158
D. Das et al. / Journal of Catalysis 223 (2004) 152–160
Table 3
Table 4
◦
Synthesis of Bisphenol-A with different catalysts at 70 C
Synthesis of Bisphenol-A with different mesoporous silica catalysts at
◦
100 C
Catalyst
Phenol conversion
(%)
Selectivity to
ꢀ
p,p -Bisphenol-A (%)
Catalyst
meq of sulfur
Phenol
Selectivity to
ꢀ
(per g of final solid) conversion p,p -Bisphenol-A
−
0
−
(%)
(%)
H-beta (Si/Al = 50)
HY (Si/Al = 11)
HZSM-5 (Si/Al = 80)
MPS-8
5.2
6.8
2.7
29.6
35.3
52.8
−
MCM-41
MPS-6
MPS-8
MPS-12
MPS-9
MPS-17
0
0
−
9.9
0.81
1.06
1.63
1.78
2.25
28.6
35.3
23.5
24.8
27.5
91.2
88.6
91.6
90.8
89.8
91.7
89.2
Amberlite-120
Reaction time 24 h, phenol:acetone molar ratio 5:1.
MPS-16
MPS-14
MPS-13
1.44
3.03
4.63
29.1
38.2
29.3
89.7
82.6
85.4
alized MCM-41 silica and other acidic zeolites are given in
Table 3. No conversion of phenol was noted in the absence of
the catalyst. Acidic zeolites like H-ZSM-5, H-Y, and H-beta
showed negligible activity due to pore-size constraints and
also mostly forms undesirable oligomeric products. The sul-
fonic acid-functionalized MCM-41 silica, MPS-8, showed
about 29% phenol conveꢀrsion with more than 92% selec-
tivity toward desired p, p -Bisphenol-A. On the other hand,
sulfonic acid resin Amberlite-120 showed slightly higher
phenol conversion (35.3%) with about 89% selectivity to the
p, pꢀ product. Apparently it appears that Amberlite-120 has
a better conversion than MPS-8. However, if the acid sites
in the two catalysts were compared we can see that MPS-8
with sulfur loadings of 1.06 meq/g has a comparable activ-
ity as Amberlite-120, which has 4 times (4.4 meq/g) more
sulfur loadings. It is noted here that the maximum achiev-
able phenol conversion at a phenol/acetone molar ratio of 5
is 40%.
Amberlite-120
4.4
39.1
86.9
Reaction time 24 h, phenol:acetone molar ratio 5:1.
fur loadings did not generally produce better catalysts. This
can be explained by considering the fact that different sulfur
species were present in these catalysts after oxidation with
H2O2. Sulfur K-edge XANES studies showed that samples
containing higher amounts of sulfur are difficult to oxidize
completely and a part of the thiol groups was incompletely
oxidized or remained at lower valent states, which are cat-
alytically inactive. For maximum catalytic activity, it is nec-
essary that the sulfur species are in the +5 state as sulfonic
acid. However, the appearance of a small peak at around
2471 eV in the XANES spectra indicates the presence of
disulfide groups. It was noted that when sulfur loadings ex-
ceed 1.5 meq/g solid a part of the sulfur atoms remains in
the reduced form even after prolonged oxidation. It appears
that at higher surface coverage nearby thiol groups interact
easily to form catalytically inactive disulfide species.
It can also be seen from Table 4 that an MCM-48-based
MPS-16 catalyst that contains slightly higher sulfur load-
ings (1.44 (meqS)/g) than MCM-41-based MPS-8 sam-
ple (1.06 (meq S)/g) showed slightly lower phenol conver-
sion. This again indicates that in samples with higher sul-
fur loadings a part of thiol groups are incompletely oxi-
dized. Furthermore, like MCM-41-based samples it appears
that at higher sulfur loadings oxidation of the thiol groups
was more difficult for MCM-48-based samples also. Con-
sequently sample of highest sulfur loading did not give
the highest catalytic activity. For example, MPS-13 con-
taining 4.63 (meq S)/g showed 29.3% phenol conversion,
while MPS-14 containing 3.03 (meq S)/g showed 38.2%
phenol conversion. Furthermore, the selectivity to the de-
sired p, pꢀ-Bisphenol-A was found to be slightly lower over
the MCM-48-based catalysts than that over MCM-41-based
catalysts. It is probably attributed to the fact that the pore di-
ameter of mesopores in cubic MCM-48 is not straight and
not as narrow-ranged as that of hexagonal MCM-41.
The other isomer of condensation, o, pꢀ-Bisphenol-A,
was also detected; however, by-products like chroman and
trisphenols were not detected at this reaction temperature.
Recently, the synthesis of Bisphenol-A over heteropolyacid-
encapsulated MCM-41 has been reported [2]. 12-Tungsto-
phosphoric acid-encapsulated MCM-41 was found to be cat-
alytic active only at temperatures 120 ◦C or above. How-
ever, phenol conversion was reported to be much lower than
that obtained with MPS samples. In addition, due to the
higher reaction temperatures, several by-products like alky-
lated phenols and chroman derivatives were formed, and the
selectivity to p, pꢀ-Bisphenol-A was less than 70%. Yadav
and Kirthivasan [1] compared the activity of different ion-
exchange resins like Amberlyst-15, -31, -XE-717p, and 12-
tungstophosphoric acid supported on K-10 clays. Activity of
K-10 clay-supported heteropolyacid was found to be sim-
ilar to that of Amberlyst-31, but much superior selectivity
to Bisphenol-A was obtained with the resin catalysts. How-
ever, the thermal stability of the resin catalysts was found to
be inferior to that of the clay catalysts, which can be used at
temperatures as high as 300 ◦C.
Table 4 shows the phenol conversion and selectivity to
p, pꢀ-Bisphenol-A at 100 ◦C for different MCM silica cata-
lysts. Phenol conversion was increased with the increase in
reaction temperature, but the activity of the catalysts was not
correlated to their sulfur content. Catalysts with higher sul-
To compare the catalytic behavior of the resin and silica-
based catalysts, the activity and selectivity of MPS-8 and
Amberlite-120 at 100 ◦C with time-on-stream are given in