Research Article

Cite This: ACS Catal. 2019, 9, 2177−2195 [Link]/acscatalysis

Unravelling the Nature of the Active Species as well as the Doping

Effect over Cu/Ce-Based Catalyst for Carbon Monoxide Preferential
Oxidation
Jichang Lu,† Jing Wang,†,‡ Qin Zou,† Dedong He,† Liming Zhang,† Zhizhi Xu,† Sufang He,§
and Yongming Luo*,†
†
Faculty of Environmental Science and Engineering, Kunming University of Science and Technology, Kunming 650500, P. R. China
‡
College of Life Science and Environment, Research Hengyang Normal University, Hengyang 421001, P. R. China
§
Research Center for Analysis and Measurement, Kunming University of Science and Technology, Kunming 650093, P. R. China
See [Link] for options on how to legitimately share published articles.

*
S Supporting Information
Downloaded via MIAMI UNIV on October 22, 2019 at [Link] (UTC).

ABSTRACT: The active sites of a mixed Cu/Ce material and the doping effect of typical element (iron, Fe) on the active
species and the catalytic behavior of Cu/Ce for CO preferential oxidation in rich H2 (CO-PROX) were investigated by in situ
diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS), in situ oxygen storage capacity measurement (OSC)
combined with designed temperature-programmed reduction (TPR), along with Raman, X-ray photoelectron spectroscopy
(XPS), X-ray diffraction (XRD), and temperature-programmed desorption/reduction of CO (CO-TPD/TPR). These results
showed that two kinds of surface active center were involved in the CuCe- and Fe-doped CuCe systems, that is, Cu+ as
adsorption sites for the chemisorption and the activation of CO molecules, the surface reactive oxygen (the highly dispersed
oxygen and surface lattice oxygen) that directly participated in the whole CO oxidation process. The addition of Fe into CuCe
sample resulted in the incorporation of Fe into CeO2 lattice forming Fe−O−Ce structure and generated more oxygen vacancies,
which not only enhanced the interaction between Cu and Ce to form more Cu+ absorption sites but also trapped the gas-phase
oxygen and promoted the release of subsurface lattice oxygen to supply more reactive oxygen. Thus, the turnover frequency
(TOF) value was increased from 3.62 × 10−2 s−1 for CuCe to 4.50 × 10−2 s−1 for Fe-doped CuCe. Moreover, with the
enhancement of the lattice oxygen migration combined with the promotional role of Fe on the water gas shift (WGS), the
capacity of the resistance to CO2 and H2O was enhanced for Fe-doping CuCe, and the corresponding stability time was largely
prolonged from 170 to 400 h, in the coexistence of CO2 and H2O.
KEYWORDS: CO-PROX, CuO-CeO2, active species, promotional role of Fe, stability

1. INTRODUCTION reduced to 10 ppm in order to avoid poisoning of the platinum

In the new energy scenario, with the increasing interest in the anode catalysts.3−5 The preferential oxidation of CO (CO-
development of more clean energy with environmentally PROX) is deemed as a promising and potential solution due to
friendly and high-efficiency ways, proton exchange membrane its economic, practical, and efficient characteristics for
fuel cells (PEMFCs) are considered as a promising source for purifying H2-rich gases.3,6,7
energy in the future because they are efficient, combustionless, In general, two kinds of CO-PROX catalysts are mainly
and virtually pollution free.1 Hydrogen suitable to feed involved (i.e., supported noble metal materials and CuO-
PEMFCs is normally generated by alcohols or hydrocarbons
steam reforming followed by the water gas shift reaction Received: October 7, 2018
(WGSR).2 However, the gas stream obtained still contains a Revised: January 18, 2019
substantial level of CO (about 1%) which must be further Published: January 24, 2019

© 2019 American Chemical Society 2177 DOI: 10.1021/acscatal.8b04035

ACS Catal. 2019, 9, 2177−2195
ACS Catalysis Research Article

CeO2-based catalysts). Noble metal materials are one kind of been extensively paid to the significance of the active oxygen
catalyst with high-performance for CO-PROX, but they suffer species on the reaction process of CO-PROX.26,32
from obstacles such as high price and the limited abundance as More importantly, any adjustment or modification in
well as low selectively at the high temperature.8−10 CuO-CeO2 nanoscale via the doping of different types of additives or
is the other kind of the most researched catalytic system for the the preparation with different methods could evoke the
elimination of CO, which is an alternative to noble metal unexpected variation in the catalytic behavior.33 As a typical
catalytic systems for the best compromise between cost, promoter, Fe oxides have high oxygen storage capacity similar
activity, selectivity, and thermal stability.6,11−14 The high to CeO2. It should be noted that Fe itself also possesses high
catalytic activity of Cu/Ce catalysts is mainly assigned to the redox ability as well as the lower redox potential of Fe2+/Fe3+
synergistic effect of copper and cerium oxides.15−18 It is known (0.77 V) than that of Ce3+/Ce4+ (1.61 V);34 thus, the
that CeO2 as the typical and promising support for various interaction between Fe2+/Fe3+, Cu2+/Cu+, and Ce3+/Ce4+
oxidation reactions has the following outstanding character- makes the changes in the structural and electronic properties
istics: one is the prominent redox ability of Ce4+/Ce3+ that can more complicated. Thus, Fe species is selected as the model
create the oxygen vacancies and induce the changes in the additives to investigate the doping effect on the active center
electron properties of active Cu species, and the other is the over CuCe sample for CO-PROX. Although a large number of
presence of abundant surface oxygen species, which usually studies have focused on the catalytic properties over Cu/Ce
involved in the oxidation reactions. Meanwhile, Cu species catalysts, the in-depth insight into the catalytic behavior for the
presented as Cu2+, Cu+, Cu0 over CeO2 surface possesses modification of Cu/Ce catalysts by iron (Fe) species in the
effective redox ability, and the synergistic interaction between reaction process of CO-PROX (especially the tolerance of
Cu and Ce oxides enables Cuδ+ species to become the active H2O and CO2 as well as the stability) is relatively less. At
adsorption sites for CO-PROX. However, since there are present, for the selective CO oxidation, some authors found
various kind of oxidized Cu species presented in the CuCe- that the incorporation of Fe species into Cu/Ce catalysts
based sample with respect to the above-mentioned features prepared by wet impregnation and single-step citrate method
(e.g., CuOx clusters, bulk CuO, Cu2+ into CeO2 lattice, Cu3+ generated better reducibility to improve the interaction
and Cu-[Ox]-Ce structure), the surface redox properties of between Cu and Ce, which leads to the higher dispersion of
CuCe sample are usually complex. The origin of the catalytic copper, thus having a positive effect on the CO oxidation
activity and the relevant active center over CuCe catalysts for activity.35,36 Other researchers deemed that the introduction of
CO-PROX remains an open question in the reported Fe additives into Cu/Ce catalysts in the certain preparation
literature. Some authors generally considered that the highly procedure promoted the generation of additional amount of
dispersed CuOx species is the catalytically active center for active sites related to CuO because of the electron−ion
both CO-PROX and CO oxidation via the specific pretreat- interactions, which improved the CO-PROX activity.37
ment (acid/base) for removing highly dispersed CuOx or However, the negative CO oxidation activity was reported
eliminating the Cu-[Ox]-Ce structure.19−22 Strongly bound for using CuO-CeO2 catalysts doped with Fe species when the
Cu-[Ox]-Ce layers were proved to promote the catalytic coprecipitated method was applied.38 They deemed that the
activity of CuCe for CO-PROX via providing the supply reversible desorption of absorbed CO was not beneficial to the
role.23−25 Further, the specific Cu ions species was proposed selective CO oxidation via the characterization of in situ
and proved to be the active sites for CO-PROX. The DRIFTS and XPS techniques. Moreover, in the field of soot
researchers found by the DRIFTS spectra with the aid of combustion,39,40 selective catalytic reduction of NH3 with
parallel time-resolved X-ray diffraction (XRD) measurements NO,34 chemical-looping reforming,41 the water gas shift
under operando conditions that interfacial Cu+ species (WGS) reaction,42 Fe species were extensively applied to
stabilized via the interaction between Cu and Ce oxides modify the redox property of pure CeO2 catalysts, and they
exhibited a strong correlation with the CO catalytic perform- found that the electronic inductive effect between Fe3+ and
ance.26,27 Through the characterization of Fourier transformed Ce4+ might enhance the redox properties for various catalytic
extended X-ray absorption fine structure (EXAFS) spectra in activity of above reactions, sometimes involving the generation
the anaerobic process, the reduction of Cu2+ species into Cu1+ of Fe−O−Ce structure when prepared by surfactant-assisted
via the elimination of the first oxygen of Cu2+ was also and coprecipitation method.34,40 In a word, the modification of
demonstrated to make an important contribution to the CO- Fe additive into CuCe sample prepared by the different
PROX reaction.28,29 Furthermore, recent research on the CO methods could largely give rise to the change in the catalytic
oxidation confirmed through X-ray absorption spectroscopy behaviors. Therefore, the investigation of the doping influence
together with density functional calculations that surface- of Fe additive on the structure, the catalytic active sites and the
substituted CuxCe1−xO2‑y phase with the stabilization of Cu3+ catalytic behavior (the tolerance of H2O and CO2 as well as
and Cu2+ into host CeO2 lattice was the active phase and Cu3+ the stability) of Cu/Ce catalyst system as well as the catalytic
was implicated in the mechanism for CO oxidation over Cu/ active species over Fe doping CuCe catalysts for CO-PROX
Ce catalysts.30,31 It is not surprising because the heterogeneous are of great necessity.
catalysis is totally a complex process, even for the relatively In our previous literature, the different ratio of Fe/Cu and
simple reactions system, which results in the fact that the the different Fe precursors were selected to be doped on the
catalytic activity is not only limited by the difference in the CuO-Ce0.8Zr0.2O2 materials for CO-PROX.43 In this work,
adsorption properties of CO over different Cuδ+ species but simplified CuO-CeO2- and Fe-doped CuO-CeO2 materials
also depends on the surface redox chemistry, the active oxygen were prepared by the simple and solvent-free route of
properties, oxygen vacancies, and so on. In view of those, the combustion synthesis. With the aid of designed in situ TPR
understanding of the roles these Cu ions play in the reaction of and the CO pulsing technique as well as in situ DRIFTS, the
CO-PROX (specific catalytic active center) over CuO-CeO2 catalytic active center for CO-PROX over CuCe and CuCe-Fe
catalysts is still in progress. More notably, little attention has were illuminated from the standpoint of the principle of
2178 DOI: 10.1021/acscatal.8b04035
ACS Catal. 2019, 9, 2177−2195
ACS Catalysis Research Article

heterogeneous catalysis, the composition−structure−activity− The detection of CO and CO2 was performed by the online
stability relationships for the introduction of Fe into CuCe detector equipped with a high sensitivity of TCD and a flame
sample were well understood. The long-term stability of about ionization detector (FID) together with a methanation reactor.
400 h was obtained for Fe-doped CuCe sample in the presence To verify the presence of the deposited carbon over samples
of H2O and CO2, which was twice as stability as CuCe sample. surface, the CO pulse was introduced at 140 °C over CuCe
This results can provide insights into understanding the active and CuCe-Fe samples (100 mg) until the amount of CO
sites over doped CuO-CeO2-based materials for CO-PROX reaches the steady state. The samples were denoted as pulsed
and give guidelines for the practical applications. CuCe and pulsed CuCe-Fe, respectively.
Temperature-programmed desorption of CO (CO-TPD)
2. MATERIALS AND METHODS was performed on the same apparatus with TCD detector.
2.1. Catalyst Preparation. CuO-CeO2 catalyst was First, 100 mg of samples (100−120 mesh) was pretreated
synthesized with a solvent-free combustion synthesis (SFCS) under the 20% O2/He at 400 °C for 1 h. After that, the
method. The method was briefly depicted in the following. temperature was cooled to 30 °C, and the flow gas was
First, the calculated amount of copper nitrate hydrate switched to 100% CO (30 mL/min). Afterward, the pretreated
(Cu(NO3)2·3H2O, AR), cerium(III) nitrate hexahydrate samples were saturated using pure CO at 30 °C for 30 min and
(Ce(NO3)3·6H2O, AR), and carbamide (CO(NH2)2, AR) then were purged with He (30 mL/min) for 60 min to remove
were added into an agate mortar without additional water and the physisorbed CO. The CO-TPD experiment was carried out
then ground completely for 10 min, where a transparent and by elevating temperature up from 30 to 300 °C with a heating
homogeneous viscous gel was obtained. Subsequently, the as- rate of 10 °C/min. CO-TPR was carried out by switching the
obtained gel was calcined at 700 °C for 20 min, denoted as carrier gas of He into pure N2 to detect the product of H2 with
CuCe, where the corresponding content of CuO was 10 wt %. TCD detector. Subsequently, 200 mg of samples was exposed
Fe-doped Cu/Ce sample was also prepared by mixing and to 10% CO/N2 at 30 °C for 30 min with a flow rate of 30 mL/
grinding the iron nitrate nonahydrate (Fe(NO3)3·9H2O, AR) min. Then, the temperature was ramped from 30 to 300 °C at
with the above-mentioned chemicals. The Fe/Cu mass ratio a ramp rate of 1 °C/min, and the gas was analyzed every 5 min
was 0.025, and the subsequent procedures were the same as assisted with a six-way gas-sampling valve.
described by CuCe catalyst, denoted as CuCe-Fe. Diffuse reflectance infrared fourier transform spectroscopy
2.2. Conventional Characterizations. XRD patterns (DRIFTS) was conducted on a Nicolet 6700 spectropho-
were performed on a Rigaku D/max-1200 instrument tometer equipped with a mercury cadmium telluride (MCT)
equipped with Cu Kα radiation at 2θ ranges between 10 and detector with a resolution of 4 cm−1. A diffuse reflection
90°. Peak identification was conducted by the software of Jade infrared cell was connected to a vacuum apparatus and was
6. Transmission electron microscopy (TEM) was performed fitted with temperature-controlled parts and a CaF2 window.
by using a Tecnai G2 TF30 S-Twin microscope operated with Prior to the measurements, the samples were first pretreated at
an accelerating voltage of 300 kV. Nitrogen sorption isotherms 300 °C for 30 min under the atmosphere of 10% O2/Ar.
were determined on a V-Sorb 2800P instrument at 77 K. Background spectra were subtracted at the target temperature
Before measurement, the samples were degassed at 200 °C under the high-purity Ar atmosphere by averaging 32 scans.
under vacuum for 3 h. X-ray photoelectron spectroscopy For every measurement, DRIFTS spectra were obtained for
(XPS) was performed on a PHI 5000 VersaProbe II with samples under the feed gas of 1% CO/1% O2/50% H2/He and
nonmonochromatic Al Kα radiation (1486.6 eV) as the then recorded from 30 to 210 °C by every 20 °C. To
excitation X-ray source. The chemical states and the investigate the deactivation reason caused by the addition of
corresponding surface element compositions of all the samples H2O, a small amount of H2O was introduced to the system via
were determined by the Casa XPS software. The charge the reaction gas stream through bubbling, and in situ DRIFTS
calibration was applied by setting the main C 1s peak at 284.6 spectra were performed over CuCe and FeCuCe catalysts from
eV. Raman spectra were detected by using a Via Reflex 90 to 210 °C by every 20 °C.
Confocal Raman Microscope with a laser (514 nm) working at 2.3. Designed Characterizations. Four kinds of designed
5 mV power. The common experiments of hydrogen H2-TPR over CuCe and CuCe-Fe samples were carried out
temperature-programmed reduction (H2-TPR) were generally and classified as follows: (1) The elimination of surface oxygen
carried out in a catalyst bed with a quartz tube reactor species (β peak-TPR). To effectively eliminate α peak
equipped with a thermal conductivity detector (TCD). Before corresponding to the highly dispersed CuOx species and
each measurement, the prepared samples were pretreated in avoid the thermal effect, the ramp rate was decreased from 10
pure O2 at 400 °C for 30 min. Next, 100 mg of the pretreated to 5 °C/min, the samples (100 mg) were first pretreated with
samples was heated in a gas flow (30 cm3/min) of 10% H2 in 10% H2/Ar (v/v, 30 mL/min) at 150 °C for 0, 1, 2, 3, 5 min,
Ar from 50 to 400 °C (800 °C) at a ramp rate of 10 °C/min. respectively, denoted as P150-0, P150-1, P150-2, P150-3, and
Oxygen storage capacity (OSC) of samples was determined P150-5. Then, the gas mixtures were switched into the
by the CO pulsing technique. The microreactor equipped with atmosphere of high-purity He followed by decreasing the
the six-way gas-sampling valve with a measuring ring (50 pretreatment temperature from 150 °C to the room temper-
μmol/L) was used to introduce the quantitative gas sample. ature. These H2-TPR measurement were finally conducted
Before testing, the sample (50 mg, 100−120 mesh) was with the procedure the same as the above-mentioned TPR
pretreated at 550 °C for 60 min in the flow of 20% (v/v)O2/ process. The optimal pretreatment conditions were found at
He with the velocity of 30 mL/min. Then, the samples were 150 °C by 2 min for CuCe (P150-2) and by 1 min for CuCe-
cooled to the desired temperature (150 and 200 °C, Fe sample (P150-1); (2) The in situ characterization of the
respectively) under the flow gas of high-purity helium (He). active oxygen species of above pretreated sample (pretreatment-
Afterward, the pure CO gas sample (99.99%) was introduced OSC). CuCe and CuCe-Fe samples were first pretreated using
every 5 min into the microreactor until reaching the plateau. 10% H2/Ar at 150 °C for 2 and 1 min, respectively. Then, the
2179 DOI: 10.1021/acscatal.8b04035
ACS Catal. 2019, 9, 2177−2195
ACS Catalysis Research Article

corresponding temperature was reduced to 30 °C, and the flow All activity tests were conducted when the steady state was
gas was switched to the flow of He to completely remove the arrived at the corresponding temperature. The tests were run
H2/Ar gas. Subsequently, the OSC measurement of pretreated at least six times on each temperature, and the average was
samples was performed at the temperature of 200 °C; (3) The determined. The inlet and outlet streams of various reactions
ex situ characterization of the oxygen migration (Activity-TPR). were determined by using an online gas chromatography
First, the samples (100 mg) were pretreated by 10% H2/Ar to (GC), which was equipped with a TCD detector with different
eliminate α peak as completed in step 1. Then, the normal carrier gas depending on the detected products. CO
catalytic activity tests of pretreated samples were performed conversion (%) and CO2 selectivity (%) were calculated
under the feed gas of 1% CO/1% O2/50% H2/He (80 mL/ according to CO and O2 consumption, as described below:
min) from 60 to 160 °C (P150-R160) and 60 to 220 °C CO conversion (%) = ([CO]in − [CO]out ) ×100/[CO]in
(P150-R220), respectively. Afterward, after the temperature
(1)
was reduced under the high-purity He, the spent CuCe and
CuCe-Fe samples were rapidly transferred to conduct the H2- CO2 selectivity (%)
TPR measurement, denoted as (P150-2-R160, P150-2-R220)
and (P150-1-R160, P150-1-R220), respectively; (4) The in situ = 0.5 × ([CO]in − [CO]out ) × 100/([O2 ]in − [O2 ]out )
characterization of the oxygen migration (OSC-TPR). First, the (2)
samples were pretreated under the flow of 20% O2/He (30 To calculate the turnover frequency (TOF) of samples, the
mL/min) at 550 °C for 60 min. After the flow gas was internal/external diffusion effects could be excluded by making
switched to high-purity He, and the temperature was decreased a Weisz−Prater analysis and a Mears analysis. On the basis of
to 200 °C, the OSC measurement of samples was carried out the assumption of hemispheric model over CuCe-based
at 200 °C where only one CO pulse with a measuring ring of catalysts, TOF values of catalysts for CO-PROX were
50 μmol/L was introduced to consume the surface initial active calculated according to the following equations:44,45
oxygen. Then, the temperature was decreased to 50 °C under
the atmosphere of high-purity He. Subsequently, the flow gas TOF (s−1) = FCO·xCO·mcat−1/(DCuO·nCuO) (3)
was switched to the 10% H2/Ar, and the in situ TPR
where mcat is the weight of catalyst (g), FCO is the flow velocity
experiments were conducted from 50 to 400 °C with a
of feed gas (mol s−1), χCO is the CO conversion, which was
ramping rate of 10 °C/min.
kept below 10% at 80 °C by increasing the flow rate assisted by
2.4. Catalytic Activity Tests. The selective oxidation of
decreasing the amount of catalysts, nCuO is the total molar
CO in H2-rich gas was conducted on a tubular reactor at the amount of CuO (mol), and DCuO is the dispersion of CuO,
temperature between 60 and 220 °C with the atmospheric which was calculated on the basis of the ratio of Cu/Ce from
pressure. First, 300 mg of catalyst (sieve fraction, 40−60 mesh) XPS results.
was placed in a quartz tubular reactor. The synthetic gas
(containing 1% CO, 1% O2, 50% H2 in volume and He as 3. RESULTS AND DISCUSSION
balance gas) was introduced to fix a total flow velocity of 80
mL/min. To investigate the influence of CO2 and H2O on the 3.1. Catalytic Activities of CuCe-Based Materials.
Figure 1 displays the conversion of CO and the selectivity of
catalytic activity and stability, the experiments for the
CO2 during the process of CO-PROX over the CuCe and
resistance to CO2 and H2O over two samples were performed
CuCe-Fe samples. From Figure 1A, CuCe-Fe catalyst exhibits
by introducing 15% CO2 and 10% H2O into the feed gas to
higher catalytic performance than that of CuCe. The reaction
bring the total flow rate to 80 mL/min. The stability tests were
temperatures T50 and T100 (corresponding to 50% and 100% of
conducted at 160 °C for CuCe and 155 °C for FeCuCe with
CO conversion) are introduced as the important criterion to
other conditions remaining the same. Moreover, the evaluate the CO oxidation activity. It can be seen that the
independent WGS reaction tests were performed over CuCe reaction temperature for T50 is 69, 77 °C, and the
and FeCuCe samples at the temperature range of 120−260 °C corresponding temperature for T100 is 99, 120 °C over
using the mixtures of 1% CO, 10% H2O, and pure N2 as CuCe-Fe, CuCe, respectively. The corresponding temperature
balance (80 mL/min). The WGS reactions over pulsed CuCe window widens from 40 °C (120−160 °C) for CuCe to 60 °C
and CuCe-Fe samples (100 mg, the carbon species was (100−160 °C) for CuCe-Fe. Those results indicates that the
formed) were also conducted using the mixtures of 10% H2O doping of Fe into CuCe catalyst promotes the CO oxidation
and pure N2 as balance with the same flow rate. activity, especially for the low temperature activity. The
In order to research the essential role of the different surface turnover frequency (TOF) and the reaction rate are calculated
active oxygen species and the oxygen mobility, two kinds of to estimate the CO oxidation activity, and the results of CuCe-
specific activity tests were also designed. For the first activity Fe and CuCe samples are compiled in Table 1. It is found that
test, fresh CuCe and CuCe-Fe samples (100 mg) were used the incorporation of Fe increases the corresponding TOF value
under the same feed gas composition with the unchanged flow (and the reaction rate) from 3.62 × 10−2 s−1 (1.13 × 10−6 mol·
rate. With respect to the second activity test, 100 mg of fresh g−1·s−1) to 4.50 × 10−2 s−1 (1.60 × 10−6 mol·g−1·s−1),
CuCe and CuCe-Fe samples were first pretreated under the testifying the promotional role of Fe species on the catalytic
flow gas of 10% H2/Ar at 150 °C for 2 and 1 min, respectively, activity of CuCe catalyst for CO-PROX. With further rising the
in order to remove the α peak corresponding to the highly temperature from 160 to 220 °C, the conversion of CO for all
dispersed oxygen in CuOx species. Then, the flow gas was the catalysts declines to the different degree due to the rapid
switched into the atmosphere of high-purity He, and the oxygen consumption rate at the high temperature as well as the
temperature was decreased to 60 °C. Afterward, the mixture occurrence of hydrogen oxidation. The selectivities of CO2 for
gas (1% CO/1% O2/50% H2/He) was introduced to conduct two catalysts are shown in Figure 1B. As seen, with increasing
the other normal activity test. the reaction temperature, the selectivity of CO2 for the
2180 DOI: 10.1021/acscatal.8b04035
ACS Catal. 2019, 9, 2177−2195
ACS Catalysis Research Article

of CuOx are not discovered in the XRD pattern of CuCe-Fe

catalyst. Two reasons could be probably considered to
interpret this phenomenon, one can expect that the doping
of Fe species might result in the incorporation of more CuOx
species into CeO2 lattice forming the solid solution. The other
can expect that the incorporation of Fe species into CeO2
lattice by replacing Cu species is presented, which facilitates
the highly redispersion of more CuOx species over the CeO2
surface, forming the stronger interacting of Cu with Ce.46,47
Compared with the diffraction diagram of CuCe, the (111)
peak shifts to a higher location for the CuCe-Fe (Figure 2,
inset). Meanwhile, the unit cell parameters calculated were
0.54126 for CuCe and 0.54016 nm for CuCe-Fe sample,
respectively. This lattice shrinkage and the shift of the position
of the (111) peak demonstrate the consequence of the
incorporation of iron or more copper species into the CeO2
lattice, due to the smaller Fe (II 0.075 nm; III 0.064 nm) and
Cu(II 0.073 nm) ionic radius compared to that of Ce(IV 0.097
nm).22 This incorporation effect might result in the formation
of the distortion in the cubic fluorite structure of CeO2, thus
generating more defective sites.48 The ratio of Cu/Ce
evaluated by the measurement of XPS (Figure 2) is used to
characterize the distribution state of Cu species over CeO2.
The corresponding result is summarized in Table 1. The Cu/
Ce atomic ratios are calculated to be 0.293 and 0.336 for CuCe
and CuCe-Fe, respectively. The higher value for CuCe-Fe
Figure 1. (a) CO conversion and (b) CO2 selectivity of CuCe and indicates the generation of more amount of surface dispersed
CuCe-Fe. Reaction conditions: mcat = 300 mg, feed gas: 1% CO, 1% Cu oxide species, demonstrating that the addition of Fe species
O2, 50% H2 and balance He, flow rate = 80 mL/min. gives rise to the incorporation of Fe into CeO2 lattice, which
leads to the migration and redistribution of partial Cu oxide
catalysts gradually decreases, probably owing to the possible species from CeO2 bulk lattice to surface and interface (as
competing reaction for converting H2 and O2 to H2O. It can be evidenced by the higher intensity of α and β TPR peaks).
concluded that the addition of Fe has a promotional effect on Moreover, the main reflections (2θ = 28.5, 33.1, 47.5, 56.3,
the catalytic activity of CuCe for CO-PROX. Therefore, the 59.1, 69.4, 76.7, and 79.1°) corresponding to the fluorite-type
following characterizations are used to investigate the promo- cubic structure of CeO2 with diffractions of (111), (200),
tional effect of Fe additive on the structural, textural, and (220), (311), (222), (400), (331), and (420) planes (ICDD
electronic properties of CuCe catalyst for CO-PROX. 34-0394) are observed for both CuCe and CuCe-Fe
3.2. Structural, Textural, and Electronic Properties of samples.49−51 It is seen Figure 2A (insert) that the character-
CuCe and CuCeFe Materials. The structural property of istic peaks of CeO2 over CuCe-Fe are much broader and have
catalyst is studied by X-ray diffraction (XRD). The a lower intensity than those of CuCe, suggesting the formation
corresponding patterns of CuCe and CuCe-Fe samples are of smaller-sized CeO2 crystallites. As seen from Table 1, the
shown in Figure 2A. The main crystallite size and the lattice corresponding crystallite size is calculated to be 9.6 nm for
parameter of CeO2 particles over two catalysts are calculated CuCe-Fe, smaller than that of CuCe (11.0 nm), suggesting the
and listed in Table 1. The small and weak diffraction peaks at addition of Fe has a certain influence on the crystallite size of
35.2 and 38.5° attributed to copper oxides (CuOx) are visible CeO2, due to the fact that the distortion of CeO2 fluorite-type
for the CuCe sample, indicating that large number of CuOx cubic phase caused by Fe additive inhibits the excessive growth
species is aggregated on the surface to be detected by XRD and of particles clusters. The textural properties of the CuCe and
a small portion of CuOx is incorporated into CeO2 lattice (the CuCe-Fe catalysts were listed in Table 1. The specific surface
lower unit cell parameter for CuCe (0.54126 nm) than that for area (SBET) of CuCe-Fe sample is higher than that of CuCe,
pure CeO2 (0.54131 nm)). However, the characteristic peaks suggesting that the modification of Fe causes an increase in the

Table 1. Textural Parameters and Structural Properties of CuCe and CuCe-Fe Catalysts
TOF × lattice crystallite concentration of
10−2 r(CO) × 10−6 parameter sizec SBET Vp oxygen vacanciese Ce3+/(Ce3++Ce4+) OII/(OI + OII)d
samples (s−1)a (mol g−1 s−1) (nm)b (nm) Cu/Ce d
(m2 g−1) (cm3 g−1) Isat/Impd (× 1021 cm−3) (3d5/2)d (%) (%)
CuCe- 4.50 1.60 0.54016 9.6 0.336 19.7 0.042 0.32 4.43 27.56 51.65
Fe
CuCe 3.62 1.13 0.54126 11.0 0.293 12.6 0.044 0.19 3.84 22.00 36.33

a
TOF value was calculated on the basis of the CO conversion at 80 °C. bLattice parameters were obtained by using MDI Jade 5.0 software
according the data of XRD. cCrystallite size determined from XRD by the Scherrer’s equation with the half width of the diffraction peak of the
(111) plane. dValues were estimated from XPS peak fitting data. eThe concentration of oxygen vacancies (N) was calculated by using Raman
spectroscopy.

2181 DOI: 10.1021/acscatal.8b04035

ACS Catal. 2019, 9, 2177−2195
ACS Catalysis Research Article

Figure 2. (A) XRD patterns of CuCe and CuCe-Fe. X-ray photoelectron spectra of CuCe and CuCe-Fe in the Ce 3d region (B), Fe 2p region (C),
and Cu 2p region (D).

surface area of the resultant catalyst (CuCe-Fe), which is due for Fe3+ around 712−713 eV in the literature,53,54 suggesting
to the smaller crystallite size of CuCe-Fe. TEM character- the change in the electron environment around Fe species.
ization is used to analyze the morphological features of CuCe This is possibly due to the fact that the incorporation of Fe
and CuCe-Fe, as shown in Figure S1-A,B, and the particles size species into CeO2 lattice enhances the interaction between Fe
distribution is shown in the corresponding insets. TEM images and Ce leading to the formation of Fe−O−Ce structure.34,39
display the existence of near-spherical, agglomerated particles Furthermore, the intensity ratios of the satellite to main peak
in two samples. The average sizes are calculated to be 9.7 nm of Fe 2p3/2 are calculated to be 1.70 for Fe2+ and 4.18 for Fe3+
for CuCe-Fe and 10.7 nm for CuCe sample, respectively, species, respectively, which is higher than that reported in
consistent with the XRD result. Furthermore, the CuCe-Fe literature (blew 1).55 As inferred from the literature,55,56
sample possesses the relatively narrow and uniform particle- according to Harrison’s relation, the intensity of the satellite
size distribution compared with that of the CuCe sample, peak of iron oxides is proportional to the value of 1−1/R7.0,
demonstrating that the addition of Fe improves the dispersion where R is the interatomic distance of Fe−O bond (Section 2
of CuCe sample. of Supporting Information (SI)). Since the incorporation of Fe
The XPS spectrum is employed to acquire further species into CeO2 lattice leads to the increment in the length
information about the surface composition and the oxida- of Fe−O bond due to the difference of ionic radius between
tion/valence state of Ce, Fe, and CuOx species. Figure 2B,C,D Fe2+ (0.075 nm), Fe3+ (0.064 nm), and Ce3+ (0.097 nm), the
display the Ce 3d, Fe 2p, and Cu 2p signal of CuCe and CuCe- satellite intensities thus significantly increase, which, in turn,
Fe samples, respectively. As shown in Figure 2B, the Ce 3d proves the formation of Fe−O−Ce structure. From Figure 2D,
XPS spectra of CuCe and CuCe-Fe samples are complex and a main and asymmetric Cu 2p3/2 peak at around 930−936 eV
contain four spin−orbit doublets. Three doublets marked as v is found for CuCe sample, which can be decomposed in two
(ca. 882.2 eV) - u (ca. 900.7 eV), v″ (ca. 888.6 eV) - u″ (ca. contributions centered at 932.2 and 933.4 eV, generally
907.6 eV) and v‴ (ca. 898.3 eV) - u‴ (ca. 916.5 eV) are associated with the Cu+ and Cu2+ species, respectively.57−59
attributed to the presence of Ce4+ species. Letters v and u refer This suggests that the coexistence of both Cu2+ and reduced
to the 3d5/2 and 3d3/2 spin−orbit components, respectively. Cu+ species in the CuCe sample where CuOx species is mainly
Meanwhile, for a practical simplification, two components of v′ presented in the form of reduced Cu+ species. The almost
(ca. 884.5 eV) and u′ (ca. 903.0 eV) are ascribed to the inexistence of satellite peaks that is located at 938−946 eV for
existence of Ce3+ species.52 Those results indicates Ce3+ and Cu2+ seems to confirm the existence of main Cu+ species. The
Ce4+ species are simultaneously presented in the CuCe and intensity ratio of the shakeup peak to the principal peak (Isat/
CuCe-Fe samples. For Fe species, as shown in the Figure 2C, Ipp) is used to characterize the valence state of Cu species. It is
the broadened and asymmetric peaks suggests that both Fe2+ generally accepted that the value below 0.57 is corresponding
and Fe3+ coexist in the CuCe-Fe sample. The binding energy at to Cu+ species, and the higher value is assigned to Cu2+
710.0 eV for Fe 2p3/2 sates followed with a satellite around species.22,28 The corresponding values are compiled in Table 1.
716.3 eV is attributed to the presence of Fe2+ species, another CuCe and CuCe-Fe samples display the ratios of 0.19 and
peak around 713.4 eV with a corresponding statellite at 719.4 0.32, proving the existence of Cu+ species. Cu LMM Auger
eV is assigned to the Fe3+ species.53,54 This Fe3+ species is spectra was performed to confirm the presence of reduced Cu
found to be shifted to a high value compared to that reported oxide species, as shown in the inset of Figure 2D. It is seen that
2182 DOI: 10.1021/acscatal.8b04035
ACS Catal. 2019, 9, 2177−2195
ACS Catalysis Research Article

the binding energy at 569.7 eV is attributed typically to the be shifted slightly toward the low frequency at 459 cm−1. This
existence of Cu+ species, followed by two other peaks phenomenon is attributed to the fact that the generation of
representing different Auger transitions involving Cu+, and Fe−O−Ce structure gives rise to the change in the metal−
no distinct signal assigned to Cu0 species at about 568 eV is oxygen bond lengths owing to the difference in the atomic
observed, demonstrating the presence of the main valence state radius between Fe3+ and Ce4+/Ce3+, forming the lattice
of reduced Cu species as Cu+.60,61 The reduced copper species distortions.34,64 The increase of lattice distortions indicates
on the sample may be originated from the strong interaction the formation of more oxygen vacancies. The oxygen vacancies
between CeO2 and the active Cu components,62 because of the concentration is calculated with the nature of a major peak at
existence of the redox balance (Ce3+ + Cu2+ ↔ Ce4+ + Cu1+),32 459 cm−1 by the spatial correlation model,64,68,69 as shown in
as confirmed by redox Ce4+/Ce3+ in the Ce 3p XPS. After the Section 3 of SI. The defect concentration of CuCe and CuCe-
addition of Fe species, the redox properties of CuCe species Fe is calculated to be 3.84 × 1021 cm−3 and 4.43 × 1021 cm−3,
has changed largely. It can be seen that the Cu 2p3/2 peak are respectively (listed in Table 1), suggesting that the addition of
broadened, and more Cu2+ ion are formed, possibly due to the Fe species increases the amount of oxygen vacancies, which
additional redox equilibrium (Fe3+ + Cu+ ↔ Fe2+ + Cu2+). The also proves the Fe doping into CeO2 lattice to generate Fe−
structural and electronic impact of Fe additive on the CuCe O−Ce structure. Moreover, the absence of any Raman band
sample are further demonstrated by Raman and O 1s XPS for Fe species at the low Raman band (217, 296, 398 cm−1)
analysis. also confirms this point.70 The further evidence for Fe doping
Raman spectroscopy is a useful tool to characterize the into CeO2 lattice to form Fe−O−Ce coordination could be
vibration of metal−oxygen bond, the defects of nanophase as achieved by looking at the binding energy of O 1s. Figure 3B
well as their surrounding chemical environment. It has proven presents a comparison of O 1s signal of CuCe and CuCe-Fe
that the doping phenomenon caused by electrons was samples. The main binding energy peak centered at 528.9 eV
associated with the change in the position of Raman and a shoulder peak around 531.3 eV are observed for CuCe
band.63,64 Therefore, the doping effect of Fe species on the sample. The main peak is ascribed to the lattice oxygen (OI),
CuCe sample is characterized by Raman analysis, and the and the shoulder peak reflects chemisorbed oxygen (OII) in
corresponding spectra of CuCe and CuCe-Fe catalysts are carbonates and hydroxyl groups.71,72 Upon Fe doping, it can be
shown in Figure 3A. The CuCe sample displays an intense seen that the main peak shifts to a higher value at 529.3 eV,
band at about 463 cm−1, attributed to the Raman active F2g which could be associated with the transformation of O
vibrational mode for oxygen atoms around the cerium ions, chemical environment from original Ce−O−Ce to Fe−O−Ce
which is an evidence for the typical cubic fluorite structure coordination (Fe containing CeO2-like solid solution).66,73 In a
CeO2.64,65 The broad peak around 599 cm−1 can be linked to word, the structural model of CuCe as well as the promotional
the oxygen vacancies/defects within the lattice of CeO2.66,67 effect of Fe additives on the structural property of CuCe are
When the Fe species is added, the intense band is observed to well-illuminated in Scheme 1.
3.3. Fe Promotional Effect on the Oxygen Mobility.
The structure−activity relationship over CuCe materials has
been previously reported and discussed extensively by large
amount of researchers.19−31 Although the presence of various
viewpoints on the cognition of the active species over CuCe
catalysts, all these studies provide insight into the possible
critical factors affecting the final catalytic performances from
different views via different methods and techniques. Thus, we
would take the active sites of CuCe-based catalysts into
account from the view of the principle of heterogeneous
catalysis. In general, the catalytic reaction process commonly
involved, first, the adsorption and the activation of reactant
molecules, then the reaction between reactants over the active
sites, as well as the desorption of products. As for CO-PROX
(including CO oxidation and H2 oxidation) and CO oxidation,
it was generally accepted that CO oxidation started with the
chemisorption and the activation of CO molecules over Cu
ions active sites. Then the reaction between Cu ions-CO
species and surface active oxygen species over Cu−Ce oxides
formed CO2 and left an oxygen vacancy. Finally, the formed
oxygen vacancies were replenished by gas-phase oxygen, which
was known to be a Mars−van Krevelen mechanism.27,74−77
Therefore, the catalytic activity of CuCe-based materials
toward CO oxidation in H2-rich is speculated to be largely
affected by two significant factors, one is the surface active
oxygen species, and the other is the Cu ions active sites.
Liable/reducible oxygen species are usually involved in the
oxidation reactions over oxide catalysts.74,75 In general, when
an reactant is oxidized at the active surface, the oxidant is
Figure 3. (A) Raman spectroscopy and (B) O 1s XPS spectra of usually a surface lattice oxygen atom instead of gas-phase
CuCe and CuCe-Fe. oxygen, which creates a surface oxygen vacancy.78,79 The
2183 DOI: 10.1021/acscatal.8b04035
ACS Catal. 2019, 9, 2177−2195
ACS Catalysis Research Article

Scheme 1. Structural Model for CuCe (a) and CuCe-Fe (b) Samples as well as the Corresponding Reaction Mechanism for
Special CO Oxidation as the Main Reaction in CO-PROX

production of oxygen vacancies plays an important role in the oxygen by H2 to produce water in the TPR process. The
formation of mobile oxygen species. As is known that the common H2-TPR profiles of the CuCe and CuCe-Fe catalysts
amount of oxygen vacancies can be estimated by several are shown in Figure 4A. For CuCe sample, H2-TPR profile
parameters, such as the ratio of Ce3+/(Ce3++Ce4+) from Ce 3d show three peaks at 161 °C (α), around 233 °C (β), and 250
XPS,80 the value of A599/A600 for Raman,81 and the ratio of °C (γ). It is generally accepted that the small α peak could be
OII/(OI+OII) from O 1s XPS.82 According to the preceding assigned to the consumption of reducible oxygen from the
result, these values are calculated to be 27.56, 0.67, 51.65 for highly dispersed CuOx species, large β peak can be ascribed to
CuCe-Fe sample, which is totally larger than that of 22.0, 0.60, the reduction of active oxygen in the strong interaction of Cu-
36.33, respectively, for CuCe sample. This would demonstrate [Ox]-Ce (or Fe−O−Ce) species, and small γ peak is attributed
that the addition of Fe into CuCe sample leads to the to the oxygen species presented in bulk CuO.19,23,83 Compared
generation of more amount of oxygen vacancies. The presence to the oxygen consumption of pure CuO characterized by one
of oxygen vacancies possibly enhances the activity of oxygen peak at 280−350 °C (Section 4 of SI) and between 300 and
(which can be evaluated by the oxygen mobility ability). The 400 °C in the literature,20,84 the decreased reduction
oxygen mobility promoted by Fe additive is further temperature for CuCe sample indicates that the oxygen
demonstrated by designed H2-TPR and OSC characterizations. species linked to Cu is activated by the interaction with
Temperature-programmed reduction by hydrogen (H2- CeO2. Upon doping Fe into the CuCe sample, the α and β
TPR) is actually used to characterize the presence of peak slightly shift to a lower temperature and the
reducible/active oxygen species due to the consumption of corresponding H2 consumption increases from 792.6 μmol/g
2184 DOI: 10.1021/acscatal.8b04035
ACS Catal. 2019, 9, 2177−2195
ACS Catalysis Research Article

It is deduced that the enhanced performance for the CO

catalytic oxidation using CuO-CeO2 as the catalyst depends on
the ability to provide the reactive oxygen to the reaction
process. Thus, the roles the oxygen mobility plays in the
reaction process becomes even more significant. The oxygen
release behavior and the oxygen mobility are further
characterized by the CO pulsing technique at 150 and 200
°C, with which oxygen storage capacity (OSC) and oxygen
storage complete capacity (OSCC) are determined and
evaluated by the production of CO2 (shown in Figure 4B)
and the consumption of CO (shown in Figure 4C). For Figure
4B, at 150 °C, the CuCe sample exhibits the low OSC vlaue of
46.4 μmol CO2/g at the first pulse with arriving the maximum
value at the fifth pulse, indicating the low oxygen mobility.
While after the addition of Fe the corresponding value is 94.8
μmol CO2/g at the first pulse with arriving the maximum value
forward to the third pulse, which provides the direct evidence
to indicate that Fe doping into CeO2 lattice largely facilitates
the migration rate between the highly dispersed oxygen, the
surface lattice oxygen, and the subsurface lattice oxygen via
oxygen vacancies.87,88 With the temperature increasing to 200
°C, the oxygen consumption rate obviously enhances, and the
oxygen storage capacity is observed to be improved by the
incorporation of Fe. These results demonstrate that Fe doping
into CuCe sample largely enhances the ability of oxygen
migration without the presence of O2, as shown in Scheme 1b.
The change in the CO consumption shown in Figure 4C is
consistent with the production of CO2, which also supports the
conclusion. Moreover, an interesting phenomenon is observed
for the generation of CO2 with the pulsing time proceeding. It
is generally speaking that the production of CO2 during the
pulsing process goes through one stage; that is, the amount of
CO2 would be generated largely at the first pulsing and
remains at a maximum value and then decrease gradually to
zero. However, in our Figure 4B, the production of CO2 goes
through two stages. That is, the amount of generated CO2
keeps first at a low value and then arrives to a maximum value;
however, afterward the value reduces to zero. These volcanic
Figure 4. (A) H2-TPR profiles of CuCe and CuCe-Fe. Comparison of
and inverted-volcanic curves observed in the production of
oxygen storage capacity (OSC) in the form of CO2 (B) and CO (C) CO2 and the consumption of CO during the pulsing process
under mild conditions of T = 150 and 200 °C for CuCe and CuCe-Fe. indicates the existence of two different types of reactive oxygen
species over CuCe and CuCe-Fe samples. One is very high
reactivity to CO and another is relative reactive activity.
for CuCe to 937.9 μmol/g for FeCuCe, indicating that Fe Combined with the TPR results, it is initially inferred that the
additive activates the surface lattice oxygen in CeO2, first active oxygen is attributed to the oxygen bound to the
promoting the oxygen migration. In other words, as before highly dispersed CuOx, and the second active oxygen is
proved that Fe doping into CeO2 lattice results in the creation assigned to the surface lattice oxygen bound to Cu-[Ox]-Ce
of more oxygen vacancies, the presented oxygen vacancies species.
possibly promote the migration ability of oxygen from CeO2 To prove the above assumption and further investigate the
subsurface to surface Cu-[Ox]-Ce and from surface to the role of the different types of reactive oxygen species on the
highly dispersed CuOx, and thus the generation of more highly activity of CO-PROX, a specific pretreatment by 10% (v/v)
reducible oxygen species,85,86 as shown in Scheme 1. The TPR H2/Ar at 150 °C (for 2 min over CuCe, P150-2, and for 1 min
peak in the higher temperature region in Figure 4A(a) over CuCe-Fe, P150-1, respectively) is used to eliminate the
supports the conclusion. It is seen that after the incorporation oxygen bound to the highly dispersed CuOx, and the
of Fe additive, the H2 consumption peak at around 400 °C corresponding TPR, OSC, and the CO oxidation activity
corresponding to the subsurface lattice oxygen over CuO- over CuCe and CuCe-Fe samples are in situ characterized
CeO2 (surface lattice oxygen of CeO2) disappears, and the without contacting any air. The corresponding results are
higher intensity of peak centered around 780 °C characterized shown in Figure 5A−F, respectively. In the TPR profiles
by the bulk lattice oxygen of CeO2 over Fe-CuCe (464.4 (Figure 5A,B), it is found that when the CuCe and CuCe-Fe
μmol/g) compared to CuCe sample (390.5 μmol/g) is samples are exposed to the 10%(v/v) H2/Ar at 150 °C for 2
observed, indicating that Fe doping facilitates the migration and 1 min, respectively, all the α peaks corresponding to the
of subsurface lattice oxygen as well as the release of bulk lattice oxygen bound to the highly dispersed CuOx are completely
oxygen over CuO-CeO2. consumed, and the β peaks characterized by the surface lattice
2185 DOI: 10.1021/acscatal.8b04035
ACS Catal. 2019, 9, 2177−2195
ACS Catalysis Research Article

Figure 5. Designed TPR characterization of CuCe (A) and CuCe-Fe (B) samples, P150-0: the fresh samples, P150-1 and P150-2: after
pretreatment at 150 °C for 1 min, P150-1-R160 and P150-2-R160: the pretreated samples were reacted from 60 to 160 °C every 20 °C, and P150-
1-R220 and P150-2-R220: the pretreated samples were reacted from 60 to 220 °C every 20 °C. OSC measurements of the pretreated CuCe and
CuCe-Fe samples calculated in the form of CO2 (C) and CO (D). Comparison of the catalytic activity of CuCe (E) and CuCe-Fe (F) before
(black line) and after (red line) pretreatment without contacting any oxygen gas. Reaction conditions: mcat = 100 mg, feed gas: 1% CO, 1% O2, 50%
H2 and balance He, flow rate = 80 mL/min.

oxygen bound to Cu-[Ox]-Ce species are almost unaffected. promotes the lattice oxygen migration rate. Thus, we assume
These pretreated samples are then in situ characterized by the that the identical CO oxidation activity before and after the H2
CO pulsing processing. It is seen from Figure 5C,D that the pretreatment is not only due to the migration of the subsurface
volcanic and inverted-volcanic curves of CuCe and CuCe-Fe lattice oxygen to surface lattice oxygen at the relative high
samples disappear, and the amount of CO2 generates at the temperature (≥100 °C) and but also due to the migration of
first pulsing with a maximum value, and then decreases gas-phase oxygen to the highly dispersed oxygen during the
gradually to zero. This result provides a proof to prove that the reaction process at the relative low temperature (<100 °C). In
oxygen consumed at the first three CO pulsing in the Figure order to prove our hypothesis, the CuCe and CuCe-Fe samples
4B,C should be related to the highly dispersed CuOx, and the after the activity test from 60 to 160 °C, and from 60 to 220
oxygen consumed after the three CO pulsings should be °C, respectively, are ex-situ characterized by H2-TPR, as
associated with Cu-[Ox]-Ce species. It seems to deduce that displayed in Figure 5A,B. As seen, when samples go through
the very high reactive oxygen is connected to the highly the activity test from 60 to 160 °C, no distinct α peak is
dispersed CuOx, which is thought to be preferentially observed for CuCe sample (P150-2-R160), and the β peak
consumed in the CO-PROX process especially at the low largely reduces without the obvious shift in the peak position,
temperature. The catalytic activities for CO-PROX over CuCe demonstrating that the highly dispersed oxygen is first
and CuCe-Fe samples evaluated in situ after the H 2 consumed at low temperature; then surface lattice oxygen
pretreatment are shown in Figure 5E,F, respectively. The CO largely participates in the reaction process at the relative high
oxidation activity is expected to reduce at the low temperature temperature region. More importantly, there is no existence of
region over CuCe catalyst, which proves that the oxygen the oxygen migration observed on CuCe in the process of CO-
bound to highly dispersed CuOx is beneficial to improve the PROX. The TPR profile of CuCe sample after the activity test
low-temperature CO oxidation activity. Although the CO from 60 to 220 °C (P150-2-R220) also confirms the point
oxidation activity decreases to some extent, the dominant because of the inexistence of α peak without the distinct shift
activity for CO-PROX is not loss, which is an indicative of the in the peak position. More precisely, only the weak oxygen
importance of the surface lattice oxygen bound to the Cu- migration is observed to be presented for CuCe sample. To
[Ox]-Ce species that participate in the CO oxidation process, our surprise, after the activity test from 60 to 160 °C, the
especially at the relative high temperature region. However, CuCe-Fe sample exhibits the appearance of α peak, and the
when Fe species is incorporated into CuCe sample, it is strange peak position is shifted from 154 to 177 °C. The β peak
that there is no decrease in the CO oxidation activity. This intensity decreases to be even lower than that of pretreated
result seems to arrive a opposite result that the oxygen bound CuCe sample, implying the significant enhancement of oxygen
to highly dispersed CuOx has no influence on the CO migration. The TPR profile of CuCe-Fe sample after the
oxidation activity. To illuminate clearly this phenomenon, the activity test from 60 to 220 °C found that the β peak is
structural properties change induced by the incorporation of completely shifted to the position at 177 °C and the peak
Fe should be considered. As evidenced by XRD, XPS, Raman, intensity largely increases, which also supports the migration of
and H2-TPR that the incorporation of Fe into CeO2 lattice surface lattice oxygen. Moreover, the weak TPR reduction peak
leads to the formation of more oxygen vacancies, which between 350 and 450 °C corresponding to the consumption of
2186 DOI: 10.1021/acscatal.8b04035
ACS Catal. 2019, 9, 2177−2195
ACS Catalysis Research Article

subsurface lattice oxygen over CeO2 is found to be disappeared pulse is introduced, the intensity of β peak corresponding to
after the addition of Fe, proving the release of subsurface bulk the surface lattice oxygen bound to Cu-[Ox]-Ce species largely
lattice oxygen. Therefore, it is concluded that the highly reduces for CuCe sample (Figure 6A), while the corresponding
dispersed oxygen linked to CuOx species is in favor of the β peak has no decreasing in the intensity for CuCe-Fe sample.
improvement of the low-temperature CO oxidation activity, This phenomenon probably suggests that the surface lattice
and because of slow consumption rate of lattice oxygen as oxygen over CuCe-Fe sample is rapidly consumed at the initial
observed by CO pulsing at 200 °C after the third time, lattice stage, and then compensated by the migration of the
oxygen is considered as a critical factor that participates in the subsurface lattice oxygen to surface due to the generation of
rate-limiting step of the whole redox reaction for the oxidation large number of oxygen vacancies via the doping of Fe into
of CO. The existence of oxygen vacancies not only traps gas- CeO2 lattice. For quantitative analysis, it is found that when
phase oxygen to form surface active oxygen species at relative the same amount of CO is introduced, the amount of the
low temperature but also provides a certain freedom for the consumed active oxygen measured by TPR analysis is 87
migration of lattice oxygen to increase the mobility of oxygen μmol/g for CuCe-Fe, which is lower than that of 106 μmol/g
at relative high temperature.78,81 Moreover, it is important that for CuCe sample, proving that partial active oxygen is
we provide an important method via OSC measurement to compensated by the migration from the subsurface lattice
determine the different types of reactive oxygen species over oxygen, where the migration would take a certain of times. In
CuCe-based catalysts. addition to the change in the β peak, some important
It should be noted that the presence of large amount of information can be obtained from the variation in the α peak
molecular oxygen in the reaction process (1%CO+1%O2+50% intensity after CO pulse was introduced. As seen, small amount
H2) and the exposure the samples to the air during the samples of highly dispersed active oxygen (23 μmol/g) is consumed for
transfer is possibly involved in the process of the lattice oxygen CuCe sample, while the incorporation of Fe results in the
migration. To exclude the influence of molecular oxygen on consumption of large number of highly dispersed active oxygen
(87 μmol/g). On the one hand, this result demonstrates that
the lattice oxygen migration and avoid the exposure to the air,
due to the incorporation of Fe species the highly dispersed
the CO pulsing technique at 200 °C combining with in situ
active oxygen is easier reactive. On the other hand, since the
H2-TPR characterization is applied to investigate the oxygen
absence of gas-phase oxygen in the CO pulsing process, the
migration. In detail, one CO pulse is introduced to consume
lower intensity of α peak over pulsed FeCuCe (Figure 6B)
the lattice oxygen at 200 °C, then H2-TPR is conducted in situ
demonstrates that the compensation of highly dispersed active
after decreasing the temperature and switching the gas to 10% oxygen after consumed at the low-temperature region (Figure
(v/v)H2/Ar in the same equipment. The corresponding H2- 5B,F) is originated from the trapping of gas-phase oxygen via
TPR profiles over CuCe and CuCe-Fe samples are shown in the oxygen vacancies rather than the migration of bulk lattice
the Figure 6A,B, respectively. Compared the temperature oxygen.
position of α with that of β peak, it is deemed that the 3.4. Chemisorption of CO over the Active Cu Species.
temperature at 200 °C is in favor of eliminate the surface The properties of the chemisorption and the activation of CO
lattice oxygen corresponding to the β peak.19 After one CO are characterized by CO-TPD,87,89 as shown in Figure 7A. It is
seen that the absorption and the activation of CO is starting at
110 °C over CuCe sample, whereas the starting temperature
decreases to 90 °C, which is consistent with the activity result.
Moreover, the CO chemisorption uptakes is higher for CuCe-
Fe than that for CuCe sample. These results indicate that the
addition of Fe species promotes the absorption and the
activation of CO on the Cu ions active sites of CuCe and
CuCe-Fe samples.
In situ DRIFT spectroscopy is used to probe the formation
of surface Cu ions adsorption sites and the CO oxidation
behavior. The bands in the 1800−2200 cm−1 region
characterized by Cuδ+-CO species as well as the wavenumber
of 1000−1800 cm−1 corresponding to the carbonate-type zone
over CuCe and CuCe-Fe samples are displayed in Figure 7B,C,
respectively. In general, the characteristic vibration frequencies
corresponding to the adsorption of CO molecule over three
Cu ions, Cu2+, Cu+ and Cu0, are characterized by the bands
between 2160 and 2210 cm−1,90,91 2080−2140 cm−1,24,83
2115−2000 cm−1,83,92,93 respectively. Notably, it was reported
that metallic copper carbonyls (i.e., Cu(755)-CO, Cu(311)-
CO, Cu(211)-CO) show frequencies as high as 2111, 2109,
2109 cm−1, respectively, and Cu(111)-CO, Cu(100)-CO,
Cu(110)-CO, show frequencies as low as 2077, 2078, 2088
cm−1, respectively.93 Considering the fact that Cu+-carbonyl
has the highest thermal stability than Cu2+-carbonyl and Cu0-
Figure 6. Designed in situ TPR characterization of CuCe (A) and carbonyl as well as the Cu(111), Cu(100), Cu(110) planes are
CuCe-Fe (B) samples before (gray line) and after (red line) one CO the more stable species,84 the bands at 2100 and 2105 cm−1 for
pulse was introduced without contacting any oxygen gas. CuCe and CuCe-Fe samples could be attributed to the CO
2187 DOI: 10.1021/acscatal.8b04035
ACS Catal. 2019, 9, 2177−2195
ACS Catalysis Research Article

in the higher intensity of the band at 2100 cm−1 over CuCe-Fe

sample, which is one of the active sites for the enhanced CO
catalytic activity. Furthermore, the frequencies at 1579 and
1294 cm−1 corresponding, respectively, to the C−O stretching
and asymmetric O−C−O stretching are visible and no distinct
difference in the intensity is observed for CuCe and CuCe-Fe
samples, indicating the formation of tridentate CO 3 2−
intermediates on the surface ceria sites.87,94 The vibrational
frequencies at 2336 and 2368 cm−1 corresponding to the
gaseous CO2 are observed for the two samples, indicating the
reduction of surface oxide species with the incoming CO.
In general, two intermediates, Cu+-CO and CO32−, are
formed on the surface of CuCe and CuCe-Fe samples. At
below 100 °C, it is distinct that the peak intensity of Cu+-CO
species is great at the initial temperature and increases
gradually as the elevating temperature, while the peak intensity
of CO32− intermediates is very low. These indicate that the CO
molecules are easily chemisorbed to generate Cu+-CO
complexes rather than CO32− intermediates, which then are
oxidized with surface active oxygen to generate CO2 at the low
temperature region,87 as shown in Scheme 1. Thus, a greater
amount of generated Cu+-CO complexes combining with the
surface easier activated oxygen would give rise to the higher
CO oxidation activity. At the high temperature between 100
and 170 °C, the peak intensity of Cu+-CO complexes gradually
weakens until it disappears, suggesting that the complexes are
rapidly decomposed because of the lower thermal stability or/
and Cu+ species are partially reduced into Cu0, and the peak
intensity of CO32− intermediates increases, indicating that CO
molecules are absorbed on Cu+ and reacted with the activated
oxygen to form CO32− intermediates.87 Further decomposition
of CO32− intermediates results in the generation of CO2.87 At
the higher temperature (above 170 °C), all the intermediates
vanish, implying that all the active oxygen are consumed with
the transformation of the active Cu+ to the inactive Cu0
species, which results in the rapid decrease of CO oxidation
activity.
3.5. Influence of Fe Additive on the Catalyst Stability
and the Resistance to CO2 and H2O of CuCe Catalyst. It
is well-known that H2O and CO2 in the hydrogen-rich gas can
impact the catalytic activities and stability of catalysts for CO-
Figure 7. (A) CO-TPD of CuCe and CuCe-Fe. In-situ DRIFTS PROX. Thus, it is practical to study the catalytic performances
spectra of CuCe (B) and CuCe-Fe (C) recorded under the reaction of CuCe- and Fe-doped CuCe samples for the CO-PROX
stream of 1% CO/1% O2/50% H2/He at the elevated temperature. reaction in hydrogen rich streams containing CO2 and H2O.
The corresponding results of CuCe and CuCe-Fe catalysts in
the existence of H2O and CO2 are displayed in Figure 8A,B,
molecule adsorbed to Cu+ sites (Cu+-carbonyl species). These respectively. As shown, although the activity performance of
Cu+-CO species are the most stable structure among three Cu two catalysts is decreased to some extent with H2O and CO2
ions structure, and are generally stabilized at the interface of presented in the feed gas (compared with Figure 1), as
CuO-CeO2 in the form of Cu-[Ox]-Ce structure or/and the observed by other researchers.22,95 However, compared with
highly dispersed CuOx close to interface.28 After the addition CuCe sample, CuCe-Fe exhibits the higher CO conversion and
of Fe species, the slight red shift from 2105 cm−1 for CuCe CO2 selectivity in the existence of H2O and CO2, especially at
sample to 2100 cm−1 for CuCe-Fe sample is observed, and 140 °C, proving that the incorporation of Fe improves the
meanwhile, the corresponding intensity of Cu+-CO peak ability of the tolerance to H2O and CO2 for CO-PROX. The
increases distinctly. It was reported that the red shift of Cu+- stability of the catalyst is one of the important parameters to
CO peak to the lower frequency was reflected by the stronger evaluate the property of the catalyst, which is essential for its
Cu−C bond due to the electron transfer from the d electrons practical utilization. In order to test the stability of CuCe-Fe, a
of the Cu+ to the CO molecule caused by the stronger Cu−Ce time-on-stream experiment was investigated with H2O and
interaction.23,81 Upon the addition of Fe into CuO-CeO2, Fe CO2 in the feed gas, as shown in Figure 8C (the stability test of
doping into CeO2 lattice forming the Fe−O−Ce structure CuCe as a comparison reference). It is found that the CuCe
gives rise to the variation in the electronic properties of Ce sample possesses the relative stabilized CO conversion and
surrounding, which strengthens the interaction between Cu CO2 selectivity before time on stream of 170 h, and then
and Ce. Thus, the stronger interaction (not too strong) results rapidly deactivates, while the CuCe-Fe sample largely prolongs
2188 DOI: 10.1021/acscatal.8b04035
ACS Catal. 2019, 9, 2177−2195
ACS Catalysis Research Article

H2O and poisoned by the carbonate-type species that are

generated through the interaction of H2O and formed CO2. In
our work, the addition of Fe into CuCe sample gives rise to the
higher resistance to H2O and CO2 for CO oxidation activity
and the superior long-term stability under the coexistence of
H2O and CO2. The following reasons are considered to explain
the excellent performance. On the one hand, due to the
promotional effect of Fe doping on the oxygen mobility (as a
larger oxygen reservoir), as preceding confirmed by in situ
TPR and OSC, enough surface reactive oxygen can be
provided, and the deactivation induced by CO2 can be relieved
to a certain degree upon the release and migration of lattice
oxygen. On the other hand, Fe-based catalysts are well-known
to be active for WGS reaction (CO + H2O = CO2 + H2) at
high temperature (>300 °C).102,103 CuCe catalysts was
reported to have good activity for WGS at the temperature
above 200 °C, while only about 5% CO conversion was
obtained at the temperature blew 200 °C.104−106 If the
addition of Fe species into CuCe catalysts could facilitate the
occurrence of WGS reaction at below 200 °C, further
improving the tolerance of H2O and CO2 as well as the
stability of catalysts, is not clear.
In general, WGS reaction over CuCe catalyst includes the
reaction of CO with surface original −OH groups of catalyst
surface as well as the H2O-assisted CO oxidation by the
formation of surface −OH groups. First, the reaction of CO
with surface original −OH groups of catalyst surface via the
WGS type reaction (CO + −OH → CO2 + 1/2H2) is
considered. Since hydrogen (H2) is a probe molecule for the
occurrence of WGS-type reaction, CO-TPR was used to detect
the presence of H2, and in situ DRIFTS is conducted to detect
the formation of intermediate species. The CO-TPR profiles
and in situ DRIFTS spectra on CuCe and FeCuCe samples are
displayed in Figure 9A−C. As for Figure 9A, it is found that
the initial temperature for producing H2 is 170 °C for CuCe
sample, while the corresponding temperature decreased to 140
Figure 8. (A) CO conversion and (B) CO2 selectivity of CuCe and °C over FeCuCe sample, suggesting the possible occurrence of
CuCe-Fe with the coexistence of H2O and CO2. (C) The stability test WGS-type reaction between surface original −OH and CO at
with time on stream for CuCe at 160 °C and CuCe-Fe at 155 °C, 140 and 160 °C due to the promotion role of Fe additive. To
respectively. mcat = 300 mg, feed gas composition: 1% CO, 1% O2, investigate the effect of adsorbed water on the WGS-type
50% H2, 15% CO2, 10% H2O and He as balance gas, flow rate = 80
mL/min.
reaction, fresh FeCuCe was pretreated at 700 °C for 1 h in the
atmosphere of pure N2. After the removal of adsorbed water
the stability time from 170 to 400 h, suggesting that the and partial OH groups, the observed temperature increases to
addition of Fe significantly strengthens the stability of CuCe 190 °C over the FeCuCe sample, indicating that the WGS-type
catalyst for CO-PROX in H2-rich gas. It was generally accepted reaction is either facilitated by the adsorbed water or inhibited
that reason for the deactivation of the Cu−Ce-based catalyst by the decreased amount of −OH groups. As known, formate
induced by the addition of CO2 into the feed gas was related to species is generally regarded as an intermediate species in
the suppression of oxygen mobility upon the generation of WGS-type reaction, which would decompose into CO2 and H2
carbonates over the CeO2 surface as well as the competitive with the temperature raising.81,104,107 In Figure 9B,C, the
adsorption of CO2 over Cu+ adsorption sites.96−98 H2O was absorbed formate species characterized by the vibrational
deemed to force a stronger inhibition effect on the CO frequencies at 2842, 2852, 2932, and 2965 cm−1 are distinctly
oxidation activity and stability, and the deactivating effect was observed for CuCe and FeCuCe samples.81,107,108 It is also
due to the blockage of Cu+ adsorption sites by the presence of seen that a larger amount of this intermediate species is formed
water vapor.99−101 To prove the deactivation reason, in situ on FeCuCe at the lower temperature, when compared with the
DRIFTS spectra was performed over CuCe and FeCuCe sample of CuCe. Meanwhile, the starting temperature for the
catalysts in the atmosphere of reaction gas (1% CO, 1% O2, decrease in peaks intensity of intermediate species over CuCe
50% H2) with the introduction of a certain amount of H2O. is 150 °C, while the corresponding temperature for FeCuCe is
The corresponding results are shown in Figure S3A,B (Section 170 °C. These results demonstrate the promotional effect of Fe
5 of SI). In general, no bands attributed to Cu+-CO species are additive on the formation of absorbed formates and on the
observed, and the intense signals between 1200 and 1700 cm−1 decomposition of intermediate into CO2 and H2 (i.e., WGS-
assigned to carbonates-type species are distinct, indicating that type reaction), which is in good accordance with the CO-TPR
Cu+ adsorption sites are possibly occupied by the presence of results.
2189 DOI: 10.1021/acscatal.8b04035
ACS Catal. 2019, 9, 2177−2195
ACS Catalysis Research Article

H2 O and CO2 of the catalysts is significant at this

temperature), 488 ppm of CO2 accompanied by 13 ppm of
H2 is generated over FeCuCe. This result demonstrates that
the reaction between CO and surface formed −OH groups is
not the main factor promoting the tolerance of H2O and CO2
of the catalysts at 140 °C because the amount of produced H2
is small at this time. Meanwhile, nearly half amount of the
generated CO2 compared with the consumed CO confirms
that the high tolerance of CO2 and H2O of FeCuCe catalyst at
140 °C (Figure 8A,B) is resulted from the possible occurrence
of the disproportionation reaction (2CO → CO2 + C). With
the temperature increasing to 160 °C (the stability tests are
performed), when 2071 ppm of CO is consumed, 1268 ppm of
CO2 accompanied by 379 ppm of H2 is formed over FeCuCe,
suggesting that 758 ppm of CO is used for the reaction
between CO and surface formed −OH groups, and 1020 ppm
of CO is used for the disproportionation reaction. These
results evidence that, although the reaction between CO and
surface formed −OH groups becomes important, the
disproportionation reaction has the slight greater impact on
the stability than the latter (1.34 vs 1.0). Thus, the longer-term
stability of FeCuCe than CuCe sample at about 160 °C is
attributed to the synergistic roles of the disproportionation
reaction with the reaction between CO and surface formed
−OH groups. With the temperature increasing from 160 °C to
the higher one, the reaction between CO and surface formed
−OH groups is gradually predominant.
To further prove the occurrence of the disproportionation of
CO as well as the subsequent reaction (C + H2O = CO + H2),
which is deemed as another pathway for WGS reaction, the
determination of the presence of carbon-containing species
over the catalyst surface is an important evidence. Therefore,
the CO pulsing technique was introduced to proceed the
disproportionation of CO where enough CO pulse was
consumed at 140 °C. The pulsed FeCuCe sample was further
reacted with 10% H2O at 140 °C for 3 h to verify the
subsequent reaction (C + H2O = CO + H2). The pulsed CuCe
and FeCuCe samples as well as reacted FeCuCe after CO
Figure 9. (A) CO-TPR profiles of fresh CuCe and FeCuCe samples pulsing were characterized by Raman technique so as to
as well as the pretreated FeCuCe sample (calcined in situ in the determine the presence and the variation of carbon species.
atmosphere of pure N2 at 700 °C for 1 h). In situ DRIFTS spectra of The Raman spectroscopy of them are shown in Figure 10C,
CuCe (B) and CuCe-Fe (C) recorded at the range of 2500−3200 and H2 concentration produced during the reaction of pulsed
cm−1 under the reaction stream of 1% CO/1% O2/50% H2/He. FeCuCe with H2O is also displayed in Figure 10D. It is found
that no Raman bands corresponding to carbon species are
Second, 10% H2O is introduced into the reaction system to found for pulsed CuCe sample, whereas pulsed FeCuCe
consider the H2O-assisted CO oxidation by the formation of sample exhibits distinct Raman peaks at 1360−1400 cm−1, and
surface −OH groups. The independent WGS reactions were 1480−1520 cm−1 as well as 1580−1640 cm−1, attributed
performed over CuCe and FeCuCe samples by using 1% CO typically to the different type of unorganized carbonaceous
and 10% H2O as reaction gas. CO conversion and products species,72,109−111 which evidence the occurrence of the
concentration of WGS reaction over two samples are shown in disproportionation of CO. After the reaction between the
Figure 10A,B, respectively. It is seen from Figure 10A that CO pulsed FeCuCe sample with deposited carbonaceous species
conversions of CuCe and FeCuCe samples are close to 0% at and H2O, the amount of deposited carbonaceous species
120 °C, and, at higher temperature, the conversion is higher obviously decreases (Figure 10C-(c)), and meanwhile, a small
for FeCuCe (10.5% at 140 °C and 20.7% at 160 °C, number of H2 is generated (Figure 10D), demonstrating
respectively) than that for CuCe (1.2% at 140 °C and 4.9% at unambiguously the consumption of carbonaceous species by
160 °C, respectively), suggesting that WGS reaction occurs at a the presence of H2O (C + H2O = CO + H2). These
temperature above 120 °C and that the addition of Fe largely carbonaceous species are generally under development, and
facilitates the WGS reaction. The concentration of products thus can be reacted with H2O to generate CO and H2 under
(CO2 and H2) during the course of WGS reaction is further the current reaction condition. However, large amount of H2
analyzed to confirm the promotional effect of Fe additive as cannot be produced possibly because that the reaction of
well as the reaction mechanism. As shown in Figure 10B, it is carbonaceous species with H2O is a rate-determining step,
found that CO with the concentration of 1046 ppm is which is favorable in the long term stability test instead. This
consumed at 140 °C (in fact, the difference in the tolerance of slow reaction rate of deposited carbonaceous species with H2O
2190 DOI: 10.1021/acscatal.8b04035
ACS Catal. 2019, 9, 2177−2195
ACS Catalysis Research Article

Figure 10. (A) CO conversion and (B) products concentration of WGS reaction over CuCe and CuCe-Fe samples. Reaction conditions: mcat =
300 mg, feed gas: 1% CO, 10% H2O, and balance N2, flow rate = 80 mL/min. (C) Raman spectroscopy of (a) pulsed CuCe (all the oxygen was
consumed by CO pulsing), (b) pulsed FeCuCe, (c) pulsed FeCuCe after the reaction with 10% H2O at 140 °C for 3 h (balance with pure N2 to
keep the flow rate of 80 mL/min). (D) H2 concentration of pulsed FeCuCe sample after the reaction with 10% H2O at 140 °C for 3 h.

and the fast reaction rate of the disproportionation reaction molecules. The modification of Fe additive resulted in the
could lead to the generation of small amount of H2 as well as incorporation of CeO2 lattice so as to induce the redox
large amount of CO2, which can also explain the situation that chemistry between Fe3+/Fe2+ and Cu1+/Cu2+, which appropri-
the selectivity of H2 decreases at 140 °C in Figure 8B (CO2 ately enhanced the adsorption ability of CO over Cu1+
selectively increases) as well as the selectivity of H2 increases at adsorption sites. Our combined in situ OSC and designed
about 160 °C in Figure 8C (CO2 selectively decreases). TPR characterizations showed that the surface highly dispersed
In summary, considering that the conversions of CO in oxygen species related to CuOx species was confirmed as one
WGS reaction are approximately 10% at 140 °C and 20% at of active species and to be involved in the CO oxidation at low
160 °C over FeCuCe, the primary causes of the high tolerance temperature. The surface lattice oxygen species assisted with
of H2O and CO2 toward FeCuCe catalyst at 140 °C and the Cu-[Ox]-Ce (Fe−O−Ce) was also proved to participate in the
long-term stability of FeCuCe around 160 °C is due to the reaction process. Because of the charge compensation and the
promotional effect of Fe additive on the lattice oxygen difference in the ionic radius between Fe and Ce ions, the
mobility, the second one is the important role of the WGS incorporation of Fe into CeO2 lattice generated plentiful
reaction, which is driven by the addition of Fe. Two reaction oxygen vacancies, which was not only in favor of trapping gas-
pathways (mechanism) are considered to be occurred during phase oxygen to form the surface highly dispersed oxygen, but
the course of WGS reaction. The disproportionation of CO is also promoted the migration of subsurface lattice oxygen to
responsible for the high tolerance of H2O and CO2 of FeCuCe surface active oxygen. The former promoted the CO catalytic
catalyst at 140 °C, and WGS reaction by the disproportiona- activity (TOF value increased from 3.62 × 10−2 s−1 for CuCe
tion pathway as well as the H2O-assisted CO oxidation by the to 4.50 × 10−2 s−1 for CuCe-Fe), and the latter in combination
formation of surface −OH groups are beneficial to the long- with the high resistance to H2O using the Fe-doped catalyst as
term stability of FeCuCe. However, the decomposition of a favorable WGS catalyst made an important contribution to
surface carbonate by H2O is not regarded as the main reason the long stability in the presence of CO2 and H2O (increased
contributing to the high tolerance of H2O and CO2 and the the lifetime from 170 to 400 h).

■
excellent stability of the catalyst, as confirmed (Section 5 of
SI). ASSOCIATED CONTENT
*
S Supporting Information
4. CONCLUSIONS The Supporting Information is available free of charge on the
In summary, CuCe- and Fe-doped CuCe catalysts were ACS Publications website at DOI: 10.1021/acscatal.8b04035.
prepared via a solvent-free combustion synthesis (SFCS)
method to investigate the active species and the Fe doping Additional transmission electron microscope (TEM),
effect. The active Cu1+ species were demonstrated by in situ H2-TPR profile of CuO, in situ DRIFTS of CuCe and
DRIFTS technique and XPS as well as CO-TPD to be FeCuCe samples in the gas stream with the addition of a
adsorption sites for the adsorption and the activation of CO certain amount of H2O, the method of calculating the

2191 DOI: 10.1021/acscatal.8b04035

ACS Catal. 2019, 9, 2177−2195
ACS Catalysis Research Article

concentration of oxygen vacancies, and the explanation carbon monoxide in excess hydrogen. Catal. Today 2002, 75, 157−
of satellite peak of Fe 2p XPS spectra (PDF) 167.

■
(14) Gamarra, D.; Munuera, G.; Hungria, A. B.; Fernandez-Garcia,
M.; Conesa, J. C.; Midgley, P. A.; Wang, X. Q.; Hanson, J. C.;
AUTHOR INFORMATION Rodriguez, J. A.; Martinez-Arias, A. Structure-activity relationship in
Corresponding Author nanostructured copper-ceria-based preferential CO oxidation cata-
*E-mail: environcatalysis222@[Link]. lysts. J. Phys. Chem. C 2007, 111, 11026−11038.
(15) Gamarra, D.; Hornés, A.; Koppàny, Z.; Schay, Z.; Munuera, G.;
ORCID Soria, J.; Martínez-Arias, A. Catalytic processes during preferential
Yongming Luo: 0000-0002-9930-261X oxidation of CO in H2-rich streams over catalysts based on copper-
Notes ceria. J. Power Sources 2007, 169, 110−116.
The authors declare no competing financial interest. (16) Ilichev, A. N.; Firsova, A. A.; Korchak, V. N. Mechanism of CO

■
oxidation in excess H2 over CuO/CeO2 catalysts: ESR and TPD
ACKNOWLEDGMENTS studies. Kinet. Catal. 2006, 47, 585−592.
(17) Skårman, B.; Grandjean, D.; Benfield, R. E.; Hinz, A.;
We gratefully acknowledge the financial supports of National Andersson, A.; Wallenberg, L. R. Carbon monoxide oxidation on
Natural Science Foundation of China (Grant Nos. 21267011, nanostructured CuOx/CeO2 composite particles characterized by
21367015, 21666013, 51704137 and U1402233). HREM, XPS, XAS, and High-energy diffraction. J. Catal. 2002, 211,

■ REFERENCES
(1) Ercolino, G.; Ashraf, M. A.; Specchia, V.; Specchia, S.
119−133.
(18) Gamarra, D.; Cámara, A. L.; Monte, M.; Rasmussen, S. B.;
Chinchilla, L. E.; Hungría, A. B.; Munuera, G.; Gyorffy, N.; Schay, Z.;
Corberán, V. C.; Conesa, J. C.; Martínez-Arias, A. A. Preferential
Performance evaluation and comparison of fuel processors integrated
with PEM fuel cell based on steam or autothermal reforming and on oxidation of CO in excess H2 over CuO/CeO2 catalysts: Character-
CO preferential oxidation or selective methanation. Appl. Energy ization and performance as a function of the exposed face present in
2015, 143, 138−153. the CeO2 support. Appl. Catal., B 2013, 130−131, 224−238.
(2) Lu, J. C.; Li, X. F.; He, S. F.; Han, C. Y.; Wan, G. P.; Lei, Y. Q.; (19) Wang, W. W.; Du, P. P.; Zou, S. H.; He, H. Y.; Wang, R. X.; Jin,
Chen, R.; Liu, P.; Chen, K. Z.; Zhang, L.; Luo, Y. M. Hydrogen Z.; Shi, S.; Huang, Y. Y.; Si, R.; Song, Q. S.; Jia, C. J.; Yan, C. H.
production via methanol steam reforming over Ni-based catalysts: Highly Dispersed Copper Oxide Clusters as Active Species in Copper-
Influences of Lanthanum (La) addition and supports. Int. J. Hydrogen Ceria Catalyst for Preferential Oxidation of Carbon Monoxide. ACS
Energy 2017, 42, 3647−3657. Catal. 2015, 5, 2088−2099.
(3) Njagi, E. C.; Genuino, H. C.; King'ondu, C. K.; Chen, C. H.; (20) Luo, M. F.; Ma, J. M.; Lu, J. Q.; Song, Y. P.; Wang, Y. J. High-
Horvath, D.; Suib, S. L. Preferential oxidation of CO in H2-rich feeds surface area CuO-CeO2 catalysts prepared by a surfactant-templated
over mesoporous copper manganese oxides synthesized by a redox method for low-temperature CO oxidation. J. Catal. 2007, 246, 52−
method. Int. J. Hydrogen Energy 2011, 36, 6768−6779. 59.
(4) Zhang, L.; Kim, J.; Zhang, J.; Nan, F.; Gauquelin, N.; Botton, G. (21) Zeng, S. H.; Zhang, W. L.; Guo, S. L.; Su, H. Q. Inverse rod-like
A.; He, P.; Bashyam, R.; Knights, S. Ti4O7 supported Ru@Pt core- CeO2 supported on CuO prepared by hydrothermal method for
shell catalyst for CO- tolerance in PEM fuel cell hydrogen oxidation preferential oxidation of carbon monoxide. Catal. Commun. 2012, 23,
reaction. Appl. Energy 2013, 103, 507−513. 62−66.
(5) Ayastuy, J.; Gutiérrez-Ortiz, M. A.; González-Marcos, J.; (22) Wang, J.; Deng, L.; He, D. D.; Lu, J. C.; He, S. F.; He, S. Y.;
Aranzabal, A.; González-Velasco, J. R. Kinetics of the Low- Luo, Y. M. A facile and rapid route to synthesize CuOx/Ce0.8Zr0.2O2
Temperature WGS Reaction over a CuO/ZnO/Al2O3 Catalyst. Ind. catalysts with high performance for CO preferential oxidation (CO-
Eng. Chem. Res. 2005, 44, 41−50. PROX). Int. J. Hydrogen Energy 2015, 40, 12478−12488.
(6) Reyes-Carmona, A.; Arango-Diaz, A.; Moretti, E.; Talon, A.; (23) Guo, X. L.; Zhou, R. X. Identification of the nano/micro
Storaro, L.; Lenarda, M.; Jimenez-López, A.; Rodriguez-Castellón, E. structure of CeO2(rod) and the essential role of interfacial copper-
CuO/CeO2 supported on Zr doped SBA-15 as catalysts for ceria interaction in CuCe(rod) for selective oxidation of CO in H2-
preferential CO oxidation (CO-PROX). J. Power Sources 2011, 196, rich streams. J. Power Sources 2017, 361, 39−53.
4382−4387. (24) Guo, X. L.; Mao, J. X.; Zhou, R. X. Influence of the copper
(7) Song, C. Fuel processing for low-temperature and high- coverage on the dispersion of copper oxide and the catalytic
temperature fuel cells Challenges, and opportunities for sustainable performance of CuO/CeO2(rod) catalysts in preferential oxidation
development in the 21st century. Catal. Today 2002, 77, 17−49. of CO in excess hydrogen. J. Power Sources 2017, 371, 119−128.
(8) Zhang, Y.; Liang, H.; Gao, X. Y.; Liu, Y. Three-dimensionally (25) Du, P. P.; Wang, W. W.; Jia, C. J.; Song, Q. S.; Huang, Y. Y.; Si,
ordered macro-porous CuO-CeO2 used for preferential oxidation of R. Effect of strongly bound copper species in copper-ceria catalyst for
carbon monoxide in hydrogen-rich gases. Catal. Commun. 2009, 10, preferential oxidation of carbon monoxide. Appl. Catal., A 2016, 518,
1432−1436. 87−101.
(9) Trimm, D. L. Minimisation of carbon monoxide in a hydrogen (26) Gamarra, D.; Belver, C.; Fernandez-Garcıa, M.; Martınez-Arias,
stream for fuel cell application. Appl. Catal., A 2005, 296, 1−11. A. Selective CO Oxidation in Excess H2 over Copper-Ceria Catalysts:
(10) Morse, J. D. Micro-fuel cell power sources. Int. J. Energy Res. Identification of Active Entities/Species. J. Am. Chem. Soc. 2007, 129,
2007, 31, 576−602. 12064.
(11) Moretti, E.; Lenarda, M.; Storaro, L.; Talon, A.; Frattini, R.; (27) Yao, S. Y.; Mudiyanselage, K.; Xu, W. Q.; Johnston-Peck, A. C.;
Polizzi, S.; Rodríguez-Castellon, E.; Jiménez-Lopez, A. Catalytic Hanson, J. C.; Wu, T. P.; Stacchiola, D.; Rodriguez, J. A.; Zhao, H. Y.;
purification of hydrogen streams by PROX on Cu supported on an Beyer, K. A.; Chapman, K. W.; Chupas, P. J.; Martinez-Arias, A.; Si,
organized mesoporous ceria-modified alumina. Appl. Catal., B 2007, R.; Bolin, T. B.; Liu, We. J.; Senanayake, S. D. Unraveling the
72, 149−156. Dynamic Nature of a CuO/CeO2 Catalyst for CO Oxidation in
(12) Liu, Z. G.; Zhou, R. X.; Zheng, X. M. Influence of preparation Operando: A Combined Study of XANES (Fluorescence) and
methods on CuO-CeO2 catalysts in the preferential oxidation of CO DRIFTS. ACS Catal. 2014, 4, 1650−1661.
in excess hydrogen. J. Nat. Gas Chem. 2008, 17, 125−129. (28) Zeng, S. H.; Wang, Y.; Ding, S. P.; Sattler, J. J. H. B.; Borodina,
(13) Avgouropoulos, G.; Ioannides, T.; Papadopoulou, C.; Batista, E.; Zhang, L.; Weckhuysen, B. M.; Su, H. Q. Active sites over CuO/
J.; Hocevar, S.; Matralis, H. K. A comparative study of Pt/γ-Al2O3, CeO2 and inverse CeO2/CuO catalysts for preferential CO oxidation.
Au/α-Fe2O3 and CuO-CeO2 catalysts for the selective oxidation of J. Power Sources 2014, 256, 301−311.

2192 DOI: 10.1021/acscatal.8b04035

ACS Catal. 2019, 9, 2177−2195
ACS Catalysis Research Article

(29) Zhang, R.; Miller, J. T.; Baertsch, C. D. Identifying the active with unprecedented catalytic performance in CO preferential
redox oxygen sites in a mixed Cu and Ce oxide catalyst by in situ X- oxidation. Catal. Commun. 2013, 32, 47−51.
ray absorption spectroscopy and anaerobic reactions with CO in (47) Cao, J. L.; Wang, Y.; Zhang, T. Y.; Wu, S. H.; Yuan, Z. Y.
concentrated H2. J. Catal. 2012, 294, 69−78. Preparation, characterization and catalytic behavior of nanostructured
(30) Elias, J. S.; Risch, M.; Giordano, L.; Mansour, A. N.; Shao- mesoporous CuO/Ce0.8Zr0.2O2 catalysts for low-temperature CO
Horn, Y. Structure, Bonding, and Catalytic Activity of Monodisperse, oxidation. Appl. Catal., B 2008, 78, 120−128.
Transition-Metal-Substituted CeO2 Nanoparticles. J. Am. Chem. Soc. (48) Vlaic, G.; Fornasiero, P.; Geremia, S.; Kaspar, J.; Graziani, M.
2014, 136, 17193−17200. Relationship between the zirconia-promoted reduction in the Rh-
(31) Elias, J. S.; Artrith, N.; Bugnet, M.; Giordano, L.; Botton, G. A.; loaded Ce0.5Zr0.5O2 mixed oxide and the Zr-O local structure. J. Catal.
Kolpak, A. M.; Shao-Horn, Y. Elucidating the Nature of the Active 1997, 168, 386−392.
Phase in Copper/Ceria Catalysts for CO Oxidation. ACS Catal. 2016, (49) Liu, Z.; Wu, Z.; Peng, X.; Binder, A.; Chai, S.; Dai, S. Origin of
6, 1675−1679. active oxygen in a ternary CuOx/Co3O4-CeO2 catalyst for CO
(32) Polster, C. S.; Nair, H.; Baertsch, C. D. Study of active sites and oxidation. J. Phys. Chem. C 2014, 118, 27870−27877.
mechanism responsible for highly selective CO oxidation in H2 rich (50) Maciel, C. G.; Silva, T. F.; Profeti, L.; Assaf, E. M.; Assaf, J. M.
atmospheres on a mixed Cu and Ce oxide catalyst. J. Catal. 2009, 266, Study of CuO/CeO2 catalyst with for preferential CO oxidation
308−319. reaction in hydrogen-rich feed (PROX-CO). Appl. Catal., A 2012,
(33) Liu, L. Q.; Zhou, F.; Wang, L. G.; Qi, X. J.; Shi, F.; Deng, Y. Q. 431−432, 25−32.
Low-temperature CO oxidation over supported Pt, Pd catalysts: (51) Jung, C. R.; Kundu, A.; Nam, S. W.; Lee, H. Doping effect of
Particular role of FeOx support for oxygen supply during reactions. J. precious metal on the activity of CuO-CeO2 catalyst for selective
Catal. 2010, 274, 1−10. oxidation of CO. Appl. Catal., A 2007, 331, 112−20.
(34) Wang, H.; Qu, Z. P.; Xie, H. B.; Maeda, N.; Miao, L.; Wang, Z. (52) Fan, J.; Wu, X.; Wu, X.; Liang, Q.; Ran, R.; Weng, D. Thermal
Insight into the mesoporous FexCe1‑xO2‑δ catalysts for selective ageing of Pt on low-surface-area CeO2-ZrO2-La2O3 mixed oxides:
catalytic reduction of NO with NH3: Regulable structure and activity. effect on the OSC performance. Appl. Catal., B 2008, 81, 38−48.
J. Catal. 2016, 338, 56−67. (53) Reddy, G. K.; Boolchand, P.; Smirniotis, P. G. Sulfur tolerant
(35) Lendzion-Bielun, Z.; Bettahar, M. M.; Monteverdi, S. Fe- metal doped Fe/Ce catalysts for high temperature WGS reaction at
promoted CuO/CeO2 catalyst Structural characterization and CO low steam to CO ratios-XPS and Mössbauer spectroscopic study. J.
oxidation activity. Catal. Commun. 2010, 11, 1137−1142. Catal. 2011, 282, 258−269.
(36) Sirichaiprasert, K.; Luengnaruemitchai, a.; Pongstabodee, S. (54) Pereira, A. L. C.; Berrocal, G. J. P.; Marchetti, S. G.; Albornoz,
Selective oxidation of CO to CO2 over Cu-Ce-Fe-O composite-oxide A.; de Souza, A. O.; Rangel, M. C. A comparison between the
catalyst in hydrogen feed stream. Int. J. Hydrogen Energy 2007, 32, precipitation and impregnation methods for water gas shift catalysts. J.
915−926. Mol. Catal. A: Chem. 2008, 281, 66−72.
(37) Firsova, A. A.; Ilichev, A. N.; Khomenko, T. I.; Gorobinskii, L. (55) Fujii, T.; de Groot, F.; Sawatzky, G.; Voogt, F. C.; Hibma, T.;
V.; Maksimov, Y. V.; Suzdalev, I. P.; Korchak, V. N. Selective Okada, K. In situ XPS analysis of various iron oxide films grown by
Oxidation of CO in the Presence of Hydrogen on CuO/CeO2 NO2-assisted molecular-beam epitaxy. Phys. Rev. B: Condens. Matter
Catalysts Modified with Fe and Ni Oxides. Kinet. Catal. 2007, 48, Mater. Phys. 1999, 59, 3195−3202.
282−291. (56) Harrison, W. A. Electronic Structure and the Properties of Solids;
(38) Zou, H. B.; Chen, S.; Liu, Z.; Lin, W. Selective CO oxidation W. H. Freeman and Company: San Francisco, 1980.
over CuO-CeO2 catalysts doped with transition metal oxides. Powder (57) Moretti, E.; Storaro, L.; Talon, A.; Lenarda, M.; Riello, P.;
Technol. 2011, 207, 238−244. Frattini, R.; de Yuso, M.; Jiménez-Lopez, L. A.; Rodríguez-Castellon,
(39) Sahoo, T. R.; Armandi, M.; Arletti, R.; Piumetti, M.; Bensaid, C. E.; Ternero, F.; Caballero, A.; Holgado, J. P. Effect of thermal
S.; Manzoli, M.; Panda, S. R.; Bonelli, B. Pure and Fe-doped CeO2 treatments on the catalytic behaviour in the CO preferential oxidation
nanoparticles obtained by microwave assisted combustion synthesis: of a CuO-CeO2-ZrO2 catalyst with a flower-like morphology. Appl.
Physico-chemical properties ruling their catalytic activity towards CO Catal., B 2011, 102, 627−637.
oxidation and soot combustion. Appl. Catal., B 2017, 211, 31−45. (58) Kundakovic, L.; Flytzani-Stephanopoulos, M. Reduction
(40) Zhang, Z. L.; Han, D.; Wei, S.; Zhang, Y. Determination of characteristics of copper oxide in cerium and zirconium oxide
active site densities and mechanisms for soot combustion with O2 on systems. Appl. Catal., A 1998, 171, 13−29.
Fe-doped CeO2 mixed oxides. J. Catal. 2010, 276, 16−23. (59) Liu, W.; Flytzani-Stephanopoulos, M. Total oxidation of
(41) He, F.; Wei, Y.; Li, H.; Wang, H. Synthesis Gas Generation by carbon-monoxide and methane over transition metal fluorite oxide
Chemical-Looping Reforming Using Ce-Based Oxygen Carriers composite catalysts II. catalyst characterization and reaction-kinetics.
Modified with Fe, Cu, and Mn Oxides. Energy Fuels 2009, 23, J. Catal. 1995, 153, 317−332.
2095−2102. (60) Peng, Y.; Shang, L.; Cao, Y.; Waterhouse, G. I.; Zhou, C.; Bian,
(42) Reddy, G. K.; Boolchand, P.; Smirniotis, P. G. Sulfur tolerant T.; Wu, L. Z.; Tung, C. H.; Zhang, T. Copper (I) cysteine complexes:
metal doped Fe/Ce catalysts for high temperature WGS reaction at Efficient earth-abundant oxidation co-catalysts for visible light-driven
low steam to CO ratios-XPS and Mössbauer spectroscopic study. J. photocatalytic H2 production. Chem. Commun. 2015, 51, 12556−
Catal. 2011, 282, 258−269. 12559.
(43) Wang, J.; Han, C.; Gao, X.; Lu, J.; Wan, G.; He, D.; Chen, R.; (61) Liu, P.; Hensen, E. J. Highly efficient and robust Au/
Chen, K.; He, S.; Luo, Y. Rapid synthesis of Fe-doped CuO- MgCuCr2 O4 catalyst for gas-phase oxidation of ethanol to
Ce0.8Zr0.2O2 catalysts for CO preferential oxidation in H2-rich acetaldehyde. J. Am. Chem. Soc. 2013, 135, 14032−14035.
streams: Effect of iron source and the ratio of Fe/Cu. J. Power (62) Avgouropoulos, G.; Ioannides, T.; Matralis, H. Influence of the
Sources 2017, 343, 437−445. preparation method on the performance of CuO-CeO2 catalysts for
(44) Zhang, X.; Tian, P.; Tu, W.; Zhang, Z.; Xu, J.; Han, Y. F. the selective oxidation of CO. Appl. Catal., B 2005, 56, 87−93.
Tuning the Dynamic Interfacial Structure of Copper-Ceria Catalysts (63) Pimenta, M. A.; del Corro, E.; Carvalho, B. R.; Fantini, C.;
by Indium Oxide during CO Oxidation. ACS Catal. 2018, 8, 5261− Malard, L. M. Comparative Study of Raman Spectroscopy in
5275. Graphene and MoS2-type Transition Metal Dichalcogenides. Acc.
(45) Jia, A. P.; Jiang, S. Y.; Lu, J. Q.; Luo, M. F. Study of Catalytic Chem. Res. 2015, 48, 41−47.
Activity at the CuO-CeO2 Interface for CO Oxidation. J. Phys. Chem. (64) Pal, P.; Singha, R. K.; Saha, A.; Bal, R.; Panda, A. B. Defect-
C 2010, 114, 21605−21610. Induced Efficient Partial Oxidation of Methane over Nonstoichio-
(46) Zhao, Z. K.; Bao, T.; Zeng, Y.; Wang, G.; Muhammad, T. metric Ni/CeO2 Nanocrystals. J. Phys. Chem. C 2015, 119, 13610−
Efficient cobalt- manganese oxide catalyst deposited on modified AC 13618.

2193 DOI: 10.1021/acscatal.8b04035

ACS Catal. 2019, 9, 2177−2195
ACS Catalysis Research Article

(65) Li, X.; Quek, X. Y.; Michel Ligthart, D. A. J.; Guo, M.; Zhang, Supported Copper Oxide Cluster Catalyst for CO Oxidation:
Y.; Li, C.; Yang, Q.; Hensen, E. M. CO-PROX reactions on copper Importance of Metal-Support Interaction. ACS Catal. 2017, 7,
cerium oxide catalysts prepared by melt infiltration. Appl. Catal., B 1313−1329.
2012, 123−124, 424−432. (84) Tang, C.; Sun, J.; Yao, X.; Cao, Y.; Liu, L.; Ge, C.; Gao, F.;
(66) Pu, Z. Y.; Liu, X. S.; Jia, A. P.; Xie, Y. L.; Lu, J. Q.; Luo, M. F. Dong, L. Efficient fabrication of active CuO-CeO2/SBA-15 catalysts
Enhanced activity for CO oxidation over Pr- and Cu-Doped CeO2 for preferential oxidation of CO by solid state impregnation. Appl.
catalysts: Effect of oxygen vacancies. J. Phys. Chem. C 2008, 112, Catal., B 2014, 146, 201−212.
15045−51. (85) Du, P. P.; Wang, W. W.; Jia, C. J.; Song, Q. S.; Huang, Y. Y.; Si,
(67) McBride, J. R.; Hass, K. C.; Poindexter, B. D.; Weber, W. H. R. Effect of strongly bound copper species in copper-ceria catalyst for
Raman and x-ray studies of Ce1‑xRExO2‑y, where RE = La, Pr, Nd, Eu, preferential oxidation of carbon monoxide. Appl. Catal., A 2016, 518,
Gd, and Tb. J. Appl. Phys. 1994, 76, 2435−41. 87−101.
(68) Trogadas, P.; Parrondo, J.; Ramani, V. CeO2 surface oxygen (86) Li, K.; Wang, H.; Wei, Y.; Yan, D. Direct conversion of
vacancy concentration governs in situ free radical scavenging efficacy methane to synthesis gas using lattice oxygen of CeO2-Fe2O3 complex
in polymer electrolytes. ACS Appl. Mater. Interfaces 2012, 4, 5098− oxides. Chem. Eng. J. 2010, 156, 512−518.
5102. (87) Bin, F.; Wei, X.; Li, B.; Hui, K. S. Self-sustained combustion of
(69) Kosacki, I.; Suzuki, T.; Anderson, H. U.; Colomban, P. Raman carbon monoxide promoted by the Cu-Ce/ZSM-5 catalyst in CO/
scattering and lattice defects in nanocrystalline CeO2 thin films. Solid O2/N2 atmosphere. Appl. Catal., B 2015, 162, 282−288.
State Ionics 2002, 149, 99−105. (88) Arandiyan, H.; Dai, H.; Deng, J.; Liu, Y.; Bai, B.; Wang, Y.; Li,
(70) Laguna, O. H.; Centeno, M. A.; Boutonnet, M.; Odriozola, J. A. X.; Xie, S.; Li, J. Three-dimensionally ordered macroporous
Fe-doped ceria solids synthesized by the microemulsion method for La0.6Sr0.4MnO3 with high surface areas: Active catalysts for the
CO oxidation reactions. Appl. Catal., B 2011, 106, 621−629. combustion of methane. J. Catal. 2013, 307, 327−339.
(71) Chen, D.; He, D.; Lu, J.; Zhong, L.; Liu, F.; Liu, J.; Yu, J.; Wan, (89) Luo, M. F.; Zhong, Y. J.; Yuan, X. X.; Zheng, X. M. TPR and
G.; He, S.; Luo, Y. Investigation of the role of surface lattice oxygen TPD studies of CuO/CeO2 catalysts for low temperature CO
and bulk lattice oxygen migration of cerium-based oxygen carriers: oxidation. Appl. Catal., A 1997, 162, 121−131.
XPS and designed H2-TPR characterization. Appl. Catal., B 2017, 218, (90) Hadjiivanov, K. I.; Vayssilov, G. N. Characterization of oxide
249−259. surfaces and zeolites by carbon monoxide as an IR probe molecule.
(72) Lu, J.; Hao, H.; Zhang, L.; Xu, Z.; Zhong, L.; Zhao, Y.; He, D.; Adv. Catal. 2002, 47, 307−511.
Liu, J.; Chen, D.; Pu, H.; He, S.; Luo, Y. The investigation of the role (91) Padley, M. B.; Rochester, C. H.; Hutchings, G. J.; King, F.
of basic lanthanum (La) species on the improvement of catalytic FTIR Spectroscopic Study of Thiophene, SO2, and CO Adsorption
activity and stability of HZSM-5 material for eliminating meth- on Cu/Al2O3 Catalysts. J. Catal. 1994, 148, 438−452.
anethiol-(CH3SH). Appl. Catal., B 2018, 237, 185−197. (92) Manzoli, M.; Monte, R. D.; Boccuzzi, F.; Coluccia, S.; Kaspar, J.
(73) Dou, X.; Sabba, D.; Mathews, N.; Wong, L. H.; Lam, Y. M.; CO oxidation over CuOx-CeO2-ZrO2 catalysts: Transient behaviour
Mhaisalkar, S. Hydrothermal Synthesis of High Electron Mobility Zn- and role of copper clusters in contact with ceria. Appl. Catal., B 2005,
doped SnO2 Nanoflowers as Photoanode Material for Efficient Dye- 61, 192−205.
Sensitized Solar Cells. Chem. Mater. 2011, 23, 3938−3945. (93) Hollins, P. The influence of surface defects on the infrared
(74) Sedmak, G.; Hocevar, S.; Levec, J. Kinetics of selective CO spectra of adsorbed species. Surf. Sci. Rep. 1992, 16, 51−94.
oxidation in excess of H2 over the nanostructured Cu0.1Ce0.9O2−y (94) Elias, J. S.; Stoerzinger, K. A.; Hong, W. T.; Risch, M.;
catalyst. J. Catal. 2003, 213, 135−150. Giordano, L.; Mansour, A. N.; Shao-Horn, Y. In Situ Spectroscopy
(75) Li, J.; Han, Y.; Zhu, Y.; Zhou, R. Purification of hydrogen from and Mechanistic Insights into CO Oxidation on Transition-Metal-
carbon monoxide for fuel cell application over modified mesoporous Substituted Ceria Nanoparticles. ACS Catal. 2017, 7, 6843−6857.
CuO-CeO2 catalysts. Appl. Catal., B 2011, 108−109, 72−80. (95) Park, J. W.; Jeong, J. H.; Yoon, W. L.; Rhee, Y. W. Selective
(76) Polster, C. S.; Nair, H.; Baertsch, C. D. Study of active sites and oxidation of carbon monoxide in hydrogen-rich stream over Cu-Ce/γ-
mechanism responsible for highly selective CO oxidation in H2 rich Al2O3 catalysts promoted with cobalt in a fuel processor for proton
atmospheres on a mixed Cu and Ce oxide catalyst. J. Catal. 2009, 266, exchange membrane fuel cells. J. Power Sources 2004, 132, 18−28.
308−319. (96) Mariño, F.; Descorme, C.; Duprez, D. Supported base metal
(77) Lykaki, M.; Pachatouridou, E.; Carabineiro, S. A.C.; Iliopoulou, catalysts for the preferential oxidation of carbon monoxide in the
E.; Andriopoulou, C.; Kallithrakas-Kontos, N.; Boghosian, S.; presence of excess hydrogen (PROX). Appl. Catal., B 2005, 58, 175.
Konsolakis, M. Michalis Konsolakis, Ceria nanoparticles shape effects (97) Inui, T.; Fujimoto, K.; Uchijima, T.; Masai, M. New Aspects of
on the structural defects and surface chemistry: Implications in CO Spillover Effect in Catalysis. Elsevier: Amsterdam, 1993; p 27.
oxidation by Cu/CeO2 catalysts. Appl. Catal., B 2018, 230, 18−28. (98) Gulari, E.; Guldur, C.; Srivannavit, S.; Osuwan, S. Co oxidation
(78) Campbell, C. T.; Peden, C. H. F. Oxygen Vacancies and by silver cobalt composite oxide. Appl. Catal., A 1999, 182, 147.
Catalysis on Ceria Surfaces. Science 2005, 309 (5735), 713−714. (99) Gamarra, D.; Martínez-Arias, A. Preferential oxidation of CO in
(79) Campbell, C. T. Waltzing with O2. Science 2003, 299 (5605), rich H2 over CuO/CeO2: Operando-DRIFTS analysis of deactivating
357. effect of CO2 and H2O. J. Catal. 2009, 263, 189−195.
(80) Liotta, L. F.; Di Carlo, D. G.; Pantaleo, G.; Venezia, A. M.; (100) Park, J. W.; Jeong, J. H.; Yoon, W. L.; Rhee, Y. W. Selective
Deganello, G. Co3O4/CeO2 composite oxides for methane emissions oxidation of carbon monoxide in hydrogen-rich stream over Cu-Ce/γ-
abatement: Relationship between Co3O4-CeO2 interaction and Al2O3 catalysts promoted with cobalt in a fuel processor for proton
catalytic activity. Appl. Catal., B 2006, 66, 217−227. exchange membrane fuel cells. J. Power Sources 2004, 132, 18.
(81) Zheng, Y.; Li, K.; Wang, H.; Wang, Y.; Tian, D.; Wei, Y.; Zhu, (101) Snytnikov, P. V.; Sobyanin, V. A.; Belyaev, V. D.; Tsyrulnikov,
X.; Zeng, C.; Luo, Y. Structure dependence and reaction mechanism P. G.; Shitova, N. B.; Shlyapin, D. A. Selective oxidation of carbon
of CO oxidation: A model study on macroporous CeO2 and CeO2- monoxide in excess hydrogen over Pt-, Ru- and Pd-supported
ZrO2 catalysts. J. Catal. 2016, 344, 365−377. catalysts. Appl. Catal., A 2003, 239, 149.
(82) Wang, J.; Han, C.; Gao, X.; Lu, J.; Wan, G.; He, D.; Chen, R.; (102) LeValley, T. L.; Richard, A. R.; Fan, M. The progress in water
Chen, K.; He, S.; Luo, Y. Rapid synthesis of Fe-doped CuO- gas shift and steam reforming hydrogen production technologies-A
Ce0.8Zr0.2O2 catalysts for CO preferential oxidation in H2-rich review. Int. J. Hydrogen Energy 2014, 39, 16983−17000.
streams: Effect of iron source and the ratio of Fe/Cu. J. Power (103) Reddy, G. K.; Kim, S. J.; Dong, J.; Smirniotis, P. G.; Jasinski, J.
Sources 2017, 343, 437−445. B. Long-term WGS stability of Fe/Ce and Fe/Ce/Cr catalysts at high
(83) Wang, W. W.; Yu, W. Z.; Du, P. P.; Xu, H.; Jin, Z.; Si, R.; Ma, and low steam to CO Ratios-XPS and Mössbauer spectroscopic study.
C.; Shi, S.; Jia, C. J.; Yan, C. H. Crystal Plane Effect of Ceria on Appl. Catal., A 2012, 415−416, 101−110.

2194 DOI: 10.1021/acscatal.8b04035

ACS Catal. 2019, 9, 2177−2195
ACS Catalysis Research Article

(104) Wang, X.; Rodriguez, J. A.; Hanson, J. C.; Gamarra, D.;

Martınez-Arias, A.; Fernandez-Garcıa, M. In Situ Studies of the Active
Sites for the Water Gas Shift Reaction over Cu-CeO2 Catalysts:
Complex Interaction between Metallic Copper and Oxygen Vacancies
of Ceria. J. Phys. Chem. B 2006, 110, 428−434.
(105) Jeong, D. W.; Jang, W. J.; Shim, J. O.; Han, W. B.; Roh, H. S.;
Jung, U. H.; Yoon, W. L. Low-temperature wateregas shift reaction
over supported Cu Catalysts. Renewable Energy 2014, 65, 102−107.
(106) Si, R.; Raitano, J.; Yi, N.; Zhang, L.; Chan, S. W.; Flytzani-
Stephanopoulos, M. Structure sensitivity of the low-temperature
water-gas shift reaction on Cu-CeO2 catalysts. Catal. Today 2012,
180, 68−80.
(107) Wu, Z.; Li, M.; Overbury, S. H. On the structure dependence
of CO oxidation over CeO2 nanocrystals with well-defined surface
planes. J. Catal. 2012, 285, 61−73.
(108) Pozdnyakova, O.; Teschner, D.; Wootsch, A.; Kröhnert, J.;
Steinhauer, B.; Sauer, H.; Toth, L.; Jentoft, F. C.; Knop-Gericke, A.;
Paál, Z.; Schlögl, R. Preferential CO oxidation in hydrogen (PROX)
on ceria-supported catalysts, part I: Oxidation state and surface
species on Pt/CeO2 under reaction conditions. J. Catal. 2006, 237,
1−16.
(109) Nemanich, R. J.; Glass, J. T.; Lucovsky, G.; Shroder, R. E.
Raman scattering characterization of carbon bonding in diamond and
diamondlike thin films. J. Vac. Sci. Technol., A 1988, 6, 1783.
(110) Ferrari, A. C.; Robertson, J. Raman spectroscopy of
amorphous, nanostructured, diamond-like carbon, and nanodiamond.
Philos. Trans. R. Soc. London A 2004, 362, 2477−2512.
(111) Botas, J. A.; Serrano, D. P.; García, A.; de Vicente, J.; Ramos,
R. Catalytic conversion of rapeseed oil into raw chemicals and fuels
over Ni- and Mo-modified nanocrystalline ZSM-5 zeolite. Catal.
Today 2012, 195, 59−70.

2195 DOI: 10.1021/acscatal.8b04035

ACS Catal. 2019, 9, 2177−2195