Plasma catalysis with perovskite-type catalysts for the removal of NO and CH4 from combustion exhausts

Article abstract of DOI:10.1016/j.jcat.2007.02.006

The removal of NO and CH₄ from quaternary gas mixtures simulating the conditions existing in real combustion exhausts has been studied with a hybrid system integrating plasma activation and a La_1-xSr_xCoO_3-d perovskite-type catalyst. The plasma reaction produces the conversion of NO into N₂ plus O₂ and the oxidation of CH₄ into CO + H₂O. Incorporation of the catalyst favors the oxidation of CH₄ into CO₂ at 190 °C. At this temperature, no oxidation of CO or CH₄ is found in a conventional catalytic reactor. A similar plasma + catalyst experiment with SiO₂ found much lower CO₂ production, indicating that the perovskite is actively involved in the oxidation of CO. The efficiency of NO removal decreased with the amount of perovskite, although this efficiency could be restored by adding carbon to the reactor. Experiments using optical emission spectroscopy (OES) and in situ X-ray photoemission spectroscopy (XPS) were carried out to gain insight into the synergetic effects found with the catalyst. OES intermediate species, including NH^*, CN^*, CO^*, and CH^*, were found in the plasma. XPS experiments of samples exposed to plasmas showed the formation of {single bond}NO_x and {single bond}CN species, indicating the active involvement of the catalyst in the reaction.

Full text of DOI:10.1016/j.jcat.2007.02.006

Journal of Catalysis 247 (2007) 288–297
www.elsevier.com/locate/jcat
Plasma catalysis with perovskite-type catalysts for the removal of NO and
CH from combustion exhausts
4
a,b
a,c
a,b
a,b
a,b,∗
J.L. Hueso , J. Cotrino , A. Caballero , J.P. Espinós , A.R. González-Elipe
a
Instituto de Ciencia de Materiales de Sevilla (CSIC – University of Seville), Avda. Américo Vespucio, 49, 41092 Seville, Spain
b
Departamento de Química Inorgánica, Universidad de Sevilla, Spain
c
Departamento de Física Atómica, Molecular y Nuclear, Universidad de Sevilla, Spain
Received 6 December 2006; revised 2 February 2007; accepted 2 February 2007
Available online 21 March 2007
Abstract
The removal of NO and CH from quaternary gas mixtures simulating the conditions existing in real combustion exhausts has been studied with
4
a hybrid system integrating plasma activation and a La
SrxCoO
perovskite-type catalyst. The plasma reaction produces the conversion of
1−x
3−d
◦
NO into N plus O and the oxidation of CH into CO + H O. Incorporation of the catalyst favors the oxidation of CH into CO at 190 C. At
2
2
4
2
4
2
this temperature, no oxidation of CO or CH is found in a conventional catalytic reactor. A similar plasma + catalyst experiment with SiO found
4
2
much lower CO production, indicating that the perovskite is actively involved in the oxidation of CO. The efficiency of NO removal decreased
2
with the amount of perovskite, although this efficiency could be restored by adding carbon to the reactor. Experiments using optical emission
spectroscopy (OES) and in situ X-ray photoemission spectroscopy (XPS) were carried out to gain insight into the synergetic effects found with
∗
∗
∗
∗
the catalyst. OES intermediate species, including NH , CN , CO , and CH , were found in the plasma. XPS experiments of samples exposed to
plasmas showed the formation of –NOx and –CN species, indicating the active involvement of the catalyst in the reaction.
©
2007 Elsevier Inc. All rights reserved.
Keywords: Plasma catalysis; Plasma chemistry; Perovskite catalyst; In situ XPS; Optical emission spectroscopy; CO oxidation; CH oxidation; NO removal; Soot
4
particles
1
. Introduction
plex gas mixtures, alone or in the presence of catalysts, is not
a well-developed subject, and much work is needed to improve
the performance of these processes for practical applications
[15]. Plasma catalytic processes are used mainly in various
decontamination processes in which the concentration of pol-
lutants is in the ppm range, including the abatement of volatile
organic compounds [16–19], oxidation of light hydrocarbons
Removal of CO, CHx, NO, SOx, and other compounds from
exhaust combustion sources is a major challenge for environ-
mental protection [1–3]. The most typical procedures used to
remove these noxious gases are based on catalysts that can
induce different selective oxidation/reduction processes [1–3].
Another possibility that has begun to be explored recently is
the combination of catalysts and plasmas to induce specific
reactions under mild thermal conditions. Cold plasmas have
typically been studied by researchers interested in the charac-
terization of the plasma parameters (e.g., electron temperature,
concentration of different active species) [4–9] and by materi-
als scientists for treating or preparing new materials [10–13] or
even catalysts [14]. In contrast, the plasma chemistry of com-
[
[
20,21], oxidation of CO [16,20], combustion of soot particles
22,23], and removal of NOx or N2O from simulated exhaust
gases [15,24–33]. Nonetheless, relatively few works have fo-
cused on the simultaneous removal of these noxious gases in
complex mixtures [15,25,34].
Most designs that use a mixed plasma + catalyst approach
are merely the online connection of two independent processes;
first, the plasma stage produces a certain reaction, and then the
catalyst stage transforms these primary products into secondary
final products [17,31,35]. Less frequently, plasma and catalysts
are combined together in such a way that the catalyst is em-
*
Corresponding author.
E-mail address: arge@icmse.csic.es (A.R. González-Elipe).
0021-9517/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2007.02.006

J.L. Hueso et al. / Journal of Catalysis 247 (2007) 288–297
289
bedded in a plasma phase rich in radicals and other activated
species [16,17]. In this latter case, despite the observation of
synergetic effects, it is not clear yet whether the role of the cat-
alyst is just to favor the combination on its surface of the active
species of the plasma, or if it participates actively in the reac-
tion. In general, in these plasma hybrid systems, the catalyst
activity is typically correlated with its surface area and poros-
ity, regardless of their chemical formulation [15,16,20].
The present work deals with the plasma catalytic removal of
NO and the oxidation of CH4 from complex mixtures consist-
ing of NO, CH4, and O2 as minority components and Ar or N2
as carrier gases. Previous plasma investigations in our labora-
tory with a microwave (MW) surfatron device in the absence
of catalysts have revealed that for such mixtures, the NO is ef-
ficiently converted into N2 and O2 [36]. Simultaneously, CH4
is oxidized to CO, a noxious gas that should be removed for
practical decontamination applications.
(a)
In the present investigation, we used a mixed plasma + cat-
alyst system in an attempt to achieve the oxidation of CO into
CO2. Our results support the utility of this experimental ap-
proach and demonstrate the synergetic effects of CO plasma +
catalyst oxidation at low temperatures. We used two types of
catalysts, a commercial high-specific surface area SiO2 catalyst
and a perovskite-type catalyst prepared in our laboratory. The
latter choice was dictated by the fact that perovskite catalysts
are well-known catalysts for CO and CH4 oxidation reactions
[
37,38]. We used the SiO2 for comparative purposes to evalu-
ate the hypothesis proposed in several papers in the sense that
only the porosity of the catalyst is controlling the efficiency
of the plasma + catalyst systems [16,20]. The main focus of
the present investigation with complex gas mixtures contain-
ing NO, CH4, and O2 is to understand the chemical and plasma
processes involved in the removal of NO and the conversion of
CO (a primary product of the plasma reaction) into CO2 when
the plasma is used in conjunction with a perovskite catalyst. In
addition, the beneficial effect of the presence of carbon together
with the catalyst for the NO removal was also addressed in the
context of this investigation. This carbon simulates the soot par-
ticles eventually present in most combustion outlets [15,25,37].
(b)
Fig. 1. (a) XRD pattern of the La Sr CoO
perovskite and scheme of
0
.5 0.5
3−d
the rhombohedral distorted geometry. (b) SEM micrograph of the perovskite
catalyst.
water) with the dissolved nitrates and produced an initially
amorphous perovskite powder. The material was collected by
a porous frit of quartz located in the outlet of the heating sys-
◦
tem. Subsequently, the powders were annealed at 600 C for
4
h, thus obtaining a crystalline perovskite with rhombohe-
2
. Experimental
dral symmetry [40], as evidenced by the X-ray diffractograms
shown in Fig. 1a. The ultrasonic nebulization method of liquid
droplets favors the formation of pseudospherical particles with
a wide range of sizes, as inferred from the scanning electron
microscopy (SEM) image shown in Fig. 1b. The mean diameter
of particles varied from 200–500 nm to 1 µm, although smaller
nanocrystalline domains of 50–60 nm were calculated by ap-
plying Scherrer’s equation to the main diffraction peaks of this
sample.
2
.1. Catalyst preparation and characterization
The two catalysts used for the present investigation are a
commercial SiO2 powder from Aldrich with a specific surface
2
−1
area of 270 m g and a La0.5Sr0.5CoO3−d perovskite pre-
pared by spray pyrolysis [39] in our laboratory, with a specific
2
−1
surface area of 13 m g . In addition, in some experiments,
the perovskite was thoroughly ground with carbon nanopartic-
ulates at a ratio of 90:10 by weight. The carbon, provided by
2
−1
2.2. Catalytic activity in the absence of plasma
Cabot, had a specific surface area of 190 m g
.
Perovskite preparation involved the uniform nebulization of
nitrate solutions containing La(NO3)3·6H2O (99.99%, Aldrich),
Co(NO3)2·6H2O (>98%, Fluka), and Sr(NO3)2 (>99%, Fluka)
prepared as a 0.1 M liquid solution of precursors. Two online
Catalytic oxidation experiments of CO and CH4 were car-
ried out at atmospheric pressure in a fixed-bed stainless steel
U-shaped reactor; 100 mg of perovskite particles were sus-
tained between quartz wool in approximately the middle of the
◦
furnaces at 250 and 600 C evaporated the solvent (distilled

290
J.L. Hueso et al. / Journal of Catalysis 247 (2007) 288–297
reactor. The temperature was controlled by a chromel–alumel
thermocouple placed at the catalyst position. The reactor was
placed vertically in the center of an electric furnace electroni-
cally controlled and provided with another thermocouple.
Before catalytic testing, the catalysts were preoxidized in a
2.4. Plasma catalytic hybrid system
The plasma experiments were carried out in a plasma re-
actor consisting of a quartz tube (3.5 mm i.d.) similar to that
described previously [36,44,45] in studies of the plasma reac-
tivity of complex gas mixtures containing NO, CH4, and O2 in
the presence of Ar [36] or N2 [45] as a carrier gas. The plasma
was ignited with a surfatron [46] device connected to a MW
power supply. A MW power of 75 W was used in all experi-
ments, and the pressure of the gases was maintained at 4 Torr.
◦
stream of synthetic air (>99% purity) at 650 C for 2 h. This
treatment ensured that the perovskite had the same oxidation
state in all experiments. The reaction mixtures were fed through
−
1
UNIT mass flow controllers with a total flow of 100 ml min
and He as gas balance. The concentrations of the reactants were
(
This low pressure is necessary to keep the plasma excited.) De-
0
.5% CH4–5% O2 and 2% CO–5% O2 in volume, respectively.
◦
tails of gas sample dosing, plasma excitation, and detection of
reactants/products (the latter by mass spectrometry) have been
provided previously [36,44,45]. In the present work, N2 or Ar
was used as a carrier gas, with 3600 ppm of CH4, 3000 ppm
The heating rate was 1 C/min up to maximum temperatures
◦
◦
of 500 C for CO oxidation and 650 C for methane oxidation.
A gas chromatograph provided with a Porapak packed column
and a thermal conductivity detector was connected online to
analyze the reaction products.
4
of NO, and 3 × 10 ppm of O2 added as minor components.
Experiments were done with both ternary (O2, CH4 plus the
carrier gas) and quaternary (O2, CH4, NO plus the carrier gas)
mixtures.
2
.3. Plasma characterization
A scheme of the reactor, composed entirely of quartz, is
shown in Fig. 2. The reactor used here was a modified version of
that used in previous investigations. The new configuration pro-
vides a way to incorporate the catalysts downstream the plasma
reaction zone. The reactor zone with the catalyst has a 1-inch
o.d. The powder was placed just below the plasma excitation re-
gion, still inside the glow zone of the plasma discharge, 10 cm
from the surfatron (see Fig. 2). This configuration allows the
active species coming from the plasma to interact directly with
the catalyst surface. The catalyst zone can be heated by an ex-
ternal furnace to check the effect of temperature on reaction
efficiency. The temperature was controlled by a thermocouple
placed at the external wall of the reactor near the catalyst re-
The activated species present in the glow discharge of the
plasma were characterized by optical emission spectroscopy
OES), a very powerful, nonintrusive diagnostic tool for detect-
(
ing in situ reactive plasma species and providing information
about the populations of atomic and molecular species in ex-
cited states [5,13,41,42]. The experimental setup consisted of
an optical fiber to collect the light coming from the plasma
glow, connected to a Jobin Ybon (HR250) scanning mono-
chromator and a Hamamatsu photomultiplier tube R928. The
emitted spectra are then collected and plotted by a computer.
This analysis was done by placing the optical fiber in the posi-
tion indicated in Fig. 2. The maximum resolution of the spectra
was 0.05 nm, and the estimated energy density [43] was ca.
gion. The amount of catalyst (perovskite or SiO ) was adjusted
2
to 100 mg, similar to the experiments with the conventional cat-
alytic reactors.
−3
1
0 W cm
.
Fig. 2. Schematic view of the plasma + catalyst reactor used for the experiments. The different parts of the system are indicated in the figure. Detail of the plasma
reactor. (For color figure the reader is referred to the web version of this article.)

J.L. Hueso et al. / Journal of Catalysis 247 (2007) 288–297
291
Table 1
Summary of CH and NO removal efficiencies and selectivity of CO forma-
4
2
tion by plasma reactions of ternary and quaternary gas mixtures
a
Gas mixtures Treatments
% CH
% NO
% CO
4
2
decomposition decomposition selectivity
yield
yield
b
b
Ar
N
Ar
N
2
2
c
CH :O
MWD
MWD
95–99
97–99
−
−
10
8
8
6
4
2
CH :O :NO
88
68
4
2
a
b
c
All treatments are referred to room temperature.
Carrier gas.
Microwave flowing discharge.
activated neutral species exist in the plasma and may react
in the gas phase. In addition, an equivalent concentration of
free electrons, usually bearing a high kinetic energy, is also
present to compensate for the positive charge of the ionized
species [48]. For the gas mixtures investigated in the present
work, Fig. 3a provides an indication of the type of intermedi-
ate species formed in the gas phase that account for the overall
reactivity of the system in the absence of catalyst. This figure
reports OES spectra of the plasma formed with the quaternary
gas mixture and Ar as carrier gas. A thorough analysis of this
type of spectra has been published recently [36,45]. For the pur-
poses of the present investigation, it is worth stating that peaks
(
a)
∗
∗
∗
∗
∗
and bands attributed to CO , OH , CH , H , and O species are
clearly distinguished in the spectra [42]. These and other simi-
lar species not yielding emission lines must be responsible for
the reactivity of the plasma when using Ar as a carrier gas. With
(b)
Fig. 3. OES spectra of the Ar + O2 + CH4 + NO (a) and N2 + O2 + CH4
+
NO (b) plasmas during reaction conditions. The insets show in an enlarged scale
some zones of the spectra to clearly distinguish bands attributed to different
species.
N used as a carrier gas, the spectrum in Fig. 3b is dominated
2
∗
by bands due to N species (second positive system) [42], with
2
∗
∗
small contributions of CN and NO species and other species
+
∗
e.g., N , NH ) not easily detectable at the scale used for
2
(
2
.5. X-ray photoemission spectroscopy characterization
the representation. The plasma reaction likely involving these
species produces the complete removal of NO, converted into
N2 and O2, and the combustion of CH4, converted mainly into
CO and H2O. We have discussed the mechanisms involved in
the corresponding plasma reactions in previous works [36,45].
X-ray photoemission spectroscopy (XPS) experiments were
carried out on a VG ESCALAB 210 apparatus provided with
a prechamber in which in situ plasma experiments can be car-
ried out. For this purpose, a plasma source was inserted in the
prechamber, in which catalyst or solid samples can be exposed
directly to the plasmas of different gas mixtures. More details
on this system were given in a recent study on the interaction
of different plasmas with the surface of carbon [47]. The spec-
tra were excited with the MgKα radiation of the X-ray source
and were calibrated in energy by referencing the binding energy
3
.2. Plasma catalysis for CH4 oxidation in ternary mixtures:
Effect of temperature
For a practical application of the plasma technology data
given in Table 1, the main results of the plasma reactions in-
dicate highly efficient NO removal but with high amounts of
CO production. In an attempt to define an integrated process
in which both NO and CH4 can be removed without produc-
ing high amounts of CO, we consider the possibility of using
the plasma in conjunction with a catalyst. Using the reactor de-
scribed in Fig. 2, we first studied the efficiency of the process
for ternary gas mixtures in the absence of NO. Table 1 shows
that plasma excitation of this ternary mixture without catalyst
produced the almost complete removal of CH4, which was con-
verted into CO + H2O. In this process, only about 10% of the
initial CH4 was converted into CO2 (cf. Table 1). In the pres-
ence of a catalyst, the amount of CH4 transformed into CO2
increased up to approximately 40% with both the perovskite
(
BE) scale of the spectra to the C 1s peak of the adventitious
carbon present on the surface of the catalyst. Experiments were
carried out after exposing the perovskite catalyst or mixtures of
catalyst + carbon to Ar + NO or N2 + NO plasmas (3000 ppm
NO in the gas mixture) for increasing periods. The pressure was
maintained at 0.04 Torr during these plasma treatments.
3
. Results
3
.1. OES analysis of plasma
A difference between conventional catalytic- and plasma-
activated reactions is that in the latter, radicals, ions, and other

292
J.L. Hueso et al. / Journal of Catalysis 247 (2007) 288–297
In connection with these plasma + catalyst results, it is inter-
esting to consider the catalytic activity of the perovskite catalyst
for conventional oxidation reactions. Fig. 5 reports the oxida-
tion yields of CO and CH4 into CO2. CO oxidation started at
◦
◦
ca. 250 C and was almost complete at 350 C. Methane oxi-
dation required much higher temperatures than CO oxidation,
◦
with an onset temperature of about 500 C. Although the oxi-
dation rate increased with temperature, only 65% methane was
◦
converted into CO2 at 650 C under our experimental condi-
tions. These results are in line with those of other similar cat-
alytic processes reported in the literature [38] and clearly sup-
port the notion that plasma activation of the perovskite catalysts
constitutes a singular process that enables the almost complete
◦
conversion of CH4 into CO2 at 200 C.
3
.3. Plasma catalysis in quaternary mixtures: Effect of
adding NO
The aim of this investigation was to define an integrated sys-
tem in which both NO and CH4 can be removed without any
significant production of CO. Along these lines, experiments
were carried out with the quaternary mixture (i.e., CH4, O2, and
NO with either Ar or N2 as a carrier gas) to determine whether
the presence of catalyst affects the overall performance of the
system. Fig. 6 compares, in the form of a bar diagram, the effi-
ciency of NO removal in three different experiments: (i) plasma
discharge without catalyst, (ii) plasma discharge in the pres-
ence of catalyst, and (iii) plasma discharge in the presence of
perovskite catalyst mixed with graphitic carbon. This latter sit-
uation simulates the effect of the soot particles that usually ac-
Fig. 4. Percentage of CH converted into CO as a function of temperature
4
2
in the plasma hybrid system in the presence of SiO and La Sr CoO
.
2
0.5 0.5
3−d
(a) Ternary mixture with Ar as carrier gas. (b) Ternary mixture with N as
2
carrier gas.
and the SiO2 catalysts (Fig. 4). It is very interesting that the
conversion efficiency was temperature-dependent and reached
◦
a value of about 90% at 190 C in experiments using either Ar
company the exhaust gases by the NO/CH removal processes.
x
or N2 as a carrier gas. It is also noteworthy that when SiO2 was
used as the catalyst, temperature had no significant effect on the
conversion yield, although differences did occur depending on
This figure clearly shows that the catalyst produced a drastic
decrease in NO removal, particularly when N was used as car-
2
rier gas. In comparison, the catalyst + carbon mixture partially
whether Ar (conversions of 40–45%) or N (conversions of ca.
restored the efficiency of the process. The effects of the inter-
action of the plasma on the surface of the catalyst were further
2
1
0%) was used as a carrier gas.
Fig. 5. Percentage of CO and CH converted into CO as a function of temperature for conventional catalytic oxidation processes.
4
2

J.L. Hueso et al. / Journal of Catalysis 247 (2007) 288–297
293
(a)
Fig. 6. NO conversion efficiencies at room temperature for quaternary mixtures
with either Ar or N as carrier gas and three different experimental situations:
2
plasma process, plasma in the presence of the perovskite catalysts and plasma
in the presence of a mixture of the perovskite catalyst plus carbon.
investigated by XPS; the results are described below and dis-
cussed in more detail in Sections 4.2 and 4.3.
Another point that must be considered when evaluating the
possible practical applications of this type of plasma + cata-
lyst process is the influence on the conversion yield of CH4
into CO2 when NO is added to the reaction mixture. Fig. 7
compares, in the form of bar diagrams, the CO2 conversion ef-
ficiencies for ternary and quaternary mixtures. For the three in-
vestigated situations, these diagrams show that, compared with
the ternary mixture results, the conversion efficiency of CH4
into CO2 was only slightly lower in the presence of NO.
(b)
Fig. 7. Effect of the NO addition on the CH into CO conversion yield at room
4
2
temperature with either Ar (a) or N (b) as carrier gas and three different ex-
2
perimental situations: plasma process, plasma in the presence of the perovskite
catalysts and plasma in the presence of a mixture of the perovskite catalyst plus
carbon.
Ar and 10 min in N2 and a subsequent decrease in both cases
at longer exposure times. In addition, a small peak at around
398 eV attributed to adsorbed –CN species also can be ob-
served for long exposure times [47,51]. Besides the N 1s peak,
the other elements present in the sample were also investigated
by this photoemission experiment; however, the Co 2p, La 3d,
O 1s, or Sr 3d spectra demonstrated no significant variation in
shape and/or intensity after these plasma exposure experiments
and are not reported here.
The results of a similar experiment carried out with a catalyst
+ carbon mixture are presented in Fig. 9. In this case, the peak
at 407.5 eV, attributed to –NO2 species, remained in the spectra
even after longer exposure times to the two plasmas. In addi-
tion, the peak at around 398 eV attributed to –CN species pro-
gressively increased in intensity with exposure time. According
to previous experiments in which pure carbon was exposed to
similar plasma mixtures [47], formation of –CN species occurs
on the surface of carbon exposed to this type of plasma and here
must be considered a consequence of the presence of carbon in
the treated samples. Meanwhile, the two other species appear-
ing at 406 and 407.5 eV can be formed on both the perovskite
and the carbon surfaces, in the latter case remaining with high
intensity after long exposure times (cf. Figs. 8 and 9). It is also
worth mentioning that, as shown in Fig. 10, plasma treatment of
the perovskite + carbon mixtures produced a significant (20%)
decrease in intensity of the C 1s peak at 284.6 eV, attributed to
3
.4. XPS investigation of plasma–catalyst interactions
The previous results demonstrate that the incorporation of
a catalyst into the plasma discharge may alter the distribu-
tion of products. In particular, Fig. 6 shows that the efficiency
of NO removal is drastically reduced by the presence of the
perovskite catalyst when compared with plasma-alone experi-
ments. To identify the reasons for this decrease and its partial
recovery when the catalyst is mixed with carbon, we carried out
a series of XPS experiments in which the catalyst or catalyst +
carbon mixtures were exposed in situ to plasmas of Ar(N ) +
2
NO. These experiments have provided information about the
intermediate species formed on the surface of the perovskite
exposed to these plasma discharges containing NO.
Fig. 8 shows a series of N 1s photoemission peaks taken
from the perovskite surface exposed to these plasmas for in-
creasing periods. The spectrum of the original sample has some
small peaks with BEs of 406–407.5 eV that can be attributed
to –NO and –NO (x ꢀ 2) adsorbed species [47,49,50]. How-
2
x
ever, the most remarkable features of this experiment are the
changes observed in the intensity of the N 1s peaks as a func-
tion of the exposure time to the plasmas. For the two plasma
mixtures, a common observation is a sharp increase in the in-
tensity of these N 1s peaks after exposure times of 20 min in

294
J.L. Hueso et al. / Journal of Catalysis 247 (2007) 288–297
Fig. 8. N 1s photoemission spectra of the perovskite exposed for increasing periods of time to a plasma of Ar + NO (left) or N2 + NO (right).
Fig. 9. N 1s photoemission spectra of the mixture perovskite + carbon exposed for increasing periods of time to a plasma of Ar + NO (left) or N + NO (right).
2
Scale factor corrections are included in some spectra.
graphitic or aliphatic carbon. The spectra in this figure present
two components, one at 284.6 eV and another at ca. 289 eV, due
tive for inducing the oxidation of CH4 into CO2. This result
contrasts with those summarized in Table 1 corresponding to
similar reactions without catalyst. The data in this table indi-
cate that methane is converted almost totally into CO and H2O
−
to COO and related species formed on the surface of carbon
−
or to –CO species formed on the surface of the perovskite.
3
(
only 10% CO2 conversion) in plasma-only conditions. Thus,
4
. Discussion
it is evident that incorporating the catalysts into the process
favors the posterior oxidation of CO into CO2. According to
Fig. 4, the efficiency of the oxidation to CO2 depends on the
type of catalyst and the type of carrier gas in the gas mixture.
The data in this figure indicate that the oxidation efficiency to
CO2 was relatively higher with Ar as a carrier gas than with
N2 as a carrier gas, particularly when working at low temper-
In what follows we will treat partial aspects of the work
carried out here separately and conclude with an overall assess-
ment of the possibilities of plasma + catalyst processes in the
removal of noxious components from exhaust gases.
4
.1. Synergy by the oxidation of CH4 (CO) with plasma +
◦
ature (i.e., 40–50 C) due to the effect of the plasma. Another
catalyst systems
significant difference in these plasma + catalyst processes is
◦
For ternary mixtures, the results summarized in Fig. 4 clearly
that at 40 C, the SiO2 rendered conversion efficiencies of CH4
demonstrate that the plasma + catalyst system is very effec-
into CO2 on the order of 40% with Ar but only 10% with N2.

J.L. Hueso et al. / Journal of Catalysis 247 (2007) 288–297
295
In summary, our plasma + catalyst experiment shows that
the oxidation of methane occurs through novel mechanisms that
imply not only the direct interaction on the catalyst surface of
∗
CO (CO ) species and activated species of oxygen, but also
the opening on the perovskite surface of new reaction routes
between them induced by both the plasma interaction and the
heating of the solid.
4
.2. Decrease in the capacity of NO removal in the presence
of the catalyst
We have found that the incorporation of the perovskite cata-
lyst produces a sharp decrease in the efficiency of NO removal
(cf. Fig. 6). In this way, the beneficial effect of the catalyst in the
oxidation of CO is detrimental with respect to the NO removal
capacity. The decreased removal efficiency must be a result of
recombination processes between plasma species occurring on
the catalyst surface. Results of the XPS experiments in which
the perovskite was exposed to Ar(N2) + NO plasmas provide
some clues as to the nature of these recombination processes.
Fig. 10. C 1s photoemission spectra of the mixture perovskite + carbon exposed
for increasing periods of time to a plasma of Ar + NO (left) or N + NO (right).
2
The enhanced CO oxidation efficiency when using Ar as a car-
rier gas can be accounted for by considering the intermediate
species detected in the plasma. The intensity of the OES bands
x 2
Figs. 8 and 9 show that –NO and –NO adsorbed species were
formed onto the surface of the perovskite exposed to these plas-
mas. The concentration of these species passed through a max-
imum and decreased when steady-state conditions are reached.
The OES spectra in Fig. 3 shows that a high concentration of
∗
∗
∗
due to CO , OH , and O is much higher with Ar than with
N2. It is also common knowledge in plasma physics that for
equivalent working conditions, Ar plasmas present higher elec-
tron temperatures and densities than nitrogen plasmas [52,53].
This is due to the existence of a wider diversity of channels for
the depletion of electrons in the case of molecular species such
as nitrogen (i.e., vibrational and rotational excitation, dissocia-
tive ionization) than in Ar (i.e., mainly metastable states) [53].
Thus, the high reactivity found at low temperatures with Ar as
carrier gas likely results from direct reactions on the surface of
∗
NO species were formed in the plasma of quaternary mixtures.
A similar situation was found for Ar(N2)–NO mixtures [36,45].
∗
These excited NO species can be very reactive toward any ox-
ide surface. In the gas phase, these intermediate species readily
dissociate into N2 and O2 as final products [36,45]. However, in
the presence of the perovskite, these species and eventually oth-
+
ers containing nitrogen (e.g., N ) likely will interact with the
2
oxide ions of the catalyst surface and give rise to the –NOx and
–NO2 adsorbed species [55] detected by XPS (cf. Figs. 8 and 9).
Decomposition of these species as an effect of the surface bom-
bardment by the other plasma species and/or free electrons in
the plasma yields NO and thus contributes to the decreased re-
moval efficiency of the system toward this minority component.
∗
∗
∗
the catalysts of CO (CO ) and O (OH ) species. The lower
concentrations of these species (and electrons) in the plasma of
the ternary mixture with N2 should be the reason for the lower
CH4 oxidation efficiency found for this gas mixture.
The results in Fig. 4 also show that the conversion yield into
CO2 increased with temperature when perovskite was used as a
catalyst. A similar enhancement was not found with SiO2. This
increased perovskite reactivity with temperature indicates that
some surface activation was induced by both the plasma and the
4.3. Effect of carbon on the removal of NO
Figs. 6 and 7 demonstrated that incorporating carbon into
the process restored its capacity to remove NO without affect-
ing the oxidation efficiency of CO/CH into CO . Soot is a
◦
effect of heating up to 200 C. In this sense, it is interesting to
note that no oxidation of either CO or CH4 was detected at T ꢀ
4
2
◦
2
00 C by conventional thermal processes (cf. Fig. 5). Although
usual component of the exhaust gases from many combustion
processes [22,23,37], and this it should have a similar role as
the carbon used in our experiment. The beneficial effect of car-
bon can be explained with the help of the results of the XPS
experiment shown in Fig. 9. According to these photoemission
spectra, besides the development of –NO and –NO species,
additional information is needed, we believe that the plasma +
catalyst synergetic effects demonstrated in Fig. 4 are related to
the formation of new active species of oxygen onto the surface
of the perovskites when they are exposed to activated species of
∗
∗
oxygen like the O or OH detected by OES (cf. Fig. 3).
Another common effect of the interaction of plasmas with
solids is the etching of their surfaces [13,14]. Cleaning or re-
moving certain species from the surfaces of the perovskites by
interaction with the plasma would provide adsorption sites for
x
2
likely formed mainly on the surface of the perovskite, the in-
teraction of the catalyst + carbon mixture with the Ar(N ) +
2
NO plasmas induced the formation of –CN species. Accord-
ing to previous XPS investigations on carbon [51], such species
are intermediates of the reduction of NO by carbon according
to a process like NO + C → –CN(ads) → N2. The detection
∗
the CO molecules (or CO species), a process known to be a
limiting step by the thermal oxidation of CO [54].

296
J.L. Hueso et al. / Journal of Catalysis 247 (2007) 288–297
of these –CN adsorbed species by XPS supports the supposi-
tion that the activity of carbon toward Ar(N2) + NO plasmas is
not affected by the presence of the catalyst. The fact that carbon
can be considered another reactant of the process is further con-
firmed by the results in Fig. 10 showing that the intensity of the
C 1s peak attributed to graphitic or aliphatic carbon decreased
with the exposure time to the Ar(N2) + NO plasma.
[2] J.N. Armor, Appl. Catal. B Environ. 1 (1992) 221.
[3] F. Garin, Catal. Today 89 (2004) 255.
[4] R. Atkinson, D.L. Baulch, R.A. Cox, R.F. Hampson, J.A. Kerr, J. Troe, J.
Phys. Chem. Ref. Data 18 (1989) 881.
[
5] E.A.H. Timmermans, J. Jonkers, A. Rodero, M.C. Quintero, A. Sola, A.
Gamero, D.C. Schram, J.A.M. van der Mullen, Spectrochim. Acta Part B
At. Spectrosc. 54 (1999) 1085.
[6] A. Ricard, C. Jaoul, F. Gaboriau, N. Gherardi, S. Villeger, Surf. Coat.
Technol. 188–189 (2004) 287.
[
[
7] J.T. Herron, D.S. Green, Plasma Chem. Plasma Process. 21 (2001) 459.
8] D.L. Baulch, C.J. Cobos, R.A. Cox, C. Esser, P. Frank, T. Just, J.A. Kerr,
M.J. Pilling, J. Troe, R.W. Walker, J. Warnatz, J. Phys. Chem. Ref. Data 21
5
. Conclusion
Plasma-catalytic processes are attracting the attention of
(1992) 411.
many researchers interested in developing new methods of
treating exhaust gases [15,16,19,20,25,27,29]. At the present
stage, most of the works in this field rely on an empirical ap-
proach in which the efficiency is empirically correlated with
the different parameters of the process. In the present work, we
aimed to develop an integrated experimental methodology that
takes into account both the plasma parameters and the process
efficiency. We used spectroscopic analysis of the plasma by
OES and that of the surface of the catalyst by XPS to evalu-
ate the active species of the plasma intervening in the reactions
and the intermediate species formed on the surface of the cat-
alyst exposed to the plasma. Combining all of this information
with the results of CH4 oxidation and NO removal, we were
able to demonstrate that the role of the perovskite catalyst in
the process is not limited to providing surface reaction sites at
which active plasma species can react among them. This effect,
which has been claimed for other plasma + catalyst processes
[
9] V. Guerra, P.A. Sa, J. Loureiro, J. Phys. D Appl. Phys. 34 (2001) 1745.
[10] A. Borrás, A. Barranco, A.R. González-Elipe, J. Mater. Sci. 41 (2006)
5220.
11] F. Gracia, F. Yubero, J.P. Holgado, J.P. Espinós, A.R. González-Elipe, T.
[
[
[
Girardeau, Thin Solid Films 500 (2006) 19.
12] A. Martín, J.P. Espinós, A. Justo, J.P. Holgado, F. Yubero, A.R. González-
Elipe, Surf. Coat. Technol. 151 (2002) 289.
13] A. Barranco, J. Cotrino, F. Yubero, J.P. Espinós, J. Benítez, C. Clerc, A.R.
González-Elipe, Thin Solid Films 401 (2001) 150.
[14] M.B. Kizling, S.G. Jaras, Appl. Catal. A Gen. 147 (1996) 1.
15] B.S. Rajanikanth, A.D. Srinivasan, V. Ravi, IEEE Trans. Dielect. Elect.
[
[
[
Insul. 12 (2005) 72.
16] F. Holzer, U. Roland, F.D. Kopinke, Appl. Catal. B Environ. 38 (2002)
163.
17] H.H. Kim, Plasma Process. Polym. 1 (2004) 91.
[18] C. Subrahmanyam, A. Magureanu, A. Renken, L. Kiwi-Minsker, Appl.
Catal. B Environ. 65 (2006) 150.
[
19] M. Magureanu, N.B. Mandache, P. Eloy, E.M. Gaigneaux, V.I. Parvulescu,
Appl. Catal. B Environ. 61 (2005) 12.
[
[
[
20] A. Rousseau, A.V. Meshchanov, J. Roepcke, Appl. Phys. Lett. 88 (2006).
21] M. Heintze, B. Pietruszka, Catal. Today 89 (2004) 21.
22] A.R. Martin, J.T. Shawcross, J.C. Whitehead, J. Adv. Oxid. Technol. 8
(2005) 126.
[
15,16,20] and observed here for SiO2, is complemented in the
case of the perovskite with its direct participation in the oxida-
◦
tion of CO into CO2 at temperatures of up to 200 C.
We also found that the presence of the perovskite is detri-
mental to the efficiency of NO removal. The interaction of the
active NO and other nitrogen plasma species with the surface
[23] J. Grundmann, S. Muller, R.J. Zahn, Plasma Chem. Plasma Process. 25
2005) 455.
[24] S. Broer, T. Hammer, Appl. Catal. B Environ. 28 (2000) 101.
(
∗
[25] J.O. Chae, J.W. Hwang, J.Y. Jung, J.H. Han, H.J. Hwang, S. Kim, V.I.
Demidiouk, Phys. Plasmas 8 (2001) 1403.
of the perovskite provides a route for the formation of new NO
molecules that compensate for those dissociated in the plasma.
A possible approach to overcoming this problem is to add car-
bon to the catalyst. This carbon acts as another reactant of the
[26] K.P. Francke, H. Miessner, R. Rudolph, Catal. Today 59 (2000) 411.
[27] T. Hammer, Plasma Sources Sci. Technol. 11 (2002) A196.
[28] K. Krawczyk, M. Mlotek, Appl. Catal. B Environ. 30 (2001) 233.
[29] H. Miessner, K.P. Francke, R. Rudolph, T. Hammer, Catal. Today 75
∗
process and, by reacting directly with the NO species, leads
(
2002) 325.
30] Y.S. Mok, D.J. Koh, K.T. Kim, I.S. Nam, Ind. Eng. Chem. Res. 42 (2003)
960.
31] R.G. Tonkyn, S.E. Barlow, J.W. Hoard, Appl. Catal. B Environ. 40 (2003)
207.
to its reduction into N2. In this way, the presence of soot parti-
cles in the exhausts of most combustion processes should not be
considered a negative factor, but rather a source of an efficient
reducing agent that, when activated with the plasma, favors the
decomposition of NO into N2.
[
[
2
[32] I.S. Yoon, R.G. Tonkyn, A.G. Panov, A.C. Kolwaite, S.E. Barlow, M.L.
Balmer, Abstr. Pap. Am. Chem. Soc. 220 (2000) U388.
[
[
[
[
33] A.M. Zhu, Q. Sun, J.H. Niu, Y. Xu, Z.M. Song, Plasma Chem. Plasma
Process. 25 (2005) 371.
34] M.X. Pei, H. Lin, W.F. Shangguan, Z. Huang, J. Environ. Sci. 17 (2005)
Acknowledgments
Financial support was provided by the Ministry of Sci-
ence and Education of Spain (projects PPQ2001-3108 and
ENE2004-01660 and a doctoral fellowship for J.L.H.). We
thank Cabot for providing the carbon nanoparticulates. We also
appreciate the collaboration of Dr. J.P. Holgado with the SEM
images and Dr. Ocaña for his helpful assistance in the develop-
ment of the spray pyrolysis experimental setup.
220.
35] H. Miessner, K.P. Francke, R. Rudolph, Appl. Catal. B Environ. 36 (2002)
53.
36] J.L. Hueso, A.R. González-Elipe, J. Cotrino, A. Caballero, J. Phys. Chem.
A 109 (2005) 4930.
[37] D. Fino, N. Russo, G. Saracco, V. Specchia, J. Catal. 217 (2003) 367.
38] M. Alifanti, N. Blangenois, M. Florea, B. Delmon, Appl. Catal. A Gen.
[
280 (2005) 255.
[
39] E. López-Navarrete, A. Caballero, V.M. Orera, F.J. Lázaro, M. Ocaña,
Acta Mater. 51 (2003) 2371.
References
[40] N. Closset, R.v. Doorn, H. Kruidhof, J. Boeijsma, Powder Diffr. 11 (1996).
[1] J.N. Armor, Catal. Today 38 (1997) 163.
[41] U. Fantz, Contrib. Plasma Phys. 44 (2004) 508.

J.L. Hueso et al. / Journal of Catalysis 247 (2007) 288–297
297
[
42] R.W.B. Pearse, A.G. Gaydon, in: The Identification of Molecular Spectra,
[48] B.M. Penetrante, S.E. Schultheis, Non-Thermal Plasma Techniques for
Pollution Control, Parts A and B, Springer, Berlin, 1993.
Champan & Hall, New York, 1976.
[
[
43] C.R. Wierenga, T.J. Morin, AIChE J. 35 (1989) 1555.
44] A. Yanguas-Gil, J.L. Hueso, J. Cotrino, A. Caballero, A.R. González-
Elipe, Appl. Phys. Lett. 85 (2004) 4004.
[49] D. Hendricksson, J. Hollande, W.L. Jolly, Inorg. Chem. 8 (1969) 2642.
[50] C.D. Batich, D.S. Donald, J. Am. Chem. Soc. 106 (1984) 2758.
[51] J.R. Pels, F. Kapteijn, J.A. Moulijn, Q. Zhu, K.M. Thomas, Carbon 33
(1995) 1641.
[52] S. Nowak, P. Groning, O.M. Kuttel, M. Collaud, G. Dietler, J. Vac. Sci.
Technol. A Vac. Surf. Films 10 (1992) 3419.
[53] E.I. Toader, Plasma Sources Sci. Technol. 13 (2004) 646.
[54] L.G. Tejuca, J.L.G. Fierro, J.M.D. Tascón, Adv. Catal. 36 (1989) 237.
[55] M.A. Peña, J.L.G. Fierro, Chem. Rev. 101 (2001) 1981.
[
[
[
45] J.L. Hueso, A.R. González-Elipe, J. Cotrino, A. Caballero, J. Phys. Chem.
A 111 (2007) 1057.
46] M. Moisan, Z. Zakrzewski, US Patent 4 906 898 (1990) to Universite de
Montreal.
47] J.L. Hueso, J.P. Espinós, A. Caballero, J. Cotrino, A.R. González-Elipe,
Carbon 45 (2007) 89.