276
G. Mitran et al. / Journal of Molecular Catalysis A: Chemical 396 (2015) 275–281
2. Experimental
2.1. Catalysts preparation
10 ◦C/min. The ammonia desorbed was bubbled through a solution
of sulfuric acid. The acid in excess was titrated with a solution of
NaOH, the amount of ammonia desorbed being then calculated. The
ammonia desorbed at temperatures lower than 350 ◦C accounted
for the weak and medium-strength acid sites while that desorbed
in the temperature range from 350 to 500 ◦C, for the strong acid
sites.
Al2O3 support was prepared from Al(NO3)3·9H2O (Fluka Ana-
lytical) by precipitation with ammonium carbonate (Lachema)
at controlled pH of 6.5. SiO2 support was prepared from
Na2SiO3·5H2O (Sigma–Aldrich) by precipitation with ammonium
chloride (Lach-Ner) at controlled pH of 6.5. SiO2–Al2O3 supports
with three different SiO2/Al2O3 mol ratios, i.e. 0.4, 1.0 and 2.5,
labeled Si(0.4)Al, Si(1.0)Al and Si(2.5)Al, respectively, were pre-
pared by coprecipitation using Al(NO3)3·9H2O and Na2SiO3 as
precursors. Ammonium chloride was used as precipitating agent
at controlled pH of 6.5 when both Al and Si ions were completely
coprecipitated. All the precipitates were separated by filtration
and washed with distilled water, dried in air at 100 ◦C for 12 h
and, finally, calcined at 600 ◦C for 4 h. WO3 was introduced at
two concentrations, 5 and 10% by weight, via incipient wetness
impregnation of supports with aqueous ammonium tungstate
(Schering–Kahlbaum) solutions containing appropriate amounts
of tungsten. After impregnation, the samples were dried in air
at 100 ◦C for 12 h and then calcined at 600 ◦C for 4 h. The 5 wt%
WO3/support and 10 wt% WO3/support samples were labeled 5W-
Al, 5W-Si, 10W-Al, 10W-Si, 10W-Si(0.4)Al, 10W-Si(1.0)Al and
10W-Si(2.5)Al, where Al and Si stand for Al2O3 and SiO2, respec-
tively.
2.3. Catalytic test
The esterification reactions of acetic acid (Chimactiv, 99.5%)
with n-butanol (Riedel-de Haën, 99.5%) were performed in a 150 mL
two-neck flask equipped with a condenser and an additional port
for sample withdrawal. The above assembly was heated using a
thermostated hotplate. The reaction was carried out at 100 ◦C with a
molar quantity of acetic acid of 0.09 and an n-butanol-to-acetic acid
molar ratio varied from 1 to 3. Cyclohexane (Riedel-de Haën, 99.5%)
was always added to the reaction mixture for water removal, the
cyclohexane-to-acetic acid molar ratio being kept at 1. The amount
of catalyst was varied between 0.5 and 1.3% of the mass of mix-
ture charge in the reaction. All the experiments were conducted
at a speed of agitation of 600 rpm to avoid diffusional limitations
as reported elsewhere [11,14]. All the catalysts used in the reac-
tion were in the powder form. Pre-adsorption experiments were
performed by premixing the catalyst with one of the reactants or
both at room temperature for 24 h followed by heating to 100 ◦C
and charging the preheated remaining reactant. Samples from the
organic layer were withdrawn at regular intervals and analyzed
with a Thermo Finnigan chromatograph using a DB-5 column and
a flame ionization detector. Under the employed conditions of
reaction butyl acetate was the only product detected. The mass
balances, calculated after a reaction time of 120 min, were always
higher than 95%.
2.2. Catalysts characterization
The crystalline phases were investigated by the X-ray diffrac-
tion (XRD) method. XRD patterns were obtained with a Philips PW
˚
3710 type diffractometer equipped with a Cu K␣ source (ꢀ = 1.54 A),
operating at 50 kV and 40 mA. They were recorded over the 5–70◦
angular range with 0.02◦ (2ꢁ) steps and an acquisition time of 1
s per point. Data collection and evaluation were performed with
PC-APD 3.6 and PC-Identify 1.0 software.
The surface areas of the catalysts were measured from the
adsorption isotherms of nitrogen at −196 ◦C using the BET method
with a Micromeritics ASAP 2020 sorptometer. The samples were
first out-gassed at 300 ◦C for 4 h in the degas port of the
adsorption apparatus. The pore size distribution curves were cal-
culated using the desorption branch of the isotherms with the
Barrett–Joyner–Halenda (BJH) method.
3.1. Characterization of the catalysts
The XRD patterns of supported WO3 on alumina and silica-
alumina (Fig. 1) showed only broad lines corresponding to
␥-alumina (PDF 10-425) and, in addition for silica-alumina-
supported samples, a halo between 15 and 35◦ (2ꢁ) due to
amorphous silica. No signal of WO3 crystals on alumina and
silica-alumina supports was observed suggesting that either the
dispersion of WO3 on the support was high or the WO3 crystallite
size was very small. On the other hand, lines corresponding to both
monoclinic (PDF 83-950) and hexagonal (PDF 33-1387) WO3 crys-
tals were present on silica support in addition to the very broad
peak between 15 and 35◦ (2ꢁ).
The textural properties of the catalysts are summarized in
Table 1. It can be observed that the specific surface areas were rela-
tively high due to the dispersion effect of porous carrier and ranged
from 184 to 377 m2 g−1. Both the surface area and the pore volume
depended on the nature of support and WO3 loading. Thus, the sur-
face areas and pore volumes of alumina-supported samples were
lower than those of silica-supported ones while, as expected, they
had intermediate values for the silica-alumina-supported samples.
At the same time, the surface areas and pore volumes of the sam-
ples with lower WO3 loading were higher compared to the samples
with higher WO3 loading. All the catalysts displayed typical type
IV nitrogen adsorption/desorption isotherms (according to IUPAC
classification) with a clear hysteresis loop characteristic of meso-
porous materials with cylindrical pores [20], as showed in Fig. 2.
The catalysts displayed well-defined pore size distributions (Fig.
S2), the average pore diameters being presented in Table 1. The pore
size distributions have been obtained from the desorption branch
Qualitative and quantitative electron probe microanalyses were
performed using a Philips XL 30 ESEM (Environmental Scanning
Electron Microscope) having EDX (Energy Dispersive X-ray) ana-
lyzer. An accelerating voltage of 20 kV was used. The powder
samples were fixed on a holder and sample compartment was evac-
uated in order to prevent the electrical charging. Electron beams
were very finely focused. Therefore, elemental analysis of very
small area on specimen surface was done in so-called spot mode. In
this case, the diameter of the microprobe beam was about 0.5 m
which can penetrate volume of the surface layer at about 5 m3.
The spot mode analysis was carried out for three points of the
same sample. Simultaneously, the surface of the sample was photo-
graphed, the micrographs obtained being presented in Fig. S1.
The acidity of the catalysts was estimated by temperature-
programmed desorption of ammonia (NH3-TPD). About 0.1 g of the
catalyst sample was dehydrated at 500 ◦C in dry air for 1 h and
purged with N2 for 0.5 h. The sample was then cooled down to
100 ◦C under the flow of N2, and NH3 was supplied to the sample
until its saturation. For desorption of the physisorbed ammonia, a
nitrogen stream was passed over the sample, at the same temper-
ature, until no more NH3 was observed in the exit flow. Finally, the
chemisorbed NH3 was desorbed in a N2 flow by increasing the tem-
perature successively up to 350 ◦C and 500 ◦C with a heating rate of