Ambient Pressure Photoemission Spectroscopy Reveals the Mechanism of Carbon Soot Oxidation in Ceria-Based Catalysts

Article abstract of DOI:10.1002/cctc.201600615

Removing soot is one of the most important challenges in minimizing the impact of combustion engines on the environment. Catalysts based on CeO₂ have proved suitable to oxidize soot owing to their capacity to store and release oxygen easily while maintaining structural integrity, although their mode of operation in a complex environment involving two solid phases (catalyst and soot) and a gas phase (oxygen) is not yet fully understood. Herein, we provide a study of the surface/subsurface of ceria–soot and ceria–zirconia–soot mixtures under working conditions by means of near-ambient-pressure photoelectron spectroscopy. Soot abatement involves two cooperative routes: one occurring at the ceria–soot interface with formation of oxygen vacancies and Ce^III and the other at the surface of soot, mediated by active superoxide species, which result from the reaction between gas-phase O₂ and oxygen vacancies. The two routes occur simultaneously and mutually reinforce each other.

Full text of DOI:10.1002/cctc.201600615

DOI: 10.1002/cctc.201600615
Communications
Ambient Pressure Photoemission Spectroscopy Reveals
the Mechanism of Carbon Soot Oxidation in Ceria-Based
Catalysts
Lluꢀs Soler,[a] Albert Casanovas,[a] Carlos Escudero,[b] Virginia Pꢁrez-Dieste,[b]
Eleonora Aneggi,[c] Alessandro Trovarelli,[c] and Jordi Llorca*[a]
Removing soot is one of the most important challenges in
minimizing the impact of combustion engines on the environ-
ment. Catalysts based on CeO2 have proved suitable to oxidize
soot owing to their capacity to store and release oxygen easily
while maintaining structural integrity, although their mode of
operation in a complex environment involving two solid
phases (catalyst and soot) and a gas phase (oxygen) is not yet
fully understood. Herein, we provide a study of the surface/
subsurface of ceria–soot and ceria–zirconia–soot mixtures
under working conditions by means of near-ambient-pressure
photoelectron spectroscopy. Soot abatement involves two co-
operative routes: one occurring at the ceria–soot interface
with formation of oxygen vacancies and CeIII and the other at
the surface of soot, mediated by active superoxide species,
which result from the reaction between gas-phase O2 and
oxygen vacancies. The two routes occur simultaneously and
mutually reinforce each other.
the dynamics of the reaction.[2] It is generally assumed that
soot oxidation proceeds through a Mars–Van Krevelen mecha-
nism, that is, lattice oxygen in the first few surface layers of
ceria is transferred onto soot, and gaseous O2 fills up the va-
cancies created on the oxide in a subsequent step.[3] However,
the mechanism of action is also associated with the availability
of adsorbed active oxygen species that spillover onto the soot
À
surface.[3a,4] Formation of paramagnetic O2 superoxide species
2À
and diamagnetic O2 peroxide species is claimed to occur if
reduced CeO2Àx is exposed to O2,[5] and it is suggested that
these are indeed the precursor surface species that are respon-
sible for soot oxidation.[3a,6] In the present study, we use oper-
ando ambient-pressure X-ray photoelectron spectroscopy (AP-
XPS) for the first time to study the surface of CeO2 and
Ce0.8Zr0.2O2 catalysts during soot oxidation. AP-XPS is a unique
surface-sensitive characterization tool essential to identify the
active species at work in reducible oxides such as ceria-based
materials, as the surface restructuring driven by the reaction
environment induces strong changes in their architecture that
cannot be followed under ultrahigh vacuum conditions.[7]
A sample of conventional ceria and a sample of Ce0.8Zr0.2O2
mixed with soot were investigated in this study. A sample of
ceria without soot was used as blank. The mixing of the cata-
lysts with carbon soot (weight ratio catalyst/soot=20:1) was
accomplished under the conventional tight contact mode for
Soot elimination constitutes a serious environmental and
health concern, as soot particles are undesired byproducts
formed in combustion processes and emitted as a main pollu-
tant from diesel engines. Usually, particle traps are used for
soot removal, and to avoid filter blocking, effective regenera-
tion systems have been developed to oxidize soot. Thermal
combustion of soot usually requires temperatures above
6008C, and catalysts play a key role in lowering the ignition
temperature. Ceria-based catalysts are among the most effec-
tive for diesel soot oxidation,[1] and incorporation of Zr and
rare earth elements, particle morphology, and the ceria–soot
interface have been demonstrated to be strongly involved in
[8]
CeO2 and under supertight contact achieved by high-energy
milling of Ce0.8Zr0.2O2 and soot for 8 h.[2b] Both materials crystal-
lize in a cubic fluorite structure and exhibit similar surface
areas and particle sizes (Table S1, Supporting Information).
Soot clumps are easily recognized in the high-resolution trans-
mission electron microscopy (HRTEM) images of the CeO2–soot
sample, whereas high-energy milling of Ce0.8Zr0.2O2 and soot re-
sulted in the formation of a core of oxide particles wrapped in
a thin carbon envelope (Figure S1), in accordance with earlier
reports.[2b,8] This greatly increases the number and quality of
contact points between the catalyst and carbon, which shifts
the combustion of soot to exceptionally low temperatures
(Table S1).
[a] Dr. L. Soler, Dr. A. Casanovas, Prof. J. Llorca
Institute of Energy Technologies and Centre for
Research in Nanoengineering
Universitat Politꢀcnica de Catalunya
Diagonal 647, ed. ETSEIB, 08028 Barcelona (Spain)
E-mail: jordi.llorca@upc.edu
[b] Dr. C. Escudero, Dr. V. Pꢁrez-Dieste
Three operando AP-XPS experiments (ALBA synchrotron
light source) were performed at 0.1 kPa over CeO2: one, heat-
ing CeO2 from room temperature up to 4508C under an at-
mosphere of argon to be used as the blank experiment; two,
heating the CeO2–soot mixture from room temperature up to
5508C under an atmosphere of argon and then replacing
argon with O2; three, heating the CeO2–soot mixture from
room temperature up to 5508C under an atmosphere of O2. It
ALBA Synchrotron Light Source
Carrer de la Llum 2–26, 08290 Cerdanyola del Vallꢀs, Barcelona (Spain)
[c] Dr. E. Aneggi, Prof. A. Trovarelli
Dipartimento Politecnico
Universitꢂ di Udine
Via del Cotonoficio 108, IT-33100 Udine (Italy)
Supporting Information and the ORCID identification number(s) for the
author(s) of this article can be found under http://dx.doi.org/10.1002/
cctc.201600615.
ChemCatChem 2016, 8, 1 – 5
1
ꢂ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Communications
is well known that heating CeO2 under an inert atmosphere
causes the formation of oxygen vacancies and the concomitant
reduction of CeIV to CeIII.[9] As expected, CeIII increased with
temperature (Figures 1a and S2). However, in the presence of
soot, ceria reduction was significantly higher (Figures 1 and 2),
and consequently, the ceria lattice oxygen atoms directly react
with soot, in accordance with the Mars–Van Krevelen mecha-
nism and with previous reports.[10] This observation is also sup-
ported by the progressive decrease in the C/Ce atomic ratio as
the temperature was increased (C/Ce=24.4 at 4508C vs. 16.0
at 5508C).
dation of residual soot (C/Ce from 16 to 2.9). Under an argon
atmosphere, the O1s signal recorded at a binding energy of
528.7 eV (Figure 3a) corresponds to ceria lattice oxygen, and
the minor contribution at a binding energy of 530.7 eV is as-
At 5508C, argon was replaced with O2, and as expected, CeIII
was reoxidized to CeIV to a large extent (Figure 2d), which re-
duced the total amount of CeIII from approximately 50 to 16%.
The introduction of oxygen on a heavily reduced ceria surface
results not only in the reoxidation of CeIII but also in deep oxi-
Figure 3. O1s spectra recorded a) over CeO2 under an argon atmosphere at
4508C and over CeO2–soot b) at 5508C under an argon atmosphere and
c) by replacing argon with O2 at 5508C. The spectrum in panel d corre-
sponds to the CeO2–soot sample heated directly under an O2 atmosphere
from ambient temperature (N O2 gas phase).
cribed to different ceria lattice oxygen environments upon va-
cancy formation.[11] Notably, in the presence of soot (Fig-
ure 3b), there is a broadening of the O1s signal at high bind-
ing energies with respect to the spectrum of bare CeO2 (Fig-
ure 3a), which points to a variety of surface/subsurface oxygen
environments that can be directly related to the higher
amount of CeIII in the CeO2–soot sample (bare soot showed an
O/C ratio of only ꢀ0.04; thus, the contribution of soot to the
O1s signal is negligible). Upon contact with oxygen, in addi-
tion to the peaks at binding energies of 538.6 and 537.5 eV
corresponding to paramagnetic O2 in the gas phase,[12] two
peaks at binding energies of 530.5 and 532.2 eV appeared (Fig-
ure 3c). According to the literature, these peaks can be attrib-
Figure 1. Amount of reduced ceria measured at different temperatures a) in
bare CeO2 under an argon atmosphere, b) in a CeO2–soot sample under an
argon atmosphere, c) in a CeO2–soot sample under an O2 atmosphere, and
d) in a Ce0.8Zr0.2O2–soot sample under argon atmosphere.
2À
À
uted to peroxide (O2
respectively.[13]
) and superoxide (O2 ) species,
In a separate experiment, the CeO2–soot sample was heated
from room temperature up to 5508C under an atmosphere of
O2 (Figures 1c and S3). In this case, the amount of CeIII species
remained low over the entire temperature range, which indi-
cated that the oxygen vacancies created at the ceria–soot in-
terface reacted rapidly with O2, as expected. As outlined
above, the reaction between molecular O2 and the oxygen va-
cancies formed active oxygen species (O2 and O2À),[14] which
2À
were clearly seen in the O1s spectra of the experiment per-
formed under an atmosphere of O2 (Figure 3d). The amount of
À
O2 species observed in the experiment performed under an
atmosphere of argon and then switched to an oxygen atmos-
phere was lower than that recorded in the experiment per-
formed under an atmosphere of O2 (O22À/O2À =16.4 vs. 4.8),
Figure 2. Ce 3d spectra recorded a) over CeO2 under an argon atmosphere
at 4508C and over CeO2–soot under an argon atmosphere b) at 4508C, c) at
5508C, and d) by replacing argon with O2 at 5508C (N argon gas phase).
&
ChemCatChem 2016, 8, 1 – 5
www.chemcatchem.org
2
ꢂ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communications
which might indicate that the reactivity of the superoxide spe-
cies toward soot oxidation is higher than that of the peroxide
species.
Our AP-XPS results support a mechanism in which ceria is
first reduced by soot at the interface between the carbon par-
ticles and the ceria surface, which releases CO2. Then the re-
duced ceria surface reacts with gaseous oxygen, which refills
the oxygen vacancies and yields peroxide/superoxide species.
These active oxygen species likely migrate to the carbon sur-
face and react efficiently to form CO2 at a much lower energy
cost. Then, a new contact point between soot and the surface
of ceria is established, and the cycle starts over again
(Scheme 1). To further validate this scheme, we studied
Figure 4. Ce3d and O1s spectra recorded a,c) over Ce0.8Zr0.2O2–soot at
5508C under an argon atmosphere and b,d) by replacing argon with O2 (N,
argon gas phase; N, O2 gas phase).
Scheme 1. Scheme of the mechanism for soot oxidation over ceria-based
catalysts. 1) Reaction between soot and ceria at the contact points. 2) For-
mation of oxygen vacancies at the interface and release of CO2. 3) Reaction
between O2 and oxygen vacancies. 4) Reoxidation of ceria surface and for-
to 23.2, in agreement with the FTIR and EPR spectroscopy
measurements of previous reports,[15] which thus supports the
À
À
mation of active oxygen species (O2 and O22À). 5) Spillover of active oxygen
idea that the superoxide O2 species formed by reaction be-
species onto soot. 6) Reaction between soot and active oxygen species and
release of CO2. The cycle starts over again at the new contact points be-
tween soot and ceria until soot is completely oxidized.
tween molecular O2 and oxygen vacancies redistributed on the
catalyst surface.
By using ambient-pressure X-ray photoelectron spectroscopy
(AP-XPS) we provided a direct measurement of the unique be-
havior of the CeIII/CeIV redox chemistry and the influence of the
reaction environment in the oxidation of carbon soot over
ceria-based catalysts. The AP-XPS technique demonstrated the
capacity of ceria to oxidize soot owing to the formation of
oxygen vacancies with reduction of CeIV to CeIII and the signifi-
cant formation of active oxygen surface species owing to the
reaction of oxygen vacancies and molecular O2. A general two-
way cooperative mechanism of soot oxidation over ceria-based
materials was, therefore, demonstrated, providing reliable in-
formation on the local characteristics of the surface/subsurface
species involved in this complex heterogeneous solid/solid/gas
catalytic process.
Ce0.8Zr0.2O2–soot mixed under supertight conditions, which
showed an extraordinarily low soot combustion temperature
(Table S1).[2b] The amount of CeIII upon heating under an argon
atmosphere was, in all cases, higher than that recorded over
the CeO2–soot sample (Figures 1b,d, 4a, and S4). The O1s
spectrum recorded under an argon atmosphere showed three
components at binding energies of 528.7, 530.2, and 532.2 eV
(Figure 4c). The peaks at binding energies of 528.7 and
530.2 eV correspond to lattice oxygen and surface oxide ions
with low coordination resulting from vacancy formation, re-
spectively. The peak at a binding energy of 532.2 eV, attributed
to superoxide species, was extraordinarily intense. At 5508C,
the argon atmosphere was replaced with O2, and the amount
of CeIII immediately decreased from approximately 60 to 23.5%
(Figure 4b). At the same time, the O1s peak of the surface
oxide ions with low coordination associated to CeIII present at
the surface/subsurface (at a binding energy of 530.2 eV) de-
creased owing to the reoxidation of ceria (Figure 4d). In addi-
tion, the superoxide signal at a binding energy of 532.1 eV do-
minated the spectrum and the O/Ce ratio increased from 17.4
Experimental Section
CeO2 and Ce0.8Zr0.2O2 were prepared by precipitation of an acidic
solution of the nitrate precursors with NH4OH in the presence of
H2O2. AP-XPS experiments were performed at the CIRCE beamline
of the ALBA synchrotron light source at a sample pressure of
0.1 kPa. A full description of the methodology is given in the
Supporting Information.
ChemCatChem 2016, 8, 1 – 5
www.chemcatchem.org
3
ꢂ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Communications
5348; d) E. Saab, E. Abi-Aad, M. N. Bokova, E. A. Zhilinskaya, A. Abouka¨ıs,
Carbon 2007, 45, 561–567.
Acknowledgements
[6] a) M. Machida, Y. Murata, K. Kishikawa, D. Zhang, K. Ikeue, Chem. Mater.
2008, 20, 4489–4494; b) M. S. Gross, M. A. Ulla, C. A. Querini, J. Mol.
Catal. A 2012, 352, 86–94.
This work was funded by Ministry of Economy and Competitive-
ness (MINECO)/Fondo Europeo de Desarrollo Regional (FEDER)
through grant ENE2015-63969-R. L.S. is grateful to Generalitat de
Catalunya for a Beatriu de Pinꢃs grant (2013 BP-B 00007). J.L. is
a Serra Hfflnter Fellow and is grateful to ICREA Academia pro-
gram. E.A. and A.T. are grateful to regione Friuli Venezia Giulia.
The AP-XPS experiments were performed at the CIRCE beamline
at ALBA Synchrotron with the collaboration of ALBA staff.
[7] a) D. E. Starr, Z. Liu, M. Hꢄvecker, A. Knop-Gericke, H. Bluhm, Chem. Soc.
Rev. 2013, 42, 5833–5857; b) N. J. Divins, I. Angurell, C. Escudero, V.
Pꢁrez-Dieste, J. Llorca, Science 2014, 346, 620–623; c) K. Mudiyanselage,
S. D. Senanayake, L. Feria, S. Kundu, A. E. Baber, J. Graciani, A. B. Vidal, S.
Agnoli, J. Evans, R. Chang, et al., Angew. Chem. Int. Ed. 2013, 52, 5101–
5105; Angew. Chem. 2013, 125, 5205–5209; d) N. J. Divins, J. Llorca,
Appl. Catal. A 2016, 518, 60–66; e) D. R. Mullins, Surf. Sci. Rep. 2015, 70,
42–85.
[8] E. Aneggi, C. De Leitenburg, J. Llorca, A. Trovarelli, Catal. Today 2012,
197, 119–126.
[9] J. P. Holgado, G. Munuera, J. P. Espinꢃs, A. R. Gonzꢅlez-Elipe, Appl. Surf.
Sci. 2000, 158, 164–171.
Keywords: ceria catalysts · heterogeneous catalysis · oxygen
storage · photoelectron spectroscopy · soot combustion
[10] E. Aneggi, N. J. Divins, C. De Leitenburg, J. Llorca, A. Trovarelli, J. Catal.
2014, 312, 191–194.
[11] R. Jain, A. Dubey, M. K. Ghosalya, C. S. Gopinath, Catal. Sci. Technol.
2016, 6, 1746–1756.
[12] S. Blomberg, M. J. Hoffmann, J. Gustafson, N. M. Martin, V. R. Fernandes,
A. Borg, Z. Liu, R. Chang, S. Matera, K. Reuter, et al., Phys. Rev. Lett.
2013, 110, 117601.
[13] a) M. Strongin, S. L. Qiu, J. Chen, C. L. Lin, E. M. McCarron, Phys. Rev. B
1990, 41, 7238–7240; b) M. Fu, X. Yue, D. Ye, J. Ouyang, B. Huang, J.
Wu, H. Liang, Catal. Today 2010, 153, 125–132; c) Y. Wei, J. Liu, Z. Zhao,
A. Duan, G. Jiang, C. Xu, J. Gao, H. He, X. Wang, Energy Environ. Sci.
2011, 4, 2959–2970.
[1] a) A. Bueno-Lꢃpez, Appl. Catal. B 2014, 146, 1–11; b) D. Fino, S. Bensaid,
M. Piumetti, N. Russo, Appl. Catal. A 2016, 509, 75–96; c) E. Aneggi, C.
de Leitenburg, A. Trovarelli in Catalysis by Ceria and Related Materials,
2nd Ed. (Eds.: A. Trovarelli, P. Fornasiero), Imperial College Press,
London, 2013, pp. 565–621.
[2] a) E. Aneggi, D. Wiater, C. De Leitenburg, J. Llorca, A. Trovarelli, ACS
Catal. 2014, 4, 172–181; b) E. Aneggi, V. Rico-Perez, C. de Leitenburg, S.
Maschio, L. Soler, J. Llorca, A. Trovarelli, Angew. Chem. Int. Ed. 2015, 54,
14040–14043; Angew. Chem. 2015, 127, 14246–14249; c) B. Bassou, N.
Guilhaume, K. Lombaert, C. Mirodatos, D. Bianchi, Energy Fuels 2010, 24,
4781–4792.
[3] a) A. Bueno-Lꢃpez, K. Krishna, M. Makkee, J. A. Moulijn, J. Catal. 2005,
230, 237–248; b) K. Harada, T. Oishi, S. Hamamoto, T. Ishihara, J. Phys.
Chem. C 2014, 118, 559–568.
[14] a) C. Li, K. Domen, K. Maruya, T. Onishi, J. Am. Chem. Soc. 1989, 111,
7683–7687; b) A. Trovarelli, Catal. Rev. 1996, 38, 439–520; c) X. Zhang,
K. J. Klabunde, Inorg. Chem. 1992, 31, 1706–1709.
[4] a) G. Mul, F. Kapteijn, C. Doornkamp, J. A. Moulijn, J. Catal. 1998, 179,
258–266; b) G. Preda, A. Migani, K. M. Neyman, S. T. Bromley, F. Illas, G.
Pacchioni, J. Phys. Chem. C 2011, 115, 5817–5822; c) J. Kullgren, K. Her-
mansson, P. Broqvist, J. Phys. Chem. Lett. 2013, 4, 604–608.
[15] a) S. Rossignol, F. Gꢁrard, D. Duprez, J. Mater. Chem. 1999, 9, 1615–
1620; b) A. Martꢀnez-Arias, M. Fernꢅndez-Garcꢀa, C. Belver, J. C. Conesa,
J. Soria, Catal. Lett. 2000, 65, 197–204.
[5] a) J. Xu, J. Harmer, G. Li, T. Chapman, P. Collier, S. Longworth, S. C.
Tsang, Chem. Commun. 2010, 46, 1887–1889; b) J. Soria, A. Martinez-
Arias, J. C. Conesa, J. Chem. Soc. Trans. 1995, 91, 1669–1678; c) V. V.
Pushkarev, V. I. Kovalchuk, J. L. D’Itri, J. Phys. Chem. B 2004, 108, 5341–
Received: May 20, 2016
Published online on && &&, 0000
&
ChemCatChem 2016, 8, 1 – 5
www.chemcatchem.org
4
ꢂ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

COMMUNICATIONS
A tale of two routes: Soot abatement
on ceria catalysts involves two coopera-
tive routes: one occurring at the ceria–
soot interface with formation of oxygen
vacancies and CeIII and the other at the
surface of soot mediated by active su-
peroxide species, which result from the
reaction between gas-phase O2 and
oxygen vacancies.
L. Soler, A. Casanovas, C. Escudero,
V. Pꢁrez-Dieste, E. Aneggi, A. Trovarelli,
J. Llorca*
&& – &&
Ambient Pressure Photoemission
Spectroscopy Reveals the Mechanism
of Carbon Soot Oxidation in Ceria-
Based Catalysts
ChemCatChem 2016, 8, 1 – 5
www.chemcatchem.org
5
ꢂ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ