), which assumes attack on
the sterically less hindered exo-oxygen of the dioxirane
Recognising the need for an efficient ketone catalyst to be
3,9
activated electronically towards attack, we initially examined
several classes of a-functionalised ketone. Of these, a-amido
ketones were attractive with regard to the easy introduction
of asymmetry, but were prone to decomposition by
3
intermediate and a spiro transition state (in accord with Yang
4
and Shi ). The hydrogen substituent on the alkene occupies the
9
Baeyer–Villiger reaction. We therefore turned our attention to
region of the fluorine substituent of the catalyst. In accord with
this model, the epoxidation of styrene, where there are two
possible hydrogens on the terminal carbon of the alkene that can
occupy the catalyst fluorine region, proceeds with lower
enantioselectivity (entry 5). Amongst the other substrates
examined so far, it is noteworthy that a,b-unsaturated esters can
be epoxidised under these conditions with moderate (but
b-amido ketones and in particular, in view of the need
eventually to prepare conformationally well-defined chiral
catalysts, tropinone derivatives. Commercially available
N-ethoxycarbonyltropinone 1 provided promising initial re-
CO2Et
N
Ph
Ph
X
Table 1 Oxone
®
epoxidation of (E)-stilbene catalysed by ketone 2
a
Ph
Ph
H
N
H
N
H
(±)-1
Conversion
(%)
(mol%)b
c
t/h
< 0.25
< 0.25
< 0.5
2
Entry
O
1
2
X = H
X = F
3
1
2
3
4
5
6
100
50
25
10
5
100
100
100
100
100
sults: using 1 (10 equiv.) and Oxone® (10 equiv.) under the
Yang MeCN–H
O conditions,10 (E)-stilbene was epoxidised
completely within 3 h. Importantly, there was no evidence for
decomposition of 1 under these conditions. Reasoning that
electron-withdrawing substituents should further increase the
activity of the catalyst,11 we prepared the a-fluoro derivative 2
@20
2
d
1
62
24
a
Alkene (0.1 mmol), Oxone® (1.0 mmol KHSO
5 3
), NaHCO (1.55 mmol),
MeCN (1.5 ml), aq. Na EDTA (1 ml of 0.4 mmol dm23 solution).
b
2
Relative
to alkene. By TLC analysis. d Measured by H NMR spectroscopy.
c
1
Chem. Commun., 1998
621

View Online
Table 2 Asymmetric epoxidation of alkenes catalysed by ketone (+)-2a
+)-2
(
Conversion
(%)
Yield
(%)
Ee (%)e
configuration
b
c
d
f
Entry
Alkene
(mol%)
t/h
< 3
< 4
< 4
< 6
< 2
24
g
1
2
3
4
5
6
(E)-Stilbene
(E)-a-Methylstilbene
Phenylstilbene
1-Phenylcyclohexene
Styrene
(E)-Methylcinnamate
10
10
10
10
10
25
100
100
100
100
100
88
100
100
97
33
33
76 R,R
73 R,R
83 R
69 R
29 R
h
i
64
64
a
Alkene (0.1 mmol), Oxone® (1.0 mmol KHSO
EDTA (1 ml of 0.4 mmol dm solution), (+)-2. b Relative
23
5
), NaHCO
3
(1.55 mmol), MeCN (1.5 ml), aq. Na
2
c
d
e
1
to alkene. Estimated by TLC. Isolated yield of epoxide product. Enantiomeric excesses were measured by H NMR spectroscopy in the presence of
Eu(hfc) as chiral shift reagent. Absolute configurations were determined by comparison to literature data (refs. 3 and 4). Determined by HPLC (Chiracel
3
OD). Measured by H NMR spectroscopy. Absolute configuration not assigned.
f
g
h
1
i
of this. Earlier 18O labelling experiments (A. Armstrong, P. A. Clarke and
A. Wood, Chem. Commun., 1996, 849) were performed in a two phase,
2 2 2
CH Cl –H O solvent system.
§
1-Chloromethyl-4-fluoro-1,4-diazoniabicyclo[2.2.2]octane bis(tetra-
fluoroborate).
¶
It is probable that the lower conversions at the 1 mol% level are due to
®
decomposition of the Oxone over the extended reaction times rather than
decomposition of the catalyst.
∑
performed using a Chiracel OD column with 100:1 hexane–Pr OH + 0.1%
The ketone had mp 57.5 °C, [a]2 +7.3 (c 1.16, CH
7
Cl
). Chiral HPLC was
D
2
2
i
3
21
TFA as eluent; detection at 224 nm; flow rate 1 cm min retention time
2
*
7 min (minor enantiomer), 29.3 min (major enantiomer).
* Details will appear in a full account of this work. We thank Dr A. J. Blake
and Dr Wan-Sheung Li of this Department for this structure determina-
tion.
1
2
T. Katsuki, Coord. Chem. Rev., 1995, 140, 189.
S. E. Denmark, D. C. Forbes, D. S. Hays, J. S. DePue and R. G. Wilde,
J. Org. Chem., 1995, 60, 1391.
3
D. Yang, Y.-C. Yip, M.-W. Tang, M.-K. Wong, J.-H. Zheng and
K.-K. Cheung, J. Am. Chem. Soc., 1996, 118, 491; D. Yang,
X.-C. Wang, M.-K. Wong, Y.-C. Yip and M.-W. Tang, J. Am. Chem.
Soc., 1996, 118, 11311.
Fig. 1 Model for the approach of a trans-disubstituted alkene to the
dioxirane derived from ketone (+)-2
4
5
(a) Y. Tu, Z.-X. Wang and Y. Shi, J. Am. Chem. Soc., 1996, 118, 9806;
(b) Z.-X. Wang, Y. Tu, M. Frohn and Y. Shi, J. Org. Chem., 1997, 62,
2
328.
promising) enantioselectivity (entry 6), albeit requiring longer
reaction times and higher catalyst loadings.
C. E. Song, Y. H. Kim, K. C. Lee, S.-G. Lee and B. W. Jin, Tetrahedron:
Asymmetry, 1997, 8, 2921.
In conclusion, we have found that a-fluoro-N-ethoxycar-
6 For earlier attempts to use chiral ketones as catalysts, see R. Curci,
M. Fiorentino and M. R. Serio, J. Chem. Soc., Chem. Commun., 1984,
155; R. Curci, L. D’Accolti, M. Fiorentino and A. Rosa, Tetrahedron
Lett., 1995, 36, 5831; D. S. Brown, B. A. Marples, P. Smith and
L. Walton, Tetrahedron, 1995, 51, 3587.
bonyltropinone is an efficient catalyst for the epoxidation of
®
alkenes by Oxone ; it can be used in low loadings and
recovered and recycled. Moreover, when prepared in enan-
tiomerically pure form, it affords high enantioselectivity for
alkene epoxidation. Attempts to prepare related bicy-
clo[3.2.1]octanone derivatives with alternative a-substitution,
in order to improve enantioselectivity further and to clarify the
factors responsible for asymmetric induction, are underway.
We thank the EPSRC, the DTI and a consortium of chemical
companies for funding this work through the LINK Asymmetric
Synthesis Second Core Programme. We also thank Mr T.
Lowdon (Hicksons) and Dr N. Johnson (Chiroscience) for
helpful discussions. We are grateful to Professor N. S. Simpkins
and Mr C. D. Jones of this Department for supplies of
N-ethoxycarbonyltropinone and of the chiral base 3, and for
helpful advice concerning the asymmetric deprotonation
chemistry.
7
For reviews of dioxirane chemistry, see R. W. Murray, Chem. Rev.,
1989, 89, 1187; W. Adam, R. Curci and J. O. Edwards, Acc. Chem. Res.,
1989, 22, 205; R. Curci, A. Dinoi and M. F. Rubino, Pure Appl. Chem.,
1995, 67, 811; R. Curci, M. Fiorentino, L. Troisi, J. O. Edwards and
R. H. Pater, J. Org. Chem., 1980, 45, 4758.
8 V. K. Aggarwal and M. F. Wang, Chem. Commun., 1996, 191;
A. Armstrong, G. Ahmed, I. Garnett and K. Goacolou, Synlett, 1997,
1075 and references cited therein; P. C. Bulman Page, G. A. Rassias,
D. Bethell and M. B. Schilling, in the press.
9
A. Armstrong and B. R. Hayter, unpublished results; A. Armstrong and
B. R. Hayter, Tetrahedron: Asymmetry, 1997, 8, 1677.
1
0 D. Yang, M.-K. Wong and Y.-C. Yip, J. Org. Chem., 1995, 60, 3887.
1
1 S. E. Denmark, Z. Wu and C. M. Crudden, ACS Meeting Abstract
3
74-ORGN, vol. 214, Las Vegas, September, 1997.
1
1
2 G. S. Lal, J. Org. Chem., 1993, 58, 2791.
3 N. J. Newcombe and N. S. Simpkins, J. Chem. Soc., Chem. Commun.,
1995, 831 and references cited therein.
Notes and References
1
4 K. Bambridge, M. J. Begley and N. S. Simpkins, Tetrahedron Lett.,
†
‡
E-mail: alan.armstrong@nottingham.ac.uk
A dioxirane is almost certainly the active species in the monophasic
1
994, 35, 3391.
MeCN–H
2
O solvent system: A. Armstrong, B. R. Hayter and P. A. Clarke,
1
8
unpublished results. See also ref. 5 for O labelling experiments in support
Received in Liverpool, UK, 2nd December 1997; 7/08695D
622
Chem. Commun., 1998