CHEMCATCHEM
FULL PAPERS
a H/Ru stoichiometry of 1:1 and a Ru surface atomic density of
in N2 to a constant weight. Then, an aliquot of approximately
25 mg of the dried catalysts was sealed in a desiccator containing
saturated NaCl aqueous solution until attaining constant weight
1
9
À2 [10]
1
.63ꢂ10 atomsm .
XRD patterns were acquired on a Bruker AXS D8 Advance X-ray dif-
fractometer using Ni-filtered CuKa radiation (l=0.15418 nm). The
tube voltage was 40 kV, and the current was 40 mA. The 2q angles
again, which represents the saturation adsorption of H O. The
2
mass difference between the two constant-weight catalysts was
the amount of H O adsorbed, which was then normalized to the
À1
2
were scanned from 108 to 708 at a speed of 48min . The narrow-
catalyst weight before water adsorption. The catalysts after satura-
À1
range XRD patterns were scanned from 388 to 488 at 0.18min .
tion adsorption of H O were immediately subjected to FTIR analy-
2
The crystallite sizes of ZrO and Ru were calculated from the inte-
2
sis. The catalyst powders (ꢀ2.0 mg) were mixed with KBr and
gral width of the strongest diffraction peaks of ZrO and Ru, re-
2
pressed into self-supporting wafers of 2 cm in diameter, which
spectively, using the Scherrer equation after correction for instru-
were fixed in the IR cell with CaF windows. The IR spectra were
[51]
2
mental broadening with the Warren procedure. In the Scherrer
scanned on a Nicolet Nexus 470 FTIR spectrometer equipped with
2
2 1/2
equation, d=0.89l/bcosq (b=(B Àb ) ), q is the diffraction angle,
B and b are the full width at the half maximum of the peak at q
and the instrumental broadening, respectively, and b is the FWHM
of the peak at q after correction for instrumental broadening. To
obtain an estimate of the contribution to the peak width by the in-
strument, quartz powder was used because of its good crystalline
À1
À1
a DTGS detector in the range of 4000–400 cm with 4 cm spec-
tral resolution by signal-averaging 32 scans.
The relative adsorption capacities of cyclohexene on the catalysts
were examined by temperature-programmed desorption on a TA
Instruments SDT-Q600 thermogravimetric analyzer coupled with
a Pfeiffer Instruments GSD 301 T2 mass spectrometer. Before the
measurements, the catalysts were immersed in cyclohexene for sat-
uration adsorption, followed by drying at ambient temperature
[52]
perfection and virtually no intrinsic line broadening.
The surface morphology and particle sizes were observed by TEM
on a JEOL JEM2011 microscope operated at 200 kV, which was cou-
pled with an EDX analyzer (Oxford INCA) for local elemental com-
position determination. The catalysts were dispersed in anhydrous
ethanol, sonicated for 10 min, and dropped onto a carbon-film-
coated copper grid. Particle-size-distribution histograms were con-
structed by measuring at least 300 NPs.
overnight under N . The amount of the catalyst used in each test
2
was ca. 8 mg. The mass spectra were obtained at a heating rate of
À1
1
4
6
0 Kmin from 303 to 1073 K under N2 with a flow rate of
À1
0 mLmin . The mass to charge (m/z) ratio used for analysis was
7 amu, which represents the largest molecular fragment cleaved
from cyclohexene.
XPS spectra were collected on a Perkin–Elmer PHI5000C instrument
with AlKa radiation as the excitation source (hn=1486.6 eV). The
catalyst protected by ethanol was mounted on the sample plate,
degassed in the pretreatment chamber at ambient temperature
overnight in vacuo, and then transferred to the analyzing chamber
Catalytic testing
The partial hydrogenation of benzene was performed in a stain-
less-steel autoclave (500 mL) with a mechanical stirrer under typical
reaction conditions.
À9
where the background pressure was <2ꢂ10 Torr. Because the
[10,13,14]
The autoclave was charged with deion-
Ru3d peak partly overlaps with the C1s peak of contaminant
carbon, all BE values were referenced to the Zr3d5/2 peak of ZrO2
at 182.2 eV with an uncertainty of Æ0.2 eV.
ized water (100 mL) containing ZnSO ·7H O (10 g) and catalyst
4
2
(1.0 g), sealed, purged with H2 for four times to expel air, and
heated under H (3.0 MPa). After the temperature reached 393 K,
2
X-ray absorption spectroscopic data at the Ru K-edge (22.117 keV)
were acquired in the fluorescence mode on the 1W1B beam line of
the Beijing Synchrotron Radiation Facility. The typical electron
beam energy was 2.5 GeV, and the current was 200 mA. The re-
duced catalyst was coated on the scotch tape, pressed into an Al
window, and inserted in the sample stage. X-ray absorption near-
benzene (50 mL) was pumped into the autoclave. The reaction
conditions were temperature of 413 K, H pressure of 5.0 MPa, and
2
stirring rate of 1000 rpm, which is sufficient to eliminate the diffu-
[10,22]
sion effect as proven by our previous works.
The reaction pro-
cess was monitored by taking a small amount of the reaction mix-
ture at intervals, followed by gas chromatographic analysis using
a PEG-20m packed column and a thermal conductivity detector.
edge spectra of the Ru/ZrO catalysts were compared after normal-
2
ization. EXAFS data were analyzed by the IFEFFIT data analysis
Catalytic activity was expressed as weight-specific activity (r ) and,
0
[53]
package according to standard procedures. The background was
removed by extrapolating the pre-edge region onto the EXAFS
region and the c(E) data were normalized with respect to the edge
further, the TOF of benzene. The former was obtained by extrapo-
lating the slope of the concentration–time curve of benzene to
zero reaction time and expressed as the amount of converted ben-
zene per minute per gram of catalyst, whereas the latter was ex-
pressed as TOF=r ꢂM /(dispersionꢂW), in which M and W are
[54]
jump step using the Athena program of the IFEFFIT package.
The normalized c(E) was transformed from energy space to k-space
0
Ru
Ru
2
with c(k) multiplied by k to compensate for the oscillations in the
molar mass of Ru and loading of Ru on the catalyst, respectively.
The selectivity to cyclohexene (SCHE) was calculated as SCHE =nCHE
nCHE+nCHA), in which nCHE and nCHA are the molar percentages of cy-
2
high-k region. Subsequently, k -weighted c(k) data in k-space rang-
/
À1
À1
ing from 3.2 to 12.1 ꢁ (for Ru foil, 3.2 to 14.0 ꢁ ) were Fourier
transformed to the R-space. The processed c(k) data were fitted in
R-space ranging from 0.8 to 2.8 ꢁ using the Artemis program of
(
clohexene and cyclohexane in the product, respectively. The initial
selectivity to cyclohexene (S ) was obtained by extrapolating the
0
[
54]
the IFEFFIT package. The theoretical parameters for the Ru–O
and Ru–Zn pairs were calculated using the FEFF6 code with RuO
slope of the selectivity–time curve of cyclohexene to zero reaction
time, in which the intercept is S0.
2
[55]
and Ru Zn as reference compounds. The values of Debye-Waller
factors are fixed at 0.005 ꢁ . From these analyses, structural pa-
1
6
2
[38]
rameters, such as N, R, and inner potential correction (DE ), were
calculated.
0
Acknowledgements
The hydrophilicity of the catalysts was determined by saturation
adsorption of water followed by gravimetric and FTIR measure-
ments. Prior to water adsorption, the catalysts were dried at 523 K
This work was supported by the National Basic Research Pro-
gram of China (2012CB224804), the National Science Foundation
of China (21073043), the Science and Technology Commission of
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 12
&
11
&
ÞÞ
These are not the final page numbers!
Zhou, Gongbing
