Total Oxidation of Formaldehyde over MnOx-CeO2 Catalysts: The Effect of Acid Treatment

Article abstract of DOI:10.1021/cs501879j

The effect of acid treatment in mixed MnO_x-CeO₂ samples has been investigated in the catalytic total oxidation of formaldehyde. The acid treatment has no effect on the textural and redox properties of the materials when Mn is stabilized in a MnO_x-CeO₂ solid solution (Mn content below 50%). However, these properties were found to be highly altered by acid treatment when the solubility limit of Mn in the ceria was exceeded (Mn content above 50%). This enabled access to the primary porosity and oxidized the manganese species to a higher oxidation state via a Mn dismutation reaction. As a result, the catalytic activity of pure manganese oxide, after chemical activation, in the oxidation of formaldehyde is greatly improved - at 100 °C, the conversion of formaldehyde is increased by a factor of 5 and the corresponding intrinsic reaction rate by 1.4. Combined in situ surface analysis unambiguously identified formate species as a result of formaldehyde oxidation at room temperature on the chemically activated pure MnO_x. The evolution of various surface species was monitored by increasing the temperature and in situ FTIR, and XPS results provided direct evidence of the desorption of monodentate formate species into formaldehyde and the oxidation of bidentate-bridging formate species. Changes in the average oxidation state of surface manganese confirmed the participation of oxygen from MnO_x in the formation of formate species at room temperature and their transformation into CO₂ and H₂O when increasing the temperature. (Figure Presented).

Full text of DOI:10.1021/cs501879j

Research Article
pubs.acs.org/acscatalysis
Total Oxidation of Formaldehyde over MnOx‑CeO2 Catalysts: The
Effect of Acid Treatment
Jhon Quiroz,† Jean-Marc Giraudon,† Antonella Gervasini,‡ Christophe Dujardin,† Christine Lancelot,†
Martine Trentesaux,† and Jean-Francois Lamonier*,†
̧
†Universite
Cedex, France
‡Universita
́
Lille1, Sciences et Technologies, Unite
́
de Catalyse et Chimie du Solide UMR CNRS 8181, 59655 Villeneuve d’Ascq
̀
degli Studi di Milano, Dipartimento di Chimica, Via Camillo Golgi 19, I-20133 Milano, Italy
S
* Supporting Information
ABSTRACT: The effect of acid treatment in mixed MnOx−CeO2 samples has
been investigated in the catalytic total oxidation of formaldehyde. The acid
treatment has no effect on the textural and redox properties of the materials when
Mn is stabilized in a MnOx−CeO2 solid solution (Mn content below 50%).
However, these properties were found to be highly altered by acid treatment
when the solubility limit of Mn in the ceria was exceeded (Mn content above
50%). This enabled access to the primary porosity and oxidized the manganese
species to a higher oxidation state via a Mn dismutation reaction. As a result, the
catalytic activity of pure manganese oxide, after chemical activation, in the
oxidation of formaldehyde is greatly improvedat 100 °C, the conversion of
formaldehyde is increased by a factor of 5 and the corresponding intrinsic
reaction rate by 1.4. Combined in situ surface analysis unambiguously identified
formate species as a result of formaldehyde oxidation at room temperature on the
chemically activated pure MnOx. The evolution of various surface species was
monitored by increasing the temperature and in situ FTIR, and XPS results provided direct evidence of the desorption of
monodentate formate species into formaldehyde and the oxidation of bidentate-bridging formate species. Changes in the average
oxidation state of surface manganese confirmed the participation of oxygen from MnOx in the formation of formate species at
room temperature and their transformation into CO2 and H2O when increasing the temperature.
KEYWORDS: formaldehyde, catalytic oxidation, volatile organic compound (VOC), MnOx−CeO2 mixed oxides, chemical activation,
indoor pollution
cases.12 However, the replacement of these materials in
environmental processes has to be considered, owing to the
well-known limitation of the use of these elements (e.g., high
price and limited resources).
1. INTRODUCTION
Formaldehyde (HCHO) is a dangerous air pollutant in several
indoor environments, including houses, offices, and industrial
buildings. Long-term exposure to this pollutant can cause
serious health problems,1 making this a critical molecule to be
Bulk and supported transition metal oxides were found to be
low-cost promising materials for low-temperature formaldehyde
eliminated to satisfy stringent environmental regulations and
oxidation.3 For these catalytic systems, the most efficient metal
improve indoor air quality.2 Due to the presence of this
pollutant in indoor environments, the abatement of HCHO at
low temperature has significant practical interest. Among the
different strategies that have been proposed to control HCHO
oxides are MnOx, CeO2, and Co3O4.11,13,14 In a recent study,
improved catalytic activity has been achieved for transition
metal oxides with the addition of other metals.15 Among the
transition metal oxides, MnOx was studied by many research
emission, heterogeneous catalytic oxidation of HCHO into non
groups, due to its low volatility and low toxicity. Manganese
harmful CO2 and H2O has been highlighted as a promising way
to remove formaldehyde with low energy consumption.3
oxides assume a wide range of stoichiometry and crystal phases
(β-MnO2, γ -MnO2, α-Mn2O3, γ -Mn2O3, α-Mn3O4, and
Mn5O8) and show excellent catalytic efficiency for HCHO
oxidation at low temperature. As two important forms of
MnOx, cryptomelane and birnessite have been investigated in
Supported noble metals are typically proposed as efficient
catalysts to oxidize HCHO at low temperature (<150 °C). For
instance, several supported noble-metals catalysts have been
developed for low-temperature formaldehyde oxidation, such as
Pt/TiO2,4,5 Pt/Al2O3,6 Pd/TiO2,7,8 Pt/MnOx-CeO2,9 Pt/
Fe2O3,10 and Au/Co3O4−CeO2.11 Among them, the catalyst
based on supported Pt exhibits superior catalytic activities at
low temperature, and even at room temperature in some
Received: November 26, 2014
Revised: February 19, 2015
© XXXX American Chemical Society
2260
DOI: 10.1021/cs501879j
ACS Catal. 2015, 5, 2260−2269

ACS Catalysis
Research Article
previous reports as suitable catalysts for HCHO oxidation.16,17
More recently, Averlant et al.18 showed that one of the most
important parameters that influences catalytic activities is the
valence state of manganese, and enhanced activities were
achieved over materials with high Mn4+ content. Consequently,
for this application, it is crucial to generate high-valence
manganese oxide materials to improve catalytic properties.
Brenet et al.19 studied the chemical activation of manganese
oxide in the form of a hausmannite phase (Mn3O4) by
treatment in acid medium using sulfuric acid. The influence of
various factors such as acid concentration and temperature have
been specified, and the manganese oxide obtained showed a
high Mn oxidation state, enhanced electrochemical properties,
and improved catalytic activities. Askar et al.20 showed that
starting from Mn2O3 and Mn3O4, a higher valence state of
manganese was also obtained through the formation of MnO2
produced by acid digestion. Depending on the type of inorganic
acid used, the obtained MnO2 structure differed: hydrochloric
acid (HCl) tended to produce the so-called γ variety while
sulfuric acid tended to produce α or γ-MnO2. The authors
explain this result by a proton-assisted rearrangement reaction
in the presence of Mn2+ ions to change γ-MnO2 into α-MnO2.
More recently, Sinha et al.21,22 have synthesized manganese
oxide mesoporous materials by using a template-assisted
method followed by chemical activation under acid treatment.
These materials have a large surface area and enhanced ability
to eliminate volatile organic compounds at low temperature.
Moreover, addition of ceria was found to further enhance the
catalytic activity of MnOx through the formation of a solid
solution, in which the synergistic interaction between
manganese and cerium oxides was explained by an oxygen
transfer mechanism.23,24
In order to obtain materials with high specific surface area
and enhanced catalytic activities, a surfactant-assisted wet-
chemistry route has been proposed for the preparation of these
catalysts. Suib et al.25 have synthesized manganese oxide with
mesoporous structures by oxidation of Mn(OH)2 in the
presence of cetyltrimethylammonium bromide (CTAB); the
solid obtained showed ordered mesoporous structure and
exceptional thermal stability (1000 °C). The synthesis and
characterization of high surface area ceria in the presence of a
cationic surfactant has also been reported by Trovarelli et
al.26,27 These ceria materials showed enhanced textural
properties and thermal resistance compared with those
prepared by conventional routes and also improved oxygen
storage and redox behavior. However, the ordering of ceria
materials was limited to a few regions. More recently, Zou et
al.28 have synthesized MnOx−CeO2 by surfactant-assisted
precipitation using CTAB as a template; the solids show a
high specific surface, and no regular arrangement was observed
by TEM, indicating that ordered mesoporous structures were
not formed in the solid.
temperature-programmed desorption, in situ XPS, and FTIR
studies of a formaldehyde saturated sample have been
performed in order to obtain more insight into HCHO
reactivity with an activated MnOx catalyst.
2. EXPERIMENTAL SECTION
2.1. Catalyst Preparation. Binary oxides (n)MnOx−(1 −
n)CeO2 (n = 0, 0.25, 0.5, 0.75 and 1) were prepared by
coprecipitation of an aqueous solution of Mn and Ce metal
nitrates with sodium hydroxide in the presence of
cetyltrimethylammonium bromide as a surfactant. In a specific
synthesis, stoichiometric quantities of Mn(NO3)2.6H2O (Alfa
Aesar 98%) and Ce(NO3)3.6H2O (Alfa Aesar 99%) were
dissolved in 150 mL of distilled water followed by the addition
of a solution of NaOH (4.8 g NaOH (Sigma-Aldrich ≥98%) in
50 mL of distilled water). The solution containing the resulting
manganese−cerium hydroxide precipitate was added to an
aqueous solution of cetyltrimethylammonium bromide (CTAB,
67 g (Sigma ≥98%) in 150 mL of distilled water). The resulting
mixture was heated to 75 °C and then stirred for 1 h. The final
gel obtained in a sealed beaker was transferred to an oven and
heated for 12 h at 75 °C. The solid residue was filtered, washed
with distilled water, dried in air, and lastly calcined at 500 °C
for 6 h (1 °C·min −1). A part of the calcined sample was treated
with an aqueous solution of H2SO4 (120 mL - 10 mol L−1) by
stirring in a beaker for 1 h. The final product was filtered, and
the residue was washed with water and dried at 105 °C. The
samples were labeled by their atomic composition as follows
(n)MnOx−(1 − n)CeO2 (n = 0, 0.25, 0.5, 0.75 and 1).
2.2. Catalyst Characterization. X-ray diffraction (XRD)
patterns were recorded on a Bruker X-ray diffractometer at
room temperature with Cu Kα radiation (λ = 1.5418 Å). The
data sets were collected in the 5−80° range with a step size of
0.02° and an integration time of 4 s. The diffraction patterns
have been indexed by comparison with the Joint Committee on
Powder Diffraction Standards (JCPDS) files.
Nitrogen adsorption/desorption isotherms were obtained at
77K on a Micromeritics ASAP 2010. Before each analysis, the
samples were pretreated at 120 °C under vacuum for 3h. The
BET and BJH analyses were used to determine the total specific
surface area and the pore size distribution of the samples,
respectively.
Transmission electronic microscopy (TEM) images were
taken on a TECNAI electron microscope operating at an
accelerating voltage of 200 kV equipped with a LaB6 crystal.
Prior to TEM observations, samples were deposited from
ethanolic solution onto holey-carbon copper grids.
Temperature-programmed reduction by H2 (H2-TPR)
measurements were performed on a Micromeritics AutoChem
apparatus; 50 mg of the sample were heated from room
temperature to 800 °C (with a heating rate of 10 °C·min−1)
using a mixture of 5% H2 in argon. These conditions
maintained K and P values at around 116 s and 16 °C,
respectively,29 for all the analyses irrespective of the relative
amounts of Mn and Ce oxides in the sample.
X-ray photoelectron spectroscopy (XPS) experiments were
performed using an AXIS Ultra DLD Kratos spectrometer
equipped with a monochromatized aluminum source (Al Kα =
1486.7 eV) and charge compensation gun. All binding energies
were referenced to the C 1s core level at 285 eV. Simulation of
the experimental photopeaks was carried out using a mixed
Gaussian/Lorentzian peak fit procedure according to the
In the present work, calcined mesoporous MnOx−CeO2
mixed oxides synthesized by a surfactant-assisted wet-chemistry
route are characterized, and the effect of their activation by acid
treatment is described. The different solids have been
extensively characterized by X-ray diffraction (XRD), nitrogen
adsorption−desorption, transmission electron microscopy
(TEM), X-ray photoelectron spectroscopy (XPS), and H2-
temperature-programmed reduction (H2-TPR), in order to
highlight the chemical activation and Mn content effects. Their
catalytic performances have been evaluated through the
complete oxidation of formaldehyde. Concomitantly, HCHO
2261
DOI: 10.1021/cs501879j
ACS Catal. 2015, 5, 2260−2269

ACS Catalysis
Research Article
Table 1. Textural Properties of the Calcined and Acid-Treated Materials
maximum pore size diameter
(nm)
crystallite size (nm)
calcined acid-treated
BET surface area (m2·g−1
)
pore volume (cm3·g−1
)
sample code (n)
calcined
acid-treated
calcined
acid-treated
calcined
acid-treated
1
31.0
5.2
29.0
4.5
20
91
82
160
112
137
75
30
25
30
15
15
5
0.13
0.33
0.23
0.35
0.23
0.26
0.43
0.25
0.36
0.25
0.75
0.5
0.25
0
3
5.3
5.1
100
130
73
30
15
15
4.8
4.8
10.0
10.0
software supplied by CasaXPS. Semiquantitative analysis
accounted for a nonlinear Shirley background subtraction.
2.3. Catalytic Activity Test. The formaldehyde (HCHO)
catalytic oxidation was performed in a fixed bed reactor (i.d. 8−
10 mm), loaded with 200 mg of the catalyst sieved between 100
and 270 mesh. The catalyst was activated in situ at 250 °C for 2
h with 20% vol O2 balanced by He, with a flow rate of 30 mL·
min−1. Gaseous formaldehyde was generated from paraformal-
dehyde (solid) in a permeation tube placed in a permeation
chamber (Dynacalibrator, VICI Metronics, Inc.). The per-
meation device chamber was maintained at a constant
temperature of 100 °C. The HCHO was then mixed with the
carrier gas O2/He leading to a gaseous mixture containing 400
ppm of HCHO, 20% vol O2 balanced by He. The total flow
rate was 100 mL·min−1, and the gas hourly space velocity was
was thermally treated from 25 to 420 °C (10 °C·min−1) under
a flow of Helium (12 mL·min−1).
In situ XPS analyses were performed using a Vacuum
Generators Escalab 220 XL spectrometer equipped with an
aluminum source (Al Kα = 1486.6 eV) and a special catalytic
cell. The HCHO saturated sample was thermally treated at
different temperatures (2 °C·min−1) from 25 to 175 °C under a
flow of argon. Once the required temperature was reached, the
sample was cooled under Ar to 25 °C, and XPS analysis was
then conducted.
3. RESULTS AND DISCUSSION
3.1. Characterization. 3.1.1. Structural, Textural, and
Morphologic Characterizations. The wide-angle powder XRD
patterns of the calcined MnOx−CeO2 samples are shown in
Figure S1a. For low Mn content (n values ranging from 0 to
0.5) samples, all the diffraction peaks can be assigned to a
fluorite structure similar to that of pure ceria (JCPDS 34394).
No peak related to a manganese oxide phase was observed,
probably due to the formation of a Mn−Ce solid solution or
the presence of amorphous manganese oxide. The inset in
Figure S1a shows that the 2θ for the (220) diffraction peak
shifts to higher angles for the (n)MnOx−(1 − n)CeO2 samples,
with increasing n. This shift can be related to the formation of a
MnOx−CeO2 solid solution in which manganese ions (Mn4+
0.054 nm, Mn3+ 0.066 nm, Mn2+ 0.080 nm) replaced Ce4+ ions
(0.094 nm).30 Based on the 2θ maximum value of the (220)
diffraction peak (Figure S1a inset), the solubility limit of
manganese ions into the ceria structure is reached for n = 0.5.
Furthermore, the diffraction patterns of high Mn content
samples showed the crystallization of a Mn3O4 hausmannite
phase (JCPDS 071841) together with the fluorite structure
(JCPDS 34394). For n = 1, in addition to Mn3O4, Mn2O3
(JCPDS 070856) and Mn5O8 (JCPDS 391218) phases were
also detected.
30 000 mL·gcat h−1. The reactor temperature was increased
−1
from 25 to 300 °C (1 °C·min−1). The effluent gas from the
reactor was analyzed online using a MicroGC (Varian)
chromatograph equipped with a thermal conductivity detector.
Separations were carried out using two columnsa 1 m COX
column channel for permanent gas analysis and a 8 m CP-Sil 5
CB column channel for the formaldehyde analysis. In the
presence of Mn-based catalysts, HCHO was selectively
transformed into CO2 independent of the reaction temperature.
The formaldehyde conversion (XHCHO, %) was estimated using
the following equation:
⎛
⎞
[HCOH]
[HCOH] + [CO ]
XHCHO(%) = 1 −
× 100
⎜
⎟
⎠
⎝
2
[HCHO] and [CO2] being the concentrations at time t. The
carbon balance of the reaction system has been checked to be
around 100% for all the conditions and catalysts tested. The
average percent error of the analyses was ca. 10%, evaluated
on repeated measurements.
2.4. Formaldehyde Adsorption and Temperature-
Programmed Surface Reaction. The HCHO adsorption
and the reactant/products desorption from the catalyst surface
were performed in the same apparatus used for the catalytic
test. The solid was first pretreated with argon at 150 °C to
remove water and surface impurities. A HCHO adsorption
curve was obtained at RT to ensure the saturation of the
catalyst surface by using mass spectrometry (Omnistar, GSD
301-Pfeiffer). The catalyst was purged with argon for 2 h to
remove physically adsorbed HCHO, and then the temperature
was increased from RT to 500 °C at a rate of 10 °C·min−1 in a
continuous flow of argon. The gaseous effluents from the
reactor were analyzed by mass spectrometry, HCHO (m/z =
30), CO2 (m/z = 44), CO (m/z = 28), and H2O (m/z = 18).
In situ IR studies were carried out with a Nicolet 460 FT-IR
spectrometer equipped with a MCT detector. The HCHO
saturated sample was pressed into self-supported pellets and
The wide-angle powder XRD patterns of the acid-treated
MnOx−CeO2 samples are displayed in Figure S1b. After acid
treatment, differences were observed only for samples with high
Mn content (n ≥ 0.75): diffraction peaks due to the Mn3O4
hausmannite phase disappeared in favor of the unique presence
of the Mn5O8 monoclinic phase. The Figure S1b inset shows
that after the acid treatment the solid solution is retained for n
≤ 0.5.
Table 1 reveals the crystallite sizes of the calcined and acid-
treated samples estimated from the Scherrer equation applied
to the (111) (2θ = 28.6°) and (220) (2θ = 47.5°) CeO2
diffraction peaks and to the (200) (2θ = 22°) Mn5O8
diffraction peak for the two cerium free samples. For the
calcined samples, the crystallite size was found to be 10 nm for
pure ceria, between 4 and 6 nm for the mixed oxides, and
around 30 nm for pure manganese oxide. After acid treatment,
the crystallite sizes remained unchanged (Table 1).
2262
DOI: 10.1021/cs501879j
ACS Catal. 2015, 5, 2260−2269

ACS Catalysis
Research Article
The BET surface area values of the pure oxides were much
lower than those of the mixed oxides (Table 1). Among the
calcined mixed oxides samples, the largest surface area was
found to be 130 m2·g−1 for the n = 0.25 sample, and this value
is consistent with the lowest crystallite size and the highest pore
volume. Interaction between Ce and Mn oxides may produce
more open and distorted structures when the two oxides are
mixed, resulting in a higher surface area for the binary oxides.
The surface area values of the manganese rich samples (n ≥
0.75) were greatly increased after the acid treatment (Table 1).
After acid treatment, Mn species dismutation can occur
according to the following reaction:19
0.75 samples (Figure 1 and Table 1). This could be explained
by the dissolution of the amorphous manganese species,
unveiling the narrow mesoporous distribution present in the
particles. For the n ≤ 0.5 samples, no significant effect to the
pore size distribution or pore volume was observed after the
H2SO4 treatment.
In order to investigate the effect of the acid treatment on the
sample morphology, TEM analysis was carried out for the n = 1
sample, in which the acid treatment strongly modifies the
surface area and pore size distribution. The calcined sample
consists of sphere-like nanoparticles, as shown in Figure 2a.
The diffraction plane of the MnOx sample can also be observed
by HRTEM (Figure 2a inset); for the calcined sample, the (2 0
Mn3O (s) + 4H+(aq)
4
0) orientation corresponding to Mn5O8 was identified (d200
=
= MnO2(s) + 2Mn2+(aq) + 2H2O(l)
0.4894 nm). After the acid treatment, the presence of smaller
particles together with larger sphere-like nanoparticles is
detected (Figure 2b). Small particles (size around 10 nm)
surround larger particles (∼100 nm), supporting the explan-
ation given above, where dissolution of amorphous manganese
species located between the particles occurs to unveil the
porosity of the larger particles.
Higher magnification images of the sample treated with
sulfuric acid are shown in Figure 2c; the acid treatment strongly
modified the size and shape of the particles from sphere-like
nanoparticles of 60 nm to small plate-like particles of 10 nm.
The particle size distribution is strongly modified after acid
treatment (Figure 2d). Indeed for the calcined and acid-treated
samples, the particle size distribution is centered at around 70
and 10 nm, respectively.
However, in this experiment, the metastable Mn5O8 phase
has been obtained instead of an MnO2 phase. Thus, a
proportion of Mn3+ ions in Mn3O4 is probably oxidized leading
to Mn5O8, stabilizing some Mn2+ ions inherited from Mn3O4.
The dissolution of Mn2+ ions may occur first from the
amorphous phases located between the crystallites, thus
revealing a higher surface area (Figure S2). Such a high
increase of the surface area was not observed for n ≤ 0.5 (Table
1). This discrepancy could be explained by the stabilization of
Mn ions in the mixed oxide. For ceria, all the data from textural
studies are similar (Table 1). This result can be explained by
the slow dissolution of CeO2 in sulfuric acid.31
Figure 1 shows the BJH pore size distribution of the calcined
and acid-treated samples. For the calcined samples, the pore
3.1.2. Redox Properties. Figure 3a shows the H2-TPR
profiles of the calcined MnOx−CeO2 samples. The H2-TPR
profile of pure ceria reveals small peaks between 200 and 400
°C, which could be assigned to the reduction of easily reducible
surface cerium.32 For the n = 0.25 and n = 0.5 samples, the
reduction peaks appear over a rather broad temperature range
(100−500 °C) with overlapping peaks characterized by two
apparent maxima at 240 and 340 °C. Several authors have
reported a similar reduction behavior in MnOx−CeO2.28,33,34
The reduction process at lower temperatures results in the
formation of a Mn−Ce−O solid solution in which the mobility
of oxygen species is greatly enhanced. For the n = 0.75 and n =
1 samples, two main reduction peaks with maxima at around
310 and 460 °C were observed. The low-temperature reduction
peak could be ascribed to the reduction of Mn5O8 and Mn2O3
into Mn3O4, whereas the high-temperature reduction peak
could be assigned to a further reduction of Mn3O4 into MnO.13
Figure 3b shows the H2-TPR profiles of the acid-treated
MnOx−CeO2 samples. The reduction profile at low temper-
ature for the n = 0.25 and n = 0.5 samples disappeared, and a
reduction peak at around 540 °C appeared. The presence of
this peak in the TPR profile of all Ce containing samples
suggested the formation of Ce(SO4)2 through the acid
treatment and reduction of sulfate species by H2 most probably
occurring in this temperature range. A strong impact of acid
treatment is observed for the n = 1 sample. In comparison with
the calcined one, the reduction took place at a much lower
temperature. A higher contribution for the first reduction peak
toward the total reduction profile is also clearly seen,
corresponding to manganese ions in higher oxidation states
(Mn5O8 instead of Mn3O4).
Figure 1. BJH pore size distribution obtained on the (n)MnOx−(1 −
n)CeO2 samples calcined (straight line) and acid-treated (dotted line).
size distribution was broad and centered at a maximum pore
diameter value of 25−30 nm, for the n ≥ 0.5 samples. For the n
= 0.25 sample, the maximum pore diameter was around 15 nm.
After acid treatment, the development of narrower pores and an
increase in pore volume were observed for the n = 1 and n =
Table 2 summarizes the total H2 consumption values
calculated from the H2-TPR profiles. For the calcined and
acid-treated samples, the total H2 consumption increased with
2263
DOI: 10.1021/cs501879j
ACS Catal. 2015, 5, 2260−2269

ACS Catalysis
Research Article
Figure 2. (a) TEM images of calcined n = 1 sample, (b) TEM images of acid-treated n = 1 sample, (c) HRTEM of acid-treated sample, and (d)
particle size distribution of the calcined and acid-treated n = 1 sample from TEM images.
the increase of Mn content in the samples. The acid treatment
mainly affected the pure manganese (n = 1) and pure cerium (n
= 0) samples for which a 39% and 350% increase in H2
consumption is observed. For the n = 1 sample, the Mn average
oxidation number increased from 3 (calcined sample) to 3.4
(acid-treated sample). This result can be correlated to the
presence of a Mn5O8 phase observed by XRD after the acid
treatment. For the n = 0 sample, the increase in H2
consumption could be explained by the reduction of sulfate
species at 540 °C.
3.1.3. XPS Analyses. The Ce 3d XP spectra (not shown)
were first investigated in order to ascertain the oxidation state
of cerium. In the Ce 3d3/2 5/2 spectra, the presence of the six
,
peaks (u, u″, u‴, v, v″, and v‴) and that of the satellite peak
located at 916.7 eV (u‴) in all Ce samples (calcined and acid-
treated) are observed. These peaks are characteristic of the
presence of Ce4+ ions.35 Chemical activation did not affect the
oxidation state of cerium.
Figure 3. H2-TPR profiles of the (n)MnOx−(1 − n)CeO2 samples:
(a) calcined and (b) acid-treated.
Table 2. Quantitative Results from H2-TPR for the
(n)MnOx−(1 − n)CeO2 Samples
Owing to the presence of unpaired electrons in Mn(II),
Mn(III), Mn(IV) species, the electronic spectrum of Mn 2p
exhibits multiplet structures, leading to difficulties in identifying
the chemical state of Mn. The Mn 2p XPS signal (not shown)
in all samples is centered at a BE of around 642 eV. For n =
0.75 and n = 1, considering both calcined and acid-treated
samples, a shoulder at low BE (639.8 eV) is observed. This
additional contribution might correspond to the divalent Mn
ions in Mn5O8 and/or Mn3O4, previously identified by XRD
analysis.
consumed H2 (mmol·g−1
)
sample code (n)
calcined
acid-treated
1
6.2
3.4
2.4
2.1
0.2
8.6
3.9
2.8
2.0
0.9
0.75
0.5
0.25
0
2264
DOI: 10.1021/cs501879j
ACS Catal. 2015, 5, 2260−2269

ACS Catalysis
Research Article
The XPS spectra of the core levels of Mn 3s were used to
estimate the average oxidation state (AOS) of Mn. The Mn
AOS has been calculated using the linear relationship between
the Mn 3s multiplet splitting energy (ΔE) and the AOS of
Mn.36 The values of Mn AOS obtained from the XPS study for
the n = 1 samples are in agreement with those obtained from
H2-TPR analyses. The variation in Mn AOS as a function of
cerium content in the samples is reported in Figure S3. For
calcined samples, the Mn AOS remained almost constant
(around 2.9) with increasing cerium content in the sample. For
the acid-treated samples, the Mn AOS was strongly enhanced
for the n = 0.75 (3.6) and n = 1 (3.9) samples. This result
suggests that the sulfuric acid reacts with the MnOx phases
present in the calcined n = 0.75 and n = 1 samples. The
formation of a Mn−Ce−O solid solution resulting from the
dispersion of manganese species in the lattice inhibits the
oxidation of manganese species once the acid treatment is
carried out.
Table 3. Surface Atomic Composition from XPS Analyses
Mn/(Mn + Ce)
S/(Mn + Ce)
acid-treated
sample code (n)
calcined
acid-treated
1
1.00
0.60
0.25
0.18
1.00
0.59
0.27
0.18
0.03
0.10
0.25
0.42
0.75
0.5
0.25
concentration is observed and can be explained by the
interaction between the sulfate and cerium species.
3.2. Reactivity. 3.2.1. Catalytic Test. Figure S4 shows the
light-off curves obtained for the HCHO conversion as a
function of temperature in the presence of the calcined (Figure
S4a) and acid-treated catalysts (Figure S4b). For the calcined
samples, the order of activity found on the basis of T50
(temperature at 50% of HCHO conversion) values is the
following: n = 0.25 > n = 0.75 > n = 0.50 > n = 1 > n = 0. For
acid-treated samples, a different behavior is observed because
O 1s XP spectra of the calcined samples are shown in Figure
4. The main line between 530 and 529.4 eV is attributed to the
T
50 decreases with increasing Mn loading: n = 1 > n = 0.75 > n
= 0.50 > n = 0.25 > n = 0. Irrespective of the reaction
temperature, for all Mn-containing samples, H2O and CO2 are
the only products observed, whereas in the presence of pure
ceria, CO (selectivity of 30% at 100% conversion), CH3−OH
and HCOOH (not quantified) are also detected.
Figure 5 shows the HCHO conversion at 100 °C as a
function of Mn/(Mn+Ce) molar surface ratio in MnOx−CeO2
Figure 5. HCHO conversion (%) evaluated at 100 °C as a function of
Mn/(Mn + Ce) molar surface ratio in calcined (straight line) and acid-
treated (dotted line) MnOx−CeO2 catalysts.
Figure 4. O 1s XP spectra of the (a) calcined (n)MnOx−(1 − n)CeO2
samples, (b) calcined and acid-treated n = 0.5 samples.
lattice oxygen (OL) in MnOx−CeO2 samples. OL BE decreased
with decreasing Mn content in the sample, owing to the change
in the O2− species environment. The shoulder on the high
binding energy side is due to the presence of chemisorbed
catalysts. The chemical composition of the calcined and acid-
treated catalysts has a significant impact on the formaldehyde
conversion. In the presence of calcined mixed oxides, the
formaldehyde conversion is much better than those of single
oxides. Tang et al.9,38 already showed evidence of the
enhancement in the formaldehyde conversion for the Mn/Ce
mixed oxides compared to pure MnOx and CeO2. For the
calcined mixed oxides, HCHO conversion activity is ranked as
follows: n = 0.25 > n = 0.75 ∼ n = 0.5. This result can be related
to the specific surface, which is largest for the n = 0.25 calcined
sample.
oxygen (O2 , O−, OH−). After chemical activation, the
−
shoulder becomes relatively more pronounced (Figure 4b),
and this can be explained by the presence of oxygen from
sulfate species, as sulfur has been observed in acid-treated
samples in the form of sulfates (S 2p BE at 169 eV).37
Surface composition values from XPS data are gathered in
Table 3. The Mn/(Mn + Ce) ratio values remain identical after
chemical activation, confirming the preservation of the solid
solution. A proportional increase in sulfur with increasing Ce
In the presence of acid-treated catalysts, the HCHO
conversion increases monotonically with the Mn content in
2265
DOI: 10.1021/cs501879j
ACS Catal. 2015, 5, 2260−2269

ACS Catalysis
Research Article
the sample (Figure 5). In comparison with calcined samples,
improvement in catalytic performances is observed for the n =
0.75 and n = 1 acid-treated samples. This can be related to the
increase of specific surface area and to the promotion of redox
properties at lower temperature due to the presence of
manganese species in a higher oxidation state. For Ce-rich
acid-treated samples (n = 0, 0.25, and 0.5), the HCHO
conversion decreases in comparison with the parent calcined
samples. Because no change in specific surface area has been
registered for these samples after the acid treatment, the
adverse effect could be explained by surface poisoning by
sulfate species such as Ce(SO4)2, previously shown by XPS and
H2-TPR analyses. This phase is probably catalytically inactive
toward HCHO oxidation.
sample was selected to investigate the adsorption and
transformation of formaldehyde upon heating under inert gas
(argon). Additionally, HCHO temperature-programmed de-
sorption, in situ XPS, and FTIR studies of samples previously
exposed to formaldehyde have been performed.
3.2.2. Formaldehyde Adsorption and Surface Reaction
Experiments. 3.2.2.1. HCHO Temperature-Programmed
Desorption. Figure 7 shows the desorption profiles of different
For a better comparison of activity among catalysts with
different surface areas, the reaction rates of HCHO oxidation
were calculated per unit of surface area at 100 °C. The
evolution observed as a function of Mn/(Mn + Ce) molar
surface ratio is plotted in Figure 6. The synergistic mechanism
Figure 7. HCHO temperature-programmed desorption on the acid-
treated MnOx sample.
product species for the n = 1 acid-treated sample after HCHO
adsorption. HCHO desorption between 50 and 100 °C can be
observed with a maximum value around 50 °C. Carbon dioxide
desorption also takes place in a single broad peak, between 50
and 250 °C, with a maximum at 180 °C. Carbon monoxide is
found to desorb in two stages between 30 and 200 °Cthe
first one occurred in the temperature range of HCHO
desorption, and the second one took place together with
CO2 desorption. Water is also found to desorb from the surface
in two broad peaks with maxima at 100 and 230 °C.
The water production in the low-temperature range can be
related to the production of OH species on the surface, as a
result of the formation of formate species with the participation
of oxygen (“O”) from the manganese oxide (step (1)). Indeed,
the formation of adsorbed formate species (HCOO(ad)) is often
proposed as an intermediate in the oxidation of form-
aldehyde.41,42
Figure 6. Intrinsic reaction rates of HCHO oxidation determined at
100 °C as a function of Mn/(Mn+Ce) molar surface ratio in calcined
(straight line) and acid-treated (dotted line) MnOx−CeO2 catalysts.
in which the interaction between manganese and cerium oxides
improves the oxygen transfer from molecular oxygen toward
active sites of MnO2 through an oxygen reservoir in CeO2 has
been proposed in different catalytic reactions.39,40 However, in
this case, the intrinsic reaction rate monotonically increases
with the Mn content in the solid, showing that (i) Mnn+ species
are more active that Ce4+ species in HCHO oxidation and (ii) a
synergistic effect between the Mn and Ce oxides in the MnOx−
CeO2 does not occur. The promoting effect in HCHO
conversion (Figure 5) observed for the calcined mixed oxides
(n = 0.25, 0.5, and 0.75) is thus mainly due to the increase in
surface area. The intrinsic reaction rate is affected after acid
treatment for n = 0, n = 0.25, and n = 0.5 catalysts, showing the
adverse effect of sulfates in HCHO oxidation. A very high
increase in intrinsic reaction rate is observed for the acid-treated
pure manganese oxide. This result underlines the significance of
the (i) Mn (AOS)XPS increase (from 2.9 to 3.85 after chemical
activation, Figure S3) and (ii) improvement in redox properties
(Figure 3b) for low-temperature HCHO oxidation.
HCHO(g) + 2“O” → HCOO(ad) + OH(ad)
(1)
2OH(ad) → H2O + “O”
(g)
(2)
At higher temperature, the formation of oxidation products
(CO or CO2) could result in the oxidation of formate species
with the participation of oxygen from the MnOx solid (steps
(3) and (4)) and the desorption of water in the reaction
between OH(ad) and H(ad) (step (5)):
HCOO → CO + OH(ad)
(ad)
(g)
(3)
(4)
(5)
HCOO → CO2(g) + H(ad)
(ad)
OH(ad) + H(ad) → H2O
(g)
In order to get more insight into the HCHO interaction with
the catalyst surface, different surface reaction experiments have
been performed. The acid-treated pure manganese oxide
The HCHO-TPD experiment strongly suggests the partic-
ipation of oxygen from manganese oxide in the reaction, giving
2266
DOI: 10.1021/cs501879j
ACS Catal. 2015, 5, 2260−2269

ACS Catalysis
Research Article
rise to the formation of surface oxidation products such as
formate species. In order to gain more understanding, in situ
Fourier transform infrared spectroscopy between RT and 200
°C was carried out on the same sample previously exposed to
formaldehyde.
species adsorbed at the surface disappeared. On the other hand,
total oxidation of HCHO to CO2 (Figure 7) was obtained
between 100 and 250 °C, underlining the specific reactivity of
bidentate-bridging formate species on the acid-treated MnOx
catalyst.
3.2.2.3. In Situ XPS Studies of the Formaldehyde-Exposed
MnOx. The C 1s signal evolution for the samples heated at
different temperatures is shown in Figure 9. The photopeak at
3.2.2.2. In Situ Fourier Transform Infrared Spectroscopy of
the Formaldehyde-Exposed MnOx. IR spectra obtained at
different temperatures are shown in Figure S5. Four intense
bands related to formate species adsorbed on the catalyst
surface41 at 2864 (νCH), 1573 (asym νOCO), 1371 (δCH), and
1355 cm−1 (sym νOCO) are observed together with a large band
near 3300 cm−1 associated with hydroxyl groups in interaction
with adsorbed formate species. This result is in perfect
agreement with step (1). These species are stable up to 215
°C. However, the relative intensity of IR bands at 1371 and
1355 cm−1 changed during the temperature-programmed
surface reaction, which suggested the presence of two different
formate species. In order to appreciate the evolution of the
formate species upon heating, the difference spectra have been
plotted (Figure 8). The bands corresponding to formate species
Figure 8. Difference between successive IR spectra recorded during
the temperature-programmed surface reaction on the acid-treated
MnOx sample (a: 39−21 °C, b: 78−39 °C, c: 121−78 °C, d: 166−121
°C, e: 215−166 °C, f: 266−215 °C, g: 319−266 °C, h: 373−319 °C).
Figure 9. C 1s XP spectra evolution of the acid-treated MnOx sample
after HCHO adsorption and heating under Ar.
lower BE (284.5 eV) is assigned to adventitious carbon (CA).
The intensity of the CA photopeak remains constant
throughout heating. The photopeak at 288.5 eV is attributed
to the presence of formate and/or formaldehyde (CF) species.
Indeed the C 1s BE of adsorbed formaldehyde on the TiO2
(110) surface was determined to be 288.5 eV.45 C 1s BE should
be of the same magnitude for both formate and formaldehyde
species, assuming the molecules are bonded end-on through
the oxygen atom. Between 50 and 100 °C, a strong decrease in
the intensity of the CF photopeak with increasing temperature
is observed (Figure 9). This can be explained by the
formaldehyde and/or monodendate formate species desorption
in this temperature range, observed by HCHO TPD and in situ
FTIR. At higher temperature, the contribution of the CF
photopeak to the C 1s signal continues to decrease, and at
the end, the CF photopeak is less intense than the CA
photopeak. The formate species oxidation into gaseous CO
and/or CO2 (steps (3) and (4)) matches well with this
observation in the temperature range 125−175 °C.
can be clearly seen between 20 and 215 °C. Moreover, the
principal bands related to v(OCO)as are shifted toward lower
wavenumbers from 1621 to 1576 cm−1. The relative intensity of
the bands present between 1300 and 1400 cm−1 also evolved.
Formate can be adsorbed in two forms on the surface (i.e.,
monodentate and bidentate-bridging).42,43 The frequency
separation between ν(OCO)as and ν(OCO)s is employed to
distinguish these two forms.44 The frequency separation at 78
°C was around 264 cm−1, whereas at 215 °C, the frequency
separation was 220 cm−1. When the frequency separation is
higher than 250 cm−1, the monodentate species prevail over the
bidentate-bridging formate species.41 This result suggests that
at low temperature (40−120 °C), the formate species are
bound to the surface in a monodentate configuration, and as
the sample was heated, the bidentate-bridging formate
configuration predominates. A similar trend was underlined
during the interaction of formic acid with amorphous
manganese oxide.43This evolution is in agreement with the
previous HCHO temperature-programmed desorption study.
On the one hand, desorption of HCHO was mainly observed
between 27 and 100 °C (Figure 7) when monodentate formate
Based on the previous XPS study, the main contribution to
the O 1s signal (Figure S6) comes from the lattice oxygen (OL)
at around 530 eV. In comparison with O 1s XP spectra of
2267
DOI: 10.1021/cs501879j
ACS Catal. 2015, 5, 2260−2269

ACS Catalysis
Research Article
MnOx without preadsorbed formaldehyde (Figure 4), the
higher binding energy shoulder appearing at 532 eV (OF) is
much more pronounced after formaldehyde adsorption. This
can be related to the presence of formate and formaldehyde
species. The shoulder (OF) becomes less pronounced with the
increase in treatment temperature, in accordance with the
formaldehyde desorption and formate oxidation.
by the Nord-Pas de Calais Region in France through the
Institute of Research in Industrial Environment (IRENI)
project and by a French−Italian research project (PHC Galilee
2011 no. 25955PK). The TEM and XPS facilities in Lille
(France) are supported by the “Conseil Regional du Nord-Pas
de Calais” and the European Regional Development Fund
(ERDF).
The evolution of Mn AOS during the study is shown in
Figure S7. After HCHO adsorption at RT, the Mn AOS is
decreased from 3.9 (Figure S3) to 3.6 (Figure S7). This
correlates perfectly to the oxidation of formaldehyde into
formate species (step (1)) with the probable participation of
lattice oxygen leading to Mn reduction. The Mn AOS evolution
is in good agreement with formaldehyde and/or formate
desorption without significant Mn AOS change from RT to 75
°C and with the formate oxidation with a decrease in Mn AOS
from 100 to 175 °C (steps (3) and (4)).
REFERENCES
■
(1) Salthamer, T.; Mentese, S.; Marutsky, R. Chem. Rev. 2010, 110,
2536−2572.
(2) Sekine, Y. Atmos. Environ. 2008, 36, 5543−5547.
(3) Quiroz, J.; Royer, S.; Bellat, J.-P.; Giraudon, J.-M.; Lamonier, J.-F.
ChemSusChem 2013, 6, 578−592.
(4) Nie, L. H.; Yu, J. G.; Li, X. Y.; Liu, G.; Jaroniec, M. Environ. Sci.
Technol. 2013, 47, 2777−2783.
(5) Zhang, C. B.; Liu, F. D.; Zhai, Y. P.; Ariga, H.; Yi, N.; Liu, Y. C.;
Asakura, K.; Flytzani-Stephanopoulos, M.; He, H. Angew. Chem., Int.
Ed. 2012, 51, 9628−9632.
4. CONCLUSION
(6) Nie, L. H.; Meng, A. Y.; Yu, J. G.; Jaroniec, M. Sci. Rep. 2013, 3,
Articleno. 3215.
In summary, acid treatment was conducted on MnOx−CeO2
mixed oxides synthesized by surfactant-assisted wet-chemistry.
The substitution of Ce4+ by Mn species in the fluorite structure
to form a solid solution was demonstrated in calcined and acid-
treated samples with a Mn solubility limit of 50% in the fluorite
structure. Textural and redox properties were found to be
strongly altered by acid treatment (10M), especially when the
Mn solubility limit was exceeded. Using H2SO4 as a reactive
agent for the acid treatment, the specific surface area and pore
volume were greatly increased owing to the dissolution of Mn2+
species present between the particles, unveiling the primary
porosity and the oxidation of manganese species in the solid
into a higher oxidation state. Among the calcined samples,
mixed Mn/Ce oxides facilitated the complete oxidation of
formaldehyde at lower temperature in comparison with pure
CeO2 and MnOx oxides, owing to an increase in specific surface
area and enhancement in the redox properties of the mixed
metal oxides solid solutions. Moreover, a mechanistic study of
formaldehyde reactivity on the best catalyst (chemically
activated pure manganese oxide) for formaldehyde oxidation
confirmed the participation of oxygen from manganese oxide
through the formation of both monodentate and bidentate-
bridging formate species, with different reactivity.
(7) Huang, H. B.; Leung, D. Y. C. ACS Catal. 2011, 1, 348−354.
(8) Zhang, C.; Li, Y.; Wang, Y.; He, H. Environ. Sci. Technol. 2014,
48, 5816−5822.
(9) Tang, X. F.; Chen, J. L.; Huang, X. M.; Xu, Y. D.; Shen, W. J.
Appl. Catal., B 2008, 81, 115−121.
(10) An, N.; Wu, P.; Li, S.; Jia, M.; Zhang, W. Appl. Surf. Sci. 2013,
285, 805−809.
(11) Ma, C. Y.; Wang, D. H.; Xue, W. J.; Dou, B. J.; Wang, H. L.;
Hao, Z. P. Environ. Sci. Technol. 2011, 45, 3628−3634.
(12) Zhang, C.; He, H. Catal. Today. 2007, 126, 345−350.
(13) Quiroz, J.; Giraudon, J.-M.; Lamonier, J.-F. Catal. Today. 2011,
176, 277−280.
(14) Liu, X.; Lu, J.; Qian, K.; Huang, W.; Luo, M. J. Rare Earths.
2009, 27, 418−424.
(15) Bai, L.; Wyrwalski, F.; Lamonier, J.-F.; Khodakov, A. Y.;
Monflier, E.; Ponchel, A. Appl. Catal., B 2013, 138, 381−390.
(16) Chen, H.; He, J.; Zhang, C.; He, H. J. Phys. Chem. C 2007, 111,
18033−18038.
(17) Zhou, L.; Zhang, J.; He, J.; Hu, Y.; Tian, H. Mater. Res. Bull.
2011, 46, 1714−1722.
(18) Averlant, R.; Royer, S.; Giraudon, J.-M.; Bellat, J.-P.; Bezverkhyy,
I.; Weber, G.; Lamonier, J.-F. ChemCatChem. 2014, 6, 152−161.
(19) Brenet, J.; Beley, M. Electrochim. Acta 1973, 18, 1003−1011.
(20) Askar, M.; Abbas, H. J. Power Sources 1994, 51, 319−330.
(21) Sinha, A.; Suzuki, K.; Takahara, M.; Azuma, H.; Nonaka, T.;
Suzuki, N.; Takahashi, N. J. Phys. Chem. C 2008, 112, 16028−16035.
(22) Sinha, A.; Susuki, K.; Takahara, M.; Azuma, H.; Nonaka, T.;
Fukumoto, K. Angew. Chem., Int. Ed. 2007, 46, 2891−2894.
(23) Imamura, S.; Shono, M.; Okamoto, N.; Hamada, A.; Ishida, S.
Appl. Catal., A 1996, 142, 279−288.
ASSOCIATED CONTENT
* Supporting Information
The following file is available free of charge on the ACS
■
S
Publications website at DOI: 10.1021/cs501879j.
Materials characterization (XRD, XPS and IR) data.
HCHO catalytic conversion vs temperature (PDF)
(24) Tang, X.; Li, Y.; Huang, X.; Xu, Y.; Zhu, H.; Wang, J.; Shen, W.
Appl. Catal., B 2006, 62, 265−273.
(25) Tian, Z.; Tong, W.; Wang, J.; Duan, N.; Krishnan, V.; Suib, S.
Science 1997, 276, 926−930.
AUTHOR INFORMATION
Corresponding Author
*E-mail: jean-francois.lamonier@univ-lille1.fr. Tel.: 33 3 20 33
77 33. Fax: 33 3 20 43 65 61.
■
(26) Terribile, D.; Trovarelli, A.; De Leitenburg, C.; Dolcetti, G.;
Llorca, J. Chem. Mater. 1997, 9, 2676−2678.
(27) Terribile, D.; Trovarelli, A.; Llorca, J.; De Leitenburg, C.;
Dolcetti, G. J. Catal. 1998, 178, 299−308.
(28) Zou, Z.-Q.; Meng, M.; Zha, Y.-Q. J. Phys. Chem. C 2010, 114,
468−477.
Notes
The authors declare no competing financial interest.
(29) Monti, D. A. M.; Baiker, A. J. Catal. 1983, 83, 323−335.
(30) Machida, M.; Uto, M.; Kurogi, D.; Kijima, T. Chem. Mater.
2000, 12, 3158−3164.
ACKNOWLEDGMENTS
■
The authors are grateful to Olivier Gardoll (Lille1 University,
France) for the H2-TPR analyses. J.Q. thanks FUNDAYACU-
CHO (Caracas, Venezuela) and the French Embassy for the
fellowship awarded. This research has been partially supported
(31) Um, M.; Miyake, M.; Hirato, T. In Green Energy and Technology;
Yao, T., Ed.; Springer: Berlin, 2011; p 165.
(32) Laachir, A.; Perrichon, V.; Badri, A.; Lamotte, J.; Catherine, E.;
́ ́ ́
Lavalley, J. C.; El Fallah, J.; Hilaire, L.; Le Normand, F.; Quemere, E.;
2268
DOI: 10.1021/cs501879j
ACS Catal. 2015, 5, 2260−2269

ACS Catalysis
Research Article
Sauvion, G. N.; Touret, O. J. Chem. Soc. Faraday Trans. 1991, 87,
1601−1609.
(33) Xingyi, W.; Qian, K.; Dao, L. Appl. Catal., B 2009, 86, 166−175.
(34) Chen, H.; Sayari, A.; Adnot, A.; Larachi, F. Appl. Catal., B 2001,
32, 195−204.
(35) Abi-aad, E.; Bechara, R.; Grimblot, J.; Aboukais, A. Chem. Mater.
1993, 5, 793−797.
(36) Gao, T.; Norby, P.; Krumeich, F.; Okamoto, H.; Nesper, R.;
Fjellvag, H. J. Phys. Chem. C 2010, 114, 922−928.
̊
(37) Smirnov, M. Y.; Kalinkin, A. V.; Pashis, A. V.; Prosvirin, I. P.;
Bukhtiyarov, V. I. J. Phys. Chem. C 2014, 118, 22120−22135.
(38) Tang, X.; Chen, J.; Li, Y.; Xu, Y.; Shen, W. Chem. Eng. J. 2006,
118, 119−125.
(39) Imamura, S.; Dol, A.; Ishido, S. Ind. Eng. Chem. Prod. Res. Dev.
1985, 24, 75−80.
(40) Qi, G.; Yang, R.; Chang, R. Appl. Catal., B 2004, 51, 93−106.
(41) Li, C.; Domen, K.; Maruya, K.; Onishi, T. J. Catal. 1990, 125,
445−455.
(42) Mao, C.; Vannice, A. J. Catal. 1995, 154, 230.
(43) Durand, J. P.; Senanayake, S. D.; Suib, S. L.; Mullins, D. R. J.
Phys. Chem. C 2010, 114, 20000−20006.
(44) Busca, G.; Lorenzelli, V. Mater. Chem. 1982, 7, 89−126.
(45) Yuan, Q.; Wu, Z.; Jin, Y.; Xu, L.; Xion, F.; Ma, Y.; Huang, W. J.
Am. Chem. Soc. 2013, 135, 5212−5219.
2269
DOI: 10.1021/cs501879j
ACS Catal. 2015, 5, 2260−2269