A novel redox-precipitation method for the preparation of α-MnO 2 with a high surface Mn4+ concentration and its activity toward complete catalytic oxidation of o-xylene

Article abstract of DOI:10.1016/j.cattod.2012.04.032

A novel redox-precipitation method was developed for the preparation of α-MnO₂, where Mn(NO₃)₂ and KOH were titrated into excess KMnO₄ solution at pH 8. For comparison, MnO _x was also prepared using a conventional precipitation method. These materials were characterized by XRD, BET, SEM/EDS, XPS, and H₂-TPR techniques, and their catalytic activities were evaluated for the complete catalytic oxidation of a typical volatile organic compound (VOCs), o-xylene. The novel method produced open porous hierarchically structured microcrystalline α-MnO₂ containing almost 100% Mn⁴⁺ ion on its surface. Whereas, the conventional precipitation method produced a mixture of MnO₂ and Mn₃O₄ with a closely packed spherical morphology containing only 31% Mn⁴⁺ ion on its surface. It was found that α-MnO₂ exhibited good low-temperature reducibility, and that it could convert 100% o-xylene into CO₂ at 220 °C at a space velocity of 8000 h^-1, 50 °C lower than the MnO_x prepared by the conventional method. The surface concentration of Mn ⁴⁺ ion in α-MnO₂ played a key role for its high catalytic activity for the complete oxidation of o-xylene. In addition, the open porous structure and the presence of small amount of potassium ion in the microcrystalline α-MnO₂ channel may also be responsible for its excellent catalytic performance. Effects of pH and calcination temperature on its catalytic activity were investigated, and optimal preparation conditions were found. Durability of the α-MnO₂ was also studied.

Full text of DOI:10.1016/j.cattod.2012.04.032

Catalysis Today 201 (2013) 32–39
Contents lists available at SciVerse ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
A novel redox-precipitation method for the preparation of ␣-MnO with a high
2
4+
surface Mn concentration and its activity toward complete catalytic oxidation
of o-xylene
a
b
c
a,∗
a
a
Yinsu Wu , Yao Lu , Chaojie Song , Zichuan Ma , Shengtao Xing , Yuanzhe Gao
a
College of Chemistry and Material Science, Hebei Normal University, Shijiazhuang 050024, China
School of Chemistry and Chemical Engineering, Ningxia University, Yinchuan 750021, China
Institute for Fuel Cell Innovation, National Research Council of Canada, Vancouver, BC V6T 1W5, Canada
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
2 3 2
A novel redox-precipitation method was developed for the preparation of ␣-MnO , where Mn(NO ) and
Received 30 December 2011
Received in revised form 22 March 2012
Accepted 1 April 2012
KOH were titrated into excess KMnO4 solution at pH 8. For comparison, MnOx was also prepared using a
conventional precipitation method. These materials were characterized by XRD, BET, SEM/EDS, XPS, and
H2-TPR techniques, and their catalytic activities were evaluated for the complete catalytic oxidation of
a typical volatile organic compound (VOCs), o-xylene. The novel method produced open porous hierar-
Available online 22 May 2012
4
+
chically structured microcrystalline ␣-MnO2 containing almost 100% Mn ion on its surface. Whereas,
Keywords:
the conventional precipitation method produced a mixture of MnO2 and Mn3O4 with a closely packed
␣
-MnO2
4
+
4+
spherical morphology containing only 31% Mn ion on its surface. It was found that ␣-MnO2 exhib-
High surface concentration of Mn ion
Redox-precipitation method
o-Xylene
◦
ited good low-temperature reducibility, and that it could convert 100% o-xylene into CO2 at 220 C at a
−1
◦
space velocity of 8000 h , 50 C lower than the MnOx prepared by the conventional method. The surface
4
+
Complete catalytic oxidation
concentration of Mn ion in ␣-MnO2 played a key role for its high catalytic activity for the complete oxi-
dation of o-xylene. In addition, the open porous structure and the presence of small amount of potassium
ion in the microcrystalline ␣-MnO2 channel may also be responsible for its excellent catalytic perfor-
mance. Effects of pH and calcination temperature on its catalytic activity were investigated, and optimal
preparation conditions were found. Durability of the ␣-MnO2 was also studied.
©
2012 Elsevier B.V. All rights reserved.
1
. Introduction
physical properties, low cost and environmentally benign features
7,11–13]. Up to now, three types of MnO2 have been used for cat-
alytic combustion of VOCs: pure (bulk) MnO , supported MnO ,
[
Volatile organic compounds (VOCs) such as benzene, toluene,
2
2
and xylene (BTX) are harmful to the environment. Therefore, the
removal of these pollutants is an important research topic for envi-
ronmental treatment systems [1,2]. Complete catalytic oxidation
techniques that convert an organic compound contaminant into
and Mn-containing mixed oxides. For example, Li and co-workers
studied TiO , Al O and SiO2 supported manganese oxide cata-
2
2
3
lysts for the oxidation of chlorobenzene (1300 ppm). They found
that MnOx/TiO2 (Mn loading, 1.9 wt.%) showed the highest cat-
alytic activity and chlorobenzene was completely oxidized at about
CO2 and H O are effective pathways to remove VOCs. Two types
2
◦
of catalysts have been used, alone or in combination, for oxidiz-
ing VOCs pollutants: supported noble metals, and transition metal
oxides. Although supported precious metals (e.g., Pt and Pd) exhib-
ited high performance at lower temperatures [3–6], the high cost
and limited availability restrict their wide application. Thus devel-
opment of efficient transition-metal-oxide-based catalysts for VOC
oxidation has attracted significant interest [7–20].
400 C [13]. Lahousse and co-workers compared the catalytic activi-
ties of commercially available ␥-MnO and Pt/TiO for VOC removal
2
2
and found that ␥-MnO2 could catalyze the complete conversion of
◦
toluene and methanol at 260 C, more active than Pt/TiO2 [15].
It is generally accepted that oxidation of hydrocarbons takes
place via a redox mechanism in which the rate determining step
is oxygen removal from the metal oxide. Thus, the reducibility
of metal oxide species and oxygen removal activity appear to be
two critical parameters determining their catalytic performance for
VOC oxidation [7]. To improve the reducibility and oxygen removal
activity of manganese oxide, binary metal oxide catalysts including
MnOx-CeO2 oxides, MnOx-CuOx oxides, MnOx-CoO and MnOx-ZrO2
oxides have been widely studied [11,16–19]. Li et al. investigated
MnO2 has been considered as an efficient material for
redox catalysis because of its polyvalent, novel chemical and
∗ Corresponding author.
E-mail address: ma zichuan@163.com (Z. Ma).
0
920-5861/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cattod.2012.04.032

Y. Wu et al. / Catalysis Today 201 (2013) 32–39
33
◦
the catalytic activities of Mn-Cu and Mn-Zr mixed oxides prepared
by a reverse microemulsion method. The Cu-Mn catalyst showed
higher activity for complete oxidation of toluene. Mn0.67-Cu0.33 (Cu
samples were calcined in air at 350–600 C for 6 h. The catalyst was
denoted as RP-MnOx (Y–Z), where Y is pH value and Z represents
◦
the calcination temperature in C.
loading 33 mol%) could catalyze the complete oxidation of toluene
For comparison, a MnOx catalyst was also prepared via a con-
ventional precipitation method. Typically, 0.2 M KOH solution was
added dropwise to Mn(NO ) solution under vigorous stirring at
◦
(
0.35 vol.%) at about 220 C [11]. Recently, our group studied the
catalytic performance of MnOx-CeO2 oxides prepared by a redox-
3
2
◦
precipitation method, and found that the MnOx-CeO catalyst could
60 C. The pH of the mixture was kept at 8.0, The precipitate was
2
◦
catalyze the conversion of o-xylene (0.05 vol. %) into CO and H O at
aged for 2 h at 60 C, and then filtered, washed with distilled water
2
2
◦
◦
2
40 C [19]. We found that MnO is the main active phase providing
and dried overnight at 100 C. The dried sample was calcined in air
at 400 C for 6 h. This catalyst was denoted as CP-MnOx (8-400).
2
◦
the oxygen species, and that CeO2 enhanced the oxygen mobil-
ity. The synergistic effects between MnO2 and CeO2 promoted the
redox behavior of the catalyst and significantly improved its activity
toward the complete catalytic oxidation of o-xylene [19].
2.2. Catalyst characterization
The reducibility of pure MnOx is dependent on its oxidation state
and morphology. Generally speaking, Mn ion with high oxidation
state may have high redox activity in which Mn ion is partially
reduced by donating oxygen, and then the reduced Mn ion is reoxi-
dized via the intake of oxygen from the gas phase [21]. Thus, MnOx
containing high content Mn4 ion may exhibit higher catalytic
performance. In addition, porous, hierarchically nanostructured
materials are potentially good candidates for next generation high-
performance catalysts, because of their high surface area and open
porous structure, both of which facilitate gas diffusion and adsorp-
tion [21]. According to our previous study, hierarchically structured
MnO2 could be prepared by controlling the precipitation kinetics
The catalysts were characterized using N2 adsorp-
tion/desorption isotherm, X-ray diffraction (XRD), X-ray
photoelectron spectroscopy (XPS), Scanning electron micro-
graphs (SEM) and Energy dispersive X-ray (EDS). The nitrogen
◦
adsorption–desorption isotherm was obtained at −196 C using
+
a Quantachrome Autosorb-1c instrument. Before measurement,
◦
the samples were degassed at 300 C for 4 h. The surface area
was calculated from the isotherms according to the BET method,
and the average pore size was calculated using the BJH method.
◦
XRD was performed in the 2ꢀ range of 20–80 with a step size of
◦
0
.05 /s using a D8-ADVANCE diffractometer (Germany) operated
at 40 kV and 40 mA with a nickel-filtered Cu K␣ (ꢁ = 0.15418 nm)
nmradiation source. XPS spectra were obtained using a SHIMADZU
ESCA-3400 spectrometer (Mg K␣ radiation). The binding energy
regions investigated were 280–300 eV (C1s), 522–542 eV (O1s),
and 630–670 eV (Mn 2p) using a C1s line (284.8 eV) of adventitious
carbon as a reference. SEM and EDS were taken on a Hitachi, S-4800
microscope. Before the SEM and EDS tests, the powder samples
were spread uniformly on carbon paste on a copper holder and
then coated with gold.
2+
and precursor adsorption in the redox reaction between Mn ions
and KMnO4 in an acidic medium at room temperature without
any templates [22]. However, this hierarchical structure consists
of MnO2 in loose aggregates, and has poor thermal stability, which
is not applicable to complete catalytic oxidation of VOCs. It was
+
reported that a certain amount of K ion could prevent the ␣-MnO2
tunnels from collapsing during the heating process [20,23].
It is also known that besides Mn4 ion, there co-exist Mn and
+
3+
Mn2+ ions in MnOx. However, so far few studies have focused on
4+
the relationship between the surface Mn concentration and its
catalytic activity toward VOCs catalytic oxidation. Also, by now,
complete BTX oxidation catalyzed by pure MnOx can only be
2
.3. H -TPR
2
◦
Temperature programmed reduction (TPR) experiments were
achieved at temperatures above 240 C [15,23]. It remains a big
◦
performed from room temperature up to 600 C under a flow of 5%
challenge to prepare MnOx that can lower down the combustion
temperature of VOCs.
−
1
H /Ar mixture (30 mL min ) over 30 mg of catalyst at a heating rate
of 10 C min . Prior to the TPR experiments, the catalysts were pre-
treated in a 20% O /He mixture at 400 C for 1 h to clean the surface
2
◦
−1
In this paper, a new redox-precipitation method was devel-
◦
oped to synthesize hierarchically structured ␣-MnO with almost
2
2
4
+
of the catalysts. A mass spectrometer (Hiden HPR20) was used for
on-line monitoring of the TPR effluent gas.
1
00% Mn ion, where Mn(NO3)2 and KOH were added into excess
KMnO4 solution at pH 8. For comparison, a MnOx catalyst contain-
4
+
3+
2+
ing Mn , Mn , and Mn ions was also prepared via a conventional
precipitation method. The two catalysts were tested for the com-
plete catalytic oxidation of o-xylene. The effects of the surface Mn4
concentration in MnOx and morphology on their catalytic activity
toward VOCs catalytic combustion were investigated. Other fac-
tors such as pH and calcination temperature were also optimized
for this novel redox-precipitation method.
2.4. Catalytic activity measurement
+
Catalytic activity was measured in a 4 mm i.d. quartz tubular
reactor. 0.25 g of catalyst supported by quartz wool was placed
in the middle of the reactor. A gas containing 700 ppm o-xylene
in simulated air (20 vol.% O , 80 vol.% N ) continuously passed
2
2
−
1
through the catalyst bed with a flow rate of 50 mL min
and
−
1
−1
W/F = 0.3 g s mL
(corresponding to a GHSV of 8000 h ). Here,
2
. Experimental
W/F is defined as the catalyst weight divided by the gas flow rate.
CO2 was the only detectable C-containing reaction product, which
passed through the TDX-01 stainless steel packed column firstly,
and was then converted into methane in a reformer furnace. No
significant amount of any partial oxidation product was detected
in the effluent. The reactant and reaction product were analyzed
with an on-line gas chromatograph equipped with two flame ion-
ization detectors (FID) in series. o-Xylene conversion (Xo-xylene) and
the yield of CO2 (YCO2 ) were calculated according to the following
equations:
2.1. Catalyst preparation
All the chemical reagents were of analytical pure grade and used
as received. The new redox-precipitation route was as follows,
a calculated amount of KMnO4 precursor, in 10% stoichiometric
excess, was dissolved in deionized water. The solution was kept at
◦
6
0 C titrated with 50% Mn(NO ) solution under vigorous stirring.
3
2
The pH of the mixture was kept constant at a value of 6.0, 8.0, and
1
0.0 ± 0.3 respectively by the addition of 0.2 M KOH solution. The
◦
(o-xylenein − o-xyleneout
)
precipitate was aged for 2 h at 60 C, and then filtered, washed
with distilled water and dried overnight at 100 C. The dried
Xo-xylene = 100 ×
(1)
◦
o-xylenein

3
4
Y. Wu et al. / Catalysis Today 201 (2013) 32–39
100
100
80
60
40
20
0
80
60
40
RP-MnOx(8-400)
CP-MnOx
RP-MnOx
CP-MnOx
(
(
(
8-400
8-400
8-400
)
)
)
20
0
160
200
240
280
320
360
o
Fig. 3. XRD patterns of RP-MnOx after heat treatment at varied temperatures of
Temperature( C)
◦
3
50–600 C.
Fig. 1. o-Xylene conversion and CO2 yield over RP-MnOx (8-400) and CP-MnOx (8-
00) catalysts. Reaction conditions: o-xylene 700 ppm, 20% O2/N2 balance, total flow
4
◦
−1
−1
66.2 . For RP-MnOx (8-400), two weak, broad peaks in the 2ꢀ range
rate 50 mL min , W/F = 0.30 g s mL
.
◦
◦
of 35–40 and 65–70 were observed. In order to determine the
phase composition of RP-MnOx (8-400), the as-prepared RP-MnOx
YCO2 = 1
00 × CO
out/8
◦
◦
2
(
8-Z) was calcined at 350 C up to 600 C (see Fig. 3). The crystalin-
ity increases as the heat treatment temperature increase. RP-MnOx
8-350) showed completely amorphous phase, which may be a mix-
(2)
o-xylene
in
(
3
. Results and discussion
.1. Effect of preparation method
Fig. 1 shows o-xylene conversion and CO2 yield versus tem-
ture of manganese oxide and manganese salts. As the calcination
◦
temperature increased to 500 C, relativity sharp diffraction peaks
3
◦
◦
◦
◦
◦
◦
at 2ꢀ = 28.6 , 37.4 , 41.8 , 49.7 , 60.1 and 69.4 appeared, which are
typical diffraction lines of ␣-MnO (JCPDS44-0141). When the sam-
2
◦
ple was further heated to 600 C, the intensities of these ␣-MnO2
perature over RP-MnOx (8-400) and CP-MnOx (8-400). The result
indicates that o-xylene conversion and CO2 yield were temper-
ature dependent and were significantly affected by the catalyst
preparation method. The RP-MnOx (8-400) catalyst showed 100%
peaks enhanced, indicating an increase in degree of crystallization
and crystallite size. It is reasonable to postulate that RP-MnOx (8-
4
00) was composed of microcrystalline structured ␣-MnO . These
2
results also illustrate that the new redox-precipitation method pro-
◦
of o-xylene conversion and CO2 yield at 220 C while the CP-MnOx
duces pure ␣-MnO2. Moreover, this ␣-MnO2 exhibits high phase
(
8-400) catalyst exhibited the same conversion efficiency and CO
◦
2
stability up to 600 C.
◦
◦
yield at 270 C, 50 C higher than the RP-MnOx (8-400) catalyst,
indicating that RP-MnOx (8-400) had higher catalytic activity than
CP-MnOx (8-400). For CP-MnOx (8-400), there is a gap between
The BET surface area and pore volume of RP-MnOx (8-400)
2
−1
3
−1
are 69.42 m g and 0.51 cm g , respectively, significantly larger
2
−1
3
−1
than those of CP-MnOx (8-400) (23.20 m g
and 0.25 cm g ).
◦
the o-xylene conversion and CO2 yield at 240–270 C. Whereas for
However, the pore size of RP-MnOx (8-400) is only 29.51 nm, sig-
nificantly smaller than that of CP-MnOx (8-400) (43.29 nm). These
results indicate that the preparation method greatly affects the
physical properties of these catalysts.
RP-MnOx (8-400), this gap is not as significant as that observed
with CP-MnOx (8-400). These results indicate that RP-MnOx (8-
4
00) exhibit excellent catalytic activity for the complete catalytic
oxidation of o-xylene.
Further insight into the influence of preparation method on
the structural features of the catalysts came from SEM images
of the RP-MnOx and CP-MnOx samples before and after calcina-
The XRD patterns of RP-MnOx (8-400) and CP-MnOx (8-400)
are shown in Fig. 2. On CP-MnOx (8-400), typical diffraction peaks
◦
◦
◦
◦
◦
of Mn O were observed at 2ꢀ of 28.7 , 30.9 , 32.3 , 36.2 , 38.3 ,
◦
3
4
tion at 400 C (Fig. 4A–D). The texture and morphology of the
◦ ◦ ◦ ◦ ◦
4
4.3 , 50.8 , 58.2 , 60.0 and 64.7 (PDF #43-1002), and very weak
diffraction peaks of MnO2 were observed at 37.1 , 47.8 , 55.9 and
as-prepared RP-MnOx are significantly different from that after cal-
◦
◦
◦
◦
cination at 400 C. The as-prepared RP-MnOx consists of a series of
nanosheets assembled into nanoflower shape with diameters of
3
00-500 nm. These nanosheets interweave together forming open
porous structure in the nanoflowers (Fig. 4A). After calcination (RP-
MnOx (8-400)), most of these nanosheets disappear, converting the
nanoflowers into porous nanospheres. In some of the nanoflow-
ers, the nanosheets agglomerate into larger petal-like sheets, and
the nanoflowers agglomerate into larger flower shaped particles,
as shown in Fig. 4B. No significant morphological change was
observed with CP-MnOx samples before (Fig. 4C) and after (Fig. 4D)
calcination. The as-prepared CP-MnOx consists of large and closely
packed spherical particles with particle size of 200–500 nm and
◦
irregularly shaped particles. After calcination at 400 C (CP-MnOx
(8-400)), these particles almost did not change. Fig. 4E and F shows
that the EDS spectra of the as-prepared RP-MnOx and CP-MnOx,
revealing that the as-prepared CP-MnOx is composed of only Mn
and O, whereas the RP-MnOx, consisted of Mn, O, as well as K.
The XRD and SEM results indicate that the composition and mor-
phology of MnOx are dependent on the reaction method. In the
Fig. 2. XRD patterns of RP-MnOx (8-400) and CP-MnOx (8-400).

Y. Wu et al. / Catalysis Today 201 (2013) 32–39
35
◦
◦
Fig. 4. SEM images (A–D) of RP-MnOx and CP-MnOx samples; for RP-MnOx before and after calcination at 400 C (A–B); for CP-MnOx before and after calcination at 400
C–D); EDS (E) patterns of RP-MnOx (8-400); EDS (F) patterns of CP-MnOx (8-400).
C
(
conventional precipitation method, dissolved oxygen is the only
oxidizer in the solution, a mixture of Mn(OH)2 and MnOx might
form quickly, which may grow and agglomerate to form large,
dense spherical particles. After calcination, the mixture was con-
verted into Mn O4 and MnO . In the redox precipitation method,
CP-MnO (8-400)
x
3
2
MnO2 was mainly produced because of the presence of power-
−
ful oxidant MnO4 . In our previous study, we have found that the
desired crystal morphologies can be obtained by manipulating the
precursor adsorption [22]. In the present redox-precipitation pro-
2
+
−
cess, when Mn was dripped into the MnO4 solution at pH 8,
the initially formed MnO was negatively charged. The following
2
2
+
added Mn might be preferentially adsorbed on the most favored
lattice face, along which direction it reacted with MnO4 . Due to the
RP-MnO (8-400)
x
−
anisotropic effect, nanosheets were formed. Then, the nanosheets
assembled into nanoflowers with diameters of 200–500 nm.
1
00
200
300
400
500
600
The H -TPR profiles of the RP-MnOx (8-400) and CP-MnOx
2
o
(
8-400) samples are shown in Fig. 5. It can be seen that the reduc-
Temperature / C
tion behaviors strongly depended on the preparation method. The
H -TPR profile of the CP-MnOx (8-400) sample exhibited a weak
2
Fig. 5. H2-TPR profiles of RP-MnOx (8-400) and CP-MnOx (8-400).

3
6
Y. Wu et al. / Catalysis Today 201 (2013) 32–39
the intense signal at 642.4 eV indicates that there was a high sur-
face atomic concentration of Mn in the RP-MnOx (8-400). For the
Mn 2p3/2
Mn 2p
(A)
4+
CP-MnOx (8-400), a broad, strong Mn 2p peak was observed, sug-
gesting Mn ion exists as more than one oxidation state [19,27].
As discussed above, the XRD and TPR results indicated two types
of manganese oxide existed in CP-MnOx (8-400). Satellite peaks
were also observed with RP-MnOx (8-400) and CP-MnOx (8-400)
at 646.0–647.0 eV, which likely originated from the charge transfer
between the outer electron shell of the ligand and the unfilled 3d
shell of Mn during the creation of the core-hole in the photoelectron
process [27]. Table 1 summarizes the surface compositions of Mn4+
in RP-MnOx (8-400) and CP-MnOx (8-400), calculated from the XPS
spectra. RP-MnOx (8-400) possessed 100% Mn4+ ions, whereas CP-
MnOx (8-400) only has and 31%. Table 1 also demonstrated that in
Mn 2p1/2
CP-MnO (8-400)
x
RP-MnO (8-2-400)
x
4+
3+
2+
CP-MnOx (8-400) Mn exists as Mn , Mn , and Mn . Machocki
et al. reported that high Mn4+ concentration is beneficial for the
catalytic combustion of hydrocarbons [28]. By comparing the cat-
alytic performance of RP-MnOx (8-400) and CP-MnOx (8-400) with
their surface Mn4+ ions content, one can see that the catalytic activ-
6
65
660
O1s
655
650
645
640
635
630
BE/eV
4+
ity is correlated with the surface Mn ion content, which means
that the large amount of Mn4+ species on the surface of RP-MnOx
(B)
(8-400) may account for its high activity and selectivity during the
CP-MnO (8-400)
x
complete catalytic oxidation of o-xylene.
The O1s spectrum is frequently used to identify the types of
surface oxygen species present in a particular oxide. The chemical
environment of oxygen in metal oxide catalysts played an impor-
tant role in their catalytic properties. Fig. 6(B) shows the XPS spectra
of O1s in RP-MnOx (8-400) and CP-MnOx (8-400). For both materi-
als, three peaks were observed, attributing to three types of oxygen
species: the peak at BE of 529–530 eV was ascribed to the lattice
oxygen (denoted as O␣), the peak at BE of 531–532 eV was ascribed
to defective oxides or surface oxygen ions with low coordination
0α
0β
RP-MnO (8-400)
x
(
denoted as O ) [19,27], and the peak at a higher BE (above 533 eV)
␤
corresponded to adsorbed water [26]. Peak deconvolution was only
performed for the O␣ and O␤ species, because the adsorbed water
peak was too weak [26]. The relative amounts of O␣ in RP-MnOx (8-
5
36
534
532
530
BE/eV
528
526
4
00) and CP-MnOx (8-400) are almost the same, 74.2% for CP-MnOx
(
8-400), and 74.5% for RP-MnOx (8-400) (see Table 1). However, the
Fig. 6. Mn 2p XPS spectra (A) and O1s XPS spectra (B) of RP-MnOx (8-400) and
CP-MnOx (8-400).
H -TPR results showed that the oxygen specie on RP-MnOx (8-400)
was more active than that of CP-MnOx (8-400).
2
The oxidation of organic molecules over metal oxide catalysts
usually proceeds according to the Mars and Van Krevelen (MVK)
mechanism [29], where the organic molecules is oxidized by the
lattice oxygen of the metal oxide, the latter being re-oxidized by gas
phase oxygen in the feed gas. Therefore, the reducibility of metal
oxide species and oxygen removal activity appear to be two crit-
ical parameters determining their catalytic performance for VOC
oxidation.
◦
reduction peak around 340 C, followed by a broad moderate inten-
◦
sity reduction peak with a maximum at 505 C. Whereas, two
intensive reduction peaks centered at lower temperatures of 338 C
and 377 C were observed for the RP-MnOx (8-400) sample. In
◦
◦
general, the low temperature reduction could be assigned to the
reduction of MnO2 to Mn O , and the high-temperature reduction
3
4
corresponded to the reduction of Mn O4 to MnO [23–25]. More-
3
over, it can be seen that the amount of consumed hydrogen over the
RP-MnOx (8-400) sample was much high than that of the CP-MnOx
Higher oxidation state of manganese species was previously
found to be preferable for oxidation reactions over the manganese-
containing catalysts. Machocki et al. [28] reported that the rate of
(
(
8-400) sample. Whereas, the reduction temperature of RP-MnOx
8-400) was extremely lower than that of CP-MnOx (8-400). TPR
methane oxidation over LaMnO displays a linear relationship with
3
the surface Mn4+/Mn ratio. Park et al. [30] also reached the same
3+
analyses showed that the redox behavior of the manganese oxide
in RP-MnOx (8-400) was potentiated, indicating improved oxygen
mobility and availability. The RP-MnOx (8-400) exhibited good low-
temperature reducibility, which may also be responsible for its high
catalytic activity in the complete oxidation of o-xylene.
conclusion that PdOx/MnO catalyst showed higher catalytic activ-
2
ity for CO oxidation than PdOx/Mn O in their comparative study of
2
3
the support effect between MnO and Mn O . Similar phenomenon
2
2
3
was observed in the present study, and the RP-MnOx (8-400) cat-
alyst that possessed almost 100% Mn4+ species and richer lattice
oxygen species resulted in much higher catalytic activity for the
complete oxidation of o-xylene. Combined the analysis of above
characterizations, we deduced that the high surface Mn4+ concen-
tration resulted in high lattice oxygen content, high surface area,
and open porous structure of RP-MnOx (8-400), which may pro-
vide more active sites and facilitate the adsorption and diffusion of
reactant molecules, resulting in fast oxygen diffusion and enhanced
mobility of lattice oxygen. Moreover, the presence of small amount
To measure the composition and oxidation state of RP-MnOx
(
8-400) and CP-MnOx (8-400), XPS spectra were taken on the
two samples, as shown in Fig. 6(A) shows the XPS spectra of Mn
p in RP-MnOx (8-400) and CP-MnOx (8-400). RP-MnOx (8-400)
showed a sharp, strong Mn 2p peak at 642.4 eV, attributed to
2
4+
Mn (641.1–642.4 eV) in MnO2 [26], indicating MnO2 species was
formed on the RP-MnOx (8-400). The sharpening of this XPS peak
may be attributed to the stabilization of only one type of man-
ganese oxide [27], in agreement with the XRD result. Furthermore,

Y. Wu et al. / Catalysis Today 201 (2013) 32–39
37
Table 1
XPS results of RP-MnOx (8-400) and CP-MnOx (8-400).
Sample
BE (eV)
Mn4+
Mn4+/(Mn4+ + Mn3+ + Mn2+) (%)
BE (eV)
O␣/(O␣ + O␤) (%)
Mn3+
Mn2+
O␣
O␤
CP-MnOx (8-400)
RP-MnOx (8-400)
642.4
642.4
641.7
–
641.2
–
31.26
100
529.8
529.9
531.5
531.6
74.2
74.5
◦
of potassium ion in the microcrystalline ␣-MnO2 may weaken the
Mn O bond and improve the reducibility of RP-MnOx (8-400) [20],
thus enhancing its catalytic activity.
Above 220 C, all three catalysts gave 100% CO2 yield. Significant
◦
difference in CO2 yields was observed between 180 and 200 C. T
(the temperature needed to reach a CO2 yield of 50%) was usu-
ally used to measure the catalytic activity in terms of CO2 yield.
The T50 of RP-MnOx (8-400) was found to be 192 C, whereas the
T50 of RP-MnOx (6-400) and RP-MnOx (10-400) was 206 C and
50
◦
3.2. Effect of the pH value
◦
◦
2
04 C respectively. These results suggest that the catalytic activity
It was reported that the pH value played important roles in
is dependent on the pH value, and that pH 8 is the optimal value.
The XRD patterns of the RP-MnOx (Y-400) (Y = 6, 8, 10) samples
are presented in Fig. 8. All three RP-MnOx (Y-400) catalysts showed
similar XRD patterns. As discussed above, for RP-MnOx (8-400),
the diffraction peaks observed in 2ꢀ ranges of 35–40 and 65–70
were attributed microcrystalline ␣-MnO2 active phase. Compared
determining the crystal structures and morphology of final prod-
ucts [31]. In our previous experiments, we found that the MnOx
synthesized via mixing KMnO4 and Mn(NO3)2 in acidic media
(
pH < 5) was not thermally stable and could not be used as gas-
◦
◦
phase catalysts. High pH may lead to the formation of Mn(OH) .
2
+
Thus pH values between 6 and 10 were used in this work. Since K
with RP-MnOx (8-400), RP-MnOx (6-400) exhibited three stronger
ions can stabilize MnOx structure, KOH was used as base reagent to
adjust the pH.
◦
◦
◦
diffraction peaks at 2ꢀ = 28.5 , 37.5 and 49.6 , also attributed to
␣
-MnO . However, the RP-MnOx (10-400) only showed one weak
Fig. 7 shows the o-xylene conversion and CO2 yield over RP-
MnOx (Y-400) catalysts prepared at pH 6, 8, and 10, and calcined at
2
◦
peak at 2ꢀ = 37.5 . At pH 6, the reaction system was a weak acidic
◦ ◦
solution, the reaction between KMnO4 and Mn(NO3)2 is relatively
4
00 C. Above 220 C, all three catalysts exhibited similar catalytic
rapid due to the distinguishing redox potential of MnO /MnO
activities for o-xylene removal, and 100% conversion of o-xylene
was achieved. The RP-MnOx (Y-400) prepared at pH 8 showed sig-
nificantly higher o-xylene conversion efficiency than the others
4
2
2+
(
1.70) and MnO /Mn (1.23), which might produce a product with
2
high crystalinity, displaying strong diffraction peaks. While at pH
10, the reaction condition is milder, and the formation of MnO2
might be slower than that at pH 8. Also Mn(OH)2 may be produced
◦
at 200 C. Similar phenomena were observed with the CO2 yield.
from this basic system, resulting in the formation of an amor-
phous phase. Most possibly, the product with a certain degree of
crystallinity might exhibit high catalyst activity, as observed with
RP-MnOx (8-400).
1
00
(
A)
8
6
4
2
0
0
0
0
0
3.3. Effect of calcination temperature
RP-MnO (6-400)
x
Calcination may change a catalyst’s morphology and further
affect its catalytic activity. Thus effect of calcination temperature
on the catalytic activity of RP-MnOx (8-Z) was investigated for the
RP-MnO (8-400)
x
RP-MnO (10-400)
x
catalytic oxidation of o-xylene. Fig. 9 shows the catalytic activities
of RP-MnOx (8-Z) catalysts that were calcined at 350–600 ◦C. As
◦
◦
the calcination temperature increased from 350 C to 400 C, the
o-xylene conversion and CO2 yield increased slightly, after that,
◦
1
60
200
240
280
o
320
360
further increase to 500 C resulted in a catalytic activity decrease,
◦
◦
◦
as evidenced from their T50 values (208 C, 196 C, and 201 C for
RP-MnOx (8-350), RP-MnOx (8-400) and RP-MnOx (8-500) respec-
Temperature( C)
◦
tively), to the material calcined at 600 C, showed an extremely low
100
(B)
catalytic activity.
80
60
40
RP-MnO (6-400)
x
RP-MnO (8-400)
x
RP-MnO (10-400
)
x
160
200
240
280
320
360
o
Temperature( C)
Fig. 7. Effect of pH value on o-xylene conversion (A) and CO2 yield (B) over RP-MnOx
Y-400) samples. Reaction conditions: o-xylene 700 ppm, 20% O2/N2 balance, total
(
−1
−1
flow rate 50 mL min , W/F = 0.30 g s mL .
Fig. 8. XRD patterns of RP-MnOx (Y-400) samples at varied pH of 6–10.

3
8
Y. Wu et al. / Catalysis Today 201 (2013) 32–39
Table 2
100
(
A)
BET surface area and activity of RP-MnOx (8-Z) for the catalytic oxidation of o-xylene
◦
at 190 C.
a
Areal rateb
(mol m min
Catalysts
Surface area
Conversion at
190 C (%)
8
6
4
2
0
0
0
0
0
2
−1
◦
−2
−1
)
(
m
g
)
−
−
−
−
9
9
9
9
RP-MnOx (8-350)
RP-MnOx (8-400)
RP-MnOx (8-500)
RP-MnOx (8-600)
60.31
69.42
57.29
43.15
7.32
10.77
8.58
3.39 × 10
4.33 × 10
4.18 × 10
3.94 × 10
6.10
a
Reaction rate was determined under the conditions which obtained a o-xylene
conversion level of below 30%.
RP-MnO (8-350)
b
x
Areal rate = [o-xylene]in × rate of flow × conversion/surface area.
RP-MnO (8-400)
x
RP-MnO (8-500)
x
could be attributed to the reduction of MnO2 to Mn O , and the
high-temperature reduction peak was assigned to the reduction
of Mn O4 to MnO. However, the two reduction peaks integrated
into single reduction peak and shifted to higher temperatures
3
4
RP-MnO (8-600)
x
3
1
60
200
240
280
320
360
o
◦
Temperature( C)
with increasing the calcination temperature up to 600 C, indicat-
ing the reducibility of MnO2 decreased as calcination temperature
increased, which also contribute to their corresponding catalytic
activity decrease.
100
(
B)
Table 2 summarized the surface area and catalytic activity over
RP-MnOx (8-Z) (Z = 350, 400, 500, and 600) at 190 C. The areal rate
reflects the intrinsic activity of catalyst, calculated from the conver-
sion efficiency of o-xylene and BET surface area. For RP-MnOx (8-Z)
8
0
0
0
0
◦
6
4
2
(
Z = 400, 500, and 600), increase in calcination temperature leads
to BET surface area decrease, and their catalytic activity decreased
correspondingly. The RP-MnOx (8-350) has a higher BET surface
area than RP-MnO (8-500) and RP-MnO (8-600). However its cat-
RP-MnO (8-350)
x
RP-MnO (8-400)
x
x
x
RP-MnO (8-500)
alytic activity is lower than the other two. This is possibly due to
its amorphous MnOx phase. The catalyst with a certain degree of
crystallinity showed high catalytic activity. The highest catalytic
activity was observed with RP-MnOx (8-400), which has the largest
BET surface area and an optimal degree of crystallinity.
x
RP-MnO (8-600)
x
0
160
200
240
280
320
360
o
Temperature( C)
The RP-MnOx (8-Z) (Z = 400, 500, 600) catalysts have similar
areal rates as shown in Table 2, suggesting that these catalysts have
the similar catalytic active center, namely, ␣-MnO2. The areal rate
Fig. 9. Effect of calcination temperature on o-xylene conversion (A) and CO2 yield
B) over RP-MnOx (8-Z) samples. Reaction conditions: o-xylene 700 ppm, 20% O2/N2
balance, total flow rate 50 mL min , W/F = 0.30 g s mL .
(
−1
−1
◦
of RP-MnOx (8-350) at 190 C was smaller than that of RP-MnOx
(
8-400), illustrating that that the active center of ␣-MnO2 was not
High calcination temperature may result in the formation of
larger degree of crystallinity and larger crystal size, as evidenced
in Fig. 3. High calcination temperature may also lead to a decrease
in reducibility of MnO2 and low BET surface area. Fig. 10 shows the
TPR profiles of the RP-MnOx (8-Z) (Z = 400, 500, 600). For the sam-
completely formed in RP-MnOx (8-350).
3
.4. Stability test for the RP-MnOx (8-400) catalyst
The stability of RP-MnOx (8-400) was tested for the isother-
◦
◦
ples calcined at 400 C and 500 C, two reduction peaks located at
◦
mal combustion of o-xylene at 220 C for 60 h. Fig. 11 shows
◦
◦
◦
◦
about 338 C, 350 C and 377 C, 386 C, respectively, can be clearly
observed. As discussed above, the low temperature reduction peak
1
00
9
0
RP-MnO (8-600)
x
80
70
60
RP-MnO (8-500)
x
RP-MnO (8-400)
x
0
10
20
30
Time(h)
40
50
60
100
200
300
400
500
600
o
Temperature / C
Fig. 11. CO2 yield versus on stream time over the RP-MnOx (8-400) catalyst. Reac-
−
1
tion conditions: o-xylene 700 ppm, 20% O2/N2 balance, total flow rate 50 mL min
,
−
1
Fig. 10. H2-TPR profiles of RP-MnOx (8-Z) samples.
W/F = 0.30 g s mL
.

Y. Wu et al. / Catalysis Today 201 (2013) 32–39
39
the result. CO2 yield remained >98% during the test, indicat-
ing. RP-MnOx (8-400) is relatively stable under these working
conditions.
Center for Eco-Environmental Science (RCEES), Chinese Academy
of Sciences.
References
4
. Conclusion
[
[
1] S.Y. Huang, C.B. Zhang, H. He, Catalysis Today 139 (2008) 15.
2] H.S. Kim, T.W. Kim, H.L. Koh, S.H. Lee, B.R. Min, Applied Catalysis A 280 (2005)
1
A new redox-precipitation method followed by calcination at
25.
[3] K. Okumura, T. Kobayashi, H. Tanaka, M. Niwa, Applied Catalysis B 44 (2003)
25.
◦
4
00 C was developed, and open porous hierarchically structured
3
␣
-MnO , was prepared. The physicochemical properties of the
2
[
[
4] J.C.S. Wu, T.Y. Chang, Catalysis Today 44 (1998) 111.
5] R.S.G. Ferreira, P.G.P. de Oliveira, F.B. Noronha, Applied Catalysis B 50 (2004)
243.
materials were characterized by the XRD, BET, SEM/EDS, XPS, and
H -TPR techniques, and the catalytic activity was tested for o-
xylene conversion. Effects of pH and calcination temperature were
studied, and it was found that pH 8 and 400 C calcination were
2
[
6] A.C. Gluhoi, B.E. Nieuwenhuys, Catalysis Today 119 (2007) 305.
◦
[7] S.M. Saqer, D.I. Kondarides, X.E. Verykios, Applied Catalysis B 103 (2011) 275.
8] M. Chen, X.M. Zheng, Journal of Molecular Catalysis A 221 (2004) 77.
[
optimal conditions. The catalyst (RP-MnOx (8-400)) showed 100%
[9] M.A. Alvarez-Merino, M.F. Ribeiro, M.J. Silva, F. Carrasco-Marin, F.J. Maldonado-
Hodar, Environmental Science and Technology 38 (2004) 4664.
10] M.C.M. Alvim-Ferraz, C.M.T.B. Gaspar, Environmental Science and Technology
◦
◦
o-xylene conversion to CO2 at 220 C, 50 C lower than the cata-
lyst prepared via the conventional precipitation method. This novel
method produced materials with 100% Mn4 ion on their surface,
whereas the conventional method could only produce materials
[
[
39 (2005) 6231.
+
11] W.B. Li, W.B. Chu, M. Zhuang, J. Hua, Catalysis Today 93–95 (2004) 205.
[12] W.B. Li, J.X. Wang, H. Gong, Catalysis Today 148 (2009) 81.
[13] Y. Liu, M.F. Luo, Z.B. Wei, Q. Xin, P.L. Ying, C. Li, Applied Catalysis B 29 (2001)
4
+
with 31% Mn ion on their surface. It is concluded that that the sur-
face concentration of Mn4 ion plays an important role for its high
catalytic activity for the catalytic combustion of o-xylene. The open
porous hierarchical structure of the ␣-MnO2 has high BET surface
area, may provide more active sites, and facilitate the adsorption
and diffusion of reactant molecules. The presence of a small amount
of potassium ions in the ␣-MnO2 channels may weaken the Mn-
O bond and promote the reducibility of the catalyst, which also
contribute to the excellent catalytic performance of the ␣-MnO2
material.
6
1.
+
[
[
[
14] T.K. Tseng, H. Chu, H.H. Hsu, Environmental Science and Technology 37 (2003)
171.
15] C. Lahousse, A. Bernier, P. Grange, B. Delmon,.P. Papaefthimiou, T. Loannides,
X. Verykios, Journal of Catalysis 178 (1998) 214.
16] T.Y. Li, S.J. Chiang, B.J. Liaw, Y.Z. Chen, Applied Catalysis B 103 (2011) 143.
[17] H.J. Li, G.S. Qi, Tana, X.J. Zhang, X.M. Huang, W. Li, W.J. Shen, Applied Catalysis
B 103 (2011) 54.
[
18] D.A. Aguilera, A. Perez, R. Molina, S. Moreno, Applied Catalysis B 104 (2011)
44.
[19] Y.S. Wu, Y.X. Zhang, M. Liu, Z.C. Ma, Catalysis Today 153 (2010) 170.
1
[
[
20] Y.S. Wu, M. Liu, Z.C. Ma, S.T. Xing, Catalysis Today 175 (2011) 196.
21] J.G. Deng, L. Zhang, H.X. Dai, H. He, C.T. Au, Journal of Molecular Catalysis A 299
This novel synthetic method is environmentally friendly, cost
effective and feasible for large scale production of highly active
catalyst for the catalytic oxidation of VOCs.
(
2009) 60.
[22] S.T. Xing, R.R. Han, Z.C. Ma, Y.S. Wu, Z.C. Zhou, CrystEngComm 13 (2011) 6033.
[23] J. Carn oˇ , M. Ferrandon, E. Bj oˇ nbom, S. Jaras, Applied Catalysis A 155 (1997) 265.
24] J. Trawczy n´ ski, B. Bielak, W. Mista, Applied Catalysis B 55 (2005) 277.
25] X.F. Tang, Y.G. Li, X.M. Huang, Applied Catalysis B 62 (2006) 265.
[
[
Acknowledgments
[26] M. Ferrandon, J. Carnö, S. Järas, E. Björnbom, Applied Catalysis A 180 (1999)
41.
1
[
27] P.R. Ettireddy, N. Ettireddy, S. Mamedov, P. Boolchand, P.G. Smirniotis, Applied
Catalysis B 76 (2007) 123.
This work was financially supported by the National Natural Sci-
ence Foundation of China (no.21147004, no.20977024), the Natural
Science Foundation of Hebei Education Department (no.2008128,
no.Z2010152) and the Foundation of Hebei Normal University
[
28] A. Machocki, T. Ioannides, B. Stasinska, W. Gac, G. Avgouropoulos, D. Delimaris,
W. Grzegorczyk, S. Pasieczna, Journal of Catalysis 227 (2004) 282.
29] S. Devaraj, N. Munichandraiah, Journal of Physical Chemistry C 112 (2008) 4406.
30] J.S. Park, D.S. Doh, K.Y. Lee, Topics in Catalysis 10 (2000) 127.
31] L. Zhou, J. Zhang, J.H. He, Y.C. Hu, H. Tian, Materials Research Bulletin 46 (2011)
1714.
[
[
[
(
no. L2010Z06). The authors appreciate valuable discussion with
Prof. Hong He and experimental help from his group in Research