Propylsulfonic acid functionalized mesoporous silica catalysts for esterification of fatty acids

Article abstract of DOI:10.1016/j.molcata.2015.10.002

The catalytic properties of 3 types of mesoporous silica SBA-15 (rope, rod and fiber), with 9.2 nm or 12.1 nm large mesopores, were examined with respect to their morphology and pore size. Commercially available Amberlyst-15 and the small pore sized MCM-41 were used for comparison. The catalysts were prepared by functionalization of the silica supports with propylsulfonic acid (Pr-SO₃H) using post-synthesis grafting with 3-mercaptopropyltrimethoxysilane as a propyl-thiol precursor. All materials remained in a well-ordered hexagonal mesoporous structure after Pr-SO₃H functionalization. The performance of the Pr-SO₃H-functionalized mesoporous silicas was evaluated in terms of their catalytic activity in the esterification of oleic acid with short (methanol) and long (glycerol) chain alcohols, i.e., to test the effect of the pore size on the substrate conversion and product yield. The synthesized catalysts were highly active and the product composition could be tuned by selective choice of the mesopore size. The Pr-SO₃H-functionalized rope-shaped SBA-15 gave the highest catalytic activity (in terms of the highest methyl oleate and triglyceride yields and oleic acid conversion level), which was higher than that obtained with the commercial Amberlyst-15 catalyst. A high acid amount, large specific surface area and a suitable pore size are the likely reasons for the high yield gained by Pr-SO₃H-functionalized rope-shaped SBA-15 silica.

Full text of DOI:10.1016/j.molcata.2015.10.002

Journal of Molecular Catalysis A: Chemical 410 (2015) 253–259
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Propylsulfonic acid functionalized mesoporous silica catalysts for
esterification of fatty acids
Sirima Jeenpadiphata, Emma M. Björkb, Magnus Odénb,
Duangamol Nuntasri Tungasmitaa,∗
a Materials Chemistry and Catalysis Research Unit, Department of Chemistry, Faculty of Science, Chulalongkorn University, 254 Phayathai Road, Bangkok
10330, Thailand
b Nanostructured Materials, Department of Physics, Chemistry and Biology, Linköping University, SE-58183, Sweden
a r t i c l e i n f o
a b s t r a c t
Article history:
The catalytic properties of 3 types of mesoporous silica SBA-15 (rope, rod and fiber), with 9.2 nm or
12.1 nm large mesopores, were examined with respect to their morphology and pore size. Commer-
cially available Amberlyst-15 and the small pore sized MCM-41 were used for comparison. The catalysts
were prepared by functionalization of the silica supports with propylsulfonic acid (Pr-SO3H) using post-
synthesis grafting with 3-mercaptopropyltrimethoxysilane as a propyl-thiol precursor. All materials
remained in a well-ordered hexagonal mesoporous structure after Pr-SO3H functionalization. The perfor-
mance of the Pr-SO3H-functionalized mesoporous silicas was evaluated in terms of their catalytic activity
in the esterification of oleic acid with short (methanol) and long (glycerol) chain alcohols, i.e., to test the
effect of the pore size on the substrate conversion and product yield. The synthesized catalysts were
highly active and the product composition could be tuned by selective choice of the mesopore size. The
Pr-SO3H-functionalized rope-shaped SBA-15 gave the highest catalytic activity (in terms of the highest
methyl oleate and triglyceride yields and oleic acid conversion level), which was higher than that obtained
with the commercial Amberlyst-15 catalyst. A high acid amount, large specific surface area and a suitable
pore size are the likely reasons for the high yield gained by Pr-SO3H-functionalized rope-shaped SBA-15
Received 6 December 2014
Received in revised form
30 September 2015
Accepted 2 October 2015
Available online 8 October 2015
Keywords:
Mesoporous silica
Acid functionalization
Esterification
Fatty acid
Glycerol
silica.
© 2015 Elsevier B.V. All rights reserved.
1. Introduction
agents, such as hexane and dodecane, in the low temperature syn-
thesis of SBA-15 [9–14] yielded materials with pore sizes that
In order to optimize the performance of mesoporous silica in
important applications, such as catalysis, separation, adsorption
and ion-exchange, new synthetic chemicals have been introduced
into their composition to alter their properties [1]. Recently, several
new types of large-pore mesoporous silica materials, such as SBA-
15, have been reported [2,3], where the material’s structure and
texture, especially the acidic sites, affect its catalytic properties [4].
The use of mesoporous silica SBA-15 as a catalyst was selected
for this study because of its well-ordered hexagonal pore struc-
ture, high surface area (690–1040 m2/g) and large-pore diameter
(5–30 nm) [5–7]. Different SBA-15 materials exhibit variations in
their surface area, pore size and wall thickness, which makes them
suitable for different applications [8]. The use of swelling alkane
varied from 9.7 to 15.7 nm depending on the alkane chain length
[8]. Different morphologies (fiber, platelet, rod and film) of meso-
porous silica can be prepared by varying the reaction conditions
during their synthesis [15–20] or by the addition of KCl or NH4F
salts [21–23].
Esterification is an acid-catalyzed reaction [24], and so to opti-
mize the reaction different types of acid functional groups have
been incorporated into the framework of mesoporous silica mate-
rials in order to increase their acidity [25]. The improvement in
the catalytic activity and selectivity in the esterification of glyc-
erol with fatty acids by the incorporation of alkyl or phenyl groups
in combination with sulfonic groups in mesoporous silica is one
recently reported example of organo-sulfonic functionalization
[24,26]. In addition, organic-inorganic hybrid silicas synthesized
by supramolecular assembly have received increasing attention
because of their high surface area, flexible pore size and potential
for controlling catalytic functionalities at a molecular level [27–31].
In this study the potential to use acid-functionalized mesoporous
∗
Corresponding author. Fax: +66 2 2541309.
E-mail addresses: duangamol.n@chula.ac.th, duangamol.n@gmail.com
(D.N. Tungasmita).
http://dx.doi.org/10.1016/j.molcata.2015.10.002
1381-1169/© 2015 Elsevier B.V. All rights reserved.

254
S. Jeenpadiphat et al. / Journal of Molecular Catalysis A: Chemical 410 (2015) 253–259
silica materials for the catalysis of esterification in the production
of biodiesel and glycerol esters was evaluated. The products can
be used as alternative fuels, or in the food, pharmaceutical and
cosmetic industries. Commonly, esterification is performed using
conventional acidic homogeneous catalysts, such as H2SO4 and HCl,
to provide a high conversion level and high yield in a short time,
but the catalyst is difficult to remove from the product [24]. There-
fore, it would be desirable to perform the esterification with an
acidic heterogeneous catalyst and then separate the catalyst from
the system to both obtain a clean product and to reuse the catalyst.
In the work described here, post-synthesis grafting of 3-
mercaptopropyltrimethoxy silane (MPTMS) into heterogeneous
mesoporous silica with 2.0–12.1 nm large pores and various mor-
phologies was performed. The mesoporous silica-propylsulfonic
acid (Pr-SO3H) catalysts subsequently obtained upon oxidation of
the thiol precursors were evaluated for their ability to catalyze the
esterification of methanol and glycerol with oleic acids. The synthe-
sized Pr-SO3H-functionalized silica catalysts were characterized by
X-ray powder diffractometry (XRD), physisorption, Fourier trans-
form infrared spectroscopy (FTIR), scanning electron microscopy
(SEM), transmission electron microscopy (TEM) and acid-base titra-
tion analyses. Finally, the relationships between the acidity, specific
surface area, mesopore size and catalytic behavior of the catalyst
are discussed.
stirring, and left static for formation of the rod morphology, the
solutions were transferred to Teflon flasks and hydrothermally
treated for 24 h at 100 ◦C. After filtration, the synthesized mate-
rials were collected and calcined at 550 ◦C in air for 5 h. The stirring
time and HCl concentration were altered in order to obtain rod-
or fiber-shaped morphologies. The rod- and fiber-shaped SBA-15
materials were denoted as rd-SBA-15 and f-SBA-15, respectively.
2.2.3. MCM-41
Hexagonal mesoporous silica (MCM-41) was prepared from
CTAB and TEOS in base solution as previously reported [33]. The
surfactants were removed by calcination at 550 ◦C in air for 5 h.
2.3. Catalyst preparation
The Pr-SO3H-functionalized silica samples were prepared using
a modified post-synthesis grafting method based on the proto-
col of Reddy et al. [34]. A mixture of 3.0 g of the respective silica
sample (rp-SBA-15, rd-SBA-15, f-SBA-15, MCM-41 and fume silica)
and 5.3 g of MPTMS was refluxed in 50 mL of dry toluene for 6 h
and the obtained solid (SBA-15-Pr-SH) was filtered and dried in
air. The -SH groups were converted to -SO3H groups by oxidation
with 30 wt.% H2O2 at 60 ◦C for 24 h and then harvested by filtra-
tion and washed with water. The materials were acidified under
continuous stirring with 50 mL of 0.2 M sulfuric acid at room tem-
perature for 2 h, filtered and dried at 100 ◦C for 4 h to obtain the
respective rp-SBA-15-Pr-SO3H, rd-SBA-15-Pr-SO3H, f-SBA-15-Pr-
SO3H, MCM-41-Pr-SO3H and fume silica-Pr-SO3H samples.
2. Experimental
2.1. Chemicals
2.4. Esterification of oleic acid with either methanol or glycerol
Tetraethyl orthosilicate (TEOS; Fluka) was used as silica pre-
cursor in the synthesis of silica mesoporous material. The tri-block
copolymer pluronic P123 (EO20PO70EO20) (Aldrich), cetyltrimethy-
lammonium bromide (CTAB; Calbiochem) were used as a template
for the synthesis of the mesoporous silica. Hydrochloric acid (HCl;
Merck) was used in the aqueous solution to dissolve the P123 tem-
plate. Heptane (Sigma–Aldrich) and ammonium fluoride (NH4F;
Fluka) were used as swelling and structure controlling agents in
the low temperature syntheses. MPTMS (Aldrich) was used as the
propyl silane precursor and 30 wt.% hydrogen peroxide (H2O2;
Merck) was used as the oxidizing agent. Methanol (Merck), oleic
acid (Aldrich) and glycerol (Merck) were the reactants for the ester-
ification reaction. Finally, commercial Amberlyst-15 in the dry form
(Rohm & Haas, France) was dried in oven at 100 ◦C for 24 h before
use. Fume silica (Aldrich) was used as a non-porous silica for com-
parison with the porous SBA-15 silica.
The esterification of oleic acid with either methanol or glycerol
was performed in a 100-mL round-bottom flask equipped with a
magnetic stirrer and a water-cooled condenser. The oleic acid and
methanol esterification was performed with a 1:9 molar ratio of
oleic acid: methanol at 60 ◦C, whereas the esterification of oleic
acid and glycerol was carried out with a 6:1 molar ratio of oleic
acid:glycerol at 110 ◦C. The reaction time was varied in the range
of 15–180 min for methanol or 15–1440 min for glycerol with 0.5%
(w/w) of catalyst (based on the reactant mass). After completion of
the reaction, the used catalyst was separated from the liquid phase
by centrifugation.
2.5. Characterization
X-ray diffractograms of the catalysts were collected with Cu
K␣ radiation using a Rigaku, Dmax 2200/Ultima+ diffractometer
equipped with a monochromator. The unit cell parameter was
determined from the peaks in the diffractograms. Nitrogen sorp-
tion measurements were performed on a BEL Japan BELSORP-mini
28SP adsorptometer. The specific surface area was calculated using
the BET (Brunauer, Emmett and Teller) method for the relative pres-
sure range of 0.1–0.2, while the pore size distribution and total pore
volume were calculated from the BJH (Barret–Joyner–Halender)
method. Fourier transform infrared spectroscopy (FTIR) was per-
formed in the range of 400–4000 cm−1 with a 4 cm−1 resolution on
a Nicolet 6700 instrument. Prior to the FTIR measurement, the pow-
der samples were vacuum-dried at 100 ◦C for 30 min and embedded
in a KBr pellet. The morphology and size of the anhydrous catalyst
particles were determined by field emission SEM (FE-SEM) using
a JEOL JSM-7001F instrument. In addition, the pore structure of
the four silica morphologies (rd-SBA-15, rp-SBA-15, f-SBA-15 and
MCM-41) and their Pr-SO3H-functionalized derivatives were inves-
tigated by TEM using a JEOL JEM-2100 instrument. The acid group
level (as OH− neutralization capacity) of the functionalized meso-
porous silica was quantified by standard acid–base titration, where
2.2. Preparation of the mesoporous materials
2.2.1. Rope-shaped SBA-15 (Rp-SBA-15)
The synthesis of rope-shaped SBA-15 (rp-SBA-15) was modified
from the previously reported procedure [5] under hydrothermal
conditions. In brief, P123 was dissolved in 2.0 M HCl solution at
room temperature and then TEOS was added to a final molar com-
position of 1.0 TEOS: 0.0165 P123: 6.95HCl: 140H2O. The solution
was kept at 40 ◦C with stirring conditions for 24 h, and then trans-
ferred to a Teflon-lined autoclave for hydrothermal treatment at
100 ◦C for 48 h. The template was removed by calcination at 550 ◦C
in air for 5 h, and the resulting synthesized rope-shaped meso-
porous silica was denoted as rp-SBA-15.
2.2.2. Rod- and fiber-shaped SBA-15 (rd-SBA-15 and f-SBA-15)
Mesoporous silica with large pores and rod- or fiber-shaped
morphologies were synthesized according to the protocols of
Johansson et al. [18,32], where the P123 and NH4F were dissolved
in HCl solutions. Heptane and TEOS were premixed, added to
the micellar solution and stirred at 20 ◦C for various times. After

S. Jeenpadiphat et al. / Journal of Molecular Catalysis A: Chemical 410 (2015) 253–259
255
shifted towards higher diffraction angles. Thus, the lattice param-
eters were decreased.
The specific surface area, total pore volume and pore diameter
of the samples were calculated from the N2 adsorption–desorption
results using the standard BET method and the BJH equation,
respectively, whilst the internal surface area was obtained from the
t-plot equation (see Table 1). The rp-SBA-15 was found to have a
BET surface area of 874 m2/g, with an internal surface area as large
as 820 m2/g and a pore diameter of 9.2 nm. After post-synthesis
functionalization, rp-SBA-15-Pr-SO3H had a BET surface area of
733 m2/g, with an internal surface area of 681 m2/g and a pore
diameter of 8.0 nm, some 1.2-fold smaller than those of the parent
rp-SBA-15 material. The rd-SBA-15, f-SBA-15 and MCM-41 materi-
als plus their Pr-SO3H-functionalized derivatives showed a similar
trend, with a smaller BET surface area (1.19- to 1.37-fold), internal
surface area (1.21- to 1.40-fold) and average pore diameter (1.15-
to 1.20-fold) than that of their respective parent materials.
With respect to the pore size of the materials, rd-SBA-15 and
f-SBA-15 had the largest pore sizes of ∼12.1 nm, followed by rp-
SBA-15 (9.2 nm) and MCM-41 (2.4 nm). This is expected because
of the alkane solvent effect [35]. The decrease in the BET surface
area, internal surface area and pore size after Pr-SO3H function-
alization indicated that the implemented preparation conditions
allowed the organo-sulfonic acid group to be incorporated both into
and on the mesoporous structure and surface at the same time. This
was also indicated by the increasing wall thickness after Pr-SO3H
functionalization (Table 1).
Fig. 1. Representative XRD patterns of mesoporous silica and propylsulfonic-
functionalized mesoporous silica samples.
a known exact mass (∼0.05 g) of the sample was added to 15 mL of
2.0 M NaCl solution and allowed to equilibrate for 30 min. Then, it
was titrated by the dropwise addition of 0.01 M NaOH (aq) [31].
The reaction mixture (supernatant) was analyzed by gas chro-
matography using a Varian CP-3800 gas chromatograph with a
CP-8 column, following the EN 14,105:2003 procedure. The methyl
oleate yield or oleic acid conversion and the mono-, di- and tri-
glyceride (MG, DG and TG, respectively) yields were calculated
based on the eicosane internal standard, and reported as the
mean 1SD from the independent repeats.
Furthermore, from the XRD and nitrogen adsorption analyses, it
was seen that each mesoporous parent sample had a different wall
thickness, resulting in different thermal stabilities. This is the effect
of the different synthesis conditions where the temperature, NH4F
and heptane affected the solubility of the surfactants, resulting in
the large pore size for the rd-SBA-15 and f-SBA-15 [21].
3. Results and discussion
3.1. Characterization of catalysts
The nitrogen adsorption–desorption isotherms (Fig. 2) for the
3 types of parent mesoporous SBA-15 silica and their respective
Pr-SO3H-functionalized derivatives exhibited characteristic type IV
isotherms, with a type 1 hysteresis loop at relative pressures of
around 0.6–0.8, which is typical for this type of material [36,37].
The rd-SBA-15 and f-SBA-15 samples exhibited a hysteresis loop at
higher relative pressures than rp-SBA-15 and MCM-41, indicating
the larger pore size of rd-SBA-15 and f-SBA-15 [38]. The position
of the adsorption branch is related to the pore size distribution;
in particular, the capillary condensation pressure is an increasing
function of the pore diameter. The capillary condensation of the
rp-SBA-15 and its functionalized derivative occurred at a lower rel-
ative pressure (p/p0 = 0.7) than that for rd-SBA-15, f-SBA-15 and
Small angle X-ray diffractograms of the Pr-SO3H-functionalized
mesoporous materials are shown in Fig. 1. All the different shaped
SBA-15 samples showed 3 characteristic peaks, which correspond
to the 100, 110 and 200 preferred orientations, respectively. How-
ever, the positions of the diffraction peaks differed between each
sample. For the untreated materials, the rd-SBA-15 had the largest
unit cell followed by f-SBA-15, rp-SBA-15 and MCM-41, respec-
tively. Comparing the Pr-SO3H-functionalized mesoporous silica
samples (rp-SBA-15-Pr-SO3H, rd-SBA-15-Pr-SO3H, f-SBA-15-Pr-
SO3H and MCM-41-Pr-SO3H) with their respective parent samples
(rp-SBA-15, rd-SBA-15, f-SBA-15 and MCM-41), the diffraction
peaks of the Pr-SO3H-functionalizedSBA-15 materialswere slightly
Table 1
Physicochemical and textural properties of the propylsulfonic acid (Pr-SO3H)-functionalized samples.
d
Catalyst
BET surface
areaa (m2/g)
Internal surface External surface Internal/ Total pore
Average pore
a0 (nm)
Wall thicknessd Acid amounte
areab (m2/g)
areab (m2/g)
external
volumec (cm3/g) diameterc (nm)
(nm)
(mmol/g)
rp-SBA-15
rp-SBA-15-Pr-SO3H
rd-SBA-15
rd-SBA-15-Pr-SO3H
f-SBA-15
f-SBA-15-Pr-SO3H
MCM-41
874
733
407
303
462
387
954
696
53
820
681
236
187
379
313
918
655
–
37
32
193
143
104
93
40
34
–
22.2
21.3
1.2
1.3
3.6
3.4
23.0
19.3
–
1.14
1.04
1.21
1.15
1.21
1.07
0.89
0.50
0.40
–
9.2
8.0
10.9
10.8
13.8
13.1
13.5
13.1
4.7
4.4
–
–
1.7
2.8
1.7
2.6
1.4
2.6
2.3
2.4
–
0.24
1.58
0.32
1.54
0.58
1.28
0.54
1.42
3.00
0.15
12.1
10.5
12.1
10.5
2.4
2.0
30.0
–
MCM-41-Pr-SO3H
Amberlyst-15f
Fume silica-Pr-SO3H 158
–
–
–
–
a
Calculated by the BET method.
b
Internal surface area = (total surface area (from t-plot)) − (external surface area (from t-plot)).
c
d
e
f
Calculated by the BJH method.
√
Wall thickness = (a0) − (pore diameter), where a0 = 2 × d(100)
/
3, d(100) = d-spacing of the (100) reflection from the XRD analysis.
Acid amount obtained by acid–base titration.
Data from the suppliers (Rohm & Haas, France).

256
S. Jeenpadiphat et al. / Journal of Molecular Catalysis A: Chemical 410 (2015) 253–259
(SO2) vibrational mode is clearly visible at around 1350 cm−1 [41].
In the FTIR spectra of the functionalized samples, the peak at
1350 cm−1 was shifted to a higher frequency (1400 cm−1) and the
S
OH stretching vibration of SO3H at 840 cm−1 was also observed.
Moreover, the S C stretching vibration of the propylsulfonic groups
was detected at 603 cm−1 [41]. These 3 characteristic peaks of the
propylsulfonic groups did not appear in the pure SBA-15 samples.
Thus, from the FTIR results it was concluded that the propylsul-
fonic groups were grafted on the surface of the mesoporous silica
samples.
The SEM micrographs revealed different anhydrous morpholo-
gies for all the samples, as shown in Fig. 4a–d. Rp-SBA-15 (Fig. 4a)
appeared as a regular dispersion containing small rod-shaped
particles that agglomerated to a rope-like structure, whilst rp-SBA-
15-Pr-SO3H (not shown) formed a higher aggregation level of small
particles with a rope-like structure, somewhat similar to the parent
rp-SBA-15. The rd-SBA-15 and its Pr-SO3H-functionalized deriva-
tive exhibited the formation of the smallest rod-like structures
(∼600 nm long with 80–200 nm rod diameter), whereas f-SBA-15
and its functionalized derivative showed fiber-like structures. In
addition, the size of the rd-SBA-15 particles was smaller than those
in the rp- and f-SBA-15 materials. On the other hand, MCM-41 had
an ordered rod-like structure of 2.5 × 0.5 ␮m sized particles, while
its functionalized derivative still exhibited the same particle shape
but was more aggregated.
Fig. 2. Representative N2 adsorption/desorption isotherms of mesoporous silica and
propylsulfonic-functionalized mesoporous silica samples.
The cylindrical pore structure in all the samples was investigated
by TEM (Fig. 4e–h), where all the samples showed well-ordered
cylindrical channels. From the TEM image of rp-SBA-15-Pr-SO3H
(Fig. 4(e)), the distance between the pores was estimated to be
close to 10 nm in the case of the SBA-15 samples, which is in
agreement with the unit cell parameter derived from the XRD data
(Table 1). Likewise, the distance between the mesopore channels of
rd-SBA-15-Pr-SO3H and f-SBA-15-Pr-SO3H was larger than that for
rp-SBA-15-Pr-SO3H and MCM-41-Pr-SO3H, which confirmed the
results from the physisorption and XRD analyses.
3.2. Catalytic activity
The esterification of a 1:9 molar ratio of oleic acid:methanol
mixture at 60 ◦C was performed with the prepared catalysts at a
0.5 wt.% catalytic loading, which has previously been found to be
optimal [42]. The catalytic reaction was evaluated in comparison
to two reference reactions; one with the commercial Amberlyst-15
using the same loading level, and one with no catalyst present. The
reaction was evaluated over 180 min during which time the prod-
uct yield increased with time without evidence of having reached
equilibrium or a maximum (Fig. 5). All catalysts displayed some
catalytic activity, significantly increasing the methyl oleate (MO)
yield obtained compared to when no catalyst was used. The rp-SBA-
15-Pr-SO3H catalyst exhibited the highest MO yield, in agreement
with its highest acid site level and largest specific surface area
(Table 1). However, all the Pr-SO3H-functionalized mesoporous sil-
ica catalysts gave a superior MO yield over that obtained with the
commercial Amberlyst-15 catalyst, despite the higher amount of
acid sites and larger pore diameter of the commercial catalyst. This
is likely to be due to their significantly higher specific surface area
and total pore volume.
Furthermore, the fume silica was chosen as non-porous silica
(Supplementary information (SI), Fig. S1 and Fig. S2) for comparison
with porous SBA-15 silica. The result showed that fume silica-Pr-
SO3H exhibited a 11% lower MO yield than the other catalysts,
presumably because of its low surface area and acidity.
The catalytic activity of the Pr-SO3H-functionalized materials
in the esterification process of oleic acid to glycerol was eval-
uated since the molecular flow depends on the size and shape
Fig. 3. Representative FT-IR spectra of the (a) rp-SBA-15, (b) rp-SBA-15-Pr-SO3H,
(c) rd-SBA-15-Pr-SO3H, (d) f-SBA-15-Pr-SO3H and (e) MCM-41-Pr-SO3H samples,
in the spectral region of 400–2000 cm−1
.
their functionalized samples (p/p0 = 0.8), indicating that rp-SBA-15
and its functionalized derivative have a smaller pore diameter than
rd- and f-SBA-15.
The net acid site values of the parent mesoporous silicas were
relatively low (from 0.24 mmol/g for rp-SBA-15 to 0.58 mmol/g for
f-SBA-15), but were significantly increased (2.2- to 6.6-fold) after
Pr-SO3H functionalization (Table 1), due to the incorporation of
the sulfonic acid groups. However, after Pr-SO3H functionalization,
the amount of acid sites was still ∼two-fold lower than that of the
commercial Amberlyst-15 catalyst.
Fig. 3 shows the FTIR spectra of the different mesoporous silica
SBA-15 and Pr-SO3H-functionalized SBA-15 samples. The 4 char-
acteristic peaks of tetrahedral silicon oxide samples were seen
at 467 cm−1 (Si
(Si
O
Si and O Si O bending vibrations), 800 cm−1
O
Si, symmetric vibration), 962 cm−1 (Si O bending vibration)
and the large band at 1090 cm−1 (Si-O-Si, asymmetric vibrations)
[39,40]. In addition, all the samples showed the O H bending
vibration of water at 1640 cm−1. In the functionalized samples,
the strongest bands of -SO3H groups and S C stretching of -
CH2SO3H groups are the best evidence of the successful grafting of
the propylsulfonic groups. In fact, the S O asymmetric-stretching

S. Jeenpadiphat et al. / Journal of Molecular Catalysis A: Chemical 410 (2015) 253–259
257
Fig. 4. Representative SEM images (20,000 × magnification) of (a) rp-SBA-15, (b) rd-SBA-15, (c) f-SBA-15, (d) MCM-41 and TEM images of (e) rp-SBA-15-Pr-SO3H, (f) rd-SBA-
15-Pr-SO3H, (g) f-SBA-15-Pr-SO3H and (h) MCM-41-Pr-SO3H.
1.76 × 1.27 nm2, 0.12 × 0.12 nm2 and 0.37 × 0.29 nm2, respectively.
of both the catalyst and its substrates. Therefore, the methanol
reactant was replaced by the larger glycerol to test the effect
of catalytic pore size on the substrate conversion and product
yield. The structural size (length × wide), determined using the
HyperChem program, of oleic acid, methanol and glycerol were
Furthermore, the structural size of MO, MG, 1,2-DG and TG prod-
ucts were 1.61 × 1.27 nm2, 1.19 × 0.91 nm2, 2.15 × 1.20 nm2 and
3.94 × 2.59 nm2, respectively. The TG product was used as the indi-
cator to study the potential effect of the mesopore size.

258
S. Jeenpadiphat et al. / Journal of Molecular Catalysis A: Chemical 410 (2015) 253–259
starting material (see example in Table S1). Furthermore, the large
pore catalysts (rp-SBA-15-Pr-SO3H, rd-SBA-15-Pr-SO3H, f-SBA-15-
Pr-SO3H and Amberlyst-15) exhibited a higher TG yield than the
small pore catalyst (MCM-41-Pr-SO3H). This result supported that
TG products could not be produced in the MCM-41-Pr-SO3H pores
due to the smaller pore size than the TG structure, and so TG is
only produced at the external surface of the material. The pore size
of the catalyst can hence in part be chosen to select what prod-
uct should be formed during the reaction. The rp-SBA-15-Pr-SO3H
catalyst gave the highest TG yield (64.3%) after a 60 min reaction
time, which was 2.5-fold higher than the yield with the commer-
cial Amberlyst-15 catalyst. When the catalytic efficiencies of the
large pore catalysts were compared, rp-SBA-15-Pr-SO3H showed
a higher TG yield compared to rd-SBA-15-Pr-SO3H and f-SBA-15-
Pr-SO3H. This is presumably due to its higher specific surface area,
internal surface area and acid amount. Both the internal surface
area and acid amount played a major role in catalytic efficiency. In
the first 45 min of reaction time, the rope- and fiber-shaped cata-
lysts, with their high internal/external surface area ratios, provided
a higher oleic conversion and TG yield compared to the rod-shaped
catalyst. However, after 45 min the rod-shaped catalyst contin-
ued to increase the esterification reaction, as did the rope-shaped
catalyst, which is attributed to their high acidic properties. Further-
more, the fiber-shaped mesoporous catalyst (f-SBA-15-Pr-SO3H)
gave a higher TG yield than the rod-shaped catalyst at reaction
times of less than 60 min.
The catalytic activity of rp-SBA-15-Pr-SO3H, Amberlyst-15, f-
SBA-15-Pr-SO3H, rd-SBA-15-Pr-SO3H and MCM-41-Pr-SO3H was
also determined from the% yield of MG, DG and TG with respect to
the number of acidic active sites, in terms of the turnover number
(TON) and turnover frequency (TOF). Here, the highest TON and TOF
of 28.4 and 1.89 (min−1) were obtained with 5 wt.% of the f-SBA-
15-Pr-SO3H catalyst at 15 min (Table 3). This result indicated that
acidic active sites of f-SBA-15-Pr-SO3H reacted fast with the reac-
tants and provided the highest net level of products (MG + DG + TG).
Moreover, this catalyst contained the lowest amount of acid sites
when compared with the other catalysts. However, for TG produc-
tion, the f-SBA-15-Pr-SO3H catalyst formed a lower TG yield than
that with rp-SBA-15-Pr-SO3H.
Fig. 5. Time course of the esterification of a 1:9 molar ratio of oleic acid:methanol
at 60 ◦C for 15–180 min with 0.5 wt.% of the different propylsulfonic acid (Pr-SO3H)-
functionalized mesoporous silica catalysts or the commercial Amberlyst-15 one,
compared to that without a catalyst.
The oleic acid and glycerol esterification was performed with
a 6:1 mole ratio of oleic acid:glycerol over a 5 wt.% catalyst load-
ing (based on total reactant weight) at 110 ◦C for 15–1440 min. The
reaction time was found to have a major influence on the obtained
conversion level and product yield, where short reaction times
(15–30 min) were insufficient to complete the esterification reac-
tion (Table 2). In fact, the esterification of glycerol (a tri-hydroxy
alcohol) with oleic acid encompasses multiple sequential esterifi-
cation reactions involving the oleic acid acyl group protonation and
glycerol hydroxyl group attraction to produce the MG followed by
esterifying the remaining hydroxyl group to DG and TG.
As the reaction time was extended, the esterification of MG
to DG and to TG was increased and the MG yield decreased. This
result can be explained by the increased nucleophilic attacks of
the remaining hydroxyl groups of the MG moiety on the acid site-
oleic acid complex on the catalyst surface. For longer reaction times
of 45–60 min, the highest oleic acid conversion level to TG was
obtained at about 99% with the rp-SBA-15-Pr-SO3H, Amberlyst-
15 and f-SBA-15-Pr-SO3H catalysts, and at 98.7% and 91.5% for
the rd-SBA-15-Pr-SO3H and MCM-41-Pr-SO3H, respectively. This
showed that the reaction can be completed within 60 min, a time at
which the equilibrium point of the reaction seemed to be attained,
if a highly efficient catalyst was used. However, the TG yield then
decreased with increasing reaction time beyond 45–60 min, due to
the backward reaction to form the intermediate MG and DGs, and
It should also be noted that the acid amount of the catalyst
depends on the number of free silanol groups of the mesoporous
parent. In case of the f-SBA-15 parent sample, the acid amount
was higher than the rod- and rope-shaped samples, perhaps due to
an easier ion-exchange capability of protons at the silanol groups.
However, after some hydroxyl groups were changed to thiol groups,
Table 2
The oleic acid conversion level and triglyceride (TG) yield in the esterification of a 6:1 oleic acid:glycerol molar ratio reaction catalyzed by a 5 wt.% catalyst loading (based on
total reactant weight) of propylsulfonic acid (Pr-SO3H)-functionalized mesoporous silica or the commercial Amberlyst-15 catalyst at 110 ◦C for up to 1440 min of reaction
time.
Reaction
Catalyst
time (min)
No catalyst
Amberlyst-
15
Oleic acid
conversion
(%)
rp-SBA-15-
Pr-SO3H
Oleic acid
conversion
(%)
f-SBA-15-
Pr-SO3H
Oleic acid
conversion
(%)
rd-SBA-15-
Pr-SO3H
Oleic acid
conversion
(%)
MCM-41-
Pr-SO3H
Oleic acid
conversion
(%)
Oleic acid
conversion
(%)
TG
yield
(%)
TG
yield
(%)
TG
yield
(%)
TG
yield
(%)
TG
yield
(%)
TG
yield
(%)
15
30
45
–
–
–
–
9.3
16.9
22.9
25.2
32.9
72.8
–
–
–
–
22.5
83.3
92.3
99.3
96.9
89.4
69.3
43.1
40.2
31.5
–
3.1
89.1
96.6
98.4
99.7
97.7
86.5
72.6
65.1
61.4
60.5
45.4
50.7
53.3
64.3
31.2
30.3
29.6
28.9
27.3
27.1
88.0
96.6
99.0
64.2
55.3
49.4
44.7
47.4
33.8
27.5
22.9
40.7
40.2
34.4
33.9
29.8
25.6
24.8
15.3
10.6
82.5
90.4
91.4
98.7
88.8
54.7
47.8
44.6
40.7
36.9
17.4
24.8
29.9
34.2
37.8
22.8
21.1
20.9
19.7
18.1
70.2
77.3
86.2
91.5
96.4
90.9
81.2
50.4
46.4
40.0
–
3.9
15.2
25.8
44.4
37.9
20.1
20.0
17.9
11.3
11.2
13.2
14.7
21.6
30.5
15.7
13.8
11.3
60
120
180
360
540
720
1440
–
0.4
1.1
1.9
2.5
14.3
Data are shown as the mean 1SD, derived from 3 independent repeats.

S. Jeenpadiphat et al. / Journal of Molecular Catalysis A: Chemical 410 (2015) 253–259
259
Table 3
The turnover number (TON) and turnover frequency (TOF) from the esterification of a 6:1 oleic acid:glycerol molar ratio reaction catalyzed by a 5 wt.% catalyst loading (based
on total reactant weight) of propylsulfonic acid (Pr-SO3H)-functionalized mesoporous silica or the commercial Amberlyst-15 catalyst at 110 ◦C for up to 1440 min of reaction
time.
Reaction time
(min)
Catalyst
Amberlyst-15
TON
rp-SBA-15-Pr-SO3H
TON
f-SBA-15-Pr-SO3H
TON
rd-SBA-15-Pr-SO3H
TON
MCM-41-Pr-SO3H
TON
TOF (min−1
)
TOF (min−1
)
TOF (min−1
)
TOF (min−1
)
TOF (min−1
)
15
30
45
3.9
11.4
10.5
10.4
9.4
8.8
7.3
4.5
4.2
0.26
0.38
0.23
0.17
0.08
0.05
0.02
0.01
0.01
0.00
19.0
19.7
19.5
17.3
18.9
16.6
14.1
12.7
12.1
11.9
1.26
0.66
0.43
0.29
0.16
0.09
0.04
0.02
0.02
0.01
28.4
24.9
23.7
15.4
12.1
11.0
10.1
11.5
8.8
1.89
0.83
0.53
0.26
0.10
0.06
0.03
0.02
0.01
0.01
24.1
23.3
21.7
22.9
17.9
11.1
9.8
9.0
8.1
7.4
1.61
0.78
0.48
0.38
0.15
0.06
0.03
0.02
0.01
0.01
23.4
24.6
26.1
26.9
27.6
25.1
21.1
14.4
13.3
11.5
1.56
0.82
0.58
0.45
0.23
0.14
0.06
0.03
0.02
0.01
60
120
180
360
540
720
1440
3.5
7.6
disulfide linkages may form in the f-SBA-15-Pr-SH structure, gener-
ating the less active sulfonic group in the f-SBA-15-Pr-SO3H sample
[39]. Therefore, the catalytic performance in esterification reaction
depends on both the particle shape and acidity of the mesoporous
material.
[3] A. Taguchi, F. Schüth, Microporous Mesoporous Mater. 77 (2005) 1–45.
[4] N. Igarashi, K.A. Koyano, Y. Tanaka, S. Nakata, K. Hashimoto, T. Tatsumi,
Microporous Mesoporous Mater. 59 (2003) 43–52.
[5] D. Zhao, J. Feng, Q. Huo, N. Melosh, G.H. Fredrickson, B.F. Chmelka, G.D.
Stucky, Science 279 (1998) 548–552.
[6] J. Fan, J. Lei, L. Wang, C. Yu, B. Tu, D. Zhao, Chem. Commun. 17 (2003)
2140–2141.
[7] J. Lei, J. Fan, C. Yu, L. Zhang, S. Jiang, B. Tu, D. Zhao, Microporous Mesoporous
Mater. 73 (2004) 121–128.
[8] J. Sun, H. Zhang, R. Tian, D. Ma, X. Bao, D.S. Su, H. Zou, Chem. Commun. 8
(2006) 1322–1324.
4. Conclusion
We show that the physical and chemical properties of meso-
porous silica materials can be modified by incorporation of an
organo-sulfonic acid group by post-synthesis grafting. The suc-
cessful incorporation of Pr-SO3H functional groups into and on the
mesoporous silica materials was confirmed by XRD, nitrogen sorp-
tion and acid titration analyses. In terms of the catalytic ability,
the esterification of methanol or glycerol with oleic acid using the
Pr-SO3H-functionalized mesoporous silica catalysts revealed that
the heterogeneous acidic catalysts had a higher catalytic activ-
ity due to the presence of the sulfonic acid groups. Furthermore,
it was observed that the substrate conversion and product yield
depended not only on the acid strength, but also on the specific
surface area, catalyst pore size and product molecular size. The
rp-SBA-15-Pr-SO3H catalyst exhibited the highest catalytic activity
for the esterification of oleic acid with either methanol or glyc-
erol, and all mesoporous supports were superior to the commercial
Amberlyst-15 catalyst and non-porous silica.
[9] M. Kruk, L. Cao, Langmuir 23 (2007) 7247–7254.
[10] H. Zhang, J. Sun, D. Ma, X. Bao, A. Klein-Hoffman, G. Weinberg, D. Su, R.
Schlögl, J. Am. Chem. Soc. 126 (2004) 7440–7441.
[11] H. Zhang, J. Sun, D. Ma, G. Weinberg, D.S. Su, X. Bao, J. Phys. Chem. B 110
(2006) 25908–25915.
[12] J. Sun, H. Zhang, D. Ma, Y. Chen, X. Bao, A. Klein-Hoffman, N. Pfänder, D.S. Su,
Chem. Commun. (2005) 5343–5345.
[13] J. Sun, D. Ma, H. Zhang, C. Wang, X. Bao, D.S. Su, A. Klein-Hoffman, G.
Weinberg, S. Mann, J. Mater. Chem. 16 (2006) 1507–1510.
[14] J. Sun, D. Ma, H. Zhang, F. Jiang, Y. Cui, R. Guo, X. Bao, Langmuir 26 (2008)
2372–2380.
[15] Z. Jin, X. Wang, X. Cui, Colloids Surf. A 316 (2008) 27–36.
[16] X.Y. Bao, X.S. Zhao, J. Phys. Chem. B 109 (2005) 10727–10736.
[17] A. Sayari, B. Han, Y. Yang, J. Am. Chem. Soc. 126 (2004) 14348–14349.
[18] E.M. Johansson, M.A. Ballem, J.M. Córdoba, M. Odén, Langmuir 27 (2011)
4994–4999.
[19] E.M. Björk, F. Söderlind, M. Odén, Langmuir 29 (2013) 13551–13561.
[20] C. Yu, J. Fan, B. Tian, D. Zhao, G.D. Stucky, Adv. Mater. 14 (2002) 1742–1745.
[21] E.M. Björk, F. Söderlind, M. Odén, J. Colloid Interface Sci. 413 (2014) 1–7.
[22] P. Schmidt-Winkel, P. Yang, D.I. Margolese, B.F. Chmelka, G.D. Stucky, Adv.
Mater. 11 (1999) 303–307.
[23] D. Zhao, J. Sun, Q. Li, G.D. Stucky, Chem. Mater. 12 (2000) 275–279.
[24] S. Miao, B.H. Shanks, J. Catal. 279 (2011) 136–143.
[25] A. C¸ itak, B. Erdem, S. Erdem, R.M. Öksüzog˘lu, J. Colloid Interface Sci. 369
(2012) 160–163.
Acknowledgements
[26] L. Sherry, J.A. Sullivan, Catal. Today 175 (2011) 471–476.
[27] A. Stein, B.J. Melde, R.C. Schroden, Adv. Mater. 12 (2000) 1403–1419.
[28] K. Nakajima, I. Tomita, M. Hara, S. Hayashi, K. Domen, J.N. Kondo, Catal. Today
116 (2006) 151–156.
[29] G.J.A.A. Soler-Illia, C. Sanchez, B. Lebeau, J. Patarin, Chem. Rev. 102 (2002)
4093–4138.
[30] A.P. Wight, M.E. Davis, Chem. Rev. 102 (2002) 3589–3614.
[31] I.K. Mbaraka, D.R. Radu, V.S.-Y. Lin, B.H. Shanks, J. Catal. 219 (2003) 329–336.
[32] E.M. Johansson, J.M. Córdoba, M. Odén, Microporous Mesoporous Mater. 133
(2010) 66–74.
The authors would like to thank the Grant for International
Research Integration: Chula Research Scholar, Ratchadaphisek-
somphot Endowment Fund, the Post-doctoral Fellowship
(Ratchadaphiseksomphot Endowment Fund) Chulalongkorn
University, the Knut and Alice Wallenberg Foundation (KAW
2012.0083), and Nanolith Sverige AB for financially supporting this
project.
[33] K.N. Tayade, M. Mishra, J. Mol. Catal. A Chem. 382 (2014) 114–125.
[34] S.S. Reddy, B.D. Raju, V.S. Kumar, A.H. Padmasri, S. Narayanan, K.S.R. Rao,
Catal. Commun. 8 (2007) 261–266.
Appendix A. Supplementary data
[35] E.M. Johansson, J.M. Córdoba, M. Odén, Mater. Lett. 63 (2009) 2129–2131.
[36] B. Rác, Á. Molnár, P. Forgo, M. Mohai, I. Bertóti, J. Mol. Catal. A Chem. 244
(2006) 46–57.
[37] M.S. Khayoon, B.H. Hameed, Appl. Catal. A Gen. 433–434 (2012) 152–161.
[38] Analysis software user’s manual, BELSORP, BEL, JAPAN, INC, 57.
[39] R. Buzzoni, S. Bordiga, G. Ricchiardi, G. Spoto, A. Zecchina, J. Phys. Chem. 99
(1995) 11937–11951.
[40] A.S. Cattaneo, C. Ferrara, D.C. Villa, S. Angioni, C. Milanese, D. Capsoni, S.
Grandi, P. Mustarelli, V. Allodi, G. Mariotto, S. Brutti, E. Quartarone,
Microporous Mesoporous Mater. 219 (2016) 219–229.
[41] G. Socrates, Infrared and Raman Characteristic Group Frequencies, third ed.,
John Wiley & Sons, Chichester, 2001, pp. 209–228.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.molcata.2015.10.
002.
References
[1] T.M. Suzuki, T. Nakamura, E. Sudo, Y. Akimoto, K. Yano, Microporous
Mesoporous Mater. 111 (2008) 350–358.
[2] A. Galarneau, D. Desplantier-Giscard, F. Di Renzo, F. Fajula, Catal. Today 68
(2001) 191–200.
[42] S. Jeenpadiphat, D.N. Tungasmita, Powder Technol. 237 (2013) 634–640.