Structure-reactivity relationship in Co3O4 promoted Au/CeO2 catalysts for the CH3OH oxidation reaction revealed by in situ FTIR and operando EXAFS studies

Article abstract of DOI:10.1039/c6ta06442f

A strong influence of the amount of the Co₃O₄ promoter on the catalytic performance in methanol oxidation of different gold catalysts supported on ceria was observed. The activity followed the order: Au/10 wt% Co₃O₄-doped CeO₂ > Au/5 wt% Co₃O₄-doped CeO₂ > Au/15 wt% Co₃O₄-doped CeO₂ > Au/CeO₂ ? Au/Co₃O₄. FTIR measurements of adsorbed CO indicate that oxidized gold sites are initially present on the activated samples and that such species are involved in the methanol reaction. Methanol oxidation performed under static conditions gave rise at 75 °C to mainly formate species on Au/CeO₂ and to a large variety of different carbonate species on Au/10 wt% Co₃O₄-doped CeO₂. FTIR and EXAFS analyses revealed that the active sites present on the best performing Au/CeO₂ catalyst added with 10 wt% Co₃O₄ are oxidized gold species, close to Co sites, at the interface with the support, which are reduced under reaction conditions. These species are able to activate and to react with oxygen giving rise to formate and carbonate species.

Full text of DOI:10.1039/c6ta06442f

View Article Online
View Journal
Journal of
M at erials Chemistry A
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: M. Manzoli, F.
Vindigni, T. Tabakova, C. Lamberti, D. Dimitrov, K. Ivanov, A. Giovanni and G. Agostini, J. Mater. Chem. A,
2016, DOI: 10.1039/C6TA06442F.
Volume
4
Number
1
7
January 2016 Pages 1–330
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Journal of
Materials Chemistry A
Accepted Manuscripts are published online shortly after
Materials for energy and sustainability
www.rsc.org/MaterialsA
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
author guidelines.
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, outlined
in our author and reviewer resource centre, 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.
ISSN 2050-7488
PAPER
Kun Chang, Zhaorong Chang et al.
Bubble-template-assisted synthesis of hollow fullerene-like
MoS nanocages as a lithium ion battery anode material
2
rsc.li/materials-a

Please do notadjustmargins
Journal of Materials Chemistry A
Page 1 of 13
View Article Online
DOI: 10.1039/C6TA06442F
Journal of Materials Chemistry A
ARTICLE
Structure-reactivity relationship in Co O promoted Au/CeO
2
3
4
catalysts for the CH OH oxidation reaction revealed by in situ FTIR
and operando EXAFS studies
3
Received 00th January 20xx,
Accepted 00th January 20xx
DOI: 10.1039/x0xx00000x
www.rsc.org/
a
b
c
b,f
d
d
e,g
M. Manzoli *, F. Vindigni , T. Tabakova , C. Lamberti , D. Dimitrov , K. Ivanov , G. Agostini
A strong influence of the amount of the Co
gold catalysts supported on ceria was observed. The activity followed the order: Au/10 wt.% Co
wt.% Co -doped CeO > Au/ 15 wt.% Co -doped CeO > Au/CeO >> Au/Co . FTIR measurements of adsorbed CO
3
O
4
promoter on the catalytic performance in methanol oxidation of different
3
O
4
-doped CeO > Au/5
2
3
O
4
2
3
O
4
2
2
3 4
O
indicate that oxidized gold sites are initially present on the activated samples and that such species are involved in the
methanol reaction. Methanol oxidation performed in static conditions gave rise at 75 °C to mainly formate species on
Au/CeO
2
and to a large variety of different carbonate species on Au/10 wt.% Co
3
O
4
-doped CeO
2
. FTIR and EXAFS analyses
are oxidized gold
revealed that the active sites present on the best performing Au/CeO
2
catalyst added with 10 wt.% Co O
3 4
species, close Co sites, at the interface with the support, which are reduced under reaction conditions. These species are
able to activate and to react with oxygen giving rise to formate and carbonate species.
selection of the proper support is a critical factor.
Introduction
Ceria is very attractive, due to its ability to maintain high metal
dispersion and to change Ce oxidation state depending on the redox
Supported gold nanoparticles attracted significant attention due to
their high catalytic activity in various oxidation reactions at low
temperatures. The catalytic oxidation of Volatile Organic
Compounds (VOCs), such as methanol, is a subject of considerable
interest due to their relevance in many industrial applications as
well as in the field of sustainable chemistry. These catalytic
processes have been considered as some of the most efficient ways
to reduce harmful emissions from various chemical industries.
During the past two decades gold-based catalysts have been proven
conditions. The rapid change of oxidation state is related to its
ability to store and release oxygen, a property measured by the
4
‘
‘oxygen storage capacity’’ (OSC), which results in rapid formation
5
and elimination of oxygen vacancy defects. These characteristics
make CeO a very interesting support for oxidation reactions.
2
In principle, the modification of ceria can increase the
concentration of oxygen vacancies and a strong influence of the
nature of the added oxide (Me= Fe, Mn, Sn) on the catalytic
6
1
performances was observed recently by some of us. For example,
to be effective for many reactions of environmental significance.
2
it was reported that the modification of CeO with different Zr
Considerable emphasis is placed on the catalyst preparation and on
7
2
amounts leads to increased catalytic activity in the CO oxidation , as
the selection of the support. The oxidation activity of gold catalysts
8
3
well as in the Water Gas Shift reaction and in the steam reforming
is closely related to the size of the Au particles and to the ability of
9
of methanol. Moreover, gold catalysts supported on iron-modified
the support to provide active oxygen species. For this reason, the
ceria oxides containing opportune molar percentages of Fe were
2
proved to be more active than unmodified Au/CeO for total and
1
0, 11
12
preferential CO oxidation
and for Water Gas Shift reaction.
The beneficial influence of cobalt oxide on the CO oxidation
1
3
2 2 3
behaviour of Au/CoOx/CeO -Al O systems was previously shown.
It was found that the catalytic activity depends on the cobalt
loading, which provides a submonolayer coverage of the support
surface, where a CoO
x
-CeO
2
interaction takes place giving rise to a
itself
synergism between Co and Ce redox properties. Indeed, Co
3
O
4
was demonstrated to be very active in the sub-ambient CO
oxidation reaction depending on the morphology and on the
preferential exposition of the [1 1 0] plane in which the highest
1
4, 15
number of Co(III) sites are present.
This journal is © The Royal Society of Chemistry 20xx
J. Materials Chemistry aName., 2013, 00, 1-3 | 1
Please do notadjustmargins

Please do notadjustmargins
Journal of Materials Chemistry A
Page 2 of 13
View Article Online
DOI: 10.1039/C6TA06442F
ARTICLE
Journal Name
Three dimensionally ordered macroporous Au/CeO
2 3 4
-Co O catalysts temperature of 60 °C, aging at the same temperature for 1 hour,
–
showed enhanced CO preferential oxidation in
H
2
-rich gas filtering, washing until removal of NO
The former paper at 80 °C. Mixed CeO –Co supports were prepared by a simple
proposed that the origin of the high CO conversion and CO mechano-chemical mixing procedure: mixture of cerium
selectivity was the presence of ionic Au in intimate contact with hydroxide and (5, 10, 15 wt%) of laboratory-made Co was
CeO . In the latter one, a mechanism involving the synergistic effect subjected to mechano-chemical milling for 30 min in a mortar and
between CeO and Co supports, which greatly accelerates the calcination at 400 °C for 2 hours. Hereafter the supports will be
3
ions and drying in vacuum
1
6
17
mixtures and formaldehyde oxidation.
2
3 4
O
2
a
3
O
4
2
2
3 4
O
surface active oxygen migration and activates the Au species was labelled as Ce5Co, Ce10Co and Ce15Co depending on cobalt oxide
proposed to explain the improvement in the catalytic HCHO loading in wt%. Gold catalysts (3 ± 0.05 wt.%) were prepared by the
oxidation. It was established that Au/CeO -CoO catalysts prepared deposition–precipitation method at constant pH 7 and temperature
2 x
by mechano-chemical activation exhibited much higher activity in 60 °C. Gold was deposited on a mixed metal oxide support
the reaction of complete benzene oxidation than that prepared preliminary suspended in water and activated in an ultrasound
1
8
using coprecipitation. The reason of this different catalytic disintegrator under vigorous stirring. Hereafter catalysts will be
behaviour was the existence of a Co-containing phase with labelled as AuCe5Co, AuCe10Co and AuCe15Co depending on cobalt
enhanced redox properties that influenced both gold and modified oxide loading in wt%.
ceria in close vicinity. Therefore, ceria supports with different The deposition-precipitation was carried out in an automated
3 4
amount of Co O were prepared by mechano-chemical activation in laboratory reactor under full control of all preparation parameters
order to optimize the content of promoter. Recently, we observed a (pH, temperature, stirring speed, reactant feed flow rates, etc.).
°
strong influence of the nature of the added oxide on the activity of After aging at 60 C within 1 h, filtering and careful washing, the
ceria-based gold catalysts for catalytic abatement of CO in waste precursors were dried under vacuum at 80 °C and calcined in air at
1
9
gases. In particular, gold catalyst supported on ceria modified by 400 °C for 2 hours. Moreover, unmodified Au/CeO
Co exhibited the best performance: 100% CO conversion degree (hereafter labeled as AuCe and Au/Co , respectively), used as
was achieved at 25 °C. The support with composition 90 wt.% CeO blank samples, were also prepared by the deposition-precipitation
and 10 wt.% Co was beneficial for the nucleation and growth of method under the conditions described above. All initial salts used
2 3 4
and Au/Co O ,
3
O
4
3 4
O
2
3
O
4
highly dispersed gold. On this system, a synergy between gold and were ‘analytical grade’.
Co-doped ceria support causes significant enhancement in the
reducibility and capability for oxygen activation, which resulted in Catalytic activity measurements
an improved oxidation activity. However, a clear correlation
3
The catalytic activity of the samples in CH OH oxidation was
between the catalytic activity with the nature and structure of the
active sites has not been fully established yet.
measured using an isothermal plug flow reactor within a different
temperature range. The following conditions were chosen: catalyst
The correlation existing between catalyst structure and catalytic
3
bed volume – 0.5 g (particle size 0.6 – 1.0 mm), inlet CH OH
2
0
activity/selectivity is the main goal in research applied to catalysis.
concentrations – 2.0 % balanced with air and space velocity 10000
In principle, EXAFS/XANES can give relevant complementary
information on the structural and electronic properties of such
-1
h . The tests were carried out by increasing the reaction
3
temperature step-wise until complete CH OH oxidation. The degree
21-25
systems.
In addition, FTIR spectroscopy represents a powerful
o
of conversion was measured at every 20 C step after stationary
conversion was achieved. The behaviour was also studied during
decrease of the temperature down to room temperature, again
step-wise. The conversion data at increasing and decreasing
temperature match well. The duration of catalytic test for each
sample was about 8 hours. The reactant and product gases were
technique to investigate the surface sites at an atomic level. In
particular, the analysis of FTIR spectra of adsorbed probe molecules
allows to understand the nature and the abundance of the exposed
active sites as well as to have detailed information on their
2
6-30
chemical environment.
Therefore, the use of these techniques
can assist in the comprehension of the parameters ruling the
unique catalytic properties of gold catalysts, as well as to
implement the knowledge in the design of new systems. Keeping in
mind these purposes, the aim of this work was an accurate EXAFS
and FTIR study on the nature of the active sites exposed at the
surface of highly dispersed gold catalysts supported on ceria
2 2 2
analysed for methanol, CO, CO , O and N by HP 5890 Series II gas-
chromatograph, equipped with flame ionization and thermal
conductivity detectors and Porapak Q and MS-5A columns.
Methods
The amount of gold in all samples was analyzed by atomic
absorption spectroscopy (Varian Vista MPX) and the values
determined were 3 (± 0.05) wt % Au.
3 4
modified by different amounts of Co O . Research efforts were
focused to achieve information on the local structure of the active
sites in terms of position of cobalt atoms in respect to the gold
metallic phase.
High resolution transmission microscopy (HRTEM) measurements
were performed using a side entry Jeol JEM 3010 (300 kV)
6
microscope equipped with a LaB filament and fitted with X-ray EDS
analysis by a Link ISIS 200 detector. For analyses, the powdered
samples were deposited on a copper grid, coated with a porous
carbon film. All digital micrographs were acquired by an Ultrascan
Experimental
Materials
Cerium hydroxide was synthesized by precipitation of aqueous 1000 camera and the images were processed by Gatan digital
solutions of Ce(NO
3
)
3
·6H
2
O with K
2
CO
3
at constant pH=9.0 and at a micrograph. A statistically representative number of Au particles
2
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do notadjustmargins

Page 3 of 13
Journal Name
i.e. 250 for each sample) was counted in order to obtain the
Jou rPn l ea al soe f dMo ant oe tr ai ad lj su s Ct mh ea rmg iins st ry A
View Article Online
DOI: 10.1039/C6TA06442F
ARTICLE
1
00
(
particle size distributions.
8
6
4
2
0
0
0
0
0
The FTIR spectra were recorded in transmission mode at a
resolution of 2.0 cm− on a Perkin-Elmer 2000 spectrometer
1
(equipped with a MCT detector) with the samples in self-supporting
pellets introduced in an AABSPEC 2000 cell allowing to run the
spectra in situ in controlled atmosphere and temperature (from
room temperature up to 600 °C).
AuCe
AuCe5Co
AuCe10Co
AuCe15Co
AuCo3O4
Ce10Co
As for the experiments at low temperature, the pelletised samples
were introduced in a cell allowing thermal treatments in controlled
atmosphere and spectrum scanning at controlled temperature
(from -180 °C up to room temperature). In order to mimic the
activation before the catalytic tests and to follow the effect of such
activation on the nature of the exposed active sites and of the
species formed during methanol oxidation reaction, the catalysts
were subjected to an oxidative treatment before the
2
0
40
60
80 100 120 140 160
o
Temperature, C
Figure 1. Temperature dependence of CH OH conversion over gold
3
measurements. The thermal treatment included heating from room
temperature (r.t.) up to 200 °C under outgassing below 10− mbar;
3
catalysts on ceria doped by 5, 10 and 15 wt% Co
O , over
3 4
unpromoted Au/CeO , Au/Co O and bare ceria doped by 10 wt%
2
3 4
an inlet of 20 mbar of oxygen and heating up to 400 °C; at 400 °C
the oxygen was changed three times (20 mbar for 10 min each run).
After that, the sample was cooled down to r.t. in oxygen and finally
outgassed at the same temperature. For each spectrum, the
spectrum of the sample before the inlet of the probe/reactants (CO,
methanol or oxygen) or of the methanol/oxygen mixture was used
as background. All reported spectra were background subtracted
and normalized to the density of the pellets.
Co O .
3
4
In particular, the activity towards the CH OH oxidation follows the
3
trend: AuCe10Co>AuCe5Co>AuCe15Co>AuCe>>Au/Co O >>Ce10Co.
3
4
The light-off curve related to bare ceria doped by 10 wt% Co O
3
4
appeared at the highest temperature, giving evidence of the key
role displayed by gold.
The best performance was observed in the case of the catalyst
EXAFS experiments were carried out on the BM23 beamline at ESRF
3
1
doped by 10 wt% Co O , over which almost 100% CH OH
3 4 3
3
facility (Grenoble, France). Fluorescence XAS spectra at the Au L -
o
conversion degree at a temperature as low as 40 C was obtained.
edge (11.919 keV) were collected with a 13-element Ge detector.
The beam energy were selected by double-crystal Si(111)
monochromator, and the third harmonic rejection was performed
The comparison with the data recently reported by Scirè et al.
o
(
about 95 % methanol conversion at 125 C over a gold supported
3
4
2
1
on ceria catalyst prepared by deposition-precipitation) puts in
evidence the promising application of AuCeO catalysts doped by
Co O . Moreover, a high defectivity of ceria exposed faces, along
by Rh coated mirror with an angle of -2.2 mrad. EXAFS signal were
3
2
2
extracted and analyzed by IFEFFIT package. The samples were
hosted inside a plug flow reactor using 1 mm quartz capillary
3
4
33
with the presence of gold nanoparticles with the smallest size and
the highest hydrogen consumption during TPR experiments (e.g.
highest oxygen mobility) were found over this AuCe10Co sample,
available on BM23 for in situ/operando measurements. The
experiments can be divided into two stages. Stage 1: the samples
have been heated from room temperature up to 200 °C by a small
1
9
according to our previous high resolution TEM characterization.
The light-off curves for CH OH oxidation of Au/CeO catalysts doped
by Co O were shifted to lower temperature with respect to that of
furnace, then they have been treated with a mixture O
2
/He (20 %
3
3
2
O
2
-He, 30 cm /min) up to 400 °C. After oxygen treatment, a CO-O -
2
3
4
1
:1/He mixture (5% CO, 5 % O
2
) was fluxed at r.t.. Stage 2: the
/He
-He, 30 cm /min) up to 400 °C. Successively the sample was
cooled down to r.t. and a vapour of methanol/O /He mixture (20 %
-He) was fluxed and finally the catalysts were heated up to 100 °C
undoped Au/CeO and to much lower temperature if compared to
2
samples were heated at 200 °C and treated with a mixture O
2
3
Au/Co O , that is also the worst catalyst. This behaviour is better
3 4
(20 % O
2
pointed out from the data in Table 1, where the T50 conversion
values (i.e. the temperature at which 50 % conversion was reached)
are contrasted. From the data reported in Table 1, the promoting
effect of Co O is evident, together with the fact that the beneficial
2
O
2
in the reaction mixture.
3
4
addition of cobalt oxide increases at low promoter content, reaches
a maximum and then declines, in agreement with the fact that gold
supported on the pure promoter phase (Au/Co O sample) is the
Results and discussion
3
4
Catalytic activity in the CH
3
OH oxidation
less active catalyst. In addition, it was found that the kinetic regime
The temperature dependence of CH
3
OH conversion degree is is not controlled by diffusional limitations. This statement is based
presented in Figure 1. The catalytic measurements indicated that on the experiments performed with two different catalyst pellets,
the amount of promoter has a strong influence on the activity of the former by using a pellet with a given weight and the latter with
ceria-based gold catalysts, particularly in defining the onset a pellet with a double weight with respect to that employed in the
temperature of the catalytic activity.
first test.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do notadjustmargins

Please do notadjustmargins
Journal of Materials Chemistry A
Page 4 of 13
View Article Online
DOI: 10.1039/C6TA06442F
ARTICLE
Journal Name
3
Table 1. Extrapolated T50 values for CH OH oxidation and calculated
-1
TOFs (s ) at two different temperatures over the catalysts.
Catalyst
CH
3
OH
TOF at 20 °C TOF at 40 °C
o
-1
-1
oxidation ( C)
(s )
(s )
AuCeO
2
53
0.00446
0.00633
0.0484
0.0167
0.0253
0.0505
0.0172
AuCe5Co
AuCe10Co
AuCe15Co
41
< 20
47
0.00458
The methanol molar flow was also doubled and a constant weight
hourly space velocity was maintained at a constant partial pressure.
The same methanol conversion to CO2 in steady state regime
indicated the absence of diffusion limitations. Table 1 also shows
-1
the Turnover frequencies (TOFs), expressed as s at two different
temperatures. The TOF values were calculated according to
3
5
previous considerations .
Gold dispersions were calculated for all samples based on the
average particle size determined by HRTEM/HAADF and reported in
19
Ref. (AuCe5Co – 2.4 nm, AuCe10Co – 1.8 nm, AuCe15Co – 2.1
nm), assuming the presence of cuboctahedral particles. According
2
Figure 2. HRTEM images collected on AuCeO (section a) and
AuCe10Co (section b), where the presence of Au nanoparticles is
put in evidence by red circles. Au particle size distributions of
3
to the catalytic trend for the CH OH conversion (see also Figure 1),
the TOF values obtained for the AuCe10Co catalyst are one order of
magnitude higher than those related to the other samples. The
temperature has a pronounced effect on the conversion: indeed,
with the exception of AuCe10Co, the obtained TOF values increase
of about four times when the temperature is raised from 20 °C up
to 40 °C. A very recent study on Au supported on bare ceria
AuCeO
number of particles. Instrumental magnification: 300000X and
00000X, respectively.
2
and AuCe10Co are reported in section c and d, where n.p.=
5
The AuCe catalyst contains gold species with heterogeneous size in
the 2-12 nm range (section a, evidenced by red circles), resulting in
an average diameter equal to 7.5 ± 4.5 nm (section c). However, the
particle size distribution reveals that the large majority of the
roundish gold nanoparticles has size around 2 nm (more than 50 %).
On the contrary, the AuCe10Co catalyst exposes gold nanoparticles
with smaller size (section b, evidenced by red circles), as indicated
by the narrow particle size distribution, being the average size equal
to 1.69 ± 0.41 nm (section d). It can be inferred that the presence of
3
6
reported the TOF values at two different temperatures. The TOF
-5
-1
-4 -1
markedly increased from 3.72 x 10
s at 45 °C to 1.02 x 10 s at
7
0 °C, putting in evidence an enhancement even more accentuate,
but in agreement with the data reported in Table 1. Moreover, the
behaviour of AuCe5Co and AuCe15Co is very similar to the
unpromoted AuCe catalyst. On the contrary, in the case of
AuCe10Co, the TOF value is almost unchanged, further showing that
1
3 4
0 wt% Co O provides the optimal composition for the catalysis
1
3 4
0 wt% Co O in the ceria support significantly affected the metal
and that for this catalyst the composition plays the major role, even
with respect to the temperature.
dispersion, exploiting a benefic effect in the stabilization of the Au
nanoparticles during the final calcination at 400 °C and therefore
favoring the formation of particles with more uniform size than in
Effect of the support composition on the gold dispersion
the case of AuCeO
enhancement in the catalytic activity observed for the catalyst
doped with 10 wt% Co
2
. These findings are in agreement with the
The catalytic activity measurements revealed almost 100% CH
3
OH
conversion degree at 40 C for the AuCe10Co catalyst. Therefore,
the addition of 10 wt% Co has improved the activity. It was
previously found that the most active catalyst was that containing a
o
3
4
O .
3
O
4
Nature of the exposed sites and in situ spectra of methanol
1
9
big amount of Au nanoparticles with average size of 1.8 nm. Since
the deposition-precipitation of gold was carried out on the already
oxidation reaction
FTIR spectroscopy measurements were carried out on the bare
AuCe and on AuCe10Co, i.e. on the most active doped sample. This
vibrational study is aimed to investigate the effect of the promoting
species on the nature and the abundance of gold exposed sites and
to have information on the nature of the surface species formed
during methanol oxidation from r.t. up to 100 °C.
3 4
doped support, an evaluation of the influence of the Co O
presence on the metal dispersion and size distribution was done by
performing new HRTEM measurements on the most promising
AuCe10Co catalyst. The results were compared with those obtained
2
for undoped AuCeO in order to put in evidence peculiar features
arisen from the different composition of the support on which the
same amount of gold has been deposited. All these data are
summarized in Figure 2. Both catalysts are supported on cubic
In Figure 3a, the FTIR spectra collected upon CO adsorption at r.t.
on unpromoted AuCe oxidised at 400 °C are shown.
2
CeO , mainly exposing the (111) face (not shown).
4
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do notadjustmargins

Page 5 of 13
Journal Name
Jou rPn l ea al soe f dMo ant oe tr ai ad lj su s Ct mh ea rmg iins st ry A
View Article Online
DOI: 10.1039/C6TA06442F
ARTICLE
line). These results indicate the absence of partially oxidized
-1 30
before reaction
after reaction
metallic gold (band at 2112 cm ) , showing that only cationic gold
clusters with the same nuclearity as those detected on AuCe are
exposed at the surface of the oxidised AuCe10Co sample before
reaction. Moreover, if the spectra are compared with those
reported in section a, the amount of these species is higher on the
Co-containing sample. The increased intensity of this absorption
despite the reduction of the CO pressure is an indication that the
small cationic gold clusters are initially covered by adsorbed oxygen
and probably are in a higher oxidized state. Indeed, the inlet of the
a
c
b
0
.1
0
.1
2
121
2
138
2
112
2250
2200
2150
2100
2050
2000 2250
2200
2150
2100
2050
2000
2
probe at r.t. produced some CO by reaction of CO and oxygen
2
140
d
adsorbed on gold clusters and big amounts of mono-, bi- and
polydentate carbonate species are observed (see Figure S1). On the
contrary, only oxidised gold species exposed at the surface of the
nanoparticles are observed after methanol oxidation reaction, as
0.1
0.1
2113
2
149
2
118
2106
-1
indicated by the presence of a band centred at 2113 cm (Figure
d). It is worth noting that no CO is adsorbed at r.t. on the Ce10Co
3
2
250
2200
2150
2100
2050
2000 2250
2200
2150
-1
Wavenumbers [cm ]
2100
2050
2000
sample, indicating that neither cerium nor cobalt cations are able to
adsorb the probe in our experimental conditions. However,
previous experiments carried out at low temperature demonstrated
-1
Wavenumbers [cm
]
Figure 3. FTIR difference spectra collected upon CO inlet (15 mbar)
-3
at r.t. and subsequent pressure decreases down to 1.0 x 10 mbar
on AuCe and on AuCe10Co before (sections a and c) and after
4+
the presence of Ce at the surface of both samples. Interestingly,
x+
the presence of bands related to Co cations was hardly
(
sections b and d) methanol oxidation reaction. The spectra are
19
hypothesized on AuCe10Co.
normalised to the density of the pellets.
The AuCe and AuCe10Co catalysts were contacted with methanol (5
mbar) and then oxygen (5 mbar) at r.t. and then the temperature
was gradually increased up to 100°C in order to follow the
formation (and the possible evolution) of surface species during
methanol oxidation by in situ FTIR spectroscopy. Notwithstanding
that in situ conditions, under static equilibrium pressure of the
reactants are far away from the ideal operando conditions where
reactants are sent in a plug-flow reactor, FTIR spectroscopy can
provide insights on both spectators and intermediate species
formed on the different catalysts. Firstly, the inlet of methanol at
r.t. on AuCe and AuCe10Co (blue lines in Figure S2) produced
different amounts of on top, doubly (two species) and triply bridged
methoxy species, as summarised in Table 2, where the vibrations
related to the methoxy species adsorbed on AuCe and on
AuCe10Co catalysts are reported. The evolution of the methoxy
species at increasing temperature is shown in Figure S2 (from
-
1
Besides a shoulder at 2112 cm , related to CO on partially oxidized
3
7
-1
metallic gold, a broad absorption at 2138 cm is observed after
the inlet of 15 mbar CO at r.t. on AuCe (blue line). Upon CO
pressure decreases at the same temperature (black lines) the band
-
1
at 2138 cm decreases in intensity without changing position,
whilst the shoulder at lower wavenumber is totally depleted. The
residual absorption is irreversible to the outgassing at r.t. (down to
-3
1
.0 x 10 mbar, red line), giving an anomalous evidence of a strong
bond between CO and the involved adsorption sites, since CO is
usually reversibly adsorbed on metallic gold. Some of us reported
on a similar band some years ago and assigned it to CO on positively
+
charged Aun clusters, where 4≤ n ≤6 stabilized on the oxidized
3
8
surface of ceria. This assignment was supported by DFT
3
9
calculations of Wu et al. , showing that the adsorption energy of
CO on small cationic, neutral, and anionic clusters progressively
increases with the electron depletion from the cluster. Also the
calculated C-O stretching frequencies supported this assignment as
they increased with increasing the positive charge on the clusters.
brown to red lines). On top and different doubly bridged methoxy
41, 42
species have been detected on both samples
, starting from
room temperature (blue line) up to 50 °C (brown lines), then a
marked increase in intensity of the bands related to on top methoxy
species is observed at 75 °C (violet lines). Methoxy species are
definitely more abundant on AuCe10Co than on AuCe, in particular
the intensity of the absorptions due to the doubly bridged methoxy
4
0
In addition, also the paper by Fielicke et al. in which the CO
+
vibrational frequencies of Au (CO)m complexes in the gas phase (3
n
≤
n ≤ 10 and 3 ≤ m ≤ 8) have been measured experimentally,
corroborated our assignment together with further spectroscopic
findings. Cationic gold clusters are still present after methanol
species with different coordination of Ce cations with oxygen ions
−1
(
bands at 1053 and 1038 cm ) seems enhanced on AuCe10Co,
-1
oxidation reaction, as signalled by the broad band at 2121 cm ,
whilst no bands related to oxidised gold species exposed at the
surface of the nanoparticles have been detected (Figure 3b). The
inlet of the CO probe at r.t. on the AuCe10Co catalyst (Figure 3c)
further indicating that the addition of cobalt leads to a lower
degree of coordination of the ceria surface sites of the doped
sample and as a consequence to an enhanced reactivity towards
the same amount of methanol. It has been reported that methoxy
-1
produced a band at 2149 cm (red line) that despite the decrease
of the CO pressure, firstly increases in intensity and red shifts at
species formed upon methanol dissociation over CeO
2
behave both
as spectators and intermediate species, depending on their
-
1
2
140 cm (orange line) and then decreases in intensity and moves
36
proximity with gold nanoparticles.
It can be proposed that on
-1
to 2135 cm (black lines down to violet line). Also in this case the
residual absorption is irreversible to the outgassing at r.t. (violet
AuCe10Co, that is the most active catalyst, the higher is the metal
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do notadjustmargins

Please do notadjustmargins
Journal of Materials Chemistry A
Page 6 of 13
View Article Online
DOI: 10.1039/C6TA06442F
ARTICLE
Journal Name
dispersion, the larger is the amount of active methoxy species
acting as reaction intermediates.
1
572
a
The above findings are also in agreement with the results of the
previous analysis of the HRTEM diffraction fringes related to the
ceria-containing phases, indicating an increase of about four times
of the spacings due to defective ceria phases for AuCe10Co
1.0
1564
1
549
371
1102
1043
1
19
(37.35%) if compared to AuCe (10.77 %).
1358
1299
1
507
The presence of more defective ceria may also favour the formation
of double bridged reactive methoxy species on the surface of the
catalyst. In addition, triply bridged methoxy species are easily
1
053
1022
856
1250
25, 26
distinguishable on AuCe10Co also at 100 °C,
whilst they are
almost depleted at the same temperature on AuCe, as indicated by
-
1
the weak bands at 1026, 2847 and 2934 cm . These features
further indicate that, due to the presence of cobalt, AuCe10Co
displays a more easy activation of methanol and an enhanced
reactivity of methoxy species if compared to AuCe.
b
The FTIR spectra collected on AuCe and on AuCe10Co in contact
with the methanol/oxygen mixture at increasing temperature are
reported in Figure 4 section a and b, respectively.
1.0
On the AuCe sample, the methanol oxidation was mainly limited to
the production of bidentate (bands at 2845, 1572, 1549, 1398, 1371
-
1
-1
cm ) and monodentate (bands at 1605 and 1250 cm ) formate
4
3
species. However, bands related to monodentate (1507 and 1358
-
1
-1
cm ), bidentate (1564, 1299, 1022, 856 cm ) and polydentate
-1
27
(
1475, 1053 and 833 cm ) carbonate species start to increase in
intensity at 75 °C, simultaneously with the decrease of the bands
related to methoxy species (see Figure 5 and Figure S2, where the
methoxy region in the 1150−950 cm− region is zoomed).
1
4000
3500
3000
1500
1000
The FTIR results put in evidence that on both catalysts methanol
reacted with the surface giving rise to different methoxy species at
r.t. and to formates at 50 °C. These species further evolved to
-
1
Wavenumbers [cm ]
mono-, bi- and polydentate carbonates by increasing the Figure 4. FTIR difference spectra collected on AuCe (section a) and
temperature at 75 °C and up to 100°C. On AuCe10Co larger AuCe10Co (section b) in contact with the methanol/oxygen mixture
amounts of carbonates are produced.
after 10 min at r.t. (blue lines), at 50 °C (brown lines), 75 °C (violet
lines), at 100 °C(orange lines) and after 20 min (red lines) at the
same temperature.
Table 2. Wavenumbers for vibrations of on top, doubly (two
species) and triply bridged methoxy species adsorbed on AuCe and
on AuCe10Co catalysts.
Differently from what previously observed in a combined steady
3
6
state isotopic transient kinetic analysis and operando FTIR study
,
Vibrational
modes
νa(CH )
cm
2
cm
δ
(CH )
νs(CH )
cm
ν(OC)
-1
cm
3
3
3
-
1
-1
-1
bands related to the presence of carbonate species were also
detected due to the adopted experimental conditions (static
On top methoxy species
3
6
conditions rather than flow conditions reported in the literature ).
The presence of carbonates indicates that formate species are
AuCe
2911
2911
2884
2881
2801
2800
1105
1105
AuCe10Co
converted into CO
2
, that evolves into surface carbonates in static
Doubly bridged methoxy species
conditions, whereas they would be flushed away in dynamic flow.
AuCe
2923
2936
2834
2838
2800
2801
1050
1053
3
6
The comparison between in situ and operando experiments
indicates that both methoxy and formate species are intermediates
and not simply spectators.
AuCe10Co
Doubly bridged methoxy species with different coordination of Ce
cations with oxygen ions
The thermal methanol decomposition was also performed on both
catalysts in order to compare the results with experiments carried
out without oxygen: it was observed that on AuCe on top methoxy
species disappeared along with the formation of formate species at
100°C, simultaneously to the conversion of doubly bridged methoxy
with different coordination into more coordinated doubly bridged
methoxy species (Figure S3). Therefore, the methoxy species with
different coordinative unsaturation could be related to the formate
AuCe
2910
2911
2841
2838
2807
2809
1039
1038
AuCe10Co
Triply bridged methoxy species
AuCe
2934
2949
2847
2836
-
-
1026
1023
AuCe10Co
6
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do notadjustmargins

Page 7 of 13
Journal Name
Jou rPn l ea al soe f dMo ant oe tr ai ad lj su s Ct mh ea rmg iins st ry A
View Article Online
DOI: 10.1039/C6TA06442F
ARTICLE
AuCe_ox
production. On the contrary, in the case of AuCe10Co no formate
species have been observed and only very intense bands due to
carbonate-like species are produced (Figure S4), whereas the
behaviour of the methoxy species is the same as observed on AuCe
for the same reaction. These results demonstrate that in the
absence of oxygen as a reactant, the AuCe10Co catalyst is able to
efficiently provide oxygen to decompose methanol, further
confirming that on this system the addition of Co causes a
significant enhancement in the reducibility, which resulted in an
a
AuCe5Co_ox
AuCe10Co_ox
AuCe15Co_ox
Au metal
-4
.0 Å
1
1
8
improved oxidation activity. However, the same reaction carried
out on the bare Ce10Co support did not give rise to any product
(Figure S5) indicating that there must be a positive synergy
involving gold to enhance oxygen mobility and activation.
0
1
2
3
4
5
R (Å)
It has been reported that promoting supported gold nanoparticles
with lanthanum oxide largely increases the hydrogen selectivity in
AuCe10Co_ox
fit
Au-O contribution
b
the partial oxidation of methanol, due to the fact that La
2 3
O was
found to facilitate the reduction of initially present oxidic gold
Au-Au contribution
4
4
-4
species. If compared with the results related to AuCe, the CO
adsorption experiments at room temperature revealed the
presence of more positivized gold sites on AuCe10Co before
methanol oxidation reaction (Figure 3c). These sites could be
responsible of the enhanced reactivity displayed by AuCe10Co and
appear also to be reduced in higher extent after reaction (Figure
1.0 Å
3d), demonstrating that an effect of the addition of cobalt on gold is
occurring.
Insights on the structure of the active sites
0
1
2
3
4
5
R (Å)
In order to investigate the effect of the Co addition to the ceria
substrate on the average local environment of supported gold (5, Figure 5.Section (a): modulus of k -weighted, phase uncorrected, FT
3
1
0 and 15 wt.%) Au L
the AuCe, AuCe5Co, AuCe10Co and AuCe15Co catalysts previously differently Co doped ceria substrate and submitted to thermal
submitted to a thermal treatment in O
at 400°C, that is the treatment in oxygen at 400°C. Corresponding imaginary parts are
activation procedure to which the catalyst underwent before the reported in Figure S5. Section (b): Comparison between the
catalytic tests.
experimental (scattered black dots) and the best fit (red full line) for
edge EXAFS experiments were performed on of the Au L -edge EXAFS spectra of the catalysts prepared on
3
3
2
3
The k -weighted, phase uncorrected, FT of the EXAFS spectra AuCe10Co_ox. Also reported are the imaginary parts of the
collected on all catalysts are shown in Figure 5, section a. The experimental data (scattered black dots) and the modulus (full
analogous spectrum collected on Au foil is also reported for lines) and the imaginary parts (dashed lines) of the individual Au-O
comparison (grey line).
(blue lines) and Au-Au (orange lines) contributions.
All spectra are characterized by two main contributions. The first
one, appearing in the 1.0-2.0 Å region and having a maximum at
about 1.55 Å, is due to the single scattering Au-O of an Au oxide The quantitative results confirm what observed previously on the
phase. The second one contributes in the 2.0-3.5 Å region and, simple basis of the intensity of the Au-O component at 1.55 Å: the
independently from the presence of two maxima at 2.39, 2.98 Å, is sample with the highest oxidized fraction is AuCe10Co, followed by
mainly due to the first shell single scattering Au-Au of supported AuCe5Co and then by AuCe15Co, that is comparable with the
4
5
metal nanoparticles. This assignment is supported by the undoped AuCe. Conversely, no trend can be put in evidence for the
imaginary parts of the FT (see Figure S6). Moreover, the AuCe10Co Au-Au contributions that are formally equivalent within the
catalyst shows the highest intensity of Au-O component due to the experimental error bars. It should be noted that in this case the
interaction between the active gold nanoparticles and the oxygen intensity of the Au-Au contribution is inversely proportional to the
atoms of the oxidic support. This EXAFS evidence is in accordance fraction of the oxidized phase discussed above but is also influenced
4
7, 48
with the FTIR characterization results here discussed and with by the different particle size distribution,
so that the two effect
previous high angle annular dark field observations that revealed cannot be directly disentangled in a straightforward manner. On
the smallest average size of gold particles on this catalyst if top of these the region between 2.0-3.5 Å is expected to host also
1
9, 46
compared to AuCe5Co and AuCe15Co.
The quantitative results obtained from the EXAFS fits on all the neglected in these refinements. This fact also explains the minor
the higher shell contributions from oxide phase, that have been
samples are summarized in Table 3.
discrepancy between experimental data and fit reported in Figure
b.
5
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do notadjustmargins

Please do notadjustmargins
Journal of Materials Chemistry A
Page 8 of 13
View Article Online
DOI: 10.1039/C6TA06442F
ARTICLE
Journal Name
The methanol oxidation reaction was then investigated on
Au10CoCe sample, being the most active catalyst: the first set of
spectra was collected at room temperature. Furthermore, starting
from the catalysts previously oxidized, another set of spectra was
recorded under operando conditions, flowing methanol and oxygen,
-4
starting from room up to 120 °C. Au/CeO
2
was measured as
1.0 Å
reference sample, too. Three subsequent χ(k) spectra were
measured at each temperature, to evaluate the experimental
49
uncertainty. The data analysis was performed on the averaged
χ(k) spectrum by non-linear fitting procedures. Figure 6 shows the
3
k weighted, phase uncorrected FT (module and imaginary part) of
Au10CoCe_met
fit
Au-Co contribution
Au-Au contribution
the AuCe10Co sample collected during methanol oxidation reaction
at 120°C (black scattered points). The signal due to the Au-O
5
0
distance at 1.55 Å disappears during reaction while, a new
contribution at about 1.8 Å is observed. The component at high
distances (2.0-3.5 Å range) is clearly due to Au-Au contributions,
1
2
3
4
R (Å)
3
whereas the new component around 1.8 Å is associated to an Au- Figure 6. Modulus of the k -weighted, phase uncorrected, FT of the
Co contribution. The EXAFS fit of the experimental data performed Au L -edge EXAFS spectrum of AuCe10Co collected during methanol
3
assuming such structural model (red line in Figure 6) clearly shows oxidation reaction at 120 °C (black scattered dots) compared with
the good agreement with the experimental spectrum, validating the the fit (red curve). Also reported are the imaginary parts of the
presence of Co in the first coordination shell of a fraction of Au experimental data (black scattered dots) and the modulus (full
atoms. The quantitative results of the fit are reported in Table 4. lines) and the imaginary parts (dashed lines) of the individual Au-Co
The EXAFS study, corroborated by FTIR measurements, performed (purple lines) and Au-Au (orange lines) contributions.
on all catalysts indicate that oxidized gold species are involved in
1
8
the methanol reaction (Figure 5 and related data analysis in Table the bands during CO and O2 interaction on all the samples is
). In addition, the EXAFS analysis of the catalyst in working shown starting from –180°C (orange lines, i.e. immediately after the
3
1
8
condition (Figure 6) suggests that the active sites present on the inlet of O2 on preadsorbed CO), during the heating (black and
most active sample (with 10 wt% Co ) are those where gold sites purple lines) and up to room temperature (blue lines).
3 4
O
are strongly interacting with Co. On these active sites reactive oxygen
species, possibly close to Co sites can be activated and participate
to the oxidative dehydrogenation reaction of methanol to formate Table 4. Result of the fit performed on AuCe10Co_met EXAFS
3
6
species . Moreover, due to the synergic effect between cobalt and spectrum.
gold, the AuCe10Co catalyst is also able to provide reactive oxygen
species in reaction conditions: reduced gold species close to oxygen
vacancies and in strong interaction with cobalt ions are formed Paths
2
2
CN
σ (Å )
Å
R
(Å)
E0 (eV)
-2
Å
eV
after reaction. Such Au species are those located at the interface Au – Co 0.3±0.1
2.09±0.03
0
.010±0.001
3.6±0.7
with the support. These features can be at the origin of the Au – Au 8.8±0.8
2.835±0.007
2
enhanced catalytic activity of AuCe10Co.
S0 = 0.85 fixed from the value extracted from the fit of the Au foil
2
0
–1
(
S
= 0.85±0.8); Δk = (3.3–10.0) Å and ΔR = (1.4–3.5) Å, resulting
Promotional role of Co on oxygen mobility
in a number of independent points larger than 11; R factor = 0.005.
Spectroscopic evidence of the effect of the support composition on
the oxygen mobility is reported in Figure 7, where the evolution of
Table 3. Results of the fit performed on oxidized samples.
2
σ
2
CN
(Å
)
R (
Å
)
E
0
(eV)
r factor
Au – O
1.2±0.3
1.7±0.3
2.2±0.4
1.2±0.2
Au – Au Au – O
Au – Au
Au – O
Au – Au
AuCe
5.5±1.1
4.6±0.5
5.0±0.6
4.7±0.3
0.006±0.001
0.004±0.002
1.96±0.02 2.854±0.012
1.95±0.01 2.848±0.006
1.94±0.02 2.852±0.008
5.3±2.2
0.002
0.006
0.010
0.003
AuCe5Co
AuCe10Co
AuCe15Co
1.95±0.01 2.850±0.004
2
2
–1
S
0
= 0.85 fixed from the value extracted from the fit of the Au foil (S
0
= 0.85±0.8); Δk = (3.3–10.0) Å and ΔR = (1.4–3.5) Å, resulting in a
number of independent points larger than 11.
8
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do notadjustmargins

Please do notadjustmargins
Journal of Materials Chemistry A
Page 9 of 13
View Article Online
DOI: 10.1039/C6TA06442F
Journal of Materials Chemistry A
ARTICLE
a 1573
a
2153
1
298
0
.5 A
0
.5 A
823
1
612
1211
2
155
2
2
121
102
1025 853
2
172
2
337
1
568
b
b
2
152
0.5 A
0
.5 A
1600
1399
1
285
1
2
341
2
355
216
820
006 851
2
158
2
127
105
1
2
173
2
2
153
c
c
0
.5 A
2
342
2326
2155
171
183
0.5 A
1
601
568
1
1392
1286
2
353
2161
2
1
213
820
51
2
2
125
103
8
2
1020
2
153
d 1606 1579 1399 1288
d
0.5 A
0
.5 A
1
220
2
157
822
1009 854
2
125
2
357
340
2326
2
173
2
2102
2
400 2350 2300 2250 2200 2150 2100 2050
1800
1600
1400
1200
1000
800
-1
-1
Wavenumbers [cm ]
Wavenumbers [cm ]
Figure 7. Evolution of FTIR difference spectra collected on AuCe (section a), AuCe5Co (section b), AuCe10Co (section c) and AuCe15Co
1
8
(
section d) oxidised at 400 °C immediately after the inlet of
times and temperature (black and purple lines) up to room temperature (blue lines) in the 2000-2400 cm (left) and in the 1800-800 cm
right) regions. The spectrum collected before CO dosage has been used to obtain the reported difference spectra, that have been
normalised to the density of the pellets.
2
O at -180 °C on preadsorbed CO (orange lines) and at increasing contact
-1
-1
(
1
8
The inlet of
2 2
O at increasing diffusion times and temperature temperature. Simultaneously, bands related to CO formation
(
black and purple lines up to blue line) over the samples previously during CO oxidation are produced at higher frequencies
5
2
saturated by CO at -180 °C (orange line) caused an erosion from the (asymmetric CO stretching or ν3 mode ). Such absorptions have
low-frequency side of the band at about 2102-2105 cm , due to CO the highest intensity in the case of AuCe10Co (section c) and their
on Au particles covered by adsorbed oxygen , a gradual decrease intensity follows the same trend as for the catalytic activity:
in intensity of the component at 2121-2127 cm due to CO AuCe10Co > AuCe5Co > AuCe15Co>AuCe. More in detail, the three
adsorbed on cationic gold clusters
the bands related to the support (component at 2150 cm due to 2326 cm , assigned to the C O O solid-like phase, accompanied by
CO in interaction OH groups and bands at 2155-2161 cm and at bands at 2340-2342 and at 2353-2357 cm , assigned to C O and
2
2
-1
3
7
-1
38-40
and the rapid depletion of isotopomers of the CO molecule were observed: a growing band at
2
-1
-1
16 18
-1
-1
16
2
-1
4+
43
18
53
171-2173 cm due to CO on different Ce sites ). Only in the case C O solid-like phase, respectively . The spectra showed that a
2
-1
18
of AuCe10Co a band at 2183 cm , tentatively assigned to CO rapid exchange between the oxygen of ceria and the O molecules
adsorbed on Co sites is observed during the heating at room coming from the gas phase occurred already at low temperature.
2
x+
51
This journal is © The Royal Society of Chemistry 20xx
J. Materials Chemistry aName., 2013, 00, 1-3 | 9
Please do notadjustmargins

Please do notadjustmargins
Journal of Materials Chemistry A
Page 10 of 13
View Article Online
DOI: 10.1039/C6TA06442F
ARTICLE
Journal Name
1
6
Moreover, the high intensity of the band ascribed to C O indicated samples. A strong influence of the amount of the Co promoter was
2
that oxygen participating in the reaction at low temperature mainly
comes from the support, putting in evidence the fact that the
catalyst promotion with Co led to increased oxygen mobility
because of the improved exchange properties of the ceria support.
The results demonstrated the enhanced ability of the Co-promoted-
ceria support to supply active lattice oxygen that is beneficial for
the reaction and were in good agreement with already published
TPR results for highest hydrogen consumption and shift ot Tmax to
lower temperature observed in the profile of AuCe10Co sample.
Here we recall that these experiments were performed on catalysts
oxidised at 400 °C, i.e. the temperature used for the activation
observed. The activity trend followed the order: AuCe10Co >
AuCe5Co > AuCe15Co >AuCe>> Au/Co TOFs have been
3
4
O .
calculated for all catalysts, and it was found that the TOF values
obtained for AuCe10Co are one order of magnitude higher than
those related to the other catalysts. Moreover, with the exception
of AuCe10Co, for which the TOF value is almost unchanged, the
conversion is strongly dependent on the temperature.
Different in situ FTIR experiments were performed. Adsorption of
CO on AuCe and AuCe10Co indicate that oxidized gold sites are
initially present on both activated samples, suggesting that such
1
7
species are involved in the methanol reaction. Upon CH
different methoxy groups are formed on both catalysts at r.t. (more
abundantly on AuCe10Co). After successive O addition they are
3
OH dosage,
before performing the catalytic tests on CH OH oxidation, see
3
Figure 1, Table 1 and related discussion.
2
consumed on both catalysts starting at about 75 °C and giving rise
to mainly formate species on AuCe and to big amounts of different
carbonate species on AuCe10Co. Moreover, the synergy between
gold and Co-promoted ceria support causes significant
enhancement of reducibility and capability for oxygen activation,
which resulted in an improved oxidation activity, as demonstrated
by the FTIR experiments carried out in the absence of oxygen. The
capability of AuCe10Co to supply active oxygen species involves the
participation of the oxygen vacancies on modified ceria as well as of
On the stability of the catalysts
The analysis of the spectra reported in Figure 7 (right) indicated
that on all the samples mainly two kinds of carbonate-like species
were formed simultaneously to the production of CO (see the
2
spectra reported in the right parts of Figure 7). Such species have
different abundance, depending on the composition of the four
-
1
catalysts. Hydrogen carbonates with peaks at 1600-1612 cm ,
-
1
-1
1
392-1399 cm and the component around 1020 cm
(υCO
3
-1
-1
modes), at 820-823 cm (πCO
3
) and at 1211-1220 cm
(δOH) were the surface of separate, highly reducible, CoO phase.
x
4
3
detected. In addition, bidentate carbonates are also produced, as As suggested by EXAFS analyses, the active sites present on the best
-1
-1
revealed by the absorptions at 1568-1579 cm , at 1285-1298 cm
performing AuCe10Co catalyst are oxidized gold species at the
interface with the support and close to Co sites, which are able to
-1
-1
43
and at 1006-1009 cm
3 3
(υCO modes) and at 851-854 cm (πCO ).
-
1
Please note that approximatively in the 1350-1400 cm region activate and to react with oxygen giving rise to the formation of
there must be the symmetric CO stretching (or mode) of the CO
formate species and of carbonate species. The presence of a
molecules giving rise to the
mode (complex band at 2320-2360 multiplet of bands in the CO asymmetric stretching region after CO
clearly reveals that the promotion of ceria by
2
ν
1
2
ν
3
2
-1
18
cm discussed above). For symmetry reason this band is IR inactive interaction with
for CO
molecules in the gas phase, but becomes IR active upon Co facilitates the exchange reaction with the
surface adsorption because of symmetry break. However, as the support. Moreover, it is demonstrated that the production of
absorption coefficient of the mode of adsorbed CO
is typically carbonate species is lowered in the presence of cobalt oxide,
one order of magnitude lower than that of the
that we can neglect their presence in the spectra reported in the of an improved stability of the Co-promoted catalysts.
right parts of Figure 7.
The correlation of the catalytic activity and of a well-defined metal
O
2
1
6
2
O atoms of the
2
ν
1
2
5
2
ν
3
2
mode , meaning despite the CO formation is increased. Such results give evidence
Summarizing this isotopically labelled IR study, it is found that the dispersion with the findings, provided by in situ FTIR and operando
presence of another oxide to ceria has a remarkable influence on EXAFS studies about the nature and structure of the active sites
the intensity of the bands in the carbonate region: on the AuCe allows to bridge the gap of experimental data previously obtained
catalyst (section a) the highest amount of hydrogen carbonate as by HAADF, XRD, FTIR, TPR and to clarify the precise nature of the
well as of bidentate carbonate species is adsorbed on the surface synergy between the gold phase and the Co-promoted ceria
after reaction if compared to the Co-promoted catalysts. Indeed, support, which resulted in an excellent CH
3
OH oxidation activity.
the intensity of such bands decreases with increasing the amount of
produced CO
of ceria and therefore the carbonate species formation results Acknowledgements
inhibited, despite the CO formation. AuCe10Co is the most active
catalyst, but also and the most stable one, due to the highest
amount of produced CO and the lowest amount of carbonate-like
2
, giving evidence that cobalt oxide lowers the basicity
2
M. Manzoli and F. Vindigni gratefully acknowledge the IIT (Istituto
Italiano di Tecnologia) for financial support (NANO GOLD Project).
The Bulgarian authors gratefully acknowledge the financial support
by the Bulgarian National Science Fund (Project DFNI T 02/4). C.
2
species present at the surface after reaction. This represents a
remarkable achievement for the promotion of the gold on ceria
catalyst with Co, as in most of the cases in catalysis an increase of Lamberti acknowledges the Mega-grant of the Russian Federation
the catalyst activity is accompanied by a faster deactivation time Government to support scientific research at the Southern Federal
and compromises must be found about these two aspects of the University, no. 14.Y26.31.0001.
promotion.
References
Conclusions
1
.
S. Scire and L. F. Liotta, Appl. Catal. B-Env., 2012, 125
22-246.
,
The catalytic performance in methanol oxidation of different gold
2
catalysts supported on ceria modified by the addition of Co
3 4
O was
investigated. Au/CeO and Au/Co were also studied as reference
2
3 4
O
1
0 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do notadjustmargins

Page 11 of 13
Journal Name
Jou rPn l ea al soe f dMo ant oe tr ai ad lj su s Ct mh ea rmg iins st ry A
View Article Online
DOI: 10.1039/C6TA06442F
ARTICLE
2
.
T. Tabakova, G. Avgouropoulos, J. Papavasiliou, M.
28.
29.
30.
C. Lamberti, A. Zecchina, E. Groppo and S. Bordiga, Chem.
Soc. Rev., 2010, 39, 4951-5001.
Manzoli, F. Boccuzzi, K. Tenchev, F. Vindigni and T.
Ioannides, Appl. Catal. B: Env., 2011, 101, 256-265.
A. Taketoshi and M. Haruta, Chem. Lett., 2014, 43, 380-
S. Bordiga, C. Lamberti, F. Bonino, A. Travert and F.
Thibault-Starzyk, Chem. Soc. Rev., 2015, 44, 7262-7341.
M. Manzoli and F. Vindigni, in Gold Catalysis, eds. L. Prati
and A. Villa, Eds., Pan Stanford Publishing Pte. Ltd.,
Singapore, 2016, ch. 8, pp. 205-252.
O. Mathon, A. Beteva, J. Borrel, D. Bugnazet, S. Gatla, R.
Hino, I. Kantor, T. Mairs, M. Munoz, S. Pasternak, F. Perrin
and S. Pascarelli, J. Synchrotron Rad., 2015, 22, 1548-
1554.
3.
4.
5.
6.
3
87.
M. Melchionna and P. Fornasiero, Materials Today, 2014,
, 349-357.
17
P. Dutta, S. Pal, M. S. Seehra, Y. Shi, E. M. Eyring and R. D.
Ernst, Chem. Mater., 2006, 18, 5144-5146.
T. Tabakova, L. Ilieva, I. Ivanov, R. Zanella, J. W. Sobczak,
W. Lisowski, Z. Kaszkur and D. Andreeva, Appl. Catal. B-
Env., 2013, 136, 70-80.
O. H. Laguna, A. Perez, M. A. Centeno and J. A. Odriozola,
Appl. Catal. B-Env., 2015, 176, 385-395.
F. Vindigni, M. Manzoli, T. Tabakova, V. Idakiev, F.
31.
32.
33.
M. Newville, J. Synchrotron Rad., 2001, 8, 96-100.
7
.
.
S. J. A. Figueroa, D. Gibson, T. Mairs, S. Pasternak, M. A.
Newton, M. Di Michiel, J. Andrieux, K. C. Christoforidis, A.
Iglesias-Juez, M. Fernandez-Garcia and C. Prestipino, J.
Appli. Cryst., 2013, 46, 1523-1527.
S. Scire, P. M. Riccobene and C. Crisafulli, Appl. Catal. B-
Env., 2010, 101, 109-117.
S. Ivanova, V. Pitchon, C. Petit and V. Caps,
ChemCatChem, 2010, 2, 556-563.
I. Castellanos, P. Bazin, S. Thomas, O. Marie and M.
Daturi, Top. Catal., 2016, 59, 337-346.
F. Boccuzzi, A. Chiorino, M. Manzoli, P. Lu, T. Akita, S.
Ichikawa and M. Haruta, J. Catal., 2001, 202, 256-267.
M. Manzoli, F. Boccuzzi, A. Chiorino, F. Vindigni, W. Deng
and M. Flytzani-Stephanopoulos, J. Catal., 2007, 245, 308-
315.
8
Boccuzzi and A. Chiorino, Appl. Catal.B-Env., 2012, 125
07-515.
,
5
34.
35.
36.
37.
38.
9.
C. Pojanavaraphan, A. Luengnaruemitchai and E. Gulari,
Int.J. Hydrogen Energy, 2013, 38, 1348-1362.
O. H. Laguna, M. A. Centeno, G. Arzamendi, L. M. Gandia,
F. Romero-Sarria and J. A. Odriozola, Catal. Today, 2010,
10.
157, 155-159.
1
1.
2.
L. Ilieva, T. Tabakova, G. Pantaleo, I. Ivanov, R. Zanella, D.
Paneva, N. Velinov, J. W. Sobczak, W. Lisowski, G. Avdeev
and A. M. Venezia, Appl. Catal. A-Gen., 2013, 467, 76-90.
T. R. Reina, S. Ivanova, V. Idakiev, J. J. Delgado, I. Ivanov,
T. Tabakova, M. A. Centeno and J. A. Odriozola, Catal. Sci.
1
Tech., 2013,
T. R. Reina, A. A. Moreno, S. Ivanova, J. A. Odriozola and
M. A. Centeno, ChemCatChem, 2012, , 512-520.
A. Alvarez, S. Ivanova, M. A. Centeno and J. A. Odriozola,
Appl. Catal. a-Gen., 2012, 431, 9-17.
X. W. Xie, Y. Li, Z. Q. Liu, M. Haruta and W. J. Shen,
Nature, 2009, 458, 746-749.
Y. Liu, B. C. Liu, Q. Wang, C. Y. Li, W. T. Hu, Y. X. Liu, P.
Jing, W. Z. Zhao and J. Zhang, J. Catal., 2012, 296, 65-76.
B. C. Liu, Y. Liu, C. Y. Li, W. T. Hu, P. Jing, Q. Wang and J.
Zhang, Appl. Catal. B-Env., 2012, 127, 47-58.
L. Ilieva, P. Petrova, T. Tabakova, R. Zanella, M. V.
Abrashev, J. W. Sobczak, W. Lisowski, Z. Kaszkur and D.
Andreeva, Catal. Today, 2012, 187, 30-38.
T. Tabakova, D. Dimitrov, M. Manzoli, F. Vindigni, P.
Petrova, L. Ilieva, R. Zanella and K. Ivanov, Catal. Comm.,
3
, 779-787.
39.
40.
X. Wu, L. Senapati, S. K. Nayak, A. Selloni and M. Hajaligol,
J. Chem. Phys., 2002, 117, 4010-4015.
A. Fielicke, G. von Helden, G. Meijer, D. B. Pedersen, B.
Simard and D. M. Rayner, J.Am. Chem. Soc., 2005, 127,
8416-8423.
1
1
1
1
1
1
3.
4.
5.
6.
7.
8.
4
41.
A. Badri, C. Binet and J. C. Lavalley, J. Chem. Soc.-Faraday
Trans., 1997, 93, 2121-2124.
C. Binet and M. Daturi, Catal. Today, 2001, 70, 155-167.
C. Binet, M. Daturi and J. C. Lavalley, Catal. Today, 1999,
50, 207-225.
B. P. C. Hereijgers, T. M. Eggenhuisen, K. P. de Jong, H.
Talsma, A. M. J. van der Eerden, A. M. Beale and B. M.
Weckhuysen, J. Phys. Chem. C, 2011, 115, 15545-15554.
J. T. Miller, A. J. Kropf, Y. Zha, J. R. Regalbuto, L. Delannoy,
C. Louis, E. Bus and J. A. van Bokhoven, J. Catal., 2006,
240, 222-234.
N. Weiher, E. Bus, L. Delannoy, C. Louis, D. E. Ramaker, J.
T. Miller and J. A. van Bokhoven, J. Catal., 2006, 240, 100-
107.
G. Agostini, A. Piovano, L. Bertinetti, R. Pellegrini, G.
Leofanti, E. Groppo and C. Lamberti, J. Phys. Chem. C,
2014, 118, 4085-4094.
G. Agostini, R. Pellegrini, G. Leofanti, L. Bertinetti, S.
Bertarione, E. Groppo, A. Zecchina and C. Lamberti, J.
Phys. Chem. C, 2009, 113, 10485-10492.
C. Lamberti, S. Bordiga, D. Arduino, A. Zecchina, F.
Geobaldo, G. Spano, F. Genoni, G. Petrini, A. Carati, F.
Villain and G. Vlaic, J. Phys. Chem. B, 1998, 102, 6382-
6390.
42.
43.
44.
45.
46.
47.
48.
49.
19.
2013, 35, 51-58.
2
2
0.
1.
H. J. Freund, Ang. Chem.-Int. Ed., 1997, 36, 452-475.
S. Bordiga, E. Groppo, G. Agostini, J. A. van Bokhoven and
C. Lamberti, Chem. Rev., 2013, 113, 1736-1850.
C. Garino, E. Borfecchia, R. Gobetto, J. A. van Bokhoven
and C. Lamberti, Coord. Chem. Rev., 2014, 277, 130-186.
J. A. van Bokhoven and C. Lamberti, in X-Ray Absorption
and X-Ray Emission Spectroscopy: Theory and
Applications eds. J. A. van Bokhoven and C. Lamberti, Eds.,
John Wiley & Sons Ltd, Hoboken, NJ, 2016, vol. 2, ch. 13,
pp. 353-383.
2
2.
3.
2
2
2
4.
5.
A. I. Frenkel, Chem. Soc. Rev., 2012, 41, 8163-8178.
A. I. Frenkel, C. W. Hills and R. G. Nuzzo, J. Phys. Chem. B,
2
001, 105, 12689-12703.
50.
51.
W. Deng, A. I. Frenkel, R. Si and M. Flytzani-
Stephanopoulos, J. Phys. Chem. C, 2008, 112, 12834-
12840.
K. I. Hadjiivanov and G. N. Vayssilov, Adv. Catal., 2002, 47,
307-511.
2
6.
7.
C. Lamberti, E. Groppo, G. Spoto, S. Bordiga and A.
Zecchina, Adv. Catal., 2007, 51, 1-74.
A. Vimont, F. Thibault-Starzyk and M. Daturi, Chem. Soc.
Rev., 2010, 39, 4928-4950.
2
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 11
Please do notadjustmargins

Please do notadjustmargins
Journal of Materials Chemistry A
Page 12 of 13
View Article Online
DOI: 10.1039/C6TA06442F
ARTICLE
Journal Name
5
2.
3.
E. Garrone, B. Bonelli, C. Lamberti, B. Civalleri, M. Rocchia,
P. Roy and C. O. Arean, J. Chem. Phys., 2002, 117, 10274-
0282.
1
5
T. Tabakova, M. Manzoli, F. Vindigni, V. Idakiev and F.
Boccuzzi, J. Phys. Chem. A, 2010, 114, 3909-3915.
1
2 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do notadjustmargins

Page 13 of 13
Journal of Materials Chemistry A
CH OH + O2
carbonate species
3
methoxy species
in situ FTIR
operando EXAFS
Au0
Au
+
CoOX
Au/CeO doped by Co O
4
CeO2
2
3