Advancing Base-Metal Catalysis: Development of a Screening Method for Nickel-Catalyzed Suzuki-Miyaura Reactions of Pharmaceutically Relevant Heterocycles

Article abstract of DOI:10.1021/acs.oprd.1c00210

Interest in replacing palladium catalysts with base metals resulted in the development of a 24-reaction screening platform for identifying nickel-catalyzed Suzuki-Miyaura reaction conditions. This method was designed to be directly applicable to process s

Full text of DOI:10.1021/acs.oprd.1c00210

pubs.acs.org/OPRD
Article
Advancing Base-Metal Catalysis: Development of a Screening
Method for Nickel-Catalyzed Suzuki−Miyaura Reactions of
Pharmaceutically Relevant Heterocycles
Matthew J. Goldfogel,* Xuelei Guo, Jeishla L. Meléndez Matos, John A. Gurak, Jr., Matthew V. Joannou,
William B. Moffat, Eric M. Simmons, and Steven R. Wisniewski
Cite This: https://doi.org/10.1021/acs.oprd.1c00210
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Interest in replacing palladium catalysts with base metals resulted in the development of a 24-reaction screening
platform for identifying nickel-catalyzed Suzuki−Miyaura reaction conditions. This method was designed to be directly applicable to
process scale-up by employing homogeneous reaction conditions alongside stable and inexpensive nickel(II) precatalysts and has
proven to be broadly suitable for complex heterocyclic substrates relevant to bioactive molecules. These advances were enabled by
the key discovery that a methanol additive greatly improves the reaction performance and enables the use of organic-soluble amine
bases. The screening platform and scale-up workflow were applied to a representative cross-coupling using the antipsychotic
perphenazine and enabled the rapid development of a gram-scale synthesis that highlighted the utility of this method and the
advantages of nickel catalysis for metal remediation.
KEYWORDS: nickel, Suzuki−Miyaura, cross-coupling, base metals, process chemistry, high-throughput experimentation,
heterocycle coupling
Scheme 1. Comparison of the Nickel- and Palladium-
Catalyzed Suzuki−Miyaura Reactions for Process Chemistry
INTRODUCTION
■
The importance of the Suzuki−Miyaura (SM) reaction to
modern organic synthesis is difficult to overestimate. Defined
as the coupling of an organo (pseudo)halide with an
organoboron reagent, its utility for forming aryl−aryl linkages
via palladium catalysis has fundamentally altered approaches to
the synthesis of active pharmaceutical ingredients (APIs).1,2
Currently, palladium-catalyzed SM reactions dominate the
industrial landscape and remain a highly active area of
academic research. However, there is an established interest
in exploring nickel catalysis as an alternative to palladium
catalysis for the SM reaction (Scheme 1A).3,4 Replacing a
precious metal (Pd) with an earth-abundant first-row transition
metal (Ni) is particularly advantageous to process chemistry,
where the benefits of reduced cost and ease of sourcing are
amplified on the multikilogram scale.5 In addition to these
practical benefits, nickel-catalyzed SM reactions have been
demonstrated with less reactive electrophilic coupling partners
that are challenging with palladium.8,9 The smaller size and
decreased electronegativity of nickel relative to palladium
enables oxidative addition to weaker electrophiles3 and has
been effectively leveraged in recent couplings of aryl
chlorides,10 fluorides,11,12 sulfonates,8,13,14 sulfamates,15 carbo-
nates,16 amides,17 and amines.18 Conversely, these benefits to
reactivity and scope are accompanied by challenges in
robustness and reliability, as nickel catalysts are more prone
to substrate inhibition and undesired side reactions such as
protodehalogenation or protodeborylation (Scheme 1B).19
Pharmaceutical molecules are particularly notable for the
prevalence of heterocycles and polar functional groups that can
a
,
Special Issue: Excellence in Industrial Organic Syn-
thesis 2021
Received: June 3, 2021
https://doi.org/10.1021/acs.oprd.1c00210
Org. Process Res. Dev. XXXX, XXX, XXX−XXX
© XXXX American Chemical Society
A

Organic Process Research & Development
pubs.acs.org/OPRD
Article
complicate the SM reaction. Heteroaryl and electron-poor
boronic acids are more susceptible to protodeborylation, and
nickel-catalyzed SM reactions are influenced by polar func-
tional groups more than the corresponding reactions with
palladium.20 As a result, nickel-catalyzed SM coupling
reactions with heterocyclic substrates are less developed
relative to those with simple arenes, and the state of the art
in scope is best exemplified in work of Hartwig21 and
Stradiotto,20 in which an impressive array of biheteroaryl
products were efficiently synthesized. However, the application
of these methods to process scale is limited by the use of
costly, preformed nickel complexes derived from expensive and
highly air-sensitive Ni(cod)2 or organometallic reagents.22,23
This counteracts the improvements in reaction cost and
scalability gained from replacing palladium with nickel.
Similarly, the vast majority of nickel-catalyzed SM reactions
employ carbonate or phosphate bases in organic solvents,
resulting in heterogeneous reactions that make robust scale-up
challenging.3
The maturity of palladium-catalyzed SM reactions and their
well-understood mechanism have made them the default
approach for industrial syntheses of APIs, yet the expanded
scope and reduced cost offered by nickel are strong drivers for
its adoption. Our interest in developing base-metal catalysis for
process chemistry applications led us to consider the
limitations of current methodology with respect to the
development of robust, cost-effective commercial processes.
Common literature conditions generally include elements that
are not ideal for scale-up (i.e., unstable and/or expensive Ni(0)
precatalysts and heterogeneous reaction conditions), and the
difficulty in coupling heterocyclic substrates suggests an
inherent challenge in utilizing base-metal catalysis. To that
end, we sought to overcome these hurdles by designing a
screening platform to specifically address process chemistry
concerns and applications. This would encompass key drivers
for implementing a nickel-catalyzed SM process, including (1)
use of an inexpensive, readily available nickel(II) salt, (2)
homogeneous reaction conditions, and (3) broad suitability for
heterocyclic substrates.
Herein is reported the invention of a rapid screening
platform to assess the viability of the nickel-catalyzed SM
reaction for nonstandard substrate pairs and a workflow to
scale the resulting conditions (Scheme 1C).24 This screening
workflow was successfully applied to the coupling of heteroaryl
(pseudo)halides with heteroarylboronic acids and used to
synthesize a variety of drug-like molecules. Additionally, the
relevance to process chemistry was demonstrated by scaling an
initial hit to a multigram-scale coupling of the antipsychotic
perphenazine.25 These advances were enabled by the key
discovery that the inclusion of methanol significantly improves
the reaction performance and enables homogeneous reaction
conditions with organic-soluble bases. Additional mechanistic
studies were conducted to examine the speciation of the active
nickel catalyst and suggest that a Ni(0)/Ni(II) cycle is active,
as the presence of an amine base promotes disproportionation
of any Ni(I) species to Ni(0) and Ni(II).
Figure 1. Structure of Tyk2 inhibitor clinical candidate BMS-986202.
The biaryl heterocycle targeted for analogue synthesis is highlighted
by the red dashed box.
SM reaction could be developed, (hetero)biaryl 3 was selected
as a challenging and representative target. Chloride coupling
partner 2 was chosen as the least reactive of the halide series,
and the heteroaryl fluoride was included to demonstrate
selectivity for the oxidative addition and provide an NMR
handle for reaction analysis (eq 1).
Initial screening demonstrated that the nickel-catalyzed SM
reaction was possible while adhering to our process-scalability
objectives; employing the soluble amine base 2-tert-butyl-
1,1,3,3-tetramethylguanidine (tBuTMG) and the inexpensive,
bench-stable precatalyst NiCl2·6H2O ($14/mol) provided 3 in
b
86 area percent (AP) based on LC−MS analysis. Inspired by
the observation of Stradiotto20 and our own studies in nickel-
catalyzed borylation,24 the inclusion of 10 vol % MeOH was
found to be vital to obtain high conversion. This discovery
suggested that a general method for homogeneous nickel-
catalyzed SM reactions might be achievable.
The initial coupling conditions discovered for 2 addressed
the key drivers for developing a robust, scalable Ni-catalyzed
SM reaction and provided a starting point for the design of a
nickel-catalyzed SM screening platform. On the basis of the
considerations outlined previously (vide supra), an effective
screening platform would need to accomplish several high-level
design goals: (1) the screen must use minimal resources to
survey the broadest possible reaction space; (2) the screen
must identify the reaction variables with the largest impact on
reaction success; (3) the platform must provide a general
workflow that functions for many substrate types; and (4) the
reaction conditions obtained from the screen must be a useful
starting point for process development and scale-up. On a
reaction-to-reaction basis, overcoming substrate variations
would require time-intensive optimization, and because no
two reactions are equivalent, we did not anticipate finding a
single universal set of conditions. This highlighted the
advantage of developing a general screening platform as a
tool for interrogating the reaction space in order to maximize
the likelihood of identifying viable reaction conditions as a
starting point for further optimization.
RESULTS AND DISCUSSION
■
Initial explorations of nickel-catalyzed SM coupling reactions
for the Bristol Myers Squibb portfolio focused on a case study
in the synthesis of analogues to the biaryl heterocycle of the
tyrosine kinase 2 (Tyk2) inhibitor BMS-986202 (Figure
1).26,27 To explore whether a homogeneous nickel-catalyzed
B
https://doi.org/10.1021/acs.oprd.1c00210
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
pubs.acs.org/OPRD
Article
With these design principles in mind, we explored the
reaction space of the model system (eq 1) through high-
throughput experimentation (HTE). This rapidly provided a
large data set comparing the class variables of ligand, solvent,
alcohol additive, base, and precatalyst (see the Supporting
Information (SI) for full data sets).28,29 The product 3 was
readily identified by LC−MS and provided a good qualitative
assessment since few byproducts related to the pyrimidine
chloride were observed. Likewise, the formation of anisole was
tracked as an important byproduct related to the known
instability of arylboronates in the presence of nickel catalysts.24
Comparison of product formation (3) and protodeborylation
(anisole) demonstrated that various mono- and bidentate
phosphine ligands (PPh3, DPPE, DPPF, DCPP, DCEPhos,
CgMe-PPh, and A-caPhos) provided improved reactivity
relative to the control case with no ligand (Figure 2).
Conversely, N-donor ligands (tB-BiPy, Phen, 4Me-Tpy)
showed little efficacy and increased protodeborylation.
efficiency were observed when the amount or identity of the
alcohol additive was varied (Figure 3). An alcohol concen-
Figure 3. Solvent composition effects on nickel-catalyzed Suzuki−
Miyaura coupling. Reaction conditions: NiCl2·6H2O (4.5 mol %),
ligand (10 mol % for monodentate, 5 mol % for bidentate), 1 (10
μmol), 2 (1.0 equiv), DBU (3.0 equiv), 2-MeTHF/alcohol (0.1 M),
70 °C.
tration of 10 vol % proved to be optimal, as the yields
decreased in the absence of MeOH (12−18 AP) and varied
widely when 40 vol % was employed (28−75 AP). Changing
the alcohol additive showed that MeOH significantly out-
performed water (H2O), ethanol (EtOH), 2,2,2-trifluorethanol
(TFE), isopropanol (iPrOH), and tert-amyl alcohol (t-
AmOH). The unexpected importance of the alcohol additive
suggested a mechanistic role, which we hypothesized to be due
to the increased Lewis acidity of the methoxy boronate ester
relative to the boronic acid, as described by Denmark et al.30
Presumably, this increase in Lewis acidity would provide a
more active transmetalating species. The identity of the alcohol
also influenced the amount of protodeborylation observed,
with MeOH forming the least anisole side product.
Development of a General Screening Platform for
Nickel-Catalyzed Suzuki−Miyaura Coupling. The diver-
sity of reactions for nickel-catalyzed SM couplings show that
the optimal conditions are highly substrate-dependent.3 Since
the goal of this work is to rapidly assess whether a nickel-
catalyzed SM reaction is appropriate for a unique substrate pair
and provide a starting point for reaction optimization via a
general workflow, the optimal conditions for the formation of 3
are less relevant than the trends in key class variables.
Unsurprisingly, HTE identified the ligand as the most
influential class variable and supported our expectation that
solubility is important for efficient product formation. The
identity and concentration of the MeOH additive significantly
impacted the yield but provided limited optionality since 10
vol % MeOH was optimal for all three ligands.
Figure 2. Results from the ligand screen for the nickel-catalyzed
Suzuki−Miyaura coupling to form 3. Reaction conditions: NiCl2·
6H2O (4.5 mol %), ligand (10 mol % for monodentate, 5 mol % for
bidentate), 1 (10 μmol), 2 (1.0 equiv), DBU (3.0 equiv), 2-MeTHF/
MeOH (0.1 M, 9:1), 70 °C.
On the basis of the large impact of the ligand, further class
variable screening was performed using three ligands in parallel
(DCPP, DPPF, and PPh3) to ensure that the observed trends
were consistent across different ligand types (e.g., bidentate vs
monodentate, aryl- vs alkylphosphine). Solvent comparisons
showed similarly high conversions provided that the substrates
were fully soluble at the reaction temperature (see SI Figure
3). Solvents that did not fully solubilize the substrates gave
poor conversion, as evidenced by the ability of toluene to
provide only 6−16 AP of 3 across three ligands, whereas
homogeneous reactions with EtOAc, MeCN, NMP, and
DMAc provided 77−99 AP.
Soluble amine bases, such as 1,8-diazabicyclo(5.4.0)undec-7-
ene (DBU) and tBuTMG, were similarly effective (94−100 AP
across three ligands) and reliably outperformed heterogeneous
K2CO3 (32−40 AP) (see SI Figure 4). This discovery provided
homogeneous reaction conditions that address the process
challenges associated with previously reported nickel-catalyzed
SM reactions that employ insoluble carbonate or phosphate
bases in nonpolar solvents such as toluene or dioxane.3 The
number of equivalents of amine base had a small impact on the
reaction performance, and a decrease from 3.0 to 1.0 equiv
reduced product formation by an average of 14 AP across the
three ligands. Conversely, large differences in reaction
The design goals for the screening platform (vide supra)
necessitate minimizing the experimental burden for assessing
the reactivity of a substrate and prompted us to limit screening
C
https://doi.org/10.1021/acs.oprd.1c00210
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
pubs.acs.org/OPRD
Article
Scheme 2. Reaction Conditions for the Screening Platform and Workflow for Substrate Assessment
to a 24-reaction array. Twelve phosphine ligands that showed
promising reactivity during HTE screening were selected to
include a variety of ligand types (monodentate, bidentate,
alkyl- vs arylphosphine) and properties (bite angle, donor
strength, steric bulk) that survey a broad range of chemical
space (Scheme 2A). These ligands were crossed with the two
successful solvents from the HTE screen with the largest
polarity difference to ideally provide homogeneous reaction
conditions for the greatest range of substrates. Other reaction
parameters were kept consistent across the screen and selected
to maximize their relevance to process scale-up: (i) DBU was
selected as the base, as it is inexpensive and more widely
available than tBuTMG; (ii) NiCl2·6H2O was selected for its
low cost, availability, and similar performance to other Ni(II)
sources (see SI Figure 5); and (iii) reactions were run at an
elevated temperature of 80 °C for 18 h to ensure high
conversion with more challenging heteroaryl substrates.
We next sought to confirm that the selected screening
conditions were relevant to a range of substrates by applying
them to a secondary test case. The coupling of 4 and 5 was
selected, as the electronics of the coupling partners are
reversed relative to the reaction to form 3: 5 is an electron-rich
heteroarene compared with pyrimidine 1, while switching the
methoxy group from the ortho position to the meta position in
4 changes it from an electron-donating to an electron-
withdrawing substituent (σp = −0.27, σm = 0.12).31 Selected
HTE screening of the ligand, solvent, and base (see the SI) was
used to tune and confirm the 12 ligands and two solvents
selected for the 24-reaction screening platform (Scheme 2A).
Applying this finalized screen generated 6 in 55 AP when
DPPF was used as the ligand with DMAc as the solvent
(Scheme 2B). The top conditions for 6 differ from those
identified for 3, which emphasizes the importance of
maintaining a broad screen rather than seeking a single general
method. This test case validated the screening protocol by
providing process-amenable conditions that can be optimized
for future reaction development.
screening platform to identify promising substrates (Scheme
2C). To minimize material requirements, the nickel-catalyzed
SM screen is run at the 10 μmol scale, and the top conditions
for successful reactions are then tested at 1 mmol to confirm
the reproducibility and scalability (Scheme 2B). The
procedure for the 10 μmol screen begins with catalyst
formation and is followed by addition of the substrates,
reaction solvent, and base prior to sealing the reaction mixture
under a nitrogen atmosphere. Reactions are set up in a
glovebox to avoid oxygen exposure, although the reaction
mixture is minimally oxygen-sensitive prior to base addition.
This workflow was validated for the scale-up and isolation of
6 and used to identify common reaction byproducts of
protodehalogenation (7), C−O coupling with methanol (8),
and dimerization (9). Further studies confirmed that these
byproducts are common across substrates and account for the
mass balance of the heteroaryl halide. Varying the identity of
the boronate indicated that the boronic acid is more reactive
than corresponding boronate ester and that these couplings do
not share optimal reaction conditions (Table 1); application of
the screening platform to p-methoxyphenylboronic acid (10),
neopentylglycolatoboronate 11, and pinacolatoboronate 12
generated 13 in 80, 48, and 31 AP, respectively. On this basis,
Table 1. Comparison of the Reactivities of Boronic Acid,
Neopentylglycolatoboronate, and Pinacolatoboronate under
the Screening Conditions
With the goal of enabling process scale-up for nickel-
catalyzed SM couplings, a workflow was developed to apply the
D
https://doi.org/10.1021/acs.oprd.1c00210
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
pubs.acs.org/OPRD
Article
Figure 4. Application of the screening platform to various aryl- and heteroarylboronic acids. General reaction conditions: NiCl2·6H2O (4.5 mol %),
ligand (10 mol % for monodentate, 5 mol % for bidentate), aryl halide 5 (10 μmol), arylboronic acid (1.0 equiv), DBU (3.0 equiv), solvent/MeOH
(0.1 M, 9:1), 80 °C. Values in parentheses represent results from 1 mmol reactions using the scale-up workflow. AP refers to LC area percentage of
the desired product relative to 5 and its associated side products. Isolated yields are denoted with a percentage symbol. See the SI for full details.
further studies were carried out using atom-efficient boronic
acids, although the screening platform also functions with
boronate esters.
boronic acid failed to provide any screening hits, which
suggests that the nucleophilic coupling partner is highly
sensitive to sterics (see SI Figure 6).
Application and Scope of the Nickel-Catalyzed
Suzuki−Miyaura Screening Method. We sought to
establish the generality of the 24-reaction screening platform
by testing an array of commercially available arylboronic acids
(Figure 4). Electronic trends were studied by comparing para-
substituted phenylboronic acids, and the results showed that
electron-rich arenes (13−15, 78−83 AP) outperform electron-
poor substrates (16−19, 30−45 AP), which matches trends
observed in previously reported nickel-catalyzed SM reac-
tions.32,33 Functional group tolerance was studied, and lead
conditions were identified for a broad range of substrates.
Arylboronic acids bearing free amines or alcohols gave C−C
coupling products 20−23 with moderate to good conversions
(57−82 AP), and those containing a nitrile (24, 59 AP), ether
(25, 88 AP), or alkene (28, 48 AP) provided screening hits.
Inclusion of an aldehyde resulted in reduced conversion (29,
12 AP), which was attributed to deleterious addition to the
electrophilic aldehyde and the increased steric constraints of
ortho substitution. Vinylboronic acids behaved similarly to
their aryl counterparts, and screening hits were found for both
exocyclic (26, 61 AP) and endocyclic (27, 58 AP) substrates.
Heterocyclic boronic acids were explored, as they are highly
relevant to structures found in APIs.34,35 N-Heterocycles
masked by a tert-butoxycarbonyl (Boc) group (30, 53 AP; 31,
41 AP) or bearing an N-methyl group (32, 82 AP; 33, 45 AP)
gave modest hits, with pyrrole and pyrazole derivatives
providing incomplete conversion. However, reactions with
unprotected imidazole or pyrazole substrates failed to generate
product (see SI Figure 6). Heterocycles with the heteroatom
adjacent to the boronic acid were challenging substrates, as
demonstrated by comparison of thiophenes 34 and 35. These
arenes differed only in their substitution pattern but provided
22 and 63 AP, respectively, which is attributed to the instability
of the boronic acid based on observed protodeborylation.
Despite reduced conversions, the 24-reaction screening
platform consistently provided lead conditions for hetero-
cycles, including furan (36, 29 AP), pyridine (37, 38 AP),
pyrimidine (38, 28 AP), and isoquinoline (39, 44 AP).
Substrates containing multiple substituents ortho to the
The importance of scalability to the utility of this screening
tool prompted us to test several of these reactions using the 1
mmol workflow to confirm reproducibility and robustness.
These yields are reported in parentheses in Figure 4 and show
that comparable conversions were obtained in most cases; the
greatest deviation was observed for 25, which was isolated in
48% yield compared with the screening result of 88 AP.
Overall, reaction conditions leading to product formation were
identified for the majority of explored arylboronic acid
substrates, and the 24-reaction screening workflow reliably
provided lead conditions for boronic acids that are likely to be
amenable to nickel-catalyzed SM coupling. Of equal
importance to our work, the screen saved experimental time
and effort by rapidly identifying challenging substrates that
would be more amenable to traditional palladium catalysis.
Next the range of electrophilic coupling partners was
explored. Screening focused on heteroarenes and drug-like
molecules that have been underdeveloped in nickel-catalyzed
SM couplings.20,21 Variations in the halide leaving group had a
minimal effect, with the halide series (Cl, Br, I) forming 6 in
56−60 AP. The minimal differences in conversion suggest that
the reaction is minimally influenced by oxidative addition,
which matches reports that nickel-catalyzed SM reactions can
employ weaker electrophiles.3 However, utilizing triflate as the
leaving group improved the conversion to 87 AP, primarily as a
result of reduced formation of side product 8. Heteroaryl
bromides and chlorides were selected for further studies
because of their ubiquity as coupling partners and relevance to
the synthesis of clinical candidates.
To ensure variety in the products, three arylboronic acids (4,
54, and 61) were crossed with a range of heteroaryl halides.
Couplings with arylboronic acid 4 were highly tolerant to
variations in the electrophilic coupling partner, and few
substrates failed to provide lead reaction conditions. N-
Heterocycles were explored, and pyridyl (40, 89 AP; 41, 25
AP; 42, 93 AP), (iso)quinoline (43, 74 AP; 44, 98 AP), and
carbazole (45, 48 AP) derivatives all provided hits. However,
increasingly electron-poor arenes resulted in reduced product
formation, as evidenced by the reduced conversions for
E
https://doi.org/10.1021/acs.oprd.1c00210
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
pubs.acs.org/OPRD
Article
Figure 5. Application of the screening platform to various aryl and heteroaryl electrophiles. General reaction conditions: NiCl2·6H2O (4.5 mol %),
ligand (10 mol % for monodentate, 5 mol % for bidentate), aryl halide (10 μmol), arylboronic acid (1.0 equiv), DBU (3.0 equiv), solvent/MeOH
(0.1 M, 9:1), 80 °C. Values in parentheses represent results from 1 mmol reactions using the scale-up workflow. AP refers to LC area percentage of
the desired product relative to the aryl halide and its associated side products. Isolated yields are denoted with a percentage symbol. aThe reaction
was run with 4.0 equiv of DBU to free-base the clomipramine·HCl salt. See the SI for full details.
Figure 6. Scale-up reaction scheme and process flow diagram for nickel-catalyzed Suzuki−Miyaura coupling.
diazene 46 (21 AP), triazene 47 (13 AP), and quinoxaline 48
(16 AP). This observation is supported by comparing 2-pyridyl
substrates 40 (89 AP) and 42 (93 AP) with inductively
withdrawing 3-pyridyl 41 (25 AP). Studies with heterocyclic
boronic acid 54 demonstrated that benzofuran 56 and
benzothiophene 57 are amenable to nickel catalysis, with hits
of 93 and 90 AP, respectively. The selectivity of oxidative
addition was demonstrated with 55, which was formed in 76
AP with complete selectivity for the aryl bromide over the
fluoride.
F
https://doi.org/10.1021/acs.oprd.1c00210
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
pubs.acs.org/OPRD
Article
Figure 7. Mechanistic studies of nickel-catalyst speciation using preformed nickel complexes and EPR analysis.
Screens pairing 4 with bioactive aryl halides were tested, and
sulfonamide-containing APIs AK-7 and diazoxide reacted to
provide 49 and 50 in 98 AP. APIs incorporating complex
heterocycles and basic functional groups were also quite
successful, and good hits were obtained with azelastine (51, 72
AP), rivaroxaban (52, 81 AP), and lapatinib (53, 41 AP).
Drug-like molecules containing aryl chlorides were tested with
model boronic acids 54 and 61 and showed no decrease in
conversion relative to aryl bromides; products derived from
clomipramine (58, 96 AP), fenofibrate (59, 92 AP; 62, 65 AP),
and perphenazine (68, 99 AP) all provided hits with high
conversion. Unexpectedly, SM coupling of glibenclamide
occurred with methanolysis to form the free sulfonamide,
affording 64 in 81 AP. Similar methanolysis was observed as
minor byproducts with 59 and 62, which suggests that nickel
catalysts may promote transesterification or transamidation.
This off-cycle reactivity should be considered as a potential risk
when evaluating a substrate pair for nickel-catalyzed SM
coupling.
Similar to the boronic acid scope, the 1 mmol scale-up
workflow was applied to several substrates to demonstrate
scalability, and the isolated yields (shown in parentheses in
Figure 5) were in line with the conversions observed in the
leading 24-well-plate screen. This substrate scope resulted in a
variety of lead conditions across substrates and demonstrates
the importance of using a screening platform rather than a
single set of optimized reaction conditions for rapid and
accurate assessment of the viability of nickel-catalyzed SM
reactions. The 24-reaction platform quickly identified com-
pounds and substrate classes that are amenable to nickel
catalysis and provided a streamlined approach to replacing
costly palladium with earth-abundant nickel.
To further demonstrate the applicability of this workflow to
process development and API synthesis, a 6 g scale coupling
reaction of the antipsychotic perphenazine (65) to form 60
was developed as a case study (Figure 6).25 Optimization of
the reaction conditions identified in the 24-reaction screen
(Figure 5) showed that a slight increase in the number of
equivalents of the arylboronic acid coupling partner signifi-
cantly improved the conversion and reduced the reaction time,
while the process mass intensity (PMI) was improved by
increasing the reaction concentration.36,37 To further improve
the PMI, a procedure for isolating 60 by crystallization was
developed. Workup began with a simple aqueous wash that
effectively removed nickel salts to <100 ppm, which is a
notable advantage as palladium remediation often requires
additional scavengers.38 Isolation was accomplished by a
solvent swap to dioxane, acidification with HCl, and filtration
of the resulting precipitate. The resulting solid was purified by
recrystallization from methanol and toluene to isolate the
desired compound in 83% yield.
Mechanistic Investigation of Nickel Speciation for
Homogeneous Nickel-Catalyzed Suzuki−Miyaura Cou-
pling with Amine Base. To gain a greater understanding of
catalyst speciation throughout the Ni-catalyzed SM cross-
coupling using DBU/MeOH conditions, several mechanistic
G
https://doi.org/10.1021/acs.oprd.1c00210
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
pubs.acs.org/OPRD
Article
studies were conducted. As an initial probe, the performances
of isolated Ni(0), Ni(I), and Ni(II) precatalysts were evaluated
in the coupling reaction. As depicted in Figure 7A, with 5 mol
% (DPPF)NiBr2, 25 was produced in 67 AP, while 5 mol %
(DPPF)NiBr and (DPPF)Ni(1,5-cod) afforded the product in
64 and 68 AP, respectively. This indicated that a common
nickel intermediate is likely accessed under the reaction
conditions regardless of the identity of the starting precatalyst,
a phenomenon previously observed in other Ni-catalyzed
Suzuki−Miyaura cross-coupling methodologies.39,40
To quantify the amount of Ni(I) generated or consumed in
these Ni-catalyzed cross-coupling reactions, aliquots of each
reaction presented in Figure 7 were removed at various time
points and analyzed by X-band electron paramagnetic
resonance (EPR) spectroscopy at both 298 and 77 K (see
the SI for low-temperature spectra). No significant signal was
detected at any point during the reaction (the 4 h time point is
shown in Figure 7B) at either 298 or 77 K, even when
(DPPF)NiBr was utilized as the precatalyst. On the basis of
these results, it is unlikely that any Ni(I) species are present in
substantial quantities during the catalytic reaction, in contrast
to what has previously been observed in other Ni-catalyzed
Suzuki−Miyaura methodologies, where off-cycle bisphosphine
Ni(I) species are present in significant amounts.39,40 However,
the presence of EPR-silent Ni(I) dimers engaging in
ferromagnetic coupling cannot be ruled out.
As (DPPF)NiBr is known to undergo disproportionation
reactions in the presence of certain ligands,39 we attempted to
determine which component of the Ni-catalyzed cross-
coupling reaction may be responsible for Ni(I) degradation.
The complex was stable in DMAc/MeOH (9:1) at 80 °C as
judged by X-band EPR spectroscopy, and the addition of 1,5-
cyclooctadiene as a potential Ni(0) trapping ligand did not
lead to reaction of the Ni(I) halide complex. However, the
addition of excess DBU and 1,5-cyclooctadiene to a solution of
(DPPF)NiBr led to disproportionation to (DPPF)Ni(1,5-cod)
and (DPPF)NiBr2 as determined by 1H and 31P NMR
spectroscopy (Scheme 3). On the basis of these results and
applied to the coupling of many drug-like aryl halides with
(hetero)arylboronic acids, and their relevance to process
chemistry was further demonstrated by the optimization and
scale-up of a nickel-catalyzed SM coupling of the drug
perphenazine. The reported screening platform provides a
reliable tool for developing process-relevant nickel-catalyzed
SM coupling reactions and will be applied in future syntheses
of active pharmaceutical ingredients.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.oprd.1c00210.
■
sı
Experimental details, data tables from HTE screening,
characterization data, NMR spectra, HRMS spectra, and
EPR spectra (PDF)
AUTHOR INFORMATION
Corresponding Author
■
Matthew J. Goldfogel − Chemical Process Development,
Bristol Myers Squibb Company, New Brunswick, New Jersey
08903, United States; orcid.org/0000-0002-6911-3068;
Email: matthew.goldfogel@bms.com
Authors
Xuelei Guo − Chemical Process Development, Bristol Myers
Squibb Company, New Brunswick, New Jersey 08903,
United States
Jeishla L. Meléndez Matos − Chemical Process Development,
Bristol Myers Squibb Company, New Brunswick, New Jersey
08903, United States
John A. Gurak, Jr. − Chemical Process Development, Bristol
Myers Squibb Company, New Brunswick, New Jersey 08903,
United States
Matthew V. Joannou − Chemical Process Development, Bristol
Myers Squibb Company, New Brunswick, New Jersey 08903,
United States; orcid.org/0000-0002-0079-7107
William B. Moffat − Chemical Process Development, Bristol
Myers Squibb Company, New Brunswick, New Jersey 08903,
United States
Eric M. Simmons − Chemical Process Development, Bristol
Myers Squibb Company, New Brunswick, New Jersey 08903,
United States; orcid.org/0000-0002-3854-1561
Steven R. Wisniewski − Chemical Process Development,
Bristol Myers Squibb Company, New Brunswick, New Jersey
08903, United States; orcid.org/0000-0001-6035-4394
Scheme 3. Disproportionation of the Catalytically Relevant
Nickel Complex Promoted by Amine Base
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.oprd.1c00210
the lack of any signal in the X-band EPR spectrum of the
reaction mixture, it is likely that nickel is restricted to a 0/II
cycle under the DBU/MeOH reaction conditions. Any Ni(I)
species formed in solution would be converted into Ni(0) and
Ni(II), which can then re-enter the catalytic cycle.
Author Contributions
The manuscript was written through contributions of all
authors. All of the authors approved the final version of the
manuscript.
CONCLUSION
Notes
■
The authors declare no competing financial interest.
We have developed and demonstrated the utility of a reaction
screening platform for identifying nickel-catalyzed SM coupling
conditions. This platform was specifically designed to be
amenable to process chemistry applications and obviates the
challenges of unstable nickel catalysts and heterogeneous
reaction conditions by utilizing a Ni(II) precatalyst and a
soluble amine base with methanol as a key additive. The
screening platform and scale-up workflow were successfully
ACKNOWLEDGMENTS
■
The authors thank the Isolation and Structural Elucidation
Group at Bristol Myers Squibb (New Brunswick, NJ) for their
analytical support and Dr. Jennifer Albaneze-Walker for the
assistance with coordinating compound isolation. The authors
also gratefully acknowledge the support and encouragement of
H
https://doi.org/10.1021/acs.oprd.1c00210
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
pubs.acs.org/OPRD
Article
Fluorosulfonates with Aryl Boronic Acids Enabled by Sulfuryl
Fluoride. ACS Catal. 2015, 5 (9), 5041−5046.
(14) Xing, T.; Zhang, Z.; Da, Y.-X.; Quan, Z.-J.; Wang, X.-C. Ni-
Catalyzed Suzuki−Miyaura Coupling Reactions of Pyrimidin-2-yl
Phosphates, Tosylates and Pivalates with Arylboronic Acids.
Tetrahedron Lett. 2015, 56 (46), 6495−6498.
(15) 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. Suzuki−Miyaura Cross-Coupling of Aryl Carbamates
and Sulfamates: Experimental and Computational Studies. J. Am.
Chem. Soc. 2011, 133 (16), 6352−6363.
(16) Barth, E. L.; Davis, R. M.; Mohadjer Beromi, M.; Walden, A.
G.; Balcells, D.; Brudvig, G. W.; Dardir, A. H.; Hazari, N.; Lant, H. M.
C.; Mercado, B. Q.; Peczak, I. L. Bis(Dialkylphosphino)Ferrocene-
Ligated Nickel(II) Precatalysts for Suzuki−Miyaura Reactions of Aryl
Carbonates. Organometallics 2019, 38 (17), 3377−3387.
(17) Shi, S.; Meng, G.; Szostak, M. Synthesis of Biaryls through
Nickel-Catalyzed Suzuki−Miyaura Coupling of Amides by Carbon−
Nitrogen Bond Cleavage. Angew. Chem., Int. Ed. 2016, 55 (24),
6959−6963.
(18) Cao, Z.-C.; Xie, S.-J.; Fang, H.; Shi, Z.-J. Ni-Catalyzed Cross-
Coupling of Dimethyl Aryl Amines with Arylboronic Esters under
Reductive Conditions. J. Am. Chem. Soc. 2018, 140 (42), 13575−
13579.
(19) Cox, P. A.; Leach, A. G.; Campbell, A. D.; Lloyd-Jones, G. C.
Protodeboronation of Heteroaromatic, Vinyl, and Cyclopropyl
Boronic Acids: pH−Rate Profiles, Autocatalysis, and Disproportiona-
tion. J. Am. Chem. Soc. 2016, 138 (29), 9145−9157.
(20) Sawatzky, R. S.; Stradiotto, M. (DPEPhos)Ni(Mesityl)Br: An
Air-Stable Pre-Catalyst for Challenging Suzuki−Miyaura Cross-
Couplings Leading to Unsymmetrical Biheteroaryls. Synlett 2018, 29
(6), 799−804.
(21) Ge, S.; Hartwig, J. F. Highly Reactive, Single-Component
Nickel Catalyst Precursor for Suzuki−Miyaura Cross-Coupling of
Heteroaryl Boronic Acids with Heteroaryl Halides. Angew. Chem., Int.
Ed. 2012, 51 (51), 12837−12841.
the CPD Senior Leadership Team for base metals research and
development.
ADDITIONAL NOTES
■
a
As pointed out by a reviewer, a complication of implementing
nickel catalysis in process is that nickel chloride, nickel nitrate,
and other nickel salts are classified as restricted substances
under the REACH regulations of the European Union (EU)
and European Chemicals Agency (ECHA), for which certain
uses are restricted in the EU, potentially limiting the
applications of nickel catalysis in an industrial setting in
Europe. The assessment of metal toxicity depends on a number
of factors, including oxidation state, solubility, and properties
of their environment, and several excellent discussions of the
topic have been published.6,7
b
In keeping with standard industrial practice for optimization,
in-process yields are assessed using “area percentage” (AP)
yields. AP values are uncorrected for the difference in LC
response factor for the starting aryl halide versus its boronic
acid derivative. These data, while approximate, give insight into
general trends and aid in determining optimal conditions for
lead reactions.
REFERENCES
■
(1) Brown, D. G.; Boström, J. Analysis of Past and Present Synthetic
Methodologies on Medicinal Chemistry: Where Have All the New
Reactions Gone? J. Med. Chem. 2016, 59 (10), 4443−4458.
(2) Magano, J.; Dunetz, J. R. Large-Scale Applications of Transition
Metal-Catalyzed Couplings for the Synthesis of Pharmaceuticals.
Chem. Rev. 2011, 111 (3), 2177−2250.
(3) Han, F.-S. Transition-Metal-Catalyzed Suzuki−Miyaura Cross-
Coupling Reactions: A Remarkable Advance from Palladium to
Nickel Catalysts. Chem. Soc. Rev. 2013, 42 (12), 5270−5298.
(4) Rosen, B. M.; Quasdorf, K. W.; Wilson, D. A.; Zhang, N.;
Resmerita, A.-M.; Garg, N. K.; Percec, V. Nickel-Catalyzed Cross-
Couplings Involving Carbon−Oxygen Bonds. Chem. Rev. 2011, 111
(3), 1346−1416.
(5) Hayler, J. D.; Leahy, D. K.; Simmons, E. M. A Pharmaceutical
Industry Perspective on Sustainable Metal Catalysis. Organometallics
2019, 38 (1), 36−46.
(22) Sawatzky, R. S.; Ferguson, M. J.; Stradiotto, M. Thieme
Chemistry Journals Awardees - Where Are They Now? Efficient
Cross-Coupling of Secondary Amines/Azoles and Activated
(Hetero)Aryl Chlorides Using an Air-Stable DPEPhos/Nickel Pre-
Catalyst. Synlett 2017, 28 (13), 1586−1591.
(23) Wender, P. A.; Smith, T. E.; Duong, H. A.; Louie, J.; Standley,
E. A.; Tasker, S. Z. Bis(1,5-Cyclooctadiene)Nickel(0). In Encyclopedia
of Reagents for Organic Synthesis; Wiley, 2015.
(6) Egorova, K. S.; Ananikov, V. P. Toxicity of Metal Compounds:
Knowledge and Myths. Organometallics 2017, 36 (21), 4071−4090.
(7) Egorova, K. S.; Ananikov, V. P. Which Metals Are Green for
Catalysis? Comparison of the Toxicities of Ni, Cu, Fe, Pd, Pt, Rh, and
Au Salts. Angew. Chem., Int. Ed. 2016, 55 (40), 12150−12162.
(8) Percec, V.; Bae, J.-Y.; Hill, D. H. Aryl Mesylates in Metal
Catalyzed Homocoupling and Cross-Coupling Reactions. 2. Suzuki-
Type Nickel-Catalyzed Cross-Coupling of Aryl Arenesulfonates and
Aryl Mesylates with Arylboronic Acids. J. Org. Chem. 1995, 60 (4),
1060−1065.
(9) Saito, S.; Sakai, M.; Miyaura, N. A Synthesis of Biaryls via
Nickel(0)-Catalyzed Cross-Coupling Reaction of Chloroarenes with
Phenylboronic Acids. Tetrahedron Lett. 1996, 37 (17), 2993−2996.
(10) Liao, Y.; Yang, W.; Wei, T.; Cai, M. A Highly Efficient and
Recyclable NiCl2(Dppp)/PEG-400 System for Suzuki−Miyaura
Reaction of Aryl Chlorides with Arylboronic Acids. Synth. Commun.
2019, 49 (9), 1134−1142.
(24) Coombs, J. R.; Green, R. A.; Roberts, F.; Simmons, E. M.;
Stevens, J. M.; Wisniewski, S. R. Advances in Base-Metal Catalysis:
Development of a Screening Platform for Nickel-Catalyzed
Borylations of Aryl (Pseudo)Halides with B2(OH)4. Organometallics
2019, 38 (1), 157−166.
(25) Lieberman, J. A. Comparative Effectiveness of Antipsychotic
Drugs: A Commentary on Cost Utility of the Latest Antipsychotic
Drugs in Schizophrenia Study (CUtLASS 1) and Clinical Anti-
psychotic Trials of Intervention Effectiveness (CATIE). Arch. Gen.
Psychiatry 2006, 63 (10), 1069−1072.
(26) Wrobleski, S. T.; Moslin, R.; Lin, S.; Zhang, Y.; Spergel, S.;
Kempson, J.; Tokarski, J. S.; Strnad, J.; Zupa-Fernandez, A.; Cheng,
L.; Shuster, D.; Gillooly, K.; Yang, X.; Heimrich, E.; McIntyre, K. W.;
Chaudhry, C.; Khan, J.; Ruzanov, M.; Tredup, J.; Mulligan, D.; Xie,
D.; Sun, H.; Huang, C.; D’Arienzo, C.; Aranibar, N.; Chiney, M.;
Chimalakonda, A.; Pitts, W. J.; Lombardo, L.; Carter, P. H.; Burke, J.
R.; Weinstein, D. S. Highly Selective Inhibition of Tyrosine Kinase 2
(TYK2) for the Treatment of Autoimmune Diseases: Discovery of the
Allosteric Inhibitor BMS-986165. J. Med. Chem. 2019, 62 (20),
8973−8995.
(27) Liu, C.; Lin, J.; Langevine, C.; Smith, D.; Li, J.; Tokarski, J. S.;
Khan, J.; Ruzanov, M.; Strnad, J.; Zupa-Fernandez, A.; Cheng, L.;
Gillooly, K. M.; Shuster, D.; Zhang, Y.; Thankappan, A.; McIntyre, K.
W.; Chaudhry, C.; Elzinga, P. A.; Chiney, M.; Chimalakonda, A.;
Lombardo, L. J.; Macor, J. E.; Carter, P. H.; Burke, J. R.; Weinstein, D.
(11) Schaub, T.; Backes, M.; Radius, U. Catalytic C−C Bond
Formation Accomplished by Selective C−F Activation of Perfluori-
nated Arenes. J. Am. Chem. Soc. 2006, 128 (50), 15964−15965.
(12) Tobisu, M.; Xu, T.; Shimasaki, T.; Chatani, N. Nickel-
Catalyzed Suzuki−Miyaura Reaction of Aryl Fluorides. J. Am. Chem.
Soc. 2011, 133 (48), 19505−19511.
(13) Hanley, P. S.; Ober, M. S.; Krasovskiy, A. L.; Whiteker, G. T.;
Kruper, W. J. Nickel- and Palladium-Catalyzed Coupling of Aryl
I
https://doi.org/10.1021/acs.oprd.1c00210
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
pubs.acs.org/OPRD
Article
S. Discovery of BMS-986202: A Clinical Tyk2 Inhibitor That Binds to
Tyk2 JH2. J. Med. Chem. 2021, 64 (1), 677−694.
(28) Selekman, J. A.; Qiu, J.; Tran, K.; Stevens, J.; Rosso, V.;
Simmons, E.; Xiao, Y.; Janey, J. High-Throughput Automation in
Chemical Process Development. Annu. Rev. Chem. Biomol. Eng. 2017,
8 (1), 525−547.
(29) Schmink, J.; Bellomo, A.; Berritt, S. Scientist-Led High-
Throughput Experimentation (HTE) and Its Utility in Academia and
Industry. Aldrichimica Acta 2013, 46, 71−80.
(30) Thomas, A. A.; Wang, H.; Zahrt, A. F.; Denmark, S. E.
Structural, Kinetic, and Computational Characterization of the Elusive
Arylpalladium(II)Boronate Complexes in the Suzuki−Miyaura
Reaction. J. Am. Chem. Soc. 2017, 139 (10), 3805−3821.
(31) Hansch, C.; Leo, A.; Taft, R. W. A Survey of Hammett
Substituent Constants and Resonance and Field Parameters. Chem.
Rev. 1991, 91 (2), 165−195.
(32) Payard, P.-A.; Perego, L. A.; Ciofini, I.; Grimaud, L. Taming
Nickel-Catalyzed Suzuki−Miyaura Coupling: A Mechanistic Focus on
Boron-to-Nickel Transmetalation. ACS Catal. 2018, 8 (6), 4812−
4823.
(33) West, M. J.; Watson, A. J. B. Ni vs. Pd in Suzuki−Miyaura sp2−
sp2 Cross-Coupling: A Head-to-Head Study in a Comparable
Precatalyst/Ligand System. Org. Biomol. Chem. 2019, 17 (20),
5055−5059.
(34) Taylor, R. D.; MacCoss, M.; Lawson, A. D. G. Rings in Drugs. J.
Med. Chem. 2014, 57 (14), 5845−5859.
(35) Vitaku, E.; Smith, D. T.; Njardarson, J. T. Analysis of the
Structural Diversity, Substitution Patterns, and Frequency of Nitrogen
Heterocycles among U.S. FDA Approved Pharmaceuticals. J. Med.
Chem. 2014, 57 (24), 10257−10274.
(36) Li, J.; Eastgate, M. D. Making Better Decisions during Synthetic
Route Design: Leveraging Prediction to Achieve Greenness-by-
Design. React. Chem. Eng. 2019, 4 (9), 1595−1607.
(37) Monteith, E. R.; Mampuys, P.; Summerton, L.; Clark, J. H.;
Maes, B. U. W.; McElroy, C. R. Why We Might Be Misusing Process
Mass Intensity (PMI) and a Methodology to Apply It Effectively as a
Discovery Level Metric. Green Chem. 2020, 22 (1), 123−135.
(38) Gallagher, W. P.; Vo, A. Dithiocarbamates: Reagents for the
Removal of Transition Metals from Organic Reaction Media. Org.
Process Res. Dev. 2015, 19 (10), 1369−1373.
(39) Guard, L. M.; Mohadjer Beromi, M.; Brudvig, G. W.; Hazari,
N.; Vinyard, D. J. Comparison of Dppf-Supported Nickel Precatalysts
for the Suzuki−Miyaura Reaction: The Observation and Activity of
Nickel(I). Angew. Chem., Int. Ed. 2015, 54 (45), 13352−13356.
(40) Mohadjer Beromi, M.; Nova, A.; Balcells, D.; Brasacchio, A. M.;
Brudvig, G. W.; Guard, L. M.; Hazari, N.; Vinyard, D. J. Mechanistic
Study of an Improved Ni Precatalyst for Suzuki−Miyaura Reactions of
Aryl Sulfamates: Understanding the Role of Ni(I) Species. J. Am.
Chem. Soc. 2017, 139 (2), 922−936.
J
https://doi.org/10.1021/acs.oprd.1c00210
Org. Process Res. Dev. XXXX, XXX, XXX−XXX