Y. Yu et al. / Journal of Molecular Catalysis A: Chemical 395 (2014) 392–397
393
[26]. In contrast, the azeotropy distillation drying is a promising
alternative because of the convenient operation and the recycling
characteristic of organic solvent used.
and then increased programmedly from 100 ◦C to 120 ◦C with a
heating ramp of 5 ◦C min−1
The conversion of MA and the selectivity to GBL were calculated
as follows:
.
In our previous work [15], Cu-CeO2-Al2O3 catalyst has showed
an excellent activity and selectivity for gas-phase hydrogenation
of MA to GBL (GHMG) at atmospheric pressure, however its sta-
bility still needs to be further improved. In this work, effects of the
residual sodium and water in the catalyst precursor on the catalytic
performance and stability of Cu-CeO2-Al2O3 catalyst for GHMG
at atmospheric pressure, and the structure–activity relationships
were investigated.
Conversion of MA = (nMA0 − nMA)/nMA0 × 100%
Selectivity to GBL = (nGBL − nGBL0)/(nMA0 − nMA) × 100%
nMA0 and nGBL0 represented the molar amount of MA and the molar
amount of GBL in the raw material of 20 wt.% MA dissolved in GBL,
respectively. nMA and nGBL represented the molar amount of MA
and the molar amount of GBL in the products after MA hydrogena-
tion, respectively.
2. Experimental
2.3. Characterization of catalyst
Cu-CeO2-Al2O3 catalyst (Cu:Ce:Al = 3:6:10, molar ratio) was
prepared by the co-precipitation method, described in our previ-
ous work [15], in which the aqueous solution of nitrates served
as the metal precursor and 1.0 mol L−1 of Na2CO3 aqueous solu-
tion served as the precipitating agent. The resultant precipitate
was divided equally into six samples by weight. The sample with-
out any washing operation was labeled as w0, and the other three
samples were washed two, four and eight times by centrifugation
using the isopyknic deionized water at 70 ◦C and labeled as w2,
w4 and w8, respectively. The residual two samples were washed
eight times using the isopyknic deionized water at 70 ◦C and then
followed by deep dehydration with two different methods. One
sample was washed two times using absolute ethanol and labeled
as w8a2. Another sample was treated using the azeotropy distil-
lation drying in a heterogeneous azeotropy distillation apparatus
[22] and labeled as w8ba. In a typical experiment, the sample was
mixed with 100 mL of benzene and then kept at 75–81 ◦C under
stirring in oil bath, in which the residual water in the sample and
benzene can be co-evaporated. The evaporated water was collected
in water trap and the evaporated benzene was refluxed by con-
denser until the constant water content in water trap. Finally, the
benzene was evacuated from the flask by the reduced pressure dis-
tillation and the powder was left in the flask. All resultant samples
by the above-mentioned different washing operations were desig-
nated as the catalyst precursors. All catalyst precursors were then
dried overnight at 120 ◦C, and calcined at 550 ◦C for 3 h and 650 ◦C
for 3 h in air successively to obtain the as-prepared catalysts. The
as-prepared catalysts were pressed and then crushed into 20–40
mesh particles before evaluation of their catalytic performance.
The chemical compositions of the as-prepared catalysts were
determined by inductively coupled plasma atomic emission spec-
troscopy (ICP-AES) using a TJA IRIS ADVANTAG 1000 instrument.
The powder XRD patterns were recorded on a Bruker AXS D8
Focus diffractometer with a CuK␣ radiation (ꢀ = 0.15406 nm) oper-
ated at 40 kV × 40 mA, and the diffraction patterns were taken in
the range of 10◦ < 2ꢁ < 80◦ at the scanning rate of 6◦ min−1. SEM
images were recorded on a JEOL JSM-6360LV scanning electron
microscope. The specific surface area was measured by nitrogen
sorption at −196 ◦C with BET method on a NOVA 4200e sur-
face area and pore size analyzer. The Cu surface area (SCu) and
dispersion (DCu = exposed Cu atoms/total Cu atoms) of the fresh
Cu-CeO2-Al2O3 catalysts were measured by N2O chemisorption at
60 ◦C assuming a molar stoichiometry of Cu:N2O = 2 and a surface
atomic density of 1.46 × 1019 Cu atoms m−2 [27]. Raman spectra
were obtained on a Renishaw inVia Reflex Raman microscope with
a CCD detector and the excitation wavelength of 785 nm. Raman
spectra were recorded at 2 cm−1 of increment in the range of
100–1000 cm−1. H2-TPR was carried out in a conventional flow sys-
tem equipment with a TCD detector. 50 mg of sample was heated
programmedly from room temperature until 850 ◦C with a heating
ramp of 10 ◦C min−1 in the atmosphere of 30 mL min−1 of 5 vol. %
H2/N2.
3. Results and discussion
The catalytic performances of Cu-CeO2-Al2O3 catalysts for
GHMG at atmospheric pressure are shown in Fig. 1. As shown in
Fig. 1, Na-rich w0 catalyst exhibited the lowest initial catalytic per-
respectively over w2 catalyst with relatively low sodium content.
Furthermore, both the conversion of MA and the selectivity to GBL
could reach 100% over w4 and w8 catalysts. According to ICP-AES
analysis shown in Table 1, the sodium contents of the as-prepared
w0, w2, w4 and w8 catalysts were 6.2 wt.%, 1.5 wt.%, 0.25 wt.% and
performance of Cu-CeO2-Al2O3 catalysts for GHMG. The negative
effect of the residual sodium in Cu-based catalyst on its catalytic
performance for methanol synthesis from CO2 hydrogenation was
also observed [19]. As shown in Table 1 and Fig. 1, three catalysts
of w8, w8a2 and w8ba, with the same sodium content and the
different water content in the precipitate before drying, showed
excellent initial catalytic performances with 100% of conversion of
MA and 100% of selectivity to GBL, but the catalytic stability of Cu-
CeO2-Al2O3 catalysts for GHMG was much different and followed
2.2. Gas-phase hydrogenation of maleic anhydride
Gas-phase hydrogenation of maleic anhydride to ␥-
butyrolactone (GHMG) over Cu-CeO2-Al2O3 catalyst was carried
out under atmospheric pressure in a quartz tubular fixed-bed
reactor (13 mm I.D., 650 mm L.). In a bottom-up sequence, the
reactor was packed with 1 mL of quartz sand pretreated at 600 ◦C in
air, 6 mL of catalyst (height = 45 mm) and then 15 mL of pretreated
quartz sand for the complete gasification of raw material (MA
dissolved in GBL, 20 wt.%). The catalyst was reduced in situ by
60 mL min−1 of 5 vol.% H2/N2 at 380 ◦C for 7 h (designated as the
fresh catalyst). Then the temperature was decreased until 240 ◦C,
and the reactor was fed by 30 mL min−1 of H2 and 0.2 h−1 of
LHSV of raw material. The products were collected in the conical
beakers cooled by ice-bath at the interval of 1 h and analyzed by
Perkin-Elmer Clarus 500 gas chromatograph equipped with a FID
detector and a SE-54 capillary column (25 m × 0.32 mm × 1.0 m).
Both temperatures of injector and detector were kept at 220 ◦C,
and the temperature of column oven was kept at 100 ◦C for 1 min