A novel route to produce thymol by vapor phase reaction of m-cresol with isopropyl acetate over Al-MCM-41 molecular sieves

Article abstract of DOI:10.1016/j.jcat.2004.03.019

Mesoporous Al-MCM-41 (Si/Al = 55 and 104) and Al,Zn-MCM-41 (Si/(Al+Zn)=52) molecular sieves were synthesized hydrothermally. The materials were characterized by XRD, TGA, TPD (pyridine), ICP-AES, nitrogen sorption, and FT-IR techniques. The catalytic performance of these materials was examined in the vapor phase alkylation of m-cresol with isopropyl acetate. The products obtained were thymol, isothymol, 2-isopropyl-5-methylphenyl acetate (2-I-5-MPA), and isopropyl-3-methylphenyl ether (I-3-MPE). The time-on-stream study was carried out over Al-MCM-41(55) at 300°C and WHSV 1.52 h^-1 wherein optimum conversion of m-cresol and selectivity to thymol were obtained. The dependence of activity and selectivity on the acid sites and hydrophobic and hydrophilic properties of the catalysts is discussed.

Full text of DOI:10.1016/j.jcat.2004.03.019

Journal of Catalysis 224 (2004) 347–357
www.elsevier.com/locate/jcat
A novel route to produce thymol by vapor phase reaction of m-cresol
with isopropyl acetate over Al-MCM-41 molecular sieves
K. Shanmugapriya, M. Palanichamy, Banumathi Arabindoo, and V. Murugesan ∗
Department of Chemistry, Anna University, Chennai-25, India
Received 22 September 2003; revised 5 March 2004; accepted 9 March 2004
Available online 20 April 2004
Abstract
Mesoporous Al-MCM-41 (Si/Al = 55 and 104) and Al,Zn-MCM-41 (Si/(Al + Zn) = 52) molecular sieves were synthesized hydrother-
mally. The materials were characterized by XRD, TGA, TPD (pyridine), ICP-AES, nitrogen sorption, and FT-IR techniques. The catalytic
performance of these materials was examined in the vapor phase alkylation of m-cresol with isopropyl acetate. The products obtained were
thymol, isothymol, 2-isopropyl-5-methylphenyl acetate (2-I-5-MPA), and isopropyl-3-methylphenyl ether (I-3-MPE). The time-on-stream
◦
−1
study was carried out over Al-MCM-41(55) at 300 C and WHSV 1.52 h wherein optimum conversion of m-cresol and selectivity to
thymol were obtained. The dependence of activity and selectivity on the acid sites and hydrophobic and hydrophilic properties of the cata-
lysts is discussed.
2004 Elsevier Inc. All rights reserved.
Keywords: Acylation; Thymol; Isothymol; Al-MCM-41; Isopropyl acetate; m-Cresol
1. Introduction
through isomerization and transalkylation reactions [16].
Mesoporous Al-MCM-41 molecular sieves have become
catalysts in widespread use in recent years in many reactions
of industrial importance. Their large pore diameter with less
diffusional constraint could discourage consecutive alkyla-
tions and deactivation by coke formation. In addition, un-
like zeolites, these materials could also permit simultaneous
entry of many reactant molecules into the pores, thus per-
mitting high conversion. Umamaheswari et al. [3] reported
vapor phase isopropylation of m-cresol with isopropyl al-
cohol over mesoporous Al-MCM-41 molecular sieves. The
selectivity of thymol was higher than that of other alkylated
products.
There is a general problem in using alcohols as alkylating
agents. They readily cluster around the protonic sites with-
out dissociation to isopropyl cation. To avoid this problem
esters can be substituted in the place of alcohols as reported
by Umamaheswari et al. [17]. It was inferred that tert-butyl
acetate is a better alkylating agent than the corresponding
alcohol in the vapor phase reaction of m-cresol. Adsorption
of ester through the steric free carbonyl group and subse-
quent liberation of tert-butyl cation are easily possible with
ester as alkylating agent. These observations and inferences
with MCM-41 have stimulated further interest in the study
of the isopropylation of m-cresol with isopropyl acetate as
Isopropylation of m-cresol continues to attract consid-
erable attention, because one of the alkylation products,
thymol, leads to menthol on hydrogenation [1–3]. Menthol
has a characteristic peppermint odor and is used as a raw
material for the production of antiseptic, local anesthetic,
antibacterial and antifungal agents, flavors, fragrances, and
preservatives [4,5]. Thymol as such exhibits activity in
protecting low-density lipoproteins [6] and as an antioxi-
dant [7–9]. It also plays an important role as intermediate in
perfumery. Isopropylation of m-cresol with isopropyl alco-
hol has been already reported over oxide catalysts [10] and
calcined Mg–Al hydrotalcites [11] and with propylene over
acidic catalysts [12]. Activated alumina [13], γ -Al2O3 [14],
Al(OC6H5)3, and ZnCl-HCl [1,2] have been used as cata-
lysts in the liquid-phase isopropylation of m-cresol. Y and β
zeolites, ZSM-5, erionite and mordenite [15], and supported
metal sulfates [1] were used in the vapor phase for such re-
actions. Synthetic silica–alumina and large-pore Y-type cat-
alysts have yielded 4-isopropyl-3-methylphenol (isothymol)
*
Corresponding author. Fax: +91-44-22200660.
E-mail address: v_murugu@hotmail.com (V. Murugesan).
0021-9517/$ – see front matter
doi:10.1016/j.jcat.2004.03.019
2004 Elsevier Inc. All rights reserved.

348
K. Shanmugapriya et al. / Journal of Catalysis 224 (2004) 347–357
the alkylating agent. In the present work we have investi-
gated this reaction using Al-MCM-41 (Si/Al = 55 and 104)
and Al,Zn-MCM-41 (Si/(Al + Zn) = 52) so that the influ-
ence of hydrophilic and hydrophobic properties and density
of acid sites of the catalysts on conversion and product se-
lectivity can be correlated.
was recorded on a Nicolet Avatar 360 FT-IR spectropho-
tometer equipped with a high-temperature vacuum chamber.
Approximately 10 mg of the sample was taken in the sam-
ple holder and dehydrated at 500 ◦C for 6 h under vacuum
(10−5 mbar). The sample was then cooled to room tempera-
ture. Then pyridine was adsorbed at room temperature. The
physically adsorbed pyridine was removed by heating the
sample at 150 ◦C under vacuum (10−5 mbar) for 30 min, the
removed material was cooled to room temperature, and the
spectrum was recorded. The acidity was calculated using the
extinction coefficients of the bands of Brønsted and Lewis
acid site adsorbed pyridine.
2. Experimental
2.1. Preparation of catalysts
All reagents, viz., sodium metasilicate, aluminum sulfate,
zinc sulfate, cetyltrimethylammoniumbromide (CTAB), sul-
furic acid, isopropyl acetate, and m-cresol were purchased
from Merck and used as such. Samples of Al-MCM-41 with
Si/Al = 55 and 104 and Al,Zn-MCM-41 with Si/(Al + Zn)
= 52 were synthesized hydrothermally using a gel com-
position of SiO2:x Al2O3:0.2 CTAB:0.89 H2SO4:120 H2O
(x varies with Si/Al and Si/(Al + Zn) ratios). In a typical
synthesis, 10.6 g of sodium metasilicate in water was com-
bined with an appropriate amount of aluminum sulfate in
distilled water, and the pH of the solution was adjusted to
10.5 using 1 M sulfuric acid with constant stirring to form
a gel. After 30 min, an aqueous solution of CTAB (struc-
ture directing agent) was added and the mixture was stirred
for 1 h at room temperature. The resulting suspension was
then autoclaved at 170 ◦C for about 12 h. After crystal-
lization, Al-MCM-41 material was recovered by filtration,
washed with distilled water, and dried at 80 ◦C for 8 h.
Al,Zn-MCM-41 with Si/(Al + Zn) = 52 was also synthe-
sized by following the same procedure using both aluminum
sulfate and zinc sulfate during initial synthesis. Mesoporous
Al-MCM-41 and Al,Zn-MCM-41 catalysts were obtained by
removing the occluded surfactant, which filled in the pores,
by calcining the samples at 550 ◦C in air for 6 h.
Surface area, pore volume, and pore size distribution of
the materials were measured by nitrogen adsorption at 77 K
with an ASAP-2010 porosimeter from Micromeritics Cor-
poration (Norcross, GA, USA). Before nitrogen adsorption–
desorption measurements, the samples was degassed at
623 K at 10−5 Torr and the specific area of the samples
was determined from the linear portion of the BET plots.
Pore size distributions were calculated from the desorption
branch of N2 adsorption–desorption isotherms using the
Barrett Joyner Halenda (BJH) algorithm (ASAP-2010 built-
in software from Micromeritics). Thermogravimetric analy-
sis (TGA) and differential thermal analysis (DTA) of the ma-
terials were performed simultaneously on a high-resolution
Mettler TA 3000 thermogravimetric analyzer. The samples
were heated in air at a heating rate of 20 ◦C/min in the tem-
perature range 25–700◦C.
29Si MAS-NMR spectra were recorded in a DRX-500FT-
NMR spectrometer at a frequency of 59.64 MHz, spinning
speed of 8 KHz, pulse length of 2.50 µs (45◦ pulse), delay
time of 10 s, and spectral width of 335 ppm. Two thousand
scans were acquired and processed with a line broaden-
ing of 50 Hz. The chemical shifts were reported with ref-
erence to trimethylsilylpropanesulfonic acid (TSP). Solid-
state 27Al MAS-NMR measurements were performed on a
Bruker MSL 400 spectrometer equipped with a magic-angle
spinning (MAS) unit. The 27Al MAS-NMR spectra were
recorded at a frequency of 104.22 MHz, a spinning rate of
8 KHz with a pulse length of 1.0 µs, a delay time of 0.2 s,
and a spectral width of 330 ppm. The total scans were 150
and the line broadening was 50 Hz. 27Al MAS-NMR chem-
ical shifts were reported in relation to the liquid solution of
aluminum nitrate.
2.2. Physicochemical characterization
The powder X-ray diffraction patterns of all the samples
were recorded with a Siemens D5005 Stereoscan diffrac-
tometer using nickel-filtered Cu-Kα radiation and a liquid
nitrogen-cooled germanium solid-state detector. The diffrac-
tograms were recorded in the 2θ range of 0.8◦–10◦ in steps
of 0.02◦ with a count time of 10 s at each point. Al and Zn
content of Al,Zn-MCM-41 and Al content of Al-MCM-41
were determined using inductively coupled plasma atomic
emission spectroscopy (ICP-AES) with the Labtum Plasma
8440 instrument.
Fourier transform infrared spectra of the materials were
recorded with a resolution of 2 cm−1 on a Nicolet (Avatar
360) instrument using the KBr pellet technique. About
15 mg of the sample was pressed (under a pressure of
2 tons/cm2) into a self-supported wafer of 13-mm diame-
ter. This pellet was used to record the infrared spectra in the
range 4000–400 cm−1. The acidity of the calcined materials
2.3. Catalytic performance: isopropylation of m-cresol
Isopropylation of m-cresol with isopropyl acetate was
carried out in a fixed-bed, vertical-flow-type reactor made
up of a borosil glass tube 40 cm in length and 2 cm in in-
ternal diameter. About 0.5 g of the catalyst was placed in
the middle of the reactor and supported on either side with
a thin layer of quartz wool and ceramic beads. The reac-
tor was heated to the requisite temperature with the help of a
tubular furnace controlled by a digital temperature controller
cum indicator. Reactants were fed into the reactor using a

K. Shanmugapriya et al. / Journal of Catalysis 224 (2004) 347–357
349
syringe infusion pump (SAGE instruments) that could be op-
erated at different flow rates. The reaction was carried out at
atmospheric pressure. The bottom of the reactor was con-
nected to a coiled condenser and a receiver to collect the
products. The products obtained in the first 10 min were dis-
carded, and the product collected after 1 h was analyzed for
identification. After each catalytic run, the catalyst was re-
generated by passing moisture- and carbon dioxide-free air
through the reactor for 6 h at 500 ◦C.
The percent conversion of m-cresol was analyzed in a
gas chromatograph (Shimadzu GC-17A) with a flame ion-
ization detector equipped with a 25-m capillary column
(crosslinked 5% phenylmethyl polysiloxane). The liquid
products were analyzed using a Perkin Elmer Auto Sys-
tem XL gas chromatograph with Turbo Mass Spectrometer
with helium as carrier gas (1 ml/h). As the percentage con-
version of m-cresol was higher with Al-MCM-41(55), all
other experiments concerning evaluation of the performance
of the catalyst were carried out only with Al-MCM-41(55).
(WHSV)−1 was calculated using the expression
synthesis conditions used, the crystallization reaction was
nonstoichiometric and a higher Si/Al ratio was noted in the
crystal [20].
Specific surface area, specific pore volume, and pore di-
ameter (BJH method) for calcined materials are presented
in Table 1. The isotherms of nitrogen adsorption for cal-
cined materials were measured at liquid nitrogen temper-
ature (77 K). The isotherms show well-defined stages and
they coincide with those already reported in the litera-
ture [21]. It is evident from these isotherms (Figs. 2a–2c) that
good mesoporous structural ordering and narrow pore size
distribution are well exhibited. The surface area of the cata-
lysts decreases in the order Al-MCM-41(104) > Al-MCM-
41(55) > Al,Zn-MCM-41(52). Similarly pore volume de-
creases in the same order due to the presence of textural
mesoporosity [22].
FT-IR spectra of the as-synthesized and calcined Al-
MCM-41 (Si/Al = 55 and 104) and Al,Zn-MCM-41 (Si/
(Al + Zn) = 52) molecular sieves are presented in Fig. 3.
The broad envelope around 3500 cm−1 is due to O–H
stretching of water, surface hydroxyl groups, and bridged
hydroxyl groups. There are less intense peaks in the spectra
of the as-synthesized samples just below 3000 cm−1 which
are assigned to symmetric and asymmetric stretching modes
of the –CH2 group of the locked-in template. Their cor-
responding bending mode is observed at 1400 cm−1. The
peaks between 500 and 1200 cm−1 are assigned to frame-
work vibrations. The intense peak at 1123 cm−1 is attributed
to the asymmetric stretching of T–O–T groups. The symmet-
ric stretching modes of T–O–T groups are observed around
800 cm−1, and the peak at 460 cm−1 is due to the bend-
ing mode of T–O–T. The peak at 963 cm−1 is assigned to
the presence of defective Si–OH groups. The symmetric and
asymmetric stretching modes of the –CH2 group of the tem-
plate are absent in the spectra of calcined samples. These
spectral features resemble those reported by previous work-
ers [23,24].
W
F
weight of catalyst
weight of flow rate
(WHSV)−1
=
=
.
Flow rate of the reactant is varied to maintain constant
(WHSV)−1 during the reaction. Selectivity to products is
defined as weight of each product obtained divided by
the weight of starting materials consumed, normalized to
100%. Conversion is the weight of starting material con-
sumed divided by weight of starting material normalized
to 100%.
3. Results and discussion
3.1. Characterization
XRD powder diffraction patterns of as-synthesized and
calcined mesoporous Al-MCM-41 (Si/Al = 55 and 104) and
Al,Zn-MCM-41 (Si/(Al + Zn) = 52) molecular sieves are
shown in Fig. 1. The XRD patterns, taken before and af-
ter calcination, confirmed the hexagonal MCM-41 phase in
the sample. After calcination, the intense peak was shifted
to higher 2θ values due to pore size contraction. As dis-
played in the figure, as-synthesized and calcined samples
exhibit an intense signal at about 1.8◦ due to [100] plane and
weak broad signals between 2.5◦ and 4◦ (2θ) due to [110],
[200], and [210] planes. These peaks confirm the hexagonal
mesophase of the material. The d100 spacing and lattice para-
meter (a0) calculated as per the literature procedure are sum-
marized in Table 1. These XRD patterns coincide with the
data already reported in the literature for mesoporous alumi-
nosilicate molecular sieves [18,19]. From ICP-AES analy-
sis the aluminum and zinc content of Al,Zn-MCM-41(52)
were found to be 0.053 and 0.057 g, respectively. The alu-
minum content of Al-MCM-41(55) and Al-MCM-41(104)
were found to be 0.097 and 0.051 g, respectively. Under the
The results of TGA of Al-MCM-41 (Si/Al = 55 and
104) and Al,Zn-MCM-41 (Si/(Al + Zn) = 52) are illus-
trated in Fig. 4. DTA traces are shown in the same figure.
In all three thermograms, there is an initial weight loss of
about 5% below 100 ◦C due to desorption of water. The
second weight loss due to decomposition and desorption
of template occurs between 180 and 300 ◦C. The percent-
age weight loss is found to 29.7, 41.0, and 35.1% for Al-
MCM-41(104), Al-MCM-41(55), and Al,Zn-MCM-41(52),
respectively. This clearly demonstrates that a larger amount
of aluminum is planted on the channel surface in Al-MCM-
41(55) than in Al-MCM-41(104). The weight loss for Al,Zn-
MCM-41(52) should be high compared with that Al-MCM-
41(55), but the value is smaller. Such a smaller value was
also observed for Cu-MCM-41 compared with Si-MCM-41
by Wang et al. [25]. If Zn2+ ions are entirely buried inside
the framework, the weight loss should be nearly equal to Al-
MCM-41(104). But the percentage weight loss is high and
hence zinc is not entirely buried inside the mesopore walls.

350
K. Shanmugapriya et al. / Journal of Catalysis 224 (2004) 347–357
Fig. 1. XRD patterns of as-synthesized and calcined mesoporous Al-MCM-41 molecular sieves: (a) Si/Al = 104, (b) Si/Al = 55, (c) Si/(Al + Zn) = 52.
Table 1
Physical properties of the catalysts
a
27
Catalyst
Si/Al
Gel ratio
50
50
100
d
a
BET surface area
Pore volume
Pore diameter
(nm)
Wall thickness
(Å)
Al MAS-NMR
b
100
0
2
3
(Å)
(Å)
(m /g)
(cm /g)
ICP
55
52
104
Al /Al
T
O
1
2
3
41.7
37.5
40.6
48.2
43.3
46.9
979
930
1044
0.68
0.66
0.66
2.79
2.57
2.83
20.3
17.6
18.6
4.1
7.4
∞
c
c
a
(1) Al-MCM-41(55), (2) Al,Zn-MCM-41(52), (3) Al-MCM-41(104).
Al MAS-NMR spectra: ratio of the intensities of 4- and 6-coordinated aluminum in the calcined catalysts.
Si/(Al + Zn) ratio.
b
c
27
29Si MAS-NMR spectra of the calcined Al-MCM-41
atmosphere. The absence of octahedral Al3+ in Al-MCM-
41(104) and the presence of the same in Al-MCM-41(55)
are attributed to the high aluminum content of the latter cat-
alyst. But Al,Zn-MCM-41(52) has nearly same Si/Al ratio
as Al-MCM-41(104). Hence the peak due to octahedral alu-
minum in the former cannot be explained on the basis of
aluminum content. Reddy et al. [29] have reported prepara-
tion of Al-MCM-41 (Si/Al = 104 and 50) materials with-
out any nonframework aluminum using aluminum sulfate as
the source. But they reported calcination of the materials
in nitrogen prior to air. Hence calcination of the materials
exclusively in air and zinc substitution may cause the frame-
work leaching in Al,Zn-MCM-41(52). Chakraborty et al.
[30] reported that water-coordinated framework aluminum
could also produce the peak at 0 ppm. Van Hooff and co-
workers [31] also reported synthesis and characterization
of Al-MCM-41, wherein incorporation of an excess of alu-
minum formed an impure crystal-phase tridimite, and the
Lewis acid site prevailed because of the octahedral non-
framework aluminum, accompanied by collapse of the struc-
ture. As Al-MCM-41(104) contains less aluminum content,
the nonframework species is not observed. Al-MCM-41(55)
contains more aluminum than Al-MCM-41(104) and Al,Zn-
materials are shown in Fig. 5. The partly resolved sig-
nal at −92 ppm is assigned to Q2 species. This signal
is less intense for Al-MCM-41(104) and more intense for
Al-MCM-41(55) and Al,Zn-MCM-41(52). The other signals
at −101 (Q3) and −106, −110 (Q4) appear similar to those
reported in the literature [26–28]. 27Al MAS-NMR spec-
tra of the samples are shown in Fig. 6, and the ratio of
the intensities of 4- and 6-coordinated aluminum was cal-
culated and is given in Table 1. The peak at 53 ppm in all
spectra is assigned to tetrahedrally coordinated framework
aluminum, and the peak at 0 ppm is assigned to octahe-
dral nonframework aluminum species [24,25]. Al-MCM-
41(104) is devoid of any nonframework Al3+ (Al2O3) as
there is no signal at 0 ppm. But Al-MCM-41(55) and Al,Zn-
MCM-41(52) have such species as there is a broad signal
at 0 ppm. Generally, materials with high aluminum content
are susceptible to framework leaching of aluminum during
calcination [28]. In the present study, calcination of the as-
synthesized materials was done in air directly without using
nitrogen in the beginning. This would also aid framework
leaching. Even for those materials of high aluminum con-
tent this problem can be avoided by calcining in a nitrogen

K. Shanmugapriya et al. / Journal of Catalysis 224 (2004) 347–357
351
Fig. 2. Adsorption isotherms of Al-MCM-41 molecular sieves: (a) Si/Al = 104, (b) Si/Al = 55, (c) Si/(Al + Zn) = 52.

352
K. Shanmugapriya et al. / Journal of Catalysis 224 (2004) 347–357
Fig. 3. FTIR spectra of as-synthesized and calcined mesoporous Al-MCM-41 molecular sieves: (a) Si/Al = 104, (b) Si/Al = 55, (c) Si/(Al + Zn) = 52.
Fig. 4. TGA/DTA analysis of Al-MCM-41 molecular sieves: (a) Si/Al
= 104, (b) Si/Al = 55, (c) Si/(Al + Zn) = 52.
29
Fig. 5.
Si MAS-NMR spectra of Al-MCM-41 molecular sieves:
(a) Si/Al = 104, (b) Si/Al = 55, (c) Si/(Al + Zn) = 52.
MCM-41(52), and produces a peak due to the octahedral alu-
minum it contains. The less intense peak at 0 ppm in Al,Zn-
MCM-41(52) spectra is due to water-coordinated framework
aluminum [30].
The acidity of the calcined materials was measured by
FT-IR spectroscopy using pyridine as probe (Fig. 7). The
samples give the expected bands due to Lewis acid-bound
(1450, 1575, and 1623 cm−1), Brønsted acid-bound (1545
and 1640 cm−1) and both Lewis and Brønsted acid-bound
(1490 cm−1) pyridine. These data coincide with those re-
ported by Climent et al. [32]. The acidity of the catalysts
was calculated using the extinction coefficients of the bands
of Brønsted and Lewis acid site adsorbed pyridine [33] and
the results are summarized in Table 2. The data indicate the
presence of both Brønsted and Lewis acid sites in Al-MCM-
41(55) and Al,Zn-MCM-41(52). These data also coincide

K. Shanmugapriya et al. / Journal of Catalysis 224 (2004) 347–357
353
Table 2
Brønsted and Lewis acidity values for mesoporous molecular sieves
Catalyst
423 K
a
a
B.A.
L.A.
Al-MCM-41(55)
Al,Zn-MCM-41(52)
Al-MCM-41(104)
7.1
8.2
3.7
9.0
9.5
1.1
a
Acidity (µmol py/g catalyst).
with the 27Al-MAS-NMR spectra of the catalysts, which
indicate the presence of both octahedral and tetrahedral alu-
minum in Al-MCM-41(55) and Al,Zn-MCM-41(52), but
only octahedral aluminum in Al-MCM-41(104). Although
the signal due to octahedral aluminum is not observed for
Al-MCM-41(104) in 27Al-MAS-NMR (Fig. 6), a slightly
smaller amount of Lewis acidity (Table 2) is also observed
as reported by Umamaheswari et al. [3].
3.2. Isopropylation of m-cresol
Isopropylation of m-cresol with isopropyl acetate was
investigated over Al-MCM-41(55), Al-MCM-41(104), and
Al,Zn-MCM-41(52) in the vapor phase. The effects of re-
action temperature, feed ratio, WHSV, and time-on-stream
were studied to optimize the parameters for better conver-
sion and selectivity of products.
27
Fig. 6.
Al MAS-NMR spectra of Al-MCM-41: (a) Si/Al = 104,
(b) Si/Al = 55, (c) Si/(Al + Zn) = 52.
3.3. Effect of temperature
The influence of reaction temperature on conversion and
selectivity was studied at 200, 250, 300, 350, and 400 ◦C.
The reactant feed ratio (m-cresol:isopropyl acetate) was
maintained at 1:3 and WHSV 1.52 h−1. The products are
thymol, isothymol, 2-I-5-MPA, and I-3-MPE (Scheme 1).
m-Cresol conversion and selectivity of the products at var-
ious temperatures are summarized in Table 3. An increase
in conversion is observed between 200 and 300 ◦C. The per-
centage conversion decreases significantly above 300 ◦C for
all catalysts.
Thymol is formed more selectively than other products
at all temperatures studied. The high selectivity of thymol
is due to chemisorption of m-cresol on the Brønsted acid
sites through its –OH group, favoring only ortho carbons for
reaction with isopropyl cation. As the second and fourth car-
bons have steric hindrance due to a methyl group, isopropyl
cation selectively attacks the sixth carbon to yield thymol.
Formation of thymol requires chemisorption of m-cresol, but
for higher m-cresol conversion greater chemisorption of iso-
propyl acetate is required. The greater than 50% m-cresol
conversion at all temperatures studied reveals the greater
adsorption of isopropyl acetate. Selectivity to isothymol is
higher at lower temperatures than at higher temperatures.
The formation of isothymol is due to the different configura-
tion of the aromatic ring on the catalyst surface as proposed
by Taylor and Ludlum [34]. Isothymol could also be formed
via isomerization of thymol over strong acid sites. Thus,
Fig. 7. Brønsted and Lewis acidity of Al-MCM-41: (a) Si/Al = 104,
(b) Si/Al = 55, (c) Si/(Al + Zn) = 52.

354
K. Shanmugapriya et al. / Journal of Catalysis 224 (2004) 347–357
Scheme 1. Reaction of m-cresol with isopropyl acetate.
Table 3
a
Effect of temperature on m-cresol conversion and product selectivity
Catalyst
Temperature
m-Cresol conversion
Selectivity of product (%)
◦
( C)
(%)
Thymol
Isothymol
2-I-5-MPA
I-3-MPE
Al-MCM-41(55)
200
250
300
350
400
200
250
300
350
400
200
250
300
350
400
68.6
75.1
80.9
70.3
60.2
60.1
65.9
70.1
58.6
50.0
53.7
56.7
62.9
57.9
49.3
71.7
76.1
83.7
87.9
91.6
63.9
67.6
70.2
76.9
82.4
57.6
61.2
63.2
67.9
71.9
9.5
8.5
4.5
3.2
2.0
9.8
7.6
6.6
5.7
4.8
13.3
9.7
9.0
8.3
7.8
3.6
4.2
7.9
5.4
3.4
10.7
11.9
14.3
9.2
15.2
11.2
3.9
3.5
3.0
15.6
12.9
8.9
8.2
6.9
16.4
14.9
10.9
10.6
8.4
Al,Zn-MCM-41(52)
Al-MCM-41(104)
5.9
12.7
14.2
16.9
13.2
11.9
a
−1
m-Cresol:isopropyl acetate (feed ratio) = 1:3. WHSV = 1.52 h
.
there are two routes by which isothymol could be formed in
this reaction. Isomerization of thymol to isothymol is greater
at low temperature (≈200 ◦C) than at 400 ◦C as reported by
earlier workers [10]. Although isomerization of thymol can
occur on the same acid sites that catalyze the formation of
thymol, the selectivity to isothymol is low. This suggests that
the sites are not acidic enough for isomerization.
Selectivity to 2-I-5-MPA increases up to 300 ◦C and then
decreases. Sakthivel et al. [35] reported the distribution of
weak, moderate, and strong acid sites for Al-MCM-41. The
weak acid sites were assigned to surface hydroxyls, and
the moderate and strong acid sites, to briged hydroxyls.
O-Acylation was observed in this study, not C-acylation.
O-Acylation is generally considered to occur on defective
sites which are less effective for adsorption of m-cresol com-
pared with bridged hydroxyl groups. These defective hy-
droxyl groups can adsorb ester by increasing the polarization
of the carbonyl group by weak protonation. The polarization
of ester is not sufficient to yield isopropyl cation, but suffi-
cient enough for nucleophilic attack by m-cresol diffusing
close to it. To verify that O-acylation of m-cresol occurs
mostly on defective sites of the catalysts, the reaction was
studied with Si-MCM-41, which has only defective acid
sites. At 300 ◦C with a feed ratio of 1:3 and a flow rate of
1.5 ml/h, the conversion of m-cresol is 47.9%, with the for-
mation of only 2-I-5-MPA thus supporting our view.
To establish that O-acylation is due to partly proto-
nated ester and not acetic acid, which is a by-product
formed during the reaction, the study was repeated with
acetic acid as one of the reactants with the feed ratio
1:3:1 (m-cresol:isopropyl acetate:acetic acid) under simi-
lar conditions. The selectivity to 2-I-5-MPA did not vary
significantly, thus confirming that it is only the ester that
gives O-acylation. The reaction was also studied with
acetic acid alone as acylating agent with the feed ratio 1:1
(m-cresol:acetic acid) under the same experimental condi-
tions. Selectivity to O-acylation was only 15.8%, suggesting

K. Shanmugapriya et al. / Journal of Catalysis 224 (2004) 347–357
355
that acetic acid is a weak acylating agent as reported in the
literature [36].
selectivity to thymol is higher with all the catalysts, but the
selectivity to other products is low with Al-MCM-41(55)
compared with the other two catalysts. Based on these ob-
servations Al-MCM-41(55) is selected as a suitable catalyst
for the reaction of m-cresol with isopropyl acetate in the va-
por phase.
Selectivity to I-3-MPE increases at lower temperatures
and gradually decreases with increase in temperature. At
200 ◦C, the maximum selectivity was found to be only 15%.
This again confirms the chemisorbed state of m-cresol on
the catalyst surface at the 1:3 feed ratio. The more appro-
priate route for the formation of I-3-MPE is the reaction
between m-cresol in the vapor phase and isopropyl cation
on the catalyst surface. The possibility of conversion of I-3-
MPE to thymol is also ruled out as its selectivity is less, even
at 200 ◦C.
3.4. Effect of feed ratio
The effect of feed ratio (m-cresol:isopropyl acetate) on
m-cresol conversion and product selectivity was investigated
over Al-MCM-41(55) with the feed ratios 1:1, 1:2, 1:3,
and 1:4 at 300 ◦C and the results are presented in Table 4.
m-Cresol conversion is found to be less than 30% at a 1:1
feed ratio. This reveals that at low feed ratio only m-cresol
is more highly adsorbed on the acid sites of the catalyst than
isopropyl acetate. The conversion increases with an increase
in isopropyl acetate content in the feed and reaches a maxi-
mum at a 1:3 feed ratio. The smaller selectivity to thymol at
the 1:1 feed ratio compared with other feed ratios is due to
isomerization of thymol, as more protonic sites are available
for chemisorption. Isothymol selectivity gradually decreases
with the increase in isopropyl acetate content in the feed.
The effect of feed ratio on conversion and product selec-
tivity over Al-MCM-41(55) was studied by increasing the
m-cresol content of the feed. The ratio was varied from 2:1
to 4:1 to 6:1 and the temperature was maintained at 300 ◦C
with a feed rate of 1 ml/h (Table 4). Conversion was almost
the same for the 1:1 and 2:1 ratios, but decreased for the 4:1
and 6:1 feed ratios. The nearly identical levels of conversion
for the 1:1 and 2:1 feed ratios illustrates the availability of
an adequate amount of isopropyl cation. Hence at feed ra-
tios 1:1 and 2:1, the adsorption of isopropyl acetate might
not be reduced by m-cresol as it is more prone to chemisorp-
tion on the catalyst surface than m-cresol. But at 4:1 and
6:1, the decrease might be due to preferential adsorption
of m-cresol over ester. The selectivity to thymol decreases,
whereas the selectivity to isothymol increases, with an in-
crease in m-cresol content in the feed.
The activity of the catalysts follows the order Al-MCM-
41(55) > Al,Zn-MCM-41(52) > Al-MCM-41(104). But
their acidity follows the order Al,Zn-MCM-41(52) > Al-
MCM-41(55) > Al-MCM-41(104) as each zinc site plants
two protons on the catalyst surface, in comparison to only
one proton for each aluminum site. The collapse of this acid-
ity order toward activity indirectly discloses the fact that
acidity is not the only factor controlling the activity of the
catalysts. In the previous report on alkylation of m-cresol
with isopropyl alcohol, Umamaheswari et al. [3] consid-
ered the hydrophilic and hydrophobicproperties of the chan-
nel surface of Al-MCM-41 molecular sieves to account for
the nearly similar conversions of m-cresol over Al-MCM-
41(55) and Al-MCM-41(104) catalysts. Similarly, Climent
et al. [32] also considered these properties to account for the
activity of the catalysts in the acetalization reaction. Based
on these reports it is suggested that Al,Zn-MCM-41(52) is
more hydrophilic than Al-MCM-41(55) due to close posi-
tioning of Brønsted acid sites. Hence, isopropyl acetate with
its hydrophobic properties can diffuse through the pores of
Al,Zn-MCM-41(52) away from the inner surface in the ab-
sence of any external force. This accounts for the low con-
version over Al,Zn-MCM-41(52) compared with Al-MCM-
41(55). In a similar way, one can expect higher conversion
over Al-MCM-41(104) as it is more hydrophobic than the
others. But the conversion is less, which is attributed to the
lower density of the acid sites in the catalyst. Even if iso-
propyl acetate can diffuse through the pores close to the
inner surface, m-cresol can also compete significantly for
chemisorption on the Brønsted acid sites of the catalyst.
As m-cresol conversion increases from 200 to 300 ◦C,
there might be an increase in the chemisorption of isopropyl
acetate on the Brønsted acid sites with increase in temper-
ature. The low m-cresol conversion at 200 ◦C indicates less
adsorption of isopropyl acetate than m-cresol. The slow de-
crease in conversion with increase in temperature could be
attributed to the scattered distribution of acid sites of the cat-
alyst that avoids multialkylation leading to coke formation.
However, the decrease in conversion above 300 ◦C is due to
blocking of active sites by coke. The catalyst becomes black
above 300 ◦C. But compared with zeolites, the large pore
diameter of the catalyst under investigation facilitates rapid
diffusion of the reactants and products and thereby main-
tains higher conversion. From Table 3 it can be inferred that
For the feed ratios 1:1, 1:2, 1:3, and 1:4, the selectivity
to thymol increases but that of isothymol decreases. But for
feed ratios 2:1, 4:1, and 6:1 the reverse trend of a decrease in
selectivity to thymol and increase in selectivity to isothymol
Table 4
a
Effect of feed ratio on m-cresol conversion and product selectivity
Selectivity of product (%)
Feed m-Cresol
ratio
conversion (%)
Thymol Isothymol 2-I-5-MPA I-3-MPE
1:1
1:2
1:3
1:4
2:1
4:1
6:1
28.9
56.8
80.9
73.2
27.4
12.9
11.9
60.2
64.5
68.7
70.9
67.3
58.5
45.8
21.6
18.7
14.5
9.4
16.1
28.6
36.2
18.2
16.0
12.9
8.7
7.1
–
–
0.8
3.9
11.0
9.5
12.9
18.0
–
a
◦
Al-MCM-41(55). Temperature = 300 C. Feed ratio = m-cresol:iso-
propyl acetate.

356
K. Shanmugapriya et al. / Journal of Catalysis 224 (2004) 347–357
is observed. It is suggested that the formation of isothymol
is due to chemisorption of m-cresol on the strong acid sites
of the catalysts, which leads to ring deactivation, favoring
direct meta substitution with respect to the –OH group. Such
ring deactivation that facilitates meta-substitution has been
reported previously [37].
also observed for 2-I-5-MPA and I-3-MPE. As m-cresol con-
version and selectivity to thymol are optimum at 1.52 h−1
the time-on-stream study was carried out at this WHSV.
,
3.6. Effect of time-on-stream
The selectivity to 2-I-5-MPA decreases when the feed ra-
tio changes from 1:1 to 2:1 and disappears at feed ratios 4:1
and 6:1, as its formation requires two consecutive processes
of alkylation and acylation. The prevention by the high
m-cresol content in the feed of two successive processes for
a single molecule in different regions of acidity is not easily
envisaged. The selectivity to I-3-MPE increases with an in-
crease in the isopropyl acetate content of the feed. The same
trend is also observed for feed ratios 2:1, 4:1, and 6:1, for
which isopropyl acetate content decreases. The formation of
ether requires reaction between chemisorbed isopropyl ac-
etate and free m-cresol in the vapor phase. When the feed
ratio is changed from 1:1 to 1:2, 1:3, and 1:4, there could
be preferential adsorption of isopropyl acetate, thus leaving
more m-cresol in the vapor phase which aids the formation
of ether. The increase in ether selectivity with feed ratios 2:1,
4:1, and 6:1 illustrates that even with these feed ratios, the
preferential adsorption of isopropyl acetate is not avoided
by the high content of m-cresol, thus giving more selectiv-
ity to ether. Hence this study clearly demonstrates that at the
feed ratio of 1:3, isopropyl acetate has a greater tendency to
be chemisorbed on the catalyst surface than does m-cresol,
and therefore it could be better exploited for both alkylation
and acylation.
The effect of time-on-stream was studied for 5 h over
Al-MCM-41(55) at 300 ◦C with a feed ratio of 1:3 and at
WHSV 1.52 h−1 (Fig. 8a). There is a rapid decrease in con-
version in the first 2 h, which is followed by a slow decrease
thereafter. In the beginning of the stream all the sites are
free, and hence multialkylation of m-cresol leading to coke
formation is rapid. But after 2 h on stream, a smaller num-
ber of sites are available for activity, and further blocking of
the active sites is reduced. Hence there is a slow decrease
in conversion. The selectivity to thymol increases with in-
crease in the stream, and the reverse trend in the selectivity
to isothymol is observed. At the end of the fourth hour of
stream, 2-I-5-MPA and I-3-MPE are not observed. The time-
on-stream study also illustrates disappearance of acylated
products after 3 h. Acetic acid is a by-product in the use
of isopropyl acetate as alkylating agent. Because the present
3.5. Effect of WHSV
The effect of WHSV on conversion and product selec-
tivity was studied at 1.01, 1.52, and 2.02 h−1 with the feed
ratio 1:3 at 300 ◦C, and the results are summarized in Ta-
ble 5. The conversion decreases with increase in WHSV.
However, the decrease is only about 10% when the WHSV
is increased from 1.01 to 1.52 h−1, whereas the decrease is
more than 40% at 2.02 h−1. The drastic decrease in conver-
sion at 2.02 h−1, is due to fast diffusion of m-cresol, which is
largely retained in the vapor phase by isopropyl acetate. The
(a)
higher selectivity to thymol was attained at WHSV 2.02 h−1
As isothymol is formed by isomerization of thymol, which is
a time-dependent process, its selectivity is high at 1.01 h−1
.
,
and then decreases gradually with the increase in WHSV.
A similar decrease in selectivity with increase in WHSV is
Table 5
a
Effect of WHSV on m-cresol conversion and product selectivity
WHSV m-Cresol
−1
Selectivity of product (%)
conversion (%)
(b)
(h
)
Thymol Isothymol 2-I-5-MPA I-3-MPE
Fig. 8. (a) Study of time-on-stream over Al-MCM-41(55). Feed ratio =
1.01
1.52
2.02
89.7
80.9
40.2
60.9
68.7
80.5
18.4
14.5
7.7
14.4
12.9
9.3
6.3
3.9
2.5
◦
1:3 (m-cresol:isopropyl acetate), temperature = 300 C, and WHSV =
−1
1.52 h . (b) Study of time-on-stream over Al-MCM-41(55). Feed ratio
◦
= 1:3 (m-cresol:isopropyl alcohol), temperature = 300 C, and WHSV =
a
◦
−1
Al-MCM-41(55). Temperature = 300 C.
1.52 h
.

K. Shanmugapriya et al. / Journal of Catalysis 224 (2004) 347–357
357
investigation involves the use of mesoporous materials, the
acetic acid formed in the vapor state easily diffuses out, and
hence its tendency to promote coke formation is not as great
as that of microporous zeolites.
Department of Science and Technology (DST), Government
of India, New Delhi.
To compare the activity of isopropyl acetate and isopropyl
alcohol in the isopropylation of m-cresol, the reaction was
also studied with the latter at 300 ◦C with a feed ratio of
1:3 and a flow rate of 1.5 ml/h. m-Cresol conversion is
found to be 51% at 1 h, which is about 29% less than that
of isopropyl acetate. This observation proves that isopropyl
acetate is a better alkylating agent than isopropyl alcohol.
As discussed previously the ease of adsorption at the steric
free carbonyl group of the ester might be the reason for
the enhanced conversion. The effect of time-on-stream with
isopropyl alcohol was also studied, and the results are illus-
trated in Fig. 8b. Conversion decreases with stream in the
first 3 h and remains almost constant thereafter, whereas the
time-on-stream study with isopropyl acetate indicates a rapid
decrease in conversion in the first 3 h on stream, followed by
a slow decrease afterward. Although the conversion is better
with isopropyl acetate than isopropyl alcohol, the selectiv-
ity to thymol is better with isopropyl alcohol than isopropyl
acetate. At the end of 3 h on stream, 100% selectivity to
thymol is obtained with 25% m-cresol conversion with iso-
propyl alcohol. But isopropyl acetate gives 85.9% selectivity
to thymol with 51.3% m-cresol conversion. Though there is
a marginally higher selectivity to thymol with isopropyl al-
cohol, the conversion of m-cresol is more than doubled with
isopropyl acetate. The major component that increases thy-
mol selectivity with isopropyl alcohol is due to the formation
of by-product water, which may deactivate the Brønsted acid
sites of the catalysts, thus preventing isomerization of thy-
mol to isothymol.
References
[1] M. Nitta, Bull. Chem. Soc. Jpn. 47 (1974) 2360.
[2] R. Stroh, R. Seydel, W. Hahn, in: W. Forest (Ed.), Newer Methods
of Preparative Organic Chemistry, vol. 2, Academic Press, New York,
1963, p. 337.
[3] V. Umamaheswari, M. Palanichamy, V. Murugesan, J. Catal. 210
(2002) 367.
[4] J.C. Leffingwell, R.E. Shackelford, Cosmet. Perfumery 89 (6) (1974)
69.
[5] R. Hopp, Recent Adv. Tobacco Sci. 19 (1993) 46.
[6] P.L. Teissedre, A.L. Waterhouse, J. Agric. Food Chem. 48 (2000)
3801.
[7] E.L. Krause, W. Ternes, Eur. Food Res. Technol. 209 (1999) 140.
[8] N.V. Yanishlieva, E.M. Marinova, M.H. Gordon, V.G. Raneva, Food
Chem. 64 (1999) 59.
[9] M. Milos, J. Mastelic, I. Jerkovic, Food Chem. 71 (2000) 79.
[10] H. Grabowska, J. Wrzyszcz, Res. Chem. Intermed. 27 (2001) 281.
[11] S. Velu, S. Sivasanker, Res. Chem. Intermed. 24 (1998) 657.
[12] T. Yamanaka, Bull. Chem. Soc. Jpn. 49 (1976) 2669.
[13] H.G. Franc, J.W. Stadelhofer, in: Industrial Aromatic Chemistry,
Springer-Verlag, Berlin/Heidelberg, 1998, p. 168.
[14] W. Biedarmann, H. Koller, K. Wedemeyer, US patent 4086283 (1978).
[15] P. Wimmer, H.J. Buysch, L. Puppe, US patent 5030770 (1991).
[16] A. Klein, K. Wedemeyer, Bayer DE-OS 2242628 (1972).
[17] V. Umamaheswari, M. Palanichamy, Arabindoo Banumathi, V. Mu-
rugesan, Indian J. Chem. A 39 (2000) 1241.
[18] J.S. Beck, J.C. Vartuli, W.J. Roth, M.E. Lernowicz, C.T. Kresge, K.D.
Schmitt, C.T.W. Chu, D.H. Olson, E.W. Sheppard, S.B. McCullen, J.B.
Higgins, J.C. Schlenker, J. Am. Chem. Soc. 121 (1992) 10834.
[19] J.H. Kin, M. Tanabe, M. Niwa, Micropor. Mater. 10 (1997) 85.
[20] M.L. Ocelli, S. Biz, A. Auroux, G.J. Ray, Micropor. Mesopor.
Mater. 26 (1998) 193.
[21] S.J. Greggand, K.S.W. Sing, in: Adsorption, Surface Area and Poros-
ity, 2nd edition, Academic Press, New York, 1982.
[22] T.R. Pauly, Y. Liu, T.J. Pinnavaia, S.J.L. Billinge, T.P. Rieler, J. Am.
Chem. Soc. 121 (1992) 8835.
4. Conclusion
[23] C.Y. Chen, H.X. Li, M.E. Davis, Micropor. Mater. 2 (1993) 17.
[24] S. Biz, M.L. Occelli, Catal. Rev. Sci. Eng. 40 (1998) 329.
[25] L. Wang, S. Velu, S. Tomura, F. Ohashi, K. Suzuki, M. Okazaki, T. Os-
aki, M. Maeda, J. Mater. Sci. 37 (2002) 801.
[26] A. Corma, Micropor. Mesopor. Mater. 4 (1997) 249.
[27] Y. Sun, Y. Yue, Z. Gao, Appl. Catal. A 161 (1997) 121.
[28] A. Matsumoto, H. Chen, K. Tsutsumi, M. Grun, K. Unger, Micropor.
Mesopor. Mater. 32 (1999) 55.
[29] K.M. Reddy, C. Song, Catal. Today 31 (1996) 137.
[30] B. Chakraborty, B. Viswanathan, Catal. Today 49 (1999) 253.
[31] M. Busio, J. Janchen, J.H.C. Van Hooff, Micropor. Mater. 37 (2002)
801.
[32] M.J. Climent, A. Corma, S. Iborra, S. Miquel, J. Primo, F. Rey,
J. Catal. 183 (1999) 76.
The reaction results allow us to conclude that conversion
and selectivity to thymol are greater at higher temperatures.
Moderate and strong acid sites promote alkylation, while
weak acid sites and silanol defects favor O-acylation. Al-
though acetic acid is one of the by-products, it is not as
reactive as ester for acylation. The reaction depends not
only on the acid sites of the catalysts; their hydrophilic and
hydrophobic properties also play an important role. Al,Zn-
MCM-41(52), with a greater density of acid sites than Al-
MCM-41(55), is less active due to its more hydrophilic na-
ture. The study reveals that isopropyl acetate can be used as a
convenient alkylating agent in the isopropylation of m-cresol
in the vapor phase over Al-MCM-41 molecular sieves.
[33] C.A. Emies, J. Catal. 141 (1993) 347.
[34] D.R. Taylor, K.H. Ludlum, J. Phys. Chem. 76 (1972) 2882.
[35] A. Sakthivel, S.K. Badamali, P. Selvam, Micropor. Mesopor. Mater. 39
(2000) 457.
[36] P. Botella, A. Corma, J.M. Lopez-Nieto, S. Valencia, R. Jacquot,
J. Catal. 195 (2000) 79.
Acknowledgment
The authors express their sincere thanks for the generous
financial support in the form of a project sponsored by the
[37] B. Rajesh, M. Palanichamy, V. Kazansky, V. Murugesan, J. Mol. Catal.
A 187 (2002) 259.