Isoamyl acetate preparation from reaction of vinyl acetate and Isoamyl alcohol catalyzed by H-ZSM-5 zeolite: a kinetic study

Article abstract of DOI:10.1016/j.mcat.2018.06.009

Microporous H-ZSM-5 zeolite having good crystallinity and with the presence of Br?nsted acid sites exclusively, was prepared by hydrothermal treatment of Expanded Perlite. The as-prepared H-ZSM-5 was used as heterogeneous catalyst for the reaction between vinyl acetate and isoamyl alcohol, as a greener alternative for the traditional preparation of isoamyl acetate. In this reaction, isoamyl acetate was generated by an acyl nucleophilic substitution mechanism, mediated by an acetylated zeolite complex, while acetaldehyde diisoamyl acetal has been produced as a secondary product due to a nucleophilic addition to acetaldehyde carbonyl carbon. The reaction was studied under different alcohol/ester molar ratio, supporting a second order kinetics when working under excess of one reactant. Moreover, an Eley-Rideal kinetic model has been suggested for a reaction mechanism of five steps, involving the ketalization of isoamyl alcohol and acetaldehyde as a parallel reaction that competes with transesterification.

Full text of DOI:10.1016/j.mcat.2018.06.009

MolecularCatalysisxxx(xxxx)xxx–xxx
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.elsevier.com/locate/mcat
Isoamyl acetate preparation from reaction of vinyl acetate and Isoamyl
alcohol catalyzed by H-ZSM-5 zeolite: a kinetic study
Pablo F. Corregidor⁎, Delicia E. Acosta, Elio E. Gonzo, Hugo A. Destéfanis
Facultad de Ingeniería, Instituto de Investigaciones para la Industria Química – INIQUI (UNSa-CONICET), Centro Científico Tecnológico (CCT) Salta, Consejo de
Investigaciones de la UNSa (CIUNSa), Universidad Nacional de Salta, Av. Bolivia 5150, A4408FVYA Salta, Argentina
A R T I C L E I N F O
A B S T R A C T
Keywords:
Microporous H-ZSM-5 zeolite having good crystallinity and with the presence of Brønsted acid sites exclusively,
was prepared by hydrothermal treatment of Expanded Perlite. The as-prepared H-ZSM-5 was used as hetero-
geneous catalyst for the reaction between vinyl acetate and isoamyl alcohol, as a greener alternative for the
traditional preparation of isoamyl acetate. In this reaction, isoamyl acetate was generated by an acyl nucleo-
philic substitution mechanism, mediated by an acetylated zeolite complex, while acetaldehyde diisoamyl acetal
has been produced as a secondary product due to a nucleophilic addition to acetaldehyde carbonyl carbon. The
reaction was studied under different alcohol/ester molar ratio, supporting a second order kinetics when working
under excess of one reactant. Moreover, an Eley-Rideal kinetic model has been suggested for a reaction me-
chanism of five steps, involving the ketalization of isoamyl alcohol and acetaldehyde as a parallel reaction that
competes with transesterification.
Transesterification
Isoamyl acetate
ZSM-5
Kinetic study
Introduction
ethyl esters are commercially accessible and thus they can be con-
veniently employed as starting materials in transesterification.
A Fischer’s esterification is a special type of reaction where a nu-
cleophilic acyl substitution take places over a carboxylic acid (RCOOH)
based on the electrophilicity of the carbonyl carbon, which usually
activates over an acidic catalyst. A possible mechanism for Fischer’s
esterification undergoes through the formation of a carbocation inter-
mediate specie (RC≡O+). This acylium ion is an electrophilic specie
which further reacts with weak nucleophilic reagents such as alcohols
to produce esters and water. This reaction was first described by Emil
Fischer in 1895 as a thermodynamically reversible process in which
products and reactants remains an equilibrium state [1].
The extensive application in industry is the crucial feature of es-
terification that particularly differentiates from other reactions [2].
Nevertheless, some carboxylic acids have a low solubility in organic
solvents which difficult to carried out a homogeneous esterification,
whereas esters are normally soluble in most of organic solvents. In that
sense, a transesterification reaction is more advantageous than the ester
synthesis from carboxylic acids and alcohols [3]. Thus, a transester-
ification is a classic organic reaction that has revealed a lot of industrial
applications and laboratory purposes. The ester-to-ester transformation
involved a reaction between an ester (R’COOR’’) and an alcohol (ROH),
this is particularly advantageous when the pair carboxylic acids are
easily altered and difficult to isolate. Some esters, especially methyl and
A transesterification reaction can also be studied as an acyl transfer-
process where an ester relocates its acyl moiety over a reactant alcohol
(acyl acceptor). This reaction remains at equilibrium in organic media
and can be shifted in the desired direction in order to improve the
conversion of reaction. In some cases, the use of a large excess of the
acyl donor or the removal of one of the products are options to drive the
process to completion. However, in practice, a more common solution is
the use of activated esters, which ensure a more or less irreversible
reaction. One of the best activated acyl donors are enol esters such as
vinyl acetate [4,5]. In this case, the resultant product is vinyl alcohol,
which simply tautomerizes to acetaldehyde [6]. The last one is a low-
boiling point aldehyde that easily evaporates above room temperature.
Moreover, the generation of a keto-compound (a weak nucleophile)
instead of an alcohol improves the non-reversibility of the process.
On the other hand, modern catalysis is moving to a clean chemistry
due to high performance and environmental impact, thus, solid cata-
lysts emerge as a competitive alternative to traditional chemical
synthesis for the production of fine chemicals. In that sense, zeolites are
well known as heterogeneous acid catalysts and also have proved good
activity in transesterification reactions [7–10]. Unlike the acylium ion
obtained in most of classical nucleophilic acyl substitutions involving
homogeneous acid catalysts, there is evidence for the formation of an
⁎
Corresponding author.
E-mail address: pcorregidor@unsa.edu.ar (P.F. Corregidor).
https://doi.org/10.1016/j.mcat.2018.06.009
Received 13 April 2018; Received in revised form 22 May 2018; Accepted 6 June 2018
2468-8231/©2018ElsevierB.V.Allrightsreserved.
Pleasecitethisarticleas:Corregidor,P.F.,MolecularCatalysis(2018),https://doi.org/10.1016/j.mcat.2018.06.009

P.F. Corregidor et al.
MolecularCatalysisxxx(xxxx)xxx–xxx
acylated surface over nanostructured heterogeneous catalyst such as
heteropoly acids [11] and zeolites [12–15]. These authors agree that
acylium cation is formed during acylation reactions over a zeolite cat-
alyst, which can be stabilized by negative charged oxygen on the active
site. This stabilized acylium ion could then react, i.e. in a bimolecular
electrophilic aromatic substitution mechanism [14].
Therefore, based in previous reports where the acylated surface of a
zeolite can perform acylation reactions, we studied the preparation of
isoamyl acetate, one of the most important flavor compounds used in
food industries because of its characteristic banana flavor, using a
transesterification reaction as an alternative to the traditional Fischer’s
esterification mechanism. Thus, we evaluated the influence of initial
concentration of reactants over a transesterification reaction, catalyzed
by the acid form of a ZSM-5 zeolite, obtained by a greener method.
Moreover, we studied some kinetics aspects related with a possible
mechanism of reaction and proposed a kinetic model for the studied
reaction, providing a molecular interpretation that complements and
enriches the experimental results.
phase of ZSM-5 zeolite, and the degree of crystallinity was calculated by
comparing the sum of the areas below the dominant peaks between 22°
and 25° of 2θ in the as-prepared zeolite, with the sum of the areas above
the reference ZSM-5 sample (ALSI-Penta SN-55, SiO2/Al2O3 = 23).
The specific surface area and the porosity of the samples were de-
termined from N2 adsorption/desorption isotherms using the multi-
point BET method. The samples were outgassed under vacuum at 623 K
for 10 h prior to the measurements. The isotherms were obtained with a
Quantachrome Quadrasorb SI automated gas adsorption system, at li-
quid nitrogen temperature. The external surface area and the micropore
volume were determined from the adsorption branch of the isotherm
using the t-plot method.
Crystal morphology and size were identified by Scanning Electron
Microscopy (SEM) with a JEOL JSM-6480 LV microscope. Before
measurement, all the samples were coated by gold sputtering.
The aluminum coordination state of the zeolite samples was con-
firmed by 27Al magic angle spinning nuclear magnetic resonance (27Al
MAS NMR). The spectra were recorded on a Bruker Advance DSX400
spectrometer operating at a magnetic field strength of 9.4 T; 36,000
scans were accumulated with a spinning frequency of 20 kHz, a pulse
length of 0.3 μs and a recycle delay of 100 ms. The 27Al signals were
referenced to an externally located 0.1 mol L−1 aqueous solution of [Al
Experimental
Materials and reagents
(H2O)6]3+
.
Expanded perlite from San Antonio de los Cobres (Northwestern
Argentina) was previously conditioned by washing with water, fol-
lowed by grinding and sieving through a 230 mesh sieve before using as
starting material. In order to obtain the appropriate SiO2/Al2O3 molar
ratio in the gel of synthesis, sodium silicate (Fisher; Na2O 27.7 wt%
with a SiO2/Na2O ratio of 2:1, was used as additional Si source. A so-
dium form of a commercial ZSM-5 zeolite (ALSI-Penta SN-55, SiO2/
Al2O3 = 23) was used as a crystallization seed.
For the transesterification reaction, a solution of vinyl acetate
(Sigma Aldrich, ≥ 99%) and isoamyl alcohol (Biopack, ≥ 98.5%) in
toluene (Merck, GC), was employed. All reagents and materials were
used as received.
29Si magic angle spinning nuclear magnetic resonance spectra (29Si
MAS NMR) were recorded by accumulating 4000 scans with a spinning
frequency of 5 kHz, a pulse width of 5 μs and a pulse delay of 60 s on a
Bruker AMX300 spectrometer working at 7.0 T. Tetramethylsilane was
used as a chemical shift reference.
The strength and distribution of acid sites were tested by adsorption
and programmed desorption of pyridine using infrared (IR) spectro-
scopy. The infrared spectra of the zeolite sample were recorded in a
Nicolet 6700 spectrometer equipped with a DTGS detector (128 scans, 2
cm−1 resolution). A self-supporting wafer (of about 15 mg) of the
sample was evacuated in a homemade vacuum infrared cell fitted with
ZnSe windows. The wafer was dried at 673 K for 1 h under vacuum.
During the cooling down of the sample, reference spectra were recorded
at 623, 523, 423 and 323 K. The evacuated sample was saturated with
about 25 mbar of pyridine vapor at 323 K. The saturated wafer was then
evacuated at 323 K for 30 min. The temperature-programmed deso-
rption of adsorbed amine was carried out at a heating rate of 5 de-
grees·min−1, maintaining the temperature at 423, 523 and 623 K for
30 min. Then, the infrared spectra of adsorbed pyridine were recorded
at these temperatures. The corresponding reference spectra were sub-
tracted from these spectra and the band intensities at 1540 and 1450
cm−1 wavenumbers were determined and assigned to bands of
Brønsted and Lewis acid sites of adsorbed pyridine, respectively. The
concentration of Brønsted and Lewis acid sites was calculated with the
integral molar extinction coefficients of pyridine infrared absorption
bands determined by Emeis [18].
Synthesis of the ZSM-5 catalyst
The Na-form of a ZSM-5 zeolite was synthesized as reported pre-
viously [16]. Basically, the expanded perlite was used as the only alu-
mina source and sodium silicate as supplementary silica source. A small
quantity (7% wt of the total amount of SiO2) of seeding Na-ZSM-5
zeolite was added into the reaction mixture with vigorous mixing. The
hydrothermal synthesis of ZSM-5 was performed at a reaction tem-
perature of 453 K for 24 h in 20 mL Teflon-lined stainless steel-auto-
claves. H2SO4 or NaOH aqueous solutions were used to adjust the pH to
10.2 and the Na2O/SiO2 desired molar ratio of the reaction mixture.
The composition of the mother liquor was 40SiO2:Al2O3:0,38-
Na2O:45H2O.
Transforming Na-ZSM-5 zeolite into the H-form
Liquid phase reaction of vinyl acetate and isoamyl alcohol
The zeolite obtained in the Na-form was previously converted to the
NH4-ZSM-5 form following the procedure described by Triantafyllidis
et al. [17]. A 10 wt% solution of NH4Cl at a ratio of 1:10 was used for
the ion exchange process. This mixture was stirred at 353 K for 1 h,
filtered and washed with Milli-Q water (182 MΩ cm). This exchange
procedure was repeated three times. The solid was collected and dried
at room temperature for 12 h. At the end, the NH4-form was converted
into the H-ZSM-5 form by heating under air flow at 723 K overnight.
The reaction of vinyl acetate with isoamyl alcohol was carried out in
10 mL tightly closed glass vials under vigorous stirring (1000 rpm) in a
multiple-well parallel reaction block. In a typical run, a solution con-
taining: vinyl acetate (0.10–1.00 mol L−1), isoamyl alcohol (0.10-
0.30 mol L−1) and toluene (solvent) was injected to a sealed vial con-
taining 110 mg of H-ZSM-5 at room temperature. Normally, a set of two
experiments (experiment A and B) were performed to evaluate the in-
fluence of concentration in the studied reaction. In the first one (ex-
periment A), the influence of the vinyl acetate concentration was varied
from 0.10 mol L−1 to 1.00 mol L−1, while keeping constant the con-
centration of isoamyl alcohol at 0.10 mol L−1. In a second experiment
(experiment B), the influence of the isoamyl alcohol concentration
(0.10-0.30 mol L−1) was evaluated while the vinyl acetate was in-
variant (0.10 mol L−1). These conditions were resumed in Table 3.
Characterization of the catalyst
The powder X-ray diffraction (XRD) patterns were recorded on a
STOE STADI P instrument using CuKα radiation (λ = 0.15415 nm).
Diffraction lines of 2θ between 5° and 45° were taken to confirm the
2

P.F. Corregidor et al.
MolecularCatalysisxxx(xxxx)xxx–xxx
The reaction mixture was stirred and heated at 363 K for 3 h.
Consequently, at defined times of reaction, an aliquot of 0.1 mL was
taken from the system (every 5 min for the first 20 min, then at 30, 40
and 60 min and finally every 30 min from 60 to 180 min of reaction).
These solutions were analyzed by gas chromatography (GC) on a Perkin
Elmer gas chromatograph, model Claurus 580, equipped with a capil-
lary column (Elite-5: 0.25 μm x 30 m x 0.25 mm i.d.) and a flame ion
detector. The temperature profile of the GC analysis was: 5 min at
313 K, from 313 to 453 K at 20 degrees min−1 and finally keeping
1.0 min at 453 K. The identity of the reactants and products was con-
firmed by gas chromatography-mass spectrometry analysis (GC–MS) on
Agilent 6890 N Gas Chromatograph (HP-5MS: 0.25 μm
x
30 m x 0.25 mm i.d.) coupled to Agilent 5973 MSN Mass Spectrometer.
A blank test consisting in a reaction system of vinyl acetate and
isoamyl alcohol in toluene with no catalyst, was carried out under the
same temperature and stirring conditions as mentioned before. Less of
3% of conversion was obtained at nearly 24 h of reaction, undoubtedly
proving that results obtained for catalyzed experiments are due to the
activity of ZSM-5.
Fig. 2. Nitrogen adsorption isotherms.
Table 1
Surface area and porosity characteristics of expanded perlite and the as-pre-
pared ZSM-5 zeolite.
Conversions of isoamyl alcohol and vinyl acetate were calculated
based on Eqs. (1) and (2), respectively:
sample
SBET
Total pore
volumea,
Vtot
Micropore
Mesopore
volume,
VMes
Average
pore ratio,
rp (nm)
m2 g−1
volume, Vmic
mol of isoamyl alcohol converted
conversion of isoamyl alcohol (%) =
(cm3 g−1
)
(cm3 g−1
)
(cm3 g−1
VMes = Vtot
- Vmic
)
initial mol of isoamyl alcohol
t-plotb D.A.
(1)
∙100
eqc
mol of vinyl acetate converted
initial mol of vinyl acetate
Expanded
Perlite
Na-ZSM-5
2
0.05
0.13
0.00
0.11
0.00
0.12
0.05
12.50d
conversion of vinyl acetate (%) =
∙100
290
0.01b - 0.02c 0.55c
(2)
a
Selectivity to isoamyl acetate was calculated according to Eq. (3).
At p/p0 = 0.99 (Gurvitch rule).
b
c
t-plot method (Halsey model).
mol of isoamyl acetate
mol of isoamyl alcohol converted
d
Sisoamyl acetate (%) =
∙100
Dubinin-Ashtakov equation. determined by BJH method.
(3)
(4)
during the thermal expansion process that was submitted. This process
generates a sponge-like structure as evidenced in Fig. 3A.
Therefore, the selectivity towards the byproduct (a ketal) is:
Sketal (%) = 100−Sisoamyl acetate
On the other hand, the H-ZSM-5 displayed a nitrogen adsorption
isotherm of type I, typical from a microporous material (Fig. 2) [19,20].
The hydrothermal treatment of Perlite generates a zeolite with a spe-
cific surface of 290 m2 g−1. Textural properties shown in Table 1 are in
good agreement with those reported elsewhere [21–24].
The X-Ray diffraction pattern (Fig. 1, curve B) revealed no crystal
phase other than ZSM-5, also having a good crystallinity as compared
with a commercial ZSM-5 zeolite (Fig. 1, curve C). The SEM micro-
photographs exhibited well defined coffin-shape particles of nearly
6–7 μm large.
Results and discussion
The H-ZSM-5 catalyst
Curve A in Fig. 1, shows the X-Ray powder diffraction pattern of
Perlite, revealing its amorphous nature, while from Fig. 2 and Table 1
we conclude that this behavior is typical from a non-porous solid
having a BET surface area of 2 m2 g−1. In a practically sense, this
material has no micropores or mesopores and its low porosity must be
explained fundamentally due the presence of macropores, generated
From the MAS NMR spectra of solid state (Fig. 4B) and according to
Eq. (5), the Si/Al molar ratio was estimated as 38.5, where ISi(nAl) is the
peak area of the Si(nAl) signal in the spectra and n is the number of
AlO4 groups linked directly with the SiO4 groups [25].
4
4
(Si/Al)NMR
=
ISi(nAl)
/
0,25∙n ∙ISi(nAl)
∑
∑
(5)
n=0
n=0
Moreover, from the 27Al MAS NMR, we consider that all Al is in-
corporated in the MFI framework as we evidenced a broad signal at 60 –
50 ppm in Fig. 4A, due to the tetrahedral coordination of Si. Also the
absence of a peak at 0 ppm, unambiguously proves that there’s no Al
extraframework (i.e. octahedral coordinated) [25,26].
The number of Brønsted and Lewis acid sites were estimated using
the integrated molar extinction coefficients (1.67 and 2.22 cm mol−1
)
determined by Emeis [18], based on the intensities of the infrared ab-
sorption bands found at 1545 cm−1 and 1450 cm−1, which correspond
to the H-bonding pyridinium ion mode 19b and the coordinatively
adsorbed pyridine [27–29], respectively. FTIR spectra of the as-pre-
pared H-ZSM-5 sample after desorption of pyridine at three different
temperatures are shown in Fig. 5. As was analyzed in our previous
Fig. 1. XRD patterns. A) Perlite, B) the as-prepared ZSM-5, C) commercial ZSM-
5.
3

P.F. Corregidor et al.
MolecularCatalysisxxx(xxxx)xxx–xxx
Fig. 3. SEM microphotographs. A) Expanded Perlite, B) the as-prepared ZSM-5.
paper [16], the concentration of Brønsted sites is 0.30 mmol of pyridine
per gram. It is clearly shown that the area of the band at about 1545
cm−1, related to Brønsted acid sites, is larger than that of the band at
1450 cm−1. Therefore, the amount of Lewis acid sites is lower than the
Brønsted one. The former is practically negligible compared to the
latter. These findings are in agreement with the presence of only a
tetrahedral Al signal in the 27Al MAS NMR spectrum and also with the
Si/Al molar ratio determined by 29Si MAS NMR.
The Brønsted acidity showed a small variation after heating and a
high amount of these acid centers still maintained pyridine bounded,
revealing a strong surface acidity (Table 2). The moderate band ob-
served at 1446 cm−1, in the spectra at 323 K, should not be confused
with the one assigned to Lewis acid sites, since the first practically
disappears when temperature increases to 423 K. This behavior explain
the desorption of weakly bounded pyridine due to interactions with
silanol groups [27], which also agree with the presence of the HO–Si
(O–Si)3 signal at −102 to −108 ppm in the 29Si MAS NMR spectra.
Therefore, after desorption of pyridine from these weak centers, a band
at 1451 cm−1 is observed due to the interactions of very small amounts
of the probe molecule with Lewis acid sites. According to these results,
we will consider in the following sections that the catalytic activity was
due to the presence of Brønsted acid centers, exclusively.
Fig. 5. FTIR spectra of pyridine absorbed on the as-prepared H-ZSM-5 sample
after pyridine desorption at different temperatures: 323, 423 and 523 K.
Table 2
Acidic properties expressed as the number of Brønsted and Lewis sites and
percentage of Brønsted acid centers (ΔBPy), referred to the initial amount at
323 K and saturated with pyridine at given temperature.
b
Temperature
(K)
Brønsted
Lewis
ΔBPy
(%)
(mmol g−1
)
(mmol.g−1
)
The acyl transfer reaction: transesterification vs. ketalization
323
423
523
623
0.30
0.30
0.26
0.21
n.a.a
∼ 0
∼ 0
∼ 0
100
100
87
The transesterification reaction between vinyl acetate and isoamyl
alcohol is shown in Scheme 1. This reaction is an example of a nu-
cleophilic acyl substitution in which vinyl acetate possesses an elec-
trophilic carbonyl carbon which also reacts with the nucleophilic iso-
amyl alcohol to give isoamyl acetate and vinyl alcohol. Although, as
every alkenyl alcohol, vinyl alcohol undergoes a tautomerization into
acetaldehyde. All reactants and products were detected in the gas
chromatogram and the GC–MS analysis. These analysis also evidenced
71
a
n.a. = not applicable.
b
ΔBPy = (Number of Brønsted sites at certain temperature/Number of
Brønsted sites at 323 K)×100.
Fig. 4. NMR spectra of the as-prepared ZSM-5
zeolite. A) 27Al-MAS NMR and B) 29Si MAS-
NMR: experimental spectra (black), calculated
model (red) and Gaussian bands corresponding
to the individual resonances (filled curves) ob-
tained by deconvolution (For interpretation of
the references to colour in this figure legend,
the reader is referred to the web version of this
article).
4

P.F. Corregidor et al.
MolecularCatalysisxxx(xxxx)xxx–xxx
Scheme 1. Reaction of vinyl acetate and iso-
amyl alcohol giving isoamyl acetate and vinyl
alcohol. The last one tautomerizes in acet-
aldehyde.
Scheme 2. Reaction path for the generation of products during the reaction of vinyl acetate and isoamyl alcohol catalyzed by H-ZSM-5.
the generation of a by-product having the IUPAC name of 1,1'-(ethyli-
denebis(oxy))bis(3-methylbutane), corresponding to the structure of
acetaldehyde diisoamyl acetal (Scheme 2).
generation of a free acetaldehyde that easily reacts with isoamyl al-
cohol in a carbonyl nucleophilic addition. On the other hand, the
generation of isoamyl acetate must occur exclusively inside the pores
since the nucleophile (isoamyl alcohol) must find the acetylated zeolite
complex, which then evolves to the desired product following an acyl
nucleophilic substitution mechanism.
Moreover, we discard the generation of ionic species in the pro-
posed reaction scheme, based in our previous results about the influ-
ence of solvents having different dielectric constant [32] over the
mentioned reaction. We found that the conversion of the reaction is
higher when using solvents with low polarity.
This secondary product is generated from the beginning of the re-
action and at first, its signal in the chromatogram grows faster than the
one corresponding to isoamyl acetate. This behavior must suggest that
the nucleophilic attack of isoamyl alcohol to the electrophilic carbonyl
carbon of acetaldehyde is faster than that of the vinyl acetate carbonyl
carbon. Moreover, as we studied the reaction under catalyst-saturated
conditions, an acyl-zeolite intermediate should be formed and thus,
acting as the real acylating agent [14,30]. Therefore, we suggest that
the acetal generation is faster than isoamyl acetate production as the
nucleophilic alcohol can easily find an electrophilic center in the
acetaldehyde carbonyl carbon. Instead of, the acetylated zeolite must
generate isoamyl acetate when the nucleophilic agent cross the channel
system and find the corresponding acyl ion interacting with the fra-
mework. Thus, a nucleophilic attack over the carbonyl carbon of the
acetyl zeolite intermediate generates the desired product (isoamyl
acetate).
Scheme 2 illustrates the proposed paths for the reaction under
study. Since several authors [12–15,30,31] have reported the genera-
tion of an acyl-zeolite complex during acylation reactions using dif-
ferent acyl donors, we suggest at first, the formation of an acetylated
surface from vinyl acetate adsorbed over a Brønsted acid center (step
1). This step also generates vinyl alcohol that quickly tautomerizes in a
free acetaldehyde in step 2. Then, the approximation of isoamyl alcohol
in step 3 has two possible electrophilic centers to attack: (a) the elec-
trophilic center of carbonyl carbon in the acetyl-zeolite complex and (b)
the electrophilic carbonyl carbon of free acetaldehyde. The reaction
that follows path (a) generates isoamyl acetate in step 4a, while the
attack according to route (b) produces a hemiacetal (step 4b) that
further reacts with another mole of the nucleophilic agent to form the
acetaldehyde diisoamyl acetal in step 5.
Influence of the concentration
Fig. 6A and B displayed graphs of conversion of alcohol and ester as
function of time for different alcohol/ester molar ratios while Fig. 6C-D
shows their respective conversion curves. These results indicate that an
excess in the concentration of ester increments the conversion of vinyl
acetate into isoamyl acetate, which can be explained considering that
under a well-sealed system, a transesterification reaction remains at
equilibrium state [2,3,33], governed by Le Châtelier Principle’s. On the
other hand, when increasing the alcohol/ester molar ratio from 1:1 to
2:1 and 3:1, at a constant concentration of vinyl acetate, the maximum
conversion reached for vinyl acetate was 74, 85 and 98%, respectively.
The global chemical reaction including both, transesterification and
ketalization can be expressed as detailed in Scheme 3.
The reaction under study does not have a simple kinetics. Thus, to
stablish the order of the reaction, we applied a graphical method
[34,35] for the determination of the exponents of the general form of
the rate Eq. (6):
v = k [alcohol]α [ester]β
(6)
In excess of vinyl acetate, the α exponent was found to be 2, since a
plot of 1/[alcohol] vs. time gives a straight line. Thus, the reaction is
second order with respect of the isoamyl alcohol concentration.
As mentioned before, the generation of the acetal seems to be the
prime reaction at the beginning. This result is congruent with the
5

P.F. Corregidor et al.
MolecularCatalysisxxx(xxxx)xxx–xxx
Fig. 6. Reaction profiles for the reaction of vinyl acetate an isoamyl alcohol. Effect of the concentration of ester (A, C) and alcohol (B, D).
Consequently, following the same methodology, the β exponent was
found to be 2, meaning that the reaction is second order with respect of
the vinyl acetate concentration when excess of isoamyl alcohol was
used. In that sense, the global order for the studied reaction when using
an excess of one reactant is 2 with respect of the limiting reagent. The
v
rate law can be expressed as showed in Eq. (7), where is the rate of
reaction (mol L−1 min−1), k is the rate constant (L mol−1 min−1) and
the concentrations of reactants in mol L−1 are presented in square
brackets.
β = 2 If [alcohol] ≫ [ester]
⎧
v = k [alcohol]α [ester]β
⎨α = 2 If [ester] ≫ [alcohol]
(7)
⎩
The selectivity of the reaction presented in Fig. 7 shows not sig-
nificate difference when considering the influence of the ester/alcohol
molar ratio. A little excess at favor of isoamyl acetate was observed
when working under equimolar conditions of alcohol and the acyl
donor. Thus, concentration of reactants seems to not have influence in
the selectivity of the reaction.
Fig. 7. Selectivity of the reaction.
catalyst under different conditions. The calculated values for TOFs are
shown in Table 4, they are in good agreement with those reported for
several reactions involving a ZSM-5 catalyst [36–40]. The catalyst has
better activity when considering an excess of one reactant, which can
be explained based in the reversibility of the reaction when maintaining
a well closed system. In fact, an excess of one reactant will modify the
equilibrium by displacing this state to the generation of products.
Moreover, in our intent to have a deep knowledge in the behavior of
the catalyst, we studied some factors that influence the reaction, i.e. the
regeneration and reuse of the catalyst. We find that it conserves ap-
proximately 70% of the activity in a fifth cycle of use, with respect to its
original activity.
Moreover, turnover frequency (TOF), defined as the number of
times that a catalyzed reaction occur per catalytic site in unit time, were
calculated by following Eq. (8) and are listed in Table 4.
TOF = nCH COOC H /(npy ∙mcat ∙t∙NA)
(8)
3
5 11
In Eq. (8), npy is the number of Brønsted acid sites in mol kg−1
determined by desorption of pyridine (py) followed by FTIR, mcat is the
mass of catalyst (kg),
nCH COOC H molecules of isoamyl acetate were obtained and NA is
t
is the reaction time (7200 s) in which the
3
5 11
Avogadro’s number (6.022·1023 acid centers mol−1).
The TOFs are of great utility to compare activities of the same
Scheme 3. Global reaction including transesterification and ketalization.
6

P.F. Corregidor et al.
MolecularCatalysisxxx(xxxx)xxx–xxx
Table 3
(from tautomerization of vinyl alcohol). At that point, it is important to
notice that a simple model can be assumed by considering that a mo-
lecule of vinyl acetate must be accompanied by another one, but also
the same result must be anticipated in view of the accompaniment of
more than one molecule. In fact, molecules associated will further be
liberated and so, they will not participate directly in the reaction but
only supporting the adequate orientation for the interaction between
vinyl acetate and the Brønsted acid center.
Conditions for the study of the influence of concentration of vinyl acetate
(Experiment A) and isoamyl alcohol (Experiment B) in the analyzed reaction.
Effect of [Ester]
(Experiment A)
Effect of [Alcohol]
(Experiment B)
Mass of H-ZSM-5 catalyst (mg)
Concentration of vinyl acetate (mol/L)
110
0.100
1.000
0.100
0.100
On the other side, we assume that all chemical reactions are irre-
versible, thus the kinetic expressions of reaction rates (9) to (11) are
respectively:
Concentration of isoamyl alcohol
(mol/L)
0.100
0.200
0.300
Temperature (K)
Alcohol/Ester molar ratio
363
1:1
1:10
ρ1 = k1 [CH3COOC2H3]2 [H−☼]
ρ3 = k3 [CH3CHO][C5H11OH]2
1:1
2:1
3:1
(13)
ρ2 = k2 [CH3CO−☼][C5H11OH]
(14)
(15)
Table 4
Moreover, considering that there’s no accumulation of CH3CO-☼
and CH3CHO, they must be consumed at same rate as they were pro-
duced, that means: ρ1 = ρ2 and ρ1 = ρ3. Thus,
Turnover frequency values calculated at 7200 s for the influence of
alcohol/ester molar ratio in the reaction of vinyl acetate and iso-
amyl alcohol.
k1 [CH3COOC2H3]2 [H−☼] = k2 [CH3CO−☼][C5H11OH]
Alcohol/ester molar ratio
TOF (x10−4 s-1
)
(16)
1:1
2:1
3:1
1:10
8.3
k1 [CH3COOC2H3]2 [H−☼] = k3 [CH3CHO][C5H11OH]2
(17)
11.6
14.7
17.2
Expressions (16) and (17) can then be rearranged as (18) and (19),
[CH3CO−☼] = k1 [CH3COOC2H3]2 [H−☼]/(k2 [C5H11OH])
(18)
[CH3CHO] = k1 [CH3COOC2H3]2 [H−☼]/(k3 [C5H11OH]2)
(19)
Kinetic model of the reaction
Thus, the total balance for Brønsted acid sites can be represented by
Eq. (20), where [H−☼]total is a constant and equal to the total amount of
acid sites.
Considering that all of the Brønsted acid centers must be occupied
by vinyl acetate molecules, the proposed mechanism of reaction follows
an Eley-Rideal model. The adsorption of isoamyl alcohol over a cata-
lytic site was not considered necessary as only the acyl donor must be
adsorbed in order to generate the acyl zeolite complex. This specie will
further suffer a nucleophilic attack from isoamyl alcohol as described
earlier in this paper.
[H−☼]total = [H−☼] + [CH3CO−☼] + [CH3CHO]
(20)
Then, the number of free Brønsted acid sites can be calculated by
replacing expression (18) and (19) in (20), resulting in Eq. (21).
k1 [CH3COOC2H3]2
k2 [C5H11OH]
k1 [CH3COOC2H3]2
k3 [C5H11OH]2
[H−☼] = [H−☼]total /(1 +
+
)
The kinetic model that fits the experimental results is the one that
consider steps in Eqs. (9)–(11).
(21)
2 CH3COOC2H3 + ☼-H → CH3CO-☼ + CH3CHO + CH3COOC2H3 (9)
The rate for the production of isoamyl acetate is given by ρ2 and so,
v
is also equals to the reaction rate ( R),
CH3CO-☼ + C5H11OH → CH3COOC5H11 + ☼-H
CH3CHO + 2 C5H11OH → CH3C(OC5H11 2 + H2O
(10)
(11)
ρ2 = v = k2 [CH3CO−☼][C5H11OH] = k1 [CH3COOC2H3]2 [H−☼]total
R
)
k1 [CH3COOC2H3]2
k2 [C5H11OH]
k1 [CH3COOC2H3]2
k3 [C5H11OH]2
/(1 +
+
)
Where, CH3COOC2H3: vinyl acetate; ☼-H: Brønsted acid centers of
the catalyst; CH3CO-☼: acetylated zeolite complex; CH3CHO: acet-
aldehyde; C5H11OH: isoamyl alcohol; CH3COOC5H11: isoamyl acetate
and CH3C(OC5H11)2: acetaldehyde diisoamyl alcohol acetal. Thus, the
addition of elemental reactions (9)–(11) gives the global reaction (12).
(22)
When the reaction is carried out in excess of isoamyl alcohol (i.e.
[CH3COOC2H3]/[C5H11OH] = 1/10,
then
[CH3COOC2H3]2/
[C5H11OH]2 = 1/100), one can approximate Eqs. (22) to (23),
v
≅ k1 [CH3COOC2H3]2 [H−☼]total = k′ [CH3COOC2H3]2
(23)
R
CH3COOC2H3 + 3 C5H11OH → CH3COOC5H11 + CH3C(OC5H11 2
)
+
H2O
(12)
Expression (23) describes the experimental behavior observed when
studying the reaction in excess of isoamyl alcohol, which means, a
second order kinetics with respects of vinyl acetate concentration.
The first stage described by reaction (9) implicates the chemisorp-
tion of a vinyl acetate molecule to a Brønsted acid center on the cata-
lyst. As we will discuss later in this section, this assumption will allow a
reasonable description of experimental results as a second order reac-
tion is needed when considering an excess of one reactant, thus de-
manding the value of 2 for the stoichiometric coefficient of vinyl
acetate in step (9). The latter assumption suggests that not only one
molecule of vinyl acetate will arrive to the acid center but also a second
molecule will accompanied the first one, positioning the first molecule
for the right interaction with the acid center. Once a molecule of vinyl
acetate interacts with an active site in the catalyst, the other one will be
closed to the first electronically attracted. At the same time, the isoamyl
acetate molecule that interacts with a Brønsted acid center, quickly
dissociates to produce the acetylated zeolite complex and acetaldehyde
As
can
be
seen
from
Eq.
(22),
when
[CH3COOC2H3] > > [C5H11OH] and considering a similar procedure
as described previously, one can ′a′ rrive to expression (24) from (22).
v
≅ k3 [H−☼]total [C5H11OH]2 = k [C5H11OH]2
(24)
R
Consequently, the kinetic expression (24) describes a second order
reaction when the reaction is carried out in excess of vinyl acetate.
Conclusions
The reaction between vinyl acetate and isoamyl alcohol was studied
over H-ZSM-5 zeolite catalyst, prepared by hydrothermal treatment of
Expanded Perlite and as a greener route for the production of isoamyl
7

P.F. Corregidor et al.
MolecularCatalysisxxx(xxxx)xxx–xxx
acetate. The as-prepared ZSM-5 zeolite offered a Si/Al molar ratio of
38.5, in which all the Al has been introduced into the framework and
has exclusively Brønsted acids centers. The transesterification reaction
that generates isoamyl acetate competes with the ketalization of acet-
aldehyde and isoamyl alcohol producing acetaldehyde diisoamyl acetal
as a secondary product. The influence of reactants concentration de-
veloped a second order reaction with respect of the concentration of
one reactant, when operated in excess of the other. A conversion of 98%
was obtained by varying properly the concentration of reactants. A
kinetic model was obtained for the reaction carried out under excess of
one reactant. The kinetic model showed a second order kinetics which
correlated well with the proposed Eley–Rideal reaction mechanism and
also agree with the generation of an acyl zeolite intermediate complex
proposed by several authors. The surface acylated intermediate further
reacts under an acyl nucleophilic substitution to generates the desired
product, while the nucleophilic addition to carbonyl carbon of acet-
aldehyde produces an acetal as a secondary product.
[14] M.L.M. Bonati, R.W. Joyner, M. Stockenhuber, Microporous Mesoporous Mater. 104
(2007) 217–224.
[15] A. Gumidyala, T. Sooknoi, S. Crossley, J. Catal. 340 (2016) 76–84.
[16] P.F. Corregidor, D.E. Acosta, H.A. Destéfanis, Sci. Adv. Mater. 6 (2014) 1203–1214.
[17] K.S. Triantafyllidis, L. Nalbandian, P.N. Trikalitis, A.K. Ladavos,
T. Mavromoustakos, C.P. Nicolaides, Microporous Mesoporous Mater. 75 (2004)
89–100.
[18] C.A. Emeis, J. Catal. 141 (1993) 347–354.
[19] K.S.W. Sing, Pure and Applied Chemistry, (1985) pp. 603.
[20] S. Suganuma, K. Nakamura, A. Okuda, N. Katada, Molecular Catalysis 435 (2017)
110–117.
[21] H. Belarbi, Z. Lounis, R. Hamacha, A. Bengueddach, P. Trens, Colloids Surf. A:
Physicochem. Eng. Aspects 453 (2014) 86–93.
[22] P. Wang, B. Shen, J. Gao, Catal. Today 125 (2007) 155–162.
[23] P. Wang, B. Shen, D. Shen, T. Peng, J. Gao, Catal. Commun. 8 (2007) 1452–1456.
[24] D.P. Serrano, J. Aguado, Á. Peral, Controlling the Generation of Hierarchical
Porosity in ZSM-5 by Changing the Silanization Degree of Protozeolitic Units, in:
A. Gédéon, P. Massiani, F. Babonneau (Eds.), Studies in Surface Science and
Catalysis, Elsevier, 2008, pp. 123–128.
[25] G. Engelhardt, D. Michel, High-Resolution Solid-State Nmr of Silicates and Zeolites,
John Wiley & Sons, Australia, Limited, 1987.
[26] L. Tortet, E. Ligner, W. Blanluet, P. Noguez, C. Marichal, O. Schäf, J.-L. Paillaud,
Microporous Mesoporous Mater. 252 (2017) 188–196.
[27] R. Buzzoni, S. Bordiga, G. Ricchiardi, C. Lamberti, A. Zecchina, G. Bellussi,
Acknowledgments
Langmuir 12 (1996) 930–940.
[28] M. Castellà-Ventura, Y. Akacem, E. Kassab, J. Phys. Chem. C 112 (2008)
19045–19054.
The authors gratefully acknowledge the financial support received
as a grant from the Council of Researches of the Universidad Nacional
de Salta (C.I.U.N.Sa.) (Projects N° 2301/0 and 2327/0).
[29] K. Sadowska, K. Góra-Marek, J. Datka, Vib. Spectrosc. 63 (2012) 418–425.
[30] M.L.M. Bonati, R.W. Joyner, G.S. Paine, M. Stockenhuber, Adsorption Studies of
Acylation Reagents and Products on Zeolite Beta Catalysts, in: E. van Steen,
M. Claeys, L.H. Callanan (Eds.), Studies in Surface Science and Catalysis, Elsevier,
2004, pp. 2724–2730.
[31] M.L.M. Bonati, R.W. Joyner, M. Stockenhuber, Catal. Today 81 (2003) 653–658.
[32] P.F. Corregidor, Faculty of Ciencias Exactas, Universidad Nacional de Salta, Salta,
Argentina, 2017, pp. 315.
[33] E. Bożek-Winkler, J. Gmehling, Ind. Eng. Chem. Res. 45 (2006) 6648–6654.
[34] J.W. Moore, R.G. Pearson, Kinetics and Mechanism, Wiley, 1961.
[35] G. Raj, Meerut-India, Chemical Kinetics, 8th ed., Krishna Prakashan media Ltd,
2010.
References
[1] E. Fischer, A. Speier, Berichte der deutschen chemischen Gesellschaft 28 (1895), pp.
3252–3258.
[2] J. Otera, J. Nishikido, Esterification: Methods, Reactions, and Applications, Wiley,
2009.
[3] J. Otera, Chem. Rev. 93 (1993) 1449–1470.
[36] H. Konno, R. Ohnaka, J.-I. Nishimura, T. Tago, Y. Nakasaka, T. Masuda, Catal. Sci.
[4] M. Paravidino, U. Hanefeld, Green Chem. 13 (2011) 2651–2657.
[5] C.C. Akoh, L.N. Yee, J. Mol. Catal. B Enzym. 4 (1998) 149–153.
[6] G.D. Yadav, A.R. Yadav, Ind. Eng. Chem. Res. 52 (2013) 10627–10636.
[7] B.S. Balaji, B.M. Chanda, Tetrahedron 54 (1998) 13237–13252.
[8] M. Sasidharan, R. Kumar, J. Mol. Catal. A Chem. 210 (2004) 93–98.
[9] D. Srinivas, R. Srivastava, P. Ratnasamy, Catal. Today 96 (2004) 127–133.
[10] H. Ogawa, T. Fujigaki, H. Saito, Bulletin of Tokyo Gakugei University. Series IV,
Mathematics and Natural Sciences 56 (2004), pp. 53–56.
[11] S.E. Collins, S.R. Matkovic, A.L. Bonivardi, L.E. Briand, J. Phys. Chem. C 115 (2011)
700–709.
Technol. 4 (2014) 4265–4273.
[37] M. Lindén, F. Babonneau, H. Amenitsch, N. Baccile, A. Riley, S. Tolbert, On the
mechanism of formation of SBA-1 and SBA-3 as studied by in situ synchrotron XRD,
in: P.M. Antoine Gédéon, B. Florence (Eds.), Studies in Surface Science and
Catalysis, Elsevier, 2008, pp. 103–108.
[38] K. Suzuki, Y. Aoyagi, N. Katada, M. Choi, R. Ryoo, M. Niwa, Catal. Today 132
(2008) 38–45.
[39] J. Wu, H. Zhu, Z. Wu, Z. Qin, L. Yan, B. Du, W. Fan, J. Wang, Green Chem. 17
(2015) 2353–2357.
[40] H. Xin, X. Li, Y. Fang, X. Yi, W. Hu, Y. Chu, F. Zhang, A. Zheng, H. Zhang, X. Li, J.
[12] A. Corma, M. JoséCliment, H. García, J. Primo, Appl. Catal. 49 (1989) 109–123.
[13] O. Kresnawahjuesa, R.J. Gorte, D. White, J. Mol. Catal. A Chem. 208 (2004)
175–185.
Catal. 312 (2014) 204–215.
8