Mehran GHIACI et al. / Chinese Journal of Catalysis, 2010, 31: 759–764
1
catalysts using 29Si, 27Al, 31P, 23Na, and H MAS NMR spec-
that the loading of H3PO4 onto Al-MCM-41 did not change the
regular arrangement of the uniform channels of the support
[17]. Interestingly, by loading approximately 30 wt% phos-
photungstic acid onto Al-MCM-41, the surface area of the
support decreases by less than 20% [10], but loading the same
support with approximately the same weight percent of H3PO4
results in the surface area decreasing to a larger extent. The
crystalline structure of the phosphotungstic acid (Keggin units)
and the amorphous structure of the phosphoric acid could
explain this difference. The 31P MAS NMR spectra of the 30
wt% H3PO4/Al-MCM-41(200) showed two peaks at δ = –10
and 0, which may correspond to the P atom in pyrophosphoric
acid or the terminal and monomeric PO43– groups, respectively.
From an analysis of the Si, Al, and P NMR signals, the forma-
tion of Si–O–Al–O–P bridges seems to be favored at the sur-
face of the Al-MCM-41 catalysts. The FT-IR spectra for the 30
wt% H3PO4/Al-MCM-41(70) containing adsorbed pyridine
shows the contribution of pyridine adducts from 1 650–1 450
cm–1. The formation of pyridinium ions, as shown by the ad-
sorption at 1 545 and 1 490 cm–1, is characteristic of Brönsted
acid sites and both Brönsted and Lewis acid sites, respectively.
A comparison of the spectra revealed that 30 wt% H3PO4/
Al-MCM-41(70) possesses a greater amount of acid sites than
the other catalysts. The concentration of Brönsted and Lewis
acid sites was calculated after evacuation at 473 K using the
extinction coefficient of the bands of Brönsted and Lewis acid
site adsorbed pyridine [18]. This revealed the presence of high
amounts of Brönsted and Lewis acid sites in 30 wt%
H3PO4/Al-MCM-41(70).
troscopy [15]. We have shown that H3PO4 loading on
AlMCM-41 induces a drastic decrease in the specific surface
area without any structural distortion of MCM-41. Therefore,
the main reason for choosing these catalysts for the t-butylation
reaction was that this reaction probably only occurs on the
external surface of the catalysts, and the acidity of the catalysts
as well as thermodynamic factors greatly affect the conversion
and selectivity.
1 Experimental
1.1 Preparation of the catalysts
The mesoporous molecular sieves Al-MCM-41 and
H3PO4/Al-MCM-41 were prepared in accordance with the
reported methods [16].
1.2 Catalytic activity
Catalytic tests were performed with the 5 wt% H3PO4/Al-
MCM-41(20) catalysts. The vapor-phase tests were carried out
using approximately 0.5 g of catalyst (20–40 mesh) in a feed
Pyrex reactor (i.d. 8 mm). The reactor was inserted into an
electric oven controlled by a J thermocouple and operated at
atmospheric pressure, and the temperature range was 383–493
K. The feed was tert-butyl alcohol (TBA):phenol (2:1 molar
ratio), which was introduced by an infusion pump (GENIE,
Kent Scientific Corporation) and nitrogen was used as the
carrier gas by regulation with a mass-feed meter. The catalyst
was previously activated in situ for 1 h at 673 K using a 25
ml/min N2 feed. In each experiment, the products were col-
lected after 8 h by condensing them in a trap containing salt-ice
and cooled to 258 K. Analyses were done off-line using a
Shimadzu gas chromatograph (model 14A) equipped with a
wide bone OV-17 (60 m) capillary column and an FID detector.
2.2 Catalytic studies
The vapor phase t-butylation of phenol with TBA was car-
ried out over Al-MCM-41 with SiO2/Al2O3 = 20, 70, and 150,
containing 5, 10, and 20 wt% H3PO4, respectively, from 383 to
493 K. The molar ratio chosen in these experiments for TBA to
phenol was 2:1. The major products were 4-tert-butyl phenol
(4-TBP), 2-tert-butyl phenol (2-TBP), and 2,4-di-tert-butyl
phenol along with less than 5% other products that were not
identified. We did not obtain any tert-butyl phenyl ether
(TBPE) even at high phenol conversion under the best reaction
conditions.
2 Results and discussion
2.1 Characterization
The physicochemical characterization of the catalysts was
similar to our earlier report [15]. However, other important
conclusions that we made from the characterization results are
outlined here. The XRD data indicated that the MCM-41
samples exhibit an ordered hexagonal structure characterized
by an intense reflection peak at a d spacing of 3.8 nm [16]. The
intensity of this low angle reflection peak did not change as the
H3PO4 loading increased. The specific surface area of the
samples calculated from the BET method ranges between
1 038 and 668 m2/g. Although the H3PO4 loading induces a
drastic decrease in the surface area, no structural distortions are
evident for Al-MCM-41. Consequently, it seems reasonable
The conversion of phenol over Al-MCM-41 with SiO2/
Al2O3 = 20, 70, and 150 was improved with an increase in
temperature. The maximum conversion was observed at 463 K
and when the temperature reached 493 K the conversion de-
creased by a few percent (Table 1).
We also observed that TBPE formed at 383 K but it was
completely absent at higher temperatures (small amounts of
ether formed at 413 K). In the early stages of the reaction, a
reversible O-alkylation occurs at a high rate because of the low
activation energy on almost all the acid sites (independent of
acid strength). The decrease in TBPE selectivity with increas-