Triphase Catalysis over Titanium-Silicate Molecular Sieves under Solvent-free Conditions: I. Direct Hydroxylation of Benzene

Article abstract of DOI:10.1006/jcat.1998.2131

Under solvent-free triphase conditions (solid catalyst TS-1, organic substrate and aqueous H₂O₂, S-L-L) a significant enhancement in the reaction rate (ca by 15-25 times) during the oxidation of benzene, compared to that obtained under conventional biphase system using an organic solvent (solid TS-1 catalyst in the presence of a cosolvent to homogenize the organic and aqueous layers, S-L), was observed. Time-dependent studies in the hydroxylation of benzene over TS-1/H₂O₂ system indicated that while in the presence of a cosolvent (like acetone, acetonitrile, or methanol) a long induction period was observed; in solvent-free conditions the induction period was almost absent. The effect of various reaction parameters such as benzene/H₂O₂ molar ratio, mode of addition of H₂O₂, reaction temperature, catalyst concentration, stirring speed, and the dilution level on benzene conversion and phenol selectivity was also studied. Around 85 ± 5 mol% H₂O₂ utilization for oxygenated benzene products (reaction time = 2 h) as well as phenol selectivity could be obtained under the present solvent-free, triphase conditions (at 333 K and benzene to H₂O₂ mole ratio = 1-3). The corresponding values under similar reaction conditions, except the presence of organic solvent (reaction time = 8 h), in conventionally used biphase system were 17 ± 3 and 85 ± 5, respectively. Further, the present solvent-free method offers distinct advantages of easier workup and being environmentally safer, because of the absence of organic solvents. Competitive adsorption experiments suggest that under triphase condition benzene competes favorably with water for diffusion inside the TS-1 channels. However, under biphase conditions benzene faces strong competition with organic solvents for diffusion inside relatively hydrophobic TS-1 channels resulting in low conversion.

Full text of DOI:10.1006/jcat.1998.2131

JOURNAL OF CATALYSIS 178, 101–107 (1998)
ARTICLE NO. CA982131
Triphase Catalysis over Titanium–Silicate Molecular Sieves
under Solvent-free Conditions
I. Direct Hydroxylation of Benzene
1
Asim Bhaumik, Priyabrata Mukherjee, and Rajiv Kumar
Catalysis Division, National Chemical Laboratory, Pune-411 008, India
Received October 23, 1997; revised April 28, 1998; accepted April 30, 1998
phase reaction) the observed reaction rates are very slow,
Under solvent-free triphase conditions (solid catalyst TS-1, or- owing to the low concentration of at least one of the reac-
ganic substrate and aqueous H2O2, S-L-L) a significant enhance- tants in each phase. To overcome this problem cosolvents
ment in the reaction rate (ca by 15–25 times) during the oxidation
of benzene, compared to that obtained under conventional biphase
system using an organic solvent (solid TS-1 catalyst in the presence
of a cosolvent to homogenize the organic and aqueous layers, S-L),
was observed. Time-dependent studies in the hydroxylation of ben-
procedure. Phase-transfer catalysts(2–4), used to overcome
zene over TS-1/H2O2 system indicated that while in the presence of
a cosolvent (like acetone, acetonitrile, or methanol) a long induction
period was observed; in solvent-free conditions the induction period
was almost absent. The effect of various reaction parameters such
are generally used to bring about a homogeneous state and
thereby completely eliminate the phase boundaries. Al-
though, the use of cosolvents often finds its utility, product
separation suffers from the drawback of complex workup
this problem, also suffer from the drawback that the cata-
lyst used promotes the formation of emulsions, giving rise
to severe problems in recycling and workup. Tetraalkylam-
as benzene/H2O2 molar ratio, mode of addition of H2O2, reaction monium cation supported resins (5, 6) and aluminas (7)
temperature, catalyst concentration, stirring speed, and the dilution have also been reported to be efficient triphase catalysts
level on benzene conversion and phenol selectivity was also stud- (solid catalyst resides at the interface of two liquid layers)
ied. Around 85 � 5 mol% H2O2 utilization for oxygenated benzene
in certain organic transformations.
During the last decade crystalline, microporous titanium
silicate, TS-1 (8, 9), has been extensively used in oxidation
of various organic substrates in the presence of dilute H2O2
using cosolvents (10–13). However, the use of environmen-
tally detrimental organic solvent creates problems related
to product separation and solvent recycle, which are energy
intensive steps. Hence, it is desirable to develop suitable
products (reaction time = 2 h) as well as phenol selectivity could
be obtained under the present solvent-free, triphase conditions
(
at 333 K and benzene to H2O2 mole ratio = 1–3). The corresponding
values under similar reaction conditions, except the presence of or-
ganic solvent (reaction time = 8 h), in conventionally used biphase
system were 17 � 3 and 85 � 5, respectively. Further, the present
solvent-free method offers distinct advantages of easier workup
and being environmentally safer, because of the absence of organic
solvents. Competitive adsorption experiments suggest that under methodology, where the oxidation reactions catalyzed by
triphase condition benzene competes favorably with water for dif- the TS-1/H2O2 system could be carried out without using
fusion inside the TS-1 channels. However, under biphase conditions organic solvents so that the whole process becomes envi-
benzene faces strong competition with organic solvents for diffusion
inside relatively hydrophobic TS-1 channels resulting in low con-
ronmentally favorable, even if comparable conversion and
selectivity, with respect to conventional processes, are ob-
tained.
Recently, we have reported that TS-1 can efficiently be
version.
�c 1998 Academic Press
used under triphase conditions in the oxidation of water-
immiscible organic substrates, such as toluene, anisole,
m-cresol, benzylalcohol, and cyclohexanol (14). Quite in-
terestingly, under triphase conditions where no organic
solvent was used (solid TS-1 + water immiscible organic
substrate + water + H2O2), a significant enhancement in ac-
tivity (4–6 times) and para-selectivity was achieved (14).
Although, phenol hydroxylation catalyzed by TS-1/H2O2
system has been extensively studied (8–12), very little is
INTRODUCTION
One of the major problems encountered in catalytic, liq-
uid phase organic transformations is the interaction be-
tween mutually immiscible reagent(s) and substrate (1).
Under heterogeneous conditions (organic phase–aqueous
1
Corresponding author. E-mail: rajiv@ems.ncl.res.in.
101
0021-9517/98 $25.00
Copyright �c 1998 by Academic Press
All rights of reproduction in any form reserved.

1
02
BHAUMIK, MUKHERJEE, AND KUMAR
known about benzene hydroxylation (13). Here, we report, matography (HP 5880, using Flame Ionization Detector).
for the first time, detailed studies using TS-1 as a very ef- The H2O2 selectivity is defined as the moles of H2O2 con-
ficient catalyst for the oxidation of benzene under solvent- sumed in the formation of oxygenated benzene products
free, triphase conditions (solid catalyst, organic, and aque- and the phenol selectivity is defined as the of moles of phe-
ous phases) with significant enhancement in the activity nol obtained among the total moles of oxygenated benzene
over conventional biphase system.
products formed in the reaction.
For competitive adsorption experiments water was used
for the dispersion of the catalyst under triphase condition,
whereas acetonotrile was used instead of water in the case
of biphase system maintaining the same conditions as em-
ployed duringtypicalreaction, except the presence ofH2O2;
EXPERIMENTAL
The catalyst TS-1wassynthesized employingour recently
developed concept of promoter induced synthesis of zeo-
lite materials (15). For synthesizing TS-1 (9, 15), first 42 g
of tetraethyl orthosilicate, TEOS (Aldrich), were added to
1.2 g of tetrapropylammonium hydroxide solution, TPA-
OH (ca 20 wt% aq., Aldrich) under vigorous stirring for
h. Then 2.24 g of tetra-n-butyl titanate, TNBT (Aldrich),
in 9 g isopropyl alcohol, IPA, were added very slowly into
the clear silicate solution followed by the addition of a so-
lution of 1.54 g H3PO4 (85% s.d. fine chem, India) in 9 g
of water and the stirring was continued for another 1 h.
The clear solution, thus obtained, was autoclaved at 433 K
for 6 h under agitation (60–62 rpm). The molar composi-
tion of the initial mixture was: 1.0 TEOS : 0.033 TNBT : 0.4
TPA-OH : 0.75 IPA : 0.067 H3PO4 : 25.0 H2O. After crystal-
lization, the solid was separated, washed thoroughly with
deionized water, dried and calcined at 773 K for 16 h in
an air flow. The calcined sample was treated with dilute
H2SO4 solution (3wt% aq.) at room temperature for 2h and
then centrifuged, washed thoroughly, and dried. This TS-1
sample was characterized through well-known techniques
such as XRD (Rigaku, D MAX III VC), SEM (JEOL, JSM
7
.5 g calcined TS-1 sample was taken in a 500-ml round
bottom flask and 50-g benzene was fed into it followed by
the addition of 250-g water. The mixture was then stirred
with the help of a magnetic stirrer at 333 K for 2 h. The
TS-1/water/benzene sample (sample B) was then filtered,
dried in air at room temperature, and finally stored in a
dessicator. The same sequence was followed during the
competitive adsorption experiment under biphase system
except that acetonitrile was used instead of water (TS-1/
acetonotrile/benzene sample (sample C). The adsorbed TS-
8
1
1
samples along with pure TS-1 (sample A) were charac-
terized for the adsorbed materials by IR spectroscopy us-
ing KBr pellet technique. For desorption experiments, 5 g
sample B or C was loaded in a tubular downflow glass reac-
tor and then flushed with dry nitrogen at 523 K for 1 h. The
desorbed material was cooled (in a liquid nitrogen trap),
collected, and analyzed by gas chromatography using ther-
mal conductivity (for sample B) or flame ionization (for
sample C) detectors.
5
2
200), FTIR (Nicolet 60 SXB spectrometer), UV-Vis (UV-
101 PC), and adsorption to ascertain that the catalyst is
RESULTS
free from impurities.
The XRD pattern of the fully crystalline TS-1 sample
The liquid phase catalytic reactions were conducted in a was characteristic of MFI topology. The Si/Ti molar ratio in
glass batch reactor. In a typical reaction, 0.1 mol of the sub- the solid was 32 as determined by chemical analysis (XRF).
strate, and x mmol of H2O2 (35 wt% aq., x = 30, 50, 67, and Figure 1 depicts the scanning electron micrograph of the
1
00) were reacted over the catalyst in the presence of water TS-1 sample synthesized using promoter H3PO4, clearly
(
under triphase) or acetonitrile as cosolvent (in biphase) showing that small crystallites (100–200 nm) with very nar-
under vigorous stirring (ca 800 rpm). For comparison, two row particle size distribution are obtained. Figure 2 depicts
sets of experiments, one in biphase system in the presence the UV-Vis spectrum of the TS-1 sample exhibiting a sharp
of acetonitrile (commonly used solvent in TS-1/H2O2 cata- absorption at ca 215 nm. The absence of any peak in the
lysis) and the other in solvent-free triphase system, were range of 260–360 nm clearly indicates the absence of extra
carried out under identical conditions. The substrate to wa- frame work TiO2 species (9–13).
ter or substrate to cosolvent ratio (wt/wt) was kept constant
Figure 3 depicts the effect of stirring speed on the con-
to maintain the same dilution level of the reaction mixture version and phenolselectivity. These experimentswere con-
in both biphase and triphase systems. The additional wa- ducted in a stainless steel Parr autoclave (Parr, USA). Un-
ter, apart from coming through aqueous H2O2, was used der triphase the conversion and phenol selectivity increase
for dispersion of the catalyst in this heterogeneous system with the stirring speed upto ca 400 rpm before leveling off.
and to maintain the dilution level. Generally, H2O2 (35% ) The effect of stirring was rather pronounced in triphase
was slowly injected drop-wise using a syringe pump (Sega, system. By contrast, in biphase conditions no significant
USA) during 30 min, unless stated otherwise. The prod- change in the conversion was observed under mild (200–
ucts were collected at various intervals of reaction times 300 rpm) and vigorous (400–800 rpm) stirring. For fur-
and were analyzed by high resolution capillary gas chro- ther experiments simple glass batch reactor with magnetic

TRIPHASE CATALYSIS OVER TITANIUM–SILICATE
103
FIG. 1. Scanning electron micrograph of TS-1 sample.
stirring (ca 800 rpm) was used as the results obtained using
Parr autoclave and the glass batch reactor, under otherwise
same condition, were comparable.
FIG. 3. Effect ofstirring(rpm) on the conversion ofbenzene, H2O2 se-
In Table 1, the effect of the substrate to H2O2 molar ra- lectivity, and phenol selectivity (mol% ) under triphase. Curves A, benzene
tio in the hydroxylation of benzene over TS-1/H2O2 sys- conversion; B, H2O2 selectivity, and C, phenol selectivity. Benzene = 0.1
mole, H2O2 = 0.033 mole, catalyst = 1.17 g, and H2O = 39 g (including that
tem under tri- and bi-phase conditions is reported. With
increasing benzene/H2O2 molar ratio, both the H2O2 selec-
tivity (utilization towards the formation of phenol, and its
further oxidation products like p-benzoquinone, catechol,
from 35 wt% H2O2). Reaction was carried out in a stainless steel Parr
(
USA) autoclave. Temperature = 333 K.
and hydroquinone) and phenol selectivity among prod- (T.O.F.) is about 15–25 times (Table 1). The benzene conver-
ucts were also increased, as expected. However, the ben- sion, H2O2 selectivity, rate of reaction (TOF), and phenol
zene conversion and H2O2 selectivity towards oxygenated selectivity increase by increasing benzene to H2O2 molar
benzenes are considerably higher (4–6 times) in triphase ratio. A small amount of secondary oxidation products (10–
than could be achieved in the presence of organic cosol- 20% ) like p-benzoquinone, catechol, and hydroquinone
vent (biphase). The corresponding increase in reaction rate formed due to further hydroxylation of primarily formed
phenol, were also obtained. In all other experiments, men-
tioned below, only triphase conditions were used.
In Fig. 4, the benzene conversion, H2O2 selectivity, and
phenol selectivity are plotted as a function of reaction time
under triphase conditions. In the presence of cosolvent the
reaction was very slow and only ca 20% H2O2 selectivity
could be obtained after 8 h (not plotted in Fig. 4). How-
ever, under the presently used triphase conditions, the re-
action was very fast from the beginning, reaching maximum
conversion level (ca 85% H2O2 selectivity) in 2 h. The se-
lectivity for phenol also increases (curve B), at the expense
of secondary products (curve C: p-benzoquinone, catechol,
and hydroquinone), with time (Fig. 4). Since in this case
the total amount of H2O2 was added in one lot at the be-
ginning of the reaction, it is quite likely that secondary ox-
idation products are also formed at an increased rate in
the beginning. However, as the reaction proceeds, with the
progressive utilization of oxidant, the relative formation of
secondary products also decreases with time.
In Table 2 the effect of temperature on the conversion
FIG. 2. UV-Vis spectrum of calcined TS-1 sample.
and product selectivity in the hydroxylation of benzene

1
04
BHAUMIK, MUKHERJEE, AND KUMAR
TABLE 1
Effect of Various Benzene to H2O2 Molar Ratios on the Oxidation of Benzene over TS-1 under Triphasea
Conv. (% )
Product selectivity, % e
d
Benz/H2O2
molar ratio
H2O2
sel. (% )
TOF
Phaseb
Exp.
Theo.
c
� 1
(h )
PH
PBQ
CA
HQ
1
1
2
3
.0
.5
.0
.0
Tri
Bi
74.4
11.6
100
100
84.8
13.7
72.5
2.9
85.6
82.0
9.3
10.4
2.0
4.6
3.1
3.0
Tri
Bi
52.3
9.8
66.7
66.7
86.2
16.6
49.1
2.4
89.5
86.9
7.5
9.0
1.2
2.5
1.8
1.6
Tri
Bi
41.1
8.7
50.0
50.0
87.2
19.2
37.2
2.1
90.4
90.1
6.1
6.0
1.4
2.3
2.1
1.6
Tri
Bi
28.8
6.4
33.3
33.3
90.5
20.6
25.5
1.5
95.0
92.8
2.9
4.9
2.8
1.4
1.3
0.9
a
Catalyst TS-1 (15 wt% with respect to substrate, H2O2 was slowly fed into the reaction mixture for a period of 30 min. Temperature 333 K,
benzene : H2O = 1 : 5 (wt/wt). Reaction time = 2 h (triphase) and 8 h (biphase).
b
In biphase acetonitrile was used as solvent, benzene : solvent = 1 : 5 (wt/wt). Almost similar results were obtained when acetone or methanol was
used as solvent in place of acetonitrile.
c
Calculated considering the utilization of one mole H2O2 in the formation of phenol + two moles for the formation of catechol, hydroquinone and
p-benzoquinone, and mole% H2O2 selectivity is based on the mole of H2O2 used.
d
Turn over frequency = moles of H2O2 converted for producing phenol + secondary products (catechol, hydroquinone, and p-benzoquinone)
per mole of Ti per h.
e
PH = phenol, PBQ = p-benzoquinone, CA = catechol, and HQ = hydroquinone.
under solvent-free triphase is shown. Upon increasing the 333 K. Hence, all other experiments were carried out at
temperature from 323 to 353 K, the rate of reaction (T.O.F.) 333 K.
increases along with the formation of secondary prod-
Table 3 illustrates the effect of continuous addition of
ucts, as expected. However, the optimum reaction tem- H2O2 on the conversion and selectivity in the oxidation of
perature, under the present conditions, was found to be benzene under triphase conditions. High H2O2 utilization
for the oxygenated benzene products could be achieved
from the beginning if the oxidant is added slowly. Table 3
also shows the amount of H2O2 added after different inter-
vals of time. Here also, phenol selectivity increases slightly
with time at the initial stages of the reaction at the cost of
secondary oxidation products.
In Table 4, the effect of catalyst concentration on benzene
hydroxylation is reported. Increasing the catalyst concen-
tration (with respect to benzene) resulted in an increase in
H2O2 utilization initially before leveling off at ca 85 � 5% .
Based on these results, ca 12.5–15 wt% catalyst with respect
TABLE 2
Effect of Reaction Temperature on the Activity and Selectivity
in the Oxidation of Benzene under Triphase
Product selectivity (% )
Temp. Benzene
H2O2
Time TOF
(h)
a
b
� 1
(h )
(
K) conv. (% ) sel. (% )
PH
PBQ CA HQ
3
3
3
3
23
33
43
53
67.8
74.4
78.6
72.3
70.5
84.8
85.7
88.2
3.0
2.0
1.5
40.2 96.0
72.5 85.6
90.7 86.5
4.0
9.3
8.7
—
2.0
2.0
3.0
—
3.1
2.8
4.4
FIG. 4. Conversion and phenol selectivity as a function of reaction
time in benzene hydroxylation catalyzed by TS-1/H2O2 system under
triphase conditions. Reaction conditions: Temperature 333 K, catalyst
TS-1 (15 wt% with respect to benzene), H2O : benzene = 5 : 1 (wt/wt),
1.0 150.8 78.4
14.2
a
Benzene : H2O2 molar ratio = 1 : 1, other conditions are same as that
benzene : H2O2 = 2 : 1 (mole/mole); A, benzene conversion; B, H2O2 ef- given in Table 1.
b
ficiency; and C, phenol selectivity among oxidized products.
Time at which maximum benzene conversion was achieved.

TRIPHASE CATALYSIS OVER TITANIUM–SILICATE
105
TABLE 3
to benzene was found to be an optimum for achieving a high
H2O2 utilization and phenol selectivity. At 20 wt% catalyst
concentration not only benzene conversion decreases
but also phenol selectivity decreases drastically, due to
enhanced secondary reactions. It is quite understandable
that, due to increased utilization of H2O2 for secondary
products, the benzene conversion has decreased.
Effect of Continuous Addition of H2O2 on the Conversion and
a
Selectivity in the Oxidation of Benzene under Triphase
Reaction time (min)
H2O2 added
15
30
45
60
90
120
12.5
25.0
37.5
50.0
75.0
100.0
(
mol% benzene)
Table 5 illustrates the effect of water concentration on
Benzene conv. (% )
9.1
91.0
19.0
92.7
30.3
92.5
41.0
92.3
63.0
91.1
80.8
b
H2O2 sel. (% )
91.3 benzene conversion and product selectivities. At a water
to benzene ratio = 5 : 1, maintaining 15 wt% catalyst w.r.t.
88.6 benzene (entries 1–6), maximum benzene conversion as
Products, mol%
Phenol, PH
p-Benzoquinone, PBQ
Catechol, CA
75.0
25.0
—
79.8
18.0
1.4
84.6
12.1
1.3
87.4
9.0
1.8
91.6
5.1
1.3
7.4
2.0
3.0
well as optimum H2O2 selectivity were observed. Although
increasing this ratio beyond 5 with increased dilution re-
sulted in higher phenol selectivity, the benzene conversion
Hydroquinone, HQ
—
0.8
2.0
2.8
2.0
a
Reaction conditions: Temperature 333 K, catalyst TS-1 (15 wt% with and H2O2 selectivity decrease considerably. However, de-
respect to benzene), H2O : benzene = 5 : 1 (mass/mass), benzene : H2O2 = crease of the H O : benzene ratio resulted in progressive
2
1
: 1 (mole/mole).
decrease of benzene conversion, H2O2 and phenol selectiv-
ities, indicating that at lower dilution secondary reactions
i.e., oxidation of phenol to catechol, hydroquinone, and
b
Conversion of H2O2 with respect to the moles of H2O2 added at that
reaction time.
(
p-benzoquinone) are facilitated. Although, the substrate to
catalyst ratio was the same in these experiments, the cata-
lyst concentration with respect to total reaction volume was
also changed considerably. Hence, in the next set of exper-
iments, the effect of water to benzene weight ratio was also
studied, keeping the catalyst concentration in the total reac-
tion volume constant (at different benzene/catalyst ratios).
These results, also given in Table 5 (entries 5, 7–9), suggest
PH PBQ CA HQ that optimum catalyst concentration with respect to both
the substrate and the total reaction mixture is needed for
TABLE 4
Effect of Catalyst Concentration on Benzene Conversion
and Product Selectivities
Products, mol%
Catalyst
conc., wt%
Conv. H2O2 sel. TOF Time
a
b
� 1
(h )
(% )
mole%
(h)
5
.0
26.8
37.3
38.1
41.1
33.6
58.6
84.7
87.6
87.2
90.3
37.6
36.2
36.0
37.2
38.6
4.0
3.0
2.5
2.0
1.5
90.8
9.2
—
—
achieving high phenol yields (i.e., high H2O2 utilization and
phenol selectivity).
1
1
1
2
0.0
2.5
5.0
0.0
86.4 11.1 1.1 1.4
85.0 12.7 1.0 1.3
90.4
6.1 1.4 2.1
65.7 19.4 6.0 8.9
DISCUSSION
a
Catalyst TS-1 in wt% with respect to the substrate.
Benzene : H2O2 molar ratio = 2 : 1 (maximum theoretical benzene
b
The present results obtained in the hydroxylation of ben-
zene showed that the reaction is very fast under triphase
conversion = 50% ); other conditions are same as Table 1.
TABLE 5
Effect of Water Concentration on Conversion and Phenol Selectivity over TS-1/H2O2 : Temp. 333 K,
H2O2 was Added in One Lot; Reaction Time = 2 h
Catalyst wt% , w.r.t.
Product selectivity, %
H2O : Benz.
(wt/wt)
Conv.a
(% )
H2O2 sel.
(% )
T.O.F.
�
1
Entry
Benz.
Total
(h
)
PH
PBQ
CA
HQ
1
2
3
4
5
6
7
8
9
1.0
2.0
3.0
4.0
5.0
7.5
1.0
3.0
5.0
15
15
15
15
15
15
5.2
9.8
15
6.0
4.3
3.3
2.8
2.4
1.7
2.4
2.4
2.4
14.3
30.1
33.4
36.2
41.1
34.6
13.0
40.0
41.0
39.9
81.3
85.7
88.3
87.2
75.9
14.5
86.1
87.8
17.1
34.7
36.6
37.8
37.2
32.4
17.8
36.7
37.5
60.3
64.9
71.7
79.8
90.4
90.3
88.9
92.9
89.9
24.0
20.1
18.0
14.8
6.1
6.1
6.7
4.6
7.2
6.3
6.0
4.1
2.0
1.4
1.4
1.8
1.4
1.1
9.4
9.0
6.2
3.0
2.1
2.2
2.6
2.1
1.8
a
Benzene : H2O2 = 2 : 1 (mole/mole), the maximum theoretical conversion being 50% .

1
06
BHAUMIK, MUKHERJEE, AND KUMAR
condition compared to that in conventional biphase condi-
In relatively hydrophobic high silica titanium silicate
tions using an organic co-solvent. Significantly higher activ- molecular sieves, organicsubstratesare expected to be com-
ity (as well as high para selectivity in the case of substituted peting more favorably with water (vis- a` -vis with organic
benzenes like toluene and anisole (14)) obtained in the solvent in biphase) for diffusion and adsorption in triphase
triphase system indicates that the reaction may be occurring condition, which is free from organic solvent. When an or-
predominantly inside the zeolite channels. However, in the ganic solvent is present along with the reactant, the diffu-
biphase system, low conversion (as well as high ortho selec- sion and adsorption of the reactant will be hindered by the
tivity in the case of substituted benzenes like toluene and solvent. If this is true, then it may explain the higher ac-
anisole (14)) probably indicates that under biphase condi- tivity in triphase compared to that in biphase. To confirm
tions the reaction takes place considerably on the external this hypothesis some competitive adsorption experiments
surface of the catalyst as well. This is further supported by under bi- and tri-phase conditions (in the absence of H2O2)
the fact that in triphase the distribution of catechol (CA) were carried out. Figure 5 illustrates the infra red spec-
and hydroquinone (HQ) was ca 60 : 40 while the same ratio tra of various TS-1 samples used in the competitive exper-
in biphase was ca 40 : 60, indicating higher para selectivity iments. The TS-1/acetonitrile/benzene sample (sample C,
�
1
in triphase (Table 1).
curve C) shows a doublet at ca 2350–2360 cm (character-
FIG. 5. F.T.-i.r. spectra of different TS-1 samples. A, pure TS-1; B, TS-1/water/benzene; C, TS-1/acetonotrile/ benzene.

TRIPHASE CATALYSIS OVER TITANIUM–SILICATE
107
istic peaks of acetonitrile (16) in nonoverlapping range with
TS-1 vibrations), whereas the TS-1/ H2O/benzene sample
SUMMARY AND CONCLUSIONS
�
sample B, curve B) shows major peaks at 3060–3120 cm
1
In summary, it has been demonstrated that using a
triphase system (solid–liquid–liquid), in the absence of an
organic cosolvent, a considerable increase (ca 20 times) in
conversion of benzene during oxidation by the TS-1/H2O2
system can be achieved. The effect of various reaction pa-
rameters such as time, temperature, benzene to oxidant mo-
lar ratio, catalyst concentration, mode of addition of H2O2,
and dilution level on conversion and selectivity was stud-
ied. Vigorous stirring was needed for the reaction to occur
in triphase.
Since titanium silicate molecular sieves are relatively hy-
drophobic in nature, the benzene reactant competes more
favorably with water for diffusion and adsorption under
triphase conditions, resulting in higher conversion. Apart
from enhancement in activity the present triphase method
offers distinct advantages in easier product separation and
thus contributes to the development of an eco-friendly pro-
cess.
(
�
1
and 1480cm , correspondingto benzene (17). Curve C also
indicates the presence of small amount of benzene (peaks
represented by asterisk, Fig. 5). It is quite clear from these
results that benzene is the main adsorbate in sample B while
in sample C acetonitrile is mainly adsorbed. Curve A rep-
resents the IR spectrum of the parent TS-1 sample.
Since, the IR measurements were carried out to confirm
the relative presence of adsorbate in the TS-1 catalyst, des-
orption experiments were conducted for quantitative mea-
surement of the adsorbates. The desorbed material from
TS-1/acetonitrile/benzene sample (spectrum C, Fig. 5) was
mainly acetonitrile (92 wt% ), benzene being only 8 wt% .
In a similar desorption experiment of TS-1/water/benzene
sample (spectrum B, Fig. 5) it was found that ca 90 wt%
benzene was desorbed, the remaining being water. The to-
tal weight loss was ca 10 wt% on the basis of dry TS-1 in both
the cases. These results clearly demonstrate that the pres-
ence of acetonitrile provides strong resistance to benzene
for diffusion inside the TS-1 channels, while such diffusional
hindrance is not encountered by benzene in triphase con-
ditions in the absence of organic solvent, resulting in high
conversions in triphase.
ACKNOWLEDGMENTS
P. Mukherjee and A. Bhaumik thank CSIR New Delhi for providing
Junior and Senior Research Fellowship, respectively. The authors are
grateful to Dr. A. V. Ramaswamy and Dr. P. Ratnasamy for helpful dis-
cussions, Dr. Mrs. N. Jacob for TG analysis, and Dr. S. G. Hegde for FTIR
analysis.
Although the above-mentioned explanation in triphase
conditions may be accepted as far as the adsorption and
diffusion of reactants inside the TS-1 channels (intra-
particle diffusion) is concerned, the transport of reactant
REFERENCES
(
(
benzene) to the solid surface through the aqueous phase
external mass transfer) poses an additional problem. Re-
1
2
. Menger, F. M., J. Am. Chem. Soc. 92, 5965 (1970).
. Starks, C. M., J. Am. Chem. Soc. 93, 195 (1971).
3. Dockx, J., Synthesis, 441 (1973).
cently, Desikan and Doraiswamy (18) have reviewed this
diffusion-reaction problem in triphase catalysis dealing
with polymer-supported phase transfer catalysts in vari-
ous types of triphase catalytic systems, including the solid–
liquid–liquid (S-L-L) type. In any such S-L-L systems the
overall activity depends on (i) external mass transfer, (ii) in-
traparticle diffusion, and (iii) intrinsic reactivity of the cata-
lytic sites. Although the last two phenomena are unique
to TS-1 (molecular sieves) and are already discussed
above in the present case of triphase catalysis over TS-
4. Dehmlow, E. V., Angew. Chem. Int. Ed. Engl. 13, 170 (1974).
5
6
7
. Regen, S. L., J. Am. Chem. Soc. 97, 5956 (1975).
. Regen, S. L., Angew. Chem. Int. Ed. Engl. 18, 421 (1979).
. Molinari, H., Montanari, F., Quici, S., and Tundo, P., J. Am. Chem.
Soc. 101, 3920 (1979).
8. Notari, B., Stud. Surf. Sci. Catal. 37, 413 (1987).
9. Thangaraj, A., Kumar, R., Mirajkar, S. P., and Ratnasamy, P., J. Catal.
30, 1 (1991).
0. Thangaraj, A., Kumar, R., and Ratnasamy, P., J. Catal. 131, 294
1991).
1. Kumar, P., Kumar, R., and Pandey, B., SYNLETT, 295 (1995).
1
1
(
1
/H2O2 system, the external mass transfer is a common and
1
most complicated problem in triphase catalysis in general 12. Kumar, R., Raj, A., Kumar, S. B., and Ratnasamy, P., Stud. Surf. Sci.
(
18).
In our present triphase conditions vigorous stirring,
Catal. A 84, 109 (1994).
3. Thangaraj, A., Kumar, R., and Ratnasamy, P., Appl. Catal. 57, L1-L3
1
1
(
1990).
4. Bhaumik, A., and Kumar, R., J. Chem. Soc. Chem. Commun., 349
1995).
needed for the reaction to occur (external mass transfer,
Fig. 3), may help the dispersion of the organic phase in the
(
aqueous one through micro-droplet formation and thereby 15. Kumar, R., Bhaumik, A., Ahedi, R. K., and Ganapathy, S., Nature 381,
facilitating the transport of the reactant to the catalyst sur-
face. Although a detailed treatment of chemical engineer-
ing and theoretical aspects of this interesting phenomenon
is obviously beyond the scope of the present communica-
298 (1996).
1
1
6. The Aldrich Library of Infrared Spectroscopy (C. J. Pouchert, Ed.),
ed. III, p. 500. Aldrich, Oklahoma City, 1981.
7. The Aldrich Library of Infrared Spectroscopy (C. J. Pouchert, Ed.),
ed. III, p. 560. Aldrich, Oklahoma City, 1981.
tion, we have tried to interpret the interesting results on the 18. Desikan, S., and Doraiswamy, L. K., Indus. Eng. Chem. Res. 34, 3524
basis of our experimental data.
(1995).