Preparation of acetates catalyzed by boric acid and/or tungstophosphoric acid-modified zirconia obtained employing polyethylene glycols as pore-forming agents

Article abstract of DOI:10.1016/j.molcata.2016.11.004

Zirconia modified with boric acid and/or tungstophosphoric acid calcined at 320 °C were prepared, characterized and used as catalysts in the production of acetates from diverse alcohols and phenols. Polyethylene glycol (PEG) of different molecular weight (400, 2000, 6000 Da) were added as low cost pore-forming agents during zirconia synthesis using zirconyl chloride as precursor and ammonium hydroxide as precipitating agent. The zirconias were impregnated with aqueous solutions of boric acid and/or tungstophosphoric acid (TPA). The borated zirconias, zirconias modified with TPA and zirconias doped with both boron and TPA were amorphous mesoporous materials with very strong acid sites, and specific surface areas S_BET of around 200, 100, and 150 m²/g, respectively. The FT-IR spectra of borated zirconias exhibited the bands of boron species, while the zirconias modified with TPA presented the characteristic bands of tungstophosphate anion, and the zirconias doped with both boron and TPA showed a degradation of the TPA anion, confirmed by ³¹P MAS-NMR. The borated zirconias and the zirconias modified with TPA gave excellent selectivity and yield in the 2-phenylethanol esterification with acetic acid. The use of these materials allows obtaining higher or similar results than those reported in the literature. Zirconias doped with both boron and TPA gave lower values, due to the transformation of the [PW₁₂O₄₀]^3? Keggin anion in to the [P₂W₂₁O₇₁]^6? and [PW₁₁O₃₉]^7? species. The reactivity towards acetylation with acetic acid of different alcohols and phenols using the best catalyst was ordered according to: primary alcohols > secondary alcohols > phenols. The reactivity difference of the alcohols and phenols was correlated with the electronic density on the oxygen atom and steric effects.

Full text of DOI:10.1016/j.molcata.2016.11.004

Accepted Manuscript
Title: Preparation of acetates catalyzed by boric acid and/or
tungstophosphoric acid-modified zirconia obtained employing
polyethylene glycols as pore-forming agents
Author: Lilian Osiglio Gabriel Sathicq Luis Pizzio Gustavo
Romanelli Mirta Blanco
PII:
S1381-1169(16)30460-5
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcata.2016.11.004
MOLCAA 10101
To appear in:
Journal of Molecular Catalysis A: Chemical
Received date:
Revised date:
Accepted date:
2-8-2016
11-10-2016
2-11-2016
Please cite this article as: Lilian Osiglio, Gabriel Sathicq, Luis Pizzio, Gustavo
Romanelli, Mirta Blanco, Preparation of acetates catalyzed by boric acid
and/or tungstophosphoric acid-modified zirconia obtained employing polyethylene
glycols as pore-forming agents, Journal of Molecular Catalysis A: Chemical
http://dx.doi.org/10.1016/j.molcata.2016.11.004
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. 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.

Preparation of acetates catalyzed by boric acid and/or tungstophosphoric acid-
modified zirconia obtained employing polyethylene glycols as pore-forming agents
Lilian Osiglio, Gabriel Sathicq, Luis Pizzio, Gustavo Romanelli, Mirta Blanco
Centro de Investigación y Desarrollo en Ciencias Aplicadas “Dr. J. J. Ronco”
(CINDECA), Departamento de Química, Facultad de Ciencias Exactas, UNLP-CCT La
Plata, CONICET, 47 N° 257, 1900 La Plata, Argentina
Corresponding author: lrpizzio@quimica.unlp.ar
GRAPHIICAL ABSTRACT

HIGHLIGHTS



Boric acid and/or tungstophosphoric acid-modified mesoporous zirconias were
prepared
PEG of different molecular weight were added to zirconia as pore-forming
agents
The new materials were tested as catalysts in acetylation of alcohols and
phenols


Yields were correlated with the electronic density on the O atom and steric
effects
Catalyst was re-used three times without losing activity.
ABSTRACT
Zirconia modified with boric acid and/or tungstophosphoric acid calcined at 320 ºC
were prepared, characterized and used as catalysts in the production of acetates from
diverse alcohols and phenols. Polyethylene glycol (PEG) of different molecular weight
(400, 2000, 6000 Da) were added as low cost pore-forming agents during zirconia
synthesis using zirconyl chloride as precursor and ammonium hydroxide as
precipitating agent. The zirconias were impregnated with aqueous solutions of boric
acid and/or tungstophosphoric acid (TPA). The borated zirconias, zirconias modified
with TPA and zirconias doped with both boron and TPA were amorphous mesoporous
materials with very strong acid sites, and specific surface areas SBET of around 200,
100, and 150 m2/g, respectively. The FT-IR spectra of borated zirconias exhibited the
bands of boron species, while the zirconias modified with TPA presented the
characteristic bands of tungstophosphate anion, and the zirconias doped with both
31
boron and TPA showed a degradation of the TPA anion, confirmed by P MAS-NMR.
The borated zirconias and the zirconias modified with TPA gave excellent selectivity
and yield in the 2-phenylethanol esterification with acetic acid. The use of these
materials allows obtaining higher or similar results than those reported in the literature.
Zirconias doped with both boron and TPA gave lower values, due to the transformation

of the [PW12O40]3− Keggin anion in to the the [P2W21O71]6− and [PW11O39]7− species. The
reactivity towards acetylation with acetic acid of different alcohols and phenols using
the best catalyst was ordered according to: primary alcohols  secondary alcohols 
phenols. The reactivity difference of the alcohols and phenols was correlated with the
electronic density on the oxygen atom and steric effects.
Keywords: Acetates, Aroma compounds, Boric acid and/or tungstophosphoric acid-
modified zirconia, Polyethylene glycols

1. Introduction
The preparation of organic esters through the esterification reaction between a
carboxylic acid and an alcohol is the most studied chemical transformation in organic
chemistry, due to the wide application of the esters that may be formed. Esters are
useful molecules because of their diverse applications, such as fragrances, flavours,
solvents, synthesis intermediates and drug products of interest. Furthermore, the
conversion of alcohols and phenols to esters is a significant synthetic transformation
that has received considerable attention [1] because it is an efficient method to protect
hydroxyl groups in multistep synthetic processes [2,3]. Among the various protecting
groups of the hydroxyl function, the acetyl group is the most appropriate in view of its
easy introduction and stability in acidic conditions [4,5].
Many esters are flavour and fragrance compounds, which are widely used in food,
cosmetic, beverage and pharmaceutical industries. In particular, the 2-phenylethyl
acetate has a pleasant rose aroma, and is widely used to add scent or flavour to
cosmetics, soaps, foods and drinks, giving the two latter a hint of fruit wine aroma [6].
Nowadays, the demand for esters has notably increased. They have usually been
obtained from plants, but some factors make synthetic production essential. Among
them, it can be mentioned that the vegetation may suffer the influence of weather or
diseases and, in addition, the quantities obtained by extraction from natural sources
are small [7].
The reaction involved to obtain esters is usually catalyzed by classical acids. There
are many updated procedures though, as reported by Shirini et al. [8], some of the
methods show varying degrees of success and limitations, such as long reaction time,
strong reaction conditions, the occurrence of side reactions, as well as low yields of the
desired product. Thus, the development of alternative clean procedures for the
acetylation reaction is highly needed. In this direction, our research group has
experience in the friendly ester synthesis using acetic acid as acetylating agent and an

alcohol or phenol with different structures. Two different catalyst families were used:
zirconium oxide doped with boron [9] and ammonium metatungstate-modified borated
zirconia [10], and in both cases the acetylation was strongly dependent on the alcohol
structure.
As is well known, the preparation of new acid solids is an important field in order to
replace the traditional acids, bearing in mind the need to achieve clean processes to
protect the environment. In this context and due to our experience in the preparation of
catalysts based on a special type of compounds, with strong acidic characteristics,
such as the heteropolycompounds with Keggin structure [e.g., 11-14], which were
tested as catalysts in different reactions [e.g., 15-20], including isoamyl acetate
production by esterification [e.g., 21,22], it was decided to add TPA to borated zirconia
in order to compare the results with samples prepared without boron addition.
Recently, an unsupported mixture of a Keggin heteropolyacid and boric acid in an ionic
liquid as solvent was used to obtain 5-hydroxymethylfurfural [23].
Moreover, as diverse pore-forming compounds, such as hydroxyacids or urea [24-
26], may be added to improve the textural properties of different oxides, in a previous
work glucose and polyethylene glycol of different molecular weight were employed as
low cost pore-forming agents, observing that polyethylene glycol may effectively be
used to obtain mesoporous zirconia with higher specific surface area and pore size
than without its addition [27].
In this context, the aim of the present study was to synthesize 2-phenylethyl acetate,
an essential aroma component for the food and cosmetic industries, by Fischer
catalyzed esterification between 2-phenylethanol and acetic acid, and then to study the
behaviour of different alcohols and phenols in catalyzed acetate production. It is known
that acetic acid is a cheap reagent and only forms water as a by-product, so it will be a
clean process. Diverse materials based on boric acid and/or tungstophosphoric acid-
modified zirconias were used as catalysts, using zirconias prepared with polyethylene
glycol of different molecular weight as pore-forming agents. The physicochemical

characteristics of the prepared catalysts and the electronic density on the oxygen
atoms in the alcohols and phenols were studied in order to correlate them with the
reaction results obtained.
2. Experimental
2.1. Sample preparation
Zirconia was prepared employing an aqueous solution of ZrOCl2.8H2O (Fluka) (0.56
mol Zr/dm3) as precursor, to which ammonium hydroxide 30% (Merck) was added
drop-wise as precipitating agent up to a final pH of 10, under constant stirring. The
effect of the addition of 10% (w/w) of polyethylene glycol (PEG) of molecular weights
400, 2000 or 6000 Da as pore-forming agents, which were added during the hydrogel
preparation, was studied. The obtained products were aged for 72 h and filtered under
vacuum. The precipitates were washed with distilled water to remove PEG and
chlorides, up to negative reaction with AgNO3 in the washings. The supports, dried at
room temperature (r.t.), were impregnated with H3BO3 (Anedra p.a.) aqueous solution
of 0.3 mol B/dm3 concentration, to obtain a boron concentration of 15 % (w/w) in the
solid, expressed as B2O3 with respect to the support. The samples to which PEG400,
PEG2000 and PEG6000 were added will be named ZrP4B, ZrP20B and ZrP60B,
respectively. After drying at r.t., the borated zirconias were impregnated with a
tungstophosphoric acid (TPA) solution of 0.04 mol W/dm3 concentration that leads to
15 %W (w/w) in the solid, thus obtaining the samples named ZrP4BTPA, ZrP20BTPA
and ZrP60BTPA. The boron and tungsten concentration in the samples were chosen
taking into account the results published in previous report [9,10]. The solids were dried
at r.t. and calcined in N2 atmosphere for 5 h at 320 ºC. Materials with TPA but without

boric acid addition calcined at the same temperature (ZrP4TPA, ZrP20TPA, ZrP60TPA
samples) were prepared for comparison purposes.
2.2. Sample characterization
The specific surface area, the pore volume and the pore size of the solids were
estimated from the adsorption-desorption isotherms of nitrogen at -196 ºC, using
Micromeritics equipment, model ASAP 2020, and previous degasification at 100 ºC for
1 h.
The X-ray diffraction (XRD) patterns were recorded between 5 and 60 °2 with Philips
PW-1732 equipment with a built-in recorder, using Cu K radiation, Ni filter, 20 mA and
40 kV in the high tension source, and a scanning rate of 1° per min.
The infrared (FT-IR) spectra were acquired employing Bruker IFS 66 equipment and
pellets of the sample in KBr. The measurements were performed in the 400-4000 cm-1
range.
For the analysis by 31P MAS–NMR, Bruker MSL-300 equipment with a sample holder
of 5 mm diameter and 10 mm in height was employed, using 5 s pulses, a repetition
time of 3 s, working at a frequency of 121.496 MHz for 31P at room temperature. The
spin rate was 2.1 kHz, and several hundred pulse responses were collected.
Phosphoric acid 85% was employed as external reference.
To obtain the micrographs of the solids, scanning electron microscopy (SEM) was
employed, using Philips equipment, model 505, at a working potential of 15 kV,
supporting the samples on graphite and metallizing with a sputtered gold film. The
images were obtained with ADDAII acquisition device, with Soft Imaging System, using
magnification between 300 and 5000x.
The materials were potentiometrically titrated in order to estimate their acidity. To this
end, 0.05 g of solid was suspended in 45 cm3 of acetonitrile. The suspension was
stirred for 3 h, and then titrated with a 0.025 N solution of n-butylamine in acetonitrile.

The potential variation was measured with Metrohm 794 Basic Titrino apparatus with a
double-junction electrode.
2.3. Catalytic evaluation
2.3.1. General
Chemicals were purchased from Fluka, Merck and Aldrich companies. The yields
refer to the isolated pure product. Products were characterized by IR and NMR
spectroscopy and mass spectrometry. Proton nuclear magnetic resonance (1H NMR)
experiments were recorded on Avance DPX 400, Bruker (400 MHz)
spectrophotometer. Carbon nuclear magnetic resonance (13C NMR) was recorded on
Avance DPX 400, Bruker (400 MHz) spectrometer with tetramethylsilane (TMS) as the
internal standard. All chemical shifts are quoted in parts per million (ppm) relative to
TMS using deuterated solvent. Coupling constants (J) are given in hertz (Hz). Signal
splitting is described as singlets (s), doublets (d), double doublet (dd), triplet (t),
quartets (q), and multiplets (m). Gas chromatography coupled with mass
spectrophotometer was done on HP 5971 instrument and mass recorded in EI mode.
Infrared (IR) spectra were acquired employing Bruker IFS 66 equipment and pellets of
the sample in KBr. Absorption maxima are quoted in wavenumbers (cm-1) and only
structurally significant peaks are listed.
The electronic density of the oxygen atoms in the alcohols and phenols was
determined after optimization of their conformation. The conformation of the
compounds was drawn by means of the “model build” modulus available in HyperChem
5.0. Each molecular structure was firstly preoptimized with the Molecular Mechanics
Force Field (MM+) procedure, and the resulting geometry was further refined by means
of the semi-empirical method PM3 (Parametric Method-3). A gradient norm limit of 0.01
kcal Å−1 was chosen.

2.3.2. Catalytic test
In all the reactions, a 25 cm3 round-bottomed flask, at the reaction temperature and
connected with a water jacketed condenser, was used. In a typical experience, the
reactor was loaded with a mixture of 1 mmol of 2-phenylethanol (122 mg), 5 mmol of
acetic acid (300 mg), 5 cm3 of toluene and 100 mg of the catalyst. The mixture was
stirred at 700 rpm. Samples were withdrawn from the organic phase at different
intervals (1, 3, 5, 7, 10 and 14 h). Each sample volume was approximately 10 mm3 and
was diluted with 2 cm3 of acetonitrile. The concentration of substrates was calculated
with the corresponding areas, using a Shimadzu HP2010 gas chromatography
instrument (Chromopack CP Sil 8 CB, 30 m x 0.32 mm ID). A similar procedure was
followed with the other studied alcohols and phenols.
After 14 h of reaction, the solution was concentrated under vacuum, and the residue
was purified by column chromatography employing silica gel as the stationary phase
and a mixture of hexane-ethyl acetate as the elution solvent. The reaction products
were identified by chromatographic comparison with standard samples. The conversion
was expressed as the ratio of moles of products to moles of initial alcohol.
2.3.4. Catalyst reuse
The stability tests were carried out running three consecutive experiments, under
the same reaction conditions. After each test, the catalyst was separated from the
reaction mixture by simple filtration, washed twice with 2 cm3 toluene, dried under
vacuum at room temperature for 4 h, and then reused.

3. Results and discussion
3.1. Catalyst characterization
The X-ray diffraction patterns of all the prepared solids presented a broad band with
maximum placed at 2 = 30º (Figure 1), indicating zirconia amorphous nature [28]. The
patterns of the samples modified with boron and/or TPA were similar to those of the
respective supports, with the absence of crystalline phases. The only exception is the
pattern of the ZrP60TPA sample, in which very low signals that may be assigned to
TPA are placed at 2 = 10º, 25º and 35º. Though the pattern of tungstophosphoric acid
depended on the hydration degree of the acid, these lines are characteristic of TPA
with a low hydration degree. Besides, it must be mentioned that peaks assigned to
boric acid or boric oxide were not present in the patterns of the samples. So, it can be
inferred that in almost all the samples the boron and TPA species are well dispersed on
the surface of the support as a non-crystalline phase, without forming the structure
needed to attain a diffraction pattern, or crystallites too small to give diffraction lines are
present. In addition, any of the different crystalline phases of zirconia were noticed,
although a stabilization effect can be induced by the presence of impurities, as was
reported by Yamaguchi [29,30].
Figure 1.
The specific surface area (SBET), determined by the Brunauer-Emmett-Teller
method, of a zirconia sample obtained without the addition of PEG during the synthesis
(Zr sample), were 146 m2/g. The microporous area (Smicro), estimated by the t-plot
method was almost negligible (12 m2/g) and its mean pore size (MPS), estimated by

the Barrett-Joyner-Halenda (BJH) method from the desorption branch of the isotherm,
was 2.1 nm. In the case of the zirconia samples (ZrP4, ZrP20, and ZrP60) obtained
adding PEG during the synthesis, a significantly increment of the MPV values was
revealed (3.0-3.1 nm). However, this increment does not depend on the polyethylene
glycol molecular weight. On the other hand, the specific surface area SBET remains
practically unchanged, slightly increases (~16%) or decrease (~20%) when PEG 400,
2000 or 6000, respectively, were used as pore forming agent. According to the Smicro
values, less than 10% of specific surface area is due to the presence of microporous.
So, these results indicate that the use of PEGs increases the mean pore size and also
reduces the presence of microporous. Additionally, higher specific surface area values
can be achieved using PEG 2000.
Both, SBET and Smicro, of the zirconias obtained by adding PEG during the synthesis
and modified with boric acid (ZrP4B, ZrP20B, and ZrP60B samples) are higher than
those of the corresponding bare zirconia (Zr sample) (Table 1). Besides, the total pore
volume (Vp) and the mean pore size, also increased. This fact may be explained
considering that boron species are well dispersed on the surface of the hydrated
zirconium oxide and inhibits particle growth during the calcination step, as was reported
for diverse oxides supported on zirconia [31].
On the other hand, the textural behaviour of the TPA modified zirconias (ZrP4TPA,
ZrP20TPA, and ZrP60TPA samples) was different to that observed for the borated
zirconias samples. The specific surface area and the pore volume are lower than those
of the corresponding supports, while the mean pore size is not modified. It can be
inferred that the tungstophosphate anions ([PW12O40]3−) are not as well dispersed as
boron species on the surface of the hydrated zirconium oxide, probably due to its
higher size (~1.2 nm) and shape (almost spherical). As result they did not inhibits
efficiently the zirconia particle growth during the calcination step.
Additionally, as result of the TPA addition to the ZrP4B, ZrP20B, and ZrP60B
samples (borated zirconias) both SBET and Vp values of ZrP4BTPA, ZrP20BTPA, and

ZrP60BTPA materials decreased, while MPS remains practically unchanged. The lower
SBET and Vp values can be due to the a partial pore blockage by the [PW12O40]3− anions.
However, these values are still higher than those obtained when zirconia was modified
only with TPA.

Table 1.
From the FT-IR spectra, it can be concluded that the washing of the prepared
supports appears to be successful, because the several characteristic bands of PEGs,
with maxima placed between 529 and 1467 cm-1, being the strongest that attributed to
the vibration of the C-O bond located at 1115 cm-1 [32], are not observed. However, an
undetectable and unimportant amount of the PEGs in the solids could not be
dismissed.
The FT-IR spectra of the solids obtained by impregnation of the zirconia with boron,
ZrP4B, ZrP20B, and ZrP60B samples, presented a wide band extended between 400
and 1750 cm-1 (Figure 2a), mostly attributed to vibrations of (O-Zr-O) bonds of zirconia
between 400 and 1150 cm-1 [33]. Some maxima are overlapped to the broad band of
zirconia. A band that appeared at 1623 cm-1 with a shoulder at 1383 cm-1 is ascribed to
the bending vibrations of the (H-O-H) and (O-H-O) bonds, respectively. In addition, the
characteristic bands of boron species, with the most intense peaks at 881 and 1190
cm-1, are less visible, which may be due to the overlapping.
With regard to the ZrP4TPA, ZrP20TPA and ZrP60TPA samples synthesized by
impregnation of the zirconias with TPA, the characteristic bands of bulk TPA are
observed, overlapped with the wide band of zirconia (Figure 2c). The FT-IR spectrum
of bulk H3PW12O40 showed bands at 1081, 982, 888, 793, 595, and 524 cm−1 (Figure
2c), assigned to the stretching vibrations of P–Oa, W–Od, W–Ob–W, W–Oc–W bonds,
and to the bending vibration of Oa–P–Oa bond, and they are in agreement with the
results reported in the literature for TPA [34]. It must be remembered that the anion of
TPA is composed by a central tetrahedral PO4 group surrounded by twelve WO6
groups and, in this context, the subscript a designates the oxygen that connects the W
addenda and the P heteroatom of the heteropolyanion, b and c denominates the corner
and the edge sharing oxygens of the twelve WO6 octahedra, and d is used to name the

terminal oxygens. The presence of the P–Oa, W–Od, W–Ob–W and W–Oc–W stretching
vibrations of [PW12O40]3− anion can be clearly observed.
In turn, the FT-IR spectra of the samples prepared by modifying the zirconias with
both boron and TPA, ZrP4BTPA, ZrP20BTPA and ZrP60BTPA samples, displayed
some special features (Figure 2b). Actually, the assignment of the observed bands is
difficult as they may be attributed to different species. The spectrum of the ZrP20BTPA
sample exhibited bands at 1080, 982, 895 and 806 cm−1, together with small bands at
1049 and 961 cm−1. The first two bands can undoubtedly be assigned to the
tungstophosphate anion, while the other bands could also be assigned to the lacunar
[PW11O39]7− anion, whose main bands are placed at 1100, 1046, 958, 904, 812, and
742 cm−1, though the presence of the [P2W21O71]6− anion, which displays bands at
1096, 1086, 1030, 983, 967, 933, 882, and 800 cm−1 [35], cannot be discarded.
With regard to the ZrP4BTPA and ZrP60BTPA samples, whose spectra present
bands with maxima at 1056, 961, 896 and 824 cm−1, and a shoulder at 1085 cm−1, the
presence of the lacunar [PW11O39]7− anion can be inferred; however, the widening of
the bands may be a result of the overlapping with the bands of the [P2W21O71]6− anion.
It must be borne in mind that the tungstophosphate anion can undergo a
transformation, following the scheme proposed by Pope [36], that is [PW12O40]3−

[P2W21O71]6−  [PW11O39]7−, attributed to the limited stability range of the [PW12O40]3−
anion with a pH increase.
Figure 2.
The 31P MAS-NMR spectra of the samples containing tungsten are shown in Figure
3. For the samples obtained by impregnating the zirconias with TPA, ZrP4TPA,
ZrP20TPA and ZrP60TPA samples, the main line is placed at −15.3 ppm, assigned to
the [PW12O40]3− anion, accompanied by a line at −13.7 ppm, which can be attributed to

the [P2W21O71]6− dimeric species [37]. Nevertheless, the line assigned to the
[PW12O40]3− anion is the most intense in all the samples, while the ratio between the
intensity of the line placed at −15.3 ppm and that of the band at −13.7 ppm decreased
in the order ZrP4TPA > ZrP60TPA > ZrP20TPA. The results are in agreement with
those obtained by FT-IR, although the presence of the [P2W21O71]6− species could not
be entirely confirmed by FT-IR.
On the other hand, the 31P MAS-NMR spectra of the samples prepared by
impregnation with both boron and TPA (ZrP4BTPA, ZrP20BTPA and ZrP60BTPA
samples) presented lines at −13.7 ppm and a signal at −11.2 ppm, which may be
assigned to the [P2W21O71]6− and [PW11O39]7− anion, respectively. The ZrP20BTPA
sample presented, in addition, a weak signal assigned to the tungstophosphate anion
at -15.0 ppm. The results are in agreement with those obtained by FT-IR.
In brief, the physicochemical characterization of the samples containing TPA
indicated that the tungstophosphate anion remains unchanged when it interacts directly
with the prepared zirconias during impregnation, while it partially or totally changes
when interacting with borated zirconias.
Figure 3.
The SEM micrographs of all the samples depicted in Figure 4 illustrated that
inhomogeneous-sized particles are present. The zirconias impregnated with boron
(ZrP4B, ZrP20B, and ZrP60B samples) showed a rather compact structure, which is
almost maintained when they are also impregnated with TPA (ZrP4BTPA, ZrP20BTPA
and ZrP60BTPA samples). Regarding the samples obtained by only impregnating TPA
on zirconia (ZrP4TPA, ZrP20TPA and ZrP60TPA samples), some small TPA
crystallites can be perceived on the support surface, though TPA addition barely
changed the structure of the parent support. The micrographs give some evidence that,

in most of the samples, a rather spongy structure is obtained when PEG2000 is
employed as pore-forming agent, though the samples presented a closer structure
when using PEG 400 or PEG6000.
It must be pointed out that the polymer PEG400 have short helices and, added
during the samples preparation, can aid to obtain materials with somewhat tight
structure. By contrast, it was reported that, for PEG2000, the PEG chains in the sol
form micelles, and the oxide particles enclose them, so aggregate type morphology is
achieved [38], which would be the cause of the slightly more spongy solids. In the case
of PEG6000, it was suggested that PEG chains in the sol tend to assemble in a
network structure, which may be attributed to interaction of the (-CH2CH2O-) groups of
PEG with hydroxyls of the hydrated zirconia, leading to a rather compact morphology.
Figure 4.
The samples were titrated with n-butylamine, method that allows one to make a
comparison of the strength of the acid sites which the solids contain, being the initial
electrode potential (Ei) an indication of the maximum strength of the sites. The
following scale is recommended to classify the acid sites strength, Ei > 100 mV (very
strong sites), 0 < Ei < 100 mV (strong sites), -100 < Ei < 0 mV (weak sites) and Ei < -
100 mV (very weak sites),.
All the borated zirconias exhibited very strong acid sites (Figure 5a), which would be
a consequence of the boron species well dispersed on the surface of the hydrated
zirconia, as observed by XRD. This can promote the acidity of the materials due to
Lewis acid sites of the zirconium atoms can be enhanced by an inductive effect and
Brønsted acid sites can be produced by hydration of the boron species [40]. The
obtained Ei values were 152, 184 and 205 mV for the ZrP4B, ZrP20B, and ZrP60B
samples, respectively. The lower Ei value of the sample obtained by impregnation of
boron solution on zirconia prepared using PEG400 may be related to the somewhat

more closed morphology of this material.
The samples containing TPA presented higher acidity than the borated zirconias.
The ZrP4TPA, ZrP20TPA and ZrP60TPA samples presented Ei values of 336, 311 and
309 mV, respectively, and the values corresponding to the ZrP4BTPA, ZrP20BTPA and
ZrP60BTPA samples were 295, 228 and 227 mV, respectively. This may be due to the
high acidic characteristics of the Keggin heteropolycompounds.
Figure 5.
3.2. Catalytic tests
The catalytic properties of the prepared samples were studied in the esterification
reaction of 2-phenylethanol with acetic acid (Scheme 1).
Scheme 1.
The catalytic behavior of all the synthesized materials, namely zirconias obtained
using PEG of different molecular weight as pore-forming agents and impregnated with
boric acid and/or TPA (ZrP4B, ZrP20B, ZrP60B; ZrP4TPA ZrP20TPA, ZrP60TPA;
ZrP4BTPA, ZrP20BTPA, and ZrP60BTPA samples), was compared. In all the tests, the
reaction conditions employed were selected based on previous results of our research
group [9,10], and the values are those indicated in Experimental.
Figure 6 shows the results, expressed as 2-phenylethanol conversion, obtained in
the esterification reaction of 2-phenylethanol with acetic acid using zirconia modified
with boric acid (ZrP4B, ZrP20B, and ZrP60B samples) as catalysts. Though at short
times the ZrP20B sample gave a lower conversion, it led to a slightly higher value from
7 h onwards; the similarity may be related to almost the same physicochemical

characteristics of the samples prepared with the zirconias obtained using PEG of
different molecular weight.
Figure 6.
On the other hand, Figure 7 shows the data obtained on the esterification reaction of
2-phenylethanol with acetic acid using as catalysts the zirconias modified by
impregnation with a TPA solution (ZrP4TPA, ZrP20TPA, and ZrP60TPA samples),
expressed as 2-phenylethanol conversion. The conversions attained were higher than
those corresponding to the zirconias doped only with boron. This fact is directly related
to the higher acid strength of these catalysts. The ZrP4TPA sample exhibited the
highest conversion; at 7 h the conversion was 96% and significantly rose up to 99%
after 14 h, without the presence of by-products. Evidently, the behavior of the ZrP4TPA
catalyst is in accordance with the fact that this sample is the most acidic material.
Figure 7.
The results of the conversion evolution in the esterification reaction of 2-
phenylethanol with acetic acid catalyzed by borated zirconias modified with TPA
(ZrP4BTPA, ZrP20BTPA, and ZrP60BTPA samples) are presented in Figure 8. Within
this group of catalysts, the ZrP20BTPA sample gave the highest conversion (58% at a
reaction time of 7 h), with an excellent selectivity. Nevertheless, the conversion results
obtained were lower than those of the other two families of catalysts when the zirconias
were doped with both boric acid and TPA. This fact must be related to the total or
partial change of the tungstophosphate anion into other species, as observed from FT-

IR and 31P MAS-NMR. It can be noted that, within this group of materials, the
ZrP20BTPA sample, which led to the highest conversion, is the only solid that showed
the presence of the undegraded tungstophosphate anion by both techniques.
Figure 8.
The catalyst that exhibited the best performance (ZrP4TPA sample) was used to
carry out a preparative synthesis of 2-phenylethyl acetate, an essential aroma
component for the food and cosmetic industries. Excellent selectivity and yield to pure
ester (98%) were obtained carrying out the reaction at 110 ºC using this most active
catalyst. The reusability of the catalyst was checked after the catalyst separation from
the reaction mixture, washing, and drying prior to its reuse. However, when the catalyst
was reused, a high yield decrease was observed, which could be attributed to a TPA
leaching from the solid. So, the borated zirconia that gave the best conversion (ZrP20B
sample) was chosen to be employed as catalyst, thus obtaining a yield to the pure
ester of 96% with excellent selectivity, as no by-product was observed, similarly to all
the reactions carried out. The reuse of this material in the reaction, following the
previous detailed procedure, was performed and, in this case, no significant loss in the
product yield was observed, the yield obtained being 95% and 94% for the second and
third cycles, respectively.
The yields obtained in this work using ZrP4TPA (Table 2, entry 1) are comparable to
those obtained by Barret, Habibi and Qiu (Table 2, entries 2-4), slightly higher than
those reported previously by our group (table 2, entry 5), and clearly superior to those
achieved by Barret in 2003 (table 2 entry 6)
Table 2.

Then, this solid was probed as catalyst in the esterification of different alcohols and
phenols with acetic acid, under the same experimental conditions, and the results of
yield to esters attained are listed in Table 3.
Table 3.
As a relationship between the sulfide oxidation easiness and the sulfur atom charge
obtained by a computational method was recently found by our research group, and
the result was associated with the nucleophilicity of the sulfur atom in the sulphide
substrate [41], in the present paper a correlation between the oxygen nucleophilicity
and the reactivity of the different tested alcohols and phenols was intended, using a
semi-empirical method. The relationship between the electronic density (De) of the
oxygen atoms, which was estimated by Hyperchem release 5, and the reaction yield to
acetylated alcohols and phenols was analyzed. The electronic density on the oxygen
atom of all the studied alcohols and phenols, which varied between -0.330 and -0.237,
is also presented in Table 3. From the results obtained, it could be observed that
alcohols (De = -0.330 to -0.326) are more reactive than phenols (De = -0.253 to -
0.237). The different reactivity between these two groups is associated with the
electronic density on the oxygen, since a higher electronic density on this atom helps in
the nucleophilic attack on the electrophilic carboxylic carbon in the acetic acid. In the
case of phenols, the diminution of electronic density is due to an electronic
delocalization by resonance in the benzene ring.
The reactivity difference between the diverse tested alcohols cannot be explained by
an electronic effect. This trend is reasonable since in all the studied alcohols the
charge on the oxygen atom (calculate by Hyperchem), which determines its
nucleophilicity, was very similar. The steric component affecting the alcohol reactivity is
also a decisive factor for Fischer acid-catalyzed esterification [9]. The steric hindrance

increases with molecular size, inducing electronic repulsion between non-bonded
atoms of the reacting molecules [42]. The reactivity of 2-phenylethanol, 2-
phenoxyethanol, and 1-hexanol (primary alcohols) and 4-methylbenzyl alcohol
(benzylic alcohol) (Table 3, entries 1-4, yield: 95%-85%) was higher than that of 1-
cyclohexanol (secondary alcohol) (Table 3, entry 5, yield: 65%)
The yield difference between primary and secondary alcohols can be attributed to
steric effects that considerably affect the acetylation rate. The presence of bulky
groups, not far from the reaction center, either in the alcohol or in the acid, slows down
the esterification rate, as the most sterically hindered alcohols give a lower yield.
Similarly, for the very bulky t-butanol (a tertiary alcohol) no reaction was detected under
similar reaction conditions (Table 3, entry 6).
On the other hand, the yield of phenol acetylation was lower than that of the primary
and secondary alcohols, reaching only values of 13% to 20% for different phenols
(Table 3, entries 7-9). The atomic charge on the oxygen atom was -0.253, -0.253, and -
0.248 for phenol, 4-methyl phenol and 4-chloro phenol, respectively. In these cases,
there is a relationship between the oxygen atom charge and the reaction yield. The 4-
nitrophenol (atomic charge on the oxygen atom: -0.237), a phenol with a strong
electron-withdrawing group, fails to give the expected acetylation product (Table 3,
entry 10).
In sum, the reactivity order towards acetylation with acetic acid using the studied
catalyst was as follows: primary alcohols  secondary alcohols  phenols.
4. Conclusions
Zirconias prepared with polyethylene glycol of different molecular weight as pore-
forming agents modified with boron or tungstophosphoric acid were very efficient and
selective catalysts for the esterification of 2-phenylethanol with acetic acid, while

samples impregnated with both boric acid and tungstophosphoric acid gave low yields.
The behavior of the latter samples is due to the total or partial change of the
tungstophosphate anion into other species. The catalyst obtained by impregnating
tungstophosphoric acid on zirconia prepared using a 400 Dalton polyethylene glycol as
pore-forming agent exhibited the best yield to pure ester (98%) with excellent selectivity
(100%), though it did not perform well during its reuse due to TPA leaching. However,
the catalyst obtained by impregnation of boric acid on zirconia prepared using a 2000
Dalton polyethylene glycol allowed overcoming this difficulty and could be reused
without a significant yield decrease.
The electronic density on the oxygen atom of all the studied alcohols (De = -0.330 to
-0.326) and phenols (De = -0.253 to -0.237), was estimated by Hyperchem release 5.
The difference in reactivity between alcohols and phenols in their esterification with
acetic acid was correlated with the electronic density of the oxygen atom. A higher
electronic density on this atom makes easier in the nucleophilic attack on the
electrophilic carboxylic carbon in the acetic acid.
The reactivity difference between the primary, secondary and tertiary tested alcohols
cannot be explained by an electronic effect since in all the studied alcohols the charge
on the oxygen atom, was very similar. In this case, the difference can be attributed to
steric effects that considerably affect the reactivity.
The use of this type of solid catalyst and acetic acid as acylating agent is a green
alternative to the classical procedures, because they lead to a cheap and clean
process, only giving water as by-product.
Acknowledgements
The authors thank G. Valle, M. Theiller and E. Soto for their experimental
contribution, and UNLP (X638) and CONICET (PIP 628) for the financial support.

References
[1] T.W. Green, P.J.M. Wuts, Protective Groups in Organic Synthesis, Wiley, New
York, 1999.
[2] P. Kumar, R. Pandey, M. Bodas, S. Dagade, M. Dongare, A. Ramaswamy, J. Mol.
Catal. A: Chem. 181 (2002) 207-213.
[3] T. Reddy, M. Narasimhulu, N. Suryakiran, Ch. Mahesh, K. Ashalatha, Y.
Venkateswarlu, Tetrahedron Lett. 47 (2006) 6825-6829.
[4] M.A. Longo, M.A. Sanromán, Food Technol. Biotechnol. 44 (2006) 335-353.
[5] L.Rajabi, R.Luque, Catal. Commun. 45 (2014) 129-132.
[6] E. Albertazzi, R. Cardillo, S. Servi, G. Zuchi, Biotechnol. Lett. 16 (1994) 491-496.
[7] H. Yan, Q. Zhang, Z. Wang, Catal. Commun. 45 (2014) 59-62.
[8] F. Shirini, M. Zolfigol, K. Mohammadi, Bull. Korean Chem. Soc. 25 (2004) 325-
327.
[9] L. Osiglio, G. Romanelli, M. Blanco, J. Molec. Catal. A: Chem. 316 (2010) 52-58.
[10] L. Osiglio, A. Sathicq, G. Romanelli, M. Blanco, J. Molec. Catal. A: Chem. 359
(2012) 97-103.
[11] L.R. Pizzio, C.V. Cáceres, M.N. Blanco, Appl. Catal. A: Gen. 167 (1998) 283-294.
[12] M.E. Chimienti, L.R. Pizzio, C.V. Cáceres, M.N. Blanco, Appl. Catal. A: Gen. 208
(2001) 7-19.
[13] V.D. Monopoli, L.R. Pizzio, M.N. Blanco, Mater. Chem. Phys. 108 (2008) 331-336.
[14] M.N. Blanco, L.R. Pizzio, Appl. Surf. Sci. 256 (2010) 3546-3553.
[15] D. P. Sawant, A. Vinu, Josena Justus, P. Srinivasu, S. B. Halligudi, J. Mol. Catal.
A: Gen 276 (2007) 150-157.
[16] D. P. Sawant, J. Justus, V. V. Balasubramanian, K. Ariga, P. Srinivasu, S.
Velmathi, S. B. Halligudi and A. Vinu, Chem. Europ. J., 14 (2008) 3200-3012.

[17] B. M. Devassy, S. B. Halligudi, S. G. Hegde, A. B. Halgeri, F. Lefebvre, Chem.
Commun., (2002) 1074-1075.
[18] E. Rafiee, N. Nobakht J. Mol. Catal. A: Chem. 398 (2015) 17-25.
[19] A. Patel, N. Narkhede, S. Singh, S. Pathan Catal. Rev. 58 (2016) 337-370.
[20] M. Gorsd, G. Sathicq, G. Romanelli, L. Pizzio, M. Blanco J. Mol. Catal. A: Chem.
420 (2016) 294-302.
[21]L. Pizzio, C. Cáceres, M. Blanco, Appl. Surf. Sci. 151 (1999), 91-101.
[22] L.R. Pizzio, M.N. Blanco, Appl. Catal. A: Gen. 255 (2003) 265-277.
[23] L. Hu, Y. Sun, L. Lin, S. Liu, Biomass Bioenergy 47 (2012) 289-294.
[24] Y. Wei, D. Jin, T. Ding, W. Shih, X. Liu, S. Cheng, Q. Fu, Adv. Mater. 10 (1998)
313-316.
[25] J. Zheng, J. Pang, K. Qiu, Y. Wei, Micropor. Mesopor. Mater. 49 (2001) 189-195.
[26] L. Pizzio, Mater. Lett. 59 (2005) 994-997.
[27] L. Osiglio, M. Blanco, Procedia Mater. Sci. 1 (2012) 491-498.
[28] B.L. Kirsch, S.H. Tolbert, Adv. Funct. Mater. 13 (2003) 281-288.
[29] J.C. Duchet, M.J. Tilliette, D. Cornet, Catal. Today 10 (1991) 507–520.
[30] T. Yamaguchi, Catal. Today 20 (1994) 199-217.
[31] B. Zhao, X. Xu, H. Ma, D. Sun, J. Gao, Catal. Lett. 45 (1997) 237-244.
[32] D. Zuo, Y. Xu, W. Xu, H. Zou, Chin. J. Polym. Sci. 26 (2008) 405-414.
[33] S. Vives, C. Guizard, L. Cot, C. Oberlin, J. Mater. Sci. 34 (1999) 3127-3135.
[34] C. Rocchiccioli-Deltcheff, R. Thouvenot, R. Franck, Spectrochim. Acta 32A (1976)
587-597.
[35] A.V. Ivanov, T.V. Vasina, V.D. Nissenbaum, L.M. Kustov, M.N. Timofeeva, J.I.
Houzvicka, Appl. Catal. A: Gen. 259 (2004) 65-72.
[36] M.T. Pope, Heteropoly and Isopoly Oxometalates, Springer-Verlag, Heidelberg,
1983, p. 58.
[37] R. Massart, R. Contant, J. Fruchart, J. Ciabrini, M. Fournier, Inorg. Chem. 16
(1977) 2916-2921.

[38] Y. Sun, Y. Wang, W. Guo, T. Wang, G. Luo, Micropor. Mesopor. Mater. 88 (2006)
31-37.
[39] Candelaria Leal Marchena, Silvina Gomez, Clara Saux, Liliana B. Pierella, Luis R.
Pizzio, Quim. Nova, 38 (2015) 518-525.
[40] D. Ravindra, Y. Nie, S. Jaenicke, G. Chuah, Catal. Today 96 (2004) 147-153.
[41] Romina Frenzel, Ángel G. Sathicq, Mirta N. Blanco, Gustavo P. Romanelli, Luis R.
Pizzio, J. Molec. Catal. A: Chem. 403 (2015) 27-36.
[42] Y. Liu, E. Lotero, J. G. Goodwin Jr., J. Catal. 243 (2006) 221-228.
Captions to the figures
Figure 1. XRD patterns of the samples obtained using PEG with different molecular
weight as pore-forming agents: a) borated zirconias, b) tungstophosphoric acid-
modified borated zirconias, c) tungstophosphoric acid-modified zirconias.
Figure 2. FT-IR spectra of the samples obtained using PEG with different molecular
weight as pore-forming agents: a) borated zirconias, b) tungstophosphoric acid-

modified borated zirconias, c) tungstophosphoric acid-modified zirconias.
Figure 3. 31P MAS-NMR spectra of the tungstophosphoric acid-modified borated
zirconias and the tungstophosphoric acid-modified zirconias obtained using PEG with
different molecular weight as pore-forming agents.
Figure 4. SEM micrographs of the samples obtained using different pore-forming
agents. Magnification: 1500x. Bar: 20 m.
Figure 5. Potentiometric titration curves of the samples obtained using PEG with
different molecular weight as pore-forming agents: a) borated zirconias, b)
tungstophosphoric acid-modified borated zirconias, c) tungstophosphoric acid-modified
zirconias.
Figure 6. Conversion obtained with boric acid-modified zirconias as catalysts in the 2-
phenylethanol esterification with acetic acid.
Figure 7. Conversion obtained with TPA-modified zirconias as catalysts in the 2-
phenylethanol esterification with acetic acid.
Figure 8. Conversion obtained with zirconias modified with both boric acid and TPA as
catalysts in the 2-phenylethanol esterification with acetic acid.

c
b
a
ZrP4TPA
ZrP4B
ZrP4BTPA
ZrP20B
ZrP60B
ZrP20TPA
ZrP60TPA
ZrP20BTPA
ZrP60BTPA
60
50
40
30
20
10
60
50
40
30
20
10
60
50
40
30
20
10
2
2
2
Figure 1

a
c
b
TPA
ZrB
ZrP4BTPA
ZrP4B
ZrP20B
ZrP4TPA
ZrP20BTPA
ZrP20TPA
ZrP60TPA
ZrP60B
B2O3
ZrP60BTPA
1800 1600 1400 1200 1000 800 600
1200
1000
800
600
1200
1000
800
600
400
Wavenumber (cm-1)
Figure 2

ZrP20TPA
-15.3
ZrP20BTPA ZrP60BTPA ZrP4TPA
-15.3
ZrP60TPA
-15.3
ZrP4BTPA
-13.7
-13.8
-13.7
-13.7
-13.7
-13.7
-11.2
-11.1
-11.5
-15.0
0
-10 -20
0
-10 -20
0
-10 -20
0
-10 -20
0
-10 -20
0
-10 -20

31P [ppm]
Figure 3