Pillared HMCM-36 zeolite catalyst for biodiesel production by esterification of palmitic acid

Article abstract of DOI:10.1016/j.molcata.2015.06.006

Layered MWW zeolite has been studied for swelling/pillaring using CTAB and silica as swelling and pillaring reagents, respectively to synthesize pillared MCM-36 material. The swelling/pillaring efficiency was evaluated based on their X-ray diffraction, N₂-physisorption, scanning electron microscopy and FTIR spectra after pyridine adsorption as indicator for acidity measurements and catalytic potential. There was an overall decrease in acid site concentration due to incorporation of inert silica pillars. However, after ion exchange, mesoporous HMCM-36 zeolite with the highest BET surface area showed increased Br?nsted and Lewis acid sites compared to the MCM-22, suggesting enhanced accessibility of acid sites for bulky reacting molecules. MCM-22 and MCM-36, as well as the ion exchanged HMCM-22 and HMCM-36 samples were tested for esterification of palmitic acid with methanol. The HMCM-36 catalyst showed high activity in palmitic acid esterification with methanol. This catalyst can be readily separated from the reaction system for re-use for at least four cycles without losing any activity suggesting potential industrial applications in biodiesel synthesis.

Full text of DOI:10.1016/j.molcata.2015.06.006

Journal of Molecular Catalysis A: Chemical 406 (2015) 159–167
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Pillared HMCM-36 zeolite catalyst for biodiesel production by
esterification of palmitic acid
Ralitsa Purovab, Katabathini Narasimharaoa,∗, Nesreen S.I. Ahmeda, Shaeel Al-Thabaitia,
Abdulmohsen Al-Shehria, Mohammed Mokhtara,∗, Wilhelm Schwiegerb
a Department of Chemistry, Faculty of Science, King Abdulaziz University, P.O. Box 80203, Jeddah 21589, Saudi Arabia
b Institute of Chemical Reaction Technology, University of Erlangen-Nuremberg, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Layered MWW zeolite has been studied for swelling/pillaring using CTAB and silica as swelling and
pillaring reagents, respectively to synthesize pillared MCM-36 material. The swelling/pillaring efficiency
was evaluated based on their X-ray diffraction, N2-physisorption, scanning electron microscopy and FTIR
spectra after pyridine adsorption as indicator for acidity measurements and catalytic potential. There
was an overall decrease in acid site concentration due to incorporation of inert silica pillars. However,
after ion exchange, mesoporous HMCM-36 zeolite with the highest BET surface area showed increased
Brönsted and Lewis acid sites compared to the MCM-22, suggesting enhanced accessibility of acid sites
for bulky reacting molecules. MCM-22 and MCM-36, as well as the ion exchanged HMCM-22 and HMCM-
36 samples were tested for esterification of palmitic acid with methanol. The HMCM-36 catalyst showed
high activity in palmitic acid esterification with methanol. This catalyst can be readily separated from
the reaction system for re-use for at least four cycles without losing any activity suggesting potential
industrial applications in biodiesel synthesis.
Received 19 March 2015
Received in revised form 3 June 2015
Accepted 4 June 2015
Available online 6 June 2015
Keywords:
Pillared HMCM-36
MCM-22(P)
Biodiesel
Esterification
Palmitic acid
© 2015 Elsevier B.V. All rights reserved.
1. Introduction
partially crystalline nanosized ZSM-5 zeolites with high surface
area and meso–macro pore volume [5]
Zeolites are solid catalysts which are used commercially in many
industrial processes [1]. Zeolites can be used as shape-selective
catalysts due to their well-defined pores with minimum kinetic
diameters similar to the size of small organic molecules [2]. How-
ever, for some chemical processes, these catalysts are required to
catalyze reactions involving larger, bulkier molecules. Researchers
studied two classes of materials to fulfill this requirement: (i)
mesoporous M41S-type materials and (ii) pillared or delaminated
layered materials. The M41S family has been extensively inves-
tigated regarding the formation, characterization of their porous
structure and catalytic applications [3]. However, these materials
possess significantly weaker acid sites than the acid sites presented
in micro porous zeolites [4]. Thus, materials with acid sites of
strength similar to those in zeolites, which are accessible to large
organic molecules, are of general interest. We recently synthesized
Pillared layered structures are built of inorganic layers with
inorganic or organic pillars appended on both sides of the sheets
[6]. These materials are potentially most attractive for catalysis,
because they combine high specific surface areas and good acces-
sibility for larger molecules to a large number of catalytic sites
[7]. MCM-22 zeolite with its MWW structure and acidic properties
holds interesting opportunities for different structure modifica-
tions and catalytic applications [8].
Currently, there is an enormous interest in the catalytic ester-
ification reaction because of its application in several industrial
processes [9]. One of the main products obtained by esterification of
long chain fatty acids is biodiesel, an attractive bio-renewable fuel
which has environmental benefits over conventional petroleum
based fuels [10]. For an alkali catalyzed transesterification, the glyc-
erides and alcohol must be free from water and FFAs since the
presence of presence of free fatty acids (FFAs) causes deactivation
of homogeneous alkali catalysts such as NaOH and KOH by forming
soaps and creating difficulties to separate the products [11]. Tak-
ing into account that some of the natural vegetable oils or animal
fats contain considerable amounts of FFAs, which interfere with the
transesterification process and must be converted into their corre-
∗
Corresponding authors. Fax: +966 26952292.
E-mail addresses: katabathini@yahoo.com, nkatabathini@kau.edu.sa
(K. Narasimharao), mmokhtar2000@yahoo.com (M. Mokhtar).
http://dx.doi.org/10.1016/j.molcata.2015.06.006
1381-1169/© 2015 Elsevier B.V. All rights reserved.

160
R. Purova et al. / Journal of Molecular Catalysis A: Chemical 406 (2015) 159–167
sponding esters before reaction, pre-esterification appears as an
essential step in the production of biodiesel from acid oils [12].
Esterification is normally carried out in the homogeneous
phase in the presence of acid catalysts such as sulfuric and
p-toluene sulfonic acids. This pretreatment step has been suc-
cessfully demonstrated using sulfuric acid [13]. Unfortunately,
use of the homogeneous sulfuric acid catalyst adds neutraliza-
tion and separation steps as well as the esterification reaction to
the process and also has several drawbacks such as equipment
corrosion, difficulty of handling, and problems separating the prod-
ucts from the catalysts [11]. The use of heterogeneous catalysts
can be an alternative to reduce biodiesel cost. Various hetero-
geneous catalysts such as zeolites (HUSY, HBEA, HMOR, HZSM-5
and HMCM-22), sulfated oxides (SnO2, ZrO2, Nb2O5 and TiO2),
heteropolyacids (12-tungstophosphoric acid) and commercial sul-
fonic resin (Amberlyst-15) were used for esterification of FFAs [14].
However, catalysts with micropores are not suitable for biodiesel
production because the micropores limit the diffusion of large
molecules with long alkyl chains [15]. Conceptually, the high acid
strength and uniform mesopores offer an unprecedented tool to
control catalytic conversion in acid catalyzed reactions [16].
The catalytic activity of MCM-22 and pillared MCM-36 material
has not been systematically explored for fatty acid esterifica-
tion, hence the present work seeks to improve our fundamental
understanding of structure, texture and acidity relations and the
concomitant catalytic activity of MCM-22 and pillared MCM-36
solid acids in the low temperature esterification of palmitic acid
with methanol for application in biodiesel production.
at 75 ◦C overnight. Pillaring was carried out with 1 g of solid dried
swollen material and 6 g of tetraethyl orthosilicate (TEOS, 98% Alfa
Aesar). The mixture was then placed in an oil bath at 80 ◦C and
stirred for 24 h. After what the samples were removed and filtered
using Whatmann’s filter paper and dried in an oven at 30 ◦C for 12 h
to remove the TEOS excess. Dried samples were then mixed with
water in a 1:10 weight ratio to undergo hydrolysis for 5 h at 40 ◦C,
after that they were filtered by gravity and dried at 30 ◦C. Pillared
samples were calcined in a muffle oven using two steps calcina-
tion procedure. In the first step heating rate was set at 1 ◦C per min
to 450 ◦C under nitrogen stream and this temperature was main-
tained for 6 h. Finally the samples were kept at 550 ◦C under air for
12 h (temperature ramp rate of 2 ◦C per min).
2.1.4. Ion exchange
MCM-22 and MCM-36 after calcination were ion-exchanged
with an excess of 1 mol L−1 NH4NO3 aqueous solution (liquid-
to-solid ratio of 10 cm3 g−1) and stirred continuously at room
temperature for 8 h. The pH of the mixture was adjusted to 7 with
NH4OH. This procedure was repeated twice. After filtering and
+
drying, the resulting NH4 forms of MCM-22 and MCM-36 were
calcined at 500 ◦C for 5 h in air to obtain HMCM-22 and HMCM-36
samples.
2.2. Characterization of materials
The elemental composition of the catalysts was determined by
ICP-MS, Optima 7300DV, PerkinElmer Corporation, USA. The sam-
ple preparation for ICP-MS is as follows; about 100 mg of catalyst
was placed in a PTFE beaker and then complete dissolution of the
sample was achieved by adding 8 mL of 40% HF, 2 mL of HNO3
and 2 mL HCl and to this 15 mL of ultra-pure water was added
and then PTFE beaker was placed in ultrasonic bath for 10 min
to obtain homogeneous dissolution. The solution was then rinsed
into a centrifuge tube and centrifuged at 3000 rpm for 3 min. The
clear supernatant was decanted and used to prepare 250 mL stock
solution.
XRD diffraction patterns were obtained from X’pert Pro diffrac-
tometer from Phillips Analytical. CuK␣ radiation (ꢀ = 1.54056 Å)
was used. The samples were measured in sample holders with a
smaller exposure area. Diffraction patterns of the samples after
synthesis, swelling and calcination were measured by using the
following program: 2ꢁ angle from 2 to 30◦ with a step of 0.02◦ and
increasing duration of 5 s per step.
Morphology of the samples (scanning electron microscopy
images) was determined by using Carl Zeiss ULTRA55 microscope
equipment at a 3 kV voltage and magnifications at magnitude
10,000.
Nitrogen-sorption measurements were performed by first car-
rying out a pre-treatment of the samples. During pre-treatment the
samples were heated with a 1 ◦C per min rate to 250 ◦C under high
vacuum, and kept at these conditions for 12 h. Analysis measure-
ments were performed using Micromeritics ASAP 2010 instrument
at a temperature of −196 ◦C.
2. Experimental
2.1. Synthesis of materials
2.1.1. MCM-22(P)
Layered precursor, MCM-22(P) was hydrothermally synthe-
sized using hexamethylenimine (HMI, 99% Aldrich), Aerosil 200
(Degussa), sodium aluminate (53% Al2O3, 43% Na2O, Riedel de
Haen), sodium hydroxide (25% solution prepared from pure pel-
lets, Merck) and deionized water. In a typical synthesis method,
solution ‘A’ was prepared by dissolving NaAlO2 (1.8 g) in water
(200.2 g) and solution ‘B’ by mixing 25% NaOH solution (20.1 g)
and water (348.1 g) and stirred for 10 min. Then both A and B were
mixed, HMI (41.0 g) added to the solution and stirred for 45 min.
After that the solution was put in water bath at 50 ◦C and Aerosil
(49.87 g) was added on portions under vigorous stirring. Finally 40 g
of H2O was added to the synthesis mixture and stirred for 2 h. The
chemical composition of the final gel was: 1 SiO2: 0.09 Na2O: 0.5
HMI: 45 H2O: 0.01 Al2O3. The synthesis was carried out in 1 L auto-
clave, stirring speed 600 rpm, crystallization temperature and time
were 135 ◦C and 8 days, respectively. The crystalline product was
collected by centrifugation at 10,000 rpm, washing with distilled
water and subsequently dried at 75 ◦C overnight.
2.1.2. Calcination of MCM-22(P) to MCM-22
MCM-22 sample was obtained by calcination of MCM-22(P) at
550 ◦C (1 ◦C min−1 rate) for 5 h under the flow of air.
DRIFT spectra of calcined catalysts obtained at room tempera-
ture using Perkin-Elmer Spectrum 100 FTIR spectrometer. Then, the
samples were subjected to a pyridine adsorption analysis. The anal-
ysis was carried out over a catalyst disk which was treated at 100 ◦C
under vacuum for 5 h. Later, the sample was treated with pyridine
vapor and finally heated at 100 ◦C under vacuum for 30 min. DRIFT
spectra were collected at room temperature. The amount of Brön-
sted and Lewis acid sites was calculated via integration of the area
of the absorption bands showing the maximum values of intensity
at 1446 and 1536 cm−1, respectively. Integrated absorbance of each
band was obtained using the appropriate software by applying the
2.1.3. Swelling and pillaring of MCM-22(P) to synthesize MCM-36
MCM-22(P) was swollen following the procedure reported by
Corma et al. and Maheshwari et al. [8]. The layered precursor was
mixed with hexadecyltrimethylammonium bromide (CTAB, 98%
Sigma–Aldrich), tetrapropylammonium hydroxide (TPAOH, 40%
Süd Chemie) and water in a weight ratio of 1.8 g MCM-22(P): 10.1 g
CTAB: 4.4 g TPAOH: 38.6 g H2O. The mixture was allowed to stir for
48 h at 40 ◦C after that the particles were recovered by repeated
cycles of centrifugation, washing with distilled water and drying

R. Purova et al. / Journal of Molecular Catalysis A: Chemical 406 (2015) 159–167
161
corresponding extinction coefficient and normalized by the weight
of the samples.
3000
2500
2000
1500
1000
500
Ammonia temperature programmed desorption (TPD) mea-
surements were performed to titrate the total number of acid sites,
using CHEMBET-3000 (Quantachrome, USA) equipped with TCD
detector, PID-controlled furnace heated flow microreactor. Calcu-
lated amount of catalyst sample (250 mg) was pretreated at 120 ◦C
for 2 h under helium gas flow (80 mL min−1). The gas flow was
switched to ammonia to saturate the sample for 1 h and then
the sample flushed by helium gas at 100 ◦C for 2 h to remove
physisorbed ammonia prior to recording the TCD signal. The des-
orption patterns were recorded by ramping the temperature of the
sample to 800 ◦C at a rate of 10 ◦C min−1 under a steady flow of
helium gas (80 mL min−1). The amount of ammonia evolved was
determined by comparing the areas desorbed from the sample with
those of known amounts of injected ammonia.
MCM-22
MCM-36
HMCM-22
HMCM-36
0
2
4
6
8
10 12 14 16 18 20 22 24 26 28 30
2θ (degrees)
Fig. 1. XRD patterns of the samples.
2.3. Esterification of palmitic acid
Table 1
Surface area and pore volume of the samples from N2-adsorption measurements.
Esterificiation reactions were performed in a stirred batch reac-
tor with samples withdrawn periodically for analysis using a
Shimadzu GC17A Gas Chromatograph fitted with a DB1 capillary
column (film thickness 0.25 mm, id 0.32 mm, length 30 m). Reac-
tions were performed at 70 ◦C using 0.05 g of catalyst, 5 mmol of
palmitic acid, 0.0025 mol (0.587 mL) hexyl ether as internal stan-
dard, and 0.3036 mol (12.5 mL) methanol. Reaction profiles were
followed for a period of 6 h and continued for 24 h to assess limiting
conversions.
Sample
Surface area (m2 g−1
)
Pore volume (cm3 g−1
)
SBET
Smicro
Smeso
Vtotal
Vmicro
Vmeso
MCM-22
HMCM-22
MCM-36
HMCM-36
492
449
907
635
399
332
385
205
93
117
522
430
0.52
0.58
0.97
0.85
0.16
0.11
0.19
0.04
0.36
0.47
0.78
0.81
Scanning electron microscopy (SEM) was employed for the
direct visualization of crystal morphology and layer structure of
the samples. SEM images of MCM-22 and MCM-36 samples before
and after ion exchange are shown in Fig. 2. Morphology of MCM-
22 samples indicating that the material consists of thin irregular
platelets that are about 500–1000 nm in diameter and 50–100 nm
in thickness. MCM-36 samples shows preservation of crystal mor-
phology after the swelling and pillaring process. The crystals retain
their platelets morphology and sharp facets as seen in Fig. 2 (the
crystals are oriented along their thinner edge in this image). Pil-
lared MCM-36 samples had more highly agglomerated parts than
MCM-22 samples; this is probably due to condensation of sur-
face Si OH groups after thermal treatment. The SEM images of
the MCM-22 samples clearly showed platelet morphology. In con-
trast, MCM-36 samples shows some large particles are formed by
the aggregation of these platelets with distinctive platelet struc-
ture. He et al. [7] attributed this change of morphology to the silica
pillars between layers of swollen MCM-22 resulting in increase of
the platelet thickness. The XRD patterns are clearly indicating that
the loss of crystallinity in case of MCM-36 samples compared to
MCM-22 samples. This is due to fact that these samples possessed
more silica than the MCM-22 samples. The observations from the
XRD measurements agree well with the changes in morphology
revealed by SEM analysis.
Nitrogen adsorption-desorption isotherms and BJH pore size
distributions of the samples are shown in Fig. 3(A) and (B), respec-
tively. The MCM-22 sample showed a type I isotherm in accordance
with the microporous nature of the material, on the other hand,
pillared MCM-36 showed type IV isotherm with a hysteresis loop
at P/P◦ = 0.4 for capillary condensation. This observation suggests
that the swelling and subsequent pillaring led to the formation
of mesopores. The textural properties of the two samples are pre-
sented in Table 1. The BET surface area (SBET) of pillared MCM-36
was substantially higher than that of MCM-22. The micropore vol-
ume (0.16 cm3 g−1) for MCM-22 sample corresponds only to 30% of
the total pore volume (0.52 cm3 g−1) and also the hysteresis loop of
this sample can be observed at p/p0 > 0.90 with asymptotic growth
to p/p0 → 1. This hysteresis loop can be indicative of the filling of
3. Results and discussion
Lamellar MWW zeolite precursor such as MCM-22(P) offers
unprecedented opportunities for creating diversity of more open
zeolite structures by expanding and modifying the interlamellar
space (as shown in Scheme 1) [17]. The swelling and pillaring pro-
cedures are easy, least destructive and most efficient expansion of
layered zeolites. In this study, CTAB solution with high pH was used
to provide swollen MCM-22 that can be converted to the pillared
MCM-36 with large pores. High pH of the swelling solution was
obtained by addition of tetrapropylammonium hydroxide to the
CTAB solution.
ICP-MS analysis was used to determine the chemical compo-
sition of the synthesized precursor. The analysis revealed that
Si/Al ratio of the MCM-22(P) sample is 34.4, which was less than
the starting composition of the synthesis substrate mixture of
Si/Al = 50. Thus, these results suggest that the incorporation of sili-
con was smaller than that expected considering its amount present
in the synthesis gel.
Fig. 1 shows the powder XRD patterns of MCM-22 and MCM-
36 samples before and after the ion exchange. The as-synthesized
MCM-22 is a two-dimensional layer with a hexagonal pore struc-
ture (Scheme 1). The XRD patterns of MCM-22 samples show a
small characteristic reflection (0 0 2) at 6.58◦, which corresponds to
the unit cell parameter ‘c’. This peak disappears in the XRD pattern
for swollen and pillared MCM-36 samples. In addition, the pillared
MCM-36 samples show (0 0 1) peak around 2ꢁ = 2◦ corresponding
to an interlayer distance of about 4.5 nm, which indicates success-
ful pillaring [7]. The two diffraction peaks corresponding to (1 0 1)
and (1 0 2) reflections appeared distinctively in MCM-22 samples,
in contrast a broad band can be observed in this region for the MCM-
36 samples, which is an indication of loss of registry along the c-axis
due to the introduction of pillars [19]. The XRD patterns of MCM-22
and MCM-36 samples are in good agreement with those previously
published report [20].

162
R. Purova et al. / Journal of Molecular Catalysis A: Chemical 406 (2015) 159–167
Scheme 1. Schematic representation of MCM-22(P) and its derivatives [18].
Fig. 2. SEM images of the samples.
the space among the particles with the adsorbate (N2 gas) and
not to the filling of mesopores. On the other hand, the MCM-36
sample showed total pore volume (0.97 cm3 g−1) with mesopores
of 0.78 cm3 g−1. The changes of micropore volume, Vmicro and
mesopore volume, Vmeso reflect structural changes caused by the
swelling followed by pillaring.
reported by Lasperas et al. [21]. The authors suggested that the loss
of surface area was caused by breakage of Si
to pore collapse.
O Si bonds leading
The BJH pore size distribution for the two samples is shown in
Fig. 3(B). The results presented in Fig. 3(B) clearly indicating that
the MCM-36 samples has mesopores, on other hand major poros-
ity exists in the micropore region for the MCM-22 samples. The
mesoporous distribution formed with a cylindrical or slit-shaped
pore model shows an intense maximum at 2.5–3.0 nm in MCM-36
A small decrease in the surface area (from 492 to 449 m2 g−1 in
case of MCM-22) was observed after ion exchange. The difference is
high in case of pillared MCM-36 sample. A similar observation was

R. Purova et al. / Journal of Molecular Catalysis A: Chemical 406 (2015) 159–167
163
2
4
6
8
10
12
14
0.0
0.2
0.4
0.6
0.8
1.0
600
(B)
HMCM-36
MCM-36
HMCM-22
MCM-22
(A)
HMCM-36
0.10
0.05
300
600
0.00
400
MCM-36
300
200
100
0.004
300
600
0.03
0.02
0.01
HMCM-22
300
100.00
800
600
400
200
0
MCM-22
0.8
600
300
2
4
6
8
10
12
14
0.0
0.2
0.4
0.6
1.0
Pore diameter [nm]
P/Po
Fig. 3. (A) N2 adsorption-desorption isotherms and (B) BJH pore size distribution of the samples.
Fig. 4. FTIR spectra of the samples in the region of (A) 500–2000 cm−1 (B) 2000–4000 cm−1
.
and these mesopores are formed in this sample due to swelling-
pillaring and calcination procedures.
attributed to the adsorbed asymmetric stretching vibration of
water molecules and surface hydroxyl groups. A small band at
3743 cm−1 due to Si OH was appeared for pillared MCM-36 sam-
ple, which is indicative of Si OH groups located on the outer
termination of SiO2 pillars present between the zeolite layers in
this sample. It is interesting to note that this band was not clearly
appeared in case of MCM-22 sample, this is probably due to very
low concentration of isolated Si-OH species on the external surface
of this sample [23]. A band at 3618 cm−1, which can be assigned
to strong Brönsted acidic bridging hydroxyl groups (Si OH Al)
[24] can be observed in both the samples, however the intensity
of this peak is relatively higher in MCM-36 sample than MCM-
22.
The FTIR spectrum of MCM-22 precursor and pillared MCM-36
samples in the range of 500–2000 cm−1 are shown in Fig. 4(A). The
bands at 1087 cm−1 and 1035 cm−1 correspond to internal asym-
metric stretching, and the bands at 595 cm−1 and 554 cm−1 are
attributed to the presence of double rings in the MCM-22 struc-
ture. The band at 805 cm−1 corresponding to external symmetrical
stretching vibration was observed in both the samples. Another
small band at 1630 cm−1 which can be ascribed to the angular
deformation of the O H bond [22] was also observed in both sam-
ples.
The presence of bands related to the symmetrical and asym-
metrical stretching of the O H groups present in the MCM-22
and MCM-36 samples was also investigated (Fig 4(B)). A broad
band around 3386 cm−1 appeared for both samples and can be
The major goal of this work is to synthesize pillared MCM-
36 material to generate a better catalyst in comparison to the
untreated zeolite.

164
R. Purova et al. / Journal of Molecular Catalysis A: Chemical 406 (2015) 159–167
Table 2
Data from DRIFT pyridine adsorption measurements of samples.
Sample
Number of acid sites (␮mol g−1
)
B/L ratio
–
Number of total sites (␮mol g−1
)
Brönsted (B)
Lewis (L)
MCM-22
HMCM-22
MCM-36
HMCM-36
0.0
4.2
0.0
13.6
15.8
10.8
10.2
13.6
20.0
10.8
44.5
0.265
–
34.3
3.362
HMCM-22
HMCM-36
MCM-36
L
B
L+B
MCM-22
HMCM-22
MCM-22
HMCM-36
100
200
300
400
500
600
700
800
Temperature (oC)
L
B
L+B
Fig. 6. NH3-temperature programmed desorption profiles of the samples.
MCM-36
formation of new Brönsted sites was clearly observed as revealed
by the intense band at 1545 cm−1 along side of intense peak due
to Lewis and Brönsted acid sites. The concentration of the Brönsted
acid (B) sites and Lewis acid (L) sites were determined based on
intensities of the bands observed at 1545 and 1443 cm−1, taking the
corresponding molar extinction coefficients (e), i.e., 0.059 0.004
and 0.084 0.003 cm2 mol−1, respectively [27]. Quantitative data
were calculated using the Lambert–Beer equation [28] listed in
Table 2. It is clearly shown from Table 2 that ion exchange resulted
increase of the Brönsted acid sites concentration. It is known that
Si/Al ratio plays an important role in the final concentration of acidic
sites in zeolites.
L= Lewis acid sites
B=Bronsted scid sites
1350
1400
1450
1500
1550
1600
Wavenumber (cm-1)
Fig. 5. FTIR spectra of the MCM-22, HMCM-22, MCM-36 and HMCM-36 samples
after the pyridine adsorption.
It is known that the presence of inert pillars in MCM-36 puts the
pillared derivative at a disadvantage by diminishing overall acidity,
while the process of swelling and pillaring may degrade the exist-
ing centers [25]. To avoid this disadvantage, we performed the ion
exchange of sodium ions with protons, which could increase the
overall acidity of the pillared MCM-36 material. The acidic proper-
ties of as synthesized and ion exchanged samples were evaluated
using FTIR spectroscopy following pyridine adsorption. The results
are shown in Fig. 5 and Table 2. As-synthesized samples showed
only peak corresponding to Lewis acid sites. In contrast, the ion
exchanged samples showed peaks due to Brönsted and Lewis acid
sites. In comparison, the ion exchanged MCM-22 sample showed
lesser number of Brönsted acid sites and almost same amount
of Lewis acid sites as pillared HMCM-36 sample. The spectra of
these two samples displayed well resolved bands at 1443, 1490
and 1545 cm−1. The band at 1443 cm−1 can be assigned to char-
acteristic of Lewis-coordinated pyridine (L), whereas the band at
1545 cm−1 was due to Brönsted coordinated pyridine (B), and the
band at 1490 cm−1 is due to Lewis and Brönsted -coordinated pyri-
dine (L + B). The assignment of these bands was in agreement with
those reported in the literature [26].
The concentration and distribution of acid sites in MCM-22
and pillared MCM-36 samples before and after ion exchange were
characterized by temperature programmed desorption (TPD) of
ammonia (Fig. 6). The TPD profiles of the MCM-22 samples showed
a maximum desorption peak at 115 ◦C and a small peak at 335 ◦C.
Unverricht et al. [29] and He et al. [7] attributed the low tempera-
ture peak to the physisorbed ammonia desorbed from the samples,
whereas the small peak at approximately 335 ◦C is assigned to the
+
desorption of NH4 ions from strong Brönsted acid sites. A broad
peak is observed in the range between 350 and 575 ◦C, with maxi-
mum near 450 ◦C in the pattern of HMCM-36 sample, which could
be ascribed to strong Brönsted acid sites generated during the
swelling and calcination procedure.
It can be easily seen that at a given Si/Al ratio in the final pillared
material, the swelling procedure gives a sample with much higher
concentration of Brönsted acid sites and with a higher proportion of
sites with medium-to-strong acidity. This result is important from
the catalytic point of view, since activity and selectivity of the cat-
alyst will be related to the total number of Brönsted acid sites and
their acid strength distribution. The acid site concentration of pil-
lared MCM-36 is significantly higher compared to MCM-22. These
results are in quite accordance with the FTIR analysis after pyridine
adsorption.
The HMCM-22 sample predominantly exhibited Lewis acidity
with a minor contribution of Brönsted acidity. However, major dif-
ferences were noticed in FTIR spectrum of HMCM-36 sample. The

R. Purova et al. / Journal of Molecular Catalysis A: Chemical 406 (2015) 159–167
165
90
80
70
60
50
40
30
20
10
0
(A)
MCM-22
HMCM-22
MCM-36
HMCM-36
MCM-22
2.0
(B)
HMCM-22
MCM-36
HMCM-36
1.5
1.0
0.5
0.0
Time (min.2)50
0
50
100
150
200
300
350
400
0
50
100
150
200
300
350
400
Time (min.2)50
Fig. 7. (A) Effect of reaction time on the conversion of palmitic acid over all the samples (B) A plot between [ln 1/(1 − x)] and reaction time (reaction temperature; 70 ◦C).
Table 3
Comparison of the kinetic performance of the functionalized mesoporous silica catalysts.
Catalyst
Apparent activation energy (kJ mol−1
)
Rate
constant (min−1) x 10−4
80 ◦
Conversion ofpalmitic acid (%)a
70 ◦
C
C
90 ◦
C
MCM-22
HMCM-22
MCM-36
HMCM-36
58.86
48.09
62.12
21.58
6.6
16.0
8.8
8.9
35.9
12.0
9.5
45.9
20.3
26
60
39
100
57.6
119.5
171.5
a
Conversion of palmitic acid after 24 h reaction, reaction temperature; 70 ◦C.
Table 4
Comparison of activity of different mesoporous catalysts in esterification of palmitic acid with methanol.
Mesoporouscatalyst
Reaction conditions
Conversion of Palmitic acid (%)
Reference
HPA/SBA-15
WOx/ZrP
WOx/ZrO2
Al-MCM-41
SO3H-SWCNHs
SBA-15-SO3H-P123
Pillared HMCM-36
30 mL of MeOH, 8 mmol of PA, temperature = 60 ◦C; catalyst wt.; 0.2 g. reaction time; 6 h
60 ◦C, 30:1 molar ratio MeOH:PA; reaction time;6 h
90
78
30
79
90
95
100
[14]
[12]
[33]
[34]
[35]
[11]
Present work
60 ◦C, 30:1 molar ratio MeOH:PA; reaction time;6 h
0.6 wt% catalyst at a 60/1 alcohol/PA ratio, 130 ◦C for 2 h
Catalyst wt., PA and MeOH = 0.15, 0.15, and 5 g; temperature = 64 ◦C; reaction time; 5 h
1:20 w/w (PA:MeOH) in oil, 85 ◦C, 3h
80 ◦C, 12.5 mL MeOH, 30:1 molar ratio MeOH:PA acid; reaction time;6 h
The catalytic activity of the as-synthesized and ion exchanged
MCM-22 and pillared MCM-36 samples for the esterification of
palmitic acid with methanol was determined. Fig. 7(A) shows the
effect of reaction time on the conversion of palmitic acid over
all the samples. We determined the degree of conversion as a
function of time under the identical reaction conditions (reaction
temperature 70 ◦C, alcohol/acid molar ratio of 60, 0.05 g catalyst).
It was observed that the catalytic activity decreases in the series:
HMCM-36> HMCM-22> MCM-36> MCM-22. This behavior can be
explained due to the HMCM-36 catalyst possessing the highest
acid strength of studied catalysts (Table 2). After 6 h of reaction, it
was observed that the palmitic acid conversion (%) is 85.3%, 42.6%
25.1 and 19.4% for the HMCM-36, HMCM-22, MCM-36 and MCM-
22 catalysts, respectively. The values of the maximum conversion
obtained by the solid HMCM-36, HMCM-22, MCM-36 and MCM-22
catalysts within 24 h are shown in Table 3. It can be clearly seen
that HMCM-36 has the highest catalytic performance.
The order of the esterification reaction was determined by fol-
lowing the classical definitions of chemical kinetics. Considering
the conditions employed in the process, palmitic acid is a limiting
reagent. Fig. 7(B) shows possible fits of the esterification reaction of
palmitic acid assuming first-order kinetics. The fitting in Fig. 7(B)
reveals a linear relation between all experimental data when [ln
1/(1 − x)] is plotted as a function of reaction time, thus establish-
ing the existence of a first-order dependence between the reaction
rate and the concentration of palmitic acid for the esterification
reaction. The regression coefficients of the straight lines show good
fits to first-order kinetics. Several reports in the literature establish
first-order kinetics for this esterification reaction [30].
It was reported that higher number of acid sites and larger pore
diameter of catalyst could contribute to the improved performance
in esterification of palmitic acid. The responsible factor of the high
activity for the HMCM-36 catalyst requires understanding of the
relative importance of its superiority in pore size and number of
acid sites. To understand these features, the esterification was per-
formed for all the catalysts at a range of temperatures from 70 to
90 ◦C (Fig. 8). The collected data was used to calculate apparent
activation energies for the catalysts.
The apparent activation energies were calculated assuming a
pseudo-first-order reaction with respect to the palmitic acid. The
calculated rate constants and apparent activation energies are sum-
marized in Table 3 for all the catalysts. A temperature increase
of 10–20 ◦C caused an increase in the rate constants. The ion
exchanged materials showed lower apparent activation energy
than the as-synthesized and calcined samples. It is interesting to
note that the apparent activation energy was found to decrease
with increasing the pore diameter in case of ion-exchanged sam-
ples. If internal diffusion was not limiting the catalyst, the apparent

166
R. Purova et al. / Journal of Molecular Catalysis A: Chemical 406 (2015) 159–167
Table 5
MCM-22
HMCM-22
MCM-36
Reuse of HMCM-36 in palmitic acid esterification.
100
80
60
40
20
0
Reaction time
Conversion of palmitic acid (%)
HMCM-36
Cycle 1
Cycle 2
Cycle 3
Cycle 4
60
15.3
31.1
53.2
69.5
77.7
85.3
14.5
30.0
52.4
68.5
76.3
84.0
14.0
29.8
51.5
67.3
75.7
83.0
13.5
29.0
51.0
66.5
74.1
82.5
120
180
240
300
360
[31] reported that Si/Al ratio of zeolites (H␤, HY, and HZSM-5) influ-
ences the esterification with alcohols. It was also observed that the
largest conversion values are obtained with low Si/Al ratios due to
the Brönsted acid sites having been incorporated into the structure
of MCM-41 [32].
70
80
90
Temperature (oC)
Table 4 presents some results on esterification of palmitic acid
over different mesoporous catalysts reported in the literature. It
is clear that the pillared HMCM-36 in the esterification reaction
offered high activity relatively at low reaction temperature. For the
esterification of palmitic acid aimed at the production of biodiesel,
HMCM-36 is active due to mesoporosity and large number of Brön-
sted acid sites. In order to study the catalytic stability of pillared
HMCM-36 catalyst, an experiment, similar to the ‘hot-filtration
experiment’ was carried out. The catalyst was suspended in the
methanol for 24 h at reaction temperature 70 ◦C under stirring
without addition of palmitic acid. After 24 h, the catalyst was sep-
arated from methanol by centrifugation, and the palmitic acid was
added to the reaction mixture. The esterification reaction was car-
ried out for 24 h, but the filtered methanol did not offered any
activity. This observation indicating that the pillared HMCM-36
catalyst behavior is clearly heterogeneous in nature and there is no
leaching of any active species. The dried catalyst was also tested for
6 h and the conversion of palmitic acid is same as the fresh catalyst.
To evaluate the reusability of the most active catalyst, HMCM-
36 the esterification of palmitic acid was repeated for four cycles
(Table 5). After each cycle of the reaction, the catalyst was sepa-
rated by filtration from the reaction mixture washed with methanol
and dried at 70 ◦C. The dried catalyst was then reused for the next
cycle. For each cycle, a fresh solution containing palmitic acid and
methanol at the same concentrations as in the first cycle was pre-
pared. A small decrease of activity was observed with each cycle,
this result probably due to loss of amount of catalyst during the
separation procedures.
Fig. 8. Effect of reaction temperature on the conversion of palmitic acid over all the
samples.
50
90
80
70
60
50
40
30
20
10
*Number of acidsites determined from FTIR pyridine adsorption
*Specific rate at reaction temp:70oC; time:6h
40
30
20
10
0
MCM-22
HMCM-22
MCM-36
HMCM-36
Fig. 9. Correlation between the number of acid sites and specific rate of all the
catalysts.
activation energies for the synthesized MCM-22 and MCM-36 sam-
ples should be the same since these two samples possesses only
Lewis acid sites with small change in number of sites. A likely cause
of the increase of apparent activation energy in these samples is due
to less pore diameter which could hinder the diffusion of reactant
molecules.
4. Conclusions
Mbaraka et al. [11] reported the importance of activated dif-
fusion in the esterification of palmitic acid over the mesoporous
catalysts. These authors also reported that materials should pos-
sess a pore size diameter at least within the range of 2–3 nm to
avoid diffusion problems. The HMCM-36 sample possessed around
3 nm pore diameter as shown in Fig. 3(B). Fig. 9 presents the corre-
lation between the total number of acid sites and specific rate for
esterification of palmitic acid at 70 ◦C. It is clear that the specific
rate of the esterification for catalyst proportional to the number
of acid sites. In addition, HCMC-36 sample possesses higher Brön-
sted/Lewis acid ratio than other samples (Table 2). This observation
coincides with the literature that esterification follows a Brönsted
acid catalyzed pathway.
Porous MCM-22 and pillared MCM-36 has been synthesized by
calcination and swelling/pillaring of MCM-22(P) material respec-
tively. It was found that swelling and pillaring processes increased
the BET surface area of MCM-36 by a factor of two and also the
microporous nature of MCM-22 changed to MCM-36 with meso
pores of around 3 nm. The ion exchanged MCM-22 and MCM-36
materials possessed more number of strong Brönsted acid sites.
It was observed that HMCM-36 sample had the highest Brön-
sted/Lewis acid site ratio. The intrinsic activities of as-synthesized
and ion exchanged materials were investigated by liquid phase
esterification of palmitic acid with methanol to produce methyl-
palmitate. Based on the specific rates, HMCM-36 showed the
highest reactivity. In addition, this reactivity was higher than that of
the other mesoporous heterogeneous catalyst. Recycling esterifica-
tion experiments suggested that the acid sites on HMCM-36 sample
were stable and offered consistent activity for four cycles. The
results of this study clearly indicating that HMCM-36 is a suitable
catalyst for the production of methyl palmitate by the esterification
It is noted that HMCM-36 sample had significantly higher num-
ber of Brönsted acid sites than HMCM-22, and also the BET surface
area of HMCM-36 was about twice as large as that of HMCM-22. It is
known that the accessibility of the acidic sites, a potentially impor-
tant characteristic is the strength of the acid site. Kirumakki et al.

R. Purova et al. / Journal of Molecular Catalysis A: Chemical 406 (2015) 159–167
167
of palmitic acid due to its superiority in number of strong Brönsted
acid sites and the mesoporous nature.
[13] J.M. Marchetti, A.F. Errazu, Biomass Bioenerg. 32 (2008) 892–895.
[14] A.I. Tropecelo, M.H. Casimiro, I.M. Fonseca, A.M. Ramos, J. Vital, J.E.
Castanheiro, Appl. Catal. A: Gen. 390 (2010) 183–189.
[15] L. Xu, Y. Wang, X. Yang, X. Yu, Y. Guo, J.H. Clark, Green Chem. 10 (2008)
746–755.
Acknowledgements
[16] M.E. Davis, Acc. Chem. Res. 26 (1992) 111–115.
[17] W.J. Roth, J. Cejka, Catal. Sci. Technol. 1 (2011) 43–53.
[18] S.-T. Yang, J.-Y. Kim, J. Kim, W.-S. Ahn, Fuel 97 (2012) 435–442.
[19] W.J. Roth, J.C., Vartuli, C.T. Kresge, in: A. Sayari, et al. (Eds.), Nanoporous
Materials II, Stud. Surf. Sci. Catal. 129 (2000) 501-506.
[20] (a) M.E. Leonowicz, S.L. Lawton, R.D. Partridge, P. Chu, M.K. Rubin, Science 264
(1994) 1910–1913;
This research work was funded by the Deanship of Scientific
Research (DSR), King Abdulaziz University, Jeddah, under grant
number (D-002-432). The authors acknowledge with thanks the
DSR’s financial support.
(b) J. Cejka, A. Krejci, N. Zikova, J. Kotrla, S. Ernst, A. Weber, Micropor.
Mesopor. Mater. 53 (2002) 121–133.
[21] M. Lasperas, H. Camboo, D. Burnel, I. Rodriguez, P. Geneste, Micropor. Mater. 7
(1996) 61–72.
[22] A. Corma, C. orell, J. Pérez-Pariente, Zeolites 15 (1995) 2–8.
[23] J.A. Lercher, C. Grundling, G. Eder-Mirth, Catal. Today 27 (1996) 353–376.
[24] J.-O. Barth, A. Jentys, J. Kornatowski, J.A. Lercher, Chem. Mater. 16 (2004)
724–730.
[25] E.J.A. Schweitzer, P.F. van den Oosterkamp, Micropor. Mesopor. Mater. 20
(1998) 397–405.
[26] B. de Rivas, C. Sampedro, M. Garcia-Real, R. Lopez-Fonseca, J.I. Gutierrez-Ortiz,
Appl. Catal. B: Environ. 129 (2013) 225–238.
[27] J.M. Jehng, A.M. Turek, I.E. Wachs, Appl. Catal. 83 (1992) 179–200.
[28] F. Delannay, Characterization Of Heterogeneous Catalysts, Marcel Dekker,
New York, 1984.
[29] S. Unverricht, M. Hunger, S. Ernst, H.G. Karge, J. Weitkamp, Stud. Surf. Sci.
Catal. 84 (1994) 37–44.
[30] A. Palani, A. Pandurangan, J. Mol. Cat. A: Chem. 245 (2006) 101–105.
[31] S.R. Kirumakki, N. Nagaraju, K.V.R. Chary, Appl. Catal. A: Gen. 299 (2006)
185–192.
References
[1] J.A. Lercher, A. Jentys, In Handbook of Microporous Solids, in: F. Schuth, K.
Sing, J. Weitkamp (Eds.), Wiley-VCH, Weinheim, Germany, 2002, pp.
1097–1145.
[2] J.-O. Barth, A. Jentys, J. Kornatowski, J.A. Lercher, Chem. Mater. 16 (2004)
724–730.
[3] J.Y. Ying, C.P. Mehnert, M.S. Wong, Angew. Chem. Int. Ed. 38 (1999) 56–77.
[4] A. Jentys, K. Kleestorfer, H. Vinek, Micropor. Mesopor. Mater. 27 (1999)
321–326.
[5] M. Mokhtar, K. Narasimharao, H.S. Harun, S.N. Basahel, I.H. Abd El-Maksod,
Ceram. Int. 39 (2013) 683–689.
[6] A. Clearfield, M. Kuchenmeister, Pillared Layered Materials, Supramolecular
Architecture, ACS Symp. Ser. No. 499, in: T. Bein (Ed.), American Chemical
Society, Washington, DC, 1992.
[7] Y.J. He, G.S. Nivarthy, F. Eder, K. Seshan, J.A. Lercher, Micropor. Mesopor.
Mater. 25 (1998) 207–224.
[8] (a) A. Corma, V. Fornes, J. Martinez-Triguero, S.B. Pergher, J. Catal. 186 (1999)
57–63;
[32] B.R. Jermy, A. Pandurangan, Appl. Catal. A: Gen. 288 (2005) 25–33.
[33] V.C. dos Santos, K. Wilson, A.F. Lee, S. Nakagaki, Appl. Catal. B: Environ. 162
(2015) 75–84.
[34] A.C. Carmo Jr., L.K.C. de Souza, C.E.F. da Costa, E. Longo, J.R. Zamian, G.N. da
Rocha Filho, Fuel 88 (2009) 461–468.
(b) S. Maheshwari, E. Jordan, S. Kumar, F.S. Bates, R.L. Penn, D.F. Shantz, M.
Tsapatsis, J. Am. Chem. Soc. 130 (2008) 1507–1516.
[9] Y. Liu, E. Lotero, J.G. Goodwin Jr., J. Catal. 243 (2006) 221–228.
[10] J.M. Marchetti, V.U. Miguel, A.F. Errazu, Fuel 86 (2007) 906–910.
[11] I.K. Mbaraka, D.R. Radu, V.S.-Y. Lin, B.H. Shanks, J. Catal. 219 (2003) 329–336.
[12] K. Narasimharao, A. Sridhar, A.F. Lee, S.J. Tavener, N.A. Young, K. Wilson,
Green Chem. 8 (2006) 790–797.
[35] C. Poonjarernsilp, N. Sano, H. Tamon, Appl. Catal B: Environ. 147 (2014)
726–732.