CHEMCATCHEM
FULL PAPERS
cient and versatile ways to modulate the chemical properties
were calcined in static air at 773 K for 4 h (heating rate=
[
12]
À1
5
Kmin ). Wet impregnation of the support (1 g) was performed
of CeO2, might enhance the reactivity of ceria, allowing op-
eration at milder conditions.
using 9m HNO3 aqueous solution of cerium ammonium nitrate
3
(
2.0 cm ; Sigma–Aldrich,>98%). The impregnated catalyst was col-
Finally, the quantitative rationalization of the influence of
lected by filtration, washed with deionized water, dried at 333 K for
the ceria loading on the TiO -supported catalysts remains an
2
1
5
2 h, and calcined in static air at 773 K for 4 h (heating rate=
Kmin ). The Lindlar catalyst evaluated in this study was provid-
open aspect at this stage. No reflections associated with CeO2
À1
were observed in the XRD patterns of the 5 and 10 wt% CeO2/
ed by Alfa Aesar (ref: 043172) and contained 4.5 wt% Pd and
3.4 wt% Pb deposited on CaCO3.
TiO samples, which indicates the very high dispersion of the
2
ceria phase on the titania carrier. XPS measurements show that
3+
the surface concentration of Ce increases linearly with the
decreased ceria loading (Figure S2). A higher amount of
oxygen vacancies is detrimental for the alkyne hydrogenation
Catalyst characterization
The Ce content was determined by ICP-OES by using a Horiba
Ultra 2 instrument after dissolution of the catalysts in a HF/HCl/
[
4a]
activity. However, the overall reaction rate should be deter-
mined by the interplay between dispersion (a priori beneficial,
HNO solution. The absence of typical hydrogenation metals was
3
higher at a low CeO loading) and degree of surface reduction
2
verified by XRF by using an Orbis Micro-EDXRF analyzer, equipped
(
detrimental, higher at a low CeO loading). The reaction rate
with a 35 kV Rh anode and a silicon drift detector. N sorption at
2
2
in acetylene hydrogenation per mole of total ceria is very simi-
lar over 5 and 20 wt% CeO /TiO , which suggests that the ben-
77 K was measured by using a Micromeritics Tristar II analyzer.
Prior to the analysis, the samples were evacuated at 473 K for 10 h.
TEM studies were performed by using an FEI Titan Cs 80–300 mi-
croscope equipped with a field-emission gun, a Gatan Tridiem
filter, and an energy-dispersive X-ray (EDX) analyzer and operated
at 300 keV. The samples were prepared by dry deposition onto
a holey-carbon-covered Cu grid. The spherical aberration was cor-
rected by using a CEOS Cs-corrector. HRTEM images were pro-
cessed to obtain the power spectra, which were used to measure
interplanar distances and angles for phase identification. For the
EFTEM of O, Ti, and Ce, the O K edge at 530 eV, Ti L edge at
2
2
efits derived from the higher dispersion of active phase in the
former catalyst are offset by the higher degree of ceria reduc-
tion in the smaller particles. For the hexyne hydrogenation, the
rate was two times higher over 5 wt% CeO /TiO , which sug-
2
2
gests the occurrence of less surface reduction in liquid-phase
operation. The above discussion is speculative but it reflects
the complexity related to the amount and type of exposed Ce
sites and calls for extensive studies to unravel detailed struc-
ture–performance relationships for these catalysts for gas- and
liquid-phase hydrogenation.
4
56 eV, and Ce N edge at 110 eV were analyzed. CeO and TiO
2 2
phases were assigned according to ICSD number 246969 and
3711, respectively. The XRD measurement was performed by
6
using a Bruker AXS D8 ADVANCE DAVINCI diffractometer equipped
with a Ni filter and a LYNXEYE position sensitive detector (CuKa1+2
radiation) in Bragg–Brentano geometry with a fixed divergence slit.
The XRD data were analyzed by whole powder pattern fitting ac-
Conclusions
We have demonstrated the highly selective character of sup-
[13a]
cording to the Rietveld method by using the TOPAS software.
Peak profiles were fitted by convolution of an instrumental contri-
ported CeO catalysts in the semi-hydrogenation of alkynes.
2
The materials were prepared by simple impregnation of the
supports with a Ce salt in an acidic medium, and the selectivity
to the alkene was not affected by the type of carrier and the
Ce loading. The ethylene selectivity in the gas-phase hydroge-
nation of acetylene reached 90%, and the results for the three-
phase hydrogenation under flow conditions were even more
impressive: full stereo- and chemoselectivity to the cis-alkene
was attained over a wide range of acetylenic substrates, that
is, oligomers and overhydrogenated products were not ob-
served, and the performance was stable for several hours on
[13b]
bution following the Fundamental Parameters approach,
with
[13c]
a sample contribution based on the Double–Voigt approach.
XPS was performed at RT by using non-monochromatized AlKa
(1486.6 eV) excitation and a hemispherical analyzer (Phoibos 150,
SPECS). The sample mounted on conductive carbon tape was
transferred to the spectrometer chamber under regular air expo-
sure. The binding energy scale was calibrated by internal referenc-
4
+
ing to the Ce3d U’’’ (916.7 eV) hybridization state of Ce to correct
for small charging effects. The Ce/Ti ratio of the sample was calcu-
lated from the Ce3d and Ti2p states after Shirley background sub-
traction by normalizing the peak areas with the corresponding
cross sections, the transmission function of the lens system at the
appropriate photoelectron kinetic energy, and with an inelastic
mean free path correction to account for the different information
stream. The CeO -based catalysts required demanding condi-
2
tions of temperature and pressure (413 K, 90 bar) because of
the intrinsic inability of ceria to activate hydrogen. However,
selectivity-wise, supported ceria outperformed the state-of-the-
art Lindlar catalyst. Consequently, these results enable the use
of a cheap metal oxide for the production of olefinic com-
pounds in fine chemical and pharmaceutical applications.
depth of the Ce3d and Ti2p states. H -TPR was performed by
2
using a Thermo TPD/R/O 1100 unit equipped with a thermal con-
ductivity detector. The catalyst (ꢁ50 mg) was loaded in the quartz
3
À1
microreactor (11 mm i.d.), pretreated in He (20 cm min ) at 473 K
for 30 min, and cooled to 323 K in He. The analysis was performed
3
À1
in 5 vol% H /He (20 cm min ), and the temperature was ramped
2
À1
from 323 to 1173 K at 5 Kmin . Differential heats of adsorption
Experimental Section
were measured by using a SETARAM MS70 Calvet calorimeter. The
calorimeter was combined with a custom-made high-vacuum and
Catalyst preparation
[13d]
gas dosing apparatus described elsewhere.
The sample was pre-
Prior to use, TiO -anatase (Sigma–Aldrich, 99.7%), Al O (Saint-
treated in O at 523 K and 1 bar for 1 h. After evacuation, a second
2
2
3
2
Gobain NorPro,>99.5%), and ZrO (Saint-Gobain NorPro,>99.8%)
pretreatment in H at 0.1 bar and 423 K for 1 h was performed to
2
2
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 8
&
6
&
ÞÞ
These are not the final page numbers!