Kinetics and mechanism of acid catalyzed alkylation of phenol with cyclohexene in the presence of styrene divinylbenzene sulfonic resins

Article abstract of DOI:10.1016/j.molcata.2011.11.025

The kinetics of phenol cyclohexylation catalyzed by sulfonic resins is studied taking into account equilibria and side reactions observed in the reaction media. The influence of different sulfonic resins has been tested and only small variation of the catalyst activity, referred to the acid amount, has been observed. Besides, the selectivity is unaffected by varying the type of the catalyst. The effect of reagent concentration on the reaction rate has been also studied, together with the reactivity of the cyclohexyl phenyl ether as intermediate. The etherification equilibrium of cyclohexyl phenyl ether has been analyzed in the presence of methanesulfonic acid, and the adsorption equilibria on a sulfonic resin of reagents and intermediates are measured, too. Starting from the evidences obtained from these studies an Eley-Rideal type kinetic model has been proposed and the fitting of the experimental data allows obtaining the kinetic constant of each stage. Good reliability of the model with the experimental data has been observed, also at high conversion, and the values of the fitting parameters are substantially constant by varying the operative variable, which is a further proof of the goodness of the model.

Full text of DOI:10.1016/j.molcata.2011.11.025

Journal of Molecular Catalysis A: Chemical 353–354 (2012) 192–203
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Kinetics and mechanism of acid catalyzed alkylation of phenol with cyclohexene
in the presence of styrene divinylbenzene sulfonic resins
∗
L. Ronchin , G. Quartarone, A. Vavasori
Department of Molecular Science and Nanosystems, University Ca’ Foscari of Venice, Dorsoduro 2137, 30123 Venice, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 11 September 2011
Received in revised form
The kinetics of phenol cyclohexylation catalyzed by sulfonic resins is studied taking into account equilibria
and side reactions observed in the reaction media. The influence of different sulfonic resins has been
tested and only small variation of the catalyst activity, referred to the acid amount, has been observed.
Besides, the selectivity is unaffected by varying the type of the catalyst. The effect of reagent concentration
on the reaction rate has been also studied, together with the reactivity of the cyclohexyl phenyl ether as
intermediate. The etherification equilibrium of cyclohexyl phenyl ether has been analyzed in the presence
of methanesulfonic acid, and the adsorption equilibria on a sulfonic resin of reagents and intermediates
are measured, too. Starting from the evidences obtained from these studies an Eley–Rideal type kinetic
model has been proposed and the fitting of the experimental data allows obtaining the kinetic constant
of each stage. Good reliability of the model with the experimental data has been observed, also at high
conversion, and the values of the fitting parameters are substantially constant by varying the operative
variable, which is a further proof of the goodness of the model.
2
4 November 2011
Accepted 25 November 2011
Available online 6 December 2011
Keywords:
Acid catalysis
Sulfonated resins
Phenol alkylation
Alkylation mechanism
Alkylation kinetics
©
2011 Published by Elsevier B.V.
1
. Introduction
a long time [1,3,13–15]. Some aspects, connected to kinetics and
selectivity of phenol cyclohexylation, however, are not completely
elucidated. In particular, the studies of Sharma and coworkers car-
ried out in the early nineties, relative to the reactivity of several
olefins toward phenol in the presence of sulfonated resins, pointed
out that the ortho–para selectivity in the ring alkylation of phenol
is strictly related to the nature of the olefin. In particular, propene
and 1-butene give an ortho–para ratio close to 2, while isobutene,
␣-methyl styrene and diisobutene give almost exclusively para
alkylation [16]. More recently, Hölderich and coworkers showed
that high para selectivity is obtained in the alkylation of phenol
with isobutene and the nature of the acid catalysts does not influ-
ence the selectivity. On the contrary, catalyst activity is influenced
by the amount and the strength of the acid sites [17]. In these stud-
ies there are no evidences of cyclohexylphenyl ether formation; in
contrast, by using cyclohexene as alkylating agent, the formation
of the ether occurs with high yield by using solid acid catalysts
with different nature [18–20]. For instance, Yadav and Kumar have
recently studied the kinetics of phenol cyclohexylation catalyzed by
many solid acids in solvent-less conditions at 333 K [18]. The time
concentrations profile shows the formation of cyclohexyl phenyl
ether, as the most abundant product (70%) after 4 h of reaction. Ring
alkylation occurs in lesser extent and the ortho–para ratio is in the
range 2–7 depending on the type of the catalyst used [18]. In that
work the best fit of the data are obtained by an Eley–Rideal kinetic
model, but several aspects of the kinetics were not studied. More
recently, Yadav and Pathre studied the cyclohexylation of guaiacol
Alkylphenols and alkyl phenyl ethers are molecules of consider-
able interest because of their industrial relevance [1–3]. As a matter
of fact, the world production of alkylphenols exceeds 460,000 t/yr.
The vast majority of alkylphenols is used to synthesize derivatives
which have applications ranging from surfactants to pharmaceu-
ticals. In additions, the use of alkylphenols in the production of
both polymer additives and monomers for engineering plastics is
expected to show a constant growth during the next years. Alkyl
phenyl ethers are compounds of great interest in the field of fine
chemicals and their production is in continuous expansion espe-
cially in the new markets [1–3].
The cation-exchange resins are used as catalysts in a large
number of industrial processes, such as the manufacture of alkyl
phenols, the esterification of carboxylic acids, the synthesis of
ethers, the hydration of alkenes, etc. [4–9]. The use of these hetero-
geneous systems is a suitable alternative to the usual procedures
in a homogeneous system in the presence of mineral acids and ion-
exchange resins appear to be ideal catalysts to convert polluting
processes into greener ones [4–12].
Phenol is an activated molecule to the electrophilic aromatic
substitutions and its mechanism of alkylation has been known for
∗ Corresponding author. Fax: +39 0412348517.
E-mail address: ronchin@unive.it (L. Ronchin).
1
381-1169/$ – see front matter © 2011 Published by Elsevier B.V.
doi:10.1016/j.molcata.2011.11.025

L. Ronchin et al. / Journal of Molecular Catalysis A: Chemical 353–354 (2012) 192–203
193
catalyzed by several solid catalysts. Also in this work, the formation
of the ether is observed but its concentration passes through a max-
imum and finally decreases to complete consumption [21,22]. Also
in this case the best fit of the data was obtained by an Eley–Rideal
kinetic model, but a consecutive rearrangement of the ether was
implemented in the model thus explaining its disappearance [21].
Quite surprisingly, there is no mention of the etherification equi-
librium, which is well known for aliphatic ether, and it is evident
also for the cyclohexylphenyl ether [23,24].
The mechanistic aspect of the electrophilic attack to the phenol
is investigated from a theoretical point of view with DFT studies by
Tang and coworkers. These authors suggested that an olefin reacts
with a sulfonic acid leading to the formation of a sulfonic ester inter-
mediate, which, in turns, reacts with phenol to form the products
of alkylation [25].
Brunauer–Emmet–Teller surface area, pore size distribution
(obtained by Barrett–Joyner–Halenda model) and total pore vol-
ume (measured at p/p = 0.98 of relative pressure) of the catalysts
0
have been determined by N2 adsorption and desorption at 94 K
using an automatic adsorption unit (Micromeritics ASAP 2010C)
[27].
The infrared spectroscopy measurement (IR) was carried out on
the solid resin and the sample was prepared as follows: a small
amount of resin (c.a. 10 mg) is grinded in an agate mortar with KBr
(c.a. 1 g). The resulting fine powder was pressed at 1000 Mpa for
5 min, and the disk was loaded on a Perkin Elmer spectrum 65 FTIR
−
1
at 2 cm of resolution.
Elemental analysis of the Amberlyst 15-NO2 catalyst has been
carried out in a PerkinElmer 2400 Series II CHN elemental Analyzer.
In this paper we study the kinetics of phenols cyclohexylation
catalyzed by some sulfonic resins in order to highlight some unclear
aspects of this reaction. In particular we try to develop a kinetic
model taking into account the different equilibria, since both the
cyclohexylphenyl ether and heterogeneous equilibria affect the
overall kinetics. In addition, the reactivity of the cyclohexyl phenyl
ether, recognized as a transient intermediate, is implemented in
the model [24].
2.4. Reactions
The reactions were performed in a well stirred baffled glass
reactor (generally at 12 Hz of stirring speed by using a Rushtone
turbine) thermostatted by circulation bath in the range 288–373 K,
containing weighed samples of solvent, reagents and catalyst at
autogenous solvent pressure (i.e. 122 kPa for 1,2-dichloroethane
at 358 K). In a typical experiment 10 mL of solution containing
1
0 mmol of phenol, 10 mmol of cyclohexene plus 5 mmol of methyl-
cyclohexane as internal standard and the desired amount of catalyst
100–500 mg) were placed into the reactor. All the operations were
2
. Experimental
(
carried out inside a glove box in order to minimize catalyst deacti-
vation by air moisture. Small amounts of the solution were drawn at
different times and the samples were analyzed by GC (Agilent 7890)
and GC–MS (Agilent 7890+), using a HP5 capillary column (300 ␮m
i.d. 30 m long, 95% methyl, 5% phenyl silicone phase). The samples
were checked also by HPLC using a Perkin Elmer 250 equipped with
a diode array LC-235 detector and a Lichrosphere 100 (RP-18, 5 ␮m)
column.
2
.1. Materials
Reagents and solvents were used after purification of the com-
mercially available samples and their purity was checked by
melting point, thin layer chromatography (TLC), high performance
liquid chromatography (HPLC), gas chromatography (GC) and gas
chromatography coupled to a mass spectroscopy (GC–MS). The sol-
vents were treated in a double bed column, filled with H SO /SiO
2
4
2
For a reliable comparison of the performances of different cata-
lysts it is essential to know if the reaction rate data are affected by
diffusion phenomena. This is verified by studying the influence of
the speed of agitation, the granulometry and the catalyst amount on
the reaction rate catalyzed by the most active catalyst at the highest
temperature of reactions (Amberlyst 36 as catalyst at 368 K). The
experimental evidences suggest that the kinetics is not influenced
by diffusion phenomena since there are no differences in the initial
reaction rate when the agitation speed vary from 10 to 16 Hz. In
addition, the initial reaction rates are strictly proportional to the
catalyst amount by using resins with different granulometry. Fur-
thermore, the inspection of Carberry and Wheeler–Weisz numbers
shows values of 0.08 and 0.35, respectively, thus suggesting a neg-
ligible influence of the diffusion phenomena on the rate of reaction
and SiO to adsorb water and impurities. The residual water content
2
was checked by HPLC analysis and its concentration is in the range
of 10–20 ppm [26]. Commercial catalysts: macro-reticular sul-
fonated styrene divinyl benzene resins Amberlyst 15 and Amberlyst
3
6, which are a trade mark of Rohm and Haas, were purchased
from Aldrich. Deuterated chloroform for NMR measurements was
purchased from Euriso-Top.
2.2. Catalysts preparation
Amberlyst 15-NO2 and Amberlyst 36-NO2 nitrated resins were
prepared as follows. In a typical preparation 5 g of styrene
divinyl benzene sulfonated resin was dispersed in 50 mL of H SO4
2
(
95–98 wt.%) and thermostatted at 273 K. The nitrating mixture was
[
28,29].
The initial reaction rate is calculated by the first derivative at
prepared by diluting 40 mmol of HNO (65 wt.%) in 50 mL of H SO
3
2
4
(
95–98 wt.%). The solution was thermostatted and added drop wise
time zero of the function obtained by the regression of cyclohexene
consumption data with a third order polynomial. In this way, we
calculate the apparent activation energy (AAE) by the slope of the
Arrhenius plot obtained from the initial rate of reaction at different
temperatures [24].
to the resin suspended in the sulfuric acid and left under gently
agitation for the time of treatment (1–3 h). Subsequently, the solid
was filtered and washed with plenty water, whose pH was continu-
ously checked until the neutrality was reached in order to minimize
the acid adsorbed. The catalyst was finally desiccated at 373 K for
7
2 h under nitrogen flow. Sulfonation on the commercial sulfonated
2.5. NMR measurements
resins (Amberlyst 15-SO3 and Amberlyst 36-SO ) was carried out
3
by the same procedure described above but in the presence of neat
H SO (98 wt.%) at 298 K for 2 h.
13C proton decoupled nuclear magnetic resonance spectra
NMR) were recorded at 75 MHz at 243 K with a Bruker Avance 300
2
4
(
spectrometer. NMR chemical shifts are internally referred to the
solvent resonance and are quoted relative to internal tetramethyl-
silane (ı = 0 ppm). All the measurements were carried out in tubes
sealed by a screw cap in order to add known amounts of methane-
sulfonic by a micro syringe. In a typical experiment 0.6 mmol of
phenol, cyclohexene or cyclohexyl phenyl ether were added to
2
.3. Catalysts characterization
The total ions exchange capacity (TIEC) of the solids was deter-
mined by potentiometric back titration with HCl of standard
solutions of NaOH after adsorption of the base on the resins.

1
94
L. Ronchin et al. / Journal of Molecular Catalysis A: Chemical 353–354 (2012) 192–203
0
.6 mL of CDCl3 and, after standard measurements, incremental
1
530
Amberlyst 15 NO2
amounts of CH SO H were added to the NMR tube cooled to 243 K
3
3
1350
0
0
0
0
0
0
0
0
.7
.6
.5
.4
.3
.2
.1
.0
and transferred into the probe pre-cooled at 243 K.
2.6. Etherification and adsorption equilibria
Amberlyst 15 SO3
The equilibrium of etherification was studied by measuring the
products of cyclohexyl phenyl ether decomposition (phenol and
cyclohexene) catalyzed by methanesulfonic acid, which is a poor
catalyst for ring alkylation reactions [24]. The observed equilibrium
constants are measured by using the same reactor with the same
reaction volume used for the reactions of alkylation, in the temper-
ature range comprised between 348 K and 368 K, in the presence
of 180 mg of methanesulfonic acid, with an initial ether concen-
Amberlyst 15
−1
tration of c.a. 0.2 mol L . When the concentration of both phenol
and cyclohexene reached their maxima (e.g. see Fig. 6), the pre-
equilibrium approximation is assumed to hold [30]. Cyclohexene,
phenol and cyclohexyl phenyl ether are then in thermodynamic
equilibrium, thus the etherification equilibrium constant can be
calculated by the concentrations of reagent and products.
4
000
3500
3000
2500
2000
1500
1000
500
-1
Wavenumbers (cm )
Fig. 1. Comparison of the infrared spectra of neat, sulfonated and nitrated Amberlyst
1
5 in KBr.
The equilibrium of adsorption on the resins is evaluated by mea-
suring the adsorption of reagents and products between 278 K and
ꢀ
i=n
i=0
2
98 K and extrapolated at the temperature of reaction. Adsorption
2
k2
2
ꢀ
=
[Y − f (X , kro, krp)]
(2)
i
i
measurements were carried out at temperatures lower than those
normally employed in the reactions to maximize the amount of the
adsorbate. In addition, under these conditions, product formation
is suppressed. In a typical experiment 800 mg of catalyst is placed
in contact with 10 mL of solution at known concentration for 1 h,
in the same reactor used for alkylation reactions (stirring speed:
The calculated error on the kinetic constants are in the range
0–50% because of the large number of fitting variables, and the
goodness of the model is analyzed by the time sequence plot of the
residuals [33].
1
1
2 Hz), and the amount of adsorbate evaluated by difference after
3
. Results and discussion
GC analysis. Langmuir adsorption model is used to fit experimental
data and the equilibrium constant was measured at low coverage
at the initial slope of fractional coverage [31].
3.1. Synthesis and characterization of catalysts
Sulfonated styrene divinyl benzene resins (AmberlystTM 15 and
2
.7. Non linear regression analysis of the kinetic model
36) have been treated with nitric acid and sulfuric acid in order to
obtain new acid catalysts to be tested in the alkylation of phenol
with cyclohexene. Average pore diameter, pore volume, BET sur-
face area and TIEC of the resins are summarized in Table 1. Average
pore diameter (AVP) does not significantly change upon treatments.
Only a moderate increase of the AVP is observed for the sulfonated
Amberlyst 15 and for the nitrated Amberlyst 36. On the contrary,
pore volume and surface area increase after the treatments partic-
ularly for Amberlyst 36. As expected, TIEC values slightly increase
only after sulfonation, most likely due to formation of new sulfonic
groups on the polymeric framework. On the contrary, nitration
diminishes the TIEC probably because of the substitution of sul-
fonic groups with nitro groups, which cannot exchange ions [14,15].
This is in agreement with the data obtained by elemental analy-
sis of the Amberlyst 15-NO2 (Table 1, entry 2), which corresponds
The regression of the data was carried out on simultane-
ous algebraic-differential equations numerically evaluated at each
experimental point (X , Y ). The minimization of the square residual
sum was achieved by a step-descent method and the convergence
was verified by reducing the step of 10 time and obtaining constant
values of the square residuals sum [32,33]. The regression function
i
i
f depends on the kinetic constants (ke, ko, kp, kro, krp, k , kdcp), which
d
are also the regression variables (see details in Section 3.2.5). The
large number of parameters employed in the model does not allow
to obtain reliable values for the whole set of kinetic constants by a
direct fitting of a single set of data. For this reason the evaluation has
been carried out by dividing the kinetic constants in three groups
of unrelated evaluation steps, which correspond to independent
experimental data. The first group of kinetic constants (ke, ko, kp,
kdcp) are those directly related to the consumption of the reagents
and are evaluated by fitting the data of reactions of cyclohexylation
of phenol employing Eq. (1). The second group (kro, krp) is evaluated
by fitting the profile of cylohexylphenyl ether rearrangement with
−
1
to a elemental composition of C26H34N that is c.a. 2 mequiv. g
of NO2 on the resin, by considering that nitrogen derives by the
nitro group. At difference of both Amberlyst 15 and Amberlyst 15-
SO3, the IR spectrum of the nitrated resins show two new sharp
−
1
−1
bands at 1530 cm and 1350 cm , assigned to asymmetric and
symmetric NO stretching of the nitro group [34] (Fig. 1). Molecular
nitric acid, its hydrates and nitrate ions are not detected by IR anal-
Eq. (2). Finally, the evaluation of k , the kinetic constant relative
d
to the cyclohexene dimerization, can be simply evaluated by the
slope of the cyclohexylcyclohexene formation standardized by the
acid concentration, since the observed reaction order is zero in the
cyclohexene concentration (see Section 3.2.4).
This procedure is possible because each set of parameters does
not influence significantly the fitting of the others, in this way, the
regression variables of each fitting results unrelated:
−
1
ysis in the characteristic range of frequencies (1600–1700 cm
)
assigned to these compounds [35]. As expected the IR spectra of
the Amberlyst 15-SO3 does not change significantly. The spectra
relative to neat Amberlyst 36, Amberlyst 36-SO3 and Amberlyst
36-NO2 are reported in Fig. 2. As a matter of fact, Amberlyst 36
and Amberlyst 36-SO3 show negligible spectral differences, while
−
1
Amberlyst 36-NO shows the two sharp bands at 1530 cm and
2
ꢀ
i=n
i=0
−
1
2
k1
)]2
1350 cm of the NO2 assigned to asymmetric and symmetric NO
stretching [34]. This is in agreement with the behavior of Amberlyst
ꢀ
=
[Y − f (X , ke, ko, kp, k
(1)
i
i
dcp

L. Ronchin et al. / Journal of Molecular Catalysis A: Chemical 353–354 (2012) 192–203
195
Table 1
Surface area, pore volume, average pore diameter and total ion exchange capacity of the catalysts.
Pore volume (cm3
g
−1
)
Surface area (m2
g
−1
)
−1
TIEC (mequiv. g )
Entry
Catalyst
AVP (nm)
1
2
3
4
5
6
Amberlyst 15
23
27
25
22
27
22
0.21
0.22
0.26
0.05
0.13
0.13
37
34
42
10
20
24
4.7
2.6
4.8
5.5
5.3
5.7
Amberlyst 15-NO2a
Amberlyst 15-SO3
Amberlyst 36
Amberlyst 36-NO2
Amberlyst 36-SO3
a
Elemental analysis of this samples shows a composition of C26H34N.
1
5, suggesting that nitration is an effective chemical modification
phenol cyclohexylation, while Fig. 4 shows the trends of formation
of the products of di-alkylation and of cyclohexene dimerization.
Finally, Fig. 5 shows the trend of phenyl cyclohexyl ether decom-
position and the relative fitting which is carried out independently
from that of the phenol cyclohexylation.
of the cross-linked styrene divinylbenzene resins structure.
3.2. Reaction profiles
Figs. 3–5 report typical time–concentrations profiles together
with the fitting curves obtained by the kinetic model described in
Section 3.2.5. Fig. 3 shows a typical time profile concentration of
0
0
0
0
0
.08
.06
.04
.02
.00
1
530
1
350
Amberlyst 36 NO2
0.3
0.2
0.1
0.0
Amberlyst 36 SO3
Amberlyst 36
0
200
400
600
Time minutes
800
1000
1200
4000
3500
3000
2500
2000
1500
1000
500
Fig. 4. Time–concentrations and fitting profiles of di-cylohexylphenols and of
cyclohexylcyclohexene in phenol cyclohexylation catalyzed by Amberlyst 15. Run
conditions: T, 358 K; phenol, 1.1 mol L ; cyclohexene, 1.2 mol L ; catalyst, 400 mg;
solvent, 1,2-dichloroethane; solution volume, 10 mL. Cyclohexylcyclohexene (ꢀ)
and di-cyclohexylphenol (ꢁ).
-1
Wavenumber (cm )
−1
−1
Fig. 2. Comparison of the infrared spectra of neat, sulfonated and nitrated Amberlyst
6 in KBr.
3
1
0
0
0
0
0
.0
.8
.6
.4
.2
.0
0.25
0.20
0
0
0
0
.15
.10
.05
.00
0
200
400
600
800
1000
1200
0
200
400
600
800
1000
1200
1400
Time minutes
Time minutes
Fig. 3. Time–concentrations and fitting profiles of reagents and products of phe-
nol cyclohexylation catalyzed by Amberlyst 15. Run conditions: T, 358 K; phenol,
Fig. 5. Time–concentrations and fitting profiles of the reactivity of the cyclo-
hexylphenyl ether in the presence of Amberlyst 15. Run conditions: T, 358 K;
cyclohexylphenyl ether, 0.2 mol L ; catalyst, 400 mg; solvent, 1,2-dichloroethane;
−1
−1
−1
1
.1 mol L ; cyclohexene, 1.2 mol L ; catalyst, 400 mg; solvent, 1,2-dichloroethane;
solution volume, 10 mL. Cyclohexene (ꢀ), phenol (ꢀ), cyclohexylphenylether (ꢁ),
-cyclohexylphenol (ꢁ), and 4-cyclohexylphenol (♦).
solution volume, 10 mL. Cyclohexene (ꢀ), phenol (ꢀ), cyclohexylphenylether (ꢁ),
2
2-cyclohexylphenol (ꢁ), and 4-cyclohexylphenol (♦).

1
96
L. Ronchin et al. / Journal of Molecular Catalysis A: Chemical 353–354 (2012) 192–203
by the slope of the Arrhenius plot of the initial reaction rate (e.g.
-
-
-
-
-
-
-
7.2
7.4
7.6
7.8
8.0
8.2
8.4
Fig. 6) and reported in Table 2, is practically constant in any case,
which suggests that the reaction path is similar for all the catalysts
employed.
Experimental data are superimposed in Fig. 7 to the simula-
tion curve of the initial reaction rates obtained by using the model
described in Section 3.2.5 (continuous lines). It appears that the
initial reaction rate depends on cyclohexene concentration with
an almost zero order (c.a. 0.1), while phenol concentration influ-
ences the rate with a power law of c.a. 1 (slightly higher). These
evidences suggest a complex reaction path where both cyclohexene
and phenol are involved in pre-equilibria [36]. This behavior is typ-
ical in heterogeneous catalyzed reactions, and can be explained by
Langmuir–Hinshelwood type mechanism, but these evidences are
not sufficient to asses a unique kinetic model. For this reason, we try
to grab some information by studying the equilibrium of etherifica-
tion in solution (Section 3.2.2) and the heterogeneous equilibrium
between reagents and catalyst sites (Section 3.2.3). In addition, by
the investigation of the cyclohexene oligomerization we model also
the side reaction that affects the overall kinetics (Section 3.2.4).
Finally, we develop a kinetic model, which takes into account all
the effects of the variation of the operative variables investigated
0
.00270
0.00275
0.00280
0.00285
-1
1
/T (K )
Fig. 6. Arrhenius plot of the initial reaction rate between 348 K and 368 K of the
reactions catalyzed by Amberlyst 15 (ꢀ) and Amberlyst 15-NO2 (ꢀ).
(see Section 3.2.5).
3.2.1. Influence of catalysts, reagents and temperature on the
initial reaction rate
In Table 2 the results obtained in the heterogeneous catalyzed
liquid phase cyclohexylation of phenol in the presence of different
sulfonic resins as catalyst are reported. We compare the conversion
and the selectivity after 4 h of reaction and the initial reaction rate
based on the cyclohexene consumption as a cumulative index of the
different kinetics. Selectivity toward O-alkylated and C-alkylated
is reported in Table 2. It is evident that products distribution does
not change significantly after 4 h of reaction by varying the cata-
lyst type. It is noteworthy, however, the selectivity in the alkylated
products is 95–97% and it does not change appreciably from 348 K
to 368 K. The main side product is cyclohexyl cyclohexene, which
can be decreased to a negligible concentration by using an excess
of phenol or by employing nitromethane as a solvent, further dis-
cussion well be in Section 3.2.4 [24].
A comparison between commercial and treated resins shows
that the latter one is generally more active. Among the treated
resins, the most active is Amberlyst 15 nitrated despite of its lower
TIEC values. Such evidence suggests a higher activity of the acid
sites of the nitrated resin but it is not clear what is the reason
of this growth. The apparent activation energy (AAE), calculated
3.2.2. Equilibrium of etherification of the cyclohexyl phenyl ether
from phenol and cyclohexene
The measurement of the equilibrium of etherification has been
carried out by following the trend of decomposition of the cyclo-
hexyl phenyl ether at 358 K in the presence of methanesulfonic acid
(Fig. 8). The use of methanesulfonic acid to promote decomposition
of phenyl cyclohexyl ether instead of other acid catalysts is due to its
low ring alkylation ability, as already discussed in a previous paper
[24]. It appears that the formation of the products of ring alkylation
proceeds very slowly compared to the decomposition of the ether.
Indeed, the initial reaction rate of decomposition and formation of
−
5
−1 −1
mequiv.H+
respectively)
−1
the cyclohexyl phenyl ether (5.2 × 10 mol L
s
−
6
−1 −1
−1
and
6 × 10 mol L
s
mequiv.H+ ,
are
larger than the formation of 2- and 4-cyclohexyl
−
7
−1 −1
−1
phenol
(3.5 × 10 mol L
s
mequiv.H+
and
−
8
−1 −1
−1
8.9 × 10 mol L
s
mequiv.H+ ,
respectively) [23]. These
evidences allow to apply the pre-equilibrium approximation to
the kinetics of the reaction catalyzed by methanesulfonic acid [29].
In this way, it is possible to calculate, at various temperatures,
the experimental equilibrium constants (Keq) by substituting the
0
0
0
0
0
.00020
.00015
.00010
.00005
.00000
0.00020
0.00015
0
0
0
.00010
.00005
.00000
0
.0
0.5
1.0
mol
L
1.5
0
.0
0.5
1.0
mol
L
1.5
cyclohexene
phenol
Fig. 7. Influence of reagents concentration on the initial reaction rate at T 358 K (dashed line) and simulation by the kinetic model (continuous line).

L. Ronchin et al. / Journal of Molecular Catalysis A: Chemical 353–354 (2012) 192–203
197
Table 2
Alkylation of phenol: conversion of cyclohexene in 1,2-dichloroethane as a solvent at 358 K. Run conditions: phenol, 1.2 mol L 1; cyclohexene, 1.2 mol L−1; catalyst, 400 mg;
−
solution volume, 10 mL; stirring speed, 12 Hz.
TIEC (mequiv. g 1
−
)
Time (min)
Conversion (%)
r0 (×10 mol L
5
−1 −1
s
mequiv.H+
−1
)
Selectivitya
alkylated (%)
AAEb (kJ mol−1
)
Catalyst
O–Cy
C–Cy
Commercial resins
Amberlyst 15
Amberlyst 36
4.7
5.5
240
240
51
52
6.79
6.28
52
45
46
53
54
55
Treated resins
Amberlyst 36-NO2
Amberlyst 36 SO3
Amberlyst 15-NO2
Amberlyst 15-SO3
5.3
5.7
2.6
4.8
240
240
240
240
60
57
58
67
6.45
6.33
10.1
8.6
47
49
44
48
50
48
52
49
55
55
55
56
a
The selectivity does not change significantly from 348 K to 368 K. O–Cy: cyclohexylphenyl ether and C–Cy: ring alkylated phenols.
AAE (apparent activation energy) is measured between 348 and 368 K.
b
Table 3
values of the concentration of cyclohexyl phenyl ether, phenol,
and cyclohexene in Eq. (3), after that the concentrations of phenol
and cyclohexene reach a maximum (see, Fig. 8). For instance, at
Influence of the temperature on the etherification equilibrium.
Cyclohexene
Phenol
(mol L
CPE
Keq (L mol −1)
T (K)
−
1
−1
−1
(mol L )
(
mol L
)
)
3
58 K after 90 min, the concentrations of cyclohexyl phenyl ether,
phenol, and cyclohexene correspond to those of equilibrium while
their relatively small and slow variations are due the formation
of o- and p-cyclohexyl-phenol. This approximation is not true in
the case of heterogeneously catalyzed reaction since adsorption
of reagents and products on catalyst sites strongly influence the
kinetics of each stage, resulting in a non equilibrium situation;
in addition, the rate of ring alkylation is comparable to that of
etherification:
0.164
0.186
0.171
0.19
0.201
0.033
0.03
0.0209
1.18
0.85
0.51
348
358
368
0.202
The associated error is 6–8% calculated from the standard deviation of the constants.
3
.2.3. Adsorption equilibria
In Fig. 10 the adsorption isotherms at 298 K of cyclohexene,
phenol, cyclohexyl phenyl ether and the combined adsorption of
cyclohexene on pre-adsorbed phenol are reported. It appears that
neat cyclohexene and cyclohexyl phenyl ether are adsorbed in a
negligible extent. On the contrary, phenol reaches a saturation
[CPE]
Keq =
(3)
(4)
[P][C]
ꢁG = −RT Ln Keq = ꢁH − T ꢁS
−1
value of 1.7 mequiv.gcat at 1 mol L of concentration and a Lang-
muir adsorption model is used to fit experimental data and the
equilibrium constant were measured at low coverage [31].
The influence of the temperature on the adsorption equilibrium
constant is reported in Fig. 11 and in Table 4. ꢁH of adsorp-
The influence of the temperature on the etherification equilibrium
constant has been reported in Table 3 and in Fig. 9. The values
of the enthalpy and of the entropy of reaction, calculated by Eq.
−1
−1 −1
(
4), are ꢁH = −44,500 J mol , ꢁS = −126 J mol
K , respectively.
−
1
tion (ꢁH = −38,000 J mol ) is quite low to account for a chemical
bond between phenol and sulfonic group, but it is compatible
to the formation of hydrogen bonds. The same adsorption model
employed for phenol is used for cyclohexene on pre-adsorbed phe-
These data are in agreement to those to those relative to aliphatic
ether such as MTBE and ETBE [37]. The precise thermodynamic
value of the equilibrium constant is beyond the scope of this work
but its estimated value is functional to the evaluation of the good-
ness of the kinetic model (see Section 3.2.5).
−
1
nol, and their saturation value reaches a value of 1.9 mequiv. g
−
1
at cyclohexene concentration of 1 mol L . The measured ꢁH of
adsorption (ꢁH = −22,000 J mol ) is lower than that found for
−
1
0.20
0.15
0.10
0.05
0.00
1
0
cyclohexene
phenol
-1
-2
ether
2-cyclohexyl phenol
4-cyclohexyl phenol
-
-
-
-
3
4
5
6
0
50
100
150
200
250
time (minutes)
0
.0026
0.0027
0.0028
0.0029
0.0030
-1
1
/T (K )
Fig. 8. Time profile concentrations of the cyclohexylphenyl ether decomposition
in the presence of CH3SO3H, evaluation of the etherification equilibrium. Run con-
−1
ditions: T, 358 K; cyclohexylphenyl ether, 0.23 mol L ; CH3SO3H, 180 mg; solvent,
,2-dichloroethane; solution volume, 10 mL.
Fig. 9. Influence of the temperature between 348 K and 368 K on the etherification
equilibrium: ꢁH = −44,500 J mol and ꢁS = −126 J mol
−
1
−1 −1
K .
1

1
98
L. Ronchin et al. / Journal of Molecular Catalysis A: Chemical 353–354 (2012) 192–203
OH
2
1
1
0
0
.0
.5
.0
.5
.0
PRODUCTS
cyclohexene on phenol
phenol
cyclohexene
PCE
O
H
0.0
0.2
0.4
0.6
0.8
1.0
H
-
1
concentration (mol L )
O
surface complex
Fig. 10. Adsorption isotherm at 298 K of phenol, cyclohexene, cyclohexene on
phenol and of cyclohexylphenyl ether on Amberlyst 15. Run conditions: catalyst,
SO
2
8
00 mg; solution volume, 10 mL; solvent, 1,2-dichoroethane.
Fig. 12. Proposed surface reaction mechanism.
phenol, suggesting a weaker interaction between olefin and the sys-
tem sulfonic resin + phenol (Fig. 10). No ether formation is observed
at 298 K. The negligible adsorption of both neat cyclohexene and
cyclohexylphenyl ether can be ascribed to the absence of strong
hydrogen bonds, which are responsible for phenol adsorption (see
later).
3
3
3
3
3
3
3
3
2
2
2
7
6
5
4
3
2
1
0
9
8
7
cyclohexene on phenol
phenol
In Table 4 the measured adsorption equilibrium constants
(obtained at 278–298 K) are reported together with those extrap-
olated by Eq. (4) from 348 K to 368 K, which will be employed in
the evaluation of the kinetic model in Section 3.2.5. Such a pro-
cedure does not allow to know the real values of the adsorption
constant at the temperatures of reaction, in fact, at those temper-
atures many reactions occur, with a different surface population
and consequently with a different free energy. However, this is
useful to fix the value of the adsorption constants in the kinetic
model with measured parameters whose values should be straight-
forward related to the real ones. In this way, it is possible to
implement a kinetic model, whose fitting variables are only kinetic
constants.
0.00335
0.00340
0.00345
0.00350
-1
0.00355
0.00360
1
/T (K )
In order to explain the trend of the initial rate of reaction
vs. both cyclohexene and phenol concentrations (Fig. 7), together
with the adsorption behavior of reagents and products, it is neces-
sary to suppose a reaction mechanism, where both a quite strong
phenol–cyclohexene adsorption and a strictly dependence of the
rate of reaction from the phenol concentration occur simultane-
ously. This can be modeled by supposing a surface complex formed
by phenol and cyclohexene adsorbed on a sulfonic group of the
resin reacting with a molecule of phenol in solution (see Fig. 12).
In this way, it is possible to explain the strong adsorption behavior
of phenol, the adsorption of cyclohexene on pre-adsorbed phenol
and their reaction orders (3/2 and zero for phenol and cyclohex-
ene, respectively). Except kinetic and adsorption data, there are no
direct experimental evidences of the formation of such a surface
complex. For this reason, we try to grab some information from
experiments in homogeneous non aqueous solution of CH3SO3H.
The interactions between CH SO H and cyclohexene in the pres-
Fig. 11. Adsorption equilibrium on Amberlyst 15 between 278 K and 298 K.
Run conditions: catalyst, 800 mg; solution volume, 10 mL; solvent, 1,2-
dichoroethane. ꢁH = −37,613 J mol−1, ꢁS = −99 J mol
−1 −1
K ,
cyclohexene on
−1
−1 −1
K .
phenol ꢁH = −22,015 J mol , ꢁS = −46 J mol
Table 4
Influence of temperature on Adsorption equilibrium constant, and extrapolation at
the temperatures of reaction.
T adsorption (K)
Phenol KHP
Cyclohexene on phenol KHPC
Measured
2
2
2
97
88
78
31
52
90
32
44
58
−
1
−1
−1 −1
ꢁ
H = −38,000 J mol
ꢁH = −21,000 J mol
1
−1
ꢁ
S = −100 J mol−
K
ꢁS = −44 J mol
K
Extrapolated using Eq. (4)
3
3
348
358
368
2.3
1.6
1.2
8.7
6.8
5.5
13
1
ence and absence of phenol are investigated. Then, C { H} are
studied by NMR at 243 K and the chemical shifts are reported
in Table 5. In the experimental conditions described, no signals
attributable to carbocation or ester formation were detected and
no reaction product was observed [38,39]. Addition of incremental
The associated error is 5–6% calculated from the linear fitting at low coverage.

L. Ronchin et al. / Journal of Molecular Catalysis A: Chemical 353–354 (2012) 192–203
199
Table 5
13
1
C { H} NMR data for CDCl3 solutions of cyclohexene, phenol, cyclohexyl phenyl ether and methanesulfonic acid at 243 K. Chemical shifts are reported in ppm.
Acid/cyclohexene (molar ratio)
Cyclohexene (chemical shifts)
CH3SO3H (chemical shifts)
Acid + cyclohexene
0
0
0
0
1
1
2
/1
127.2
127.2
127.1
127.1
127.1
127.1
127.1
24.9
24.9
24.9
24.9
24.9
24.9
24.9
22.3
22.3
22.3
22.3
22.3
22.3
22.3
.25/1
.5/1
.75/1
/1
.5/1
/1
39.3
39.3
39.3
39.3
39.3
39.3
39.0 bra
39.0 br
38.8 br
38.9 br
38.9 br
38.8 br
38.5
38.5
38.5
38.4
Acid/cyclohexene (molar ratio)
Acid + phenol
Phenol (chemical shifts)
CH3SO3H (chemical shifts)
0
0
0
0
1
1
2
/1
154.3
153.8
153.3
152.9
152.8
153.0
152.9
129.7
129.7
129.7
129.7
129.7
129.7
129.7
121.0
121.3
121.6
121.8
121.8
121.7
121.7
115.1
.25/1
.5/1
.75/1
/1
.5/1
/1
115.1
115.1
115.1
115.0
115.1
115.1
39.1
39.1
39.0
39.0
39.0
39.0
Acid/cyclohexyl phenyl (molar ratio)
Acid + cyclohexyl phenyl ether
Cyclohexyl phenyl ether (chemical shifts)
CH3SO3H (chemical shifts)
0
0
0
0
1
1
2
/1
157.1
156.5
156.7
156.7
156.7
156.7
156.7
129.3
129.4
129.4
129.4
129.4
129.4
129.4
120.1
120.8
120.5
120.7
120.6
120.6
120.6
115.3
115.9
115.7
115.7
115.7
115.7
115.7
31.5 br
25.3
25.2
25.3
25.2
25.2
25.2
25.2
23.9 br
23.9 br
23.8 br
23.8 br
23.8 br
23.8 br
.25/1
.5/1
.75/1
/1
.5/1
/1
31.5 br
31.5 br
31.5 br
31.5 br
31.5 br
31.5 br
39.1
39.1
39.1
39.1
39.1
39.1
38.8 br
38.8 br
38.8 br
38.8 br
238 br
Acid/cyclohexene (molar ratio)
Acid + cyclohexene + phenol
Cyclohexene (chemical shifts)
Phenol (chemical shifts)
CH3SO3H (chemical shifts)
0
0
0
0
1
1
2
/1/1
127.2
127.2
127.1
127.1
127.1
127.1
127.1
24.9
24.9
24.8
24.8
24.8
24.8
24.8
22.3
154.4
153.6
153.4
153.2
153.1
153.0
153.0
129.7
129.7
129.7
129.7
129.7
129.7
129.6
120.9
121.3
121.4
121.5
121.6
121.6
121.7
115.1
115.1
115.1
115.1
115.0
115.0
115.0
.25/1/1
.5/1/1
.75/1/1
/1/1
.5/1/1
/1:1
22.3
22.3
22.2
22.2
22.2
22.2
39.0
39.0
38.9
38.9
39.0
38.9
br: broad.
−1
amounts of CH SO H to a 1 mol L
CDCl3 solution of cyclohex-
These results obtained in CDCl3 in the presence of methanesul-
3
3
ene did not cause any meaningful variation of the olefin chemical
shifts. Cyclohexene appeared to be insensitive to the acid addition
also in the presence of phenol, while the variations of chemical
shifts on increasing the acid concentration for phenol O-bonded
carbon atom and, to a lesser extent, for the para-carbon atom agree
with the formation of an adduct between CH SO H and phenol.
fonic acid are in agreement to the formation of hydrogen bonded
phenol to the sulfonic group of the resin thus allowing its adsorp-
tion on the catalyst. Additionally, the negligible interaction of both
the olefin and the ether with the methane sulfonic acid is in agree-
ment to their poor tendency to adsorb on the sulfonic resin. On
the contrary, the results of the NMR measurements in solutions do
not allow any explanation relative to the large values of cyclohex-
ene adsorption on pre-adsorbed phenol. However, the quite low
values of the ꢁH of adsorption of cyclohexene measured on the
resin–phenol system suggest a specific physical interaction of the
cyclohexene with the sulfonic resin. As a matter of fact, the only
measurable interaction of methanesulfonic acid with phenol and
cyclohexene is physical in nature, probably for this reason, equilib-
ria in solution are not easily measurable by NMR.
3
3
Measurements on phenol/acid mixtures showed that the interac-
tion between these two species is not sensitive to the presence of
1
3
cyclohexene. It is noteworthy that in the absence of phenol the
C
1
{
H} NMR spectra of CH SO H show a variable number of signals
3
3
in the range 38–39 ppm depending upon acid concentration. When
a low quantity of acid is present in the NMR tube a sharp singlet
at 39.3 ppm and a broad signal around 39.0 ppm are detectable.
On increasing CH SO H the relative intensity of the broad sig-
3
3
nal increases and a new sharp singlet around 38.5 ppm appears,
whose relative intensity grows on acid addition. These observa-
tions are indicative of the formation of quite complex mixtures of
methanesulfonic acid adducts in the absence of phenol. When also
phenol is present in solution, instead, only a sharp singlet around
3
.2.4. Oligomerization of cyclohexene
Regarding side reactions, cyclohexene oligomerization (practi-
cally only dimerization) is the one which may significantly affect
the kinetics and for this reason it must be accounted for in the
overall kinetic model. In Fig. 13 the comparison of the trends of
cyclohexene dimerization in the absence and in the presence of
phenol is reported. The reaction has been carried out in the absence
of phenol with Amberlyst 15 as a catalyst at 358 K and in 1,2-
dichloroethane as a solvent. Cyclohexene disappears with an initial
3
9.0 ppm is detectable in the acid range considered, this confirming
the existence of relatively strong interactions between CH SO H
3
3
and phenol. Finally, cyclohexyl phenyl ether chemical shifts are not
significantly affected by the presence of methanesulfonic acid in the
described experimental conditions, as deducible from Table 5. The
data reported suggest only a very weak interaction between ether
and CH SO H.
−5 −1 −1
mequiv.H+ , which is 25 times
reaction rate of about 3 × 10
s
3
3
higher than that measured in the presence of phenol. In addition,

2
00
L. Ronchin et al. / Journal of Molecular Catalysis A: Chemical 353–354 (2012) 192–203
(
HP)
OH
(
P)
OH
H
KHP
1
0
0
.2
.8
.4
1
2
+ (H)
(
C)
(HPC)
OH
H
KHPC
(HP)
+
OH
(
CPE)
cyclohexene disappearing without phenol
dimer formation with phenol
Keq
O
3
+
0
.02
.00
ke
0
4
5
(
HPC) + (P)
(CPE) + (HP)
0
50
100
150
200
250
300
ke
time (minutes)
(CPE)
+
(H)
(C)
+
(HP)
Fig. 13. Cyclohexene dimerization/oligomerization in the absence and in the pres-
ence of phenol. Run conditions: cyclohexene, 1.2 mol L ; catalyst, 400 mg; solution
volume, 10 mL. Phenol 1 mol L when phenol is present.
(
oCP)
−1
OH
−1
OH
ko
+ (HP)
6
(HPC) +
we found a complex mixture of isomers and oligomers, very dif-
ferent to the composition of products formed in the presence of
phenol. It is noteworthy that in both cases an almost linear profile
is observed, thus suggesting the constancy of the active specie as
the concentration of cyclohexene decrease.
The lower reaction rate, observed in the reaction carried out
with phenol and its different product distribution, is likely due to
the inhibiting effect of the phenol on the cyclohexene oligomeriza-
tion, for the steric hindrance of the species adsorbed on the surface
of the catalyst. A detailed study of this reaction is beyond the scope
of this work, but now it is clear that the dimerization of cyclohex-
ene affects the overall kinetic and an equation for explaining this
reaction must be included in the kinetic model.
(pCP)
OH
kp
HO
+ (HP)
(
HPC)
+
7
8
O
kro
(oCP)
(
H)
O
krp
H)
(
pCP)
9
(
3
.2.5. Kinetic model and evaluation of the kinetic parameters
In this section we try to develop a kinetic model starting from
(
DCP)
OH
OH
the experimental evidences described in previous sections. As a
matter of fact, an Eley–Rideal type mechanism, where phenol in
solution reacts with a surface electrophilic complex, is proposed
also by Yadav and Kumar [18]. They studied the kinetics of reaction
in solvent-less conditions, but in that paper some aspects relative
to adsorption and etherification equilibria are not considered [18].
Starting from those studies, we implement both the etherification
and the adsorption equilibria in the kinetic model, thus allowing a
comprehensive explanation of the phenomena (see Fig. 5).
In Scheme 1 the proposed reaction stages are reported: reac-
tions (1) and (2) are the surface equilibrium, between reagent and
the surface acid site (H); (3) and (4) are the equations of the ether-
ification equilibrium, in which direct and reverse have different
interactions with catalyst surface; reactions (5)–(7) are the alkyla-
tion of phenol; (8) and (9) are the rearrangement of the cyclohexyl
phenyl ether; finally steps (10) and (11) are the dicyclohexylation
of phenol and the dimerization of cyclohexene, respectively. Eqs.
kdcp
+
(HP)
1
1
0
1
+ (HPC)
(CYCY)
kd
+
(
H)+(HP)+(HPC)
Scheme 1. Reactions stages of phenol cyclohexylation on sulfonic resins. (H) is the
proton of the sulfonic resin.
(iv) cyclohexyl phenyl ether is negligibly adsorbed on catalyst
surface;
(v) cyclohexyl phenyl ether decomposition and rearrangement
are catalyzed by acid sites (H) and both reactions are of first
order on ether concentration in agreement to the negligible
adsorption behavior;
(
6)–(15) derive from the previous stages and by taking into consid-
(vi) di-cyclohexylation occurs by a consecutive alkylation by
Eley–Rideal mechanism between the surface complex and the
alkylphenol in solution;
eration the following assumptions:
(
i) all the reactions are acid catalyzed, and the sites of the catalyst
are energetically equivalent at any grade of coverage;
(vii) cyclohexene dimerization is catalyzed by all kinds of acid sites
and shows a zero apparent order of reaction in the cyclohex-
ene concentration;
(viii) a cumulative mass balance for reagents and acid sites (Eq.
(14)) needs in order to reach numerical convergence of the
system of algebraic and differential equations [40].
(
ii) phenol is adsorbed, pure cyclohexene is negligibly adsorbed
but is adsorbed in the presence of phenol;
(
iii) phenol attacks the cyclohexene–phenol complex on the cat-
alyst surface by an Eley–Rideal mechanism;

L. Ronchin et al. / Journal of Molecular Catalysis A: Chemical 353–354 (2012) 192–203
201
[
HP]
KHP =
(5)
(6)
[H][P]
[HPC]
KHPC = [
HP][C]
d[P]
dt
−
= ke[HPC][P] − ker[CPE][H] + k [HPC][P] + kp[HPC][P]
0
+
kdcp[HPC]([oCP] + [pCP])
(7)
d[CPE]
dt
=
ke[HPC][P] − ker[CPE][H] − kro[CPE][H] − krp[CPE][H]
(8)
d[oCP]
dt
=
=
kro[CPE][H] + ko[HPC][P]
(9)
d[pCP]
dt
krp[CPE][H] + kp[HPC][P]
(10)
(11)
d[C]
−
= ke[HPC][P] − ker[CPE]H + ko[HPC][P] + kp[HPC][P]
dt
+
kdcp[HPC]([oCP] + [pCP]) + k ([HPC] + [HP] + [H])
d
d([DCP])
=
kdcp[HPC]([oCP] + [pCP])
(12)
(13)
dt
d([CYCY])
=
kd([HPC] + [HP] + H)
dt
[
P ] + [C ] + [Ht] = [P] + [C] + [H] + 3[HPC] + 2[HP] + 2[CPE]
0
0
+
2[oCP] + 2[pCP] + 3[DCP] + 2[CICI]
(14)
Eq. (15) represents the rate of formation of cyclohexylphenyl ether
and it is zero at the maximum of the concentration profile of the
cyclohexylphenyl ether (see Fig. 4). In this way, ker can be calculated
directly, from Eq. (16). This is obtained from Eq. (15) by neglecting
the terms kro [CPE] [H] − krp [CPE] [H], which are relative to the
ether rearrangement and being much smaller compared to those
of the etherification. For this reason ker is not considered a fitting
variable but an implicit function of ke, KHP, KHPC and Keq (adsorption
and etherification equilibrium constant). In this way, ker is calcu-
lated from Eqs. (3), (5), (6) and (16) from available equilibrium and
kinetic constant, thus reducing the number of the fitting variables
by one:
d[CPE]
dt
0
=
−
= ke[HPC][P] − ker[CPE][H] − kro[CPE][H]
krp[CPE][H]
(15)
(16)
d[CPE]
dt
0
=
= ke[HPC][P] − ker[CPE][H]
The numerical solution of the complete set of equations ((3) and
8)–(16)) allows the fitting of the experimental data [32,33,40].
(
Typical results have been reported in Figs. 3–5. It is noteworthy,
the good agreement between the effect of phenol and cyclohex-
ene concentration on the initial reaction rate and its simulation
obtained by using the fitted kinetic constants. A further proof of
the goodness of the model is the detailed description of the differ-
ences in the ortho/para selectivity observed between the reactions
of phenol cyclohexylation and cyclohexylphenyl ether rearrange-
ment [24]. This behavior is related to the amount of products of ring
alkylation formed by the two routes. Indeed, it is well known that
ortho isomer is favored via rearrangement and this is evident by

2
02
L. Ronchin et al. / Journal of Molecular Catalysis A: Chemical 353–354 (2012) 192–203
Table 7
Fitting results of the kinetic model applied to cyclohexyl phenyl ether decomposition at substrate concentration of 0.2 mol L−1.
3
−1 −1
3
−1 −1
3
−1 −1
3
−1 −1
s )
T (K)
kro (×10 L mol
s
)
εro (×10 L mol
s
)
krp (×10 L mol
s
)
εrp (×10 L mol
348
358
368
1.25
1.57
1.83
0.11
0.13
0.15
0.28
0.37
0.47
0.025
0.036
0.044
Table 8
Activation energy of the kinetic constants of the model.
observing that the ratio kro/krp ≈ 4 while ko/kp ≈ 2 (Tables 6 and 7)
14,15].
In Tables 6 and 7 the results of the fitting with Amberlyst 15
[
Constant
Intercept (J mol 1
−
K
−1
)
Slope (J mol
−1
)
as catalyst and at different operation conditions are reported. The
evaluation of the model by fitting the experimental data shows
that all the parameters are consistent. In fact, the kinetic constants
remain practically unaffected by varying reagent concentration
ke
ker
ko
kp
kd
kdcp
kro
krp
47
110
150
160
110
110
37
−29,000
−48,000
−68,000
−74,000
−53,000
−58,000
−20,000
−27,000
(Table 6); as expected, their values rise as temperature increases.
The cumulative Arrhenius plot for all the kinetic constants of the
model are reported in Fig. 14 and their numerical values are in
Table 8. Even though, the model allows a comprehensive explana-
tion of all the aspects of the reaction, its complexity would require
a much larger number of experiments in order to have numerical
values with statistical meaning for each kinetic constant and the
respective activation energy. In any case, in order to take the mea-
sure of the goodness of the model the time sequence plots of the
residuals are studied. For instance, it can be seen in Fig. 15 that the
residuals plot of a typical kinetics are well behaved around zero
suggesting a good reliability of the model [33].
43
-
-
-
-
-
-
20
30
40
50
60
70
4. Conclusions
We tested commercial and modified macroreticular sulfonic
TM
TM
styrene divinylbenzene resins (Amberlyst
15 and Amberlyst
3
6) in the cyclohexylation of phenol. The modification of these
materials by nitration and sulfonation allowed obtaining new cat-
alysts with activities higher than the commercial materials. The
reaction showed a complex kinetic path, which is characterized
by the formation of the cyclohexyl phenyl ether as an equilibrium
intermediate. The influence of the reagents on the initial reaction
rate shows a zero reaction order for cyclohexene and 3/2 for phenol.
In addition, adsorption equilibrium of phenol on the resin is strong
while cyclohexene is faintly adsorbed in the absence of phenol.
Also cyclohexyl phenyl ether is poorly adsorbed on the resins. Fur-
thermore, the adsorption of cyclohexene on pre-adsorbed phenol is
almost strong as phenol adsorption. Starting from these evidences
an Eley–Rideal mechanism is proposed where a surface complex
sulfonic group-phenol–cyclohexene reacts with phenol in solution
to give both the ether and the products of ring alkylation. The
rearrangement of the ether and its decomposition are, instead, cat-
alyzed by the free acid sites. Parallel independent reaction path is
proposed for cyclohexene dimerization. The model satisfactorily
fits the data giving a comprehensive explanation of the different
aspects of the reaction.
0
.00272
0.00276
0.00280
0.00284
0.00288
-1
1
/T (K )
Fig. 14. Arrhenius plot between 348 K and 368 K of the kinetics constants calculated
by the fitting results of the proposed kinetic model.
0
0
0
0
0
0
0
.06
.04
.02
.00
.02
.04
.06
Notations
P
C
H
P0
C0
Ht
oCP
pCP
CPE
HP
HPC
phenol concentration (mol L 1
−
)
cyclohexene concentration (mol L−1)
free H+ concentration (mol L
−1
)
initial phenol concentration (mol L−1)
initial cyclohexene concentration (mol L−1)
total H+ concentration (mol L
−1
)
−
1
2-cyclohexyl phenol concentration (mol L
)
4-cyclohexyl phenol concentration (mol L−1)
0
200
400
600
Time minutes
800
1000
1200
−
1
cyclohexylphenyl ether (mol L
)
concentration of adsorbed phenol (mol L−1);
concentration of the surface complex phenol cyclohexene
Fig. 15. Time sequence plot of the fitting residuals for a typical reactions kinetics.
−
1
(mol L
)

L. Ronchin et al. / Journal of Molecular Catalysis A: Chemical 353–354 (2012) 192–203
203
−
dcp
cYcY
Keq
KHP
KHPC
di-cyclohexylphenols concentration (mol L 1
)
[2] J.F. Lorenc, G. Lambeth, W. Scheffer, Kirk Ohtmer “Encyclopedia of Chemical
Technology”, vol. 2, Wiley, 2001.
cyclohexylcyclohexene concentration (mol L 1)
−
[
3] W. Keim, M. Röper, Ulmann’s Encyclopedia of Industrial Chemistry, 6th ed.,
Wiley, 1998.
−
1
constant of etherification equilibrium (L mol
)
phenol adsorption equilibrium constant (L mol−1)
[4] M.A. Harmer, Q. Sun, Appl. Catal. A: Gen. 221 (2001) 45–62.
[
[
[
5] W.F. Hölderich, G. Heitmann, Catal. Today 38 (1997) 227–233.
6] A. Mitsutani, Catal. Today 73 (2002) 57–63.
7] M.M. Sharma, React. Funct. Polym. 26 (1995) 3–23.
adsorption equilibrium constant of cyclohexene on pre-
−1
adsorbed phenol (L mol
)
−
1
−1
)
s )
ke
ker
ko
kinetic constant of ether formation (L mol
s
[8] H. Zhang, S.M. Mahajani, M.M. Sharma, T. Sridhar, Chem. Eng. Sci. 57 (2002)
315–322.
−
1
−1
kinetic constant of ether destruction (L mol
[
9] P.F. Siril, H.E. Cross, D.R. Brown, J. Mol. Catal. A: Chem. 279 (2008) 63–68.
kinetic constant of 2-cyclohexylphenol formation
[
10] R.A. Rajadhyaksha, D.D. Chaudhari, Ind. Eng. Chem. Res. 26 (1987) 1276–1280.
−1
−1
)
(
L mol
s
[11] A. de Klerk, R.J.J. Nel, Ind. Eng. Chem. Res. 46 (2007) 7066–7072.
[12] R. Amandi, K. Scovell, P. Licence, T.J. Lotz, M. Poliakoff, Green Chem. 9 (2007)
797–801.
13] N.I. Shuikin, E.A. Viktorova, Russ. Chem. Rev. 29 (1960) 560–570.
14] K. Schofield, Aromatic Nitration, Cambridge University Press, Cambridge,
1980.
15] P.B.D. De La Mare, J.H. Ridd, Aromatic Substitution, Butterworths Scientific
Publications, London, 1959.
16] B. Chaudhuri, M.M. Sharma, Ind. Eng. Chem. Res. 30 (1991) 227–231.
kp
kinetic constant of 4-cyclohexylphenol formation
−1
−1
)
(
L mol
s
[
[
kro
krp
kinetic constant of ether rearrangement to 2-
−
1
−1
)
cyclohexylphenol formation (L mol
s
[
kinetic constant of ether rearrangement to 4-
−
1
−1
s )
cyclohexylphenol formation (L mol
[
kd
kdcp
kinetic constant of cyclohexene dimerization (s−1)
kinetic constant of dicylohexylation product formation
[17] E. Modrogan, M.H. Valkenberg, W.F. Hölderich, J. Catal. 76 (2009) 177–187.
[
[
[
18] G.D. Yadav, P. Kumar, Appl. Catal. A: Gen. 286 (2005) 61–70.
19] R. Anand, K.U. Gore, B.S. Rao, Catal. Lett. 81 (2002) 33–41.
20] K. Wilson, D.J. Adams, G. Rothenberg, J.H. Clark, J. Mol. Catal. A: Chem. 159
(2000) 309–314.
21] G.D. Yadav, G.S. Pathre, Ind. Eng. Chem. Res. 46 (2007) 3119–3127.
22] G.D. Yadav, G.S. Pathre, J. Mol. Catal. A: Chem. 243 (2006) 77–84.
23] J.F. Izquierdo, F. Cunill, M. Vila, M. Jhorra, J. Tejero, Ind. Eng. Chem. Res. 33 (1994)
2830–2835.
24] L. Ronchin, L. Toniolo, A. Vavasori, J. Mol. Catal. A (2011).
25] Q. Ma, D. Chakraborty, F. Faglioni, R.P. Muller, W.A. Goddard, T. Harris, C. Camp-
bell, Y. Tang, J. Phys. Chem. A 110 (2006) 2246–2252.
[26] B. Bjoerkqvist, H. Toivonen, J. Chromatogr. 178 (1979) 271–276.
[27] S.J. Greg, K.S.W. Sing, Adsorption Surface Area and Porosity, 2nd ed., Academic
Press, 1982.
−1
−1
(
L mol
s )
εe
εer
εo
εp
εdcp
εd
calculated error of ke kinetic constant
calculated error of ker kinetic constant
calculated error of ko kinetic constant
calculated error of kp kinetic constant
calculated error of kdcp kinetic constant
calculated error of kd kinetic constant
calculated error of kro kinetic constant
calculated error of krp kinetic constant
[
[
[
[
[
εro
εrp
[
[
[
28] G.W. Roberts, in: P.N. Rylander, H. Greenfield (Eds.), Catalysis in Organic Syn-
Acknowledgements
thesis, Academic Press, 1976, pp. 1–44.
29] P.A. Ramachadran, R.V. Chaudhari, Three-Phase Catalytic Reactors, Gordon and
Breach Sciences Publisher, 1982.
Financial support by Ca’ Foscari University of Venice is gratefully
acknowledged (Ateneo fund 2009). A thank to Dr. Davide Montin
for some preliminary experiments carried out during his degree in
Industrial Chemistry. Finally, a special thank to Mr. Claudio Tortato
for the helpful discussions.
30] M. Rae, M.N. Berberan-Santos, Chem. Phys. 280 (2002) 283–293.
[31] J.J. Kipling, Adsorption of Solution of Non Electrolytes, Academic Press, London,
965, p. 25.
32] G. Quartarone, L. Ronchin, C. Tortato, A. Vavasori, Int. J. Chem. Kinet. 41 (2008)
07–112.
1
[
1
[33] D.M. Bates, D.G. Watts, Nonlinear Regression Analysis & Its Applications, Wiley,
New York, 1988.
[
34] M. West-Nielsen, P.M. Dominiak, K. Wozniak, P. Erik Hansen, J. Mol. Struct. 789
2006) 81–91.
35] G. Ritzhaupt, J.P. Devlin, J. Phys. Chem. 95 (1991) 90–95.
Appendix A. Supplementary data
(
[
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.molcata.2011.11.025.
[36] J.B. But, Reaction Kinetics and Reactor Design, 2nd ed., Marcel Dekker Inc., NY,
000, p. 169.
37] M. Vila, F. Cunill, J.F. Izquierdo, J.T.M. Iborra, Chem. Eng. Commun. 124 (1993)
23–232.
[38] G.A. Olah, A.M. White, D.H. O’Brien, Chem. Rev. 70 (1970) 561–591.
2
[
2
References
[
39] G.A. Olah, G.K. Surya Prakash, J. Sommer, Superacids, J. Wiley, 1985, p. 90.
40] http://reference.wolfram.com/mathematica/tutorial/NDSolveIntroductory
TutorialDAEs.html.
[
[1] H.U. Hammershaimb, T. Imai, G.J. Thompson, B.V. Vora, Kirk Ohtmer “Encyclo-
pedia of Chemical Technology”, vol. 2, Wiley, 2001.