The influence of Ba addition on thermal stability and catalytic activity of Cu-based mixed oxide

Article abstract of DOI:10.1016/j.cattod.2020.06.052

Catalytic reduction of NO by CO was studied over La_2-xBa_xCuO₄ (x = 0.0 or 0.4) mixed oxides prepared by coprecipitation method. The catalysts were characterized by X ray diffraction (XRD), O₂ temperature-programmed desorption (O₂-TPD), H₂ temperature-programmed reduction (H₂-TPR) and X-ray absorption near edge structure (XANES). NO + CO reaction was carried out in a fixed-bed flow reactor, from room temperature to 500 °C, at atmospheric pressure. Both fresh catalysts were active for the NO + CO reaction but when were submitted to thermal aging, the barium free catalyst significantly lost its activity. The partial substitution of La by Ba enhanced the thermal stability and although the activity of the catalyst has decreased, it remains high, approaching 100 % at 500 °C. These results can be associated with reducible copper species observed by XANES in the thermal aged catalyst containing Ba.

Full text of DOI:10.1016/j.cattod.2020.06.052

Catalysis Today xxx (xxxx) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
The influence of Ba addition on thermal stability and catalytic activity of Cu-
based mixed oxide
Karen Gonçalves Rachele, Luz Amparo Palacio, Fatima Maria Zanon Zotin*
Programa de Pós-graduação em Engenharia Química, UERJ, Instituto de Química, Rua São Francisco Xavier, 524 – Maracanã, Rio de Janeiro, CEP 20550-900, RJ, Brazil
A R T I C L E I N F O
A B S T R A C T
Keywords:
x 4
Catalytic reduction of NO by CO was studied over La2-xBa CuO (x = 0.0 or 0.4) mixed oxides prepared by
La
Barium
NO reduction
2
CuO
4
coprecipitation method. The catalysts were characterized by X ray diffraction (XRD), O2 temperature-pro-
grammed desorption (O -TPD), H temperature-programmed reduction (H -TPR) and X-ray absorption near
2
2
2
edge structure (XANES). NO + CO reaction was carried out in a fixed-bed flow reactor, from room temperature
to 500 °C, at atmospheric pressure. Both fresh catalysts were active for the NO + CO reaction but when were
submitted to thermal aging, the barium free catalyst significantly lost its activity. The partial substitution of La
by Ba enhanced the thermal stability and although the activity of the catalyst has decreased, it remains high,
approaching 100 % at 500 °C. These results can be associated with reducible copper species observed by XANES
in the thermal aged catalyst containing Ba.
2
N O
Thermal stability
1. Introduction
emission control for more than 45 years [19]. High activity and thermal
stable structure make these materials appropriate catalysts for large-
Three-way catalyst (TWC) system consists of a complex mixture of
scale applications [20]. In perovskite (ABO
3
) and perovskite-like oxides
noble metals such as Pd, Pt and Rh, promoter oxides and stabilizer
agents [1,2]. The NO reduction by CO is one of the reactions that occur
in this system. However, a high selectivity to N is only achieved at high
2
temperatures (400 °C or higher) and depends on the catalyst formula-
tion and the extent of deactivation. At low temperatures, nitrous oxide
is preferably formed and emitted in the environment, since the catalyst
with A BO structure, the valence of cations of B-site, as well as the
2
4
number of non-stoichiometric oxygen can be controllable by substitu-
tion of cations A or B, without destroying the structure of the matrix
[21]. The perovskite type mixed oxides containing lanthanum have
shown good catalytic properties, furthermore, it is possible to synthe-
size multi-component perovskites through partial cation substitutions
[22,23].
is not effective in converting N
One of the factors contributing to the N
catalysts is the presence of noble metals [3–5]. Studies using commer-
2
O to N
2
under these conditions.
2
O emission from these
La
tion and its structure is an orthorhombic (a =5.363 Å, b =5.409 Å, c
=13.17 Å) distortion of the tetragonal structure of K NiF [24,25]. It
consists of Cu octahedron and La polyhedron coordinated to nine
oxygen atoms [26] and is related to that of the LaCuO perovskite [27],
in which Cu environment is the same, but different for La. Fig. 1 shows
a structural model of La CuO (ICSD 56528) and LaCuO (ICSD 9428),
build with Diamond [28]. As can be seen, the Cu octahedron in the
NiF type structure has more neighboring La atoms than in the per-
ovskite one. Consequently, a displacement of the layer of the perovskite
(LaCuO ) whose structure is rhombohedral, results in a structure with
tetragonal symmetry (La CuO ) [30].
A property of great interest of perovskites or perovskite-type mixed
2 4
CuO catalyst, have already been tested in NOx reduction reac-
2
cial TWC catalyst (bench test) confirmed the N O emission from the
2
4
catalyst under real conditions [6,7].
In order to minimize the nitrous oxide emissions and for economic
reasons, the replacement of noble metals by transition metals in TWC is
highly advisable [8–17]. The use of copper has been investigated since
its electronic structure is similar to that of noble metals, such as silver
and gold, with completely filled d-subshells, in addition to being more
environmentally friendly and lower cost. It was noticed that copper
metal species act as effective active sites for NO chemical adsorption
and reduction [18]. Despite the numerous studies carried out, few of
these catalysts had their selectivity and thermal stability (due to high
operating temperatures [19]) evaluated.
3
2
4
3
K
2
4
3
2
4
−
2-
oxides is just the oxygen ions mobility (O , O ), which depends mainly
on the existence of defects in the catalyst crystalline lattice. As the
Perovskites related materials have been studied for vehicular
⁎
Corresponding author.
E-mail address: fzotin@uerj.com.br (F.M.Z. Zotin).
https://doi.org/10.1016/j.cattod.2020.06.052
Received 11 February 2020; Received in revised form 31 May 2020; Accepted 12 June 2020
0920-5861/ © 2020 Elsevier B.V. All rights reserved.
Please cite this article as: Karen Gonçalves Rachele, Luz Amparo Palacio and Fatima Maria Zanon Zotin, Catalysis Today,
https://doi.org/10.1016/j.cattod.2020.06.052

K.G. Rachele, et al.
Catalysis Today xxx (xxxx) xxx–xxx
Fig. 1. Structural model in the view toward plane 110 of (a) La
2
CuO
4
and (b) LaCuO
3
. Octahedra correspond to Cu, gray balls to La and black balls to O.
oxygen ions are large, they are transported mostly by mechanisms in-
volving vacancies and in a few rare cases by mechanisms involving
interstitial oxygen [29–31]. The increase in the ionic radius can lead to
an increment of the oxygen ion transport due to a decrease in the
oxygen bond length and an increase of vacancy concentration [23,32].
Partial replacement of lanthanum by barium in the A-sites of these
materials is considered a promising alternative not only due to its low
coupled plasma optical emission spectrometry (ICP-OES) technique,
performed using a Spectro brand equipament, model Arcos. The surface
area of the samples was determined using a Micromeritics ASAP 2020
analyzer. Prior to the analysis, the samples were degassed under 50
mTorr vacuum at 300 °C for 2 h. X-ray diffraction patterns were ob-
tained by the powder method and collected on a Rigaku Miniflex II
equipment, using Cu Kα radiation (30 kV/15 mA), step size of 0.05°
−
1
−1
oxidation state (+2), but also for its potential capacity to NO
x
storage
step with step time of 2.0 s step , collected in a 2θ range from 10 to
[
33–35]. Partial replacement by barium in LaMnO perovskites has
3
80°.
been shown to directly affect the redox properties and the lattice
oxygen mobility, promoting oxygen vacancies formation and the rapid
migration of lattice oxygen from the bulk to the surface [36]. It is also
accepted that barium addition can also suppress the catalysts sintering
2 2
O temperature programmed desorption (O -TPD) measurements
were carried out in a multipurpose unit AutoChem 2920 Micromeritics
coupled to a mass spectrometer Pfeiffer Vacuum OmniStar, following
the masses 18, 28, 32 and 44. Initially, 50 mg of the catalyst was pre-
−
1
[
37]. According to Johnson [38] and Cao et al. [39], the alumina sin-
treated at 300 °C for 40 min under He flow at a rate of 10 °C min . The
tering was significantly reduced using barium as additive. Alumina
textural properties were improved with Ba addition in low contents.
Thus, the present study aims to evaluate the influence of La partial
sample was cooled down to room temperature (RT) and then performed
the oxygen adsorption under 5 % O /He at a rate of 10 °C min from
2
−1
room temperature to 800 °C, kept at this temperature for 30 min.
Subsequently, the sample was cooled down under the same gas mixture
flow. Upon reaching room temperature, the catalyst was purged with
He for 30 min. Finally, the sample was heated under He flow at a rate of
2 4
substitution by Ba on La CuO oxide performance in the catalytic re-
duction of NO by CO focusing on activity, selectivity (N
and thermal stability.
2
O formation)
−1
1
0 °C min up to 1000 °C to record the TPD spectra. The desorbed O
quantification was carried out considering a prior calibration with the
mixture containing 5% O /He, using a known volume loop. H tem-
perature-programmed reduction (H -TPR) analyses were also per-
2
2. Experimental
2
2
2
The mixed oxides with nominal formulas La
2
CuO
4
and
formed in the multipurpose unit AutoChem 2920 Micromeritics.
Initially, 100 mg of sample was pretreated in argon flow at 300 °C for
La1.6Ba0.4CuO3.8 (20 % molar replacement of La by Ba) were synthe-
sized by coprecipitation method adapted from the method described by
6
0 min. After cooling down to room temperature, a catalyst reduction
Liu et al. [40]. Two distinct solutions, Na
2
CO
3
/NaOH with Na con-
−1
was performed with a mixture 10 % H
heating rate of 10 °C min
2
/Ar flowing at 30 mL min , at a
−
1
centration equal to 3.0 mol L and the metal nitrates of La/Ba and Cu
−1
up to 500 °C.
−1
with a concentration of 1.0 mol L . The solutions were added drop-
wise and simultaneously to a beaker containing 100 mL of distilled
water. The addition was carried out by dosing pumps and the pH was
controlled to maintain it constant around 10. It was obtained a gelati-
nous precipitate, which was washed with distilled water until reaching
neutral pH and then was washed using 150 mL of ethanol to eliminate
the sodium ions. The obtained precipitate was dried at 75 °C for 12 h.
X-ray absorption near edge structure (XANES) spectra were mea-
sured at the D06A-DXAS beamline of the National Synchrotron Light
Laboratory (LNLS) in order to determine the copper oxidation state
under the reaction conditions. A Si (111) monochromator was used to
select the X-ray beam from the synchrotron light produced by the 1.37
GeV electron storage ring with a maximum stored current of 200 mA.
The Cu K-edge absorption spectra were recorded in the transmission
mode, in a range of photon energy from 8950 to 9150 eV, using a CCD
camera. The reactor consisted of a quartz tube, where the pellet was
placed in a sample holder, positioned perpendicular to the gas flow but
not necessarily percolated by it. The reaction was carried out with a
stoichiometric gas mixture (NO (1%), CO (1%), balance in He); from
2 4
The precursors of La CuO and La1.6Ba0.4CuO3.8 oxides were calcined in
air for 4 h at 650 °C and 700 °C, respectively. Calcination temperatures
were determined according to the thermogravimetric analysis results.
The catalysts were subjected to thermal treatment at 900 °C for 4 h, in
order to evaluate their thermal stability. The fresh catalysts were named
LaCu and LaBaCu and the catalysts submitted to the thermal treatment,
LaCu900 and LaBaCu900.
−1
room temperature to 500 °C at a heating rate of 5 °C min . A pre-
treatment at 150 °C under He stream for 1 h was performed to remove
The chemical composition of the solids was measured by inductively
2

K.G. Rachele, et al.
Catalysis Today xxx (xxxx) xxx–xxx
the physisorbed water in the catalysts. For LaBaCu sample, an experi-
Table 2
Amount of desorbed oxygen and surface areas of the catalysts.
ment using a reducing pretreatment (H
conducted.
2
flow, 500 °C, 1 h) was also
cat−1)
2
Catalyst
Amount of desorbed oxygen (μmol O
2
g
Surface area (m /g)
XANES spectra were normalized using Athena software from the
IFEFFIT software package [41]. Some spectra obtained at intermediate
reaction temperatures were chosen and a linear combination was per-
formed using the initial and final spectra as standards. Thereafter,
XANES spectrum obtained at the end of the reaction was quantified
α (T < 600 °C)
β (T > 600 °C)
LaCu
–
45
–
870
327
948
432
2.5
2.2
0
LaBaCu
LaCu900
LaBaCu900
using the initial spectrum and the standards (CuO, Cu
2
O and metallic
146
0
Cu), in order to determine the quantities of copper species and initial
oxide present in the catalysts at the end of the reaction.
area values generally lower than 10 m2 g−1 [23,40]. The partial re-
placement of La by Ba did not cause significant changes in the surface
area values that remained very low or negligible.
The diffractograms of the perovskite-type mixed oxides are shown
in Fig. 2. For the LaCu sample, only the presence of the orthorhombic
Catalytic tests were carried out in a fixed-bed flow reactor with a
vertical furnace (PID temperature control) and a HP6890 chromato-
graph coupled. External and internal diffusion limitations were ex-
perimentally evaluated with one of the most active catalysts available
in our laboratory. Regarding the internal diffusion effects, catalyst ac-
tivities were measured for several catalyst particle sizes which allowed
us to define a granulometric range free from internal diffusion limita-
tions. For external diffusion limitations, the conversion was measured
for different flowrates at the same space time, which provided us a
flowrate that ensured the absence of external diffusion problems. The
size of the catalyst particles and the flow rates used in the catalytic test
of the present work were chosen in order to guarantee that diffusion
problems did not limit the catalytic processes.
The tests were accomplished with 120 mg of the sample, previously
sieved (40–60 mesh size fraction). The catalysts, in the selected gran-
ulometry, were diluted in silicon carbide in a mass ratio of 1:3. This
procedure minimizes the formation of hot spots inside the reactor and
contributes to the reproducibility of the tests. Prior to the catalytic
activity tests, the catalysts were either pretreated under He stream at
La
2
CuO
4
phase is verified (ICSD 56528). According to the literature,
CuO with a K NiF structure exhibit a tetragonal sym-
although La
2
4
2
4
metry, a structural distortion may occur, which means that an orthor-
hombic and not tetragonal structure can be obtained [24]. After
thermal aging, the LaCu900 sample remained showing a characteristic
2 4
diffractogram of the orthorhombic La CuO , but it also exhibits small
peaks at 35.4, 38.9 and 48.9° (2θ) belonging to CuO (ICSD 291393). It
indicates that occurs a segregation of CuO crystalline phase after cat-
alyst exposure to more severe conditions. The chemical analysis and
XRD results suggest that LaCu900 is a mixture of La
According to the stoichiometry proposed in Eq. (1), the La
2
CuO
4
and CuO.
CuO /CuO
2
4
ratio is 2.3 (0.7/0.3). Therefore, it is possible that these two phases are
already present in the fresh LaCu catalyst, but in this case, CuO would
be amorphous or with crystalline fractions undetectable by XRD.
1
50 °C for 1 h to remove the physisorbed water or subjected to a re-
duction pretreatment (H flow, 500 °C, 1 h). The catalytic activity
evaluation was carried out using a mixture containing 1% CO and 1%
La1.4CuO3.1 → 0.7 La
2
CuO
4
+ 0.3 CuO
(1)
2
The LaBaCu diffractogram shows that the partial substitution of La
by Ba induced a small change in the crystalline phases present in the
mixed oxide. Although the diffractogram remains predominantly
−1
NO in He, under 225 mL min
min
flow and temperature ramp of 2 °C
from room temperature to 500 °C.
−
1
characteristic of orthorhombic La
different peaks are observed in 23.8, 27.6, 34.6 and 46.9° (2θ) char-
acteristic of BaCO (ICSD 15196).
After thermal aging of LaBaCu, the characteristic peaks of BaCO
2 4
CuO perovskite-type mixed oxide,
3
. Results and discussion
3
3
3
.1. Characterization of the catalysts
practically disappeared, but new diffraction peaks appear at 2θ values
of 22.7, 32.4, 39.9, 46.4 and 57.6°, indicating that the thermal aging
promoted the formation of a new phase. This phase was identified as
The chemical analysis of the solids by ICP-OES is presented in
Table 1. In addition to the mass percentages of each element, the molar
ratios are also presented. For LaCu sample, the La/Cu molar ratio is
the tetragonal La1.5Ba1.5Cu2.94
O6.84 phase (ICSD 62923), with a tetra-
gonal YBa Cu structure, that has triple layers of perovskite with
2
3 7
O
equal to 1.40, below the desired value of 2 (for La
2
CuO
4
). Almost the
defects, especially oxygen vacancies [42]. These results indicate that up
to 700 °C part of the Ba was not inserted into the mixed oxide structure,
but with the thermal aging at 900 °C, the Ba atoms from BaCO un-
3
derwent a solid-state reaction giving rise to a La-Ba-Cu mixed oxide.
Therefore, in the LaBaCu900 sample a mixture of the orthorhombic
same ratio (La/Cu) was obtained for LaBaCu sample, but in this case,
the Ba incorporation increased the (La + Ba)/Cu molar ratio, which
reached the value of 1.8. The Ba/La molar ratio was 0.2 instead 0.40
(
the nominal value). From these molar ratios and considering oxidation
state for La, Ba and Cu as +3, +2 and +2, respectively, it was pro-
posed a formula for each solid, showed in Table 1. The results indicated
that larger atoms (La e Ba) are more difficult to incorporate from the
solution to the solid during the coprecipitation process.
La
2
CuO
4
and tetragonal La
x
Ba3-xCu
3
O
7
phases was obtained. Even
6.84 phase, this
though the XRD results showed the La1.5Ba1.5Cu2.94
formula does not necessarily represent the composition of the phase,
but certainly indicates that this crystalline phase corresponds to the
O
The BET surface area values for the fresh samples were less than 10
2 3 7
structure YBa Cu O .
m2
g
−1
as observed in Table 2. Thermal aged samples presented neg-
As previously mentioned, the transport of oxygen ions is of great
importance for catalytic performance of perovskite related materials
and can be associated with the presence of oxygen vacancies [29–31].
ligible area. Previous studies shown that perovskite or perovskite-like
oxides calcined in the temperature range of 650–950 °C present surface
Table 1
Chemical composition of the solids.
Composition (% w/w)
Molar ratio
La/Cu
Proposed formula
La
Cu
Ba
Ba/La
(La + Ba)/Cu
LaCu
LaBaCu
63.6
57.4
20.8
17.5
–
10.1
1.4
1.5
–
0.2
1.4
1.8
La1.4CuO3.1
La1.5Ba0.3CuO3.55
3

K.G. Rachele, et al.
Catalysis Today xxx (xxxx) xxx–xxx
Fig. 2. X-ray diffractograms of the fresh and thermal aged catalysts and patterns. At the botton, stick representation shows the orthorhombic La
2
CuO
4
pattern (ICSD
56528). * is CuO (ICSD 291393), + is BaCO
3
(ICSD 15196) and o is La1.5Ba1.5Cu2.94
O6.84 (ICSD 62923) phases.
2
Studies related to oxygen adsorption and desorption (O -TPD) are used
for this purpose. In general, two types of peaks are exhibited in the
desorption curves, the peak at low temperature is attributed to oxygen
adsorbed on surface vacancies (α-oxygen), and the one observed at
higher temperature (β-oxygen), is related to lattice oxygen or oxygen
species that occupy the inner vacancies originated from cation sub-
stitution in the perovskite [43–46].
According to Fig. 3, LaCu catalyst presented only an intense peak at
8
50 °C attributed to β-oxygen desorption, adsorbed on vacancies in the
bulk structure. This result is in agreement with those obtained by Zhu
et al. [47] in which the perovskite type La CuO mixed oxide did not
2
4
show α-oxygen desorption. However, the partial substitution by Ba in
LaCu caused structural changes, creating oxygen vacancies on the sur-
face of the catalyst. A peak in the range of 200–400 °C, characteristic of
α-oxygen desorption, is observed in LaCuBa, indicating that the Ba
addition led to the formation of vacancies on the catalyst surface.
Seiyama et al. [48] also used O
on the surface and in the bulk of the La1-xSr
2
-TPD technique to characterize defects
CoO type perovskite
x
3
catalysts (obtained from isostructural substitutions of lanthanum by
strontium). For x equal to 0.4, a large increase in α-oxygen desorption
was observed. The authors suggested that the partial substitution of
3
+
La
by a low valence cation (+2) leads to the creation of surface
vacancies in the perovskites by a charge compensation effect.
From Fig. 3, it can still be seen that the partial replacement of La by
Ba was unfavorable to the formation of oxygen vacancies in the bulk,
with a decrease of desorption peak at 850 °C for LaBaCu sample. Studies
2
involving O -TPD in perovskite-type mixed oxides have indicated that
β-oxygen desorption peak might be more associated with B-site cation.
However, the A-site partial replacement might interfere with B-site
electronic configuration and consequently have an influence on β-
oxygen desorption [46,49]. In order to verify the influence of the B
cations nature on the β-oxygen desorption, Seiyama et al. [48] per-
Fig. 3. O
2
-TPD profiles of the catalysts.
formed O
Mn and Co). LaCoO
tributed to large amounts of β-oxygen desorption, while LaCrO
2
-TPD experiments with LaMO
3
perovskites (M = Ni, Cr, Fe,
presented wide peaks at-
and
3
, LaMnO and LaNiO
3
3
3
4

K.G. Rachele, et al.
Catalysis Today xxx (xxxx) xxx–xxx
LaFeO
3
showed no desorption peak. According to the authors, the
lower than that of the first one. Probably, Cu+ species are very stable
oxygen desorption associated with vacancies creation in the bulk is
related to the partial reduction of the B cations, on the surface and
within the perovskite crystalline lattice. Thus, the oxygen desorption
from perovskite mixed oxides might depend on many factors, as the
crystalline structure, oxygen deficiency and reducibility of the B-site
cation [46].
For the aged samples, the thermal treatment did not cause sig-
nificant changes in the unsubstituted LaCu sample (LaCu900), which
maintained the same desorption profile with a single peak at approxi-
mately 850 °C, characteristic of desorbed oxygen of vacancies in the
bulk. However, thermal aging promoted changes in the desorption
profile of the Ba-containing sample. Oxygen desorption peaks are pre-
sent at low (around 400 °C) and high (around 1000 °C) temperature,
related to the formation of both surface and bulk vacancies, respec-
tively. The β-oxygen peak (above 900 °C) can be associated with both,
the oxygen adsorbed in the bulk that desorbed, and the oxygen output
from the crystalline lattice. The thermal aging of LaBaCu sample
even at high temperature and under H
2
atmosphere [47,52–54].
consumption relative to
According to Velasquez et al. [51], the H
2
the first reduction stage, between 230 and 320 °C, might be associated
with other possibilities (eq. 2–6). Thus, in addition to the reduction of
2+
1+
Cu
during the preparation, a partial reduction of La
type phase (LaCuO ) may be occurring (eq. 2). The second peak at 470
°C might be also relative to the delafossite-type phase to Cu° [51].
to Cu
and the reduction of a possible CuO segregated, formed
2
CuO to a delafossite-
4
2
2
La
2
CuO
4
+ H
2
→ 2LaCuO
2
+ La
2
O
3
+ H
2
O (230 a 320 °C)
(2)
(3)
(4)
(5)
(6)
2La CuO + H → Cu O + 2La O + H O (230 a 320 °C)
2
4
2
2
2
3
2
0
CuO + H2 → Cu + H2 (230 a 320 °C)
0
Cu2O + H2 → 2Cu + H2O (420 a 550 °C)
0
2LaCuO + H → 2Cu + La O + H O (420 a 550 °C)
2
2
2
3
2
For LaBaCu catalyst, in addition to the high temperature peak in-
(
x 3 7
LaBaCu900) favored the formation of tetragonal La Ba3-xCu O phase,
terpreted previously, are observed two other H
2
consumption peaks in
as observed by XRD (Fig. 2). The presence of this new crystalline phase
might have contributed to increase the oxygen mobility, since in
YBaCu O structure type the oxygen content easily changes with tem-
3 7
the first reduction stage, below 350 °C. Clearly, different reducible
species might have been formed in the catalyst. As well as for La CuO
sample, it is likely that extra-framework Cu species are also present in
LaBaCu, probably as CuO. These species are reduced at higher tem-
peratures and apparently contribute to the formation of the second peak
2
4
perature [50].
The oxygen quantification in these desorption stages is presented in
Table 2. With thermal aging, the samples showed an increase in the
amount of α- and/or β-oxygen desorbed. Probably, the high tempera-
tures reorganized oxygen atoms in the structure, including the possi-
bility of generating new crystalline phases, in which the surface oxygen
atoms would migrate, releasing vacancies in the surface region. It is
noteworthy that the catalyst LaCu after aging (LaCu900) presented a 9
percent increase in bulk oxygen vacancies. Besides, the barium addition
proved essential for the formation of surface vacancies.
Temperature programmed reduction results are shown in Fig. 4. For
2
LaCu and LaBaCu samples, the TPR profiles showed H consumptions
between 170 and 350 °C and a less intense consumption starting at
approximately 410 °C, remaining to 500 °C, the TPR final temperature.
3
of reduction, at around 300 °C. The presence of BaCO verified by XRD
indicates that not all barium has been incorporated into the perovskite-
type mixed oxide structure. ICP results showed a molar ratio (La + Ba)/
Cu equal to 1.8 (lower than the theoretical ratio equal to 2) confirming
Ba segregation. Thus, the change in the reduction profile is possibly due
to the different Cu species, including those dispersed and in the per-
ovskite-type mixed oxide structure. This differentiated behavior is re-
lated to the Ba addition that also provided a decrease in the initial
temperature of the reduction. This might indicate that the barium
presence promoted the formation of oxygen vacancies at lower tem-
2
peratures, as verified by O -TPD, improving the catalyst redox prop-
erties. Therefore, the partial substitution of lanthanum by barium ap-
pears to increase the surface mobility of lattice oxygens favoring the
LaBaCu surface reducibility, but this effect did not reach the bulk.
The most intense H
2
consumption for LaCu sample occurred in a single
2
+
step, with a single peak attributed to the reduction of Cu
species to
Cu . The last peak at 470 °C is related to the reduction of Cu+ to Cu°
+
The H
sented in Table 3. When comparing the H
to the first peak of LaCu sample (2794 μmol g ) with that of the La-
2
consumption during the reduction of the catalysts are pre-
[
51]. However, H consumption relative to the second peak is much
2
2
consumption corresponding
−
1
−
1
BaCu sample (2091 μmol g ), a consumption decrease is observed,
which is in agreement with the chemical analysis that presented a
higher Cu content for LaCu sample. It is possible to verify that for both
samples, mainly for LaBaCu, the TPR hydrogen consumption is lower
than the theoretical consumption required to reduce all Cu present
−1
−1
(
3265 μmol g
for LaCu and 2751 μmol g
for LaBaCu, calculated
according to the results of ICP-OES). The reduction percentage at the
end of the TPR was 94 % for the LaCu catalyst and 86 % for LaCuBa.
Possibly not all Cu present in the samples is reduced to metallic copper.
2+
Cu
in the CuO form is generally reduced to the metallic state by H
2
below 500 °C, however in the case of perovskite-type mixed oxides, as
Table 3
Quantitative results of H
2
-TPR experiments.
Consumption of H
(μmol g−1)
Catalyst
Maximum
temperature (°C)
2
Peak 1
268
Peak 2
Peak 1 Peak 2 Total Theoretical % red
LaCu
LaCu900
LaBaCu
468
477
2794
2314
2091
2345
283
277
3077 3265
2314 3265
2368 2751
2345 2751
94
71
86
85
a
329
242
a
LaBaCu900 292
a
Despite showing more peaks, the program identified only the main peak for
2
Fig. 4. H -TPR profiles of the catalysts.
both aged samples.
5

K.G. Rachele, et al.
Catalysis Today xxx (xxxx) xxx–xxx
Fig. 5. CuK-edge XANES spectra of the LaBaCu under H
mentioned above, Cu+ may be stable even under high temperatures
2
treatment at the temperature range of 30 to 500 °C.
through the derivative of the absorption curves (see Fig. S2). It can be
seen the peak at 8983 eV include mainly the peak corresponding to
[
47,53,54].
In order to corroborate the TPR results, the reducing treatment was
2
Cu O and a very small part of metallic Cu. In the 8985−8999 eV region
performed and monitored through Cu-K edge XANES for LaBaCu
sample, whose spectra are presented in Fig. 5. It is observed that Cu
only begins to be reduced from 380 °C and the significant reduction
occurs in the temperature range between 380 and 500 °C. This dis-
crepancy in relation to the conventional TPR results is probably due to
the different flow dynamics, since in the TPR reactor the reagents flow
crosses the catalytic bed, whereas in XANES experiments, not all the
flow crosses the pellet, in which the catalyst is diluted with boron ni-
tride. Thus, it is reasonable to expect changes in the reduction kinetics.
Anyway, the evolution of copper oxidation state is clearly observed. A
detailed analysis of the spectra obtained during the reduction process
shows a more intense decrease of the peak at 8999 eV, called peak A.
The wider and less intense peak at 9013 eV, called peak B undergoes
minor changes. According to the literature, these two peaks are related
a CuO contribution also appears. Therefore, it can be concluded that at
+
500 °C the spectrum is mainly composed of Cu species and in minor
proportion, of Cu2+ and Cu° ones.
The XANES results also provided an assessment of copper oxidation
state present in the samples after preparation. It is known that the re-
placement of lanthanum by a lower valence element induces the for-
mation of oxygen vacancies [57], which can change the oxidation state
of one of the metals involved. However, there is also a consensus that
the surface oxygen binding energy is of great importance, because
oxygen with weaker bond strength (labile oxygen) favors the formation
of vacancies [58].
In fact, the introduction of barium in the LaCu catalyst generates an
unbalanced charge in the solid since lanthanum (3+) is replaced by
barium (2+). The XANES spectra of LaBaCu catalyst at room tem-
perature (Fig. S1) are quite similar to the CuO pattern, indicating that
copper is mainly in the 2+ oxidation state. Thus, in the case of this
perovskite-type material, barium promoted the generation of more la-
2
+
to the Cu
[11,55,56].
As the reduction temperature increases, the peak at approximately
+
8
983 eV appears, called peak C, which is attributed to Cu and/or Cu0
species [58].
Aiming for a better understanding of the copper species evolution,
the first spectrum (at room temperature) and the last one (at 500 °C)
were analyzed separately. The sample spectra in the presence of H at
room temperature and at 500 °C were plotted together with CuO, Cu
2
bile oxygen species (as observed by the results of TPD-O ), which also
contributes to the reduction of LaBaCu, as observed by the TPR results.
2
2
O
3.2. Catalytic performance
and Cu foil standard spectra. These spectra are shown in Figs. S1 and S2
of the Supplementary Information. It is possible to note through the
normalized and derivative spectra that the position of the main peak is
the same for CuO and for the sample at room temperature and does not
Catalytic performance results are presented in Fig. 6. The behavior
of CO conversion curves was similar to the NO conversion curves for all
catalysts (the light-off curves for CO conversion are shown in Supple-
mentary Information, Fig. S3), indicating that CO and NO react with
2
match with the other two standards (Cu O and Cu foil). Our results
were compared with the ones of Tranquada et al. [55], who carried out
an exhaustive XAS studies in order to verify the copper valence in
2 2
each other and are converted into N and CO . It was verified that fresh
catalysts (LaCu and LaCuBa) have similar activities. The variations
observed are very small and practically within the experimental error
margin. However, LaCu900 sample reached only 59 % of NO conver-
sion at 500 °C, indicating that it loses activity after thermal aging. The
LaBaCu900 sample showed to be slightly more active at lower tem-
peratures, which can be attributed to the presence of surface vacancies
that provides easily accessible adsorption sites, facilitating the reaction,
compounds with K
2 4 x
NiF type structure, such as La2-x(Ba,Sr) CuO4-y
(
being x from 0 to 0.3). The XANES profiles obtained for these samples
are similar to the LaBaCu of our work. Tranquada et al. [55] compared
the XANES spectra of their samples with calculations and experimental
standards and concluded that the lack of variation in the main peak
2
+
position(in c.a. 8999 eV) of the CuO, Cu
2
ion (CuCl solution) stan-
dards and La2-x(Ba,Sr) CuO4-y samples, was sufficient evidence for
x
as observed by O
observed by the H
2
-TPD results. This behavior is also consistent with that
-TPR data. LaBaCu900 catalyst has been reduced at
9
2+
confirming the 3d configuration of Cu in the samples. The shoulder
in approximately 8987 eV, being stronger in CuO standard than LaBaCu
sample, is a feature of the fine structure, that depends on the radial
2
lower temperatures than the LaCu900. Thus, although the partial sub-
stitution by Ba did not promote an increase in the catalytic activity of
the fresh sample, the Ba addition improved thermal stability as evi-
denced by the LaBaCu900 catalytic behavior, which reached 97 % of
NO conversion, compared to 56 % of barium-free catalyst (LaCu900).
In order to explain the catalysts performance in the reduction re-
action of NO by CO, X-ray absorption near edge structure studies were
4
distances of neighboring atoms. In CuO and LaCuO structures, the
copper is in distorted octahedral coordination, but in CuO the distortion
is higher, inducing this small difference in the XANES profiles.
At 500 °C, however, the Cu reduction can be more easily observed
6

K.G. Rachele, et al.
Catalysis Today xxx (xxxx) xxx–xxx
2
Fig. 6. (a) Light-off curves for NO conversion and (b) N O yield of fresh and
thermal aged catalysts. Pretreatment: He flow at 150 °C for 1 h. Catalytic test:
−
1
−1
1
% CO and 1% NO in He, 225 mL min , 2 °C min
from RT to 500 °C.
Reactor: fixed bed where the sample is in power form.
carried out on freshly prepared and thermal aged samples. The Cu-K
edge XANES spectra obtained at the beginning and at the end of the
Fig. 7. Distribution of copper species (Cu2+, Cu+ and Cu°) in fresh and thermal
aged samples. Pretreatment: He flow at 150 °C for 1 h. Catalytic test: stoi-
chiometric gas mixture 1:1 of NO and CO in He, samples in pellet form, 5 °C
−
1
reaction, following a temperature ramp of 5 °C min up to 500 °C, are
shown in Fig. S4 of Supplementary Information. All XANES spectra
show similar profiles at room temperature, exhibiting a sharp increase
of absorption, giving an intense and narrow peak at 8999 eV and a less
−
1
min
from RT to 500 °C. Reactor: tube where the sample in a pellet form is
arranged in a sample holder.
2
+
intense peak at 9013 eV, related to the Cu [11,55,56]. At the end of
the reaction, (500 °C), the samples had their profile changed, except the
LaCu900 catalyst whose profiles at 500 °C showed little changes. The
intensity of peak A decreased and one shoulder began to appear in the
region of 8982−8985 eV (peak C) [56]. The decrease in peak A in-
dicates a change in the copper electron density, maintaining the same
absence of these species during the reaction. According to Fig. 6(a), the
LaBaCu900 catalytic behavior is completely different from that ob-
served for LaCu900, reaching 97 % of NO conversion at 500 °C even
after thermal aging. In relation to the fresh catalyst (LaBaCu), it also
performs well.
Regarding the formation of reduced species during the reaction
(Fig. 7), the reduction of copper species in LaCuBa900 occurs, but at
higher temperatures compared to LaBaCu, explaining the differences
observed in the catalytic performance of these catalysts. This result is in
accordance with the TPR profiles (Fig. 4). For LaBaCu900, its reduction
starts at temperatures slightly below those of the fresh sample, how-
ever, it stabilizes between 200 and 270 °C and advances only at higher
temperatures (above 270 °C) compared to the reduction profile of the
LaBaCu sample.
structural arrangement. This behavior, togetherwith the appearance of
the peak C, is characteristic of the copper species reduction to Cu+ and/
2
or Cu° [11,55]. These results are in agreement with the H -TPR data,
+
which indicated that Cu can be stable even under high temperatures
in the presence of reducing agent.
The quantification results acquired by linear combination of the
CuO, Cu
Fig. 7. As explained before (H
NiF type structure has particular features slightly different from the
2
O, metallic Cu and the initial catalyst species are shown in
2
reduction monitored by XANES), the
K
2
4
The better performance of LaBaCu900 catalyst in comparison with
CuO, and for this reason, the initial catalyst (at room temperature) was
included in the linear combination.
According to Fig. 7, LaBaCu activity starts at lower temperature,
around 250 °C, while LaCu only at around 350 °C. By increasing the
LaCu900 might be associated with the presence of YBa
2 3 7
Cu O type
structure in the phase La1.5Ba1.5Cu2.94 6.84 (proposed formula), which
O
has higher Ba content than LaBaCu, besides the presence of oxygen
vacancies.
temperature, the distribution of copper species in these catalysts differs
significantly. For LaCu catalyst, only Cu° is formed, whereas Cu+ is
In order to evaluate the nitrous oxide formation during the NO re-
duction by CO, in Fig. 6(b) are shown the N
thermal aged catalysts. All catalysts exhibited a low N
maximum of 10 % at 405 °C for LaCu catalyst. The fresh catalysts have
O production initiated as soon as the NO conversion starts; it reaches
a maximum yield that decreases until N O is no longer produced. For
thermal aged samples, N O is still produced even at 500 °C, but the
2
O yield curves of fresh and
additionally formed in LaBaCu. As the catalytic performance of both
catalysts compared at high temperatures are similar (Fig. 6), this might
indicate that Cu° promotes the reaction.
2
O yield with a
N
2
Some changes in the spectra (Fig. 7) were also observed when
comparing the catalysts before and after thermal aging. The biggest
difference is between LaCu and LaCu900 catalysts, with the latter not
being reduced during the reaction. This means that for LaCu900, the
copper species present in the beginning of the reaction (practically only
2
2
yield values are lower than the fresh samples.
The in-situ XANES results showed that during the reaction, between
2
+
2
+
320 and 420 °C, the fresh samples exhibit mainly the reduction of Cu
Cu ) were maintained up to 500 °C. As the reduced copper species are
more active, the poor performance of LaCu900 is explained by the
+
to Cu and Cu°, which corresponds to the temperature range where
O is formed. Above this range, copper is predominantly present in the
N
2
7

K.G. Rachele, et al.
Catalysis Today xxx (xxxx) xxx–xxx
the reaction conditions, but with the reducing pretreatment, active sites
were generated, justifying the significant improvement in the activity of
the LaCu900 after pretreatment with H
Concerning to the N O generation, when catalysts are subjected to
reducing pretreatment, N O appears at lower temperatures, whereas
2
.
2
2
the NO conversion is also anticipated. Even so, the yield values remain
low (maximum of 13 % for LaCu catalyst) and much lower compared to
those obtained in catalysts based on noble metals. Except for the
LaCu900 sample, all catalysts had nitrogen production finished before
2
500 °C. Considering that N O is an intermediate in the NO reduction by
CO, its formation will depend on the adsorption equilibria on the dif-
ferent copper species as well as on the relative rates of formation and
decomposition to N .
2
A simplified kinetic study was also carried out using the conversion
results of the light-off curves and are described in detail in the
Supplementary Information. A pseudo-first order reaction rate model
was used, which can be a good approximation considering that the
experiments were performed at low partial pressures of NO and CO, as
well as under a fixed NO/CO molar ratio [59]. Although we are aware
that the light-off curves may be affected by eventual changes in the
catalyst over the temperature domain, a good consistency of the results
was verified, indicating that the first order model is a reasonable ap-
proximation. The results regarding the kinetic constant calculated for
each catalyst confirm the conclusions obtained by the light-off curves
for both pretreatments performed. For aged catalysts, the barium-con-
taining catalyst is more active; it has undergone less deactivation which
2
Fig. 8. (a) Light-off curves for NO conversion and (b) N O yield of the fresh and
thermal aged catalysts. Pretreatment: H flow at 500 °C for 1 h. Catalytic test:
2
−
1
−1
1
% CO and 1% NO in He, 225 ml min , 2 °C min
from RT to 500 °C.
Reactor: fixed bed where the sample is in power form.
2
confirms its stabilizing role. For the catalysts pretreated in H , there is a
decrease in the activation energy (Fig. S7(b) and S7(d), and Table S1)
and an increase in the activity of the catalysts after a pretreatment
under H , when compared with the pretreated catalysts at 150 °C in
2
helium. This indicate a change in the nature of the active sites after
treatment with hydrogen, with the formation of reduced copper species
since the beginning of the reaction.
reduced form, and the N
2
O yield decreases.
Regarding to the thermal aged samples, for LaCu900, N
2
O continues
to be produced even at 500 °C (Fig. 6(b)), but the yield values are much
lower than the fresh samples as well as the NO conversions obtained by
the catalysts.
4. Conclusions
3.3. Effect of reducing treatment on catalytic performance
LaCu and LaBaCu fresh catalysts showed good activity in the NO
Previous in situ XANES results had already shown that Cu+/Cu0
reduction by CO reaction. The Ba addition caused structural changes
creating oxygen vacancies on the surface and in the bulk of the catalyst
that promoted the reducibility and activity of the fresh catalysts mainly
at low temperatures. Another important role of barium was the im-
provement of the catalyst thermal resistance, maintaining their re-
species are formed during the NO reduction by CO reaction and remain
present even under high temperatures working as an active site for the
reaction. Therefore, the presence of reduced copper species might be
contributing to the better catalytic performance of the samples, espe-
cially at lower temperatures. Thus, it was decided to carry out a re-
ducing pretreatment prior to the reaction and evaluate the catalyst
performance under these conditions.
Fig. 8(a) shows the results of NO conversion to the catalysts with
reducing pretreatment (the light-off curves for CO conversion are
shown in Supplementary Information, Fig. S5). Comparing Figs. 6 and
ducibility even after thermal aging. Based on in-situ XANES results,
LaBaCu900 presented the higher reducibility with Cu+ and Cu° for-
mation during the reaction. The use of a reducing pretreatment pro-
vided active sites formation, justifying the significant improvement in
catalytic activity. The N
fresh and aged) but in lower yields when compared with noble metal-
based catalysts. The decrease in the activation energy of the pretreated
catalysts at 500 °C under H flow indicates a change in the active sites
presence of reduced copper species since the beginning of the reaction)
2
O generation was also observed in all catalysts
(
8
, the previously reduced samples were more active throughout the
temperature range investigated. In this condition, the fresh catalyst
partially replaced with Ba presented better performance, being active at
lower temperature, just after 100 °C. The most of Cu species in this
2
(
catalyst were reduced to metallic copper after the pretreatment with H
Fig. S2), leading to a collapse of the perovskite-type mixed oxide
2
Author statement
(
structure. Thus, the higher activity of the previously reduced LaBaCu
catalyst, even at lower temperatures, is related to the Cu species present
in the material. At higher temperatures, above 250 °C, these catalysts
showed similar behavior, probably due to the formation of the same
copper species in both samples during the reaction.
All authors declare that they have seen and approved the content of
the submitted manuscript. The paper presents original work not pre-
viously published in similar form and not currently under consideration
by another Journal.
The conception and design of study, acquisition of data, analysis
and/or interpretation of data, drafting the manuscript and critical re-
vision had the participation of all the authors
The thermal aged catalysts also showed a conversion improvement,
+
since, as in the fresh samples, Cu and Cu° were formed after the re-
ducing treatment, working as active sites for the reaction. A more sig-
nificant improvement in catalytic activity was observed for the
LaCu900 catalyst, whose NO conversion starts at lower temperature
than LaBaCu900 and at 500 °C it jumped from 59 to 85 %. According to
the XANES spectra, LaCu900 sample lost its reduction capacity under
Declaration of Competing Interest
The authors declare that they have no known competing financial
8

K.G. Rachele, et al.
Catalysis Today xxx (xxxx) xxx–xxx
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
[23] M.A. Peña, J.L.G. Fierro, Chem. Rev. 101 (2001) 1981–2017.
[
[
24] J.M. Longo, P.M. Raccah, J. Solid State Chem. 6 (1973) 526–531.
25] T. Yokoya, H. Kumigashira, A. Chainani, M. Kasaib, J. Electron Spectros. Relat.
Phenomena 78 (1996) 171–174.
[26] A.F. Wells, Structural Inorganic Chemistry, 4th ed., Clarendon Press, Oxford, 1975.
27] G. Demazeau, C. Parent, M. Pouchard, P. Hagenmuller, Mat. Res. Bull. 7 (1972)
Acknowledgments
[
9
13–920.
We thank to FAPERJ (111.063/2013) by the financial support of
this work. The authors would like to thank the Laboratório Nacional de
Luz Síncroton (Campinas, Brazil)for project approval and financial
support during the DXAS (project 20150337) measurements. Fatima
Zotin thanks CNPqfor the Research Productivity Scholarship.
[
28] H. Putz, K. Brandenburg, Diamond - Crystal and Molecular Structure Visualization,
(2018).
[29] H. Tanaka, M. Misono, Curr. Opin. Solid State Mater. Sci. 5 (2001) 381–387.
[
30] P. Porta, S. De Rossi, M. Faticanti, G. Minelli, I. Pettiti, L. Lisi, M. Turco, J. Solid
State Chem. 146 (1999) 291–304.
31] D. Ferri, L. Forni, Appl. Catal. B Environ. 16 (1998) 119–126.
32] C. De La Calle, A. Aguadero, J.A. Alonso, M.T. Fernández-Díaz, Solid State Sci. 10
[
[
(
2008) 1924–1935.
Appendix A. Supplementary data
[
33] W.S. Epling, J.E. Parks, G.C. Campbell, A. Yezerets, N.W. Currier, L.E. Campbell,
Catal. Today 96 (2004) 21–30.
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.cattod.2020.06.052.
[34] Y. Doi, M. Haneda, M. Ozawa, J. Mol. Catal. A, Chem. 383–384 (2014) 70–76.
[
35] S. Kumar, Y. Teraoka, A.G. Joshi, S. Rayalu, N. Labhsetwar, J. Mol. Catal. A Chem.
48 (2011) 42–54.
36] A.G. Bhavani, W.Y. Kim, J.S. Lee, ACS Catal. 3 (2013) 1537–1544.
3
[
References
[37] L.F. Liotta, G. Deganello, J. Mol. Catal. A Chem. 204–205 (2003) 763–770.
[
[
[
[
[
38] M.F.L. Johnson, J. Catal. 123 (1990) 245–259.
39] Y. Cao, R. Ran, X. Wu, D. Weng, Appl. Catal. A Gen. 501 (2015) 17–26.
40] Y. Liu, H. Zheng, J. Liu, T. Zhang, Chem. Eng. J. 89 (2002) 213–221.
41] M. Newville, J. Synchrotron Radiat. 8 (2001) 322–324.
[
1] P. Kočí, V. Nevoral, M. Záhrubský, M. Kubíček, M. Marek, Chem. Eng. Sci. 59
(2004) 5597–5605.
[2] S. Roy, M. Hegde, G. Madras, Appl. Energy 86 (2009) 2283–2297.
42] S. Sato, I. Nakada, T. Kohara, Y. Oda, H. Daidoji, Acta Crystallogr. Sect. C 45 (1989)
[3] P. Granger, J.P. Dacquin, F. Dhainaut, C. Dujardin, Stud. Surf. Sci. Catal. 171 (2007)
3
47–348.
2
91–324.
[43] N. Russo, D. Fino, G. Saracco, V. Specchia, J. Catal. 229 (2005) 459–469.
[44] N.A. Merino, B.P. Barbero, P. Grange, L.E. Cadús, J. Catal. 231 (2005) 232–244.
[45] N.A. Merino, B.P. Barbero, P. Eloy, L.E. Cadús, Appl. Surf. Sci. 253 (2006)
[
[
4] M. Odaka, N. Koike, H. Suzuki, Chemosphere - Glob. Change Sci. 2 (2000) 413–423.
5] D.M.F. Alves, F.S. Alves, D.G. Ribeiro, S.S. Montani, F.R.L. Faulstich,
M.J.B. Cardoso, C.A. Henriques, F.M.Z. Zotin, Chem. Eng. J. 223 (2013) 239–245.
6] M. Daemme, L. Penteado, R. Zotin, F. Errera, SAE Tech. Pap. Ser. 32–0032 (2014).
7] L.C. Daemme, R. De Arruda Penteado, C. Furlan, M. Errera, F.M.Z. Zotin, SAE Tech.
Pap. 2013 (36–01) (2013).
1
489–1493.
[
[
[
[
46] Z. Zhao, X. Yang, Y. Wu, Appl. Catal. B Environ. 8 (1996) 281–297.
47] J. Zhu, Z. Zhao, D. Xiao, J. Li, X. Yang, Y. Wu, J. Mol. Catal. A Chem. 238 (2005)
3
5–40.
48] T. Selyama, N. Yamazoe, K. Eguchi, Ind. Eng. Chem. Prod. Res. Dev. 24 (1985)
9–27.
[8] Y. Okamoto, H. Gotoh, K. Hishida, H. Aritani, T. Tanaka, S. Yoshida, Appl. Surf. Sci.
[
1
21–122 (1997) 509–512.
1
[9] L. Liu, Z. Yao, B. Liu, L. Dong, J. Catal. 275 (2010) 45–60.
[
[
49] Y. Yokoi, H. Uchida, Catal. Today 42 (1998) 167–174.
50] H. Song, M. Huang, D. Yang, X. Hu, Y. Li, Mater. Sci. Eng. B Solid-State Mater. Adv.
Technol. 137 (2007) 284–288.
[10] Y. Hu, L. Dong, M. Shen, D. Liu, J. Wang, W. Ding, Y. Chen 31 (2001) 61–69.
[11] F. Amano, S. Suzuki, T. Yamamoto, T. Tanaka, Appl. Catal. B Environ. 64 (2006)
2
82–289.
[
[
[
[
51] M. Velasquez, A. Santamaria, C. Batiot-Dupeyrat, Appl. Catal. B Environ. 160-161
[
[
12] L. Ma, M.F. Luo, S.Y. Chen, Appl. Catal. A Gen. 242 (2003) 151–159.
13] V. Parvulescu, P. Grange, B. Delmon, V.I. Pârvulescu, P. Grange, B. Delmon, Catal.
Today 46 (1998) 233–316.
(
2014) 606–613.
52] Y. Zhu, D. Wang, F. Yuan, G. Zhang, H. Fu, Appl. Catal. B Environ. 82 (2008)
55–263.
53] J. Liu, Z. Zhao, C. ming Xu, A. jun Duan, T. Meng, X. jun Bao, Catal. Today 119
2007) 267–272.
54] J. Liu, Z. Zhao, C.M. Xu, A.J. Duan, G.Y. Jiang, J. Phys. Chem. C 112 (2008)
930–5941.
2
[14] D. Lopes, F. Zotin, L.A. Palacio, Appl. Catal. B Environ. 237 (2018) 327–338.
[15] C.D.O.P. Teixeira, S.D.S. Montani, L.A. Palacio, F.M.Z. Zotin, Dalton Trans. 47
(
(
2018) 10989–11001.
16] C.L. Oliveira Corrêa, Y.E. Licea, L. Amparo Palacio, F.M. Zanon Zotin, Catal. Today
89 (2017) 133–142.
[
5
2
[
[
55] J.M. Tranquada, S.M. Heald, A.R. Moodenbaugh, 36 (1987) 5263–5274.
56] S.S. Maluf, P.A.P. Nascente, C.R.M. Afonso, E.M. Assaf, Appl. Catal. A Gen. 413–414
[17] V. Muñoz, F.M.Z. Zotin, L.A. Palacio, Catal. Today 250 (2015) 173–179.
[18] R.L. Martins, M.A.S. Baldanza, M. Schmal, J. Phys. Chem. B 105 (2001)
(2012) 85–93.
1
0303–10307.
[
[
57] L.G. Tejuca, J.L.G. Fierro, J.M.D. Tascón, Adv. Catal. 36 (1989) 237–328.
58] R.J.H. Voorhoeve, J.P. Remeika, L.E. Trimble, A.S. Cooper, F.J. Disalvo,
P.K. Gallagher, J. Solid State Chem. 14 (1975) 395–406.
[
[
[
19] W.F. Libby, Science (80-.) 171 (1971) 499–500.
20] M. Alifanti, M. Florea, V.I. Pârvulescu, Appl. Catal. B Environ. 70 (2007) 400–405.
21] J. Zhu, Z. Zhao, D. Xiao, J. Li, X. Yang, Y. Wu, Ind. Eng. Chem. Res. 44 (2005)
[
59] P. Granger, C. Dujardin, J.F. Paul, G. Leclercq, J. Mol. Catal. A Chem. 228 (2005)
4
227–4233.
2
41–253.
[22] J. Zhu, A. Thomas, Appl. Catal. B Environ. 92 (2009) 225–233.
9