Dinuclear iridium(III) complexes containing bibenzimidazole and their application to water photoreduction

Article abstract of DOI:10.1021/cs500296s

An efficient three-component catalytic system for visible-light-induced production of hydrogen from water was developed based on new dinuclear iridium photosensitizers (PSs), [Ir(tfdpyb)Cl]₂(BiBzIm) (P1) and [Ir(tfmppy)₂]₂BiBzIm (P2) [tfdpyb = 1,3-di(2-pyridyl)-4,6- bis(trifluoromethyl)benzene, tfmppy = 2-(4-(trifluoromethyl)phenyl)-pyridine, BiBzIm = 2,2′-bibenzimidazole]. These iridium complexes were fully characterized by ¹H NMR and ESI-MS, and their photophysical properties and electrochemical behaviors were also investigated. To compare with this new type of iridium compounds, the mononuclear analogues Ir(tfdpyb)(BiBzImH)Cl (P3), Ir(tfmppy)₂(BiBzImH) (P4), and Ir(dpyx)(BiBzImH)Cl (P5) (dpyx = 1,3-di(2-pyridyl)-4,6-dimethylbenzene) were also synthesized. The absorption spectra of dinuclear and mononuclear complexes are similar mainly in terms of their shape in the visible-light region, but as expected, the dinuclear species are more intense (approximately twice) compared with the corresponding mononuclear compounds. These complexes were tested as PSs with regard to their capacity to reduce water in the photocatalytic system for hydrogen production together with a series of water reduction catalysts and triethanolamine (TEOA) as a sacrificial electron donor. Turnover numbers up to 3780 for the dinuclear iridium complex P1 and 1020 for the mononuclear iridium compound P3 were obtained under the identical conditions. The results indicate that dinuclear iridium compounds can behave as PSs to be used for reducing water to hydrogen, and their activity was superior to that of the mononuclear compounds in this system. This work provides us a more general architectural guideline for constructing metal complexes as light-harvesting materials for visible-light-induced hydrogen production.

Full text of DOI:10.1021/cs500296s

Research Article
pubs.acs.org/acscatalysis
Dinuclear Iridium(III) Complexes Containing Bibenzimidazole and
Their Application to Water Photoreduction
Jian-Guang Cai,†,‡ Zhen-Tao Yu,*,† Yong-Jun Yuan,† Feng Li,‡ and Zhi-Gang Zou†
†National Laboratory for Infrared Physics, Chinese Academy of Sciences; Eco-Materials and Renewable Energy Research Center,
Department of Materials Science and Engineering, Nanjing University, Kunshan Innovation Institute of Nanjing University, No. 22,
Hankou Road, Nanjing, Jiangsu 210093, P. R. China
‡School of Chemical Engineering, Nanjing University of Science and Technology, No. 200, Xiaolingwei Street, Nanjing, Jiangsu
210094, P. R. China
S
* Supporting Information
ABSTRACT: An efficient three-component catalytic system for visible-light-induced
production of hydrogen from water was developed based on new dinuclear iridium
photosensitizers (PSs), [Ir(tfdpyb)Cl]2(BiBzIm) (P1) and [Ir(tfmppy)2]2BiBzIm (P2)
[tfdpyb = 1,3-di(2-pyridyl)-4,6-bis(trifluoromethyl)benzene, tfmppy = 2-(4-
(trifluoromethyl)phenyl)-pyridine, BiBzIm = 2,2′-bibenzimidazole]. These iridium
1
complexes were fully characterized by H NMR and ESI-MS, and their photophysical
properties and electrochemical behaviors were also investigated. To compare with this
new type of iridium compounds, the mononuclear analogues Ir(tfdpyb)(BiBzImH)Cl
(P3), Ir(tfmppy)2(BiBzImH) (P4), and Ir(dpyx)(BiBzImH)Cl (P5) (dpyx = 1,3-di(2-
pyridyl)-4,6-dimethylbenzene) were also synthesized. The absorption spectra of
dinuclear and mononuclear complexes are similar mainly in terms of their shape in
the visible-light region, but as expected, the dinuclear species are more intense (approximately twice) compared with the
corresponding mononuclear compounds. These complexes were tested as PSs with regard to their capacity to reduce water in the
photocatalytic system for hydrogen production together with a series of water reduction catalysts and triethanolamine (TEOA)
as a sacrificial electron donor. Turnover numbers up to 3780 for the dinuclear iridium complex P1 and 1020 for the mononuclear
iridium compound P3 were obtained under the identical conditions. The results indicate that dinuclear iridium compounds can
behave as PSs to be used for reducing water to hydrogen, and their activity was superior to that of the mononuclear compounds
in this system. This work provides us a more general architectural guideline for constructing metal complexes as light-harvesting
materials for visible-light-induced hydrogen production.
KEYWORDS: hydrogen evolution, iridium, photochemistry, photosensitizer, water splitting
opportunities to alleviate the above-mentioned limitations. It
has been widely recognized that the electronic properties and
chemical attributes of the iridium complexes can be finely tuned
by systematic control of the nature and position of the
appropriate functional substituents on the chelating ligands that
are bound to the metal center.24 The ligand properties can be
manipulated to design metal complexes with particular
functionality and may also strongly influence the course and
efficiency of the photochemical electron transfer involving the
metal complexes.
Through the use of transition metal catalysts and an external
electron donor, the photosensitizing behavior of many iridium
complexes has typically been evaluated for the photogeneration
of hydrogen using mononuclear bis-cyclometalated complexes,
[Ir(C∧N)2(N∧N)]+ (C∧N = 2-arylpyridine; N∧N = neutral
bipyridyl ligands), because of their relatively facile synthetic
accessibility and their good efficacy. Unfortunately, a general
INTRODUCTION
■
Transition-metal-based photoactive complexes are garnering
increasing amounts of interest as important components, i.e.,
efficient light-harvesting photosensitizers (PSs), in artificial
systems for solar energy conversion.1−6 A key characteristic that
makes these complexes unique is their capacity to absorb low-
energy visible light and thereby access the energetic metal-to-
ligand charge-transfer excited states capable of photoactiva-
tion.7−9 In designing such synthetic light-harvesting species for
solar energy conversion strategies, it is imperative to explore
systems that maximize the efficiency of photon absorption,
whereas the construction of structural complexity is to be
minimized.10−14 Among these systems, iridium-containing
complexes carrying bidentate ligands, such as 2,2′-bipydirdine
and their derivatives, express this unusual utility due to their
relevant photophysical and photoredox properties for such
applications.15−23 The development of such chromophores
suffers from relatively weak absorption in the visible region,
which impairs the overall functional performance and efficiency
of the solar energy conversion system. The tailoring of
materials under synthetic design control will provide unique
Received: March 5, 2014
Revised: April 24, 2014
Published: May 2, 2014
© 2014 American Chemical Society
1953
dx.doi.org/10.1021/cs500296s | ACS Catal. 2014, 4, 1953−1963

ACS Catalysis
Research Article
problem with the use of this type of light-absorbing molecule is
the inherently low stability caused by the easy rupture of weak
Ir−N bonding in the Ir-bipyridyl moiety, which leads to the
termination of catalysis during irradiation.25 Improving the
stability properties can be achieved to a certain degree by
structural modifications, such as the introduction of vinyl group
or pyridyl pendants at the bipyridyl ligand backbone of the
iridium complex.26−28 Additionally, an alternative approach has
been developed to minimize the photolability of the
chromophore using neutral tris-cyclometalated iridium com-
plexes.29,30 The implementation of tris-cyclometalation pro-
vides such complexes with improved long-term stability along
with extensive electronic couplings between the d-orbital of the
Ir center and π-orbital of the ligands. In this approach, the Ir−C
bond between the central Ir metal and the ligand is sufficiently
strong to survive the necessary changes in oxidation state that
occur during a photochemical reaction. Clearly, the exploration
of complexes with new structural motifs aimed to facilitate
photoinduced electron-transfer processes is worth pursuing to
contribute to a better understanding of the structure−activity
correlations of this class of materials.
Scheme 1. Chemical Structure of Iridium Compounds P1−
P5
In the course of our studies related to the design and
development of new potential light-harvesting materials that
can play an active role in hydrogen production, we have
become interested in conjugated organic unit-bridged dimetallic
iridium assemblies with the intention of achieving improved
optical characteristics and the associated performance improve-
ments in such processes. Dinuclear complexes are of particular
interest because they display exquisite electronic and photo-
chemical properties that are distinctly different from those of
their mononuclear counterparts.31−42 However, the use of this
class of compounds for fundamental studies of the photo-
chemical conversion of solar energy remains limited, most likely
due to the synthetic difficulties involved. In the design of such
systems, the suitable bridging ligands are responsible for
separating the photoactive terminals and ensuring electronic
coupling between the metal partners.
The use of 2,2′-bibenzimidazole (BiBzImH2) has been
extensively studied as a symmetrical two-bidentate ligand for
the formation of dinuclear complexes with five-membered
chelate rings, particularly with Ru, either as heterometallic or
homometallic systems.43−46 Related complexes of the BiB-
zImH2 ligand exhibit enhanced electron transport ability and
notable electronic coupling between two metals due to the
great electron motilities of the bridging ligand. These studies
motivated us to explore BiBzImH2 as a bridging ligand in
connection with pairs of iridium complex fragments for the
construction of appealing dinuclear iridium complexes, which is
appropriate from structural and photophysical viewpoints. In
this context, two new dinuclear iridium complexes (compounds
P1 and P2 shown in Scheme 1), which use a BiBzImH2 ligand
to link together two cyclometalated Ir(III) species, have been
synthesized. The electrochemical and photophysical properties
measured for the dinuclear compounds are discussed in
comparison with those obtained for the mononuclear analogues
P3, P4, and P5, containing one terdentate cyclometalating
ligand that can bind to iridium in an N∧C∧N coordination
mode or two ortho-metalating ligands. The resulting dinuclear
compounds exhibit significantly different photophysical and
electrochemical attributes with respect to their corresponding
mononuclear complexes. In particular, the present report
indicates that photocatalytic hydrogen production is possible
using such dinuclear complexes as P1 and P2 as the active PS
components and is rationalized in terms of the photophysical
properties of the dinuclear complexes.
RESULTS AND DISCUSSION
■
Synthesis and Characterization. To date, few examples
of dinuclear iridium complexes incorporating a π-conjugated
bridging ligand have been reported in the literature, and all of
these complexes are cationic.31−42 The straightforward
synthetic strategy used in the present study to prepare the
target complexes and reference complexes is outlined in
Scheme 2. A similar two-step strategy has previously been
successfully applied to the synthesis of mononuclear iridium
complexes. The molecules 1,3-di(2-pyridyl)-4,6-bis-
(trifluoromethyl)benzene (tfdpybH), 1,3-di(2-pyridyl)-4,6-di-
methylbenzene (dpyxH), and 2-(4-(trifluoromethyl)phenyl)-
pyridine (tfmppy) were prepared as the cyclometalating ligands
according to a literature procedure by the traditional palladium-
catalyzed Still cross-coupling method.47,48 As reported, the
complexation of 1,3-di(2-pyridyl)benzene with iridium primar-
ily preferred a bidentate mode of coordination (C∧N), rather
than the terdentate structure (N∧C∧N). The introduction of
substituents at the C4 and C6 positions of the central ring
resulted in the formation of the terdentate-bound product.49
The use of dpyxH allows us to prepare different iridium
complexes for understanding the effects of substituents on the
photophysical properties of the desired complexes.
The first step involves the preparation of a chloride-bridged
Ir(III) dimer as the precursor by reacting the cyclometalating
ligand with IrCl3·3H2O in aqueous 2-ethoxyethanol according
to the Nonoyama protocol. We attempted the synthesis of the
target chloro complexes of iridium(III) from the conventional
bridge-splitting reaction of the respective dichloro-bridged
intermediates with ca. 2.2 equiv of BiBzImH2 in the presence of
a proton scavenger, Na2CO3, in CH2Cl2/CH3OH (v/v = 1:1)
under reflux. The reaction of BiBzImH2 with the dimer
intermediate of either tfdpybH or tfmppy led to mixtures of
mono- and dinuclear iridium complexes, which were easy to
separate by a silica column to obtain the resulting iridium
complexes [Ir(tfdpyb)Cl]2 (BiBzIm) (P1), [Ir-
(tfmppy)2]2BiBzIm (P2), Ir(tfdpyb)(BiBzImH)Cl (P3), and
Ir(tfmppy)2(BiBzImH) (P4). In both reactions, the cyclo-
1954
dx.doi.org/10.1021/cs500296s | ACS Catal. 2014, 4, 1953−1963

ACS Catalysis
Research Article
Scheme 2. Synthetic Route of Iridium Complexes P1−P5
metalation proceeded smoothly and cleanly; however, attempts
to prepare the related complexes containing the dpyxH ligand
by the same method were unsuccessful. This finding was most
likely due to the low solubility of the intermediate of dpyxH in
the reaction solvent. In this case, even with a longer reaction
time, mainly starting materials were recovered. In an alternative
approach, the chlorides in the dimer precursor based on dpyxH
were substituted with BiBzImH2 in the presence of silver triflate
(AgOTf) as a chloride scavenger,50 followed by column
chromatography, resulted in the clean formation of the
mononuclear complex Ir(dpyx)(BiBzImH)Cl (P5) as a sole
product. After the complexation, the protons on the BiBzImH2
in all of the mononuclear complexes were partially deproto-
nated. Several attempts were made to react the mononuclear
BiBzImH-containing complexes with the second metal center,
such as RhCl3·3H2O, Pd(CH3CN)2Cl2, and K2PtCl4; however,
efforts to form the heterometallic complexes by the second
deprotonation with Na2CO3 were unsuccessful with the original
mononuclear precursors being recovered. All of the iridium
complete the coordination sphere of the metals. After
recrystallization, all samples were used for the following studies.
Photophysical Properties. The electronic absorption
spectra of all of the complexes were recorded at room
temperature in CH2Cl2 solutions, as shown in Figure 1. The
absorption spectra of all of the complexes were featured by the
most intense absorption bands in the ultraviolet region at
wavelengths below 390 nm. This was attributed to the spin-
allowed 1π−π* transition localized on both coordinated
aromatic ligands within each complex, with reference to the
1
complexes were fully identified by a combination of H NMR
spectroscopy and mass spectrometry, the results of which are
given in the Experimental Section, along with the single crystal
X-ray diffraction of the chloride-free neutral complexes P2 and
P4 containing tfmppy ligands (see the Supporting Informa-
tion). Moreover, we used HRMS analyses to confirm that the
negatively charged chloride group exists as an auxiliary ligand
for the monodeprotonated iridium complexes P3 and P5 to
Figure 1. Electronic absorption spectra of all complexes (10−5 M) in
CH2Cl2.
1955
dx.doi.org/10.1021/cs500296s | ACS Catal. 2014, 4, 1953−1963

ACS Catalysis
Research Article
previous well-established photophysical studies of the iridium-
(III) diimine complexes.31−42 These bands are less intense with
maxima at lower energies (400−430 nm), most likely arising
from the metal-to-ligand charge-transfer (MLCT) transition in
the singlet manifold. It can be observed that the absorption
spectra of dinuclear and mononuclear complexes are mainly
similar in terms of their shape in the visible-light region;
however, the dinuclear species are twice as intense compared to
the corresponding mononuclear compound, which corresponds
with the number of independent iridium-containing cores. Such
an increase in absorption allows for the ability to capture more
solar energy for this type of light-harvesting material. The
lower-energy band tailed toward the lower energies from 460 to
Table 1. Photophysical and Electrochemical Properties of
Studied Iridium Complexes P1−P5
d
E0−0
a
b
b
c
c
PS
λem (nm)
568
τ
[μs] Φ [%] Eox [V] Ered [V]
[eV]
P1
P2
P3
P4
P5
0.29
11.8
37.7
55.8
28.8
+0.68
+1.03
+0.73
+0.81
+0.74
−0.89
−1.45
−0.84
−1.56
−1.07
2.23
2.42
2.22
2.47
2.51
500, 532
578
1.28
1.55
2.01
0.22
496, 527
512, 540 (sh)
a
Measurements in air-equilibrated CH3OH at room temperature.
Measurements in N2 atmosphere in CH3OH at room temperature.
b
For P5, the fluorescence intensity was too low to quantify a quantum
c
yield (Φ). Redox potentials measured in deoxygenated CH3CN with
3
550 nm. This band is likely associated with MLCT; however,
0.1 M n-Bu4NPF6 as the supporting electrolyte; potentials measured vs
3
mixed with π−π* transition character, this is based on the
Ag/AgCl coupled and converted to normal hydrogen electrode
d
observation of similar absorptions of the corresponding iridium
complexes. The energy and intensity of the MLCT transitions
is usually affected by the nature of ligand architectures in the
iridium complexes. A difference in optical density and peak
maxima for the present complexes is observed, indicating the
electron-donation capability of the cyclometalating aryl
moieties. The lower energy MLCT band in the terdentate
N∧C∧N-containing complexes is significantly red-shifted with
respect to the C∧N-containing complexes. The red shift is due
to a lowering of the energy of the acceptor ligand π* orbital
upon increased electronic delocalization (stability) throughout
the entirety of the three aromatic rings of the N∧C∧N ligand.
The iridium(III) complexes display intense emission in
solution at ambient air and room temperature, with the
exception of P5. For mononuclear complex P5, no emission
was detected, suggesting that it was sensitive to oxygen and
subsequently completely quenched in air. The normalized
luminescence spectra of these investigated complexes in an N2-
saturated methanol solution are presented in Figure 2, and the
(NHE); only the first reduction and oxidation wave were shown. E0−0
values were calculated from the intersection of the normalized
absorption and the emission spectra.
change in maximum emission also indicated that the formation
of the dinuclear complexes exert an influence on the energy of
the emitting states. However, the photoluminescence spectra of
complexes P2 and P4 have a similar shape and display vibronic-
structured emission bands. The luminescence of terdentate
complexes P1 (λmax = 568 nm) and P3 (λmax = 578 nm) are
substantially red-shifted in comparison to those observed for
complexes P2 (λmax = 500 nm) and P4 (λmax = 496 nm). As in
the absorption spectra, this can be attributed to a lowering of
the energy of the N∧C∧N ligand π* orbital due to increased
conjugation as a result of the introduction of an additional
pyridyl ring compared to the bidentate cyclometalating ligand.
Little shift in the value of λmax was observed in the emission for
these iridium complexes in CH2Cl2 under N2 atmosphere. We
also determined the excited-state lifetime (τ) and the absolute
quantum yield (Φ) of the complexes in a dilute N2-saturated
CH3OH solution (10−5 M), the results of which are
summarized in Table 1. The values are consistent with those
previously observed for cyclometalated iridium complexes.31−42
The lifetimes of the dinuclear complexes P1 (0.29 μs) and P2
(1.28 μs) are notably shorter than those of their corresponding
mononuclear complexes P3 (1.55 μs) and P4 (2.01 μs) under
the same conditions.
Electron-transfer quenching of the excited states of
complexes in solution at room temperature, with TEOA,
Co(bpy)32+, and Rh(dtb-bpy)3+ acting as quenchers, were
conducted to probe the influence of the excited state of the
complexes on the quenching process. We detected the emission
spectral changes upon the addition of increasing amounts of
TEOA, Co(bpy)32+, or Rh(dtb-bpy)3+ to the acetone/water
(4:1 v/v) solution of complexes P1−P4 (see the Supporting
Information). P5 yielded notably weak emission in air-
equilibrated solution and thus did not allow reliable
quantification of the quenching process. The results of these
experiments are presented in Table 2. Notably, all of the
dinuclear and mononuclear iridium complexes P1−P4 are
Figure 2. Emission spectra of all complexes (10−5 M, in N2) in
methanol at room temperature.
corresponding photophysical data are collected in Table 1. In
contrast to the absorption spectra, the emission spectra of the
dinuclear and mononuclear complexes differ significantly. The
emission colors given by these iridium complexes range from
green to orange (496−578 nm), with the emission energies in
the order of P4 > P2 > P5 > P1 > P3. The color shifts
depended on the organic ligand chromophores with the
difference in the ligand field strength. The terdentate complexes
P1 and P3 show a broad, structureless emission band centered
at λmax = 568 (P1) and 578 nm (P3), indicating that the
easily oxidatively quenched by Co(bpy)3 or Rh(dtb-bpy)3+,
2+
following good Stern−Volmer behavior. The terdentate
complexes P1 and P3 can also be reductively quenched by
TEOA following good Stern−Volmer behavior, whereas
compounds P2 and P4 are not reductively quenched by
TEOA, similar to other neutral cyclometalated iridium
complexes.29,30 The observed quenching behavior of P1 or
P3 is similar to that observed for other charged complexes with
3
emission can be attributed to the prevalent MLCT state. The
1956
dx.doi.org/10.1021/cs500296s | ACS Catal. 2014, 4, 1953−1963

ACS Catalysis
Research Article
energy levels are similar. These results are supported by the
DFT calculations below, which indeed show similar HOMO
energies for the two complexes. In contrast to the oxidation
process, the reduction involving the addition of an electron to
the lowest unoccupied molecular orbital (LUMO) was thought
to occur at the surrounding electronegative ligands with little
contribution from the iridium metal center. Upon reduction, all
of the terdentate complexes showed the first wave with
potentials in the range of −0.84 V to −1.07 V vs NHE, which
were significantly positively shifted compared to those of the
bidentate cyclometalated iridium complexes (−1.45 V to −1.56
V).
Table 2. Quenching Kinetics of Studied Iridium
Complexes.
a
2+
kq Co(bpy)3
kq Rh(dtb-bpy)3+
(M−1 s−1
kq TEOA
PS
(M−1 s−1
)
)
(M−1 s−1
)
P1
P2
P3
P4
1.59 × 1010
3.36 × 109
3.42 × 1010
6.96 × 1010
2.00 × 1010
2.73 × 109
2.71 × 109
1.94 × 109
2.62 × 107
no quenching
4.52 × 109
no quenching
b
P5
a
b
Measured in acetone/water (4:1 v/v) at room temperature. P5 gave
very weak emission in the air-saturated solution.
The above interpretations are further supported through
DFT calculations. The electronic distribution between the
HOMO and LUMO for each complex indicates an obvious
intramolecular charge separation, as illustrated in Table 3. The
an N∧N ancillary ligand such as [Ir(C∧N)2(N∧N)]+.21,22 The
phenomenon is likely related to the fact that compounds P1
and P3 contain one monodentate halide ligand for possible
interaction with TEOA in the quenching process. The results
indicated that chromophores P1 and P3 undergo oxidative, as
well as reductive, quenching of their respective excited states.
The value of the quenching constant (kq) for each complex
with each of the two quenchers was determined from the
lifetimes in the absence of quencher using the Stern−Volmer
Table 3. HOMO and LUMO Distributions of Iridium
Complexes P1−P5
2+
equation. The rate for quenching by Co(bpy)3 was at least 1
order of magnitude faster than the rate by TEOA for terdentate
complexes, which is in agreement with the observation for the
cases of charged complexes with an N∧N ancillary ligand such
as [Ir(C∧N)2(N∧N)]+.18
Electrochemical Properties. Electrochemical measure-
ments were performed to determine the redox potentials of
all five complexes in deoxygenated MeCN, the results of which
are shown in Table 1. The electrochemical properties of these
iridium compounds are too complex to understand exactly the
oxidation and reduction involving the transfer of electrons,
which have been observed in previous reports on the metal
complexes based on the BiBzIm ligand.43−46 In this study, only
the first oxidation and reduction were discussed, which play an
important role in the photochemical processes. The cyclic
voltammograms of the five compounds exhibit irreversible first
oxidation potentials in the range of 0.68−1.03 V versus NHE.
According to the previously published works concerning
analogous iridium cyclometalates and on the basis of the
quantum mechanical calculations (vide infra), the oxidation
could mainly be associated with iridium-centered orbitals and σ
band Ir−C orbitals derived from the removal of an electron
belonging to the highest occupied molecular orbital (HOMO).
The first oxidation potential of the dinuclear compound P1
(E1/2 = 0.68 V vs NHE) is lower than that of the mononuclear
compound P3 (E1/2 = 0.73 V vs NHE), which may be due to
the electronic coupling between two metals, leading to a
decrease of the electronic density on the metal. This evidence
also indicates that the BiBzIm bridging ligand in the dinuclear
complex has electron-donor ability.46 The results are associated
with the observation that the mononuclear compound P3 is
more efficiently quenched by TEOA than the dinuclear
complex; however, the first oxidation potential of the dinuclear
complex P2 (E1/2 = 1.03 V vs NHE) is higher than that of the
mononuclear complex P4 (E1/2 = 0.81 V vs NHE). These
results indicate that the BiBzIm bridging ligand in the dinuclear
complex P2 may be an electron acceptor. The observed
oxidation potential of P3 (E1/2 = 0.73 V vs NHE) appeared
close to that of P5 (E1/2 = 0.74 V vs NHE), which suggests that
the contributions from the surrounding ligand with different
electron donor/acceptor abilities of P3 and P5 to the HOMO
HOMO differs depending on the strength of the bonding
properties of the ligand backbones. For the dinuclear complexes
P1 and P2, the HOMO is formed mainly from a combination
of Ir 5d orbitals and the entire conjugated system of the BiBzIm
bridging ligand, while the HOMO for the mononuclear
complexes P3−P5 originates from a combination of Ir 5d
orbitals and half (cyclometalated fragments) of the BiBzIm
ligand. Among the chloro complexes, there is significant
1957
dx.doi.org/10.1021/cs500296s | ACS Catal. 2014, 4, 1953−1963

ACS Catalysis
Research Article
electron density on the monodentate halide ligand in the
HOMO, which is expressed experimentally by the close
oxidation potential between P3 and P5. In contrast, the
LUMO for N∧C∧N-containing complexes and C∧N-containing
complexes are mainly centered on the one N∧C∧N and two
C∧N ligands with little involvement of the Ir orbitals,
respectively, which is consistent with the observed first
reduction potentials staying in a narrow range in the related
Ir complexes. On the basis of the results of DFT calculations,
the excited state in these complexes can be approximately
viewed as charge-transfer states mixed with MLCT/LLCT
(ligand-to-ligand charge transfer) character.
Moreover, this condition may inhibit hydrogen production
because the cobalt hydride intermediate species, which are
proposed to be the active species involving the observed
photocatalysis, will be too stable to release hydrogen if the pH
is too high.51,52 On the contrary, lowering the pH value reduces
the electron-donating ability of the protonated TEOA, offering
an electron to the PS and thus reducing the reaction activity.
For all of the reported complexes, we measured their catalytic
effectiveness in the reduction of water to hydrogen with a
typical end-point reached after an irradiation time of 72 h for
the dinuclear complexes P1 and P2 and 32 h for the
mononuclear complexes P3−P5. These data are collected in
Table 4. The photocatalytic capacity of the dinuclear complexes
Hydrogen Production Experiment. For applications in
light-induced hydrogen production, the photoactive complex as
a PS should have visible-light sensitivity and function as an
efficient charge transfer sensitizer to the catalyst for water
reduction. The electrochemical and photophysical results
clearly indicated that all of the complexes have the potential
to work as PSs in photochemical processes to reduce water into
hydrogen. The dinuclear complexes are of particular interest
because their absorption in the visible region is almost twice as
intense as the mononuclear compounds. Such behavior is
expected for compounds acting as PSs to make significant
contributions to the system for solar energy conversion. To
identify the dinuclear complexes as PSs, the photocatalytic
reaction activities for hydrogen generation were assessed and
compared to those of the mononuclear species, by adopting a
three-component catalytic system in the presence of the water
reduction catalyst (WRC) and TEOA as a sacrificial electron
donor in a mixture of acetone and water. Quantitative
determination of the generation of H2 was conducted by GC
analysis of the gas phase of the reaction system using a
molecular sieve column with Ar as the carrier gas (for details,
see the Experimental Section). The reaction turnover numbers
[TONs = n(H)/n(PS)] were calculated according to one-
electron transfer processes. The experimental results with value
variables within less than 10% have repeatedly proven to be
reproducible. No hydrogen was detected in the control
experiments when PS, TEOA, or WRC was excluded from
the system, demonstrating that all of the components are
indispensable for the production of hydrogen in this system.
Moreover, when the reaction was performed in the dark, no
hydrogen was detected by GC analysis.
Initially, production of hydrogen occurred using [Co(bpy)3]-
Cl2 as a WRC upon irradiating deaerated solutions under visible
light (Xe lamp with a cutoff filter, 300 W, λ > 420 nm). In a
typical photocatalytic experiment, the three-component system
was used in a 100 mL solution of water/acetone (1:4)
containing PS (10 μM), [Co(bpy)3]Cl2 (0.33 mM), and TEOA
(0.19 M) at a pH value lower than the pKa of TEOA, pH = 8.0.
The conditions were identical to those used for hydrogen
photogeneration in our previous reports,29,30 which have been
optimized for the catalytic action of related PS compositions.
The acetone/water mixed solvent was used to solubilize all of
the components. The adjustment of pH value of the solution is
accomplished by addition of hydrochloric acid. According to
the previous related studies, the effect of the pH is complex and
control of pH at the optimum level is required. The systems
employing a cobalt catalyst, such as [Co(bpy)3]Cl2, have shown
a pH-dependent activity for hydrogen evolution and perform as
a more active WRC at approximately pH 8. At higher pH
values, there is lower proton concentration and hydrogen
production becomes more thermodynamically unfavorable.
Table 4. Photocatalytic H2-Producing Activity of Different
Photosensitizers
a
run
PS
H2 (μmol)
TON
b
1
P1
P2
P3
P4
P5
73
67
23
34
6
146
134
46
b
2
c
3
c
4
68
c
5
12
d
6
[Ir(tfm-ppy)2(dtb-bpy)]+
74
148
a
Photocatalytic conditions: 10 μM PS, 0.33 mM [Co(bpy)3]Cl2, and
0.19 M TEOA in acetone/water (4:1 v/v, 100 mL, pH 8.0) upon
b
irradiation (Xe lamp; 300 W, λ > 420 nm). Irradiation for 72 h.
c
d
Irradiation for 32 h. Data taken from ref 23, where tfm-ppy =4-
trifluoromethyl-2-phenylpyridine and dtb-bpy =4,4′-di-tert-butyl-2,2′-
dipyridyl.
were superior to those of the mononuclear complexes under
the given reaction conditions. When subjected to photolysis for
a longer period of time, after 32 h irradiation, the dinuclear
complexes P1 and P2 continued producing H2, whereas the
activity of the mononuclear complexes P3 and P4 slowed down
substantially. After the hydrogen production experiment with
P3 had proceeded for 48 h, the system achieved a total
hydrogen production of 27 μmol. For P5, no more hydrogen
was produced upon extending the irradiation time. These
results indicate that there is a correlation between the
photophysical properties and catalytic activity. All fluorescing
complexes are active for photocatalytic hydrogen production,
while the low activity of P5 was consistent with both the
fluorescence intensity and the lifetime of the excited state. The
catalytic activity and stability of the systems were improved
when using the dinuclear complexes as PSs, which benefited
from the photophysical properties and the relative symmetry of
the molecular structures of these complexes. However, the
TON was far less than for other previously reported
mononuclear iridium-based photocatalytic systems,53 in which
synthetic modification and screening has been well established
for improvement in catalytic activity and efficiency. Addition-
ally, the TON depends strongly on the experimental
conditions, such as the reaction scale, the experimental setup,
the incident light power density, and so on. It should be taken
into account that the reaction scale used in our experiments is
different from most of the previous related reports. Our results
are comparable to those of the related neutral homoleptic and
heteroleptic cyclometalated Ir(III) complexes, which were
measured in the same setup and identical conditions (10 μM
PS, 100 mL solutions).23,29,30 As described before, the kq for the
TEOA (P1 and P3) is lower than that for the Co catalyst;
1958
dx.doi.org/10.1021/cs500296s | ACS Catal. 2014, 4, 1953−1963

ACS Catalysis
Research Article
however, in the electron transfer step, reductive quenching still
prevails over oxidative quenching because of the much higher
concentration of TEOA (0.19 M) in the photoreactions
compared with that of the catalyst (0.33 mM). For P2 and
P4, an oxidative quenching pathway in the photochemically
driven step leads to H2 evolution.
To improve the TON, a series of WRCs were evaluated
under identical experimental conditions to those of the cobalt
catalyst as described above. Among the iridium complexes
tested, P1 yielded the highest TONs and was used in
subsequent studies compared to the corresponding mono-
nuclear complex P3. The results are reported in Table 5. When
ligand on the iridium center may modulate the cationic charge
on the complexes in a controlled fashion, resulting in
complexes with certain behavior as the cationic iridium
complexes under the special conditions.54 Subsequently, we
chose [Rh(dtb-bpy)3](PF6)3 as the WRC to study its activity
toward hydrogen production. With P1, a total of 214 μmol H2
was obtained with a TON of 428 after 32 h of irradiation.
Though the long-term stability of the photocatalytic process
was reduced, the system yielded more hydrogen when using
[Rh(dtb-bpy)3](PF6)3 instead of [Co(bpy)3]Cl2 as a WRC,
which has previously been observed in the molecular system of
hydrogen production.30 Additionally, the use of CoCl2 as a
WRC was also attempted. For complex P3, in particular, it is
expected that the complex spontaneously assembles with a
Co(II) ion, making use of the free binding sites to form a
supramolecular Ir−Co complex, as reported in the literature.9
Unfortunately, this attempt failed as no appreciable amount of
H2 was obtained after 8 h of irradiation under the present
conditions. The reason may be that the N−H proton on the
BiBzImH ligand restrains the interactions between the Co
metal center and the BiBzImH ligands. From the two series of
experiments altering the WRC, [Rh(dtb-bpy)3](PF6)3 was
found to be the more active reduction catalyst than the others
(K2PtCl4, Co(bpy)3Cl2, or CoCl2) with these iridium
complexes.
Table 5. Photocatalytic H2-Producing Activity of P1 or P3
a
Using Different Catalysts
b
run
PS
catalyst
t (h)
H2 (μmol)
TON
1
2
3
4
5
6
7
8
P1
P1
P1
P1
P3
P3
P3
P3
[Co]
72
5
73
35
0
146
70
0
K2PtCl4
CoCl2
[Rh]
8
32
32
5
214
23
9
428
46
18
0
[Co]
K2PtCl4
CoCl2
[Rh]
8
0
32
165
330
a
Photocatalytic conditions: 10 μM PS, 0.33 mM catalyst and 0.19 M
In light-induced catalytic experiments, the solvent plays a
profound role in the activity and stability of the catalytic system.
The ligating solvent influences the catalytic reaction by its
ability to participate in the binding of the redox intermediates.
This medium can also affect the photophysical properties of the
excited state of PSs, such as energy, fluorescence quantum yield,
lifetime and the rate of intermolecular electron transfer between
PS and WRC. This effect is considered to be closely associated
with the efficiency of hydrogen generation. A series of
photocatalytic experiments based on the Ir−Rh-TEOA system
by changing the ratio of acetone/water was carried out to study
the effect of solvents with component concentrations of 10 μM
P1 or P3, 0.33 mM [Rh(dtb-bpy)3](PF6)3, and 0.19 M TEOA
at pH 8.0. For P1, a drastic solvent dependence for catalysis
was observed, as shown in Figure 3A. When using the co-
solvent acetone/water (9:1), the system with P1 had the best
performance with a total of 298 μmol H2 (TON = 596) formed
after 32 h irradiation. If using a higher water content in the
acetone solution, the activity declined significantly. This result
may be related to the intermolecular electron transfer rate,
TEOA in acetone/water (4:1 v/v, 100 mL, pH 8.0) upon irradiation
(Xe lamp, 300 W, λ > 420 nm). [Co] = Co(bpy)3Cl2. [Rh] = Rh(dtb-
bpy)3(PF6)3.
b
the catalysis was carried out using the colloidal platinum
catalyst generated in situ from K2PtCl4 under irradiation, both
systems of P1 and P3 are active for hydrogen production. In
the first hour, the system generated 23 μmol H2 while using P1,
after which the activity decreased significantly. After 5 h of
irradiation, hydrogen production ceased, yielding 35 μmol H2
in total. A dark solid particle was observed in the solution after
the photocatalytic experiment, which may be aggregation of the
colloidal platinum catalyst according to related reports. The
results indicated that the systems were effective for hydrogen
production through a reductive quenching pathway, which is
compatible with the observed substantial quenching behavior.
The chlorido-containing complexes, P1 and P3, are seemingly
neutral through the incorporation of negatively charged
chlorido groups into the coordination sphere, confirmed by
the ESI-MS data. However, the inclusion of a labile chloride
Figure 3. Photoinduced hydrogen production with different solvents at pH 8.0. Conditions: 0.33 mM [Rh(dtb-bpy)3](PF6)3, 0.19 M TEOA, 10 μM
P1 (A) or P3 (B), 100 mL solution (Xe lamp, 300 W, λ > 420 nm).
1959
dx.doi.org/10.1021/cs500296s | ACS Catal. 2014, 4, 1953−1963

ACS Catalysis
Research Article
which is sensitive to changes in the dielectric constant of the
solution. When using another polar solvent, acetonitrile, to
replace acetone, the activity diminished, with the system
yielding only 37 μmol H2 after 32 h irradiation for P1
(acetonitrile/water 4:1 v/v), which is a result of PS
decomposition promoted by the ligating ability of the solvent.20
When the photocatalytic reaction ceased, UV−vis spectra was
conducted on a sample of the reaction mixture in acetone/
water (9:1, v:v). The disappearance of the characteristic
absorbance attributed to P1 was observed compared with
that prior to photolysis, suggesting the decomposition of P1
occurs as hydrogen evolution. The evidence in regard to the
instability of PS will be further discussed in the sections below.
This finding was confirmed by the formation of a black solid
after the photolysis.25 However, when we performed the
experiments with the corresponding mononuclear compound
P3, the variation of the amount of water in the acetone solution
did not have an obvious impact on the photocatalytic activity
under the given conditions, as shown in Figure 3B. Similarly,
the system did not retain photocatalytic activity in the mixed
solvent solution of water/acetonitrile.
accounted for the photodegradation of the excited-state
sensitizer and thus the limited stability of the reaction system.8
We also investigated the effect of the concentration of
compounds P1 and P3 (1−20 μM) on the light-induced
hydrogen production while the concentrations of TEOA (0.19
M) and catalyst kept constant at 0.19 M and 0.33 mM,
respectively, in the mixed solvent solution of acetone/water
2+
(4:1 v/v, pH = 8.0). In the case of Co(bpy)3 as a catalyst
(Table 6), when the photolysis experiment was conducted at
Table 6. Photocatalytic H2-Producing Activity of Various
Concentrations of P1 or P3 in the Presence of Co(bpy)3Cl2
a
as a Catalyst
run
PS
concn (μM)
t (h)
H2 (μmol)
TON
1
2
3
4
5
6
7
8
P1
P1
P1
P1
P3
P3
P3
P3
1
5
72
72
72
72
32
32
32
32
87
66
1740
264
146
286
480
88
10
20
1
73
286
24
5
22
10
20
23
46
In a separate experiment, the stability of P1 during photolysis
was tested. At the end of a given hydrogen-producing cycle, the
hydrogen was pumped out. As shown in Figure 4, the addition
36
36
a
Photocatalytic conditions: a specified amount of PS, 0.33 mM
Co(bpy)3Cl2 and 0.19 M TEOA in acetone/water (4:1 v/v, 100 mL,
pH 8.0) upon irradiation (Xe lamp, 300 W, λ > 420 nm).
low concentrations of P1 or P3, there was little influence on the
overall evolution of hydrogen. The observation indicated that
the activity of such systems may be intrinsically limited by the
efficiency of the catalyst. Using 1 μM PS, a TON of 1740 based
on P1 can be achieved, whereas only 480 TON H2 was
obtained when P3 was empolyed under the same conditions.
Along with [Rh(dtb-bpy)3](PF6)3 as a catalyst in the mixed
solvent solution of acetone/water (9:1 v/v, pH = 8.0), as
shown in Figure 5, increasing the PS concentration increased
the total amount of hydrogen evolved for the system,
suggesting that the activity was limited by the PS concentration.
Using 1 μM PS, the system afforded a 3780 TON with respect
to P1 (1 μM), while the system based on P3 was still much less
efficient (TON = 1020) in the same conditions. When the
concentration of PS was increased to 20 μM, a total of 598 and
495 μmol H2 was produced for P1 and P3 in the catalytic
system, respectively. The apparent quantum efficiency for the
hydrogen production reaction was 1.02% and 0.45% at 420 nm
over an irradiation time of 24 h for P1 and P3, respectively (10
μM PS). Although it is known that molar absorptivity of the
light-harvesting complex is essential for good performance of
the solar-to-fuel conversion systems, there was little apparent
correlation between photocatalytic behaviors and the photo-
physical properties of the sensitizers.17,19 On the basis of the
photophysical and electrochemical studies of P1 and P3, molar
absorptivity of the complex appears to be beneficial for the
efficiency of the catalytic system presented here.
Figure 4. Hydrogen evolution vs time in an acetone/water (4:1 v/v)
solvent containing 10 μM P1, 0.33 mM Co(bpy)32+, and 0.19 M
TEOA (pH = 8, λ > 420 nm). Trace A: the first 72 h irradiation. Trace
B: further irradiation for 24 h after the injection of additional
Co(bpy)32+ (0.33 mM). Trace C: further irradiation for 72 h after the
injection of additional P1 (10 μM).
of an extra 1 equiv of catalyst failed to resume hydrogen
production under illumination; only 4.4 μmol H2 was obtained
after irradiation of 24 h. When P1 was readded (Figure 4, the
blue curve), the activity was recovered with 244 TON H2 after
an additional 72 h of irradiation. The presence of the extra
catalyst led to the increased performance of the system. The
results implied that the chromophore decomposition, rather
than catalyst deactivation, was the main reason for the cease of
hydrogen evolution after illumination. The electrospray
ionization-mass spectrometry (ESI-MS) analysis of the reaction
mixture after the photocatalytic reaction clearly showed that the
peak of complex P1 (m/z = 1445.06) had disappeared (see the
Supporting Information), indicating the occurrence of the
decomposition of P1 under the photocatalytic conditions,
which was consistent with the above experimental results. The
favored reductive quenching pathway in the present systems
CONCLUSIONS
■
In this work, the synthesis of a new type of dinuclear
iridium(III) complexes, the bridging ligands of which consist of
two bidentate N−N chelates, were described. These charge-
neutral complexes were thoroughly characterized, with the
results indicating that they exhibit more intense absorption in
visible light compared to their mononuclear analogues and have
sufficient reduction potential to transfer an electron onto the
1960
dx.doi.org/10.1021/cs500296s | ACS Catal. 2014, 4, 1953−1963

ACS Catalysis
Research Article
Figure 5. Photoinduced hydrogen production of various concentrations of P1 (A) or P3 (B) in acetone/water (9:1 v/v, 100 mL, pH 8.0).
Conditions: 0.33 mM [Rh(dtb-bpy)3](PF6)3 and 0.19 M TEOA (Xe lamp, 300 W, λ > 420 nm).
water reduction catalyst. The increase of absorption in the
visible region of these chromophores appears most suitable for
the application for light-induced photochemical reactions. The
properties of the chlorido-containing complexes in solutions
were different from others due to the presence of the labile
chloride ligand. All of the complexes, including dinuclear and
mononuclear compounds, were tested as photosensitizers for
hydrogen generation with the help of a series of water
reduction catalysts in photocatalytic experiments under pH = 8
conditions. The catalytic results indicated that enhancement of
the photocatalytic reduction of water into hydrogen in the
visible region was achieved by the dinuclear iridium complexes.
However, the catalytic performance based on these new
systems is far lower, in contrast to the previous reports. The
development of new dinuclear coordination compounds can be
anticipated to produce further superior results if additional
structural modification and screening are performed. The
results of this work, in connection with the dinuclear light-
harvesting complexes, provide new information concerning the
photophysical and structural requirements needed to build
supramolecular metal complexes that can be used as PSs for
artificial photosynthetic systems.
refined by the full-matrix least-squares method on F2 with
anisotropic thermal parameters for all non-hydrogen atoms.
Poor crystal structure refinements for P2 cannot be improved
due to relatively low-quality crystals. Density functional
calculations (B3LYP/LANL2DZ) were performed by using
Guassian 09. The numerical calculations in this paper were
performed on the IBM Blade cluster system at the High
Performance Computing Center (HPCC) of Nanjing Uni-
versity.
H2 Evolution Experiments. All H2 production experiments
were evaluated in a Pyrex vessel with a side visible light
irradiation using a 300 W xenon lamp equipped with a cutoff
filter (radiation wavelength >420 nm), as described previ-
ously.21,22 The reactor was charged with 10 μM PS, 0.33 mM
catalyst, and 0.19 M TEOA (2.5 mL) in an aqueous solution
(100 mL) with gentle magnetic stirring. The pH value was
adjusted to approximately 8.0 by the addition of concentrated
hydrochloric acid. Prior to light irradiation, the system was
evacuated successively before being backfilled with argon. The
evolved gas was periodically detected on a gas chromatograph
equipped with a thermally conductive detector (Shimadzu GC-
8A, argon as a carrier gas and MS-5A column). The apparent
quantum efficiency (QE) for the photocatalytic reaction was
measured under irradiation of monochromatized light at 420
nm.21
EXPERIMENTAL SECTION
■
General Measurements. The absorbance measurement
was performed using a Varian Cary 50 UV−vis spectropho-
tometer for solutions. Emission spectra and luminescence
quenching experiments were performed on a Varian Cary
Eclipse fluorescence spectrophotometer. The fluorescence
lifetime and the absolute quantum yield were acquired with a
Horiba Jobin-Yvon FluoroMax-4 TCSPC spectrometer. 1H
NMR spectra were recorded on a Bruker AMX-500 MHz
instrument. Mass spectra were obtained on a Thermo LCQ
Fleet ESI-mass spectrometer. The electrochemical properties of
these complexes in anhydrous acetonitrile were investigated in
a traditional three-electrode cell consisting of a glassy carbon
disk (3 mm diameter disk) working electrode, an auxiliary
platinum wire, and an Ag/AgCl (saturated KCl) reference
electrode using a PARSTAT-2273 advanced electrochemical
system. Diffraction intensity data for single crystals of
complexes P2 and P4, which were obtained by slow diffusion
of hexane into a CH2Cl2 solution, were collected at room
temperature on a Bruker Smart CCD diffractometer equipped
with graphite monochromated Mo Kα radiation (λ = 0.71073
Å). The structures were solved by the direct method and
Synthetic Procedure for Iridium Complexes. The
corresponding dichloro-bridged dimers were synthesized
using a known method.55 BiBzImH2 was also synthesized as
reported in the literature.56 Commercial chemicals were used as
supplied.
[Ir(tfdpyb)Cl]2(BiBzIm) (P1) and Ir(tfdpyb)(BiBzImH)Cl
(P3). A mixture of [Ir(tfdpyb)Cl)]2 (400 mg, 0.38 mmol), 2,2′-
bibenzimidazole (192 mg, 0.82 mmol), and Na2CO3 (396 mg,
3.8 mmol) in 30 mL of CH3OH/CH2Cl2 (1:1) was refluxed
under an atmosphere of nitrogen for 24 h. After cooling to
room temperature, the solvent was removed under reduced
pressure, and the crude product was purified by chromatog-
raphy on silica (hexane/EtOAc (1:1) to EtOAc) to afford the
products P1 (122 mg) and P3 (212 mg).
P1: 1H NMR (DMSO, 500 MHz): δ 8.46 (4H, d, J = 8.3), δ
8.27 (2H, s), δ 8.12 (4H, t, J = 7.7), δ 7.94 (4H, d, J = 5.3), δ
7.87 (2H, d, J = 8.0), δ 7.39 (4H, t, J = 6.5), δ 7.04 (2H, t, J =
7.5), δ 6.86 (2H, t, J = 7.8), δ 5.37 (2H, d, J = 8.4). ESI-MS
(CH3OH): m/z 1445.25 [M + Na]+; calcd for
C50H26N8F12Cl2NaIr2, 1445.06.
1961
dx.doi.org/10.1021/cs500296s | ACS Catal. 2014, 4, 1953−1963

ACS Catalysis
Research Article
P3: 1H NMR (DMSO, 500 MHz): δ 8.65 (1H, d, J = 8.2), δ
8.38 (2H, d, J = 8.4), δ 8.22 (1H, s), δ 8.15 (2H, d, J = 5.3), δ
8.05 (2H, t, J = 7.9), δ 7.84 (1H, d, J = 8.2), δ 7.54 (1H, t, J =
7.7), δ 7.46 (1H, t, J = 7.6), δ 7.41 (2H, t, J = 6.6), δ 7.35 (1H,
d, J = 8.1), δ 6.80 (1H, t, J = 7.5), δ 6.55 (1H, t, J = 7.7), δ 5.13
(1H, d, J = 8.4). ESI-MS (CH3OH): m/z 829.42 [M + H]+;
calcd for C32H19N6ClF6Ir, 829.08. HRMS (CH3OH): m/z
829.0921 [M + H]+; calcd for C32H19N6ClF6Ir, 829.0894.
[Ir(tfmppy)2]2BiBzIm (P2) and Ir(tfmppy)2(BiBzIm)(P4).
Similar to the above procedure, the reaction starting from
[Ir(tfmppy)2Cl]2 (300 mg, 0.25 mmol), 2,2′-bibenzimidazole
(129 mg, 0.55 mmol), and Na2CO3 (198 mg, 1.8 mmol) in 30
mL CH3OH/CH2Cl2 (1:1) gave the products P2 (82 mg) and
P5 (182 mg).
ACKNOWLEDGMENTS
■
This work was financially supported by the National Basic
Research Program of China (Grant No. 2013CB632400), the
National Science Foundation of China (Grant No. 20901038),
and the Fundamental Research Funds for the Central
Universities. We are also grateful to the Scientific Research
Foundation for the Returned Overseas Chinese Scholars, State
Education Ministry.
REFERENCES
■
(1) Frischmann, P. D.; Mahata, K.; Wurthner, F. Chem. Soc. Rev.
̈
2013, 42, 1847−1870.
(2) Sun, L.; Hammarstrom, L.; Åkermark, B.; Styring, S. Chem. Soc.
̈
P2: 1H NMR (DMSO, 500 MHz): δ 8.39 (4H, d, J = 8.2), δ
8.11 (4H, d, J = 8.2), δ 7.97 (4H, t, J = 7.8), δ 7.62 (4H, d, J =
5.7), δ 7.38 (4H, t, J = 8.0), δ 7.15 (4H, t, J = 6.6), δ 6.66 (4H,
m, J = 3.1), δ 6.61 (4H, s), δ 5.87 (4H, m, J = 3.1). ESI-MS
(CH3OH): m/z 1507.08 [M + H]+; calcd for C62H37N8F12Ir2,
1507.21.
Rev. 2001, 30, 36−49.
(3) Artero, V.; Chavarot-Kerlidou, M.; Fontecave, M. Angew. Chem.,
Int. Ed. 2011, 50, 7238−7266.
(4) Wang, M.; Na, Y.; Gorlov, M.; Sun, L. Dalton Trans. 2009, 6458−
6467.
(5) Krassen, H.; Ott, S.; Heberle, J. Phys. Chem. Chem. Phys. 2011, 13,
47−57.
(6) Fukuzumi, S.; Yamada, Y.; Suenobu, T.; Ohkubo, K.; Kotani, H.
Energy Environ. Sci. 2011, 4, 2754−2766.
(7) Luo, S. P.; Mejía, E.; Friedrich, A.; Pazidis, A.; Junge, H.; Surkus,
A. E.; Jackstell, R.; Denurra, S.; Gladiali, S.; Lochbrunner, S.; Beller, M.
Angew. Chem., Int. Ed. 2013, 52, 419−423.
(8) Han, Z.; McNamara, W. R.; Eum, M. S.; Holland, P. L.;
Eisenberg, R. Angew. Chem., Int. Ed. 2012, 51, 1667−1670.
(9) Jasimuddin, S.; Yamada, T.; Fukuju, K.; Otsuki, J.; Sakai, K. Chem.
Commun. 2010, 46, 8466−8468.
(10) Zhang, P.; Jacques, P. A.; Chavarot-Kerlidou, M.; Wang, M.;
Sun, L.; Fontecave, M.; Artero, V. Inorg. Chem. 2012, 51, 2115−2120.
(11) Zhang, P.; Wang, M.; Na, Y.; Li, X.; Jiang, Y.; Sun, L. Dalton
Trans. 2010, 39, 1204−1206.
(12) Li, X.; Wang, M.; Zheng, D.; Han, K.; Dong, J.; Sun, L. Energy
Environ. Sci. 2012, 5, 8220−8224.
(13) Wang, M.; Chen, L.; Sun, L. Energy Environ. Sci. 2012, 5, 6763−
6778.
P4: 1H NMR (DMSO, 500 MHz): δ 8.33 (2H, d, J = 8.2), δ
8.13 (2H, d, J = 8.2), δ 7.90 (2H, t, J = 7.4), δ 7.76 (2H, d, J =
5.5), δ 7.56 (2H, d, J = 8.1), δ 7.38 (2H, d, J = 7.9), δ 7.25 (2H,
t, J = 6.6), δ 7.08 (2H, t, J = 7.6), δ 6.78 (2H, t, J = 7.6), δ 6.61
(2H, s), δ 5.89 (2H, d, J = 8.3). ESI-MS (CH3OH): m/z 871.42
[M + H]+, calcd for C38H24N6F6Ir: m/z 871.15. HRMS
(CH3OH): m/z 871.1635 [M + H]+; calcd for C38H24N6F6Ir,
871.1597.
Ir(dpyx)(BiBzImH)Cl (P5). A mixture of [Ir(dpyx)Cl]2 (200
mg, 0.19 mmol), 2,2′-bibenzimidazole (98 mg, 0.42 mmol), and
AgOTf (261 mg, 0.88 mmol) in toluene (30 mL) was refluxed
under a nitrogen atmosphere for 24 h. After the precipitated
AgCl was filtered and washing with CH2Cl2, the solutions were
combined, and the solvent was removed under reduced
pressure. The residue was purified by chromatography on
silica (CH3OH/CH2Cl2 (1:10)) to give a yellow solid. The
solid can be purified by recrystallization from the solution (10:1
CH2Cl2/CH3OH), giving a yellow solid (112 mg).
(14) Wenger, O. S. Coord. Chem. Rev. 2009, 253, 1439−1457.
(15) Gartner, F.; Boddien, A.; Barsch, E.; Fumino, K.; Losse, S.;
̈
Junge, H.; Hollmann, D.; Bruckner, A.; Ludwig, R.; Beller, M. Chem.
̈
P5: 1H NMR (DMSO, 500 MHz): δ 8.68 (1H, d, J = 8.0), δ
8.14 (2H, d, J = 8.2), δ 7.88 (2H, d, J = 4.9), δ 7.80 (1H, d, J =
8.0), δ 7.75 (2H, t, J = 3.8), δ 7.50 (1H, t, J = 7.5), δ 7.43 (1H,
t, J = 7.4), δ 7.32 (1H, d, J = 7.9), δ 7.15 (1H, s), δ 7.10 (2H, t,
J = 6.3), δ 6.78 (1H, t, J = 7.3), δ 6.50 (1H, t, J = 7.5), δ
5.41(1H, d, J = 8.2), δ 2.86 (6H, s). ESI-MS (CH3OH): m/z
721.42 [M + H]+; calcd for C32H25N6ClIr, 721.14. HRMS
(CH3OH): m/z 721.2597 [M + H]+; calcd for C32H25N6ClIr,
721.1459.
Eur. J. 2011, 17, 6425−6436.
(16) Gartner, F.; Cozzula, D.; Losse, S.; Boddien, A.; Anilkumar, G.;
Junge, H.; Schulz, T.; Marquet, N.; Spannenberg, A.; Gladiali, S.;
̈
Beller, M. Chem.Eur. J. 2011, 17, 6998−7006.
(17) Gartner, F.; Denurra, S.; Losse, S.; Neubauer, A.; Boddien, A.;
̈
Gopinathan, A.; Spannenberg, A.; Junge, H.; Lochbrunner, S.; Blug,
M.; Hoch, S.; Busse, J.; Gladiali, S.; Beller, M. Chem.Eur. J. 2012, 18,
3220−3225.
(18) Goldsmith, J. I.; Hudson, W. R.; Lowry, M. S.; Anderson, T. H.;
Bernhard, S. J. Am. Chem. Soc. 2005, 127, 7502−7510.
(19) Brooks, A. C.; Basore, K.; Bernhard, S. Inorg. Chem. 2013, 52,
5794−5800.
ASSOCIATED CONTENT
■
(20) Tinker, L. L.; Bernhard, S. Inorg. Chem. 2009, 48, 10507−10511.
(21) Yuan, Y. J.; Yu, Z. T.; Chen, X. Y.; Zhang, J. Y.; Zou, Z. G.
Chem.Eur. J. 2011, 17, 12891−12895.
S
* Supporting Information
Full experimental details and additional characterization data
including emission decay kinetics, CVs, and Stern−Volmer plot
for emission quenching for all complexes in this work. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(22) Yuan, Y. J.; Zhang, J. Y.; Yu, Z. T.; Feng, J. Y.; Luo, W. J.; Ye, J.
H.; Zou, Z. G. Inorg. Chem. 2012, 51, 4123−4133.
(23) Yu, Z. T.; Yuan, Y. J.; Cai, J. G.; Zou, Z. G. Chem.Eur. J. 2013,
19, 1303−1310.
(24) You, Y.; Nam, W. Chem. Soc. Rev. 2012, 41, 7061−7084.
(25) Tinker, L. J.; McDaniel, N. D.; Curtin, P. N.; Smith, C. K.;
Ireland, M. J.; Bernhard, S. Chem.Eur. J. 2007, 13, 8726−8732.
(26) Metz, S.; Bernhard, S. Chem. Commun. 2010, 46, 7551−7553.
(27) DiSalle, B. F.; Bernhard, S. J. Am. Chem. Soc. 2011, 133, 11819−
11821.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: yuzt@nju.edu.cn.
Notes
(28) Hansen, S.; Pohl, M. M.; Klahn, M.; Spannenberg, A.; Beweries,
T. ChemSusChem 2013, 6, 92−101.
The authors declare no competing financial interest.
1962
dx.doi.org/10.1021/cs500296s | ACS Catal. 2014, 4, 1953−1963

ACS Catalysis
Research Article
(29) Yuan, Y. J.; Yu, Z. T.; Gao, H. L.; Zou, Z. G.; Zheng, C.; Huang,
W. Chem.Eur. J. 2013, 19, 6340−6349.
(30) Yuan, Y. J.; Yu, Z. T.; Cai, J. G.; Zheng, C.; Huang, W.; Zou, Z.
G. ChemSusChem 2013, 6, 1357−1365.
(31) Donato, L.; McCusker, C. E.; Castellano, F. N.; Zysman-
Colman, E. Inorg. Chem. 2013, 52, 8495−8504.
(32) Auffrant, A.; Barbieri, A.; Barigelletti, F.; Lacour, J.; Mobian, P.;
Collin, J.-P.; Sauvage, J.-P.; Ventura, B. Inorg. Chem. 2007, 46, 6911−
6919.
(33) Auffrant, A.; Barbieri, A.; Barigelletti, F.; Collin, J.-P.; Flamigni,
L.; Sabatini, C.; Sauvage, J.-P. Inorg. Chem. 2006, 45, 10990−10997.
(34) Welter, S.; Lafolet, F.; Cecchetto, E.; Vergeer, F.; De Cola, L.
ChemPhysChem 2005, 6, 2417−2427.
(35) Plummer, E. A.; Hofstraat, J. W.; De Cola, L. Dalton Trans.
2003, 2080−2084.
(36) Indelli, M. T.; Bura, T.; Ziessel, R. Inorg. Chem. 2013, 52, 2918−
2926.
́
(37) Andreiadis, E. S.; Imbert, D.; Pecaut, J.; Calborean, A.; Ciofini,
I.; Adamo, C.; Demadrille, R.; Mazzanti, M. Inorg. Chem. 2011, 50,
8197−8206.
(38) Barcina, J. O.; Herrero-García, N.; Cucinotta, F.; De Cola, L.;
Contreras-Carballada, P.; Williams, R. M.; Guerrero-Martínez, A.
Chem.Eur. J. 2010, 16, 6033−6040.
(39) McCusker, C. E.; Hablot, D.; Ziessel, R.; Castellano, F. N. Inorg.
Chem. 2012, 51, 7957−7959.
́ ́
(40) Costa, R. D.; Fernandez, G.; Sanchez, L.; Martín, N.; Ortí, E.;
Bolink, H. J. Chem.Eur. J. 2010, 16, 9855−9863.
(41) Tsuboyama, A.; Takiguchi, T.; Okada, S.; Osawa, M.; Hoshino,
M.; Ueno, K. Dalton Trans. 2004, 1115−1116.
(42) Whittle, V. L.; Williams, J. A. G. Inorg. Chem. 2008, 47, 6596−
6607.
(43) Hags, M.; Matsumura-Inoue, T.; Yamabe, S. Inorg. Chem. 1987,
26, 4148−4154.
(44) Rillema, D. P.; Sahai, R.; Matthews, P.; Edwards, A. K.; Shaver,
R. J.; Morgan, L. Inorg. Chem. 1990, 29, 167−175.
(45) Haga, M.; Bond, A. M. Inorg. Chem. 1991, 30, 475−480.
(46) Yin, J.; Elsenbaumer, R. L. Inorg. Chem. 2007, 46, 6891−6901.
(47) Brulatti, P.; Gildea, R. J.; Howard, J. A. K.; Fattori, V.; Cocchi,
M.; Williams, J. A. G. Inorg. Chem. 2012, 51, 3813−3826.
(48) Zhu, Y. C.; Zhou, L.; Li, H. Y.; Xu, Q. L.; Teng, M. Y.; Zheng, Y.
X.; Zuo, J. L.; Zhang, H. J.; You, X. Z. Adv. Mater. 2011, 23, 4041−
4046.
(49) Wilkinson, A. J.; Puschmann, H.; Howard, J. A. K.; Foster, C. E.;
Williams, J. A. G. Inorg. Chem. 2006, 45, 8685−8699.
(50) Wilkinson, A. J.; Goeta, A. E.; Foster, C. E.; Williams, J. A. G.
Inorg. Chem. 2004, 43, 6513−6515.
(51) Zhang, P.; Wang, M.; Dong, J.; Li, X.; Wang, F.; Wu, L.; Sun, L.
J. Phys. Chem. C 2010, 114, 15868−15874.
(52) Marinescu, S. C.; Winkler, J. R.; Gray, H. B. Proc. Natl. Acad. Sci.
U.S.A. 2012, 109, 15127−15131.
(53) Whang, D. R.; Sakai, K.; Park, S. Y. Angew. Chem., Int. Ed. 2013,
52, 11612−11615.
(54) Pandrala, M.; Li, F.; Feterl, M.; Mulyana, Y.; Warner, J. M.;
Wallace, L.; Keene, F. R.; Collins, J. G. Dalton Trans. 2013, 42, 4686−
4694.
(55) Nonoyama, M. Bull. Chem. Soc. Jpn. 1974, 47, 767−768.
(56) Huang, W. K.; Cheng, C. W.; Chang, S. M.; Lee, Y. P.; Diau, E.
W. G. Chem. Commun. 2010, 46, 8992−8994.
1963
dx.doi.org/10.1021/cs500296s | ACS Catal. 2014, 4, 1953−1963