Cobalt-catalyzed reductive allylation of alkyl halides with allylic acetates or carbonates

Article abstract of DOI:10.1002/anie.201104390

An efficient method for the direct allylation of alkyl halides catalyzed by simple cobalt(II) bromide has been developed. This reaction, using a variety of substituted allylic acetates or carbonates, provides the linear product as the major product. It displays broad substrate scope and good functional group tolerance. Copyright

Full text of DOI:10.1002/anie.201104390

Communications
DOI: 10.1002/anie.201104390
Cross-Coupling
Cobalt-Catalyzed Reductive Allylation of Alkyl Halides with Allylic
Acetates or Carbonates
Xin Qian, Audrey Auffrant, Abdellah Felouat, and Corinne Gosmini*
Transition-metal-catalyzed allylic alkylations, using a broad
range of metal complexes, have been intensively studied
because of their potential applications in the synthesis of new
olefinic compounds in particular for total synthesis.[1] Soft
Table 1: Optimized Reaction Conditions.
nucleophiles are usually used in Pd-,[1] Mo-,[2] Ir-,[3] Ru-,[4] Rh-
[5]
,
Pt-,[6] and even Fe-catalyzed[7] allylic substitutions. Ni,[8]
Co,[9] and Cu[10] catalysts allow the use of hard nucleophiles
such as alkylzinc or Grignard reagents, but limited functional
group compatibility and/or poor regioselectivity can be
observed if the system is not designed carefully. To avoid
handling the air- and moisture-sensitive organomagnesium
and organozinc reagents, straightforward procedures, which
do not require organometallic reagents, are highly desirable
and many have now been developed.[11] To the best of our
knowledge, direct transition-metal-catalyzed alkyl–allyl
cross-couplings using in situ generated catalytic organome-
tallic reagents are still unknown. However, a few years ago,
we reported a related Co-catalyzed coupling reaction of aryl
halides with allylic acetates;[12] these reactions in the presence
of an appropriate reducing reagent, gave allylaromatic
compounds. Such allylic carboxylates, whilst less reactive
than allyl halides, are much more environmentally friendly.
Given our previous experience with the direct Co-
catalyzed functionalization, including alkylation,[11c] of aryl
halides[13] we were interested to take the chemistry further,
and herein we report a new and general method for direct
reductive cross-coupling of allylic acetates with alkyl halides
using a CoBr2/Mn system with an acetonitrile/pyridine solvent
mixture. The approach accommodates a variety of simple and
functionalized alkyl halides and substituted allylic compounds
and is experimentally straightforward. Indeed it uses off-the-
shelf reagents without any particular precautions against air
and moisture. First, we investigated the use of the readily
available yet poorly reactive ethyl 4-bromobutanoate with a
simple allyl acetate as the electrophile. The major challenge
here lies in promoting cross-coupling rather than the for-
mation of reduction and homocoupling products. A combi-
nation of factors enabled us to overcome these difficulties
(Table 1). The reaction conditions we established as standard
afforded an excellent yield within 3 hours (Table 1, entry 1).
A 5 mol% catalyst loading gave the same result but over a
period of 16 hours, (Table 1, entry 2) and a 20 mol% CoBr2
loading accelerated the reaction (2 hours) but gave a higher
Entry
Deviation from Standard Conditions
Yield [%][a]
1
2
3
4
5
6
7
8
None
90
CoBr2 (5 mol%)
CoBr2 (20 mol%)
Co(acac)2 (10 mol%)
Mn (1.9 equiv)
Zn (3.8 equiv) instead of Mn
DMF instead of CH3CN
Allyl acetate (1 equiv)
No pyridine
pyridine (2 mL)
Bipyridine instead of pyridine
PPh3 instead of pyridine
No TFA
43 (91)[b]
77
–
17[a]
–
trace
47 (67)[c]
27
9
10
11
12
13
14
15
67
18[b]
trace[b]
–
358C
508C
trace
43[a]
[a] Yields were calculated by GC analysis using dodecane as an internal
standard. [b] The reaction time is 16 h. After 16 h, there may be still some
starting material. [c] Used 1.1 equiv of allyl acetate. acac=acetoaceto-
nate, DMF=N,N’-dimethylformamide, TFA=trifluoroacetic acid
quantity of the alkyl dimer according to GC analysis (Table 1,
entry 3). Co(acac)2 showed no catalytic activity (Table 1,
entry 4). Reducing the amount of Mn dust decreased the
reaction rate and the yield (Table 1, entry 5), while replacing
Mn by Zn dust resulted in no formation of cross-coupling
product (Table 1, entry 6). Equally, no cross-coupling product
was detected upon changing CH3CN for DMF (Table 1,
entry 7). An excess of the allyl acetate was required to drive
the reaction to completion because of the formation of a p-
allyl Co complex (Table 1, entry 8); the pyridine appears to be
important in stabilizing the low-valent Co intermediate
because cross-coupling yields decreased in its absence
(Table 1, entry 9). Replacing pyridine by bipyridine or
triphenylphosphine gave poor yields, with more than 50%
alkyl halide remaining unconsumed (Table 1, entries 11 and
12). The Co/Mn system requires activation by trifluoroacetic
acid (TFA) for the formation of the low-valent Co inter-
mediate, and attempts to run the reaction in the absence of
this activator gave no cross-coupling product (Table 1,
entry 13). At 358C, almost no reaction occurred (Table 1,
entry 14) and conversion remained low at 508C, with the alkyl
halide being only partially consumed even after 16 h (Table 1,
entry 15).
[*] X. Qian, Dr. A. Auffrant, A. Felouat, Dr. C. Gosmini
Laboratoire Hꢀtꢀroꢀlꢀments et Coordination, Ecole Polytechnique
CNRS, 91128 Palaiseau Cedex (France)
E-mail: corinne.gosmini@polytechnique.edu
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201104390.
10402
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 10402 –10405

Other substrates were tested in place of allylic acetate,
including allyl alcohol, alkyl acetates (benzyl acetate
included), aryl acetates/carbonates, and prop-2-yn-1-yl ace-
tate. Unfortunately, no reaction was observed in any of these
cases.
With these results in hand, we first screened various alkyl
halides with allyl acetate. The results reported in Table 2 and
Table 3 demonstrate that the reaction has a good functional-
group tolerance. Functional groups such as nitriles, esters, a
dioxane, a carbamate, and chlorine are nicely tolerated
(Table 2 and Table 3)[14] and the reaction also proceeded
well with long-chain alkyl halides (Table 2, entries 5 and 6).
The coupling of secondary alkyl bromides (either cyclic or
acyclic) was achieved in high yield (Table 2, entries 7–9) and
even the tertiary alkyl bromide 1j afforded the product 3j in
moderate quantities (Table 2, entry 10). Generally, the reac-
tion reaches completion within 4–6 h, although coupling with
tertiary alkyl halides required longer reaction times. The
results are therefore in general accord with the suggestion by
Oshima[9c] that cobalt catalysts are superior to Ni[15] and Cu[16]
for the coupling of two quaternary carbon centers.
Next we investigated the scope of allyl acetates (Table 3).
trans-Crotyl acetate (2b) coupled with primary and secondary
alkyl halides in good albeit slightly lower yields than those
obtained with unsubstituted allyl acetate (Table 3, entries 1–
3).
The formation of the isomeric a and g products from
substituted allyl acetate in allyl–alkyl cross-coupling reactions
is known to occur in the Cu-catalyzed system[10] and both
Table 3: Scope of allylic acetates.
Table 2: Scope of alkyl halides.
Entry
Allylic acetate
Alkyl-X
Product 4
(Ratio 4’/4’’)
Yield
[%][a]
Entry
Alkyl-X
Product 3
Yield
[%][a]
1
2
2b
2b
1a
1b
4a (85:15)
4b (89:11)
63
72
1
2
3
1a
1b
1c
3a
3b
3c
90
88
90
3
4
2b
2c
1g
1b
4c (95:5)
75[b]
88[b]
4d
5
2d
1b
4e (78:22)
67
6
7
2e
2e
1a
1b
Cl(CH2)4Br
1j
CN(CH2)2Cl
1k
4 f (>99:1)
4g (>99:1)
71
81
4
5
6
1d
C10H21Br
1e
3d
3e
3 f
70
8
9
2e
2e
4h (>99:1)
4i (>99:1)
77
98
75[b,c]
71[b,c]
C16H33Br
1 f
10
11
2e
2 f
1l
4j (>99:1)
4k (92:8)[c]
68
52
7
8
1g
1h
3g
3h
81[a] (85)[d]
1b
85[b]
12
13
2g
2h
1b
1b
4l
38
56
4b (12:82)
9
1i
1j
3i
3j
70[c,e,f]
14
15
2h
2i
1j
4m (15:85)
66
76
10
60[b,c,e]
1a
4n[d]
[a] Yield of the isolated product. [b] The yields were determined by
corrected GC using dodecane as an internal standard. [c] The reaction
time was 10 h. [d] Yield from the cyclohexyliodide, as calculated using
GC. [e] Mixture of alkyl–H and alkyl–alkyl. [f] The yields were determined
by 1H NMR spectroscopy.
1
[a] Yield of isolated 4 and 4’. [b] The yield was determined by H NMR
spectroscopy. [c] In this case the minor product comes from an attack at
the methyl-substituted carbon atom (e position). [d] Mixture of bis and
mono g-alkylated products bearing an acetate group in a 81:19 ratio.
Angew. Chem. Int. Ed. 2011, 50, 10402 –10405
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 10403

Communications
Table 4: Scope of allylic carbonates and alkyl halides.
products were detected in our reactions. However, the linear
product 4’ always dominated (with a minimal proportion of
78%; Table 3, entry 5). The 4’/4’’ ratio was determined by
1H NMR spectroscopy. But-3-en-2-yl acetate (2c) reacted
with the primary alkyl halide 1b (Table 3, entry 4) but not the
secondary alkyl halide 1g. More sterically hindered acetates,
such as prenyl acetate (2d), reacted with 1b and provided the
cross-coupling product in a good yield (Table 3, entry 5), but
again no coupling product was observed with secondary alkyl
bromide 1g. Excellent yields were obtained using (E)-
cinnamyl acetate (2e; Table 3, entries 6–9). The alkylation
reaction of 1-bromo-4-chlorobutane resulted in a selective
attack at the bromide site, thus affording (E)-7-chlorohept-1-
en-1-ylbenzene (4h) in good yield (Table 3, entry 8). Inter-
estingly, when a chloro group was present b to the nitrile, an
excellent yield of the cross-coupled product 4i was obtained
(Table 3, entry 9). Having observed that no cross-coupling
occurs with bromoalcohols, we employed the acylated alcohol
1l with success (Table 3, entry 10). Importantly, no branched
coupling product could be detected with the phenyl-substi-
tuted allylic acetate 2e. The conjugated allylic acetate 2 f
could also be used, although the yields were lower than those
obtained from allyl or cinnamyl acetates (Table 3, entry 11).
Next we investigated the reactivity of secondary allylic
acetates. The cyclohex-2-en-1-yl acetate (2g) reacted with
1b to give the product in poor yield; the high reactivity of the
primary alkyl halide leads to the formation of by-products
(Table 3, entry 12). The acyclic secondary allylic acetate 2h
reacted with both primary and tertiary alkyl halides to give
mainly the linear coupling product in moderate yield (Table 3,
entries 13 and 14). Unsurprisingly, double alkylation of cis-
1,4-diacetoxy-2-butene with 1a was the main reaction
observed; the reaction proceeded with retention of stereo-
chemistry to give the Z product in good yield (Table 3,
entry 15).
During the screening of alkyl halides, we found that both
the presence of electron-withdrawing substituents, such as
nitrile or ester groups, in the b position relative to the reactive
bromo functionality and the use of benzyl chloride prevented
the coupling reaction. This prompted us to employ the more
reactive series of allyl carbonates. After minor modifications
to the standard protocol allyl carbonates, including crotyl
carbonate and cinnamyl carbonate, were successfully coupled
to such halides (Table 4). In the case of trans-crotyl carbonate,
the reaction with a primary alkyl halide gave primarily the
terminal coupling product (Table 4, entry 2). Note that with
bulkier cinnamyl carbonates, only the linear product was
detected (Table 4, entry 4). The reaction also worked effi-
ciently with the secondary alkyl halide, cyclohexyl iodide
(Table 4, entry 3). However, the more-reactive a-substituted
alkyl halides, such as ethyl 2-chloro/bromo acetate, could not
be coupled with allylic carbonates. To our best of our
Entry
1
Allylic
carbonate
Alkyl-X
Product
(5’/5’’)
Yield
[%]
3a
1l
5a
95[a]
2
3
3b
3b
1m
1g
5b (87:13)[c]
4c[c] (89:11)
82[b]
70[a]
4
3d
1n
4i (>99:1)[c]
93[b]
[a] Yield determined by GC using dodecane as an internal standard.
[b] Combined yield of isolated 5’ and 5’’. [c] Ratio of linear/branched.
coupling product [Eq. (1)]. Moreover, the addition of the free
radical 2,2,6,6-tetramethylpiperine-1-oxyl (TEMPO) before
the alkyl halides inhibited the cross-coupling reaction. These
results point towards the involvment of an alkyl radical
intermediate in the activation process of the alkyl halide .
Our current mechanistic hypothesis is presented in
Scheme 1. Initial reduction of the CoII precatalyst should
furnish a catalytically active low-valent Co species. Subse-
quent oxidative addition to the allyl acetate forms an allyl Co
intermediate that is again subjected to reduction by manga-
nese dust. This allyl Co complex reacts with an alkyl halide to
give an allyl alkyl Co complex through the formation of an
alkyl radical. Then reductive elimination occurs to furnish the
cross-coupling product along with the regeneration of the
active species.
À
knowledge, very few reports deal with C C coupling of
these reactive alkyl halides.[17]
A few experiments were conducted to provide some
insight into the mechanism of this allyl–alkyl cross coupling
reaction. When bromomethylcyclopropane was reacted with
(E)-cinnamyl acetate, the ring-opened product (E)-hepta-1,6-
dien-1-ylbenzene was detected by GC as the sole cross-
Scheme 1. Postulated mechanism for the direct allylation of alkyl
halides.
10404 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 10402 –10405

[6] A. J. Blacker, M. L. Clark, M. S. Loft, J. M. J. Williams, Chem.
Commun. 1999, 913 – 914.
In summary, a new route for the direct allylation of
various alkyl halides catalyzed by simple cobalt(II) bromide
was developed. This method is very straightforward, and
environmentally friendly. It is efficient for the coupling of a
large variety of alkyl halides (primary, secondary, and
tertiary) with substituted allylic acetates and carbonates and
provides good to excellent yields with good functional group
tolerance. Moreover, in the case of substituted allyl acetates,
the reaction affords the linear product as the major or the sole
product. Both sterically hindered secondary allyl acetates and
secondary and tertiary alkyl halides are acceptable as
substrates. Further studies to extend the scope of this
method and to gain detailed insight into its mechanism are
currently underway in our laboratory.
[7] a) B. Plietker, Angew. Chem. 2006, 118, 1497 – 1501; Angew.
Chem. Int. Ed. 2006, 45, 1469 – 1473; b) M. Holzwarth, A.
Dieskau, M. Tabassam, B. Plietker, Angew. Chem. 2009, 121,
7387 – 7391; Angew. Chem. Int. Ed. 2009, 48, 7251 – 7255.
[8] a) N. Nomura, T. V. Rajan Babu, Tetrahedron Lett. 1997, 38,
1713 – 1716; b) S. Son, G. C. Fu, J. Am. Chem. Soc. 2008, 130,
2756 – 2757.
[9] a) C. Kishan Reddy, P. Knochel, Angew. Chem. 1996, 108, 1812 –
1813; Angew. Chem. Int. Ed. Engl. 1996, 35, 1700 – 1701; b) T.
Tsuji, H. Yorimitsu, K. Oshima, Angew. Chem. 2002, 114, 4311 –
4313; Angew. Chem. Int. Ed. 2002, 41, 4137 – 4139; c) H.
Ohmiya, T. Tsuji, H. Yorimitsu, K. Oshima, Chem. Eur. J.
2004, 10, 5640 – 5648.
[10] a) C. A. Falciola, K. Tissot-Croset, A. Alexakis, Angew. Chem.
2006, 118, 6141 – 6144; Angew. Chem. Int. Ed. 2006, 45, 5995 –
˘
5998; b) E. Erdik, M. KoÅoglu, Tetrahedron Lett. 2007, 48, 4211 –
Received: June 24, 2011
4214; c) M. Sai, H. Yorimitsu, K. Oshima, Bull. Chem. Soc. Jpn.
2009, 82, 1194 – 1196; d) A. M. Lauer, F. Mahmud, J. Wu, J. Am.
Chem. Soc. 2011, 133, 9119 – 9123.
Published online: September 14, 2011
À
Keywords: alkyl halides · allylation · C C coupling · cobalt ·
.
[11] a) M. Amatore, C. Gosmini, Synlett 2009, 1073 – 1076; b) A.
Moncomble, P. L. Floch, C. Gosmini, Chem. Eur. J. 2009, 15,
4770 – 4774; c) M. Amatore, C. Gosmini, Chem. Eur. J. 2010, 16,
5848 – 5852; d) D. A. Everson, R. Shrestha, D. J. Weix, J. Am.
Chem. Soc. 2010, 132, 920 – 921; e) M. R. Prinsell, D. A. Everson,
D. J. Weix, Chem. Commun. 2010, 46, 5743 – 5745; f) X. Yu, T.
Yang, S. Wang, H. Xu, H. Gong, Org. Lett. 2011, 13, 2138 – 2141;
for stoichiometric reactions, see g) D. C. Billington, Chem. Soc.
Rev. 1985, 14, 93 – 120; M. Durandetti, J. Y. Nedelec, J. Pꢀrichon,
J. Org. Chem. 1996, 61, 1748 – 1755.
[12] P. Gomes, C. Gosmini, J. Pꢀrichon, Org. Lett. 2003, 5, 1043 –
1045.
[13] C. Gosmini, A. Moncomble, Isr. J. Chem. 2010, 50, 568 – 576.
[14] No coupling product was isolated from the reaction of a
substrate containing a ketone.
[15] a) A. Joshi-Pangu, C.-Y. Wang, M. R. Biscoe, J. Am. Chem. Soc.
2011, 133, 8478 – 8481; b) C. Lohre, T. Drçge, C. Wang, F.
Glorius, Chem. Eur. J. 2011, 17, 6052 – 6055.
[16] L. Hintermann, L. Xiao, A. Labonne, Angew. Chem. 2008, 120,
8370 – 8374; Angew. Chem. Int. Ed. 2008, 47, 8246 – 8250.
[17] M. Durandetti, C. Gosmini, J. Pꢀrichon, Tetrahedron 2007, 63,
1146 – 1153.
homogeneous catalysis
[1] a) B. M. Trost, M. L. Crawley, Chem. Rev. 2003, 103, 2921 – 2943;
b) Z. Lu, S. Ma, Angew. Chem. 2008, 120, 264 – 303; Angew.
Chem. Int. Ed. 2008, 47, 258 – 297; c) R. Jana, T. P. Pathak, M. S.
Sigman, Chem. Rev. 2011, 111, 1417 – 1492.
[2] a) B. M. Trost, M. H. Hung, J. Am. Chem. Soc. 1983, 105, 7757 –
7759; b) B. M. Trost, G. B. Tometzki, M. H. Hung, J. Am. Chem.
Soc. 1987, 109, 2176 – 2177; c) G. C. Lloyd-Jones, A. Pfaltz,
Angew. Chem. 1995, 107, 534 – 536; Angew. Chem. Int. Ed. Engl.
1995, 34, 462 – 464; d) B. M. Trost, Y. Zhang, J. Am. Chem. Soc.
2007, 129, 14548 – 14549.
[3] a) B. Bartels, G. Helmchen, Chem. Commun. 1999, 741 – 742;
b) R. Takeuchi, N. Shiga, Org. Lett. 1999, 1, 265 – 267.
[4] a) Y. Matsushima, K. Onitsuka, T. Kondo, T. Mitsudo, S.
Takahashi, J. Am. Chem. Soc. 2001, 123, 10405 – 10406; b) K.
Onitsuka, Y. Matsushima, S. Takahashi, Organometallics 2005,
24, 6472 – 6474.
[5] T. Hayashi, A. Okada, T. Suzuka, M. Kawatsura, Org. Lett. 2003,
5, 1713 – 1715.
Angew. Chem. Int. Ed. 2011, 50, 10402 –10405
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 10405