Total oxidation of ethanol over Au/Ce0.5Zr0.5O2 cordierite monolithic catalysts

Article abstract of DOI:10.1016/j.apcata.2016.05.004

The aim of this work was to propose the methods for gold introduction during the preparation of monolithic catalysts and to investigate their effect on catalyst properties. Two types of catalysts were prepared: (i) monoliths washcoated with gold/ceria-zirconia powder, and (ii) gold deposited on the monoliths washcoated with ceria-zirconia powder. An important part of the work was the characterization of the catalysts, in particular Au particle size and redox properties. Catalytic performance and selectivity were evaluated using ethanol gas-phase oxidation. It was shown that the enhanced reducibility of the catalysts with higher Au dispersion leads to improved catalytic performance.

Full text of DOI:10.1016/j.apcata.2016.05.004

Applied Catalysis A: General 522 (2016) 130–137
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Total oxidation of ethanol over Au/Ce0.5Zr0.5O2 cordierite monolithic
catalysts
Pavel Topkaa,∗, Mariana Klementováb
a Institute of Chemical Process Fundamentals of the Czech Academy of Sciences, v. v. i., Rozvojová 2/135, 165 02 Praha 6, Czech Republic
b University of West Bohemia, New Technologies − Research Centre, 306 14 Plzenˇ, Czech Republic
a r t i c l e i n f o
a b s t r a c t
Article history:
The aim of this work was to propose the methods for gold introduction during the preparation of mono-
lithic catalysts and to investigate their effect on catalyst properties. Two types of catalysts were prepared:
(i) monoliths washcoated with gold/ceria-zirconia powder, and (ii) gold deposited on the monoliths
washcoated with ceria-zirconia powder. An important part of the work was the characterization of the
catalysts, in particular Au particle size and redox properties. Catalytic performance and selectivity were
evaluated using ethanol gas-phase oxidation. It was shown that the enhanced reducibility of the catalysts
with higher Au dispersion leads to improved catalytic performance.
Received 15 December 2015
Received in revised form 2 May 2016
Accepted 3 May 2016
Available online 4 May 2016
Keywords:
Oxidation
Volatile organic compounds
Gold
© 2016 Elsevier B.V. All rights reserved.
Ceria-zirconia mixed oxide
Ethanol
Monoliths
1. Introduction
lined by Scirè and Liotta [8]. The support could not only affect the
amount of gold anchored on the surface, the size and the shape
Volatile organic compounds (VOC) are known as one of the
major contributors to air pollution. They may have short and long-
term adverse health effects, act as greenhouse gases and control
the rate of oxidant formation during the production of ground-level
ozone, which is the primary constituent of the photochemical smog
[1,2]. Catalytic oxidation is regarded as an efficient, cost-effective
and environmentally friendly way to treat VOC emissions [3]. It
is due to the fact that the use of catalysts leads to lower reac-
tion temperatures and avoids the generation of dangerous reaction
by-products like NOx [4].
The most commonly used VOC oxidation catalysts are based on
supported noble metals or transition metal oxides. Metal oxides,
mainly of Co, Cu, Ni and Mn, have the advantage of having lower cost
and greater resistance to deactivation by poisoning [5]. However,
noble metal catalysts are generally preferred because of their high
activity, excellent stability and good selectivity to CO2 [6].
Since the pioneering work by Haruta et al. [7], gold catalysts sup-
ported on different metal oxides have been reported to be efficient
catalysts for oxidation reactions. The importance of the nature of
the support for the catalytic activity of gold catalysts was under-
of gold particles, but also participate in the reaction pathway when
reducible oxides are used. Ceria is a very suitable support because of
its ability to maintain a high gold dispersion and to take part in the
oxidation reaction providing active oxygen species via Mars-van
Krevelen mechanism [9]. The high activity of gold/ceria catalysts
towards oxidation reactions has been attributed to the enhanced
reactivity of CeO2 surface oxygens induced by the metal, which
causes a decrease in the strength of the surface Ce-O bonds adjacent
to the active metal atoms thus leading to a higher surface lattice
oxygen mobility of ceria [10]. The oxidation activity has been found
to depend on both the capacity of the oxide to provide active lattice
oxygens and the oxygen activation on the active metal [11].
More recently, ceria-zirconia mixed oxides were proposed as
supports for the VOC oxidation catalysts [12,13]. It was demon-
strated that the partial substitution of cerium with zirconium
leads to improved redox properties, higher oxygen storage capac-
ity and increased catalytic activity at low temperatures [14]. We
have shown that depending on the type of VOC, the introduc-
tion of gold can increase catalytic performance and/or selectivity
of ceria-zirconia catalysts in the oxidation of chlorobenzene,
dichloromethane, ethanol and toluene [15–17].
These previous studies were carried out using powder catalysts.
Nevertheless, the implementation of catalysts in industry requires
the use of suitable supports, such as ceramic or metallic honey-
∗
Corresponding author.
E-mail address: topka@icpf.cas.cz (P. Topka).
http://dx.doi.org/10.1016/j.apcata.2016.05.004
0926-860X/© 2016 Elsevier B.V. All rights reserved.

P. Topka, M. Klementová / Applied Catalysis A: General 522 (2016) 130–137
131
comb monoliths, due to high pressure drop of powder catalysts at
high flow rates. Martínez Tejada et al. [18] recently demonstrated
that ceria-based gold catalysts can be successfully deposited on fer-
ritic stainless steel and aluminium monoliths. However, ceramic
cordierite monoliths are often preferred over metallic due to high
mechanical stability and low thermal expansion coefficient, which
resulted in their successful application in automotive catalytic con-
verters and in industry. To the best of our knowledge, ceria or
ceria-zirconia supported gold catalysts washcoated on cordierite
monoliths have not yet been studied.
The aim of this work was to investigate the preparation of
cordierite monolithic catalysts with gold supported on ceria-
zirconia as an active phase. The Au/Ce0.5Zr0.5O2 catalysts were
prepared via direct anionic exchange method, which appears to
be particularly interesting as it leads to highly active oxidation cat-
alysts. Washcoating with ready-made gold/ceria-zirconia catalysts
and deposition of gold onto ceria-zirconia washcoat were com-
pared and the catalytic performance in ethanol total oxidation was
correlated with physicochemical properties of the catalysts.
powder diffraction (XRD), transmission electron microscopy (TEM),
X-ray photoelectron spectroscopy (XPS), N2 physisorption, tem-
perature programmed reduction by hydrogen (H2-TPR), and
temperature programmed desorption (TPD) of NH3 and CO2.
Chemical analysis of catalysts was done in ALS Czech Republic
using an ICP-AES analyzer.
Powder X-ray diffraction data were obtained using a Bruker D8
Discover powder diffractometer equipped with a LynxEye detector
and a primary double-crystal Si monochromator providing CuK␣1
radiation. The data were refined using the TOPAS software. LaB6
(NIST standard reference material #660) was employed as a refer-
ence material for modelling the instrumental broadening.
Transmission electron microscopy (TEM) and scanning TEM
(STEM) was carried out on a JEOL JEM 2200FS microscope operating
at 200 kV (autoemission Shotky gun, point resolution 0.19 nm) with
an in-column energy filter, a HAADF detector, and an EDX silicon
drift detector Oxford Instruments X-Max attached. Images were
recorded on a Gatan CCD camera with resolution 2048 × 2048 pix-
els using the Digital Micrograph software package. EDX analyses
were acquired and treated in the INCA software package. Pow-
der samples were dispersed in ethanol and the suspension was
treated in ultrasound for 5 min. A drop of very dilute suspension
was placed on a holey-carbon-coated copper grid and allowed to
dry by evaporation at ambient temperature.
2. Experimental
2.1. Catalyst preparation
XPS spectra were recorded on a Kratos ESCA 3400 photo-
electron spectrometer equipped with a polychromatic Mg X-Ray
source (Mg K␣, 1253.4 eV) under the pressure of 5 × 10−7 Pa. Before
the measurement, the samples were sputtered with Ar+ ions at
500 V with 10 mA current for 30 s to remove superficial layers. All
binding energies were calculated taking as reference the C–(C,H)
component of the C 1s peak fixed at 284.8 eV. The overlapping
spectral features were resolved into individual components using
the damped non-linear least squares method and the lines of
Gaussian-Lorentzian shape. Prior to fitting the Shirley background
was subtracted.
N2 physisorption on catalysts was performed using Micromerit-
ics ASAP 2020 instrument after drying at 105 ◦C under 1 Pa
vacuum for 24 h. The adsorption-desorption isotherms of nitrogen
at −196 ◦C were treated by the standard Brunauer–Emmett–Teller
(BET) procedure to calculate the specific surface area SBET . The sur-
face area of mesopores Smeso and the volume of micropores Vmicro
were determined by t-plot method using Lecloux-Pirard standard
isotherm [21]. The total pore volume Vtotal was determined from
the amount of nitrogen adsorbed at nitrogen relative pressure
p/p0 = 0.99.
Temperature-programmed reduction (TPR) measurements of
the catalysts (0.025 g) were performed with a H2/N2 mixture
(10 mol% H2), flow rate 50 mL min−1 and linear temperature
increase 20 ◦C min−1 up to 1 000 ◦C. A change in H2 concentra-
tion was detected with a mass spectrometer Omnistar 300 (Pfeiffer
Vacuum). Reduction of the grained CuO (0.16–0.32 mm) was per-
formed in each experiment to calculate absolute values of hydrogen
consumed during reduction.
Temperature-programmed desorption (TPD) of NH3 and CO2
was carried out to examine acid and basic properties of the cat-
alysts surface. The measurements were accomplished with 0.1 g
of a catalyst in the temperature range 20–1000 ◦C using flow rate
20 mL min−1, helium as a carrier gas and NH3 or CO2 as an adsorb-
ing gas. Prior to the measurement, each sample was calcined in
helium at 500 ◦C, then cooled to 30 ◦C and an excess of NH3 or CO2
(ten doses, 840 ␮L each) was applied on the sample. Then, the sam-
ple was flushed with helium for 1 h to remove physically adsorbed
NH3 or CO2 and after that heating rate of 20 ◦C min−1 was applied.
A change in NH3 or CO2 concentration was detected with a mass
spectrometer Omnistar 300 (Pfeiffer Vacuum). During the experi-
The cylindrical cordierite monoliths of diameter 26 mm, length
20 mm, and average weight 6.32 g, with 200 cpsi and square chan-
nels (Céramiques Techniques Industrielles, France) were calcined
in a batch furnace in static air at 500 ◦C for 4 h before further treat-
ments.
The washcoated monoliths were prepared by manually con-
trolled dipping of a monolith in a slurry composed of diluted nitric
acid solution (1.4 wt.%) and 15 wt.% of ceria-zirconia mixed oxide
(Ce0.5Zr0.5O2, < 50 nm particle size (BET), Aldrich) [19]. Typically,
several dips were needed to obtain the desired amount of washcoat
loading. In each cycle the excess slurry was removed by blowing
air through the channels and then the samples were dried at 120 ◦C
and calcined in air at 300 ◦C for 4 h. The monoliths washcoated with
ceria-zirconia mixed oxide were labelled as CeZr.
Powder gold catalyst was prepared by the direct anionic
exchange method of gold species with hydroxyl groups of the sup-
port [20]. 2.25 × 10−4 M aqueous solution of HAuCl4 was heated
up to 70 ◦C and the ceria-zirconia support was introduced in the
amount corresponding to nominal Au loading 2 wt.%. After 1 h
of thermostating and vigorous stirring, the suspension was cen-
trifuged and the catalyst was washed with 4 M ammonia solution at
25 ◦C for 1 h. After drying in an oven at 120 ◦C overnight, the catalyst
was calcined in air at 300 ◦C for 4 h. Finally, the prepared powder
catalyst was washcoated on the cordierite monolith as described
above. The monoliths washcoated with gold powder catalyst were
labelled as AuCeZr.
Alternatively, gold was deposited on the monolith already wash-
coated with ceria-zirconia mixed oxide, which was prepared using
the above-described procedure. 2.25 × 10−4 M aqueous solution of
HAuCl4 was heated up to 70 ◦C and a monolith with ceria-zirconia
washcoat was introduced. The nominal Au loading was 2 wt.%. After
1 h of thermostating and vigorous stirring, the prepared catalyst
was removed and washed with 4 M ammonia solution at 25 ◦C for
1 h. After drying in an oven at 120 ◦C overnight, the catalyst was
calcined in air at 300 ◦C for 4 h. These catalysts were labelled as
Au/CeZr.
2.2. Characterization
The catalysts were characterized by atomic emission spec-
troscopy with inductively coupled plasma atomic (ICP-AES), X-ray

132
P. Topka, M. Klementová / Applied Catalysis A: General 522 (2016) 130–137
ments the following mass contributions m/z were collected: 2-H2,
16-NH3, 18-H2O, 28-N2, 32-O2, 44-N2O/CO2, and 46-NO2.
2.3. Catalytic experiments
Catalytic reaction was carried out in a glass reactor in the tem-
perature range from 50 to 300 ◦C (the furnace temperature was
linearly increased with the rate of 2 ◦C min−1). The monolithic cat-
alyst was examined at 20 m3 kg−1 h−1 space velocity without any
pretreatment. Temperature was measured inside the reactor before
and after the monolith and the average temperature was taken as
catalyst temperature. The inlet concentration of ethanol in the air
was 0.8 g/m3. Before the experiment the catalyst was kept under
the feed stream until the outlet concentration became constant.
Reaction mixture was analyzed by a gas chromatograph Hewlett-
Packard 6890 and concentrations of CO and CO2 were monitored
using a Siemens Ultramat 23 infrared analyzer. Conversion C and
selectivity to carbon dioxide S were calculated as given elsewhere
[15]. Temperature T50 (the temperature at which 50% conversion
was observed) was employed as a measure of catalyst performance.
The catalytic activity of gold catalysts was also assessed in terms
of turn-over frequency (TOF). The TOF was calculated in moles of
converted ethanol per mol of surface gold atoms employing the
average Au particle size determined by electron microscopy. The
TOF was evaluated at 60 ◦C, the ethanol conversion was 5.1% for
the AuCeZr catalyst (space velocity 20 m3 kg−1 h−1) and 12.5% for
the Au/CeZr catalyst (space velocity 75 m3 kg−1 h−1). The appar-
ent activation energy was calculated for ethanol conversions lower
than 30% using the Arrhenius plot ln(ln(1-C)/-1) = f(1/T). Reaction
by-products were identified using GC–MS analysis and the catalyst
selectivity was evaluated as the selectivity to CO2 at 95% and 99%
conversion (S95 and S99, respectively).
Fig. 1. X-ray diffraction patterns of CeZr (A), AuCeZr (B) and Au/CeZr (C) catalysts.
both gold catalysts. The mean size of Au crystallites determined
using Scherrer equation was 23 nm for the AuCeZr catalyst. In the
case of the Au/CeZr catalyst, the gold crystallites were not detected.
Adsorption of nitrogen at −196 ◦C showed that the adsorption
isotherms of the support and catalysts are similar and correspond
to the combination of I and IV type isotherms according to IUPAC
classification [24], which is typical for samples containing both
micropores and mesopores. From Table 1 it is seen that after the
deposition of gold surface area of the catalysts decreased (from
56 m2/g to 36–38 m2/g), which is in line with earlier studies [20].
AuCeZr and Au/CeZr catalysts were studied by STEM to measure
the size distribution of Au nanoparticles and observe their disper-
sion on ceria-zirconia support. Loading of Au nanoparticles in both
catalysts is so low that selected-area electron diffraction (SAED)
did not reveal their presence. In the TEM mode, the detection is
complicated mainly by diffraction contrast. This is illustrated in
Fig. 2A and B, where Au nanoparticles can be only hardly discerned.
Therefore, the HRTEM images were acquired (Fig. 2C, D) to distin-
guish Au nanoparticles from Ce0.5Zr0.5O2 by the spacing of lattice
fringes. Moreover, Fig. 3 shows the STEM results, where contrast
depends on the atomic number by the factor of 3/2 (Zr = 40, Ce = 58,
Au = 79). Here, the particles can be clearly discerned. In addition,
every particle counted in the particle analysis was checked by
EDX spectroscopy. The particle size distribution histograms were
plotted based on measurements of 100 Au particles in each sam-
ple. Maximum length of particles was measured as well as the
corresponding perpendicular length. As the particles are mostly
isometric (similar results of size distribution were obtained for
both catalysts), the geometric average corresponding to the diam-
eter of a circle with same area is shown in Fig. 3 together with
HAADF images illustrating the dispersion of Au nanoparticles. Both
samples display lognormal size distribution. The AuCeZr catalyst
contains Au nanoparticles of an average size of 13.2 nm with a wide
size distribution (from 3.4 to 34.8 nm) while the Au particles from
the Au/CeZr catalyst display a narrow size distribution (from 2.1 to
11.6 nm, with the exception of one particle with the diameter of
17.3 nm) and the average size of 6.7 nm. Moreover, the Au particles
3. Results
The Au content in the prepared catalysts determined by ICP-AES
analysis was 1.56 wt.% Au for the AuCeZr catalyst and 1.76 wt.% Au
for the Au/CeZr catalyst.
X-ray powder diffractograms of parent ceria-zirconia mixed
oxide and both types of gold catalysts are shown in Fig. 1. The
XRD data are consistent with that reported for CexZr1-xO2 solid
solutions [22]. The well-known ceria-zirconia phase diagram indi-
cates that with the addition of zirconia, cubic ceria is progressively
transformed into tꢀꢀ, tꢀꢀ, t, m phase with metastable pyrochlore-like
structures. In comparison with pure CeO2 (diffractions at 28.7◦,
33.3◦, 47.8◦ and 56.8◦), the diffraction lines in Fig. 1 are shifted
to higher angles (29.2◦, 33.8◦, 48.6◦, and 57.8◦) due to the incorpo-
ration of Zr. The position of the (111) diffraction line of the cubic
phase at 29.2◦ corresponds to the composition of the ceria-zirconia
solid solution Ce0.5Zr0.5O2 [22]. For the homogeneous Ce0.5Zr0.5O2
the structure is tꢀꢀ, i.e. cubic for the cation position but the oxygens
are displaced from their regular position. However, the splitting of
diffraction lines at higher angles (Fig. 1) points to the presence of at
least two solid solutions with slightly different cell constants [23].
The mean size of ceria-zirconia crystallites was not influenced by
the addition of gold and was 11 nm for parent ceria-zirconia and
Table 1
Physicochemical properties of investigated catalysts.
Catalyst Smeso
(m2 g−1
SBET
Vtotal
(cm3
H2-TPRa
NH3-TPDa
(mmol g−1
CO2-TPDa
(mmol g−1
)
Ce 3d5/2 (V)
(eV)
Au 4f7/2
(eV)
−1
)
(m2 g−1
)
g
)
(mmol g−1
)
)
liq
CeZr
AuCeZr 24
Au/CeZr 28
37
56
38
36
0.227
0.164
0.195
0.88
0.85
1.02
0.06
0.08
0.11
0.12
0.08
0.10
885.4 (29%) Ce3+902.1 (71%) Ce4+
–
885.4 (25%) Ce3+902.1 (75%) Ce4+ 84.2
885.4 (27%) Ce3+902.1 (73%) Ce4+ 84.3
a
corresponding to total peak area between 25 and 900 ◦C.

P. Topka, M. Klementová / Applied Catalysis A: General 522 (2016) 130–137
133
Fig. 2. Electron microscopy images of AuCeZr (A, C) and Au/CeZr (B, D) catalysts.
are evenly dispersed in both samples with a higher concentration
of particles in the Au/CeZr catalyst.
tion of ceria-zirconia with gold did not change the characteristics
of the support. AuCeZr and Au/CeZr catalysts exhibited similar
binding energy of Au 4f7/2 (84.2 and 84.3 eV, respectively), which
corresponds to metallic gold [26]. In both cases, the Au 4f7/2 peak
possessed additional shoulders at 85.3 and 85.4 eV, respectively.
Quantitative fits led to a composition of about 88% metallic and
12% of partially oxidized Au species for the AuCeZr catalyst; 73%
metallic and 27% of partially oxidized Au species was observed in
the Au/CeZr catalyst. The atomic concentration of Au was 2.0% in
the AuCeZr catalyst and 3.2% in the Au/CeZr catalyst.
The acidity of investigated catalysts was studied using
temperature-programmed desorption of ammonia (NH3-TPD). The
desorption features from CeZr, AuCeZr and Au/CeZr catalyst are
shown in Fig. 5. Desorption of ammonia proceeded mainly in the
temperature range from 25 to 300 ◦C. It shows that the catalysts
possess only weak acid sites. The surface species responsible for this
acidity are coordinatively unsaturated Ce4+ and Zr4+ ions, which
possess an uncompensated positive charge and create Lewis acidity
[25]. For all catalysts, a distinct desorption peak can be recognized
at 102 ◦C with a shoulder at ∼184 ◦C. Thus, both gold catalysts
as well as pure ceria-zirconia exhibited similar strength of acid
sites. The amount of acid sites corresponding to total peak area
between 25 and 900 ◦C in the NH3-TPD profiles is shown in Table 1.
Ammonia adsorbs on the surface of the catalysts, therefore it may
be more correct to compare the amount of acid sites taking into
account the different surface area of the catalysts SBET . The acid
site density (amount of NH3 adsorbed per m2 of catalyst surface)
The reducibility of the catalysts was examined by temperature-
programmed reduction using hydrogen as a reducing agent (Fig. 4).
The pristine ceria-zirconia showed two peaks centred at 353 ◦C and
513 ◦C that can be ascribed to the reduction of surface cerium and
bulk cerium, respectively. In the case of the AuCeZr catalyst, both
low-temperature and high-temperature reduction peak of cerium
are shifted to lower temperatures (213 ◦C and 273 ◦C, respectively).
This shows that the reduction of Ce4+ was promoted by gold
nanoparticles. Even larger shift of the two reduction peaks to lower
temperatures can be seen with the Au/CeZr catalyst, exhibiting a
main reduction peak at 161 ◦C with a shoulder at 91 ◦C. This cata-
lyst was practically reduced in a single step, which points to a high
mobility of the lattice oxygen ions [25]. The extent of the reduction
was larger for the Au/CeZr catalyst with a H2 uptake of 1.02 mmol/g,
while the AuCeZr and CeZr catalysts exhibited similar values (0.85
and 0.88 mmol/g, respectively, Table 1). This shows a marked effect
of preparation method on the reducibility of ceria-zirconia support,
which is connected with Au particle size, as evidenced by electron
microscopy (Fig. 3). The reduction of the support surface strongly
anchored with finely dispersed Au particles occurs at lower tem-
perature and the reduction of the support being in contact with
larger gold agglomerates and/or unaffected by Au, occurs at higher
temperature [20].
The surface composition of the catalysts was analyzed by XPS.
The results are summarized in Table 1. The catalysts exhibited
amount of Ce3+ between 25 and 29%. It is seen that the modifica-

134
P. Topka, M. Klementová / Applied Catalysis A: General 522 (2016) 130–137
Fig. 3. Au particle size distribution and HAADF images of AuCeZr (A, B) and Au/CeZr (C, D) catalysts.
Fig. 4. H2-TPR profiles of Au/CeZr (A), AuCeZr (B) and CeZr (C) catalysts.
Fig. 5. NH3-TPD profiles of Au/CeZr (A), AuCeZr (B) and CeZr (C) catalysts.
increased in the order CeZr < AuCeZr < Au/CeZr (0.0010 mmol/m2,
0.0020 mmol/m2 and 0.0029 mmol/m2, respectively).
and 900 ◦C in the CO2-TPD profiles is shown in Table 1. The density
of basic sites (amount of CO2 adsorbed per m2 of catalyst surface)
increased in the order CeZr ∼ AuCeZr < Au/CeZr (0.0022 mmol/m2,
0.0020 mmol/m2 and 0.0029 mmol/m2, respectively).
The catalytic performance of the catalysts in the total oxidation
of ethanol is summarized in Table 2. In the case of monoliths wash-
coated with ceria-zirconia mixed oxide, the temperature of 50%
conversion (T50) determined by light-off experiments decreased
linearly with increasing washcoat loading up to ∼15 wt.%. The cata-
lysts with washcoat loading 19.5 and 21.5 wt.% exhibited the same
The basicity of the catalysts was characterized by temperature-
programmed desorption of carbon dioxide (CO2-TPD). The patterns
of CO2 desorption from CeZr, AuCeZr and Au/CeZr catalyst are
shown in Fig. 6. The CeZr and Au/CeZr catalysts exhibited a des-
orption peak at 110 ◦C. With the AuCeZr catalyst, a small shift of
the main desorption peak to 126 ◦C was revealed. Nevertheless, all
studied catalysts possessed only basic sites of low strength. The
amount of basic sited corresponding to total peak area between 25

P. Topka, M. Klementová / Applied Catalysis A: General 522 (2016) 130–137
135
Fig. 6. CO2-TPD profiles of Au/CeZr (A), AuCeZr (B) and CeZr (C) catalysts.
Fig. 7. Evolution of by-products during ethanol oxidation over the Au/CeZr catalyst:
♦ ethanol, ꢁ ethylene, ᭹ acetaldehyde, ᮀ acetic acid, ꢀ ethyl acetate.
Table 2
Performance of catalysts in ethanol oxidation (0.8 g/m3 in air, space velocity
20 m3 kg−1 h−1, temperature ramp 2 ◦C min−1).
dation over Au/TiO2 catalysts with gold particle size 4.0 nm was
significantly shifted to lower temperatures in comparison with
that for 9.1 nm. In terms of turn-over frequency, the Au/CeZr cat-
alyst with ∼ 20 wt.% washcoat loading would be approximately
four times more active than its AuCeZr analogue (the TOF was
1.3 × 10−2 s−1 and 3.2 × 10−3 s−1, respectively). However, it should
be noted that the TOF was computed in moles of converted ethanol
per mol of surface gold atoms, i.e. the contribution to the activ-
ity from the bare support was neglected; moreover, not all surface
Au atoms but only perimeter of Au particles may be active in the
reaction [30,31]. Finally, the selectivity of the Au/CeZr catalysts to
CO2 was comparable with the AuCeZr catalysts and the same by-
products were observed, with the exception of carbon monoxide.
The detailed distribution of by-products is shown in Fig. 7.
Catalyst
Washcoat (wt.%)
T50(◦C)
T90(◦C)
S95(%)
S99(%)
CeZr
CeZr
CeZr
CeZr
4.1
8.8
13.3
19.5
21.9
5.5
274
256
240
233
233
142
129
112
113
120
80
303
281
271
267
262
210
196
198
204
201
133
125
84
86
88
90
95
35
18
54
59
55
44
47
97
97
98
98
98
57
55
71
74
71
76
73
CeZr
AuCeZr
AuCeZr
AuCeZr
AuCeZr
AuCeZr
Au/CeZr
Au/CeZr
9.2
14.7
16.5
20.0
20.9
21.1
74
T50 (233 ◦C). Selectivity to CO2 can be even more important than
the catalytic performance due to the fact that some by-products
formed during the oxidation may be more detrimental to the envi-
ronment than the initial compounds. For example, acetaldehyde
and/or acetic acid can be formed during ethanol oxidation [27]. It
is known that over noble metal catalysts, ethanol oxidation may
follow two parallel reaction paths − direct oxidation to CO2, and
partial oxidation to acetaldehyde, which is then oxidized to CO2
[28]. In ethanol oxidation over CeZr catalysts, acetaldehyde and
acetic acid were observed as main by-products, together with traces
of ethylene and carbon monoxide. The selectivity to CO2 at 95%
conversion increased with increasing washcoat loading (from 84
to 95%). At 99% conversion, the selectivity to carbon dioxide was
97–98% (Table 2).
With the AuCeZr catalysts (monoliths washcoated with
gold/ceria-zirconia powder), the catalytic performance increased
linearly with increasing washcoat loading up to ∼15 wt.%. For the
AuCeZr catalyst with 20 wt.% washcoat loading, the T50 was 120 ◦C,
i.e. 113 ◦C less than for its CeZr analogue. However, the selectivity to
CO2 was much lower due to the formation of various by-products.
Acetaldehyde and ethylene were detected as main by-products,
together with small amounts of acetic acid and ethyl acetate and
traces of carbon monoxide. Nevertheless, it should be noted that
the selectivity reported in Table 2 was determined using light-off
experiments and that at prolonged reaction times much higher
selectivity to CO2 can be achieved (89–100%).
4. Discussion
The AuCeZr and Au/CeZr catalysts were prepared by the direct
anionic exchange method of gold species with hydroxyl groups
of the support. The nominal Au loading and Au concentration in
the impregnation solution were the same for both types of the
catalysts. Thus, the different Au particle size distribution (larger
Au particles observed in the AuCeZr catalyst) might be connected
with some other detail of catalysts preparation. In order to eluci-
date the possible effect of nitric acid solution that was used during
the preparation of the AuCeZr catalysts, the following experiment
was done. The Au/CeZr catalyst was washed for 2 h with diluted
nitric acid solution (1.4 wt.%) under dynamic conditions, dried at
120 ◦C overnight and calcined at 300 ◦C for 4 h. After that a catalytic
test was made. The difference between the standard Au/CeZr cat-
alyst and that subjected to washing with acid solution was in the
frame of the experimental error (the T50 was by 4 ◦C lower). There-
fore, it can be concluded that the use of diluted nitric acid solution
during the catalysts preparation was not the factor which resulted
in different properties of the AuCeZr catalysts. Another aspect of
the preparation procedure that might influence Au particle size
distribution can be the mass transport resistance of the washcoat
layer. For the preparation of the powder AuCeZr catalyst, a very
fine ceria-zirconia powder was used (particle size < 50 nm by BET).
Therefore, a high fraction of the support’s surface was immedi-
ately accessible for the anionic exchange with gold anions from the
impregnation solution. On the other hand, when a monolith already
washcoated with ceria-zirconia was immersed into the impregna-
tion solution, it can be expected that initially only a superficial layer
of the ceria-zirconia washcoat would be readily accessible for the
The Au/CeZr catalysts prepared by depositing of gold on the
monolith already washcoated with ceria-zirconia exhibited better
catalytic performance than the catalysts prepared using gold/ceria-
zirconia powder (Table 2). In comparison with AuCeZr catalysts, the
T50 was by 40 ◦C lower. Similar results were reported by Santos
et al. [29], who showed that the light-off curve of ethanol oxi-

136
P. Topka, M. Klementová / Applied Catalysis A: General 522 (2016) 130–137
Fig. 9. Arrhenius plot for ethanol oxidation over AuCeZr (᭹) and Au/CeZr (᭿) cata-
lysts.
Fig. 8. Correlation between T50 in ethanol oxidation and temperature of surface
cerium reduction in H2-TPR experiments for the Au/CeZr, AuCeZr and CeZr catalysts.
anionic exchange with gold anions, while the exchange in the bulk
of the washcoat would be hindered by mass transport resistances.
By H2-TPR measurements (Fig. 4) it was shown that the addi-
tion of Au to ceria-zirconia significantly promoted the reduction
of surface cerium. Moreover, there was a considerable difference
between H2-TPR profiles of AuCeZr and Au/CeZr catalysts. The
H2-TPR peaks of Au/CeZr catalyst are shifted to lower temper-
atures in comparison with the AuCeZr catalysts (Fig. 4). Thus,
there is a good correlation between H2-TPR profiles and catalytic
performance of investigated catalysts (Fig. 8). The temperature
of the main reduction peak of the catalysts increased in the
order Au/CeZr > AuCeZr > CeZr (161, 213 and 353 ◦C, respectively).
At the same time, the catalytic performance in ethanol oxi-
dation decreased in the same order (the T50 was 80, 120 and
233 ◦C, respectively). A similar correlation between reducibility
and catalytic performance was observed for ethanol oxidation over
Pt/Ce0.5Zr0.5O2 catalysts [32].
It was demonstrated earlier that the catalytic performance
in alcohol oxidation can be influenced by noble metal loading
and/or particle size [33–35]. Minicò et al. [36] concluded that
the high activity of Au/Fe2O3 catalyst in ethanol oxidation was
related to the capacity of highly dispersed gold to weaken the Fe–O
bond thus increasing the mobility of the lattice oxygen which is
involved in ethanol oxidation probably through a Mars–van Kreve-
len reaction mechanism. Similarly, Solsona et al. [37,38] attributed
the improved catalytic performance of the Au/Co3O4 catalysts in
ethanol oxidation to the enhancement of the redox properties of
the support oxide induced by gold due to the introduction of sur-
face defects at the gold/oxide interface, involved in VOC activation.
Recently, Holz et al. [39] proposed two types of active sites for
ethanol oxidation over Au/TiO2: (i) the perimeter of Au nanoparti-
cles, and (ii) the support. They reported that the presence of highly
dispersed Au nanoparticles induced a new oxidation pathway. The
second type of active sites located on the support was likely to
follow the Mars–van Krevelen mechanism. The presence of Au
nanoparticles also increased the activity of the support due to its
enhanced reducibility. Similarly to the Au/TiO2 catalysts, Mars–van
Krevelen mechanism implying lattice oxide ions as the active oxy-
gen species together with oxygen activation on the perimeter of
Au nanoparticles is likely to apply also to the ethanol oxidation
over the investigated Au/Ce0.5Zr0.5O2 catalysts. A more detailed
study is needed to deconvolute the contribution to the catalytic
performance from Au particles and from the support.
ter of all nanoparticles in the Au/CeZr catalyst would be 4.4 times
larger than in the AuCeZr catalyst. Consequently, a higher number
of active sites would be available on the perimeter of Au nanoparti-
cles and the metal-support interaction would be more pronounced,
which is in line with obtained H2-TPR and catalytic results. The
Au/CeZr catalyst with average Au particle size of 6.7 nm exhib-
ited lower temperature of surface cerium reduction (161 ◦C) than
the AuCeZr catalyst with average size of Au nanoparticles 13.2 nm
(213 ◦C). Dobrosz-Gómez et al. [20] reported earlier that the reduc-
tion of the support surface strongly anchored with finely dispersed
Au particles occurs at lower temperature than the reduction of
the support being in contact with larger gold agglomerates. More-
over, the H2-TPR experiments suggest that the Au/CeZr catalyst
not only has a lower reduction temperature but also appeared to
have improved oxygen storage capacity (evidenced by the higher
H2 consumption). While the AuCeZr and CeZr catalysts exhibited
similar values of hydrogen consumption (0.85 and 0.88 mmol/g,
respectively, Table 1), the Au/CeZr catalyst reached 1.02 mmol/g.
It is known that the improved oxygen storage capacity may con-
tribute to higher catalytic performance of the catalysts. However,
from Fig. 8 it is seen that the catalytic performance correlated
with reducibility of the catalysts, while the effect of higher oxy-
gen storage capacity of the Au/CeZr catalyst in comparison with
the other catalysts was not pronounced. Concluding, the exam-
ples of CeZr and AuCeZr catalysts showed that the enhanced redox
properties appear to promote the oxidation rates significantly.
Therefore, it can be assumed that the significantly higher catalytic
activity observed for the Au/CeZr catalysts is mainly connected
with the further enhanced surface redox properties of these cat-
alysts. In order to disclose the possible difference in the nature of
active sites, the Arrhenius plot was constructed for the AuCeZr and
Au/CeZr catalysts with ∼ 20 wt.% washcoat loading (Fig. 9). How-
ever, the apparent activation energy was the same for both types
of the catalysts (84 kJ/mol). This points to the same nature of the
active sites in both AuCeZr and Au/CeZr catalysts. Therefore, the
increased catalytic activity of the Au/CeZr catalyst may be ascribed
to improved redox properties due to the presence of highly dis-
persed Au nanoparticles, which possess a higher number of active
sites as discussed above.
It was shown earlier that acid-base properties of VOC oxidation
catalysts represent an important parameter for their catalytic per-
formance. The changes in acidity and basicity of the catalyst surface
can influence adsorption-desorption equilibria of reactants and/or
reaction mechanism [40]. Recently, it was proposed that the Lewis
acidity of the metal center is a key factor influencing the perfor-
mance of some types of catalysts in ethanol oxidation [41]. On the
other hand, the acid sites usually decrease the selectivity to CO2
If we consider the average size of Au nanoparticles in the present
Au/Ce0.5Zr0.5O2 catalysts, hemispherical shape of the particles and
actual Au loading, the Au/CeZr catalyst would contain 8.6 times
more Au particles than the AuCeZr catalyst, and the total perime-

P. Topka, M. Klementová / Applied Catalysis A: General 522 (2016) 130–137
137
due to the formation of by-products (mainly acetaldehyde) and a
References
decreased acidity thus may lead to improved selectivity [42]. In our
case, all studied catalysts exhibit the same strength of acid sites. The
density of acid sites increased in the order CeZr < AuCeZr < Au/CeZr
(0.0010 mmol/m2, 0.0020 mmol/m2 and 0.0029 mmol/m2, respec-
tively). The increased amount of acid sites after gold introduction
can be attributed to coordinatively unsaturated Ce4+ and Zr4+ ions,
which are in contact with gold nanoparticles. On the other hand, the
AuCeZr and Au/CeZr catalysts exhibited similar selectivity to CO2
(S99 was 71–76%), while the CeZr catalyst was much more selective
(S99 = 98%). Thus, the differences in selectivity among the studied
catalysts can be tentatively ascribed to different reaction mech-
anism over gold catalysts in comparison with pure Ce0.5Zr0.5O2,
while the effect of catalyst acidity would be negligible. Concerning
the basicity of the catalysts, it was reported that increased basicity
can positively influence the activity of the catalyst due to strong
adsorption of the oxidized alcohol [40]. In the present case, all
studied catalysts exhibited similar strength of basic sites. The den-
sity of basic sites increased in the order CeZr ∼ AuCeZr < Au/CeZr
(0.0022 mmol/m2, 0.0020 mmol/m2 and 0.0029 mmol/m2, respec-
tively), while the catalytic performance increased in the order
CeZr < AuCeZr < Au/CeZr (the T50 was 233, 120 and 80 ◦C, respec-
tively). However, taking into account the different nature of the
studied catalysts (the presence vs. the absence of gold, the differ-
ent Au particle size in the case of gold catalysts), the possible effect
of amount of basic sites on catalytic activity cannot be judged.
[1] R.G. Derwent, M.E. Jenkin, S.M. Saunders, M.J. Pilling, P.G. Simmonds, N.R.
Passant, G.J. Dollard, P. Dumitrean, A. Kent, Atmos. Environ. 37 (2003)
1983–1991.
[2] M.M. Galloway, P.S. Chhabra, A.W.H. Chan, J.D. Surratt, R.C. Flagan, J.H.
Seinfeld, F.N. Keutsch, Atmos. Chem. Phys. 9 (2009) 3331–3345.
[3] F.I. Khan, A.K. Ghoshal, J. Loss Prev. Process Ind. 13 (2000) 527–545.
[4] A.C. Gluhoi, B.E. Nieuwenhuys, Catal. Today 119 (2007) 305–310.
[5] P.O. Larsson, A. Andersson, Appl. Catal. B 24 (2000) 175–192.
[6] J.J. Spivey, Ind. Eng. Chem. Res. 26 (1987) 2165–2180.
[7] M. Haruta, T. Kobayashi, H. Sano, N. Yamada, Chem. Lett. 16 (1987) 405–408.
[8] S. Scirè, L.F. Liotta, Appl. Catal. B 125 (2012) 222–246.
[9] S. Scirè, S. Minicò, C. Crisafulli, C. Satriano, A. Pistone, Appl. Catal. B 40 (2003)
43–49.
[10] Q. Fu, A. Weber, M. Flytzani-Stephanopoulos, Catal. Lett. 77 (2001) 87–95.
[11] S. Minicò, S. Sciré, C. Crisafulli, R. Maggiore, S. Galvagno, Appl. Catal. B 28
(2000) 245–251.
[12] J.I. Gutiérrez-Ortiz, B. de Rivas, R. López-Fonseca, J.R. González-Velasco, Appl.
Catal. A 269 (2004) 147–155.
[13] J.I. Gutierrez-Ortiz, B. de Rivas, R. Lopez-Fonseca, J.R. Gonzalez-Velasco, Appl.
Catal. B 65 (2006) 191–200.
[14] M. Boaro, C. de Leitenburg, G. Dolcetti, A. Trovarelli, J. Catal. 193 (2000)
338–347.
ˇ
[15] J. Gaálová, P. Topka, L. Kaluzˇa, O. Solcová, Catal. Today 175 (2011) 231–237.
[16] L. Mateˇjová, P. Topka, L. Kaluzˇa, S. Pitkäaho, S. Ojala, J. Gaálová, R.L. Keiski,
Appl. Catal. B 142 (2013) 54–64.
[17] P. Topka, R. Delaigle, L. Kaluzˇa, E.M. Gaigneaux, Catal. Today 253 (2015)
172–177.
[18] L.M. Martínez Tejada, M.I. Domínguez, O. Sanz, M.A. Centeno, J.A. Odriozola,
Gold Bull. 46 (2013) 221–231.
[19] L.F. Liotta, A. Longo, G. Pantaleo, G. Di Carlo, A. Martorana, S. Cimino, G. Russo,
G. Deganello, Appl. Catal. B 90 (2009) 470–477.
[20] I. Dobrosz-Gómez, I. Kocemba, J.M. Rynkowski, Appl. Catal. B 88 (2009) 83–97.
[21] A. Lecloux, J.P. Pirard, J. Colloid Interface Sci. 70 (1979) 265–281.
[22] L. Lan, S. Chen, Y. Cao, M. Zhao, M. Gong, Y. Chen, J. Colloid Interface Sci. 450
(2015) 404–416.
5. Conclusions
[23] L.F. Liotta, A. Macaluso, A. Longo, G. Pantaleo, A. Martorana, G. Deganello,
Appl. Catal. A 240 (2003) 295–307.
[24] S.J. Gregg, K.S.W. Sing, Adsorption, Surface Area and Porosity, Academic Press
New York, 1982.
[25] B. de Rivas, R. López-Fonseca, M.A. Gutiérrez-Ortiz, J.I. Gutiérrez-Ortiz, Appl.
Catal. B 104 (2011) 373–381.
[26] R. Leppelt, B. Schumacher, V. Plzak, M. Kinne, R.J. Behm, J. Catal. 244 (2006)
137–152.
[27] G. Avgouropoulos, E. Oikonomopoulos, D. Kanistras, T. Ioannides, Appl. Catal.
B 65 (2006) 62–69.
[28] A.A. Barresi, G. Baldi, Chem. Eng. Comm. 123 (1993) 31–42.
[29] V.P. Santos, S.A.C. Carabineiro, P.B. Tavares, M.F.R. Pereira, J.J.M. Órfão, J.L.
Figueiredo, Appl. Catal. B 99 (2010) 198–205.
[30] L.M. Petkovic, S.N. Rashkeev, D.M. Ginosar, Catal. Today 147 (2009) 107–114.
[31] M.S. Avila, C.I. Vignatti, C.R. Apesteguía, T.F. Garetto, Chem. Eng. J. 241 (2014)
52–59.
Cordierite monolithic catalysts with gold nanoparticles sup-
ported on Ce0.5Zr0.5O2 were prepared either by washcoating with
powder gold/ceria-zirconia (catalyst AuCeZr) or by depositing of
gold on the monolith washcoated with ceria-zirconia (catalyst
Au/CeZr). The AuCeZr catalyst possessed wide Au particle size
distribution centred at 13.2 nm, while the Au/CeZr catalyst exhib-
ited narrow Au particle size distribution centred at 6.7 nm. The
gold catalysts were compared with pure Ce0.5Zr0.5O2. In gas-phase
oxidation of ethanol, the monolith washcoated only with ceria-
zirconia converted 50% of ethanol at 233 ◦C. The AuCeZr catalyst
reached the same conversion at 120 ◦C and the Au/CeZr catalyst at
80 ◦C. A good correlation was observed between H2-TPR profiles
and catalytic performance of the catalysts. The superior catalytic
performance of the Au/CeZr catalyst was ascribed to the presence of
highly dispersed Au nanoparticles, which enhanced the reducibility
of the catalyst.
[32] P. Topka, L. Kaluzˇa, J. Gaálová, Chem. Pap. 70 (2016) 898–906.
ˇ
[33] K. Soukup, D. Petrásˇ, P. Topka, P. Slobodian, O. Solcová, Catal. Today 193
(2012) 165–171.
ˇ
[34] K. Soukup, P. Topka, V. Hejtmánek, D. Petrásˇ, V. Valesˇ, O. Solcová, Catal. Today
236 (2014) 3–11.
ˇ
[35] L. Mateˇjová, P. Topka, K. Jirátová, O. Solcová, Appl. Catal. A 443 (2012) 40–49.
[36] S. Minicò, S. Scirè, C. Crisafulli, R. Maggiore, S. Galvagno, Appl. Catal. B 28
(2000) 245–251.
[37] B.E. Solsona, T. Garcia, C. Jones, S.H. Taylor, A.F. Carley, G.J. Hutchings, Appl.
Catal. A 312 (2006) 67–76.
Acknowledgements
The financial support of the Grant Agency of the Czech Republic
(project No. 13-24186P) is gratefully acknowledged. The authors
thank Dr. P. Dytrych and M. Kosˇtejn for XPS analysis and Dr.
K. Jirátová for TPR and TPD measurements. M. K. would like to
acknowledge the financial support from the CENTEM project, Reg.
No. CZ.1.05/2.1.00/03.0088, cofunded by the ERDF as part of the
Ministry of Education, Youth and Sports OP RDI programme and,
in the follow-up sustainability stage, supported through CENTEM
PLUS (LO1402) by financial means from the Ministry of Education,
Youth and Sports under the National Sustainability Programme I.
[38] B. Solsona, E. Aylón, R. Murillo, A.M. Mastral, A. Monzonís, S. Agouram, T.E.
Davies, S.H. Taylor, T. Garcia, J. Hazard. Mater. 187 (2011) 544–552.
[39] M.C. Holz, K. Tolle, M. Muhler, Catal. Sci. Technol. 4 (2014) 3495–3504.
[40] R. Beauchet, P. Magnoux, J. Mijoin, Catal. Today 124 (2007) 118–123.
[41] P. Gonzalez-Navarrete, M. Schlangen, X.N. Wu, H. Schwarz, Chem. Eur. J. 22
(2016) 3077–3083.
[42] J. Ludvíková, K. Jirátová, J. Klempa, V. Boehmová, L. Obalová, Catal. Today 179
(2012) 164–169.