Carbon black and propylene oxidation over Ru/CoxMg yAl2Oz catalysts

Article abstract of DOI:10.1016/j.crci.2012.12.013

The effects of Co_xMg_yAl₂O_z mixed oxides composition and ruthenium addition on the oxidation of propylene and carbon black (CB) were investigated. Different reactive cobalt and ruthenium oxide species were formed following calcination at 600 °C. The addition of ruthenium was beneficial for the CB oxidation under ''loose contact'' conditions and for propylene oxidation when the cobalt content was intermediate to low. The calculated activation energy for CB oxidation was decreased from 151 kJ mol^-1 for the uncatalyzed reaction to 111 kJ mol^-1 over the best catalyst.

Full text of DOI:10.1016/j.crci.2012.12.013

C. R. Chimie 16 (2013) 868–871
Contents lists available at SciVerse ScienceDirect
Comptes Rendus Chimie
www.sciencedirect.com
Full paper/M e´ moire
Carbon black and propylene oxidation over Ru/Co Mg Al O catalysts
x
y
2 z
a,
a
a,b,c
d
a
Samer Aouad *, Amal Aoun , Mira Skaf
, Madonna Labaki , John El Nakat ,
a
a
b,c
b,c
Bilal El Khoury , Pierre Obeid , Edmond Abi-Aad , Antoine Abouka ı¨ s
a
Department of Chemistry, Faculty of Sciences, University of Balamand, P.O. Box 100, Deir El Balamand, Kelhat, Tripoli, Lebanon
University Lille Nord de France, 59000 Lille, France
b
c
ULCO, LCE, 59000 Dunkerque, France
d
Laboratory of Physical Chemistry of Materials, Faculty of sciences II, Lebanese University, PO Box 90656, Fanar, Lebanon
A R T I C L E I N F O
A B S T R A C T
Article history:
The effects of Co Mg Al O mixed oxides composition and ruthenium addition on the
x
y
2 z
Received 27 September 2012
Accepted after revision 11 December 2012
Available online 16 January 2013
oxidation of propylene and carbon black (CB) were investigated. Different reactive cobalt
and ruthenium oxide species were formed following calcination at 600 8C. The addition of
ruthenium was beneficial for the CB oxidation under ‘‘loose contact’’ conditions and for
propylene oxidation when the cobalt content was intermediate to low. The calculated
ꢀ
1
Keywords:
activation energy for CB oxidation was decreased from 151 kJ mol for the uncatalyzed
Cobalt
ꢀ1
reaction to 111 kJ mol over the best catalyst.
Heterogeneous catalysis
Magnesium
ß 2012 Acad e´ mie des sciences. Published by Elsevier Masson SAS. All rights reserved.
Oxidation
´
´
R E S U M E
Supported catalysts
Ruthenium
x y 2 z
Les effets de la composition des oxydes mixtes Co Mg Al O ainsi que leur dopage par du
ruth e´ nium sur l’oxydation catalytique du noir de carbone et du prop e` ne ont e´ t e´ e´ tudi e´ s. La
calcination des solides a` 600 8C a contribu e´ a` la formation de diff e´ rentes esp e` ces r e´ actives
de cobalt et de ruth e´ nium. L’addition de ruth e´ nium s’est av e´ r e´ e b e´ n e´ fique pour l’oxydation
du noir de carbone sous les conditions de « contact faible » ainsi que pour l’oxydation de
prop e` ne lorsque la teneur du solide en cobalt est relativement faible. L’ e´ nergie d’activation
ꢀ
1
pour l’oxydation du noir de carbone est pass e´ e de 151 kJ mol en absence de catalyseur a`
ꢀ
1
1
11 kJ mol en pr e´ sence du solide le plus performant.
ß 2012 Acad e´ mie des sciences. Publi e´ par Elsevier Masson SAS. Tous droits r e´ serv e´ s.
1
. Introduction
Industrial chimneys, gasoline and diesel engines emit
presenting high oxygen storage capacity were investigated
for their use in catalytic oxidation reactions [4–6].
However, these latter are relatively costly and they may
present some environmental adverse effects. Cobalt,
magnesium and aluminium mixed oxides synthesized by
hydrotalcite route (HT) and used as supports for different
metals are potential candidates to deal with cost and
toxicity problems. They were tested in several catalytic
reactions as the oxidation of carbon black (CB) [7] and VOC
[8]. Moreover, supported precious metals such as Pt and Pd
are well established as efficient catalysts for VOCs
combustion [9] but unfortunately are very expensive
and are susceptible to poisonning. On the other hand,
ruthenium oxide based catalysts are cheaper and have
showed good reactivity in acetic acid, propene and CO
considerable amounts of particulate matter (PM) as well as
volatile organic compounds (VOC) [1]. One of the principal
solutions to reduce their emissions is the catalytic
oxidation technique [2,3]. Several mixed oxides systems
*
Corresponding author.
E-mail addresses: samer.aouad@balamand.edu.lb (S. Aouad),
amal.aoun@balamand.edu.lb (A. Aoun), mira.skaf@balamand.edu.lb
(
M. Skaf), m.labaki@ul.edu.lb (M. Labaki), john.nakat@balamand.edu.lb
J.E. Nakat), bilal.elkhoury@balamand.edu.lb (B.E. Khoury),
(
pobeid@balamand.edu.lb (P. Obeid), abiaad@univ-littoral.fr (E. Abi-Aad),
aboukais@univ-littoral.fr (A. Abouka ı¨ s).
1
631-0748/$ – see front matter ß 2012 Acad e´ mie des sciences. Published by Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.crci.2012.12.013

S. Aouad et al. / C. R. Chimie 16 (2013) 868–871
869
oxidation reactions and in various catalytic reactions such
3. Results and discussion
as: water-gas shift, ammonia synthesis and reduction of
NO [2,10]. Indeed, the association of ruthenium oxide and
ceria establishes a successful catalytic system in oxidation
reactions [11,12]. In fact, under oxidative conditions, Ru is
3.1. XRD characterization
XRD patterns of freshly calcined Ru/Co
x
Mg
catalyst
pattern presents diffraction lines corresponding to the
MgO periclase phase (JCPDS N845-0946) and the MgAl
mixed oxides phase (JCPDS N873-1959). Moreover, diffrac-
tion lines characteristic of RuO in a tetragonal phase
y 2 z
Al O
transformed to RuO
2
, showing highly desirable reactivity
6 2 z
catalysts are shown in Fig. 1. The Ru/Mg Al O
and stability and having a lower cost than other noble
metals. However, few if any studies have been devoted to
the total oxidation of volatile organic compounds and
carbon black in the presence of ruthenium supported on
cobalt, magnesium and aluminium mixed oxides.
2 4
O
2
(JCPDS N840-1290) are observed on the same pattern. For
In this study, Ru/Co Mg Al O
x y 2 z
solids were developed and
cobalt-containing catalysts, the patterns present diffrac-
their reactivitywas evaluated inCB and propyleneoxidation
reactions in order to consider their use as a potential
solution for some types of air pollution emissions.
tion lines corresponding to Co
CoAl (JCPDS N844-0160) and Co
3
O
4
(JCPDS N842-1467),
AlO (JCPDS N838-
2
O
4
2
4
0814) crystallized in a spinel phase. It is not possible to
distinguish between these different mixed oxides using
the XRD technique since they give similar patterns. It is
important to note that despite the presence of the same
ruthenium amount, the diffraction lines corresponding to
2
. Experimental
x y 2 z
Four different Co Mg Al O mixed oxides supports (x
and y are the numbers of cobalt and magnesium atoms
2 2 4 2 z
RuO became less intense for the Ru/Co Mg Al O catalyst
respectively in one formula unit) with an atomic ratio
and totally disappeared for higher cobalt contents. This
result shows that ruthenium oxide species tend to form
well-crystallized agglomerates when magnesium content
is high. On the other hand, the presence of cobalt enhances
the dispersion or ruthenium oxide species or even
facilitates the incorporation of some ruthenium atoms
into the spinel phase described above.
2+
2+
3+
(Co + Mg )/Al = 3 were prepared and stabilized as
detailed in [7]. An adequate quantity of Ru(NO)(NO
3 3
)
solution (1.7 wt.% Ru) was added to 1 g of the calcined
supports already dispersed in 50 mL of water to obtain
a nominal metallic Ru content of 1 wt.% in the final catalyst.
The obtained mixture is then stirred during
2 h
for maturation, and left overnight to dry in the oven at
6
6
0 8C. Catalysts were thermally stabilized by calcination at
00 8C (1 8C min ) under air flow (2L h ) during 4 h.
3.2. Catalytic results
ꢀ1
ꢀ1
X-ray diffraction (XRD) experiments were performed at
Fig. 2 shows the parameter T50% that represents the
ambient temperature on a Bruker D8 Advance diffractom-
˚
temperature at which 50% of carbon black or propylene is
eter using Cu K
a
radiation (1.5405 A). The diffraction
converted in the presence of Co
x
Mg
y
Al
2
O
z
supports or Ru/
Mg Al
patterns were indexed by comparison with the JCPDS files.
Temperature programmed reduction (TPR) was carried out
on a ZETON ALTAMIRA apparatus. Hydrogen (5 vol.% in Ar)
was passed through a U-shaped reaction tube containing
Co Mg Al catalysts. It is observed that Co
x
y
2
O
z
x
y
2 z
O
supports do not show significant catalytic reactivity in the
CB oxidation test under ‘‘loose contact’’ conditions.
ꢀ
1
the catalyst under atmospheric pressure at 30 mL min
.
,
s
ꢀ
1
The tube was heated with an electric furnace at 5 8C min
s
Ru/Co
and the amount of H
2
consumed is monitored with a
s
s
s
s
s
s
s
thermal conductivity detector.
s
s
The oxidation of carbon black CB (N330 Degussa:
6
Al
2
O
z
z
z
2
ꢀ1
S
sp = 76 m g
,
elementary analysis: 97.23 wt.% C;
s
0
.73 wt.% H; 1.16 wt.% O; 0.19 wt.% N; 0.45 wt.% S), was
*
*
Ru/Co
Ru/Co
^
4
Mg
Mg
2
Al
2
2
O
O
studied by simultaneous thermogravimetric (TG)
—
*
*
differential scanning calorimetry (DSC) analysis using a
Labsys Evo apparatus. Before test, 10 wt.% of CB and
*
^
2
4
Al
9
0 wt.% of catalyst were manually shaken for 5 min inside a
plastic tube to obtain ‘‘loose contact’’ conditions (L) or
grinded for 10 min using a mortar and a pestle to obtain
Ru/Mg
*
6
Al
2
O
z
‘‘tight contact’’ conditions (T). An amount of 10 mg of the
*
mixture were then loaded in an alumina crucible and
*
ꢀ
1
*
heated from room temperature up to 900 8C (5 8C min
)
*
ꢀ1
3 6
under air flow of 50 mL min . Propylene (C H 6000 ppm)
oxidation was carried out in a U-shaped catalytic fixed bed
micro-reactor (diameter: 1 cm/catalytic bed height:
20
30
40
50
60
70
80
2 θ (°)
2
a
mm), coupled to a Varian 3600 gas chromatograph using
double flame ionization and thermal conductivity
Fig. 1. XRD patterns for Ru/Co
x
Mg
(JCPDS N844-0160)
JCPDS N838-0814) mixed oxides; ‘‘*’’: RuO
y
Al
2
O
z
catalysts. ‘‘s’’: Co
3
O
4
4
(
(
JCPDS N842-1467),
CoAl
2
O
4
and Co AlO
2
detection. An amount of 100 mg of a catalyst were used
2
phase (JCPDS 40-1290);
mixed oxides
ꢀ
1
with a total gas flow (air + propylene) of 100 mL min and
‘
‘^’’: MgO periclase phase (JCPDS N845-0946); MgAl
(JCPDS N873-1959).
2 4
O
ꢀ1
a heating rate of 1 8C min
.

8
70
S. Aouad et al. / C. R. Chimie 16 (2013) 868–871
However, the Ru/Co
under the same contact conditions. For example,
substantial 70 8C decrease from 629 8C (without any
catalyst) down to 559 8C for Ru/Co Al was observed.
This can be explained by a ‘‘spill over’’ mechanism of O
over Ru oxide species. In fact the dissociative adsorption of
x
Mg
y
Al
2
O
z
catalysts decreased T50%
parameter (A) were calculated based on the method
a
described in [14]. In this method, the required data are
directly retrieved from TG curves and used in the below
equation:
6
2 z
O
2
ꢀꢀ
D
m
ꢁ
E
a
ꢀn
ln
m
¼ lnA ꢀ
RT
ꢀ6
oxygen over metallic ruthenium is low (ꢁ10 ), but when
a very small quantity of RuO is formed on the surface, the
D
t
2
where m is the actual mass of the sample undergoing
reaction, mthemass variation duringthetimeinterval t, R
the perfect gas constant, T the absolute temperature and n an
adjustment parameter. The assumption of n = 1 is made for
reaction of oxide formation is self catalyzed because the
dissociative adsorption coefficient of oxygen over rutheni-
um oxide is high (0.7) [13]. Moreover, in loose contact, the
mobility of ruthenium oxide species plays a major role
D
D
this analysis according to [14]. Computed E
are reported in Table 1. The activation energy for the non-
catalyzed reaction (Al used for dilution as it is inert in the
a
and lnA values
2
knowing that the Tamman temperature of RuO is close to
the observed temperature range in the carbon black
oxidation under ‘‘loose contact’’. In addition, a decrease
of T50% is observed when the amount of cobalt is increased.
Under ‘‘tight contact’’ conditions, it is clearly observed that
all prepared solids are significantly reactive in the total
oxidation of CB and that T50% are much lower than those
obtained for ‘‘loose contact’’ conditions. This is due to the
grinding that decreased the aggregates sizes, increasing
the number of contact points between catalytic sites and
CB. Moreover, it appears that ruthenium addition did not
contribute significantly to the catalytic reactivity under
2 3
O
considered reaction [15]) is in line with values already
reported in the literature for the combustionof carbonaceous
materials [16]. The presence of ruthenium decreased the
ꢀ1
same value to around 138 kJ mol under ‘‘loose contact’’
conditions. This is a substantial decrease keeping in mind
thatthefraction of molecules withenough energy to undergo
efficient collisions and lead to products is an exponential
function. For the ‘‘tight contact’’ mixture, the activation
energy is even lower for reasons that were stated above.
Concerning propylene oxidation, it is to be noted that
total conversion is not achieved before 690 8C [12]. The
support that showed the best reactivity in total propylene
‘‘tight contact’’ conditions. In fact, the grinding leads to
several direct contact points between cobalt oxide species
and carbon particulates, which makes the oxidation of
these latter possible at lower temperatures and making the
promoting effect of ruthenium less important. Thus, the
presence of ruthenium can be considered as a precursor for
the initiation of CB oxidation at low temperature and for
oxidation is Co
to ꢂ550 8C for the non-catalyzed reaction [12]. Compared
to the other supports, Mg Al presents the lowest
6 2 z
Al O with a T50% = 182 8C (Fig. 2) compared
6
2 z
O
reactivity. For supports with high cobalt contents, (x = 4
and x = 6), the addition of ruthenium did not significantly
enhance the catalytic reactivity. However, for lower cobalt
contents (x = 2 and x = 0), ruthenium addition contributed
to lower T50% (Fig. 2). This can be explained by the fact that
the weight percentage of ruthenium is small compared to
‘‘loose contact’’ conditions. Once the reaction is initiated,
cobalt oxides take over to complete the redox cycle. In
order to evaluate the contribution of the best catalyst (Ru/
6 2 z a
Co Al O ), the activation energy (E ) and the Arrhenius
that of cobalt for high
contribution is negligible compared to that of cobalt.
While, for low values, the catalytic reactivity of
ruthenium oxides becomes important compared to less
reactive cobalt oxide species (on an atom to atom basis)
and therefore T50% values were decreased. It is worth
x values and therefore its
CB (L) + supp.
CB (T) + supp.
Propylene + supp.
CB (L) + cat.
CB (T) + cat.
x
Propylene + cat.
7
6
5
4
3
2
1
00
00
00
00
00
00
00
2
mentioning that the selectivities towards CO and CO were
monitored in both CB and propylene oxidation reactions
and were found to be equal to 0% and 100%, respectively,
whatever the catalyst used.
The catalytic behavior of the different solids may be
partially explained by the XRD results presented at the
beginning of this section. In fact, catalysts with high cobalt
content showed the best reactivity owing to the good
dispersion of ruthenium oxide species, which initiate the
different reactions. However, the contribution of the
Table 1
Activation energies and Arrhenius pre-exponential factor for the non-
catalyzed and catalyzed CB oxidation.
0
2
4
6
ꢀ
1
y (CoxMgyAl2Oz)
Reaction
Al + CB
E
a
(kJ mol
)
lnA
2
O
3
151
138
111
18.5
17.3
17
Fig. 2. Effect of the support composition on T50% for the carbon black and
Ru/Co
Ru/Co
6
Al
Al
2
+ CB (L)
+ CB (T)
propylene oxidation reactions over Co
solids.
x y 2 z x y 2 z
Mg Al O and Ru/Co Mg Al O
6
2

S. Aouad et al. / C. R. Chimie 16 (2013) 868–871
871
support composition to the catalytic reactivity is difficult
to elaborate using the XRD results.
supports. This decrease is greater for higher cobalt
contents showing the formation of a greater amount of
easily reducible species in the presence of ruthenium. The
TPR results make it clear that the oxidation reduction
behavior of the solids is responsible for the catalytic
reactivity in the oxidation reactions. In fact, for higher
cobalt contents, more easily reducible species are present
and participate in the catalytic process. The addition of
ruthenium made the reduction of those species possible at
even lower temperatures, which allowed the earlier
initiation of the oxidation reactions.
3.3. TPR Results
In order to gain more insight in the catalytic process-
taking place during the reactions, a TPR study was done on
the different calcined solids. Fig. 3 shows TPR profiles of
calcined Co
x
y 2 z x y 2 z
Mg Al O supports and Ru/Co Mg Al O cat-
alysts. Mg Al
6
2
does not show any reduction peak due to the
non-reducibility of magnesium oxides species in the
studied temperature range. Cobalt containing supports
present two reduction peaks. The first peak (I) is situated in
the 270 8C to 370 8C range while the second peak (II)
appears in the 600 8C to 850 8C range. These two peaks are
4. Conclusion
x y 2 z
Co Mg Al O mixed oxides are cheap and reactive in
attributed to the reduction of Co
3
O
4
and Co-Al mixed
carbon black (tight contact) and propylene oxidation
reactions. However, under real conditions, the contact
between soot and the catalyst is more represented by the
‘‘loose contact’’ mixing. To promote the reactivity of these
supports under ‘‘loose contact’’ conditions, the addition of
a small amount of ruthenium (1 wt.%) can be considered.
The presence of ruthenium oxide species initiates the
oxidation reaction at lower temperatures especially when
cobalt content is low.
oxides species, respectively. The area of those peaks
increases with the increase of Co content. The maximum
temperature of peak II is shifted towards lower tempera-
tures when the amount of cobalt increases due to kinetic
considerations that suggest that the reduction tempera-
ture decreases when the amount of reactive sites increases.
The Ru/Mg
peak attributed to the reduction of RuO
6
Al
2
O
z
solid showed one composite reduction
species with
x
different cluster sizes. Cobalt-containing catalysts showed
two reduction peaks. The first peak (I) in the 150 8C to
Disclosure of interest
2
30 8C temperature range is attributed to simultaneous
reduction of ruthenium and cobalt oxide species. The
second peak (II) in the 280 8C to 420 8C temperature range
corresponds to the reduction of Co–Al mixed oxides
The authors declare that they have no conflicts of
interest concerning this article.
2 4
species that behave similarly to CoAl O spinel [17] that
Acknowledgments
will be reduced into Co(0). The modification of the
supports with ruthenium shifted the Co oxides species
reduction peaks to lower temperatures. This shift affected
both reduction peaks, (I) and (II), observed for calcined
The authors are very grateful to the Lebanese National
Council for Scientific Research for financial support,
contract 02-11-09.
The authors would like to thank the BRG 4/2010 for
financial support.
I
II
References
Ru/Co Al O
6
2
z
[
[
1] C. Chan, X. Yao, Atm. Env. 42 (2008) 1.
2] E. Meszaros, Fundamentals of atmospheric aerosol chemistry, Akade-
mia Kiado, Budapest, 1999.
Co Al O
6
2
z
[3] D. Homsi, S. Aouad, J. El Nakat, B. El Khoury, P. Obeid, E. Abi-Aad, A.
Abouka ı¨ s, Catal. Commun. 12 (2011) 776.
Ru/Co Mg Al O
[4] P. Papaefthimiou, T. Ioannides, X.E. Verykios, Appl. Catal. B 13 (1997) 175.
4
2
2
z
[5] G.L. Markaryan, L.N. Ikryannikova, G.P. Muravieva, A.O. Turakulova, B.G.
Kostyuk, E.V. Lunina, Colloid. Surface A 151 (1999) 435.
Co Mg Al O
4
2
2
z
[6] P. Bera, K.C. Patil, V. Jayaram, G.N. Subbanna, M.S. Hegde, J. Catal. 196
2000) 293.
(
Ru/Co Mg Al O
[7] A. Aoun, S. Aouad, H. El Nakat, B. El Koury, E. Abi Aad, A. Abouka ı¨ s, Phys.
Procedia. 21 (2011) 1.
2
4
2
z
[
8] C. Gennequin, S. Kouassi, L. Tidahy, R. Cousin, J.-F. Lamonier, G. Gar c¸ on,
P. Shirali, F. Cazier, A. Abouka ı¨ s, S. Siffert, C. R. Chimie 13 (2010) 494.
9] N. Nevers, Air Pollution Control, McGRAW-Hill, Singapore, 2000.
Co Mg Al O
2
4
2
z
[
Ru/Mg Al O
[10] S. Damyanova, B. Pawelec, K. Arishtirova, M.V. Martinez Huerta, J.L.
Fierro, Appl. Catal. A 337 (2008) 86.
6
2
z
[11] D. Terribile, A. Trovarelli, C. Leitenburg, A. Primavera, G. Dolcetti, Catal.
Mg Al O
6
2
z
Today 47 (1999) 133.
12] S. Aouad, E. Abi-Aad, A. Aboukais, Appl. Catal. B 88 (2009) 249.
13] A. B o¨ ttcher, H. Niehus, Phys. Rev. B 60 (1999) 14396.
[
[
5
0
250
450
650
850
[14] G.A. Stratakis, A.M. Stamatelos, Combust. Flame 132 (2003) 157.
[15] M.N. Bokova, C. Decarn, E. Abi-Aad, A.N. Pryakhin, V.V. Lunin, A.
Abouka ¨ı s, Thermochim. Acta 428 (2005) 165.
Temperature (ºC)
[16] E. Cauda, D. Mescia, D. Fino, G. Saracco, V. Specchia, Ind. Eng. Chem. Res.
44 (25) (2005) 9549.
Fig. 3. H
Co Mg Al
2
-TPR profiles for the different Co
x y 2 z
Mg Al O supports and Ru/
x
y
2
O
z
catalysts.
[17] Z. Wang, Z. Jiang, W. Shangguan, Catal. Commun. 8 (2007) 1659.