Carbon - Supported platinum catalyst for n-hexane dehydrogenation

110 Journal of Chemistry, Vol. 41, No. 1, P. 110 - 114, 2003 CARBON - SUPPORTED PLATINUM CATALYST FOR N-HEXANE DEHYDROGENATION Received 01-7-2002 Nguyen Thi Dung, Tran Manh Tri Institute of Chemical Technology, Hochiminh City Summary Activated carbon (AC) from coconut shells is used as support for platinum catalysts. A series of catalysts based on platinum and Pt doped Sn were prepared by impregnation on different supports like AC and SiO2. It was shown that the nature of the support has a significant influence on the activity and selectivity to alkenes of the catalysts in the dehydrogenation of n- hexane to alkenes. The higher activity and selectivity of the carbon supported platinum catalyst can be explained by higher dispersion of platinum particles of the catalysts. The addition of Sn led higher stability and activity and that may be related to the formation of a PtSn alloy on the surface and the modification of interaction between the alkenes and the metal sites, making it weaker and keeping the metallic surface more free of cokes by migration of that to the carbon surface. I - Introduction Platinum supported catalysts are widely used in many industrial processes including hydrogenation, dehydrogenation, and reforming of n-alkanes. Its chemical stability in both oxidizing and reducing conditions makes this metal an ideal catalyst in many applications. Catalytic dehydrogenation of alkanes is of increasing importance due to the growing demand for linear alkenes as raw materials for the production of biodegradable detergents. Bimetallic Pt-Sn/Al2O3 catalysts have been extensively studied by many research groups [1 - 4] due to their importance in industrial petrochemical processes. Many of these studies have focused on hydrocarbon conversion such as isomerization, aromatization, dehyrocyc- lization and dehydrogenation of light alkanes. Timofeeva et al. [6] studied Pt/Al2O3 and Li- doped Pt/Al2O3 catalysts and showed that the selectivity to monoalkenes is dependent on the presence of Li but is not dependent on the concentration. Castrol [7] found that bimetallic Pt-Sn catalyst doped by alkali metals has a better yield in alkenes and lower selectivity towards gases and aromatics compared to monometallic platinum. Bimetallic Pt-Ge and Pt-Pb catalysts show a similar performance compared to Pt-Sn. However, their selectivity to alkenes is lower. Yang et al. [8] indicated that the oxidation states of tin and the possibility of alloy formation between platinum and tin seems to be depended on the method of preparation, the nature of supports and the loading of the catalytic component. Recent studies have showed that, the nature of the support is one of the important characteristics of great influence on activity and selectivity of the catalyst . Activated carbon as a catalyst support is of increasing interest today due to its unique properties like stability in both acidic and basic media and thermal resistance but only a few 111 publications are devoted to the investigation of the selective dehydrogenation of n-alkanes to alkenes over carbon supported platinum catalysts. The aim of the following study is the preparation and characterization of platinum catalysts supported on activated carbon obtained from coconut shells of Vietnam in comparison with supported Pt/SiO2 catalysts in the selective dehydrogenation of n-hexane to alkenes as a model reaction II - Experiment 1. Preparation and characterization of the AC support The active carbon was prepared by carbonization and activation of coconut shells [10] according to the following procedure. The coconut shells were impregnated with a solution of phosphoric acid dried at 115oC for 2 h and then carbonized by heating up to 850oC for 3 h in a furnace under flowing nitrogen. The carbon obtained was activated under CO2 at 850oC for 4 h, then washed with deionized water and dried at 115oC overnight. The surface area and the porosity of the support were determined from adsorption isotherms of nitrogen which were measured on a Micrometric ASAP-200-V20. SiO2 (Degussa Aerosil-type silica) has been selected as a support for comparison purposes. The results are summarized in table 1. Table 1: The textural properties of the activated carbon (AC) and the SiO2 Supports Surface area BET, m2/g Pore volume, cm3/g AC SiO2 1026 230 0.29 0.90 2. Preparation of catalysts The platinum-supported catalysts (0.5wt% Pt) were prepared by impregnation of the supports with a solution of H2PtCl6 (0.02 mol/l). The support was contacted with the impregnating solution under stirring at room temperature for 2 h. Following impregnation, the catalysts were filtered, dried overnight at 110oC and then calcined at 450oC in flowing nitrogen for 2 h. Subsequently Sn was added by impregnation from a solution of SnCl2 (0,04 mol/l) following the same procedure. The composition of the catalysts used in the catalytic tests are given in table 2 (For selected catalysts the metal dispersions were determined by transmission electron microscopy) carried out with a Philips EM-420 Instrument operated at 120 kV. Table 2: Composition of the catalysts used Support Pt weight, % Sn weight, % Pt/Sn (atomic ratio) AC SiO2 0.52 0.52 0.55 0.55 0 0.71 0 0.71 1 : 0 1 : 0.43 1 : 0 1 : 0.43 3. Catalytic test Dehydrogenation of n-hexane was investigated in a micro-flow reactor at atmospheric pressure. A feed of hydrogen saturated with n-hexane vapor was generated by bubbling hydrogen (30 ml/min) through a thermostated saturator at 15oC. All catalysts used were reduced in-situ prior to the reaction at 500oC in flowing hydrogen (20 ml/min) for 4 h. After that the reactor was brought to reaction temperature (500oC) and then contacted with the feed consisting of hydrogen and n-hexane with molar ratio of 6.6 and a n-hexane partial pressure of 100 Pa. Product analysis was performed on line by gas-chromatography (Fisons Instrument 8130-00) equipped with a FID detector and a capillary column DB-5 from W&J (30 m, 0.32 mm). Catalytic measurements were carried out at 500oC. The specific rate (R) was calculated as mole of n-hexane transformed per second per gram of Pt: mole/sec.g (Pt). 112 Conversion (X) was defined as mole of n- hexane converted per mole of n-hexane in the feed (%mole) and the selectivity (S) as moles of n-hexenes obtained per moles of total n-hexane converted (%mole). III - Results and discussion 1. The effect of the support Blank tests were carried out with the supports AC and SiO2 which showed no activity. The results of the catalytic activity evaluated at 500oC for the two Pt (0.5%) containing catalysts are summarized in table 3. Table 3: Catalytic activity and selectivity of supported platinum catalysts in the dehydrogenation of n-hexane at 500oC Phexane = 1.33.10 4 Pa, mcat.= 0.5 g of catalysts; H2/nC6 = 6.6 Pt(0,52wt%)/AC Pt(0,55wt%)/SiO2On stream, min Conversion, %mole Selectivity, %mole Conversion, %mole Selectivity, %mole 1 30 60 90 120 150 180 210 240 12.4 10.5 9.8 7.7 6.9 6.8 6.7 6.8 6.6 80.5 79.8 78.2 78.9 78.1 77.6 78.5 77.8 78.6 8.5 6.2 5.1 4.2 3.4 2.8 2.1 1.9 1.8 65.7 62.3 60.8 61.2 61.8 60.9 61.5 60.8 60.1 From the results, it can be seen that the change of the support led to significant changes in the dehydrogenation activity of the platinum catalyst. The carbon supported Pt catalyst shows a higher activity and selectivity compared with the SiO2 supported catalysts. This is in agreement with results reported by Guerrenro-Ruiz [13] who showed that the activity of the carbon supported catalyst was higher compared to the silica supported catalyst (conversion 6.3% and 0.9% respectively). The results for both catalysts Pt/AC and Pt/ SiO2 showed the deactivation of the catalysts. The conversion and selectivity decreased with time on stream. The carbon supported catalyst remained more stable after 2 h reaction time. In our case the selectivity to alkenes of Pt/SiO2 was higher than that indicated in [11] 5% and 35%, respectively). The difference may be caused by different reaction conditions, methods of preparation and comparison and composition of the catalyst. The deactivation of the silica-supported platinum catalyst can be explained by the accumulation of carbonaceous species that leads to the loss of platinum surface area. For the carbon supported catalyst the better stability according to Llorca [11] could be caused by the easy adsorption of alkenes from the metallic sites and by a less favorable polymerization type reaction under the same reaction conditions. On the other hand, the observed results may also be related to the dispersion of metal particles. From the TEM results shown that the Pt/C catalyst has a very homogeneous particle size distribution. No Pt particles could be seen above the resolution limit of the microscope i.e. They are below 1 nm (Fig. 2). This is in agreement with results published by Meriaudeau et al. [5] who found very small particles (about 1 nm) of Pt supported on NaY. In the case of the Pt/SiO2 sample, large particle sizes between 5 nm and 30 nm were observed (Fig. 3). According to Cortright et al. [12] the dissociative adsorption of n-hexane is the rate limiting step that controls the dehydrogenation 113 0 5 10 15 20 25 30 35 40 0 30 60 90 120 150 180 210 240 time(min) C on ve rs io n (% m ol ) Pt/AC Pt-Sn/AC Pt-Sn/SiO2 Pt/SiO2 Fig. 1: Conversion of n-hexane to hexene on different catalysts Fig. 2: TEM of Pt/C catalyst Fig. 3: TEM of Pt/SiO2 catalyst reaction. The adsorption and dissociation of n- hexane should take place favorably on the small particles of Pt. This phenomenon can be explained by the electron deficiency of small Pt particles that leads to strong electronic affinity to the electron of the hydrocarbon. On large particles of Pt the adsorption and dissociation n-hexane is more difficultly because of the weak electron affinity [14]. 2 The effect of the Sn To examine the effect of tin on the catalytic properties, a series of experiments were carried out and the data showed that the activity and the selectivity to alkenes increased when tin was added to both Pt catalysts (Fig. 1). The activity increased three times compared to the monometallic catalysts for both catalyst and the stability was improved. In the case of the supported carbon catalyst the activity slightly decreased from 33.5% to 29.5% while for the silica supported catalyst a considerable loss in activity from 25.5% to 10.2% after 4 h was observed. The selectivity increased for both catalysts Pt/AC and PtSiO2 from 80.5% to 97.2% and from 65.7 to 80.5%, respectively. The selectivity remained nearly stable for 4 h time on stream. Both bimetallic Pt-Sn catalysts demons- trated more stability than monometallic catalyst, but for carbon supported that characteristic was better. After 4 h, activity of Pt-Sn/C decreased 0.8 times while for Pt- Sn/SiO2 that diminual 2.5 times. 114 It was shown repeatedly that the addition of Sn led to an increase of the catalytic activity and the selectivity of the catalyst. Some studies indicated that alloy is formed on silica with different modifications depending on the content of tin. According to results reported by Yang et al. [8] a PtSn alloy was obtained on the support Al2O3 high interaction between Sn and Al2O3 was performed, while on the carbon support the particle of Pt and Sn were separate with small size, and a weak interaction between Sn and C was observed. The higher activity and selectivity to alkenes and the better stability of the bimetallic compared to the monometallic catalysts in our investigation may be related to a decreasing of coke formation by modification of the interaction between metallic site and alkenes. The obtained data also lend support to the idea that the coke formed in the reaction may migrate more easily from the metallic site to the carbon support which as the same nature. In the other hand, for carbon supported Pt-Sn catalysts the higher activity and selectivity compared with silica supported catalyst may be due to its higher dispersion of platinum and the decrease of coke formation on metallic sites. IV - Conclusions 1. The active carbon from coconut shells of Vietnam was shown to be a good support for metallic catalysts in selective dehydrogenation of n-hexane to alkenes. 2. The carbon supported catalyst demonstrated higher dispersion, compared to the silica supported catalyst. No particle of Pt above 2 nm was observed for Pt/AC while for the Pt/SiO2 it was 5 nm - 30 nm. This may be the reason for the higher activity and selectivity of the carbon supported catalyst. 3. The effect of Sn addition to supported platinum catalyst should be related to the modification of the strength of interaction between metal sites and alkenes that lead to a decrease of coke. The higher activity and selectivity of the carbon supported catalysts suggests the idea that the obtained coke could migrate more easily to the carbon support due to their similar nature. Acknowledgement: We are grateful to the VOLSKWAGEN- STIFTUNG (.AZ I /75855) for financial support. We thank Prof. Dr. Nils Jaeger, IAPC of Bremen University for helpful discussion and TEM measurement. References 1. R. Srinivasan, B. H Davis. Platinum Met. Rev. Vol. 36, P. 151 (1992). 2. Lin Liwu, Zhangtao, Zang Jinling. Appl. Catal. A 67, P. 11 (1990). 3. K. Balakrishnan, J. Shwank. J. Catal., Vol. 127, P. 827 (1991). 4. B. Bursh. J. Catal., Vol. 71, P. 348 (1981). 5. P. Meriaudeau, C. Naccache, A. Thangara, C. L. Bianchi, R. Carli,V. Vishvanathan, S. Narayanan. J. Catal., Vol. 154, P. 345 - 354 (1995). 6. E. A Timofeeva, A. P. Tuypaev and G .B. Isagulyants, Izv. Akad. Nauk. SSSR Ser. Khim., Vol. 9, P. 2067 (1981). 7. A. A. Castro. Catal. Lett., 22, P. 123 (1993). 8. Yang Weishen, Lin Liwu, Fan Yining, Zang Jingling, Catal. Lett., 12, P. 267 (1992). 9. F. Rodringuez-Reinoso, C. Moreno Castilla, A. Guerrero-Ruiz, I. Rodriguez- Ramos, J. D. Lopez Gonzalez. Appl. Catal. A 15, P. 293 (1985). 10. Tran Manh Tri, Bui Thanh Huong, Nguyen Kim Dung. The Communication of the Institute of Chemistry. HCM City, P. 18 (1992). 11. Jordi Llorca, Narcis Homs, Jose-Luis G.Fierro, Joaquim Sales, Pilar Ramirez de la Piscina. J. Catal., Vol. 166, P. 44 (1997). 12. R. D. Cortright, J. A Dumesic. J. Catal., Vol. 157, P. 576 (1995). 13. A. Guerrero-Ruiz, B. Bachiller-Baeza, Rodriguez-Ramos. Appl. Catal. A 173, P. 231 (1998). 14. Tran Manh Tri, J. Massardier, P. Gallezot, B. Imelik. Preprints of the 7th International Congress on Catalysis, (Tokyo, July 1-2, 1980, Japan) A1 - A26, P. 266.