Cobalt-catalyzed C-H borylation

Article abstract of DOI:10.1021/ja500712z

A family of pincer-ligated cobalt complexes has been synthesized and are active for the catalytic C-H borylation of heterocycles and arenes. The cobalt catalysts operate with high activity and under mild conditions and do not require excess borane reagents. Up to 5000 turnovers for methyl furan-2-carboxylate have been observed at ambient temperature with 0.02 mol % catalyst loadings. A catalytic cycle that relies on a cobalt(I)-(III) redox couple is proposed.

Full text of DOI:10.1021/ja500712z

Communication
pubs.acs.org/JACS
Cobalt-Catalyzed C−H Borylation
Jennifer V. Obligacion, Scott P. Semproni, and Paul J. Chirik*
Department of Chemistry, Princeton University, Princeton, New Jersey 08544, United States
S
* Supporting Information
of readily synthesized and modular pincer-type ligands that are
active for catalytic C−H borylation under thermal conditions.
The base metal-catalyzed C−H functionalization process likely
operates via a Co(I)−(III) redox cycle, distinct from known
Ir(III)−(V) pathways.
ABSTRACT: A family of pincer-ligated cobalt complexes
has been synthesized and are active for the catalytic C−H
borylation of heterocycles and arenes. The cobalt catalysts
operate with high activity and under mild conditions and
do not require excess borane reagents. Up to 5000
turnovers for methyl furan-2-carboxylate have been
observed at ambient temperature with 0.02 mol % catalyst
loadings. A catalytic cycle that relies on a cobalt(I)−(III)
redox couple is proposed.
Inspired by recent reports of iron-17−19 and cobalt-catalyzed
alkene hydroboration,20 we reasoned that if a sufficiently
electron-rich cobalt center was generated, C−H activation may
be enabled and coupled to C−B bond formation. To test this
hypothesis, cobalt alkyl complexes with both aryl- and alkyl-
substituted redox-active bis(imino)pyridine ligands21 were
initially evaluated for the catalytic borylation of 2-methylfuran
with pinacolborane (HBPin) (Figure 1). Analysis of the products
arbon−hydrogen bonds are abundant in most organic
1
C
molecules, and methods for their direct, selective catalytic
functionalization are attractive as an efficient means to access
more complex molecular entities.1−3 The transition-metal-
catalyzed borylation of heterocycles and arenes has emerged as
one of the most effective and efficient methods for C−H
functionalization, in part due to the versatility of the resulting aryl
boronate esters in synthesis.4−6 Catalytic borylation activity has
been observed with several transition-metal complexes, although
in most cases the synthetic utility is limited by poor activity or
photochemical methods required for catalyst activation.4,7
Iridium diene complexes such as [Ir(COD)OMe]2 (COD =
1,5-cyclooctadiene) in combination with substituted bipyridine
or phenanthroline ligands are the most widely used catalysts for
C−H borylation.8 Silica-supported monophosphine rhodium
and iridium catalysts as well as iridium nanoparticles in ionic
liquids have also been reported and offer reactivity and selectivity
advantages over their soluble counterparts.9−12 Despite their
widespread application, the number of precious metal catalysts
reported in the literature that enable efficient turnover is limited.7
Thus, the discovery of new, readily accessible and modular
catalyst platforms is attractive as rational changes to the
electronic and steric profile of the catalyst may expand the
scope and utility of metal-catalyzed C−H borylation.
by H NMR spectroscopy and GC-MS established selective
borylation of the 5-position of the heterocycle. With the aryl-
substituted compound 1, 59% conversion of the starting material
was observed after 72 h at 23 °C. Introduction of a cyclohexyl
imine substituent increased the activity of the cobalt catalyst such
Base metal compounds are of interest because of their
potential cost and environmental advantages but more
importantly due to their distinct electronic structures and ability
to enable new mechanistic pathways that may ultimately result in
new reactivity.13 Cyclopentadienyl iron N-heterocyclic car-
bene,14 Fe2O3 nanoparticles,15 and heterobimetallic copper
complexes16 have recently been reported for the borylation of
aromatic compounds. While important precedents for base
metal-catalyzed C−H functionalization, the catalyst loadings are
high, and the substrate scope is limited. Also the operating
conditions, either continued irradiation or use of sacrificial
olefins, present practical disadvantages as compared to the
established iridium-catalyzed methods. Here we describe new
families of cobalt alkyl complexes supported by different classes
Figure 1. Evaluation of cobalt pincer complexes for the catalytic
borylation of five-membered heterocycles with HBPin. Reported values
are % conversions with respect to the five-membered heterocycle as
determined by gas chromatography using mesitylene as an internal
standard. Product ratios were determined by NMR spectroscopy, and
the values in parentheses are isolated yields.
Received: January 27, 2014
Published: March 3, 2014
© 2014 American Chemical Society
4133
dx.doi.org/10.1021/ja500712z | J. Am. Chem. Soc. 2014, 136, 4133−4136

Journal of the American Chemical Society
Communication
that quantitative conversion was observed in the same time
frame, suggesting that more electron-donating ligands22 increase
the rate of catalytic C−H functionalization.
To generate a more electron-rich base metal center, the
supporting pincer ligand was modified by replacing the relatively
π-acidic imine arms with more σ-donating phosphines. The
cobalt alkyl complexes, 3 and 4 were synthesized by
straightforward alkylation of the corresponding cobalt chloride
complexes,23−25 and the solid-state structure of 3 was confirmed
by X-ray diffraction (Figure S1). The [CH2SiMe3] hydrocarbyl
substituent was chosen due to its commercial availability and
demonstrated success in cobalt catalysis.26 In the solid state, 3 is
1
diamagnetic yet the H NMR spectrum recorded in toluene-d8
exhibits paramagnetically shifted and broadened resonances
(Figures S2 and S3). Cooling the sample to −75 °C resulted in
sharpening of the peaks and shifting to expected diamagnetic
references values. This behavior is likely a result of thermal
population of a low-lying triplet state upon warming.
Both 3 and 4 were active for the borylation of 2-methylfuran,
reaching >98% and 40% conversion after 72 h, respectively.
Importantly, no solvent was needed, and efficient conversion and
high yields were observed with stoichiometric quantities of
HBPin. Cobalt precatalyst 3 also functions using B2Cat2 (Cat =
catechol) as the boron source. In the presence of 1 mol % of 3,
the selective borylation of 2-methylfuran was accomplished in
THF solution over the course of 1 h at 80 °C.27,28
Figure 2. Cobalt-catalyzed borylation of pyridines with 3. Conditions A
with 0.5 equiv B2Pin2 and B with 1.0 equiv of B2Pin2. Reported numbers
are % conversions of the corresponding pyridine derivative determined
by GC analysis using mesitylene as an internal standard, and the value in
parentheses is the isolated yield. Product ratios were determined by
NMR analysis. Values under the % conversions are selectivities, and the
numbers in parentheses denote the position of borylation. Selectivities
were the same for both conditions A and B.
To further evaluate catalyst performance, the borylation of a
family of representative five-membered heterocycles containing
oxygen, sulfur, and nitrogen was studied with 2 and 3. The
phosphine-substituted pincer precatalyst, 3, was superior in each
case, likely due to a more electron-rich cobalt center, which
facilitates oxidative addition of the C−H bonds of the substrate.
Up to 5000 turnovers have been observed with the borylation of
methyl furan-2-carboxylate at 23 °C over the course of 6 h in the
presence of 0.02 mol % of 3. Isolated yields for the borylation
products of 2-methylfuran, methyl furan-2-carboxylate, and
benzofuran were 91%, >99%, and >99%, respectively. The
bis(imino)pyridine compound, 2 proved incompatible with
sulfur-containing heterocycles perhaps a result of increased
electrophilicity. Precatalyst 2 was effective for the borylation of
furans and N-methyl indole, although in some cases competing
substrate reduction was observed, consistent with the known
hydrogenation activity of this class of cobalt alkyl.29,30 With 3,
exclusive borylation of the position adjacent to the heteroatom
was observed with each substrate correlating with the relative
acidity of the C−H bonds.4,31 For parent furan and thiophene,
the borylation of the initial product was sufficiently rapid that the
2,5-disubstituted product was also observed.
The performance of 3 for the catalytic borylation of a family of
substituted pyridines was also examined. These substrates were
of interest due to their prevalence in pharmaceuticals32 and the
enabling impact that direct C−H functionalization methods have
on elaboration of lead compounds in medicinal chemistry. Each
catalytic reaction was conducted in a 0.55 M THF solution of
substrate in the presence of B2Pin2 and 3 mol % 3 at 80 °C
(Figure 2). Experiments in solution were selected over neat
conditions as substituted pyridines are often solids or valuable
and therefore sometimes available in only limited quantities.
Two sets of catalytic conditions were selected that varied by
the ratio of B2Pin2 to the substrate. The first, termed condition A,
employed 0.5 equiv of B2Pin2 (1 BPin unit per pyridine) and
generated H2 as the reaction byproduct. Using this protocol,
HBPin is formed during catalytic turnover and is known to be a
less effective borylation agent, accounting for the reduced activity
of the catalyst. These conditions are useful, however for the
differentiation of the relative rates of reaction among the different
pyridines. Condition B employed 1 equiv of B2Pin2 (2 BPin
equivalents per pyridine) relative to substrate and was designed
to maintain the diboron species as the active borylating agent,
forming HBPin as a byproduct.
The cobalt-catalyzed borylation of 2,6-lutidine with 3 mol % of
3 selectively borylated the 4-position of the ring, consistent with
functionalization of the most sterically accessible C−H bond.
Notably, 2,4-lutidine and 2-methoxy-4-methyl pyridine also
underwent smooth catalytic borylation in the presence of 3 to
selectively yield the 2-borylated products. To our knowledge, a
successful iridium-catalyzed borylation of these two substrates
has not been published. In our hands, applying the reported
published conditions for the iridium-catalyzed borylation of 2,6-
lutidine33 for the 2,4-isomer furnished a complex mixture of
products. The desired 2-borylated product accounted for only
28% of the mixture, highlighting one of the benefits of the cobalt-
catalyzed method. One curiosity of the cobalt method is the
higher yields observed with condition A during the same time
frame as condition B. The origin of this anomaly under
investigation.
Pyridines bearing substituents at the 2-position such as methyl,
methoxy, dimethylamino, or phenyl also underwent smooth
catalytic borylation, all yielding monoborylated products as a
mixture of positional isomers (Figure 2). In each case, the aryl
boronate ester resulting from borylation of the 4-position was the
major product. For 2-phenylpyridine, this product was obtained
in an equimolar ratio to the 5-position isomer. No borylation of
the phenyl ring was observed under these conditions. For the Me
and NMe2 functionalized substrates, small quantities of 6-
position borylation was observed, demonstrating the ability of
cobalt to functionalize positions adjacent to sp2 hybridized
nitrogen atoms.
The cobalt-catalyzed borylation of pure hydrocarbon arenes
was also studied with 3 (Figure 3). Conditions employed a 20:1
4134
dx.doi.org/10.1021/ja500712z | J. Am. Chem. Soc. 2014, 136, 4133−4136

Journal of the American Chemical Society
Communication
catalytic borylation, and monitoring borylations by NMR
spectroscopy reveals this compound is the catalyst resting state.
A deuterium-labeling experiment was also conducted. The
cobalt-catalyzed borylation of 2-methylfuran was conducted with
DBPin in the presence of 1 mol % of 3. Analysis of the starting
furan and the resulting borylated product by quantitative 13C
NMR spectroscopy at 40% conversion established no deuterium
incorporation into either organic molecule. Analysis of the
isotopic composition of the remaining pinacolborane revealed
formation of HBPin, and sampling of the headspace and analysis
by 1H NMR spectroscopy also confirmed formation of HD gas
(Figure S6). It is also likely that D2 gas was formed. These results
are consistent with reversible oxidative-addition and reductive
elimination of the B−H(D) bonds.
A working mechanism consistent with the experimental
observations is presented in Figure 4. Following formation of
Figure 3. Arene borylation with 3 and B2Pin2. Reported values are %
conversions of B2Pin2 determined by GC using mesitylene as an internal
standard. Product ratios were determined via NMR analysis. Isolated
yields are reported in parentheses.
ratio of arene to B2Pin2 and 1 mol % of 3 at 80 °C for 24 h. The
reported conversions are based on the consumption of B2Pin2,
and no other boron products were observed. The borylation of
benzene and toluene proceeded smoothly. For the latter, a 70:30
selectivity of meta- and para-substituted products was obtained.
The functionalization of meta-xylene required a slightly higher
catalyst loading of 3 mol % and exclusively occurred at the 5-
position of the ring. These data are consistent with borylation
occurring at the most sterically accessible C−H bonds, selectivity
likely arising from the C−H oxidative addition step in the
catalytic cycle. Electron-poor arenes such as fluorobenzene were
borylated more readily than electron-rich examples such as
anisole likely as a result of the relative acidity of the C−H bonds,
similar to the trend observed in iridium-catalyzed reactions.4
Unfortunately, reasonable turnover was observed for this
substrate class only when the reactions were conducted in neat
arene.
Figure 4. Proposed mechanism for cobalt-catalyzed borylation with 3
using methylfuran as a representative substrate.
trans-(iPrPNP)CoH2(BPin), reductive elimination of H2 either by
isomerization to the cis isomer or by dissociation of a phosphine
ligand36 generates the active species in the cycle. The
isomerization pathway is shown for illustrative purposes. The
1
observation of a broad cobalt hydride signal by H NMR
The rich catalytic chemistry associated with 3 prompted
investigation into the mechanism of turnover. Treatment of a
diethyl ether solution of the cobalt alkyl compound with 2 equiv
of HBPin resulted in loss of one equivalent of Me3SiCH2BPin
with concomitant formation of a new cobalt species identified as
spectroscopy at 23 °C suggests that a dynamic process may be
operative. Following H2 elimination, the resulting cobalt(I) boryl
complex, (iPrPNP)CoBPin promotes the oxidative addition of a
C−H bond of the substrate.37 Reductive elimination of the B−C
bond furnishes the observed product and oxidative addition of
trans-(iPrPNP)CoH2(BPin). The toluene-d8 1H NMR spectrum
exhibits a single broadened hydride resonance at −8.71 ppm at
23 °C; cooling the same solution below 0 °C resulted in
sharpening of the signal to a well-resolved triplet (Figure S5).
Single 31P and 11B NMR resonances were observed at 103.5 and
41.1 ppm, respectively, consistent with formation of a single
isomer of the cobalt product. Single crystals suitable for X-ray
diffraction were obtained from a concentrated diethyl ether
solution stored at −35 °C. The data were of sufficient quality
such that the two cobalt-hydrides were located and freely refined
providing definitive structural evidence for formation of a
Co(III) oxidative addition product and the trans-dihydride
isomer. Notably, this species is an active catalyst precursor for
HBPin regenerates cis-(iPrPNP)CoH2(BPin). It is also plausible
that the cobalt hydride (iPrPNP)CoH formed from B−H
reductive elimination from (iPrPNP)CoH2(BPin) could promote
the oxidative addition of the C−H bond. An alternative pathway
highlighting this possibility is presented in Figure S7.
One notable feature of the proposed mechanism is that the
base metal-catalyzed C−H functionalization operates via a
Co(I)−(III) redox couple. These electronic structure require-
ments are distinct from iridium, where a cycle involving Ir(III)−
(V) intermediates is now well-established.4,7,35 The cobalt
intermediate responsible for C−H activation, likely the boryl
4135
dx.doi.org/10.1021/ja500712z | J. Am. Chem. Soc. 2014, 136, 4133−4136

Journal of the American Chemical Society
Communication
compound (iPrPNP)CoBPin but also possibly (iPrPNP)CoH, is
four-coordinate and coordinatively unsaturated and presents the
opportunity, by virtue of vacant coordination sites, for substrate
directed reactions.34 These mechanistic principles serve as the
centerpiece for further interrogation and future catalyst develop-
ment and optimization.
(24) Khaskin, E.; Diskin-Posner, Y.; Weiner, L.; Leitus, G.; Milstein, D.
Chem. Commun. 2013, 49, 2771.
(25) Benito-Garagorri, D.; Kirchner, K. Acc. Chem. Res. 2008, 41, 201.
(26) Friedfeld, M. R.; Shevlin, M.; Hoyt, J. M.; Krska, S. W.; Tudge, M.
T.; Chirik, P. J. Science 2013, 342, 1076.
(27) Boller, T. M.; Murphy, J. M.; Hapke, M.; Ishiyama, T.; Miyaura,
N.; Hartwig, J. F. J. Am. Chem. Soc. 2005, 127, 14263.
(28) Liskey, C. W.; Wei, C. S.; Pahls, D. R.; Hartwig, J. F. Chem.
Commun. 2009, 45, 5603.
(29) Knijnenburg, Q.; Horton, A. D.; Heijden, H. V. D.; Kooistra, T.
M.; Hetterscheid, D. G. H.; Smits, J. M. M.; de Bruin, B.; Budzelaar, P. H.
M.; Gal, A. W. J. Mol. Catal. A: Chem. 2005, 232, 151.
(30) Monfette, S.; Turner, Z. R.; Semproni, S. P.; Chirik, P. J. J. Am.
Chem. Soc. 2012, 134, 4561.
(31) Tajuddin, H.; Harrisson, P.; Bitterlich, B.; Collings, J. C.; Sim, N.;
Batsanov, A. S.; Cheung, M. S.; Kawamorita, S.; Maxwell, A. C.; Shukla,
L.; Morris, J.; Lin, Z.; Marder, T. B.; Steel, P. G. Chem. Sci. 2012, 3, 3505.
(32) DeGoey, D. A.; Grampovnik, D. J.; Flentge, C. A.; Flosi, W. J.;
Chen, H.-j.; Yeung, C. M.; Randolph, J. T.; Klein, L. L.; Dkehtyar, T.;
Colletti, L.; Marsh, K. C.; Stoll, V.; Mamo, M.; Morfitt, D. C.; Nguyen,
B.; Schmidt, J. M.; Swanson, S. J.; Mo, H.; Kati, W. M.; Molla, A.; Kempf,
D. J. J. Med. Chem. 2009, 52, 2571.
(33) Murphy, J. M.; Liao, X.; Hartwig, J. F. J. Am. Chem. Soc. 2007, 129,
15434.
(34) Tamura, H.; Yamazaki, H.; Sato, H.; Sakaki, S. J. Am. Chem. Soc.
2003, 125, 16144.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental details and crystallographic data. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
pchirik@princeton.edu
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Princeton University for financial support. S.P.S.
thanks the Natural Sciences and Engineering Research Council
of Canada for a predoctoral fellowship. We also thank Dr. Crisita
Atienza for the initial preparation of 1 and Boropharm for a
generous gift of B2Pin2.
(35) Preshlock, S. M.; Plattner, D. L.; Maligres, P. E.; Krska, S. W.;
Maleczka, R. E.; Smith, M. R. Angew. Chem., Int. Ed. 2013, 52, 12915.
(36) Xu, H.; Bernskoetter, W. H. J. Am. Chem. Soc. 2011, 133, 14956.
(37) Semproni, S. P.; Atienza, C. C. H.; Chirik, P. J. Chem. Sci.; DOI:
10.1039/C4SC00255E.
REFERENCES
■
(1) Godula, K.; Sames, D. Science 2006, 312, 67.
(2) Rouquet, G.; Chatani, N. Angew. Chem., Int. Ed. 2013, 52, 11726.
(3) McMurray, L.; O’Hara, F.; Gaunt, M. J. Chem. Soc. Rev. 2011, 40,
1885.
(4) Mkhalid, I.; Barnard, J. H.; Marder, T. B.; Murphy, J. M.; Hartwig, J.
F. Chem. Rev. 2010, 110, 890.
(5) Hall, D. G. Boronic Acids; Wiley-VCH: Weinheim, Germany, 2005.
(6) Dick, G. R.; Woerly, E. M.; Burke, M. D. Angew. Chem., Int. Ed.
2012, 51, 2667.
(7) Hartwig, J. F. Acc. Chem. Res. 2011, 45, 864.
(8) Preshlock, S. M.; Ghaffari, B.; Maligres, P. E.; Krska, S. W.;
Maleczka, R. E.; Smith, M. R. J. Am. Chem. Soc. 2013, 135, 7572.
(9) Kawamorita, S.; Murakami, R.; Iwai, T.; Sawamura, M. J. Am. Chem.
Soc. 2013, 135, 2947.
(10) Kawamorita, S.; Miyzaki, T.; Iwai, T.; Ohmiya, H.; Sawamura, M.
J. Am. Chem. Soc. 2012, 134, 12924.
(11) Kawamorita, S.; Miyzaki, T.; Ohmiya, H.; Iwai, T.; Sawamura, M.
J. Am. Chem. Soc. 2011, 133, 19310.
(12) Yinghuai, Z.; Chenyan, K.; Peng, A. T.; Emi, A.; Monalisa, W.;
Louis, L. K.; Hosmane, N. S.; Maguire, J. A. Inorg. Chem. 2008, 47, 5756.
(13) Chirik, P. J.; Weighardt, K. Science 2010, 327, 5967.
(14) Hatanaka, T.; Ohki, Y.; Tatsumi, K. Chem. Asian J. 2010, 5, 1657.
(15) Yan, G.; Jiang, Y.; Kuang, C.; Wang, S.; Liu, H.; Zhang, Y.; Wang,
J. Chem. Commun. 2010, 46, 3170.
(16) Mazzacano, T. J.; Mankad, N. P. J. Am. Chem. Soc. 2013, 135,
17258.
(17) Obligacion, J. V.; Chirik, P. J. Org. Lett. 2013, 15, 2680.
(18) Zhang, L.; Peng, D.; Leng, X.; Huang, Z. Angew. Chem., Int. Ed.
2013, 52, 3676.
(19) Greenhalgh, M. D.; Thomas, S. P. Chem. Commun. 2013, 49,
11230.
(20) Obligacion, J. V.; Chirik, P. J. J. Am. Chem. Soc. 2013, 135, 19107.
(21) Knijnenburg, Q.; Hetterscheid, D.; Kooistra, T. M.; Budzelaar, P.
H. M. Eur. J. Inorg. Chem. 2004, 1204.
(22) Bowman, A. C.; Milsmann, C.; Bill, E.; Lobkovsky, E.;
Weyhermuller, T.; Wieghardt, K.; Chirik, P. J. Inorg. Chem. 2010, 49,
̈
6110.
(23) Trovitch, R. J.; Lobkovsky, E.; Chirik, P. J. Inorg. Chem. 2006, 45,
7252.
4136
dx.doi.org/10.1021/ja500712z | J. Am. Chem. Soc. 2014, 136, 4133−4136