Kinetics and process parameter studies in highly selective air oxidation of side-chain alkyl groups in picolines, 2-methylnaphthalene, and pseudocumene

Article abstract of DOI:10.1021/op980047t

Picolines, 2-methylnaphthalene, and pseudocumene were oxidized by air in acetic acid medium. Process parameters and kinetics of the reaction were studied from the viewpoint of proces research and development. Use of lithium chloride as the promoter was worth considering for this oxidation to obtain a higher rate of reaction. At 170 °C and at a reactant concentration of 15% w/v, 52% conversion of β-picoline with a selectivity of 97% was achieved in 8 h.

Full text of DOI:10.1021/op980047t

Organic Process Research & Development 1999, 3, 227−231
Kinetics and Process Parameter Studies in Highly Selective Air Oxidation of
Side-Chain Alkyl Groups in Picolines, 2-Methylnaphthalene, and Pseudocumene
Sudip Mukhopadhyay and Sampatraj B. Chandalia*
UniVersity Department of Chemical Technology, UniVersity of Mumbai, Matunga, Mumbai, India
Abstract:
Scheme 1. Air oxidation of picolines, 2-methylnaphthalene,
and pseudocumene
Picolines, 2-methylnaphthalene, and pseudocumene were oxi-
dized by air in acetic acid medium. Process parameters and
kinetics of the reaction were studied from the viewpoint of
proces research and development. Use of lithium chloride as
the promoter was worth considering for this oxidation to obtain
a higher rate of reaction. At 170 °C and at a reactant
concentration of 15% w/v, 52% conversion of â-picoline with
a selectivity of 97% was achieved in 8 h.
Introduction
Pyridine carboxylic acids, naphthalene carboxylic acids,
and trimellitic acid have great relevance in organic process
industries as intermediates for pharmaceuticals and fine
chemicals. In general, these are synthesized by air oxidation
1-3
4-6
7,8
of picolines, methylnaphthalenes, and pseudocumenes
using different transition metal catalysts and bromide salts
as promoters.
There is little information regarding these air oxidations,
and what is available is mostly confined to the patent
literature. Thus, this work is an attempt to study the process
parameters to establish the most suitable process conditions
and a more realastic kinetic interpretation in the air oxidation
of picolines, methylnaphthalenes, and pseudocumenes using
cobalt and manganese acetates as the catalyst and LiCl as a
promoter to synthesize the respective carboxylic acids
Experimental Procedure. Predetermined amounts of the
catalyst, promoter, and reactant were mixed with solvent,
and the solution was shaken thoroughly to make it homo-
geneous. The reactor was pressurized with air to the desired
pressure. The reactor was then heated to the desired
temperature, and flow of air was started. After the reaction
was allowed to proceed for the predetermined period, the
reactor was allowed to cool to the room temperature, and
the pressure was released. A complete diagram of the bubble
column reactor is shown in Figure 1.
Analytical Details. The reaction mixture was filtered and
distilled under vacuum to remove acetic acid and then diluted
with water, and organic compounds were extracted in toluene
or benzene. The organic layer, after thorough washing with
water, was taken for analysis.
(Scheme 1).
Experimental Section
Materials. The materials used and their technical grade
are tabulated below:
technical
grade
technical
grade
material used
material used
diglyme
1,4-dioxane
cobalt(II) acetate
R-picoline
L.R.
L.R.
L.R.
L.R.
Fluka
A.R.
technical grade
L.R.
L.R.
â-picoline
γ-picoline
pseudocumene
manganese(II) acetate L.R.
2
-methylnaphthalene
acetic acid
sodium bromide
lithium chloride
L.R.
L.R.
(
(
(
1) Bhattacharya, G. R. Indian Chem. Eng. 1982, 24 (1), 46.
2) Inoue, H. T. Jpn. Kokai 76, 29, 483, 1976; Chem. Abstr. 1976, 85, 94233.
3) Ahmed, M.; Chaudhary, T. A.; Azam, M. P. J. Sci. Ind. Res. 1987, 30 (3),
(6) Sato, T.; Ito, I.; Takeda, K. (Sumikin Coke Co. Ltd.). Jpn. Kokai Tokkyo
1
82-84.
Koho JP 90 240 047, 1990.
(
(
4) Sato, T.; Ito, I.; Takeda, K. Jpn. Kokai Tokkyo Koho JP 90 160 745, 1990.
5) Tachibana, Y.; Tate, K.; Ono, M.; Takei, N.; Miki, A.; Taniguchi, H.;
Shirato, Y.; Shimura, M.; Fukui, Y. (Nippon Kokan K. K.). Jpn. Kokai
Tokkyo Koho JP 90 164 844, 1990.
(7) Schammel, W. P.; Green, M. R. (Amoco Corp.). U.S. Patent US 4, 845,
275, 1989.
(8) Darin, J. K.; Bemis, A. G. (Amoco Corp.). U.S. Patent US 4, 895, 978,
1990.
1
0.1021/op980047t CCC: $18.00 © 1999 American Chemical Society and Royal Society of Chemistry
Vol. 3, No. 3, 1999 / Organic Process Research & Development
•
227
Published on Web 05/06/1999

Figure 2. Effect of air flow rate on overall conversion. Reaction
conditions: reactant concentration, 15% w/v; air pressure, 16
atm; solvent, acetic acid; reaction volume, 300 mL. Key to
symbols (L/min): [, 0.6; 9, 1.2; 4, 1.8; ×, 2.4; +, 3.0; and O,
3.36.
The carboxylic acid present in the reaction mixture was
analyzed by taking a measured volume of the extracted
organic layer. The extract was titrated with standard metha-
nolic sodium hydroxide to estimate the amount of carboxylic
acid.
Figure 1. Experimental setup for liquid phase oxidation by
air: 1, compressor; 2, needle valves; 3, reactor; 4, heating
element; 5, thermometer pocket; 6, sparger; 7, condenser; 8,
pressure gauge; and 9, rotameter.
Estimation of Unreacted Substrate. The unreacted
substrate was analyzed by GC. The conditions are given
below:
Results and Discussions
The oxidation of the alkyl heterocyclics in acetic acid
media using cobalt acetate as catalyst and a promoter such
as sodium bromide is fairly well established and is practiced
on a large scale. Though the technology has matured, the
information is mostly confined to the patent literature. From
the limited information available, it is not easy to appraise
the relative merits of using one or the other promoter, solvent,
substrate concentration, etc. A significant part of the work
was carried out in order to study the reaction parameters,
using â-picoline as a model compound.
column used
column dimensions
stainless steel column
3.2 mm × i.d. × 2 m (10% OV-17 on
Chromosorb-W)
carrier gas
detector
nitrogen (flow rate, 30 mL/min)
FID
oven temperature
injector temperature
detector temperature
250 °C
300 °C
300 °C
Estimation of Acid. A measured volume of the organic
extract was washed with sodium hydroxide to extract the
carboxylic acid as sodium salts. The acid layer was neutral-
ized with mineral acid to precipitate the carboxylic acid. The
latter was dried and weighed. The organic extract was
directly titrated with standard sodium hydroxide to estimate
the amounts of carboxylic acids when solvents other than
acetic acid were used.
Estimation of Aldehyde. Estimation of aldehyde was
done by oximation reaction. In a typical procedure, an
accurately measured organic layer was diluted with methanol,
and the pH of the solution was adjusted to 3.0. A solution
of hydroxyl amine hydrochloride in 90% methanol was made,
and the pH was adjusted to 3.0 separately. Both of the
solutions were mixed and kept at 45 °C for 1 h.
Material Balance. In a typical study, a flow rate of 50
mL/cm was maintained. To ascertain whether mass-transfer
effects have been eliminated, air flow was increased from
.0 to 3.36 L/min; there was almost no change in the
conversion obtained. At air flow rates less than 3.0 mL/min
Figure 2), the conversion was relatively poor, presumably
because of the effects of the mass transfer. Therefore, in all
the subsequent runs, air flow rate was maintained at 3.0
L/min so that effects of mass transfer were eliminated, and
the results obtained represent the kinetics of the process
3
3
(
(
Table 1.
Effect of Air Pressure. In a liquid phase oxidation by
air, usually the reaction is carried out at a high pressure.
Such a high pressure is required in order to keep reactants
in the liquid phase and to maintain a sufficient partial
pressure of oxygen.
In this work, the pressure was varied from 6 to 18 atm
Figure 3). At a pressure of 6 atm, the rate of reaction was
The liberated HCl was titrated potentiometricaly to pH
3.0 using 0.1 N sodium hydroxide solution. The amount of
the aldehyde present in the sample was calculated as follows:
(
too slow, and only 17% conversion of â-picoline was
obtained. When pressure was increased to 16 atm, marked
improvement in the rate of reaction was noticed. However,
with a further increase in pressure to 18 atm, the conversion
remained more or less the same. Hence, a pressure of about
16 atm was suitable for the process.
VNM
grams of aldehyde )
1
000
where V is the volume of sodium hydroxide solution required
in mL), N is the normality of the sodium hydroxide solution,
and M is the molecular weight of the aldehyde.
(
228
•
Vol. 3, No. 3, 1999 / Organic Process Research & Development

Table 1. Material balancea
material
gmol
% â-picoline accounted for
â-picoline (taken)
aldehyde
acid
unreacted â-picoline
unaccounted â-picoline 0.0140
0.4838
0.0053
0.1160
0.3479
100
1.09
23.97
71.92
2.9
total
0.4833
99.89
a
Reaction conditions: substrate concentration, 15% w/v; air flow rate, 3.0
L/min; air pressure, 16 atm; cobalt acetate, 0.033 gmol/L; manganese acetate,
0
.033 gmol/L; sodium bromide, 0.05 gmol/L; reaction time, 8 h; reaction
temperature, 170 °C; solvent, acetic acid; reaction volume, 300 mL.
Figure 5. Effect of initial concentration of reactant on overall
conversion. Reaction conditions: reaction temperature, 170 °C;
air flow rate, 3.0 L/min; air pressure, 16 atm; solvent, acetic
acid; reaction volume, 300 mL. Key to symbols (% w/v): [,
15; 0, 30; and 4, 45.
Figure 3. Effect of air pressure on overall conversion. Reaction
conditions: reactant concentration, 15% w/v; air pressure, 16
atm; reaction temperature, 170 °C; solvent, acetic acid; reaction
volume, 300 mL. Key to symbols (atm): [, 8; 0, 12; 4, 16;
and ×, 18.
Figure 6. Effect of temperature on overall conversion. Reaction
conditions: reactant concentration, 15% w/v; air flow rate, 3.0
L/min; air pressure, 16 atm; solvent, acetic acid; reaction
volume, 300 mL. Key to symbols (°C): [, 140; 0, 170; 4, 190;
and ×, 210.
Subsequent reactions were carried out by taking â-picoline
as the starting material.
Effect of Initial Reactant Concentration. The concen-
tration of the â-picoline was varied from 15 to 45% w/v. It
was observed that changing the concentration had very little
influence on the rate of reaction of â-picoline (Figure 5).
All the reactions were thus carried out at a substrate
concentration of 15% w/v.
Effect of Temperature. The reaction temperature was
varied from 140 to 210 °C. The reaction temperature was
found to have a significant effect on the reaction rate (Figure
Figure 4. Effect of period of reaction on overall conversion.
Reaction conditions: reactant concentration, 15% w/v; tem-
perature, 170 °C; air pressure, 16 atm; solvent, acetic acid;
reaction volume, 300 mL. Key to symbols: [, â-picoline; 0,
γ-picoline; 4, methyl-2-naphthalene; and ×, pseudocumene.
6
). At 140 °C, the rate of oxidation of â-picoline was very
poor, and in 8 h, the conversion to acid was 10.78%. When
the temperature was increased to 170 °C, there was a marked
improvement in the reaction rate, and the conversion to acid
increased to 24%. A further increase in temperature increased
the conversion, but the fact that the selectivity was as little
as 47% may be due to decomposition of the heterocyclic
ring (Figure 7). At higher temperatures, corrosion due to
sodium bromide, the promoter, was significant. In view of
this, a reaction temperature of 170 °C appeared to be the
most suitable for this process.
Effect of Period of Reaction. The reaction was studied
for different times. It was observed that, with an increase in
the period of reaction, the conversion also increased (Figure
4). The rate of reaction for different starting material was in
the following order:
2-methylnaphthalene > γ-picoline > pseudocumene >
â-picoline
Vol. 3, No. 3, 1999 / Organic Process Research & Development
•
229

Figure 7. Effect of temperature on overall conversion and
selectivity. Reaction conditions: reactant concentration, 15%
w/v; air flow rate, 3.0 L/min; air pressure, 16 atm; time, 8 h;
solvent, acetic acid; reaction volume, 300 mL. Key to symbols:
A
Figure 8. -ln(1 - X ) vs t for different substrates. Reaction
conditions: reactant concentration, 15% w/v; air flow rate, 3.0
L/min; air pressure, 16 atm; temperature, 170 °C; solvent, acetic
acid; reaction volume, 300 mL. Key to symbols: [, â-picoline;
9, pseudocumene; 2, γ-picoline; and ×, methyl-2-naphthalene.
[, conversion; and 0, selectivity.
Table 2. Effect of solvent
%
overall
% selectivity
%
conversion
name of
solvent
conversion
with respect to
of â-picoline to aldehyde to acid acid formed
acetic acid
diglyme
27.59
14
17
1.1
1.4
1.3
24
11.62
13.77
87
83
81
1,4-dioxane
a
Reaction conditions: reaction temperature, 170 °C; air flow rate, 3.0 L/min;
air pressure, 16 atm; cobalt acetate, 0.033 gmol/L; manganese acetate, 0.033
gmol/L; sodium bromide, 0.05 gmol/L; reaction time, 8 h; reaction volume, 300
mL.
A
Figure 9. -ln(1 - X ) vs t at different substrate concentra-
Table 3. Effect of promoters
tions. Reaction conditions: air flow rate, 3.0 L/min; air
pressure, 16 atm; temperature, 170 °C; solvent, acetic acid;
reaction volume, 300 mL. Key to symbols (% w/v): [, 15; 9,
%
overall % selectivity with
name of promoter
no promoter
paraldehyde
sodium bromide
lithium chloride
sodium bromide (50% w/w) +
lithium chloride (50% w/w)
conversion
respect to acid
30; and 2, 45.
8
83
85
87
97
96
12
27.59
52
46
a
Reaction conditions: reaction temperature, 170 °C; air flow rate, 3.0 L/min;
air pressure, 16 atm; cobalt acetate, 0.033 gmol/L; manganese acetate, 0.033
gmol/L; sodium bromide, 0.05 gmol/L; reaction time, 8 h; reaction volume, 300
mL.
Effect of Solvent. With a view toward using the catalyst
system under consideration with cobalt acetate, manganese
acetate, and sodium bromide with solvents other than acetic
acid as the medium, the use of diglyme and 1,4-dioxane was
considered. In all cases, the results obtained were somewhat
satisfactory (Table 2).
Effect of Promoters. Different type of promoters, such
as sodium bromide, lithium chloride, and paraldehyde, were
used. It was observed that, under the reaction conditions,
lithium chloride was the best promoter. But lithium chloride
is costlier than sodium bromide, so a 50/50 (% w/v) mixture
of sodium bromide and lithium chloride was proved to be
the best to achieve a fast reaction rate. Sodium bromide and
lithium chloride are known to be corrosive. To avoid the
use of a titanium-lined reactor, paraldehyde was tried as the
promoter, but the conversion of â-picoline was as low as
A
Figure 10. -ln(1 - X ) vs t at different temperatures. Reaction
conditions: reactant concentration, 15% w/v; air flow rate, 3.0
L/min; air pressure, 16 atm; solvent, acetic acid; reaction
volume, 300 mL. Key to symbols (°C): [, 140; 9, 170; 4, 190;
and ×, 210.
12%, as compared to 52% (Table 3) when lithium chloride
was used.
Optimum Conditions for â-Picoline Oxidation. The
following conditions were found to be optimal: air flow rate,
3 L/min; air pressure, 16 atm; cobalt acetate, 0.033 gmol/L;
manganese acetate, 0.033 gmol/L; promoter, LiCl; temper-
ature, 170 °C; time, 8 h. These conditions resulted in 52%
conversion of â-picoline, with 97% selectivity.
230
•
Vol. 3, No. 3, 1999 / Organic Process Research & Development

10 - , and 9.6 × 10 s-1 for a particular air flow rate,
5
-5
×
air pressure, and catalyst loading.
From the Arrhenius plot, the activation energy was found
to be 15.9 kcal/mol (Figure 11).
Conclusions
â- and γ-picoline can be readily oxidized to the corre-
sponding acids, but not R-picoline, since at this temperature
R-picolinic acid is susceptible to decarboxylation.
Pseudocumene. which is difficult to oxidize, can be
oxidized to trimellitic acid with a sizable yield. Lithium
chloride seems to be the best promoter under the reaction
conditions.
One can produce 2-naphthanoic acid by this process
scheme. The rate of reaction of 2-methylnaphthalene is higher
than those of picolines and pseudocumenes under the same
reaction conditions.
Figure 11. Arrhenius plot. Reaction conditions: air flow rate,
.0 L/min; air pressure, 16 atm; solvent, acetic acid; reaction
volume, 300 mL.
3
Kinetic Interpretation. The reactions were carried out
at a high air flow rate and air pressure to ensure that the
mass transfer has been eliminated completely.
The reaction follows a first-order kinetics.
A plot of -ln(1 - X
) shows that these oxidation reactions are first order with
respect to the substrate. A plot of -ln(1 - X ) vs t) for
A
) vs t for different substrates (Figure
8
Acknowledgment
A
S.M. is grateful to the University Grants Commissions,
New Delhi, for the award of a senior research fellowship.
different concentrations of â-picoline (Figure 9 further
supports the fact that it is a first-order kinetics.
At different temperatures, -ln(1 - X
A
) vs t is plotted
Received for review June 1, 1998.
OP980047T
(
Figure 10), and from the slopes, k values at different
-
6
-5
temperatures are found to be 6.0 × 10 , 1.1 × 10 , 3.39
Vol. 3, No. 3, 1999 / Organic Process Research & Development
•
231