Zeolites and Related Materials: Trends, Targets and Challenges Proceedings of 4th International FEZA Conference A. Gédéon, P. Massiani and F. Babonneau (Editors) © 2008 Elsevier B.V. All rights reserved.
Oxidative dehydrogenation of propane over FeBEA catalysts 1 1
Naveen Kumar Sathu, 1Petr Sazama, 2Valentin Valtchev, 3Bohumil Bernauer, Zdenek Sobalik*
J. Heyrovsky Institute of Physical Chemistry, v.v.i, Academy of Sciences of the Czech Republic,-182 23 Prague 8, Czech Republic; 2Laboratoire de Matériaux à Porosité Contrôlée CNRS, Université de Haute Alsace, 68093 Mulhouse Cedex, France; 3 Institute of Chemical Technology, Prague, Technická 5, 166 28 Praha 6, C.Z. E- mail: [email protected]
Abstract Catalytic Oxidative DeHydrogenation of Propane (ODHP) by N2O over Fe-BEA catalysts has been studied in continuous feed regime against time and with periodic changing of in-let feed composition under reaction-regeneration cycles. The influence of process parameters on the propylene production has been evaluated, namely N2O/propane ratio and crystal size of the parent zeolite. The particle size (micro- vs. nano-particles) of the catalyst has been shown to be a crucial parameter for propylene yields and catalyst selectivity at higher temperatures. The nano-sized catalyst exhibits superior propylene formation in comparison with micro-sized one. Time dependent performance of ODHP was carried out using reactant streams containing N2O or N2O with O2. Reactivation of the catalysts by oxidation pulses revealed that short pulses of oxygen are sufficient to regenerate the active sites and stabilize propylene yield. Keywords: Fe/BEA, Particle size, N2O concentration, Regeneration
1. Introduction Catalytic oxidative dehydrogenation of propane by N2O (ODHP) over Fe-zeolite catalysts represents a potential process for simultaneous functionalization of propane and utilization of N2O waste as an environmentally harmful gas. The assumed structure of highly active Fe-species is presented by iron ions balanced by negative framework charge, mostly populated at low Fe loadings. These isolated Fe sites are able to stabilize the atomic oxygen and prevent its recombination to a molecular form, and facilitate its transfer to a paraffin molecule . A major drawback of iron zeolites in ODHP with N2O is their deactivation by accumulated coke, leading to a rapid decrease of the propylene yield. In recent years, there has been a growing interest in the synthesis and application of nano-scale zeolites. Zeolites with a crystal size smaller than 100 nm are the potential replacement for existing zeolite catalysts and can be used in novel environmentally benign catalytic processes. It is well known that the crystal size of zeolites has a great effect on their catalytic properties. The improved catalytic activity and selectivity as well as lower coke formation and better durability can be obtained over nano-sized zeolite crystals .
N.K. Sathu et al.
In the present study the effect of zeolite particle size (micro- vs. nano-particles) and N2O concentrations on ODHP is studied over Fe modified BEA zeolites. The feasibility of periodic reactivation over Fe-BEA catalysts by oxygen pulses for continuous C3H6 production has been also evaluated.
2. Experimental 2.1. Catalyst preparation Fe-zeolites were prepared using the NH4 form of BEA Si/Al = 13.5. Parent BEA zeolite (average particle size of 300 nm or 1m) was dried at 150 °C for 4 h and then mixed with a solution of FeCl3 in acetyl acetone. After 12 h of mixing, excess of the solution was removed, the solid was dried at room temperature and heated under vacuum at 350 °C for 4 h. A sample was washed with distilled water and dried in an air at room temperature. Then, the remaining organic species in the Fe-zeolites was removed by calcination at 450 °C in air for 10 h. The produced catalysts contain 0.6 wt% of Fe. This preparation procedure predominantly provides iron introduction into cationic sites . Two types of catalysts were prepared, Fe-BEA with a particle size of 1 m (Fe/m-BEA) and Fe-BEA with particle size of 300 nm (Fe/n-BEA). 2.2. Catalytic activity tests Catalytic experiments were carried out using plug flow fixed bed micro reactor. The micro reactor was loaded with 50 mg of catalyst (sieve 0.6-0.3 mm) placed between two plugs of quartz wool. Prior to catalytic experiment, a catalyst was pretreated in a flow of He and oxygen stream at 550 °C for 1 h and then cooled to reaction temperature in a flow of pure He, after stabilization of the reaction temperature. He stream was then replaced by a reaction mixture. The ODHP was investigated using feeds containing 1.5 vol % N2O, 1.5 vol % C3H8 (regime RS-1), 3 vol % N2O, 1.5 vol % C3H8 (regime RS-2), or 1.5 vol % N2O and 1.5 vol % C3H8, 0.2 vol % O2 (regime RS-3). In all catalytic experiments the GHSV was maintained at 60,000 h1. The regeneration procedure consisted of 10 or 5 min pulses of O2 at GHSV 30,000 h-1 in each run. The evolution of ODHP with time-on-stream has been followed at 400 °C. In a typical experiment the feed reactant stream of constant composition was maintained over the catalysts for 2 h, whereas in reaction-regeneration experiments, the feed mixture was replaced after 10 min of the reaction by oxygen pulses. The product mixture was analyzed with on-line connected gas chromatograph using mol sieve and HP-PLOT Q capillary column to separate permanent gases followed by detection with TCD; saturated and unsaturated hydrocarbons were separated using PLOT Q and DB-VRX capillary columns and detected by FID.
3. Results and discussion 3.1. Performance of ODHP without regeneration Figures 1 shows the catalytic performance of the Fe-BEA catalysts in the temperature range of 250-550 C. It is clear from the figure that propylene yield depends on particle size of the parent BEA zeolite. Effect of the N2O concentration has been analyzed under reaction regimes RS-1 and RS-2. Increase in N2O concentration resulted in the same propene yields but increased the N2O conversion and decreased the selectivity toward propylene. At higher temperature has been obtained increases in the formation of the molecular oxygen which further accelerates production of the undesired carbon oxides. Thus, at lower feed concentration of N2O, i.e. at 1:1 feed ratio of reactants (RS-1), formation of carbon oxides is suppressed and the selectivity of ODHP reaction is
Oxidative dehydrogenation of propane over Fe-BEA catalysts
higher. With both reaction streams increase of temperature would provide decrease in propylene selectivity due to total oxidation of propane to carbon oxides. Nevertheless, comparison of the performance of the catalysts with different particle sizes shown clearly that at higher temperatures the Fe-BEA with nano crystal size display improved catalytic activity as well as propylene yields (see Fig. 1). Increase in the propylene yield indicated over nano-BEA could be probably attributed to the enhancement of the mass transfer of the product in the zeolite porous structure or better utilization of the zeolite channels by the reactant molecules. Thus use of nano-zeolites increases the selectivity of the desired product due to easier desorption of product molecule from the local environment of zeolite channels which prevents pore blocking and over-oxidation. 9
Yield of C3H6 (%)
Yield of C3H6 (%)
Dark symbol (RS-1) Open symbol (RS-3)
Fe/m-BEA (RS-1) Fe/n-BEA (RS-1) Fe/m-BEA (RS-2)
4 3 2 1 250
Figure 1. ODHP: propylene yields against temperature
8 6 4
10 20 30 40 50 60 70 80 90 100 110 120
-1 Time on stream (min )
Figure 2. ODHP vs. time on stream over Fe/m-BEA
Figure 2 illustrate the changes in propylene yield with time using pure N2O or N2O/O2 mixture. Use of the N2O alone (RS-1) induce a rapid decrease in the propylene yield, together with decrease of the propane and N2O conversions, probably due to formation of carbonaceous deposits. Experiment under mixed N2O/O2 stream (regime RS-3) shown that presence of a small amount of oxygen does not prevent deactivation of the catalyst but indicate some stabilization of the propylene yields after longer time periods. 3.2. ODHP performance under cyclic reaction-regeneration Reaction-regeneration studies were carried out with the aim to stabilize the propylene yield for longer periods and prevent the catalyst deactivation. Regeneration of the catalyst was analyzed using a repeated reaction-regeneration periods, with O2 pulse with different time intervals after production periods. Figure 3, gives the propylene yields with periodic elimination of carbon species during 5 or 10 min O2 pulses. Results obtained under such dynamic conditions, combining 10 minutes production and regeneration periods, show effective stabilization of the propylene yields under such process conditions. Using shorter regeneration periods of 5 min would provide a gradual decrease in propylene yield. The complete reaction profile of cyclic reactionregeneration with 10 min O2 pulse (see Figure 4), evidence stabilization of the catalytic active sites and suppression of the coke formation at 400 °C under ODHP reaction. Similar catalytic activity profiles under cyclic reaction-regeneration reveals that deactivation of the Fe-BEA catalysts is fully reversible under such conditions . Thus
N.K. Sathu et al.
Fe-BEA channels are completely recovered and accessible for the reactant molecules after the relatively short oxidation pulse. 60
Yield of C3H6 (%)
5 min O2 pulse
10 min O2 pulse
40 30 20 10
12.5 12.0 10 20 30 40 50 60 70 80 90 100 110 120
Time on stream (min-1)
Figure 3. Propylene yields on reactionregeneration with time depended O2 pulse
C3H6 selectivity C3H6 conversion
CO2 selectivity N2O conversion
10 20 30 40 50 60 70 80 90 100 110 120
Time on stream (min-1)
Figure 4. Complete ODHP on reactionregeneration cycles with 10 min O2 pulse.
4. Conclusion Catalytic oxidative dehydrogenation of propane by N2O over Fe-BEA catalysts of a different crystal dimension and its stability with time-on-stream has been analyzed. The performance of ODHP with good propylene yields were produced at equilibrium concentration of N2O/propane ratio of 1:1. It could be concluded that the equimolar ratio of reactants minimizes the formation of carbon oxides and optimizes the propylene selectivity. The nano-sized catalyst exhibits better C3H6 yields with in comparison to the micro-sized catalysts at higher temperatures. Addition of a small amount of oxygen stream to the reaction mixture increases N2O and propane conversions but decreases the propylene selectivity and increase CO2 formation. On the other hand, reactivation of the catalysts by oxidation pulses proved full reversibility of the catalyst showing 10 min pulse are sufficient to regenerate the active sites. Thus a stable propylene yield has been achieved under cyclic reaction-regeneration process. Acknowledgement The authors acknowledge the support of the IDECAT, and N.K.S., B.B., Z.S. also of the GACR 106/06/1254.
References  S. Kameoka, T. Nobukawa, S. Tanaka, S. Ito, K. Tomishige and K. Kunimori,
Phys. Chem. Chem. Phys. 5 (2003) 3328  S.C. Larsen. J. Phys. Chem. C. 111 (2007) 18464  Z. Sobalik, Z. Tvaruzkova, A. Vondrova, and M. Schwarze, Stud. Surf. Sci. Catal.
162 (2006) 889  J. Perez-Ramirez, A. Gallardo-Llamas, C. Daniel and C. Mirodatos, Chem. Eng. Sci. 59 (2004) 5535