Eosin Y as a Direct Hydrogen-Atom Transfer Photocatalyst for the Functionalization of C?H Bonds

Article abstract of DOI:10.1002/anie.201803220

Eosin Y, a well-known economical alternative to metal catalysts in visible-light-driven single-electron transfer-based organic transformations, can behave as an effective direct hydrogen-atom transfer catalyst for C?H activation. Using the alkylation of C?H bonds with electron-deficient alkenes as a model study revealed an extremely broad substrate scope, enabling easy access to a variety of important synthons. This eosin Y-based photocatalytic hydrogen-atom transfer strategy is promising for diverse functionalization of a wide range of native C?H bonds in a green and sustainable manner.

Full text of DOI:10.1002/anie.201803220

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Eosin Y as a Direct Hydrogen Atom Transfer Photocatalyst for
the Functionalization of C-H Bonds
Authors: Xuan-Zi Fan, Jia-Wei Rong, Hao-Lin Wu, Quan Zhou, Hong-
Ping Deng, Jin Da Tan, Cheng-Wen Xue, Li-Zhu Wu, Hai-
Rong Tao, and Jie Wu
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201803220
Angew. Chem. 10.1002/ange.201803220
Link to VoR: http://dx.doi.org/10.1002/anie.201803220
http://dx.doi.org/10.1002/ange.201803220

10.1002/anie.201803220
Angewandte Chemie International Edition
COMMUNICATION
Eosin Y as a Direct Hydrogen Atom Transfer Photocatalyst for the
Functionalization of CH Bonds
Xuan-Zi Fan,† Jia-Wei Rong,† Hao-Lin Wu, Quan Zhou, Hong-Ping Deng, Jin Da Tan, Cheng-Wen Xue,
Li-Zhu Wu, Hai-Rong Tao, and Jie Wu*
Abstract: Eosin Y, a well-known economical alternative to metal
catalysts in visible-light-driven single-electron transfer-based organic
transformations, can behave as an effective direct hydrogen atom
transfer catalyst for CH activation. Using the alkylation of CH
bonds with electron-deficient alkenes as a model study revealed an
extremely broad substrate scope, enabling easy access to a variety
of important synthons. This eosin Y-based photocatalytic hydrogen
atom transfer strategy is promising for diverse functionalization of a
wide range of native CH bonds in a green and sustainable manner.
performing direct HAT,[2a] being restricted to the families of
benzophenones, quinones, polyoxometalates,[6] and
few
a
others, such as the uranyl cation[3e] (Scheme 1b). Moreover,
each of these catalysts has specific limitations. Activation of
benzophenones and quinones normally requires ultraviolet (UV)
light. Furthermore, a high catalyst loading of diaryl ketones is
generally required as the ketyl radical intermediate formed in situ
can undergo side reactions, such as self-dimerization.[1a] Both
polyoxometalates and uranyl cation require the use of toxic
transition metals. The former requires UV light,[6] and the latter
suffers from a limited substrate scope.[3e] Therefore, it would be
ideal to discover a direct HAT catalyst that is readily available,
metal-free, activated by visible-light (hv 400-750 nm), and able
to avoid side reactions.
Dramatic developments in photocatalysis over the past decade
have enabled previously inaccessible transformations.[1] In
addition to single-electron transfer (SET) and energy transfer,
hydrogen atom transfer (HAT) has been frequently involved in
photocatalysis, which can activate substrates without the
limitation of redox potentials, offering enormous opportunities for
CH activations.[2] Upon light absorption, HAT catalysis is
normally achieved in three different ways.[2a] The first strategy is
the direct HAT process, where the activated photocatalyst
behaves as a HAT catalyst to abstract a H atom from a
substrate. The catalytic cycle is subsequently turned over
through a reverse hydrogen atom transfer (RHAT) to one of the
generated intermediates.[3] The second mode is the indirect HAT
process, where the excited photocatalyst activates another
catalyst through SET or energy transfer, and the latter promotes
the following HAT process.[4] The last route is the proton-coupled
electron transfer (PCET) process, which involves a concerted
transfer of an electron and a proton.[5] Among these three
pathways, direct HAT catalysis represents the most reagent-
and redox-economical method, as both the indirect HAT and
PCET processes require other additives.
Photo-induced direct HAT catalysis has endowed with a large
scope, allowing for the selective introduction of a plethora of
functional groups in place of the original CH bonds, e.g., via
alkylation,[3d]
vinylation,[3j]
alkynylation,[3i]
cyanation,[3m]
and
formylation,[3b]
carboxylation,[3f]
halogenation,[3e,k,l]
oxidation[3h] (Scheme 1a). However, the major bottleneck to its
wide application is the limited photocatalysts capable of
[*]
X.-Z. Fan, Dr. J.-W. Rong, Q. Zhou, Dr. H.-P. Deng, J. D. Tan, C.-W.
Xue, Dr. J. Wu
Department of Chemistry, National University of Singapore
3 Science Drive 3, Republic of Singapore, 117543
E-mail: chmjie@nus.edu.sg
H.-L. Wu, Dr. L.-Z. Wu
Key Laboratory of Photochemical Conversion and Optoelectronic
Materials, Technical Institute of Physics and Chemistry
The Chinese Academy of Sciences, Beijing, 100190 (China)
Dr. H.-R. Tao
Scheme 1. Direct HAT catalysis for CH functionalization.
College of Chemistry, Beijing Normal University
Beijing, 100875 (China)
These authors contribute equally to this work.
Xanthene dyes have been successfully used as catalysts for
photoredox reactions. These dyes are typically inexpensive,
easy to handle, completely absorb in the visible-light region, and
possess excellent photocatalytic performance.[1a,7] Even though
photochemical hydrogen abstraction properties of quinones
[†]
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201803220
This article is protected by copyright. All rights reserved.

10.1002/anie.201803220
Angewandte Chemie International Edition
COMMUNICATION
have long been known,[1a] xanthene dyes, possessing structural
similarity with quinones, have never been utilized as direct HAT
photocatalysts. We speculate that after absorbing a photon, the
excited xanthene molecule may undergo HAT with a CH bond
to form a relatively stable radical intermediate due to both
captodative[8] and steric effects, which is unlikely to participate in
a side coupling reaction, thus enabling a more effective RHAT
process (Scheme 1c).
regioselectively on the α-oxy carbon of ambroxide, delivering
product 28 in good yield with the alkyl substituent installed at
equatorial position in the major isomers.
To explore the possibility of using xanthene dyes as direct
HAT catalysts, we selected the alkylation of CH bonds via
radical addition to electrophilic alkenes as a model reaction. This
conceptually simplest transformation represents the most
economically appealing process for CC bond formation.
However, current reported examples suffer from at least one
drawback such as limited scope, requiring UV light, high catalyst
loading, or poor reactivity/selectivity.[2]
We initiated the evaluation of various xanthene dyes by using
tetrahydrofuran (THF) as the CH partner and solvent in the
presence of a dicyanide electron-deficient alkene 1 under white
LED light (Table S1). A 50 oC water bath was applied to increase
the reaction efficiency. To our delight, 2 mol% eosin Y could
efficiently catalyze this transformation to deliver quantitative
yield of alkylation product 2 in a three-hour reaction period. The
heavy-atom effect contributing to effective intersystem crossing
for a long-lived triplet excited state was essential, as fluorescein
and rhodamine B gave very low product yields. Notably,
employment of the dianionic form of eosin Y as the catalyst
resulted in a significantly reduced reactivity. Moreover, the
reaction using eosin Y was also effective in acetone or tBuOH
as the solvent with 10 equiv of THF. Finally, no product
formation was detected in the absence of either photocatalyst or
light, demonstrating the need for these components.
We next investigated the scope of CH bonds that can be
activated by catalytic eosin Y under visible-light. The alkylation
reactions were conducted with 5 equiv of CH partners and
electron-deficient olefin 1 (1 equiv) in the presence of 2 mol%
eosin Y in acetone (0.2 M) under white LED irradiation in a 60 oC
water bath. As shown in Scheme 2, ethers including 1,4-
dioxane, tetrahydropyran, and 3,3-dimethyloxetane afforded the
Giese adducts (3 to 5) in good yields. Only the 2-alkylation
product 6, which can be applied as an inexpensive formyl
equivalent, was detected upon switching to 1,3-dioxolane as the
CH partner.[9] The reactions of thioethers or aliphatic amides
proceeded smoothly to afford products 7-10 in very good yields,
with alkylation exclusively at the carbon adjacent to the
heteroatom. Alcohol substrates successfully produced the
coupling products (11 to 14) in excellent yields. These products
would undergo cyclization when treated with silica gel to give
enamino-ketonitriles 15 to 18, which are important synthons for
constructing complex bioactive heterocycles.[10] Aldehydes can
be applied as latent nucleophilic handles to achieve umpolung
reactivity, accessing hydroacylation-type products (19 to 22) that
can be further derivatized to 1,3-dicarbonyl compounds.[11] The
direct alkylation of allylic or benzylic substrates proceeded
smoothly (23 to 26). Notably, a simple unfunctionalized alkane,
cyclohexane, could be direct alkylated using this method,
although with a lower efficiency (27, 32%). Importantly, our
method has shown the potential for site-selective CH activation,
as compounds 6, 10, and 24 were exclusively generated in the
presence of at least one additional hydridic proton in the
substrates. Furthermore, the value of this protocol is
demonstrated by its applicability to the late-stage
functionalization of natural products. The HAT acted
Scheme 2. Scope of CH partners. [a] Isolated yields. [b] D.r. based on the
analysis of the 1H NMR spectra of crude reaction mixtures. [c] CH partners as
the solvent. [d] Yields were determined from the crude 1H NMR using 1,3,5-
trimethoxybenzene as an internal standard. [e] tBuOH as the solvent.
Experiments probing the scope of methylene-malononitriles
revealed that sterically demanding tetra-substituted olefins
smoothly delivered the desired products 29 and 30 in good
yields. Further evaluation of the generality of the alkylation
protocol with regard to the electron-deficient alkenes
demonstrated an extremely broad substrate scope (Scheme 3a).
Substrates including unsaturated aldehydes, ketones, esters,
lactones, amides, imides, sulfones, and nitro compounds were
all good candidates, delivering good yields of alkylation products
(31 to 39). Vinylpyridines can also be utilized as building blocks
for alkylation to introduce the biologically important pyridine
moiety (40).[12] Substituents including electron-rich or electron-
deficient aryl rings (41-43), heterocycles (44, 45), and alkyl
chains (46) were all tolerated. Diene Michael acceptors could be
successfully applied to give 1,4-addition type products (47). Both
the wide scopes of CH partners and the electron-deficient
alkenes were attributed to the HAT process which overcomes
the redox limitation associated with the SET process.[11]
To demonstrate the synthetic applicability of this methodology,
the alkylations were amenable to scale-up to gram quantities
This article is protected by copyright. All rights reserved.

10.1002/anie.201803220
Angewandte Chemie International Edition
COMMUNICATION
using continuous-flow reactors (Scheme 3b). Furthermore, the
CH alkylation could be effectively performed using water as
solvent (Scheme S1), enabling an even greener process.
the active catalyst, which fundamentally differed from the SET
process induced by anionic eosin Y.[14]
T H F solution
B lack LED
B lue LED
G reen LED
W hite LED
neu tral E Y
m o no an ionic EY
dianionic EY
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
350
400
450
500
550
600
W avelen gth (nm )
0.05
0.05
EY in TH F
E Y +olefin in TH F
32 9 nm
3 6 6 n m
329 nm
366 nm
543 nm
54 3 nm
0.00
0.00
0
5
 s
 s
0
s
10 s
20 s
30 s
80 s
20 s
50 s
80 s
-0.05
-0.05
400
500
600
700
800
400
500
600
700
800
W avelength (nm )
W a velength (n m )
Scheme 4. Plausible mechanism with supporting evidence. a) Emission of
light sources and UV-Vis of different forms of eosin Y in THF. b) CH
alkylation in THF under different light sources. c) Transient absorption spectra
of eosin Y in THF and d) eosin Y + phenyl vinyl sulfone 50 in THF (degassed,
excitation at 470 nm). e) Plausible mechanism.
Scheme 3. Scope of electron-deficient olefins and continuous-flow synthesis.
[a] Isolated yields. D.r. based on the analysis of 1H NMR spectra of the
reaction crude mixture. [b] tBuOH as the solvent with 10 equiv of THF.
Both the redox potential analysis and fluorescence quenching
studies (Figure S11) indicated that the CH activation was not
induced by electron- or energy-transfer processes. Laser flash-
photolysis was further employed to elucidate the transient
intermediates during the reaction process. As shown in Scheme
4c, upon laser excitation at 470 nm, the THF solution of neutral
eosin Y alone immediately exhibited a strong bleaching of the
ground state from 350 nm to 520 nm, and two new absorption
peaks at 329 nm and 543 nm decayed with very similar lifetimes
(20.6 and 21.3 μs, Figures S9). We assigned them to the triplet
state *eosin Y. Importantly, a new absorption peak emerged
after 20 μs at 366 nm with an exceptional long lifetime (>4 ms),
which we proposed to be the stable radical intermediate of eosin
YH generated from HAT between *eosin Y and THF. The
lifetime of 366 nm absorption decreased to 1.0 ms after adding
alkene 50 into the mixture, which further supported our
hypothesis that eosin YH would be consumed after adding the
olefin substrate (Scheme 4d and Figure S9). Both light on/off
experiments (Figure S13) and the calculated quantum yield ( =
0.40) suggested a nonchain pathway.[15] To gain further insight
into the reaction mechanism, deuterium labeling study was
Eosin Y can exist in four different structures in solution by
acid-base equilibration: the spirocyclic form, the neutral form, the
monoanionic form, and the dianionic form, exhibiting different
absorption properties with visible-light (Scheme 4a, Figures S3,
S4). It has been illustrated that the anionic forms of eosin Y were
the active catalysts that promote SET-based redox catalytic
cycles in majority of previous studies, where neutral eosin Y was
considered inactive.[13] In order to elucidate the actual species of
eosin Y in the HAT-based catalytic cycle, various control
experiments were performed. CH alkylation in THF was
conducted under various light sources, and blue LED afforded
the highest conversion (Scheme 4b), which coincided with the
maximum absorption of neutral eosin Y (Scheme 4a). The same
reaction in acetone under different light wavelengths (Table S3),
catalyst-free reactions (Table S5), reaction monitoring by UV-Vis
(Figure S5), and acid/base additive studies (Tables S7-S8,
Figures S6-S7), indicated that the neutral eosin Y species was
This article is protected by copyright. All rights reserved.

10.1002/anie.201803220
Angewandte Chemie International Edition
COMMUNICATION
conducted by treating THF and d8-THF with alkene 50 in two
different vessels or in one vessel under the optimized conditions,
resulting in kH/kD 1.6 and 2.4 respectively. The KIE data
indicated that the CH cleavage was reversible and might occur
before the rate-determining step.[16]
a promising platform to tune the strength of HAT ability for site-
selective and stereoselective CH activation.
Acknowledgements
A plausible mechanistic pathway was proposed in light of all
the experimental data (Scheme 4e). The formation of a carbon-
centered radical was promoted by visible-light-activated *eosin Y
through a HAT process. The derived carbon radical was
subsequently trapped by an electron-deficient alkene to
selectively form radical adduct B. The RHAT process between
eosin Y-H II to radical B exhibited a high free energy barrier
(32.3 kcal/mol) based on DFT calculation (path a).[17] Instead,
another THF molecule and radical B might undergo a reversible
HAT process (19.2 kcal/mol) to deliver the desired alkylation
product, followed by RHAT between THF radical A and eosin Y-
H II (19.9 kcal/mol) to regenerate ground state eosin Y catalyst
(path b). However, we were unable to exclude the possibility of
path a at current stage based on the deuterium labeling study
(Scheme S2).[17]
The authors thank Mr. Feng Hong-Yu (NUS) for assistance with
the development of chemistry. We are grateful for the financial
support provided by the National University of Singapore and the
Ministry of Education (MOE) of Singapore (R-143-000-645-112,
R-143-000-665-114), GSK-EDB (R-143-000-687-592), and
A*STAR RIE2020 AME (R-143-000-690-305).
Keywords: eosin Y • photocatalysis • organocatalysis • CH
functionalization • hydrogen atom transfer
[1] For selected reviews, see: a) N. A. Romero, D. A. Nicewicz, Chem. Rev.
2016, 116, 10075; b) D. Ravelli, S. Protti, M. Fagnoni, Chem. Rev. 2016,
116, 9850; c) C. K. Prier, D. A. Rankic, D. W. C. MacMillan, Chem. Rev.
2013, 113, 5322; d) J. Xuan, W.-J. Xiao, Angew. Chem. Int. Ed. 2012,
51, 6828; Angew. Chem. 2012, 124, 6934; e) J. M. R. Narayanam, C. R.
J. Stephenson, Chem. Soc. Rev. 2011, 40, 102.
[2]
a) L. Capaldo, D. Ravelli, Eur. J. Org. Chem. 2017, 2056; b) D. Ravelli,
D. Dondi, M. Fagnoni, A. Albini, Chem. Soc. Rev. 2009, 38, 1999.
a) V. De Waele, O. Poizat, M. Fagnoni, A. Bagno, D. Ravelli, ACS
Catal. 2016, 6, 7174; b) S. Kamijo, G. Takao, K. Kamijo, M. Hirota, K.
Tao, T. Murafuji, Angew. Chem. Int. Ed. 2016, 55, 9695; Angew. Chem.
2016, 128, 9847; c) J. J. Murphy, D. Bastida, S. Paria, M. Fagnoni, P.
Melchiorre, Nature 2016, 532, 218; d) S. Kamijo, G. Takao, K. Kamijo,
T. Tsuno, K. Ishiguro, T. Murafuji, Org. Lett. 2016, 18, 4912; e) J. G.
West, T. A. Bedell, E. J. Sorensen, Angew. Chem. Int. Ed. 2016, 55,
8923; Angew. Chem. 2016, 128, 9069; f) N. Ishida, Y. Masuda, S.
Uemoto, M. Murakami, Chem. Eur. J. 2016, 22, 6524; g) J. G. West, D.
Huang, E. J. Sorensen, Nat. Commun. 2015, 6, 10093; h) S. Kamijo, K.
Tao, G. Takao, H. Tonoda, T. Murafuji, Org. Lett. 2015, 17, 3326; i) M.
Nagatomo, S. Yoshioka, M. Inoue, Chem. Asian J. 2015, 10, 120; j) Y.
Amaoka, M. Nagatomo, M. Watanabe, K. Tao, S. Kamijo, M. Inoue,
Chem. Sci. 2014, 5, 4339; k) C. W. Kee, K. F. Chin, M. W. Wong, C. H.
Tan, Chem. Commun. 2014, 50, 8211; l) J.-B. Xia, C. Zhu, C. Chen, J.
Am. Chem. Soc. 2013, 135, 17494; m) S. Kamijo, T. Hoshikawa, M.
Inoue, Org. Lett. 2011, 13, 5928; n) D. Dondi, S. Protti, A. Albini, S. M.
Carpio, M. Fagnoni, Green. Chem. 2009, 11, 1653.
[3]
Scheme 5. Eosin Y-based HAT for CH functionalizations. [a] With 2.5 equiv
of NaOAc as an additive.
The eosin Y-based direct HAT process provides an extremely
convenient and green pathway for CH activation, who’s
synthetic utility can be extended far beyond alkylation. Our very
preliminary results illustrated that this visible-light-mediated CH
activation protocol can be extended to vinylation, allylation,
arylation, and cyanation (Scheme 5).
[4]
[5]
For selected recent examples, see: a) M. H. Shaw, V. W. Shurtleff, J. A.
Terrett, J. D. Cuthbertson, D. W. C. MacMillan, Science 2016, 352,
1304; b) S. Mukherjee, B. Maji, A. Tlahuext-Aca, F. Glorius, J. Am.
Chem. Soc. 2016, 138, 16200.
For selected recent examples, see: a) G. J. Choi, Q. Zhu, D. C. Miller,
C. J. Gu, R. R. Knowles, Nature 2016, 539, 268; b) J. C. K. Chu, T.
Rovis, Nature 2016, 539, 272.
In summary, we have developed a visible-light-mediated
alkylation of CH bonds with eosin Y as the catalyst. This
transformation accommodates an extremely broad substrate
scope, and is distinguished by its operational simplicity, green
protocol, and amenability to large-scale synthesis via
continuous-flow technology. A variety of synthons can be easily
achieved by this method, which will likely find wide industrial
application. Moreover, to the best of our knowledge, this study
represents the first example of using xanthene dyes as direct
HAT photocatalysts. Its metal-free, readily available, and low-
costing nature, in addition to light absorption in the visible region
and unlikely side reactions, makes eosin Y an ideal direct HAT
photocatalyst, with great promise for CH activation with a
diverse range of functionalities. Unlike the well-established
anionic eosin Y-based photoredox process, neutral eosin Y is
the active catalyst to promote the photo-HAT transformation.
Moreover, the easy structural modification of xanthenes provides
[6]
[7]
[8]
D. Ravelli, S. Protti, M. Fagnoni, Acc. Chem. Res. 2016, 49, 2232.
D. P. Hari, B. König, Chem. Commun. 2014, 50, 6688.
H. G. Viehe, Z. Janousek, R. Merenyi, L. Stella, Acc. Chem. Res.1985,
18, 148.
[9]
M. K. Nielsen, B. J. Shields, J. Liu, M. J. Williams, M. J. Zacuto, A. G.
Doyle, Angew. Chem. Int. Ed. 2017, 56, 7191; Angew. Chem. 2017,
129, 7297.
[10] T. Saito, T. Obitsu, T. Kondo, T. Matsui, Y. Nagao, K. Kusumi, N.
Matsumura, S. Ueno, A. Kishi, S. Katsumata, Y. Kagamiishi, H. Nakai,
M. Toda, Bioorg. Med. Chem. 2011, 19, 5432.
[11] R. Zhou, H. Liu, H. Tao, X. Yu, J. Wu, Chem. Sci. 2017, 8, 4654.
[12] a) K. N. Lee, Z. Lei, M.-Y. Ngai, J. Am. Chem. Soc. 2017, 139, 5003; b)
L. Capaldo, M. Fagnoni, D. Ravelli, Chem. Eur. J. 2017, 23, 6527.
[13] a) M. Majek, A. J. Wangelin, Acc. Chem. Res. 2016, 49, 2316; b) M.
Majek, F. Filace, A. J. Wangelin, Beilstein J. Org. Chem. 2014, 10, 981.
[14] Even though good conversions have been achieved with most
substrates tested within 24 h using white LED, we expected that blue
LED would afford more efficient transformations. See Table S4 for
comparison studies.
This article is protected by copyright. All rights reserved.

10.1002/anie.201803220
Angewandte Chemie International Edition
COMMUNICATION
[15] M. A. Cismesia, T. P. Yoon, Chem. Sci. 2015, 6, 5426.
[16] E. M. Simmons, J. F. Hartwig, Angew. Chem. Int. Ed. 2012, 51, 3066;
Angew. Chem. 2012, 124, 3120.
[17] See Supporting Information for detailed control experiment study, DFT
calculation, and mechanistic analysis.
This article is protected by copyright. All rights reserved.

10.1002/anie.201803220
Angewandte Chemie International Edition
COMMUNICATION
Entry for the Table of Contents (Please choose one layout)
COMMUNICATION
Xuan-Zi Fan,† Jia-Wei Rong,† Hao-Lin
Wu, Quan Zhou, Hong-Ping Deng, Jin
Da Tan, Cheng-Wen Xue, Li-Zhu Wu,
Hai-Rong Tao, and Jie Wu*
Page No. – Page No.
Eosin Y as a Direct Hydrogen Atom
Transfer Photocatalyst for the
Functionalization of CH Bonds
We discovered that neutral eosin Y could be employed as an effective direct hydrogen atom transfer
photocatalyst to activate a wide range of native CH bonds in a green and sustainable fashion.
This article is protected by copyright. All rights reserved.