Extended Hydrogen Bond Networks for Effective Proton-Coupled Electron Transfer (PCET) Reactions: The Unexpected Role of Thiophenol and Its Acidic Channel in Photocatalytic Hydroamidations

Article abstract of DOI:10.1021/jacs.0c08673

Preorganization and aggregation in photoredox catalysis can significantly affect reactivities or selectivities but are often neglected in synthetic and mechanistic studies, since the averaging effect of flexible ensembles can effectively hide the key activation signatures. In addition, aggregation effects are often overlooked due to highly diluted samples used in many UV studies. One prominent example is Knowles's acceleration effect of thiophenol in proton-coupled electron transfer mediated hydroamidations, for which mainly radical properties were discussed. Here, cooperative reactivity enhancements of thiophenol/disulfide mixtures reveal the importance of H-bond networks. For the first time an in-depth NMR spectroscopic aggregation and H-bond analysis of donor and acceptor combined with MD simulations was performed revealing that thiophenol acts also as an acid. The formed phosphate-H+-phosphate dimers provide an extended H-bond network with amides allowing a productive regeneration of the photocatalyst to become effective. The radical and acidic properties of PhSH were substituted by Ph2S2 and phosphoric acid. This provides a handle for optimization of radical and ionic channels and yields accelerations up to 1 order of magnitude under synthetic conditions. Reaction profiles with different light intensities unveil photogenerated amidyl radical reservoirs lasting over minutes, substantiating the positive effect of the H-bond network prior to radical cyclization. We expect the presented concepts of effective activation via H-bond networks and the reactivity improvement via the separation of ionic and radical channels to be generally applicable in photoredox catalysis. In addition, this study shows that control of aggregates and ensembles will be a key to future photocatalysis.

Full text of DOI:10.1021/jacs.0c08673

pubs.acs.org/JACS
Article
Extended Hydrogen Bond Networks for Effective Proton-Coupled
Electron Transfer (PCET) Reactions: The Unexpected Role of
Thiophenol and Its Acidic Channel in Photocatalytic
Hydroamidations
Nele Berg, Sebastian Bergwinkl, Patrick Nuernberger, Dominik Horinek, and Ruth M. Gschwind*
Cite This: J. Am. Chem. Soc. 2021, 143, 724−735
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Preorganization and aggregation in photoredox catalysis can
significantly affect reactivities or selectivities but are often neglected in synthetic
and mechanistic studies, since the averaging effect of flexible ensembles can
effectively hide the key activation signatures. In addition, aggregation effects are often
overlooked due to highly diluted samples used in many UV studies. One prominent
example is Knowles’s acceleration effect of thiophenol in proton-coupled electron
transfer mediated hydroamidations, for which mainly radical properties were
discussed. Here, cooperative reactivity enhancements of thiophenol/disulfide
mixtures reveal the importance of H-bond networks. For the first time an in-depth
NMR spectroscopic aggregation and H-bond analysis of donor and acceptor
combined with MD simulations was performed revealing that thiophenol acts also as
an acid. The formed phosphate-H+-phosphate dimers provide an extended H-bond
network with amides allowing a productive regeneration of the photocatalyst to
become effective. The radical and acidic properties of PhSH were substituted by Ph2S2 and phosphoric acid. This provides a handle
for optimization of radical and ionic channels and yields accelerations up to 1 order of magnitude under synthetic conditions.
Reaction profiles with different light intensities unveil photogenerated amidyl radical reservoirs lasting over minutes, substantiating
the positive effect of the H-bond network prior to radical cyclization. We expect the presented concepts of effective activation via H-
bond networks and the reactivity improvement via the separation of ionic and radical channels to be generally applicable in
photoredox catalysis. In addition, this study shows that control of aggregates and ensembles will be a key to future photocatalysis.
compounds can be fine-tuned independently allowing the
activation of strong bonds which are not easily accessible via
conventional redox processes.36−38,47
INTRODUCTION
■
Light dependent as well as light independent proton-coupled
electron transfer (PCET) reactions play an essential role in
biological systems,1−13 e.g. for the O2 generation in the
photosystem II.14 Their potential was also successfully applied
in electrochemistry,2,15−24 transition metal chemistry,1,2528 and
photochemical transformations.1,2,29−45 In synthetic organic
chemistry the activation of strong covalent bonds, such as C−
H, O−H, and N−H, is a very challenging task to access
important synthetic building blocks. To overcome their high
energy barriers in an atom economic and environmentally
friendly manner, the combination of PCET and photocatalysis
paved the way to an efficient reaction design. Especially, the
photocatalytic concerted multisite PCET (MS-PCET) was
found to be successful.2,36,46 There, a decoupled proton and
electron transfer event in one elementary step omits highly
energetic intermediates, which usually arise in conventional
hydrogen atom transfer (HAT). In the oxidative MS-PCET, a
proton and an electron originating from the same donor bond
are transferred to a base and to a separate oxidant.1,2,18 As a
result, the proton and electron affinities of the two acceptor
Thus, in the field of photocatalytic organic chemistry, using
PCET, the group of Knowles transformed ketones into cyclic
alcohols,34,42,48 activated N−H bonds in extensive hydro-/
carboamination and hydro-/carboamidation protocols, opened
cyclic alcohols via PCET mediated alkoxy radical gener-
ation,31,32,35,41,43,44,49 and succeeded even in the alkylation of
remote C−H bonds via an amide PCET process.33 More
recently, they developed an aliphatic C−H bond activation50
and a PCET driven NH3 synthesis.51 Even for the
deracemization of ureas, PCET is involved.52 Further coupling
reactions were recently designed including an aldehyde−olefin
Received: August 12, 2020
Published: January 11, 2021
https://dx.doi.org/10.1021/jacs.0c08673
J. Am. Chem. Soc. 2021, 143, 724−735
© 2021 American Chemical Society
724

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
coupling via reductive ketyl radical formation,29 the generation
of pyridine based heterocycles,53 and the copper-based
photocatalytic coupling of ketones.54 Moreover, the PCET
concept was employed in visible light mediated selective C−C
bond cleavage of lignines55 and the C−H functionalization of
cyclic ethers.56
In such light driven oxidative PCET reactions the formation
of a D−H···A hydrogen bond (H-bond) between the D−H
bond to be cleaved and the applied base A (Figure 1a) is stated
formation of a H-bond was even found to suppress a HAT
mechanism due to steric reasons.62 As expected, charge
assisted H-bonds and solvents promoting strong H-bonds
accelerate charge transfer events for an efficient PCET
process.63,64 Moreover, using IR spectroscopy, H-bond
equilibrium constants were determined for a TEMPOH/
pyridine complex.65
Computational studies on model systems in biology,
biochemistry, solution chemistry, and electrochemistry corro-
borated the essential role of H-bond formation for PCET, e.g.
enhancing the vibronic donor−acceptor coupling.3,12,17,18,66,67
For example, the PCET driven aerobic respiratory chain3,17
and effects of antioxidants68−70 were shown to occur via the
formation of H-bond mediated aggregates.
In contrast, experimental studies are scarce in the field of
photocatalytic PCET reactions. Via EPR spectroscopy, a strong
H- bonded complex between histidine and a benzoquinone
derivative (TolSQ) was detected.71 Furthermore, combined
NMR, transient absorption, and emission spectroscopic
investigations demonstrated a correlation between H-bond
strengths and charge transfer kinetics in model systems, using
exclusively 1H chemical shift changes for the H-bond
analysis.63,64 Later, by applying a combination of H NMR
1
spectroscopy and cyclic voltammetry, this trend was further
corroborated for an electrochemical reaction.72 Via H NMR
1
titration experiments a phenol−pyridine interaction was
identified.73 Knowles and Alexanian even examined a non-
covalent interaction between the iridium photocatalyst and a
phosphate base to be responsible for successful C−H bond
activation.50
One prominent example deviating from the general
mechanistic concept of concerted MS-PCET is the accelerating
effect of thiophenol in the oxidative photocatalytic hydro-
amidation of phenylamides developed by Knowles and co-
workers (Figure 1b).32 In the proposed mechanism, the excited
state of the photocatalyst acts as electron acceptor and a
phosphate base as the proton acceptor for the activation of the
amide substrate in a H-bonded amide/base preaggregate,
followed by the intramolecular addition of the formed amidyl
radical to the olefin. In a final HAT step involving thiophenol
(PhSH) as the hydrogen atom donor, the lactam product is
furnished in up to 95% yield; however phenol (PhOH), which
is a commonly known HAT transfer reagent, was unproduc-
tive.32 PhSH as the HAT donor was applied in other synthetic
protocols,74−76 and other groups showed that the S−H bond
can be directly activated via PCET reactions.77−79 The
chemoselectivity factors for the selective cleavage of the
stronger amide N−H group was recently explained by Knowles
via a combination of H-bond networks and PCET kinetics.80
However, the luminescence quenching experiments in these
studies were conducted at highly diluted concentrations not
addressing the complex aggregation and H-bond networks
expected for ion pairs and amides in apolar organic
solvents.32,80 An in-depth kinetic study of Nocera revealed
the importance of the overall performance of the PCET/back-
electron transfer (BET) equilibria for quantum efficiency in
these hydroamidations.81 By adding disulfides they allowed for
an off-cycle equilibrium, could reduce the BET rate, and
increased the concentration of the key amidyl radical available
for the downstream reaction. However, both studies neglect
the complex H-bond situation present at synthetic conditions
and its potential to influence the PCET, BET, and
corresponding back-proton transfer (BPT). Here, the central
Figure 1. (a) An H-bonded preaggregate is proposed as the key
interaction for a selective cleavage of strong bonds via PCET in
literature. (b) The photocatalytic hydroamidation of phenylamides
developed by Knowles was chosen as the model system. Its postulated
light driven mechanism involves an excited state iridium catalyst and a
phosphate base for the crucial PCET step, thiophenol as the HAT
donor and a highly effective unproductive BET. The N−H cleavage in
the presence of weaker S−H bonds is surprising concerning the
BDFEs (see main text). (c) Our studies reveal the potential to
selectively access the radical and ionic properties of PhSH by the
formation of extended H-bonded aggregates allowing for a productive
back oxidation of Ir(II) prior to radical cyclization.
to be responsible for lowering the reaction energy
barrier.1,2,37,36,47,51−58 Especially, the proton transfer requires
an overlap of the vibronic states and thus a minimized donor−
acceptor distance. Hence, the formation of strong donor−
acceptor H-bonded preaggregates is proposed for the initiation
of PCET.1,2,46 Furthermore, the geometry and strength of the
H-bond seem to be essential as shown for competing
mechanisms, HAT versus PCET in benzyl/toluene pairs and
phenoxy/phenol aggregates.59,60 Internal H-bonds of the latter
aggregate were shown to accelerate the coupled electron
proton transfer rate in a pH independent manner.61 The
725
https://dx.doi.org/10.1021/jacs.0c08673
J. Am. Chem. Soc. 2021, 143, 724−735

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 2. (a) The in situ NMR kinetics revealed a drastic acceleration of the overall PCET efficiency for mixtures of PhSh/Ph2S2 hinting at a H-
bond effect in terms of cooperativity (top). The fast product formation (∼100% after ∼10 h) for the PhSH containing photocatalytic system and
very low yields using PhOH (∼1.2% after ∼10 h) were confirmed by in situ NMR kinetics, too (bottom). The trend of the H-bond donor chemical
shifts deviates from the PCET reactivities: (b) For the NMR spectroscopic H-bond investigations, 15N-phenylpent-4-enamide (amide),
tetrabutylammonium ditert-butylphosphate (base), thiophenol (PhSH) and phenol (PhOH) were used. The H-bond donor side was studied via
(c) 1H and (d) 15N NMR chemical shift and scalar coupling analysis in CD2Cl2 at 180 K and 1:1/1:1:1 mixtures were used for the multicomponent
samples. For the complete spectra, see SI. The amidyl 15N−H low field shift in both 1D spectra in the presence of base confirmed the existence of
an H-bond, which is weakened after adding PhSH. In contrast, PhSH is mainly free in solution as the δ-values are only barely affected. (e) A
1
significant H low field shift of the phenol OH proton indicates its incorporation into the H-bonded complex, but the amide chemical shifts are
similar to the PhSH containing samples.
question is the role and effect of thiophenol on this H-bond
network and the overall reaction rate.
hydroamidation protocol of Knowles and co-workers.32 For
the in situ reaction kinetics 3,3-dimethyl-N-phenylpent-4-
enamide was synthesized in order to exclude self-HAT. To
get access to the H-bond donor side, a 15N labeled N-
phenylpent-4-enamide was synthesized. As a proton acceptor
species, tetrabutylammonium di-tert-butylphosphate was used
which showed excellent reactivity inside the photocatalytic
reaction at room temperature (see SI chapter 3.1−3.4) and
enables simplification of the spectra due to a reduced amount
of NMR signals compared to the originally used n-butyl
phosphate base. For the H-bond and aggregation studies, 1:1
mixtures (50 mM) of amide and phosphate as well as 1:1:1
mixtures of amide, phosphate, and thiol or phenol were
prepared in dichloromethane-d2 (CD2Cl2) and measured at
300 and 180 K.82 Due to chemical exchange reduction at low
temperature, specific H-bonded aggregates are visible on the
NMR time scale.
Therefore, in this study first initial rate kinetics of mixtures
of PhSH and diphenyldisulfide (Ph2S2) were investigated.
These showed cooperativity effects beyond the additivity
expected for the mere radical effect and underpinned the
importance of the H-bond network. Next, for the first time an
in-depth NMR spectroscopic analysis of the donor and
acceptor side of the H-bonds of this PCET driven hydro-
1
amidation reaction was performed using low temperature H,
1
15N, 31P chemical shifts and JNH scalar couplings (Figure 1c).
These and diffusion measurements reveal an unexpected
modulation of the H-bonded aggregates by thiophenol.
Furthermore, these special aggregates were corroborated by
MD-simulations, which is to the best of our knowledge the first
time such a combined NMR/MD approach in chemical
photoredox catalysis has been undertaken. Last but not least,
PhSH was replaced by a combination of Ph2S2 and phosphoric
acid, which allowed the ionic and the radical properties to be
dissected and optimized separately. Luminescence quenching
studies and reaction profiles with different light intensities were
employed to dissect the various mechanistic steps and to reveal
a productive regeneration pathway of Ir(II) to Ir(III) being
affected by the extended H-bond network.
Furthermore, dichloromethane was applied by the group of
Knowles and is also suitable for our low temperature NMR
studies.31,32,83
Initial Rate Kinetics Reveal Importance of H-Bonds.
For phenol and thiophenols as additives, huge reactivity
differences32 were observed previously, which can be attributed
to their pronounced distinctions in radical and HAT
properties, potential catalyst inhibition, side reactions of
phenol, or unproductive PCET activation of phenol. Our
initial rate kinetics shown in Figure 2a (bottom) confirmed this
reactivity difference. For phenol, no catalyst inhibition by the
RESULTS AND DISCUSSION
■
Model System. The employed PCET model system in this
work is directly derived from the synthetic conditions of the
726
https://dx.doi.org/10.1021/jacs.0c08673
J. Am. Chem. Soc. 2021, 143, 724−735

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
product and only marginal byproduct formation were found
but significantly different H-bond networks were found. Thus,
both different radical properties and different H-bond
networks may contribute to dissimilar reactivities of phenol
and thiophenol. The radical/HAT properties of thiophenol
and phenol are so different that it is difficult to dissect the
contributions of radical properties and the H-bond network
between these two. Therefore, we investigated Ph2S2, PhSH,
and mixtures of both. In agreement with Nocera’s results, the
system with 1 equiv of disulfide shows a faster reaction rate
than that with 1 equiv of PhSH corroborating the importance
of the off-cycle radical equilibrium for a reduced BET (Figure
2a; top). Most astonishing however are the data of the
mixtures of Ph2S2 and PhSH (80:20 and 70:30). When
considering mere radical effects, reaction rates between those
of pure Ph2S2 and PhSH are expected in terms of additivity.
However, compared to pure Ph2S2, significantly higher reaction
rates, i.e. nonlinear effects, were detected. This indicates that
aggregation may play a pivotal role in this reaction type, which
is typical for complex H-bond networks and beyond mere
radical properties of isolated molecules. We interpreted this
special acceleration as a hint that H-bonds may really be
important and may affect the PCET, the BPT back reaction, or
other parts of the mechanistic cycle. The principal importance
of the H-bond binding constants for this reaction was already
outlined by Qiu and Knowles in terms of chemoselectivity.80
However, in their study highly diluted samples were used,
which study the interactions between single molecules but not
within the complex aggregates existing under synthetic
conditions (see DOSY studies below).
(12.34/138.65 ppm, 89.6/89.6 Hz) indicating a by far stronger
H-bond of the amide to the phosphate in accordance to their
H-bond donor and acceptor abilities (see SI chapter 4.4).90
The addition of PhSH slightly weakens the 15N−H···O−P H-
bond, which was identified by the reduction of the chemical
shifts and a small increase of the coupling constant (11.45/
137.91 ppm, 89.991/90.0 Hz). In contrast to the amide signals,
1
the H chemical shift changes of the thiol S−H proton with
and without base and amide are quite small (3.73−4.07 ppm;
see Figure 2c, orange signal) and, hence, denote minimal
1
interaction with the phosphate base or the amide. Via H,1H
NOESY experiments, we further confirmed that the average
amide−phosphate interaction is slightly reduced upon addition
of thiophenol (see SI chapter 4.6.3).
Next, the situation with PhOH as the HAT donor was
investigated, which showed a drastically reduced reactivity
compared to PhSH as the additive (see Figure 2a, bottom).
The phenolic O−H proton signal, which is located at 6.79 ppm
for pure phenol, is significantly shifted to higher ppm values
(12.26 ppm) in the presence of amide and base indicating that
phenol is incorporated in a complex network of considerably
strong H-bonds (Figure 2e). Simultaneously, the amidyl 15N−
H doublet shows in principle similar chemical shift and scalar
coupling values (11.84/137.95 ppm, 89.7/90.6 Hz) compared
1
to the amide/base/PhSH sample in both H and 15N spectra
even with a trend toward slightly stronger amide−phosphate
H-bonds. As a result, the classical H-bond strength analysis
1
using exclusively H chemical shifts as the sensor suggests a
similar H-bond activation of the amide in the presence of
PhSH and PhOH and fails to explain the huge reactivity
difference of these two HAT donors (see Figure 2a), which
previously was mainly attributed to their different radical and
HAT properties, potential catalyst inhibition, or side reactions
of phenol.80 The only detectable difference is the inclusion of
phenol into the H-bond network in contrast to mainly free
thiophenol. However, this simplified donor-only observation of
H-bonds can also hide the full situation.
H-Bond Analysis by NMR. Therefore, in-depth H-bond
studies were conducted next. In general, the formation of a H-
bond influences the electronic environment of both, the H-
bond donor and acceptor sites. The shift of the proton toward
the acceptor leads to a change in electron density distribution
by deshielding of donor and proton and shielding of the
acceptor.84−86 Thus, the strength of this noncovalent bond can
be read out via NMR chemical shifts and scalar couplings.83
Limbach and co-workers established a method for the
determination of the H-bond strengths in a biological system
H-Bond Acceptor and Phosphate Dimer Formation.
Therefore, next the H-bond acceptor side of these mixtures
was analyzed by 31P NMR measurements. According to the
using low temperature H and 15N chemical shift correla-
trend demonstrated via H and 15N NMR, the movement of
1
1
tions.83 This concept was already successfully applied in our
group for the analysis of H-bonds in Brønsted acid
catalysis.87,88 Moreover, we recently investigated the impor-
tance of H-bond mediated interactions for the selective
activation of strong single C−F bonds.89
the proton toward the base predicts a shielding of the acceptor
and therefore a high field shift of the 31P signal of the base.
Furthermore, the diminished H-bond interaction after adding
PhSH is expected to be reflected in a back shift to low field. As
shown in Figure 3a, the 31P signal of the pure phosphate anion
(−7.44 ppm; bottom) was shifted toward high field in the
presence of the amide (−8.19 ppm; above), which is in full
accordance with the previous results. However, upon addition
of PhSH to the pure base, the phosphate signal was even
further high field shifted to −8.69 ppm, which is in strong
H-Bond Donor and Reactivity Profile. First, the H-bond
1
donor side was analyzed via H and 15N NMR chemical shifts
and scalar coupling studies. An overview of the analyzed
compounds is depicted in Figure 2b. The formation of a H-
bond between amide and phosphate base is expected to go
hand in hand with a low field shift for both the amidyl proton
and nitrogen as well as a reduced 15N−H scalar coupling. In
1
contrast to the H spectra indicating only very weak H-bonds
to PhSH in all thiol containing samples (vide supra). Thus,
upon addition of PhSH another very strong H-bond has to be
created independently of the thiol proton itself. This is also
1
Figure 2c and 2d, the details of the H and 15N NMR spectra
of pure 15N-phenylpent-4-enamide (bottom), the 15N-amide/
base (middle), and the 15N-amide/base/thiophenol (top)
mixtures are shown. For the applied concentration of 50 mM,
the pure amide itself exists as an oligomer (for detailed studies,
see SI chapter 4.6.4) and the signals of the 15N−H
functionality at 8.45/134.49 ppm (89.9/91.3 Hz) refer
therefore to the amide−amide H-bond (Figure 2c,d). In the
presence of phosphate base, both amide proton and nitrogen
are drastically low field shifted and the coupling is reduced
1
corroborated by the appearance of a H signal at 16.43 ppm
(PhSH: 3.94 ppm) for all PhSH/base containing mixtures.
Since in our group dimers of chiral phosphoric acids were
1
found to produce H-bond signals with H chemical shifts
around 16.00 ppm,88 we suspected that a proton being H-
bonded inside a phosphate dimer might cause this signal.
Indeed, by adding a high excess of H2O to the pure base in a
1
control experiment, H and 31P signals at approximately equal
727
https://dx.doi.org/10.1021/jacs.0c08673
J. Am. Chem. Soc. 2021, 143, 724−735

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Therefore, the protonation of the phosphate by PhSH creates a
classical network of phosphates H-bonded to both amide and
phosphoric acid (N−H···A−···H−A; with A being the
phosphate). These structures could also be verified by MD
simulations (vide infra). The higher electron density on the
phosphates detected in these aggregates (see Figure 3a)
indicates that a significant amount of phosphates is
incorporated in extended H-bond networks with at least two
H-bonds. The actual strengths of the N−H ···O−P H-bond is
difficult to assess, since the DOSY experiments and the MD
simulations discussed below show a release of the ammonium
cations from the aggregate as well as a reduction of the overall
amount of N−H moieties incorporated in H-bonds in the
presence of phosphate dimers. In addition, an averaging of the
chemical shifts with weaker bifurcated H-bond aggregates and/
or a dynamic hopping of the amide from one phosphate to the
other within dimer 1 may occur. Such proton exchange on the
NMR time scale is corroborated by the line broadening of the
1
N−H proton in the H NMR spectrum (Figure 2c, top). The
bifurcation motif could directly be found in our MD derived
snapshots (vide infra). Nevertheless, the interaction of the
phosphate dimers with the amides and the resulting change of
the overall aggregation is clearly visible from the 31P spectra.
In contrast, for the amide/base/phenol mixture only a very
broad 31P signal over a ppm region of −7.77 to −11.16 ppm
was detected but no 1H signal at around 16.50 ppm (see Figure
3c). This indicates no protonation of phosphate and in
consequence no phosphate dimer formation with phenol in
accordance with the relative pKa values of thiophenol and
phenol (pKa (PhSH): 10; pKa (PhOH): 18; in DMSO).92 The
high field position of the base signal indicates high electron
density also for this complex potentially induced by extended
networks with phenol.
The presented data show that thiophenol is able to partially
protonate the phosphate in an acid−base reaction and to form
H-bond mediated P−O···H···O−P dimers, while with phenol
as the additive no phosphate protonation occurs. However, the
1H and 31P chemical shifts as average parameters of the H-
bond strengths are similar in both cases. This is an indication
that besides H-bond strengths also the overall molecular
arrangement is important for reactivity.
Aggregation Analysis by NMR. Therefore, the nature of
the different aggregates was analyzed in detail via diffusion
ordered spectroscopy (DOSY) at 300 and 180 K. From the
obtained self-diffusion coefficients, which refer to averaged
values for the species in the entire solution, the viscosity
corrected volumes V (DOSY) were calculated (for details see
SI chapter 4.6.2). In Figure 4, the volumes derived from DOSY
for the amide/base, amide/base/PhSH, and the amide/base/
PhOH mixtures as well as the literature derived monomer
volumes according to Bondi93 are summarized.
At both temperatures large aggregates were detected as
expected for ion pairs and molecules (amides), which can form
H-bonds in apolar organic solvents. The modulation of
aggregation between ambient and low temperature follows
the expected trends. The amide tends toward higher aggregates
at 180 K, while the ion pair shows higher complexation at
room temperature due to the reduced dielectric constant of the
solvent (see Figure 4 and Table S1, SI).
Figure 3. Phosphate dimer 1 possessing high electron density was
identified as the crucial complex enabling strong H-bonds to amide:
(a) The H-bond acceptor side was studied by 31P NMR chemical shift
analysis in CD2Cl2 at 180 K, and 1:1/1:1:1 mixtures were used for the
multicomponent samples. For complete spectra, see SI. The high field
shift of the phosphate after adding amide confirms the existence of an
amide-base complex. Against the 1H/15N chemical shift trend (Figure
2c,d), the phosphate signal is further high field shifted in the presence
of PhSH (red box). We assume dimer 1 is created by partial
protonation of the base by PhSH (see SI chapter 4.1), which was also
formed in a base/H2O control experiment (top). The main signal
(1:3 ratio) of the amide/base/PhSH mixture appeared at maximum
high field position indicating high amounts of amide H-bonded in
1
aggregate 2. (b) The low field H signals at ∼16.5 ppm for PhSH
containing as well as for the base/H2O samples represent the
hydrogen-bridged proton inside 1. (c) The broad 31P signal of the
amide/base/PhOH sample corroborates a phenol containing large
1
aggregate and the absence of dimer 1 is verified by H NMR.
positions appeared (16.54/-8.59 ppm; Figure 3a, top), albeit in
small concentrations. In addition, in mixtures of PhSH and
base the occurrence of thiophenolate species was identified by
NMR spectroscopy (a detailed characterization is given in
chapter 4.1, SI). Thus, the addition of PhSH creates a new
species, the phosphate dimer 1, with a strong intermolecular
P−O···H···O−P H-bond and high electron density being
located at the 31P nuclei identified by the high field shift. In the
mixture of amide, base, and PhSH, the phosphate dimer 1 is
present as well (see Figure 3a).
In addition, the main and sharper 31P signal of the amide
included complex is even further high field shifted (−9.05
ppm) indicating the highest electron density is located on the
phosphate. This chemical shift reduction excludes a simple
change in the concentration of the original amide−base
complex and confirms a modulation of the overall aggregation.
However, at both temperatures the high amount of
aggregation exemplifies that it is of utmost importance for
the interpretation of H-bond networks in PCET reactions to
check the aggregation situation at the concentrations used in
728
https://dx.doi.org/10.1021/jacs.0c08673
J. Am. Chem. Soc. 2021, 143, 724−735

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
DOSY derived value of NBu4 (V = 986 Å3) indicates that
upon addition of PhSH even a higher percentage of the cation
is pushed away from the phosphate complex, which may
support higher electron densities on phosphate as well. Of
+
+
course, also the formation of thiolate−NBu4 complexes may
explain the small volumes to some extent. For the pure PhSH,
the volume is within the range of the monomer, which is in full
accordance with the poor H-bond donor ability.
Amide/Phosphate/Phenol. In contrast, the measurement
of the amide/base/phenol sample yielded immense values for
both the amide and phenol species (V(DOSY) = 2096 and
1966 Å3). Thus, both compounds are incorporated in large
aggregates (see aggregate 3 in Figure 4). As a result, the
amide−phosphate and the phenol−phosphate H-bonds coexist
competitively in large ensembles. The reduced activity using
PhOH for the hydroamidation indicates that both H-bond
competition and aggregation are potentially detrimental to
reactivity as was shown for aggregation in flavin photo-
catalysis.95
These data show that PhSH generates a phosphate dimer
with increased electron density and a H-bond network to
amides within smaller aggregates, which partially release
ammonium cations while phenol produces larger aggregates
with H-bonds, which are competitive to those of the amide.
Molecular-Dynamics (MD) Simulations. In order to
obtain more details about the essential substrate activating
aggregates, MD simulations were performed. While the
underlying classical force fields often require care for
quantitative predictions of solution structures, they can give
a qualitative account of the manifold of expected and
unexpected aggregates that are present in complex mixed
solutions. In Figure 5, selected PCET relevant snapshots of a
Figure 4. Reduced aggregation in the presence of PhSH and release of
ammonium: An overview of the main aggregates identified by H,
1
15N, 31P NMR, DOSY, and MD simulations is given. The tables
summarize the DOSY derived volumes and the estimated van der
Waals monomer values according to Bondi.93 Large amide-base
aggregation (top) shown as complex 4 with partial release of the
ammonium counterion was identified. In addition to dimer 1 the main
amide containing aggregate 2 with mostly released counterion and an
immense disaggregation was identified for the PhSH containing
mixture. In contrast, for the amide/base/PhOH sample, the volumes
verify the formation of a large PhOH containing aggregate 3. [a]The
value is assumed to be higher, as the phosphate signal is not
completely baseline separated from the adjacent ammonium CH2
groups. [b]Complete overlap of signals made the analysis of the
phosphate diffusion impossible.
catalysis and not in highly diluted samples usually used for UV
luminescence quenching.
Amide/Base and Amide/Base/Thiophenol. In the
following only the aggregation at low temperature is discussed,
as this can be analyzed by both H-bond study and MD
simulations. For the amide/base sample, large volumes were
obtained for both the amide (V(DOSY) = 1900 Å3) and
phosphate (V(DOSY) = 1681 Å3) confirming the existence of
an amide−base complex, proposed as aggregate 4 (Figure 4).94
Figure 5. The PCET relevant amide containing aggregate 2 was
verified by MD simulations: MD snapshots of a mixture containing N-
phenylpent-4-enamide, di-tert-butyl phosphate anion, tetrabutyl-
ammonium cation, tert-butyl phosphoric acid, thiophenolate, H2O,
and H3O+ show the H-bond mediated amide-dimer complex 2 (top)
and an additional extended H-bond motif with two amides (bottom).
The low volume of NBu4 indicates (V(DOSY) = 1389 Å3) a
+
partial dissociation from complex 4. In the presence of PhSH,
surprisingly, an immense reduction of the amide complexation
was obtained (ΔV(DOSY) = 591 Å3). This suggests less self-
aggregation and partial release of the substrate from the
phosphate complex. This unexpected result also explains the
simulation containing N-phenylpent-4-enamide, a di-tert-butyl
phosphate anion, a tetrabutylammonium cation, thiophenolate,
tert-butyl phosphoric acid, H2O, and H3O+ are depicted. The
acid was used in order to be able to mimic the phosphate−H+
interaction. For detailed information about the simulations, see
chapter 5 in the SI. In general, the existence of phosphate
dimer 1 was verified. The crucial complex 2, which is formed
via attachment of the amidyl N−H group to the phosphate
dimer, could also be directly found (Figure 5; top). In
1
reduction of the H and 15N chemical shifts after addition of
PhSH. Due to the low time resolution of NMR spectroscopy,
the signals of the highly activated N−H moieties in H-bonds
with the P−O···H···O−P dimers are averaged with non-
activated N−H moieties of pure amides and the real activation
is masked by the crowd of aggregates. Only in the nonaveraged
signal of the phosphate−amide interaction the extended H-
bond network is reflected. In addition, the comparably small
729
https://dx.doi.org/10.1021/jacs.0c08673
J. Am. Chem. Soc. 2021, 143, 724−735

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
addition, hydrogen bridged to this complex by N−H···OC, a
second amide was found to contribute to the extended H-bond
network in a cooperative way. Also sterically demanding
complexes with one amide attached to the phosphate and one
on the phosphoric acid of dimer 1 could be detected, which
showed geometrically distorted, i.e. weaker, H-bonds (see
Scheme S59). Furthermore, a significant reduction of the
absolute number of H-bonds between amide and phosphate
was calculated upon addition of PhSH.
Considering the averaging function of the ensemble, these
findings explain the overall reduction of the amide−phosphate
H-bond strength in the presence of PhSH indicated by the
1H/15N high field shift of the 15N−H doublet representing all
amide-containing species inside the solution. Thus, also the
MD simulations corroborate significant changes of the overall
aggregates in the presence of protons originating from the
acidic channel of thiophenol.96
Multisite Effect of PhSH within the Mechanism. The
mechanism of the photocatalyzed hydroamidation is extremely
complex, and the PhSH additive can act both as an acid and as
a radical. Therefore, the presence of PhSH has an impact on
multiple rates at different stages of the reaction (see violet
boxes for proton channels and gray boxes for radical channels
of PhSH in Figure 6). The protonated phosphate dimers
Acceleration Effects Substituting PhSH by Diphenyl-
disulfide and Phosphoric Acid. In case the radical and the
ionic pathway influence the hydroamidation reaction in-
dependently from each other, it should be possible to replace
the radical part of PhSH and the acidic part by two additives.
Given the complex mechanism shown in Figure 6, it is key for
further understanding to avoid additional players in the cycle.
Therefore, Ph2S2 was selected to represent the radical part of
PhSH and phosphoric acid was chosen to replace the acidic
part of PhSH. Both Ph2S2 and phosphoric acid were detected
by NMR in the reaction mixtures with PhSH as the exclusive
additive. This separation of the properties of PhSH into two
additives opens the unique opportunity to modulate the
relative amount of acid and radicals by using different ratios.97
Indeed, the initial rate kinetics of amide, Ph2S2, and different
mixtures of phosphate and phosphoric acid show that the
addition of acid accelerates the photoredox catalytic reaction
(Figure 7) up to three times, which is even higher than the
Figure 7. Acceleration effects substituting PhSH by phosphoric acid
and Ph2S2 presented by initial rate kinetics of amide, Ph2S2, and
different amounts of base and acid; (a) in 1:1:1 mixtures of amide/
Ph2S2/acid+base acceleration effects up to a factor of 3 validate the
effect of the proton network on the photoredox catalytic reaction; (b)
under synthetic conditions (1:0.1:0.2 of amide/Ph2S2/acid+base) the
selective optimization of ionic and radical channels allows even an
acceleration by 1 order of magnitude.
acceleration effect of PhSH with its partially acidic function
(see Figure 2a, top; di-n-butyl phosphoric acid was used
because of instability of the tert-butyl acid).
Figure 6. Multisite effects of PhSH on the mechanism in
hydroamidation reactions. Mechanistic steps affected by the acidic
properties of PhSH are highlighted with violet boxes, those affected by
the radical properties with gray boxes. Ionic and radical channels are
intertwined via the partial deprotonation of PhSH. The rates
presented originate from previous studies by Nocera and co-
workers.81
In addition, again the proton signature of the phosphate
dimer was detected, corroborating the assignment (see Figure
S26). Under synthetic reaction conditions the acceleration
effect is even more pronounced (see Figure 7b). Compared to
the reaction with base and disulfide only the real reaction is
accelerated up to a factor of ∼4 (the values are derived from
the comparison of the slopes; see chapter 3.6 and 3.7 in the
SI). Compared to the initially published reaction conditions in
the presence of PhSH (see Figure 2a, bottom, and brown curve
in Figure 7b) even a factor of ∼10 is achieved.32 As a result,
these mechanistic investigations show that the radical and the
acidic properties of PhSH can be successfully replaced by a
combination of disulfide and acid. In addition, by applying
these two additives, the radical and the acidic properties can be
selectively modulated and optimized leading to a reactivity
improvement of 1 order of magnitude. We expect that this will
have a huge impact on the developments in synthetic
chemistry.
themselves and the resulting aggregates discussed above can
influence the initial PCET step as well as the BPT via potential
proton trapping and/or differences in their aggregate size.
PhSH as the radical may impact the radical off-cycle
equilibrium as well as the two HAT steps. In addition, the
amount of thiolate anions reduces the availability of PhSH for
the radical channels; thus, the radical and ionic channel
intertwine. Therefore, PhSH and its derivatives were regarded
as unfavorable additives to separate these effects and additives
were selected which allow for separating the radical and the
ionic pathways at least in the ground state.
730
https://dx.doi.org/10.1021/jacs.0c08673
J. Am. Chem. Soc. 2021, 143, 724−735

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Effect of Phosphoric Acid on the Mechanism. Next, we
investigated which steps in the mechanism contribute to the
observed acceleration upon addition of phosphoric acid. First,
the effect of the proton network on the photoreaction was
tested via luminescence quenching experiments at relevant
synthetic concentrations (for details, see chapter 6, SI).
Previous studies determined very fast and highly effective
PCET rates close to the diffusion limit.80,81 Indeed, the
emission decays of the excited photocatalyst were fastest
without acid as the additive (see Figure S62). The
luminescence decays show that addition of phosphoric acid
significantly slows down the quenching of the photocatalyst’s
excited state, as expected for quenching species becoming
bulkier with increasing amounts of acid. This is in agreement
with DOSY measurements showing larger aggregates in the
presence of phosphoric acid (see Table S1) and excludes the
photostep effecting the acceleration of the overall reaction (for
mechanistic scheme, see Figure 8a). Since acid slows down the
PCET step, the observed acceleration can be caused by either
an acceleration of the productive radical cyclization or a
reduction of the unproductive BET/BPT back reaction.
Therefore, next the availability of the photogenerated amidyl
radical was tested, which is the key intermediate for the
productive reaction. We achieved this by measuring NMR
reaction profiles, applying different light intensities during the
reaction (see Figure 8b). The exclusive change of light
intensity allows the different steps in the mechanism regarding
the amount of amidyl radical to be dissected. The PCET
generation of the radical should be proportional to the light
intensity. The rate constant of the cyclization step should be
independent of the light intensity and constant at least within
one reaction profile. As a result, the rate of the cyclization,
which can be read out in the slope of the overall reaction
profile is expected to be proportional to the amount of
photogenerated amidyl radical available for the productive
reaction.
The blue curve in Figure 8b shows the response of the
overall reaction with disulfide and base. After reducing the light
intensity to 50% or to 25%, a small persistence of the previous
reaction rate is observed for about half a minute. An
acceleration of the cyclization should reduce the amount of
the photogenerated amidyl radical. In contrast the reduction of
the BET/BPT should increase the amount of the amidyl
radical intermediate.
This is in contrast to our previous studies, in which direct
responses to light reduction were detected (upon light
reduction the systems relaxed below the NMR time resolution
to the new photochemical equilibrium situation even in the
case of reaction intermediates).98 This indicates the formation
of an amidyl reservoir feeding the cyclization over a longer
time span. Such an accumulation of the amidyl radical
photogenerated at higher light intensity in the presence of
Ph2S2 is in agreement with the mechanistic proposal and the
relative reaction rates previously proposed by Nocera and co-
workers (see Figure 6), in which the amidyl radical is stored as
the S−N intermediate shown in Figure 8.81 Thus, in the
following the term amidyl radical reservoir describes both the
amidyl radical itself and the amidyl radical stored in this S−N
intermediate. The PCET step is very fast, while the amidyl
cyclization is the rate-determining step of the reaction. In case
the BET/BPT back reaction can be slowed down, an excess
formation of amidyl radical is expected. This was achieved by
Ph2S2 as the additive providing an off-cycle equilibrium to the
S−N intermediate. This amidyl reservoir seems to be now
visible in the persisting slope corresponding to the preceding
light intensity (and in the slower initial buildup in the first
minutes, see also Figure 7). In principle, different times for this
persistence of the slope depending on the preceding light
intensity are expected; however, the slow time resolution of the
NMR spectra (0.5 min per spectrum) is not sufficient in the
case of Ph2S2 as the exclusive additive.
Upon applying a 30:70 mixture of phosphoric acid and base,
again a significant acceleration at 100% light was observed.
After reducing the light to 50% intensity, a persistence of the
initial reaction rate over 3 min can be observed (green curve in
Figure 8b). This directly shows that, with acid as the additive, a
significantly larger amidyl reservoir is formed. In addition, this
indirectly proves the presence of the S−N intermediate, since
the pure amidyl radical would be consumed by far faster by the
cyclization reaction (2.4 × 104 s−1).81 By reducing the light
intensity from 50% to 25%, this persistence of the previous rate
is repeated but the length is shorter (2.5 min) as anticipated
for lower light intensities. This pattern is corroborated by the
third switching of the light intensity: Again, the previous slope
Figure 8. (a) Mechanistic proposal using Ph2S2 and phosphoric acid
instead of PhSH; (b) light intensity dependent reaction profiles
reflecting conditions of a mixture of amide/base/Ph2S2 (blue curve)
and a mixture of amide, base+acid (70:30), and Ph2S2 (green curve).
The observed persistence times after reducing the light intensity
indicate a large amidyl radical reservoir in the presence of acid. This as
well as the light intensity dependent persistence times reveal a
productive pathway of regenerating Ir(III). For details, see text.
731
https://dx.doi.org/10.1021/jacs.0c08673
J. Am. Chem. Soc. 2021, 143, 724−735

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
continues, but for a shorter time. These data show that with
acid such a huge excess of amidyl radical is formed (and stored
in the S−N intermediate) and that even the slow time
resolution of NMR spectroscopy is sufficient to resolve the
additive-dependent offset between the photogenerated for-
mation of an amidyl radical reservoir and its reaction in the
downstream dark reaction.
After reduction of the light from 100% to 50%, around 10%
conversion is observed at the initial rate, which is significantly
more than the 2% of Ir(III) catalyst available in this reaction
(even in case the ongoing conversion at 50% is subtracted).
This substantial size of the amidyl reservoir indicates that a
productive regeneration pathway of Ir(II) to Ir(III) has to be
present prior to the cyclization reaction. Initially this
regeneration was proposed by Knowles and co-workers via a
sequential ET/PT involving PhS· and phosphoric acid after the
cyclization step (see Figure 1b). Considering the off-cycle
equilibrium to the S−N intermediate creating PhS· and the
presence of additional protons in close proximity within the
phosphate dimers, we propose this additional Ir(III)
regeneration pathway directly after the PCET step as shown
in Figure 8a.
replaceable by disulfide and acid, allowing for the first time an
individual tuning of the radical and ionic channel and yielding
accelerations up to a factor of ∼10 under synthetic conditions.
(3) Advanced NMR studies in combination with MD-
simulations revealed activation modes of photoredox catalysis
within a complex ensemble of H-bond assisted ion pairs. (4)
Light intensity dependent reaction profiles allowed the
accumulation of a photogenerated radical species to be traced
indirectly and the correlation of the overall reaction rate to
individual mechanistic steps.
We expect that all aspects of this study, ensembles and
extended H-bond networks, proton transfer pathways, and
tuning of ionic and radical channels, will play crucial roles in
future photoredox catalysis.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/jacs.0c08673.
■
sı
Detailed experimental procedures, sample preparation,
NMR spectroscopic data, MD simulation and emission
experiments (PDF)
In principle, this shortcut should be also possible in the
presence of disulfide and without additional acid since
phosphoric acid is generated in the PCET step and PhS· is
available from the off-cycle equilibrium to the S−N
intermediate. However, the reaction profiles only with disulfide
in Figure 8b show only a very short persistence of the previous
reaction rate upon light reduction. In contrast, with 30% acid a
persistence up to 3 min is observed. This shows clearly the
importance of additional protons, which are in close proximity
(phosphate dimers) to the reaction center which is key to an
effective and productive regeneration of the reduced Ir(II)
catalyst back to the initial Ir(III) state.
Overall, reaction profiles with variation of the light intensity
are used to dissect the effect of acid on the different
mechanistic steps within the catalytic cycle. In the presence
of acid, persistence times of the reaction rate at higher light
intensities were observed over minutes after attenuating the
light, indicating the formation of a large amidyl radical
reservoir. This shows that additional protons in close proximity
are able to open up a productive shortcut for the photocatalyst
regeneration prior to the slow cyclization step.
AUTHOR INFORMATION
Corresponding Author
■
Ruth M. Gschwind − Institute of Organic Chemistry,
University of Regensburg, D-93053 Regensburg, Germany;
orcid.org/0000-0003-3052-0077;
Email: ruth.gschwind@chemie.uni-regensburg.de
Authors
Nele Berg − Institute of Organic Chemistry, University of
Regensburg, D-93053 Regensburg, Germany
Sebastian Bergwinkl − Institute of Physical and Theoretical
Chemistry, University of Regensburg, D-93053 Regensburg,
Germany
Patrick Nuernberger − Institute of Physical and Theoretical
Chemistry, University of Regensburg, D-93053 Regensburg,
Germany; orcid.org/0000-0002-4690-0229
Dominik Horinek − Institute of Physical and Theoretical
Chemistry, University of Regensburg, D-93053 Regensburg,
Germany; orcid.org/0000-0002-0958-0043
Complete contact information is available at:
https://pubs.acs.org/10.1021/jacs.0c08673
CONCLUSION
■
In conclusion, this study shows that complex aggregation and
H-bond networks are potentially present under synthetic
conditions in photoredox catalysis, which affect key properties
in terms of reactivity. Due to highly diluted samples in many
UV studies, these effects have been often overlooked so far.
Regarding the puzzling activation effect of thiophenol in PCET
hydroamidation reactions, it was shown that the interplay of
sophisticated NMR studies, MD simulations, initial rate
kinetics, and light-dependent reaction profiles allow the
disclosure of activation patterns which are hidden by the
average properties of the ensembles.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the German Science Foundation (DFG; GRK 1626
Chemical Photocatalysis; GRK 2620 Ion Pair Effects) for
financial and intellectual support.
REFERENCES
■
(1) Huynh, M. H. V.; Meyer, T. J. Proton-Coupled Electron
Transfer. Chem. Rev. 2007, 107 (11), 5004−5064.
To our knowledge, the combination of techniques within
this study reveals quite a number of aspects in photocatalysis
for the first time: (1) An extended H-bond network was
directly detected as the key element in photocatalysis, here
allowing a productive shortcut for the regeneration of the
reduced photocatalyst to become effective. (2) The broadly
used PhSH additive in photocatalysis was shown to be
(2) Weinberg, D. R.; Gagliardi, C. J.; Hull, J. F.; Murphy, C. F.;
Kent, C. A.; Westlake, B. C.; Paul, A.; Ess, D. H.; McCafferty, D. G.;
Meyer, T. J. Proton-Coupled Electron Transfer. Chem. Rev. 2012, 112
(7), 4016−4093.
̈
(3) Kaila, V. R. I.; Verkhovsky, M. I.; Wikstrom, M. Proton-Coupled
Electron Transfer in Cytochrome Oxidase. Chem. Rev. 2010, 110
(12), 7062−7081.
732
https://dx.doi.org/10.1021/jacs.0c08673
J. Am. Chem. Soc. 2021, 143, 724−735

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
(4) Reece, S. Y.; Nocera, D. G. Proton-Coupled Electron Transfer in
Biology: Results from Synergistic Studies in Natural and Model
Systems. Annu. Rev. Biochem. 2009, 78 (1), 673−699.
(5) Migliore, A.; Polizzi, N. F.; Therien, M. J.; Beratan, D. N.
Biochemistry and Theory of Proton-Coupled Electron Transfer.
Chem. Rev. 2014, 114 (7), 3381−3465.
(6) Weinberg, D. R.; Gagliardi, C. J.; Hull, J. F.; Murphy, C. F.;
Kent, C. A.; Westlake, B. C.; Paul, A.; Ess, D. H.; McCafferty, D. G.;
Meyer, T. J. Proton-Coupled Electron Transfer. Chem. Rev. 2012, 112,
4016.
(7) Meyer, T. J.; Huynh, M. H. V.; Thorp, H. H. The Possible Role
of Proton-Coupled Electron Transfer (PCET) in Water Oxidation by
Photosystem II. Angew. Chem., Int. Ed. 2007, 46 (28), 5284−5304.
(8) Stubbe, J.; Nocera, D. G.; Yee, C. S.; Chang, M. C. Y. Radical
Initiation in the Class I Ribonucleotide Reductase: Long-Range
Proton-Coupled Electron Transfer? Chem. Rev. 2003, 103 (6), 2167−
2202.
(24) Li, B.; Zhao, J.; Onda, K.; Jordan, K. D.; Yang, J.; Petek, H.
Ultrafast Interfacial Proton-Coupled. Science 2006, 311 (March),
1436−1440.
(25) Symes, M. D.; Surendranath, Y.; Lutterman, D. A.; Nocera, D.
G. Bidirectional and Unidirectional PCET in a Molecular Model of a
Cobalt-Based Oxygen-Evolving Catalyst. J. Am. Chem. Soc. 2011, 133
(14), 5174−5177.
(26) Costentin, C.; Drouet, S.; Robert, M.; Saveant, J.-M. A Local
Proton Source Enhances CO2 Electroreduction to CO by a Molecular
Fe Catalyst. Science 2012, 338 (6103), 90−94.
(27) Markle, T. F.; Darcy, J. W.; Mayer, J. M. A New Strategy to
Efficiently Cleave and Form C−H Bonds Using Proton-Coupled
Electron Transfer. Sci. Adv. 2018, 4 (7), eaat5776.
(28) Liu, T.; Guo, M.; Orthaber, A.; Lomoth, R.; Lundberg, M.; Ott,
̈
S.; Hammarstrom, L. Accelerating Proton-Coupled Electron Transfer
of Metal Hydrides in Catalyst Model Reactions. Nat. Chem. 2018, 10
(8), 881−887.
(29) Fava, E.; Nakajima, M.; Nguyen, A. L. P.; Rueping, M.
Photoredox-Catalyzed Ketyl-Olefin Coupling for the Synthesis of
Substituted Chromanols. J. Org. Chem. 2016, 81 (16), 6959−6964.
(30) Lefebvre, Q.; Hoffmann, N.; Rueping, M. Photoorganocata-
lysed and Visible Light Photoredox Catalysed Trifluoromethylation of
Olefins and (Hetero)Aromatics in Batch and Continuous Flow. Chem.
Commun. 2016, 52 (12), 2493−2496.
(9) Hatcher, E.; Soudackov, A. V.; Hammes-Schiffer, S. Proton-
Coupled Electron Transfer in Soybean Lipoxygenase. J. Am. Chem.
Soc. 2004, 126 (18), 5763−5775.
(10) Wang, Y.; Chen, H.; Makino, M.; Shiro, Y.; Nagano, S.;
Asamizu, S.; Onaka, H.; Shaik, S. Theoretical and Experimental
Studies of the Conversion of Chromopyrrolic Acid to an Antitumor
Derivative by Cytochrome P450 StaP: The Catalytic Role of Water
Molecules. J. Am. Chem. Soc. 2009, 131 (19), 6748−6762.
(11) Bonin, J.; Robert, M. Photoinduced Proton-Coupled Electron
Transfers in Biorelevant Phenolic Systems. Photochem. Photobiol.
2011, 87 (6), 1190−1203.
(31) Choi, G. J.; Knowles, R. R. Catalytic Alkene Carboaminations
Enabled by Oxidative Proton-Coupled Electron Transfer. J. Am.
Chem. Soc. 2015, 137 (29), 9226−9229.
(32) Miller, D. C.; Choi, G. J.; Orbe, H. S.; Knowles, R. R. Catalytic
Olefin Hydroamidation Enabled by Proton-Coupled Electron Trans-
fer. J. Am. Chem. Soc. 2015, 137 (42), 13492−13495.
(12) Kaila, V. R. I. Long-Range Proton-Coupled Electron Transfer in
Biological Energy Conversion: Towards Mechanistic Understanding
of Respiratory Complex I. J. R. Soc., Interface 2018, 15 (141),
20170916.
(33) Choi, G. J.; Zhu, Q.; Miller, D. C.; Gu, C. J.; Knowles, R. R.
Catalytic Alkylation of Remote C−H Bonds Enabled by Proton-
Coupled Electron Transfer. Nature 2016, 539 (7628), 268−271.
(34) Rono, L. J.; Yayla, H. G.; Wang, D. Y.; Armstrong, M. F.;
Knowles, R. R. Enantioselective Photoredox Catalysis Enabled by
Proton-Coupled Electron Transfer: Development of an Asymmetric
Aza-Pinacol Cyclization. J. Am. Chem. Soc. 2013, 135 (47), 17735−
17738.
(13) Mora, S. J.; Odella, E.; Moore, G. F.; Gust, D.; Moore, T. A.;
Moore, A. L. Proton-Coupled Electron Transfer in Artificial
Photosynthetic Systems. Acc. Chem. Res. 2018, 51 (2), 445−453.
(14) Tommos, C.; Babcock, G. T. Oxygen Production in Nature: A
Light-Driven Metalloradical Enzyme Process. Acc. Chem. Res. 1998,
31 (1), 18−25.
(35) Yayla, H. G.; Wang, H.; Tarantino, K. T.; Orbe, H. S.; Knowles,
R. R. Catalytic Ring-Opening of Cyclic Alcohols Enabled by PCET
Activation of Strong O−H Bonds. J. Am. Chem. Soc. 2016, 138 (34),
10794−10797.
(36) Miller, D. C.; Tarantino, K. T.; Knowles, R. R. Proton-Coupled
Electron Transfer in Organic Synthesis: Fundamentals, Applications
and Opportunities. Top. Curr. Chem. 2016, 374 (3), 1−59.
(37) Gentry, E. C.; Knowles, R. R. Synthetic Applications of Proton-
Coupled Electron Transfer. Acc. Chem. Res. 2016, 49 (8), 1546−1556.
(38) Yayla, H. G.; Knowles, R. R. Proton-Coupled Electron Transfer
in Organic Synthesis: Novel Homolytic Bond Activations and
Catalytic Asymmetric Reactions with Free Radicals. Synlett 2014, 25
(21), 2819−2826.
(39) Qiu, G.; Knowles, R. R. Rate−Driving Force Relationships in
the Multisite Proton-Coupled Electron Transfer Activation of
Ketones. J. Am. Chem. Soc. 2019, 141 (6), 2721−2730.
(40) Hoffmann, N. Proton-Coupled Electron Transfer in Photo-
redox Catalytic Reactions. Eur. J. Org. Chem. 2017, 2017 (15), 1982−
1992.
(41) Nguyen, L. Q.; Knowles, R. R. Catalytic C−N Bond-Forming
Reactions Enabled by Proton-Coupled Electron Transfer Activation
of Amide N−H Bonds. ACS Catal. 2016, 6 (5), 2894−2903.
(42) Tarantino, K. T.; Liu, P.; Knowles, R. R. Catalytic Ketyl-Olefin
Cyclizations Enabled by Proton-Coupled Electron Transfer. J. Am.
Chem. Soc. 2013, 135 (27), 10022−10025.
́
(15) Costentin, C.; Robert, M.; Saveant, J.-M. Electrochemical
Approach to the Mechanistic Study of Proton-Coupled Electron
Transfer. Chem. Rev. 2010, 110 (12), 2145−2179.
(16) Madhiri, N.; Finklea, H. O. Potential-, PH-, and Isotope-
Dependence of Proton-Coupled Electron Transfer of an Osmium
Aquo Complex Attached to an Electrode. Langmuir 2006, 22 (25),
10643−10651.
(17) Hammes-Schiffer, S.; Soudackov, A. V. Proton-Coupled
Electron Transfer in Solution, Proteins, and Electrochemistry. J.
Phys. Chem. B 2008, 112 (45), 14108−14123.
(18) Hammes-Schiffer, S. Proton-Coupled Electron Transfer:
Moving Together and Charging Forward. J. Am. Chem. Soc. 2015,
137 (28), 8860−8871.
(19) Ghosh, S.; Soudackov, A. V.; Hammes-Schiffer, S. Electro-
chemical Electron Transfer and Proton-Coupled Electron Transfer:
Effects of Double Layer and Ionic Environment on Solvent
Reorganization Energies. J. Chem. Theory Comput. 2016, 12 (6),
2917−2925.
́
(20) Saveant, J. M. Electrochemical Approach to Proton-Coupled
Electron Transfers: Recent Advances. Energy Environ. Sci. 2012, 5 (7),
7718−7731.
́
(21) Costentin, C.; Saveant, J. M. Theoretical and Mechanistic
Aspects of Proton-Coupled Electron Transfer in Electrochemistry.
Curr. Opin. Electrochem. 2017, 1 (1), 104−109.
́
(22) Saveant, J.-M. Concerted Proton-Electron Transfers: Funda-
(43) Zhu, Q.; Graff, D. E.; Knowles, R. R. Intermolecular Anti-
Markovnikov Hydroamination of Unactivated Alkenes with Sulfona-
mides Enabled by Proton-Coupled Electron Transfer. J. Am. Chem.
Soc. 2018, 140, 741−747.
(44) Ota, E.; Wang, H.; Frye, N. L.; Knowles, R. R. A Redox Strategy
for Light-Driven, Out-of-Equilibrium Isomerizations and Application
mentals and Recent Developments. Annu. Rev. Anal. Chem. 2014, 7
(1), 537−560.
́
(23) Costentin, C.; Robert, M.; Saveant, J. M.; Tard, C. Breaking
Bonds with Electrons and Protons. Models and Examples. Acc. Chem.
Res. 2014, 47 (1), 271−280.
733
https://dx.doi.org/10.1021/jacs.0c08673
J. Am. Chem. Soc. 2021, 143, 724−735

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
to Catalytic C-C Bond Cleavage Reactions. J. Am. Chem. Soc. 2019,
141, 1457.
(64) Damrauer, N. H.; Hodgkiss, J. M.; Rosenthal, J.; Nocera, D. G.
Observation of Proton-Coupled Electron Transfer by Transient
Absorption Spectroscopy in a Hydrogen-Bonded, Porphyrin Donor-
Acceptor Assembly. J. Phys. Chem. B 2004, 108, 6315−6321.
(65) Morris, W. D.; Mayer, J. M. Separating Proton and Electron
Transfer Effects in Three-Component Concerted Proton-Coupled
Electron Transfer Reactions. J. Am. Chem. Soc. 2017, 139 (30),
10312−10319.
(45) Wenger, O. S. Proton-Coupled Electron Transfer with
Photoexcited Metal Complexes. Acc. Chem. Res. 2013, 46 (7),
1517−1526.
(46) Hammes-Schiffer, S.; Stuchebrukhov, A. A. Theory of Coupled
Electron and Proton Transfer Reactions. Chem. Rev. 2010, 110 (12),
6939−6960.
(47) Mayer, J. M.; Rhile, I. J. Thermodynamics and Kinetics of
Proton-Coupled Electron Transfer: Stepwise vs. Concerted Pathways.
Biochim. Biophys. Acta, Bioenerg. 2004, 1655 (1−3), 51−58.
(48) Zhao, K.; Yamashita, K.; Carpenter, J. E.; Sherwood, T. C.;
Ewing, W. R.; Cheng, P. T. W.; Knowles, R. R. Catalytic Ring
Expansions of Cyclic Alcohols Enabled by Proton-Coupled Electron
Transfer. J. Am. Chem. Soc. 2019, 141 (22), 8752−8757.
(49) Nguyen, S. T.; Zhu, Q.; Knowles, R. R. PCET-Enabled Olefin
Hydroamidation Reactions with N-Alkyl Amides. ACS Catal. 2019, 9
(5), 4502−4507.
(66) Hammes-Schiffer, S.; Iordanova, N. Theoretical Studies of
Proton-Coupled Electron Transfer Reactions. Biochim. Biophys. Acta,
Bioenerg. 2004, 1655, 29−36.
(67) Hammes-Schiffer, S.; Stuchebrukhov, A. A. Theory of Coupled
Electron and Proton Transfer Reactions. Chem. Rev. 2010, 110 (12),
6939−6960.
(68) Di Meo, F. Di; Lemaur, V.; Lazzaroni, R.; Duroux, J.; Olivier,
Y.; Trouillas, P. Free Radical Scavenging by Natural Polyphenols:
Atom versus Electron Transfer. J. Phys. Chem. A 2013, 117, 2082−
2092.
(50) Morton, C. M.; Zhu, Q.; Ripberger, H.; Troian-Gautier, L.;
Toa, Z. S. D.; Knowles, R. R.; Alexanian, E. J. C−H Alkylation via
Multisite-Proton-Coupled Electron Transfer of an Aliphatic C−H
Bond. J. Am. Chem. Soc. 2019, 141 (33), 13253−13260.
(51) Wang, D.; Loose, F.; Chirik, P. J.; Knowles, R. R. N−H Bond
Formation in a Manganese(V) Nitride Yields Ammonia by Light-
Driven Proton-Coupled Electron Transfer. J. Am. Chem. Soc. 2019,
141 (12), 4795−4799.
(69) Hanthorn, J. J.; Valgimigli, L.; Pratt, D. A. Incorporation of
Ring Nitrogens into Diphenylamine Antioxidants: Striking a Balance
between Reactivity and Stability. J. Am. Chem. Soc. 2012, 134 (20),
8306−8309.
(70) DiLabio, G. A.; Johnson, E. R. Lone Pair-π and π-π Interactions
Play an Important Role in Proton-Coupled Electron Transfer
Reactions. J. Am. Chem. Soc. 2007, 129 (19), 6199−6203.
(71) Yuasa, J.; Yamada, S.; Fukuzumi, S. One-Step versus Stepwise
Mechanism in Protonated Amino Acid-Promoted Electron-Transfer
Reduction of a Quinone by Electron Donors and Two-Electron
Reduction by a Dihydronicotinamide Adenine Dinucleotide Ana-
logue. Interplay between Electron Transfer and Hydrogen Bonding. J.
Am. Chem. Soc. 2008, 130 (17), 5808−5820.
(72) Alligrant, T. M.; Alvarez, J. C. The Role of Intermolecular
Hydrogen Bonding and Proton Transfer in Proton-Coupled Electron
Transfer. J. Phys. Chem. C 2011, 115 (21), 10797−10805.
(73) Nomrowski, J.; Wenger, O. S. Photoinduced PCET in
Ruthenium-Phenol Systems: Thermodynamic Equivalence of Uni-
and Bidirectional Reactions. Inorg. Chem. 2015, 54 (7), 3680−3687.
(74) Nguyen, T. M.; Nicewicz, D. A. Anti-Markovnikov Hydro-
amination of Alkenes Catalyzed by an Organic Photoredox System. J.
Am. Chem. Soc. 2013, 135 (26), 9588−9591.
(75) Margrey, K. A.; Nicewicz, D. A. A General Approach to
Catalytic Alkene Anti-Markovnikov Hydrofunctionalization Reactions
via Acridinium Photoredox Catalysis. Acc. Chem. Res. 2016, 49 (9),
1997−2006.
(76) Nguyen, T. M.; Manohar, N.; Nicewicz, D. A. Anti-
Markovnikov Hydroamination of Alkenes Catalyzed by a Two-
Component Organic Photoredox System: Direct Access to Phenethyl-
amine Derivatives. Angew. Chem., Int. Ed. 2014, 53 (24), 6198−6201.
(77) Gagliardi, C. J.; Murphy, C. F.; Binstead, R. A.; Thorp, H. H.;
Meyer, T. J. Concerted Electron Proton Transfer (EPT) in the
Oxidation of Cysteine. J. Phys. Chem. C 2015, 119 (13), 7028−7038.
(78) Kuss-Petermann, M.; Wenger, O. S. Mechanistic Diversity in
Proton-Coupled Electron Transfer between Thiophenols and Photo-
excited [Ru(2,2′-Bipyrazine)3]2+. J. Phys. Chem. Lett. 2013, 4 (15),
2535−2539.
(52) Shin, N. Y.; Ryss, J. M.; Zhang, X.; Miller, S. J.; Knowles, R. R.
Light-Driven Deracemization Enabled by Excited-State Electron
Transfer. Science 2019, 366, 364−369.
(53) Bissonnette, N. B.; Ellis, J. M.; Hamann, L. G.; Romanov-
Michailidis, F. Expedient Access to Saturated Nitrogen Heterocycles
by Photoredox Cyclization of Imino-Tethered Dihydropyridines.
Chem. Sci. 2019, 10 (41), 9591−9596.
́
(54) Caron, A.; Morin, E.; Collins, S. K. Bifunctional Copper-Based
Photocatalyst for Reductive Pinacol-Type Couplings. ACS Catal.
2019, 9 (10), 9458−9464.
(55) Wang, Y.; Liu, Y.; He, J.; Zhang, Y. Redox-Neutral
Photocatalytic Strategy for Selective C−C Bond Cleavage of Lignin
and Lignin Models via PCET Process. Sci. Bull. 2019, 64 (22), 1658−
1666.
(56) Zhang, M.; Yang, L.; Yang, H.; An, G.; Li, G. Visible Light
Mediated C(Sp3)-H Alkenylation of Cyclic Ethers Enabled by Aryl
Ketone. ChemCatChem 2019, 11 (6), 1606−1609.
(57) Hammes-Schiffer, S. Proton-Coupled Electron Transfer:
Classification Scheme and Guide to Theoretical Methods. Energy
Environ. Sci. 2012, 5 (7), 7696−7703.
(58) Hammes-Schiffer, S. Theoretical Perspectives on Proton-
Coupled Electron Transfer Reactions. Acc. Chem. Res. 2001, 34 (4),
273−281.
(59) Mayer, J. M.; Hrovat, D. A.; Thomas, J. L.; Borden, W. T.
Proton-Coupled Electron Transfer versus Hydrogen Atom Transfer in
Benzyl/Toluene, Methoxyl/Methanol, and Phenoxyl/ Phenol Self-
Exchange Reactions. J. Am. Chem. Soc. 2002, 124, 11142−11147.
(60) Skone, J. H.; Soudackov, A. V.; Hammes-Schiffer, S.
Calculation of Vibronic Couplings for Phenoxyl/Phenol and
Benzyl/Toluene Self-Exchange Reactions: Implications for Proton-
Coupled Electron Transfer Mechanisms. J. Am. Chem. Soc. 2006, 128
(51), 16655−16663.
(79) Gesmundo, N. J.; Grandjean, J. M. M.; Nicewicz, D. A. Amide
and Amine Nucleophiles in Polar Radical Crossover Cycloadditions:
Synthesis of γ-Lactams and Pyrrolidines. Org. Lett. 2015, 17 (5),
1316−1319.
̈
́
(61) Sjodin, M.; Irebo, T.; Utas, J. E.; Lind, J.; Merenyi, G.;
̈
Åkermark, B.; Hammarstrom, L. Kinetic Effects of Hydrogen Bonds
(80) Qiu, G.; Knowles, R. R. Understanding Chemoselectivity in
Proton-Coupled Electron Transfer: A Kinetic Study of Amide and
Thiol Activation. J. Am. Chem. Soc. 2019, 141, 16574−16578.
(81) Ruccolo, S.; Qin, Y.; Schnedermann, C.; Nocera, D. G. General
Strategy for Improving the Quantum Efficiency of Photoredox
Hydroamidation Catalysis. J. Am. Chem. Soc. 2018, 140 (44),
14926−14937.
on Proton-Coupled Electron Transfer from Phenols. J. Am. Chem. Soc.
2006, 128 (40), 13076−13083.
(62) Litwinienko, G.; Ingold, K. U. Solvent Effects on the Rates and
Mechanisms of Reaction of Phenols with Free Radicals. Acc. Chem.
Res. 2007, 40 (3), 222−230.
(63) Young, E. R.; Rosenthal, J.; Hodgkiss, J. M.; Nocera, D. G.
Comparative PCET Study of a Donor-Acceptor Pair Linked by
Ionized and Nonionized Asymmetric Hydrogen-Bonded Interfaces. J.
Am. Chem. Soc. 2009, 131 (22), 7678−7684.
(82) Gschwind, R. M.; Armbruster, M.; Zubrzycki, I. Z. NMR
̈
Detection of Intermolecular NH···OP Hydrogen Bonds between
734
https://dx.doi.org/10.1021/jacs.0c08673
J. Am. Chem. Soc. 2021, 143, 724−735

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Guanidinium Protons and Bisposphonate Moieties in an Artificial
Arginine Receptor. J. Am. Chem. Soc. 2004, 126 (33), 10228−10229.
(83) Sharif, S.; Denisov, G. S.; Toney, M. D.; Limbach, H.-H. NMR
Studies of Coupled Low-and High-Barrier Hydrogen Bonds in
Pyridoxal-5′-Phosphate-Enzyme Model Systems in Polar Solution. J.
Am. Chem. Soc. 2007, 129 (11), 6313−6327.
(84) Steiner, T. The Hydrogen Bond in the Solid State. Angew.
Chem., Int. Ed. 2002, 41 (1), 48−76.
(85) Prins, L. J.; Reinhoudt, D. N.; Timmerman, P. Noncovalent
synthesis using hydrogen bonding. Angew. Chem. Int. Ed. 2001, 40
(13), 2382−2426.
(86) Harris, T. K.; Zhao, Q.; Mildvan, A. S. NMR Studies of Strong
Hydrogen Bonds in Enzymes and in a Model Compound. J. Mol.
Struct. 2000, 552 (1), 97−109.
(87) Greindl, J.; Hioe, J.; Sorgenfrei, N.; Morana, F.; Gschwind, R.
M. Brønsted Acid Catalysis - Structural Preferences and Mobility in
Imine/Phosphoric Acid Complexes. J. Am. Chem. Soc. 2016, 138,
15965−15971.
(88) Sorgenfrei, N.; Hioe, J.; Greindl, J.; Rothermel, K.; Morana, F.;
Lokesh, N.; Gschwind, R. M. NMR Spectroscopic Characterization of
Charge Assisted Strong Hydrogen Bonds in Brønsted Acid Catalysis.
J. Am. Chem. Soc. 2016, 138 (50), 16345−16354.
̈
(89) Chen, K.; Berg, N.; Gschwind, R.; Konig, B. Selective Single
C(Sp3)-F Bond Cleavage in Trifluoromethylarenes: Merging Visible-
Light Catalysis with Lewis Acid Activation. J. Am. Chem. Soc. 2017,
139, 18444−18447.
(90) Hunter, C. A. Quantifying Intermolecular Interactions:
Guidelines for the Molecular Recognition Toolbox. Angew. Chem.,
Int. Ed. 2004, 43 (40), 5310−5324.
(91) The broad 15N−H signal was processed using a Gaussian
window function for better resolution (Figure 2b).
(92) Bordwell, F. G.; McCallum, R. J.; Olmstead, W. N. Acidities
and Hydrogen Bonding of Phenols in Dimethyl Sulfoxide. J. Org.
Chem. 1984, 49 (49), 1424−1427.
(93) Bondi, A. Van Der Waals Volumes and Radii. J. Phys. Chem.
1964, 68 (3), 441−451.
(94) Notably, the ammonium is partially separated from dimer 1 in
the thiophenol/base mixture (for comparison of the DOSY values, see
SI).
́
́
̌
(95) Dadova, J.; Kummel, S.; Feldmeier, C.; Cibulkova, J.; Pazout,
̈
̈
R.; Maixner, J.; Gschwind, R. M.; Konig, B.; Cibulka, R. Aggregation
Effects in Visible-Light Flavin Photocatalysts: Synthesis, Structure,
and Catalytic Activity of 10-Arylflavins. Chem. - Eur. J. 2013, 19 (3),
1066−1075.
(96) The comparison of MD derived diffusion coefficients of the
different species yielded additional information about the thiopheno-
late. As the values for the thiolate are significantly smaller than those
of PhSH, the complexation of the thiolate was verified.
(97) Please be aware that Ph2S2 and acid can produce PhSH upon
irradiation with blue light. That means that, in photocatalysis acid,
Ph2S2 and PhSH can interconvert into each other. Nevertheless,
separation into acid and Ph2S2 allows for a selective tuning of the
proton density and the H-bond networks in the ground state.
̈
(98) Bartling, H.; Eisenhofer, A.; Konig, B.; Gschwind, R. M. The
Photocatalyzed Aza-Henry Reaction of N-Aryltetrahydroisoquino-
lines: Comprehensive Mechanism, H●- versus H+-Abstraction, and
Background Reactions. J. Am. Chem. Soc. 2016, 138 (36), 11860−
11871.
735
https://dx.doi.org/10.1021/jacs.0c08673
J. Am. Chem. Soc. 2021, 143, 724−735