An efficient cobalt(I)-catalysed reformatsky reaction using α-chloro esters

Article abstract of DOI:10.1002/adsc.200600378

An efficient cobalt(I)-catalysed Reformatsky reaction using α-chloro esters has been developed. The catalyst is prepared by reducing the cobalt(II) chloride (5%)/1,2-bis(diphenylphosphino)ethane (dppe)(5%)/zinc iodide (10%) system with zinc metal in acetonitrile in the presence of both the α-chloro ester and the carbonyl compound; good to excellent conversions to β-hydroxy esters are obtained at room temperature in 2.5 h.

Full text of DOI:10.1002/adsc.200600378

UPDATES
DOI: 10.1002/adsc.200600378
An Efficient Cobalt(I)-Catalysed Reformatsky Reaction using a-
Chloro Esters
Marco Lombardo,a,* Alessandra Gualandi,a Filippo Pasi,a and Claudio Trombinia,*
a
Dipartimento di Chimica “G. Ciamician”, Università degli Studi di Bologna,
via Selmi 2, 40126, Bologna, Italy
Fax : (+39)-051-209-9456; e-mail: marco.lombardo@unibo.it or claudio.trombini@unibo.it
Received: July 26, 2006
Dedicated to Professor Vincenzo Balzani on the occasion of his 70th birthday.
Abstract: An efficient cobalt(I)-catalysed Refor-
matsky reaction using a-chloro esters has been de-
veloped. The catalyst is prepared by reducing the
cobalt(II) chloride (5%)/1,2-bis(diphenylphosphi-
able induction time, while the adoption of activation
techniques on the zinc surface resulted in clean reac-
tions at or below room temperature. Several of the
techniques to activate the zinc surface proposed so far
have been recently reviewed;[1] they include i) the
chemical activation of commercial zinc powder with
acidic or oxidising agents, ii) the physical activation of
the metal by means of ultrasound and microwaves, iii)
the in situ production of highly reactive metal species
by the reduction of zinc halides with alkali metals or
potassium-graphite, iv) and the use of electrochemical
techniques. Other low-valent metals have been also
exploited in the Reformatsky reaction instead of zinc;
Orsini proposed cobalt(0)-phosphine complexes to
achieve Reformatsky-like reactions,[3] while a very
recent example based on the use of the FeBr2/Mn
couple was reported by PØrichon.[4]
no)ethane (dppe)ACHTRE(UGN 5%)/zinc iodide (10%) system
with zinc metal in acetonitrile in the presence of
both the a-chloro ester and the carbonyl com-
pound; good to excellent conversions to b-hydroxy
esters are obtained at room temperature in 2.5 h.
Keywords: catalysis; a-chloro esters; cobalt(I); b-
hydroxy esters; Reformatsky reaction
Introduction
Alternatively, a low-valent metal was used to per-
The Reformatsky reaction is the zinc-promoted addi- form the oxidative addition of the a-halo ester to first
tion of a-halo esters 1 to carbonyl compounds 2 to afford a metal enolate that reacted with a zinc(II)
give b-hydroxy esters 3 (Scheme 1).[1]
species (e.g., zinc halide or dialkylzinc) to produce a
Despite its usefulness as a complement of the zinc enolate through an alternative reaction chan-
Grignard reaction, which cannot be applied to a-halo nel.[5]
esters, nitriles, amides, etc., since its discovery in
Following this approach, here we report a very effi-
1887[2] the Reformatsky reaction presented several cient Reformastsky reaction protocol, catalysed by an
problems in terms of yields, reproducibility, and com- in situ formed cobalt(I) species and using a-chloro
peting reactions, such as ester or aldehyde self-con- esters, without appreciable side-product formation.
densation or adduct dehydration. The metal was early
identified as the main factor responsible for the some-
what erratic performance of Reformatsky reactions Results and Discussion
which, when carried out with partially passivated zinc,
required long heating in aromatic solvents with a vari- The chemistry of Co(I) directed to the construction of
carbon-carbon bonds is recently enjoying an interest-
ing revival.
Significant examples were proposed by Hilt who
exploited the CoACHTRE[UNG 1,2-bis-(diphenylphosphino)ethane
(dppe)]Br2 (1.5%)/Bu4NBH4/ZnI2 system in a 1:1:3
ratio to catalyse the [4+2] homo-Diels–Alder cyclo-
[6]
addition of acyclic 1,3-dienes with alkynes in CH2Cl2
Scheme 1.
and the redox system CoACHRTE(UNG dppe)Br2 (10%)/Zn(0)/ZnI2
Adv. Synth. Catal. 2007, 349, 465 – 468
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
465

Marco Lombardo et al.
UPDATES
Table 1. The role of ZnI2 in the reaction of 6 and benzalde-
in the 1:5:5 ratio to catalyse the Diels–Alder reaction
of propargylic amines with butadienes in CH2Cl2,[7] as
well as the cyclotrimerisation of alkynes to polysusb-
stituted benzenes.[8] Snyder exploited an analogous
route to Co(I) using the CoI2 (5%)/dppe/Zn(0)/ZnI2
system in the 1:1:1:3 ratio to catalyse the [4+2+2]
homo-Diels–Alder cycloaddition of substituted nor-
bornadienes with butadiene at room temperature in
CH2Cl2.[9]
hyde.[a]
Run
ZnI2 (%)
8a, Yield [%][b]
1
2
3
0
5
10
10
75
93
[a]
CoCl2 (5%)/dppe (5%)/Zn(0) (1.5 equivs.) in CH3CN at
208C for 2.5 h.
Recent studies by PØrichon and Gosmini[10] demon-
strated that Co(I) species, generated in acetonitrile
using the Co(II)/Zn(0) redox couple, insert into the
carbon-halogen bond of aryl, vinyl and allyl halides;
through the intermediacy of allyl- and arylzinc com-
plexes, and useful carbon-carbon bond forming reac-
tion protocols were developed. Exploiting the process
developed by Gosmini on allyl acetate, we recently
proposed a simple Co(I)-catalysed homoaldol reac-
tion of allylidene dipivalate 4 with aldehydes leading
to configurationally pure (Z)-4-hydroxy-alk-1-en-1-yl
pivalates 5 (Scheme 2).[11]
[b]
Isolated yields after purification by flash chromatography
on silica.
Table 2. Effect of different ligands in the Co(I)-catalysed
Reformatsky reaction of 6 with benzaldehyde.[a]
Run
Ligand (5%)
8a, Yield [%][b]
1
2
3
4
-
17
<5
93
95
2,2’-bipyridyl
dppe
dppp[c]
Inspired by the foregoing cobalt chemistry, we
started searching for a practical way to catalyse the
otherwise very sluggish Reformatsky reaction of a-
chloro esters, using an in situ prepared Co(I) species.
An optimisation study was carried out on the ethyl a-
chloroacetate (6)/benzaldehyde system (7a, R=Ph),
and, after a number of combinations of experimental
conditions had been tested (Scheme 3), we realised
that the reaction occurred in more than 90% yield
[a]
CoCl2 (5%)/ligand (5%)/ZnI2 (10%)/Zn(0) (1.5 equivs.)
in CH3CN at 208C for 2.5 h.
Isolated yields after purification by flash chromatography
on silica.
[b]
[c]
1,3-Bis(diphenylphosphino)propane.
using the CoCl2 (5%)/ACHTRE(UNG dppe)/ZnI2 system in the 1:1:2
ratio, and 1.5 equivalents of commercial zinc powder
in CH3CN. It is worth noticing that the reaction did
not proceed at all in the absence of cobalt salts.
Scheme 4.
Among all the experiments performed in the opti-
misation stage, Table 1 and Table 2 collect those
aiming at verifying the importance of ZnI2 and of sup- known to give the ion pair [Co2+][ZnX4]2À,[10a] much
ACHTREUNG
plementary ligands for cobalt, respectively. In all the more easily reduced by Zn metal to a transient Co(I),
runs of Table 1 and Table 2 temperature and reaction whose lifetime is increased by the presence of ZnX2
time were set at 208C and 2.5 h, respectively. The role and of supplementary ligands, such as the acetonitrile
of ZnI2 in favouring the catalyst formation is apparent itself, as well as phosphines, diimines, etc.
(Table 1); the interaction of CoX2 with ZnX2 is
The dramatic effect of a supplementary diphos-
phine was unambiguously confirmed by results shown
in Table 2. 1,4-Bis(diphenylphosphino)butane (dppb)
was also tested, but being insoluble in CH3CN, the
same result as for run 1 was observed.
Once having established the optimum conditions
for the coupling of ethyl chloroacetate (6) and benzal-
dehyde, we checked the same reaction protocol with a
series of representative aromatic and aliphatic alde-
hydes and ketones (Scheme 4). In the results collected
in Table 3, half of the runs presented chemical yields
ꢀ90%. The lowest yield (run 8) was due to pivalalde-
hyde; a,b-unsaturated aldehydes (runs 9 and 10) af-
forded yields in the 70–75% range. In all the experi-
ments, dehydration was the only side reaction, howev-
er its extent was variable and always limited to less
Scheme 2.
Scheme 3.
466
www.asc.wiley-vch.de
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 465 – 468

Cobalt(I)-Catalysed Reformatsky Reaction using a-Chloro Esters
UPDATES
Table 3. Co(I)-catalysed Reformatsky reaction of ethyl
chloroacetate and carbonyl compounds, using CoCl2 (5%)/
dppe (5%)/ZnI2 (10%)/Zn(0) (1.5 equivs.) in CH3CN at
208C for 2.5 h.
Run
R1R2C=O
8, Yield [%][a]
1
2
3
4
5
6
7
8
Benzaldehyde
8a, 93
8b, 98
8c, 90
8d, 79
8e, 66
8f, 94
8g, 93
8h, 59
8i, 69
8j, 76
8j, 85
8k, 95
4-Chlorobenzaldehyde
4-Methoxybenzaldehyde
4-Fluorobenzaldehyde
2-Naphthaldeyde
Cyclohexanecarboxaldehyde
Octanal
Pivalaldehyde
9
(E)-Crotonaldehyde
(E)-Cinnamaldehyde
Cyclohexanone
Scheme 6.
10
11
12
Acetophenone
[a]
A plausible synthetic reaction mechanism is shown
in Scheme 6. Complex 11 is a simplified picture of the
catalytically active Co(I) species – acetonitrile mole-
cules complete the coordination shell of Co – which
insert into the carbon-chlorine bond of 6 to give the
Isolated yields after purification by flash chromatography
on silica.
CoACHTRE(UNG III) enolate 12. Transmetallation by the zinc hal-
ides present in the reaction mixture frees a classical
Reformatsky reagent and the catalyst precursor 13
which, finally, is reduced to 11 by zinc. Actually, we
cannot rule out a slightly modified catalytic cycle
where the CoACTHRE(UNG III) enolate 12 undergoes reduction to
Scheme 5.
a Co(II) enolate in the presence of zinc, leaving after
the transmetallation step a Co(II) precatalyst species
analogous to 13.
than 5%. Pinacol coupling products of the carbonyl
compounds were never observed, while using metals Conclusions
with a higher reduction potential (e.g., manganese)
pinacolisation intereferes with the Reformatsky reac- We have reported a cobalt-catalysed modification of
tion.[4]
the Reformatsky reaction, affording an efficient cross
The reaction was then applied to a-chloro esters 9a coupling of a-chloro esters and carbonyl compounds
and 9b with model aldehydes under the same reaction to give b-hydroxy esters in high isolated yields.
conditions applied in Table 3 (Scheme 5 and Table 4). Making a-chloro esters competitive to a-bromo esters
Good to excellent yields were again observed, while, in Reformatsky reactions is advantageous both under
as generally observed in Reformatsky reactions, zinc economic (the prices of a-chloro esters are from 50%
enolates do not display interesting diastereoselectivi- to two-fold lower than those of their brominated
ties when a prochiral enolate reacts with a prochiral counterparts) and toxicological terms. a-Bromo esters
carbonyl compound.
are, indeed, classified as much more toxic and can-
Table 4. Co(I)-catalysed Reformatsky with substituted a-chloro esters, using CoCl2 (5%)/dppe (5%)/ZnI2 (10%)/Zn(0)
(1.5 equivs.) in CH3CN at 208C for 2.5 h.
Run
9
RCHO
10, Yield [%][a]
10, syn/anti
1
2
3
4
9a
9a
9b
9b
Benzaldehyde
Cyclohexanecarboxaldehyde
Benzaldehyde
10a, 95
10b, 80
10c, 96
10d, 98
56:44
55:45
60:40
51:49
Cyclohexanecarboxaldehyde
[a]
Isolated yields after purification by flash chromatography on silica.
Adv. Synth. Catal. 2007, 349, 465 – 468
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
467

Marco Lombardo et al.
UPDATES
References
cerogenic agents than their chloro counterparts owing
to their higher reactivity as alkylating agents.
[1] M. Lombardo, C. Trombini, The Chemistry of Zinc
Enolates, in: The Chemistry of Organozinc Componds,
Part II, (Eds.: Z. Rappoport, I. Marek), John Wiley
and Sons, Chichester, 2006, pp. 797–861.
Experimental Section
[2] a) S. N. Reformatsky, Ber. dtsch. chem. Ges. 1887, 20,
1210; b) R. L. Shriner, Org. React. 1942, 1, 1;
c) D. G. M. Diaper, A. Kuksis, Chem. Rev. 1959, 59, 89.
[3] F. Orsini, G. Sello, Curr. Org. Synth. 2004, 1, 111.
[4] M. Durandetti, J. PØrichon, Synthesis 2006, 1542.
[5] a) K. Kanai, H. Wakabayashi, T. Honda, Org. Lett.
2000, 2 , 2549; b) J. C. Adrian, Jr., M. L. Snapper, J.
Org. Chem. 2003, 68, 2143; c) P. G. Cozzi, E. Rivalta,
Angew. Chem. Int. Ed. 2005 44, 3600.
[6] a) G. Hilt, F.-X. du Mesnil, Tetrahedron Lett. 2000, 41,
6757; b) G. Hilt, K. I. Smolko Angew. Chem. Int. Ed.
2003, 42, 2795; c) G. Hilt, J. Janikowski, W. Hess
Angew. Chem. Int. Ed. 2006 45, 5204.
Synthesis of Ethyl 3-Hydroxy-3-phenylpropanoate
(8a, Table 1, Entry1); General Procedure
Commercial CoCl2·6H2O (0.012 g, 0.05 mmol) is flamed
under a positive argon pressure until the colour turns from
pink to bright blue. Anhydrous ZnI2 (0.032 g, 0.10 mmol) is
added and again the flask is flamed under a positive argon
pressure. The salts are dissolved in freshly distilled acetoni-
trile (2 mL) and to this solution dppe (0.020 g, 0.05 mmol),
ethyl chloroacetate 6 (0.127 mL, 1.2 mmol) and benzalde-
hyde (0.102 mL, 1 mmol) are added. The temperature is set
at 208C with a thermostated bath, and commercial zinc
powder (0.98 g, 1.5 mmol) is added. The reaction mixture is
vigorously stirred at the same temperature for 2.5 h,
quenched with saturated NH4Cl aqueous solution (1 mL)
and filtered on a short pad of Celiteꢁ. The filtered solution
is dried on Na2SO4, evaporated under reduced pressure and
the residue is purified by flash-chromatography on silica
eluting with cyclohexane/ethyl acetate 8:2 to afford pure 8a
as an oil; yield: 0.181 g (93%).
[7] G. Hilt, F. Galbiati, Synlett 2005, 829.
[8] a) G. Hilt, W. Hess, T. Vogler, C. Hengst, J. Organomet.
Chem. 2005, 690, 5170; b) G. Hilt, T. Vogler, W. Hess,
F. Galbiati, Chem. Commun. 2005, 1474.
[9] a) B. Ma, J. K. Snyder, Organometallics 2002, 21, 4688;
b) B. Ma, J. K. Snyder, Tetrahedron Lett. 2005, 46, 703.
[10] a) S. Seka, O. Buriez, J. PØrichon, Chem. Eur. J. 2003,
9, 3597; b) M. Amatore, C. Gosmini, J. PØrichon, Eur.
J. Org. Chem. 2005, 989; c) C. Gosmini, J. PØrichon,
Org. Biomol. Chem. 2005, 3, 216, and references cited
therein; d) C. Gosmini, M. Amatore, S. Claudel, J. PØri-
chon, Synlett 2005, 2171; e) L. Polleux, E. Labbe, O.
Buriez, J. PØrichon, Chem. Eur. J. 2005, 11, 4678; f) I.
Kazmierski, C. Gosmini, J.-M. Paris, J. PØrichon, Synlett
2006, 881; g) M. Amatore, C. Gosmini, J. PØrichon, J.
Org. Chem. 2006, 71, 6130.
Acknowledgements
This work was supported by MIUR(Rome) PRIN project
grant: “Sintesi e Stereocontrollo di Molecole Organiche per lo
Sviluppo di Metodologie Innovative di Interesse Applicativo”.
[11] M. Lombardo, S. Licciulli, F. Pasi, G. Angelici, C.
Trombini, Adv. Synth. Catal. 2005, 347, 2015.
468
www.asc.wiley-vch.de
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 465 – 468