Kinetics and Mechanisms of Transalkylation and Disproportionation of meta-Diethylbenzene by Triflic Acid Catalyst

Article abstract of DOI:10.1002/kin.10152

The kinetics of transalkylation and isomerization of meta-diethylbenzene in the presence of benzene using triflic acid as a catalyst has been investigated. High catalytic activity of the triflic acid catalyst was observed in homogeneous liquid-phase reactions. On the basis of the product distribution obtained, transalkylation, disproportionation, and isomerization reactions have been considered and the main product of the reaction was ethylbenzene. These reactions are conducted in a closed liquid batch reactor with continuous stirring under dry nitrogen and atmospheric pressure over the temperature range of 288-308 K. The main transalkylation, disproportionation, and isomerization reactions occurred simultaneously and were considered as elementary reactions. The apparent activation energy of the transalkylation reaction was found to be 35.5 kj/mol, while that of disproportionation reaction was 42.3 kj/mol. The reproducibility of the experimental product distribution occurred with an average relative error of ±2%.

Full text of DOI:10.1002/kin.10152

Kinetics and Mechanisms
of Transalkylation and
Disproportionation
of meta-Diethylbenzene
by Triflic Acid Catalyst
1
2
1
2
S. M. AL-ZAHRANI, M. C. AL-KINANY, K. I. AL-HUMAIZI, S. H. AL-KHOWAITER
1
Chemical Engineering Department, King Saud University, P.O. Box 800, Riyadh 11421, Saudi Arabia
Petroluem and Petrochemical Research Institute, KACST, P.O. Box 6086, Riyadh 11421, Saudi Arabia
2
Received 30 October 2002; accepted 12 June 2003
DOI 10.1002/kin.10152
ABSTRACT: The kinetics of transalkylation and isomerization of meta-diethylbenzene in the
presence of benzene using triflic acid as a catalyst has been investigated. High catalytic ac-
tivity of the triflic acid catalyst was observed in homogeneous liquid-phase reactions. On the
basis of the product distribution obtained, transalkylation, disproportionation, and isomeriza-
tion reactions have been considered and the main product of the reaction was ethylbenzene.
These reactions are conducted in a closed liquid batch reactor with continuous stirring under
dry nitrogen and atmospheric pressure over the temperature range of 288–308 K. The main
transalkylation, disproportionation, and isomerization reactions occurred simultaneously and
were considered as elementary reactions. The apparent activation energy of the transalkylation
reaction was found to be 35.5 kJ/mol, while that of disproportionation reaction was 42.3 kJ/mol.
The reproducibility of the experimental product distribution occurred with an average relative
error of ± 2%. ꢀC 2003 Wiley Periodicals, Inc. Int J Chem Kinet 35: 555–563, 2003
INTRODUCTION
pounds of ethylbenzene (EB) was produced in the
United States, which ranked 19th among all chemicals
produced in the United States, representing an increase
of 27% from the 1994 production volume of 10.76
billion pounds. In 1998, the global operating rate of
EB was 83%. The operating rate for facilities in North
America averaged 90%, in Western Europe 85% and
in Asia 84%. Japan accounted for about 50% of the
Asian EB demand in 1998. Overall EB demand for
the world increased at an average annual rate of 4.3%
from 1998 to 2003, resulting in global EB demand of
nearly 25 million metric tons. In 2003 consumption is
expected to grow the fastest in the Middle East and
Dialkylbenzenes include xylene, diethylbenzene, and
diisopropylbenzene which are very important raw
materials for many intermediates of commodity
petrochemicals and valuable fine chemicals such as
monomers for engineering plastics, polyesters, inter-
mediates for detergents, pharmaceuticals, agricultural
products, and explosives [1]. In 1995, 13.66 billion
Correspondence to: S. M. Al-Zahrani; e-mail: szahrani@
ksu.edu.sa.
ꢀ
c 2003 Wiley Periodicals, Inc.

556
AL-ZAHRANI ET AL.
South America. Almost all of the EB produced in the
world is used in the manufacture of styrene.
Streitweisere and Reif [12] investigated the kinetics
of transalkylation of radiolabeled and optically labeled
EB in benzene with a gallium tribromide–hydrogen
bromide catalyst. Moore and Wolf [13] have demon-
strated from a study of disproportionation and isomer-
Transalkylation and disproportionation are the two
major practical processes for the utilization of aromat-
ics. They are coined as “alkyl group transfer reactions.”
The catalytic rearrangement of alkyl groups present in
alkyl aromatic hydrocarbons to provide one or more
products suitable for use in the petroleum and chem-
ical industries has heretofore been effected by a wide
variety of catalysts [1–8]. Such processes are com-
monly used in conversion of diethylbenzene (DEB),
which yields EB, isomers of DEB (ortho-, para-, and
meta-), triethylbenzene, and tetraethylbenzene. Cur-
rent EB production processes are based on commercial
catalysts that are effective above 473 K. Developing
new catalysts that work at lower temperatures will in-
crease the operational and economical benefits of these
processes.
Acidic halides such as aluminum chloride, alu-
minum bromide, boron trifluoride–hydrogen fluoride
mixtures, etc. have been used in the rearrangement of
alkyl benzenes to provide valuable intermediates that
find utility in the synthesis of rubber, plastic, fibers, and
dyes. Triflic acid, owing to its catalyst stability, which
is far superior to that of other acids, and its resistance
to oxidation and reduction reactions, is particularly a
valuable reactant and solvent. It does not split off fluo-
rine ions, even in the presence of strong nucleophiles.
Triflic acid has been used in the plastic industry as
an oligomerization–polymerization catalyst, as well as
for the production of electrically conductive polymers.
Also it has been used as a protonization catalyst in the
fuel industry besides its utilization in the pharmaceuti-
cal, sugar, and vitamins industries. In addition to triflic
acid and its derivatives, in the form of acid halides,
anhydrides and esters are also used. Mixed anhydrides
formed from triflic acid are extremely strong Friedel–
Crafts alkylating agents. The salts of triflic acids are
used in microelectronics, polymer membrane technol-
ogy, and in preparative organic synthesis.
1
4
ization reactions of ethylbenzene–1- C in AlBr3–HBr
that para alkylation is dominant followed by an in-
tramolecular 1,2-shift of an ethyl group to give the
thermodynamically most stable isomer. Bakoss et al.
[14] have measured the rates of disproportionation of
EB and m-DEB using triflic acid as catalyst at room
temperature and indicated that the EB disproportion-
ates very rapidly, whereas m-DEB reacts at a conve-
niently measurable rate and the reactions obey first-
order kinetics. Al-Zahrani et al. [15] have studied the
kinetics of transalkylation and isomerization of o-DEB
in the presence of benzene using trifluoromethanesul-
phonic acid as a catalyst. Power law type model has
been tested for the main transalkylation and dispropor-
tionation reactions, while the isomerization reactions
followed a first-order mechanisms. The apparent acti-
vation energy of the transalkylation reaction was found
to be 50 kJ/mol, while that of disproportionation reac-
tion was 29 kJ/mol. Recently Cavani et al. [16] inves-
tigated the transalkylation of DEB with benzene over
ꢀ-zeolite in the liquid phase over temperature ranges
of 513–573 K. These authors confirmed that EB was
formed via the direct reaction of DEB transalkylation
with benzene.
The aim of the present work is to develop a kinetic
model for the transalkylation, disproportionation, and
isomerization of m-DEB using triflic acid catalyst. The
model will be used to understand the physiochemical
nature of the process and to interpret the experimental
results via developing rate equations for the transalky-
lation of m-DEB with benzene, and disproportionation
of DEB and isomerization reactions.
EXPERIMENTAL
Materials
Transalkylation reaction for the synthesis of EB
is usually carried out on an industrial scale, employ-
◦
ing aluminum chloride or trifluoroboron at 100 C or
◦
zeolite catalysts at temperatures higher than 200 C.
The materials used in this reaction include benzene,
m-DEB, trifluoromethanesulphonic acid (triflic acid)
as reactants and hexane as a solvent. Benzene was ob-
tained from Sigma with purity of 99.9%. It was also
purified according to the standard method and dried
prior to use over sodium wire. Triflic acid was a com-
mercial sample obtained from Fluka Chemie with pu-
The transalkylation of DEB with benzene has also
been investigated over superacidic catalyst, such as
Nafion-H [9]. The major reactions include isomeriza-
tionandtransalkylationwiththeirrelativecontributions
controlled by the nature of the catalyst and the reaction
conditions. Considerable efforts have been aimed at
the isomerization and disproportionation of DEB iso-
mers in the absence and presence of benzene to pro-
duce EB using trifluromethanesulphonic acid catalyst
3
rity of 98% and density of 1.696 gm/cm . The acid
was purified by double distillation under dry nitrogen
at reduced pressure immediately before use. Hexane
(HPLC grade) was used as solvent for reaction product
[10,11].

TRANSALKYLATION AND DISPROPORTIONATION OF meta-DIETHYLBENZENE
557
extraction and was obtained from Fluka. m-DEB was
RESULTS AND DISCUSSION
also obtained from Fluka with a purity of 97% and was
used directly without further purification. The physical
and chemical properties of triflic acid, m-DEB, and EB
are listed in Table I.
m-DEB reacted with benzene in the presence of tri-
flic acid catalyst. Triflic acid has a great advantage of
providing a homogeneous medium of high acidity and
is much more easily handled than the Friedel–Crafts
catalytic system [14]. The reaction proceeds via rapid
transalkylation to EB and disproportionation to tri-
ethylbenzene as well as isomerization to o- and p-DEB.
Figure 1 shows conversion of m-DEB at three different
temperatures versus reaction time. Obviously increas-
ing the reaction time will increase the m-DEB conver-
Reaction System
The reaction was performed in a closed liquid batch
reactor with continuous stirring under dry nitrogen en-
vironment and atmospheric pressure over the tempera-
◦
ture range of 15–35 C. The batch reactor was equipped
◦
with a PID temperature controller to monitor the reac-
tion temperature. The reaction procedure involved first
flushing the reactor with nitrogen, and then introduc-
ing weighted sample according to the required ratio of
m-DEB, benzene, and catalyst. The reaction mixture
was heated to the chosen temperature under continuous
stirring. The reaction continued for 6 h. The analysis
of the reaction mixture was carried out using a Varian
sion. After 6 h of reaction time 30% of m-DEB at 15 C,
◦
◦
50% at 25 C, and 60% at 35 C were converted. Dur-
ing the reaction time, small amounts of triethylbenzene
were found among the reaction products. This indicates
that a part of m-DEB undergoes the disproportionation
reaction (R2) to EB and triethylbenzene. P-DEB and
o-DEB isomers were also found among the reaction
products [23].
3
400 chromatograph equipped with a 60 m × 0.32 mm
id capillary column (phase DBI, film thickness of
ꢁm) and a flame ionization detector. The instrument
In accordance with the product distribution ob-
tained, the system can be described with the following
reactions:
1
was calibrated by external standardization method us-
ing standard mixture of benzene, ethylbenzene, m-,
p-, and o-DEB, 1,3,5-triethylbenzene, and 1,2,4,5-
tetraethylbenzene. The ratios of m-DEB to benzene
Transalkylation of m-DEB with benzene (R1):
(B) and to catalysts (i.e., m-DEB/B/catalyst) ranged
from 1:1:1 to 1:6:1 at a pressure of 1 atm. In all the
different experiments, the products distributions were
followed by analyzing samples of the organic layer at
(
R1)
0
.5-h intervals after quenching with deionized water
and extracting with hexane.
Disproportionation of m-DEB (R2):
Table I Chemical and Physical Properties of
Reactants, Products, and Catalyst
Triflic acid
Formula and structure
Molecular weight
Appearance at 20 C
CF3SO3H
150.08 gm/mol
Colorless to yellowish
liquid, pungent
◦
(
R2)
3
Density
Boiling point
1.7 g/cm
162 C
◦
Neutralization equivalent
149–151.5 (min. 99%)
Isomerization reactions to o- (R3) and p-DEB (R4):
(content)
meta-Diethylbenzene (m-DEB)
Formula and structure
Appearance
C6H4(C2H5)2
Colorless limpid liquid
181.1
−84.2
◦
Boiling point ( C)
◦
Melting point ( C)
Ethylbenzene (EB)
◦
Boiling point ( C)
136.2
−95.0
◦
Melting point ( C)
(
R3)

558
AL-ZAHRANI ET AL.
Figure 1 The conversion of m-DEB vs. reaction time at m-DEB/B/catalyst 1:1:1 (molar ratio) at three different temperatures.
and the isomerization reactions are expressed by
r3 = k f 3Cm-DEB − kb3Co-DEB
r4 = k f 4Cm-DEB − kb4Cp-DEB
r5 = k f 5Co-DEB − kb5Cp-DEB
(3)
(4)
(5)
(R4)
Based on the given kinetic scheme, the components
mass balance equations are shown below in the differ-
ential form:
Isomerization reactions of o- and p-DEB (R5):
Consumption rate of m-DEB:
dCm-DEB
=
−r1 − 2r2 − r3 − r4
(6)
(7)
dt
Consumption rate of benzene:
(R5)
dCB
=
−r1
dt
In the kinetic model developed here, the transalky-
lation, disproportionation, and isomerization reac-
tions are treated as elementary reactions. The rate of
transalkylation reaction is expressed in the following
form:
Formation rate of ethylbenzene:
dCEB
=
2r1 + r2
(8)
(9)
dt
2
r1 = k f 1Cm-DEBCB − kb1CEB
(1)
Formation rate of o-DEB:
while the rate of disproportionation reaction of m-DEB
is given by
dCo-DEB
=
r3 − r5
2
r2 = k f 2Cm-DEB − kb2CEBCTEB
(2)
dt

TRANSALKYLATION AND DISPROPORTIONATION OF meta-DIETHYLBENZENE
559
Figure 2 The EB concentration vs. reaction time at m-DEB/B/catalyst 1:1:1 (molar ratio) at three different temperatures.
Formation rate of p-DEB:
disproportionation of p- or o-DEB are also neglected
because of the small concentrations reported here for
both p- and o-DEB and also to avoid complexity of the
reaction network.
The kinetic model for this batch reaction system
is in the form of ordinary differential equations (i.e.,
Eqs. 5–11). The integral form of the above differential
equations was used to determine the kinetic parame-
ters because of high m-DEB conversion obtained in this
work. The objective is to determine the kinetic parame-
ters listed on the rate equations of the kinetic schemes,
which will be used to predict the m-DEB and benzene
dCp-DEB
=
r4 + r5
(10)
(11)
dt
Formation rate of triethylbenzene (TEB):
dCTEB
=
r2
dt
The direct transalkylations from p- or o-DEB are
neglected reactions in this work. Other reactions like
Figure 3 The o-DEB concentration vs. reaction time at m-DEB/B/catalyst 1:1:1 (molar ratio) at three different temperatures.

560
AL-ZAHRANI ET AL.
Figure 4 The p-DEB concentration vs. reaction time at m-DEB/B/catalyst 1:1:1 (molar ratio) at three different temperatures.
conversions and EB and other product selectivities and
yield. Therefore, the problem requires a combination
of nonlinear differential equation solver and nonlin-
ear regression analysis technique. For this objective, a
computer program was developed to solve the problem,
which combined an integrator such as the Runge–Kutta
method with the least-square approximation technique
to solve for the best fitting kinetic parameters. More
than 30 experimental data points at each temperature
have been used to determine these kinetic parameters,
which require the solution of the above six nonlinear
differential equations simultaneously for these param-
eters to be obtained.
The simulation results reveal that the rate of back-
ward reactions for the transalkylation and dispropor-
tionation steps are small compared to the forward re-
actions and therefore can be neglected. These results
are in agreement with the recent findings reported
Figure 5 The TEB concentration vs. reaction time at m-DEB/B/catalyst 1:1:1 (molar ratio) at three different temperatures.

TRANSALKYLATION AND DISPROPORTIONATION OF meta-DIETHYLBENZENE
561
Table II Example of Measured Product Distribution
mol%) at m-DEB/B/Catalyst 1:1:1 (Molar Ratio) at Three
Different Temperatures
For the main disproportionation reaction the rate con-
stant is given by the formula
(
ꢀ−
ꢁ
4967
Product Distribution
mol%; at t = 4 h)
5
k f 2 = 2.02371 × 10 × exp
(13)
(
T
Compound
T = 288 K T = 298 K T = 308 K
The forward and backward rate constants for the iso-
merization reactions [i.e. reactions (3)–(5)] are given
below:
EB
o-DEB
12.6
0.3
24.2
0.60
25.7
0.7
p-DEB
TEB
Unreacted m-DEB
Unreacted benzene
1.3
0.98
34.5
46.5
3.61
2.45
27.4
3.7
2.6
23.4
41.0
ꢀ−
ꢁ
4441
5
k f 3 = 2.1445 × 10 × exp
(14)
(15)
(16)
(17)
(18)
(19)
T
ꢀ−
ꢁ
41.6
2231
3
Total
96.2%
99.86%
97.1%
kb3 = 5.2402 × 10 × exp
T
ꢀ−
ꢁ
5472
6
k f 4 = 6.71987 × 10 × exp
T
by Ngandjui et al. [17]. The kinetic expression re-
action obeys first-order with respect to m-DEB and
benzene, while the rate of the disproportionation re-
action obeys second-order kinetics over the studied
range. The kinetic parameters for the above reaction
scheme are listed in Table I. Bakoss [14] reported
that the disproportionation of DEB obeys first-order
kinetics.
ꢀ−
ꢀ−
ꢀ
ꢁ
ꢁ
ꢁ
1092
1
kb4 = 1.74726 × 10 × exp
T
4017
5
k f 5 = 2.73633 × 10 × exp
T
−1455
1
kb5 = 4.39459 × 10 × exp
T
For all of the kinetic parameters listed above the
Arrhenius form was substituted before starting the
simulation. For the main transalkylation reaction the
reaction rate constants has the following Arrhenius
expression:
Figures 2 and 3 show the model predictions (solid
lines) along with the experimental values (symbols)
for EB and TEB, respectively, versus reaction time at
ratio 1:1:1 and temperatures of 288, 298, and 308 K.
These figures indicate that the kinetic model predicts
fairly well the number of moles with time. Increasing
the temperature will lead to an increase in both EB and
ꢀ−
ꢁ
4270
4
k f 1 = 3.6602 × 10 × exp
(12)
T
Figure 6 The m-DEB conversion vs. EB selectivity at DEB/B/catalyst 1:1:1 (molar ratio) and T = 298 K.

562
AL-ZAHRANI ET AL.
Figure 7 The benzene concentration vs. reaction time at m-DEB/B/catalyst 1:1:1 (molar ratio) at three different temperatures.
TEB. The selectivity for EB almost doubled when the
temperature increased from 288 to 298 K, while it was
increased only by 10% when the temperature increased
further from 288 to 308 K. Figures 4 and 5 show the
mole predictions along with the experimental data for
mol% of o-DEB and p-DEB respectively. The mol%
of o-DEB varied between 0 and 0.6%, while that of p-
DEB varied between 0 and 5%. These figures indicated
that the isomerization reactions are small compared
to the transalkylation reaction and these results are in
agreement with the results obtained by Cavani et al.
is accompanied with a slight drop in EB selectivity.
Figure 7 shows the model predictions of mol% for ben-
zenealongwiththeexperimentalvaluesversusreaction
time at ratio 1:1:1 and the above-mentioned tempera-
tures. Higher temperature leads to higher conversion
of benzene. The conversion of benzene increases with
increasing the reaction time.
Most of the studies reported in the literature on aro-
matics transalkylation were conducted in the gaseous
phase over zeolite catalysts; however, few studies were
conducted in the liquid phase. The activation energies
for such reactions are listed in Table III. From the
results illustrated in Table III it can be observed that
the activation energy calculated in this work falls in
the range of the ones reported in several publications
for DEB or toluene transalkylation in both liquid and
gaseous phases.
[16]. An example for the measured molar distribution
of the products with time is shown in Table II.
Figure 6 shows the m-DEB conversion versus the
EB selectivity at 288 K. When the conversion increases
from 0 to 55%, the selectivity drops only from 84 to
7
8%, which indicates that increasing the conversion
Table III Comparison of Activation Energies for Transalkylation and Disproportionation Over Different Catalytic
Systems
Activation Energy
(kJ/mol)
Authors [Ref.]
Reaction
Phase
Catalyst
Ngandjui et al. [17]
Corma et al. [18]
Betrame et al. [19]
Dooley et al. [20]
Bhaskar and Do [21]
Transalkylation of DEB-BZ
Transalkylation of aromatics
Transalkylation of toulene
Transalkylation of toulene
Liquid
Y(Si/Al = 13.8)
81
107
85.8
87 ± 10
100–121
53
Gaseous HYD-3(Si/Al = 12.8)
Gaseous USY-3 (Si/Al = 12) Various zeolite
Gaseous
Y
Disproportionation of toluene Gaseous HZSM-5
Nayak and Riekert [22] Disproportionation of toluene Gaseous Pentasil (Si/Al = 36)
Bakoss et al. [14]
Al-Zahrani et al. [15]
Cavani et al. [16]
This work
Disproportionation of m-DEB Liquid
Triflic acid
Triflic acid
ꢀ-Zeolite
66 ± 2
Transalkylation of o-DEB
Transalkylation of DEB
Transalkylation of m-DEB
Liquid
Liquid
Liquid
50
130
35.5
Triflic acid

TRANSALKYLATION AND DISPROPORTIONATION OF meta-DIETHYLBENZENE
563
CONCLUSIONS
8. Cejka, J.; Wichetrlova, B.; Bednarova, S. Appl Catal, A
991, 79, 215.
. Olah, G. A.; Kaspi, J.; Nouveau, J. DE Chieme 1978,
(6), 585.
0. Al-Kinany, M. C.; Al-Khowaiter, S. H. In Proceedings of
nd Middle East and Refining and Petrochemicals Con-
1
9
Adequate kinetic models describing the transalkyla-
tion, disproportionation, and isomerization of m-DEB
over triflic catalyst were derived. Experiments over
a wide variety of experimental conditions were per-
formed in a liquid batch reactor with continuous stir-
ring under dry nitrogen and atmospheric pressure over
the temperature range of 288–308 K and kinetic pa-
rameters were estimated. The predicted conversion and
product selectivities were found to agree well with the
experimental data. In addition, such catalytic systems
were found to be highly stable with a very good activity
under lower reaction temperature environments.
2
1
1
2
ference and Exhibition, Petrotech’98, Bahrain, 1998;
Vol. 1, p. 523.
1. Al-Kinany, M. C.; Al-Khowaiter, S. H. In Proceedings of
the 2nd International Symposium on Reaction Kinetics
and the Development of Catalytic Processes; Froment,
G. F.;Waugh, K. C. (Eds.);Brugge, Belgium, April1999;
Vol. 122, p. 375.
12. Streitweisere, A.; Reif, L. J Am Chem Soc 1963,
8
5, 3454.
1
1
3. Moore, G. G.; Wolf, A. P. J Org Chem 1966, 31, 1106.
4. Bakoss, H. J.; Roberts, R. M. G.; Sadri, A. R. J Org Chem
1
982, 47, 4053.
The authors thank Mr. Ali Al-kharji, Mr. Fahad Al-Malki,
and Mr. Rasheed Al-Rasheed for their assistance in the ex-
perimental works.
1
1
1
1
5. Al-Zahrani, S. M.; Al-kinany, M. C.; Alhumaizi, K. I.;
Al-khowaiter, S. H. Chem Eng Proc 2002, 41, 321.
6. Cavani, F.; Corazzari, M.; Bencini, E.; Goffredi, G. Appl
Catal, A 2002, 226, 31.
7. Tiako Ngandjui, L. M.; Louhibi, D.; Thyrion, F. C.
Chemi Eng Processing 1997, 36, 133.
8. Corma, A.; Liopis, F.; Monton, J. B. J Catal 1993, 140,
384.
19. Betrame, P.; Beltrame, P. L.; Carniti, P.; Zuretti, G.;
Forni, L.; Zuretti, G. Zeolites 1985, 5, 400.
20. Dooley, K. M.;Brignac, S. D.;Price, G. L. IndEngChem
Res 1990, 29, 789.
BIBLIOGRAPHY
1
2
3
4
5
. McFarlane, A. C. Oil Gas J 1976, 99.
. Dwyer, F. G.; Lewis, P. J. Chem Eng 1976, 83(1), 90.
. Jones, E. K. Oil Gas J 1960, 58(9), 80.
. Grote, H. W. Oil Gas J 1958, 56, 73.
. Kecker, K. A.; Karge, H. G.; Streubel, W. D. J Catal
1
973, 28, 403.
21. Bhaskar, G. V.; Do, D. D. Ind Eng Chem Res 1990,
29, 355.
22. Nayak, V. S.; Riekert, L. Appl Catal, A 1986, 23, 403.
23. Klemm, E.; Emig, G. Microporous Mater 1997,
12, 281.
6
7
. Morita, Y.; Matsumoto, H.; Kiruma, T.; Kato, F.;
Takayasu, M. Bull Jpn Pet Inst 1973, 15, 37.
. Shamshoum, E. S.; Schuler, T. R.; Ghosh, A. K. US
Patent 5 227 558, 1993.