Synthesis and Characterization of Strong Cyclometalated Iridium Photoreductants for Application in Photocatalytic Aryl Bromide Hydrodebromination

Article abstract of DOI:10.1021/acscatal.9b02759

A series of potent bis-cyclometalated iridium photoreductants with electron-rich β-diketiminate (NacNac) ancillary ligands is described. Structure-property analysis reveals that substituent modification of the NacNac ligands has a large effect on the ground-state Ir^IV/Ir^III potential, which shifts cathodically as the NacNac is made more electron-rich. In addition, the excited-state Ir^IV/?Ir^III potentials are ca. 300-500 mV more negative than that of fac-Ir(ppy)₃ (ppy = 2-phenylpyridine), indicating that these compounds have much more reducing excited states. Rate constants for excited-state electron transfer between these photosensitizers and benzophenone are -2-3 times faster than fac-Ir(ppy)₃, demonstrating that these complexes are both kinetically and thermodynamically more potent for excited-state electron transfer. We use these photosensitizers to optimize a simple reaction procedure for photocatalytic debromination of aryl bromide substrates, which requires only the photosensitizer, blue light, and an amine base, without silanes or other additives that are used in previously reported methods.

Full text of DOI:10.1021/acscatal.9b02759

Subscriber access provided by RUTGERS UNIVERSITY
Article
Synthesis and Characterization of Strong Cyclometalated
Iridium Photoreductants for Application in
Photocatalytic Aryl Bromide Hydrodebromination
Jong-Hwa Shon, Steven Sittel, and Thomas S Teets
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.9b02759 • Publication Date (Web): 07 Aug 2019
Downloaded from pubs.acs.org on August 8, 2019
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 14
ACS Catalysis
1
2
3
4
5
6
Synthesis and Characterization of Strong Cyclometalated Iridium
Photoreductants for Application in Photocatalytic Aryl Bromide
Hydrodebromination
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
Jong-Hwa Shon, Steven Sittel, and Thomas S. Teets*
University of Houston, Department of Chemistry, 3585 Cullen Blvd., Room 112, Houston, TX 77204-5003, USA.
KEYWORDS. Photoredox catalysis; iridium; photochemistry; hydrodehalogenation; electron transfer; quenching
ABSTRACT: A series of potent bis-cyclometalated iridium photoreductants with electron-rich β-diketiminate (NacNac) ancillary ligands is
IV
III
described. Structure-property analysis reveals that substituent modification of the NacNac ligands has a large effect on the ground-state Ir /Ir
IV
III
potential, which shifts cathodically as the NacNac is made more electron rich. In addition, the excited-state Ir /*Ir potentials are ca. 300–500
mV more negative than that of fac-Ir(ppy) (ppy = 2-phenylpyridine), indicating that these compounds have much more reducing excited
states. Rate constants for excited-state electron transfer between these photosensitizers and benzophenone are ~2–3 times faster than fac-
Ir(ppy) , demonstrating these complexes are both kinetically and thermodynamically more potent for excited-state electron transfer. We use
3
3
these photosensitizers to optimize a simple reaction procedure for photocatalytic debromination of aryl bromide substrates, which requires
only the photosensitizer, blue light, and an amine base, without silanes or other additives that are used in previously reported methods.
Introduction
the poor photostability is detrimental to catalyst performance. More
recently, researchers have recognized that cyclometalated iridium
complexes are a class of photocatalysts more reactive and more
Visible-light photosensitizers, which can be organic molecules or
metal-based coordination compounds, absorb light energy and
convert it to a chemical potential via charge separation. The excited-
state electron-hole pair render the excited state both more oxidizing
and more reducing than the ground state, enabling rich electron-
transfer chemistry. Excited-state charge transport is important in
2
+
stable than [Ru(bpy)
prepared an iridium complex, [Ir(dF(CF
3
] derivatives. Bernhard and coworkers have
+
3
)ppy) (dtbbpy)]
2
(df(CF
3
)ppy = 2-(2,4-difluorophenyl)-5-trifluoromethylpyridine;
dtbbpy = 4,4ʹ-di-tert-butyl-2,2ʹ-bipyridine), which has a similarly
1
–3
4
reducing and significantly more oxidizing excited state than
renewable energy and catalysis applications, which span from
small-molecule activation reactions and organic synthesis
2
+ 12
[
Ru(bpy) ] . This complex is a versatile photoredox catalyst
3
4
–9
10,11
currently used by many other groups for a variety of
applications to the synthesis or degradation of large polymers.
4
transformations. Neutral cyclometalated iridium complexes, in
One of the most attractive features of photocatalysis in synthetic
applications, often referred to as “photoredox catalysis,” is that
generation of the reactive excited-state requires only input of light
energy under otherwise ambient reaction conditions. As such,
highly-reactive intermediates can be generated in a controlled
fashion, and problems associated with incompatibility of redox
reagents and temperature-induced decomposition can be
circumvented.
particular the archetypal compound fac-Ir(ppy) (ppy = 2-
3
phenylpyridine), have excited states that are more reducing than
2
+
13
[Ru(bpy)
3
]
by almost 1 V. As a result, these compounds have
been especially prominent for reductive transformations, allowing a
larger substrate scope for reactions involving C–X (X = halogen,
oxygen) cleavage. This principle is best exemplified by work from
the Stephenson group, showing effective hydrodeiodination of a
range of substrates using a combination of the photoreductant fac-
The “ambidextrous” nature of photosensitizers, i.e. the
phenomenon that they are both stronger oxidants and reductants in
their excited state, is best exemplified in [Ru(bpy)
bipyridine), the most well-studied metal-based photosensitizer. The
excited-state redox potentials for [Ru(bpy)
1
4
Ir(ppy) and sacrificial electron/hydrogen donors. In spite of these
3
successes, it remains a challenge to perform visible-light photoredox
catalysis on unactivated organobromide or organochloride
substrates, which are at least 200 mV more difficult to reduce than
2
+
3
] (bpy = 2,2ʹ-
2
+
III
II
3
] are E(Ru /*Ru ) =
1
5
their iodide counterparts. There have been some recent successes
+
−1.2 V vs ferrocenium/ferrocene (Fc /Fc) (−0.8 V vs. SCE) and
in engaging organobromide substrates in photoredox catalysis,
although these do require an excess of silanol or silane additive.
II
I
+
E(*Ru /Ru ) = 0.4 V vs Fc /Fc (0 V vs. SCE), allowing electron-
transfer reactions to/from the redox partner. These pathways are
respectively known as oxidative or reductive quenching, and both
can be operative in catalytic transformations. Although there are
1
6,17
As expansive as the field of photoredox catalysis has become, there
is comparatively little effort in catalyst development. Our group
recognized that possibility that new types of photoredox catalysts,
2
+
many examples of photoredox reactions using [Ru(bpy) ] as the
3
with even more reducing excited states than fac-Ir(ppy) , could
3
catalyst, there are some limitations of this widely used sensitizer. Its
modest reactivity, especially for reductive transformations, limits the
types of transformations and the number of viable substrates, and
facilitate development of photoredox reactions on a wider range of
substrates and with simpler reaction conditions. Along these lines,
ACS Paragon Plus Environment

ACS Catalysis
our group has recently reported three different complexes of the type dimers [Ir(C^N)
Page 2 of 14
(C^N= ppy, tbppy, and ppz). The
2
(μ-Cl)]
2
R
R
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
reaction of alkali metal NacNac salts with the cyclometalated iridium
dimer is complete in a couple of hours at room temperature, and the
products are easily separated by precipitation using diethyl ether or
n-hexane, without any column chromatography required.
Ir(ppy)
2
(NacNac ), where NacNac is a N, Nʹ-diphenyl-β-
diketiminate ligand with the 2- and 4-positions of the NacNac
backbone varied (R = Me, NMe , OEt). We showed in this work
that these complexes, by virtue of the electron-rich NacNac ligands
which destabilize the HOMO and shift the ground-state Ir /Ir
1
8
2
IV
III
potentials to more negative values, are more potent as
photoreductants by ~300–400 mV when compared to fac-Ir(ppy)
3
.
This work also showed us that the redox properties are strongly
dependent on the NacNac substitution, motivating us to pursue
substituent modifications at other positions of the NacNac
framework to better understand structure-property relationships in
this series of compounds and pursue improved next-generation
photoreductants. In this study, we present a set of 10 well-
characterized bis-cyclometalated iridium photoreductants with
NacNac ancillary ligands, adding to the three already introduced by
our group. Modifications to the N and Nʹ substituents of the
NacNac, as well as methylation of the central 3-position of the
NacNac backbone, are shown to significantly influence redox and
photophysical properties. In addition, modifications of the
cyclometalating (C^N) ligands on iridium are also explored as
avenues to further optimize excited-state redox properties. The
kinetics of excited-state electron transfer with the electron-acceptor
benzophenone (BP) are determined from Stern−Volmer quenching
experiments. The values of quenching rate are dependent on the
driving force of the electron-transfer reaction, as predicted by
Marcus theory. Finally, we use these strong photoreductants as
catalysts for visible-light-promoted catalytic hydrodebromination
reactions of aryl bromide substrates. This improved method
operates with simple amine bases as the sole sacrificial reagent and
does not require silane mediators. Supported by Stern-Volmer
quenching studies with these substrates, the catalytic reactivity is
consistent with an oxidative quenching mechanism for
hydrodebromination, and illustrates the concept that these more
reactive photocatalysts can enable transformations on a wide
substrate scope with relatively simple conditions.
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
Results
Synthesis and structures of the iridium complexes. The chemical
structures of all nine β-diketiminate (NacNac) ligand precursors and
all 13 bis-cyclometalated iridium complexes prepared from these
NacNac ligand precursors are shown in Chart 1. The three
18
compounds 1–3 were reported in our previous study and the other
ten compounds (4–13) are newly prepared. We sought several
modifications to the structures of 1–3 as a means of further
influencing and optimizing the key photosensitizer attributes.
Complexes 4–9 incorporate one or more modifications onto the
NacNac ligand, whereas 10–13 involve modification of the
cyclometalating ligand. Several of the complexes include electron-
donating substituents (Me, OMe, or NMe ) on the NacNac N-
2
Chart 1
phenyl rings (4–7), with complex 7 also including a methyl group
on the central backbone position. In 8 and 9 we installed the
nonaromatic N and N′ substituents OMe (8) and cyclohexyl (9).
Finally, in 10–13 we used cyclometalating ligands besides ppy,
specifically 2-(4-tert-butyl-phenyl)-pyridine (tbppy, 10 and 11) and
Structures of five of the compounds were verified by X-ray
crystallography and are shown in Figure 1. Structures of complexes
18
5
, 9, 10 and 12 are akin to the previously reported structures of 3,
and several other cyclometalated iridium NacNac compounds from
19–21
1
-phenylpyrazole (ppz, 12 and 13). To synthesize the iridium
our group,
which have approximate C symmetry with a planar
2
complexes, we used potassium or lithium salts of each modified
NacNac ancillary ligand combined with chloride-bridged iridium
NacNac core. On the other hand, the crystal structure of 11 is very
similar to the structure of 2, with a nonplanar “puckered”
18
ACS Paragon Plus Environment

Page 3 of 14
conformation of the L2 ligand that results in loss of C
ACS Catalysis
symmetry.
(CH ) onto the ligand backbone results in a ca. 140–190 mV
2
3
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
cathodic shift in redox potential, as judged by comparing 2, 11, and
13 to 1, 10, and 12, respectively. We previously noted that complex
3, with the substitution of ethoxy groups onto the NacNac
backbone, results in a ca. 100 mV of anodic shift of oxidation
potential comparing to 1, the opposite effect of dimethylamino
substitution. By comparing 6 and 7, which differ with respect to the
substitution at the central 3-position of the NacNac backbone, we
This phenomenon is only observed in the solid-state, with solution
NMR spectra consistent with approximate C symmetry. We do not
2
believe that this puckering behavior arises from the steric bulk of the
dimethylamino group on the NacNac backbone, since the structures
of iridium complexes with trifluoro-methyl-substituted NacNac
1
9
ligands remain C symmetric.
2
IV
III
see a substantial cathodic shift of E(Ir /Ir ) (ca. 230 mV) in 7,
where the backbone is methylated. Other modifications involved
alteration of the N-phenyl substituents on the NacNac, which
likewise have significant effects on redox potentials. The effects of
phenyl substituents generally track with the Hammet substituent
constant σ, with more electron-donating substituents resulting in
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
IV
III
more cathodically shifted Ir /Ir potentials. Comparing compound
1 (−0.07 V; +0.33 V vs. SCE) to 4 (−0.11 V; +0.29 V vs. SCE) and
6 (−0.10 V; +0.30 V vs. SCE) reveals very modest effects of adding
4-methoxy (4) or 2,6-dimethyl (6) substituents to the NacNac
phenyl rings. The effect is more pronounced in 4-dimethylamino-
substituted 5 (−0.23 V; +0.17 V vs. SCE). Replacing the phenyl rings
with more electron-rich nonaromatic substituents also has
demonstrable effects. Complex 8 with N-OMe substitution,
IV
III
E(Ir /Ir ) = −0.15 V (+0.25 V vs. SCE), is modestly shifted relative
to phenyl-substituted 1, and the most pronounced effect of all comes
from replacing the aromatic substituents with cyclohexyl rings,
which results in a potential of −0.39 V (+0.01 V vs. SCE) for complex
9. Switching from the ppy C^N ligand to other C^N ligand such as
tbppy (10, 11) or ppz (12, 13) generally has a small impact on the
oxidation potential; the largest perturbation is seen in complex 10,
which is easier to oxidize by 60 mV when compared to 1, but in the
rest of the pairs with the same NacNac and different C^N ligands the
IV
III
differences in Ir /Ir potential are no more than 30 mV. Sweeping
cathodically, the reduction features of most complexes are
irreversible, except 1 and 4, and the peak potentials for all first
reduction events occur at very negative potentials, near −2.7 V (−2.3
III
II
V vs. SCE). Thus, the LUMO energy and the formally Ir /Ir
reduction potentials are not been perturbed much by changing the
NacNac or C^N ligands.
Figure 1. Molecular structures of complexes 5, 9, 10, 11, and 12,
determined by single-crystal X-ray diffraction. H atoms and solvent
molecules are omitted.
Photophysical properties. UV-vis absorption spectra were
recorded in dry acetonitrile solution and are overlaid in Figure 2. We
have additionally recorded spectra of the free NacNac ligands,
1
shown in Figure S2. All complexes exhibit strong (π→π*) transitions
Electrochemical properties. Having previously observed
in the UV region (λ < 280 nm). Overlapped metal-to-ligand charge
significant effects of NacNac backbone substituents on redox
1
3
transfer bands, MLCT and MLCT, dominate the low-energy
regions of the spectra. All complexes also have a band centered
around 400 nm, tentatively ascribed to a NacNac π→π* transition as
IV
III
potentials, in particular the Ir /Ir couples which reflect relative
HOMO energies, one of our primary interests in this study is to
determine the effects of other NacNac modifications on the
electrochemical properties of this photosensitizer library. As
discussed below, the ground-state redox potentials are one of the
two determinants of the excited-state redox potentials, so these
trends in redox behavior provide key insight into how to control and
optimize excited-state redox properties. The redox properties of the
cyclometalated iridium complexes were measured via cyclic
voltammetry in acetonitrile solution with a ferrocene internal
reference. Overlaid voltammograms are collected in Figure S1, and
such pronounced bands are not present in isoelectronic complexes
21,22
with other ancillary ligands.
Furthermore, the position of this
band is sensitive to the NacNac structure, and tends to red shift
significantly as the ligand backbone becomes more electron rich via
dimethylamino substitution (2, 11, and 13) or additional
methylation (7). Significantly, cyclometalated iridium complexes
with ppy and tbppy C^N ligand have intense visible absorption
3
−1
−1
bands with ε > 5 × 10 M cm at 400 nm and significant absorption
intensity tailing out to 500−550 nm. This substantial visible
absorption is a prerequisite for visible-light photosensitization and is
comparable to many other iridium complexes commonly used in
IV
III
Table 1 includes a summary of the Ir /Ir potentials. The first
oxidation potentials are notably affected by the modified NacNac
ligands. Installing dimethylamino (NMe
2
) groups instead of methyl
4
photoredox catalysis. One detriment to switching to the
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 14
Table 1. Summary of ground-state redox potentials, UV-vis absorption spectra, photoluminescence properties, and excited-state redox
potentials.
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
IV
III
IV
III
E(Ir /*Ir ) / V
E(Ir /Ir ) / V
emission, λ / nm
UV-vis Absorption
PLa
a
vs. Fc /Fc
Φ
τ / μs
E
T1 / eVe
vs. Fc /Fc
vs. SCE) g
+
+
3
–1
–1 a
λ / nm (ε/10 M cm )
293 Ka
77 Kd
(vs. SCE) g
(
1
2
3
−0.07 (+0.33)
−0.26 (+0.14)
+0.03 (+0.43)
264 (39), 357 (12), 383 (11), 460 (sh) (1.8)
269 (27), 298 (21), 393 (6.1), 511 (1.9)
595
634
571
522, 554 (sh)
560, 594 (sh)
515, 540 (sh)
0.053
0.16
0.23
0.20
0.76
2.0
2.4
2.3
2.4
−2.5 (−2.1)
−2.6 (−2.2)
−2.4 (−2.0)
264 (62), 342 (24), 413 (sh) (6.1), 456 (sh) (3.2)
2
60 (14), 290 (sh) (8.7), 353 (3.0), 385 (sh) (2.9),
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
1
2
3
4
5
6
7
8
9
0
4
5
6
7
8
9
−0.11 (+0.29)
−0.23 (+0.17)
−0.10 (+0.30)
−0.33 (+0.07)
−0.15 (+0.25)
−0.39 (+0.01)
−0.13 (+0.27)
576
536, 575 (sh)
547, 589 (sh)
553, 590 (sh)
592, 633 (sh)
557, 600 (sh)
607, 658 (sh)
528, 566 (sh)
0.021
0.52
2.4
2.3f
2.3
2.2
2.4
2.0
2.4
−2.5 (−2.1)
−2.5 (−2.1)
−2.4 (−2.0)
−2.5 (−2.1)
−2.6 (−2.2)
−2.4 (−2.0)
−2.5 (−2.1)
435 (0.5)
259 (14), 289 (sh) (7.2), 357 (sh) (3.4), 385 (3.8),
b
b
b
4
35 (0.5), 488 (sh) (0.3)
2
59 (50), 287 (sh) (34), 308 (sh) (23), 365 (12),
605
623c
632
661
575
0.18
0.012c
0.022
0.040
0.045
0.53
391 (15), 500 (1.4)
258 (36), 288 (sh) (26), 319 (sh) (15), 358 (7),
0.63c
0.40
0.35
0.15
3
91 (sh) (8.5), 421 (9.5), 536 (1.3)
2
56 (23), 285 (sh) (13), 350 (4.7), 382 (sh, 4.0),
4
75 (sh) (0.8)
65 (36), 288 (sh) (26), 361 (10) 391 (Sh), 506
2.4)
66 (17), 295 (sh) (10), 356 (4.8), 386 (4.2),
85 (0.5)
2
(
2
1
0
1
4
1
−0.27 (+0.13)
−0.10 (+0.30)
269 (15), 304 (sh) (10), 391 (2.3), 494 (0.6)
245 (22), 307 (7.5), 308 (6.7)
594
558, 593 (sh)
533
0.12
0.51
2.3
−2.6 (−2.2)
−2.4 (−2.0)
b
b
b
2.3f
12
1
3
−0.25 (+0.15)
240 (39), 302 (22), 400 (5.9)
637
570
0.017
0.35
2.2f
−2.4 (−2.0)
a
b
c
d
e
In MeCN unless otherwise noted. Not luminescent at 293 K. In THF. In butyronitrile. Determined from intersection point of UV-vis
f
g
+
absorption and emission, unless otherwise noted. Determined from first vibronic peak in 77-K emission spectrum. Corrected from Fc /Fc =
0.40 V vs. Standard calomel electrode.
2
3
a)7
0
(b)40
in Chart 1.
(
4
5
8
9
1
2
3
60
3
0
Figure 3 shows overlaid photoluminescence spectra of 4–13,
measured at room temperature (298 K) and cooled in liquid
nitrogen (77 K). Complexes 4–9 are Ir(ppy) (NacNac) analogues
which include modified NacNac ligands designed to be more
5
4
3
2
0
0
0
0
20
10
0
2
10
0
Me
electron-rich than the parent NacNac ligand used in complex 1.
3
00 400 500 600 700
300
400

500
600
Most of these modifications induce a significant red-shift in the
room-temperature emission spectrum, and all result in a red-shift of
the 77-K maximum, consistent with destabilized HOMO energies in
these more electron-rich complexes and thus a smaller HOMO–
LUMO gap. Excepting 5 and 12, which are negligibly luminescent at
room temperature, the room-temperature emission maximum of
complex 9 is the most red-shifted at 661 nm (15,100 cm ), which is
shifted by 2,400 cm compared to the emission maximum of 3 (571
nm, 17,500 cm ), which has the most blue-shifted emission in this
/ nm

/ nm
6
7
(d)40
(
c)50
10
11
12
13
40
30
30
2
1
0
0
0
20
1
0
−1
0
−1
300
400
 / nm
500
600
300
400
 / nm
500
600
−1
study. By cooling down to 77 K and using butyronitrile as the
solvent, the low-temperature emission (red line in Figure 3) could
be obtained in a rigid solvent glass. Cooling to 77 K induces a large
rigidochromic blue shift in the emission maximum. The large
Figure 2. UV-vis absorption spectra of (a) 1–3; (b) 4, 5, 8, and 9; (c) 6
and 7; (d) 10–13. All absorption spectra were recorded in acetonitrile.
cyclometalating ligand ppz is a decrease in visible absorption, and for
12 and 13 the absorption completely tails off before 500 nm and is
24
rigidochromic shift, which has been studied by many other groups,
very weak beyond 400 nm. In general, there are no significant
differences between the spectra of the various NacNac free ligands,
as shown in Figure S2 of the Supporting Information. Most NacNac
ligands absorb in the near-UV region, with maxima between ca. 300
and 350 nm. The exception is L2H, which only absorbs high-energy
light <300 nm, likely because L2H exists in the less conjugated keto
form, unlike the other ligands which are in the enol form as described
is evidence for significant charge-transfer character, MLCT and/or
ligand-to-ligand charge transfer (LLʹCT), in the excited state.
Previous computational work on cyclometalated iridium NacNac
complexes with other C^N ligands revealed that the HOMO in
such complexes is almost exclusively NacNac-centered, thus we
20
hypothesize there is significant LLʹCT excited-state character. The
ACS Paragon Plus Environment

Page 5 of 14
ACS Catalysis
+
photoluminescence lifetime (τ
0
) and quantum yield (Φ
0
) were also
Fc /Fc couple and SCE, are summarized in Table 1. A comparison
of the values shows that while the ground-state Ir /Ir potential
IV
III
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
measured in acetonitrile and are included in Table 1. The quantum
yields are low to moderate and vary from negligible (5 and 12) to a
maximum of 0.23 in 3.
varies over a substantial range in this series of compounds, the
IV
III
excited-state Ir /*Ir potentials all fall into a relatively narrow
range. Nevertheless, the excited-states of our complexes are ca. 300–
500 mV more reducing than the excited-state of fac-Ir(ppy)
–1
Energy / cm
2
0000 18000 16000 14000
IV
III
(E*(Ir /*Ir ) = –2.1 V (–1.7 V vs. SCE)), which is the strongest
photoreductant commonly used in photoredox catalysis. Among the
NacNac complexes, the iridium complexes 2 and 11 have the most
negative excited-state reduction potential, ca. −2.6 V (–2.2 V vs.
SCE).
13
12
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
11
A selection of the complexes was evaluated in Stern-Volmer
quenching experiments, to determine the rates of photoinduced
electron transfer with the model substrate benzophenone (BP).
10
−•
Benzophenone has a reduction potential E(BP/BP ) = −2.2 V (–1.8
9
V vs. SCE), indicating there should be a small but positive driving
force for photoinduced electron transfer from all of complexes 1–13.
Furthermore, the reduction potential of benzophenone is similar to
those of many organobromide substrates, so it is a good test
substrate to determine whether excited-state electron-transfer to
organobromides is thermodynamically and kinetically viable. The
quenching reaction via PET can be traced by monitoring the steady-
8
7
6
5
4
state emission (Φ
0
/Φ
q
) and time-resolved (τ
0
q
/τ ) emission
experiment as a function of benzophenone concentration, both
shown in Figure 4. Both the steady-state emission quantum yield and
the lifetime decrease as benzophenone is added, and the linear and
overlaid Stern-Volmer plots for both datasets support that the
reaction occurs via dynamic quenching, in which the quencher and
excited iridium complex react in a bimolecular fashion. The slope of
500
600
700
800

/ nm
Figure 3. Overlaid room-temperature (298 K, black) and low-
temperature (77 K, red) emission spectra of complexes 4–13. All the
measurements were obtained from acetonitrile media for room-
temperature emission and butyronitrile for low-temperature emission.
Complexes 5 and 12 do not emit at room temperature. Emission spectra
of 1–3 can be found in our previous report.18
the Stern-Volmer plot divided by the native lifetime (τ , lifetime
0
without any quencher) gives the quenching rate constant (k ).
q
Some of the complexes were excluded from the quenching study
due to poor photoluminescence intensity and others due to
instability under the conditions of the quenching experiments, but
those tested demonstrated rapid quenching with benzophenone.
Photoinduced electron transfer (PET) quenching study. The
IV
III
excited-state potential (ESP), E(Ir /*Ir ), can be estimated using
photophysical and electrochemical properties that we have obtained
as described above. As shown in Equation 1, the two parameters that
We have previously demonstrated that fac-Ir(ppy)
3
is quenched by
9
−1 −1
benzophenone with a rate constant k
q
= 1.9 × 10 M s , and that
IV
III
IV
III
25
E*(Ir /*Ir ) = E(Ir /Ir ) – ET1
(1)
complexes 1–3 exhibit quenching rate constants ~2–3× faster, with
18
1
and 2 being the fastest. In addition to having similar excited-state
determine the ESP are the first oxidation potential of the complex,
IV
III
redox potentials, all of the newly tested complexes (8, 10, 11, and
E(Ir /Ir ), and the excited-state energy, ET1. ET1 should reflect all of
the available excited-state energy, and thus cannot be simply
determined from the room-temperature emission maximum, the
position of which is influenced by the vibronic coupling between the
excited state and ground state. In the literature, the most common
method for estimating ET1, often abbreviated E0–0, is to locate the
crossover point between the low-energy UV-vis absorption band
and the room-temperature photoluminescence. This method is used
to report the ET1 values collected in Table 1, excepting complexes 5
and 12, which have no room-temperature emission, and 13, which
does not have a well-defined crossover point. For these three
complexes, ET1 is estimated from the first maximum in the 77-K
photoluminescence spectrum, which is also a well-established
method for estimating triplet-state energy and one we have used in
1
3) have similar quenching rate constants as 1 and 2, falling in the
9
−1 −1
narrow range of 4.6–6.0 × 10 M s . These values are all
significantly larger than that for fac-Ir(ppy) , showing that the
3
increase in electron-transfer driving force correlates with an increase
26
in electron-transfer rate constant, as predicted by Marcus theory.
4
3
2
1
2 (ex: 530nm)
0
/
11
0
/


8
o
13
10
1
8,21
0.0
0.2
0.4
0.6
0.8
1.0
previous reports. Neither of the two methods for estimating ET1
are rigorously accurate, but by comparing these two common
methods we estimate the error in ET1 is ±0.1 eV.
[
BP] / mM
Figure 4. Stern-Volmer plots of excited-state quenching by
benzophenone (λex: 380 nm for 10; 400 nm for 8, 11, 13; 530 nm for 2).
The ESPs of all complexes described here, referenced to the
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 14
As part of this expanded study of electron-transfer quenching by work, we focused on optimizing photocatalytic debromination
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
these complexes we also show that when complex 2 is excited with
conditions without silane mediators, focusing our attention instead
530 nm excitation (other experiments were typically conducted with
on comparisons of different catalysts, the effects of catalyst loading,
9
−1 −1
380, 400, or 420-nm excitation), k is 4.8 × 10 M s . This value is
q
Catalytic reactions are summarized in Table 2, and control reac-
tions are summarized in Table S2 in the Supporting Information. No
product formation is observed in the absence of photosensitizer,
base, or light, and yields are attenuated significantly when the reac-
tions are run in air.
similar to the observed value with 420-nm excitation and indicates
that rapid electron-transfer can occur when complex 2 is excited
anywhere within its visible absorption manifold.
Stern-Volmer quenching experiments were also carried out with
five aryl bromide substrates (S1–S5), which were also examined in
photocatalytic hydrodebromination reactions (see below). The
Table 2. Optimization of hydrodebromination reaction and
catalysts screening.
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
Stern-Volmer plots and associated quenching rate constants (k ) are
q
summarized in Figure 5 below. Quenching of these substrates occurs
with rate constants that are at least one order of magnitude smaller
than quenching of benzophenone. There does not appear to be an
obvious correlation between how electron-rich or electron-poor the
substrate is and its quenching rate constant. The substrate methyl 4-
bromo-3-methyl benzoate (S1) has the fastest rate constant, k
q
= 4.6
% Yield
Entry
catalyst (mol%)
Base
T / °C
7
−1 −1
24h/50hf
×
10 M s , and the slowest belongs to bromomesitylene (S5, k
q
=
6
−1 −1
7.8 × 10 M s ), suggesting there may be a steric component to the
1
2
3
4
5
6
7
8
9
2 (1%)
2 (1%)
2 (1%)
2 (1%)
2 (1%)
2 (2.5%)
2 (5%)
2 (2.5%)
TBAa
20
20
20
20
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
33/51
30/40
Trace/16
42/62
61
observed quenching rate. Nevertheless, the quenching rate
constants for all of these substrates are larger than that of TMEDA,
which is used as the base/sacrificial reductant in photoredox
catalytic reactions described below.
DIPEAb
iPAc
TMEDAd
TMEDA
TMEDA
TMEDA
TMEDAe
TMEDA
TMEDA
TMEDA
TMEDA
TMEDA
TMEDA
TMEDA
TMEDA
TMEDA
TMEDA
Photocatalytic hydrodebromination. The highly reducing
photosensitizers described here were investigated in a photoredox
reaction of a challenging aryl bromide substrate, methyl 4-bromo-3-
methyl benzoate. A closely related substrate, lacking the methyl
group at the phenyl 3-position, was shown by Stephenson et al. to be
86
84
debrominated in high yields using the cationic photocatalyst
87
+
[
Ir(ppy)
2
(dtbbpy)] ,
which
requires
2.2
equiv
of
16
Ir(ppy)
Ir(ppy)
3
(1%)
66
tris(trimethylsilyl)silane (TTTMSS) additive to operate. In this
3.0
2.5
2.0
1.5
1.0
10
3
(2.5%)
94
S1
S2
11
1 (1%)
4 (1%)
25
12
22
S4
S3
13
5 (1%)
30
S5
14
15
16
17
18
19
6 (1%)
45
TMEDA
12 (1%)
9 (2.5%)
10 (2.5%)
11 (2.5%)
13 (2.5%)
11
0
10 20 30 40 50
S] / mM
[
82
24
31
TMEDA
15
a
tributylamine b diisopropylethylamine isopropylamine d tetrameth-
c
e
f
ylethylenediamine, 6 equiv Yield was calculated from GC or NMR with
1,2,4,5-tetramethylbenzene as the internal reference. Catalyst 8 was not
tested due to its poor thermal stability.
To choose the proper sacrificial reductant/hydrogen atom donor
for the reaction, several amine bases were screened, as shown in
entries 1–4 in Table 2. Initial screens were conducted with 1 mol%
of complex 2 as the photocatalyst, 20 °C reaction temperature, with
Figure 5. Stern-Volmer plots of excited-state quenching by selected aryl
bromides (S1–S5) and TMEDA base with complex 2, and calculated
3
equiv of base under blue LED illumination. Under these
q
quenching rate constants (k ). For all experiments, λex = 390 nm.and the
conditions, we identified TMEDA as the best choice for a base,
optimum base/sacrificial reductant to use for this transformation.
which yielded 62% of the debrominated product after 50h (entry 4).
ACS Paragon Plus Environment

Page 7 of 14
ACS Catalysis
perturb the excited-state redox properties, our strategy is to make the
Increasing the temperature to 45 °C gave the same yield of product
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
in only 24 h (entry 5). Increasing catalyst loading to 2.5% (entry 6)
increased the yield to 86%, and a further increase to 5 mol% loading
of 2 led to no further increase in yield (entry 7). Similarly, increasing
the amount of TMEDA to 6 equiv, still using 2.5 mol% catalyst, did
not influence the product yield (entry 8).
NacNac ancillary ligand more electron rich. Our previous work on
complexes 1–3 illustrated a structural analogy between the iridium-
NacNac chelate ring and the structure of a six-membered aromatic
ring, as summarized in Figure 6. In our previous work, we focused on
substituent effects at the 2- and 4-positions of the backbone, labeled
as “R” in Figure 6 and analogous to the meta positions of a 6-
membered aromatic ring. In the present work, we also include also
modifications to the substituents on the nitrogen atoms (Rʹin Figure
Encouraged by these results, we investigated whether the more
common photocatalyst fac-Ir(ppy)
3
would also promote
debromination. As mentioned, Stephenson’s work previously
16
6), and at the center 3-position of the backbone (Rʹʹ), which are
showed that debromination required a silane mediator, but in their
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
respectively analogous to the ortho and para substituents of the
aromatic ring. Consideration of the Hammet substituent constants
(σ) allows us to predict which modifications will increase the
electron-richness of the NacNac ancillary ligand.
work they used a cationic iridium catalyst which is not as strongly
reducing. As shown in entries 9 and 10, fac-Ir(ppy) outperforms
3
catalyst 2 in this reaction, yielding 66% product after 24 h with 1
mol% loading, and 94% when the catalyst loading is increased to 2.5
mol%. Finally, as entries 11–19 show, most of the other
photosensitizers described in this work are effective catalysts for
visible-light promoted debromination, catalysts 2 and 9 rivaling the
C^N ligand
Metalloaromatic HOMO
N
ortho
R'
meta
R
R'
R
C
para
R''
performance of fac-Ir(ppy)
3
in these reactions. The silane mediator
N
Ir
R''
tris(trimethylsilyl)silane (TTMSS) is not detrimental to catalysis,
and with of 2.2 equivalents of TTMSS (Table S2, entry 5) the
product yield is 98%, as compared to the 86% using otherwise
identical conditions without TTMSS (Table 2, entry 6). Thus, silane
mediators may slightly improve the hydrodebromination yield, but
are not strictly required. Finally, as described in the Supporting
Information (Figure S3 and page S7), we are able to set up these
reactions on the benchtop and scale them up to 2 mmol of substrate.
We obtained a nearly identical yield of P1 (82%) under these
conditions when compared to the smaller-scale screen where
reagents were handled in a glovebox (86% yield, entry 6 of Table 2).
N
C
Ir d orbital
R'
R
R
R'
N
-donating
Figure 6. Modification of π-donating ancillary ligands via electron-rich
substituents. The p-orbital conjugation status of NacNac ligand is
20
described by calculation result from our previous study .
The effects of the NacNac ligand substitution are most clearly
IV
III
discerned by comparing the formally Ir /Ir redox potentials,
which reflect relative HOMO energies. The series of compounds
where C^N = ppy and R = Me, i.e. the 2- and 4-positions of the
NacNac backbone are methyl groups, allows us to evaluate the
effects of different substituents by comparing to parent complex 1
Using the optimized conditions, photocatalytic reactions of four
other aryl bromide substrates were carried out as described in
Scheme 1, using 2 or 9 as the catalyst. Near-quantitative yields were
achieved with the electron-deficient substrate 1-bromo-3,5-
bis(trifluoromethyl)benzene (S2), and the more electron-rich
substrate 3-bromoanisole (S3) could be acted upon by catalyst 9
with good yields of the hydrodebromination product.
Hydrodebromination of the fluorinated substrate S4 occurs with
moderate yields, and the sterically encumbered substrate
bromomesitylene (S5) can also be debrominated under these
reaction conditions, with moderate yields of mesitylene observed.
IV
III
(
E(Ir /Ir ) = −0.07 V). Figure 7 summarizes the perturbations of
IV
III
E(Ir /Ir ) caused by adding various electron-donating substituents
to complex 1. As expected, increasing the electron-richness of the
NacNac ligand destabilizes the HOMO and cathodically shifts the
redox potential, although the magnitude depends on the type of
substituent and its location. Addition of OMe (σ
p
= –0.27) to the 4-
position of the N-phenyl rings results in a modest 40-mV shift of the
IV
III
potential (complex 4, E(Ir /Ir ) = −0.11 V), but the more
electron-rich NMe
2
(σ = –0.83) substituents in 5 cause a much
p
III
IV
larger shift (E(Ir /Ir ) = −0.23 V). The effect of 2,6-dimethyl
substitution on the N-phenyls is also quite small, with the potential
in complex 6 (−0.10 V) only 30-mV shifted from the parent complex
1. Comparison of complex 6 to 7 shows that methylation of the 3-
position of the NacNac backbone, i.e. para in Figure 6, induces a
large 230-mV shift in potential, in spite of methyl groups only being
modestly electron-donating (σ = –0.17). However, previous DFT
p
calculations from our group on other bis-cyclometalated iridium
NacNac complexes showed that the NacNac core participates in the
HOMO much more than the peripheral aryl groups, which is
consistent with the observation that backbone methylation in 7 has
a much larger effect than N-phenyl substituents in 4–6. Another
substitution we investigated was to replace the N-phenyl rings (i.e.
the ortho positions in Figure 6) with electron-donating nonaromatic
substituents. Methoxy-substituted analogue 8 has a potential (−0.15
V) only modestly shifted from 1, but replacing the N-phenyl rings
Scheme 1. Hydrodebromination reactions of S2–S5. The catalyst
used (2 or 9) is indicated in parentheses after the percent yield.
Discussion
In this study, we have explored further insights into our ligand
design strategy for potent bis-cyclometalated iridium
1
8
photoreductants, which we recently introduced. To further
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 14
0
.0
0.1
-0.2
0.3
-0.4
We have also verified, through Stern-Volmer quenching
2
,6-Me2
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
6
experiments using select members of the photosensitizer library with
benzophenone as the electron acceptor, that these NacNac
complexes undergo excited-state electron transfer with rate constant
OMe
-
1
4-OMe
4
8
4
-NMe2
2–3 times faster than fac-Ir(ppy)
electron transfer the compounds we describe here are
thermodynamically and kinetically more potent than fac-Ir(ppy)
3
. Thus, in terms of excited-state
Cy
5
-
7
9
3
,
Methylate
Replace N-phenyl Add N-phenyl
E
3-position of
backbone
the most reducing of the commonly used photoredox catalysts.
These attributes motivated us to begin screening the library of
NacNac-based photosensitizers for catalytic applications.
with nonaromatic
group
substituents
-
0.5
-0.6
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
To demonstrate the catalytic efficacy of the photosensitizer
library, we screened them for the hydrodebromination of model aryl
bromide substrates, reasoning that the enhanced excited-state
reactivity of these complexes would enable efficient activation of
challenging substrates. Previous methods for aryl bromide
hydrodebromination required addition of excess silane mediators,
Figure 7. Plot showing the effects of various NacNac substituents on the
Ir /Ir redox potentials in Ir(ppy)
IV
III
2
(NacNac) derivatives.
with cyclohexyl groups in 9 proved to have the largest effect of all,
IV
III
shifting the Ir /Ir couple to −0.39 V.  ese observations suggest
that the basicity of the NacNac nitrogen donors, expected to be
much higher in complex 9, plays a large role in determining the
HOMO energy.
16
which generates a lot of unrecoverable waste byproducts. Starting
with 1 mol% complex 2 as the photocatalyst, the most reducing of
18
our initial set of Ir(ppy)
2
(NacNac) complexes, we optimized
The above analysis of redox potentials shows that the structure of
the NacNac ligand has a significant effect on the HOMO energy, and
that electron-donating groups can drive the Ir /Ir potential to
conditions for photocatalytic debromination of methyl 4-bromo-3-
methyl benzoate. Good yields of the debrominated methyl 3-
methylbenzoate product were obtained, with TMEDA identified as
the best base/sacrificial reductant we tried. Encouraged by these
IV
III
more negative values. On its own, this would result in a more
IV
III
negative excited-state Ir /*Ir potential, per Equation 1 above.
However, as we previously noted when introducing complexes 1–3,
there is a tradeoff between the two terms in Equation 1, such that a
cathodically shifted ground-state potential is accompanied by a
decrease in the triplet excited-state energy, ET1. This same tradeoff is
observed in the much larger suite of photosensitizers described here,
as shown in Figure 8. Note that we estimated ET1 to only one decimal
place, owing to the inherent error in determining that parameter (see
above), so the complexes are clustered over only a few excited-state
energies, but nonetheless the tradeoff between the two terms is
results, we tried the same reaction conditions with fac-Ir(ppy) as the
3
photocatalyst, noting that in previous work the Stephenson group
+
had used the cationic catalyst [Ir(ppy)
2
(dtbbpy)] , which is far less
IV
III
reducing (E(Ir /*Ir ) = −1.3 V; –0.9 V vs. SCE) than fac-Ir(ppy)
The hydrodebromination reaction proceeds similarly with fac-
Ir(ppy) as the catalyst, and these results show that
3
.
3
hydrodebromination is possible without silane mediators, provided
that a sufficiently reducing photocatalyst is used. We examined the
remaining photosensitizers as catalysts under analogous reaction
conditions, excepting 7 and 8 which are too unstable for catalytic
applications, and found that all resulted in productive
hydrodebromination. There are significant differences in
performance between the different catalysts (see Table 2), though
we don’t see an obvious correlation between product yield and ESP.
We are unable to determine from the available data if the differences
in performance are due to inherent differences in reaction rate or
better stability of some of the catalysts, but nevertheless we can
IV
III
apparent. As a result, the excited-state redox potentials, E(Ir /*Ir ),
determined via Equation 1, all fall in a relatively narrow range for the
NacNac complexes described here, between −2.4 and −2.6 V (–2.0
and –2.2 V vs. SCE). Nonetheless, all of these photosensitizers are
more potent excited-state reductants than fac-Ir(ppy) , and the
3
tunable ground-state redox properties may also play a role in
catalysis applications, influencing back-electron transfer and
ground-state redox processes involving sacrificial reagents.
conclude empirically that complex 2, fac-Ir(ppy) , and complex 9
3
0
.4
performed the best in this particular reaction.
fac-Ir(ppy)3
0.3
A mechanism of the hydrodehalogenation reaction, based on
oxidative quenching of the photocatalyst excited state, is proposed
in Scheme 2. We have shown via control experiments three
components are required for the catalytic reaction: photosensitizer,
sacrificial agent (electron and proton donor), and light (see Table
S2). Examining the Stern-Volmer quenching data in Figure 5, under
the typical operating conditions of the catalytic reactions (~0.033 M
substrate, ~0.10 M TMEDA), quenching rates for the substrates are
competitive with or faster than the quenching rate for TMEDA. At
one extreme, the quenching rate for S1 is 28× the quenching rate for
TMEDA, and at the other extreme the quenching rate for S5 is 0.37×
the quenching rate for TMEDA. With this data in mind we can’t
explicitly rule out a reductive quenching mechanism, but we do
expect quenching by the aryl bromide substrate to be at least
competitive if not kinetically preferred. Thus, the mechanism in
Scheme 2 starts with absorption of a photon to generate the excited
0
0
0
.2
.1
.0
3
1
-
-
-
0.1
0.2
0.3
12
2
4
8
10
E
5
1
13
1
7
9
-0.4
2.0
2.1
2.2
2.3
2.4
2.5
ET1 / eV
IV
III
Figure 8. Plot of the ground-state Ir /Ir potential vs. the triplet
excited-state energy, for the cyclometalated iridium
photosensitizers described here.
T1
E ,
ACS Paragon Plus Environment

Page 9 of 14
ACS Catalysis
state of the iridium catalyst, *Ir(III). From there, electron-transfer to
the quantum yield for photoinduced electron transfer to the
substrate:
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
the substrate generates the substrate radical anion and the oxidized
−
photocatalyst, denoted Ir(IV). The substrate radical anion loses Br
ꢄꢅꢆsubstrateꢇ
ꢀꢁꢂ
ꢃ
(2)
to form the aryl radical. The TMEDA (or other amine) serves as a
sacrificial reductant and hydrogen atom donor, donating an electron
to the oxidized catalyst to regenerate the Ir(III) ground state,
followed by hydrogen atom transfer to the aryl radical to generate
the arene product. Studies of different bases showed that tertiary
amines like TMEDA give better outcomes than secondary amines,
which outperform primary amine bases. Tertiary amines are known
to be stronger reductants and hydrogen atom donors than primary
and secondary amines, so this trend further supports the role of the
base in the catalytic cycle.
ꢄ ꢆsubstrateꢇ ꢊ ꢄ′ ꢆTMEDAꢇ ꢊ ꢄ ꢊ ꢄ
ꢋꢉ
ꢅ
ꢈ
ꢉ
Using catalyst 2 and typical catalytic reaction conditions (0.033 M
substrate, 0.10 M TMEDA), we estimate electron-transfer quantum
yields ranging between 0.12 (S5) and 0.91 (S1). The quantum
yields for hydrodebromination product formation are probably
much lower, since it is unlikely that every electron-transfer event to
the substrate results in productive turnover. This analysis does
indicate that the strongly reducing photocatalysts we describe here
give rise to efficient electron transfer, though further improvements
in catalyst design could lead to small increases in electron-transfer
rates for some of the substrates. Further interrogation, namely
measurement of photocatalytic quantum yields, will reveal whether
catalytic turnover is limited by electron transfer or by one of the dark
reactions that occurs subsequent to electron transfer.
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
Conclusions
In this work we performed a thorough study of ground and excited-
state redox properties of bis-cyclometalated iridium complexes with
substituted β-diketiminate ancillary ligands. We showed that the
IV
III
ground-state Ir /Ir redox potentials and the triplet-state energies
are very sensitive to the substitution pattern of the NacNac ligand.
All of the compounds are more potent excited-state reductants than
fac-Ir(ppy) , with excited-state redox potentials ~300–500 mV more
3
Scheme 2. Proposed mechanism of hydrodebromination reaction
via oxidative quenching pathway.
negative. To demonstrate a catalytic application of these
compounds, we screened them as catalysts for the photocatalytic
debromination of aryl bromide substrates, and found that our
That said, we cannot rule out alternative pathways for catalysis,
especially for the dark electron and hydrogen-atom transfer steps
that occur after oxidative quenching. For example, one possibility is
instead of the TMEDA reducing the oxidized catalyst (Ir(IV)), it
could transfer a hydrogen atom to the aryl radical, generating the
product and a neutral TMEDA radical. The TMEDA radical could
then undergo electron transfer with Ir(IV) to regenerate the catalyst
and form the iminium cation. In addition, a radical chain process is
also possible, whereby the aryl radical reacts with TMEDA to form
the product via hydrogen-atom transfer, and the TMEDA radical
thus formed activates the aryl bromide to form the aryl radical and
the iminium byproduct. The Stern-Volmer data suggests oxidative
quenching as a viable pathway, but other possibilities do exist.
catalysts as well as the popular fac-Ir(ppy) catalyst enable
3
productive conversion using much simpler reaction conditions than
were previously described. We expect that this suite of highly
reactive, well-characterized photosensitizers will enable many future
discoveries in photoredox catalysis, especially for substrates that are
difficult to activate.
Experimental Section
Materials. All reactions were executed in a nitrogen-filled
glovebox operating at < 1 ppm of O
2
and < 1 ppm of H O. Solvents
2
for reactions and optical measurements were dried by the method of
27
Grubbs , passing through dual alumina columns on a commercial
solvent purification system (SPS), and stored over 3Å molecular
sieves. All NMR solvents were dried and stored over 3Å molecular
We did not measure the photocatalytic quantum yield for the
2
8,29
hydrodebromination
reactions,
and
there
are
no
sieves. The β-diketiminate ligands L1H, L2H, and L3H,
their
hydrodebromination quantum yields reported in the literature to
benchmark against. However, the Stern-Volmer quenching data in
potassium salts L1K, L2K, and L3K, and the respective iridium
complexes1, 2, and 3 were prepared according to our previously
1
8
Figure 5, in combination with the k
r
and knr values determined from
reported procedures. The β-diketiminate ligands L4H–L6H, L8H,
and L9H were prepared following a general procedure, via
condensation of 2,4-pentanedione with the appropriate aniline or
amine in the presence of 4-toluenesulfonic acid. The ligand L7H and
the photophysical properties in Table 1, allow us to estimate the
electron-transfer quantum yields for the substrates studied here
under catalytic conditions. We do note that Stern-Volmer
experiments were conducted at room temperature and optimized
catalytic reactions at 45 °C, so these numbers may not be
quantitatively reliable but they do give some insight into the
efficiency of electron transfer. In the catalytic reaction mixture, the
3
0
its potassium salt L7K were prepared as previously described. C^N
ligands, [Ir(C^N)
phenylpyridine,
phenylpyrazole ), benzylpotassium , and fac-Ir(ppy)
2
(μ-Cl)]
2
starting materials (C^N=2-
3
1
2-(4-tert-butyl-phenyl)pyridine ,
1-
3
2
33
13
3
were
possible excited-state deactivation pathways are radiative decay (k
r
),
obtained according to the various literature procedures.
Benzophenone was purchased from Aldrich and was used without
further purification. The substrates for hydrodebromination
reactions were obtained commercially and used without further
purification.
nonradiative decay (knr), and electron-transfer to the substrate or
TMEDA. The rates of the electron transfer processes are given by
the respective k
q
(Figure 5) multiplied by the concentration of that
species. With this information, Equation 2 can be used to estimate
ACS Paragon Plus Environment

ACS Catalysis
Physical Methods. NMR spectra were recorded at room
in 20 mL of Et
Page 10 of 14
O. The solution was stirred for 30 min and cooled in
2
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
the glovebox cold well with liquid nitrogen for 30 min. After removal
from the cold well, benzyl potassium (130 mg, 0.998 mmol, 1.00
equiv) was added to the solution. The reaction mixture was stirred
at room temperature for 1 h. During this time, the red color of the
benzyl potassium disappeared, and a yellow precipitate formed. The
precipitate was collected by vacuum filtration and washed with
hexane. Finally, the product was dried in vacuum. Yield: 330 mg
temperature using a JEOL ECA-500 or ECA-600 NMR
spectrometer. UV−vis absorption spectra were recorded in MeCN
solutions in screw-capped quartz cuvettes using an Agilent Carey 60
UV−vis spectrophotometer. Emission spectra were recorded using a
Horiba FluoroMax-4 spectrofluorometer, using appropriate long-
pass filters to exclude stray excitation light from detection. Samples
for room-temperature emission were housed in 1 cm quartz cuvettes
with septum-sealed screw caps, and samples for low-temperature
emission were contained in a custom quartz EPR tube with high-
vacuum valve and immersed in liquid nitrogen using a finger dewar.
Emission quantum yields were determined relative to a standard of
tetraphenylporphyrin in toluene, which has a reported fluorescence
1
8
(94.4 %). H NMR (600 MHz, THF-d ) δ: 6.78–6.82 (m, 8H, ArH),
4.75 (s, 1H, ArNC(Me)CHC(Me)NAr), 3.70 (s, 6H, OCH ), 1.87
(s, 6H, CH
Preparation of L5K. L5H (336 mg, 0.999 mmol) was dissolved
in 20 mL of Et O. The solution was stirred for 30 min and cooled in
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
3
)
2
3
4
quantum yield (Φ ) of 0.11 . We have determined the quantum
F
the glovebox cold well with liquid nitrogen for 30 min. After removal
from the cold well, benzyl potassium (130 mg, 0.998 mmol, 1.00
equiv) was added to the solution. The reaction mixture was stirred
at room temperature for 1 h. During this time, the red color of the
benzyl potassium disappeared, and a yellow precipitate formed. The
precipitate was collected by vacuum filtration and washed with
hexane. Finally, the product was dried in vacuum. Yield: 365 mg
yield (Φ
x
) by using the equation, Φ
x
= Φref × (slope /sloperef) ×
x
2
(�MeCN/�toluene) , where the slope comes from the best-fit line of
integrated photoluminescence vs. absorption at the excitation
wavelength. Luminescence lifetimes were measured with a Horiba
DeltaFlex Lifetime System, using 453 nm pulsed diode excitation.
Emission wavelengths were selected by using appropriate long-pass
filters, and the decay trace was fit using the instrument’s analysis
software. Cyclic voltammetry (CV) measurements were performed
with a CH Instruments 602E potentiostat interfaced with a nitrogen
glovebox via wire feed-throughs. Samples were dissolved in
1
8
1
(
97.4%). H NMR (600 MHz, THF-d ) H NMR (600 MHz, THF-
8
d ) δ: 6.63–6.76 (m, 8H, ArH), 4.70 (s, 1H,
ArNC(Me)CHC(Me)NAr), 2.84 (s, 12H, N(CH
CH
Preparation of L6K. L6H (306 mg, 0.999 mmol) was dissolved
in 20 mL of Et O. The solution was stirred for 30 min and cooled in
3
)
2
), 1.86 (s, 6H,
3
)
acetonitrile with 0.1 M TBAPF as a supporting electrolyte. A 3 mm
6
diameter glassy carbon working electrode, a platinum wire counter
electrode, and a silver wire pseudo-reference electrode were used.
Potentials were referenced to an internal standard of ferrocene.
2
the glovebox cold well with liquid nitrogen for 30 min. After removal
from the cold well, benzyl potassium (130 mg, 0.998 mmol, 1.00
equiv) was added to the solution. The reaction mixture was stirred
at room temperature for 1 h. During this time, the red color of the
benzyl potassium disappeared, and a yellow precipitate formed. The
precipitate was collected by vacuum filtration and washed with
Stern-Volmer Quenching Studies. All samples were prepared in
a glove box to avoid any quenching by oxygen. Stock solutions of the
iridium complexes, BP, and aryl bromides were prepared in dry
MeCN solvent, and the iridium complex was diluted in a quartz
–6
cuvette to make 1.0–3.0 × 10 M solution. The concentration of the
hexane. Finally, the product was dried in vacuum. Yield: 310 mg
–
3
quencher stock solution was between 2.0 and 3.0 × 10 M, and
aliquots of 5–10 μL were introduced to the cuvette. UV-vis
absorption spectra, photoluminescence lifetimes, and steady-state
emission spectra were recorded before and after the addition of each
quencher aliquot. For lifetime measurements the samples were
excited at 453 nm, and for steady-state emission spectra, 420 nm
excitation was used. Also, to rule out any quenching by adventitious
oxygen, the emission spectra and lifetime were measured repeatedly
after the addition of the quencher to ensure stable values.
1
(
90.0 %). H NMR (400 MHz, C
6
6
D ) δ: 7.05 (d, J = 7.6 Hz, 4H,
ArH), 6.84 (t, J = 7.4 Hz, 2H, ArH), 4.54 (s, 1H,
ArNC(Me)CHC(Me)NAr), 1.90 (s, 12H, Ar-CH ), 1.62 (s, 6H,
CH ).
Preparation of L8Li. L8H (a mixture of isomer, 158 mg) was
dissolved in 10.0 mL of Et O and kept in the freezer for 1 hour. Then,
.7 equiv of 1.6 M of n-BuLi in Hexane solution was added to the
3
3
2
0
solution. 1 hour later, the precipitate was filtered and washed with
hexane. Yield: 75 mg (45.7%). H NMR (600 MHz, C
1
6
D
6
) δ: 4.72
), 1.60
General procedure for blue-light promoted aryl bromide
hydrodebromination. A 2.5 dram vial was filled with catalytic
amounts (1~5 mol %) of photosensitizer, the amine reagent (3 or 6
equiv), and 0.1 mmol of the substrate methyl 3-methyl-4-
bromobenzoate. After adding 3 mL of dry MeCN, the vial was
secured with a rubber-septa-lined plastic cap. The home-built
photoreactor (see photo in Figure S3) consists of a glass bowl
wrapped with blue LED strips, purchased from Creative Lighting
Solutions (Model: Sapphire Blue LED Tape -12vdc), and wrapped
on the outside with aluminum foil. To maintain the reaction
temperature, the vessel was filled with water, heated on a stirring
hotplate, monitored with a thermometer. After the specified
reaction time, an aliquot of the reaction solution was withdrawn
using micro-syringe and analyzed with GC or NMR.
(
s, 1H, MeONC(Me)CHC(Me)NOMe), 3.60 (s, 6H, OCH
s, 6H, CH ).
Preparation of L9Li. L9H (264 mg, 0.999 mmol) was dissolved
in 10.0 mL of Et O and kept in the freezer for 1 hour. Then, 1.1 equiv
3
(
3
2
of 1.6 M of n-BuLi in Hexane solution was added to the solution. 1
1
hour later, the precipitate was filtered and washed with hexane. H
NMR in C
6
6
D indicates successful deprotonation of L9H, with the
characteristic downfield N–H resonance no longer present.
−
However, there are two sets of peaks for L9 in a 1:1 ratio, indicating
either two isomeric products in a 1:1 ratio or formation of a cluster
–
with two inequivalent chemical environments for L9 . Yield: 122 mg
1
(
45%).
H
NMR (500 MHz,
C
4.34
6
D
6
)
δ: 5.10 (s, 1H,
(s, 1H,
CyNC(Me)CHC(Me)NCy), 3.48 (t, J = 11 Hz, 2H), 3.48 (tt, J =
1, 4.0 Hz, 2H), 2.55 (s, 6H, CH3), 2.19 (d, J = 11.5 Hz, 4H), 2.02
CyNC(Me)CHC(Me)NCy),
Preparation of L4K. L4H (310 mg, 0.999 mmol) was dissolved
1
ACS Paragon Plus Environment

Page 11 of 14
ACS Catalysis
(
d, J = 14 Hz, 4H), 1.83–1.67 (m, 10H), 1.76 (s, 6H, CH3), 1.57– method using [Ir(ppy)
2
(μ-Cl)]
2
(100 mg, 1 equiv., 0.0933 mmol)
1
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
1.43 (m, 10H), 1.25–1.10 (m, 12H).
and L7K (72 mg, 2.1 equiv., 0.20 mmol) Yield: 70 mg (46%). H
NMR (500 MHz, C
6
D ) δ: 9.26 (d, J = 5.5 Hz, 2H, ArH), 7.09 (d, J
6
General Procedure for the Preparation of bis-Cyclometalated
= 8.5 Hz, 2H, ArH), 6.99 (d, J = 7.5 Hz, 2H, ArH), 6.89 (t, J = 8.0
Hz, 2H, ArH), 6.73 (m, 4H, ArH), 6.62 (m, 4H, ArH), 6.54 (t, J =
6.5 Hz, 2H, ArH), 6.46 (t, J = 6.0 Hz, 2H, ArH), 6.04 (d, J = 7.5 Hz,
Iridium β-Diketiminate Complexes. In a glovebox, the chloro-
bridged dimer [Ir(C^N)
2
(μ-Cl)] was suspended in 2 mL of THF,
2
and 2.1 equivalents of the respective NacNac potassium or lithium
salt was dissolved in 6 mL of THF. The solution of the NacNac
ligand was added to the chloro-bridged dimer suspension via pipet.
The mixture was stirred for 1 h at room temperature, during which
the color darkened from yellow to orange or red. The solution was
concentrated in vacuo, and the crude product was extracted into 2
mL of toluene and filtered. The solution was again concentrated, and
trituration with 2 mL of diethyl ether released the solid yellow
product, which was washed with 2 × 2 mL of pentane and dried
under vacuum.
2H, ArH), 2.16 (s, 6H, CH
3
), 2.09 (s, 3H, Ar-CH
). C{ H} NMR (126 MHz, C ) δ:
3
), 1.63 (s, 6H,
1
3
1
CH
3
), 1.33 (s, 6H, Ar-CH
3
6
D
6
170.8, 159.6, 153.3, 152.5, 150.6, 144.9, 135.8, 133.4, 132.8, 131.4,
128.1, 127.7, 127.2 123.1, 123.0, 119.9, 119.3, 117.8, 98.4, 21.09,
20.9, 19.0, 18.0. UV-vis(MeCN) λ/nm (ε/M cm ) 258 (36000),
288 (26000), 319 (15000), 358 (7000), 391 (8500), 421 (9500),
536 (1300).
−1
−1
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
Preparation of Ir(ppy)
using [Ir(ppy) (μ-Cl)] (100 mg, 1 equiv., 0.0933 mmol) and L8Li
33 mg, 2.1 equiv., 0.20 mmol) was dissolved in 6 mL of THF. In a
slight departure from the general procedure, both solutions were
kept in the freezer (−35 °C) for 1 h prior to mixing. The remaining
2
(L8) (8). Prepared by the general route,
2
2
(
Preparation of Ir(ppy)
2
(L4) (4). Prepared by the general
strategy using [Ir(ppy) (μ-Cl)]
and L4K (70 mg, 2.1 equiv., 0.20 mmol) Yield: 50 mg (33%). H
NMR (400 MHz, C
=
Hz, 2H, ArH), 6.62 (d, J = 6.4 Hz, 2H, ArH), 6.54 (t, J = 5.2 Hz, 2H,
ArH), 6.51 (t, J = 5.2 Hz, 2H, ArH), 6.4 (t, J = 5.6 Hz, 2H, ArH), 6.33
2
2
(100 mg, 1 equiv., 0.0933 mmol)
1
1
procedure and workup are identical. Yield: 45 mg (34%). H NMR
6
D ) δ: 9.35 (d, J = 4.8 Hz, 2H, ArH), 7.17 (d, J
6.4 Hz, 2H, ArH), 6.99 (d, J = 6.4 Hz, 2H, ArH), 6.96 (t, J = 6.4
6
(
500 MHz, C
6
6
D ) δ: 9.00 (d, J = 5.5 Hz, 2H, ArH), 7.42 (d, J = 6.5
Hz, 2H, ArH), 7.29 (d, J = 8.0 Hz, 2H, ArH), 6.86 (t, J = 7.5 Hz, 2H,
ArH), 6.82 (t, J = 7.5 Hz, 2H, ArH), 6.73 (m, 4H, ArH), 6.40 (t, J =
6.0 Hz, 2H, ArH), 3.57 (s, 1H, MeONC(Me)CHC(Me)NOMe),
(
d, J = 6.8 Hz, 2H, ArH), 6.17 (m, 4H, ArH), 5.35 (d, J = 6.4 Hz, 2H,
13
1
2
C
1
.89 (s, 6H, OCH
) δ: 169.6, 156.5, 155.0, 150.2, 145.0, 135.6, 133.0, 128.8,
23.6, 120.8, 120.6, 118.1, 79.8, 60.3, 19.9. UV-vis(MeCN) λ/nm
3
), 1.83 (s, 6H, CH
3
). C{ H} NMR (126 MHz,
ArH), 4.75 (s, 1H, ArNC(Me)CHC(Me)NAr), 3.14 (s, 6H,
OCH
1
1
1
3
1
6
D
6
3
), 1.74 (s, 6H, CH
70.1, 158.5, 156.6, 155.3, 150.3, 146.3, 143.8, 135.4, 132.9, 125.8,
24.8, 123.4, 120.3, 119.3, 118.1, 112.9(d), 95.9, 54.7, 25.2.; UV-
3
). C{ H} NMR (101 MHz, C
6
D
6
) δ:
−1
−1
(ε/M cm ) 256 (23000), 285 (13000), 350 (4700), 382 (4000),
−1
−1
475 (800).
vis(MeCN) λ/nm (ε/M cm ) 260 (14000), 290 (8700), 353
3000), 385 (sh) (2900), 435 (500)
Preparation of Ir(ppy) (L5) (5). Prepared by the general
method using [Ir(ppy) (μ-Cl)]
(
Preparation of Ir(ppy)
2
(L9) (9). This compound was prepared
(μ-Cl)] (100 mg, 1 equiv., 0.0933
by a different route. [Ir(ppy)
2
2
2
mmol) was suspended in 20 mL of THF in a Teflon-capped glass
tube. A solution of L9Li (53 mg, 2 equiv., 0.20 mmol) in 10 mL of
THF was added via pipette. The resulting suspension was heated to
2
2
(100 mg, 1 equiv., 0.0933 mmol)
1
and L5K (75 mg, 2.1 equiv., 0.20 mmol). Yield: 67 mg (43%). H
NMR (600 MHz, THF-d ) δ: 9.21 (d, J = 5.4 Hz, 2H, ArH), 7.74 (t,
J = 8.4 Hz, 2H, ArH), 7.69 (d, J = 8.4 Hz, 2H, ArH), 7.21 (t, J = 6.0
Hz, 2H, ArH), 7.03 (d, J = 8.4 Hz, 2H, ArH), 6.32 (d, J = 9.0 Hz, 2H,
ArH), 6.27 (t, J = 7.8 Hz, 2H, ArH), 6.08 (m, 4H, ArH), 5.81 (d, J =
8
7
0 °C for 3 days. The resulting red mixture was concentrated in
vacuo. The dark-red residue was extracted into 20 mL of toluene and
the solution was filtered through Celite. The toluene was removed
in vacuo and the residue was crystallized using 2 mL Et O and 10 mL
of pentane at −35 °C. The supernatant liquid was removed via
pipette and the solid residue was washed three more times with
2
7
2
.8 Hz, 2H, ArH), 5.71 (d, J = 7.2 Hz, 2H, ArH), 5.03 (d, J = 8.4 Hz,
H, ArH), 4.42 (s, 1H, ArNC(Me)CHC(Me)NAr), 2.56 (s, 12H,
1
3
1
6
N(CH
3
)
2
), 1.51 (s, 6H, CH ). C{ H} NMR (151 MHz, THF-d )
3
pentane. The product was dried in vacuo and obtained as an orange
δ: 170.0, 158.8, 156.1, 150.5, 146.6, 144.0, 143.76, 135.6, 132.6,
27.4, 125.0, 122.8, 120.5, 118.4, 118.0, 112.7, 112.3, 94.8, 40.7. UV-
1
solid. Yield: 56 mg (37%). H NMR (400 MHz, C
6
6
D ) δ: 9.27 (d, J
1
−
1
−1
= 4.8 Hz, 2H, ArH), 7.36 (d, J = 7.2 Hz, 2H, ArH), 7.32 (d, J = 8.4
Hz, 2H, ArH), 6.91 (t, J = 8.0 Hz, 2H, ArH), 6.77 (t, J = 8.0 Hz, 2H,
ArH), 6.67 (t, J = 7.2 Hz, 2H, ArH), 6.50 (m, 4H, ArH), 4.27 (s, 1H,
CyNC(Me)CHC(Me)NCy), 3.34 (t, J = 12.0 Hz, 2H, CH), 1.96 (s,
vis(MeCN) λ/nm (ε/M cm ) 259 (14000), 289 (7200), 357
3.400), 385 (3800), 435 (500), 488 (300)
Preparation of Ir(ppy) (L6) (6). Prepared by the general
method using [Ir(ppy) (μ-Cl)]
(
2
2
2
(100 mg, 1 equiv., 0.0933 mmol)
6
H, CH
CH ), 1.43 (q, J = 16 Hz, 4H, CH
.83 (m, 6H, CH
101 MHz, C
3
), 1.80 (d, J = 9.6 Hz, 2H, CH
2
), 1.70 (q, J = 12 Hz, 2H,
),1.26 (t, J = 12.8 Hz, 4H, CH ),
). C{ H} NMR
1
and L6K (69 mg, 2.1 equiv., 0.20 mmol) Yield: 60 mg (39%). H
2
2
2
NMR (400 MHz, C
8.0 Hz, 2H, ArH), 6.96 (d, J = 8.0 Hz, 2H, ArH), 6.90 (t, J = 8.0
Hz, 2H, ArH), 6.70–6.62 (m, 4H, ArH), 6.58–6.52 (m, 6H, ArH),
6
D ) δ: 9.44 (d, J = 6.0 Hz, 2H, ArH), 7.08 (d, J
6
13
1
0
(
2
), 0.50 (q, J = 12 Hz, 2H, CH
2
=
6
D
6
) δ: 169.6, 159.1, 157.9, 151.6, 144.8, 135.5, 132.2,
1
2
29.2, 123.8, 120.2, 119.9, 117.9, 99.4, 66.4, 34.7, 34.2, 26.9, 26.6,
6.3, 25.1. UV-vis(MeCN) λ/nm (ε/M cm ) 265 (36000), 288
6
1
0
C
1
.45 (t, J = 7.2 Hz, 2H, ArH), 6.11 (d, J = 8.0 Hz, 2H, ArH), 4.89 (s,
H, ArNC(Me)CHC(Me)NAr), 2.20 (s, 6H, Ar-CH ), 1.57 (d, J =
). C{ H} NMR (101 MHz,
) δ: 170.5, 159.9, 153.6, 151,8 150.2, 144.9, 135.9, 133.0, 132.3,
31.4, 123.3, 122.9, 120.1, 119.4, 117.7, 97.2, 24.9, 21.4, 17.4. UV-
−1
−1
3
(
sh, 26000), 361 (10000) 391 (Sh, 9300), 506 (2400).
Preparation of Ir(tbppy) (L1) (10). Prepared by the general
method using [Ir(tbppy) (μ-Cl)]
1
3
1
.8 Hz, 6H, CH
3
) 1.3 (s, 6H, Ar-CH
3
2
6
D
6
2
2
(130 mg, 1 equiv., 0.100 mmol)
1
−1
−1
and L1K (58 mg, 2.0 equiv., 0.20 mmol). Yield: 75 mg (44%). H
vis(MeCN) λ/nm (ε/M cm ) 259 (50000), 287 (34000), 308
23000), 365 (12000), 391 (15000), 500 (1400).
Preparation of Ir(ppy) (L7) (7). Prepared by the general
NMR (400 MHz, C
6
6
D ) δ: 9.37 (d, J = 4.4 Hz, 2H, ArH), 7.23 (d, J
(
=
8.0 Hz, 2H, ArH), 7.03–6.97 (m, 4H, ArH), 6.79–6.72 (m, 4H,
2
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 14
ArH), 6.56–6.50 (m, 8H, ArH), 5.97 (d, J = 2.0 Hz, 2H, ArH), 5.50
(t, J = 2.0 Hz, 2H, ArH), 4.74 (s, 1H, ArNC(Me)CHC(Me)NAr),
Notes
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
The authors declare no competing financial interest.
1
3
1
1.70 (s, 6H, CH
MHz, C ) δ: 170.1, 157.4, 156.9, 152.9, 150.4(d), 141.2, 135.4,
129.3, 129.0, 124.9, 124.5, 122.7, 122.3, 119.8, 117.5, 116.8 95.8,
3
), 0.89 (s, 18H, C(CH ) ), C{ H} NMR (101
3
3
6
D
6
ASSOCIATED CONTENT
−1
−1
Supporting Information. Crystallographic summary tables, cyclic volt-
ammograms, additional experimental details for photoredox reactions,
and NMR spectra for all new compounds. This material is available free
of charge via the Internet at http://pubs.acs.org. CCDC 1894972–
1894976 contains the supplementary crystallographic data in CIF for-
mat, available for download from https://www.ccdc.cam.ac.uk/struc-
tures/.
33.8, 30.9, 25.2. UV-vis(MeCN) λ/nm (ε/M cm ) 266 (17000),
295 (10000), 356 (4800), 386 (4200), 485 (500).
Preparation of Ir(tbppy)
2
(L2) (11). Prepared by the general
method using [Ir(tbppy) (μ-Cl)]
2
2
(130 mg, 1 equiv., 0.100 mmol)
1
and L2K (70 mg, 2.0 equiv., 0.20 mmol). Yield: 67 mg (37%). H
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
NMR (500 MHz, C
6
D
6
) δ: 8.87 (d, J = 4.5 Hz, 2H, ArH), 7.37 (d, J
=
8.0 Hz, 2H, ArH), 7.19 (d, J = 8.0 Hz, 2H, ArH), 6.93 (d, J = 8.0
Hz, 2H, ArH), 6.85−6.78 (m, 6H, ArH), 6.64 (t, J = 7.0 Hz, 2H,
ArH), 6.56 (s, 2H, ArH), 6.32 (broad m, 6H, ArH), 4.10 (s, 1H,
ACKNOWLEDGMENT
We thank the University of Houston, the National Science Foundation
(CHE-1846831), and the Welch Foundation (Grant No. E-1887) for
funding this research and Prof. Loi Do for access to the gas
chromatography.
PhNC(NMe
2
)CHC(NMe
2
)NPh), 2.29 (s, 12H, N(CH
3
) ), 1.04 (s,
2
13
1
1
1
1
8H, C(CH
3
)
3
). C{ H} NMR (126 MHz, C ) δ: 169.7, 167.8,
6
D
6
57.9, 154.3, 151.7, 150.8, 142.2, 135.1, 129.5, 128.3, 123.5,
19.8,118.9, 117.3, 116.6, 84.4, 40.9, 34.1, 31.2. UV-vis(MeCN)
−1
−1
REFERENCES
(1)
Teets, T. S.; Nocera, D. G. Solar Energy Supply and Storage for the Legacy
and Nonlegacy Worlds. Chem. Rev. 2010, 110 (11), 6474–6502.
https://doi.org/10.1021/cr100246c.
λ/nm (ε/M cm ) 269 (15000), 304 (sh) (10000), 391 (2300), 494
600).
Preparation of Ir(ppz)
procedure using [Ir(ppz) (μ-Cl)]
(
Cook, T. R.; Dogutan, D. K.; Reece, S. Y.; Surendranath, Y.;
2
(L1) (12). Prepared by the general
(100 mg, 1 equiv., 0.0973 mmol)
2
2
L1K (67 mg, 2.1 equiv., 0.20 mmol). A slightly different workup
procedure was used, triturating the crude product with 2 mL of
hexane before washing with 2 × 2 mL of hexane and drying under
(
2)
Chem.
https://doi.org/10.1039/c1cc12390d.
(3) Youngblood, W. J.; Lee, S.-H. A.; Maeda, K.; Mallouk, T. E. Visi-
Teets, T. S.; Nocera, D. G. Photocatalytic Hydrogen Production.
Commun. 2011, 47 (33), 9268.
1
vacuum to give a pale-yellow solid product. Yield: 45 mg (31%). H
ble Light Water Splitting Using Dye-Sensitized Oxide Semiconductors. Acc
Chem Res 2009, 42, 1966.
(4)
Photoredox Catalysis with Transition Metal Complexes: Applications in Or-
ganic Synthesis. Chem. Rev. 2013, 113 (7), 5322–5363.
https://doi.org/10.1021/cr300503r.
NMR (400 MHz, C
2.0 Hz, 2H, ArH), 6.81–6.80 (m, 2H, ArH), 6.66 (d, J = 7.6 Hz,
H, ArH), 6.58 (t, J = 7.2 Hz, 4H, ArH), 6.48 (dt, J = 8.0, 1.2 Hz, 2H,
ArH), 6.39 (dd, J = 7.6, 1.2 Hz, 2H, ArH), 6.34 (dt, J = 7.2, 1.6 Hz,
6
6
D ) δ: 7.90 (d, J = 1.6 Hz, 2H, ArH), 7.14 (d, J
=
2
Prier, C. K.; Rankic, D. A.; MacMillan, D. W. C. Visible Light
2H, ArH), 6.10 (t, J = 2.4 Hz, 2H, ArH), 6.00 (dd, J = 7.2, 1.2 Hz,
2
H, ArH), 5.52 (broad s, 2H, ArH), 4.66 (s, 1H,
(
5)
Corcoran, E. B.; Pirnot, M. T.; Lin, S.; Dreher, S. D.; DiRocco, D.
13
1
PhNC(Me)CHC(Me)NPh), 1.66 (s, 6H, CH
3
). C{ H} NMR
A.; Davies, I. W.; Buchwald, S. L.; MacMillan, D. W. C. Aryl Amination Us-
ing Ligand-Free Ni(II) Salts and Photoredox Catalysis. Science 2016, 353
(
101 MHz, C
6
D
6
) δ: 157.0, 153.4, 143.2, 139.0, 138.3, 134.8, 125.7,
(
6296), 279–283. https://doi.org/10.1126/science.aag0209.
(6) Beatty, J. W.; Stephenson, C. R. J. Amine Functionalization via
1
25.0, 124.6, 124.1, 122.5, 119.9, 109.8, 106.4, 95.7, 24.7. UV-
−1
−1
vis(MeCN) λ/nm (ε/M cm ) 268 (24000), 357 (8000), 388
Oxidative Photoredox Catalysis: Methodology Development and Complex
Molecule Synthesis. Acc. Chem. Res. 2015, 48 (5), 1474–1484.
https://doi.org/10.1021/acs.accounts.5b00068.
(7)
catalytic CO
mogeneous Catalyst. J. Am. Chem. Soc. 2014, 136 (48), 16768–16771.
https://doi.org/10.1021/ja510290t.
(8000), 470 (1000).
Preparation of Ir(ppz)
2
(L2) (13). Prepared by the general
(100 mg, 1 equiv., 0.0973 mmol)
method using [Ir(ppz)
2
(μ-Cl)]
2
Bonin, J.; Robert, M.; Routier, M. Selective and Efficient Photo-
and L2K (56 mg, 2.1 equiv., 0.20 mmol). A slightly modified workup
procedure was used, trituring the crude product with 2 mL of hexane
and washing the resulting solid with 2 × 2 mL of hexane before
2
Reduction to CO Using Visible Light and an Iron-Based Ho-
(
8)
defluorination: Facile Access to Partially Fluorinated Aromatics. J. Am.
Chem. Soc. 2014, 136 (8), 3002–3005.
https://doi.org/10.1021/ja500031m.
9) McAtee, R. C.; McClain, E. J.; Stephenson, C. R. J. Illuminating
Senaweera, S. M.; Singh, A.; Weaver, J. D. Photocatalytic Hydro-
drying under vacuum to give a pale-green solid product. Yield: 84 mg
1
(
53%). H NMR (500 MHz, C
6
6
D ) δ: 7.31 (d, J = 1.5 Hz, 2H, ArH),
7.03 (d, J = 3.0 Hz, 2H, ArH), 6.92 (broad t, 4H, ArH), 6.75–6.65
(m, 8H, ArH), 6.50 (broad s, 4H, ArH), 6.48 (d, J = 7.0 Hz, 2H,
(
ArH),
5.80
(s,
2H,
)NPh), 2.28 (s, 12H, N(CH
) δ: 167.9, 155.3 143.8, 139.4, 134.8,
25.2, 124.1 120.3 120.1 110.3 105.4 83.7, 40.9 (Three peaks are
ArH),
4.09
(s,
1H,
Photoredox Catalysis. Trends Chem. 2019,
https://doi.org/10.1016/j.trechm.2019.01.008.
1
(1), 111–125.
PhNC(NMe
2
)CHC(NMe
2
3
)
2
).
1
3
1
C{ H} NMR (126 MHz, C
6
D
6
(10) Chen, M.; Zhong, M.; Johnson, J. A. Light-Controlled Radical
Polymerization: Mechanisms, Methods, and Applications. Chem. Rev. 2016,
116 (17), 10167–10211. https://doi.org/10.1021/acs.chemrev.5b00671.
(11) Nguyen, J. D.; Matsuura, B. S.; Stephenson, C. R. J. A Photo-
chemical Strategy for Lignin Degradation at Room Temperature. J. Am.
Chem. Soc. 2014, 136 (4), 1218–1221. https://doi.org/10.1021/ja4113462.
1
overlapped with the benzene peak and were not located). UV-
−1
−1
vis(MeCN) λ/nm (ε/M cm ) 240 (39000), 302 (22000), 400
5900)
(
(
12) Lowry, M. S.; Goldsmith, J. I.; Slinker, J. D.; Rohl, R.; Pascal, R.
AUTHOR INFORMATION
A.; Malliaras, G. G.; Bernhard, S. Single-Layer Electroluminescent Devices
Corresponding Author
*E-mail: tteets@uh.edu.
ACS Paragon Plus Environment

Page 13 of 14
ACS Catalysis
and Photoinduced Hydrogen Production from an Ionic Iridium(III) Com-
plex. Chem. Mater. 2005, 17 (23), 5712–5719.
https://doi.org/10.1021/cm051312+.
13) Dedeian, K.; Djurovich, P. I.; Garces, F. O.; Carlson, G.; Watts,
R. J. A New Synthetic Route to the Preparation of a Series of Strong Photo-
reducing Agents: Fac-Tris-Ortho-Metalated Complexes of Iridium(III)
with Substituted 2-Phenylpyridines. Inorg. Chem. 1991, 30 (8), 1685–1687.
https://doi.org/10.1021/ic00008a003.
(29) Land, M. A. New Electron-Rich Diketiminate Ligands: An Ex-
perimental and Computational Investigation on the Isolation of Reactive
Species, Saint Mary’s University: Thesis, 2016.
(30) Rodriguez, M. M.; Bill, E.; Brennessel, W. W.; Holland, P. L. N2
Reduction and Hydrogenation to Ammonia by a Molecular Iron-Potassium
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
(
Complex.
Science
2011,
334
(6057),
780–783.
https://doi.org/10.1126/science.1211906.
(31) Liu, W.; Zhu, Y.; Li, C.-J. Pd-Catalyzed Homo Cross-Dehydro-
genative Coupling of 2-Arylpyridines by Using I2 as the Sole Oxidant. Syn-
thesis 2016, 48 (11), 1616–1621. https://doi.org/10.1055/s-0035-
1561399.
(32) Correa, A.; Bolm, C. Ligand-Free Copper-CatalyzedN-Arylation
of Nitrogen Nucleophiles. Adv. Synth. Catal. 2007, 349 (17–18), 2673–
2676. https://doi.org/10.1002/adsc.200700408.
(33) Bailey, P. J.; Coxall, R. A.; Dick, C. M.; Fabre, S.; Henderson, L.
C.; Herber, C.; Liddle, S. T.; Loroño-González, D.; Parkin, A.; Parsons, S.
The First Structural Characterisation of a Group 2 Metal Alkylperoxide
Complex: Comments on the Cleavage of Dioxygen by Magnesium Alkyl
(
14) Nguyen, J. D.; D’Amato, E. M.; Narayanam, J. M. R.; Stephen-
son, C. R. J. Engaging Unactivated Alkyl, Alkenyl and Aryl Iodides in Visible-
Light-Mediated Free Radical Reactions. Nat. Chem. 2012, 4 (10), 854–859.
https://doi.org/10.1038/nchem.1452.
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
(
15) Fry, A. J.; Krieger, R. L. Electrolyte Effects upon the Polaro-
graphic Reduction of Alkyl Halides in Dimethyl Sulfoxide. J. Org. Chem.
976, 41 (1), 54–57. https://doi.org/10.1021/jo00863a012.
16) Devery, J. J.; Nguyen, J. D.; Dai, C.; Stephenson, C. R. J. Light-
Mediated Reductive Debromination of Unactivated Alkyl and Aryl Bro-
mides. ACS Catal. 2016, (9), 5962–5967.
https://doi.org/10.1021/acscatal.6b01914.
1
(
6
Complexes. Chem.
-
Eur. J. 2003,
9
(19), 4820–4828.
(17) Le, C.; Chen, T. Q.; Liang, T.; Zhang, P.; MacMillan, D. W. C. A
Radical Approach to the Copper Oxidative Addition Problem: Trifluoro-
methylation of Bromoarenes. Science 2018, 360 (6392), 1010–1014.
https://doi.org/10.1126/science.aat4133.
https://doi.org/10.1002/chem.200305053.
(34) Seybold, P. G.; Gouterman, M. Porphyrins: XIII: Fluorescence
Spectra and Quantum Yields. J. Mol. Spectrosc. 1969, 31 (1–13), 1–13.
(
18) Shon, J.-H.; Teets, T. S. Potent Bis-Cyclometalated Iridium Pho-
toreductants with β-Diketiminate Ancillary Ligands. Inorg. Chem. 2017, 56
(
24), 15295–15303. https://doi.org/10.1021/acs.inorgchem.7b02859.
(19) Maya, R. M.; Maity, A.; Teets, T. S. Fluorination of Cyclomet-
alated Iridium β-Ketoiminate and β-Diketiminate Complexes: Extreme Re-
dox Tuning and Ligand-Centered Excited States. Organometallics 2016, 35
(
17), 2890–2899. https://doi.org/10.1021/acs.organomet.6b00453.
20) Lai, P.-N.; Brysacz, C. H.; Alam, M. K.; Ayoub, N. A.; Gray, T.
(
G.; Bao, J.; Teets, T. S. Highly Efficient Red-Emitting Bis-Cyclometalated
Iridium Complexes. J. Am. Chem. Soc. 2018, 140 (32), 10198–10207.
https://doi.org/10.1021/jacs.8b04841.
(21) Radwan, Y. K.; Maity, A.; Teets, T. S. Manipulating the Excited
States of Cyclometalated Iridium Complexes with β-Ketoiminate and β-
Diketiminate Ligands. Inorg. Chem. 2015, 54 (14), 7122–7131.
https://doi.org/10.1021/acs.inorgchem.5b01401.
(
22) Lamansky, S.; Djurovich, P.; Murphy, D.; Abdel-Razzaq, F.;
Kwong, R.; Tsyba, I.; Bortz, M.; Mui, B.; Bau, R.; Thompson, M. E. Synthesis
and Characterization of Phosphorescent Cyclometalated Iridium Com-
plexes.
https://doi.org/10.1021/ic0008969.
23) Connelly, N. G.; Geiger, W. E. Chemical Redox Agents for Or-
Inorg.
Chem.
2001,
40,
1704–1711.
(
ganometallic Chemistry. Chem. Rev. 1996, 96 (2), 877–910.
https://doi.org/10.1021/cr940053x.
(
24) Lees, A. J. The Luminescence Rigidochromic Effect Exhibited by
Organometallic Complexes: Rationale and Applications. Comments Inorg.
Chem. 1995, 17 (6), 319–346.
https://doi.org/10.1080/02603599508032711.
25) Juris, A.; Balzani, V.; Barigelletti, F.; Campagna, S.; Belser, P.; von
(
Zelewsky, A. Ru(II) Polypyridine Complexes: Photophysics, Photochemis-
try, Electrochemistry, and Chemiluminescence. Coord. Chem. Rev. 1988, 84,
85–277. https://doi.org/10.1016/0010-8545(88)80032-8.
(26) Marcus, R. A. Chemical and Electrochemical Electron-Transfer
Theory. Annu. Rev. Phys. Chem. 1964, 15 (1), 155–196.
https://doi.org/10.1146/annurev.pc.15.100164.001103.
(
27) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J. Safe and Convenient Procedure for Solvent Purification. Or-
ganometallics 1996, 15 (5), 1518–1520.
https://doi.org/10.1021/om9503712.
28) Fürstner, A.; Alcarazo, M.; Krause, H. TetrakisS(Dimethyla-
mino)Allene. Org. Synth. 2009, 86, 298–307.
https://doi.org/10.15227/orgsyn.086.0298.
(
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 14
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
Insert Table of Contents artwork here
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
ACS Paragon Plus Environment