The role of m-ZrO2 in the selective oxidation of ethanol to acetic acid employing PdO/m-ZrO2

Article abstract of DOI:10.1016/j.molcata.2015.09.012

Catalysts based on Pd (0.6 wt.% of PdO) supported on ZrO₂ and α-Al₂O₃ (powder and pellets) were prepared by impregnation. Their performances were evaluated in the oxidation of ethanol to acetic acid employing a Berty reactor. The N₂ physisorption, H₂ chemisorption, cyclohexane dehydrogenation reaction, temperature programmed reduction, UV-vis diffuse reflectance spectroscopy, temperature programmed desorption of ethanol and optical microscopy were employed to characterize these catalysts. The PdO/Al₂O₃ catalyst was used as a reference. It was verified that PdO/m-ZrO₂ reaches high selectivity to acetic acid at high conversion. The PdO sites are the active ones for PdO/Al₂O_3, whereas in the case of PdO/m-ZrO₂ catalyst both PdO and m-ZrO₂ are in charge of this synthesis. The catalytic behavior of m-ZrO₂ can be associated with the adsorption and spillover of the oxygenate species from this oxide to PdO and vice-versa. Moreover, the oxidation of acetaldehyde (intermediate of the reaction) occurs on PdO and also on m-ZrO₂. The results show that m-ZrO₂ is more than a support. Indeed, this oxide works together with PdO generating a very active system for ethanol oxidation to acetic acid.

Full text of DOI:10.1016/j.molcata.2015.09.012

Journal of Molecular Catalysis A: Chemical 410 (2015) 177–183
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
The role of m-ZrO in the selective oxidation of ethanol to acetic acid
2
employing PdO/m-ZrO2
a
a
a
a
Sonia Letichevsky , Priscila C. Zonetti , Philipe P.P. Reis , Johnatan Celnik ,
b
a
a,∗
Carlos R.K. Rabello , Alexandre B. Gaspar , Lucia G. Appel
a
Laboratório de Catálise, Instituto Nacional de Tecnologia, Av. Venezuela 82/518, Rio de Janeiro, RJ 21081-312, Brazil
CENPES/Petrobras, Ilha do Fundão, Quadra 7, Cidade Universitária, 21949-900 Rio de Janeiro, RJ, Brazil
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 31 July 2015
Received in revised form
Catalysts based on Pd (0.6 wt.% of PdO) supported on ZrO2 and ␣-Al O (powder and pellets) were
2
3
prepared by impregnation. Their performances were evaluated in the oxidation of ethanol to acetic
acid employing a Berty reactor. The N2 physisorption, H2 chemisorption, cyclohexane dehydrogenation
reaction, temperature programmed reduction, UV–vis diffuse reflectance spectroscopy, temperature pro-
grammed desorption of ethanol and optical microscopy were employed to characterize these catalysts.
The PdO/Al2O3 catalyst was used as a reference. It was verified that PdO/m-ZrO2 reaches high selectivity
to acetic acid at high conversion. The PdO sites are the active ones for PdO/Al2O3, whereas in the case
of PdO/m-ZrO2 catalyst both PdO and m-ZrO2 are in charge of this synthesis. The catalytic behavior of
m-ZrO2 can be associated with the adsorption and spillover of the oxygenate species from this oxide to
PdO and vice-versa. Moreover, the oxidation of acetaldehyde (intermediate of the reaction) occurs on
PdO and also on m-ZrO2. The results show that m-ZrO2 is more than a support. Indeed, this oxide works
together with PdO generating a very active system for ethanol oxidation to acetic acid.
1
5 September 2015
Accepted 18 September 2015
Available online 25 September 2015
Keywords:
Acetic acid
Zirconia
Alumina
Palladium
Ethanol
©
2015 Elsevier B.V. All rights reserved.
1
. Introduction
on ZrO2 monoclinic (m-ZrO ) is a very active, selective and stable
2
catalyst for the ethanol selective oxidation to acetic acid.
Acetic acid is one of the most relevant commodities of the chem-
Zirconia (ZrO ) monoclinic, which has been used as support of
2
ical industry. It is used in the production of acetic anhydride, vinyl
acetate monomer (VAM), terephthalic acid (PTA) and acetate esters.
These chemicals are employed in the manufacture of plastics, deter-
gents, paints, rubbers, solvents, textile fibers and pharmaceutical
products.
At present, ethanol can be considered as not only a promising
renewable raw material for the Chemical Industry, but also a very
special platform molecule. It can generate many different chem-
icals, including ethyl acetate [1–4], 1-butanol [5,6], acetone [7],
propene [8], isobutene [9], 1,3 butadiene [10] in a one-pot reaction
employing multifunctional catalysts.
Pd and other metals, shows a very relevant role in several reactions
due to its basicity, acidity, superficial vacancies and oxygen mobil-
ity. Below, some examples of m-ZrO2 catalytic performances are
exhibited.
The Ni/m-ZrO2 catalyst is a very promising system for the
elimination of acetaldehyde from H2 rich streams generated
from ethanol steam reforming. Silva et al. [18] employing the
TPSR of acetaldehyde decomposition, the thermodesorption of
acetaldehyde followed by FTIR and catalytic tests verified that the
performance of Ni/m-ZrO2 is not only related to its high activity,
but also to the large number of active sites located on the metal
and mainly on the support.
Nowadays, acetic acid is mainly produced by the carbonylation
of methanol. However, it is possible to generate acetic acid from
ethanol using catalysts based on Au, Ru, Mo, V and Pd [11–17].
Recently, our group has found that the PdO catalyst supported
Physical mixtures comprising m-ZrO2 and Cu/ZnO/Al O3 cata-
2
lysts were employed in the ethyl acetate synthesis from ethanol.
According to the authors, acetaldehyde generation occurs by
ethanol dehydrogenation on Cu. Afterwards, this aldehyde migrates
to m-ZrO2 and reacts with ethoxide species adsorbed on this oxide
generating hemiacetal, which is then dehydrogenated to ethyl
acetate [3]. Employing the TPD of acetaldehyde and ethanol, these
authors proposed that the m-ZrO2 performance in the ethyl acetate
∗ Corresponding author. Fax: +55 21 21231165.
E-mail address: lucia.appel@int.gov.br (L.G. Appel).
http://dx.doi.org/10.1016/j.molcata.2015.09.012
1
381-1169/© 2015 Elsevier B.V. All rights reserved.

1
78
S. Letichevsky et al. / Journal of Molecular Catalysis A: Chemical 410 (2015) 177–183
synthesis is related to the interaction of this oxide with acetalde-
hyde (spillover) and also to the ethoxide species generation, which
in turn is associated with the m-ZrO2 high density of strong basic
sites.
The measurements of samples surface area (powder catalysts)
were carried out using an ASAP 2020 Micromeritics unit. The sam-
ples were dried at 150 C for 24 h, and then submitted to an in situ
◦
◦
treatment under vacuum at 150 C for 2 h. The catalysts were cooled
◦
One-pot ethyl acetate synthesis was also studied using physical
and the N2 adsorption was carried out at −196 C.
mixtures of PdO/m-ZrO2 catalyst with m-ZrO . It was verified that
The UV–vis diffuse reflectance spectra (powder catalysts) in
the range of 190–800 nm were recorded using a Varian Cary 500
spectrophotometer equipped with a Harrick diffuse reflectance
accessory with Praying Mantis geometry. The supports were used
as references.
2
acetaldehyde is generated by the ethanol oxidation on PdO catalyst.
Afterwards, it spillovers toward m-ZrO2 where the condensation
reaction between this aldehyde and the ethoxide species occurs
[
2].
The acetone synthesis from ethanol was described employing
Cu/ZnO/Al O and m-ZrO2 physical mixture and in situ DRIFTS
The TPR analyses of the catalysts (powder) were performed in an
AutoChem II 2920 Micromeritics unit. Initially, the catalysts were
2
3
◦
−1
experiments. Rodrigues et al. [7] proposed the following mecha-
nism: firstly, ethanol is adsorbed on m-ZrO2 and Cu, generating
treated at 400 C for 60 min under synthetic air flow (30 ml min ).
The analyses were performed using an amount of catalyst corre-
−
1
◦
ethoxides species. These species adsorbed on m-ZrO spillover to Cu
sponding to 6 mg of Pd, 10% H /N (50 ml min ), from −50 C to
2
2
2
◦
◦
−1
◦
catalyst, and then acetaldehyde is synthesized by dehydrogenation.
250 C, at 10 C min , and then remaining at 250 C for 60 min.
The TPD of ethanol analyses were performed using a micro
reactor system coupled to a QMS200 Balzers mass quadrupole spec-
trometer employing 500 mg of catalyst (powder). The samples were
After that, this aldehyde spillovers from Cu/ZnO/Al O3 to m-ZrO2
2
and acetates species are formed by the interaction between the
acetaldehyde and the oxygen of the m-ZrO surface. Finally, acetone
2
◦ −1
pretreated at 150 C, 5% O /He (40 ml min ) for 30 min. Then, they
2
is generated by the ketonization of the acetates species on m-ZrO2.
This condensation occurs assisted by strong basic sites of m-ZrO2.
This work shows that m-ZrO2 is able to oxidize acetaldehyde to
acetate species and promote the ketonization reaction [7].
◦
−1
◦
−1
were heated up to 400 C in 5% O /He (40 ml min , 10 C min ) for
2
1 h. The ethanol adsorption was carried out at room temperature
◦
for 1 h. Ethanol adsorption was accomplished at 30 C (saturator)
The Pd based catalyst supported on m-ZrO2 is very active in the
methane combustion. Ciaparu et al. [19,20] employing temperature
programmed isotopic exchange of oxygen experiments suggested
that this reaction occurs by Mars and Van Krevelen mechanism.
They also proposed that the hydroxyls generated from the water of
the methane oxidation hinder the reoxidation of the PdO surface
using He as the carrier gas. The TPD measurements were per-
◦
−1
◦
formed by heating the samples at 20 C min up to 500 C, under
−
1
He flow (40 ml min ), and then remaining at this temperature
for 30 min. The following fragments, m/z = 2 (H ), m/z = 16 (CH ),
2
4
m/z = 27 (C H ), m/z = 28 (CO), m/z = 29 (C H O), m/z = 31 (C H O)
2
4
2
4
2
6
and m/z = 44 (CO ), related to products and byproducts of the reac-
2
by air (O ). Then, the O from the m-ZrO lattice reoxidizes PdO. This
tion were continuously monitored by the QMS200 Balzers mass
quadrupole spectrometer. The intensities of these fragments were
mathematically treated in order to eliminate contributions of more
than one species.
2
2
O migration from m-ZrO toward PdO also avoids the nucleation of
2
the reduced Pd. This phenomenon occurs at low temperatures such
◦
as 430 C [19,20].
Taking the catalytic behavior of m-ZrO2 described above into
account, it can be foreseen that this oxide might also show a major
role in the synthesis of acetic acid from ethanol employing PdO/m-
Cyclohexane dehydrogenation rates of the prepared catalysts
(powder) were employed as an indirect measurement of the Pd
metallic surface area [22]. This reaction was carried out at atmo-
spheric pressure using a conventional fixed-bed micro reactor
monitored by on-line gas chromatography. The H /C H ratio was
ZrO . Therefore, the aim of this work is to study the role of m-ZrO2
2
in the selective oxidation of ethanol to acetic acid, contributing to
improve the description of the catalytic behavior of this oxide.
2
6
12
◦
13. The cyclohexane vapors were generated in a saturator (12 C)
−
◦
1
using H as the carrier gas (90 ml min ). The samples were reduced
2
−
1
◦
−1
with H2 flow (30 ml min ) at 260 C (10 C min ) for 1 h. The
dehydrogenation of cyclohexane rates were measured under differ-
2
. Methodology
◦
ential conditions at 260 C. An Agilent HP6890N with FID and TCD
2
.1. Catalysts synthesis
detectors was used to determine the reactant and product com-
positions. Very low amount of cyclohexene was generated being
benzene the main product.
The catalysts were prepared as “eggshell pellets” [21] by wet
impregnation using Pd(NO3)2 and commercial supports, ␣-Al O3
and m-ZrO2, supplied by Norpro, which were employed as pel-
The measurements of the metallic surface area of powder and
pellets samples were performed in an ASAP 2010 Micromeritics
2
◦
lets. The shape of the ␣-Al O3 pellets was cylindrical with around
unit. The samples were dried at 150 C for 24 h, and then were
2
−
1
◦
◦
−1
0
.40 cm of length and 0.20 cm of diameter. The shape of the
m-ZrO2 pellets was also cylindrical with around 0.44 cm and
.26 cm of length and diameter, respectively. The samples were
reduced in situ with H2 flow (30 ml min ) at 260 C (10 C min )
for 1 h. After that, they were submitted to an in situ treatment under
◦
0
vacuum. The H2 adsorption was carried out at 75 C. The isotherms
−1
◦
−1
◦
calcined under air (60 mL min ) at 0.5 C min until 250 C and
1
were obtained employing the following pressures: 40, 100, 200,
250, 300, 350 Torr for 2 h each. The equation below was used for
determining the metallic surface area (Sm):
◦
−1
◦
0 C min until 400 C for 10 h.
Two powder catalysts, PdO/Al O3 and PdO/ZrO2 were prepared
2
by incipient wetness impregnation employing the same Pd precur-
sor and commercial supports mentioned above. These oxides were
grinded to powder (d < 0.074 nm) before deposition of the metal.
ꢀ
ꢁ
Vs
23
Sm =
× F × 6.123 × 10 × S
W × 22414
−
1
◦
−1
The samples were calcined under air (60 ml min ) at 0.5 C min
◦
◦
−1
◦
Vs = volume sorbed (cm3 at STP), W = sample weight (g),
F = stoichiometry factor (2, dissociative), S = surface area of one Pd
until 250 C and 10 C min until 400 C for 10 h.
−
16
2
atom (7.87 × 10
cm ).
2.2. Characterization
Optical microscopy analysis was performed using a Confocal
Axio CSM 700—Zeiss. The cylinders were cut into small disks. The
thickness measurements of the PdO layer were obtained employing
the software ImageJ.
The composition of each sample (powder and pellets catalysts)
was determined by X-ray fluorescence (XRF) on a S8 Tiger (Bruker).

S. Letichevsky et al. / Journal of Molecular Catalysis A: Chemical 410 (2015) 177–183
179
Fig. 1. (A) Micrographs of the cross sections of the prepared PdO/Al2O3 pellet. The black layer is related to PdO. (B) Micrographs of the cross sections of the prepared PdO/ZrO2
pellet. The black layer is related to PdO.
The X-ray diffraction (XRD) was performed using a Rigaku Mini-
prepared catalysts (pellets). The results showed that this reactor
was able to generate reliable data which can be used for the descrip-
tion of the performance of the prepared catalysts.
flex instrument with Cu␣ radiation (ꢀ = 1.542 Å) operated at 40 kV
◦
◦
and 30 mA. The angular range varied from 10 to 90 , with incre-
ments of 0.02 and 10 s per step.
◦
Therefore, Pd based catalysts were prepared employing pellets
of the same oxides used in the case of the powder catalysts. In
order to avoid internal diffusion issues, the preparation was con-
ducted in a way to coat the external surface of the pellets generating
eggshell pellets [21]. Fig. 1A and B shows the disks generated from
the PdO/Al2O3 and PdO/ZrO2 pellets (cylinders), respectively. The
PdO (black layer) is only observed in the outermost layer of both
pellets. The PdO layer thicknesses of the PdO/ZrO2 and PdO/Al2O3
pellets are around 111 and 81 ␮m, respectively.
Many characterization techniques normally employed for pow-
der catalysts are not appropriate for pellets. The preferential paths
which might appear when using micro reactors are the main rea-
son. Moreover, the modification of active phases may occur when
these pellets are ground.
Thus, for better understanding the catalytic performances of the
pellets some characterization data of the powder catalysts (TPD of
ethanol, UV–vis spectroscopy, TPR and XRD) should also be consid-
ered.
Table 1 depicts the palladium oxide content (wt.%) of the pre-
pared catalysts (powder and pellets). This table also shows that the
surface area of PdO/ZrO2 catalyst (powder) is ten times larger than
the one of the PdO/Al2O3 catalyst. This occurs due to the specific
surface area values of the supports.
2
.3. The oxidation of ethanol
The catalytic tests were carried out using an Autoclave Engi-
neers Micro Berty Reactor. This reactor can be viewed as a small
packed bed” one with an extremely high recycle ratio. Thus, it
“
behaves like a continuously stirred tank reactor (CSTR) [23]. This
reactor avoids external diffusion issues. The experimental con-
ditions, i.e., gas stream composition, stirring speed of the Berty
reactor, temperatures, pressure, flow rate and mass of the cata-
lyst were N :O :C H5OH = 77.6:19.4:3.0 v/v%, 1500 rpm, 140 and
2
2
2
◦ −1
1
75 C, 4 bar, 34 ml min (STP) and 200 mg, respectively. Ethanol
was admitted into the reactor by a Gilson HPLC pump coupled to
◦
a vaporizer (150 C). Synthetic air was used as the carrier gas. The
−
1
◦
catalysts were pretreated with synthetic air (20 ml min ) at 200 C
for 2 h. The ethanol consumption rates were calculated using the
mass balance of a CSTR. The rates of acetic acid, acetaldehyde or
carbon dioxide formation were obtained multiplying the ethanol
consumption rates by the selectivity to these products. Conversion
and selectivities values were obtained after 15 h (time on stream).
No deactivation was observed. The products were analyzed on-
line with a GC (Agilent HP6890) equipped with TCD and FID and
−
1
a Hayesep Q column using He as the carrier gas (15.7 ml min ).
On the one hand, ␣-Al2O3 surface does not show acid neither
◦
◦
−1
2
−1
The oven was programmed from 120 to 200 C at 5 C min . All
the compounds observed were calibrated. The retention times
of methane, ethane, acetaldehyde, ethanol, acetic acid and ethyl
acetate were 0.70, 1.55, 5.65, 8.29, 15.19 and 16.84 min, respec-
tively.
basic sites. Its surface is lower than 10 m g . From the catalytic
point of view, it is an inert oxide. On the other hand, m-ZrO2 (mon-
oclinic) is a mesoporous oxide that shows basic and acid sites [3].
2
−1
Its specific surface area is around 100 m g , and it shows surface
anionic vacancies. It can be foreseen that these physicochemical
properties might influence the interaction of the Pd precursor with
these oxides and also, the behavior of these catalysts in the ethanol
oxidation.
3
. Results and discussion
Some ethanol oxidation experiments with PdO/ZrO2, PdO/Al O3
catalysts (both in powder) were carried out using a fixed bed
Fig. 2 displays the UV–vis spectra of the PdO/ZrO2 and PdO/Al2O3
powder catalysts. Very broad peaks can be observed with maxima
2
reactor. The former shows high acetic acid yield (71%) at the follow-
◦
ing conditions: N :O :C H5OH = 77.6:19.4:3.0 v/v%, 175 C, 25 mg,
2
2
2
Table 1
−
1
2
0 ml min . The supports Al O3 and ZrO2 were also evaluated.
Alumina does not react and the latter shows very low activity gen-
erating tiny amounts of acetaldehyde.
When performing diffusion tests [24], it was verified that due to
the high catalytic activity of the catalysts the influence of the exter-
nal diffusion was very relevant under these conditions. Thus, a Berty
reactor was employed in order to avoid diffusion problems and
properly describe the catalytic behavior. External diffusion tests
were carried out once again employing a Berty reactor and the
PdO loading (L), metallic surface area (Sm), cyclohexane dehydrogenation rate (R),
specific surface area (S) and amount of ethanol adsorbed on the catalysts (A).
2
Catalysts
Shape
PdO/ZrO2
Powder
PdO/Al2O3
Powder
PdO/ZrO2
Pellet
PdO/Al2O3
Pellet
L/wt.%
Sm/m gcat
0.6
0.35
0.074
95
0.6
0.08
0.020
9
0.6
0.28
0.6
0.08
2
−1
−
1
−1
h
R/molCH gcat
2
−1
S/m
g
−
1
A/␮molethanol gcat
2803
9

1
80
S. Letichevsky et al. / Journal of Molecular Catalysis A: Chemical 410 (2015) 177–183
3
−1
) value when PdO is considered
the theoretical (210.5 cm g
Pd
the only reducible species of this catalyst, according to the UV–vis
results. Indeed, the TPR profile of Al O does not show H consump-
2
3
2
tion.
The TPR profile of ZrO2 (not shown) at the temperature range
displayed in Fig. 3 depicts a very small amount of H2 consump-
PdO/ -Al O
tion (tiny broad peak). The TPR profile of PdO/ZrO shows two
2
3
2
◦
3
−1
peaks, one at 8 C (79.7 cm g
) and another one at higher
−1
). Considering the reduction profile
Pd
3
temperature (248.7 cm g
Pd
of PdO/Al O3 and the UV––vis results, the peak at low tempera-
2
ture can be associated with PdO and the one at high temperature
◦
(
112 C) with the dispersed Pd species. The H consumption of
2
3
−1
PdO/ZrO2 (328.4 cm g
value (210.5 cm g
) is much higher than the theoretical
−1
). Hence, the reduction (TPR) of PdO/ZrO2
Pd
PdO/m-ZrO2
3
Pd
was repeated and continuously monitored employing a MS detec-
◦
tor. The generation of CO and CO2 around 110 C was observed.
4
00
600
800
◦
Thus, the reduction at 112 C can be attributed not only to dis-
/
nm
persed Pd species but also to the reduction of carbonate species.
Indeed, the generation of carbonate species on the ZrO2 surface is
well described in the literature [28]. It occurs due to the presence
of basic sites that show strong and medium strength on the surface
Fig. 2. UV spectra of the powder prepared catalysts.
of this oxide, which interact with CO from the air. These carbonate
2
species generate an increase of H consumption not only as a result
2
of their reduction but also due to the conductivity changes related
to the presence of CO and CO2 in the gas mixture that reached the
TCD detector. Thus, the UV–vis spectroscopy and the TPR experi-
ments suggest the presence of PdO and dispersed Pd species on the
ZrO2 surface.
PdO/ -AlO
2
3
Table 1 shows the cyclohexane dehydrogenation rates (R) of the
powder prepared samples. These data are directly proportional to
the amount of metallic Pd on the catalytic surface [22]. As the reduc-
PdO/ZrO2
◦
tion occurs at very low temperature (260 C), it can be suggested
that the amount of Pd on the surface does not change when the cat-
alysts are oxidized. Thus, the cyclohexane dehydrogenation rates
(R) are directly proportional to the amount of Pd on the catalytic
surface. The ratio between the cyclohexane dehydrogenation rates
of PdO/ZrO2 and PdO/Al O3 is around 3.7. As these catalysts con-
tain the same amount of Pd, this result can be associated with the
0
100
200
2
o
temperature/ C
presence of dispersed Pd species on PdO/ZrO . Moreover, taking
2
Fig. 3. TPR of the powder prepared catalysts.
into account the values of the specific surface area, the same Pd
concentration of both catalysts, the cyclohexane dehydrogenation
rates (Table 1) and the physicochemical properties of ␣-Al O , it
2
3
around 364 and 353 nm, respectively. According to Boukha et al.
25], the absorptions around 360 nm can be attributed to the d–d
transition band of Pd from PdO. Both spectra also show shoul-
ders at around 450 nm, which can be ascribed to PdO2 or large
PdO particles [26]. The PdO/ZrO2 catalyst also depicts a band at
can be proposed that the average PdO particle size of PdO/Al O3 is
2
[
much larger than the one of PdO/ZrO2.
The ratio of the metallic surface area (Sm) between PdO/ZrO2
and PdO/Al O3 (powder) measured by H2 chemisorption is around
2
4
.4 (Table 1), which is similar to the one observed by the cyclo-
2
50 nm, which might be assigned to dispersed Pd species having
hexane dehydrogenation rates measurements. Again, it is possible
strong interactions with the support. This absorption is related to
the charge transfer from the oxygen of the support to the d orbitals
of Pd [25–27].
The XRD measurements of the prepared catalysts were
carried out. The reflections observed are related to the sup-
to suggest that the PdO particle size of the Al O3 catalyst is much
2
higher than the one of PdO/ZrO2.
It was verified that the specific surface area and the Pd con-
centration of the pellets samples are similar to the powder ones.
Moreover, the metallic surface area of the pellets catalysts was also
measured by H2 chemisorption (Table 1). The powder and pellet
ports, i.e., ␣-Al O and m-ZrO2 (monoclinic). Palladium species
2
3
were not observed due to their low concentration in the
catalysts.
samples of Pd/Al O3 catalysts show the same value, whereas for
2
the ZrO2 catalysts, there is a small difference between their values.
Therefore, it can be inferred that the shape of the catalyst seems
not to modify significantly the Pd dispersion.
The TPR profiles of the prepared catalysts (powder) are depicted
in Fig. 3. The PdO/Al O catalyst profile shows only one peak at
2
3
◦
3
−1
) matches
Pd
6
C. The H2 consumption of PdO/Al O3 (210.2 cm g
2
Table 2
−
1
−1
Ethanol consumption rate, acetic acid, acetaldehyde and carbon dioxide synthesis rates (␮molgcat min ) and also selectivities to acetic acid (SAA) and carbon dioxide (SCD)
of the catalysts PdO/Al2O3 and PdO/ZrO2 at 175 C (pellets).
◦
Catalyst
Ethanol consumption rate
Acetic acid synthesis rate
158
Acetaldehyde synthesis rate
Carbon dioxide synthesis rate
22
SAA %
SCD %
11
PdO/ZrO2
PdO/Al2O3
200
137
10
32
79
53
73
9
7

S. Letichevsky et al. / Journal of Molecular Catalysis A: Chemical 410 (2015) 177–183
181
A
B
-7
6.0 X10
100
100
-5
90
H O
2.0X10
2
H O
2
acetic acid
acetaldehyde
CO2
ethyl acetate
methane
conversion
80
70
60
50
40
80
60
40
20
0
CO
-
7
4.0 X10
CH4
CO2
30
CO2
20
-5
1.0X10
CH COOH
10
3
0
-
7
1
40
175
140
175
2.0 X10
CH CHO
temperature / ºC
temperature / ºC
3
CH CHO
3
Fig. 4. (A) Ethanol conversion and selectivities of the prepared PdO/Al2O3 pellet. (B)
Ethanol conversion and selectivities of the prepared PdO/ZrO2 pellet. The WHSV,
−
1
−1
−1
,
the flow rate and the gas mixture composition are 10 l h gcat , 34 ml min
N2:O2:C2H5OH = 77.6:19.4:3.0 v/v%, respectively.
C H OH
2
5
C H OH
2
5
0
.0
Table 2 exhibits the ethanol consumption rates and also the
acetaldehyde, acetic acid and carbon dioxide generation rates
employing a Berty reactor at 175 C (pellets). This table also displays
the selectivities to acetic acid and carbon dioxide (SAA and SCD). As
can be verified the PdO/ZrO2 catalyst, not only shows a higher rate
of acetic acid generation but also that it is 50% more selective to
0.0
0
250 500 7i s5o 0t hermic
temperature / ºC
0
250 500 i7s 5o 0t hermic
◦
o
temperature / C
Fig. 5. (A) TPD spectra of ethanol over the powder prepared PdO/Al2O3 catalyst. The
spectrum related to C2H5OH is multiplied by 4 and CH3CHO by 10. (B) TPD spectra
of ethanol over the powder prepared PdO/ZrO2 catalyst. The spectrum related to
C2H5OH is multiplied by 4, CH3CHO by 10, CH3COOH by 20, CO2 by 10, CO by 2.
acetic acid than the PdO/Al O3 catalyst. Additionally, it also shows
2
a higher rate of carbon dioxide generation and selectivity to carbon
dioxide when compared with the alumina one.
Gaspar et al. prepared a PdO/ZrO2 catalyst by ionic exchange.
This catalyst showed almost only dispersed Pd species and low
rates calculations (Table 2). Although it is not possible to compare
directly the selectivity of these catalysts due to their different con-
versions, these figures depict results that it is worth describing.
For both catalysts, as the temperature increases, the conversion
of ethanol and the selectivity to acetic acid and carbon dioxide
increase as well, whereas the selectivity to acetaldehyde decreases.
This result is in line with the steps of the ethanol oxidation mecha-
nism. Small amounts of methane can be also observed, which might
be generated by the decarbonylation of acetaldehyde. This reac-
tion also produces carbon monoxide which is oxidized to carbon
dioxide. Ethyl acetate, which is generated by the condensation of
selectivity to acetic acid [2]. Therefore, the high rate of PdO/ZrO to
2
acetic acid cannot be associated with the presence of dispersed Pd
species.
The ratio between the ethanol consumption rates of PdO/ZrO2
and PdO/Al O is 1.5. Comparing this value with the metallic sur-
2
3
face areas ratio of these samples (3.5), it can be inferred that
PdO/Al O shows a higher rate of ethanol consumption per unit of
2
3
Pd on the surface (activity) than PdO/ZrO (Table 2). As described in
2
the methane combustion literature [29], the larger the size of PdO
particle, the higher the oxidation activity of the catalyst.
ethanol with acetaldehyde on the ZrO surface, is also observed [3].
2
On the one hand, the ratio between the acetaldehyde genera-
The selectivity to this ester decreases as the temperature increases.
As can be inferred, the PdO/ZrO2 catalyst shows a higher yield
(higher conversion and selectivity) to acetic acid than PdO/Al O .
These results are in line with the ethanol consumption and acetic
acid formation rates depicted in Table 2.
tion rates of PdO/ZrO2 and PdO/Al O3 is 0.3 (Table 2). This value
2
is much lower than the one obtained for the ethanol consump-
tion (1.5). These results show that PdO/Al O is very active for the
2
3
2
3
acetaldehyde generation when compared with PdO/ZrO . It is well
2
known that, first, this alcohol is oxidized to acetaldehyde. After that,
the aldehyde is oxidized to acetic acid. Finally, this acid can be oxi-
dized to CO2 [12,15]. This behavior can be associated not only with
Fig. 5A and B shows the TPD of ethanol on the PdO/Al O3 and
2
PdO/ZrO2 powder catalysts, respectively. The TPD of ethanol on
PdO/Al O (Fig. 5A) depicts that the ethanol and acetaldehyde des-
2
3
the PdO particle size of Al O3 catalyst but also with the possibility
of the PdO/ZrO2 catalyst keeping this aldehyde on its surface (see
orption occur at low temperatures. Hydrogen was not observed.
Thus, this aldehyde is generated by the oxidative dehydrogenation
of ethanol on PdO and concurrently this oxide is reduced by ethanol.
2
selectivities to acetic acid and CO , and also their synthesis rates
2
◦
on PdO/ZrO ), thus decreasing its acetaldehyde generation rate.
The methane desorption occurs at 173 C with a small shoulder
2
◦
◦
On the other hand, the ratio between the acetic acid generation
at 226 C, whereas carbon dioxide desorption can be observed
◦
rates of PdO/ZrO2 and PdO/Al O3 is 2.3. This result shows that the
at 229 C with a shoulder at 182 C. Acetaldehyde can be decar-
bonylated (CC bond scission) generating equimolecular amounts of
2
PdO/ZrO2 catalyst is more active for the acetaldehyde oxidation
than PdO/Al O . The same behavior can be observed in the case of
◦
methane and CO [30]. The methane peak at 173 C and the shoulder
of CO (182 C) suggest that the decarbonylation occurs on Pd sites
followed by the oxidation of CO to CO . The Pd species are gen-
2
3
◦
0
CO . It was also verified that ZrO (without PdO) shows very low
2
2
2
0
activity and it generated mainly ethylene at these experimental
conditions.
2
erated by ethanol oxidative dehydrogenation. The peak of CO2 at
◦
Fig. 4A and B show some catalytic results related to the oxi-
dation of ethanol employing the prepared pellets samples at 140
229 C should be mainly assigned to the oxidation of acetaldehyde.
◦
The small peak related to CH4 (226 C) shows that the decarbony-
◦
and 175 C. The higher temperature results were employed in the
lation is less relevant at this temperature. These results suggest,

1
82
S. Letichevsky et al. / Journal of Molecular Catalysis A: Chemical 410 (2015) 177–183
firstly, that acetaldehyde might be adsorbed according to two dif-
The higher rate of acetic acid formation exhibited by PdO/m-
ferent configurations (ꢁ and ꢁ ) on the catalyst [31] and, secondly,
ZrO2 when compared with PdO/Al O3 might be due to higher
1
2
2
that the oxidation of ethanol to acetaldehyde occurs by Mars and
Van Krevelen mechanism [32].
number of active sites of the former. These sites are mainly related
to the oxidation of acetaldehyde on m-ZrO . This behavior can be
2
The TPD of ethanol on PdO/ZrO2 shows spectra with much
higher intensities when compared with the ones of PdO/Al O .
associated with the O superficial mobility of m-ZrO2.
As it is well described in the literature, m-ZrO2 shows strong
acid and basic sites. The former are related to the adsorption of
ethoxide species and acetaldehyde on m-ZrO2 whereas the latter
to the ethoxide species formation.
Despite of the spillover of ethoxide species from m-ZrO2 to PdO
and acetaldehyde from PdO to m-ZrO2 the rates of ethanol con-
2
3
Indeed, Table 1 shows that PdO/ZrO adsorbs 40 times more ethanol
2
than PdO/Al O . This occurs because ZrO is able to adsorb ethanol
2
3
2
and other oxygenates on its acid sites [3,7]. Fig. 5B shows a broad
◦
peak of ethanol desorption at around 150 C, whereas acetaldehyde
◦
exhibits an even broader peak (275–350 C) at higher tempera-
tures. It was observed that ZrO2 shows very low activity for the
ethanol oxidation. Therefore, this synthesis can be associated with
the oxidative dehydrogenation of ethanol on PdO generating Pd , as
sumption and acid acetic generation of PdO/m-ZrO are higher than
the one of the alumina catalyst, showing that these steps should not
be slow.
2
0
proposed for the alumina catalyst. Thus, ethanol or ethoxide species
[
7] adsorbed on ZrO2 might spillover to PdO and then be oxidized
4. Conclusions
to acetaldehyde.
According to Pajonk [33], the spillover phenomenon involves
the transport of active species (spillover species) sorbed or formed
on a donor (first phase) onto the acceptor (another phase). This
author mentions that this phenomenon can happen to oxygenates,
for instance, methoxy groups.
PdO/ZrO is a very promising catalyst for the generation of acetic
2
acid from ethanol. It reaches high selectivity to acetic acid at high
conversion. Dispersed Pd species and PdO are observed on the
PdO/ZrO2 surface. In the case of PdO/Al O only PdO species are
2
3,
detected. The PdO sites are the active ones for PdO/Al O whereas,
2
3,
Comparing the PdO/Al O aldehyde spectrum (Fig. 5A) with the
in the case of PdO/ZrO2 catalyst, both PdO and ZrO2 are in charge
of this synthesis. The catalytic behavior of m-ZrO2 can be associ-
ated with the adsorption and spillover of the oxygenate species
from this oxide to PdO and vice-versa. Moreover, the oxidation of
2
3
broad and high intensity one of PdO/ZrO2 (Fig. 5B), it can be sug-
gested that acetaldehyde is not only adsorbed on PdO but mainly on
ZrO2 (high desorption temperatures). Hence, this aldehyde might
spillover from PdO toward the support. In point of fact, previous
works of our group had already reported the spillover of acetalde-
hyde on ZrO2 [3,7], see Section 1.
acetaldehyde occurs both on PdO and m-ZrO . The results show that
2
m-ZrO2 is more than a support. Indeed, this oxide works in coop-
eration with PdO generating a very active system for the ethanol
oxidation to acetic acid.
◦
Fig. 5B also shows the acetic acid spectrum (133–284 C). The
acetaldehyde oxidation to acetate species (precursor of the acetic
acid) by ZrO2 was observed by Rodrigues et al. [7] employing IR
spectroscopy. It was suggested that the O of the ZrO2 surface oxi-
dizes this aldehyde. This same reaction might occur during the TPD
experiment (Fig. 5B). The TPD of ethanol over PdO/Al O does not
exhibit an acetic acid spectrum showing that the acetaldehyde syn-
thesis reduces PdO making it impossible for this catalyst, at this
experimental condition, to oxidize acetaldehyde to acetic acid. As
Acknowledgments
The authors are grateful to the LACAT/INT staff for the charac-
terization analyses and the LaH2S staff for the optical microscopy
results as well.
2
3
References
PdO/ZrO2 employs the oxygen of ZrO , i.e., it generates acetate
2
species, this oxide seems to be able to synthesize acetic acid. In addi-
tion, methane combustion studies employing PdO/ZrO2 showed
that the O of the ZrO2 lattice can re-oxidize Pd [19]. Therefore,
[1] A.B. Gaspar, A.M.L. Esteves, F.M.T. Mendes, F.G. Barbosa, L.G. Appel, Appl.
Catal. A Gen. 363 (2009) 109–114.
[
2] A.B. Gaspar, F.G. Barbosa, S. Letichevsky, L.G. Appel, Appl. Catal. A Gen. 380
(1–2) (2010) 113–117.
0
PdO might be partly recovered during the TPD experiments, and
thus, producing more acetaldehyde and acetic acid which might
contribute to the broad peaks mentioned above. The TPD of ethanol
on PdO/ZrO2 also exhibits CO and CO2 signals (CH4 was observed
but not shown). The C C and C O bond scissions of the acetalde-
hyde and carboxylate species produce these compounds which are
[3] P.C. Zonetti, J. Celnik, S. Letichevsky, A.B. Gaspar, L.G. Appel, J. Mol. Catal. A
Chem. 334 (1–2) (2011) 29–34.
[
4] A.G. Sato, D.P. Volanti, D.M. Meira, S. Damyanova, E. Longo, J.M.C. Bueno, J.
Catal. 307 (2013) 1–17.
[5] D.L. Carvalho, R.R. de Avillez, M.T. Rodrigues, L.E.P. Borges, L.G. Appel, Appl.
Catal. A Gen. 415–416 (2012) 96–100.
[
6] D.L. Carvalho, L.E.P. Borges, L.G. Appel, P.R. la Piscina, N. Homs, Catal. Today
13 (2013) 115–121.
2
0
associated with the presence of metal (Pd ) on the catalyst surface
[7] C.P. Rodrigues, P.C. Zonetti, C.G. Silva, A.B. Gaspar, L.G. Appel, Appl. Catal. A
Gen. 458 (2013) 111–118.
[
30].
Taking the TPD of ethanol on PdO/Al O , the catalysts charac-
[
8] S. Mizuno, M. Kurosawa, M. Tanaka, M. Iwamoto, Chem. Lett. 41 (2012)
92–894.
2
3
8
terization data and the catalytic test results (Table 2, PdO/Al O )
2
3
[9] C. Liua, J. Sun, C. Smith, Y. Wang, Appl. Catal. A Gen. 467 (2013) 91–97.
[10] E.V. Makshina, W. Janssens, B.F. Sels, P.A. Jacobs, Catal. Today 198 (2012)
into account, it can be proposed that PdO synthesizes acetaldehyde
by the oxidative dehydrogenation of ethanol. In the presence of air,
this oxide is also able to oxidize acetaldehyde to acetic acid.
In short, during the reaction ethanol is adsorbed on PdO and
3
38–344.
11] F. Gon c¸ alves, P.R.S. Medeiros, J.G. Eon, L.G. Appel, Appl. Catal. A Gen. 193 (1–2)
2000) 195–202.
[12] P.R.S. Medeiros, J.G. Eon, L.G. Appel, Catal. Lett. 69 (1–2) (2000) 79–82.
[
(
[
[
13] T. Takei, N. Iguchi, M. Haruta, New J. Chem. 35 (2011) 2227–2233.
14] B. Jørgensen, S.B. Kristensen, A.J. Kunov-Kruse, R. Fehrmann, C.H. Christensen,
A. Riisager, Top. Catal. 52 (2009) 253–257.
ZrO . Then, ethanol is oxidized to acetaldehyde on PdO. This alco-
2
hol, i.e., the ethoxide formed on ZrO2 might spillover from ZrO2
to PdO being oxidized to acetaldehyde. After that, this aldehyde is
[15] X. Li, E. Iglesia, Chem. Eur. J. 13 (2007) 9324–9330.
[
16] L.G. Appel, C., Moraes, S.E.C., Pereira, Z. Guimarães, P.R.S. Medeiros, BR
,901,776 (1990), to Instituto Nacional de Tecnologia.
17] M.C. Greca, C. Moraes, A.M. Segadães, Appl. Catal. A Gen. 216 (2001) 267–276.
oxidized to acetic acid on PdO or it migrates toward ZrO and is oxi-
2
8
dized on this oxide to acetate species which desorb as acetic acid.
Oxygen (air) reoxidizes the catalyst. ZrO2 might also promote the
reoxidation of PdO [19].
It was observed that m-ZrO2 interact with PdO precursors pro-
ducing Pd disperse species, which seems not to be very relevant for
this reaction.
[
[18] D.C.D. da Silva, S. Letichevsky, L.E.P. Borges, L.G. Appel, Int. J. Hydrog. Energy
0 (2015) 8706–8712.
19] D. Ciaparu, F. Bozon-Verduraz, L. Pfefferle, J. Phys. Chem. B 106 (2002)
434–3442.
4
[
3
[20] D. Ciuparu, M.R. Lyubovsky, E. Altman, L.D. Phefferle, A. Datye, Catal. Rev. 44
(2002) 593–649.

S. Letichevsky et al. / Journal of Molecular Catalysis A: Chemical 410 (2015) 177–183
183
[21] A.V. Nelmark, L.I. Kheifez, V.B. Fenelonov, Ind. Eng. Chem. Prod. Res. Dev. 20
[27] D. Ciuparu, A. Ensuque, F. Bozon-Verduraz, Appl. Catal. A Gen. 326 (2007)
(
1981) 439–450.
130–142.
[
[
[
22] D.C.R.M. Santos, L. Madeira, F.B. Passos, Catal. Today 149 (2010) 401–406.
23] F.G. Botes, W. Bohringer, Appl. Catal. A Gen. 267 (2004) 217–225.
24] J.F. Le Page, J. Cosyns, P. Courty, E. Freund, J.P. Franck, Y. Jacquin, B. Juguin, C.
Marcilly, G. Martino, J. Miquel, R. Montarnal, A. Sugier, H. Van Ladeghem,
Applied Heterogeneous Catalysis: Design Manufacture Use of Solid Catalysis,
first ed., Technip, Paris, 1987.
[28] K. Pokrovski, K.T. Jung, A.T. Bell, Langmuir 17 (2001) 4297–4303.
[29] Y.H. Chin, D.E. Resasco, in: J.J. Spivey (Ed.), Catalysis—Specialists Periodical
Reports, vol. 14, The Royal Society of Chemistry, Cambridge, 1999, pp. 1–39.
[30] A.F.F. de Lima, R.C. Colman, F.M.Z. Zotin, L.G. Appel, Int. J. Hydrog. Energy 35
(2010) 13200–13205.
[31] M. Mavrikakis, M.A. Barteau, J. Mol. Catal. A Chem. 131 (1998) 135–147.
[32] P. Mars, D. Van Krevelen, Chem. Eng. Suppl. 3 (1954) 41–57.
[33] G.M. Pajonk, in: G. Ertl, H. Knozinger, F. Schuth, J. Weitkamp (Eds.), Handbook
of Heterogeneous Catalysis, vol. 3, first ed., Wiley–VCH, Weinheim, 1997, pp.
1064–1077.
[
[
25] Z. Boukha, M. Kacimi, M. Ziyad, A. Ensuque, F. Bozon-Verduraz, J. Mol. Catal. A
2
70 (2007) 205–213.
26] D. Tessier, A. Rakai, F. Bozon-Verduraz, J. Chem. Soc. Faraday Trans. 88 (1992)
41–749.
7