Cu-Catalyzed Couplings of Heteroaryl Primary Amines and (Hetero)aryl Bromides with 6-Hydroxypicolinamide Ligands

Article abstract of DOI:10.1021/acs.oprd.9b00195

A family of 6-hydroxypicolinamide ligands have been identified as effective supporting ligands for Cu-catalyzed couplings of heteroaryl bromides and chlorides with heteroaryl primary amines. The C-N couplings are carried out at 80-120 °C in DMSO or sulfolane using K₂CO₃ or K₃PO₄ as the base with 2-10 mol % CuI and supporting ligand. The strength of the base was found to have an impact on the chemoselectivity and rate. The use of K₂CO₃ as the base enabled selective C-N coupling of aryl bromides over aryl chlorides with 2-5 mol % Cu at 80-120 °C. With K₃PO₄ as the base, aryl chlorides are capable of undergoing C-N coupling, though 5-10 mol % Cu is required at 120-130 °C. Members of the ligand family are straightforward to prepare in one step from 6-hydroxypicolinic acid and the corresponding anilines.

Full text of DOI:10.1021/acs.oprd.9b00195

Article
Cite This: Org. Process Res. Dev. XXXX, XXX, XXX−XXX
pubs.acs.org/OPRD
Cu-Catalyzed Couplings of Heteroaryl Primary Amines and
(
Hetero)aryl Bromides with 6‑Hydroxypicolinamide Ligands
David J. Bernhardson, Daniel W. Widlicka, and Robert A. Singer*
Chemical Research and Development, Pfizer Worldwide Research and Development, Groton Laboratories, Eastern Point Road,
Groton, Connecticut 06340, United States
ABSTRACT: A family of 6-hydroxypicolinamide ligands have been identified as effective supporting ligands for Cu-catalyzed
couplings of heteroaryl bromides and chlorides with heteroaryl primary amines. The C−N couplings are carried out at 80−120
°
C in DMSO or sulfolane using K CO or K PO as the base with 2−10 mol % CuI and supporting ligand. The strength of the
2
3
3
4
base was found to have an impact on the chemoselectivity and rate. The use of K CO as the base enabled selective C−N
2
3
coupling of aryl bromides over aryl chlorides with 2−5 mol % Cu at 80−120 °C. With K PO as the base, aryl chlorides are
3
4
capable of undergoing C−N coupling, though 5−10 mol % Cu is required at 120−130 °C. Members of the ligand family are
straightforward to prepare in one step from 6-hydroxypicolinic acid and the corresponding anilines.
KEYWORDS: Cu-catalyzed C−N coupling, heteroaryl primary amines, (hetero)aryl bromides, 6-hydroxypicolinamides
INTRODUCTION
aminopyridine moiety was selected because it is fairly common
to assemble potential therapeutic agents with this functional
group or closely related heterocycles. The substituted
bromoarene 1 was selected for ease of monitoring the
substrate and its derived impurities. The analytical responses
measured included dehalogenation of the aryl bromide to give
■
Metal-catalyzed cross-couplings are among the most frequently
used methods for the assembly of pharmaceutical targets. This
1
methodology spans C−C, C−N, C−O, and C−S couplings,
and of these, C−N couplings are often used for the preparation
2
of clinically relevant drug candidates. While Pd-catalyzed
4
, hydroxylation of the aryl bromide to form 5, diarylation of
variants are the most frequently used because of their high
efficiency and broad scope, there is a need to develop
3
the amine to afford 6, and arylation of the supporting ligand.
After examination of a large number of ligands using 5−10
mol % CuI in DMSO with K PO as the base at 120 °C, it was
comparable methods with nonprecious metals because of the
increasing cost and limited supply of Pd. Cu is among the
nonprecious metals capable of carrying out C−N couplings
3
4
observed that BFMO and N,N′-bis(thiophen-2-ylmethyl)-
15
4
oxalamide (BTMO) were very difficult ligands to outperform
(Table 1, entries 2 and 3). These Cu catalysts provided ex-
cellent reactivity and moderate selectivity for the desired
monoarylated product 3 (∼6:1 ratio of 3 to 6). In comparison
to BFMO and BTMO, dicarbonyl, diamine, salicylamide,
amino acid, and amino alcohol ligands were relatively
ineffective (Figure 3) on the basis of their inferior reaction
rates and ultimately lower product conversions. Among the
ligands examined that gave surprisingly poor results was L21.
In this case, shifting the carbonyl group toward the periphery
completely eliminated the reactivity. This suggests that
maintaining a carbonyl for binding as a five-membered chelate
analogous to Pd. Historically, Cu-catalyzed couplings have
5
been developed using ligands derived from α-amino acids,
6
7
8
1
,2-diamines, 1,3-dicarbonyls, and other motifs (Figure 1).
More recently, Ma’s group has advanced ligand design to the
9
10
use of oxalamides and amides derived from proline (Figure
2
). This family of ligands (Figure 2) has shown tremendous
11,12
utility for C−N, C−O, and C−S couplings.
A survey of C−N couplings carried out by Medicinal
Sciences at Pfizer found that the vast majority involved
heteroaryl halides and heteroaryl primary amines. There are
1
3,14
few reports for this category of couplings,
generally focused on coupling of anilines or primary amines
with aryl halides. Most notable of these reports is the use of
and they have
16,17
in the catalytic cycle is important.
From a further survey of
1
4
ligands, it became apparent that the catalyst maintained high
activity when binding as a chelate through nitrogen, preferably
through two amides. Regarding the ligands designed by Ma’s
group, the oxalamide moiety was the most critical design
element. Other secondary binding features, such as a pendent
furan or thiophene, had minor impacts on the reactivity and
selectivity. The replacement of the furan-2-ylmethyl group with
a hindered arene provided comparable reactivity (Table 1,
N,N′-bis(furan-2-ylmethyl)oxalamide (BFMO) as a ligand,
which was exemplified by Ma’s group to be effective in Cu-
catalyzed couplings of heteroaryl bromides with anilines and
secondary amines. Of all of the couplings reported with BFMO
as a ligand, there was only one example involving a heteroaryl
1
4
primary amine. Our ultimate goal was to identify the optimal
ligand and a general set of conditions for the Cu-catalyzed
couplings of heteroaryl bromides with heteroaryl primary
amines.
Special Issue: Honoring 25 Years of the Buchwald-Hartwig
Amination
RESULTS AND DISCUSSION
■
The model system chosen for initial screening (Scheme 1)
involved aryl bromide 1 and 2-aminopyridine (2). The 2-
Received: May 1, 2019
1
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.oprd.9b00195
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Figure 1. First generation of ligands employed in Cu-catalyzed couplings.
Figure 2. Second generation of ligands developed by Ma’s group for Cu-catalyzed couplings.
Scheme 1. Model C−N Coupling Studied with the Ligand Library
entry 4), while less hindered aryl groups exhibited slightly
reduced reactivity (Table 1, entry 5). The choice of the base
was found to be critical as well. When K PO was replaced
resulted in lower rates of competitive ligand arylation while
maintaining relatively high reactivity for the desired coupling of
1 and 2 (Table 1, entry 7). Unfortunately, ligands derived from
6-oxopipecolic acid are not as practical to prepare as those
derived from pyroglutamic acid because of the higher cost of
the starting material. However, this provided a good lead for
ligand design.
3
4
with K CO , the oxalamide ligands (BFMO, BTMO, and
2
3
N,N′-bis(2,4,6-trimethoxyphenyl)oxalamide (BTMPO)) pos-
sessed no reactivity for the coupling of 1 and 2.
From the broad ligand screen, pyroglutamamides (such as
L35, shown in Figure 4) emerged as efficient scaffolds for Cu
catalysts. Pyroglutamamides had practical appeal as ligands
because of their low cost. However, with weakly nucleophilic 2
as a substrate, N-arylation of the ligand pyrrolidinone moiety
was highly competitive. The competitive arylation of L35 led
to catalyst deactivation and stalling of the reactions at partial
As an alternative to ligands possessing two amide donors,
competitive reactivity was achieved when one of the two
amides was replaced with a nitrogen heterocycle such as a
pyridine (L37), indole, pyrrole, or benzimidazole. AbbVie
disclosed the use of a picolinamide ligand (Figure 5) for the
19
synthesis of dasabuvir, an HCV polymerase inhibitor with a
related C−N coupling. Following this precedent, picolinamide
ligands such as L37 were found to have sufficient reactivity but
18
conversion (Table 1, entry 6). Increasing the ring size so that
ligands were derived from 6-oxopipecolic acid (such as in L36)
B
DOI: 10.1021/acs.oprd.9b00195
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Table 1. Comparison of Ligand Reactivitiesa,b
the desired monoarylated product 3. Even more impressive
was the observation that the hydroxypicolinamides (L40 and
L41) significantly surpassed the oxalamide ligands (BFMO,
BTMO, and BTMPO) in terms of reactivity and selectivity.
The members of the family of 6-hydroxypicolinamide
ligands 10 (Figure 6) were prepared in a single step by
coupling of 6-hydroxypicolinic acid (7) with the corresponding
anilines 8 by propylphosphonic anhydride (T3P) activation
2
h
15 h
IPC
24 h IPC (% total
impurities)
entry
ligand
IPC
1
2
3
4
5
6
7
8
9
0
1
2
3
4
none
<1%
37%
39%
35%
13%
68%
75%
5%
14%
95%
99%
8%
6%
92%
99%
97%
80%
71%
99%
48%
47%
99%
99%
37%
8%
9%
BTMO
BFMO
BTMPO
99% (19%)
99% (15%)
99% (17%)
99% (25%)
71% (12%)
99% (9%)
67% (28%)
48% (16%)
99% (1%)
99% (7%)
70%
c
des-ortho-OMe BTMPO
(
Scheme 2). The amidation was significantly slower with more
L35
L36
L37
L38
L40
L41
hindered and electron-deficient anilines. Upon completion of
the amidation, heating with methanol hydrolyzed the
remaining phosphonate species bound to the desired ligand.
Usually the reaction was quenched with aqueous hydroxide
followed by distillation to displace the THF and ethyl acetate
with methanol. The phosphonate bound to the desired ligand
1
1
1
1
1
L16 (8-hydroxyquinoline)
L8 (phenanthroline)
L6 (DMEDA)
(
9) was visible by UPLC, allowing the phosphonate hydrolysis
<1%
<1%
11%
9%
to be monitored. For most of the ligands, the hydrolysis
required heating for 1−8 h in methanol under alkaline
conditions. Once the hydrolysis was complete, the liberated
ligand 10 was directly crystallized with the addition of water.
The family of 6-hydroxylpicolinamide ligands 10 displayed
structure−activity relationships similar to those of Ma’s
7%
a
Conditions: 5 mol % CuI, 10 mol % ligand, 1.5 equiv of K PO , 2 M
3
4
b
DMSO relative to 1 (1 mmol), 1.1 equiv of 2, 120 °C. In-process
control (IPC) reported as a conversion, excluding impurities or side
c
products. Structure of des-ortho-OMe-BTMPO:
14,15,21
oxalamide ligands (Table 2).
For the arylamide moiety,
ortho substituents were essential to maintain a high level of
activity. Preparing the parent 6-hydroxy-N-phenylpicolinamide
(
L47) from aniline provided significantly less reactivity than
L46 or L43. Introducing a second ortho substituent (L40 and
L42) further improved the reactivity and selectivity for the
desired monoarylated product 3. Electron-donating groups on
the arylamide increased the reactivity (L41) while electron-
withdrawing groups reduced it (L45). Despite the lower
reactivity with L45, the model coupling reaction still managed
to reach completion in a reasonable time frame and provided a
relatively clean impurity profile. For the pyridyl ring, shifting
the hydroxyl group to the 4-position and introducing a 6-
methyl group or fused ring led to severe inhibition of the
reactivity (L48). From this study, both L40 and L42 were
deemed to be optimal as ligands for the coupling of 1 and 2.
However, L40 was chosen for optimization studies and further
possessed relatively poor selectivity for the desired product 3
2
0
versus diarylated product 6 (Table 1, entry 8). The
introduction of a methoxy group ortho to the pyridine (or
pyrimidine) nitrogen improved selectivity for monoarylation
(
1
entry 9), but Cu complexes derived from L37 or L38 required
0 mol % Cu to achieve at least 90% conversion. It was then
hypothesized that swapping the methoxy group for a hydroxy
group would provide ligands with improved reactivity
analogous to L36 since this would enable the pyridine nitrogen
to become an anionic donor. As expected, the 6-hydrox-
ypicolinamides (L40 and L41) were far superior ligands,
providing more reactive Cu complexes with high selectivity for
Figure 3. Ineffective ligands for the Cu-catalyzed coupling of 1 and 2.
C
DOI: 10.1021/acs.oprd.9b00195
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Figure 4. Early ligands examined for optimization of design.
Figure 5. Further optimization of the ligand design.
Figure 6. Library of 6-hydroxypicolinamide ligands 10.
Scheme 2. Preparation of 6-Hydroxypicolinamide Ligands
development because of its preparation from inexpensive
commodity chemicals.
corresponding higher levels of 5. Because K PO is highly
3
4
hygroscopic, water was typically present when this base was
used and led to significant levels of hydroxylation (5−10%) in
early studies. To illustrate this further, the use of aqueous
KOH (Table 3, entry 7) afforded 5 as the major product,
suggesting that the catalyst could be efficiently used for
hydroxylation if intended. With dry DMSO (≤0.05% water)
Optimization of the Reaction Conditions and Process
Understanding. Although the 6-hydroxypicolinamide ligands
limited undesired diarylation (6) and dehalogenation (4),
competitive hydroxylation of the aryl bromide (5) became
more prevalent. The extent of hydroxylation was found to
correlate directly to the level of water present in the reaction
mixture. Spiking with higher levels of water led to
22
and anhydrous K PO , hydroxylation was not a significant
3
4
issue (<1% 5) and typically did not require additional
D
DOI: 10.1021/acs.oprd.9b00195
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Table 2. Comparison of the Reactivities of 6-
Hydroxypicolinamide Ligands 10
volume ratio of DMSO to toluene at 110−120 °C was ideal for
a
,
b
24
minimizing hydroxylation in the presence of water.
Of the inorganic bases, K PO (Table 3, entry 4)
3
4
entry
ligand
2 h IPC
15 h IPC (% total impurities)
demonstrated the highest reactivity other than Cs CO
2
3
1
2
3
4
5
6
7
8
9
L40
L41
L42
L43
L44
L45
L46
L47
L48
L49
L50
93%
99%
99%
84%
65%
28%
85%
35%
1%
99% (1%)
99% (7%)
99% (1%)
99% (4%)
95% (13%)
99% (4%)
99% (11%)
99% (16%)
16%
(
Table 3, entry 3) and aqueous KOH (Table 3, entry 7).
Both Cs CO and aqueous KOH had the liability of promoting
competitive hydroxylation because of the presence of water in
those reagents. While Cs CO possessed high reactivity and
could be potentially dried, it was considered less desirable for
large-scale manufacture as cesium is a less abundant and more
expensive heavy metal. As a viable alternative to K PO ,
K CO (Table 3, entry 1) was effective with this ligand family
1
base is required. While KOtBu or NaOtBu could be used as
the base, they possess incompatibility with DMSO as a safety
concern, especially with the relatively high temperatures
2
3
2
3
3
4
2
3
0 and could be useful for instances when a slightly milder
1
1
0
1
77%
40%
99% (11%)
79% (5%)
a
Conditions: 5 mol % CuI, 10 mol % ligand, 1.5 equiv of K PO , 2 M
3
4
b
DMSO relative to 1 (1 mmol), 1.1 equiv of 2, 120 °C. In-process
control (IPC) reported as a conversion, excluding impurities or side
products.
25
required for the couplings. As an alternative to DMSO, the
alkoxide bases were evaluated in alcohol solvents but were still
less reactive than K PO under those conditions.
3
4
In general, potassium salts were more reactive than sodium
salts. This trend was observed with carbonates (Table 3, entry
Table 3. Comparison of Bases Used in the Coupling of 1
and 2
,
a c
1
vs entry 2) and phosphates (Table 3, entry 4 vs entry 5)
entry
base
3 h IPC
51%
11%
99% (13%)
82%
15 h IPC
24 h IPC
alike. Amine bases (Et N, iPr NEt, and TMEDA) were
3
2
c
c
1
2
3
4
5
6
7
K CO3
Na CO3
99% (1%)
99% (1%)
ineffective, but DBU (Table 3, entry 8) was able to promote
the coupling relatively slowly, although the reaction appeared
to stall at partial conversion. As an additional complication,
DBU was observed to undergo coupling with the aryl bromide
2
30%
38%
2
c
c
c
c
c
c
2
3
Cs CO3
99% (13%)
99% (13%)
c
c
K PO4
99% (1%)
99% (1%)
26
3
Na PO4
27%
25%
46%
39%
48%
39%
as a side reaction.
b
b
b
KOH (pellets)
aqueous KOH
DBU
The strength of the base was observed to influence both the
99% (76%)
30%
99% (76%)
40%
99% (76%)
40%
reactivity and chemoselectivity. In this respect, K PO
3
4
8
provided more rapid catalysis than K CO , but K CO
2
3
2
3
a
Conditions: 5 mol % CuI, 10 mol % L40, 1.5 equiv of base, 1 M in
provided selectivity toward exclusive coupling of aryl bromides
over aryl chlorides. Early screening work with K PO showed
b
DMSO relative to 1 (1 mmol), 1.1 equiv of 2, 120 °C. A mixture of
regioisomers was obtained. In-process control (IPC) reported as a
3
4
c
that aryl chlorides proceeded to couple at relatively low rates
conversion, excluding impurities or side products; the extent of the
hydroxylation impurity is reported as a percentage of the total product
distribution.
(
48 h at 120−130 °C with 5−10 mol % Cu) compared with
aryl bromides (2−6 h at 120 °C with 2−5 mol % Cu).
Switching to the use of K CO as the base ceased coupling
2
3
with aryl chlorides entirely, even when chlorobenzene was used
as a cosolvent.
precautions. With dry DMSO and substrates, K PO and
3
4
K CO were found to be equivalent in maintaining low levels
Upon examination of solvents for the coupling reactions,
DMSO, sulfolane, and other polar aprotic solvents were found
to be ideal (Table 4, entries 1−3). The reaction was
significantly slower when a large proportion of toluene or
2
3
of competitive hydroxylation (<1%). When the solvent or
substrates introduced significant levels of water (>0.2% by Karl
Fischer analysis (KF) of the resulting reaction mixture), a
practical solution to minimize the competitive hydroxylation
was to conduct the couplings above the boiling point of water
or to blend toluene with DMSO or sulfolane to facilitate
27
anisole was blended with DMSO. Primary and secondary
alcohols provided competitive levels of reactivity with the
catalyst. Under some circumstances, primary alcohols were
23
28
removal of water by distillation. The use of a 5:1 or 10:1
found to undergo competitive C−O coupling, though
a
Table 4. Comparison of Solvents Used in the Coupling of 1 and 2
entry
solvent
DMSO
sulfolane
DMAC
1-butanol
ethylene glycol
4-methyl-2-pentanol
cyclohexanol
tert-amyl alcohol
2 h IPC
58%
67%
5 h IPC
99%
97%
18 h IPC
c
1
2
3
4
5
6
7
8
99% (1%)
99% (2%)
99% (1%)
c
c
53%
79% (9%)
86%
c
c
c
b
c
c
99% (17%)
99% (67%)
50%
99% (20%)
20%
99% (17%)
99% (67%)
92% (10%)
99% (20%)
51%
b
b
85% (44%)
30%
98%
c
15%
a
Conditions: 5 mol % CuI, 10 mol % L40, 1.5 equiv of K PO , 2 M in solvent relative to 1 (1.0 mmol), 1.1 equiv of 2, 120 °C; in-process control
3
4
b
(
IPC) reported as a conversion, excluding impurities or side products. Extent of C−O coupling reported as a percentage of the product mixture.
c
Extent of total impurities reported as a percentage of the product distribution.
E
DOI: 10.1021/acs.oprd.9b00195
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
secondary and tertiary alcohols did not. For the reaction
conditions explored in Table 4, 1-butanol (entry 4) did not
undergo ether formation. In contrast to 1-butanol, ethylene
glycol (Table 4, entry 5) heavily favored C−O coupling under
these conditions. Tertiary alcohol solvents gave poor
reactivity (Table 4, entry 8). Even though primary and
secondary alcohols appeared to be ideal as solvents, generally
more diarylation of the amine was observed compared with
using polar aprotic solvents (DMSO, DMAC, or sulfolane);
however, conditions could potentially be optimized to
minimize this side reaction. With aryl chlorides as substrates,
observed that electron-rich products possessing the 4-amino-
pyrazole moiety were capable of undergoing dimerization with
Cu(II). In these cases, where Cu(II) must be present to enable
31
dimerization, the addition of sodium ascorbate was found to
suppress this side reaction. When coupling reactions were
initiated, frequently the reaction mixtures would transition
through a green color at the start, suggesting the presence of
Cu(II), and rapidly became a brown color shortly after
2
9
3
2
heating. The catalyst initiation proceeded well under
nitrogen, but when couplings were conducted in vials without
nitrogen inertion, the catalyst failed to initiate in some
instances. In these rare cases, addition of sodium ascorbate
1
-butanol was generally more reactive than DMSO and,
33
surprisingly, did not lead to any appreciable levels of side
products, including C−O coupling (hydroxylation or ether
formation).
enabled reliable catalyst activation. In general, the addition of
sodium ascorbate ensured that the reactions achieved
34
maximum reaction rates more efficiently.
The ligand to Cu molar ratio was found to have a profound
impact on the reactivity and selectivity. As the ligand to Cu
molar ratio was increased beyond 1:1, the reaction rate
The addition of KI to reactions allowed halogen exchange to
proceed, converting the aryl bromide to the corresponding aryl
iodide. When 0.50 equiv of KI was introduced into the reaction
mixture, typically about 10−20% of the aryl bromide was
converted to the aryl iodide as the aryl halides were consumed
in the coupling. The addition of KI appeared to slow the
reaction progress, and in cases where hydroxylation was highly
competitive (such as coupling of aqueous ammonia or a
secondary amine), it reduced the extent of hydroxylation by
3
0
increased. For example, a 1:1 ligand to Cu ratio at a Cu
loading of 2.5 mol % gave 21% conversion to 3 after 6 h; the
same transformation achieved 62% conversion with a 2:1
ligand to Cu ratio and >99% conversion with a 4:1 ligand to
Cu ratio. In addition to its impact on the reaction rate, the
ligand to Cu ratio influenced the extent of competing
hydroxylation (formation of 5). Maintaining the ligand to Cu
ratio near 1:1 suppressed hydroxylation (Table 5, entries 1 and
35
about 30−50%. Using the corresponding aryl iodide had this
same effect of slowing the reaction progress relative to the aryl
bromide but gave the same product distribution as the aryl
Table 5. Examination of the Ligand to CuI Molar Ratio ,
a b
36
bromide.
Discussion of the Substrate Scope. A variety of aryl and
heteroaryl bromides were examined in the coupling with 2-
aminopyridine (Figure 7) using 4 mol % CuI and 6 mol %
L40. The couplings were generally carried out at 120−130 °C
for efficiency, but in many cases the couplings of (hetero)aryl
bromides could be conducted at 80−100 °C with longer
reaction times. Unhindered bromides performed well,
providing complete reactions within 4−12 h, while those
possessing an ortho substituent typically did not proceed well
in the coupling. For example, 4-bromochlorobenzene (11f)
and 4-bromoanisole (11k) coupled cleanly, but the ortho-
substituted isomers (11g and 11l) failed to proceed in the
coupling regardless of the base chosen. When the solvent was
switched to 1-butanol, 11l proceeded to couple more
efficiently, reaching 42% conversion in 16 h and 76%
conversion in 48 h with 4 mol % CuI at 120 °C. In the case
of 11c, coupling proceeded despite the o-pyrazole substituent,
most likely because the pyrazole served as a directing group for
Cu. A number of heterocycles such as 6-bromoquinoline (11a)
coupled in good yield, but the isomer 8-bromoquinoline (11b)
failed to couple well and favored hydroxylation. Highly
electron-rich and unprotected substrates such as 3-iodoinda-
zole and 6-bromoindazole failed to undergo coupling, most
likely because of the deprotonation of the indazole nitrogen,
which could impede oxidative addition or bind the Cu catalyst
to form a stable complex.
entry
L40:CuI
3 h IPC
7 h IPC
24 h IPC
1
2
3
4
5
0.6:1
1.3:1
2:1
64% (<1%)
66% (<1%)
85% (1%)
86% (2%)
94% (5%)
72% (1%)
81% (1%)
95% (3%)
99% (1%)
99% (5%)
77% (1%)
88% (1%)
99% (4%)
99% (3%)
99% (5%)
c
2:1
4:1
a
Conditions: 4 mol % CuI, L40 as indicated, 1 M in DMSO relative
to 1 (1.0 mmol), 1.1 equiv of 2, 1.5 equiv of K CO , and 0.10 equiv of
sodium ascorbate. In-process control (IPC) reported as a
conversion, with the amount of 5 formed as a percentage of the
product distribution reported in parentheses. No sodium ascorbate
2
3
b
c
was used.
2
), but when the ligand to Cu ratio was increased to 2:1 and
higher, hydroxylation became more problematic (Table 5,
entries 3−5). The extent of competitive hydroxylation
increased significantly with solvents possessing higher water
levels and followed this same trend relating to the ligand to Cu
molar ratio. For example, DMSO with >0.2% water by KF
afforded 1−2% 5 at a 2:1 ligand to Cu ratio but 10−20% 5 at a
4
:1 ligand to Cu ratio. As the overall reaction concentration for
the coupling of 1 and 2 was increased, the reaction rate was
observed to increase with similar product distributions. For
optimal performance, the reaction concentration was typically
maintained at 1−2 M and the ideal ligand to Cu molar ratio
was typically in the range of 1.5:1 to 2:1.
Additives were examined to further understand their impact
on the process. Introducing sodium ascorbate as an additive
had a minimal effect on the reaction rate and purity profile for
this set of substrates and conditions (Table 5, compare entries
Aryl chlorides coupled at a much lower rate than the
corresponding aryl bromides. In general, the aryl chlorides did
not appear to have the same tendency toward diarylation of the
amine or competitive hydroxylation of the aryl halide. It was
observed that switching the solvent from DMSO to 1-butanol
provided a higher rate of catalysis for coupling of aryl chlorides
with primary amines at 120 °C without any observed C−O
coupling. Increasing the reaction temperature to 130 °C with
DMSO or sulfolane as the solvent significantly increased the
3
and 4). Nevertheless, sodium ascorbate was included for this
study to ensure reliable activation of the Cu catalyst for
comparison of the kinetics and product distributions across all
time points. When other substrates were explored, it was
F
DOI: 10.1021/acs.oprd.9b00195
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Figure 7. Couplings of 2-aminopyridine with various (hetero)aryl halides. Reaction conditions: 4 mol % CuI, 6 mol % L40, 1 M in DMSO, 1.5
b
c
equiv of K CO , 1.1 equiv of amine, 120 °C, 16 h. Notes: HPLC/UPLC purity assay with K CO as the base. HPLC/UPLC purity assay with
2
3
2
3
d
e
f
K PO as the base. Isolated yield with K CO as the base. Isolated yield with K PO as the base. Modified reaction conditions: 8 mol % CuI, 16
3
4
2
3
3
4
mol % L40, 2 M in DMSO, 1.5 equiv of K PO , 130 °C, 24 h.
3
4
reaction rate for aryl chlorides and other substrates that were
less capable of undergoing oxidative addition with the Cu
catalyst. When 8 mol % CuI and 16 mol % L40 were used with
Scheme 3. Dimer Formation of 4-Aminopyrazoles
1
.5 equiv of K PO in DMSO, 11j was able to achieve 66%
3 4
conversion within 24 h and 90% conversion in 48 h. In
comparison, chloropyridine 11m was able to reach >90%
conversion under these same conditions in 12−15 h with
minimal side products (<1%). From this perspective, activated
aryl chlorides seemed to be optimal substrates with reasonable
reactivities and cleaner product profiles.
level similar to the Cu loading was found to be sufficient to
suppress this side reaction.
To demonstrate the efficiency of the process further,
examples 11a and 11h were scaled up to 5 g using 2 mol %
Cu and 4 mol % L40. In the case of 11a, the reactions in
DMSO and sulfolane were compared and achieved nearly
In general, the aminopyrazoles were prone to many side
reactions, leading to lower yields in some cases (such as 13c).
Other than suppressing the dimerization of the pyrazoles with
sodium ascorbate, the reaction conditions were not optimized.
With the use of sodium ascorbate, a variety of heteroaryl
bromides were coupled with 4-aminopyrazoles in modest to
excellent yields (Figure 8). In the case of 13g, the yield was
lower because the ortho substituent impeded the catalysis.
This pyrazole dimerization side reaction was observed when
4-bromopyrazole derivatives (such as in example 11h in Figure
7) were coupled with amines because this also furnished 4-
(hetero)arylaminopyrazole products (13). In the case of
coupling of 11h with 2, the addition of sodium ascorbate
suppressed the observed dimerization to <1%. As an additional
liability, dehalogenation was observed for 11h as a major
competing side reaction rather than hydroxylation, which had
not been the case for any of the other substrates examined.
Limiting the level of water to <0.1% seemed to minimize
dehalogenation to <1%, analogous to hydroxylation which
occurred with most other substrates. It was also found that the
loading of sodium ascorbate was correlated to the extent of
dehalogenation for coupling of substrate 11h.
37
identical reaction rates and impurity distributions. For 11a,
the reaction stalled consistently at 50−60% conversion because
of arylation of L40. Recharging the reaction mixture with
another 4 mol % loading of L40 allowed the reaction to reach
>
99% conversion within a few hours. This demonstrates the
ability of recharging the ligand to Cu in the presence of the
substrates and product to successfully reactivate the catalyst.
38
Upon reaction completion, the reaction mixtures were diluted
with MeOH and aqueous ammonia to directly crystallize the
desired products. Aqueous ammonia facilitated the removal of
Cu during the crystallization. The ligand (L40) remained in
the mother liquor or in an aqueous phase if a phase separation
was performed for the workup. From this exercise, the
products were isolated in good yields (76% and 90% for 12a
and 12h, respectively).
Other heteroaryl amines were examined under similar
conditions with a focus on aminopyrazoles. Couplings with
highly electron-rich 4-amino-3-methoxymethylpyrazole were
explored and were found to proceed rapidly with 2 mol % Cu.
As the reactions approached completion, pyrazole dimers (14)
consistently formed at levels ranging from 5% to 50% (Scheme
Couplings with 3-aminopyrazoles were found to be higher-
yielding than those with 4-aminopyrazoles because these
products did not undergo dimerization or other side reactions.
The reactions were carried out with 4 mol % CuI and 6 mol %
L40 in DMSO with K CO as the base (Figure 9). As observed
3
). It was speculated that the presence of Cu(II) could
promote the oxidative dimerization reaction of the amino-
pyrazole products. The addition of sodium ascorbate at a
39
2
3
G
DOI: 10.1021/acs.oprd.9b00195
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Figure 8. Couplings of (hetero)aryl bromides with 4-aminopyrazoles. Reactions conditions: 2 mol % CuI, 4 mol % L40, 1 M in DMSO, 2 equiv of
K CO , 1.1 equiv of amine, and 0.10 equiv of sodium ascorbate at 80 °C (examples 13a, 13d, and 13e) or 100 °C (examples 13b, 13c, 13f, and
2
3
b
c
1
3g). Notes: HPLC/UPLC purity assay in the reaction mixture. Isolated yield.
Figure 9. Couplings with 3-amino-N-methylpyrazole. Reaction conditions: 4 mol % CuI, 6 mol % L40, 1 M in DMSO, 1.5 equiv of K CO , 1.1
2
3
b
c
equiv of amine, 110 °C. Notes: HPLC/UPLC purity assay. Isolated yield.
Figure 10. Other couplings of heteroaryl bromides with heteroaryl amines. Reaction conditions: 4 mol % CuI, 6 mol % L40, 1 M in DMSO, 1.5
b
c
equiv of K CO , 1.1 equiv of amine, 110 °C. Notes: HPLC/UPLC purity assay. Isolated yield.
2
3
with other substrate sets, aryl bromides with ortho substituents
impeded catalysis (such as in 14g) unless the ortho substituent
was a directing group (such as in 14b). Because K CO was
used as the base, 14f was observed to form cleanly as the sole
product with no competing coupling at the chloride.
A variety of other heteroaryl primary amines were examined
in couplings, mainly with 6-bromoquinoline (11a) and 2-
bromo-6-methoxypyridine (11i) (Figure 10). The highly
activated substrate 11i outperformed 11a in each case. The
reactions were much slower for the aminopyrimidines
examined (15a, 15d, and 15g) because of the relatively poor
nucleophilicity. For each of the pyrimidine couplings there was
a strong preference for the C−N coupling over the potential
C−O coupling, but the reactions achieved only modest
conversions under these unoptimized conditions. In the cases
with unprotected pyrazoles (15b, 15e, and 15h−j), com-
petitive C−N coupling was observed to proceed at both
nitrogens of the pyrazole ring. The typical isomer ratio was
2
3
H
DOI: 10.1021/acs.oprd.9b00195
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Figure 11. Couplings of alkyl amines and ammonia with (hetero)aryl bromides. Reaction conditions: 4 mol % CuI, 6 mol % L40, 1 M in DMSO,
b
c
d
1
.5 equiv of K PO , 1.1 equiv of amine, 120 °C. Notes: HPLC/UPLC purity assay. Isolated yield. Modified reaction conditions: 4 mol % CuI, 8
3 4
mol % L40, 1 M in DMSO, 1.5 equiv of K CO , 110 °C.
2
3
3
0:1:1, favoring the coupling at the desired primary amine
CONCLUSIONS
■
rather than at the two pyrazole nitrogens. The unprotected
pyrazoles had lower reaction rates and did not reach reaction
completion, suggesting that protection or alkylation of the
pyrazole is warranted for greater efficiency, as observed in
other examples (Figures 8 and 9).
The scope of couplings with L40 and CuI was briefly
examined for primary and secondary alkyl amines (Figure 11).
Couplings of (hetero)aryl bromides with benzylamine were
complete in 1−2 h because of the high nucleophilicity.
Couplings of aryl bromides with benzyloxycarbonyl (Cbz)-
protected piperazine were also fairly rapid with K PO as the
A new family of 6-hydroxypicolinamide ligands was developed
for the Cu-catalyzed coupling of heteroaryl bromides and
heteroaryl primary amines. The reaction conditions were
highly optimized for the coupling of 2-aminopyridine (2) with
various heteroaryl bromides using CuI and L40. L40 was
selected for study because it provided high catalytic perform-
ance, delivered a clean product profile, and was practical to
prepare. A number of key process parameters were identified
that impact both the reaction rate and the product distribution.
Among the most influential process parameters were the
choice of solvent, base, reaction temperature, and concen-
tration in addition to the ligand to Cu molar ratio. Even
though DMSO was the most widely utilized solvent for the
studies, sulfolane appeared to provide virtually identical
3
4
base, but it was observed that hydroxylation with subsequent
diaryl ether formation was highly competitive, which resulted
in modest purity and yields for 16a−c. Even though the
couplings proceeded efficiently at 80 °C, this did not provide
any benefit to the product distribution. Only when KI as was
included as an additive was the level of hydroxylation reduced
by about 50%. Coupling of Cbz-protected piperazine with aryl
chloride 11j to furnish 16a was significantly slower than with
the corresponding bromide (1) but was relatively clean with no
significant C−O coupling (<1%) in sulfolane (99% conversion
25
performance and has better stability with base, making it a
more viable choice for process development. In a number of
examples, 1-butanol and other alcohols were potential
alternatives to polar aprotic solvents for these Cu-catalyzed
C−N couplings. The use of sodium ascorbate provided more
reliable activation of the catalyst, especially for experiments
carried out in vials without nitrogen inertion. A variety of other
heteroaryl amines were surveyed to illustrate the potential
scope with this family of ligands. For the unoptimized
substrates studied, additional development of reaction
conditions could be carried out to improve the reaction
profiles and increase the isolated yields. Reaction optimization
should be possible through design of experiment (DoE)
studies, which is likely to be necessary given the complexity of
interactions among the various process parameters. Optimiza-
tion of the ligand design for other applications is facilitated
through the highly modular design and will be reported in due
course.
48 h at 10 mol % Cu). This same transformation in DMSO
had a similar reaction rate but formed 8% diaryl ether, similar
to coupling with 1. In contrast to what was observed with
primary amines, the reaction between Cbz-protected piper-
azine and 11j was much slower in 1-butanol (only 14%
conversion over 20 h and 35% conversion after 48 h) and had
competitive side reactions.
The coupling of ammonia with aryl halides has been
demonstrated to be highly efficient with oxalamide ligands and
other systems. Bromination of arenes or heteroarenes
followed by coupling with ammonia is a valuable alternative
to nitration. In particular, nitration of pyrazoles and related
heterocycles can produce potentially hazardous intermedi-
4
0
4
1
EXPERIMENTAL SECTION
ates. The coupling of 11h with ammonia was demonstrated
to afford 16f with 88% purity using K CO as the base in
■
General. All of the screening reactions were carried out in
screw-capped glass vials without nitrogen inertion. For the
substrate scope studies and scale-up trials (to prepare 12a and
12h), the reactions were conducted in dry reaction vessels
under an atmosphere of dry nitrogen. All of the reagents and
solvents were used as received without further purification
unless otherwise specified.
N-(2,6-Dimethylphenyl)-6-hydroxypicolinamide
(L40). To a 100 mL reaction vessel were charged 6-
hydroxypicolinic acid (10.0 g, 71.8 mmol), THF (100 mL),
and triethylamine (30 mL, 216 mmol, 3 equiv) at 25 °C. This
2
3
DMSO (1 M) at 110 °C over 24 h. The only impurity present
with 16f was the product of dehalogenation of 11h (at a 12%
level). Increasing the concentration of the substrates (2 M
relative to 11h) resulted in higher levels of dehalogenation and
other side reactions for the coupling with ammonia.
Dehalogenation seemed to be common problem for 11h as a
side reaction, and this was also observed in the coupling with
benzylamine to yield 16e. As observed in other examples with
1
1h, increasing the loading of sodium ascorbate was correlated
to an increase in dehalogenation for this substrate.
I
DOI: 10.1021/acs.oprd.9b00195
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
initially dissolved the picolinic acid, and a new, thicker white
suspension formed. 2,6-Dimethylaniline (9.32 mL, 75.4 mmol,
125.8, 126.2, 126.7, 126.8, 127.2, 129.0, 133.2, 134.2, 141.3,
146.4, 162.4, 163.1.
6-Hydroxy-N-(o-tolyl)picolinamide (L46). Prepared
analogously to L40 on the same scale of operations. L46
1
2
.05 equiv) was charged, and the resulting mixture was held at
5 °C; the charge of the aniline helped break up the slurry. To
the reaction mixture was charged T3P (50 wt %) in ethyl
acetate (60 mL, 100.6 mmol, 1.40 equiv) dropwise over 30
min. The exotherm was well-controlled by charging the T3P
slowly. As the addition progressed, the reaction mixture
became homogeneous. The mixture was held at 25 °C
following addition for 30 min, warmed to 65 °C over 2 h,
and held at 65 °C for 16 h. The jacket was heated to 85 °C to
begin distillation, and the mixture was distilled to a volume of
(14.8 g, 90% yield) was isolated as white crystals (mp 190−
1
192 °C). H NMR (400 MHz, DMSO-d ): δ 2.34 (s, 3H),
6
6.88 (d, J = 8.59 Hz, 1H), 7.08−7.16 (m, 1H), 7.25 (t, J = 7.61
Hz, 1H), 7.29 (d, J = 7.41 Hz, 1H), 7.52 (d, J = 7.02 Hz, 1H),
7.84 (t, J = 7.82 Hz, 1H), 7.92 (d, J = 7.41 Hz, 1H), 10.01 (s,
1
3
1H). C NMR (100 MHz, DMSO-d ): δ 18.0, 113.0, 115.8,
6
122.7, 125.3, 126.9, 130.0, 130.9, 136.3, 141.3, 146.4, 162.0,
162.9.
8
0 mL with a pot temperature of ∼71 °C. The reaction
6-Hydroxy-N-phenylpicolinamide (L47). Prepared anal-
ogously to L40 on the same scale of operations. L47 (14.6 g,
mixture was cooled to 50 °C temporarily and quenched with
water (40 mL) and aqueous 11.5 M KOH (10 mL, 115 mmol,
95% yield) was isolated as white crystals (mp 269−272 °C).
1
1
.6 equiv) to adjust the pH to 10−11. The distillation was
H NMR (400 MHz, DMSO-d ): δ 6.83 (d, J = 8.59 Hz, 1H),
6
resumed under atmospheric conditions, and MeOH (80 mL)
was charged in portions during the distillation (or as a
constant-volume displacement). As the distillation continued,
the pH dropped to 7 (volume at 120 mL). When a volume of
about 110 mL was reached, the pH had dropped to 5−6. When
a volume of about 100 mL was reached, crystals formed
7.14 (t, J = 7.34 Hz, 1H), 7.34−7.45 (m, 3H), 7.75−7.82 (m,
1
3
3H), 10.23 (s, 1H). C NMR (100 MHz, DMSO-d ): δ 100.1,
6
112.3, 116.9, 120.4, 124.5, 129.3, 138.6, 141.9, 162.0, 162.9.
General Procedure for Coupling of Primary Amines
with Heteroaryl Bromides. To a 25 mL reaction tube were
charged CuI (0.02−0.04 equiv) and L40 (0.04−0.08 equiv) in
DMSO (2−4 mL/g). After the mixture was held for at least 10
min at 25−35 °C, potassium carbonate or potassium
phosphate (1.5 equiv) was charged. The reaction mixture
was held at 25−30 °C for at least 10 min. To the reaction
mixture was charged the heteroaryl bromide (5.00 mmol, 1
equiv) followed by the amine (5.50 mmol, 1.1 equiv). In cases
with electron-rich heterocyclic amines, sodium ascorbate (0.10
equiv) was charged (it should also be used for reactions
conducted without nitrogen inertion). The reaction mixture
was heated to 120 °C and held until the reaction assay
indicated >99% conversion. The reaction mixture was cooled
to 50 °C. To the reaction mixture were charged aqueous
ammonia (5 mL/g) and water (5 mL/g). For direct
crystallization of the product, MeOH (2−3 mL/g) was
charged, followed by aqueous ammonia (2−3 mL/g) and
water (2−3 mL/g), and the mixture was held at 50−60 °C for
30−60 min and then cooled to room temperature to filter the
crystals. For aqueous workup, ethyl acetate or toluene (10 mL/
g) was charged to the aqueous mixture to extract the desired
product. The layers were separated, and the organic phase was
concentrated to either crystallize the desired product or purify
the desired product by silica gel chromatography.
(
UPLC tracking showed that the phosphonate hydrolysis had
proceeded to completion over this time during the distillation)
and had a pot temperature of 72−73 °C. To the mixture were
charged water (30 mL) and MeOH (20 mL), which thickened
the slurry. (Note: charging much more MeOH can start to
redissolve the crystals). The reaction mixture was held at reflux
for 1 h. The slurry was cooled to 20 °C over at least 2 h. The
suspension thickened upon cooling. The slurry was held for 15
min at 20 °C. The crystals were filtered. The transfer was
washed with 1:1 water/MeOH and dried with suction for 1 h.
The crystals were transferred to a vacuum oven at 45 °C. L40
(
15.63 g, 90% yield) was isolated as white crystals (mp 235−
1
2
6
7
37 °C). H NMR (400 MHz, DMSO-d ): δ 2.18 (s, 6H),
6
.79 (d, J = 8.59 Hz, 1H), 7.10−7.17 (m, 3H), 7.35 (d, J =
.02 Hz, 1H), 7.75 (dd, J = 8.59, 7.02 Hz, 1H), 9.84 (s, 1H).
1
3
C NMR (100 MHz, DMSO-d ): δ 18.6, 111.0, 117.2, 127.3,
6
1
28.3, 135.0, 135.8, 140.9, 145.0, 161.7, 162.8.
6
-Hydroxy-N-(2,4,6-trimethoxyphenyl)picolinamide
(
L41). Prepared analogously to L40 on the same scale of
operations. L41 (17.1 g, 78% yield) was isolated as purple
1
crystals (mp 160−162 °C). H NMR (400 MHz, DMSO-d ):
δ 3.74 (s, 6H), 3.81 (s, 3H), 6.31 (s, 2H), 6.78 (d, J = 8.59 Hz,
6
N-(4-(tert-Butyl)phenyl)pyridin-2-amine (3). 1H NMR
1
H), 7.36 (d, J = 5.85 Hz, 1H), 7.74 (t, J = 7.81 Hz, 1H), 9.13
1
3
(
s, 1H). C NMR (100 MHz, DMSO-d ): δ 55.9, 56.2, 91.5,
(400 MHz, DMSO-d ): δ 1.27 (s, 9H), 6.69 (dd, J = 6.83,
6
6
1
07.1, 111.0, 117.2, 140.9, 145.0, 156.9, 160.1, 161.6, 162.8.
-Hydroxy-N-(2-methylnaphthalen-1-yl)picolinamide
L42). Prepared analogously to L40 on the same scale of
5.27 Hz, 1H), 6.80 (d, J = 8.20 Hz, 1H), 7.28 (d, J = 7.86 Hz,
2H), 7.50−7.58 (m, 3H), 8.11 (dd, J = 4.88, 1.37 Hz, 1H),
6
1
3
(
8.89 (s, 1H). C NMR (100 MHz, DMSO-d ): δ 31.8, 34.3,
6
operations. L42 (15.2 g, 76% yield) was isolated as yellow
110.8, 114.3, 118.5, 125.6, 137.6, 139.5, 143.2, 147.7, 156.6.
Scale-Up of N-(Pyridin-2-yl)quinolin-6-amine (12a). To a
reaction vessel under nitrogen were charged CuI (97.2 mg,
0.02 equiv) and L40 (242 mg, 0.04 equiv) in sulfolane (10 mL,
∼2 mL/g). This gave a light-brown suspension at first that
became a solution eventually and was held for 20 min at 25−
35 °C. To the reaction mixture were charged K CO (5.24 g,
1
crystals (mp 239−242 °C). H NMR (400 MHz, DMSO-d ):
6
δ 2.37 (s, 3H), 6.84 (d, J = 8.59 Hz, 1H), 7.41−7.56 (m, 4H),
7
.77−7.90 (m, 3H), 7.94 (d, J = 7.31 Hz, 1H), 10.26 (s, 1H).
1
3
C NMR (100 MHz, DMSO-d ): δ 18.8, 113.0, 115.8, 123.3,
6
1
1
25.8, 126.9, 127.5, 128.4, 129.2, 130.9, 131.1, 132.8, 133.4,
36.3, 141.0, 162.0, 162.9.
2
3
6
-Hydroxy-N-(naphthalen-1-yl)picolinamide (L43).
1.50 equiv), 11a (5.20 g, 25.0 mmol), and 2-aminopyridine
(2.59 g, 1.1 equiv, 27.5 mmol). The reaction mixture became a
yellowish-brown color and was heated to 110−120 °C. Upon
heating the reaction mixture turned a brown color. Heating
was continued for about 4 h, at which point a reaction assay
indicated that the reaction was 57% complete. The progress
had stopped because of arylation of the ligand. Additional L40
Prepared analogously to L40 on the same scale of operations.
L43 (15.6 g, 87% yield) was isolated as off-white crystals (mp
1
1
75−179 °C). H NMR (400 MHz, DMSO-d ): δ 6.93 (d, J =
6
8
.59 Hz, 1H), 7.45−7.74 (m, 4H), 7.78−7.94 (m, 2H), 8.00
(
d, J = 7.81 Hz, 1H), 8.10 (d, J = 7.61 Hz, 2H), 10.77 (s, 1H).
1
3
C NMR (100 MHz, DMSO-d ): δ 113.0, 116.0, 120.2, 122.0,
6
J
DOI: 10.1021/acs.oprd.9b00195
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
242 mg, 0.04 equiv) was charged, and heating was resumed.
Article
(
7.48 (t, J = 7.41 Hz, 2H), 7.49−7.54 (m, 1H), 7.77 (s, 1H),
7.78 (d, J = 8.4 Hz, 2H), 8.18 (dd, J = 5.07, 1.56 Hz, 1H), 8.64
After another 4 h of heating at 110−120 °C, a reaction assay
indicated that the reaction was >99% complete. The reaction
mixture was cooled to 60 °C. To the reaction mixture was
charged MeOH (20 mL, ∼4 mL/g) followed by aqueous
ammonia (20 mL, ∼4 mL/g) and water (20 mL, ∼4 mL/g).
The reaction mixture was held at 55−60 °C for 1 h and then
cooled to 22 °C over 1 h. The slurry was held at 22 °C for 2 h
and then filtered. The transfer was washed with 2:1 water/
MeOH and dried in a vacuum oven at 45 °C for 16 h. 12a
1
3
(s, 1H), 9.04 (s, 1H). C NMR (100 MHz, DMSO-d ): δ
6
109.5, 113.2, 115.6, 117.6, 125.4, 126.4, 129.5, 133.1, 137.0,
139.9, 147.6, 155.3.
6-Methoxy-N-(pyridin-2-yl)pyridin-2-amine (12i). 1H
NMR (400 MHz, DMSO-d ): δ 3.86 (s, 3H), 6.26 (d, J =
6
7.80 Hz, 1H), 6.85 (dd, J = 7.41, 4.68 Hz, 1H), 7.21 (d, J =
8.20 Hz, 1H), 7.55 (t, J = 8.00 Hz, 1H), 7.65 (ddd, J = 8.59,
7.02, 1.95 Hz, 1H), 7.81 (d, J = 8.59 Hz, 1H), 8.21 (dd, J =
5.07, 1.17 Hz, 1H), 9.54 (s, 1H). C NMR (100 MHz,
DMSO-d ): δ 53.1, 100.4, 103.2, 111.7, 115.8, 137.5, 140.3,
1
3
(
4.20 g, 76% yield) was isolated as light-tan crystals (mp 159−
1
1
6
8
8
62 °C). H NMR (400 MHz, DMSO-d ): δ 6.84 (dd, J =
6
6
.63, 5.46 Hz, 1H), 6.96 (d, J = 8.20 Hz, 1H), 7.42 (dd, J =
.20, 4.29 Hz, 1H), 7.64 (t, J = 7.70 Hz, 1H), 7.83 (dd, J =
.98, 2.34 Hz, 1H), 7.91 (d, J = 9.37 Hz, 1H), 8.20 (d, J = 7.80
147.4, 152.7, 154.2, 162.4.
N-(4-(tert-Butyl)phenyl)-3-methoxy-1-methyl-1H-pyrazol-
4-amine (13a). H NMR (400 MHz, CDCl ): δ 1.29 (s, 9H),
1
3
Hz, 1H), 8.23−8.31 (m, 1H), 8.52 (d, J = 1.95 Hz, 1H), 8.68
3.74 (br s, 3H), 3.92 (s, 3H), 4.60−4.95 (br s, 1H), 6.60−6.77
(
br d, J = 2.73 Hz, 1H), 9.43 (s, 1H). 1 C NMR (100 MHz,
3
(m, 2H), 7.14−7.24 (m, 3H). C NMR (100 MHz, CDCl ): δ
13
3
DMSO-d ): δ 111.7, 111.8, 115.4, 122.0, 124.0, 129.4, 129.7,
31.5, 33.9, 39.2, 56.2, 108.7, 113.0, 125.9, 126.6, 141.1, 144.3,
157.8.
6
1
35.2, 137.9, 140.2, 144.2, 147.8, 147.9, 156.1.
N-(2-(1H-Pyrazol-1-yl)phenyl)pyridin-2-amine (12c). 1H
6-Methoxy-N-(3-methoxy-1-methyl-1H-pyrazol-4-yl)-
1
NMR (400 MHz, DMSO-d ): δ 6.52 (t, J = 2.15 Hz, 1H),
pyridin-2-amine (13b). H NMR (400 MHz, CDCl ): δ 3.74
6
3
6
.74−6.80 (m, 2H), 7.14 (td, J = 7.71, 1.37 Hz, 1H), 7.36 (t, J
(s, 3H), 3.90 (s, 3H), 3.94 (s, 3H), 5.76 (br s, 1H), 6.08 (dd, J
= 14.05, 7.80 Hz, 2H), 7.35 (t, J = 7.81 Hz, 1H), 7.51 (s, 1H).
=
7.42 Hz, 1H), 7.50−7.57 (m, 2H), 7.83 (d, J = 1.56 Hz, 1H),
1
3
8
.06−8.11 (m, 2H), 8.19 (d, J = 1.95 Hz, 1H), 9.12 (s, 1H).
C NMR (100 MHz, CDCl ): δ 39.1, 53.3, 56.3, 98.4, 98.7,
3
1
3
C NMR (100 MHz, DMSO-d ): δ 106.9, 110.0, 115.2, 122.3,
107.3, 125.0, 139.9, 155.5, 155.9, 163.6.
6
1
1
22.4, 124.2, 127.7, 130.3, 131.0, 133.7, 137.7, 140.6, 147.6,
3-Methoxy-1-methyl-N-(1-phenyl-1H-pyrazol-4-yl)-1H-
1
55.2.
pyrazol-4-amine (13c). H NMR (400 MHz, CDCl ): δ 3.72
3
1
2
-Phenyl-N-(pyridin-2-yl)thiazol-4-amine (12d). H NMR
(s, 3H), 3.95 (s, 3H), 7.10 (s, 1H), 7.19−7.25 (m, 1H), 7.37−
1
3
(
400 MHz, DMSO-d ): δ 6.79 (dd, J = 6.44, 5.66 Hz, 1H),
7.47 (m, 3H), 7.50 (s, 1H), 7.57−7.68 (m, 2H). C NMR
6
7
.04 (d, J = 8.20 Hz, 1H), 7.46−7.55 (m, 2H), 7.55−7.63 (m,
(100 MHz, CDCl ): δ 39.1, 56.3, 113.2, 115.7, 118.2, 123.2,
3
1
H), 7.68 (s, 1H), 7.92 (dd, J = 7.81, 1.56 Hz, 2H), 8.21 (dd, J
125.6, 129.3, 131.9, 133.1, 140.3, 155.8.
=
5.07, 1.56 Hz, 1H), 10.14 (s, 1H). 13C NMR (100 MHz,
N-(3-Methoxy-1-methyl-1H-pyrazol-4-yl)quinolin-6-
1
DMSO-d ): δ 95.7, 110.8, 114.6, 125.6, 129.2, 130.0, 133.1,
amine (13d). H NMR (400 MHz, CDCl
): δ 3.79 (s, 3H),
3
6
137.1, 147.1, 151.0, 154.3, 163.3.
3.93 (s, 3H), 5.21 (br s, 1H), 6.80−6.86 (m, 1H), 7.16−7.32
1
N-(4-Chlorophenyl)pyridin-2-amine (12f). H NMR (400
(m, 3H), 7.89 (d, J = 7.41 Hz, 1H), 7.93 (d, J = 8.98 Hz, 1H),
1
3
MHz, DMSO-d ): δ 6.76 (dd, J = 6.63, 4.68 Hz, 1H), 6.82 (d,
J = 8.20 Hz, 1H), 7.28 (d, J = 8.98 Hz, 2H), 7.57 (td, J = 7.80,
8.63 (dd, J = 8.63, 4.29 Hz, 1H). C NMR (100 MHz,
CDCl ): δ 39.3, 56.3, 105.0, 107.4, 120.8, 121.3, 127.6, 129.8,
6
3
1
1
.95 Hz, 1H), 7.73 (d, J = 8.59 Hz, 2H), 8.16 (dd, J = 4.68,
130.2, 134.1, 143.4, 145.1, 146.5, 158.2.
13
.56 Hz, 1H), 9.16 (s, 1H). C NMR (100 MHz, DMSO-d ):
N-(4-Chlorophenyl)-3-methoxy-1-methyl-1H-pyrazol-4-
6
1
δ 111.0, 114.6, 119.2, 128.3, 129.1, 137.3, 140.7, 147.1, 155.5.
amine (13e). H NMR (400 MHz, CDCl
): δ 3.75 (s, 3H),
3
Scale-Up of N-(1-Phenyl-1H-pyrazol-4-yl)pyridin-2-amine
3.91 (s, 3H), 6.62 (d, J = 8.59 Hz, 2H), 7.11 (d, J = 8.59 Hz,
1
3
(
(
12h). To a reaction vessel under nitrogen were charged CuI
77.5 mg, 0.02 equiv) and L40 (194 mg, 0.04 equiv) in DMSO
10 mL, ∼2 mL/g). This gave a light-brown solution that was
2H), 7.18 (s, 1H). C NMR (100 MHz, CDCl
107.6, 114.4, 122.8, 127.5, 128.9, 145.6, 158.1.
3-Methoxy-N-(4-methoxyphenyl)-1-methyl-1H-pyrazol-4-
3
): δ 39.2, 56.2,
(
1
held for 10 min at 25−35 °C. To the reaction mixture were
charged K PO (6.50 g, 1.50 equiv), 11h (4.70 g, 20.0 mmol),
amine (13f). H NMR (400 MHz, CDCl
): δ 3.73 (s, 3H),
3
3.75 (s, 3H), 3.92 (s, 3H), 4.69 (s, 1H), 6.64−6.72 (m, 2H),
3
4
1
3
and 2-aminopyridine (2.07 g, 1.1 equiv, 22.0 mmol). The
6.73−6.81 (m, 2H), 7.15 (s, 1H). C NMR (100 MHz,
CDCl ): δ 39.2, 55.8, 56.3, 109.5, 114.7, 114.7, 126.4, 140.9,
reaction mixture became a green color and was heated to 110−
3
1
20 °C. Upon heating the reaction mixture turned a brown
152.7, 157.7.
color. After about 1 h of heating, sodium ascorbate (99 mg,
.025 equiv) was charged to the reaction mixture. Heating was
3-Methoxy-N-(2-methoxyphenyl)-1-methyl-1H-pyrazol-4-
1
0
amine (13g). H NMR (400 MHz CDCl
3
): δ 3.75 (s, 3H),
continued for about 4 h, at which point a reaction assay
indicated that the reaction was >99% complete. The reaction
mixture was cooled to 60 °C. To the reaction mixture was
charged MeOH (20 mL, ∼4 mL/g) followed by aqueous
ammonia (20 mL, ∼4 mL/g) and water (40 mL, ∼8 mL/g).
The reaction mixture was held at 60−65 °C for 1 h and then
cooled to 22 °C over 1 h. The slurry was held at 22 °C for 1 h
and then filtered. The transfer was washed with 2:1 water/
MeOH and dried in a vacuum oven at 45 °C for 16 h. 12h
3.89 (s, 3H), 3.93 (s, 3H), 4.95−5.87 (br s, 1H), 6.66−6.75
13
(m, 2H), 6.78−6.86 (m, 2H), 7.20 (s, 1H). C NMR (100
MHz, CDCl ): δ 39.2, 55.4, 56.2, 108.0, 109.7, 111.1, 117.5,
3
121.0, 126.7, 136.5, 146.6, 157.9.
N-(1-Methyl-1H-pyrazol-3-yl)isoquinolin-7-amine (14a).
1
H NMR (400 MHz, DMSO-d ): δ 3.80 (s, 3H), 5.90 (d, J
6
= 2.34 Hz, 1H), 7.34 (dd, J = 8.20, 4.29 Hz, 1H), 7.56 (dd, J =
9.37, 2.73 Hz, 1H), 7.57 (d, J = 2.34 Hz, 1H), 7.83 (d, J = 8.98
Hz, 1H), 7.89 (d, J = 2.34 Hz, 1H), 8.05 (d, J = 8.20, 1H), 8.57
1
3
(
4.26 g, 90% yield) was isolated as light-tan crystals (mp 108−
(dd, J = 4.29, 1.56 Hz, 1H), 8.87 (s, 1H). C NMR (100
1
1
10 °C). H NMR (400 MHz, DMSO-d ): δ 6.67 (t, J = 6.24
MHz, DMSO-d ): δ 38.5, 94.0, 106.5, 121.4, 122.1, 129.3,
6
6
Hz, 1H), 6.75 (d, J = 8.59 Hz, 1H), 7.26 (t, J = 7.02 Hz, 1H),
129.5, 131.1, 133.9, 141.6, 142.9, 146.2, 150.7.
K
DOI: 10.1021/acs.oprd.9b00195
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
N-(2-(1H-Pyrazol-1-yl)phenyl)-1-methyl-1H-pyrazol-3-
= 6.24 Hz, 1H), 7.17 (t, J = 7.41 Hz, 1H), 7.23 (t, J = 7.02 Hz,
1H), 7.29−7.36 (m, 3H), 7.38−7.44 (m, 4H), 7.67 (d, J = 7.80
1
amine (14b). H NMR (400 MHz, DMSO-d ): δ 3.74 (s, 3H),
6
1
3
5
(
7
1
.92 (d, J = 2.34 Hz, 1H), 6.58 (t, J = 2.15 Hz, 1H), 6.87−6.91
Hz, 2H), 7.75 (s, 1H). C NMR (100 MHz, DMSO-d ): δ
6
m, 1H), 7.24−7.31 (m, 1H), 7.42 (dd, J = 8.00, 1.37 Hz, 1H),
49.9, 110.5, 116.9, 124.7, 126.7, 127.6, 128.2, 129.3, 131.2,
136.5, 140.0, 140.2.
.55 (d, J = 2.34 Hz, 1H), 7.84−7.88 (m, 1H), 7.87−7.88 (m,
13
1
H), 8.25 (d, J = 2.73 Hz, 1H), 8.75 (s, 1H). C NMR (100
1-Phenyl-1H-pyrazol-4-amine (16f). H NMR (400 MHz,
MHz, DMSO-d ): δ 38.4, 94.7, 106.8, 116.0, 118.6, 123.5,
DMSO-d
(s, 1H), 7.42 (t, J = 7.81 Hz, 2H), 7.68 (d, J = 9.37, 3H).
NMR (100 MHz, DMSO-d ): δ 112.6, 117.5, 125.2, 129.8,
6
): δ 4.19 (br s, 2H), 7.19 (t, J = 7.38 Hz, 1H), 7.27
6
1
3
1
26.1, 128.1, 131.0, 131.4, 136.6, 140.5, 149.6.
-Methoxy-N-(1-methyl-1H-pyrazol-3-yl)pyridin-2-amine
C
6
6
1
133.2, 134.0, 140.6.
(
14c). H NMR (400 MHz, DMSO-d ): δ 3.72 (s, 3H), 3.82
6
(
6
=
s, 3H), 6.07 (d, J = 7.80 Hz, 1H), 6.39 (d, J = 2.34 Hz, 1H),
.68 (d, J = 7.80 Hz, 1H), 7.43 (t, J = 7.81 Hz, 1H), 7.49 (d, J
1.95 Hz, 1H), 9.16 (s, 1H). 13C NMR (100 MHz, DMSO-
d6): δ 38.2, 52.9, 95.2, 98.0, 100.7, 130.6, 139.9, 149.3, 153.8,
62.7.
N-(1-Methyl-1H-pyrazol-3-yl)-2-phenylthiazol-4-amine
AUTHOR INFORMATION
Corresponding Author
*E-mail: robert.a.singer@pfizer.com.
ORCID
Daniel W. Widlicka: 0000-0001-9565-7619
Robert A. Singer: 0000-0001-9730-1261
Notes
■
1
1
(
(
(
14d). H NMR (400 MHz, DMSO-d ): δ 3.74 (s, 3H), 5.80
6
d, J = 1.95 Hz, 1H), 7.06 (s, 1H), 7.43−7.52 (m, 4H), 7.89
13
dd, J = 8.00, 1.37 Hz, 2H), 9.42 (s, 1H). C NMR (100
The authors declare no competing financial interest.
MHz, DMSO-d ): δ 38.3, 91.1, 92.5, 125.5, 129.2, 129.9,
6
1
31.0, 133.2, 150.5, 153.2, 163.5.
N-(4-Chlorophenyl)-1-methyl-1H-pyrazol-3-amine (14f).
ACKNOWLEDGMENTS
■
1
H NMR (400 MHz, DMSO-d ): δ 3.74 (s, 3H), 5.75 (d, J
This work was influenced by ongoing efforts of the Non-
Precious Metal Catalysis Alliance between AbbVie, Asymchem,
Boehringer-Ingelheim, and Pfizer. The authors thank Sebastien
Monfette, Scott Bagley, and Robert Hicklin for helpful
discussions and demonstration of this methodology at Pfizer.
6
=
2.34 Hz, 1H), 7.19 (d, J = 8.98 Hz, 2H), 7.35 (d, J = 8.98
1
3
Hz, 2H), 7.51 (d, J = 1.95 Hz, 1H), 8.58 (s, 1H). C NMR
(
100 MHz, DMSO-d ): δ 38.3, 93.5, 116.0, 120.9, 128.4,
6
1
31.0, 142.6, 150.8.
N-(5-Methyl-1H-pyrazol-3-yl)quinolin-6-amine (15h). H
NMR (400 MHz, DMSO-d ): δ 2.22 (s, 3H), 5.72 (s, 1H),
1
REFERENCES
6
■
7
1
.33 (dd, J = 8.39, 4.10 Hz, 1H), 7.55 (dd, J = 8.98, 2.34 Hz,
H), 7.80 (d, J = 8.98 Hz, 1H), 7.91−7.98 (m, 1H), 8.05 (d, J
(1) (a) Transition Metal-Catalyzed Couplings in Process Chemistry:
Case Studies from the Pharmaceutical Industry; Magano, J., Dunetz, J.
R., Eds.; Wiley-VCH, 2013. (b) Busacca, C. A.; Fandrick, D. R.; Song,
J. J.; Senanayake, C. H. The Growing Impact of Catalysis in the
Pharmaceutical Industry. Adv. Synth. Catal. 2011, 353, 1825−1864.
=
(
7.80 Hz, 1H), 8.56 (dd, J = 4.29, 1.56 Hz, 1H), 8.72−8.81
m, 1H), 11.78 (br s, 1H). 13C NMR (100 MHz, DMSO-d ):
6
δ 10.7, 93.2, 106.5, 121.4, 122.1, 129.2, 129.5, 133.9, 138.4,
41.8, 142.8, 146.1, 151.1.
-Methoxy-N-(5-methyl-1H-pyrazol-3-yl)pyridin-2-amine
(
c) Roughley, S. D.; Jordan, A. M. The Medicinal Chemist’s Toolbox:
1
An Analysis of Reactions Used in the Pursuit of Drug Candidates. J.
Med. Chem. 2011, 54, 3451−3479. (d) Magano, J.; Dunetz, J. R.
Large-Scale Applications of Transition Metal-Catalyzed Couplings for
the synthesis of Pharmaceuticals. Chem. Rev. 2011, 111, 2177−2250.
(e) Torborg, C.; Beller, M. Recent Applications of Palladium-
Catalyzed Coupling Reactions in the Pharmaceutical, Agrochemical,
and Fine Chemical Industries. Adv. Synth. Catal. 2009, 351, 3027−
6
1
(
3
15i). H NMR (400 MHz, CDCl ): δ 2.28 (s, 3H), 3.92 (s,
H), 6.01 (s, 1H), 6.20 (d, J = 7.80 Hz, 1H), 6.53 (d, J = 7.80
3
Hz, 1H), 7.09−7.23 (br s, 1H), 7.42 (t, J = 7.80, 1H), 9.64−
9
9
.93 (br s, 1H). 13C NMR (100 MHz, CDCl ): δ 11.9, 53.4,
3.9, 99.8, 100.3, 140.2, 142.0, 148.4, 153.4, 163.4.
3
3
(
043.
5
-Methyl-N-(1-phenyl-1H-pyrazol-4-yl)-1H-pyrazol-3-
2) (a) Vitaku, E.; Smith, D. T.; Njardarson, J. T. Analysis of the
1
amine (15j). H NMR (400 MHz, CDCl ): δ 2.25 (s, 3H),
3
Structural Diversity, Substitution Patterns, and Frequency of Nitrogen
Heterocycles among U.S. FDA Approved Pharmaceuticals. J. Med.
Chem. 2014, 57, 10257−10274. (b) Tasler, S.; Mies, J.; Lang, M.
Applicability Aspects of Transition Metal-Catalyzed Aromatic
Amination Protocols in Medicinal Chemistry. Adv. Synth. Catal.
2007, 349, 2286−2300. (c) Cooper, T. W. J.; Campbell, I. B.;
Macdonald, S. J. F. Factors Determining the Selection of Organic
Reactions by Medicinal Chemists and the Use of These Reactions in
Arrays (Small Focused Libraries). Angew. Chem., Int. Ed. 2010, 49,
5
.59 (s, 1H), 6.38−6.80 (br s, 1H), 7.19−7.26 (m, 1H), 7.41
(
t, J = 8.00 Hz, 2H), 7.59 (s, 1H), 7.66 (dd, J = 8.78, 0.98 Hz,
13
2
1
H), 8.07 (s, 1H). C NMR (100 MHz, CDCl ): δ 11.2, 91.6,
15.6, 118.3, 125.6, 128.5, 129.2, 132.9, 140.3, 140.7, 153.6.
3
Benzyl 4-(4-(tert-Butyl)phenyl)piperazine-1-carboxylate
1
(
3
16a). H NMR (400 MHz, DMSO-d ): δ 1.24 (s, 9H),
6
.04−3.08 (m, 4H), 3.54 (br s, 4H), 5.12 (s, 2H), 6.88 (d, J =
7
.68 Hz, 2H), 7.24 (d, J = 8.59 Hz, 2H), 7.32−7.40 (m, 5H).
8
082−8091. (d) Fischer, C.; Koenig, B. Palladium- and Copper-
1
3
C NMR (100 MHz, DMSO-d ): δ 31.8, 34.1, 43.9, 49.3,
6.8, 116.4, 126.1, 128.1, 128.4, 128.9, 137.3, 142.3, 149.1,
54.9.
6
mediated N-aryl Bond Formation Reactions for the Synthesis of
Biological Active Compounds. Beilstein J. Org. Chem. 2011, 7, 59−74.
(e) Schlummer, B.; Scholz, U. Palladium-Catalyzed C−N and C−O
Coupling-A Practical Guide from an Industrial Vantage Point. Adv.
Synth. Catal. 2004, 346, 1599−1626.
6
1
Benzyl 4-(6-Methoxypyridin-2-yl)piperazine-1-carboxy-
1
late (16c). H NMR (400 MHz, CDCl ): δ 3.51−3.57 (m,
3
(
3) (a) Ruiz-Castillo, P.; Buchwald, S. L. Applications of Palladium-
Catalyzed C−N Cross-Coupling Reactions. Chem. Rev. 2016, 116,
2564−12649. (b) Surry, D. S.; Buchwald, S. L. Dialkylbiaryl
Phosphines in Pd-Catalyzed Amination: A User’s Guide. Chem. Sci.
011, 2, 27−50. (c) Hartwig, J. F. Evolution of a Fourth Generation
Catalyst for the Amination and Thioetherification of Aryl Halides.
4
H), 3.61−3.67 (m, 4H), 3.87 (s, 3H), 5.18 (s, 2H), 6.12 (d, J
=
7.81 Hz, 1H), 6.17 (d, J = 8.20 Hz, 1H), 7.29−7.49 (m, 6H).
1
1
3
C NMR (100 MHz, CDCl ): δ 43.5, 45.0, 53.0, 67.2, 98.3,
3
9
8.8, 127.9, 128.0, 128.5, 136.6, 140.2, 155.3, 157.9, 163.1.
N-Benzyl-1-phenyl-1H-pyrazol-4-amine (16e). H NMR
2
1
(
400 MHz, DMSO-d ): δ 4.15 (d, J = 6.24 Hz, 2H), 5.32 (t, J
Acc. Chem. Res. 2008, 41, 1534−1544. (d) Bariwal, J.; Van der Eycken,
6
L
DOI: 10.1021/acs.oprd.9b00195
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
E. C−N Bond Forming Cross-Coupling Reactions: an Overview.
Chem. Soc. Rev. 2013, 42, 9283−9303. (e) Beletskaya, I. P.;
Cheprakov, A. V. The Complementary Competitors: Palladium and
Copper in C−N Cross-Coupling Reactions. Organometallics 2012, 31,
(11) For a review, see: Bhunia, S.; Pawar, G. G.; Kumar, S. V.; Jiang,
Y.; Ma, D. Selected Copper-Based Reactions for C−N, C−O, C−S,
and C−C Bond Formation. Angew. Chem., Int. Ed. 2017, 56, 16136−
16179.
(12) Patent applications worldwide have been submitted for the use
of the oxalamide ligands in Cu-catalyzed couplings. The U.S. patent
application is US20180207628A1, which was published in July 2018.
(13) (a) Guo, X.; Rao, H.; Fu, H.; Jiang, Y.; Zhao, Y. An Inexpensive
and Efficient Copper Catalyst for N-Arylation of Amines, Amides and
Nitrogen-Containing Heterocycles. Adv. Synth. Catal. 2006, 348,
7753−7808.
(4) (a) Ley, S. V.; Thomas, A. W. Modern Synthetic Methods for
Copper-Mediated C(aryl)-O, C(aryl)-N, and C(aryl)-S Bond
Formation. Angew. Chem., Int. Ed. 2003, 42, 5400−5449. (b) Monnier,
F.; Taillefer, M. Catalytic C−C, C−N, and C−O Ullmann-Type
Coupling Reactions. Angew. Chem., Int. Ed. 2009, 48, 6954−6971.
2
197−2202. (b) Altman, R. A.; Anderson, K. W.; Buchwald, S. L.
(c) Ma, D.; Cai, Q. Copper/Amino Acid Catalyzed Cross-Couplings
Pyrrole-2-carboxylic Acid as a Ligand for the Cu-Catalyzed Reactions
of Aryl and Vinyl Halides with Nucleophiles. Acc. Chem. Res. 2008, 41,
of Primary Anilines with Aryl Halides. J. Org. Chem. 2008, 73, 5167−
1
450−1460. (d) Sambiagio, C.; Marsden, S. P.; Blacker, A. J.;
5
(
169.
McGowan, P. C. Copper Catalysed Ullmann Type Chemistry: from
14) Bhunia, S.; Kumar, S. V.; Ma, D. N,N′-Bisoxalamides Enhance
Mechanistic Aspects to Modern Development. Chem. Soc. Rev. 2014,
the Catalytic Activity in Cu-Catalyzed Coupling of (Hetero)aryl
4
(
3, 3525−3550.
Bromides with Anilines and Secondary Amines. J. Org. Chem. 2017,
5) (a) Ma, D.; Zhang, Y.; Yao, J.; Wu, S.; Tao, F. Accelerating Effect
8
(
2, 12603−12612.
Induced by the Structure of α-Amino Acid in the Copper-Catalyzed
Coupling Reaction of Aryl Halides with α-Amino Acids Synthesis of
Benzolactam-V8. J. Am. Chem. Soc. 1998, 120, 12459−12467. (b) Ma,
D.; Xia, C. CuI-Catalyzed Coupling Reaction of β-Amino Acids or
Esters with Aryl Halides at Temperature Lower Than That Employed
in the Normal Ullmann Reaction. Facile Synthesis of SB-214857. Org.
Lett. 2001, 3, 2583−2586. (c) Ma, D.; Cai, Q.; Zhang, H. Mild
Method for Ullmann Coupling Reaction of Amines and Aryl Halides.
Org. Lett. 2003, 5, 2453−2455. (d) Ma, D.; Cai, Q. N,N-Dimethyl
Glycine-Promoted Ullmann Coupling Reaction of Phenols and Aryl
Haildes. Org. Lett. 2003, 5, 3799−3802. (e) Zhang, H.; Cai, Q.; Ma,
D. Amino Acid Promoted CuI-Catalyzed C−N Bond Formation
between Aryl Halides and Amines or N-Containing Heterocycles. J.
Org. Chem. 2005, 70, 5164−5173. (f) Guo, X.; Rao, H.; Fu, H.; Jiang,
Y.; Zhao, Y. An Inexpensive and Efficient Copper Catalyst for N-
Arylation of Amines, Amides and Nitrogen-Containing Heterocycles.
Adv. Synth. Catal. 2006, 348, 2197−2202. (g) Cai, Q.; et al. Amino
Acid-Promoted Ullmann-Type Coupling Reactions and their
Applications in Organic Synthesis. Pure Appl. Chem. 2009, 81,
15) BTMO and BFMO were also the subject of couplings of aryl
chlorides and amides or nitrogen heterocycles. See: (a) De, S.; Yin, J.;
Ma, D. Copper-Catalyzed Coupling Reaction of (Hetero)Aryl
Chlorides and Amides. Org. Lett. 2017, 19, 4864−4867. (b) Pawar,
G. G.; Wu, H.; De, S.; Ma, D. Copper(I) Oxide/N,N’-Bis[(2-
furyl)methyl]oxalamide-Catalyzed Coupling of (Hetero)aryl Halides
and Nitrogen Heterocycles at Low Catalytic Loading. Adv. Synth.
Catal. 2017, 359, 1631−1636.
(
16) For the oxalamide ligands it has been suggested by
computational and structure−activity relationship studies that one
of the amides can switch binding from the amide nitrogen to the
carbonyl oxygen as needed during the catalytic cycle. See: Chan, V. S.;
Krabbe, S. W.; Li, C.; Sun, L.; Liu, Y.; Nett, A. J. Identification of an
Oxalamide Ligand for Copper-Catalyzed C−O Couplings from a
Pharmaceutical Compound Library. Chem. Cat Chem. 2019,
DOI: 10.1002/cctc.201900393.
(17) This binding mode for oxalamide−Cu complexes has been
observed in single-crystal X-ray structures. See: Zilic, D.; Rakvin, B.;
Milic, D.; Pajic, D.; Dilovic, I.; Cametti, M.; Dzolic, Z. Crystal
Structures and Magnetic Properties of a Set of Dihalobridged
Oxalamidato Copper(II) Dimers. Dalton Trans. 2014, 43, 11877−
2
(
27−234.
6) (a) Klapars, A.; Huang, S.; Buchwald, S. L. A General and
Efficient Copper Catalyst for the Amidation of Aryl Halides. J. Am.
Chem. Soc. 2002, 124, 7421−7428. (b) Surry, D. S.; Buchwald, S. L.
Diamine Ligands in Copper-Catalyzed Reactions. Chem. Sci. 2010, 1,
1
(
1887.
18) Ligand arylation was found in this case and in other examples
to be a competitive process throughout the reaction time frame. The
rate of ligand arylation was found to be dependent on the nature of
the aryl halide, amine, and ligand involved, in addition to the reaction
conditions. It was noted that certain substrates tended to promote
ligand arylation to a greater extent than others.
1
(
3−31.
7) (a) Xia, N.; Taillefer, M. A Very Simple Copper-Catalyzed
Synthesis of Anilines by Employing Aqueous Ammonia. Angew.
Chem., Int. Ed. 2009, 48, 337−339. (b) Lv, X.; Bao, W. A A-Keto
Ester as a Novel, Efficient, and Versatile Ligand for Copper(I)-
Catalyzed C−N, C−O, and C−S Coupling Reactions. J. Org. Chem.
(19) (a) Krueger, A. C.; et al. Aryl Uracil Inhibitors of Hepatitis C
Virus NS5B Polymerase: Synthesis and Characterization of Analogs
with a Fused 5,6-Bicyclic Ring Motif. Bioorg. Med. Chem. Lett. 2013,
2
007, 72, 3863−3867. (c) Shafir, A.; Buchwald, S. L. Highly Selective
Room-Temperature Copper-Catalyzed C−N Coupling Reactions. J.
2
3, 3487−3490. (b) Barnes, M. D.; Shekhar, S.; Dunn, T. B.;
Am. Chem. Soc. 2006, 128, 8742−8743.
Barkalow, J. H.; Chan, V. S.; Franczyk, T. S.; Haight, A. R.;
Hengeveld, J. E.; Kolaczkowski, L.; Kotecki, B. J.; Liang, G.; Marek, J.
C.; McLaughlin, M. A.; Montavon, D. K.; Napier, J. J. Discovery and
Development of Metal-Catalyzed Coupling Reactions in the Synthesis
of Dasabuvir, an HCV-Polymerase Inhibitor. J. Org. Chem. 2019, 84,
(
8) (a) Cristau, H.-J.; Cellier, P. P.; Spindler, J.-F.; Taillefer, M. Mild
Conditions for Copper-Catalyzed N-Arylation of Pyrazoles. Eur. J.
Org. Chem. 2004, 2004, 695−709. (b) Fagan, P. J.; Hauptman, E.;
Shapiro, R.; Casalnuovo, A. Using Intelligent/Random Library
Screening to Design Focused Libraries for the Optimization of
Homogeneous Catalysts: Ullmann Ether Formation. J. Am. Chem. Soc.
4
(
873−4892.
20) For diarylation impurity 6, L37 provided nearly equal amounts
2
(
000, 122, 5043−5051.
of two regioisomers of 6 (which would include arylation of the
pyridine nitrogen), while the more hindered 6-hydroxypicolinamides
L40 and L41 produced only one major isomer of 6 as an impurity
(most likely the isomer shown).
(21) As observed with Ma’s reports, hindered aryl amides and more
electron-rich amides provided more efficient catalysis. On the basis of
the structure−activity relationship of the picolinamide ligands,
coordination of Cu most likely occurs much of the time through
the two nitrogens of the ligands.
9) (a) Zhou, W.; Fan, M.; Yin, J.; Jiang, Y.; Ma, D. CuI/Oxalic
Diamide Catalyzed Coupling Reaction of (Hetero)aryl Chlorides and
Amines. J. Am. Chem. Soc. 2015, 137, 11942−11945. (b) Gao, J.;
Bhunia, S.; Wang, K.; Gan, L.; Xia, S.; Ma, D. Discovery of N-
(
Naphthalen-1-yl)-N’-alkyl Oxalamide Ligands Enables Cu-Catalyzed
Aryl Amination with High Turnovers. Org. Lett. 2017, 19, 2809−
812.
10) Zhao, J.; Niu, S.; Jiang, X.; Jiang, Y.; Zhang, X.; Sun, T.; Ma, D.
A Class of Amide Ligands Enable Cu-Catalyzed Coupling of
Hetero)aryl Halides with Sulfinic Acid Salts under Mild Conditions.
J. Org. Chem. 2018, 83, 6589−6598.
2
(
(22) We found that free-flowing Redi-Dri reagent-grade anhydrous
K PO from Sigma (product no. RDD019) provided optimal results
(
3
4
regarding water level minimization.
M
DOI: 10.1021/acs.oprd.9b00195
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
(
23) Water was removed by distillation, which reduced the operating
stable Cu(II) complex. The use of sodium ascorbate facilitated
activation of the catalyst in these instances.
(34) Various Cu salts were examined, and even though some salts
were slower to initiate catalysis, they all provided similar conversions
and product distributions. With the addition of sodium ascorbate,
minimal differences were observed among Cu salts. For example,
volume, though toluene could be replaced to maintain a constant
volume during the distillation if desired. For trials on a small scale,
this was done by allowing water to distill from vials through a small
orifice in the lid, such as a disposable needle inserted into a rubber
seal of the lid.
(24) From a processing perspective, it could be ideal to distill the
Cu O was consistently slow to initiate catalysis compared with CuI,
2
toluene at the start of the process in the absence of the Cu catalyst
and, once the water is removed by distillation, to add the Cu catalyst
to the reaction mixture and continue heating in the absence of
elevated levels of water.
but when sodium ascorbate was added, the reaction immediately
proceeded at a rate similar to that with CuI.
(35) Phase-transfer reagents (Bu
NI, Bu NBr, Bu NCl, and
4
4 4
Bu NBF ) were examined and had minimal positive effect on the
4
4
(
25) (a) Wang, Z.; Richter, S. M.; Gates, B. D.; Grieme, T. A. Safety
reaction rate. Only sodium ascorbate appeared to influence the
reaction rate in instances where the presence of unproductive Cu(II)
may have a greater tendency to impede the catalyst activation and
reaction rate or product distribution.
Concerns in a Pharmaceutical Manufacturing Process Using Dimethyl
Sulfoxide (DMSO) as a Solvent. Org. Process Res. Dev. 2012, 16,
1994−2000. (b) Yang, Q.; Cabrera, P. J.; Li, X.; Sheng, M.; Wang, N.
(36) Surprisingly, use of the aryl iodide in place of the aryl bromide
X. Safety Evaluation of the Copper-Mediated Cross-Coupling of 2-
Bromopyridines with Ethyl Bromodifluoroacetate. Org. Process Res.
Dev. 2018, 22, 1441−1447. (c) Wang, Z.; Richter, S. M.; Bellettini, J.
R.; Pu, Y.-M.; Hill, D. R. Safe Scale-Up of Pharmaceutical
Manufacturing Processes with Dimethyl Sulfoxide as the Solvent
and a Reactant or a Byproduct. Org. Process Res. Dev. 2014, 18, 1836−
did not improve the product distribution in cases where the addition
of KI to the aryl bromide appeared to do so.
(37) The coupling with 11a was found to be highly sensitive to the
concentration of the Cu catalyst relative to the substrates. Impurities
were minimized at a concentration of 1 M. At 2 M, diarylation of the
amine began to predominate. At 0.5 M, dehalogenation and
hydroxylation began to be more competitive with the desired
coupling.
1
(
842.
26) DBU was found to undergo coupling with the aryl bromide and
after workup hydrolyzed to give the adduct shown below:
(38) This observation suggests that the order of addition may not
matter in many cases. Typically, the ligand and CuI were combined in
the solvent prior to addition of any substrates to ensure optimal
formation of the catalyst. Trials in which the CuI and ligand were
added to the substrates last provided comparable results, suggesting
that the order of addition may not matter. It is suspected that with
strongly binding substrates the premixing of the ligand and CuI may
be more important, but this has not been observed.
(
27) Using toluene or anisole as a minor solvent component with
(39) Oxidative dimerization of (hetero)arenes promoted by Cu has
DMSO or sulfolane did not lead to significant erosion in the reaction
rate, but when toluene or anisole was used as the main solvent by
volume, the reaction rate decreased substantially, such that a reaction
that was typically complete in 3−6 h now required 24−30 h.
been studied. See: (a) Morgan, B. J.; Xie, X.; Phuan, P.-W.;
Kozlowski, M. C. Enantioselective synthesis of Binaphthyl Polymers
Using Chiral Asymmetric Phenolic Coupling Catalysts: oxidative
Coupling and Tandem Glaser/Oxidative Coupling. J. Org. Chem.
(28) Primary alcohols coupled readily with aryl halides under certain
2
007, 72, 6171−6182. (b) Do, H.-Q.; Daugulis, O. An Aromatic
Glaser-Hay Reaction. J. Am. Chem. Soc. 2009, 131, 17052−17053.
c) Monguchi, D.; Yamamura, A.; Fujiwara, T.; Somete, T.; Mori, A.
reaction conditions, especially those involving KOtBu in the alcohol.
Cu-catalyzed C−O couplings with 10 and related ligands will be
reported in due course.
(
Oxidative Dimerization of Azoles via Copper(II)/Silver(I)-Catalyzed
CH Homocoupling. Tetrahedron Lett. 2010, 51, 850−852. (d) Li, Y.;
Jin, J.; Qian, W.; Bao, W. An Efficient and Convenient Cu(OAc)2/Air
Mediated Oxidative Coupling of Azoles via C-H Activation. Org.
Biomol. Chem. 2010, 8, 326−330. (e) Zhu, M.; Fujita, K.; Yamaguchi,
R. Efficient Synthesis of Biazoles by Aerobic Oxidative Homocoupling
of Azoles Catalyzed by a Copper(I)/2-Pyridonate Catalytic System.
Chem. Commun. 2011, 47, 12876−12878. (f) Allen, S. E.; Walvoord,
R. R.; Padilla-Salinas, R.; Kozlowski, M. C. Aerobic Copper-Catalyzed
Organic Reactions. Chem. Rev. 2013, 113, 6234−6458.
(
29) 1-Butanol was observed to undergo significant C−O coupling
with 1 under more dilute conditions with higher amounts of K PO .
3
4
In contrast, sec-butanol performed more closely to cyclohexanol, with
high reactivity and relatively high amounts of diarylated amine but no
sign of C−O coupling. Under the conditions of Table 4, 1-butanol
showed no C−O coupling but did undergo significant diarylation to
yield 6 as a major impurity. Additional optimization will be required
to increase the efficiency of C−O couplings with this ligand family 10.
It is likely that at lower temperatures and through the use of sodium
ascorbate to allow facile catalyst activation, 1-butanol could be used
efficiently with fewer side reactions.
(
40) (a) Fan, M.; Zhou, W.; Jiang, Y.; Ma, D. Assembly of Primary
Hetero)Arylamines via CuI/Oxalic Diamide-Catalyzed Coupling of
Aryl Chlorides and Ammonia. Org. Lett. 2015, 17, 5934−5937.
b) Gao, J.; Bhunia, S.; Wang, K.; Gan, L.; Xia, S.; Ma, D. Org. Lett.
017, 19, 2809−2812. (c) Schranck, J.; Tlili, A. Transition-Metal-
Catalyzed Monoarylation of Ammonia. ACS Catal. 2018, 8, 405−418.
d) Lang, F.; Zewge, D.; Houpis, I. N.; Volante, R. P. Amination of
Aryl Halides Using Copper Catalysis. Tetrahedron Lett. 2001, 42,
251−3254. (e) Xia, N.; Taillefer, M. A Very Simple Copper-
(
(30) This reactivity trend correlating to the ligand to Cu ratio varied
in magnitude with different substrate combinations. While this effect
was almost absent in some cases, in others the rate enhancement was
amplified dramatically.
(
2
(31) It has been shown that Cu(II) salts can be present in cross-
(
couplings. See: (a) Sambiagio, C.; Munday, R. H.; Marsden, S. P.;
Blacker, A. J.; McGowan, P. C. Picolinamides as Effective Ligands for
Copper-Catalysed Aryl Ether Formation: Structure-Activity Relation-
ships, Substrate Scope and Mechanistic Investigations. Chem. - Eur. J.
3
Catalyzed Synthesis of Anilines by Employing Aqueous Ammonia.
Angew. Chem., Int. Ed. 2009, 48, 337−339. (f) Jiang, L.; Lu, X.; Zhang,
H.; Jiang, Y.; Ma, D. CuI/4-Hydro-L-Proline as a More Effective
Catalytic System for Coupling of Aryl Bromides with N-Boc
Hydrazine and Aqueous Ammonia. J. Org. Chem. 2009, 74, 4542−
2
014, 20, 17606−17615. (b) Sherborne, G. J.; Adomeit, S.; Menzel,
R.; Rabeah, J.; Bruckner, A.; Fielding, M. R.; Willans, C. E.; Nguyen,
̈
B. N. Origins of High Catalyst Loading in Copper(I)-Catalysed
Ullmann-Goldberg C−N Coupling Reactions. Chem. Sci. 2017, 8,
4
(
546.
7
203−7210.
41) 4-Nitropyrazoles have the potential to be explosive compounds.
(
32) These observations relating to Cu(II) are consistent with those
previously reported for Cu-catalyzed C−N couplings in ref 31.
33) Catalyst activation appeared to be more problematic with
substrates capable of chelating Cu, most likely rendering a relatively
Halogenation and coupling with ammonia (or a surrogate) provides a
viable alternative to avoid this issue.
(
N
DOI: 10.1021/acs.oprd.9b00195
Org. Process Res. Dev. XXXX, XXX, XXX−XXX