CHEMCATCHEM
FULL PAPERS
Catalyst characterization
under ambient conditions in the form of self-supported wafers.
In situ measurements were performed by placing the sample into
a quartz capillary connected to a flow system and attached to
The BET surface areas and pore size distributions were determined
from the analysis of the N adsorption–desorption isotherms of the
2
a heating wire. The sulfidation was performed under 10% H S in
2
oxidic precursors at ꢀ1968C. A PMI Automated BET Sorptomatic
H at 4008C for 1 h, and spectra were recorded every 1008C. After-
2
1
900 Series instrument (Thermo Finnigan) was used to perform the
wards the sample was cooled to RT, and the flow was switched to
N2 to record further spectra. Finally, the sample was heated to
experiments. Prior to the adsorption, the supported samples were
evacuated at 2508C for 2 h, and the unsupported catalyst precur-
sor was evacuated at 1208C for 4 h. Elemental analysis was per-
formed at the Microanalytical Laboratory of the TU Mꢄnchen.
4
008C again under a flow of synthetic air to acquire the final
spectra.
The diffuse reflectance technique was applied to collect UV/Vis/
near IR spectra by using an Avantes AvaSpec-2048 fiber optic spec-
trometer equipped with a CCD detector array. A combined deuteri-
um and halogen light source, Ava Light-DH-S-BAL, was used in
combination with a fiber optic cable FCR-7V400-2-SR-HT. Spectra of
oxide catalyst precursors and reference materials were recorded
under ambient conditions. The samples were placed as powders in
a Teflon sample holder that provided 1 mm sample thickness.
NO adsorption was performed as a pulse experiment at RT to
probe the concentration of adsorption sites and average edge dis-
persion in the studied metal sulfide catalysts. A detailed descrip-
[15]
tion of the experiment can be found elsewhere.
The crystal structure of the samples was determined by analysis of
the powder XRD patterns. The crystallographic phases were identi-
[20]
fied by using the inorganic crystal structure database (ICSD). The
The structural properties of the oxide catalyst precursors, the sul-
fided catalysts, and the materials during the catalyst sulfidation
were studied in situ by X-ray absorption spectroscopy (XAS) at the
X1 beamline at Hasylab, DESY, Hamburg, Germany. The data set
was completed with experiments performed on the BM 26A—
DUBBLE, (Dutch-Belgian) beamline at the ESRF, Grenoble, France.
Spectra were recorded in the transmission mode at the Mo K-edge
XRD patterns were collected by using a Philips X’Pert System (CuKa
radiation, 0.1542 nm) using a NiK filter and a solid-state detector
b
(
X’Celerator). The operating conditions were 45 kV and 40 mA. The
prepared catalysts were measured with a step size of 0.0178 and
a scan time of 115 s per step. Selected reference materials were
measured over 5 min with a step size of 0.0178 and a scan time of
1
0 s per step.
(20000 eV) using Si (311) crystals and at the Ni K-edge (8333 eV)
The Scherrer equation was used to determine the stacking degree
of the sulfide slabs in the unsupported catalyst in the knowledge
that the diffraction at 2q=148 corresponds to the (002) plane
with an interplanar distance of 6.1 ꢂ (distance between the Mo
using Si (111) crystals in the monochromator, respectively. The
contributions of the higher harmonics were minimized by detun-
ing the second crystal of the monochromator to 60% of the maxi-
mum intensity. Energy calibration was performed with Mo and Ni
metal foils, respectively, measured simultaneously with the sam-
ples. The samples were prepared as self-supported wafers placed
in a stainless-steel in situ flow cell. The measured reference com-
pounds were mixed with cellulose to achieve a total absorption of
mx=1.5. The spectra of the oxide precursors, sulfided catalysts, and
reference compounds used for EXAFS analysis were collected in He
layers in MoS ) [Eq. (1)]:
2
K ꢃ l
L ¼
ð1Þ
Dð2qÞ ꢃ cos q
in which L is the mean size of ordered (crystalline) domain, K is the
Scherrer shape factor (0.9), l is the X-ray wavelength used, q is the
measured Bragg angle, and D(2q) is the line broadening [rad]. To
calculate the line broadening Equation (2) was used:
flow and at liquid N temperature (LNT) to minimize thermal vibra-
2
tions. At least two spectra of each sample were averaged to en-
hance the signal-to-noise ratio. After the EXAFS measurements of
the oxidic catalyst precursors, quick XAFS scans with a resolution
of 180 s were recorded continuously during the catalyst sulfidation
Dð2 qÞ ¼ FWHMꢀ0:1
ð2Þ
ꢀ
1
under a flow of 10% H S in H with a heating rate of 58Cmin up
2
2
to 4008C followed by an isothermal period of 1 h.
in which FWHM is the full width at the half maximum [rad], and
.1 is the typical instrument parameter.
Quick XAFS data were processed by using XANES dactyloscope
0
[61]
software (version 2012/4).
The scattering contributions in the
Electron microscopy measurements were performed in the trans-
mission mode coupled with selective area electron diffraction
pre- and postedge of the background were removed from the X-
ray absorption by using a third-order polynomial function, and all
spectra were normalized to the average postedge height of one.
The XANES and EXAFS data were analyzed by using IFEFFIT using
(
TEM-SAD) and in scanning mode at high resolution coupled with
energy-dispersive X-ray spectroscopy (HR-SEM-EDX). Samples of
the catalysts were ground and ultrasonically dispersed in ethanol.
Drops of the dispersion were applied to a copper-carbon grid. TEM
and SAD measurements were performed by using a JEOL JEM-2011
TEM instrument with an accelerating voltage of 120 keV. For the
HR-SEM and EDX mapping measurements, a high-resolution field
emission (FE) SEM JSM 7500 F (JEOL) instrument with EDX (Oxford)
was used. The micrographs were recorded with a secondary elec-
tron imaging (SEI) detector and an accelerating voltage of 5 keV.
[62,63]
the Horae package (ATHENA and ARTEMIS, version 1.2.11).
After the removal of the background absorption and normaliza-
3
tion, the oscillations were weighted with k and Fourier trans-
ꢀ1
formed within the limit of 3.5–14 ꢂ for the oxide precursor and
ꢀ1
k=3–12.0 ꢂ for the sulfided catalyst. The local environments of
the Mo and Ni atoms in the sulfided catalysts were determined in
k space from the EXAFS data. Single and multiple scattering contri-
butions for MoꢀS, MoꢀMo, NiꢀS, and NiꢀNi (phase shifts and back-
[64]
scattering amplitudes) were calculated with FEFF by using the
geometries of the crystallographic information files (cif) of the
Raman spectra were obtained with a 514 nm Ar laser by using a Re-
nishaw Raman Spectrometer (Type 1000) equipped with a charge-
coupled device (CCD) detector and a Leica microscope DM LM.
Prior to the measurements, calibration was performed with
[20]
2
ICSD. The amplitude reduction factor S0 was determined from
reference compounds and found to be 1.16 for Mo and 1.00 for
Ni. For Ni-MoS /g-Al O and Ni-MoS /unsupported, the EXAFS data
2 2 3 2
at the Mo K-edge and at the Ni K-edge were simultaneously fitted
ꢀ1
a Si(111) crystal. The wavenumber accuracy was within 1 cm . The
oxidic catalyst precursors and reference materials were analyzed
ꢁ
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 485 – 499 497