Catalytic reactions of NO over 0-7 mol% Ba/MgO catalysts: II. Reduction with CH4 and CO

Article abstract of DOI:10.1006/jcat.1999.2649

Barium oxide supported on magnesium oxide (Ba/MgO) is an effective catalyst at elevated temperatures for the selective catalytic reduction (SCR) of nitric oxide, with methane as the reductant. The rate of the SCR reaction is greater than either the rate of reduction with CH₄, in the absence of O₂, or the rate of decomposition of NO. At 650°C the rate of the SCR reaction over 1 mol% Ba/MgO was 0.19 μmol g^-1s^-1 with 1% NO, 0.25% CH₄, and 0.5% O₂. The SCR reaction is believed to occur via gas-phase CH₃· radicals that are formed at the surface. First-order kinetics with respect to CH₄ is consistent with radical formation being the rate-determining step. The activities of the catalysts and the overall reaction order are influenced by CO₂, which is a product of the reaction, but also would be present in an exhaust stream. The inhibiting effect of CO₂ is much more significant for a catalyst containing 4 mol% Ba than for one containing 1 mol% Ba. In the latter case, the MgO support modifies the basicity of the BaO. In the absence of O₂, CO also is a reductant for NO, but if O₂ is present, the nonselective oxidation of CO occurs. That is, CO reacts preferentially with O₂ instead of NO.

Full text of DOI:10.1006/jcat.1999.2649

Journal of Catalysis 188, 32–39 (1999)
Article ID jcat.1999.2649, available online at http://www.idealibrary.com on
Catalytic Reactions of NO over 0–7 mol% Ba/MgO Catalysts
II. Reduction with CH and CO
4
1
Shuibo Xie, Michael P. Rosynek, and Jack H. Lunsford
Department of Chemistry, Texas A&M University, College Station, Texas 77843
Received January 12, 1999; revised July 20, 1999; accepted July 26, 1999
workers (6, 7), who were the first to study these catalysts
Barium oxide supported on magnesium oxide (Ba/MgO) is an for SCR of NO, noted that they also were active and se-
effective catalyst at elevated temperatures for the selective catalytic lective for the oxidative coupling of methane, a reaction
reduction (SCR) of nitric oxide, with methane as the reductant. The
rate of the SCR reaction is greater than either the rate of reduction
that is known to involve CH3� radicals (9). Furthermore,
they observed that the presence of NO strongly inhibited
with CH4, in the absence of O2, or the rate of decomposition of NO.
the formation of the coupling products, C2H4 and C2H6,
�
At 650 C the rate of the SCR reaction over 1 mol% Ba/MgO was
which is consistent with the reaction of NO with the CH3�
�
1
� 1
0
.19 �mol g
s with 1% NO, 0.25% CH4, and 0.5% O2. The SCR
radicals. Methyl radicals have also been postulated as in-
termediates in the SCR of NO with CH4 over Co–ZSM-5;
however, there is less direct evidence for formation of rad-
icals over this catalyst (10–12). In fact, we have attempted
reaction is believed to occur via gas-phase CH3� radicals that are
formed at the surface. First-order kinetics with respect to CH4 is
consistent with radical formation being the rate-determining step.
The activities of the catalysts and the overall reaction order are
influenced by CO2, which is a product of the reaction, but also without success to detect CH3� radicals over Co–ZSM-5
would be present in an exhaust stream. The inhibiting effect of CO2 by using a variable ionization energy mass spectrometer
is much more significant for a catalyst containing 4 mol% Ba than (13).
for one containing 1 mol% Ba. In the latter case, the MgO support
modifies the basicity of the BaO. In the absence of O2, CO also is a
reductant for NO, but if O2 is present, the nonselective oxidation of
CO occurs. That is, CO reacts preferentially with O2 instead of NO.
It was demonstrated in our laboratory that the reaction
of CH4 with O2 indeed results in the formation of CH3�
radicals over Sr/La2O3 and Ba/MgO catalysts (8, 14). These
radicals emanate into the gas phase, where they may couple
�
c 1999 Academic Press
or react with NO, if present. The concentration of radicals
decreased when the argon diluent was replaced by nitric ox-
ide. Moreover, when CH3� radicals, derived from the ther-
mal decomposition of azomethane, were allowed to react
with NO in the gas phase, nitrosomethane, CH3NO, was
detected by mass spectrometry (8). This reaction occurred
Key Words: nitric oxide; methane; barium oxide; selective cata-
lytic reduction.
1
. INTRODUCTION
�
at 25 C, but CH3NO was detected in decreasing amounts
Methane is one of the more desirable reductants for the
removal of NO because of its abundance and availability;
however, methane is difficult to activate. Nevertheless, a
number of different catalysts, including cobalt-exchanged
zeolites (1–3), gallium- and indium-loaded H–ZSM-5 (4),
and Ce/Ag–ZSM-5 (5), have been reported to be active and
selective for the selective catalytic reduction (SCR) of NO
with CH4 at moderate temperatures.
Among the various types of materials that have activ-
ity for the SCR of NO with CH4, the alkaline earth and
rare-earth oxides (e.g., Li/MgO and Sr/La2O3) are unique
in that the mechanism of the reaction appears to involve
surface-generated methyl radicals (6–8). Vannice and co-
�
at temperatures up to 800 C.
In the study reported here, the more general properties
of Ba/MgO catalysts containing up to 7 mol% Ba were de-
termined, with emphasis on the kinetics of the reaction. The
SCR reaction with CH4 was investigated, as well as the re-
action with CH4 and CO in the absence of O2. Particular
attention was given to the negative effect that CO2 has on
these reactions, in part because of the role that CO2 plays
in the interpretation of kinetic results, but also because of
the presence of large concentrations of CO2 in any practical
application of these catalysts for the removal of NO from
combustion products. No attempt was made to determine
the inhibiting effect of SO2 or H2O. The results of these
reduction reactions are compared with those obtained dur-
ing the direct decomposition of NO over the same set of
catalysts, as described in Part I (15).
1
To whom correspondence should be addressed. E-mail: Lunsford@
mail.chem.tamu.edu.
32
0021-9517/99 $30.00
Copyright �c 1999 by Academic Press
All rights of reproduction in any form reserved.

CATALYTIC REACTIONS OF NO OVER BaMgO CATALYSTS, II
33
TABLE 1
exhibited the largest specific activity, in part because of its
smaller surface area (Table 1). On a per gram basis, the
1 mol% Ba/MgO catalyst was the most active, as shown in
Activation Energies of NO Reduction by Methane
a
over Ba/MgO Catalysts
�
Fig. 1B, and the NO conversion at 800 C was 40% , which is
Surface
area
about six times the conversion for NO decomposition over
the same catalyst.
Activation energy (kcal/mol)
2
� 1
c
d
e
Catalysts
MgO
(m g )b NO + CH4
NO + CH4 + O2
NO + CO
With 0.25% CH4 added to 1% NO, the N2 formation rate
�
at 800 C over the 7 mol% Ba/MgO catalyst was 0.040 �mol
38.3
27.8
13.3
4.6
31.7
32.2
30.8
30.7
23.7
31.8
36.8
29.4
28.2
31.2
27.2
24.7
�
�
2
2
� 1
1
4
7
% Ba/MgO
% Ba/MgO
% Ba/MgO
m
m
s , which may be compared to a value of 0.015 �mol
� 1
s
for the direct decomposition of NO over the same
�
catalyst at 800 C (15). The rate for the direct decomposition
a
�
is not negligible compared to the N formation rate during
the reaction of NO with CH4; however, the direct decompo-
sition of NO is strongly suppressed by the presence of CO2
Kinetic data were obtained in the temperature range 600–800 C and
with NO conversion less than 20% .
2
b
�
After reaction of 1% NO with 0.25% CH4 and 0.5% O2 at 800 C for
1
hr.
(
see below), which is an inevitable product after CH4 is in-
c
d
e
Reagent gas composition, 1% NO + 0.25% CH4.
Reagent gas composition, 1% NO + 0.25% CH4 + 0.5% O2.
Reagent gas composition, 1% NO + 1% CO.
troduced. Therefore, the reduction reaction is the primary
pathway for the formation of N when CH is present.
2
4
2
. EXPERIMENTAL
3.2. Selective Catalytic Reduction of NO with CH4
in the Presence of O2
The catalysts, reactor, and analytical methods were the
�
At temperatures <750 C, the presence of 0.5% O2 in
same as those reported in Part I (15). Reactant gases were
obtained by mixing 4% NO/He, 1% CH4/He or 1% CO/He,
and 5.4% O2/He. Pure He was used as a makeup gas to keep
the total flow rate at 40 ml/min. Typically 200 mg catalyst
was used. The surface areas of the catalysts used are re-
ported in Table 1.
the reagent stream further enhanced the N2 formation rate
during NO reduction by CH4, as shown in Fig. 2A. For ex-
ample, with the 7 mol% Ba/MgO catalyst at 700 C, the N2
�
�
2
� 1
formation rate increased from 0.008 �mol m
s in the
The CH4 conversion was based on either the disappear-
ance of CH4 or the formation of CO2. Only trace amounts of
CO were detected asproductsduringNO reduction byCH4.
In the presence of O2, there are two main reaction pathways
that are responsible for the disappearance of CH4,
CH4 + 2NO + O2 → N2 + CO2 + 2H2O
CH4 + 2O2 → CO2 + 2H2O.
[1]
[2]
Reaction [1] is taken to be the selective oxidation reac-
tion, which produced N2, and in Reaction [2], only CO2 and
H2O are formed. The selectivity of CH4 is defined as the
percentage of CH4 reacted via reaction [1]; i.e., S(CH4) =
[
N2]/([CO2]1 + [CO2]2) � 100 = [N2]/[CO2]total � 100. Selec-
tivities defined in this manner are a factor of two larger
than those obtained using the definition of Vannice and co-
workers (16). Unless stated otherwise, rates were obtained
by assuming differential conditions up to ca. 30% conver-
sion, which is a reasonable approximation considering the
apparent orders of reaction (see below).
3
. RESULTS
3
.1. Reduction of NO by CH4 in the Absence of O2
FIG. 1. Nitrogen formation rate (A) and NO conversion (B) as a func-
tion of temperature during the reaction of 1% NO + 0.25% CH4 over MgO
and Ba/MgO catalysts: (᭿) MgO, (ᮀ) 0.5 mol% Ba/MgO, (᭹) 1 mol%
Ba/MgO, (᭺) 4 mol% Ba/MgO, (᭡) 7 mol% Ba/MgO. Catalyst weight,
Even in the absence of added oxygen, the reduction of
NO with CH4 occurs at a moderate rate, as shown in Fig. 1A.
Among the materials studied, the 7 mol% Ba/MgO catalyst 0.2 g; flow rate, 40 ml/min.

3
4
XIE, ROSYNEK, AND LUNSFORD
�
sion was observed at temperatures up to 875 C. At tem-
peratures that are higher than the maximum, other reac-
tions occur, such as the direct decomposition of NO, and
the relative rates of these reactions also will define the
maximum.
In the selective catalytic reduction of NO by CH4, two
reactions compete for methane, with Reaction [1] being
desirable and Reaction [2] being undesirable. The CH4 se-
lectivity as a function of reaction temperature is shown
in Fig. 3B. For the 4 mol% Ba/MgO catalyst the car-
bon balance was within 95% of the theoretical value. The
1
mol% Ba/MgO sample had exceptionally high CH4
selectivity (>80% ) over the entire temperature range that
was studied, and the selectivity approached 100% at tem-
�
peratures >700 C. This high selectivity provides a signifi-
cant advantage for the practicalutilization ofsuch a catalyst.
At both larger and smaller barium loadings, and for tem-
peratures corresponding to a reasonable level of NO con-
�
version (>700 C), the selectivities were less, with 7 mol%
Ba/MgO being the least selective.
3
.3. Reduction of NO by CO
Carbon monoxide is an active reductant for NO over
FIG. 2. Reaction rates (A) and NO conversions (B) as a function
of temperature during the reaction of 1% NO + 0.25% CH4 + 0.5% O2 Pt–Rh–Pd three-way catalysts (17). With Ba/MgO as the
over MgO and Ba/MgO catalysts: (᭿) MgO, (ᮀ) 0.5 mol% Ba/MgO,
catalyst, CO was not observed as a product during the SCR
of NO by CH4. The absence of CO may result from the fact
(
᭹) 1 mol% Ba/MgO, (᭺) 4 mol% Ba/MgO, (᭡) 7 mol% Ba/MgO.
�
2
� 1
s in the presence of O2.
absence of O2 to 0.022 �mol m
Again, the specific activity was the largest for the 7 mol%
Ba/MgO catalyst, but when the conversions were compared
for a fixed mass of catalyst, as shown in Fig. 2B, the NO
conversion was the largest over the 1 mol% Ba/MgO cata-
�
lyst. At 650 C, the rate of reaction over this catalyst was
�
1
� 1
0
.19 �mol g s .
As the temperature was increased above ca. 750 C, the
�
NO conversion no longer increased continuously, and at
a certain temperature it went through a maximum. With
the 1 mol% Ba/MgO catalyst, the maximum was not as
pronounced. A similar phenomenon has been reported for
the SCR reaction over many other catalysts and is due
to the fact that the CH4 conversion approaches 100% at
a particular temperature. For the Ba/MgO catalysts, this
is made evident by comparing the results of Fig. 3A with
the temperatures of the maxima, which were between 750
�
and 850 C, as shown in Fig. 2B. Obviously, the tempera-
ture at which the maximum NO conversion occurs depends
on the catalytic activity and selectivity for CH4 oxidation,
as well as the loading of the catalyst. The maximum for
the 1 mol% Ba/MgO catalyst was at a lower temperature,
which is consistent with the fact that the CH4 conversion
approached 100% at a lower temperature (Fig. 3A). Since
the reaction is CH4 limited, the concentration of CH4 and
O2 may affect the temperature maximum as well. With
FIG. 3. Nitric oxide and methane conversions during the reaction
of 1% NO + 0.25% CH4 + 0.5% O2 over MgO and Ba/MgO catalysts:
(
᭿) MgO, (ᮀ) 0.5 mol% Ba/MgO, (᭹) 1 mol% Ba/MgO, (᭺) 4 mol%
an excess of methane (25% ), no maximum in NO conver- Ba/MgO, (᭡) 7 mol% Ba/MgO. Catalyst weight, 0.2 g; flow rate, 40 ml/min.

CATALYTIC REACTIONS OF NO OVER BaMgO CATALYSTS, II
35
�
CO conversion was nearly 100% for T > 700 C (Fig. 5B),
confirms that CO reacts preferentially with O2, rather
than with NO. This is supported by the fact that the
CO conversion was less when only NO and CO were
present as reagents (i.e., no O2). In summary, the small de-
crease in NO conversion that resulted from the addition
of CO to the standard SCR mixture with CH4 (Fig. 5A)
is a result of the poisoning effect of CO2. The rapid reaction
of CO with O2 led to a higher concentration of CO2 in the
gas stream. Similarly, with the mixture of 1% NO + 0.25%
CO + 0.5% O2, all of the CO rapidly reacted to form CO2,
and even the direct decomposition of NO to form N2 and
O2 was poisoned. This set of experiments establishes that
CO is not a primary intermediate in the SCR of NO with
CH4.
3
.4. Inhibition by CO2
Carbon dioxide strongly inhibits the oxidative coupling
of methane over basic catalysts due to the formation of
inactive surface carbonates (18). The effect of CO2 on
NO decomposition and reduction by CH4 was studied by
varying the flow rate of a 10% CO2/He mixture while
FIG. 4. Reaction rates (A) and NO conversions (B) as a function
of temperature during the reaction of 1% NO + 1% CO over MgO and
Ba/MgO catalysts: (᭿) MgO, (᭹) 1 mol% Ba/MgO, (᭺) 4 mol% Ba/MgO,
� 1
keeping the total flow rate of the reagents at 40 ml min
.
From the results shown in Fig. 6A, it is evident that CO2
had a negative effect on the N2 formation rate for the
(
᭡) 7 mol% Ba/MgO. Catalyst weight, 0.2 g; flow rate, 40 ml/min.
that no CO was formed, or that CO rapidly reacted with
NO or O2 so that it existed only in trace amounts. To clarify
the role of CO in the SCR reaction over Ba/MgO catalysts,
the reaction of NO with CO was also studied. The overall
reaction of NO with CO is
2CO + 2NO → N2 + 2CO2.
[3]
With a stoichiometric mixture containing 1% NO and 1%
CO, the results shown in Fig. 4 were obtained over MgO
and the Ba/MgO catalysts. Again, the 7 mol% Ba/MgO
catalyst had the largest specific rate for N2 formation, but
the 1 mol% Ba/MgO was the most active on a per gram
basis, and resulted in the largest NO conversion (Fig. 4B).
A comparison of the results of Figs. 1 and 4 indicates that for
the larger amount of CO (1% CO versus 0.25% CH4), the
N2 formation rates were greater with CO than with CH4.
However, when the same amounts of CO and CH4 (0.25% )
were employed, the rate of NO reduction was larger with
CH4 as the reductant (see below).
The NO conversions obtained with several different gas
mixtures are compared in Fig. 5A for a 4 mol% Ba/MgO
catalyst. The highest conversion was obtained for the stan-
dard mixture of 1% NO + 0.25% CH4 + 0.5% O2. Add-
ition of 0.25% CO to this mixture actually caused a slight
decrease in conversion at T < 800 C. With the deletion
of CH4 from the mixture, almost no nitric oxide was re-
duced. This result, together with the observation that the 0.5% O2. Catalyst, 0.2 g 4 mol% Ba/MgO.
FIG. 5. Comparison ofNO and CO conversionswith different reagent
gases over 200 mg of 4 mol% Ba/MgO catalyst. Gas composition:
�
(
᭺) 1% NO + 0.25% CH4 + 0.5% O2, (᭹) 1% NO + 0.25% CO, (᭡) 1%
NO + 0.25% CH4 + 0.25% CO + 0.5% O2, (᭢) 1% NO + 0.25% CO +

3
6
XIE, ROSYNEK, AND LUNSFORD
of CO2 as the product of NO reduction by CH4 may ex-
plain why the reaction rates of NO with CH4 and NO with
CH4 and O2 were reduced to a smaller extent when CO2
was added to the gases. As shown in Fig. 6B, the 1 mol%
Ba/MgO catalyst was much less susceptible to CO2 poison-
ing during the reduction of NO with CH4 in the presence of
O2 than was the 4 mol% Ba/MgO catalyst, which is consis-
tent with a decrease in basicity of the material (see below).
3
.5. Kinetic Parameters for NO Reduction by CH4
The activation energies for N2 formation with mixtures
consisting of 1% NO + 0.5% CH4, 1% NO + 0.25% CH4 +
0
.5% O2, and 1% NO + 0.25% CO over MgO and Ba/MgO
catalysts were determined and the results are given in
Table 1. For the simple reduction with CH4 the activation
�
1
energy was 31 � 1 kcal mol for MgO and the three
Ba/MgO catalysts. When O2 was added, the variation was
�
1
greater, with a maximum of 36.8 kcal mol occurring for
the 4 mol% Ba/MgO catalyst. A similar activation energy
of 39 kcal mol was observed for CH4 oxidation coupling
over this catalyst (18), which is consistent with the common
role of CH3� radicals in the coupling and the SCR reactions.
Agreement in activation energies for the two reactions also
�
1
FIG. 6. The effect of added CO2 on the reaction rates of NO decom-
position and reduction with CH4. (A) 4 mol% Ba/MgO catalyst (0.3 g): was found for the 1 mol% Ba/MgO catalyst. The activation
�
1
(
᭡) 1% NO, (᭹) 1% NO + 0.25% CH4, (᭿) 1% NO + 0.25% CH4 + 0.5% energy of 29.4 kcal mol for the SCR reaction over the
O2. (B) 1 mol% Ba/MgO catalyst (0.2 g): (4) 1% NO, (᭺) 1% NO + 0.25%
7
mol% Ba/MgO, however, was unexpectedly low when
compared to a value of 43 kcal mol (18) for the coupling
reaction over 8 mol% Ba/MgO catalyst. The difference
�
CH4, (ᮀ) 1% NO + 0.25% CH4 + 0.5% O2. T = 750 C.
� 1
reagent gas compositions 1% NO, 1% NO + 0.25% CH4, could have resulted, in part, from the temperature-
and 1% NO + 0.25% CH4 + 0.5% O2, but the decrease was dependent poisoning effect of CO2. In the coupling reac-
much more dramatic for the direct decomposition of NO tion, the amount of CO2 formed was much larger and the
over the 4 mol% Ba/MgO catalyst. In this case, the N2 for- effect was more significant at the higher Ba loadings. With
�
� 3
� 2
mation rate at 750 C decreased from 5.1 � 10 �mol m
CO as the reductant, one might have expected a signif-
�
1
� 3
� 2 � 1
s
s
to 0.8 � 10 �mol m
upon the addition of 0.25% icantly different activation energy because of a different
mechanism. Such was the case for the 4 mol% Ba/MgO
CO2 to the reagent gas.
Thus, compared to the direct decomposition of NO, the catalysts, but not for the 1 mol% Ba/MgO catalyst.
reduction of NO by CH4 was less susceptible to the CO2 poi- To determine the reaction orders during NO reduction
soning, especially when O2 was present. For the reduction of over the 4 mol% Ba/MgO catalyst, it was necessary to ac-
NO by CH4 in the absence of O2, the reaction rate was sen- count for the CO2 poisoning effect on the activity. The
sitive to the presence of CO2 for concentrations up to 1% , amount of CO2 formed during NO reduction by CH4 var-
but the CO2 poisoning effect was less significant when the ied with reaction conditions such as the reaction tempera-
CO2 concentration was greater than 1% . In the presence of ture and the concentrations of oxygen and methane. Larger
O2, even with 4% CO2 in the feed gas, the N2 formation rate amounts of CO2 are expected when the CH4 concentration
over the 4 mol% Ba/MgO catalyst remained at a relatively is increased, if the conversion of CH4 is constant. In order
�
3
� 2 � 1
high value of 6.7 � 10 �mol m s , which is about half to overcome the effects that result from a variation in the
of the N2 formation rate without the addition of CO2. Even amount of CO2, an excess of CO2 (2.5% ) was introduced
when CO2 was not added to the reagents, it was, of course, into the feed gas to minimize the effect of CO2 produced
present as a product when CH4 was introduced as the re- during the reaction. Thus the reaction became pseudo zero
�
ductant. In the absence and presence of O2 at 750 C, the order with respect to CO2 (Fig. 6). Another advantage in
products contained ca. 0.08 and 0.22% CO2, respectively, at determining the reaction orders for NO reduction in an ex-
�
1
the normal flow rate of 40 ml min . The activity for N2 for- cess amount of CO2 is that the NO decomposition is almost
�
2
mation increased by nearly a factor of 4.5 to 4.2 � 10 �mol completely suppressed with 2.5% CO2 added. That is, the
�
2
� 1
� 1
m
s
when the flow rate was 480 ml min , at which point contribution of N2 formation from the direct decomposition
the product gas contained only 0.07% CO2. The existence of NO is minimized.

CATALYTIC REACTIONS OF NO OVER BaMgO CATALYSTS, II
37
In these experiments, the concentration of CH4 was kept
constant at 0.25% while the concentration of NO was var-
ied between 0.5 and 3.5% . When the reaction order with
respect to CH4 was determined, the NO concentration was
1
% and the CH4 concentration was varied between 0.075
and 0.75% . The initial concentration of O2, when present,
was 0.5% . The conversions for NO and CH4 were kept less
than 20% in experiments determining the kinetic orders.
The experimental results are summarized in Table 2.
In the absence of O2, the reaction order with respect to
CH4 in the reaction of NO with CH4 was almost the same
in the absence and presence of added CO2, with values of
0
.25 and 0.24, respectively. When the NO concentration was
fixed and the CH4 concentration was varied, the change in
the concentration of CO2 formed varied only slightly in
the absence of added CO2. There is, however, a difference
in reaction orders with respect to NO, with and without
added CO2. The NO order was higher (0.88 vs. 0.52) in the
presence of added 2.5% CO2. When the CH4 concentration
was fixed and the NO concentration was increased, more
CO2 was formed, and the increase in reaction rate was offset
by a larger inhibiting effect of CO2. The results in Table 2
also suggest that the N2 formation rate during the reaction
of NO with CH4 is less sensitive to the change in the CH4
concentration than that in the NO concentration.
FIG. 7. Nitrogen formation rates as a function of O2 concentration
during the reaction of 1% NO with 0.25% CH4 and different concentra-
tions of O2: (᭿) N2 formation rate, (᭹) CH4 conversion. Catalyst, 0.2 g
�
4
mol% Ba/MgO; T = 750 C.
amounts of O2 to a mixture of 1% NO + 0.25% CH4 in-
creased the NO conversion up to a maximum with 0.5% O2,
which was the amount used in the experiments described
above. Larger O2 concentrations resulted in a decrease in
the NO conversion, largely because the conversion of CH4
approached 100% . When the added O2 was greater than
With the reaction mixture consisting of NO, CH4, and
O2, the reaction orders with respect to NO and CH4, de-
termined in the presence of added CO2, were significantly
higher than those obtained in the absence of added CO2
2
% , the NO conversion was even less than that observed
(
Table 2). In contrast to the reaction of NO with CH4 in
with 1% NO + 0.5% CH4 in the absence of O2. Therefore,
a positive reaction order with O2 during the reaction of NO
with CH4 and O2 would be obtained if the O2 concentration
were kept lower than 0.5% , while a negative value would
be observed if larger O2 concentrations were employed.
the absence of O2, the reaction orders for both NO and
CH4 were affected by the presence of added CO2. The re-
action orders with respect to CH4 were 0.68 in the absence
of added CO2 and 1.02 with 2.5% CO2 added in the feed gas.
The reaction orders with respect to NO were 0.21 and 0.36,
respectively, without and with the added CO2. The different
reaction orders with and without O2 suggest that the reac-
tion mechanisms may be different for these two reactions.
No attempt has been made to determine quantitatively
the reaction order with respect to O2 since the effect of
O2 on the N2 formation rate depends on the concentra-
tion used. The order with respect to O2 may be either pos-
itive or negative, as shown in Fig. 7. The addition of small
4
. DISCUSSION
As described in the Introduction, there is evidence that
CH3� radicals, formed on the surface of Ba/MgO, may be
involved in the SCR of NO (14). The CH3� radicals subse-
quently react with NO in the gas phase to produce N2O,
and ultimately N2. The following mechanism appears to be
consistent with the evidence obtained for these catalysts:
→
O2 + 2s� 2Os
[4]
[5]
TABLE 2
CH4 + Os → CH3�+ OHs (RDS)
OHs → H2O + Os
CH3�+ NO → CH3NO
CH3NO + NO →→ N2O + CH3O�
N2O + s → N2 + Os
CH3O�+ O2 →→ CO2 + H2O
CH3�+ O2 →→ CO2 + H2O.
Reaction Orders of NO and CH4 over a 4 mol% Ba/MgO Catalysta
2
[6]
NO + CH4
NO CH4
NO + CH4 + O2
NO CH4
[7]
[8]
Without added CO2
0.52 � 0.02 0.25 � 0.01 0.21 � 0.07 0.68 � 0.03
[9]
With 2.5% CO2 added 0.88 � 0.03 0.24 � 0.01 0.36 � 0.01 1.02 � 0.04
[10]
[11]
a
Reaction orders were obtained with conversions for NO and CH4 less
than 20% .

3
8
XIE, ROSYNEK, AND LUNSFORD
The surface species, Os and OHs, are probably present as cals would desorb, yet almost none were detected. In ad-
�
�
ions (i.e., Os and OH ). If the active form of oxygen is a dition, NO and CH4 would be only weakly adsorbed on
s
peroxide ion, as might be the case for Ba/MgO (19), one these closed shell oxides at the reaction temperatures em-
could write an analogous reaction mechanism. The forma- ployed. Considering the fact that Ba/MgO reacts with NO
tion of CH3� radicals is suggested to be the rate-determining to form nitrate and nitrite ions that exist at elevated tem-
step (RDS), which is consistent with the reaction being first peratures (15), we propose that gas-phase CH4 reacts with
order with respect to CH4. The reaction of CH3� radicals these polyatomic anions to reduce the nitrogen. The reac-
with NO to form CH3NO must be rapid since C2H6 and tions through which N2 is formed are not known, but they
C2H4, the products of methyl radical coupling, were not de- probably involve ionic intermediates. One cannot exclude
tected. Nitrosomethane may react with NO through a series the possibility that, even in the presence of O2, a nonradical
of steps to form N2O (20). Another possible pathway for mechanism accounts for part of the NO conversion.
the formation of N2 may be through a bimolecular reac-
tion of CH3NO; however, this is probably a minor pathway,
considering the fact that the concentration of NO is much
larger than that of CH3NO. The 0.36 order with respect to
5
. CONCLUSIONS
The SCR of NO with CH over a series of Ba/MgO cata-
4
NO may be the result of several reactions. As shown pre- lysts occurs at a significantly higher rate than either the
viously, NO had a negative effect on CH3� radical forma- direct decomposition of NO or the reduction of NO in
tion (8), which could result from the competitive reactions the absence of O . The SCR reaction is partially poisoned
2
of NO with the Os species. But subsequent reactions (e.g., by CO , which would be present as a product of combus-
2
Reactions [7] and [8]) would require the presence of NO in tion. This poisoning effect is less on the catalyst containing
the formation of N2.
1 mol% Ba than on the one containing 4 mol% Ba because
It was demonstrated that N2O is a reaction intermediate at low loadings the basicity of the barium oxide is mod-
during NO decomposition over Ba/MgO catalysts (15). In erated by the less basic MgO support. When the effect of
the SCR ofNO with CH4, the role ofN2O asan intermediate CO is taken into account, the reaction becomes first order
2
is uncertain, although it was detected in small amounts in with respect to CH , which is consistent with CH � radicals
4
3
�
the product stream at reaction temperatures <700 C. The being formed in the rate-determining step. In the absence
presence of CH4 inhibited the decomposition of N2O. One of O , the order with respect to methane decreased to 0.24,
2
could replace Reaction [8] by
which suggested a change in the reaction mechanism. The
details of this mechanism are not known, but it probably
involves the reaction of CH4 with ionic NOx species, rather
than adsorbed NO.
CH3NO + NO →→ N2 + CH3O �,
[8a]
2
which would not lead to the formation of N2O.
Since the formation of CH3� radicals is believed to be a
common feature in both the SCR of NO and the oxidative
coupling of CH4, it is of interest to compare a common fea-
ture ofthe Ba/MgO catalystsfor these reactions, namely, the
resistance of the 1 mol% Ba/MgO catalyst to poisoning by
CO2. This resistance to poisoning is of practical significance
since CO2 is a major product of combustion. At low BaO
loadings (<2 mol% ), the basicity of the Ba/MgO catalysts,
as determined by the temperature-programmed desorption
of CO2, decreased as a result of interaction with the much
less basic MgO support (18). Hence, the 1 mol% Ba/MgO
catalyst was much less susceptible to CO2 poisoning than
the 4 mol% Ba/MgO catalyst.
The mechanism for the reduction of NO with CH4, in the
absence of O2, is less certain because CH3� radicals were de-
tected only in very small concentrations over Ba/MgO and
Sr/La2O3 catalysts when O2 was replaced by NO (8, 14).
Vannice et al. (21) tentatively adopted a mechanism that
agreed with their kinetic results over La2O3 and Sr/La2O3.
In this mechanism it was proposed that adsorbed NO and
CH4 reacted in the rate-determining step to form adsorbed
HNO and CH3� radicals. Based on results obtained with
CH4 and O2 (8, 14), it was expected that the CH3� radi-
ACKNOWLEDGMENT
This research was supported by the National Science Foundation under
Grant CHE-9520806.
REFERENCES
1
2
. Li, Y., and Armor, J. N., J. Catal. 145, 1 (1994).
. Li, Y., and Armor, J. N., in “Natural Gas Conversion II” (H. E. Curry-
Hyde and R. F. Howe, Eds.), p.103. Elsevier, Amsterdam, 1994.
. Kim, Y. G., Kim, Y. C., Hong, S. B., Kim, M. H., Kim, Y. P., and Uh,
Y. S., Catal. Lett. 57, 179 (1999).
4. Kikuchi, E., Ogura, M., Terasaki, I., and Goto, Y., J. Catal. 161, 469
1996).
. Li, Z. J., and Flytzani-Stephanopoulos, M., Appl. Catal. A Gen. 165,
5 (1997).
. Zhang, X., Walters, A. B., and Vannice, M. A., J. Catal. 146, 568
1994).
7. Zhang, X., Walters, A. B., and Vannice, M. A., Appl. Catal. B Environ.
, 321 (1996).
3
(
5
6
1
(
7
8
. Xie, S., Ballinger, T. H., Rosynek, M. P., and Lunsford, J. H., Stud. Surf.
Catal. 101, 711 (1996). [11th International Congress on Catalysis]
. Lunsford, J. H., Angew. Chem. Intl. Ed. Engl. 34, 970 (1995).
9
1
0. Cowan, A. D., D u¨ mpelmann, R., and Cant, N. W., J. Catal. 151, 356
(1995).

CATALYTIC REACTIONS OF NO OVER BaMgO CATALYSTS, II
39
1
1. Lukyanov, D. B., Sill, G., d’Itri, J., and Hall, W. K., J. Catal. 153, 265
1995).
16. Zhang, X., Walters, A. B., and Vannice, M. A., J. Catal. 155, 290 (1995).
17. Armor, J. N., Appl. Catal. B Environ. 1, 221 (1992).
(
1
2. Lukyanov, D. B., Lombardo, E. A., Sill, G. A., d’Itri, J., and Hall, W. K., 18. Dissanayake, D., Lunsford, J. H., and Rosynek M. P., J. Catal. 143, 286
J. Catal. 163, 447 (1996).
(1993).
1
1
1
3. Xie, S., and Lunsford, J. H., unpublished results.
4. Xie, S., Rosynek, M. P., and Lunsford, J. H., Catal. Lett. 43, 1 (1997).
5. Xie, S., Rosynek, M. P., and Lunsford, J. H., J. Catal. 188, 24 (1999).
19. Mestl, G., Rosynek, M. P., and Lunsford, J. H., J. Phys. Chem. B 102,
154 (1998).
20. Johnston, T., and Heicklen, J., J. Phys. Chem. 70, 3089 (1966).
21. Vannice, M. A., Walters, A. B., and Zhang, X., J. Catal. 159, 119 (1996).
[Part I of this series]