Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
94
Y. Gu et al. / Journal of Catalysis 301 (2013) 93–102
nickel sulfate might, in theory, be a perfect catalyst for this reaction
under aerobic conditions, owing to the combination of moderate
acid sites and enhanced oxidation properties.
resolution of 4 cmꢀ1. Brønsted and Lewis acid sites were identified
by bands at around 1540 and 1450 cmꢀ1, respectively.
FT–IR in transmission mode was employed to identify the con-
formation of nickel sulfate, for which a DTGS detector was used.
Prior to analysis, the sample was ground together with pure, dry
spectroscopic grade KBr to a fine powder, and then, the mixture
was transferred to a compression die, in which it was converted
to a wafer transparent to infrared light through high pressure.
The thickness of the thin wafers was kept constant through con-
trolling preparation parameters. The spectrum of a KBr wafer
was collected initially as a background reference. All spectra in this
section were obtained by subtraction of the corresponding back-
ground reference.
The oxidation properties of 29NiSO4-550 were investigated by
means of H2 TPR–MS. The sample (about 0.1 g) was pretreated at
550 °C for 2 h under flowing helium to ensure complete removal
of impurities. After cooling down to 80 °C, the carrier gas was
switched to a mixture of 10 vol.% H2/N2 (30 mL/min) and the sys-
tem kept at 80 °C till stability of TCD signals. A 10 °C/min temper-
ature ramp was then started from 80 °C up to the final isotherm at
360 °C, kept for 1.5 h, during which TCD signals were recorded as a
function of temperature, and the effluent gas was analyzed by
means of MS. The mass numbers (m/z) 2, 18, 32, 34, 64, and 80
were used for H2, H2O, O2, H2S, SO2, and SO3, respectively.
To understand the difference in catalytic performance of cata-
lysts calcined at various temperatures, TG–DTA–MS was employed
to study structural changes of nickel sulfate during calcination. The
sample (approximately 20 mg) was heated in alumina crucibles
from room temperature to 750 °C at a rate of 10 °C/min, under
high-purity helium (100 mL/min), during which evolved gases
were introduced to a mass spectroscope to obtain evolution curves.
The gas lines between TG and MS were kept at 100 °C to avoid con-
densation of gaseous products.
2. Experimental
2.1. Catalyst preparation
Supported nickel sulfate catalysts were prepared by conven-
tional impregnation of mesoporous silica (specific surface
area = 386.8 m2/g, average pore diameter = 9.6 nm) with an aque-
ous solution of NiSO4ꢁ6H2O, followed by calcination at various tem-
peratures for 2 h in air. A prescribed amount of NiSO4ꢁ6H2O was
dissolved in deionized water, and silica with particle size ranging
from 100 to 150 lm was added into the solution. The slurry was
stirred vigorously and evaporated at 65 °C until dryness. Then,
the obtained solid was dried at 120 °C overnight before calcination.
Hereafter, silica-supported nickel sulfate is denoted as aNiSO4-T,
where ‘‘a%’’ indicates the loading amount and ‘‘T’’ represents the
calcination temperature. Supported ammonium sulfate and nickel
oxide catalysts were prepared following the procedures described
above.
The modification of ammonium sulfate (AS) and cesium carbon-
ate was conducted over 29NiSO4-550 through an incipient wetness
method. The required amount of promoter was dissolved in a vol-
ume of deionized water corresponding to the pore volume of
29NiSO4-550, which was then added dropwise to the 29NiSO4-
550 sample during intensive mixing. The resulting catalyst precur-
sor was dried at 120 °C overnight and subsequently calcined in air
at 550 °C for 2 h.
2.2. Characterization
Nickel content in the liquid product was determined by atomic
absorption spectrometry (AAS) in order to ascertain loss of active
component attributed to flushing with high-temperature steam.
XPS data were obtained using an ESCALab250 electron spec-
trometer from Thermo Scientific Corporation with monochromatic
Several characterization techniques, including nitrogen
adsorption, powder X-ray diffraction (XRD), scanning electron
microscopy (SEM), Fourier transform infrared spectroscopy (FT–
IR), temperature-programmed desorption of CO2 (CO2 TPD),
temperature-programmed desorption of ammonia (NH3 TPD),
temperature-programmed reduction of H2 with mass spectroscopy
(H2 TPR–MS), and thermogravimetric analysis with mass spectros-
copy (TG–MS), were used to determine the structure and acid–base
properties of nickel sulfate catalysts.
150 W Al Ka radiation. Pass energy for the narrow scan was 30 eV.
The base pressure was about 6.5 ꢂ 10ꢀ11 kPa. The binding energies
were referenced to the Si2p line at 103.6 eV or the Al2p line at
74.7 eV from the carrier.
BET surface area, average pore diameter, and pore volume were
measured by an Autosorb-iQ Automated Surface Area and Pore Size
Analyzer using the nitrogen adsorption method. Prior to measure-
ment, all samples were evacuated at 200 °C for 4 h at a pressure of
1.0 ꢂ 10ꢀ3 kPa to ensure complete removal of adsorbed moisture.
XRD was used to identify the crystalline structure of catalysts using
an X’pert PRO MPD diffractometer (PANalytical Company, Nether-
2.3. Catalytic reaction
Gas-phase dehydration of glycerol was conducted in a vertical
fixed-bed reactor (8 mm i.d.) under atmospheric pressure using
1 g of catalyst. A preheater was located on top of the reactor to
vaporize the feed (vaporization temperature 250 °C). In the mean-
time, the reactor had provided an additional preheating zone (spir-
al chute, about 10 cm in length) to ensure complete vaporization.
Prior to reaction, catalysts were pretreated at reaction temper-
ature (340 °C) in flowing nitrogen for about 1 h. The reaction feed,
an aqueous solution with 20 wt.% glycerol, was then introduced
into the system with carrier gas (20 mL/min) from the top of the
preheater by an HPLC pump at a fixed flow rate of 0.13 mL/min.
The reaction was carried out at a gas hourly space velocity (GHSV)
of glycerol of 873 hꢀ1, which was much higher than in previously
published experiments. And the GHSV was defined as the gaseous
volume flow rate of vaporized glycerol at 340 °C and 101.325 kPa
divided by the volume of catalyst (ca. 1 mL). During reaction, the
effluent was collected in a receiver located at the exit of the reac-
tor, which was kept at 0 °C by means of a mixture of water and ice.
A certain amount of ethanol was loaded into the receiver to ensure
efficient capture of products. The reaction was carried out for sev-
eral hours, and products were collected every hour for analysis.
lands) with Cu K
a radiation (40 kV, 40 mA). The X-ray diffracto-
grams were recorded from 5° to 75° at a speed of 5°/min. The
morphology of different catalysts before and after reaction was
studied by S-4800 SEM (Hitachi Company, Japan).
The acidity (basicity) of supported nickel sulfate was evaluated
by TPD of ammonia (carbon dioxide). Detailed operation
procedures have been described previously [8]. Nevertheless, pre-
treatment temperature was adjusted to catalyst calcination tem-
perature. Consequently, all TPD profiles were baseline-corrected
to remove the contribution of impurities emitted at temperatures
higher than the pretreatment temperature. The acidity measure-
ments were also performed by FT–IR after pyridine adsorption.
All spectra were collected using a diffuse reflectance accessory
connected to an infrared spectrometer (Nicolet NEXUS 670) that
was equipped with a Mercury Cadmium Telluride (MCT) detector.
Each spectrum involved the accumulation of 64 scans at
a