Sensitization-Initiated Electron Transfer for Photoredox Catalysis

Article abstract of DOI:10.1002/anie.201703004

Photosynthetic organisms exploit antenna chromophores to absorb light and transfer excitation energy to the reaction center where redox reactions occur. In contrast, in visible-light chemical photoredox catalysis, a single species (i.e., the photoredox catalyst) absorbs light and performs the redox chemistry. Mimicking the energy flow of the biological model, we report a two-center photoredox catalytic approach in which the tasks of light energy collection and electron transfer (i.e., redox reactions) are assigned to two different molecules. Ru(bpy)₃Cl₂ absorbs the visible light and transfers the energy to polycyclic aromatic hydrocarbons that enable the redox reactions. This operationally simple sensitization-initiated electron transfer enables the use of arenes that do not absorb visible light, such as anthracene or pyrene, for photoredox applications. We demonstrate the merits of this approach by the reductive activation of chemical bonds with high reduction potentials for carbon–carbon and carbon–heteroatom bond formations.

Full text of DOI:10.1002/anie.201703004

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201703004
German Edition: DOI: 10.1002/ange.201703004
Photocatalysis Very Important Paper
Sensitization-Initiated Electron Transfer for Photoredox Catalysis
Indrajit Ghosh, Rizwan S. Shaikh, and Burkhard Kçnig*
Abstract: Photosynthetic organisms exploit antenna chromo-
phores to absorb light and transfer excitation energy to the
reaction center where redox reactions occur. In contrast, in
visible-light chemical photoredox catalysis, a single species
(i.e., the photoredox catalyst) absorbs light and performs the
redox chemistry. Mimicking the energy flow of the biological
model, we report a two-center photoredox catalytic approach
in which the tasks of light energy collection and electron
transfer (i.e., redox reactions) are assigned to two different
molecules. Ru(bpy)3Cl2 absorbs the visible light and transfers
the energy to polycyclic aromatic hydrocarbons that enable the
redox reactions. This operationally simple sensitization-initi-
ated electron transfer enables the use of arenes that do not
absorb visible light, such as anthracene or pyrene, for photo-
redox applications. We demonstrate the merits of this approach
by the reductive activation of chemical bonds with high
reduction potentials for carbon–carbon and carbon–hetero-
atom bond formations.
times and relatively poor photochemical quantum yields, the
restricted redox energy gain upon visible-light photoexcita-
tion limits the overall performance of a photoredox catalyst
with respect to the substrate scope. Furthermore, in accord-
ance with the common notion that the more reductive
a catalyst, the less oxidative it is, synthetically demanding
chemical modifications enhance the redox equivalence gain in
one direction at the expense of the other (oxidative/reductive
or vice versa).[9]
Inspired by the natural photosynthetic systems, we
envisioned using strongly absorbing molecules for light
energy harvesting and funneling the excitation energy to
a highly reducing molecule that does not absorb visible light.
The redox potential of the latter species may be utilized to
drive challenging redox reactions that are not feasible or less
efficient with a single photocatalyst. The effective conversion
of visible-light energy into redox equivalents in different
photoredox catalytic systems is compared in Scheme 1. The
energies of blue (455 nm) and green photons (530 nm) of
262 kJmolÀ1 or 2.72 eV and 226 KJmolÀ1 or 2.34 eV, respec-
tively, are insufficient for the activation of many chemical
bonds via direct photoexcitation. However, converting the
same energy into redox equivalents enables the activation of
such bonds through single-electron-transfer redox processes.
The many applications of visible-light photoredox catal-
ysis for the activation of chemical bonds that have been
reported over the last decade document the practicality of
such energy transduction processes.[3–7,10] However, as the
electronic rearrangements of the photoredox catalysts upon
light absorption commence the energy transduction process-
es, the intrinsic properties, in particular, the absorption and
ground-state redox potentials of the photoredox catalysts,
define how much light energy can be converted into redox
energy (Rehm–Weller equation).[11] As depicted in Scheme 1,
the maximum reduction potential available in a typical
Ru(bpy)32+-based photoredox catalytic system never exceeds
a value of À1.33 V (the inherent ground-state reduction
P
hotosynthetic organisms transform light energy into chem-
ical free energy through a series of energy-transducing
reactions. Visible light is harvested by antenna pigments,
such as chlorophyll b and b-carotene, and transferred to the
reaction center pigment, chlorophyll a, to drive photosyn-
thetic reactions.[1–2] This strategy of using strongly absorbing
antenna molecules for visible-light collection and weakly
absorbing redox centers to drive chemical reactions enables
the efficient conversion of light energy into redox energy for
the simultaneous oxidation of water to molecular oxygen and
the reduction of NADP+ to NADPH. In contrast, visible-
light-mediated photoredox catalysis,[3–7] an emerging field in
synthetic organic chemistry, uses visible light to drive
chemical reactions, but relies on the use of the same molecule
(i.e., a photoredox catalyst) for both visible-light absorption
and the conversion of the light energy into redox energy to
initiate redox reactions. This excludes the application of many
chromophores that have extremely high redox potentials, but
do not absorb visible light (e.g., polycyclic aromatic hydro-
carbons for reduction reactions; c.f., sodium naphthalenide)
in photoredox catalysis, and leads to a strong dependence on
the inherent redox potentials of typical photoredox catalysts
for the conversion of visible light into the maximum available
redox energy.[8] Aside from the typically very long reaction
+
potential of Ru(bpy)3 ).[3] In contrast, the photoredox cata-
lytic approach based on sensitization-initiated electron trans-
fer (SenI-ET) uses the intrinsic properties (absorption and
redox potentials) of two different molecules for the con-
version of light energy into redox energy through a bimolec-
ular energy-transfer process. Here, although the energy of
visible light is insufficient to access the excited states of the
redox-active species (the DE for photoexcitation), the relative
energy of its excited state with respect to the visible-light-
absorbing sensitizer allows its excitation via a simple photo-
induced energy-transfer process. As depicted in Scheme 1,
such an energy transduction allows accessing the reduction
potentials of polycyclic aromatic hydrocarbons, such as
pyrene or anthracene. Even more interestingly, the reduction
potentials of the radical anions of the polycyclic aromatic
[*] Dr. I. Ghosh, Dr. R. S. Shaikh, Prof. Dr. B. Kçnig
Universitꢀt Regensburg
Fakultꢀt fꢁr Chemie und Pharmazie
93040 Regensburg (Germany)
E-mail: burkhard.koenig@ur.de
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under https://doi.org/10.
1002/anie.201703004.
Angew. Chem. Int. Ed. 2017, 56, 1 – 7
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
Scheme 1. Biological (left), typical photoinduced (PET, middle), and sensitization-initiated electron transfer (SenI-ET, right) photoredox processes.
The relative redox potential gains upon photoexcitation using the same visible-light energy for PET and SenI-ET processes are also depicted. Upon
photoexcitation using the same light energy (i.e., 2.0 eV), the SenI-ET catalytic method accumulates 0.8 V more redox energy than the typical
photoredox PET cycle (see text).
hydrocarbons (PAHs; e.g., pyrene: À2.1 V vs. SCE, which is
The SenI-ET approach to photoredox catalysis is partic-
ularly suitable for the activation of carbon–(pseudo)halogen
bonds in aryl chlorides, bromides,[15–19] or pseudohalides for
the following reasons: 1) Aryl halides and pseudohalides do
not absorb visible light, and therefore cannot be activated by
direct excitation with visible light. 2) Their high reduction
potentials are beyond the reach of many typical photoredox
catalysts including Ru(bpy)32+, but lower than the reduction
potentials of the radical anions of polyaromatic hydrocar-
bons.[10,20] 3) Owing to the strong carbon–halogen bonds and
the two-step bond-dissociation kinetics,[20] very long reaction
times are required even with very good electron donors[21] or
highly reducing conPET photoredox catalytic systems[22] for
0.8 V more negative than the reduction potential of Ru-
+
(bpy)3 ) can be accessed by using 530 nm light in a process
that requires approximately 120 KJmolÀ1 or 1.2 eV less
energy than the energy required for direct excitation. Note
that pyrene absorbs only up to about 350 nm (energy ca.
342 kJmolÀ1 or 3.5 eV).
Diffusion-controlled sequential energy- and electron-
transfer processes, however, impose two challenges for the
development of such two-center photoredox catalytic systems
for synthetic applications (herein, we report photoredox
catalytic reduction reactions): 1) The energy transfer from the
energy-harvesting molecule to the redox-active (and visible-
light-inactive) molecule should be faster, ideally by one order
of magnitude or higher, than single-electron transfer from the
sacrificial electron donor that is present in the system (see
Scheme 1). 2) Upon energy transfer, the triplet energy
acceptor should be quickly transformed into its radical
anion by electron transfer from the sacrificial electron
donor. With these mechanistic challenges in mind, we began
2+
our investigations with Ru(bpy)3 as the light-harvesting
complex (i.e., sensitizer) and various PAHs as triplet energy
acceptors as they have exceptionally high reduction poten-
2+
tials.[12] To our delight, the luminescence of Ru(bpy)3 was
quenched efficiently by a range of PAHs, such as anthracene
(An), pyrene (Py), and 9,10-diphenylanthracene (DPA; see
Figure 1 and the Supporting Information for luminescence
quenching experiments). Furthermore, the luminescence
2+
quenching of Ru(bpy)3 via energy transfer by the inves-
tigated PAHs was faster by at least one order of magnitude
than electron transfer from N,N-diisopropylethylamine
(DIPEA).[13,14] Triphenylene or naphthalene, however, were
ineffective triplet acceptors owing to their high triplet-state
energies (see Figure 1). Among the successful PAH quench-
ers, we selected Py for further investigations owing to its
superior ground-state reduction potential (ca. À2.1 V, see
Figure 1. Top: Chemical structures of Ru(bpy)3Cl2·H2O and the PAHs
investigated herein. Ground-state reduction potentials and triplet
energies are also given.[12] The depicted reduction potentials are
against the saturated calomel electrode (SCE). Bottom: Spectroscopic
investigations. A) Changes in the luminescence spectra (in this case,
intensity; lex =470 nm) of Ru(bpy)32+ upon successive addition of
pyrene in DMSO. In the insets, changes in the absorption spectra of
2+
Ru(bpy)3 upon successive addition of Py (i) and the absorption
À
spectra of Ru(bpy)32+, Py, and An (ii) are shown. B) Stern–Volmer plots
Figure 1), and observed relatively fast C H arylation reaction
2+
kinetics (see below and Table S1 in the Supporting Informa-
tion).
for the quenching of Ru(bpy)3 with different PAHs (An, Py, Naph,
and Tpn), DIPEA, and 2-bromobenzonitrile.
2
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 7
These are not the final page numbers!

Angewandte
Communications
Chemie
carbon–halogen bond activation. 4) The reduction potentials
of both catalytic components are utilized for photoredox
cascade reactions as the catalytic components are connected
by diffusion-controlled energy transfer. Single-electron trans-
fer to the (pseudo)halide substrates followed by (pseudo)ha-
lide anion extrusion generates (hetero)aryl radicals for the
À
À
formation of new C C or C heteroatom bonds in the
presence of suitable radical trapping reagents.
The reaction conditions were optimized by irradiating
a
mixture of 2-bromobenzonitrile (model substrate),
Ru(bpy)32+, Py, N-methylpyrrole (trapping reagent), and
different sacrificial electron donors with blue light (lex =
455 Æ 15 nm; see Table S2). To our delight, when the reaction
was conducted in the presence of Ru(bpy)32+ (1.0 mol%), Py
(5.0 mol%), and a slight excess of DIPEA (1.4 equiv) in
À
DMSO at room temperature (258C), the C H-arylated
product 1a was obtained within about 2.5 h and isolated in
86% yield (GC yield: 92%; see Table S1). When the Ru-
(bpy)32+ and Py loadings were reduced, longer reaction times
were required (see Table S2). With respect to the sacrificial
electron donor, DIPEA was slightly more effective than
triethylamine (Et3N), and more effective than tributylamine
(Bu3N, probably owing to the lower solubility of Bu3N in
DMSO). A series of control experiments (i.e., omitting each
individual component) confirmed that all components, that is,
Ru(bpy)32+, Py, the electron donor, and visible-light irradi-
ation, are essential (Table S1, entries 1–8) for the photoredox
À
Figure 2. The C H arylation products obtained from (hetero)aryl
halides with arenes and biologically important heterocycles. Yields of
the isolated products are given.
À
catalytic C H arylation reactions to take place. Interestingly,
À
the reductive C H arylation reactions were also effective and
gave the products in similar yields when the reaction mixtures
were irradiated with green LEDs (lex = 530 Æ 15 nm) as
potential of 5-bromo-3-methylindole is too high (see the
Supporting Information for the cyclic voltammogram) for the
activation of the carbon–halogen bond under these condi-
tions. However, when it is used as a trapping partner for the
2+
Ru(bpy)3 absorbs across a broad range of the visible
spectrum. However, longer reaction times were necessary
2+
owing to the lower extinction coefficient of Ru(bpy)3 at
À
530 nm. Py could be recovered during the isolation of the
aryl radicals, the C H-arylated product 2 f could be obtained
À
C H-arylated product 1a (see the Supporting Information).
in good yield (see Figure 2). When 2-bromo-1-methyl-1H-
pyrrole was used as the trapping reagent, product 2j was
isolated in 69% yield.
With the optimized reaction conditions in hand, which
require only the mixing of substrates, Ru(bpy)32+, Py, DIPEA,
trapping reagent, and irradiation with visible light, we then
explored the scope of this SenI-ET photoredox catalytic
process with different (hetero)aryl bromides as (hetero)aryl
radical precursors. A range of bromo(hetero)arenes reacted
very efficiently under this reductive SenI-ET cross-coupling
catalytic protocol (Figure 2). Aryl bromides with functional
groups, for example, -CN, -COCH3, -CHO, and -CO2Et, were
suitable arylating reagents (64–86% yield). Nitrogen-con-
taining heteroaromatic substrates, such as bromo-substituted
quinoline, pyridine, and pyrimidine, served as heteroarylating
reagents (58–78% yield). Benzothiazole was functionalized at
the C2 position in good yields. The C2 position of 2,4-
Considering the ground-state reduction potential of PyCÀ,
we anticipated that the redox power of this catalytic system
exceeds the reduction potentials of substituted aryl chlorides.
À
Consequently, we explored the SenI-ET method for C H
arylation reactions using (hetero)aryl chlorides as the (heter-
o)aryl radical precursors. Aside from aryl chlorides with
functional groups such as -CN or -CO2Me (see Figure 2),
chlorinated N-heterocycles, such as pyridines, quinolines, and
thiazoles, could be readily used as (hetero)aryl radical
precursors (61–87% yield). The reaction times were slightly
longer with the aryl chlorides than with the corresponding
bromides (compare the reaction times of 1a and 3a) because
À
À
À
dibromothiazole could be activated selectively for C H
of the stronger C Cl bonds and the relatively slow C Cl bond
arylation, leaving the other C Br bond intact for further
cleavage kinetics.[20] Importantly, this method can also be used
À
À
synthetic modifications.
for the activation of carbon–chlorine bonds for C H arylation
reactions in the presence of brominated trapping reagents. A
new carbon–carbon bond is formed by C Cl bond activation,
and simultaneously, a new C Br bond is introduced for
further functionalization (3d in Figure 2). Remarkably, the
SenI-ET method also enables the conversion of aryl tri-
flates[23] into aryl radicals upon visible-light excitation,
Interestingly, this redox-potential-dependent activation of
carbon–halogen bonds in organohalides allows the use of
halogenated (hetero)arenes as trapping reagents for the
functionalization of carbon–halogen bonds, introducing
another carbon–halogen bond for further synthetic modifica-
tions of the coupling product. For example, the reduction
À
À
Angew. Chem. Int. Ed. 2017, 56, 1 – 7
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
line were phosphonylated (5d and 5e) in excellent yields. The
use of other phosphites, such as (MeO)3P, also gave the
corresponding phosphonylated products in excellent yields
(5g, 81% yield). The phosphonylated product 6b was isolated
in 76% yield when the aryl triflate was used as the aryl radical
precursor.
Partially fluorinated aryl bromides were also used as the
À
C H arylating reagents using this photoredox catalytic
method.[24] For example, when 1,3-bis(trifluoromethyl)-5-
bromobenzene or 3-bromo-5-fluorobenzotrifluoride were
introduced as the precursors of the 1,3-bis(trifluoromethyl)-
benzene and 5-fluorobenzotrifluoride radicals, respectively, in
the presence of Ru(bpy)32+, only traces of the desired
products were obtained. The SenI-ET method, however,
provided products 7a, 7b, and 7c in very good yields.
The experimental results along with recent literature
reports[10,18,22] strongly support our hypothesis that the C H
À
arylation reactions proceed via radical mechanisms. In
addition, when 2-bromobenzonitrile was irradiated in the
presence of 1,1-diphenylethylene, a radical scavenger,[25] the
coupling product 8a was isolated in 26% yield (see Figure 3
and the Supporting Information for further details), providing
strong evidence for a radical reaction. The measured photo-
chemical quantum yield of 0.12 Æ 0.02 (12 Æ 2%) of the
photocatalytic reaction provides further support that discrete
Figure 3. Photoredox catalytic cascade reactions, photo-Arbuzov reac-
À
tions using aryl halides and trialkyl phosphites, and C H arylation
reactions using partially fluorinated aryl bromides.
photoredox electron-transfer events take place.[20,26] The C H
À
arylation reactions did not proceed in the dark, and the yields
were very low when the reactions were performed without
one of the catalytic components. In the presence of air
(Table S1, entry 7 and Figure S18), the yield was very low. The
extending the scope of aryl radical precursors that can be used
in visible-light photoredox catalysis (6a and 6c in Figure 3).
As the SenI-ET catalytic protocol operates via diffusion-
controlled interactions, we envisioned that the intrinsic redox
potential of Ru(bpy)3+ (ca. À1.33 V) and the redox potential
of the catalytic system could be used sequentially to explore
synthetically important photoredox cascade reactions with
polychlorinated aromatic substrates. For example, when 2,4,6-
2+
luminescence of Ru(bpy)3 was quenched efficiently in the
presence of Py, and the quenching efficiency was at least one
order of magnitude higher than the quenching efficiency of
2+
Ru(bpy)3 with DIPEA (see Figure 1 and Figure S2). The
reactions did not proceed in the presence of PAHs (with even
higher reduction potentials compared to Py) that possess
2+
trichloropyrimidine was used as a substrate in the presence of
higher triplet state energies than Ru(bpy)3 (see Figure 1)
2+
2+
À
Ru(bpy)3 and DIPEA, the C H-arylated product 4a was
and do not quench the luminescence of Ru(bpy)3 .
obtained in good yield. However, when the same substrate
was reacted under the SenI-ET photoredox catalytic con-
ditions, the doubly arylated product 4b was obtained. More-
over, two different coupling partners can be sequentially used
to yield the disubstituted compounds 4c and 4d, but isolation
of the intermediate is preferable to avoid the homocoupling
product.
All experiments, spectroscopic investigations, and litera-
ture reports[27] support the catalytic cycle proposed in
2+
Figure 4. Upon visible-light photoexcitation, Ru(bpy)3
transfers its energy to Py (as a representative PAH, see
Figure 1). The excited Py is then reductively quenched by
DIPEA to generate PyCÀ and the radical cation of DIPEA
+
(DIPEAC ). PyCÀ in its ground state transfers one electron to
We then explored the SenI-ET method for carbon–
heteroatom bond formation using nucleophiles as the trap-
ping partners for the aryl radicals (as in SRN1 reactions)[17]
under visible-light irradiation. In this case, the challenges are
competition with fast hydrogen atom abstraction of the aryl
radical from the solvent and the radical cation of DIPEA, and
the stability of the sensitizer in the presence of nucleophiles.
We therefore selected (EtO)3P as the coupling partner as its
addition to aryl radicals is very fast. The expected phospho-
nylated products were obtained in excellent yields (products
isolated in up to 89% yield; Figure 3) when aryl bromides and
chlorides were irradiated in the presence of (EtO)3P using the
SenI-ET catalytic protocol. Aside from aryl halides, hetero-
aryl halides such as 5-bromopyrimidine and 3-bromoquino-
the (hetero)aryl halide, yielding the (hetero)aryl radical
precursor (Het)ArXCÀ and regenerating neutral Py to com-
plete the catalytic cycle. Fragmentation of (Het)ArXCÀ yields
the (hetero)aryl radical, which reacts with (hetero)arenes,
À
alkenes, or nucleophiles ((EtO)3P) to yield C C and
À
C heteroatom coupling products. In a competing pathway,
the aryl radical abstracts a hydrogen atom either from
DIPEA·+ or from the solvent (in this case DMSO) to give
undesired reduction products in minimal amounts and
diisopropylamine (as confirmed by GC and GC-MS analysis
of the crude reaction mixture). The inefficient photoredox
2+
reaction of excited-state Ru(bpy)3 and DIPEA competes
with the energy transfer from Ru(bpy)3 to Py but the rate
2+
constant is smaller by at least one order of magnitude.
4
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 7
These are not the final page numbers!

Angewandte
Communications
Chemie
[1] R. E. Blankenship, Molecular mechanisms of photosynthesis,
2nd ed., Wiley-Blackwell, Hoboken, 2014.
[2] T. Mirkovic, E. E. Ostroumov, J. M. Anna, R. Van Grondelle,
Govindjee, G. D. Scholes, Chem. Rev. 2017, 117, 249 – 293.
[3] C. K. Prier, D. A. Rankic, D. W. C. MacMillan, Chem. Rev. 2013,
113, 5322 – 5363.
[4] T. P. Yoon, M. A. Ischay, J. N. Du, Nat. Chem. 2010, 2, 527 – 532.
[5] J. M. R. Narayanam, C. R. J. Stephenson, Chem. Soc. Rev. 2011,
40, 102 – 113.
[6] N. A. Romero, D. A. Nicewicz, Chem. Rev. 2016, 116, 10075 –
10166.
À
Figure 4. Proposed catalytic cycle for the SenI-ET catalytic C H aryla-
tion reactions using N-methylpyrrole as a representative trapping
reagent. For the mechanism of the phosphonylation of aryl halides and
pseudohalides, see Figure S29.
[7] J. Xuan, W. J. Xiao, Angew. Chem. Int. Ed. 2012, 51, 6828 – 6838;
Angew. Chem. 2012, 124, 6934 – 6955.
[8] For example, although both Ru(bpy)32+ and Ir(ppy)3 are excited
at similar visible-light wavelengths, depending on the solvent,
the maximum available reduction potentials of the catalytic
systems differ by the difference of their intrinsic ground-state
reduction potentials of ca. 0.9 V.
Notably, PyCÀ could also eject an electron upon photoexcita-
tion.[27] However, we excluded this possibility considering the
observed substrate scope and the use of non-aqueous solvents.
In conclusion, sensitization-initiated photoredox electron
transfer enhances the conversion of visible-light energy into
redox energy for synthetic applications. The efficient funnel-
ing of the absorbed light energy enables the use of low-energy
green light (lex = 530 nm, E = 2.3 eV) to access the redox
potentials of highly reducing polycyclic aromatic hydrocar-
bons, such as pyrene or anthracene (ca. 0.8 V more negative
[9] Compare the reduction and oxidation potentials of Ru(bpy)32+
and Ru(bpz)32+ or fac-Ir(ppy)3 and Ir[dF(CF3)ppy]2(dtbbpy)+ in
acetonitrile. For the respective reduction and oxidation poten-
tials, see Ref. [3].
[10] I. Ghosh, T. Ghosh, J. I. Bardagi, B. Konig, Science 2014, 346,
725 – 728.
[11] D. Rehm, A. Weller, Isr. J. Chem. 1970, 8, 259 – 271.
[12] M. Montalti, A. Credi, L. Prodi, M. T. Gandolfi, Handbook of
photochemistry, 3rd ed., CRC, Taylor & Francis, Boca Raton,
2006.
+
than that of Ru(bpy)3 ) that only absorb in the UV region
(labs(Py) ꢀ 350 nm, E ꢁ 3.5 eV) for photoredox applications.
Aryl halides (including aryl chlorides and heteroaryl chlor-
ides) and pseudohalides are activated for carbon–carbon and
carbon–heteroatom bond-forming reactions under extremely
mild reaction conditions. The reaction times are short owing
to a good photochemical quantum efficiency of 12%. The
described SenI-ET method is a minimalistic chemical model
of biological photosynthesis and extends the span of reaction
methods in visible-light photoredox catalysis. Furthermore, it
enables the use of extremely reducing molecules that do not
absorb visible light for the activation of stable chemical bonds
under visible-light irradiation.
[13] M. Wrighton, J. Markham, J. Phys. Chem. 1973, 77, 3042 – 3044.
[14] S. P. Pitre, C. D. McTiernan, H. Ismaili, J. C. Scaiano, J. Am.
Chem. Soc. 2013, 135, 13286 – 13289.
[15] The generation of aryl radicals from aryl halides under photo-
redox catalysis is achieved by using strong bases, such as
KOtBu,[16] or nucleophiles under UV (lex ꢀ 350 nm) irradiation
(SRN1)[17] and in the presence of an excess of highly reactive
neutral organic reducing agents, such as N2,N2,N12,N12-tetra-
methyl-7,8-dihydro-6H-dipyrido[1,4]diazepine-2,12-diamine
and UV-A (365 nm) irradiation as introduced by Murphy and
co-workers.[21] Highly reducing conPET catalytic systems[10] and
Ir complexes[18,19] are also efficient in generating aryl radicals
from aryl halides but typically require longer reaction times.
[16] M. E. Budꢀn, J. F. Guastavino, R. A. Rossi, Org. Lett. 2013, 15,
1174 – 1177.
[17] R. A. Rossi, A. B. Pierini, A. B. Penenory, Chem. Rev. 2003, 103,
71 – 167.
Acknowledgments
[18] J. D. Nguyen, E. M. DꢁAmato, J. M. R. Narayanam, C. R. J.
Stephenson, Nat. Chem. 2012, 4, 854 – 859.
[19] H. Kim, C. Lee, Angew. Chem. Int. Ed. 2012, 51, 12303 – 12306.
[20] L. Pause, M. Robert, J. M. Saveant, J. Am. Chem. Soc. 1999, 121,
7158 – 7159.
[21] E. Cahard, F. Schoenebeck, J. Garnier, S. P. Y. Cutulic, S. Z.
Zhou, J. A. Murphy, Angew. Chem. Int. Ed. 2012, 51, 3673 –
3676; Angew. Chem. 2012, 124, 3733 – 3736.
We thank the Deutsche Forschungsgemeinschaft (GRK 1626)
for financial support. We also thank Dr. Rudolf Vasold and
Regina Hoheisel (University of Regensburg) for assistance
with GC-MS and cyclic voltammetry, respectively.
Conflict of interest
[22] I. Ghosh, L. Marzo, A. Das, R. Shaikh, B. Konig, Acc. Chem. Res.
2016, 49, 1566 – 1577.
[23] The reduction potentials of aryl triflates are very similar to those
of aryl halides (see Figure S10).
The authors declare no conflict of interest.
[24] Polyfluorinated aryl radicals (e.g., the pentafluorobenzene
radical) can be generated from their bromide precursors owing
to their extremely low reduction potentials, which are due to the
electron-withdrawing effect of the fluorine atoms, using Ru-
À
Keywords: C H arylation · electron transfer · energy transfer ·
photocatalysis · polycyclic aromatic hydrocarbons
2+
(bpy)3 or Eosin Y.
Angew. Chem. Int. Ed. 2017, 56, 1 – 7
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

Angewandte
Communications
Chemie
[25] W. Liu, H. Cao, H. Zhang, H. Zhang, K. H. Chung, C. A. He,
H. B. Wang, F. Y. Kwong, A. W. Lei, J. Am. Chem. Soc. 2010,
132, 16737 – 16740.
an extremely efficient overall photoredox catalytic process
combining several discrete energy- and photoredox electron-
transfer events.
[26] Note that five consecutive events (light absorption, energy
transfer, electron transfer from the sacrificial electron donor,
electron transfer to the substrate, and stepwise cleavage of the
carbon–halide bonds)[20] are involved in the generation of the
aryl radicals. Therefore, the quantum yield of 12% demonstrates
[27] C. Kerzig, M. Goez, Chem. Sci. 2016, 7, 3862 – 3868.
Manuscript received: March 22, 2017
Final Article published: && &&, &&&&
6
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 7
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Photocatalysis
Inspired by nature: In the presented two-
center photoredox catalytic approach,
Ru(bpy)3Cl2 absorbs visible light and
transfers the energy to polycyclic aro-
matic hydrocarbons, which in turn enable
an efficient redox process. This method
was used for the activation of (hetero)aryl
halides for the formation of carbon–
carbon and carbon–heteroatom bonds.
I. Ghosh, R. S. Shaikh,
B. Kçnig*
&&&& — &&&&
Sensitization-Initiated Electron Transfer
for Photoredox Catalysis
Angew. Chem. Int. Ed. 2017, 56, 1 – 7
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7
These are not the final page numbers!