Enantioselective cyclopropanation of styrene catalysed by copper(I) complexes with chiral oxazolines

Article abstract of DOI:10.1016/S0040-4020(00)00144-7

A large set of copper complexes, prepared in situ from copper(I)- triflate and a variety enantiopure oxazoline ligands, was assessed as chiral catalysts in the enantioselective cyclopropanation of styrene by ethyl diazoacetate. Enantioselectivities up to 60% and up to 52%, respectively, for trans- and cis-2-phenylcyclopropanecarboxylate were observed. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4020(00)00144-7

TETRAHEDRON
Pergamon
Tetrahedron 56 (2000) 2889±2893
Enantioselective Cyclopropanation of Styrene Catalysed by
Copper(I) Complexes with Chiral Oxazolines
*
Giorgio Chelucci , Maria G. Sanna and Sera®no Gladiali
*
Dipartimento di Chimica, Universit aÁ di Sassari, via Vienna 2, I-07100 Sassari, Italy
Received 4 December 1999; accepted 28 January 2000
AbstractÐA large set of copper complexes, prepared in situ from copper(I)-tri¯ate and a variety enantiopure oxazoline ligands, was
assessed as chiral catalysts in the enantioselective cyclopropanation of styrene by ethyl diazoacetate. Enantioselectivities up to 60% and
up to 52%, respectively, for trans- and cis-2-phenylcyclopropanecarboxylate were observed. q 2000 Elsevier Science Ltd. All rights
reserved.
Introduction
improved stereocontrol of the reaction by removing com2-
petitive reaction paths which may reduce enantio®delity,
high e.e.'s can be obtained also with ligands devoid of this
topology. In the cyclopropanation of ole®ns by diazoesters,
During the last decade, enantiopure oxazoline ligands have
emerged as an effective class of chiral inducers in a wide
variety of transition metal catalysed asymmetric processes.
High catalytic activities and enantiomeric excesses (e.e.'s)
have been obtained in hydrosilylation, allylic alkylation,
H-transfer reduction and cyclopropanation, mainly using
for instance, the Ru(II)-complex with the C -symmetry
2
tridentate ligand 2,6-bis(oxazolinyl)pyridine (pybox) 1 is a
quite active and enantioselective catalyst,3 but more
recently it has been shown that the unsymmetrical bis(ox-
C -symmetric chiral ligands in conjunction with suitable
2
transition metal ions.
azolinyl)pyridine
(Scheme 1).
2
provides similar stereoselection
1
4
Although C -symmetric oxazolines usually allow for an
2
These results stimulated us to investigate the behaviour of
Scheme 1.
Keywords: copper catalysts; asymmetric cyclopropanation; oxazoline ligands.
*
Corresponding authors. Fax: 139-79-229559; e-mail: chelucci@ssmain.uniss.it; e-mail: gladiali@ssmain.uniss.it
0040±4020/00/$ - see front matter q 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(00)00144-7

2
890
G. Chelucci et al. / Tetrahedron 56 (2000) 2889±2893
Scheme 2.
C -symmetric oxazolinylpyridines 3 in the cyclopropana-
1
model for the evaluation of the ef®ciency of these oxazoline
ligands in the copper(I)-catalysed asymmetric cyclo-
propanation of ole®ns.
tion of styrene. These ligands are capable of excellen5t
asymmetric inductions in Rh-catalysed hydrosilylation
and in the Pd-catalysed allylic substitution reactions. We
6
have further pursued our study to evaluate the utility in this
reaction of the potentially tridentate and tetradentate oxazo-
line derivatives 4 and 5, respectively.
In this paper we report the results obtained in the application
of a set of 28 enantiopure oxazoline derivatives as ligands
in copper(I)-catalysed asymmetric cyclopropanation of
styrene.
The reaction was carried out at room temperature by slow
addition of ethyl diazoacetate to a solution of styrene in
methylene chloride containing the copper(I)-oxazoline
catalyst. This was prepared in situ by adding the appropriate
amount of ligand to copper(I) tri¯uoromethanesulfonate±
benzene complex [Cu(OTf)(C H ) ].
Results and Discussion
Synthesis of the ligands
All but two of the oxazoline ligands used throughout this
study were already known and have been prepared accord-
ing to the procedures reported in the literature. The oxa-
zolinylpyridine 3i and the oxazolinylphenanthridine 3o
which were not yet known, have been expressly prepared
for this study by heating under re¯ux a chlorobenzene solu-
tion of the 2-cyano-6-(1,1-dimethylethyl)pyridine (6) or of
the 2-cyanobenzo[h]quinoline (7) with (S)-tert-leucinol in7
the presence of a catalytic amount of zinc chloride
6
6 0.5
At the outset of the catalytic screening, our attention was
focussed on ligands 3 (Scheme 1) bearing substituents of
different steric demand or electronic property on the oxa-
zoline and/or pyridine rings. This follows from the con-
sideration that the introduction of a substituent on the
6-position of the pyridine ring of oxazolinylpyridines is
the cause of a chiral switch in the Rh-catalysed hydrosilyl-
9
ation and exerts a dramatic effect on the enantioselectivity
of the Pd-catalysed allylic alkylation.
(
Scheme 2).
6
Copper(I)-catalysed asymmetric cyclopropanation
The ®rst set of catalytic runs has been performed with
ligands 3a±o substituted with groups of increasing steric
demand on the pyridine and/or the oxazoline ring (Scheme 3).
8
The reaction of styrene with ethyl diazoacetate to give the
trans and cis cyclopropanes 8 and 9 was chosen as the
Scheme 3.

G. Chelucci et al. / Tetrahedron 56 (2000) 2889±2893
2891
Table 1. Enantioselective cyclopropanation of styrene by ethyl diazoacetate using ligands 3a±o
a
Yield (%) 819
b
trans/cis ratio 8:9
b
c
Entry
Ligand
% e.e.
Con®guration
8
9
8
9
1
2
3
4
5
6
7
8
9
3a
3b
3c
3d
3e
3f
3g
3h
3i
3j
3k
3l
3m
3n
3o
67
74
76
66
88
89
78
77
89
71
75
77
66
73
69
68:32
68:32
68:32
68:32
70:30
71:29
64:36
61:39
62:38
77:23
69:31
69:31
67:33
75:25
67:33
2
4
5
4
9
7
1R, 2R
1S, 2S
1R, 2R
1S, 2S
1S, 2S
1R, 2R
1S, 2S
1S, 2S
1S, 2S
1R, 2R
1S, 2S
1S, 2S
1S, 2S
1R, 2R
1R, 2R
1S, 2R
1R, 2S
1S, 2R
1R, 2S
1R, 2S
1S, 2R
1R, 2S
1R, 2S
1R, 2S
1S, 2R
1R, 2S
1R, 2S
1S, 2R
1R, 2S
1S, 2R
34
43
29
29
27
60
12
39
45
20
51
21
39
44
42
30
50
51
23
39
41
6
10
11
12
13
14
15
15
8
a
b
c
Isolated yield, based on ethyl diazoacetate, for the mixture of trans and cis cyclopropanes.
Determined by GC analysis (see Experimental).
Assignment according to Ref. 13.
From the results reported in Table 1, the following con-
siderations can be made on the mutual cross-effects of the
substitution pattern.
the substituent on the pyridine ring increases from methyl
to t-butyl, as in the case of 3d±h (entries 4±8) and 3e±i
(entries 5±9). Substitution of the hydrogen by a benzo-
fused ring, as in the case of 3a±j (entries 1±10), 3b±k
(entries 2±11), 3j±m (entries 10±13), 3k±n (entries 11±
14) and 3l±o (entries 12±15), has contrasting effects.
(
i) The diastereoselectivity of the reaction is negligibly
affected by the structure of the ligand.
ii) The enantioselectivity of the reaction is positively
(
affected by the presence of substituents of increasing
steric demand on the stereogenic carbon of the oxazoline
ring as in the case of 3a±b (entries 1±2), 3d±e (entries
These results indicate that steric encumbering in the
proximity of the pyridine nitrogen donor aids a more
ef®cient transmission of the chiral information from the
oxazoline pendant of the ligand. We may speculate that
the presence of a substituent on the 6-position can push
the metal closer to the oxazoline nitrogen in the transition
state or that it can prevent the approach of the substrate from
the pyridine side.
3
3
±5), 3h±i (entries 8±9), 3j±k±l (entries 10±12) and
m±n (entries 13±14). The sole exception to this trend
was observed with 3o which was less stereoselective than
n (entries 15±14).
iii) The introduction of an achiral substituent in the
-position of the pyridine ring has a pronounced effect
3
(
6
on the enantioselectivity of the reaction. Substitution of
the hydrogen by a methyl (3a±c; entries 1±3 and 3b±d;
entries 2±4) or by a p-tolyl group (3a±f; entries 1±6 and
3b±g; entries 2±7) leads to a sharp increase of the e.e.
The stereoselectivity improves further when the size of
The next set of catalytic experiments was devoted to clear
out the presence, if any, of an `electronic control' on the
outcome of the reaction. This follows from recent observa-
tions that even substituents remote from the active site of the
chiral ligand can in¯uence the catalytic activity and the
Table 2. Enantioselective cyclopropanation of styrene by ethyl diazoacetate using ligands 3p±u
a
Yield (%) 819
b
trans/cis ratio 8:9
b
c
Entry
Ligand
% e.e.
Con®guration
8
9
8
9
1
2
3
4
5
6
3p
3q
3r
3s
3t
78
74
88
75
75
86
66:34
73:21
65:35
64:36
63:37
61:39
0
4
10
26
50
8
0
8
3
19
45
11
±,±
±,±
1R, 2R
1R, 2R
1R, 2R
1S, 2S
1S, 2S
1R, 2S
1R, 2S
1R, 2S
1R, 2S
1R, 2S
3u
a
Isolated yield, based on ethyl diazoacetate, for the mixture of trans and cis cyclopropanes.
Determined by GC analysis (see Experimental).
Assignment according to Ref. 13.
b
c

2
892
G. Chelucci et al. / Tetrahedron 56 (2000) 2889±2893
Table 3. Enantioselective cyclopropanation of styrene by ethyl diazoacetate using ligands 4 and 5
a
Yield (%) 819
b
trans/cis ratio 8:9
b
c
Entry
Ligand
% e.e.
Con®guration
8
9
8
9
1
2
3
4
4a
4b
4c
4d
67
88
78
77
62:38
60:40
58:42
58:42
9
0
2
0
1R, 2R
1R, 2R
1R, 2R
1R, 2R
±,±
2
6
2
1S, 2R
±, ±
1R, 2S
41
5
6
7
5a
5b
5c
78
76
78
62:38
61:39
55:45
41
7
17
52
8
40
1S, 2S
1R, 2R
1S, 2S
1R, 2S
1R, 2S
1R, 2S
a
Isolated yield, based on ethyl diazoacetate, for the mixture of trans and cis cyclopropanes.
Determined by GC analysis (see Experimental).
Assignment according to Ref. 13.
b
c
enantioselectivity of the process through their electronic
effect.
(41%) for the cis-diastereoisomer was recorded in the
presence of the ligand 4d possessing two stereocentres on
the oxazoline ring. This last result seems to support the view
that the oxazoline nitrogen donor is not involved in the
coordination at the metal centre and acts as a mere chiral
target.
6
a,10
To this purpose, a variety of oxazoline ligands
substituted with electron-withdrawing (3p±r) or electron-
releasing (3q±s) groups in 4-position of the pyridine ring
were assessed in the cyclopropanation of styrene. While for
these derivatives we can con®dently assume a pure elec-
tronic effect for the substituent, steric and electronic proper-
ties can hardly be separated in ligands 3t±u where the
carbon adjacent to the pyridine nitrogen is substituted
In the series of C -symmetric pyridyloxazolines, the
2
epimeric ligands 5a±b were exploited in order to explore
the chiral matching±mismatching of the stereogenic
elements appended onto the pyridine and oxazoline rings.
The best enantioselection, both for the cis- and the trans-
isomer, was obtained with the (R,S,S,R)-compound 5a.
(
Table 2).
The data reported in Table 2 show that a 4-chloro sub-
stituent has a negligible effect on the e.e. (3p±r; entries
1
±3), whereas a methoxy substituent in the same position
leads to a sharp increase of the stereoselectivity (3q±s;
entries 2±4) and even to a chiral switch in comparison
with the unsubstituted oxazolines 3a±b (entries 1±2,
Table 1). A slight positive effect of the electron-releasing
substituent is observed even when a methyl substituent is
present in position-6 (compare entries 5 of this table and
Table 1). Finally, ligand 3u (entry 6) gives an e.e. lower than
the one recorded with the roughly related ligand 3d (entry 4,
Table 1). This provides further con®rmation to the negative
in¯uence of electron-withdrawing groups.
The use of the epimeric (S,S,S,S)-derivative 5b, which has
opposite con®guration at the oxazoline stereocentres, gave
products of much lower e.e. (entries 5±6) and brought about
a chiral switch in the trans-cyclopropane. This indicates that
the stereoselection process is mainly dictated by the stero-
genic elements of the dioxolane moiety. This assumption is
corroborated by the result obtained with the (S,S)-ligand 5c
which has no stereocentre in the oxazoline rings. In this
reaction the e.e.'s were slightly lower than the ones obtained
with 5a, but the sense of the chiral induction for both cis and
trans isomers was the same.
Notably, the trend observed with our pyridyloxazolines is
opposite to the case of 4-substituted pybox 1 (Scheme 1)
where electron-withdrawing substituents provide higher
The results obtained with ligands 4 and 5 lend support to the
view that these ligands bind to the metal in a bidentate rather
than in a tridentate fashion (Table 3). Coordination seems to
occur most probably at the pyridine nitrogen donors. In this
case the oxazoline substituent should act merely as a chiral
pendant without being directly involved in the coordination.
1
0b
e.e.'s than electron-donating groups.
1
The oxazolinylphenanthrolines 4 and the C -symmetric
1
2
1
2
pyridyloxazolines 5 were the last set of chiral modi®ers
we have tested in the cyclopropanation of styrene.
Compounds 4 and 5 are both capable of bidentate and
terdentate binding to the metal centre (tetradentate coordi-
nation of 5 to Cu(I) is possible as well, but should inactivate
the catalyst). Moreover, for both ligands two different
modes of bidentate coordination are conceivable: the
In conclusion, we have assessed in the Cu(I)-catalysed
enantioselective cyclopropanation of styrene a large variety
of chelating oxazoline ligands which were investigated with
respect to their steric and electronic features. The results
obtained indicate that the ef®ciency of these ligands is
moderate, the best e.e. being 60% given by the bis-t-butyl
substituted derivative 3i. This value is much higher than the
®
rst involving both the pyridine nitrogen donors and
the second involving only one pyridine and the oxazo-
line nitrogen.
1
4
one obtained with C -symmetric bis-oxazolines (3±8%),
2
but it is far from the best e.e.'s obtained with C -symmetric
2
1
bipyridines (92±98%). Thus, the C -symmetry hybrid
5
Ligands 4 were poor chiral inducers and gave negligible
e.e.'s, much lower than the ones observed with the
structurally related ligands 3m±n (compare with entries
1
pyridine±oxazoline ligands seem to retain half of the posi-
tive and half of the negative properties of their C -symmetry
counterparts.
2
13±14 of Table 1), in all cases but one. A moderate e.e.

G. Chelucci et al. / Tetrahedron 56 (2000) 2889±2893
2893
Experimental
added to a suspension of [Cu(OTf)(C H ) ] (8 mg,
6 6 0.5
3
2 mmol) in CH Cl (2.5 ml). After 30 min, the mixture
2 2
General
was ®ltered through packed adsorbent cotton under argon,
and to the ®ltrate was added styrene (1.59 ml, 13.87 mmol).
Then a solution of ethyl diazoacetate (285 mg, 2.5 mmol) in
CH Cl (2.5 ml) was added dropwise over a period of 1 h.
Gas chromatographic analyses were performed on a HP
900 chromatograph using He (60 kPa) as the carrier gas.
5
2
2
The trans/cis ratio and enantiomeric excesses of ethyl trans-
and cis-2-phenyl-cyclopropane-1-carboxylates 8 and 9 were
determined by GC analysis on a 25 m£0.25 mm capillary
column coated with diethyl-t-butylsilyl b-cyclodextrin
operated at 608C for 5 min, then programmed at 38C/min
to 1608C. The con®gurations of the prevailing enantiomers
The mixture was stirred for 12-14 h and then concentrated in
vacuo. The residue was puri®ed by column chromatography
on silica gel (hexane/ethyl acetateˆ15:1) to afford a mixture
of ethyl trans- and cis-2-phenyl-cyclopropane-1-carboxy-
lates as a colourless oil. This mixture was analysed by
GC. The diastereomeric and enantiomeric excess was deter-
1
3
were attributed according to the literature.
mined by capillary GC with a chiral column [T : 33.2
r
(
and 31.8 (1S,2R) min for cis 9].
1S,2S) and 33.5 (1R,2R) min for trans 8; T : 31.4 (1R,2S)
r
Copper(I) tri¯uoromethanesulfonate benzene complex
Cu(OTf)(C H ) ] and ethyl diazoacetate were purchased
from Aldrich. The ligands 3a±e, 3f±h, 3j±n, 3p±t,
[
6
6 0.5
6
a
6b
6b
6c
6
c
11
12
3u, 4a±d, 5a±c, 2-cyano-6-(1,1-dimethylethyl)pyri-
dine (6)
6
b
6b
and 2-cyanobenzo[h]quinoline (7)
prepared according to reported procedures.
were
Acknowledgements
This work was ®nancially supported by the Ministero
dell'Universit aÁ e della Ricerca Scienti®ca e Tecnologica,
Roma.
General procedure for the preparation of oxazolinyl-
methylpyridines 3
In a 25 ml two-necked ¯ask, zinc chloride (14 mg,
0.10 mmol) was melted under high vacuum and cooled
under argon. After cooling to room temperature, chloro-
benzene (12 ml) was added, followed by the nitrile
References
(
2 mmol) and the amino alcohol (3.0 mmol). The resulting
1
. For a recent review: Ghosh, A. K.; Mathivan, P.; Cappiello, J.
Tetrahedron: Asymmetry 1998, 9, 1.
. Nishiyama, H.; Itoh, Y.; Matsumoto, H.; Park, S.-B.; Itoh, K.
J. Am. Chem. Soc. 1991, 113, 2223.
mixture was heated under re¯ux for the proper time (vide
infra) and then the solvent was evaporated under reduced
pressure. The residue was dissolved in CH Cl (6 ml) and
2
2
2
the resulting solution was washed with water (3£4 ml). The
3. Whitesell, J. K. Chem. Rev. 1989, 89, 1581.
4. Nishiyama, H.; Soeda, N.; Tomonari, N.; Motoyama, Y.
aqueous solution was extracted with CH Cl (6 ml), the
2
2
combined organic phases were dried over anhydrous
Na SO and the solvent evaporated. The residue was puri-
®
acetate/petroleum etherˆ2:1).
Tetrahedron: Asymmetry 1998, 9, 2865.
2
4
5. Brunner, H.; Obermann, U. Chem. Ber. 1989, 122, 499.
6. (a) Chelucci, G.; Medici, S.; Saba, A. Tetrahedron: Asymmetry
1997, 8, 3183. (b) Chelucci, G.; Medici, S.; Saba, A. Tetrahedron:
ed by chromatography on a silica gel column (eluent: ethyl
Asymmetry 1999, 10, 543. (c) Chelucci, G.; Deriu, S.; Saba, A.;
Valenti, R. Tetrahedron: Asymmetry 1999, 10, 1457. (d) Chelucci,
G.; Deriu, S.; Pinna, G. A.; Saba, A.; Valenti, R. Tetrahedron:
Asymmetry 1999, 10, 3803.
(
(
0
S)-2-[4,5-Dihydro-4-(1,1-dimethylethyl)oxazol-2-yl]-6-
1,1-dimethylethyl)pyridine (3i). Reaction time: 48 h;
2
5
1
.42 g (46%); oil; [a] 2124.4 (c 0.6, CHCl ); H NMR
D 3
(
CDCl ) d: 7.94 (dd, 1H, Jˆ7.5, 0.9 Hz), 7.66 (t, 1H,
3
7
1
8
. Bolm, C.; Veickhardt, K.; Zehnder, M.; Ranff, T. Chem. Ber.
Jˆ7.5 Hz), 7.45 (dd, 1H, Jˆ7.5, 0.9 Hz), 4.47 (dd, 1H,
Jˆ10.2, 8.4 Hz), 4.31 (t, 1H, Jˆ8.4 Hz), 4.09 (dd, 1H,
Jˆ10.2, 8.4 Hz), 1.40 (s, 9H), 0.97 (s, 9H). Anal. Calcd
for C H N O: C, 73.81; H, 9.29; N, 10.76;. Found: C,
991, 124, 1173.
. For recent reviews on cyclopropanation: Doyle, M. P.;
Protonova, M. N. Tetrahedron 1998, 54, 7919; Singh, V. K.;
DattaGupta, A.; Sekar, G. Synthesis 1997, 137.
1
6
24
2
7
3.91; H, 9.15; N, 10.84;
9. Brunner, H.; Brandl, P. Tetrahedron: Asymmetry 1991, 2, 919.
10. (a) Clyne, D. S.; Mermet-Bouvier, Y. C.; Nomura, N.;
(S)-2-[4,5-Dihydro-4-(1,1-dimethylethyl)oxazol-2-yl]-
benzo[h]quinoline (3o). Reaction time: 24 h; 0.494 g
RajanBabu, T. V. J. Org. Chem. 1999, 64, 7601 and references
therein. (b) Park, S.-B; Murata, K.; Matsumoto, H.; Nishiyama, H.
Tetrahedron: Asymmetry 1995, 6, 2487.
2
5
1
(
(
8
81%); mp 1158C; [a] 285.1 (c 0.5, CHCl ); H NMR
D 3
CDCl ) d: 9.39 (d, 1H, Jˆ7.2 Hz), 8.38 (d, 1H, Jˆ8.1 Hz),
3
1
1. Gladiali, S.; Pinna, L.; Delogu, G.; Graf, E.; Brunner, H.
.23 (d, 1H, Jˆ8.1 Hz), 7.92±7.68 (m, 5H), 4.59 (dd, 1H,
Jˆ10.5, 9.0 Hz), 4.44 (t, 1H, Jˆ9.0 Hz), 4.22 (dd, 1H,
Jˆ10.5, 9.0 Hz), 1.03 (s, 9H). Anal. Calcd for C H N O:
Tetrahedron: Asymmetry 1990, 1, 937.
12. Chelucci, G. Tetrahedron: Asymmetry 1997, 8, 2667.
13. Fritschi, H.; Leutenegger, U.; Pfaltz, A. Helv. Chim. Acta
2
0
20
2
C, 78.92; H, 6.62; N, 9.2. Found: C, 78.64; H, 6.45; N, 9.35.
1988, 71, 1553.
14. Evans, D. A.; Woerpel, K. A.; Hinman, M. M.; Faul, M. M.
Asymmetric cyclopropanation of styrene: typical
procedure
J. Am. Chem. Soc. 1991, 113, 726.
1
5. Ito, K.; Katsuki, T. Tetrahedron Lett. 1993, 34, 2661.
A solution of the ligand (34 mmol) in CH Cl (2.5 ml) was
2
2