Chemicals from ethanol - The acetone one-pot synthesis

Article abstract of DOI:10.1016/j.apcata.2013.03.028

Ethanol is a platform molecule that can generate many different products employing one-pot processes and multifunctional catalysts. The synthesis of acetone from ethanol is a good example of these processes. Some authors studied the generation of acetone from ethanol and proposed mechanisms which are not fully in agreement. This work using catalyst physical mixtures and IR spectroscopy (DRIFTS) is consistent with the generation of acetone from ethanol by ketonization of carboxylates which are obtained by dehydrogenation of this alcohol followed by oxidation. These results confirm that physical mixtures of catalysts can be a valuable tool in order to study the mechanism of multi-step reactions.

Full text of DOI:10.1016/j.apcata.2013.03.028

Applied Catalysis A: General 458 (2013) 111–118
Contents lists available at SciVerse ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Chemicals from ethanol—The acetone one-pot synthesis
Clarissa P. Rodrigues, Priscila C. Zonetti, Camila G. Silva,
Alexandre B. Gaspar, Lucia G. Appel∗
Divisão de Catálise e Processos Químicos, Instituto Nacional de Tecnologia, Av. Venezuela 82/518, CEP 21081-312, Rio de Janeiro, RJ, Brazil
a r t i c l e i n f o
a b s t r a c t
Article history:
Ethanol is a platform molecule that can generate many different products employing one-pot processes
and multifunctional catalysts. The synthesis of acetone from ethanol is a good example of these processes.
Some authors studied the generation of acetone from ethanol and proposed mechanisms which are not
fully in agreement. This work using catalyst physical mixtures and IR spectroscopy (DRIFTS) is consistent
with the generation of acetone from ethanol by ketonization of carboxylates which are obtained by
dehydrogenation of this alcohol followed by oxidation. These results confirm that physical mixtures of
catalysts can be a valuable tool in order to study the mechanism of multi-step reactions.
© 2013 Elsevier B.V. All rights reserved.
Received 29 November 2012
Received in revised form 17 March 2013
Accepted 23 March 2013
Available online xxx
Keywords:
Acetone
Ethanol
Zirconia
Ketonization
1. Introduction
formic acid. More acetone can be produced by the ketonization of
the acetic acid.
Nowadays, the Brazilian and American huge production of
ethanol, and mainly the future large generation of this alcohol from
cellulosic residues offer a great opportunity for the development of
new processes and products employing renewable raw materials.
One very interesting example is the one-pot acetone synthesis from
ethanol (Eq. (1)). Acetone is an important solvent and it is also used
in the syntheses of drugs and polymers.
Nakajima et al. [3–5] studied many different catalytic systems
for the synthesis of acetone from ethanol. They obtained very high
activities and selectivities employing Fe-Zn and Zn-Ca mixed oxide
catalysts. They proposed that the acetone formation from ethanol
is an acid-base bifunctional catalytic reaction. According to their
suggestions, ethanol is dehydrogenated to acetaldehyde. After that,
acetate species (CH3COO−) are obtained by the reaction of this alde-
hyde with the superficial hydroxyl species. The following step is the
condensation of these acetate species. In addition the authors sug-
gest that the basicity of the catalyst enhances the nucleophilicity
of the superficial hydroxyls and that the acidity stabilizes the tran-
sition states with negative charges. They also proposed that the
synthesis of the acetate species is the rate-determining step of this
system.
Idriss and Seebauer [6] and Yee et al. [7] studying the synthesis
of acetone from ethanol proposed two different routes. The first one
is related to the acetyl disproportionation reaction, which occurs in
the presence of a metal, for instance Pd. Ethanol is dehydrogenated
to acetaldehyde and this aldehyde to acetyl species; afterwards,
two acetyl molecules react generating one molecule of acetone and
one of CO. The second pathway involves the dehydrogenation of
ethanol generating acetaldehyde followed by the oxidation of this
aldehyde to acetate species on the oxide surface. Then, the coupling
of the two carboxylates species, to yield one molecule of acetone
and one molecule of CO2 occurs. According to these same authors,
the surface of the catalyst must be capable of donating its oxygen in
order to oxidize the acetaldehyde and also be able to accommodate
two acetate molecules for the condensation reaction.
2CH3CH2OH + H2O → CH3COCH3 + CO2 + 4H2
(1)
Nishiguchi et al. [1] investigated the steam reforming of ethanol
employing Cu supported on CeO2. They observed acetone as one
of the byproducts. They proposed that acetone formation occurs
by the aldol condensation of acetaldehyde producing acetaldol
(3-hydroxybutyraldehyde), which is oxidized by the O of the sup-
port. After that, this compound decomposes itself producing carbon
dioxide, acetone and hydrogen.
Murthy et al. [2] studied the generation of acetone from ethanol
employing Fe2O3-ZnO, Fe2O3-CaO and Fe2O3-Mn in the presence
of water. They proposed that, firstly, ethanol is dehydrogenated
to acetaldehyde. Then, the aldol condensation of this aldehyde
generates acetaldol. According to these authors, this compound is
dehydrogenated to 1,3-dicarbonyl, which can be decomposed by
hydrolysis generating acetic acid and acetaldehyde or acetone and
∗
Corresponding author. Tel.: +55 21 21231166; fax: +55 21 21231165.
E-mail addresses: lucia.appel@int.gov.br, lucia.appel@gmail.com (L.G. Appel).
0926-860X/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2013.03.028

112
C.P. Rodrigues et al. / Applied Catalysis A: General 458 (2013) 111–118
Recently, some works have shown that physical mixtures of
and the spectra were collected at 100, 200, 300 and 400 ◦C. The data
obtained were treated and the presence of water was removed from
the spectra employing Thermo Nicolet OMNICTM software.
catalysts are simple and effective tools to describe multi-step reac-
tions. These systems were employed successfully by our group in
several reactions, as for example: the one-step ethyl acetate syn-
thesis from ethanol [8,9], the methanation reactions [10] and the
dimethyl ether synthesis [11].
2.3. Catalytic tests
In summary, the various suggestions made by these authors
which are related to the mechanism of the acetone synthesis from
ethanol are not fully in agreement. Therefore, using physical mix-
tures and IR spectroscopy, this work aims to further study the
one-step process which generates acetone from ethanol.
The catalytic tests were performed using a conventional sys-
tem with a fixed bed reactor at atmospheric pressure. Ethanol
and water vapors were generated by passing N2 through two
saturators, one at 4.9 ◦C and the other at 53.6 ◦C, respectively. A
N2:H2O:C2H5OH = 91:8:1 mol% mixture was fed at 70 mL min−1
.
When only ZrO2 (Figs. 2A, 4) or CZA (Fig. 2A and B) were employed
as catalyst the weight was 50 mg and 25 mg, respectively. The
weight of ZrO2 in CZA + ZrO2 (1:1) (Figs. 1, 2B and 5) and CZA + ZrO2
(1:2) (Figs. 2A, B and 3) physical mixtures were 25 mg and 50 mg,
respectively. The weight of CZA in the CZA + ZrO2 (1:1) (Figs. 1,
2B and 5) and CZA + ZrO2 (1:2) (Figs. 2A, B and 3) physical mix-
tures were 25 mg. The samples were previously dried under N2
flow at 130 ◦C for 30 min and reduced under H2/N2 flow (10%, v/v,
100 mL min−1) at 450 ◦C for 1 h. The catalytic tests were carried out
at 400 ◦C. One exception was the study related to the influence of
the temperature in the catalytic behavior (Fig. 1). In this case, when
the test was carried out at 475 ◦C (Fig. 1), the catalyst was reduced
at that same temperature. The products were analyzed on-line with
gas chromatograph GC Agilent 6890 equipped with two detectors
(thermal conductivity and flame ionization) and a Porapak-Q/60 ft
column using He as the carrier gas. The samples were analyzed
every 23 min during 12 h on stream.
Catalytic tests at isoconversion (∼80%) employing ZrO2, CZA,
CZA + ZrO2 (1:2) (Fig. 2A) were carried out using the same amount
of catalyst described above. The flow rate of the gaseous mixture
was changed in order to reach the isoconversion. All the other
experimental conditions remained unchanged.
In order to evaluate the role of water and CZA in the acetone
synthesis, two catalytic tests were carried out. The first employed
(Fig. 2B) CZA + ZrO2 (1:2) and a gaseous mixture without water
(N2:C2H5OH = 99:1 mol%), whereas the second one (Fig. 4) replaced
ethanol by acetaldehyde (N2:H2O:C2H4O = 91:8:1 mol%) employ-
ing ZrO2 as catalyst. All the other experimental conditions remained
unchanged.
2. Experimental
2.1. Catalysts
Binary physical mixtures comprising of Cu/ZnO/Al2O3 (CZA) and
one oxide (ZrO2 (monoclinic), Al2O3 and MgO) were employed. The
CZA, ZrO2, and Al2O3 were commercial samples. The oxides were
supplied by NORPRO. The composition of CZA is 60% CuO, 30% ZnO
and 10% Al2O3. The MgO sample was synthesized by the precipita-
tion method [12]. The physical mixtures are named as CZA + oxide
(1:1). The numbers are related to the weight ratio between CZA and
the oxides. The catalyst and the oxide were gently mixed employ-
ing a small beaker and a thin glass stick. The particle size of the
catalysts and oxides was smaller than 0.053 mm.
2.2. Characterization
The acid sites of the oxides were analyzed by pyridine thermo
desorption followed by IR spectroscopy using a Nicolet Magna
Spectrophotometer. The spectra were recorded using thin self-
supporting wafers (∼20 mg). The samples were pretreated at 500 ◦C
for 2 h under high vacuum, submitted to dry air pulses at the same
temperature for 1 h and exposed again to high vacuum for 30 min.
Pyridine was adsorbed at room temperature for 30 min at 2 Torr.
The spectra were collected after desorption at 25 ◦C for 1 h under
high vacuum. The values of the acid density were obtained consid-
ering the absorption around 1445 cm−1 at 25 ◦C. The extinction
coefficients published by Tamura et al. [13] were employed for the
quantification of the acid sites.
Catalytic tests at isoconversion (∼35%) employing physical mix-
tures CZA + oxides (1:1) were carried out using ZrO2, MgO and Al2O3
The basic sites were analyzed by temperature-programmed
desorption of CO2 (TPD of CO2) using a micro reactor coupled to
a Micromeritics Autochem II 2920. The oxides (500 mg) were pre-
treated under N2 flow (50 mL min−1) at 130 ◦C for 30 min, at 450 ◦C
for 1 h. The CO2 adsorption was conducted at room temperature
for 1 h (25 mL min−1). The TPD of CO2 was carried out heating the
sample at 10 ◦C min−1 up to 400 ◦C under He flow (50 mL min−1).
The signal was monitored using a TCD detector. Desorption pro-
files were deconvoluted in order to quantify the weak, medium and
strong basic sites. The basic sites related to a curve which shows
its maximum value at temperatures below 170 ◦C are assigned as
weak; those between 170 and 270 ◦C as medium basic sites; and
finally, those above 270 ◦C as strong ones [12].
Diffuse Reflectance Infrared Fourier Transform Spectroscopy
(DRIFTS) analyses have been carried out employing a Nicolet Spec-
trometer 6700 with a MCT/B detector equipped with a diffuse
reflectance assembly chamber (Spectra-Tech) and ZnSe window.
All spectra were obtained using 4 cm−1 resolution and 64 scans.
ZrO2 and CZA + ZrO2 (1:1) were pretreated in situ under He flow
(24 mL min−1) at 130 ◦C for 30 min and reduced under H2 flow
(24 mL min−1) at 450 ◦C for 1 h. Then, the samples were cooled
under He flow. The spectra of the samples after pretreatment were
collected as background spectra. The gaseous mixture composed
of He, H2O and C2H5OH (24 mL min−1) was introduced in the cell
Fig. 1. Ethanol conversion and selectivities of CZA + ZrO2 (1:1) physical mixture
versus temperature. The flow rate, the gas mixture composition, the weights of CZA
and ZrO2 are 70 mL min−1, N2:H2O:C2H5OH = 91:8:1 mol%, 25 mg, 25 mg, respec-
tively.

C.P. Rodrigues et al. / Applied Catalysis A: General 458 (2013) 111–118
113
Fig. 4. Acetaldehyde conversion and selectivities of ZrO2 versus time on stream.
The flow rate, the gas mixture composition, temperature and the mass of ZrO2 are
70 mL min−1, N2:H2O:C2H4O = 91:8:1 mol%, 400 ◦C and 50 mg, respectively.
Fig. 2. (A) Selectivities of different catalytic system at isoconversion. The gas mix-
ture composition, temperature, the weight of CZA and the weight of ZrO2 are
N2:H2O:C2H5OH = 91:8:1 mol%, 400 ◦C, 25 mg and 50 mg, respectively. These same
weights are employed in the physical mixture CZA + ZrO2 (1:2). The flow rate was
changed in order to reach the isoconversion. (B) Ethanol conversion and selectivities
of different catalytic system. The flow rate, the gas mixture composition, tempera-
ture and the weight of CZA are 70 mL min−1, N2:H2O:C2H5OH = 91:8:1 mol%, 400 ◦
C
and 25 mg, respectively. When ZrO2 is employed in CZA + ZrO2 (1:1) and CZA + ZrO2
(1:2) physical mixtures its weight is 25 mg. and 50 mg, respectively.
Fig. 5. Ethanol conversion and selectivities of CZA + ZrO2 (1:1), CZA + MgO(1:1),
CZA + Al2O3 (1:1) physical mixtures at isoconversion. The flow rate, the gas mix-
ture composition and temperature are 70 mL min−1, N2:H2O:C2H5OH = 91:8:1 mol%
and 400 ◦C, respectively.
in order to evaluate the role of the acid and basic properties of these
oxides in the acetone synthesis (Fig. 5). All samples were tested
at the same conditions described above, except for the mass of
CZA + Al2O3 (1:1) which was 20 mg for both CZA and Al2O3.
The ethanol conversion was defined as the ratio of the moles of
ethanol consumed to the moles of ethanol introduced in the feed.
The definition of the selectivity to one specific compound is the
ratio of the number of carbon moles consumed to synthesize this
compound to the total number of carbon moles consumed. The car-
bon balance was higher than 90% for all the catalytic systems tested,
except for the data obtained using acetaldehyde as reagent.
The rates of ethanol consumption, acetone and acetaldehyde
formation at 375 ◦C were obtained employing differential con-
ditions (5 mg of CZA, 10 mg of CZA + ZrO2 (1:1) and 15 mg of
CZA + ZrO2 (1:2) and 90 mL mim−1 of the gaseous mixture).
3. Results and discussion
The ethanol conversion and selectivity versus the reaction
temperature for CZA + ZrO2 (1:1) physical mixture are shown
in Fig. 1. Increasing the temperature, the selectivity to acetone
and the ethanol conversion increase as well, while the selectiv-
ity toward acetaldehyde decreases showing that this aldehyde
might be an intermediate of the acetone synthesis. The selectiv-
ity to carbon dioxide, a byproduct of this reaction, follows the
same trend observed for acetone at temperatures below 450 ◦C.
Hydrogen (not shown) is also generated during the acetone syn-
thesis at stoichiometric values. The selectivity to acetone shows a
Fig. 3. Ethanol conversion and selectivities of CZA + ZrO2 (1:2) physical mixture
versus time on stream. The flow rate, the gas mixture composition, temperature
and the weight of CZA and ZrO2 are 70 mL min−1, N2:H2O:C2H5OH = 91:8:1 mol%,
400 ◦C, 25 mg and 50 mg, respectively.

114
C.P. Rodrigues et al. / Applied Catalysis A: General 458 (2013) 111–118
maximum at around 430 ◦C. Increasing the temperature, the selec-
tivity toward acetone decreases and the selectivity to 1-butene
increases. The latter, might be synthesized by the aldol conden-
sation of acetaldehyde. The literature has described this reaction
as follows: two molecules of acetaldehyde condense generating
acetaldol (3-hydroxybutyraldehyde), which dehydrates to croton-
aldehyde. Then, this aldehyde is hydrogenated to n-butanol. Finally,
the dehydration of this alcohol produces 1-butene [14].
Table 1
Density of weak and medium basic sites (M), density of strong basic sites (S), density
of basic sites (T) and density of the acid sites (A).
Oxides
M (␮mol g−1
)
S (␮mol g−1
)
T (␮mol g−1
)
A (␮mol g−1
)
Al2O3
MgO
ZrO2
199
219
126
–
409
90
199
628
216
186
–
92
Fig. 2A shows the selectivity of different catalytic systems at
isoconversion (∼80%). When ZrO2 or CZA are employed the main
products are ethylene and acetaldehyde, respectively. As it is well
described in the literature ZrO2 shows pairs of acid and basic Lewis
sites [9] which promote the dehydration of ethanol to ethylene. On
the other hand, acetaldehyde is synthesized by the dehydrogena-
tion of ethanol, which occurs on the Cu0 sites of CZA [9]. At the
experimental conditions employed (Fig. 2A) this reaction almost
reaches the thermodynamic equilibrium [15]. As well known, it
also produces hydrogen. As Fig. 2A shows these two catalysts also
generate small amounts of acetone and carbon dioxide. The phys-
ical mixture of these oxides (CZA + ZrO2 (1:2)) shows a completely
different catalytic behavior when compared with the ones of ZrO2
and CZA. Adding ZrO2 to CZA, the selectivity to acetone and also
to carbon dioxide increases while the selectivity to acetaldehyde
decreases.
Fig. 2B shows the distribution of the products and also the con-
version of ethanol using different catalytic systems under the same
experimental conditions this time around. The CZA catalyst shows
low conversion, which is far from the thermodynamic equilibrium
[15]. Adding ZrO2 to CZA (CZA + ZrO2 (1:1)) the ethanol conversion
rises, the selectivity to acetone and carbon dioxide also increases
while the selectivity to acetaldehyde decreases. Doubling the con-
centration of ZrO2 in the physical mixture (CZA + ZrO2 (1:2)), the
ethanol conversion and the selectivity to acetone and carbon diox-
ide increase even further whilst the selectivity to acetaldehyde
decreases. Thus, the acetone synthesis might occur on the ZrO2
surface.
The estimation of the ethanol consumption rates employing
CZA, CZA + ZrO2 (1:1) and CZA + ZrO2 (1:2) leads to almost the same
value (1.5 mmolethanolgCZA−1 min−1) at 375 ◦C. The CZA catalyst
synthesizes only acetaldehyde and the physical mixtures acetalde-
hyde, acetone and carbon dioxide. The ratios between the rates of
acetone and acetaldehyde generation are 0, 0.1 and 0.2 for CZA,
CZA + ZrO2 (1:1) and CZA + ZrO2 (1:2), respectively. These results
show again that acetone is obtained on ZrO2. When increasing the
amount of ZrO2 in the physical mixture the rate of acetone synthesis
increases, as well.
Zonetti et al. [9] using CZA + ZrO2 observed that acetaldehyde
generated by CZA migrate from this catalyst to ZrO2 and under-
goes a condensation reaction, i.e., the ethyl acetate synthesis on
this oxide. This phenomenon might also occur during the acetone
synthesis. In other words, acetaldehyde is generated on CZA and
then, it migrates toward the ZrO2 surface. After that, acetone and
CO2 are synthesized on the oxide surface.
first few hours the ethanol conversion and the selectivities to ace-
tone and carbon dioxide show a slight decrease, whereas there is an
increase of acetaldehyde concentration. After that, the deactivation
rate is very low.
In order to evaluate the role of CZA in the acetone synthesis, the
mixture of ethanol plus water was changed to acetaldehyde plus
water employing, in this case, only ZrO2 as catalyst. Fig. 4 depicts
the catalytic behavior with time on stream. The selectivity to ace-
tone and carbon dioxide are lower than the ones observed in Fig. 3,
and they decrease with time on stream. Moreover, the conversion
of acetaldehyde follows the same behavior. In the meantime, the
selectivity to crotonaldehyde increases. Thus, it can be inferred that,
for the acetone synthesis, the role of CZA is not only the dehydro-
genation of ethanol which generates acetaldehyde. Crotonaldehyde
is an intermediate of the aldol condensation. As acetaldehyde is
employed instead of ethanol, no hydrogen is available. In this case,
1-butene is not observed because hydrogenations of crotonalde-
hyde and butyraldehyde followed by dehydration of n-butanol can
not occur. The competition between the aldol condensation and
the acetone synthesis is clearly demonstrated by these results.
This phenomenon occurs due to the fact that acetaldehyde is an
intermediate of both syntheses. It can be suggested that the high
deactivation rate observed can be associated with the aldol con-
densation of crotonaldehyde.
Fig. 5 depicts the catalytic behavior of the following physical
mixtures: CZA + Al2O3 (1:1), CZA + MgO (1:1) and CZA + ZrO2 (1:1)
at isoconversion. Table 1 shows the acid and basic sites densi-
ties of these three oxides. The three physical mixtures generate
acetaldehyde, ethylene, carbon dioxide and acetone, however, with
different selectivities. Acetaldehyde is the product with the highest
concentration for all the three catalytic systems mentioned above.
The CZA + Al2O3 (1:1) mixture produces the highest selectivity to
ethylene. This result can be associated with its acid properties. This
oxide not only shows the highest concentration of acid sites, but
also the strongest ones [9] when comparing with MgO and ZrO2.
Despite the low density of the MgO acid sites (Fig. 5), this oxide is
able to adsorb oxygenates as depicted by the results obtained in this
work and the data available in the literature [16]. The CZA + MgO
physical mixture shows the highest selectivity to acetaldehyde.
As it is well known, the H abstraction is the slowest step of the
dehydrogenation of alcohols employing oxides as catalysts. There-
fore, this result can be associated with the high density of the
strong basic sites of MgO. The CZA + ZrO2 mixture shows the highest
selectivity toward acetone and carbon dioxide. However, as Fig. 5
depicts, this behavior can neither be directly correlated with the
acid nor the basic properties of ZrO2.
Fig. 2B depicts that when CZA + ZrO2 (1:2) is employed without
water, the main products obtained are ethylene and acetaldehyde.
The catalytic behavior seems to be related only to the contribution
of individual component of the physical mixture. Comparing the
result of this test with the one with water, it can be inferred that the
presence of water changes completely the behavior of the physical
mixture. Thus, water plays a major role in the acetone synthesis
from ethanol.
Fig. 3 depicts the catalytic behavior of CZA + ZrO2 (1:2) when
this system is in contact with ethanol and water as a function of
time on stream. The main products of this reaction are acetone,
carbon dioxide and acetaldehyde, as already mentioned. At conver-
sion around 90% the selectivity to acetone reaches 64%. During the
Taking the above mentioned results and the literature into
consideration, the following steps can be suggested for the syn-
thesis of acetone from ethanol: first, acetaldehyde is generated
on CZA catalyst by ethanol dehydrogenation [9]. According to Abe
et al. [17], ZrO2 is able to oxidize acetaldehyde to acetate species.
Moreover, Yamada et al. [18] and Parida et al. [19] showed that ace-
tone can be synthesized by the condensation of two carboxylate
species (ketonization reaction). Therefore, it is possible to suggest
that acetaldehyde migrates to ZrO2 by spillover or by desorp-
tion/adsorption, and afterwards, reacts with this oxide generating
acetate species. Then, these species adsorbed on ZrO2 condensate

C.P. Rodrigues et al. / Applied Catalysis A: General 458 (2013) 111–118
115
generating acetone and carbon dioxide. Furthermore, Phatak et al.
[20] employing the density functional theory showed that water
can be dissociated on Cu0. Later, Voss et al. [21] showed that the dis-
sociation of water on the Cu0 sites generate oxidant species. Thus,
water is dissociated by CZA and, the species obtained (O, OH), might
migrate and reoxidize the surface of ZrO2 which was reduced by the
acetate synthesis. The reoxidation of ZrO2 surface does not require
formal changes in the oxidation state of this oxide. It means addi-
tion of O to the O-vacancies. Indeed, the oxygen mobility of ZrO2 is
very well described for the methane oxidation [22]. Consequently,
CZA is in charge not only of the ethanol dehydrogenation, but also
of the synthesis of the oxidative species which regenerate the ZrO2
surface. These species might be generated on CZA, and then migrate
to the ZrO2 oxide. It is worth stressing that water does not partici-
pate in the dehydrogenation of ethanol, neither in the oxidation of
acetaldehyde nor in the ketonization reaction. The results suggest
that the presence of water is only related to reoxidation of ZrO2.
It is important to notice that ZrO2 surface is reduced before the
reaction (pretreatment). Thus, in the absence of water (Fig. 2), the
acetone synthesis does not occur since the oxide (ZrO2) is not able
to oxidize acetaldehyde.
As it was previously observed (Fig. 4), when only ZrO2 and the
acetaldehyde + water mixture are used, the selectivity to acetone
obtained is low compared with the results of CZA + ZrO2 (1:2). As
no Cu sites are available to generate the oxidant species, ZrO2 might
be only slightly reoxidized by water. Thus, this system synthesizes
low amounts of acetone.
The ketonization mechanism is not completely described yet
[18,19]. However, it is known that strong basic sites are relevant in
order to generate C-C bonds. Moreover, the acetone synthesis from
ethanol requires acid sites to adsorb acetaldehyde. The MgO oxide
shows a higher density of strong basic sites than ZrO2 (Table 1).
Both oxides are able to adsorb acetaldehyde. Fig. 5 depicts that the
CZA + ZrO2 (1:1) shows a much higher selectivity to acetone than
MgO. Therefore, this result might be associated with the mobility
of oxygen in the ZrO2 lattice.
Fig. 6. In situ DRIFTS spectra recorded at increasing temperatures on ZrO2 (a) 100 ◦C,
(b) 200 ◦C, (c) 300 ◦C and (d) 400 ◦C.
In other words, the highest selectivity to acetone is generated by
CZA + ZrO2 when comparing the catalytic behavior of the physical
mixtures composed of CZA with the ZrO2, Al2O3 and MgO. However,
ZrO2 does not show the highest amount of strong basic sites, nei-
ther the highest amount of acid sites (Table 1). This result suggests
that the slowest step of this reaction on the oxide is not related to
these properties. Considering the mechanism proposed, this result
suggests that the ZrO2 oxi-reduction properties might be the most
relevant for the reaction on the oxide.
Aiming to better describe the synthesis of acetone from ethanol,
CZA + ZrO2 (1:1) and ZrO2 were analyzed by DRIFTS.
Figs. 6 and 7 show the spectra obtained when ZrO2 is exposed
to a gaseous mixture of ethanol and water at four different temper-
atures.
Fig. 6 shows absorptions at 1160 and 1072 cm−1 which can be
attributed to the ꢀ(C O) modes for type I and type II of the ethoxide
species adsorbed on the ZrO2 surface. Increasing the temperature
the intensity of their absorptions slightly decreased [7,23].
Fig.
7
shows the infrared spectral features in the
1800–1200 cm−1 range. As many vibrations can be observed in
this region, the identification of the species is not straightforward.
At 100 ◦C (Fig. 7a) the absorption at 1473 cm−1 can be associated
with ethoxide species (ı(CH2)). At high temperature this absorption
shifts to 1462 cm−1, which can also be assigned to acetate species
(ıas(CH3), white line and numbers). The bands around 1577 cm−1
,
1413 cm−1 and 1340 cm−1, whose intensities increase as the tem-
perature increases, can be attributed to ꢀas(OCO), ꢀs(OCO) and
ꢁ(CH3) of the acetate species [7,23] (white lines and numbers),
respectively. A band at 1380 cm−1 can also be observed (Fig. 7a).
This absorption can be attributed to ethoxide species (ıs(CH3)).
Fig. 7. In situ DRIFTS spectra recorded at increasing temperatures on ZrO2 (a) 100 ◦C,
(b) 200 ◦C, (c) 300 ◦C and (d) 400 ◦C.

116
C.P. Rodrigues et al. / Applied Catalysis A: General 458 (2013) 111–118
Fig. 9. In situ DRIFTS spectra recorded at increasing temperatures on CZA + ZrO2
(1:1) (a) 100 ◦C, (b) 200 ◦C, (c) 300 ◦C and (d) 400 ◦C.
Fig. 8. In situ DRIFTS spectra recorded at increasing temperatures on CZA + ZrO2
Fig. 8a (100 ◦C) depicts three absorptions; one at 1162 cm−1
,
(1:1) (a) 100 ◦C, (b) 200 ◦C, (c) 300 ◦C and (d) 400 ◦C.
another at 1062 cm−1 and a very tiny one at 1099 cm−1. The first
one can be associated with the ethoxide species adsorbed on ZrO2
(ꢀ(C O), type I), as shown in Fig. 6. According to Yang et al. [25],
absorptions at 1101 and 1052 cm−1 can be assigned to ethox-
ide species adsorbed on Cu/ZnO catalyst, being the last one more
intense. Thus, the band at 1099 cm−1 can be assigned to the ethox-
ide species adsorbed on CZA. The broad band at 1062 cm−1 can be
attributed to a combination of ꢀ(C O) modes of the adsorbed ethox-
ide species on both ZrO2 (1072 cm−1, Fig. 6) and CZA (1052 cm−1).
Increasing the temperature up to 200 ◦C, (Fig. 8b) the band at
1062 cm−1 shifts to 1054 cm−1, its intensity decreases, and a shoul-
der around 1033 cm−1 appears, which might be assigned to ꢁ(CH3)
of acetate species adsorbed on CZA or ZrO2 [7,25]. In addition, the
intensity of the band at 1099 cm−1 increases, whereas the one at
1162 cm−1 decreases. At low temperatures, the ethoxide species
are adsorbed on both ZrO2 and CZA. Increasing the temperature
up to 200 ◦C, the amount of ethoxide species adsorbed on CZA
increases, whereas the amount of ethoxide species adsorbed on
ZrO2 decreases. At 300 ◦C and 400 ◦C ethoxide species adsorbed on
ZrO2 (ꢀ(C O), type I) disappear and the amount of ethoxide species
on CZA decreases (1054 cm−1). Comparing Fig. 6c with Fig. 8c, it is
possible to infer that the ethoxide species on ZrO2 did not des-
orb but rather migrate to CZA. Taking the results of catalytic test
(Fig. 2) into account it can be suggested that the ethoxide species
react on CZA generating acetaldehyde. Thus, when ZrO2 is in con-
tactwithCZAthelatterchangesthebehaviorof theethoxidespecies
adsorbed on ZrO2.
However, when the temperature increases its intensity decreases
and it almost disappears at 400 ◦C. Thus, this absorption and tiny
peaks at 1049 and 1090 cm−1 (Fig. 6a) can be attributed to ethanol
adsorbed.
The bands at 1658 cm−1 (ꢀ(C C), Fig. 7c), 1725 cm−1 (ꢀ(C O))
and C–H vibrations at 2705 and 2743 cm−1 (not shown) suggest that
crotonaldehyde is adsorbed on ZrO2 at temperatures above 200 ◦C
[24]. This compound is observed when ZrO2 and acetaldehyde are
employed as catalyst and reagent, respectively (Fig. 4).
It can be observed that increasing the temperature, the intensity
of the band at 1528 cm−1 (Fig. 7b–d) increases as well. At 400 ◦C
(Fig. 7d) a tiny band at 1311 cm−1 is also observed. Both absorptions
can be attributed to carbonate monodentate species [7].
These results show that ZrO2 in contact with ethanol and water
is able to synthesize carbonates, ethylene, acetates, ethoxides and
crotonaldehyde species. Carbonates are generated by the interac-
tion of CO2 with the ZrO2 basic sites. As already proposed, CO2
might be obtained by the condensation of acetates which also
produces acetone. Thus, it is possible to verify that the condensation
of acetates occurs on the ZrO2 surface.
Fig. 6 depicts that ethoxide species are observed even at high
temperatures. According to Zonetti et al. [9], ZrO2 shows a low
activity for the acetaldehyde synthesis. Moreover, ethylene is the
main product obtained when ZrO2 is employed as catalyst (Fig. 2).
Thus, the rate of the acetates condensation on ZrO2 is low due to
the low concentration of acetaldehyde.
Figs. 8 and 9 show the IR spectra obtained when the mixture
CZA + ZrO2 is exposed to the ethanol and water gaseous mixture at
four different temperatures.
Fig. 9a also shows ethoxides species at 1475 ı(CH2) and
1380 cm−1 ıs(CH3). The intensity of the latter decreases as the
temperature increases, showing that these species are consumed
during the reaction. The former shifts to lower values (1463 cm−1
,

C.P. Rodrigues et al. / Applied Catalysis A: General 458 (2013) 111–118
117
Table 2
Elementary surface reaction
Type
References
I
II
III
V
CH3CH2OH(g) + *1 = CH3CH2OH*1
CH3CH2OH*1 + *1 = CH3CH2O*1 + H*1
CH3CH2O*1 + *1 = CH3CHO*1 + H*1
CH3CHO*1 + *2 = CH3CHO*2 + *1
CH3CHO*1 = CH3CHO(g) + *1
CH3CHO(g) + *2 = CH3CHO*2
CH3CHO*2 + OL = CH3CHOO*2 + V0
CH3CHOO*2 + *2 = CH3COO*2 + H*2
2CH3COO*2 + *2 = CH3COCH3*2 + CO2*2 + O*2
CH3COCH3*2 = CH3COCH3(g) + *2
CO2*2 = CO2(g) + *2
Ethanol adsorption on Cu0
[6,7]
Ethoxides formation on Cu0
[6,7,32,33]
[32,33]
[9]
Ethoxy ␣-hydrogen abstraction on Cu0
Acetaldehyde spillover from Cu0 to ZrO2
Acetaldehyde desorption from Cu0
Acetaldehyde adsorption on ZrO2
Acetaldehyde oxidation on ZrO2
Formation of acetates on ZrO2
Acetone synthesis (ketonization)
Acetone desorption
VI
This work
This work
[16,20]
[16,20]
[6,7,17,18]
This work
This work
[20]
[19,20]
[20]
[19,20]
[34]
VII
VIII
IX
X
XI
XII
XIII
XIV
XV
XV
XVI
XVIII
XIX
Carbon dioxide desorption
H2O(g) + *1 = H2O*1
Water adsorption on Cu0
H2O*1 + *1 = OH*1 + H*1
Water dissociation
Hydroxyl disproportionation
Hydroxyl dissociation
2 OH*1 = H2O*1 + O*1
OH*1 = O*1 + H*1
O*1 + *2 = O*2 + *1
Oxygen spillover from Cu0 to ZrO2
Dihydrogen formation on Cu0 and ZrO2
Regeneration of the oxygen of the ZrO2 lattice
2H* = H2(g) + 2*
[20]
This work
V
0 + O*2 = OL + *2
*1: site on Cu0; *2: site on ZrO2; *: both sites on Cu0 and ZrO2; V0: O vacancies; OL: oxygen of the ZrO2 lattice.
white line and numbers) and its intensity increases as the tempera-
ture increases as well up to 300 ◦C, suggesting that the absorptions
of ethoxide species are replaced by the ones of acetate species
(ıas(CH3)) [7].
All the species observed by IR are associated with the aldol con-
densation (acetaldehyde, crotonaldehyde and 1-butene), acetate
condensation (acetaldehyde, acetates, carbonates and acetone) and
ethanol dehydration (ethene).
Carboxylates species (1585, 1463, 1412, 1338 cm−1) can be
observed on CZA + ZrO2 (Fig. 9, white numbers and lines) [7].
These bands increase their intensity as temperature increases up
to around 300 ◦C. These carboxylates are the same observed in the
case of ZrO2 (Fig. 7). According to Yang et al. [25], acetate species
adsorbed on Cu/ZnO can be observed at 1562 and 1451 cm−1. Ana-
lyzing the spectra it is impossible not to consider these bands.
However, the major contribution seems to be related to the ZrO2
surface. Therefore, acetaldehyde synthesized on CZA migrates to
ZrO2 and generates acetates interacting with the O of the ZrO2
lattice.
It is worth stressing that, the role of the aldol condensation in the
acetone synthesis is not yet clear. Some authors [1] proposed that
acetaldol, the precursor of crotonaldehyde, is oxidized to acetone
and CO2 by the catalyst support in this case ZrO2. Our DRIFTS stud-
ies were able to identify crotonaldehyde. Nevertheless, there is no
clear indication of the presence of acetaldol on the catalyst surface.
Despite the fact that ZrO2 shows O mobility at high temperatures,
it is not an oxidant catalyst. Indeed, ZrO2 is able to synthesize ace-
tone together with CZA (Fig. 2). Therefore, in the case of CZA + ZrO2
this oxidation mechanism seems not to be very promising. Besides,
some authors [2] suggest that acetaldol is dehydrogenated to 1,3
dicarbonyl, which undergoes a cleavage that generates acetic acid
and acetaldehyde or acetone and formic acid. They suggest that
acetic acid undergoes condensation to acetone and CO2. They sup-
port this mechanism by the presence of formic and acetic acid in the
products and the observation of 1,3 dicarbonyl by FTIR. However,
the results obtained in this study do not show formic and acetic
acid neither 1,3 dicarbonyl. Therefore, as suggested by the results
depicted in Fig. 1 and 4, the aldol condensation should be a parallel
reaction of the acetate condensation.
Fig. 9d and Fig. 9c also show vibrations around 1528 and
1309 cm−1. Their intensities increase at temperatures higher than
200 ◦C. According to the data shown in Fig. 7 these bands can be
assigned to carbonates on ZrO2. Carbonates on CZA should also be
considered (1551, 1534, 1428, 1345 cm−1, [25]). However, compar-
ing Fig. 7 with Fig. 9, the contribution of ZrO2 seems to be much
more relevant. Thus, the acetates on the ZrO2 surface condensate
and generate acetone and CO2. The carbonates are obtained by the
interaction of CO2 with the basic sites of ZrO2. Moreover, the acetate
condensation occurs above 200 ◦C (see Fig. 1).
Taking the results obtained in this work into account, Table 2
shows the main steps of the mechanism suggested for the genera-
tion of acetone from ethanol in one-step.
Fig. 9 also depicts small bands around 1760 cm−1, 1725 cm−1
,
1686 cm−1, 1663 cm−1 and 1626 cm−1. The band at 1760 cm−1
(ꢀ(C O)), which is observed at low temperature, can be associ-
ated with acetaldehyde. The absorptions at 1725 cm−1 (ꢀ(C O))
and 1663 cm−1 (ꢀ(C C)), observed at temperatures higher than
200 ◦C, can be assigned to crotonaldehyde adsorbed species. Tiny
bands at 2704 cm−1 and 2743 cm−1 (not shown) are related to
CH3 vibrations which confirm the presence of acetaldehyde and
crotonaldehyde [26–28]. The band at 1626 cm−1 (ꢀ(C C)) might
tentatively be assigned to olefins (1-butene and/or ethylene)
4. Conclusion
All in all, these results show that acetone can be produced in a
one-pot synthesis from ethanol according to the following steps:
ethoxide species are generated on ZrO2 and also on CZA by the
adsorption of ethanol; the species adsorbed on the oxide migrate
to CZA; the ethoxide species on CZA are dehydrogenated; acetalde-
hyde migrates by spillover or desorption/adsorption from CZA to
ZrO2; acetates are generated by the interaction of the O of the ZrO2
lattice with acetaldehyde; the acetates species react producing ace-
tone, hydrogen and CO2; acetone desorbs, whereas CO2 desorbs
or reacts with the ZrO2 basic sites generating carbonates; finally,
water is dissociated on Cu producing O and OH species which
migrate and reoxidize ZrO2. Acetaldehyde is an intermediate of
both the aldol and acetate condensation. The former generates
1-butene at high temperatures whereas the latter produces ace-
tone at lower temperatures. These results show that the CZA + ZrO2
adsorbed on Cu◦ [29]. As a large shoulder is observed at 1587 cm−1
,
the rate of the olefins synthesis might increase at 300 ◦C. Above
300 ◦C, two bands at 1686 and 1169 cm−1 (Fig. 8) can be observed.
These absorptions can be attributed to acetone adsorbed. [30,31].
The IR analysis corroborates with the mechanism proposed
above. Taking the results obtained into account, Table 2 shows the
main steps of the mechanism suggested for the generation of ace-
tone from ethanol in one-step. It is interesting to observe that, the
ketonization reaction generates O (Table 2, X). However, more O is
necessary for the total reoxidation of ZrO2 surface.

118
C.P. Rodrigues et al. / Applied Catalysis A: General 458 (2013) 111–118
physical mixture is a promising system for the one-pot acetone
synthesis from ethanol. They also confirm that physical mixtures
of catalysts can be a valuable tool in order to study the mechanism
of multi-step reactions.
[12] D.L. Carvalho, R.R. de Avillez, M.T. Rodrigues, L.E.P. Borges, L.G. Appel, Appl.
Catal., A 415–416 (2012) 96–100.
[13] M. Tamura, K. Shimizu, A. Satsuma, Appl. Catal., A 433–434 (2012) 135–145.
[14] M. León, E. Díaz, S. Ordónez, Catal. Today 164 (2011) 436–442.
[15] W. Dai, L. Ren, in: G. Ertl, H. Knozinger, F. Schuth, J. Weitkamp (Eds.), Handbook
of Heterogeneous Catalysis, vol. 7, 2nd ed., Wiley–VCH, Weinheim, 2008, pp.
3256–3265 (Chapter 14).
[16] R.O. Kagel, R.G. Greenler, J. Chem. Phys. 49 (1968) 1638–1647.
[17] H. Abe, K. Maruya, K. Domen, T. Onishi, Chem. Lett. 11 (1984) 1875–1878.
[18] Y. Yamada, M. Segawa, F. Sato, T. Kojima, S. Sato, J. Mol. Catal. A 346 (2011)
79–86.
[19] K. Parida, H.K. Mishra, J. Mol. Catal. A 139 (1999) 73–80.
[20] A.A. Phatak, W.N. Delgas, F.H. Ribeiro, W.F. Schneider, J. Phys. Chem. C 113
(2009) 7269–7276.
[21] B. Voss, N.C. Schjødt, J.D. Grunwaldt, S.I. Andersen, J.M. Woodley, Appl. Catal.,
A 402 (2011) 69–79.
Acknowledgment
We are grateful to eng. Lilia Fernanda Cardoso Souza Ribeiro for
technical support.
References
[1] T. Nishiguchi, T. Matsumoto, H. Kanai, K. Utani, Y. Matsumura, W.-J. Shen, S.
Imamura, Appl. Catal., A 279 (2005) 273–277.
[2] R.S. Murthy, P. Patnaik, P. Sidheswaran, M. Jayamani, J. Catal. 109 (1988)
298–302.
[3] T. Nakajima, H. Nameta, S. Mishima, I. Matsuzaki, K. Tanabe, J. Mater. Chem. 4
(6) (1994) 853–858.
[4] T. Nakajima, T. Yamaguchi, K. Tanabe, J. Chem. Soc. Chem. Commun. (1987)
394–395.
[5] T. Nakajima, K. Tanabe, T. Yamaguchi, I. Matsuzaki, S. Mishima, Appl. Catal. 52
(1989) 237–248.
[6] H. Idriss, E.G. Seebauer, J. Mol. Catal. A 152 (2000) 201–212.
[7] A. Yee, S.J. Morrison, H. Idriss, J. Catal. 186 (1999) 279–295.
[8] A.B. Gaspar, F.G. Barbosa, S. Letichevsky, L.G. Appel, Appl. Catal., A 380 (2010)
113–117.
[9] P.C. Zonetti, J. Celnik, S. Letichevsky, A.B. Gaspar, L.G. Appel, J. Mol. Catal. A 334
(2011) 29–34.
[10] D.C.D. da Silva, S. Letichevsky, L.E.P. Borges, L.G. Appel, Int. J. Hydrogen Energy
37 (2012) 8923–8928.
[22] D. Ciuparu, F. Bozon-Verduraz, L. Pfefferle, J. Phys. Chem.
3434–3442.
B 106 (2002)
[23] C. Binet, M. Daturi, Catal. Today 70 (2001) 155–167.
[24] M. Singh, N. Zhou, D.K. Paul, K.J. Klabunde, J. Catal. 260 (2008) 371–379.
[25] R. Yang, Y. Fu, Y. Zang, N. Tsubaki, J. Catal. 228 (2004) 23–35.
[26] V.V. Ordomsky, V.L. Sushkevich, I.I. Ivanova, J. Mol. Catal. A 333 (2010) 85–93.
[27] J.C.D. Jesús, F. Zaera, Surf. Sci. 430 (1999) 99–115.
[28] H. Idriss, C. Diagne, J.P. Hindermann, A. Kiennemann, M.A. Barteau, J. Mol. Catal.
A 152 (2000) 201–212.
[29] G. Hübner, G. Rauhut, H. Stoll, E. Roduner, Phys. Chem. Chem. Phys. 4 (2002)
3112–3121.
[30] R.P. Young, N. Sheppard, J. Catal. 7 (1967) 223–233.
[31] R.M. Silverstein, G.C. Bassler, T.C. Morril, Spectrometric Identification of Organic
Compounds, fifth ed., Wiley, New York, 1991.
[32] J. Llorca, N. Homs, P.R. De la Piscina, J. Catal. 227 (2004) 556–560.
[33] J.M. Guil, N. Homs, J. Llorca, P.R. De la Piscina, J. Phys. Chem. B 109 (2005)
10813–10819.
[34] F. Rößner, in: G. Ertl, H. Knozinger, F. Schuth, J. Weitkamp (Eds.), Handbook
of Heterogeneous Catalysis, vol. 3, 2nd ed., Wiley-VCH, Weinheim, 2008, pp.
1574–1585 (Chapter 5).
[11] F.S. Ramos, A.M.D. Farias, L.E.P. Borges, J.L. Monteiro, M.A. Fraga, E.F. Sousa-
Aguiar, L.G. Appel, Catal. Today 101 (2005) 39–44.