Discovery and Elucidation of Counteranion Dependence in Photoredox Catalysis

Article abstract of DOI:10.1021/jacs.9b01885

Over the past decade, there has been a renewed interest in the use of transition metal polypyridyl complexes as photoredox catalysts for a variety of innovative synthetic applications. Many derivatives of these complexes are known, and the effect of ligand modifications on their efficacy as photoredox catalysts has been the subject of extensive, systematic investigation. However, the influence of the photocatalyst counteranion has received little attention, despite the fact that these complexes are generally cationic in nature. Herein, we demonstrate that counteranion effects exert a surprising, dramatic impact on the rate of a representative photocatalytic radical cation Diels-Alder reaction. A detailed analysis reveals that counteranion identity impacts multiple aspects of the reaction mechanism. Most notably, photocatalysts with more noncoordinating counteranions yield a more powerful triplet excited state oxidant and longer radical cation chain length. It is proposed that this counteranion effect arises from Coulombic ion-pairing interactions between the counteranion and both the cationic photoredox catalyst and the radical cation intermediate, respectively. The comparatively slower rate of reaction with coordinating counteranions can be rescued by using hydrogen-bonding anion binders that attenuate deleterious ion-pairing interactions. These results demonstrate the importance of counteranion identity as a variable in the design and optimization of photoredox transformations and suggest a novel strategy for the optimization of organic reactions using this class of transition metal photocatalysts.

Full text of DOI:10.1021/jacs.9b01885

Subscriber access provided by Drexel University Libraries
Article
Discovery and Elucidation of Counteranion
Dependence in Photoredox Catalysis
Elliot P Farney, Steven J Chapman, Wesley B. Swords,
Marco D. Torelli, Robert J Hamers, and Tehshik P. Yoon
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.9b01885 • Publication Date (Web): 21 Mar 2019
Downloaded from http://pubs.acs.org on March 21, 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 8
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Discovery and Elucidation of Counteranion Dependence in Photore-
dox Catalysis
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Elliot P. Farney, Steven J. Chapman, Wesley B. Swords, Marco D. Torelli, Robert J. Hamers, Tehshik P. Yoon*
Department of Chemistry, University of Wisconsin–Madison, 1101 University Avenue, Madison, Wisconsin 53706
ABSTRACT: Over the past decade, there has been a renewed interest in the use of transition metal polypyridyl complexes as photoredox
catalysts for a variety of innovative synthetic applications. Many derivatives of these complexes are known, and the effect of ligand modifications
on their efficacy as photoredox catalysts has been the subject of extensive, systematic investigation. However, the influence of the photocatalyst
counteranion has received little attention, despite the fact that these complexes are generally cationic in nature. Herein, we demonstrate that
counteranion effects exert a surprising, dramatic impact on the rate of a representative photocatalytic radical cation Diels–Alder reaction. A
detailed analysis reveals that counteranion identity impacts multiple aspects of the reaction mechanism. Most notably, photocatalysts with more
non-coordinating counteranions yield a more powerful triplet excited state oxidant and longer radical cation chain length. It is proposed that
this counteranion effect arises from Coulombic ion-pair interactions between the counteranion and both the cationic photoredox catalyst and
the radical cation intermediate, respectively. The comparatively slower rate of reaction with coordinating counteranions can be rescued by using
hydrogen-bonding anion binders that attenuate deleterious ion-pairing interactions. These results demonstrate the importance of counteranion
identity as a variable in the design and optimization of photoredox transformations and suggest a novel strategy for the optimization of organic
reactions using this class of transition metal photocatalysts.
community of scholars interested in the use of these complexes as
photoredox catalysts in organic chemistry.
Introduction
Ruthenium(II) polypyridyl complexes have been among the most
widely studied molecular photocatalysts for a variety of applications.
The photophysical, electrochemical, and physical properties of this
Ligand effects well-characterized
CF3
N
2X–
2X–
2X–
class of luminescent transition metal complexes have been exten-
F3C
1
sively characterized. They generally exhibit strong absorbance in the
N
N
CF3
CF3
N
N
N
N
N
N
N
N
N
N
N
N
N
visible spectrum, feature high intersystem crossing efficiency, and
can participate in a diverse range of photoinduced electron- and en-
ergy-transfer processes. Because of these attractive features, Ru(II)
photocatalysts were instrumental in the early development of solar
N
N
N
N
RuII
N
RuII
N
RuII
N
F3C
N
2
fuels technologies; in addition, some of the best light-harvesting
CF3
sensitizers for dye-sensitized solar cells belong to this family of com-
1 [Ru(bpy)3](X)2
2 [Ru(bpz)3](X)2
3 [Ru(btfmb)3](X)2
plexes. Over the past decade, the recognition that Ru(bpy)32+ and its
3
Effect of counteranion structure poorly documented
analogues are also useful photocatalysts for organic transformations
4
has stimulated a renewal of interest in photochemical synthesis. Be-
CF3
cause of the exceptional utility of Ru(bpy)32+ in so many diverse ap-
plications, numerous structurally varied Ru(II) polypyridyl photo-
catalysts have been prepared, and the effects of ligand modifications
F
P
F
F
F
F
B
O3S CF3
F
CF3
4
5
3a X = BArF
3b X = TfO–
3c X = PF6
–
–
on catalyst properties are well-understood (Figure 1).
4
The effect of counteranion structure on the photoactivity of these
cationic complexes, on the other hand, has not been subject to simi-
lar systematic study. In this paper, we document the discovery of the
unexpected impact counteranion identity plays on the efficiency of a
CF3
CF3
O3S
O2C
O3S
CH3
CF3
–
CF3
–
radical cation Diels–Alder cycloaddition,
a representative
3d X = ArFSO3
3e X = TsO–
3f X = ArFCO2
photoredox transformation. We rationalize the observed rate in-
crease as the consequence of (1) a change in the photocatalyst
ground-state electrochemical properties, (2) a significant shift in its
triplet-state energy, and (3) an increase in the efficiency of radical
cation chain propagation. The results reported herein suggest that
this counterion effect may be an unappreciated but important phe-
nomenon in many photoredox reactions. Understanding the impact
of this experimental variable, therefore, should benefit the growing
Figure 1. Structurally varied Ru(II) photocatalysts.
Results and Discussion
Counterion Effects in Radical Cation Cycloadditions
Several years ago, we reported that visible light photoredox catal-
ysis offered an efficient means to conduct radical cation
Diels–Alder cycloadditions between a wide range of electron-rich
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 8
styrenes and diverse dienes.6 The highly electron-deficient
series of salts bearing a variety of lipophilic counteranions (3a–f)
and assessed their activities in a model photoreaction.
[Ru(bpz)3](BArF )2 complex7 proved to be a potent photocatalyst
1
2
3
4
5
6
7
8
4
for this transformation, providing excellent rates and yields at ambi-
The results of this initial catalyst screen for the radical cation
Diels–Alder cycloaddition between anethole (4) and isoprene (5)
are summarized in Table 1. In general, these various photocatalysts
each promoted the reaction, but surprisingly, the rates of reaction
varied dramatically depending on counteranion identity. While the
cycloaddition of 4 and 5 is complete in 20 min using 1 mol% of
8
ent temperatures with as little as 0.5 mol% of photocatalyst. Our
proposal for the mechanism of this reaction is briefly summarized in
Scheme 1. Photoexcitation of Ru(bpz)32+ with visible light results in
the efficient formation of a long-lived redox-active triplet state. The
electron-deficient bpz ligands render the photoexcited catalyst a
substantially stronger oxidant (+1.4 V vs SCE) than the parent
BArF – catalyst 3a, the reaction proceeds to only 55% yield after 24 h
4
2+
Ru(bpy)3 catalyst (+0.89 V vs SCE), enabling the one-electron
with the analogous triflate catalyst 3b, 14% yield with the 3,5-bis(tri-
9
–
fluoromethyl)benzenesulfonate (ArFSO3 ) catalyst 3d, and only 2%
photooxidation of anethole (4, +1.1 V vs SCE). The resulting alkene
radical cation undergoes rapid [4+2] cycloaddition with diene 5 to
afford product radical cation 6•+.9 Formation of the neutral product
can occur by one of two mechanisms: either radical chain-propagat-
ing oxidation of another equivalent of alkene 4 or by chain-terminat-
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
yield with the tosylate catalyst 3e. No conversion was observed in
this timeframe using the carboxylate complex 3f. Thus, there ap-
pears to be a correlation between the rate of product formation and
the non-coordinating nature of the catalyst counteranions.12 Una-
ware of any previous report of similarly dramatic counterion effects
on the rate of organic photoredox transformations, we elected to in-
vestigate the origins of this phenomenon.
+
ing oxidation of the reduced Ru(bpz)3 catalyst.10 The latter process
regenerates the photoactive Ru(II) state of the catalyst and closes
the catalytic cycle.
Scheme 1. Proposed Photocatalytic Radical Cation Diels–Alder Cy-
cloaddition Mechanism
Table 1. Counteranion Effect on the Rate of Radical Cation
Diels–Alder Cycloaddition.a
MeO
MeO
MeO
visible light
[Ru(bpz)3](BArF
2 (0.5 mol%)
MeO
[Ru(btfmb)3](X)2
)
4
(1 mol%)
+
CH2Cl2
CH2Cl2
23 W CFL
6
Me
4
Me
Me
Me
Me
6
Me
4
5
Me
hν
entry
catalyst
time
yield 6b
RuII
MeO
4
4
unreacted 4
RuII*
b
–
1
2
3
4
5
3a (BArF
)
20 min
24 h
98%
55%
14%
2%
0%
37%
80%
95%
100%
4
RuI
3b (TfO–)
3d (ArFSO3
3e (TsO–)
MeO
–
5
)
24 h
Me
24 h
4
6
–
3f (ArFCO2
)
24 h
0%
Me
Me
Me
a
Despite the efficiency and broad scope of this reaction, the
General conditions: anethole (0.06 mmol), isoprene (0.18 mmol),
CH2Cl2 (0.08 M), [Ru] (1 mol%). Each reaction was subjected to 3
2+
Ru(bpz)3 chromophore suffers from limited solubility in
freeze-pump-thaw cycles prior to irradiation from a 23 W CFL bulb. b 1
NMR yields referenced to trimethyl(phenyl)silane internal standard.
H
non-polar organic solvents. Empirical screening indicated that the
reaction proceeds more rapidly in these solvents and resulted in the
use of CH2Cl2 in optimized reaction conditions. Nevertheless, the
solubility of the bpz complex in CH2Cl2 is modest, and reactions
conducted at moderate catalyst loadings are often visibly heteroge-
neous. These constraints limited our ability to systematically inves-
tigate catalyst structure on the rate of photoredox reactions. We thus
became interested in the use of strongly oxidizing photocatalysts
with greater lipophilicity that might be freely soluble in low-dielec-
tric solvents.
To begin to understand this effect, we performed
Stern–Volmer analyses of the relationship between the concentra-
tion of anethole (4) in CH2Cl2 and the photoluminescence intensity
–
–
of the BArF –, PF6 , and ArFSO3 complexes of Ru(btfmb)32+. This in-
4
vestigation demonstrated that the degree of excited-state quenching
between the photoexcited catalyst and the organic substrate (i.e., the
Stern–Volmer constant, Ksv) decreased by two orders of magnitude
from the least coordinating counteranion, BArF –, to the most coor-
4
–
dinating counteranion in this study, ArFSO3 (Figure 2A). This is
As a starting point for these studies, we prepared a series of pho-
tocatalysts based upon the Ru(btfmb)32+ chromophore (Figure 1, 3;
btfmb = 4,4’-bis(trifluoromethyl)-2,2’-bipyridyl). The photophysi-
cal and electrochemical properties of the homoleptic
[Ru(btfmb)3](PF6)2 complex were previously investigated in ace-
tonitrile by Furue and Kamachi.11 Given the oxidizing potential re-
ported for its excited state (+1.3 V vs SCE), we hypothesized that
Ru(btfmb)32+ would be an effective photooxidative catalyst for elec-
tron-rich styrenes such as 4. Moreover, we hoped that the lipophilic
CF3 substituents would improve the solubility of the photocatalyst
in non-polar organic solvents compared to Ru(bpz)32+. In order to
maximize the organic solubility of this chromophore, we prepared a
–
consistent with the markedly superior reactivity of the BArF com-
4
plex. The value of Ksv is dependent both upon the excited-state life-
time of the photocatalyst (t) and the bimolecular electron-transfer
rate constant (kq); Ksv = tkq. To deconvolute whether the large
change in the Stern–Volmer constant arises primarily from a change
in catalyst triplet lifetime or in the electron-transfer rate constant, we
measured t for each Ru(btfmb)32+ complex (Table 2). These results
show that while the counteranion does have an influence on triplet
lifetime, the effect is relatively small — approximately two-fold over
the range of counteranions investigated. The impact on kq, therefore,
–
is much larger, spanning two orders of magnitude from ArFSO3 3d
ACS Paragon Plus Environment

Page 3 of 8
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. A. Stern–Volmer plots for excited-state quenching of catalysts 3a–c in CH2Cl2. B/C. Effect of counteranion identity on excited-
state properties of [Ru(btfmb)3](X)2, X is indicated in the legend. (B) Absorption (solid line) and photoluminescence (dashed line) spectra
in MeCN. (C) Absorption (solid line) and photoluminescence (dashed line) spectra in CH2Cl2.
to BArF – 3a. Thus, the unanticipated conclusion from these prelim-
are often ideal for applications in organic synthesis. Figure 2C shows
absorption and emission spectra for the same series of catalysts in
CH2Cl2. The impact of counteranion identity on the absorption
4
inary studies is that the identity of the photocatalyst counteranion
can impact a photooxidative reaction by dramatically altering the in-
trinsic rate constant of bimolecular electron transfer to the photo-
catalyst excited state.
spectrum
in
this
relatively
non-polar solvent is modest, suggesting that if differences in Cou-
lombic interactions exert any influence on the ground-state proper-
ties of the photocatalyst or on the singlet excited state, it is a small
effect. In contrast, the photoluminescence spectra of the various
Table 2. Spectroscopic Properties and Stern–Volmer Quench-
ing Constantsa
complexes differ markedly. Most notably, the lmax of photolumines-
cence varies by 52 nm from the least coordinating (BArF –, 573 nm)
4
entry catalyst
KSV
(x 102 M-1)
kq
λabs
λem
τ
to the most coordinating (TsO–, 625 nm) counteranion, corre-
sponding to a substantial energy difference of 4.2 kcal/mol (0.18
eV).
(x 108 M-1 s-1)
(ns)
(nm) (nm)
1
2
3
3a
3c
3d
453
458
458
573
605
618
17
4.2
520
860
950
33
4.9
Thus, altering the identity of the counteranion produces an unex-
pectedly large change in the energy of the emissive triplet state of
Ru*(btfmb)32+. One would expect these changes to be reflected in
excited-state redox potentials. To quantify this effect, we measured
0.61
0.64
a Data acquired in CH2Cl2.
2+
the one-electron reduction potential E(Ru2+/+) of the Ru(btfmb)3
This finding was surprising. While the importance of the bipyridyl
ligand structure in the design and optimization of photocatalytic re-
actions is well appreciated,13 the effect of the catalyst counterion on
photocatalytic reaction rates has received significantly less attention.
Meyer and coworkers have examined ion-pairing effects on the pho-
tophysics of Ru(II) polypyridyl chromophores.14 These investiga-
tions show that addition of Cl– to solutions of
[Ru(bpy)2(deeb)](PF6)2 in CH2Cl2 results in the formation of tight
complexes in CH2Cl2. Each measurement was made using a match-
ing n-Bu4N+X– salt as a supporting electrolyte in order to avoid com-
plications arising from counteranion exchange. Counteranions of
lower Lewis basicity resulted in significant anodic shifts in the
ground-state potentials, with the largest and most significant effect
observed for the BArF – counteranion (Table 3). This effect can also
4
be rationalized as a consequence of ion pairing where the one-elec-
tron reduction of the least electrostatically stabilized BArF – complex
4
ion pairs and
a
concomitant decrease in triplet
15
is more energetically favorable than the tightly ion-paired tosylate
complex. To calculate the excited-state redox potential, we made the
excited-state energy and lifetime. However, neither the impact of
structurally complex organic counteranions on the photochemical
properties of Ru(II) complexes, nor the effects of ion pairing on the
rate of synthetic photocatalytic applications, have been systemati-
commonly utilized assumption ,18 that the Gibbs free energy change
17
for the S0 to T1 transition is represented by the energy of the corre-
sponding photoluminescence maximum (DGES). The catalytically
relevant first triplet excited-state reduction potential E(Ru2+*/+) can
then be approximated from the sum of DGES and E(Ru2+/+). As the
data in Table 3 show, these potentials span a range of 480 mV (11
16
cally explored.
Spectroscopic, Electrochemical, and Quantum Yield
Studies
We have collected absorption, emission, and electrochemical data
kcal/mol), with the BArF – complex having the most positive reduc-
4
tion potential of +1.52 V vs SCE. The conclusion from these studies,
therefore, is that the degree of ion pairing has a synergistic effect on
both the excited-state triplet energy and on the ground-state electro-
chemical potential, leading to a large net dependence of
photooxidant strength on the identity of the catalyst counteranion.
These results are consistent with the experimentally observed effect
–
for a representative series of Ru(btfmb)32+ complexes bearing BArF
4
, PF6 , ArFSO3 , and TsO– counteranions. These data are depicted in
Figure 2B and C. First, we obtained spectral data for these complexes
in acetonitrile (Figure 2B). Both the absorption and emission spec-
tra are superimposable in MeCN, consistent with the attenuated im-
pact of ion pairing in high-dielectric solvents. On the other hand,
Coulombic effects are more significant in non-polar solvents, which
–
–
of
counteranion
identity
on
the
radical
cation
Diels–Alder reaction described above. The most non-coordinating
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 8
counteranion (BArF –) results in the largest driving force for pho-
toinduced electron transfer, consistent with a faster rate of photoin-
itiation and a shorter reaction time.
In the radical cation Diels–Alder reaction, the radical cation inter-
mediates (4•+ and 6•+) would also be expected to exist as ion pairs,
and the most reasonable counteranion would be that introduced by
the photocatalyst.16 We wondered if this ion-pairing interaction
might also affect the dynamics of the product-forming cycloaddition
and chain propagation steps as well as the photoinitiation step. To
investigate this question, we utilized the same protocol we previ-
ously described for estimating the chain length in radical cation cy-
4
1
2
3
4
5
6
7
8
Table 3. Ground- and Excited-State Redox Potentials for
Ru(btfmb)32+ in CH2Cl2. a
/+
entry
catalyst
3a
E(Ru2+/+
–0.65 V
–0.89 V
–0.93 V
–0.95 V
)
E(Ru2+
*
)
ΔGES
1
2
3
4
2.17 eV
2.05 eV
2.01 eV
1.99 eV
+1.52 V
+1.16 V
+1.08 V
+1.04 V
cloadditions.10 First, we measured the reaction quantum yield with
–
the BArF – and ArFSO3 catalysts through chemical actinometry (see
3c
4
9
Supporting Information). The quantum yield using the BArF – cata-
3d
4
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
lyst 3a (1 mol%) was measured to be F = 26, comparable to the
3e
value we determined for the corresponding [Ru(bpz)3](BArF )2 cat-
4
aElectrochemical potentials were measured through cyclic voltamme-
try in a standard three-electrode set-up, scan rate = 100 mV/s. A 100
mM solution of a n-Bu4N+X– salt that matched the photocatalyst coun-
teranion was used as a supporting electrolyte. Potentials were corrected
to SCE through an external ferrocene reference.
–
alyst in previous studies.10 However, when ArFSO3 complex 3d was
utilized as the photocatalyst, we measured a significantly decreased
quantum yield of F = 0.35. To correct for the differing efficiency of
photoinitiation by photocatalysts with different excited-state oxida-
tion potentials, we divided the measured quantum yield values by
the quenching fraction. The quenching of 3a by anethole was highly
efficient (Q > 0.99), and thus the estimated average chain length is
the same as the quantum yield (CL = 26). The quenching fraction
for 3d was lower (Q = 0.83), but the resulting average chain length
was still calculated to be quite low (CL = 0.42). Thus, in addition to
influencing the ground and excited states of the photocatalyst, the
counteranion impacts the efficiency of the subsequent radical chain
reaction.
In order to rationalize the impact of counteranion identity on the
2+
triplet excited-state energy of the Ru(btfmb)3 chromophore, we
propose an explanation based upon an empirical physical model for
charge redistribution between the ground and electronically excited
states of this canonical class of transition metal photocatalysts
2+
(Scheme 2). The ground state of the Ru(btfmb)3 chromophore
has D3 symmetry and consequently cannot support a permanent di-
pole moment. On the other hand, the emissive states of Ru(II)*
tris(bipyridyl) complexes are understood to be metal-to-ligand
charge-transfer (MLCT) triplets, and considerable experimental ev-
idence supports the contention that the transferred electron is local-
ized to a single ligand without significant delocalization across the
These results have several significant implications. First, coun-
teranion identity is a previously underappreciated variable in the op-
timization of organic photoredox reactions that has the potential to
dramatically impact the success and efficiency of synthetically useful
organic reactions. Second, counteranion effects impact multiple as-
pects of the photocatalytic mechanism, including the energy of the
2+
other two ligands.19 Thus, electronically excited Ru*(btfmb)3 is
best
conceptualized
as
a
C2-symmetric,
charge-separated state with an oxidized Ru(III) core and a single re-
duced btfmb•– ligand (Scheme 2). This lower-symmetry MLCT
state would therefore be expected to have a very large dipole mo-
ment. Meyer has estimated the dipole moment of the triplet
reactive
triplet
excited
state,
the
rate
of
the
electron-transfer photoinitiation event, and the dynamics of non-
photochemical product-forming radical chain propagation events.
Finally, because the strength of ion-pairing interactions is sensitive
to solvent dielectric, these counterion effects are expected to be most
important in the relatively non-polar organic solvents that are often
optimal for synthetic applications. Such effects may be important in
a much wider range of organic photoredox reactions than previously
appreciated.
2+
Ru*(bpy)3 state to be approximately 14 D.20 If the photocatalyst
exists largely in an ion-paired state in non-polar solvents, stabilizing
charge–dipole interactions should have a larger effect on the triplet
excited state than they do on the ground state. One would further
expect that more strongly coordinating anions, which produce
tighter ion pairs, would better stabilize the triplet excited state. Fi-
nally, a strong solvent dependence would be consistent with this
model, as charge–dipole interactions are attenuated by increasing
solvent dielectric.
Hydrogen-Bonding Anion Binders as Co-Catalysts
The model proposed above suggests that the rate of photocata-
lytic Diels–Alder cycloaddition is strongly influenced by Coulombic
interactions between the counteranion and both the cationic photo-
catalyst and the radical cation intermediates. As a further test of this
model, we hypothesized that other strategies for disrupting ion pair-
ing might be used to exert a similar effect. In particular, we drew in-
spiration from a concept pioneered by Jacobsen: hydrogen-bonding
organocatalysts can accelerate reactions involving various cationic
Scheme 2. Representation of the ground state (no dipole) and tri-
plet excited state (significant dipole) for [Ru(btfmb)3](X)2.
CF3
CF3
F3
C
F3C
N
N
CF3
CF3
CF3
CF3
reactive intermediates by binding their associated counteranions.21,
N
N
N
N
N
N
N
N
22
hν
RuII
N
RuIII
N
We hypothesized that the tight ion pairing between a Lewis basic
F3
C
F3C
2+
counteranion and Ru*(btfmb)3 could be disrupted by addition of
an appropriate hydrogen-bonding anion binder, recapitulating the
rate increases we observed using weakly coordinating counterani-
ons.
CF3
CF3
D3 symmetry
no permanent dipole
C2 symmetry
large dipole moment
Our investigations focused on the use of sulfonate complex 3d as
a photoredox catalyst for the radical cation Diels–Alder reaction. As
ACS Paragon Plus Environment

Page 5 of 8
Journal of the American Chemical Society
described previously, 3d is markedly less reactive than the optimal
BArF – complex 3a. Several recent reports have shown that C3v-sym-
1
2
3
4
4
metric thiophosphotriamide 7 is an effective hydrogen-bond donor
for binding sulfonate anions,23 and we imagined that the sequestra-
–
tion of the ArFSO3 counteranion by 7 might attenuate its propensity
to participate in tight ion-pairing interactions. To test this hypothe-
sis, we conducted Diels–Alder cycloadditions using 1 mol% of 3d in
the presence and absence of 7 (Table 4). These experiments
showed a large rate increase for the Diels–Alder cycloaddition upon
addition of just 20 mol% of 7. Under these conditions, the reaction
is complete within 2 h, while only 5% yield of 6 is formed at the same
timepoint in the absence of the anion binder. Notably, there is no
observable formation of cycloadduct upon irradiation in the pres-
ence of 7 without photocatalyst 3d. This demonstrates that the
thiophosphotriamide is not photocatalytically active, and the im-
provement in photoredox activity thus arises from a synergistic co-
catalytic effect.
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 4. Effect of Ion-binder Co-catalyst 7 on Radical Cation
Diels–Alder reaction.
CF3
CF3
S
P
CF3
N
H
CF3
F3C
N
H
N
H
7
CF3
ion binder 7
[Ru(btfmb)3](Ar FSO3)2
MeO
MeO
+
Figure 3. (A) Effect of anion-binding co-catalyst 7 on photolumines-
cence of 3d. (B) Stern–Volmer plot in the absence and presence of ion
binder 7.
CH2Cl2
23 W CFL
Me
Me
Me
6
Me
4
5
entry
1
[Ru]
H-bond
time
2 h
yield 6
cat 3d
cat 7
The influence of added thiophosphotriamide 7 replicated the ef-
fect of weakly coordinating anions in other regards as well. First,
Stern–Volmer analysis indicates that the rate of quenching of 3d by
anethole is significantly faster upon the addition of the thiophos-
photriamide (Figure 3B). This is consistent with the observed in-
crease in triplet excited-state energy with greater concentrations of
7. Second, we observed a substantial effect on the radical cation
chain length (see Supporting Information). The addition of 20
mol% of 7 yielded a 20-fold increase in both the quantum yield and
apparent radical chain length of the reaction, suggesting that anion
binder 7 can influence the dynamics of the chain process by disrupt-
ing ion pairing.
1 mol%
20 mol%
99%
2
3
1 mol%
none
none
2 h
2 h
5%
0%
20 mol%
We also investigated whether co-catalyst 7 had an influence on the
–
photophysical properties of ArFSO3 photocatalyst 3d consistent
with our proposed model (Figure 3A). The addition of 7 to 3d in-
duced a large hypsochromic shift in the photoluminescence maxi-
mum. This shift was in the direction of the emission maximum of
–
BArF catalyst 3a, consistent with the expectation that added 7
4
would decrease the extent of ion pairing. In contrast, the addition of
Thus, we have been able to recapitulate the observed effect of
7 to the BArF – complex 3a yielded no change in the emission maxi-
–
non-coordinating anions on the photocatalytic activity of ArFSO3
4
mum, even at 50-fold excess of 7 relative to 3a. This experiment sup-
ports the contention that the effect arises from a specific interaction
between the thiophosphotriamide and the sulfate counteranion, ra-
ther than an interaction with some other component of the reaction
mixture or a general medium effect. To further support this conten-
tion, we investigated interaction in CD2Cl2 using 1H NMR spectros-
catalyst 3d using hydrogen-bonding co-catalyst 7. The thiophos-
photriamide disrupts ion pairing by binding the sulfonate coun-
teranion, which results in significant increases to the photocatalyst’s
triplet excited-state energy, the rate of photoinduced electron trans-
fer, the length of the radical cation chain process, and the overall ef-
ficiency of the photocatalytic Diels–Alder process. These results
support the contention that Coulombic effects can be more signifi-
cant in photoreactions than previously appreciated. Moreover, these
results suggest that the use of anion-binding organocatalysts could
be a conceptually orthogonal strategy for optimization of the grow-
ing class of synthetically useful photoredox transformations.
–
copy. Titration of thiophosphotriamide 7 with n-Bu4N+ ArFSO3 re-
sulted in a significant shift of the aromatic C–H resonances of 7. A
fit of these data to a 1:1 binding model provided an association con-
stant of 1.3 x 106.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 8
(Supporting Information. General experimental preparation, synthe-
Conclusion
1
2
3
4
5
6
7
8
sis of Ru(II) polypyridyl catalysts, reaction conditions, UV-Vis and pho-
toluminescence spectroscopy, cyclic voltammetry, and NMR character-
ization. The Supporting Information is available free of charge on the
ACS Publications website.
The studies summarized above suggest several important implica-
tions for the design, understanding, and optimization of photocata-
lytic processes. First, counterion effects can exert a significant im-
pact on the observed rate of radical cation reactions initiated by pho-
toredox catalysis. The degree of ion pairing between the coun-
teranion and both the Ru(II) photoredox catalyst and the oxidized
radical cation intermediate can influence the efficiency of multiple
steps in the mechanism of these reactions. Thus, these studies indi-
cate that modulating the degree of ion pairing is an important unex-
plored variable in the optimization of this class of transformations,
and we have described two complementary approaches that can suc-
cessfully increase the overall rate of a radical cation cycloaddition by
several orders of magnitude. Second, the Coulombic interactions
that are the putative origin of these effects are most significant in rel-
atively non-polar solvents such as those that are often optimal for
synthetic applications. This could indicate that counteranion iden-
tity is a particularly important variable for organic photoredox reac-
tions compared to better-established applications of Ru(II) photo-
redox catalysts in solar energy conversion and biology, where the use
of water or other high-dielectric solvents might mask the impact of
ion pairing. As interest in the use of this class of transition metal pho-
toredox catalyst for synthetic applications continues to grow, a
deeper understanding of the impact of ion-pairing effects will be crit-
ical to developing a complete, detailed understanding of the mecha-
nisms in this class of synthetically useful transformations.
AUTHOR INFORMATION
Corresponding Author
* E-mail: tyoon@chem.wisc.edu
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Notes
ORCID:
Steven J. Chapman: 0000-0003-1570-4580
Wesley B. Swords: 0000-0002-2986-326X
Marco D. Torelli: 0000-0002-7691-6491
Robert J. Hamers: 0000-0003-3821-9625
Tehshik P. Yoon: 0000-0002-3934-4973
ACKNOWLEDGMENT
This work was supported by the NIH (GM095666). The NMR and
mass spectroscopy facilities at UW–Madison are funded in part by NIH
(S10OD020022-1), NSF (CHE-1048642), the University of Wiscon-
sin, and a generous gift from Paul J. and Margaret M. Bender.
REFERENCES
ASSOCIATED CONTENT
1
(a) Kalyanasundaram, K. “Photophysics, Photochemistry and Solar En-
ergy Conversion with Tris(bipyridyl)ruthenium(II) and Its Analogues,”
Coord. Chem. Rev. 1982, 46, 159–244. (b) Roundhill, D. M. “Photochemis-
try, Photophysics, and Photoredox Reactions of Ru(bpy)3 and Related
2+
Complexes,” in Modern Inorganic Chemistry: Photochemistry and Photophys-
ics of Metal Complexes, Springer: New York, 1994, pp. 165–215.
2
Lehn, J. M.; Sauvage, J. P. “Chemical Storage of Light Energy. Catalytic
Generation of Hydrogen by Visible Light or Sunlight. Irradiation of Neutral
Aqueous Solutions,” Nouv. J. Chim. 1977, 1, 449–451.
3
O’Regan, B.; Grätzel, M. “A Low-Cost, High-Efficiency Solar Cell Based
on Dye-Sensitized Colloidal TiO2 Films,” Nature 1991, 353, 737–740.
4
For reviews, see: a) Narayanam, J. M. R.; Stephenson, C. R. J. “Visible
Light Photoredox Catalysis: Applications in Organic Synthesis,” Chem. Soc.
Rev. 2011, 40, 102–113. b) Prier, C. K.; Rankic, D. A.; MacMillan, D. W. C.
“Visible Light Photoredox Catalysis with Transition Metal Complexes: Ap-
plications in Organic Synthesis,”Chem. Rev. 2013, 113, 5322–5363. c)
Romero, N. A.; Nicewicz, D. A. "Organic Photoredox Catalysis,” Chem. Rev.
2016, 116, 10075–10166. d) Twilton, J.; Le, C.; Zhang, P.; Shaw, M. H.; Ev-
ans, R. W.; MacMillan, D. W. C. “The Merger of Transition Metal and Pho-
tocatalysis,” Nature Rev. Chem. 2017, 1, 0052. e) Zou, Y. Q.; Hörmann, F.
M.; Bach, T. “Iminium and Enamine Catalysis in Enantioselective Photo-
chemical Reactions,” Chem. Soc. Rev. 2018, 47, 278–290. f) Silvi, M.; Mel-
chiorre, P. “Enhancing the Potential of Enantioselective Organocatalysis
with Light,” Nature 2018, 554, 41–49.
5
Juris, A.; Balzani, V.; Barigelletti, F.; Campagna, S.; Belser, P.; von Zelew-
sky, A. "Ru(II) Polypyridine Complexes: Photophysics, Photochemistry,
Electrochemistry, and Chemiluminescence," Coord. Chem. Rev. 1988, 84,
85–277.
6
(a) Lin, S.; Ischay, M. A.; Fry, C. G.; Yoon, T. P. "Radical Cation Diels-
Alder Cycloadditions by Visible Light Photocatalysis," J. Am. Chem. Soc.
2011, 133, 19350–19353. (b) Lin, S.; Padilla, C. E.; Ischay, M. A.; Yoon, T.
ACS Paragon Plus Environment

Page 7 of 8
Journal of the American Chemical Society
H.; Li, S; Zhang, M.–T.; Luo, S. “A Chiral Ion-Pair Photoredox Organocat-
alyst: Enantioselective Anti-Markovnikov Hydroetherification of Alkenols,”
Org. Chem. Front., 2017, 4, 1037–1041. (c) Wang, H.; Ren, Y.; Wang, K.;
Man, Y.; Xiang, Y.; Li, N.; Tang, B. “Visible Light-Induced Cyclization Re-
actions for the Synthesis of 1,2,4–Triazolines and 1,2,4–Triazoles,” Chem.
Commun., 2017, 53, 9644–9647. (d) Gentry, E. C.; Rono, L. J.; Hale, M. E.;
Matsuura, R.; Knowles, R. R. “Enantioselective Synthesis of Pyrroloin-
dolines via Noncovalent Stabilization of Indole Radical Cations and Appli-
cations to the Synthesis of Alkaloid Natural Products,” J. Am. Chem. Soc.,
2018, 140, 3394–3402.
1
2
3
4
5
6
7
8
P. "Visible Light Photocatalysis of Intramolecular Radical Cation Diels–Al-
der Cycloadditions," Tetrahedron Lett. 2012, 53, 3073–3076. (c) Lin, S. S.;
Lies, S. D.; Gravatt, C. S.; Yoon, T. P. “Radical Cation Cycloadditions Using
Cleavable Redox Auxiliaries,” Org. Lett. 2017, 19, 368–371.
7
(a) Crutchley, R. J.; Lever, A. B. P. “Ruthenium(II) Tris(bipyrazyl) Di-
cation – A New Photocatalyst,” J. Am. Chem. Soc. 1980, 102, 7128–7129. (b)
Rillema, D. P.; Allen, G.; Meyer, T. J.; Conrad, D. “Redox Properties of Ru-
thenium(II) Tris Chelate Complexes Containing the Ligands 2,2'-bipyra-
zine, 2,2'-bipyridine, and 2,2'-bipyrimidine,” Inorg. Chem. 1983, 22, 1617–
1622.
17
Adamson, Q. W.; Namnath, J.; Shastry, V. J.; Slawson, V. “Thermody-
9
namic Inefficiency of Conversion of Solar Energy to Work,” J. Chem. Ed.
1984, 61, 221.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
8
Several other photocatalytic radical cation Diels–Alder reactions have
18
Flamigni, L.; Barbieri, A.; Sabatini, C.; Ventura, B.; Barigelletii, F. “Pho-
been reported using alternate photocatalysts. For leading references, see: (a)
Stevenson, S. M.; Shores, M. P.; Ferreira, E. M. “Photooxidizing Chromium
Catalysts for Promoting Radical Cation Cycloadditions,” Angew. Chem. Int.
Ed. 2015, 54, 6506–6510. (b) Pitre, S. P.; Scaiano, J. C.; Yoon, T. P. “Photo-
catalytic Indole Diels–Alder Cycloadditions Mediated by Heterogenous
Platinum-Modified Titanium Dioxide,” ACS Catal. 2017, 7, 6440–6444. (c)
Gieseler, A.; Steckhan, E.; Wiest, O; Knoch, F. “Photochemically Induced
Radical Cation Diels–Alder Reaction of Indole and Electron-Rich Dienes,”
J. Org. Chem. 1991, 56, 1405–1411.
tochemistry and Photophysics of Coordination Compounds: Iridium,” Top.
Curr. Chem. 2007, 281, 143–203.
19
(a) Forster, M.; Hester, R. E. “Resonance Raman Investigation of Elec-
2+
tronically Excited Ru(bipyridine)3 Using a CW Laser,” Chem. Phys. Lett.
1981, 81, 42–47. (b) Bradley, P. G.; Kress, N.; Hornberger, B. A.; Dallinger,
R. F.; Woodruff, W. H. “Vibrational Spectroscopy of the Electronically Ex-
cited State. 5. Time-Resolved Resonance Raman Study of Tris(bipyridine)
Ruthenium(II) and Related Complexes. Definitive Evidence for the Local-
ized MLCT State,” J. Am. Chem. Soc. 1981, 103, 7441–7446. (c) Ceulemans,
A.; Vanquickenborne, L. G. “On the Charge-Transfer Spectra of Iron(II)-
and Ruthenium(II)-Tris(2,2'-bipyridyl) Complexes,” J. Am. Chem. Soc.
1981, 103, 2238–2241. (d) Ohsawa, Y.; DeArmond, M. K.; Hanck, K. W.;
Morris, D. E.; Whitten, D. G.; Neveux, P. E. “Spatially Isolated Redox Orbit-
als: Evidence from Low-Temperature Voltammetry,” J. Am. Chem. Soc.
1983, 105, 6522–6524.
9
Bauld, N. L. “Cation Radical Cycloadditions and Related Sigmatropic
Reactions,” Tetrahedron 1989, 45, 5307–5363.
10
Cismesia, M. A.; Yoon, T. P. “Characterizing Chain Processes in Visible
Light Photoredox Catalysis,” Chem. Sci. 2015, 6, 5426–5434.
11
Furue, M.; Maruyama, K.; Oguni, T.; Naiki, M.; Kamachi, M. “Trifluo-
romethyl-substituted 2,2'-bipyridine Ligands. Synthetic Control of Excited-
State Properties of Ruthenium(II) Tris-Chelate Complexes,” Inorg. Chem.
1992, 31, 3792–3795.
20
Kober, E. M.; Sullivan, B. P.; Meyer, T. J. “Solvent Dependence of Metal-
to-Ligand Charge-Transfer Transitions. Evidence for Initial Electron Local-
ization in MLCT Excited States of 2,2’-Bipyridine Complexes of Ruthe-
nium(II) and Osmium(II),” Inorg. Chem. 1984, 23, 2098–2104.
12
Strauss, S. H. “The Search for Larger and More Weakly Coordinating
Anions” Chem. Rev. 1993, 93, 927.
13
(a) Ischay, M. A.; Ament, M. S.; Yoon, T. P. “Crossed Intermolecular
21
(a) Reisman, S. E.; Doyle, A. G.; Jacobsen, E. N. “Enantioselective Thi-
[2+2] Cycloaddition of Styrenes by Visible Light Photocatalysis,” Chem. Sci.
2012, 2046–2050. (b) Douglas, J. J.; Nguyen, J. D.; Cole, K. P.; Stephenson,
C. R. J. “Enabling Novel Photoredox Reactivity via Photocatalyst Selection,”
Aldrichim. Acta 2014, 47, 15–25.
ourea-Catalyzed Additions to Oxocarbenium Ions,” J. Am. Chem. Soc. 2008,
130, 7198–7199. (b) Birrell, J. A.; Desrosiers, J.-N.; Jacobsen, E. N. “Enanti-
oselective Acylation of Silyl Ketene Acetals through Fluoride Anion-Binding
Catalysis,” J. Am. Chem. Soc. 2011, 133, 13872–13875. (c) Wasa, M.; Liu, R.
Y.; Roche, S. P.; Jacobsen, E. N. “Asymmetric Mannich Synthesis of α-Amino
Esters by Anion-Binding Catalysis,” J. Am. Chem. Soc. 2014, 136, 12872–
12875.
14
(a) Ward, W. M.; Farnum, B. H.; Siegler, M.; Meyer, G. J. “Chloride Ion-
Pairing with Ru(II) Polypyridyl Compounds in Dichloromethane,” J. Phys.
Chem. A 2013, 117, 8883–8894. (b) Wehlin, S. A. M.; Troian-Gautier, L.;
Li, G.; Meyer, G. J. “Chloride Oxidation by Ruthenium Excited-States in So-
lution,” J. Am. Chem. Soc., 2017, 139, 12903–12906. (c) Troian-Gautier, L.;
Wehlin, S. A. M.; Meyer, G. J. “Photophysical Properties of Tetracationic
Ruthenium Complexes and Their Ter-Ionic Assemblies with Chloride,” In-
org. Chem., 2018, 57, 12232–12244.
22
Gansäuer has recently reported the use of hydrogen bond-donating co-
catalysts to influence the rate of titanocene-mediated electron transfer reac-
tions. See: Liedtke, T.; Spannring, P.; Riccardi, L.; Gansäuer, A. “Mecha-
nism-Based Condition Screening for Sustainable Catalysis in Single-Elec-
tron Steps by Cyclic Voltammetry,” Angew. Chem. Int. Ed. 2018, 57, 5006–
5010.
1
15
The chemical shifts associated with the ligands in the H NMR of the
photocatalysts vary as a function of counteranion identity and are consistent
with observations made by Meyer. These data are summarized in the Sup-
porting Information. See: Li, G.; Swords, W. B.; Meyer, G. J. “Bromide
Photo-oxidation Sensitized to Visible light in Consecutive Ion Pairs,” J. Am.
Chem. Soc. 2017, 139, 14983–14991.
23
(a) Cranwell, P. B.; Hiscock, R.; Haynes, C. J. E.; Light, M. E.; Wells, N.
J.; Gale, P. A. “Anion Recognition and Transport Properties of Sulfamide-,
Phosphoric Triamide- and Thiophosphoric Triamide-Based Receptors,”
Chem. Commun. 2013, 49, 874–876. (b) Borovika, A.; Tang, P. I.; Klapman,
S.; Nagorny, P. “Thiophosphoramide-Based Cooperative Catalysts for
Brønsted Acid Promoted Ionic Diels–Alder Reactions,” Angew. Chem. Int.
Ed. 2013, 52, 13424–13428.
16
Four recent reports have described the influence of chiral counteranions
on the enantioselectivity of photocatalytic radical cation reactions: (a)
Morse, P. D.; Nguyen, T. M.; Cruz, C. L.; Nicewicz, D. A. “Enantioselective
Counter-Anions in Photoredox Catalysis: The Asymmetric Cation Radical
Diels–Alder Reaction,” Tetrahedron 2018, 74, 3266–3272. (b) Yang, Z.; Li,
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 8
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
+
–
Wavelength (nm)
Potential
8
ACS Paragon Plus Environment