Modular ligands in asymmetric synthesis. Copper-mediated cyclopropanation

Article abstract of DOI:10.1016/j.tetlet.2003.12.061

The chiral imidazoline/copper catalyst system efficiently mediates asymmetric intermolecular cyclopropanations. Complexes derived from (R,R)- or (S,S)-1,1-diphenylethylenediamine, cyclic ketones, and Cu(I) or Cu(II) triflates were compared. The reaction b

Full text of DOI:10.1016/j.tetlet.2003.12.061

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 1781–1783
Modular ligands in asymmetric synthesis. Copper-mediated
cyclopropanation
Dora Tepfenhart, Lionel Moisan, Peter I. Dalko* and Janine Cossy*
Laboratoire de Chimie Organique associꢀe au CNRS, ESPCI, 10 rue Vauquelin, 75231 Paris Cedex 05, France
Received 10 July 2003; revised 2 December 2003; accepted 11 December 2003
Abstract—The chiral imidazoline/copper catalyst system efficiently mediates asymmetric intermolecular cyclopropanations. Com-
plexes derived from (R,R)- or (S,S)-1,1-diphenylethylenediamine, cyclic ketones, and Cu(I) or Cu(II) triflates were compared. The
reaction between ())-menthyl diazoacetate and 1,1-diphenylethylene affords cyclopropane carboxylates in up to 80% yield and with
up to 78% de.
ꢀ 2004 Elsevier Ltd. All rights reserved.
The concept of a modular system, which allows quick
access to a great diversity of ligands from simple com-
ponents, continues to attract considerable interest.
Recently, we developed an efficient system, which cata-
lyzes asymmetric Diels–Alder reactions between Dani-
shefskyꢀs diene and a variety of dienophiles.1 The
catalyst preparation is based on the condensation of 1,2-
diamines of type 1 with cycloalkanones 2, which affords
imidazolidines 3 in equilibrium with the open form 4
(Scheme 1). The corresponding bis-imine 5 and the
starting diamine 1 are also present in the reaction mix-
ture.2 We anticipated that metals such as Cu(I) or Cu(II)
could shift this equilibrium toward the metallacyclic
form 6 through the formation of a bidentate complex
(Scheme 1).
A remarkable advantage of this system is the great
number of ligand structures which can be generated
without multi-step syntheses. Although there is no
precedent for the use of this type of ligand in enantio-
catalytic reactions, literature analogies suggest that these
complexes may be active in a number of reactions.3 In
continuation of earlier work done on imidazolidine-
based chiral catalyst systems, asymmetric cyclopropa-
nation was examined (Scheme 2).4 The addition of ())-
menthyl diazoacetate 8 to 1,1-diphenylethylene 7 was
(CH2)n
(CH2)n
O
(CH2)n
H
N
Ph
Ph
2
Ph
Ph
Ph
Ph
Ph
Ph
N
N
2
N
NH2
NH2
(CH2)n
N
H
NH2
4
3
5
1
(CH2)n
2a: n = 1
Cu(I) or Cu(II)
(CH2)n
2b: n = 2
2c: n = 3
Ph
Ph
N
N
Cu(I) or Cu(II)
H2
6
Scheme 1.
Keywords: Catalysis; Catalysts; Cyclopropanation; Diazo compounds; Stereoselection.
* Corresponding authors. Tel.: +33-140-794412; fax: +33-140-794660; e-mail addresses: peter.dalko@espci.fr; janine.cossy@espci.fr
0040-4039/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2003.12.061

1782
D. Tepfenhart et al. / Tetrahedron Letters 45 (2004) 1781–1783
O
O
O
6 (5%)
H
H
Ph
Ph
R*O2C
+
+
+
O
O
O
CH2Cl2, rt
4Å MS
CO2R*
N2
Ph
Ph
Ph
Ph
7
9b
10
8
9a
1) NaOH (aq), then H+
2) CH2N2
O
O
H
H
OMe
11a
OMe
Ph
Ph
Ph
Ph
11b
Scheme 2.
selected as the model reaction.5 Both Cu(I) and Cu(II)
complexes were examined as catalysts.
when the cyclopentanone-derived catalyst 6b was used
(entry 3 vs entries 1 and 7). Although the Diels–Alder
and Mukaiyama-aldol reactions worked best with cat-
alyst 6c in a 1:1 ratio of ligand–copper salt,1 the asym-
metric cyclopropanation reaction gave better yields and
selectivity in the presence of catalyst 6b in a 2:1 ratio of
ligand–copper salt (entries 2–4). The selectivity of the
reaction was not further ameliorated by increasing the
ligand–copper salt ratio above 2. In fact, the high
ligand–copper ratio decreased the yield of the cyclo-
propanation reaction (entry 4).
Complexes derived from Cu(I) were tested first. The
effect of the ligand structure, particularly the influence
of the ketone component and the ligand/metal stoichio-
metry, on the diastereoselectivity was examined.
Accordingly, ligands 3a–c, derived from (R,R)-diphen-
ylethylenediamine (1), cyclobutanone (2a), cyclopenta-
none (2b), or cyclohexanone (2c), and (CuOTf)2ÆC6H6,
leading to 6a–c were tested. A solution of ())-menthyl
diazoacetate in dichloromethane was added by syringe
pump to a mixture of 1,1-diphenylethylene and catalyst
(5%) in dichloromethane (Scheme 2).6 The reaction
afforded the desired cyclopropane carboxylate 9 as a
mixture of inseparable diastereoisomers (Table 1),
The match/mismatch effect of the chiral ester with the
asymmetric catalyst was also studied (entry 5).10 When
the ligand derived from (1S,2S)-diphenylethylene-
diamine and cyclopentanone was used in the presence of
())-menthyl diazoacetate, the diastereomer 9b was
obtained as the major isomer although with a slight
erosion in the diastereoselectivity (de ¼ 50% vs 62%;
Table 1, entries 3 and 5). This fact demonstrates that the
stereochemistry of the cyclopropanation of 7 using chi-
ral diazoacetate 8 is not substrate but ligand controlled.
1
whose ratio was established either by GC–MS or by H
NMR. It is worth noting that variable amounts (15–
22%) of dimenthyl fumarate 10 were also obtained from
dimerization of the diazoester. These results are sum-
marized in Table 1. The absolute configuration of the
newly formed stereocenter was determined after sapon-
ification of the cyclopropylcarboxylate esters 9, followed
by esterification and comparison with the optical rota-
tion of the (R)-carboxycyclopropane derivative 11a.7
Independently, the absolute configuration of the major
isomer 9a was ascertained by GC comparison with a
sample of known configuration prepared according to
Nishiyamaꢀs procedure.8;9
The preparation of the catalyst also affects the selectivity
of the reaction. In particular, the time of complexation
of the catalyst (ꢁagingꢀ) markedly affected the reaction.
Better selectivity was observed when the catalyst was
used after 3 days of complexation compared to the one
used after the usual 1 day preparation (entries 3 and 6).
Catalysts derived from (1R,2R)-diphenylethylenedi-
amine mediated the formation of the (R)-cyclopropyl-
carboxylate ester 9a as the major product independently
of the nature of the ketone used to prepare the catalyst.
As depicted in Table 1, the best results were obtained
Complexes derived from Cu(II) were tested likewise
(Table 2). Catalysts were prepared from (1S,2S)-di-
phenylethylenediamine, cyclopentanone, and Cu(OTf)2,
and reactions were run according to the same procedure
used for the Cu(I) complexes.11 Although both Cu(I)
Table 1
Entry
111
2
Ligand:Cu(I) (cat.)
Time of complexation (days)
9a+9b Yield (%)
9a:9b (Dr)
1
2
3
4
5
6
7
(R,R)-1
(R,R)-1
(R,R)-1
(R,R)-1
(S,S)-1
(R,R)-1
(R,R)-1
2a
2b
2b
2b
2b
2b
2c
2:1 (6a)
1:1 (6b)
2:1 (6b)
6:1 (6b)
2:1 (6b)
2:1 (6b)
2:1 (6c)
1
1
1
1
1
3
1
31
64
84
39
67
68
50
(77:23)
(79:21)
(81:19)
(81:19)
(25:75)
(89:11)
(75:25)

D. Tepfenhart et al. / Tetrahedron Letters 45 (2004) 1781–1783
1783
Table 2
References and notes
a
Entry
2
Ligand:Cu(II)12 Tc (%)
(cat.)
9a+9b 9a:9b
Yield (%) (Dr)
1. Dalko, P. I.; Moisan, L.; Cossy, J. Angew. Chem. Int. Ed.
2002, 41, 625.
2. Chapuis, C.; Gauvreau, A.; Klaebe, A.; Lattes, A.; Perie,
J. J. Bull. Soc. Chim. Fr. 1973, 977.
3. For related catalyst systems see: (a) Crabtree, R. H. Chem.
Commun. 1999, 1611; (b) Reetz, M. T. Angew. Chem., Int.
Ed. 2001, 40, 284.
1
2
3
4
2b
2b
2a
2c
1:1 (6b)
2:1 (6b)
2:1 (6a)
2:1 (6c)
50
50
45
90
18
25
14
45
(79:21)
(85:15)
(77:23)
(75:25)
a Tc ¼ conversion.
4. For recent works on catalytic cyclopropanation reactions
see: (a) Doyle, M. P. In Catalytic Asymmetric Synthesis;
Ojima, I., Ed.; Wiley-VCH: New York, 2000; p 191; (b)
Lydon, K. M.; McKervey, M. A. In Comprehensive
Asymmetric Catalysis; Jacobsen, E. N., Pfaltz, A.,
Yamamoto, H., Eds.; Springer: Berlin, 1999; Vol. II, p
540; (c) Pfaltz, A. In Comprehensive Asymmetric Catalysis;
Jacobsen, E. N., Pfaltz, A., Yamamoto, H., Eds.; Spring-
er: Berlin, 1999; Vol. II, p 513; (d) Charette, A. B.; Lebel,
H. In Comprehensive Asymmetric Catalysis; Jacobsen, E.
N., Pfaltz, A., Yamamoto, H., Eds.; Springer: Berlin,
1999; Vol. II, p 581; (e) Catalytic Asymmetric Carbene
Transfer Reactions. Davies, H. M. L. Ed.; Tetrahedron:
Asymmetry 2003, 14, 763; (f) Lebel, H.; Marcoux, J.-F.;
Molinaro, C.; Charette, A. Chem. Rev. 2003, 103, 977.
5. Chiral ligands were made by mixing the carbonyl com-
pound 2a–c with the (R,R)- or (S,S)-1,2-diphenylethylene-
diamine. Catalysts were prepared by mixing equimolar
amounts of Cu(OTf)2 or (CuOTf)2ÆC6H6, respectively,
with chiral ligands 3a–c in dichloromethane, and allowing
the mixture to stand for at least 1 day. The complexes were
blue-green to purple.
Table 3
a
Entry
Ligand 3a ee
(%)
Tc (%)
9a+9b Yield
(%)
9a:9b (Dr)
1
2
3
4
50.0
71.0
85.0
99.9
99
95
99
99
81
65
64
84
(59:41)
(66:34)
(75:25)
(81:19)
a Tc ¼ conversion.
and Cu(II) derived salts showed similar trends in stereo-
selectivity, Cu(II) complexes were less reactive and
afforded lower conversions.
The relationship between the optical purity of the catalyst
and the diastereomeric ratio of the products was probed
(Table 3). In this study, complexes derived from Cu(II)
triflate and ligand 3a, which came from (1R,2R)- or
(1S,2S)-diphenylethylenediamine and cyclopentanone,
were mixed and used in the desired ratio. As in the pre-
vious experiments, the stoichiometry of the complex was
set to a 2:1 ratio of ligand–Cu(II) salt and the catalyst was
allowed to stand for 1 day in order to ensure complexa-
tion. The analysis of the data shows a clear negative
nonlinear effect (NLE). These data may be explained
either by the formation of aggregates or by the partici-
pation of two or more ligands in the active complex.13
6. The selectivity of the reaction depends marginally on the
temperature in the range of 0–40 ꢁC. The transformation
was, however, highly sensitive to lower temperatures. At
)20 ꢁC and below, the cyclopropanation reaction was
inhibited.
20
D
7. (a) ½aꢀ )78.6 (c 0.21, acetone); (b) Walborsky, H. M.;
ꢀ
Hornyak, F. M. J. Am. Chem. Soc. 1955, 77, 6026; (c)
Altmann, L. J. J. Am. Chem. Soc. 1969, 91, 5163.
8. GLC–MS were performed with a HP 6890 GC apparatus
equipped with a 12 m · 0.20 mm dimethylpolysiloxane
capillary column, linked to a HP 5973 EIMS. The mixure
showed two signals: 11.77 min (9b) and 11.81 min (9a),
respectively.
9. Nishiyama, H.; Itoh, Y.; Sugawara, Y.; Matsumoto, H.;
Aoki, K.; Itoh, K. Bull. Chem. Soc. Jpn. 1995, 68, 1247.
10. Catalyst was prepared from (S,S)-1,2-diphenylethylene-
diamine and cyclopentanone under standard conditions.
11. Catalysts were prepared from imidazolidines 3a–c and
(CuOTf)2ÆC6H6, and were used after 24 h of mixing.
12. Catalysts were prepared from imidazolidines 3a–c and
Cu(OTf)2, and were used after 24 h of mixing.
In summary, complexes derived from chiral imidazol-
idines and copper(I) or copper(II) triflates were shown
to mediate asymmetric cyclopropanation reactions. By
using 1,1-diphenylethylene and ())-menthyl diazoacet-
ate, the reaction afforded the corresponding cyclopro-
pane carboxylates in good yield and diastereoselectivity.
Although the diastereoselectivity trends of the cyclo-
propanation were similar in both cases, higher yields
were observed when Cu(I) complexes were used. This
work forms the basis for the use of imidazoline-based
systems in diazoester-mediated cyclopropanations.
13. Brunel, J.-M.; Luukas, T. O.; Kagan, H. B. Tetrahedron:
Asymmetry 1998, 9, 1941, and references cited therein.