Tricyclometalated iridium complexes as highly stable photosensitizers for light-induced hydrogen evolution

Article abstract of DOI:10.1002/chem.201300146

The development of an efficient and stable artificial photosensitizer for visible-light-driven hydrogen production is highly desirable. Herein, a new series of charge-neutral, heteroleptic tricyclometalated iridium(III) complexes, [Ir(thpy)₂(bt

Full text of DOI:10.1002/chem.201300146

DOI: 10.1002/chem.201300146
Tricyclometalated Iridium Complexes as Highly Stable Photosensitizers for
Light-Induced Hydrogen Evolution
[
a]
[a]
[a]
[a]
Yong-Jun Yuan, Zhen-Tao Yu,* Hong-Lin Gao, Zhi-Gang Zou,*
[
b]
[b]
Chao Zheng, and Wei Huang
Abstract: The development of an effi-
cient and stable artificial photosensitiz-
er for visible-light-driven hydrogen
production is highly desirable. Herein,
a new series of charge-neutral, hetero-
longer wavelengths for these iridium
complexes. We describe new molecular
platforms that are based on these neu-
tral iridium complexes for the produc-
tion of hydrogen through visible-light-
induced photocatalysis over an extend-
ed period of time in the presence of
addition of cobalt-based catalysts when
hydrogen evolution ceased. Our results
demonstrated that the dissociation of
2+
the [Co ACHTGNUTERNN(UNG bpy) ] catalyst contributed to
3
the loss of catalytic activity and limited
the long-term catalytic performance of
the systems. The properties of the neu-
tral complexes are compared in detail
to those of two known non-neutral
leptic tricyclometalated iridium
complexes, [Ir(thpy) (bt)] (1–4; thpy=
,2’-thienylpyridine, bt=2-phenylben-
ACHTUNGTNERNUNG( III)
ACHTUNGTRENNUNG
2
2+
2
[Co
A
H
U
G
R
N
N
(bpy) ]
and
triethanolamine
3
zothiazole and its derivatives), were
systematically synthesized and their
structural, photophysical, and electro-
chemical properties were established.
Three solid-state structures were stud-
ied by X-ray crystallographic analysis.
This design offers the unique opportu-
nity to drive the metal-to-ligand
charge-transfer (MLCT) band to
(TEOA). The maximum amount of hy-
drogen was obtained under constant ir-
radiation over 72 h and the system
could regenerate its activity upon the
bpy-type complexes, [Ir ACHUTNGERNNUG( thpy) ACHTUNGTRENNUNG( dtb-
2
+
+
bpy)] (5) and [Ir ACHUTNGRENNUG( ppy)2 ACHTUNGTRENNGNU( dtb-bpy)] (6;
ppy=2-phenylpyridine, dtb-bpy=4,4’-
di-tert-butyl-2,2’-dipyridyl). This work
is expected to contribute toward the
development of long-lasting solar hy-
drogen-production systems.
Keywords: iridium · hydrogen pro-
duction · photochemistry · sensitiz-
ers · water splitting
[
3–7]
Introduction
synthesis.
Typical solar-energy-conversion schemes
employ photoactive and redox-active chromophores that
II
II
III
The direct conversion of solar energy into usable and stor-
able chemical energy holds great promise for resolving the
surging demand for energy and for addressing the environ-
contain transition metals, such as Ru , Pt , and Ir , thus
closely mimicking the natural photosynthetic light-harvest-
ing machinery in capturing and utilizing solar energy. The
photonic energy that is absorbed by these chromophores
provides the reductive driving force for splitting water into
[1,2]
mental concerns of today’s society.
Among the currently
available options for the generation of sustainable energy,
solar-light-initiated electron-transfer reactions that break
water into “green”, energy-rich hydrogen are of particular
interest because they offer insight into the fundamental
energy-collection and -transfer processes in artificial photo-
[8–16]
molecular hydrogen.
The effectiveness of the chromo-
phores’ action is strongly associated with their beneficial
phtotophysical and electrochemical properties, which can be
fine-tuned through synthetic control over the ligand struc-
ture.
A number of materials were evaluated and charged iridi-
+
um complexes of the type [Ir
A
H
U
G
R
N
U
G
A
H
U
G
E
N
N
(NN)] , which contained
2
[
a] Y.-J. Yuan, Dr. Z.-T. Yu, H.-L. Gao, Prof. Dr. Z.-G. Zou
National Laboratory of Solid State Microstructures and
Eco-Materials and Renewable Energy Research Center
Department of Materials Science and Engineering
Nanjing University and Kunshan Innovation Institute
of Nanjing University
No. 22, Hankou Road, Nanjing, Jiangsu 210093 (P.R. China)
Fax: (+86)-25-8368-6632
E-mail: yuzt@nju.edu.cn
two CN cyclometalating ligands, such as 2-phenylpyridine,
and a neutral NN ligand, such as 2,2’-bipyridine (bpy), were
shown to have better catalytic efficiency and activity in the
photochemical water-reducing system than ruthenium com-
[17–27]
plexes.
Owing to the NN ligand, such species have large
molar extinction coefficients, which can contribute to the
light-harvesting efficiency of the photocatalytic conversion
of water into hydrogen. From a synthetic point of view, the
functional modification of the NN ligand still remains a
zgzou@nju.edu.cn
[
b] Dr. C. Zheng, Prof. Dr. W. Huang
Displays and Institute of Advanced Materials
Nanjing University of Posts and Telecommunications
No. 9, Wenyuan Road, Nanjing, Jiangsu 210046 (P.R. China)
[28]
challenge.
Unfortunately, the practical exploitation of
such a photocatalytic process still suffers from the instability
of the molecular photosensitizer (PS), thereby requiring fur-
ther optimization through innovative chemical design. The
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201300146.
6340
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 6340 – 6349

FULL PAPER
facile dissociation of the labile NN moiety from the metal
center on the reducing PS often occurs during the conver-
[18]
sion. The incorporation of vinyl groups or pendant pyridyl
moieties onto the NN ligand of the iridium complexes sta-
bilizes metal-catalyst colloids and facilitates electron trans-
[20,21]
fer, thus resulting in improved hydrogen production.
Despite these encouraging efforts, all of these robust molec-
ular platforms do not provide a long-term, effective solution
for the design of stable molecular photocatalytic systems.
In this regard, efficient and stable PSs that are best-suited
for achieving the sustained photoelectrochemical conversion
of solar energy are appealing. Recently, we reported that
neutrally charged amidinate-containing iridium complexes
afforded marked activities and stabilities for photoinduced
2
+
hydrogen production by using [Co ACTHUNGTERNN(UNG bpy) ] as a water-reduc-
3
tion catalyst (WRC) in aqueous/organic media. Neverthe-
less, the amidinate ligands that are incorporated into the
Scheme 1. Structures of iridium complexes 1–6 that were investigated in
1
2
3
1
2
3
1
3
3
this study (1: R =R =R =H; 2: R =CF , R =R =H; 3: R =R =H,
irid
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
ium complexes are believed to provide relatively weak
2
1
2
3
3 2 3
R =CF ; 4: R =NMe , R =H, R =CF ).
ligandÀmetal bonding interactions within these complexes.
As a consequence, theses photocatalytic systems are unsta-
ble but still undergo similar photodeactivation after 72 h of
were similar in structure to other cationic iridium complexes
were synthesized and characterized. With compounds 5 and
6, the photocatalytic system reveals limited stability, proba-
bly owing to the formation of the catalytically inactive spe-
cies that originate from the degradation of the PS during
catalysis. The electronic structures of the complexes and the
nature of their excited states were found to influence their
photocatalytic hydrogen-production behavior of these iridi-
um complexes.
[16]
continuous irradiation. Central to our strategy to suppress
the liberation of the NN ligand from the excited PSs is the
replacement of a neutral IrÀN bond with an anionic IrÀC
bond, which may improve the stability of the PSs through
strong covalent IrÀC bonding interactions. More important-
ly, this approach offers the unique opportunity to drive the
metal-to-ligand charge-transfer (MLCT) band to longer
[29,30]
wavelengths.
As a result, their excited states can special-
ly promote visible-light-driven photoredox processes.
Although tricyclometalated iridium complexes are intrigu-
ing as luminophores, because of their excellent photophysi-
cal properties, they have rarely been investigated in the con-
text of solar-light-to-energy conversion. Continuing our in-
vestigations in the field of photoactive metal complexes for
solar-energy conversion, herein, we report that a new series
of neutral tricyclometalated iridium complexes, [Ir-
Results and Discussion
Synthesis and characterization: Heteroleptic tricyclometalat-
ed iridium complexes 1–4 were synthesized at low tempera-
[32]
tures in a three-step reaction process. Detailed synthetic
procedures and characterization data are provided in the
Experimental Section. The tailored functionalization of the
bt ligand leads to greater synthetic control over the structur-
al and electronic properties of the corresponding iridium
complexes. For comparison, cationic iridium complexes 5
and 6 were prepared by employing a two-step bridge-split-
ting synthesis route, followed by anion metathesis with
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(thpy) (bt)] (thpy=2,2’-thienylpyridine, bt=2-phenylbenzo-
2
thiazole and its derivatives; Scheme 1), can undergo a visi-
ble-light-induced process, thus leading to hydrogen produc-
2+
tion in a homogeneous system by using [Co ACHTUNGTRENN(UNG bpy) ] as a
3
proton-reduction catalyst and TEOA as a sacrificial electron
donor. These systems provide the first stable catalytic plat-
form for solar hydrogen production, in which tricyclometal-
[33]
NH PF , which have been reported in the literature. All
4
6
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
ated iridium complexes act as photoinitiating systems. In
of the iridium AHCUTNGTERNNUN(G III) complexes were fully characterized by
+
contrast to cationic iridium complexes, [Ir
decrease in overall charge with an increasing number of
A
H
U
G
R
N
N
(CN)
A
T
N
T
E
N
N
(NN)] , the
NMR spectroscopy and by elemental analysis and the solid-
state structures of complexes 1, 2, and 4 were also confirmed
by single-crystal X-ray diffraction. Single crystals suitable
for X-ray diffraction were obtained by the vapor diffusion
of n-hexane into concentrated solutions of the complexes in
CH Cl over a period of several days at room temperature.
2
metalÀcarbon s bonds enhances the stability of the neutral
PSs. We found that the excited state of the iridium com-
plexes (PS*) could only be oxidatively quenched by [Co-
2+
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(bpy) ] through electron transfer, which demonstrated the
3
2
2
stability of the PS and, thus, the extended lifetime of this
The crystal-structure data and structure-refinement parame-
ters for these complexes are provided Table 1; the images
were generated by using the program Olex2. As shown in
Figure 1, a representative crystal structure of compound 1
reveals an expected octahedral geometry around the iridium
center, which consists of two thpy ligands with cis-C,C and
[16,31]
photocatalytic system.
on photocatalytic activity and stability in this system, two
additional reference iridium complexes, [Ir(thpy)2 A( dtb-
(dtb-bpy)] (6; dtb-bpy=4,4’-di-tert-
To elucidate the ligation effects
[34]
A
H
U
G
R
N
U
G
CHTUNGTRENNUNG
+
+
bpy)] (5) and [Ir
ACHTUNGTRENNUNG( ppy) ACHTUGNTRENNNUG
2
butyl-2,2’-dipyridyl, ppy=2-phenylpyridine; Scheme 1), that
Chem. Eur. J. 2013, 19, 6340 – 6349
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6341

Z.-T. Yu, Z.-G. Zou et al.
Table 1. Crystal data and structure refinements for complexes 1, 2, and 4.
trans-N,N dispositions and one bt ligand in a facial arrange-
ment.
1
2
4
The experimental UV/Vis absorption spectra of com-
plexes 1–4 in CH Cl show intense absorption bands (e>ꢁ
molecular
formula
C
62
H
40Ir
2
N
6
S
6
·H
2
O
C
32
H
19
F
3
IrN
3
S
3
C
68 48 6 2 8 6 2
H F Ir N S ·H O
2
2
4
À1
À1
M
w
1463.78
291(2)
0.71073
790.88
291(2)
0.71073
tetragonal
I4 /a
34.04000(10)
34.04000(10)
10.0020(6)
90.00
1685.92
291(2)
0.71073
monoclinic
P2
9.3570(15)
24.6901(17)
29.1673(12)
90.00
1
0 m cm ) in the ultraviolet region, centered at 298 nm
T [K]
l [ꢂ]
(
Figure 2), which can be assigned to spin-allowed ligand-
1
centered (LC) pÀp* transitions. The next, weaker, broad
band, which reaches approximately 480 nm into the visible
region, can be ascribed to a spin-allowed metal-to-ligand
crystal system monoclinic
space group
a [ꢂ]
b [ꢂ]
c [ꢂ]
a [8]
P2
8.1460(14)
18.5380(11)
8.9170(15)
90.00
1
1
1
b [8]
g [8]
V [ꢂ ]
102.361(2)
90.00
1315.3(3)
1.848
90.00
90.00
11589.5(7)
1.813
94.503(3)
90.00
6717.6(12)
1.667
3
À3
1
calcd [gcm
]
absorption
5.343
4.872
4.210
coefficient
À1
[
mm
]
F
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(000)
scan range q
8]
714
2.19–28.00
6144
2.39–25.02
3304
2.25–23.49
[
index range
À10ꢁhꢁ9
À41ꢁhꢁ41
À41ꢁkꢁ20
À11ꢁlꢁ12
33537
À11ꢁhꢁ11
À28ꢁkꢁ30
À35ꢁlꢁ35
54377
À20ꢁkꢁ22
0
ꢁlꢁ10
total reflns
10676
unique reflns 5026
int =0.0371)
4713
(Rint =0.0907)
25791 (Rint =0.0464)
(
R
data/
5026/1/362
5690/0/367
25791/1/1693
restraints/
parameters
GOF on F
Figure 2. Absorption spectra of complexes 1–4, together with those of
reference complexes 5 and 6 for comparison; all spectra were recorded
in CH Cl at room temperature.
2
1.013
1.062
1.088
2
2
R1, wR
I>2s(I)]
R index (all
2
0.0430 and 0.1051 0.0420 and
0.0473 and 0.1014
[
0.1072
0.0543 and 0.1140 0.0516 and
0.1130
1
charge-transfer ( MLCT) transition mixed with some ligand-
0.0559 and 0.1055
1
data)
to-ligand charge-transfer ( LLCT) character, even including
the lower-energy absorption tail that appears above 480 nm,
which probably corresponds to MLCT transitions. The in-
CCDC-893767 (1), CCDC-893768 (2), and CCDC-893769 (4) contain the
supplementary crystallographic data for this paper. These data can be ob-
tained free of charge from The Cambridge Crystallographic Data Centre
via www.ccdc.cam.ac.uk/data_request/cif.
3
troduction of a CF group onto the bt ligand in complexes 2
3
1
and 3 causes a slight spectroscopic blue-shift of the MLCT
transitions relative to compound 1. Figure 2 also shows the
electronic absorption spectra of compounds 5 and 6 for
comparison; the molar absorptivities of compound 6, which
are within the range 350–500 nm, are relatively lower than
those of other complexes. A significantly blue-shift of the
long-wavelength side of the MLCT absorption band was ob-
served compared with that of complexes 1–4, thus indicating
an increased HOMO–LUMO energy gap in compounds 5
and 6, as expected from the smaller number of metalÀ
carbon s bonds with respect to those in neutral iridium com-
plex. This feature has been well-documented in comparative
studies on cyclometalated ruthenium and iridium complexes.
Upon excitation at lꢀ350 nm, all of the complexes show
strong luminescence, both in their solid state and in organic
solution at room temperature (Figure 3 and Table 2). The lu-
minescence spectrum of compound
1 in air-saturated
CH Cl solution has an emission maximum at 583 nm, with a
2
2
shoulder at approximately 560 nm (t=44 ns), whereas the
emission band of CF -substituted complexes 2 and 3 are
3
structure-less and red-shifted to 611 nm (t=58 ns) and
5
95 nm (t=52 ns), respectively, owing to the prevalent
3
Figure 1. Molecular structure of complex 1.
MLCT character of the excited state. Moreover, compound
6342
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 6340 – 6349

Iridium Photosensitizers for Light-Induced Hydrogen Evolution
FULL PAPER
this increase in emission quantum efficiency (F). The emis-
sion quantum efficiencies of complexes 1–6 range from 0.12
to 0.64 at room temperature in de-aerated CH Cl , relative
2
2
to a fac-[Ir ACHTNUGTERNN(UNG ppy) ] standard (F=0.40), with compound 6
3
being the highest (Table 2), which suggests a different elec-
tronic nature of the luminescent triplet state. As for highly
effective PSs, the values are likely to be of importance for
the overall efficiency of the light-harvesting process (see
below). All of the neutral complexes in this study give very
similar emission efficiencies, except for compound 1, and
are much less emissive than compound 6, which may be re-
lated to the smaller HOMO–LUMO energy gaps in com-
pounds 1–4.
Density functional theory (DFT) calculations reveal that
the electron densities in the HOMOs of compounds 1–4 are
mainly located around the iridium atom, mixed with the
p orbital of the thpy ligands, with a small contribution from
the bt ligand, whereas the electron densities of the LUMOs
primarily reside on the p* orbital that is delocalized over
the whole bt skeleton, with minor contributions from the
2 2
Figure 3. Normalized emission spectra of complexes 1–4 in CH Cl at
room temperature; for clarity, the spectra of reference complexes 5 and 6
are not shown.
[37]
4
has an emission maximum at 555 nm, with a shoulder at
irid
A
H
U
G
E
N
N
ium center and the thpy fragments (Table 3).
The
approximately 590 nm (t=46 ns). The spectroscopic shifts
are consistent with the electrochemical data of these com-
plexes (Table 2). These short luminescence lifetimes in air-
saturated CH Cl solution at ambient temperature are not
energy levels of both the HOMO p orbitals and the LUMO
p* orbitals of these complexes are decreased by the attach-
ment of the electron-withdrawing CF substituent onto the
3
bt ligand. However, because the p* orbital is more localized
on the bt unit, the decrease in the energy level of the p* or-
bital will be larger, consequently leading to a smaller energy
gap and an accompanying red-shift of the emission spec-
trum. In contrast to what is computed for compound 1, for
which both of the cyclometalating thpy ligands contribute to
the HOMO, only one thpy ligand contributes to this orbital
in complex 5. The LUMO density in compound 5 is mainly
located on the ancillary dtb-bpy ligand, with some extension
onto the other thpy ligand. Moreover, both the HOMO and
LUMO orbitals have some p character, with some overlap
2
2
surprising, partially owing to the photochemical quenching
[
35]
of the excited state. However, these values are still suffi-
ciently long for these species to enable photochemical elec-
tron-transfer events to occur prior to the decay of the excit-
[36]
ed state. The observed PL spectra shows that oxygen has
a strong quenching effect on the emission intensity of com-
pound 1 (see the Supporting Information, Figure S1). Under
oxygen-free conditions, the observed lifetimes of the neutral
iridium complexes were considerably longer (0.79–2.89 ms)
than the values under ambient (air-equilibrated) conditions,
which confirmed that the emission arose from a triplet state.
The photophysical data of compounds 5 and 6 are consistent
[38]
on the iridium center. As with compound 6, the HOMO
and LUMO distributions are separated by the metal center.
The HOMO contains substantial contribution from the iridi-
um center, as well as from the phenyl p orbitals of the cyclo-
metalating ppy ligands, whereas the LUMO is mainly dis-
tributed on the ancillary dtb-bpy ligand. The spatial separa-
tion of the HOMO and LUMO in compound 6 allows for
the generation of long-lived, charge-separated excited states
[33]
with previous reported values for these compounds. The
emission maximum (570 nm) of compound 6 in CH Cl is
2
2
blue-shifted compared with the previously reported value
581 nm in MeCN), as expected for an emission with signifi-
(
cant MLCT character. The shorter emission lifetime of com-
pound 6 compared to those of compounds 1–4 is contrary to
Table 2. Photophysical and electrochemical properties of iridium complexes 1–6.
À
[d]
+
[d]
F[e,f]
Compound
l
abs
l
em
E
[V]
ox
E
[V]
red
[b]
E
[eV]
0À0
E
A
H
U
G
R
N
G
E
A
H
U
G
R
N
U
G
t
t
[
a]
[a]
[b]
[c]
[a]
[e]
[nm]
[nm]
[ns]
[ms]
1
2
3
4
5
6
300, 389
300, 388
300, 385
299, 414
278, 416
263, 380
560(sh), 583
611
595
555, 590(sh)
546(sh), 586
570
1.03
1.12
1.13
0.94
1.48
1.57
À1.66
À1.45
À1.51
À1.76
À1.08
À1.11
2.19
2.09
2.15
2.28
2.44
2.54
0.53
0.64
0.62
0.52
1.36
1.43
À1.16
À0.97
À1.02
À1.34
À0.96
À0.97
44
58
52
46
–
2.25
0.79
1.64
2.89
1.35
0.72
0.24
0.14
0.15
0.18
0.12
0.64
–
[
a] Recorded in air-equilibrated CH
2
Cl
2
at room temperature. [b] Redox potentials of complexes 1–6, measured in deoxygenated MeCN with 0.1m
À1
nBu NPF
4
6
as the supporting electrolyte at a scan rate of 100 mVs relative to Ag/AgCl as a reference electrode; the value was adjusted with respect to
the normal hydrogen electrode (0.197 V vs. NHE); only the first reduction is shown. [c] E0À0 values were estimated from the high-side energy of the
À
+
emission band at 10–20% of its maximum intensity. [d] E
temperature. [f] Measured with [Ir ACHTUNGETRNN(UNG ppy)
3
A
H
U
G
E
N
N
A
H
U
G
R
N
U
G
2
Cl
2
at room
Chem. Eur. J. 2013, 19, 6340 – 6349
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6343

Z.-T. Yu, Z.-G. Zou et al.
Table 3. Plot of the HOMO and LUMO distributions for complexes 1–6.
Complex
1
2
3
4
5
6
HOMO
LUMO
upon photoexcitation. Such a well-separated state is desir-
ACHTUNGTRENNUNGa ble for the development of efficient artificial photosynthet-
signed to occur on the ancillary NN ligand, as corroborated
by the DFT results.
ic systems. Based on the results of the theoretical calcula-
tions, the emission in these mixed-ligand systems can be at-
tributed to strong mixing between the MLCT and LC excit-
ed states.
Hydrogen-production experiments: With the abovemen-
tioned analysis in hand, we subsequently employed the ex-
cited state of the iridium complexes in photochemical proc-
esses. A series of photocatalysis experiments were per-
formed in an acetone/water solution that contained [Co-
Electrochemical properties: The electrochemical behavior
of complexes 1–4 was investigated in deoxygenated MeCN
by cyclic voltammetry. Their CV properties were compared
to reference complexes 5 and 6 and were summarized in
Table 2. All of the neutral complexes showed oxidation
waves with potentials in the region 0.94–1.13 V versus a
normal hydrogen electrode (NHE), which were mostly de-
termined by the iridium center and the thpy ligands. The
higher oxidation potentials for compounds 2 and 3 com-
pared to that of parent complex 1 are primarily owing to the
2+
AHCTUNGTNERNU(G bpy) ] as a WRC species and TEOA as a sacrificial elec-
3
tron donor to study the activities of these new complexes in
the visible-light-driven photogeneration of hydrogen. No H2
was detected in the control experiments when PS, TEOA, or
2+
[Co ACHTUNGTERNNN(UG bpy) ] was excluded from the system. Moreover, when
3
the reaction was performed in the dark, no formation of hy-
drogen was observed.
As shown in Figure 4, in the presence of the neutral com-
plex the yield of hydrogen increased almost linearly with the
electron-withdrawing ability of the CF substituent on the bt
3
ligand, thus echoing how the bt ligand appears to be in-
volved in the HOMO levels, as supported by the DFT calcu-
lations. Moreover, these complexes exhibited one or more
reduction waves between À1.45 and À1.76 V versus NHE,
which are primarily attributed to the bt ligand. Complexes 5
and 6 showed more-positive oxidation waves (at 1.48 and
1.57 V versus NHE, respectively), which were expected to
occur on the iridium metal site, together with additional
contributions from the cyclometalating ligands, than those
[38]
of the neutral iridium complexes. The oxidation potential
of compound 5 was anodically shifted by 0.45 V relative to
that of compound 1 as a consequence of the influence of the
diminished electron density at the metal center and the dis-
parity of the overall charge on the complexes. On the other
hand, the first reduction waves for benchmark complexes 5
and 6 were found at about À1.08 and À1.11 V versus NHE,
respectively, which were significantly anodically shifted com-
pared to those of compounds 1–4. This reduction was as-
Figure 4. Photoinduced hydrogen production by using complexes 1–6.
2+
Conditions: PS (5 mm), [Co
A
H
U
G
R
N
N
(bpy)
3
]
(0.33 mm), and TEOA (0.38m) in
acetone/water (4:1 v/v, 100 mL, pH 8.0) upon irradiation (Xe lamp; l>
420 nm).
6344
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 6340 – 6349

Iridium Photosensitizers for Light-Induced Hydrogen Evolution
FULL PAPER
reaction time from the start of the irradiation (Xe lamp, l>
different from that of the performance in terms of hydrogen
formation for these complexes as PSs, which is dependent
on both the light-harvesting process and the quenching rate
constants. In addition, the addition of excess water or ace-
tone to solution of compounds 1–4 did not cause fluores-
cence quenching, thus indicating that there are no simulta-
neous interactions in these cases. However, as observed for
420 nm) up to about 60 h and gradually slowed down there-
after. After the cessation of the reaction (after 72 h), the
total amount of H produced was 87 mmol for compound 1,
2
6
7
4 mmol for compound 2, 74 mmol for compound 3, and
8 mmol for compound 4. These results indicated that the
hydrogen-generation capability of these neutral complexes
was only slightly affected by the nature of the substituents
on the PSs and followed the order: 1>4>3>2. These re-
sults were in agreement with the excited-state lifetimes for
these complexes. In contrast with the results observed for
compounds 1–4, the systems with charged complexes 5 and
other cationic complexes with an NN ancillary ligand, such
+
as [Ir
A
H
U
G
R
N
U
G
2A
H
U
G
R
N
U
G
can be readily quenched not only by TEOA but also by [Co-
2+
AHCTUNGTNERNUG( bpy) ] as a result of their suitable excited-state redox po-
3
tential, thus indicating that electron-transfer quenching of
the PS excited state can occur both oxidatively and reduc-
tively. In nitrogen-saturated CH Cl solutions, the quenching
6
yielded 38 mmol and 110 mmol of H , respectively. After
2
about 50 h of irradiation, no hydrogen production was ob-
served, which was partly a result of the shorter lifetime of
the photoexcited state of both compounds 5 and 6. For com-
2
2
behavior of all of these complexes by either TEOA or [Co-
2+
AHCTUNGTNERNUG( bpy) ] was the same as that in air-equilibrated solutions.
3
2+
2+
parison, the photocatalytic performance of [Ru
A
H
U
G
E
N
N
(dmphen) ]
The simultaneous addition of TEOA and [Co ACHTUNGTRENNUNG( bpy) ]
3
3
(
dmphen=4,7-dimethy-1,10-phenanthroline) was also tested
slightly increased the quenching efficacy for compound 6,
whereas, for compound 1, cooperative quenching was negli-
gible (see the Supporting Information, Figures S9–S11).
under the same conditions and this PS reached a maximum
volume of 43 mmol H after irradiation for 36 h.
2
These results demonstrate that these neutral complexes
can act as light-absorbing units in the photogeneration of
hydrogen. To reduce the protons, the electrons from the
photoexcited iridium complexes were delivered to the cata-
Photocatalytic H production is a relatively complex proc-
2
ess that involves multicomponent photochemical reactions.
Many factors, such as the photophysical and photoelectro-
chemical properties of the PS and the WRC species, the
electron donor, and the reaction conditions, can affect the
efficiency and performance of a photocatalytic reaction. As
mentioned above, under our experimental conditions with
an irradiation time of 72 h, benchmark cationic complex 5
exhibited slightly inferior catalytic activity with respect to
compounds 1–4; however, compound 6 exhibited the highest
activity among the molecules in this study. The poorly emit-
ting character and the poor absorption ability of compound
5 limit the utility of this complex as a PS for hydrogen pro-
duction. Although complex 6 exhibited a fairly weak absorp-
tion in the visible region, it was still very active as a chromo-
phore for hydrogen photoproduction; we ascribe this result
to the outstanding emission quantum yield of the excited
state of compound 6, which is related to the light-harvesting
process and its ability to undergo both oxidative and reduc-
tive quenching.
[39–42]
lytic cobalt center through oxidative quenching,
as sup-
ported by the fluorescence-quenching experiments. The lu-
minescence quenching of complexes 1–6 was systematically
examined in air-equilibrated solutions at room temperature.
The excited state of iridium complexes 1–4 cannot be reduc-
tively quenched upon the addition of hydroxy amines, such
as TEOA, even in large excess, because of the unfavorable
thermodynamics of electron transfer between the PS frag-
[43]
ment and TEOA. This result can be verified by compari-
son of the estimated excited-state-reduction potentials E
A
H
U
G
R
N
U
G
À
S ), which ranged from 0.52 to 0.64 V versus NHE, for
these neutral complexes with the oxidation potential of
[44]
TEOA (0.84 V versus NHE; Table 2). In addition, when
the catalysis was performed with colloidal platinum instead
2+
of [Co ACHTUNGTRENNUNG( bpy) ] as the reduction catalyst under otherwise
3
identical experimental conditions, no hydrogen was formed,
which indicated that reductive quenching by TEOA did not
occur in these cases. Thermodynamic analysis based on the
Influence of pH value and PS concentration on photocata-
lytic hydrogen production: By using more-active complex 1,
+
estimated excited-state-oxidation potential E
A
H
U
G
R
N
U
G
neutral complexes, which ranged from À0.97 to À1.34 V
versus NHE, thus suggested that the light-induced state of
these PSs has sufficient driving force to allow for electron
the effect of pH value on the activity of the H production
2
was also investigated. Figure 5 shows that H production is
2
most active in weakly basic solutions (pH 8.0, near the pKa
II
II
transfer to the Co centers (the reduction potential of Co /
of TEOA), similar to previously reported systems that used
I
[45]
[46–48]
Co is À0.95 V versus NHE). This speculation was further
cobalt catalysts.
At higher pH values, the amount of H2
demonstrated by the fact that these excited states could be
efficiently oxidatively quenched by [Co
that was produced decreased, possibly because of the small-
2+
A
H
U
G
R
N
U
G
er driving force for H production, owing to the lower con-
3
2
linear Stern–Volmer behavior (see the Supporting Informa-
centration of protons. At lower pH values, the decrease in
tion, Figures S2–S5), with quenching rate constants (K ) in
H production is attributed to the protonation of TEOA,
2
q
the order: K (2)>K (3)>K (1)>K (4), all within the same
which is a less-effective electron donor.
We also examined the effect of the concentration of com-
pound 1 on the photoinduced hydrogen production whilst
q
q
q
q
9
À1 À1
order of magnitude (ꢁ10 m s ; see the Supporting Infor-
mation, Figure S6). These observations indicated that elec-
tron transfer only comes from the PS excited state through
2+
keeping the concentrations of [Co
A
H
U
G
R
N
U
G
and TEOA
3
an oxidative-quenching mechanism. The trend in K value is
fixed, as shown in Figure 6. Clearly, increasing the concen-
q
Chem. Eur. J. 2013, 19, 6340 – 6349
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6345

Z.-T. Yu, Z.-G. Zou et al.
Figure 5. Influence of pH value on the photoinduced hydrogen evolution
2
+
by using complex 1 (5 mm). Conditions: [Co
A
H
U
G
R
N
N
(bpy)
3
]
(0.33 mm) and
Figure 7. Plot of the quantum yield (QY) of hydrogen production of com-
pound 1 as a function of the wavelength of the incident light. QY is de-
fined as twice the amount of hydrogen that is produced divided by the
number of photons that are absorbed at a given wavelength. Conditions:
TEOA (0.38m) in acetone/water (4:1 v/v, 100 mL) upon irradiation (Xe
lamp, l>420 nm).
2+
1
(50 mm), [Co
4:1 v/v, 100 mL, pH 8.0) upon irradiation for 24 h (l=350, 380, 400, 420,
and 440 nm). Inset: Excitation spectrum of compound 1 for emission at
83 nm in CH Cl
A
H
U
G
E
N
N
(bpy)
3
]
(0.33 mm), and TEOA (0.38m) in acetone/water
(
5
2
2
.
activity on the wavelength of incident light over complex 1
was investigated to determine the incident-photon-to-hydro-
gen-transfer yield of our system. Figure 7 shows a plot of
the quantum yield (QY) of hydrogen production over com-
pound 1 as a function of the wavelength of the incident
monochromatic light in the range 350–440 nm. The QY
showed a similar trend to the excitation spectrum, which
clearly indicated that the hydrogen-production reaction was
indeed driven by the excited-state of compound 1. The ob-
served QY value reached a maximum of 1.62% at 350 nm
over an irradiation time of 24 h (light intensity:
À2
0
.13 mWcm ) and gradually decreased with the lengthening
Figure 6. Photoinduced hydrogen production of various concentrations of
2
+
of the wavelength within the range 350–440 nm, owing to
the decrease in the energy of the absorbed photon.
compound 1 at pH 8.0. Conditions: [Co
A
H
U
G
R
N
N
(bpy)
3
]
(0.33 mm) and TEOA
(
0.38m) in acetone/water (4:1 v/v, 100 mL) upon irradiation (Xe lamp,
l>420 nm).
Stability of the PS for photocatalytic hydrogen production:
As is evident from the kinetics plots (Figure 4), all of the
neutral complexes exhibit far better durability than most of
the previously reported iridium-based systems over longer
irradiation times, which may be due to both the nature of
the PSs and to the oxidative-quenching pathway that is in-
volved in the electron-transfer process. However, our cata-
lytic systems seemed to undergo a deactivation process after
more than 60 h of photocatalysis, which may be related to
tration of the PS increased the rate and overall yield of hy-
drogen generation. With compound 1, the total PS turnover
number (TONs=n(H)/n(PS)) reached a maximum of ap-
proximately 602 at a concentration of 1 mm. At higher con-
centrations of compound 1, a larger amount of H was pro-
2
duced, thus indicating that the hydrogen formation was di-
rectly correlated to the concentration of the PS. When the
concentration of compound 1 was increased to 50 mm, a total
of 360 mmol of hydrogen was produced. Under the same
conditions, a system with 50 mm of complex 5 was only capa-
2+
the inactivation of the PS and/or the [Co ACHTUNGTERNNNU(G bpy) ] catalyst.
3
Complexes 1, 5, and 6 were chosen for the following studies.
When hydrogen production ceased, the decomposition prod-
ucts were collected from the reaction mixture of the system
with compound 1 (50 mm) as the PS after the photolysis ex-
periments and were determined by GCMS (see the Support-
ing Information, Figure S12). In this case, a dominant peak
at m/z=156 was observed, which was characteristic of the
bpy ligand and clearly revealed its dissociation from the
ble of producing 210 mmol of H . However, compound 6 still
2
showed more-efficient photocatalytic activity (875 mmol of
H produced) than compound 1.
2
Incident-photon-to-hydrogen-transfer efficiency: Under the
optimized conditions, the dependence of the photocatalytic
6346
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 6340 – 6349

Iridium Photosensitizers for Light-Induced Hydrogen Evolution
FULL PAPER
2
+
[
Co
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(bpy) ] units. Remarkably, no signs of free ligands that
the species occurred during the photocatalytic reaction. Re-
markably, enhanced activity and improved long-term photo-
stability of this system were achieved by the second addition
3
belonged to the PS were detected. Based on these observa-
tions, the dissociation of the [Co ACHTNUGTERNN(UNG bpy) ] catalyst contribut-
3
2+
2
+
ed to the loss of catalytic activity and limited the long-term
catalytic performance of the systems. For comparison, simi-
lar experiments were conducted for the reference systems
under identical conditions. As expected, MS data of the col-
lected decomposition products from the reactions with com-
plexes 5 or 6 clearly revealed that both the bpy (m/z=156)
and dtb-bpy ligands (m/z=268) were released during the
course of the photolysis (see the Supporting Information,
Figures S13 and S14). These observations were consistent
with previous reports in which both the catalyst and the PS
underwent decomposition when hydrogen production
ceased in the system that contained [Ir
PS.
To further determine the factors that were responsible for
the deterioration of the photocatalytic activity in these
abovementioned systems, each of the three essential compo-
nents was added again under the same conditions at a later
of [Co
ACHTUNGTNERNGNU( bpy) ] (Figure 8, curve d). The addition of extra
3
2+
[Co(bpy) ] appeared to slow the decomposition of WRC,
AHCTUNGTRENNUNG
3
[49]
thus resulting in a higher net amount of H2. This system
was subjected to five consecutive photochemical runs with
intermittent evacuation over a period of 360 h (Figure 8,
curves e–h). Continuous hydrogen evolution was clearly ob-
served in subsequent runs, which indicated that, when the
2
+
concentration of [Co ACHTUNGTERNN(UNG bpy) ] was high enough, hydrogen
3
[50]
generation lasted longer. The amount of H evolution in
2
the third (Figure 8, curve f) and subsequent runs was smaller
than that in the first run but the evolution continued to pro-
ceed steadily. The decrease over consecutive runs was prob-
ably caused by the decrease in the concentration of residual
+
A
H
U
G
R
N
U
G
A
H
U
G
E
N
N
(NN)] as the
2
2+
[Co ACHTUNGERTNNUNG( bpy) ] , owing to partial decomposition, as identified
3
by MS analysis of the reaction mixture after long-term visi-
ble-light irradiation. The decrease in the activity might also
be related to the decrease in pH value from 8.0 to about 7.0,
which is indicative of the consumption of TEOA during the
photolysis. The overall amount of hydrogen that was
evolved during all of the runs was 544.6 mmol (TON=2178),
time in the reaction to determine whether H could be re-
2
gained. The system with complex 1 provided much-higher
hydrogen-evolution performance than other neutral com-
plexes and, thus, it was selected to estimate the relative re-
quirements in these processes (Figure 8). Once the catalytic
activity had ceased, the addition of TEOA or compound 1
which far exceeded the amount with compound
6
(110 mmol). Notably, gas evolution was still observed upon
irradiation for up to 500 h and a larger TON is expected for
further irradiation. Thus, these combined results suggest
that the PS species is stable when subjected to photolysis for
a long time, which is a result of the increased binding affini-
ty of the ligands at the iridium center. This system has rela-
tively good stability compared to other reported homogene-
ous systems, which is critically important for achieving an ef-
ficient and robust production of hydrogen and provides a
sustainable solution for the implementation of such long-
term photocatalytic processes.
could not effectively resume H production; only 1.6 and
2
2
.0 mmol H were obtained after irradiation for 36 h, respec-
2
tively. However, the addition of the same amount of [Co-
(bpy) ] catalyst as that used in the photocatalytic reaction
was sufficient to restart H2 production and showed com-
pletely recovered activity after irradiation for a further 72 h
2+
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
3
(
Figure 8, curve c), thus suggesting that decomposition of
Conclusion
In conclusion, a new series of neutral tricyclometalated iridi-
um complexes is presented, including their photophysical
characterization and theoretical studies, and they are pro-
posed as new, promising PSs with superior long-term reac-
tivity for visible-light-driven H production in levels that are
2
beyond those of classical cationic iridium complexes. This
improvement is attributed to the process taking place
through an oxidative-quenching pathway. The stability is
also attributed to the three cyclometalating ligands, which
have high affinity for these iridium complexes. To improve
this system further, additional stabilization of the water-re-
ducing cobalt-based catalyst is required, which limits the
photocatalytic activity. This study represents a significant ad-
vancement in the rational design of photocatalytic systems
with enhanced longevity for solar-fuel production and also
opens up new perspectives for the development of innova-
tive molecular alternatives to artificial photosynthetic sys-
tems.
Figure 8. Photoinduced hydrogen production by using compound
1
(
5 mm), together with regeneration by adding extra TEOA or [Co-
2
+
A( bpy)
C
H
T
U
N
G
T
R
E
N
N
U
N
G
3
]
at various times. a) Hydrogen evolution from the start of irradi-
ation; b) the addition of TEOA at 72 h to the photoinactive mixture;
2
+
c) the addition of extra [Co
A
H
U
G
R
N
N
3
]
at 72 h; d) the second addition of
2
+
extra [Co
3
ACHTUNGTRENNUNG( bpy) ]
2
+
tion every 72 h (dashed line). Conditions: [Co
A
H
U
G
R
N
N
(bpy)
3
]
(0.33 mm) and
TEOA (0.38m) in acetone/water (4:1 v/v, 100 mL, pH 8.0) upon irradia-
tion (l>420 nm).
Chem. Eur. J. 2013, 19, 6340 – 6349
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6347

Z.-T. Yu, Z.-G. Zou et al.
1
19.4, 119.3, 118.8, 117.5 ppm; elemental analysis calcd (%) for
Experimental Section
IrC32
N
3
S
3
H
19
F
3
: C 48.59, H 2.42, N 5.31; found: C 48.10, H 2.54, N 5.41.
[
(
Ir
4): Yield: 36%; H NMR (500 MHz, CDCl
H), 7.50–7.62 (m, 7H), 7.35 (s, J=6.5 Hz, 1H), 7.24 (m, J=6.5 Hz, 2H),
.82 (t, J=6.5 Hz, 1H), 6.71 (s, 2H), 6.54 (s, 1H), 6.40 (d, J=8.5 Hz,
H), 2.86 ppm (s, 6H); C NMR (500 MHz, CDCl ): d=179.9, 162.7,
3
62.3, 148.9, 148.2, 137.0, 135.3, 134.6, 134.1, 128.1, 127.7, 119.6, 119.4,
19.2, 117.9, 117.8, 117.7 ppm; elemental analysis calcd (%) for
ACHTUNGTRENN(UNG thpy) {2-(N,N-dimethyl-4-phenyl)-5-(trifluoromethyl)benzothiazole}]
2
General: All of the synthetic reactions were performed under an atmo-
ACHTUNGTRENNUNGs phere of dry nitrogen by using standard Schlenk techniques or in a
glove box. The workup and purification procedures were performed in
air.
1
3
): d=7.73 (d, J=8.5 Hz,
3
6
1
1
1
1
3
General synthetic procedure for the iridium complexes: Iridium com-
plexes 1–4 were prepared according to a similar procedure as that out-
[
32]
lined in the Supporting Information, Scheme S1. Herein, only the syn-
thesis of [Ir(thpy) (2-phenylbenzothiazole)] is described in detail.
{Ir(thpy) Cl} ]: According to a literature procedure, a flask was charg-
ed with IrCl ·H (298 mg, 1.0 mmol), 2,2’-thienylpyridine (355 mg,
.2 mmol), and a 3:1 mixture of 2-methoxyethanol and water (20 mL).
The resulting mixture was heated at reflux (1408C) for 24 h under a N
IrC34
4 3 24 3
N S H F : C 48.97, H 2.90, N 6.72; found: C 48.41, H 3.09, N 6.86.
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
2
Complexes 5 and 6 were prepared according to literature procedures.
The characterization data were consistent with the published data for
these compounds. Details for their synthesis and characterization are
available in the Supporting Information.
[
51]
[
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
2
2
3
2
O
2
2
Theoretical calculations: All calculations were performed with the Gaus-
sian 09 program by using density functional theory (DFT). The geometric
and energy optimizations of the complexes were performed by using the
B3LYP/LANL2DZ basis set. The numerical calculations were performed
on the IBM Blade cluster system at the High-Performance Computing
Center (HPCC) of Nanjing University.
atmosphere. After cooling to RT, water (50 mL) was added and the re-
sulting precipitate was collected by filtration. Further purification was
performed by column chromatography silica gel (EtOAc/n-hexane, 3:1)
1
to afford a yellow powder in 84% yield. H NMR (500 MHz, CDCl
3
): d=
9
7
2
.02 (d, J=8.5 Hz, 2H), 7.64 (t, J=6.5 Hz, 2H), 7.53 (d, J=6.5 Hz, 2H),
.08 (d, J=6.5 Hz, 2H), 6.64 (t, J=6.5 Hz, 2H), 5.94 ppm (d, J=8.5 Hz,
H).
General procedure for photocatalytic hydrogen production: All of the
H -production experiments were evaluated by using a gas-closed circula-
2
tion system in a Pyrex vessel with side visible-light irradiation from a Xe
lamp (300 W) that was equipped with a cut-off filter (l>420 nm). The re-
[
Ir
solved in hot MeCN (150 mL). A solution of AgPF
in MeCN (50 mL) was added and the mixture was heated at reflux under
a N atmosphere in the dark for 2 h. The reaction was filtered to remove
the AgCl precipitate. The resulting filtrate was concentrated under re-
duced pressure and Et O (50 mL) was added. The product was collected
by filtration and dried under vacuum to give a yellow product (375 mg,
3% yield). H NMR (500 MHz, CDCl ): d=8.92 (d, J=6.5 Hz, 2H),
3
.93 (t, J=6.5 Hz, 2H), 7.66 (d, J=6.5 Hz, 2H), 7.30 (d, J=6.5 H, 2H z),
.25 (t, J=6.5 Hz, 2H), 5.99 (d, J=8.5 Hz, 2H), 3.29 ppm (s, J=6.5 Hz,
H).
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(thpy)
2
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(NCCH
3
)
2
]PF
6
:
[{Ir
A
H
U
G
R
N
U
G
2
Cl}
2
]
(300 mg, 0.3 mmol) was dis-
6
(150 mg, 0.6 mmol)
2+
3
actor was charged with the PS, the catalyst ([Co ACTHUNGTRENNUG( bpy) ] ), and TEOA
2
and was promoted with LiCl in aqueous solution under gentle magnetic
stirring. The specific pH value was adjusted with concentrated HCl. Prior
to visible-light irradiation, the system was successively evacuated before
being backfilled with argon gas. The evolved gases were periodically de-
tected in situ by GC that was equipped with a thermal-conductivity de-
tector (Shimadzu GC-8A, carrier gas: argon, column: MS-5A). The varia-
tion in hydrogen evolution was less than 10% upon repeated experi-
ments.
2
1
9
7
7
6
[
Ir
(NCCH
.64 mmol) in o-dichlorobenzene (30 mL) was heated at 1008C under a
atmosphere for 4 days. The crude product was purified by column
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(thpy)
2
(2-phenylbenzothiazole)] (1):
A
2
mixture of [Ir ACHNUTGTNERNUG( thpy) -
The apparent quantum yield for hydrogen production was measured
under the same photocatalytic reaction conditions and was determined as
the ratio between the number of reacted electrons during the course of
hydrogen evolution and the number of incident photons of a defined
wavelength. The number of incident photons can be calculated from the
intensity of the light, which was measured on a Newport Model 840 opti-
cal power meter with a 1 cm photodiode detector. The photocatalytic hy-
drogen reactions were performed for 24 h by using monochromatic light
that was irradiated at the following wavelengths: 350, 380, 400, 420, and
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
3
)
2
]PF (300 mg, 0.40 mmol) and 2-phenylbenzothiazole (134 mg,
6
0
N
2
chromatography on silica gel (n-hexane to elute the o-dichlorobenzene,
followed by hexanes/dichloromethane, 1:1) to afford the product as a
bright-orange/red band. After recrystallization from CH
compound was isolated as bright-orange crystals in 50% yield.
H NMR (500 MHz, CDCl ): d=7.82 (d, J=8.5 Hz, 1H), 7.71 (d, J=
.5 Hz, 2H), 7.58 (m, J=6.5 Hz, 2H), 7.50 (m, J=6.5 Hz, 3H), 7.26 (d,
2
Cl
2
/n-hexane,
2
1
1
3
6
4
40 nm. The light source was a Xe lamp that was equipped with various
J=6.5 Hz, 1H), 7.22 (dd, J=6.5 Hz, 2H), 7.09 (t, J=8.5 Hz, 1H), 6.94 (t,
J=6.5 Hz, 1H), 6.88 (d, J=6.5 Hz, 2H), 6.80 (t, J=8.5 Hz, 1H), 6.67 (m,
J=8.5 Hz, 2H), 6.39 (d, J=8.5 Hz, 1H), 6.34 ppm (t, J=8.5 Hz, 1H);
band-pass filters. Details of the method for calculating the quantum yield
are presented in the Supporting Information.
1
3
C NMR (500 MHz, CDCl
35.4, 134.5, 130.7, 128.2, 127.7, 126.8, 126.0, 124.7, 122.5, 120.3, 119.4,
3 3 20
19.2, 118.8, 117.5 ppm; elemental analysis calcd (%) for IrC31N S H :
3
): d=148.9, 148.4, 140.1, 138.2, 136.7, 136.6,
1
1
Acknowledgements
C 51.50, H 2.79, N 5.81; found: C 51.10, H 2.99, N 5.72.
Ir(thpy) (4-trifluoromethyl-2-phenylbenzothiazole)] (2): Yield: 40%;
H NMR (500 MHz, CDCl ): d=7.86 (d, J=6.5 Hz, 1H), 7.74 (d, J=
.5 Hz, 1H), 7.70 (d, J=6.5 Hz, 1H), 7.58 (t, J=6.5 Hz, 1H), 7.52 (dd,
[
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
2
This work was financially supported by the National Basic Research Pro-
gram of China (2013CB632400), the National Science Foundation of
China (20901038), and the Fundamental Research Funds for the Central
Universities (1106021342 and 1117021302). We are also grateful to the
Scientific Research Foundation for the Returned Overseas Chinese
Scholars, State Education Ministry.
1
3
6
J=6.5 Hz, 4H), 7.32 (t, J=8.5 Hz, 1H), 7.25 (d, J=6.5 Hz, 1H), 7.22 (d,
J=6.5 Hz, 1H), 7.15 (t, J=8.5 Hz, 1H), 7.12 (d, J=8.5 Hz, 2H), 7.11 (d,
J=8.5 Hz, 1H), 6.82 (t, J=8.5 Hz, 1H), 6.68 (t, J=6.5 Hz, 1H), 6.33 ppm
1
3
(
1
1
dd, J=6.5 Hz, 2H); C NMR (500 MHz, CDCl
3
): d=148.7, 148.3, 136.9,
35.3, 134.2, 133.1, 128.8, 128.0, 127.1, 125.5, 125.1, 122.7, 119.6, 119.4,
3 3 19 3
19.0, 117.8, 117.6 ppm; elemental analysis calcd (%) for IrC32N S H F :
[
[
1] A. J. Esswein, D. G. Nocera, Chem. Rev. 2007, 107, 4022–4047.
2] N. Armaroli, V. Balzani, Angew. Chem. 2007, 119, 52–67; Angew.
Chem. Int. Ed. 2007, 46, 52–66.
C 48.59, H 2.42, N 5.31; found: C 48.40, H 2.49, N 5.26.
Ir(thpy) (3-trifluoromethyl-2-phenylbenzothiazole)] (3): Yield: 34%;
H NMR (500 MHz, CDCl ): d=7.86 (s, 1H), 7.82 (d, J=6.5 Hz, 1H),
[
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
2
1
3
[3] J. Barber, Chem. Soc. Rev. 2009, 38, 185–196.
[4] M. G. Walter, E. L. Warren, J. R. McKone, S. W. Boettcher, Q. X.
Mi, E. A. Santori, N. S. Lewis, Chem. Rev. 2010, 110, 6446–6473.
[5] H. Krassen, S. Ott, J. Heberle, Phys. Chem. Chem. Phys. 2011, 13,
47–57.
7
4
1
1
6
1
.68 (d, J=6.5 Hz, 1H), 7.55 (t, J=6.5 Hz, 1H), 7.51 (dd, J=6.5 Hz,
H), 7.30 (d, J=8.5 Hz, 1H), 7.23 (d, J=6.5 Hz, 1H), 7.21 (d, J=6.5 Hz,
H), 7.09 (t, J=6.5 Hz, 1H), 7.05 (d, J=8.5 Hz, 1H), 7.00 (d, J=6.5 Hz,
H), 6.80 (t, J=6.5 Hz, 1H), 6.68 (dd, J=6.5 Hz, 2H), 6.33 ppm (dd, J=
1
3
.5 Hz, 2H); C NMR (500 MHz, CDCl
3
): d=159.5, 148.9, 148.4, 140.8,
[6] J. H. Alstrum-Acevedo, M. K. Brennaman, T. J. Meyer, Inorg.
Chem. 2005, 44, 6802–6827.
38.2, 136.7, 135.5, 134.5, 130.7, 128.2, 127.7, 126.7, 126.0, 124.5, 120.3,
6348
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 6340 – 6349

Iridium Photosensitizers for Light-Induced Hydrogen Evolution
FULL PAPER
[
[
7] M. Wang, Y. Na, M. Gorlov, L. Sun, Dalton Trans. 2009, 6458–6467.
8] A. Fihri, V. Artero, M. Razavet, C. Baffert, W. Leibl, M. Fontecave,
Angew. Chem. 2008, 120, 574–577; Angew. Chem. Int. Ed. 2008, 47,
[30] F. M. Hwang, H. Y. Chen, P. S. Chen, C. S. Liu, Y. Chi, C. F. Shu,
F. I. Wu, P. T. Chou, S. M. Peng, G. H. Lee, Inorg. Chem. 2005, 44,
1344–1353.
5
64–567.
[31] Z. Han, W. R. McNamara, M. S. Eum, P. L. Holland, R. Eisenberg,
Angew. Chem. Int. Ed. 2012, 51, 1667–1670.
[
9] S. Tschierlei, M. Karnahl, M. Presselt, B. Dietzek, J. Guthmuller, L.
Gonzꢃlez, M. Schmitt, S. Rau, J. Popp, Angew. Chem. 2010, 122,
[32] K. A. McGee, K. R. Mann, Inorg. Chem. 2007, 46, 7800–7809.
[33] M. S. Lowry, W. R. Hudson, R. A. Pascal, Jr., S. Bernhard, J. Am.
Chem. Soc. 2004, 126, 14129–14135.
[34] O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard, H.
Puschmann, J. Appl. Crystallogr. 2009, 42, 339–341.
[35] K. A. King, P. J. Spellane, R. J. Watts, J. Am. Chem. Soc. 1985, 107,
1431–1432.
[36] B. Happ, A. Winter, M. D. Hager, U. S. Schubert, Chem. Soc. Rev.
2012, 41, 2222–2255.
[37] P. J. Hay, J. Phys. Chem. A 2002, 106, 1634–1641.
[38] Md. K. Nazeeruddin, R. T. Wegh, Z. Zhou, C. Klein, Q. Wang, F.
De Angelis, S. Fantacci, M. Grꢄtzel, Inorg. Chem. 2006, 45, 9245–
9250.
4
073–4076; Angew. Chem. Int. Ed. 2010, 49, 3981–3984.
10] J. L. Dempsey, J. R. Winkler, H. B. Gray, J. Am. Chem. Soc. 2010,
32, 16774–16776.
11] S. Fukuzumi, T. Kobayashi, T. Suenobu, Angew. Chem. 2011, 123,
54–757; Angew. Chem. Int. Ed. 2011, 50, 728–731.
12] H. Ozawa, M. Haga, K. Sakai, J. Am. Chem. Soc. 2006, 128, 4926–
927.
13] T. A. White, B. N. Whitaker, K. J. Brewer, J. Am. Chem. Soc. 2011,
33, 15332–15334.
[
[
[
[
1
7
4
1
[
[
14] P. Du, R. Eisenberg, Energy Environ. Sci. 2012, 5, 6012–6021.
15] B. Probst, A. Rodenberg, M. Guttentag, P. Hamm, R. Alberto,
Inorg. Chem. 2010, 49, 6453–6460.
[
[
[
16] Z. T. Yu, Y. J. Yuan, J. G. Cai, Z. G. Zou, Chem. Eur. J. 2013, 19,
[39] C. V. Krishnan, B. S. Brunschwig, C. Creutz, N. Sutin, J. Am. Chem.
Soc. 1985, 107, 2005–2015.
[40] R. Ziessel, I. Hawecker, J. M. Lehn, Helv. Chim. Acta 1986, 69,
1065–1084.
[41] V. Artero, M. Chavarot-Kerlidou, M. Fontecave, Angew. Chem.
2011, 123, 7376–7405; Angew. Chem. Int. Ed. 2011, 50, 7238–7266.
[42] N. Sutin, C. Creutz, E. Fujita, Comments Inorg. Chem. 1997, 19, 67–
92.
[43] B. Ma, P. I. Djurovich, M. E. Thompson, Coord. Chem. Rev. 2005,
249, 1501–1510.
1
303–1310.
17] J. I. Goldsmith, W. R. Hudson, M. S. Lowry, T. H. Anderson, S.
Bernhard, J. Am. Chem. Soc. 2005, 127, 7502–7510.
18] L. J. Tinker, N. D. McDaniel, P. N. Curtin, C. K. Smith, M. J. Ireland,
S. Bernhard, Chem. Eur. J. 2007, 13, 8726–8732.
[19] L. L. Tinker, S. Bernhard, Inorg. Chem. 2009, 48, 10507–10511.
[20] S. Metz, S. Bernhard, Chem. Commun. 2010, 46, 7551–7553.
[21] B. F. DiSalle, S. Bernhard, J. Am. Chem. Soc. 2011, 133, 11819–
1
1821.
[
[
[
[
22] Y. J. Yuan, Z. T. Yu, X. Y. Chen, J. Y. Zhang, Z. G. Zou, Chem. Eur.
[44] H. Sun, M. Z. Hoffman, J. Phys. Chem. 1994, 98, 11719–11726.
[45] H. A. Schwarz, C. Creutz, N. Sutin, Inorg. Chem. 1985, 24, 433–439.
[46] J. L. Dempsey, B. S. Brunschwig, J. R. Winkler, H. B. Gray, Acc.
Chem. Res. 2009, 42, 1995–2004.
[47] P. Du, J. Schneider, P. Jarosz, R. Eisenberg, J. Am. Chem. Soc. 2006,
128, 7726–7727.
[48] C. F. Leung, S. M. Ng, C. C. Ko, W. L. Man, J. Wu, L. Chen, T. C.
Lau, Energy Environ. Sci. 2012, 5, 7903–7907.
[49] X. Wang, S. Goeb, Z. Ji, N. A. Pogulaichenko, F. N. Castellano,
Inorg. Chem. 2011, 50, 705–707.
J. 2011, 17, 12891–12895.
23] Y. J. Yuan, J. Y. Zhang, Z. T. Yu, J. Y. Feng, W. J. Luo, J. H. Ye,
Z. G. Zou, Inorg. Chem. 2012, 51, 4123–4133.
24] S. Jasimuddin, T. Yamada, K. Fukuju, J. Otsuki, K. Sakai, Chem.
Commun. 2010, 46, 8466–8468.
25] F. Gꢄrtner, S. Denurra, S. Losse, A. Neubauer, A. Boddien, A. Go-
pinathan, A. Spannenberg, H. Junge, S. Lochbrunner, M. Blug, S.
Hoch, J. Busse, S. Gladiali, M. Beller, Chem. Eur. J. 2012, 18, 3220–
3
225.
[
26] F. Gꢄrtner, D. Cozzula, S. Losse, A. Boddien, G. Anilkumar, H.
Junge, T. Schulz, N. Marquet, A. Spannenberg, S. Gladiali, M.
Beller, Chem. Eur. J. 2011, 17, 6998–7006.
[50] T. M. McCormick, B. D. Calitree, A. Orchard, N. D. Kraut, F. V.
Bright, M. R. Detty, R. Eisenberg, J. Am. Chem. Soc. 2010, 132,
15480–15483.
[
[
[
27] P. Zhang, M. Wang, Y. Na, X. Li, Y. Jiang, L. Sun, Dalton Trans.
[51] M. Nonoyama, Bull. Chem. Soc. Jpn. 1974, 47, 767–768.
2
010, 39, 1204–1206.
28] V. Marin, E. Holder, R. Hoogenboom, U. S. Schubert, Chem. Soc.
Rev. 2007, 36, 618–635.
29] S. H. Wadman, J. M. Kroon, K. Bakker, R. W. A. Havenith, G. P. M.
van Klink, G. van Koten, Organometallics 2010, 29, 1569–1579.
Received: January 15, 2013
Published online: March 20, 2013
Chem. Eur. J. 2013, 19, 6340 – 6349
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6349