Dialkyl Ether Formation by Nickel-Catalyzed Cross-Coupling of Acetals and Aryl Iodides

Article abstract of DOI:10.1002/anie.201503936

A new substrate class for nickel-catalyzed C(sp³) cross-coupling reactions is reported. α-Oxy radicals generated from benzylic acetals, TMSCl, and a mild reductant can participate in chemoselective cross-coupling with aryl iodides using a 2,6-bis(N-pyrazolyl)pyridine (bpp)/Ni catalyst. The mild, base-free conditions are tolerant of a variety of functional groups on both partners, thus representing an attractive C-C bond-forming approach to dialkyl ether synthesis. Characterization of a [(bpp)NiCl] complex relevant to the proposed catalytic cycle is also described.

Full text of DOI:10.1002/anie.201503936

.
Angewandte
Communications
International Edition: DOI: 10.1002/anie.201503936
German Edition: DOI: 10.1002/ange.201503936
Synthetic Methods
Dialkyl Ether Formation by Nickel-Catalyzed Cross-Coupling of
Acetals and Aryl Iodides**
Kevin M. Arendt and Abigail G. Doyle*
Abstract: A new substrate class for nickel-catalyzed C(sp3)
cross-coupling reactions is reported. a-Oxy radicals generated
from benzylic acetals, TMSCl, and a mild reductant can
participate in chemoselective cross-coupling with aryl iodides
using a 2,6-bis(N-pyrazolyl)pyridine (bpp)/Ni catalyst. The
mild, base-free conditions are tolerant of a variety of functional
À
groups on both partners, thus representing an attractive C C
bond-forming approach to dialkyl ether synthesis. Character-
ization of a [(bpp)NiCl] complex relevant to the proposed
catalytic cycle is also described.
O
ver the past decade, tremendous advances have been
made in the field of nickel-catalyzed cross-coupling,[1] includ-
ing the development of stereoconvergent alkyl cross-coupling
reactions[2] and chemoselective cross-electrophile couplings.[3]
These two synthetic methods have recently been the subject
of mechanistic studies,[4] which have revealed a pathway that
involves a radical-chain mechanism. Specifically, the Weix
and Fu groups concluded that oxidative addition of an organic
halide to nickel(I) can occur by a bimetallic mechanism
wherein a nickel(II) halide and transient organic radical are
produced (Figure 1a). This organic radical is subsequently
captured by another nickel(II) species to furnish a nickel(III)
À
intermediate primed for C C bond formation. This ability to
engage highly reactive organic radicals in cross-coupling
reactions presents exciting new possibilities for organic
synthesis, especially if alternative mechanisms can be
exploited for radical generation.
Figure 1. Nickel-catalyzed C(sp3) cross-coupling with radicals.
TMS=trimethylsilyl.
The Weix group recently reported a nickel/titanium
catalyst system for mild and regioselective arylation of
epoxides with aryl halides. This system demonstrated that
radical generation and nickel-catalyzed coupling can be
accomplished by two different catalysts.[5] Our group, in
collaboration with the MacMillan group, subsequently de-
scribed a platform wherein a photoredox catalyst was able to
supply radical partners to a nickel-catalyzed coupling reac-
limitations inherent to transmetalation with C(sp3) organo-
boron reagents by generating radicals from these reaction
partners utilizing
a related photoredox/nickel catalysis
approach.[7]
In an effort to advance the synthetic scope of alkyl cross-
coupling reactions further, we questioned what alternative
radical-generating processes might be coupled with the nickel
machinery. It is known that the combination of a Lewis acid
and reductant can generate a-oxy radicals from acetals.[8] We
envisioned that successful coupling of these radicals with aryl
tion.[6] This approach enabled a-amino C CO2H and C H
À
À
3
À
bonds to serve as handles for C(sp ) C bond formation. In
a concurrent study, the Molander group was able to overcome
À
and heteroaryl halides would deliver an attractive C C bond-
forming synthesis of dialkylethers (Figure 1b). Furthermore,
the transformation would enable the use of simple, abundant,
and stable acetals as handles for coupling reactions.
[*] K. M. Arendt, Prof. A. G. Doyle
Department of Chemistry, Princeton University
120 Washington Road, Princeton, NJ 08544 (USA)
E-mail: agdoyle@princeton.edu
Dialkylethers are prevalent in pharmaceuticals and bio-
logically active compounds.[9] However, approaches to their
[**] Financial support from NIGMS (R01 GM100985) is gratefully
acknowledged. We thank Brian Schaefer (Princeton University) for
assistance with EPR and computational data, and Phillip D. Jeffrey
(Princeton University) for X-ray structure determination.
3
À
synthesis typically rely upon C(sp ) O bond formation (e.g.
Williamson ether synthesis) and remain limited because of the
requirements for strong base, high temperature, and/or
unhindered substrates.[10] There are fewer examples of C C
À
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201503936.
bond-forming approaches to dialkyl ethers, and these often
9876
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 9876 –9880

Angewandte
Chemie
rely on the intrinsic reactivity of the electrophilic or
intermediacy of Vand its formation using Zn as a reductant.[17]
3
nucleophilic reaction partner.[11] C(sp ) C bond-forming
With a catalytic amount of nickel and 3 equivalents TMSCl,
cross-coupling to form the desired ether A was observed.
However, the selectivity for A varied significantly depending
on the identity of the ligand on nickel. Phosphine and amine
ligands commonly employed in cross-coupling were not
efficient in the reaction (entries 6–11). After extensive
ligand evaluation, it was found that highest cross-selectivity
could be achieved using the tridentate amine ligand 2,6-bis(N-
pyrazolyl)pyridine (bpp), which delivered the desired cross-
coupled product in 91% yield (entry 1).
À
reactions that feature the modularity of a cross-coupling
reaction but are applicable to heteroatom-containing, phar-
maceutically relevant structures like ethers are rare in the
field. Bode and co-workers have reported Lewis acid
catalyzed additions of Ar-BF3K salts to acetals.[12] However,
3
À
the conditions are restricted to primary C(sp ) C bond
formation. Molander and co-workers have shown that a-oxy
BF3K salts couple efficiently with aryl chlorides under
palladium catalysis.[13] However, these nucleophiles are not
readily available and the conditions for cross-coupling require
high temperature and strong base. Thus, we recognized that
a nickel-catalyzed approach from readily available acetals
and aryl halides could offer unprecedented mildness and
generality for the preparation of this structural motif.[14]
As shown in Figure 1c, we envisioned a catalytic cycle
initiated by oxidative addition of Ni0 (I) into an aryl halide.
The resulting NiII intermediate II would intercept an a-oxy
radical (V) generated from the acetal III, a Lewis acid, and
reductant such as Zn (Ered1/2 = À1.0 V vs SCE in MeCN),
delivering the NiIII adduct VI. Subsequent reductive elimi-
nation would afford the dialkylether VIII, releasing the NiI
species VII. Completion of the catalytic cycle requires
turnover of the nickel catalyst by a reductant to regenerate
I (Ered1/2[NiII/Ni0] = À1.2 V vs SCE in DMF).[15,16]
The bpp ligand has seen limited application in other
nickel-catalyzed cross-coupling reactions.[19] As such, we
sought to elucidate the structure and properties of a bpp-
ligated nickel species relevant to the catalytic reaction. Based
on our proposed mechanism, we chose to prepare the
nickel(I) chloride complex 1 (see VII in Figure 1c). Compro-
portionation of [Ni(cod)2] and NiCl2·DME in the presence of
2 equivalents of bpp provided the monomeric chloride
complex 1 as dark green crystals (Figure 2a). Importantly,
1 is catalytically competent, thus delivering dialkyl ether A in
64% yield under standard conditions. A single-crystal X-ray
diffraction study established that 1 adopts a slightly distorted
octahedral geometry in the solid state and associates as a m-
chloride dimer upon recrystallization from ethanol (Fig-
ure 2b).[20] The complex 1 exhibits an axial EPR signal with
gk = 2.262 and g? = 2.070, consistent with its formulation as
a d9 nickel(I) complex, rather than as a d8 nickel(II) species
with the radical residing in the bpp ligand (Figure 2c).
Computational experiments confirm that the unpaired spin
is localized exclusively on nickel in a dx2Ày2 orbital.[17] These
features are remarkably similar to those of [(terpy)NiBr] and
[(terpy)NiCl], as reported by Vicic[21] and Wieghardt[22]
respectively, with the exception that the terpy complexes
are monomeric and square-planar in the solid state. The redox
potential of [(bpp)NiCl] (Ered1/2 = À0.8 V vs. SCE in MeCN) is
also within 200 mV of those for [(terpy)NiI] halides.[23] Taken
together, these data do not provide an obvious parameter to
discriminate the selectivity observed between bpp and terpy
in this chemistry. Since the ligands themselves have more
distinct redox potentials (terpy: Ered1/2 = À2.2 V vs SCE in
MeCN; bpp: Ered1/2 = À0.9 V vs SCE in MeCN), it is possible
that the reaction selectivity arises at stages in the catalytic
cycle wherein oxidation/reduction of nickel is ligand-based.
With optimized reaction conditions in hand, we next
explored the generality of the transformation. High cross-
selectivity was observed with a variety of aryl iodides
(Table 2). Substitution around the aromatic ring was well
tolerated (3,4), though ortho-tolyl iodide (5) suffered from
rapid formation of 2,2’-bitolyl. Both electron-rich (7) and
electron-poor (9) aryl iodides coupled with equal levels of
efficiency.[24] This reactivity represents a significant advance
as methods to form alkyl ethers via oxocarbenium ions
typically rely on the inherent reactivity of the nucleophile and
are thus limited to electron-rich coupling partners.[11a,b] Nitro-
gen-containing heterocycles, which are common in pharma-
ceuticals, undergo the reaction in good yield (12 and 13).
Protic as well as Lewis-basic functionality is tolerated (14 and
15) and in the case of carbonyl-containing aryl iodides, no
We began our investigations by exploring the coupling of
benzaldehyde dimethylacetal with iodobenzene using TMSCl
as a Lewis acid and Zn as a reductant (Table 1).[18] In the
absence of a nickel catalyst, the acetal converts into a mixture
of rac- and meso-pinacol dimers (B), presumably via the a-
oxy radical V (entry 2). Indeed, trapping experiments using
(2,2,6,6-tetramethylpiperidin-1-yl)oxyl (TEMPO) deliver
77% yield of the TEMPO adduct, also supporting the
Table 1: Reaction optimization studies.
Entry
Deviation from standard conditions
Yield [%][a]
A
B
C
1
2
3
none
91
0
0
3
13
0
3
0
0
no NiCl2·DME
no Zn0
4
5
6
7
8
9
10
11
12
no TMSCl
1 equiv PhI
0
55
5
0
2
25
2
PPh3 instead of bpp[b]
dppbz instead of bpp
bpy instead of bpp
terpy instead of bpp
Ph-Box instead of bpp
Ph-PyBox instead of bpp
Mn0 instead of Zn0
30
28
10
11
39
27
15
35
10
31
35
0
4
38
10
2
0
23
38
35
[a] Determined by GC using dodecane as a quantitative internal
standard. [b] 20 mol% ligand loading. Box=bis(oxazoline),
DMA=N,N-dimethylacetamide, DME=dimethoxyethane.
Angew. Chem. Int. Ed. 2015, 54, 9876 –9880
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 9877

.
Angewandte
Communications
Figure 2. a) Comproportionation to form [bpp(Ni)Cl] (1). b) ORTEP diagram of dimeric 1. Ellipsoids shown shown at 30% probability and H
À
atoms omitted for clarity (except for ethanol O H atoms). Selected bond lengths [Š]: Ni–Cl1 2.333; Ni–Cl2 2.484; Ni–N3 2.012; Ni–N1 2.009;
Ni–O1 2.078. Selected bond angles [deg]: N3-Ni-Cl 173.8; N5-Ni-Cl 101.3; N1-Ni-N5 154.9. c) X-band EPR spectrum of 1 frozen in MeCN glass at
10 K. Black line: experimental spectrum, red line: simulation using Easyspin. Parameters: microwave frequency 9.3742949 GHz; microwave power
2.0 mW; modulation amplitude 0.1 mT/100 kHz. cod=1,5-cyclooctadiene, THF=tetrahydrofuran.
Table 2: Scope of aryl iodides.[a]
Table 3: Scope of acetals.[a]
[a] Yield is that of the isolated product (0.50 mmol). [b] 4.0 equiv NaBF4
added. [c] Vinyl triflate was employed.
[a] Yield is that of the isolated product (0.50 mmol scale). [b] 4.0 equiv
NaBF4 added. [c] Reaction run at 608C. [d] p-Iodoanisole was employed
as the coupling partner.
evidence of products arising from ketyl formation are
observed, even in the presence of excess Lewis acid and
reductant. Finally, we were able to extend the scope to vinyl
substrates: a cyclohexenyl vinyl triflate reacts with good
efficiency to provide the desired allylic ether 16.
Several of the trends observed with the aryl iodides were
recapitulated with the acetals (Table 3). Substituents that are
able to coordinate to the metal are well tolerated at the
ortho position, whereas simple alkyl groups decrease the
efficiency of the reaction (17 versus 5). When more than one
halide is present, the cross-coupling proceeds selectively at
the aryl iodide (22). Lewis-basic functionality is tolerated
(21). Electron-rich (10), electron-neutral (4), and electron-
poor (23) acetals perform well under these reaction con-
ditions. Recently, the MacMillan group reported a powerful
photoredox/organocatalysis approach to form similar product
types.[11c,d] This method, however, is limited to easily reduced,
electron-deficient aryl nitrile coupling partners. The current
strategy allows us to bring together two carbon fragments
independent of their electronic nature. Furthermore,
À
although nickel-catalyzed C O coupling of benzylic ethers
is well-known,[25] we have not observed the products arising
9878 www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 9876 –9880

Angewandte
Chemie
[4] a) S. Biswas, D. J. Weix, J. Am. Chem. Soc. 2013, 135, 16192;
from subsequent arylation of the benzylic ether products in
Table 2 and Table 3.
Pleasingly other types of benzylic acetals can be used to
arrive at structurally distinct products (Figure 3). Phthalan-
derived acetal 25 undergoes coupling with iodobenzene to
b) N. D. Schley, G. C. Fu, J. Am. Chem. Soc. 2014, 136, 16588.
[5] a) Y. Zhao, D. J. Weix, J. Am. Chem. Soc. 2014, 136, 48; Also see:
b) L. K. G. Ackerman, L. L. Anka-Lufford, M. Naodovic, D. J.
Weix, Chem. Sci. 2015, 6, 1115; c) Y. Zhao, D. J. Weix, J. Am.
Chem. Soc. 2015, 137, 3237.
[6] Z. Zuo, D. T. Ahneman, L. Chu, J. A. Terrett, A. G. Doyle,
D. W. C. MacMillan, Science 2014, 345, 437.
[7] J. C. Tellis, D. N. Primer, G. A. Molander, Science 2014, 345, 433.
[8] B. Hatano, K. Nagahashi, S. Habaue, Chem. Lett. 2007, 36, 1418.
[9] a) S. D. Roughley, A. M. Jordan, J. Med. Chem. 2011, 54, 3451;
b) R. Silvestri, M. Artico, S. Massa, T. Marceddu, F. De Montis,
P. La Colla, Bioorg. Med. Chem. Lett. 2000, 10, 253; c) G.
De Martino, G. La Regina, A. Di Pasquali, R. Ragno, A.
Bergamini, C. Ciaprini, A. Sinistro, G. Maga, E. Crespan, M.
Artico, R. Silvestri, J. Med. Chem. 2005, 48, 4378; d) J. N. N.
Eildal, J. Andersen, A. S. Kristensen, A. M. Jørgensen, B. Bang-
Andersen, M. Jørgensen, K. Strømgaard, J. Med. Chem. 2008, 51,
3045; e) D. Ameen, T. J. Snape, MedChemComm 2013, 4, 893.
[10] a) A. W. Williamson, J. Chem. Soc. 1852, 4, 229; b) J. Barluenga,
M. Tomµs-Gamasa, F. Aznar, C. ValdØs, Angew. Chem. Int. Ed.
2010, 49, 4993; Angew. Chem. 2010, 122, 5113.
[11] a) P. N. Moquist, T. Kodama, S. E. Schaus, Angew. Chem. Int. Ed.
2010, 49, 7096; Angew. Chem. 2010, 122, 7250; b) Y. Luan, K. S.
Barbato, P. N. Moquist, T. Kodama, S. E. Schaus, J. Am. Chem.
Soc. 2015, 137, 3233; c) K. Qvortrup, D. A. Rankic, D. W. C.
MacMillan, J. Am. Chem. Soc. 2014, 136, 626; d) D. Hager,
D. W. C. MacMillan, J. Am. Chem. Soc. 2014, 136, 16986.
[12] a) C.-V. T. Vo, T. A. Mitchell, J. W. Bode, J. Am. Chem. Soc.
2011, 133, 14082; for addition of alkynyl BF3K salts to generate
Figure 3. Access to dialkyl ether-containing bioactive molecules.
generate isobenzofuran 26, which represents the core of
a number of SSRI pharmaceuticals.[9d,e] Cross-coupling of
dioxolane 27 furnishes ether 28, which possesses an ethoxy
alcohol tether that could be easily elaborated to motifs found
in numerous bioactive compounds.[9b] The ethoxy alcohol
handle can also serve as a traceless directing group for an
additional round of nickel-catalyzed cross-coupling, ulti-
mately affording tri- or diarylalkanes, which are valuable
pharmacophores in their own right.[26] These results highlight
opportunities to apply this C(sp3) coupling reaction to late-
stage diversification of bioactive compounds. While exo- and
endocyclic acetals are competent reaction partners, acyclic
acetals other than dimethyl acetals have thus far proven
unreactive under the outlined coupling conditions.
À
secondary C C bonds, see: b) T. A. Mitchell, J. W. Bode, J. Am.
Chem. Soc. 2009, 131, 18057.
[13] G. A. Molander, S. R. Wisniewski, J. Am. Chem. Soc. 2012, 134,
16856. Also see Ref. [7].
[14] MacMillan and co-workers have recently reported vinylation of
a-oxy acids with aryl iodides to deliver allylic ethers: A. Noble,
S. J. McCarver, D. W. C. MacMillan, J. Am. Chem. Soc. 2015,
137, 624.
[15] M. Durandetti, M. Devaud, J. Perichon, New J. Chem. 1996, 20,
659.
[16] Molander, Kozlowski, and co-workers have proposed an alter-
nate mechanism that could be operative, wherein Ni0 intercepts
a radical intermediate and the subsequent NiI alkyl undergoes
oxidative addition with aryl halides: O. Gutierrez, J. C. Tellis,
D. N. Primer, G. A. Molander, M. C. Kozlowski, J. Am. Chem.
Soc. 2015, 137, 4896.
[17] See the Supporting Information for more discussion of substrate
scope, experimental details, and mechanistic investigations.
[18] Cheng and co-workers reported a nickel-catalyzed coupling of
aromatic aldehydes and aryl bromides which delivers benzhydryl
alcohol products by a distinct mechanism: K. K. Majumdar, C.-
H. Cheng, Org. Lett. 2000, 2, 2295.
[19] a) S. W. Smith, G. C. Fu, Angew. Chem. Int. Ed. 2008, 47, 9334;
Angew. Chem. 2008, 120, 9474; b) G. D. Jones, J. L. Martin, C.
McFarland, O. R. Allen, R. E. Hall, A. D. Haley, R. J. Brandon,
T. Konovalova, P. J. Desrochers, P. Pulay, D. A. Vicic, J. Am.
Chem. Soc. 2006, 128, 13175.
In conclusion we have demonstrated that bench-stable
and abundant benzylic acetals participate in formal C(sp ) O
cross-coupling with aryl iodides in the presence of a nickel
catalyst, Lewis acid, and reductant. This mild and functional-
3
À
À
group tolerant C C bond-forming approach to dialkyethers
provides access to a variety of acylic and cyclic products in
good yields with high chemoselectivity. We have shown that
an a-oxy radical is generated under the reaction conditions
and propose that this radical intercepts a [(bpp)nickel]
complex to selectively generate the cross-coupled product.
Keywords: cross-coupling · ethers · nickel · radicals ·
synthetic methods
[20] CCDC 1062498 (1) contains the supplementary crystallographic
data for this manuscript. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre.
[21] J. T. Ciszewski, D. Y. Mikhaylov, K. V. Holin, M. K. Kadirov,
O. S. Budnikova, D. A. Vicic, Inorg. Chem. 2011, 50, 8630.
[22] M. Wang, J. England, T. Weyhermüller, K. Wieghardt, Eur. J.
Inorg. Chem. 2015, 2015, 1511.
How to cite: Angew. Chem. Int. Ed. 2015, 54, 9876–9880
Angew. Chem. 2015, 127, 10014–10018
[1] S. Z. Tasker, E. A. Standley, T. F. Jamison, Nature 2014, 509, 299.
[2] For a recent review, see: E. C. Swift, E. R. Jarvo, Tetrahedron
2013, 69, 5799.
[3] a) D. A. Everson, D. J. Weix, J. Org. Chem. 2014, 79, 4793; b) For
a review of metal-catalyzed cross-electrophile couplings, see: T.
Moragas, A. Correa, R. Martin, Chem. Eur. J. 2014, 20, 8242.
[23] Experiments to determine if [(bpp)NiCl] (1) can generate the a-
oxy radical V in a radical chain mechanism have thus far been
inconclusive. Specifically, reaction of the acetal A, TMSCl, and
Angew. Chem. Int. Ed. 2015, 54, 9876 –9880
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 9879

.
Angewandte
Communications
1 does not produce the pinacol byproduct B. Trapping experi-
ments with TEMPO, A, TMSCl, and 1 lead to bleaching of 1. See
Ref. [4,b] for a similar observation.
[26] a) M. A. Greene, I. M. Yonova, F. J. Williams, E. R. Jarvo, Org.
Lett. 2012, 14, 4293; b) B. L. H. Taylor, M. R. Harris, E. R. Jarvo,
Angew. Chem. Int. Ed. 2012, 51, 7790; Angew. Chem. 2012, 124,
7910.
[24] It was determined that some electron-deficient aryl iodides
required the addition of NaBF4 to achieve high efficiency. See:
G. A. Molander, K. M. Traister, B. T. O’Neill, J. Org. Chem.
2014, 79, 5771.
[25] For an example, see: B.-T. Guan, S.-K. Xiang, B.-Q. Wang, Z.-P.
Sun, Y. Wang, K.-Q. Zhao, Z.-J. Shi, J. Am. Chem. Soc. 2008, 130,
3268, and references therein.
Received: April 29, 2015
Published online: July 14, 2015
9880 www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 9876 –9880