Ambient pressure XPS and IRRAS investigation of ethanol steam reforming on Ni-CeO2(111) catalysts: An: in situ study of C-C and O-H bond scission

Article abstract of DOI:10.1039/c6cp01212d

Ambient-Pressure X-ray Photoelectron Spectroscopy (AP-XPS) and Infrared Reflection Absorption Spectroscopy (AP-IRRAS) have been used to elucidate the active sites and mechanistic steps associated with the ethanol steam reforming reaction (ESR) over Ni-CeO₂(111) model catalysts. Our results reveal that surface layers of the ceria substrate are both highly reduced and hydroxylated under reaction conditions while the small supported Ni nanoparticles are present as Ni⁰/Ni_xC. A multifunctional, synergistic role is highlighted in which Ni, CeO_x and the interface provide an ensemble effect in the active chemistry that leads to H₂. Ni⁰ is the active phase leading to both C-C and C-H bond cleavage in ethanol and it is also responsible for carbon accumulation. On the other hand, CeO_x is important for the deprotonation of ethanol/water to ethoxy and OH intermediates. The active state of CeO_x is a Ce³⁺(OH)_x compound that results from extensive reduction by ethanol and the efficient dissociation of water. Additionally, we gain an important insight into the stability and selectivity of the catalyst by its effective water dissociation, where the accumulation of surface carbon can be mitigated by the increased presence of surface OH groups. The co-existence and cooperative interplay of Ni⁰ and Ce³⁺(OH)_x through a metal-support interaction facilitate oxygen transfer, activation of ethanol/water as well as the removal of coke.

Full text of DOI:10.1039/c6cp01212d

PCCP
View Article Online
View Journal
PAPER
Ambient pressure XPS and IRRAS investigation of
ethanol steam reforming on Ni–CeO2(111) catalysts:
an in situ study of C–C and O–H bond scission†
Cite this:DOI: 10.1039/c6cp01212d
Zongyuan Liu,ab Tomas Duchon,c Huanru Wang,a David C. Grinter,a
´ˇ
ˇ
Iradwikanari Waluyo,a Jing Zhou,d Qiang Liu,e Beomgyun Jeong,e
Ethan J. Crumlin,e Vladimır Matolın,c Dario J. Stacchiola,a Jose A. Rodriguezab and
´
´
´
Sanjaya D. Senanayake*a
Ambient-Pressure X-ray Photoelectron Spectroscopy (AP-XPS) and Infrared Reflection Absorption
Spectroscopy (AP-IRRAS) have been used to elucidate the active sites and mechanistic steps associated
with the ethanol steam reforming reaction (ESR) over Ni–CeO2(111) model catalysts. Our results reveal
that surface layers of the ceria substrate are both highly reduced and hydroxylated under reaction condi-
tions while the small supported Ni nanoparticles are present as Ni0/NixC. A multifunctional, synergistic
role is highlighted in which Ni, CeOx and the interface provide an ensemble effect in the active chemistry
that leads to H2. Ni0 is the active phase leading to both C–C and C–H bond cleavage in ethanol and it is
also responsible for carbon accumulation. On the other hand, CeOx is important for the deprotonation of
ethanol/water to ethoxy and OH intermediates. The active state of CeOx is a Ce3+(OH)x compound that
results from extensive reduction by ethanol and the efficient dissociation of water. Additionally, we gain an
important insight into the stability and selectivity of the catalyst by its effective water dissociation, where
the accumulation of surface carbon can be mitigated by the increased presence of surface OH groups.
The co-existence and cooperative interplay of Ni0 and Ce3+(OH)x through a metal–support interaction
facilitate oxygen transfer, activation of ethanol/water as well as the removal of coke.
Received 22nd February 2016,
Accepted 29th March 2016
DOI: 10.1039/c6cp01212d
www.rsc.org/pccp
bonds in hydrocarbon oxygenates16–20 and have been reported
to exhibit activity and/or selectivity comparable to those of
Introduction
Traditionally, the steam reforming of ethanol (ESR: C2H5OH + expensive noble metals such as Rh, Pt and Pd.21–23
3H2O - 6H2 + 2CO2) is an important process in the chemical
The behavior of ceria supported nickel nanoparticles is quite
industry, while more recently it has been employed in fuel cell different from that of bulk Ni, a system which is very active for
applications, where it provides an alternative route to obtain CO methanation and undergoes rapid deactivation by coke
renewable hydrogen through the reforming of hydrocarbon formation during the steam reforming process.8,21,24 Ni and
oxygenates (e.g. bio-ethanol), from sustainable sources such ceria, when combined in the same catalyst, have the ability to
as biomass.1,2 Metal supported oxide catalysts have shown activate both ethanol (C–C and C–H bonds) and H2O (O–H), and
significant activity for this reaction.1,3–15 Notably, Ni-oxide selectively extract H2, without the production of CH4 or other
based materials have emerged as promising catalysts for the C–O byproducts (aldehydes or olefins) provided certain phenom-
reforming reaction due to the ability to activate C–C and C–H enological factors are maintained. This includes compositional,
geometric and morphological effects of keeping Ni loading low,
small in size, well dispersed and in good contact with ceria.
a Chemistry Department, Brookhaven National Laboratory, Upton, New York 11973,
Both experimental and DFT calculation results suggest that
USA. E-mail: ssenanay@bnl.gov
b Department of Chemistry, Stony Brook University, Stony Brook, NY, 11794, USA
the small Ni nanoparticles on CeO2(111) are activated, electro-
nically perturbed and can thus influence catalytic chemistry
through metal–support interactions.21,25,26 Hydroxyl groups gen-
erated from these strong metal–support interactions could play
an important role in redox chemistry such as in the water-gas
shift (WGS) reaction and in suppressing deactivation through
coke formation.26
c
ˇ
ˇ ´
Faculty of Mathematics and Physics, Charles University in Prague, V Holesovickach 2,
Praha 8, Czech Republic
d Department of Chemistry, University of Wyoming, Laramine, WY 82071, USA
e Advanced Light Source, Lawrence Berkeley National Laboratory, Berkeley, CA 94720,
USA
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c6cp01212d
This journal is ©the Owner Societies 2016
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
We have previously studied Ni and CeOx based systems as estimated to be B0.15 ML per min. The Ni coverage in all
powder catalysts and identified them to be active for the ESR experiments presented here is approximated to be 0.1–0.15 ML.
reaction, and we also adopted strategies for protecting the Scanning tunneling microscopy (STM) studies show that the
active Ni centers in solid solutions of the NixCeyOz type.8,21 as-annealed Ni particles around this coverage are uniformly
We attributed the sustained activity and good stability to the dispersed on the CeO2(111) film with an average size of 2.7 nm
suppression of sintering under harsh reaction conditions, in width and 0.5 nm in height.33,34
which leads to the sustained contact between Ni and CeOx
and the prevention of coking. Furthermore, previous studies by
Duhamel et al. on hydrogenated Ni–ceria powder catalysts
Results and discussion
proposed that hydroxyls played a key role as interfacial stabi-
Ceria alone is not catalytically active for the ESR process,
lizers and thus aid in keeping the metal and oxide in contact
however hydrocarbon oxygenates such as ethanol and H2O
react with both Ce4+ and Ce3+. Previously, studies under UHV
through an active Ni–OH–CeOx phase that significantly
improved the catalytic activity and stability.27,28 However, no
conditions have identified the ability of Ce3+ to dissociate
H2O into OH, while both Ce3+ and Ce4+ can produce ethoxy
studies have been able to unequivocally confirm this process,
and obtain an in situ surface sensitive perspective into the role
(CH3CH2O–) species through deprotonation of ethanol. It has
also been well established that ethanol is a strong reducing
of hydroxyl groups in the formation of C-intermediates and
agent of ceria converting Ce4+ to Ce3+, while water is a poor
the removal of coke. Extending from the previous ultra-high
vacuum (UHV) study, we present here the use of Ambient
oxidant capable of healing oxygen vacancies through the libera-
tion of H2.35 We will start by unraveling the chemistry of
Pressure X-ray Photoelectron Spectroscopy (AP-XPS) and Infra-
red Reflection Absorption Spectroscopy (IRRAS) to identify the
ethanol and H2O at elevated pressures over CeO2(111).
catalytic chemistry including surface species and active phases
under steady state reaction conditions (40–240 mTorr). By a
sequential and systematic study, we have strived to gain more
insights into the elementary steps of ethanol/water activation,
surface carbon formation and coke removal through the strong
metal–support interaction of a low loading of small Ni nano-
particles on ceria.
Ethanol reaction on CeO2(111)
Fig. 1 shows C 1s and O 1s AP-XP spectra obtained from the
CeO2(111) surface under a pressure of 40 mTorr of ethanol with
and without adding 200 mTorr of H2O at 300 K. With an
exposure of 40 mTorr of ethanol, the C 1s region shows the
formation of ethoxy species on the CeO2(111) surface. The two
peaks located at 287.2 eV and 286.1 eV are attributed to the
methylene (–CH2–O–) and methyl carbon (–CH3) in the ethoxy
species, respectively.15,33,36 In agreement with previous UHV
studies of alcohol adsorption on CeO2(111),15,33,36,37 ethanol
adsorbs dissociatively at 300 K, and no molecularly adsorbed
ethanol is observed on the surface above room temperature.
After mixing with 200 mTorr of water, a small shoulder appears
along with the ethoxy features at a higher binding energy of
288.2 eV, and the relative intensity of the two ethoxy peaks
changes. This indicates that even at room temperature some
reaction occurs between the two reactants, which likely results
Experimental section
AP-XPS studies were performed at the Advanced Light Source in
Berkeley, CA (beamline 9.3.2). A VG Scienta R4000 HiPP analyzer
was used for XPS analysis.29 The O 1s region was probed with a
photon energy of 650 eV, and the C 1s, Ni 3p, and Ce 4d regions
with a photon energy of 490 eV and a resolution of B0.2 eV. The
Ce 4d photoemission lines were used for binding energy calibra-
tion based on the 122.8 eV satellite features.30 The beam effect was
checked by repeated recording of the spectra of the surface
exposed to an X-ray beam at room temperature, and no beam
damage was detected. AP-IRRAS was performed in a combined
UHV/elevated-pressure cell system using a Bruker IFS 66v/S
spectrometer.31,32 Spectra were collected at a resolution of
4 cmꢀ1 by using a grazing angle incidence. The ethanol source
(Pharmco-aaper, ACS/USP grade) was freeze-pumped-thawed in
liquid nitrogen to remove gaseous contaminants.
Ce metal was evaporated onto a Ru single crystal (0001) held
at 700 K in the presence of 5 ꢁ 10ꢀ7 Torr O2, and then annealed
to 800 K for 10 min at the same O2 pressure.33 The ceria films
were estimated to be ca. 4 nm thick (B10 layers of O–Ce–O)
based on the attenuation of the Ru 3d XPS signal with 650 eV
incident photon energy, which can eliminate the interference of
Ru 3d3/2 (284.0 eV) to the C 1s signal. Ni was vapor deposited on
the as-prepared ceria film at 300 K under vacuum and then
annealed to 700 K for 1 min. Ni coverage calibration has been
described elsewhere,33 and the deposition rate of Ni was
Fig. 1 C 1s and O 1s spectra of the CeO2(111) film under 40 mTorr of
ethanol vapor at 300 K with/without adding H2O (200 mTorr).
Phys. Chem. Chem. Phys.
This journal is ©the Owner Societies 2016

View Article Online
PCCP
Paper
in the formation of a small amount of dioxyethylene species well as variations of partial pressures of each gas. Therefore,
through a partial transition from the mono-dentate methylene considering the complexity of the gas mixture (reactants and
carbon (CH3CH2–O–) of ethoxy to bi-dentate dioxy-carbon products), we represent the gas phase feature with only one
(CH3CHO2–) of dioxyethylene.37,38
single broad peak. On the other hand, the broad peak at
In the O 1s region, before adding water, the spectrum can be 532.5 eV that was previously assigned to –OH/ethoxy species
fitted with the lattice oxygen from ceria at 530.4 eV, O from gas gradually decreases in intensity from 500 K and shifts to a
phase ethanol at 534.9 eV and another feature at 532.5 eV. The higher binding energy of 533.0 eV at 600 K. Taking into
peak at 532.5 eV can be assigned to the O from ethoxy species as consideration that other hydrocarbon oxygenate species are
found in ethoxy (CH3CH2O–) and hydroxyl groups (–OH) that present in the C 1s spectrum in addition to ethoxy at higher
are formed upon binding of deprotonated H to lattice O of ceria temperature, this region can have co-existence of species such
due to the dissociative adsorption of ethanol. After adding as O–C–O from dioxyethylene38 and the carboxylate oxygen
200 mTorr of H2O into the reaction to make a B5 : 1 (H2O/EtOH) (–COO) from acetate species41 that also appear in a similar
vapor mixture, deconvolution of the spectrum shows an extra gas region B533.0 eV. The formation of these species is herein
phase feature at 536.0 eV for water vapor.39,40 Moreover, the attributed to the further oxidation of ethoxy species. Further-
peak intensity of the ceria lattice O is significantly attenuated, more, a new feature centered at 531.8 eV appears at 600 K and
which is mainly a result of the increased scattering of photo- dominates at 700 K while the lattice oxygen appears considerably
electrons due to the higher concentration of gas as well as attenuated. Since most of the carbon species have desorbed by
surface adsorbed species between the ceria film and the this temperature based on the C 1s spectrum, the peak at 531.8 eV
analyzer cone.
cannot be assigned to any C-containing species. As will be
The catalyst was then heated from 300 to 700 K by 100 K discussed below, we tentatively assign this peak to the Ce(III)
increments in an ambient of 40 mTorr ethanol and 200 mTorr hydroxide (Ce3+(OH)x) formed in the hydroxyl-rich background.
H2O, and the corresponding AP-XPS results for the C 1s, O 1s
By looking at the Ce 4d region in Fig. 2, the relative degree of
and Ce 4d regions are shown in Fig. 2. In the C 1s spectra, most reduction of the CeO2(111) film can be identified by comparing
of the ethoxy species at 287.2 eV and 286.1 eV recombine with the intensities of the peaks at 122.8 and 126.1 eV, labeled as X0 00
the surface hydroxyls and gradually desorb from the surface and W0 0 0, respectively.30,43 These two features are solely asso-
up to 700 K:
ciated with Ce4+ spin–orbit splitting. Herein, one can observe
that the CeO2(111) film has started to reduce from 500 K to
700 K, and fitting of the Ce 4d spectrum at 700 K indicates that
the surface of the ceria film consists of B40% Ce3+. Therefore
the peak at 531.8 eV in the O 1s region could be assigned
to either O from the Ce2O3 or Ce3+(OH)x-like compound formed
in 240 mTorr of the ethanol/H2O vapor mixture. However,
in previous studies of ceria surfaces by Mullins et al. and
Matolin et al., the reported peak shift from CeO2 to Ce2O3 is
0.3–0.8 eV.30,44 In our case, we observe a 1.4 eV shift from the
lattice oxygen of CeO2, a position that is in the range of
reported hydroxide compounds.45,46 Furthermore, in our pre-
vious study of the ceria film surface with the same extent of
reduction (B40%),33 no feature at 531.8 eV other than slight
broadening of the lattice oxygen peak was evident. Thus, the
assignment of this peak to Ce2O3 can be ruled out, and it is
most likely Ce3+(OH)x hydroxide that is formed by hydroxyla-
tion of surface and subsurface layers of the film during the
exposition of the highly reduced ceria surface to the 240 mTorr
ethanol/water environment.
CH3CH2O–(ad) + –OH(ad) - CH3CH2OH(g) + O(lattice)
However, two additional features appear as a result of the
increasing temperature at 288.2 and 289.8 eV. These two features
can be attributed to dioxyethylene (CH3CHO2–) and acetate
(C2H3OO–) species, respectively, and they occur as a consequence
of the oxidation of ethoxy species.38,41 No additional carbon
species are evident.
In the O 1s region, the most prevalent peak is that of the
lattice O at 530.4 eV. The fitting of the gas phase features, for
ethanol and water, was difficult as these features change in
both the position and intensity under reaction conditions and
temperatures. The changes of the gas phase peak have been
observed previously,42 and they are related to several factors
including the changes in concentration of reactants and pro-
ducts, the work function at the near surface of the sample as
In the results discussed above, no evidence of C–C bond
scission was observed, indicating that ceria alone is not likely to
perform the most essential step in catalytically activating the
ESR reaction, which is consistent with previous studies on both
model and powder systems.8,33 However, ethanol does react
with ceria to produce hydrocarbon oxygenates such as acet-
aldehyde, and ceria can be strongly reduced by ethanol to give
rise to Ce3+ and its associated O vacancies. The creation of Ce3+
and O vacancies will in turn dissociate H2O into –OH groups
and even give rise to the formation of cerium hydroxide
compounds. Furthermore, it is worth noting that despite the
Fig. 2 C 1s, O 1s and Ce 4d spectra of the CeO2(111) surface under
ethanol steam reforming conditions (40 mTorr ethanol + 200 mTorr H2O).
This journal is ©the Owner Societies 2016
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
fact that the 1 : 5 ratio favors a strongly oxidizing reactant
mixture, the role of ethanol as a strong reducing agent is far
greater than the oxidizing strength of water.
Ethanol reaction on Ni–CeO2(111)
Having established the surface chemistry over CeO2(111), we
now turn to the role of Ni nanoparticles supported on CeO2(111).
From our previous studies we have established that the most
catalytically active and stable Ni prevails at low coverage and low
dimensionality, while maintaining a good contact with the ceria
support.21,26 Thus we investigated the ethanol reaction on the
Ni–CeO2(111) surface with a small coverage of Ni (YNi E 0.15 ML).
Following the same procedure of experimentation as that on the
bare ceria film, 40 mTorr of ethanol and 200 mTorr of water were
introduced into the chamber at 300 K and the surface was
subsequently stepwise heated up to 700 K. First, the Ce 4d/Ni
3p regions of AP-XPS results are presented in Fig. 3a and b. The Ni
nanoparticles initially remain as Ni2+ on CeO2(111) at 300 K, in
agreement with our previous study of Ni deposition on ceria.33,47
Upon heating to 700 K, nearly all of the Ce4+ contributions in the
Ce 4d spectrum disappeared, and simultaneously the supported
NiO nanoparticles are reduced to metallic Ni as evident in the Ni
3p region. The reduction extent derived from Ce 4d spectra with/
without Ni deposition is also compared in Fig. 3c. In contrast to
the bare ceria film, the B90% reduction of the Ni–CeO2(111) film
demonstrates that the top few layers of the catalyst are almost
completely reduced to Ce3+ under the steam reforming condi-
tions. This reduction is likely due to several effects, including the
ability of Ni and ceria to dissociate H2O, the strong interaction
between Ni and ceria, and a possible strain in the ceria due to
doping with Ni, which can make the oxide lattice favor the
formation of Ce3+ and O vacancies.21,26
Fig. 4 C 1s and O 1s spectra of the Ni–CeO2(111) surface under ethanol
steam reforming conditions (40 mTorr ethanol + 200 mTorr H2O).
indicative of carbonate groups is observed, revealing that the
C–C bond cleavage has occurred along with further oxidation to
carbonates (–CO32ꢀ).48–51 The formation of carbonates was not
seen in our previous UHV study of ethanol adsorption on
Ni–CeO2(111) likely due to the low pressure of ethanol/water
and the insufficient oxidative environment to give rise to the
complete oxidation from hydrocarbon to carbonate (CO32ꢀ).
Similarly, in the O 1s region, peaks can be clearly seen for
lattice O of ceria (530.4 eV) and the gas phase mixture of water/
ethanol (535.9 eV) at 300 K. Analogous to the reaction on the
bare ceria film and considering the complexity of having multi-
ple surface species, a broad peak (fwhm = 2.1 eV) centered at
532.5 eV can be identified, representing a combination of
surface hydroxyl and hydrocarbon oxygenate related species
(e.g. surface ethoxy, dioxyethylene, acetate, etc.). Upon heating
to 700 K, this peak attenuates and shifts to 533.2 eV, as a result
of the development of surface hydroxyl/ethoxy to more oxidative
species. Moreover, the lattice oxygen peak of CeO2 appears to be
attenuated and has nearly vanished by 700 K, while a peak at
531.8 eV, similar to the one observed on the bare ceria film,
emerges at 500 K and finally dominates at 700 K. This again
suggests the formation of a hydroxide compound. Recalling
that a heavy reduction of the top layers of the Ni–CeO2(111) film
occurred as shown in Fig. 3c, one can conclude that the extent
of hydroxylation and concentration of cerium hydroxide layers
are much more significant compared to those of the bare ceria,
as practically all the lattice O signal from the CeO2(111) film has
been covered. Additionally, we have attempted to ascertain the
extent of the surface hydroxide formation by XPS depth profiling
of this surface (see Fig. S1, ESI†). The oxygen species from
multiple layers of the CeO2(111) film were probed with varying
incident photon energies (550–850 eV), tuning the escape depth of
photoelectrons from the O 1s core level, elucidating the location
of Ce3+ and OH species within the first few layers of the ceria film.
The richest zone of OH and Ce3+ species appears approximately in
the top 1 nm, or B3–4 atomic monolayers of O–Ce–O of this
surface. This provides complementary information about the
For the surface species evolution, Fig. 4 depicts the C 1s/O 1s
AP-XPS results for the heated surface. In the C 1s spectra,
starting from 300 K, a small amount of acetate (289.8 eV) as
well as dioxyethylene species (288.2 eV) is formed along with
the ethoxy species (287.2 eV and 286.1 eV), which matches the
results from the bare ceria film. However, the difference in
ethanol reaction between Ni–CeO2(111) and the CeO2(111) film
becomes apparent when heating up to 700 K. A peak at 290.3 eV
Fig. 3 (a and b) Ce 4d and Ni 3p spectra of the Ni–CeO2(111) surface
under ethanol steam reforming conditions (40 mTorr EtOH + 200 mTorr
H2O), (c) Surface Ce3+ concentration comparison between CeO2(111) and
Ni–CeO2(111) catalysts.
Phys. Chem. Chem. Phys.
This journal is ©the Owner Societies 2016

View Article Online
PCCP
Paper
surface state of Ni–CeOx catalysts, in comparison to previous (DRIFTS) study of powder Ni–CeO2,8 the formation of acetates
powder catalysts where we only elucidated the bulk state of the species at room temperature and its subsequent transformation
catalyst and ascertained less than 30% of the bulk to have to carbonates species at higher temperature are evident in both
undergone reduction.8,52 These results suggest that the first few cases, but all the features of the powder system are more
surface layers are not only reduced but also remain hydroxylated pronounced and could be retained at higher temperature, prob-
with a high dispersion of small Ni nanoparticles.
ably due to the fact that the multiple crystal facets of powder
The AP-IRRAS results under similar reaction conditions ceria affect the adsorption/desorption selectivity. In addition, a
further confirm the surface species observed in AP-XPS experi- small peak at 2680 cmꢀ1 corresponding to the n(OD) derived
ments, as shown in Fig. 5. With only ethanol, characteristic infrared from the dissociation of D2O is resolved at 700 K.55 Correlating
peaks for ethoxy species are identified on the Ni–CeO2(111) film this with the Ce(OH)x demonstrated in the AP-XPS results above,
by the CO bands at 1121 cmꢀ1 for monodentate n(CO) and at the observation of hydroxyls (–OD) verifies the participation of
1061 cmꢀ1 for bidentate n(CO) or r(CH3), whereas the CH water in the formation of cerium hydroxide while still acting as
stretching bands are found at 2960 cmꢀ1 and 2925 cmꢀ1 for the feedstock for the reforming reaction.
nas(CH3) and nas(CH2), respectively.53 Moreover, the peaks at
Thus, combining data of ambient pressure XPS and IRRAS,
1444 and 1380 cmꢀ1 are assigned to the bending mode of it is clear that Ni–CeO2(111) is active for ethanol steam reform-
d(CH2) and d(CH3). After adding water (D2O), significant changes ing at B700 K in contrast to the bare ceria film surfaces. Also
were observed compared to the surface with pure ethanol. note that Ni nanoparticles are present as metallic Ni0, on the
A water gas phase peak at 2780 cmꢀ1 shows up, while the highly reduced and hydroxylated ceria surface, indicating that
regions at 1350–1450 cmꢀ1 and 1050–1150 cmꢀ1 become broad the active phase for ethanol steam reforming includes both
and not well resolved, probably due to the co-existence of metallic Ni and Ce3+(OH)x. Nevertheless, how these two com-
multiple bands from ethoxy species and other hydrocarbon ponents correlate with each other and their roles in the steam
oxygenates (acetate/dioxyethylene). A new peak that is indicative reforming reaction under steady state conditions are not
of the nas(COO) band from acetate species emerged at 1580 cmꢀ1
,
obvious. Therefore, we performed additional experiments,
in agreement with the AP-XPS observation in Fig. 4. Upon an initially exposing the surface to ethanol and then sequentially
increase of the temperature up to 700 K, the nearly complete adding water at 700 K, while the catalyst is at its most active
disappearance of the CHx stretching band region at B2900 cmꢀ1 state, as will be discussed below.
suggests the dehydrogenation of CHx surface species to produce
hydrogen, while a broad hump centered at 1450 cmꢀ1 is attrib-
Ethanol reaction on Ni–CeO2(111): adding water in sequence
uted to the small amount of carbonate species, as observed in
On a freshly prepared surface with similar coverage of Ni
the 700 K C 1s spectrum in Fig. 4.54 Compared with the previous
Diffuse Reflectance Infrared Fourier Transform Spectroscopy
presence of 40 mTorr of ethanol first. As observed previously,
(YNi E 0.12 ML), the sample was stepwise heated in the
the surface was drastically reduced from 600 K, and metallic Ni
was formed (see Fig. S2, ESI†). The corresponding O 1s/C 1s
spectra are shown in Fig. 6. The most important differences
with respect to the ethanol steam reforming reaction in Fig. 4
are the C 1s spectra above 600 K. In the absence of water, the
reaction gave rise to a very complex C 1s spectrum where
Fig. 5 AP-IRRAS spectra of the Ni–CeO2(111) surface under ethanol
Fig. 6 C 1s and O 1s spectra of the Ni–CeO2(111) surface under a back-
steam reforming conditions (10 mTorr ethanol + 40 mTorr D2O).
ground of 40 mTorr ethanol at elevated temperature from 300–700 K.
This journal is ©the Owner Societies 2016
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
numerous surface species formed simultaneously as a transi- the enlarged peak at 533.2 eV for carbonate/carboxylate species
tion state by 600 K. The peaks appearing from 288.0 eV to after the addition of water, indicating the re-oxidation of the
290.0 eV are attributed to carbon oxygenate species (–COx) as surface carbon species. This result reveals that water and its
indicated in Fig. 6, probably bound to several parts of the Ni associated hydroxyl groups play a vital role in the removal of
and ceria components. More importantly, at binding energies surface carbon. Intuitively based on this, the H2 liberated from
below 285.0 eV, three new features associated with non- each molecule of water associated with the oxidation is also
oxygenated carbon appeared at 284.8 eV, 283.4 eV and 282.0 eV. potentially a major contributor to the yield of H2 in the overall
The one at 284.8 eV is assigned to surface carbon (C0), indicating mass balance of the reaction.
that coke formation occurred on the surface, while the other
Previous studies have identified that the formation of hydro-
two are attributed to the formation of nickel carbide species xyl groups is favored on reduced ceria,35 and these hydroxyl
(Ni3C/NixC).33,56 Nickel carbide species were similarly observed groups are likely to impart a strong interaction with the
in our previous study on both powder and model Ni–CeO2 oxide supported metal particles.60 Additionally, Ni/Ni2+ on the
systems.8,33 They are also reported as precursors for the growth CeOx(111) substrate can also readily dissociate water to form
of graphene on nickel.57,58 Nickel carbides are metastable and hydroxyls on the surface.26 Furthermore, our earlier isotope
will further decompose into metallic Ni and surface carbon study of the reaction of ethanol on the Ni–CeOx(111) surface
(coke) by 733 K.59 Herein, a more prominent surface carbon under UHV showed that there is oxygen transfer between ceria
feature at 284.8 eV formed at 700 K as evident in the C 1s and surface carbon on Ni nanoparticles, and a significant
spectra, which confirms the importance of metallic nickel, as amount of D2 and CO is produced in the temperature pro-
the active agent responsible for C–C and C–H bond breaking in grammed desorption (TPD) data after introducing D2O into the
the decomposition of ethoxy related intermediates. Moreover, reaction.33 Herewith, considering all the factors mentioned
this experiment reveals that metallic nickel is also responsible above and also noting that top few layers of the ceria film were
for the surface carbon accumulation and that it can drive the fully reduced and hydroxylated at 700 K, this in situ observation
reaction to an unselective reaction pathway in an ethanol rich of oxidation and coke removal by the addition of 200 mTorr of
reactant stream. The chemistry of water and OH formation in water reveals that, under steady state reforming conditions, the
the ESR reaction appears to be pivotal, and may lead to the oxygen or hydroxyl transfer between water and surface carbon
protection of the active sites and mitigate deactivation.1
occurs in the form of hydroxyls and it could go through cerium
To gain further insights into the effect of water and its hydroxide as a possible intermediate. Moreover, the existence
associated hydroxyl groups on the reaction in an in situ man- of metal–support interactions between Ni and ceria after
ner, we sequentially introduced 200 mTorr of water into the depositing a very small amount of Ni (YNi o 0.15 ML) could
ethanol ambient while the surface was kept at 700 K, as shown further facilitate this oxygen transfer.21,26 Therefore, the possible
in Fig. 7. In the C 1s spectra, the instant decrease of the intensity mechanistic pathway on Ni–Ceria could be interpreted as:
of the surface carbon peak at 284.8 eV upon the introduction of
water is evident, whereas the intensity of the carbon oxygenate
related peaks (288.0–289.0 eV) increased, clearly demonstrating
the oxidation process through the complete dissociation of H2O
and incorporation of O. This is also evident in the O 1s region as
On ceria sites:
CH3CH2OH - CH3CH2O(ad) + H(ad);
H2O - OH(ad) + H(ad);
CH3CH2O(ad) + OH/O(lattice) - C2OxHy (e.g. CH3COO–) + nH(ad);
2H(ad) - H2(gas)
On Ni sites or Ce–Ni interfaces:
C2OxHy - COx(ad) + CHy(ad);
COx(ad) - CO/CO2(gas);
CHy + xOH(ad) - CO/CO2 + (x + y)H(ad) and 2H(ad) - H2(gas)
Coking process:
2CO - C (or coke) + CO2;
CHy - C (or coke) + yH(ad).
Fig. 7 C 1s and O 1s spectra of the Ni–CeO2(111) surface under a back-
ground of 40 mTorr ethanol at 700 K before and after adding 200 mTorr
of water.
On the other hand, the reappearance of nickel carbide
species (Ni3C/NixC) after introducing water in the C 1s spectra
Phys. Chem. Chem. Phys.
This journal is ©the Owner Societies 2016

View Article Online
PCCP
Paper
3 P. R. d. l. Piscina and N. Homs, Chem. Soc. Rev., 2008, 37,
2459–2467.
of Fig. 7 points at a possible re-oxidation pathway of surface
carbon through a strong interaction with Ni:
4 J. Llorca, N. s. Homs, J. Sales and P. R. r. de la Piscina,
J. Catal., 2002, 209, 306–317.
OH(ad) + xNi0 + coke - HO–NixC - xNi0 + CO/CO2 + H(ad)
5 V. Palma, F. Castaldo, P. Ciambelli and G. Iaquaniello, Appl.
Catal., B, 2014, 145, 73–84.
6 I. I. Soykal, H. Sohn, D. Singh, J. T. Miller and U. S. Ozkan,
ACS Catal., 2014, 4, 585–592.
7 F. Aupretre, C. Descorme, D. Duprez, D. Casanave and
D. Uzio, J. Catal., 2005, 233, 464–477.
8 W. Xu, Z. Liu, A. C. Johnston-Peck, S. D. Senanayake,
G. Zhou, D. Stacchiola, E. A. Stach and J. A. Rodriguez,
ACS Catal., 2013, 3, 975–984.
Thus, under steady state reaction conditions, it is likely that two
competing processes will take place on the surface: (1) the
accumulation of coke due to C–C/C–H bond cleavage by metallic
Ni and (2) the re-oxidation of surface carbon by hydroxyls provided
by the interaction between water and ceria in the form of
Ce3+(OH)x. Both processes could go through a Ni carbide inter-
mediate phase. More importantly, the redox properties of the oxide
as well as the synergistic interplay between ceria and low loading
of Ni are crucial in order to sustain the stability and selectivity of
the reaction by facilitating the oxygen transfer from water to ceria
and then to surface carbon on nickel nanoparticles.
9 P. Y. Sheng, G. A. Bowmaker and H. Idriss, Appl. Catal., A,
2004, 261, 171–181.
10 P. Ciambelli, V. Palma and A. Ruggiero, Appl. Catal., B, 2010,
96, 18–27.
11 D. K. Liguras, D. I. Kondarides and X. E. Verykios, Appl.
Catal., B, 2003, 43, 345–354.
Conclusions
We have studied the steam reforming of ethanol over a
Ni–CeO2(111) catalyst with a combination of in situ AP-XPS and
AP-IRRAS. Under reaction conditions, metallic Ni is the active
phase leading to both C–C and C–H cleavages of ethanol as well as
the carbon accumulation or coking, while reduced ceria assists in
the deprotonation of ethanol and water with subsequent genera-
tion of hydroxyls which are crucial for further reaction with CHx or
coke. We have obtained from this study a highly surface sensitive
and dynamic perspective of the ESR reaction that reveals that the
near surface layers of the ceria substrate remain highly reduced
and hydroxylated as Ce3+(OH)x, and the coke formation/removal
may occur through a Ni carbide intermediate phase, which was
not established in previous in situ studies of powder Ni–CeOx
systems. We have also presented an important insight into the
stability and selectivity of the catalyst: the accumulation of surface
carbon can be mitigated by the co-existence and cooperative
interplay of small Ni0 particles and Ce3+(OH)x, facilitating hydro-
xyl/oxygen transfer through a strong metal–support interaction.
12 B. Zhang, X. Tang, Y. Li, W. Cai, Y. Xu and W. Shen, Catal.
Commun., 2006, 7, 367–372.
13 H.-S. Roh, A. Platon, Y. Wang and D. L. King, Catal. Lett.,
2006, 110, 1–6.
14 A. L. Alberton, M. M. V. M. Souza and M. Schmal, Catal.
Today, 2007, 123, 257–264.
15 S. D. Senanayake, K. Mudiyanselage, A. Bruix, S. Agnoli,
J. Hrbek, D. Stacchiola and J. A. Rodriguez, J. Phys. Chem. C,
2014, 118, 25057–25064.
16 S. Wang, X. Li, F. Zhang, Q. Cai, Y. Wang and Z. Luo,
Int. J. Hydrogen Energy, 2013, 38, 16038–16047.
17 Z. Guo, S. Wang, L. Guo and X. Li, BioResources, 2011, 6,
4092–4102.
18 M. Dan, M. Mihet, Z. Tasnadi-Asztalos, A. Imre-Lucaci,
G. Katona and M. D. Lazar, Fuel, 2015, 147, 260–268.
19 S. Gates, J. Russell and J. Yates, Surf. Sci., 1986, 171,
111–134.
20 J. Xu, X. Zhang, R. Zenobi, J. Yoshinobu, Z. Xu and
J. T. Yates, Surf. Sci., 1991, 256, 288–300.
21 G. Zhou, L. Barrio, S. Agnoli, S. D. Senanayake, J. Evans,
Acknowledgements
´
A. Kubacka, M. Estrella, J. C. Hanson, A. Martınez-Arias,
The research carried out at Brookhaven National Laboratory
was supported by the U.S. Department of Energy, Office of
Science and Office of Basic Energy Sciences under contract No.
DE-SC0012704. The collaborative work with Charles University
in Prague is part of the KONTAKT project that was supported by
the Ministry of Education of the Czech Republic (LH15272).
The Advanced Light Source is supported by the Director, Office
of Science, Office of Basic Energy Sciences, of the U.S. Department
of Energy under Contract No. DE-AC02-05CH11231.
´
´
M. Fernandez-Garcıa and J. A. Rodriguez, Angew. Chem.,
2010, 122, 9874–9878.
22 F. Frusteri, S. Freni, L. Spadaro, V. Chiodo, G. Bonura,
S. Donato and S. Cavallaro, Catal. Commun., 2004, 5,
611–615.
23 J. P. Breen, R. Burch and H. M. Coleman, Appl. Catal., B,
2002, 39, 65–74.
24 J. Carrasco, L. Barrio, P. Liu, J. A. Rodriguez and
M. V. Ganduglia-Pirovano, J. Phys. Chem. C, 2013, 117,
8241–8250.
25 S. D. Senanayake, J. Evans, S. Agnoli, L. Barrio, T.-L. Chen,
J. Hrbek and J. A. Rodriguez, Top. Catal., 2011, 54, 34–41.
References
´
´
ˇ
1 L. V. Mattos, G. Jacobs, B. H. Davis and F. B. Noronha, 26 J. Carrasco, D. Lopez-Duran, Z. Liu, T. Duchon, J. Evans,
´
´
Chem. Rev., 2012, 112, 4094–4123.
2 G. W. Huber, S. Iborra and A. Corma, Chem. Rev., 2006, 106,
4044–4098.
S. D. Senanayake, E. J. Crumlin, V. Matolın, J. A. Rodrıguez
and M. V. Ganduglia-Pirovano, Angew. Chem., Int. Ed., 2015,
54, 3917–3921.
This journal is ©the Owner Societies 2016
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
ˇ
ˇ´
´
27 C. Pirez, M. Capron, H. Jobic, F. Dumeignil and 44 T. Duchon, F. Dvorak, M. Aulicka, V. Stetsovych, M. Vorokhta,
´
´
ˇ
´
´
L. Jalowiecki-Duhamel, Angew. Chem., Int. Ed., 2011, 50,
D. Mazur, K. Veltruska, T. Skala, J. Myslivecek, I. Matolınova
´
and V. Matolın, J. Phys. Chem. C, 2014, 118, 357–365.
10193–10197.
28 L. Jalowiecki-Duhamel, C. Pirez, M. Capron, F. Dumeignil 45 I. G. Casella, M. R. Guascito and M. G. Sannazzaro,
and E. Payen, Int. J. Hydrogen Energy, 2010, 35, 12741–12750. J. Electroanal. Chem., 1999, 462, 202–210.
29 M. E. Grass, P. G. Karlsson, F. Aksoy, M. Lundqvist, 46 N. Mora, E. Cano, J. L. Polo, J. M. Puente and J. M. Bastidas,
B. Wannberg, B. S. Mun, Z. Hussain and Z. Liu, Rev. Sci.
Instrum., 2010, 81, 053106.
30 D. Mullins, S. Overbury and D. Huntley, Surf. Sci., 1998, 409,
307–319.
Corros. Sci., 2004, 46, 563–578.
47 Y. Zhou, J. M. Perket and J. Zhou, J. Phys. Chem. C, 2010,
114, 11853–11860.
´
´
48 A. Neitzel, Y. Lykhach, V. Johanek, N. Tsud, T. Skala,
´
31 F. M. Hoffmann, Surf. Sci. Rep., 1983, 3, 107–192.
32 K. Mudiyanselage and D. J. Stacchiola, In-situ Characteriza-
K. C. Prince, V. Matolın and J. Libuda, J. Phys. Chem. C,
2015, 119, 13721–13734.
tion of Heterogeneous Catalysts, John Wiley & Sons, Inc., 49 P. M. Albrecht, D.-e. Jiang and D. R. Mullins, J. Phys. Chem. C,
2013, pp. 209–239. 2014, 118, 9042–9050.
33 Z. Liu, T. Duchon, H. Wang, E. W. Peterson, Y. Zhou, S. Luo, 50 S. D. Senanayake and D. R. Mullins, J. Phys. Chem. C, 2008,
ˇ
´
J. Zhou, V. Matolın, D. J. Stacchiola, J. A. Rodriguez and
112, 9744–9752.
S. D. Senanayake, J. Phys. Chem. C, 2015, 119, 18248–18256. 51 S. D. Senanayake, D. Stacchiola, P. Liu, C. B. Mullins,
34 Y. Zhou and J. Zhou, J. Phys. Chem. C, 2012, 116, 9544–9549.
J. Hrbek and J. A. Rodriguez, J. Phys. Chem. C, 2009, 113,
35 D. R. Mullins, P. M. Albrecht, T.-L. Chen, F. C. Calaza,
19536–19544.
M. D. Biegalski, H. M. Christen and S. H. Overbury, J. Phys. 52 Z. Liu, W. Xu, S. Yao, A. C. Johnson-Peck, F. Zhao,
´
P. Michorczyk, A. Kubacka, E. A. Stach, M. Fernandez-
Chem. C, 2012, 116, 19419–19428.
´
36 D. R. Mullins, S. D. Senanayake and T. L. Chen, J. Phys.
Chem. C, 2010, 114, 17112–17119.
Garcıa, S. D. Senanayake and J. A. Rodriguez, J. Catal.,
2015, 321, 90–99.
37 D. R. Mullins, M. D. Robbins and J. Zhou, Surf. Sci., 2006, 53 A. Yee, S. J. Morrison and H. Idriss, J. Catal., 1999, 186, 279–295.
600, 1547–1558.
38 D. R. Mullins and P. M. Albrecht, J. Phys. Chem. C, 2013, 117,
14692–14700.
39 K. Mudiyanselage, S. D. Senanayake, L. Feria, S. Kundu,
A. E. Baber, J. Graciani, A. B. Vidal, S. Agnoli, J. Evans,
R. Chang, S. Axnanda, Z. Liu, J. F. Sanz, P. Liu,
54 C. Li, Y. Sakata, T. Arai, K. Domen, K.-i. Maruya and
T. Onishi, J. Chem. Soc., Faraday Trans. 1, 1989, 85, 929–943.
55 K. Mudiyanselage, S. D. Senanayake, P. J. Ramirez,
S. Kundu, A. Baber, F. Yang, S. Agnoli, S. Axnanda, Z. Liu,
J. Hrbek, J. Evans, J. A. Rodriguez and D. Stacchiola, Top.
Catal., 2015, 58, 271–280.
J. A. Rodriguez and D. J. Stacchiola, Angew. Chem., Int. Ed., 56 I. Czekaj, F. Loviat, F. Raimondi, J. Wambach, S. Biollaz and
2013, 52, 5101–5105. A. Wokaun, Appl. Catal., A, 2007, 329, 68–78.
40 G. Ketteler, S. Yamamoto, H. Bluhm, K. Andersson, 57 R. Rameshan, L. Mayr, B. Klotzer, D. Eder, A. Knop-Gericke,
¨
¨
¨
D. E. Starr, D. F. Ogletree, H. Ogasawara, A. Nilsson and
M. Havecker, R. Blume, R. Schlogl, D. Y. Zemlyanov and
M. Salmeron, J. Phys. Chem. C, 2007, 111, 8278–8282.
S. Penner, J. Phys. Chem. C, 2015, 119, 26948–26958.
´
41 F. C. Calaza, T.-L. Chen, D. R. Mullins, Y. Xu and S. H. 58 R. S. Weatherup, L. D’Arsie, A. Cabrero-Vilatela, S. Caneva,
Overbury, Catal. Today, 2015, 253, 65–76.
42 S. Axnanda, M. Scheele, E. Crumlin, B. Mao, R. Chang,
R. Blume, J. Robertson, R. Schloegl and S. Hofmann, J. Am.
Chem. Soc., 2015, 137, 14358–14366.
S. Rani, M. Faiz, S. Wang, A. P. Alivisatos and Z. Liu, Nano 59 T. Tanaka, K. N. Ishihara and P. H. Shingu, Metall. Trans. A,
Lett., 2013, 13, 6176–6182. 1992, 23, 2431–2435.
43 D. R. Mullins, P. V. Radulovic and S. H. Overbury, Surf. Sci., 60 R. Addou, T. P. Senftle, N. O’Connor, M. J. Janik, A. C. T. van
1999, 429, 186–198.
Duin and M. Batzill, ACS Nano, 2014, 8, 6321–6333.
Phys. Chem. Chem. Phys.
This journal is ©the Owner Societies 2016