Safe, Scalable, Inexpensive, and Mild Nickel-Catalyzed Migita-Like C?S Cross-Couplings in Recyclable Water

Article abstract of DOI:10.1002/anie.202013017

A new approach to C?S couplings is reported that relies on nickel catalysis under mild conditions, enabled by micellar catalysis in recyclable water as the reaction medium. The protocol tolerates a wide range of heteroaromatic halides and thiols, including alkyl and heteroaryl thiols, leading to a variety of thioethers in good isolated yields. The method is scalable, results in low residual metal in the products, and is applicable to syntheses of targets in the pharmaceutical area. The procedure also features an associated low E Factor, suggesting a far more attractive entry than is otherwise currently available, especially those based on unsustainable loadings of Pd catalysts.

Full text of DOI:10.1002/anie.202013017

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Safe, Scalable, Inexpensive, and Mild Nickel-Catalyzed Migita-
like C–S Cross-Couplings in Recyclable Water
Authors: Bruce Howard Lipshutz, Tzu-Yu Yu, Haobo Pang, Yilin Cao,
and Fabrice Gallou
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202013017
Link to VoR: https://doi.org/10.1002/anie.202013017

10.1002/anie.202013017
Angewandte Chemie International Edition
COMMUNICATION
Safe, Scalable, Inexpensive, and Mild Nickel-Catalyzed Migita-like
C–S Cross-Couplings in Recyclable Water
Tzu-Yu Yu,[a] Haobo Pang,[a] Yilin Cao,[a] Fabrice Gallou,[b] and Bruce H. Lipshutz*[a]
[a]
[b]
T.-Y. Yu, H, Pang, Y. Cao, B. H. Lipshutz
Department of Chemistry and Biochemistry, University of California, Santa Barbara, CA 93106 (USA)
E-mail: lipshutz@chem.ucsb.edu
F. Gallou
CHAD, Novartis Pharma AG
4057 Basel, Switzerland
Abstract: A new approach to C–S couplings is reported that relies on
nickel catalysis under mild conditions, enabled by micellar catalysis in
recyclable water as reaction medium. The protocol tolerates a wide
range of heteroaromatic halides and thiols, including alkyl and
heteroaryl thiols, leading to a variety of thioethers in good isolated
yields. The method is scalable, results in low residual metal in the
products, and is applicable to syntheses of targets in the
pharmaceutical area. The procedure also features an associated low
E Factor, suggesting a far more attractive entry than is otherwise
currently available, especially those based on unsustainable loadings
of Pd catalysts.
S
OH
N
N
N
O
N
S
N
O
Cl
Br
S
N
N
OH
O
S
S
N
Cl
Cl
Cl
O
N
Clothiapine
neuroleptic
Butoconazole
antifungal
Lesinuard
uric acid reabsorption inhibitor
Zofenopril
anticonvulsant
S
S
O
O
N
NH
H
O
P
O
OH
O
HO
O
O
N
HO
HO
N
S
Cl
O
O
NH
NH
N
O
P
P
O
N
N
Cl Cl
O
O
HO
OH
Thioethers are widely distributed throughout nature, including
being found in numerous physiologically active compounds.
Unfortunately, Migita cross-couplings that lead to carbon-sulfur
(C–S) bond formation remain challenging in several ways,
including the typically high loadings of endangered Pd[1] catalysts
attributed to strong coordination of thiolates to the metal,
oftentimes leading to catalyst deactivation and hence, overall low
efficiency.[2] Perhaps not surprisingly, therefore, development of
methodologies aimed at construction of C–S bonds remains a
topic of considerable interest.[3] Much of the effort, however,
focuses on use of precious and expensive metals, such as
palladium,[4] iridium,[5] rhodium,[6] and along with nickel[7] and
ruthenium,[7] high temperatures are typically needed in waste-
generating organic solvents.[8] Recently, alternative routes have
emerged that call for milder conditions using photoredox
catalysis,[5][9] and less costly metals such as copper,[10] cobalt,[11]
or nickel.[5][7][13] Nonetheless, they oftentimes rely on the presence
of additional expensive metals (e.g., Ir),[5] can involve forcing
conditions, and always require organic solvents, sometimes to the
strict exclusion of moisture.[7] Moreover, most rarely include direct
applications to highly functionalized products, especially those
characteristic of pharmaceuticals (Fig. 1). And while recent
reports involving electrochemical approaches look enticing,[12]
they also tend to involve high loadings of metal catalyst and are
run in dipolar aprotic media, e.g., DMF, which are either being
phased out or are currently prohibited, e.g., under REACH in the
EU.[14] Clearly, an alternative protocol that replaces dangerously
flammable and toxic organic solvents with safe, recyclable water,
and that minimizes the investment of both energy and a metal
catalyst that is also subject to recycling would help greatly in
working towards an environmentally responsible solution to most
of these challenging issues. In this report we describe such a
process that relies on low levels of base metal (nickel) catalysis
enabled by aqueous micellar catalysis. This new technology
involves a readily available catalyst, is used under mild conditions,
N
S
Cl
CF3
N
Fenticonazole
antifungal
Cangrelor
antiplatelet
Febantel
anthelmintic
Figure 1. Selected examples of therapeutic agents bearing
aromatic/heteroaromatic thioethers.
and is scalable; hence, it is both environmentally responsible as
well as available for immediate use.
Pre-ligated nickel in the form of Ni(Phen)2Br2 (Phen =
phenanthroline) was selected as pre-catalyst to test its activity in
the coupling between decanethiol and 4-methoxyiodobenzene in
2 wt % TPGS-750-M aqueous solution (Table 1). Remarkably, the
desired C−S cross-coupling occurred smoothly using only a 5%
molar excess of thiol. Complete conversion occurred in the
presence of only 2 mol % of this Ni(II) species, in the presence of
zinc nanopowder (2 equiv) and anhydrous K3PO4 (2 equiv; entry
1). Inferior results were noted upon reducing the amount of pre-
catalyst to 0.35 mol % or below (entries 2–4). Decreasing the
base to 1.2 equivalents, fortunately, did not affect conversion
(entry 5), but the necessity of its presence for (presumably)
activation of the nucleophile was established (entry 6). Running
this coupling at room temperature (22 °C) gave a significant
reduction in yield over the same period of time at 45 °C (entry 7).
Notably, screening the loading of Zn in the 0.1−1.0 equivalent
range confirmed that only 0.25 equivalents were required (entries
8–11). By contrast, previous literature using the Ni/Zn
combination involves super-stoichiometric zinc.[15] When air
remained within the reaction vessel (entry 13) reaction efficiency
was reduced, as thiols, not surprisingly, were oxidized under
these mildly basic conditions (pH = 7–9) to disulfides.[16]
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202013017
Angewandte Chemie International Edition
COMMUNICATION
Table 1. Optimization of equivalents of catalyst and base.[a]
Table 2. Screening of pre-catalyst and base.[a]
Ni catalyst
Zn nanopowder (0.25 equiv)
base 1.2 equiv
(0.7 mol%)
I
S
(
)
Ni(Phen)2Br2 (mol %)
Zn nanopowder (equiv)
K3PO4 (equiv)
+
SH
2 wt % TPGS-750-M/H2O (0.5
45 oC, 20 h, Ar
M)
I
S
O
O
4
+
2
5
2 wt % TPGS-750-M/H2O (0.5 M)
O
(1.5 equiv)
45 oC, 20 h, Ar
SH
(1.05 equiv)
O
Entry
Base
Ni source
Ligand
Yield [%][b]
1
3
2
1
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
KOt-Bu
Cs2CO3
K3PO4
K3PO4
K3PO4
NiBr2
NiBr2
NiBr2
NiBr2
NiBr2
NiBr2
Ni(OAc)2
NiCl2
NiI2
Phen
bpy
57 (66)
Entry
1
Ni(Phen)2Br2
2 mol %
Zn (equiv)
K3PO4 (equiv)
Conv. [%][b]
100
77
2
59
2
2
3
DPPF
DPPB
DPPE
19
2
0.35 mol %
0.07 mol %
0.0035 mol %
2 mol %
2
2
4
5
3
2
2
79
5
5
4
2
2
53
6
neocuproine
Phen
Phen
Phen
Phen
Phen
Phen
–
0
5
2
1.2
0
100
0
7
18
6
2 mol %
2
8
8
7
2 mol %
2
2
6[c]
9
7
8
0.7 mol %
0.7 mol %
0.7 mol %
0.7 mol %
0.7 mol %
0.7 mol %
1
1.2
1.2
1.2
1.2
1.2
1.2
100
99
10
11
12
13
14
NiBr2
NiBr2
NiBr2
–
94 (82)
74 (71)
0[c]
9
0.5
0.25
0.1
0.5
0.5
10
11
12
13
94 (96)
83
0
100
62[d]
NiBr2
Phen
0[d]
[a] Scale of reaction: 0.25 mmol of 4-iodoanisole and 2 wt % TPGS-750-M/H2O
(0.5 mL), Ni : ligand = 1 : 2 molar ratio. [b] Yield determined by 1H NMR using
1,3,5-trimethylbenzene as internal standard. Isolated yields in parenthesis. [c]
Run without Zn. [d] Mn, ascorbic acid, or PMHS used instead of Zn.
[a] Scale of reaction: 0.25 mmol of 4-iodoanisole and 2 wt % TPGS-750-M/H2O
(0.5 mL). [b] Conversion determined by 1H NMR. Isolated yields in parenthesis.
[c] Run at rt (22 °C). [d] Run in air; disulfide formed.
In comparison with other ligands, bipyridine gave similar yields
to those obtained using phenanthroline (Table 2, entries 1–2).
Phosphine ligands, however, including DPPF, DPPE, and DPPB,
led to significantly lower yields (entries 3–5). Locating two methyl
groups on phenanthroline (i.e., neoocuproine) also impaired
reaction efficiency (entry 6). Moreover, the nature of the
counterion in the initial nickel salt was found to be crucial, as
switching from bromide to chloride, iodide, or acetate afforded
inferior results (entries 7–9). Alternative bases, including both
Cs2CO3 and KOt-Bu, were found to be suitable choices upon
screening (entries 10–11). Control studies revealed that both the
nickel catalyst and zinc powder were essential elements (entries
12–13). The reaction failed completely when Zn was replaced by
manganese,[13c] or other weak reducing agents[17] (e.g., ascorbic
acid or polymethylhydrosiloxane (PMHS); entry 14).
Table 3. Catalyst screening.[a]
Ni catalyst
Zn nanopowder (0.25 equiv)
O
O
K3PO4 (1.2 equiv)
O
SH
O
+
I
2 wt % TPGS-750-M/H2O (0.5 M)
45 oC, 20 h, Ar
S
6
7
8
Entry
[Ni]
Catalyst
Conv. [%][b]
1
2
3
4
5
1 mol %
1 mol %
1 mol %
1 mol %
0.7 mol %
Pre-mixed NiBr2 : Phen = 1 : 1[c]
Pre-mixed NiBr2 : Phen = 1 : 2[c]
NiBr2 : Phen = 1 : 2[d]
87
88
81
[e]
recrystallized Ni(Phen)2Br2
94
[e]
recrystallized Ni(Phen)2Br2
95 (95)
Further evaluation of reaction conditions indicated that pre-
ligation of Ni performed in limited amounts of acetonitrile gave the
desired coupling product in 87-88% yield (Table 3, entries 1–2).
Adding the catalyst and ligands directly, but separately, into the
reaction flask gave a slightly lowered yield (81%; entry 3). Use of
recrystallized catalyst had a significant impact on the loading,
which could be further reduced to only 0.70 mol %, while the
isolated yield jumped to 95% (entries 4–5). An X-ray crystal
structure of this octahedrally configured Ni(II) pre-catalyst
Ni(Phen)2Br2 is shown in Figure 2.
[a] Scale of reaction: 0.25 mmol of 5-iodo-2-furaldehyde and 0.5 mL of 2 wt %
TPGS-750-M/H2O. [b] Conversion determined by 1H NMR. [c] NiBr2 and
phenanthroline were pre-ligated in acetonitrile and then dried. [d] NiBr2 and
phenanthroline were used as received. [e] Ni(Phen)2Br2 was recrystallized from
EtOAc/hexane.
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202013017
Angewandte Chemie International Edition
COMMUNICATION
Table 4. Evaluation of the reaction medium.
(a) reaction 1[a]
Ni(Phen)2Br2
Zn nanopowder (0.25 equiv)
base 1.2 equiv
(0.7 mol%)
I
S
(
)
+
SH
solvent (0.5 M)
45 oC, 20 h, Ar
O
O
5
4
2
(1.5 equiv)
Entry
Solvent
Yield [%][b]
Figure 2. X-ray structure of Ni(Phen)2Br2 (CCDC 1993760). Hydrogen atoms
are omitted for clarity. Refined formula: C24H16Br2N4Ni, formula weight Mr:
578.90. Selected bond lengths (Å): Ni1–Br1, 2.530; Ni1–Br2, 2.596; Ni1–N1,
2.105; Ni1–N2 2.063; Ni1–N3 2.092; Ni1–N 2.061.
base = K3PO4
base = KOt-Bu
1
2
3
DMF
32
6
99
59
90
acetonitrile
2 wt % TPGS-750-M
To demonstrate the crucial role played by the nanoreactors
formed by the surfactant in water, comparison reactions in organic
solvents (Table 4a) were studied. Under optimized conditions,
nanoreactors present in aqueous solution performed
competitively with commonly used, dipolar organic solvents such
as DMF and acetonitrile. As shown in Table 4b, several
commercially available nonionic surfactants were also evaluated.
When water alone, or 2 wt % PEG 2000 in water was used, lower
conversions were seen, especially in the case of more challenging
reaction partners (reaction 2), mainly due to substrate polarity
leading to adherence to the stirrer (Figure S1). The presence of
surfactants emulsified these materials, raising the overall
conversion from 56% to ca. 95%. A control experiment using this
electron deficient pyridinyl bromide in reaction 2 showed, in the
absence of metals, the reactivity is significantly reduced (Scheme
S1). In the case of reaction 3, designer surfactant TPGS-750-M
led to superior results compared to other amphiphiles, although
the background reaction in its absence (i.e., “on water”)[18] was
somewhat competitive in this particular case (entry 1).
Under optimized conditions, the scope of these C−S couplings
was extensively explored. As summarized in Table 5, most
combinations of thiols and aryl iodides/bromides afforded coupled
products in good-to-excellent yields.[19] Although alkyl thiols have
previously been problematic in such cross-couplings due to their
strong nucleophilicity that can compete with a ligand on
palladium,[13d,20] both primary (leading to products 3, 8, 11, 15, 16,
26, 31, 32) and secondary alkyl (affording products 20, 29) thiols
gave the desired thioethers to the extent of 55–96%. The reaction
also took place smoothly when more complex heterocyclic thiols
were involved, including use of a thiazoline giving 19, an
oxadiazole leading to 21, a thiadiazole producing 22, a triazole
arriving at 23, a benzothiazole yielding 24), and a furan generating
32.
58
(b) reaction 2[c]
Ni(Phen)2Br2 (1.4 mol %)
Zn nanopowder (0.25 equiv)
Cs2CO3 (1.2 equiv)
Br
S
BocHN
SH
+
BocHN
N
N
aqueous medium (0.5
55 oC, 4 h, Ar
M)
NO2
NO2
11
9
10
(1.05 equiv)
(c) reaction 3[c]
Ni(Phen)2Br2 (1.4 mol %)
Zn nanopowder (0.25 equiv)
Cs2CO3 (1.2 equiv)
Br
S
SH
S
S
+
N
N
aqueous medium (0.5
55 oC, 2 h, Ar
M)
F
F
12
(1.05 equiv)
14
13
Entry
Aqueous medium
Conversion [%] [d]
reaction 2
reaction 3
1
2
3
4
5
6
7
Pure water
56
55
96
95
95
94
95
61
20
6
2 wt % PEG 2000/H2O
2 wt % Brij 30/H2O
2 wt % Triton X 100/H2O
2 wt % PTS 600/H2O
2 wt % Tween 60/H2O
2 wt % TPGS-750-M/H2O
15
19
13
66
[a] Scale of reaction: 0.25 mmol of 4-iodoanisole (2) in the designated medium
(0.5 mL). [b] Yield determined by 1H NMR using 1,3,5-trimethylbenzene as
internal standard. Isolated yield in parentheses. [c] 0.125 mmol 5-nitro-2-
bromopyridine or 4-fluoroiodobenzene in the designated aqueous medium (0.25
mL). [d] Conversion1 determined by proton NMR.
Nickel-based catalysts have been known to lose activity due
to metal chelation by the presence of heteroatoms within either or
both reaction partners, as well as by the products formed.[22]
Hence, it is noteworthy that heterocyclic aryl iodides, such as
Insofar as the aryl halide coupling partner is concerned, aryl
iodides, either electron-donating (e.g., containing methoxy; see
products 1 and 5), or electron-withdrawing (e.g., bearing fluoro,
as in 15 and trifluoromethyl, as in 19), delivered the corresponding
thioethers in good-to-excellent yields (70–96%). Perhaps more
importantly, aryl iodides bearing reducible functional groups, such
as ketone (in 16 and 21), aldehyde (see thioethers 8, 17, and 29),
and ester (in 26 and 31), were tolerated notwithstanding the
presence of zinc nanopowder.[21] Aryl bromides, whether activated
(e.g., leading to products 11 and 28-33) or not (see product 3)
appear to undergo coupling under otherwise identical conditions.
those containing
a furan (see 8), thiophene (as in 18),
methylenedioxybenzenepyrazole (product 22), or pyrazole ring
(products 20 and 25), all participated with similar efficiency.
Moreover, a variety of heteroaryl bromides leading to products 28
to 33, likewise, can be used when the loading of nickel pre-
catalyst was increased from 0.7 to 1.4 mol % (or to 3 mol % in
rare cases).
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202013017
Angewandte Chemie International Edition
COMMUNICATION
Table 5. Substrate scope.
Ni(Phen)2Br2 (0.7–1.4 mol %)
100
90
80
70
60
50
40
30
20
10
0
91
Zn nanopowder (0.25–0.50 equiv)
base (1.2 equiv)
X
S
84
our method
80
R
Het
(X = Br, I)
78
Het
+
SH
R
2 wt % TPGS-750-M/H2O (0.5 M)
45–55 oC, overnight, Ar
literature
(1.05 equiv)
(base = K3PO4, Cs2CO3, or KOtBu)
54
aryl iodides
41
O
F
S
S
O
C10H21
S
C10H21
29
O
S
O
3; X = I, 96%i
5; 82%j
8; 95%i
15; 89%i
X = Br, 88%f,i
S
S
O
S
0
O
S
O
O
Cl
NH
N
N
16; 86%c,i
17; 88%c,i
18; 87%c,i
N
S
S
S
S
O
S
S
S
S
O
N
N
N
N
O
O
CF3
19; 70%b,c,e,i
N
O
N
20; 79%b,c,e,i
21; 85%b,c,e,g,j
34
35
36
37
O
S
S
O
S
N
S
N
O
N
Figure 3. Selected examples of therapeutic agents bearing aromatic thioethers.
Direct comparison with literature conditions. Prior art conditions: For 34 and 35:
[Ir] (2 mol %), NiCl2 glyme (10 mol %), dtbbpy (15 mol %), pyridine (2 equiv), 34
W blue-LED, MeCN (0.1 M).[5a] For 36: CuI (10 mol %), NaOt-Bu (1 equiv), 100-
wat Hg lamp, MeCN (0.3 M), 0 °C, 5 h.[10a] For 37: CoI2(dppe) (1 mol %), Zn (1.5
equiv), pyridine (1 equiv), MeCN (0.25 M), 80 °C, 10 h.[11a] Current conditions:
1.4 mol % Ni(Phen)2Br2, 0.25 equiv Zn, 1.2 equiv base, stirred in 2 wt % TPGS-
750-M/H2O (0.5 mL), 55 °C, 23 h. For 36 and 37: 0.7 mol % Ni(Phen)2Br2 was
used, stirred at 45 °C, 18 h. Isolated yields.
N
O
CF3
N
N
S
HN
24; 85%b,c,e,i
22; 89%b,c,e,i
23; 73%b,c,e,g,i
O
O
O
F
S
N
N
Cl
O
S
O
N
BocHN
N
S
25; 26%b,i
26; 80%b,k
27; 89%f,h,k
aryl bromides
NO2
S
S
S
O
S
N
O
BocHN
S
N
N
O
11; 90%b,e,k
28; 74%f,k
29; 57%b,f,k
30; 54%d,k
O
O
H
S
N
S
N
S
N
O
N
N
N
Br
Br
31; 55%e,k
32; 85%b,d,k
33; 99%f,h,k
As a meaningful application of this chemistry in water, the
penultimate intermediate to Pfizer’s antitumor agent axitinib (43)
was prepared using aqueous micellar catalysis technology
(Scheme 1a). Commercially available disulfide 38 was reduced
with NaBH4 in aqueous TPGS-750-M forming thiol 39, generated
in gram quantities. Subsequent coupling with aryl iodide 40 led to
the benzopyrazole-protected sulfide, taking place in water under
very mild conditions rather than the traditional use of harmful
amide solvents at the reported high temperatures.[23] Without
isolation, methanolic HCl was added to remove the THP group to
afford 41 in 69% isolated yield over both steps.
[a] Reaction conditions unless otherwise noted: 0.25 mmol aryl bromide/iodide,
0.263 mmol thiol, 0.70 mol % Ni(Phen)2Br2, 0.25 equiv Zn, 1.2 equiv base,
stirred in 2 wt % TPGS-750-M/H2O (0.5 mL), 45 °C. Isolated yields; see SI for
details. [b] Run at 55 °C. [c] 0.5 equiv Zn. [d] 1 mol % Ni(Phen)2Br2.[e] 1.4 mol %
Ni(Phen)2Br2. [f] 3 mol % Ni(Phen)2Br2. [g] 10 v/v % DMSO added. [h] 10 v/v %
EtOAc added. [i] K3PO4. [j] KOt-Bu. [k] Cs2CO3.
Particularly instructive were side-by-side comparison studies
involving the preparation of problematic sulfides chosen from prior
art. Examples included cases leading to pyrazoles 34 and 35, as
well as ortho-substituted aromatics 36 and 37. As summarized in
Figure 3, photocatalysis conditions leading to compounds 34 and
35, calling for 2 mol % of an iridium catalyst together with 10
mol % nickel(II), in addition to being run in organic solvents with a
costly catalyst.[5a] Conditions associated with formation of both 36
involved 10 mol % CuI,[10a] while super-stoichiometric zinc at high
temperature (80 °C) were needed to synthesize compound 37,[11a]
neither of which is synthetically competitive.
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202013017
Angewandte Chemie International Edition
COMMUNICATION
(a) Gram-scale synthesis en route to axitinib; residual metal analysis
O
Eosin Y (5 mol %)
NBS (2 equiv)
NH
HN
Br
O
NH
SH
O
Br
O
S
S
O
NaBH4, 10 % v/v THF
rt, white light
I
N
Br
Br
2 wt % TPGS-750-M/H2O
N
2 wt % TPGS-750-M/H2O
(quantitative)
(0.5
M)
48
49
40
2.1 g (6.4 mmol)
39
38
O
1.1 g (6.7 mmol)
HS
O
O
Br
Ni(Phen)2Br2 (1 mol %)
Zn nanopowder (0.25 equiv)
O
O
S
O
NH
1)
iodide 40
Ni(Phen)2Br2 (2.5 mol %)
Zn nanopowder (0.3 equiv)
K3PO4 (1.2 equiv)
H
N
S
Cs2CO3 (1.2 equiv)
I2, KOH
10 % v/v NMP, 60 o
45 o
C
N
50; 76%
10 % v/v acetone, 45 o
2 wt % TPGS-750-M/H2O
C
C
2 wt % TPGS-750-M/H2O
2)
1
M
HCl, MeOH, 60 o
C
41
1.2 g (4.4 mmol)
(69%, 2 steps)
residual Ni in 41: 9.8 ppm
FDA allowance: <25 ppm
Scheme 2. 1-Pot, 2-reaction sequence.
O
NH
O
NH
S
H
H
The aqueous reaction mixture, used throughout at substrate
global concentration of 0.50 M containing 2 wt % TPGS-750-M,
could be re-used in three additional reactions (Scheme 3) thereby
further minimizing generation of waste water.[27] By virtue of the
sub-stoichiometric amount of zinc required, reaction mixtures
stirred easily, and “in-flask” extraction with minimal and
recoverable MTBE allowed for isolation of the desired product.
The E Factor[28] based on the organic solvent employed was only
4.6, suggesting, overall, a sustainable process is in hand for
thioether bond formation.
S
N
N
ref. 23
N
N
I
42 (66%)
N
43 (axitinib)
kidney cancer treatment
($297 million/y)
(b) Synthesis of a vortioxetine intermediate; residual metal analysis
HN
Cl
Ni(Phen)2Br2 (1.4 mol %)
Zn nanopowder (0.25 equiv)
K3PO4 (1.2 equiv )
I
N
S
Cl
S
ref. 24
+
2 wt % TPGS-750-M/H2O
HS
(0.5
M C
), 55 o
Ni(Phen)2Br2 (1.0 mol %)
Zn nanopowder (0.25 equiv)
Cs2CO3 (1.2 equiv)
44
45
46 (71%)
47 (vortioxetine)
antidepressant
($826 million/y)
residual Ni in 46: 1.0 ppm
FDA allowance: ≤25 ppm
+
S
F
I
F
C10H21 SH
TPGS-750-M/H2O (0.5
45 oC, 20 h, Ar
M)
C10H21
1
15
51
mass of total waste
mass of product
E Factor =
Scheme 1. Synthetic routes towards intermediates en route to axitinib and
vortioxetine.
E Factor (organic solvent) = 4.6
E Factor (including base, water) = 7.6
94
100
50
0
93
92
88
Importantly, based on ICP-MS results, residual nickel found in
axitinib precursor 41 was 9.8 ppm, which is well under the limit
permitted under FDA guidelines (≤25 ppm). Iodination could also
be achieved under aqueous surfactant conditions, thereby
reducing the amount of NMP by 90% relative to the literature route.
recycling
study:
[23]
Further extension to key precursor 46 of the antidepressant
initial
1st
reaction recycle recycle recycle
2nd
3rd
vortioxetine[24] (47; Scheme 1b) could also be realized in 71%
yield. Similar analysis of this product indicated only 1.0 ppm
residual nickel.
Thioether formation can be demonstrated in tandem with
photocatalysis in aqueous solution (Scheme 2). Initially, 4-bromo-
phenylacetylene was converted to α,α-dibromoketone 49 under
visible light irradiation, following the method recently developed
by Handa and co-workers.[25] Subsequent nickel catalyzed
thioetherification could then be effected concurrent with
monodebromination facilitated by base present in the aqueous
solution.[26] Noteworthy is the observation that the α-bromoketone
product 50 remained intact under these aqueous micellar
conditions.
Scheme 3. Determination of E Factors, and recycling study.
In summary, a mild and recyclable Ni-catalyzed C−S cross-
coupling reaction has been developed that takes place in
recyclable water under environmentally responsible micellar
aqueous conditions. The process effectively promotes thio-
etherification at sp2 carbon centers with both aryl and alkyl thiols
that is not only general, but is also tolerant of a variety of
functionality in either or both reaction partners. Opportunities to
use the aqueous medium for tandem, 1-pot applications are also
demonstrated. Overall, this new technology offers an inexpensive,
safe, scalable, and reliable route to aryl/heteroaryl sulfides that
may find, in particular, applications to active pharmaceutical
ingredients (APIs).
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202013017
Angewandte Chemie International Edition
COMMUNICATION
[16]
[17]
G. A. Bagiyan, I. K. Koroleva, N. V. Soroka, A. V. Ufimtsev, Russ. Chem.
Bull. 2003, 52, 1135–1141.
Acknowledgements
P.-Z. Wang, J.-R. Chen, W.-J. Xiao, Org. Biomol. Chem. 2019, 17,
6936–6951.
Financial support provided by the NSF (CHE 1856406) and
Novartis is warmly acknowledged with thanks. Support was also
provided, in part, by an NSF Major Research Instrumentation
award, MRI-1920299, for the purchase of NMR equipment. The
authors also acknowledge use of the ICP-MS facility within the UC
Center for Environmental Implications of Nanotechnology in CNSI
at UCLA.
[18]
[19]
G. La Sorella, G. Strukul, A. Scarso, Green Chem. 2015, 17, 644–683.
The only limitation observed thus far concerns aryl nitro-containing
pyridines. That is, for those that contain a bromide, competing coupling
at the nitro group (i.e., via loss of the NO2 moiety) appears to take place,
leading to a mixture of products.
[20]
[21]
[22]
J. P. Stambuli, in Catalysis Series (Ed.: T. Colacot), Royal Society Of
Chemistry, Cambridge, 2014, pp. 254–275.
T. Mandal, S. Jana, J. Dash, Euro. J. Org. Chem. 2017, 2017, 4972–
4983.
Keywords: aqueous micellar catalysis • axitinib • designer
a) L. L. Hegedus, R. W. McCabe, in Studies in Surface Science and
Catalysis (Eds.: B. Delmon, G.F. Froment), Elsevier, 1980, pp. 471–
505; b) G. Wilkinson, R. D. Gillard, J. A. McCleverty, Comprehensive
surfactant • Migita cross-coupling • nickel-catalysis
Coordination Chemistry. the Synthesis, Reactions, Properties
Applications of Coordination Compounds Vol. 6, Pergamon, Oxford,
1987.
&
[1]
[2]
T. Scattolin, E. Senol, G. Yin, Q. Guo, F. Schoenebeck, Angew. Chem.
Int. Ed. 2018, 57, 12425–12429; Angew. Chem. 2018, 130, 12605–
12609.
[23]
[24]
a) L.-H. Zhai, L.-H. Guo, Y.-H. Luo, Y. Ling, B.-W. Sun, Org. Process
Res. Dev. 2015, 19, 849–857; b) B. P. Chekal, S. M. Guinness, B. M.
Lillie, R. W. McLaughlin, C. W. Palmer, R. J. Post, J. E. Sieser, R. A.
Singer, G. W. Sluggett, R. Vaidyanathan, G. J. Withbroe, Org. Process
Res. Dev. 2014, 18, 266–274.
a) J. Oudar, Catalysis Reviews 1980, 22, 171–195; b) A. Kolpin, G.
Jones, S. Jones, W. Zheng, J. Cookson, A. P. E. York, P. J. Collier, S.
C. E. Tsang, ACS Catal. 2017, 7, 592–605.
[3]
[4]
I. P. Beletskaya, V. P. Ananikov, Chem. Rev. 2011, 111, 1596–1636.
a) T. Norris, K. Leeman, Org. Process Res. Dev. 2008, 12, 869–876; b)
J. M. Ganley, C. S. Yeung, J. Org. Chem. 2017, 82, 13557–13562; c)
S. S. M. Bandaru, S. Bhilare, J. Cardozo, N. Chrysochos, C. Schulzke,
Y. S. Sanghvi, K. C. Gunturu, A. R. Kapdi, J. Org. Chem. 2019, 84,
8921–8940; d) J. F. Hartwig, Acc. Chem. Res. 2008, 41, 1534–1544; f)
M. A. Fernández-Rodríguez, Q. Shen, J. F. Hartwig, J. Am. Chem. Soc.
2006, 128, 2180–2181; g) J. Xu, R. Y. Liu, C. S. Yeung, S. L. Buchwald,
ACS Catal. 2019, 9, 6461–6466; h) M. Cong, Y. Fan, J.-M. Raimundo,
Y. Xia, Y. Liu, G. Quéléver, F. Qu, L. Peng, Chem. Eur. J. 2013, 19,
17267–17272; i) “A. C. Jones, W. I. Nicholson, H. R. Smallman, D. L.
Z. Boros, L. Nagy-Győr, K. Kátai-Fadgyas, I. Kőhegyi, I. Ling, T. Nagy,
Z. Iványi, M. Oláh, G. Ruzsics, O. Temesi, B. Volk, J. Flow. Chem. 2019,
9, 101–113.
[25]
[26]
L. Finck, J. Brals, B. Pavuluri, F. Gallou, S. Handa, J. Org. Chem. 2018,
83, 7366–7372.
a) S. Sadhukhan, B. Baire, Advanced Synthesis & Catalysis 2018, 360,
298–304; b) K. K. Rajbongshi, D. Hazarika, P. Phukan, Tetrahedron
2016, 72, 4151–4158.
[27]
[28]
B. H. Lipshutz, N. A. Isley, J. C. Fennewald, E. D. Slack, Angew. Chem.
Int. Ed. 2013, 52, 10592–10958; Angew. Chem. 2013, 125, 11156–
11162.
Browne,
Org.
Lett.”,
can
be
found
under
https://doi.org/10.1021/acs.orglett.0c02418, 2020.
[5]
[6]
[7]
[8]
[9]
M. S. Oderinde, M. Frenette, D. W. Robbins, B. Aquila, J. W. Johannes,
J. Am. Chem. Soc. 2016, 138, 1760–1763.
R. A. Sheldon, Green Chem. 2017, 19, 18–43.
S. D. Timpa, C. J. Pell, O. V. Ozerov, J. Am. Chem. Soc. 2014, 136,
14772–14779.
B. A. Vara, X. Li, S. Berritt, C. R. Walters, E. J. Petersson, G. A.
Molander, Chem. Sci. 2018, 9, 336–344.
T. Migita, T. Shimizu, Y. Asami, J. Shiobara, Y. Kato, M. Kosugi, Bull.
Chem. Soc. Jpn. 1980, 53, 1385–1389.
a) B. Liu, C.-H. Lim, G. M. Miyake, J. Am. Chem. Soc. 2017, 139,
13616–13619; b) C. Uyeda, Y. Tan, G. C. Fu, J. C. Peters, J. Am. Chem.
Soc. 2013, 135, 9548–9552.
[10]
a) C. Uyeda, Y. Tan, G. C. Fu, J. C. Peters, J. Am. Chem. Soc. 2013,
135, 9548–9552; b) R. S. Schwab, D. Singh, E. E. Alberto, P. Piquini,
O. E. D. Rodrigues, A. L. Braga, Catal. Sci. Technol. 2011, 1, 569; c) R.
Xu, J.-P. Wan, H. Mao, Y. Pan, J. Am. Chem. Soc. 2010, 132, 15531–
15533; d) H.-J. Xu, Y.-F. Liang, X.-F. Zhou, Y.-S. Feng, Org. Biomol.
Chem. 2012, 10, 2562; e) F. Y. Kwong, S. L. Buchwald, Org. Lett. 2002,
4, 3517–3520; f) A. Rostami, A. Rostami, A. Ghaderi, J. Org. Chem.
2015, 80, 8694–8704; g) S. S. Bahekar, A. P. Sarkate, V. M. Wadhai,
P. S. Wakte, D. B. Shinde, Cat. Comm. 2013, 41, 123–125.
[11]
[12]
a) Y.-C. Wong, T. T. Jayanth, C.-H. Cheng, Org. Lett. 2006, 8, 5613–
5616; b) F. M. Moghaddam, R. Pourkaveh, Cat. Commun. 2017, 94,
33–37.
a) D. Liu, H.-X. Ma, P. Fang, T.-S. Mei, Angew. Chem. Int. Ed. 2019,
58, 5033–5037; Angew. Chem. 2019, 131, 5087–5091; b) Y. Wang, L.
Deng, X. Wang, Z. Wu, Y. Wang, Y. Pan, ACS Catal. 2019, 9, 1630–
1634;
[13]
a) R. Sikari, S. Sinha, S. Das, A. Saha, G. Chakraborty, R. Mondal, N.
D. Paul, J. Org. Chem. 2019, 84, 4072–4085; b) A. R. Martin, D. J.
Nelson, S. Meiries, A. M. Z. Slawin, S. P. Nolan, Eur. J. Org. Chem.
2014, 2014, 3127–3131; c) S.-C. Lee, H.-H. Liao, A. Chatupheeraphat,
M. Rueping, Chem. Eur. J. 2018, 24, 3608–3612; d) K. D. Jones, D. J.
Power, D. Bierer, K. M. Gericke, S. G. Stewart, Org. Lett. 2018, 20,
208–211.
[14]
[15]
“N,N-dimethylformamide - Substance Information - ECHA,” A. C. Jones,
W. I. Nicholson, H. R. Smallman, D. L. Browne, Org.
Lett. 2020, 22, 7433–7438.
S. Kim, M. J. Goldfogel, M. M. Gilbert, D. J. Weix, J. Am. Chem. Soc.
2020, 142, 9902–9907.
6
This article is protected by copyright. All rights reserved.

10.1002/anie.202013017
Angewandte Chemie International Edition
COMMUNICATION
Entry for the Table of Contents
X
• low catalyst loading (down to 0.7 mol %)
• inexpensive and bench stable catalyst
• active pharmaceutical intermediates
• recyclable aqueous reaction medium
• scalable
R
SH
1.05 equiv
(R = aryl, alkyl)
+
Het
(X = Br, I)
• low E Factor
Ni
Zn
S
R
Het
>25 examples
up to 99% yield
TPGS-750-M/H2O
Aryl sulfides made easy… in water under mild conditions, using a ligated base metal catalyst, and very little of it! New, Ni-catalyzed
C-S bond formation leading to functionalized products can now be obtained in good yields, and with residual levels of metal
contamination below FDA guidelines without additional processing.
7
This article is protected by copyright. All rights reserved.