Access to functionalized cyclopropylcarbinyl compounds from homoallylic ethers via zirconocene intermediates

Article abstract of DOI:10.1016/j.tetlet.2004.06.122

Cyclopropylcarbinylzirconium complexes were generated from homoallylic ethers and zirconocene (Cp₂ZrCl₂/2n-BuLi). They underwent further in situ transformations that is, bromination-substitution, transmetallation-functionalization and insertion reactions to afford various cyclopropylcarbinyl derivatives. In contrast, ring-opening products could be selectively obtained by treating the zirconium complex with MeLi, prior to functionalization.

Full text of DOI:10.1016/j.tetlet.2004.06.122

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 6449–6451
Access to functionalized cyclopropylcarbinyl compounds from
homoallylic ethers via zirconocene intermediates
Jean-Luc Vasse and Jan Szymoniak*
´ ´ ´
UMR 6519––Reactions Selectives et Applications, CNRS––Universite de Reims Champagne-Ardenne,
UFR Sciences, BP 1039, 51687 REIMS Cedex 2, France
Received 25 May 2004; accepted 29 June 2004
Available online 20 July 2004
Abstract—Cyclopropylcarbinylzirconium complexes were generated from homoallylic ethers and zirconocene (Cp2ZrCl2/2n-BuLi).
They underwent further in situ transformations that is, bromination–substitution, transmetallation–functionalization and insertion
reactions to afford various cyclopropylcarbinyl derivatives. In contrast, ring-opening products could be selectively obtained by treat-
ing the zirconium complex with MeLi, prior to functionalization.
Ó 2004 Elsevier Ltd. All rights reserved.
The synthetic methods employed for preparing cyclo-
propanes are typically based upon the direct cycloprop-
anation of alkenes.1 However, in the last decade some
new approaches to cyclopropanes have been developed.
Among them, the Kulinkovich synthesis of cyclopro-
panols from esters and the de Meijere synthesis of terti-
ary cyclopropylamines from amides should be noticed.2
We have recently reported zirconium- and titanium-
mediated reactions that allow the preparation of pri-
Scheme 1.
mary cyclopropylamines from nitriles,3 cyclopropanes
from aldehydes and ketones,4 as well as from allylic5
and homoallylic6 ethers. The last reaction is depicted
the metal–carbon bond, followed by a fast ring opening
in Scheme 1. It proceeds through the c-elimination of
of the resulting cyclopropylcarbinyl radical.8 We
the alkoxy group to afford the cyclopropylcarbinylzirco-
thought that non ring-opening transformations from A
nium intermediate A (Eq. 1). Cyclopropylcarbinyl–
homoallyl rearrangement takes place occasionally (A
could be carried out, due to the unfavoured homolytic
cleavage of the zirconium–carbon bond.9 In order to test
predominant over B, Eq. 2), depending on the substrate
structure.
this hypothesis, we first reacted the homoallylic ether 1a
with (1-butene)ZrCp2, preformed in situ from Cp2ZrCl2
and n-BuLi.10 The exchange of 1-butene with 1a fol-
lowed by the c-elimination of the OMe group led to
the cyclopropylcarbinylzirconocene 2a. The hydrolysis
(deuterolysis) of 2a afforded the cyclopropane 3a
(X=H or D) in 80% isolated yield (Scheme 2).6
We envisioned to explore in situ transformations from A
with the aim of converting homoallylic ethers directly
into cyclopropylcarbinyl derivatives. So far, a few syn-
thetic applications of cyclopropylcarbinylmetal com-
plexes (M=Fe, Co, Pd, Ni, Rh, Mn, Pt) have been
reported, dealing with ring-opening reactions.7 Such
reactions typically proceed via homolytic cleavage of
As expected, the zirconium complex 2a underwent the
bromonolysis reaction without ring opening, thus con-
firming the heterolytic cleavage of the C–Zr bond in this
reaction.11 Thus, treatment of 2a with NBS in THF at
Keywords: Cyclopropylcarbinyl compounds; Homoallylic ethers; Zirco-
nium.
20 °C provided the cyclopropylcarbinyl bromide 4a in
74% yield (NMR yield starting from the ether 1a).12
The bromide 4b was obtained in the analogous manner
*
Corresponding author. Tel.: +33(0)-3269-13244; fax: +33(0)-3326-
913166; e-mail: jan.szymoniak@univ-reims.fr
0040-4039/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.06.122

6450
J.-L. Vasse, J. Szymoniak / Tetrahedron Letters 45 (2004) 6449–6451
Scheme 2.
from 1b. Although cyclopropylcarbinyl–homoallyl rear-
rangement has been demonstrated to operate starting
from 1b,6 no acyclic olefinic bromide was obtained in
this case. This might be rationalized by assuming that
Curtin–Hammett conditions apply, in which more rapid
bromonolysis takes place from A than from B (Scheme
1, Eq. 2).13
that in our case, both insertion and transmetallation
can be applied to transform cyclopropylcarbinylzircon-
ocenes into functionalized cyclopropanes. Firstly, reac-
tion of 2a with BnNC was carried out in toluene at
40°C for 6h. Hydrolysis of the reaction mixture and
subsequent reduction of the resulting crude aldehyde
afforded the alcohol 6 in 36% yield (starting from 1a)
(Scheme 3).17 Secondly, transmetallation to copper
using 2a and a catalytic amount of CuI allowed further
acylation and allylation reactions.18 In this way, ketone
7 and homoallylcyclopropane 8 were formed, the latter
being accompanied by the ring-opening compound 9.
The transmetallation of 2a using HgCl2 was also
performed to afford the mercury derivative 10 (Scheme
3) that can be isolated and characterized by NMR
spectroscopy.
The cyclopropylcarbinyl bromides 4a–b underwent sub-
sequent displacement reactions with mild nonbasic
nucleophiles. Carbon-, oxygen-, nitrogen- and sulfur-
based nucleophilic reagents could be employed (Table
1). These reactions were performed under mild condi-
tions to afford functionalized cyclopropanes 5a–f in
moderate to good yields.14 A possible access to various
functionalized cyclopropylcarbinyl derivatives from
readily available homoallylic ethers should be syntheti-
cally useful.
Finally, we tried to favour reactions with ring opening
from the cyclopropylcarbinylzirconocene 2a, and found
that such reactions can be carried out by activating the
complex with MeLi. Thus, treatment of 2a with MeLi
at low temperature and subsequent addition of
PhCHO or PhCOClafforded, respectiveyl, the aclohol
11 and ketone 12 in moderate yields (Scheme 4). These
Alkylzirconocenes are rather unreactive towards typical
carbon-based electrophiles. However, by using these
complexes, C–C bond constructions can be achieved
through carbenoid insertion into the C–Zr bond,15 or
transmetallation to various metals.16 We have noticed
Table 1. Preparation of the bromides 4a–b and their reactions with nucleophiles
Entry
Ether
Nu
5
Yield, % (trans:cis)a
1
2
3
4
5
6
1a
2b
1a
1a
1a
2b
N3
N3
5a
5b
5c
5d
5e
5f
60
47 (1.6:1)
CN
68
24
OPh
SO2Ph
SO2Ph
43
33 (1.6:1)
a Isolated yields.
Scheme 3. Reagents and conditions: (i) BnNC; (ii) HCl(3N); (iii) NaBH
.
4

J.-L. Vasse, J. Szymoniak / Tetrahedron Letters 45 (2004) 6449–6451
6451
3521; (c) Grovenstein, E.; Black, K. W.; Goel, S. C.;
Hughes, R. L.; Northrop, J. H.; Streeter, D. L.; Van
Derveer, D. J. Org. Chem. 1989, 54, 1671.
8. For the use of cyclopropylcarbinyl radicals in synthesis
and as radical clocks, see: (a) Newcomb, M.; Radicals in
Organic Synthesis; Sibi, M. P., Ed.; Wiley, 2001; Vol. 1,
Chapter 3.3; (b) Gansga¨uer, A.; Pierobon, M.; Radicals in
Organic Synthesis; Sibi, M. P., Ed.; Wiley, 2001; Vol. 2; (c)
Zard, S. Z. In Radical Reactions in Organic Synthesis;
Compton, R. G., Davies, S. G., Evans, J., Eds.; Oxford
University Press: Oxford, 2003, Chapters 3.2 and 3.3.
9. (a) Only recently, a radical mechanism has been postulated
for some particular zirconocene-mediated reactions, see:
Fujita, K.; Nakamura, T.; Yorimitsu, H.; Oshima, K. J.
Am. Chem. Soc. 2001, 123, 3137; (b) Fujita, K.; Yorimitsu,
H.; Oshima, K. Synlett 2002, 337.
10. For reviews of ÔCp2ZrÕ chemistry, see: (a) Negishi, E.;
Takahashi, T. Bull. Chem. Soc. Jpn. 1998, 71, 755; (b)
Negishi, E.; Kondakov, D. Y. Chem. Soc. Rev. 1996, 26,
417; (c) Negishi, E.; Takahashi, T. Acc. Chem. Res. 1994,
27, 124.
11. A nonradicalmechanism is asol consistent with the
electrophilic halogenation reported for simpler alkylzirco-
nocenes, see: (a) Hart, D. W.; Schwartz, J. J. Am. Chem.
Soc. 1974, 96, 8115; (b) Bertelo, C. A.; Schwartz, J. J. Am.
Chem. Soc. 1976, 98, 262.
Scheme 4.
reactions possibly proceed through ring opening from
an intermediate zirconium ate complex. They involve a
substituted homoallyl anion equivalent and lead to the
creation of a quaternary carbon centre.
In conclusion, we have generated cyclopropylcarbinyl-
zirconocenes from homoallylic ethers and studied their
in situ transformations into functionalized cyclopropyl-
carbinylcompounds. A remarkabel feature of these
reactions is that they occurred without ring opening.
However, ring-opening products could be obtained
selectively by treating the zirconium intermediate with
MeLi prior to functionalization.
References and notes
1. (a) Brookhart, M.; Studabaker, W. B. Chem. Rev. 1987,
87, 411; (b) Salaun, J. In The Chemistry of the Cyclopropyl
12. Similarly, the reaction of 2a with I2 gave the correspond-
ing iodide; however, this compound appeared unstable.
13. Starting from an homoallylic ether bearing two alkyl
groups, complex reaction mixtures containing alkenes
were obtained.
¨
Group; Rappoport, Z., Ed.; Wiley: New York, 1987; (c) de
Meijere, A., Ed.Houben-Weyl, Methods of Organic
Chemistry; Thieme: Stuttgart, 1997; Vol. 117.
1
2. For reviews see: (a) de Meijere, A.; Kozhushkov, S. I.;
Savchenko, A. I. In Titanium and Zirconium in Organic
Synthesis; Marek, I., Ed.; Wiley–VCH: Weinheim, 2002,
Chapter 11; (b) Kulinkovich, O. G.; de Meijere, A. Chem.
Rev. 2000, 100, 2789; (c) Kulinkovich, O. G. Chem. Rev.
2003, 103, 2597.
3. (a) Bertus, P.; Szymoniak, J. Chem. Commun., 2001, 1792;
(b) Bertus, P.; Szymoniak, J. J. Org. Chem. 2002, 67, 3965;
(c) Laroche, C.; Bertus, P.; Szymoniak, J. Tetrahedron
Lett. 2003, 44, 2485; (d) Bertus, P.; Szymoniak, J. Synlett,
2003, 265; (e) Bertus, P.; Szymoniak, J. J. Org. Chem.
2003, 68, 7133.
14. Selected data for 5a: H NMR (CDCl3, 250MHz): d 1.31
(m, 2H), 1.88 (m, 1H), 2.98 (ABX system, 2H), 7.03–7.43
(m, 10H); 13C NMR (CDCl3, 62.5MHz): d 17.7, 23.1,
34.6, 51.5, 125.1, 125.8, 126.7, 127.3, 127.5, 129.2, 139.3,
144.7.
15. (a) Dixon, S.; Whitby, R. J. In Titanium and Zirconium in
Organic Synthesis; Marek, I., Ed.; Wiley–VCH: Wein-
heim, 2002, Chapter 3; (b) Dixon, S.; Fillery, S. M.;
Kasatkin, A.; Norton, D.; Tomas, E.; Whitby, R. J.
Tetrahedron 2004, 60, 1401.
16. (a) Wipf, P.; Jahn, W. Tetrahedron 1996, 52, 12853; (b)
Lipshutz, B. H.; Pfeiffer, S. S.; Nogson, K.; Tomioka, T.
In Titanium and Zirconium in Organic Synthesis; Marek,
I., Ed.; Wiley–VCH: Weinheim, 2002, Chapter 4.
4. (a) Bertus, P.; Gandon, V.; Szymoniak, J. Chem. Commun.
2000, 171; (b) Gandon, V.; Bertus, P.; Szymoniak, J. Eur.
J. Org. Chem. 2000, 3713.
1
17. Selected data for 6: H NMR (CDCl3, 250MHz): d 0.96
5. Gandon, V.; Szymoniak, J. Chem. Commun., 2002, 1308.
6. Gandon, V.; Laroche, C.; Szymoniak, J. Tetrahedron Lett.
2003, 44, 4827.
7. (a) de Meijere, A., Ed.Houben-Weyl, Methods of Organic
Chemistry; Thieme: Stuttgart, 1997; Vol. E17, p 2686, and
references cited therein; (b) Only a few examples of
reactions without ring opening have been reported, see:
Carr, D. B.; Schwartz, J. J. Am. Chem. Soc. 1979, 101,
(m, 1H); 1.25 (m, 2H), 1.69 (m, 2H), 3.66 (td, J= 6.5, 2.0,
Hz, 2H), 7.04–7.38 (m, 10H); 13C NMR (CDCl3,
62.5MHz): d 20.8, 23.3, 34.5, 35.2, 63.2, 121.2, 126.9,
128.2, 128.7, 128.8, 131.0, 142.0, 147.6; HRMS: found
239.1426, calcd for C17H19O+ 239.1436 (24, M+1).
18. This result is noteworthy, since transmetallations are
generally more difficult with alkyl- than with alkenylzirco-
nocenes.