Nickel hydroxo complexes as intermediates in nickel-catalyzed Suzuki-Miyaura cross-coupling

Article abstract of DOI:10.1021/om5001327

The synthesis, characterization, and reactivity of intermediates formed in the Ni-catalyzed Suzuki-Miyaura cross-coupling (SMC) of an aryl chloride are described. Oxidative addition of 1-chloro-4-trifluoromethylbenzene (1) to a mixture of Ni(cod)₂ and PCy₃ afforded NiCl(4-CF ₃Ph)(PCy₃)₂ (2), which then cleanly provided dimeric [Ni(4-CF₃Ph)(U-OH)(PCy₃)]₂ (3) by reaction with aqueous KOH. Reactivity studies of 2 and 3 with phenylboronic acid (4) revealed that, while 2 affords only traces of the biphenyl coupling product after 24 h, the same reaction with 3 is complete within minutes at room temperature. In contrast, the reaction of 3 with potassium phenyltrihydroxyborate (6) is much slower than that with boronic acid 4, and significantly lower yields of the cross-coupling product are obtained. We show that formation of the hydroxo species 3 is the rate-determining step in the present SMC.

Full text of DOI:10.1021/om5001327

Communication
pubs.acs.org/Organometallics
Nickel Hydroxo Complexes as Intermediates in Nickel-Catalyzed
Suzuki−Miyaura Cross-Coupling
Alec H. Christian,† Peter Muller,‡ and Sebastien Monfette*,†
̈
†Worldwide Research and Development, Pfizer Inc., Eastern Point Road, Groton, Connecticut 06340, United States
‡Department of Chemistry 6-331, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, United States
S
* Supporting Information
ABSTRACT: The synthesis, characterization, and reactivity of
intermediates formed in the Ni-catalyzed Suzuki−Miyaura cross-
coupling (SMC) of an aryl chloride are described. Oxidative
addition of 1-chloro-4-trifluoromethylbenzene (1) to a mixture
of Ni(cod)2 and PCy3 afforded NiCl(4-CF3Ph)(PCy3)2 (2),
which then cleanly provided dimeric [Ni(4-CF3Ph)(μ−OH)-
(PCy3)]2 (3) by reaction with aqueous KOH. Reactivity studies
of 2 and 3 with phenylboronic acid (4) revealed that, while 2
affords only traces of the biphenyl coupling product after 24 h,
the same reaction with 3 is complete within minutes at room
temperature. In contrast, the reaction of 3 with potassium phenyltrihydroxyborate (6) is much slower than that with boronic acid
4, and significantly lower yields of the cross-coupling product are obtained. We show that formation of the hydroxo species 3 is
the rate-determining step in the present SMC.
ransition-metal-mediated cross-coupling reactions have
halides22,24 but has remained poorly studied. It has been
Ttransformed the way chemists assemble molecules. While proposed that complexation of the base anion to the metal
center precedes transmetalation during the Ni-catalyzed SMC
of aryl halides,22,25 although observation or isolation of such
intermediates has, to the best of our knowledge, not been
reported. Convincing reports by Hartwig and Amatore have
described the intermediacy of Pd−hydroxo complexes as the
species undergoing transmetalation in Pd-catalyzed SMC
carried out in the presence of water.26,27 A similar pathway
could be operative for Ni and may explain the differences in
yield when anhydrous vs hydrated bases are used.28 As part of
our ongoing efforts to understand the mechanism of Ni-
catalyzed SMC for future pharmaceutical applications, we
became interested in the potential role of water within this
transformation. Herein we report the syntheses of intermedi-
ates formed during a Ni-catalyzed SMC of an aryl halide and
explore their reactivity toward a typical arylboronic acid and a
borate. This study reveals that the role of water in the Ni-
catalyzed SMC is to form a Ni−hydroxo complex that
undergoes rapid transmetalation with the boronic acid.
As a starting point in our study, we chose Ni(cod)2 and PCy3
as the catalyst system, given its importance in the SMC of both
aryl halides and activated phenols.25b We selected 1-chloro-4-
trifluoromethylbenzene (1) as the electrophile because the
electron-withdrawing trifluoromethyl group should facilitate
oxidative addition while also serving as a useful NMR handle.
Addition of 1 to a red-orange THF solution of Ni(cod)2 and
PCy3 at room temperature caused an instantaneous color
the vast majority of the catalysts used in such transformations
are centered around a Pd atom, there has been significant
interest in the use of cheaper, more abundant first-row
elements1 such as Ni,2−10 Fe,11 Cu,12−16 Co,11,17,18 Mn,19
and Cr20 as alternatives to Pd. As Suzuki−Miyaura cross-
coupling (SMC) is the most heavily used cross-coupling
reaction in the pharmaceutical industry,21 we became attracted
by the potential advantages of replacing Pd with Ni in the
preparation of pharmaceutically relevant structures. However,
despite the plethora of examples documented in a laboratory
setting, only a few examples of Ni-catalyzed cross-coupling have
been used on a process-relevant scale (>100 mmol),21 none of
which were SMC transformations. There are two major
drawbacks to the application of Ni catalysis in SMC on scale:
(1) the relatively high catalyst loading that is normally
employed (3−10 mol % Ni),4 which tends to offset the
generally lower cost of Ni vs Pd and can potentially complicate
the isolation of cross-coupling products with low residual metal
content, and (2) the limited substrate scope, particularly for
heteroaryl substrates. Importantly, recent advances in catalyst
design show encouraging results in broadening the substrate
scope while maintaining low catalyst loadings.4
Understanding the mechanism of the Ni-catalyzed SMC is
key to improving the efficiency of the reaction and could result
in superior catalyst turnover numbers. The mechanism of this
reaction is believed to proceed in a way analogous to that for
Pd: oxidative addition, transmetalation, and then reductive
elimination.22,23 Of these steps, transmetalation is thought to be
the turnover-limiting step in the Ni-catalyzed SMC of aryl
Received: February 5, 2014
Published: April 18, 2014
© 2014 American Chemical Society
2134
dx.doi.org/10.1021/om5001327 | Organometallics 2014, 33, 2134−2137

Organometallics
Communication
change to yellow-orange (Scheme 1). Formation of the
anticipated NiCl(4-CF3Ph)(PCy3)2 (2) was complete within
be responsible for the low conversion observed. Indeed, when a
yellow-orange THF solution of 2 was treated with 20 equiv of
aqueous KOH (final ratio 20:1 THF:H2O), a faint color change
to yellow was observed after stirring for 2 h at room
temperature; conversion to the new diamagnetic product 3
averaged 75% at that time. A prolonged reaction time of 18 h
showed complete conversion. As examination of the crude
reaction mixture by 31P{1H} NMR spectroscopy only revealed
free PCy3 in addition to 3 (1:1 ratio); we propose that 3 is a
dinuclear compound of the type [Ni(4-CF3Ph)(μ-OH)-
(PCy3)]2. This structure is supported by NMR, IR, and
combustion analyses and is also consistent with previously
reported dimeric Ni complexes.32−34 Scaling up the reaction to
1.5 g of 2 allowed isolation of 3 as a fine yellow powder in 95%
Scheme 1. Synthesis of Ni Complexes 2 and 3
1
12 h, as judged by 31P{1H} NMR analysis. Complex 2 was
isolated in 71% yield after workup; its structure is supported by
NMR, X-ray, and combustion analyses. Diamagnetic 2 is
indefinitely stable under a dinitrogen atmosphere and is stable
for at least 4 months as a solid in air at room temperature, as
yield after workup. The H NMR spectrum of 3 shows a
diagnostic resonance for the bridging hydroxo protons at −3.80
ppm in C6D6, in good agreement with other Ni complexes
containing bridging hydroxo ligands.32,33 A single, sharp
resonance for the hydroxide protons implies that 3 is present
only as the anti isomer in C6D6.35 Addition of up to 50 equiv of
PCy3 did not promote anti/syn isomerization or the formation
of a monomeric hydroxide complex of the type Ni(4-
CF3Ph)(OH)(PCy3)2. Similar results were obtained by
Boncella and co-workers for the related [Ni(Mes)(μ-OH)-
(PMe3)]2.33 Infrared spectroscopy revealed a strong absorption
at 3650 cm−1 (Nujol) that we assign to the νO−H band of 3.
With complexes 2 and 3 in hand, we wanted to confirm that
they are competent catalysts for the SMC of aryl halide 1 and
phenylboronic acid (4). Thus, formation of 4-trifluoromethyl-
biphenyl (5) was monitored over time using 5 mol % of
Ni(cod)2/PCy3 as a benchmark catalyst system in addition to
complexes 2 and 3 (Figure 2). Reactions were carried out at
1
judged by H and 31P NMR spectroscopy.
X-ray-quality crystals of 2 were obtained by cooling a
concentrated diethyl ether solution to −35 °C for 2 days. The
asymmetric unit contains one molecule of 2 and one molecule
of diethyl ether (Figure 1). Complex 2 adopts a slightly
Figure 1. Thermal ellipsoid plot of 2 with ellipsoids at the 50%
probability level. Hydrogen atoms and diethyl ether solvent molecule
are omitted for clarity. Key bond lengths (Å) and angles (deg):
Ni(1)−P(1), 2.2492(3); Ni(1)−P(2), 2.2645(3); Ni(1)−C(1),
1.8844(9); Ni(1)−Cl(1), 2.2295(3); C(1)−Ni(1)−Cl(1), 168.05(3);
P(1)−Ni(1)−P(2), 167.840(11); C(1)−Ni(1)−P(1), 91.65(3);
C(1)−Ni(1)−P(2), 93.73(3).
distorted square planar geometry, as evidenced by the C−Ni−
Cl and P−Ni−P bond angles of 168.1 and 167.8°, respectively,
both of which are significantly shy of the ideal 180°. These
angles are nevertheless fully consistent with a related complex
recently reported by Standley and Jamison.29 The remaining
bond lengths and angles around the Ni center are similar to
those of related complexes.30
Initial attempts to replace the chloride atom within 2 with
MOH (M = Li, Na, K, Cs) showed incomplete conversion
along with the formation of an unidentified, insoluble black
residue after stirring for 24 h at both room temperature and 50
°C (see the Supporting Information). Campora commented on
the difficult salt metathesis reaction between NiCl(Me)(dippe)
(dippe = 1,2-bis(diisopropylphosphino)ethane) and MOH.31
We hypothesized that the low solubility of MOH in THF could
Figure 2. Yield of 5 over time. Conditions: 5 mol % of [Ni], 180 mM
1, 1.5 equiv of 4, 3 equiv of K3PO4·H2O, THF, 22 °C, N2 atmosphere.
Values are 3% in replicate runs.
room temperature in THF using 1.5 equiv of 4 and 3 equiv of
K3PO4·H2O. Samples were removed periodically and quenched
with DMSO, and the yield of biphenyl 5 was assessed
quantitatively by ultraperformance liquid chromatography
(UPLC) against an internal standard. With all three catalysts,
the yield of biphenyl 5 reaches 75% and 100% at 24 and 48 h,
respectively. The rates of formation of biphenyl 5 as a function
of time were identical within experimental error for all three
2135
dx.doi.org/10.1021/om5001327 | Organometallics 2014, 33, 2134−2137

Organometallics
Communication
catalyst systems. No induction period was noticed with either 2
or 3, consistent with the presence of both complexes “on” the
catalytic cycle.
Having established the ability of 2 and 3 to catalyze a SMC
reaction, we next studied the stoichiometric reaction of 2 and 3
with boronic acid 4 to better understand the individual steps of
the catalytic cycle (Figure 3). When the halide complex 2 was
reaction of a Ni−hydroxo complex with a boronic acid. This
result is fully consistent with the results of Hartwig and
Amatore using a Pd catalyst.26,27 The relevance of the hydroxo
species 3 during the SMC between aryl chloride 1 and 4 was
established by reaction of the oxidative addition product 2 with
15 equiv of K3PO4·H2O in THF-d8. The initially orange
solution of complex 2 became slightly paler over 15 h, and
31P{1H} NMR analysis revealed the formation of hydroxo 3 in
15% yield (Figure S1, Supporting Information).38 The only
other species detected by 31P NMR spectroscopy were
unreacted 2 and free PCy3, the coproduct from the formation
of dimeric 3. The low yield of 3 obtained using K3PO4·H2O is
consistent with the large excess of KOH required for its
isolation from 2 in good yield (vide supra). The high reactivity
of hydroxo 3 with boronic acid 4 (Figure 3) coupled with the
observation of halide 2 as the catalyst resting state by 31P{1H}
NMR spectroscopy (Figure S2) suggests that formation of
hydroxo 3 is the turnover-determining step in the catalytic
cycle. Moreover, conversion of 2 to 3, and the formation of 5
from chloride 2 and boronic acid 4 (Figure 3), both occur over
a time scale of hours. This supports that formation of 3
accounts for the time scale of catalysis. Thus, strategies targeted
at accelerating the chloride to hydroxide ligand exchange,
possibly through ancillary ligand design, may help in improving
reaction rate and perhaps reducing catalyst loadings.
The foregoing describes high-yield routes to catalytically
relevant intermediates in the SMC of aryl halide 1. Comparison
of the reactivities of halide complex 2 and hydroxo 3 revealed
that only the latter undergoes transmetalation at an appreciable
rate. We also showed that the boronic acid undergoes
transmetalation, while the borate is substantially slower and is
sometimes unreactive. The present study revealed future
avenues to explore such as ways to accelerate the rate of
formation of the key Ni−hydroxo complexes through ligand
design. Such experiments are currently underway and will be
reported in due course.
Figure 3. Yield of 5 over time. Conditions: 17 mM [Ni], 1.5 equiv of 4
(solid line) or 6 and 18-crown-6 (dashed line), THF, 22 °C, N2
atmosphere. Values are 3% in replicate runs.
treated with boronic acid 4 in the absence of base at both room
temperature and 60 °C, only a trace amount of biphenyl 5 was
detected at 6 h. Longer reaction times did not improve
conversions. Similarly, the use of rigorously dried K3PO4 led to
no observable reaction after 24 h at both room temperature and
60 °C. In sharp contrast to the reactivity of 2, a quantitative
yield of biphenyl 5 was obtained in less than 2 min at room
temperature when hydroxo complex 3 was treated with boronic
acid 4 even in the absence of base and/or added water.36
Consistent with this observation, the reaction of complex 2
with boronic acid 4 in the presence of 3 equiv of K3PO4·H2O
reaches full conversion at 6 h at room temperature. These
results clearly show the importance of water in promoting the
SMC reaction. It has been proposed that water may promote
the reaction by formation of a borate species, such as potassium
trihydroxyborate 6, which could then undergo transmetalation
with the Ni complex.22 Our results suggest that this is unlikely
as (1) it was previously demonstrated by Lloyd-Jones and co-
workers that the equilibrium between the boronic acid and
borate favors the former at low water concentration,37 as is the
case here, and (2) reaction of either 2 or 3 with borate 6 and
18-crown 6 (to solubilize 6) leads to significantly less product
than with boronic acid 4 (Figure 3). The reduced reactivity is
not due to base-mediated protodeboronation,37 as analysis of a
quenched sample by UPLC showed residual boronic acid but
no traces of benzene, the product from protodeboronation of 6.
However, we do note that the color of the reaction mixture
changed from the original yellow-orange to a very pale green
using borate 6, potentially signaling the formation of catalyti-
cally inactive Ni(OH)2 species.28
ASSOCIATED CONTENT
* Supporting Information
■
S
Text, figures, tables, and a CIF file giving general consid-
erations, synthetic procedures, conversions of 2 to 3,
representative procedures for catalytic reactions and for
stoichiometric experiments, 31P{1H} NMR spectra of the
reaction of 2 with K3PO4·H2O, 31P{1H} NMR spectra of the
SMC of 1 with 4, and crystallographic data for 2. This material
is available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*E-mail for S.M.: sebastien.monfette@pfizer.com.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors acknowledge the following Pfizer colleagues: Javier
Magano, Chris McWilliams, and Joel Hawkins for helpful
discussions, Brian Samas for assistance with the crystallographic
analysis, Ronald Morris for assistance with mass spectrometry
analysis, and Joshua Denette for analytical support. This work
has been influenced through the ongoing efforts of the Non-
Precious Metal Catalysis Alliance among Pfizer, Boehringer-
Ingelheim, and GlaxoSmithKline.
The above data suggest that the kinetically more accessible
pathway to transmetalation in a Ni-catalyzed SMC involves
2136
dx.doi.org/10.1021/om5001327 | Organometallics 2014, 33, 2134−2137

Organometallics
Communication
(27) Amatore, C.; Jutand, A.; Le Duc, G. Chem. Eur. J. 2011, 17,
2492−2503.
REFERENCES
■
(1) For recent reviews on the use of non-precious metals in cross-
coupling, see: (a) Han, F.-S. Chem. Soc. Rev. 2013, 42, 5270−5298.
(b) Mesganaw, T.; Garg, N. K. Org. Process Res. Dev. 2013, 17, 2029−
2039. (c) Jana, R.; Pathak, T. P.; Sigman, M. S. Chem. Rev. 2011, 111,
1417−1492. (d) Knochel, P.; Thaler, T.; Diene, C. Isr. J. Chem. 2010,
50, 2547−2557.
(2) Nielsen, D. K.; Huang, C.-Y.; Doyle, A. G. J. Am. Chem. Soc.
2013, 135, 13605−13609.
(3) Huang, C.-Y.; Doyle, A. G. J. Am. Chem. Soc. 2012, 134, 9541−
9544.
(4) Ge, S.; Hartwig, J. F. Angew. Chem., Int. Ed. 2012, 51, 12837−
12841.
(5) Binder, J. T.; Cordier, C. J.; Fu, G. C. J. Am. Chem. Soc. 2012,
134, 17003−17006.
(6) Wisniewska, H. M.; Swift, E. C.; Jarvo, E. R. J. Am. Chem. Soc.
2013, 135, 9083−9090.
(28) Inada, K.; Miyaura, N. Tetrahedron 2000, 56, 8657−8660.
(29) Standley, E. A.; Jamison, T. F. J. Am. Chem. Soc. 2013, 135,
1585−1592.
(30) A search of the Cambridge crystallographic database for
complexes of the type NiCl(Ar)(PR3)2 revealed average Ni−P, Ni−C,
and Ni−Cl bond lengths of 2.196(9), 1.892(6), and 2.219(6) Å,
respectively.
(31) Campora, J.; Matas, I.; Palma, P.; Graiff, C.; Tiripicchio, A.
Organometallics 2005, 24, 2827−2830.
(32) Carmona, E.; Marin, J. M.; Palma, P.; Paneque, M.; Poveda, M.
L. Inorg. Chem. 1989, 28, 1895−1900.
(33) VanderLende, D. D.; Abboud, K. A.; Boncella, J. M. Inorg. Chem.
1995, 34, 5319−5326.
(34) We were unsuccessful in obtaining corroborating evidence for
the dimeric structure of 3 by mass spectrometry and thus cannot
completely rule out higher nuclearity species.
(7) (a) Ramgren, S. D.; Hie, L.; Ye, Y.; Garg, N. K. Org. Lett. 2013,
15, 3950−3953. (b) Leowanawat, P.; Resmerita, A.-M.; Moldoveanu,
C.; Liu, C.; Zhang, N.; Wilson, D. A.; Hoang, L. M.; Rosen, B. M.;
Percec, V. J. Org. Chem. 2010, 75, 7822−7828. (c) Leowanawat, P.;
Zhang, N.; Safi, M.; Hoffman, D. J.; Fryberger, M. C.; George, A.;
Percec, V. J. Org. Chem. 2012, 77, 2885−2892.
(35) In THF-d8, a 7:1 anti:syn isomeric ratio is observed.
(36) Karl Fischer titration of the THF solvent revealed a residual
water content of 9−12 ppm. Assuming 1 equiv of water is consumed
per turnover, the amount of water provided by the solvent would only
allow for formation of 13% of biphenyl product 5.
(37) Butters, M.; Harvey, J. N.; Jover, J.; Lennox, A. J. J.; Lloyd-Jones,
G. C.; Murray, P. M. Angew. Chem., Int. Ed. 2010, 49, 5156−5160.
(38) Integration against an internal standard showed loss of about
10−15% of the total integration, suggesting either decomposition and/
or formation of unidentified paramagnetic complex(es).
(8) Chen, G.-J.; Han, F.-S. Eur. J. Org. Chem. 2012, 2012, 3575−
3579.
(9) Wang, Z.-X.; Liu, N. Eur. J. Inorg. Chem. 2012, 2012, 901−911.
(10) Liu, C.; Tang, S.; Liu, D.; Yuan, J.; Zheng, L.; Meng, L.; Lei, A.
Angew. Chem., Int. Ed. 2012, 51, 3638−3641.
(11) Kuzmina, O. M.; Steib, A. K.; Markiewicz, J. T.; Flubacher, D.;
Knochel, P. Angew. Chem., Int. Ed. 2013, 52, 4945−4949.
(12) Ren, P.; Stern, L.-A.; Hu, X. Angew. Chem., Int. Ed. 2012, 51,
9110−9113.
(13) Kabir, M. S.; Lorenz, M.; Namjoshi, O. A.; Cook, J. M. Org. Lett.
2010, 12, 464−467.
(14) Ziegler, D. T.; Choi, J.; Munoz-Molina, J. M.; Bissember, A. C.;
Peters, J. C.; Fu, G. C. J. Am. Chem. Soc. 2013, 135, 13107−13112.
(15) Saha, D.; Chatterjee, T.; Mukherjee, M.; Ranu, B. C. J. Org.
Chem. 2012, 77, 9379−9383.
(16) Uyeda, C.; Tan, Y.; Fu, G. C.; Peters, J. C. J. Am. Chem. Soc.
2013, 135, 9548−9552.
(17) Araki, K.; Inoue, M. Tetrahedron 2013, 69, 3913−3918.
(18) Li, B.; Wu, Z.-H.; Gu, Y.-F.; Sun, C.-L.; Wang, B.-Q.; Shi, Z.-J.
Angew. Chem., Int. Ed. 2011, 50, 1109−1113.
(19) Cahiez, G.; Duplais, C.; Buendia, J. Angew. Chem., Int. Ed. 2009,
48, 6731−6734.
(20) Steib, A. K.; Kuzmina, O. M.; Fernandez, S.; Flubacher, D.;
Knochel, P. J. Am. Chem. Soc. 2013, 135, 15346−15349.
(21) Magano, J.; Dunetz, J. R. Chem. Rev. 2011, 111, 2177−2250.
(22) Liu, L.; Zhang, S.; Chen, H.; Lv, Y.; Zhu, J.; Zhao, Y. Chem.
Asian J. 2013, 8, 2592−2595.
(23) Lei, X.; Obregon, K. A.; Alla, J. Appl. Organomet. Chem. 2013,
27, 419−424.
(24) Transmetalation is also thought to be the rate-determining step
in the SMC of aryl esters, carbamates, and sulfamates. See:
(a) Quasdorf, K. W.; Antoft-Finch, A.; Liu, P.; Silberstein, A. L.;
Komaromi, A.; Blackburn, T.; Ramgren, S. D.; Houk, K. N.; Snieckus,
V.; Garg, N. K. J. Am. Chem. Soc. 2011, 133, 6352−6363. (b) Li, Z.;
Zhang, S.-L.; Fu, Y.; Guo, Q.-X.; Liu, L. J. Am. Chem. Soc. 2009, 131,
8815−8823.
(25) Percec and Han also proposed that complexation of the base
anion to the metal center precedes transmetalation during the SMC of
activated phenols. See: (a) Percec, V.; Bae, J.-Y.; Hill, D. H. J. Org.
Chem. 1995, 60, 1060−1065. (b) Rosen, B. M.; Quasdorf, K. W.;
Wilson, D. A.; Zhang, N.; Resmerita, A.-M.; Garg, N. K.; Percec, V.
Chem. Rev. 2011, 111, 1346−1416. (c) Chen, G.-J.; Huang, J.; Gao, L.-
X.; Han, F.-S. Chem. Eur. J. 2011, 17, 4038−4042.
(26) Carrow, B. P.; Hartwig, J. F. J. Am. Chem. Soc. 2011, 133, 2116−
2119.
2137
dx.doi.org/10.1021/om5001327 | Organometallics 2014, 33, 2134−2137