72
H. Fan et al. / Journal of Molecular Catalysis A: Chemical 425 (2016) 68–75
3. Results and discussion
3.1. Characterization of KH550-modified-SiO2
To introduce ligands for the loading of the metal complexes,
SiO2 was primarily modified via a silylation reaction between the
surface hydroxyl groups on SiO2 and C2H5O groups in KH550. The
typical Fourier transform infrared spectroscopy (FTIR) absorption
peaks of SiO2 and KH550-modified-SiO2 can be observed in Fig. 1 (as
shown in Fig. 1A). As observed in the spectrum of KH550-modified-
SiO2, 2924 cm−1 and 2850 cm−1 were the characteristic bands of
the symmetric and antisymmetric vibrations of CH2 connected
to the silane chains, while two bands at 1483 cm−1 and 1405 cm−1
were assigned to the bending vibrations of CH2. These results ver-
ified the successful incorporation of organic carbon chain, existing
in KH550, onto the surface of SiO2. The typical vibrations of N
H
antisymmetric stretching vibration at 1093 cm−1 of Si-O-Si bonds
generated from the abundant Si-OH groups on the surface of SiO2
and the triethoxy groups in KH550. However, the existence of N
element in KH550-modified-SiO2 was corroborated by X-ray pho-
toelectron spectroscopy (XPS) (as seen in Fig. 1B). Additional signals
of N were clearly observed in the XPS spectrum of KH550-modified-
SiO2. These results indicated the successful bonding of aminosilane
onto the surface of SiO2 through the silylation reaction. The remain-
ing bands at 799 cm−1, 3430 cm−1 and 1640 cm−1 were ascribed to
the systematic stretching vibration of Si-O and the antisymmetric
stretching and bending vibrations of water molecular, respectively.
Fig. 5. The effect of hydrogen pressure on the catalytic performance of Ru-NH2-
SiO2 Reaction conditions: 0.5 g of 2 wt.% reduced Ru-NH2-SiO2, 10 ml methanol, 0.4 g
DMO, 80 ◦C, t = 24 h.
The FTIR spectra were recorded by a BRUKER TENSOR27 FTIR
spectrophotometer (Bruker, Germany) equipped with a DTGS
detector. The samples were grinded to fine powder and mixed care-
fully with KBr, then pressed into translucent disks. The spectra were
recorded from 400 cm−1 to 4000 cm−1
.
X-ray photoelectron spectroscopy (XPS) was measured on a VG
MiltiLab 2000 spectrometer with Al K␣radiation and a multichan-
nel detector. Prior to the test, the samples were reduced at 500 ◦C
in 1 atm H2 for 2 h. The obtained binding energies were calibrated
using the C1s peak at 284.6 eV as the reference. The experiment
error was given with 0.1 eV.
3.2. Characterization of Ru–NH2-SiO2 catalyst
TEM and HRTEM measurements were determined by a field-
emission transmission electron microscopy (FETEM, JEM-2011F)
operated at 200 kV voltages. Prior to test, the reduced catalysts were
suspended in ethanol with an ultrasonic dispersion for 15 min, and
deposited on copper grids coated with amorphous carbon films.
The content of Ru in the samples was ascertained by ICP optical
emission spectroscopy (Perkin Elmer Optima 2100DV).
As their isolated electron pairs, the introduced NH2 groups can
strongly bond with Ru sites though coordination. Therefore they act
as anchoring sites for Ru complexes and highly improve the disper-
sion of Ru sites on the surface of SiO2 (as shown in Scheme 1b). The
physicochemical properties of catalysts are listed in Table 1.
3.2.1. Physicochemical properties Ru–NH2-SiO2 catalyst
Compared to SiO2, the BET area, pore size and pore volume of
Ru/SiO2 and Ru-NH2-SiO2 had slightly changed. The particles size
of Ru species was much smaller on Ru-NH2-SiO2 than those on
Ru/SiO2.
2.2.5. Catalyst performance tests
The catalytic tests were performed in a batch reactor (50 ml)
with magnetic stirring. For a specific procedure, the reactor was
fed with DMO (0.4 g), methyl alcohol (10 ml), catalyst (0.5 g), then
sealed and purged by H2 to exclude the air for several times. After
that, the reactor was fulfilled with H2 to reaction pressure and
heated to object temperature. When reaction was over, the reac-
tor was quenched in cool water. The products were separated from
catalyst by centrifuging and analyzed by a GC instrument with a
FID detector. The catalyst was recycled as following procedure:
after each run, the catalyst was separated by centrifuging and thor-
oughly washed with methyl alcohol for three times, then reused
for another run under the identical conditions. The conversion and
selectivity of products were calculated based on the following equa-
tions:
3.2.2. XPS analysis of Ru–NH2-SiO2 catalyst
To study the structure and chemical environment of the Ru-
Fig. 2A, the N signal appeared in the spectrum of Ru-NH2-SiO2 cat-
alyst, while absent in the spectrum of Ru/SiO2. It could also be
observed that a positive shift of 0.50 eV for N 1s appeared in the
spectrum of Ru-NH2-SiO2 compared with that in KH550-midified-
SiO2 (as shown in Fig. 2B). These results indicated the formation of
NH2-Si-O-Si structure and the coordination of Ru3+ sites and NH2
groups [17][17a]. The chemical state of Ru0 on the Ru/SiO2 and
Ru-NH2-SiO2 catalysts was also studied. As previously reported,
the Ru3d5/2 XPS peak for Ru0 is at 280.2 eV. In this work, the BE
value of Ru0 on Ru/SiO2 catalyst is 280.30 eV, while a negative
shift of 0.50 eV was obtained on Ru-NH2-SiO2 catalyst, which was
279.80 eV. This result reflected that some electrons were trans-
ferred from amino ligands to Ru0, leading to an electron-rich state
of Ru0 (as shown in Fig. 2C). The electron-rich state of Ru0 could
Amount of DMO after reaction mol
(
(
)
)
Conversion % = 100 −
( )
× 100
Total amount DMO in the feed mol
Amount of a product mol
(
)
(
Selectivity % =
( )
× 100
)
Total amount of all products mol
Turnover frequency (TOF) was calculated to explore the intrinsic
catalytic activity. It was estimated by the total Ru loadings:
effectively activate
sites are readily to yield electronic feedback to anti orbitals of the
O [25].
C
O bonds as the adequate electrons on Ru0
Number of DMO molecular converted
TOF =
Number of surface Ruatoms × reaction time, h
(
)
(
)
C