Study of the Ru/Ce system in the oxidation of carbon black and volatile organic compounds

Article abstract of DOI:10.1134/S0023158407060122

The relationship between the state of Ru on CeO₂ and catalytic activity in the oxidation of carbon black (CB) and some volatile organic compounds (VOCs) was investigated for Ru/CeO₂ catalysts prepared by wet impregnation. It was demo

Full text of DOI:10.1134/S0023158407060122

ISSN 0023-1584, Kinetics and Catalysis, 2007, Vol. 48, No. 6, pp. 835–840. © MAIK “Nauka /Interperiodica” (Russia), 2007.
Published in Russian in Kinetika i Kataliz, 2007, Vol. 48, No. 6, pp. 893–898.
MECHANISMS
OF CATALYTIC REACTIONS
Study of the Ru/Ce System in the Oxidation of Carbon Black
and Volatile Organic Compounds1
S. Aouad, E. Saab, E. Abi-Aad*, and A. Aboukaïs
Laboratoire de Catalyse et Environnement E.A. 2598, Université du Littoral—Côte d’Opale,
MREI-1, 145 av. Maurice Schumann, Dunkerque, 59140, France
*e-mail: abiaad@univ-littoral.fr
Received June 30, 2006
Abstract—The relationship between the state of Ru on CeO2 and catalytic activity in the oxidation of carbon
black (CB) and some volatile organic compounds (VOCs) was investigated for Ru/CeO2 catalysts prepared by
wet impregnation. It was demonstrated that the addition of ruthenium to ceria significantly improved the reac-
tivity of the latter. The temperature programmed reduction (TPR) experiments of Ru/CeO2 showed that the oxy-
gen species of RuO2 was reduced at low temperatures. In addition, Electronic Paramagnetic Resonance (EPR)
studies of outgassed samples at different temperatures showed an anisotropic signal indicating that Ru(IV) was
reduced to intermediate valence states like Ru(III) before its total reduction to metallic Ru. It was concluded
that Ru–O–Ce bonds in the well-dispersed Ru species are highly fragile and its mobile oxygen is the active spe-
cies in the catalytic oxidation process.
DOI: 10.1134/S0023158407060122
1
1. INTRODUCTION
change oxidation state during operation (i.e., CeO2 to
CeO2 – x) maintaining structural integrity, thus allowing
oxygen uptake and release to occur easily. Ruthenium
oxide catalysts have been demonstrated to be active in
acetic acid [11] and propane [12] oxidation reactions,
but only few studies investigated its reactivity in diesel
soot oxidation.
The deleterious effect on human health of particu-
lates present in diesel engine and industrial emissions
has become a subject of great concern and has resulted
in the formulation of restrictive legislation both in the
EU and the US [1–5]. Although physical filtration of
the particulate from the emissions is possible, this solu-
tion has several drawbacks. Such physical traps suffer
from blocking and need periodic regeneration. In the
same time, various industrial processes continuously
generate air flows containing volatile organic com-
pounds (VOCs). Given the toxicity of these compounds
and considering their impact on the greenhouse effect,
it is necessary to treat these air flows before release into
the atmosphere [6]. Emission control strategies include
changes and improvements in the fuel used, engine
modifications, or exhaust after-treatment systems.
These latter provide satisfactory results for compliance
with the emission regulations where the engine modifi-
cations or fuel improvements are not sufficient. Using
oxidation catalysts in order to oxidize diesel soot is one
of the promising after-treatment systems. These cata-
lysts also provide significant reductions in the carbon
monoxide (CO) and hydrocarbons (HC) emissions.
They were reported to be active in PM reduction espe-
cially when low sulfur fuel is used. Recently, it has been
reported that the use of supports based on cerium oxide
confers interesting properties to soot combustion cata-
lysts due to high availability of surface oxygen and high
surface reducibility [7–10]. The success of oxygen stor-
age systems based on ceria is due to their ability to
In this paper, we describe the activity of ceria-sup-
ported ruthenium catalysts towards the oxidation of
carbon black (CB) particulates, propane, toluene, and
ortho-xylene. Catalysts containing various amounts of
ruthenium were prepared, and their catalytic activity
and CO2 selectivity have been investigated.
2. EXPERIMENTAL
2.1. Solids Preparation
Cerium hydroxide Ce(OH)4 was prepared by precip-
itation from cerium(III) nitrate hexahydrated solution
Ce(NO3)3 · 6H2O with an alkali solution of NaOH. The
resulting hydroxide Ce(OH)4 was filtered, washed, and
dried for about 20 h in a drying oven at 100°ë, then it
was calcined at 500°ë to obtain ceria CeO2. Different
volumes
of
ruthenium(III)
nitrosyl
nitrate
Ru(NO)(NO3)3 solution have been impregnated on ceria
in order to obtain solids with different ruthenium load-
ings (0.67, 1, 1.5, 2, 3, and 5 wt %). These solids are
designated as RuxCe (x = Ru wt %). After drying at
100°ë for about 20 h, solids were stabilized by calcina-
tion under dried air (35 ml min–1) at 600°ë for 4 h. Ele-
1
This article was submitted by the authors in English.
835

836
AOUAD et al.
room temperature, 100, 200, and 300°C. A 1 K min–1
temperature rise allowed reaching the treatment tem-
perature which was maintained for 1 h.
Table 1. Specific surface area (Ssp) and elementary content
(wt %) of the solids
Sample Ssp, m2/g Ru, wt % Ce, wt % O, wt %
CeO2
93
89
86
82
78
71
64
45
–
–
–
2.3. Activity and Selectivity Tests
Ru0.67Ce
RuCe
0.56
0.85
1.31
1.79
2.53
4.15
–
77.51
76.85
78.71
77.47
76.02
74.65
–
21.93
22.3
19.98
20.74
21.45
21.2
–
The catalytic test towards the combustion of CB
N330 (DEGUSSA) was studied by simultaneous ther-
mogravimetric (TG)–differential scanning calorimetry
(DSC) analysis with a NETZSCH STA 409 apparatus.
Before the test, 20 wt % of CB and 80 wt % of catalyst
were mixed slightly in an appropriate cell for 10 min,
“loose contact,” or heavily in a ball miller for 40 min-
utes, “tight contact.” Even though Neeft et al. [13]
found that the practical physical contact between soot
and the catalyst in a filter is much closer to a “loose
contact” condition, “tight contact” results were dis-
played to be of use as a comparison scale.
Ru1.5Ce
Ru2Ce
Ru3Ce
Ru5Ce
RuO2
mentary analyses were performed on the obtained sol-
ids and element weight percentages are given in Table 1.
Thirty milligrams of this mixture were loaded in an
alumina crucible and heated from room temperature up
to 1000°C (5 K min–1) under air flow of 75 ml min–1.
The VOCs (propene, toluene, and ortho-xylene) oxida-
tion was realized in a catalytic micro reactor coupled to
a Varian 3600 gas chromatograph using a double detec-
tion FID and TCD. A hundred milligrams of the catalyst
were tested in the presence of VOCs (air + propene =
100 ml min–1; 6000 ppm of propene or 1800 ppm of tol-
uene or 2000 ppm of ortho-xylene).
2.2. Characterization Techniques
Temperature programmed reduction (TPR) analyses
were performed on an ALTAMIRA AMI 200 apparatus
with a flow type reactor. Hydrogen (5 vol % in Ar) was
passed through a reaction tube containing the catalyst at
30 ml min–1. The tube was heated with an electric fur-
nace, and the amount of H2 consumed was monitored
with a TC detector.
Electron paramagnetic resonance (EPR) measure-
ments were performed at different temperatures from
−196°C up to room temperature on a EMX BRUKER
spectrometer with a cavity operating at a frequency of
As CB is consumed during the reaction, we can only
calculate the total selectivity which is given by the ratio
nCO /(nCO + nCO) where nCO and nCO represent the
2
2
2
~9.5 GHz (X band). The magnetic field was modulated mole numbers of emitted gas during the combustion.
at 100 kHz, and the power supply was sufficiently low CB/catalyst mixtures are introduced in a calcination
to avoid saturation effects. The g values were deter- reactor, and then the temperature is raised by a mean of
5 K min–1. Gas samples are taken with syringes during
the thermal phenomenon and mainly around the Tmax
temperature (Tmax determined by an earlier DSC-TG
test of the mixture). Then the syringe’s content is
injected in a gas chromatography (GC) VARIAN 3600.
The GC consists of two filled columns, one PORA-
PACK Q (column 1) and the other consists of a molec-
ular sieve 5 Å (column 2), connected with each other by
a six-way commutation valve. By extracting the com-
pounds only on the PORAPACK Q, a first composite
peak is detected, followed by a second peak corre-
sponding to CO2. The O2, N2, and CO gas are then ori-
ented to the molecular sieve 5 Å to be separated then
analyzed by a TCD to determine catalyst selectivity in
CB oxidation.
mined from precise frequency and magnetic field val-
ues. Vacuum treatments (3 × 10–5 mbar) of Ru/CeO2
catalysts were performed at different temperatures:
Weight loss, %
DTA signal, arb. units
0
T
max
TG
5
10
15
20%
DSC
3. RESULTS AND DISCUSSION
20
3.1. Carbon Black (CB) Catalytic Combustion
0
200
400
600
800
1000
Figure 1 shows the example of DSC-TG curves of
an “80% CeO2 + 20% CB” “loose contact” mixture. A
weight loss of ~20 wt % is observed, corresponding to
the combustion of the total mass of carbon black. Tmax
Temperature, °C
Fig. 1. Example of DSC-TG curves of “80% CO + 20%
CB” “loose contact” mixture.
2
KINETICS AND CATALYSIS Vol. 48 No. 6 2007

STUDY OF THE Ru/Ce SYSTEM IN THE OXIDATION OF CARBON BLACK
837
Table 2. Tmax for different 80% RuxCe + 20% CB mixtures
is the temperature of the maximum of the DSC-curve
and corresponds to the temperature of the highest rate
of CB combustion. Catalytic properties of solids are
evaluated by comparing these temperatures with those
of CB combustion without the catalyst in the same con-
ditions. Table 2 shows Tmax obtained for different “cat-
alyst + CB” mixtures. The total combustion of CB with-
out any catalyst was performed using SiC as an inactive
support and has exhibited only one exothermic peak
with a maximum of the combustion temperature at
about 629°C. For “tight contact” mixtures, a decrease
of about 250°C of the CB Tmax value was observed in
the presence of ceria. This was not the case of the
“CeO2 + CB” “loose contact” mixture as a nonsignifi-
cant temperature shift was observed (17°C). Thus, even
if ceria was shown as a very active support in the CB
combustion in “tight contact” mixtures [14], it appears
that it does not favor this same reaction under “loose
contact” conditions. However, when ruthenium is
added to ceria, a decrease of about 100 to 120°C of the
temperature values was observed for “loose contact”
mixtures, and about 270 to 280°C for “tight contact”
mixtures, showing that the presence of ruthenium
enables the soot particle combustion at lower tempera-
tures. Nevertheless, the catalytic reactivity is stable for
ruthenium loading ≥1.5% as Tmax remains unchanged at
about 490°C (“loose contact”) and 345°C (“tight con-
tact”) despite the increase of Ru loading from 1.5 to
5%. In the case of CB combustion without a catalyst,
87% CO2 selectivity is observed, while there is no CO
production during the same combustion in the presence
of different catalysts. RuxCe catalysts have shown a
100% selectivity for CO2 formation in the two types of
mixtures. In fact, ceria is entirely selective to CO2 for-
mation because of several of its properties in oxidation
catalysis like its capacity to capture and release oxygen
easily [15]. The interaction between the ruthenium and
Tmax, °C
Mixtures
“loose contact”
“tight contact”
SiC + CB
629
612
506
492
490
490
493
490
–
CeO2 + CB
Ru0.67Ce + CB
RuCe + CB
Ru1.5Ce + CB
Ru2Ce + CB
Ru3Ce + CB
Ru5Ce + CB
375
358
350
345
349
350
352
the support probably maintains or improves oxygen
transfer making its permanent presence on the solid
surfaces easier.
3.2. Catalyst Performance in VOCs Oxidation
The catalytic activity results of CeO2 and Ru/CeO2
are presented in Fig. 2 in terms of propene conversion
versus temperature. While CeO2 gives a light-off curve
with a T50% ≈ 390°ë, all RuxCe catalysts showed a
T50% ≈ 175–185°ë. CeO2 is less active than Ru/CeO2
catalysts, and a gain of about 200°C is obtained in the
oxidation reaction. On the other hand, the catalytic
activity of Ru1.5Ce was also investigated in the oxida-
tion of two other VOCs, toluene and o-xylene
(1800 and 2000 ppm, respectively). Figure 3 shows
light-off curves for propene, toluene, and o-xylene over
Ru1.5Ce. T50% for propene, toluene, and o-xylene con-
version are respectively 175, 215, and 230°C. Knowing
that total conversion of these VOCs alone does not
occur before 600°C and that propene conversion did not
Conversion, %
Conversion, %
100
100
80
80
CeO2
3
RuCe
60
Ru Ce
1.5
60
2
1
Ru Ce
3
Ru Ce
5
40
20
0
40
20
0
100
150
200
250
300
350
400
Temperature, °C
50
100 150 200 250 300 350 400
Temperature, °C
Fig. 3. Propene, toluene, and ortho-xylene conversion vs.
Fig. 2. Propene conversion vs. reaction temperature over
reaction temperature over the Ru Ce catalyst. (1) Propene,
1.5
ceria and different Ru Ce.
(2) toluene; (3) ortho-xylene.
x
KINETICS AND CATALYSIS Vol. 48 No. 6 2007

838
AOUAD et al.
As in the CB oxidation reaction, no CO formation in
VOCs catalytic oxidation reactions was observed.
135
3.3 Characterization of Catalysts
Table 1 shows the specific area of the different solids
after their calcination at 600°C. It was observed that the
specific area of RuxCe catalysts is lower than that of
ceria and it decreases with increase of the ruthenium
content. This is probably caused by ruthenium occupa-
tion of ceria surface sites due to the impregnation
method or the formation of ruthenium oxide clusters on
the support surface. In fact, the ruthenium oxide RuO2
specific area (45 m2/g) is lower than cerium oxide area
(93 m2/g) which could explain the decrease of specific
areas with increase of the ruthenium content.
TPR study was realized for all the catalysts after
their calcination at 600°C. Figure 4 shows hydrogen
consumption profiles of the different catalysts. Accord-
ing to the literature [11], ruthenium(IV) in RuO2 is
totally reduced into metallic Ru0 under hydrogen
(1.9 vol % in Ar) flow at temperatures close to 110°C.
RuO2 hydrogen consumption is found by calculating
the surface going between 25 and 250°C for each cata-
lyst. The theoretical values for this same consumption
are calculated while being based on the following reac-
tion:
66
127
Ru5Ce
68
68
129
Ru3Ce
Ru2Ce
70
Ru1.5Ce
RuCe
74
72
RuO2 + 2H2
Ru0 + 2H2O (Ru4+
Ru0).
Ru0.67Ce
225
Table 3 recapitulates the values of hydrogen con-
sumption necessary for the reduction of various RuxCe
catalysts. A smaller amount of H2 consumption than the
theoretical value may indicate that the reduced state Ru
is formed even during calcinations at 600°ë [11]. The
first reduction peak (Fig. 4) observed at low tempera-
tures for each catalyst corresponds to the reduction of
ruthenium oxide [11, 18]. Moreover, this first peak is
broad and seems to be composed of two or more peaks.
This can be due to the fact that ruthenium takes differ-
ent states of valence while reducing from Ru4+ (RuO2)
to Ru0. When the quantity of ruthenium increases, its
reduction occurs at lower temperatures. Moreover, for
Ru2Ce, there is the presence of a rather broad peak at
129°ë, this peak also exists and is broader and more
intense for Ru3Ce and Ru5Ce at 127 and 135°ë, respec-
tively. These peaks probably correspond to a part of the
ruthenium oxide that agglomerated on the surface of
ceria, which explains its reduction at higher tempera-
tures. On the other hand, for contents lower than ~2%,
this peak does not exist, which indicates the absence of
25
125
Temperature, °C
Fig. 4. TPR profiles for the different calcined catalysts.
exceed 22% over alumina at 500°C [16], the high
RuxCe activity observed can be attributed to a strong
synergetic effect obtained by combining Ru and CeO2.
In fact, in a previous work, it was shown that, for the same
content of Ru (wt %), a better activity of Ru–Ce–Al cat-
alysts is obtained when the ceria content in the support
increases [17]. This can confirm the presence of a syn-
ergetic effect between Ru species and ceria, which is
less expressed when Ru is in interaction with other sup-
ports. It is evident to note that in the propene oxidation
reactions, all catalysts showed approximately the same
activity, which is in accordance with the results
observed in CB oxidation.
Table 3. Hydrogen consumption for the catalysts
Theoretical/experi-
Ru0.67Ce
RuCe
Ru1.5Ce
Ru2Ce
Ru3Ce
Ru5Ce
mental data
Th [H2] (mmol g–1)
Exp [H2] (mmol g–1)
0.132
0.127
0.198
0.190
0.297
0.246
0.396
0.325
0.594
0.494
0.990
0.844
KINETICS AND CATALYSIS Vol. 48 No. 6 2007

STUDY OF THE Ru/Ce SYSTEM IN THE OXIDATION OF CARBON BLACK
839
agglomerated ruthenium. Starting from a content of
approximately 2%, the quantity of agglomerated ruthe-
nium increases with increase of the Ru content in the
solid, since Ru3Ce and Ru5Ce peaks at 127 and 135°C
are more intense and broader than the Ru2Ce peak at
129°C. Consequently, for Ru2Ce, Ru3Ce, and Ru5Ce,
the ceria surface is saturated with ruthenium oxide
which causes cluster formation of this oxide. Finally, as
was shown earlier, the reactivity of catalysts in carbon
black and VOCs oxidation is stable for catalysts con-
taining 1.5% and more of Ru. This can be confirmed by
TPR study, which has shown the formation of RuO2
clusters for Ru percentages exceeding 1.5%. Then, it
was considered that Ru species with reduction peaks at
low temperature (~70°C) are in tight interaction with
the support “Ru–O–Ce” and seem to be responsible for
the activity of the catalysts in carbon black and VOCs
oxidation. This reactivity is due to the weakness of the
Ru–O bonds [11], which allows the mobility of oxygen
and then an easy reduction of Ru oxide species at low
temperatures and thus the complete oxidation of CB
and VOCs.
g
zz = 2.047
4
g
yy = 1.926
g⊥ = 1.967
g
xx = 1.827
g|| (B)
3
g|| (A)
2
1
3000
3200
3400
Magnetic field, G
3600
3800
4000
Fig. 5. EPR spectra of Ru Ce treated under vacuum at dif-
ferent temperatures: (1) room temperature, (2) 100,
(3) 200, and (4) 300°C.
3
ment temperature reaches 100°C. These data can be
correlated with the presence of a double peak at low
temperature in TPR indicating that the reduction of
ruthenium Ru(IV) to give Ru(0) passes by different
valence states among which Ru(III) is responsible for
the EPR signal.
In order to deepen the reducibility study of RuxCe
catalysts, EPR measurements were carried out after a
vacuum treatment of catalysts at high temperatures,
knowing that a vacuum treatment can cause a partial
reduction of the catalysts. Figure 5 shows EPR spectra
recorded at room temperature for the Ru3Ce catalyst
treated at different temperatures. The EPR spectrum
presents two anisotropic signals of axial symmetry
(called A and B) characterized by the following param-
eters: g⊥ = 1.967, g|| = 1.939 for A, and g⊥ = 1.967, g|| =
1.946 for B. This spectrum is similar to those observed
in the literature [19, 20] where it was attributed to Ce3+
ions (f 1 ions : ge > g⊥ > g||) or more exactly to a trapped
electron in a Ce4+ free orbital which confers to ceria a
paramagnetic character.
4. CONCLUSIONS
In conclusion, the RuxCe catalysts are active in the
total oxidation of carbon black and VOCs. Even in the
case of a “loose contact,” these doped ruthenium cata-
lysts allowed the combustion of the CB at temperatures
close to ~490°C, largely lower than those obtained with
ceria (612°C) under practical conditions. These cata-
lysts were also very reactive under “tight contact” con-
ditions with a gain of about 280°C in CB oxidation.
They all also showed important reactivity in VOCs oxi-
dation. This reactivity is due to the weakness of the Ru–
O bonds, which allows the mobility of oxygen, an easy
reduction of ruthenium oxide species at low tempera-
tures, and thus the complete oxidation CB and VOCs.
Moreover, the reactivity of catalysts increases with the
ruthenium content until a content of 1.5%. In fact, for
this quantity of ruthenium, there is ruthenium oxide sat-
uration on the ceria surface, which explains stabiliza-
tion of catalyst reactivity from ≥1.5% out of ruthenium.
These catalysts were also shown to be selective in the
oxidation reactions with only CO2 as product, which is
promising for the use of these catalysts in the struggle
to control the carbonaceous particle pollution.
From a vacuum treatment temperature of 100°ë, a
new signal for which intensity increases with the
increase of the vacuum treatment temperature is
observed. In addition, this new signal becomes sharper
when the treatment temperature increases. It consists of
three distinct components; it seems thus that it is a phe-
nomenon of general anisotropy. The g factor values of
this signal are the following: gxx = 1.827, gyy = 1.926,
and gzz = 2.047 then giso = (gxx + gyy + gzz)/3 = 1.933.
Moreover, the vacuum treatment does not have much
influence on ceria. Consequently, the effect observed
(general anisotropy) is due to the ruthenium which
undergoes a partial reduction during this vacuum treat-
ment to pass from Ru(IV) to Ru(III). Indeed, the elec-
tronic configuration of Ru(III) is “4d5,” then it pos-
sesses 5 single electrons which contributes to the EPR
signal.
EPR analysis confirms the TPR results obtained pre-
viously by the fact that it shows the reduction of ruthe-
nium at low temperatures because the orthorhombic
signal due to ruthenium appears as the vacuum treat-
REFERENCES
1. Spiro, T.G. and Stigliani, W.M., Chemistry of the Envi-
ronment, London: Prentice Hall, 1996.
KINETICS AND CATALYSIS Vol. 48 No. 6 2007

840
AOUAD et al.
2. Meszaros, E., Fundamentals of Atmospheric Aerosol 12. Hosokawa, S., Fujinami,Y., and Kanai, H., J. Mol. Catal.
Chemistry, Budapest: Akadémiai Kiadó, 1999.
3. Parkinson, G., Chem. Eng., 2000, p. 25.
4. Diesel Emissions and Lung Cancer: Epidemiology and
Quantitative Risk Assessment, Cambridge: Health
Effects Institute, 1999.
5. Seaton, A., MacNee, W., Donaldson, K., and Godden, D.,
Paniculate Air Pollution and Acute Health Effects, 1995,
vol. 345, p. 176.
6. Salvador, S., Commandre, J.-M., and Kara, Y., Appl.
Thermal Eng., 2006, vol. 26, p. 2355.
A: Chem., 2005, vol. 240, p. 49.
13. Neeft, J.P.A., VanPruissen O.P., Makkee M., and Moul-
ijn J.A., Appl. Catal., B, 1997, vol. 12, p. 21.
14. Flouty, R., Abi-Aad, E., Siffert, S., and Aboukais, A.,
Appl. Catal., B, 2003, vol. 46, p. 145.
15. Van Doorn, J., Varould, J., Meriaudeau, P., Perrichon, V.,
Chevrier, M., and Gauthier, C., Appl. Catal., B, 1992,
vol. 1, p. 117.
16. Gluhoi, A.C., Bogdanchikova, N., and Nieuwenhuys, B.E.,
J. Catal., 2005, vol. 232, p. 96.
7. Miro’, E.E., Ravelli, F., Ulla, M.A., Cornaglia, L.M.,
17. Aouad, S., Saab, E., Abi Aad, E., and Aboukaïs, A.,
ande Querini, C.A., Catal. Today, 1999, vol. 53, p. 631.
Catal. Today, 2007, vol. 119, p. 273.
8. Pisarello, M.L., Milt, V., Peralta, M.A., Querini, C.A.,
18. Pinto, P., Calhorda, M.J., Straub, T., Miikkulainen, V.,
Raty, J., Suvanto, M., and Pakkanen, T.A., J. Mol. Catal.
A: Chem., 2001, vol. 170, p. 215.
and Miro’, E.E., Catal. Today, 2002, vol. 75, p. 465.
9. Milt, V.G. and Querin, C.A., Miro’, E.E., and Ulla, M.A.,
J. Catal., 2003, vol. 220, p. 424.
19. Abi-Aad, E., Bechara, R., Grimblot, J., and Aboukaïs, A.,
10. Setiabudi, J., Chen, G., Makkee, M., and Moulijn, J.A.,
Chem. Mater., 1993, vol. 5, p. 793.
Appl. Catal., B, 2004, vol. 51, p. 9.
11. Hosokawa, S., Kanai, H., Utani, K., Taniguchi,Y., Saito,Y., 20. Fierro, J.L.G., Soria, J., Sanz, J., and Rojo, M.J., J. Solid
and Imamura, S., Appl. Catal., B, 2003, vol. 45, p. 181.
State Chem., 1987, vol. 66, p. 154.
KINETICS AND CATALYSIS Vol. 48 No. 6 2007