A cascade cross-coupling hydrogen evolution reaction by visible light catalysis

Article abstract of DOI:10.1021/ja408486v

Cross-dehydrogenative-coupling reaction has long been recognized as a powerful tool to form a C-C bond directly from two different C-H bonds. Most current processes are performed by making use of stoichiometric amounts of oxidizing agents. We describe her

Full text of DOI:10.1021/ja408486v

Communication
pubs.acs.org/JACS
A Cascade Cross-Coupling Hydrogen Evolution Reaction by Visible
Light Catalysis
Qing-Yuan Meng,† Jian-Ji Zhong,† Qiang Liu,†,‡ Xue-Wang Gao,† Hui-Hui Zhang,† Tao Lei,† Zhi-Jun Li,†
Ke Feng,† Bin Chen,† Chen-Ho Tung,† and Li-Zhu Wu*,†
†Key Laboratory of Photochemical Conversion and Optoelectronic Materials, Technical Institute of Physics and Chemistry &
University of Chinese Academy of Sciences, Chinese Academy of Sciences, Beijing 100190, PR China
‡State Key Laboratory of Applied Organic Chemistry, College of Chemistry and Chemical Engineering, Lanzhou University, Lanzhou
730000, PR China
S
* Supporting Information
In this communication, we wish to report a new type of
ABSTRACT: Cross-dehydrogenative-coupling reaction
cross-coupling reaction for C−C bond construction, namely,
has long been recognized as a powerful tool to form a
C−C bond directly from two different C−H bonds. Most
current processes are performed by making use of
stoichiometric amounts of oxidizing agents. We describe
here a new type of reaction, namely cross-coupling
hydrogen evolution (CCHE), with no use of any sacrificial
oxidants, and only hydrogen (H2) is generated as a side
product. By combining eosin Y and a graphene-supported
RuO2 nanocomposite (G-RuO2) as a photosensitizer and a
catalyst, the desired cross-coupling products and H2 are
achieved in quantitative yields under visible light
irradiation at room temperature.
cross-coupling hydrogen evolution (CCHE), which activates
C−H bonds to afford a cross-coupling product and an
equivalent amount of H2 in good to excellent yields with no
use of any sacrificial oxidants (Scheme 1, this work). Herein, an
organic dye eosin Y is employed as a photosensitizer to initiate
cross-coupling of amines with nucleophiles by visible light
catalysis, and at the same time, a graphene-supported RuO2
nanocomposite3i (G-RuO2) is selected as a catalyst to capture
the electron and proton eliminated from the C−H bonds of the
substrates. As will be discussed later, this was found to be the
case. With visible light irradiation for 20 h, the desired cross-
coupling products are achieved smoothly at ambient condition,
accompanying with H2 formation in a quantitative yield (up to
96%). Spectroscopic study and product analysis provide direct
evidence on the photoinduced electron transfer and H2
evolution for this CCHE transformation.
Because of the prevalent skeleton of isoquinoline and indole
in natural products,4 our initial study focused on cross-coupling
reaction of N-phenyl-1,2,3,4-tetrahydroisoquinoline (1a) and
indole (2a) that has been studied by using a photocatalyst and a
sacrificial oxidant6i at room temperature. First, catalytic
amounts of eosin Y (5 mol %) and G-RuO2 (0.3 mol %)
were added into the solution of 1a and 2a, and then the
nitrogen-purged solution was irradiated by a high-pressure
mercury lamp (500 W) with light wavelength longer than 450
nm for 10 h. To our delight, in the absence of a sacrificial
oxidant the desired cross-coupling product was obtained in a
moderate yield with 30% conversion of 1a. More importantly,
H2 was generated in a 25% yield based on the consumption of
1a (Table 1, entry 1). Increasing the concentration of eosin Y
and irradiation time could significantly improve the perform-
ance. The cross-coupling product 3a (89%) and H2 (76%) were
obtained with 90% conversion of 1a after 20 h irradiation
(Table 1, entries 2−3). Further increasing the amount of eosin
Y from 20 to 50 mol %, however, resulted in lower conversion
and yield of the transformation (Table 1, entries 4−6),
probably due to light-filter effect at high concentration of
eosin Y. In addition, the amount of indole was also found to
he design of mild, general, and efficient methods for C−C
Tbond construction is an essential topic of synthetic
chemistry.1 Cross-dehydrogenative coupling (CDC) reaction is
one of the most powerful tools to make a C−C bond directly
from two different C−H bonds under oxidative conditions.2,3
Such a coupling reaction avoids the prefunctionalization and
defunctionalization that have been part of the traditional
synthetic design, and thus reduces the number of steps to the
target molecule. Over the past decade, this straightforward
reaction has spurred tremendous research effort, and there have
been numerous advances in the substrate scope, functional
group tolerance, and a range of different catalysts for improving
the transformation.3 However, hydrogen (H2) is not usually the
byproduct because the thermodynamics of making a C−C
bond with loss of H2 is unfavorable. As a result, an appropriate
sacrificial oxidant (Ox) is always required for the CDC reaction
(Scheme 1, previous work).
Scheme 1
Received: August 16, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja408486v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
a
of 832 for 8 h of irradiation (Figure S2 (SI)), which is greater
than that of the best system achieved by RuO2 for H2 evolution
so far (TON = 120).8a For systematic comparison, RuO2·nH2O
and Al2O3-supported RuO2 (Al2O3−RuO2)8b were further
prepared as the catalyst for CCHE reaction. And the results
indicated that G-RuO2 is much more efficient than RuO2·nH2O
and Al2O3−RuO2 under the same condition (Table S1 (SI),
entries 7−8). Clearly, the electronic conductivity of the
graphene9 may accelerate the electron transfer between eosin
Y and RuO2 on the surface of graphene during the reaction.
Next, we carried out an experiment with a deuterated
substrate to identify the source of H2 in the reaction. When
deuterated substrate 1a-D was selected to react with indole 2a
under the same condition, the conversion and efficiency were
significantly decreased, indicating that the dissociation of
proton from amine 1a-D is a rate-determining step for the
CCHE process. When D2O was used as the solvent, D2 was
generated instead of H2 as the only byproduct with no
alteration of the reaction efficiency, which suggests that the
released protons from the substrates are quickly exchanged with
D2O (Scheme 2). From the above results, we could speculate
that G-RuO2 is able to accept the protons of the substrates.
Table 1. Optimization of the Reaction Conditions
yield
b
c
d
entry eosin Y (equiv %) time (h) conv (%)
3a (%)
H2 (%)
1
2
3
4
5
6
7
8
9
5
10
10
20
20
20
20
20
20
20
30
70
90
68
84
85
90
94
98
50
62
89
77
81
77
92
94
95
25
78
76
58
75
52
83
88
90
20
20
15
25
50
20
20
20
e
f
g
a
Conditions: 0.1 mmol 1a, 0.2 mmol 2a, 0.0003 mmol G-RuO2, and
corresponding amount of eosin Y in 5 mL of H2O under N2,
irradiation of λ > 450 nm at room temperature. Corresponding to 1a.
Based on 1a and determined by NMR using 4-nitroacetophenone as
b
c
d
e
f
an internal standard. Based on 1a. 2.5 equiv of 2a. 3.0 equiv of 2a.
g
4.0 equiv of 2a.
Scheme 2. Deuterium Experiments for CCHE Reactions
influence the CCHE process to some extent (Table 1, entries
7−9), and as the mixed solvents were used to replace water, the
adverse effects occurred (Table S1, Supporting Information
(SI)). It is of significance that under an optimal condition
almost complete conversion of 1a (98%) and excellent yields
for 3a (95%) and H2 (90%), respectively, were achieved in the
CCHE transformation (Table 1, entry 9). The absence of either
eosin Y or G-RuO2, however, led to a negligible conversion of
1a under the same conditions (Table S1 (SI)). Moreover, no
conversion could be detected when the reaction was conducted
in the dark (Table S1 (SI)). These results suggest that light,
eosin Y and G-RuO2 are all essential for the reaction.
It has long been known that thermodynamics of making a
C−C bond with the loss of H2 is typically unfavorable and thus
requires an external driving force, stoichiometric sacrificial
oxidants,3 such as peroxides, quinones and molecular oxygen.
In particular, the use of visible light5,6 has been recently
demonstrated promising in aerobic CDC reactions, where
molecular oxygen is necessary to accept the electron and
proton eliminated from the substrates. In the current study,
however, the whole reaction undergoes under an inert
atmosphere. And the absence of oxygen led to negligible
product formation when the aqueous solution of eosin Y, 1a,
and 2a was irradiated by visible light (λ > 450 nm) (Table S1
(SI)). Nevertheless, a catalytic amount of G-RuO2 (0.3 mol %)
was able to produce 3a and an equivalent amount of H2 from
the same reaction vessel with excellent yields (Table 1, entry 9).
Obviously, G-RuO2 plays a crucial role in operating this unique
CCHE reaction.
To understand the primary process of the reaction, we
examined the possibility of G-RuO2 to function as a catalyst for
H2 evolution. It was noted that H2 could evolve when
nucleophile indole was absent from the reaction system, but
in the absence of G-RuO2 no H2 was detected (Figure S1 and
Table S1 (SI)). To confirm G-RuO2 responsible for H2
evolution, we used tertiary amine triethanolamine (TEOA), a
typical sacrificial electron donor for photosynthesis of H2,7 to
replace 1a. The system, containing eosin Y, G-RuO2, and
TEOA, evolved H2 immediately with turnover number (TON)
The following question is whether G-RuO2 could be an
electron acceptor. To answer this question, a flash-photolysis
study was carried out in a degassed aqueous solution at room
temperature. Thanks to the rich spectroscopic property of eosin
Y,6e,10 we obtained valuable information on the intermediates
during irradiation. Upon laser excitation by 532 nm light, a
strong negative bleach of the ground-state absorption of eosin
Y at ∼530 nm and characteristic absorptions at ∼560 nm with a
lifetime of 115 μs were seen immediately (Figure S3-a (SI)),
the latter of which is consistent with that of the triplet excited
10
3
state of eosin Y, [eosin Y]*. When TEOA was added into
the solution of eosin Y, new absorptions appeared with a
maximum at ∼400 nm in addition to the strong absorptions of
3[eosin Y]* (Figure S3-b (SI)). With reference of spectroscopic
character of eosin Y,10 the new species at ∼400 nm is ascribed
to reduced eosin Y radical anion, [eosin Y]•−. From the kinetics
probed at 400 nm, we inferred the lifetime of [eosin Y]•− being
94.3 μs. In contrast, direct introduction of G-RuO2 into the
3
solution of eosin Y caused the absorptions of [eosin Y]*
unchanged (112 μs, Figure S3-c (SI)) indicative the oxidative
3
quenching [eosin Y]* by G-RuO2 is negligible. Alternatively,
with the addition of G-RuO2 into a mixture of eosin Y and
B
dx.doi.org/10.1021/ja408486v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
a
TEOA in aqueous solution, the system exhibited almost the
same absorptions of [eosin Y]•− (Figure S3-d (SI)), but the
lifetime of absorption at 400 nm decreased to 54.6 μs. This
finding provided direct evidence on the electron transfer from
[eosin Y]•− to G-RuO2.
Table 2. Scope of 1,2,3,4-Tetraisoquinolines
On the basis of the above results, we suggest a general
mechanism of this CCHE reaction (Scheme 3). Upon visible
yield
b
c
d
entry
R1
R2
conv (%)
3 (%)
H2 (%)
Scheme 3. Proposed Mechanism
1
2
3
4
5
6
7
8
H
H
H
H
94
3a, 94(80)
3b, 96(78)
3c, 83(58)
3d, 89(56)
3e, 96(82)
3f, 91(74)
3g, 89(70)
0
88
CH3
OCH3
H
85
96
35
43
OCH3
H
21
80
F
91
96
H
Cl
83
76
H
Br
80
73
H
CN
trace
trace
a
Conditions: 0.1 mmol 1, 0.3 mmol 2a, 0.0003 mmol G-RuO2, and
0.02 mmol eosin Y in 5 mL of water under N2, irradiation of λ > 450
b
c
nm at room temperature. Corresponding to 1. Based on 1 and
determined by NMR using 4-nitroacetophenone as an internal
standard, isolated yields are given in parentheses. Based on 1.
d
light irradiation, eosin Y is pumped to its singlet excited state
1[eosin Y]* that quickly transfers to its triplet state 3[eosin Y]*
with the lifetime of 115 μs. With the addition of tertiary amine
1a into the solution, an electron transfer from 1a to 3[eosin Y]*
state results in the formation of cation radical [1a]•+ and eosin
Y radical anion [eosin Y]•−, as evidenced by transient
absorption spectra (Figure S3 (SI)). The generated [1a]•+
further releases a proton into water and then is oxidized to
afford an iminium ion intermediate. Subsequently, the
nucleophilic addition to the iminium gives rise to the cross-
coupling product 3a. On the other hand, the formed radical
anion [eosin Y]•− is restored to its ground state by G-RuO2 in
water. The shortened lifetime of [eosin Y]•− from 94.3 to 54.6
μs suggests an effective electron transfer from the [eosin Y]•−
to G-RuO2, which can react with the protons delivered by
[1a]•+ to produce H2 in water. Because the deuterium
experiments indicate that the dissociation of proton from 1a
is a rate-determining step and D2 was produced when D2O was
used as a solvent, together with the fact that the reaction
efficiency is significantly decreased when water was replaced by
mixed organic solvents, we believe that water is of significance
in mediating the proton exchange in the cross-coupling process.
With understanding the reaction mechanism, we further
explored the generality of the CCHE reaction on the scope of
N-phenyl-1,2,3,4-tetrahydroisoquinolines (1). As shown in
Table 2, most of the desired crossing-coupling products 3
and H2 are obtained in good to excellent yields. In particular,
the products containing chloro- or bromo- functionalities can
serve as potential intermediates for further organic trans-
formations (Table 2, entries 6−7). However, the strong
electronic effect of the substituent retards the reaction
dramatically (Table 2, entries 3−4), probably due to the
reduction of the electrophilicity of imine cation generated
during reaction. Meanwhile, a stronger electron-withdrawing
group at the 4-position of N-phenyl tetrahydroisoquinoline
makes the photoinduced electron transfer from N-phenyl
tetrahydroisoquinoline to eosin Y thermodynamically unfavor-
able, thereby leading to no reaction at all (Table 2, entry 8).
Good to excellent yields were also obtained by using a variety
of substituted nucleophilic indoles. Note that the CCHE
reaction always occurs at the 3-position of indoles no matter
which position is substituted by different electronic groups
(Table 3). A good result was achieved when 7-methyl indole
a
Table 3. Scope of Indoles
a
Conditions: 0.1 mmol 1a, 0.3 mmol 2, 0.0003 mmol G-RuO2, and
0.02 mmol eosin Y in 5 mL of water under N2, irradiation of λ > 450
b
nm at room temperature. Based on 1a and determined by NMR
using 4-nitroacetophenone as an internal standard, isolated yields are
c
given in parentheses. Corresponding to 1a.
was used (Table 3, 3q). Because of the lower electron density
for the nucleophilie addition, a strong electron-withdrawing
substituent at indole displays a relatively lower reactivity (Table
3, 3m and 3p). More nucleophilic substrates, such as malonate
C
dx.doi.org/10.1021/ja408486v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
(4) (a) Scott, J. D.; Williams, R. M. Chem. Rev. 2002, 102, 1669.
(b) Somei, M.; Yamada, F. Nat. Prod. Rep. 2005, 22, 73.
(c) Siengalewicz, P.; Rinner, U.; Mulzer, J. Chem. Soc. Rev. 2008, 37,
2676.
(5) (a) Yoon, T. P.; Ischay, M. A.; Du, J. Nat. Chem. 2010, 2, 527.
(b) Narayanam, J. M. R.; Stephenson, C. R. J. Chem. Soc. Rev. 2011,
40, 102. (c) Xuan, J.; Xiao, W.-J. Angew. Chem., Int. Ed. 2012, 51, 6828.
(d) Prier, C. K.; Rankic, D. A.; MacMillan, D. W. C. Chem. Rev. 2013,
113, 5322.
esters and phosphite ester, were also used to replace indoles for
the CCHE reaction. Satisfactorily, the reaction proceeds well
resulting in the formation of the crossing-coupling products
and H2 in moderate to excellent yields (Scheme S1 (SI)).
In summary, we have succeeded in developing a new type of
reaction, cross-coupling hydrogen evolution CCHE, by visible
light catalysis. The cascade reaction is accomplished under an
inert atmosphere by activation of C−H bonds via dehydrogen-
ation that does not require any sacrificial oxidants, and an
equivalent amount of H2 is generated as an only side product.
Combining eosin Y and G-RuO2 as the photosensitizer and the
catalyst, the CCHE reaction undergoes smoothly to afford the
cross-coupling products and H2 in good to excellent yields.
Spectroscopic study and product analysis demonstrate the
photoinduced electron transfer from tertiary amine to eosin Y
to generate [eosin Y]•−, which further delivers an electron to
G-RuO2 for reduction of protons to H2 and regeneration of
eosin Y. This work constitutes the first example of photo-
catalytic dehydrogenative cross-coupling reaction to form a C−
C bond by two different C−H bonds with concomitant
emission of H2. The operationally simple and general mode of
activation is suitable for a broad range of reactants. We hope
this CCHE reaction will become a useful method to construct
C−C bonds for cleaner, safer, and more atom-economic
organic transformation.
(6) (a) Condie, A. G.; Gonzal
Am. Chem. Soc. 2010, 132, 1464. (b) Hari, D. P.; Konig, B. Org. Lett.
2011, 13, 3852. (c) Xie, Z.; Wang, C.; deKrafft, K. E.; Lin, W. J. Am.
́
ez-Gom
́
ez, J. C.; Stephenson, C. R. J. J.
̈
Chem. Soc. 2011, 133, 2056. (d) Mohlmann, L.; Baar, M.; Rieß, J.;
̈
Antonietti, M.; Wang, X.; Blechert, S. Adv. Synth. Catal. 2012, 354,
1909. (e) Liu, Q.; Li, Y.-N.; Zhang, H.-H.; Chen, B.; Tung, C.-H.; Wu,
L.-Z. Chem.Eur. J. 2012, 18, 620. (f) To, W.-P.; Tong, G. S.-M.; Lu,
W.; Ma, C.; Liu, J.; Chow, A. L.-F.; Che, C.-M. Angew. Chem., Int. Ed.
2012, 51, 2654. (g) Mitkina, T.; Stanglmair, C.; Setzer, W.; Gruber,
M.; Kisch, H.; Konig, B. Org. Biomol. Chem. 2012, 10, 3556. (h) Zhu,
̈
S.; Rueping, M. Chem. Commun. 2012, 48, 11960. (i) Zhong, J.-J.;
Meng, Q.-Y.; Wang, G.-X.; Liu, Q.; Chen, B.; Feng, K.; Tung, C.-H.;
Wu, L.-Z. Chem.Eur. J. 2013, 19, 6443.
(7) Esswein, A. J.; Nocera, D. G. Chem. Rev. 2007, 107, 4022.
(8) (a) Maeda, K.; Saito, N.; Lu, D.; Inoue, Y.; Domen, K. J. Phys.
Chem. C 2007, 111, 4749. (b) Zeng, H. C.; Pang, X. Y. Appl. Catal., B
1997, 13, 113.
(9) Geim, A. K.; Novoselov, K. S. Nat. Mater. 2007, 6, 183.
(10) Joselevich, E.; Willner, I. J. Phys. Chem. 1995, 99, 6903.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures, methods, and product character-
ization. This material is available free of charge via the Internet
at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
lzwu@mail.ipc.ac.cn
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We are grateful for financial support from the Ministry of
Science and Technology of China (2013CB834804,
2013CB834505, and 2014CB239402), the National Natural
Science Foundation of China (21090343 and 91027041), and
the Chinese Academy of Sciences.
REFERENCES
■
(1) (a) Corey, E. J.; Cheng, X. M. The Logic of Chemical Synthesis;
John Wiley & Sons: New York, 1989; pp 1−91. (b) Newhouse, T.;
Baran, P. S.; Hoffmann, R. W. Chem. Soc. Rev. 2009, 38, 3010.
(2) (a) Li, C.-J. Acc. Chem. Res. 2008, 42, 335. (b) Yeung, C. S.;
Dong, V. M. Chem. Rev. 2011, 111, 1215. (c) Sun, C.-L.; Li, B.-J.; Shi,
Z.-J. Chem. Rev. 2011, 111, 1293. (d) Liu, C.; Zhang, H.; Shi, W.; Lei,
A. Chem. Rev. 2011, 111, 1780.
(3) (a) Murahashi, S.-I.; Komiya, N.; Terai, H.; Nakae, T. J. Am.
Chem. Soc. 2003, 125, 15312. (b) Li, Z.; Li, C.-J. J. Am. Chem. Soc.
2004, 126, 11810. (c) Stuart, D. R.; Fagnou, K. Science 2007, 316,
1172. (d) Hull, K. L.; Sanford, M. S. J. Am. Chem. Soc. 2007, 129,
11904. (e) Jia, Y.-X.; Kundig, E. P. Angew. Chem., Int. Ed. 2009, 48,
̈
1636. (f) Shu, X.-Z.; Yang, Y.-F.; Xia, X.-F.; Ji, K.-G.; Liu, X.-Y.; Liang,
Y.-M. Org. Biomol. Chem. 2010, 8, 4077. (g) Leow, D.; Li, G.; Mei, T.-
S.; Yu, J.-Q. Nature 2012, 486, 518. (h) Boess, E.; Schmitz, C.;
Klussmann, M. J. Am. Chem. Soc. 2012, 134, 5317. (i) Meng, Q.-Y.;
Liu, Q.; Zhong, J.-J.; Zhang, H.-H.; Li, Z.-J.; Chen, B.; Tung, C.-H.;
Wu, L.-Z. Org. Lett. 2012, 14, 5992.
D
dx.doi.org/10.1021/ja408486v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX