Synthesis and catalytic water oxidation activities of ruthenium complexes containing neutral ligands

Article abstract of DOI:10.1002/chem.201100274

Two dinuclear and one mononuclear ruthenium complexes containing neutral polypyridyl ligands have been synthesised as pre-water oxidation catalysts and characterised by ¹H and ¹³C NMR spectroscopy and ESI-MS. Their catalytic water oxidation properties in the presence of [Ce(NH ₄)₂(NO₃)₆] (Ce^IV) as oxidant at pH 1.0 have been investigated. At low concentrations of Ce ^IV (5 mM), high turnover numbers of up to 4500 have been achieved. An ¹⁸O-labelling experiment established that both O atoms in the evolved O₂ originate from water. Combined electrochemical study and electrospray ionisation mass spectrometric analysis suggest that ligand exchange between coordinated 4-picoline and free water produces Ru aquo species as the real water oxidation catalysts. Copyright

Full text of DOI:10.1002/chem.201100274

DOI: 10.1002/chem.201100274
Synthesis and Catalytic Water Oxidation Activities of Ruthenium Complexes
Containing Neutral Ligands
[
a]
[a]
[b]
[a]
[c]
Yunhua Xu, Lele Duan, Torbjçrn ꢀkermark, Lianpeng Tong, Bao-Lin Lee,
[
d]
[c]
[a, d]
Rong Zhang, Bjçrn ꢀkermark, and Licheng Sun*
Abstract: Two dinuclear and one
mononuclear ruthenium complexes
containing neutral polypyridyl ligands
have been synthesised as pre-water oxi-
dation catalysts and characterised by
pH 1.0 have been investigated. At low
concentrations of Ce (5 mm), high
ment established that both O atoms in
the evolved O2 originate from water.
Combined electrochemical study and
electrospray ionisation mass spectro-
metric analysis suggest that ligand ex-
change between coordinated 4-picoline
and free water produces Ru aquo spe-
cies as the real water oxidation cata-
lysts.
IV
turnover numbers of up to 4500 have
1
8
been achieved. An O-labelling experi-
1
13
H and C NMR spectroscopy and
ESI-MS. Their catalytic water oxida-
Keywords: homogeneous catalysis ·
isotopic labeling · neutral ligands ·
ruthenium · water oxidation
tion properties in the presence of [Ce-
IV
ACHTUNGTRENNUNG( NH ) ACHTUNGTRENNUNG( NO ) ] (Ce ) as oxidant at
4 2 3 6
Introduction
mainly been focused either on manganese-based com-
[3–7]
plexes,
um-based complexes,
(Co, Ir, Fe) have also been investigated.
which mimics natureꢀs preference, or on rutheni-
[4–24]
As global energy demand continues to increase, the devel-
opment of sustainable and environmentally friendly energy
resources has become a big challenge for scientists. Solar
energy is believed to be one of the most promising ways to
solve this problem. Through light-driven water splitting,
for example, solar energy can be converted into dihydro-
gen—a clean and valuable fuel. The most challenging step
in this artificial photosynthesis process has been found to be
water oxidation, now intensely studied by many research
groups. Considerable efforts have been made to develop
synthetic analogues of the oxygen-evolving complex (OEC)
although other metal complexes
[25–29]
Although only
a very few of the manganese complexes that have been syn-
thesised have shown moderate to low catalytic activity for
[
1]
[30–34]
water oxidation,
many ruthenium complexes with high
[4–24]
catalytic activities have been developed.
As early as the
1980s, Meyer and co-workers reported a few dinuclear
ruthenium complexes (such as the “blue dimer” A,
[35]
Scheme 1) as water oxidation catalysts,
in 2004 Llobet
and co-workers described a new type of dinuclear complex
(B, Scheme 1) that contains no Ru-O-Ru motif but is still
[11]
in photosystem II (PSII), which contains a Mn Ca cluster as
active for water oxidation, whereas in 2005 Thummel and
co-workers presented a series of dinuclear ruthenium com-
plexes (such as C, Scheme 1) capable of catalysing water ox-
4
[2]
the water oxidation catalyst. These model systems have
[12]
[
a] Dr. Y. Xu, L. Duan, L. Tong, Prof. L. Sun
Department of Chemistry
School of Chemical Science and Engineering
Royal Institute of Technology (KTH)
idation. In 2008, Sartorel and Hill independently reported
a tetraruthenium-oxo oxygen-evolving catalyst containing
[16,17]
all-inorganic POM ligands.
Several research groups have recently also reported that
mononuclear ruthenium complexes can efficiently catalyse
1
00 44 Stockholm (Sweden)
Fax : (+46)8-791-2333
E-mail: lichengs@kth.se
[14,36]
water oxidation.
Negatively charged ligands can lower
[
b] Dr. T. ꢁkermark
the oxidation potentials of metal complexes and thus stabi-
[37]
Department of Materials and Environmental Chemistry
Arrhenius Laboratory, Stockholm University,
lise the high oxidation states of the metal ions, and so we
have employed carboxylate ligands to synthesise highly effi-
cient water oxidation catalysts. We have previously reported
1
06 91 Stockholm (Sweden)
[
c] B.-L. Lee, Prof. B. ꢁkermark
Department of Organic Chemistry, Arrhenius Laboratory
Stockholm University, 106 91 Stockholm (Sweden)
[
20–22]
dinuclear (D and E, Figure 1)
nium complexes,
and mononuclear ruthe-
[
18,19,23]
which indeed promise high activities.
[
d] Dr. R. Zhang, Prof. L. Sun
In addition to manganese- and ruthenium-based complexes,
in 2008 Bernhard and co-workers reported iridium-based
State Key Laboratory of Fine Chemicals
DUT-KTH Joint Education
and Research Centre on Molecular Devices
Dalian University of Technology (DUT)
Dalian 11602 (P.R. China)
aquo complexes that can also catalyse homogeneous water
IV
oxidation in the presence of Ce
dant with turnover numbers (TONs) of over 2000.
ACHTUNGTNERUNNG( NH ) ACHTNUGTERNNUN(G NO ) (Ce ) as oxi-
4 2 3 6
[27]
Although negatively charged carboxylate ligands can use-
fully improve the catalytic properties of metal complexes,
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201100274.
9520
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 9520 – 9528

FULL PAPER
1
–3, Scheme 2) containing neu-
tral polypyridyl ligands and ex-
plore their catalytic activities in
IV
the presence of Ce as sacrifi-
cial electron acceptor. As ex-
pected, we gained enhanced
catalytic activity relative to the
complexes C.
Results and Discussion
Synthesis and characterisation:
The ligand L1 was prepared by
the route shown in Scheme 3,
from 4,4’-dimethyl-2,2’-bipyri-
dine (4) as the starting material.
The compound 4 was oxidised
with
meta-chloroperbenzoic
acid (m-CPBA) to form the N-
oxide 5, which was further
treated with trimethylsilyl cya-
nide in the presence of carba-
moyl chloride to yield 6-cyano-
Scheme 1. Structures of some reported Ru-based water oxidation catalysts.
[38]
they have some drawbacks. When the carboxylate ligands
are coordinated with manganese, for example, insoluble
oligomers or polymers are easily formed, giving rise to seri-
ous problems for homogeneous water oxidation. In this re-
spect, improvement of the catalytic activities of metal com-
plexes containing neutral ligands is of interest as well. In the
case of the complexes C, electron-donating groups at the 4-
pyridyl positions have been shown to enhance the catalytic
2,2’-bipyridine (6). This was subsequently treated with hy-
drazine in ethanol to form the dihydro base 7, which was
further oxidised with nitric acid in acetic acid to afford the
[
39]
tetrazine 8 as a pink solid. Bubbling of acetylene through
a solution of 8 in DMF at reflux provided the bis-tridentate
ligand 3,6-bis-(4,4’-dimethyl-2,2’-bipyrid-6-yl)pyridazine (L1)
[39]
as a white solid.
The synthesis of L2 is shown in Scheme 4. The compound
11 could be obtained in a good yield by a procedure similar
to that described above. A reaction between 11 and cyclo-
pentanone at 1108C in toluene in the presence of morpho-
line and silica gel afforded the ligand L2 in a quantitative
[12]
activities. To enhance the electron-donating ability of its
polypyridyl bridging ligand, we modified Thummelꢀs ligands
(
C) by removing the uncoordinated pyridyl motifs and intro-
ducing alkyl groups on the equatorial polypyridyl ligand.
Here we present several ruthenium complexes (complexes
[40]
yield.
The syntheses of the com-
plexes 1 and 2 were carried out
by treatment of the ligands with
cis-Ru
followed by addition of excess
-methylpyridine in the pres-
ence of triethylamine and LiCl
see also Schemes 3 and 4). Ion
ACHTUNGTNERNUNG( dmso) Cl in ethanol,
4 2
4
(
exchange with NH PF6 gave
4
precipitates, which were puri-
fied by column chromatography
to afford the desired products
in 35–45% yields.
The complex 3 was prepared
by treatment of the ligand L2
with cis-[Ru ACHTUNGTENRUNG( dmso) Cl ] in etha-
4 2
nol to give a black precipitate.
This precipitate was shown to
be al-
[Ru(L2)
A
H
U
G
R
N
U
G
2
Scheme 2. Molecular structures of the complexes 1–3 and the ligands L1 and L2.
though the L2/Ru AHCNUTGTERGUN(NN dmso) Cl
4
2
Chem. Eur. J. 2011, 17, 9520 – 9528
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9521

L. Sun et al.
tures of the complexes 1–3
were also confirmed by ESI-MS
1
13
data. All H and C NMR and
mass spectral data are consis-
tent with both 1 and 2 having
symmetric structures, each con-
taining a m-Cl bridge, similar to
the dinuclear ruthenium com-
plexes reported by Thummel
[12]
and co-workers.
In the mass
spectrum of 1, the peak corre-
3
+
sponding to [Mꢀ3PF ]
was
6
found at m/z=351.7380 (calcd:
3
51.7384). For the complex 2,
3
+
peaks
for
[Mꢀ3PF ]
,
6
2
+
+
[
Mꢀ2PF ]
and [MꢀPF ]
6
6
were
observed
at
m/z=346.16,
592.06
and
1
328.93, respectively.
Scheme 3. Synthesis of the complex 1.
Absorption spectroscopy: UV/
Vis spectra of the complexes 1–
3
in acetonitrile are shown in
Figure 1 and the electronic absorption data are summarised
in Table 1.
The complex 1 has an electronic spectrum similar to those
[12]
reported by Thummel for the complexes C.
region, there are three absorption maxima at 275, 310 and
45 nm assigned to ligand-centred (LC) p–p* transitions. In
In the UV
3
the visible region, three relatively weak absorption maxima
were observed at 437, 483 and 567 nm. They are assigned to
metal-to-ligand charge transfer (MLCT) transitions involv-
ing electron promotions from the Ru d orbitals to the ligand
Scheme 4. Syntheses of the complexes 2 and 3.
ratio in the reaction mixture had been 1:2. This intermediate
was further heated at reflux with excess 4-picoline in an eth-
anol/water mixture (9:1), followed by purification to afford
the complex 3 in 52% yield.
All complexes and compounds except for the tetrazines 8
and 11 were characterised by NMR spectroscopy. The struc-
Figure 1. UV/Vis spectra of the complexes 1–3 in acetonitrile.
Table 1. Electronic absorption and electrochemical data for the complexes 1–3.
4
ꢀ1
ꢀ1 [a]
[b]
[c]
Complex
l
max [nm (e/10 m cm )]
E
p
1/2 [V] (DE [mV])
1
2
3
275 (4.94), 310 (4.98), 345 (4.16), 437 (1.14), 483 (1.09), 567 (0.84)
276 (3.35), 315 (4.47), 345 (4.00), 436 (1.16), 479 (1.23), 639 (0.67)
278 (3.75), 327 (4.85), 400 (0.61), 463 (0.54), 536 (0.46)
ꢀ0.98 (57), ꢀ0.47 (59), 1.46 (61), 1.89 (47)
ꢀ0.96 (80), ꢀ0.50 (92), 1.44 (93), 1.95 (80)
ꢀ0.85 (78), 1.46 (92)
ꢀ
1
[
a] In CH
3
CN solution. [b] Versus NHE. [c] n=0.1 Vs ; DE
p
=EpaꢀEpc
.
9522
www.chemeurj.org
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 9520 – 9528

Synthesis and Catalytic Water Oxidation Activities of Ru Complexes
FULL PAPER
p* orbitals. The complex 2 shows similar absorption to the
complex 1 in the MLCT bands, but has a much smaller ex-
tinction coefficient than 1 in the 250–300 nm region. In the
UV region the monomeric complex 3 displays absorption
strength similar to that of the complex 2 but much weaker
absorption than the complexes 1 and 2 in the visible region.
Electrochemistry: The electrochemistry of the complexes 1–
3
was studied by cyclic voltammetry (CV) and differential
pulse voltammetry (DPV). The electrochemical data are
also given in Table 1. All potentials reported here are versus
NHE.
In acetonitrile: The CV of 1 (Figure 2a) in dry acetonitrile
has two reversible one-electron oxidation waves at E1/2
=
1
.46 and 1.89 V, which is similar to the values reported for
[12]
the related complex C (Figure 2, R=CH ). They are as-
signed to the oxidations of the Ru
to Ru2
3
II,II
II,III
II,III
to Ru2
and Ru2
2
III,III
, respectively. In addition to these two oxidation
peaks, two reversible one-electron reduction waves were
also observed at E1/2 =ꢀ0.47 and ꢀ0.98 V. These are as-
signed to the ligand-based reductions of the complex. Addi-
tional ligand-based reduction could also be observed at
E =ꢀ1.60 V (DPV peak potential) as shown in Figure 2b.
pk
II,II
2
II,III
For the complex 2, the oxidation of the Ru
to Ru
2
occurs at E1/2 =1.44 V (Figure 3a), 20 mV lower than in the
II,III
III,III
ꢀ1
case of 1, whereas the oxidation of Ru2
to Ru2
is at
Figure 3. CVs (u=0.1 Vs ) a) of 2 (1 mm), and b) of 3 (1 mm) in dry ace-
tonitrile with Bu NPF (0.1m) as the supporting electrolyte, glassy carbon
disk as the working electrode and Ag/AgNO as the reference electrode.
4
6
3
E1/2 =1.95 V, 60 mV higher than in the case of 1. The ligand-
based reductions of 2 were observed at E1/2 =ꢀ0.50 and
ꢀ
0.96 V (see also Figure 3a). The complex 3 shows that a
III/II
Ru
redox couple arises at E1/2 =1.46 V and a ligand-based
reduction at E1/2 =ꢀ0.85 V (Figure 3b).
Notably, both the first and the second oxidation potentials
of the dimeric complexes 1 and 2 are about 0.5 V higher
than the corresponding ones of the complex E, which has a
[22]
similar structure but contains negatively charged ligands.
The electrochemical behaviour of the dinuclear ruthenium
complexes in the presence of a small amount of water was
also studied. When water (3%, v/v) was added to the aceto-
nitrile solution of 1, the CV (Figure 4, solid curve) in the po-
tential range below 2.0 V displayed well-defined reversible
redox waves similar to those seen in the dry acetonitrile.
Above about 2.2 V, a potential slightly higher than the oxi-
II,III
III,III
dation of Ru2
to Ru2
, however, a very large irreversi-
ble oxidation peak was observed, clearly indicating catalytic
water oxidation by the complex 1.
In aqueous solution (pH 1): In aqueous CF SO H (pH 1.0),
3
3
only one redox couple for each complex was observed in the
0
1
.15–1.65 V region (Figure 5). The redox couples at E1 =
/2
III,II/II,II
.23, 1.21 are related to Ru
in 1 and 2, respectively.
ꢀ
1
These values are higher than those for the complexes D and
Figure 2. a) CV (u=0.1 Vs ) and b) DPV of 1 (1 mm) in dry acetonitrile
with Bu NPF (0.1m) as the supporting electrolyte, glassy carbon disk as
[20,22]
III/II
E.
1.23 V.
The Ru
redox couple of 3 was detected at E1/2
=
4
6
3
the working electrode and Ag/AgNO as the reference electrode.
Chem. Eur. J. 2011, 17, 9520 – 9528
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9523

L. Sun et al.
ꢀ1
IV
Figure 4. CV (u=0.1 Vs ) of 1 (1 mm) in an acetonitrile/water (100:3)
mixture with Bu NPF (0.1m) as the supporting electrolyte, glassy carbon
disk as the working electrode and Ag/AgNO as the reference electrode.
Figure 6. Kinetics of oxygen evolution induced by Ce (330 mm) in aque-
ous CF SO H (pH 1.0, 3 mL) in the presence of catalyst (13.3 mm). The
curves were measured with an O -sensor and calibrated by GC.
4
6
3
3
3
2
The TONs of 1 and 2 are more than twice as high as
[13]
those of Thummelꢀs dimeric complexes (TONs<700),
showing that they are indeed more efficient catalysts than
the complexes C. There could be two reasons for this cata-
lytic efficiency difference. Firstly, the uncomplexed nitrogen
in the naphthyridine moieties of Thummelꢀs complexes C
(
see Scheme 1) was intended to provide hydrogen bonding
[12]
with water. Under pH 1 conditions, however, this uncom-
plexed nitrogen is probably protonated and so would
become an electron-withdrawing group, which would de-
crease the catalytic activities of the complexes C in a similar
way as the electron-withdrawing group (CF ) at the axial po-
3
sition of pyridines, as observed by Thummel and co-work-
[12]
ers. Secondly, the electron-donating alkyl groups in the li-
gands L1 and L2 can facilitate oxidation processes and thus
enhance the catalytic performance of complexes 1 and 2.
On the other hand, the TONs and TOFs of 1 and 2 are
ꢀ1
Figure 5. CVs (u=0.1 Vs ) of the complexes 1 (saturated), 2 (saturated)
and 3 (1 mm) in aqueous CF SO H (pH 1.0) with glassy carbon disk as
3
3
the working electrode and Ag/AgCl as the reference electrode.
[22]
only 10 to 40% of those of the complex E, showing that
the complexes 1 and 2 are much less stable and less active
than the complex E. The improvement in the catalytic activ-
ity of the metal complexes achieved by introduction of alkyl
groups is therefore far less than that achieved by use of neg-
atively charged ligands. This is because the anionic ligands
are more efficient electron-donating groups, capable of re-
ducing the charge accumulation on the metal complexes
generally and of lowering the oxidation potentials of the
metal ions, and thus of stabilising the high oxidation states
of the metal ions and providing a low-energy pathway for
oxidising water.
Catalytic activity in water oxidation: The catalytic activities
of the complexes 1–3 in water oxidation were investigated
IV
with use of Ce as oxidant in aqueous CF SO H (pH 1.0)
3
3
solutions [Eq. (1)].
IV
catalyst
III
4
Ce þ 2H O ƒƒƒ! 4Ce þ O þ 4Hþ
ð1Þ
2
2
Evolved oxygen in the headspace of the reaction flask
was measured by gas chromatography (GC) and the reac-
tion kinetics were monitored with an oxygen sensor.
Figure 6 shows the kinetics of oxygen evolution catalysed by
the complexes 1–3. When a solution of the complex 1 or 2
When the complex 3 was used as catalyst, unlike 1 and 2,
it needed a long induction period (ca. 3 h) to produce de-
tectable amount of oxygen (see also Figure 6). After a reac-
tion time of 22 h, a TON of 568 was obtained for 3. This
result indicates that the complex 3 is not a true water oxida-
tion catalyst and that ligand exchange between the 4-meth-
ylpyridyl ligand and free water molecules occurs to form the
real water oxidation catalyst. This phenomenon has also
in acetonitrile was added to a deaerated water solution
IV
(
pH 1.0, adjusted with CF SO H) containing Ce (330 mm),
3 3
dioxygen gas was produced immediately. In the absence of
catalysts, no oxygen evolution was found under otherwise
the same conditions. The initial turnover frequencies
ꢀ
1
ꢀ1
(
TOFs) are 0.09 s for 1 and 0.10 s for 2. After 22 h,
2
+
TONs of 1100 and 1400 were obtained for 1 and 2, respec-
been observed in the case of [Ru
ACHTUNGTNERNUNG( tpy) ACHTUGNTRENNUNG( pic) ] (tpy=terpyri-
3
tively.
dine; pic=4-methylpyridine), which has a structural similari-
9524
www.chemeurj.org
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 9520 – 9528

Synthesis and Catalytic Water Oxidation Activities of Ru Complexes
FULL PAPER
[14]
2+
ty to 3. Most probably, complex 3 and [Ru AHCUTNGREUNNG( tpy) HCATUNGTRNNEUN(G pic) ]
3
have similar reaction mechanisms.
Previously, the true catalyst corresponding to [RuCAHTUNGTRENNUNG
2
+
2+
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(pic) ]
(trans-[Ru ACHTUNTGRENNUNG( tpy) ACHNUTGERTNNUNG( pic) AHCTGNRUETNNUG( OH )] )
2 2
3
2
+
evolution is first order in trans-[Ru
plying that a mononuclear catalytic pathway is involved.
In the light of our findings and of fruitful mechanistic stud-
ACHTUNGTRENNUN(G tpy) ACHTUNGETRUNNNG( pic) ACHTGRUNETNUN(G OH )] , im-
2 2
[14]
2
+
[8,36]
ies on [Ru
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(tpy)
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(bpy)
A
T
G
R
N
N
(OH )]
(bpy=bipyridine),
we
2
propose that 3 needs ligand exchange between coordinated
pic and free water to generate the Ru-aqua complex, which
in turn catalyses water oxidation in the manner of water nu-
cleophilic attack on the Ru=O species.
16/16
16/18
Figure 8.
O
2
(&) and
O
2
(&) evolution recorded by mass spectrosco-
In our previous work we have demonstrated that low con-
centrations of Ce benefit the catalytic activities of the two
complexes D and E, very probably because the ligand de-
py. An Ar-degassed solution of 1 in acetonitrile (0.15 mL) was injected
IV
I
V
1
8
into a mixture of Ce (100 mg), CF
3
SO
3 2
H (5 mL) and O-enriched H O
1
8
(
0.3 mL, 10% O) (under Ar, ꢁ75 mbar).
composition is significantly reduced in the presence of a low
IV [22]
concentration of Ce . Here, we also conducted the cata-
1
8
IV
lytic reaction in the presence of a relatively low concentra-
riched H O (10% O, 0.3 mL), CF SO H (5 mL) and Ce
2
3
3
IV
18/18
tion of Ce (5 mm). As expected, the TONs for all three
(100 mg). (The trace for
actly, due to the presence of Ar.) The observed ratio of
O could not be determined ex-
2
3
6
complexes were increased dramatically: TONs of 3500, 4500
and 2000 were achieved for 1, 2 and 3, respectively
evolved O with masses 34 and 32 was found to be 0.23,
2
(
Figure 7). These data imply that the performances of our
which is very close to the theoretical ratio (2:9=0.22) based
on the assumption that both O atoms of O come from H O
2
2
1
8
containing 10% OH . This ratio did not change with time,
2
so oxygen exchange between the oxidant and water can be
ruled out. The conclusion that can be drawn from this result
is that water is the only source of the evolved oxygen in
water oxidation catalysed by 1.
ESI-MS analysis: ESI-MS has been used to probe reaction
[36b]
intermediates under catalytic conditions.
We conducted
ESI-MS measurements to isolate any intermediate related
IV
to the 2·Ce catalytic system in situ. Figure 9 and Figure S1
in the Supporting Information show the mass spectra of
IV
identified species after addition of Ce (200 equiv) to aque-
ous complex 2. We observed the singly charged species
+
[
[
Mꢀ3PF +2CF SO ]
at 1337.10 (calcd: 1337.09),
IV
6
3
3
Figure 7. Kinetics of oxygen evolution induced by Ce (5 mm) in aqueous
+
Mꢀ3PF +O+2CF SO ] at 1353.13 (calcd: 1353.08) and
6
3
3
CF
3
SO
3
H (pH 1.0, 30 mL) in the presence of the catalyst (0.2 mm). The
-sensor and calibrated by GC.
+
curves were measured with an O
2
[Mꢀ3PF +2O+2CF SO ] at 1369.05 (calcd: 1369.08), as
6
3
3
well
[
as
the
doubly
charged
species
2
+
Mꢀ3PF ꢀpic+CF SO ] at 547.59 (calculated: 547.54) and
6
3
3
2+
catalysts might be further improved under milder catalytic
conditions, such as in an electrolyser. Immobilisation of
these catalysts might therefore be an attractive way to pro-
duce active water oxidation electrodes based on molecular
catalysts.
[Mꢀ3PF ꢀ2pic+CF SO ]
at 501.05 (calculated: 501.01),
6
3
3
where M refers to the formula of complex 2. We note that a
few MS peaks are unidentifiable. After addition of ascorbic
acid (80 equiv) to the above solution, the MS peaks for
+
+
[Mꢀ3PF +O+2CF SO ]
and [Mꢀ3PF +2O+2CF SO ]
6
3
3
6
3
3
still remain, revealing that these two species are ligand-oxi-
dised products rather than seven-coordinate Ru oxo species.
From these observations, we propose ligand-exchange
18
O-Isotopic labelling studies: An essential question in water
oxidation is whether or not the oxygen atoms in the evolved
1
8
O originate from the water. The use of O-labelled water
species such as Ru OH and Ru ACHNUTGTRNNEG(U OH ) (Scheme 5) as the
2 2 2 2 2
true water oxidation catalysts for the 2·Ce catalytic system.
2
IV
together with mass spectroscopy enabled us to determine
the source of the O atoms in the evolved O , by evaluation
The aquo ligand could lower the oxidation potentials of the
2
of the ratio between the oxygen isotopes in the evolved O .
metal centre through proton-coupled electron transfer
2
1
6/16
IV
Figure 8 shows the traces of evolved O with m/z 32 ( O2)
(PCET), potentially allowing Ce to drive Ru OH2 and
2
2
16/18
and 34 ( O ) on addition of a solution of 1 (0.15 mL) in
Ru2
ACHTUNGTNERNGNU( OH ) to oxidise water. At this stage, however, we
2 2
2
acetonitrile to an aqueous solution containing isotope-en-
could not exclude the possibility that water coordinates to
Chem. Eur. J. 2011, 17, 9520 – 9528
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9525

L. Sun et al.
change is necessary for 1–3 to form the true real water oxi-
dation catalysts. Ligand modification demonstrated that the
catalytic activities could be promoted by introducing elec-
tron-donating functional groups on the neutral ligands. Al-
though these complexes exhibit relatively high turnover
numbers, they are much less efficient catalysts for water oxi-
dation than the complexes with negatively charged ligands.
These results, together with our previous investigations,
could provide useful insights in the design of efficient water
oxidation catalysts by ligand modification.
Experimental Section
Materials: All chemicals were purchased from Sigma–Aldrich or Lancas-
ter and used as received. All solvents were dried by standard methods
[38]
when needed. Compound 9 was prepared by the literature method.
NMR spectroscopy: 1D and 2D NMR spectra were recorded with a
Bruker Avance 400/500 MHz spectrometer.
Mass spectrometry: Mass spectrometry measurements were performed
either with a Q-Tof Micro mass spectrometer (HRMS, Micromass, Man-
chester, UK) or with a LCQ ADVANTAGE MAX (Finnigan) mass spec-
IV
trometer. The 2·Ce catalytic system was analysed by Q-Tof Micro mass
IV
spectrometry: typically a Ce solution (20 mm in triflic acid, pH 1.0) was
added to complex 2 (2 mm) in acetonitrile (0.2 mL) with vigorous stirring,
during which the solution of 2 changed from deep purple to light brown;
after about 10 min, the resulting solution was injected into the mass spec-
trometer.
Figure 9. Observed (curves) and simulated (bars) isotope pattern. Upper:
2
+
[
[
Mꢀ3PF
6
ꢀpic+CF
3
SO
3
]
observed: 547.59; calculated: 547.54. Lower:
observed: 501.05; calculated: 501.01.
2
+
Mꢀ3PF
6
ꢀ2pic+CF
3
SO
3
]
UV/Vis absorption spectroscopy: The UV/Vis absorption spectrum was
measured with a CARY 300 Bio UV/Vis spectrophotometer.
Electrochemistry: Cyclic voltammetry and differential pulse voltammetry
were performed in acetonitrile and water, with an Autolab potentiostat
and GPES electrochemical interface (Eco Chemie), a glassy carbon disk
(
diameter 3 mm, freshly polished) as the working electrode and a plati-
num wire in a compartment separated from the bulk solution by a fritted
disk as counter-electrode. The electrolyte used was Bu NPF in acetoni-
trile (0.1m) or a mixture of aqueous CF SO H (pH 1) and acetonitrile
4
6
3
3
+
(
2:1, v/v). The reference electrode was Ag/Ag in AgNO
3
in acetonitrile
(
0.01m) or Ag/AgCl in aqueous KCl (3m). Potentials versus NHE are cal-
+
culated by adding +0.630 V to the potentials versus the Fc /Fc couple in
acetonitrile and by use of the Ru
3
+/2+
3 2
redox couple of Ru ACHUTNGTRNENG(U bpy) Cl as
standard with E1/2 =1.26 V versus NHE in aqueous solution. Half-wave
potentials (E1/2) were determined by cyclic voltammetry as the averages
of the anodic and cathodic peak potentials [E1/2 =(Epa+Epc)/2]. Reversi-
p p
bility was determined from the peak-to-peak separations DE (DE =
2 2 2 2 2
Scheme 5. Structures of Ru OH and Ru ACHUNTTGRENNUNG( OH ) (isomers not shown).
E
paꢀEpc) and the ratio of the anodic to cathodic peak currents (ipa/ipc).
The glassware used was oven-dried, assembled and flushed with argon
while hot. Before all measurements, solvent-saturated argon was bubbled
through the stirred solutions and the samples were kept under argon
during measurements.
the ruthenium metal as a seventh ligand without loss of pi-
colines.
Oxygen evolution analysis: The oxygen produced in the gas phase was
measured with a 3000 A Micro GC (Agilent Technologies) fitted with a
thermal conductive detector and a 5 ꢁ molecular sieve column (12 mm/
3
20 mm/10 m) allowing auto sampling and analysis at 668C. Helium was
Conclusion
used as carrier gas. Data were calibrated with a standard gas mixture and
air as the standard. In a typical run, a reaction flask containing a solution
of Ce in aqueous CF SO H solution (pH 1.0) was connected to the GC
3 3
IV
The three Ru-based precatalysts 1–3 for Ce -driven water
IV
oxidation have been synthesised and characterised. Under
optimised catalytic conditions, large TONs of 3500, 4500 and
sampling system through a septum. The reaction mixture was stirred and
deaerated with helium for 20 min. The oxygen content in the headspace
of the flask was measured by GC, after which the catalyst in acetonitrile
solution was injected into the flask through the septum by syringe. After
2000 have been achieved with complexes 1, 2 and 3, respec-
tively. Isotope labelling studies are consistent with the atoms
in the evolved oxygen originating only from water. Detailed
electrochemical and ESI-MS studies suggest that ligand ex-
2
4 h, the O
evolved was analysed. No leakage from air in the reaction flask was
found by monitoring the N content by GC.
2
content was measured again by GC. The amount of oxygen
2
9526
www.chemeurj.org
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 9520 – 9528

Synthesis and Catalytic Water Oxidation Activities of Ru Complexes
FULL PAPER
The kinetics were followed simultaneously with the aid of an Ocean
Optics oxygen sensor (FOXY-OR125-G) with a multifrequency phase
fluorimeter (MFPF-100) connected to a PC. The sensor probe was insert-
ed into the reaction flask described above through a septum. The oxygen
content versus the reaction time was recorded with the oxygen sensor
and calibrated with the data obtained from GC.
tion mixture had disappeared, it was cooled down to room temperature.
DMF was removed, and the crude product was purified by column chro-
matography on silica gel with acetone in CH
2
Cl
the desired product (0.8 g, 83% yield). H NMR (400 MHz, CDCl
2.53 (s, 6H), 2.59 (s, 6H), 7.20 (d, J=4.5 Hz, 2H), 8.39 (s, 2H), 8.43 (s,
2
(4%) as eluent to give
1
3
): d=
1
3
2H), 8.60 (d, J=4.9 Hz, 2H), 8.67 (s, 2H), 8.92 ppm (s, 2H); C NMR
(100 MHz, CDCl ): d=21.80, 21.86, 122.43, 122.88, 123.37, 125.29, 125.62,
1
8
O-Labelling isotope ratio mass spectrometry: Mass spectrometry was
3
1
48.43, 149.41, 149.86, 153.00, 156.10, 156.23, 158.72 ppm.
employed to perform simultaneous measurement of the concentrations of
various isotopomers of oxygen produced by the reaction mixture contain-
3,6-Bis-(2’,2’’-bipyrid-6’-yl)-1,2-dihydro-1,2,4,5-tetrazine (10): This com-
pound was prepared by a procedure similar to that used for compound 7.
Treatment of compound 9 (9.36 g, 19.7 mmol) with hydrazine hydrate
1
8
[20]
ing O-labelled water, as described previously.
(
4 2 3 6
Ce ACHTUNRTGENNUN(G NH ) ACHTUNGERTNNUN(G NO )
1
8
18
100 mg, 0.18 mmol), O-labelled water (10% O) (0.3 mL) and
CF
3
SO
3
H (5 mL) were mixed in the reaction vessel. Air in the reaction
(4.5 mL, 55%, 79.6 mmol) in ethanol afforded compound 10 (2.0 g, 52%
1
system was removed with a rough pump, prior to filling with argon to
keep the pressure at ꢁ100 mbar. After 15–20 min, an Ar-degassed solu-
tion of the catalyst in acetonitrile (0.15 mL) was injected into the reaction
vessel. The generated gases were measured and recorded against time.
yield). H NMR (500 MHz, CDCl
3
): d=7.34–7.37 (m, 2H), 7.87 (td, J=
8.0, 2.0 Hz, 2H), 7.93 (t, J=8.0 Hz, 2H), 8.09 (d, J=8.0 Hz, 2H), 8.49 (d,
J=7.5 Hz, 2H), 8.52 (d, J=8.0 Hz, 2H), 8.70 (d, J=5.0 Hz, 2H),
8.73 ppm (s, 2H).
4
,4’-Dimethyl-2,2’-bipyridine N-oxide (5): The mono N-oxide 5 was pre-
3,6-Bis-(2’,2’’-bipyrid-6’-yl)-1,2,4,5-tetrazine (11): This compound was pre-
pared by a procedure similar to that used for compound 8. Oxidation of
pared by the procedure used for the similar compound in ref. [38]. A sus-
pension of 3-chloroperbenzoic acid (m-CPBA, 13.8 g, 50–60%, 40–
3
compound 10 (3.576 g, 23.88 mmol) with concentrated HNO (10 mL,
4
1
CH
8 mmol) in CH
08C to a solution of 4,4’-dimethyl-2,2’-bipyridine (4, 8.0 g, 43.4 mmol) in
Cl (100 mL). The mixture was stirred at room temperature over-
night. Aqueous Na CO (10%, 100 mL) was added at 08C and the mix-
ture was stirred for 30 min. The organic phase was separated, washed
once more with aqueous Na CO (10%, 100 mL) and dried over solid
Na CO . The solvent was removed and the crude product was purified by
column chromatography on SiO with MeOH in CH Cl (5%) as eluent
to afford the pure product (5.7 g, 66% yield) as a white solid. H NMR
500 MHz, CDCl ): d=2.42 (s, 3H), 2.46 (s, 3H), 7.09 (dd, J=6.6, 2.6 Hz,
H), 7.19 (dd, J=5.0, 0.6 Hz, 1H), 7.96 (d, J=2.4 Hz, 1H), 8.22 (d, J=
2
Cl
2
(100 mL) was added dropwise over 50 min at 5–
65%) in acetic acid (100 mL) at 08C gave compound 11 (8.5 g, 91%
yield).
2
2
Ligand L2: This ligand was prepared by a method similar to that report-
[40]
2
3
ed by Schubert and co-workers. A mixture of compound 11 (330 mg,
0
1
.846 mmol), cyclopentanone (143 mg, 1.7 mmol), morpholine (140 mg,
.6 mmol) and silica gel (210 mg) in toluene was heated at 1108C for 5 h.
2
3
2
3
After the mixture had been allowed to cool to RT, a white precipitate
had formed. The mixture was filtered through celite and washed thor-
oughly with MeOH in CH Cl (10%). The slightly yellow filtrate was col-
2 2
2
2
2
1
(
3
lected and solvent was removed under reduced pressure. The desired
1
6
1
product was obtained in 100% yield as a white powder. H NMR
.6 Hz, 1H), 8.59 (d, J=5.0 Hz, 1H), 8.75 ppm (d, J=0.6 Hz, 1H).
(
3
500 MHz, CDCl ): d=2.30 (quint., J=7.8 Hz, 2H), 3.76 (t, J=7.7 Hz,
6
-Cyano-4,4’-dimethyl-2,2’-bipyridine (6): Compound 6 was prepared by
4H), 7.34–7.37 (m, 2H), 7.89 (td, J=7.6, 1.6 Hz, 2H), 8.05 (t, J=7.8 Hz,
2H), 8.51 (d, J=7.9 Hz, 2H), 8.54 (d, J=7.7 Hz, 2H), 8.73 ppm (d, J=
6.9 Hz, 4H); C NMR (125 MHz, CDCl ): d=24.5, 34.1, 121.1, 123.5,
3
the procedure used for the similar compound in ref. [38]. Me
3
SiCN
CAUTION! Highly toxic cyanide equivalent: handle with care! 4.6 mL,
1
3
(
3
CH
3
.39 g, 34 mmol) was added at room temperature to a solution of 5 in
Cl (50 mL). After 5 min, dimethylcarbamoyl chloride (3.15 mL,
4 mmol) was added. The mixture was stirred at room temperature for 6
days. Aqueous NaHCO (10%, 50 mL) was added and the mixture was
stirred for 20 min. The organic phase was separated, washed once more
with aqueous NaHCO (10%, 50 mL) and then with brine, and dried
over solid Na CO . The solvent was removed and the crude product was
123.8, 137.0, 137.9, 145.5, 149.3, 155.1, 155.3, 155.4, 156.3 ppm.
2
2
Complex 1: A mixture of the ligand L1 (100 mg, 0.226 mmol) and cis-
[
Ru
reflux for one day. Water (10 mL), 4-methylpyridine (1.1 mL,
1.28 mmol), LiCl (50 mg, 1.19 mmol) and triethylamine (0.5 mL) were
added, and the mixture was further heated at reflux for 2 days. Ethanol
was removed and excess NH PF was added. The precipitate was separat-
ed and purified by column chromatography on Al to give the desired
]acetone): d=2.09
s, 12H), 2.58 (s, 6H), 2.76 (s, 6H), 6.70 (d, J=6 Hz, 8H), 7.72 (d, J=
4 2
ACHTUNGTRNNGE(U dmso) Cl ] (240 mg, 0.496 mmol) in ethanol (80 mL) was heated at
3
1
3
2
3
4
6
purified by recrystallisation from a mixture of diethyl ether and hexane
to afford the pure product (4.69 g, 70% yield) as a white solid. H NMR
2
O
3
1
1
product (120 mg, 36% yield). H NMR (500 MHz, [D
(
6
(
1
500 MHz, CDCl
3
): d=2.49 (s, 3H), 2.52 (s, 3H), 7.22 (d, J=5.0 Hz,
H), 7.55 (s, 1H), 8.32 (s, 1H), 8.50 (s, 1H), 8.56 ppm (d, J=5.0 Hz, 1H).
,6-Bis-(4’,4’’-dimethyl-2’,2’’-bipyrid-6’-yl)-1,2-dihydro-1,2,4,5-tetrazine
6
Hz, 8H), 7.95 (d, J=5.5 Hz, 2H), 8.27 (s, 2H), 8.42 (s, 2H), 8.94 (s,
1
3
3
2H), 9.23 (s, 2H), 9.42 ppm (d, J=5.5 Hz, 2H); C NMR (125 MHz,
[D ]acetone): d=20.60, 21.20, 21.73, 125.49, 125.56, 126.92, 127.60,
129.03, 131.45, 148.64, 150.70, 151.43, 152.72, 153.44, 157.73, 159.72,
(
7): This compound was prepared by the similar procedure in ref. [39].
Hydrazine hydrate (4.0 g, 98%, 78.3 mmol) was added to a suspension of
(4.16 g, 19.92 mmol) in ethanol (70 mL). The mixture became a clear
6
3
+
6
160.34, 163.72 ppm; HRMS (ESI): m/z: calcd for [Mꢀ3PF
6
]
: 351.7384;
Cl
3
N (1 mL), 4-picoline (1 mL) and LiCl
brown solution on heating. It was gently heated at reflux for 24 h. The re-
sulting yellow precipitate was filtered and washed with ethanol. The
crude product was recrystallised from ethanol to give the dihydro base
found: 351.7380.
Complex 2: The ligand L2 (214 mg, 0.5 mmol) and cis-[Ru
(
heated at reflux overnight. Et
A
T
N
T
E
N
G
4
2
]
1
(
1.67 g, 37%). H NMR (500 MHz, CDCl
3
): d=2.39 (s, 6H), 2.40 (s, 6H),
7
.08 (dd, J=5.0, 0.8 Hz, 2H), 7.82 (s, 2H), 8.21 (s, 2H), 8.25 (s, 2H), 8.46
(
90 mg) were added, and the mixture was further heated at reflux for
(
d, J=5.0 Hz, 2H), 8.68 ppm (s, 2H).
2
4 h. Most of the solvent was removed, and the residue was washed with
3
,6-Bis-(4’,4’’-dimethyl-2’,2’’-bipyrid-6’-yl)-1,2,4,5-tetrazine (8): This com-
ether and purified by chromatography on silica gel, with use of a mixture
of acetonitrile/H O/KNO (sat.) (30:1:1) as eluent, and finally treated
with NH PF , to afford the title complex (330 mg, 45% yield). H NMR
(500 MHz, CD CN): d=2.03 (s, 12H), 2.79 (quint, J=7.8 Hz, 2H), 4.09
[
39]
pound was prepared by a procedure similar to that in the literature.
2
3
1
Concentrated nitric acid (8 mL, 65%) was added dropwise to a suspen-
sion of 7 (1.67 g, 3.72 mmol) in acetic acid (60 mL) cooled with an ice-
water bath. The mixture immediately turned red. After 30 min, excess ice
was added and the mixture was neutralised by the addition of sodium bi-
carbonate. The precipitate was filtered and washed with ethanol to give
the tetrazine 8 as a red solid (0.975 g, 59% yield). No NMR or MS data
are available, due to the poor solubility of the product. The crude prod-
uct was used in the next reaction without purification.
4
6
3
(t, J=7.9 Hz, 4H), 6.53 (d, J=6.2 Hz, 8H), 7.44 (d, J=6.4 Hz, 8H), 7.98–
8.08 (m, 8H), 8.16 (d, J=8.0 Hz, 2H), 8.69 (d, J=8.4 Hz, 2H), 9.34 ppm
1
3
3
(d, J=5.3 Hz, 2H); C NMR (125 MHz, CD CN): d=20.6, 24.7, 34.5,
124.1, 124.6, 126.7, 130.3, 130.9, 135.3, 140.3, 150.8, 151.0, 151.3, 152.9,
158.4, 160.0, 160.6, 161.4 ppm. ESI-MS: m/z: 346.16 (calcd for
3
+
2+
[Mꢀ3PF
6
]
: 346.39), 592.06 (calcd for [Mꢀ2PF
6
]
: 592.07), 1328.93
+
(
calcd for [MꢀPF
[Ru(L2)(dmso)Cl
6
] : 1329.11).
Ligand L1: This compound was prepared by a procedure similar to that
in ref. [39]. Acetylene was slowly bubbled through a solution of 8 (0.97 g,
A
H
U
G
R
N
U
G
2
]:
A
solution of cis-[Ru
A
H
U
G
R
N
U
G
4
2
]
2
.17 mmol) in DMF (25 mL) at reflux. After the red colour of the reac-
0.198 mmol) and L2 (42 mg, 0.098 mmol) in EtOH (4 mL) was heated at
Chem. Eur. J. 2011, 17, 9520 – 9528
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9527

L. Sun et al.
7
08C for 0.5 h. A black precipitate was obtained by filtration, washed
[13] Z. Deng, H.-W. Tseng, R. Zong, D. Wang, R. Thummel, Inorg.
Chem. 2008, 47, 1835–1848.
[14] H.-W. Tseng, R. Zong, J. T. Muckerman, R. Thummel, Inorg. Chem.
2008, 47, 11763–11773.
[15] J. T. Muckerman, D. E. Polyansky, T. Wada, K. Tanaka, E. Fujita,
Inorg. Chem. 2008, 47, 1787–1802.
[16] Y. V. Geletii, B. Botar, P. Kçerler, D. A. Hillesheim, D. G. Musaev,
C. L. Hill, Angew. Chem. 2008, 120, 3960–3963; Angew. Chem. Int.
Ed. 2008, 47, 3896–3899.
1
with EtOH and dried in vacuo (yield 82 mg, 96%). H NMR (500 MHz,
CDCl ): d=10.13 (d, J=5.19 Hz, 1H), 8.74 (d, J=4.18 Hz, 1H), 8.61 (d,
J=7.69 Hz, 1H), 8.40 (d, J=7.78 Hz, 1H), 8.23 (d, J=8.22 Hz, 1H), 8.19
d, J=12.37 Hz, 1H), 8.17 (d, J=12.67 Hz, 1H), 8.10 (t, J=8.29 Hz, 1H),
3
(
7
7
.95 (t, J=8.00 Hz, 1H), 7.89–7.81 (m, 2H), 7.52 (t, J=6.40 Hz, 1H),
.38 (t, J=6.50 Hz, 1H), 3.79 (s, 6H), 3.73 (t, J=7.50 Hz, 2H), 3.56 (t,
J=7.48 Hz, 2H), 2.38 ppm (quint, J=7.02 Hz, 2H); ESI-MS: m/z: 679.0
+
+
(
calcd for [M+H] : 679.0) and 701.0 (calcd for [M+Na] : 701.0).
Complex 3: A mixture of [Ru(L2)(dmso)Cl ] (44 mg, 0.064 mmol) and 4-
picoline (0.5 mL) in EtOH/H O (10 mL, 9:1 v/v) was deoxygenated (N
and heated at reflux overnight. The solvent was evaporated and the resi-
due was dissolved in H O (5 mL), to which excess NH PF aqueous solu-
[
17] A. Sartorel, M. Carraro, G. Scorrano, R. D. Zorzi, S. Geremia, N. D.
A
H
U
G
R
N
U
G
2
McDaniel, S. Bernhard, M. Bonchio, J. Am. Chem. Soc. 2008, 130,
2
2
)
5
006–5007.
18] L. Duan, A. Fischer, Y. Xu, L. Sun, J. Am. Chem. Soc. 2009, 131,
0397–10399.
19] L. Duan, Y. Xu, P. Zhang, M. Wang, L. Sun, Inorg. Chem. 2010, 49,
09–215.
20] Y. Xu, T. ꢁkermark, V. Gyollai, D. Zou, L. Eriksson, L. Duan, R.
Zhang, B. ꢁkermark, L. Sun, Inorg. Chem. 2009, 48, 2717–2719.
21] Y. Xu, L. Duan, L. Tong, B. ꢁkermark, L. Sun, Chem. Commun.
[
[
[
[
2
4
6
1
tion had been added. The formed precipitate was collected and washed
with water. The crude product was purified by column chromatography
2
on silica gel with acetonitrile/KNO
was collected and solvents were removed. The resulting solid was dis-
solved in a small amount of CH CN/H O and then NH PF was added.
Evaporation of CH CN resulted in the formation of precipitate, which
was washed with water and dried under vacuum to provide a red solid
3 2
/H O 40:1:1 as eluent. The red band
3
2
4
6
3
2
010, 46, 6506–6508.
1
[22] Y. Xu, A. Fischer, L. Duan, L. Tong, E. Gabrielsson, B. ꢁkermark,
(
37 mg, 52%). H NMR (400 MHz, MHz, CD
3
CN): d=8.74 (m, 2H),
L. Sun, Angew. Chem. 2010, 122, 9118–9121; Angew. Chem. Int. Ed.
8
8
2
.62 (d, J=7.93 Hz, 1H), 8.54–8.47 (m, 4H), 8.38 (d, J=5.22 Hz, 2H),
.36 (d, J=8.32 Hz, 1H), 8.20 (t, J=7.81, 7.81 Hz, 1H), 8.17–8.08 (m,
H), 7.98 (t, J=7.03 Hz, 1H), 7.84–7.78 (m, 1H), 7.53–7.44 (m, 3H), 7.34
2
010, 49, 8934–8937.
[
[
23] L. Duan, Y. H. Xu, M. Gorlov, L. P. Tong, S. Andersson and L. C.
Sun, Chem. Eur. J. 2010, 16, 4659–4668.
24] Y. V. Geletii, Z. Huang, Y. Hou, D. G. Musaev, T. Lian, C. L. Hill, J.
Am. Chem. Soc. 2009, 131, 7522–7523.
(
d, J=5.33 Hz, 4H), 6.82 (d, J=5.78 Hz, 4H), 3.91 (t, J=7.74 Hz, 2H),
3
2
9
.65 (t, J=7.72 Hz, 2H), 2.60 (s, 3H), 2.41 ( quint, J=7.27 Hz, 2H),
.15 ppm (s, 6H); ESI-MS: m/z: 404.52 (calcd for [Mꢀ2PF
2
+
6
] : 404.63),
+
[25] M. W. Kanan, D. G. Nocera, Science 2008, 321, 1072–1075.
53.87 (calcd for [MꢀPF
6
] : 954.22).
[
26] Q. Yin, J. M. Tan, C. Besson, Y. V. Geletii, D. G. Musaev, A. E. Kuz-
netsov, Z. Luo, K. I. Hardcastle, C. L. Hill, Science 2010, 328, 342–
3
45.
[
[
27] N. D. McDaniel, F. J. Coughlin, L. L. Tinker, S. Bernhard, J. Am.
Chem. Soc. 2008, 130, 210–217.
28] J. F. Hull, D. Balcells, J. D. Blakemore, C. D. Incarvito, O. Eisen-
stein, G. W. Brudvig, R. H. Crabtree, J. Am. Chem. Soc. 2009, 131,
8730–8731.
Acknowledgements
This work was supported by the Swedish Research Council, the K&A
Wallenberg Foundation, the Swedish Energy Agency and the China
Scholarship Council.
[29] W. C. Ellis, N. D. McDaniel, S. Bernhard, T. J. Collins, J. Am. Chem.
Soc. 2010, 132, 10990–10991.
[
30] J. H. Alstrum-Acevedo, M. K. Brennaman, T. J. Meyer, Inorg.
Chem. 2005, 44, 6802–6827.
[
[
1] E. Amouyal, Solar Energy Mater. Solar Cells 1995, 38, 249–276.
2] K. N. Ferreira, T. M. Iverson, K. Maghlaoui, J. Barber, S. Iwata, Sci-
ence 2004, 303, 1831–1838.
[31] Y. Shimazaki, T. Nagano, H. Takesue, B.-H. Ye, F. Tani, Y. Naruta,
Angew. Chem. 2004, 116, 100–102; Angew. Chem. Int. Ed. 2004, 43,
98–100.
[32] A. K. Poulsen, A. Rompel, C. J. McKenzie, Angew. Chem. 2005, 117,
7076–7080; Angew. Chem. Int. Ed. 2005, 44, 6916–6920.
[33] J. Limburg, J. S. Vrettos, H. Chen, J. C. de Paula, R. H. Crabtree,
G. W. Brudvig, J. Am. Chem. Soc. 2001, 123, 423–430.
[34] R. Tagore, R. H. Crabtree, G. W. Brudvig, Inorg. Chem. 2008, 47,
1815–1823.
[
3] G. C. Dismukes, R. Brimblecombe, G. A. N. Felton, R. S. Pryadun,
J. E. Sheats, L. Spiccia, G. F. Swiegers, Acc. Chem. Res. 2009, 42,
1
935–1943.
[
[
4] M. Yagi, M. Kaneko, Chem. Rev. 2001, 101, 21–35.
5] C. W. Cady, R. H. Crabtree, G. W. Brudvig, Coord. Chem. Rev. 2008,
2
52, 444–455.
6] L. L. Tinker, N. D. McDaniel, S. Bernhard, J. Mater. Chem. 2009, 19,
328–3337.
[
[
[
3
[35] S. W. Gersten, G. J. Samuels, T. J. Meyer, J. Am. Chem. Soc. 1982,
104, 4029–4030.
7] R. Brimblecombe, G. C. Dismukes, G. F. Swiegers, L. Spiccia,
Dalton Transactions 2009, 9374–9384.
8] J. J. Concepcion, J. W. Jurss, M. K. Brennaman, P. G. Hoertz,
A. O. T. Patrocinio, N. Y. Murakami Iha, J. L. Templeton, T. J.
Meyer, Acc. Chem. Res. 2009, 42, 1954–1965.
9] F. Bozoglian, S. Romain, M. Z. Ertem, T. K. Todorova, C. Sens, J.
Mola, M. Rodrꢃguez, I. Romero, J. Benet-Buchholz, X. Fontrodona,
C. J. Cramer, L. Gagliardi, A. Llobet, J. Am. Chem. Soc. 2009, 131,
[36] a) J. J. Concepcion, J. W. Jurss, J. L. Templeton, T. J. Meyer, J. Am.
Chem. Soc. 2008, 130, 16462–16463; b) D. J. Wasylenko, C. Ganesa-
moorthy, M. A. Henderson, B. D. Koivisto, H. D. Osthoff, C. P. Ber-
linguette, J. Am. Chem. Soc. 2010, 132, 16094–16106; c) S. Masaoka,
K. Saki, Chem. Lett. 2009, 38, 182–183; d) M. Yoshida, S. Masaoka,
K. Saki, Chem. Lett. 2009, 38, 702–703.
[37] R. Lomoth, P. Huang, J. Zheng, L. Sun, L. Hammarstrçm, B. ꢁker-
mark, S. Styring, Eur. J. Inorg. Chem. 2002, 2965–2974.
[38] T. Norrby, A. Bçrje, L. Zhang, B. ꢁkermark, Acta Chem. Scand.
1998, 52, 77–85.
[
1
5176–15187.
[
10] X. Sala, I. Romero, M. Rodrꢃguez, L. Escriche, A. Llobet, Angew.
Chem. 2009, 121, 2882–2893; Angew. Chem. Int. Ed. 2009, 48, 2842–
2
852.
[39] W. A. Butte, F. H. Case, J. Org. Chem. 1961, 26, 4690–4692.
[40] R. Hoogenboom, B. C. Moore, U. S. Schubert, J.Org.Chem. 2006, 71,
4903–4909.
[
[
11] C. Sens, I. Romero, M. Rodriguez, A. Llobet, T. Parella, J. Benet-
Buchholz, J. Am. Chem. Soc. 2004, 126, 7798–7799.
12] R. Zong, R. P. Thummel, J. Am. Chem. Soc. 2005, 127, 12802–
Received: January 25, 2011
1
2803.
Published online: July 15, 2011
9528
www.chemeurj.org
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 9520 – 9528