Solketal synthesis from ketalization of glycerol with acetone: A kinetic study over a sulfated zirconia catalyst

Article abstract of DOI:10.1016/j.cattod.2020.10.005

In this work, a series of acid catalysts were synthetized from a commercial zirconium oxide sulfated with a 0.5 M H₂SO₄ solution by wet impregnation. The characterization results show a correlation between the calcination temperature and the acid sites generated on the materials. Among the catalysts prepared, the sulfated zirconia calcined in air at 400 °C (Zr-S-400), with a molar ratio S/Zr = 0.23 was the most active one due to its larger acid density and greater acid strength caused by the generation of new Br?nsted sites. The Zr-S-400 catalyst exhibited an initial reaction rate of 0.0497 mol.min^?1. g^?1, and achieved a glycerol conversion of 80 % in 1 h of reaction at 40 °C (glycerol:acetone molar ratio = 1:6). The Zr-S-400 material remained stable after four catalytic cycles, demonstrating the stability of the superficial sulfate species (S/Zr ～ 0.2). In addition, the thermodynamics and kinetics of the reaction were evaluated, as well as the influence of some operating conditions such as the molar ratio of reactants and the water content in the reaction mixture. The following standard molar reaction properties were obtained: ΔHo = -11.6 ± 1.1 kJ.mol^?1 and ΔGo = 4.0 ± 0.1 kJ.mol^?1. Taking into account that the adsorption of water on this catalyst did not affect the number of acid sites available, a simple pseudo-homogeneous kinetic expression was developed and successfully adjusted to the experimental data in the range under study. Based on this model, the estimated activation energy of the reaction was 88.1 ± 8.9 kJ.mol^?1.

Full text of DOI:10.1016/j.cattod.2020.10.005

Journal Pre-proof
Solketal synthesis from ketalization of glycerol with acetone: A kinetic
study over a sulfated zirconia catalyst
´
Julian A. Vannucci, Nora N. Nichio, Francisco Pompeo
PII:
S0920-5861(20)30678-7
DOI:
https://doi.org/10.1016/j.cattod.2020.10.005
CATTOD 13158
Reference:
To appear in:
Catalysis Today
Received Date:
Revised Date:
Accepted Date:
18 May 2020
19 August 2020
1 October 2020
Please cite this article as: Vannucci JA, Nichio NN, Pompeo F, Solketal synthesis from
ketalization of glycerol with acetone: A kinetic study over a sulfated zirconia catalyst, Catalysis
Today (2020), doi: https://doi.org/10.1016/j.cattod.2020.10.005
This is a PDF file of an article that has undergone enhancements after acceptance, such as
the addition of a cover page and metadata, and formatting for readability, but it is not yet the
definitive version of record. This version will undergo additional copyediting, typesetting and
review before it is published in its final form, but we are providing this version to give early
visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal
pertain.
© 2020 Published by Elsevier.

Solketal synthesis from ketalization of glycerol with acetone: A kinetic study over a
sulfated zirconia catalyst
Julián A. Vannuccia,b, Nora N. Nichioa,b, Francisco Pompeoa,b,*
a Centro de Investigación y Desarrollo en Ciencias Aplicadas (CINDECA), Facultad de Ciencias Exactas,
Universidad Nacional de La Plata (UNLP) – CONICET, Calle 47, 257, CP 1900, La Plata, Argentina.
b
Facultad de Ingeniería, Universidad Nacional de La Plata (UNLP), Calle 1 esq. 47, CP 1900, La Plata,
Argentina.
*fpompeo@quimica.unlp.edu.ar (Francisco Pompeo)
Graphical abstract
Highlights
 SO4/ZrO2 were prepared by a simple wet impregnation technique.
 A simple pseudo-homogeneous kinetic expression was developed and validated.
 The activation energy of the reaction was estimated as 88.1 ± 8.9 kJ/mol.

Water largely affects the reaction rate and the conversion at equilibrium state.
Abstract
In this work, a series of acid catalysts were synthetized from a commercial zirconium oxide
sulfated with a 0.5 M H2SO4 solution by wet impregnation. The characterization results
show a correlation between the calcination temperature and the acid sites generated on the
1

materials. Among the catalysts prepared, the sulfated zirconia calcined in air at 400 °C (Zr-
S-400), with a molar ratio S/Zr = 0.23 was the most active one due to its larger acid density
and greater acid strength caused by the generation of new Brönsted sites. The Zr-S-400
catalyst exhibited an initial reaction rate of 0.0497 mol.min-1.g-1, and achieved a glycerol
conversion of 80% in 1 hour of reaction at 40°C (glycerol:acetone molar ratio =1:6). The
Zr-S-400 material remained stable after four catalytic cycles, demonstrating the stability of
the superficial sulfate species (S/Zr ~ 0.2). In addition, the thermodynamics and kinetics of
the reaction were evaluated, as well as the influence of some operating conditions such as
the molar ratio of reactants and the water content in the reaction mixture. The following
standard molar reaction properties were obtained: ΔHº = -11.6 ± 1.1 kJ.mol-1 and ΔGº = 4.0
± 0.1 kJ.mol-1. Taking into account that the adsorption of water on this catalyst did not
affect the number of acid sites available, a simple pseudo-homogeneous kinetic expression
was developed and successfully adjusted to the experimental data in the range under study.
Based on this model, the estimated activation energy of the reaction was 88.1 ± 8.9 kJ.mol-
1
.
Keywords: Solketal, Sulfated zirconia, Batch reactor, Kinetic model.
1. Introduction
Glycerol is a promising biomass platform molecule having several applications in many
industries. Nowadays, the development of new technological ways to add value to glycerol
has become a research hotspot. Many studies have focused on the valorization of glycerol
through catalytic processes such as ketalization, dehydration, oxidation, and reforming. In
particular, the condensation reaction of glycerol with carbonyl compounds to produce
oxygenated compounds has received great attention [1]. Among the different glycerol
ketals, solketal is highly valuable for its potential applications as a green solvent, plasticizer
in the polymer industry [2], and solubilizing and suspending agent in biodegradable
systems for the controlled release of medicinally active substances [3]. In addition, it can be
used as a fuel additive to increase the octane number and reduce gum formation [4].
Solketal is normally synthesized by the ketalization of glycerol with acetone (Scheme 1).
2

O
OH
O
O
H2O
+
OH
+
HO
Acetone
Glycerol
HO
Solketal
Scheme 1: Ketalization of glycerol with acetone
The miscibility of both reactants, glycerol and acetone, is poor. However, as the reaction
progresses, the produced solketal acts as a cosolvent, improving the solubility of glycerol in
the acetone phase [5]. Thus, an excess of acetone is often used as a solubility enhancer [5]
and to shift the chemical equilibrium toward the formation of solketal [2]. Despite these
problems, several reports have shown that the reaction could be successfully carried out
with heterogeneous catalysts such as ion exchange resins [6], heteropolyacids [7], acid
clays [8], mesostructured silicas [9-11], and zeolites [12-14]. In comparison to the
aforesaid catalysts, promoted metal oxides offer several advantages, since they are stable,
regenerable, and active. Li et al. synthesized layered crystalline α-zirconium phosphates
and studied the effect of the calcination temperature over the acid properties of the material.
The results showed that materials with a relation P/Zr ~ 2 possess higher surface density of
acid sites and stability when calcined at temperatures below 300°C. The conversion of
glycerol decreased from 86 to 45% with the increasing calcination temperature from 200 to
600ºC, which was attributed to the decomposition of the active species Zr(HPO4)2∙H2O to
ZrP2O7 [15].
Miao et al. synthesized a series of ordered mesoporous titanium phosphate (M-TiPO)
materials with controllable composition by a simple one-pot method. With the
incorporation of highly dispersed P species in the structure of M-TiPO, the bonds in the
material transferred from Ti-O-Ti to Ti-O-P bonds, thus improving the acid properties of
the material. The M-TiPO catalyst with a P/Ti ratio = 0.75 exhibited the best catalytic
effect, achieving a glycerol conversion of 91% and a selectivity to solketal of 94% [16].
Recently, Li et al. prepared a series of zirconium organophosphates with different quantities
of fenil groups following a hydrothermal method. The materials showed a hydrophobic
3

character which was enhanced with an increasing amount of phenyl groups. The best
activity was attributed to both abundant acidity and optimal hydrophilia/hydrophobicity
balance [17].
In reactions requiring acid materials, sulfate species represent one of the most widely used
promoters of metal oxides [18,19]. The sulfated zirconia catalyst has been employed in
many industrially important vapor-phase reactions such as dehydration of ethanol [20],
alkylation of isobutane with 2-butene [21], and isomerization of hydrocarbons [22-24].
Furthermore, it has been found to be very active in liquid-phase reactions, such as Friedel-
Crafts monoalkylation [18], synthesis of aromatics gem-dihalides [25], stereocontrolled
glycosidation [26], etc.
The application of sulfated zirconia to the ketalization of glycerol with acetone has been
previously explored [27-29]. Reddy et al. compared diverse promoters of zirconia, such as
W, Mo, and sulfate ions, concluding that the sulfated promoted zirconia showed the highest
catalytic activity due to its higher surface area and number of acidic sites. Although their
study demonstrated the efficacy of sulfated zirconia in glycerol ketalization, the stability of
the catalyst and the kinetics of the reaction were not evaluated [27].
Nanda et al. compared a zirconium sulfate catalyst in the synthesis of solketal in a
continuous flow reactor with different commercial materials. According to the yield to
solketal and glycerol conversion, the activity of these catalysts followed the sequence
Amberlyst Wet ~ Zeolite ~ Amberlyst Dry > Zirconium Sulfate > Montmorillonite >
Polymax. The stability of these catalysts was analyzed and a slight decrease in the activity
was found, likely due to the loss of its acidity after a long time on stream [29].
The objectives of this work were to develop an efficient catalyst for the synthesis of
solketal and determine the thermodynamic and kinetic parameters, which are essential tools
for understanding the reactive system, designing reactors, and scaling up the process. For
this purpose, commercial ZrO2 was promoted with H2SO4 following a wet impregnation
technique and employed as a catalyst to carry out the ketalization of glycerol with acetone.
Once the catalyst was proven efficient and stable, several experimental data were collected
in order to determine a kinetic expression that describes the reaction behavior accurately.
Characterization techniques, such as XRD, N2 adsorption-desorption, potentiometric
4

titration, XPS, and Pyridine FTIR, were employed to relate the catalytic activity results to
the presence of acid sites.
2. Experimental
2.1. Materials
Glycerol (99.5%), absolute ethanol (99.5%) and sulfuric acid (98%) were purchased from
Cicarelli. Acetone (99.5%), n-propanol (99.3%), and acetonitrile (99.8%) were purchased
from Anedra. Solketal (98.2%) was purchased from TCI Chemicals, and powder zirconium
oxide was purchased from MEL Chemicals.
2.2. Catalyst preparation
A commercial zirconium oxide powder, namely Zr, was subjected to a thermal treatment at
600 °C for 5 hours, resulting in sample Zr600. Both materials, Zr and Zr600, were promoted
with a proper amount of 0.5M H2SO4 solution, following a wet impregnation technique.
Excess water was evaporated on a water-bath at 60 °C assisted by ultrasound, oven-dried,
and calcined in air at 400 (Zr-S-400, Zr600-S-400) and 600 °C (Zr-S-600) for 5 hours.
2.3. Catalyst characterization
The XRD patterns in the 2θ angular range from 10 to 70° were obtained with a Philips 3710
X’Pert using CuKα radiation, with a 0.04°/min step. The identification of the XRD phases
present in the sample was aided by JCPDS data files.
Textural properties were determined by nitrogen adsorption-desorption isotherms at the
temperature of liquid nitrogen (-196°C) in a Micrometrics ASAP 2020 instrument. Before
adsorption, the samples were evacuated by heating at 100 °C in vacuum, with a pressure
lower than 4 Pa for 12 hours. The specific surface area was calculated according to the
Brunauer-Emmett-Teller (BET) equation, in the relative pressure range 0.05-0.35 [30]. The
pore size distribution was obtained by the Barret-Joyner-Halenda (BJH) method, using the
adsorption branch and assuming slit-shape pore geometry [31].
The X-ray photoelectron spectroscopy (XPS) measurements were performed using a non-
monochromatic Al Kα source (XR50, Specs GmbH) and a hemispherical electron energy
5

analyzer (PHOIBOS 100, Specs GmbH). O 1s, Zr 3d, S 2p, and C 1s lines were monitored.
C 1s at 284.6 eV was used as a charging reference. Spectra were analyzed with CasaXPS
software.
The amount and strength of acid sites were estimated by potentiometric titration. A known
mass (0.05 g) of the solids was suspended in acetonitrile and kept under stirring for 3 hours.
The suspension was titrated with a solution of n-butylamine in acetonitrile (0.05 M) at 0.05
ml min-1. The electrode potential variation was obtained on a digital pH meter (Metrohm
794 Basic Titrino apparatus with a double junction electrode).
Pyridine FTIR was used to characterize the nature of the acid sites. Spectra were recorded
with a Thermo Nicolet iS10. To create a clean surface for the analysis, the catalysts were
heated to 400 °C under vacuum conditions and maintained for 1 hour. Thereafter, the
catalyst was cooled to room temperature to collect the background spectra from the clean
surface. After background collection, the adsorption was performed on the cell at 25 °C.
The excess pyridine was desorbed with a vacuum from room temperature to 400 °C with
100 °C steps. Brönsted and Lewis acid quantification was performed from 1545 cm-1 and
1450-1460 cm-1 bands, respectively, using the literature data on the integrated molar
extinction coefficients [32].
2.4. Catalytic activity measurements
The ketalization of glycerol was carried out in a 200 ml Büchi glass batch reactor equipped
with a manometer, magnetic stirrer, thermostatic bath, gas inlet, and release valves. Before
the catalytic evaluation, experiments were performed to verify the negligible contribution
of the reaction in the absence of a catalyst and in the absence of external and internal
diffusion limitations.
In a typical experiment, acetone, glycerol, and the catalyst are loaded in a reactor. Then, the
reactor is pressurized to 0.2 MPa with N2 and heated to the desired temperature. As
previously mentioned, the use of an excess of acetone improves the amount of solubilized
glycerol. Molar ratios of glycerol to acetone in the range of 1:2 to 1:20 have been reported
in the literature [33]. However, molar ratios greater than 1:10 do not generate significant
impact on catalytic performance [34-36]. Concerning the catalyst, mass values between
0.5% and 5% wt. related to glycerol are commonly used [6, 8, 27, 37, 38]. In our
6

experiments, we employed a catalyst amount in the range of 0.3% to 2.5% wt. of the total
mass of glycerol and molar ratios glycerol:acetone in the range of 1:4 to 1:8.
Once the experiment was over, the reactor was cooled to 20 °C and the catalyst was
separated by centrifugation and filtration. The reactants and products were analyzed by gas
chromatography with a Shimadzu GCMS-QP505A, equipped with a PE-Elite-Wax
capillary column, and an FID detector, using n-propanol as an external standard.
The conversion was determined with the following equation:
(푖푛푖푡푖푎푙 푔푙푦푐푒푟표푙 푚표푙푒푠 − 푓푖푛푎푙 푔푙푦푐푒푟표푙 푚표푙푒푠)
푋% =
푖푛푖푡푖푎푙 푔푙푦푐푒푟표푙 푚표푙푒푠
and the selectivity to solketal with
푠표푙푘푒푡푎푙 푚표푙푒푠
푆% =
(푖푛푖푡푖푎푙 푔푙푦푐푒푟표푙 푚표푙푒푠 − 푓푖푛푎푙 푔푙푦푐푒푟표푙 푚표푙푒푠)
2.5. Thermodynamic and Kinetic studies
The equilibrium thermodynamic constant, Keq, can be determined by the following equation
푎푆표푙푘. 푎푊
푎퐺푙푦. 푎퐴푐
(1)
(2)
퐾푒푞
=
푎푖 = 훾푖. 푥푖
The activity coefficients (훾푖) at different temperatures and compositions were calculated by
UNIFAC [39] with the groups listed in Table S1 (Supplementary Information).
The reaction mechanism proposed in the literature [40-42] is represented in Scheme 2. In
the first step, glycerol and acetone interact forming a hemiketal (3-(2-hydroxypropan-2-
yloxy)propane-1,2-diol) [40]. The second step differs depending on whether the reaction is
conducted under Brönsted acid sites or Lewis acid sites. In the presence of Brönsted acid
sites, a carbenium ion is stabilized and activated for a nucleophilic attack from one of the
alcoholic groups of glycerol, leading to the formation of solketal or the six-membered ring
ketal (2,2-dimethyl-1,3-dioxan-5-ol) [41]. The presence of Lewis acid sites coordinates and
activates the tertiary alcohol of the hemiketal. Then, an intramolecular reaction with one of
the alcoholic groups leads to the formation of solketal and the isomer [42].
7

Although both solketal and 2,2-dimethyl-1,3-dioxan-5-ol isomer are produced in the
reaction, selectivities toward solketal higher than 80% have been reported by several
authors. This could be related to the presence of a methyl group in the axial position of the
six-membered ring of 2,2-dimethyl-1,3-dioxan-5-ol, making this molecule less
thermodynamically stable than the solketal molecule [43].
H C
3
+
C
H C
3
H2O
+
O
H+
+
-
H
H3C
H3C
HO
OH
O
H
H C
3
CH
O
(3)
3
H C
3
CH
O
(2)
3
O
OH
O
O
+
O
OH
HO
+
(1)
HO
OH
(3)
Acetone (Ac)
A+
Glycerol (Gly)
OH
2,2-dimethyl-1,3-dioxan-5-ol
HO
Solketal (Slk)
A+
O–
H3C
H3C
(2)
O
H2O
+
CH+
OH
Scheme 2: Reaction mechanism proposed for the ketalization of glycerol with acetone over acid
catalysts (adapted from Calvino-Casilda et al. [40]).
The reaction rate of the glycerol ketalization could be expressed in the form of a Langmuir-
Hinshelwood-Hougen-Watson (LHHW) model [44,45]. The mechanism is based on the
adsorption of both reactants (glycerol and acetone) on the catalyst surface, followed by
three consecutive reactions (Scheme 2) and the desorption of the products (water and
solketal).
The kinetic parameters were estimated by implementing the Orthogonal Distance
Regression algorithm for nonlinear curve fitting using OriginLab software. The differences
between the predicted values for the variation of the glycerol concentration over time and
the experimental data were minimized using the chi-squared criterion.
The Mears criterion and Weisz-Prater criterion were used to evaluate external and internal
diffusion limitations. For this purpose, the diffusivity coefficients for each component, the
diffusivity coefficient of the multicomponent system, and the mass transfer coefficient were
estimated with the Scheibel correlation [46], Perkins and Geankoplis correlation [47] and
the Hixson and Baum correlation [48], respectively.
8

3. Results and discussion
3.1. Catalytic Activity
Since the ketalization of glycerol with acetone requires the presence of strong acid sites,
this reaction represents an additional characterization technique to evidence these active
sites. The non-promoted materials, samples Zr and Zr600, were not active in this reaction,
which shows their lack of strong acid sites. By contrast, all sulfated samples were active in
the reaction. Table 1 presents the initial reaction rate and selectivities for each catalyst.
(determined in the first 10 minutes of the reaction). In all reactions, only solketal and 2,2-
dimethyl-1,3-dioxan-5-ol were identified as reaction products. Some authors have reported
the presence of subproducts such as 4-hydroxy-4-methyl-2-pentanone [49], 1,2,3-
propanetriol monoacetate, 2-(1-hydroxypropan-2-yloxy)-propane-1,2-diol and 3-(1-
hydroxypropan-2-yloxy)-propane-1,2-diol [50 - 52], in the presence of a catalyst with redox
properties such as Nb [49], Co, Ni, or Pt [50, 51]. Therefore, the absence of these properties
in our catalyst improves the selectivity toward solketal. These results show that the pre-
calcined and sulfated Zr600-S-400 sample presents two-thirds of the catalytic activity in
comparison with sample Zr-S-400. In addition, sample Zr-S-600 was the least active. The
characterization of the results presented in Section 3.2 shows the relationship between this
catalytic behavior and the acidic and textural properties of the sample.
Table 1: Initial reaction rate. Experimental conditions: 10 min reaction, 40 °C; 0.2 MPa
N2, glycerol: acetone molar ratio = 1:6, 0.6 wt.% (catalyst:glycerol).
Initial reaction rate
(mol.min-1.g-1)
0.0497
SSolketal (%) S2,2-dimethyl-1,3-dioxan-5-ol (%)
Sample
Zr-S-400
Zr-S-600
86
88
81
14
12
19
0.0217
Zr600-S-400 0.0361
9

Sample Zr-S-400 was the most active and it was, therefore, selected for the kinetic study.
The study of the stability of the Zr-S-400 catalyst represents an important factor in every
kinetic determination. For this purpose, four activity cycles of the same catalyst batch were
performed. After each run, the reaction mixture was separated from the catalyst by
centrifugation and filtration, and the catalyst was reintroduced into the reactor (without any
treatment). After these four runs (Figure 1), the catalyst showed a 16% decrease over its
initial activity. Since the separation process could lead to small losses of fine catalyst
powders, the result obtained in this study is remarkable, evidencing the stability of the
material. From these results, a kinetic study of the ketalization of glycerol with an acetone
reaction was performed using the Zr-S-400 catalyst.
70
60
50
40
30
20
10
0
1
2
3
4
Run
Figure 1: Stability of Zr-S-400. Experimental conditions: 40 °C; 0.2 MPa N2,
glycerol:acetone molar ratio = 1:6, 0.3 wt.% (catalyst: glycerol) and an 80 min reaction in
each run.
3.2. Superficial characterization
10

The N2 adsorption-desorption isotherms obtained for the samples are shown in the
Supplementary Information (Figure S1). All samples show type IV isotherms with a
hysteresis loop of H3 type according to the IUPAC classification. Table 2 summarizes the
values of specific areas, average pore size, and total pore volume, for all the samples
prepared.
Table 2: Textural and acidic properties
BET
Potentiometric titration
Sample
BET Surface Pore volume Pore diameter E0
meq n-butylamine.g-1
(m2.g-1)
105
(cm3.g-1)
0.34
(nm)
5.1
(mV)
63
Zr
0.20
0.15
0.75
0.52
0.52
Zr600
49
58
86
0.30
0.23
0.29
0.13
12.3
5.8
5.9
11
134
560
310
510
Zr-S-400
Zr-S-600
Zr600-S-400 24
Figure 2 presents the XRD patterns of commercial zirconia and sulfate promoted zirconium
oxides. It can be observed that all materials exhibit both characteristic peaks from the
monoclinic (2θ = 28.19°, 31.48°, 34.17°, 50.13°) and tetragonal zirconia phase (2θ =
30.18°), the monoclinic phase being dominant (JCPDS 86-1451, JCPDS 81-1544). Given
that all diffractograms show the same characteristic peaks, it is possible to conclude that
sulfation and subsequent heat treatment do not substantially modify the crystal structure.
However, some morphological changes are evidenced. Calcination of the commercial
11

zirconium oxide at 600 °C (sample Zr600) results in a drop in surface area and, therefore, an
increase in the pore diameter from 5.1 to 12.3 nm.
Regarding the sulfated samples, Zr-S-400 and Zr600-S-400 present smaller surface areas
than the unsulfated oxides, which would indicate the incorporation of sulfated species. By
contrast, the larger surface area of the sulfated sample Zr-S-600 in comparison to sample
Zr-S-400 could be evidence of the loss of sulfate groups caused by an increase in the
calcination temperature.
Figure 2: X-ray diffraction patterns of commercial zirconia and sulfate promoted
zirconium oxides. M: monoclinic phase, T: tetragonal phase.
By means of potentiometric titration with n-butylamine, it is possible to estimate the
strength and number of acid sites present in a solid. It is considered that the initial potential
(E0) indicates the maximum strength of the acid sites, and the value from which the plateau
is reached is indicative of the total number of acid sites present in the titrated solid [53].
Potentiometric titration curves are shown in the Supplementary Information (Figure S2),
12

whereas the results are summarized in Table 2. It can be observed that all sulfated samples
have higher acidity than the Zr and Zr600 samples and that an increase in the calcination
temperature (from 400 to 600 °C) leads to a decrease in the number of acid sites in the
sample. The latter behavior could be related to the loss of sulfate species at temperatures
above 600 °C [54].
The superficial coverage of sulfate species was explored by XPS. The results of fresh Zr-S-
400 and that reused in 4 runs show an S/Zr atomic ratio of 0.23 and 0.18, respectively
(Figure S3). These values are similar, within the error of the technique, demonstrating the
stability of the superficial sulfate species under the conditions studied.
An IR analysis of pyridine adsorbed on the Zr-S-400 sample was performed to determine
the type of active site of the catalyst. For sulfated zirconia, traditional spectral interpretation
attributes absorption peaks at 1607, 1574, and 1445 cm-1 to Lewis sites [55-57], 1490 cm-1
to Lewis and Brönsted sites [58], and 1640, 1610, and 1540 cm-1 to Brönsted acid sites
[59]. Figure 3 shows the FTIR spectra of the Zr-S-400 sample. This spectrum produced
peaks at 1445, 1490, 1540, 1574, 1610, and 1640 cm-1, showing the presence of both Lewis
and Brönsted acid sites in the material.
13

0,5
0,4
0,3
0,2
0,1
0,0
L
100°C
200°C
300°C
400°C
L+B
L+B
B
B
L
1700
1650
1600
1550
1500
1450
Wavenumber (cm-1)
Figure 3: FTIR spectra of pyridine adsorbed on Zr-S-400 and evacuated at different
temperatures. L = Lewis; B = Brönsted.
For comparison, the FTIR spectra of Zr is shown in the Supplementary Information (Figure
S4). The concentrations of both sites were calculated from the intensities of the bands at
1540 cm-1 (B) and 1440 cm-1 (L), in the spectra obtained at different temperatures for each
material (Table 3). It is observed that the sulfation process generates sites with Brönsted
acidity and slightly increases the acidity of Lewis acid sites in Zr-S-400.
Table 3: Concentration (µmol.g-1) of acid sites as calculated from adsorption/desorption of
pyridine followed by IR absorption spectroscopy.
Concentration of acid sites (µmol.g-1)
T (°C)
14

Zr
Zr-S-400
Lewis Brönsted Lewis Brönsted
100
200
300
400
41
33
15
-
-
-
-
-
53
24
18
9
37
7
2
-
In view of the characterization results obtained, it is evident that the catalytic activity of the
materials is linked to its acid properties. From the results shown in Table 1, the turnover
frequency (TOF) of materials Zr-S-400, Zr-S-600 and Zr600-S-400 were estimated as 66.3,
14.4 and 69.4 min-1, respectively. Samples Zr-S-400 and Zr600-S-400 have practically the
same initial potential in the titration experiments (E0 of 560 mV and 510 mV, respectively)
indicating similar acid strength. This acid strength, which is due to the generation of new
Brönsted acid sites in the sulfation process, is partially lost when the material is calcined at
600°C, leading to a decrease in the catalytic activity in sample Zr-S-600 [54].
3.3. Estimation of thermodynamic parameters
In order to determine the thermodynamic parameters, the reaction was carried out with the
Zr-S-400 catalyst and in sufficient time to ensure no changes in the composition. Figure 4
presents the linearization of the experimental data. With these results, the dependence of
1
(
)
the equilibrium constant with temperature could be expressed as 푙푛 퐾푒푞 = 1397 푇 −
6.304.
The estimated values were ΔH0 = -11.6 ± 1.1 kJ.mol-1, and ΔG0 = 4.0 ± 0.1 kJ.mol-1, which
show the exothermic nature of the reaction. Cornejo et al. [60] conducted a thermodynamic
study considering an ideal behavior of the system and obtained a ΔH0 = -6.6 ± 0.2 kJ.mol-1
and ΔG0 = -0.4 ± 0.1 kJ.mol-1. While their study predicts an exergonic character, ours
predicts an endergonic character, highlighting the significance of considering the non-
ideality of the reactive mixture in these studies.
Other studies have determined thermodynamic equilibrium parameters using ethanol as a
solvent. Nanda et al. [44] determined a ΔH0 = -30.1 ± 1.6 kJ.mol-1 and ΔG0 = -2.1 ± 0.1
kJ.mol-1, considering ideal behavior. Moreira et al. [45] estimated the thermodynamic
15

properties on the basis of activities, taking into account the non-ideal behavior of the
mixture. The obtained values were ΔH0 = -20.1 ± 1.1 kJ.mol-1 and ΔG0 = 1.4 ± 0.1 kJ.mol-1.
Although the thermodynamic parameters of Moreira et al. could not be easily compared
with our results, they also determined an endergonic character.
-1,7
-1,8
-1,9
-2,0
0,00310
0,00315
0,00320
0,00325
0,00330
1/T (K-1)
Figure 4: Linearization of ln (Keq) vs 1/T.
3.4 Reaction kinetic study
Kinetic model
To determine the kinetic model, the following reaction steps were taken:
Table 4: Steps involved in the kinetic model.
Step
Reaction
퐴푐 + ∗ ↔ 퐴푐∗
퐺푙푦 + ∗ ↔ 퐺푙푦∗
퐴푐∗ + 퐺푙푦∗ ↔ 퐻퐴∗ + ∗
퐻퐴∗ + ∗ ↔ 퐼푛푡∗ + 푊∗
퐼푛푡∗ ↔ 푆표푙푘∗
1
2
3
4
5
16

푆표푙푘∗ ↔ 푆표푙푘 + ∗
6
푊∗ ↔ 푊 + ∗
7
The * symbol represents an active site of the catalyst, and the * superscript indicates that
the component is adsorbed on the catalyst surface.
The reaction starts with the adsorption of both reactants on the catalyst surface. Then, a
surface reaction between adsorbed glycerol (Gly*) and acetone (Ac*) leads to the formation
of the hemiketal (3-(2-hydroxypropan-2-yloxy) propane-1,2-diol (HA*). In step 4, a short-
lived intermediate (Int*) and water are formed, followed by a subsequent reaction, where
the intramolecular reaction of the intermediate leads to the formation of Solketal (Solk*).
The final step of the reaction is the desorption of solketal and water from the catalyst
surface.
In our reaction experiments with Zr-S-400, we did not detect the presence of the hemiketal
even at low glycerol conversions. Using in situ Raman studies, Calvino-Casilda et al.
detected the presence of an incipient formation of the hemiketal without a catalyst, and
almost total selectivity to solketal in the presence of a catalyst [40]. Therefore, we consider
that step 3 (formation of the hemiketal) is the controlling one.
The different equations employed (3-9) to develop the model are presented below:
휃퐴푐
퐾1 =
(3)
휃푉. 푎퐴푐
휃퐺푙푦
퐾2 =
(4)
(5)
(6)
휃푉. 푎퐺푙푦
푟 = 퐾3+휃퐴푐휃퐺푙푦 − 퐾3−휃퐻퐴휃푉
휃퐼푛푡. 휃푊
퐾4 =
휃푉. 휃퐻퐴
휃푆표푙푘
퐾5 =
(7)
(8)
휃퐼푛푡
휃푉. 푎푆표푙푘
퐾6 =
휃푆표푙푘
17

휃푉. 푎푊
휃푊
퐾7 =
(9)
where θ represents the fraction of active sites occupied by each species and θV the free
active sites. From equations 3 and 4, it is possible to obtain the following expression:
2
휃퐴푐휃퐺푙푦 = 퐾1. 퐾2. 푎퐴푐. 푎퐺푙푦휃푉
(10)
(11)
From equations 6, 7, 8, and 9, the following expression for θHA is obtained:
푎푆표푙푘. 푎푊. 휃푉
휃퐻퐴
=
퐾4. 퐾5. 퐾6. 퐾7
Furthermore, taking into account the balance of active sites:
푛
1 = ∑ 휃푖
(12)
(13)
푖=1
The reaction rate could be expressed from equations (3-12) as:
푎푆표푙푘. 푎푊
푎퐺푙푦. 푎퐴푐
−
퐾푒푞
푟 = 푘
푛
2
(
∑
)
퐾 . 푎푖
1 +
푖
푖=1
where k is the kinetic constant, Ki is the adsorption equilibrium constant of each
component, and Keq is the equilibrium constant of the reaction.
In the literature, several articles consider that water is the most strongly adsorbed
component, and the adsorption of the remaining species could be neglected [44,45,61]. If
we apply this consideration, the expression of the reaction rate is simplified to:
푎푆표푙푘. 푎푊
푎퐺푙푦. 푎퐴푐
−
퐾푒푞
(14)
푟 = 푘
2
(
)
1 + 퐾푊. 푎푊
It is well known that the adsorption of water on sulfated zirconia catalysts breaks the
coordination of the Zr (IV) species bonded to sulfate species in order to bring Brönsted acid
sites [62]. Therefore, the presence of water does not affect the number of active sites
available, and the reaction rate could be further simplified to:
푎푆표푙푘. 푎푊
(15)
푟 = 푘 (푎퐺푙푦. 푎퐴푐
−
)
퐾푒푞
18

Estimation of kinetic parameters
The Mears criterion values were below 10-5, and the Weisz-Prater criterion values below
10-8, confirming that both external and internal diffusion limitations are negligible on the
operation conditions. Temperature, acetone, and water concentration were varied
systematically. Figure 5 shows the results of the kinetic model and the experimental ones. It
is observed that the behavior at different temperatures is linked to the exothermic character
of the reaction.
80
60
30 °C (Model)
35 °C (Model)
40 °C (Model)
40
50 °C (Model)
30 °C (Exp.)
35 °C (Exp.)
20
40 °C (Exp.)
50 °C (Exp.)
0
0
40
80
120
160
200
240
280
Time (min)
Figure 5: Effect of temperature on glycerol conversion. Experimental conditions: 0.2 MPa
N2, glycerol: acetone molar ratio = 1:6, 0.6 wt.% (catalyst: glycerol), Zr-S-400 catalyst.
Table 4 shows the estimated kinetic constant for each temperature. The estimation was
performed by fitting the experimental data with the mathematical model (equation 15,
Experimental Section). It is worth mentioning that these values were obtained assuming
that none of the species is strongly adsorbed. To confirm this assumption, the experimental
data were also adjusted considering water and solketal as strongly adsorbed species. Since
19

the kinetic parameters obtained were inconsistent, the consideration that none of the species
involved are strongly adsorbed on the catalyst is acceptable.
Table 5: Estimated kinetic constant at different temperatures.
Temperature (°C)
k (mol.g-1.min-1)
0.04544 ± 0.0061
0.07126 ± 0.0079
0.11516 ± 0.0093
0.39720 ± 0.0460
30
35
40
50
From the results shown in Table 4, the activation energy value (Ea) was estimated as 88.1 ±
8.9 kJ.mol-1, and the pre-exponential factor (k0) as 6.55 x 1013 mol.g-1.min-1. These results
are within the range of values obtained by other authors. Esteban et al. [61] estimated a
value of Ea = 124.0 ± 12.9 kJ.mol-1 assuming an Eley-Rideal mechanism on sulphonic ion-
exchange resin in the absence of solvent. Nanda et al. [44] and Moreira et al. [45] reported
Ea = 55.6 ± 3.1 kJ.mol-1 and Ea = 69.0 ± 6.6 kJ/mol, respectively assuming an LHHW
mechanism considering water as the most adsorbed species and using Amberlyst-35 and
ethanol as a solvent. Rossa et al. [63] estimated an Ea = 44.8 ± 1.2 kJ.mol-1 assuming a
pseudo-homogeneous model and using an H-BEA zeolite. In the matter of homogeneous
catalysis, da Silva et al. [64] and Ji et al. [65] reported the Ea of 26.0 kJ.mol-1 using salt
Fe(NO3)3.9 H2O and the Ea of 28.2 kJ.mol-1 using ionic liquid [P(C4H9)3C14H29][TsO],
respectively.
Figures 6 A and B show the effect of the initial glycerol to acetone molar ratio on the
kinetics and thermodynamics of the reaction. As shown in these figures, the increase in
acetone concentration not only increases the conversion reached in the equilibrium but also
20

improves the reaction kinetics. This effect is more evident at lower catalyst content (Figure
6 B).
B
A
80
60
40
20
0
80
60
40
20
0
1:4 (Model)
1:6 (Model)
1:8 (Model)
1:4 (Exp.)
1:6 (Exp.)
1:8 (Exp.)
1:4 (Model)
1:6 (Model)
1:8 (Model)
1:4 (Exp.)
1:6 (Exp.)
1:8 (Exp.)
0
40 80 120 160 200 240 280
0
40 80 120 160 200 240 280
Time (min)
Time (min)
Figure 6: Effect of the glycerol to acetone molar ratio in the feed on glycerol conversion.
Experimental conditions: 0.2 MPa N2, 40 °C, Zr-S-400 catalyst. A: 0.6 wt.% (catalyst:
glycerol); B: 0.3 wt.% (catalyst: glycerol).
Since one of the main impurities in the glycerol feedstock is the presence of water, it is
essential to study the influence of this component on the reaction. In our study, water was
added to the reaction mixture to simulate different glycerol feedstocks. Figure 7 shows the
effect of water on the reaction rate at 5%, 10%, and 20% wt. of the total mass of glycerol.
The presence of water in the reaction generates more extreme effects on the reaction rate
and the conversion at equilibrium state. Feedstocks presenting 20% wt. of water
significantly reduce the reaction rate, and the conversion at equilibrium decays from 80% to
63%. This highlights the need to use purified glycerol feedstock, with water concentration
below 10% for better efficiencies in the chemical process.
21

80
60
40
20
0
0% (Model)
5% (Model)
10% (Model)
20% (Model)
0% (Exp.)
5% (Exp.)
10% (Exp.)
20% (Exp.)
0
40
80
120
160
200
240
280
Time (min)
Figure 7: Effect of water composition in the feed on glycerol conversion. Experimental
conditions: 0.2 MPa N2, 40 °C, glycerol: acetone molar ratio = 1:6, 0.6 wt.% (catalyst:
glycerol), Zr-S-400 catalyst.
4. Conclusions
The sulfation of a commercial zirconium oxide, by wet impregnation with a 0.5M H2SO4
solution, allowed us to obtain an active catalyst for the glycerol ketalization in the liquid
phase. The characterization results showed a correlation between the calcination
temperature and the acid sites generated on the materials. Among the catalysts studied, the
Zr-S-400 material was the most active one and presented the largest acid density and
greatest acid strength caused by the generation of new Brönsted sites, as determined by
potentiometric titration and Pyridine FTIR. This catalyst exhibited a glycerol conversion of
80% in 1 hour of reaction at 40°C with a glycerol to acetone molar ratio = 1:6 and 0.6 wt%
catalyst:glycerol.
The chemical equilibrium was analyzed using the UNIFAC method to determine the
activity coefficients of each species. The thermodynamic parameters, standard enthalpy and
22

Gibbs free energies, were determined to be -11.6 ± 1.1 kJ.mol-1 and 4.0 ± 0.1 kJ.mol-1,
respectively.
Taking into account that the adsorption of water on this catalyst does not affect the number
of acid sites available, a simple pseudo-homogeneous kinetic expression was developed and
successfully adjusted to the experimental data in the range studied. Based on this model, the
estimated activation energy of the reaction was 88.1 ± 8.9 kJ.mol-1.
The study of the sensitivity of the reaction to the presence of water showed that the
presence of water in the reactive mixture generates negative effects on the thermodynamics
and kinetics of the reaction and, more importantly, it was determined that the concentration
of water in the feedstock should be lower than 10% to keep the efficiency of the process.
Credit Author Statement
Julián A. Vannucci carried out all the experimental work, which was conceived and designed with Nora
Nichio. Francisco Pompeo participated in the discussion of the results and the writing of the paper.
Declaration of interests
The authors declare that they have no known competing financial interests or personal relationships
that could have appeared to influence the work reported in this paper.
Acknowledgments
This research work was possible due to the funds received from “Consejo Nacional de
Investigaciones Científicas y Técnicas” (CONICET-PIP 0065), “Universidad Nacional de
La Plata” (UNLP-I248). The doctoral fellowship granted by CONICET to Julian A.
Vannucci is greatfully acknowledged.
References
[1] J. A. Melero, G. Vicente, G. Morales, M. Paniagua, J. Bustamante, Fuel, 89 (2010)
2011-2018.
[2] M. R. Nanda, Y. Zhang, Z. Yuan, W. Qin, H. S. Ghaziaskar, C. Xu, Renewable and
Sustainable Energy Reviews, 56 (2016) 1022-1031.
[3] K. Fraatz, D. Mertin, I. Heep, US Patent 8231903B2, 2012.
23

[4] C. J. A. Mota, C. X. A. da Silva, N. Rosenbach Jr, J. Costa, F. da Silva, Energy Fuels, 24
(2010) 2733-2736.
[5] J. Esteban, A. J. Vorholt, A. Behr, M. Ladero, F. Garcia-Ochoa, Journal of Chemical &
Engineering Data, 59 (2014) 2850-2855.
[6] J. Esteban, F. García-Ochoa, M. Ladero, Green Processing and Synthesis, 6 (2017) 79-
89.
[7] P. Ferreira, I. M. Fonseca, A. M. Ramos, J. Vital, J. E. Castanheiro, Applied Catalysis B:
Environmental, 98 (2010) 94-99.
[8] M. N. Timofeeva, V. N. Panchenko, V. V. Krupskaya, A. Gil, M. A. Vicente, Catalysis
Communications, 90 (2017) 65-69.
[9] G. Vicente, J. A. Melero, G. Morales, M. Paniagua, E. Martín, Green Chemistry, 12
(2010) 899-907.
[10] L. Li, D. Cani, P. P. Pescarmona, Inorganica Chimica Acta, 431 (2015) 289-296.
[11] Z. Li, Z. Miao, X. Wang, J. Zhao, J. Zhou, W. Si, S. Zhuo, Fuel, 233 (2018) 377-387.
[12] H. Serafim, I. M. Fonseca, A. M. Ramos, J. Vital, J. Castanheiro, Chemical
Engineering Journal, 178 (2011) 291-296.
[13] S. S. Priya, P. Selvakannan, K. V. Chary, M. L. Kantam, S. K. Bhargava, Molecular
Catalysis, 434 (2017) 184-193.
[14] L. H. Vieira, L. G. Possato, T. F. Chaves, S. H. Pulcinelli, C. V. Santilli, L. Martins,
Molecular Catalysis, 458 (2018) 161-170.
[15] X. Li, Y. Jiang, R. Zhou, Z. Hou, Applied Clay Science, 174 (2019) 120-126.
[16] Z. Miao, Z. Li, M. Liang, J. Meng, Y. Zhao, L. Xu, J. Mu, J. Zhou, S. Zhuo, W. Si,
Chemical Engineering Journal, 381 (2020) 122594.
[17] X. Li, Y. Jiang, R. Zhou, Z. Hou, Applied Clay Science, 189 (2020) 105555.
[18] V. S. Marakatti, S. Marappa, E. M. Gaigneaux, New Journal of Chemistry, 43 (2019)
7733-7742.
[19] J. R. Sohn, D. H. Seo, Catalysis Today, 87 (2003) 219-226.
[20] A. I. Ahmed, S. A. El-Hakam, S. E. Samra, A. A. EL-Khouly, A. S. Khder, Colloids
and Surfaces A: Physicochemical and Engineering Aspects, 317 (2008) 62-70.
[21] A. Corma, M. I. Juan-Rajadell, J. M. López-Nieto, A. Martinez, C. Martínez, Applied
Catalysis A: General, 111 (1994) 175-189.
24

[22] X. Li, K. Nagaoka, J. A. Lercher, Journal of Catalysis, 227 (2004) 130-137.
[23] N. Essayem, Y. Ben Taârit, C. Feche, P. Y. Gayraud, G. Sapaly, C. Naccache, Journal
of Catalysis, 219 (2003) 97-106.
[24] A. Corma, J. M. Serra, A. Chica, Catalysis Today, 81 (2003) 495-506.
[25] A. Wolfson, O. Shokin, D. Tavor, Journal of Molecular Catalysis A: Chemical, 226
(2005) 69-76.
[26] H. Nagai, K. Kawahara, S. Matsumura, K. Toshima, Tetrahedron Letters, 42 (2001)
4159-4162.
[27] P. S. Reddy, P. Sudarsanam, B. Mallesham, G. Raju, B. M. Reddy, Journal of Industrial
and Engineering Chemistry, 17 (2011) 377-381.
[28] V. T. Vasantha, N. J. Venkatesha, S. Z. M. Shamshuddin, J. Q. D'Souza, B. G. V.
Reddy, ChemistrySelect, 3 (2018) 602-608.
[29] M. R. Nanda, Z. Yuan, W. Qin, H. S. Ghaziaskar, M. A. Poirier and C. Xu, Applied
Energy, 123 (2014) 75-81.
[30] S. Brunauer, P. H. Emmett, E. Teller, Journal of the American Chemical Society, 60
(1938) 309-319.
[31] E. P. Barrett, L. G. Joyner, P. P. Halenda, Journal of the American Chemical Society,
73 (1951) 373-380.
[32] C. Emeis, Journal of Catalysis, 141 (1993) 347-354.
[33] L. Roldán, R. Mallada, J. M. Fraile, J. A. Mayoral, M. Menéndez, Asia-Pacific Journal
of Chemical Engineering, 4 (2009) 279-284.
[34] O. Ilgen, S. Yerlikaya, F. O. Akyurek, Periodica polytechnica chemical engineering, 61
(2017) 144-148.
[35] S. Sandesh, A. B. Halgeri, G. V. Shanbhag, Journal of Molecular Catalysis A:
Chemical, 401 (2015) 73-80.
[36] S. Zhang, S. Zhao, Y. Ao, Applied Catalysis A: General, 496 (2015) 32-39.
[37] P. Manjunathan, S. P. Maradur, A. B. Halgeri, G. V. Shanbhag, Journal of Molecular
Catalysis A: Chemical, 396 (2015) 47-54.
[38] J. Kowalska-Kus, A. Held, M. Frankowski, K. Nowinska, Journal of Molecular
Catalysis A: Chemical, 426 (2017) 205-212.
25

[39] A. Fredenslund, J. Gmehling, P. Rasmussen, Vapor-liquid Equilibria Using Unifac: A
Group-Contribution Method, Elsevier, 1977.
[40] V. Calvino-Casilda, K. Stawicka, M. Trejda, M. Ziolek, M. A. Bañares, The Journal of
Physical Chemistry, 118 (2014) 10780-10791.
[41] B. Mallesham, P. Sudarsanam, G. Raju, B. M. Reddy, Green Chemistry, 15 (2013) 478-
489.
[42] L. Li, T. I. Korányi, B. F. Sels, P. P. Pescarmona, Green Chemistry, 14 (2012) 1611-
1619.
[43] L. P. Ozorio, R. Pianzolli, M. B. S. Mota, C. J. A. Mota, Journal of the Brazilian
Chemical Society, 23 (2012) 931-937.
[44] M. R. Nanda, Z. Yuan, W. Qin, H. S. Ghaziaskar, M. A. Poirier, C. C. Xu, Fuel, 117
(2014) 470-477.
[45] M. N. Moreira, R. P. V. Faria, A. M. Ribeiro, A. E. Rodrigues, Industrial &
Engineering Chemistry Research, 58 (2019) 17746-17759.
[46] E. G. Scheibel, Industrial & Engineering Chemistry, 46 (1954) 1569-1579.
[47] L. R. Perkins, C. J. Geankoplis, Chemical Engineering Science, 24 (1969) 1035-1042.
[48] A. W. Hixson, S. J. Baum, Industrial & Engineering Chemistry, 33 (1941) 478-485.
[49] A. Feliczak-Guzik, I. Nowak, Microporous and Mesoporous Materials, 277 (2019) 301-
308.
[50] I. S. Gomes, D. C. de Carvalho, A. C. Oliveira, E. Rodríguez-Castellón, S.
Tehuacanero-Cuapa, P. T. C. Freire, J. M. Filho, G. D. Saraiva, F. F. de Sousa, R. Lang,
Chemical Engineering Journal, 334 (2018) 1927-1942.
[51] A. L. G. Pinheiro, J. V. C. do Carmo, D. C. Carvalho, A. C. Oliveira, E. Rodríguez-
Castellón, S. Tehuacanero-Cuapa, L. Otubo, R. Lang, Fuel Processing Technology, 184
(2019) 45-56.
[52] D. C. de Carvalho, A. C. Oliveira, O. P. Ferreira, J. M. Filho, S. Tehuacanero-Cuapa,
A. C. Oliveira, Chemical Engineering Journal, 313 (2017) 1454-1467.
[53] L. R. Pizzio, M. N. Blanco, Microporous and Mesoporous Materials, 103 (2007) 40-47.
[54] G. X. Yan, A. Wang, I. E. Wachs, J. Baltrusaitis, Applied Catalysis A, General, 572
(2019) 210-225.
26

[55] B. T. Loveless, A. Gyanani, D. S. Muggli, Applied Catalysis B: Environmental, 84
(2008) 591-597.
[56] C. D. Miranda M., A. E. Ramírez S., S. G. Jurado, C. R. Vera, Journal of Molecular
Catalysis A: Chemical, 398 (2015) 325-335.
[57] C. Breitkopf, S. Matysik, H. Papp, Applied Catalysis A: General, 301 (2006) 1-8.
[58] W. H. Chen, H. H. Ko, A. Sakthivel, S. J. Huang, S. H. Liu, A. Y. Lo, T. C. Tsai, S. B.
Liu, Catalysis Today, 116 (2006) 111-120.
[59] K. T. Wan, C. B. Khouw, M. E. Davis, Journal of Catalysis, 158 (1996) 311-326.
[60] A. Cornejo, M. Campoy, I. Barrio, B. Navarrete, J. Lázaro, Reaction Chemistry &
Engineering, 4 (2019) 1803-1813.
[61] J. Esteban, M. Ladero, F. García-Ochoa, Chemical Engineering Journal, 269 (2015)
194-202.
[62] M. Hino, M. Kurashige, H. Matsuhashi, K. Arata, Thermochimica Acta, 441 (2006) 35-
41.
[63] V. Rossa, Y. S. P. Pessanha, G. C. Díaz, L. D. T. Câmara, S. B. C. Pergher, D. A. G.
Aranda, Industrial & Engineering Chemistry Research, 56 (2017) 479-488.
[64] M. J. da Silva, A. A. Rodrigues, P. F. Pinheiro, Fuel, 276 (2020) 118164.
[65] Y. Ji, T. Zhang, X. Gui, H. Shi, Z. Yun, Chinese Journal of Chemical Engineering, 28
(2020) 158-164.
27