Trifluoromethanesulfonic acid immobilized on zirconium oxide obtained by the sol-gel method as catalyst in paraben synthesis

Article abstract of DOI:10.1016/j.apcata.2011.04.018

The parabens, alkylic esters of p-hydroxybenzoic acid, were synthesized using trifluoromethanesulfonic acid immobilized on zirconium oxide (zirconia) as catalyst. The oxide was obtained by the sol-gel method, using urea as a pore-forming agent. After removing urea by extraction with water, the solid was dried and then calcined at 100, 205, 310 and 425 °C for 24 h. Afterward, it was impregnated with trifluoromethanesulfonic acid in toluene at reflux and leached to remove the weakly adsorbed acid. Mesoporous materials were obtained, whose mean pore diameter increased with the temperature of the thermal treatment of the support, while the specific surface area and the amount of acid bonded to the support decreased. The samples are crystalline from 400 °C and are thermally stable up to 250 °C. The catalysts have strong acidity and the number of acid sites decreased with the acid content in the support. The catalytic activity in the synthesis of propyl paraben, expressed as moles of ester formed at 5 h/mol acid in the catalyst, decreased when samples obtained from supports thermally treated at higher temperature were used. The activity also slightly diminished for the synthesis of different parabens in the order propyl > ethyl > methyl ester.

Full text of DOI:10.1016/j.apcata.2011.04.018

Applied Catalysis A: General 400 (2011) 91–98
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Trifluoromethanesulfonic acid immobilized on zirconium oxide obtained by the
sol–gel method as catalyst in paraben synthesis
Marina Gorsd, Luis Pizzio, Mirta Blanco∗
Centro de Investigación y Desarrollo en Ciencias Aplicadas “Dr. J.J. Ronco” (CINDECA), Departamento de Química, Facultad de Ciencias Exactas, UNLP-CCT La Plata, CONICET, 47 No
257, 1900 La Plata, Argentina
a r t i c l e i n f o
a b s t r a c t
Article history:
The parabens, alkylic esters of p-hydroxybenzoic acid, were synthesized using trifluoromethanesulfonic
acid immobilized on zirconium oxide (zirconia) as catalyst. The oxide was obtained by the sol–gel method,
using urea as a pore-forming agent. After removing urea by extraction with water, the solid was dried and
then calcined at 100, 205, 310 and 425 ◦C for 24 h. Afterward, it was impregnated with trifluoromethane-
sulfonic acid in toluene at reflux and leached to remove the weakly adsorbed acid. Mesoporous materials
were obtained, whose mean pore diameter increased with the temperature of the thermal treatment of
the support, while the specific surface area and the amount of acid bonded to the support decreased. The
samples are crystalline from 400 ◦C and are thermally stable up to 250 ◦C. The catalysts have strong acidity
and the number of acid sites decreased with the acid content in the support. The catalytic activity in the
synthesis of propyl paraben, expressed as moles of ester formed at 5 h/mol acid in the catalyst, decreased
when samples obtained from supports thermally treated at higher temperature were used. The activity
also slightly diminished for the synthesis of different parabens in the order propyl > ethyl > methyl ester.
© 2011 Elsevier B.V. All rights reserved.
Received 30 November 2010
Received in revised form 11 April 2011
Accepted 16 April 2011
Available online 22 April 2011
Keywords:
Trifluoromethanesulfonic acid
Zirconium oxide
Sol–gel
Catalyst
Parabens
1. Introduction
which generate a large amount of acid waste. At the same time, they
allow for easy product and catalyst separation, thus decreasing the
The parabens are esters of p-hydroxybenzoic acid, which are
widely used as preservatives in food and also in cosmetics. They
are antimicrobial agents, being more active against fungi and yeast
rather than against bacteria.
The esters of p-hydroxybenzoic acid were obtained with the
purpose of replacing salicylic and benzoic acids, both of which have
the drawback of being effective only in the highly acid pH range,
assuming that the esters might be advantageous in this aspect since
they have a wide pH range [1]. Recently, excellent reviews have
been published about parabens, which deal not only with their
properties and uses, but also report toxicological and risk assess-
ment studies [2–4].
The antimicrobial action of p-hydroxybenzoic acid esters
increases with the chain length, but the solubility decreases; hence,
in practice, the esters with shorter chain length are commonly used.
The parabens are obtained through the reaction of the corre-
sponding alcohol with p-hydroxybenzoic acid, usually catalyzed by
acids. Due to the increasing awareness about environmental care, it
is advisable to use solid catalysts to avoid the use of mineral acids,
need of the classical separations by distillation or extraction.
The heterogeneization of trifluoromethanesulfonic acid
(CF3SO3H) by immobilization on a suitable support is an interest-
ing alternative to contribute to the field of clean processes. This
acid, also named triflic acid, has a highly acidic nature and an
excellent thermal stability; it also has good resistance to reductive
and oxidative dissociation, with no generation of fluoride ions [5].
The trifluoromethanesulfonic acid was used as an efficient
homogeneous catalyst in many organic reactions, such as polymer-
ization and acylation [6,7], among others, but has environmental
disadvantages because the recovery of triflic acid (Tri) from the
reaction mixture results in the formation of large amounts of waste.
Papers dealing with Tri heterogeneization are scarce, though its
immobilization on zirconium hydroxide for its use in the biphenyl
benzoylation reaction [8], and in the synthesis of flavones and
chromones immobilized on carbon [9] or titania [10] has been
reported.
With the purpose of making a contribution to a little explored
field and as a continuation of previous work, the preparation and
characterization of new catalysts obtained by impregnation of tri-
fluoromethanesulfonic acid on zirconia prepared by the sol–gel
method, using zirconium propoxide as precursor, and urea as a
nonsurfactant low-cost pore-forming agent, is reported.
∗
Corresponding author. Tel.: +54 221 421 1353; fax: +54 221 421 1353.
E-mail address: mnblanco@quimica.unlp.edu.ar (M. Blanco).
0926-860X/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.apcata.2011.04.018

92
M. Gorsd et al. / Applied Catalysis A: General 400 (2011) 91–98
Table 1
The effect of the calcination temperature of the support on the
Trifluoromethanesulfonic acid amount in the catalysts.
physicochemical, acidic and textural characteristics of the catalysts
was studied. The catalytic behavior of the prepared materials in
paraben synthesis was firstly studied through the esterification
of p-hydroxybenzoic acid with propyl alcohol in a conventional
batch glass reactor at atmospheric pressure. In addition, the effi-
ciency of the obtained catalysts was tested in the esterification of
p-hydroxybenzoic acid with methanol, ethanol and n-propanol in
an autoclave reactor at the self-generated pressure. The results are
correlated with the catalyst properties.
Sample
NTri (mmol CF3SO3H/g)
WTri (mmol CF3SO3H/g)
TriZrT100
TriZrT200
TriZrT300
TriZrT400
0.91
0.53
0.41
0.10
0.88
0.49
0.38
0.11
Fourier transform infrared (FT-IR) spectra of the samples were
recorded using Bruker IFS 66 FT-IR equipment, pellets with
approx. 1% w/w of the sample in KBr, in a measuring range of
400–4000 cm−1
.
2. Experimental
The acidity of the solids was measured by means of potentiomet-
ric titration. A known mass of solid was suspended in acetonitrile
and stirred for 3 h. Then, the suspension was titrated with 0.05 N
n-butylamine in acetonitrile solution at a flow of 0.05 cm3/min,
measuring the electrode potential variation in a digital pH meter
Hanna 211 with a double-junction electrode.
2.1. Support preparation
Zirconium propoxide (Aldrich, 26.6 g) was mixed with absolute
ethanol (Merck, 336.1 g) and stirred for 10 min to obtain a homo-
geneous solution under N2 at room temperature. Then 0.47 cm3 of
0.28 M HCl aqueous solution was dropped slowly into the above
mixture to catalyze the sol–gel reaction.
2.4. Paraben synthesis
After 3 h, an appropriate amount of urea–alcohol–water (1:5:1
weight ratio) solution was added to the hydrolyzed solution under
vigorous stirring to act as template. The amount of added solution
was fixed in order to obtain a template concentration of 10% by
weight in the final material.
The gel was kept in a beaker at room temperature up to dry-
ness. The solid was ground into powder and extracted with distilled
water for three periods of 24 h, in a system with continuous stir-
ring, to remove urea. Afterwards, portions of the obtained solid
were calcined at 100, 205, 310, or 425 ◦C for 24 h, thus obtaining the
supports identified as ZrT100, ZrT200, ZrT300, and ZrT400, respectively.
The esterification of p-hydroxybenzoic acid with propyl alcohol
to obtain propyl paraben was carried out in liquid phase at 100 ◦C
in a 50 ml batch glass reactor equipped with a condenser and a
magnetic stirrer. The reagent mixture was heated under stirring
and then the catalyst was added. An n-propanol:p-hydroxybenzoic
acid:catalyst molar ratio of 10:1:0.1 was used. Samples were peri-
odically taken and analyzed by gas chromatography in a Shimadzu
GC-14B chromatograph with TC detector, and using dodecane as
internal standard.
The esterification was also performed in a 15 ml autoclave reac-
tor heated up to 100 ◦C, at the self-generated pressure, using the
same reagent molar ratio mentioned above, not only to evaluate the
catalyst performance in this type of reactor for the esterification of
p-hydroxybenzoic acid with n-propanol, but also with ethanol and
methanol, in order to obtain propyl, ethyl and methyl parabens.
After 5 h under reaction, the analysis of the reaction mixture was
performed by gas chromatography.
2.2. Catalyst preparation
Trifluoromethanesulfonic acid (0.01 mol, Alfa Aesar) was added
dropwise to a mixture of ZrTX (2 g) and toluene (20 cm3, Merck)
at 90 ◦C under nitrogen atmosphere; then it was further refluxed
for 2 h. Next, the sample was cooled, filtered, washed with acetone
(Mallinckrot AR) and dried at 100 ◦C for 24 h.
The solids were then extracted with
a
mixture of
3. Results and discussion
dichloromethane and diethyl ether (40 cm3/g of solid) for three
periods of 4 h in order to remove the acid weakly attached to the
support. Afterwards, they were dried again at 100 ◦C for 24 h. The
amount of trifluoromethanesulfonic acid retained in the solids
was determined by C and S elemental analysis with a Carlo Erba
The amount of trifluoromethanesulfonic acid attached to the
support (NTri), calculated from the elemental analysis, decreased
with the increment of the thermal treatment temperature of the
support (Table 1). This effect may be explained if the interac-
tion between the trifluoromethanesulfonic acid and zirconia is
assumedtobeofelectrostatic type due toproton transfer to the–OH
groups on the support surface with subsequent water evolution
EA1108. The samples were named TriZrT100, TriZrT200, TriZrT300
and TriZrT400
,
.
2.3. Solid characterization
The characterization of the thermal properties of the supports
and the catalysts was performed by thermogravimetric (TGA) and
differential scanning calorimetric (DSC) analyses, with a Shimadzu
DT 50 thermal analyzer. The measurements were carried out under
argon, with a 25–50 mg sample, and a heating rate of 10 ◦C/min,
using quartz cells as sample holders and ␣-Al2O3 as reference.
The textural properties of the solids were determined from
N2 adsorption–desorption isotherms at the liquid-nitrogen tem-
perature. They were obtained using Micromeritics ASAP 2020
equipment. The samples were previously degassed at 100 ◦C for
2 h.
X-ray diffraction (XRD) patterns of the solids were recorded with
Philips PW-1732 equipment, using Cu K␣ radiation, Ni filter, 20 mA
and 40 kV in the high voltage source, a 5–60◦ 2ꢀ scanning angle
range, and a scanning rate of 1◦ per min.
Scheme 1.

M. Gorsd et al. / Applied Catalysis A: General 400 (2011) 91–98
93
calcined at 205, 310, and 425 ◦C showed similar characteristics.
This exothermic peak appeared at 494 and 487 ◦C in the dia-
grams corresponding to the TriZrT200 and TriZrT300 samples, and
it is absent in that of the TriZrT400 sample up to 500 ◦C (Fig. 2).
A similar shift of the exothermal peak maximum was previously
reported for samples containing molybdenum oxide or copper
oxide on zirconia [11], for zirconia doped with nickel or alu-
minum [12], and for zirconia with sulfate or chromate addition
[13]. These authors agree in that the stability increase of the
hydrated zirconium oxide is related to an inhibitory effect in zir-
conia crystallization induced by a good dispersion of this type of
additive.
491
434
365
TriZrT100
ZrT100
Additionally, in all the samples containing trifluoromethane-
sulfonic acid, another exothermal peak appeared in the range
350–370 ◦C, ascribed to acid elimination. The intensity of this peak
diminished in parallel with the NTri decrease.
171
69
The TGA diagram of the ZrT100 support (Fig. 3) showed that dehy-
dration takes place in three main steps. The first step, below 120 ◦C,
is due to the evolution of physically adsorbed water and the sec-
ond, in the temperature range 120–250 ◦C, to the loss of structural
water. Above 250 ◦C, a small weight loss takes place, which can be
assigned to the decomposition of Zr(OH)4 into ZrO2.
The TGA patterns showed that the catalysts are thermally stable
up to 250 ◦C, and from this temperature on, acid evolution begins.
From the comparison of the TriZrT100 catalyst with the ZrT100
support (Fig. 3), it is clearly observed that the evolution of the
adsorbed acid takes place at temperatures between 250 ◦C and
400 ◦C, an effect that appeared in the DSC diagram as an exother-
mic peak with a maximum at 365 ◦C (Fig. 1). A similar behavior
was observed in the TGA diagrams of the catalysts obtained from
the supports calcined at 205, 310 and 425 ◦C (Fig. 4).
100
200
300
400
500
600
Temperature (°C)
Fig. 1. DSC diagrams of the ZrT100 and TriZrT100 samples.
(Scheme 1). So, as a result of the support dehydroxylation during
the thermal treatment, the amount of OH groups to be protonated
decreases, and therefore NTri diminishes.
The DSC diagram of the ZrT100 support (Fig. 1) showed two
endothermic peaks at 69 and 171 ◦C, attributed to the loss of phys-
ically adsorbed water, and the partial dehydroxylation of the solid,
respectively. There was also an exothermic peak at 434 ◦C, which
was assigned to zirconia transformation from an amorphous to a
crystalline phase.
In the DSC diagram of the TriZrT100 catalyst (Fig. 1), the last
peak appeared shifted to higher temperatures (491 ◦C), indicating
that the trifluoromethanesulfonic acid retards zirconia crystal-
lization. The samples obtained by impregnation of the supports
The amount of acid estimated from the TGA diagrams (WTri
)
is in close agreement with the amount of acid firmly attached to
the support (NTri) calculated from the elemental analysis, (Table 1).
494
TriZrT200
TriZrT300
TriZrT400
487
368
350
372
200
400
600
200
400
600
200
400
600
Temperature (ºC)
Fig. 2. DSC diagrams of the TriZrT200, TriZrT300 and TriZrT400 catalysts.

94
M. Gorsd et al. / Applied Catalysis A: General 400 (2011) 91–98
Table 2
Textural properties of the supports calcined at different temperatures.
1,00
0,95
0,90
0,85
0,80
a
b
Sample
SBET
(m2/g)
SMicro
VP
(cm3/g)
Micropore
volumea
(cm3/g)
DP
(m2/g)
(nm)
ZrT100
ZrT200
ZrT300
ZrT400
192
132
78
88
50
17
0
0.18
0.16
0.11
0.07
0.05
0.02
0.01
0
3.7
4.7
5.5
ZrT100
20
14.2
a
Micropore specific surface area and volume estimated by the t-plot method.
Mean pore diameter obtained from the specific BET surface area.
b
TriZrT100
mesopores present in the material, as was reported for MCM-41
mesoporous materials [14].
The specific surface area (SBET) of the supports, together
with the average pore diameter (DP) obtained from the
Brunauer–Emmett–Teller pore size distribution, the total pore
volume (VP) estimated from the value corresponding to a ratio
p/p0 = 0.98, the micropore specific surface area (SMicro) and the
micropore volume, both estimated from the t-plot method, are
gathered in Table 2.
100
200
300
400
500
600
Temperature (°C)
Fig. 3. TGA diagrams of the ZrT100 and TriZrT100 samples.
The values of DP were higher than 3.7 nm, so mainly mesoporous
materials were obtained, with VP being significantly higher than
the micropore volume. While DP increased with the temperature
of the thermal treatment, both SBET and SMicro, and VP decreased
with the increment of the calcination temperature. At the same
time, the micropore contribution to the total value of surface and
volume also diminished. On the other hand, the textural proper-
ties of the catalysts were almost the same as those of the supports,
thus indicating that they are not altered by trifluoromethanesul-
fonic acid addition, in a similar way to the report about the same
acid supported on titanium dioxide [15].
WTri was calculated using the following expression:
WTri = (WTriZrTx − WZrTx)/FWTri
where WTriZrTx = weight loss of TriZrTx catalyst from 250 ◦C to 600 ◦C
(mg/g). WZrTx = weight loss of ZrTx support from 250 ◦C to 600 ◦C
(mg/g). FWTri = CF3SO3H molecular weight.
The textural properties of the samples were determined from
the N2 adsorption–desorption isotherms at the liquid-nitrogen
temperature. The isotherms obtained can be classified as type IV,
characteristic of mesoporous materials, hysteresis being hardly
visible, which is attributed to an ordered arrangement of the
The XRD patterns of the support calcined at 100, 205, and 310 ◦C
(Fig. 5) showed a broad band at 2ꢀ = 30◦ attributed to their low
TriZrT200
TriZrT400
TriZrT300
1,00
0,95
0,90
0,85
1,00
0,95
0,90
0,85
1,00
0,95
0,90
0,85
200
400
600
200
400
600
200
400
600
Temperature (°C)
Fig. 4. TGA diagrams of the TriZrT200, TriZrT300 and TriZrT400 catalysts.

M. Gorsd et al. / Applied Catalysis A: General 400 (2011) 91–98
95
T
ZrT100
TriZrT100
T
TriZrT200
T
M
M
M
ZrT400
TriZrT300
ZrT300
TriZrT400
ZrT200
ZrT100
4000
3000
2000
1000
Wavenumber (cm-1)
10
20
30
40
50
60
2
Fig. 6. FTIR spectra of the ZrT100 support, and TriZrT100, TriZrT200, TriZrT300, and
TriZrT400 samples.
Fig. 5. XRD diagrams of the ZrT100, ZrT200, ZrT300, and ZrT400 supports. (T): peaks
assigned to the tetragonal phase. (M): peaks assigned to the monoclinic phase.
is reached indicates the total number of acid sites. The acid
strength of these sites may be classified according to the following
scale: Ei > 100 mV (very strong sites), 0 < Ei < 100 mV (strong sites),
−100 < Ei < 0 (weak sites) and Ei < −100 mV (very weak sites) [20].
crystallinity. In turn, the pattern of the ZrT400 support showed peaks
with maxima at 2␪ = 28.5, 30.5, 31.7, 35.0, 35.5 and 50.6◦ assigned
to the metastable tetragonal phase of zirconia, which is the main
phase present though it is accompanied by a minor amount of the
monoclinic phase of zirconia, as indicated in Fig. 5.
The potentiometric titration results indicated that the TriZrT100
,
TriZrT200, TriZrT300, and TriZrT400 catalysts present very strong acid
sites, with Ei potential values between 650 mV and 700 mV, and that
the potential decreased more quickly when the temperature of the
thermal treatment increased (Fig. 7a), indicating a total number of
acid site decrease at higher temperatures. Values of 4.5, 3.5, 3 and
The XRD patterns of the catalysts (TriZrT100, TriZrT200, TriZrT300
,
and TriZrT400) showed features similar to those of the correspond-
ing supports used for their preparation. This fact indicates that the
impregnation of hydrated zirconia with trifluoromethanesulfonic
acid did not appreciably affect the crystalline characteristics.
On the other hand, the FT-IR spectra of the zirconia gel after
being leached with distilled water did not present any of the
characteristic bands of urea, as that assigned to the C=O group
at 1660 cm−1 [16], showing that the template removal by water
extraction was effective.
2 meq amine/g solid for TriZrT100, TriZrT200, TriZrT300, and TriZrT400
,
respectively, were obtained. On the other hand, the potentiometric
titration curves of the supports were very similar to each other,
with Ei being around 360 mV (Fig. 7b).
So, the acid strength of the catalysts is nearly independent of
NTri, and much higher than that of the supports. In addition, the
The FT-IR spectrum of the ZrT100 sample (Fig. 6) displayed an
intense band between 3600 cm−1 and 3200 cm−1 and another one
at 1600 cm−1, assigned to the stretching vibrations of hydroxo- and
aquo-OH, and to the bending vibration of (H–O–H) and (O–H–O)
present in the structure of the solid, respectively [17–19]. Also, in
the energy interval below 850 cm−1, the wide band due to Zr–O
stretching vibration is observed.
FT-IR spectra of ZrT200, ZrT300, and ZrT400 supports showed fea-
tures similar to those observed for the ZrT100 sample, though the
intensity of the bands at 1600 cm−1 and 3600 cm−1 decreased when
the calcination temperature increased, in parallel with the reduc-
number of acid sites decreased in an almost linear relation with
the NTri decrease. As was previously stated, NTri was dependent on
the hydroxylation degree of the support resulting from the thermal
treatment applied, which not only affected the SBET of the materials
but also, and particularly, the number of sites in which the triflic
acid can firmly interact.
The esterification reaction is a type of organic reaction of con-
siderable industrial interest, and it is generally carried out in liquid
phase. The mechanism of this reaction homogeneously catalyzed
by acids is well known [21], and involves the protonation of the
carboxylic oxygen, giving a delocalized carbocation, which can be
nucleophilically attacked by an alcohol. The loss of a proton of the
adduct produces the intermediary. The protonation of each hydrox-
ylic oxygen leads to water elimination to give the ester. The addition
of alcohol excess or water extraction from the reaction medium
favors the esterification by displacing the equilibrium.
The mechanism of the reaction carried out employing hetero-
geneous catalysis can follow an Eley–Rideal (ER) of unique site
or a Langmuir–Hinshelwood (LH) of dual site mechanism [22,23],
which depends on the type of material employed as catalyst and
the working conditions. For exchange resins, it was suggested that
the esterification in liquid phase occurs between the protonated
adsorbed alcohol and the acid in the liquid phase [24,25] or by a
dual site mechanism [26]. In turn, for zeolites, mesoporous solids,
zirconia modified with different additives, such as tungstated zir-
conia, it was reported that the reaction proceeds by a unique site
tion of the SBET
.
The FT-IR spectrum of the TriZrT100 sample displayed bands at
1267 cm−1, 1180 cm−1, and 1040 cm−1, in addition to those present
in the ZrT100 sample (Fig. 6). The first two bands are ascribed
to the S=O stretching mode and the last one is assigned to the
C–F stretching mode [16], thus showing the presence of trifluo-
romethanesulfonic acid adsorbed on the support. These bands are
also present in the spectra of TriZrT200, TriZrT300, and TriZrT400 cata-
lysts, although with lower intensity as a result of the lower amount
of triflic acid adsorbed on the support.
The acidity measurements of the catalysts by means of poten-
tiometric titration with n-butylamine let us estimate the number
of acid sites and their acid strength. It was suggested that the ini-
tial electrode potential (Ei) indicates the maximum acid strength
of the sites, and the value of meq amine/g solid where the plateau

96
M. Gorsd et al. / Applied Catalysis A: General 400 (2011) 91–98
800
700
600
500
400
300
200
100
TriZrT100
TriZrT200
TriZrT300
TriZrT400
ZrT100
ZrT200
ZrT300
ZrT400
1
2
3
4
5
1
2
3
4
5
meq n-butylamine/g
Fig. 7. Potentiometric titration curves of the TriZrT100, TriZrT200, TriZrT300, and TriZrT400 samples (a), and those of the respective supports (b).
mechanism, with the acid adsorbed on the catalytic site, followed
by reaction with the alcohol from the liquid phase [23,27,28].
The prepared catalysts were tested in the esterification of
p-hydroxybenzoic acid with n-propyl alcohol (Scheme 2). The syn-
thesis of propyl paraben was carried out in a conventional batch
glass reactor at atmospheric pressure and 100 ◦C, with a high alco-
hol:acid molar ratio to favor the displacement of the reaction and
a catalyst amount optimized from exploratory results.
It was observed that the amount of p-hydroxybenzoic acid
converted into the ester continuously increased with the reac-
tion time (Fig. 8). Furthermore, after 5 h, the reached conversion
depended strongly on the CF3SO3H content, decreasing from 80%
to 7% (Table 3) when NTri decreased from 0.91 mmol to 0.10 mmol
CF3SO3H/g. On the other hand, the experiences performed using
the supports as catalyst showed a negligible conversion (<2%).
Under the working conditions, it was found that the catalysts are
selective, and only traces of propyl ether were detected by means
of MS-GC.
The specific catalytic activity (SCA) of the samples,
expressed as moles of ester formed at 5 h/mol CF3SO3H
in the catalyst, decreased slightly in the following order
TriZrT100 > TriZrT200 > TriZrT300 > TriZrT400 (Table 3), an order-
ing similar to that followed by the number of acid sites present in
the catalysts. A similar conclusion was reported by Landge et al.
[29] for toluene benzoylation using Zr-TMS functionalized with
triflic acid as catalyst.
After separating the catalyst TriZrT100 from the reaction
medium, it was washed with ethanol and dried, and then it was
reused several times with a slight loss of catalytic activity. While the
amount of p-hydroxybenzoic acid converted into propyl paraben
with the fresh catalyst was 80%, a value of 77% was obtained in the
first reuse and 75% in the second reuse. The TriZrT100 catalyst after
the second reuse displayed SBET, NTri and Ei values similar to those
of the fresh one.
On the other hand, the tests performed with the purpose of
comparing the catalytic behavior of the TriZrT100 catalyst in the syn-
thesis of the propyl, ethyl and methyl esters of p-hydroxybenzoic
acid were carried out in an autoclave reactor, because it was not
100
TriZrT100
TriZrT200
TriZrT300
80
60
40
20
TriZrT400
Table 3
Catalytic behavior of the catalysts in propyl paraben synthesis.
Sample
p-hydroxybenzoic
acid conversion (%)
SCAa (mol
ester/mol
CF3SO3H)
TriZrT100
TriZrT200
TriZrT300
TriZrT400
80
45
32
7
8.8
8.6
7.6
7.1
50
100
150
200
250
300
Time (min)
Fig. 8. Conversion of p-hydroxybenzoic acid in propyl paraben synthesis using the
TriZrT100, TriZrT200, TriZrT300, and TriZrT400 samples as catalysts.
a
SCA: specific catalytic activity at 5 h.

M. Gorsd et al. / Applied Catalysis A: General 400 (2011) 91–98
97
Scheme 2.
100
80
60
40
20
0
The acid amount depended on the thermal treatment to which
the support was subjected. As the increase of the calcination tem-
perature produces the support dehydroxylation, the amount of
−OH groups available to firmly adsorb trifluoromethanesulfonic
acid decreased, thus decreasing the acid amount.
The prepared catalysts presented very strong acid sites whose
number, determined by potentiometric titration, correlated well
with the amount of acid firmly attached to the support (NTri).
Both the acid and textural properties of the catalysts showed
that adequate solids to be used as catalysts in acid reactions were
obtained. Particularly, the conversion measured in the esterifica-
tion of p-hydroxybenzoic acid with n-propanol for the synthesis
of propyl paraben showed that the conversion definitely depended
on the number of acid sites present in the catalysts. This charac-
teristic was, in turn, a result of the preparation conditions of the
catalysts, the thermal treatment of the support being the most
important parameter, which affected the textural properties and
also the amount of trifluoromethanesulfonic acid firmly adsorbed
on the support, as a consequence of the dehydroxylation produced
when the temperature was increased.
methyl ester
ethyl ester
propyl ester
Fig. 9. p-hydroxybenzoic acid conversion in the synthesis of methyl, ethyl and
propyl parabens.
With regard to the esterification reaction to obtain parabens
with shorter alkylic chain, slightly lower conversion was obtained
under the working conditions used. However, it can be concluded
that, in all cases, the prepared materials were selective and led to
very good conversion values, thus making them suitable candidates
to be used in replacement of classical acids utilized for esterification
reactions.
possible to obtain the ethyl and methyl parabens in the previous
conditions. The results obtained at 5 h under reaction were 98%
conversion for the obtainment of propyl paraben, 95% for ethyl
paraben, and 82% for methyl paraben (Fig. 9).
As can be observed, under the employed conditions, in which the
same ratio was used for the three alcohols, the conversion decreases
for the alcohols with shorter chain.
According to Climent et al. [30], a conversion increase with
the increment of the alcohol chain length in this type of reac-
tion is due to a higher alcohol hydrophobicity, thus allowing a
better alcohol interaction with the carbonylic carbon of the car-
bocation formed by the carboxylic acid adsorption on the catalyst.
It must also be taken into account that the conversion decrease
observed in obtaining the parabens corresponding to alcohols with
shorter chain can be the result of the lower boiling point of the
alcohols, which showed the following ordering: n-propyl alcohol
(97 ◦C) > ethyl alcohol (78.4 ◦C) > methyl alcohol (64.7 ◦C).
In brief, the desired products were selectively obtained with
very good yields using trifluoromethanesulfonic acid adsorbed on
zirconia as catalyst.
Acknowledgements
The authors thank G. Valle and E. Soto for their experimental
contribution, and CONICET and UNLP for the financial support.
References
[1] T. Sabalitschka, in: T.E. Furia (Ed.), CRC Handbook of Food Additives, Vol. 1, CRC
Press, Cleveland, Ohio, USA, 1975, pp. 115–184.
[2] M.G. Soni, G.A. Burdock, S.L. Taylor, N.A. Greenberg, Food Chem. Toxicol. 39
(2001) 513–532.
[3] M.G. Soni, S.L. Taylor, N.A. Greenberg, G.A. Burdock, Food Chem. Toxicol. 40
(2002) 1335–1373.
[4] M.G. Soni, I.G. Carabin, G.A. Burdock, Food Chem. Toxicol. 43 (2005) 985–1015.
[5] D.S. Sood, S.C. Sherman, A.V. Iretskii, J.C. Kenvin, D.A. Schiraldi, M.G. White, J.
Catal. 199 (2001) 149–153.
[6] R.D. Howells, J.D.Mc. Cown, Chem. Rev. 77 (1977) 69–92.
[7] M. Chamoumi, D. Brunel, F. Fajula, P. Geneste, P. Moreau, J. Solofo, Zeolites 14
(1994) 282–289.
[8] M. Chidambaram, C. Venkatesan, P.R. Rajamohanan, A.P. Singh, Appl. Catal. A:
Gen. 244 (2003) 27–37.
[9] D.O. Bennardi, G.P. Romanelli, J.C. Autino, L.R. Pizzio, Catal. Commun. 10 (2009)
576–581.
[10] D.O. Bennardi, G.P. Romanelli, J.C. Autino, L.R. Pizzio, Appl. Catal. A: Gen. 324
(2007) 62–68.
4. Conclusions
The parabens were prepared using trifluoromethanesulfonic
acid immobilized on zirconia. The oxide was obtained by the sol–gel
method, using urea as a pore-forming agent, and it was a mainly
mesoporous solid. When the calcination temperature of the sup-
port was increased, the specific surface area and the microporous
contribution to this area decreased, and at the same time the mean
pore diameter increased. The impregnation of the supports with
triflic acid, and the subsequent leaching with dichloromethane and
diethyl ether led to solid catalysts with CF3SO3H tightly adsorbed
on the support.
[11] B.-Y. Zhao, X.-P. Xu, H.-R. Ma, D.-H. Sun, J.-M. Gao, Catal. Lett. 45 (1997)237–244.
[12] J.C. Duchet, M.J. Tilliette, D. Cornet, Catal. Today 10 (1991) 507–520.
[13] T. Yamaguchi, Catal. Today 20 (1994) 199–218.
[14] S. Ajaikumar, A. Pandurangan, J. Mol. Catal. A: Chem. 266 (2007) 1–10.
[15] L.R. Pizzio, Mater. Lett. 60 (2006) 3931.
[16] L.J. Bellamy, The Infrared Spectra of Complex Molecules, Wiley, New York, 1960.

98
M. Gorsd et al. / Applied Catalysis A: General 400 (2011) 91–98
[17] S. Patel, N. Purohit, A. Patel, J. Mol. Catal. A: Chem. 192 (2003) 195–202.
[18] J. Rubio, J.L. Otero, M. Villegas, P. Duran, J. Mater. Sci. 32 (1997) 643.
[19] J.A.R. Van Veen, F.T.G. Veltmaat, G. Jonkers, J. Chem. Soc., Chem. Commun.
(1985) 1656.
[25] C.S.M. Pereira, S.P. Pinho, V.M.T.M. Silva, A.E. Rodríguez, Ind. Eng. Chem. Res.
47 (2008) 1453–1463.
[26] H.T.R. Teo, B. Saha, J. Catal. 228 (2004) 174–182.
[27] Y. Liu, E. Lotero, J.G. Goodwin Jr., J. Catal. 242 (2006) 278–286.
[20] V.M. Fuchs, L.R. Pizzio, M.N. Blanco, Eur. Polym. J. 44 (2008) 801–807.
[21] I. Roberts, H.C. Urey, J. Am. Chem. Soc. 60 (1938) 2391.
[22] L.R. Pizzio, M.N. Blanco, Appl. Catal. A: Gen. 255 (2003) 265–277.
[23] D.E. López, K. Suwannakarn, J.G. Goodwin Jr., D.A. Bruce, Ind. Eng. Chem. Res.
47 (2008) 2221–2230.
[28] S.R. Kirumakki, N. Nagaraju, K.V.R. Chary, Appl. Catal. A: Gen. 299 (2006)
185.
[29] S.M. Landge, M. Chidambaram, A.P. Singh, J. Mol. Catal. A: Chem. 213 (2004)
257–266.
[30] M.J. Climent, A. Corma, S. Iborra, S. Miquel, J. Primo, F. Rey, J. Catal. 183 (1999)
76.
[24] M. Altiokka, A. Citak, Appl. Catal. A: Gen. 239 (2003) 141–148.