Potential of supported copper and potassium oxide catalysts in the combustion of carbonaceous particles

Article abstract of DOI:10.1023/B:KICA.0000038089.98976.95

Different oxide carriers (TiO₂ and ZrO₂) as supports for low amounts of Cu²⁺ and K⁺ species (2 wt % as equivalent oxide) were tested in the catalytic oxidation of carbon black. The K-Cu/oxide catalysts were shown to have a lower soot combustion temperature than K/oxide, Cu/oxide, and pure oxide carriers. The K-Cu/ZrO₂ catalyst was found to be the most active; it exhibited activity in a loose contact nearly similar to that obtained in a tight contact mode. Physicochemical characterization by EPR, XPS, and TPR revealed the interaction of K⁺ species with Cu²⁺ species and the ZrO₂ carrier in K-Cu/ZrO₂ as well as a strongly distorted Cu²⁺ species on the ZrO₂ surface. The potassium ions ensure promoting effects towards the contact between the carbon black and the catalyst surface. Although potassium ions were found to lower the reducibility of the cupric oxide species, the oxidation rate of carbon black increased in the presence of K/oxide and K-Cu/oxide.

Full text of DOI:10.1023/B:KICA.0000038089.98976.95

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Kinetics and Catalysis, Vol. 45, No. 4, 2004, pp. 580–588.From Kinetika i Kataliz, Vol. 45, No. 4, 2004, pp. 614–621.
Original English Text Copyright © 2004 by Courcot, Pruvost, Zhilinskaya, Aboukaïs.
CATALYTIC REACTION MECHANISMS
Potential of Supported Copper and Potassium Oxide Catalysts
in the Combustion of Carbonaceous Particles1
D. Courcot, C. Pruvost, E. A. Zhilinskaya, and A. Aboukaïs
^
Laboratoire de Catalyse et Environnement, E.A. 2598, Université du Littoral–Cote d’Opale, MREID, 145, av. M. Schumann,
59140 Dunkerque, France
e-mail: courcot@univ-littoral.fr
Received December 23, 2002
Abstract—Different oxide carriers (TiO2 and ZrO2) as supports for low amounts of Cu2+ and K+ species
(2 wt % as equivalent oxide) were tested in the catalytic oxidation of carbon black. The K–Cu/oxide catalysts
were shown to have a lower soot combustion temperature than K/oxide, Cu/oxide, and pure oxide carriers. The
K–Cu/ZrO2 catalyst was found to be the most active; it exhibited activity in a loose contact nearly similar to
that obtained in a tight contact mode. Physicochemical characterization by EPR, XPS, and TPR revealed the
interaction of K+ species with Cu2+ species and the ZrO2 carrier in K–Cu/ZrO2 as well as a strongly distorted
Cu2+ species on the ZrO2 surface. The potassium ions ensure promoting effects towards the contact between the
carbon black and the catalyst surface. Although potassium ions were found to lower the reducibility of the
cupric oxide species, the oxidation rate of carbon black increased in the presence of K/oxide and K–Cu/oxide.
1
INTRODUCTION
of active components could be released from the cata-
lyst under operating conditions.
Nowadays, worldwide research efforts are under-
taken to lower particulate emissions from diesel
engines. The small size of diesel particles (≤2 µm) may
be linked to a number of health problems by its ability
to penetrate the body through the respiratory system.
Among the technologies considered for diesel particulate
removal, one way is the use of filters carrying a suitable
catalyst, which enables simultaneous soot filtration and
combustion in the temperature range 350–400°C.
In this work, we have investigated the potential of
stable oxide systems based on supported Cu and/or K
oxide species using titania TiO2 or zirconia ZrO2 as the
oxide carriers. Even if the activity of these systems
appeared to be lower than different salts, interesting
properties could arise from their higher stability. More-
over, the Cu–K/TiO2 and Cu–K/ZrO2 catalysts studied
in this work contain low equivalent amounts of CuO
or/and K2O (2 wt %) in order to keep a strong interac-
tion between the oxide carrier surface and the active
species after calcination. The activity of K–Cu/oxide
was tested in the oxidation of carbon black following
both “tight” and “loose” contact conditions. Their cata-
lytic behavior has been correlated to the physicochem-
ical properties, which were investigated by electron
paramagnetic resonance (EPR), temperature pro-
grammed reduction (TPR), and X-ray pholoelectron
spectroscopy (XPS).
The catalytic combustion of diesel particles is in
principle a solid–solid process where the contact
between the catalyst and the soot to be burned out
markedly influences the reaction kinetics. For example,
Neeft et al. [1, 2] showed that Fe2O3 is an active catalyst
in the “tight” contact mode but shows hardly any activ-
ity in “loose” contact. On the contrary, MoO3 and Sb2O3
exhibit activity in loose contact that is only slightly less
than what is observed in the tight contact mode. Labora-
tory preparation of soot–catalyst mixtures by milling
yields “tight” contact, whereas, in a particulate filter, the
soot–catalyst contact will typically be loose [3].
EXPERIMENTAL
In recent studies [4–6], the use of salts and mixed
compounds such as Cu–K–Mo–Cl, Cu–K–V, Cu–K–
V–Cl, or K0.7Cu0.3VO3 + KCl was proposed. The high
activity of these systems at temperatures lower than
400°C has been explained by the formation of low
eutectics that melts at the operating temperature, which
enables one to obtain a good contact between the active
phase and the carbon particles. However, the stability of
such catalytic systems is sometimes poor since vapors
The titanium oxide carrier was prepared from the
hydrolysis of titanium(IV) isopropoxide diluted in
2-propanol following a sol-gel method as described in
[7]. Zirconium oxide carrier was obtained by the pre-
cipitation method adding an ammonia solution to a
ZrOCl2 solution. The precipitate was washed, filtered,
and then dried at 100°C overnight. The resulting solids
were calcined under a flow of dry air for 4 h at 600°C.
After calcination treatment, ZrO2 and TiO2 possessed a
specific area of 61 and 6 m2/g, respectively. Supported
1
This article was submitted by the authors in English.
0023-1584/04/4504-0580 © 2004 MAIK “Nauka /Interperiodica”

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
POTENTIAL OF SUPPORTED COPPER AND POTASSIUM OXIDE CATALYSTS
581
Mass loss, %
DSC signal, arb. units
copper systems (Cu/oxide) and supported potassium
systems (K/oxide) were obtained, respectively, by
impregnation of a Cu(NO3)2 solution and a KHCO3
solution onto the calcined supports (TiO2 and ZrO2). A
coimpregnation procedure using the previous precur-
sors was performed to prepare the K–Cu/oxide sys-
tems. After drying, all solids were calcined for 4 h at
600°C. The amounts of CuO or/and K2O in the calcined
catalysts reached almost 2 wt %.
Carbonaceous particulate materials used for the
tests were Carbon Black (CB) N330 provided by
DEGUSSA. An elementary analysis revealed the fol-
lowing composition expressed in wt % [C, 97.23; H,
0.73; O, 1.16; N, 0.19; and S, 0.45]. In addition, CB
possesses a BET surface area of 80 m2/g. Before the
catalytic tests, CB (6 wt %) and catalyst (94 wt %) were
mixed in a ball miller for 40 min, designating a tight
contact between the carbon and the catalyst surface. To
obtain another (CB–catalyst) mixture corresponding to
the loose contact conditions, CB (6 wt %) and catalyst
(94 wt %) were simply mixed inside a zirconia ball
using a teflon stirring rod (light mixing). These latter
conditions allowed us to evaluate the activity of oxide
catalysts in experimental conditions similar to those of
real application, since the soot particles emitted
through the exhaust line can only be stopped by the cat-
alyzed filter surface. This contact mode corresponds to
loose contact conditions [2, 3].
(‡)
Ti
10
8
100
98
96
94
92
90
88
6
Tf
4
2
0
exo
86
300
400
500
600
T, °C
Mass loss, %
Temperature, °C
(b)
400
300
200
100
98
96
94
25 50 75 100 125 150 175 200 225
Time, min
The catalytic combustion was studied by simulta-
neous TG-DSC analysis with a NETZSCH STA 409
apparatus. About 50 mg of the CB–catalyst mixtures
were loaded into an alumina crucible and heated from
25 to 800°C using a heating rate of 5°C/min in air flow
(75 ml/min). As shown in Fig. 1a, by processing the
experimental data, the onset temperature (Ti) could be
derived (i.e., the temperature at which combustion
starts). The value of Ti is calculated from the TG curve
as the intersection between the baseline and the tangent
of the curve traced at the inflection point. Following a
similar method, the temperature (Tf) at which the com-
bustion of carbon is complete was obtained. Additional
catalytic tests were performed under isothermal condi-
tions considering the CB–catalyst mixtures in loose
contact. Samples were kept under an air flow
(75 ml/min) and heated to 400°C (5°C/min). The
weight loss rate was used as a measure for the CB oxi-
dation rate (Fig. 1b) in agreement with the experimental
approach described by van Setten et al. [8]. Once the
combustion starts, the oxidation rate progressively
increases. Then, a part of the TG curve that is practically
linear permits one to calculate the oxidation rate value
Fig. 1. Examples of thermal analysis curves recorded for the
catalytic test of (CB–catalyst) mixtures: (a) nonisothermal
conditions, (b) isothermal conditions.
Electron paramagnetic resonance (EPR) measure-
ments of the calcined catalysts were performed at
−196°C on an EMX Bruker spectrometer, using a cav-
ity operating with a frequency of ~9.5 GHz (X-band).
The magnetic field was modulated at 100 kHz, and the
power supply was sufficiently small to avoid saturation
effects. The g values are determined from the precise
frequency and magnetic field values.
X-ray photoelectron spectroscopy (XPS) spectra
were recorded on a Leybold–Heraeus LHS 10 spec-
trometer equipped with an AlKα anticathode (hν =
1486.6 eV). The binding energies were determined by
using the Zr 3d5/2 peak at 182 eV as standardization for
ZrO2-based catalysts and the Ti2p3/2 peak at 458.5 eV
for TiO2-based catalysts.
Temperature programmed reduction experiments
that corresponds to a maximum (w, µg/(gCB initial s)). The were performed on a Micromeritics Autochem 1900
oxidation rate then decreases to zero. The analysis of apparatus. The catalyst sample (75 mg) was first treated
combustion products was performed evaluating the in situ under O2 diluted in argon flow at 450°C for 1 h.
CO/(CO + CO2) molar ratio from a Varian 3600 chro- After cooling under argon flow, reduction treatment
matograph coupled to the thermobalance.
(3% H2 in argon) was performed with a total flow of
KINETICS AND CATALYSIS Vol. 45 No. 4 2004

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
582
COURCOT et al.
Mass, %
Mass, %
(‡)
(b)
100
100
90
90
1'
1
2
3
exo
exo
2'
3'
4'
4
300
400
500
600
700
300
400
500
600
700
Temperature, °C
Temperature, °C
Fig. 2. (a) TG-DSC curves obtained for the tests of (CB + TiO -based catalyst) mixtures in air flow: mixtures prepared following
2
loose contact conditions: (1) TiO , (2) Cu/TiO , (3) K/TiO , (4) K–Cu/TiO ; (b) TG-DSC curves obtained for the tests of (CB +
2
2
2
2
ZrO -based catalyst) mixtures in air flow: mixtures prepared following loose contact conditions: (1') ZrO , (2') Cu/ZrO ,
2
2
2
(3') K/ZrO , (4') K–Cu/ZrO .
2
2
30 mL/min from 25 to 700°C with a heating rate of nearly 160°C between the test performed with
K−Cu/TiO2 and TiO2 and nearly 200°C between the
test performed with K–Cu/ZrO2 and ZrO2. The catalyst
K–Cu/ZrO2 appeared then as the most reactive system for
the oxidation of carbon particles studied in this work.
10°C/min.
RESULTS
Catalytic Combustion of Carbon Black
In addition, we have compared these catalytic test
results with the ones obtained when CB–catalyst mix-
tures resulted from ball miller agitation (“tight con-
tact”). The corresponding results are displayed in
Fig. 3. In the case of pure oxide carrier and Cu/oxide
systems (i.e., ZrO2, TiO2, Cu/TiO2, and Cu/ZrO2), the
conversion of carbon black occurred in a lower temper-
ature range for the mixtures (CB–catalyst) in tight con-
tact than for the mixtures in loose contact. It indicates
that the contact between carbon black and the catalyst
surface has to be created by the physical force obtained
from ball milling in order to obtain a lower reaction
temperature.
Figures 2a and 2b show TG-DSC curves obtained
for the test of (CB–TiO2-based catalysts) and (CB–
ZrO2-based catalysts) mixtures, respectively, which
correspond to a loose contact mode. For the noncata-
lyzed reaction, we observed that the combustion of CB
in air flow occurred regularly between Ti = 530°C and
Tf = 635°C. The results obtained in the presence of pure
titania or pure zirconia indicated the low reactivity of
these oxide carriers. In the case of Cu/oxide catalysts,
the presence of a copper oxide phase only slightly
improved the reactivity of the corresponding systems.
Nevertheless, we can observe (Fig. 2) that the lowering
in reaction temperature is greater with the TiO2-based
catalyst than the ZrO2-based catalyst. Indeed, the reac-
tion temperature range for Cu/ZrO2 and ZrO2 exhibits
only little difference.
For catalysts containing potassium ions, the results
obtained revealed in each case higher activity in loose
contact in comparison with copper-based catalysts
(Cu/oxide) tested in tight contact conditions. Moreover,
In the case of K/oxide catalysts, we obtained that the Fig. 3a shows a similar reactivity of K/TiO2 toward the
oxidation reaction occurred between Ti = 440°C and carbon oxidation whatever the contact mode; the reac-
Tf = 500°C for K/TiO2 and between Ti = 405°C and Tf = tivity in loose contact is nearly similar for Cu–K/TiO2
505°C for K/ZrO2. These results indicate that the potas- and K/TiO2. It can be seen that the conversion of carbon
sium oxide species provide nearly identical effects can be obtained at a lower temperature if the mixing of
toward the reactivity in the oxidation of carbon black CB and the catalyst is performed by ball milling with
whatever the nature of oxide carrier ZrO2 or TiO2. The K–Cu/TiO2. On the contrary, the catalytic behavior of
presence of a potassium oxide species leads to a high K–Cu/ZrO2 is different since its reactivity is high and
improvement of the reactivity of the catalytic systems. very similar whatever the contact mode. These compar-
For the catalytic tests performed in the presence of isons clearly show that a high reactivity of potassium-
K−Cu/oxide, the conversion of carbon occurred in a containing catalysts can be obtained (Ti = 370°C, Tf =
low-temperature range. In fact, it is noteworthy to 445°C) in this oxidation reaction even if a loose contact
observe the shift toward lower temperatures, which was exists between the carbon black and the catalyst sur-
KINETICS AND CATALYSIS Vol. 45 No. 4 2004

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
POTENTIAL OF SUPPORTED COPPER AND POTASSIUM OXIDE CATALYSTS
583
(‡)
(b)
TiO2
ZrO2
Cu/TiO2
Cu/ZrO2
K/TiO2
K/ZrO2
K–Cu/TiO2
K–Cu/ZrO2
300
400
500
600
700
300
400
500
600
700
Temperature, °C
Temperature, °C
Fig. 3. Effect of metal oxide catalysts in tight (- - -) and loose (—) contact mode on the combustion temperature of carbon black:
(a) TiO -based catalysts, (b) ZrO -based catalysts.
2
2
face. To sum up, Cu/TiO2 and Cu–K/TiO2 are active and all potassium-containing catalysts revealed that
only CO2 was formed.
catalysts mainly in tight contact mode. The K–Cu/ZrO2
catalyst, on the contrary, exhibits a high activity in
loose contact mode that is closely similar to the activity
in tight contact.
Due to the differences in catalytic behavior evi-
denced for the TiO2-based and ZrO2-based catalysts,
investigations on the physicochemical properties of the
catalysts will be undertaken in order to reveal eventual
differences between the copper and potassium sup-
ported species stabilized in these systems and to get a
better understanding of the reaction mechanism
involved in the oxidation of carbon black in the pres-
ence of such catalysts.
To compare the oxidation rate of CB in the presence
of the different catalysts, CB oxidation tests were per-
formed under isothermal conditions at 400°C (Table 1).
This temperature value was chosen taking into account
the results obtained under nonisothermal conditions. At
400°C, although low oxidation rates should be expected
for pure oxide supports, the effect of the copper and/or
potassium species appeared to be of main interest. Table 1
gathers the CB oxidation rate values obtained in the pres-
ence of the different catalysts. It appeared that Cu/TiO2
allows one to obtain a higher CB oxidation rate than
Cu/ZrO2. The two K/oxide catalysts provided a similar
oxidation rate close to 110 µg/(gCB initial s). The results
clearly show that the highest rates are obtained for cat-
alysts containing both copper and potassium species.
However, an important difference in the oxidation rate is
revealed between K–Cu/TiO2 (173 µg/(gCB initial s)) and
K–Cu/ZrO2 (379 µg/(gCB initial s)). Thus, these data show
that potassium species have a role in providing high
oxidation rates in the CB oxidation reaction. Finally,
the interest in combining copper and potassium species
in such oxide systems is clearly revealed.
Characterization of the Catalysts
1. Electron paramagnetic resonance (EPR). Fig-
ure 4a shows the X-band EPR spectra of Cu/TiO2 and
K–Cu/TiO2 catalysts. For Cu/TiO2, the spectra relative
to Cu2+ species are the superimposition of three signals
denoted H, L, and I. All these signals were studied and
attributed earlier [7]. A signal with well-resolved
hyperfine structure (H signal) is characterized by
parameter values gzz = 2.32, gyy = 2.09, gxx = 2.10, Azz =
75 G, Ayy = 26 G, and Axx = 19 G. This signal has been
Table 1. Oxidation rates of CB determined from tests per-
formed under isothermal conditions at 400°C: (CB-catalyst)
mixtures in loose contact mode
Regarding the combustion products, it can be men-
tioned first that a selectivity of CO/(CO + CO2) of 5.2%
was obtained for the combustion of CB performed in
the absence of catalyst. In the case of loose contact and
tight contact CB–catalyst mixtures, carbon monoxide
(CO) was detected in low amounts in the presence of
TiO2 (loose, 4.5%; tight, 4.1%), ZrO2 (loose, 3.6%;
tight, 3.2%), and Cu/ZrO2 (loose, 3.1%; tight, 2.8%).
On the contrary, whatever the contact mode, the analy-
sis of combustion products in the presence of Cu/TiO2
Oxidation rate
of CB,
Oxidation rate
of CB,
TiO2-based
catalysts
ZrO2-based
catalysts
µg/(gCB initial s)
µg/(gCB initial s)
TiO2 support
Cu/TiO2
12
48
ZrO2 support
Cu/ZrO2
6
15
K/TiO2
109
173
K/ZrO2
116
379
K–Cu/TiO2
K–Cu/ZrO2
KINETICS AND CATALYSIS Vol. 45 No. 4 2004