- Email: [email protected]
Effect of additives doping on catalytic properties of Mg3(VO4)2 catalysts in oxidative dehydrogenation of cyclohexane
G Model CATTOD-8246; No. of Pages 7
ARTICLE IN PRESS Catalysis Today xxx (2012) xxx–xxx
Contents lists available at SciVerse ScienceDirect
Catalysis Today journal homepage: www.elsevier.com/locate/cattod
Effect of additives doping on catalytic properties of Mg3 (VO4 )2 catalysts in oxidative dehydrogenation of cyclohexane M. Jin a , P. Lu a , G.X. Yu a,∗ , Z.M. Cheng b , L.F. Chen c , J.A. Wang c a
School of Chemistry and Environmental Engineering, Jianghan University, Wuhan, Hubei 430056, PR China State Key Laboratory of Chemical Engineering, UNILAB Research Centre of Chemical Reaction Engineering, East China University of Science and Technology, Shanghai 200237, PR China c Escuela Superior de Ingeniería Química e Industrias Extractivas (ESIQIE), Instituto Politécnico Nacional, Av. Politécnico S/N, Col. Zacatenco, 07738 México, D.F., Mexico b
a r t i c l e
i n f o
Article history: Received 7 February 2012 Received in revised form 21 August 2012 Accepted 1 September 2012 Available online xxx Keywords: Cyclohexane Oxidative dehydrogenation Alkali/alkaline earth metal Cyclohexene Modified catalyst
a b s t r a c t Effects of additives comprising alkali and alkaline earth metals (Li, Na, K and Ca) introduced to Mg3 (VO4 )2 on their structures, physicochemical properties and the catalytic behaviors in the oxidative dehydrogenation of cyclohexane were investigated. The characterization and the experimental results showed that the additive does not affect markedly the structure of the catalyst, but blocks the active sites and hinders the reducibility of the active species thus decreasing the catalytic activity. Moreover, the additive could enhance the selectivity to cyclohexene by enhancing the nucleophilicity, the redox property as well as the type and number of the oxygen species on the catalyst surface. The doped catalysts could give the catalytic activity and selectivity to cyclohexene in the orders of K- < Na- ≈ Ca- < Li- < non-doped and non-doped < Li< Na- < K- < Ca-, respectively. Among the doped catalysts, Ca–Mg3 (VO4 )2 catalyst demonstrated a yield of cyclohexene of 8.2%, which was a promotion compared with 6.4% of Mg3 (VO4 )2 . © 2012 Published by Elsevier B.V.
1. Introduction The oxidative dehydrogenation (ODH) of cyclohexane to cyclohexene is a challenging pursuit in both economic and scientific terms. Previous investigations of the catalysts used in the ODH reaction of cyclohexane demonstrated Mg3 (VO4 )2 has been considered as a favorite catalyst for producing a promising yield of cyclohexene [1–3]. However, a major challenge in the commercial development of cyclohexane ODH is further improvement in cyclohexene yield since the main cause of the selectivity limitation arises from the conversion–selectivity relationship [4,5]. One of the solutions for the catalyst is to ensure easy desorption of intermediate product cyclohexene from the catalyst surface. While, the difficulty or the ease of cyclohexene desorption depends on the properties of the catalyst surface. Therefore, an effective way for improving the selectivity is to modify the catalyst surface by the introduction of additive, especially the alkali/alkaline earth metal [6–11]. The introduction of alkali/alkaline earth metal to V2 O5 /Al2 O3 [6,7], V2 O5 /TiO2 [8,9], V2 O5 /SiO2 , V2 O5 /MgO [10,11] and V Mg O catalysts [12] has been carried out in the ODH of lower alkanes to produce the corresponding alkenes. Most of the additives, such as Ca, Li, Na and Cs, played a promoting effect on
∗ Corresponding author. Tel.: +86 027 84226806. E-mail address: [email protected] (G.X. Yu).
the selectivity to the objective alkene which could be explained by modification of acid–base properties and the reducibility of the active phases by alkalis. While the other additive, such as K, was found to decrease the selectivity to the objective product since K was found to inhibit the formation of Mg3 (VO4 )2 from a mixture of MgO and NH4 VO3 and/or trace K was segregated onto the surface of magnesium vanadates [13–16]. Compared with the ODH of lower alkanes, Patcas et al. [17,18] only investigated the effect of alkali metal addition on the catalytic activity in the ODH of cyclohexane. They first modified NiO/Al2 O3 catalyst by alkali (Na, K and Cs) and found no promotion effect owing to a lower active species content and a larger amount of bulk nickel aluminate leaded by the modification. Then they modified the egg-shell NiO/Al2 O3 catalyst by Li and proposed Li could enhance the catalytic activity and the selectivity to cyclohexene by decreasing the formation of bulk nickel aluminates. Besides, it is a pity that few or no studies have been reported on the influence of alkali/alkaline earth metal doping on catalytic property of Mg3 (VO4 )2 in the ODH of cyclohexane. In this work, the catalytic performance of the alkali/alkaline metal (Li, Na, K and Ca) doping on Mg3 (VO4 )2 was studied in the ODH of cyclohexane. N2 -adsorption, XRD, H2 -TPR and XPS techniques were applied to give further information on the structural, the reducibility of active species as well as the type and distribution of the surface oxygen species of the catalyst. Furthermore, the relationship among the reducibility of the active species, the type and
0920-5861/$ – see front matter © 2012 Published by Elsevier B.V. http://dx.doi.org/10.1016/j.cattod.2012.09.019
Please cite this article in press as: M. Jin, et al., Effect of additives doping on catalytic properties of Mg3 (VO4 )2 catalysts in oxidative dehydrogenation of cyclohexane, Catal. Today (2012), http://dx.doi.org/10.1016/j.cattod.2012.09.019
G Model CATTOD-8246; No. of Pages 7
ARTICLE IN PRESS M. Jin et al. / Catalysis Today xxx (2012) xxx–xxx
2
distribution of the surface oxygen species and the catalytic behavior of the catalysts were investigated in detail.
Ca-Mg3(VO 4)2
2. Experimental K-Mg 3(VO 4)2
2.1. Catalyst preparation Mg3 (VO4 )2 was synthesized via the citrate complexation method, which has been described in detail in the literature [2,3]. First, a transparent solution of Mg(NO3 )2 ·6H2 O and NH4 VO3 with Mg/V atomic ratio of 3/2 was prepared. Then citric acid was added in the mixing solution with a 20% excess over the number of ionic equivalents of cations. Subsequently, ammonia was added to adjust the pH value of 4.8. The solution obtained was evaporated at 343 K to form a gel, and the gel was dried subsequently in a vacuum oven overnight at 343 K. Finally, the resulting solid was ground and heated up to 673 K for 18 h at a constant rising rate of 1 K/min to decompose the gel precursor, then calcined in the air at 823 K for 6 h. The final sample was crushed and sieved to obtain a particle size of 20–40 mesh. The doped samples were denoted as A-, where A were Li, Na, K and Ca, respectively. In this work, the atomic ratio A/V of 0.1 was adopted, which was based on the following aspects [19]: (1) the K/V ratio of 0.1 was shown in previous studies as optimal for V-based catalysts; (2) the A/V ratio of 0.1 can avoid formation of mixed A V O bulk compounds and/or trace A onto the catalyst surface; (3) the evident modification of the catalytic properties can be observed at the A/V ratio of 0.1. The doped samples were prepared by incipient impregnation. The additive was introduced by adding the appropriate amount of the corresponding nitrate solution onto Mg3 (VO4 )2 particles. After impregnation, the obtaining sample was dried at 393 K overnight and calcined in the air at 823 K for 6 h. 2.2. Catalyst characterization BET surface area of the as-prepared sample was measured by N2 physisorption at 77.3 K (ASAP 2010, Micromeritics). The XRD pattern was measured using Cu K␣ radiation (D/MAX 2550 X-ray diffractometer) at 40 kV and 40 mA. Actual composition of the sample was determined on an ICP-AES equipment (Varian 710 ES). XPS measurements were carried out on a PHI 5300/ESCA system with Al K␣ as a radiation source (Perkin-Elmer), the correction of the surface charge electricity effect was using C1s (284.60 eV) and the range of the spectrum of scanning energy was from 0 to 1200 eV. XPSPEAK software was used for the deconvolution and analysis the XPS characterization results. The temperature-programmed reduction (H2 -TPR) was measured on a Micromeritics AutoChem II 2920 apparatus. Before reduction, about 0.20 g sample was pre-treated in air at 423 K for 40 min to remove the adsorbed water, followed by cooling to 313 K under Ar flow (30 ml/min). Subsequently, the sample was reduced with a 10 vol.% H2 /Ar mixture (30 ml/min) by temperature programming from 313 to 1273 K at a rate of 10 K/min. 2.3. Catalytic test The catalyst activity test was carried out isothermally at atmospheric pressure in a fixed-bed microreactor made of stainless steel with an inner diameter of 9 mm. About 0.5 g of as-prepared catalyst diluted with inert quartz sands at a mass ratio of 1:4 was loaded into the reactor so as to ensure a uniform catalyst distribution. In each run, the catalyst was first pretreated in an air stream of 50 ml/min at 773 K for 1 h, and then adjusted to the reaction conditions. The system was allowed to stabilize for 1.5 h before the first product sample was taken for analysis. The liquid products were analyzed
Na-Mg3(VO 4)2
Li-Mg 3(VO 4)2
Mg 3(VO4)2
10
20
30
40
50
60
70
80
2 Fig. 1. XRD patterns of the catalysts.
by a HP 6890 gas chromatography equipped with a PEG 20,000 column. 3. Results and discussion 3.1. Textural and structural properties Examination of the surface areas of the samples reveals that the doping treatment produces a modification during the preparation procedure. The doped catalysts show a lower BET surface area with respect to the non-doped one (Table 1); meanwhile, the results show that the surface area decreases along with the ion size increasing, which could be explained by the formation of O− A+ on the catalyst surface [20,21]. As shown in Fig. 1, no features of crystalline additive vanadates or additive metal involving material can be observed in the doped catalysts, which is different from the research of Kung and Kung [13,14]. The XRD characterization result suggests that the additive coordinates with the surface active species in modifying its structure instead of forming bulk compounds. 3.2. Characterization results of TPR As shown in Fig. 2, the temperature of maximum hydrogen consumption (Tmax ) is 970 K for Mg3 (VO4 )2 , while it shifts to 979 (Li-), 1002 (Na-), 1054 (K-), and 1023 K (Ca-) for the doped catalysts, respectively. The shift to higher temperature could be due to the stronger interaction between the active species and the additive, which is in agreement with the views of Valenzuela et al. [12], and Balderas-Tapia et al. [22]. It is obviously that the reducibility of the active species decreases in the order of Mg3 (VO4 )2 > Li- > Na- > Ca> K-. In addition, based on the previous researches [24,25], Tmax ranged from 873 to 973 K is responsible for the adsorption of the incompletion reduction oxygen Oı− (0 < ı < 2), and the value ranged from 973 to 1173 K is involved in the adsorption of lattice oxygen. Therefore, the oxygen species on the doped-catalysts surface are the mixture of the lattice oxygen and Oı− (0 < ı < 2) species.
Please cite this article in press as: M. Jin, et al., Effect of additives doping on catalytic properties of Mg3 (VO4 )2 catalysts in oxidative dehydrogenation of cyclohexane, Catal. Today (2012), http://dx.doi.org/10.1016/j.cattod.2012.09.019
ARTICLE IN PRESS
G Model CATTOD-8246; No. of Pages 7
M. Jin et al. / Catalysis Today xxx (2012) xxx–xxx
3
Table 1 Catalytic performance of cyclohexane ODH over the catalysts. SBET (m2 /g)
Catalysts
Mg3 (VO4 )2
Li-
Na-
K-
Ca-
ICP (mol) Mg/V/A
Temperature (K)
Conversion (%)
Selectivity (%)
Cyclohexane
Cyclohexene
COx
Yield (%) Cyclohexane
7.8 15.5 21.1
65.7 41.5 27.2
16.2 20.7 28.2
5.12 6.43 5.73
17.5
1.5/1/–
673 723 773
15.5
1.5/1/0.1
673 723 773
5.8 11.5 18.0
77.6 61.4 42.1
11.2 17.3 25.3
4.50 7.06 7.57
14.8
1.5/1/0.1
673 723 773
4.9 10.8 17.1
79.9 65.9 43.5
10.3 16.9 24.1
3.92 7.12 7.43
14.3
1.5/1/0.1
673 723 773
4.2 10.0 15.4
81.8 69.6 49.5
9.9 16.5 23.6
3.43 6.96 7.62
15.2
1.5/1/0.1
673 723 773
4.7 10.6 16.5
82.1 69.5 49.9
10.2 16.8 23.5
3.86 7.37 8.23
Experimental condition: cyclohexane flow rate of 0.107 mol/h, air flow rate of 100 ml/min, atmospheric pressure.
Even though quantification of H2 -TPR profiles does not give precise information about V oxidation states, it can provide some useful indications. According to the total hydrogen consumption, the average valences of vanadium species in the samples are probably +4.69 (K-), +4.71 (Na-), +4.72 (Li-), +4.70 (Ca-) and +4.74 (Mg3 (VO4 )2 ), respectively. It is indicated that the type of the additive has an effect on the V oxidation states, which is in same with the view proposed by Sloczynski [23]. In addition, no feature of the reduction peaks of crystalline additive vanadates or additive metal involving material, which confirms the result of the XRD characterization. 3.3. Characterization results of XPS When the additive is an ion with a larger ionic radius than that of V5+ /V4+ /Mg2+ , it might enter into the interstitial site of the catalyst lattice rather than replace the cations in the bulk crystal [26]. In order to investigate the composition of the catalyst surface, possible Ca-Mg3(VO4)2
TCD Signal/a.u.
K-Mg3(VO4)2
Na-Mg3(VO4)2
3.4. Catalytic performances
Li-Mg3(VO4)2
Mg3(VO4)2
273
473
variations in the vanadium oxidation state, changes of the electronic density of V O and the type and the distribution of the oxygen species, XPS were recorded shown in Figs. 3 and 4 and Table 2. Fig. 3 demonstrated XPS spectra of V2p on the catalysts surface. The assignment of bands is a complex problem, as can be seen from the different results of V4+ and V5+ reported in literature [27,28]. For V2 O4 , binding energy values of 515.5, 515.7 and 516.3–516.6 eV have been assigned, and for V2 O5 , values between 517.1 and 518.0 eV have been assigned. It can be suggested that bands with a high binding energy value can be assigned to V5+ while those with a low binding energy value may be attributed to V4+ . The results listed in Table 2 of the peak resolution of V species show that there is a small shift to lower binding energies in V2p 3/2 (V 2p 3/2 ) and V2p 1/2 (V 2p 1/2 ) of the doped catalysts with respect to the nondoped one, which indicates that there should be an increase of the electron density around vanadium atoms. Considering the limitation in the assignment of the bands, an approximate value of the atomic ratios of V5+ /V4+ can be obtained. XPS characterization also gives some information of the oxygen species on the catalyst surface shown in Fig. 4. Based on the peak resolution of O1s , oxygen species on the catalyst surface involves the lattice (O2− ) and adsorbed oxygen species (O− and O2 − ), which is in a good agreement with the H2 -TPR characterization. Meanwhile, the amount of three different oxygen species O2− , O− and O2 − (denoted as OI , OII , and OIII ) can be obtained in Table 2 [29,30], indicating that the lattice oxygen species increases is in the order of Mg3 (VO4 )2 < Li- < Na- < K- < Ca-.
673
873
1073
Temperature/K Fig. 2. H2 -TPR profiles of the catalysts.
1273
It is well known that the reaction is accepted to proceed by the Mars–van Krevelen reaction mechanism [31], in which adsorbed cyclohexane reacts with lattice oxygen and the reduced metal oxide reacts with adsorbed, dissociated oxygen. Besides the main product cyclohexene, benzene, a trace of 1,3-cyclohexadiene and the nonselective oxidation production COx are also observed. Table 1 presents the catalytic conversions and the selectivity to cyclohexene over all the samples at various temperatures. Given a similar surface area of the doped samples, the comparison is made in terms of the conversion of cyclohexane and the selectivity to cyclohexene. The doped catalysts shows a lower conversion of cyclohexane and a higher selectivity to cyclohexene; meanwhile, the conversion of cyclohexane increases and the selectivity to cyclohexene decreases with elevation of reaction temperature,
Please cite this article in press as: M. Jin, et al., Effect of additives doping on catalytic properties of Mg3 (VO4 )2 catalysts in oxidative dehydrogenation of cyclohexane, Catal. Today (2012), http://dx.doi.org/10.1016/j.cattod.2012.09.019
ARTICLE IN PRESS
G Model CATTOD-8246; No. of Pages 7
M. Jin et al. / Catalysis Today xxx (2012) xxx–xxx
4
Li-Mg3(VO4)2
Intensity/CPS
Intensity/CPS
Mg3(VO4)2
512
514
516
518
520
522
524
526
528
512
514
516
Binding energy/eV
518
520
522
524
526
528
Binding energy/eV
K-Mg3(VO4)2
Intensity/CPS
Intensity/CPS
Na-Mg3(VO4)2
512
514
516
518
520
522
524
526
528
512
514
516
Binding energy/eV
518
520
522
524
526
528
Binding energy/eV
Intensity/CPS
Ca-Mg3(VO4)2
512
514
516
518
520
522
524
526
528
Binding energy/eV Fig. 3. XPS spectra supplementary information of V2p on the catalysts.
Please cite this article in press as: M. Jin, et al., Effect of additives doping on catalytic properties of Mg3 (VO4 )2 catalysts in oxidative dehydrogenation of cyclohexane, Catal. Today (2012), http://dx.doi.org/10.1016/j.cattod.2012.09.019
ARTICLE IN PRESS
G Model CATTOD-8246; No. of Pages 7
M. Jin et al. / Catalysis Today xxx (2012) xxx–xxx
5
Li-Mg3(VO4)2
Intensity/CPS
Intensity/CPS
Mg3(VO4)2
526
528
530
532
534
536
526
528
Binding energy/eV
530
532
534
536
Binding energy/eV
Na-Mg3(VO4)2
Intensity/CPS
Intensity/CPS
K-Mg3(VO4)2
526
528
530
532
534
536
526
528
Binding energy/eV
530
532
534
536
Binding energy/eV
Intensity/CPS
Ca-Mg3(VO4)2
526
528
530
532
534
536
Binding energy/eV Fig. 4. XPS spectra supplementary information of O1s on the catalysts.
Please cite this article in press as: M. Jin, et al., Effect of additives doping on catalytic properties of Mg3 (VO4 )2 catalysts in oxidative dehydrogenation of cyclohexane, Catal. Today (2012), http://dx.doi.org/10.1016/j.cattod.2012.09.019
G Model CATTOD-8246; No. of Pages 7
ARTICLE IN PRESS M. Jin et al. / Catalysis Today xxx (2012) xxx–xxx
6 Table 2 XPS results for the catalysts. Catalysts
Mg3 (VO4 )2 LiNaKCa-
V5+
V4+
V5+/ V4+
BE O1s
OI/ OII/ OIII
BEL V2p 3/2
BEL V2p 1/2
BEL V 2p 3/2
BEL V 2p 1/2
(mol%)
OI
OII
OIII
(mol%)
517.9 517.7 517.6 517.6 517.5
525.8 525.6 525.5 525.6 525.3
515.8 515.6 515.8 515.7 515.4
524.9 524.8 524.7 524.7 524.5
2.75 2.58 2.47 2.30 2.39
530.4 530.4 530.4 530.5 530.3
531.8 531.8 531.9 532.4 531.9
532.9 532.8 532.7 533.6 532.8
85.4/8.4/6.2 86.3/7.7/6.0 88.1/9.5/2.4 89.1/8.9/2.0 90.4/7.1/2.5
as a typical behavior for the ODH over oxide catalysts. Among the samples, K-catalyst shows the lowest conversion of cyclohexane and Ca-catalyst shows the highest selectivity to cyclohexene and the highest yield of cyclohexene, as shown in Table 1. Fig. 5 demonstrates the model of cyclohexane ODH reactionshowing of the catalyst surface. The introduction of additive would geometrical block a certain amount of active sites, which leads to a lower conversion of cyclohexane. Moreover, it is interesting to find that the effect becomes more pronounced as the ionic radius of the additives increase from Li to Na, Ca and K, explaining the differentiation among the catalytic activities in the same order of K- < Ca- ≈ Na- < Li- < non-doped. Due to the high electron density at bond, cyclohexene is considered nucleophilicity. If the catalyst surface is of almost the same property, the selectivity to cyclohexene can be improved. In the previous studies [32,33], the introduction of the additive should effect the distribution of electronic cloud densities around vanadium and oxygen atoms; meanwhile, if the additive is an ion of lower electronegativity than that of vanadium, oxygen atoms in the active species become more nucleophilicity and vanadium less electrophilicity. Accordingly, the alkali/alkaline metal increase the nucleophilicity of the catalyst surface through its coordination with surface vanadium species. As a result, compared with the non-doped catalyst, the doped one could lead to a lower cyclohexane conversion and higher cyclohexene selectivity, which is in accordance with the experimental results listed in Table 1. According to the Mars–van Krevelen mechanism, gaseous oxygen participates in the reaction only after adsorption in other parts of the catalyst and then migrates through the lattice to the active sites by the reoxidation between V4+ and V5+ . Therefore, the coexistence of V4+ and V5+ on the surface is necessary for the ODH, as shown in Table 2. Among the vanadium species, V5+ O in an octahedral structure is responsible for providing the center of oxidation for the adsorbed alkyl species, which contributes to a high conversion of cyclohexane and a high selectivity to COx , whereas V4+ O in a tetrahedral structure is responsible for activating cyclohexane and adsorbing oxygen, which attributes a high selectivity to cyclohexene. As a result, a lower atomic ratio of V5+/ V4+ on the catalyst surface suggests a lower conversion of cyclohexane and a higher selectivity to cyclohexene, as shown in Tables 1 and 2.
Furthermore, the lower reducibility of the active species and the lower V oxidation states are essential to a better catalytic performance [34,35]. According to the results of H2 -TPR characterization, the additive plays a beneficial effect on cyclohexene selectivity. However, the selectivity to cyclohexene showed in Table 1 follows the order of Mg3 (VO4 )2 < Li- < Na- < K- < Ca-, which is different from that of the reducibility of the active species. It can be suggested that the selectivity to cyclohexene does not depend clearly on the reducibility of the active species. According to the research of Valenzuela et al. [36] and Sugiyama et al. [35], when the additive was doped into Mg3 (VO4 )2 , unbalanced charges and lattice distortion occur on its surface. From XPS characterization of the samples, a small shift to lower binding energies in V2p 3/2 (V 2p 3/2 ) and V2p 1/2 (V 2p 1/2 ) implies an increase of the electron density around vanadium atoms, which result in a easy desorption of cyclohexene from the catalyst surface [37]. Besides, alkali/alkaline earth metal is known to facilitate oxygen dissociation in metal-based catalyst by transferring an electron to metal center, which is then back-donated to the 1 g antibonding orbital of the oxygen molecule [38]. A similar effect may be present in metal oxide system, where dissociated oxygen species may be easily incorporated into the lattice as O2− [7]. Among the three types of oxygen species [39–41], the lattice oxygen species O2− are mainly responsible for selective oxidation, whereas the adsorbed oxygen species O− and O2 − are responsible for the activation of cyclohexane and deep oxidation to the formation of COx . As a result, the less adsorbed oxygen species (O− + O2 − ) the catalyst has, the higher selectivity to cyclohexene can be obtained. Accordingly, the selectivity to cyclohexene should be increased as the amount of lattice oxygen species O2− in the order of Mg3 (VO4 )2 < Li< Na- < K- < Ca-. As a matter of fact, the order of the selectivity to cyclohexene listed in Table 1 shows a good agreement with that of the amount of lattice oxygen species of all catalysts. Then, it is suggested the selectivity to cyclohexene appear to be mostly influenced by the type and number of the oxygen species on the catalyst surface. In addition, the results also reveal that the contributions of the lattice oxygen in the catalysts to the reaction should be strongly influenced by the introduction of the additive, especially Ca. Due to the different valence with the alkali metal, the incorporation of Ca2+ into Mg3 (VO4 )2 not only affects the abstraction/the incorporation
Fig. 5. Model of the ODH of cyclohexane reaction showing of the catalysts.
Please cite this article in press as: M. Jin, et al., Effect of additives doping on catalytic properties of Mg3 (VO4 )2 catalysts in oxidative dehydrogenation of cyclohexane, Catal. Today (2012), http://dx.doi.org/10.1016/j.cattod.2012.09.019
G Model CATTOD-8246; No. of Pages 7
ARTICLE IN PRESS M. Jin et al. / Catalysis Today xxx (2012) xxx–xxx
of oxygen species from and into the lattice vacancy but also influences the stabilization of the catalyst structure during the redox process. 4. Conclusions The addition of alkali/alkaline earth metal to Mg3 (VO4 )2 affects the physical, chemical nature and the catalytic behavior in the ODH of cyclohexane to cyclohexene. During the coordination of the additive with the surface active species, the additive blocked the active sites and hindered the reducibility of the active species thus decreasing the catalytic activity. Moreover, the additive could improve the selectivity to cyclohexene by enhancing the nucleophilicity, the redox property as well as the type and number of the oxygen species as investigated by H2 -TPR and XPS characterization. However, the selectivity to cyclohexene did not depend clearly on the reducibility of the active species, but appears to be strongly influenced by the acid-basic property as well as the type and number of the oxygen species on the catalyst surface. As a result, the catalytic activity and the selectivity to cyclohexene increased in the orders of K- < Na- ≈ Ca< Li- < Mg3 (VO4 )2 and Mg3 (VO4 )2 < Li- < Na- < K- < Ca-, respectively. Among the doped catalysts, Ca-catalyst could give a relatively higher yield of cyclohexene of 8.2% compared with the nonmodified one. Acknowledgements The present work was conducted under support from the Doctoral Scientific Research Fund of Jianghan University (No. 2010018) and the Project of Education Department of Hubei Province (Q20123402). References [1] [2] [3] [4]
M.C. Kung, H.H. Kung, Journal of Catalysis 128 (1991) 287. M. Jin, Z.M. Cheng, Catalysis Letters 131 (2009) 266. M. Jin, Z.M. Cheng, Y.L. Gao, X.C. Fang, Materials Letters 63 (2009) 2055. A. Christodoulakis, M. Machli, A.A. Lemonidou, S. Boghosian, Journal of Catalysis 222 (2004) 293. [5] S. Blanco, S.R.G. Carrazán, V. Rives, Applied Catalysis A – General 342 (2008) 93. [6] T. Blasco, J.M. Lopez Nieto, Applied Catalysis A – General 157 (1997) 117. [7] B. Grzybowska, P. Mekˇss, R. Grabowski, K. Wcisło, Y. Barbaux, L. Gengembre, Studies in Surface Science and Catalysis 82 (1994) 151.
7
˜ [8] R. Grabowska, B. Grzybowska, K. Samson, J. Słoczynski, J. Stoch, K. Wcisło, Applied Catalysis A 125 (1995) 129. [9] A.A. Lemonidou, L. Nalbandian, I.A. Vasalos, Catalysis Today 61 (2000) 333. ´ [10] A. Klisinska, K. Samson, I. Gressel, B. Grzybowska, Applied Catalysis A – General 309 (2006) 10. ´ S. Loridant, B. Grzybowska, J. Stoch, I. Gressel, Applied Catalysis A [11] A. Klisinska, – General 309 (2006) 17. [12] R.X. Valenzuela, E.A. Mamedov, V.C. Corberán, Reaction Kinetics and Catalysis Letters 55 (1995) 213. [13] H.H. Kung, M.C. Kung, Applied Catalysis A – General 157 (1997) 105. [14] M.C. Kung, H.H. Kung, Journal of Catalysis 134 (1992) 668. [15] W.D. Harding, H.H. Kung, V.L. Kozhevnikov, K.R. Poeppelmeier, Journal of Catalysis 144 (1993) 597. [16] S. Sugiyama, T. Hashimoto, Y. Morishiita, N. Shigemoto, H. Hayashi, Applied Catalysis A – General 270 (2004) 253. [17] F. Patcas, D. Hönicke, Catalysis Communications 6 (2005) 23. [18] F. Patcas, W. Krysmann, D. Hönicke, F.C. Buciuman, Catalysis Today 69 (2001) 379. [19] B. Grzybowska, I. Gressel, K. Samson, K. Wcisło, J. Stoch, M. Mikołajczyk, F. Dautzenberg, Polish Journal of Chemistry 75 (2001) 1513. [20] M. Houalla, F. Delannay, B. Delmon, Journal of Physical Chemistry 85 (12) (1981) 1704. [21] S.L. Chen, H.L. Zhang, J. Hu, C. Contescu, J.A. Schwarz, Applied Catalysis A 73 (1991) 289. [22] L. Balderas-Tapia, I. Hernández-Pérez, P. Schacht, L.R. Córdova, G.G. AguilarRíos, Catalysis Today 107–108 (2005) 371. [23] J. Sloczynski, Applied Catalysis A 146 (1996) 401. [24] T. Chen, W.Z. Li, C.Y. Yu, Chinese Journal of Catalysis 18 (1997) 160. [25] C.Y. Yu, W.Z. Li, W.J. Feng, Studies in Surface Science and Catalysis 75 (1993) 1119. ´ Topics in Catalysis 11–12 (2000) 23. [26] B. Grzybowska-Swierkosz, [27] D. Siew Hew Sam, V. Soenen, J.C. Volta, Journal of Catalysis 123 (1990) 417. [28] X.T. Gao, P. Ruiz, Q. Xin, X.X. Guo, B. Delmon, Journal of Catalysis 148 (1994) 56. [29] H.X. Yuan, J.M. Ouyang, Spectroscopy and Spectral Analysis 27 (2007) 395. [30] A.J. Xu, G.T. Zhaori, M.L. Jia, Q. Lin, Spectroscopy and Spectral Analysis 27 (2007) 2134. [31] P. Mars, D.W. Van Krevenlen, Chemical Engineering Science 3 (Suppl.) (1954) 341. [32] M.C. Biesinger, L.W.M. Lau, A.R. Gerson, R.S.C. Smart, Applied Surface Science 257 (2010) 887. [33] J.C. Vedrine, J.M.M. Millet, J.C. Volta, Catalysis Today 32 (1996) 115. [34] O.V. Safonova, M. Florea, J. Bilde, P. Delichere, J.M.M. Millet, Journal of Catalysis 268 (2009) 156. [35] S. Sugiyama, T. Osaka, Y. Hirata, K.I. Sotowa, Applied Catalysis A – General 312 (2006) 52. [36] R.X. Valenzuela, E.A. Mamedov, V. Cortes Corberan, Reaction Kinetics and Catalysis Letters 55 (1995) 213. ´ [37] A. Klisinska, A. Haras, K. Samson, M. Witko, B. Grzybowska, Journal of Molecular Catalysis A: Chemical 210 (2004) 87. [38] M. Witko, R. Grybos, R. Tokarz-Sobieraj, Topics in Catalysis 38 (2006) 75. [39] M. Chen, X. Zheng, Journal of Molecular Catalysis A 201 (2003) 161. [40] X. Wang, Y.C. Xie, Applied Catalysis B 35 (2001) 85. [41] S. Deng, H.Q. Li, S.G. Li, Y. Zhang, Journal of Molecular Catalysis A: Chemical 268 (2007) 169.
Please cite this article in press as: M. Jin, et al., Effect of additives doping on catalytic properties of Mg3 (VO4 )2 catalysts in oxidative dehydrogenation of cyclohexane, Catal. Today (2012), http://dx.doi.org/10.1016/j.cattod.2012.09.019
ARTICLE IN PRESS Catalysis Today xxx (2012) xxx–xxx
Contents lists available at SciVerse ScienceDirect
Catalysis Today journal homepage: www.elsevier.com/locate/cattod
Effect of additives doping on catalytic properties of Mg3 (VO4 )2 catalysts in oxidative dehydrogenation of cyclohexane M. Jin a , P. Lu a , G.X. Yu a,∗ , Z.M. Cheng b , L.F. Chen c , J.A. Wang c a
School of Chemistry and Environmental Engineering, Jianghan University, Wuhan, Hubei 430056, PR China State Key Laboratory of Chemical Engineering, UNILAB Research Centre of Chemical Reaction Engineering, East China University of Science and Technology, Shanghai 200237, PR China c Escuela Superior de Ingeniería Química e Industrias Extractivas (ESIQIE), Instituto Politécnico Nacional, Av. Politécnico S/N, Col. Zacatenco, 07738 México, D.F., Mexico b
a r t i c l e
i n f o
Article history: Received 7 February 2012 Received in revised form 21 August 2012 Accepted 1 September 2012 Available online xxx Keywords: Cyclohexane Oxidative dehydrogenation Alkali/alkaline earth metal Cyclohexene Modified catalyst
a b s t r a c t Effects of additives comprising alkali and alkaline earth metals (Li, Na, K and Ca) introduced to Mg3 (VO4 )2 on their structures, physicochemical properties and the catalytic behaviors in the oxidative dehydrogenation of cyclohexane were investigated. The characterization and the experimental results showed that the additive does not affect markedly the structure of the catalyst, but blocks the active sites and hinders the reducibility of the active species thus decreasing the catalytic activity. Moreover, the additive could enhance the selectivity to cyclohexene by enhancing the nucleophilicity, the redox property as well as the type and number of the oxygen species on the catalyst surface. The doped catalysts could give the catalytic activity and selectivity to cyclohexene in the orders of K- < Na- ≈ Ca- < Li- < non-doped and non-doped < Li< Na- < K- < Ca-, respectively. Among the doped catalysts, Ca–Mg3 (VO4 )2 catalyst demonstrated a yield of cyclohexene of 8.2%, which was a promotion compared with 6.4% of Mg3 (VO4 )2 . © 2012 Published by Elsevier B.V.
1. Introduction The oxidative dehydrogenation (ODH) of cyclohexane to cyclohexene is a challenging pursuit in both economic and scientific terms. Previous investigations of the catalysts used in the ODH reaction of cyclohexane demonstrated Mg3 (VO4 )2 has been considered as a favorite catalyst for producing a promising yield of cyclohexene [1–3]. However, a major challenge in the commercial development of cyclohexane ODH is further improvement in cyclohexene yield since the main cause of the selectivity limitation arises from the conversion–selectivity relationship [4,5]. One of the solutions for the catalyst is to ensure easy desorption of intermediate product cyclohexene from the catalyst surface. While, the difficulty or the ease of cyclohexene desorption depends on the properties of the catalyst surface. Therefore, an effective way for improving the selectivity is to modify the catalyst surface by the introduction of additive, especially the alkali/alkaline earth metal [6–11]. The introduction of alkali/alkaline earth metal to V2 O5 /Al2 O3 [6,7], V2 O5 /TiO2 [8,9], V2 O5 /SiO2 , V2 O5 /MgO [10,11] and V Mg O catalysts [12] has been carried out in the ODH of lower alkanes to produce the corresponding alkenes. Most of the additives, such as Ca, Li, Na and Cs, played a promoting effect on
∗ Corresponding author. Tel.: +86 027 84226806. E-mail address: [email protected] (G.X. Yu).
the selectivity to the objective alkene which could be explained by modification of acid–base properties and the reducibility of the active phases by alkalis. While the other additive, such as K, was found to decrease the selectivity to the objective product since K was found to inhibit the formation of Mg3 (VO4 )2 from a mixture of MgO and NH4 VO3 and/or trace K was segregated onto the surface of magnesium vanadates [13–16]. Compared with the ODH of lower alkanes, Patcas et al. [17,18] only investigated the effect of alkali metal addition on the catalytic activity in the ODH of cyclohexane. They first modified NiO/Al2 O3 catalyst by alkali (Na, K and Cs) and found no promotion effect owing to a lower active species content and a larger amount of bulk nickel aluminate leaded by the modification. Then they modified the egg-shell NiO/Al2 O3 catalyst by Li and proposed Li could enhance the catalytic activity and the selectivity to cyclohexene by decreasing the formation of bulk nickel aluminates. Besides, it is a pity that few or no studies have been reported on the influence of alkali/alkaline earth metal doping on catalytic property of Mg3 (VO4 )2 in the ODH of cyclohexane. In this work, the catalytic performance of the alkali/alkaline metal (Li, Na, K and Ca) doping on Mg3 (VO4 )2 was studied in the ODH of cyclohexane. N2 -adsorption, XRD, H2 -TPR and XPS techniques were applied to give further information on the structural, the reducibility of active species as well as the type and distribution of the surface oxygen species of the catalyst. Furthermore, the relationship among the reducibility of the active species, the type and
0920-5861/$ – see front matter © 2012 Published by Elsevier B.V. http://dx.doi.org/10.1016/j.cattod.2012.09.019
Please cite this article in press as: M. Jin, et al., Effect of additives doping on catalytic properties of Mg3 (VO4 )2 catalysts in oxidative dehydrogenation of cyclohexane, Catal. Today (2012), http://dx.doi.org/10.1016/j.cattod.2012.09.019
G Model CATTOD-8246; No. of Pages 7
ARTICLE IN PRESS M. Jin et al. / Catalysis Today xxx (2012) xxx–xxx
2
distribution of the surface oxygen species and the catalytic behavior of the catalysts were investigated in detail.
Ca-Mg3(VO 4)2
2. Experimental K-Mg 3(VO 4)2
2.1. Catalyst preparation Mg3 (VO4 )2 was synthesized via the citrate complexation method, which has been described in detail in the literature [2,3]. First, a transparent solution of Mg(NO3 )2 ·6H2 O and NH4 VO3 with Mg/V atomic ratio of 3/2 was prepared. Then citric acid was added in the mixing solution with a 20% excess over the number of ionic equivalents of cations. Subsequently, ammonia was added to adjust the pH value of 4.8. The solution obtained was evaporated at 343 K to form a gel, and the gel was dried subsequently in a vacuum oven overnight at 343 K. Finally, the resulting solid was ground and heated up to 673 K for 18 h at a constant rising rate of 1 K/min to decompose the gel precursor, then calcined in the air at 823 K for 6 h. The final sample was crushed and sieved to obtain a particle size of 20–40 mesh. The doped samples were denoted as A-, where A were Li, Na, K and Ca, respectively. In this work, the atomic ratio A/V of 0.1 was adopted, which was based on the following aspects [19]: (1) the K/V ratio of 0.1 was shown in previous studies as optimal for V-based catalysts; (2) the A/V ratio of 0.1 can avoid formation of mixed A V O bulk compounds and/or trace A onto the catalyst surface; (3) the evident modification of the catalytic properties can be observed at the A/V ratio of 0.1. The doped samples were prepared by incipient impregnation. The additive was introduced by adding the appropriate amount of the corresponding nitrate solution onto Mg3 (VO4 )2 particles. After impregnation, the obtaining sample was dried at 393 K overnight and calcined in the air at 823 K for 6 h. 2.2. Catalyst characterization BET surface area of the as-prepared sample was measured by N2 physisorption at 77.3 K (ASAP 2010, Micromeritics). The XRD pattern was measured using Cu K␣ radiation (D/MAX 2550 X-ray diffractometer) at 40 kV and 40 mA. Actual composition of the sample was determined on an ICP-AES equipment (Varian 710 ES). XPS measurements were carried out on a PHI 5300/ESCA system with Al K␣ as a radiation source (Perkin-Elmer), the correction of the surface charge electricity effect was using C1s (284.60 eV) and the range of the spectrum of scanning energy was from 0 to 1200 eV. XPSPEAK software was used for the deconvolution and analysis the XPS characterization results. The temperature-programmed reduction (H2 -TPR) was measured on a Micromeritics AutoChem II 2920 apparatus. Before reduction, about 0.20 g sample was pre-treated in air at 423 K for 40 min to remove the adsorbed water, followed by cooling to 313 K under Ar flow (30 ml/min). Subsequently, the sample was reduced with a 10 vol.% H2 /Ar mixture (30 ml/min) by temperature programming from 313 to 1273 K at a rate of 10 K/min. 2.3. Catalytic test The catalyst activity test was carried out isothermally at atmospheric pressure in a fixed-bed microreactor made of stainless steel with an inner diameter of 9 mm. About 0.5 g of as-prepared catalyst diluted with inert quartz sands at a mass ratio of 1:4 was loaded into the reactor so as to ensure a uniform catalyst distribution. In each run, the catalyst was first pretreated in an air stream of 50 ml/min at 773 K for 1 h, and then adjusted to the reaction conditions. The system was allowed to stabilize for 1.5 h before the first product sample was taken for analysis. The liquid products were analyzed
Na-Mg3(VO 4)2
Li-Mg 3(VO 4)2
Mg 3(VO4)2
10
20
30
40
50
60
70
80
2 Fig. 1. XRD patterns of the catalysts.
by a HP 6890 gas chromatography equipped with a PEG 20,000 column. 3. Results and discussion 3.1. Textural and structural properties Examination of the surface areas of the samples reveals that the doping treatment produces a modification during the preparation procedure. The doped catalysts show a lower BET surface area with respect to the non-doped one (Table 1); meanwhile, the results show that the surface area decreases along with the ion size increasing, which could be explained by the formation of O− A+ on the catalyst surface [20,21]. As shown in Fig. 1, no features of crystalline additive vanadates or additive metal involving material can be observed in the doped catalysts, which is different from the research of Kung and Kung [13,14]. The XRD characterization result suggests that the additive coordinates with the surface active species in modifying its structure instead of forming bulk compounds. 3.2. Characterization results of TPR As shown in Fig. 2, the temperature of maximum hydrogen consumption (Tmax ) is 970 K for Mg3 (VO4 )2 , while it shifts to 979 (Li-), 1002 (Na-), 1054 (K-), and 1023 K (Ca-) for the doped catalysts, respectively. The shift to higher temperature could be due to the stronger interaction between the active species and the additive, which is in agreement with the views of Valenzuela et al. [12], and Balderas-Tapia et al. [22]. It is obviously that the reducibility of the active species decreases in the order of Mg3 (VO4 )2 > Li- > Na- > Ca> K-. In addition, based on the previous researches [24,25], Tmax ranged from 873 to 973 K is responsible for the adsorption of the incompletion reduction oxygen Oı− (0 < ı < 2), and the value ranged from 973 to 1173 K is involved in the adsorption of lattice oxygen. Therefore, the oxygen species on the doped-catalysts surface are the mixture of the lattice oxygen and Oı− (0 < ı < 2) species.
Please cite this article in press as: M. Jin, et al., Effect of additives doping on catalytic properties of Mg3 (VO4 )2 catalysts in oxidative dehydrogenation of cyclohexane, Catal. Today (2012), http://dx.doi.org/10.1016/j.cattod.2012.09.019
ARTICLE IN PRESS
G Model CATTOD-8246; No. of Pages 7
M. Jin et al. / Catalysis Today xxx (2012) xxx–xxx
3
Table 1 Catalytic performance of cyclohexane ODH over the catalysts. SBET (m2 /g)
Catalysts
Mg3 (VO4 )2
Li-
Na-
K-
Ca-
ICP (mol) Mg/V/A
Temperature (K)
Conversion (%)
Selectivity (%)
Cyclohexane
Cyclohexene
COx
Yield (%) Cyclohexane
7.8 15.5 21.1
65.7 41.5 27.2
16.2 20.7 28.2
5.12 6.43 5.73
17.5
1.5/1/–
673 723 773
15.5
1.5/1/0.1
673 723 773
5.8 11.5 18.0
77.6 61.4 42.1
11.2 17.3 25.3
4.50 7.06 7.57
14.8
1.5/1/0.1
673 723 773
4.9 10.8 17.1
79.9 65.9 43.5
10.3 16.9 24.1
3.92 7.12 7.43
14.3
1.5/1/0.1
673 723 773
4.2 10.0 15.4
81.8 69.6 49.5
9.9 16.5 23.6
3.43 6.96 7.62
15.2
1.5/1/0.1
673 723 773
4.7 10.6 16.5
82.1 69.5 49.9
10.2 16.8 23.5
3.86 7.37 8.23
Experimental condition: cyclohexane flow rate of 0.107 mol/h, air flow rate of 100 ml/min, atmospheric pressure.
Even though quantification of H2 -TPR profiles does not give precise information about V oxidation states, it can provide some useful indications. According to the total hydrogen consumption, the average valences of vanadium species in the samples are probably +4.69 (K-), +4.71 (Na-), +4.72 (Li-), +4.70 (Ca-) and +4.74 (Mg3 (VO4 )2 ), respectively. It is indicated that the type of the additive has an effect on the V oxidation states, which is in same with the view proposed by Sloczynski [23]. In addition, no feature of the reduction peaks of crystalline additive vanadates or additive metal involving material, which confirms the result of the XRD characterization. 3.3. Characterization results of XPS When the additive is an ion with a larger ionic radius than that of V5+ /V4+ /Mg2+ , it might enter into the interstitial site of the catalyst lattice rather than replace the cations in the bulk crystal [26]. In order to investigate the composition of the catalyst surface, possible Ca-Mg3(VO4)2
TCD Signal/a.u.
K-Mg3(VO4)2
Na-Mg3(VO4)2
3.4. Catalytic performances
Li-Mg3(VO4)2
Mg3(VO4)2
273
473
variations in the vanadium oxidation state, changes of the electronic density of V O and the type and the distribution of the oxygen species, XPS were recorded shown in Figs. 3 and 4 and Table 2. Fig. 3 demonstrated XPS spectra of V2p on the catalysts surface. The assignment of bands is a complex problem, as can be seen from the different results of V4+ and V5+ reported in literature [27,28]. For V2 O4 , binding energy values of 515.5, 515.7 and 516.3–516.6 eV have been assigned, and for V2 O5 , values between 517.1 and 518.0 eV have been assigned. It can be suggested that bands with a high binding energy value can be assigned to V5+ while those with a low binding energy value may be attributed to V4+ . The results listed in Table 2 of the peak resolution of V species show that there is a small shift to lower binding energies in V2p 3/2 (V 2p 3/2 ) and V2p 1/2 (V 2p 1/2 ) of the doped catalysts with respect to the nondoped one, which indicates that there should be an increase of the electron density around vanadium atoms. Considering the limitation in the assignment of the bands, an approximate value of the atomic ratios of V5+ /V4+ can be obtained. XPS characterization also gives some information of the oxygen species on the catalyst surface shown in Fig. 4. Based on the peak resolution of O1s , oxygen species on the catalyst surface involves the lattice (O2− ) and adsorbed oxygen species (O− and O2 − ), which is in a good agreement with the H2 -TPR characterization. Meanwhile, the amount of three different oxygen species O2− , O− and O2 − (denoted as OI , OII , and OIII ) can be obtained in Table 2 [29,30], indicating that the lattice oxygen species increases is in the order of Mg3 (VO4 )2 < Li- < Na- < K- < Ca-.
673
873
1073
Temperature/K Fig. 2. H2 -TPR profiles of the catalysts.
1273
It is well known that the reaction is accepted to proceed by the Mars–van Krevelen reaction mechanism [31], in which adsorbed cyclohexane reacts with lattice oxygen and the reduced metal oxide reacts with adsorbed, dissociated oxygen. Besides the main product cyclohexene, benzene, a trace of 1,3-cyclohexadiene and the nonselective oxidation production COx are also observed. Table 1 presents the catalytic conversions and the selectivity to cyclohexene over all the samples at various temperatures. Given a similar surface area of the doped samples, the comparison is made in terms of the conversion of cyclohexane and the selectivity to cyclohexene. The doped catalysts shows a lower conversion of cyclohexane and a higher selectivity to cyclohexene; meanwhile, the conversion of cyclohexane increases and the selectivity to cyclohexene decreases with elevation of reaction temperature,
Please cite this article in press as: M. Jin, et al., Effect of additives doping on catalytic properties of Mg3 (VO4 )2 catalysts in oxidative dehydrogenation of cyclohexane, Catal. Today (2012), http://dx.doi.org/10.1016/j.cattod.2012.09.019
ARTICLE IN PRESS
G Model CATTOD-8246; No. of Pages 7
M. Jin et al. / Catalysis Today xxx (2012) xxx–xxx
4
Li-Mg3(VO4)2
Intensity/CPS
Intensity/CPS
Mg3(VO4)2
512
514
516
518
520
522
524
526
528
512
514
516
Binding energy/eV
518
520
522
524
526
528
Binding energy/eV
K-Mg3(VO4)2
Intensity/CPS
Intensity/CPS
Na-Mg3(VO4)2
512
514
516
518
520
522
524
526
528
512
514
516
Binding energy/eV
518
520
522
524
526
528
Binding energy/eV
Intensity/CPS
Ca-Mg3(VO4)2
512
514
516
518
520
522
524
526
528
Binding energy/eV Fig. 3. XPS spectra supplementary information of V2p on the catalysts.
Please cite this article in press as: M. Jin, et al., Effect of additives doping on catalytic properties of Mg3 (VO4 )2 catalysts in oxidative dehydrogenation of cyclohexane, Catal. Today (2012), http://dx.doi.org/10.1016/j.cattod.2012.09.019
ARTICLE IN PRESS
G Model CATTOD-8246; No. of Pages 7
M. Jin et al. / Catalysis Today xxx (2012) xxx–xxx
5
Li-Mg3(VO4)2
Intensity/CPS
Intensity/CPS
Mg3(VO4)2
526
528
530
532
534
536
526
528
Binding energy/eV
530
532
534
536
Binding energy/eV
Na-Mg3(VO4)2
Intensity/CPS
Intensity/CPS
K-Mg3(VO4)2
526
528
530
532
534
536
526
528
Binding energy/eV
530
532
534
536
Binding energy/eV
Intensity/CPS
Ca-Mg3(VO4)2
526
528
530
532
534
536
Binding energy/eV Fig. 4. XPS spectra supplementary information of O1s on the catalysts.
Please cite this article in press as: M. Jin, et al., Effect of additives doping on catalytic properties of Mg3 (VO4 )2 catalysts in oxidative dehydrogenation of cyclohexane, Catal. Today (2012), http://dx.doi.org/10.1016/j.cattod.2012.09.019
G Model CATTOD-8246; No. of Pages 7
ARTICLE IN PRESS M. Jin et al. / Catalysis Today xxx (2012) xxx–xxx
6 Table 2 XPS results for the catalysts. Catalysts
Mg3 (VO4 )2 LiNaKCa-
V5+
V4+
V5+/ V4+
BE O1s
OI/ OII/ OIII
BEL V2p 3/2
BEL V2p 1/2
BEL V 2p 3/2
BEL V 2p 1/2
(mol%)
OI
OII
OIII
(mol%)
517.9 517.7 517.6 517.6 517.5
525.8 525.6 525.5 525.6 525.3
515.8 515.6 515.8 515.7 515.4
524.9 524.8 524.7 524.7 524.5
2.75 2.58 2.47 2.30 2.39
530.4 530.4 530.4 530.5 530.3
531.8 531.8 531.9 532.4 531.9
532.9 532.8 532.7 533.6 532.8
85.4/8.4/6.2 86.3/7.7/6.0 88.1/9.5/2.4 89.1/8.9/2.0 90.4/7.1/2.5
as a typical behavior for the ODH over oxide catalysts. Among the samples, K-catalyst shows the lowest conversion of cyclohexane and Ca-catalyst shows the highest selectivity to cyclohexene and the highest yield of cyclohexene, as shown in Table 1. Fig. 5 demonstrates the model of cyclohexane ODH reactionshowing of the catalyst surface. The introduction of additive would geometrical block a certain amount of active sites, which leads to a lower conversion of cyclohexane. Moreover, it is interesting to find that the effect becomes more pronounced as the ionic radius of the additives increase from Li to Na, Ca and K, explaining the differentiation among the catalytic activities in the same order of K- < Ca- ≈ Na- < Li- < non-doped. Due to the high electron density at bond, cyclohexene is considered nucleophilicity. If the catalyst surface is of almost the same property, the selectivity to cyclohexene can be improved. In the previous studies [32,33], the introduction of the additive should effect the distribution of electronic cloud densities around vanadium and oxygen atoms; meanwhile, if the additive is an ion of lower electronegativity than that of vanadium, oxygen atoms in the active species become more nucleophilicity and vanadium less electrophilicity. Accordingly, the alkali/alkaline metal increase the nucleophilicity of the catalyst surface through its coordination with surface vanadium species. As a result, compared with the non-doped catalyst, the doped one could lead to a lower cyclohexane conversion and higher cyclohexene selectivity, which is in accordance with the experimental results listed in Table 1. According to the Mars–van Krevelen mechanism, gaseous oxygen participates in the reaction only after adsorption in other parts of the catalyst and then migrates through the lattice to the active sites by the reoxidation between V4+ and V5+ . Therefore, the coexistence of V4+ and V5+ on the surface is necessary for the ODH, as shown in Table 2. Among the vanadium species, V5+ O in an octahedral structure is responsible for providing the center of oxidation for the adsorbed alkyl species, which contributes to a high conversion of cyclohexane and a high selectivity to COx , whereas V4+ O in a tetrahedral structure is responsible for activating cyclohexane and adsorbing oxygen, which attributes a high selectivity to cyclohexene. As a result, a lower atomic ratio of V5+/ V4+ on the catalyst surface suggests a lower conversion of cyclohexane and a higher selectivity to cyclohexene, as shown in Tables 1 and 2.
Furthermore, the lower reducibility of the active species and the lower V oxidation states are essential to a better catalytic performance [34,35]. According to the results of H2 -TPR characterization, the additive plays a beneficial effect on cyclohexene selectivity. However, the selectivity to cyclohexene showed in Table 1 follows the order of Mg3 (VO4 )2 < Li- < Na- < K- < Ca-, which is different from that of the reducibility of the active species. It can be suggested that the selectivity to cyclohexene does not depend clearly on the reducibility of the active species. According to the research of Valenzuela et al. [36] and Sugiyama et al. [35], when the additive was doped into Mg3 (VO4 )2 , unbalanced charges and lattice distortion occur on its surface. From XPS characterization of the samples, a small shift to lower binding energies in V2p 3/2 (V 2p 3/2 ) and V2p 1/2 (V 2p 1/2 ) implies an increase of the electron density around vanadium atoms, which result in a easy desorption of cyclohexene from the catalyst surface [37]. Besides, alkali/alkaline earth metal is known to facilitate oxygen dissociation in metal-based catalyst by transferring an electron to metal center, which is then back-donated to the 1 g antibonding orbital of the oxygen molecule [38]. A similar effect may be present in metal oxide system, where dissociated oxygen species may be easily incorporated into the lattice as O2− [7]. Among the three types of oxygen species [39–41], the lattice oxygen species O2− are mainly responsible for selective oxidation, whereas the adsorbed oxygen species O− and O2 − are responsible for the activation of cyclohexane and deep oxidation to the formation of COx . As a result, the less adsorbed oxygen species (O− + O2 − ) the catalyst has, the higher selectivity to cyclohexene can be obtained. Accordingly, the selectivity to cyclohexene should be increased as the amount of lattice oxygen species O2− in the order of Mg3 (VO4 )2 < Li< Na- < K- < Ca-. As a matter of fact, the order of the selectivity to cyclohexene listed in Table 1 shows a good agreement with that of the amount of lattice oxygen species of all catalysts. Then, it is suggested the selectivity to cyclohexene appear to be mostly influenced by the type and number of the oxygen species on the catalyst surface. In addition, the results also reveal that the contributions of the lattice oxygen in the catalysts to the reaction should be strongly influenced by the introduction of the additive, especially Ca. Due to the different valence with the alkali metal, the incorporation of Ca2+ into Mg3 (VO4 )2 not only affects the abstraction/the incorporation
Fig. 5. Model of the ODH of cyclohexane reaction showing of the catalysts.
Please cite this article in press as: M. Jin, et al., Effect of additives doping on catalytic properties of Mg3 (VO4 )2 catalysts in oxidative dehydrogenation of cyclohexane, Catal. Today (2012), http://dx.doi.org/10.1016/j.cattod.2012.09.019
G Model CATTOD-8246; No. of Pages 7
ARTICLE IN PRESS M. Jin et al. / Catalysis Today xxx (2012) xxx–xxx
of oxygen species from and into the lattice vacancy but also influences the stabilization of the catalyst structure during the redox process. 4. Conclusions The addition of alkali/alkaline earth metal to Mg3 (VO4 )2 affects the physical, chemical nature and the catalytic behavior in the ODH of cyclohexane to cyclohexene. During the coordination of the additive with the surface active species, the additive blocked the active sites and hindered the reducibility of the active species thus decreasing the catalytic activity. Moreover, the additive could improve the selectivity to cyclohexene by enhancing the nucleophilicity, the redox property as well as the type and number of the oxygen species as investigated by H2 -TPR and XPS characterization. However, the selectivity to cyclohexene did not depend clearly on the reducibility of the active species, but appears to be strongly influenced by the acid-basic property as well as the type and number of the oxygen species on the catalyst surface. As a result, the catalytic activity and the selectivity to cyclohexene increased in the orders of K- < Na- ≈ Ca< Li- < Mg3 (VO4 )2 and Mg3 (VO4 )2 < Li- < Na- < K- < Ca-, respectively. Among the doped catalysts, Ca-catalyst could give a relatively higher yield of cyclohexene of 8.2% compared with the nonmodified one. Acknowledgements The present work was conducted under support from the Doctoral Scientific Research Fund of Jianghan University (No. 2010018) and the Project of Education Department of Hubei Province (Q20123402). References [1] [2] [3] [4]
M.C. Kung, H.H. Kung, Journal of Catalysis 128 (1991) 287. M. Jin, Z.M. Cheng, Catalysis Letters 131 (2009) 266. M. Jin, Z.M. Cheng, Y.L. Gao, X.C. Fang, Materials Letters 63 (2009) 2055. A. Christodoulakis, M. Machli, A.A. Lemonidou, S. Boghosian, Journal of Catalysis 222 (2004) 293. [5] S. Blanco, S.R.G. Carrazán, V. Rives, Applied Catalysis A – General 342 (2008) 93. [6] T. Blasco, J.M. Lopez Nieto, Applied Catalysis A – General 157 (1997) 117. [7] B. Grzybowska, P. Mekˇss, R. Grabowski, K. Wcisło, Y. Barbaux, L. Gengembre, Studies in Surface Science and Catalysis 82 (1994) 151.
7
˜ [8] R. Grabowska, B. Grzybowska, K. Samson, J. Słoczynski, J. Stoch, K. Wcisło, Applied Catalysis A 125 (1995) 129. [9] A.A. Lemonidou, L. Nalbandian, I.A. Vasalos, Catalysis Today 61 (2000) 333. ´ [10] A. Klisinska, K. Samson, I. Gressel, B. Grzybowska, Applied Catalysis A – General 309 (2006) 10. ´ S. Loridant, B. Grzybowska, J. Stoch, I. Gressel, Applied Catalysis A [11] A. Klisinska, – General 309 (2006) 17. [12] R.X. Valenzuela, E.A. Mamedov, V.C. Corberán, Reaction Kinetics and Catalysis Letters 55 (1995) 213. [13] H.H. Kung, M.C. Kung, Applied Catalysis A – General 157 (1997) 105. [14] M.C. Kung, H.H. Kung, Journal of Catalysis 134 (1992) 668. [15] W.D. Harding, H.H. Kung, V.L. Kozhevnikov, K.R. Poeppelmeier, Journal of Catalysis 144 (1993) 597. [16] S. Sugiyama, T. Hashimoto, Y. Morishiita, N. Shigemoto, H. Hayashi, Applied Catalysis A – General 270 (2004) 253. [17] F. Patcas, D. Hönicke, Catalysis Communications 6 (2005) 23. [18] F. Patcas, W. Krysmann, D. Hönicke, F.C. Buciuman, Catalysis Today 69 (2001) 379. [19] B. Grzybowska, I. Gressel, K. Samson, K. Wcisło, J. Stoch, M. Mikołajczyk, F. Dautzenberg, Polish Journal of Chemistry 75 (2001) 1513. [20] M. Houalla, F. Delannay, B. Delmon, Journal of Physical Chemistry 85 (12) (1981) 1704. [21] S.L. Chen, H.L. Zhang, J. Hu, C. Contescu, J.A. Schwarz, Applied Catalysis A 73 (1991) 289. [22] L. Balderas-Tapia, I. Hernández-Pérez, P. Schacht, L.R. Córdova, G.G. AguilarRíos, Catalysis Today 107–108 (2005) 371. [23] J. Sloczynski, Applied Catalysis A 146 (1996) 401. [24] T. Chen, W.Z. Li, C.Y. Yu, Chinese Journal of Catalysis 18 (1997) 160. [25] C.Y. Yu, W.Z. Li, W.J. Feng, Studies in Surface Science and Catalysis 75 (1993) 1119. ´ Topics in Catalysis 11–12 (2000) 23. [26] B. Grzybowska-Swierkosz, [27] D. Siew Hew Sam, V. Soenen, J.C. Volta, Journal of Catalysis 123 (1990) 417. [28] X.T. Gao, P. Ruiz, Q. Xin, X.X. Guo, B. Delmon, Journal of Catalysis 148 (1994) 56. [29] H.X. Yuan, J.M. Ouyang, Spectroscopy and Spectral Analysis 27 (2007) 395. [30] A.J. Xu, G.T. Zhaori, M.L. Jia, Q. Lin, Spectroscopy and Spectral Analysis 27 (2007) 2134. [31] P. Mars, D.W. Van Krevenlen, Chemical Engineering Science 3 (Suppl.) (1954) 341. [32] M.C. Biesinger, L.W.M. Lau, A.R. Gerson, R.S.C. Smart, Applied Surface Science 257 (2010) 887. [33] J.C. Vedrine, J.M.M. Millet, J.C. Volta, Catalysis Today 32 (1996) 115. [34] O.V. Safonova, M. Florea, J. Bilde, P. Delichere, J.M.M. Millet, Journal of Catalysis 268 (2009) 156. [35] S. Sugiyama, T. Osaka, Y. Hirata, K.I. Sotowa, Applied Catalysis A – General 312 (2006) 52. [36] R.X. Valenzuela, E.A. Mamedov, V. Cortes Corberan, Reaction Kinetics and Catalysis Letters 55 (1995) 213. ´ [37] A. Klisinska, A. Haras, K. Samson, M. Witko, B. Grzybowska, Journal of Molecular Catalysis A: Chemical 210 (2004) 87. [38] M. Witko, R. Grybos, R. Tokarz-Sobieraj, Topics in Catalysis 38 (2006) 75. [39] M. Chen, X. Zheng, Journal of Molecular Catalysis A 201 (2003) 161. [40] X. Wang, Y.C. Xie, Applied Catalysis B 35 (2001) 85. [41] S. Deng, H.Q. Li, S.G. Li, Y. Zhang, Journal of Molecular Catalysis A: Chemical 268 (2007) 169.
Please cite this article in press as: M. Jin, et al., Effect of additives doping on catalytic properties of Mg3 (VO4 )2 catalysts in oxidative dehydrogenation of cyclohexane, Catal. Today (2012), http://dx.doi.org/10.1016/j.cattod.2012.09.019