Oxidative dehydrogenation of ethane over BaF2LaOF catalysts

Oxidative dehydrogenation of ethane over BaF2LaOF catalysts

~ ELSEVIER APPLIED CATALYSS I AG : ENERAL Applied Catalysis A: General 133 (1995) 263-268 Oxidative dehydrogenation of ethane over BaF2-LaOF cataly...

318KB Sizes 0 Downloads 75 Views

~ ELSEVIER

APPLIED CATALYSS I AG : ENERAL

Applied Catalysis A: General 133 (1995) 263-268

Oxidative dehydrogenation of ethane over BaF2-LaOF catalysts X.P. Zhou, Z.S. Chao, J.Z. Luo, H.L. Wan *, K.R. Tsai Department of Chemistry, Xiamen University,Xiamen 361005, People's Republic of China Received 25 November 1994; revised 29 June 1995; accepted 19 July 1995

Abstract The oxidative dehydrogenation of ethane was investigated on LaOF and BaF2-LaOF catalysts. It was found that BaF2-LaOF was more effective than LaOF for the catalytic conversion of ethane to ethylene. A selectivity of 74% for ethylene was obtained at 55% ethane conversion over 8mol-%BaF 2LaOF compared with a selectivity of 58.5% for ethylene at 44.6% ethane conversion over LaOF under the same conditions: reaction temperature 660°C, C 2 H 6 : O 2 = 67.7:32.3 and a feed gas flow-rate of 90 ml/min. It was also found that part of the ethylene in the products was produced by the thermal cracking of ethane. X-ray diffraction results showed that, when the molar percentage of BaF2 in BaF2LaOF was less than 18%, only tetragonal LaOF was observed, and when the molar percent of BaF2 in BaF2-LaOF increased from 18% to 50%, tetragonal LaOF and lattice-contracted BaF 2 were observed.

Keywords: BaF2-LaOF catalyst; Oxidative dehydrogenation of ethane; Ethane

I. Introduction

The catalytic oxidative dehydrogenation of alkanes to alkene is important both in industry and for fundamental study. In the last decades, highly selective oxidative processes for the production of butadiene [ 1 ], isoprene [ 1,2], and acrolein [3] from mono-alkenes have been developed. But the oxidative dehydrogenation of alkanes to the corresponding alkenes, including the catalytic oxidative dehydrogenation of ethane (ODE) to ethylene has not been developed successfully. Since Thorsteinson et al. [4] reported the oxidative dehydrogenation of ethane over MoV-O catalysts, only a few papers have been published in this area. Catalysts developed were generally Mo-O, V-O, Mo-V-O or other metal oxide promoted * Corresponding author. Tel. ( + 86-592) 2086405, fax. (+86-592) 2086116. 0926-860X/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDIO926-860X(95)O01 85 -9

264

X.P. Zhou et al. /Applied Catalysis A: General 133 (1995) 263-268

M o - V - O catalysts [4-7] and alkali metal oxide promoted MgO catalyst [8]. A relatively more efficient lithium-promoted magnesium oxide catalyst and Li ÷ MgO-C1- catalyst have been reported by Lunsford and co-workers [8,9]. However, due to the loss of Li ÷ or CI-, the catalytic activity decreased faster. In the present paper, an account of the oxidative dehydrogenation of ethane over BaF2 promoted LaOF is given.

2. Experimental LaOF was prepared by grinding an equal molar ratio of LaF3 and La203 in a mortar. The mixture was then pressed into pellets under a pressure of 300 kg/cm 2, and calcined at 900°C for 4 h. The X-ray diffraction (XRD) measurement proved that the resulting material was tetragonal LaOF. BaFz/LaOF catalysts were prepared by using appropriate amounts of BaF2 and LaOF according to the same procedure as above for LaOF preparation. After calcination, the pellets of LaOF and BaF2-LaOF were crushed and sieved to a grain size of 40-80 mesh before use. The reactions were performed in a fixed-bed quartz reactor (I.D. 0.8 cm) under atmospheric pressure. 0.5 ml of catalysts were loaded in the middle part of the reactor. The rest of the reactor was filled with quartz sand of grain size 20 to 40 mesh. A 102G-D gas chromatograph equipped with a thermal conductivity detector was employed to analyze the gaseous effluent. A 5 molecular sieve column was used to analyze 02 and CO, and a Porapok Q column to analyze CH4, CO2, C2H4 and C2H6. The specific surface area of the catalysts was measured on a Sorptomatic-1900 using N2 as the adsorbate by the BET method at liquid nitrogen temperature. Before the measurement, the samples were treated at 300°C for at least 3 h at a pressure P < 8 Pa. The phase analysis of the catalysts was carried out on a Rigaku Rotaflex D/maxC XRD system using Cu Ka (A = 1.5406 ~,) radiation.

3. Results and discussion In the catalytic performance evaluation, no dilute gas was used in any of the reactions. All the data were collected after 6 h on stream. Table 1 shows the promotion of BaF2 to LaOF, and the effect of BaF2 content on catalytic activity and selectivity of the catalysts. When the reactor was filled with quartz sand alone, ethane conversion was below 5.5% at 720°C. This result indicates that the gasphase reaction of ethane with molecular oxygen may be inhibited by using quartz sand and that, below 700°C, the noncatalytic gas-phase reaction between ethane and molecular oxygen was basically negligible. It can be seen that LaOF is an

X.P. Zhou et al. /Applied Catalysis A: General 133 (1995) 263-268

265

Table 1 ODE performance of BaF2-LaOF with different BaF2 content a Catalyst

Quartz sand LaOF 6%BaF2-LaOF 8%BaF2-LaOF 10%BaF2-LaOF 12%BaF2-LaOF 14%BaF2-LaOF 18%BaF2-LaOF 22%BaF2-LaOF 26%BaF2-LaOF 30%BaF2-LaOF 50%BaF2-LaOF

T(°C)

700 720 660 660 680 660 680 660 680 660 680 680 680 680 660 680 660 680

Xo:b (%)

31.69 31.21 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Selectivity (%) CO

CH 4 CO 2

C2H4

Conversion of ethane (%)

0 _ 0 13.29 10.28 9.67 2.88 8.24 8.28 10.18 8.08 8.06 9.84 7.16 10.12 9.29 10.33 7.94 8.23

0 0 4.01 3.57 4.02 3.79 5.86 4.18 5.10 3.12 3.61 3.59 3.66 2.46 3.47 4.88 3.53 2.98

74.18 79.30 58.52 68.99 69.89 74.04 68.30 70.65 68.94 68.48 70.63 66.89 71.63 67.36 66.43 65.75 64.01 66.72

2.95 5.22 44.63 54.70 57.29 55.20 57.31 57.79 63.64 50.16 56.16 58.42 57.63 51.17 52.37 55.79 46.18 51.47

25.81 20.64 24.19 17.16 16.41 19.29 17.58 17.63 15.78 20.33 17.69 19.68 17.56 20.46 20.81 19.04 24.52 22.07

Yield of ethylene (%) 2.19 4.14 26.11 37.74 40.04 40.87 39.14 40.83 43.87 34.35 39.67 35.73 41.28 34.47 34.79 36.68 29.56 34.34

a Reaction conditions: C2H6:O2 = 67.7:32.2: flow rate of feed gas = 90 ml/min. Xo2: the molar percentage concentration of 02 in the effluent.

active ODE catalyst at 660°C, but its selectivity for ethylene formation is limited to 58.52%. When BaF2 was added to LaOF, both ethane conversion and ethylene selectivity increased significantly. BaF2 contents from 6 mol-% to 18 mol-%, give the highest ethane conversions and ethylene selectivities. A life-span test with a 14mol-%BaF2-LaOF catalyst showed no decrease in catalytic activity and ethylene selectivity during 26 h on stream. The catalytic performance evaluation also indicates that the reaction over LaOF or BaF2-LaOF is an oxygen-limited reaction. Table 2 shows the ODE results for 6mol-%BaF2-LaOF at different reaction temperatures. As can be seen, when the reaction temperature was increased from 580°C to 640°C, ethane conversion increased from 34.85% to 50.65%, and the selectivities of CO, CH 4 and C2H4 changed slightly, while CO2 selectivity decreased from 21.13% to 18.46%; Meanwhile the percentage content of 02 in the effluent decreased. When the reaction temperature reached 660°C, all the oxygen was consumed. With further increase of temperature from 660°C to 700°C, the conversion of ethane increased from 54.70% to 62.64%, suggesting that 7.94% of the ethane conversion may have resulted from thermal cracking of ethane. No measurements were made to determine if coke was formed under these conditions. From the data in Table 3, it can be seen that within the gas flow-rate range used, higher ethane conversions can be obtained at higher gas hourly space velocity (GHSV). However there is no simple correlation between the ethylene selectivity and the GHSV, which may indicate that a hot spot existed in the reactor, especially

X.P. Zhou et al. /Applied Catalysis A: General 133 (1995) 263-268

266

Table 2 Effect of reaction temperature on ODE performance a

T (°C)

580 600 620 640 660 680 700

Xo2h ( % )

14.20 9.97 4.43 2.39 0 0 0

Selectivity (%) CO

CH~

CO 2

C2H 4

8.08 7.22 8.19 8.62 10.28 9.67 13.93

2.77 2.96 3.11 3.38 3.57 4.02 4.70

21.13 20.12 19.05 18.46 17.16 16.41 14.05

68.03 69.69 69.65 69.54 68.99 69.89 67.33

Conversion of ethane

Yield ofethylene

(%)

(%)

34.85 40.41 45.50 50.65 54.70 57.29 62.64

23.71 28.16 31.69 35.22 37.34 40.04 42.17

a Reaction conditions: C 2 H 6 : O 2 = 67.4:32.6: 6%BaF2-LaOF catalyst; the flow-rate of feed gas = 90 ml/min. b See Table 1.

at the higher GHSV. At the hot spot of the catalyst layer, the temperature may be 100°C to 150°C higher than elsewhere in the catalyst bed. At such high temperature, when oxygen was used up, some of the residual ethane might be easily cracked to methane, ethylene and hydrogen in a catalytic or non-catalytic way. From Table 1 and Table 4, it can be seen that when BaF2 was added to LaOF, both the catalytic activity and the specific surface area of catalysts increased. Thus the enhancement of catalytic activity might be at least partially related to surface area increases. As shown in Table 1, the selectivity to ethylene over BaF2-LaOF catalysts was also higher than that over LaOF, while the specific surface area of the BaF2-LaOF catalyst was larger than that of LaOF as mentioned above. Thus the increase in ethylene selectivity over BaF2-LaOF catalysts can not be attributed to Table 3 Catalytic performance at different gas hourly space velocity (GHSV) over some of BaF2-LaOF a Catalyst

14%BaF2-LaOF 18%BaF2-LaOF 22%BaFz-LaOF 26%BaF2-LaOF 30%BaF2-LaOF 50%BaF2-LaOF

R~ ( ml / min )

90 200 90 200 90 200 90 200 200 387 90 200

Xo, c (%)

0 0 0 0 0 0 0 0 0 0 0 0

Selectivity (%)

Conversion of

Yield of

ethane

ethylene

CO

CH 4

CO 2

c2n 4

(%)

(%)

10.24 7.82 7.44 11.35 12.92 10.09 9.25 11.91 9.46 8.68 9.09 8.28

3.11 4.03 2.76 7.88 2.54 6.54 2.62 6.64 4.52 8.77 2.38 5.71

22.87 15.23 19.94 12.36 19.20 12.73 21.34 14.76 14.15 11.73 23.84 13.03

63.78 72.92 69.86 68.41 65.33 70.65 66.79 66.68 71.87 70.83 64.69 72.98

47.29 59.20 51.39 74.54 48.48 69.82 47.96 67.21 65.46 80.82 46.35 65.61

30.16 43.17 35.90 50.99 31.67 49.33 32.03 44.82 47.04 57.24 29.98 47.88

a Reaction conditions: reaction temperature = 640°C, C2H6:O 2 = 67.4:32.6. Rx: flow-rate of feed gas, ml/min. c See Table I footnote b.

X.P. Zhou et al. / Applied Catalysis A: General 133 (1995) 263-268

267

Table 4 BET specific surface area of catalyst Catalyst

Specific surface area

Catalyst

Specific surface area

(m2/g) LaOF 6%BaF2/LaOF 8%BaF2/LaOF 10%BaFz/LaOF 12%BaF2/ LaOF 14%BaF2-LaOF

(m2/g)

2.94 9.13 5.20 6.53 4.71 5.05

18%BaF2/LaOF 22%BaF2/LaOF 26%BaF2/LaOF 50%BaFJLaOF BaF2

6.30 6.12 4.36 3.61 2.67

a reduction in surface area. The higher activity for ethane conversion and selectivity for ethylene on BaF3-LaOF than on LaOF may be caused by the dispersion and/ or the formation of the BaF2 phase as well as the relative abundances of O~- and O - in BaFa-LaOF systems compared to LaOF systems, for which preliminary in situ Raman and EPR spectra have been obtained [ 10]. The XRD results (see Table 5) reveal only the tetragonal LaOF phase, for BaF2 contents from 6 mol-% to 14 mol-%. With BaF2 contents from 18% to 50%, in addition to tetragonal LaOF, there was also lattice contracted cubic BaF2. These results suggest that when the BaF2 content is lower than 18 mol-%, all of the BaFz may be dispersed in LaOF, with the formation of a BaFa-containing LaOF phase (BCLP). In the BCLP, some of La 3+ lattice points might be replaced by Ba 2+ ions, leading to the formation of anion vacancies or O - species as described by Zhou et al. [ 11 ]. On the other hand, when the content of BaF2 is above 18%, the LaOF lattice can not accommodate all the BaF2, and thus a separate BaF2 phase Table 5 XRD results of fresh and used catalysts

Catalyst

LaOF 6%BaFJLaOF 8%BaF2/LaOF 10%BaF2/LaOF 12%BaF2/LaOF 14%BaFJLaOF 18%BaF2/LaOF 22%BaF2/LaOF 26%BaF2/LaOF 30%BaFJLaOF 50%BaF2/LaOF

Phase Fresh catalysts

Used catalysts

t (h) ~

(T) a LaOF (T) LaOF (T) LaOF (T) LaOF (T) LaOF (T) LaOF (T) LaOF (T) LaOF; (T) LaOF; (T) LaOF; (T) LaOF;

(T) (T) (T) (T) (T) (T)

10 10 12 10 26

(C) (C) (C) (C) (C)

BaF2(I/lo < 21 )b BaF2(I/lo<21) BaF2(l/lo<42) BaF2(l/lo<45) BaF2(l/lo<45)

LaOF LaOF LaOF LaOF LaOF LaOF

(T)LaOF; (T)LaOF; (T)LaOF;

(C) BaF2 (1/1o< 17) (C) BaF2 (1/Io< 30) (C) BaF2 (1/lo<29)

13 10 11 12

a T: tetragonal; C: lattice contracted cubic; t: time on stream. b I/I0: the relative diffraction intensity of the ( 111 ) crystal face of cubic BaF2 in BaF2-LaOF which has a value of 100% in JCPDS. c Time catalysts are used for ODE on stream.

268

X.P. Zhou et al./ Applied Catalysis A: General 133 (1995) 263-268

forms. In this kind of BaF2 lattice, some of Ba 2+ may be substituted by La 3+ ions, with a lattice contraction of the BaF2 phase, in consideration of the more positive charges and a smaller ionic size of La 3÷ than Ba 2+. The XRD results also show that after the catalysts were used for 10 to 26 h, their structure changed only slightly, except when the BaF2 molar percentage concentration was higher than 18%, the peak intensity of the BaF2 phase became weaker after reaction. These results also indicate the stability of the BaF2-LaOF catalyst for these periods on stream.

4. Conclusion The addition of BaF2 to LaOF substantially improves the catalytic properties of LaOF for ODE, and leads to the formation of a BaF2- containing LaOF phase and a lattice-contracted cubic BaF2 phase. The enhancement of activity of the BaF2LaOF systems for ODE may be due to increases in the specific surface area and the number of anionic vacancies, while the selectivity improvement may be related to the dispersion of BaF2 in BaF2-LaOF systems, the isolation of surface active center and, possibly, more adsorbed 022 and O - species in the BaF2-LaOF systems.

Acknowledgements This work was supported by the National Natural Science Foundation and the State Education Commission of China. The catalysts used in this paper have been patented in China.

References [ 1] H.H. Voge, W.E. Armstrong and L.B. Ryland, US Patent 3 110 746 ( 12 November 1963). [2] J.L. Callahan, B. Gertisser and R. Grasselli, US Patent 3 260 768 ( 12 July 1966). [3] F. Veath, J.L. Callahan, E.C. Milberger and R.W. Foreman, in Actes du Deuxieme Congress International de Catalyse, Paris, 1960, Vol. 2, 1961, p. 2647. [4] E.M. Thorsteinson, T.P. Wilson, F.G. Young and P.H. Kasai, J. Catal., 52 (1978) 116. [5] T.J. Yang and J.H. Lunsford, J. Catal., 63 (1980) 505. [6] A. Erodhelyi and F. Solymosi, J. Catal., 123 (1990) 31. [7] R. Burch and R. Swarnakar, Appl. Catal., 70 (1990) 129. [8] E. Morales and J.H. Lunsford, J. Catal., 118 (1989) 255. [ 9 ] $3. Conway and J.H. Lunsford, J. Catal., 131 ( 1991 ) 513. [ 10] Z-S. Chao, Ph.D. Thesis, Xiamen University of China, 1995, Ch. 5~6. [ 11 ] X.P. Zhou, Z.S. Chao, W.Z. Weng, W.D. Zhang, S.J. Wang, H.L. Wan and K.R. Tsai, Catal. Lett., 29 (1994) 177.