Zirconocene-catalyzed reactions of alkenyl-substituted enol ethers

Article abstract of DOI:10.1002/anie.200601959

(Chemical Equation Presented) A straightforward approach: An unprecedented zirconocene-catalyzed isomerization-magnesation reaction of readily available alkene-substituted enol ethers allows the direct synthesis of functionalized Grignard reagents. Cp = c

Full text of DOI:10.1002/anie.200601959

Communications
Rearrangement
DOI: 10.1002/anie.200601959
Zirconocene-Catalyzed Reactions of Alkenyl-
Substituted Enol Ethers**
JosØ Barluenga,* Lucía lvarez-Rodrigo,
FØlix Rodríguez, and Francisco J. Faæanµs
Organozirconocene complexes[1] have become very useful
tools in organic chemistry since the pioneering works of
Schwartz[2] and Negishi.[3] Despite the extensive advances in
this area that have been accomplished in recent years, very
few catalytic processes involving the use of zirconocene
complexes have been described.[4] Particularly noteworthy are
those catalytic carbometalative processes developed mainly
by Negishi[5] and Hoveyda.[6] Some years ago, we initiated a
study of new cross-coupling reactions between organozirco-
nocene complexes and heterosubstituted alkenes (principally
enol ethers).[7] In this context, the extensive work by Marek
and co-workers should also be remarked upon.[8] In an
attempt to further develop this interesting chemistry, we
decided to investigate intramolecular versions of the reaction
of alkene–zirconocene complexes and enol ethers
(Scheme 1).[9] Thus, we envisioned that alkenyl-substituted
Scheme 1. Proposed intramolecular version of the reaction of alkene–
zirconocene complexes and enol ethers. Cp=cyclopentadienyl.
enol ethers such as 1 could coordinate to a zirconocene
complex generated in situ to give species 2, which could react
through a carbocyclization process to give 3.These new
complexes would evolve by a b-elimination of the alkoxy
group to finally give 4, which could be further functionalized
by reaction with electrophiles.Thus, the general reaction from
1 to 4 would suppose an interesting isomerization–metalation
[*] Prof. Dr. J. Barluenga, L. lvarez-Rodrigo, Dr. F. Rodríguez,
Dr. F. J. Faæanµs
Instituto Universitario de Química Organometµlica “Enrique
Moles”, Unidad Asociada al C.S.I.C.
Universidad de Oviedo
Juliµn Clavería 8, 33006 Oviedo (Spain)
Fax: (+34)98-510-3450
E-mail: barluenga@uniovi.es
[**] This research was supported by the DGICYT (BQU-2001-3853). The
authors thank the MEC (FPU predoctoral grant for L.A.-R.) and the
MEC and Fondo Social Europeo (contract under the “Ramón y
Cajal” program for F.R.). Helpful comments from a referee are
deeply acknowledged.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
6362
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 6362 –6365

Angewandte
Chemie
process.The final goal of our investigation was the develop-
ment of a catalytic method, and herein we describe a new
zirconocene-catalyzed isomerization–magnesation reaction
of alkene-substituted enol ethers.
At the outset of this study, the reaction was tested with
stoichiometric amounts of zirconocene dichloride.Thus, the
reactive low-valent zirconocene species were generated by
the method developed by Negishi by mixing zirconocene
dichloride [ZrCl2Cp2] with two equivalents of a Grignard or
an organolithium reagent.This reaction gives the correspond-
ethylmagnesium bromide was used either in diethyl ether or
THF (Table 1, entries 1 and 2).However, 5a was obtained in
78% yield as a 1:1 mixture of diastereoisomers when
propylmagnesium chloride was used in THF.A small
amount of 6a was also isolated (5% yield; Table 1, entry 3).
Similar results, although with slightly lower yields, were
obtained when this reaction was performed in diethyl ether
(Table 1, entry 4).Finally, 5a was obtained in 88% yield (1:1
mixture of diastereoisomers) by using butyllithium (Table 1,
entry 5).However, the selectivity of the reaction improved
and a 5:1 mixture of diastereoisomers was obtained when the
reaction was carried out in diethyl ether (78% yield; Table 1,
entry 6).[10] It is of note that formation of 6a was completely
suppressed by using butyllithium.
ing alkene–zirconocene complex, which after
a ligand
exchange reaction would generate the necessary complex 2.
Initially, the reaction was studied on the model alkenyl-
substituted enol ether 1a (Scheme 2 and Table 1).Thus, 1a
Once we had found the right conditions to direct the
reaction to the isomerized compound 5a, thus avoiding the
formation of undesired 6a, we decided to investigate the
reaction under catalytic conditions with zirconium.We
speculated that under catalytic conditions, the excess of
organometallic reagents (Grignard or organolithium com-
pounds) could react with intermediate 4 (see Scheme 1)
through a double transmetalation reaction, thereby regener-
ating the catalytic zirconium species and forming a new
organometallic reagent related to 4 (with magnesium or
lithium instead of zirconium).
To find the optimum conditions, model compound 1a was
treated with an excess of the corresponding organometallic
compound in the presence of catalytic amounts of commer-
cially available zirconocene dichloride under various reaction
conditions (Scheme 3 and Table 2).As the best results in the
stoichiometric reactions were reached by using butyllithium,
we started our catalytic studies with this organometallic
Scheme 2. Zirconocene-mediated reaction of enol ether 1a.
Table 1: Products obtained from the zirconocene-mediated reaction of
enol ether 1a and different Grignard or organolithium reagents.
Entry
RM
Solvent
Product
Yield [%][a]
d.r. of 5a[b]
1
2
3
4
5
6
EtMgBr
EtMgBr
PrMgCl
PrMgCl
BuLi
THF
Et2O
THF
Et2O
THF
Et2O
6a
6a
5a/6a
5a/6a
5a
90
86
78/5
66/5
88
–
–
1:1
1:1
1:1
5:1[c]
BuLi
5a
78
[a] Yield based on starting material 1a. [b] Determined by 1H NMR
spectroscopic analysis of the reaction crude product. [c] The major
diastereoisomer was found to be the 1,3-anti isomer (see the Supporting
Information).
was treated with one equivalent of [ZrCl2Cp2] and two
equivalents of different Grignard or organolithium reagents
at À788C.The reaction was allowed to reach room temper-
ature and after 1 h was quenched with silica gel.Thus, the
expected alkenol 5a, which is formed by simple hydrolysis of
the corresponding zirconium complex 4 (see Scheme 1), was
obtained together with different amounts of alcohol 6a.
Formation of 6a could be explained by following the
mechanism proposed by Marek and co-workers for the
synthesis of alkenyl–zirconium complexes from acyclic enol
ethers.[8c,i] Thus, we suppose the generation of vinyl–zirconium
complex 7a, as the by-product 6a and a molecule of ethene
are generated after hydrolysis.
Scheme 3. Zirconocene-catalyzed reaction of enol ether 1a.
Table 2: Zirconocene-catalyzed reaction of enol ether 1a and different
Grignard or organolithium reagents.
Entry RM
[ZrCl2Cp2] Solvent Additive Products Yield d.r.
[mol%]
[%][a] of 5a
1
2
3
4
5
6
7
BuLi
5–25
25
5
15
15
5
THF[b] –polymer – –
EtMgBr
EtMgBr
PrMgCl
PrMgCl
PrMgCl
PrMgCl
THF[b]
THF[b]
Et2O
THF
–
–
–
–
PPh3
PPh3
6a/8a
6a/8a
5a/6a
5a/6a
5a
25/66
5/85
63/12 2:1
76/15 1:1
62
82
–
–
[c]
[c]
Et2O
THF
2:1
1:1
5
5a
As shown in Table 1, the formation of the desired
compound 5a or by-product 6a was highly influenced by
the structure of the Grignard or organolithium reagent used.
Thus, we observed the exclusive formation of 6a when
[a] Yield of the isolated products based on starting material 1a.
[b] Similar results were found when Et2O was used as the solvent.
[c] 10 mol% of PPh3 was used.
Angew. Chem. Int. Ed. 2006, 45, 6362 –6365
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6363

Communications
reagent.However, we did not detect the formation of 5a in
any case attempted and mainly observed polymerization of
the starting material (Table 2, entry 1).In consonance with
the results of the stoichiometric reactions, by using ethyl-
magnesium bromide we observed the formation of 6a along
with different quantities of the new alcohol 8a, depending on
the amount of catalyst used (Table 2, entries 2 and 3).
Formation of 8a is easily explained by a simple zirconium-
catalyzed carbomagnesiation of the double bond of 6a and
further hydrolysis.Fortunately, we observed the formation of
the desired alkenol 5a together with a small amount of 6a
(Table 2, entries 4 and 5) after only 1 h at room temperature
when propylmagnesium chloride was employed in the pres-
ence of 20 mol% of [ZrCl2Cp2].Both diethyl ether and THF
were found to be appropriate solvents; however, a slightly
higher diastereoselectivity was found in diethyl ether and
higher yield in THF.The reaction times became longer when
the amount of catalyst was lowered, and 24 h were required
for complete conversion of the starting material when
5 mol% of [ZrCl2Cp2] was used.We tried some reactions in
the presence of an additive[11] in an attempt to totally avoid
the formation of 6a and to achieve complete conversions in
shorter times.To our delight, under the optimal reaction
conditions (5 mol% of [ZrCl2Cp2], 10 mol% PPh3), starting
compound 1a was completely transformed into isomerized
alcohol 5a after 8 h at room temperature in THF (82% yield
of the isolated product, 1:1 mixture of diastereoisomers;
Table 2, entry 7).The formation of 6a was not observed under
these conditions.Again, slightly higher selectivity, although
lower yield, was found when diethyl ether was used as the
solvent (Table 2, entry 6).
Scheme 4. Proposed mechanism for the zirconocene-catalyzed reac-
tion of 1a.
With an effective zirconocene-catalyzed reaction in hand,
we investigated the scope of the process.Table 3 shows the
results of the reactions performed under the optimized
catalytic conditions previously found for 1a.The reaction
worked for a range of alkene-substituted enol ethers 1, and
the formed Grignard reagent could be trapped with different
typical electrophiles for organomagnesium reagents.The final
rearranged products 5 were isolated in high yield, but
unfortunately the reaction gave mixtures of both possible
diastereoisomers (d.r. ꢀ 1:1, except for starting material 1e
d.r. = 14:1; however, the yield in this case was lower).
Attempts to improve the selectivity by changing the reaction
A proposed mechanism for the
catalytic reaction is shown in
Table 3: Zirconocene-catalyzed reaction of enol ethers 1 in the synthesis of functionalized alcohols 5.
Scheme 4.Initially, the zirconocene
dichloride reacts with two equiva-
lents of propylmagnesium chloride
to form the corresponding dipro-
pylzirconocene complex 9, which
further evolves, as described by
Negishi, to give the propene–zirco-
nocene complex 10 with the loss of a
molecule of propane.This sixteen-
electron complex reacts in the pres-
ence of the alkene-substituted enol
ether 1a through a ligand-exchange
process, thus releasing a molecule
of propene and forming the eight-
een-electron zirconocene complex
2a.A carbocyclization reaction fur-
nishes the bicyclic complex 3a.
Subsequent elimination of zirco-
nium alkoxide affords the inter-
mediate 4a.[12] Finally, a double
Entry
R1
R2
Starting diene
E+(E)
Product
Yield [%][a]
1
2
3
Ph
Cy
Cy
H
H
H
1a
1b
1b
H2O(H)
H2O(H)
D2O(D)
5a
5b
5c
82
93
85
4
Cy
H
1b
5d
5e
56[b]
5
Cy
H
1b
67
5
6
heptyl
heptyl
H
H
1c
1c
H2O(H)
D2O(D)
5 f
5g
94
84
7
heptyl
H
1c
5h
64[b]
8
9
10
Ph(CH2)2
Ph(CH2)2
H
H
H
Ph
1d
1d
1e
H2O(H)
D2O(D)
H2O(H)
5i
5j
5k
90
88
52[c]
magnesium–zirconium
exchange
[a] Yield of the isolated products based on starting material 1; unless noted, an inseparable 1:1 mixture
of the two possible diastereoisomers was obtained. [b] Both diastereoisomers from the 1:1 mixture were
readily separated. [c] In this case, the formation of several other by-products was observed, and the
major diastereoisomer was found to be (2R*,3R*)-3-methyl-2-phenyl-4-penten-1-ol (d.r. =14:1; see the
Supporting Information). E=electrophile.
gives the final organomagnesium
compound 11 and regenerates the
dipropylzirconocene complex 9,
which initiates
a new catalytic
cycle.[11]
6364 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 6362 –6365

Angewandte
Chemie
3411; m) D.P.Lewis, P.M.Muller, R.J.Whitby, R.V.H.Jones,
Tetrahedron Lett. 1991, 32, 6797.
conditions were unsuccessful until now.For this reason,
current efforts are being directed to find the appropriate
catalyst or reaction conditions to achieve higher selectivities.
In summary, we have developed a new zirconocene-
catalyzed isomerization–magnesation reaction of readily
available alkene-substituted enol ethers.[13] The net process
supposes the migration of the vinyl moiety of the enol ether to
the internal position of the initial alkene and the magnesation
of the terminal carbon atom of that initial alkene moiety
(Scheme 5).The global reaction may also be considered as an
[5] For some recent papers, see: a) Z.Tan, E.Negishi,
Angew.
Chem. 2004, 116, 2971; Angew. Chem. Int. Ed. 2004, 43, 2911;
b) E.Negishi, Z.Tan, B.Liang, T.Novak, Proc. Natl. Acad. Sci.
USA 2004, 101, 5782; c) S.Huo, J.Shi, E.Negishi, Angew. Chem.
2002, 114, 2245; Angew. Chem. Int. Ed. 2002, 41, 2141.
[6] a) R.R.Cesati III, J.de Armas, A.H.Hoveyda, Org. Lett. 2002,
4, 395; b) J.de Armas, A.H.Hoveyda, Org. Lett. 2001, 3, 2097;
c) J.de Armas, S.P.Kolis, A.H.Hoveyda,
J. Am. Chem. Soc.
2000, 122, 5977; d) A.H.Hoveyda in Titanium and Zirconium in
Organic Synthesis (Ed.: I. Marek), Wiley-VCH, Weinheim,
Germany, 2002, p.180.
[7] a) J.Barluenga, F.Rodríguez, L.lvarez-Rodrigo, F.J.Faæanµs,
Chem. Soc. Rev. 2005, 34, 762; b) J.Barluenga, A.Fernµndez, L.
lvarez-Rodrigo, F.Rodríguez, F.J.Faæanµs, Synlett 2005, 2513;
c) J.Barluenga, L.lvarez-Rodrigo, F.Rodríguez, F.J.Faæanµs,
Angew. Chem. 2004, 116, 4022; Angew. Chem. Int. Ed. 2004, 43,
3932; d) J.Barluenga, F.Rodríguez, L.lvarez-Rodrigo, J.M.
Zapico, F.J. Faæanµs, Chem. Eur. J. 2004, 10, 109; e) J.
Barluenga, F.Rodríguez, L.lvarez-Rodrigo, FJ..Faæanµs,
Chem. Eur. J. 2004, 10, 101.
Scheme 5. General reaction of the zirconocene-catalyzed isomeriza-
tion–magnesation process.
[8] a) N.Chinkov, A.Levin, I.Marek, Angew. Chem. 2006, 118, 479;
Angew. Chem. Int. Ed. 2006, 45, 465; b) I.Marek, N.Chinkov, A.
Levin, Synlett 2006, 501; c) S.Farhat, I.Zouev, I.Marek,
Tetrahedron 2004, 60, 1329; d) N.Chinkov, S.Majumdar, I.
Marek, J. Am. Chem. Soc. 2003, 125, 13258; e) S.Farhat, I.
Marek, Angew. Chem. 2002, 114, 1468; Angew. Chem. Int. Ed.
2002, 41, 1410; f) N.Chinkov, H.Chechik, S.Majumdar, A.
Liard, I.Marek, Synthesis 2002, 2473; g) N.Chinkov, S.
Majumdar, I.Marek, J. Am. Chem. Soc. 2002, 124, 10282;
h) A.Liard, J.Kaftanov, H.Chechik, S.Farhat, N.Morlender-
Vais, C.Averbuj, I.Marek, J. Organomet. Chem. 2001, 624, 26;
i) A.Liard, I.Marek, J. Org. Chem. 2000, 65, 7218.
equivalent of the vinylmagnesation of terminal olefins, which
is still a rather difficult challenge in the field of carbometal-
lation of alkenes.This unprecedented process is one of the
few examples of zirconocene-catalyzed reactions described in
the literature.The reaction supposes a simple methodology to
access to functionalized Grignard reagents which could find
application in organic synthesis programs.
Received: May 17, 2006
Published online: August 23, 2006
[9] For interesting zirconocene-induced cocyclization–elimination
reactions, see: a) D.R.Owen, R.J.Whitby, Synthesis 2005, 2061;
Keywords: Grignard reagents · homogeneous catalysis ·
rearrangement · synthetic methods · zirconium
.
b) M.Kotora, G.Gao, Z.Li, T.Takahashi,
Tetrahedron Lett.
2000, 41, 7905; c) D.B. Millward, R.M. Waymouth, Organo-
metallics 1997, 16, 1153.
[10] The reaction led to a complex mixture of unidentified com-
pounds when less polar solvents, such as toluene or hexane, were
used.Similar variations on the diastereoselectivity by changing
the solvent (from diethyl ether to THF) have been observed in
related works; this solvent effect has been attributed to the
chelation between the zirconium center and the solvent; for
details, see: A.F.Houri, M.T.Didiuk, Z.Xu, N.R.Horan, A.H.
Hoveyda, J. Am. Chem. Soc. 1993, 115, 6614.
[11] Addition of catalytic amounts of PPh3 has led to important
improvements in related processes; see for example ref.[4k].
The role played by the phosphine moiety in these processes is
unclear; however, we hypothesize the possibility of a stabiliza-
tion of sixteen-electron zirconocene complexes such as 9 and/or
10 (see Scheme 4).As these complexes are the real catalytic
species of the reaction, their stabilization by coordination with
the phosphine allows a longer half-life time of the catalyst and
so, a lower zirconium loading is possible.
[12] Alternatively, a similar mechanism that involves the formation
of a zirconate species by addition of PrMgCl to 3a followed by
Mg–Zr exchange, elimination of Mg alkoxide, further reaction
with another molecule of PrMgCl, and a second transmetalation
step to form 11 and 9 might also be suggested; see ref.[6].
[13] Starting enol ethers used in this work are very readily available
on a large scale by Pd-catalyzed O-vinylation of the correspond-
ing homoallyl alcohol; see: K.F.W. Hekking, F.L. van Delft,
F.P.J.T.Rutjes, Tetrahedron 2003, 59, 6751.
[1] a) New Aspects of Zirconium-Containing Organic Compounds,
Topics in Organomet. Chem., Vol. 10 (Ed.: I. Marek), Springer,
Berlin, 2005; b) Titanium and Zirconium in Organic Synthesis
(Ed.: I. Marek), Wiley-VCH, Weinheim, Germany, 2002.
[2] For some representative examples, see: a) J.Schwartz, JA. .
Labinger, Angew. Chem. 1976, 88, 402; Angew. Chem. Int. Ed.
Engl. 1976, 15, 333; b) C.A.Bertelo, J.Schwartz, J. Am. Chem.
Soc. 1976, 98, 262; c) D.W.Hart, T.F.Blackburn, J.Schwartz,
Am. Chem. Soc. 1975, 97, 679; d) D.W.Hart, J.Schwartz, J. Am.
Chem. Soc. 1974, 96, 8115.
J.
[3] E.Negishi, T.Takahashi, Acc. Chem. Res. 1994, 27, 124, and
references cited therein.
[4] Zirconocene complexes are known to catalyze polymerization
reactions; for some other interesting zirconocene-catalyzed
reactions, see: a) Y.Ikeuchi, T.Taguchi, Y.Hanzawa,
J. Org.
Chem. 2005, 70, 4354; b) J.Terao, S.A.Begum, A.Oda, N.
Kambe, Synlett 2005, 1783; c) V.Gandon, P.Bertus, J.Szymo-
niak, Eur. J. Org. Chem. 2001, 3677; d) P.Wipf, S.Ribe, Org. Lett.
2001, 3, 1503; e) Y.Ura, R.Hara, T.Takahashi, Chem. Commun.
2000, 875; f) F.Grepioni, S.Grilli, G.Martelli, D.Saoia,
J. Org.
Chem. 1999, 64, 3679; g) Y.Yamamura, M.Mori, Tetrahedron
Lett. 1999, 40, 3221; h) K.Kasai, Y.Liu, R.Hara, T.Takahashi,
Chem. Commun. 1998, 1989; i) D.B. Millward, R.M. Way-
mouth, Organometallics 1997, 16, 1153; j) G.Dawson, CA. .
Durrant, G.G. Kirk, R.J. Whitby, R.V.H. Jones, M.C.H.
Standen, Tetrahedron Lett. 1997, 38, 2335; k) N.Uesaka, M.
Mori, K.Okamura, T.Date, J. Org. Chem. 1994, 59, 4542; l) T.
Takahashi, D.Y.Kondakov, N.Suzuki, Organometallics 1994, 13,
Angew. Chem. Int. Ed. 2006, 45, 6362 –6365
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6365