Characterization of Ag and Ba-modified manganese oxide catalysts: Unraveling the factors leading to their enhanced CH4 oxidation activity

Article abstract of DOI:10.1039/b101116m

Catalysts based on manganese oxide and modified with Ag (AgMn0.10) or Ba (BaMn0.10) were prepared and applied to CH₄ deep oxidation. The catalysts were characterized by means of N₂-BET, XRD, XPS, CH₄-TPR and O₂-TPD techniques. Catalytic evaluation results show that both AgMn0.10 and BaMn0.10 exhibit higher CH₄ oxidation activity than the unmodified MnO_x. However, comparative studies on CO oxidation reveal that only AgMn0.10 displays markedly improved activity for this reaction; the activity of BaMn0.10 remains unchanged compared to unmodified MnO_x. In AgMn0.10, a large amount of more mobile and reactive oxygen species is formed, which is thought to be the main reason for its enhanced CH₄ and CO oxidation activity. Whereas, in BaMn0.10, it is found that the mobility of the lattice oxygen is not increased, but somewhat decreased, thus resulting in no enhanced CO oxidation activity on this catalyst. However, the basicity of the lattice O^2- is thought to be increased via Ba addition, which is favorable for the activation and rupture of the first C-H bond in CH₄, the rate-determining step for this reaction. Indeed, this modification effect of Ba is believed to be the predominant factor responsible for the enhanced CH₄ oxidation activity of BaMn0.10.

Full text of DOI:10.1039/b101116m

View Article Online / Journal Homepage / Table of Contents for this issue
Characterization of Ag and Ba-modiÐed manganese oxide catalysts:
unraveling the factors leading to their enhanced CH oxidation
4
activity
Xiang Wang*¤ and You-chang Xie
Institute of Physical Chemistry, Peking University, Beijing, 100871, P. R. China
Received (in Montpellier, France) 1st February 2001, Accepted 2nd April 2001
First published as an Advance Article on the web 1st June 2001
Catalysts based on manganese oxide and modiÐed with Ag (AgMn0.10) or Ba (BaMn0.10) were prepared and
applied to CH deep oxidation. The catalysts were characterized by means of N -BET, XRD, XPS, CH -TPR and
4
2
4
O -TPD techniques. Catalytic evaluation results show that both AgMn0.10 and BaMn0.10 exhibit higher CH
2
4
oxidation activity than the unmodiÐed MnO . However, comparative studies on CO oxidation reveal that only
AgMn0.10 displays markedly improved activity for this reaction; the activity of BaMn0.10 remains unchanged
x
compared to unmodiÐed MnO . In AgMn0.10, a large amount of more mobile and reactive oxygen species is
x
formed, which is thought to be the main reason for its enhanced CH and CO oxidation activity. Whereas, in
4
BaMn0.10, it is found that the mobility of the lattice oxygen is not increased, but somewhat decreased, thus
resulting in no enhanced CO oxidation activity on this catalyst. However, the basicity of the lattice O2~ is thought
to be increased via Ba addition, which is favorable for the activation and rupture of the Ðrst CÈH bond in CH , the
4
rate-determining step for this reaction. Indeed, this modiÐcation e†ect of Ba is believed to be the predominant
factor responsible for the enhanced CH oxidation activity of BaMn0.10.
4
CH , the main component of natural gas, is an attractive
expectation that this will provide some directions for further
4
alternative fuel due to its high H to C ratio, low SO and NO
developing catalysts for CH deep oxidation. In an attempt to
x
x
4
emissions and its ready availability around the world.1h3
elucidate the modiÐcation e†ects of Ag and Ba on manganese
However, CH is reported to be a much stronger greenhouse
oxide, CO oxidation was also performed over the samples.
The catalysts were characterized by di†erent physicochemical
techniques. In this study, the relations between the physi-
cochemical properties of the catalysts and their catalytic activ-
ity are discussed.
4
gas than CO .4 Although the emission of CH from di†erent
2
4
sources, e.g. industry effluents and automobile exhausts, is not
specially regulated now, it is expected to be restricted in the
near future. Therefore, the utilization of CH to provide
4
energy or its elimination from the atmosphere to reduce air
pollution is desirable.
Experimental
Traditional catalysts used for high temperature combustion
of hydrocarbons and volatile organic compounds (VOCs) are
supported noble metals such as Pt, Pd, Rh or combinations
thereof.5 Pd-based catalysts are found to be the most active
Catalyst preparation
The high speciÐc surface area MnO precursor (265 m2 g~1)
was prepared via the stoichiometric reaction of aqueous
for CH oxidation.5h8 However, these catalysts are reported
2
4
not to be active enough for CH deep oxidation at relatively
4
KMnO and Mn(NO ) solutions.19,20 Typically, 0.24 M
lower temperatures in these applications.2,3,8 Currently, the
4
4
3 2
KMnO solution was dripped into 50% Mn(NO ) which was
search for catalysts applicable to these processes is an impor-
tant area of research. Furthermore, because of the limited
availability of noble metals, over recent years, there has been a
trend to develop combustion catalysts with low Pd loading, or
without precious metals and based on composite base metal
oxides. Manganese oxides have been extensively studied for
di†erent reactions.9h16 Our previous work has shown that the
3 2
under stirring. A brown precipitate, which has been proven to
be ultraÐne a-MnO powder,21 was rapidly formed. On com-
pletion of the reaction, the pH of the mixture is ca. 2.0. This
mixture was then vacuum Ðltered and the precipitate washed
repeatedly with distilled water, until the water passing
through it was neutral. Afterwards, the product was dried at
2
1
10 ¡C overnight and used as the precursor to prepare the
catalysts.
AgÈMn and BaÈMn binary oxide catalysts with a M/Mn
addition of Ag or Ba to high speciÐc surface area MnO sig-
2
niÐcantly enhances the CH deep oxidation activity of the
4
modiÐed catalysts.17,18 However, the reasons leading to this
mol ratio (M \ Ag and Ba) of ca. 0.1 were prepared by
impregnating the MnO precursor with a calculated amount
improved activity are not yet clearly understood. Therefore,
the primary objective in this study is to clarify the modiÐ-
cation e†ects of Ag and Ba on manganese oxides, with the
2
3 2
of AgNO and Ba(NO ) aqueous solution. After evaporation
3
to dryness in a water bath, both of the samples were calcined
in static air atmosphere at 600 ¡C for 4È6 h. The catalysts are
denoted AgMn0.10 and BaMn0.10. The unmodiÐed MnO
¤
Present address: Department of Chemical Engineering, Towne
Building, Room 311A, University of Pennsylvania, Philadelphia, PA
9104, USA. E-mail: wangx2=seas.upenn.edu
x
was prepared by direct calcination of the MnO precursor
2
1
under the same conditions.
964
New J. Chem., 2001, 25, 964È969
DOI: 10.1039/b101116m
This journal is ( The Royal Society of Chemistry and the Centre National de la Recherche ScientiÐque 2001

View Article Online
Catalyst characterization
evaluated at a GHSV of 21 000 h~1 with ascending tem-
perature. The volume composition of the feed gas was 2.5%
CO in air balance. The conversion of CO was calculated from
the change in the peak area before and after passage over the
catalyst bed.
BrunauerÈEmmettÈTeller (BET) surface areas of the samples
were measured by nitrogen adsorptionÈdesorption at 77 K
with an ST-30 instrument.
XRD patterns were recorded on a BD-90 X-ray di†ractom-
eter with Cu-Ka radiation of 40 kV ] 20 mA and Ni Ðlter.
The scan step was 0.1¡ with a preset counting time of 4 s.
XPS experiments were carried out on a VG-ESCA-LAB-5
electron spectrometer using Al-Ka radiation (10 kV, 40 mA).
The spectra were obtained at ambient temperature with a
chamber pressure of 10~8 Torr. The surface O/Mn atomic
ratios of the catalysts were calculated from the integrated
areas of the Mn2p3@2 and O1s peaks after the necessary cali-
bration.
Results
Activity evaluation
CH deep oxidation. Fig. 1(A) shows the light-o† curves of
4
CH oxidation over the modiÐed and unmodiÐed catalysts. In
4
this work, CH is considered to be completely oxidized when
4
conversion reaches 98%. Obviously, AgMn0.10 and
BaMn0.10 exhibit higher CH conversion than the unmod-
iÐed MnO at the same reaction temperatures, especially
below 500 ¡C. With increasing reaction temperature, the di†er-
ence in CH conversion between the two modiÐed catalysts
and MnO becomes less, although the CH conversions over
the modiÐed catalysts are still higher. Among the three
samples, BaMn0.10 displays the highest CH conversion at a
certain temperature. The complete oxidation of CH occurs at
580 ¡C over MnO , while it takes place at 540 ¡C over the two
modiÐed samples.
Both AgMn0.10 and BaMn0.10 were subjected to a 10 h
durability test at 540 ¡C in ca. 4% H O (results not shown).
No deactivation occurred, indicating they are stable catalysts
CH temperature programmed reduction (CH -TPR) was
4
4
4
carried out using a CH /N (5.5%) gas mixture. The tem-
4
2
x
perature was increased from 24 to 800 ¡C at a rate of 10 ¡C
min~1. Prior to entering into the catalyst bed, the gas Ñow
was puriÐed with a MnO oxygen trap and a Molecular Sieve
4
x
4
5
A H O trap. Generally, the catalysts were diluted with
2
quartz powder to 10% (wt). 100 mg of this mixture containing
4
ca. 10 mg pure catalyst powder was used for the experiments.
4
The main reaction products in this experiment are H O and
2
CO ; the formation of CO is negligible, as also attested to by
2
x
Stobbe et al.22 We followed the reaction by measuring the
CH consumption with a thermal conductivity detector
4
2
(
TCD). To rule out the e†ect of the H O and CO produced
2
2
on the signal, an additional Molecular Sieve 5A trap was
attached to the outlet of the reactor to adsorb them. When the
trap is saturated, the baseline becomes noisy and some nega-
tive H O or CO peaks can be observed. With this method,
for CH oxidation.
4
2
2
we made sure that the obtained proÐles are really based on
CH consumption, and not perturbed by the products formed.
4
O
temperature programmed desorption (O -TPD) was
2
2
performed on the same set-up used for CH -TPR with a ramp
4
of 10 ¡C min~1 using high purity He as the carrier gas (25 ml
min~1). Generally, 40 mg pure sample was used here. Prior to
the experiment, the samples were calcined again in dry air at
600 ¡C for 1 h, then cooled down to room temperature, and
purged with a high purity N Ñow for ca. 30 min.
2
Activity evaluation
Catalytic evaluation was carried out in a U-shaped Ðxed-bed
microreactor (i.d. \ 6 mm) with a continuous downÑow. The
samples were pressed under 8 MPa for 2 min to form pellets,
then crushed and sieved. Typically, 0.2 ml 40È60 mesh cata-
lysts were used. To avoid channeling, ca. 1 ml of porcelain
particles of the same size were loaded above the catalyst bed.
A blank experiment conÐrmed the lack of activity of these
porcelain particles under the reaction conditions employed. A
K-type thermocouple was placed on top of the catalyst bed (in
contact with the catalyst) to monitor the reaction temperature.
To measure the light-o† behavior of the catalysts, all data
were collected with ascending temperature. The volume com-
position of the feed gas was CH 1.5%, O 18%, balanced by
4
2
high purity N . The total feed Ñow rate was 70 ml min~1
2
corresponding to a gas hourly space velocity (GHSV) of
2
1 000 h~1. The reactants and products were analyzed with an
on-line 1102G gas chromatograph equipped with a TCD on a
Porapak Q column for CH and CO , and on a Molecular
Sieve 5A column for CH and CO. In order to collect stable
4
2
4
and reproducible data, before analysis, the reaction at each
temperature over all of the studied catalysts was stabilized for
at least 30 min, then 3È5 injections were made. The Ñow rate
of the carrier gas He was 30 ml min~1. CH conversion was
4
calculated from the change in the peak area before and after
passage over the catalyst bed.
CO oxidation was performed on the same set-up and with
similar methods. Typically, 0.2 ml 40È60 mesh catalysts were
Fig. 1 CH
4 deep oxidation over the catalysts: (A) light-o† curves;
(B) Arrhenius plots.
New J. Chem., 2001, 25, 964È969
965

View Article Online
The steady-state CH oxidation rates on the three catalysts,
which were achieved under di†erential conditions, are com-
pared in Fig. 1(B) as Arrhenius plots. Apparently, the two
samples. It is evident that the surface areas of the modiÐed
samples do not play a key role in their enhanced CH4 deep
oxidation activity.
4
modiÐed catalysts show higher activity than MnO , as
demonstrated by their higher reaction rates. The activation
XRD was used to analyze the bulk phase composition of
the samples, with the results shown in Fig. 3. For clarity, the
crystalline phases detected in each sample are also listed in
Table 1. The predominant phase of MnO is a-Mn O
x
energies on MnO , AgMn0.10 and BaMn0.10 are 84, 103 and
x
9
1
4 kJ mol~1, respectively. Considering the results in both Fig.
x
2 3
(A) and (B), the CH oxidation activity sequence of the cata-
lysts is BaMn0.10 [ AgMn0.10 [MnO .
(Bixbyte), together with a trace amount of b-MnO
(Pyrolusite). This indicates that part of the lattice oxygen was
4
2
x
desorbed from the MnO precursor, with the metastable a-
2
CO oxidation. To elucidate the reasons leading to the
MnO transforming mainly into more stable a-Mn O and a
2
2 3
enhanced CH oxidation activity over the modiÐed catalysts,
trace amount of b-MnO .21 As a result, the oxidation state of
4
2
CO oxidation over the three catalysts was also performed for
Mn decreased from ]4 to ]3. In AgMn0.10 and BaMn0.10,
comparison. The light-o† behaviors of CO oxidation over dif-
ferent samples are shown in Fig. 2. Interestingly, although
a-Mn O is still one of the main crystalline phases. However,
2
3
besides this component, some new crystalline phases such as
Ag Mn O , BaMn O and BaMnO are clearly detected in
the corresponding samples. Obviously, during the calcination,
BaMn0.10 displays the highest activity in CH oxidation
4
2
8 16
8 16
3
among the three samples, it does not show improved CO oxi-
dation activity over MnO . Both catalysts exhibit nearly the
Ag and Ba reacted with the MnO precursor to form these
x
2
same light-o† behaviors, with complete CO oxidation taking
new composite oxides. It is noted that, instead of ]3 as in
place at 140 ¡C. In contrast, AgMn0.10 shows markedly
MnO , the oxidation state of a considerable amount of the
x
higher CO oxidation activity than MnO and BaMn0.10. It is
Mn is stabilized at ]4 in the two modiÐed samples due to the
x
clear that the steeply ascending light-o† curve of AgMn0.10 is
formation of these composite oxides.
separated from those of MnO and BaMn0.10 by 30 to 40 ¡C.
x
The complete oxidation of CO occurs at 90 ¡C over
XPS analysis
AgMn0.10, which is 50 ¡C lower than that on MnO and
x
The surface O/Mn atomic ratios of the samples measured by
XPS are listed in Table 1. These ratios are 1.47, 1.75 and 1.61
BaMn0.10. Based on steady-state CO oxidation rates under
di†erential conditions, the activation energies on MnO ,
x
for MnO , AgMn0.10 and BaMn0.10, respectively. Within the
AgMn0.10 and BaMn0.10 are 41, 45 and 46 kJ mol~1, respec-
x
experimental error, the O/Mn ratio 1.47 of MnO is equal to
tively. The ranking of the CO oxidation activities of the cata-
x
the calculated ratio of 1.50 for a-Mn O , the main crystalline
lysts is AgMn0.10 [ BaMn0.10 \ MnOx .
2
3
phase of this sample, indicating the surface composition of
N -BET and XRD measurements
MnO is the same as in the bulk. In AgMn0.10, the main
2
x
components are a-Mn O and Ag Mn O
8 16 , as evidenced by
2
3
2
The speciÐc surface areas of the catalysts are listed in Table 1.
The value for MnO is 23 m2 g~1, which is a little higher than
x
that of the two modiÐed samples. Our previous work indicates
that the high speciÐc surface area MnO precursor used in
2
this work consists mainly of ultraÐne a-MnO powder.21
2
Clearly, during the 600 ¡C calcination, the metastable a-MnO
2
sintered and transformed into more stable crystalline phases,
accompanied by a sharp decrease in the surface areas of the
Fig. 3 XRD patterns of the catalysts: (…) a-Mn O ; (=) b-MnO ;
2
3
2
Fig. 2 CO oxidation over the catalysts.
(+) BaMn O ; (K) BaMnO ; (L) Ag Mn O
16 .
8
16
3
2
8
Table 1 Physicochemical properties of the catalysts
BET surface
area/m2 g~1
Phase composition
(XRD)
Surface O/Mn atomic
ratio (XPS)
Catalyst
MnO
AgMn0.10
BaMn0.10
23
18
20
a-Mn O , b-MnO
1.47
1.75
1.61
x
2 3 2
a-Mn O , Ag Mn O
2
3
2
8
16
a-Mn O , BaMnO , BaMn O
2
3
3
8 16
966
New J. Chem., 2001, 25, 964È969

View Article Online
XRD. The O/Mn ratios of these two crystalline phases are 1.5
and 2.0, respectively. The measured surface O/Mn ratio of
AgMn0.10 is 1.75. This implies that both a-Mn O and
2
3
Ag Mn O are present on its surface, with slightly more
8 16
Ag Mn O than a-Mn O . As to BaMn0.10, the O/Mn
2
2
8 16
2 3
ratio of 1.61 indicates that the main component of the surface
is a-Mn O . However, a small amount of BaMn O could
2
3
8 16
also be present while the presence of BaMnO is negligible
3
since this phase has an O/Mn ratio of 3.0. The non-uniform
composition of the two modiÐed samplesÏ surfaces possibly
contributes to their enhanced activity for CH oxidation.
4
CH -TPR results
4
Previous work has substantiated that the formation of more
reducible or mobile oxygen species in a catalyst can improve
its oxidation activity.1,23 Therefore, in this study, CH -TPR
4
was performed to investigate the nature of the samples (Fig.
4). It is worth noting here that the initial brown color of the
samples changed to green after reduction, which is typical for
MnO, indicating that this is the main reduction product of the
sample.22,24
The main reduction peak of MnO is located at 605 ¡C,
x
together with a smaller peak at 500 ¡C. These two peaks can
be assigned to the reduction of a-Mn O , the main com-
Fig. 5 O -TPD proÐles of the catalysts.
2
3
2
ponent of this sample, to MnO with Mn O as the interme-
3
4
diate.22 In the proÐle of AgMn0.10, only a more intense peak
is observed at 540 ¡C, indicating the presence of a greater
amount of more reducible oxygen species in this sample than
In comparison, the two oxygen desorption peaks of
BaMn0.10 appear at 560 ¡C and 690 ¡C. The assignment of
these peaks is not very clear, but they are believed to relate to
the formation of BaÈMn composite oxides in the catalyst. The
relatively greater total integrated area of the two peaks indi-
in MnO . In comparison, in the reduction curve of BaMn0.10,
x
a wide peak at 530 ¡C and a narrow peak at 675 ¡C are found.
The CH consumption of the 530 ¡C peak is less than that of
4
cates the amount of desorbed oxygen is larger than in MnO .
AgMn0.10 and MnO in the same temperature region.
x
Although CH -TPR showed that the reducibility of the major
4
x
Overall, the mobility of the lattice oxygen in BaMn0.10 is not
part of the lattice oxygen of BaMn0.10 is not improved, but
improved, but somewhat decreased.
decreased. This cannot exclude the presence of a small
amount of oxygen species that can be desorbed more easily,
perhaps due to the non-uniform composition of the catalyst.
Interestingly, AgMn0.10 exhibits a very intense oxygen
desorption peak at 665 ¡C, together with a small shoulder at
525 ¡C. It is evident that a large amount of a more easily
desorbed oxygen species is present in this sample, possibly due
to the formation of Ag Mn O . The presence of a large
O -TPD results
2
Fig. 5 shows O -TPD proÐles for the catalysts. All the
samples exhibit two oxygen desorption peaks. The two peaks
2
of MnO are situated at 425 and 740 ¡C. The low-temperature
x
peak can be assigned to the desorption of adsorbed oxygen,11
while the 740 ¡C peak is attributed to the desorption of lattice
2
8 16
oxygen from Mn O to form Mn O .10
quantity of more mobile and active oxygen species could be
the main reason leading to the exceptional CO oxidation
activity of this catalyst.9h11 Moreover, this could also contrib-
2
3
3 4
ute to its enhanced CH deep oxidation activity.
4
Discussion
The results in the present work show that both Ag and Ba as
additives can improve the activity of manganese oxide for
CH oxidation, with the activity sequence being BaMn0.10 [
4
AgMn0.10 [MnO . However, for CO oxidation, only Ag can
x
markedly enhance the activity of manganese oxide; Ba does
not change the reactivity of the prepared catalyst. This seems
to imply that for e†ective CH and CO oxidation, di†erent
4
functions are required of the catalysts due to the di†erent
structures of the CH and CO molecules.
4
When studying Ag as a catalyst, Sachtler et al.25 found that
CO readily reacts with adsorbed oxygen on the surface of
metallic Ag. Ag0 particles or partially oxidized metallic Ag
were also proposed by some other authors as active sites for
hydrocarbon oxidation.3,26 However, in this study, although
we have no evidence for the absence of Ag0 crystallites in the
samples, neither is there any proof of their presence. It is thus
obvious that the enhanced CH and CO oxidation activity of
4
AgMn0.10 cannot be ascribed to the existence of metallic Ag
particles.
Kanungo et al.20,27 have studied the physicochemical
Fig. 4 CH -TPR proÐles of the catalysts.
properties of MnO modiÐed with CuO, namely Hopcalite
4
2
New J. Chem., 2001, 25, 964È969
967

View Article Online
catalyst. They attributed the main reason leading to its
remarkable CO oxidation activity to the formation of
CuMn O on the surface of this catalyst. The structure of
Conclusions
The addition of Ag and Ba to manganese oxide can improve
2
4
the CH oxidation activity of the resulting catalysts, with the
this
CuMn O
spinel
can
be
described
as
4
2
4
activity sequence BaMn0.10 [ AgMn0.10 [ MnO . However,
Cu`[Mn3`Mn4`]O 2~.27,28 They suggested that the
coexistence of Mn3` and Mn4` in the structure resulted in
x
4
a comparative study showed that only Ag can markedly
enhance the CO oxidation activity of manganese oxide, while
the activity of BaMn0.10 is basically unchanged compared to
ready electron transfer between Mn in di†erent oxidation
states, since this can occur without the cations changing their
positions. As a result, this catalyst shows superior CO oxida-
tion activity.20,27 Interestingly, in this study, composite oxides
such as Ag Mn O , BaMn O and BaMnO are formed in
that of MnO .
x
Based on our characterization results, it is believed that the
enhanced CH oxidation activity of AgMn0.10 is attributable
4
2
8 16
8 16
3
to the presence of a large amount of mobile and reactive
AgMn0.10 and BaMn0.10, respectively. Similarly to
oxygen species, perhaps as a consequence of the formation of
Ag Mn O in this catalyst. This conclusion is further sup-
CuMn O , both Mn3` and Mn4` are present in Ag Mn O
2
4
2
8 16
and BaMn O . On the surface of AgMn0.10, XPS indicates
2
8 16
8
16
that a considerable amount of Ag Mn O coexists with a-
8 16
Mn O . Therefore, possibly due to the same inherent reason
ported by its superior CO oxidation activity, as previous
studies have shown that mobile and reactive oxygen species
are critical for e†ective CO oxidation over a catalyst.
2
2
3
as Hopcalite catalyst, AgMn0.10 shows also remarkable CO
However, the mobility and reactivity of oxygen species are
not as crucial for CH oxidation as for CO oxidation, though
they are important. The rupture of the Ðrst CÈH bond of CH
is very difficult and generally regarded as the rate-determining
step. In this study, it was found that the mobility and reacti-
vity of the oxygen species in BaMn0.10 is not improved over
oxidation activity and enhanced CH deep oxidation activity.
4
However, for BaMn0.10, though XPS also shows the
4
coexistence of BaMn O and a-Mn O on its surface, the
4
8
16
2 3
amount of BaMn O is very low. Therefore, it seems that
8
16
the enhanced CH oxidation activity of BaMn0.10 cannot
4
simply be ascribed to easy electron transfer between Mn3`
MnO , thus resulting in no improvement in the CO oxidation
and Mn4` in the structure.
x
activity of this catalyst. However, the basicity of the lattice
In the present work, CH -TPR and O -TPD demonstrated
4
2
O2~ can be enhanced via Ba addition, which is believed to be
the presence of a large amount of more reducible and mobile
oxygen species in AgMn0.10. Indeed, this could relate to the
formation of Ag Mn O in this sample, and result from the
favorable for the activation of the Ðrst CÈH bond in CH . It
4
is thus concluded that the CH oxidation activity of
4
2
8 16
BaMn0.10 is markedly enhanced due to this modiÐcation
facile electron transfer between Mn3` and Mn4`. The pres-
ence of more mobile and reactive oxygen species has been
proven previously to be crucial for the signiÐcantly enhanced
CO oxidation activity of AgÈMn composite oxide catalysts,
which are able to increase the rates of the overall reaction
e†ect of Ba.
Acknowledgement
remarkably.9h11 We thus attribute the enhanced CH and CO
Financial support from the National Science Foundation of
China is greatfully acknowledged by the authors.
4
oxidation activity of AgMn0.10 to the presence of more
mobile and reactive oxygen species in this sample.
However, in this study, there is no proof to show that the
mobility of the lattice oxygen of BaMn0.10 is improved; on
References
the contrary, CH -TPR results demonstrate that the
1
2
3
Y. Li and J. N. Armor, Appl. Catal., B, 1994, 3, 275.
4
reducibility and activity is somewhat decreased. This could be
W. S. Epling and G. B. HoÑund, J. Catal., 1999, 182, 5.
Lj. Kundakovic and M. Flytzani-Stephanopoulos, Appl. Catal. A,
the reason why the CO oxidation activity is not enhanced but
1
999, 183, 35.
remains unchanged when compared with MnO . In contrast,
x
as mentioned above, BaMn0.10 displays the best CH oxida-
4
4
5
E. S. Rubin, R. N. Cooper, R. A. Frosch, T. H. Lee, G. Marland,
A. H. RosenÐeld and D. D. Stine, Science, 1992, 257, 148.
M. M. Zwinkel, S. G. Jaras and P. C. Menon, Catal. Rev. Sci.
Eng., 1993, 35, 319.
tion activity among the three samples. This seems to imply
that the addition of Ba produces some suitable sites or cata-
lytic functions for e†ective CH oxidation.
6
7
C. F. Cullis and B. M. Willat, J. Catal., 1983, 83, 267.
R. F. Hicks, H. Qi, M. L. Young and R. G. Lee, J. Catal., 1990,
4
CH is well known to be the most difficult to oxidize
4
1
22, 280.
among all of the hydrocarbons, since the molecule contains
8
9
K. Sekizawa, H. Widjaja, S. Maeda, Y. Ozawa and K. Eguchi,
Appl. Catal., A, 2000, 200, 211.
only very strong CÈH bonds.1 The Ðssion of the Ðrst CÈH
bond could be the rate-determining step for CH oxidation,
S. Imamura, H. Sawada, K. Uemura and S. Ishida, J. Catal.,
1988, 109, 198.
4
and needs high temperature.29h32 Indeed, the activation of
CH should be more difficult than CO on the catalysts, as
10 M.-F. Luo, X.-X. Yuan and X.-M. Zheng, Appl. Catal., A, 1998,
4
175, 121.
demonstrated by the higher activation energy of CH as
4
opposed to CO oxidation. When studying CH oxidative
4
1
1 K.-S. Song, S.-K. Kang and S. D. Kim, Catal. L ett., 1997, 49, 65.
12 M. I. Zaki, M. A. Hasan, L. Pasupulety, N. E. Fouad and H.
coupling, Campbell et al.32 found that the relative activity of
Knozinger, New J. Chem., 1999, 23, 1197.
M. Baldi, V. S. Escribano, J. M. G. Amores, F. Milella and G.
rare-earth metal oxides parallels their basicity. It was pointed
out that this could be due to the better ability of more basic
1
3
Busca, Appl. Catal., B, 1998, 17, L175.
14 D. Mehandjiev, E. Zhecheva, G. Ivanov and R. Ioncheva, Appl.
Catal., A, 1998, 167, 277.
metal oxides to abstract one hydrogen atom from a CH mol-
4
ecule to form a methyl radical,32h34 the important interme-
1
5
G. J. Hutchings, A. A. Mirzaei, R. W. Joyner, M. R. H. Siddqui
and S. H. Taylor, Catal. L ett., 1996, 43, 21.
diate in the reaction. More recently, it has also been suggested
that for efficient activation of CH in combustion, highly basic
4
16 R. Burch, S. Chalker, G. D. Squire and S. C. Tsang, J. Chem.
Soc., Faraday T rans., 1990, 86, 1607.
17 X. Wang and Y. C. Xie, Fenzi Cuihua, 1998, 12, 312.
18 X. Wang and Y. C. Xie, React. Kinet. Catal. L ett., 2000, 71, 263.
sites are required in a catalyst.29 Previous work has already
demonstrated that the alkalization of metal oxides such as
manganese oxides or tin oxide with Li, K or Ba can improve
the basicity of their lattice oxygen or form more basic
sites.32h35 We thus speculate that in this study, the predomi-
1
2
2
9
0
1
E. C. Pitzer and J. C. W. Frazer, J. Phys. Chem., 1941, 45, 761.
S. B. Kanungo, J. Catal., 1979, 58, 419.
X. Wang and Y. C. Xie, React. Kinet. Catal. L ett., 2000, 70, 43.
nant reason leading to the enhanced CH oxidation activity of
4
22 E. R. Stobbe, B. A. De Boer and J. W. Geus, Catal. T oday, 1999,
the Ba-modiÐed manganese oxide catalysts could actually be
47, 161.
attributed to their improved basicity, which is favorable for
23 H. Kung and M. Kung, Adv. Catal., 1984, 33, 159.
24 I. R. Leith and M. G. Howden, Appl. Catal., 1988, 37, 75.
the activation of the Ðrst CÈH bonds in CH .
4
968
New J. Chem., 2001, 25, 964È969

View Article Online
2
5
C. Backx, C. P. M. De Groot, P. Biloen and W. M. H. Sachtler,
Surf. Sci., 1983, 128, 81.
31 V. R. Choudhary and V. H. Rane, J. Catal., 1991, 130, 411.
32 K. D. Campbell, H. Zhang and J. H. Lunsford, J. Phys. Chem.,
1988, 92, 750.
2
2
6
7
K. Bethke and H. H. Kung, J. Catal., 1997, 172, 93.
G. M. Schwab and S. B. Kanungo, Z. Phys. Chem., Neue Folge,
33 I. A. Kursina, L. N. Kurina, A. I. Galanov and S. I. Galanov,
Catal. T oday, 1998, 42, 263.
1
977, 107, 109.
A. P. B. Sinha, N. R. Sanjana and A. B. Biswas, J. Phys. Chem.,
958, 62, 191.
R. Burch, D. J. Crittle and M. J. Hayes, Catal. T oday, 1999, 47,
29.
J. Au-Yeung, K. Chen, A. T. Bell and E. Iglesia, J. Catal., 1999,
88, 132.
28
29
30
34 S. I. Galanov, L. N. Kurina, A. I. Galanov, A. A. Davydov and
V. N. Belousova, Catal. T oday, 1995, 24, 293.
1
35 M. I. Zaki, M. A. Hasan, L. Pasupulety, N. E. Fouad and K.
Kumari, New. J. Chem., 1998, 22, 875.
2
1
New J. Chem., 2001, 25, 964È969
969