Mechanistic Analysis of Metallaphotoredox C-N Coupling: Photocatalysis Initiates and Perpetuates Ni(I)/Ni(III) Coupling Activity

Article abstract of DOI:10.1021/jacs.0c05901

The combined use of reaction kinetic analysis, ultrafast spectroscopy, and stoichiometric organometallic studies has enabled the elucidation of the mechanistic underpinnings to a photocatalytic C-N cross-coupling reaction. Steady-state and ultrafast spect

Full text of DOI:10.1021/jacs.0c05901

Subscriber access provided by CARLETON UNIVERSITY
Article
A Mechanistic Analysis of Metallaphotoredox C–N Coupling:
Photocatalysis Initiates and Perpetuates Ni(I)/Ni(III) Coupling Activity
Nicholas A. Till, Lei Tian, Zhe Dong, Gregory Scholes, and David W. C. MacMillan
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.0c05901 • Publication Date (Web): 04 Aug 2020
Downloaded from pubs.acs.org on August 4, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 12
Journal of the American Chemical Society
1
2
3
4
5
6
A Mechanistic Analysis of Metallaphotoredox C–N Coupling: Photo-
catalysis Initiates and Perpetuates Ni(I)/Ni(III) Coupling Activity
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
†
§
†
§
,†
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Nicholas A. Till , Lei Tian , Zhe Dong , Gregory D. Scholes , and David W. C. MacMillan*
†
Merck Center for Catalysis at Princeton University, Princeton, New Jersey 08544, United States
§
Department of Chemistry, Princeton University, Princeton, New Jersey 08544, United States
ABSTRACT: The combined use of reaction kinetic analysis, ultrafast spectroscopy, and stoichiometric organometallic studies has
enabled the elucidation of the mechanistic underpinnings to a photocatalytic C–N cross-coupling reaction. Steady-state and ultrafast
spectroscopic techniques were used to track the excited-state evolution of the employed iridium photocatalyst, determine the resting
states of both iridium and nickel catalysts, and uncover the photochemical mechanism for reductive activation of the nickel co-
catalyst. Stoichiometric organometallic studies along with a comprehensive kinetic study of the reaction, including rate-driving force
analysis, unveiled the crucial role of photocatalysis in both initiating and sustaining a Ni(I)/Ni(III) cross-coupling mechanism. The
insights gleaned from this study further enabled the discovery of a new photocatalyst providing a >30-fold rate increase.
Ir/Ni metallaphotoredox C–N coupling
Introduction
Over the past 50 years, transition metal-catalyzed cross-cou-
pling reactions have risen to the forefront of synthetic technol-
ogy, enabling rapid access to molecular complexity in medici-
nal chemistry, process chemistry, as well as materials science
Br
broad scope
Ir
Ni
R
R
highly scalable
N
mechanism
unclear
DABCO
H
N
R
R
1
–3
settings. In order to promote a wide array of bond-forming
processes through cross-coupling, the synthetic community has
sought catalysts capable of undertaking a sequence of elemen-
tary organometallic steps in a generic and robust manner. To
answer this call, major advances in Pd-catalyzed C–C and C–N
cross-coupling technologies have resulted from the rational de-
how does photocatalysis switch on challenging C–N bond formation?
-
10
0
.5
4–6
7
sign of ligands. Rates of oxidative addition, transmetalla-
-11
1.5
R²
=
0.99672
8
9
tion, and reductive elimination with a given transition metal
catalyst can all be tuned to achieve high reaction efficiencies
through the judicious modulation of ligand electronic and steric
properties. In particular, the ligand design work by Buchwald
and Hartwig in the area of C–N cross-coupling stands as a quin-
tessential example of this approach in the context of an espe-
cially important reaction to the practicing synthetic chemist.
More specifically, tuning the steric and electronic parameters of
-12
-
1.45
-1.35
-1.25
reaction kinetics
stoichiometric studies
ultrafast spectroscopy
R
R
>30-fold rate increase
N
H
+
Ar–Br
+
–
6
PF
Me
1
0–13
H N
N
Ir
N
IrIII
dark
cycle
2
mono- and bidentate phosphines
has enabled the often chal-
NiII
NiI
NiIII
N
lenging C–N reductive elimination step, imparting the palla-
dium-catalyzed arylation of amines with both high efficiency
N
2
H N
14,15
Me
and broad scope.
Ar–NR
2
In recent years photocatalysis has emerged as a complemen-
tary alternative to ligand design, wherein visible light can
revised mechanism
improved catalyst
“
switch on” otherwise sluggish or inaccessible elementary or-
Figure 1. A mechanistic basis for C–N cross-coupling.
1
6
ganometallic steps such as transmetallation, oxidative addi-
using a simple Ni(II) catalyst, with no exogenous ligand, and a
mild organic base in combination with an Ir photocatalyst at a
1
7
18
tion, and reductive elimination. The merger of transition
metal and photocatalysis has proven to be particularly advanta-
geous in the context of facilitating challenging reductive elimi-
loading of 200 ppm to effect efficient and scalable C–N cross-
23,24
coupling with a broad substrate scope.
In this way, the use
1
8,19
20
21,22
nation events to forge C–O , C–N , and C–CF
3
bonds,
of visible light photoactivation, rather than ligand design, ap-
pears to promote an otherwise challenging C–N bond-for-
mation. Given its synthetic relevance and operational simplic-
ity, a detailed mechanistic description of this C–N cross-cou-
pling would be valuable to both understanding and designing
wherein photocatalysis provides access to high-valent or ex-
cited-state metal complexes (Ni(III), Ni(II)*, Cu(III), or
Au(III)) poised for rapid bond-formation. This reaction design
principle was recently applied to the N-arylation of amines,
ACS Paragon Plus Environment

Journal of the American Chemical Society
dual-catalytic systems. More broadly speaking, continued ex- Table 1. Control experiments for the C–N coupling reaction.
Page 2 of 12
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
pansion of the repertoire of elementary organometallic steps
available through visible light excitation requires a clear mech-
anistic understanding of the fundamental photochemical pro-
cesses underlying the observed catalytic activity.
5 mol% NiBr
•3H
O
H
2
2
Br
N
0
.02 mol% photocatalyst 1
hexyl
H
2
N
hexyl
DMF, 23 °C, 36 h
F3C
F3C
DABCO (1.8 equiv)
aryl bromide 2
hexylamine
coupled product 3
A key step in understanding photoinduced Ir/Ni-catalyzed
cross-coupling pathways is determining the photochemical ba-
sis for Ni co-catalyst activation. Within the context of Ir(III) and
Ru(II) photocatalysts typically employed in nickel/photoredox
cross-coupling reactions, the long lifetimes (up to 2 µs), high
triplet energies (often >2 eV), and high driving forces for re-
entry
conditions
yielda
+
–
PF
6
Me
1
2
3
4
5
no change
no light
no 1
81%
<1%
<1%
<1%
14%
N
Ir
N
N
N
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
red
n*
n–1
ductive quenching (E1/2 [M /M ] > +1 V vs. SCE) of these
Me
no NiBr
2 2
•3H O
photocatalysts’ excited-states allow for a multitude of diffu-
2
5,26
no DABCO
sional quenching pathways.
In keeping with these photo-
photocatalyst 1
a
physical characteristics, photoactivation of the nickel co-cata-
lyst can take place through either energy transfer or electron
transfer (ET), depending on the photophysical and electrochem-
ical properties of the quenchers present in solution. In practice,
both of these general quenching mechanisms have been ob-
served: energy transfer to access a Ni(II)* excited-state has
been shown to promote C–O reductive elimination in a stoichi-
Performed with 1.0 equiv aryl bromide 2, 1.5 equiv hexyla-
19
mine. Yields were determined by F NMR analysis.
mechanism underlying the overall photocatalytic process, we
first sought to determine the photochemical fate of the excited-
state iridium catalyst under synthetically relevant reaction con-
ditions. To this end, a comprehensive approach to establishing
the major deexcitation pathway for the photocatalyst was taken.
More specifically, all reaction components were evaluated as
potential quenchers for the long-lived excited-state (τ = 870 ns,
Table S2, Figure S103) of photocatalyst 1 using steady-state
emission measurements. A Stern-Volmer quenching analysis of
photocatalyst 1 with DABCO, the soluble organic base em-
ployed in this reaction, revealed its high efficiency for quench-
2
7
ometric context, and initiate C–N bond-formation in a related
28
catalytic reaction reported by Miyake. On the other hand, re-
ductive quenching to access a high-valent Ni(III) intermediate
is thought to enable C–O bond-formation in the coupling of al-
1
8,29
cohols and aryl halides.
While the primary photocatalyst quenching mechanism can
be crucial in determining the mode of Ni catalysis, subsequent
diffusional reactions can dramatically influence the eventual
bond-forming process. As one example, recent work by Doyle
has demonstrated that initially accessed Ni(II)* states can un-
dergo subsequent electron transfer events to access Ni(III) and
9
–1 –1
q
ing Ir(III)* photoluminescence (k = 2.75(0.07) × 10 M s ,
Figure 2A). Comparing this quenching efficiency with that of
all remaining reaction components reveals the dominant role
that DABCO plays in photocatalyst quenching ([DABCO] =
4
q
86 mM, Φ (DABCO) >0.99, page S26). While these results
3
0
Ni(I) species. Additionally, Nocera and coworkers have
shown that initial electron transfer quenching by an organic
base indirectly facilitates the formation of a Ni(III) complex ca-
clearly demonstrate that nearly every photon absorbed by the
Ir(III) photocatalyst is eventually funneled into this DABCO
quenching pathway, it does not make clear that this quenching
pathway is relevant to product formation. More specifically,
this analysis leaves open the possibility of a low quantum yield
process occurring rarely, but nonetheless leading to the ob-
served product-forming reactivity. To further interrogate the
relevance of this dominant quenching pathway to product for-
mation, initial rate measurements were carried out at various
concentrations of DABCO. In the case that a low quantum yield
quenching pathway is actually responsible for product for-
mation, lower DABCO concentrations should increase the
2
9
pable of undergoing reductive elimination. These considera-
tions are particularly relevant to Ni-catalyzed C–X bond-for-
mation,
wherein
Ni(0)/Ni(II),
Ni(I)/Ni(III),
and
Ni(0)/Ni(II)/Ni(III)/Ni(I) mechanisms have been variously pro-
posed, and the relevance of these pathways may not be solely
dependent on the initial photocatalyst quenching mechanism.
Given these studies, we recognized that it would be necessary
to not only determine the initial deexcitation mechanism for the
photocatalyst, but to further identify the ensuing energy or elec-
tron transfer pathways leading to the required C–N bond-form-
ing step. For these reasons, a detailed mechanistic analysis in-
volving kinetic, spectroscopic, and stoichiometric studies was
undertaken to determine the specific roles of both iridium and
nickel catalysts in this synthetically valuable C–N cross-cou-
pling protocol. As a result, this in-depth study revealed that the
Ni(0)/Ni(II)/Ni(III)/Ni(I) mechanism we originally proposed
for the photoredox C–N cross-coupling is unlikely to be opera-
tive, and instead supports a reassignment to a Ni(I)/Ni(III) path-
way. This finding not only clarified the mechanistic understand-
ing of the photoredox amination reaction, but also provided cru-
cial insight necessary for developing superior catalysts.
A
B
kq(DABCO) = 1.33(0.03) × 109 s–1
DABCO promotes catalysis
k+DABCO
kq(all other reactants) = 9.5(0.7) × 105 s–1
=
9
k-DABCO
Photochemistry of the Iridium Catalyst
Φ (DABCO) > 0.99
q
The aryl amination reaction shown in Table 1 requires both
light and photocatalyst to achieve synthetically useful levels of
C–N coupling (Table 1, entries 1–3). These control reactions
make clear the crucial role of the photoexcited iridium catalyst
in promoting this reaction. To initiate our study of the
Figure 2. A. Stern-Volmer luminescence quenching of 1 by
DABCO. Standard error of the mean (n = 3) given in
parentheses. B. Initial rates of product formation in the C–N
coupling reaction with (orange) and without DABCO (blue).
Error bars represent standard error of the mean (n = 2).
ACS Paragon Plus Environment

Page 3 of 12
Journal of the American Chemical Society
A
B
C
1
2
3
4
5
6
7
8
9
0 mV
–1700 mV
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
D
E
F
minor, low Φ
major, high Φ
4 *IrIII + DABCO
DABCO•+ + IrII
5
hν
NiII
1
IrIII + NiI
IrIII + DABCO
1
Figure 3. A. Difference spectra at various pump-probe delays from a ns-µs transient absorption measurement of 1 in DMF absent of
quencher (λpump = 400 nm) and temporal evolution profile monitored at λ = 500 nm (inset). B. Pump-probe difference spectra and
temporal evolution profile monitored at λ = 540 nm (inset) for a mixture of 1 and DABCO (486 mM). C. Absorption spectra of 1 in
DMF at various applied voltages (0 mV to –1700 mV vs. SCE). D. Difference spectrum taken from B at 1000 ns pump-probe delay
blue) and difference spectrum taken from C (A(E = –1600 mV) – A(E = 0 mV)) (orange) normalized at λ = 850 nm and resulting
difference spectrum from these two traces (blue – orange, inset). E. Pump-probe difference spectra and temporal evolution profile
monitored at λ = 540 nm (inset) for a mixture of 1, DABCO (486 mM), aryl bromide 2 (270 mM), hexylamine (405 mM), and
(
2 2
NiBr •3H O (13.5 mM). F. Schematic representation of photochemical pathways for Ir(III) catalyst 1 and Ni(II) precatalyst.
quantum yield of the product-forming pathway, and thus the
measured initial rate of product formation. Instead, the reaction
rate decreases as a function of decreasing DABCO concentra-
tion (Figure S4), culminating in a full 9-fold decrease and sig-
nificantly diminished final yield upon complete removal of
DABCO from the reaction (Figure 2B and Table 1, entry 5),
corroborating the relevance of DABCO quenching to the over-
all reactivity.
state absorption (ESA) feature forms within the rise time of the
instrument, and follows a monoexponential decay with a life-
time consistent with long-lived Ir(III)* 4 (τ = 850 ns), wherein
recovery of the baseline occurs within 10 µs (Figure 3A). The
excited-state dynamics of 1 in the presence of a catalytically-
relevant DABCO concentration (486 mM) differ markedly
from the iridium-only dynamics. A positive feature distinct
from the Ir(III)* ESA forms within the rise time of the instru-
ment (consistent with the DABCO quenching rate determined
by Stern-Volmer analysis) and exhibits distinct peaks at 510
and 540 nm (Figure 3B). This feature then decays on the late
microsecond timescale, and positive signal still persists on the
early millisecond timescale ([ΔA540(t = 1000 µs)/ΔA540(t = 0)]
We next turned our attention to the mechanistic steps follow-
ing the primary quenching process. Given the oxidizing power
of the Ir(III)* excited-state 4 (E1/2 [Ir /Ir ] = +1.08 V vs.
SCE, Table S2) and relatively reducing nature of DABCO
(E1/2 [DABCO /DABCO] = +0.71 V vs. SCE, Figure S65) we
posited that DABCO quenching is reductive, giving rise to one-
electron reduced Ir(II) photocatalyst 5 and the one-electron ox-
idized form of DABCO. The highly reducing and oxidizing na-
red
III* II
red
•+
=
2%) (Figure 3B inset). Spectroelectrochemistry was per-
formed on 1 to obtain an authentic Ir(II)/Ir(III) difference spec-
trum (Figure 3C). Increasingly negative applied potentials sur-
red
III II
red
III II
passing the reduction potential of 1 (E1/2 [Ir /Ir ] = –1.41 V
vs. SCE, Figure S56) led to the growth of a positive feature with
peaks centered at 510 and 540 nm. Comparison of the resulting
Ir(II)/Ir(III) difference spectrum (obtained by subtracting the
spectrum obtained at E = 0 mV from the E = –1600 mV spec-
trum) with the difference spectrum at an early pump-probe de-
lay (t = 1000 ns) reveals broad agreement in peak shape and
position, however, with a marked discrepancy in the 410-540
nm region (Figure 3D). In this wavelength region, significant
positive signal in the pump-probe difference spectrum remains
unaccounted for by the presence of Ir(II) 5 only. Subtracting the
spectroelectrochemical data from the pump-probe data pro-
duces a broad positive peak centered around 460 nm, which
tures of these two species (E1/2 [Ir /Ir ] = –1.41 V vs. SCE,
red
•+
Figure S56, E1/2 [DABCO /DABCO] = +0.71 V vs. SCE, Fig-
ure S65), makes it possible that back electron transfer could rap-
idly follow the initial quenching electron transfer. This would
return Ir(III) 1 in its ground state form and neutral, closed-shell
DABCO, and potentially outcompete cage escape, thereby pre-
venting any ensuing diffusional processes required for cataly-
sis. In order to confirm the presence of these proposed Ir(II) and
DABCO radical cation electron transfer products, as well as as-
sess their temporal profiles, nanosecond-microsecond pump-
probe spectroscopy was carried out alongside spectroelectro-
chemistry with iridium photocatalyst 1. In a nanosecond pump-
probe experiment with 1 in the absence of quencher, an excited-
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 12
form. To probe this hypothesis and interrogate the relevance of
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Ir(II) complex 5 to the overall dual-catalytic amination reaction,
an array of Ir(III) photocatalysts (1, 6–12) with systematically
–
6
PF
5′
4
′
N
red
III II
R
R
varied reduction potentials (ranging from E1/2 [Ir /Ir ] = –0.77
V to –1.41 V) were employed in the reaction, and the rate of
product formation measured (Figure 4). In this SAR series, only
N
N
Ir+
N
4
5
red
III II
highly reducing Ir(II) photocatalysts (E1/2 [Ir /Ir ] ≤ –1.26 V)
gave rise to detectable levels of product formation (Figure 4),
providing evidence that only sufficiently reducing Ir(II) species
are able to effect the challenging precatalyst reduction step
Ir(III) photocatalysts
II
I
i
(E [Ni /Ni ] = –1.43 V vs. SCE, Figure S67, see page S31 for
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
E1/2red (V vs. SCE) kobs (μMs–1
entry
photocatalyst
[Ir(ppy) (5,5ʹ-Me bpy)]PF (1)
)
further discussion). Furthermore, the reducing power of the
Ir(II) species was found to correlate significantly with reaction
1
2
3
4
5
6
7
8
– 1.41
– 1.41
– 1.37
– 1.29
– 1.26
– 0.90
– 0.88
– 0.77
25.2(0.6)
23(1)
red
III II
2
2
6
rate, wherein a plot of ln(kobs) vs. E1/2 [Ir /Ir ] exhibits a linear
relationship over a small driving force range of 150 meV (Fig-
ure 4). A plot of kobs vs. [Ni] (Figure S8) shows a strong rate
dependence on nickel concentration, a finding which is con-
sistent with the Ir(II) intermediate directly reducing the Ni(II)
precatalyst, given that the rate of single electron transfer be-
tween Ir(II) and Ni(II) should depend not only on the reducing
power of Ir(II), but also on the concentration of Ni(II). Taking
this latter result together with the strong dependence of the re-
action rate on Ir(II) reducing power (Figure 4) it appears evident
that the Ir(II) intermediate is relevant to the overall photoredox
C–N coupling, and that the reduction of the Ni(II) precatalyst is
likely involved in determining the overall rate of product for-
mation.
[Ir(ppy) (4,4ʹ-(OMe) bpy)]PF (6)
2
2
6
[Ir(ppy) (4,4ʹ-Me bpy)]PF (7)
16.9(0.2)
9.0(0.1)
7.0(0.3)
NR
2
2
6
[Ir(ppy) (bpy)]PF6 (8)
2
[Ir(ppy) (5,5ʹ-F bpy)]PF (9)
2
2
6
[Ir(ppy) (4,4ʹ-(CO Me) bpy)]PF (10)
2
2
2
6
[Ir(ppy) (4,4ʹ-(CF ) bpy)]PF (11)
NR
2
3 2
6
[Ir(ppy) (5,5ʹ-(CF ) bpy)]PF (12)
NR
2
3 2
6
Figure 4. Structures and reduction potentials for a panel of
Ir(III) photocatalysts and initial rate measurements for the pho-
tocatalytic C–N coupling between hexylamine and ArBr 2.
Error bars represent standard error of the mean (n = 2).
matches well with previously reported visible absorption spec-
3
1
tra for the DABCO radical cation (Figure 3D, inset). Taken
together, these data support the originally posited electron
transfer products of the initial quenching event, namely reduced
state Ir(II) 5 and DABCO . Moreover, the long-lived nature of
the signal demonstrates that a significant portion of the electron
transfer products undergo solvent cage escape, and are therefore
capable of diffusional processes potentially necessary for facil-
itating C–N cross-coupling.
In order to investigate the impact of the low efficiency Ni
reduction process on the resting states of the iridium and nickel
catalysts, steady-state absorption measurements of the dual-cat-
alytic amination were carried out. The absorption spectrum for
the C–N coupling under the standard reaction conditions (fol-
lowing irradiation at 450 nm) bears remarkable similarity to the
absorption spectrum reported for an analogous Ni(II) amine
•
+
28
complex by Miyake and coworkers. This led to a tentative as-
signment of the Ni(II) catalyst in this system as octahedral
tetrakis(hexylamino)nickel(II) dibromide (13, Figure 5, orange
trace). Moreover, the absorption spectrum for the complete C–
N coupling protocol (containing all of the reaction components)
can be reconstructed from a simple sum of independently meas-
ured Ni(II) and Ir(III) spectra (Figure 5, green and purple traces,
To confirm the presence of Ir(II) 5 and to probe its subse-
quent diffusional electron transfer pathways under the catalyti-
cally relevant reaction conditions, a second pump-probe exper-
iment was carried out with photocatalyst 1, DABCO (486 mM),
and the remaining reaction components, namely: aryl bromide
2 2
2 (270 mM), hexylamine (405 mM), and NiBr •3H O (13.5
mM). In this experiment, the excited-state features and dynam-
ics are remarkably similar to those observed when DABCO is
the only quencher present (Figure 3E). While this result clearly
demonstrates the formation of Ir(II) 5 under the exact reaction
conditions employed in our protocol, the absence of measurable
reactivity towards the remaining reaction components suggests
that either i) Ir(II) 5 is not relevant to the catalytic activity, or
ii) its role in initiation is necessary, yet occurs with low quan-
tum efficiency.
PF6–
Me
N
Br
N
N
hexyl–H N
NH –hexyl
2
2
Ir+
Ni
hexyl–H N
NH –hexyl
2
2
Br
N
1
3
Me
1
catalytic reaction
Role of Ir(II) in C–N Bond-Formation
Ir(III) 1 + Ni(II) 13
In considering the role that Ir(II) intermediate 5 might play
in the dual-catalytic C–N coupling, we recognized that reduc-
tion of the Ni(II) precatalyst (NiBr •3H O) to a low-valent form
2 2
would be critical in facilitating the oxidative addition step in
this C–N cross-coupling process. More specifically, a low-va-
lent Ni(I) species would be competent to undergo oxidative ad-
dition with aryl bromide 2, whereas its Ni(II) congener would
be unable to undergo the prohibitively slow Ni(II)/Ni(IV) oxi-
dative addition. Within this mechanism, we reasoned that a
plausible role for the identified Ir(II) intermediate would be in
reducing the Ni(II) precatalyst to an active, low-valent Ni(I)
Figure 5. Steady-state UV/vis spectra of Ni(II) catalyst 13 (or-
ange), Ir(III) photocatalyst 1 (blue), their linear combination
(green) and the full reaction mixture (purple) following 450 nm
irradiation for 5 minutes.
3
2
ACS Paragon Plus Environment

Page 5 of 12
Journal of the American Chemical Society
A
mechanism A
B
mechanism B
C
mechanism C
DABCO•+
1
2
3
4
5
6
7
8
9
Ar–Br
Ar–Br
NiII
2
2
Ar–Br
2
DABCO
dark
dark
photo
NiI
NiIII
Ni0
NiII
Ni0
NiIII
I
III
0
II
I
III
Ni /Ni
Ni /Ni
Ni /Ni
IrIII
1
NiI
IrII
Ar–NHR 3
Ar–NHR 3
5
Ar–NHR 3
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
D
E
H
N
2
H
N
Br
Pr
FcBF4
Et
NiII
Pr
N
H
2
30 minutes
no light
1
1
4
5
F
3
C
F3C
CF3
CF
3
1
4
14
16, 68% yielda
H
Br
N
FcBF4
N
NiII
N
H
30 minutes
F
3
C
F
C
CF
3
3
CF
3
no light
1
5
15
17, 70% yielda
NiII
– stable towards reductive elimination
– rapid C–N reductive elimination
R
e-
NiII
NiIII
FcBF4
Ar
N
NiIII
R
Figure 6. A. General Ni(I)/Ni(III) cross-coupling mechanism showing elementary steps involved in oxidation state changes. B. Gen-
eral Ni(0)/Ni(II) cross-coupling mechanism. C. General Ni(0)/Ni(II)/Ni(III)/Ni(I) mechanism involving photoinduced reductive elim-
ination. D. Cyclic voltammetry data for Ni(II) aryl amine complexes 14 and 15 in DMF (0.1 M NBu
4 6
PF ). E. Solid state structures
and oxidatively-induced reductive elimination reactions with independently synthesized Ni(II) aryl amine complexes 14 and 15.
a
19
ORTEP structures with 70% probability ellipsoids (para-CF
3
groups disorder omitted for clarity). Yields determined by F NMR.
respectively), confirming the resting states of the two metal cat-
alysts as photocatalyst 1 and Ni(II) complex 13 (Figure 5). This
finding, that the nickel catalyst exists predominantly in its
Ni(II) oxidation state, is consistent with the spectroscopic ob-
servation that Ir(II) formation is efficient, but that the subse-
quent Ni(II) reduction step is extremely inefficient. Further-
more, the spectroscopic evidence that photocatalyst 1 is pre-
dominantly in its ground state Ir(III) form demonstrates that
back electron transfer from Ir(II) to the DABCO radical cation
is likely rapid, thereby preventing any buildup in concentration
of the reduced Ir(II) species.
oxidative addition precedes a photocatalytic oxidatively in-
duced reductive elimination and Ni(I)/Ni(0) reduction (Figure
6C). Given the high efficiency of quenching Ir(III)* 4 by
DABCO (Φ > 0.99), the photoinduced reductive elimination
step implicated in mechanism C would likely proceed through
electron relay, wherein DABCO rather than Ir(III)* serves to
oxidize the Ni(II) center. Precise mechanistic details on the Ni
oxidation states implicated in these various potential pathways
are often challenging to obtain, but pioneering stoichiometric
2
3
q
•+
3
3
34
work by Hillhouse, and more recent studies by Zargarian
have demonstrated that while Ni(II)/Ni(0) reductive elimination
from Ni(II) amido complexes is slow at room temperature, their
Ni(III) oxidation state congeners undergo rapid C–N bond-for-
mation with the same thermal exposure. This stark contrast in
bond-formation rates suggests the Ni(III) oxidation state may
be responsible for promoting the otherwise challenging C–N
bond-formation step in the presently studied amination reac-
tion, especially considering the absence of a ligand designed for
Mechanism of C–N Reductive Elimination
After studying the primary photochemical decay pathway of
the excited-state iridium catalyst, and its subsequent role in re-
ductively initiating Ni catalysis, we turned our attention to un-
derstanding the precise nature of the Ni-catalyzed C–N bond-
forming step. In particular, three general mechanisms are often
invoked in Ni-catalyzed N– and O–arylation reactions: A) a
Ni(I)/Ni(III) cycle, which proceeds through Ni(I)/Ni(III) oxida-
tive addition with the aryl halide partner and a subsequent
Ni(III)/Ni(I) C–X (X = N, O) bond-forming reductive elimina-
tion (Figure 6A), B) a Ni(0)/Ni(II) cycle, wherein the analogous
Ni(0)/Ni(II) oxidative addition and Ni(II)/Ni(0) reductive elim-
35
facilitating C–N reductive elimination. In order to probe the
role of the Ni(II) or Ni(III) oxidation state in promoting C–N
reductive elimination, stoichiometric organometallic studies
with Ni complexes bearing catalytically-relevant ligand spheres
were undertaken. To this end, Ni(II) amine complexes 14 and
5 with two different amine coupling partners were prepared.
These Ni(II) complexes can be prepared in 3 steps or fewer
Scheme S2), and are indefinitely stable in the solid form.
1
inations
take
place
(Figure
6B),
or
C)
a
Ni(0)/Ni(II)/Ni(III)/Ni(I)/Ni(0) process, wherein Ni(0)/Ni(II)
(
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 12
+
Br–
0
.4 mol% photocatalyst 1
H
N
NiII (13)
H
N
2
2
H
N
2
ArBr (2)
Et
*IrIII (4)
IrII (5)
IrIII (1) + NiI/0
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Pr
Pr
Pr
propylamine (405 mM)
product (3)
NiII
hexylamine
N
H
DMF, 23 °C, 5 h
F3C
CF3
Φq ≈ 1
Φ
Ir ≤ 1
Φ
Ni
F
3
C
DABCO (486 mM)
CF
3
1
8 (13.5 mM)
16, 69% yielda
Φtot > 1
× ΦIr × ΦNi
Figure 7. Stoichiometric photoinduced reductive elimination
with Ni(II) complex 18. Yield determined by F NMR.
a
19
Φ
tot = Φ
ΦNi > 1
q
Moreover, they are stable toward C–N reductive elimination in
DMF solution both in the dark and under 450 nm irradiation
(Tables S7–8). Cyclic voltammetry was then employed to ex-
amine the oxidative stabilities of Ni(II) complexes 14 and 15:
both complexes display irreversible oxidation waves (E
0.70 V and +0.69 V vs. SCE for 14 and 15, respectively). This
electrochemical behavior begged the question of whether a
rapid intramolecular reductive elimination process after oxida-
tion at the electrode serves as the chemical basis for the irre-
versible nature of the two oxidation events (Figure 6D). To test
this hypothesis, Ni(II) complexes 14 and 15 were treated with a
Figure 8. Schematic representation of mechanistic components
of photoredox C–N coupling and associated quantum yields.
overall product formation (Φtot), Ir(III)* quenching to form
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Ir(II) (Φ
product-forming nickel-catalyzed process (ΦNi). In this rela-
tionship, the total quantum yield of product formation (Φtot
should be equivalent to the product of quantum yields for the
individual components of the catalytic mechanism (Φ , ΦIr, and
Ni, Figure 8). A key feature distinguishing mechanisms A and
q
), Ni catalyst initiation by Ir(II) (ΦIr), and that of the
i
=
+
)
q
Φ
C lies in the necessity for light input in mechanism C to achieve
catalyst turnover, whereas mechanism A requires no photonic
input beyond the initiation step. In terms of theoretical maxima
for quantum efficiencies of these processes (ΦNi), mechanism C
requires one photon per turnover, limiting its maximum quan-
tum yield to 1 (ΦNi ≤ 1), whereas mechanism A has no theoret-
ical limit in this regard. Based on this analysis, obtaining the
quantum yield of the Ni-catalyzed product-forming cycle
within the photoredox C–N coupling (ΦNi) could provide a
straightforward means for differentiating between the two gen-
eral mechanisms. In particular, a quantum yield for the nickel
cycle above 1 (ΦNi > 1) would indicate that mechanism C is not
the only operative mechanism, and a quantum yield signifi-
cantly surpassing unity would indicate that C can only be min-
imally operative.
4
mild single-electron oxidant (FcBF , ferrocenium tetrafluorob-
red
III
II
orate, E1/2 [Fe /Fe ] = +0.49 V vs. SCE). Under these condi-
tions, both 14 and 15 undergo efficient reductive elimination to
give the expected C–N coupled products (16 and 17) within
minutes (68% yield of 16 from 14 and 70% yield of 17 from 15,
Figure 6E). These results underscore the rapid nature with
which C–N bond-formation can take place under the appropri-
ate conditions, and the importance of accessing the Ni(III) oxi-
dation state in promoting this elementary organometallic step.
In the context of mechanisms A–C, a key feature distinguishing
mechanisms A and C from mechanism B lies in the details of
the reductive elimination process. Mechanism B relies on a
Ni(II)/Ni(0) reductive elimination to forge the C–N coupled
product, which is apparently a low efficiency step under these
conditions. In contrast, both mechanisms A and C remain as
plausible pathways, given the high efficiency of reductive elim-
ination from Ni(III).
In order to make an assessment of ΦNi based on the afore-
mentioned analysis, measurements and estimates for all quan-
tum yield values aside from ΦNi were carried out. We first rec-
ognized that Ir(III)* quenching efficiency by DABCO is suffi-
In considering the two mechanisms involving C–N reductive
elimination from Ni(III) (mechanisms A and C), we first eval-
uated mechanism C in a stoichiometric setting. In order to ac-
cess the Ni(III) oxidation state, mechanism C invokes a pho-
toinduced oxidation process from an intermediate Ni(II) com-
plex (Figure 6C). To assess the viability of this photoinduced
reductive elimination step, Ni(II) complex 14 was exposed to
photocatalyst 1, 450 nm irradiation, and catalytically-relevant
concentrations of amine and DABCO. At the high amine con-
centration relevant to the catalytic conditions (405 mM), nickel
complex 14 is present predominantly in its cationic tris(amino)
aryl bromide form, 18 (Figure S42). Under these conditions,
Ni(II) complex 18 undergoes efficient C–N coupling to give
aniline 16 (Figure 7, 69% yield), and control experiments reveal
that amine association does not dramatically affect this reduc-
tive elimination efficiency (Tables S7 and S10). In total, this
result clearly demonstrates that photocatalyst 1 is capable of in-
ducing C–N bond-formation from an otherwise stable Ni(II)
aryl amine complex, a step which is crucial to the viability of
mechanism C (Figure 6C).
q
ciently high to be disregarded in the calculation (Φ > 0.99,
Figure 2A). While a quantitative evaluation of ΦIr cannot be
made based on the lack of observable Ni(II) reduction by pump-
probe measurements (Figure 3E, Figure S46B), this value nec-
essarily has an upper bound of 1 and is likely considerably
lower, based on the absence of quenching in comparing the
transient absorption decay kinetics of Ir(II) 5 with and without
5 mol% NiBr
2
•3H
2
O
H
Br
N
0
.02 mol% photocatalyst 1
hexyl
H
2
N
hexyl
DMF, 23 °C
R
R
DABCO (1.8 equiv)
aryl bromide
hexylamine
coupled product
CN
CF3
Cl
H
Unfortunately, based on stoichiometric reductive elimination
experiments alone, mechanisms A and C still cannot be distin-
guished. Instead, a closer examination of the quantum efficien-
cies of C–N cross-coupling, and those of the component mech-
anistic steps leading to the final product enables discrimination
between the two competing mechanisms. In more detail, a rela-
tionship can be constructed between the quantum yields of
Me
Figure 9. Dependence of the rate of C–N coupling on the aryl
bromide para-substituent Hammett electronic parameter.
ACS Paragon Plus Environment

Page 7 of 12
Journal of the American Chemical Society
A
Ar–Br
*
IrIII + DABCO
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
4
Ni(II) reduction by Ir(II) is inefficient,
and rate-determining
dark
Ni(II) 13
DABCO•+ + IrII
IrIII + NiI
NiIII
1
9
cycle
5
1
20
Ni(I)/Ni(III) oxidative addition is slow
multiple pathways for Ni(I) deactivation
hν
Ar–NHR
IrIII + DABCO
1
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
B
C
off
light off: thermal deactivation
light on: re-activation
off on
I
DABCO•+
ArBr
II
I
Ni Br
+
+
+
DABCO + Ni Br2
Ni Br
off on
I
II
I
Ni Br
Ar• + Ni Br2
Ni Br
I
NiIII
II
I
Ni Br
2Ni Br2
Ni Br
on
Figure 10. A. Overview of photochemical and organometallic mechanisms underlying photoredox C–N coupling with rate-determin-
ing steps highlighted (orange and blue). B. Plot of product concentration vs. time with and without irradiation where indicated (440
nm, photoNMR). C. Proposed mechanisms for oxidative deactivation of active Ni(I) catalyst. See Figure S48 for further details.
Ni(II) (Figure S46B). While in the case of mechanism A single
measured for a variety of electronically diverse 4-substituted
aryl bromides. A Hammett correlation could be constructed
from these measurements, giving a strong positive correlation
(ρ = +4.5), indicating that during the Ni(I)/Ni(III) oxidative ad-
dition transition state, there is significant negative charge
buildup on the electrophilic aryl bromide, and that this oxida-
tive addition step plays a significant role in determining the
electron transfer between Ir(II) and Ni(II) would directly pro-
36
duce Ni(I), subsequent disproportionation or further reduction
would be required to access Ni(0) in mechanism C. This added
mechanistic complexity makes the upper bound for ΦIr in mech-
anism C lower than the corresponding value for mechanism A,
but does not affect the final analysis. Finally, Φtot can be meas-
ured through standard reaction kinetics measurements and
chemical actinometry (Φtot up to 1.1 for hexylamine coupling,
Tables S3–S4, and 5.4 for pyrrolidine coupling, Table S4). Tak-
ing these values together reveals that ΦNi is greater than unity
3
7
overall rate of C–N coupling (Figure 9). This finding is con-
sistent with a mechanistic scenario in which the active Ni(I)
species can either undergo oxidative addition, or a competitive
•
+
oxidative deactivation process, such as oxidation by DABCO ,
or comproportionation with Ni(III), either of which would con-
vert this Ni(I) species back into the observed Ni(II) resting state,
(ΦNi ≥ 1.1 for hexylamine and ΦNi ≥ 5.4 for pyrrolidine). More-
over, the vanishingly low value for ΦIr indicated by the Ni(II)-
quenched transient absorption spectroscopy measurements sug-
gests that ΦNi significantly passes the theoretical limit for a one-
photon-per-turnover mechanism. This result not only rules out
mechanism C as the only mechanism, but makes clear that this
mechanism can only be minimally responsible for product for-
mation. Moreover, this highly efficient nickel-catalyzed prod-
uct formation pathway is consistent with mechanism A, which
requires no photonic input for turnover. More specifically, this
finding indicates that while inefficient Ni(II) reduction by Ir(II)
hampers formation of high concentrations of active Ni(I), the
activated nickel catalyst is able to undergo multiple turnovers
to arrive at high overall total quantum efficiencies (Φtot > 1).
p
13. Within this competition, increasing σ values for the aryl
bromide partner would favor the desired oxidative addition
pathway, thus resulting in higher quantum yields of product for-
mation (Table S4).
Catalyst Development Studies
Taken together, these spectroscopic investigations, stoichio-
metric experiments, and kinetic studies demonstrate the photo-
chemical mechanism for Ni co-catalyst activation by Ir(III)
photocatalyst 1 and further support a Ni(I)/Ni(III) cycle as be-
ing responsible for the ensuing C–N bond-formation step (Fig-
ure 10A). In more detail, photocatalyst 1 accesses an oxidizing
red
III* II
excited-state under 450 nm irradiation (E1/2 [Ir /Ir ] = +1.08
Role of C–Br Oxidative Addition
V vs. SCE), which subsequently undergoes rapid exergonic
electron transfer quenching with DABCO (kET = 1.34 × 10 s ,
ΔG = –0.37 eV), giving rise to long-lived (>100 µs) Ir(II) state
5 and DABCO (Figure 10A). While this Ir(II) intermediate
predominantly undergoes a highly exergonic charge recombi-
9
–1
While the aforementioned spectroscopic and kinetic evi-
dence demonstrates the importance of Ni(II) reduction by Ir(II)
intermediate 5 in promoting the C–N cross-coupling, and stoi-
chiometric reductive elimination experiments suggest the C–N
bond-forming step is efficient and rapid, it is unclear from these
studies what role the oxidative addition step plays in determin-
ing the reaction efficiency. To better understand the impact of
this elementary organometallic step on the efficiency of C–N
coupling, we examined the influence of aryl halide electronics
on the reaction rate. Initial rates of product formation were
•
+
•+
nation with DABCO (ΔG = –2.12 eV), its reduction of Ni(II)
catalyst 13 (ΔG ≈ 0 eV), albeit with low efficiency, initiates a
Ni(I)/Ni(III) cycle responsible for C–N bond-formation be-
tween the amine and aryl halide coupling partners. In particular,
Ni(I) species 19 (produced through reduction by Ir(II)) can un-
dergo C–Br oxidative addition with aryl bromide 2 and C–N
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 12
red
III* II
A
+
–
+
–
(E1/2 [Ir /Ir ] = +1.05 to +0.73 V vs. SCE, Table S2). In ac-
cordance with the overall mechanistic hypothesis, iridium pho-
tocatalysts with lower reduction potentials, and hence higher
driving forces for Ni(II) reduction, should give rise to higher
rates of product formation provided that these catalysts are sim-
ultaneously sufficiently oxidizing to access their Ir(II) forms
through DABCO oxidation. In such a rate-driving force rela-
tionship, increased reaction rate would be expected to accom-
pany reducing power increases until a curve flattening as the
diffusion limit for electron transfer is reached. Indeed, these
new catalysts gave rise to higher levels of reaction efficiency,
culminating in a 37-fold increase in initial rate and >10-fold in-
crease in the quantum yield of product formation with catalyst
22 when compared to photocatalyst 21, originally reported for
the photoredox C–N coupling (Figure 11A–C). Moreover, pho-
tocatalyst 22 provides a significantly improved full reaction
course profile (Figure S18). These studies provide guiding prin-
ciples for photocatalyst design directed towards both initiating
and perpetuating Ni(I)/Ni(III) cross-coupling activity.
F
3
C
PF
6
PF
6
F
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Me
tBu
N
Ir
H
H
N
2
N
Ir
N
N
N
F
F
N
tBu
N
N
2
N
Me
F
3
F C
2
1 (original catalyst)
22 (improved catalyst)
E1/2red = –1.23 V
E1/2red = –1.91 V
B
C
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
2
Φ(21) = 0.072
Φ(22) = 1.1
2
1
kobs(22)
=
37
kobs(21)
Having explored the mechanistic underpinnings of the pho-
toredox aryl amination and developed a higher efficiency pho-
tocatalyst for this cross-coupling reaction, we next turned our
attention to examining the relevance of these mechanistic in-
sights to end-user medicinal chemistry applications of C–N
cross-coupling. The importance of C–N cross-coupling to me-
dicinal chemistry applications in particular has motivated
mechanistic studies on this class of reactions historically with
Figure 11. A. Structures and reduction potentials for originally
reported catalyst 21 and improved catalyst 22. B. Plot of ln(kobs
)
vs. Ir(III)/Ir(II) reduction potential over a large driving force
range (680 meV) including new photocatalysts. C. Comparison
of C–N coupling performances of catalysts 21 and 22 with 4-
bromobenzotrifluoride and hexylamine coupling partners. Error
bars represent standard error of the mean (n = 2).
3
9
bond-forming reductive elimination with bound hexylamine to
furnish the coupled product 3. In this process, the reductive
elimination step is rapid, whereas oxidative addition is slow and
impacts the overall rate of coupling.
Pd-catalyzed methods, and more recently in the context of
40,41
nickel catalysis.
Based on this, it became clear that for the
mechanistic insights gained from this study to have the most
relevance, their applicability in complex, drug-like settings
would need to be demonstrated. More specifically, the presence
of numerous heteroatoms, sensitive functional groups, and
other metal-coordinating sites in pharmaceutical compounds
and their synthetic precursors requires that a robust cross-cou-
pling method carry out the desired bond-forming process with-
out interference from a diverse set of competing functionalities.
Taking this consideration together with our catalyst optimiza-
tion studies makes it clear that while photocatalyst 22 operates
with a higher rate and quantum efficiency than its predecessor
with a set of model substrates (Figure 11), its generality would
be of crucial importance in determining its superiority as a pho-
tocatalyst for C–N cross-coupling.
Having determined the role of the photocatalyst in initiating
the nickel-catalyzed cross-coupling process, we questioned
whether the resulting Ni(I)/Ni(III) cycle could perpetuate con-
tinuously without further irradiation, given that no photonic in-
put is required for Ni catalyst turnover (Figure 10A). To probe
this hypothesis, the light dependence of the reaction was exam-
ined. Product formation rapidly ceases in the absence of light,
indicating that the Ni(I)/Ni(III) cycle does not continuously per-
petuate under these conditions, and instead deactivates rapidly
(< 30 seconds) (Figure 10B). This finding is consistent with the
vanishingly low level of Ni(I) buildup (below the detection
limit) observed by steady-state UV/vis spectroscopy during the
course of the reaction (Figure 5). While the deactivation mech-
anisms at play in the Ni(I)/Ni(III) cycle are not entirely clear,
electron recombination between the Ni(I) catalyst and
Over the past 5 years, a set of 18 particularly challenging
aryl halide cross-coupling partners constructed by Dreher and
coworkers (referred to as an informer library) has emerged as a
standardized means for benchmarking the generality of new
•
+
DABCO , halogen atom abstraction from aryl halide 2 by Ni(I)
38
42
1
9, and Ni(I)/Ni(III) comproportionation would all return the
Ni(II)Br resting state 13 observed spectroscopically (Figure
0C).
Recognizing that reduction of Ni(II) complex 13 is crucial
metal-catalyzed cross-coupling protocols. This set of aryl hal-
ide substrates was specifically constructed based on the molec-
ular complexities of the cross-coupling partners, including
many potential metal-coordination sites and reactive function-
alities (Figure S43). This effort resulted in a set of aryl halides
with molecular complexity mirroring that of drug substances
and intermediates, making this evaluation technique particu-
larly relevant to medicinal chemistry applications. Given these
properties, we recognized that this benchmarking technique
would be an apt method for evaluating the generality of photo-
catalyst 22. To this end, C–N coupling reactions with each
member of the aryl halide informer library were carried out with
photocatalyst 22. Gratifyingly, the desired C–N-coupled prod-
ucts are obtained with the vast majority of aryl halide library
members under these photocatalytic conditions (15 of 18 aryl
halides). While a number of these aryl halides give rise to only
2
1
not only to initiate the nickel-catalyzed cross-coupling mecha-
nism, but also in perpetuating this process, we endeavored to
increase the efficiency of Ni(II) reduction through photocatalyst
design (Figure 11). More specifically, Ir(III) photocatalysts of
increasing electron density on the π*-localized bipyridine lig-
and were prepared, and their photophysical and electrochemical
properties characterized. These new photocatalysts exhibit the
red
III II
expected increase in reducing power (E1/2 [Ir /Ir ] = –1.52 to
1.91 V vs. SCE, Table S2), while still maintaining sufficient
oxidizing powers to render quenching by DABCO exergonic
–
ACS Paragon Plus Environment

Page 9 of 12
Journal of the American Chemical Society
Br
1
2
3
4
5
6
7
8
9
CO2Me
N
OMe
N
CO2Et
O
O
N
N
O
N
N
CO Bn
O
2
Et
N
N
N
N
O
N
NH
F
I
N
OH
Br
N
O
Me
O
Br
OH
X2
X8
X12
X14
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13 X14 X15 X16 X17 X18 Hits
21, 450 nm LEDs
1
4
2
2, 450 nm LEDs
15
– satisfies hit criteria; useful
lead result for optimization
0
%
1–15%
16–40%
> 40%
Figure 12. Product formation levels for the cross-coupling reaction of piperidine with the 18 aryl halide informer library members
under light-mediated photoredox conditions with photocatalysts 21 and 22, and structures of selected library members shown above.
low levels of product formation (1–15% yield, Figure S43) un-
der the action of photocatalyst 22, these results serve as prom-
ising leads for further optimization, given that medicinal chem-
istry campaigns often originate from low, but non-zero levels
of product formation. In comparing the performance of catalyst
more broadly provides an opportunity to systematically tune the
electrochemical and photophysical properties of the photocata-
lyst for a given C–N cross-coupling reaction as demonstrated in
our photocatalyst SAR studies (Figure 11). This feature serves
as a point of modularity, akin to the use of ligand modulation
which has rendered Pd-catalyzed C–N coupling so robust in
medicinally relevant contexts.
2
2 to that of the originally reported catalyst 21, the redesigned
photocatalyst 22 gives broadly similar levels of reactivity
across the informer library, but with modestly expanded gener-
ality (X6 can be coupled with catalyst 22, but not 21, Figure
12). In total, improved photocatalyst 22 exhibits a dramatically
improved reaction rate and quantum efficiency without sacri-
ficing the generality of the originally reported photoredox C–N
coupling protocol. Within the context of larger scale applica-
tions of C–N coupling, a recent report by Corcoran et al. on the
kilogram-scale application of this photoredox amination in flow
further highlights the importance of minimizing residence time
and energy input by maximizing reaction rate and quantum
Conclusion
The use of steady-state and time-resolved spectroscopic tech-
niques has long provided a powerful approach to elucidating
photophysical and photochemical reaction mechanisms. At the
same time, traditional kinetic and stoichiometric experiments
are often employed to gain important mechanistic insight into
synthetically important transition metal-catalyzed cross-cou-
pling reactions. In this study, their combined use has proven
particularly advantageous, wherein this multi-faceted approach
has helped uncover the photochemical and organometallic
mechanisms underlying the nickel/photoredox dual-catalytic
amination of aryl bromides. Of particular relevance to the prac-
ticing organic chemist, these studies have demonstrated the role
of photocatalysis in this C–N cross-coupling reaction: not only
does the photocatalyst initiate the reaction through Ni(II) reduc-
tion, but its presence ensures that reactivity perpetuates through
continually reducing the resting-state Ni(II) catalyst to its ac-
tive, low-valent form. Moreover, analysis of the overall quan-
tum yield of product formation and low quantum yield of initi-
ation reveals that the active Ni catalyst in this system operates
at high efficiency. These studies have further pinpointed the in-
efficient catalytic steps present in this C–N cross-coupling re-
action, knowledge which proved indispensable in designing
new photocatalysts capable of superior performances in a syn-
thetically meaningful context. Overall, this study may provide
a roadmap to mechanistic analysis and catalyst improvement in
the context of dual-catalytic photoredox cross-coupling reac-
tions.
24
yield, respectively .
While the precise mechanistic details underlying the high
generality of the photoredox C–N cross-coupling are not en-
tirely clear at present, light-mediated formation of the active
Ni(I) catalyst rather than stoichiometric reduction to low-valent
nickel appears to enable high C–N cross-coupling activity
across a diverse set of aryl halide partners. Mechanistically, the
use of an appropriate photocatalyst and visible light irradiation
may provide a route for continual generation of a highly active
Ni(I) catalyst capable of carrying out challenging oxidative ad-
dition steps, even if this catalyst often returns to its Ni(II) rest-
ing state 13 through the aforementioned deactivation mecha-
nisms (Figure 10C). More specifically, maintaining a low con-
centration of active Ni(I) may disfavor bimolecular dimeriza-
tion and further aggregation to form nickel-black, a process
which has recently been shown to detrimentally affect nickel-
4
3
catalyzed C–N cross-coupling efficiency. By avoiding the for-
mation of unreactive nickel-black, this photocatalytic mode of
catalyst activation generates a highly active low-valent nickel
complex capable of performing challenging oxidative addition
steps with relatively unreactive aryl halides and in the presence
of multiple coordinating functional groups (Figure 12). Beyond
these specific mechanistic features, the use of photocatalysis
ASSOCIATED CONTENT
Supporting Information
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 12
The Supporting Information is available free of charge on the ACS
(8) Clarke, M. L.; Frew, J. J. R. Ligand Electronic Effects in
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Publications website.
Homogeneous Catalysis Using Transition Metal Com-
plexes of Phosphine Ligands. In Organometallic
Crystallographic Information (CIF)
Experimental procedures, characterization data, and spectroscopic
data (PDF)
Chemistry;
2009;
pp
19–46.
https://doi.org/10.1039/B801377M.
(
(
9) Nielsen, M. C.; Bonney, K. J.; Schoenebeck, F. Compu-
tational Ligand Design for the Reductive Elimination
of ArCF3 from a Small Bite Angle PdII Complex: Re-
markable Effect of a Perfluoroalkyl Phosphine. An-
gewandte Chemie International Edition 2014, 53 (23),
AUTHOR INFORMATION
Corresponding Author
*dmacmill@princeton.edu
5
903–5906. https://doi.org/10.1002/anie.201400837.
Notes
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
10) Ruiz-Castillo, P.; Blackmond, D. G.; Buchwald, S. L.
Rational Ligand Design for the Arylation of Hindered
Primary Amines Guided by Reaction Progress Kinetic
Analysis. J. Am. Chem. Soc. 2015, 137 (8), 3085–3092.
https://doi.org/10.1021/ja512903g.
The authors declare no competing financial interests.
ACKNOWLEDGMENT
This work was supported as part of BioLEC, an Energy Frontier
Research Center funded by the U.S. Department of Energy, Office
of Science under Award # DE-SC0019370 and kind gifts from
Merck, BMS, Pfizer, Janssen, Genentech, and Eli Lilly. We
acknowledge the Princeton Catalysis Initiative for supporting this
work. N.A.T. and L.T. acknowledge Princeton University, E. Tay-
lor and the Taylor family for an Edward C. Taylor Fellowship. We
thank Phil Jeffrey of Princeton University for X-ray crystallo-
graphic structure determination. We thank Megan H. Shaw and
Emily B. Corcoran for their help in carrying out initial studies.
(
(
11) Hartwig, J. F. Palladium-Catalyzed Amination of Aryl
Halides: Mechanism and Rational Catalyst Design.
Synlett
1997,
1997
(4),
329–340.
https://doi.org/10.1055/s-1997-789.
12) Olsen, E. P. K.; Arrechea, P. L.; Buchwald, S. L. Mech-
anistic Insight Leads to a Ligand Which Facilitates the
Palladium-Catalyzed Formation of 2-(Hetero)Aryla-
minooxazoles and 4-(Hetero)Arylaminothiazoles. An-
gewandte Chemie International Edition 2017, 56 (35),
10569–10572.
REFERENCES
https://doi.org/10.1002/anie.201705525.
(
13) Klinkenberg, J. L.; Hartwig, J. F. Slow Reductive Elim-
ination from Arylpalladium Parent Amido Complexes.
J. Am. Chem. Soc. 2010, 132 (34), 11830–11833.
https://doi.org/10.1021/ja1023404.
(14) Dennis, J. M.; White, N. A.; Liu, R. Y.; Buchwald, S. L.
Breaking the Base Barrier: An Electron-Deficient Pal-
ladium Catalyst Enables the Use of a Common Soluble
Base in C–N Coupling. J. Am. Chem. Soc. 2018, 140
(
(
1) Miyaura, Norio.; Suzuki, Akira. Palladium-Catalyzed
Cross-Coupling Reactions of Organoboron Com-
pounds. Chem. Rev. 1995, 95 (7), 2457–2483.
https://doi.org/10.1021/cr00039a007.
2) JohanssonꢀSeechurn, C. C. C.; Kitching, M. O.; Colacot,
T. J.; Snieckus, V. Palladium-Catalyzed Cross-Cou-
pling: A Historical Contextual Perspective to the 2010
Nobel Prize. Angewandte Chemie International Edition
(13),
4721–4725.
2012,
51
(21),
5062–5085.
https://doi.org/10.1021/jacs.8b01696.
https://doi.org/10.1002/anie.201107017.
(15) Dennis, J. M.; White, N. A.; Liu, R. Y.; Buchwald, S. L.
Pd-Catalyzed C–N Coupling Reactions Facilitated by
Organic Bases: Mechanistic Investigation Leads to En-
hanced Reactivity in the Arylation of Weakly Binding
Amines. ACS Catal. 2019, 9, 3822–3830.
https://doi.org/10.1021/acscatal.9b00981.
(16) Tellis, J. C.; Primer, D. N.; Molander, G. A. Single-Elec-
tron Transmetalation in Organoboron Cross-Coupling
by Photoredox/Nickel Dual Catalysis. Science 2014,
345 (6195), 433–436. https://doi.org/10.1126/sci-
ence.1253647.
(
(
3) Torborg, C.; Beller, M. Recent Applications of Palla-
dium-Catalyzed Coupling Reactions in the Pharmaceu-
tical, Agrochemical, and Fine Chemical Industries. Ad-
vanced Synthesis & Catalysis 2009, 351 (18), 3027–
3
043. https://doi.org/10.1002/adsc.200900587.
4) Miura, M. Rational Ligand Design in Constructing Effi-
cient Catalyst Systems for Suzuki–Miyaura Coupling.
Angewandte Chemie International Edition 2004, 43
(
17),
https://doi.org/10.1002/anie.200301753.
(5) Lundgren, R. J.; Stradiotto, M. Addressing Challenges in
Palladium-Catalyzed Cross-Coupling Reactions
Through Ligand Design. Chemistry – A European
Journal 2012, 18 (32), 9758–9769.
https://doi.org/10.1002/chem.201201195.
2201–2203.
(17) Kainz, Q. M.; Matier, C. D.; Bartoszewicz, A.; Zultanski,
S. L.; Peters, J. C.; Fu, G. C. Asymmetric Copper-Cat-
alyzed C-N Cross-Couplings Induced by Visible Light.
Science
2016,
351
(6274),
681–684.
https://doi.org/10.1126/science.aad8313.
(
6) Bruno, N. C.; Tudge, M. T.; Buchwald, S. L. Design and
Preparation of New Palladium Precatalysts for C–C
and C–N Cross-Coupling Reactions. Chem. Sci. 2013,
(18) Terrett, J. A.; Cuthbertson, J. D.; Shurtleff, V. W.; Mac-
Millan, D. W. C. Switching on Elusive Organometallic
Mechanisms with Photoredox Catalysis. Nature 2015,
524 (7565), 330. https://doi.org/10.1038/nature14875.
(19) 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
4
(3), 916–920. https://doi.org/10.1039/C2SC20903A.
(
7) Joost, M.; Zeineddine, A.; Estévez, L.; Mallet−Ladeira,
S.; Miqueu, K.; Amgoune, A.; Bourissou, D. Facile Ox-
idative Addition of Aryl Iodides to Gold(I) by Ligand
Design: Bending Turns on Reactivity. J. Am. Chem.
Soc.
2014,
136
(42),
14654–14657.
(6323),
380–385.
https://doi.org/10.1126/sci-
https://doi.org/10.1021/ja506978c.
ence.aal2490.
ACS Paragon Plus Environment

Page 11 of 12
Journal of the American Chemical Society
(
20) Liang, Y.; Zhang, X.; MacMillan, D. W. C. Decarboxy- (32) Bour, J. R.; Camasso, N. M.; Sanford, M. S. Oxidation
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
lative Sp 3 C–N Coupling via Dual Copper and Photo-
redox Catalysis. Nature 2018, 559 (7712), 83–88.
https://doi.org/10.1038/s41586-018-0234-8.
of Ni(II) to Ni(IV) with Aryl Electrophiles Enables Ni-
Mediated Aryl–CF3 Coupling. J. Am. Chem. Soc.
2015,
137
(25),
8034–8037.
(
21) Le, C.; Chen, T. Q.; Liang, T.; Zhang, P.; MacMillan, D.
W. C. A Radical Approach to the Copper Oxidative
Addition Problem: Trifluoromethylation of Bro-
moarenes. Science 2018, 360 (6392), 1010–1014.
https://doi.org/10.1126/science.aat4133.
22) Kim, S.; Toste, F. D. Mechanism of Photoredox-Initiated
C–C and C–N Bond Formation by Arylation of
IPrAu(I)–CF3 and IPrAu(I)–Succinimide. J. Am.
Chem. Soc. 2019, 141 (10), 4308–4315.
https://doi.org/10.1021/jacs.8b11273.
https://doi.org/10.1021/jacs.5b04892.
(33) Koo, K.; Hillhouse, G. L. Carbon-Nitrogen Bond For-
mation by Reductive Elimination from Nickel(II) Am-
ido Alkyl Complexes. Organometallics 1995, 14 (9),
4421–4423. https://doi.org/10.1021/om00009a054.
(34) Cloutier, J.-P.; Zargarian, D. Functionalization of the
Aryl Moiety in the Pincer Complex (NCN)NiIIIBr2:
Insights on NiIII-Promoted Carbon–Heteroatom Cou-
pling. Organometallics 2018, 37 (9), 1446–1455.
https://doi.org/10.1021/acs.organomet.8b00103.
(
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(23) Corcoran, E. B.; Pirnot, M. T.; Lin, S.; Dreher, S. D.; Di-
Rocco, D. A.; Davies, I. W.; Buchwald, S. L.; MacMil-
lan, D. W. C. Aryl Amination Using Ligand-Free Ni(II)
Salts and Photoredox Catalysis. Science 2016, 353
(35) Lavoie, C. M.; Stradiotto, M. Bisphosphines: A Promi-
nent Ancillary Ligand Class for Application in Nickel-
Catalyzed C–N Cross-Coupling. ACS Catal. 2018, 8
(8),
7228–7250.
(
6296),
279–283.
https://doi.org/10.1126/sci-
https://doi.org/10.1021/acscatal.8b01879.
ence.aag0209.
(36) Beattie, D. D.; Lascoumettes, G.; Kennepohl, P.; Love,
J. A.; Schafer, L. L. Disproportionation Reactions of an
Organometallic Ni(I) Amidate Complex: Scope and
Mechanistic Investigations. Organometallics 2018, 37
(9), 1392–1399. https://doi.org/10.1021/acs.organ-
omet.8b00074.
(37) Hansch, Corwin.; Leo, A.; Taft, R. W. A Survey of Ham-
mett Substituent Constants and Resonance and Field
Parameters. Chem. Rev. 1991, 91 (2), 165–195.
https://doi.org/10.1021/cr00002a004.
(38) Tsou, T. T.; Kochi, J. K. Mechanism of Oxidative Addi-
tion. Reaction of Nickel(0) Complexes with Aromatic
Halides. J. Am. Chem. Soc. 1979, 101 (21), 6319–6332.
https://doi.org/10.1021/ja00515a028.
(39) Shekhar, S.; Ryberg, P.; Hartwig, J. F.; Mathew, J. S.;
Blackmond, D. G.; Strieter, E. R.; Buchwald, S. L.
Reevaluation of the Mechanism of the Amination of
Aryl Halides Catalyzed by BINAP-Ligated Palladium
Complexes. J. Am. Chem. Soc. 2006, 128 (11), 3584–
3591. https://doi.org/10.1021/ja045533c.
(40) Ge, S.; Green, R. A.; Hartwig, J. F. Controlling First-
Row Catalysts: Amination of Aryl and Heteroaryl
Chlorides and Bromides with Primary Aliphatic
Amines Catalyzed by a BINAP-Ligated Single-Com-
ponent Ni(0) Complex. J. Am. Chem. Soc. 2014, 136
(4), 1617–1627. https://doi.org/10.1021/ja411911s.
(41) Sun, R.; Qin, Y.; Nocera, D. G. General Paradigm in
Photoredox Nickel-Catalyzed Cross-Coupling Allows
for Light-Free Access to Reactivity. Angewandte
Chemie International Edition 2020, 59 (24), 9527–
9533. https://doi.org/10.1002/anie.201916398.
(42) Kutchukian, P. S.; Dropinski, J. F.; Dykstra, K. D.; Li,
B.; DiRocco, D. A.; Streckfuss, E. C.; Campeau, L.-C.;
Cernak, T.; Vachal, P.; Davies, I. W.; Krska, S. W.;
Dreher, S. D. Chemistry Informer Libraries: A
Chemoinformatics Enabled Approach to Evaluate and
Advance Synthetic Methods. Chem. Sci. 2016, 7 (4),
2604–2613. https://doi.org/10.1039/C5SC04751J.
(43) Gisbertz, S.; Reischauer, S.; Pieber, B. Overcoming Lim-
itations in Dual Photoredox/Nickel-Catalysed C–N
Cross-Couplings Due to Catalyst Deactivation. Nature
Catalysis 2020, 1–10. https://doi.org/10.1038/s41929-
020-0473-6.
(
24) Corcoran, E. B.; McMullen, J. P.; Lévesque, F.; Wismer,
M. K.; Naber, J. R. Photon Equivalents as a Parameter
for Scaling Photoredox Reactions in Flow: Translation
of Photocatalytic C−N Cross-Coupling from Lab Scale
to Multikilogram Scale. Angewandte Chemie Interna-
tional Edition 2020, 59 (29), 11964–11968.
https://doi.org/10.1002/anie.201915412.
25) Lowry, M. S.; Goldsmith, J. I.; Slinker, J. D.; Rohl, R.;
Pascal, R. A.; Malliaras, G. G.; Bernhard, S. Single-
Layer Electroluminescent Devices and Photoinduced
Hydrogen Production from an Ionic Iridium(III) Com-
plex. Chem. Mater. 2005, 17 (23), 5712–5719.
https://doi.org/10.1021/cm051312+.
26) Arias-Rotondo, D. M.; McCusker, J. K. The Photophys-
ics of Photoredox Catalysis: A Roadmap for Catalyst
Design. Chem. Soc. Rev. 2016, 45 (21), 5803–5820.
https://doi.org/10.1039/C6CS00526H.
(27) Tian, L.; Till, N. A.; Kudisch, B.; MacMillan, D. W. C.;
Scholes, G. D. Transient Absorption Spectroscopy Of-
fers Mechanistic Insights for an Iridium/Nickel-Cata-
lyzed C–O Coupling. J. Am. Chem. Soc. 2020, 142
(
(
(
10),
4555–4559.
https://doi.org/10.1021/jacs.9b12835.
(
28) Kudisch, M.; Lim, C.-H.; Thordarson, P.; Miyake, G. M.
Energy Transfer to Ni-Amine Complexes in Dual Cat-
alytic, Light-Driven C–N Cross-Coupling Reactions. J.
Am. Chem. Soc. 2019, 141 (49), 19479–19486.
https://doi.org/10.1021/jacs.9b11049.
(29) Sun, R.; Qin, Y.; Ruccolo, S.; Schnedermann, C.;
Costentin, C.; Daniel G. Nocera. Elucidation of a Re-
dox-Mediated Reaction Cycle for Nickel-Catalyzed
Cross Coupling. J. Am. Chem. Soc. 2019, 141 (1), 89–
9
3. https://doi.org/10.1021/jacs.8b11262.
(
30) Shields, B. J.; Kudisch, B.; Scholes, G. D.; Doyle, A. G.
Long-Lived Charge-Transfer States of Nickel(II) Aryl
Halide Complexes Facilitate Bimolecular Photoin-
duced Electron Transfer. J. Am. Chem. Soc. 2018, 140
(
8), 3035–3039. https://doi.org/10.1021/jacs.7b13281.
(
31) Balakrishnan, G.; Keszthelyi, T.; Wilbrandt, R.; Zwier,
J. M.; Brouwer, A. M.; Buma, W. J. The Radical Cation
and Lowest Rydberg States of 1,4-Diaza[2.2.2]Bicy-
clooctane (DABCO). J. Phys. Chem. A 2000, 104 (9),
1
834–1841. https://doi.org/10.1021/jp993052n.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 12
mechanistic study
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
IrIII + DABCO
*
NiII
dark
DABCO•+ + IrII
NiI
NiIII
hν
cycle
IrIII + DABCO
catalyst design
+
–
6
PF
Me
H
2
N
N
N
Ir
> 30-fold rate increase
N
N
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
>
10-fold Φ increase
N
H
2
Me
1
2
ACS Paragon Plus Environment