- Email: [email protected]
Effect of pH in the preparation of ZnFe2O4 for oxidative dehydrogenation of n-butene to 1,3-butadiene: Correlation between catalytic performance and surface acidity of ZnFe2O4
Available online at www.sciencedirect.com
Catalysis Communications 9 (2008) 1137–1142 www.elsevier.com/locate/catcom
Effect of pH in the preparation of ZnFe2O4 for oxidative dehydrogenation of n-butene to 1,3-butadiene: Correlation between catalytic performance and surface acidity of ZnFe2O4 Howon Lee a, Ji Chul Jung a, Heesoo Kim a, Young-Min Chung b, Tae Jin Kim b, Seong Jun Lee b, Seung-Hoon Oh b, Yong Seung Kim b, In Kyu Song a,* a
School of Chemical and Biological Engineering, Institute of Chemical Processes, Seoul National University, Shinlim-dong, Kwanak-ku, Seoul 151-744, South Korea b SK Energy Corporation, Yuseong-ku, Daejeon 305-712, South Korea Received 15 August 2007; received in revised form 25 October 2007; accepted 25 October 2007 Available online 30 October 2007
Abstract A series of zinc ferrite (ZnFe2O4) catalysts were prepared by a co-precipitation method with a variation of pH value (pH 3–12), and applied to the oxidative dehydrogenation of n-butene to 1,3-butadiene. It was found that a zinc ferrite phase was formed at pH 6–12. Conversion of n-butene, selectivity for 1,3-butadiene, and yield for 1,3-butadiene were very high at low pH value (pH 6–10), but were drastically decreased with increasing pH value (pH 11–12). NH3-TPD experiments were conducted to correlate the acid property with the catalytic performance of zinc ferrite catalysts. It was revealed that the catalytic performance was closely related to the surface acidity of the catalysts. Conversion of n-butene was increased with increasing surface acidity of the catalysts. Among the catalysts tested, the zinc ferrite catalyst prepared at pH 9 with the largest surface acidity showed the best catalytic performance in the oxidative dehydrogenation of n-butene. Ó 2007 Elsevier B.V. All rights reserved. Keywords: Zinc ferrite; n-Butene; 1,3-Butadiene; Oxidative dehydrogenation; Effect of pH; Surface acidity
1. Introduction Selective oxidation of olefins has been considered to be an important subject for the production of a number of chemical intermediates in the petrochemical industries [1–4]. In particular, oxidative dehydrogenation of n-butene has attracted much attention as a promising process for producing 1,3-butadiene [5–7], which is an important raw material for manufacturing a large number of chemical products. Various metal ferrite catalysts such as magnesium ferrite [8–11], zinc ferrite [9,10,12], cobalt ferrite [10,13], and copper ferrite [13] have been employed in the oxidative dehydrogenation of n-butene. Among these cata*
Corresponding author. Tel.: +82 2 880 9227; fax: +82 2 889 7415. E-mail address: [email protected] (I.K. Song).
1566-7367/$ - see front matter Ó 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.catcom.2007.10.023
lysts, zinc ferrite has been widely investigated as an efficient catalyst for this reaction [14–18]. The oxidative dehydrogenation of n-butene is strongly affected by various catalytic properties. It is known that the major factors determining the catalytic performance in the oxidative dehydrogenation of n-butene are different depending on the catalyst system. A number of investigations on the reaction mechanism for oxidative dehydrogenation of n-butene over ferrite-type catalysts have been made in order to elucidate the catalytic active sites [11,19,20]. It has been generally accepted that the reaction mechanism by way of p-allyl intermediate is the most feasible reaction pathway for the formation of 1,3-butadiene from n-butene [9,10,16,21]. According to the reaction mechanism [16,22], the oxidative dehydrogenation of n-butene over ferrite-type catalyst
1138
H. Lee et al. / Catalysis Communications 9 (2008) 1137–1142
consists of five sequential elementary steps; (i) chemisorption of n-butene, (ii) abstraction of a-hydrogen from nbutene to form p-allyl intermediate, (iii) abstraction of one more hydrogen atom from p-allyl intermediate to form 1,3-butadiene, (iv) dissociation of oxygen in the gas phase to make up oxygen vacancy in the catalyst, and (v) reoxidation of multivalent cations. Many researchers agree that the abstraction of a-hydrogen from n-butene is the ratedetermining step in the oxidative dehydrogenation of n-butene [10,21,22]. It has also been reported that acid property of the catalyst plays a key role in abstracting an a-hydrogen atom from n-butene to form p-allyl intermediate [10,23–25]. Therefore, it is believed that acid property of ferrite-type catalyst would be an important factor determining the catalytic performance in the oxidative dehydrogenation of n-butene. Co-precipitation method has been generally used in the preparation of zinc ferrite catalyst. It has been reported that pH value during the co-precipitation step is an important factor for the successful formation of active phase in the preparation of zinc ferrite catalyst [17,26–29]. This implies that pH value in the preparation of zinc ferrite catalyst by a co-precipitation method can serve as a crucial factor determining the catalytic performance of zinc ferrite catalyst in the oxidative dehydrogenation of n-butene. Nonetheless, no systematic investigation on the effect of pH value in the preparation of zinc ferrite catalyst on the catalytic performance in the oxidative dehydrogenation of n-butene has been reported yet. In this work, a series of zinc ferrite (ZnFe2O4) catalysts were prepared by a co-precipitation method with a variation of pH value (pH 3–12) for use in the oxidative dehydrogenation of n-butene to 1,3-butadiene. The acid properties of zinc ferrite catalysts were measured by NH3-TPD experiments. A correlation between catalytic performance and acid property of zinc ferrite catalysts was then established. 2. Experimental 2.1. Preparation of zinc ferrite (ZnFe2O4) catalysts A series of zinc ferrite (ZnFe2O4) catalysts were prepared by a co-precipitation method with a variation of pH value during the co-precipitation step. Known amounts of zinc chloride (ZnCl2 from Sigma–Aldrich) and iron chloride (FeCl3 Æ 6H2O from Sigma–Aldrich) were dissolved in distilled water. The metal precursor solution and an aqueous sodium hydroxide solution (3N) were then added dropwise into distilled water for co-precipitation under vigorous stirring. During the co-precipitation step, pH value of the mixed solution was precisely controlled using aqueous sodium hydroxide solution. The pH value was varied from 3 to 12 with an interval of 1 in the preparation of zinc ferrite (ZnFe2O4) catalysts. After the resulting solution was stirred vigorously at room temperature for 12 h, it was aged overnight at room temperature. The precipitate
was filtered and washed to obtain a solid product. The solid product was dried at 175 °C for 16 h, and finally, it was calcined at 650 °C for 6 h to yield the zinc ferrite (ZnFe2O4) catalyst. 2.2. Characterization The formation of zinc ferrite (ZnFe2O4) catalysts was confirmed by XRD (MAC Science, M18XHF-SRA) measurements. The atomic ratios of constituent metal components were determined by ICP-AES (Shimadz, ICP1000IV) analyses. The surface areas of zinc ferrite catalysts were measured using a BET apparatus (Micromeritics, ASAP 2010). The acid properties of zinc ferrite catalysts were measured by NH3-TPD experiments. Each catalyst (0.1 g) was charged into a tubular quartz reactor of the conventional TPD apparatus. The catalyst was pretreated at 200 °C for 1 h under a flow of helium (20 ml/min) to remove any physisorbed organic molecules. Twenty millilitres of NH3 was then pulsed into the reactor every minute at room temperature under a flow of helium (5 ml/min), until the acid sites were saturated with NH3. The physisorbed NH3 was removed by evacuating the catalyst sample at 50 °C for 1 h. The furnace temperature was increased from room temperature to 600 °C at a heating rate of 5 °C/min under a flow of helium (10 ml/min). The desorbed NH3 was detected using a GC-MSD (Agilent, MSD-6890N GC). 2.3. Oxidative dehydrogenation of n-butene The oxidative dehydrogenation of n-butene to 1,3-butadiene was carried out in a continuous flow fixed-bed reactor in the presence of air and steam. Water was sufficiently vaporized by passing through a pre-heating zone and continuously fed into the reactor together with n-butene and air. C4 raffinate-3 containing 57.9 wt% n-butene (1-butene (7.5 wt%) + trans-2-butene (33.9 wt%) + cis-2butene (16.5 wt%)) was used as a n-butene source, and air was used as an oxygen source (nitrogen in air served as a carrier gas). The feed composition was fixed at nbutene:oxygen:steam = 1:0.75:15. Prior to the catalytic reaction, each catalyst was pretreated at 470 °C for 1 h with an air stream. The catalytic reaction was carried out at 420 °C. The GHSV (gas hourly space velocity) was fixed at 475 h1 on the basis of n-butene. Reaction products were periodically sampled and analyzed with gas chromatographs. Conversion of n-butene and selectivity for 1,3butadiene were calculated on the basis of carbon balance as follows. Yield for 1,3-butadiene was calculated by multiplying conversion and selectivity. moles of n-butene reacted Conversion of n-butene ¼ moles of n-butene supplied Selectivity for 1; 3-butadiene moles of 1; 3-butadiene formed ¼ moles of n-butene reacted
H. Lee et al. / Catalysis Communications 9 (2008) 1137–1142
1139
Table 1 Atomic ratios and BET surface areas of zinc ferrite catalysts prepared at different pH
3. Results and discussion 3.1. Formation of zinc ferrite (ZnFe2O4) catalysts The formation of zinc ferrite (ZnFe2O4) catalysts was confirmed by XRD measurements. Fig. 1 shows the XRD patterns of zinc ferrite catalysts prepared at different pH (pH 6–12). Each phase was identified by its characteristic diffraction peaks using JCPDS. It is interesting to note that a different phase was formed depending on the pH value during the co-precipitation step. Only a pure aFe2O3 phase was formed in the catalysts prepared at pH 3–5, indicating that acidic condition (pH 3–5) during the co-precipitation step was not favorable for the formation of zinc ferrite catalyst. In other words, the catalysts prepared at pH 3–5 are not the zinc ferrite catalysts. Therefore, any investigations on the catalysts prepared at pH 3–5 were excluded in this work, and our major studies on the zinc ferrite catalysts for oxidative dehydrogenation of n-butene were focused on the catalysts prepared at pH 6– 12. It is noticeable that a mixed phase of a-Fe2O3 and ZnFe2O4 was detected in the catalyst prepared at pH 6, and that only a small portion of a-Fe2O3 phase was observed in the catalyst prepared at pH 7. However, the catalysts prepared at pH 8–12 retained a pure zinc ferrite (ZnFe2O4) phase. The atomic ratios of Fe/Zn in the catalysts prepared at different pH are summarized in Table 1. The atomic ratios of Fe/Zn in the catalysts prepared at pH 8–12 were in the range of 2.07–2.29, in good agreement with the theoretical value of 2.0. On the other hand, those in the catalysts prepared at pH 6–7 were much higher than 2.0, due to the presence of a-Fe2O3 phase. This result is well consistent with the XRD result (Fig. 1). The BET surface areas of zinc ferrite catalysts prepared
pH
Atomic ratio of Fe/Zn
BET surface area (m2/g)
6 7 8 9 10 11 12
3.41 2.44 2.29 2.15 2.08 2.07 2.08
24.9 17.5 31.3 22.8 8.66 24.8 19.2
at different pH are also summarized in Table 1. The BET surface areas were measured three times for each sample to confirm consistency and reproducibility, and the average values are listed in Table 1. However, the BET surface areas of the catalysts showed no consistent trend with respect to pH value. 3.2. Catalytic performance of zinc ferrite (ZnFe2O4) catalysts Fig. 2 shows the catalytic performance of zinc ferrite catalysts in the oxidative dehydrogenation of n-butene at 420 °C after a 6 h-reaction, plotted as a function of pH value. Conversion of n-butene and selectivity for 1,3-butadiene were very high at low pH value (pH 6– 10), but were drastically decreased with increasing pH value (pH 11–12). Yield for 1,3-butadiene also showed the same trend as conversion of n-butene and selectivity for 1,3-butadiene with respect to pH value. Among the catalysts tested, the zinc ferrite catalyst prepared at pH 9 showed the best catalytic performance. The yield for 1,3-butadiene over the catalyst prepared at pH 9 was found to be 80.2%.
-Fe2O3
ZnFe2O4 pH=12
Intensity (A. U.)
pH=11
pH=10
pH=9
pH=8
pH=7
pH=6
10
20
30
40
50
60
70
2 Theta (degree) Fig. 1. XRD patterns of zinc ferrite catalysts prepared at different pH (pH 6–12).
80
H. Lee et al. / Catalysis Communications 9 (2008) 1137–1142
90
100
80
90
70
80
60
70
50
60
40
50
30
40
20
30
10
b
90 80
Yield for 1,3-butadiene (%)
Conversion of n-butene (%)
a
Selectivity for 1,3-butadiene (%)
1140
70 60 50 40 30 20 10 0
20 6
7
8
9
10
11
12
6
7
8
9
pH
10
11
12
pH
It is noticeable that the catalysts prepared at pH 6–7 exhibited high catalytic performance, although these catalysts retained a a-Fe2O3 phase that was less active in the oxidative dehydrogenation of n-butene. However, this catalytic behavior can be explained by the synergistic effect of a-Fe2O3 and ZnFe2O4. It was previously reported that the phase cooperation caused by the close contact between aFe2O3 and ZnFe2O4, which facilitated oxygen transfer from a-Fe2O3 phase to ZnFe2O4 phase, led to the enhanced catalytic performance in the oxidative dehydrogenation of n-butene [15]. 3.3. Correlation between catalytic performance and surface acidity
Mass signal intensity (A. U.)
Fig. 2. Catalytic performance of zinc ferrite catalysts in the oxidative dehydrogenation of n-butene at 420 °C after a 6 h-reaction, plotted as a function of pH value: (a) conversion of n-butene and selectivity for 1,3-butadiene, and (b) yield for 1,3-butadiene.
4x10
4
3x10
4
2x10
4
1x10
4
pH=9
pH=11
pH=12
0 0
100
200
300
400
500
600
Temperature (oC)
An attempt has been made to correlate the catalytic performance (Fig. 2) with the BET surface area (Table 1) of zinc ferrite catalysts. However, the catalytic performance was not directly correlated with the BET surface area of zinc ferrite catalysts. Furthermore, we measured the oxygen mobility of zinc ferrite catalysts using temperature-programmed technique with an attempt to find a correlation between catalytic performance and oxygen mobility. However, no significant difference in oxygen mobility between zinc ferrite catalysts was observed. This indicates that the oxygen mobility is not a crucial factor determining the catalytic performance in the oxidative dehydrogenation of nbutene, as far as zinc ferrite catalysts examined in this work are concerned. In order to investigate the effect of acid properties on the catalytic performance of zinc ferrite catalysts, NH3-TPD experiments were conducted over the catalysts. Fig. 3 shows the NH3-TPD profiles of selected zinc ferrite catalysts. All the catalysts showed a broad NH3-TPD peak. The zinc ferrite catalysts exhibited a significant difference in total acidity (peak area) with respect to pH value. However, no great difference in acid strength (peak temperature) was observed with respect to pH value. We attempted to correlate the catalytic performance with the acid properties of zinc ferrite catalysts. Unfortunately, acid
Fig. 3. NH3-TPD profiles of selected zinc ferrite catalysts.
strength was not correlated with the catalytic performance of zinc ferrite catalysts. Furthermore, total acidity was not directly correlated with the catalytic performance of zinc ferrite catalysts (Fig. 4). However, a reliable correlation between conversion of n-butene and surface acidity of the catalyst was observed. The surface acidity was defined as the amount of adsorbed NH3 per unit surface area of a catalyst. Fig. 5 shows a comprehensive correlation between conversion of n-butene and surface acidity of zinc ferrite catalysts. The correlation clearly shows that the conversion of n-butene was increased with increasing surface acidity of the catalyst. The yield for 1,3-butadiene also showed the same trend as the conversion of n-butene with respect to surface acidity. Among the catalysts tested, the zinc ferrite catalyst prepared at pH 9 with the largest surface acidity showed the best catalytic performance. As mentioned earlier, the acid property of zinc ferrite catalysts plays an important role in activating n-butene by the abstraction of a-hydrogen to form p-allyl intermediate [10,23–25]. Furthermore, it has been reported that sufficient acid sites are required to adsorb and activate n-butene (a base molecule)
H. Lee et al. / Catalysis Communications 9 (2008) 1137–1142 100
Conversion of n-butene (%)
90 80
pH=7
pH=10
pH=9 pH=8
pH=6
70 60 50 40 pH=11
30 20
pH=12
10 0 10
20
30
40
50
60
Acknowledgements
Fig. 4. A possible correlation between conversion of n-butene and total acidity of zinc ferrite catalysts.
100
This work was supported by SK Energy Corporation (POST-BK21 Program) and Korea Energy Management Corporation (2005-01-0090-3-010). References
90
Conversion of n-butene (%)
(ZnFe2O4) phase. In the oxidative dehydrogenation of nbutene, the catalytic performance of zinc ferrite catalysts was very high at low pH value (pH 6–10), but was drastically decreased with increasing pH value (pH 11–12). Among the catalysts tested, the zinc ferrite catalyst prepared at pH 9 showed the best catalytic performance. It was revealed from NH3-TPD experiments that the catalytic performance was closely related to the surface acidity of the catalysts. The catalytic performance was increased with increasing surface acidity of the catalysts. The largest surface acidity of zinc ferrite catalyst prepared at pH 9 was responsible for its highest catalytic performance in the oxidative dehydrogenation of n-butene.
70
Total acidity (μmol-NH3)
pH=9 pH=10 pH=8 pH=6 pH=7
80 70 60 50 40 pH=11
30 20
1141
pH=12
10 0 0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
2.4
Surface acidity (μmol-NH3/m2) Fig. 5. A correlation between conversion of n-butene and surface acidity of zinc ferrite catalysts.
[10]. Therefore, it is reasonable to expect that a catalyst with large surface acidity (with large capacity to activate n-butene per unit surface area) is favorable in the oxidative dehydrogenation of n-butene. It is concluded that the surface acidity of zinc ferrite catalysts is a crucial factor determining the catalytic performance in the oxidative dehydrogenation of n-butene. 4. Conclusions A series of zinc ferrite (ZnFe2O4) catalysts were prepared by a co-precipitation method with a variation of pH value (pH 3–12) for use in the oxidative dehydrogenation of n-butene to 1,3-butadiene. It was found from XRD and ICP-AES analyses that a zinc ferrite phase was not formed at pH 3–5, while a mixed phase of a-Fe2O3 and ZnFe2O4 was formed at pH 6–7. On the other hand, the catalysts prepared at pH 8–12 retained a pure zinc ferrite
[1] S.C. Oh, H.P. Lee, H.T. Kim, K.O. Yoo, Korean J. Chem. Eng. 16 (1999) 543. [2] Y.H. Kim, H.S. Yang, Korean J. Chem. Eng. 17 (2000) 357. [3] H.H. Kung, Ind. Eng. Chem. Prod. Res. Dev. 25 (1986) 171. [4] N.J. Jebarathinnam, M. Eswaramoorthy, V. Krishnasamy, Appl. Catal. A 145 (1996) 57. [5] F. Hong, B.L. Yang, L.H. Schwartz, H.H. Kung, J. Phys. Chem. 88 (1984) 2525. [6] B.J. Liaw, D.S. Cheng, B.L. Yang, J. Catal. 118 (1989) 312. [7] J.C. Jung, H. Kim, Y.S. Kim, Y.-M. Chung, T.J. Kim, S.J. Lee, S.-H. Oh, I.K. Song, Appl. Catal. A 317 (2007) 244. [8] B.L. Yang, D.S. Cheng, S.B. Lee, Appl. Catal. 70 (1991) 161. [9] R.J. Rennard, W.L. Kehl, J. Catal. 21 (1971) 282. [10] H.H. Kung, M.C. Kung, Adv. Catal. 33 (1985) 159. [11] M.A. Gibson, J.W. Hightower, J. Catal. 41 (1976) 431. [12] J.A. Toledo-Antonio, N. Nava, M. Martı´nez, X. Bokhimi, Appl. Catal. A 234 (2002) 137. [13] W.R. Cares, J.W. Hightower, J. Catal. 23 (1971) 193. [14] H.H. Kung, B. Kundalkar, M.C. Kung, W.H. Cheng, J. Phys. Chem. 84 (1980) 382. [15] M. Zhang, R. Lan, J. Liu, X. Chen, W. Zhou, J. Chem. Soc. Faraday Trans. 88 (1992) 637. [16] F.E. Massoth, D.A. Scarpiello, J. Catal. 21 (1971) 294. [17] F.-Y. Qiu, L.-T. Weng, E. Sham, P. Ruiz, B. Delmon, Appl. Catal. 51 (1989) 235. [18] J.A. Toledo-Antonio, P. Bosch, M.A. Valenzuela, A. Montoya, N. Nava, J. Mol. Catal. A 125 (1997) 53. [19] F.V. Aliev, E.N. Amirbekov, A.D. Kuliev, V.Y. Danyushevskii, V.D. Nissenbaum, V.I. Yakerson, A.M. Rubinshtein, Kinet. Katal. 22 (1981) 329. [20] V. Nicolescu, R. Manaila, M.A. Chaar, G. Filoti, Revue Roum. Chim. 31 (1986) 99. [21] M.A. Gibson, J.W. Hightower, J. Catal. 41 (1976) 420. [22] W.-Q. Xu, Y.-G. Yin, G.-Y. Li, S. Chen, Appl. Catal. A 89 (1992) 131. [23] C.F. Cullis, D.J. Hucknall, in: G.C. Bond, G. Webb (Eds.), A Specialist Periodical Report: Catalysis, vol. 5, Royal Chemical Society, London, 1982, p. 273. [24] R.K. Grasselli, J.D. Burrington, Adv. Catal. 30 (1981) 133. [25] J.C. Jung, H. Lee, H. Kim, Y.-M. Chung, T.J. Kim, S.J. Lee, S.-H. Oh, Y.S. Kim, I.K. Song, Catal. Commun. (2007), doi:10.1016/ j.catcom.2007.07.031.
1142
H. Lee et al. / Catalysis Communications 9 (2008) 1137–1142
[26] B. Jeyadevan, C.N. Chinnasamy, K. Shinoda, K. Tohji, J. Appl. Phys. 93 (2003) 8450. [27] G. Vaidyanathan, S. Sendhilnathan, R. Arulmurugan, J. Magn. Magn. Mater. 313 (2007) 293.
[28] I.H. Gul, A.Z. Abbasi, F. Amin, M. Anis-ur-Rehman, A. Maqsood, J. Magn. Magn. Mater. 311 (2007) 494. [29] R.K. Sharma, O. Suwalka, N. Lakshmi, K. Venugopalan, A. Banerjee, P.A. Joy, Mater. Lett. 59 (2005) 3402.
Catalysis Communications 9 (2008) 1137–1142 www.elsevier.com/locate/catcom
Effect of pH in the preparation of ZnFe2O4 for oxidative dehydrogenation of n-butene to 1,3-butadiene: Correlation between catalytic performance and surface acidity of ZnFe2O4 Howon Lee a, Ji Chul Jung a, Heesoo Kim a, Young-Min Chung b, Tae Jin Kim b, Seong Jun Lee b, Seung-Hoon Oh b, Yong Seung Kim b, In Kyu Song a,* a
School of Chemical and Biological Engineering, Institute of Chemical Processes, Seoul National University, Shinlim-dong, Kwanak-ku, Seoul 151-744, South Korea b SK Energy Corporation, Yuseong-ku, Daejeon 305-712, South Korea Received 15 August 2007; received in revised form 25 October 2007; accepted 25 October 2007 Available online 30 October 2007
Abstract A series of zinc ferrite (ZnFe2O4) catalysts were prepared by a co-precipitation method with a variation of pH value (pH 3–12), and applied to the oxidative dehydrogenation of n-butene to 1,3-butadiene. It was found that a zinc ferrite phase was formed at pH 6–12. Conversion of n-butene, selectivity for 1,3-butadiene, and yield for 1,3-butadiene were very high at low pH value (pH 6–10), but were drastically decreased with increasing pH value (pH 11–12). NH3-TPD experiments were conducted to correlate the acid property with the catalytic performance of zinc ferrite catalysts. It was revealed that the catalytic performance was closely related to the surface acidity of the catalysts. Conversion of n-butene was increased with increasing surface acidity of the catalysts. Among the catalysts tested, the zinc ferrite catalyst prepared at pH 9 with the largest surface acidity showed the best catalytic performance in the oxidative dehydrogenation of n-butene. Ó 2007 Elsevier B.V. All rights reserved. Keywords: Zinc ferrite; n-Butene; 1,3-Butadiene; Oxidative dehydrogenation; Effect of pH; Surface acidity
1. Introduction Selective oxidation of olefins has been considered to be an important subject for the production of a number of chemical intermediates in the petrochemical industries [1–4]. In particular, oxidative dehydrogenation of n-butene has attracted much attention as a promising process for producing 1,3-butadiene [5–7], which is an important raw material for manufacturing a large number of chemical products. Various metal ferrite catalysts such as magnesium ferrite [8–11], zinc ferrite [9,10,12], cobalt ferrite [10,13], and copper ferrite [13] have been employed in the oxidative dehydrogenation of n-butene. Among these cata*
Corresponding author. Tel.: +82 2 880 9227; fax: +82 2 889 7415. E-mail address: [email protected] (I.K. Song).
1566-7367/$ - see front matter Ó 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.catcom.2007.10.023
lysts, zinc ferrite has been widely investigated as an efficient catalyst for this reaction [14–18]. The oxidative dehydrogenation of n-butene is strongly affected by various catalytic properties. It is known that the major factors determining the catalytic performance in the oxidative dehydrogenation of n-butene are different depending on the catalyst system. A number of investigations on the reaction mechanism for oxidative dehydrogenation of n-butene over ferrite-type catalysts have been made in order to elucidate the catalytic active sites [11,19,20]. It has been generally accepted that the reaction mechanism by way of p-allyl intermediate is the most feasible reaction pathway for the formation of 1,3-butadiene from n-butene [9,10,16,21]. According to the reaction mechanism [16,22], the oxidative dehydrogenation of n-butene over ferrite-type catalyst
1138
H. Lee et al. / Catalysis Communications 9 (2008) 1137–1142
consists of five sequential elementary steps; (i) chemisorption of n-butene, (ii) abstraction of a-hydrogen from nbutene to form p-allyl intermediate, (iii) abstraction of one more hydrogen atom from p-allyl intermediate to form 1,3-butadiene, (iv) dissociation of oxygen in the gas phase to make up oxygen vacancy in the catalyst, and (v) reoxidation of multivalent cations. Many researchers agree that the abstraction of a-hydrogen from n-butene is the ratedetermining step in the oxidative dehydrogenation of n-butene [10,21,22]. It has also been reported that acid property of the catalyst plays a key role in abstracting an a-hydrogen atom from n-butene to form p-allyl intermediate [10,23–25]. Therefore, it is believed that acid property of ferrite-type catalyst would be an important factor determining the catalytic performance in the oxidative dehydrogenation of n-butene. Co-precipitation method has been generally used in the preparation of zinc ferrite catalyst. It has been reported that pH value during the co-precipitation step is an important factor for the successful formation of active phase in the preparation of zinc ferrite catalyst [17,26–29]. This implies that pH value in the preparation of zinc ferrite catalyst by a co-precipitation method can serve as a crucial factor determining the catalytic performance of zinc ferrite catalyst in the oxidative dehydrogenation of n-butene. Nonetheless, no systematic investigation on the effect of pH value in the preparation of zinc ferrite catalyst on the catalytic performance in the oxidative dehydrogenation of n-butene has been reported yet. In this work, a series of zinc ferrite (ZnFe2O4) catalysts were prepared by a co-precipitation method with a variation of pH value (pH 3–12) for use in the oxidative dehydrogenation of n-butene to 1,3-butadiene. The acid properties of zinc ferrite catalysts were measured by NH3-TPD experiments. A correlation between catalytic performance and acid property of zinc ferrite catalysts was then established. 2. Experimental 2.1. Preparation of zinc ferrite (ZnFe2O4) catalysts A series of zinc ferrite (ZnFe2O4) catalysts were prepared by a co-precipitation method with a variation of pH value during the co-precipitation step. Known amounts of zinc chloride (ZnCl2 from Sigma–Aldrich) and iron chloride (FeCl3 Æ 6H2O from Sigma–Aldrich) were dissolved in distilled water. The metal precursor solution and an aqueous sodium hydroxide solution (3N) were then added dropwise into distilled water for co-precipitation under vigorous stirring. During the co-precipitation step, pH value of the mixed solution was precisely controlled using aqueous sodium hydroxide solution. The pH value was varied from 3 to 12 with an interval of 1 in the preparation of zinc ferrite (ZnFe2O4) catalysts. After the resulting solution was stirred vigorously at room temperature for 12 h, it was aged overnight at room temperature. The precipitate
was filtered and washed to obtain a solid product. The solid product was dried at 175 °C for 16 h, and finally, it was calcined at 650 °C for 6 h to yield the zinc ferrite (ZnFe2O4) catalyst. 2.2. Characterization The formation of zinc ferrite (ZnFe2O4) catalysts was confirmed by XRD (MAC Science, M18XHF-SRA) measurements. The atomic ratios of constituent metal components were determined by ICP-AES (Shimadz, ICP1000IV) analyses. The surface areas of zinc ferrite catalysts were measured using a BET apparatus (Micromeritics, ASAP 2010). The acid properties of zinc ferrite catalysts were measured by NH3-TPD experiments. Each catalyst (0.1 g) was charged into a tubular quartz reactor of the conventional TPD apparatus. The catalyst was pretreated at 200 °C for 1 h under a flow of helium (20 ml/min) to remove any physisorbed organic molecules. Twenty millilitres of NH3 was then pulsed into the reactor every minute at room temperature under a flow of helium (5 ml/min), until the acid sites were saturated with NH3. The physisorbed NH3 was removed by evacuating the catalyst sample at 50 °C for 1 h. The furnace temperature was increased from room temperature to 600 °C at a heating rate of 5 °C/min under a flow of helium (10 ml/min). The desorbed NH3 was detected using a GC-MSD (Agilent, MSD-6890N GC). 2.3. Oxidative dehydrogenation of n-butene The oxidative dehydrogenation of n-butene to 1,3-butadiene was carried out in a continuous flow fixed-bed reactor in the presence of air and steam. Water was sufficiently vaporized by passing through a pre-heating zone and continuously fed into the reactor together with n-butene and air. C4 raffinate-3 containing 57.9 wt% n-butene (1-butene (7.5 wt%) + trans-2-butene (33.9 wt%) + cis-2butene (16.5 wt%)) was used as a n-butene source, and air was used as an oxygen source (nitrogen in air served as a carrier gas). The feed composition was fixed at nbutene:oxygen:steam = 1:0.75:15. Prior to the catalytic reaction, each catalyst was pretreated at 470 °C for 1 h with an air stream. The catalytic reaction was carried out at 420 °C. The GHSV (gas hourly space velocity) was fixed at 475 h1 on the basis of n-butene. Reaction products were periodically sampled and analyzed with gas chromatographs. Conversion of n-butene and selectivity for 1,3butadiene were calculated on the basis of carbon balance as follows. Yield for 1,3-butadiene was calculated by multiplying conversion and selectivity. moles of n-butene reacted Conversion of n-butene ¼ moles of n-butene supplied Selectivity for 1; 3-butadiene moles of 1; 3-butadiene formed ¼ moles of n-butene reacted
H. Lee et al. / Catalysis Communications 9 (2008) 1137–1142
1139
Table 1 Atomic ratios and BET surface areas of zinc ferrite catalysts prepared at different pH
3. Results and discussion 3.1. Formation of zinc ferrite (ZnFe2O4) catalysts The formation of zinc ferrite (ZnFe2O4) catalysts was confirmed by XRD measurements. Fig. 1 shows the XRD patterns of zinc ferrite catalysts prepared at different pH (pH 6–12). Each phase was identified by its characteristic diffraction peaks using JCPDS. It is interesting to note that a different phase was formed depending on the pH value during the co-precipitation step. Only a pure aFe2O3 phase was formed in the catalysts prepared at pH 3–5, indicating that acidic condition (pH 3–5) during the co-precipitation step was not favorable for the formation of zinc ferrite catalyst. In other words, the catalysts prepared at pH 3–5 are not the zinc ferrite catalysts. Therefore, any investigations on the catalysts prepared at pH 3–5 were excluded in this work, and our major studies on the zinc ferrite catalysts for oxidative dehydrogenation of n-butene were focused on the catalysts prepared at pH 6– 12. It is noticeable that a mixed phase of a-Fe2O3 and ZnFe2O4 was detected in the catalyst prepared at pH 6, and that only a small portion of a-Fe2O3 phase was observed in the catalyst prepared at pH 7. However, the catalysts prepared at pH 8–12 retained a pure zinc ferrite (ZnFe2O4) phase. The atomic ratios of Fe/Zn in the catalysts prepared at different pH are summarized in Table 1. The atomic ratios of Fe/Zn in the catalysts prepared at pH 8–12 were in the range of 2.07–2.29, in good agreement with the theoretical value of 2.0. On the other hand, those in the catalysts prepared at pH 6–7 were much higher than 2.0, due to the presence of a-Fe2O3 phase. This result is well consistent with the XRD result (Fig. 1). The BET surface areas of zinc ferrite catalysts prepared
pH
Atomic ratio of Fe/Zn
BET surface area (m2/g)
6 7 8 9 10 11 12
3.41 2.44 2.29 2.15 2.08 2.07 2.08
24.9 17.5 31.3 22.8 8.66 24.8 19.2
at different pH are also summarized in Table 1. The BET surface areas were measured three times for each sample to confirm consistency and reproducibility, and the average values are listed in Table 1. However, the BET surface areas of the catalysts showed no consistent trend with respect to pH value. 3.2. Catalytic performance of zinc ferrite (ZnFe2O4) catalysts Fig. 2 shows the catalytic performance of zinc ferrite catalysts in the oxidative dehydrogenation of n-butene at 420 °C after a 6 h-reaction, plotted as a function of pH value. Conversion of n-butene and selectivity for 1,3-butadiene were very high at low pH value (pH 6– 10), but were drastically decreased with increasing pH value (pH 11–12). Yield for 1,3-butadiene also showed the same trend as conversion of n-butene and selectivity for 1,3-butadiene with respect to pH value. Among the catalysts tested, the zinc ferrite catalyst prepared at pH 9 showed the best catalytic performance. The yield for 1,3-butadiene over the catalyst prepared at pH 9 was found to be 80.2%.
-Fe2O3
ZnFe2O4 pH=12
Intensity (A. U.)
pH=11
pH=10
pH=9
pH=8
pH=7
pH=6
10
20
30
40
50
60
70
2 Theta (degree) Fig. 1. XRD patterns of zinc ferrite catalysts prepared at different pH (pH 6–12).
80
H. Lee et al. / Catalysis Communications 9 (2008) 1137–1142
90
100
80
90
70
80
60
70
50
60
40
50
30
40
20
30
10
b
90 80
Yield for 1,3-butadiene (%)
Conversion of n-butene (%)
a
Selectivity for 1,3-butadiene (%)
1140
70 60 50 40 30 20 10 0
20 6
7
8
9
10
11
12
6
7
8
9
pH
10
11
12
pH
It is noticeable that the catalysts prepared at pH 6–7 exhibited high catalytic performance, although these catalysts retained a a-Fe2O3 phase that was less active in the oxidative dehydrogenation of n-butene. However, this catalytic behavior can be explained by the synergistic effect of a-Fe2O3 and ZnFe2O4. It was previously reported that the phase cooperation caused by the close contact between aFe2O3 and ZnFe2O4, which facilitated oxygen transfer from a-Fe2O3 phase to ZnFe2O4 phase, led to the enhanced catalytic performance in the oxidative dehydrogenation of n-butene [15]. 3.3. Correlation between catalytic performance and surface acidity
Mass signal intensity (A. U.)
Fig. 2. Catalytic performance of zinc ferrite catalysts in the oxidative dehydrogenation of n-butene at 420 °C after a 6 h-reaction, plotted as a function of pH value: (a) conversion of n-butene and selectivity for 1,3-butadiene, and (b) yield for 1,3-butadiene.
4x10
4
3x10
4
2x10
4
1x10
4
pH=9
pH=11
pH=12
0 0
100
200
300
400
500
600
Temperature (oC)
An attempt has been made to correlate the catalytic performance (Fig. 2) with the BET surface area (Table 1) of zinc ferrite catalysts. However, the catalytic performance was not directly correlated with the BET surface area of zinc ferrite catalysts. Furthermore, we measured the oxygen mobility of zinc ferrite catalysts using temperature-programmed technique with an attempt to find a correlation between catalytic performance and oxygen mobility. However, no significant difference in oxygen mobility between zinc ferrite catalysts was observed. This indicates that the oxygen mobility is not a crucial factor determining the catalytic performance in the oxidative dehydrogenation of nbutene, as far as zinc ferrite catalysts examined in this work are concerned. In order to investigate the effect of acid properties on the catalytic performance of zinc ferrite catalysts, NH3-TPD experiments were conducted over the catalysts. Fig. 3 shows the NH3-TPD profiles of selected zinc ferrite catalysts. All the catalysts showed a broad NH3-TPD peak. The zinc ferrite catalysts exhibited a significant difference in total acidity (peak area) with respect to pH value. However, no great difference in acid strength (peak temperature) was observed with respect to pH value. We attempted to correlate the catalytic performance with the acid properties of zinc ferrite catalysts. Unfortunately, acid
Fig. 3. NH3-TPD profiles of selected zinc ferrite catalysts.
strength was not correlated with the catalytic performance of zinc ferrite catalysts. Furthermore, total acidity was not directly correlated with the catalytic performance of zinc ferrite catalysts (Fig. 4). However, a reliable correlation between conversion of n-butene and surface acidity of the catalyst was observed. The surface acidity was defined as the amount of adsorbed NH3 per unit surface area of a catalyst. Fig. 5 shows a comprehensive correlation between conversion of n-butene and surface acidity of zinc ferrite catalysts. The correlation clearly shows that the conversion of n-butene was increased with increasing surface acidity of the catalyst. The yield for 1,3-butadiene also showed the same trend as the conversion of n-butene with respect to surface acidity. Among the catalysts tested, the zinc ferrite catalyst prepared at pH 9 with the largest surface acidity showed the best catalytic performance. As mentioned earlier, the acid property of zinc ferrite catalysts plays an important role in activating n-butene by the abstraction of a-hydrogen to form p-allyl intermediate [10,23–25]. Furthermore, it has been reported that sufficient acid sites are required to adsorb and activate n-butene (a base molecule)
H. Lee et al. / Catalysis Communications 9 (2008) 1137–1142 100
Conversion of n-butene (%)
90 80
pH=7
pH=10
pH=9 pH=8
pH=6
70 60 50 40 pH=11
30 20
pH=12
10 0 10
20
30
40
50
60
Acknowledgements
Fig. 4. A possible correlation between conversion of n-butene and total acidity of zinc ferrite catalysts.
100
This work was supported by SK Energy Corporation (POST-BK21 Program) and Korea Energy Management Corporation (2005-01-0090-3-010). References
90
Conversion of n-butene (%)
(ZnFe2O4) phase. In the oxidative dehydrogenation of nbutene, the catalytic performance of zinc ferrite catalysts was very high at low pH value (pH 6–10), but was drastically decreased with increasing pH value (pH 11–12). Among the catalysts tested, the zinc ferrite catalyst prepared at pH 9 showed the best catalytic performance. It was revealed from NH3-TPD experiments that the catalytic performance was closely related to the surface acidity of the catalysts. The catalytic performance was increased with increasing surface acidity of the catalysts. The largest surface acidity of zinc ferrite catalyst prepared at pH 9 was responsible for its highest catalytic performance in the oxidative dehydrogenation of n-butene.
70
Total acidity (μmol-NH3)
pH=9 pH=10 pH=8 pH=6 pH=7
80 70 60 50 40 pH=11
30 20
1141
pH=12
10 0 0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
2.4
Surface acidity (μmol-NH3/m2) Fig. 5. A correlation between conversion of n-butene and surface acidity of zinc ferrite catalysts.
[10]. Therefore, it is reasonable to expect that a catalyst with large surface acidity (with large capacity to activate n-butene per unit surface area) is favorable in the oxidative dehydrogenation of n-butene. It is concluded that the surface acidity of zinc ferrite catalysts is a crucial factor determining the catalytic performance in the oxidative dehydrogenation of n-butene. 4. Conclusions A series of zinc ferrite (ZnFe2O4) catalysts were prepared by a co-precipitation method with a variation of pH value (pH 3–12) for use in the oxidative dehydrogenation of n-butene to 1,3-butadiene. It was found from XRD and ICP-AES analyses that a zinc ferrite phase was not formed at pH 3–5, while a mixed phase of a-Fe2O3 and ZnFe2O4 was formed at pH 6–7. On the other hand, the catalysts prepared at pH 8–12 retained a pure zinc ferrite
[1] S.C. Oh, H.P. Lee, H.T. Kim, K.O. Yoo, Korean J. Chem. Eng. 16 (1999) 543. [2] Y.H. Kim, H.S. Yang, Korean J. Chem. Eng. 17 (2000) 357. [3] H.H. Kung, Ind. Eng. Chem. Prod. Res. Dev. 25 (1986) 171. [4] N.J. Jebarathinnam, M. Eswaramoorthy, V. Krishnasamy, Appl. Catal. A 145 (1996) 57. [5] F. Hong, B.L. Yang, L.H. Schwartz, H.H. Kung, J. Phys. Chem. 88 (1984) 2525. [6] B.J. Liaw, D.S. Cheng, B.L. Yang, J. Catal. 118 (1989) 312. [7] J.C. Jung, H. Kim, Y.S. Kim, Y.-M. Chung, T.J. Kim, S.J. Lee, S.-H. Oh, I.K. Song, Appl. Catal. A 317 (2007) 244. [8] B.L. Yang, D.S. Cheng, S.B. Lee, Appl. Catal. 70 (1991) 161. [9] R.J. Rennard, W.L. Kehl, J. Catal. 21 (1971) 282. [10] H.H. Kung, M.C. Kung, Adv. Catal. 33 (1985) 159. [11] M.A. Gibson, J.W. Hightower, J. Catal. 41 (1976) 431. [12] J.A. Toledo-Antonio, N. Nava, M. Martı´nez, X. Bokhimi, Appl. Catal. A 234 (2002) 137. [13] W.R. Cares, J.W. Hightower, J. Catal. 23 (1971) 193. [14] H.H. Kung, B. Kundalkar, M.C. Kung, W.H. Cheng, J. Phys. Chem. 84 (1980) 382. [15] M. Zhang, R. Lan, J. Liu, X. Chen, W. Zhou, J. Chem. Soc. Faraday Trans. 88 (1992) 637. [16] F.E. Massoth, D.A. Scarpiello, J. Catal. 21 (1971) 294. [17] F.-Y. Qiu, L.-T. Weng, E. Sham, P. Ruiz, B. Delmon, Appl. Catal. 51 (1989) 235. [18] J.A. Toledo-Antonio, P. Bosch, M.A. Valenzuela, A. Montoya, N. Nava, J. Mol. Catal. A 125 (1997) 53. [19] F.V. Aliev, E.N. Amirbekov, A.D. Kuliev, V.Y. Danyushevskii, V.D. Nissenbaum, V.I. Yakerson, A.M. Rubinshtein, Kinet. Katal. 22 (1981) 329. [20] V. Nicolescu, R. Manaila, M.A. Chaar, G. Filoti, Revue Roum. Chim. 31 (1986) 99. [21] M.A. Gibson, J.W. Hightower, J. Catal. 41 (1976) 420. [22] W.-Q. Xu, Y.-G. Yin, G.-Y. Li, S. Chen, Appl. Catal. A 89 (1992) 131. [23] C.F. Cullis, D.J. Hucknall, in: G.C. Bond, G. Webb (Eds.), A Specialist Periodical Report: Catalysis, vol. 5, Royal Chemical Society, London, 1982, p. 273. [24] R.K. Grasselli, J.D. Burrington, Adv. Catal. 30 (1981) 133. [25] J.C. Jung, H. Lee, H. Kim, Y.-M. Chung, T.J. Kim, S.J. Lee, S.-H. Oh, Y.S. Kim, I.K. Song, Catal. Commun. (2007), doi:10.1016/ j.catcom.2007.07.031.
1142
H. Lee et al. / Catalysis Communications 9 (2008) 1137–1142
[26] B. Jeyadevan, C.N. Chinnasamy, K. Shinoda, K. Tohji, J. Appl. Phys. 93 (2003) 8450. [27] G. Vaidyanathan, S. Sendhilnathan, R. Arulmurugan, J. Magn. Magn. Mater. 313 (2007) 293.
[28] I.H. Gul, A.Z. Abbasi, F. Amin, M. Anis-ur-Rehman, A. Maqsood, J. Magn. Magn. Mater. 311 (2007) 494. [29] R.K. Sharma, O. Suwalka, N. Lakshmi, K. Venugopalan, A. Banerjee, P.A. Joy, Mater. Lett. 59 (2005) 3402.