[Link].

org/acscatalysis Research Article

Elucidating the Oxygen Activation Mechanism on Ceria-Supported

Copper-Oxo Species Using Time-Resolved X‑ray Absorption
Spectroscopy
Olga V. Safonova,* Alexander Guda, Yury Rusalev, René Kopelent, Grigory Smolentsev, Wey Yang Teoh,
Jeroen A. van Bokhoven, and Maarten Nachtegaal
Cite This: ACS Catal. 2020, 10, 4692−4701 Read Online
See [Link] for options on how to legitimately share published articles.

ACCESS Metrics & More Article Recommendations *

sı Supporting Information

ABSTRACT: Copper-ceria finds applications in various energy-

Downloaded via KAROLINSKA INST on July 18, 2020 at [Link] (UTC).

related and environmental catalysts. However, the versatile

structure and complex redox activity of this material entangle
uncovering structure−activity relationships and distinguishing
active species from spectators. In this work, we monitored the
dynamic structure of the active sites in a catalyst containing highly
dispersed copper-oxo species on ceria during low-temperature CO
oxidation using time-resolved X-ray absorption spectroscopy. We
quantitatively demonstrate that the CO oxidation mechanism
below 90 °C involves an oxygen intermediate strongly bound to the active sites as well as the redox activity of Cu2+/Cu+ and Ce4+/
Ce3+ couples. The redox activity of cerium is much lower than that of copper; however, both metals change their oxidation states in
concert, indicating that oxygen activation involves copper−oxo species in close interaction with ceria. In addition to short-lived Cu+
and Ce3+ intermediates that are generated in the CO oxidation cycle, long-lived Cu+ and Ce3+ species appear in the catalyst under
the working conditions. We demonstrate that they do not participate in the main low-temperature CO oxidation mechanism, which
is mediated by a strongly bound oxygen intermediate. Finally, our results confirm the high potential of element-specific time-resolved
X-ray spectroscopy methods combined with a non-steady-state experimental strategy to uncover the mechanisms of catalytic
processes in complex multicomponent systems.
KEYWORDS: copper-ceria, oxygen activation mechanism, CO oxidation, operando spectroscopy, time-resolved XAS,
non-steady-state kinetics

1. INTRODUCTION versatility of Cu-CeO2 represents a great spectroscopic

Copper-ceria (Cu-CeO2) is one of the most versatile alternatives challenge for elucidating the catalytic mechanisms at the atomic
to precious metal catalysts1 with wide ranging applications that scale. Given its stoichiometric simplicity, CO oxidation is a well-
include the water-gas shift (WGS) reaction,2,3 methanol adopted model reaction to probe the catalytic oxidation
synthesis,4 oxidation and preferential oxidation of CO and mechanisms because the limited intermediate pathways allow
volatile organic carbons,5−7 and the reduction of nitrogen oxides for obtaining an unobscured view in interpreting the mechanistic
(NOx).8−10 The versatility of the catalyst can be traced back to principles. In addition, this reaction has wide implications for
the dynamic and highly dispersed nature of the copper species some of the most fundamental reactions, including WGS and
on the ceria surface, giving rise to interesting, albeit moderately syngas conversions, exhaust treatment and air pollution
understood, redox properties. Furthermore, the synergetic mitigation, and the purification of hydrogen (i.e., preferential
chemical and electronic interactions between copper and ceria oxidation of CO traces).
have been found to be favorable in catalyzing many of the surface Prior kinetics studies on CO oxidation over Cu-CeO2 found
reactions as mentioned above. that the low-temperature reactivity is similar in the absence or
Under oxidizing conditions and at low loading, the copper excess of hydrogen.5,7,21,22 The high selectivity for CO oxidation
species on ceria tend to be stabilized as cationic monomers and
oligomers. However, on subjection to reducing conditions above Received: February 1, 2020
200 °C, the copper species segregate into two-dimensional, Revised: March 19, 2020
partially metallic nanoparticles.3,11,12 In fact, both copper and Published: March 25, 2020
cerium can vary their oxidation states under conditions of a
specific catalytic process.3,4,13−20 Although highly beneficial in
mediating surface reactions, the structural and electronic

© 2020 American Chemical Society [Link]

4692 ACS Catal. 2020, 10, 4692−4701
ACS Catalysis [Link]/acscatalysis Research Article

has been explained as due to the preferential adsorption of CO (ii) The redox activity of cerium is much lower than that of
on copper sites.17 The reaction order with respect to CO copper; however, both metals change their oxidation
decreases from one to zero with increasing CO concentration, states in concert, indicating that oxygen activation
suggesting an equilibrium coverage of the active sites by CO involves copper-oxo species in close interaction with ceria.
during CO oxidation. The reaction order with respect to oxygen (iii) Long-lived Cu+ and Ce3+ species exist on the surface
is typically close to zero (±0.15), which indicates that the during low-temperature CO oxidation, but they do not
majority of oxygen intermediates participating in the reaction participate in the main reaction mechanism.
are strongly bound to the surface.5 Evidence for a Mars−van
Krevelen (MvK) mechanism of CO oxidation was found in 2. EXPERIMENTAL SECTION
transient experiments through the observation of significant CO For this study, we selected a well-established Cu-CeO2 catalyst,
conversion in the absence of gas-phase oxygen.21,23 At the same containing highly dispersed copper-oxo species which is highly
time, other mechanisms of CO oxidation involving weakly active for CO oxidation, prepared by the flame spray pyrolysis
bound oxygen species, for example, peroxide species activated method.12 This Cu-CeO2 (4 wt % Cu) catalyst has a specific
on Cu+,24,25 could not be fully excluded on the basis of kinetics surface area of 93 m2/g, and details about its synthesis,
experiments alone. characterization, and activity assessments can be found
The CO oxidation mechanism at the atomic scale remains elsewhere.12 X-ray photoelectron spectroscopy (XPS), X-ray
uncertain due to the coexistence of various oxidation states of diffraction (XRD), transmission electron microscopy (TEM),
copper and cerium in the catalyst structure, which depends on Raman spectroscopy, electron paramagnetic resonance (EPR)
the synthesis method, copper loading, reaction conditions, and spectroscopy, and elemental analysis suggested preferential
the pretreatment history. In situ infrared spectroscopy studies by segregation of copper oligomers of different nuclearity on the
Gamarra et al.26 and Polster et al.21 demonstrated a correlation surface of ceria without the formation of crystalline CuOx
between the CO oxidation activity of different Cu-CeO2 phases. Systematic variation of the copper concentration
catalysts and the intensity of the Cu+ carbonyl peak. These between 0.5 and 12 wt % demonstrated a maximum of
results suggest that CO is either directly activated on Cu+ to preferential CO oxidation activity for 4 wt % of copper
participate in the CO oxidation reaction or Cu+ carbonyl species coinciding with the strongest signal for Cu2+ dimers in EPR.
appears after reduction of Cu2+ to Cu+ during the CO oxidation In this work, we characterized this catalyst by scanning
cycle. It is also not certain which redox couple compensates the transmission electron microscopy (STEM) and in situ/
negative charge created after the removal of lattice oxygen operando XAS. STEM combined with energy-dispersive X-ray
according to the MvK mechanism. Several possibilities were spectroscopy (EDXS) was performed on a Talos microscope
suggested in the literature, including Cu2+/Cu+,13−17 Ce4+/ (FEI) with a high brightness field emission gun operated at an
Ce3+,5,13 Cu+/Cu0,16 and Cu3+/Cu2+ 18 redox couples. A acceleration potential of 200 kV. Four EDX detectors allowed
“synergetic” mechanism considering fast O2− diffusion from recording EDXS maps within 10−20 min.
cerium to the copper sites, where CO oxidation takes place, was The scheme of the setup for the in situ/operando time-
also proposed.16 In situ and operando experiments using X-ray resolved XAS studies used in our work is shown in Figure 1. A
absorption spectroscopy (XAS),15,27,28 and infrared17,21,26 and
electron paramagnetic resonance (EPR)16 spectroscopies
demonstrate that both copper and cerium can change their
oxidation states under relevant conditions and thus can
potentially participate in the oxygen activation mechanism.
However, the rates of oxidation and reduction of copper and
cerium sites were never quantified and correlated to the CO2
formation rate,29 making the discrimination between possible
mechanisms and verification of species involved in the catalytic
cycle impossible.
In our previous works,30,31 we demonstrated using transient
oxygen cutoff experiments (cycling between a mixture of CO
and oxygen and one of CO) combined with operando time-
resolved X-ray emission spectroscopy methods that the Ce3+
formation rate is kinetically coupled to the rate-determining step
of low-temperature CO oxidation on a Pt/CeO2 catalyst. These
studies indicated that lattice oxygen at the metal−support Figure 1. Scheme of the experimental setup for in situ/operando time-
interface reacts with CO adsorbed on platinum, generating two resolved XAS studies of heterogeneous catalysts in a plug-flow reactor
Ce3+ ions, which are rapidly reoxidized to Ce4+ in the presence of cell.
gas-phase oxygen. Here we used a similar approach to clarify the
mechanism of oxygen activation on highly dispersed copper− sieved fraction of 4 wt % Cu-CeO2 catalyst (12−65 mg)
oxo species on ceria at low temperature, i.e. combining transient containing 100−150 μm particles was loaded in a plug-flow
oxygen cutoff experimentation with X-ray absorption spectros- reactor cell.32 The catalyst was precalcined in 4% oxygen at 300
copy (XAS) at the Cu K and Ce L3 edges. Our results °C. The gas flow was passing through the catalyst in a vertical
quantitatively demonstrate for the first time the following. direction. We used two cartridge heaters (Tuerk-Hillinger) to
control the cell temperature (24−400 ± 1 °C). The temperature
(i) CO oxidation below 90 °C is mediated by a strongly of the catalyst bed was measured by an additional thermocouple
bound oxygen intermediate and by Cu2+/Cu+ and Ce4+/ placed in the middle of the catalyst bed. We used 150 μm thick
Ce3+ redox couples. graphite foils as X-ray windows. The gas-feeding setup consisted
4693 [Link]
ACS Catal. 2020, 10, 4692−4701
ACS Catalysis [Link]/acscatalysis Research Article

Figure 2. (a) Cu K edge XANES spectra of 4 wt % Cu/CeO2 catalyst after calcination in 4% oxygen at 300 °C and in 1% CO at 80 °C in comparison to
Cu2+, Cu+, and Cu0 references. (b) Cu K edge XANES spectra of 4 wt % Cu/CeO2 catalyst in A1 and A2 states and the A1−A2 difference spectrum. (c)
Fourier-transformed Cu K edge EXAFS spectra of 4 wt % Cu/CeO2 catalyst after calcination in 4% oxygen at 300 °C and after exposure to 1% CO at 80
°C in comparison to the corresponding spectra for CuO, Cu2O, and Cu. (d) Ce L3 edge XANES spectra of 4 wt % Cu/CeO2 catalyst in 1% CO and 4%
oxygen at 66 °C (B3 reference) together with the reference spectra of 1.5 wt % Pt/CeO2 catalyst in 1% CO (B1 reference) and in 4% oxygen (B2
reference) and the B1−B2 difference.

of mass flow controllers (El-Flow, Bronkhorst) connected to gas typically kept below 30% to ensure uniform reactivity of the
bottles of 5% CO in argon (CO 4.7 purity, argon 5.0 purity), whole catalyst bed. The state of the catalyst was probed by XAS
21% oxygen in argon (oxygen 4.5 purity, argon 5.0 purity), and in the middle of the catalyst bed. Figure S1 demonstrates how
argon (4.8 purity). During transient experiments, we used two fast the gas atmosphere in the reactor cell can be exchanged.
three-way solenoid valves (Parker) to switch between two gas While switching between two gas mixtures, pure argon versus
mixtures (1% CO and 4% oxygen in argon vs 1% CO in argon) 10% krypton in argon, we monitored the rate of krypton removal
passing through the cell and bypass with a constant flow (50 from the reactor following the decay of the m/z 86 signal of
mL/min). The pressure difference between the cell and the krypton. The exponential decay fit of the resulting curve
bypass was less than 100 mbar, and we further reduced it to 2 demonstrates a characteristic time of 1.4 s.
mbar by a needle valve back-pressure regulator placed at the 2.1. Time-Resolved X-ray Absorption Spectroscopy.
bypass. We analyzed the residual gas at the cell exit using a mass The XAS experiments at the Ce L3 and Cu K edges were
spectrometer (OmniStar GSD 320, Pfeiffer) and monitored the performed at the SuperXAS beamline of the Swiss Light Source
mass to charge (m/z) ratios of 28, 32, 40, and 44 corresponding (PSI, Villigen, Switzerland). The incident photon beam was
to CO, oxygen, argon, and CO2, respectively. We quantified the selected by a Si (111) channel-cut monochromator from the
CO conversion using the CO2 signal (m/z 44) divided by that of polychromatic beam coming from a 2.9 T superbend magnet.
argon (m/z 40). A 1% CO2 in argon gas mixture was used for the The rejection of higher harmonics and the collimation were
calibration of the mass spectrometer. achieved by a silicon-coated collimating mirror at 2.5 mrad
During kinetic tests, we analyzed the catalytic activity of 4 wt located before the monochromator, while focusing was achieved
% Cu-CeO2 in a flow of gas mixtures containing 0.25−4% CO by a rhodium-coated torroidal mirror at 2.5 mrad. For the
and 3−16% oxygen at 39−90 °C. The CO conversion was detection of Ce L3 and Cu K-edge XAS, we calibrated the
4694 [Link]
ACS Catal. 2020, 10, 4692−4701
ACS Catalysis [Link]/acscatalysis Research Article

Figure 3. (a) Steady-state CO oxidation rate over 4 wt % Cu/CeO2 catalyst as a function of temperature. (b) Steady-state CO oxidation rate over 4 wt
% Cu/CeO2 catalyst as a function of CO and oxygen concentrations. (c) Evolution of CO conversion and Cu+ concentration during a repeated
temporal oxygen cutoff experiment at 60 °C involving cycling between 1% CO (120 s) and 1% CO and 4% oxygen (60 s) after equilibration in 1% CO
and 4% oxygen. Gray fields correspond to 1% CO and 4% oxygen; white fields correspond to 1% CO. (d) Comparison of CO2 formation rates during
the tenth cycle of oxygen cutoff experiments and under steady-state conditions.

beamline with vanadium (K edge at 5465 eV) and copper (K component and linear combination analyses were done using
edge at 8979 eV) foils, respectively. The size of the X-ray beam the Fitit software.34,35
on the sample was about 1.5 mm in the horizontal direction and To compare the local structure of copper in the Cu-CeO2
0.5 mm in the vertical direction with a total flux of about (3−7) catalyst to that of the reference samples, we pretreated the
× 1011 ph/s. catalyst under specific conditions, cooled it down to 39−40 °C
Cu foil, CuO, Cu2O, and Cu(NO3)2·6H2O standards were in the same gas atmospheres and measured in situ the extended
measured in transmission mode. Due to the strong X-ray X-ray absorption fine structure (EXAFS) spectra at the Cu K
absorption in the 4 wt % Cu-CeO2 catalyst, in situ/operando edge. We analyzed the EXAFS spectra using the Demeter
experiments were performed in fluorescence mode (Figure 1). software package36 and fitted the Fourier-transformed k3-
The fluorescence signal was detected with a five-element silicon weighted signal for k = 3−11 Å−1 with dk = 1 and R = 1−3 Å
drift detector (SGX) and processed by XIA electronics. We used with dR = 0.5.
titanium- and nickel-containing filters to suppress the elastic
scattering while measuring the Ce Lα and Cu Kα fluorescence 3. RESULTS AND DISCUSSION
signals, respectively. To measure XAS with 0.5 s time resolution, 3.1. State of Copper and Catalytic Activity. Figure 2a
we used the DXP-XMAP mapping mode of the XIA data compares the Cu K edge XANES spectra of the Cu-CeO2
acquisition system described in detail in our previous work.33 catalyst calcined at 300 °C in 4% oxygen to those of various
For normalization of the fluorescence signal to the intensity of Cu2+, Cu+, and Cu0 standards. The spectrum of the calcined
the incoming beam, we used an ionization chamber placed catalyst closely resembles that of CuO, which has Cu2+ sites in
before the sample. For quantification of the Ce3+ concentration square-planar oxygen coordination. By comparison, the Cu-
we used an additional reference sample (1.5 wt % Pt/CeO2 (NO3)2·6H2O standard containing Cu2+ cations in the
catalyst described in detail elsewhere30). To compare the distorted-octahedral environment of [Cu(H2O)6]2+ demon-
reactivities of cerium and copper, we performed similar in situ/ strates the absence of a shoulder at ca. 8985 eV (corresponding
operando experiments at the Ce L3 and Cu K edges. Principal to the dipole-allowed 1s to 4pz electronic transition) and has a
4695 [Link]
ACS Catal. 2020, 10, 4692−4701
ACS Catalysis [Link]/acscatalysis Research Article

much higher white line peak at ca. 8996 eV. The absence of a Figures 3a,b shows the results of kinetic characterization of
peak at ca. 9016 eV for the 4 wt % Cu-CeO2 catalyst is typical of the Cu-CeO2 under steady-state CO oxidation conditions. The
highly dispersed Cu2+ sites on the surface of ceria crystal- apparent activation energy measured in the temperature range of
lites.26,37 Fitting of the Cu K-edge EXAFS spectrum of the 39−75 °C is close to 62 kJ/mol (Figure 3a). The CO conversion
catalyst (Figure 2c and Figure S2 and Table S1) indicates of on increases as a function of CO concentration, and the reaction
average ca. 3.9 oxygen neighbors around copper at a distance of order with respect to CO measured at 64 °C in 4% of oxygen and
1.929 Å. This distance is slightly shorter in comparison to the 0.25−4% CO is close to 0.65 (Figure 3b). Changes in the oxygen
Cu−O distances reported for CuO (1.95−1.96 Å) and similar to concentration only weakly affect the CO conversion, while the
those reported for highly dispersed copper on ceria.11,38,37,39 reaction order with respect to oxygen at 64 °C in 1% CO and 3−
The second coordination shell around copper (Cu−O−Cu) can 16% oxygen is close to −0.2. These results suggest that, during
be fitted by 0.7 copper neighbor at ca. 2.87 Å (Table S1). This CO oxidation, the active sites are only partially covered by CO,
highly dispersed nature of copper oxo species on the ceria and this correlates with the relatively weak CO adsorption on an
surface is in agreement with STEM images of the catalyst before oxidized surface containing mainly Cu2+. A slightly negative
and after operando experiments (Figure S3). Previously, EPR reaction order with respect to oxygen indicates that oxygen
data suggested the predominant presence of Cu2+ dimers in this mildly poisons the sites for CO adsorption. Overall, the activity
catalyst12 that correlate well with the small number of copper and kinetic parameters obtained in our study are similar to those
neighbors in the second coordination shell of copper as reported in the literature for highly dispersed copper on ceria
estimated by EXAFS analysis. These results are furthermore in under similar reaction conditions.5,21,22,38
agreement with earlier works suggesting stabilization of highly To uncover the involvement of the redox activity of copper
dispersed Cu2+ species on the ceria surface under oxidizing and cerium in the mechanism of CO oxidation, we wanted to
conditions as described by the wetting phenomenon11 that was correlate the rates of CO2 formation to the rates of oxidation and
confirmed theoretically.18,20 reduction of copper and cerium atoms in the Cu-CeO2 catalyst.
We also analyzed the Cu K edge XAS spectra of the Cu-CeO2 Therefore, we performed oxygen cutoff experiments30 by first
catalyst upon exposure to CO at 80 °C (Figure 2a,c). Fitting of exposing the catalyst to 1% CO and 4% oxygen and then
the Cu K edge EXAFS spectrum of the Cu-CeO2 catalyst switching off the oxygen supply, leaving the catalyst in 1% CO.
exposed to 1% CO at 80 °C and cooled to 39 °C under the same Doing so allows measuring the rates of formation and decay of
atmosphere (Figure S2 and Table S1) did not detect any Cu− copper and cerium intermediates. We repeated the experiments
Cu coordination corresponding to metallic copper. Such a several times and at several temperatures below 90 °C. The
neighboring shell at ca. 2.5 Å was observed in the literature for experiments are given in Table S2. In the next sections we will
discuss (i) the catalytic activity of the Cu-CeO2 during oxygen
Cu-CeO2 only after exposure to CO and hydrogen above 200
cutoff experiments, (ii) time-resolved chemical speciation of
°C.3,11,38 Interestingly, also the Cu−Cu scattering path detected
copper and cerium, and finally (iii) the low-temperature CO
for the oxidized catalyst disappears upon exposure to CO. We
oxidation mechanism that can be formulated on the basis of
can suggest that removal of oxygen from the Cu−O−Cu
these experiments.
ensemble increases the static disorder because copper atoms get
3.2. Catalytic Activity during Oxygen Cutoff Experi-
an additional degree of freedom. This effect, however, can also ments. Figure 3c shows the evolution of CO conversion over
be related to the k range of our EXAFS data being too short (up the Cu-CeO2 catalyst at 60 °C during six periodic oxygen cutoff
to 11 Å−1), which is due to the general difficulties in measuring cycles, which we performed after equilibration of the catalyst in a
high-quality EXAFS data for a low concentration of the element 1% CO and 4% oxygen gas mixture for ca. 2100 s. The changes in
of interest in a highly absorbing matrix, such as CeO2. The the Cu+ concentration measured simultaneously using operando
number of oxygen neighbors around copper in the Cu-CeO2 Cu K edge XANES are also shown (vide infra). The first cycle in
catalyst upon exposure to CO at 80 °C decreased to 2.5 without Figure 3c differs from the others, but afterward both CO
a significant change in the Cu−O bond length (1.94 Å). Such a conversion and Cu+ concentration evolve reproducibly. For the
coordination suggests a linear (2-fold) coordination of Cu+ first and the tenth oxygen cutoff cycles, we evaluated two
similar to that in Cu2O, except that in Cu2O the length of the parameters: (i) the rate of CO2 formation in the 1% CO and 4%
Cu−O bonds is significantly shorter (1.85 Å). The Cu K edge oxygen flow just before switching off the oxygen supply and (ii)
XANES spectrum of the catalyst acquired in 1% CO at 80 °C the initial CO2 formation rate in the absence of oxygen. The
(Figure 2a) resembles that of Cu2O but has a smaller shoulder at results are given in Table S2. To estimate the initial reaction rate
the edge (at 8981 eV) corresponding to the 1s to 4pz transition. in 1% CO, we integrated the corresponding CO2 formation rate
The intensity of this shoulder increases for linearly coordinated during 120 s exposure in 1% CO, fitted the resulting curve with
Cu+ sites due to a large contribution of the copper pz orbital an exponential decay function, and determined the first
located perpendicularly to the O−Cu−O bond. Figure 2a derivative at the initial condition right after the switch (Figure
further shows that in the spectrum of a Cu+ complex with S4). Figure 3d compares these rates (evaluated from the tenth
phenanthroline ([Cu(phen)2]+),40 where copper is coordinated cycle) to the steady-state CO2 formation rate for the same
by four nitrogen ligands in a tetrahedral geometry, the catalyst, already shown in Figure 3a. The initial rate of CO2
aforementioned shoulder completely disappears. This happens formation in 1% CO is only about 1.5 times higher than the
since the pz orbital is mixed with py and px orbitals and not steady-state CO2 formation rate and shows a similar apparent
separated energetically from the other p orbital manifolds. Thus, activation energy (ca. 62−65 kJ/mol). This suggests that CO
one may reasonably conclude that copper sites in the Cu-CeO2 oxidation involves a strongly bound oxygen intermediate, which
catalyst exposed to 1% CO at 80 °C are mainly in the Cu+ state also oxidizes CO in the absence of oxygen in the gas phase. At
with a nonideal linear oxygen coordination. At this point, the the same time, during oxygen cutoff cycles we observed that the
presence of minor quantities of Cu2+ and Cu0 cannot be CO2 formation rate in the presence of oxygen is systematically
excluded. higher than that under the steady-state conditions and has a
4696 [Link]
ACS Catal. 2020, 10, 4692−4701
ACS Catalysis [Link]/acscatalysis Research Article

Figure 4. (a) Evolution of Cu K edge XANES for 4 wt % Cu/CeO2 catalyst during periodic oxygen-cutoff experiments, where at each energy point the
catalyst was exposed to 1% CO (120 s) and then to 1% CO and 4% oxygen (60 s). (b) Comparison of Cu K edge XANES spectra for 4 wt % Cu/CeO2
catalyst in A1 and A2 states with the components 1 and 2 obtained by the PCA analysis of the time-resolved spectra. (c) Evolution of Cu+ concentration
in 4 wt % Cu/CeO2 catalyst during oxygen cutoff experiments at 39, 60, and 81 °C (during the tenth cycle). Solid lines show the fits of Cu+
concentration increase in 1% CO by single exponential decay functions and dashed lines indicate the initial Cu+ formation rates at the moment of
oxygen cutoff. (d) Evolution of Ce3+ concentration in 4 wt % Cu/CeO2 catalyst during oxygen cutoff experiments at 66 and 83 °C (average of 19
cycles). Solid lines show the fits of Ce3+ concentration increase in 1% CO by single-exponential decay functions, and dashed lines indicate the initial
Ce3+ formation rates at the moment of oxygen cutoff.

higher apparent activation energy of 84 kJ/mol (Figure 3d). (Figure 3b). At the same time, CO adsorbs better on reduced
This is related to an activity spike appearing after pre-exposure of Cu+. Thus, a higher concentration of reduced copper could
the catalyst to the reducing atmosphere of CO. The spike was enhance CO adsorption and accelerate the catalytic process
most prominent at 80 °C (Figure S5) but gradually decreased at involving a strongly bound oxygen intermediate. However, we
60 °C (Figure S6) and almost disappeared at 39 °C (Figure S7). observed from the oxygen cutoff experiments that the rate of CO
Moreover, during the oxygen cutoff cycles at 80 °C, we observed oxidation increases much more strongly in the presence than in
an increase in the catalyst bed temperature up to 86 °C upon the absence of oxygen (Figure 3d). Thus, an additional CO
switching on oxygen, which can partially explain the origin of the oxidation mechanism temporarily appearing after exposure of
activity spike (Figure S5). Below 60 °C, the catalyst temperature the catalyst in CO may involve rather weakly bound oxygen
was constant but its activity was still increasing after exposure in intermediates that are present on the surface in equilibrium with
CO and then slowly decaying (Figure S8). Indications for an the gas-phase oxygen and disappear when the oxygen supply is
increased CO2 formation rate after exposure of the Cu-CeO2 cut off. For this mechanistic study, it is important that during
catalysts to CO can be found in the literature,17,23,41−43 where oxygen-cutoff experiments the measurements of the initial rate
the release of carbonates and oxidation of carbonyls stored on of CO oxidation in the absence of oxygen and its apparent
the catalyst surface during pretreatment in CO can explain this activation energy are similar to those observed for steady-state
effect. CO oxidation over the preoxidized Cu-CeO2 catalyst.
We do not rule out, at this juncture, the possibility of an 3.3. Time-Resolved Chemical Speciation of Copper
additional catalytic CO oxidation mechanism appearing in the and Cerium. To quantify changes in the copper state in Cu-
catalyst upon the exposure in CO. It is known that CO is not CeO2 catalyst during oxygen-cutoff experiments as probed by
bound well to Cu2+; therefore, during CO oxidation on a the operando time-resolved Cu K edge XANES, we used the
preoxidized Cu-CeO2 catalyst the reaction order in CO is high corresponding spectra of the catalyst measured under steady-
4697 [Link]
ACS Catal. 2020, 10, 4692−4701
ACS Catalysis [Link]/acscatalysis Research Article

Figure 5. (a) Comparison of the initial rates of CO2, Cu+, and Ce3+ formation in 1% CO during oxygen-cutoff experiments at different temperatures.
(b) Scheme explaining the redox activity of active sites during low-temperature CO oxidation on copper-ceria catalysts. (c) Evolution of absolute
amounts of Cu+ and Ce3+ in 4 wt % Cu/CeO2 catalyst during oxygen-cutoff experiments at similar temperatures. (d) Evolution of Cu+ and Ce3+
amounts in 4 wt % Cu/CeO2 catalyst on relative scales.

state conditions, between the time-resolved experiments. Figure performed the majority of the time-resolved Cu K edge
2b shows such Cu K-edge XANES spectra of the Cu-CeO2 experiments at a fixed incident energy of 8981 eV, as it was
catalyst in the most reduced state (A1 reference) and the most sufficient for chemical speciation in a two-component system
oxidized state (A2 reference) used for quantitative analysis. The containing Cu2+ and Cu+. The difference between the Cu K-
A1 reference XANES spectrum was measured at 80 °C in 1% edge XANES spectra of the A1 and A2 references indicates that
CO and the A2 reference at 39 °C in a flow of 1% CO and 4% 8981 eV is the optimal energy for quantification of Cu+ and Cu2+
oxygen. The Cu K edge XANES of A2 is very similar to that for concentrations in this system. Further details about quantifica-
the catalyst after calcination in 4% oxygen at 300 °C (Figure S9). tion of the Cu+ concentration are given in the Supporting
As verified by XANES and EXAFS analyses (Figure 2 and Figure Information.
S2 and Table S1), the precalcined catalyst comprises mainly of We applied a similar approach for time-resolved speciation of
Cu2+ species. Therefore, we assume that the A2 reference also cerium using Ce L3 edge XANES.33 However, the concentration
corresponds to Cu2+. We have also established above that the of Ce3+ in the Cu-CeO2 catalyst under static conditions at the
catalyst exposed to 1% CO at 80 °C (A1 reference) contains studied temperature range was too low to use these data for
mainly Cu+ species (Figure 2). This correlates to the literature quantitative analysis. Therefore, as references, we used the Ce
suggesting that formation of Cu0 in CO is not very probable for L3-edge XANES spectra of a 1.5 wt % Pt/CeO2 catalyst30,44
highly dispersed copper-ceria catalysts below 90 °C.26,38,37 To measured previously in the same cell under static conditions at
confirm that the A1 reference does not contain significant 150 °C in 1% CO (B1 reference containing 11% of Ce3+) and in
amounts of Cu0, we also analyzed the time-resolved Cu K edge 4% oxygen (B2 reference containing no Ce3+) shown in Figure
XANES spectra of the Cu-CeO2 catalyst during periodic oxygen- 2d. Further details on quantification of the Ce3+ concentration
cutoff cycling at 81 °C by principal component analysis (PCA). are given in the Supporting Information. Considering the
Figure 4a shows the resulting 360 XANES Cu K edge spectra complexity of this analysis involving self-absorption correction,
collected with 0.5 s time resolution. PCA revealed only two use of standards other than the catalyst itself, and averaging of
components, which perfectly match the spectra and the A1 and XAS signals obtained during several gas switching cycles, we
A2 references (Figure 4b and Figure S10). Thus, we finally expect that the error bar on Ce3+ concentration can be as high as
assigned the A1 reference to the Cu+ state. Consequently, we 100%. This would result in uncertainties of about a factor of 2 for
4698 [Link]
ACS Catal. 2020, 10, 4692−4701
ACS Catalysis [Link]/acscatalysis Research Article

Ce3+ formation rates but would not affect the value of the the above observation based on the initial rates. Importantly,
corresponding apparent activation energy. when they are depicted on the relative scales, the kinetics of the
3.4. Mechanism of Low-Temperature CO Oxidation redox transformations of both copper and cerium (Figure 5d)
over Cu-CeO2. Figure 4c compares the evolution of the Cu+ demonstrate excellent agreement. In other words, copper and
concentration in the Cu-CeO2 catalyst during oxygen cutoff cerium should be chemically interacting, forming the same
experiments at different temperatures during the tenth cycle active site. To the best of our knowledge, this work is the first to
(the results of the first cycle are shown in Figure S13). Figure 4d demonstrate experimentally (in a quantitative manner) the
summarizes the evolution of the Ce3+ concentration under concerted interactions of copper and cerium in the activation of
similar conditions. Both Cu+ and Ce3+ concentrations increase oxygen and its transfer to carbon monoxide (Figure 5b).
after oxygen cutoff to 1% CO and decrease again in 1% CO and
4% oxygen. The initial rates of Cu+ and Ce3+ formation in CO 4. CONCLUSIONS
are slower than their decay in CO and oxygen. This agrees with Time-resolved X-ray absorption spectroscopy provides unique
the kinetics experiments indicating that, under steady-state insight into the mechanism of oxygen activation on highly
conditions, the active sites are largely covered by oxygen, leading dispersed copper-ceria catalysts. A specially designed non-
to a low order of reaction with respect to oxygen. Some long- steady-state experimental strategy combined with time-resolved
lived Cu+ and Ce3+ species formed in CO tend to decay slowly in XAS allowed comparing the reactivity of copper and cerium
the presence of oxygen and their concentration is dependent on atoms during the CO oxidation cycle. Oxygen activation
the temperature and pretreatment history (Table S2); they do mediated by a strongly bound oxygen intermediate involves
not play any role in the activation of strongly bound oxygen the redox activity of Cu2+/Cu+ and Ce4+/Ce3+ couples. Cu2+ and
intermediates. Ce4+ reduction rates are kinetically coupled to the rate-
To verify the involvement of Cu2+/Cu+ and Ce4+/Ce3+ redox determining step of CO oxidation. The extent of participation
couples in the mechanism of low-temperature CO oxidation, we of the Ce4+/Ce3+ couple in CO oxidation is about 10 times lower
quantified the initial rates of Cu+ and Ce3+ formation after than that of Cu2+/Cu+. However, both metals react in concert,
oxygen cutoff. We fitted the Cu+ and Ce3+ concentration profiles likely being a part of the same active site. Long-lived Cu+ and
during 120 s in 1% CO by an exponential decay function (Figure Ce3+ species also appear during the CO oxidation process. While
4c,d) and assigned the first derivatives of the resulting curves at their origin and reactivity require clarification, we prove that
the initial moment to the initial rates of Cu+ and Ce3+ formation, they do not activate oxygen to form a strongly bound oxygen
respectively.30 Figure 5a compares the resulting initial CO2, Cu+, intermediate.

■
and Ce3+ formation rates in CO. The Cu+ formation rate is ca.
1.7 times higher than the CO2 formation rate, while the Ce3+ ASSOCIATED CONTENT
formation rate is ca. 5.6 times lower than the CO2 formation rate.
Even when 2-fold uncertainties for Ce3+ formation rates are
*
sı Supporting Information

considered, Ce3+ forms much more slowly than CO2. The The Supporting Information is available free of charge at
stoichiometry of CO oxidation involving lattice oxygen (O2−) [Link]
implies that two electrons should be formed per one CO2 Analysis methods of time-resolved XAS data, results of
molecule: EXAFS data analysis, concentrations of Cu+ and Ce3+ and
activity data and corresponding rates obtained from time-
O2 − + CO = CO2 + 2e (1) resolved XAS experiments, kinetics of gas exchange in the
To fulfill the electron balance, these electrons should be experimental cell, EXAFS fits at the Cu K edge, TEM
transferred either on copper images of 4 wt % Cu-CeO2 catalyst, evolution of CO2
concentration in 1% CO after oxygen cutoff, evolution of
Cu 2 + + e = Cu+ (2) CO conversion and changes in the catalyst bed temper-
or on cerium ature during oxygen cutoff experiments, long-term decay
of CO2 conversion after switching on the oxygen supply,
Ce 4 + + e = Ce3 + (3) comparison of Cu K edge XANES spectra of the catalyst
A comparison of the initial rates (Figure 5a) suggests that the in the oxidized state, results of PCA analysis on Cu K edge
total number of electrons participating in reaction 1 is about 1.9 XANES, linear correlation between Cu K-edge XANES
times higher than the number of CO2 molecules, matching the and Cu+ concentration, Ce L3 XANES spectra of 4 wt %
stoichiometry of this reaction. About 91% of electrons were Cu-CeO2 catalyst with those of the reference samples, and
derived from the reduction of Cu2+ to Cu+, while only 9% of evolution of Cu+ concentration in 4 wt % Cu-CeO2
them were derived from the reduction of Ce4+ to Ce3+. Note that catalyst during the first cycle of an oxygen-cutoff
the latter process does not necessarily point to the direct experiment at different temperatures (PDF)
oxidation of CO by the Ce4+/Ce3+ couple, which would not take
place in CO at such low temperatures.45 Rather, Ce3+ species
appear as the result of electron exchange between copper and
■ AUTHOR INFORMATION
Corresponding Author
cerium redox couples or/and lattice oxygen mobility between Olga V. Safonova − Paul Scherrer Institute, 5232 Villigen PSI,
the neighboring sites schematically depicted in Figure 5b Switzerland; [Link]/0000-0002-6772-1414;
describing the mechanism. The effect is further verified in Figure Email: [Link]@[Link]
5c,d by comparing the kinetics of Cu+ and Ce3+ formation and
decay during oxygen cutoff experiments performed at similar Authors
temperatures on absolute and relative scales. On the absolute Alexander Guda − The Smart Materials Research Institute,
scale (Figure 5c), the more prominent increase in Cu+ Southern Federal University, Rostov-on-Don 344090, Russian
concentration in comparison with that of Ce3+ clearly confirms Federation
4699 [Link]
ACS Catal. 2020, 10, 4692−4701
ACS Catalysis [Link]/acscatalysis Research Article

Yury Rusalev − The Smart Materials Research Institute, Southern (5) Liu, W.; Flytzanistephanopoulos, M. Total Oxidation of Carbon-
Federal University, Rostov-on-Don 344090, Russian Federation Monoxide and Methane over Transition Metal Fluorite Oxide
René Kopelent − Paul Scherrer Institute, 5232 Villigen PSI, Composite Catalysts: II. Catalyst Characterization and Reaction-
Switzerland Kinetics. J. Catal. 1995, 153 (2), 317−332.
Grigory Smolentsev − Paul Scherrer Institute, 5232 Villigen PSI, (6) Kašpar, J.; Fornasiero, P.; Hickey, N. Automotive Catalytic
Converters: Current Status and Some Perspectives. Catal. Today 2003,
Switzerland
77 (4), 419−449.
Wey Yang Teoh − School of Chemical Engineering, The (7) Sedmak, G.; Hocevar, S.; Levec, J. Kinetics of Selective CO
University of New South Wales, Sydney, New South Wales 2052, Oxidation in Excess of H-2 over the Nanostructured Cu0.1Ce0.9O2-
Australia; [Link]/0000-0002-4400-4578 (y) Catalyst. J. Catal. 2003, 213 (2), 135−150.
Jeroen A. van Bokhoven − Paul Scherrer Institute, 5232 Villigen (8) Nagai, Y.; Dohmae, K.; Nishimura, Y. F.; Kato, H.; Hirata, H.;
PSI, Switzerland; Institute for Chemistry and Bioengineering, Takahashi, N. Operando XAFS Study of Catalytic NO Reduction over
ETH Zurich, 8093 Zürich, Switzerland; [Link]/0000- Cu/CeO2: The Effect of Copper−Ceria Interaction under Periodic
0002-4166-2284 Operation. Phys. Chem. Chem. Phys. 2013, 15 (22), 8461−8465.
Maarten Nachtegaal − Paul Scherrer Institute, 5232 Villigen PSI, (9) Zabilskiy, M.; Djinović, P.; Tchernychova, E.; Tkachenko, O. P.;
Switzerland; [Link]/0000-0003-1895-9626 Kustov, L. M.; Pintar, A. Nanoshaped CuO/CeO2Materials: Effect of
the Exposed Ceria Surfaces on Catalytic Activity in N2O Decom-
Complete contact information is available at: position Reaction. ACS Catal. 2015, 5 (9), 5357−5365.
[Link] (10) Zhang, R.; Teoh, W. Y.; Amal, R.; Chen, B.; Kaliaguine, S.
Catalytic Reduction of NO by CO over Cu/CexZr1−xO2 Prepared by
Author Contributions Flame Synthesis. J. Catal. 2010, 272 (2), 210−219.
W.Y.T. prepared and characterized the catalyst. R.K. and Y.R. (11) Tschöpe, A.; Markmann, J.; Zimmer, P.; Birringer, R. Chadwick.
performed the ex situ catalytic tests. A.G., O.V.S., and R.K. N2O Temperature-Programmed Oxidation and EXAFS Studies on the
performed in situ/operando XAS experiments and analyzed the Dispersion of Copper in Ceria-Supported Nanocatalysts. Chem. Mater.
results. O.V.S. and A.G. designed the experiments. G.S. 2005, 17 (15), 3935−3943.
contributed to the development of the fluorescence detected (12) Kydd, R.; Teoh, W. Y.; Wong, K.; Wang, Y.; Scott, J.; Zeng, Q.-
time-resolved XAS method. O.V.S. supervised the project. All H.; Yu, A.-B.; Zou, J.; Amal, R. Flame-Synthesized Ceria-Supported
Copper Dimers for Preferential Oxidation of CO. Adv. Funct. Mater.
authors were involved in the discussion of the results and writing
2009, 19 (3), 369−377.
of the manuscript. All authors have given approval to the final (13) Martínez-Arias, A.; Fernández-García, M.; Gálvez, O.; Coronado,
version of the manuscript. J. M.; Anderson, J. A.; Conesa, J. C.; Soria, J.; Munuera, G. Comparative
Notes Study on Redox Properties and Catalytic Behavior for CO Oxidation of
The authors declare no competing financial interest. CuO/CeO2 and CuO/ZrCeO4 Catalysts. J. Catal. 2000, 195 (1),

■
207−216.
ACKNOWLEDGMENTS (14) Harrison, P. G.; Ball, I. K.; Azelee, W.; Daniell, W.; Goldfarb, D.
Nature and Surface Redox Properties of Copper(II)-Promoted
We thank Dr. Frank Krumeich from ETH Zurich for detailed Cerium(IV) Oxide CO-Oxidation Catalysts. Chem. Mater. 2000, 12,
STEM analysis of the catalyst and help with the interpretation of 3715.
the results. We thank Dr. Davide Ferri, Dr. Vitaly Sushkevich, (15) Gamarra, D.; Fernández-García, M.; Belver, C.; Martínez-Arias,
and Dr. Maxim Zabilskiy (Paul Scherrer Institute) for fruitful A. Operando DRIFTS and XANES Study of Deactivating Effect of CO2
discussions. We thank J. Imbao and D. Kuzmenko (Paul on a Ce0.8Cu0.2O2 CO-PROX Catalyst. J. Phys. J. Phys. Chem. C 2010,
Scherrer Institute) for help during beamtimes. We thank the 114 (43), 18576−18582.
Swiss Light Source for granting beamtime at the SuperXAS (16) Wang, F.; Büchel, R.; Savitsky, A.; Zalibera, M.; Widmann, D.;
beamline. We gratefully acknowledge financial support for Pratsinis, S. E.; Lubitz, W.; Schüth, F. In Situ EPR Study of the Redox
O.V.S. and R.K. from the Swiss National Science Foundation Properties of CuO−CeO2 Catalysts for Preferential CO Oxidation
(PROX). ACS Catal. 2016, 6 (6), 3520−3530.
Grant No. 200021_140750. Y.R. acknowledges the Russian (17) Kydd, R.; Ferri, D.; Hug, P.; Scott, J.; Teoh, W. Y.; Amal, R.
Science Foundation project No. 17-72-10245 for travel support. Temperature-Induced Evolution of Reaction Sites and Mechanisms
A.G. acknowledges the President’s Grant of Russian Federation during Preferential Oxidation of CO. J. Catal. 2011, 277 (1), 64−71.
for young scientists MK-2730.2019.2. (18) Elias, J. S.; Artrith, N.; Bugnet, M.; Giordano, L.; Botton, G. A.;

■ REFERENCES
(1) Konsolakis, M. The Role of Copper−Ceria Interactions in
Kolpak, A. M.; Shao-Horn, Y. Elucidating the Nature of the Active
Phase in Copper/Ceria Catalysts for CO Oxidation. ACS Catal. 2016, 6
(3), 1675−1679.
Catalysis Science: Recent Theoretical and Experimental Advances. (19) Bu, Y.; Er, S.; Niemantsverdriet, J. W. H.; Fredriksson, H. O.A.
Appl. Catal., B 2016, 198, 49−66. Preferential Oxidation of CO in H2 on Cu and Cu/CeOx Catalysts
(2) Li, Y.; Fu, Q.; Flytzani-Stephanopoulos, M. Low-Temperature Studied by in Situ UV−Vis and Mass Spectrometry and DFT. J. Catal.
Water-Gas Shift Reaction over Cu- and Ni-Loaded Cerium Oxide 2018, 357, 176−187.
Catalysts. Appl. Catal., B 2000, 27 (3), 179−191. (20) Monte, M.; Munuera, G.; Costa, D.; Conesa, J. C.; Martinez-
(3) Chen, A.; Yu, X.; Zhou, Y.; Miao, S.; Li, Y.; Kuld, S.; Sehested, J.; Arias, A. Near-Ambient XPS Characterization of Interfacial Copper
Liu, J.; Aoki, T.; Hong, S.; Camellone, M. F.; Fabris, S.; Ning, J.; Jin, C.; Species in Ceria-Supported Copper Catalysts. Phys. Chem. Chem. Phys.
Yang, C.; Nefedov, A.; Wöll, C.; Wang, Y.; Shen, W. Structure of the 2015, 17 (44), 29995−30004.
Catalytically Active Copper−Ceria Interfacial Perimeter. Nat. Catal. (21) Polster, C. S.; Nair, H.; Baertsch, C. D. Study of Active Sites and
2019, 2 (4), 334. Mechanism Responsible for Highly Selective CO Oxidation in H2 Rich
(4) Graciani, J.; Mudiyanselage, K.; Xu, F.; Baber, A. E.; Evans, J.; Atmospheres on a Mixed Cu and Ce Oxide Catalyst. J. Catal. 2009, 266
Senanayake, S. D.; Stacchiola, D. J.; Liu, P.; Hrbek, J.; Sanz, J. F.; (2), 308−319.
Rodriguez, J. A. Highly Active Copper-Ceria and Copper-Ceria-Titania (22) Moreno, M.; Bergamini, L.; Baronetti, G. T.; Laborde, M. A.;
Catalysts for Methanol Synthesis from CO2. Science 2014, 345 (6196), Mariño, F. J. Mechanism of CO Oxidation over CuO/CeO2 Catalysts.
546−550. Int. J. Hydrogen Energy 2010, 35 (11), 5918−5924.

4700 [Link]
ACS Catal. 2020, 10, 4692−4701
ACS Catalysis [Link]/acscatalysis Research Article

(23) Sedmak, G.; Hočevar, S.; Levec, J. Transient Kinetic Model of (39) Wang, X.; Rodriguez, J. A.; Hanson, J. C.; Gamarra, D.; Martínez-
CO Oxidation over a Nanostructured Cu0.1Ce0.9O2−y Catalyst. J. Arias, A.; Fernández-García, M. Unusual Physical and Chemical
Catal. 2004, 222 (1), 87−99. Properties of Cu in Ce1-XCuxO2 Oxides. J. Phys. Chem. B 2005, 109
(24) Baber, A. E.; Yang, X.; Kim, H. Y.; Mudiyanselage, K.; Soldemo, (42), 19595−19603.
M.; Weissenrieder, J.; Senanayake, S. D.; Al-Mahboob, A.; Sadowski, J. (40) Lockard, J. V.; Kabehie, S.; Zink, J. I.; Smolentsev, G.; Soldatov,
T.; Evans, J.; Rodriguez, J. A.; Liu, P.; Hoffmann, F. M.; Chen, J. G.; A.; Chen, L. X. Influence of Ligand Substitution on Excited State
Stacchiola, D. J. Stabilization of Catalytically Active Cu+ Surface Sites Structural Dynamics in Cu(I) Bisphenanthroline Complexes. J. Phys.
on Titanium−Copper Mixed-Oxide Films. Angew. Chem., Int. Ed. 2014, Chem. B 2010, 114 (45), 14521−14527.
53 (21), 5336−5340. (41) Avgouropoulos, G.; Ioannides, T. Adsorption and Reaction of
(25) Yang, X.; Kattel, S.; Xiong, K.; Mudiyanselage, K.; Rykov, S.; CO on CuO−CeO2 Catalysts Prepared by the Combustion Method.
Senanayake, S. D.; Rodriguez, J. A.; Liu, P.; Stacchiola, D. J.; Chen, J. G. Catal. Lett. 2007, 116 (1), 15−22.
Direct Epoxidation of Propylene over Stabilized Cu+ Surface Sites on (42) Avgouropoulos, G.; Ioannides, T. TPD and TPSR Study of CO
Titanium-Modified Cu2O. Angew. Chem., Int. Ed. 2015, 54 (41), Interaction with CuO−CeO2 Catalysts. J. Mol. Catal. A: Chem. 2008,
11946−11951. 296 (1), 47−53.
(26) Gamarra, D.; Belver, C.; Fernández-García, M.; Martínez-Arias, (43) Goguet, A.; Meunier, F. C.; Tibiletti, D.; Breen, J. P.; Burch, R.
A. Selective CO Oxidation in Excess H2 over Copper−Ceria Catalysts: Spectrokinetic Investigation of Reverse Water-Gas-Shift Reaction
Identification of Active Entities/Species. J. Am. Chem. Soc. 2007, 129 Intermediates over a Pt/CeO2 Catalyst. J. Phys. Chem. B 2004, 108
(40), 12064−12065. (52), 20240−20246.
(27) Yao, S.; Mudiyanselage, K.; Xu, W.; Johnston-Peck, A. C.; (44) Safonova, O. V.; Guda, A. A.; Paun, C.; Smolentsev, N.; Abdala,
Hanson, J. C.; Wu, T.; Stacchiola, D.; Rodriguez, J. A.; Zhao, H.; Beyer, P. M.; Smolentsev, G.; Nachtegaal, M.; Szlachetko, J.; Soldatov, M. A.;
K. A.; Chapman, K. W.; Chupas, P. J.; Martínez-Arias, A.; Si, R.; Bolin, Soldatov, A. V.; van Bokhoven, J. A. Electronic and Geometric Structure
T. B.; Liu, W.; Senanayake, S. D. Unraveling the Dynamic Nature of a of Ce3+ Forming Under Reducing Conditions in Shaped Ceria
CuO/CeO2 Catalyst for CO Oxidation in Operando: A Combined Nanoparticles Promoted by Platinum. J. Phys. Chem. C 2014, 118 (4),
Study of XANES (Fluorescence) and DRIFTS. ACS Catal. 2014, 4 (6), 1974−1982.
1650−1661. (45) Luo, M.-F.; Zhong, Y.-J.; Yuan, X.-X.; Zheng, X.-M. TPR and
(28) Elias, J. S.; Stoerzinger, K. A.; Hong, W. T.; Risch, M.; Giordano, TPD Studies of CuOCeO2 Catalysts for Low Temperature CO
L.; Mansour, A. N.; Shao-Horn, Y. In Situ Spectroscopy and Oxidation. Appl. Catal., A 1997, 162 (1), 121−131.
Mechanistic Insights into CO Oxidation on Transition-Metal-
Substituted Ceria Nanoparticles. ACS Catal. 2017, 7 (10), 6843−6857.
(29) Oyama, S. T.; Li, W. Absolute Determination of Reaction
Mechanisms by in Situ Measurements of Reaction Intermediates. Top.
Catal. 1999, 8 (1−2), 75−80.
(30) Kopelent, R.; van Bokhoven, J. A.; Szlachetko, J.; Edebeli, J.;
Paun, C.; Nachtegaal, M.; Safonova, O. V. Catalytically Active and
Spectator Ce3+ in Ceria-Supported Metal Catalysts. Angew. Chem., Int.
Ed. 2015, 54 (30), 8728−8731.
(31) Kopelent, R.; van Bokhoven, J. A.; Nachtegaal, M.; Szlachetko, J.;
Safonova, O. V. X-Ray Emission Spectroscopy: Highly Sensitive
Techniques for Time-Resolved Probing of Cerium Reactivity under
Catalytic Conditions. Phys. Chem. Chem. Phys. 2016, 18 (47), 32486−
32493.
(32) Chiarello, G. L.; Nachtegaal, M.; Marchionni, V.; Quaroni, L.;
Ferri, D. Adding Diffuse Reflectance Infrared Fourier Transform
Spectroscopy Capability to Extended X-Ray-Absorption Fine Structure
in a New Cell to Study Solid Catalysts in Combination with a
Modulation Approach. Rev. Sci. Instrum. 2014, 85 (7), No. 074102.
(33) Guda, A. A.; Bugaev, A. L.; Kopelent, R.; Braglia, L.; Soldatov, A.
V.; Nachtegaal, M.; Safonova, O. V.; Smolentsev, G. Fluorescence-
Detected XAS with Sub-Second Time Resolution Reveals New Details
about the Redox Activity of Pt/CeO2 Catalyst. J. Synchrotron Radiat.
2018, 25 (4), 989.
(34) Smolentsev, G.; Soldatov, A. V. FitIt: New Software to Extract
Structural Information on the Basis of XANES Fitting. Comput. Mater.
Sci. 2007, 39 (3), 569−574.
(35) Smolentsev, G.; Guilera, G.; Tromp, M.; Pascarelli, S.; Soldatov,
A. V. Local Structure of Reaction Intermediates Probed by Time-
Resolved x-Ray Absorption near Edge Structure Spectroscopy. J. Chem.
Phys. 2009, 130 (17), 174508.
(36) Ravel, B.; Newville, M. ATHENA, ARTEMIS, HEPHAESTUS:
Data Analysis for X-Ray Absorption Spectroscopy Using IFEFFIT. J.
Synchrotron Radiat. 2005, 12 (4), 537−541.
(37) Zhang, R.; Miller, J. T.; Baertsch, C. D. Identifying the Active
Redox Oxygen Sites in a Mixed Cu and Ce Oxide Catalyst by in Situ X-
Ray Absorption Spectroscopy and Anaerobic Reactions with CO in
Concentrated H2. J. Catal. 2012, 294, 69−78.
(38) Elias, J. S.; Stoerzinger, K. A.; Hong, W. T.; Risch, M.; Giordano,
L.; Mansour, A. N.; Shao-Horn, Y. In Situ Spectroscopy and
Mechanistic Insights into CO Oxidation on Transition-Metal-
Substituted Ceria Nanoparticles. ACS Catal. 2017, 7, 6843.

4701 [Link]
ACS Catal. 2020, 10, 4692−4701