Académique Documents
Professionnel Documents
Culture Documents
A R T I C L E I N F O
A B S T R A C T
Article history:
Received 20 December 2008
Received in revised form 29 January 2009
Accepted 29 January 2009
Available online 6 February 2009
Co-oxidation of p-xylene (PX) and p-toluic acid (p-TA) to terephthalic acid (TA) in water with Mn acetate
catalyst has been studied. Mn acetate showed better catalytic activity than Co acetate and no synergism
was observed between two catalysts. During the reaction, PX was transformed to p-TA while p-TA was
converted to TA by a radical reaction. In the co-oxidation, radicals generated from facile PX-to-p-TA
conversion would assist more difcult p-TA to TA conversion. Addition of CO2 in the oxygen promoted
the TA formation and suppressed the CO2 formation from the combustion of reactants. Additional
transition metals such as Ni and Ti also promoted the catalytic activity of Mn catalyst. Furthermore, a
synergism was observed between added CO2 and the added transition metal. Finally, the yield of
terephthalic acid reached about 95% with Mn + (Ti, Ni or Cr) catalyst system and CO2 addition.
2009 Elsevier B.V. All rights reserved.
Keywords:
Terephthalic acid
Co-oxidation
Water solvent
Manganese acetate
CO2 effect
1. Introduction
Puried terephthalic acid (TA) is an important petrochemical
product and the main source of polyethylene terephthalate, a
major material for textiles and bottles. About 42 million tons of TA
was produced in the world [1] and 5.6 million tons in Korea in 2006
[2]. Today most of TA in the world is produced by the Amoco-MC
process. In this process, p-xylene (PX) is oxidized by air at about
200 8C, 20 atm with soluble Co/Mn catalysts, acetic acid solvent
and a Br promoter. After oxidation, the nal cake called crude
terephthalic acid contains about 3000 ppm of 4-formylbenzoaldehyde (4-CBA) and 500 ppm of p-toluic acid (p-TA) as impurities.
Since the content of impurities is too high for polymerization, the
cake is hydrogenated to convert 4-CBA to p-TA in a subsequent
purication process [3]. Upon hydrogenation, p-TA is separated
from TA by crystallization. The nal puried TA contains less than
25 ppm of 4-CBA and 150 ppm of p-TA.
The Amoco-MC process is highly efcient and optimized with
extended experience of commercial operation. But it causes serious
environment, economy and safety problems by using the bromine
promoter and the acetic acid solvent. For example, anti-corrosive
materials like Ti are needed for the equipment because of the highly
corrosive bromine and acetic acid, which raises the initial investment
and maintenance cost. Further, this process needs various acetic acid
* Corresponding author. Tel.: +82 54 279 2266; fax: +82 54 279 5528.
** Corresponding author.
E-mail address: jlee@postech.ac.kr (J.S. Lee).
0926-860X/$ see front matter 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.apcata.2009.01.037
recovery and recycling units from its aqueous solution, which add
investment and operating cost further. In addition, a large amount of
acetic acid is lost in real plants by combustion and handling and fresh
acetic acid is replenished to compensate for the loss. Moreover, in the
overall TA manufacturing process, PX oxidation is carried out in
acetic acid solvent but TA purication in water. Thus PX oxidation
and TA purication units should be strictly separated. It makes the
whole process more complicated and increases the operation cost
and risk. In particular, liquid bromine compounds cause serious
equipment corrosion and gaseous compounds bring environmental
problems in vent streams [4]. So installation of bromine scrubber is
essential to mitigate these problems in the downstream, which
pushes up the production cost further.
Many attempts have been made to remove the bromine and
acetic acid from the TA production process since 1970s [5,6].
Hronec and Ilavsky [7] proposed the co-oxidation of PX and p-TA in
water to produce TA at a high yield (more than 90%). Ishii and
Sakaguchi [8] carried out PX oxidation with N-hydroxyphthalimide (NHPI) as a promoter instead of Br, and the yield of TA was
near 95%. Dunn et al. [9] studied PX oxidation in a supercritical
water solution, which produced TA with about 80% yield. There
have also been a few reports on the improvement of Amoco-MC
type oxidation of PX, i.e. Co/Mn/Br catalysts in acetic acid. Thus,
Yoo et al. [10] found a promotional effect of carbon dioxide added
to the reaction mixture of the Amoco-MC oxidation.
Among the proposed methods of PX oxidation without Br
promoter and/or acetic acid solvent, we considered that cooxidation of PX and p-TA in water was a most promising one. It
needs neither obnoxious Br promoter nor corrosive acetic acid
55
(2)
(1)
Fig. 1. Produced TA vs. mole fraction of Co in catalyst solution: reaction conditions
PX
6.4 g
(0.06 mol),
p-TA
12.24 g
(0.09 mol),
water
8 g,
Co(OAc)24H2O + Mn(OAc)24H2O = 0.36 mmol (xed), 195 8C, 22 atm (O2 8 atm,
water vapor 11 atm and PX vapor 3 atm), reaction time 3 h.
56
Table 2
Composition of deutrated compound in each intermediate.a.
Reaction time (min)
30
60c
p-TA
4-CBA
TA
10.2
9.4
8.8
5.4
0
3.0
0
2.8
a
PX 7 g (66.0 mmol), deutrated PX 1 g (9.1 mmol), p-TA 15.3 g (112.5 mmol),
water 6.5 g, Mn(OAc)24H2O 0.225 g (0.9 mmol) 195 8C, 27 atm (O2 8 atm, N2 5 atm,
water vapor 11 atm, PX vapor 3 atm), reaction time 3 h.
b
PX conversion: 42.3%.
c
PX conversion: 80.1%.
Fig. 2. The amounts of PX and TA vs. the reaction time. The amount of p-TA shows
the change from the initial amount. Inset shows the time change of oxygen
consumption: reaction conditions PX 8 g (75.5 mmol), p-TA 15.3 g (112.5 mmol),
water 6.5 g, Mn(OAc)24H2O 0.225 g (0.9 mmol), 195 8C, 27 atm (O2 8 atm, N2 5 atm,
water vapor 11 atm and PX vapor 3 atm).
the results are consistent with the initial report of Hronec and
Ilavsky [7] and only Mn(II) acetate was used as a catalyst in the
subsequent study unless stated otherwise.
The time course of the reaction is presented in Fig. 2 together with
the oxygen consumption prole (inset). The rapid conversion of PX
was observed at the beginning, but the reaction rate of PX conversion
became slower with reaction time and reached almost zero after
80 min. The amount of p-TA increased slightly initially with reaction
time and hit a broad maximum at 30 min, and then decreased
gradually. The formation of TA was observed from the beginning
with a steady rate up to 80 min, and then the rate became slower but
the formation continued until the end of the reaction. Oxygen
consumption continued at a steady rate up to about 80 min and then
almost stopped after then. Thus 80 min seems to be a proper reaction
time to convert PX to TA without much change in the amount of pTA. Further reaction beyond 80 min converted only p-TA to TA,
probably using oxygen dissolved in water.
Several control experiments were carried out to elucidate the
nature of this co-oxidation reaction. First, when 2,6-di-tert-butyl4-methoxyphenol (Aldrich, 97%) was added to the reaction system
as a radical scavenger, the reaction did not occur at all. This is
evidence that the co-oxidation of PX and p-TA in water is a freeradical reaction. From the relative study with substituted toluenes,
Hronec and Ilavsky [7] also proposed that the reaction proceeded
through a free-radical mechanism. This differs from the known
catalytic oxidation chemistry of alkyl aromatics in acetic acid
solvent, where an electron transfer mechanism prevails and Co(II)
acetate is more effective than Mn(II) acetate [16].
Table 1 lists results of other control experiments to show the
unique role of p-TA. Under the standard condition with Mn(II)
acetate in water without p-TA addition, PX conversion was 48.4%,
Table 1
Result of the co-oxidation of PX with various materials.a.
Reactant
PX conversion (%)
Formed TA (g)
PX 188 mmol
PX 75.5 mmol + p-TA 112.5 mmol
p-TA 188 mmol
PX 75.5 mmol + 4-CBA 112.5 mmol
PX 75.5 mol + TA 112.5 mol
48.4
100.0
20.0
4.1
0.4
13.8
0.5
7.6
a
Reactant 188 mmol, water 6.5 g, Mn(OAc)24H2O 0.225 g (0.9 mmol), 180 8C,
27 atm (O2 8 atm, N2 5 atm, water vapor 11 atm, PX vapor 3 atm), reaction time 3 h.
but the amount of formed TA was very small (0.4 g). Thus the
majority of PX was converted only to p-TA and a small fraction of it
was converted all the way to TA. When both PX and p-TA were
present as reactants, PX conversion was 100%, and TA was formed
with an apparent yield (based on PX consumption) of 111%. The yield
higher than 100% indicates that a signicant fraction of TA came
from p-TA. When p-TA alone was oxidized at the same condition, the
reactivity was negligible. When 4-CBA, another important intermediate in PX-to-TA conversion was used, substantial degrees of PX
conversion and TA formation were observed. But the effect was not
as good as the combination of p-TA and PX. Finally, TA itself did not
show any help in the conversion of PX. The results have
demonstrated the unique effectiveness of p-TA/PX co-oxidation to
transform PX to TA in water with Mn(II) acetate as the sole catalyst.
In another experiment, about 10% of deuterated PX (C6D4(CH3)2)
was mixed with the normal PX and co-oxidized with normal p-TA
under the standard conditions. Table 2 shows the isotope distributions in each product at reaction times of 30 and 60 min. Note that PX
is transformed to TA through these intermediates, i.e. PX ! ptolualdehyde ! p-TA ! 4-CBA ! TA. Most of the isotopes were
found in p-tolualdehyde and p-toluic acid at 30 min and then, they
spread out to the next step molecules with time of the reaction. The
result indicates that most of TA is originated from initial p-TA, while
PX is transformed to p-TA as shown below:
(3)
57
Table 3
Effects of CO2 addition replacing a part of oxygen.a.
Total pressure
(atm)
CO2 fraction in
gas phase
O2 fraction in
gas phase
Residual PX
after reaction (g)
p-Toluic
acid (g)
4-CBA (g)
TA (g)
Generated COx
from reaction (g)
27
30
33
37
0.000
0.109
0.184
0.259
0.296
0.267
0.242
0.216
2.24
2.05
1.85
2.94
12.54
10.66
10.15
13.97
0.86
0.76
0.77
0.61
10.93
11.69
12.43
8.02
1.12
0.78
0.70
0.35
a
PX 8 g (75.5 mmol), p-TA 15.3 g (112.5 mmol), water 6.5 g, Mn(OAc)24H2O 0.225 g (0.9 mmol), 195 8C, reaction time 3 h. The base case was O2 8 atm, N2 5 atm, water
vapor 11 atm and PX vapor 3 atm.
PX (g)
p-Toluic
acid (g)
4-CBA (g)
TA (g)
Generated
COx (g)
Base case
Ni
Zr
Ti
Fe
Cr
2.24
2.39
2.00
1.87
3.00
1.95
12.54
10.11
10.96
11.27
10.52
11.28
0.86
0.79
0.88
0.69
1.09
0.74
10.93
11.22
11.01
12.09
10.59
11.59
1.12
1.17
1.14
1.30
1.24
1.25
a
PX 8 g (75.5 mmol), p-TA 15.3 g (112.5 mmol), water 6.5 g, Mn(OAc)24H2O
0.225 g (0.9 mmol), 195 8C, 27 atm (O2 8 atm, N2 5 atm, water vapor 11 atm and PX
vapor 3 atm), co-catalyst 0.045 mmol, reaction time 3 h.
Table 5
Effects of various Ti concentrations in manganese acetate-catalyzed co-oxidation of
PX and p-TA in water.a.
Amount of Ti
(mmol)
PX (g)
p-Toluic
acid (g)
4-CBA (g)
TA (g)
Generated
COx (g)
0.0000
0.0225
0.0450
0.0675
2.24
2.05
1.87
2.25
12.54
10.11
11.27
10.30
0.86
0.74
0.69
0.68
10.93
11.41
12.09
12.02
1.22
1.47
1.30
1.20
a
PX 8 g (75.5 mmol), p-TA 15.3 g (112.5 mmol), water 6.5 g, Mn(OAc)24H2O
0.225 g (0.9 mmol), 195 8C, 27 atm (O2 8 atm, N2 5 atm, water vapor 11 atm and PX
vapor 3 atm), TiCl4, reaction time 3 h.
(4)
PX (g)
p-Toluic
acid (g)
4-CBA (g)
TA (g)
Generated
COx (g)
Base case
Ni
Zr
Ti
Fe
Cr
2.24
1.90
2.10
1.85
2.02
2.10
12.54
10.04
10.21
10.24
9.98
10.16
0.86
0.66
0.82
0.59
0.92
0.64
10.93
12.91
11.48
14.02
12.88
13.66
1.12
0.86
0.84
0.91
0.84
0.89
a
PX 8 g (75.5 mmol), p-TA 15.3 g (112.5 mol), water 6.5 g, Mn(OAc)24H2O
0.225 g (0.9 mmol), 195 8C, 33 atm (O2 8 atm, N2 5 atm, water vapor 11 atm, PX
vapor 3 atm and CO2 6 atm), co-catalyst 0.045 mmol, reaction time 3 h.
58
Acknowledgements
This work has been supported by National Research Laboratory
program, Korean Institute of Industrial Technology Evaluation &
Planning and Korean National Cleaner Production Center, and
sponsored by Samsung Petrochemical Corporation.
References
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
http://www.hyosung.com/eng/product/chemical/tpa.jsp.
http://www.kpia.or.kr/eng/statistics/product04.htm.
US Patent 3,584,039 (June 8, 1971), to Standard Oil Company.
C.M. Li, J. Hou, Y. Yu, H. Hou, Environ. Sci. 12 (2000) 266.
US Patent 4,278,810 (July 14, 1981), to Labona S.A.
US Patent 4,334,086 (June 8, 1982), to Labona S.A.
M. Hronec, J. Ilavsky, Ind. Eng. Chem. Prod. Res. Dev. 21 (1982) 455.
Y. Ishii, S. Sakaguchi, Catal. Today 117 (2006) 105.
J.B. Dunn, D.I. Urquhart, P.E. Savage, Adv. Synth. Catal. 344 (2002) 385.
J.S. Yoo, S.H. Jeong, K.H. Lee, Y.S. Park, Appl. Catal. A 223 (2002) 239.
M. Hronec, J. Ilavsky, Ind. Eng. Chem. Prod. Res. Dev. 24 (1985) 787.
M. Hronec, J. Ilavsky, Oxid. Commun. 9 (1986) 227.
M. Hronec, J. Ilavsky, React. Kinet. Catal. Lett. 33 (1987) 323.
W. Partenheimer, Catal. Today 23 (1995) 69.
Y. Cheng, X. Li, L. Wang, Q. Wang, Ind. Eng. Chem. Prod. Res. Dev. 45 (2006) 4156.
R. Sheldon, J.K. Kochi, Adv. Catal. 25 (1976) 272.
K. Weissermal, H.J. Arpe, Industrial Organic Chemistry, Weinheim, 1978,, p. 342.
M. Wei, G.T. Misue, D.H. Busch, B. Subramaniam, J. Am. Chem. Soc. 124 (2002)
2513.
US Patent 6,194,607 (February 27, 2001), to Samsung General Chemicals Co.
J.S. Yoo, P.S. Lin, S.D. Elne, Appl. Catal. A 105 (1993) 259.
Korean Patent 10-0427298 (April 2, 2004), to Korea Research Institute of Chemistry.
W. Partenheimer, J. Mol. Catal. A 206 (2003) 131.
W. Partenheimer, J. Mol. Catal. A 206 (2003) 105.