Selective catalytic reduction of nitric oxide by methane in the presence of oxygen over CaO catalyst

Article abstract of DOI:10.1006/jcat.1999.2414

The selective catalytic reduction of nitric oxide by methane in the presence of oxygen was studied over CaO catalyst in the temperature range 550-850°C. The nitric oxide conversion-versus-temperature profile was found to depend on oxygen feed concentration, while the selectivity to N₂ formation was found to be 100% and independent of oxygen feed concentration in the range 1-10 mol%. The kinetic study has shown that the reaction order with respect to NO and CH₄ is 1.0 and 0.5, respectively. On the other hand, the reaction order with respect to O₂ was found to be slightly positive for O₂ feed concentrations up to 1 mol% and slightly negative at higher concentrations. The apparent activation energy of the reaction in the presence of oxygen was found to be 14.6 kcal/mol. The addition of 2.5 mol% CO₂ and 5 mol% H₂O in the feed stream had a considerable inhibiting effect on NO conversion in the range 550-650°C, while a small inhibiting effect of CO₂ and a positive effect of H₂O were observed in the range 700-850°C. Temperature-programmed desorption (TPD) of NO revealed the presence of two well-resolved NO peaks in the temperature range 250-650°C, a behavior also observed during a temperature-programmed oxidation experiment under O₂/He flow. However, during temperature-programmed surface reaction under CH₄/He flow the NO desorption spectrum showed a significant shift of the stronger adsorbed NO species toward lower temperatures. Preadsorbed CO₂ on the CaO surface was found to largely affect the chemisorption of NO and its desorption kinetics during TPD. The amount of deposited carbon-containing species on the catalyst surface during reaction was determined with transient titration experiments and it was found to be small.

Full text of DOI:10.1006/jcat.1999.2414

Journal of Catalysis 183, 323–335 (1999)
Article ID jcat.1999.2414, available online at http://www.idealibrary.com on
Selective Catalytic Reduction of Nitric Oxide by Methane
in the Presence of Oxygen over CaO Catalyst
,1
K. D. Fliatoura, X. E. Verykios, C. N. Costa,† and A. M. Efstathiou†
Department of Chemical Engineering, University of Patras, GR-26500 Patras, Greece; and †Department
of Natural Sciences, University of Cyprus, P.O. Box 537, Nicosia, Cyprus
Received September 16, 1998; revised January 6, 1999; accepted January 6, 1999
trolling this hazardous emission from power stations (1, 2),
but the use of NH3 as a reducing agent poses problems
of storage, transportation, high cost, and equipment cor-
rosion. Recently, the selective catalytic reduction of nitric
oxide with hydrocarbons (HC-SCR) has received increased
attention from both academic and industrial points of view,
because this may offer an effective way of removing NO
even in the presence of a large excess of O2 (3). Nitric ox-
ide reduction using CH4 as a reducing agent was studied as
early as 1984 over Rh/Al2O3 catalysts (4). A variety of ion-
exchanged zeolites, including Cu-, Fe-, Pt-, Co-, Ga-, Ce-,
and H-exchanged zeolites have been found to be active
for this reaction (5–11) but their suitability is limited due
to a volcano-like activity dependence on temperature and
their deactivation in the presence of large amounts of wa-
ter. Currently, ethylene, propane, propylene, and n-octane
hydrocarbons have been investigated for lean-NOx reduc-
tion (12–16). Methane is regarded as a nonselective reduc-
ing agent because it is difficult to activate due to the very
strong C–H bond which necessitates high reaction temper-
atures (10, 17, 18). On the other hand, with the abundance
of natural gas and its widespread use as a fuel by many util-
ities, it certainly would be desirable to use methane as a
reducing agent. Thus, the investigation of new catalytic sys-
tems that can effectively reduce NO with CH4 at relatively
high temperatures has attracted attention (19–23).
The selective catalytic reduction of nitric oxide by methane
in the presence of oxygen was studied over CaO catalyst in the
temperature range 550–850 C. The nitric oxide conversion-versus-
temperature profile was found to depend on oxygen feed concentra-
tion, while the selectivity to N2 formation was found to be 100% and
independent of oxygen feed concentration in the range 1–10 mol%.
The kinetic study has shown that the reaction order with respect
to NO and CH4 is 1.0 and 0.5, respectively. On the other hand, the
reaction order with respect to O2 was found to be slightly positive
for O2 feed concentrations up to 1 mol% and slightly negative at
higher concentrations. The apparent activation energy of the re-
action in the presence of oxygen was found to be 14.6 kcal/mol.
The addition of 2.5 mol% CO2 and 5 mol% H2O in the feed stream
had a considerable inhibiting effect on NO conversion in the range
550–650 C, while a small inhibiting effect of CO2 and a positive
effect of H2O were observed in the range 700–850 C. Temperature-
programmed desorption (TPD) of NO revealed the presence of two
well-resolved NO peaks in the temperature range 250–650 C, a
behavior also observed during a temperature-programmed oxida-
tion experiment under O2/He flow. However, during temperature-
programmed surface reaction under CH4/He flow the NO desorp-
tion spectrum showed a significant shift ofthestronger adsorbed NO
species toward lower temperatures. Preadsorbed CO2 on the CaO
surface was found to largely affect the chemisorption of NO and its
desorption kinetics during TPD. The amount of deposited carbon-
containing species on the catalyst surface during reaction was de-
termined with transient titration experiments and it was found to
c
be small.
1999 Academic Press
The reactivity of CH3 radicals with NO and NO2 is
well established in homogeneous gas-phase reactions (24–
27). Vannice and co-workers hypothesized that an increase
in methyl radical concentration would enhance the over-
all reaction between CH4 and NO and they chose several
methane oxidative coupling catalytic systems to investigate.
They found that MgO, Li/MgO, and several rare earth ox-
ide catalysts are active for either decomposition of NO or
its reduction by methane in both the absence and presence
of oxygen (19–23). However, the true reaction pathways
of NO reduction with CH4 in the presence of oxygen over
metal oxide catalytic surfaces remain an open question to
investigate.
Key Words: nitric oxide reduction by CH4; NO reduction by hy-
drocarbon; nitrogen oxides; methane; calcium oxide; nitric oxide
reduction kinetics.
1. INTRODUCTION
Many processes have been developed in recent years for
the removal of nitrogen oxides in flue gases from station-
ary and mobile sources. Selective catalytic reduction of NOx
with ammonia is among the most effective methods for con-
1 To whom correspondence should be addressed. Fax: (3572) 339060.
E-mail: efstath@ucy.ac.cy.
Campbell et al. (28) have indicated that the activities of
oxidative coupling of methane (OCM) metal oxides for the
323
0021-9517/99 $30.00
c
Copyright
1999 by Academic Press
All rights of reproduction in any form reserved.

324
FLIATOURA ET AL.
abstraction of hydrogen atom from the methane molecule direct and converted from NO2, wasblended in the chamber
to form methyl radical can best be related to the basicity of with ozone produced by an ozone generator from oxygen.
the oxides, the more basic oxides being more active. Based The intensity of the emitted light was in a linear relation-
on this finding, in the present work nitric oxide reduction by ship with the nitric oxide concentration. The photomulti-
methane was investigated over pure CaO, which is one of plier tube enhanced and transduced the emitted light into a
the most basic oxides being used successfully for the OCM current. The latter then went to an electrometer and wasdis-
reaction. According to a recent review paper (29) concern- played on the front panel of the analyzer. Separate readouts
ing the current state of research on lean-NOx catalysis, no of NO alone and NO plus NO2 combined were available.
reference is made to a published work for the HC-SCR of
The catalytic and kinetic experiments were performed
NO reaction over CaO surfaces. On the other hand, CO using a feed mixture that consisted of NO in the range 500–
and H2 have been studied as reducing agents of NO over 6000 ppm, O2 in the range 0–10 mol% , and CH4 in the range
the CaO surface (30–33).
0.5–4 mol% , with He as the balance gas. Under these ex-
In the present work, the catalytic behavior of CaO for the perimental conditions, reaction of NO with O2 to form NO2
nitric oxide conversion by methane in the range 550–850 C in the gas lines (held at 120 C) outside the reactor was not
was studied in the presence of excess oxygen in the range observed. Before any data were taken, the catalyst sample
1–10 mol% . The kinetic behavior of this reaction system was pretreated at 800 C in 20% O2/He mixture at a flow
was tested for NO feed concentrations in the range 500– rate of 50 ml/min for 1 h. The amount of catalyst sample
6000 ppm and for CH4 and O2 feed concentrations in the used was varied in the range 50–300 mg and the reaction
range 0.5–4 mol% . The stability of the catalyst investi- temperature in the range 500–850 C. The volume flow rate
gated and the effect of H2O and CO2 in the feed stream (F) of reactant gas mixture was kept constant at 50 ml/
on NO reduction activity are also presented and discussed. min (ambient), yielding a W/F ratio in the range 1–6
Temperature-programmed desorption (TPD) experiments 10 3 g min/ml. For the experiments to study the effects of
with NO have been conducted to study the interaction of H2O in the feed stream on NO conversion, a feed gas mix-
adsorbed NO with the catalyst surface (i.e., bond strength, ture of about 5 mol% H2O/He was prepared by flowing He
kinds of adsorbed NO), while the surface basicity of the through a deionized water saturator held at 33 C. The exit
CaO catalyst was studied by conducting TPD experiments stream was then mixed with NO/He, CH4/He, and O2/He
of CO2. The effects of gaseous oxygen and methane on gas streams at the appropriate mass flow rates to produce
the desorption properties of preadsorbed NO were in- the desired feed gas composition.
vestigated by conducting TPSR experiments with O2/He,
For kinetic measurements, the conversion of NO was al-
CH4/He, and CH4/O2/He gas mixtures. In addition, the ef- ways kept below 15% . Nitrogen reaction rates were then
fects of preadsorbed CO2 on the desorption kinetics of NO calculated from the product analyses using the differen-
were studied by experiments on TPD of NO.
tial reactor approximation, i.e., rate (mol g 1 min 1) = NT
yi/W, where NT is the total molar flow rate (mol/min), yi
is the molar fraction of component i (i.e., N2) expressed
in ppm 10 6, and W is the weight of the catalyst (g). For
the present catalytic system, methane reacts with NO and
O2 according to the following competitive reaction scheme:
2. METHODS
The catalyst used in the present study was prepared by
calcining CaO of high purity (99.95% , Alpha Products) for
24 h at 900 C. Due to this pretreatment of the catalyst, no
further sintering was expected during reaction. The BET
surface area after calcination was found to be less than
0.5 m2/g.
2NO + CH4 + O2 → N2 + CO2 + 2H2O,
CH4 + 2O2 → CO2 + 2H2O.
[1]
[2]
The flow system used for conducting steady-state kinetic
and catalyst evaluation measurements at atmospheric pres-
sure consisted ofa flowmeasuringand controlsystem (MKS
Instruments, Model 247C), mixing chambers, a quartz fixed-
bed microreactor, an on-line gas chromatograph (GC)
(Shimadzu 9A), and an on-line NOx stack gas emissions an-
alyzer (Teledyne Brown Engineering, Model 911/912). The
flow system, the microreactor, and the analysis conditions
The selectivity, , for CH4 reduction of NO to N2, i.e., the
ratio of the consumption rate of CH4 for the NO reduction
(reaction [1]) to the total consumption rate ofCH4 (reaction
[1] + reaction [2]) is calculated based on
0.5 yNf O XNO
(% ) =
100,
[3]
yCf H XCH
4
4
pertaining to the gas chromatograph have been described where yNf O and yCf H are the inlet molar fractions of NO and
4
in detail elsewhere (34).
CH4, respectively, and XNO and XCH are the NO and CH4
4
The NOx gas analyzer used the principle of chemilumi- conversions, respectively. The value of 0.5 is the stoichio-
nescence. The heart ofthe instrument wasa photomultiplier metric ratio of NO to CH4 (reaction [1]). The fraction of
tube assembly with a reaction chamber. Nitric oxide, both CH4 combusted by O2 (reaction [2]) is given by 1
.

SELECTIVE CATALYTIC REDUCTION OF NO BY CH4
325
FIG. 1. (a) Conversions of NO, CH4, and O2 reactants as a function of temperature. Feed composition: NO = 2000 ppm, CH4 = 2.5 mol% , O2 =
2.5 mol% . (b) Conversions of NO, CH4, and O2 reactants as a function of temperature. Feed composition: NO = 2000 ppm, CH4 = 2.5 mol% , O2 =
3
3
8 mol% . W/F = 6 10 g/min/cm .
For the determination of reaction orders, a power rate W/F ratio of 6 10 3 g min/ml. This value corresponds to
law of the following form was used:
approximately 10000 h 1 (GHSV).
Figure 1a shows that for 2.5 mol% O2 in the feed stream,
NO conversion continuously increases with reaction tem-
perature to reach the value of practically 100% at 850 C.
At temperatures above 750 C, oxygen and methane con-
versions remain practically constant, obviously due to the
remaining small concentrations of unreacted NO (reac-
tion [1]). In this temperature region the conversion of CH4
reaches the value of approximately 37% . Unlike zeolite-
based catalysts, no bendover in activity is observed over the
CaO catalyst under these experimental conditions (Fig. 1a).
The conversion-versus-temperature curves of Fig. 1b
which correspond to 8 mol% O2 in the feed stream are
significantly different from those of Fig. 1a. In this case,
NO conversion decreased significantly above 700 C. In the
aforementioned experiments of Fig. 1, N2, CO2, and H2O
were the only reaction products observed. The N and C
mass balances closed each within 2.5% . No higher hydro-
carbons due to possible oxidative coupling of methane were
detected in any circumstances, neither NO2 nor N2O.
Table 1 reports values of the selectivity parameter, , as a
function of reaction temperature for the experiments pre-
sented in Fig. 1. The highest selectivity value obtained is
about 11% , and the lowest, 1% . These results obviously
are not the desired ones since a large portion of CH4 fed
into the reactor is not used for the desired reduction of
NO. On the other hand, it has to be stressed that for prac-
tical lean-deNOx conditions the concentration of O2 must
be at least 5 mol% . Thus, to increase NO conversion for
rate = k PCXH PNYO POZ .
[4]
4
2
The turnover frequencies (TOFs) of NO reduction by CH4
were calculated after dividing the rate ( mol/g s) by the
irreversible NO uptake ( mol/g) at 50 C, the latter deter-
mined by TPD experiments.
TPD, temperature-programmed oxidation (TPO), and
temperature-programmed surface reaction (TPSR) with
CH4 and CH4/O2/He of preadsorbed NO and various other
transient experiments were conducted in a specially de-
signed flow system which has been described elsewhere
(35). Chemical analysis of gaseous streams during tran-
sients was done with an on-line mass spectrometer (VG
Quadrupoles, SXP Elite) equipped with fast-response inlet
capillary and data acquisition systems. The gaseous respo-
nses obtained by mass spectrometry were calibrated using
standard mixtures. The gases during transient experiments
were used at the flow rate of 40–50 ml/min (ambient). For
TPD and TPSR measurements, the mass numbers (m/z)
15, 30, and 44 were used for CH4, NO, and CO2 (or N2O),
respectively. Very small amounts of N2O but not of any
NO2 (m/z = 46) were detected during the TPSR experi-
ments with O2/He and CH4/O2/He mixtures.
3. RESULTS AND DISCUSSION
3.1. Catalyst Performance and Stability
NO reduction by CH4 in the presence of oxygen was stud- a constant value of GHSV, the concentration of the re-
ied over CaO in the temperature range 550–850 C using ducing agent (CH4) must increase; the dependence of the
2000 ppm NO, 2.5 mol% CH4, and two different O2 concen- rate of NO reduction on methane pressure is positive (see
trations (2.5 and 8.0 mol% ) in the feed stream at a constant Section 3.2). The fact that 100% conversion of NO to N2

326
FLIATOURA ET AL.
650 C and using a GHSV = 10,000 h 1, conversions as high
TABLE 1
as 50% were obtained up to 10% O2 in the feed stream
(Fig. 2).
Selectivities for CH4 Reduction of NO to N2 According
to the Results of Fig. 1 and Eq. [3]
The catalytic results shown in Figs. 1 and 2 are related to
oxygen effects on important elementary steps of the reac-
tion mechanism of the CH4/NO/O2 reaction over the CaO
catalytic surface. Acke et al. (33) recently reported results
from mechanistic studies of the NO + CO and NO + H2 re-
actions over the CaO surface. The authors proposed that
formation of N2 from NO is based on the following steps:
(i) an adsorbed N2O2 or N2O22 species forms on a partially
reduced CaO surface; (ii) the N–O bond in the aforemen-
tioned adsorbed dimers of NO breaks, leading to the for-
mation of an N2O intermediate product and to reoxidation
of the CaO surface; (iii) at temperatures higher than 500 C,
decomposition of N2O to N2 can be catalyzed by both the
reduced and the unreduced CaO surfaces. In their study,
Acke et al. (33) explained the role of reducing agents of
CO and H2 as being that of creating oxygen vacancies by
an oxygen abstraction step via the formation of CO2 and
H2O, respectively.
In the present case with CaO for the NO/CH4/O2 reac-
tion, there is not yet conclusive experimental evidence for
the mechanism of the reaction at hand. Recently, Vannice
et al. (22) reported a kinetic and mechanistic study related
to the NO/CH4/O2 reaction over La2O3-based catalysts.
The authors proposed 13 elementary steps, where the rate-
determining step was claimed to be the surface reaction of
adsorbed CH4 and NO2 to form methyl species. Adsorbed
NO2 is formed by the surface reaction of adsorbed NO and
O species, while adsorbed CO and OH species are also
found as intermediates in the proposed mechanism. The
adsorbed CO species is formed via the surface reaction of
Selectivity, (% )
NO = 2000 ppm, CH4 = 2.5% , NO = 2000 ppm, CH4 = 2.5% ,
T ( C)
O2 = 2.5 mol%
O2 = 8 mol%
550
600
650
700
750
800
850
4.9
5.0
5.7
6.8
8.0
11.3
5.2
3.1
2.2
1.6
10.0
10.8
1.0
0.09
has been achieved under the conditions shown in Fig. 1a
should be pointed out.
A comparison of the selectivity values reported by
Vannice and co-workers (20) over La2O3- and Sm2O3-
based catalysts should be made. In that work (20) the NO,
CH4, and O2 feed concentrations used were 18000 ppm,
0.45 mol% , and 1 mol% , respectively. At 550 C the authors
reported selectivity values in the range 35–65% over the
above-mentioned metal oxide catalysts. However, Vannice
and co-workers (20) used a yNO/yCH ratio of 4 in the feed
4
stream, to be compared with the value of 0.08 used in the
present work (50 times lower).
The effect of oxygen feed concentration on the catalyst
activity of nitric oxide reduction with methane was deter-
mined by performing experiments at 650 and 770 C using
2000 ppm NO and 2.5 mol% CH4, while the concentration
of O2 was varied between 0 and 10 mol% . In the absence of
gaseous oxygen, methane and NO react over CaO accord-
ing to the reaction
4NO + CH4 → 2N2 + CO2 + 2H2O.
[5]
Figure 2 shows that at 650 C the conversion of NO re-
mained almost constant, even in the presence of excess O2
in the feed stream of 10 mol% . On the other hand, at the
higher temperature of 770 C the addition of oxygen up to
concentrations of 10 mol% had a strong inhibiting effect.
It must be noted that the stability of NO conversion versus
O2 feed concentration at 650 C shown in Fig. 2 is a very
remarkable result. To our knowledge, it is the first time that
behavior at 650 C (Fig. 2) is reported in the literature over
metal oxide catalytic systems for the CH4/NO/O2 reaction.
Vannice and co-workers studied the effect of oxygen (up to
concentrations of 3 mol% O2) on NO reduction with CH4
1
over La2O3 catalyst (19). After using a GHSV = 12,400 h
(a value similar to that used in the present work), 1.4% NO
and 0.35% CH4 in the feed stream, at the reaction temper-
ature of 650 C there was a maximum in the NO conversion
( 10% ) at about 0.5% O2, while at 3% O2 the conversion of
NO reached 6% . However, in the present work with CaO at
FIG. 2. Dependence of nitric oxide conversion on oxygen feed con-
centration at T = 650 and 770 C. Feed composition: NO = 2000 ppm,
3
3
CH4 = 2.5 mol% ; W/F = 6 10 g/min/cm .

SELECTIVE CATALYTIC REDUCTION OF NO BY CH4
327
NCO and NO intermediate species. For the present CaO,
Investigations of the OCM reaction over La2O3/CaO
OCM studies (36) have indicated that CO is produced to- catalysts have shown that CO2, to a certain extent, pre-
gether with the desired C2-hydrocarbon products. As men- vents the adsorption of O2 on the surface (38). Moreover, it
tioned earlier, no detectable concentrations of these prod- was shown (39) that pretreatment of the catalyst with CO2,
ucts were observed in the present catalytic reaction system. which leads to the formation of surface carbonates, inhibits
Explanations for the absence of C2-hydrocarbons in the markedly the reversible activation of oxygen. These results
NO/CH4/O2 reaction over OCM metal oxide surfaces have indicate that activation of O2 involves the same basic sites
been given elsewhere (19). In addition, it should be pointed as activation of CO2. Knowing that the role of dioxygen
out that adsorbed CO speciesmight be formed directlyfrom is to promote the activation of CH4 via the formation of
the reaction of CH4/O2 over the present CaO surface.
methyl radicals, the observed inhibition of NO conversion
Based on the above-mentioned discussion, if the reoxi- by CO2 (Fig. 3a) could also be due to the reduced activation
dation step of the catalytic surface is an important step in of methane.
the sequence of elementary steps of the reaction at hand,
Figure 3b presents results of the effects of 5 mol% H2O in
then this could explain the large decrease in NO conversion the feed stream containing 2000 ppm NO, 2.5 mol% CH4,
with increasing O2 feed concentration observed at 770 C.
Increasing the O2 gas-phase concentration is expected to
decrease the surface coverage of oxygen vacant sites and,
therefore, the rate of N2O2 or N2O22 species formation,
the latter being likely important intermediate species for
the formation of N2. On the other hand, the results in Fig. 2
pertaining to T = 650 C cannot be explained based on the
above-mentioned reasoning which might be used to explain
the corresponding results obtained at T = 770 C. It seems
that at the lower temperature of 650 C other additional el-
ementary steps become more important than the surface
reoxidation step. It is suggested that at 650 C the rate of
reoxidation of the catalytic sites by gaseous oxygen is ap-
parently very slow, while oxidation of these sites by NO
or the intermediate NO2 species is much faster. The large
decrease in NO conversion with increasing O2 feed concen-
tration at 770 C must also be due to the CH4 combustion
reaction [2] mentioned before, the rate of the latter being
much larger at 770 C than 650 C.
One of the major products of reactions [1] and [2] is car-
bon dioxide which is present in significant concentrations
under integral reactor conditions. It is well known that cal-
cium oxide reacts readily with CO2 to form various kinds of
surface carbonate species and bulk CaCO3 as well (37). A
study of the interaction of CO2 with the catalyst surface and
its effect on the conversion of NO is therefore of interest.
The results reported in Fig. 3a reveal a decrease in NO con-
version when 2.5 mol% CO2 in the feed stream is used. The
extent of reduction in NO conversion, however, varies with
the reaction temperature. In the range 550–650 C, a larger
inhibition of the NO reduction by CO2 occurs as compared
with that obtained in the range 700–850 C. From transient
experiments to be presented (see Section 3.3), it is observed
that NO adsorption takes place over the CaO surface in the
temperature range 25–650 C. In addition, preadsorbed CO2
at 100 C causes a significant decrease in NO chemisorption
at 50 C. From these results it is suggested that the inhibiting
FIG. 3. (a) Effect of 2.5 mol% CO2 in the feed stream on NO con-
version as a function of reaction temperature. (b) Effect of 5 mol% H2O
in the feed stream on NO conversion as a function of reaction tempera-
effect of CO2 under reaction conditions is due to compet-
itive adsorption of CO2 and NO on the same surface sites
of CaO and/or to reduced adsorption of NO due to the
formation of a stable calcium carbonate layer.
ture. Feed composition: NO = 2000 ppm, CH4 = 2.5 mol% , O2 = 2.5 mol% ;
3
3
W/F = 6 10 g/min/cm .

328
FLIATOURA ET AL.
FIG. 4. Stability test over the CaO catalyst. Feed composition: NO =
FIG. 5. Arrhenius plot of the rate of nitrogen production over the
3
3
2000 ppm, CH4 = 2.5 mol% , O2 = 2.5 mol% ; W/F = 6 10 g/min/cm .
CaO catalyst. Feed composition: NO = 2000 ppm, CH4 = 2.5 mol% , O2 =
2.5 mol% .
and 2.5 mol% O2 (balance He). A small inhibiting effect
of H2O on NO reduction is observed in the range 550–
650 C, while a slight increase in NO conversion occurs in
the range 700–850 C. The latter result is of practical impor-
tance. Based on the results in Fig. 3, CaO appears to have
very good catalytic performance in the range 700–850 C un-
der practical lean-deNOx conditions when methane is used
as a reducing agent.
The variation ofNO conversion to N2 asa function oftime
on stream at 700 C is illustrated in Fig. 4. In this experiment,
the composition of the feed gas mixture was 2000 ppm NO,
2.5 mol% CH4, and 2.5 mol% O2. The initial conversion
of NO was about 60% and decreased slightly with time on
stream to approximately 55% after 50 h of continuous re-
action. Thus, the CaO catalyst is observed to exhibit good
stability with time-on-stream under the above-mentioned
reaction conditions. It should also be emphasized that se-
lectivity to N2 remained at the 100% level during the entire
course of the experiment.
to CH4, the partial pressures of NO and O2 were kept con-
stant at 2 and 25 mbar, respectively, whereas the CH4 partial
pressure was varied between 5 and 40 mbar (0.5–4 mol% ).
A period of 30 min on stream was allowed at each temper-
ature before any gas sample was taken for analysis.
Figures 6 and 7 present results of the rate of N2 formation
as a function of NO and CH4 feed concentrations, respec-
tively, at 650 and 770 C. Based on these results, the reac-
tion orders with respect to NO and CH4 (in the presence
of 2.5 mol% O2) were found to be 1.0 and 0.5, respectively,
3.2. Steady-State Kinetic Study
Figure 5 shows the Arrhenius plot of the rate of NO
reduction to N2 formation for NO feed concentration of
2000 ppm and for CH4 and O2 feed concentrations of
2.5 mol% . Results were obtained in the temperature range
550–700 C at NO conversions less than 15% . The apparent
activation energy determined was found to be 14.6 kcal/mol
(SD = 0.057).
In determining the reaction order with respect to nitric
oxide, the partial pressures of CH4 and O2 were kept con-
stant at 25 mbar (2.5 mol% ), whereas the partial pressure of
NO was varied between 1 and 6 mbar (1000 and 6000 ppm).
FIG. 6. Dependence of nitrogen production rate on nitric oxide feed
concentration at T = 650 and 770 C. Feed composition: CH4 = 2.5 mol% ,
Similarly, for determining the reaction order with respect O2 = 2.5 mol% .

SELECTIVE CATALYTIC REDUCTION OF NO BY CH4
329
that shown in Fig. 8 at 650 C has been reported by Vannice
and co-workers (19) for the La2O3 catalyst. However, for
the Li/MgO catalyst the presence of O2 had a strong inhibit-
ing effect (21).
The kinetic results presented in Fig. 8 could at first be con-
sidered as contradicting the catalytic results of Fig. 2. The
relatively small decrease in the rate of N2 formation with
increasing O2 feed concentration at 770 C is significantly
different from the large decrease in NO conversion with
increasing O2 feed concentration (in the range 0–4 mol% ).
The reason is that under the integral reactor conditions of
the experiments presented in Fig. 2 the rate of reaction
varies significantly along the catalyst bed due largely to the
variation in methane concentration (large methane conver-
sions are achieved). On the other hand, in the case of the
kinetic experiments (Fig. 8) the conversion of methane is
kept small. The increase in reaction rate with increasing O2
feed concentration in the range 0–1 mol% at 650 C (see
FIG. 7. Dependence of nitrogen production rate on methane feed Fig. 8) is compensated by the decrease in the rate along
concentration at T = 650 and 770 C. Feed composition: NO = 2000 ppm,
the catalyst bed due again to the large decrease in methane
concentration for the integral reactor experiments shown in
Fig. 2. Another reason for the additionaldecrease in the rate
of NO consumption along the catalyst bed with increasing
oxygen feed concentration is the effect of CO2 (see Fig. 3a)
which becomes larger in the case of the integral reactor ex-
periments (large methane conversion and CO2 formation
due to reaction [2]).
The present kinetic results have provided information
concerning the orders of reaction with respect to the three
reactant species, namely, CH4, O2, and NO, and also the
apparent activation energy of the reduction of NO with
CH4 in the presence of O2. These results are compared and
O2 = 2.5 mol% .
at both reaction temperatures. Under the feed conditions
of 2.5 mol% CH4, 2.5 mol% O2, and 6000 ppm of NO,
TOF values of 0.018 and 0.064 s 1 at 650 and 770 C, respec-
tively, were calculated. A comparison is made between the
TOF values reported by Vannice and co-workers (20) for
Sr/La2O3 and Li/MgO catalysts and the values obtained for
the present CaO under the reaction conditions of 1.8 mol%
NO, 0.46 mol% CH4, and 1 mol% O2 in the feed stream at
T = 500 C. After an extrapolation is made from the con-
ditions used for CaO (NO = 0.6 mol% , CH4 = 2.5 mol% ,
O2 = 2.5 mol% , T = 650 C) to the conditions reported by
3
1
Vannice and co-workers (20), a value of 1.65 10
s
is
obtained for the CaO. This value is comparable to the value
3
1
of 2.3 10
s
reported for Sr/La2O3 and much higher
than the value of 0.32 10 3 s 1 reported for the Li/MgO
catalyst (20).
The effect of O2 concentration on the rate of NO reduc-
tion by CH4 has also been determined at 650 and 770 C. In
this case, 2 mbar NO and 25 mbar CH4 were used in the feed
stream, while the O2 concentration was varied from 0 to
40 mbar (0–4 mol% ). The corresponding experimental re-
sults obtained are presented in Fig. 8. It is evident that at
650 C the rate of N2 formation increased with oxygen con-
centration up to 1 mol% and then practically remained con-
stant up to oxygen concentrations of 4 mol% . The reaction
order at O2 concentrations below 1 mol% was found to
be 0.25, while at higher concentrations it was found to be
slightly negative. At the higher temperature of 770 C, the
kinetic behavior was very similar to that observed at 650 C.
The reaction order at O2 concentrations below 1.5 mol%
was found to be practically zero, while at higher concentra-
FIG. 8. Dependence of nitrogen production rate on oxygen feed con-
centration at T = 650 and 770 C. Feed composition: NO = 2000 ppm,
tions it was found to be slightly negative. Behavior similar to CH4 = 2.5 mol% .

330
FLIATOURA ET AL.
discussed with respect to those obtained over similar cata- ent activation energy of the overall reaction is observed for
lytic surfaces, such as MgO, Li/MgO, La2O3, and Sr/La2O3 these catalytic materials.
(21, 22). At 650 C, the reaction order with respect to CH4
3.3. Transient Studies
over the La2O3 and Sr/La2O3 catalysts was found to be 0.62
and 0.87, respectively (22), to be compared with the value
of 0.5 determined in this work over the CaO surface. On the
other hand, there is a large difference in the reaction order
with respect to NO between the La2O3 and CaO surfaces.
In the case of La2O3 (22), a reaction order of 0.53 was ob-
tained, to be compared with the value of 1.0 obtained in the
present work over CaO. This result may indicate that the
nature of the chemisorption step of NO and of subsequent
elementary steps that lead to N2 formation is different over
the two catalytic surfaces. It could also be true that the
nature of the rate-determining step might be different over
the two catalytic surfaces. A significant difference between
the kinetic behavior of Li/MgO (21) and that of the present
CaO catalyst appears in the reaction order with respect to
O2. In the case ofLi/MgO catalysts, reaction ordersbetween
0.43 and 0.73 have been reported, to be compared with
the value of 0.25 obtained for the CaO surface. It is impor-
tant to note that these reaction orders have been obtained
after using similar O2 feed concentrations. On the other
hand, in the case of La2O3 and Sr/La2O3 catalysts (22) val-
ues similar to those for CaO were obtained, namely, 0.5 and
0.26, respectively.
As the feed oxygen concentration increases above
1 mol% , it was found in the present work that the reaction
order with respect to O2 becomes practically zero at 650 C
and slightly negative at 770 C, results largely different from
those observed at O2 concentrations below 1 mol% . These
results demonstrate that the partial pressure of gaseous
oxygen controls important elementary steps that contain
adsorbed oxygen species. It is noted that the nature of the
rate-determining step may shift depending on the value of
the partial pressure of oxygen. A mechanistic scheme based
on a sequence of elementary steps has appeared in the lit-
erature for the present reaction system (22). It consists of
13 elementary steps where adsorbed atomic oxygen species
participate in important steps from NO to form N2, CO2,
and H2O.
Various kinds of transient experiments were conducted
to study qualitatively and quantitatively the chemical inter-
action of the NO molecule with the surface of CaO. Figure 9
shows the TPD and TPO response curves of desorbed NO
under He and O2/He flow, respectively. Following adsorp-
tion of NO from a 2 mol% NO/He mixture at 50 C for
15 min, the feed was changed to He for 5 min to purge the
lines of the apparatus and to remove the gaseous NO from
the microreactor. Then, the temperature was increased to
850 C at the rate of 20 C/min under He flow in the case of
TPD, or under 1 mol% O2/He flow in the case of TPO, while
the nitric oxide (NO) and nitrous oxide (N2O) desorbed
from the catalyst surface were continuously monitored by
mass spectrometry. Two distinct well-resolved desorption
peaks were observed at 350 and 550 C in both cases, while
the peak maximum intensity of the second peak is found to
be lower than that of the first peak. In the case of TPD in He
flow, a small broad peak in the range 50–250 C is observed
(Fig. 9). In addition, some broadening of the second TPD
peak at the higher temperatures is observed during the TPO
experiment and a very small amount of N2O is formed in the
range 490–750 C (not shown in Fig. 9). The total amount
of equivalent nitric oxide corresponding to the TPD re-
sponse was found to be 32.3 mol/g, and the amount corre-
sponding to the TPO response was found to be practically
the same (32.1 mol/g). The amounts of NO desorbed
The apparent activation energy of the CH4/NO/O2 reac-
tion over the present CaO catalyst was found to be about
15 kcal/mol, a value significantly lower than that obtained
over La2O3 and Sr/La2O3 catalysts [26 kcal/mol (22)]. It
should be pointed out that in the present case of CaO, the
O2 feed concentration used in the experiments for deter-
mining the apparent activation energy was 2.5 mol% , while
in the case of La2O3-based catalysts it was 1 mol% . Ac-
cording to the previous discussion concerning the effects
of O2 partial pressure on the oxygen reaction order, and
taking into account the large difference in the reaction or-
der of NO exhibited by the CaO and La2O3-based catalysts,
FIG. 9. Temperature-programmed desorption (TPD) and oxidation
(TPO) of nitric oxide responses obtained according to the following gas
delivery sequences: (a) 2 mol% NO/He (50 C, 15 min) → He (50 C,
10 min) → TPD ( = 20 C/min); (b) 2 mol% NO/He (50 C, 15 min) → He
(50 C, 10 min) → TPO (1 mol% O2/He, = 20 C/min). Amount of cata-
it seems reasonable that a significant change in the appar- lyst used: 0.3 g; Q = 40 cm3/min.

SELECTIVE CATALYTIC REDUCTION OF NO BY CH4
331
TABLE 2
Amount of Nitric Oxide Desorbed as Determined by Various
Temperature-Programmed Transient Experiments
Amount desorbed
NO desorbed, first peak
(% )
Experiment
TPD
(
mol/g)
32.3
32.1
32.1
31.1
60
60
72
65
TPO
TPSR with CH4/He
TPSR with CH4/O2
during the various temperature-programmed experiments
are given in Table 2.
Figure 10 shows comparative results of TPD and TPSR
experiments with CH4. The latter experiment was perfor-
med by flowing a 1 mol% CH4/He mixture over the cata-
lyst on which NO had been adsorbed at 50 C. In the case
of TPSR with CH4, two well-resolved NO peaks are ob-
served: the first one was centered at 350 C and the second
one at 500 C. The amount of nitric oxide corresponding to
the first peak is approximately twice the amount of the sec-
ond peak. The total amount of desorbed NO is found to
be 32.1 mol/g. Very small amounts of CO2 were measured
in the range 550–850 C due to the reaction of CH4 with
the surface lattice oxygen of CaO. Although the amount
of desorbed nitric oxide during the experiments presented
in Fig. 10 is found to be the same (Table 2), qualitatively
the results do exhibit certain differences. It is obvious that
in the presence of CH4 the desorption of NO observed at
FIG. 11. Temperature-programmed desorption (TPD) and tempera-
ture-programmed surface reaction (TPSR) with CH4/O2/He of nitric ox-
ide responses obtained according to the following gas delivery sequences:
(a) 2 mol% NO/He (50 C, 15 min) → He (50 C, 10 min) → TPD (
=
20 C/min); (b) 2 mol% NO/He (50 C, 15 min) → He (50 C, 10 min) →
==
TPSR with CH4/O2/He (CH4 O2 = 1 mol% , = 20 C/min). Amount of
catalyst used: 0.3 g; Q = 40 cm3/min.
higher temperatures in He flow is now shifted to lower tem-
peratures, while the largest amount desorbed is observed
in the lower temperature region of 250–400 C (Fig. 10).
Figure 11 shows comparative responses of TPD and
==
TPSR experiments with CH4/O2/He (CH4 O2 = 1 mol% )
following NO chemisorption at 50 C for 15 min. In the
case of TPSR with CH4/O2, two kinds of NO species are
probed, based on the two resolved peaks observed at 350
and 500 C, as in the case of TPD and TPSR with methane.
However, in the present case the ratio of the amount of NO
desorbed corresponding to the two peaks is different. Very
small amounts of N2O but larger amounts of CO2 were ob-
served in the range 550–850 C (not shown in Fig. 11). The
CO2 formed is due to the combustion of CH4. The quan-
titative results of desorbed NO for this TPSR experiment
are presented in Table 2.
The results shown in Figs. 9–11 provide clear informa-
tion concerning the influence of O2 and CH4 molecules on
the chemisorbed state of NO over the CaO surface. The
presence of gaseous O2 leaves unaffected the kinetics of
desorption of the first kind of chemisorbed NO, while it af-
fects only slightly the kinetics of the high-temperature des-
orption of the second kind of chemisorbed NO. However,
in the presence of gaseous CH4, the kinetics of desorption
of chemisorbed NO drastically changes; there exists a sig-
nificant shift in the peak maximum temperature (TM) of
the high-temperature desorption peak and a large increase
in the maximum desorption rate in the case of the low-
temperature NO desorption peak. In the first case, there is
a weakening of the chemisorption bond between the NO
FIG. 10. Temperature-programmed desorption (TPD) and tempera-
ture-programmed surface reaction (TPSR) with CH4 of nitric oxide re-
sponses obtained according to the following gas delivery sequences:
(a) 2 mol% NO/He (50 C, 15 min) → He (50 C, 10 min) → TPD (
=
20 C/min); (b) 2 mol% NO/He (50 C, 15 min) → He (50 C, 10 min) →
TPSR with CH4 (1 mol% CH4/He, = 20 C/min). Amount of catalyst
used: 0.3 g; Q = 40 cm3/min.

332
FLIATOURA ET AL.
molecule and the CaO surface induced by the presence of kinetic study of the NO/CH4/O2 reaction over La2O3 and
gaseous CH4. Sr/La2O3 catalysts (22), it was found that a rate-determining
Detailed description of the modes of NO adsorption over step involving the surface reaction of adsorbed CH4 and
metal oxide surfaces is still a controversial issue. This sub- NO2 to generate methyl species was one of the restrictions
ject concerns the direction of electron transfer between the of the kinetic model that best described the experimental
solid surface and the orbitals of the NO molecule (40–42), results. The characterization of the surface basicity of CaO
for example, whether an NO species is formed via an elec- in terms of strength distribution and amount therefore be-
tron transfer from the metal oxide to the antibonding or- comes important.
bital of NO, thus weakening the N–O bond, or the opposite
The basicity of the surface of the present CaO catalyst
occurs by forming NO+ species. In addition, it is not yet was studied by TPD of preadsorbed CO2 at 100 C, at a
clear whether NO adsorption occurs with the N atom near- heating rate of 20 C/min and using helium as the carrier
est the surface or with the O atom coordinated to anionic gas. Figure 12a shows the CO2 TPD spectrum obtained. A
vacancies (41).
According to the recent work of Acke et al. (33), forma-
tion of NO dimers (N2O2 or N2O22 ) over a CaO surface
could occur, and these species have been considered as im-
portant intermediate species for the reduction of NO by
CO or H2 (33). Assuming that formation of NO dimers
does occur over the present CaO surface, following NO
chemisorption at low temperatures, then one possible ex-
planation for the weakening of the NO adsorption bond
is the existence of repulsions between adjacent adsorbed
H3C –H + and N2O2 or N2O22 species. The former species
are formed on adsorption of CH4 onto Ca2+–O2 sites (28,
43), while the latter species are formed on adsorption of
NO onto oxygen vacancy sites (33, 41). Another possible
explanation for the weakening of the NO adsorption bond
in the presence of methane might be the existence of an
induced electron transfer within the chemisorbed state of
NO by adsorbed CH4 resulting in a reduced overlapping
of the bonding orbitals of the molecular NO and the CaO
surface. However, such a mechanism needs to be proved by
providing experimental evidence.
The appearance of two well-resolved NO desorption
peaks and a third small peak in the range 50–250 C strongly
suggests the heterogeneity of the present CaO surface.
Based on the discussion above, it is likely that molecu-
larly adsorbed NO and N2O2 or N2O22 species (nitric oxide
dimer species) give rise to the appearance of the desorption
peaks of NO shown in Figs. 9–11. Two nitric oxide desorp-
tion peakshave also been reported in the case ofLa2O3 (20).
The importance of surface basicity in the abstraction of
hydrogen atoms from methane molecules to form methyl
radicals has been pointed out by many investigators (28,
n+
43). It has been proposed that an acid–base pair (M OL2 C
)
LC
on the metal oxide surface is involved in the abstraction of
the H atom from the methane molecule, according to the
following scheme: When methane interacts with an acid–
base pair having enough strength, it undergoes heterolytic
FIG. 12. (a) Temperature-programmed desorption (TPD) of CO2 res-
C–H bond rupture, resulting in a CH3 ion attached on the ponse obtained according to the gas delivery sequence 5 mol% CO2/He
acid site, while the H+ ion interacts with the basic site. In
the presence of O2, an electron transfer from the carbanion
CH3 to O2 resulting in an O2 species is expected to take
place, while the methyl radical can be released into the gas
(100 C, 15 min) → He (100 C, 15 min) → TPD ( = 20 C/min). (b) Temp-
erature-programmed desorption of NO and CO2 responses obtained acc-
ording to the gas delivery sequence 5 mol% CO2/He (100 C, 15 min) →
He (100 C, 15 min) → cool down in He flow to 50 C → 2 mol% NO/He
(50 C, 15 min) → He (50 C, 10 min) → TPD ( = 20 C/min). Amount of
phase or be oxidized on the catalyst surface. In a recent catalyst used: 0.3 g; Q = 40 cm3/min.

SELECTIVE CATALYTIC REDUCTION OF NO BY CH4
333
single peak of desorbed CO2 in the temperature range 400
to 800 C with peak maximum at TM = 680 C and a small
shoulder to the rising part of it appears. By integrating
the TPD response curve, the amount of adsorbed CO2 is
found to be 155 mol/gcat. A comparison of this result with
those obtained from similar basicity measurements over
rare-earth metal oxides (43) leads to the conclusion that
the present CaO sample has much higher surface basicity
per gram of solid than the group of catalysts examined by
Vannice et al. (22, 23) for NO reduction with CH4. Catalytic
activity tests over La2O3 performed under the same exper-
imental reaction conditions as for CaO in this work (44) re-
vealed that the CaO catalyst is more active than the La2O3
catalyst, especially at high temperatures. This may proba-
bly be related to the surface basicity differences exhibited
by the two catalytic surfaces. In fact, La2O3 was found to
have a significantly smaller amount of basic sites (17 mol
CO2/gcat) and also weaker basic sites (desorption of CO2 in
the range 250–650 C) than the CaO surface (44).
FIG. 13. Temperature-programmed hydrogenation (TPH), tempera-
ture-programmed oxidation (TPO), and temperature-programmed des-
orption (TPD) of adsorbed carbon-containing species formed during 1 h
of CH4/NO/O2/He reaction at 700 C over the CaO catalyst. Gas deliv-
ery sequence: CH4/NO/O2/He (700 C, 1 h) → He (700 C, 5 min) → cool
down in He flow to 500 C → 1% H2/He (TPH) or 1% O2/He (TPO) or He
(TPD), = 30 C/min. Amount of catalyst used: 0.3 g; Q = 40 cm3/min.
It is of interest to study the effect of preadsorbed CO2 on
NO chemisorption on the CaO surface since carbon dioxide
is considered a major product of the present reaction and,
also, it remains bound on the CaO surface even at 750 C
(Fig. 12a). Figure 12b presents the desorption spectra of NO
and CO2 in a He TPD experiment accordingto the following
experimental sequence. Carbon dioxide was adsorbed at
100 C for 15 min followed by a He purge to 100 C and increased at the rate of 30 C/min to 900 C to carry out a
subsequent cooling of the reactor to 50 C. The feed gas TPH experiment. A CH4 peak is shown in Fig. 13 that is cen-
mixture was then changed to NO/He for 15 min followed tered at about 520 C tailing out to 850 C. Integration of the
by a switch to He for 10 min. The temperature of the catalyst CH4 response resulted in an equivalent amount of carbon-
sample was then increased to 900 C at the rate of 20 C/min containing species of 15 mol/gcat. Figure 13 also shows
in He flow.
the CO2 response obtained during a TPO experiment with
Figure 12b shows clearly that a large decrease in the carbon-containing species formed on exposure of the cata-
amount of desorbed NO takes place when compared with lyst to the reaction mixture at 700 C for 1 h, as in the case
the result of Fig. 9 (6 mol/gcat vs 32.3 mol/gcat), while at of the TPH experiment. A rather symmetric peak is ob-
the same time there occurs a change in the shape of the served (TM = 750 C) suggesting one kind of carbon species.
desorption spectrum of NO. On the other hand, the fea- In this TPO experiment, it was found that the amount of
tures of the desorption spectrum of CO2 remain practically carbon removed as CO2 is 9.7 mol/gcat. Note that signif-
the same as those observed in the CO2 TPD spectrum of icant quantities of CO2 appeared at temperatures higher
Fig. 12a. The results in Figs. 9, 12a, and 12b strongly suggest than 650 C during TPO, while a very small CH4 response
that NO and CO2 compete for the same adsorption sites, signal was observed above 650 C during TPH. Figure 13
while the presence of adsorbed CO2 influences the binding also shows results of a TPD experiment in He flow of
energy of adsorbed NO on the CaO surface. These results carbon-containing species formed during the reaction un-
can provide some explanation of the inhibiting role of CO2 der conditions similar to those pertaining to the TPO exper-
in NO conversion under reaction conditions (Fig. 3a).
iment. The total amount of equivalent carbon correspond-
The present work also reports results of the transient ing to the CO2 TPD response was found to be 8.2 mol/gcat
,
hydrogenation and oxidation of carbon-containing species which is similar to the amount of CO2 measured from the
accumulated on the surface of CaO following reaction at TPO response.
700 C with CH4/NO/O2/He mixture. Figure 13 presents re-
The position and shape of the CO2 TPD and TPO re-
sults of a temperature-programmed hydrogenation (TPH) sponses shown in Figs. 12a and 13 probe for the presence
experiment with carbon species formed during reaction at of a carbonate-like species (i.e., ionic carbonate, unidentate
700 C for 1 h. Following reaction, the feed was changed to or bidentate carbonate, bicarbonate) that might be formed
pure He at 700 C for 5 min followed by cooling of the re- on the CaO surface after reaction with the NO/CH4/O2/He
actor under He flow to 500 C. The feed was then changed mixture at 700 C for 1 h. If it is assumed that the only
to 1 mol% H2/He and the temperature of the catalyst was carbon-containing species formed on the catalyst surface

334
FLIATOURA ET AL.
during reaction is that having a carbonate-like nature, then fluence the kinetics of NO desorption during a TPD exper-
explanations must be provided for the different quantities iment.
of equivalent carbon measured by the TPH and TPD exper-
imentsofFig. 13. It isspeculated whether the larger quantity
ACKNOWLEDGMENTS
of equivalent carbon measured by TPH could be due to the
presence of additional adsorbed carbon-containing species,
such as CN– and/or HCN– . The latter species may not
be reactive toward either oxygen or surface reactions under
He flow to give CO2. The above-mentioned intermediate
species have been proposed for the present reaction over
La2O3-based catalysts (22). Despite the true chemical com-
position of the carbon-containing species formed under the
present reaction conditions, it should be noted that a very
small amount is accumulated on the CaO surface during 1 h
of reaction at 700 C. This is believed to be the main rea-
son for the very good stability of CaO under the reaction
conditions investigated (Fig. 4).
Financial support by the Commission of the European Community
(Contract EV5V-CT92-0234) and the Research Committee of the Uni-
versity of Cyprus is gratefully acknowledged.
REFERENCES
1. Bosch, H., and Janssen, F. J. J. G., Catal. Today 2, 369 (1988).
2. Beechman, J. W., and Hegedus, L. L., Ind. Eng. Chem. Res. 30, 969
(1991).
3. Amiridis, M. D., Zhang, T., and Farrauto, R. J., Appl. Catal. B 10, 203
(1996).
4. Hardee, J. R., and Hightower, J. W., J. Catal. 86, 137 (1984).
5. Iwamoto, M., Yokoo, S., Sakai, K., and Kagawa, S., J. Chem. Soc.
Faraday Trans. 1 77, 1629 (1981).
6. Iwamoto, M., Yahiro, H., Shundo, S., Yu, Y., and Mizuno, N., Appl.
Catal. 69, L15 (1991).
4. CONCLUSIONS
7. Iwamoto, M., and Hamada, H., Catal. Today 10, 57 (1991).
8. Sato, S., Hirabayashi, H., Yahiro, H., Mizuno, N., and Iwamoto, M.,
Catal. Lett. 12, 193 (1992).
The following conclusions can be derived from the results
of the present work:
9. Li, Y., and Armor, J. N., Appl. Catal. B 1, L31 (1992).
CaO was found to be active for nitric oxide reduction 10. Li, Y., and Armor, J. N., Appl. Catal. B 2, 239 (1993).
11. Witzel, F., Sill, G. A., and Hall, W. K., J. Catal. 149, 229 (1994).
12. Hamada, H., Kintaichi, Y., Sasaki, M., Ito, T., and Tabata, M., Appl.
Catal. 64, L1 (1990).
13. Hamada, H., Kintaichi, Y., Sasaki, M., Ito, T., and Tabata, M., Appl.
by methane in the presence of oxygen and 100% selective
toward N2 formation in the temperature range 550–850 C.
TOF values for the NO/CH4/O2 reaction over the pre-
sent CaO catalyst compare favorably with the values re-
Catal. 70, L15 (1991).
ported in the literature over the best rare-earth metal oxide 14. Sato, S. S., Hirabay, H., Yahiro, H., Mizuno, N., and Iwamoto, M., Catal.
Lett. 12, 193 (1992).
15. Inaba, M., Kintaichi, Y., and Hamada, H., Catal. Lett. 36, 223 (1996).
16. Burch, R., Fornasiero, P., and Watling, T. C., J. Catal., in press.
17. Harrison, B., Wyatt, M., and Gough, K. G., “Catalysis,” Vol. 5, pp. 127–
catalysts.
The catalytic performance of CaO and the observed
reaction rates of N2 formation are considered much better
than those reported in the literature for MgO and Li/MgO
catalysts.
The addition of oxygen at concentrations as high as
10 mol% has no effect on CaO catalyst activity at 650 C.
However, at 770 C a significant decrease in NO conversion
is evident.
171. Roy. Soc. of Chemistry, London, 1982.
18. Maitra, A. M., Campbell, I., and Tyler, R. J., Appl. Catal. 85, 27 (1992).
19. Zhang, X., Walters, A. B., and Vannice, M. A., Appl. Catal. B 4, 237
(1994).
20. Zhang, X., Walters, A. B., and Vannice, M. A., J. Catal. 155, 290 (1995).
21. Zhang, X., Walters, A. B., and Vannice, M. A., J. Catal. 146, 568 (1994).
22. Vannice, M. A., Walters, A. B., and Zhang, X., J. Catal. 159, 119 (1996).
Carbon dioxide (2.5 mol% in the feed stream) is found 23. (a) Zhang, X., Walters, A. B., and Vannice, M. A., Appl. Catal. B
7, 321 (1996). (b) Shi, C., Walters, A. B., and Vannice, M. A., Appl.
Catal. B 14, 175 (1997).
24. Miller, J. A., and Bowman, C. T., Prog. Energy Combust. Sci. 15, 287
to inhibit the reduction of nitric oxide, especially in the
range 500–650 C. However, the inhibiting effect of 5 mol%
H2O in the feed stream is small in the range 500–650 C,
(1989).
while the influence of water on NO conversion in the range
700–850 C appears to be slightly positive.
25. Yamada, F., Slafle, I. R., and Gutman, D., Chem. Phys. Lett. 83, 409
(1981).
26. Davies, J. W., Grant, N. J. B., and Pilling, M. J., J. Chem. Soc. Faraday
Trans. 87, 2317 (1991).
27. Thorne, L. R., Branch, M. C., Chandler, D. W., Kee, R. J., and Miller,
J. A., in “Proceedings, 21st Symposium (International) on Comb.,”
p. 965. The Comb. Inst., 1986.
28. Campbell, K. D., Zhang, H., and Lunsford, J. H., J. Phys. Chem. 92,
750 (1988).
Temperature-programmed desorption studies of nitric
oxide revealed two well-resolved desorption peaks, a result
suggesting two kinds of adsorbed NO species on the CaO
surface. The presence of 1 mol% O2 has no effect on the
desorption profile of NO, while the presence of 1 mol%
CH4 causes a shift of the high-temperature NO desorption
to lower temperatures (decrease in the binding energy of
NO with the CaO surface).
29. Fritz, A., and Pitchon, V., Appl. Catal. B 13, 1 (1997).
30. Tsujimura, M., Furusawa, T., and Kunii, D., J. Chem. Eng. Japan 16,
132 (1983).
Preadsorbed carbon dioxide is found to largely de-
crease the chemisorption of NO at 50 C and also to in-
31. Bertrand, R. R., Hoke, R. C., Shaw, H., and Skopp, A., Am. Chem.
Soc. Div. Fuel Chem. 18, 25 (1973).

SELECTIVE CATALYTIC REDUCTION OF NO BY CH4
335
32. Dam-Johansen, K., Hansen, P. F. B., and Rasmussen, S., Appl. Catal.
B 5, 283 (1995).
33. Acke, F., Panas, I., and Stromberg, D., J. Phys. Chem. B 102, 5127
(1998).
34. Efstathiou, A. M., and Fliatoura, K., Appl. Catal. B 6, 35 (1995).
35. Efstathiou, A. M., Papageorgiou, D., and Verykios, X. E., J. Catal. 141,
612 (1993).
36. Efstathiou, A. M., and Verykios, X. E., Appl. Catal. A 151, 109
(1997).
39. Lacombe, S., Zanthoff, H., and Mirodatos, C., J. Catal. 155, 106 (1995).
40. Iwamoto, M., in “Studies in Surface Science and Catalysis: Fu-
ture Opportunities in Catalytic Separation Technology” (M. Misono,
Y. Moro-oka, and S. Kuimura, Eds.), p. 121. Elsevier, Amsterdam,
1990.
41. Hightower, J. W., and van Leirsburg, D. A., in “The Catalytic Chem-
istry of Nitric Oxides” (R. L. Klimisch and J. G. Larson, Eds.). Plenum,
New York, 1975.
42. Gandhi, H. S., and Shelef, M., J. Catal. 28, 1 (1973).
43. Choudhary, V. R., and Rane, V. H., J. Catal. 130, 411 (1991).
44. Fliatoura, K. D., and Verykios, X. E., in preparation.
37. Philipp, R., and Fujimoto, K., J. Phys. Chem. 96, 9035 (1992).
38. Yang, T., Feng, L., and Shen, S., J. Catal. 145, 384 (1994).