Carbonylative Coupling of Alkyl Zinc Reagents with Benzyl Bromides Catalyzed by a Nickel/NN2 Pincer Ligand Complex

Article abstract of DOI:10.1002/anie.201710089

An efficient catalytic protocol for the three-component assembly of benzyl bromides, carbon monoxide, and alkyl zinc reagents to give benzyl alkyl ketones is described, and represents the first nickel-catalyzed carbonylative coupling of two sp³-carbon fragments. The method, which relies on the application of nickel complexed with an NN₂-type pincer ligand and a controlled release of CO gas from a solid precursor, works well with a range of benzylic bromides. Mechanistic studies suggest the intermediacy of carbon-centered radicals.

Full text of DOI:10.1002/anie.201710089

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Carbonylative Coupling of Alkyl Zinc Reagents with Benzyl
Bromides Catalyzed by an NN2 Pincer Ligand Nickel Complex
Authors: Troels Skrydstrup, Thomas Andersen, Aske Donslund, and
Karoline Neumann
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201710089
Angew. Chem. 10.1002/ange.201710089
Link to VoR: http://dx.doi.org/10.1002/anie.201710089
http://dx.doi.org/10.1002/ange.201710089

10.1002/anie.201710089
Angewandte Chemie International Edition
COMMUNICATION
Carbonylative Coupling of Alkyl Zinc Reagents with Benzyl
Bromides Catalyzed by an NN2 Pincer Ligand Nickel Complex
Thomas L. Andersen,[a] Aske S. Donslund,[a] Karoline T. Neumann[a] and Troels Skrydstrup[a]*
and C(sp3)-electrophile, we examined an alternative approach
Abstract: An efficient catalytic protocol for the three-
for carrying out such catalytic carbonylations with Ni-complexes
that would be less sensitive to the concentration of CO. We
reasoned that Ni(II) pincer complexes may be suitable due to the
strong and three-point binding of the pincer ligand to the metal
center, thereby possibly preventing multiple CO binding on some
of the reactive intermediates. Herein, we report the application of
such a Ni(II) pincer complex for the successful carbonylative
Negishi coupling between alkyl zinc reagents with benzyl
bromides in a CO atmosphere applying our two-chamber
technology (Scheme 1).[9] Many of these fast coupling reactions
prove to be high yielding and function at near ambient
temperature. A prominent feature of these reactions is the
viability of sp3-derived building blocks, which are problematic
components in the analogous plethora of Pd-catalyzed
carbonylation reactions, due to the inherent tendency of
palladium alkyl complexes to undergo β-hydride elimination.[10,11]
component assembly of benzyl bromides, carbon monoxide
and alkyl zinc reagents to benzyl alkyl ketones is described,
representing the first Ni-catalyzed carbonylative coupling of
two sp3-carbon fragments. The method, which relies on the
application of NN2-type pincer ligand nickel complexes and a
controlled release of CO gas from a solid precursor, works
well with a range of benzylic bromides. Mechanistic studies
suggest the intermediacy of carbon-centered radicals.
Over 40 years ago, Richard Heck and coworkers reported
their pioneering work on the Pd-catalyzed carbonylation reaction
of aryl halides with N- and O-nucleophiles.[1] Since then, these
carbonylation reactions have been expanded to the use of C-
nucleophiles as well, and today three-component versions of the
Negishi, Suzuki, Sonogashira, Stille and Hiyama couplings with
carbon monoxide have become common-use.[2] In contrast to
palladium, carbonylation reactions mediated by nickel catalysts
are far less prominent, which is surprising considering the
current and high focus from the chemical community to develop
catalysts based on earth abundant transition metals.
N
N
NiII Cl
XZn
Alkyl
O
C
N (cat.)
Ar
Alkyl
Ar
Br
= sp3-carbon
CO (1.5 equiv.)
generated ex situ
The underdeveloped state of Ni-catalyzed carbonylative
cross-coupling is undoubtedly related to the strong binding
affinity of CO to nickel leading to toxic Ni-carbonyl complexes
(e.g. Ni(CO)4).[3] A number of early examples on the use of
stoichiometric Ni(CO)4 in carbonylation reactions do exist,[4] but
the use of this and related species for catalytic purposes is
limited.[5] Recently, a few groups have relied on the use of CO
surrogates for the slow release of this diatomic gas as a means
to counteract the formation of inactive Ni-carbonyl species.[6] For
example, Ogoshi and coworkers successfully applied
phenylformate in a Ni(0)-catalyzed aza-Pausen-Khand reaction
for accessing γ-lactams, whereas direct exposure to CO gas
hampered the carbonylative cycloaddition.[7] Similarly, the
groups of Troupel and Weix relied on Fe(CO)5 as the CO-source
for demonstrating the concept of Ni(0)-catalyzed carbonylative
cross-electrophile coupling between aryl and alkyl halides.[8]
Scheme 1. General strategy for Ni-catalyzed carbonylative coupling of sp3-
carbon fragments.
We started our work, by investigating a set of nickel(II) pincer
1
complexes (NiL1–NiL3)[12–15] by H and 13C NMR spectroscopy
for their ability to mediate a sequence of discrete reaction steps
(Scheme 2). These steps consisted of transmetallation with an
organometallic reagent (e.g. propyl zinc bromide) followed by
migratory insertion with 13CO to produce a Ni-acyl species
(Scheme 2). Efficient transmetallation was observed in all three
cases, but only the use of NiL3 led to the formation of a stable
acyl species, as indicated by a singlet resonating at δ = 270.5
ppm in 13C NMR spectrum. Both NiL3 and the acyl species were
found to be stable over a prolonged period of exposure to
excess 13CO in THF (1 h, 30−50 °C). Furthermore, we found that
the in situ generated acyl species was consumed by the addition
of benzyl bromide, with concomitant formation of 13C-2a as the
major species as determined by 13C NMR spectroscopy. Without
optimization, 13C-2a could be isolated in a 40% yield by this
approach.
In an effort to identify suitable conditions for the first Ni-
catalyzed carbonylative coupling between a C(sp3)-nucleophile
[a]
T. L. Andersen, A. S. Donslund, K. T. Neumann, Prof. Dr. T.
Skrydstrup
Carbon Dioxide Activation Center (CADIAC), Department of
Chemistry and the Interdisciplinary Nanoscience Center (iNANO),
Aarhus University
To convert these stoichiometric experiments to a catalytic
protocol, we commenced optimization studies with NiL3 in the
carbonylative coupling between BnBr and nPr-ZnBr. For this
purpose, a two-chamber setup was employed, where CO gas
could be generated in a controlled rate from a solid precursor in
order to limit the exposure of the catalytic system to excess CO
(see Supporting Information). Conducting the reaction at 50 °C
in butyronitrile and THF, employing NiL3 (5 mol%) and 1.5
Gustav Wieds Vej 14, 8000 Aarhus C (Denmark)
Email: ts@chem.au.dk
Supporting information and the ORCID identification number(s) for
the author(s) can be found under http://dx.doi.org/XXXX.
This article is protected by copyright. All rights reserved.

10.1002/anie.201710089
Angewandte Chemie International Edition
COMMUNICATION
N
N
NiII
N
ligand amine moiety was investigated (entries 7–12). The best
results were obtained by applying the N,N-dimethyl analog NiL5
(7.5 mol%), which in combination with the addition of BnBr over
a 4 h period, furnished the desired ketone 2a in an isolated yield
of 92% (entry 13). Remarkably, when NiL5 was employed in the
reaction sequence depicted in Scheme 2, it was possible to
isolate 13C-2a in 99% yield (see Supporting Information). Control
experiments revealed that the absence of a Ni-based catalyst
leads to no product formation. Moreover, the use of NiCl2•DME
as a Ni(0)-precursor did not result in any conversion of the
starting bromide. The latter finding supports the suggestion that
low valent Ni-carbonyl species are not involved in the
transformation.
13
C
13CO
O
C
PrZnBr
BnBr (1)
N
NiII13
N
N
NiII Cl
N
O
(1.1 equiv)
N
N
13C-2a - 40%
(for NiL3)
δ = 270.5 ppm
with NiL3
13CO
13CO
no rxn
PPh2
NR2
NR2
N
L
L
8
NMe2
NiII Cl
NMe2
3
4
5
6
7
NEt2
NCy2
NMe2
NBn2
N
NiII
N
N
Cl
O
N
NiII Cl
NR2
9
N
PPh2
O
N
NiL1
NiL2
NiL3–9
Scheme 2. Stoichiometric reaction sequence and pincer complexes.
equivalents of CO with stirring for 5 h, led to the formation of 2a
in a 13% yield (Table 1, entry 1). Performing the reaction solely
in THF afforded only trace amounts of the product (entry 2). In
both cases considerable amounts of side products were
observed including direct coupling to 2b and homocoupling to
2c. In most reactions, minor quantities of the product 2d could
also be observed along with traces of dibenzylketone.
CO (1.5 equiv.)
ZnBr
NiL5 (7.5 mol%)
Br
C
R
R
1.25 equiv
[0.5 M in THF]
30 oC, 4 h + 30 min
THF/PrCN [1:1.7 mL]
O
+
0.4 mmol
3a–26a
C
C
O
C
O
O
tBu
MeS
4a - 90%
3a - 94%
5a - 81%
OMe
C
Table 1. Initial results and optimization.[a]
C
O
C
O
O
MeO
O
7a - 60%
O
CO (1.5 equiv)
C
6a - 88%
8a - 74%
NiL (5–10 mol%)
O
Br
O
2a
nPr-ZnBr (1.25 equiv)
C
O
C
O
C
(0.5 M in THF)
O
T (oC), t (h)
PrCN (1.7 mL)
C
O
O
NC
N
1 - 0.4 mmol
11a - 58%
9a - 68%
10a - 58%
2b
2c
2d
C
O
C
O
C
O
MeO
F3CO
O
F3C
F
Deviations from
std. conditions
NiL
2a/2b/2c/2d
14a - 86%
#
t [h]
T [°C]
CF3
12a - 83%
13a - 82%
[%][b]
F
F
[mol%]
NiL3[5]
NiL3[5]
NiL3[5]
C
C
O
C
1
2
3
none
5
5
5
50
50
30
13/38/10/3
2/17/29/1
35/2/2/16
O
O
Br
O
Br
in THF
F
F
17a - 89%
16a - 89%
15a - 80%
Cl
none
Cl
C
O
C
C
4
5
none
5
5
25
30
30
30
30
30
30
30
30
30
NiL3[5]
NiL3[10]
NiL3[10]
NiL4[10]
NiL5[10]
NiL6[10]
NiL7[10]
NiL8[10]
NiL9[10]
NiL5[7.5]
22/1/1/18
36/21/6/4
40/1/1/4
O
Br
O
Cl
18a - 71%
none
20a - 83%
19a - 89%
Cl
C
6
slow add. of 1
slow add. of 1
slow add. of 1
slow add. of 1
slow add. of 1
slow add. of 1
slow add. of 1
slow add. of 1
2
C
O
C
O
O
N
7
2
0/13/2/0
TsO
I
23a - 74%
Cl
21a - 71%
22a - 45%
Ph
8
2
56/11/2/1
50/10/6/1
45/12/1/0
44/22/1/0
56/15/2/0
92[c]/1/0/0
C*
O
S
C
O
C
9
2
N
N
O
F3C
N
25a - 60%
10
11
12
13
2
[
13C]25a - 60%
Ph
26a - 52%
24a - 61%
Cl
2
Other substrate
classes
O
N
Br
2
Br
Br
I
4+½
8% ketone
formation
no
no
conversion
no conversion
O
conversion
[a] All reactions were conducted in a two-chamber setup, in which CO was
released from a solid precursor in one chamber; see Supporting Information
for full detail. [b] Yields determined by GC using dodecane as internal
standard. [c] Isolated yield.
Scheme 3. Scope of benzyl bromides. All reactions were conducted in a two-
chamber setup, in which CO was released from a solid precursor in one
chamber; see Supporting Information for full detail. Yields are isolated and are
an average of two separate reactions.
The selectivity and yield for 2a could be increased
considerably (35% yield) by lowering the reaction temperature to
30–35 °C (entry 3). Attempts at lower reaction temperatures
(entry 4) or increasing the amount of catalyst (entry 5) were not
beneficial. As compound 2c[16] was suspected to be derived from
a side reaction involving unreacted BnBr, we attempted to
diminish this pathway by maintaining a low concentration of the
starting benzyl bromide. Thus, by adding a stock solution of 1 to
a stirred solution of NiL3 and nPr-ZnBr over 2 h, 2a was formed
in a 40% yield while side product formation was minimal (entry
6). Next, the effect of the substitution pattern on the pincer
The substrate scope for the catalytic reaction is
illustrated in Scheme 3. In general, many functionalities on the
aromatic ring of benzyl bromide are well-tolerated under
unaltered reaction conditions providing the desired ketones in
good to excellent yields. This includes alkyl (3a and 4a), sulfur
and oxygen (5a, 6a and 8a) substituents, as well as 2-allyl,
ester, acetamide, ketone and cyano groups (7a, 9a–11a). In
contrast, neither a nitro group nor a N-Boc protected amine were
compatible with the developed conditions (not shown). A range
of polyfluorinated benzylic bromides proved to be excellent
This article is protected by copyright. All rights reserved.

10.1002/anie.201710089
Angewandte Chemie International Edition
COMMUNICATION
substrates in the carbonylative transformation, affording the
desired products 12a–15a. Halides on the aromatic ring such as
bromide and chloride were untouched (compounds 16a–20a),
thereby allowing for the possibility of sequential cross-coupling
reactions. An aryl tosylate could also be employed to furnish the
corresponding ketone 21a. On the other hand, the use of a
substrate with an aryl iodide proved less optimal, but
nevertheless provided the carbonylative product 22a in a
reproducible fashion and an acceptable yield. Finally, the
feasibility of bromomethylated heteroaromatics as coupling
partners was tested affording ketones 23a–26a in satisfactory
yields. As demonstrated by the synthesis of compound [13C]25a,
the use of a two-chamber system for ex situ generation of
carbon monoxide provides a convenient platform for isotope
labeling, by simply substituting the CO releasing molecule for a
13C-labeled analog.
addition pathway[19] where single electron transfer from a
nickel(II) alkyl complex to organic halides occurs to form two
distinct nickel(III) species. A similar mechanistic scenario could
operate in the present case (Scheme 5a). Here, the catalytic
cycle would be initiated by transmetallation of the starting
nickel(II) complex A to give a nickel(II) alkyl complex B that
inserts CO to yield the corresponding nickel(II) acyl complex C.
a] proposed catalytic cycle - bimetallic oxidative addition pathway
NQ
R-ZnBr ZnBrCl
C
N
NiII Cl
Ni(I)/Ni(III)
comproportionation
NR2
NQ
A
N
NiII
NR2
CO
R
NQ
NiIII
NR
NQ
X
F
N
NiI
N
D
O
B
C
2 R
NR2
Attempts to employ (1-bromoethyl)benzene in the catalytic
coupling led to no overall conversion. Presumably the increased
bulk of this secondary bromide is not viable. Similarly, no
reactivity was observed with a primary alkyl iodide or bromide.
Cinnamyl bromide provided the linear carbonylation product in
8% isolated yield, while the corresponding alkylation product
was formed in 16%, suggesting that this substrate class may be
suitable, with further optimization.
red.
O
C
NQ
NiII
eilm.
O
C
R
X
Ar
Ar
R
Ar
N
NQ
NiIII
NR
bimetallic
ox. add.
NR2
N
R
H2C
C
C
2 O
Ar
E
C
b] cyclopropyl radical probe
standard conditions
no rxn
O
F3C
F3C
F3C
Br
35 - 26%[b]
Br
+
standard
conditions
[b]
CO (1.5 equiv.)
33
R-ZnBr
34 - 56%
R
Br
CF3
CF3
CF3
pressure
NiL5 (7.5–12.5 mol%)
C
O
+
1.25 equiv
[0.5M in THF]
tBu
tBu
30 oC, 4 hrs + 30 min
THF/PrCN [1:1.7 mL]
c] carbon monoxide pressure profile
[a]
[b]
tBu
0.4 mmol
27–33a
(2a[%] : 2b[%])
profile
Ni (7.5%)
nPr-ZnBr
Br
+
C
O
C
O
C
O
(without CO)
profile 1
profile 2
(0:62)
(3:56)
(88:3)
1.25 equiv
[0.5 M in THF]
standard conditions
with various CO
pressure profiles
tBu
O
tBu
O
EtO2C
1 - 0.4 mmol
27a - 76%
28a - 87%
29a - 86%
R
profile 3
no conversion
C
O
C
O
C
O
tBu
tBu
tBu
R= cyclopropyl: 32a - 47%
R= cyclohexyl: 33a - ~ 20%[a]
30a - 79%
31a - 90%
Scheme 4. Scope of alkyl zinc reagents. All reactions were conducted in a
two-chamber setup, in which CO was released from a solid precursor in one
chamber; see Supporting Information for full detail. Yields are isolated and are
an average of two separate reactions. [a] Compound not isolated.
A selection of alkyl zinc reagents was also tested in the
coupling with 4-tert-butylbenzyl bromide under unaltered
conditions (Scheme 4). Pleasingly, all of the primary alkyl zinc
reagents examined proved to be efficient coupling partners,
whereby the best results could be obtained with a small increase
in the catalytic loading of NiL5. This afforded the ketones 27a–
30a bearing substituents such as an ethyl carboxylate, an acetal
and a terminal olefin in good to excellent yields. We also found
that increased sterical hindrance on the C2-position of the alkyl
zinc reagent is tolerated as demonstrated in the synthesis of 31a
in 90% yield. Nevertheless, with secondary alkyl zinc reagents
the scope becomes more limited. For instance, the cyclopropyl
benzyl ketone 32a could be prepared in a 47% isolated yield, but
the use of a cyclohexyl zinc reagent led to less than 20% of 33a.
Stoichiometric experiments have shown that the main issue for
this compound is a non-optimal transmetallation step.[17]
d] deactivation by carbon monoxide
Ni (7.5%)
CO (equiv) 37 [b]
R-ZnBr
1.25 equiv
[0.5M in THF]
+
CO (equiv)
1.5 equiv no rxn
I
37
standard
0 equiv
37%
36 - 0.4 mmol
conditions
Scheme 5. Mechanistic assays and proposed mechanism. See Supporting
Information for full detail. [a] Yield determined by GC using dodecane as
internal standard. [b] Isolated yield.
Single electron transfer from this species to the
electrophile leads to the nickel(III) complex D and a benzylic
radical. Recombination of the radical with a second nickel(II) acyl
species C provides a formal nickel(III) acyl-alkyl complex E,
which undergoes reductive elimination to expel the product while
forming
a nickel(I) species F. Comproportionation with D
generates the two nickel(II) complexes A and C, which can
The mechanism of a related nickel(II) pincer complex
catalyzed Kumada coupling has been investigated thoroughly by
the group of Xile Hu.[18] They propose a bimetallic oxidative
reenter the catalytic cycle.[20]
This article is protected by copyright. All rights reserved.

10.1002/anie.201710089
Angewandte Chemie International Edition
COMMUNICATION
Several experiments were performed to support the
involvement of radical intermediates. The carbonylative coupling
between benzyl bromide and nPr-ZnBr was inhibited by the
addition of 1,4-dinitrobenzene (20 mol%), which serves as a
reversible electron acceptor (See Supporting Information).[21]
Furthermore, employing 33 as the electrophile, led to ring
opening of the cyclopropyl moiety to form 35 in a 26% yield,
being consistent with the proposed intermediacy of a benzylic
radical (Scheme 5b).[22] The main product from this reaction was
the primary bromide 34, however, control reactions confirmed
that this structure is not a viable precursor for 35 under the
employed conditions.
We are deeply appreciative of generous financial support from
the Danish National Research Foundation (grant no. DNRF118),
the Lundbeck Foundation, and Aarhus University.
Keywords: • Carbonylation • nickel catalysis • Pincer ligand
[1]
a) A. Schoenberg, R. F. Heck, J. Am. Chem. Soc. 1974, 96, 7761–
1764. b) A. Schoenberg, R. F. Heck, J. Org. Chem. 1974, 39, 3327–
3331. c) A. Schoenberg, I. Bartoletti, R. F. Heck, J. Org. Chem. 1974,
39, 3318–3326.
[2]
[3]
[4]
a) A. Brennführer, H. Neumann, M. Beller, Angew. Chem. Int. Edit.
2009, 48, 4114–4133. b) X. F. Wu, H. Neumann, M. Beller, Chem. Soc.
Rev. 2011, 40, 4986–5009.
a) M. F. Semmelhack, P. M. Helquist, Org. Synth. 1972, 52, 115–121.
b) The Merck Index. 10th ed. Rahway, New Jersey: Merck Co., Inc.,
1983, 93.
The role of carbon monoxide in the proposed mechanism
was examined by analyzing the effect of the reaction pressure
profile in bar on the outcome of the coupling between benzyl
bromide and nPr-ZnBr (Scheme 5c). As expected, the only
major outcome observed in the absence of CO was direct
alkylation to give 2b in a 62% yield. Curiously, we found that a
very similar product distribution was formed when CO (approx. 5
mmol) was applied from a balloon, giving rise to only trace
amounts of the carbonylation product under such low pressure
conditions (profile 1). In contrast to these results, the slow
release of CO, which are employed under the optimized
conditions, lead to near perfect selectivity between 2a and 2b,
the former of which could be isolated in this case in 88% yield
(profile 2). Modulating the conditions to ensure rapid CO-release
leads to catalyst inactivation, and no overall conversion of the
starting bromide is observed (profile 3). Thus, it seems that
although the starting nickel(II) halide (A) and the proposed
nickel(II) acyl species (C) are stable to excess CO gas, the
system remains vulnerable to high concentrations CO. We
suggest that the proposed Ni(I) and Ni(III) intermediates react
with CO to generate inactive off-cycle species. On the other
hand, when the partial pressure of CO is too low, direct
alkylation becomes the dominant pathway.
a) N. L. Bauld, Tetrahedron Lett. 1963, 27, 1841–1845. b) G. P.
Chiusoli, L. Cassar, Angew. Chem. Int. Edit. 1967, 6, 124–133. c) E.
Yoshisato. S. Tsutsumi, J. Org. Chem. 1968, 33, 869–871. d) I. Rhee,
M. Ryang, T. Watanabe, H. Omura, S. Murai, N. Sonoda, Synthesis,
1977, 11, 776–777. e) E. R. H. Jones, G. H. Whitham, M. C. Whiting, J.
Chem. Soc. 1957, 0, 4628–4633.
[5]
a) Q. L. Wang, C. Chen, Tetrahedron Lett. 2008, 49, 2916–2922. b) L.
Cassar, M. Foa, J. Organomet. Chem. 1973, 51, 381–393. c) I. Amer,
H. Alper, J. Org. Chem. 1988, 53, 5147–5148. d) H. Arzoumanian, F.
Cochini, D. Nuel, N. Rosas, Organometallics 1993, 12, 1871–1875. e)
H. Arzoumanian, M. Jean, D. Nuel, A. Cabrera, J. L. G. Gutierrez, N.
Rosas, Organometallics 1995, 14, 5438–5441. f) N. Satyanarayana, H.
Alper, I. Amer, Organometallics 1990, 9, 284–286. g) H. Alper, I. Amer,
G. Vasapollo, Tetrahedron Lett. 1989, 30, 2615–2616. h) H. Alper, G.
Vasapollo, Tetrahedron Lett. 1989, 30, 2617–2618. i) J. Tjutrins, J. L.
Shao, V. Yempally, A. A. Bengali, B. A. Arndtsen, Organometallics
2015, 34, 1802–1805.
[6]
For the use of DMF as CO source in Ni-catalyzed carbonylation for the
synthesis of symmetrical biaryl ketones, see: a) Y. Li, D.-H. Tu, B.
Wang, J-Y. Lu, Y.-Y. Wang, Z.-T. Liu, Z.-W. Liu, J. Lu, Org. Chem.
Front. 2017, 4, 569–572. For the synthesis of phthalimides, see: X. Wu,
Y. Zhao, H. Ge, J. Am. Chem. Soc. 2015, 137, 4924–4927.
a) Y. Hoshimoto, T. Ohata, Y. Sasaoka, M. Ohashi, S. Ogoshi, J. Am.
Chem. Soc. 2014, 136, 15877–15880. b) Y. Hoshimoto, K. Ashida, Y.
Sasaoka, R. Kumar, K. Kamikawa, X. Verdaguer, A. Riera, M. Ohashi,
S. Ogoshi, Angew. Chem. Int. Edit. 2017, 56, 8206–8210.
a) A. C. Wotal, R. D. Ribson, D. J. Weix, Organometallics 2014, 33,
5874–5881. b) M. Oçafrain, M. Devaud, M. Troupel, J. Perichon, J.
Chem. Soc. Chem. Comm. 1995, 2331–2332. c) E. Dolhem, M.
Oçafrain, J. Y. Nédélec, J. M. Troupel, Tetrahedron 1997, 53, 17089–
17096.
[7]
[8]
Finally, we found that no conversion of 3-iodo-1-phenyl
propane (36) occurred under standard carbonylation conditions
(Scheme 5d), but in the absence of CO, the corresponding
alkylation took place in 37% yield. This suggests that the
nickel(II)
alkyl
complex
B,
being
generated
upon
transmetallation, is capable of activating both alkyl and benzyl
halides, whereas the corresponding nickel(II) acyl complex C, is
only reactive in presence of more activated benzylic substrates.
Most likely, this is a consequence of the electronic differences
between the two complexes invoked by the binding of CO.
[9]
S. D. Friis, A. T. Lindhardt, T. Skrydstrup, Acc. Chem. Res. 2016, 49,
594–605.
[10] The use of primary alkyl iodides as electrophiles in Pd-catalyzed
carbonylation reactions has been studied by the Ryu group through the
concept of light mediated Atom Transfer Carbonylation (ATC) as a way
to surpress β-hydride elimination upon oxidative addition. For a review
see: S. Sumino, A. Fusano, T. Fukuyama, I. Ryu, Acc. Chem. Res.
2014, 47, 1563–1574. Furthermore, Sargant and Alexanian have
recently reported a Pd-catalyzed alkoxycarbonylation, of secondary
alkyl bromides. B. T. Sargant, E. J. Alexanian, J. Am. Chem. Soc. 2016,
138, 7520–7523.
[11] The use of alkyl derived organometallic reagents in Pd-catalyzed
carbonylative cross-coupling is limited, for examples, see; a) R. F. W.
Jackson, D. Turner, M. H. Block, J. Chem. Soc., Perkin Trans 1, 1997,
865–870. b) P. H. Lee, S. W. Lee, K. Lee, Org. Lett. 2003, 5, 1103–
1106. c) N. A. Bumagin, P. G. More, I. P. Beletskaya, J. Organomet.
Chem. 1989, 365, 379–387. d) Y. Wakita, T. Yasunaga, M. Kojima, J.
Organomet. Chem. 1985, 288, 261–268. e) T. Ishiyama, M. Murata, A.
Suzuki, N. Miyaura, J. Chem. Soc. Chem. Comm. 1995, 295–296.
In conclusion, a NN2-type pincer ligand nickel(II) complex
has been employed along with a controlled release of CO gas, in
order to facilitate Ni-catalyzed carbonylation to produce a range
of benzyl alkyl ketones. The method represents the first
carbonylative assembly of two sp3-hybridized carbon fragments,
which is catalyzed by nickel. Further studies are underway to
identify other nickel based pincer complexes for expanding the
scope of this reaction to other alkyl halides.
Acknowledgments
This article is protected by copyright. All rights reserved.

10.1002/anie.201710089
Angewandte Chemie International Edition
COMMUNICATION
[12] O. Vechorkin, Z. Csok, R. Scopelliti, X. L. Hu, Chem. Eur. J. 2009, 15,
3889–3899.
[13] a) N. Grüger, H. Wadepohl, L. H. Gade, Dalton Trans. 2012, 41,
14028–14030. b) S. Kumar, G. Mani, S. Mondal, P. K. Chattaraj, Inorg.
Chem. 2012, 51, 12527–12539. c) G. T. Venkanna, S. Tammineni, H.
D. Arman, Z. J. Tonzetic, Organometallics 2013, 32, 4656–4663.
[14] V. Soni, R. A. Jagtap, R. G. Gonnade, B. Punji, ACS Catal. 2016, 6,
5666–5672.
[15] Inspiration to NiL3 comes from Chatani and coworkers’ extensive
studies on Ni-catalyzed C–H activation applying
a N,N’-bidentate
directing group. For a recent and comprehensive overview on this work,
see: N. Chatani, Top. Organomet. Chem. 2016, 56, 19–46.
[16] Products of the type 2d are formed in a non-catalyzed reaction between
benzyl bromides and the product ketone in the presence of nPr-ZnBr.
[17] When subjecting NiL3 (0.1 mmol) to cyclohexylzinc bromide (1 equiv.)
at room temperature in THF-d8 under argon, the solution turned black
within minutes. Analysis by 1H NMR after 1 and 16 h revealed a
complex reaction mixture, with only minimal transmetallation product.
[18] a) J. Breitenfeld, J. Ruiz, M. D. Wodrich, X. L. Hu, J. Am. Chem. Soc.
2013, 135, 12004–12012. b) J. Breitenfeld, M. D. Wodrich, X. L. Hu,
Organometallics 2014, 33, 5708–5715.
[19] a) K. C. MacLeod, J. L. Conway, L. M. Tang, J. J. Smith, L. D.
Corcoran, K. H. D. Ballem, B. O. Patrick, K. M. Smith, Organometallics
2009, 28, 6798–6806. b) D. Zhu, P. H. M. Budzelaar, Organometallics
2010, 29, 5759–5761.
[20] A mechanistic manifold cannot be excluded, which involves a nickel(IV)
species. The ligand rather than the metal center may be oxidized in a
Ni(IV) pathway, as previously proposed in ref. 18a.
[21] TEMPO was found to be an unsuitable radical trap, as this species
reacts with the nickel(II) alkyl complex B, leading to O-alkylation of
TEMPO. A similar observation has previously been made: See ref. 18a.
[22] K. E. Poremba, N. T. Kadunce, N. Suzuki, A. H. Cherney, S. E.
Reisman, J. Am. Chem. Soc. 2017, 139, 5684–5687.
This article is protected by copyright. All rights reserved.

10.1002/anie.201710089
Angewandte Chemie International Edition
COMMUNICATION
COMMUNICATION
Thomas L. Andersen, Aske S. Donslund,
Karoline T. Neumann, Troels
Skrydstrup*
N
R
Br
O
C
N
NiII Cl
R
O
Alkyl
Page No. – Page No.
CO
+
NMe2
NN2 pincer ligand based
nickel(II) catalyst
= C(sp3)
XZn
Alkyl
+
Carbonylative Coupling of Alkyl Zinc
Reagents with Benzyl Bromides
Catalyzed by an NN2 Pincer Ligand
Nickel Complex
Mild reaction conditions
Easy-to-prepare pincer catalyst
Near-stoichiometric CO
Radical intermediates
Hold tight: NN2 type pincer ligand nickel(II) complexes are employed as catalysts
in the carbonylative coupling of benzyl bromides and alkyl zinc reagents. The
method relies on the controlled release of CO gas from a solid precursor, and thus
circumvents the commonly encountered issue of Ni-catalyst deactivation upon CO-
saturation. With this approach, the first examples of Ni-catalyzed carbonylative
couplings of two sp3-hybridized carbon centers are reported.
This article is protected by copyright. All rights reserved.