Photoredox Nickel-Catalyzed C-S Cross-Coupling: Mechanism, Kinetics, and Generalization

Article abstract of DOI:10.1021/jacs.0c11937

Photoredox-mediated nickel-catalyzed cross-couplings have evolved as a new effective strategy to forge carbon-heteroatom bonds that are difficult to access with traditional methods. Experimental mechanistic studies are challenging because these reactions involve multiple highly reactive intermediates and perplexing reaction pathways, engendering competing, but unverified, proposals for substrate conversions. Here, we report a comprehensive mechanistic study of photoredox nickel-catalyzed C-S cross-coupling based on time-resolved transient absorption spectroscopy, Stern-Volmer quenching, and quantum yield measurements. We have (i) discovered a self-sustained productive Ni(I/III) cycle leading to a quantum yield φ > 1; (ii) found that pyridinium iodide, formed in situ, serves as the dominant quencher for the excited state photocatalyst and a critical redox mediator to facilitate the formation of the active Ni(I) catalyst; and (iii) observed critical intermediates and determined the rate constants associated with their reactivity. Not only do the findings reveal a complete reaction cycle for C-S cross-coupling, but the mechanistic insights have also allowed for the reaction efficiency to be optimized and the substrate scope to be expanded from aryl iodides to include aryl bromides, thus broadening the applicability of photoredox C-S cross-coupling chemistry.

Full text of DOI:10.1021/jacs.0c11937

pubs.acs.org/JACS
Article
Photoredox Nickel-Catalyzed C−S Cross-Coupling: Mechanism,
Kinetics, and Generalization
Yangzhong Qin,‡ Rui Sun,‡ Nikolas P. Gianoulis, and Daniel G. Nocera*
Cite This: J. Am. Chem. Soc. 2021, 143, 2005−2015
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Photoredox-mediated nickel-catalyzed cross-couplings have evolved as a new effective strategy to forge carbon−
heteroatom bonds that are difficult to access with traditional methods. Experimental mechanistic studies are challenging because
these reactions involve multiple highly reactive intermediates and perplexing reaction pathways, engendering competing, but
unverified, proposals for substrate conversions. Here, we report a comprehensive mechanistic study of photoredox nickel-catalyzed
C−S cross-coupling based on time-resolved transient absorption spectroscopy, Stern−Volmer quenching, and quantum yield
measurements. We have (i) discovered a self-sustained productive Ni(I/III) cycle leading to a quantum yield Φ > 1; (ii) found that
pyridinium iodide, formed in situ, serves as the dominant quencher for the excited state photocatalyst and a critical redox mediator to
facilitate the formation of the active Ni(I) catalyst; and (iii) observed critical intermediates and determined the rate constants
associated with their reactivity. Not only do the findings reveal a complete reaction cycle for C−S cross-coupling, but the
mechanistic insights have also allowed for the reaction efficiency to be optimized and the substrate scope to be expanded from aryl
iodides to include aryl bromides, thus broadening the applicability of photoredox C−S cross-coupling chemistry.
Thioethers are frequently encountered in natural bio-
products and pharmaceuticals. The potential bioactivity of
thioethers in the treatment of diseases such as cancer and
HIV31−35 has motivated interest in developing effective
methodologies for their synthesis.25,34−43 Recently, visible-
light-driven photoredox nickel catalysis has provided a new
strategy to forge C−S bonds between (hetero)aryl iodides and
thiols under mild conditions,25 thus representing an important
advance over traditional methods requiring high temperatures
or strong bases.34−39 Unlike other photoredox nickel-catalyzed
cross-couplings between aryl bromides and alcohols, amines, or
carboxylic acids,20−22 where the nucleophile engages the nickel
INTRODUCTION
■
Photoredox generation of transition metal intermediates has
emerged as a powerful strategy for promoting transformations
that are otherwise difficult to access thermally.1−8 A tandem
approach of employing a photoredox catalyst to drive
transition metal catalysis has been fruitful,9−15 especially for
effecting nickel-catalyzed carbon−heteroatom bond formation
under mild conditions using simple and inexpensive
ligands.16−25 However, despite the prolific reports of new
nickel photoredox methods, there have been relatively few
experimental mechanistic studies due to reaction complex-
ity.26−30 The variety of potential reaction pathways in
photoredox cross-coupling systems together with the accessi-
bility of multiple oxidation states obfuscates the precise redox
chemistry between the photocatalyst and the cross-coupling
catalyst and, consequently, the nickel redox levels responsible
for supporting catalysis. Lacking such a mechanistic under-
standing can impede reaction development and optimization,
which often rely on time-consuming and labor-intensive trial-
and-error reaction screening.
Received: November 14, 2020
Published: January 19, 2021
https://dx.doi.org/10.1021/jacs.0c11937
J. Am. Chem. Soc. 2021, 143, 2005−2015
© 2021 American Chemical Society
2005

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
catalyst through a redox-neutral metalation−deprotonation
sequence, the coupling of aryl iodides to thiols has been
proposed to proceed through a “radical” pathway via the
photogeneration of thiyl radicals (Scheme 1A),25 thus
substrate scope of this methodology, which has been restricted
to only aryl iodides, to include aryl bromides, allowing for
access to a class of electrophiles with higher availability and
lower cost.
EXPERIMENTAL SECTION
Scheme 1. (A) Proposed Mechanisms for Photoredox-
Mediated Nickel-Catalyzed Aryl Thiolation and (B) Self-
Sustained Mechanism Driven by a Redox Mediator as
Deduced from This Work
■
General Considerations. All samples were prepared in a
nitrogen-filled glovebox with commercial reagents and anhydrous
acetonitrile stored over activated 3 Å molecular sieves. Ir(III) (=
[Ir(dF-CF3-ppy)2(dtbbpy)][PF6] and dtbbpy = 4,4′-di-tert-butyl-2,2′-
dipyridyl) was purchased from Strem Chemicals and used as received.
Ni(II) (= (dtbbpy)NiCl2) was prepared in situ from (dme)NiCl2 and
dtbbpy, both of which were purchased from Sigma-Aldrich and used
as received. Thiols were obtained from Oakwood Chemical. 1-
Methylpyridinium iodide was prepared from a metathesis reaction
between 1-methylpyridinium chloride and sodium iodide in
acetonitrile. Typically, reaction solutions were stored in a 20 mL
glass vial and sealed with electric tape. For convenience of
presentation, Table 1 lists the compositions of key solutions used in
Table 1. Composition of Key Solutions Used in Kinetics
Studies
solution
composition
a
S1
S2
S3
S4
Ni(II) (10 mM) + py (200 mM) + thiol 2 (150 mM)
Ir(III) (150 μM) + pyHI (25 mM)
S2 + py (200 mM) + thiol 2 (150 mM)
circumventing the formation of free thiolate equivalents that
are known to coordinatively inhibit catalysis.16,35 Conversely,
computational studies have suggested an alternative “oxidation
state modulation” mechanism (Scheme 1A) involving oxidative
quenching of the excited iridium photocatalyst and the
formation of a nickel thiolate from the deprotonation of
coordinated thiol.44 Both mechanisms invoke closed photo-
cycles (quantum yield Φ < 1) requiring the involvement of
Ni(I), Ni(II), and Ni(III) intermediates for every turn-
over.25,44 This is in contradistinction to recent investigations
of photoredox-mediated nickel-catalyzed cross-coupling of aryl
bromides with alcohols,29,45,46 amines,30,45,46 and carboxylic
acids,45 wherein Φ > 1 owing to a self-sustained Ni(I/III) dark
cycle. Whether a self-sustained Ni(I/III) cycle is also
responsible for aryl thiolation warrants investigation because
a thermally sustained cycle will involve unique reaction
intermediates that engender optimization strategies distinct
from both of the previously proposed mechanisms (Scheme
1A).25,44
We now report a comprehensive mechanistic study on the
photoredox-mediated nickel-catalyzed C−S cross-coupling
(Scheme 1B). Through a combination of time-resolved
photophysical and photochemical techniques, we find that
the photoredox mechanism is characterized by 12 rate
constants associated with a thermally sustained Ni(I/III)
cycle with Φ > 1. Nanosecond transient absorption spectros-
copy permits the observation of the in situ formation and
subsequent reactivity of a Ni(I) intermediate, along with a side
reaction leading to thiyl radical formation. Furthermore, we
identify the hitherto underappreciated, but nonetheless critical,
roles of pyridinium iodide (pyHI), which is produced as a
byproduct of the cross-coupling. We show that pyridinium
iodide is essential in facilitating the photoredox transformation
by (i) quenching the excited photocatalyst; (ii) preventing the
formation of nickel thiolate complexes that competitively
absorb light; and (iii) acting as a redox mediator to efficiently
generate Ni(I) from off-cycle Ni(II) species, thereby sustaining
a productive Ni(I/III) catalytic cycle. By leveraging these
mechanistic insights, we present a strategy to expand the
a
S3 + Ni(II) (10 mM)
a
Thiol 2 = 4-methoxybenzyl mercaptan.
this study. The photocatalytic reactions were carried out on solutions
as reported previously25 except that, rather than purging the head
space with nitrogen, all samples described herein were prepared in a
glovebox. Additionally, in place of a 34 W blue LED excitation source,
solutions were illuminated with a Kessil A160WE Tuna Blue light
source at a short distance (Figure S1A) and constantly agitated with a
magnetic stirrer and cooled with a fan. For reactions at 55 °C, a hot
plate equipped with a thermocouple was used. UV−vis spectra were
measured with a Cary 5000 spectrometer (Agilent) and blank-
corrected against the solvent. NMR spectra were recorded on an
Agilent DD2 spectrometer (600 MHz) or a Varian/Inova
spectrometer (500 MHz). The product yields were obtained based
on the 1H NMR spectra referenced to prequantified 1,3-benzodioxole
as the internal standard.
Stern−Volmer Quenching Studies. Steady-state emission
spectra were obtained using a fluorimeter (Photon Technology
International, Model QM4). Steady-state Stern−Volmer quenching
studies were carried out by measuring the steady-state emission
intensity (I) at 500 nm and exciting the photocatalyst Ir(III) at 370
nm. The dynamic Stern−Volmer quenching studies were carried out
by exciting solutions at 430 nm and measuring the lifetime (τ) of the
photocatalyst excited state, *Ir(III), at 500 nm using the laser setup
described below. The quenching ratio (I0/I or τ0/τ) and the Stern−
Volmer constant (Ksv) were determined by the relation
I0/I or τ0/τ = 1 + Ksv[quencher]
(1)
where I0 and τ0 are the emission intensity and lifetime in the absence
of quencher, respectively. The quenching rate (kq) is given by kq =
Ksv/τ0. Depending on the experimental conditions and the presence of
adventitious oxygen in the sample, the measured τ0 varies marginally
(see Figure S4).
Reaction Progress and Quantum Yield Measurements.
Reaction solutions were prepared with acetonitrile-d3 and stored in
J. Young NMR tubes (1 mL each) in a N2-filled glovebox. For samples
exposed to air, the screw cap of the NMR tube was opened for ∼5 s
and then closed, and the solution was mixed by turning the NMR
tube over repeatedly. This procedure was repeated two more times.
All NMR tubes were placed in a 3-D printed NMR tube holder which
2006
https://dx.doi.org/10.1021/jacs.0c11937
J. Am. Chem. Soc. 2021, 143, 2005−2015

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
was located at the center of the cylindrically arranged 24 W blue LED
strip lights (Figure S1B) to ensure that each sample receives the same
amount of illumination. A fan was mounted at the top of the
apparatus to cool the samples. The product yields were obtained at
Table 2. Photoredox-Mediated Nickel-Catalyzed Aryl
Thiolation and Stern−Volmer Quenching Studies
a
1
different times based on the H NMR spectra. For quantum yield
measurements, the output from a 150 W Xe arc lamp (Newport
67005 arc lamp housing and 69907 universal arc lamp power supply)
was passed through a 435 nm band-pass filter (FWHM = 10 nm), and
a lens (f = 40 mm) was used to focus the light onto the reaction
solution contained within a 1 cm cuvette. The power reaching the
sample was attenuated by neutral density filters and measured with an
Ophir ORION/PD power meter equipped with a PD-300-ROHS
head sensor. The photon flux was calibrated against a potassium
ferrioxalate standard based on a published procedure.47 Each
quantum yield measurement was performed in triplicate.
Time-Resolved Emission and Transient Absorption Spec-
troscopy. A Quanta-Ray Nd:YAG laser (SpectraPhysics) produced
laser pulses at 355 nm at a repetition rate of 10 Hz and a time width
of 8 ns. The 355 nm laser pulses were either directly used or passed to
a MOPO (SpectraPhysics) to produce laser pulses at 430 nm for
sample excitation. A 75 W xenon-arc lamp (PTI, Model A1010) was
used to generate white probe light for transient absorption (TA)
measurements. Both excitation and probe beams were focused and
overlapped onto the sample, which was typically contained in a 1 cm
cuvette. The emission of the photocatalyst and the transmitted probe
light were directed to a Triax 320 spectrometer, and the signal was
detected with a photomultiplier tube (PMT) (Hamamatsu) coupled
to a 1 GHz oscilloscope (LeCroy, Model 9384CM) or a CCD camera
(Andor Technology). Further details of the laser spectroscopic setup
can be found elsewhere.48 The fitting models for the TA kinetic traces
are provided in the Supporting Information.
K
SV/M−1
K
SV/M−1
kq/M−1 s−1
reagent
(steady-state)
(dynamic)
(dynamic)
b
pyridine (py)
Ni(II)
1a
−
−
200(8)
1.3(1)
2.4(2)
7.6(5)
−
c
N.A.
7.8(3) × 107
5.1(4) × 105
1.1(1) × 106
3.4(2) × 106
1.5(1)
1.9(1)
7.1(4)
2
2 with py
pyHI
d
53383(3370)
54480(2243)
2.1(1) × 1010
a
All reagent concentrations are referenced to a 0.1 M concentration of
1a. No quenching observed. Not measured. kq,ET in kinetic
b
c
d
modeling.
decay at 500 nm (λexc = 430 nm). For a fresh reaction solution,
the measured lifetime was τ = 450 ns (Figure 1A). However,
for a solution that was illuminated for 20 min under blue light,
the lifetime fell below our instrumental time resolution (8 ns),
indicating the in situ generation and accumulation of a
UV−Vis Study with (dtbbpy)NiCl2 in the Presence of
Thiolate. A 50 mL glass bottle was charged with (dme)NiCl2
(43.9 mg, 0.200 mmol) and dtbbpy (80.5 mg, 0.300 mmol). MeCN
(20 mL) was then added, and the reaction mixture was stirred for 30
min at room temperature to form Ni(II). A 5 mL aliquot was drawn
and added to a 20 mL scintillation vial charged with potassium (4-
methoxyphenyl)methanethiolate (4.8 mg, 0.025 mmol), which was
prepared from the corresponding thiol and potassium tert-butoxide,
and the reaction mixture was stirred at room temperature for 2 h. The
mixture was then filtered using a 0.2 μm PTFE syringe filter and
diluted 10 times with MeCN. A UV−vis spectrum was recorded on
the resulting solution.
RESULTS
■
Quenching Studies. We undertook mechanistic inves-
tigation of the C−S cross-coupling by examining which
reagents listed in Table 2 react with *Ir(III) per steady-state
and dynamic Stern−Volmer quenching. Pyridine does not
quench *Ir(III) (Figure S2). In the case of Ni(II), the steady-
state Ksv cannot be accurately measured due to its strong and
broad absorption, which overlaps with that of Ir(III) and thus
requires inner-filter corrections (Figure S3 and Section B.1 in
the Supporting Information). Steady-state Stern−Volmer
quenching plots for reagents in Table 2 are shown in Figure
S4. Ni(II) shows dynamic quenching with a rate constant of kq
= 7.8(3) × 107 M−1 s−1, which is ∼2 orders of magnitude
greater than that of any other individual component in the
original reaction solution, including 4-iodotoluene (1a) and 4-
methoxybenzyl mercaptan (2) (Table 2). Thiol compound 2
in the presence of 200 mM pyridine exhibits a quenching rate
∼3 times higher than that of 2 alone. However, when the
concentrations of the reactants are considered, the overall
contributions to the quenching of *Ir(III) by the reactants
(Ni(II), 1a, and 2) in the starting solution are comparable.
The lifetime of *Ir(III) in the reaction solution can be
explicitly determined by monitoring the compound’s emission
Figure 1. (A) Time-resolved emission decay of the excited
photocatalyst *Ir(III) in a fresh and illuminated sample, the latter
of which was subjected to 20 min blue light irradiation under standard
reaction conditions (Table 2, top panel). The scatter plots and solid
lines show the raw data and single-exponential fittings, respectively.
The fitted lifetime τ is significantly lower for an illuminated sample
when compared to a fresh one. This change in *Ir(III) lifetime is
accompanied by a change in the color of the reaction solution, as
shown in the inset. (B) UV−vis absorption spectra of solution S1
(Table 1), S1 with an additional 100 mM pyHI at 0 and 18 h in the
dark, and a solution containing Ni(II) with RS− (potassium (4-
methoxyphenyl)methanethiolate), the spectrum for which is
normalized with that of S1 at 500 nm.
2007
https://dx.doi.org/10.1021/jacs.0c11937
J. Am. Chem. Soc. 2021, 143, 2005−2015

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
byproduct that was an efficient quencher. Control experiments
based on the 19F NMR spectra (Figure S13) for solutions
before and after illumination showed no appreciable change,
suggesting that Ir(III) is stable during the reaction and the
lifetime change is not due to nonradiative pathways associated
with photocatalyst degradation. In particular, the presence of
an iodide substrate, thiol as a proton source (from its acid−
base chemistry and PCET involvement in photoredox
reactions49−51), and pyridine suggests the possibility of
generating pyridinium iodide (pyHI). As shown in Table 2,
pyHI is an exceptional quencher of *Ir(III); the quenching
rate constant of kq,ET = 2.1(1) × 1010 M−1 s−1 is over 2 orders
of magnitude higher than that of Ni(II). A similar quenching
rate was obtained for tetrabutylammonium iodide (TBAI)
(Figure S4D), suggesting that this unusually high quenching
rate constant is due to the photooxidation of iodide (I−) by
*Ir(III).
Effect of Thiolate and pyHI on (dtbbpy)NiCl2. The
initial reaction solution was orange (Figure 1A, inset), which
we later discovered was similar to the color of solution S1 (S1
= 10 mM Ni(II), 200 mM pyridine, and 150 mM thiol 2 in
MeCN). To identify the compound responsible for the orange
color, we prepared a solution of Ni(II) in the presence of
potassium (4-methoxyphenyl)methanethiolate. The UV−vis
absorption spectrum of this solution (Figure 1B, blue trace) is
nearly identical to that of solution S1 (Figure 1B, red trace),
suggesting that the orange color is due to a nickel thiolate
compound. When the complete reaction solution (Table 2,
top) was photolyzed, the initial orange color gradually
disappeared, and the solution turned yellow (Figure 1A,
inset), diagnostic of Ir(III) solutions. This result suggested
protonolysis of the nickel thiolate complex present in the initial
solution. We hypothesized that the Brønsted acid pyHI is
responsible for this protonolysis reaction, and this acid
accumulated as the reaction progressed. Indeed, addition of
100 mM pyHI to solution S1 prompted a color change (Figure
1B, red to red-orange trace) that continued to lighten in color
over 18 h of storage in the dark (Figure 1B, orange trace).
Additionally, if pyHI was added to the initial reactant solution
before thiol, the orange color, indicative of the nickel thiolate,
was not observed. These results suggest that pyH+ (pKa = 3.4
in DMSO)52 can protonate any thiolate equivalents formed in
situ, thereby avoiding the formation of the nickel thiolate
complex, which can cause a strong inner filter effect.
Reaction Progress and Quantum Yield Measure-
ments. The effect of pyHI on the C−S cross-coupling
between 1a and 2 was examined under different conditions by
monitoring the product yield with time (Figure 2A, details
provided in the Experimental Section). Under reaction
conditions similar to those of the published method,25 where
O2 was present, the reaction showed an induction period in the
first 20 min (Figure 2A, red trace) that was less prominent for
an identical sample without air exposure (Figure 2A, black
trace). However, the induction period diminished after adding
50 mM pyHI (Figure 2A, blue and teal traces). The sample
without air exposure exhibited the highest initial photoreaction
rates (Figure 2A, blue trace). Although the product yield
(Figure 2A) is within error and not distinguished at the early
stages of reaction (≤6 min), a clear overall trend is established
at later times of reaction (at times >6 min) that lies well
outside the standard deviation of 5−15% as reflected in the
quantum yield measurements (Figure 2B).
Figure 2. (A) Aryl thiolation progress for a standard reaction solution
with the addition of 50 mM pyHI, exposure to air, or both. (B) Power
dependence of the quantum yield for aryl thiolation between 4-
methoxybenzyl mercaptan and 4-iodotoluene (1a) or 4′-iodoaceto-
phenone with and without additional 50 mM pyHI (λexc = 435 nm).
Lines are included to highlight the trend.
To further quantify the reaction efficiency and gain
mechanistic insight, we measured the quantum yield for the
C−S cross-coupling with and without pyHI (Figure 2B, black
and gray traces). The addition of pyHI (50 mM) results in an
increase in the quantum yield by 50% (from 0.12(2) to
0.18(1)) with an incident power of 7.4 mW. The quantum
yield also increases with decreasing incident power of the
irradiation source; changing the incident power from 7.4 mW
to 90 μW results in a 375% increase in quantum yield (from
0.12(2) to 0.45(3) in the absence of pyHI), which is
accentuated in the presence of pyHI (increase of 594%, from
0.18(1) to 1.07(11)). A quantum yield in excess of 1 (Φ > 1)
was further corroborated by using 4′-iodoactophenone as a
substrate (Figure 2B, red trace). With this more electron-
deficient aryl halide, the quantum yield increases from 0.28(2)
at 7.4 mW to 2.04(6) at 90 μW. Significantly, these values of Φ
> 1 establish the existence of a thermally self-sustained catalytic
cycle29,45,46 for product formation (vide infra).
Transient Intermediates and Reaction Kinetics. Nano-
second TA spectroscopy was employed to interrogate the
photoredox process as well as identify the reaction
intermediates and define the kinetics of a complete photoredox
cycle. Upon addition of 25 mM pyHI or TBAI to a solution of
150 μM Ir(III), TA features at ∼400, 525, and 720 nm were
observed 30 ns after photoexcitation (Figure 3A). The TA
spectrum consist of two components (Figure 3A, inset): the
reduced photocatalyst, Ir(II), and I2•−. The spectrum of Ir(II)
(maxima at 400 and 525 nm) has been previously determined
by spectroelectrochemistry.29 Subtraction of the Ir(II)
2008
https://dx.doi.org/10.1021/jacs.0c11937
J. Am. Chem. Soc. 2021, 143, 2005−2015

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
clean back electron transfer reaction between I2•− and Ir(II)
to form 2I− and Ir(III). However, with pyHI, the TA feature at
525 nm (Ir(II)) decays much faster than that at 720 nm
(I2•−), suggesting that the disappearance of Ir(II) is
predominantly due to the reaction between Ir(II) and pyH+
to form Ir(III) and pyH•. Similarly, in the presence of 1-
methylpyridinium iodide, a faster decay of Ir(II) was also
observed (Figure S5), ruling out the possibility that the
reaction between Ir(II) and pyH+ required a proton. By
monitoring the decay of Ir(II) at 525 nm with TBAI as the
quencher, we extracted a rate constant for the back electron
transfer reaction between I2•− and Ir(II) (Figure 3B, black
curve and Section B.3.1 in the Supporting Information) of
kBET1 = 9.4(2) × 109 M−1 s−1, consistent with an appreciable
driving force (E1/2(Ir(III/II)) = −1.74 V29 and E1/2(I2•−/ I−)
< 0.30 V54 vs Fc+/Fc). For the case of pyHI (Figure 3B, red
curve), and accounting for kBET1 (Section B.3.2 in the
Supporting Information), an electron transfer rate constant
of kET1 = 1.14(3) × 108 M−1 s−1 is extracted for the reaction
between pyH+ and Ir(II), which is also thermodynamically
favored (E° = −1.7 V vs Fc+/Fc for the pyH+/pyH•
couple,55,56 where 0.4 V was used to convert reference
potential from SCE to Fc+/Fc). Note, despite kBET1 being
nearly 2 orders of magnitude larger than kET1, we observed
faster reaction of Ir(II) with pyH+ than with I2•− due to the
higher concentration of pyH+ (25 mM pyH+ versus less than
10 μM I2•−). Additionally, I2•−, monitored at 700 nm (Figure
3B, blue curve), decays more slowly for the solution containing
pyHI versus the one containing TBAI due to the fast
disappearance of Ir(II) in the former, which attenuates the
back reaction between I2•− and Ir(II). Since I2•− is known to
disproportionate, we attribute the slower decay at 700 nm to
the disproportionation of I2•− to form I3− and I−, for which we
extracted a rate constant of 2kdisp = 2.9(1) × 109 M−1 s−1
(Section B.3.2 in the Supporting Information), consistent with
reported rate constants.53,57
With the reaction and kinetics for solution S2 as a reference,
the kinetics of S2 solutions containing more components of
the C−S cross-coupling reaction were examined. The addition
of 150 mM thiol 2 to solution S2 resulted in little change to
the TA spectra and kinetics (Figures S6 and S7). However, the
addition of both 200 mM pyridine and 150 mM 2 to solution
S2, resulting in solution S3 (= S2 + 150 mM 2 + 200 mM py),
accelerates the decay of I2•− (Figure 4). We posit that the
faster decay may result from proton-coupled electron transfer
(PCET) between compound 2 and I2•− in the presence of
pyridine as a base to form thiyl radical 2•, 2I−, and pyH+,
which may occur in either a concerted or a stepwise
fashion.58,59 To assess this contention, we chose to replace 2
in S3 with thiophenol (PhSH) because the PhS• radical
exhibits absorption in the visible region, allowing us to observe
the radical product if PCET does indeed occur. The inset of
Figure 4 shows the difference of TA spectra measured at 6.3 μs
between solutions of S2 + 150 mM PhSH in the absence and
presence of 200 mM pyridine (Figures S7A and S7B,
respectively); subtracting the spectrum without pyridine from
that with pyridine furnishes a spectral profile (Figure 4 inset,
gray trace and Figure S7C) with a band maximum at 450 nm,
which is similar to the absorption profile of the PhS• radical
(Figure 4 inset, red trace) as obtained independently by
photolyzing a solution of 6 mM diphenyl disulfide (PhSSPh)
with laser pulses at 355 nm. The PCET rate constants for thiol
oxidation were derived to be kPCET = 2.6 (1) × 105 M−1 s−1 for
Figure 3. (A) TA spectra of a solution containing 150 μM Ir(III) and
25 mM TBAI or pyHI (λexc = 430 nm). The inset shows that the
corresponding TA spectrum at 30 ns can be deconvolved into
contributions from Ir(II) (red) and I2•− (blue). The delay times listed
in the bottom panel also apply to the top panel. (B) The TA kinetic
traces probed at 525 and 700 nm predominantly show the decay of
Ir(II) and I2•−, respectively. The scatter plots and the solid lines show
the raw data and model fittings, respectively (see the Supporting
Information for details).
spectrum from the TA spectrum recorded at 30 ns yields the
blue trace (Section B.2 in the Supporting Information), which
matches that of I2•− (maxima at 395 and 720 nm).53 These
results confirm that I− quenches *Ir(III) by electron transfer
to form Ir(II) and I•, the latter of which reacts facilely with
I−.53 Based on the Stern−Volmer quenching study (Table 2),
the electron transfer rate between *Ir(III) and I− is kq,ET
2.1(1) × 1010 M−1 s−1.
=
In addition to the role of iodide, we also investigated the role
of pyH+. The TA spectrum for solution S2 (S2 = 150 μM
Ir(III) + 25 mM pyHI) shows clear distinctions in time
evolution for Ir(II) as compared to when pyHI was replaced
with TBAI (Figure 3A). For the solution with TBAI, the TA
features at 525 nm (predominantly Ir(II)) and 720 nm
(predominantly I2•−) both decay at a similar rate, suggesting a
2009
https://dx.doi.org/10.1021/jacs.0c11937
J. Am. Chem. Soc. 2021, 143, 2005−2015

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 4. TA kinetic trace probed at 700 nm for solution S2 (150 μM
Ir(III) and 25 mM pyHI) with 150 mM thiophenol, 150 mM
thiophenol in the presence of 200 mM pyridine, and 150 mM 4-
methoxybenzyl mercaptan in the presence of 200 mM pyridine (λexc
=
430 nm). The faster decay for solutions containing pyridine is due to
PCET between I2•− and thiol with pyridine as a base. The scatter
plots and solid lines show the raw data and model fittings, respectively
(see the Supporting Information for details). The inset shows the
difference TA spectrum at 6.3 μs (gray) for solutions of S2 with 150
mM thiophenol in the absence and presence of 200 mM pyridine; this
difference spectrum matches the TA spectrum of thiophenoxyl radical
(red) obtained independently from directly exciting diphenyl disulfide
at λexc = 355 nm.
thiol 2 and kPCET = 3.6(1) × 105 M−1 s−1 for PhSH (Section
B.3.3 in the Supporting Information). The observation of the
thiyl radicals by TA is consistent with their presence in C−S
cross-coupling, as ascertained from radical trapping studies.25
We next interrogated the role of Ni in the photoredox cycle
by adding 10 mM Ni(II) to S3 (S4 = S3 + 10 mM Ni(II)).
The measured TA spectra for solution S4 are shown in Figure
5A. Subtracting the TA spectrum of solution S3 at 7.5 μs from
that of S4 yielded the difference TA spectrum (Figure 5B),
which clearly shows two absorption features with maxima at
425 and 600 nm. These spectral features have been observed
previously in photoredox aryl etherification29 and arise from a
Ni(I) intermediate. To further assess our assignment of the
absorption bands in Figure 5B to a Ni(I) intermediate, we
monitored the comproportionation between Ni(0) [Ni(0) =
(dtbbpy)Ni(cod)] and Ni(II) (Section A.2 in the Supporting
Information). Electron paramagnetic resonance (EPR) meas-
urement of the comproportionation product supported the
presence of a new Ni(I) species (Figure S8A). Furthermore,
the comproportionation solution showed an absorption
spectrum (Figure S8B) distinct from that of Ni(0) or Ni(II)
(Figure S8C), with an absorption band in the 600 nm region.
Exposure of the same solution to air led to the disappearance
of the 600 nm band to give the Ni(II) absorption spectrum.
To assess whether the Ni(I) species formed by the
comproportionation reaction accounts for the TA difference
spectrum shown in Figure 5B, the difference absorption
spectrum for comproportionated solutions before and after air
exposure (i.e., Ni(I) spectrum − Ni(II) spectrum) was
computed (Figure S8D) and found to be similar to the
difference TA spectrum shown in Figure 5B (red curve), thus
supporting the formation of Ni(I) under photoredox
conditions.
Figure 5. (A) TA spectra (λexc = 430 nm) of solution S4 (150 μM
Ir(III), 25 mM pyHI, 200 mM pyridine, 150 mM 4-methoxybenzyl
mercaptan, and 10 mM Ni(II)). (B) The difference TA spectrum at
7.5 μs for solution S3 and S4 (parent spectra shown in the inset),
revealing the presence of a putative Ni(I) intermediate. (C) TA
kinetic trace measured at 600 nm for solution S3 (black) and S4
(red). The additional rising feature on the red curve suggests the
formation of Ni(I) with a time constant of 1.56 μs. (D) TA kinetic
trace measured at 600 nm for solution S4 with 0.1 and 0.5 M 4-
iodotoluene, 1a. The faster decay for solutions with higher
concentrations of aryl iodide implies the oxidative addition of aryl
iodide to the Ni(I). The scatter plots and solid lines show the raw
data and model fittings, respectively (see the Supporting Information
for details). The inset shows the linear fit to extract the oxidative
addition rate constant kOA of 4-iodotoluene, 1a.
Accordingly, the kinetics of the Ni(I) intermediate, for
which the 600 nm band offered a direct signature, were
defined. Figure 5C shows the decay kinetics for the 600 nm
signal of solution S4. A slight decay in signal over 0.5 μs is
followed by a rise over 5 μs and then a subsequent slow decay
lasting several hundreds of microseconds (Figure 5C, red
trace). This evolution of the signal contrasts the immediate
drop in signal at 600 nm observed for solution S3 (Figure 5C,
black trace) where no nickel complex was present. An
2010
https://dx.doi.org/10.1021/jacs.0c11937
J. Am. Chem. Soc. 2021, 143, 2005−2015

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
additional control experiment on a similar solution without
thiol 2 but with the Ni(II) complex (S4 − 150 mM 2) also
showed a decay−rise−decay feature, suggesting that the
formation of Ni(I) species is not due to thiol 2 (Figure
S9D, black trace). Significantly, the initial fast decay and slow
rise indicate that the formation of Ni(I) occurs through the
action of an intermediate that is not Ir(II), as its formation is
slower than the disappearance of Ir(II). This result establishes
that Ni(I) is not generated from the direct reduction of Ni(II)
by Ir(II). As noted in Figure 3A, pyH• is present during the
photoredox transformation and thus may act as a reductant for
Ni(II). Consistent with this contention, the rate constant for
the reaction between pyH• and Ni(II) was determined to be
kET2 = 6.4(5) × 107 M−1 s−1 whether in the presence or
absence of thiol (Section B.3.4 in the Supporting Information).
From modeling (Section B.3.5 in the Supporting Information)
the slow decay over 5 to 200 μs, kBET2 = 8.5(4) × 109 M−1 s−1,
is ascribed to the back reaction between Ni(I) and I2•− to
form Ni(II) and 2I−.
fit furnishes a bimolecular OA rate constant of kOA = 2.5(2) ×
104 M−1 s−1. We emphasize that the obtained rate (kOA) may
offer an upper limit for the OA if other unknown reactions
exist between Ni(I) and ArI. Given the quantitative product
yield and high observed quantum yields (Figure 2), OA is
likely the major reaction here. The observed kOA is also
consistent with previously reported values,61 despite the
differences in the ligand environment. Due to the absence of
any salient signals in the TA spectrum, we were unable to
measure the rates of ligand exchange or reductive elimination
at Ni(III) following OA.
Reaction Optimization and Generalization. In line
with previous observations,25 we found that the substrate scope
for the C−S cross-coupling reaction was limited to aryl iodides
(Table S1) under the conditions shown in Table 2. Guided by
our mechanistic insights, which show the important, but
previously unidentified, roles of pyridinium and iodide in the
photoredox cycle, we found that the addition of pyHI and a
slightly elevated temperature of 55 °C enable the C−S cross-
coupling between 4-bromotoluene (1b) and thiol 2 (Table 3,
Given the reported reduction potential for Ir(II) (E1/2
=
−1.74 V vs Fc+/Fc),29 it should be able to directly reduce
Ni(II) in the absence of pyH+. To test this hypothesis, we
obtained the TA spectrum of a solution containing 150 μM
Ir(III), 25 mM TBAI, 200 mM py, and 10 mM Ni(II). We
monitored the absorption maximum of Ir(II) at 525 nm, which
was generated immediately following the initial quenching and
then decayed slowly due to the back reaction (Figures S9A and
S9B, red curves). Monitoring Ni(I) at 600 nm, a decay-and-
rise feature was not observed on a short time scale (<5 μs)
(Figure S9D, red trace). However, an additional TA feature
appeared on a longer time scale (>50 μs) after Ir(II) and I2•−
had decayed significantly (shown by the comparison of the red
curves in Figures S9C and S9D). This additional TA feature is
likely due to the reaction between Ir(II) and Ni(II) in the
presence of pyridine to form Ir(III) and Ni(I), the rate
constant of which was extracted to be kET3 = 2.5(5) × 106 M−1
s−1 (Figure S10 and Section B.3.5 in the Supporting
Information).
Table 3. Photoredox-Mediated Nickel-Catalyzed Aryl
Thiolation Using Aryl Bromide
a
The rate constant for the direct reduction of Ni(II) by Ir(II)
in the presence of pyridine is much smaller than that of the
pyH•-mediated process. However, in the absence of pyridine,
Ir(II) can effectively reduce Ni(II) directly to form Ni(I) with
a rate constant of kET4 = 3.0(4) × 107 M−1 s−1 (Figure S11 and
Section B.3.5 in the Supporting Information). These results
suggest that pyridine makes Ni(II) less amenable toward
reduction. Indeed, the absorption spectrum of Ni(II) showed
clear differences with and without 200 mM pyridine (Figure
S12A). We further carried out cyclic voltammetry (CV)
measurements on Ni(II) in the absence and presence of 200
mM pyridine. Despite the irreversibility of the reduction wave,
its onset shifted cathodically by ∼200 mV when pyridine was
present (Figure S12B), consistent with the smaller reduction
rate (kET3) observed for Ni(II) in the presence of pyridine.
With the formation pathways of Ni(I) determined, we
further probed its oxidative addition (OA) reactivity60 with 4-
iodotoluene (1a) by monitoring the TA kinetics of Ni(I) at
600 nm. The Ni(I) decay became faster when the
concentration of 1a was increased from 0 to 0.5 M (Figures
5C and 5D). Specifically, with 0.5 M 1a, the signal drops to
nearly zero after 100 μs. With fitting (Section B.3.4 in the
Supporting Information), we extract the apparent OA rate
constant, kOA[ArI], as a function of the aryl iodide
concentration (Figure 5D, inset), where the slope of the linear
a
Top panel shows the reaction used in photochemical kinetics studies
with the bromo analogue of 1a; bottom panel shows generalization to
a range of substrates.
top). Similarly, high yields were also obtained when TBAI was
added since pyH+ is able to be generated in situ (Table S1).
Finally, multiple aryl bromides and thiols (Table 3, bottom)
were tested. The optimized strategy applied well to aryl
bromides with electron-withdrawing groups (4−8), electron-
donating groups (1b, 9), and aryl heterocycles (10−12). Aryl
thiols (13−15) and alkyl thiols (2, 16−18) were also well
accommodated by the strategy. Moreover, the amino acid
2011
https://dx.doi.org/10.1021/jacs.0c11937
J. Am. Chem. Soc. 2021, 143, 2005−2015

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 6. Complete reaction mechanism and key rate constants for photoredox-mediated nickel-catalyzed aryl thiolation. aPCET reaction between
b
c
d
I2•− and compound 2. PCET reaction between I2•− and thiolphenol. Rate was reported by ref 53. For the reaction with 200 mM pyridine.
eOxidative addition of 4-iodotoluene. NiII = LnNiIIX2; NiI = LnNiIX. ET = electron transfer; PT = proton transfer.
cysteine (18) shows a high yield, indicating potential
applicability for biomolecule preparation. Therefore, the
generality of our optimization strategy was confirmed.
decrease due to higher concentrations of the intermediates
being generated at higher excitation powers. This was indeed
observed, as the quantum yield increased significantly as the
power of the excitation source was diminished (Figure 2B).
Such an observation is in line with our previous results for
photoredox aryl etherification, where the Ni(I/III) cycle is
enhanced as the deactivation of the on-cycle nickel
intermediates is attenuated at lower powers.29 However, the
photoredox aryl etherification and the C−S cross-coupling
reaction reported here access the on-cycle Ni(I) catalyst in
distinct ways. The direct reduction pathway, which is
predominant in aryl etherification, becomes inefficient for
generating Ni(I) in C−S cross-coupling due to the retarded
reaction (kET3) between Ir(II) and Ni(II) as compared to the
competing back reaction (kBET1) between Ir(II) and I2•−
(Figure 3 and Figure S10). The pyH+ that is formed in situ
provides an efficient pathway to transport electrons from Ir(II)
to Ni(II) through pyH+/pyH• mediators to form Ni(I) and
hence sustain the productive Ni(I/III) cycle (Figure 6).
Although pyH+ has often been used as a redox mediator in
CO2 reduction55,56,62,63 and, recently, in nickel-catalyzed aryl
amination with catalytic amounts of Zn metal as a reductant,64
our discovery reveals an underappreciated role of pyH+ as a
viable mediator in photoredox chemistry.
The mechanism in Figure 6 differs from the previously
reported “radical” and “oxidation state modulation” mecha-
nisms25,44 (Scheme 1A) in critical ways. These mechanisms
propose the production of Ni(I) by reduction of Ni(II) for
each turnover in the cycle. This restricts the quantum yield to
be Φ ≤ 1, which is inconsistent with our measurements
(Figure 2B). Additionally, these two previously proposed
mechanisms invoked a reductive quenching of *Ir(III) by thiol
to generate Ir(II) and thiyl radical cation25 or, alternatively,
oxidative quenching of *Ir(III) by Ni(II) to generate Ni(I)
and Ir(IV) (Scheme 1A).44 Based on the relative kq values in
Table 2, we found that the initial step subsequent to the
photoexcitation of Ir(III) predominantly involves the reduc-
tive quenching of *Ir(III) by I− to generate Ir(II) and I•. The
rapid sequestration of I• by I− to generate I2•− with a rate
constant close to the diffusional limit (k0)53 carries the benefit
of suppressing the back-electron transfer between I• and
Ir(II), which is propagated to the self-sustaining Ni(I/III)
catalytic cycle (Figure 6, green cycle) via pyridinium redox
DISCUSSION
■
Figure 6 summarizes the elementary reactions, reaction
intermediates, and critical rate constants for a comprehensive
description of the photoredox nickel-catalyzed cross-coupling
between aryl halides and thiols. The productive reaction
pathway consists of the following steps: (i) photoexcitation of
Ir(III) to generate *Ir(III); (ii) reductive quenching of
*Ir(III) by I− to generate Ir(II) and I• (Figure 6, red cycle);
(iii) reduction of pyH+ by Ir(II) to generate a pyridyl radical
pyH• and Ir(III); (iv) reduction of the Ni(II) precatalyst by
pyH• to form a Ni(I) species (Figure 6, blue cycle); (v)
oxidative addition of aryl halide to Ni(I) to form a putative
Ni(III) aryl halide complex; and (vi) ligand exchange on
Ni(III) and subsequent reductive elimination to release
product and reform Ni(I) (Figure 6, green cycle). In addition
to the major pathway of generating Ni(I) from step (iv), a
minor pathway was also identified that (vii) generates Ni(I)
from the direct reaction between Ir(II) and Ni(II).
The important role of the reactivity of downstream products
originating from photogenerated I• is shown in the orange
cycle of Figure 6. Following the initial quenching, I• complexes
with I− to form I2•−,53 which reacts with Ir(II) or Ni(I) via
back electron transfer, leading to the deactivation of these two
key intermediates. Alternatively, I2•− may oxidize thiol in the
presence of pyridine via PCET to generate a thiyl radical and
pyH+, both of which eventually undergo back reaction with
Ni(I) in what is likely a multistep process involving electron
and proton transfer (Figure 6, purple cycle). Additionally, I2•−
can also disproportionate to I− and I3 (Figure 3B), the latter
−
of which can also oxidize Ni(I). However, under actual
reaction conditions, this disproportionation becomes negligible
given the fast back electron transfer and PCET reactions.
The mechanism we have elucidated herein highlights the
importance of accessing and perpetuating a thermally sustained
and productive Ni(I/III) cycle, which is evidenced by the
larger-than-one quantum yields observed at low powers
(Figure 2B). Due to the bimolecular nature of all the back-
electron transfer reactions, the overall energy efficiency will
2012
https://dx.doi.org/10.1021/jacs.0c11937
J. Am. Chem. Soc. 2021, 143, 2005−2015

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
mediation (Figure 6, blue cycle). Moreover, the thiyl radical in
the radical mechanism, whose presence was inferred from
reported trapping experiments25 and corroborated by our TA
studies (Figure 4), was previously proposed to be generated by
direct quenching of *Ir(III) and on-cycle with nickel cross-
coupling catalysis (Scheme 1A, top path).25 In contrast, we
discovered that the radical may be generated by its PCET
reaction with I2•− (Figure 4) and, most importantly, is not
required for the nickel cross-coupling catalytic cycle (Figure
6).
Our mechanistic insights, together with well-defined kinetics
of the complete photoredox cycle, enable the optimization of
the cross-coupling reaction as follows: (1) The induction
period in the reaction, which is in part due to the absence of
efficient quenchers of *Ir(III) to generate Ir(II) (Table 2
except pyHI), is significantly diminished by the addition of 50
mM pyHI (Figure 2A); (2) the overall reaction efficiency and
quantum yields are improved by introducing pyHI to mediate
the formation of the on cycle Ni(I) intermediate; and (3)
notably, adding iodide to the initial reaction circumvents the
requirement of substrate to provide an iodide source, thus
allowing for aryl bromides to become effective substrates
(Table 3). This expansion of the substrate scope increases the
utility of the methodology since aryl bromides typically exhibit
greater stability and commercial availability when compared to
their iodo-counterparts, which may exhibit unwanted reactivity
in multistep synthetic pathways.
Materials and methods, data analysis, photoredox
catalysis setups, quenching studies, TA kinetics, controls,
and reaction optimizations (PDF)
AUTHOR INFORMATION
■
Corresponding Author
Daniel G. Nocera − Department of Chemistry and Chemical
Biology, Harvard University, Cambridge, Massachusetts
02138, United States; orcid.org/0000-0001-5055-320X;
Email: dnocera@fas.harvard.edu
Authors
Yangzhong Qin − Department of Chemistry and Chemical
Biology, Harvard University, Cambridge, Massachusetts
02138, United States; orcid.org/0000-0002-2450-521X
Rui Sun − Department of Chemistry and Chemical Biology,
Harvard University, Cambridge, Massachusetts 02138,
United States; orcid.org/0000-0002-3894-8425
Nikolas P. Gianoulis − Department of Chemistry and
Chemical Biology, Harvard University, Cambridge,
Massachusetts 02138, United States
Complete contact information is available at:
https://pubs.acs.org/10.1021/jacs.0c11937
Author Contributions
‡These authors contributed equally.
Funding
CONCLUSION
■
This work was supported by the National Science Foundation
under grant CHE-1855531.
Combining Stern−Volmer quenching studies, quantum yield
measurements, and nanosecond TA spectroscopy, we have
identified the productive reaction pathways along with critical
reaction intermediates and rate constants for photoredox
nickel-catalyzed cross-coupling between thiols and aryl halides.
We have found that a self-sustained Ni(I/III) cycle is operative
for product formation, in contrast to previously proposed
closed photocycles involving Ni(I), Ni(II), and Ni(III) states.
In addition to identifying the productive cycle, we also
determined that pyHI formed in situ serves three crucial roles
in facilitating the cross-coupling reaction: (1) I− serves as an
effective quencher for *Ir(III) to form I• and the highly
reducing Ir(II); (2) pyH+ serves as an electron shuttle between
Ir(II) and Ni(II) to form Ir(III) and Ni(I) through the
intermediacy of pyH•; and (3) pyH+ prevents formation of
nickel thiolate complexes, which would interfere with the light
absorption of Ir(III) via an inner filter effect. Knowledge of
these reaction pathways and the roles of pyHI allowed us to
optimize the reaction efficiency and expand the substrate scope
from aryl iodides to include aryl bromides, thus broadening the
applicability of photoredox C−S cross-coupling chemistry. The
broader deployment of redox mediators such as I−/I2•− and
pyH+/pyH• may merit further investigation and can provide a
general strategy for future photoredox methods development.
In summary, this study demonstrates an example of how
mechanistic understanding of complex photoredox systems can
provide information on the optimization and development of
photoredox reaction methodologies.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Drs. Adam Rieth, Miguel Gonzalez, and Qilei Zhu
for insightful discussions.
REFERENCES
■
(1) Twilton, J.; Le, C.; Zhang, P.; Shaw, M. H.; Evans, R. W.;
MacMillan, D. W. C. The Merger of Transition Metal and
Photocatalysis. Nat. Rev. Chem. 2017, 1, 0052.
(2) Tellis, J. C.; Kelly, C. B.; Primer, D. N.; Jouffroy, M.; Patel, N.
R.; Molander, G. A. Single-Electron Transmetalation via Photoredox/
Nickel Dual Catalysis: Unlocking a New Paradigm for sp3-sp2 Cross-
Coupling. Acc. Chem. Res. 2016, 49, 1429−1439.
(3) Ghosh, I.; Marzo, L.; Das, A.; Shaikh, R.; Konig, B. Visible Light
Mediated Photoredox Catalytic Arylation Reactions. Acc. Chem. Res.
2016, 49, 1566−1577.
(4) Levin, M. D.; Kim, S.; Toste, F. D. Photoredox Catalysis Unlocks
Single-Electron Elementary Steps in Transition Metal Catalyzed
Cross-Coupling. ACS Cent. Sci. 2016, 2, 293−301.
(5) Diccianni, J. B.; Diao, T. N. Mechanisms of Nickel-Catalyzed
Cross-Coupling Reactions. Trends Chem. 2019, 1, 830−844.
(6) Hossain, A.; Bhattacharyya, A.; Reiser, O. Copper’s Rapid Ascent
in Visible-Light Photoredox Catalysis. Science 2019, 364, eaav9713.
(7) McAtee, R. C.; McClain, E. J.; Stephenson, C. R. J. Illuminating
Photoredox Catalysis. Trends Chem. 2019, 1, 111−125.
(8) Hockin, B. M.; Li, C. F.; Robertson, N.; Zysman-Colman, E.
Photoredox Catalysts Based on Earth-Abundant Metal Complexes.
Catal. Sci. Technol. 2019, 9, 889−915.
(9) Tlahuext-Aca, A.; Hopkinson, M. N.; Daniliuc, C. G.; Glorius, F.
Oxidative Addition to Gold(I) by Photoredox Catalysis: Straightfor-
ward Access to Diverse (C,N)-Cyclometalated Gold(III) Complexes.
Chem. - Eur. J. 2016, 22, 11587−11592.
ASSOCIATED CONTENT
■
sı
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/jacs.0c11937.
2013
https://dx.doi.org/10.1021/jacs.0c11937
J. Am. Chem. Soc. 2021, 143, 2005−2015

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
(10) Reed, N. L.; Herman, M. I.; Miltchev, V. P.; Yoon, T. P.
Photocatalytic Oxyamination of Alkenes: Copper(II) Salts as
Terminal Oxidants in Photoredox Catalysis. Org. Lett. 2018, 20,
7345−7350.
(29) Sun, R.; Qin, Y.; Ruccolo, S.; Schnedermann, C.; Costentin, C.;
Nocera, D. G. Elucidation of a Redox-Mediated Reaction Cycle for
Nickel-Catalyzed Cross Coupling. J. Am. Chem. Soc. 2019, 141, 89−
93.
(30) Till, N. A.; Tian, L.; Dong, Z.; Scholes, G. D.; MacMillan, D.
W. C. Mechanistic Analysis of Metallaphotoredox C-N Coupling:
Photocatalysis Initiates and Perpetuates Ni(I)/Ni(III) Coupling
Activity. J. Am. Chem. Soc. 2020, 142, 15830−15841.
(31) Jarrett, J. T. The Biosynthesis of Thiol- and Thioether-
Containing Cofactors and Secondary Metabolites Catalyzed by
Radical S-Adenosylmethionine Enzymes. J. Biol. Chem. 2015, 290,
3972−3979.
(32) Ilardi, E. A.; Vitaku, E.; Njardarson, J. T. Data-Mining for Sulfur
and Fluorine: An Evaluation of Pharmaceuticals to Reveal
Opportunities for Drug Design and Discovery. J. Med. Chem. 2014,
57, 2832−2842.
(33) Scott, K. A.; Njardarson, J. T. Analysis of US FDA-Approved
Drugs Containing Sulfur Atoms. Top. Curr. Chem. 2018, 376.
DOI: 10.1007/s41061-018-0184-5.
(34) Feng, M. H.; Tang, B. Q.; Liang, S. H.; Jiang, X. F. Sulfur
Containing Scaffolds in Drugs: Synthesis and Application in
Medicinal Chemistry. Curr. Top. Med. Chem. 2016, 16, 1200−1216.
(35) Eichman, C. C.; Stambuli, J. P. Transition Metal Catalyzed
Synthesis of Aryl Sulfides. Molecules 2011, 16, 590−608.
(36) Fernandez-Rodriguez, M. A.; Shen, Q. L.; Hartwig, J. F. A
General and Long-Lived Catalyst for the Palladium-Catalyzed
Coupling of Aryl Halides with Thiols. J. Am. Chem. Soc. 2006, 128,
2180−2181.
(37) Murata, M.; Buchwald, S. L. A General and Efficient Method
for the Palladium-Catalyzed Cross-Coupling of Thiols and Secondary
Phosphines. Tetrahedron 2004, 60, 7397−7403.
(38) Ke, F.; Qu, Y. Y.; Jiang, Z. Q.; Li, Z. K.; Wu, D.; Zhou, X. G. An
Efficient Copper-Catalyzed Carbon-Sulfur Bond Formation Protocol
in Water. Org. Lett. 2011, 13, 454−457.
(39) Qiao, Z. J.; Wei, J. P.; Jiang, X. F. Direct Cross-Coupling Access
to Diverse Aromatic Sulfide: Palladium-Catalyzed Double C-S Bond
Construction Using Na2S2O3 as a Sulfurating Reagent. Org. Lett.
2014, 16, 1212−1215.
(40) Wang, X.; Cuny, G. D.; Noel, T. A Mild, One-Pot Stadler-
Ziegler Synthesis of Arylsulfides Facilitated by Photoredox Catalysis
in Batch and Continuous-Flow. Angew. Chem., Int. Ed. 2013, 52,
7860−7864.
(41) Liu, B.; Lim, C. H.; Miyake, G. M. Visible-Light-Promoted C-S
Cross-Coupling via Intermolecular Charge Transfer. J. Am. Chem. Soc.
2017, 139, 13616−13619.
(42) Jiang, M.; Li, H. F.; Yang, H. J.; Fu, H. Room-Temperature
Arylation of Thiols: Breakthrough with Aryl Chlorides. Angew. Chem.,
Int. Ed. 2017, 56, 874−879.
(43) Sikari, R.; Sinha, S.; Das, S.; Saha, A.; Chakraborty, G.; Mondal,
R.; Paul, N. D. Achieving Nickel Catalyzed C-S Cross-Coupling under
Mild Conditions Using Metal Ligand Cooperativity. J. Org. Chem.
2019, 84, 4072−4085.
(44) Ren, H.; Li, G. F.; Zhu, B.; Lv, X. D.; Yao, L. S.; Wang, X. L.;
Su, Z. M.; Guan, W. How Does Iridium(III) Photocatalyst Regulate
Nickel(II) Catalyst in Metallaphotoredox-Catalyzed C-S Cross-
Coupling? Theoretical and Experimental Insights. ACS Catal. 2019,
9, 3858−3865.
(11) Ackerman, L. K. G.; Alvarado, J. I. M.; Doyle, A. G. Direct C-C
Bond Formation from Alkanes Using Ni-Photoredox Catalysis. J. Am.
Chem. Soc. 2018, 140, 14059−14063.
(12) Shimomaki, K.; Murata, K.; Martin, R.; Iwasawa, N. Visible-
Light-Driven Carboxylation of Aryl Halides by the Combined Use of
Palladium and Photoredox Catalysts. J. Am. Chem. Soc. 2017, 139,
9467−9470.
(13) Yoo, W. J.; Tsukamoto, T.; Kobayashi, S. Visible-Light-
Mediated Chan-Lam Coupling Reactions of Aryl Boronic Acids and
Aniline Derivatives. Angew. Chem., Int. Ed. 2015, 54, 6587−6590.
(14) He, J.; Chen, C. Y.; Fu, G. C.; Peters, J. C. Visible-Light-
Induced, Copper-Catalyzed Three-Component Coupling of Alkyl
Halides, Olefins, and Trifluoromethylthiolate to Generate Trifluor-
omethyl Thioethers. ACS Catal. 2018, 8, 11741−11748.
(15) Zhang, G. T.; Liu, C.; Yi, H.; Meng, Q. Y.; Bian, C. L.; Chen,
H.; Jian, J. X.; Wu, L. Z.; Lei, A. W. External Oxidant-Free Oxidative
Cross-Coupling: A Photoredox Cobalt-Catalyzed Aromatic C-H
Thiolation for Constructing C-S Bonds. J. Am. Chem. Soc. 2015,
137, 9273−9280.
(16) Hartwig, J. F. Carbon-Heteroatom Bond Formation Catalysed
by Organometallic Complexes. Nature 2008, 455, 314−322.
(17) Park, B. Y.; Pirnot, M. T.; Buchwald, S. L. Visible Light-
Mediated (Hetero)Aryl Amination Using Ni(II) Salts and Photoredox
Catalysis in Flow: A Synthesis of Tetracaine. J. Org. Chem. 2020, 85,
3234−3244.
(18) Lim, C. H.; Kudisch, M.; Liu, B.; Miyake, G. M. C-N Cross-
Coupling via Photoexcitation of Nickel-Amine Complexes. J. Am.
Chem. Soc. 2018, 140, 7667−7673.
(19) Key, R. J.; Vannucci, A. K. Nickel Dual Photoredox Catalysis
for the Synthesis of Aryl Amines. Organometallics 2018, 37, 1468−
1472.
(20) Corcoran, E. B.; Pirnot, M. T.; Lin, S. S.; Dreher, S. D.;
DiRocco, D. A.; Davies, I. W.; Buchwald, S. L.; MacMillan, D. W. C.
Aryl Amination Using Ligand-Free Ni(II) Salts and Photoredox
Catalysis. Science 2016, 353, 279−283.
(21) Terrett, J. A.; Cuthbertson, J. D.; Shurtleff, V. W.; MacMillan,
D. W. C. Switching on Elusive Organometallic Mechanisms with
Photoredox Catalysis. Nature 2015, 524, 330−334.
(22) Welin, E. R.; Le, C.; Arias-Rotondo, D. M.; McCusker, J. K.;
MacMillan, D. W. C. Photosensitized, Energy Transfer-Mediated
Organometallic Catalysis through Electronically Excited Nickel(II).
Science 2017, 355, 380−384.
(23) Kim, T.; McCarver, S. J.; Lee, C.; MacMillan, D. W. C.
Sulfonamidation of Aryl and Heteroaryl Halides through Photo-
sensitized Nickel Catalysis. Angew. Chem., Int. Ed. 2018, 57, 3488−
3492.
(24) Jouffroy, M.; Kelly, C. B.; Molander, G. A. Thioetherification
via Photoredox/Nickel Dual Catalysis. Org. Lett. 2016, 18, 876−879.
(25) Oderinde, M. S.; Frenette, M.; Robbins, D. W.; Aquila, B.;
Johannes, J. W. Photoredox Mediated Nickel Catalyzed Cross-
Coupling of Thiols with Aryl and Heteroaryl Iodides via Thiyl
Radicals. J. Am. Chem. Soc. 2016, 138, 1760−1763.
(26) Shields, B. J.; Kudisch, B.; Scholes, G. D.; Doyle, A. G. Long-
Lived Charge-Transfer States of Nickel(II) Aryl Halide Complexes
Facilitate Bimolecular Photoinduced Electron Transfer. J. Am. Chem.
Soc. 2018, 140, 3035−3039.
(45) Sun, R.; Qin, Y.; Nocera, D. G. General Paradigm in
Photoredox Ni-Catalyzed Cross-Coupling Allows for Light-Free
Access to Reactivity. Angew. Chem., Int. Ed. 2020, 59, 9527−9533.
(46) Qin, Y. Z.; Martindale, B. C. M.; Sun, R.; Rieth, A. J.; Nocera,
D. G. Solar-Driven Tandem Photoredox Nickel-Catalysed Cross-
Coupling Using Modified Carbon Nitride. Chem. Sci. 2020, 11,
7456−7461.
(47) Hatchard, C. G.; Parker, C. A. A New Sensitive Chemical
Actinometer - II. Potassium Ferrioxalate as a Standard Chemical
Actinometer. Proc. R. Soc. London, Ser. A 1956, 235, 518−536.
(48) Holder, P. G.; Pizano, A. A.; Anderson, B. L.; Stubbe, J.;
Nocera, D. G. Deciphering Radical Transport in the Large Subunit of
(27) Ting, S. I.; Garakyaraghi, S.; Taliaferro, C. M.; Shields, B. J.;
3
Scholes, G. D.; Castellano, F. N.; Doyle, A. G. d-d Excited States of
Ni(II) Complexes Relevant to Photoredox Catalysis: Spectroscopic
Identification and Mechanistic Implications. J. Am. Chem. Soc. 2020,
142, 5800−5810.
(28) Yin, H. L.; Fu, G. C. Mechanistic Investigation of
Enantioconvergent Kumada Reactions of Racemic Alpha-Bromoke-
tones Catalyzed by a Nickel/Bis(Oxazoline) Complex. J. Am. Chem.
Soc. 2019, 141, 15433−15440.
2014
https://dx.doi.org/10.1021/jacs.0c11937
J. Am. Chem. Soc. 2021, 143, 2005−2015

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Class I Ribonucleotide Reductase. J. Am. Chem. Soc. 2012, 134,
1172−1180.
(49) Miller, D. C.; Choi, G. J.; Orbe, H. S.; Knowles, R. R. Catalytic
Olefin Hydroamidation Enabled by Proton-Coupled Electron Trans-
fer. J. Am. Chem. Soc. 2015, 137, 13492−13495.
(50) Qiu, G. Q.; Knowles, R. R. Understanding Chemoselectivity in
Proton-Coupled Electron Transfer: A Kinetic Study of Amide and
Thiol Activation. J. Am. Chem. Soc. 2019, 141, 16574−16578.
(51) Ruccolo, S.; Qin, Y. Z.; Schnedermann, C.; Nocera, D. G.
General Strategy for Improving the Quantum Efficiency of Photo-
redox Hydroamidation Catalysis. J. Am. Chem. Soc. 2018, 140,
14926−14937.
(52) Kolthoff, I. M.; Chantooni, M. K.; Bhowmik, S. Dissociation
Constants of Uncharged and Monovalent Cation Acids in Dimethyl
Sulfoxide. J. Am. Chem. Soc. 1968, 90, 23−28.
(53) Gardner, J. M.; Abrahamsson, M.; Farnum, B. H.; Meyer, G. J.
Visible Light Generation of Iodine Atoms and I-I Bonds: Sensitized I−
−
Oxidation and I3 Photodissociation. J. Am. Chem. Soc. 2009, 131,
16206−16214.
(54) Wang, X. G.; Stanbury, D. M. Oxidation of Iodide by a Series of
Fe(III) Complexes in Acetonitrile. Inorg. Chem. 2006, 45, 3415−
3423.
(55) Keith, J. A.; Carter, E. A. Theoretical Insights into Pyridinium-
Based Photoelectrocatalytic Reduction of CO2. J. Am. Chem. Soc.
2012, 134, 7580−7583.
(56) Lim, C. H.; Holder, A. M.; Hynes, J. T.; Musgrave, C. B.
Reduction of CO2 to Methanol Catalyzed by a Biomimetic Organo-
Hydride Produced from Pyridine. J. Am. Chem. Soc. 2014, 136,
16081−16095.
(57) Elliot, A. J. A Pulse-Radiolysis Study of the Reaction of OH
−
with I2 and the Decay of I2 . Can. J. Chem. 1992, 70, 1658−1661.
(58) Cukier, R. I.; Nocera, D. G. Proton-Coupled Electron Transfer.
Annu. Rev. Phys. Chem. 1998, 49, 337−369.
(59) Reece, S. Y.; Nocera, D. G. Proton-Coupled Electron Transfer
in Biology: Results from Synergistic Studies in Natural and Model
Systems. Annu. Rev. Biochem. 2009, 78, 673−699.
(60) Tsou, T. T.; Kochi, J. K. Mechanism of Oxidative Addition -
Reaction of Nickel(0) Complexes with Aromatic Halides. J. Am.
Chem. Soc. 1979, 101, 6319−6332.
(61) Amatore, C.; Jutand, A. Rates and Mechanisms of Electron-
Transfer Nickel-Catalyzed Homocoupling and Carboxylation Reac-
tions. An Electrochemical Approach. Acta Chem. Scand. 1990, 44,
755−764.
(62) Cole, E. B.; Lakkaraju, P. S.; Rampulla, D. M.; Morris, A. J.;
Abelev, E.; Bocarsly, A. B. Using a One-Electron Shuttle for the
Multielectron Reduction of CO2 to Methanol: Kinetic, Mechanistic,
and Structural Insights. J. Am. Chem. Soc. 2010, 132, 11539−11551.
(63) Lebegue, E.; Agullo, J.; Belanger, D. Electrochemical Behavior
of Pyridinium and N-Methyl Pyridinium Cations in Aqueous
Electrolytes for CO2 Reduction. ChemSusChem 2018, 11, 219−228.
(64) Han, D. Y.; Li, S. S.; Xia, S. Q.; Su, M. C.; Jin, J. Nickel-
Catalyzed Amination of (Hetero)Aryl Halides Facilitated by a
Catalytic Pyridinium Additive. Chem. - Eur. J. 2020, 26, 12349−
12354.
2015
https://dx.doi.org/10.1021/jacs.0c11937
J. Am. Chem. Soc. 2021, 143, 2005−2015