Co3O4 particles grown over nanocrystalline CeO2: Influence of precipitation agents and calcination temperature on the catalytic activity for methane oxidation

Article abstract of DOI:10.1039/c4cy01158a

Crystalline cobalt oxides were prepared by a precipitation method using three different precipitation agents, (NH₄)₂CO₃, Na₂CO₃ and CO(NH₂)₂. Cobalt oxide nanoparticles corresponding to a Co₃O₄ loading of 30 wt% were also deposited over high-surface area nanocrystalline ceria by the same precipitation agents. The effect of calcination temperature, 350 or 650 °C, on the morphological and structural properties was evaluated. Characterization by BET, XRD, SEM, TEM, Raman spectroscopy, H₂-TPR, XPS and NH₃-TPD was performed and the catalytic properties were explored in the methane oxidation reaction. The nature of the precipitation agent strongly influenced the textural properties of Co₃O₄ and the Co₃O₄-CeO₂ interface. The best control of the particle size was achieved by using CO(NH₂)₂ that produced small and regular crystallites of Co₃O₄ homogeneously deposited over the CeO₂ surface. Such a Co₃O₄-CeO₂ system precipitated by urea showed enhanced low-temperature reducibility and high surface Co³⁺ concentration, which were identified as the key factors for promoting methane oxidation at low temperature. Moreover, the synergic effect of cobalt oxide and nanocrystalline ceria produced stable full conversion of methane in the entire range of investigated temperature, up to 700-800°C, at which Co₃O₄ deactivation usually occurs.

Full text of DOI:10.1039/c4cy01158a

View Article Online
View Journal
Catalysis
Sc ience &
Technology
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: H. Wu, G.
Pantaleo, G. Di Carlo, S. Guo, G. Marcì, P. Concepcion, A. M. Venezia and L. F. Liotta, Catal. Sci. Technol.,
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/catalysis

Page 1 of 38
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01158A
Co O4 particles grown over nanocrystalline CeO : influence of
3
2
precipitation agent and calcination temperature on catalytic activity for
methane oxidation.
1, 2
1
3
1,2
4
5
H. Wu , G. Pantaleo , G. Di Carlo , S. Guo , G. Marcì , P. Concepción , A. M.
1
1
Venezia and L. F. Liotta *
1
Istituto per lo Studio dei Materiali Nanostrutturati (ISMN), Consiglio Nazionale delle Ricerche
(CNR), Via Ugo La Malfa 153, 90146, Palermo, Italy.
2
Department of Applied Physics, Northwestern Polytechnical University (NPU), Youyi Xilu 127,
7
10072, Xi’an, Shaanxi, PR China.
3
Istituto per lo Studio dei Materiali Nanostrutturati (ISMN), Consiglio Nazionale delle Ricerche
(CNR), Via Salaria km 29300, 00015 Monterotondo Stazione, Roma, Italy.
4
“SchiavelloꢀGrillone”Photocatalysis Group. Dipartimento di Energia, Ingegneria dell’informazione
e modelli Matematici (DEIM), Università di Palermo, Viale delle Scienze, 90128 Palermo, Italy.
5
Instituto de Tecnología Química, UPVꢀCSIC, Campus de la Universidad Politécnica de Valencia,
Avenida de los Naranjos s/n, 46022 Valencia, Spain.
2 3 4
Keywords: nanocrystalline CeO ; precipitation method; control of Co O crystallite size; reduction
properties; methane oxidation activity.
*
Corresponding author. Tel.: +39 091 6809371; fax: +39 091 6809399.
E-mail address: liotta@pa.ismn.cnr.it (L.F. Liotta).
1

Catalysis Science & Technology
Page 2 of 38
View Article Online
DOI: 10.1039/C4CY01158A
Abstract
Crystalline cobalt oxides were prepared by precipitation method using three different precipitation
agents, (NH CO , Na CO and CO(NH . Cobalt oxide nanoparticles corresponding to Co
4
)
2
3
2
3
2
)
2
3 4
O
loading of 30 wt% were also deposited over highꢀsurface area nanocrystalline ceria by the same
precipitation agents. The effect of calcination temperature, 350 or 650 °C, on the morphological and
structural properties was evaluated. Characterizations by BET, XRD, SEM, TEM, Raman
2 3
spectroscopy, H ꢀTPR, XPS and NH ꢀTPD were performed and the catalytic properties were
explored in the methane oxidation reaction. The nature of precipitation agent strongly influenced
the textural properties of Co O and the Co O ꢀCeO interface. The best control of the particle size
3
4
3
4
2
was achieved by using CO(NH
2
)
2
that produced small and regular crystallites of Co
3 4
O
homogeneously deposited over CeO
2
surface. Such Co ꢀCeO system precipitated by urea
3
O
4
2
3+
showed enhanced lowꢀtemperature reducibility and high surface Co concentration, which were
identified as the key factors for promoting methane oxidation at lowꢀtemperature. Moreover, the
synergic effect between cobalt oxide and nanocrystalline ceria produced stable full conversion of
methane in the entire range of investigated temperature, up to 700ꢀ800 °C, at which Co O
3
4
deactivation usually occurs.
1
. Introduction
Cobalt oxide catalysts have been extensively investigated during the last two decades as possible
substitutes to precious metals for their catalytic properties in several oxidation reactions for waste
gases abatement [1ꢀ11] and diesel soot combustion as well [12, 13]. Recently a growing scientific
attention has been paid to Co
abatement of unburned methane released from natural gas vehicles (NGVs). To this aim, different
combinations Co on various oxide supports have been proposed [5, 14, 15]. Highly dispersed
Co particles promoted by ZrO have been claimed as active sites in methane oxidation over lowꢀ
content cobalt catalysts [5]. Nanosized Co /γꢀAl catalysts with defective structure, prepared
3 4
O ꢀbased oxides as highly efficient catalysts for the complete
3
O
4
3
O
4
2
3
O
4
2 3
O
by a combination of wetness impregnation and combustion synthesis, exhibited high activity in the
total oxidation of methane [14]. The promotional effect in methane oxidation of manganese doped
cobalt oxides was ascribed to crystal defections into the spinel structure of Co O , which probably
3
4
2+
caused an increased amount of Co ions responsible for catalytic activity [15].
CoO ꢀCeO systems have been widely addressed in the last years for different applications,
depending on the CoꢀCe composition, preparation method and on the nature of preꢀtreatments [8,
x
2
1
6ꢀ19]. We have reported that Co
O
3 4
ꢀCeO
2
coprecipitated oxides are effective catalysts for methane
2

Page 3 of 38
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01158A
oxidation in stoichiometric oxygen concentration and the best activity was attained in
correspondence of Co/Ce atomic ratio ∼1 [20, 21]. In this sample, the presence of highly dispersed
Co
3 4 2
O particles in good contact with CeO was claimed as the key factor for the improved redox
properties and catalytic performances. High lattice oxygen mobility and formation of surface
anionic vacancies in such Co
activity [22, 23].
3 4 2
O ꢀCeO system were also relevant to propene and toluene oxidation
On these grounds, Co O ꢀbased catalysts are definitely good candidate for oxidation reactions,
3
4
however, deactivation processes induced by highꢀtemperature, such as loss of surface area, thermal
decomposition to the less active CoO and interactions of cobalt with the matrix leading to inactive
phases (i.e., CoAl O , CoTiO , CoZr O0.69), limit its application to high temperatures [5, 21, 24,
2
4
3
3
2
5].
Therefore, deposition of Co
3
O
4
active sites over a high surface area carrier is attractive for the
achievement of high catalytic performances and longꢀterm stability.
It has been reported that the properties of cerium oxide can change drastically when the particle size
is decreased to the nanosize region, turning into a very active support strongly interacting with gold
[
26, 27]. It was, therefore, clear to us that such nanocrystalline ceria would be a good candidate for
depositing Co nanoparticles.
The preparation of Co nanocrystals with different morphology is widely documented, the
morphology strongly depending on the experimental conditions, reaction temperature, nature of
precipitation agent [11, 28ꢀ31]. Synthesis of Co nanoparticles with small diameter, in the range
0ꢀ40 nm, has been reported by homogeneous precipitation of cobaltꢀhydroxideꢀcarbonate with
3
O
4
3 4
O
3
O
4
2
urea at 110 °C starting from cobalt nitrate precursor [29]. The synthesis procedures allowed to
control also crystals shape and crystals orientation plane which have been claimed as key factors in
CO and CH
The present work focuses on the effect of the precipitation agent, Na
on the nature of cobalt species, cobaltꢀhydroxide carbonate and/or Co
powder at 120 °C. Their successive thermal conversion to crystalline Co O spinel and the influence
4 3 4
oxidation activity of Co O nanocrystals [11, 31].
2
CO
3 4 2 3 2 2
, (NH ) CO , CO(NH ) ,
3
O
4
, detected after drying the
3
4
of the calcination temperature and duration time, 350 °C for 3 h or 650 °C for 5 h, on the physicꢀ
chemical properties was investigated. The influence of nanocrystalline ceria as support for cobalt
species growing and crystallization was also investigated preparing by sodiumꢀ, ammonium
carbonate and urea, three mixed oxides with composition Co
3
O
4
(30wt%)ꢀCeO
2
. The so obtained
oxides were characterized by BET, XRD, SEM, TEM, Raman spectroscopy, H
2
ꢀTPR, XPS and
NH ꢀTPD techniques. The catalytic properties were evaluated in the methane oxidation in
3
stoichiometric oxygen concentration.
3

Catalysis Science & Technology
Page 4 of 38
View Article Online
DOI: 10.1039/C4CY01158A
2
. Experimental
All chemicals (SigmaꢀAldrich) were of analytical grade and were used as received without further
purification.
2
.1. Synthesis of nanocrystalline CeO
2
A colloidal dispersion of CeO
2
3 4
nanoparticles was prepared by thermolysis of an acidified Ce(NO )
solution followed by reꢀdispersion, according to a previously described method [26, 27]. The
dispersion consisting of discrete CeO nanoparticles that display a spherical morphology was
2
purified and concentrated using an ultraꢀfiltration cell equipped with 3KD membrane. The
purification was monitored by the residual acidity of the dispersion, determined by an acid titration
of the supernatant after ultraꢀcentrifugation at 50000 rpm for 6 h. After evaporation in air at room
temperature, the resulting solid sample was calcined at 500 °C for 6 h using a programmable oven,
ꢀ1
heating in air at a rate of 1 °C min was performed in order to favour complete crystallization of the
fluorite structure, according to the published procedure [26, 27].
2
.2. Synthesis of Co
Bare Co oxides were prepared by precipitation method with three different precipitation agents,
Na CO , (NH CO and CO(NH , starting from Co(NO ⋅6H O, as cobalt precursor. In a typical
preparation for obtaining 2 g of Co oxide, Na CO or (NH CO solution (0.5 M) was added
drop by drop, under mild heating at 80 °C, to the proper volume of cobalt nitrate water solution
1M) until pH = 8.5. The as formed precipitate was aged overnight in the mother liquor at the
3 4 3 4 2
O and Co O (30wt%)-CeO oxides
3
O
4
2
3
4
)
2
3
2
)
2
3
)
2
2
3
O
4
2
3
4
)
2
3
(
constant temperature of 80 °C. The third preparation was carried out using homogeneous
precipitation method by urea hydrolysis under stirring at 90 °C overnight. In this method urea in
excess (molar ratio: urea/cobalt nitrate= 100) was added. The obtained solid, labeled as CoNa, CoNH4
and Courea, was filtered and washed with distilled water several times until neutrality. Then it was
dried overnight at 120 °C in oven, then divided into two portions and calcined in static air at 350 °C
ꢀ1
for 3 h or at 650 °C for 5 h, respectively, (heating rate 3°C min ) in order to form crystalline Co O
3
4
spinel.
3 4
By the same precipitation method with three agents, Co O with loading 30wt% was deposited over
nanocrystalline CeO . Drying at 120 °C and calcination at 350 °C and at 650 °C were performed as
2
above described. The obtained oxides were labeled as Co30CeNa, Co30CeNH4 and Co30Ceurea
,
respectively. For all samples herein discussed the drying or calcination temperature is added as
subscript to the name label (i.e. CoNa350 and Co30CeNa350 mean bare Co O and
3
4
Co
3
O
4
(30wt%)/CeO
2
precipitated by Na
2
CO
3
solution and calcined at 350 °C).
4

Page 5 of 38
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01158A
2
.3. Characterization
Specific surface area (SSA) measurements were performed using a Sorptomatic 1900 Carlo Erba
Instrument, by physical adsorption of N at the temperature of liquid nitrogen (ꢀ196 °C), using the
BET method in the standard pressure range of 0.05ꢀ0.30 p/p [32].
2
0
Xꢀray diffraction (XRD) patterns were recorded with a D 5005 XꢀRay Diffractometer (SIEMENS)
using Cu Kα radiation coupled with a graphite monochromator. The assignment of the various
crystalline phases was based on the Inorganic Crystal Structure database (ICSD, FIZ Karlsruhe)
[
3 4
33]. Mean crystallite size (d) of Co O phase was calculated from the line broadening of the most
intense reflection using the Scherrer equation [34]. The estimated error was ±10%.
Scanning electron microscopy (SEM) images and energyꢀdispersive Xꢀray (EDX) analyses were
obtained using a FEI quanta 200 ESEM microscope, operating at 20 kV on specimens after being
coated with a thin layer of gold.
TEM images were obtained with a Philips CM10 microscope operating at 100 kV. A portion of
2 2
grounded solid was suspended in CH Cl and sonicated for 1 min. Then, a drop was extracted under
sonication and deposited on the copper grid and dried before observation.
Raman spectra were acquired with a Renishaw Raman In Via spectrometer equipped with a Leika
DM LM microscope and a 785 nm HPNIR diode laser as an excitation source. The laser power at
the sample was 30mW. A 50x objective of 8 mm optical length was used to focus the depolarized
laser beam onto a 3–5 µm spot on the sample surface and to collect the backscattered light. The
ꢀ
1
Raman scattering was collected in a staticꢀscan mode in the 100–3000 cm spectral region with a
ꢀ1
resolution better than 4 cm . Ten scans were accumulated for each spectrum during a total scanning
time of 100 s.
Temperatureꢀprogrammed reduction (TPR) experiments were carried out with a Micromeritics
Autochem 2910 apparatus equipped with a thermal conductivity detector (TCD). In the TPR
ꢀ1
experiments, a gas mixture H
2
(5 vol.% in Ar, 30 ml min ) was flowed over the sample preꢀtreated
with O (5 vol.% in He) at 550 °C for 30 min, heating from room temperature to 1000 °C (rate 10
2
ꢀ1
°
C min ). In order to avoid any influence of mass transfer effects on the shape of the TPR profiles,
the absence of interꢀ and intraꢀparticle mass transfer limitations was accurately checked by
employing appropriate K values.
The acidity of oxide catalysts was determined by temperatureꢀprogrammed desorption of ammonia
3
(NH ꢀTPD) experiments by using a Micromeritics Autochem 2910 apparatus equipped with a
thermal conductivity detector (TCD) and a mass quadrupole spectrometer (Thermostar, Balzers).
The sample amount of 0.3 g was preꢀtreated in He flow at 200 °C for 1 h. Then, after cooling down
5

Catalysis Science & Technology
Page 6 of 38
View Article Online
DOI: 10.1039/C4CY01158A
to room temperature, ammonia adsorption was performed by admitting a flow of 5% NH
3
/He
ꢀ1
stream (30 ml min ) for 1 h. In order to remove all the physically adsorbed ammonia, the sample
ꢀ1
was purged by flowing 100 ml min He at 100 °C for 1 h, then after cooling down to room
ꢀ1
temperature, ammonia desorption was started by flowing He (30 ml min ) and heating up to 500 °C
ꢀ1
(
rate of 10 ºC min ), holding time at 500 °C for 30 min. All the gases desorbed from the sample
were monitored by online TCD and QMS analysis. The total acidity of the catalyst was calculated
by integration of the NH desorption profile referred to the QM signal, at mass 17.
3
The Xꢀray photoelectron spectroscopy analyses were performed with a VG Microtech ESCA 3000
Multilab, equipped with a dual Mg/Al anode. The spectra were excited by the unmonochromatised
Al Kα source (1486.6 eV) run at 14 kV and 15 mA. The analyser operated in the constant energy
(CAE) mode. For the individual peak energy regions, a pass energy of 20 eV set across the
hemispheres was used. Survey spectra were measured at 50 eV pass energy. The sample powders
were studied as pellets, mounted on a doubleꢀsided adhesive tape. The pressure in the analysis
ꢀ8
chamber was in the range of 10 Torr during data collection. The constant charging of the samples
was removed by referencing all the energies to the C 1s set at 285.1 eV, arising from the
adventitious carbon. The invariance of the peak shapes and widths at the beginning and at the end
of the analyses ensured absence of differential charging. Analyses of the peaks were performed with
the software provided by VG, based on nonꢀlinear least squares fitting program using a weighted
sum of Lorentzian and Gaussian component curves after background subtraction according to
Shirley and Sherwood [35, 36]. Atomic concentrations were calculated from peak intensity using
the sensitivity factors provided with the software. The binding energy values are quoted with a
precision of ± 0.15 eV and the atomic percentage with a precision of ± 10%.
2
.4. Catalytic tests
Methane combustion catalytic tests were performed in stoichiometric conditions by flowing a gas
reaction mixture consisting of 0.3 vol.% of CH and 0.6 vol.% of O in He (λ =1) [20]. The
experiments were carried out over 50 mg of catalyst (fraction sieved between 180 and 250 ꢀm) at a
4
2
ꢀ1
flow rate of 50 ml min (STP), equivalent to a weight hourly space velocity (WHSV) of 60000 ml
ꢀ1
ꢀ1
g h by using a quartz U shaped reactor with a inner diameter of 12 mm, electrically heated in a
furnace. The reaction temperature was monitored by a Kꢀtype thermocouple in contact with the
catalytic bed, long 12 mm. Before the catalytic tests the samples were preꢀtreated ‘‘in situ’’ under
2
flowing O (5 vol.% in He) at 350 °C for 1/2 h. Activities were measured by increasing the
temperature from 200 to 600 or 800 °C by steps of 50 °C, hold time 30 min at the final set
temperature. Catalysts calcined at 350 °C have been tested up to 600 °C, while samples calcined at
6

Page 7 of 38
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01158A
6
50 °C were investigated up to 800 °C. The inlet and outlet gas compositions were analyzed by on
line mass quadrupole (ThermostarTM, Balzers), in order to follow the evolution of all the species,
CH , CO, CO , H , H O, O . Moreover, the concentration of CO and CO species was checked by
an IR analyser (ABB Uras 14), calibrated in the range 0–3000 ppm for CO and 0–10 000 ppm for
CO . The reaction products of methane oxidation were CO and H O, no significant CO (max 10
4
2
2
2
2
2
2
2
2
ppm) was detected in the overall range of temperature. Carbon balance was close to ± 5% in all the
catalytic tests.
3
3
. Results and discussion
.1. Structural and textural properties of catalysts
Fig. 1 shows powder XRD patterns of asꢀprepared cobalt oxides species dried at 120 °C. The weak
diffraction intensities of CoNa120 indicate that such sample prepared by precipitation with Na CO
solution is not well crystallized. The Courea120 shows a better crystallinity and peaks characteristics
of the orthorhombic cobaltꢀhydroxideꢀcarbonate phase Co(OH)1.0(CO O, in good
2
3
3 2
)0.5·0.11H
agreement with literature data [29, 31]. The hkl reflections shown in Fig. 1 can be indexed perfectly
to this phase. Surprisingly, diffraction peaks typical of the Co O fcc phase (space group Fd3m) at
3
4
2
θ = 19.0°, 31.3°, 37.0°, 38.6°, 44.9°, 55.8°, 59.4° and 65.4°, corresponding to the (111), (220),
311), (222), (400), (422), (511) and (440) crystal planes [33], were observed in the XRD pattern of
the sample CoNH4120. Lowꢀcrystalline features of cobaltꢀhydroxideꢀcarbonate species were also
identified as the main background components. In a previous work [37] where (NH CO was used
as precipitation agent, cobaltꢀhydroxideꢀcarbonate was obtained as the only compound. The
crystallization of Co occurring in our experimental conditions could be ascribed to the longer
(
4
)
2
3
3
O
4
2
+
3+
ageing procedure at 80 °C herein employed, that could favour Co →Co oxidation.
2
In Fig. 2 powder XRD patterns of the cobalt species precipitated over nanocrystalline CeO and
dried at 120 °C are displayed. Compared with the powder XRD patterns of bare cobaltꢀhydroxide
carbonate and Co
3
O
4
compounds, the presence of CeO
2
3
seems do not affect the nature of such
and CO(NH , where weakly crystalline
species, especially for samples precipitated by (NH
)
4 2
CO
2 2
)
peaks of Co O and of the orthorhombic cobaltꢀhydroxideꢀcarbonate phase were detected,
3
4
2
respectively. The pattern of Co30CeNa120 shows mainly CeO signals, although small features of
cobaltꢀhydroxideꢀcarbonate species can be seen. By comparing Figs. 1 and 2, it is worth noting the
effect of ceria in dispersing cobalt phases.
After calcination at 350 °C all cobalt phases crystallized into Co
cobalt oxides and as well for the Co30Ce samples. Calcination at higher temperature, 650 °C,
makes the Co O peaks sharper, in particular, for the Co sample. In Fig. 3a,b, powder XRD
3 4
O , as detected by XRD for bare
3
4
NH4
7

Catalysis Science & Technology
Page 8 of 38
View Article Online
DOI: 10.1039/C4CY01158A
patterns of bare cobalt oxides after calcination at 350 and 650 °C are shown. The same trend is
observed for the Co30Ce catalysts after calcinations (see Fig. S1), although the nanocrystalline
ceria plays a fundamental role in maintaining supported Co
bare oxides. These results indicate that the crystal phases (cobaltꢀhydroxideꢀcarbonate and/or
Co ) formed by precipitation method are related to the used precipitant. Moreover, the
3 4
O crystallites better dispersed than
3
O
4
crystallinity is also function of the precipitation agent, the sharpest diffraction peaks of the Co
3 4
O
spinel phase were observed for CoNH4 and for Co30Ce NH4 samples.
3 4
Mean crystallite sizes (d) of Co O phase calculated from the (311) diffraction line are shown in
Table 1 for Co and Co30Ce series calcined at 350 and 650 °C. For samples calcined at 350 °C d
values are ranging between 10 and 25 nm. For both series the calcination at 650 °C caused a
4 2 3
pronounced sintering of crystallites, in particular for samples precipitated by (NH ) CO . Those
prepared by Na CO and CO(NH ) maintained relatively smaller sizes, between 40 and 110 nm.
2
3
2 2
Nanocrystalline CeO
2
3 4
stabilized significantly Co O particles against sintering, in particular for the
Co30Ceurea650 and Co30CeNa650. Moreover, by a perusal of Table 1, it emerges a less pronounced
increase of the d values for Co30Ce650 samples as compared to the corresponding Co650 series,
3 4
especially for Co O precipitated in the presence of sodium carbonate and urea, suggesting the
highest stabilizing effect of nanocrystalline ceria for those samples.
Similar to the trend found for Co O crystallite sizes, the BET surface area values for bare cobalt
3
4
oxides and for the Co30Ce samples depend on the precipitation agent and decreased by increasing
the calcination temperature from 350 to 650 °C, as shown in Table 1. All Co30Ce samples calcined
2
at 350 °C have specific surface area higher than 100 m /g that diminished to about the half values
2
for those calcined at 650 °C. The specific surface area of the nanocrystalline CeO was equal to 198
2
2
m /g. Co
3
O
4
oxides treated at 350 °C showed surface area around 30ꢀ40 m /g and suffered stronger
2
thermal shrinkage than ceriaꢀsupported oxides, BET values decreasing to ∼5ꢀ10 m /g upon
calcination at 650 °C.
In order to have a better understanding of the morphological and structural differences between the
various samples prepared by different precipitation agents a careful microscopic investigation by
SEM and TEM analysis was performed on all the samples.
In Fig. 4 SEM images of nanocrystalline CeO
2 3 4
, Co O and Co30Ce samples calcined at 350 °C and
precipitated by (NH ) CO and by urea, respectively, are shown. CeO powder shows uniform
4
2
3
2
morphology and appears constituted of nanoparticles of ca. 25 nm size agglomerated in nearly
spherical particles with ca. 3 µm of diameter (Fig. 4a,b). The shape of bare oxides, Courea350 and
CoNH4350 is different. Courea350 consists of homogeneous nanorodꢀshaped particles with size between
3
0 to 50 nm, while for CoNH4350 inhomogeneously agglomerated nanoparticles were detected (Fig.
8

Page 9 of 38
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01158A
4
c,d). SEM observation for Co30Ceurea350 evidenced that CeO
2
nanoparticles are homogenously
covered by nanorodꢀshaped particles, indicating the precipitation by urea ensures the controlled
growth of Co nanoparticles onto the CeO surface (Fig. 4e). On the contrary, Co30CeNH4350 fails
to show a uniform distribution of Co nanoparticles over ceria (Fig. 4f). EDX analyses confirmed
3
O
4
2
3
O
4
the so far discussed data, indicating for Co30Ceurea350 that the average atomic percentages of Co is
slightly higher (59 ±10) than the nominal one (48), consequently, the average atomic percentage of
Ce (41±10) is slightly lower than the nominal value (52). However, the values are quite
reproducible on different agglomerates present in a large area of ∼ 500 ꢀm per side. Conversely Co
and Ce were quite heterogeneously distributed from agglomerate to agglomerate for Co30CeNH4350
with atomic percentages of Co and Ce highly fluctuating with respect to the nominal values. Based
on the so reported data, we have considered worthy of interest to focus the attention on the sample
Co30Ceurea and to investigate the effect of the temperature of treatment, i.e, 120, 350 and 650 °C, on
the morphological properties. In Fig. 5, the SEM micrographs are shown. After drying at 120 °C the
sample Co30Ceurea consists of well defined nanorods with diameter around 20 nm and length
ranging from 500 nm up to about 1.5 ꢀm. Such morphology is typical of cobalt hydroxide
carbonates, as previously reported [31]. The sample calcined at 350 °C maintains the rodꢀlike
morphology but the nanorods are much shorter (maximum length ∼500 nm) with particles size
between 30 to 50 nm that further increased up to 60ꢀ80 nm after calcination at 650 °C.
Fig. 6 shows the TEM images corresponding to the nanocrystalline CeO
crystallites, with an average size below 10 nm can be observed. In Fig. 7 TEM images of Courea
calcined at 350 °C and 650 °C are displayed. According with the literature [29] Co shows a
2
support. Very small
3
O
4
beadꢀlike structure with nanoparticles interconnected along the longitudinal directions of the rods.
The increase of the crystallite size when increasing the calcination temperature is clearly evidenced
3 4
and in agreement with the particle sizes (d) of Co O phase calculated by the Scherrer equation (see
Table 1).
TEM images of Co30Ce catalysts prepared by precipitation with Na CO , (NH ) CO and
2
3
4 2
3
CO(NH
quite heterogeneous, with large agglomerations of CeO
with the Co O crystallites (Fig. 8 aꢀd). This effect is more evident in the case of Co30Ce
2
)
2
and calcined at 350 °C are compared in Fig. 8. Co30CeNa and Co30CeNH4 appear to be
2
particles, which are not directly interacting
3
4
NH4
characterized by bigger Co
Raman results below reported.
When the precipitation of Co O was carried out using urea with calcination at 350 °C, we can
3
O
4
crystallites (Fig. 8 c,d). Both features are in agreement with the
3
4
3 4 2
observe a much more homogeneous distribution of the Co O and CeO particles, a better
interaction between these oxides and small particle size (15ꢀ25 nm) (Fig. 8 e,f). The higher degree
9

Catalysis Science & Technology
Page 10 of 38
View Article Online
DOI: 10.1039/C4CY01158A
of homogeneity of these materials detected by TEM was in agreement with the Raman results. After
calcination at 650 °C larger Co crystals were obtained resulting in segregation from the ceria
support, in particular for Co30CeNH4 that sintered in crystallites bigger than 100 nm, while for
Co30Ceurea the crystal size was lower and the perimeter interface Co ꢀCeO , regarded as a
fundamental parameter in catalysis, was still clearly detectable (see Fig. S2).
To further analyze the structure of Co and Co30Ce oxides as a function of the precipitation
3
O
4
3
O
4
2
3
O
4
agent and calcinations temperature, Raman spectra have been registered. In Fig. S3 the Raman
spectra corresponding to CoNa350, CoNH4350 and Courea calcined at 350 and 650°C are displayed. In
all cases the five characteristic bands assigned to the active modes of the Co O spinel crystalline
3
4
phase [38, 39] were observed. No additional bands, corresponding to other phases, such as CoO,
were found [40]. Fig. S4 shows the spectra obtained for the Co30CeNa, Co30CeNH4 and Co30Ceurea
calcined at 350 °C. In order to investigate the effect of calcinations temperature, the spectrum of
Co30Ceurea650 is also reported. Four different spectra are presented for each sample, corresponding
to different mappings carried out during the spectra collection, in order to show the possible
heterogeneity of the materials.
3 4
Compared with Fig. S3 we can see that the bands corresponding to the spinel structure of Co O are
ꢀ1
still present, and an additional band appears at around 460 cm , slightly before, or in some cases
even overlapping the band corresponding to the E Raman active mode of Co O . This band has
g
3
4
2
been assigned to the F2g Raman active mode of the fluorite structure in CeO , and its intensity is
known to decrease significantly with Co loading [41].
In the case of the sample prepared by precipitation with Na
broad, according with the presence of small crystallites, and the spectra differ from one map to the
other (see Fig. S4a). In fact, the Co bands are almost absent in the red spectrum corresponding
to map2, whereas the green spectrum (map3) does not present the band assigned to CeO
2 3 3 4
CO , the Co O bands are considerably
3
O
4
2
.
When (NH
4
2
) CO
3
is used as precipitation agent (see Fig. S4b), sharper and more intense bands
phase, and a small CeO peak appears only in one of the maps (map4).
bands, but more intense
band at 460 cm can be hardly detected as a shoulder of
the E Co O phonon mode, and only in two of the spectra (maps 2 and 3). Based on the smaller
appear for the Co
3
O
4
2
3 4
The sample prepared by means of urea (see Fig. S4c) presents broad Co O
ꢀ1
2 3 2
than in the case of Na CO , and the CeO
g
3
4
differences between the different mappings it can be concluded that this sample is more
homogeneous. These properties are maintained after calcination of the sample Co30Ceurea650 (Fig.
S4d). The above reported Raman results confirm XRD, SEM and TEM data.
3
.2. Temperature-programmed reduction
1
0

Page 11 of 38
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01158A
High bulk oxygen mobility (through the Marsꢀvan Krevelen mechanism) and formation of highly
active oxygen species activated by the oxygen vacancies are widely recognized as the main factors
influencing the catalytic activity of Co
42]. In order to get insights into the effect of different precipitation agents and as well on the
interaction between Co and CeO on the reduction properties, H ꢀTPR experiments were
performed on the prepared Co and Co30Ce oxides. In Fig. 9 the reduction curves are shown.
With respect to the reduction process of Co , there are various controversial viewpoints. Arnoldy
and Moulijn reported one broad peak for the reduction of Co [43]. A single reduction peak at
34 °C was found also by Luo et al. [44]. However, more commonly the reduction of Co O was
3 4 4
O and CoxCe for CO, CH and VOCs catalytic oxidation
[
3
O
4
2
2
3
O
4
3 4
O
3
O
4
4
3
4
described as a twoꢀstep process involving the intermediate reduction of CoO [42]. These different
results could be explained considering that the reduction behaviour of Co is highly dependent on
3 4
O
the preparation method and dispersion of particles. Large particles are usually reduced to metallic
cobalt in a single step, while for nanoparticles often a twoꢀstep process occurs [18, 45, 46]. Two
reduction peaks, labelled as peak I and peak II, in the range of temperature ∼200ꢀ 400 °C, were
detected in the TPR profiles of the prepared Co
3 4
O oxides calcined at 350 and at 650 °C (see Fig. 9
a,b). Confirming our previous data and in agreement with other literature results, the peak I was
3
+
2+
associated with the reduction of Co to Co , and the peak II was ascribed to the reduction of CoO
to metallic cobalt [20, 47, 48]. For a given calcination temperature, 350 or 650 °C, moving from
CoNa having small particles to CoNH4 characterized by bigger crystallites, both reduction peaks
shifted to higher temperature, indicating a decreased reducibility. Moreover, increasing the
calcinations temperature from 350 to 650 °C, produced a further shift to higher temperature. Larger
3 4
Co O crystallites and lower BET surface areas (Table 1) induced by the calcinations at 650 °C can
account for the decreased reducibility. Independently on the preparation method and on the
calcination temperature, the overall hydrogen consumption related to the two peaks area (Fig. 9 a,b)
2 3 4
was equal to 370 ± 5 ml. Being the theoretical H consumption for the total reduction of Co O to
0
Co equal to ∼ 406 ml H
2
/g (calculated at 1 atm and 298 K) the measured value is within the
experimental error ±10% ).
The TPR profiles of Co30Ce oxides calcined at 350 °C (Fig. 9 c) are characterized by twoꢀthree
main reduction peaks in the range of temperature between 150 and 600 °C. An additional peak at
about 740 °C is present in all the reduction profiles. The sample Co30Ceurea exhibited the highest
reducibility at low temperature with a first peak centred at 221 °C with a shoulder at 183 °C and a
second intense peak at 453°C. For Co30CeNa the reduction occurs in a wide range of temperature,
with three broad signals, centred at 275, 334 and at 500 °C. While for Co30CeNH4, characterized by
3 4
larger Co O particles, the reduction is completed in a shorter range of temperature with three close
1
1

Catalysis Science & Technology
Page 12 of 38
View Article Online
DOI: 10.1039/C4CY01158A
peaks at 291, 345 and 382 °C. In Fig. 9 c the reduction profile for the nanocrystalline ceria is also
reported for comparison. Two main hydrogen consumptions, at 370 °C and at around 770 °C,
respectively, were detected. The low temperature peak was due to the reduction of ceria surface,
while at high temperature the reduction of the bulk takes place [49].
According to the literature [18, 44, 50, 51], the following assignments for the reduction of Co
3 4
O
can be done: small peaks below 200 °C are attributed to the reduction of surface oxygen species
3
+
adsorbed on oxygen vacancies; peaks in the range 240ꢀ320 °C are due to the reduction of Co to
2
+
Co for Co
3 4 2 3 4
O at the interface with CeO ; between 320ꢀ480 °C reduction in single step of Co O to
Co occurs for isolated clusters not interacting with ceria; finally, in the range 480ꢀ700 °C reduction
2
+
0
of Co to Co is accomplished for CoO
x
species interacting with ceria.
Moreover, it is likely that ceria reduction is in turn promoted by the interaction with Co
3
O
4
and
therefore surface reduction could overlap with Co to Co reduction, while the bulk reduction of
isolated ceria (not strongly interacting with Co ) takes place at around 740 °C. The so far listed
3
+
2+
3
O
4
assignments of reduction peaks are supported by the hydrogen consumptions listed in Table 2. The
total consumption was comparable within the experimental error for all the Co30Ce samples,
however, different volumes were consumed in the three ranges of temperature identified in Table 2.
For the sample Co30Ceurea350 the lowꢀtemperature peak corresponds to 41 ml H
2
/g that exceeds the
theoretical value of about 30 ml/g (calculated at 1 atm and 298 K) expected for the reduction of
3
+
2+
Co to Co . Only a very small hydrogen consumption was observed in the range 300ꢀ400 °C (0.8
ml/g), while the hydrogen consumed between 400ꢀ550 and 680ꢀ820 °C well accounts for the
2
+
0
reduction of Co to Co and for bulk reduction of ceria. On this basis, it could be inferred that in
3
+
Co30Ceurea350 ceria surface reduction occurs between 100ꢀ300 °C overlapping with Co reduction.
Moreover, on the basis of values in Table 2 and in agreement with our previous findings [20] it
appears that Co
at temperature as low as 400ꢀ550 °C. On the other hand the good dispersion of Co
at the interface with ceria and the absence of independent bulkꢀlike Co phase was further
confirmed by the very low intensity of the peak in the range 300ꢀ400 °C [18]. The small peak at
3
O
4
nanoparticles finely dispersed over ceria catalyse the reduction of the bulk CeO
2
3
O
4
nanoclusters
3
O
4
around 740 °C, corresponding to 2.5 ml H
ceria clusters.
2
/g has been assigned to the reduction of few isolated
For Co30CeNa350 the reduction profile can be explained as before discussed for Co30Ceurea350,
although some differences emerge from a careful examination of Fig, 9 c: the first reduction peak
(at 275 °C) is shifted to higher temperature with respect to Co30Ceurea350, moreover in the range
3
00ꢀ400 °C surface reduction of ceria, not interacting with Co , occurs.
3 4
O
1
2

Page 13 of 38
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01158A
On the other hand, for Co30CeNH4350, in the range of temperature between 300ꢀ400 °C, the
contribution to the reduction profile of ceria surface and as well of isolated Co
3 4
O crystallites is
much more pronounced than for the previous samples.
These findings suggest that for Co30Ce oxides the mean crystallite size of Co
3
O
4
phase is not the
ꢀCeO seems to
only parameter influencing the reduction profile, in fact the interaction Co
prevail.
3
O
4
2
3 2
For Co30Ce samples calcined at 650 °C, reduction features typical of Co O4, surface and bulk CeO
were singly detected (Fig. 9 d). Co reduction was achieved in the range 200ꢀ380 °C, surface and
3
O
4
bulk ceria reduction occurred at around 400ꢀ440 °C and at about 740 °C, respectively. The
hydrogen consumptions listed in Table 2 are consistent with the above assignments. The presence
of two reduction peaks at low temperature detected only for Co30Ceurea650 is in agreement with the
presence of relatively small Co O crystallites that are reduced through a twoꢀstep process.
3
4
Moreover, the hydrogen consumption registered for such sample between 200ꢀ380 °C is higher than
the theoretical one expected for complete reduction of 30wt% of Co O to metallic cobalt (∼ 122 ml
3
4
2
H /g calculated at 1 atm and 298 K), suggesting that some ceria surface reduction is overlapping
3
+
2+
0
with the process Co to Co to Co .
2 3 4
Conversely, isolated clusters not interacting with CeO and/or larger Co O crystallites, such as
those detected for Co30CeNa650 and Co30CeNH4650, give rise to TPR curves quite similar to those of
the bare Co samples calcined at the same temperature. Reduction of surface and bulk ceria
occurred with two distinct peaks between 400ꢀ500 °C and 680ꢀ820 °C.
3
O
4
3
.3. X-ray Photoelectron Spectroscopy (XPS)
In order to get more insights into the surface chemical composition of Co30Ce oxides as a function
of the preparation method and calcinations temperature, XPS analyses were carried out on samples
calcined at 350 and 650 °C. The data in terms of Co2p and Ce3d binding energies, Co/Ce atomic
3
+
ratio and relative surface content of Co are summarized in Table 3. For all samples, no extra
detectable peaks due to impurities, like residual sodium ions, were observed. For all Co30Ce
samples, the Co 2p XP spectra exhibited two main peaks at about 795.0 ± 0.2 and 780.0 ± 0.3 eV
corresponding to the Co 2p1/2 and Co 2p3/2 spinꢀorbital peaks, respectively. For the mixed valence
Co3O4 a spin−orbit splitting value of 15.2 eV has been reported [18, 20, 45]. Here the spinꢀorbit
3 4
value is of 15.0 ± 0.1 eV close to that of Co O . Moreover, the shakeꢀup satellite with a low
intensity at ca.7.0 eV from the main Co 2p3/2 spin orbit components is also characteristic of pure
Co O [52]. In Fig. S5 experimental and fitted XP spectra for two selected samples, Co30Ce
urea350
3
4
and Co30CeNH4350, are displayed as an example and are also representative of the other samples.
1
3

Catalysis Science & Technology
Page 14 of 38
View Article Online
DOI: 10.1039/C4CY01158A
As it results from Co2p3/2 binding energies listed in Table 3, no change in the oxidation state of
cobalt induced by the precipitation agent or by the calcination temperature was observed for
3
+
3+
3+
2+
Co30Ce series. The relative surface content of Co
(Co /(Co +Co ) (%) was ∼ 46% for all the
samples and this value was higher than that reported for analogous Co O /ZSMꢀ5 systems [53]. The
3
4
results in Table 3 show that the precipitation agent plays an important role on the surface
composition in terms of Co/Ce atomic ratio. For instance, Co30CeNa350 and Co30CeNH4350 have
relatively low surface Co/Ce ratio (3.8 and 1.8, respectively), while for Co30Ceurea350 a considerable
3 4
cobalt enrichment (Co/Ce = 6.3) on the surface was observed. Anyhow, according to the Co O
preparation method (namely precipitation over the ceria surface), a cobalt enrichment on the surface
occurs and the surface atomic ratios were always higher than the nominal value (Co/Ce = 0.92).
After calcinations at 650 °C, due to Co
3 4
O crystallites sintering and specific surface area shrinkage
the amount of Co detected on the surface generally decreased, less dramatically for Co30Ceurea650
.
XPS analysis was carried out also to characterize nanocrystalline ceria. The spectrum (not shown)
was typical of pure ceria spectrum [49] and was fitted with eight peaks corresponding to four pairs
of spinꢀorbit doublets arising from the multiplicity of final states reached during the Ce 3d
photoionization process [54]. In Table 3 Ce3d5/2 binding energy values are listed. Interestingly, a
3
+
4+
3+
constant Ce /( Ce + Ce ) atomic ratio equal to ∼16% was found in all samples and as well in the
3+
pure nanocrystalline ceria, analysed for comparison. This high percentage of Ce , which was not
induced by the Xꢀray exposure as checked by recording the spectra after different exposure time, is
indicative of high oxygen vacancies content likely related to the peculiar nanocrystalline structure
[
20].
3
.4. Methane oxidation tests
The methane oxidation activity of bare Co
C are shown in Fig. 10 aꢀd in terms of CH
3
O
4
and Co30Ce catalysts calcined at 350 °C and at 650
°
4
conversion (%) curves versus temperature. In Tables 4
and 5 the T50 and T90 values (the reaction temperature at which 50% or 90% of methane conversion
is reached) are listed for samples calcined at 350 and at 650 °C, respectively. Reaction rates (ꢀmol
ꢀ1
ꢀ1
CH ⋅s ⋅gcatalyst ) calculated for conversions < 20% are also reported in Tables 4 and 5. Among
4
3 4
Co O oxides calcined at 350 °C and at 650 °C, the activity increased in the order CoNH4 < Courea <
Co , the latter showing the lowest T50 and T90 temperatures for methane conversion and the
Na
highest reaction rate. XRD and BET characterization showed that CoNa has smaller crystallites and
higher surface area than the other bare oxides. Moreover, the TPR reducibility decresed moving
from CoNH4 to Co . These factors can account for the trend of activity, according with several
Na
1
4

Page 15 of 38
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01158A
literature results highlighting that crystals size, shape and crystals orientation planes are key factors
in CH oxidation activity of Co nanocrystals [5,14, 31].
4
3 4
O
Looking at Co30Ce catalysts calcined at 350 °C and at 650 °C, the catalytic activity for methane
oxidation increased in the order Co30CeNH4 < Co30CeNa < Co30Ceurea (Tables 4 and 5) and seems
not only governed by the Co
as the homogeneous distribution and mutual interaction between Co
such mutual interaction is not surprising considering recent results on the peculiar structure of
Co crystallites encapsulated by nanosized ceria that maximized the interaction between the two
3
O
4
crystallite size, but rather by a combination of several factors, such
3
O
4
and CeO . The key role of
2
3
O
4
oxides with unique redox and catalytic properties [18].
XRD and BET characterizations performed on selected spent Co350 and Co30Ce350 catalysts,
recovered after catalytic tests performed up to 600 °C, that is a higher temperature than the
calcination treatment, showed a slight increase of the mean crystallite size, in the same trend
observed for the samples calcined at 650 °C, without any dramatic structural and morphological
change.
The catalytic data of the Co series and Co30Ce samples herein described are in good agreement
with our previous results [20, 21] recorded in the same WHSV conditions on precipitated Co
3 4
O
and Co30Ce oxides with comparable crystallite size and surface area. By comparison with other
ꢀ1 ꢀ
literature results [31] the present Co O catalysts are apparently less performing (0.225 µmol g s
cat
3
4
1
3 4
at 300 °C for CoNa650) than Co O nanocrystals synthesized via hydrothermal process and
characterized by different shapes, like nanosheets, nanobelts and nanocubes (0.544, 0.424 and 0.359
ꢀ1
ꢀ1
µmol gcat s , respectively, calculated at 273ꢀ284 °C). The superior catalytic activity of such pure
Co nanocrystals may be correlated with the more reactive crystal planes exposed on the surface.
On the other hand, different specific rates of methane conversion may depend on different
3
O
4
ꢀ1
ꢀ1
ꢀ1
experimental conditions (WHSV = 60000 ml g h in the present case, GHSV = 40000 h [31]).
For methane oxidation on cobalt oxideꢀbased catalysts, a number of parameters, such as crystallite
size, specific surface area, reducibility of the cobalt species, nature of the support and the cobalt
content on the surface, are thought to influence the catalytic activity [20, 21, 53].
ꢀ1
ꢀ1
In Fig. 11 a,b, the reaction rates (ꢀmol CH ⋅s ⋅gcatalyst ) versus the Co O (wt%) and versus the
4
3
4
Co/Ce (XPS atomic ratio) are displayed for the Co and Co30Ce samples. The precipitation agent
and calcination temperature seriously influenced the catalytic activity. Besides the samples
precipitated by (NH
4 2 3
) CO showing relatively low activity and performing similarly, the activity of
bare Co and Co30Ce oxide precipitated with the same agent significantly differed as a function
3 4
O
of the calcination temperature and Co/Ce surface atomic ratio. The highest activity was registered
for Co30Ce oxides prepared by urea according to the high reducibility observed at lowꢀtemperature
1
5

Catalysis Science & Technology
Page 16 of 38
View Article Online
DOI: 10.1039/C4CY01158A
3 4 2
(Fig. 9) and as well to the homogeneous distribution of the small Co O particles over CeO
strongly interacting with them (Fig. 8).
According with the literature, the combustion of methane over the cobalt oxide catalyst follows a
MarsꢀVan Krevelen mechanism and depends on the fast migration of oxygen ions through the
lattice of cobalt oxide [53]. Thus, the oxygen buffer by release/uptake of oxygen through redox
4
+
3+
processes involving the Ce /Ce couple contributed to the methane oxidation at high temperature.
Accordingly, nanocrystalline ceria showed appreciable activity starting from about 450 °C giving
5
0% of methane conversion at 526 °C and almost full conversion at 700 °C (see Table 5, Fig. 10 d).
Moreover, stable methane conversions were registered for all the Co30Ce catalysts above 700 °C
although it is well know that at high temperature, the active phase, Co partially decomposes to
3
O
4
the less active CoO oxide [20,21]. In the present case, the attainment of stable catalytic conversions
is likely due to the coꢀcatalytic effect of CeO at 700 °C (see Fig. 10 d). On the contrary, for bare
2
3 4 4
Co O catalysts an evident CH conversion decay was observed above 700 °C (see Fig. 10 c,d).
Except for the appearance of the CoO phase detected in the XRD patterns of spent Co650 and
Co30Ce650 catalysts recovered after the catalytic tests performed at 800 °C, no sintering
phenomena or morphological changes were noticed suggesting a good stability in the reaction
conditions for samples calcined at 650 °C for 5h.
3
.5. Temperature-programmed desorption of NH
It has been, recently, demonstrated the key role of surface Lewis acidity of nanocrystalline Co
oxides for catalytic oxidation of chlorinated VOCs [55]. A high Lewis acidity evaluated by means
of NH ꢀTPD analysis was found to be relevant for promoting the conversion of 1,2ꢀdichloroethane.
Moreover, Qiu et al. have proposed the close relationship between the catalytic activity of catalyst
in SCR and its surface adsorption level of NH [48]. A direct proportion between the NH
adsorption capacity and the abundance of Co on the surface of Co nanocatalysts was found.
3
3
O
4
3
3
3
3
+
3
O
4
A linear relationship between specific reaction rates of methane conversion and Co/Ce surface
atomic ratio (Fig. 11 b) has been reported so far, confirming our previous results on the catalytic
3
+
role of Co species in methane oxidation activity [20ꢀ23]. In order to get further insight into the
influence of Lewis acidity on the catalytic performances of Co O ꢀCeO systems, NH ꢀTPD
3
4
2
3
experiments were carried out over selected samples, Co
3
O
4
, Co30Ce calcined at 350 and at 650°C.
Nanocrystalline ceria has been measured for comparison. Some representative NH
3
ꢀTPD profiles
are displayed in Fig. 12. The total acidity values expressed as mmol of desorbed NH per gram of
3
catalyst and calculated by integrating the area of the TPD curves, are listed in Table 6. The most
intense features were observed for the Co30Ceurea350 showing a first broad peak between 150ꢀ350
1
6

Page 17 of 38
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01158A
°C and a second and less intense signal extended to the entire hold time at 500 °C. According with
the literature [55], peaks at T≤ 200 °C can be attributed to the presence of weak acid sites, while
desorption occurring between 300ꢀ500 °C was associated with strong acid sites. The desorption
profile of Courea350 was characterized by a very large and continuous peak in the range 200ꢀ500 °C,
suggesting also in this case the presence of both, weak and strong acid sites. However, by
comparing the two curves of Courea350 and Co30Ceurea350 it is evident that the abundance and
strength of acid sites of pure Co
oxide that in turn is much more acid than the parent CeO
deposition over ceria of regular and highly dispersed Co O nanoparticles by precipitation with urea
3
O
4
strongly differed from that of the corresponding Co
3 4 2
O ꢀCeO
2
. These findings suggest that the
3
4
3+
3
is an effective method to form abundant Co ions responsible for the high NH adsorption capacity.
In contrast to Co30Ceurea350, the acid capacity of Co30CeNH4350 was much lower. The acidity
decreased for all samples after calcination at 650 °C (see Table 6), in agreement with the Co O4
3
crystallite sintering generally observed for bare oxides and Co30Ce samples as well. The
Co30Ceurea650 was the sample that preserved a larger amount of acid sites.
Conclusions
The synthetic procedure, precipitation by (NH
of cobalt species as determined after XRD characterization of 120 °C dried powders (formation of
cobaltꢀhydroxideꢀcarbonate or Co spinel species).
4
Upon calcination at 350 °C cobalt crystallized into pure Co O spinel phase for all samples, despite
4 2 3 2 3 2 2
) CO , Na CO and CO(NH ) influenced the nature
3
O
4
3
on the precipitation agent and chemical composition. Textural (crystallite size, specific surface area)
and catalytic properties strongly depended on the nature of precipitation agent. Homogeneous
precipitation by Na
2
CO
3
and CO(NH
2
)
2
led to regular and controlled Co
3 4
O particle size, while
bigger and heterogeneous crystals were formed with (NH
4
)
2
CO . When the precipitation of cobalt
3
oxides was carried out in presence of ceria by sodium or ammonium carbonate, the sample mapping
carried out by TEM and SEM analyses appears to be quite heterogeneous, with large
agglomerations of CeO
This effect is more evident in the case of the sample precipitated by (NH
Co O crystallites. On the contrary, the homogeneous precipitation via urea appears the most
2
particles, which are not directly interacting with the Co
3 4
O crystallites.
4
)
2
CO characterized by big
3
3
4
adequate method for preparing regular nanocrystalline Co
over ceria and strongly interacting with them.
3 4
O particles homogeneously distributed
The peculiar structure and morphology of nanocrystalline ceria plays a fundamental role in
stabilizing Co active phase against sintering and in promoting its activity. This trend was
common to all Co ꢀCeO mixed oxides.
3
O
4
3
O
4
2
1
7

Catalysis Science & Technology
Page 18 of 38
View Article Online
DOI: 10.1039/C4CY01158A
3 4 2
The excellent catalytic activity of the Co O ꢀCeO catalyst precipitated by urea can be attributed to
a combination of several factors, such as small crystallite size which give rise to easily accessible
3
+
3+
active sites, easy reducibility of Co at low temperature, high surface Co content at the Co
3 4
O ꢀ
CeO interface. The mutual interaction between the two oxides guarantees full methane conversion
2
above 700 °C.
Acknowledgements
The authors acknowledge the financial support provided by the Network of Excellence IDECAT ꢀ
Integrated Design of Catalytic Nanomaterials for a Sustainable Production and by COST Action
CM 1104.
The authors are sincerely grateful to Prof. A. Corma of Instituto de Tecnologia Quimica (CSICꢀ
UPV), University Politecnica de Valencia for valuable suggestions and discussions.
Dr. Francesco Giordano from ISMNꢀCNR is acknowledged for XRD analyses.
The China Scholarship Council is sincerely acknowledged for supporting Wu’s and Guo’s
scholarship. H. Wu and S. Guo thank the Excellent Doctorate Foundation, Doctorate Foundation of
Northwestern Polytechnical University and the Scholarship Award for Excellent Doctoral Student
granted by Ministry of Education, P.R. China.
References
[
[
[
[
[
1] J.J. Spivey, Ind. Eng. Chem. Res. 1987, 26 , 2165ꢀ2180.
2] M.M. Zwinkels, S.G. Jaras, P.G. Menon, Catal. Rev. Sci. Eng. 1993, 35, 319ꢀ358.
3] J. Jansson, J. Catal. 2000, 194, 55ꢀ60.
4] V.G. Milt, M.A. Ulla, E.A. Lomardo, Catal. Lett. 2000, 65, 67ꢀ73.
5] T.ꢀC. Xiao, S.ꢀF. Ji, H.ꢀT. Wang, K.S. Coleman, M.L.H. Green, J. Mol. Catal. A 2001, 175,
1
11ꢀ123.
[
[
[
[
[
6] J. Kirchnerova, M. Alifanti, B. Delmon, Appl. Catal. A 2002, 231, 65ꢀ80.
7] M. Kang, M.W. Song, C.H. Lee, Appl. Catal. A 2003, 251, 143ꢀ156.
8] C.ꢀW. Tang, C.ꢀC. Kuo, M.ꢀC. Kuo, C.ꢀB. Wang, S.ꢀH. Chien, Appl. Catal. A 2006, 309, 37ꢀ43.
9] C.ꢀW. Tang, M.ꢀC. Kuo, C.ꢀJ. Lin, C.ꢀB. Wang, S.ꢀH. Chien, Catal. Today 2008, 131, 520ꢀ525.
10] J. Łojewska, A. Kołodziej, T. Łojewski, R. Kapica, J. Tyczkowski, Appl. Catal. A 2009, 366,
2
06ꢀ211.
[
[
11] X. Xie, Y. Li, ZhiꢀQ. Liu, M. Haruta, W. Shen, Nature 2009, 458, 746ꢀ749.
12] V.G. Milt, C.A. Querini, E.E. Miró, M.A. Ulla, J. Catal. 2003, 220, 424ꢀ432.
1
8

Page 19 of 38
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01158A
[
13] M. Dhakad, T. Mitshulhashi, S. Rayalu, P. Doggali, S. Bakardjiva, J. Subrt, D. Fino, H.
Haneda, N. Labhsetwar, Catal. Today 2008, 132, 188ꢀ193.
[
[
[
[
[
14] U. Zavyalova, P. Scholz, B. Ondruschka, Appl. Catal. A 2007, 323, 226ꢀ233.
15] J. Li, X. Liang, S. Xu, J. Hao, Appl. Catal. B 2009, 90, 307ꢀ312.
16] M.M. Natile, A. Glisenti, Chem. Mater. 2005, 17, 3403ꢀ3414.
17] Li Xue, C. Zhang, H. He, Y. Teraoka, Appl. Catal B 2007, 75, 167ꢀ174.
18] J.ꢀY. Luo, M. Meng, X. Li, X.ꢀG. Li, Y.ꢀQ. Zha, T.ꢀD. Hu, Y.ꢀN. Xie, J. Zhang, J. Catal. 2008,
2
54, 310ꢀ324.
[
[
19] C.ꢀW. Tang, W.ꢀY. Yu, C.ꢀJ. Lin, C.ꢀB. Wang, S.ꢀH. Chien, Catal. Lett. 2007, 116, 161ꢀ166.
20] L.F. Liotta, G. Di Carlo, G. Pantaleo, A.M. Venezia, G. Deganello, Appl. Catal. B 2006, 66,
2
17ꢀ227.
[
[
21] L.F. Liotta, G. Di Carlo, G. Pantaleo, G. Deganello, Appl. Catal. B 2007, 70, 314ꢀ322.
22] L.F. Liotta, M. Ousmane, G. Di Carlo, G. Pantaleo, G. Deganello G. Marcì, L. Retailleau, A.
GiroirꢀFendler, Appl. Catal. A 2008, 347, 81ꢀ88.
23] L.F. Liotta, M. Ousmane, G. Di Carlo, G. Pantaleo, G. Deganello, A. Boreave, A. Giroirꢀ
Fendler, Catal. Lett. 2009, 127, 270ꢀ276.
24] L. F. Liotta, G. Pantaleo, A. Macaluso, G. Di Carlo and G. Deganello, Appl. Catal. A 2003,
45, 167ꢀ177.
25] I.Yu. Zavaliy, R.V. Denys, R. Cerny, I.V. Koval'chuk, G.Wiesinger, G. Hilscher, J. Alloys
Comp. 2005, 386, 26ꢀ34.
26] S. Carrettin, P. Concepción, A. Corma, J.M.L. Nieto, V.F. Puntes, Angew. Chem. Int. Ed. 2004,
3, 2538ꢀ2540.
[
[
2
[
[
4
[
[
[
[
[
[
27] J. Guzman, S. Carrettin, A. Corma, J. Am. Chem. Soc. 2005, 127, 3286ꢀ3287.
28] T. Ishikawa, E. Matijević, Colloid Polym. Sci. 1991, 269, 179ꢀ186.
29] R. Xu, H.C. Zeng, J. Phys. Chem. B 2003, 107, 12643ꢀ12649.
30] B. Li, Y. Xie, C. Wu, Z. Li, J. Zhang, Mater. Chem. Phys. 2006, 99, 479ꢀ486.
31] L. Hu, Q. Peng, Y. Li, J. Am. Chem. Soc. 2008, 130, 16136ꢀ16137.
32] S.J. Gregg, K.S. Sing, Adsorption, Surface Area and Porosity, 2nd ed., Academic Press, San
Diego, 1982.
[
[
33] Inorganic Crystal Structure Database, ICSD, FIZ Karlsruhe, 2014.
34] H.P. Klug, L.E. Alexander, Xꢀray Diffraction Procedures for Polycrystalline and Amorphous
Materials, Wiley, New York, 1954.
[
[
35] D.A. Shirley, Phys. Rev. B 5 (1972) 4709ꢀ4714.
36] P.M.A. Sherwood, Practical Surface Analysis, in: D. Briggs, M.P. Seah (Eds.), Wiley, New
York, 1990, p.181.
1
9

Catalysis Science & Technology
Page 20 of 38
View Article Online
DOI: 10.1039/C4CY01158A
[
37] T. Baird, K. C. Campbell, P. J. Holliman, , R. W. Hoyle, D. Stirling, B. P. Williams, M.
Morris, J. Mater. Chem. 1997, 7, 319.
[
[
[
[
[
[
[
[
[
38] V.G. Hadjiev, M.N. Iliev, I.V. Vergilov, J. Phys. C: Solid State Phys. 1988, 21, L199ꢀL201.
39] C. W. Tang, C. B. Wang, S. H. Chein, Thermochim. Acta 2008, 473, 68ꢀ73.
40] H.ꢀH. Chou, H.Y. Fan, Phys. Rev. B 1976, 13, 3924ꢀ3938.
41] J. Liu, Z. Zhao, J. Wang, C. Xu, A. Duan, G. Jian, Q. Yang, Appl. Catal. B 2008, 84, 185ꢀ195.
42] L. F. Liotta, H. J. Wu, G. Pantaleo, A. M. Venezia, Catal. Sci. Technol. 2013, 3, 3085ꢀ3102.
43] P. Arnoldy, J. A. Moulijn, J. Catal. 1985, 93, 38ꢀ54.
44] J. Y. Luo, M. Meng, Y. Q. Zha, L. H. Guo, J. Phys. Chem. C 2008, 112, 8694ꢀ8701.
45] Q. Y. Yan, X. Y. Li, Q. D. Zhao, G. H. Chen, J. Hazard. Mater. 2012, 209ꢀ210, 385ꢀ391.
46] G. Jacobs, Y. Y. Ji, B. H. Davis, D. Cronauer, A. J. Kropf, C. L. Marshall, Appl. Catal. A 2007,
3
33, 177ꢀ191.
47] A. Y. Khodakov, A. G. Constant, R. Bechara, F. Villain, J. Phys. Chem. B. 2001, 105, 9805ꢀ
811.
[
9
[
[
48] B. Meng, Z. K. Zhao, X. Z. Wang, J. J. Liang, J. S. Qiu, Appl. Catal. B 2013, 129, 491ꢀ500.
49] A.M. Venezia, G. Pantaleo, A. Longo, G. Di Carlo, M.P. Casaletto, L.F. Liotta, G. Deganello,
J. Phys. Chem. B 2005, 109, 2821ꢀ2827.
[
[
50] C. W. Tang, C. C. Kuo, M. C. Kuo, C. B. Wang, S. H. Chien, Appl. Catal. A 2006, 309, 37ꢀ43.
51] W. Song, A. S. Poyraz, Y. Meng, Z. Ren, S.ꢀY. Chen, S. L. Suib, Chem. Mater. 2014, 26,
4
629ꢀ4639.
[
[
52] H. J. Wu, L. D. Wang, Z. Y. Shen, J. H. Zhao, J. Mol. Catal. A 2011, 351, 188ꢀ195.
53] Z. Z. Zhu, G. Z. Lu, Z. G. Zhang, Y. Guo, Y. L. Guo, Y. Q. Wang, ACS Catal. 2013, 3 , 1154ꢀ
1
164.
[
[
54] A. Pfau, K.D. Schierbaum, Surf. Sci. 1994, 321, 71ꢀ80.
55] B. de Rivas, R. LópezꢀFonseca, C. JiménezꢀGonzález, J.I. GutiérrezꢀOrtiz, J. Catal. 2011, 281,
8
8ꢀ97.
2
0

Page 21 of 38
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01158A
3 4
Table 1. Mean crystallite size of the Co O phase and specific surface area for all the
prepared samples, bare Co O and Co30Ce oxides
3
4
2
2
Sample
d (nm)
BET (m /g)
Sample
d (nm)
BET (m /g)
Courea350
CoNa350
19.0
37.0
Co30Ceurea350 14.0
122.0
127.0
108.0
53.0
Co30CeNa350
1
2
6.0
4.0
43.0
10.0
CoNH4350
Courea650
Co30CeNH4350
31.0
19.0
Co30Ceurea650
110.0
8.0
61.0
CoNa650
Co30CeNa650
40.0
7
1
2.0
10.0
64.0
CoNH4650
Co30CeNH4650
45.0
5.9
121.0
47.0
1

Catalysis Science & Technology
Page 22 of 38
View Article Online
DOI: 10.1039/C4CY01158A
3+
2
Table 2. H uptakes and relative Co contents of Co30Ce catalysts prepared by
precipitation method using three different precipitating agents and calcined at 350 or
at 650 °C
Range of temperature (°C)
1
00-300
300-400
V (ml g )
102.0
400-550
680-820
-
1 a
-1 a
Sample
Vtotal (ml g )
Co30Ceurea350 41.0
Co30CeNa350
0.8
2.5
4.0
5.7
146.3
147.0
144.7
45.0
98.0
118.0
Co30CeNH4350 21.0
Range of temperature (°C)
2
00-380
400-500 680-820
V (ml g )
-1 a
-1 a
Sample
Vtotal (ml g )
Co30Ceurea650
Co30CeNa650
133.0
125.0
124.0
4.0
8.5
9.5
12.0
13.0
14.0
149.0
146.5
147.5
Co30CeNH4650
a
All values are calculated at 1 atm and 298 K
2

Page 23 of 38
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01158A
Table 3. XPS data of Co30Ce catalysts prepared by precipitation method using three
different precipitating agents and calcined at 350 or at 650 °C.
3
+
3+
2+
Sample
Co 2p3/2 (eV)
Ce 3d5/2 (eV)
Surface
Co/Ce
atomic ratio
6.3
Co /(Co +Co ) (%)
a
Co30Ceurea350
Co30CeNa350
Co30CeNH4350
Co30Ceurea650
779.9
882.0
884.4
882.2
884.6
882.2
884.6
882.4
884.8
882.4
884.8
881.9
884.3
47.0
46.0
45.0
45.0
48.0
47.0
7
81.3
779.6
81.8
779.6
80.9
779.7
3.8
1.8
4.3
1.5
0.8
7
7
7
7
81.5
79.9
Co30CeNa650
781.7
80.1
782.0
7
Co30CeNH4650
a
The nominal value Co/Ce is equal to 0.92
3

Catalysis Science & Technology
Page 24 of 38
View Article Online
DOI: 10.1039/C4CY01158A
Table 4. Temperatures of 50 and 90 % of CH conversion (T and T ) and reaction
4
50
90
rates (r) calculated at 250 °C (conversions < 20%) for over Co O and Co30Ce
3
4
catalysts prepared by using three different precipitating agents and calcined at 350 °C.
For comparison, the values obtained for nanocrystalline ceria are also listed.
-
1 -1
T (°C)
T (°C)
Sample
4
r (µmol CH gcat s )
5
0
90
0
.153
.281
Courea
371
354
487
460
0
CoNa
0
0
0
.098
.373
.215
CoNH4
3
92
512
450
490
Co30Ceurea
Co30CeNa
343
366
409
0
0
.102
Co30CeNH4
CeO2
531
-
.0004
526
4

Page 25 of 38
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01158A
4
Table 5. Temperatures of 50 and 90 % of CH conversion (T50 and T90) and reaction
rates (r) calculated at 300 °C (conversions < 20%) for Co O and Co30Ce catalysts
3
4
prepared by using three different precipitating agents and calcined at 650 °C. For
comparison, the values obtained for nanocrystalline ceria are also listed.
-
1 -1
T (°C)
T (°C)
Sample
r (µmol gcat s )
5
0
90
0.133
0.225
0.082
Courea
472
449
490
583
550
623
CoNa
CoNH
4
0
.266
.156
Co30Ceurea 451
Co30CeNa 474
549
593
0
0
0
.082
Co30CeNH4
CeO2
5
5
18
26
649
653
.0049
5

Catalysis Science & Technology
Page 26 of 38
View Article Online
DOI: 10.1039/C4CY01158A
3 3 4
Table 6. NH -TPD data of selected Co O and Co30Ce catalysts prepared by using
three different precipitating agents, calcined at 350 and at 650°C. For comparison, the
values calculated for nanocrystalline ceria are also listed.
Samples
Total acidity
mmol NH
-
1
(
3
g
cat
)
Courea350
0.21
0.35
0.19
0.09
0.14
0.26
0.10
Co30Ce urea350
Co30CeNH4350
CeO
2
Courea650
Co30Ceurea650
Co30CeNH4650
6

Page 27 of 38
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01158A
CoNa120
CoNH4120
Co O ICSDreference
3
4
Courea120
10
20
30
40
50
60
70
2θ (°)
3 4
Fig. 1. XRD patterns of cobalt-hydroxide carbonate and/or Co O species prepared by precipitation
method using three different precipitating agents and dried at 120 °C.
1

Catalysis Science & Technology
Page 28 of 38
View Article Online
DOI: 10.1039/C4CY01158A
°
Co O
3 4
#
Co(OH) (CO ) 0.11H O
1.0 3 0.5 2
*
CeO2
*
*
*
*
#
#
Co30CeNa120
Co30CeNH4120
°
°
*
°
*
°
°
*
*
Co30Ceurea120
#
#
# #
#
#
#
10
20
30
40
50
60
70
2θ
(°)
Fig. 2. XRD patterns of cobalt-hydroxide carbonate and/or Co O species over nanocrystalline
3
4
CeO
2
deposited by precipitation method using three different precipitating agents and dried at
1
20 °C.
2

Page 29 of 38
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01158A
°
Co O4
3
(a)
°
°
°
CoNa350
°
CoNH4350
Courea350
15
20
25
30
35
40
45
50
2θ
(°)
°
Co O4
3
(b)
°
°
CoNa650
°
°
°
CoNH4650
Courea650
1
5
20
25
30
35
40
45
50
2θ (°)
3 4
Fig. 3. XRD patterns of bare Co O oxides, prepared by precipitation method using three different
precipitating agents, after calcination at (a) 350 °C and (b) 650 °C.
3