Reduction of aryl halides by consecutive visible light-induced electron transfer processes

Article abstract of DOI:10.1126/science.1258232

Biological photosynthesis uses the energy of several visible light photons for the challenging oxidation of water, whereas chemical photocatalysis typically involves only single-photon excitation. Perylene bisimide is reduced by visible light photoinduced electron transfer (PET) to its stable and colored radical anion.We report here that subsequent excitation of the radical anion accumulates sufficient energy for the reduction of stable aryl chlorides giving aryl radicals, which were trapped by hydrogen atom donors or used in carbon-carbon bond formation. This consecutive PET (conPET) overcomes thecurrent energetic limitation of visible light photoredox catalysis and allows the photocatalytic conversion of less reactive chemical bonds in organic synthesis .

Full text of DOI:10.1126/science.1258232

Reduction of aryl halides by consecutive visible light-induced
electron transfer processes
Indrajit Ghosh et al.
Science 346, 725 (2014);
DOI: 10.1126/science.1258232
This copy is for your personal, non-commercial use only.
If you wish to distribute this article to others, you can order high-quality copies for your
colleagues, clients, or customers by clicking here.
Permission to republish or repurpose articles or portions of articles can be obtained by
following the guidelines here.
The following resources related to this article are available online at
www.sciencemag.org (this information is current as of November 6, 2014 ):
Updated information and services, including high-resolution figures, can be found in the online
version of this article at:
http://www.sciencemag.org/content/346/6210/725.full.html
Supporting Online Material can be found at:
http://www.sciencemag.org/content/suppl/2014/11/05/346.6210.725.DC1.html
This article cites 35 articles, 4 of which can be accessed free:
http://www.sciencemag.org/content/346/6210/725.full.html#ref-list-1
This article appears in the following subject collections:
Chemistry
http://www.sciencemag.org/cgi/collection/chemistry
Science (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by the
American Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. Copyright
2014 by the American Association for the Advancement of Science; all rights reserved. The title Science is a
registered trademark of AAAS.

RESEARCH
| REPORTS
At the boundary between two topologically
distinct segments of edge, localized exotic topo-
logical zero modes arise that give rise to topo-
logically protected degeneracies and projective
non-Abelian statistics (12). In the case of the Z2
sRVB state, the domain wall between e and m
edges localizes a Majorana fermion zero mode,
with the following physical consequences: Let
us consider the case of the superconductivity-
induced e edge, where an electron can coherently
enter the QSL as a fermionic spinon. If this pro-
cess occurs in the vicinity of the domain wall be-
tween an e and m edge, then the fermionic spinon
can also emit or absorb a vison from the m edge,
thus becoming a bosonic spinon. In other words,
the Majorana fermion zero mode is a source or
sink of fermion parity, allowing the electron to
coherently enter into the spin liquid as a bosonic
spinon. If the fermionic spinon in the bulk of the
QSL can decay into a vison and a bosonic spinon,
then this geometry (Fig. 1B) will allow the Tomasch
oscillations to be observed in the tunneling con-
ductance. Similar considerations show that when
the e edge is induced by magnetism, the electron
can enter into the spin liquid as a bosonic holon
in the vicinity of the e-m domain wall (Fig. 1D).
The considerations outlined here suggest ways
to tune through the topological phase transition
that separates the e and m edges, such as by
applying a magnetic field to the edge of an easy-
plane QSL. This can be done by taking a thin
sample and shielding the bulk of the QSL by
sandwiching it between two superconductors.
At the critical field for the edge quantum phase
transition, there will be enhanced thermal tran-
sport through the edge, leading to a nonzero
15. M. Levin, Phys. Rev. X 3, 021009 (2013).
description of such phases should be adequate for present
purposes: More formally, it gives a leading order description
of the edge modes in the vicinity of an edge quantum critical
point where the edge gaps, but not the bulk gaps, are
vanishingly small.
16. A. Kapustin, N. Saulina, Nucl. Phys. B 845, 393–435 (2011).
17. N. H. Lindner, E. Berg, G. Refael, A. Stern, Phys. Rev. X 2,
041002 (2012).
18. D. J. Clarke, J. Alicea, K. Shtengel, Nat. Commun. 4, 1348 (2013).
19. M. Cheng, Phys. Rev. B 86, 195126 (2012).
34. W. J. Tomasch, Phys. Rev. Lett. 15, 672–675 (1965).
35. W. L. McMillan, P. W. Anderson, Phys. Rev. Lett. 16, 85–87
(1966).
20. M. Punk, D. Chowdhury, S. Sachdev, Nat. Phys. 10, 289–293
(2014).
21. T.-H. Han et al., Nature 492, 406–410 (2012).
22. Y. Qi, C. Xu, S. Sachdev, Phys. Rev. Lett. 102, 176401 (2009).
23. These were identified long ago in the toric code model (38)
in terms of rough versus smooth edges (39), although the
generic necessity to realize one or the other in resonating
valence bond states and their general independence of lattice
structure has only recently been fully appreciated.
24. D. S. Rokhsar, S. A. Kivelson, Phys. Rev. Lett. 61, 2376–2379
(1988).
25. T. Senthil, O. Motrunich, Phys. Rev. B 66, 205104 (2002).
26. M. Freedman, C. Nayak, K. Shtengel, K. Walker, Z. Wang,
Ann. Phys. 310, 428–492 (2004).
27. X.-G. Wen, Quantum Field Theory of Many-Body Systems
(Oxford Univ. Press, Oxford, 2004).
28. H.-C. Jiang, Z. Wang, L. Balents, Nat. Phys. 8, 902–905 (2012).
29. H.-C. Jiang, H. Yao, L. Balents, Phys. Rev. B 86, 024424 (2012).
30. S. Yan, D. A. Huse, S. R. White, Science 332, 1173–1176 (2011).
31. S. Depenbrock, I. P. McCulloch, U. Schollwöck, Phys. Rev. Lett.
109, 067201 (2012).
32. See the supplementary materials on Science Online.
33. If the resulting gaps are large, the field theory description of
the edge is no longer controlled, but because the most
important qualitative features of a gapped state do not
depend on the magnitude of the gap, the field theory
36. S. Kivelson, D. Rokhsar, Phys. Rev. B 41, 11693–11696 (1990).
37. M. Barkeshli, H. Yao, S. A. Kivelson, Phys. Rev. B 87, 140402
(2013).
38. A. Kitaev, Ann. Phys. 303, 2–30 (2003).
39. S. B. Bravyi, A. Y. Kitaev, http://arxiv.org/abs/quant-ph/
9811052.
ACKNOWLEDGMENTS
This work was supported by NSF-DMR grant 1265593 (S.K.), the
Israel Science Foundation, the Israel-U.S. Binational Foundation,
the Minerva Foundation, and the German-Israeli Foundation (E.B.).
M.B. thanks X.-L. Qi for recent collaborations on related topics.
We thank E. Altman, C. Nayak, M. Freedman, Z. Wang,
K. Walker, M. Hastings, M. P. A. Fisher, B. Bauer, and T. Grover
for discussions.
SUPPLEMENTARY MATERIALS
www.sciencemag.org/content/346/6210/722/suppl/DC1
Supplementary Text
References (40–45)
12 March 2014; accepted 2 October 2014
10.1126/science.1253251
PHOTOCHEMISTRY
Reduction of aryl halides by
consecutive visible light-induced
intercept at low temperatures in the thermal
2
2
kB
conductance: limT→0k=T ¼ NLc p , where NL
electron transfer processes
3
h
is the number of layers in the QSL, c = 1/2 is the
central charge of the edge at the critical point,
and kB is Boltzmann’s constant. Because neither
the trivial paramagnet nor the doubled-semion
QSL have topologically distinct types of gapped
boundaries, the observation of a topological quan-
tum phase transition at the edge of a gapped in-
sulating spin system would prove the existence of
a fractionalized spin-liquid state and rule out the
doubled-semion state. The present considerations
are readily extended to other sorts of topologi-
cally ordered states, such as those that occur in
fractional quantum Hall systems.
Indrajit Ghosh,* Tamal Ghosh,* Javier I. Bardagi, Burkhard König†
Biological photosynthesis uses the energy of several visible light photons for the
challenging oxidation of water, whereas chemical photocatalysis typically involves only
single-photon excitation. Perylene bisimide is reduced by visible light photoinduced
electron transfer (PET) to its stable and colored radical anion. We report here that
subsequent excitation of the radical anion accumulates sufficient energy for the reduction
of stable aryl chlorides giving aryl radicals, which were trapped by hydrogen atom donors
or used in carbon-carbon bond formation. This consecutive PET (conPET) overcomes the
current energetic limitation of visible light photoredox catalysis and allows the
photocatalytic conversion of less reactive chemical bonds in organic synthesis.
isible light provides sufficient energy to
promote challenging chemical reactions.
Biological photosynthesis as the omni-
present example uses a visible portion of
the solar spectrum to separate charges by
single-photon excitation of dye molecules, such
as redox-active coordination compounds [e.g.,
Ru(bpy)32+ or Ir(ppy)3] (3, 5), conjugated organics
(e.g., eosin Y) (6), or inorganic semiconductors
(e.g., CdS) (7) mediates photoinduced electron
or energy transfer process to substrates.
A recent application is the generation of highly
reactive aryl radicals, which are useful arylating
reagents in synthesis, by photoinduced electron
transfer (PET) from photoredox catalysts to suit-
able precursors followed by bond scission (8, 9).
However, the choice of aryl radical precursors is
currently limited to electron-poor arenes, such as
diazonium (6, 10) or iodonium (11) salts, or in a
few cases aryl iodides (9, 12), with weakly bound
leaving groups, due to the accessible reducing
REFERENCES AND NOTES
1. P. W. Anderson, Mater. Res. Bull. 8, 153–160 (1973).
2. S. A. Kivelson, D. S. Rokhsar, J. P. Sethna, Phys. Rev. B 35,
8865–8868 (1987).
3. N. Read, S. Sachdev, Phys. Rev. Lett. 66, 1773–1776 (1991).
4. X. G. Wen, Phys. Rev. B 44, 2664–2672 (1991).
5. T. Senthil, M. P. A. Fisher, Phys. Rev. B 62, 7850–7881 (2000).
6. R. Moessner, S. L. Sondhi, Phys. Rev. Lett. 86, 1881–1884 (2001).
7. L. Balents, M. P. A. Fisher, S. M. Girvin, Phys. Rev. B 65, 224412
(2002).
8. A. Seidel, Phys. Rev. B 80, 165131 (2009).
9. L. Balents, Nature 464, 199–208 (2010).
10. M. Barkeshli, X.-L. Qi, Phys. Rev. X 2, 031013 (2012).
11. M. Barkeshli, C.-M. Jian, X.-L. Qi, Phys. Rev. B 87, 045130 (2013).
12. M. Barkeshli, C.-M. Jian, X.-L. Qi, Phys. Rev. B 88, 241103(R)
(2013).
V
electron transfer, providing the energy for water
oxidation. This transformation (water to oxygen,
protons, and electrons) requires the cumulative
absorption of four photons (1). In past decades,
visible light–mediated chemical photoredox ca-
talysis has emerged into a conceptually related
valuable method for organic synthesis (2–4). Here,
Institute of Organic Chemistry, University of Regensburg,
D-93040 Regensburg, Germany.
*These authors contributed equally to this work. †Corresponding
13. S. Beigi, P. W. Shor, D. Whalen, Commun. Math. Phys. 306,
663–694 (2011).
14. A. Kitaev, L. Kong, Commun. Math. Phys. 313, 351–373 (2012).
author. E-mail: burkhard.koenig@ur.de
SCIENCE sciencemag.org
7 NOVEMBER 2014 • VOL 346 ISSUE 6210 725

RESEARCH
| REPORTS
Fig. 1. Chemical structure of the photocatalyst PDI, one electron reduction of PDI to its radical anion, and effects of Et3N and 4′-bromoacetophenone
on its photophysical properties. (A) Changes in the fluorescence spectra (in this case, intensity; lEx = 455 nm) of PDI upon successive addition of Et3N in
DMF. In the insets, changes in the fluorescence spectra of PDI upon addition of (i) 4′-bromoacetophenone, and changes in the absorption spectra of PDI upon
addition of (ii) Et3N, and (iii) 4′-bromoacetophenone are shown. (B) Formation of the PDI radical anion (PDI•–) upon photoexcitation (lEx = 455 nm) of PDI in
the presence of Et3N. In the inset, regeneration of neutral PDI from PDI•– upon exposure to air is shown (see also fig. S5).
power of typical visible light photoredox catalysts
(13, 14). Two mechanistic scenarios for the aryl
radical generation can be considered: (i) oxidative
quenching of the excited photoredox catalyst by
the aryl radical precursor, which must be exer-
gonic or, if the lifetime of the excited state is long
and the subsequent bond cleavage fast and ir-
reversible, at least thermo neutral (15); (ii) oxida-
tion of the photoreduced catalyst in its ground
state (fig. S1, left), which can be slightly ender-
gonic. The energy conferred by visible light excita-
tion for subsequent reduction chemistry is limited
by the energy of a single absorbed photon. The
energy of blue photons (440 nm) of 270 KJ/mol
or 2.8 eV defines a maximum theoretical energy
threshold between the donor (i.e., photocatalyst)
and acceptor (i.e., substrate). In addition, part of
the accessible energy is always lost due to inter-
system crossing and reorganization of the excited
states of the photocatalysts by nonradiative path-
ways. In the case of Ru complexes, this loss is
~0.6 eV (3). As a consequence, the available en-
ergy of typical photocatalysts just reaches the reduc-
tion potential (E°) of aryl iodides (9), defining the
current synthetic scope of photoredox catalysis.
Here, we report a practical approach to over-
come the limitations of visible light–mediated
chemical photocatalysis by using the energies of
two photons in one catalytic cycle (16). Photo-
catalytic alkylation or arylation reactions as re-
ported by MacMillan (3, 17), Stephenson (18), Yoon
(2), and others (6) employ a typical PET process
(fig. S1, left). The excited dye becomes a stronger
oxidant (and reductant) and is converted into its
radical anion, which activates substituted benzyl
Fig. 2. Photoreduction of aryl halides. Yields (%) and reaction times (hours) are given.
726 7 NOVEMBER 2014 • VOL 346 ISSUE 6210
sciencemag.org SCIENCE

RESEARCH
| REPORTS
bromides (3, 19), alpha bromo carbonyl com-
pounds (20), aryl iodides (9, 12), diazonium (6),
and iodonium (11) salts. However, compounds
that are less reactive (e.g., aryl bromides and chlo-
rides) (21–23) due to a more negative reduction
potential, higher carbon-halide bond dissociation
energy, and a different, stepwise cleavage mech-
anism (23) are not accessible by this process using
typical photocatalysts and, more importantly, vis-
ible light. Our approach is inspired by the Z scheme
of biological photosynthesis, which has already
been used in water photooxidation (24) but, sur-
prisingly, has not yet been applied in organic
synthesis. The energy of a second visible light
excitation can be added to the process if the
radical anion of the dye is reasonably stable in
the ground state, colored, and thus can be ex-
cited again by visible light (fig. S1, right).
istic combination of thermal- and photostability
and optical and redox properties (25)—fulfill the
requirements of such a biomimetic organic dye–
based catalytic system. Among different perylene
diimides N,N-bis(2,6-diisopropylphenyl)perylene-
3,4,9,10-bis(dicarboximide) (PDI) (see Fig. 1 for
chemical structure) was selected due to its better
solubility in N,N´-dimethylformamide (DMF) and
dimethyl sulfoxide (DMSO) (solvents used in this
study). Upon irradiation with blue light (455 nm)
in the presence of triethylamine (Et3N) as electron
donor, PDI forms a colored radical anion PDI•–
(Fig. 1 and figs. S3 to S5) that can again be excited
by visible light (26). In the absence of oxygen, the
radical anion is very stable. Spectroscopic inves-
tigations confirmed that electron transfer from
Et3N to PDI requires photoexcitation (Fig. 1 and
fig. S4) (27).
a mixture of 4′-bromoacetophenone, PDI (5 mol%),
and Et3N with blue light (455 nm) (see table S1).
In DMSO as solvent, the reduction product aceto-
phenone was obtained in 4 hours with 47% photo-
reduction yield (entry 16, table S1). Continuous
irradiation of the reaction mixture for 8 hours
gave 69% yield. Use of DMF as solvent gave com-
parable or slightly better yields (see entry 10 in
Fig. 2 and table S1) in a shorter reaction time of
4 hours. Control experiments confirmed that
PDI, electron donor, and light irradiation are
necessary for the photoreduction reaction to
occur (entries 1 to 6 in table S1).
Using the optimized conditions, the reaction
scope was explored with a range of substituted
aryl bromides giving the corresponding reduc-
tion products in good to nearly quantitative yields
(Fig. 2). The reduction potentials of NMe2- and
OMe-substituted aryl bromides are too high to
be reached employing these photoreduction con-
ditions, whereas p-nitro-substituted aryl bromides
have such a low fragmentation rate (22) that back
electron transfer becomes dominant. The photo-
reduction reactions could also be performed under
sunlight (entry 10 in Fig. 2) or with 530-nm light-
emitting diodes as the absorption spectrum of PDI
spans a broad portion of the visible spectrum
(fig. S3). Substituted aryl iodides having slightly
lower reduction potentials than aryl bromides (22)
gave comparable photoreduction yields (see entries
1 to 6 in Fig. 2). Notably, an aryl–iodine bond was
chemoselectively reduced in the presence of a
bromine substituent (entry 6 in Fig. 2).
Perylene diimides—a class of fluorescent dye
molecules that have been used as pigments, col-
orants, photoreceptors, and, more recently, as
electronic materials because of their character-
A synthetic application of this catalytic system
is the photoreduction of aryl halides, including
aryl chlorides, using visible light irradiation. The
reaction conditions were optimized by irradiating
A commercially available catalyst, N,N′-bis
(3-pentyl)perylene-3,4,9,10-bis(dicarboximide) (for
the chemical structure, see fig. S2), gave similar
yields when the reaction mixtures were irradiated
for 8 hours (see entries 14 and 15 in table S1). The
slightly slower reaction rate is attributed to its
poorer solubility.
Based on the reduction potential of PDI/PDI•–
[–0.37 V versus saturated calomel electrode (SCE)]
and the E0–0 transition energy of PDI•–, we esti-
mated a reducing power of the excited state
PDI•–* according to the Rehm-Weller equation
(28) that reaches or exceeds the reduction po-
tentials of substituted aryl chlorides (26, 29).
This class of compounds, although easily acces-
sible and relatively inexpensive, has not been
considered in visible light photocatalysis because
of their low reactivity due to high reduction po-
tentials, high carbon-chlorine bond energies,
and a stepwise fragmentation mechanism. To
the best of our knowledge, the reduction of aryl
chlorides has only been achieved using strong
bases (30, 31), such as potassium tert-butoxide,
or nucleophiles under ultraviolet (UV) (lEx
≤
350 nm) irradiation (SRN1) (32)and in the presence
of an excess of highly reactive neutral organic re-
ducing agents, such as N2,N2,N12,N12-tetramethyl-
7,8-dihydro-6H-dipyrido[1,4]diazepine-2,12-diamine
(for the chemical structure, see fig. S2) and UV-A
(365 nm) irradiation as introduced by Murphy
(21). The consecutive PET (conPET) process gen-
erates the strong reducing PDI•–* in situ by two
subsequent visible light excitations starting from
air-stable PDI. This avoids the use of highly air- and
Fig. 3. C–H aromatic substitution reactions of aryl halides with substituted pyrroles and
intramolecular addition to an alkene.
SCIENCE sciencemag.org
7 NOVEMBER 2014 • VOL 346 ISSUE 6210 727

RESEARCH
| REPORTS
radicals are obtained from stable, and in the case
of aryl chlorides, inexpensive bulk chemicals, under
very mild and metal-free reaction conditions.
REFERENCES AND NOTES
1. J. J. Concepcion, R. L. House, J. M. Papanikolas, T. J. Meyer,
Proc. Natl. Acad. Sci. U.S.A. 109, 15560–15564 (2012).
2. D. M. Schultz, T. P. Yoon, Science 343, 1239176 (2014).
3. C. K. Prier, D. A. Rankic, D. W. C. MacMillan, Chem. Rev. 113,
5322–5363 (2013).
4. J. Xuan, W. J. Xiao, Angew. Chem. Int. Ed. 51, 6828–6838 (2012).
5. F. Teplý, Collect. Czech. Chem. Commun. 76, 859–917 (2011).
6. D. P. Hari, P. Schroll, B. Koenig, J. Am. Chem. Soc. 134,
2958–2961 (2012).
7. D. Ravelli, D. Dondi, M. Fagnoni, A. Albini, Chem. Soc. Rev. 38,
1999–2011 (2009).
Fig. 4. Proposed catalytic mechanism. S–H, solvents.
8. D. P. Hari, B. Koenig, Angew. Chem. Int. Ed. 52, 4734–4743 (2013).
9. J. D. Nguyen, E. M. D’Amato, J. M. R. Narayanam,
C. R. J. Stephenson, Nat. Chem. 4, 854–859 (2012).
10. D. P. Hari, T. Hering, B. Koenig, Angew. Chem. Int. Ed. 53, 725 (2014).
11. Y. X. Liu et al., Synlett 24, 507–513 (2013).
moisture-sensitive donor molecules, which require
strict inert reaction conditions, UV-A irradiation,
and strongly basic conditions that are incom-
patible with many functional groups. Aryl chlorides
bearing electron withdrawing groups gave the cor-
responding reduction products with good to ex-
cellent yields in photocatalytic conditions (entries
13 to 17 in Fig. 2) that require only mixing of sub-
strates, PDI, Et3N, and irradiation with visible light.
Next, we applied the aryl radical intermediates
for C–C bond-forming arylation reactions. Chal-
lenge in this case is competition from fast hy-
drogen abstraction of the aryl radical from the
solvent and the radical cation of Et3N (33). We
therefore selected as the reaction partner N-
heterocyclic pyrroles, which were found to have
high reaction rates in the addition of radicals.
Substantial amounts of the expected arylation
product were indeed obtained by irradiating aryl
halides in the presence of N-methylpyrrole and
catalytic amounts of PDI. The reduction product
is a minor by-product but dominates when fu-
ran or thiophene are used as reaction partner.
Changing the solvent from DMF, which favors
the reduction product, to DMSO improved the
yields greatly. Isolated yields of functionalized
N-methylpyrrole derivatives obtained from differ-
ent substituted aryl halides are depicted in Fig. 3.
As in the photoreduction reaction, the C–H aryl-
ation reaction with N-methylpyrrole could also
be performed with N,N′-bis(3-pentyl)perylene-
3,4,9,10-bis(dicarboximide) (entry 6 in Fig. 3).
The reaction scope was extended to other pyrrole
derivatives affording arylated products in good
to excellent yields (Fig. 3).
The photoreduction of aryl halides appears to
proceed via a radical mechanism (23, 34), as evi-
denced by the conversion of 2-(allyloxy)-1,3,5-
tribromobenzene to the 5-exo cyclization product
5,7-dibromo-3-methyl-2,3-dihydrobenzofuran
(entry 12, Fig. 3), which implies a radical inter-
mediate (12, 34). Furthermore, the reduction reac-
tion of 4′-bromoacetophenone in the presence
of 2,2,6,6-tetramethylpiperidinoxyl (TEMPO) gave
the expected TEMPO adduct (see the supplemen-
tary materials). The formation of a PDI dianion,
which could be formed via a two-electron reduc-
tion (35), was not detected under the reaction
conditions (compare Fig. 1B and fig. S4 with
fig. S3; fig. S3 shows the absorption spectrum
of the electrochemically generated PDI dianion).
The photoreduction of 4′-bromoacetophenone
was minimal in air, preventing the formation
of the PDI radical anion (entry 6 in table S1,
Fig. 1B, and fig. S5). In the absence of light, no
reduction product is obtained: 4′-bromoaceto-
phenone added to a photochemically gener-
ated PDI radical anion (by photo-irradiating
the mixture of PDI and Et3N) and kept in the
dark for 4 hours was not converted (entries 9
and 11 in table S1). When the reaction mixture
was then illuminated with 455-nm light, aceto-
phenone was obtained in yields comparable to
the normal photoreduction protocol. Reduc-
tion of 4′-bromoacetophenone also did not occur
when the substrate was added to a chemically
generated [using (Et4N)2S2O4 as chemical reduc-
tant of PDI] radical anion (see entry 10 in table
S1). Degradation products of the catalyst formed
during the course of the reaction may still con-
tribute to substrate conversion because only the
perylene core is required (compare PDI and N,N′-
bis(3-pentyl)perylene-3,4,9,10-bis(dicarboximide):
The substituents in the amide nitrogens of per-
ylene have almost no influence on the photo-
physical properties and are mainly introduced to
increase the solubility of perylene diimides) (25).
All experiments support the proposed catalytic
cycle shown in Fig. 4. Excited PDI* is reductively
quenched by Et3N to give PDI•– and the radical
cation of triethylamine (Et3N•+) (27). Upon the
second excitation, PDI•–* reduces the substrate yield-
ing the aryl radical precursor (ArX•–) and regen-
erating the neutral PDI. Fragmentation of ArX•–
yields the aryl radical, which abstracts a hydro-
gen atom from either Et3N•+ or solvents to yield
the reduction products, or reacts with unsat-
urated compounds yielding C–C coupling pro-
ducts. Gas chromatography–mass spectrometry
(GC-MS) analysis of the crude product mixture
confirmed the formation of diethylamine, and
hydrogen atom abstraction reduction reactions
in D7-DMF gave deuterated products (see the
supplementary materials).
12. H. Kim, C. Lee, Angew. Chem. Int. Ed. 51, 12303–12306 (2012).
13. The excited state redox potential of fac-Ir(ppy)3 is estimated
to be –1.73 versus SCE (9). Some metal complexes with even
stronger reducing power in the excited state have been
reported but have not been applied to photoredox catalysis (14).
14. S. B. Harkins, J. C. Peters, J. Am. Chem. Soc. 127, 2030–2031 (2005).
15. The Rehm-Weller equation estimates the reduction potential
of a photoredox catalyst in its excited state from the ground-
state redox potential and the E0-0 transition energy (28).
16. For a recent example of two-step photoexcitation (Z scheme)
of heterogeneous semiconductor in water splitting, see (24).
17. M. T. Pirnot, D. A. Rankic, D. B. C. Martin, D. W. C. MacMillan,
Science 339, 1593–1596 (2013).
18. L. Furst, B. S. Matsuura, J. M. R. Narayanam, J. W. Tucker,
C. R. J. Stephenson, Org. Lett. 12, 3104–3107 (2010).
19. J. M. R. Narayanam, J. W. Tucker, C. R. J. Stephenson, J. Am.
Chem. Soc. 131, 8756–8757 (2009).
20. D. A. Nicewicz, D. W. C. MacMillan, Science 322, 77–80 (2008).
21. E. Cahard et al., Angew. Chem. Int. Ed. 51, 3673–3676 (2012).
22. C. Costentin, M. Robert, J. M. Savéant, J. Am. Chem. Soc. 126,
16051–16057 (2004).
23. L. Pause, M. Robert, J. M. Saveant, J. Am. Chem. Soc. 121,
7158–7159 (1999).
24. Y. Sasaki, H. Kato, A. Kudo, J. Am. Chem. Soc. 135, 5441–5449
(2013).
25. F. Würthner, Chem. Commun. 2004 (14), 1564–1579 (2004).
26. D. Gosztola, M. P. Niemczyk, W. Svec, A. S. Lukas,
M. R. Wasielewski, J. Phys. Chem. A 104, 6545–6551 (2000).
27. M. J. Tauber, R. F. Kelley, J. M. Giaimo, B. Rybtchinski,
M. R. Wasielewski, J. Am. Chem. Soc. 128, 1782–1783 (2006).
28. D. Rehm, A. Weller, Isr. J. Chem. 8, 259–271 (1970).
29. C. Ramanan, A. L. Smeigh, J. E. Anthony, T. J. Marks,
M. R. Wasielewski, J. Am. Chem. Soc. 134, 386–397 (2012).
30. M. E. Budén, J. F. Guastavino, R. A. Rossi, Org. Lett. 15,
1174–1177 (2013).
31. Y. Cheng, X. Gu, P. Li, Org. Lett. 15, 2664–2667 (2013).
32. R. A. Rossi, A. B. Pierini, A. B. Peñéñory, Chem. Rev. 103,
71–168 (2003).
33. Z. Chami, M. Gareil, J. Pinson, J. M. Saveant, A. Thiebault,
J. Org. Chem. 56, 586–595 (1991).
34. M. Newcomb, in Encyclopedia of Radicals in Chemistry,
Biology and Materials, C. S. Chatgilialoglu, A. Studer, Eds.
(Wiley, Chichester, UK, 2012)pp. 107–124.
35. E. Shirman et al., J. Phys. Chem. B 112, 8855–8858 (2008).
ACKNOWLEDGMENTS
We thank the Deutsche Forschungsgemeinschaft (GRK 1626)
for financial support. J.I.B. thanks the Alexander von Humboldt
Foundation for a scholarship. We thank R. Vasold for his assistance
in GC-MS measurements.
SUPPLEMENTARY MATERIALS
www.sciencemag.org/content/346/6210/725/suppl/DC1
Materials and Methods
Supplementary Text
Figs. S1 to S5
Table S1
References (36–39)
Two conPET steps using perylene diimide
dyes accumulate the energy from two visible light
excitations. The process is a minimalistic chem-
ical model of the Z scheme in biological pho-
tosynthesis and extends the scope of visible light
photocatalysis to aryl chlorides. Highly reactive aryl
2 July 2014; accepted 1 October 2014
10.1126/science.1258232
728 7 NOVEMBER 2014 • VOL 346 ISSUE 6210
sciencemag.org SCIENCE