The Effect of Rh on the Interaction of Co with Al2O3 and CeO2 Supports

Article abstract of DOI:10.1007/s10562-016-1809-3

Abstract: Co and Rh+Co catalysts supported on Al₂O₃ and CeO₂ were investigated by temperature programmed reduction (TPR) and X-ray photoelectron spectroscopy (XPS). The CeO₂ defects, resulted from the effects of the metals, were further analyzed by Raman spectroscopy and optical absorption. Although the interaction of Co with these two supports is fairly different, it can be concluded that Rh inhibits the strong Co-support interaction. It was revealed that Co over Al₂O₃ forms mainly Co²⁺ compounds, only a smaller fraction of cobalt is in metallic state. After the addition of 0.1?% Rh, Co₃O₄ like species is dominant, the amount of metallic state increased after reduction. Over CeO₂ the Co dissolution into the support was inhibited by Rh. A wide range of TPR results proved the stepwise reduction of Co, which was promoted by the addition of Rh. By Rh the entire mechanism of this process was altered that can be observed even by XPS. On the basis of the Raman and the optical measurements we concluded that the metals induce defect sites on the CeO₂ surface, and these appear as similar features on the spectra of Co and Rh containing samples, thus their density depends on the metal loading and not on the metal type. CeO₂ has a bandgap of 3.27?eV, which is not altered by the metals, but an electronic contact was detected between the metals and CeO₂ by photovoltammetry. The increased number of metallic species is mainly responsible for the higher catalytic activity and for the enhanced hydrogen selectivity in the stream reforming of ethanol. Graphical Abstract: [Figure not available: see fulltext.]

Full text of DOI:10.1007/s10562-016-1809-3

Catal Lett
DOI 10.1007/s10562-016-1809-3
The Effect of Rh on the Interaction of Co with Al2O3 and CeO2
Supports
1
1
•
Erika Varga1 Kornelia Baan Gergely Ferenc Samu1,2 Andras Erdohelyi
•
•
•
´
´
Albert Oszko Zoltan Konya
´
}
1
3,4
1,4
•
•
´
´
´
´
Janos Kiss
Received: 13 April 2016 / Accepted: 7 July 2016
Ó Springer Science+Business Media New York 2016
Abstract Co and Rh?Co catalysts supported on Al2O3
and CeO2 were investigated by temperature programmed
reduction (TPR) and X-ray photoelectron spectroscopy
(XPS). The CeO2 defects, resulted from the effects of the
metals, were further analyzed by Raman spectroscopy and
optical absorption. Although the interaction of Co with these
two supports is fairly different, it can be concluded that Rh
inhibits the strong Co-support interaction. It was revealed
that Co over Al2O3 forms mainly Co2? compounds, only a
smaller fraction of cobalt is in metallic state. After the
addition of 0.1 % Rh, Co3O4 like species is dominant, the
amount of metallic state increased after reduction. Over
CeO2 the Co dissolution into the support was inhibited by Rh.
A wide range of TPR results proved the stepwise reduction of
Co, which was promoted by the addition of Rh. By Rh the
entire mechanism of this process was altered that can be
observed even by XPS. On the basis of the Raman and the
optical measurements we concluded that the metals induce
defect sites on the CeO2 surface, and these appear as similar
features on the spectra of Co and Rh containing samples, thus
their density depends on the metal loading and not on the
metal type. CeO2 has a bandgap of 3.27 eV, which is not
altered by the metals, but an electronic contact was detected
between the metals and CeO2 by photovoltammetry. The
increased number of metallic species is mainly responsible
for the higher catalytic activity and for the enhanced
hydrogen selectivity in the stream reforming of ethanol.
Graphical Abstract
Keywords Rh-Co Á CoAl2O4 Á CeO2 defects Á TPR Á
Bandgap Á Photovoltammetry
1 Introduction
´
& Janos Kiss
jkiss@chem.u-szeged.hu
In heterogeneous catalysis the application of complex
catalysts, such as supported or bimetallic systems, has
resulted in significant progress in efficiency and even sta-
bility [1–4]. The role of the different oxide supports is not
only to hinder the sintering of the metal particles, but also
to have electronic and steric contact with them [5].
Moreover, some important reaction steps take place over
the support or the metal-support interface, such as the
ethylene or acetaldehyde formation from ethanol, the oxi-
dation of CO, and the dissociation of water [6–9].
1
Department of Physical Chemistry and Materials Science,
´
´
University of Szeged, Aradi vertanuk tere 1, Szeged 6720,
Hungary
2
3
4
MTA-SZTE ‘‘Lendu¨let’’ Photoelectrochemistry Research
´
Group, Rerrich B. ter 1, Szeged 6720, Hungary
Department of Applied and Environmental Chemistry,
´
University of Szeged, Rerrich B. ter 1, Szeged 6720, Hungary
MTA-SZTE Reaction Kinetics and Surface Chemistry
´
Research Group, Rerrich B. ter 1, Szeged 6720, Hungary
123

E. Varga et al.
In addition to the support effect, the physical, chemical
and catalytic behavior of a metal can be gently tuned by a
second metal, especially by noble metal promoters. Thus,
by the addition of trace amounts of the precious component
to a cheaper metal, like transition metals, catalyst with high
activity at low costs could be developed. The effect of the
promoter manifests in the activation of hydrogen during
pretreatments causing enhanced reduction, changing the
particle size distribution or they can induce electronic
interaction as well as alloy formation [10, 11].
2 Experimental
The catalysts were prepared by impregnating the supports
(Al2O3, Degussa P110 C1, 100 m2 g-1 or CeO2, Alfa
Aesar, 43 m2 g-1) with the aqueous solution of Co(NO3)2
to yield a nominal metal content of 2 or 10 wt%. The
impregnated powders were dried at 383 K and calcined at
973 K. The bimetallic Rh–Co samples were prepared by
sequential impregnation (impregnation with Co and calci-
nation first, then impregnation with Rh). Before the mea-
surements, the samples were oxidized at 673 K in flowing
O2 for 20 min and reduced at 773 K in H2 for 60 min.
Temperature programmed reduction experiments were
carried out by a BELCAT-A instrument, in a reactor
(quartz tube 9 mm outer diameter) that was externally
heated. Before the measurements, the catalysts were treated
in O2 at 673 K for 30 min. Thereafter, the samples were
cooled in flowing Ar to room temperature and let the
temperature equilibrate for 15 min. The oxidized sample
was flushed with Ar containing 10 % H2, the reactor was
heated linearly at a rate of 20 K min-1 up to 1373 K and
the H2 consumption was detected by a thermal conductivity
detector.
Regarding the base metal, beside the highly efficient Ni,
Co has gained attention in the last few years. It should be
mentioned that Co supported on Al2O3 or SiO2 is at least as
active in the Fischer–Tropsch synthesis as the classical Fe
systems, although the reaction pathways are different [12].
According to some former results this reaction is enhanced
with increasing the Co particle size and the ratio of the
metallic cobalt, as the active form is the Co0 and the small
particles tend to oxidize [13, 14]. From this point of view
the addition of noble metal promoters can facilitate the
reaction by enhancing the reduction. Similar considerations
can be made with the dry reforming of CH4 [15, 16].
However, in ethanol involving reactions the acidic Al2O3
favors the dehydration of the molecule and the polymer-
ization of the resulting ethylene, causing significant carbon
deposits [17, 18].
XP spectra were taken with a SPECS instrument
equipped with a PHOIBOS 150 MCD 9 hemispherical
electron energy analyzer, using Al Ka radiation
(hm = 1486.6 eV). The X-ray gun was operated at 210 W
(14 kV, 15 mA). The analyzer was operated in the FAT
mode, with the pass energy set to 20 eV. For data acqui-
sition and evaluation both manufacturer’s (SpecsLab2) and
commercial (CasaXPS, Origin) software were used. The
binding energy scale was corrected by fixing the Ce 3d u000
peak to 916.6 eV or the Al 2p to 74.7 eV.
Co has been tried in the stream reforming of ethanol
(SRE), too. At low Co content the product distribution was
determined by the support, the main product was ethylene
when supported on Al2O3, but acetone and a high portion
of H2 were formed over Co/CeO2. The latter effect was
improved by the addition of 0.1 % Rh [19]. In the literature
there is a disagreement relating to the active centers of Co:
some authors suggest that Co0 is responsible for coke
formation thus blocks the reaction. Others claim that
metallic Co is the necessary form for the SRE [20, 21]. In
some cases both Co0 and oxidized Co were present during
the reaction [22, 23]. Nevertheless, it is clear that the
addition of the noble metals enhances the reduction of
CoOx to metallic Co and the efficiency is improved in the
SRE and oxidative SRE [24–26].
For XPS studies, the powder samples were pressed into
pellets with ca. 1 cm diameter and a few tenth of mm
thickness. Sample treatments were carried out in a high-
pressure cell connected to the analysis chamber via a gate
valve. After the pre-treatment, the samples were cooled
down to room temperature in flowing nitrogen. Then, the
high-pressure cell was evacuated; the sample was trans-
ferred to the analysis chamber in high vacuum (i.e., without
contact to air), where the XP spectra were recorded.
Raman spectra of slightly pressed powder samples were
measured in 180o backscattered geometry at 532 nm laser
excitation (an incident laser power of 10 mW) using a
Thermo Scientific DXR Raman microscope. Scans were
integrated at 4 cm-1 resolution until the desired signal-to-
noise ratio of 1000 or better was achieved (typically 30 s).
UV–vis diffuse reflectance spectroscopy (DRS) was
employed to characterize the optical properties of the
synthetized materials. The diffuse reflectance UV–vis
spectra of the powdered samples were recorded by an
Despite the large number of studies published in this
topic, the accurate nature of the relationship between Co
and noble metals in bimetallic systems is still not fully
understood. In the present paper we show some compara-
tive results of Al2O3 and CeO2 supported Co and Rh–Co
catalysts, and demonstrate how the addition of Rh promoter
changes the oxidation state of Co and its interaction with
the support. As CeO2 is a promising candidate as a pho-
tocatalyst, we carried out some electrochemical measure-
ments to reveal the effects of the Co and Rh on the
electronic structure of this support.
123

The Effect of Rh on the Interaction of Co with Al2O3 and CeO2 Supports
Avantes AvaSpec2048 Fiber Optic Spectrometer equipped
with an Avasphere-50 type integrating sphere. The optical
bandgap of the prepared materials was estimated by
deriving the appropriate Tauc-plots using the following
equation:
transiently dropped, which was also seen as an increase in
the acetaldehyde selectivity. A possible reason is the
acetaldehyde desorption and recombination to ethanol [19].
Figure 1b shows the main products (hydrogen and ethy-
lene). When 0.1 % Rh promoter was added the hydrogen
selectivity markedly increased and ethylene formation
significantly dropped (Fig. 1c). Between 600 and 800 K,
small amounts of methane, CO, CO2, and acetaldehyde
were also detected (not shown). It is important to mention
that in presence of Rh acetone formation was not found. It
is worthy of note that the by increasing the amount of
cobalt to 10 %, the activity and hydrogen selectivity did
not increase significantly [19]. For the CeO2 support
(without cobalt), initially only acetaldehyde was formed,
but between 650–850 K the main product was ethylene
(Fig. 1d) in contrast to the metal containing samples. From
700 to 900 K also other products such as acetone, hydro-
gen, CO2, CO, CH4, ethane and ethylene were formed. The
presented catalytic data indicate that not only the acid–base
character of support but the reducibility of the oxide and
also the nature of interaction between the metals and the
support are important.
À
Á
1
n
ðhmaÞ ¼ A hm À Eg
where h is the Planck-constant, m is the photon frequency, a
is the absorption coefficient, A is a proportionality con-
stant, Eg is the bandgap and n is a constant related to the
nature of the electronic transition. All materials exhibited a
direct transition, where n = 0.5.
X-ray diffractometry (Rigaku MiniFlexII; CuKa) and
electron diffraction technique were used for crystal struc-
ture and crystallinity determination. The ceria crystallites
was determined by Scherrer’s method for the (111)
reflection.
Electrochemical measurements were performed on an
Autolab PGSTAT302 instrument, in a classical one-com-
partment, three-electrode electrochemical cell. The various
semiconductor nanoparticles were spray coated on Glassy
carbon electrodes and were used as working electrodes. A
large Pt foil counter-electrode and an Ag/AgCl/3 M KCl
reference electrode completed the cell setup. The light
source was a 300 W Hg–Xe arc lamp (Hamamatsu L8251).
The radiation source was placed 2 cm away from the
working electrode surface. Photovoltammetry profiles were
recorded in 0.1 M Na2SO3 electrolyte, using a slow
potential sweep (2 mV s-1) in conjunction with interrupted
irradiation (0.1 Hz) on the semiconductor coated elec-
trodes. All the above procedures were performed at the
laboratory ambient temperature (293 2 K). For more
experimental details see for example ref [27].
In order to get closer to the understanding of the effect
of Rh promoter in Co-based alumina and ceria catalysts we
investigate the interaction of Rh and Co with the supports.
The TPR, a dynamic method, gives information about the
reducibility of the catalysts and helps identify the species
formed after reduction in XPS. In Fig. 2a the TPR results
of Al2O3 supported Co–Rh catalysts are depicted. It can be
seen that Rh in 1 % Rh/Al2O3 becomes metallic at 395 K,
but in the case of 2 % Co (curve b) hardly any peak cor-
responding to the reduction of Co oxides could be detected
in the low temperature region. This is in agreement with
the previous observation that at small metal contents
probably mainly Co-aluminate compounds are formed with
low reducibility, while at higher (10 w%) loading the
Co3O4 is dominant [28, 29]. Cobalt-support compounds
can be formed during the catalytic reactions, too [30]. The
only signal at 1187 K was attributed to the reduction of
very small Co particles having strong interaction with
Al2O3 as well as to mixed metal-support oxides [31, 32].
After adding 0.1 % Rh to this sample, this process was
enhanced and occurred at lower temperature. When the Rh
content was increased to 1 %, two additional peaks were
detected: the Rh was reduced at 421 K, and the Co at
569 K.
3 Results and Discussion
It turned out in former studies that small amounts of Rh
additives (0.1–0.5 %) dramatically influenced the product
formation in the CH4?CO2 reaction on Co/alumina [15]
and in the stream reforming of ethanol (SRE) on Co/ceria
[19]; in both cases the selectivity of hydrogen significantly
increased. In the SRE in the medium temperature range of
650–800 K practically only ethylene was detected on 2 %
Co/alumina (Fig. 1a). Above 850 K, ethylene was still the
dominant species, but H2, CO, CH4, acetaldehyde and a
small amount of CO2 were also formed. On 2 % Co/CeO2,
at low conversion up to 500 K acetaldehyde and acetone
were detected. From 500 to 700 K, the acetaldehyde
selectivity attenuated the selectivities of H2, ethylene and
CO2 increased moderately and acetone was also present.
At *800 K, the conversion and the H2 selectivity
At 10 % Co content a broad feature with at least three
components was detected, indicating the multiple mecha-
nism of Co reduction over Al2O3. In general, TPR curves
of Co/Al2O3 samples in this region consists of two com-
ponents: the reduction of Co3O4 takes place first, followed
by the CoO to Co0 process [33]. This behavior is especially
characteristic of the supported Co samples [34]. Taken into
123

E. Varga et al.
Fig. 1 H2 (Diamond) and
ethylene (Circle) selectivity
data as well as EtOH
(a)
100
(b)
100
conversion—during the SRE
reaction over 2 % Co/Al2O3 (a),
2 % Co/CeO2 (b), 0.1 %
Rh ? 2 % Co/CeO2 (c) and
CeO2 (d). Before catalytic test
experiments the catalysts were
reduced with H2 at 773 K for
1 h. 1:3 ratio of ethanol–water
was used
80
60
40
20
0
80
60
40
20
0
400 500 600 700 800 900 1000 1100
Temperature (K)
400 500 600 700 800 900 1000 1100
Temperature (K)
(c)
(d)
100
100
80
60
40
20
0
80
60
40
20
0
400 500 600 700 800 900 1000 1100
Temperature (K)
400 500 600 700 800 900 1000 1100
Temperature (K)
(a)
(b)
consideration the former results, we ascribe these features
to the peaks located at 680 and 785 K, respectively.
Components at lower temperatures can be the contributions
of Co3O4 particles having only weak interaction with the
support [35]. When Rh was added, first an intensive peak
emerged at 625 K next to the main peak at 785 K, then at
1 % Rh the maximum moved to 573 K, and only a
shoulder remained at 647 K. The feature at 421 K origi-
nated from Rh.
540
587
573
647
454
1065
421
421
584
1120
g
f
g
f
413
643
567
625
785
680
1087
1117
1080
1146
e
554
587
569
461
e
d
c
b
a
d
c
b
417
531
1095
1191
From these results it can be concluded that by increasing
the noble metal content, a systematic shift to lower tem-
peratures was observed in case of the peak located above
1000 K. In the lower temperature region shifts and peak
structure changes also took place. Fundamentally we
attribute this enhanced reducibility to the H spill over
phenomenon [23, 36]. However, on the basis of the XPS
spectra, we suggest that Rh somehow weakens the inter-
action of Co with the support. Previously Ru and Pt have
also been found to inhibit the formation of hardly reducible
Co-Al2O3 compounds [37, 38], and Rh was shown to
hinder the dissolution of Co into the CeO2 support [39].
That is why the reduction of the CoO-Al2O3 compounds
occurs sooner, and there are more available support-inde-
pendent Co3O4. The forms of the oxides also differ at low
and high metal contents.
617
550
391
1080
395
a
400 600 800 1000 1200 1400
Temperature (K)
400 600 800 1000 1200 1400
Temperature (K)
Fig. 2 TPR results of Al2O3 (a) and CeO2 supported (b) Co–Rh
catalysts: 1 % Rh (a), 2 % Co (b), 0.1 % Rh?2 % Co (c), 1 %
Rh?2 % Co (d), 10 % Co (e) 0.1 % Rh?10 % Co (f) and 1 %
Rh?10 % Co (g)
attributed to the bulk reduction of CeO2 but the surface
reduction at around 750–780 K is not well resolved on
these samples [19], probably because the reducing effect of
the metals could take place even during the preparation,
before the reduction. Upon adding 2 or 10 % of Co, a well-
resolved doublet was detected corresponding to the step-
wise mechanism of Co3O4 reduction, in the same
On CeO2 the Rh interacts with H2 approximately at the
same temperature as on Al2O3. The feature at 1080 K is
123

The Effect of Rh on the Interaction of Co with Al2O3 and CeO2 Supports
temperature range as on Al2O3. When the 2 % Co sample
was promoted with 0.1 % Rh, the TPR peak of the Rh
shifted to higher values. We suppose that it has some
contribution from the Co, as only an additional minor peak
could be seen at 531 K. In the case of 1 % Rh?2 % Co, the
first peak moved to higher temperature and both peaks
gained intensity.
core–shell like structure. Rhodium at 0.1 % amount can be
well dispersed over the whole surface, even in the Co-
support interface, weakening their interaction.
When 10 % Co was supported on CeO2, after oxidation
Co3O4 was detected at 779.7 eV regardless of the Rh
addition. When reduced, the promotion of Rh was evident
from the higher intensity of the metallic peak (Co0) at
777.8 eV (Fig. 3b). The two-step reduction of Co3O4 is
indicated by the existence of the satellite on 10 % Co, and
the higher CoOx binding energy on 0.1 % Rh ? 10 % Co,
By increasing the Co content, the reduction peaks
moved to somewhat higher temperatures. In case of Co/
Al2O3 lower metal loading results in stronger interaction
with the support, but with CeO2 there is a different effect.
In our former paper it was demonstrated that Co is well
dispersed and upon reduction it dissolves into the support
and disrupts it [39]. We also found that this phenomenon is
enhanced at higher Co content. It means that CeO2 will
have larger surface (higher dispersity) to interact with Co at
higher metal loadings.
that are characteristic of Co2?
.
For the summarizing of the XPS data the surface
metallic content with respect to the whole metallic content
plus support molar content (M/(support ?M) were calcu-
lated. We also determined the ratio of metallic cobalt to all
cobalt content (Co0/Coall). The data are collected in the
Table 1. On alumina support the of Co0/Coall ratio
increased from 11.3 to 25 % when the Co content was
changed from 2 to 10 %. When 0.1 % Rh was added to
10 % Co this value increased slightly further to 28.6 %.
Significantly higher promotional effect was observed on
ceria support. In the case of 0.1 % Rh?10 % Co/CeO2, the
Co0/Coall is 53.1 % comparing to the 10 % Co/CeO2 where
this value is only 33.1 %. The other important observation
is that the Rh enriches on the surface. While the Rh/(sup-
port ?Mall) is 0.5 % on 0.1 % Rh/CeO2, this value is
significantly higher, 3.8 % on 0.1 % Rh?10 % Co/CeO2.
Such kind of enrichment was not calculated for Co (see
Table 1). We may conclude that the increased number of
Rh atoms and the presence of metallic Co species are
responsible for the higher catalytic activity. It is important
to emphasize that the formation of metallic sites are
When Rh was added to 10 % Co, its reduction took
place at higher temperature (413 and 454 K) and the Co
peak shifted to lower values (584 K at 0.1 % Rh, the two
components are not resolved). On the 1 % Rh?10 % Co
sample there is a significant intensity increase due to the
enhanced reduction of Co3O4.
In general, the catalytic reaction take place at elevated
temperatures. It has been shown that heating even in inert
atmosphere induces oxidation state changes due to the
enhanced oxygen mobility [39]. Thus, characterization
should be done after higher temperature treatments to
understand better the real catalysts.
The cobalt oxidation states were determined by XPS
after reduction at 773 K. The Rh XPS was also monitored.
The Rh enhanced reducibility was well detected. On 10 %
Co supported over Al2O3 (Fig. 3a) the Co 2p3/2 component
appeared at 781.3 eV with a satellite at 786.7 eV. The
separation of the two 2p components was 15.7 eV. These
features indicate the presence of Co2? compounds. When
reduced, a new peak at 778.6 eV emerged due to the
metallic Co. After the addition of 0.1 % Rh, the peak of the
oxidized sample was located 0.8 eV lower than on the
unpromoted one. In addition, the shake-up satellite signif-
icantly lost intensity and also shifted to lower values. The
1/2-3/2 peak separation was 15.2 eV, which is more
characteristic of diamagnetic Co3? complexes [40, 41]. All
these observations prove that after oxidation 10 % of Co is
in Co2? state on Al2O3, but in the presence of Rh it forms
preferably Co3O4. The formation of Co2RhO4 in the reac-
tion of Co3O4 and Rh2O3 cannot be excluded either [42].
However, its reduction after that is also promoted, as the
metallic peak was more intensive. When reduced, the peak
assigned to the oxidized state was also found Co3? like,
which would indicate different surface reduction mecha-
nisms with and without Rh. As the peak intensities did not
change after Rh addition we exclude the possibility of
(a)
(b)
781.3
778.6
777.8
779.7
782.1
780.6
d
d
c
c
b
b
a
a
795 790 785 780 775 770 765
Binding energy (eV)
795 790 785 780 775 770 765
Binding energy (eV)
Fig. 3 XP spectra of Al2O3 (a) and CeO2 (b) supported catalysts
after oxidation at 673 K for 30 min and after reduction at 773 K for
1 h. 10 % Co oxidized (a) and reduced (b), 0.1 % Rh?10 % Co
oxidized (c) and reduced (d)
123

E. Varga et al.
necessary for C–C bond rupture in SRE on ceria supported
catalysts which finally leads to hydrogen production
[19, 23, 42].
defect concentration, although Rh and Co interacts differ-
ently with CeO2.
The value of the bandgap is also a very important
property of the catalytic systems. It has been presented that
the addition of doping agents or the formation of Ce3?
centers decreases the bandgap [44]. A decreased bandgap
in partially reduced CeO2 and its enhanced photocatalytic
activity has already been published, too [45, 46]. In our
samples untreated CeO2 had intensive absorption at
3.71 eV (Fig. 5). When 1 % Rh, 2 % Co or 1 % Rh?2 %
Co were added, additional absorptions appeared at lower
energies attributed to both the plasmonic effect of the metal
nanoparticles, and the eventually formed midgap states of
CeO2. Because the local maxima were located at the same
positions in all samples, we suppose that this feature is
primarily caused by the defect structure of CeO2. In line
with that XPS confirmed higher Ce3? ratio in the presence
of the metals; for pure CeO2 it was 10 %, and in the 1 %
Rh?2 % Co/CeO2 it was 15 % (Fig. 5b), thus the addition
of the metals alters significantly the CeO2 surface structure
and its catalytic properties (Fig. 1). We suppose that for the
observation of Rh plasmonic feature our metal particles are
too small [47, 48]. The incorporation of metal ions would
suppress the Ce3? formation by replacing it and the cor-
responding absorptions decrease [49, 50]. In our case this
can be ruled out as an opposite trend of the intensities was
observed. As we postulated in our previous work [39], after
the reduction of Co/CeO2 the majority of Co remains in
ionic state. Part of the absorption feature at around
550–600 nm may contain some Co2? contribution
(Fig. 5a). Recently it was found that the attachment of
Co2? onto TiO2 nanorod produced an absorption maximum
at *580 nm in the UV–Vis absorbance spectrum [51].
The direct bandgaps of CeO2, 1 % Rh/CeO2 and 2 %
Co/CeO2 varied around 3.27 eV. The high non-character-
istic absorption made it impossible to calculate the bandgap
of 1 % Rh ? 2 % Co/CeO2.
The catalytic results show that the character of the
support has a great influence on the SRE catalysis; using
acidic support like alumina, the ethylene formation is
favorable, while on basic ceria supported catalysts, other
routes, including hydrogen formation are allowanced
[6, 19]. On the other hand the metals (in the present case
Rh and Co) have influence on the reducibility of the sup-
port (in the present case on ceria). XPS results (Table 2)
show that the Rh and Co (and Rh?Co) interact with ceria
and increase its reducibility. XRD experiments pointed out
that Co induces the disruption of ceria, its particle size
decreased dramatically (Table 2). We suppose that the
change of particle size of ceria also has an effect for cat-
alytic activity of Co-based ceria catalysts.
We believe that beside the increased number of atoms in
metallic state in the surface layer, the defect concentration
and changes in Fermi level region may also contribute to
the observed catalytic effects. In order to get more infor-
mation about this phenomenon, some Raman measure-
ments were carried out on 1 % Rh and 2 % Co samples
reduced at different temperatures. In Fig. 4 the reference
spectrum of CeO2 can be seen, with a narrow peak at
464 cm-1 corresponding to the triply degenerate F2 g
mode of CeO2 with fluorite structure. After the addition of
either Co or Rh this feature significantly broadened and
shifted to 439 cm-1. In addition, new peaks and shoulders
appeared at 225 (2TA mode), 602 (defect induced, D
mode) and 1175 cm-1 (2LO mode) that we attribute defi-
nitely to the effect of metals. It could indicate metal ion
incorporation, O vacancy formation and particle size
decrease [43]. Our spectra show very similar structure in
every case, suggesting that the reduction temperature and
the type of the metal defect have minor effect on the final
Table 1 Metallic contents and the Co0/Coall ratio on the different
catalysts calculated from the XP spectra
Table 2 Ceria reducibility (Ce3?/Ceall) and CeO2 particle size on
different catalysts
Samples
M/(support?M)
Co0/Coall
2 % Co/CeO2
3.3
25.8
33.1
–
Samples
Ce3?/Ceall
CeO2 particle size (nm)
10 % Co/CeO2
14.0
CeO2
11
10
14
12
15
16
14
17
15
27.6
17.1
10.7
27.6
26.7
26.5
17.0
11.1
24.1
0.1 % Rh/CeO2
1 % Rh/CeO2
0.5
2 % Co/CeO2
2.2
–
10 % Co/CeO2
0.1 % Rh?2 % Co/CeO2
5.4 (Co)
2.0 (Rh)
13.4 (Co)
3.8 (Rh)
1.9
29.1
0.1 % Rh/CeO2
1 % Rh/CeO2
0.1 % Rh?10 % Co/CeO2
53.1
5 % Rh/CeO2
0.1 % Rh?2 % Co/CeO2
0.1 % Rh?10 % Co/CeO2
1 % Rh?2 % Co/CeO2
2 % Co/Al2O3
11.3
25.0
28.6
10 % Co/Al2O3
2.5
0.1 % Rh?10 % Co/Al2O3
1.7 (Co)
123

The Effect of Rh on the Interaction of Co with Al2O3 and CeO2 Supports
439
225
d
c
602
1175
f
e
d
c
b
a
b
464
a
-0.6
-0.4
-0.2
0.0
0.2
0.4
0
400
800
1200
1600
Raman shift (cm-1)
E (V) vs. Ag/AgCl/3M NaCl
Fig. 6 Photovoltammetry profiles of CeO2 (a), 1 % Rh/CeO2
reduced at 423 K (b), 2 % Co/CeO2 reduced at 773 K (c), and 1 %
Rh?2 % Co/CeO2 reduced at 773 K (d) recorded in 0.1 M Na2SO3
solution at a sweep rate of 2 mV.s-1
Fig. 4 Raman spectra of CeO2 (a), 2 % Co/CeO2 reduced at 423 K
(b) and 773 (c), 1 % Rh reduced at 423 K (d) and 773 K (e), 1 %
Rh?2 % Co reduced at 423 K (f)
(a)
(b)
u'"
1.97 eV
2.33 eV
the surface defects over CeO2 depends on the metal
quantity, rather than on the type of the metal.
2.63 eV
v'"
u"
1.80 eVd
3.71 eV
Figure 6 clearly shows that all samples behave as n-type
semiconductors with the photocurrent signals anodic in
polarity. The photocurrent density of CeO2 slightly
decreased when the metal nanoparticles were added. It
means that the separation of the photoexcited electron–hole
pairs worsened, especially in the presence of Co [53]. More
importantly, the onset potential of the photovoltammogram
can be used to estimate the Fermi level of the n-CeO2
[27].Thus the position of the conduction band edge was
estimated for all the samples. The fact that there is a
considerable shift in the Fermi level in the case of the metal
containing samples (50, 150, 80 mV for Co, Rh, and Co/
Rh, respectively) indicate an intimate contact between the
CeO2 and the metal particles allowing a certain extent of
Fermi-level equilibration [54].
v
u"
u'
c
b
a
v"
u0
v'
v0
b
a
920 910 900 890 880
Binding energy (eV)
300 400 500 600 700 800
Wavelength (nm)
Fig. 5 UV–vis absorption spectra (a) of CeO2 (a), 1 % Rh/CeO2
reduced at 423 K (b) 2 % Co/CeO2 reduced at 773 K (c) and 1 %
Rh?2 % Co/CeO2 reduced at 773 K (d). Ce 3d spectra (b) of CeO2
and 1 % Rh?2 % Co/CeO2 after reduction at 773 K
4 Summary
In our previous work [39, 52] we demonstrated that after
the reduction of ceria supported Rh and Co samples,
enhanced surface reduction of the support was observed. In
accordance with this fact Fig. 5b shows that Co and Rh
together could increase considerably the number of Ce3?
species on the surface. In this process the surface diffusion
may play an important role [38]. Taking into account the
Raman and optical results, we propose that the density of
Al2O3 and CeO2 supported Co catalyst were characterized
with and without Rh promoter. TPR and XPS results
revealed that Rh fundamentally changed the reduction
mechanism of Co: not only the reduction took place at
lower temperatures, but the intermediate Co forms were
different. Moreover, Rh inhibits the formation of strong
Co-support interaction. Using ceria support we may
123

E. Varga et al.
´
}
23. Varga E, Ferencz Z, Oszko A, Erdohelyi A, Kiss J (2015) J Mol
conclude that the Rh enrichment in the surface layer and
the presence of metallic Co are responsible for the higher
catalytic conversion and for the enhanced hydrogen
selectivity in the SRE reaction. Over the n-type semicon-
ductor CeO2 metal induced defects were evidenced by
Raman spectroscopy and DRS. Co and Rh did not change
the bandgap, but metal-support electronic interaction was
observed by photovoltammetry. Comparing the catalytic
results with the catalyst characterization results, we sup-
pose that in addition to the presence of the metals, the
metal induced changes on the support surface, like
increased O-vacancy density and Fermi-level equilibration
also influence the catalytic process.
Catal A 397:127–133
24. Profeti LPR, Ticianelli EA, Assaf EM (2008) J Power Sources
175:485–489
25. Espinal R, Taboada E, Molins E, Chimentao RJ, Medina F,
Llorca J (2013) Top Catal 56:1660–1671
26. Pereira EB, Piscina PR, Marti S, Homs N (2010) Energy Environ
Sci 3:486–492
´
27. Kormanyos A, Thomas A, Huda MN, Sarker P, Liu JP, Poudyal
´
N, Janaky C, Rajechwar K (2016) J Phys Chem C. doi:10.1021/
acs.jpcc.1025b12738
28. Chin RL, Hercules DM (1982) J Phys Chem 86:360–367
29. Liotta LF, Carlo GD, Pantaleo G, Venezia AM, Deganello G
(2006) Appl Catal B 66:217–227
30. Jacobs G, Ma W, Davis BH (2014) Catalysts 4:49–76
31. Ali S, Zabidi NAM, Subbarao D (2011) Chem Centr J 5:68
32. Das TK, Jacobs G, Patterson PM, Conner WA, Li J, Davis BH
(2003) Fuel 82:805–815
´
Acknowledgments The authors acknowledge Dr. Csaba Janaky for
the discussion about the photoelectrochemical results and Levente
´ ´
Koppany Juhasz for the Raman spectroscopic measurements.
33. Ewbank JL, Kovarik L, Kenvina C, Sievers C (2014) Green
Chem 16:885–896
34. Bulavchenko OA, Cherepanova SV, Malakhov VV, Dovlitova
LS, Ishchenko AV, Tsybulya SV (2009) Kinet Catal 50:192–198
35. Simionato M, Assaf EM (2003) Mater Res 6:535–539
36. Cook KM, Poudyal S, Miller JT, Bartholomew C, Hecker WC
(2012) Appl Catal A 449:69–80
References
37. Kogelbauer A, Goodwin JJG, Oukaci
160:125–133
38. Guczi L, Hoffer T, Zsoldos Z, Zyade S, Maire G, Garin F (1991)
R (1996) J Catal
1. Breyssea M, Afanasiev P, Geantet C, Vrinat M (2003) Catal
Today 86:5–16
2. Saeidi S, Amin NAS, Rahimpour MR (2014) J CO2 Util 5: 66–81
3. Enger BC, Lodeng R, Holmen A (2010) Catal Lett 134:13–23
J Phys Chem 95:802–808
´
´
39. Varga E, Pusztai P, Ovari L, Oszko A, Erdohelyi A, Papp C,
Steinru¨ck H-P, Konya Z, Kiss J (2015) Phys Chem Chem Phys
´
}
´
4. Pusztai P, Puskas R, Varga E et al (2014) Phys Chem Chem Phys
16:26786–26797
´
17:25166–27157
40. Frost CD, McDowell CA, Woolsey IS (1972) Chem Phys Lett
17:320–323
5. Mullins DR (2015) Surf Sci Rep 70:42–85
6. Mattos LV, Jacobs G, Davis BH, Noronha FB (2012) Chem Rev
112:4094–4123
41. Schmid M, Kaftan A, Steinru¨ck HP, Gottfried JM (2012) Surf Sci
606:945–949
42. Cascales C, Rasines I (1984) Mater Chem Phys 10:199–203
43. Daniel M, Loriant S (2012) J Raman Spectrosc 43:1312–1319
44. Khan MM, Ansari SA, Pradhan D, Han DH, Lee J, Cho MH
(2014) Ind Eng Chem Res 53:9754–9763
45. Aslam M, Oamar MT, Soomro MT, Ismail IMI, Salah N,
Almeelbi T, Gonda A, Hameed A (2016) Appl Catal B
180:391–402
46. Verma R, Samdarshi SK, Bojja S, Paul S, Choudhury B (2015)
Sol Energ Mater Sol Cells 141:414–422
7. Liu W, Flytzani-Stephanopoulos M (1995) J Catal 153:317–332
8. Rodriguez JA, Liu P, Hrbek J, Evans J, Manuel P (2007) Angew
Chem Int Ed 46:1329–1332
9. Mullins DR, Albrecht PM, Chen T-L et al (2012) J Phys Chem C
116:19419–19428
10. Guczi L, Boskovic G, Kiss E (2010) Chem Rev 52:133–203
11. Zhang Y, Chen L, Bai G, Li Y, Yan X (2005) J Catal
236:176–180
12. Riedel T, Claeys M, Schulz H et al (1999) Appl Catal A
186:201–213
13. Melaet G, Lindeman AE, Somorjai GA (2014) Top Catal
57:500–507
´
47. Watson AM, Zhang X, de la Osa RA, Sanz JM, Gonzalez F,
Moreno F, Finkelstein G, Liu J, Everitt HO (2015) Nano Lett
15:1095–1100
14. Chu W, Chernavskii PA, Gengembre L, Parkina GA, Fongarland
P, Khodakov AY (2007) J Catal 252:215–230
´ ´
48. Zhang X, Li P, Barreda A, Gutierrez Y, Gonzalez F, Moreno F,
Everitt HOJ (2016) Nanoscale Horiz 1:75–80
´
´
´
´
15. Ferencz Z, Baan K, Oszko A, Konya Z, Kecskes T, Erdohelyi A
}
(2014) Catal Today 228:123–130
16. Ruckenstein E, Wang HY (2002) J Catal 205:289–293
17. Badlani M, Wachs IE (2001) Catal Lett 75:137–149
49. Balestrieri M, Colis S, Gallart M, Schember G, Ziegler M, Gilliot
P, Dinia A (2015) J Mater Chem C 3:7014–7021
50. Wang Z, Quan Z, Lin J (2007) Inorg Chem 46:5237–5242
51. Balasanthiran C, Hoefelmeyer JD (2014) Chem Commun
50:5721–5724
´
´
´
}
18. Toth M, Varga E, Oszko A, Baan K, Kiss J, Erdohelyi A (2016)
Appl Catal A 411:377–387
´
}
´
´
´
´
19. Ferencz Z, Erdohelyi A, Baan K, Oszko A, Ovari L, Konya Z,
Papp C, Steinru¨ck H-P, Kiss J (2014) ACS Catal 4:1205–1218
20. Martono E, Vohs JM (2012) J Catal 291:79–86
21. Batista MS, Santos RKS, Assaf EM, Assaf JM, Ticinalli EA
(2003) J Power Sources 124:99–103
´
´
}
´
52. Varga E, Pusztai P, Oszko A, Baan K, Erdohelyi A, Konya Z,
Kiss J (2016) Langmuir 32:2761–2770
53. Li B, Gu T, Wang J, Wang P, Wang J, Yu JC (2014) ACS Nano
8:8152–8162
54. Subramanian V, Wolf EE, Kamat VP (2004) J Am Chem Soc
126:4943–4950
22. Lin SSY, Kim DH, Engelhard MH, Ha SY (2010) J Catal
273:229–235
123