Diastereoselective titanium-mediated construction of cis-2,3-ring annelated 1-(2′-chloroethyl)cyclopropanols

Article abstract of DOI:10.1039/b211642a

Titanium (IV)-mediated cyclopropanation induced by reaction of cycloalkylmagnesium bromides 1a-e in the presence of titanium tetraisopropoxide with ethyl β-chloropropionate 2 yielded cis-fused 1-(2′-chloroethyl)cyclopropanols 3b-e with diastereoselectivity highly depending on the ring size. Unexpected rearrangement products 5d,e were isolated in some cases.

Full text of DOI:10.1039/b211642a

Diastereoselective titanium-mediated construction of cis-2,3-ring
annelated 1-(2A-chloroethyl)cyclopropanols†
Frédéric Lecornué and Jean Ollivier*
Laboratoire des Carbocycles, UMR 8615, Institut de Chimie Moléculaire et des Matériaux dAOrsay, Bât.
420, Université de Paris-Sud, 91405 Orsay, France. E-mail: jollivie@icmo.u-psud.fr; Fax: +33 1 69 15 62
78; Tel: +33 1 69 15 72 52
Received (in Cambridge, UK) 27th November 2002, Accepted 22nd January 2003
First published as an Advance Article on the web 29th January 2003
Titanium(IV)-mediated cyclopropanation induced by reac-
tion of cycloalkylmagnesium bromides 1a–e in the presence
of titanium tetraisopropoxide with ethyl b-chloropropionate
2 yielded cis-fused 1-(2A-chloroethyl)cyclopropanols 3b–e
with diastereoselectivity highly depending on the ring size.
Unexpected rearrangement products 5d,e were isolated in
some cases.
was not significantly modified; the exo isomer was always
observed as the major product.
Unexpected results were obtained with cyclohexylmagne-
sium bromide 1d (entry 4). In fact, although the expected
cyclopropanol 3d (n = 6) was isolated as a single exo-
diastereomer, the main product surprisingly was 1-cyclohex-
ylcyclopropanol 5d (n = 6).
Competing routes could explain the formation of these two
products. The first path (Scheme 1, path A) would proceed
through the classical Kulinkovich reaction to lead to the fused
cyclopropanol 3d. In this case, formation of the intermediate
titanafurane A (on the cyclohexyl moiety) could be slower than
nucleophilic attack of the cyclohexylmagnesium bromide9 on
the ester function, leading to the anionic intermediate B (path
B).
1-Vinylcyclopropanols and their derivatives (esters, trime-
thylsilylethers, etc.) are smoothly and efficiently prepared from
1-(2A-chloroethyl)cyclopropanols.1 They constitute very useful
intermediates for numerous transformations including
C3?C4–8 ring expansions2 and C3?C10,15,20 ring enlarge-
ments.3 Moreover, the corresponding sulfonic esters undergo
regio- and diastereoselective nucleophilic or electrophilic
substitution via s- or p-1,1-ethyleneallylmetal complexes4
allowing large synthetic applications recently reported.
The titanium(IV)-promoted Kulinkovich cyclopropanation,5
to now, appears to be the most efficient method for the
preparation of substituted cyclopropanols; however, this proce-
dure has not been applied to the direct synthesis of various cis-
fused cyclopropane derivatives. In fact, cycloalkylmagnesium
halides (cyclopentyl and cyclohexyl) have been used for
hydroxycyclopropanation of alkenes involving ligand ex-
change.5a,6 Surprisingly, Cha et al.6a claimed that such
Grignard reagents did not yield any cyclopropanation products
in reaction with esters under the Kulinkovich conditions, but de
Meijere et al.7 reported one reaction of formation in low yield of
a fused cyclopropanol from ethyl acetate and cyclohex-
ylmagnesium halide with titanium tetraisopropoxide. We
decided to extend this reaction by the study of the reactivity of
various cycloalkylmagnesium bromides (from three to seven-
membered ring sizes) with ethyl b-chloropropionate 2 promoted
by Ti(OiPr)4 thus allowing the synthesis of 1,2,3-trisubstituted
cyclopropanols.
We suggest that intermediate B undergoes an internal SN2
reaction on the terminal chloride (probably assisted by the
Table 1 Reaction of cycloalkylmagnesium bromidesa 1a–e with ethyl b-
chloropropionate 2 and Ti(OiPr)4
Grignard
Entry
1
n
reagent
eq. Ti(OiPr)4 Products (%)
3
1
b
2
3
4
5
0.2
1
Formation of products 3a–e and by-products 4a, 5d–e as well
as the relationship between diastereoselectivity and ring size are
summarized in Table 1.
The reaction of cyclopropylmagnesium bromide 1a (entry 1)
according to the Kulinkovich protocol did not give the expected
fused cyclopropanol 3a (n = 3), but led only to the product 4a
(n = 3), resulting from the classical addition of the Grignard
reagent to the ester 2.
c
From cyclobutylmagnesium bromide 1b (entry 2), cyclopro-
panol 3b (n = 4) was isolated; only the exo diastereomer was
detected. To our knowledge, such easy formation of 5-alkyl-
bicyclo[2.1.0]pentan-5-ol has never been reported in the
literature.
With cyclopentylmagnesium bromide 1c (entry 3), the
reaction led to a mixture of (exo/endo: 63/37) 6-(2A-chloroethyl)
bicyclo[3.1.0]hexan-6-ol 3c (n = 5).8 The two diastereomers
were separated by chromatography on silica gel; however, even
on varying the experimental conditions (temperature, equiva-
lents of titanium and Grignard species,…), the stereoselectivity
b
4
5
6
7
1
b
0.2
a Reactions were optimized with 4 eq. of Grignard reagents at room
temperature in Et2O–THF. b Pure exo. c Mixure exo/endo 63+37.
† Electronic supplementary information (ESI) available: NMR, IR, MS and
HRMS data. See http://www.rsc.org/suppdata/cc/b2/b211642a/
584
CHEM. COMMUN., 2003, 584–585
This journal is © The Royal Society of Chemistry 2003

Scheme 3
equivalent of titanium tetraisopropoxide gave an unexploitable
mixture.
The interest of this new strategy was exemplified by the total
synthesis of the unknown 8-exo-amino bicyclo[5.1.0]octane-
8-carboxylic acid 8. Thus, after tosylation and dehydrochlorina-
tion of the alcohol 3e, the resulting 1-vinylcyclopropyltosylate
7 successively underwent palladium(0)-catalyzed azidation,
oxidation of the double bond and reduction of the azide function
following a previously reported procedure13 to give the required
fused aminoacid 8 in 56% yield (Scheme 3).
In conclusion, we have developed a new and concise
diastereoselective construction of fused 1-alkylcyclopropanols.
The ability of such compounds to yield 1-vinylcyclopropanols
opens a wide range of other useful applications which are under
current investigation.
Scheme 1
Notes and references
1 I. Sylvestre, J. Ollivier and J. Salaün, Tetrahedron Lett., 2001, 42, 4991
and references cited therein.
presence of a titanium species) generating an unstable oxetane
intermediate C which would afford the 1-cyclohexylcyclopro-
panol 5d after ring contraction and hydrolysis. Such ring
contraction of oxetane to cyclopropyl moiety has never been
reported previously. However, another mechanism could also
involve intermediate B, which in the presence of Ti(OiPr)4,
would undergo a known halogen metal exchange with the
Grignard reagent10 leading to the intermediate D capable of
undergoing cyclization11 to the cyclopropanol 5d. Moreover, in
using only 0.2 equivalents of Ti(OiPr)4, the yield was hardly
decreased (20%), but the ratio 3d/5d reached 50+50. It must be
underlined that when the reaction was carried out without
Ti(OiPr)4, only tertiary alcohol 4d (n = 6) was isolated;
furthermore, treatment of the cyclopropanol 3d by Ti(OiPr)4
and cyclohexylmagnesium bromide did not yield alcohol 5d (n
= 6).
To endorse the formation of the intermediate D, we treated
chloroketone 6 (Scheme 2), which was prepared independently
by known procedures, by 4 equivalents of cyclohexylmagne-
sium bromide 1d in the presence of titanium tetraisopropoxide.
Surprisingly, with 1 equivalent of Ti(OiPr)4, only cyclopropa-
nol 5d was isolated. Such formation of cyclopropanol from b-
haloketone has previously been reported in the literature, but
using SmI2.12
2 J. Salaün, Top. Curr. Chem., 1988, 144, 1.
3 J. Schnaubelt, A. Ullmann and H.-U. Reissig, Synlett, 1995, 1223; A.
Ullmann, H.-U. Reissig, H. -U. and O. Rademacher, Eur. J. Org. Chem.,
1998, 20, 2541.
4 J. Ollivier, N. Girard and J. Salaün, Synlett, 1999, 1539; A. Stolle, J.
Ollivier, P. P. Piras, A. de Meijere and J. Salaün, J. Am. Chem. Soc.,
1992, 114, 4051.
5 (a) O. G. Kulinkovich and A. de Meijere, Chem. Rev., 2000, 100, 2789;
(b) O. G. Kulinkovich, S. V. Sviridov, D. A. Vasilevski and T. S.
Pritytskaya, Zh. Org. Khim., 1989, 25, 2244; (c) O. G. Kulinkovich, S.
V. Sviridov and D. A. Vasilevski, Synthesis, 1991, 234; (d) O. G.
Kulinkovich, S. V. Sviridov, D. A. Vasilevski, A. I. Savchenko and T.
S. Pritytskaya, Zh. Org. Khim., 1991, 27, 294.
6 (a) J. Lee, H. J. Kim and J. K. Cha, J. Am. Chem. Soc., 1996, 118, 4198;
(b) J. Lee, C. H. Kang, H. J. Kim and J. K. Cha, J. Am. Chem. Soc., 1996,
118, 291; (c) O. L. Epstein and O. G. Kulinkovich, Tetrahedron Lett.,
1998, 39, 1823.
7 V. Chaplinsky, H. Winsel, M. Kordes and A. de Meijere, Synlett, 1997,
111.
8 The configuration of these cyclopropanols has been assigned on the
basis of 2D-NOESY experiments.
9 However, it has been previously reported that alkylmagnesium halides
react faster with Ti(OiPr)4 than with a ketone or an ester : N. Morlender-
Vais, J. Kaftanov and I. Marek, Synthesis, 2000, 917; F. Sato, H. Urabe
and S. Okamoto, Chem. Rev., 2000, 100, 2835.
10 While to our knowledge, no cases with chloride derivative have been
related in the literature, i-PrMgI and i-PrMgBr are widely used in
halogen magnesium exchange reactions: M. Rottländer, L. Boymond, L.
Bérillon, A. Leprêtre, G. Varchi, S. Avolio, H. Laaziri, G. Quéquiner, A.
Ricci, G. Cahiez and P. Knochel, Chem. Eur. J., 2000, 6, 767 and
references cited therein.
Cycloheptylmagnesium bromide 1e (entry 5) gave almost
exclusively the expected exo fused cyclopropanol 3e (n = 7);
only traces (2%) of 1-cycloheptylcyclopropanol 5e (n = 5)
were detected, suggesting the same mechanism previously
proposed for the six-membered ring, but in this case, use of one
11 Some analogy can be observed in the intramolecular cyclization of b-
halo-substituted acetals: J.-W. Huang, C.-D. Chen and M.-k. Leung,
Tetrahedron Lett., 1999, 40, 8647.
12 S.-h. Fukuzawa, Y. Niimoto and S. Sakai, Tetrahedron Lett., 1991, 32,
7691; T. Imamoto, T. Hatajima, N. Takayima, T. Takeyama, Y. Kamiya
and T. Yoschizawa, J. Chem. Soc., Perkin Trans. 1, 1991, 3127.
13 S. Racouchot, I. Sylvestre, J. Ollivier, Y. Y. Kozyrkov, A. Pukin, O. G.
Kulinkovich and J. Salaün, Eur. J. Org. Chem., 2002, 24, 2160.
Scheme 2
CHEM. COMMUN., 2003, 584–585
585