Lithium and calcium based perovskite type oxides for ethylic transesterification

Article abstract of DOI:10.1016/j.cattod.2016.06.018

ABO₃ basic catalysts were synthesized by the nitrate-b-alanine solution combustion method at about 400 °C. The powders were formed by combustion route in times up to 5 min and presented high crystallinity and a single-phase, without the calcination step. XRD analyses showed that the oxides containing calcium (CaZrO₃ and Ca_0.9Li_0.6Zr_0.9O₃) presented perovskite structure with the orthorhombic system, whereas a monoclinic structure was observed in the oxide Li₂ZrO₃. The catalysts were characterized by FEG-SEM, TPD-CO₂, TGA, N₂ physisorption (BET) and ICP OES. The catalytic activity of the materials was tested in the transesterification of both methyl acetate (model reaction) and soybean oil with ethanol to produce biodiesel. Ca_0.9Li_0.6Zr_0.9O₃ was the most active oxide in the model reaction, reaching a maximum conversion in ethyl acetate of 92%, followed by the oxide Li₂ZrO₃, with the conversion of 64%. The CaZrO₃ catalyst conversely presented negligible catalytic activity. Reuse tests have indicated that the catalysts containing Li can be reused for up to six reaction cycles, with an average conversion of around 50% up to its sixth reuse. Ca_0.9Li_0.6Zr_0.9O₃ also shows promising for the biodiesel production since its FAEEs yield was good (60% m/m) under mild reaction conditions (molar ratio oil/ethanol 1:12 and 50 °C for 30 min). It was noticed that the incorporation of Li to calcium zirconate has resulted in more efficient and promising catalysts for the transesterification, what is possibly related to the textural and basic alterations occurred with the Li insertion. Although the metallic ions leaching of the most active materials has been detected, no reaction was observed in homogeneous phase.

Full text of DOI:10.1016/j.cattod.2016.06.018

G Model
CATTOD-10262; No. of Pages10
ARTICLE IN PRESS
Catalysis Today xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Lithium and calcium based perovskite type oxides for ethylic
transesterification
Alexandra M. Gonc¸ alvesa,b,∗, Renata A.B. Lima-Corrêab, José M. Assafb,
Ana R.A. Nogueirac
a Chemistry Department, Federal University of São Carlos, Rodovia Washington Luís, Km 235, CEP 13565-905, São Carlos, SP, Brazil
b Chemical Engineering Department, Federal University of São Carlos, Rodovia Washington Luís, Km 235, CEP 13565-905, São Carlos, SP, Brazil
c Embrapa Pecuária Sudeste, São Carlos, SP, Brazil
a r t i c l e i n f o
a b s t r a c t
Article history:
ABO3 basic catalysts were synthesized by the nitrate-b-alanine solution combustion method at about
400 ◦C. The powders were formed by combustion route in times up to 5 min and presented high crys-
tallinity and a single-phase, without the calcination step. XRD analyses showed that the oxides containing
calcium (CaZrO3 and Ca0.9Li0.6Zr0.9O3) presented perovskite structure with the orthorhombic system,
whereas a monoclinic structure was observed in the oxide Li2ZrO3. The catalysts were characterized by
FEG-SEM, TPD-CO2, TGA, N2 physisorption (BET) and ICP OES. The catalytic activity of the materials was
tested in the transesterification of both methyl acetate (model reaction) and soybean oil with ethanol to
produce biodiesel. Ca0.9Li0.6Zr0.9O3 was the most active oxide in the model reaction, reaching a maximum
conversion in ethyl acetate of 92%, followed by the oxide Li2ZrO3, with the conversion of 64%. The CaZrO3
catalyst conversely presented negligible catalytic activity. Reuse tests have indicated that the catalysts
containing Li can be reused for up to six reaction cycles, with an average conversion of around 50% up to
its sixth reuse. Ca0.9Li0.6Zr0.9O3 also shows promising for the biodiesel production since its FAEEs yield
was good (60% m/m) under mild reaction conditions (molar ratio oil/ethanol 1:12 and 50 ◦C for 30 min). It
was noticed that the incorporation of Li to calcium zirconate has resulted in more efficient and promising
catalysts for the transesterification, what is possibly related to the textural and basic alterations occurred
with the Li insertion. Although the metallic ions leaching of the most active materials has been detected,
no reaction was observed in homogeneous phase.
Received 1 December 2015
Received in revised form 6 June 2016
Accepted 7 June 2016
Available online xxx
Keywords:
Heterogeneous catalysis
Biodiesel
Perovskite
Transesterification
Lithium
Calcium
© 2016 Elsevier B.V. All rights reserved.
1. Introduction
harder to separate, and whose yields are usually lower about those
obtained during the methanolysis, once methoxide ion is more
The Biodiesel is a non-toxic, biodegradable fuel, from a renew-
able source, which is an important alternative for fossil fuel for use
in a sustainable approach, involving economic and environmentally
friendly aspects [1,2]. This biofuel is obtained by the transesterifi-
cation reaction of vegetable oils or animal fats with short chain
alcohols such as methanol and ethanol, in the presence of cata-
lysts [3,4]. The biodiesel synthesis through transesterification by
ethanolysis has been calling attention to the methanolysis process
because ethanol comes from renewable and non-toxic sources [2].
Besides that, Brazil is one of the biggest global producers of this
alcohol. In contrast, the use of ethanol leads to mixtures that are
reactive than the ethoxide [5].
The industrial biodiesel production employs homogeneous
basic catalysts, as alkoxides and sodium or potassium hydroxide
in the presence of methanol [6,7], mainly because they promote
the transesterification reaction efficiently at relatively low tem-
peratures [8]. However, a considerable inconvenience in the use
of these catalysts is a possible formation of emulsions and saponi-
fication resulting in a complex mixture of difficult separation [7,9].
Due to the possibility of the reuse of the catalyst and simplification
on the biodiesel purifying steps, several studies have been giv-
ing focus on the heterogeneous catalysis [8–10]. Oxides of alkaline
earth metals have been considered since they have demonstrated
less solubility in reaction mixtures and less corrosion of metallic
equipments. However, simple oxides such as CaO present consider-
ably low catalytic activity in comparison with that obtained in the
current homogeneous process, besides the possibility of reacting
∗
Corresponding author at: Chemistry Department, Federal University of São Car-
los, Rodovia Washington Luís, Km 235, CEP 13565-905, São Carlos, SP, Brazil.
E-mail address: alemary321@gmail.com (A.M. Gonc¸ alves).
http://dx.doi.org/10.1016/j.cattod.2016.06.018
0920-5861/© 2016 Elsevier B.V. All rights reserved.
Please cite this article in press as: A.M. Gonc¸ alves, et al., Lithium and calcium based perovskite type oxides for ethylic transesterification,
Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.06.018

G Model
CATTOD-10262; No. of Pages10
ARTICLE IN PRESS
2
A.M. Gonc¸ alves et al. / Catalysis Today xxx (2016) xxx–xxx
with atmospheric water and CO2 that causing deactivation [11]. In
this context, the use of mixed metal oxides is presented as an alter-
native to modifying the basic properties, improving the activity and
stability of the simple metal oxides [12].
such as the transesterification [28–30]. Watkins et al. investigated
the use of the Li-CaO oxide in the transesterification reaction and
observed increase of the basic force of the simple oxide (CaO) with
the addition of lithium leading a higher catalytic activity [31]. In
a recent research, zirconium oxides doped with alkaline metals
were applied in simultaneous esterification and transesterification
reactions. Among the evaluated oxides, Li/ZrO2 with a structure
associated with the Li2ZrO3 resulted in yields of fatty acid methyl
esters (FAMEs) higher than 90% under optimized conditions of reac-
tion [32].
In this context, the objectives of this research were to evaluate
the catalytic activity and stability of ABO3-type mixed oxides syn-
thesized by the combustion method. Besides that, it was proposed
to verify the viability of a Li/Ca-based perovskite-type oxide in the
biodiesel production by ethylic transesterification reaction.
Among the multi-metal oxides with possible applicability in the
transesterification, it can be mentioned the ABO3-type perovskites,
where A is an alkaline earth metal, alkaline or rare earth, and B
a transition metal. These compounds are versatile and its activity
highly improved through partial substitutions on the A and/or B
sites with few alterations in its structure [13]. Among the meth-
ods that can be applied, the synthesis by combustion reaction has
been highlighted as a promising for perovskite ABO3-type oxides
preparation, being it is a simple, efficient, and low-cost method. In
previous research, Ianos and Barvinschi obtained a CaZrO3 powder
with high crystallinity and good homogeneity, and with a larger
surface area by using a nitrate-alanine system in the combus-
tion reaction instead the synthesis of oxide via ceramic method
[14]. The combustion method using glycine-nitrate solution was
also employed by Boobalan et al. in the production of nanometric
CaZrO3 of orthorhombic perovskite with a single-phase crystalline
structure [15]. Li2ZrO3 was prepared from a modified combustion
method, where LiOH was employed with ZrO2, in the presence of
urea as a fuel. Nevertheless, the combustion with urea was not
shown as an efficient method for the formation of the referred
oxide [16]. Also studied as catalyst, CaZrO3 is currently being used
in a wide broad of applications, such as ceramic pigment [17,18],
ionic conductor for solid electrodes [19], and multi-layer ceramic
capacitors [20]. Its widespread use is due to its valuable thermal,
electrical, mechanical, and luminescence properties, such as high
melting point (2340 ◦C) and dielectric permittivity, and low dissi-
pation factor [15,21]. Likewise, some authors [22,23] have reported
for Li2ZrO3 an excellent thermophysical, chemical and mechanical
stability, which makes this a potential material for nuclear fusion
reactors [16,22].
A review of the literature reveals that there are few studies
dedicated to evaluating the activity of this type of material in the
transesterification reaction and biodiesel production. A perovskite-
type oxide with orthorhombic structure, Na0.1Ca0.9TiO3, was
obtained by alkaline hydrothermal synthesis and was investi-
gated in the biodiesel production (optimized conditions: ratio
oil/methanol of 1:36; T = 60 ◦C and 8 h of reaction), resulting in
a yield of 93% [24]. Kawashima et al. investigated A-B-O type
heterogeneous basic catalysts in the methanolic transesterifica-
tion of rapeseed oil and observed that the oxides containing
calcium, among them CaZrO3, presented enough activity and cat-
alytic stability [25]. A new phase of CaZrO3, formed by CaO-ZrO2,
and synthesized by combustion reaction with nitrate and urea
was employed in the biodiesel production from soybean oil with
methanol (molar ratio equals to 1:25 and 65 ◦C). The results showed
that by increasing the molar ratio Ca/Zr, the catalytic activity is
increased, being it associated with the higher number of basic
sites on the catalyst surface [26]. In a more recent study, Kesic´
et al. performed a study on catalytic performance of the single-
phase perovskites CaZrO3, CaTiO3, CaMnO3 and Ca2Fe2O5 in the
methanolic transesterification of sunflower oil. These catalysts
were prepared by calcination of mechanochemically treated sto-
ichiometric mixtures of CaCO3-ZrO2, CaCO3-TiO2, CaCO3-MnO2
and CaCO3-Fe2O3. Under mild reaction conditions (6:1 molar ratio
methanol/oil, 10 wt.% of catalyst and T = 60 ◦C), insignificant or even
no catalytic activity was obtained. In contrast, when the transester-
ification reaction was carried out at 165 ◦C, molar ratio of 10:1 and
2 wt.% of pure perovskite, oil yields greater than 90% were achieved
after 3 h of reaction [27].
2. Material and methods
2.1. Materials
All the solutions were prepared using analytical grade reagents
and ultrapure water obtained in a MilliQ® (Millipore, Billerica,
MA, USA) system. For the synthesis of oxides, the following
reagents were used: ␤-alanine (99%, Sigma Aldrich), calcium nitrate
tetrahydrate (99–103%, Sigma Aldrich), zirconyl nitrate (99%, Sigma
Aldrich) e lithium nitrate (≥ 99%, Acros Organics). In the catalytic
tests and chromatographic analysis were used methyl acetate (99%,
Sigma-Aldrich), ethanol (anhydrous, 99.9% UV-IR-HPLC, Panreac),
methyl heptadecanoate (≥ 99% Sigma-Aldrich), n-heptane (anhy-
drous, 99% Sigma-Aldrich) and commercial soybean oil (purchased
from the market). For wet digestion were applied nitric acid (65%
v/v) and chloridric acid (37% v/v), distilled in quartz sub-boiling
equipment, and hydrogen peroxide (30% m/v) (Merck). Reference
solutions for the analytical calibration were prepared from the
stock solutions 1000 mgL−1 of Zr (SpecSol), Ca (Sigma-Aldrich) and
Li (Sigma-Aldrich).
2.2. Catalysts preparation by solution combustion
The ␤-alanine 0.8 mol L−1 solution was prepared with 5 mL
of deionized water and used as fuel. To synthesize the oxide
CaZrO3, Ca(NO3)2·4H2O and ZrO(NO3)2 were used as oxidant,
adding 0.03 mol of each nitrate to the ␤-alanine solution under
continuous stirring until complete dissolution. The oxide Li2ZrO3
was prepared following the same procedure by mixing 0.06 mol of
LiNO3 and 0.03 mol of ZrO(NO3)2 to the ␤-alanine solution. Simi-
larly to produce the calcium and lithium containing mixed oxide
(previously represented by CaLiZrO3) were used Ca(NO3)2·4H2O,
ZrO(NO3)2 and LiNO3 in molar quantities of 0.02, 0.03 and 0.04,
respectively. Each clear transparent solution was introduced into
the quartz bottles then placed in a heating mantle at 400 ◦C to start
the solution combustion, forming voluminous powders with low
density, which were pressed in an agate mortar.
2.3. Catalysts characterization
The phase composition was determined by powder X-ray
diffraction (XRD) in a Rigaku Multiflex diffractometer using radi-
ation Cu-K␣ (␭ = 0.15418 nm). Data were collected over a 2␪ range
of 10−80◦ with steps of 0.02◦␪ at a scanning speed of 2◦␪ min−1
.
The crystalline phases were identified using reference patterns of
the Joint Committee on Powder Diffraction Standards (JCPDS). The
crystallite size was calculated based on XRD using the Scherrer’s
equation [33]. The structural information was obtained by refine-
ment of crystal structures by the Rietveld method using the GSAS
Some works have also shown that the incorporation of lithium
to oxides promotes a high basicity and, consequently, an increase
in the performance of the compounds in base-catalyzed reactions
Please cite this article in press as: A.M. Gonc¸ alves, et al., Lithium and calcium based perovskite type oxides for ethylic transesterification,
Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.06.018

G Model
CATTOD-10262; No. of Pages10
ARTICLE IN PRESS
A.M. Gonc¸ alves et al. / Catalysis Today xxx (2016) xxx–xxx
3
Table 1
1100 rpm. These vials (semi-reactors) were disposed equidistantly
on a PTFE Teflon base into the jacketed reactor, with the capac-
ity to accommodate up to 6 semi-reactors, attached to the ultra
thermostatic bath to temperature control. The catalytic activity of
the oxides CaLiZrO3 and CaZrO3 was previously studied in three
consecutive reaction cycles under the following conditions: tem-
perature of 50 ◦C, molar ratio methyl acetate/ethanol at 1:6, 10 wt.%
of catalyst (related to the total mass) and 30 min of reaction.
Homogeneous contribution for the most active catalyst
(CaLiZrO3) was investigated. The catalysts were separated from the
reaction mixture by centrifugation, followed by filtering after 5 min
of reaction. The solution composition was analyzed by GC-FID, and
the remaining solution was exposed again to the same reaction
conditions. Representative aliquots were withdrawn periodically
(10, 30, 60, and 90 min) for chromatographic analysis. Simultane-
ously a kinetic study of the reaction was carried out under the
same conditions of synthesis and analysis. The influence of the
ratio ester/ethanol was also evaluated with this oxide using three
different molar proportions (1:6, 1:9, and 1:12).
From these outcomes previously observed, the activity and sta-
bility of the catalysts CaZrO3, Li2ZrO3 e CaLiZrO3 were then studied
by reuse tests for a higher number of batches (6 consecutive reac-
tion cycles). The following conditions were employed: temperature
of 50 ◦C, molar ratio methyl acetate/ethanol at 1:12, 10 wt.% of cat-
alyst and 30 min of reaction. Catalysts were separated from the
reaction mixture by centrifugation in two stages for 5 min each
(at 5000 and 9000 rpm, respectively). The recovered solids were
dried in their semi-reactor at 50 ◦C before being reused, and the
liquid phase was analyzed by GC-FID. Besides that, in order to ver-
ify the occurrence of leaching, the determination of Ca, Li, and Zr
was performed in liquid phase obtained after each reaction cycle
by ICP OES with axial view employing the following wavelengths:
Ca (422.673 nm), Li (610.365 nm), and Zr (349.619 nm). The sam-
ple preparation also consisted of an aliquot of the liquid phase in
a glass bottle and heated to 60 ◦C to eliminate organic solvents,
avoiding the extinction of the argon plasma during the analysis.
After solvent evaporating, the HNO3 solution (1% v/v) was added to
the remaining solid for dissolution.
The catalytic activity in the transesterification reaction of soy-
bean oil was carried out using the most active catalyst in the
tests with model reaction. The same batch-type reaction system
used in the model transesterification was employed under the
following conditions: temperature at 50 ◦C, duration of 30 min,
molar ratio 1:12 (oil/ethanol), and 10 wt.% of catalyst (related
to the oil mass). The catalyst was mixed with the ethanol in
the semi-reactor and after homogenization, the oil was added.
After the end of the reaction, the catalyst was easily separated
by centrifugation at 5000 rpm for 5 min. The purification of the
product consisted in centrifugation the liquid phase for 5 min at
8000 rpm and washing with hot water at 50 ◦C followed by cen-
trifugation for biodiesel separation from the reaction mixture.
Quantitative determination of the yield of fatty acid ethyl esters was
performed by GC-FID equipped with Rtx®-Wax capillary column
(30 m × 0.32 mm × 0.25 ␮m), according to the European Standard
EN14103, using methyl heptadecanoate 10 mg mL−1 (diluted in n-
heptane) as an internal standard.
Heating program employed for the microwave assisted acid digestion of the
catalysts.
Temperature (◦C)
Ramp (◦C min−1
)
Hold (min)
Fan
1
2
120
150
160
0
15
15
20
–
02:00
02:00
15:00
20:00
0
1
1
2
3
4a
a
Cooling.
program. The refinement Rietveld of the XRD results was per-
formed with the structural data ICSD as initial parameters. The
Fourier transform infrared spectroscopy (FT-IR) adsorption spec-
tra were obtained using an FT-IR spectrophotometer (Vertex 70,
Bruker) employing the KBr pellet method in the range of wavenum-
ber in 4000–400 cm−1. The surface area was measured by the
BET method in the Micromeritics ASAP 2020 equipment. Mor-
phology of the materials was investigated using a field-emission
gun scanning electronic microscope (FEG-SEM, Jeol JSM-6701F).
Thermogravimetric analysis (TGA) was performed on DRG–60H
Shimadzu (Simultaneous DTA-TG apparatus) equipment. The pow-
ders were heated under N2 atmosphere (50 mL min−1) until 1000 ◦C
at a rate of 10 ◦C min−1. The heating behavior of the precursor
solution also was investigated by TGA until 800 ◦C at the same
conditions. The basic properties of the catalysts were evaluated
by temperature-programmed desorption of CO2 (TPD-CO2) on an
AutoChem II Chemisorption Analyser Micromeritics, equipped with
a thermal conductivity detector (TCD). The samples (around 50 mg)
were pre-treated under He atmosphere (50 mL min−1) and heated
at a rate of 20 ◦C min−1 until 1000 ◦C to promote an impurity
removal. After cooling to 100 ◦C, the adsorption of CO2 was car-
ried out for 5 min. Physically adsorbed CO2 was removed for 1 h
by the flow of He at 30 mLmin−1. Finally, the CO2 desorption was
quantified using TCD by heating the sample at a rate of 10 ◦C min−1
to 1000 ◦C. A microwave-assisted acid digestion of the solids was
performed applying the heating program presented in Table 1 using
microwave oven (Anton Paar Multiwave® 3000 GmbH) equipped
with 16 closed vessels of modified polytetrafluoroethylene (PTFE-
TFM) and operated with fixed power of 1200 W. It was added 4.5 mL
of HNO3 and 1.5 mL of HCl, both concentrated, in the reaction
vessel containing 50 mg of the sample, keeping at room temper-
ature for 20 min for a pre-digestion and release of the acid vapor
excess. Then were added 2 mL of H2O2, closing the vessels after
10 min. After complete mineralization, the solutions were diluted
in deionized water for elemental analysis. The metals concentration
were quantified by inductively coupled plasma optical emission
spectrometry (ICP OES), in an equipment Thermo Fisher Scientific,
model iCAP6000 double-view Series, employing axial vision and
the wavelength of: Ca (317.9 nm), Li (610.3 nm), and Zr (349.6 nm).
2.4. Catalytic tests in transesterification reaction: activity and
stability
The activity and catalytic stability of the synthesized oxides
were evaluated in the model reaction of transesterification
between methyl acetate and ethanol. These low molar mass
reagents can be rapidly identified by gas chromatography,
allowing a good analytical frequency in the evaluation of the
synthesized materials. The conversions into ethyl acetate and
methanol were verified in a gas chromatograph with flame
ionization detection (GC-FID, Shimadzu GC-2010AF), equipped
with injector type split/splitless and Rtx®-1 capillary col-
umn (30 m × 0.32 mm × 3 ␮m). The reaction was performed in
headspace vials (Supelco®) of 20 mL hermetically sealed with a
screw cap and septum of PTFE/silicon using magnetic stirring at
3. Results and discussion
3.1. Catalyst characterization
The solution combustion is a fast and inexpensive method.
The exothermic auto-sustainable reactions provide nanocrys-
talline compounds with high purity and good homogeneity as the
high temperature generated by the flame allow the expelling of
Please cite this article in press as: A.M. Gonc¸ alves, et al., Lithium and calcium based perovskite type oxides for ethylic transesterification,
Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.06.018

G Model
CATTOD-10262; No. of Pages10
ARTICLE IN PRESS
4
A.M. Gonc¸ alves et al. / Catalysis Today xxx (2016) xxx–xxx
Thermal analysis of the precursor compounds (Fig. 2) showed
the temperatures of transformations in the formation of the oxides
containing lithium. DTA curve of the precursor of CaLiZrO3 (Fig. 2a)
indicates that an endothermic process takes place between 60
and 120 ◦C, with a mass loss around 30% as seen in the TG curve,
which could be associated with water removal. The ignition pro-
cess occurred at about 212–246 ◦C indicated by exothermic peaks
and mass loss of about 70%, typical of combustion reactions. This
quick mass loss can be associated to the decrepitation phenomenon
[35], which after the combustion reaction the sample undergoes
expansion forming a fine, fluffy powder that is either expelled or
ejected from the crucible. DTA curve of the Li2ZrO3 precursor solu-
tion (Fig. 2b) showed the occurrence of an endothermic process
between 60 and 120 ◦C associated again with a mass loss of 30%.
The ignition of the combustion reaction occurred between 240 and
265 ◦C with a total mass loss after this exothermic process. In pre-
vious research, it was found that the ignition of the combustion
process for CaZrO3 formation occurs between 245 and 255 ◦C [14],
confirming the results obtained in this work.
The thermal stability of the oxides containing lithium was also
investigated, and the TG/DTA curves are showed in Fig. 3. Total
mass loss of about 8% occurred up 700 ◦C. CaZrO3 presented similar
results. In previous paper was reported weight loss of 9% up to
600 ◦C for CaZrO3 perovskite [15]. Initially, a little mass loss was
observed for the synthesized oxides, probably due to the adsorbed
water, since Li+ ions are moisture sensitive [36]. However, at the
range of 500–800 ◦C, it is possible to observe the occurrence of a
mass loss around 4%, possibly due to the presence of others species
or by the occurrence of some chemical transformation.
Fig. 1. Selected images obtained from a time-lapse sequence showing the formation
of Li2ZrO3 by combustion reaction.
temporary impurity [15]. The synthesis by combustion involves the
exothermic reaction of organic nitrogenated fuel with an oxidant,
normally the metallic nitrates forming voluminous powders highly
homogeneous [34]. Ianos and Barvinschi noticed the formation of a
pure and crystalline powder of CaZrO3 from the combustion reac-
tion while the use of ceramic method resulted in the formation
of several crystalline phases [14]. In this work, the formation of
CaZrO3 by combustion was observed after 5 min of reaction, while
to obtain CaLiZrO3 and Li2ZrO3, the maximum reaction time was
4.45 min. In the formation of lithium-containing oxides, the mix-
ture boiling started at a time below 1.00 min having combustion
imminence after 4.10 min. The ignition process started at 4.34 min
being evidenced as an extremely incandescent area, and its end-
ing was registered for t = 4.44 min. The combustion produced a fine
and voluminous white powder. The previously mentioned times
were obtained through visual observation of the experiments dur-
ing the performance of the reactions, and they are represented in
Fig. 1, that shows images obtained at different reaction times in the
Li2ZrO3 synthesis.
The FT-IR was then employed to investigate the spectra of
the powders (Fig. 4). The IR spectrum shows no evidence for the
100
80
60
40
20
0
240
(a)
(b)
265
DTA
246
TG
TG
212
DTA
75
98
100
0
200
300
400
500
600
700
100
200
300
400
500
600
700
Temperature (°C)
Temperature (°C)
Fig. 2. TG and DTA curves of the precursor solutions of (a) CaLiZrO3 and (b) Li2ZrO3.
TG
100
DTA
98
96
94
92
90
TG
DTA
0
200
400
600
800
1000
0
200
400
600
800
1000
Temperature (°C)
Temperature (°C)
Fig. 3. TG and DTA curves of the oxides (a) CaLiZrO3 and (b) Li2ZrO3.
Please cite this article in press as: A.M. Gonc¸ alves, et al., Lithium and calcium based perovskite type oxides for ethylic transesterification,
Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.06.018

G Model
CATTOD-10262; No. of Pages10
ARTICLE IN PRESS
A.M. Gonc¸ alves et al. / Catalysis Today xxx (2016) xxx–xxx
Table 2
5
Lithium zirconate
Loadings of Ca, Li and Zr (wt.%) in the oxides (mean S.D., n = 3).
Catalyst
Metal Loadings (wt.%)
Ca
1620
Li
Zr
860
CaZrO3
Li2ZrO3
CaLiZrO3
22.3 0.60
–
21.5 1.08
–
50.4 0.68
60.4 0.47
48.7 1.90
3450
1500
1440
9.16 0.08
2.40 0.24
550
Calcium-lithium zirconate
100% (Zr) related to the nominal levels, indicating efficiency in the
decomposition of the oxides and good accuracy of the method. With
the concentration obtained for the evaluated metals, was possible
to predict that Ca0.9Li0.6Zr0.9O3 was the molecular formula of the
calcium/lithium zirconate. The limits of quantification (LOQs) were
calculated considering the concept of Background Equivalent Con-
centration (BEC). The LOQs obtained for Ca, Li, and Zr were 0.097,
0.024 and 0.009 mg L−1, respectively.
Calcium zirconate
Fig. 5 shows the XRD of oxides synthesized by combustion,
where it is possible to notice the presence of intense and symmet-
rically defined diffraction peaks, indicating a high crystallinity, a
typical characteristic of perovskites structures [13]. Besides, it is
revealed a crystalline single-phase structure, showing the forma-
tion of pure oxides without calcination. The CaZrO3 sample showed
a single main peak with 2␪ at 31.55, suggesting orthorhombic
4000
3500
3000
2500
2000
1500
1000
500
-1
Wave number (cm )
Fig. 4. FT-IR spectrum of the oxides.
presence of organic intermediates in the samples, except for the
band at around 3450 cm−1, which is related to the O H stretching
vibration mode of water molecules [37,38], all peaks are charac-
teristics of the materials studied [37,39]. The bands in the region
between 860 and 400 cm−1 indicating the presence of M O bonds
[40], while the absorption peaks at approximately 500 cm−1 are
assigned to the Zr O vibration [40,41]. However, the CO2 vibration
peaks are observed in the spectrum at 1620–1500 cm−1, corre-
sponding to C O bond [38,42], most notably for Li2ZrO3 and
CaLiZrO3 oxides. Some studies have found that Li2ZrO3 is a good
adsorbent of CO2 [42–44]. In these studies, a mechanism of reac-
tion at high temperature proposes the formation of Li2CO3 on the
surface and a dense ZrO2 shell in the middle covering the unreacted
Li2ZrO3. The effects of CO2 sorption promote the Li+ and O2− forma-
tion in the reversible reaction with Li2ZrO3 [42–44]. Therefore, the
mass changes observed for oxides containing lithium in the TGA at
500–800 ◦C may be attributed to chemical structural transforma-
tions, which occur due to adsorbed atmospheric CO2 according to
FT-IR analysis. The DTA curves for CaLiZrO3 showed an exothermic
process in the mentioned region. Hwang et al. [42] noted a signifi-
cant weight change during CO2 sorption and desorption process at
500–800 ◦C.
Ceramic materials such as multi-metal oxides are hardly soluble
compounds. Nevertheless, the acid digestion microwave-assisted
of the powders, employing a mixture of HNO3 and HCl at the
inverted aqua regia (3:1) plus H2O2 enabled the achievement of
clear solutions without the presence of precipitates, indicating
the total samples solubilization. The use of diluted acids was not
possible due to precipitation of zirconium compound insoluble.
Work developed by Niemelä et al. pointed that the use of aqua
regia associated to the microwave radiation presented efficiency
in the digestion of catalytic materials based on Al2O3, Al2O3–TiO2
and Al2O3–CeO2, not being observed any considerable difference
related to the method employing hydrofluoric acid [45]. The ele-
mental determination obtained by ICP OES (see Table 2) resulted
in a stoichiometric average percentage of 101% (Ca), 100% (Li) and
Fig. 5. XRD of the prepared samples by solution combustion synthesis.
Please cite this article in press as: A.M. Gonc¸ alves, et al., Lithium and calcium based perovskite type oxides for ethylic transesterification,
Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.06.018

G Model
CATTOD-10262; No. of Pages10
ARTICLE IN PRESS
6
A.M. Gonc¸ alves et al. / Catalysis Today xxx (2016) xxx–xxx
Table 3
BET surface area (SBET), diameter pore (DBET) and crystallite size (DXRD) of the
catalysts.
Catalysts
Physical Structures
SBET (m2 g−1
)
DBET (nm)
DXRD (nm)
CaZrO3
Li2ZrO3
CaLiZrO3
22.7
2.30
4.70
14.8
9.1
10.2
43.9
210.4
116.7
symmetry (JCPDS: 76-2401) [14]. According to Merino et al.
it is possible to identify the orthorhombic and rhombohedral
symmetries by analyzing the diffraction peak with the highest
intensity. In the orthorhombic structures the main peak is simple,
meanwhile in the rhombohedral a more intense peak is a doublet
[46]. XRD of Li2ZrO3 presented a monoclinic structure relative
to this pure oxide (JCPDS: 16-0263) [44]. The diffractions for the
oxide Ca0.9Li0.6Zr0.9O3 presented a high similarity to the perovskite
profile with formula Ca0.97Li0.03Ta0.03Zr0.97O3 (JCPDS 82-1926) and
orthorhombic crystalline system [47]. It was not found in the spe-
cialized literature any other material already synthesized whose
composition was closer to that obtained with this perovskite mixed
oxide. The refined orthorhombic lattice parameters for the CaZrO3
(data 37264-ICSD) are a = 5.7669 Å, b = 8.0242 Å, c = 5.5990 Å
resulting in V = 258.113 Å3 and d = 4.65 g cm−3 with ␹2 = 2.364%
and Rwp2 = 4.63%. The refinement of Ca0.9Li0.6Zr0.9O3 (data 78030-
ICSD) also orthorhombic converged with a = 5.758563 Å,
b = 8.013703 Å, c = 5.587543 Å, V = 257.851 Å3 and d = 4.388 g cm−3
,
␹2 = 5.975% and Rwp2 = 8.02%. Concerning the monoclinic Li2ZrO3
(data 35236-ICSD) the parameters obtained are a = 5.424987 Å,
b = 9.029775 Å, c = 5.419316 Å, V = 244.923 Å3 and d = 4.152 g cm−3
with ␹2 = 5.897% and Rwp2 = 4.37%. Concerning the mono-
clinic Li2ZrO3 (data 35236-ICSD), the parameters obtained are
a = 5.424987 Å, b = 9.029775 Å, c = 5.419316 Å, V = 244.923 Å3 and
d = 4.152 g cm−3 with ␹2 = 5.897% and Rwp2 = 4.37%. The results
are compatible with the reference data. The Li2ZrO3 showed lower
unit cell volume and density values compared to others oxides. It is
observed that the presence of Li in the perovskite structure CaZrO3
resulted in a decrease of lattice parameters. Smith and West [47]
reported a decrease in unit cell parameters, which reduces the size
of the cavity formed by the BO6 octahedral, accommodating both
Ca and Li atoms and the distortion of the octahedral decreases
with increase the Li content.
The specific surface area (BET method), pore diameter and crys-
tallite size of all studied catalysts are showed in Table 3. It is
observed that the lithium incorporation to form Ca0.9Li0.6Zr0.9O3
caused a considerable reduction in the value of the specific area
and the pore size (reductions of around 79% and 31%, respectively,
related to the oxide CaZrO3), but an increase on the crystallite,
which went of 43.9–116.7 nm. Many authors have associated these
morphological changes caused by the addition of Li+ to the sintering
effect that occurs at high temperatures [28,29]. As a result, particles
with smoother appearance are formed, and there is a considerable
reduction in the specific surface area of the catalysts [48]. Within
the nanometric materials, Li2ZrO3 showed a higher crystallite size
(210.4 nm) and smaller specific area values (2.30 m2 g−1) and pores
diameter (9.1 nm).
Fig. 6. FEG-SEM micrographs of the oxides (a) CaZrO3, (b) Ca0.9Li0.6Zr0.9O3 and (c)
Li2ZrO3.
Fig. 6 shows micrographs obtained by FEG-SEM to investigate
the morphology of the materials where it is initially possible to
notice that all the oxides particles are in nanometric scale. It seems
that the average diameter of the particles tends to increase with
the increase of lithium load in the sample, that meets the specific
area and crystallite size results previously shown (see Table 3). The
formation of bigger particles can lead to the reduction of the spe-
cific surface area where the addition and/or substitution of an atom
by another can cause an electronic alteration as a result of the load
and structural change, due to the difference of the atomic radius,
influencing the particle size and its area value [49]. It is possible
to observe that the oxides containing calcium (Fig. 6a) are formed
by highly clustered particles in the form of nanometric granules,
rounded, angular and irregular. These oxides present spherical
shape, typical of perovskites structures [50]. However, the Li2ZrO3
micrograph (Fig. 6c) shows the presence of molten particles,
Please cite this article in press as: A.M. Gonc¸ alves, et al., Lithium and calcium based perovskite type oxides for ethylic transesterification,
Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.06.018

G Model
CATTOD-10262; No. of Pages10
ARTICLE IN PRESS
A.M. Gonc¸ alves et al. / Catalysis Today xxx (2016) xxx–xxx
7
Table 4
Ca Li Zr O
Basic sites density of the perovskite-type catalysts.
3
.9 .6 .9
Catalysts
Density of base sites nb (␮mol m−2
)
Weak nb
Medium nb
Strong nb
Total nb
CaZrO3
Li2ZrO3
CaLiZrO3
0.70
1.03
0.04
3.10
5.17
7.04
0.58
29.78
7.85
4.38
35.98
14.93
CaZrO3
Li ZrO
2
3
sites (29.78 ␮mol m−2). Although the oxide Ca0.9Li0.6Zr0.9O3 pre-
sented a significant amount of basic sites (3.4 times higher than the
sites density obtained for CaZrO3), it presented lower total density
when compared to Li2ZrO3.
3.2. Catalytic tests
The catalytic activity of CaZrO3 and Ca0.9Li0.6Zr0.9O3 (also rep-
resented as CaLiZrO3) was previously evaluated in the model
transesterification reaction in three successive reaction cycles to
verify their applicability. The results obtained in these trials showed
that the use of the catalyst CaZrO3 did not lead to considerable con-
version values, considering that there was almost no activity in the
3rd cycle. In similar reaction conditions, the CaLiZrO3 showed high
conversions (82% in the 1 st reaction cycle) in small time like 30 min,
apparently presenting as a promising catalyst for the proposed
application. Using an oxide with Ca/Zr, and molar ratio similar to
CaZrO3 investigated in this work, Xia et al. [26] found low activity
in the transesterification of soybean oil with methanol (15% FAME
yield at 3 h of reaction), even with the use of more drastic reac-
tion conditions (methanol to oil molar ratio = 25:1 and temperature
of 65 ◦C). Our findings are corroborated with those of Kesic´ et al.
who obtained no catalytic activity for methanolysis of sunflower
oil using CaZrO3 at 60 ◦C, molar ratio of methanol to oil of 6:1 and
10 wt.% of catalyst [27]. According to these authors, the high cat-
alytic activity reported for CaZrO3 at 60 ◦C in the Kawashima’s work
[25] may be explained because this oxide was a mixture of CaO and
CaZrO3 phases, and even a small amount of CaO can be active for
the transesterification reaction.
100
200
300
400
500
600
700
Temperature (°C)
Fig. 7. TPD of CO2 profiles of the perovskite-type catalysts.
forming bigger agglomerate with a dense appearance (sintering
effect). Cruz et al. also observed the formation of dense and agglom-
erated particles, presenting sizes between 0.5 and 1 ␮m in the
synthesis of Li2ZrO3 by combustion using LiOH and urea [16]. Thus
according to the characterizations performed it was found the for-
mation of perovskite oxides such as CaZrO3 and Ca0.9Li0.6Zr0.9O3.
The TPD- CO2 technique was applied to evaluate the basicity
of the catalysts. Usually, the higher the desorption temperature of
CO2 the higher will be the force of the basic sites. Furthermore, the
desorption peaks area is related to the density of the basic sites on
the catalysts surface [29,51]. Fig. 7 shows the TPD- CO2 profiles of
three synthesized oxides.
It is initially noticed a CO2 desorption peak at around 150 ◦C for
all the materials, possibly related to the basic weak sites. CO2 des-
orption peaks were also detected in the regions of 350 ◦C and 550 ◦C,
which was attributed to medium and strong basic sites, respec-
tively. CaZrO3 presented wide bandwidth of desorption in the
regions of weak and medium strength basic sites around 150 and
350 ◦C, respectively, and the strong sites with very low-intensity
bands at around 550 ◦C. In Ca0.9Li0.6Zr0.9O3 TPD can be observed
two intense and well-defined peaks: the first one centered at
approximately 300 ◦C (medium strength basic sites region) and the
second one, of higher intensity, with maximum at around 550 ◦C,
revealing this oxide to be mainly composed of strong basic sites. On
the other hand, Li2ZrO3 also presented intense desorption bands in
the regions of 350 and 500 ◦C, but more enlarged.
From these results, a study of the homogeneous contribution for
the most active catalyst (CaLiZrO3) was carried out, seeking to prove
the catalytic heterogeneity. Simultaneously and under the same
conditions, the kinetic of the reaction was also evaluated being both
presented in Fig. 8.
The CO2 molecule is slightly acid and, being so it is prone to
adsorption in basic sites, making it possible the quantification of
these sites by relating them with the quantity of CO2 adsorbed.
It is important to highlight that the stronger this interaction the
higher will be the site force and, consequently, the higher the nec-
essary temperature for the desorption of CO2. The quantity of CO2
desorbed in the oxides was determined from the integration of
the TPD-CO2 curves on Fig. 7, that were deconvoluted in three
bands with maximum desorption at the ranges of 120–150 ◦C (weak
bases characteristic), 250–350 ◦C (medium strength sites), and
500–600 ◦C (strong basic sites). According to this, it was possible to
determine the density of basic sites (nb) associated with the force of
the sites. The obtained results are presented in Table 4. These out-
comes indicate that CaZrO3 presented a small nb (4.38 ␮mol m−2),
prevailing the medium force sites (3.10 ␮mol m−2). The Li2ZrO3 cat-
alyst was the one that demonstrated the highest quantity of basic
sites (35.98 ␮mol m−2), being expressive the density of strong basic
Fig. 8. Kinetics of transesterification in the presence of the CaLiZrO3. Homoge-
neous catalysis contribution after catalyst removal from the reaction medium (50 ◦C,
10 wt.% of catalyst, methyl acetate/ethanol ratio = 1:6.
Please cite this article in press as: A.M. Gonc¸ alves, et al., Lithium and calcium based perovskite type oxides for ethylic transesterification,
Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.06.018

G Model
CATTOD-10262; No. of Pages10
ARTICLE IN PRESS
8
A.M. Gonc¸ alves et al. / Catalysis Today xxx (2016) xxx–xxx
100
92
90
80
70
60
50
40
30
20
10
0
90
85
82
67
61
56
49
45
01:06
01:09
01:12
1°
2°
3°
Cycle
Fig. 9. Influence of the methyl acetate/ethanol molar ratio on ethyl acetate conver-
sion (%) for CaLiZrO3.
Fig. 10. Reuse cycles for transesterification reaction in the presence of the catalysts
(50 ◦C, 10 wt.% of catalyst, methyl acetate/ethanol molar ratio = 1:12 and 30 min).
Sheldon et al. claim that a rigorous heterogeneity proof can
be obtained by the separation of the catalysts before the reac-
tion is completed by filtering at the reaction temperature, followed
by monitoring the activity of the filtered product. However, the
authors highlight that the proof of the heterogeneous catalysis
does not indicate the absence of leaching [52]. Thus, it is possi-
ble to infer from the occurrence of heterogeneous catalysis or if
there is the homogeneous contribution of eventually leached ions.
According to Fig. 8 it is possible to observe that after the removal
of the catalyst from the reaction mixture, the conversion remained
at around 30%. This result proves that the reaction does not ‘go
further’ after the removal of the oxide, indicating that there is no
homogeneous contribution of leached species and that the reaction
occurs in a heterogenous phase. The kinetic study demonstrated
that after 30 min of reaction, the conversions have no considerable
alterations, remaining at around 85%, and indicating that within
this optimal time the reaction gets into a chemical equilibrium.
At the light of the previous results, new catalytic tests were car-
ried out with the oxide CaLiZrO3 in a total reaction time of 30 min
and temperature of 50 ◦C varying the ester/alcohol proportion in
three different values (1:6, 1:9 and 1:12). The data in Fig. 9 indi-
cates that the use of the ratio ester/ethanol at 1:12 resulted in
greater yields up to the 3rd reaction cycle (61% of conversion into
ethyl acetate) and greater stability. Nonetheless, the proportion
1:12 of methyl acetate/ethanol was the most advantageous to be
established for the next experiments.
The influence of the ester/alcohol proportion was also verified
in the reaction with soybean oil for obtaining the biodiesel intend-
ing to evaluate if the results would be equivalent. Molar ratios
oil/alcohol equal to 1:6, 1:12, and 1:24 were assessed (the ratio 1:9
was not chosen since the result obtained during the transesterifica-
tion model was similar to that one found at 1:6). It was found that
the ratio oil/alcohol at 1:12 was one more time the most advan-
tageous, being obtained a total content of fatty acid ethyl esters
(FAEEs) of 60% m/m, while with the use of other ratios this content
was between 40 and 50% m/m.
It is important to note that the heterogeneous catalysis during
the transesterification of oil with ethanol is of difficult execution
at lower yields than when performed by using methanol. Besides
that, the temperature conditions and reaction time here adopted
were milder than many other works involving the ethanol trans-
esterification. A review of the literature shows only a few studies
dedicated to evaluate the employment of the studied catalysts in
the biodiesel synthesis. In a study involving the use of several basic
catalysts containing calcium, among them CaZrO3, it was found a
low activity during the first cycle of methanol transesterification,
what is in agreement with the results obtained in present work.
However, in the second operation it was obtained an 88% esters
yield, being it necessary a time of 10 h of reaction. It is supposed that
this activation from the second operation occurs by the formation of
calcium methoxide by the catalyst reaction with methanol, which
are compounds presenting high basicity and reactivity [25]. Kaur
and Ali affirmed that Li2ZrO3 synthesized by wet chemical route
demonstrated high catalytic efficiency in the simultaneous esteri-
fication and transesterification of several raw material with molar
ratio oil/ethanol of 1:15. However, a higher temperature (75 ◦C)
and long reaction times were used along the execution of these
analytical tests [32].
The catalytic stability of CaZrO3, Li2ZrO3 and CaLiZrO3 in the
model transesterification reaction was investigated by the exe-
cution of reuse tests in 6 successive reaction cycles, employing
optimized reaction conditions: ester/alcohol ratio of 1:12, 50 ◦C
and time of 30 min. According to the results presented in Fig. 10,
CaZrO3 has not presented any relevant catalytic activity, corrobo-
rating the results obtained during the initial tests. In fact, this oxide
presented low density of basic sites (see Table 4), what can justify
its low catalytic activity. CaLiZrO3 presented a high activity, with
92% of conversion during the 1 st reaction cycle. After the 1 st cycle,
however, the catalyst suffered some deactivation with the loss of
31% of conversion until the 3rd cycle. From this cycle on, the cat-
alyst maintains an average conversion of 48% up to the 6th cycle,
what gives it a moderate stability considering the relative standard
deviation. The Li2ZrO3 oxide also presented satisfactory catalytic
activity results with 64% of conversation during the 1 st cycle. A
reduction in the average conversion of 17% is observed during the
2nd cycle and after this one, the tendency is to remain unaltered. It
must be highlighted that from the 3rd cycle on, the conversions
in ethyl acetate are practically equal for both oxides containing
lithium. The quantity of basic sites for Li2ZrO3 was greater than for
the oxide CaLiZrO3. However, this latter presented higher catalytic
activity during the initial cycles. It is believed that the interaction
between calcium and lithium atoms in the perovskite structure
can have allowed this higher catalytic activity. Though CaLiZrO3
was the most active catalyst, even presenting surface area just a bit
bigger than the Li2ZrO3. Then, the catalytic activity of these mate-
rials is not justified exclusively by the surface area. Ding et al. [53]
Please cite this article in press as: A.M. Gonc¸ alves, et al., Lithium and calcium based perovskite type oxides for ethylic transesterification,
Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.06.018

G Model
CATTOD-10262; No. of Pages10
ARTICLE IN PRESS
A.M. Gonc¸ alves et al. / Catalysis Today xxx (2016) xxx–xxx
9
Table 5
In the present study, even after finding the leaching, it was not
possible to observe the formation of active species in a homogenous
phase as show the results of Fig. 8. In a recent review, Lee and Wil-
son affirmed that the leaching of alkaline and alkaline earth metals
associated with the reuse of catalysts in heterogeneous reaction
to biodiesel production remains a challenge [9]. Therefore, it is of
fundamental importance to determine the concentration of met-
als leached also in the biodiesel to specify and evaluate whether
the profile of the phenomenon of leaching observed in the model
transesterification reaction is similar in the presence of oil. Thus,
modifications in the synthesis method can be performed to com-
bine to the elevated activity a higher stability of the perovskite-type
catalysts.
Concentrations of Ca, Li and Zr (mg/kg) in the liquid phase after each reaction cycle
(mean S.D., n = 3).
Catalysts
Loading (mg/kg)
Cycle
Ca
Li
Zr
0.64 0.43
0.60 0.49
0.18 0.10
0.04 0.03
29.6 33.2
0.23 0.14
Ca.9Li.6Zr.9O3
1◦
2◦
3◦
4◦
5◦
6◦
109 42.7
59.9 9.55
28.0 5.79
19.3 2.17
40.7 0.10
27.5 0.09
471 139
226 33.6
92.0 0.01
90.2 46.7
103 28.7
36.0 0.21
Li2ZrO3
1◦
2◦
3◦
4◦
5◦
6◦
–
–
–
–
–
–
89.3 26.6
66.4 15.8
59.8 26.1
79.4 4.03
68.1 21.0
53.7 18.9
0.26 0.29
0.16 0.18
0.16 0.09
1.49 1.36
0.29 0.37
0.37 0.34
4. Conclusions
The synthesis of multi-metal perovskite-type oxides by combus-
tion reaction proved to be adequate and advantageous to prepare
catalysts of high purity and crystallinity without calcination step.
XRD and total elemental contents analyses showed the formation
of CaZrO3 and Ca0.9Li0.6Zr0.9O3 perovskite oxides and also Li2ZrO3
of the monoclinic structure.
Method applied for the oxides digestion that combined
HNO3 HCl-H2O2 provided results with good approximation to
nominal contents where low detection limits were observed. So,
this method for digestion of zirconium oxides with hard decom-
position showed to be efficient and a good alternative to those
methods that employ the use of hydrofluoric acid or fusion for the
decomposition of ceramic materials.
observed no direct correlation between the BET surface area of the
ZrO2 catalysts with their catalytic activity. Some studies showed a
decrease in catalytic activity with the Li content increase above of
optimum level loading [31,32,54]. The lithium content determined
in the Li2ZrO3 is almost four times higher than for the CaLiZrO3 per-
ovskite, which can be related to the maximum activity in the first
reaction cycles attributed to this oxide. Concerning the overload-
ing of Li, which causes reduction in catalytic activity for Li-CaO and
Li2ZrO3 [32,54], it was found that the electronic state of Li is crit-
ical to maintaining the high activity in transesterification reaction
with the use of Li-CaO catalysts, considering that the Li 1 s binding
energy is sensitive to electronic state [31].
The elementary composition of the liquid phase of each reac-
tion cycle was evaluated to verify the existence of leached species
of the catalyst for the reaction mixture in a way to complement
the results of homogeneous and stability contribution. According
to Sheldon et al., it is possible to consider three hypotheses in the
context of leaching: (i) it occurs but not associated with the for-
mation of active homogeneous species; (ii) the metals are leached
resulting in some homogeneous contribution, and (iii) the metals
are not leached and the catalysis is truly heterogeneous [52]. The
results obtained by ICP OES for the catalysts containing lithium are
presented in Table 5. As the CaZrO3 resulted in a non-significant
catalytic activity for the transesterification, the data related to this
catalyst was not explored.
Table 5 indicates low quantity of leached species about the
total levels of Zr, Ca, and Li in the catalysts, being the maximum
obtained values equal to approximately 60, 22, and 9 wt.%, respec-
tively. Nevertheless, it is highlighted that these values deserve to be
investigated since it is possible that there has occurred ions concen-
tration by uncontrolled evaporation of the filtered material before
the analysis, since this was not monitorated immediately after the
reaction and that the mixture of solvents is highly volatile. Among
the investigated catalysts, it is noticed that one higher leaching
of the metals occurred for the CaLiZrO3, once higher calcium and
lithium levels were found in the reaction mixture after the two first
reaction cycles. According to the stability data (Fig. 9), the catalytic
activity decreases after two initial cycles of around 30%, possibly
due to severe metals loss to the solution. On the other hand, after
the 3rd cycle the quantity of leached species is reduced, what con-
tributes to a certain stability up to the sixth cycle. Table 5 also shows
that the zirconium leaching was very low in all the materials, sug-
gesting that the catalytic activity is not directly connected to this
element. Low LOQs were obtained in mg kg−1 for Ca, Li, and Zr,
equal to 0.851, 0.046, and 1.07, respectively. However, the zirco-
nium concentration was lower than the LOQ, but are close to the
LOD (0.321 mg kg−1), showing an approximate value of the leached
content, which is insignificant.
The catalysts containing Li and Ca presented higher activity and
stability in the transesterification reaction between methyl acetate
and ethanol (average conversion of 89% in 30 min of reaction for
the first two cycles), having presented better catalytic performance
when compared to those oxides containing only one of the given
metals. Reaction tests revealed that the CaZrO3 catalyst presents
negligible activity in the model transesterification while Li2ZrO3
presented a conversion of 64% in the first cycle, being the average
conversion kept at around 50% up to the 6th cycle. Thus, it is possible
to conclude that the incorporation of lithium to the zirconate pro-
vided an efficient and promising catalysts for transesterification.
This performance can be related to the alteration in the texture
and basic characteristics caused by insertion of Li. The presence
of lithium in the catalyst causes physical alterations, such as the
reduction of the BET surface area and the increase in the crystallite
size, in the number and force of the sites.
The leaching of metals was detected at the different catalysts
being more pronounced for Ca0.9Li0.6Zr0.9O3. Even so, the contribu-
tion of homogeneous catalysis was not observed from the leached
ions, indicating that the reaction occurs in a heterogeneous phase.
The gradual deactivation seen in the perovskite catalysts after
successive reuse cycles possibly occur due to the leaching of the
metallic ions, a recurrent phenomenon in the basic heterogeneous
catalysis. Reuse tests showed that the yield decrease is higher until
the 3rd cycle with the most active catalyst, probably because of the
higher quantity of leached species during the first batches. How-
ever, it is noteworthy that the catalysts containing lithium remain
active for at least 6 reaction cycles, indicating moderate catalytic
stability. Besides that, the metals leaching observed from those cat-
alysts is a characteristic of heterogeneous basic catalysts still being
a challenge to be bypassed.
The Ca0.9Li0.6Zr0.9O3 catalyst presented an excellent perfor-
mance in the transesterification of the soybean oil using mild
reaction conditions (50 ◦C and 30 min), seeming it to be very
efficient and promising for biodiesel production by ethylic trans-
esterification. Finally, the Ca0.9Li0.6Zr0.9O3 oxide being a versatile
Please cite this article in press as: A.M. Gonc¸ alves, et al., Lithium and calcium based perovskite type oxides for ethylic transesterification,
Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.06.018

G Model
CATTOD-10262; No. of Pages10
ARTICLE IN PRESS
10
A.M. Gonc¸ alves et al. / Catalysis Today xxx (2016) xxx–xxx
material enables the execution of further studies involving substi-
tutions by other elements in its structure, aiming the increase in
activity and especially in catalytic stability.
[24] L.F. Man, W.T. Wong, K.F. Yung, J. Clust. Sci. 23 (2012) 873–884, http://dx.doi.
org/10.1007/s10876-012-0475-x.
[25] A. Kawashima, K. Matsubara, K. Honda, Bioresour. Technol. 99 (2008)
3439–3443, http://dx.doi.org/10.1016/j.biortech.2007.08.009.
[26] S. Xia, X. Guo, D. Mao, Z. Shi, G. Wu, G. Lu, RSC Adv. 4 (2014) 51688–51695,
http://dx.doi.org/10.1039/C4RA11362D.
Acknowledgements
ˇ
ˇ
[27] Z. Kesic´, I. Lukic´, M. Zdujic´, C. Jovalekic´, V. Veljkovic´, D. Skala, Fuel Process.
Technol. (2016) 162–168, http://dx.doi.org/10.1016/j.fuproc.2015.11.018.
[28] J. Rynkowski, A. Kaz´mierczak, A. Praz´mowska-Wilanowska, T. Paryjczak,
React. Kinet. Catal. Lett. 58 (1996) 169–175, http://dx.doi.org/10.1007/
B F020 71121.
The authors are grateful to the Conselho Nacional de Desen-
volvimento Científico
e Tecnológico (CNPq; process number
500251/2013-4) and to the Instituto Nacional de Ciências e Tec-
nologias Analíticas Avanc¸ adas (INCTAA), Brazil, for the support of
this research.
[29] C.S. Castro, D. Cardoso, P.A.P. Nascente, J.M. Assaf, Catal. Lett. 141 (2011)
1316–1323, http://dx.doi.org/10.1007/s10562-011-0650-y.
[30] V.K. Díez, C.R. Apesteguía, J.I. Di Cosimo, J. Catal. 240 (2006) 235–244, http://
dx.doi.org/10.1016/j.jcat.2006.04.003.
[31] R.S. Watkins, A.F. Lee, K. Wilson, Green Chem. 6 (2004) 335–340, http://dx.
doi.org/10.1039/b404883k.
[32] N. Kaur, A. Ali, Appl. Catal. A Gen. 489 (2015) 193–202, http://dx.doi.org/10.
1016/j.apcata.2014.10.013.
[33] B.D. Cullity, Elements of x-ray Diffraction, Addison-Wesley Pub Comp. Inc.,
EUA, 1978.
[34] J. McKittrick, L.E. Shea, C.F. Bacalski, E.J. Bosze, Displays 19 (1999) 169–172,
http://dx.doi.org/10.1016/S0141-9382(98)00046-8.
[35] P.M. Crnkovic, W.L. Polito, C.G. da S. Filho, F.E. Milioni, J.D. Pagliuso, Quim.
Nova. 27 (2004) 58–61, http://dx.doi.org/10.1590/S0100-
40422004000100012.
[36] R.K. Kotnala, J. Shah, B. Singh, H. Kishan, S. Singh, S.K. Dhawan, A. Sengupta,
Sens. Actuat. B Chem 129 (2008) 909–914, http://dx.doi.org/10.1016/j.snb.
2007.10.002.
[37] C.S. Prasanth, H. Padma Kumar, R. Pazhani, S. Solomon, J.K. Thomas, J. Alloys
Compd. 464 (2008) 306–309, http://dx.doi.org/10.1016/j.jallcom.2007.09.098.
[38] L. Loembe, Z. Fu, W. Wang, H. Wang, IJRET 4 (2015) 1–7, http://dx.doi.org/10.
15623/ijret.2015.0406001.
[39] V. Singh, S. Watanabe, T.K.G. Rao, K. Al-shamery, M. Haase, Y. Jho, J. Lumin.
132 (2012) 2036–2042, http://dx.doi.org/10.1016/j.jlumin.2012.03.027.
[40] M. Villanueva-Ibán˜ez, M.A. Hernández-Pérez, P.N. Rivera Arzola, M.A.
Flores-González, Matéria 18 (2013) 1501–1509, http://dx.doi.org/10.1590/
S1517-70762013000400011.
[41] B. Liu, X. Lin, L. Zhun, X. Wang, D. Xu, Ceram. Int. 40 (2014), http://dx.doi.org/
10.1016/j.ceramint.2014.03.025.
[42] K.-S. Hwang, Y.-H. Lee, S. Hwangbo, Mater. Sci-Pol. 25 (2007), http://dx.doi.
org/10.1016/05 2007 165lee.
[43] G. Tilekar, K. Shinde, K. Kale, R. Raskar, A. Gaikwad, Front. Chem. Sci. Eng. 5
(2011) 477–491, http://dx.doi.org/10.1007/s11705-011-1107-y.
[44] R. Xiong, J. Ida, Y.S. Lin, Chem. Eng. Sci. 58 (2003) 4377–4385, http://dx.doi.
org/10.1016/S0009-2509(03)00319-1.
[45] M. Niemelä, S. Pitkäaho, S. Ojala, R.L. Keiski, P. Perämäki, Microchem. J. 101
(2012) 75–79, http://dx.doi.org/10.1016/j.microc.2011.11.001.
[46] N.A. Merino, B.P. Barbero, P. Ruiz, L.E. Cadús, J. Catal. 240 (2006) 245–257,
http://dx.doi.org/10.1016/j.jcat.2006.03.020.
[47] R.I. Smith, A.R. West, J. Solid State Chem. 108 (1994) 29–36, http://dx.doi.org/
10.1006/jssc.1994.1005.
[48] C.S. Castro, C. Ferreti, J.I. Di Cosimo, J.M. Assaf, Fuel 103 (2013) 632–638,
http://dx.doi.org/10.1016/j.fuel.2012.06.072.
[49] J.I. Gutiérrez-Ortiz, B. De Rivas, R. López-Fonseca, J.R. González-Velasco, Appl.
Catal. A Gen. 269 (2004) 147–155, http://dx.doi.org/10.1016/j.apcata.2004.04.
014.
References
[1] L. Meher, D. Vidyasagar, S. Naik, Renew. Sustain. Energy Rev. 10 (2006)
248–268, http://dx.doi.org/10.1016/j.rser.2004.09.002.
[2] E.F. Aransiola, T.V. Ojumu, O.O. Oyekola, T.F. Madzimbamuto, D.I.O.
Ikhu-Omoregbe, Biomass Bioenerg. 61 (2014) 276–297, http://dx.doi.org/10.
1016/j.biombioe.2013.11.014.
[3] A. Murugesan, C. Umarani, T.R. Chinnusamy, M. Krishnan, R. Subramanian, N.
Neduzchezhain, Renew. Sustain. Energy Rev. 13 (2009) 825–834, http://dx.
doi.org/10.1016/j.rser.2008.02.003.
[4] S.K. Hoekman, A. Broch, C. Robbins, E. Ceniceros, M. Natarajan, Renew. Sustain.
Energy Rev. 16 (2012) 143–169, http://dx.doi.org/10.1016/j.rser.2011.07.143.
[5] C. Brunschwig, W. Moussavou, J. Blin, Prog. Energy Combust. Sci. 38 (2012)
283–301, http://dx.doi.org/10.1016/j.pecs.2011.11.001.
[6] B. Moser, Biofuels (2011) 229–266, http://dx.doi.org/10.1007/s11627-009-
9204-z.
[7] L. Lin, Z. Cunshan, S. Vittayapadung, S. Xiangqian, D. Mingdong, Appl. Energy
88 (2011) 1020–1031, http://dx.doi.org/10.1016/j.apenergy.2010.09.029.
[8] A. Sivasamy, K.Y. Cheah, P. Fornasiero, F. Kemausuor, S. Zinoviev, S. Miertus,
ChemSusChem 2 (2009) 278–300, http://dx.doi.org/10.1002/cssc.200800253.
[9] A.F. Lee, K. Wilson, Catal. Today 242 (2015) 3–18, http://dx.doi.org/10.1016/j.
cattod.2014.03.072.
[10] Z. Helwani, M.R. Othman, N. Aziz, J. Kim, W.J.N. Fernando, Appl. Catal. A-Gen.
363 (2009) 1–10, http://dx.doi.org/10.1016/j.apcata.2009.05.021.
[11] S. Yan, C. DiMaggio, S. Mohan, M. Kim, S.O. Salley, K.Y.S. Ng, Top. Catal. 53
(2010) 721–736, http://dx.doi.org/10.1007/s11244-010-9460-5.
[12] D.-W. Lee, Y.-M. Park, K.-Y. Lee, Catal Surv. Asia 13 (2009) 63–77, http://dx.
doi.org/10.1007/s10563-009-9068-6.
[13] M.R. Goldwasser, M.E. Rivas, E. Pietri, M.J. Pérez-Zurita, M.L. Cubeiro, A.
Grivobal-Constant, G. Leclercq, J. Mol. Catal. A Chem. 228 (2005) 325–331,
http://dx.doi.org/10.1016/j.molcata.2004.09.030.
[14] R. Iano&#x015F, P. Barvinschi, J. Solid State Chem. 183 (2010) 491–496,
http://dx.doi.org/10.1016/j.jssc.2009.12.015.
[15] K. Boobalan, A. Varun, R. Vijayaraghavan, K. Chidambaram, U.K. Mudali,
Ceram. Int. 40 (2014) 5781–5786, http://dx.doi.org/10.1016/j.ceramint.2013.
11.017.
[16] D. Cruz, H. Pfeiffer, S. Bulbulian, Solid State Sci. 8 (2006) 470–475, http://dx.
doi.org/10.1016/j.solidstatesciences.2006.01.003.
[17] R. Ianos, C. Pacurariu, I. Lazau, S. Ianosev, Z. Ecsedi, R. Lazau, P. Barvinschi, J.
Therm. Anal. Calorim. 94 (2008) 343–348, http://dx.doi.org/10.1007/s10973-
008-9104-1.
[50] L. Bedel, A. Roger, C. Estournes, A. Kiennemann, Catal. Today. 85 (2003)
207–218, http://dx.doi.org/10.1016/S0920-5861(03)00388-2.
[51] J. Zhu, X. Peng, L. Yao, D. Tong, C. Hu, Catal. Sci. Technol. 2 (2012) 529, http://
dx.doi.org/10.1039/c1cy00333j.
[52] R.A. Sheldon, M. Wallau, I.W.C.E. Arends, U. Schuchardt, Acc. Chem. Res. 31
(1998) 485–493, http://dx.doi.org/10.1021/ar9700163.
[18] C. Gargori, S. Cerro, R. Galindo, A. García, M. Llusar, G. Monrós, Ceram. Int. 38
(2012) 4453–4460, http://dx.doi.org/10.1016/j.ceramint.2012.02.019.
[19] R.A. Davies, M.S. Islam, J.D. Gale, Solid State Ionics. 126 (1999) 323–335,
http://dx.doi.org/10.1016/S0167-2738(99)00244-1.
[20] S. Marinel, M. Pollet, G. Desgardin, J. Mater. Sci. − Mater. El. 13 (2002)
149–155, http://dx.doi.org/10.1023/A:1014381300373.
[53] Y. Ding, H. Sun, J. Duan, P. Chen, H. Lou, X. Zheng, Catal. Commun. 12 (2011)
606–610, http://dx.doi.org/10.1016/j.catcom.2010.12.019.
[54] J. Boro, L.J. Konwar, D. Deka, Fuel Process. Technol. (2014) 72–78, http://dx.
doi.org/10.1016/j.fuproc.2014.01.022.
[21] Z. Li, W.E. Lee, S. Zhang, J. Am. Ceram. Soc. 90 (2007) 364–368, http://dx.doi.
org/10.1111/j. 1551 -2916.2006.01383. x.
[22] H. Pfeiffer, P. Bosch, S. Bulbulian, Mater. Chem. Phys. 78 (2002) 558–561,
http://dx.doi.org/10.1016/S0254-0584(02)00382-6.
[23] L. Montanaro, J.P. Lecompte, J. Mater. Sci. 27 (1992) 3763–3769, http://dx.doi.
org/10.1007/B F005 4545 3.
Please cite this article in press as: A.M. Gonc¸ alves, et al., Lithium and calcium based perovskite type oxides for ethylic transesterification,
Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.06.018