Unraveling the Chemical State of Cobalt in Co-Based Catalysts during Ethanol Steam Reforming: An in Situ Study by Near Ambient Pressure XPS and XANES

Article abstract of DOI:10.1021/acscatal.8b02666

The steam reforming of ethanol (ESR) has been studied by near ambient pressure XPS (NAP-XPS), extended X-ray absorption fine structure (EXAFS), and X-ray absorption near edge structure (XANES) under in situ conditions in the ALBA synchrotron facility at 200-580 °C and S/C = 3 over different cobalt-based catalysts that showed different catalytic performances: Co₃[Si₂O₅]₂(OH)₂ (Co-talc), [Co₂Mg₄Al₂(OH)₁₆]CO₃·4H₂O (Co-hydrotalcite shortened to Co-HT) calcined at 550 °C, and Co₃O₄ (Co-spinel). Both Co-spinel and Co-talc yield to a greater or lesser degree metallic cobalt under ESR conditions. While Co-spinel shows a complete reduction to metallic cobalt under the conditions used for the XANES measurements, the more bulk-sensitive Co-talc sample exhibits only a partial reduction. On the other hand, under the ESR conditions used with the NAP-XPS, a more surface sensitive technique, the results indicate a higher reduction degree for the Co-talc sample in comparison to the Co-spinel sample. In contrast, the catalyst prepared from the Co-HT does not show metallic cobalt traces under the experimental conditions used with both techniques. On comparison of these three cobalt-based catalysts, the stable operation exhibited by Co-HT under ESR reaction conditions is justified by the absence of metallic cobalt formation under in situ conditions, which is identified as being responsible for the carbon deposition phenomenon that triggers the deactivation suffered by most cobalt-based catalysts during ESR.

Full text of DOI:10.1021/acscatal.8b02666

Subscriber access provided by University of South Dakota
Article
Unraveling the chemical state of Cobalt in Co-based catalysts during Ethanol
Steam Reforming: an in situ study by Near Ambient Pressure XPS and XANES
Cristian Huck-Iriart, Lluís Soler, Albert Casanovas, Carlo Marini, Jordi Prat, Jordi Llorca, and Carlos Escudero
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.8b02666 • Publication Date (Web): 06 Sep 2018
Downloaded from http://pubs.acs.org on September 7, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
Unraveling the chemical state of Cobalt in Co-
based catalysts during Ethanol Steam Reforming:
an in situ study by Near Ambient Pressure XPS
and XANES
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Cristián Huck-Iriart1,2, Lluís Soler3, Albert Casanovas3, Carlo Marini1, Jordi Prat1, Jordi
Llorca3,* Carlos Escudero1,*
1ALBA Synchrotron Light Source, Carrer de la Llum 2–26, 08290 Cerdanyola del Vallès,
Barcelona (Spain)
2Escuela de Ciencia y Tecnología, Universidad Nacional de San Martín (UNSAM),
Campus Miguelete, 25 de Mayo y Francia, 1650 San Martín, Provincia de Buenos Aires
(Argentina)
3Institute of Energy Technologies, Department of Chemical Engineering and Barcelona
Research Center in Multiscale Science and Engineering, Universitat Politècnica de
Catalunya, EEBE, Eduard Maristany 10-14, 08019 Barcelona, (Spain)
*Corresponding authors. E-mails: jordi.llorca@upc.edu, cescudero@cells.es
1
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 44
1
2
3
4
Keywords: cobalt talc, cobalt hydrotalcite, cobalt spinel, ethanol steam reforming, in situ
5
6
spectroscopy, NAP-XPS, XANES
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ABSTRACT
The steam reforming of ethanol (ESR) has been studied by Near Ambient Pressure XPS
(NAP-XPS), Extended X-ray Absorption Fine Structure (EXAFS) and X-ray Absorption
Near Edge Structure (XANES) under in situ conditions in the ALBA synchrotron facility at
200-580ºC and S/C=3 over different cobalt-based catalysts that showed different catalytic
performance: Co3[Si2O5]2(OH)2 (Co-talc), [Co2Mg4Al2(OH)16]CO3·4H2O (Co-hydrotalcite
shortened as Co-HT) calcined at 550ºC, and Co3O4 (Co-spinel). Both, Co-spinel and Co-
talc yield to a greater or lesser degree metallic cobalt under ESR conditions. While the Co-
spinel shows a complete reduction to metallic cobalt under the conditions used for the
XANES measurements, more bulk-sensitive, the Co-talc sample exhibits only a partial
reduction. On the other hand under the ESR conditions used with the NAP-XPS, a more
surface sensitive technique, the results indicate a higher reduction degree for the Co-talc
sample as compared to the Co-spinel. In contrast, the catalyst prepared from the Co-HT
does not show metallic cobalt traces under the experimental conditions used with both
techniques. Comparing these three cobalt-based catalysts, the stable operation exhibited by
Co-HT under ESR reaction conditions is justified by the absence of metallic cobalt
formation under in situ conditions, which is identified as responsible for the carbon
deposition phenomena that triggers the deactivation suffered by most cobalt-based catalysts
during ESR.
2
ACS Paragon Plus Environment

Page 3 of 44
ACS Catalysis
1
2
3
4
1. INTRODUCTION
5
6
7
8
The steam reforming of ethanol (ESR) is a well-known process used for generating
hydrogen (eq. 1). It is particularly interesting for the on-site and on-board production of
hydrogen due to the widespread use of ethanol and its safe transport and storage1,2. The
steam reforming of renewable bioethanol (ethanol produced from biomass by fermentation)
can be considered CO2 neutral because the CO2 yielded in the reaction is compensated by
the CO2 fixed by the biomass during its growth. Therefore, ESR is emerging as a cleaner
alternative to generate hydrogen compared to the fossil fuels currently used. Recently, the
use of microreaction technologies and catalytic membrane reactors have allowed strong
process intensification, which in turn has increased the interest of ESR for feeding fuel cells
in transportation and electronic applications3–6.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
C2H5OH + 3 H2O D 6 H2 + 2 CO2
(1)
The reaction network associated to the ESR is complex, both from the mechanistic and
the thermodynamic points of view7. The intermediate steps involved include a variety of
reaction types (dehydration, dehydrogenation, reforming, WGS, methanation, etc.). They
show different dependency relationships on reaction conditions and their promotion
depends on different functionalities of the catalyst. In addition, formation of coke
precursors from several of the compounds involved makes deactivation an important issue
for the practical application of this process, which can be partly solved by the addition of
alkaline promoters8–10. All these factors explain the need for extensive research of
appropriate catalysts. Many studies have been focused on the ESR using supported nickel,
cobalt and noble metal catalysts11–15. An efficient catalyst for hydrogen production from
ethanol has to dissociate the C-C bond at reasonably low temperatures, it has to maintain
low the CO concentration and be stable under catalytic operation.
3
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 44
1
2
3
4
5
6
7
8
9
Noble metal-based catalysts perform well for ESR; they are stable and exhibit high
activity16–18. However, they are expensive and usually need high temperatures to be active.
The main reaction mechanism involves the decomposition of ethanol into a mixture of
hydrogen, carbon monoxide and methane (eq. 2), followed by the steam reforming of the
produced methane (eq. 3)19. Additionally, the water gas shift (WGS) reaction balances CO
and CO2 rendering extra hydrogen amounts (eq. 4). Nickel catalysts are inexpensive and
follow the same reaction scheme, but they normally suffer from sintering and deactivation
by carbon deposition1. In contrast, cobalt-based catalysts can operate at much lower
temperature levels when compared with noble metal- and nickel-based catalysts, typically
at 400-550ºC, since they do not yield methane as an intermediate species in the reaction
mechanism, which can only be reformed at high temperature20–27. Over cobalt-based
catalysts, ethanol is first dehydrogenated into a mixture of hydrogen and acetaldehyde (eq.
5), and then acetaldehyde reacts with steam to yield hydrogen and carbon monoxide (eq. 6),
which participate in the WGS (eq. 4). Dimethyl ketone can be also formed via condensation
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
of acetaldehyde. A complete network of reactions can be found in literature7,17
.
C2H5OH D H2 + CO + CH4
CH4 + H2O D 3 H2 + CO
CO + H2O D H2 + CO2
(2)
(3)
(4)
(5)
(6)
C2H5OH D H2 + CH3CHO
CH3CHO + H2O D 3 H2 + 2 CO
An important advantage of conducting the ESR at lower temperature is that the WGS
equilibrium favors the formation of hydrogen and CO2 at the expense of CO and water (eq.
4), thus maximizing the production of H2 and reducing the volume of the WGS units
normally implemented downstream the reformer. However, most cobalt catalysts also
4
ACS Paragon Plus Environment

Page 5 of 44
ACS Catalysis
1
2
3
4
suffer from severe deactivation during ESR due to extensive carbon deposition11–13,28–30
,
5
6
7
8
particularly under realistic loads of ethanol. The coke formation on the catalyst surface has
been related to the presence of metallic Co sites that would act as nucleation points for its
growth but there still is controversy among different reports. Essential for this matter is the
understanding of the role of the cobalt oxidation state.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In a pioneering study by in situ magnetic measurements, it was first demonstrated that
metallic cobalt is a very active species for the reforming of ethanol31. A Co/ZnO catalyst
showed 92% of reduced cobalt under ESR conditions, mainly as small superparamagnetic
nanoparticles. However, on the same catalyst, in situ FTIR with CO as a probe molecule
revealed that, in addition to metallic cobalt, oxidized Co species were also present at the
surface, thus suggesting that the redox pair Co0DCoδ+ is responsible for the activity of
cobalt in ESR32. Since then, the exact role of Co0 and Coδ+ during ESR has been
investigated over several cobalt-based catalysts with a variety of in situ and operando
techniques with great success, and it has been recognized that the oxidation state of Co
depends on many different factors like the catalyst structure, particle size, support material
and composition of the gas phase in the reaction environment, among others33–42. In
particular, considerable efforts have been addressed towards the Co/CeO2 system due to the
large oxygen storage capacity and high oxygen mobility exhibited by ceria that prevent
extensive carbon deposition during ESR. It has been reported that proper metal-support
interaction allows only partial reduction of cobalt and leads to a superior catalytic
performance for the production of hydrogen. Specifically, Coδ+ with O adatoms supplied by
a reducible support facilitates the oxidative dehydrogenation of ethoxide to acetaldehyde,
while the reduced Co0 subsequently decomposes the acetaldehyde to reforming products43.
5
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 44
1
2
3
4
5
6
7
8
9
However, inconsistent results regarding the role of the oxidized cobalt species have been
presented for non-reducible supports. Over Al2O3 a balanced Co0:Coδ+ ratio of about 3:1
has been reported to be optimum for stable ESR performance39, but over MgO it has been
suggested that Coδ+ should be minimized to avoid methanation activity44. To conclude,
results have highlighted the importance of the oxidation state of cobalt in determining the
activity, selectivity and stability of cobalt-based catalysts for ESR, but no conclusion has
been made regarding the different inorganic oxide supports and their interactions with
cobalt, which calls for more investigations upon the active sites in working cobalt catalysts
under ESR conditions. Therefore, the study of the cobalt oxidation state of Co-based
catalysts under in situ and operando conditions is crucial to evaluate the catalyst behavior
under more realistic ESR conditions and can offer new insights regarding its role in the
activation/deactivation phenomena.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The aim of this work is the study of Co-spinel, Co-hydrotalcite (shortened Co-HT from
now on) and Co-talc in bulk and at the surface employing in situ X-ray Absorption Near
Edge Structure (XANES), in situ Extended X-ray absorption fine structure (EXAFS) and
Near Ambient Pressure X-ray Photoemission Spectroscopy (NAP-XPS), in order to
understand the behavior of these catalysts under ESR conditions. Co-spinel (Co3O4) has
been selected as a standard material whereas Co-talc (Co3[Si2O5]2(OH)2) and Co-HT
([Co2Mg4Al2(OH)16]CO3·4H2O) have been chosen due to their radically different catalytic
performance in ESR. Over Co-talc and Co-spinel, metallic Co nanoparticles are formed
easily under ESR conditions at 400ºC, which rapidly detach from the catalyst support and
originate carbon nanotubes, nanofibers and platelets. The carbon deposition rate measured
at S/C=3 is ca. 30 mgcarbon gcatalyst–1 h–1 for both, Co-spinel and Co-talc45. At the same time,
6
ACS Paragon Plus Environment

Page 7 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
selectivity to methane and higher hydrocarbons increases at the expense of the reforming
products (H2 and COx)46,47. In contrast, the catalyst derived from Co-HT shows an excellent
performance for ESR at 550ºC (H2 yield of 92%) and no metallic cobalt is apparently
formed48. In this case, the carbon deposition rate measured at S/C=2 is only about 1 mgcarbon
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
–1
gcatalyst h–1 48.This interesting result allows designing cobalt-based catalysts for ESR
without coke deposition by placing in appropriate environments Coδ+ active species. Long-
term catalytic tests (300 h) under high loads of ethanol and commercial bioethanol have
shown stable operation over hydrotalcite-derived cobalt catalysts without carbon deposition
49,50
.
2. MATERIALS AND METHODS
2.1. Catalyst preparation
Co-spinel, Co3O4, was prepared by the citrate method from Co(NO3)2·6H2O and citric
acid as it has been previously described51. Co-talc, Co3[Si2O5]2(OH)2, was obtained by
following
the
procedure
described
in
the
literature47.
Co-HT,
[Co2Mg4Al2(OH)16]CO3·4H2O, was prepared by co-precipitation at constant pH (10 0.2)
by adding an aqueous solution of NaOH/Na2CO3 (2 M) onto an aqueous solution of
CoCl2·6H2O, Mg(NO3)2·6H2O and Al(NO3)3·9H2O. The slurry was aged at 25ºC for 15 h
under vigorous stirring and the resulting precipitate was thoroughly washed to eliminate
any Na residue and calcined at 823 for 3 h49. The resulting material was constituted by a
mixture of CoAl spinel and CoO strongly interacting with MgO, as reported in literature48.
2.2. Characterization techniques
7
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 44
1
2
3
4
2.2.1. X-ray absorption near edge structure (XANES)
5
6
7
8
XANES measurements at Co K edge (7709 eV) were performed in quick-EXAFS
transmission mode at the CLAESS beamline52 of the ALBA Synchrotron Light Source. The
synchrotron radiation emitted by a multi-pole wiggler was first collimated vertically by a
first mirror and than monochromatised with a liquid nitrogen cooled Si (111) double
crystal. The monochromatised X-ray beam was then focused down to the sample at about
800 µm x 500 µm (H x V) by changing the radius of curvature of a vertically focusing
mirror. The higher harmonics contribution to desired energies was minimized by
appropriately setting the rejection angles of the two mirrors in the optical hutch. The
incoming and outgoing photon fluxes were measured by ionization chambers filled with an
appropriate mixture of N2 and Kr gases. Beamline energy was calibrated on the edge
position of a Cobalt foil. The ethanol and water mixture for the ESR was prepared in a
bubbler using He as a carrier gas in order to have an ethanol:water ratio of 1:6 in the reactor
chamber (S/C=3). Reactants were introduced into the reactor chamber at atmospheric
pressure and controlling the flow (20 mL/min).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.2.2. Extended X-ray absorption fine structure (EXAFS)
The EXAFS data at the Co K edge were collected at CLAESS beamline52 of the ALBA
Synchrotron Light Source, using the same optical configuration and setup employed in the
XANES experiments. EXAS data reduction has been performed according to standard
procedure using DEMETER software package53. For each spectrum, the raw data has been
normalized by calculating and subtracting pre-edge and post-edge backgrounds as low
order polynomial smooth curves. Successively the EXAFS oscilation signal has been
8
ACS Paragon Plus Environment

Page 9 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
extracted. The useful k-range available for the EXAFS analysis was between 2.8 Å-1 and
12.5 Å-1. A sine window has been used to truncate the EXAFS signal and a k-square
weighted Fourier transformed (FT) of it was used for visualization of the real space
distribution of Co local structure.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.2.3. Near ambient pressure X-ray photoemission spectroscopy (NAP-XPS)
The X-ray photoemission spectra were recorded at the NAPP branch from CIRCE, an
undulator beamline with a photon energy range 100-2000 eV at the ALBA Synchrotron
Light Source54. The acquisition was performed using a PHOIBOS 150 NAP electron
energy analyzer (SPECS GmbH) equipped with four differential pumping stages and a set
of electrostatic lenses which enable to perform XPS measurements with the sample at
pressures from ultra-high vacuum (UHV, with a base pressure of 2−5x10−10 mbar) and up
to 20 mbar. The take-off angle was approximately 50º and the angle between the analyzer
axis and the incoming synchrotron radiation horizontal linear polarization vector was 54.7º,
the magic angle. The spectra were acquired with 20 µm exit slit and the diameter of the X-
ray spot was estimated to be in the order of 100 µm x 65 µm (H x V). Two different photon
energies were used: 965 and 1250 eV, to provide variable surface sensitivity, i.e. the
inelastic mean free paths (IMFP) were close to 0.5 and 0.9 nm respectively55. The spectra
were acquired with 20 eV pass energy with an overall (analyzer and monochromator)
energy resolution lower than 300 meV full width at half-maximum (FWHM): 276 meV for
hν= 965 eV and 291 meV for 1250 eV. For the data analysis, the spectra were calibrated
with respect to C1s (at a binding energy of 284.7 eV) and fitted by Gaussian/Lorentzian
convolution functions using the CASAXPS software. The background was modeled by a
9
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 44
1
2
3
4
5
6
7
8
9
Shirley background function. The ethanol and water mixture for the ESR was prepared in a
bubbler like for the XANES experiments, but using Ar as carrier gas, in order to maintain
an ethanol:water ratio of 1:6 in the analysis chamber (S/C=3). Reactants were introduced
into the analysis chamber by means of a leak valve. Another leak valve was used to feed
H2. The NAP-XPS experiments were performed at a sample pressure of 0.1 mbar. The
samples were mounted on a stainless steel plate and were heated using an infrared laser
with λ=808 nm. The sample temperature was monitored with a K-type thermocouple in
direct contact with the samples.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3. RESULTS AND DISCUSSION
3.1. In situ XANES
The K-edge X-ray Absorption Near Edge Structure (XANES) of cobalt has been often
used to characterize the bulk of Co-based materials due to the possibility of detecting
spectral changes in the near edge features related to the local structural and electronic
properties of the compounds. These spectral variations appear as resonances characterized
by different energy positions and/or intensities in the pre-peak and absorption edge56.
Figure 1 shows the evolution of the Co K-edge XANES spectra of the three different Co-
based materials studied, i.e. Co3O4, Co-talc and Co-HT, under ESR reaction conditions.
Additionally, the spectra of five different cobalt compounds are presented in the bottom of
the plot to be used as references: (a) LiCoO2, (b) Co3O4, (c) CoO, (d) Co(OH)2 and (e)
metallic Co foil. When increasing the sample temperature from room temperature (RT) to
550ºC under ESR reaction conditions, Co K-edge XANES reveals significant changes in
the cobalt electronic structure which indicate a reduction process for all the samples
10
ACS Paragon Plus Environment

Page 11 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
studied. The spinel phase Co3O4 exhibits a mixture of Co oxidation states where two thirds
from the total Co ions correspond to trivalent cations in octahedral positions while one third
corresponds to divalent cations occupying tetrahedral sites51,57. No significant changes are
observed in the pre-edge features when heating from 250ºC up to 400ºC and only partial
reduction to CoO is detected at temperatures around 400ºC, but above 500ºC, the formation
of metallic cobalt phase masks the contribution of the cations (Figure 1). For the Co-talc
system there are not clear spectral changes up to 500ºC and from this temperature to 550ºC,
the spectral features reveal that part of the cobalt is reduced to metallic phase, as shown in
Figure 1, where a clear shoulder at ca. 7709 eV assigned to metallic cobalt appears.
However, a different behavior is observed for the Co-HT system. In this case, like for the
Co3O4 reference sample, there is a partial reduction to Co(II) already detected at 300ºC but
there is no further reduction to metallic Co at higher temperatures. The final state for the
Co-HT is characterized by an increase of the intensity of the white line and the resonance
transitions and multiple scattering contributions seem to correspond to a combination of
CoO and Co(OH)2 phases. Nevertheless, some features like the pre-edge and oscillations
intensities are not fully reproduced by linear combination. Unlike in the other studied
samples, the pre-edge peak for the Co-HT remains with the same shape even at 550ºC (see
Figure 1) and there is no indication of reduction to metallic Co. In fact, the Co-HT sample
has been previously calcined and its thermal decomposition yields a well dispersed mixture
of CoAl spinels50, which is similar to cobalt spinels, but cobalt is in a more complex
environment surrounded by other oxide phases (MgO, Al2O3)48.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
11
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 44
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. In situ Co K-edge XANES during ESR reaction at different temperatures for the
three tested catalysts: Co3O4, Co-talc and Co-HT. The spectra for each temperature are
displayed with different colors: 250 (black), 300 (green), 400 (orange), 500 (blue) and
550ºC (red). In the bottom the spectra for five reference compounds are plotted (all at room
temperature): (a) LiCoO2 (black), (b) Co3O4 (green), (c) CoO (orange), (d) Co(OH)2 (blue)
and (e) Co foil (red).
12
ACS Paragon Plus Environment

Page 13 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
The similarities in the spectral features for different chemical environments and the
complexity of the reduction pathway in samples such as Co-talc and Co-HT could lead to
misinterpretation of the experimental results. However, the cobalt average oxidation state
can be easily estimated considering the energy edge shift in our samples and other standard
cobalt compounds and using integral methods as described by Capehart et. al.58. This
approach has been already applied in order to follow the reduction process of Co in bulk
materials59. Figure 2a shows the calibration curve obtained using metallic Co, CoO, Co3O4
and LiCoO2 as reference compounds and employing this integral method (a more detailed
description of its application can be found in the supplementary information). In this way it
is possible to get the evolution of the average oxidation state of the samples studied under
ESR conditions and different temperatures (Figure 2b). Co3O4 shows a slight decrease of
the average oxidation state from 250ºC to 400ºC which correlates to the partial formation of
CoO described from the spectra evolution in Figure 1, but from 400ºC to 500ºC a dramatic
reduction to metallic state is observed. The average charge of Co ions in the Co-talc sample
at 250ºC is close to 2.1 0.2 probably due to different local environments and strong
interactions within the silicate matrix that would also explain the different spectral features
observed with respect to the CoO reference sample. Then, when increasing the temperature
to 500ºC, the Co oxidation state decreases down to 1.4 0.2, which indicates the partial
reduction to metallic Co and therefore the presence of a mixture of chemical phases as
shown in the near edge features (Figure 1). On the other hand, Co-HT exhibits an oxidation
state close to 3 at 250ºC which smoothly decreases one unit at 400ºC and no further
reduction is observed at higher temperatures. For this sample there are no insights of
metallic Co formation as stated when discussing the XANES spectra.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 44
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. (a) Calibration curve employing different oxide references. These values have
been calculated using integral methods as described by Capehart et. al.58 (see
supplementary information). The curve slope was 2.26 0.08 with R2 = 0.996. (b) Average
oxidation state of the bulk material under ESR reaction conditions as a function of
temperature.
More detailed information has been obtained for Co3O4, Co-talc and Co-HT systems
through chemical speciation by linear combination of XANES references. Co3O4, CoO and
metallic Co were employed to analyze the Co3O4 reduction under ESR conditions (Figure
14
ACS Paragon Plus Environment

Page 15 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
3a). The references used to analyze the behavior of Co-talc under ESR conditions were Co-
talc (the same sample measured at room temperature under He atmosphere), CoO and
metallic Co (Figure 3b) while the selected references to analyze the chemical changes of
Co-HT were the original Co-HT, CoO and Co-HT after H2 reduction (named as Reduced
Co-HT), as illustrated in Figure 3c. The selection of these references was based on the
similarities of their spectral features with the ones shown by the samples studied at different
temperatures. For both samples, Co3O4 and Co-talc, the cobalt reduction to the metallic
form is through a Co(II) intermediate (most likely CoO). For Co3O4 the formation of
metallic Co at 500ºC is already close to 100% (Figure 3a). On the other hand, for Co-talc,
previous crystallographic measurements reported for similar nanocomposites revealed that
phase segregation takes place through delamination of the talc nanolayers under ESR
reaction conditions47. Figure 3b shows that Co-talc treated at 550ºC still has around 32% of
the cobalt inside the matrix, 40% as CoO and 28% as metallic Co. From the results at this
temperature, the average oxidation state calculated for Co is 1.47 0.06 (this value depends
on the formal charge assigned to the Co in the Co-talc, in this case considered as +2.1)
which is in agreement with the value obtained by the integral method. Co-HT, initially with
an average oxidation state of approximately +2.8 (named as Co-HT in Figure 3c), yields
Reduced Co-HT (as named in Figure 3c) with an average oxidation state close to +2 even
below 400ºC. In this case, it was not possible to complete the spectral fitting with only
these two contributions (Co-HT and Reduced Co-HT) and a third component was required
in order to improve the linear fit. The features for the third contribution were very similar to
the CoO phase and this contribution was more important at high temperatures than near
400ºC. Although it has been remarked the complexity in which different phases can coexist
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
15
ACS Paragon Plus Environment

ACS Catalysis
Page 16 of 44
1
2
3
4
5
6
in this sample under different conditions, this assumption worked well for explaining the
reduction behavior in a simple way.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Relative component composition for (a) Co3O4, (b) Co-talc and (c) Co-HT
samples under ESR reaction conditions as a function of temperature (R2 is higher than
0.996 in all cases).
3.2. In situ EXAFS
Figure 4a shows the raw EXAFS data acquired in situ under ESR reaction conditions and
different temperatures for Co3O4 and Figure 4b their corresponding Fourier Transform
(FT). Following the results of the linear combination analysis of the XANES, and
comparing to the reference spectra, the Co3O4 pure fraction progressively decreases in
favor of CoO and metallic Co. Coherently the EXAFS signals can be analyzed as a
combination of these three phases. The analysis of the other samples, Co-talc and Co-HT is
complex due to the multiple phases present. Under similar reaction conditions than for
Co3O4, the raw EXAFS data and the corresponding FT for Co-talc and Co-HT are shown in
Figure 5a,b and Figure 6a,b respectively. In a similar way than Co3O4, Co-talc EXAFS
signals change significantly during reaction and, in agreement with the results of the linear
16
ACS Paragon Plus Environment

Page 17 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
combination analysis of the XANES, the Co-talc fraction also progressively decreases in
favor of CoO and metallic Co, but to a lesser extent that the Co3O4 sample. Nevertheless,
unlike Co3O4 and Co-talc, Co-HT EXAFS signals don’t change significantly during the
ESR reaction at different temperatures, which again is in agreement with the XANES
results exposed above and supports the no formation of metallic Co for this sample.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. (a) Raw EXAFS data for the Co3O4 sample under ESR reaction conditions and at
increasing temperatures. (b) Corresponding Fourier Transformations. The spectra for each
temperature are displayed with different colors: 250 (black), 400 (orange) and 550ºC (red).
In the bottom the spectra for four reference compounds are plotted (all at room
temperature): LiCoO2 (black), Co3O4 (green), CoO (orange) and (e) Co foil (red).
17
ACS Paragon Plus Environment

ACS Catalysis
Page 18 of 44
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. (a) Raw EXAFS data for the Co-talc sample under ESR reaction conditions and
at increasing temperatures. (b) Corresponding Fourier Transformations. The spectra for
each temperature are displayed with different colors: 250 (black), 400 (orange) and 550ºC
(red). In the bottom the spectra for four reference compounds are plotted (all at room
temperature): LiCoO2 (black), Co3O4 (green), CoO (orange) and (e) Co foil (red).
18
ACS Paragon Plus Environment

Page 19 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. Raw EXAFS data for the Co-HT sample under ESR reaction conditions and at
increasing temperatures. (b) Corresponding Fourier Transformations. The spectra for each
temperature are displayed with different colors: 250 (black), 400 (orange) and 550ºC (red).
In the bottom the spectra for four reference compounds are plotted (all at room
temperature): LiCoO2 (black), Co3O4 (green), CoO (orange) and (e) Co foil (red).
3.3. In situ Near Ambient Pressure XPS (NAP-XPS)
The surface and subsurface regions constitute the active part of the catalyst in catalytic
heterogeneous reactions since it is the part that interacts with the reactant gases. X-ray
Photoemission Spectroscopy (XPS) is a technique traditionally used under high vacuum
conditions that allows studying the composition and electronic structure of surfaces. Since
19
ACS Paragon Plus Environment

ACS Catalysis
Page 20 of 44
1
2
3
4
5
6
7
8
9
the results from experiments performed ex situ under vacuum conditions can differ from
those obtained under more realistic environments, XPS, among other techniques, has been
further improved in order to break or diminish the so-called pressure gap60. In the case of
XPS, the implementation of a differentially pumping system and a set of electrostatic lenses
have enabled to work at near ambient pressures with good statistics61–63. The possibility of
studying samples under different gas mixtures has proved that the surface composition
directly depends on the reaction environment, among other factors, pointing out the need to
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
measure under operando or in situ conditions18,64
.
In situ NAP-XPS measurements were done in order to study the chemical state and local
environment of cobalt during the ESR reaction thus providing information about its surface
chemistry and complement the bulk information obtained by XANES. In order to
understand the catalysts behavior under the ethanol/water reaction atmosphere used for
ESR reaction, where H2 is generated creating a local reducing environment, the catalysts
were also exposed in a separate experiment (with a different pellet made from the same
sample) to pure H2 to compare the final state for both experimental conditions. For all the
samples, spectra were acquired under in situ conditions at 0.1 mbar and room temperature,
and then the temperature was increased while the pressure was kept constant.
The C 1s signal did not show significant changes under ESR reaction conditions (Figure
S5) for all the samples studied. Taking into account the amount of carbon deposition
measured in our previous catalytic tests45–48, longer reaction times are required to follow the
evolution of the C 1s signal, which was not possible due to beamtime limitation.
The interpretation of Co XP spectra is not straightforward due to peak asymmetries,
binding energy (BE) overlap of the metallic and oxide species and their satellites, the
different complex environments present in Co-talc and Co-HT samples and the charging
20
ACS Paragon Plus Environment

Page 21 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
effects. In order to minimize the uncertainty and error in the peak assignment process, the
data treatment has been done according to the procedure reported by Biesinger et. al.65
which describes in detail the curve fitting procedure for various chemical states of Co-
containing samples. Following this reference, only the more intense components and its
satellites have been considered in a more simplified way to facilitate the interpretation of
the results. In spite of the complications to measure non-conductive samples like Co-talc, or
samples exhibiting complex environments, like Co-talc and Co-HT, a clear evolution is
observed in all cases, as illustrated below.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3.3.1. Co3O4
Figure 7 shows the XP spectra of Co3O4 under H2 (Figure 7a,b) and ESR conditions
(Figure 7c,d) with two different incident photon energies to provide information at different
depths: 965 eV (Figure 7a,c) and 1250 eV (Figure 7b,d). There is a common trend in both
gas environments, a progressive reduction of Co3O4 when heating. At 200ºC, the spectra
under ESR mixture and H2 show the typical shape for the Co3O4 cubic normal spinel
structure, which exhibits one third of tetrahedrally coordinated Co2+ and two thirds of
octahedrally coordinated Co3+. Nevertheless, at the highest temperature tested (580ºC) the
samples are reduced to CoO (or hydroxide), and metallic cobalt traces are detected during
ESR conditions but not under H2 atmosphere. The peak associated to the main Co 2p3/2
component of Co3O4 (779.6 eV) is sharper and at slightly lower energy than the one
corresponding to CoO (780.0 eV and similar values for the hydroxide species)65. Besides,
Co3O4 exhibits a weak shake-up satellite at higher binding energies (around 10 eV from the
main peak) while CoO has a characteristic strong satellite at 786 eV (around 6eV above the
main peak). This strong satellite observed in the CoO photoemission is explained by the
21
ACS Paragon Plus Environment

ACS Catalysis
Page 22 of 44
1
2
3
4
5
6
7
8
9
octahedrally coordinated, high-spin Co2+ oxides, where charge is transferred from the
ligand to the 3d orbital enabling different final states in the photoemission process. Instead,
for Co3O4 the satellite peak is weaker because the two thirds of octahedrally coordinated
Co3+ states do not contribute to charge transfer and only the remaining one third
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
tetrahedrally coordinated Co2+ give rise to shake-up peaks66,67
.
Figure 7. In situ near ambient pressure Co 2p3/2 XPS for Co3O4 exposed to H2 (a,b) and
ESR atmosphere (c,d). The experiments were carried out at different temperatures as
indicated in the right panel and with two different photon incidence energies: 965eV (a,c)
and 1250 eV (b,d). Co3O4: blue lines, CoO: black lines, metallic Co in orange, in grey the
background and in red the envelope.
The evolution of the surface relative composition (Co3O4 vs CoO) under a particular gas
environment with the temperature (Figure 8) does not show significant differences at
22
ACS Paragon Plus Environment

Page 23 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
different penetration depths, only a slight surface oxidation under ESR conditions at 200ºC
that disappears progressively when increasing the temperature. However, comparing the
evolution under H2 with respect to ESR conditions, the sample is more reduced when
exposed to ESR reaction conditions under the temperature range tested. These results
indicate that at the temperatures at which the material is active for hydrogen production the
ESR environment is more reducing than pure H2. Even at lower temperatures, where the
dehydrogenation of ethanol takes place, the sample is overall more reduced under ESR. It is
worth noting that traces of metallic Co are detected at 580ºC using 965 eV incidence
energy (this was confirmed by increasing the temperature to 660ºC, see Figure S2 in
supporting information) while this metallic phase is not observed under H2 gas.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 8. Relative surface composition (Co3O4 vs CoO) as a function of temperature and
penetration depth for: (a) pure H2, and (b) ESR reaction mixture.
23
ACS Paragon Plus Environment

ACS Catalysis
Page 24 of 44
1
2
3
4
5
6
7
8
9
Figure 9 shows the O1s spectra of Co3O4 obtained with 965 eV photon energy, therefore
displaying the most surface sensitive results. The spectra are composed by the contributions
from the oxygen lattices of Co3O4 and CoO but the energy separation between the O1s peak
of both oxides is less than 0.2 eV and it is not possible to resolve them (both with a BE
close to ~530 eV). Other contributions to the O1s signal arise from hydroxide or hydrated
oxide species, around ~531.5 eV, and from surface bonded water or organic alcohols
around 532.5 eV65. At 200ºC the reduction of cobalt by pure H2 produces water that
hydrates the surface and about 40% of the total O1s total area correspond to adsorbed H2O
and hydroxide species while ESR promotes hydroxide formation increasing this percentage
to 60%. When the temperature is increased to 580ºC, these contributions decay to 13-14%
both under H2 and ESR gas environments and the spectra obtained are equivalent (Figure
9).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
24
ACS Paragon Plus Environment

Page 25 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 9. In situ near ambient pressure O1s XPS (acquired with 965 eV photon energy) of
Co3O4 exposed to: (a) H2, and (b) ESR atmosphere. The experiments were carried out at
different temperatures as indicated in the right panel. The contribution of water or alcohol
bounded to the surface region is displayed in green and the contribution of the hydroxide
(or hydrated) oxide in orange. The black line represents the Co3O4 contribution and the blue
line the CoO. The weight of each compound was estimated by the Co 2p3/2 fitting
procedure described before65.
3.3.2. Co-talc
25
ACS Paragon Plus Environment

ACS Catalysis
Page 26 of 44
1
2
3
4
5
6
7
8
9
In the Co-talc sample, the cobalt ions are embedded in a complex matrix formed by the
talc structure and this fact justifies the broad peaks and satellites obtained for Co 2p spectra
(Figure 10). Under both gas environments studied, at lower temperatures (200ºC) - unlike
Co3O4 - there is already a clear satellite which indicates that most of the cobalt corresponds
to high spin Co(II), which is in agreement with the structure of the Co-talc, where the metal
occupies octahedral sites with a Co2+ oxidation state. However, the peaks are broad due to
charging and to the presence of different local environments of the Co atoms within the talc
matrix. When heating under H2 gas there are changes in the spectra that could be ascribed
to Co reorganization or migration up to 400ºC, but it is difficult to assign different
components. In this case it can only be noted a clear evolution to metallic Co at higher
temperatures (500 - 580ºC).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Under ESR reaction conditions and high temperatures (400 - 500ºC) there is a phase
segregation process between the talc sheets and CoO (which exhibits a sharper satellite)
and then at 580ºC CoO is partially reduced to metallic cobalt, as illustrated in Figure 10c,d.
Preliminary transmission electron microscopy studies on similar Co-talc catalysts exposed
to ESR reaction conditions (ex situ) revealed the partial formation of metallic Co NP of 3
nm size47. This CoO phase transition is not so easily detected under H2. In this case the
reduction to metallic Co at 580ºC is not partial as observed for ESR conditions, but almost
complete. Furthermore, it is worth noting that the metallic Co contribution is larger at
higher photon energies (bigger penetration length) for both reactive environments. It has
been reported that the formation of a very stable 3 to 5 nm thick CoO phase takes place
during redox reactions of Co NPs by a complex mechanism which can inhibit the reaction
path to metallic nanoparticles (or reoxidize them into Co3O4)68. The reduction of cobalt
under H2 gas involves cation migration and metallic NP formation, while under ESR
26
ACS Paragon Plus Environment

Page 27 of 44
ACS Catalysis
1
2
3
4
reaction conditions, in situ XPS suggests the formation of a CoO phase before reduction
5
6
which could partially hamper the evolution to metallic Co.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
As previously mentioned, in the case of the Co3O4 sample metallic Co is only detected as
a trace under the ESR reaction conditions studied, therefore the formation of metallic Co
NP from the Co-talc sample exposed to the same experimental conditions could be
explained through a combination of surface energy and mechanical tension due to the
silicate layers confinement. While for the Co3O4 sample the reduction degree is higher
under ESR conditions than under H2 atmosphere; for the Co-talc sample the measurements
under H2 result in a higher reduction degree as compared to ESR reaction conditions. In
addition, the high water partial pressure in the ESR atmosphere could also hinder the
formation of SiO2, which is formed during the reduction of the Co-talc structure.
27
ACS Paragon Plus Environment

ACS Catalysis
Page 28 of 44
1
2
3
4
5
6
7
8
9
Figure 10. In situ near ambient pressure Co 2p3/2 XPS for Co-talc exposed to H2 (a,b) and
ESR atmosphere (c,d) The experiments were carried out at different temperature as
indicated in the right panel and with two different photon incidence energies: 965eV (a,c)
and 1250 eV (b,d). Co+2: black lines, metallic Co: orange lines. In grey the background and
in red line the envelope.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3.3.3. Co-HT
Figure 11 shows the analysis of the XP Co 2p spectra for the Co-HT sample acquired
with two different photon energies under H2 gas (Figure 11a,b) and ESR conditions (Figure
11c,d). In all cases there is a clear reduction from Co(III) to Co(II) indicated by a
remarkable increase of the satellite attributed to high spin Co(II) between 500ºC and 580ºC,
but there is no evidence of metallic Co in all the experimental conditions tested. Three
contributions were used in order to fit the experimental data: a Co(III) contribution at low
energies (~779 eV); a low spin Co(II) from spinel structure at ~780 eV with a small
satellite 9.5-10eV shifted respected from the main peak; and finally, a high spin Co(II) like
CoO oxide with a 2p3/2 transition at ~781eV with a satellite with 6eV difference respected
from the main peak.
Above 500ºC there are no significant differences between the spectra obtained under both
reaction conditions except for the width of the Co 2p peaks, which are larger when
measuring under pure H2. The thermal treatment in H2 reducing media could lead to
CoAl2O4 formation, which exhibits around 1 eV shift of the Co 2p to higher energies with
respect to the CoO contribution69. Instead, in ESR atmosphere the Co 2p peaks are
narrower and this fact could be attributed to a thin CoO or Co(OH)2 surface layer, which
28
ACS Paragon Plus Environment

Page 29 of 44
ACS Catalysis
1
2
3
34
should be more active for hydrogen production than CoAl2O4 . The presence of MgO
could lead to complex dehydrogenation and dehydration routes depending on the
temperature conditions70, and the generated by-products along with the presence of water
could produce a partial segregation of the cobalt oxide from the main matrix. XP spectra of
the Al 2p and Mg 2s regions did not change significantly during the experiments and they
are assigned to their respective oxides (data not shown).
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 11. In situ near ambient pressure Co 2p3/2 XPS of Co-HT exposed to H2 (a,b) and
ESR atmosphere (c,d). The experiments were carried out at different temperatures as
indicated in the right panel and with two different photon energies: 965eV (a,c) and 1250
eV (b,d). Co+2 in tetrahedral environments: black lines, high spin Co+2 in octahedral
positions: green lines, Co3+: blue lines. In grey the background and in red line the envelope.
4. CONCLUSIONS
29
ACS Paragon Plus Environment