Mild Cobalt-Catalyzed Negishi Cross-Couplings of (Hetero)arylzinc Reagents with (Hetero)aryl Halides

Article abstract of DOI:10.1002/anie.201510665

A catalytic system consisting of CoCl₂?2 LiCl (5 mol %) and HCO₂Na (50 mol %) enables the cross-coupling of various N-heterocyclic chlorides and bromides as well as aromatic halogenated ketones with various electron-rich and -poor ar

Full text of DOI:10.1002/anie.201510665

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201510665
German Edition: DOI: 10.1002/ange.201510665
Cross-Coupling
Mild Cobalt-Catalyzed Negishi Cross-Couplings of (Hetero)arylzinc
Reagents with (Hetero)aryl Halides
Diana Haas+, Jeffrey M. Hammann+, Ferdinand H. Lutter, and Paul Knochel*
Abstract: A catalytic system consisting of CoCl2·2LiCl
(5 mol%) and HCO2Na (50 mol%) enables the cross-cou-
pling of various N-heterocyclic chlorides and bromides as well
as aromatic halogenated ketones with various electron-rich and
-poor arylzinc reagents. The reactions reached full conversion
within a few hours at 258C.
2-bromoquinoline (1a) with para-anisylzinc chloride (2a,
1.2 equiv) in the presence of CoCl2·2LiCl (5 mol%) and
observed the formation of the desired cross-coupling product
(3a) in 65% yield. However, aside from the desired cross-
coupling, we observed extensive side reactions, including
homocoupling. To improve the reaction outcome, we were
inspired by the recent work of Miller and co-workers, who
showed that the addition of potassium formate plays an
important role in Suzuki reactions.[13] We anticipated that this
salt could generate a more selective cobalt species with equal
or superior catalytic activity.
To our delight, the addition of HCO2Na (50 mol%) led to
an improved yield of 88% of the isolated product (Scheme 1).
Preliminary kinetic studies showed that the main effect of
HCO2Na is to considerably reduce the occurrence of side
reactions, thus, as anticipated, leading to more selective cross-
T
ransition-metal-catalyzed cross-coupling reactions are val-
2
2
À
uable approaches for the formation of C(sp ) C(sp ) bonds
and are of great interest for the synthesis of biologically active
molecules.[1] The Negishi cross-coupling, in particular, has
attracted a lot of attention as a wide range of polyfunctional
zinc reagents are available,[2] and transmetalations with
transition-metal catalysts are fast and efficient. Mostly Pd
and Ni catalysts have been used to perform such
C(sp ) C(sp ) cross-couplings; however, price[3] and toxicity[4]
issues have encouraged the search for alternative metal
sources, such as cobalt[5,6] and iron.[7] Bedford and co-workers
demonstrated that the use of iron(I) complexes in Negishi
cross-couplings leads to great efficiency.[8] Gosmini and
BØgouin showed that organozinc reagents can be generated
in situ and cross-coupled with heteroaryl halides in a one-pot
procedure.[9] Yoshikai and co-workers reported impressive
organometallic cascade reactions, where arylzinc reagents
were generated by Co catalysis and then underwent cross-
coupling with aryl iodides in the presence of a Pd catalyst.[10]
Furthermore, Hayashi and co-workers reported cobalt-cata-
lyzed asymmetric C(sp ) C(sp) couplings.
Recently, we showed that arylzinc reagents that are
prepared by directed metalation undergo smooth
C(sp ) C(sp ) cross-couplings with primary and secondary
alkyl halides.[12] Whereas this reaction proceeds under mild
conditions, it suffers from a limited scope with respect to the
arylzinc reagent, and the extension of this approach to
C(sp ) C(sp ) couplings under the reported reaction condi-
tions was difficult.
2
2
À
Scheme 1. Cobalt-catalyzed cross-coupling of 2-bromoquinoline (1a)
and para-anisylzinc chloride (2a) with and without sodium formate.
2
[11]
À
2
3
couplings.[14] This effect proved to be general, and a broader
À
screen of reaction conditions using 2-bromopyridine (1b)
[6c]
showed that cobalt halides,[6f] Co(acac)2, and Co(acac)3
gave good results (Table 1, entries 1–5). In particular,
CoCl2·2LiCl,[15] which is conveniently soluble in THF,
afforded product 3b in excellent yield in the presence of
HCO2Na (entry 7). Interestingly, when sodium pivalate
(tBuCO2Na = PivONa) was used as an additive, the reaction
was equally efficient, showing that HCO2Na does not act as
a reducing agent, but rather as a ligand.[16] Control experi-
ments using ultrapure CoCl2 (99.99%) confirmed that the Co
salts are the active catalysts and that metal impurities do not
play a role (compare with entry 2). Polar solvents, such as
N,N’-dimethylpropylene urea (DMPU),[6b] or the use of
a typical additive, such as N,N,N’,N’-tetramethylethylenedi-
amine (TMEDA),[15] did not improve the reaction (entries 9
and 10). The use of Co(acac)3 instead of CoCl2·2LiCl was not
advantageous (entry 11). Furthermore, we confirmed that
HCO2Na alone did not catalyze this coupling by additional
metal impurities (entry 12). Additional control experiments
2
2
À
Herein, we report a new set of reaction conditions that
enables smooth cross-couplings of various arylzinc reagents
with (hetero)aryl chlorides or bromides within a few hours at
room temperature. In a preliminary experiment, we treated
[*] M. Sc. D. Haas,[+] M. Sc. J. M. Hammann,[+] B. Sc. F. H. Lutter,
Prof. Dr. P. Knochel
Department of Chemistry
Ludwig-Maximilians-Universität München
Butenandtstrasse 5–13, Haus F, 81377 Munich (Germany)
E-mail: paul.knochel@cup.uni-muenchen.de
[+] These authors contributed equally to this work.
Supporting information for this article can be found under http://dx.
doi.org/10.1002/anie.201510665.
Angew. Chem. Int. Ed. 2016, 55, 3809 –3812
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3809

Angewandte
Communications
Chemie
Table 1: Optimization of the reaction conditions for the cobalt-catalyzed
Table 2: Cobalt-catalyzed cross-coupling reactions between 2-chlori-
nated aromatic ketones and arylzinc reagents.
2
2
À
C(sp ) C(sp ) cross-coupling of 2-bromopyridine (1b).
Entry
1
Electrophile
Zinc reagent
Product[a]
Entry
Catalyst
CoCl2
Additive
Homocoupling [%]
Yield [%][b]
1c (X=Cl)
1d (X=Br)
2b
4a: 65% (X=Cl)
4b: 74% (X=Br)
1
2
–
–
14
24
82
76
[c]
CoCl2
3
CoBr2
–
14
78
4
5
6
7
8
9
10
11
12
13
14
Co(acac)2
Co(acac)3
CoCl2·2LiCl
CoCl2·2LiCl
CoCl2·2LiCl
CoCl2·2LiCl
CoCl2·2LiCl
Co(acac)3
–
–
–
–
10
14
13
12
14
12
7
16
72
2
3
4
80
83 (79)[d]
HCO2Na
PivONa
DMPU
TMEDA
HCO2Na
HCO2Na
–
94 (87)[d]
1c
1e
2c
2d
4c: 67%
4d: 73%
96
84
51
89
–
–
CrCl2
FeCl2
traces
traces
traces
17
–
[a] MgCl2 and LiCl were omitted for the sake of clarity. 5 mol% of the
indicated catalyst and 50 mol% of the indicated additive were used.
[b] Determined by GC analysis using undecane (C11H24) as an internal
standard. [c] 99.99% purity. [d] Yield of isolated product.
1e
1 f
2e
2a
2 f
2g
4e: 98%
4 f: 76%
4g: 68%
4h: 89%
5
6
7
indicated that Cr[17] and Fe[18] salts are not good catalysts for
this reaction (entries 13 and 14).
The reaction scope of this cross-coupling proved to be
quite broad. Thus, 2-halogenated, functionalized aceto- and
benzophenones (1c–1j) underwent the cobalt-catalyzed
cross-coupling with a range of aryl- and heteroarylzinc
reagents (2a–2h), yielding the corresponding ketones (4a–
4k) in 61–98% yield (Table 2). 2-Chloro- and 2-bromoaceto-
phenone (1c, 1d) reacted with zinc reagents bearing various
functional groups, providing the expected products (4a–4c) in
65–74% yield (entries 1 and 2). Interestingly, zinc reagents
with a dimethylamino substituent or a cyano group (2d, 2e)
were well tolerated as they reacted with 2-chlorobenzophe-
none (1e) to give 4d and 4e in 73 and 98% yield, respectively
(entries 3 and 4). Remarkably, various other 2-chlorinated
aromatic ketones (1 f–1j) underwent the cross-coupling with
both electron-rich and -poor arylzinc reagents, yielding the
desired products in 61–89% yield (entries 5–10). Heterocyclic
zinc reagents, such as 2h, provided the new ketones 4i and 4k
in 61 and 62% yield, respectively (entries 8 and 10).
1g
1h
8
1h
1i
2h
2g
4i: 61%
4j: 73%
9
10
Furthermore, a range of 2,3-disubstituted N-heterocyclic
chlorides can also be readily employed in this reaction
(Table 3, entries 1–4). p-MeOC6H4ZnCl (2a) reacted
smoothly with ethyl 2-chloronicotinate (1k), leading to the
2,3-disubstituted pyridine 3c in 70% yield (entry 1). The
coupling of the electron-rich arylzinc reagents 2i and 2d with
methyl 2-chloronicotinate (1l) afforded pyridines 3d and 3e
in 71 and 91% yield, respectively (entries 2 and 3). Similarly,
the cross-coupling of 2-chloronicotinonitrile (1m) and an
organozinc reagent bearing a OMOM group (MOM =
methoxymethyl) led to the desired pyridine 3 f in 60% yield
(entry 4). Furthermore, 2,5-disubstituted N-heterocyclic
1i
2h
4k: 62%
[a] Yield of isolated product. TBS=tert-butyldimethylsilyl.
chlorides were also good substrates, delivering the diarylated
pyridines 3g–3l in 73–89% yield (entries 5–10).
Moreover, halogenated quinolines, pyrimidines, and tri-
azines also proved to be good substrates for this cross-
coupling (Table 4). Substituted quinolines, such as 2-bromo-
quinoline-3-carbonitrile (1q) and ethyl 2-bromoquinoline-4-
3810
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 3809 –3812

Angewandte
Communications
Chemie
Table 3: Cobalt-catalyzed cross-coupling reactions between 2-chloropyr-
Table 4: Cobalt-catalyzed cross-coupling reactions between N-heterocyclic
idines and arylzinc reagents.
halides and arylzinc reagents.
Entry
Electrophile
Zinc reagent
Product[a]
Entry
1
Electrophile
Zinc reagent
Product[a]
1
1k
2a
3c: 70%
1q
2 f
3m: 92%
2
3
4
5
2
3
1l
1l
2i
2d
2j
3d: 71%
3e: 91%
3 f: 60%
3g: 89%
1r
2a
2 f
3n: 65%
3o: 92%
1s
1m
1n
4
5
1t
2l
3p: 86%
3q: 76%
2a
6
7
1u
2c
1n
2c
3h: 88%
6
7
1v
2l
3r: 65%
3s: 61%
1o
1o
1p
1p
2b
2k
2 f
2a
3i: 86%
3j: 73%
3k: 84%
3l: 84%
8
9
1w
2m
[a] Yield of isolated product.
10
[a] Yield of isolated product. Bn=benzyl.
carboxylate (1r), can be rapidly coupled with the organozinc
reagents 2 f and 2a to provide the corresponding arylated
quinolines in 92 and 65% yield, respectively (3m and 3n;
entries 1 and 2). Pyrimidines, which are common scaffolds of
pharmaceuticals,[19] were readily obtained by the coupling of
2-chloro- or 2-bromopyrimidine (1s–1v) in 65–92% yield
(entries 3–6). Triazines are also of great importance as
building blocks for materials and agrochemicals.[20] The
cross-coupling of 2-chloro-4,6-diphenyl-1,3,5-triazine (1w)
with the arylzinc reagent 2m led to the desired product 3s
in 61% yield (entry 7). The cross-coupling of 4,6-dichloro-
pyrimidine (5) with p-MeOC6H4ZnCl (2a; 2.4 equiv)
afforded the arylated compound 6 in 73% yield on gram
scale (Scheme 2).
Scheme 2. Cobalt-catalyzed cross-coupling reaction between 4,6-
dichloropyrimidine (5) and p-MeOC6H4ZnCl (2a).
catalyst deactivation is often observed owing to chelation of
the reagents with the catalyst.[21] However, in the presence of
THF-soluble CoCl2·2LiCl (5 mol%) and sodium formate
(50 mol%), the cross-coupling of 2-bromopyrimidine (1s)
with (1-methyl-1H-indol-5-yl)zinc chloride (2h) or thiophen-
3-ylzinc chloride (2n) proceeded smoothly to afford the
heteroaryl compounds 7a and 7b in 72 and 61% yield,
respectively (Scheme 3).
In summary, we have developed a new, practical, cobalt-
2
2
À
The synthesis of heteroaryl–heteroaryl cross-coupling
products is a challenge. When Pd or Ni catalysts are used,
catalyzed, sodium formate promoted C(sp ) C(sp ) cross-
coupling reaction between N-heterocyclic chlorides and
Angew. Chem. Int. Ed. 2016, 55, 3809 –3812
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3811

Angewandte
Communications
Chemie
Ciapetti, M. Brellier, A. GuØrinot, J. Cossy, Org.
Lett. 2014, 16, 6160; i) O. M. Kuzmina, A. K.
Steib, S. Fernandez, W. Boudot, J. T. Markiewicz,
P. Knochel, Chem. Eur. J. 2015, 21, 8242; j) J. Li,
L. Ackermann, Angew. Chem. Int. Ed. 2015, 54,
8551; Angew. Chem. 2015, 127, 8671; k) M.
Moselage, N. Sauermann, S. C. Richter, L. Acker-
mann, Angew. Chem. Int. Ed. 2015, 54, 6352; l) J.
Wu, N. Yoshikai, Angew. Chem. Int. Ed. 2016, 55,
336; Angew. Chem. 2016, 128, 344.
Scheme 3. Heteroaryl–heteroaryl cross-coupling reactions between 2-bromopyrimidine
(1s) and heteroarylzinc reagents.
[7] For selected reviews on iron-catalyzed reactions,
see: a) C. Bolm, Nat. Chem. 2009, 1, 420; b) A.
Fürstner, Angew. Chem. Int. Ed. 2009, 48, 1364;
Angew. Chem. 2009, 121, 1390; c) W. M. Czaplik,
bromides as well as halogenated aromatic ketones with
ˇ
M. Mayer, J. Cvengros, A. Jacobi von Wangelin, ChemSusChem
2009, 2, 396; d) E. Nakamura, N. Yoshikai, J. Org. Chem. 2010,
75, 6061.
various (hetero)arylzinc reagents. The use of sodium formate
was the key to the success of these cobalt-catalyzed cross-
couplings. Further studies, including mechanistic investiga-
tions, are currently underway in our laboratories.
[8] For representative examples, see: a) C. J. Adams, R. B. Bedford,
E. Carter, N. J. Gower, M. F. Haddow, J. N. Harvey, M. Huwe, M.
ngeles Cartes, S. M. Mansell, C. Mendoza, D. M. Murphy, E.
Neeve, J. Nunn, J. Am. Chem. Soc. 2012, 134, 10333; b) R. B.
Bedford, E. Carter, P. M. Cogswell, N. J. Gower, M. F. Haddow,
J. N. Harvey, D. M. Murphy, E. C. Neeve, J. Nunn, Angew. Chem.
Int. Ed. 2013, 52, 1285; Angew. Chem. 2013, 125, 1323.
[9] J.-M. BØgouin, C. Gosmini, J. Org. Chem. 2009, 74, 3221.
[10] B.-H. Tan, J. Dong, N. Yoshikai, Angew. Chem. Int. Ed. 2012, 51,
9610; Angew. Chem. 2012, 124, 9748.
Acknowledgements
We cordially thank Prof. Konstantin Karaghiosoff for fruitful
discussions. We would also like to thank the DFG (SFB749,
B2) for financial support. We also thank Rockwood Lithium
GmbH for the generous gift of chemicals.
[11] a) T. Sawano, K. Ou, T. Nishimura, T. Hayashi, Chem. Commun.
2012, 48, 6106; b) T. Sawano, K. Ou, T. Nishimura, T. Hayashi, J.
Org. Chem. 2013, 78, 8986.
Keywords: cobalt · cross-coupling · heteroarenes · heterocycles ·
Negishi coupling
[12] J. M. Hammann, D. Haas, P. Knochel, Angew. Chem. Int. Ed.
2015, 54, 4478; Angew. Chem. 2015, 127, 4560.
[13] W. D. Miller, A. H. Fray, J. T. Quatroche, C. D. Sturgill, Org.
Process Res. Dev. 2007, 11, 359.
How to cite: Angew. Chem. Int. Ed. 2016, 55, 3809–3812
Angew. Chem. 2016, 128, 3873–3877
[14] See the Supporting Information for details.
[15] a) J. M. Hammann, A. K. Steib, P. Knochel, Org. Lett. 2014, 16,
6500; b) G. Cahiez, C. Chaboche, C. Duplais, A. Giulliani, A.
Moyeux, Adv. Synth. Catal. 2008, 350, 1484.
[1] a) N. Miyaura, Cross-Coupling Reactions. A Practical Guide,
Springer, Berlin, 2002; b) Metal-Catalyzed Cross-Coupling
Reactions (Eds.: F. Diederich, A. de Meijere), Wiley-VCH,
Weinheim, 2004; c) Organotransition Metal Chemistry (Ed.: J. F.
Hartwig), University Science Books, Sausalito, California, 2010.
[2] Handbook of Functionalized Organometallics: Applications in
Synthesis (Ed.: P. Knochel), Wiley-VCH, Weinheim, 2004.
[3] a) World market prices for Pd: $770 per pound; b) world market
price for Co: $13 per pound; 11.02.2015.
[4] a) Handbook on the Toxicology of Metals (Eds.: L. Friberg, G. F.
Nordberg, V. B. Vouk), Elsevier, Amsterdam, 1986; b) M. N.
Hughes, Comprehensive Coordination Chemistry, Vol. 6 (Eds.:
G. Wilkinson, R. D. Gillard, J. A. McCleverty), Pergamon Press,
Oxford, 1987, pp. 643 – 648; c) Nickel and the Skin: Absorption,
Immunology, Epidemology, and Metallurgy (Eds.: J. J. Hosty-
nek, H. I. Maibach), CRC, Boca Raton, FL, 2002.
[16] We propose that the halide ligands in CoCl2·2LiCl are replaced
by carboxylate ligands (HCO2À), which provides a more selec-
tive cobalt catalyst. In fact, halide ligands mostly lead to
tetrahedral coordination (see, for example: N. S. Gill, F. B.
Taylor, Inorg. Synth. 1967, 9, 136), whereas carboxylate ligands
lead to an octahedral coordination (see, for example: a) A.
Kaufmann, C. Afshar, M. Rossi, D. E. Zacharias, J. P. Glusker,
Struct. Chem. 1993, 4, 191; b) G. Aromí, A. S. Batsanov, P.
Christian, M. Helliwell, A. Parkin, S. Parsons, A. A. Smith, G. A.
Timco, R. E. P. Winpenny, Chem. Eur. J. 2003, 9, 5142).
Furthermore, whereas CoCl2·2LiCl in THF is a blue solution,
the addition of HCO2Na leads to a pink solution, which is typical
for a coordination change.
[17] A. K. Steib, O. M. Kuzmina, S. Fernandez, D. Flubacher, P.
Knochel, J. Am. Chem. Soc. 2013, 135, 15346.
[5] For selected reviews on cobalt-catalyzed reactions, see a) C.
Gosmini, J.-M. BØgouin, A. Moncomble, Chem. Commun. 2008,
3221; b) G. Cahiez, A. Moyeux, Chem. Rev. 2010, 110, 1435.
[6] For representative examples, see: a) H. Ohmiya, H. Yorimitsu,
K. Oshima, Chem. Lett. 2004, 33, 1240; b) T. J. Korn, P. Knochel,
Angew. Chem. Int. Ed. 2005, 44, 2947; Angew. Chem. 2005, 117,
3007; c) T. Hatakeyama, S. Hashimoto, K. Ishizuka, M. Naka-
mura, J. Am. Chem. Soc. 2009, 131, 11949; d) J.-M. BØgouin, M.
Rivard, C. Gosmini, Chem. Commun. 2010, 46, 5972; e) O. M.
Kuzmina, A. K. Steib, D. Flubacher, P. Knochel, Angew. Chem.
Int. Ed. 2013, 52, 4945; Angew. Chem. 2013, 125, 5045; f) C. E. I.
Knappke, S. Grupe, D. Gärtner, M. Corpet, C. Gosmini, A.
Jacobi von Wangelin, Chem. Eur. J. 2014, 20, 6828; g) M. Corpet,
X.-Z. Bai, C. Gosmini, Adv. Synth. Catal. 2014, 356, 2937; h) B.
BarrØ, L. Gonnard, R. Campagne, S. Reymond, J. Marin, P.
[18] O. M. Kuzmina, A. K. Steib, D. Flubacher, P. Knochel, Org. Lett.
2012, 14, 4818.
[19] M. E. Welsch, S. A. Snyder, B. R. Stockwell, Curr. Opin. Chem.
Biol. 2010, 14, 347.
[20] a) H. Zhong, H. Lai, Q. Fang, J. Phys. Chem. C 2011, 115, 2423;
b) R. V. Patel, Y.-S. Keum, S. W. Park, Mini-Rev. Med. Chem.
2014, 14, 768.
[21] a) C. Kaes, A. Katz, M. W. Hosseini, Chem. Rev. 2000, 100, 3553;
b) Comprehensive Coordination Chemistry II, Vol. 1 (Eds.: J. A.
McCleverty, T. J. Meyer), Elsevier, Oxford, 2004.
Received: November 17, 2015
Revised: January 11, 2016
Published online: February 16, 2016
3812
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 3809 –3812