Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy journal homepage: www.elsevier.com/locate/saa
Spectroscopic characterization of Co3O4 catalyst doped with CeO2 and PdO for methane catalytic combustion q P.J. Jodłowski a,⇑, R.J. Je˛drzejczyk b, A. Rogulska b, A. Wach b, P. Kus´trowski b, M. Sitarz c, T. Łojewski a, A. Kołodziej d,e, J. Łojewska b a
Faculty of Chemical Engineering and Technology, Cracow University of Technology, Warszawska 24, 31-155 Kraków, Poland Jagiellonian University, Faculty of Chemistry, Ingardena 3, 30-060 Kraków, Poland Faculty of Materials Science and Ceramics, AGH University of Science and Technology, al. Mickiewicza 30, 30-059 Kraków, Poland d Institute of Chemical Engineering, Polish Academy of Sciences, Bałtycka 5, 44-100 Gliwice, Poland e Faculty of Civil Engineering, Opole University of Technology, Katowicka 48, 45-061 Opole, Poland b c
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
lRaman, XPS and EDX methods used
to analyze catalyst material distribution. Palladium doped cobalt catalyst shows amazingly high activity in methane combustion. Surface of palladium doped catalyst covered with highly dispersed PdO.
a r t i c l e
i n f o
Article history: Received 7 January 2014 Received in revised form 1 May 2014 Accepted 11 May 2014 Available online xxxx Keywords: Methane Combustion Structured catalysts Cobalt oxide Palladium promoted catalysts
a b s t r a c t The study deals with the XPS, Raman and EDX characterization of a series of structured catalysts composed of cobalt oxides promoted by palladium and cerium oxides. The aim of the work was to relate the information gathered from spectroscopic analyses with the ones from kinetic tests of methane combustion to establish the basic structure–activity relationships for the catalysts studied. The most active catalyst was the cobalt oxide doped with little amount of palladium and wins a confrontation with pure palladium oxide catalyst which is commercially used in converters for methane. The analyses Raman and XPS analyses showed that this catalyst is composed of a cobalt spinel and palladium oxide. The quantitative approach to the composition of the catalysts by XPS and EDX methods revealed that the surface of palladium doped cobalt catalyst is enriched with palladium oxide which provides a great number of active centres for methane combustion indicated by kinetic parameters. Ó 2014 Elsevier B.V. All rights reserved.
Introduction
q
Selected paper presented at XIIth International Conference on Molecular spectroscopy, Kraków – Białka Tatrzan´ska, Poland, September 8–12, 2013. ⇑ Corresponding author. Tel.: +48 12 628 27 60. E-mail address: [email protected] (P.J. Jodłowski).
From a wide spectrum of methods which are used in waste gases cleaning, catalytic combustion seems to be the most effective approach especially in the case of low concentration of methane and carbon monoxide. These pollutants appear in several
http://dx.doi.org/10.1016/j.saa.2014.05.027 1386-1425/Ó 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: P.J. Jodłowski et al., Spectroscopic characterization of Co3O4 catalyst doped with CeO2 and PdO for methane catalytic combustion, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy (2014), http://dx.doi.org/10.1016/j.saa.2014.05.027
2
P.J. Jodłowski et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy xxx (2014) xxx–xxx
stationary sources utilizing biogas that can be obtained from biomass or methane as for example from a shale gas or coal mines. It seems that the conditions of the catalytic combustion are less demanding than those of classic homogeneous combustion: undoubtedly catalytic process runs at lower temperature and more flexible air/fuel ratio [1]. For methane combustion, catalysts based on palladium oxide are the most likely used and a lot of attention has been given to such systems in the literature, as described in [2]. Besides
palladium, which is claimed to be the most active in this reaction, a great deal of research has been performed to test different metal oxides. In the literature high catalytic activity of Co3O4, AgO, CuO, MnO, Cu/CeO2, Ag/CeO2, MnOX/ZrO2 was reported as for example in Ref. [3]. Moreover, great activity of mixed metal oxide systems was also reported elsewhere [4]. For example, the catalytic activity of the composite catalysts were studied during the combustion of CH4 (3000 ppm) in a mixture of oxygen and helium. High activity in methane combustion was also fulfilled by using a mixed
Table 1 Catalysts preparation parameters. Catalyst
Number of active agents
Solution used for immersion
Impregnation time (h)
Co1 Co1Pd0.001 Co1Ce1 Pd0.001
1 2 2 1
1 M Co(NO3)2 1 M Co(NO3)2 0.001 M Pd(NO3)2 1 M Co(NO3)2 1 M Ce(NO3)3 0.001 M Pd(NO3)2
1 1 1 1
Table 2 Semi-quantitative results of XPS and EDX analysis. Method
Catalyst
Wt.% Al
At.%
⁄
Ce
Co
Pd
Al⁄
Ce
Co
Pd
XPS
Co1 Co1Pd0.001 Co1Ce1 Pd0.001
69.4 75.4 30.0 11.5
– – 0.0 –
30.6 0.0 70.0 –
– 24.6 – 88.5
83.2 92.4 48.3 33.8
– – 0.0 0.0
16.8 0.00 51.7 0.00
– 7.63 0.00 66.17
EDX
Co1 Co1Pd0.001 Co1Ce1 Pd0.001
49.6 ± 0.3 66.2 ± 0.3 9.9 ± 0.3 96.6 ± 0.3
– – 18.70 ± 0.50 –
50.4 ± 0.5 31.1 ± 0.5 71.4 ± 1.2 –
– 2.7 ± 0.3 – 3.4 ± 0.3
68.21 ± 0.02 81.60 ± 0.03 21.42 ± 0.06 99.12 ± 0.39
– – 7.80 ± 0.02 –
31.79 ± 0.01 17.54 ± 0.02 70.78 ± 0.03 –
– 0.85 ± 0.08 – 0.88 ± 0.10
Fig. 1. Raman spectra of prepared samples.
Please cite this article in press as: P.J. Jodłowski et al., Spectroscopic characterization of Co3O4 catalyst doped with CeO2 and PdO for methane catalytic combustion, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy (2014), http://dx.doi.org/10.1016/j.saa.2014.05.027
P.J. Jodłowski et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy xxx (2014) xxx–xxx
manganese-based catalysts and catalyst systems of Mg–Mn, Ca– Mn, Sr, and Ba–Mn–Mn [5]. For these systems the best activity has been demonstrated by Mg–Mn catalyst, obtained by impregnation method. The application of mixed M–Pd/Al2O3 (where M: Co, Cr, Cu, Fe, Mn, Ni) catalysts for methane oxidation was reported in [6]. In this study, palladium oxide was impregnated onto mixed oxide supports (Al2O3–MOx), with final 1.1 wt.% palladium. During catalytic combustion of methane (1 vol% in air), the best catalytic properties were presented by mixed Ni and Co catalysts. The aim of this study is to show applicability and limitations of spectroscopic methods devoted to surface analyses in search for basic structure–activity correlations of the mixed oxides systems studies in catalytic combustion of methane. A direct motivation here is to answer a question why the palladium doped cobalt catalyst has shown higher activity than its counterpart composed of pure palladium oxide? Experimental Sample preparation A series of cobalt catalysts differing in composition were prepared using impregnation method. In all cases precalcinated kanthal steel (FeCrAl) washcoated with c-Al2O3 was used as a support. The whole procedure of preparing the support can be found in our previous work [7]. In order to deposit catalyst precalcined and whashcoated kanthal sheets were immersed in aqueous solutions of cobalt, palladium and cerium nitrates for 1 h at room
3
temperature. Catalysts were dried at room temperature for 12 h and then calcined at 550 °C for 6 h (with 3 °C/min). Multicomponent catalysts were prepared by subsequent impregnations and calcinations as described above. Thus prepared catalysts are listed in Table 1 (the name represents active agents used for catalyst preparation in reverse order of deposition). Surface analyses X-ray Photoelectron Spectroscopy was performed using the ESCA Prevac spectrometer equipped with a hemispheric analyser of charged particles XPS and AES (VG Scienta R3000), X-ray tube was equipped with two anticathodes Mg/Al (power Mg/Al = 400/ 600 W) and the X-ray monochromator with the radiation source (single anticathode – Al). XPS spectra were calibrated to the carbon component C1s binding (with energy of 285.0 eV). The fitting of high resolution spectra was provided through the CasaXPS software. The composition and chemical surrounding of the sample surface were investigated on the basis of the areas and binding energies of C 1s, O 1s, Al 2p, Pd 3d, Co 2p and Ce 3d photoelectron peaks. The percentage atomic concentrations were calculated by taking into account the Relative Sensitivity Factors (RSF) from the default CasaXPS library. XPS measurements were performed directly without sample preparation, hence the signal acquisition can be associated with the surface of catalysts. Raman analysis was carried out using the spectrometer (JOBIN YVON LABRAM HR), equipped with three lasers, He–Ne (633 nm), Ar (514 nm), and HeCd (325 nm). The spectra of prepared catalysts
Fig. 2. XPS spectra of prepared samples.
Please cite this article in press as: P.J. Jodłowski et al., Spectroscopic characterization of Co3O4 catalyst doped with CeO2 and PdO for methane catalytic combustion, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy (2014), http://dx.doi.org/10.1016/j.saa.2014.05.027
4
P.J. Jodłowski et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy xxx (2014) xxx–xxx
were taken using the 514 nm line, which provided good Raman signal. Prior to the spectra acquisition, the samples were dehydrated in a flow of 50 cm3/min He, at 500 °C for 1 h using the Linkam CCR – 1000 reactor. Microstructure of the produced materials was examined by means of scanning electron microscope (SEM, Nova Nano SEM 200, FEI Company) with an attachment for chemical analysis of specimen in microareas with energy dispersive X-ray spectroscopy (EDX, EDAX). The analyses were carried out in low vacuum condition in secondary electron mode. Samples were covered with carbon layer. The percentage metal concentration were calculated using the RSF values from the default equipment software library. Prior to the SEM/EDX analysis, the catalysts were calcined at 550 °C for 6 h (with ramp 3 °C/min). The results of quantitative analyses of catalyst surface composition by XPS and SEM/EDX methods are presented in Table 2 and expressed as a concentration of metals on the catalyst surface in at.% and wt.%. Kinetic experiments The kinetic parameters, such as Arrhenius parameters and a kinetic equation of methane combustion were determined using a CSTR gradientless reactor. Prior to the catalytic tests, all samples were activated in a 100 cm3/min flow of synthetic air (Air Products) at 500 °C for 1 h. The catalytic experiments were carried out using a reaction mixture containing 4000 ppm of methane, 10% oxygen and He as a balance. In all catalytic experiments, the reaction effluent gases i.e. CH4, CO, CO2, and H2O were analyzed using FTIR photoacoustic gas analyser GASERA PA 101. The gas flow for all types of experiments was 200 cm3/min. The reaction rate and Arrhenius parameters were calculated form a mass balance for an ideal CSTR reactor.
The Arrhenius equation parameters such as pre-exponential factor, k1, and activation energy, Ea, were calculated for the temperature range 600–730 K. The mean reaction rate was calculated from several measurements, after a steady conditions in the reactor were achieved. The Arrhenius parameters were calculated at temperatures 600–730 K, where the linear regression of the fitted function ln(kr) = f(1/T) was represented by the best square correlation coefficients. The methane reaction rate order was obtained varying the methane content from 1500 to 4000 ppm, with constant amount of 10% O2 and He as balance.
Results and discussion The surface structural properties were studied using Raman spectroscopy and X-ray Photoelectron Spectroscopy. The results are presented in Fig. 1. The Raman spectra obtained for a reference sample of palladium catalyst, Pd0.001 (Fig. 1D) revealed two characteristic bands at 419 and 648 cm1 that can be attributed to two vibrational normal modes Eg and B1g of PdO, respectively, however, in comparison with the bands at 431 and 625 cm1 of PdO from Fig. 1D sample at presented in [8] they are shifted. The additional bands at 273 and 330 cm1, can be attributed to secondary Raman scattering or the resonance effects [8]. The PdO characteristic bands also appear on the spectra of bimetallic cobalt–palladium catalyst, Co1Pd0.001 (Fig. 1C). However, the band at 648 cm1 is tangled both with a strong band 500–600 cm1 region, which is attributed to amorphous cobalt oxide [9] and the strong band at 668 cm1 of cobalt spinel. The cobalt oxide spinel has a characteristic vibrational pattern with frequencies at 476, 516, 613 and 685 cm1 (Fig. 2, column 4), which according to Raman study on Co3O4 monocrystal by Hadjiev et al. [10] can be assigned to the
(A) Pd0.001
(B) Co1Pd0.001
(C) Co1Ce1
(D) Co1
Fig. 3. SEM micrographs of prepared samples.
Please cite this article in press as: P.J. Jodłowski et al., Spectroscopic characterization of Co3O4 catalyst doped with CeO2 and PdO for methane catalytic combustion, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy (2014), http://dx.doi.org/10.1016/j.saa.2014.05.027
P.J. Jodłowski et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy xxx (2014) xxx–xxx
Eg, F2g, F2g and A1g symmetry, respectively. This assignment has been confirmed in Refs. [9,11]. Thus the characteristic bands of cobalt spinel oxide can also be found on the spectra of Co1 and Co1Ce1 catalysts (Fig. 1A and B). The Raman spectra of the mixed cobalt–cerium catalyst revealed the band at 456 cm1. This band is typical of F2g vibrational mode of CeO2 phase as previously reported in Ref. [12]. All spectra indicate also the presence of aAl2O3 with Eg and A1g stretching modes at 384 and 418 cm1, respectively, which are clearly seen on the spectra of Co1 and Co1Pd0.001 samples (Fig. 1A and B). These have also been found in Ref. [13] where metal oxides such as WO3, MoO3, NiO and CoO upon a V2O5/Al2O3 samples were analyzed. As c-Al2O3 does not give Raman transitions signal coming from a-Al2O3 should come from alumina that is formed during the kanthal steel calcination. This observation confirms that Raman spectroscopy is not fully a surface sensitive method and also provides informations from subsurface regions as in the discussed case of multicomponent supported catalysts. The XPS spectra of palladium containing catalysts (Pd0.001 and Co1Pd0.001) has also proved the presence of the palladium oxide on the catalyst surface. The characteristic bands of Pd 3d5/2 at 334.5 eV and Pd 3d3/2 at 342.2 eV have been assigned to PdO according to Ref. [14]. An important message to be sent here is that the cobalt oxide was not detected by XPS on Co1Pd0.001 sample which seems contradictory to the Raman analyses results. However, taking into account the fact that XPS electrons come from few monolayers from the underneath of the sample it can be inferred that the palladium oxide is located on the top layer of the catalyst. This difference in both analytical methods can be used to gather complimentary information from the material layer deposit on the metallic support. The XPS spectra of cobalt catalysts (Co1 and Co1Ce1) shows the characteristic bands of the spinel type cobalt oxide at 780 eV and at 795 eV which has been attributed to Co 2p3/2, and Co 2p1/2 core levels [15]. The asymmetry which can be noted for the Co 2p3/2 and Co 2p1/2 bands and revealed by their fitting presented in Fig. 2, can be a results of the shifts of the binding energy of Co3+ and Co2+ in octahedral and tetrahedral voids [15,16]. The appearance of the band at 795 eV may also suggest the presence of CoO at the catalyst surface. It is due to the fact, that at ultra-high vacuum conditions the spinel cobalt oxide may decompose to CoO [17]. The presence of the band at 795 eV can also be accounted for by strong alumina–cobalt interactions favouring the formation of CoAl2O4 [18]. However, the presence of CoAl2O4 was not confirmed by the Raman spectroscopy. The XPS analysis of Co1Ce1 did not detect cerium oxide on the catalysts surface which indicates that cerium oxide was located under the spinel cobalt oxide layer. The surface topography of prepared structured catalysts was examined using the SEM microscopy. The pictures of the catalysts surfaces are presented in Fig. 3. Those results has revealed significant differences between the catalysts surfaces after the impregnation. In all cases, except Co1Ce1, the support surface is fully covered by c-Al2O3 with an average grain size 2–10 lm. The sample promoted with cerium has revealed the sponge-like topography, where it was difficult to distinguish the support pores. Moreover, after the cerium nitrate chemisorption, the catalysts surface looked extremely smooth, as if it was laminated. The impregnation with other metals did not cause such an effect. The XPS and EDX spectroscopies were also applied for quantitative analysis of the surface composition of the prepared samples. The results are shown in Table 2. A comparison of the results collected by both methods shows the distribution of the catalyst material in the surface layers. It can be clearly seen for the single component samples Co1 which spotted with XPS gave lower relative concentration of Co than EDX. This indicates that the surface is
5
still enriched with alumina. In the case of Pd0.001 the situation is different and both results suggests that the surface contains more palladium(II) oxide than alumina. Since cerium oxide gave no signal in the XPS analysis of Co1Ce1 catalyst cerium must be located under the surface layer of cobalt spinel. In the EDX analysis which is able to reach deeper than XPS cerium was detected. His sample surface seems to be enriched with cobalt oxide as referred to pure Co1 sample. In the case of palladium promoted cobalt catalyst (Co1Pd0.001), the results suggests that the cobalt oxide located at the surface was fully covered by palladium oxide which creates a very thin layer as palladium contribution is very low in the EDX view. The differences between the metal oxide (Al2O3, Co3O4, PdO and CeO2) distribution on the catalysts surface does not only seem to be a result of the impregnation sequence of the active material on the structured support. This can be also connected to the physicochemical differences of the metal oxides. It is well known that the noble metals have higher volatility than the non-noble metals,
Fig. 4. Kinetic results of methane catalytic combustion over various structured catalysts (A) reaction rate normalized to geometric area of reactor internals (further used in reactor modeling), (B) Arrhenius plot derived from kinetic results presented in (A) and (C) concentration dependence on the reaction rate expressed in linearized coordinates.
Please cite this article in press as: P.J. Jodłowski et al., Spectroscopic characterization of Co3O4 catalyst doped with CeO2 and PdO for methane catalytic combustion, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy (2014), http://dx.doi.org/10.1016/j.saa.2014.05.027
6
P.J. Jodłowski et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy xxx (2014) xxx–xxx
Table 3 Arrhenius parameters and reaction order results of methane catalytic combustion over various structured catalysts. Sample
k1 (m s1)
Eapp (kJ mol1)
R2
Reaction order
R2
Co1Pd0.001 Pd0.001 Co1Ce1
136.46 1.61 25.93
60.06 42.49 63.90
0.993 0.949 0.972
0.98 1.00 0.97
0.999 0.999 0.999
which in fact can cause a migration of the noble-metals to the catalyst surface [19]. This can be partly avoided by the addition of cerium oxide, which is commonly used to improve the catalysts thermal stability [4]. The results of kinetic experiments are presented in Fig. 4. Fig. 4A and B shows the light-off curves represented by the rate of the reaction vs. temperature. To compare the intrinsic catalyst activity, the reaction rate was normalized to the catalyst geometrical area. That kind of approach is commonly used in the structured reactors description [20]. The best catalytic properties were represented by palladium promoted cobalt catalyst, Co1Pd0.001 (Fig. 4A) even in comparison with pure palladium catalyst which is used in commercial catalytic converters. The worst catalytic properties in methane combustion were demonstrated by mixed cobalt–cerium catalyst. The results for cobalt catalyst, Co1, were not presented in Fig. 3, because the catalytic activity during the methane combustion was negligibly low. The catalytic activity of prepared structured catalysts may also be described by comparing the Arrhenius equation parameters. These are presented in Fig 4B and Table 3. When comparing the activation energy, the activity sequence of the catalysts shows the descending order Pd0.001 > Co1Pd0.001 > Co1Ce1 > Co1, whereas when comparing the values of the pre-exponential constant from Arrhenius equation, the activity changes in the following order: Co1Pd0.001 Co1Ce > Pd0.001. This indicates that the number of collisions is the highest for the Co1Pd0.001 probably due to the highest number of active centres available for methane activation. The results of the determination of the reaction order of methane catalytic combustion are presented in Table 3 and Fig. 4C. The values of the reaction order were obtained using the linear regression. In all cases the value of the slope of the line in Fig. 4C was close to the unity. The statistical error was represented by R2. The fitting statistic represented by R2 is close to unity, which proves excellent linear fitting and low experimental errors. Conclusions The aim of this paper was to obtain and describe the structure of mono- and multi-oxide catalysts deposited on metallic supports and tested for methane catalytic combustion and to provide fundamental structure–activity relationships. Methane combustion catalytic tests has revealed higher catalytic properties of palladium promoted cobalt catalyst in comparison with pure palladium catalyst that is commercially used in catalytic converters. This can be due to lower activation energy of palladium doped cobalt catalyst and much higher pre-exponential factor than for the reference palladium catalysts. The latter can be accounted for by the larger number of active sites available for the methane on the palladium doped cobalt catalyst. The structural
analysis of the catalysts by Raman and XPS spectroscopies revealed that an active form of cobalt in the catalyst is cobalt spinel and palladium present in its oxidized PdO form. Pure cobalt oxide is, however, inactive in methane combustion. This together with the fact that on ceria doped cobalt catalysts the XPS has shown only cobalt oxide form can account for its lowest activity in methane combustion. The quantitative analysis of the catalysts by XPS and EDX methods differing in the depth of the signal penetration of the samples, showed differences in surface distributions of the active metal oxides. For both palladium containing samples: pure and palladium doped cobalt, the surface is enriched with palladium. This is consistent with the high value of pre-exponential factor of the palladium doped cobalt catalysts and can be related to the higher melting point of palladium oxide than cobalt oxide. The nature of the amazing activity enhancement of the bimetal oxide system is still unknown and calls for further investigation. The methods spectroscopic methods used in this study (XPS, Raman, EDX) were able to differentiate between the very surface region to which XPS is sensitive and subsurface region typical of Raman analyses. Comparison of quantitative results of surface composition by XPS and EDX was used to show the differences in active material distribution within its layers deposited on the metal supports. Acknowledgements Financial support for this work was provided by the Polish National Science Centre – Project No. 2011/03/N/ST8/05129 and also by BRIDGE Programme Grant (No. 2010-1/4) from the Foundation for Polish Science, co-financed by the EU. The authors are grateful to SPB 811/N-COST/2010/0 for sponsoring optical parts to Raman system. We are also thankful for the START scholarship within the Foundation for Polish Science. References [1] R. Litto, R.E. Hayes, H. Sapoundjiev, A. Fuxman, F. Forbes, B. Liu, et al., Catal. Today 117 (2006) 536–542. [2] J. Lee, D. Trimm, Fuel Process. Technol. 42 (1995) 339–359. [3] T. Choudhary, S. Banerjee, V. Choudhary, Appl. Catal. A Gen. 234 (2002) 1–23. [4] L. Liotta, G. Dicarlo, G. Pantaleo, G. Deganello, Catal. Commun. 6 (2005) 329– 336. [5] X. Wang, Y. Xie, React. Kinet. Catal. Lett. 71 (2000) 263–271. [6] H. Widjaja, K. Sekizawa, K. Eguchi, H. Arai, Catal. Today 47 (1999) 95–101. [7] P.J. Jodłowski, J. Kryca, M. Iwaniszyn, R. Je˛drzejczyk, J. Thomas, a. Kołodziej, et al., Catal. Today 216 (2013) 276–282. [8] O. Demoulin, M. Navez, E.M. Gaigneaux, P. Ruiz, A.-S. Mamede, P. Granger, et al., Phys. Chem. Chem. Phys. 5 (2003) 4394–4401. [9] M.A. Vuurman, D.J. Stufkens, A. Oskam, O. Goutam, I.E. Wachs, J. Chem. Soc. Faraday Trans. 92 (1996) 3259–3265. [10] V.G. Hadjiev, M.N. Iliev, I.V. Vergilov, J. Phys. Chem. C Solid State Phys. 21 (1988) L199–L202. [11] J. Tyczkowski, R. Kapica, J. Łojewska, Thin Solid Films 515 (2007) 6590–6595. [12] Z. Cmjak, B. Orel, Sol. Energy Mater. Sol. Cells 40 (1996) 205–219. [13] S.P.S. Porto, R.S. Krishnan, J. Chem. Phys. 47 (1967) 1009–1012. [14] J.M. Tura, P. Regull, L. Surf. Interface Anal. 11 (1988) 447–449. [15] S.C. Petitto, E.M. Marsh, G.a. Carson, M.a. Langell, J. Mol. Catal. A Chem. 281 (2008) 49–58. [16] R. Berenguer, A. La Rosa-Toro, C. Quijada, E. Morallo, J. Phys. Chem. C 112 (2008) 16945–16952. [17] A. Jnioui, M. Eddouasse, A. Amariglio, J.J. Ehrhardt, M. Alnot, J. Lambert, et al., J. Catal. 106 (1987) 144–165. [18] M. Oku, K. Hirokawa, J. Electron Spectros. Relat. Phenom. 8 (1976) 475–481. [19] J.H. Lee, D.L. Trimm, Fuel Process. Technol. 42 (1995) 339–359. [20] A. Cybulski, J.A. Moulijn, Structured Catalysts and Reactors, Taylor & Francis, 2006.
Please cite this article in press as: P.J. Jodłowski et al., Spectroscopic characterization of Co3O4 catalyst doped with CeO2 and PdO for methane catalytic combustion, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy (2014), http://dx.doi.org/10.1016/j.saa.2014.05.027
Contents lists available at ScienceDirect
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy journal homepage: www.elsevier.com/locate/saa
Spectroscopic characterization of Co3O4 catalyst doped with CeO2 and PdO for methane catalytic combustion q P.J. Jodłowski a,⇑, R.J. Je˛drzejczyk b, A. Rogulska b, A. Wach b, P. Kus´trowski b, M. Sitarz c, T. Łojewski a, A. Kołodziej d,e, J. Łojewska b a
Faculty of Chemical Engineering and Technology, Cracow University of Technology, Warszawska 24, 31-155 Kraków, Poland Jagiellonian University, Faculty of Chemistry, Ingardena 3, 30-060 Kraków, Poland Faculty of Materials Science and Ceramics, AGH University of Science and Technology, al. Mickiewicza 30, 30-059 Kraków, Poland d Institute of Chemical Engineering, Polish Academy of Sciences, Bałtycka 5, 44-100 Gliwice, Poland e Faculty of Civil Engineering, Opole University of Technology, Katowicka 48, 45-061 Opole, Poland b c
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
lRaman, XPS and EDX methods used
to analyze catalyst material distribution. Palladium doped cobalt catalyst shows amazingly high activity in methane combustion. Surface of palladium doped catalyst covered with highly dispersed PdO.
a r t i c l e
i n f o
Article history: Received 7 January 2014 Received in revised form 1 May 2014 Accepted 11 May 2014 Available online xxxx Keywords: Methane Combustion Structured catalysts Cobalt oxide Palladium promoted catalysts
a b s t r a c t The study deals with the XPS, Raman and EDX characterization of a series of structured catalysts composed of cobalt oxides promoted by palladium and cerium oxides. The aim of the work was to relate the information gathered from spectroscopic analyses with the ones from kinetic tests of methane combustion to establish the basic structure–activity relationships for the catalysts studied. The most active catalyst was the cobalt oxide doped with little amount of palladium and wins a confrontation with pure palladium oxide catalyst which is commercially used in converters for methane. The analyses Raman and XPS analyses showed that this catalyst is composed of a cobalt spinel and palladium oxide. The quantitative approach to the composition of the catalysts by XPS and EDX methods revealed that the surface of palladium doped cobalt catalyst is enriched with palladium oxide which provides a great number of active centres for methane combustion indicated by kinetic parameters. Ó 2014 Elsevier B.V. All rights reserved.
Introduction
q
Selected paper presented at XIIth International Conference on Molecular spectroscopy, Kraków – Białka Tatrzan´ska, Poland, September 8–12, 2013. ⇑ Corresponding author. Tel.: +48 12 628 27 60. E-mail address: [email protected] (P.J. Jodłowski).
From a wide spectrum of methods which are used in waste gases cleaning, catalytic combustion seems to be the most effective approach especially in the case of low concentration of methane and carbon monoxide. These pollutants appear in several
http://dx.doi.org/10.1016/j.saa.2014.05.027 1386-1425/Ó 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: P.J. Jodłowski et al., Spectroscopic characterization of Co3O4 catalyst doped with CeO2 and PdO for methane catalytic combustion, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy (2014), http://dx.doi.org/10.1016/j.saa.2014.05.027
2
P.J. Jodłowski et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy xxx (2014) xxx–xxx
stationary sources utilizing biogas that can be obtained from biomass or methane as for example from a shale gas or coal mines. It seems that the conditions of the catalytic combustion are less demanding than those of classic homogeneous combustion: undoubtedly catalytic process runs at lower temperature and more flexible air/fuel ratio [1]. For methane combustion, catalysts based on palladium oxide are the most likely used and a lot of attention has been given to such systems in the literature, as described in [2]. Besides
palladium, which is claimed to be the most active in this reaction, a great deal of research has been performed to test different metal oxides. In the literature high catalytic activity of Co3O4, AgO, CuO, MnO, Cu/CeO2, Ag/CeO2, MnOX/ZrO2 was reported as for example in Ref. [3]. Moreover, great activity of mixed metal oxide systems was also reported elsewhere [4]. For example, the catalytic activity of the composite catalysts were studied during the combustion of CH4 (3000 ppm) in a mixture of oxygen and helium. High activity in methane combustion was also fulfilled by using a mixed
Table 1 Catalysts preparation parameters. Catalyst
Number of active agents
Solution used for immersion
Impregnation time (h)
Co1 Co1Pd0.001 Co1Ce1 Pd0.001
1 2 2 1
1 M Co(NO3)2 1 M Co(NO3)2 0.001 M Pd(NO3)2 1 M Co(NO3)2 1 M Ce(NO3)3 0.001 M Pd(NO3)2
1 1 1 1
Table 2 Semi-quantitative results of XPS and EDX analysis. Method
Catalyst
Wt.% Al
At.%
⁄
Ce
Co
Pd
Al⁄
Ce
Co
Pd
XPS
Co1 Co1Pd0.001 Co1Ce1 Pd0.001
69.4 75.4 30.0 11.5
– – 0.0 –
30.6 0.0 70.0 –
– 24.6 – 88.5
83.2 92.4 48.3 33.8
– – 0.0 0.0
16.8 0.00 51.7 0.00
– 7.63 0.00 66.17
EDX
Co1 Co1Pd0.001 Co1Ce1 Pd0.001
49.6 ± 0.3 66.2 ± 0.3 9.9 ± 0.3 96.6 ± 0.3
– – 18.70 ± 0.50 –
50.4 ± 0.5 31.1 ± 0.5 71.4 ± 1.2 –
– 2.7 ± 0.3 – 3.4 ± 0.3
68.21 ± 0.02 81.60 ± 0.03 21.42 ± 0.06 99.12 ± 0.39
– – 7.80 ± 0.02 –
31.79 ± 0.01 17.54 ± 0.02 70.78 ± 0.03 –
– 0.85 ± 0.08 – 0.88 ± 0.10
Fig. 1. Raman spectra of prepared samples.
Please cite this article in press as: P.J. Jodłowski et al., Spectroscopic characterization of Co3O4 catalyst doped with CeO2 and PdO for methane catalytic combustion, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy (2014), http://dx.doi.org/10.1016/j.saa.2014.05.027
P.J. Jodłowski et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy xxx (2014) xxx–xxx
manganese-based catalysts and catalyst systems of Mg–Mn, Ca– Mn, Sr, and Ba–Mn–Mn [5]. For these systems the best activity has been demonstrated by Mg–Mn catalyst, obtained by impregnation method. The application of mixed M–Pd/Al2O3 (where M: Co, Cr, Cu, Fe, Mn, Ni) catalysts for methane oxidation was reported in [6]. In this study, palladium oxide was impregnated onto mixed oxide supports (Al2O3–MOx), with final 1.1 wt.% palladium. During catalytic combustion of methane (1 vol% in air), the best catalytic properties were presented by mixed Ni and Co catalysts. The aim of this study is to show applicability and limitations of spectroscopic methods devoted to surface analyses in search for basic structure–activity correlations of the mixed oxides systems studies in catalytic combustion of methane. A direct motivation here is to answer a question why the palladium doped cobalt catalyst has shown higher activity than its counterpart composed of pure palladium oxide? Experimental Sample preparation A series of cobalt catalysts differing in composition were prepared using impregnation method. In all cases precalcinated kanthal steel (FeCrAl) washcoated with c-Al2O3 was used as a support. The whole procedure of preparing the support can be found in our previous work [7]. In order to deposit catalyst precalcined and whashcoated kanthal sheets were immersed in aqueous solutions of cobalt, palladium and cerium nitrates for 1 h at room
3
temperature. Catalysts were dried at room temperature for 12 h and then calcined at 550 °C for 6 h (with 3 °C/min). Multicomponent catalysts were prepared by subsequent impregnations and calcinations as described above. Thus prepared catalysts are listed in Table 1 (the name represents active agents used for catalyst preparation in reverse order of deposition). Surface analyses X-ray Photoelectron Spectroscopy was performed using the ESCA Prevac spectrometer equipped with a hemispheric analyser of charged particles XPS and AES (VG Scienta R3000), X-ray tube was equipped with two anticathodes Mg/Al (power Mg/Al = 400/ 600 W) and the X-ray monochromator with the radiation source (single anticathode – Al). XPS spectra were calibrated to the carbon component C1s binding (with energy of 285.0 eV). The fitting of high resolution spectra was provided through the CasaXPS software. The composition and chemical surrounding of the sample surface were investigated on the basis of the areas and binding energies of C 1s, O 1s, Al 2p, Pd 3d, Co 2p and Ce 3d photoelectron peaks. The percentage atomic concentrations were calculated by taking into account the Relative Sensitivity Factors (RSF) from the default CasaXPS library. XPS measurements were performed directly without sample preparation, hence the signal acquisition can be associated with the surface of catalysts. Raman analysis was carried out using the spectrometer (JOBIN YVON LABRAM HR), equipped with three lasers, He–Ne (633 nm), Ar (514 nm), and HeCd (325 nm). The spectra of prepared catalysts
Fig. 2. XPS spectra of prepared samples.
Please cite this article in press as: P.J. Jodłowski et al., Spectroscopic characterization of Co3O4 catalyst doped with CeO2 and PdO for methane catalytic combustion, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy (2014), http://dx.doi.org/10.1016/j.saa.2014.05.027
4
P.J. Jodłowski et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy xxx (2014) xxx–xxx
were taken using the 514 nm line, which provided good Raman signal. Prior to the spectra acquisition, the samples were dehydrated in a flow of 50 cm3/min He, at 500 °C for 1 h using the Linkam CCR – 1000 reactor. Microstructure of the produced materials was examined by means of scanning electron microscope (SEM, Nova Nano SEM 200, FEI Company) with an attachment for chemical analysis of specimen in microareas with energy dispersive X-ray spectroscopy (EDX, EDAX). The analyses were carried out in low vacuum condition in secondary electron mode. Samples were covered with carbon layer. The percentage metal concentration were calculated using the RSF values from the default equipment software library. Prior to the SEM/EDX analysis, the catalysts were calcined at 550 °C for 6 h (with ramp 3 °C/min). The results of quantitative analyses of catalyst surface composition by XPS and SEM/EDX methods are presented in Table 2 and expressed as a concentration of metals on the catalyst surface in at.% and wt.%. Kinetic experiments The kinetic parameters, such as Arrhenius parameters and a kinetic equation of methane combustion were determined using a CSTR gradientless reactor. Prior to the catalytic tests, all samples were activated in a 100 cm3/min flow of synthetic air (Air Products) at 500 °C for 1 h. The catalytic experiments were carried out using a reaction mixture containing 4000 ppm of methane, 10% oxygen and He as a balance. In all catalytic experiments, the reaction effluent gases i.e. CH4, CO, CO2, and H2O were analyzed using FTIR photoacoustic gas analyser GASERA PA 101. The gas flow for all types of experiments was 200 cm3/min. The reaction rate and Arrhenius parameters were calculated form a mass balance for an ideal CSTR reactor.
The Arrhenius equation parameters such as pre-exponential factor, k1, and activation energy, Ea, were calculated for the temperature range 600–730 K. The mean reaction rate was calculated from several measurements, after a steady conditions in the reactor were achieved. The Arrhenius parameters were calculated at temperatures 600–730 K, where the linear regression of the fitted function ln(kr) = f(1/T) was represented by the best square correlation coefficients. The methane reaction rate order was obtained varying the methane content from 1500 to 4000 ppm, with constant amount of 10% O2 and He as balance.
Results and discussion The surface structural properties were studied using Raman spectroscopy and X-ray Photoelectron Spectroscopy. The results are presented in Fig. 1. The Raman spectra obtained for a reference sample of palladium catalyst, Pd0.001 (Fig. 1D) revealed two characteristic bands at 419 and 648 cm1 that can be attributed to two vibrational normal modes Eg and B1g of PdO, respectively, however, in comparison with the bands at 431 and 625 cm1 of PdO from Fig. 1D sample at presented in [8] they are shifted. The additional bands at 273 and 330 cm1, can be attributed to secondary Raman scattering or the resonance effects [8]. The PdO characteristic bands also appear on the spectra of bimetallic cobalt–palladium catalyst, Co1Pd0.001 (Fig. 1C). However, the band at 648 cm1 is tangled both with a strong band 500–600 cm1 region, which is attributed to amorphous cobalt oxide [9] and the strong band at 668 cm1 of cobalt spinel. The cobalt oxide spinel has a characteristic vibrational pattern with frequencies at 476, 516, 613 and 685 cm1 (Fig. 2, column 4), which according to Raman study on Co3O4 monocrystal by Hadjiev et al. [10] can be assigned to the
(A) Pd0.001
(B) Co1Pd0.001
(C) Co1Ce1
(D) Co1
Fig. 3. SEM micrographs of prepared samples.
Please cite this article in press as: P.J. Jodłowski et al., Spectroscopic characterization of Co3O4 catalyst doped with CeO2 and PdO for methane catalytic combustion, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy (2014), http://dx.doi.org/10.1016/j.saa.2014.05.027
P.J. Jodłowski et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy xxx (2014) xxx–xxx
Eg, F2g, F2g and A1g symmetry, respectively. This assignment has been confirmed in Refs. [9,11]. Thus the characteristic bands of cobalt spinel oxide can also be found on the spectra of Co1 and Co1Ce1 catalysts (Fig. 1A and B). The Raman spectra of the mixed cobalt–cerium catalyst revealed the band at 456 cm1. This band is typical of F2g vibrational mode of CeO2 phase as previously reported in Ref. [12]. All spectra indicate also the presence of aAl2O3 with Eg and A1g stretching modes at 384 and 418 cm1, respectively, which are clearly seen on the spectra of Co1 and Co1Pd0.001 samples (Fig. 1A and B). These have also been found in Ref. [13] where metal oxides such as WO3, MoO3, NiO and CoO upon a V2O5/Al2O3 samples were analyzed. As c-Al2O3 does not give Raman transitions signal coming from a-Al2O3 should come from alumina that is formed during the kanthal steel calcination. This observation confirms that Raman spectroscopy is not fully a surface sensitive method and also provides informations from subsurface regions as in the discussed case of multicomponent supported catalysts. The XPS spectra of palladium containing catalysts (Pd0.001 and Co1Pd0.001) has also proved the presence of the palladium oxide on the catalyst surface. The characteristic bands of Pd 3d5/2 at 334.5 eV and Pd 3d3/2 at 342.2 eV have been assigned to PdO according to Ref. [14]. An important message to be sent here is that the cobalt oxide was not detected by XPS on Co1Pd0.001 sample which seems contradictory to the Raman analyses results. However, taking into account the fact that XPS electrons come from few monolayers from the underneath of the sample it can be inferred that the palladium oxide is located on the top layer of the catalyst. This difference in both analytical methods can be used to gather complimentary information from the material layer deposit on the metallic support. The XPS spectra of cobalt catalysts (Co1 and Co1Ce1) shows the characteristic bands of the spinel type cobalt oxide at 780 eV and at 795 eV which has been attributed to Co 2p3/2, and Co 2p1/2 core levels [15]. The asymmetry which can be noted for the Co 2p3/2 and Co 2p1/2 bands and revealed by their fitting presented in Fig. 2, can be a results of the shifts of the binding energy of Co3+ and Co2+ in octahedral and tetrahedral voids [15,16]. The appearance of the band at 795 eV may also suggest the presence of CoO at the catalyst surface. It is due to the fact, that at ultra-high vacuum conditions the spinel cobalt oxide may decompose to CoO [17]. The presence of the band at 795 eV can also be accounted for by strong alumina–cobalt interactions favouring the formation of CoAl2O4 [18]. However, the presence of CoAl2O4 was not confirmed by the Raman spectroscopy. The XPS analysis of Co1Ce1 did not detect cerium oxide on the catalysts surface which indicates that cerium oxide was located under the spinel cobalt oxide layer. The surface topography of prepared structured catalysts was examined using the SEM microscopy. The pictures of the catalysts surfaces are presented in Fig. 3. Those results has revealed significant differences between the catalysts surfaces after the impregnation. In all cases, except Co1Ce1, the support surface is fully covered by c-Al2O3 with an average grain size 2–10 lm. The sample promoted with cerium has revealed the sponge-like topography, where it was difficult to distinguish the support pores. Moreover, after the cerium nitrate chemisorption, the catalysts surface looked extremely smooth, as if it was laminated. The impregnation with other metals did not cause such an effect. The XPS and EDX spectroscopies were also applied for quantitative analysis of the surface composition of the prepared samples. The results are shown in Table 2. A comparison of the results collected by both methods shows the distribution of the catalyst material in the surface layers. It can be clearly seen for the single component samples Co1 which spotted with XPS gave lower relative concentration of Co than EDX. This indicates that the surface is
5
still enriched with alumina. In the case of Pd0.001 the situation is different and both results suggests that the surface contains more palladium(II) oxide than alumina. Since cerium oxide gave no signal in the XPS analysis of Co1Ce1 catalyst cerium must be located under the surface layer of cobalt spinel. In the EDX analysis which is able to reach deeper than XPS cerium was detected. His sample surface seems to be enriched with cobalt oxide as referred to pure Co1 sample. In the case of palladium promoted cobalt catalyst (Co1Pd0.001), the results suggests that the cobalt oxide located at the surface was fully covered by palladium oxide which creates a very thin layer as palladium contribution is very low in the EDX view. The differences between the metal oxide (Al2O3, Co3O4, PdO and CeO2) distribution on the catalysts surface does not only seem to be a result of the impregnation sequence of the active material on the structured support. This can be also connected to the physicochemical differences of the metal oxides. It is well known that the noble metals have higher volatility than the non-noble metals,
Fig. 4. Kinetic results of methane catalytic combustion over various structured catalysts (A) reaction rate normalized to geometric area of reactor internals (further used in reactor modeling), (B) Arrhenius plot derived from kinetic results presented in (A) and (C) concentration dependence on the reaction rate expressed in linearized coordinates.
Please cite this article in press as: P.J. Jodłowski et al., Spectroscopic characterization of Co3O4 catalyst doped with CeO2 and PdO for methane catalytic combustion, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy (2014), http://dx.doi.org/10.1016/j.saa.2014.05.027
6
P.J. Jodłowski et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy xxx (2014) xxx–xxx
Table 3 Arrhenius parameters and reaction order results of methane catalytic combustion over various structured catalysts. Sample
k1 (m s1)
Eapp (kJ mol1)
R2
Reaction order
R2
Co1Pd0.001 Pd0.001 Co1Ce1
136.46 1.61 25.93
60.06 42.49 63.90
0.993 0.949 0.972
0.98 1.00 0.97
0.999 0.999 0.999
which in fact can cause a migration of the noble-metals to the catalyst surface [19]. This can be partly avoided by the addition of cerium oxide, which is commonly used to improve the catalysts thermal stability [4]. The results of kinetic experiments are presented in Fig. 4. Fig. 4A and B shows the light-off curves represented by the rate of the reaction vs. temperature. To compare the intrinsic catalyst activity, the reaction rate was normalized to the catalyst geometrical area. That kind of approach is commonly used in the structured reactors description [20]. The best catalytic properties were represented by palladium promoted cobalt catalyst, Co1Pd0.001 (Fig. 4A) even in comparison with pure palladium catalyst which is used in commercial catalytic converters. The worst catalytic properties in methane combustion were demonstrated by mixed cobalt–cerium catalyst. The results for cobalt catalyst, Co1, were not presented in Fig. 3, because the catalytic activity during the methane combustion was negligibly low. The catalytic activity of prepared structured catalysts may also be described by comparing the Arrhenius equation parameters. These are presented in Fig 4B and Table 3. When comparing the activation energy, the activity sequence of the catalysts shows the descending order Pd0.001 > Co1Pd0.001 > Co1Ce1 > Co1, whereas when comparing the values of the pre-exponential constant from Arrhenius equation, the activity changes in the following order: Co1Pd0.001 Co1Ce > Pd0.001. This indicates that the number of collisions is the highest for the Co1Pd0.001 probably due to the highest number of active centres available for methane activation. The results of the determination of the reaction order of methane catalytic combustion are presented in Table 3 and Fig. 4C. The values of the reaction order were obtained using the linear regression. In all cases the value of the slope of the line in Fig. 4C was close to the unity. The statistical error was represented by R2. The fitting statistic represented by R2 is close to unity, which proves excellent linear fitting and low experimental errors. Conclusions The aim of this paper was to obtain and describe the structure of mono- and multi-oxide catalysts deposited on metallic supports and tested for methane catalytic combustion and to provide fundamental structure–activity relationships. Methane combustion catalytic tests has revealed higher catalytic properties of palladium promoted cobalt catalyst in comparison with pure palladium catalyst that is commercially used in catalytic converters. This can be due to lower activation energy of palladium doped cobalt catalyst and much higher pre-exponential factor than for the reference palladium catalysts. The latter can be accounted for by the larger number of active sites available for the methane on the palladium doped cobalt catalyst. The structural
analysis of the catalysts by Raman and XPS spectroscopies revealed that an active form of cobalt in the catalyst is cobalt spinel and palladium present in its oxidized PdO form. Pure cobalt oxide is, however, inactive in methane combustion. This together with the fact that on ceria doped cobalt catalysts the XPS has shown only cobalt oxide form can account for its lowest activity in methane combustion. The quantitative analysis of the catalysts by XPS and EDX methods differing in the depth of the signal penetration of the samples, showed differences in surface distributions of the active metal oxides. For both palladium containing samples: pure and palladium doped cobalt, the surface is enriched with palladium. This is consistent with the high value of pre-exponential factor of the palladium doped cobalt catalysts and can be related to the higher melting point of palladium oxide than cobalt oxide. The nature of the amazing activity enhancement of the bimetal oxide system is still unknown and calls for further investigation. The methods spectroscopic methods used in this study (XPS, Raman, EDX) were able to differentiate between the very surface region to which XPS is sensitive and subsurface region typical of Raman analyses. Comparison of quantitative results of surface composition by XPS and EDX was used to show the differences in active material distribution within its layers deposited on the metal supports. Acknowledgements Financial support for this work was provided by the Polish National Science Centre – Project No. 2011/03/N/ST8/05129 and also by BRIDGE Programme Grant (No. 2010-1/4) from the Foundation for Polish Science, co-financed by the EU. The authors are grateful to SPB 811/N-COST/2010/0 for sponsoring optical parts to Raman system. We are also thankful for the START scholarship within the Foundation for Polish Science. References [1] R. Litto, R.E. Hayes, H. Sapoundjiev, A. Fuxman, F. Forbes, B. Liu, et al., Catal. Today 117 (2006) 536–542. [2] J. Lee, D. Trimm, Fuel Process. Technol. 42 (1995) 339–359. [3] T. Choudhary, S. Banerjee, V. Choudhary, Appl. Catal. A Gen. 234 (2002) 1–23. [4] L. Liotta, G. Dicarlo, G. Pantaleo, G. Deganello, Catal. Commun. 6 (2005) 329– 336. [5] X. Wang, Y. Xie, React. Kinet. Catal. Lett. 71 (2000) 263–271. [6] H. Widjaja, K. Sekizawa, K. Eguchi, H. Arai, Catal. Today 47 (1999) 95–101. [7] P.J. Jodłowski, J. Kryca, M. Iwaniszyn, R. Je˛drzejczyk, J. Thomas, a. Kołodziej, et al., Catal. Today 216 (2013) 276–282. [8] O. Demoulin, M. Navez, E.M. Gaigneaux, P. Ruiz, A.-S. Mamede, P. Granger, et al., Phys. Chem. Chem. Phys. 5 (2003) 4394–4401. [9] M.A. Vuurman, D.J. Stufkens, A. Oskam, O. Goutam, I.E. Wachs, J. Chem. Soc. Faraday Trans. 92 (1996) 3259–3265. [10] V.G. Hadjiev, M.N. Iliev, I.V. Vergilov, J. Phys. Chem. C Solid State Phys. 21 (1988) L199–L202. [11] J. Tyczkowski, R. Kapica, J. Łojewska, Thin Solid Films 515 (2007) 6590–6595. [12] Z. Cmjak, B. Orel, Sol. Energy Mater. Sol. Cells 40 (1996) 205–219. [13] S.P.S. Porto, R.S. Krishnan, J. Chem. Phys. 47 (1967) 1009–1012. [14] J.M. Tura, P. Regull, L. Surf. Interface Anal. 11 (1988) 447–449. [15] S.C. Petitto, E.M. Marsh, G.a. Carson, M.a. Langell, J. Mol. Catal. A Chem. 281 (2008) 49–58. [16] R. Berenguer, A. La Rosa-Toro, C. Quijada, E. Morallo, J. Phys. Chem. C 112 (2008) 16945–16952. [17] A. Jnioui, M. Eddouasse, A. Amariglio, J.J. Ehrhardt, M. Alnot, J. Lambert, et al., J. Catal. 106 (1987) 144–165. [18] M. Oku, K. Hirokawa, J. Electron Spectros. Relat. Phenom. 8 (1976) 475–481. [19] J.H. Lee, D.L. Trimm, Fuel Process. Technol. 42 (1995) 339–359. [20] A. Cybulski, J.A. Moulijn, Structured Catalysts and Reactors, Taylor & Francis, 2006.
Please cite this article in press as: P.J. Jodłowski et al., Spectroscopic characterization of Co3O4 catalyst doped with CeO2 and PdO for methane catalytic combustion, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy (2014), http://dx.doi.org/10.1016/j.saa.2014.05.027
