Selective synthesis of p-hydroxybenzaldehyde by liquid-phase catalytic oxidation of p-cresol

Article abstract of DOI:10.1021/op0498619

Liquid-phase oxidation of p-cresol over insoluble cobalt oxide (Co ₃O₄) catalyst under elevated pressure of air gave 95% selectivity to p-hydroxybenzaldehyde, an important flavoring intermediate. The selectivity to p-hydroxybenzaldehyde could be enhanced by manipulating the concentrations of p-cresol, sodium hydroxide, and catalyst and the partial pressure of oxygen in such a way that the byproducts normally encountered in this oxidation process were eliminated or minimized significantly.

Full text of DOI:10.1021/op0498619

Organic Process Research & Development 2004, 8, 873−878
Selective Synthesis of p-Hydroxybenzaldehyde by Liquid-Phase Catalytic
Oxidation of p-Cresol
Chandrashekhar V. Rode, Mahesh V. Sonar, Jayprakash M. Nadgeri, and Raghunath V. Chaudhari*
Homogeneous Catalysis DiVision, National Chemical Laboratory, Dr. Homi Bhabha Road, Pune - 411008, India
Abstract:
well as the product quality. The process reported by
Nishizawa et al.5 involves oxidation of p-cresol using soluble
as well as heterogeneous cobalt-based catalysts which gave
70-90% conversion of p-cresol with a selectivity to PHB
in the range of 60-70%. Thereafter, several attempts were
made to produce p-hydroxybenzaldehyde with high conver-
sion of p-cresol using cobalt chloride as the main catalyst.6-8
Recently, catalysts containing cobalt and/or some other metal
such as Cu or Mn supported on molecular sieves, carbon, or
resins have been reported for p-cresol oxidation.9-11 The
preparation of such catalysts is either a multistep procedure
or the cobalt metal leaches out under reaction conditions. It
is also important to note that in almost all reports, p-cresol
oxidation has been carried out under atmospheric conditions
with a longer reaction time (8-16 h) using large excesses
of catalyst (up to 8 mol %) and solvent methanol. The
selectivity of PHB is affected due to the formation of side
products such as p-hydroxybenzyl alcohol, p-hydroxybenzyl
methyl ether (PHBME), p-hydroxybenzoic acid and tarry
material.8 We thought that the formation of side products
could be eliminated by enhancing the rate of oxidation and
conversion of p-cresol as well as the intermediate p-
hydroxybenzaldehyde using particularly higher partial pres-
sure of oxygen. Therefore, the objectives of this work were
(i) to asses the suitability of the insoluble tricobalt tetraoxide
as a catalyst rather than soluble cobaltous chloride and (ii)
to optimize reaction conditions under elevated pressure of
oxygen. In this contribution, we demonstrate that Co3O4 is
an excellent heterogeneous catalyst for p-cresol oxidation,
which can be very easily separated and reused for subsequent
recycles. The effects of various reaction parameters such as
catalyst loading, substrate concentration, temperature, partial
pressure of oxygen, solvent and base concentrations on the
conversion of p-cresol, and selectivity of p-hydroxybenzal-
dehyde were also investigated. It was found that increase in
substrate concentration (up to 24%) caused increase in PHB
selectivity, while increase in partial pressure of oxygen led
to a decrease in PHB selectivity. The catalyst turnover
number (TON) was found to be much higher than that
reported in the literature.
Liquid-phase oxidation of p-cresol over insoluble cobalt oxide
(Co3O4) catalyst under elevated pressure of air gave 95%
selectivity to p-hydroxybenzaldehyde, an important flavoring
intermediate. The selectivity to p-hydroxybenzaldehyde could
be enhanced by manipulating the concentrations of p-cresol,
sodium hydroxide, and catalyst and the partial pressure of
oxygen in such a way that the byproducts normally encountered
in this oxidation process were eliminated or minimized signifi-
cantly.
Introduction
p-Hydroxybenzaldehyde (PHB) is an important intermedi-
ate for the manufacture of vanillin, a widely used flavoring
agent, trimethoxybenzaldehyde, various agrochemicals, and
pharmaceuticals such as semi-synthetic penicillin, amoxicil-
lin, and the antiemetic drug trimethobenzamide.1-3 p-
Hydroxybenzaldehyde is also used as an additive for metal-
plating brighteners, electroplating, and in perfumes and in
liquid crystals.3 Conventionally, p-hydroxybenzaldehyde is
synthesized by the following methods:
(i) saligenin process in which base-catalyzed reaction of
formaldehyde with phenol gives p-hydroxybenzyl alcohol
which is oxidized over platinum or palladium to give
p-hydroxybenzaldehyde;3
(ii) Reimer-Tiemann process in which phenol reacts with
chloroform and aqueous sodium hydroxide to give benzal
chlorides, which are rapidly hydrolyzed by the alkaline
medium to give salicylaldehyde as a major product and PHB
as a byproduct;
(iii) classical method of oxidation of alkyl benzenes by
using chromic acid or potassium permanganate.4
From an industrial point of view such methods have
several disadvantages such as use of relatively expensive
reagents generating large amounts of inorganic salts which
cause serious effluent problems and inefficiency of these
processes due to lower yields of the desired product PHB.
The liquid-phase autocatalytic oxidation of p-cresol has
major advantages in minimizing the effluent disposal prob-
lems and improvement of the overall process economics as
(5) Nishizawa, K.; Hamada, K.; Aratani, T. EP 0012939, 1979.
(6) Sharma, S. N.; Chandalia, S. B. J. Chem. Technol. Biotechnol. 1990, 49,
141.
(7) Lin, P.; Alper, H. J. Mol. Catal. 1992, 72, 143
(8) Marko, E.; Triendl, L. Catal. Lett. 1992, 46, 345.
(9) Peeters, M. P. J.; Busio, M.; Leijten, P. Appl. Catal., A 1994, 118, 51.
(10) Yumin, L.; Shetian, L.; Kaizheng Xingkai, Y., Yue, W. Appl. Catal., A
1998, 169, 127.
* To whom correspondence should be addressed. Fax: 91 20 25893260.
E-mail: rvc@ems.ncl.res.in.
(1) Sigeru, T.; Hideo, T.; Takasi, S.; Mitstio, A. J. Org. Chem. 1979, 44, 3305.
(2) Rao, D. V.; Stuber, F. A. Synthesis 1983, 4308.
(3) Mitchell, S. Kirk-Othmer Encyclopedia of Chemical Technology, 4th ed.;
Willy-Interscience: New York, 1998; Vol. 13, pp 1030-1042.
(4) Hudlicky, M. Oxidations in Organic Chemistry, American Chemical Society
National Meeting, Washington, DC, 1990.
(11) Wang, F.; Yang, G.; Zhang, W.; Wu, W.; Xu, J. Chem. Commun. 2003,
1172.
10.1021/op0498619 CCC: $27.50 © 2004 American Chemical Society
Published on Web 10/16/2004
Vol. 8, No. 6, 2004 / Organic Process Research & Development
•
873

Experimental Section
mixture was acidified by concentrated hydrochloric acid until
the pH of the mixture became 1-2, so that the sodium salts
of product and reactant were liberated. This aqueous reaction
mixture was extracted with ethyl acetate. The ethyl acetate
layer containing organic compounds was concentrated and
then eluted on a 60-120-mesh silica gel column to recover
the unreacted p-cresol, the product p-hydroxybenzaldehyde,
and other side products such as p-hydroxybenzyl alcohol,
p-hydroxybenzyl methyl ether, and the p-hydroxybenzoic
acid.
Materials. p-Cresol was supplied by Loba Chemie, while
sodium hydroxide was obtained from Merck. Analytical
grade methanol and cobalt oxide (Co3O4) were supplied by
M/s SD Fine chemicals, Pune (India).
Experimental Setup. The oxidation reactions were car-
ried out in a 300 cm3 capacity high-pressure Hastelloy reactor
supplied by Parr Instruments Co., U.S.A. The reactor was
connected to an air reservoir held at a pressure higher than
that of the reactor. The pressure drop in the reservoir vessel
(monitored using a transducer) was equivalent to the oxygen
consumed; hence, the reactor was cooled, inert gas (N2) was
vented out, and the reactor was again pressurized by air after
attaining the temperature. This procedure was repeated until
the reaction was over as indicated by absorption of stoichio-
metric amount of oxygen. The initial rate of reaction was
calculated in the region where change in partial pressure of
oxygen was <0.13 MPa.
A Hewlett-Packard model 1050 liquid chromatograph
equipped with an ultraviolet detector was employed for the
analysis. HPLC analysis was performed on a 25 cm RP-18
column supplied by Hewlett-Packard. The products and
reactant were detected using a UV detector at λmax ) 210
nm using 30% methanol-water as mobile phase at a column
temperature of 40 °C and flow rate of 1 mL/min. Samples
of 10 µL were injected into the column using an autosampler
HP 1100.
Procedure. In a typical oxidation experiment, 50 cm3 of
methanol, 17.777 gm of NaOH (mole ratio of p-cresol:NaOH
) 4) was heated in a flask with a reflux condenser until
most of the NaOH was dissolved; then 12 gm (2.2 kmol/
m3) of p-cresol was added. This reaction mixture was charged
to a 300 cm3 Parr autoclave provided with provisions such
as gas inlet and vent, liquid sampling, safety rupture disk,
cooling coil, stirrer, and thermowell. A weighed amount of
catalyst (Co3O4) was added, and the reaction mixture was
heated to 80-100 °C. When the desired temperature was
attained, the reactor was pressurized with air up to 600 psig,
and then the reaction was started by agitating at 900 rpm.
When the pressure dropped to an extent of 20% correspond-
ing to the O2 content in air, the reaction mixture was cooled
to 25 °C, and the rest of the gas (i.e., unabsorbed nitrogen)
was vented out. Again the reactor was heated to the reaction
temperature, and the reactor was pressurized with air to 600
psig. The same procedure was repeated until the stoichio-
metric amount of oxygen was absorbed. The progress of the
reaction was monitored by the observed pressure drop in the
reservoir vessel as a function of time. Liquid samples were
also withdrawn from time to time and analyzed by HPLC
for reactant and product concentration. After the reaction was
over, the contents were cooled to room temperature and
discharged.
Results and Discussion
Although a homogeneous CoCl2‚6H2O catalyst has been
widely used for p-cresol oxidation, we found that, after the
oxidation reaction, cobaltous chloride gets converted to a
solid-black residue which separates out from the reaction
crude and it cannot be reused for subsequent reactions.
Cobaltous chloride (CoCl2) in the presence of O2 and water
gets oxidized to give a black residue of a mixture of cobalt
oxide and hydroxide which cannot be reused again. There-
fore, we used solid tricobalt tetraoxide as a catalyst for
p-cresol oxidation. A few preliminary experiments showed
that 0.1 g of catalyst was sufficient for complete conversion
of p-cresol 1 to p-hydroxybenzaldehyde 3 (Scheme 1). In
these experiments, the oxidation reaction was monitored by
liquid-phase analysis as a function of time, and the various
products formed were the following: p-hydroxybenzyl
alcohol 2, p-hydroxybenzaldehyde 3, p-hydroxybenzyl meth-
yl ether 4, p-hydroxybenzoic acid 5, and a coupling product
of p-cresol. From the observed concentration-time profile
(see Figure 1) it was clear that, initially, p-hydroxybenzyl
alcohol is formed and, with the progress of the reaction,
p-hydroxybenzaldehyde started building up. Also, the forma-
tion of p-hydroxybenzyl methyl ether is greater when a higher
methanol concentration was used. Based on these observa-
tions and other data obtained at other reaction conditions,
the reaction pathway of p-cresol oxidation is shown in
Scheme 1. The results in the following sections are discussed
on the basis of conversion of p-cresol, selectivity, yield, and
turnover frequency which were calculated as follows.
% conversion )
initial concentration of p-cresol - final concentration of p-cresol
×
initial concentration of p-cresol
100 (1)
concentration of product formed
concentration of p-cresol consumed
% selectivity )
× 100
(2)
concentration of product formed
concentration of p-cresol charged to reaction
% yield )
× 100
(3)
concentration of p-cresol consumed
concentration of catalyst (active metal)
TON )
(4)
Workup Procedure for the Recovery of Pure Product.
A quantity of 80-100 cm3 of water was added to the reaction
crude to make the sodium salt of phenolic products com-
pletely soluble, and then the catalyst could be separated
quantitatively by filtration. Methanol was distilled off on a
rotavap. After complete removal of methanol, the reaction
Further work on the effects of various parameters were
studied over Co3O4 catalyst in the presence of sodium
hydroxide. In each experiment, a final sample was analyzed
by HPLC to calculate the conversion of p-cresol and
selectivity of p-hydroxybenzaldehyde, which agreed well
with the oxygen absorption data and recovered p-hydroxy-
874
•
Vol. 8, No. 6, 2004 / Organic Process Research & Development

Figure 1. Concentration-time profile. Reaction conditions:
temperature 373 K; mole ratio of NaOH to p-cresol 4; catalyst
1.5 kg/m3; agitation 900 rpm; total reaction volume 7.2 × 10-5
m3; pO2 0.1 MPa.
Figure 2. Effect of partial pressure of oxygen on initial rate
of p-cresol oxidation. Reaction conditions: p-cresol concentra-
tion 1.5 kmol/m3; mole ratio of NaOH to p-cresol 4; catalyst
1.5 kg/m3; temperature 373 K; total reaction volume 7.2 × 10-5
m3.
Scheme 1
Figure 3. Effect of oxygen partial pressure on conversion and
product selectivity. Reaction conditions: p-cresol concentration
1.5 kmole/m3; mole ratio of NaOH to p-cresol 4; catalyst 1.5
kg/m3; temperature 373 K; total reaction volume 7.2 × 10-5
m3; reaction time 3 h.
Table 1. Range of operating conditions
1
2
3
4
5
6
7
8
partial pressure of O2
temperature effect
substrate concentration
mole ratio of NaOH to p-cresol
catalyst loading
substrate-to-solvent ratio
total reaction volume
agitation speed
0.1-1.5 MPa
333-393 K
0.49-1.53 kmol/m3
1-5
0.38-3.03 kg/m3
0.13-0.24
partial pressure on the conversion of p-cresol and selectivity
of p-hydroxybenzaldehyde is shown in Figure 3. It was found
that the conversion of p-cresol was in the range of 90-93%
at lower oxygen partial pressure while, beyond 0.8 MPa as
the partial pressure of O2, conversion stabilized around 98%.
The selectivity to p-hydroxybenzaldehyde was about 95%
for oxygen partial pressure of 0.57-0.85 MPa, while beyond
0.85 MPa of pO2 selectivity to PHB decreased drastically
(to 70%) due to formation of side products, viz., p-
hydroxybenzoic acid and a tarry product, in higher amount.
The increase in oxygen partial pressure retards the desorption
of p-hydroxybenzaldehyde from the catalyst surface, facili-
tating further oxidation of p-hydroxybenzaldehyde to p-
hydroxybenzoic acid as well as formation of a tarry product
due to a condensation type of reaction between the PHB and
p-cresol, leading to a decrease in selectivity of p-hydroxy-
benzaldehyde.
7.2 × 10-5 m3
900 rpm
benzaldehyde. In each experiment g98% conversion was
observed. Table 1 presents the range of various process
parameters studied.
Effect of Oxygen Partial Pressure. Oxidation experi-
ments carried out by varying the air pressure allowed to study
the effect of oxygen partial pressure on the initial rate of
oxidation as well as on the selectivity of p-hydroxybenz-
aldehyde. For this purpose, the partial pressure of oxygen
was varied in the range of 0.1-1.5 MPa at 373 K. The initial
rate increased linearly with increase in partial pressure of
oxygen up to 0.8 MPa, beyond which the rate was found to
remain constant as shown in Figure 2. Effect of oxygen
Vol. 8, No. 6, 2004 / Organic Process Research & Development
•
875

Figure 4. Effect of temperature on conversion and selectivity.
Reaction conditions: p-cresol concentration 1.5 kmol/m3; mole
ratio of NaOH to p-cresol 4; catalyst 1.5 kg/m3 gm; pO2 0.83
MPa; total reaction volume 7.2 × 10-5m3; reaction time 3 h.
Figure 5. Effect of temperature on rate of oxidation of p-cresol.
Reaction conditions: p-cresol concentration 1.5 kmol/m3; mole
ratio of NaOH to p-cresol 4; catalyst 1.5 kg/m3; pO2 0.83 MPa;
total reaction volume 7.2 × 10-5 m3.
Effect of Temperature. The effect of temperature on both
conversion of p-cresol and selectivity to PHB was studied
in the temperature range of 333-393 K and results are
presented in Figure 4. It was found that the conversion of
p-cresol increased from 24% to 98% with increase in
temperature from 333 to 363 K and remained almost constant
at 98% with further increase in temperature up to 393 K.
The selectivity to PHB increased from 17 to 95% with
increase in temperature from 333 to 373 K, while it remained
constant with further increase in temperature up to 393 K.
The selectivity to p-hydroxybenzyl alcohol and that of other
byproducts was high at lower temperature and decreased
gradually with increase in temperature. The high concentra-
tion of p-hydroxybenzyl alcohol observed at lower temper-
atures clearly indicates that p-hydroxybenzyl alcohol is the
first intermediate formed in p-cresol oxidation as shown in
Scheme 1. The formation of p-hydroxybenzyl methyl ether
is also substantial (8-10%) at lower temperatures (333 and
343 K), while it decreased to a minimum value of 1.2% at
the highest temperature (393 K) studied. The considerable
amount of formation of other byproducts at lower temper-
atures might be due to the fact that NaOH is not completely
dissolved at lower temperatures; hence, complete conversion
of p-cresol to its sodium salt does not take place. This leaves
some free p-cresol available to form some other oxidation-
coupling products under the reaction conditions. The initial
rate of reaction increased with increase in temperature, and
the activation energy evaluated from the Arrhenius plot
(Figure 5) was found to be 22.6 kJ mol-1, indicating that
the reaction may be diffusion controlled under the conditions
of the present study.
Figure 6. Effect of substrate concentration on rate of oxidation
of p-cresol. Reaction conditions: mole ratio of NaOH to p-cresol
4; catalyst 1.5 kg/m3; temperature 373 K; pO2 0.83 MPa; total
reaction volume 7.2 × 10-5 m3.
was found to increase from 54 to 95% with increase in
p-cresol concentration from 0.49 to 1.53 kmol/m3. Initially,
at lower concentration of p-cresol (0.49 kmol/m3), a sub-
stantial amount of p-hydroxybenzyl methyl ether (>20%)
was observed which decreased gradually with an increase
in substrate concentration. Since, the total reaction volume
is kept constant at lower substrate concentration, solvent
(methanol) concentration was high, causing a nucleophilic
attack of CH3O- on the benzyl carbon of the intermediate
p-hydroxybenzyl alcohol, leading to the formation of p-
hydroxybenzyl methyl ether.
Effect of Sodium Hydroxide Concentration. Since the
phenolic hydroxyl group of p-cresol interferes in the oxida-
tion process, it is necessary to convert p-cresol into its sodium
phenolate salt in the presence NaOH. Hence, it was important
to study the effect of NaOH concentration in the p-cresol
oxidation process. These experiments were carried out by
varying the mole ratio of NaOH to p-cresol from 2 to 5 at a
constant p-cresol concentration of 1.5 kmol/m3, and the
results are shown in Figure 8. It was observed that the
Effect of Substrate Concentration. Typical results on
the effect of initial concentration of substrate in the range
from 0.49 to 1.53 kmol/m3 at 373 K are shown in Figure 6.
It was found that the initial rate of reaction decreased by
about 36% with 4-fold increase in p-cresol concentration up
to 1.53 kmol/m3. As can be seen from Figure 7, p-cresol
conversion was almost constant for all concentrations of
p-cresol studied in this work. However, the selectivity to PHB
876
•
Vol. 8, No. 6, 2004 / Organic Process Research & Development

Figure 7. Effect of substrate concentration on conversion and
selectivity. Reaction conditions: mole ratio of NaOH to p-cresol
4; catalyst 1.5 kg/m3; pO2 0.83 MPa; total reaction volume 7.2
× 10-5 m3; reaction time 3 h.
Figure 9. Effect of catalyst loading on rate of oxidation of
p-cresol. Reaction conditions: mole ratio of NaOH to p-cresol
4; temperature 373 K; pO2 0.83 MPa; total reaction volume 7.2
× 10-5 m3.
Scheme 2
the rate of reaction was independent of catalyst loading,
indicating that external gas-liquid mass-transfer resistance
may be significant under these experimental conditions.13
Further experiments on effect of agitation speed and quan-
titative analysis of initial rate data for ascertaining the
importance of mass-transfer resistance is in progress for
kinetic studies. A plot of selectivity of p-hydroxybenzalde-
hyde vs catalyst loading for the same range of conditions is
also shown Figure 10. At a very low catalyst loading (0.153
kg/m3), p-cresol conversion was 60% and increased with
increase in catalyst loading. The selectivity to PHB was also
low (58%) for lower catalyst loading and gradually increased
to 95% with increase in catalyst loading to 1.53 kg/m3. Major
formation of an intermediate (p-hydroxybenzyl alcohol) was
observed for lower catalyst concentrations due to suppression
of further oxidation of p-hydroxybenzyl alcohol to PHB. At
a higher catalyst loading of 3.03 kg/m3, formation of
p-hydroxybenzoic acid and other byproducts was observed
in appreciable amounts due to further oxidation of PHB.
Effect of Solvent. To asses the suitability of a solvent
for p-cresol oxidation, three common solvents were screened,
viz., water, ethanol, and methanol. As can be seen from
Figure 11, the conversion of p-cresol was marginally affected,
while selectivity pattern was dramatically altered for water
and ethanol as oxidation solvents. In case of water, PHB
formation was only 2% with major amounts of other
oxidation byproducts due to the fact that NaOH had very
high solubility in water and it was not available for protecting
Figure 8. Effect of mole ratio of NaOH to substrate on
conversion and selectivity. Reaction conditions: p-cresol con-
centration 1.5 kmol/m3; catalyst 1.5 kg/m3; temperature 373
K; pO2 0.83 MPa; total reaction volume 7.2 × 10-5 m3; reaction
time 3 h.
conversion of p-cresol increased from 69 to 98.5% with
increase in NaOH to p-cresol mole ratio from 2 to 4, and it
remained constant with further increase in mole ratio to 5.
At a lower value of NaOH to p-cresol mole ratio 2, selectivity
to PHB was minimum (11%) while that of other byproduct
was very high (>76%). Even when NaOH to p-cresol mole
ratio was 3, byproduct formation was more than 28%. Lower
selectivities of PHB under these conditions were due to the
oxidative dimerization of p-cresol which appears to give
dihydroxy diphenyl compound 612 (Scheme 2), as evidenced
by NMR analysis.12
Effect of Catalyst Loading. The effect of catalyst loading
on the initial rate of reaction was studied in the range of
0.38-3.03 kg/m3 at 373 K and 0.83 MPa of oxygen partial
pressure, and the results are presented in Figure 9. The initial
rate was found to increase by about 3 times when the catalyst
loading was increased from 0.38 to 1.5 kg/m3 beyond which
(13) Rode, C. V.; Chaudhari, R. V. Ind. Eng. Chem. Res. 1994, 33, 3, 1645.
(14) Gao, Z.; Shao, ya. J.; Wang, S.; Yang, P. Appl. Catal., A 2001, 209, 27.
(12) Moore, R. F.; Waters, W. A. J. Org. Chem. 1954, 19, 3.
Vol. 8, No. 6, 2004 / Organic Process Research & Development
•
877

Figure 12. Catalyst recycle study. Reaction conditions: p-
cresol concentration 1.5 kmol/m3; temperature 373 K; catalyst
loading 1.5 kg/m3; pO2 0.83 MPa; total reaction volume 7.2 ×
10-5 cm3; reaction time 3 h.
Figure 10. Effect of catalyst loading on conversion and
selectivity. Reaction conditions: p-cresol concentration 1.5 kmol/
m3; temperature 373 K; pO2 0.83 MPa; total reaction volume
7.2 × 10-5 m3; reaction time 3 h.
in Figure 12. The catalyst was found to retain its activity
even after the third recycle without affecting the conversion
of p-cresol; however, selectivity for PHB was marginally
decreased by 2-3%. This lowering of selectivity could be
due to the handling of losses of catalyst in the subsequent
recycle runs. The overall turnover number (TON) for the
catalyst for our process was found to be 1 × 103 which is
higher than the literature value of 3.25 × 102 for a resin-
supported cobalt oxide catalyst.14 The catalyst recycle studies
also showed the absence of any leaching of cobalt in the
solution under reaction conditions.
Conclusions
Liquid-phase oxidation of p-cresol under elevated pressure
of air was carried out using Co3O4 as a catalyst in the
presence of sodium hydroxide to give p-hydroxybenzalde-
hyde with selectivity as high as 95% and almost complete
conversion of p-cresol. As compared to air oxidation of
p-cresol under atmospheric conditions, elevated pressure of
air completely converted the intermediate p-hydroxybenzyl
alcohol to PHB. A systematic study on the effect of reaction
conditions on conversion and selectivity to p-hydroxyben-
zaldehyde was carried out. The selectivity to PHB was
strongly dependent on substrate concentration, temperature,
oxygen partial pressure, mole ratio of NaOH to substrate,
and catalyst loading. It was found that the high substrate
concentration (24%) eliminated the formation of a byproduct,
p-hydroxybenzyl methyl ether, while a mole ratio of NaOH
to substrate concentration below 3 gave rise to formation of
a dihydroxy diphenyl compound causing a decrease in
selectivity to PHB. The catalyst was successfully used for
three recycle experiments, giving a total TON of 1.023 ×
103, which is about 3 times higher than the literature value.
Figure 11. Effect of solvent on conversion and selectivity.
Reaction conditions: p-cresol concentration 1.5 kmol/m3; tem-
perature 373 K; catalyst loading 1.5 kg/m3 m.; pO2 0.83 MPa;
total reaction volume 7.2 × 10-5 m3; reaction time 3 h; (1)
water, (2) ethanol, (3) methanol.
the -OH group of p-cresol. For ethanol as a solvent, PHB
selectivity was somewhat higher than that in water but still
around 38%. The ethanol used was not in completely
anhydrous form, and it may have contained water in
appreciable amount; hence, free -OH of p-cresol is again
available to form other oxidation byproducts. From this study,
it was confirmed that methanol was the only suitable solvent
that gave 95% selectivity to PHB with almost complete
conversion of p-cresol.
Catalyst Recycle. The catalyst recycle experiments were
carried out in the following way: after the first oxidation
run with the fresh catalyst, it was filtered out and dried in
an oven at 373 K for 3 h and was recharged to the reactor
for the subsequent run. This procedure was followed for three
subsequent oxidation experiments, and the results are shown
Received for review July 28, 2004.
OP0498619
878
•
Vol. 8, No. 6, 2004 / Organic Process Research & Development