- Email: [email protected]
SiO2 catalyst
G Model CATTOD-8804; No. of Pages 10
ARTICLE IN PRESS Catalysis Today xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today journal homepage: www.elsevier.com/locate/cattod
Support material variation for the Mnx Oy -Na2 WO4 /SiO2 catalyst M. Yildiz a , U. Simon b , T. Otremba a , Y. Aksu c , K. Kailasam d , A. Thomas d , R. Schomäcker a , S. Arndt a,∗ a
Technische Universität Berlin, Institut für Chemie, Technische Chemie, Straße des 17. Juni 124, 10623 Berlin, Germany Technische Universität Berlin, Institut für Werkstoffwissenschaften und -technologien, Fachgebiet Keramische Werkstoffe, Hardenbergstraße 40, 10623 Berlin, Germany c Akdeniz University, Faculty of Engineering, Department of Material Science and Engineering, Dumlupinar Bulvari, 07058 Antalya, Turkey d Technische Universität Berlin, Institut für Chemie, Funktionsmaterialen, Hardenbergstraße 40, 10623 Berlin, Germany b
a r t i c l e
i n f o
Article history: Received 29 June 2013 Received in revised form 5 December 2013 Accepted 7 December 2013 Available online xxx Keywords: Oxidative coupling of methane OCM Mnx Oy -Na2 WO4 /SiO2 Support material variation Silica TiO2 -rutile
a b s t r a c t Mnx Oy -Na2 WO4 /SiO2 is an active catalyst for the oxidative coupling of methane, with a remarkable stability and a suitable performance for an industrial application. Mnx Oy -Na2 WO4 was supported on different support materials and the catalytic activity was investigated with a parallel reactor system, allowing a direct comparison of all results. Considering the C2 yield and the potential practical application, SiO2 is indeed the best support material for this catalyst. However, the comparison in an X-S plot with a reference Mnx Oy -Na2 WO4 /SiO2 catalyst indicates that SiC, Fe2 O3 -based oxides and TiO2 -rutile are also promising support materials. © 2014 Elsevier B.V. All rights reserved.
1. Introduction The oxidative coupling of methane (OCM, see Eq. (1)) is an attractive direct route for the conversion of methane into value added compounds. aCH4 + bO2 → cC2 H6 + dC2 H4 + eH2 O
(1)
In the search for a suitable catalyst, hundreds of materials have been tested in the past [1–5]. The Mnx Oy -Na2 WO4 /SiO2 catalyst has attracted a lot of attention as an OCM catalyst for a practical application [6–9,14]. The literature about this material has recently been reviewed by our research group [3]. The stability of this material is remarkable and confirmed by different research groups [10–15]. Moreover, the catalytic performance (CH4 conversions of 20–30% at C2 selectivities of approximately 70%) makes this catalyst system a potential candidate for practical applications.
∗ Corresponding author. Tel.: +49 30 31424973. E-mail addresses: [email protected] (M. Yildiz), [email protected] (U. Simon), [email protected] (T. Otremba), [email protected] (Y. Aksu), [email protected] (K. Kailasam), [email protected] (A. Thomas), [email protected] (R. Schomäcker), [email protected], [email protected] (S. Arndt).
The formation of a surface cluster species of tetrahedral WO4 species with one W O bond and three W O Si bonds were suggested and presumed to be the active site by Jiang et al. [16]. Wu et al. showed that the tetrahedral WO4 is distorted, most probably due to the presence of Na ions [17]. It was reported that at the surface of the reduced catalyst in the presence of gas phase O2 , F-centers (an oxygen ion vacancy with two trapped electrons) exist, which are able to activate molecular O2 above 80 ◦ C, forming lattice oxygen. A redox mechanism with W O Si bonds and electron transfer between a tungsten species and F-centers was suggested [18], which was later modified to a two metal site model, including an oxygen spillover from the Mn2 O3 to the Na2 WO4 [19]. The two site model seemed to be in accord with XAFS (X-ray Absorption Fine Structure) and XPS (X-Ray Photoelectron Spectroscopy) studies before and after the reaction [20]. The exact composition and functionality of the active centers are still not clear. Palermo et al. [21] studied the oxides of Mn, Na and W supported on SiO2 alone and/or in combination. They reported that amorphous SiO2 undergoes a phase transition to highly crystalline ␣-cristobalite during calcination at 750 ◦ C, far below the usual transition temperature of 1500 ◦ C. This is caused by the presence of Na [21]. A suitable catalyst requires all three metal atoms/oxides, so it was concluded that strong synergies are present between these three metal oxides and a Na-induced crystallization of the support material occurred, converting it from an active and unselective
0920-5861/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.cattod.2013.12.024
Please cite this article in press as: M. Yildiz, et al., Support material variation for the Mnx Oy -Na2 WO4 /SiO2 catalyst, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2013.12.024
G Model CATTOD-8804; No. of Pages 10
ARTICLE IN PRESS M. Yildiz et al. / Catalysis Today xxx (2014) xxx–xxx
2
Table 1 The determined factors [fi = ki (CD4 )/ki (CH4 )] for the reaction constants obtained from CD4 and CH4 with 100 mg catalyst at 750 ◦ C. Reprinted from Yilidz, M., Arndt, S., Simon, U., Otremba, Y. A. T., Berthold, A., Görke, O., Thomas, A., Schubert, H., Schomäcker, R., 2012. Mn-Na2 WO4 /SiO2 – an industrial catalyst for methane coupling. In: Ernst, S., Balfanz, U., Buchholz, S., Lichtscheidl, J., Marchionna, M., Nees F., Santacesaria E., (Eds.), Reducing the Carbon Footprint of Fuels and Petrochemicals, Deutsche Wissenschaftliche Gesellschaft für Erdöl, Erdgas und Kohle e.V., pp. 125–132., with permission from DGMK. fi
f1 f2 f3 f4 f5
Feed gas composition (CH4 :O2 :N2 ) 2:1:4
4:1:4
8:1:4
0.6 0.7 0.3 0.5 0.2
0.5 0.7 0.4 0.5 0.2
0.6 0.7 0.5 0.4 0.2
material into an inert support material. However, the exact active site could not be revealed. Wang et al. compared Mnx Oy -Na2 WO4 /SiO2 , Mnx Oy Na2 WO4 /MgO and NaMnO4 /MgO [22]. Due to the identical catalytic performance of all three materials, it was suggested that a common active site, consisting of Na–O–Mn species exists. This is inconsistent with the suggestion of Wu et al. [17], who proposed that the W O bonds of a distorted WO4 tetrahedron could be the active site. Besides determination of the active center, the production of large amounts of this catalyst is important. Therefore, Simon and co-workers developed an up-scaled preparation of this catalyst [14], which is crucial for any practical application. A variation of the composition has been investigated, trying to find substitutes for Mn [23–26], for Na [27–30] and for W [31–33], respectively. It was found that any other elements are also active, but the combination of the oxides of Mn–Na–W was superior to all tested catalysts. Despite the research performed till now, the role of the SiO2 support material remains unclear. Moreover, there are only very few attempts using other support materials than SiO2 . Liu et al. reported the successful application of SiC as support material [34], giving a similar performance compared to SiO2 . Yu et al. reported a study of different support materials. The applied support materials were rare earth oxides (La2 O3 , CeO2 , Pr6 O11 , Nd2 O3 , Sm2 O3 , Dy2 O3 and Yb2 O3 ) and SiO2 for Na2 WO4 as active component [31]. CeO2 and Pr6 O11 were mainly total oxidation catalysts. Covered with Na2 WO4 they exhibited good C2 selectivities. However, it is unclear why typical support materials such as Al2 O3 , TiO2 , etc., were not used in these experiments. Moreover, the applied rare earth oxides are active OCM catalysts themselves, Sm2 O3 and Nd2 O3 in particular. A blank experiment with only support material was not reported in this study. Generally, it is questionable if supporting an active OCM catalyst on a material which is an active OCM catalyst itself, leads to results, which could significantly increase the knowledge on the supported material due to expectable complex interactions. In the previous work, we performed kinetic isotope effect experiments with CD4 for Mnx Oy -Na2 WO4 /SiO2 catalyst to contribute to the understanding of the reaction pathways and the origin of COx products [35]. The results of these experiments in Table 1 indicate that the most influenced reactions were the consecutive reactions which convert the C2 products towards the total oxidation products (Fig. 1). The kinetic isotope effect is as discussed in the literature [36–45] agrees with the observed results. The consecutive reactions might be suppressed by reaction engineering (e.g. applying new reactor concepts, such as membrane reactor) [46,47]. However, oxidation of CH4 molecules towards COx products, because of unselective catalysts or unselective sites of the selective catalysts,
Fig. 1. Simplified reaction network [2]. Adapted from Ref. [2].
can be prevented only by a development of a new highly selective material or an improvement of a catalyst. One issue which has not been thoroughly studied for the Mnx Oy NaWO4 catalyst is the effect of support materials other than SiO2 . Therefore, in the present manuscript we investigated a detailed variation of different metal oxides as support material for the Mnx Oy -Na2 WO4 in order to study their influence on the catalytic performance. Moreover, we studied in detail how the different components of the active materials (Mnx Oy and Na2 WO4 ) act on these support materials alone. Furthermore, a test with pure Mnx Oy -Na2 WO4 was performed to study the catalytic activity without any support material. All catalytic materials have been studied in a parallel reactor under identical conditions, allowing a direct comparison of the obtained results; this is important because results reported in different publications in literature are often not comparable, due to very different reaction conditions [48]. 2. Experimental part 2.1. Catalyst preparation The prepared catalysts, the loading of the active components and the abbreviations used in the figures and tables are presented in Table 2. Manganese is present in the form of manganese oxides or Mn-containing mixed oxides, however, the loading is calculated for pure Mn. The applied support materials (some of them e.g. La2 O3 , CaO, MgO, are also known active components themselves in the OCM [49]), their origin and specific surface area are shown in Table 3. 2.1.1. WO3 /support Supported tungsten oxide catalysts were prepared via a wet impregnation method of the different support materials (Table 3) with the appropriate aqueous solution of the (NH4 )10 H2 (W2 O7 )6 ·xH2 O (Aldrich, 99.99%, the number of hydrate was assumed four for calculations) at room temperature. Samples were dried in air overnight at 65 ◦ C. The calcination process was performed in air, heating the samples in 4 h from room temperature to 750 ◦ C (with a heating rate of approximately 3 K/min) and holding the temperature for 1 h at 750 ◦ C, consecutively. The prepared catalysts were then ground into fine powder and sieved. The particle size of the catalysts used for the reaction was ≤200 m. Table 2 Prepared catalysts, applied loadings and used abbreviations. Catalyst
Abbreviation
Blank 5 wt% WO3 /Support 5 wt% Na2 WO4 /Support 2 wt% Mnx Oy /Support 2 wt% Mnx Oy –5 wt% Na2 WO4 /Support
Support W/Support Na/W/Support Mn/Support Mn/Na/W/Support
Please cite this article in press as: M. Yildiz, et al., Support material variation for the Mnx Oy -Na2 WO4 /SiO2 catalyst, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2013.12.024
G Model CATTOD-8804; No. of Pages 10
ARTICLE IN PRESS M. Yildiz et al. / Catalysis Today xxx (2014) xxx–xxx
Table 3 The origins and specific surface areas of the used support materials. Support material
Origin of support material
BET (m2 /g)
La2 O3 CaO Al2 O3 ZrO2 SiO2 SiC MgO Fe2 O3 Fe3 O4 SrO TiO2 -rutile TiO2 -anatase
Sigma Aldrich, min. 99.9% Riedel-de Haën BASF 3% Yttria stabilized, Sigma Aldrich BASF Sigma Aldrich Sigma Aldrich, ≤ 99% Sigma Aldrich Alfa Aesar, 98% metal basis Alfa Aesar, 99.5% metal basis Aldrich BASF
3 6 338 105 120 1 124 40 39 2 20 127
WO3 (g) × 100% WO3 (g) + Support(g)
2.2.1. ICP The determination of the metal loading was performed via inductively coupled plasma atomic emission spectroscopy (ICPAES), using a Horiba Scientific ICP Model Ultima 2.
(2)
2.1.2. Na2 WO4 /support Supported sodium tungstate catalysts were prepared with the same method as described in Section 2.1.1. Na2 WO4 ·2H2 O (SigmaAldrich, min. 99%) was used for impregnating the support and the loading was calculated according to Eq. (3). Na2 WO4 (g) Na2 WO4 (wt%) = × 100% Na2 WO4 (g) + Support(g)
(3)
2.1.3. Mnx Oy /support Supported manganese oxide catalysts were prepared with the same method as described in Section 2.1.1. The applied Mnx Oy precursor was (CH3 COO)2 Mn·4H2 O (Fluka, >99%). The loading of Mnx Oy was calculated according to Eq. (4). Mnx Oy (wt%) =
Mn(g) × 100% Mn(g) + Support(g)
(4)
2.1.4. Mnx Oy -Na2 WO4 /support The Mnx Oy -Na2 WO4 active compounds on different support materials were prepared via an adapted two-step wet impregnation method [15]. First, the different support materials were impregnated with an aqueous solution containing appropriate concentration of (CH3 COO)2 Mn·4H2 O (Fluka, >99%) at room temperature. These materials were then dried in air at 65 ◦ C overnight. After that, the obtained materials were impregnated with an aqueous solution containing the appropriate concentration of Na2 WO4 ·2H2 O (Sigma-Aldrich, min. 99%) at room temperature. The preparation was then followed by the same drying, calcination and sieving procedures as described in Section 2.1.1. The loading of Na2 WO4 was calculated according to the Eq. (5) and the Mnx Oy loading was calculated according to Eq. (6). Na2 WO4 (wt%) =
Mnx Oy (wt%) =
Na2 WO4 (g) × 100% Mn(g) + Na2 WO4 (g) + Support(g)
Mn(g) × 100% Mn(g) + Na2 WO4 (g) + Support(g)
themselves. The catalytic performance of the bare untreated support materials as purchased is omitted, because this information was considered irrelevant for the purpose of this study. 2.2. Catalyst characterization
The loading of WO3 was calculated from the recipe according to Eq. (2). WO3 (wt%) =
3
(5)
2.2.2. BET The specific surface area was determined by a Micromeritics Gemini III 2375 Surface Area Analyzer, using N2 adsorption at −196 ◦ C. Before measuring, the samples were degassed at 300 ◦ C and 0.15 mbar for at least 30 min. The surface areas were calculated by the method of Brunauer, Emmett and Teller (BET). 2.2.3. XRD Powder X-Ray diffractograms were obtained (CuK␣1 radiation, wavelength 0.154 nm) using Bruker AXS D8 ADVANCE X-ray diffractometer. The angle variation was performed from 2 to 90◦ , with a step size of 0.008◦ . The diffractograms were analyzed with the program Diffrac.suite EVA. 2.3. Catalytic tests A 6-fold parallel reactor (Integrated Lab Solutions Berlin and Premex Reactor AG) was used for the determination of the catalytic performance, with packed-bed, linear, tubular reactors made of quartz glass. For each catalytic run, 50 mg catalyst were diluted with approximately 1.5 ml quartz sand (quartz sand: purchased from Merck, already washed with HCl and calcined, ca. 60% of the particles have the size 0.2–0.8 mm). Below and above the catalyst bed, a small amount of pure quartz sand was put to ensure proper heat transfer. The particle size of the catalyst was below 200 m in each experiment to exclude internal mass transfer effects. The applied reaction conditions were: 750 ◦ C reactor temperature, 60 ml/min gas flow and a feed gas composition of CH4 :O2 :N2 = 4:1:4 (methane and synthetic air as oxygen source). The analysis was performed with a gas chromatograph Agilent 7890 A, equipped with a flame ionization detector (FID), a thermal conductivity detector (TCD), an HP-PLOT/Q and an HP Molsieve column. He (Air Liquide, purity 99.999%) was used as carrier gas. The analyzed compounds were O2 , N2 , CO2 via TCD, CH4 , C2 H4 and C2 H6 via FID. N2 was used as internal standard. The reproducibility of conversion (X) and selectivity (S) is sufficient. The conversion and selectivity were calculated from a mass balance based on the inlet and outlet concentration of reactants and products, see Eqs. (7) and (8). The carbon balance was always higher than 95%. The desired reaction products of the oxidative coupling of methane are C2 H4 and C2 H6 . The selectivity of these two products was discussed as a sum, the C2 selectivity. X=
(6) SA =
2.1.5. Blank The described procedure in Section 2.1.1 was applied to the support materials exactly the same way, without adding any of the active compounds but only distilled water. These samples were prepared for blank tests for the activity of the support materials
Products
Products + Unconverted Reactant
Product A Reaction Products
(7)
(8)
Under the applied reaction conditions (reactor with quartz sand, no catalyst), as described above, the results of thermal OCM reaction were: O2 conversion: 1.2%, CH4 conversion: 0.2% at 93.8% C2 selectivity.
Please cite this article in press as: M. Yildiz, et al., Support material variation for the Mnx Oy -Na2 WO4 /SiO2 catalyst, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2013.12.024
ARTICLE IN PRESS
G Model CATTOD-8804; No. of Pages 10
M. Yildiz et al. / Catalysis Today xxx (2014) xxx–xxx
4
Table 4 The intended and determined loadings of the selected catalysts. Active components
Intended loading
Mn (wt%) Na (wt%) W (wt%)
2.00 0.78 3.13
Determined loading Mn/Na/W/SiO2
Mn/Na/W/TiO2 -rutile
Mn/Na/W/SiC
2.16 0.85 2.95
2.02 0.73 3.12
1.99 0.65 3.04
Table 5 The specific surface areas in m2 /g of the prepared fresh catalysts and spent Mnx Oy -Na2 WO4 /Support catalysts in the OCM (Pure Support: Support material without any treatment. Blank Material: Support material after applying the preparation method of the catalysts but without any active components, see Section 2.1.5., S: Support material). Support material
Pure support
Blank material
W/S
Na/W/S
Mn/S
Mn/Na/W/S
Mn/Na/W/S (spent)
La2 O3 CaO Al2 O3 ZrO2 SiO2 SiC MgO Fe2 O3 Fe3 O4 SrO TiO2 -rutile TiO2 -anatase
3 6 338 105 120 1 124 40 39 2 20 127
6 12 217 51 113 ≤1 37 3 12 3 11 29
3 7 186 50 52 ≤1 73 10 9 2 7 22
5 3 260 46 16 ≤1 21 5 4 4 8 30
10 10 222 49 104 2 22 6 10 4 10 6
9 7 242 36 11 ≤1 15 5 5 3 6 13
7 8 94 1 5 3 11 3 3 2 4 7
3. Results & discussion 3.1. Characterization 3.1.1. Elemental analysis Due to the large number of catalysts, it was not possible to determine all loadings for all prepared materials via chemical analysis. Therefore, the loading for a few chosen catalysts was determined (Table 4) and found to agree with the intended loading, within the range of experimental errors. 3.1.2. Specific surface area The BET surface areas of all prepared materials are shown in Table 5. Surface areas of the selected metal oxides vary between very low (e.g. La2 O3 , CaO, SiC, SrO) and high values (Al2 O3 , ZrO2 , SiO2 , MgO, TiO2 -anatase). The surface areas of the blank samples are lower than those of pure support materials because of the calcination process, except for La2 O3 , CaO and SrO. There is a strong reduction of the surface area upon the addition of the active component, except for La2 O3 , Al2 O3 , CaO, SiC and SrO. Especially for SiO2 , the most substantial decrease occurs in the samples which contain Na, because of the phase transformation of the amorphous SiO2 into ␣-cristobalite [3]. As expected the surface areas of most Mnx Oy -Na2 WO4 /Support catalysts decrease after 16 h time on stream, while the surface areas of CaO and SiC supported catalysts increase. The strongest reductions happen for Al2 O3 and ZrO2 supported samples. 3.1.3. Phase analysis A summary of the phase analysis of the detected XRD patterns for the pure support materials, the blank samples, the supported single metal oxides and the mixed metal oxides are presented in Tables 6–8. XRD patterns and phase analysis of all blank samples and catalysts can be found in the supporting information (Appendix A). Generally, the prepared materials are multi-phase materials, leading to numerous different phases being present. Therefore, only unambiguously identified patterns are presented. This does not exclude the presence of other phases, either in minor amounts or amorphous states. Moreover, many prepared catalysts are amorphous, prohibiting a structural analysis via XRD.
For the pure support material, as purchased and blank, the expected phases are found. Additionally, for certain compounds, such as La2 O3 , CaO or SrO, hydroxides and carbonates are also present, for MgO, only Mg(OH)2 and no carbonates are found. This could be due to storage in air. TiO2 in the rutile form, also contains TiO2 in the anatase form. Treating the purchased support material with the preparation procedure obtaining the blank samples, does not lead to any observable changes in the material. Loading only WO3 on the support materials, led again to the formation of hydroxides and carbonates for a number of support materials. The formation of WO3 is detected for Al2 O3 , SiO2 , SiC, Fe2 O3 , Fe3 O4 and TiO2 -rutile. However, for a number of support materials, the formation of tungstate compounds is found, e.g. La10 W22 O81 , CaWO4 , Ca3 WO6 , MgWO4 , Sr3 WO6 and SrWO4 . Supporting WO3 on TiO2 -anatase and ZrO2 results in amorphous materials, indicating a strong structural change of the whole support material. Similar to that of WO3 , loading of Mnx Oy on the support materials leads to the formation of hydroxides and carbonates. Mn2 O3 is observed as a separate phase for a number of support materials. The formation of mixed metal oxides is observed with Mn on only
Table 6 The detected phases for the different support and blank materials. Support materials Support material
Pure support material
Blank material
La2 O3 CaO Al2 O3 ZrO2 SiO2 SiC MgO Fe2 O3 Fe3 O4 SrO TiO2 -rutile TiO2 -anatase No support
La2 O3 , La(OH)3 CaO, Ca(OH)2 , CaCO3 Amorphous Y0.15 Zr0.85 O1.93 Amorphous SiC MgO, Mg(OH)2 Fe2 O3 Fe2 O3 SrO, SrCO3 , Sr(OH)2 ·H2 O Rutile, anatase Anatase –
La2 O3 , La(OH)3 CaO, Ca(OH)2 , CaCO3 Amorphous Y0.15 Zr0.85 O1.93 Amorphous SiC MgO Fe2 O3 Fe2 O3 SrO, SrCO3 , Sr(OH)2 ·H2 O Rutile, anatase Anatase –
Please cite this article in press as: M. Yildiz, et al., Support material variation for the Mnx Oy -Na2 WO4 /SiO2 catalyst, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2013.12.024
G Model CATTOD-8804; No. of Pages 10
ARTICLE IN PRESS M. Yildiz et al. / Catalysis Today xxx (2014) xxx–xxx
5
Table 7 The detected phases for the different supported single metal oxides. Single metal oxides Support material
W/Support
Mn/Support
La2 O3 CaO Al2 O3 ZrO2 SiO2 SiC MgO Fe2 O3 Fe3 O4 SrO TiO2 -rutile TiO2 -anatase No support
La2 O3 , La(OH)3 , La10 W22 O81 CaO, Ca(OH)2 , CaWO4 , Ca3 WO6 WO3 , amorphous Amorphous WO3 , amorphous SiC, WO3 MgO, Mg(OH)2 , MgCO3 , MgWO4 WO3 , Fe2 O3 WO3 , Fe2 O3 SrO, SrCO3 , Sr3 WO6 , SrWO4 Ti4 H2 O9.1 .9H2 O, rutile, anatase, WO3 Amorphous WO3
La2 O3 , La(OH)3 CaO, Ca(OH)2 , Mn2 O3 , Ca2 MnO4 Amorphous Y0.15 Zr0.85 O1.93 Mn2 O3 or MnMn6 SiO12 , amorphous SiC, Mn2 O3 MgO, Mg(OH)2 Fe2 O3 Fe2 O3 SrO, SrCO3 , Sr(OH)2 .H2 O Rutile, anatase Anatase, rutile, Mn2 O3 Mn2 O3
Mnx Oy /CaO. Mnx Oy does not change TiO2 -anatase and ZrO2 into amorphous materials, like WO3 . When supporting Na2 WO4 , only for SiO2 , SiC, Fe2 O3 and TiO2 rutile detectable amounts of Na2 WO4 were found. The known transition of amorphous SiO2 into ␣-cristobalite and tridymite phases is found while TiO2 -anatase is turned into an amorphous material. Supporting Mnx Oy -Na2 WO4 , hydroxides or carbonates are only found for MgO, La2 O3 and SrO. Mixed oxides of support-Mn or support-W phases are found for CaO and SrO. Al2 O3 , ZrO2 and TiO2 -anatase support materials lead the Mnx Oy -Na2 WO4 catalysts to totally amorphous materials. However on TiO2 -rutile supported catalyst, Na2 WO4 , TiO2 -anatase and TiO2 -rutile phases were detected to the contrary of the Mnx Oy -Na2 WO4 /TiO2 -anatase. Fe2 O3 phase is dominating for Mnx Oy -Na2 WO4 /Fe3 O4 , but for Fe2 O3 supported catalyst Na2 WO4 phase is also present additionally. On SiO2 and SiC supported catalysts mainly same active component phases (Na2 WO4 and Mn2 O3 ) were detected. Amorphous and unselective silica support material transformed into ␣-cristobalite phase after preparation process for the Mnx Oy Na2 WO4 /SiO2 . For the Mnx Oy /SiO2 and Mnx Oy -Na2 WO4 /SiO2 catalysts, patterns identified as Mn2 O3 might also be explained as braunite (MnMn6 SiO12 ) phase. Since patterns of Mn2 O3 and braunite are very similar and overlap, it is too difficult to distinguish whether the patterns belong to Mn2 O3 or braunite. It can be seen in the phase analysis of the catalysts (Appendix A) that in some catalysts both amorphous contributions and crystalline phases are present together. However, certain oxides, such as La2 O3 or Fe3 O4 , Al2 O3 and TiO2 -anatase do not show any distinct patterns when the material is supported on them. A reason could be either that the supported material remains amorphous or that the
support material reacts with the metal oxides forming a new compound in low concentration or in an amorphous state, prohibiting any detection via XRD. 3.2. Catalysis A good method to compare different catalysts is to determine via kinetic modeling the best performance which can be achieved over a range of operating conditions. However, this is in any case time consuming and many catalysts are instable, making a comparison difficult. In the present manuscript, we measured the catalytic activity only for one set of reaction conditions (feed gas composition, gas flow, mass of catalyst, reactor temperature) but for 16 h time on stream. Because especially for the very temperaturesensitive OCM reaction it is really a challenging issue to compare the results with reported results in literature due to very different conditions [48]. We are aware that this comparison is not the optimal way for a catalyst selection. However, the obtained results can be used to exclude catalysts with an inappropriate performance, for example due to instability or very low C2 selectivities. Moreover, an estimation if a support material is a candidate for further investigations, or not, should also be possible. 3.2.1. Stability The stability of a catalyst is a crucial point for its application potential. However, due to the large amounts of measured data, it is not feasible to present the time on stream trajectories for all prepared catalysts here. The complete data are shown in Appendix B. However, in order to provide information about the stability of the catalysts here, the first data point for the CH4 conversion, obtained approximately 60 min after reaching the desired reactor temperature, is compared with the last data point for the CH4 conversion,
Table 8 The detected phases for the different supported mixed metal oxides, Na2 WO4 and Mnx Oy -Na2 WO4 . Mixed metal oxides Support material
Na/W/Support
Mn/Na/W/Support
La2 O3 CaO Al2 O3 ZrO2 SiO2 SiC MgO Fe2 O3 Fe3 O4 SrO TiO2 -rutile TiO2 -anatase No support
La2 O3 , La(OH)3 CaO, Ca3 WO6 Amorphous Y0.15 Zr0.85 O1.93 SiO2 (cristobalite, tridymite), Na2 WO4 , Na4 WO5 SiC, Na2 WO4 MgO Fe2 O3 , Na2 WO4 Fe2 O3 SrO, Sr(OH)2 , SrCO3 , Sr(OH)2 .H2 O Rutile, anatase, Na2 WO4 Amorphous Na2 WO4
La2 O3 , La(OH)3 CaO, Ca3 WO6 , Ca2 MnO4 Amorphous Amorphous SiO2 (cristobalite, tridymite), Mn2 O3 or MnMn6 SiO12 , Na2 WO4 , Na4 WO5 SiC, Mn2 O3 , Na2 WO4 MgO, Mg(OH)2 , Mn2 O3 Fe2 O3 , Na2 WO4 Fe2 O3 SrO, SrCO3 , Sr(OH)2 .H2 O, Sr(OH)2 , Sr2 MnO4 , MnO Rutile, anatase, Na2 WO4 Amorphous Mn2 O3 , Na2 WO4
Please cite this article in press as: M. Yildiz, et al., Support material variation for the Mnx Oy -Na2 WO4 /SiO2 catalyst, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2013.12.024
ARTICLE IN PRESS
G Model CATTOD-8804; No. of Pages 10
M. Yildiz et al. / Catalysis Today xxx (2014) xxx–xxx
6
Table 9 The residual catalytic activity for CH4 conversion in % after 16 h time on stream. Support material
Blank material
W/Support
Mn/Support
Na/W/Support
Mn/Na/W/Support
La2 O3 CaO Al2 O3 ZrO2 SiO2 SiC MgO Fe2 O3 Fe3 O4 SrO TiO2 -rutile TiO2 -anatase No support
98 80 82 100 22 60 42 16 75 90 74 54 –
51 55 125 80 67 200 200 32 16 85 83 100 100
104 94 103 105 36 64 101 40 77 102 57 33 74
68 79 101 49 112 88 160 63 83 72 80 180 100
88 98 108 67 89 121 108 82 89 115 82 38 71
obtained after approximately 16 h time on stream. The result is the residual catalytic activity. The calculated data is presented in Table 9. The catalytic activity of some catalysts, like WO3 /SiC, Na2 WO4 /TiO2 -anatase, Na2 WO4 /MgO increases with time on stream. This is probably caused by structural changes during the reaction. In Figs. 2–4, XRD patterns of some fresh and spent catalysts are shown for catalysts with a strongly increased catalytic activity after 16 h time on stream. Some of them demonstrate structural changes different from fresh catalysts. It was detected for the XRD patterns of WO3 /SiC that the ratio of SiC to WO3 has changed. For the Na2 WO4 /TiO2 -anatase, a phase transformation from amorphous state into anatase, rutile and Na2 WO4 phases occurred during reaction, showing very strong crystallization. The patterns of Na2 WO4 /MgO catalysts before and after reaction are exactly the same, no changes are observed. The stability of the blank samples was high, except for SiO2 , MgO, Fe2 O3 and TiO2 -anatase. WO3 /Fe2 O3 and WO3 /Fe3 O4 catalysts strongly deactivated, while the activity of the Al2 O3 , SiC and MgO supported WO3 samples increased. SrO supported Mnx Oy catalyst was stable during the reaction whereas Mnx Oy /SiO2 , Mnx Oy /TiO2 -anatase and Mnx Oy /Fe2 O3 lost their activity substantially. Although supported Na2 WO4 catalysts were stable, except for the ZrO2 supported sample, with the combination of three active components, nearly all supported catalysts were highly stable except Mnx Oy -Na2 WO4 /ZrO2 and Mnx Oy -Na2 WO4 /TiO2 anatase which strongly deactivated. Among all catalysts Al2 O3 and MgO supported samples did not exhibit any sign of deactivation. However, the unselective performance of Al2 O3 supported catalysts is an important disadvantage with respect to the
desired C2 products. Furthermore no interpretation can be derived for the La2 O3 , Mnx Oy /La2 O3 , Mnx Oy -Na2 WO4 /La2 O3 , Mnx Oy /CaO, Mnx Oy -Na2 WO4 /CaO, Mnx Oy /Al2 O3 , Mnx Oy -Na2 WO4 /Al2 O3 , ZrO2 , Mnx Oy /ZrO2 and Mnx Oy /MgO catalysts because of full O2 conversion (Appendix B). 3.2.2. Comparison of catalytic performances For the blank support material, WO3 /Support, Na2 WO4 /Support, Mnx Oy /Support and Mnx Oy -Na2 WO4 /Support the catalytic performances after approximately 16 h are presented in Fig. 8. In Fig. 5 the CH4 conversions are presented, in Fig. 6 the C2 selectivities and in Fig. 7 the C2 yields. The CH4 conversions of Al2 O3 , SiO2 , SiC, Fe2 O3 , SrO, TiO2 -rutile and TiO2 -anatase support materials are rather poor, on the other hand La2 O3 , CaO, ZrO2 , MgO and Fe3 O4 are active themselves, but C2 selectivities of these active support materials are very low. It becomes evident, that certain support materials (i.e. La2 O3 , CaO or MgO) are by no means inert carriers. In particular, La2 O3 exhibits a high activity, which is hardly changed upon the loading of the metal oxides, except for WO3 . However, most of the other support materials show a rather low performance. Supporting WO3 , does not lead to any significant CH4 conversion. In fact, it does decrease the conversion of the active support materials. This is especially prominent for La2 O3 , ZrO2 , MgO and Fe3 O4 , while it is not distinct for SiO2 [50], SiC and CaO support materials. WO3 improved slightly the activity only for SrO. Since most of the supported WO3 catalysts have low CH4 conversion, their C2 selectivities are very high. However, supporting Na2 WO4 , the CH4 conversion increases compared to supporting WO3 . La2 O3 , CaO and Al2 O3 supported Na2 WO4 catalysts are highly active in comparison to the other
Fig. 2. XRD patterns of W/SiC before and after 16 h time on stream (䊉: WO3 ,
SiC).
Please cite this article in press as: M. Yildiz, et al., Support material variation for the Mnx Oy -Na2 WO4 /SiO2 catalyst, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2013.12.024
ARTICLE IN PRESS
G Model CATTOD-8804; No. of Pages 10
M. Yildiz et al. / Catalysis Today xxx (2014) xxx–xxx
7
Fig. 3. XRD patterns of Na/W/TiO2 -anatase before and after 16 h time on stream (: Na2 WO4 , : TiO2 -anatase, : TiO2 -rutile).
Fig. 4. XRD patterns of Na/W/MgO before and after 16 h time on stream (: MgO).
Na2 WO4 /Support samples, but their C2 selectivity is very poor. When Al2 O3 , SiO2 and SiC supported Na2 WO4 catalysts are compared to the blank samples of the corresponding support materials, their CH4 conversions increase explicitly in contrast to the other catalysts.
Supporting only Mnx Oy , leads to relatively high CH4 conversion. For Mnx Oy on SiO2 and MgO, this has already been reported [51–53]. The CH4 conversions over La2 O3 , CaO, Al2 O3 , ZrO2 , MgO and Fe3 O4 supported manganese oxide catalysts are outstanding, some of them are even higher than conversion of the
25
CH4 Conversion [%]
20
15
10
5
0 Lanthana Calcium oxide
Alumina Zirconia Silica Silicon carbide
Mn/Na/W/Support Mn/Support Na/W/Support W/Support Support
Magnesia Iron (II,III) Iron (III) oxide oxide
Strona TitaniaRule
Titania-
Anatase Without Support
Fig. 5. The CH4 conversion determined after approximately 16 h time on stream for the different materials. Reaction conditions: 750 ◦ C, 50 mg catalyst, flow rate of 60 ml/min and feed gas composition of CH4 :O2 :N2 = 4:1:4.
Please cite this article in press as: M. Yildiz, et al., Support material variation for the Mnx Oy -Na2 WO4 /SiO2 catalyst, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2013.12.024
ARTICLE IN PRESS
G Model CATTOD-8804; No. of Pages 10
M. Yildiz et al. / Catalysis Today xxx (2014) xxx–xxx
8
90 80
C2-Selecvity [%]
70 60 50 40 30 20 10 0 Lanthana Calcium oxide
Alumina Zirconia
Mn/Na/W/Support Mn/Support Na/W/Support W/Support Support
Silica Silicon carbide
Magnesia Iron (II,III) Iron (III) oxide oxide
Strona TitaniaRule
TitaniaAnatase
Without Support
Fig. 6. The C2 selectivity determined after approximately 16 h time on stream for the different materials. Reaction conditions: 750 ◦ C, 50 mg catalyst, flow rate of 60 ml/min and feed gas composition of CH4 :O2 :N2 = 4:1:4.
Mnx Oy -Na2 WO4 /Support catalysts. However the C2 selectivities are mostly drastically reduced, total oxidation is dominating. Supporting Mnx Oy -Na2 WO4 , it becomes evident that the combination of these three metal oxides results in the best catalytic performance, c.f. C2 yield. La2 O3 , CaO, Al2 O3 , ZrO2 and SiO2 supported trimetallic catalysts show the highest CH4 conversions amongst all Mnx Oy -Na2 WO4 /Support samples. While the wellknown catalyst Mnx Oy -Na2 WO4 /SiO2 is highly selective towards C2 products, for the La2 O3 , CaO, Al2 O3 , ZrO2 supported trimetallic catalysts total oxidation products are prevailing. Considering the C2 yield, presented in Fig. 7, it can be seen that SiO2 is not the only suitable support material. Support materials, such as La2 O3 , Al2 O3 , CaO or ZrO2 exhibit a similar C2 yield than SiO2 , but their CH4 conversion is much higher at much lower C2 selectivity. This catalytic behavior is not beneficial for a practical application. However, TiO2 -rutile and SiC exhibit a similar performance than SiO2 , with relatively low conversions at high C2 selectivities.
Regarding catalyst components without support material, Mnx Oy shows substantial activity itself with poor C2 selectivity, whereas the CH4 conversion over WO3 and Na2 WO4 is extremely low. Furthermore, it is interesting to note, that Mnx Oy -Na2 WO4 without support material also displays a considerable activity and selectivity, questioning the actual contribution of the support material. The results of the catalytic tests of all tested materials are summarized in a conversion selectivity diagram (Fig. 8) by comparing the obtained data points with the trajectory of a reference Mnx Oy Na2 WO4 /SiO2 catalyst, which is described in more detail in [14]. If only the number of active sites changes, the resulting data points would fall on the reference trajectory. However, if the nature of active sites is changed, the data points might be below (aggravation) or above (improvement) the reference trajectory, indicating possible changes of the catalytic properties. WO3 /Support and Na2 WO4 /Support catalysts are concentrated mostly in the low CH4 conversion region with 0–3%. Mnx Oy /Support
10 9 8 C2-Yield [%]
7 6 5 4 3 2 1 0 Lanthana
Calcium oxide
Alumina
Zirconia
Silica
Silicon carbide
Magnesia
Iron (II,III) oxide
Iron (III) oxide
Strona
TitaniaRule
Mn/Na/W/Support Mn/Support Na/W/Support W/Support Support TitaniaAnatase
Without Support
Fig. 7. The C2 yield determined after approximately 16 h time on stream for the different materials. Reaction conditions: 750 ◦ C, 50 mg catalyst, flow rate of 60 ml/min and feed gas composition of CH4 :O2 :N2 = 4:1:4.
Please cite this article in press as: M. Yildiz, et al., Support material variation for the Mnx Oy -Na2 WO4 /SiO2 catalyst, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2013.12.024
G Model CATTOD-8804; No. of Pages 10
ARTICLE IN PRESS M. Yildiz et al. / Catalysis Today xxx (2014) xxx–xxx
9
Fig. 8. Active components on the different support materials were compared in an X-S diagram for the oxidative coupling of methane. Results were recorded after approximately 16 h time on stream for each data point. Reaction conditions: 750 ◦ C, 50 mg catalyst, flow rate of 60 ml/min and feed gas composition of CH4 :O2 :N2 = 4:1:4.
catalysts and bulk materials (pure support materials) exhibit high conversions but they are mostly unselective towards C2 products. Even if they showed high or low CH4 conversion, nearly all data points are below the reference trajectory. For Mnx Oy -Na2 WO4 supported on SrO, MgO, CaO, La2 O3 , Al2 O3 and ZrO2 , the performance is also worse in comparison to the reference trajectory. Despite their often high CH4 conversion, they usually lack C2 selectivity. Moreover, some of these materials strongly deactivate, e.g. ZrO2 . This means that the prepared catalysts contain active sites with higher tendency towards COx than the reference catalysts. The materials supported on TiO2 (rutile or anatase), Fe2 O3 , Fe3 O4 , SiC and SiO2 are either on the reference trajectory or well above, indicating that they could also be suitable support materials. If stability of the catalytic performance is additionally taken into account, then TiO2 -anatase has actually to be excluded. However, without knowing the exact nature of the active sites it is possible to say, that they are at least as good if not better than that of the reference Mnx Oy -Na2 WO4 /SiO2 . That could mean, that their active sites are either less unselective or less active for the consecutive combustion of the C2 products. 4. Summary & conclusion For the Mnx Oy -Na2 WO4 , a variety of different support materials has been studied. Most of the materials show a catalytic performance inferior to the reference Mnx Oy -Na2 WO4 /SiO2 and/or deactivate, although Mnx Oy -Na2 WO4 without a support material exhibits also significant catalytic activity, questioning suggestions of the active centers like W O Si bonds. However, since quartz sand has been used to dilute the catalyst, it cannot be excluded that the active component has smeared over the SiO2 particles, although it seems unlikely. La2 O3 and CaO exhibit a high activity as pure support materials, making them unattractive for fundamental research, since it will be difficult to distinguish the contribution of the support material and the active phase. Among the remaining materials only SiO2 , SiC, TiO2 -rutile and the Fe2 O3 -based support materials show catalytic performances higher than the reference Mnx Oy -Na2 WO4 /SiO2 . Considering the C2 yield, SiO2 results in the best catalyst, however, the other materials
have active sites which are as suitable (SiC, Fe-based) or even more suitable (TiO2 -rutile) for the formation of C2 , but seem to be present in lower numbers. Acknowledgments This work is part of the Cluster of Excellence “Unifying Concepts in Catalysis” coordinated by the Technische Universität Berlin. Financial support by the Deutsche Forschungsgemeinschaft (DFG) within the framework of the German Initiative for Excellence is gratefully acknowledged. We would like to thank Prof. Dr. Helmut Schubert† for his precious contributions. Mr. Yildiz is obliged to the Ministry of Education of the Republic of Turkey for fundings. We also thank Harald Link for ICP measurements and Maria Unterweger for XRD measurements. We are also obliged to Mr. Schiele and the workshop for their support with the equipment. Moreover, we would like to thank Evonik and BASF for supplying support material. Appendices A and B. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/ 10.1016/j.cattod.2013.12.024. References [1] A.M. Maitra, I. Campbell, R.J. Tyler, Appl. Catal. A Gen. 85 (1992) 27–46. [2] E. Kondratenko, M. Baerns, Oxidative coupling of methane, in: G. Ertl, H. Knözinger, F. Schüth, J. Weitkamp (Eds.), Handbook of Heterogeneous Catalysis, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, 2008, pp. 3010–3023. [3] S. Arndt, T. Otremba, U. Simon, M. Yildiz, H. Schubert, R. Schomäcker, Appl. Catal. A Gen. 425–426 (2012) 53–61. [4] S. Arndt, G. Laugel, S. Levchenko, R. Horn, M. Baerns, M. Scheffler, R. Schlögl, R. Schomäcker, Catal. Rev. 53 (2011) 424–514. [5] U. Zavyalova, M. Holena, R. Schlögl, M. Baerns, ChemCatChem 12 (2011) 1935–1947. [6] H. Liu, X. Wang, D. Yang, R. Gao, Z. Wang, J. Yang, J. Nat. Gas Chem. 17 (2008) 59–63. [7] M. Makri, C.G. Vayenas, Appl. Catal. A Gen. 244 (2003) 301–310. [8] S. Jaˇso, S. Sadjadi, H.R. Godini, U. Simon, S. Arndt, O. Görke, A. Berthold, H. Arellano-Garcia, H. Schubert, R. Schomäcker, G. Wozny, J. Nat. Gas Chem. 21 (2012) 534–543.
Please cite this article in press as: M. Yildiz, et al., Support material variation for the Mnx Oy -Na2 WO4 /SiO2 catalyst, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2013.12.024
G Model CATTOD-8804; No. of Pages 10 10 [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31]
ARTICLE IN PRESS M. Yildiz et al. / Catalysis Today xxx (2014) xxx–xxx
S. Jaˇso, H. Arellano-Garcia, G. Wozny, J. Chem. Eng. 171 (2011) 255–271. X. Fang, S. Li, J. Lin, Y. Chu, J. Mol. Catal. (China) 6 (1992) 427–433. X. Fang, S. Li, J. Lin, J. Gu, D. Yan, J. Mol. Catal. (China) 6 (1992) 255–262. X. Wang, J. Zhang, D. Yang, C. Zhang, J. Lin, S. Li, Shiyou Huagong 26 (1997) 361–367. J. Lin, J. Gu, D. Yang, C. Zhang, Y. Yang, Y. Chu, S. Li, Shiyou Huagong 24 (1995) 293–298. U. Simon, O. Görke, A. Berthold, S. Arndt, R. Schomäcker, H. Schubert, J. Chem. Eng. 168 (2011) 1352–1359. S. Pak, P. Qiu, J. Lunsford, J. Catal. 179 (1998) 222–230. Z. Jiang, C. Yu, X. Fang, S. Li, H. Wang, J. Phys. Chem.-US 97 (1993) 12870–12875. J. Wu, S. Li, J. Phys. Chem.-US 99 (1995) 4566–4568. J. Wu, S. Li, J. Niu, X. Fang, Appl. Catal. A Gen. 124 (1995) 9–18. Z. Jiang, H. Gong, S. Li, Stud. Surf. Sci. Catal. 112 (1997) 481–490. Y. Kou, B. Zhang, J. Niu, S. Li, H. Wang, T. Tanaka, S. Yoshida, J. Catal. 173 (1998) 399–408. A. Palermo, J. Vazquez, A. Lee, M. Tikhov, R. Lambert, J. Catal. 177 (1998) 259–266. D. Wang, M. Rosynek, J. Lunsford, J. Catal. 155 (1995) 390–402. A. Malekzadeh, A. Khodadadi, M. Abedini, M. Amini, A. Bahramian, A. Dalai, Catal. Commun. 2 (2001) 241–247. A. Malekzadeh, A. Dalai, A. Khodadadi, Y. Mortazavi, Catal. Commun. 9 (2008) 960–965. Z. Gholipour, A. Malekzadeh, R. Hatami, Y. Mortazavi, A. Khodadadi, J. Nat. Gas Chem. 19 (2010) 35–42. A. Malekzadeh, M. Abedini, A. Khodadadi, M. Amini, H. Mishra, A. Dalai, Catal. Lett. 84 (2002) 45–51. S. Ji, T. Xiao, S. Li, L. Chou, B. Zhang, C. Xu, R. Hou, A. York, M. Green, J. Catal. 220 (2003) 47–56. A. Palermo, J.H. Vazquez, R. Lambert, Catal. Lett. 68 (2000) 191–196. A. Malekzadeh, A. Khodadadi, A. Dalai, M. Abedini, J. Nat. Gas Chem. 16 (2007) 121–129. A. Tyunyaev, G. Nipan, T. Kol’tsova, A. Loktev, V. Ketsko, A. Dedov, I. Moiseev, Russ. J. Inorg. Chem. 54 (2009) 664–667. Z. Yu, X. Yang, J. Lunsford, M. Rosynek, J. Catal. 154 (1995) 163–173.
[32] S. Mahmoodi, M. Ehsani, S. Goreishi, J. Ind. Eng. Chem. 16 (2010) 923–928. [33] S. Hou, Y. Cao, W. Xiong, H. Liu, Y. Kou, Ind. Eng. Chem. Res. 45 (2006) 7077–7083. [34] H. Liu, D. Yang, R. Gao, L. Chen, S. Zhang, X. Wang, Catal. Commun. 9 (2008) 1302–1306. [35] M. Yildiz, S. Arndt, U. Simon, Y. Aksu, T. Otremba, A. Berthold, O. Görke, A. Thomas, H. Schubert, R. Schomäcker, DGMK-Tagungsbericht 2012–3 (2012) 125–132. [36] N.W. Cant, C.A. Lukey, P.F. Nelson, R.J. Tylerb, J. Chem. Soc., Chem. Commun. (1988) 766–768. [37] N.W. Cant, C.A. Lukey, P.F. Nelson, J. Catal. 124 (1990) 336–348. [38] N.W. Cant, E.M. Kennedy, J. Phys. Chem. 97 (1993) 1445–1450. [39] C.A. Mims, R.B. Hall, K.D. Rose, G.R. Myers, Catal. Lett. 2 (1989) 361–368. [40] C.A. Mims, R. Mauti, J. Phys. Chem. 98 (1994) 13357–13372. [41] P.F. Nelson, C.A. Lukey, N.W. Cant, J. Phys. Chem. 92 (1988) 6176–6179. [42] P.F. Nelson, C.A. Lukey, N.W. Cant, J. Catal. 120 (1989) 216–230. [43] P.F. Nelson, J. Phys. Chem. 94 (1990) 3756–3761. [44] P.F. Nelson, E.M. Kennedy, N.W. Cant, Isotopic labelling studies of the mechanism of the catalytic oxidative coupling of methane, in: A. Holmen et al. (Eds.), Natural Gas Conversion, Elsevier Science Publishers B.V., 1991, pp. 89–95. [45] C. Shi, M. Xu, M.P. Rosynek, J.H. Lunsford, J. Phys. Chem. 97 (1993) 216–222. [46] S. Bhatia, C.Y. Thien, A.R. Mohamed, J. Chem. Eng. 148 (2009) 525–532. [47] T.P. Tiemersma, A.S. Chaudhari, F. Gallucci, J.A.M. Kuipers, M.S. Annaland, Chem. Eng. Sci. 82 (2012) 232–245. [48] M. Baerns, J. Ross, Catalytic chemistry of methane conversion, in: J.M. Thomas, K.I. Zamaraev (Eds.), Perspectives in Catalysis, Blackwell Science, IUPAC International Union of Pure and Applied Chemistry, 1992, pp. 315–335. [49] Y. Amenomiya, V.I. Birss, M. Goledzinowski, J. Galuszka, A.R. Sanger, Catal. Rev.Sci. Eng. 32 (3) (1990) 163–227. [50] J. Wu, S. Li, J. Phys. Chem. 99 (1995) 4566–4568. [51] A. Gaffney, C. Jones, J. Leonard, J. Sofranko, H. Withers, Stud. Surf. Sci. Catal. 38 (1987) 523–529. [52] J. Sofranko, J. Leonard, C. Jones, J. Catal. 103 (1987) 302–310. [53] J. Sofranko, J. Leonard, C. Jones, A. Gaffney, H. Withers, Catal. Today 3 (1988) 127–135.
Please cite this article in press as: M. Yildiz, et al., Support material variation for the Mnx Oy -Na2 WO4 /SiO2 catalyst, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2013.12.024
ARTICLE IN PRESS Catalysis Today xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today journal homepage: www.elsevier.com/locate/cattod
Support material variation for the Mnx Oy -Na2 WO4 /SiO2 catalyst M. Yildiz a , U. Simon b , T. Otremba a , Y. Aksu c , K. Kailasam d , A. Thomas d , R. Schomäcker a , S. Arndt a,∗ a
Technische Universität Berlin, Institut für Chemie, Technische Chemie, Straße des 17. Juni 124, 10623 Berlin, Germany Technische Universität Berlin, Institut für Werkstoffwissenschaften und -technologien, Fachgebiet Keramische Werkstoffe, Hardenbergstraße 40, 10623 Berlin, Germany c Akdeniz University, Faculty of Engineering, Department of Material Science and Engineering, Dumlupinar Bulvari, 07058 Antalya, Turkey d Technische Universität Berlin, Institut für Chemie, Funktionsmaterialen, Hardenbergstraße 40, 10623 Berlin, Germany b
a r t i c l e
i n f o
Article history: Received 29 June 2013 Received in revised form 5 December 2013 Accepted 7 December 2013 Available online xxx Keywords: Oxidative coupling of methane OCM Mnx Oy -Na2 WO4 /SiO2 Support material variation Silica TiO2 -rutile
a b s t r a c t Mnx Oy -Na2 WO4 /SiO2 is an active catalyst for the oxidative coupling of methane, with a remarkable stability and a suitable performance for an industrial application. Mnx Oy -Na2 WO4 was supported on different support materials and the catalytic activity was investigated with a parallel reactor system, allowing a direct comparison of all results. Considering the C2 yield and the potential practical application, SiO2 is indeed the best support material for this catalyst. However, the comparison in an X-S plot with a reference Mnx Oy -Na2 WO4 /SiO2 catalyst indicates that SiC, Fe2 O3 -based oxides and TiO2 -rutile are also promising support materials. © 2014 Elsevier B.V. All rights reserved.
1. Introduction The oxidative coupling of methane (OCM, see Eq. (1)) is an attractive direct route for the conversion of methane into value added compounds. aCH4 + bO2 → cC2 H6 + dC2 H4 + eH2 O
(1)
In the search for a suitable catalyst, hundreds of materials have been tested in the past [1–5]. The Mnx Oy -Na2 WO4 /SiO2 catalyst has attracted a lot of attention as an OCM catalyst for a practical application [6–9,14]. The literature about this material has recently been reviewed by our research group [3]. The stability of this material is remarkable and confirmed by different research groups [10–15]. Moreover, the catalytic performance (CH4 conversions of 20–30% at C2 selectivities of approximately 70%) makes this catalyst system a potential candidate for practical applications.
∗ Corresponding author. Tel.: +49 30 31424973. E-mail addresses: [email protected] (M. Yildiz), [email protected] (U. Simon), [email protected] (T. Otremba), [email protected] (Y. Aksu), [email protected] (K. Kailasam), [email protected] (A. Thomas), [email protected] (R. Schomäcker), [email protected], [email protected] (S. Arndt).
The formation of a surface cluster species of tetrahedral WO4 species with one W O bond and three W O Si bonds were suggested and presumed to be the active site by Jiang et al. [16]. Wu et al. showed that the tetrahedral WO4 is distorted, most probably due to the presence of Na ions [17]. It was reported that at the surface of the reduced catalyst in the presence of gas phase O2 , F-centers (an oxygen ion vacancy with two trapped electrons) exist, which are able to activate molecular O2 above 80 ◦ C, forming lattice oxygen. A redox mechanism with W O Si bonds and electron transfer between a tungsten species and F-centers was suggested [18], which was later modified to a two metal site model, including an oxygen spillover from the Mn2 O3 to the Na2 WO4 [19]. The two site model seemed to be in accord with XAFS (X-ray Absorption Fine Structure) and XPS (X-Ray Photoelectron Spectroscopy) studies before and after the reaction [20]. The exact composition and functionality of the active centers are still not clear. Palermo et al. [21] studied the oxides of Mn, Na and W supported on SiO2 alone and/or in combination. They reported that amorphous SiO2 undergoes a phase transition to highly crystalline ␣-cristobalite during calcination at 750 ◦ C, far below the usual transition temperature of 1500 ◦ C. This is caused by the presence of Na [21]. A suitable catalyst requires all three metal atoms/oxides, so it was concluded that strong synergies are present between these three metal oxides and a Na-induced crystallization of the support material occurred, converting it from an active and unselective
0920-5861/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.cattod.2013.12.024
Please cite this article in press as: M. Yildiz, et al., Support material variation for the Mnx Oy -Na2 WO4 /SiO2 catalyst, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2013.12.024
G Model CATTOD-8804; No. of Pages 10
ARTICLE IN PRESS M. Yildiz et al. / Catalysis Today xxx (2014) xxx–xxx
2
Table 1 The determined factors [fi = ki (CD4 )/ki (CH4 )] for the reaction constants obtained from CD4 and CH4 with 100 mg catalyst at 750 ◦ C. Reprinted from Yilidz, M., Arndt, S., Simon, U., Otremba, Y. A. T., Berthold, A., Görke, O., Thomas, A., Schubert, H., Schomäcker, R., 2012. Mn-Na2 WO4 /SiO2 – an industrial catalyst for methane coupling. In: Ernst, S., Balfanz, U., Buchholz, S., Lichtscheidl, J., Marchionna, M., Nees F., Santacesaria E., (Eds.), Reducing the Carbon Footprint of Fuels and Petrochemicals, Deutsche Wissenschaftliche Gesellschaft für Erdöl, Erdgas und Kohle e.V., pp. 125–132., with permission from DGMK. fi
f1 f2 f3 f4 f5
Feed gas composition (CH4 :O2 :N2 ) 2:1:4
4:1:4
8:1:4
0.6 0.7 0.3 0.5 0.2
0.5 0.7 0.4 0.5 0.2
0.6 0.7 0.5 0.4 0.2
material into an inert support material. However, the exact active site could not be revealed. Wang et al. compared Mnx Oy -Na2 WO4 /SiO2 , Mnx Oy Na2 WO4 /MgO and NaMnO4 /MgO [22]. Due to the identical catalytic performance of all three materials, it was suggested that a common active site, consisting of Na–O–Mn species exists. This is inconsistent with the suggestion of Wu et al. [17], who proposed that the W O bonds of a distorted WO4 tetrahedron could be the active site. Besides determination of the active center, the production of large amounts of this catalyst is important. Therefore, Simon and co-workers developed an up-scaled preparation of this catalyst [14], which is crucial for any practical application. A variation of the composition has been investigated, trying to find substitutes for Mn [23–26], for Na [27–30] and for W [31–33], respectively. It was found that any other elements are also active, but the combination of the oxides of Mn–Na–W was superior to all tested catalysts. Despite the research performed till now, the role of the SiO2 support material remains unclear. Moreover, there are only very few attempts using other support materials than SiO2 . Liu et al. reported the successful application of SiC as support material [34], giving a similar performance compared to SiO2 . Yu et al. reported a study of different support materials. The applied support materials were rare earth oxides (La2 O3 , CeO2 , Pr6 O11 , Nd2 O3 , Sm2 O3 , Dy2 O3 and Yb2 O3 ) and SiO2 for Na2 WO4 as active component [31]. CeO2 and Pr6 O11 were mainly total oxidation catalysts. Covered with Na2 WO4 they exhibited good C2 selectivities. However, it is unclear why typical support materials such as Al2 O3 , TiO2 , etc., were not used in these experiments. Moreover, the applied rare earth oxides are active OCM catalysts themselves, Sm2 O3 and Nd2 O3 in particular. A blank experiment with only support material was not reported in this study. Generally, it is questionable if supporting an active OCM catalyst on a material which is an active OCM catalyst itself, leads to results, which could significantly increase the knowledge on the supported material due to expectable complex interactions. In the previous work, we performed kinetic isotope effect experiments with CD4 for Mnx Oy -Na2 WO4 /SiO2 catalyst to contribute to the understanding of the reaction pathways and the origin of COx products [35]. The results of these experiments in Table 1 indicate that the most influenced reactions were the consecutive reactions which convert the C2 products towards the total oxidation products (Fig. 1). The kinetic isotope effect is as discussed in the literature [36–45] agrees with the observed results. The consecutive reactions might be suppressed by reaction engineering (e.g. applying new reactor concepts, such as membrane reactor) [46,47]. However, oxidation of CH4 molecules towards COx products, because of unselective catalysts or unselective sites of the selective catalysts,
Fig. 1. Simplified reaction network [2]. Adapted from Ref. [2].
can be prevented only by a development of a new highly selective material or an improvement of a catalyst. One issue which has not been thoroughly studied for the Mnx Oy NaWO4 catalyst is the effect of support materials other than SiO2 . Therefore, in the present manuscript we investigated a detailed variation of different metal oxides as support material for the Mnx Oy -Na2 WO4 in order to study their influence on the catalytic performance. Moreover, we studied in detail how the different components of the active materials (Mnx Oy and Na2 WO4 ) act on these support materials alone. Furthermore, a test with pure Mnx Oy -Na2 WO4 was performed to study the catalytic activity without any support material. All catalytic materials have been studied in a parallel reactor under identical conditions, allowing a direct comparison of the obtained results; this is important because results reported in different publications in literature are often not comparable, due to very different reaction conditions [48]. 2. Experimental part 2.1. Catalyst preparation The prepared catalysts, the loading of the active components and the abbreviations used in the figures and tables are presented in Table 2. Manganese is present in the form of manganese oxides or Mn-containing mixed oxides, however, the loading is calculated for pure Mn. The applied support materials (some of them e.g. La2 O3 , CaO, MgO, are also known active components themselves in the OCM [49]), their origin and specific surface area are shown in Table 3. 2.1.1. WO3 /support Supported tungsten oxide catalysts were prepared via a wet impregnation method of the different support materials (Table 3) with the appropriate aqueous solution of the (NH4 )10 H2 (W2 O7 )6 ·xH2 O (Aldrich, 99.99%, the number of hydrate was assumed four for calculations) at room temperature. Samples were dried in air overnight at 65 ◦ C. The calcination process was performed in air, heating the samples in 4 h from room temperature to 750 ◦ C (with a heating rate of approximately 3 K/min) and holding the temperature for 1 h at 750 ◦ C, consecutively. The prepared catalysts were then ground into fine powder and sieved. The particle size of the catalysts used for the reaction was ≤200 m. Table 2 Prepared catalysts, applied loadings and used abbreviations. Catalyst
Abbreviation
Blank 5 wt% WO3 /Support 5 wt% Na2 WO4 /Support 2 wt% Mnx Oy /Support 2 wt% Mnx Oy –5 wt% Na2 WO4 /Support
Support W/Support Na/W/Support Mn/Support Mn/Na/W/Support
Please cite this article in press as: M. Yildiz, et al., Support material variation for the Mnx Oy -Na2 WO4 /SiO2 catalyst, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2013.12.024
G Model CATTOD-8804; No. of Pages 10
ARTICLE IN PRESS M. Yildiz et al. / Catalysis Today xxx (2014) xxx–xxx
Table 3 The origins and specific surface areas of the used support materials. Support material
Origin of support material
BET (m2 /g)
La2 O3 CaO Al2 O3 ZrO2 SiO2 SiC MgO Fe2 O3 Fe3 O4 SrO TiO2 -rutile TiO2 -anatase
Sigma Aldrich, min. 99.9% Riedel-de Haën BASF 3% Yttria stabilized, Sigma Aldrich BASF Sigma Aldrich Sigma Aldrich, ≤ 99% Sigma Aldrich Alfa Aesar, 98% metal basis Alfa Aesar, 99.5% metal basis Aldrich BASF
3 6 338 105 120 1 124 40 39 2 20 127
WO3 (g) × 100% WO3 (g) + Support(g)
2.2.1. ICP The determination of the metal loading was performed via inductively coupled plasma atomic emission spectroscopy (ICPAES), using a Horiba Scientific ICP Model Ultima 2.
(2)
2.1.2. Na2 WO4 /support Supported sodium tungstate catalysts were prepared with the same method as described in Section 2.1.1. Na2 WO4 ·2H2 O (SigmaAldrich, min. 99%) was used for impregnating the support and the loading was calculated according to Eq. (3). Na2 WO4 (g) Na2 WO4 (wt%) = × 100% Na2 WO4 (g) + Support(g)
(3)
2.1.3. Mnx Oy /support Supported manganese oxide catalysts were prepared with the same method as described in Section 2.1.1. The applied Mnx Oy precursor was (CH3 COO)2 Mn·4H2 O (Fluka, >99%). The loading of Mnx Oy was calculated according to Eq. (4). Mnx Oy (wt%) =
Mn(g) × 100% Mn(g) + Support(g)
(4)
2.1.4. Mnx Oy -Na2 WO4 /support The Mnx Oy -Na2 WO4 active compounds on different support materials were prepared via an adapted two-step wet impregnation method [15]. First, the different support materials were impregnated with an aqueous solution containing appropriate concentration of (CH3 COO)2 Mn·4H2 O (Fluka, >99%) at room temperature. These materials were then dried in air at 65 ◦ C overnight. After that, the obtained materials were impregnated with an aqueous solution containing the appropriate concentration of Na2 WO4 ·2H2 O (Sigma-Aldrich, min. 99%) at room temperature. The preparation was then followed by the same drying, calcination and sieving procedures as described in Section 2.1.1. The loading of Na2 WO4 was calculated according to the Eq. (5) and the Mnx Oy loading was calculated according to Eq. (6). Na2 WO4 (wt%) =
Mnx Oy (wt%) =
Na2 WO4 (g) × 100% Mn(g) + Na2 WO4 (g) + Support(g)
Mn(g) × 100% Mn(g) + Na2 WO4 (g) + Support(g)
themselves. The catalytic performance of the bare untreated support materials as purchased is omitted, because this information was considered irrelevant for the purpose of this study. 2.2. Catalyst characterization
The loading of WO3 was calculated from the recipe according to Eq. (2). WO3 (wt%) =
3
(5)
2.2.2. BET The specific surface area was determined by a Micromeritics Gemini III 2375 Surface Area Analyzer, using N2 adsorption at −196 ◦ C. Before measuring, the samples were degassed at 300 ◦ C and 0.15 mbar for at least 30 min. The surface areas were calculated by the method of Brunauer, Emmett and Teller (BET). 2.2.3. XRD Powder X-Ray diffractograms were obtained (CuK␣1 radiation, wavelength 0.154 nm) using Bruker AXS D8 ADVANCE X-ray diffractometer. The angle variation was performed from 2 to 90◦ , with a step size of 0.008◦ . The diffractograms were analyzed with the program Diffrac.suite EVA. 2.3. Catalytic tests A 6-fold parallel reactor (Integrated Lab Solutions Berlin and Premex Reactor AG) was used for the determination of the catalytic performance, with packed-bed, linear, tubular reactors made of quartz glass. For each catalytic run, 50 mg catalyst were diluted with approximately 1.5 ml quartz sand (quartz sand: purchased from Merck, already washed with HCl and calcined, ca. 60% of the particles have the size 0.2–0.8 mm). Below and above the catalyst bed, a small amount of pure quartz sand was put to ensure proper heat transfer. The particle size of the catalyst was below 200 m in each experiment to exclude internal mass transfer effects. The applied reaction conditions were: 750 ◦ C reactor temperature, 60 ml/min gas flow and a feed gas composition of CH4 :O2 :N2 = 4:1:4 (methane and synthetic air as oxygen source). The analysis was performed with a gas chromatograph Agilent 7890 A, equipped with a flame ionization detector (FID), a thermal conductivity detector (TCD), an HP-PLOT/Q and an HP Molsieve column. He (Air Liquide, purity 99.999%) was used as carrier gas. The analyzed compounds were O2 , N2 , CO2 via TCD, CH4 , C2 H4 and C2 H6 via FID. N2 was used as internal standard. The reproducibility of conversion (X) and selectivity (S) is sufficient. The conversion and selectivity were calculated from a mass balance based on the inlet and outlet concentration of reactants and products, see Eqs. (7) and (8). The carbon balance was always higher than 95%. The desired reaction products of the oxidative coupling of methane are C2 H4 and C2 H6 . The selectivity of these two products was discussed as a sum, the C2 selectivity. X=
(6) SA =
2.1.5. Blank The described procedure in Section 2.1.1 was applied to the support materials exactly the same way, without adding any of the active compounds but only distilled water. These samples were prepared for blank tests for the activity of the support materials
Products
Products + Unconverted Reactant
Product A Reaction Products
(7)
(8)
Under the applied reaction conditions (reactor with quartz sand, no catalyst), as described above, the results of thermal OCM reaction were: O2 conversion: 1.2%, CH4 conversion: 0.2% at 93.8% C2 selectivity.
Please cite this article in press as: M. Yildiz, et al., Support material variation for the Mnx Oy -Na2 WO4 /SiO2 catalyst, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2013.12.024
ARTICLE IN PRESS
G Model CATTOD-8804; No. of Pages 10
M. Yildiz et al. / Catalysis Today xxx (2014) xxx–xxx
4
Table 4 The intended and determined loadings of the selected catalysts. Active components
Intended loading
Mn (wt%) Na (wt%) W (wt%)
2.00 0.78 3.13
Determined loading Mn/Na/W/SiO2
Mn/Na/W/TiO2 -rutile
Mn/Na/W/SiC
2.16 0.85 2.95
2.02 0.73 3.12
1.99 0.65 3.04
Table 5 The specific surface areas in m2 /g of the prepared fresh catalysts and spent Mnx Oy -Na2 WO4 /Support catalysts in the OCM (Pure Support: Support material without any treatment. Blank Material: Support material after applying the preparation method of the catalysts but without any active components, see Section 2.1.5., S: Support material). Support material
Pure support
Blank material
W/S
Na/W/S
Mn/S
Mn/Na/W/S
Mn/Na/W/S (spent)
La2 O3 CaO Al2 O3 ZrO2 SiO2 SiC MgO Fe2 O3 Fe3 O4 SrO TiO2 -rutile TiO2 -anatase
3 6 338 105 120 1 124 40 39 2 20 127
6 12 217 51 113 ≤1 37 3 12 3 11 29
3 7 186 50 52 ≤1 73 10 9 2 7 22
5 3 260 46 16 ≤1 21 5 4 4 8 30
10 10 222 49 104 2 22 6 10 4 10 6
9 7 242 36 11 ≤1 15 5 5 3 6 13
7 8 94 1 5 3 11 3 3 2 4 7
3. Results & discussion 3.1. Characterization 3.1.1. Elemental analysis Due to the large number of catalysts, it was not possible to determine all loadings for all prepared materials via chemical analysis. Therefore, the loading for a few chosen catalysts was determined (Table 4) and found to agree with the intended loading, within the range of experimental errors. 3.1.2. Specific surface area The BET surface areas of all prepared materials are shown in Table 5. Surface areas of the selected metal oxides vary between very low (e.g. La2 O3 , CaO, SiC, SrO) and high values (Al2 O3 , ZrO2 , SiO2 , MgO, TiO2 -anatase). The surface areas of the blank samples are lower than those of pure support materials because of the calcination process, except for La2 O3 , CaO and SrO. There is a strong reduction of the surface area upon the addition of the active component, except for La2 O3 , Al2 O3 , CaO, SiC and SrO. Especially for SiO2 , the most substantial decrease occurs in the samples which contain Na, because of the phase transformation of the amorphous SiO2 into ␣-cristobalite [3]. As expected the surface areas of most Mnx Oy -Na2 WO4 /Support catalysts decrease after 16 h time on stream, while the surface areas of CaO and SiC supported catalysts increase. The strongest reductions happen for Al2 O3 and ZrO2 supported samples. 3.1.3. Phase analysis A summary of the phase analysis of the detected XRD patterns for the pure support materials, the blank samples, the supported single metal oxides and the mixed metal oxides are presented in Tables 6–8. XRD patterns and phase analysis of all blank samples and catalysts can be found in the supporting information (Appendix A). Generally, the prepared materials are multi-phase materials, leading to numerous different phases being present. Therefore, only unambiguously identified patterns are presented. This does not exclude the presence of other phases, either in minor amounts or amorphous states. Moreover, many prepared catalysts are amorphous, prohibiting a structural analysis via XRD.
For the pure support material, as purchased and blank, the expected phases are found. Additionally, for certain compounds, such as La2 O3 , CaO or SrO, hydroxides and carbonates are also present, for MgO, only Mg(OH)2 and no carbonates are found. This could be due to storage in air. TiO2 in the rutile form, also contains TiO2 in the anatase form. Treating the purchased support material with the preparation procedure obtaining the blank samples, does not lead to any observable changes in the material. Loading only WO3 on the support materials, led again to the formation of hydroxides and carbonates for a number of support materials. The formation of WO3 is detected for Al2 O3 , SiO2 , SiC, Fe2 O3 , Fe3 O4 and TiO2 -rutile. However, for a number of support materials, the formation of tungstate compounds is found, e.g. La10 W22 O81 , CaWO4 , Ca3 WO6 , MgWO4 , Sr3 WO6 and SrWO4 . Supporting WO3 on TiO2 -anatase and ZrO2 results in amorphous materials, indicating a strong structural change of the whole support material. Similar to that of WO3 , loading of Mnx Oy on the support materials leads to the formation of hydroxides and carbonates. Mn2 O3 is observed as a separate phase for a number of support materials. The formation of mixed metal oxides is observed with Mn on only
Table 6 The detected phases for the different support and blank materials. Support materials Support material
Pure support material
Blank material
La2 O3 CaO Al2 O3 ZrO2 SiO2 SiC MgO Fe2 O3 Fe3 O4 SrO TiO2 -rutile TiO2 -anatase No support
La2 O3 , La(OH)3 CaO, Ca(OH)2 , CaCO3 Amorphous Y0.15 Zr0.85 O1.93 Amorphous SiC MgO, Mg(OH)2 Fe2 O3 Fe2 O3 SrO, SrCO3 , Sr(OH)2 ·H2 O Rutile, anatase Anatase –
La2 O3 , La(OH)3 CaO, Ca(OH)2 , CaCO3 Amorphous Y0.15 Zr0.85 O1.93 Amorphous SiC MgO Fe2 O3 Fe2 O3 SrO, SrCO3 , Sr(OH)2 ·H2 O Rutile, anatase Anatase –
Please cite this article in press as: M. Yildiz, et al., Support material variation for the Mnx Oy -Na2 WO4 /SiO2 catalyst, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2013.12.024
G Model CATTOD-8804; No. of Pages 10
ARTICLE IN PRESS M. Yildiz et al. / Catalysis Today xxx (2014) xxx–xxx
5
Table 7 The detected phases for the different supported single metal oxides. Single metal oxides Support material
W/Support
Mn/Support
La2 O3 CaO Al2 O3 ZrO2 SiO2 SiC MgO Fe2 O3 Fe3 O4 SrO TiO2 -rutile TiO2 -anatase No support
La2 O3 , La(OH)3 , La10 W22 O81 CaO, Ca(OH)2 , CaWO4 , Ca3 WO6 WO3 , amorphous Amorphous WO3 , amorphous SiC, WO3 MgO, Mg(OH)2 , MgCO3 , MgWO4 WO3 , Fe2 O3 WO3 , Fe2 O3 SrO, SrCO3 , Sr3 WO6 , SrWO4 Ti4 H2 O9.1 .9H2 O, rutile, anatase, WO3 Amorphous WO3
La2 O3 , La(OH)3 CaO, Ca(OH)2 , Mn2 O3 , Ca2 MnO4 Amorphous Y0.15 Zr0.85 O1.93 Mn2 O3 or MnMn6 SiO12 , amorphous SiC, Mn2 O3 MgO, Mg(OH)2 Fe2 O3 Fe2 O3 SrO, SrCO3 , Sr(OH)2 .H2 O Rutile, anatase Anatase, rutile, Mn2 O3 Mn2 O3
Mnx Oy /CaO. Mnx Oy does not change TiO2 -anatase and ZrO2 into amorphous materials, like WO3 . When supporting Na2 WO4 , only for SiO2 , SiC, Fe2 O3 and TiO2 rutile detectable amounts of Na2 WO4 were found. The known transition of amorphous SiO2 into ␣-cristobalite and tridymite phases is found while TiO2 -anatase is turned into an amorphous material. Supporting Mnx Oy -Na2 WO4 , hydroxides or carbonates are only found for MgO, La2 O3 and SrO. Mixed oxides of support-Mn or support-W phases are found for CaO and SrO. Al2 O3 , ZrO2 and TiO2 -anatase support materials lead the Mnx Oy -Na2 WO4 catalysts to totally amorphous materials. However on TiO2 -rutile supported catalyst, Na2 WO4 , TiO2 -anatase and TiO2 -rutile phases were detected to the contrary of the Mnx Oy -Na2 WO4 /TiO2 -anatase. Fe2 O3 phase is dominating for Mnx Oy -Na2 WO4 /Fe3 O4 , but for Fe2 O3 supported catalyst Na2 WO4 phase is also present additionally. On SiO2 and SiC supported catalysts mainly same active component phases (Na2 WO4 and Mn2 O3 ) were detected. Amorphous and unselective silica support material transformed into ␣-cristobalite phase after preparation process for the Mnx Oy Na2 WO4 /SiO2 . For the Mnx Oy /SiO2 and Mnx Oy -Na2 WO4 /SiO2 catalysts, patterns identified as Mn2 O3 might also be explained as braunite (MnMn6 SiO12 ) phase. Since patterns of Mn2 O3 and braunite are very similar and overlap, it is too difficult to distinguish whether the patterns belong to Mn2 O3 or braunite. It can be seen in the phase analysis of the catalysts (Appendix A) that in some catalysts both amorphous contributions and crystalline phases are present together. However, certain oxides, such as La2 O3 or Fe3 O4 , Al2 O3 and TiO2 -anatase do not show any distinct patterns when the material is supported on them. A reason could be either that the supported material remains amorphous or that the
support material reacts with the metal oxides forming a new compound in low concentration or in an amorphous state, prohibiting any detection via XRD. 3.2. Catalysis A good method to compare different catalysts is to determine via kinetic modeling the best performance which can be achieved over a range of operating conditions. However, this is in any case time consuming and many catalysts are instable, making a comparison difficult. In the present manuscript, we measured the catalytic activity only for one set of reaction conditions (feed gas composition, gas flow, mass of catalyst, reactor temperature) but for 16 h time on stream. Because especially for the very temperaturesensitive OCM reaction it is really a challenging issue to compare the results with reported results in literature due to very different conditions [48]. We are aware that this comparison is not the optimal way for a catalyst selection. However, the obtained results can be used to exclude catalysts with an inappropriate performance, for example due to instability or very low C2 selectivities. Moreover, an estimation if a support material is a candidate for further investigations, or not, should also be possible. 3.2.1. Stability The stability of a catalyst is a crucial point for its application potential. However, due to the large amounts of measured data, it is not feasible to present the time on stream trajectories for all prepared catalysts here. The complete data are shown in Appendix B. However, in order to provide information about the stability of the catalysts here, the first data point for the CH4 conversion, obtained approximately 60 min after reaching the desired reactor temperature, is compared with the last data point for the CH4 conversion,
Table 8 The detected phases for the different supported mixed metal oxides, Na2 WO4 and Mnx Oy -Na2 WO4 . Mixed metal oxides Support material
Na/W/Support
Mn/Na/W/Support
La2 O3 CaO Al2 O3 ZrO2 SiO2 SiC MgO Fe2 O3 Fe3 O4 SrO TiO2 -rutile TiO2 -anatase No support
La2 O3 , La(OH)3 CaO, Ca3 WO6 Amorphous Y0.15 Zr0.85 O1.93 SiO2 (cristobalite, tridymite), Na2 WO4 , Na4 WO5 SiC, Na2 WO4 MgO Fe2 O3 , Na2 WO4 Fe2 O3 SrO, Sr(OH)2 , SrCO3 , Sr(OH)2 .H2 O Rutile, anatase, Na2 WO4 Amorphous Na2 WO4
La2 O3 , La(OH)3 CaO, Ca3 WO6 , Ca2 MnO4 Amorphous Amorphous SiO2 (cristobalite, tridymite), Mn2 O3 or MnMn6 SiO12 , Na2 WO4 , Na4 WO5 SiC, Mn2 O3 , Na2 WO4 MgO, Mg(OH)2 , Mn2 O3 Fe2 O3 , Na2 WO4 Fe2 O3 SrO, SrCO3 , Sr(OH)2 .H2 O, Sr(OH)2 , Sr2 MnO4 , MnO Rutile, anatase, Na2 WO4 Amorphous Mn2 O3 , Na2 WO4
Please cite this article in press as: M. Yildiz, et al., Support material variation for the Mnx Oy -Na2 WO4 /SiO2 catalyst, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2013.12.024
ARTICLE IN PRESS
G Model CATTOD-8804; No. of Pages 10
M. Yildiz et al. / Catalysis Today xxx (2014) xxx–xxx
6
Table 9 The residual catalytic activity for CH4 conversion in % after 16 h time on stream. Support material
Blank material
W/Support
Mn/Support
Na/W/Support
Mn/Na/W/Support
La2 O3 CaO Al2 O3 ZrO2 SiO2 SiC MgO Fe2 O3 Fe3 O4 SrO TiO2 -rutile TiO2 -anatase No support
98 80 82 100 22 60 42 16 75 90 74 54 –
51 55 125 80 67 200 200 32 16 85 83 100 100
104 94 103 105 36 64 101 40 77 102 57 33 74
68 79 101 49 112 88 160 63 83 72 80 180 100
88 98 108 67 89 121 108 82 89 115 82 38 71
obtained after approximately 16 h time on stream. The result is the residual catalytic activity. The calculated data is presented in Table 9. The catalytic activity of some catalysts, like WO3 /SiC, Na2 WO4 /TiO2 -anatase, Na2 WO4 /MgO increases with time on stream. This is probably caused by structural changes during the reaction. In Figs. 2–4, XRD patterns of some fresh and spent catalysts are shown for catalysts with a strongly increased catalytic activity after 16 h time on stream. Some of them demonstrate structural changes different from fresh catalysts. It was detected for the XRD patterns of WO3 /SiC that the ratio of SiC to WO3 has changed. For the Na2 WO4 /TiO2 -anatase, a phase transformation from amorphous state into anatase, rutile and Na2 WO4 phases occurred during reaction, showing very strong crystallization. The patterns of Na2 WO4 /MgO catalysts before and after reaction are exactly the same, no changes are observed. The stability of the blank samples was high, except for SiO2 , MgO, Fe2 O3 and TiO2 -anatase. WO3 /Fe2 O3 and WO3 /Fe3 O4 catalysts strongly deactivated, while the activity of the Al2 O3 , SiC and MgO supported WO3 samples increased. SrO supported Mnx Oy catalyst was stable during the reaction whereas Mnx Oy /SiO2 , Mnx Oy /TiO2 -anatase and Mnx Oy /Fe2 O3 lost their activity substantially. Although supported Na2 WO4 catalysts were stable, except for the ZrO2 supported sample, with the combination of three active components, nearly all supported catalysts were highly stable except Mnx Oy -Na2 WO4 /ZrO2 and Mnx Oy -Na2 WO4 /TiO2 anatase which strongly deactivated. Among all catalysts Al2 O3 and MgO supported samples did not exhibit any sign of deactivation. However, the unselective performance of Al2 O3 supported catalysts is an important disadvantage with respect to the
desired C2 products. Furthermore no interpretation can be derived for the La2 O3 , Mnx Oy /La2 O3 , Mnx Oy -Na2 WO4 /La2 O3 , Mnx Oy /CaO, Mnx Oy -Na2 WO4 /CaO, Mnx Oy /Al2 O3 , Mnx Oy -Na2 WO4 /Al2 O3 , ZrO2 , Mnx Oy /ZrO2 and Mnx Oy /MgO catalysts because of full O2 conversion (Appendix B). 3.2.2. Comparison of catalytic performances For the blank support material, WO3 /Support, Na2 WO4 /Support, Mnx Oy /Support and Mnx Oy -Na2 WO4 /Support the catalytic performances after approximately 16 h are presented in Fig. 8. In Fig. 5 the CH4 conversions are presented, in Fig. 6 the C2 selectivities and in Fig. 7 the C2 yields. The CH4 conversions of Al2 O3 , SiO2 , SiC, Fe2 O3 , SrO, TiO2 -rutile and TiO2 -anatase support materials are rather poor, on the other hand La2 O3 , CaO, ZrO2 , MgO and Fe3 O4 are active themselves, but C2 selectivities of these active support materials are very low. It becomes evident, that certain support materials (i.e. La2 O3 , CaO or MgO) are by no means inert carriers. In particular, La2 O3 exhibits a high activity, which is hardly changed upon the loading of the metal oxides, except for WO3 . However, most of the other support materials show a rather low performance. Supporting WO3 , does not lead to any significant CH4 conversion. In fact, it does decrease the conversion of the active support materials. This is especially prominent for La2 O3 , ZrO2 , MgO and Fe3 O4 , while it is not distinct for SiO2 [50], SiC and CaO support materials. WO3 improved slightly the activity only for SrO. Since most of the supported WO3 catalysts have low CH4 conversion, their C2 selectivities are very high. However, supporting Na2 WO4 , the CH4 conversion increases compared to supporting WO3 . La2 O3 , CaO and Al2 O3 supported Na2 WO4 catalysts are highly active in comparison to the other
Fig. 2. XRD patterns of W/SiC before and after 16 h time on stream (䊉: WO3 ,
SiC).
Please cite this article in press as: M. Yildiz, et al., Support material variation for the Mnx Oy -Na2 WO4 /SiO2 catalyst, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2013.12.024
ARTICLE IN PRESS
G Model CATTOD-8804; No. of Pages 10
M. Yildiz et al. / Catalysis Today xxx (2014) xxx–xxx
7
Fig. 3. XRD patterns of Na/W/TiO2 -anatase before and after 16 h time on stream (: Na2 WO4 , : TiO2 -anatase, : TiO2 -rutile).
Fig. 4. XRD patterns of Na/W/MgO before and after 16 h time on stream (: MgO).
Na2 WO4 /Support samples, but their C2 selectivity is very poor. When Al2 O3 , SiO2 and SiC supported Na2 WO4 catalysts are compared to the blank samples of the corresponding support materials, their CH4 conversions increase explicitly in contrast to the other catalysts.
Supporting only Mnx Oy , leads to relatively high CH4 conversion. For Mnx Oy on SiO2 and MgO, this has already been reported [51–53]. The CH4 conversions over La2 O3 , CaO, Al2 O3 , ZrO2 , MgO and Fe3 O4 supported manganese oxide catalysts are outstanding, some of them are even higher than conversion of the
25
CH4 Conversion [%]
20
15
10
5
0 Lanthana Calcium oxide
Alumina Zirconia Silica Silicon carbide
Mn/Na/W/Support Mn/Support Na/W/Support W/Support Support
Magnesia Iron (II,III) Iron (III) oxide oxide
Strona TitaniaRule
Titania-
Anatase Without Support
Fig. 5. The CH4 conversion determined after approximately 16 h time on stream for the different materials. Reaction conditions: 750 ◦ C, 50 mg catalyst, flow rate of 60 ml/min and feed gas composition of CH4 :O2 :N2 = 4:1:4.
Please cite this article in press as: M. Yildiz, et al., Support material variation for the Mnx Oy -Na2 WO4 /SiO2 catalyst, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2013.12.024
ARTICLE IN PRESS
G Model CATTOD-8804; No. of Pages 10
M. Yildiz et al. / Catalysis Today xxx (2014) xxx–xxx
8
90 80
C2-Selecvity [%]
70 60 50 40 30 20 10 0 Lanthana Calcium oxide
Alumina Zirconia
Mn/Na/W/Support Mn/Support Na/W/Support W/Support Support
Silica Silicon carbide
Magnesia Iron (II,III) Iron (III) oxide oxide
Strona TitaniaRule
TitaniaAnatase
Without Support
Fig. 6. The C2 selectivity determined after approximately 16 h time on stream for the different materials. Reaction conditions: 750 ◦ C, 50 mg catalyst, flow rate of 60 ml/min and feed gas composition of CH4 :O2 :N2 = 4:1:4.
Mnx Oy -Na2 WO4 /Support catalysts. However the C2 selectivities are mostly drastically reduced, total oxidation is dominating. Supporting Mnx Oy -Na2 WO4 , it becomes evident that the combination of these three metal oxides results in the best catalytic performance, c.f. C2 yield. La2 O3 , CaO, Al2 O3 , ZrO2 and SiO2 supported trimetallic catalysts show the highest CH4 conversions amongst all Mnx Oy -Na2 WO4 /Support samples. While the wellknown catalyst Mnx Oy -Na2 WO4 /SiO2 is highly selective towards C2 products, for the La2 O3 , CaO, Al2 O3 , ZrO2 supported trimetallic catalysts total oxidation products are prevailing. Considering the C2 yield, presented in Fig. 7, it can be seen that SiO2 is not the only suitable support material. Support materials, such as La2 O3 , Al2 O3 , CaO or ZrO2 exhibit a similar C2 yield than SiO2 , but their CH4 conversion is much higher at much lower C2 selectivity. This catalytic behavior is not beneficial for a practical application. However, TiO2 -rutile and SiC exhibit a similar performance than SiO2 , with relatively low conversions at high C2 selectivities.
Regarding catalyst components without support material, Mnx Oy shows substantial activity itself with poor C2 selectivity, whereas the CH4 conversion over WO3 and Na2 WO4 is extremely low. Furthermore, it is interesting to note, that Mnx Oy -Na2 WO4 without support material also displays a considerable activity and selectivity, questioning the actual contribution of the support material. The results of the catalytic tests of all tested materials are summarized in a conversion selectivity diagram (Fig. 8) by comparing the obtained data points with the trajectory of a reference Mnx Oy Na2 WO4 /SiO2 catalyst, which is described in more detail in [14]. If only the number of active sites changes, the resulting data points would fall on the reference trajectory. However, if the nature of active sites is changed, the data points might be below (aggravation) or above (improvement) the reference trajectory, indicating possible changes of the catalytic properties. WO3 /Support and Na2 WO4 /Support catalysts are concentrated mostly in the low CH4 conversion region with 0–3%. Mnx Oy /Support
10 9 8 C2-Yield [%]
7 6 5 4 3 2 1 0 Lanthana
Calcium oxide
Alumina
Zirconia
Silica
Silicon carbide
Magnesia
Iron (II,III) oxide
Iron (III) oxide
Strona
TitaniaRule
Mn/Na/W/Support Mn/Support Na/W/Support W/Support Support TitaniaAnatase
Without Support
Fig. 7. The C2 yield determined after approximately 16 h time on stream for the different materials. Reaction conditions: 750 ◦ C, 50 mg catalyst, flow rate of 60 ml/min and feed gas composition of CH4 :O2 :N2 = 4:1:4.
Please cite this article in press as: M. Yildiz, et al., Support material variation for the Mnx Oy -Na2 WO4 /SiO2 catalyst, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2013.12.024
G Model CATTOD-8804; No. of Pages 10
ARTICLE IN PRESS M. Yildiz et al. / Catalysis Today xxx (2014) xxx–xxx
9
Fig. 8. Active components on the different support materials were compared in an X-S diagram for the oxidative coupling of methane. Results were recorded after approximately 16 h time on stream for each data point. Reaction conditions: 750 ◦ C, 50 mg catalyst, flow rate of 60 ml/min and feed gas composition of CH4 :O2 :N2 = 4:1:4.
catalysts and bulk materials (pure support materials) exhibit high conversions but they are mostly unselective towards C2 products. Even if they showed high or low CH4 conversion, nearly all data points are below the reference trajectory. For Mnx Oy -Na2 WO4 supported on SrO, MgO, CaO, La2 O3 , Al2 O3 and ZrO2 , the performance is also worse in comparison to the reference trajectory. Despite their often high CH4 conversion, they usually lack C2 selectivity. Moreover, some of these materials strongly deactivate, e.g. ZrO2 . This means that the prepared catalysts contain active sites with higher tendency towards COx than the reference catalysts. The materials supported on TiO2 (rutile or anatase), Fe2 O3 , Fe3 O4 , SiC and SiO2 are either on the reference trajectory or well above, indicating that they could also be suitable support materials. If stability of the catalytic performance is additionally taken into account, then TiO2 -anatase has actually to be excluded. However, without knowing the exact nature of the active sites it is possible to say, that they are at least as good if not better than that of the reference Mnx Oy -Na2 WO4 /SiO2 . That could mean, that their active sites are either less unselective or less active for the consecutive combustion of the C2 products. 4. Summary & conclusion For the Mnx Oy -Na2 WO4 , a variety of different support materials has been studied. Most of the materials show a catalytic performance inferior to the reference Mnx Oy -Na2 WO4 /SiO2 and/or deactivate, although Mnx Oy -Na2 WO4 without a support material exhibits also significant catalytic activity, questioning suggestions of the active centers like W O Si bonds. However, since quartz sand has been used to dilute the catalyst, it cannot be excluded that the active component has smeared over the SiO2 particles, although it seems unlikely. La2 O3 and CaO exhibit a high activity as pure support materials, making them unattractive for fundamental research, since it will be difficult to distinguish the contribution of the support material and the active phase. Among the remaining materials only SiO2 , SiC, TiO2 -rutile and the Fe2 O3 -based support materials show catalytic performances higher than the reference Mnx Oy -Na2 WO4 /SiO2 . Considering the C2 yield, SiO2 results in the best catalyst, however, the other materials
have active sites which are as suitable (SiC, Fe-based) or even more suitable (TiO2 -rutile) for the formation of C2 , but seem to be present in lower numbers. Acknowledgments This work is part of the Cluster of Excellence “Unifying Concepts in Catalysis” coordinated by the Technische Universität Berlin. Financial support by the Deutsche Forschungsgemeinschaft (DFG) within the framework of the German Initiative for Excellence is gratefully acknowledged. We would like to thank Prof. Dr. Helmut Schubert† for his precious contributions. Mr. Yildiz is obliged to the Ministry of Education of the Republic of Turkey for fundings. We also thank Harald Link for ICP measurements and Maria Unterweger for XRD measurements. We are also obliged to Mr. Schiele and the workshop for their support with the equipment. Moreover, we would like to thank Evonik and BASF for supplying support material. Appendices A and B. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/ 10.1016/j.cattod.2013.12.024. References [1] A.M. Maitra, I. Campbell, R.J. Tyler, Appl. Catal. A Gen. 85 (1992) 27–46. [2] E. Kondratenko, M. Baerns, Oxidative coupling of methane, in: G. Ertl, H. Knözinger, F. Schüth, J. Weitkamp (Eds.), Handbook of Heterogeneous Catalysis, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, 2008, pp. 3010–3023. [3] S. Arndt, T. Otremba, U. Simon, M. Yildiz, H. Schubert, R. Schomäcker, Appl. Catal. A Gen. 425–426 (2012) 53–61. [4] S. Arndt, G. Laugel, S. Levchenko, R. Horn, M. Baerns, M. Scheffler, R. Schlögl, R. Schomäcker, Catal. Rev. 53 (2011) 424–514. [5] U. Zavyalova, M. Holena, R. Schlögl, M. Baerns, ChemCatChem 12 (2011) 1935–1947. [6] H. Liu, X. Wang, D. Yang, R. Gao, Z. Wang, J. Yang, J. Nat. Gas Chem. 17 (2008) 59–63. [7] M. Makri, C.G. Vayenas, Appl. Catal. A Gen. 244 (2003) 301–310. [8] S. Jaˇso, S. Sadjadi, H.R. Godini, U. Simon, S. Arndt, O. Görke, A. Berthold, H. Arellano-Garcia, H. Schubert, R. Schomäcker, G. Wozny, J. Nat. Gas Chem. 21 (2012) 534–543.
Please cite this article in press as: M. Yildiz, et al., Support material variation for the Mnx Oy -Na2 WO4 /SiO2 catalyst, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2013.12.024
G Model CATTOD-8804; No. of Pages 10 10 [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31]
ARTICLE IN PRESS M. Yildiz et al. / Catalysis Today xxx (2014) xxx–xxx
S. Jaˇso, H. Arellano-Garcia, G. Wozny, J. Chem. Eng. 171 (2011) 255–271. X. Fang, S. Li, J. Lin, Y. Chu, J. Mol. Catal. (China) 6 (1992) 427–433. X. Fang, S. Li, J. Lin, J. Gu, D. Yan, J. Mol. Catal. (China) 6 (1992) 255–262. X. Wang, J. Zhang, D. Yang, C. Zhang, J. Lin, S. Li, Shiyou Huagong 26 (1997) 361–367. J. Lin, J. Gu, D. Yang, C. Zhang, Y. Yang, Y. Chu, S. Li, Shiyou Huagong 24 (1995) 293–298. U. Simon, O. Görke, A. Berthold, S. Arndt, R. Schomäcker, H. Schubert, J. Chem. Eng. 168 (2011) 1352–1359. S. Pak, P. Qiu, J. Lunsford, J. Catal. 179 (1998) 222–230. Z. Jiang, C. Yu, X. Fang, S. Li, H. Wang, J. Phys. Chem.-US 97 (1993) 12870–12875. J. Wu, S. Li, J. Phys. Chem.-US 99 (1995) 4566–4568. J. Wu, S. Li, J. Niu, X. Fang, Appl. Catal. A Gen. 124 (1995) 9–18. Z. Jiang, H. Gong, S. Li, Stud. Surf. Sci. Catal. 112 (1997) 481–490. Y. Kou, B. Zhang, J. Niu, S. Li, H. Wang, T. Tanaka, S. Yoshida, J. Catal. 173 (1998) 399–408. A. Palermo, J. Vazquez, A. Lee, M. Tikhov, R. Lambert, J. Catal. 177 (1998) 259–266. D. Wang, M. Rosynek, J. Lunsford, J. Catal. 155 (1995) 390–402. A. Malekzadeh, A. Khodadadi, M. Abedini, M. Amini, A. Bahramian, A. Dalai, Catal. Commun. 2 (2001) 241–247. A. Malekzadeh, A. Dalai, A. Khodadadi, Y. Mortazavi, Catal. Commun. 9 (2008) 960–965. Z. Gholipour, A. Malekzadeh, R. Hatami, Y. Mortazavi, A. Khodadadi, J. Nat. Gas Chem. 19 (2010) 35–42. A. Malekzadeh, M. Abedini, A. Khodadadi, M. Amini, H. Mishra, A. Dalai, Catal. Lett. 84 (2002) 45–51. S. Ji, T. Xiao, S. Li, L. Chou, B. Zhang, C. Xu, R. Hou, A. York, M. Green, J. Catal. 220 (2003) 47–56. A. Palermo, J.H. Vazquez, R. Lambert, Catal. Lett. 68 (2000) 191–196. A. Malekzadeh, A. Khodadadi, A. Dalai, M. Abedini, J. Nat. Gas Chem. 16 (2007) 121–129. A. Tyunyaev, G. Nipan, T. Kol’tsova, A. Loktev, V. Ketsko, A. Dedov, I. Moiseev, Russ. J. Inorg. Chem. 54 (2009) 664–667. Z. Yu, X. Yang, J. Lunsford, M. Rosynek, J. Catal. 154 (1995) 163–173.
[32] S. Mahmoodi, M. Ehsani, S. Goreishi, J. Ind. Eng. Chem. 16 (2010) 923–928. [33] S. Hou, Y. Cao, W. Xiong, H. Liu, Y. Kou, Ind. Eng. Chem. Res. 45 (2006) 7077–7083. [34] H. Liu, D. Yang, R. Gao, L. Chen, S. Zhang, X. Wang, Catal. Commun. 9 (2008) 1302–1306. [35] M. Yildiz, S. Arndt, U. Simon, Y. Aksu, T. Otremba, A. Berthold, O. Görke, A. Thomas, H. Schubert, R. Schomäcker, DGMK-Tagungsbericht 2012–3 (2012) 125–132. [36] N.W. Cant, C.A. Lukey, P.F. Nelson, R.J. Tylerb, J. Chem. Soc., Chem. Commun. (1988) 766–768. [37] N.W. Cant, C.A. Lukey, P.F. Nelson, J. Catal. 124 (1990) 336–348. [38] N.W. Cant, E.M. Kennedy, J. Phys. Chem. 97 (1993) 1445–1450. [39] C.A. Mims, R.B. Hall, K.D. Rose, G.R. Myers, Catal. Lett. 2 (1989) 361–368. [40] C.A. Mims, R. Mauti, J. Phys. Chem. 98 (1994) 13357–13372. [41] P.F. Nelson, C.A. Lukey, N.W. Cant, J. Phys. Chem. 92 (1988) 6176–6179. [42] P.F. Nelson, C.A. Lukey, N.W. Cant, J. Catal. 120 (1989) 216–230. [43] P.F. Nelson, J. Phys. Chem. 94 (1990) 3756–3761. [44] P.F. Nelson, E.M. Kennedy, N.W. Cant, Isotopic labelling studies of the mechanism of the catalytic oxidative coupling of methane, in: A. Holmen et al. (Eds.), Natural Gas Conversion, Elsevier Science Publishers B.V., 1991, pp. 89–95. [45] C. Shi, M. Xu, M.P. Rosynek, J.H. Lunsford, J. Phys. Chem. 97 (1993) 216–222. [46] S. Bhatia, C.Y. Thien, A.R. Mohamed, J. Chem. Eng. 148 (2009) 525–532. [47] T.P. Tiemersma, A.S. Chaudhari, F. Gallucci, J.A.M. Kuipers, M.S. Annaland, Chem. Eng. Sci. 82 (2012) 232–245. [48] M. Baerns, J. Ross, Catalytic chemistry of methane conversion, in: J.M. Thomas, K.I. Zamaraev (Eds.), Perspectives in Catalysis, Blackwell Science, IUPAC International Union of Pure and Applied Chemistry, 1992, pp. 315–335. [49] Y. Amenomiya, V.I. Birss, M. Goledzinowski, J. Galuszka, A.R. Sanger, Catal. Rev.Sci. Eng. 32 (3) (1990) 163–227. [50] J. Wu, S. Li, J. Phys. Chem. 99 (1995) 4566–4568. [51] A. Gaffney, C. Jones, J. Leonard, J. Sofranko, H. Withers, Stud. Surf. Sci. Catal. 38 (1987) 523–529. [52] J. Sofranko, J. Leonard, C. Jones, J. Catal. 103 (1987) 302–310. [53] J. Sofranko, J. Leonard, C. Jones, A. Gaffney, H. Withers, Catal. Today 3 (1988) 127–135.
Please cite this article in press as: M. Yildiz, et al., Support material variation for the Mnx Oy -Na2 WO4 /SiO2 catalyst, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2013.12.024