Transesterification of triacetin and castor oil with methanol catalyzed by supported polyaniline-sulfate. A role of polymer morphology

Article abstract of DOI:10.1016/j.apcata.2013.01.022

Polyaniline-sulfate deposited on three different carriers was studied for transesterification of triacetin and castor oil with methanol at mild reaction conditions (temperature of 55 °C). Multi-wall carbon nanotubes (CNT), carbon and silica were coated with polyaniline sulfate (ca. 20 wt%) during polymerization of aniline. Because of different textural and hydrophobic properties of the carriers, the polymer coatings of various morphologies were obtained as evidenced by the electron microscopy technique. A uniform coating of CNT with polymer resulted in the most extended polymer structure. Nanorods of polymer forming branched dendritic structures appeared in the other two carbon and silica carriers. The acid capacity and the strength of acid sites were similar in all studied catalysts. All the samples were found to be active solid acid catalysts in methanolysis of both studied triglycerides and CNT-coated polyaniline sulfate exhibited the highest activity. The course of reaction during methanolysis of triacetin on CNT-containing catalyst was similar to that in the presence of soluble sulfuric acid. On the other hand, a partial blockage of active sites was observed in carbon and silica coated with polyaniline sulfate. A blockage effect was ascribed to strong interaction of acid sites with more polar reagents among them diacetin and glycerol. These interactions were facilitated by aggregated fibriral morphology of the polymer coating resulting in locally high density of acid sites.

Full text of DOI:10.1016/j.apcata.2013.01.022

Applied Catalysis A: General 455 (2013) 92–106
Contents lists available at SciVerse ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Transesterification of triacetin and castor oil with methanol catalyzed
by supported polyaniline-sulfate. A role of polymer morphology
A. Drelinkiewicza,∗, Z. Kalemba-Jajea, E. Lalika, A. Zie˛baa, D. Muchaa,
E.N. Konyushenkob, J. Stejskalb
a Institute of Catalysis and Surface Chemistry, Polish Academy of Sciences, Niezapominajek 8, 30-239 Kraków, Poland
b Institute of Macromolecular Chemistry, Academy of Sciences of the Czech Republic, Heyrovsky Sq. 2, 162 06 Prague 6, Czech Republic
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 27 August 2012
Received in revised form 18 January 2013
Accepted 22 January 2013
Available online 7 February 2013
Polyaniline-sulfate deposited on three different carriers was studied for transesterification of triacetin and
castor oil with methanol at mild reaction conditions (temperature of 55 ◦C). Multi-wall carbon nanotubes
(CNT), carbon and silica were coated with polyaniline sulfate (ca. 20 wt%) during polymerization of aniline.
Because of different textural and hydrophobic properties of the carriers, the polymer coatings of various
morphologies were obtained as evidenced by the electron microscopy technique. A uniform coating
of CNT with polymer resulted in the most extended polymer structure. Nanorods of polymer forming
branched dendritic structures appeared in the other two carbon and silica carriers. The acid capacity and
the strength of acid sites were similar in all studied catalysts. All the samples were found to be active
solid acid catalysts in methanolysis of both studied triglycerides and CNT-coated polyaniline sulfate
exhibited the highest activity. The course of reaction during methanolysis of triacetin on CNT-containing
catalyst was similar to that in the presence of soluble sulfuric acid. On the other hand, a partial blockage
of active sites was observed in carbon and silica coated with polyaniline sulfate. A blockage effect was
ascribed to strong interaction of acid sites with more polar reagents among them diacetin and glycerol.
These interactions were facilitated by aggregated fibriral morphology of the polymer coating resulting
in locally high density of acid sites.
Keywords:
Transesterification
Triacetin
Castor oil
Polyaniline
© 2013 Elsevier B.V. All rights reserved.
1. Introduction
(Scheme 1) (PANI-S) was active and stable acid catalyst for the
transesterification of triglycerides (triacetin and castor oil) and
Biodiesel fuel consisting of methyl esters of fatty acids is
obtained in transesterification of triglycerides with methanol
catalyzed by strong acids or bases. Owing to environmental, tech-
nological and economic reasons, the substitution of homogeneous
catalysts by heterogeneous ones is a desirable goal. The advantage
of solid catalysts consists of the ease of product separation and
recycling of the catalyst. Recent studies concentrated mostly on
solid acid catalysts as they are able to catalyze both transesterifi-
cation of triglycerides and esterification of free fatty acids present
in oil feedstock. A variety of solid acid catalysts has already been
tested [1–5] and the sulfonic acid-based catalysts seem to be the
most promising candidates. The examples include sulfate treated
inorganic oxides such as zirconia, titania, tin oxide [6–10]. Another
class of solid acid catalysts are organosulfonic groups-containing
polymers such as Nafion, Amberlyst [5,11,12], various polystyrenes
[13,14] as well as mesoporous silica materials [15–17]. Our pre-
vious work showed that polyaniline doped with sulfuric acid
esterification of fatty acid (ricinoleic acid) with methanol [18,19].
Polyaniline salts such as hydrochloride, sulfate, nitrate were also
reported by Palaniappan et al. [20–23] to be active catalysts for
esterification of carboxylic acids (lauric, caprylic, acetic) and trans-
esterification of ketoester (acetoacetate). Polyaniline sulphate was
observed to be much more active than nitrate and hydrochloride
salts [23]. Moreover, polyaniline sulfate showed almost fully
stable activity evidenced by no essential change in the catalyst
composition (leaching of sulphate) during the reactions [21–23].
The term “polyaniline” (PANI) refers to a polymer obtained by
the oxidative polymerization of aniline. Due to nitrogen-containing
groups, polyaniline exhibits basic character which allows easy
doping of polymer with protonic acids to give the “polyaniline
salt” (hydrochloride, sulfate etc.) [24]. Polyaniline salts are easy to
prepare, handle, exhibit good environmental and thermal stability
(temp up. to ca. 300 ◦C), are insoluble in most of organic solvents
and they are organic semiconductors. These properties, and in par-
ticular electrical conductivity, make polyaniline salts a potential
candidate to be used in optical and microelectronic devices, etc.
From catalytic point of view, an interesting property of polyaniline
sulfate is its high content of acid sites (4.75 mmol/g) [18,19] which
∗
Corresponding author. Tel.: +48 126395114; fax: +48 124251923.
E-mail address: ncdrelin@cyf-kr.edu.pl (A. Drelinkiewicz).
0926-860X/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2013.01.022

A. Drelinkiewicz et al. / Applied Catalysis A: General 455 (2013) 92–106
93
Scheme 1. The structure of polyaniline base and polyaniline sulfate.
is close to that of Amberlyst-15 commonly considered to be a cata-
lyst of high acid capacity. On the other hand, PANI-sulfate exhibits
relatively low specific surface area (25.9 m2/g) [19] and does not
swell in contact with reagents such as methanol, methyl esters or
triglycerides. This results in not enough satisfactory utilization of
active sites in the PANI-S powder catalyst.
etc. These features could determine deposition of polymer more
or less deeply inside the pore structure of the carrier. This could
influence the access of active sites for the reactants and seems to
be especially important in the case of relatively large triglyceride
molecules, reactants in transesterification reaction.
Polyaniline sulfate-coated carbon samples were tested in our
previous work [18]. Present studies are extended to other car-
riers, namely silica and multi-wall carbon nanotubes (CNT). The
research is based on the polymerization of polyaniline sulfate in the
presence of supports resulting in materials showing acid proper-
ties. The so-prepared catalysts have been thoroughly characterized
by means of different experimental techniques including FTIR,
XRD, nitrogen adsorption–desorption isotherms, ammonia sorp-
tion and microscopy (optical and SEM). Methanolysis is studied
for triacetin (glycerol triacetate) (TACT) and castor oil. Reactivity
of polyaniline sulfate-coated silica (SiO2-PANI-S) and CNT (CNT-
PANI-S) is compared with that of previously studied polyaniline
sulfate-coated carbon (C-PANI-S) sample. The results obtained in
the present work demonstrate that apart from textural properties
(surface area, porosity) the support-sensitive morphology of poly-
mer coating plays an essential role in catalytic reactivity of studied
samples.
However, strong adhering ability of polyaniline towards variety
of materials (including glass, ceramics, inorganic oxides, metals,
polymer microspheres, plastics, fibres, membranes) allowed depo-
sition of polymer overlayer (submicroemter thickness) on the
material surface [25]. The obtained polyaniline coatings exhib-
ited an excellent resistance to solvent washing and manual
scratching by sharp objects. Deposition of polyaniline salts was
widely explored to modify electrical conductivity of the materi-
als. Therefore these studies concentrated mostly on the role of
polymerization conditions in electrical conductivity of polyaniline
salts-coated samples. Polyaniline coatings could be produced by
means of electrochemical as well as chemical procedures, among
them by “in situ” adsorption polymerization of aniline performed in
the presence of the materials. In this procedure the PANI-salts over-
layer is formed by the immersion of the supporting material into
an aqueous acidic solution of aniline containing an oxidant [26–30].
The coatings of polyaniline salts could exhibit various morpholo-
gies ranging from microspherical, rice-grain, coral-like cylinders,
nanotubes and nanofibers. Nanofibers are often branched forming
various networks, arrays and bundles.
2. Experimental
2.1. Catalysts preparation
The hydrophobous properties of the carrier surface were
observed to play an important role in the process of polyaniline
coating formation [26,31]. The coatings produced on hydropho-
bic surfaces like carbon were more uniform thus exhibiting more
extended conformation. The tendency to develop the granular
morphology of polyaniline was more pronounced for hydrophilic
surfaces, such as glass, silica [31–36]. This was explained by a pref-
erential adsorption of the reactive intermediates such as aniline
cation radicals and oligomers onto the hydrophobic surface. This
adsorption might change as a result of altered electron-density
distribution caused by oligomers interactions with the surface of
carriers.
From catalytic point of view, not only the morphology of
polyaniline-coating but also the location of polymer throughout the
particles of carrier seems to be important. However, to the best of
our knowledge, no systematic studies were performed to examine
a role of textural properties of carrier, like porosity, pore diameter,
Polyaniline sulfate was deposited on carbon (Kemisorb, Kemira),
silica (SP18 Kemira) and multi-wall carbon nanotubes (CNT). The
physicochemical characteristics of CNT sample are described in
previous papers [37,38]. Shortly, the multiwalled carbon nano-
tubes were manufactured by the firms L.MWNCTs-2040, Conyan
Biochemical Technology, Taipei, Taiwan, diameter 40–80 nm in
diameter, length 5–15 ␮m, and specific surface area 40–300 m2/g
[37].
Deposition of polymer was carried out in situ during the prepa-
ration of polyaniline by the method described earlier in detail
[39]. The polymerization was performed by the oxidation of 0.1 M
aniline sulfate with 0.25 ammonium peroxydisulfate in 50 vol.%
ethanol–water at room temperature. Support was dispersed in
ethanol, aniline sulfate was added and dissolved, followed by the
equal volume aqueous solution of ammonium peroxydisulfate.
A 100 cm3 of this reaction mixture when used without support

94
A. Drelinkiewicz et al. / Applied Catalysis A: General 455 (2013) 92–106
generates approximately ∼2 g of PANI-sulfate. If 8 g of silica is
suspended in the reaction mixture at the beginning of reaction,
polyaniline salt coats the surface of support with a thin submi-
crometre film. The content of polymer in the resulting composite
material is thus ca. 20 wt%. The same applies to the coating of
other supports, CNT and carbon material. General information on
the principles of coating are described in a recent review article on
this topic [26]. For ∼20 wt% polymer loading, 200 cm3 of reaction
mixture was used per 20 g of support. Next day the coated support
was separated, rinsed with 0.1 M sulfuric acid and acetone. The
collected samples were dried at room temperature in air and then
in desiccator over silica gel. Prior to catalytic test the samples were
dried at temperature of 60 ◦C under reduced pressure.
2.3. Catalytic tests
The transesterification of triacetin (Fluka) with methanol was
carried out in a 100 cm3 glass reactor at atmospheric pressure fol-
lowing the procedure reported in our previous publication [19].
Reactor was equipped with a reflux condenser, magnetic stirrer,
and a tube for sampling the solution. In catalytic experiment, tri-
acetin or castor oil, methanol and internal standard (toluene) were
introduced to the reactor, heated up to a given temperature and
then the catalyst was added.
The transesterification of triacetin (TACT) with methanol was typ-
ically performed at temperature of 55 ◦C using 2.6 cm3 of triacetin
and 16.2 cm3 of methanol, i.e. at methanol to triacetin molar ratio
(MR) of 29. The concentration of catalysts varied from 5.3 g/dm3
to 15.8 g/dm3, which corresponded to 0.33–1 wt% relative to the
mass of triacetin. In the course of catalytic tests the samples of reac-
tion mixture were periodically taken and analyzed by GC method
following the method described earlier [19]. From the data of GC
analysis the triacetin conversion (CTG), and the yield of diacetin,
monoacetin (Yi) were calculated as follows:
Elemental analysis and thermogravimetric techniques were
used to evaluate the content of PANI-sulfate in the obtained sam-
ples.
2.2. Characterization of catalysts
The specific surface areas of samples were calculated from the
nitrogen adsorption–desorption isotherms at 77 K in an Autosorb-
1, Quantachrome equipment. Prior to the measurements, the
samples were preheated and degassed, under vacuum at 60 ◦C for
18 h. Pore size distribution was calculated using the BJH model
based on nitrogen desorption isotherm. The FT-IR spectra were
recorded using Bruker-Equinox 55 spectrometer and standard KBr
pellets technique. Morphology of samples was studied by means of
Field Emission Scanning Electron Microscope JEOL JSM – 7500 F.
Elemental analysis was performed by means of CHNS-VarioEL
III apparatus (Elementar Analysensysteme Hanau-Germany). The
thermogravimetric measurements were performed using a NET-
ZSCH STA 409 Luxx instrument within the temperature from 10
to 1000 ◦C (heating rate 30 ◦C/min) in an air atmosphere. Micro-
scopic observations of catalysts before and after catalytic test were
performed using inverted optical microscope (Nikon, Japan).
Microcalorimetric measurements of ammonia sorption were
performed to determine the capacity and the strength of acid
sites. Gas flow-through MICROSCAL microcalorimeter was used.
The instrument measures the rate of heat evolution accompanying
a solid–gas interaction. A sample is placed in a small microcalori-
metric cell (7 mm in diameter) and the measurement is carried
out in a flow-through mode. After thermally equilibrating the
system, ammonia was adsorbed from the nitrogen carrier gas con-
taining 3% of NH3 flowing through the cell at 3 cm3/min under
normal pressure. Concurrently with the heat evolution, the uptake
of ammonia from the carrier gas was measured by thermocon-
ductivity detector (downstream detector, DSD). The completion
of adsorption was signalled by cessation of the exothermic effect
measured by microcalorimeter as well as by a plateau being
reached in the downstream detector. At this point the pure nitrogen
carrier was switched on and the endothermic effect of desorp-
tion was measured by calorimeter whereas the amount of the
desorbed NH3 was detected by DSD. The irreversible uptake of
ammonia was calculated as a difference between the total uptake
detected on the sorption and the release of ammonia observed
on the subsequent desorption stage of the cycle. Similarly, the
molar heats of the irreversible ammonia sorption were calculated
as a difference of the total thermal effects of the sorption and
desorption.
NTG,0 − NTG
CTG
=
× 100%
(1)
(2)
NTG,0
Ni
NTG,0
Yi =
× 100%
where NTG,0 and NTG are the moles of triacetin initially present in the
reactor and the moles of triacetin remaining at time t, respectively.
Ni presents the number of moles of diacetin (NDI) or monoacetin
(NMONO) in the reaction mixture at time t.
It was shown in our previous article that PANI-base powder did
not exhibit activity for the methanolysis of triacetin. Furthermore,
in the absence of catalyst only a trace conversion of triacetin (0.9%
after 30 min of reaction) was observed [18,19].
The transesterification of castor oil with methanol was performed
at 55 ◦C using 6 g of castor oil, 7.6 cm3 of methanol (MR = 29) and
0.6 g of catalyst. This amount of catalyst corresponds to catalysts
concentration in the reaction mixture equal to 43 g/dm3.
Transesterification of castor oil at the same conditions as for tri-
acetin (concentration of catalyst, 15.8 g/dm3) resulted in very low
yield of methyl esters. This is a consequence of definitively lower
rate of reaction involving bulky triglycerides of castor oil. There-
fore, in methanolysis of castor oil higher concentration of catalysts
equal to 43 g/dm3 was used. This concentration was chosen in our
previous experiments [40].
The progress of reaction was monitored by GC analysis of methyl
esters formed as described before [40]. Castor oil is entirely com-
posed of a triglyceride of ricinoleic acid (87.44 wt%), traces of
glycerol (below 0.1 wt%), di- and mono-glycerides of ricinoleic acid,
and free ricinoleic acid (below 0.1 wt%). From the GC analysis of
methyl esters apart from the dominating triglycerides of ricinoleic
acid, low amounts of triglycerides of other fatty acids (linoleic
5.05 wt%, oleic 3.88 wt%, stearic 1.4 wt%, palmitic 1.28 wt%, linolenic
0.56 wt%, and other acids 0.39 wt%) were determined and the aver-
age molecular weight of castor oil was calculated to be 928 g/mol
[40]. In discussion of catalytic results, the yield of methyl esters
formed in methanolysis of castor oil was taken into consideration,
identically to the method widely used in the case of vegetable oils.
As a measure of catalyst activity initial rate of methyl esters
(ME) formation (below 10%) related to 1 g of catalyst was assumed
[r (ME), mol ME/min g].
For selected samples, the capacity of acid sites was also deter-
mined by acid–base titration. The amount of ca. 0.1 g of each sample
was suspended in 20 cm3 of NaOH (0.1 M) for 24 h at room tem-
perature. The liquid sample obtained after polymer filtration was
subsequently titrated with 0.1 M HCl. The completeness of PANI-
salt deprotonation after treatment by NaOH was checked by the
FT-IR technique.
The catalytic tests were performed 2-times (in selected cases
3-times) and the average values are reported.
Esterification of ricinoleic acid with methanol was carried in the
same reactor using 3 g of ricinoleic acid (Fluka) and 11.8 cm3 of
methanol (MR = 29) and 0.66 g of catalyst at temperature of 60 ◦C.

A. Drelinkiewicz et al. / Applied Catalysis A: General 455 (2013) 92–106
95
The progress of reaction was monitored by GC using the same ana-
lytical procedure as that in castor oil methanolysis.
3. Results and discussion
The supports used in the present work for deposition of polyani-
line sulfate differ essentially in hydrophobic character and in
textural properties (Table 1). Both carbon-based supports are
more hydrophobic compared to silica support. The carbon support
exhibits the highest surface area and the highest contribution of
microporous structure. The other two CNT and SiO2 supports are
materials of lower specific surface area and mostly mesoporous
structures.
In order to evaluate the content of polymer the catalysts were
subjected to elemental analysis and thermogravimetric experi-
ments. The data obtained by elemental analysis are not informative
owing to high contribution of carbon originating from C and
CNT supports. Furthermore, the content of sulfur determined by
elemental analysis in all three samples is low (0.15–0.32 wt%)
being close to the detection limit of the used technique. Previous
investigations devoted to thermal behaviour of CNT coated with
polyaniline sulphate [37] showed that thermal decomposition of
polyaniline sulphate started at temperature of ca. 200 ◦C and was
complete at 650 ◦C. The CNT material was thermally stable up to ca.
700 ◦C and become completely decomposed above 750 ◦C [37].
The content of polymer in the catalysts determined by thermo-
gravimetric analysis as the loss of mass upon heating up to 600 ◦C
is given in Table 2. This technique shows the content of polymer in
the carbon supported C-PANI-S sample equal to 24.1 wt%. The con-
tent of ca. 20.2 wt% was determined in this sample as the increase
of mass upon polyaniline deposition [18]. As shown in Table 2, the
content of polyaniline sulphate in the catalysts ranges from 12.9 to
24.1 wt%.
The capacity of acid sites (mmol H+/g of catalyst) is given in
Table 2. Before discussing the acidity data some comment relative
to the structure of polyaniline sulfate is needed. The catalysts were
obtained by the polymerization of aniline in the presence of sulfu-
ric acid. Recent results of Ayad et al. [41] demonstrated that two
kinds of polyaniline sulfate may be formed (Scheme 1). At lower
acidity of the medium, PANI-sulfate is formed by the association
of one molecule of sulfuric acid per two imine nitrogen (struc-
ture 1), while at higher acidity two molecules of sulfuric acid may
be introduced yielding PANI-hydrogen sulfate (structure 2). The
values of H+ capacity calculated assuming the formation of both
polyaniline sulfate structures are summarized in Table 2. The capac-
ities of protons in CNT-PANI-S and SiO2-PANI-S catalysts (1.16 and
1.24 mmol H+/g, respectively) are ca. two times higher than those of
structure 1 but are close to those predicted by structure 2. Slightly
higher value obtained for CNT supported sample may arise form
the acid sites present in the CNT material.
Fig. 1. FT-IR spectra of initial supports and polyaniline sulfate coated fresh and
recovered catalysts.
The most characteristic bands of polyaniline are located at 1498
and 1589 cm−1. They correspond to benzene and quinone ring-
stretching deformations, the C C in benzenoid and C N and C
C
bonds in quinonoid units, respectively [24]. Upon protonation of
nitrogen groups and formation of polyaniline salts, the quinoid and
benzenoid bands, both are shifted to lower wavenumbers, ca. 1479
and 1559 cm−1. Other spectral feature originating from the vibra-
tions of the doped segments of polyaniline chain is strong band at
1125 cm−1 and the peak at 1310 cm−1, related to the C N defor-
mation mode. All these spectral features characteristic of PANI-salt
(the bands located at 1555, 1472, 1308 and 1122 cm−1) can be
clearly recognized in the spectrum of carbon supported C-PANI-
S catalyst. They are less distinct in silica-supported SiO2-PANI-S
catalyst because of strong overlapping bands originating from pure
silica material. These are the peaks in the range of 1000–1300 cm−1
The content of H+ in carbon supported C-PANI-S sample
(1.41 mmol H+/g) although locates between that for structures 1
and 2, it seems to be nearest to the predicted value of struc-
ture 2. Thus, although both structures of polyaniline sulphate may
appear, structure 2 corresponding to polyaniline hydrosulfate is
the dominating one in the studied catalysts. However, for simplic-
ity, all the catalysts are termed as “polyaniline-sulfate” – based
samples.
representing Si
O Si stretching.
In the present work FT-IR spectroscopy is used to confirm
the presence of polyaniline sulfate in the catalysts. The spec-
tra of catalysts and supporting materials are displayed in Fig. 1.
Owing to relatively low content of polymer in the catalysts the
bands originating from the polyaniline-sulfate are relatively weak.
Nevertheless, spectral features originating from the protonated
polyaniline can be recognized in the spectra of all studied catalysts.
In the spectrum of CNT supported catalyst number of bands
originating from CNT material appears. However, apart from them,
two weak bands, characteristic of benzenoid and quinoid units of
polyaniline-salt, located at 1477 and 1570 cm−1 appear. Moreover,
strong band at 1125 cm−1 as well as the band at 1308 cm−1 due to
protonated nitrogen groups can be easily recognized. For clarity, the
spectrum registered for the deprotonated sample (treated by 0.1 M

96
A. Drelinkiewicz et al. / Applied Catalysis A: General 455 (2013) 92–106
Table 1
Physicochemical properties of catalysts.
Sample
Surface area
[m2/g]
Porosity
[cm3/g]
Pore diameter
[nm]
Micro pore volume
[cm3/g]
Meso pore volume
[cm3/g]
Carbon (C)
C-PANI-S
Silica (SiO2)
SiO2-PANI-S
CNT
734
347
325
288
173
20
0.39
0.13
1.07
0.68
0.37
0.11
1–2
1–2
9.8
6.7
3.8
3.7
0.199
0.037
0.014
0.008
0.013
0
0.191
0.093
1.056
0.672
0.357
0.110
CNT-PANI-S
NH3 (aq) for overnight) is also reported. It can be observed that in
the spectrum of polyaniline base-containing sample, the bands of
benzenoid units shifted to higher frequency, 1505 cm−1, whereas
the strong band at 1125 cm−1 vanished at the expense of the band
at 1146 cm−1, characteristic of polyaniline-base.
Thus, the FT-IR spectra of all studied catalysts exhibit the
spectral features characteristic of PANI-salt. The band originat-
ing from the vibration of S-O groups is commonly observed at
1030–1050 cm−1, i.e. at the position which is superimposed with
the strong band at 1125 cm−1 arising from the protonated segments
of polyaniline. This makes its observation difficult.
the polyaniline sulfate-coated CNT sample no microporous struc-
ture appears and its specific surface area (20 m2/g) is dramatically
lower than that of initial sample (173 m2/g). The comparison of
nitrogen adsorption–desorption isotherms for initial and PANI-S-
coated CNT shows no essential change in their shape. As a result
final CNT-PANI-S sample is mesoporous material of average pore
diameter of 3.7 nm compared to 3.8 nm before polymer deposi-
tion. These changes in textural properties of CNT due to polyaniline
deposition are consistent with the literature data [38].
Our previous results [19] demonstrated that surface area of
carbon support decreased strongly from 734 to 347 m2/g due to
polyaniline sulphate deposition. As the data in Table 1 show this
is accompanied by strong decrease in pore volume (from 0.39 to
0.13 cm3/g) and in particular decreases the micropores volume.
These results are consistent with the data reported by Avlyanov [42]
who observed that surface area of carbon coated with ca. 20 wt% of
PANI was about one half less than the surface area of initial carrier.
Data provided by the electron microscopy technique (SEM)
demonstrate that the morphology of polyaniline sulfate coating
is to high extent determined by the type of support (Figs. 3–6).
The polymer nanorods agglomerated into interconnected networks
forming branched fibrous structures can be seen for carbon and sil-
ica supported samples (Figs. 3 and 4). A more detailed inspection
(images registered at high magnification ×50,000) reveals typical
for polyaniline granular surface of nanorods in both catalysts. How-
ever, some differences in the shape, diameter, thickness and the
extent of aggregation of nanofibiral structures in polymer coatings
formed on silica and carbon materials exist. In carbon coated C-
PANI-S catalyst, the nanorodes of polymer are relatively short and
most of them are almost uniform in diameter 50–70 nm (Fig. 3).
The nanorodes of polymer are remarkably thicker (90–100 nm
in diameter) in the SiO2-PANI-S catalyst and the branched den-
dritic structures are more aggregated (Fig. 4). Apart from these
branched structures, a few rice-type polymer aggregates can be
also observed. This shows a more compact morphology of PANI-S
coating in silica-coated sample. The observed tendency to forming
polymer coating of more compact morphology on the silica surface
is consisted with the results of other authors [31,34] who observed
that glass and silica surfaces produced polyaniline “island”. This
may suggest that PANI-S coating is deposited mostly onto outer
most surfaces of silica particles and may explain definitively lower
reduction in specific surface area and porosity of silica when com-
pared with changes observed for carbon supports.
As described before, among all three supports carbon is material
of the highest specific surface area and the highest contribu-
tion of microporous structure. The nitrogen adsorption–desorption
isotherm of carbon support is typical for microporous materials
(Fig. 2). Other two silica and CNT carriers exhibit lower spe-
cific surface area with lower contribution of micropores. The
nitrogen adsorption–desorption isotherms and pore distribution
diagrams show that both these supports are essentially meso-
porous materials (Fig. 2). The isotherms for SiO2 and CNT supports
exhibit a hysteresis loop which is characteristic of mesoporous
solids. The adsorption branch of isotherm for initial SiO2 shows
a sharp inflection. The sharpness of this step indicates relatively
high uniformity of the mesopore size in the SiO2 support which
is calculated to be 9.8 nm. Deposition of PANI-S affects textu-
ral properties of all studied supports. This is manifested by the
reduction of specific surface area and porosity of the supports.
The contribution of both micropores and mesopores decrease due
to the coating of the supports with polyaniline sulfate overlayer
(Table 1).
However, the observed changes are different depending on
the type of support material. This is evidenced by comparison of
N2 adsorption–desorption isotherms for initial and PANI-S-coated
samples (Fig. 2). The changes in textural properties are the weak-
est in the case of silica support, i.e. material of less hydrophobous
character. The reduction in specific surface area of silica particles
is low (from 325 to 288 m2/g) being the lowest among all studied
supports. The N2 adsorption–desorption isotherm of SiO2-PANI-S
sample (Fig. 2) exhibits a narrow hysteresis loop thus showing some
changes in the porosity evidenced also by the pore distribution
diagram.
The changes in surface area caused by deposition of polyaniline
sulfate are considerably more distinct for carbon-based supports.
The most dramatic changes can be observed for CNT support. Sur-
face area and porosity of CNT material are dramatically reduced. In
The morphology of polyaniline sulfate in CNT differs essentially
from those in the carbon and silica – supported catalysts. The
Table 2
Acidic properties of studied catalysts.
Sample
PANI-S
[wt%]
Heat of sorption
kJ/mol NH3
Calculated acid capacity
mmol H+/g catalyst
Measured acid capacity
mmol H+/g catalyst
Struct. I
Struct. II
Acid-base titration
NH3 sorption
C-PANI-S
SiO2-PANI-S
CNT-PANI-S
24.1
16.4
12.9
83.3
60.7
76.0
1.04
0.70
0.56
1.72
1.20
0.92
1.25
1.25
1.10
1.41
1.24
1.16

A. Drelinkiewicz et al. / Applied Catalysis A: General 455 (2013) 92–106
97
700
600
500
400
300
200
100
0
0,04
0,02
0,00
0,030
SiO2
600
0,020
500
1
10
100
1000
0,010
0,000
Pore diameter [nm]
400
300
200
100
0
1
10
100
1000
Pore diameter [nm]
C
SiO2-PANI-S
C-PANI-S
0,00
0,20
0,40
0,60
0,80
1,00
0,00
0,20
0,40
0,60
0,80
1,00
Relative pressure (P/P0)
Relative pressure (P/P0)
0,0063
240
0,0042
0,0021
0,0000
180
120
60
1
10
100
1000
Pore diameter [nm]
CNT
CNT-PANI-S
0,20
0
0,00
0,40
0,60
0,80
1,00
Relative pressure (P/P0)
Fig. 2. Nitrogen adsorption – desorption isotherms and pore size distribution diagrams for initial supports and PANI-S coated samples.
micrographs presenting morphology of pristine and polyani-
line sulfate coated CNT samples are displayed in Figs. 5 and 6,
respectively. Previously reported data revealed that coating of
CNT with polyaniline took place only at the outer surface of the
CNT and the obtained coating was almost uniform [37,38]. In
the obtained composite polyaniline localized exclusively at the
surface of nanotubes created uniform layer capsulating individual
particles of the supporting material and forming a structure
labelled “fibre in a jacket” [37,38]. According to previous studies,
the polymerization of aniline inside the CNT was hindered by the
restricted access of reactants to the interior of the CNT [38]. This
type of polyaniline overlayer is consisted with the literature data
revelling the complete and uniform coating of CNT with various
conductive polymers, among them polyaniline salts [43,44].

98
A. Drelinkiewicz et al. / Applied Catalysis A: General 455 (2013) 92–106
Fig. 3. SEM micrographs of C-PANI-S catalyst.
Fig. 5. SEM micrographs of initial CNT support.
The SEM micrographs (Fig. 5) show that initial CNT support
whereas no fibrous network of polyaniline sulfate appears. Fur-
thermore, no cavities in the nanotubes appear and some traces
of polymer precipitate can be also observed. This demonstrates
uniform coating of CNT with polymer leading to much extended
polyaniline sulfate coating. This type of polymer coating may
consists of nanotubes 30–50 nm in diameter and their length is
extended to several micrometres. The cavities in the nanotubes can
be easily visible. Some nanotubes have no discernible cavity and
could be rather regarded as nanorods. The PANI-S-coated material
does not lose quasi-one-dimensional structure (Fig. 6). Polyaniline-
coated CNT become thicker, 40–60 nm in diameter. Coating of
CNT manifests by a more roughness surface of coated nanotubes
Fig. 4. SEM micrographs of SiO2-PANI-S catalyst.
Fig. 6. SEM micrographs of CNT-PANI-S catalyst.

A. Drelinkiewicz et al. / Applied Catalysis A: General 455 (2013) 92–106
99
explain strong reduction of CNT surface area and porosity due to
the polymer deposit.
100
80
60
40
20
0
Thus, textural and hydrophobous features of support mate-
rial influence the process of polyaniline sulfate deposition. More
hydrophobous carbon-based supports produce better extended
overlayers of polyaniline sulfate than that formed on less
hydrophobous silica material. This is evidenced by more dramatic
reduction of specific surface area and porosity of both carbon-based
supports compared to that for silica carrier. The most extended
polymer coating is formed on CNT support.
CNT-PANI-S
C-PANI-S
The acid capacity (mmol H+/g of catalyst) was determined using
ammonia sorption technique. For selected catalysts the acid–base
titration method was also involved to determine the acid centres
content. The obtained data are collected in Table 2. From the sorp-
tion experiments the heat of ammonia sorption (kJ/mol NH3) was
also determined. This value is assumed to be a measure of the
strength of acid centres in the samples. As shown in Table 2, stud-
ied catalysts exhibit similar acid characteristics. The content of
H+ ranges from 1.16 to 1.41 mmol/g. The heat of ammonia sorp-
tion is within similar range, 60.7–83.3 kJ/mol. However, the heat
obtained for SiO2-PANI-S catalyst (60.7 kJ/mol) is slightly lower
when compared with both carbon-based samples. This may suggest
that interactions of structural units of polyaniline with silica surface
are somewhat different than interactions with carbon materials.
SiO2-PANI-S
0
4
8
12
16
20
24
28
time [h]
100
CNT-PANI-S
3.1. Catalytic experiments
80
60
40
20
0
As described before, PANI-sulfate-based catalysts were studied
in our previous works [18,19]. The PANI-sulfate powder catalyst
and carbon supported catalysts with various loading of polyaniline-
sulfate (13–26 wt%) were tested. It was established that the PANI-S
powder as well as PANI-S-coated carbon samples could be used
repeatedly without essential loss in catalytic activity. In the present
work catalytic properties of PANI-S-coated CNT and silica are
compared with that of previously studied C-PANI-S catalyst with
capacity of acid sites 1.41 mmol H+/g. Here, methanolysis experi-
ments are carried out at the same conditions as those in previous
experiments. The methanol to triacetin molar ratio of 29:1 is used.
In order to evaluate any possible contribution of some PANI-
S soluble in methanol in the catalytic performance, the catalysts
were stirred for 3 h at 55 ◦C under methanol reflux. Then, methanol
obtained after filtration was used in methanolysis of triacetin. The
triacetin conversion of ca. 1% was observed after 30 min of reac-
tion. Our previous results showed a trace conversion of triacetin
(ca. 0.9% after 30 min of reaction) in blank catalytic experiment car-
ried out in the absence of catalyst [18,40]. Moreover, similarly to
previously used procedure, the obtained methanol filtrate was also
analyzed for the presence of sulfate ions using a sensitive analytical
test with Ba2+ ions. After addition of BaCl2 solution no precipitate
was observed. It seems therefore that no free sulfuric acid appears
in the catalysts.
C-PANI-S
SiO2-PANI-S
0
4
8
12
16
20
24
28
time [h]
Fig. 7. The conversion of TACT and the yield of methyl ester against reaction
time. Reaction conditions: catalyst concentration 15.8 g/dm3, temperature of 55 ◦C,
methanol: TACT = 29:1 molar ratio.
0.120 × 10−4 [mol (ME) min−1 g−1 (cat)] was obtained on PANI-S
powder catalyst.
The time conversion plots for the methanolysis of triacetin on
PANI-S coated catalysts demonstrate that all studied catalysts are
active in the reaction (Fig. 7). The most active is CNT-PANI-S catalyst
whereas the activity of other two catalysts is lower and PANI-S-
coated carbon exhibits the lowest activity. The same relation is
preserved when the yield of methyl esters is taken into consid-
eration (Fig. 7). The comparison of initial activities related to mass
of catalyst [expressed as mol TACT min−1 g−1 (catalyst)] (Table 3)
clearly demonstrates that CNT-supported catalyst is ca. 5-times
more active than the less active carbon-supported sample.
Fig. 8 shows the comparison of triacetin methanolysis in the
presence of the most active CNT-PANI-S catalyst and in homoge-
neous reaction catalyzed by dissolved sulfuric acid. According to
the literature and our previous results [18] from the very beginning
of homogeneous reaction catalyzed by sulfuric acid both partial
The activity of PANI-S coated catalysts is also compared with
that of previously studied PANI-S powder and Amberlyst-15 cat-
alysts. As a measure of catalyst activity, initial rate of triacetin
conversion (below 10% conversion) calculated per mass of catalyst
is assumed. The initial rates calculated per unit mass of catalyst are
collected in Table 3. As described before, the capacity of acid sites
in the PANI-S powder and in Amberlyst-15 are similar 4.75 and
4.70 mmol H+/g, respectively. The initial rate of triacetin conver-
sion in the presence of Amberlyst-15 (2.26 × 10−4 mol min−1 g−1
)
was determined to be only slightly higher than the rate on PANI-S
(1.76 × 10−4 mol min−1 g−1) powder catalysts [19]. The same rela-
tion was obtained in methanolysis of castor oil. The initial rate of
methyl ester (ME) formation on Amberlyst-15 was determined to
be 0.125 × 10−4 [mol (ME) min−1 g−1 (cat)]. A similar rate equal to

100
A. Drelinkiewicz et al. / Applied Catalysis A: General 455 (2013) 92–106
Table 3
Activity data obtained in methanolysis of triacetin and castor oil.
Catalyst
TACT methanolysisb
Castor oil methanolysisc
Conversion
(3 h) [%]
Initial rate [mol TACT
Initial rate [mol TACT
Initial rate [mol (ME)
min−1 g−1 (cat)] × 10−4
min−1 mol−1 H+] × 10−2
min−1 g−1 (cat)] × 10−4
Amberlyst-15
PANI-S powder
C-PANI-S
50.2
42.8
4.7
2.26
1.76
0.14
0.13a
0.42
0.38a
0.78
0.62a
4.80
3.70
1.00
0.98a
3.36
3.26a
6.70
6.30a
0.125
0.120
0.011
4.4a
SiO2-PANI-S
CNT-PANI-S
15.5
13.9a
31.4
28.7a
0.009
0.048
a
Data obtained in recycling use of the catalysts. The samples were treated with methanol/THF mixture.
Reaction conditions: temperature 55 ◦C, catalyst concentration 15.8 g/dm3, methanol:TACT = 29:1.
Reaction conditions: temperature 55 ◦C, catalyst concentration 43 g/dm3, methanol:TACT = 29:1.
b
c
glycerides diacetin and monoacetin are formed and their contents
pass through the maximum. The maximum yield of diacetin corre-
sponds to 13.6% and is attained at ca. 45% triacetin conversion. The
maximum yield of monoacetin is observed at higher triacetin con-
version, ca. 90%. In final solution, the triacetin conversion higher
than 95% could be attained. As Fig. 8 shows the course of con-
secutive reactions on CNT-PANI-S catalyst is similar to that under
homogeneous conditions. The maximum yield of diacetin (15.6%)
formed on CNT-PANI-S is only slightly higher than that in homoge-
neous reaction (13.6%). The difference between these two catalysts
is more pronounced relative to the monoacetin. The amount of
monoacetin formed on CNT-PANI-S catalyst is higher than that
under homogeneous conditions.
However, the course of consecutive reactions during methanol-
ysis of triacetin on CNT-PANI-S catalyst differs essentially from
those on other two C-PANI-S and SiO2-PANI-S catalysts. This is
evidenced by the conversion plots in Fig. 7. In the presence of CNT-
PANI-S catalyst the maximum triacetin conversion attains 91%. On
the other hand, in the presence of other two catalysts, after con-
sumption of ca. 30–40% of triacetin, the rate of its further conversion
starts to slow down remarkably. This effect suggests a blockage of
active sites in these two catalysts. To better recognize this effect it
is interesting to follow the formation of partial glycerides, diacetin
and monoacetin. As shown in Fig. 9 the change in their content is
strongly dependent on the type of catalyst. The yields of both partial
glycerides diacetin and monoacetin are the lowest on CNT-PANI-
S catalyst and as shown in Fig. 8 they pass through the maxima.
The contents of both partial glycerides are remarkably higher on
C-PANI-S and SiO2-PANI-S catalysts. Within the whole triacetin
conversion the yield of diacetin is the highest on microporous
carbon-supported C-PANI-S catalyst. At 40% TACT conversion a dis-
tinct plateau corresponding to diacetin yield of ca. 30% is reached
and after that the content of diacetin practically does not decrease.
The effect of plateau shows that although diacetin is formed, it is not
further reacted to monoacetin as well as the reverse transforma-
tion of diacetin to triacetin is strongly inhibited. This demonstrates
almost complete blockage of the active sites in the C-PANI-S cata-
lyst. As a result, the yield of methyl ester is definitively lower than
that predicted from the obtained conversion of triacetin. This is
evidenced by the graph showing the yield of methyl ester (Fig. 8).
On SiO2-PANI-S catalyst the yield of diacetin is lower than that
on C-PANI-S (Fig. 9) thus suggesting that although a blockage of
active sites appears, it is definitively weaker.
0,04
0,03
0,02
0,01
TACT
ESTER
DACT
MACT
CNT-PANI-S
0,00
0,04
0
5
10
15
20
time [h]
TACT
ESTER
DACT
MACT
0,03
0,02
0,01
0,00
H2SO4
0
1
2
3
4
5
time [min]
The reaching of plateau by diglyceride has been observed by
Cantrell et al. [45] in methanolysis over alkaline solid catalysts,
Mg–Al hydrotalcites as well as CaO and MgO catalysts [46]. This
effect was ascribed to a detrimental role of glycerol/partial glyce-
rides interactions with the active sites of alkaline catalysts which
resulted in the creation of glycerol/glycerides film onto the solid
Fig. 8. Reagents distribution in methanolysis of triacetin catalyzed by CNT-PANI-S
and soluble H2SO4. Methanolysis of TACT on CNT-PANI-S: catalyst concentra-
tion 15.8 g/dm3, temperature of 55 ◦C, methanol: TACT = 29:1 molar ratio. Reaction
catalyzed by soluble sulfuric acid: H2SO4 concentration 1,62 × 10−2 mol/dm3, tem-
perature of 55 ◦C, methanol: TACT = 29:1 molar ratio.

A. Drelinkiewicz et al. / Applied Catalysis A: General 455 (2013) 92–106
101
35
30
25
20
15
10
5
consideration as a measure of reaction rate. When the content of
both catalysts increases the transesterification occurs at a faster
rate. The reaction rate increases almost proportionally with the
growth in the concentration of the CNT-PANI-S catalyst within
the range employed in the studies (Fig. 10). According to Pappu
et al. [48] who studied kinetic model for triglyceride transesterifi-
cation on Amberlyst-15, the linear plot shows that the catalyst is
effectively distributed and active throughout the reaction mixture.
On the other hand, although the TACT methanolysis is faster
as the concentration of C-PANI-S catalyst increases, there is no
proportional growth.
Fig. 11 demonstrates the influence of reaction temperature on
the conversion of triacetin. The data obtained in the presence of
CNT-PANI-S catalyst are compared with those on C-PANI-S cata-
lyst. When the temperature increases from 35 to 55 ◦C the rate
of TACT conversion remarkably grows on CNT-PANI-S catalyst.
On the other hand, the temperature of reaction has definitively
weaker effect in the case of C-PANI-S catalyst. From the initial
rates of TACT transesterification (below 15% conversion) the acti-
vation energy is determined using Arrhenius plots reported in
Fig. 11. For both studied catalysts, the linear relationships are
obtained within the temperature range employed in experiments.
By linear regression analysis, the apparent activation energy is
evaluated to be 54.5 kJ/mol for TACT conversion on CNT-PANI-S
catalyst. Definitively lower value of activation energy, 14.3 kJ/mol
is calculated for C-PANI-S catalyst. The activation energy obtained
on CNT-PANI-S catalyst is close to the data obtained on powder
PANI-S (50.5 kJ/mol [19]) as well as to those reported by other
authors for methanolysis catalyzed by acid catalysts. For triacetin
methanolysis on Nafion SAC-13 composite activation energy of
48.5 kJ/mol was reported, which was comparable to that for trigly-
cerides methanolysis catalyzed by soluble acids (p-toluenesulfonic
50.7 kJ/mol, sulfuric 46.1 kJ/mol) [12,49]. Thus, apparent activation
energy for TACT methanolysis obtained on CNT-PANI-S catalyst is
within the range predicted by the other authors for reaction under
mass-transfer free conditions.
C-PANI-S
SiO2-PANI-S
CNT-PANI-S
0
0
10
20
30
40
50
60
conversion TACT [%]
35
CNT-PANI-S
30
25
20
15
10
5
C-PANI-S
SiO2-PANI-S
The effects observed for C-PANI-S catalyst namely very weak
influence of concentration of catalyst and reaction temperature
may suggest diffusion limitation. During transesterification of nat-
ural oils mass-transfer limitations may occur because the reaction
system contains a methanol phase, an oil phase and a solid-catalyst
phase. With the increase of methyl esters yield, biodiesel can
change the phase equilibrium and promote the mutual dissolving of
oils and methanol. When the produced biodiesel makes methanol
dissolve all oils completely, the system becomes a liquid–solid
reaction. Therefore, the reaction system contains a heterogeneous
liquid–liquid–solid reaction stage and a liquid–solid reaction stage.
However, reports on kinetics studies have focused mostly on homo-
geneous catalysts. Recently published few papers focusing on
kinetics on solid base as the catalyst demonstrated that methanoly-
sis of soybean oil on CaO, SrO was controlled by both mass-transfer
and reaction. However, the mass transfer resistance was higher
than the reaction resistance. The authors concluded that the main
mass transfer resistance lay on the surface of the catalyst parti-
cles [50]. Mass-transfer limitations resulting due to microporous
structure of Dowex-type resin were also observed to be crucial for
methanolysis of coconut oil [51].
In the present reaction mixture no phase separation exists,
because all reagents, triacetin, methanol and methyl esters are
well miscible. However, liquid–solid mass-transfer limitation may
be encountered and as described before, accumulation of glyc-
erol/partial glycerides on the active sites may be a barrier for the
transport of reactants to/from the active sites.
In order to follow catalysts blockage effect the recovered cata-
lysts are subjected to FT-IR and optical microscopy experiments.
The samples of catalysts were taken from the reaction mixture
0
0
20
40
60
80
100
conversion TACT [%]
Fig. 9. The yield of diacetin (YDI) and monoacetin (YMONO) in methanolysis catalyzed
by CNT-PANI-S, C-PANI-S and SiO2-PANI-S samples. Reaction conditions: catalyst
concentration 15.8 g/dm3, temperature of 55 ◦C, methanol: TACT = 29:1 molar ratio.
catalyst. The presence of this glycerol/glycerides film could be a
mass transfer barrier to the methanol and triglycerides as well
as it could facilitate the reverse reactions [46]. On the other
hand, esterification of glycerol with fatty acids to form monoglyc-
erides catalyzed by acid sites in Amberlyst resin was significantly
inhibited due to strong interaction of glycerol with the active sites.
This was result of specific solvation properties of polyols (glycerol)
leading to the formation of system of hydrogen bonds with the
participation of glycerol OH-groups and the acid groups of catalyst
[47].
Thus, in view of literature reports, the observed effect of
“plateau” reaching by diacetin may be ascribed to the accumula-
tion of glycerol/partial glycerides on the catalyst surface. This can be
result of strong interactions of more polar reagents with acid sites
of polyaniline-sulfate. However, textural properties of catalyst like
porosity as well as polymer morphology may also contribute to the
blockage effect. These effects are discussed in the next paragraph.
The effect of catalyst concentration is studied using CNT-PANI-S
and C-PANI-S samples (Fig. 10). In experiments, the amount
of catalyst was varied keeping other parameters constant. The
TACT conversion obtained after reaction time of 3 h is taken into

102
A. Drelinkiewicz et al. / Applied Catalysis A: General 455 (2013) 92–106
50
50
40
30
20
10
0
CNT-PANI-S
3
15,8 g/dm
C-PANI-S
40
30
3
10,5 g/dm
20
15,8g/dm3
10
3
5,3 g/dm
10,5g/dm3
5,3g/dm3
0
0
1
2
3
4
5
6
7
0
1
2
3
4
5
6
time [h]
time [h]
40
CNT-PANI-S
30
20
10
0
C-PANI-S
0
5
10
15
20
catalyst concentration [g/dm3 ]
Fig. 10. The effect of CNT-PANI-S and C-PANI-S catalysts concentration. Reaction conditions: temperature of 55 ◦C, methanol: TACT = 29:1 molar ratio.
(after 10 h of reaction) and allowed to stay in air up to dryness. The
FTIR spectra of fresh and recovered (not washed catalysts) are dis-
played in Fig. 1. For the comparison, the spectra of triacetin, diacetin
and glycerol are also reported (Fig. 12). A similar set of bands char-
acteristic of C O in ester groups can be observed in the spectra of
both glycerides. They are located at frequency of 1040, 1369 and
1740 cm−1. However, the position of the most intense band differs
in the frequency. This band is observed at 1212 cm−1 in the triacetin
spectrum and at higher frequency, i.e. 1230 cm−1 in the spectrum of
diacetin. Moreover, in the spectrum of diacetin a broad band at ca.
3400 cm−1 originating from the hydrogen bonds can be observed.
In the FT-IR spectrum of glycerol apart from a strong and broad
peak arising from the hydrogen bonds (at ca. 3400 cm−1) also an
intense band at 1040 cm−1 together with a not well resolved bands
within the range 1400–1600 cm−1 appear.
In the spectrum of recovered C-PANI-S catalyst apart from the
bands originating from the catalyst also a few new bands aris-
ing from the glycerides/glycerol can be observed. Well observable
are the bands at 1725, 1380 and 1240 cm−1 as well as a strong
band at 3400 cm−1 characteristic of hydrogen bonds. Their presence
and in particular the one at 1240 and 3400 cm−1 can be ascribed
to diacetin. However, it cannot be excluded from these spectral
features that apart from diacetin, glycerol also appears in the recov-
ered C-PANI-S catalyst. Similar spectral effects can be observed
for recovered CNT-PANI-S catalyst. In the spectrum of this catalyst
apart from a number of bands originating from the polyaniline and
CNT support also the band at 1748 cm−1 arising from the glycerides
appears. However, similarly to the previous spectrum for C-PANI-
S catalyst the band at 1230 cm−1 characteristic of diacetin can be
also recognized. It should be pointed out that in the spectrum of
recovered CNT-PANI-S catalyst the position of the bands charac-
teristic of protonated polyaniline does not change relative those in
the spectrum of initial catalyst. This indicates that polymer is still
in the protonated form and the acid sites are not removed during
the catalytic tests. It should be pointed out that spectral observa-
tions for SiO2-PANI-S catalyst are practically impossible owing to a
presence of strong band originating form silica material.
The microscopic images registered for initial and recovered cat-
alysts are displayed in Fig. 13. The optical microscopy was used
because this technique allows observing the samples without spe-
cial treatment procedure like vacuum treatment during the SEM
investigations. It can be seen that in all three recovered catalysts
most of the particles are glued together forming number of various
agglomerates. The particles of recovered carbon and silica sup-
ported catalysts seem to be more agglomerated than the particles
of CNT-containing catalyst.
Recycling experiments were carried out by recovering the
used catalysts samples and reusing them with fresh reagents in

A. Drelinkiewicz et al. / Applied Catalysis A: General 455 (2013) 92–106
103
100
80
60
40
20
0
100
C-PANI-S
CNT-PANI-S
80
60
40
o
55 C
o
55 C
o
o
35 C
45 C
o
45 C
20
o
35
0
0
2
4
6
8
10
12
0
2
4
6
8
10
12
time [h]
time [h]
-0,5
-1,5
-2,5
-3,5
-4,5
CNT-PANI-S
y = -6996,8x + 20,223
R2 = 0,999
C-PANI-S
y = -1816,1x + 1,8576
R2 = 0,99
0,003
0,0031
0,0032
0,0033
1/T [K-1]
Fig. 11. Methanolysis of triacetin in the presence of CNT-PANI-S and C-PANI-S catalysts. The effect of reaction temperature on the catalyst activity and Arrhenius plot.
Reaction conditions: catalyst concentration 15.8 g/dm3, methanol: TACT = 29:1 molar ratio.
subsequent reaction. The catalysts after first catalytic test were
subjected to simple regeneration by washing with methanol/THF
mixture followed by drying the samples in contact with air.
The obtained conversion of triacetin (Table 3) shows the activity
drop of ca. 10–15%. Similar effect what was observed in our previ-
ous work devoted to PANI-S powder catalyst [29]. It was observed
that methanolysis of triacetin resulted in slight reduction of acid
capacity in the polyaniline sulfate powder and carbon supported
PANI-S with low content of polymer (13.6 wt%). After successive
reaction cycles using PANI-S powder and this C-PANI-S sample an
acid capacities were determined to be 88.6 and 82.2% from the
initial ones [19]. However, the activity of fresh and reused cata-
lysts expressed as the rate (mol TACT/min) per active site (mol H+)
was almost stable. For sulfonic acid-based catalysts the reduction of
H+ content during methanolysis of triglycerides and esterification
with methanol was also observed by other authors [5,16]. This was
ascribed to the possible esterification of the sulfonic acid groups to
methyl sulfonates.
Here, the content of acid sites is determined for the samples
of recovered catalysts regenerated with methanol/THF mixture.
The obtained capacities 1.31, 1.16 and 0.98 mmol H+/g for C-PANI-
S, SiO2-PANI-S and CNT-PANI-S catalysts respectively, are slightly
lower than those in fresh catalysts. As mentioned above, the activ-
ity of recovered catalysts drop by ca. 10–15%. As the data in Table 3
show the activity of fresh and reused catalysts expressed as the rate
(mol TACT/min) per active site (mol H+) decreases only slightly.
Thus, the observed drop in activity upon recycling experiments
could be result of reduction in the content of acid sites, similarly
to the observations of other authors and previous effects obtained
for PANI-S powder catalyst [19]. These results indicate that organic
material including partial glycerides/glycerol appears on the cat-
alysts separated from the reactor. However, simple regeneration
of recovered catalysts by the THF/methanol washing procedure
restored almost completely their activity.
Let us compare the activity of PANI-S coated catalysts in
the methanolysis of vegetable oil, castor oil. As a measure of

104
A. Drelinkiewicz et al. / Applied Catalysis A: General 455 (2013) 92–106
SiO2-PANI-S catalysts are definitively less active and their activity
does not substantially differ. This latter effect may be explained tak-
ing into account that the access of active sites for large triglyceride
molecules present in castor oil is more difficult and the observed
activity could be mainly related to the centres in outermost surface
of the catalysts particles.
The esterification of ricinoleic acid with methanol is studied
using CNT-PANI-S and C-PANI-S catalysts. The conversion of rici-
noleic acid plotted against the esterification time is displayed in
Fig. 15. At the same loading of catalysts (42.9 g/dm3) the activity of
CNT-PANI-S catalyst is ca. 3.5 times higher than that of C-PANI-S
sample. This activity relation between CNT and C-supported cat-
alysts in esterification of ricinoleic acid is analogous to that in
methanolysis of castor oil. The CNT-PANI-S catalyst is ca. 3.5–4
times more active than other two, silica and carbon-supported sam-
ples.
The comparison of activities given in Table 3 demonstrates that
the initial activity of all catalysts with polyaniline sulfate coatings
in both transesterification reactions is lower than that of PANI-S
powder catalyst. However, if the activities related to the proton
capacity are taken into consideration, the best performance can be
observed for CNT-PANI-S catalyst. This may suggest lower utiliza-
tion of active sites in all other catalysts and among them in the
PANI-S powder sample.
Fig. 12. FT-IR spectra of triacetin, diacetin and glycerol.
catalyst activity initial rate of methyl ester formation (r(ME),
mol min−1 g−1) was assumed (Table 3). The yield of methyl esters
against reaction time is shown in Fig. 14. Similar activity relation
can be observed as that in methanolysis of triacetin. The CNT-PANI-
S catalyst exhibits the highest activity. The other two C-PANI-S and
Fig. 13. Images registered by optical microscope for fresh catalysts (left) and the samples recovered after reaction (react.) (right).

A. Drelinkiewicz et al. / Applied Catalysis A: General 455 (2013) 92–106
105
10
8
with a difference in the morphology of polymer coating and its loca-
tion throughout the catalysts particles. Porous structure of catalysts
could be also considered to be variable determining the observed
activity/blockage effects.
CNT-PANI-S
As−shown in Scheme 1 the protons in –NH+– groups or in
HSO4 are located throughout the polymer network. As a result,
the access of active sites for the reactants could be determined
by the morphology of polymer. However, polymer morphology
could also make difficult to some extent the migration of prod-
uct molecules from the active sites. In both less active C-PANI-S
and SiO2-PANI-S catalysts the coatings of polyaniline sulfate exhibit
similar morphology consisting of more or less aggregated fibrous
and branched networks. This type of morphology could limit the
number of readily accessible acidic sites because of the proton
location deeply inside the branched polymer nanorods. As a result,
accumulation of more polar products such as glycerol/partial gly-
cerides may be facilitated thus giving a barrier for the reaction. The
latter process could be farther facilitated by the microporous struc-
ture as in carbon-supported catalyst. On the other hand, although
fibrous network of polymer also appears in the silica-containing
catalyst blockage of active sites is weaker. This may be associated
with lower contribution of microporous structure (Table 1) as well
as deposition of polymer mostly onto outer surface of silica parti-
cles. It should be pointed out that the blockage of present catalysts
manifests very strongly because the protons are located throughout
the polyaniline sulfate coating only thus giving high surface den-
sity of acid sites. Although this surface location of active sites leads
to easy access for the triglycerides molecules it also facilitates a
blockage of active sites due to accumulation of more polar reagents.
Thus, aggregated and fibrous network of polyaniline sulfate has an
adverse influence on catalyst performance in terms of both activity
and blockage due to facilitated accumulation of glycerides/glycerol.
Maybe, fibrous branched structures of polyaniline sulfate creates
a specific spatial organization of the polymer chains resulting in
locally increased density of active sites (protons). This facilitates
interactions between active sites and polar reagents and in particu-
lar with glycerol. On the other hand, a combination of well extended
polyaniline sulfate coating together with the mesoporous structure
of CNT-PANI-S sample provides conditions for almost uniform sur-
face distribution of active sites. This gives easily accessible active
sites for the reactants and almost effectively prevents the accumu-
lation of polar reagents. These features make CNT material coated
with polyaniline sulfate an interesting catalyst for methanolysis of
triacetin, castor oil as well as for esterification of ricinoleic acid.
Hence, the obtained results are consistent with the literature sug-
gestions that catalysts with mesoporous structure are preferred for
the transesterification of triglycerides.
6
4
C-PANI-S
2
SiO2-PANI-S
0
0
2
4
6
8
10
time [h]
Fig. 14. The yield of methyl esters against reaction time in methanolysis of
castor oil in the presence of CNT-PANI-S, C-PANI-S and SiO2-PANI-S cata-
lysts. Reaction conditions: catalyst concentration 43 g/dm3, temperature of 55 ◦C,
methanol:triglyceride = 29:1 molar ratio.
The obtained data demonstrate that interaction between partial
glycerides/glycerol and active sites occur in all studied catalysts.
However, in the C-PANI-S and SiO2-PANI-S catalysts they result
in the effect of “plateau” reaching by partial glycerides and block-
age of catalysts. It facilitates the reverse reactions and gives lower
conversion of triglycerides. On the other hand, although organic
species are also observed on CNT-PANI-S catalyst, their presence
does not lead to a blockage of catalyst activity. The CNT material
coated with polyaniline sulfate is observed to be the most active
catalysts among the studied samples.
As shown in Table 2, the capacity of acid sites and their strength
are of similar range in the studied catalysts with polyaniline sul-
fate coatings. Therefore, the observed difference in the catalysts
performance in terms of activity and blockage could be associated
100
CNT-PANI-S
C-PANI-S
80
4. Conclusions
60
40
20
0
Polyaniline-sulfate deposited on three different carriers, namely
multi-wall carbon nanotubes (CNT), carbon and silica exhibited
activity for the transesterification of triacetin and castor oil with
methanol. Because of different textural and hydrophobic proper-
ties of these carriers, the polymer coatings of various morphologies
were obtained. The acid capacity and the strength of acid sites were
similar in all studied catalysts.
Uniform coating of CNT with polymer resulted in the most
extended polymer coating and the highest activity of the CNT-
PANI-S catalyst in all tested reactions. The course of reaction during
methanolysis of triacetin on CNT-containing catalyst was similar
to that in the presence of soluble sulfuric acid and the maximum
yield of methyl esters ca. 95% was attained. The properties of CNT-
PANI-S catalyst can be attributed to a combination of mesoporous
structure together with well extended polymer coating. This gives
easily accessible active sites for the reactants and does not hinder
0
1
2
3
4
5
6
7
8
time [h]
Fig. 15. The yield of methyl ester against reaction time in esterification of
ricinoleic acid with methanol in the presence of CNT-PANI-S and C-PANI-S cata-
lysts. Reaction conditions: catalyst concentration 43 g/dm3, temperature of 60 ◦C,
methanol:ricinoleic acid = 29:1 molar ratio.

106
A. Drelinkiewicz et al. / Applied Catalysis A: General 455 (2013) 92–106
the migration of reagents from the catalyst. Nanorods of polymer
forming branched dendritic structures were formed in carbon and
silica carriers. In the presence of these catalysts the accumulation
of partial glycerides/glycerol was observed. This was attributed to
the aggregated morphology of polymer coating resulting in a locally
high density of acid sites. Microporous structure of catalysts as in
carbon-supported sample facilitated also this products accumula-
tion effect. Too high density of the active sites, especially inside
the pore structure could facilitate interaction with polyol, such as
glycerol, which might lead to the accumulation of reagents with
consequent blockage of active sites.
[17] L. Sherry, J.A. Sullivan, Catal. Today 175 (2011) 471–476.
[18] A. Zieba, A. Drelinkiewicz, E.N. Konyushenko, J. Stejskal, Appl. Catal., A 383
(2010) 169–181.
[19] A. Zieba, A. Drelinkiewicz, P. Chmielarz, L. Matachowski, J. Stejskal, Appl. Catal.,
A 387 (2010) 13–25.
[20] S. Palaniappan, M. Sai Ram, Green Chem. 4 (2002) 53–55.
[21] M. Sai Ram, S. Palaniappan, J. Mol. Catal., A 201 (2003) 289–296.
[22] S. Palaniappan, A. John, J. Mol. Catal., A 233 (2005) 9–15.
[23] S. Palaniappan, R.C. Shekhar, Polym. Adv. Technol. 15 (2004) 140–143.
[24] E.T. Kang, K.G. Neoh, K.L. Tan, Prog. Polym. Sci. 23 (1998) 277–324.
[25] A. Malinauskas, Polymer 42 (2001) 3957–3972.
[26] J. Stejskal, I. Sapurina, M. Trchova, Prog. Polym. Sci. 35 (2010) 1420–1481.
[27] J. Stejskal, J. Prokesˇ, M. Trchová, React. Funct. Polym. 68 (2008) 1355–1361.
[28] J. Stejskal, D. Hlavatá, P. Holler, M. Trchová, J. Prokesˇ, I. Sapurina, Polym. Int. 53
(2004) 294–300.
[29] A. Riede, M. Helmstedt, I. Sapurina, J. Stejskal, J. Colloid Interface Sci. 248 (2002)
413–418.
Acknowledgement
[30] I. Sapurina, A. Osadchev, B.Z. Volchek, M. Trchová, A. Riede, J. Stejskal, Synth.
Met. 129 (2002) 29–37.
Studies have been partly supported by the EU Human Capital
Operation Program, Polish Project No. POKL.04.0101-00-434/08-
00.
[31] Z. Huang, P.-Ch. Wang, A.G. MacDiarmid, Lamgmuir 13 (1997) 6480–6484.
[32] C.G. Wu, Y.R. Yeh, J.Y. Chen, Y.H. Chiou, Polymer 42 (2001) 2877.
[33] S. Fedorova, J. Stejskal, Langmuir 18 (2002) 5630.
[34] P.-C. Wang, Z. Huang, A.G. MacDiarmid, Synth. Met. 101 (1999) 852–853.
[35] S.S. Pandey, W. Takashima, M. Fuchiwaki, K. Kaneto, Synth. Met. 135–136
(2003) 59–60.
References
[36] Y. Yang, M. Wan, J. Mater. Chem. 11 (2001) 2022.
[37] E.N. Konyushenko, J. Stejskal, M. Trchova, J. Hradil, J. Kovarova, J. Prokes, M.
Cieslar, J.-Y. Hwang, K.-H. Chen, I. Sapurina, Polymer 47 (2006) 5715–5723.
[38] I. Sapurina, M.E. Kompan, A.G. Zabrodski, J. Stejskal, M. Trchova, Russ. J. Elec-
trochem. 43 (2007) 528–536.
[39] J. Stejskal, M. Trchova, S. Fedorova, I. Sapurina, J. Zemek, Langmuir 19 (2003)
3013–3018.
[40] A. Zieba, L. Matachowski, E. Lalik, A. Drelinkiewicz, Catal. Lett. 127 (2009)
183–194.
[41] M.M. Ayad, E.A. Zaki, J. Stejskal, Thin Solid Films 515 (2007) 8381–8385.
[42] I.K. Avlyanov, Synth. Met. 102 (1999) 1272–1273.
[43] M.G. Deng, B.C. Yang, Y.D. Hu, J. Mater. Sci. 40 (5021) (2005), PANI-CNT.
[44] B. Philip, J. Xie, J.K. Abraham, V.K. Varadan, Polym. Bull. 53 (127) (2005), PANI-
CNT.
[45] D.G. Cantrell, L.J. Gillie, A.F. Lee, K. Wilson, Appl. Catal. A: Gen. 287 (2005)
183–190.
[46] C.S. MacLeod, A.P. Harvey, A.F. Lee, K. Wilson, Chem. Eng. J. 135 (2008)
63–70.
[1] E. Lotero, Y. Liu, D.E. Lopez, K. Suwannakarn, D.A. Bruce, J.G. Goodwin Jr., Ind.
Eng. Chem. Res. 44 (2005) 5353–5363.
[2] M. Zabeti, W.M.A. Daud, M.K. Aroua, Fuel Process. Technol. 90 (2009) 770–777.
[3] M. Di Serio,.R. Tesser, L. Pengmei, E. Santacesaria, Energy Fuels 22 (2008)
207–218.
[4] Z. Helwani, M.R. Othman, A. Aziz, J. Kim, W.J.N. Fernando, Appl. Catal., A 363
(2009) 1–10.
[5] J.A. Melero, J. Iglesias, G. Morales, Green Chem. 11 (2009) 1285–1308.
[6] J. Jitputti, B. Kitiyanan, P. Rangsunvigit, K. Bunyakiat, L. Attanatho, P. Jenvanit-
panjakul, Chem. Eng. J. 116 (2006) 61–66.
[7] D.E. Lopez, J.G. Goodwin Jr., D.A. Bruce, E. Lotero, Appl. Catal., A 295 (2005)
97–105.
[8] R.M. de Almedia, L.K. Noda, N.S. Goncalves, S.M.P. Menegethetti, M.R.
Meneghetti, Appl. Catal., A 347 (2008) 100–105.
[9] S. Furuta, H. Matsuhashi, K. Arata, Catal. Commun. 5 (2004) 721–723.
[10] S. Furuta, H. Matsuhashi, K. Arata, Biomass Bioenerg. 30 (2006) 870–873.
[11] M.A. Harmer, Q. Sun, Appl. Catal., A 221 (2001) 45–62.
[12] D.E. Lopez, J.G. Goodwin Jr., D.A. Bruce, J. Catal. 245 (2007) 381–391.
[13] B.M.E. Russbueldt, W.F. Hoelderich, Appl. Catal., A 362 (2009) 47–57.
[14] S.M. de Rezende, M. de Castro Reis, M.G. Reid, P.L. da Silva Jr., F.M.B. Coutinho,
R.A. da Silva San Gil, E.R. Lachter, Appl. Catal., A 349 (2008) 198–203.
[15] I.K. Mbaraka, K.J. McGuire, B.H. Shanks, Ind. Eng. Chem. Res. 45 (2006)
3022–3028.
[47] Y. Pouilloux, S. Abro, C. Vanhove, J. Barrault, J. Mol. Catal. A; 149 (1999)
243–254.
[48] V.K.S. Pappu, A.J. Yanez, L. Peerboom, E. Muller, C.T. Lira, D.J. Miller, Bioresour.
Technol. 102 (2011) 4270–4272.
[49] S. Furukawa, Y. Uehara, H. Yamasaki, Bioresour. Technol. 101 (2010)
3325–3332.
[50] X. Liu, X. Piao, Y. Wang, S. Zhu, J. Phys. Chem. A 114 (2010) 3750–3755.
[51] Ch.E.T. Co, M.C. Tan, J.A.R. Diamante, L.R.C. Yan, Catal. Today 174 (2011) 54–58.
[16] J.A. Melero, L.F. Bautista, G. Morales, J. Iglesias, D. Briones, Energy Fuels 23
(2009) 539–547.