Synthesis of linear phenyldodecanes by the alkylation of benzene with 1-dodecene over non-zeolitic catalysts

Article abstract of DOI:10.1021/op000044s

Linear alkylbenzenes (LAB) are typically manufactured by the alkylation of benzene and α-olefin, employing HF or AlCl₃ as catalyst. LAB are the precursors of linear alkylbenzene sulphonates (LABS) used in a variety of industries. Various acid catalysts are being explored by different researchers, and zeolites are claimed to be effective. The isomer distribution depends strongly on the type and nature of the catalyst and reaction conditions. The liquid-phase alkylation of benzene with 1-dodecene was examined by using several non-zeolites based on clays, pillared clays, and clay-supported heteropolyacids such as dodecatungstophosphoric acid (DTP), dodecatungstosilicic acid (DTS), and dodecamolybdophosphoric acid (DMP). The activities and selectivities of K-10 clay, 20% w/w heteropolyacids (DTP, DMP, and DTS) supported on K-10, Filtrol-24, Al pillared clay, 20% w/w DTP/silica,10% AlCl₃/10% FeCl₃/K-10, DTP, Cr-exchanged K-10, sulphated zirconia, Zr-exchanged K-10, and 20% DTP/activated carbon were evaluated. It was found that 20% w/w DTP/K-10 clay offered the best conversion with favourable product distribution. A molar ratio of 10:1 benzene/1-dodecene favoured the formation of linear dodecylbenzenes (LAB). However, with decreasing benzene concentration, the formation of didodecylbenzenes increased. The best parameters for the alkylation were established. A mechanistic and kinetic model was developed and validated against experimental data. Benzene alkylation was also accomplished with 1-octene, 1-decene, and 1-tetradecene under otherwise similar sets of conditions. It was found that the rate of benzene alkylation decreased with an increase in the chain length of α-olefin.

Full text of DOI:10.1021/op000044s

Organic Process Research & Development 2002, 6, 263−272
Synthesis of Linear Phenyldodecanes by the Alkylation of Benzene with
1-Dodecene over Non-Zeolitic Catalysts
G. D. Yadav*,† and N. S. Doshi
Department of Chemical Engineering, UniVersity Institute of Chemical Technology (UICT), UniVersity of Mumbai,
Matunga, Mumbai - 400 019 India
Abstract:
isomer distribution in the product LAB strongly depends on
the nature and distribution of acidity (Brønsted or Lewis)
and amount of catalyst, solvent, temperature and the position
of the double bond in the starting olefin.6-12 The position of
the phenyl group and the alkyl chain length in LABS are
important factors in determining their surface active proper-
ties and biodegradability.13,14 Both HF and AlCl3 being
homogeneous acid catalysts lead to formation of all 5 isomers
from 2- to 6-phenyldodecanes but with varying distribution8
and these catalysts are a source of pollution and corrosion
of reactors and pose disposal problems. Further, its being
hazardous and highly corrosive, HF requires special handling
and equipment design. Thus there is a tremendous scope to
develop ecofriendly, reusable, noncorrosive and cheap
catalysts for the alkylation of benzene with R-olefins.
Alkylation of benzene with 1-dodecene was found to yield
mainly monoalkylated product with alumina pillared clay as
catalyst,15 whereas silica gel was shown to be superior
support for silicotungstic acid (STA) than alumina and
silica-alumina and 20% w/w/STA/silica gel led to a conver-
sion of 98% with 90% selectivity to monoalkylated product.16
Different crystalline aluminosilicates are claimed to be active
catalysts for alkylation of aromatics.17-20. For instance,
alkylation of benzene was effectively carried out with
propene, 1-butene and 1-decene over rare earth exchanged
X and Y zeolites21 and with an industrial mixture of higher
Linear alkylbenzenes (LAB) are typically manufactured by the
alkylation of benzene and r-olefin, employing HF or AlCl3 as
catalyst. LAB are the precursors of linear alkylbenzene sul-
phonates (LABS) used in a variety of industries. Various acid
catalysts are being explored by different researchers, and
zeolites are claimed to be effective. The isomer distribution
depends strongly on the type and nature of the catalyst and
reaction conditions. The liquid-phase alkylation of benzene with
1-dodecene was examined by using several non-zeolites based
on clays, pillared clays, and clay-supported heteropolyacids such
as dodecatungstophosphoric acid (DTP), dodecatungstosilicic
acid (DTS), and dodecamolybdophosphoric acid (DMP). The
activities and selectivities of K-10 clay, 20% w/w heteropolyacids
(DTP, DMP, and DTS) supported on K-10, Filtrol-24, Al
pillared clay, 20% w/w DTP/silica,10% AlCl3/10% FeCl3/K-
10, DTP, Cr-exchanged K-10, sulphated zirconia, Zr-exchanged
K-10, and 20% DTP/activated carbon were evaluated. It was
found that 20% w/w DTP/K-10 clay offered the best conversion
with favourable product distribution. A molar ratio of 10:1
benzene/1-dodecene favoured the formation of linear dodecyl-
benzenes (LAB). However, with decreasing benzene concentra-
tion, the formation of didodecylbenzenes increased. The best
parameters for the alkylation were established. A mechanistic
and kinetic model was developed and validated against experi-
mental data. Benzene alkylation was also accomplished with
1-octene, 1-decene, and 1-tetradecene under otherwise similar
sets of conditions. It was found that the rate of benzene
alkylation decreased with an increase in the chain length of
r-olefin.
(1) Anstett, R. M.; Munger, P. A.; Rubinfeld, J. J. Am. Oil Chem. Soc. 1966,
43, 25.
(2) Rubinfeld, J.; Emery, J. M.; Cross, H. D. J. Am. Oil Chem. Soc. 1964, 41,
8223.
(3) Rubinfeld, J.; Emery, J. M.; Cross, H. D. Ind. Eng. Chem. Prod. Res. DeV.
1965, 4, 33.
(4) Gray, F. W. J. Org. Chem. 1955, 20, 511.
(5) Olson, A. C. Ind. Eng. Chem. 1960, 52, 833.
Introduction
(6) Swisher, R. D.;. Kaelle, E.F; Liu, S. J. Org. Chem. 1961, 26, 4066.
(7) Allen, R. H.; Yats, D. J. Am. Chem. Soc. 1961, 83, 2799.
(8) Alul, H. R. Ind. Eng. Chem. Prod. Res. DeV. 1967, 7, 7.
(9) McCaulay, D. A. In Friedel-Crafts and Related Reactions; Olah, G. Ed.;
Interscience: New York, 1964; Vol. II, Chapter 24.
(10) Lappin, G. R.; Wagner, J. D. In Alpha Olefins Application Handbook;
Lappin, G. R., Ed.; Dekker: New York, 1989; Chapter 8, p 250.
(11) Matheson, K. L.; Matson, T. P. J. Am. Oil Chem. Soc. 1983, 60, 9.
(12) Cavalli, L.; Zatta, A. Proceedings 3rd CESIO World Surfactant Congress;
London, 1-5 June, 1992; p 105.
(13) Larson, R. J. J. Am. Oil Chem. Soc. 1993, 70, 645.
(14) Baumgartner, F. N. Ind. Eng. Chem. 1954, 46, 1349.
(15) Mokaya, R.; Jones, W. Chem. Commun. 1994, 929.
(16) Sebulsky, R. T.; Henke, A. M. Ind. Eng. Chem. Process Des. DeV. 1971,
10, 272.
Linear alkylbenzene sulphonates (LABS), containing
different carbon atoms in the alkyl chain, are the most
important biodegradable surfactants that are used in formula-
tions of light-duty (C11-C12) and heavy-duty (C12-C14)
detergents, lubricants (C16-C24), and surfactant flooding
(C16) in tertiary oil recovery. LABS are manufactured by
sulphonation of linear alkylbenzenes (LAB) which in turn
are traditionally synthesised by the Friedel-Crafts alkylation
of benzene with R-olefins1-5 by employing HF or AlCl3 as
catalyst or with chloroalkanes by using AlCl3. The final
(17) O’Kelly, A. A.; Kellett, J. Ind. Eng. Chem. 1947, 39, 154.
(18) Cullinane, N. M.; Chard, S. J. J. Chem. Soc. 1948, 804.
(19) Kutz, W. M.; Corson, B. B. J. Am. Chem. Soc. 1946, 68, 1477.
(20) Thomas, C. L. Ind. Eng. Chem. 1949, 41, 2564.
(21) Venuto, P. B.; Hamilton, L. A.; Wise, J. J.; Landis, P. S. J. Catal. 1966, 5,
81.
* To whom correspondence should be sent. E-mail: gdyadav@yahoo.com.
† Current address: Johansen-Crosby Visiting Professor of Chemical Engineer-
ing, Department of Chemical Engineering and Material Science, Michigan State
University, East Lansing, MI-48824. Telephone: 517-355-9291. Fax: 517-432-
1105. E-mail: yadavg@egr.msu.edu.
10.1021/op000044s CCC: $22.00 © 2002 American Chemical Society
Published on Web 04/23/2002
Vol. 6, No. 3, 2002 / Organic Process Research & Development
•
263

alkenes over â-zeolite, H-mordenite and ZSM-12 in Sivasank-
er’s laboratory22,23 including a recent report24 on alkylation
of benzene with 1-dodecene on H-USY, HY, H-mordenite,
Mg-H-mordenite, Fe-Mg-H-mordenite and H-ZSM-5.
Mixtures of LABS of different chain length are preferred
in detergents in order to handle any of several end uses in
which a particular formulation may be employed.10 The effect
of carbon chain length and phenyl isomer distribution on
the properties of LABS was studied and a comparison made
of high and low 2-phenyl LABS homologues11 to establish
that the 2-phenyl content had little effect on LABS perfor-
mance in both light- and heavy-duty detergent applications,
and the carbon number chain size was far more important.
However, the biodegradability of 2-phenyl and 3-alkane
isomer is higher than the central ones such as 4-, 5- and 6-
phenyldodecanes.13 Some catalysts do not produce 6-isomer
at all.22,25 Baumgartner14 claimed that optimum detergency
power occurs, when the linear alkyl chain contains 11 to 12
carbon atoms. In the case of n-phenyldocanes and their
sulphonates, the position of benzene ring in the alkyl chain
affects the ease of biodegradability, with external isomers
being more readily degraded than the internal ones.25,26 Thus
2-, 3-, and 4- phenyldodecanes would be preferred for
making LABS from biodegradability viewpoint in compari-
son with the 5- and 6- isomers. The ratio of total concentra-
tions of 2-, 3- and 4- isomers to that of 5- and 6- isomers
(with no branched phenyldocanes or didoceylbenzenes) is
taken as a yardstick of biodegradability and environmental
acceptability27 and less content of 5- and 6-isomers is
beneficial for sulphonation and formulation. The longer the
alkyl chain in a LAB the better is the degradation and
2-phenyldodecane is the most biodegradable in all homo-
logues.
clays, pillared clays, clay supported heteropolyacids, and
sulphated zirconia with a view to find the most active catalyst
and favourable product distribution useful for making LABS.
Some of these catalysts can be synthesised more easily and
cheaply and have proved to be better than zeolites for liquid-
phase reactions. This paper delineates the evaluation of
several catalysts including kinetic modelling.
Experimental Section
Chemicals and Catalysts. 1-Dodecene, 1-octene, 1-decene
and 1-tetradecene (99% pure) were obtained from Albemarle
Corporation, USA. Authentic samples of phenyldodecanes
and phenyldidocanes were obtained from Indian Petrochemi-
cal Ltd, Baroda, India. Benzene (A. R.), dodecatungstophos-
phoric acid (DTP), dodecatungstosilicic acid (DTS), dode-
camolybdophosphoric acid (DMP) were obtained from s.d.
Fine Chemicals Pvt. Ltd, Mumbai, India. K-10 montmoril-
lonite clay was obtained from Fluka, Germany. Filtrol-24
was procured from Engelhard Inc.
The following catalysts were prepared by well-developed
procedures and characterised in this laboratory. 20% w/w
heteropolyacids (DTP, DTS and DMP) supported on K-10
clay,28 Al pillared clay,28 20% DTP/silica,30 10%AlCl3/10%
FeCl3/K-10,28 Cr-exchanged K-10,28 sulphated zirconia,29 Zr-
exchanged K-1028 and 20% DTP/activated carbon.30
Apparatus and Reaction Procedure. All experiments
were conducted in a batch mode by using a 100 mL capacity
Parr autoclave. A standard experiment consisted of 0.041
mol of 1-dodecene and 0.41 mol of benzene and a known
quantity of catalyst (typically 0.05 g/mL). The autoclave was
gradually heated to 150 °C and stirring was then started at
1000 rpm, under autogenous-pressure. Samples were with-
drawn at regular intervals starting from initial time and
analysed by gas chromatography.
From the foregoing it is evident that there is a scope in
developing newer catalytic processes for LAB manufacture,
particularly since there is no open literature on kinetics and
product distribution using non-zeolitic catalysts which may
be cheaper. We have reported the novelty and efficacy of
clay supported heteropolyacids in a large number of pro-
cesses, with reference to a variety of industries such as
pharmaceuticals and drugs, rubber chemicals, dyestuff,
agrochemicals, and perfumery and flavour chemicals.28-39
Thus the aim of the current research was to evaluate the
applicability of a variety of non-zeolitic solid acids, such as
Analysis. A gas chromatograph (Perkin-Elmer Model
8500) with a flame ionisation detector was used with 2m ×
0.003 m column, packed with 10% OV 17/chromosorb WHP.
Synthetic mixtures were used for calibration and quantifica-
tion of the collected data. After the catalyst was filtered,
products were separated by fractional distillation. Benzene
was removed first followed by unreacted 1-dodecene and
then alkylbenzenes were distilled out under vacuum. Iden-
tification was made through GC-MS as well as by compari-
son with authentic material.
(22) Sivasanker, S.; Thangaraj A. J. Catal. 1992, 138, 386.
(23) Sivasanker, S. In Studies in Surface Science and Catalysis; Guisnet, M.,
Ed.; Elsevier: Amsterdam, 1993; Vol. 75, p397.
Results and Discussion
(24) Wang, B.; Lee, C. W.; Cai, T. X.; Park, S. E. Catal. Lett. 2001, 76, 99.
(25) Bhatia, M.; Singh, H. D. J. Biosci. 1996, 21, 487.
(26) Johnson, S. J.; Barry, D. A.; Christofi, N.; Patel, D. Land Contam. Reclaim.
2001, 9, 279.
Comparison of Activities of Different Catalysts. Several
catalysts were employed to evaluate their efficacy in the
alkylation of benzene with 1-dodecene at a mol ratio of 10:1
benzene/1-dodecene. Table 1 lists the efficacy of the various
catalysts in order of their activity with the corresponding
selectivity to 2- and 3-phenyldodecanes. The K-10 clay-based
catalysts were more effective, which included heteropolyacids
(DTP, DMP, and DTS) supported on K-10, Filtrol-24 clay,
Al-pillared K-10 clay, and K-10 clay itself. Twenty percent
(27) Takada, H.; Ishiwatari, R. EnViron. Sci. Technol. 1990, 24, 86.
(28) Yadav, G. D.; Kirthivasan, N. J. Chem. Commun. 1995, 203.
(29) Kumbhar, P. S.; Yadav, G. D. Chem. Eng. Sci. 1989, 44, 2535.
(30) Thorat, T. S.; Yadav, V. M.; Yadav, G. D. Appl. Catal., A 1992, 90, 73.
(31) Yadav, G. D.; Mehta, P. H. Ind. Eng. Chem. Res. 1994, 33, 2198.
(32) Yadav, G. D.; Thorat, T. S. Ind. Eng. Chem. Res. 1996,35, 721.
(33) Yadav, G. D.; Bokade, V. V. Appl. Catal., A 1996, 147, 299.
(34) Yadav, G. D.; Kirthivasan, N. Appl. Catal., A 1997, 154, 29.
(35) Yadav, G. D.; Satoskar, D. V. J. Chem. Technol. Biotechnol. 1997, 69,
438.
(36) Yadav, G. D.; Krishnan, M. S. Org. Process Res. DeV. 1998, 2, 95.
(37) Yadav, G. D.; Krishnan, M. S. Ind. Eng. Chem. Res. 1998, 37, 3358.
(38) Yadav G. D.; Nair, J. J. Chem. Commun., 1998, 2370.
(39) Yadav, G. D.; Kirthivasan, N. In Fundamental and Applied Aspects of
Chemically Modified Surfaces; Blitz, J. P., Little, C. B., Eds.; Royal Society
of Chemistry: London, 1999; pp 254-269.
264
•
Vol. 6, No. 3, 2002 / Organic Process Research & Development

Table 1. Effect of different catalysts on alkylation of
benzene with 1-dodecenea
selectivity of
selectivity of
conversion 2-phenyldodecane 3-phenyldodecane
run
catalyst
of C12
%
(%)
(%)
1
2
3
4
5
6
7
8
20% w/w DTP/K-10
20% w/w DMP/K-10
20% w/w DTS/K-10
Filtrol-24
Al pillared clay
K-10
20% w/w DTP/silica
10% w/wAlCl3/
10% w/w FeCl3/K-10
97
92
91
85
80
71
57
32
32
28
29
22
31
27
35
45
62
52
50
47
47
30
42
20
9
Al-exchanged K-10
22
12
12
10
∼0
33
30
30
30
0
42
41
32
35
0
10 Cr-exchanged K-10
11 sulphated zirconia
12 Zr-exchanged K-10
13 20% w/w/DTP/carbon
a Conditions: 1-dodecene:benzene, 1:10 mol, temperature 150 °C, catalyst
loading: 0.05 g/cm3, speed: 1000 rpm. Autoclaved. DTP: dodecatungstophos-
phoric acid. DTS: dodecatungstosilicic acid. DMP: dodecamolybdophosphoric
acid.
Figure 1. Alkylation of benzene with 1-dodecene over 20%
DTP/K-10. Distribution of linear phenyldodecanes in the reac-
tion mixture. Reaction conditions: benzene:1-dodecene ) 10:
1, 150 °C, 1000 rpm, autoclaved. 20% DTP/K-10 catalyst
loading: 0.05 g/mL.
Scheme 1. Possible series and parallel reactions during
alkylation of benzene with 1- dodecene
Alul8 for HF-catalysed homogeneous reaction after thermo-
dynamic equilibrium had been reached. Two reasons were
advocated for different product distributions for homoge-
neous catalysis, by both HF and AlCl3: (i) existence of two
equilibrium steps due to isomerisation, one for the olefin
and another for the phenyldodecanes and (ii) the difference
in the rates of isomerisation of the olefin and the phenyl-
dodecanes. HF catalyses the isomerisation of 1-dodecene into
an equilibrium mixture of olefins but does not equilibrate
the phenyldodecanes even at 55 °C; on the other hand AlCl3
does so. However, this ratio could be altered by using a
suitable catalyst;16-20 and shape selectivity22 and the nature
of acidity 23,24 have a profound impact on product distribution.
Thus, the secondary carbon atom of the alkyl chain is
first attached to benzene, and further products can be formed
due to the proton shifts caused by the catalyst (Scheme 1).
After monoalkylation, dialkylated product can be formed
under a certain set of conditions such as high concentration
of olefin in comparison with benzene concentration and
reaction temperature. The literature suggested that the amount
of dialkylate could be reduced by using a molar excess of
benzene over 1-dodecene (above 5:1) and by controlling the
temperature and the type of acidity of the catalyst and its
loading.
In the current case, at a mol ratio of 10:1 benzene/1-
dodecene, over 20% DTP/K-10, the most active catalyst, only
four isomers of linear phenyldodecanes were formed with
97% conversion of 1-dodecene with a selectivity of 32% to
2-phenyldodecane and 62% to 3-phenyldocane with minor
amounts of 4- and 5-isomers (Figure 1). 6-Phenyldodecane
was not detected, and it can be explained as follows.
6-Phenyldocane is bulky molecule22-24 and tends to block
pores of microporous catalysts; with solid acids its formation
can be suppressed. With mordenite H-M,22 Mg-H-M24 and
Fe-H-M,24 no 6-isomer was formed, and less than 2% of both
4- and 5-phenyldodecanes were formed even at 100%
DTP/K-10 was found to be the best with 97% conversion of
1-dodecene, all leading to only the monoalkylated dodecyl-
benzenes. It is relevant to throw light on product distribution
and reaction mechanism. The alkylation of benzene with
olefins proceeds through the carbenium ion mechanism
(Scheme 1). In the case of 1-dodecene there is a possibility
of formation of six carbenium ions of the olefin whose
relative stability increases as the C-number increases, the
least stable being the 1-phenyldodecane isomer. In fact, due
to its being highly unstable, 1-phenyldodecane has not been
detected by any researcher for both homogeneous and
heterogeneous catalysts, and 2-phenyldodecane is the first
stable isomer. Thus, for homogeneous catalysts the thermo-
dynamic isomer distribution would be expected to contain
increasing amounts of 2- through 6- isomers, the 2-isomer
being the smallest in quantity. This was indeed found by
Vol. 6, No. 3, 2002 / Organic Process Research & Development
•
265

conversion, whereas H-ZSM-5 was ineffective.24 Thus, K-10
clay pore structure can be suitably adjusted to get favourable
product distribution.
In the current studies, 10 of the 13 catalysts were
montmorillonite K-10 clay-based, and another was Filtrol-
24 clay. After supporting with the heteropolyacids (HPA),
the pore size of K-10 is significantly reduced due to
deposition of the Keggin-structured HPA on the accessible
pore walls, and they occupy many small-sized pore junctions,
thereby reducing the accessibility of some networks.39 HPAs
are more acidic than acid-treated clay. Consequently, the
distribution of acidity is also changed significantly among
Brønsted and Lewis acid centres, both of which are required
for alkylation. Because of this structural change, 20% DTP/
K-10 was found to be the most effective catalyst than K-10
in a number of our studies.28,30,39 The surface areas of the
catalysts by nitrogen BET method for K-10 clay and DTP/
K-10 were measured as 230 and 107 m2/g, respectively. Since
both the catalysts were pretreated in similar way prior to
analysis, the reduction in surface area of DTP/K-10 is due
to the blockage of the smaller pores by the active species.
Indeed, the pore structure of DTP/K-10 is suitable for
formation of smaller isomers (2- and 3-phenyldodecanes),
which could get diffused in and out of the pore networks
better than higher-numbered ones such as 6-isomers. Thus,
there is a shape selectivity, and a greater amount of 3-isomer
was formed than of 2-isomer with time. There is a transition-
state shape selectivity, and the amounts of 4- and 5- isomers
are therefore very low. The bulkiest isomer is therefore was
not formed. With K-10 clay as catalyst per se, 71%
conversion of 1-dodecene was obtained, all leading to
dodecylbenzene isomers with 27% selectivity of 2-phenyl-
dodecane. Other heteropolyacids such as DTS and DMP
supported on K-10 also showed better activity than K-10
alone. Exchanged clays such as Al exchanged K-10, Zr
exchanged K-10 and Cr exchanged K-10 showed much lower
activity. There is a clear synergism between the clay and
heteropolyacids which was reported earlier for etherifica-
tion.28 Another interesting catalyst is FeCl3-AlCl3/K-10. It
gave 45% selectivity to 2-phenyldodecane, but the conversion
was only 32% in 4 h.
How both type of acidity and pore network accessibility
affect the product distribution can be explained with DTP/
carbon and sulphated zirconia as catalysts. Carbon-based
DTP did not give any measurable concentration of any
isomers, whereas sulphated zirconia was less effective than
20% DTP/K-10. Sulphated zirconia is a superacid vis-a`-vis
DTP/K-10 and has a much wider pore-size distribution and
thus can lead to higher rates initially, leading to the formation
of bulkier molecules such as didodecylbenzenes. However,
it was not more active due to the type and distribution of
acidity, and indeed there could be some formation of
didocylbenzene which could block the accessibility, thereby
reducing the acidity. The BET surface area of sulphated
zirconia had dropped from 96 to 70 m2/g at the end of the
reaction. When DTP was supported on carbon, there was
no conversion of 1-dodecene, perhaps due to strong intra-
particle diffusion resistance offered by the bulky molecules
Figure 2. Effect of calcination temperature on benzene
alkylation with 1-dodecene. C12:C6H6, 1:10 mol; catalyst DTP/
K-10, 0.05 g/mL; temperature, 150 °C; speed, 1000 rpm;
reaction time, 4 h.
and pore blocking. Pores of carbon are narrow slits, and
accessibility can be affected drastically when bulky molecules
are used.
Thus, in all further experiments 20% DTP/K-10 was used
as the catalyst. The possibilities of formation of some heavier
alkylate as byproducts under different conditions particularly
when benzene to 1-dodecene ratio was 2:1 are given in
Scheme 1 and described below.
1. Dialkylation: It occurs by alkylation of linear dode-
cylbenzene with a second 1-dodecene molecule (reaction 2).
R1, R2, R3 and R4 are linear alkyl chains. The above depiction
of DDB includes different isomers.
2. Oligomerisation of 1-dodecene followed by benzene
alkylation (reactions 3 and 4).
It was found that the main products were linear alkyl-
benzenes whenever the mol ratio of benzene to 1-dodecene
was 10:1, and it was used in all experiments in the current
investigation. Oligomerization of 1-dodecene itself and the
benzene alkylation due to oligomerised 1-dodecene were not
at all observed.
Effect of Calcination Temperature. The activity of 20%
DTP/K-10 can be a function of temperature because of some
water of hydration bound to the catalyst. Hence, the catalyst
was calcined at four different temperatures, namely, 200, 250,
285, and 400 °C for 3 h. As shown in Figure 2, the
conversion of 1-dodecene remained unchanged up to 285
°C. Conversion to monoalkylated products was 97% with
34% selectivity for 2-phenyldodecane and 62% for 3-phenyl-
dodecane. The conversion to monoalkylated products was
reduced to 69% when the catalyst was calcined at 400 °C.
It was due to the loss of interlayer water in the clay at high
temperature, which had resulted in collapse of interlayer of
clay and also due to sintering and decomposition of DTP
above 300 °C. Under such circumstances loss of both the
effective area of the support and the active sites would occur,
resulting in net loss of activity. Independent experiments on
the pore-size distribution measurements and XRD as a
function of temperature had showed that the pore size gets
drastically reduced and that the catalyst activity drops
substantially.39
266
•
Vol. 6, No. 3, 2002 / Organic Process Research & Development

Since 20% DTP/K-10 calcined at 285 °C was the best
catalyst, the effect of various parameters on the rate of
reaction, and hence conversion, was studied under otherwise
similar conditions. Characterization of the catalyst is reported
earlier by us,33,34,39 and a brief summary is provided here.
As stated earlier, the surface areas of the catalysts by nitrogen
BET method for K-10 clay and DTP/K-10 were measured
as 230 and 107 m2/g, respectively. Since both the catalysts
were pretreated in similar way prior to analysis, the reduction
in surface area of DTP/K-10 is due to the blockage of the
smaller pores by the active species. It appears that the active
species are held in a few junctions of such dimensions from
which the access to smaller pores is denied, thereby leading
to the decrease in accessible surface area. The average
particle size of both K-10 and DTP/K-10, as determined by
image analysis, was in the range of 600 µm. The XRD
studies of DTP/K-10 had indicated that in the impregnation
process, the clay had lost some of its crystallinity compared
to K-10. The cation-exchange capacity (CEC) of the catalysts
was determined by the method to observe that DTP/K-10
possesses higher CEC than K-10. This may be due to some
additional surface protons, which come by way of het-
eropolyacid impregnation, which may play a role on account
of easy availability for the exchange reaction. We have earlier
reported39 that on the basis of the actual quantity of DTP
loaded on the support in comparison with the same unloaded
catalyst or the support K-10 alone, the turnover frequency
is greater in the case of the supported catalyst. Thus, there
is a synergism between DTP and K-10, leading to higher
activity. In the case of DTP/K-10 there are lot of dispersed
particles on the surface of the support K-10 as revealed by
SEM photomicrographs. FTIR studies of K-10 showed that
there were Si-O and Si-O-Al linkages and that the OH
groups are bonded to the Al atoms, whereas in DTP/K-10
the presence of H3O+ (Brønsted acidity) and the linkage of
phosphorus was indicated.
Effect of Speed of Agitation. The influence of external
mass-transfer resistance was studied for the transfer of the
limiting reactant 1-dodecene to the external surface of the
catalyst particle at different speeds of agitation at 150 °C
for a catalyst loading of 0.05 g/cm3. The effect of speed of
agitation was studied at 250, 700, 800,900, 1000, and 1200
rpm and it was found that the conversion of 1-dodecene had
increased from 250 to 700 rpm only marginally. From 700
to 1200 rpm the conversion and product distribution was
practically similar in all the cases. Thus, the external
resistance to mass transfer was absent beyond 700 rpm.
However, to be on safer side and to avoid attrition of the
catalyst, further experiments were conducted at 1000 rpm.
Assessment of External and Internal Resistance to
Mass Transfer. This is a typical solid-liquid slurry reaction
involving the transfer of 1-dodecene (D) and benzene (B)
from the bulk liquid phase to the catalyst wherein external
mass transfer of reactants to the surface of the catalyst
particle, followed by intraparticle diffusion, adsorption,
surface reactions and desorption take place. The influence
of external solid-liquid mass transfer resistance must be
ascertained before a true kinetic model could be developed.
Since benzene was always used in excess, there is likelihood
of mass transfer resistance to the transfer of 1-dodecene from
the bulk liquid phase to the external surface of catalyst
particle.
At steady state, the rate of mass transfer (rD) per unit
volume of the liquid phase (mol‚cm-3‚s-1) is given by:
rD ) kSL-Dap{[D0] - [Ds]}
(5)
(rate of transfer of D from bulk liquid to
external surface of the catalyst particle)
) kSL-Bap{[B0] - [Bs]}
(6)
(rate of transfer of B from the bulk liquid phase to
the external surface of the catalyst particle)
) robs
(7)
) rate of reaction within the
catalyst particle
where, kSL is the solid-liquid mass-transfer coefficient of
the concerned species (cm/s), ap is the external surface area
of the particle (cm2/cm3 liquid-phase volume), subscript o
and s denote the bulk liquid phase and external surface
concentrations respectively, in mol/cm3.
Depending on the relative magnitudes of external resis-
tance to mass transfer and reaction rates, different controlling
mechanisms have been put forward.29 When the external
mass-transfer resistance is small, then the following inequal-
ity holds,
1
1
1/robs
.
and
(8)
kSL-Dap[D0]
kSL-Bap[B0]
The observed rate robs could be given by three types of
models wherein the contribution of intraparticle diffusion
resistance could be accounted for by incorporating the
effectiveness factor η. These models are (a) the power law
model if there is very weak adsorption of reactant species,
(b) Langmuir-Hinshelwood-Hougen-Watson model, (c)
Eley-Rideal model.
It is therefore necessary to study the effects of speed of
agitation, catalyst loading, and particle size to ascertain the
absence of external and intraparticle resistance so that a true
intrinsic kinetic equation could be used.
According to eq 8, it is necessary to calculate the rates
of external mass transfer of both 1-dodecene (D) and benzene
(B) and compare them with the rate of reaction.
For a typical spherical particle, the particle surface area
per unit liquid volume is given by
6w
(FPdP)
aP )
(9)
where w ) catalyst loading g/cm3 of liquid phase, Fp )
density of particle g/cm3, and dp ) particle diameter, cm.
For the maximum catalyst loading used (0.08 g/cm3) for
a particle size (dp) of 0.06 cm, in the current studies, ap )
9.41 cm2/cm3 liquid phase. The liquid-phase diffusivity
Vol. 6, No. 3, 2002 / Organic Process Research & Development
•
267

values of the reactants D and B, denoted by DDB and DBD
,
were calculated by using the Wilke-Chang equation40 at 150
°C as 5.2 × 10-5 and 8.1 × 10-5 cm2/s, respectively. The
solid-liquid mass transfer coefficients for both A and B were
calculated from the limiting value of the Sherwood number
(e.g., ShD ) kSL-Ddp/DDB) of 2. The actual Sherwood numbers
are typically higher by order of magnitude in well-agitated
systems but for conservative estimations a value of 2 is taken.
The solid-liquid mass-transfer coefficients kSL-A and kSL-B
values were obtained as 1.73 × 10-3 and 2.70 × 10-3 cm/s,
respectively. The initial rate of reaction was calculated from
the conversion profiles. A typical calculation shows that for
a typical initial rate of reaction was calculated as 1.64 ×
10-7 mol‚cm-3‚s-1. Therefore, putting the appropriate values
in eq 8, the respective values are:
Figure 3. Effect of catalyst loading on benzene alkylation with
1-dodecene. C12:C6H6, 1:10 mol; catalyst DTP/K-10, 0.05 g/mL,
temperature, 150 °C; speed, 1000 rpm; reaction time, 4 h.
µm, practically the same conversion and selectivity was
obtained. This also supported the theoretical calculations. A
further proof of the absence of the intraparticle diffusion
resistance was obtained through the study of the effect of
temperature, and it will be discussed later.
6.097 × 106 . 7.49 × 104 and 4.80 × 103
The above inequality demonstrates that there is an absence
of resistance due to the solid-liquid external mass transfer
for both the species D and B and the rate may be either
surface-reaction-controlled or intraparticle-diffusion-con-
trolled. Therefore, the effects of catalyst loading at a fixed
particle size and temperature were studied to ascertain the
influence of intraparticle resistance.
The average particle diameter of the catalyst used in the
reactions was 0.06 cm, and thus a theoretical calculation was
done on the basis of the Wiesz-Prater criterion to assess
the influence of intraparticle diffusion resistance.41
According to the Wiesz-Prater criterion, the dimension-
less parameter Cwp which represents the ratio of the intrinsic
reaction rate to intraparticle diffusion rate, can be evaluated
from the observed rate of reaction, the particle radius (Rp),
effective diffusivity of the limiting reactant (De) and con-
centration of the reactant at the external surface of the
particle.
Effect of Catalyst Loading. Figure 3 shows the effect
of catalyst loading on the alkylation of benzene with
1-dodecene. The catalyst loading was varied over 8-fold from
0.01 to 0.08 g/cm3. It was found that the yield of dodecyl-
benzenes increased with increase in the catalyst loading. Rate
of alkylation was very slow with 0.01 g/cm3 catalyst loading,
and only 35% conversion of 1-dodecene was obtained. When
catalyst loading was increased from 0.05 to 0.08 g/cm3, the
rate of reaction was almost the same, and 97% yields of
dodecylbenzene were obtained in both cases. This could be
because of the fact that beyond a certain loading, there
existed an excess of catalyst sites greater than actually
required by the reactant molecules. Hence, there was leveling
off of the reaction rate. As shown by eqs 5 and 6, at steady
state, the rate of external mass transfer (i.e., from the bulk
liquid phase in which D and B are located with concentration
[D0] and [B0], respectively) to the exterior surface of the
catalyst is proportional to aP, the exterior surface area of the
catalyst where the concentrations of D and B are [DS] and
[BS], respectively. For a spherical particle, aP is also
proportional to w, the catalyst loading per unit liquid volume
as shown by eq 9. It is possible to calculate the values of
[DS] and [BS]. For instance,
2
(i) If Cwp ) robsFpRP /De [Ds] . 1
(10a)
then the reaction is limited by severe
internal diffusion resistance,
(ii) If Cwp , 1
(10b)
then the reaction is intrinsically kinetically controlled
kSL-Aap {[Do] - [DS]} ) robs at steady state
) 1.64 × 10-7 mol‚cm-3‚s-1
The effective diffusivity of 1-dodecene (De-D) inside the
pores of the catalyst was obtained from the bulk diffusivity
(DDB), porosity (ꢀ) and tortuosity (τ) as 6.5 × 10-6 cm2/s
where De-A ) DABꢀ/τ. In the present case, the value of Cwp
was calculated as 0.0235 for the initial observed rate, and
therefore the reaction is intrinsically kinetically controlled.
However, in a separate experiment, the particles of DTP/K-
10 were crushed and sieved and used for the reaction. At
the same loading of catalyst, for average particle size of 125
Thus, putting the appropriate values, it is seen that [DS]
≈ [D0], similarly [BS] ≈ [B0]. Thus, any further addition of
catalyst is not going to be of any consequence for external
mass transfer.
Effect of Concentration of Reactants. The concentra-
tions of benzene and 1-dodecene were the important vari-
ables. An excess of benzene favoured only the formation of
monoalkylated derivatives (Table 2), whereas when the mol
ratio of benzene/1-dodecene was decreased from 1:10 to 1:5
as shown in Figure 4, the conversion of 1-dodecene was
reduced from 97 to 92%. However, this was marginal, and
(40) Reid, R. C.; Prausnitz, M. J.; Sherwood, T. K. The Properties of Gases
and Liquids, 3rd ed.; McGraw-Hill: New York, 1977.
(41) Fogler, H. S. Elements of Chemical Reaction Engineering; Prentice-Hall:
New Delhi, 1995.
268
•
Vol. 6, No. 3, 2002 / Organic Process Research & Development

Table 2. Effect of mol ratio on alkylation of benzene with
1-dodecenea
% product distribution
C12:C6H6 mol % conversion of C12
D.B.
D.D.B.
1:10
1:5
1:2
2:1
5:1
97
92
78
67
55
100
100
86
49
40
-
-
14
51
60
a Conditions: catalyst 20% w/w DTP/K-10 0.05 g/cm3, temperature 150 °C,
speed 1000 rpm, reaction time 4 h. Autoclaved.
Figure 6. Effect of mol ratio on yields of mono- and dialkylated
products in benzene alkylation with 1-dodecene. C12:C6H6, 2:1
mol; catalyst DTP/K-10, 0.05 g/mL, temperature, 150 °C; speed,
1000 rpm; reaction time, 4 h.
Figure 4. Effect of mol ratio on benzene alkylation with
1-dodecene. Catalyst DTP/K-10, 0.05 g/mL, temperature, 150
°C; speed, 1000 rpm; reaction time, 4 h.
Figure 7. Effect of mol ratio on yields of mono- and dialkylated
products in benzene alkylation with 1-dodecene. C12:C6H6, 5:1
mol; catalyst DTP/K-10, 0.05 g/mL, temperature, 150 °C; speed,
1000 rpm; reaction time, 4 h.
reactant when 1-dodecene-to-benzene mol ratio was in-
creased to 2:1. Conversion to the alkylated product was 67%
and selectivities of monoalkylated and dialkylated products
were 49 and 51%, respectively. When the 1-dodecene-to-
benzene mol ratio was changed to 5:1 (Figure 7) the
selectivity of didodecylbenzene had increased to 60% with
55% conversion of 1-dodecene. Thus, in the presence of
excess of 1-dodecene, the rate formation of didodecylbenzene
was higher than the formation of dodecylbenzene.
Reaction Mechanism and Modeling. In the absence of
any external or internal mass transfer limitations, the reaction
would be controlled by chemisorption, surface reaction or
desorption of products all on the internal pore surface of the
catalyst. The initial experiments showed that amongst all
possible isomers, only 2-, 3-, 4-, and 5-phenyldodecanes were
formed. The formation of these isomers was likely either (i)
by the isomerisation of 1-dodecene to 2-, 3-, 4-, and
5-dodecenes when the carbenium ion is formed upon
chemisorption of 1-dodecene on the catalyst surface or (ii)
by the isomerisation of 1-phenyldodecane which is formed
first. However, the concentration profiles of the various
phenyldodecanes showed that the possibility (ii) did not arise.
Figure 5. Effect of mol ratio on yields of mono- and dialkylated
products in benzene alkylation with 1-dodecene. C12:C6H6, 1:2
mol; catalyst DTP/K-10, 0.05 g/mL, temperature, 150 °C; speed,
1000 rpm; reaction time, 4 h.
the formation of didodecylbenzene was not still observed.
Didodecylbenzene was formed only when the 1-dodecene-
to-benzene mol ratio was further reduced to 1:2. The rate of
formation of didodecylbenzene was very slow compared to
the formation of dodecylbenzene (Figure 5). With a decrease
in the mol ratio of 1-dodecene to benzene, the formation of
didodecylbenzene had increased, particularly below a mol
ratio of 1:5. Figure 6 shows the conversions of the limiting
Vol. 6, No. 3, 2002 / Organic Process Research & Development
•
269

Scheme 2. Mechanism of formation of linear
phenyldodecanes
-dC
dt
-rB )
B ) (k5C2-D+ + k C + + k C + + k C +)C
6 3-D 7 4-D 8 5-D B
) (k5k1C1-DCH+ + k k k C1-DCH+ + k k k k C1-DDH
+
+
6
3
1
7 3 2 1
k8k4k3k2k1C1-DCH+)CB
) (k′5 + k′6 + k′7 + k′8)C1-DCBCH
(20)
+
where, with the corresponding k as forward rate constants,
the following constants are defined.
k′5 ) k5K1
(21)
(22)
(23)
(24)
(25)
k′6 ) k6K2K1
k′7 ) k7K3K2K1
k′8 ) k8K4K3K2K1
kp ) k′5 + k′6 + k′7 + k′8, overall pseudo-constant.
The net rate of formation of m-PD is given by:
dC
m-PD ) kpC1-DCBCH+ - k Cm-PDCn-D
(26)
+
15
dt
The formation of DBB is given by:
dC
DDB ) k15Cm-PDCn-D
(27)
+
dt
The rates of formations of individuals phenyldodecanes
can also found
dC
r2-PD
)
2-PD ) k′5C1-DCBCH
(28)
(29)
(30)
(31)
+
dt
dC
r3-PD
r4-PD
r5-PD
)
)
)
3-PD ) k′6C1-DCBCH
+
+
+
dt
dC
4-PD ) k′7C1-DCBCH
By denoting acidic sites on the catalyst surface as H+
irrespective of their distinction between Brønsted and Lewis
sites, the following reactions are thus depicted in Scheme 2.
The formation of didodecylbenzene can take place as per
the following:
dt
dC
5-PD ) k′8C1-DCBCH
dt
The concentration of the acid centres will be in terms of
g-catalyst per cm3 organic phase. Thus, k′5, k′6, k′7, k′8, and
K15
kp will have units of cm6‚mol-1‚s-1‚gcat-1
.
m-PD + n-D+ y z DDB + H+
(19)
Validation of Model with Experimental Data and
Establishment of Kinetics. When the mol ratio of benzene
to 1-dodecene was greater than 5, it could be safely assumed
to be a pseudo-first-order reaction for a fixed catalyst
concentration. Thus, eq 20 can be written as,
where m ) 2, 3, 4, 5 and n ) 2, 3, 4, 5, Ks are equilibrium
constants.
As shown in Figure 1, when benzene was taken in excess
with 20% DTP/K-10 as the catalyst, only the phenyldode-
canes are formed without any didodecylbenzene or oligomers
1-dodecene. However, when the mol ratio of 1-dodecene to
benzene was 1:2, there was substantial formation of didode-
cylbenzene (Table 2).
-dC
dt
B ) kpC1-DCB w
0
-dC
)
1-D ) (kpCB w)C1-D (32)
0
dt
It is thus clear that the carbenium ion formation as shown
by eqs 11-14 are equilibrium reactions. Analysis of products
had shown that in the cases of reactions 15-18, only the
forward rates were significant initially and reaction 19 did
occur.
which in terms of fractional conversion and integration leads
to
-ln(1 - X) ) kpCB0wt ) k0wt ) k1t
(33)
where X ) fractional conversion of 1-dodecane. A plot of
the left-hand side terms of eq 33 against t will be a straight
The net rate of forward reactions of benzene is:
270
•
Vol. 6, No. 3, 2002 / Organic Process Research & Development

Figure 8. Plot of -ln(1 - X) vs time at different catalyst
loading.
Figure 10. Effect of temperature on benzene alkylation with
with 1-dodecene. C12:C6H6, 1:10 mol; catalyst DTP/K-10, 0.05
g/mL, temperature, 150 °C; speed, 1000 rpm; reaction time, 4
h.
Figure 9. Plot of k1 vs catalyst loading.
line passing through origin with a slope of k1 ) k0w )
kpCB w, from which kp could be established. Furthermore,
0
k′5, k′6, k′7, and k′8 can be found from initial rate data.
A plot of -ln(1 - X) versus t is made for different catalyst
loadings in Figure 8 which are all straight lines passing
through origin, the slopes of which are plotted against catalyst
loading (w) in Figure 9, which shows that the rates are
linearly proportional to catalyst loading.
Figure 11. Plot of -ln(1 - X) vs time at different tempera-
tures.
From Figure 1, the initial rates of formation of 2-, 3-, 4-,
and 5- phenyldodecanes were obtained and used to calculate
the corresponding individual rate constants. The rate con-
stants k′5, k′6, k′7, and k′8 for the formation of different
isomers were found to be, 0.00372, 0.0109, 0.00014, and
0.00006 cm‚6gmol-1‚gcat-1‚s-1, respectively at 150 °C.
Again from the slope of Figure 9 the overall reaction rate
constant kp at 150 °C is found to be 0.015 cm‚6gmol-1‚gcat-1‚
s-1, which is very close to the sum of the individual reaction
rate constants confirming the validity of the model.
Effect of Temperature. At low temperature (120 °C),
the conversion of 1-dodecene was 43%. It was found that
no dialkylated product was found even when the temperature
was increased from 120 to 150 °C (Figure 10). This was
because of the presence of excess benzene in the reaction
mixture. Plots of -ln(1 - X) versus t at different tempera-
tures are shown in Figure 11 for 1:10 mol/mol of 1-dodecene
to benzene. The slopes of the straight lines represent rate
constant kp at different temperatures. The Arrhenius plot for
ln(kp) versus 1/T (K-1) is shown in Figure 12 from which
the apparent activation energy of the reaction is found to be
20.06 kcal/mol.
Figure 12. Arrhenius plot.
Reusability of Catalyst. Reusability of 20% DTP/K-10
and K-10 was examined for three experiments. After each
run, both the catalysts were thoroughly washed with hexane
and dried at 120 °C for 2 h prior to reuse. In the presence of
Vol. 6, No. 3, 2002 / Organic Process Research & Development
•
271

Figure 13. Reusability of DTP/K-10 on benzene alkylation with
with 1-dodecene. Olefin:benzene, 1:10 mol; temperature, 150
°C; catalyst load, 0.05 g/mL; rpm, 1000; reaction time 2 h.
Figure 14. Alkylation of benzene with different olefins. Olefin:
benzene, 1:10 mol; temperature, 150 °C; catalyst load, 0.05
g/mL; rpm, 1000; reaction time 2 h.
Table 3. Reusability of 20% DTP/K-10 on alkylation of
benzene with 1-dodecenea
heteropolyacids such as dodecatungstophosphoric acid (DTP),
dodecatungstosilicic acid (DTS), and dodecamolybdophos-
phoric acid (DMP) supported on K-10 clay, Filtrol-24, Al-
pillared clay, 20% DTP/silica,10% AlCl3/10% FeCl3/K-10,
DTP, Cr-exchanged K-10, Sulphated-zirconia, Zr-exchanged
K-10, and 20% DTP/activated carbon. Of all other catalysts
used 20% w/w DTP/K-10 clay led to best activity and
selectivity to 2- and 3-phenyldodecanes. A mol ratio for
benzene/1-dodecene of 10:1, favoured the formation of linear
phenyldodecanes. However, with decreasing benzene con-
centration, the formation of didodecylbenzene had increased.
A mechanistic model was built and validated against
experimental data.
run
% conversion of C12
% 2 - P.D.
% 3 - P.D.
1
2
3
97
92
86
62
62
60
34
32
32
a Conditions: catalyst loading 0.05 g/cm3, temperature 150 °C, speed 1000
rpm, reaction time 4 h. Autoclaved.
DTP/K-10, the conversion of 1-dodecene was reduced to an
extent of 11% from the first run to the third run (Figure 13),
whereas K-10 clay deactivated to an extent of 23% after three
runs without much change in the selectivity of 2-phenyl-
dodecane as shown in Table 3.
Alkylation of Benzene with Other Olefins. To inves-
tigate the effect of chain length of R-olefins, the benzene
alkylation was also carried out with 1-octene, 1-decene, and
1-tetradecene under otherwise similar conditions. It was
found that the rate of benzene alkylation decreased with an
increase in the chain length of R-olefin (Figure 14). With
1-octene and 1-decene the rate of reaction was very fast
compared to that of 1-tetradecene; 98% conversions of both
1-octene and 1-decene were obtained in 50 and 120 min,
respectively. With 1-tetradecene, being a bulky molecule,
the rate of reaction was very slow, and conversion to
monoalkylated product (tetradecylbenzene) was only 49%.
Acknowledgment
Research support from Department of Science and
Technology (DST), and All Indian Council for Technical
Education (AICTE) and Darbari Seth Professorship Endow-
ment are gratefully acknowledged. G.D.Y. acknowledges the
hospitality of the Michigan State University as the Johansen
Crosby Visiting Chair Professor of Chemical Engineering
during 2001-2002.
Conclusions
Received for review April 10, 2000.
OP000044S
The alkylation of benzene with 1-dodecene was examined
by using 13 different catalysts such as K-10 clay, 20% w/w
272
•
Vol. 6, No. 3, 2002 / Organic Process Research & Development