G Model
CATTOD-10275; No. of Pages10
ARTICLE IN PRESS
2
M. Li et al. / Catalysis Today xxx (2016) xxx–xxx
contribution of the support in modifying Au structure, reactant acti-
vation and overall surface reaction mechanism is far from resolved.
In this study we compare the catalytic action of Au nanoparticles on
oxides (␥-Al2O3, ZrO2, TiO2, CeO2, ␣-Fe2O3 and Fe3O4) with distinct
redox character in the hydrogenation of CH O (benzaldehyde)
and NO2 (nitrobenzene) and correlate performance with cata-
lyst structure. We propose surface reaction mechanisms to account
for the role of support reducibility in governing CH O and NO2
activation and product selectivity.
method. Pore volume was measured using the Micromeritics Gem-
ini VII 2390p system. Prior to analysis, samples were outgassed
at 423 K for 1 h in N2. Total pore volume was obtained at a rel-
ative N2 pressure (P/P0) = 0.95. X-ray diffractograms (XRD) were
recorded on a Bruker/Siemens D500 incident X-ray diffractometer
using Cu K␣ radiation. Samples were scanned at 0.02◦ step−1 over
the range 20◦ ≤ 2 ≤ 80◦ and the diffractograms identified against
the JCPDS-ICDD reference standards, i.e. Au (04-0784), ␥-Al2O3
(10-0425), anatase-TiO2 (A-TiO2, 21–1272), rutile-TiO2 (R-TiO2,
21–1276), monoclinic-ZrO2 (M-ZrO2, 37–1784), tetragonal-ZrO2
(T-ZrO2, 50–1089), CeO2 (43–1002), ␣-Fe2O3 (hematite, 33-0664)
and Fe3O4 (magnetite, 19-0629). X-ray photoelectron spectro-
scopic (XPS) analysis was performed on a VG ESCA spectrometer
equipped with monochromatised Al K␣ radiation (1486 eV). The
sample was adhered to conducting carbon tape, mounted in the
sample holder and subjected to ultra-high vacuum conditions
(<10−8 Torr). Full range surveys (Au 4f5/2 and 4f7/2 spectra) were
collected where the binding energies (BE) were calibrated with
respect to the C 1s peak (284.5 eV). The Au 4f spectra were fitted
with abstraction of the Shirley background using the Gaussian-
Lorentzian function in XPSPEAK 41. Gold particle morphology (size
and shape) was examined by transmission (TEM, JEOL JEM 2011)
and scanning transmission (STEM, JEOL 2200FS field emission
gun-equipped unit) electron microscopy, employing Gatan Digi-
tal Micrograph 1.82 for data acquisition/manipulation. Samples for
analysis were dispersed in acetone and deposited on a holey car-
bon/Cu grid (300 Mesh). The surface area weighted mean Au size
2. Experimental
2.1. Catalyst preparation and activation
The supports employed in this study were obtained from
commercial sources (␥-Al2O3 (Puralox, Condea Vista), TiO2 (P25,
Degussa) and CeO2 (Grace Davison)) or synthesised (␣-Fe2O3,
Fe3O4 and ZrO2) as described elsewhere [12,20]. Supported Au
catalysts were prepared by deposition-precipitation using urea
(Riedel-de Haën, 99%) as basification agent. An aqueous solution of
urea (100-fold excess) and HAuCl4 (3–7 × 10−3 M, 400 cm3, Sigma
Aldrich, 99%) was added to the support (10–30 g). The suspension
was stirred and heated (2 K min−1) to 353 K where the pH progres-
sively increased (to 6.5–8.0) as a result of urea decomposition:
NH2 CO NH2 + 3H2OT=353K
→
2NH+4 + 2OH− + CO2
(1)
The solid obtained was separated by filtration, washed with dis-
tilled water until Cl free (from AgNO3 test) and dried (2 K min−1
(d) was based on a count of at least 300 particles according to
)
ꢀ
in 45 cm3 min−1 He at 373 K for 5 h. The catalyst precursors were
sieved (ATM fine test sieves) to mean particle diameter = 75 m and
activated at 2 K min−1 to 423–673 K in 60 cm3 min−1 H2. The cata-
lysts were cooled to ambient temperature and passivated in 1% v/v
O2/He for off-line characterisation.
nidi3
i
d =
(2)
ꢀ
nidi2
i
where ni is the number of particles of diameter di.
2.2. Catalyst characterisation
2.3. Catalytic procedure
Gold content was measured by atomic absorption spectroscopy
from the diluted extract in aqua regia (25% v/v HNO3/HCl). The
pH associated with the point of zero charge (pHpzc) of the support
was determined using the potentiometric mass titration technique
described in detail elsewhere [21]. Temperature programmed
reduction (TPR), H2 chemisorption/temperature programmed des-
orption (TPD), O2 chemisorption and specific surface area (SSA)
measurements were conducted on the CHEM-BET 3000 (Quan-
tachrome) unit equipped with a thermal conductivity detector
(TCD) for continuous monitoring of gas composition and the
TPR WinTM software for data acquisition/manipulation. Samples
were loaded into a U-shaped Pyrex quartz cell (3.76 mm i.d.)
and heated in 17 cm3 min−1 (Brooks mass flow controlled) 5% v/v
H2/N2 at 2 K min−1 to 423–673 K for supported Au catalysts and
to 1073–1273 K for the supports where the effluent gas passed
through a liquid N2 trap. The activated samples were swept with
65 cm3 min−1 N2 for 1.5 h, cooled to reaction temperature (413 K)
and subjected to a H2 (BOC, >99.98%) pulse (10 l) titration pro-
cedure. Samples were cooled to ambient temperature, thoroughly
flushed in N2 (65 cm3 min−1) to remove weakly bound H2 and sub-
hold until the signal returned to baseline. Oxygen (BOC, 99.9%)
pulse (50 l) titration at 413 K post-TPR was employed to deter-
mine the extent of support reduction where any contribution from
Au to total O2 adsorption is negligible [22]. SSA (reproducible to
8%) was recorded in 30% v/v N2/He with undiluted N2 (BOC,
99.9%) as internal standard. At least three cycles of N2 adsorption-
desorption were employed using the standard single point BET
Catalyst testing was carried out at atmospheric pressure, in
situ after activation, in a continuous flow fixed bed tubular reac-
tor (i.d. = 15 mm) at 413–573 K under conditions of negligible
heat/mass transport limitations. A layer of borosilicate glass beads
served as preheating zone, ensuring the organic reactant was
vaporised and reached reaction temperature before contacting
the catalyst (10–40 mg). Isothermal conditions ( 1 K) were main-
tained by diluting the catalyst bed with ground glass (75 m).
Reaction temperature was continuously monitored by a thermo-
couple inserted in a thermowell within the catalyst bed. Reactants
(benzaldehyde (Fluka, ≥98%), nitrobenzene (Riedel-de Haën, ≥99%)
or benzyl alcohol (Riedel-de Haën, ≥99%)) were delivered as an
ethanolic (Sigma Aldrich, ≥99%) solution to the reactor via a
glass/teflon air-tight syringe and teflon line using a micropro-
cessor controlled infusion pump (Model 100 kd Scientific) at a
fixed calibrated flow rate. Reactions were conducted in a co-
current flow of reactant with H2 (BOC, >99.98%, 60 cm3 min−1
)
at GHSV = 2 × 104 h−1. The molar Au to inlet organic molar feed
rate (n/F) spanned the range 1.2 × 10−3–3.7 × 10−3 h. In blank tests,
passage of each reactant in a stream of H2 through the empty
reactor or over the support did not result in any detectable con-
version. The reactor effluent was collected in a liquid nitrogen
trap for subsequent analysis using a Perkin-Elmer Auto System XL
gas chromatograph equipped with a programmed split/splitless
injector and a flame ionisation detector (FID), employing a DB-
1 (50 m × 0.33 mm i.d., 0.20 m film thickness) capillary column
(J&W Scientific). Data acquisition and manipulation were per-
formed using the TurboChrom Workstation Version 6.3.2 (for
Please cite this article in press as: M. Li, et al., Effect of support redox character on catalytic performance in the gas phase hydrogenation