L.-Z. Zeng et al. / Catalysis Communications 68 (2015) 84–87
85
peak at 5.90 ppm with an area in a 2:3 ratio to that of the methyl group
on the dendritic structure motif. This peak is attributed to the coordinat-
ed H2O on the complex (Fig. S4).
The UV–vis absorption profiles of the complexes were investigated
in 0.1 M HNO3 aqueous solution or in acetonitrile (Figs. 1a, S5 and S6).
Both complexes exhibited characteristic broad and intense long-
wavelength absorptions, which are attributed to the metal-to-ligand
charge transfer (MLCT) in nature. Compared with 1, complex 2
displayed blue-shifted MLCT absorption due to an increase in the
¯
⁎
dπ–π energy gap caused by the Cl to H2O ligand exchange process
[22,23]. Cyclic voltammetry (CV) measurements of 1 in pH 1.0 displayed
one redox event at E1/2 = 0.97 V vs normal hydrogen electrode
(NHE) was observed in a 0–1.7 V potential sweep window and this
redox process is assigned to RuIII/II (Fig. S7). For complex 2, three ox-
idation peaks were observed by CV and DPV at 0.93, 1.21 and 1.68
and are attributed to the RuIII/II, RuIV/III and RuV/IV processes, respectively
(Figs. 1b and S8). These values are lower than those determined for
[Ru(bpy)(tpy)(H2O)]2+ in literature (1.04, 1.23 and 1.80) [10]. Previous
insights into the electron withdrawing and donating effects of function-
al groups on the [Ru(bpy)(tpy)(H2O)]2+ motif have revealed that
electron-donating groups can cause an increase in the π-back bonding
of the tpy ligand to the bpy ligand and lower the oxidation potentials
of the resulting complexes. This behavior eventually leads to an increase
in catalytic activity but at a compromise to the catalytic stability [17]. It
is therefore expected from the electrochemical data obtained that en-
hanced catalytic activity can be achieved, and are ascribed to the in-
crease in the π-back bonding to the bpy ligand by the conjugated tris-
terpyridine ligand. Additionally, the onset potential of the catalytic
curve for complex 2 in pH 1.0 HNO3 is almost 1.55 V vs NHE, 200 mV
lower than the reduction potential of CAN (~1.75 V vs NHE), which
meant the possibility of the utility of CAN as oxidant to drive complex
2 for chemical water oxidation.
Scheme 1. Simplified molecular structures of ligand and complexes.
Dendritic tris-terpyridine ligand: 1H NMR (400 MHz, CDCl3): δ =
8.72 (d, 6H, J = 4.0 Hz; Ha), 8.681 (d, 6H, J = 7.6 Hz; Hd), 8.414 (s,
6H; He), 7.89–7.82 (m, 6H; Hc), 7.36–7.29 (m, 6H; Hb), 1.91 (s, 9H; HI).
Complex 1: ESI-MS: 564.0554 [M-3(PF6)]3+ 1H NMR (400 MHz,
.
acetone-d6): δ = 10.25 (d, 3H, J = 5.0 Hz), 8.76–8.55 (m, 15H), 8.49
(t, 3H, J = 6.8 Hz), 8.277 (t, 3H, J = 8 Hz), 8.00–7.82 (m, 9H), 7.77–
7.64 (m, 9H), 7.51 (d, 3H, J = 5.2, 37.6 Hz), 7.31 (t, 6H, J = 6.0 Hz),
6.99 (t, 3H, J = 6.0 Hz), 2.50 (s, 9H).
Complex 2: 1H NMR (400 MHz, acetone-d6): δ = 9.87 (d, 3H, J =
10.2 Hz), 9.07–8.80 (m, 15H), 8.67 (t, 3H, J = 6.8 Hz), 8.53 (t, 3H, J =
6.4 Hz), 8.29–8.03 (m, 15H), 7.88 (t, 3H, J = 7.6 Hz), 7.72 (d, 3H, J =
4.0 Hz), 7.59 (t, 6H, J = 4.0 Hz), 7.17 (t, 3H, J = 3.6 Hz), 5.90 (m, 6H),
2.52 (s, 9H).
2.2. Chemical water oxidation
Ce(IV)-driven water oxidation for complex 1 and 2 was investigated
and showed in Fig. 2. Firstly, water oxidation catalyzed by complex 1
was performed at various concentrations (Fig. 2a). Oxygen evolution
steadily increased over the time period of 12 h in the presence of excess
amounts of CAN, over 63 μmol of O2 was achieved at a catalyst concen-
tration of 0.05 mM. In addition, by linear fitting of the oxygen evolved in
the first 10,000 s, we found that the initial rate of oxygen evolution and
the concentration of the catalyst can be fitted into a single exponential
function with an order of 0.75 (R2 = 0.998). The obtained value deviates
from first order linear relationship (1.0, R2 = 0.991) which is commonly
known for mononuclear catalysts, and the catalytic behavior may be
tentatively regarded as pseudo-first-order. The reason for the
deviation can be explained on that the substitution of all three Cl li-
gands by H2O on catalyst 1 is a gradual process, which may vary at
different 1 concentrations, thereby leading various moiety ratios of
[Ru(bpy)(tpy)(H2O)]2+/[Ru(bpy)(tpy)(Cl)]+ (Fig. 2b). Complex 2
showed a great deal of enhanced catalytic activity but decreased stability;
at a catalyst concentration of 0.01 mM, oxygen evolution reached a pla-
teau at around 4 h, producing 30 μmol of O2 (Fig. 2c). Finally, when sys-
tematically comparing the catalytic activities of complexes 1 and 2 with
that of mononuclear [Ru(tpy)(bpy)(OH2)]2+, we found that the catalytic
activity of 2 was more than two fold 1 or [Ru(tpy)(bpy)(OH2)]2+ for
equimolar ruthenium amounts (Fig. 2d). Over 17.2 μmol of O2 was
produced by 2 in the first 5000 s while only 8.2 μmol and 7.9 μmol of
O2 were obtained by 1 and [Ru(tpy)(bpy)(OH2)]2+, respectively. Under
the present experimental conditions, a TOF value of 3.44 × 10−2 s−1
was obtained for 2 for the first 5000 s of reaction. In addition,
30.47 μmol of O2 evolved over 5 h which gives a calculated turnover
number (TON) value of 305 at a concentration of 0.01 mM of complex
2. At equimolar ruthenium amounts, complex 1 and the mononuclear
[Ru(bpy)(tpy)(H2O)]2+ exhibited similar catalytic water oxidation
trends, the latter showing slightly better performance after 10,000 s.
These results demonstrate the possibility of enhancing the catalytic ef-
ficiency of classical [Ru(tpy)(bpy)(OH2)]2+ catalysts by introducing a
In a typical experiment, 100 μL 1.0 mM catalyst solution of 1 or 2 was
injected into 10 mL pH = 1.0 HNO3 aqueous solution containing 548 mg
of (NH4)2Ce(NO3)6 (CAN) in a degassed cell with 21 mL volume, the ox-
ygen evolution curve over time was reflected using an ocean optics
FOXY-OR125G probe. The turnover numbers (TON) were calculated
by TON = n(O2)/n(catalyst) and the turnover frequency (TOF) was cal-
culated by TOF = TON/t(s).
2.3. Photocatalytic oxidation of alcohols
A quartz tube (15 mL) containing 1.0 mM of Ru(bpy)3Cl2, 0.01 mM
of catalyst, 10 mM of substrate, 20 mM of [Co(NH3)5Cl]Cl2 and 0.1 M
phosphate buffer was irradiated (λ N 420 nm) using light-emitting di-
odes (LEDs) (30 × 1 W) for 8 h in degassed aqueous solution (5 mL)
at pH 6.8 and room temperature. The resulting solution was extracted
with CH2Cl2 (3 × 20 mL). The organic fraction was dried over anhydrous
Na2SO4 and then evaporated to give the crude product. The results were
obtained by 1H NMR spectroscopy with quantitative analyses via the
ratio of integrated peak intensities of products to that of corresponding
substrates.
3. Results and discussion
Structures of the compounds were characterized by nuclear magnet-
ic resonance (NMR) and electrospray ionization mass spectrometry (EI-
MS) (Fig. S1–4). The 1H NMR spectrum of 1 showed fair characteristic
resonances of one stereoselective polypyridyl complex with only one
double peak observed at 10.25 ppm (Fig. S2), and the high-resolution
mass spectral peaks at m/z = 564.0554 are ascribed to the [M −
3(PF6)]3+ species (Fig. S3). Formation of 2 was performed by adopting
an established procedure in which complex 1 was refluxed with
AgClO4 in acetone/water solution for over 3 days and then precipitated
by NH4PF6 [25]. The 1H NMR spectrum showed the appearance of a new
Zeng, Ling-Zhen
