CATALYSIS, KINETICS AND REACTION ENGINEERING Chinese Journal of Chemical Engineering, 21(7) 730—735 (2013) DOI: 10.1016/S1004-9541(13)60537-6
Propane Dehydrogenation over a Commercial Pt-Sn/Al2O3 Catalyst for Isobutane Dehydrogenation: Optimization of Reaction Conditions* Farnaz Tahriri Zangeneh, Saeed Sahebdelfar** and Mohsen Bahmani
Catalyst Research Group, Petrochemical Research and Technology Company, National Petrochemical Company, 14358-84711, Tehran, Iran Abstract The applicability of a commercial Pt-Sn/Al2O3 isobutane dehydrogenation catalyst in dehydrogenation of propane was studied. Catalyst performance tests were carried out in a fixed-bed quartz reactor under different operating conditions. Generally, as the factors improving propane conversion decrease the propylene selectivity, the optimal operating condition to maximize propylene yield is expected. The optimal condition was obtained by the experimental design method. The investigated parameters were temperature, hydrogen/hydrocarbon (H2/HC) ratio and space velocity, being changed in three levels. Constrains such as the susceptibility of the catalyst components to sintering or phase transformation were also taken into account. Activity, selectivity and stability of the catalyst were considered as the measured response factors, while the space-time-yield (STY) was considered as the variable to be optimized due to its commercial interest. A STY of 16 mol·kg−1·h−1 was achieved under the optimal conditions of T = 620 °C, H2/HC = 0.6 and, weight hourly space velocity (WHSV) = 2.2 h−1. Single carbon-carbon bond rupture was found to be the main route for the formation of lower hydrocarbon byproducts. Keywords Pt-Sn/Al2O3 catalyst, dehydrogenation, propane, isobutane
Propylene is an important raw material for the production of many petrochemicals such as polypropylene, acroleine, and acrylic acid. It is also a valuable feedstock for the production of clean high-octane gasoline blending stocks, required by the reformulated gasoline. It can be converted to alkylate via alkylation of isobutane, polygasoline (a high-octane gasoline from polymerization of light olefins) via oligomerization and oxygenates such as diisopropyl ether (DIPE) by converting isopropanol or mixture of isopropanol with propylene. Alkylation is a process for combining isobutane with light olefins, typically propylene and butylenes in the presence of an acidic catalyst, usually hydrogen fluride (HF) or sulfuric acid. Alkylate from propylene can reduce D-86 T50 values (50% point distillation temperature) to conform to tighter reformulation standards. However, getting the full benefit of propylene alkylation requires separate olefin processing (each olefin is optimally alkylated separately), which can improve product properties and lower operating costs. Increased alkylate demand to meet reformulated gasoline requirements necessitates an increase in light olefins . Although propylene has been mostly produced as a by-product of steam crackers along with ethylene, an important change in propylene supply is occurring through a shift to on-purpose production methods . Among these, propane dehydrogenation (PDH) has received attention. Propane dehydrogenation is an endothermic reaction accompanied with an increase in volume and is limited by chemical equilibrium:
ZZX C3 H 6 + H 2 C3 H8 YZZ
0 ΔH 298 = 124 kJ·mol−1 (1)
Therefore, high temperature and low pressure are necessary to achieve high equilibrium conversion. Despite the simple chemistry, industrial implementation of dehydrogenation is very complicated due to side reactions such as deep dehydrogenation, cracking, polymerization and coke formation. Some technologies utilize diluents such as steam or hydrogen to reduce coke formation and increase equilibrium conversion through decreasing hydrocarbon partial pressure. In the UOP Oleflex process, hydrogen is used as a diluent which, increases catalyst lifetime to a few days, despite a small decrease of thermodynamic driving force . Commercial catalysts used for light paraffin dehydrogenation are either based on chromium oxide [4, 5] or platinum [6-12]. Both types exhibit good performance for light paraffin dehydrogenation. However, due to coke formation and rapid catalyst deactivation, continuous catalyst regeneration is necessary. Different technologies were developed to cope with this problem [13, 14]. Because of decreasing isobutene demand due to the phase-out of methyl tertiary butyl ether (MTBE) as an octane enhancer, a suitable revamp of isobutane dehydrogenation plants is to modify them for propane dehydrogenation to propylene which still has an expanding market. The optimum operating conditions, however, are different for these two reactions. Unfortunately, all factors increasing propane conversion (such as higher reactor temperature, longer contact time and lower H2/HC ratio) tend to decrease propylene selectivity and catalyst lifetime as well.
Received 2011-03-30, accepted 2012-12-30. * Supported by the Petrochemical Research & Technology Co. of National Petrochemical Co. ** To whom correspondence should be addressed. E-mail: [email protected]
Chin. J. Chem. Eng., Vol. 21, No. 7, July 2013
Therefore, optimal operating conditions should be sought for the catalyst to exhibit best performance. The purpose of this work is to obtain optimal conditions for propane dehydrogenation over a commercial isobutane dehydrogenation catalyst using the method of experimental design. The experimental results are analyzed and compared to those for isobutane dehydrogenation over the same catalyst based on data from the literature. The findings would be of interest for revamping isobutane dehydrogenation plants to PDH plants. 2 2.1
EXPERIMENTAL Characterization techniques
The commercial Pt-Sn/Al2O3 isobutane dehydrogenation catalyst sample (Pt = 0.5%, Sn = 0.7%, by mass) was supplied by Procatalyse Co. The textural properties of the catalyst were measured by means of nitrogen adsorption-desorption experiments (ASAP2010, Micromeritics). Specific total surface area was calculated using the Brunauer-Emmett-Teller (BET) equation. The mesopore size distribution was calculated from the desorption branch using the Barrett-JoynerHalenda (BJH) method. 2.2
Catalyst performance tests were carried out in a fixed-bed quartz reactor [internal diameter (ID) = 15 mm] under atmospheric pressure using a mixture of propane and hydrogen as the feed. In each run, 1.5 g of the catalyst was loaded. The reactor effluent was analyzed for C1-C3 hydrocarbons with an online gas chromatograph (Agilent 6890N), equipped with thermal conductivity detector (TCD) and flame ionization detector (FID), one hour after setting the operating condition to ensure steady operation. Test conditions were specified according to the normal conditions of commercial plants and experimental design method discussed below. Table 2
Design of experiments
Design of experiments (DOE) allows for the study of the effects several factors may have on a process . The logical procedure for executing DOE is to define the process to be optimized, to determine the response factors, to determine process input variables and their levels on which experimental design and testing are based on, and to analyze the results, and finally to make conclusions and recommendations. The objective of this work is to optimize the operating conditions for propane dehydrogenation reaction. The activity, selectivity and stability of the catalyst were selected as the measured response factors. However, the variable to be optimized is the space-timeyield (STY) which is of critical importance in commercial operation. In a PDH process, the important input variables (factors) that influence product composition are as follows: (1) Temperature, which significantly influences the conversion, selectivity and lifetime of the catalyst. (2) H2/HC molar ratio, which influences the equilibrium conversion, selectivity and stability of the catalyst. (3) Space velocity, which influences the residence time, selectivity and productivity. The factor levels were selected such that the resulting response measured was in the nominal values. According to the operating conditions typical to commercial plants for propane dehydrogenation, factor levels were selected as given in Table 1. Table 1 Factor
Values of important factors in three levels Temperature/°C
H2/HC molar ratio
The combinations of factor levels, corresponding to the conditions under which responses measured were defined accordingly. The entire set of runs or the “design” is shown in Table 2 according to one factor
Experimental conditions used in experimental design
H2/HC molar ratio
short time runs for investigation of influence of temperature
short time runs for investigation of influence of H2/HC ratio
short time runs for investigation of influence of space velocity
Chin. J. Chem. Eng., Vol. 21, No. 7, July 2013
at a time method. The rows show nine runs of design for experiments and the columns show the level values of each factor. The runs including severe operating conditions such as high temperatures or low H2/HC molar ratio were continued for long time-on-streams to check the influence on catalyst stability as well. The experiments were performed randomly. Randomization helps to minimize or eliminate the effects of extraneous factors that may be present in an experiment. 3 3.1
RESULTS AND DISCUSSION Catalyst characterization
Figure 1 shows a unimodal pore size distribution of the catalyst with a peak at nearly 10 nm. The textural properties of the catalyst from BET measurements are summarized in Table 3. The high surface area implies the predominance of γ-alumina phase in the support.
Figure 1 The dV/dD plot (volume versus pore diameter) of the catalyst Table 3
Summary of textural test results
average pore diameter/nm 3
total pore volume/m ·g
However, the selectivity to propylene remains rather unaffected with only a slight decrease within the investigated range. At higher temperature one may expect a larger decrease in propylene selectivity due to noncatalytic gas-phase reactions bringing about coke formation and catalyst deactivation. Figure 3 shows catalyst stability test results in term of selectivity to propylene and C1 and C2 hydrocarbons at 600 and 620 °C. The lower selectivity to propylene at higher temperature can be attributed to enhanced formation of lower hydrocarbons. Lower hydrocarbons could be formed both via single or multiple carbon-carbon bond rupture of propane molecule due to cracking and hydrogenolysis reactions. However, a careful look at Fig. 3 also reveals that the C2 selectivities are nearly twice as those for methane under the same conditions, implying that single C C bond scission is the predominant route for the formation of lower hydrocarbon by-products.
The average pore diameter of the catalyst (10.6 nm) is much larger than the kinetic diameter of propane (0.43 nm ), indicating that there is no steric hindrance for propane molecules to access to most active sites. The kinetic diameter of propane is even smaller than that of isobutane (0.50 nm). In fact, a recent study has shown that propane dehydrogenation reaction is neither external-diffusion nor internal-diffusion limited over the same catalyst . 3.2
Figure 2 Influence of temperature on catalyst performance (WHSV = 2 h−1, H2/HC = 1) ◆ propane conversion; ■ propylene selectivity; ▲ propylene yield
3.2.1 Effect of reaction temperature Figure 2 shows the effect of temperature on catalyst performance. Since the reaction is endothermic, a increase in temperature enhances the conversion of propane due to both kinetic and thermodynamic factors.
Figure 3 Results of catalyst stability tests (WHSV = 2 h−1, H2/HC = 1) selectivity of propylene at T = 620 °C; ◆ selectivity of propylene at T = 600 °C; ¾ selectivity of C2 at T = 620 °C; ■ selectivity of C2 at T = 600 °C; ● selectivity of C1 at T = 620 °C; ▲ selectivity of C1 at T = 600 °C
Figures 4 and 5 show that the catalyst exhibits a reasonable stability for propane dehydrogenation both at 600 °C and 620 °C. Considering the favorable influence of temperature on product yield which is
Chin. J. Chem. Eng., Vol. 21, No. 7, July 2013
Figure 4 Results of catalyst stability tests (WHSV = 2 h−1, H2/HC = 1) T = 600 °C; ■ T = 620 °C
Figure 5 Results of catalyst stability tests (WHSV = 2 h−1, H2/HC = 1) ◆ T = 600 °C; ■ T = 620 °C
much more prominent compared to its adverse influence on catalyst stability (Figs. 2 and 4, respectively), 620 °C was selected as the optimum temperature. It is noted that the catalyst is susceptible to platinum sintering and phase transformation of the γ-alumina support at higher temperature, both of which reduce catalyst lifetime. Consequently, higher temperature was not examined in this work. 3.2.2 Effect of hydrogen/propane mole ratio Figure 6 shows the influence of H2/HC molar ratio on the performance of the catalyst. While propylene selectivity remained rather constant, propane conversion and propylene yield decreased with the increase of H2/HC ratio. An increase in the partial pressure of hydrogen not only decreases the thermodynamic driving force, but also kinetically reduces the rate of dehydrogenation reaction due to the competitive adsorption of hydrogen with propane on platinum active sites . In addition, lower values of H2/HC ratio may accelerate catalyst deactivation. Therefore, depending on the paraffin under consideration and its coke formation tendency, an optimum H2/HC ratio should exist. Figure 7 shows the long-term stability of the catalyst at different levels of H2/HC ratios. No severe catalyst deactivation at lower level was observed. Therefore, from Fig. 6 and reaction conditions for PDH, the ratio H2/HC = 0.6 was selected as an optimal value.
Figure 6 Influence H2/HC on catalyst performance (WHSV = 2 h−1, T = 600 °C) ◆ propane conversion; ■ propylene selectivity; ▲ propylene yield
Figure 7 Results of stability test (WHSV = 2.0 h−1; T = 600 °C) ◆ H2/HC = 0.6; ■ H2/HC = 1
3.2.3 Effect of the space-velocity Figure 8 shows the short-term performance test results in terms of STY versus WHSV. Nearly a straight line with only slight downward curvature was obtained. This was an indication of a mainly thermodynamiccontrolled reaction regime. As Fig. 8 shows, the highest STY was achieved at WHSV = 2.2 h−1. This space velocity was selected as an optimal value by the virtue of practically constant propane conversion (33%-35%) in the range of interest. However, higher space velocities should result in the shift to the kinetic-controlled regime and decreasing propane conversion which means higher cost of separation and feed recycle, which is not desirable.
Figure 8 Effect of space velocity on STY of propylene (T = 600 °C; H2/HC = 1.0)
Chin. J. Chem. Eng., Vol. 21, No. 7, July 2013
3.2.4 Optimal operating conditions Using the results of Figs. 2-8, the optimal operating conditions for catalytic runs were obtained. The results are indicative of a better performance of the catalyst under T = 620 °C, H2/HC = 0.6 and WHSV = 2.2 h−1 which corresponds to a STY of about 16 mol·kg−1·h−1. To confirm these results, a long-term performance test was performed under these conditions (Fig. 9). The present test duration (approximately 5 d) approximates one catalyst cycle (7 d) in moving-bed reactors of the Oleflex process before catalyst regeneration in the continuous catalyst regeneration (CCR) unit. The observed catalyst decay rate is acceptable, and to cope with it, the catalyst loading and average reactor temperature is increased in the successive reactor.
Figure 9 Results of stability test in terms of propane conversion and propylene yield under optimal condition (WHSV = 2.2 h−1, T = 620 °C, H2/HC = 0.6) ◆ propane conversion; ■ propylene yield
Comparison with isobutane dehydrogenation
It is noteworthy to compare the dehydrogenation test results for propane and isobutane over the same catalyst. For isobutane dehydrogenation under the approximately same conditions as a commercial practice (T = 575 °C, H2/HC = 0.5 and WHSV = 2.0 h−1), the isobutane conversion and isobutene yield were 55% and 48%, respectively , which are higher than those for propane dehydrogenation under the optimal conditions obtained in this work. Compared to isobutane dehydrogenation, propane dehydrogenation needs much higher reaction temperature which can be explained by the fact that the equilibrium constant for paraffin dehydrogenation decreases significantly as the carbon number decreases . This higher operating temperature is accompanied with slightly higher H2/HC ratio and WHSV to reduce the severity of reaction conditions. When propane conversions depicted in Fig. 9 are converted to catalyst activity and plotted versus time on stream on semi-log axes, a straight line is obtained, implying an independent first-order decay law (Fig. 10). The observed deactivation rate is 0.018% h−1 for propane under the optimal conditions, compared to 0.013 h−1 for isobutane dehydrogenation under its own reaction conditions . The observed higher deactivation rate
Figure 10 Catalyst activity versus time-on-stream (WHSV = 2.2 h−1, T = 620 °C, H2/HC = 0.6)
for propane dehydrogenation compared to isobutane dehydrogenation can be evidently attributed to the much higher operating temperature employed in the former to achieve acceptable conversions. This also explains the need for higher H2/HC and lower space velocity in PDH reaction to compensate this effect in part. Coke formation is the main cause of deactivation of Pt-based dehydrogenation catalysts. A study on dehydrogenation of lower hydrocarbon on these catalysts has shown that propane produced coke in a serial pathway via propadiene, while reactants with five to eight carbon atoms produced coke in serial and parallel pathways . Similarly, the reaction network is much more complicated for isobutane. Byproducts could be propylene (through C C bond rupture) and the conjugated diene (1,3-butadiene through isomerizationdehydrogenation) with the latter known as an important coke precursor. This can explain the higher tendency of isobutane to coke formation and catalyst deactivation than propane under the same operating conditions, imposing lower operating temperature for acceptable catalyst life in isobutane dehydrogenation. 4
The commercial Pt-Sn/Al2O3 isobutane dehydrogenation catalyst exhibits a good performance for propane dehydrogenation, although under different but more severe conditions. The optimal condition for propane dehydrogenation was obtained by one factor at a time method and it was found to be T = 620 °C, H2/HC = 0.6 and WHSV = 2.2 h−1. Despite higher operating temperature, the conversion for propane dehydrogenation was lower than that for isobutane dehydrogenation, mainly due to thermodynamic limitations. The deactivation rate was higher for propane which could be attributed to the higher operating temperature. REFERENCES 1
Letzsch, W.S., “Deep catalytic cracking, the new light olefin generator”, In: Handbook of Petroleum Refining Processes, Meyres,
Chin. J. Chem. Eng., Vol. 21, No. 7, July 2013
2 3 4 5
6 7 8 9
R.A., 3rd edition, McGraw Hill, New York (2004) Heinritz-Adrian, M., Wenzel, S., Youssef, F., “Advanced propane dehydrogenation”, Petroleum Jechonology Quarterly, 84-96 (2008). Bhasin, M.M., McCain, J.H., Vora, B.V., Imai, T., Pujado, P.R., “Dehydrogenation and oxydehydrogenation of paraffins to olefins”, Appl. Catal. A- Gen., 221, 397-419 (2001). Suzuki, I., Kaneko, Y., “Dehydrogenation of propane over chromia-alumina-potassium-oxide catalyst”, J. Catal., 47, 239-242 (1977). Gascón, J., Téllez, C., Herguido, J., Menéndez, M., “Propane dehydrogenation over a Cr2O3/Al2O3 catalyst: Transient kinetic modeling of propene and coke formation”, Appl. Catal. A- Gen., 248, 105-116 (2003). Jablonski, E.L., Castro, A.A., “Effect of Ga addition to Pt/Al2O3 on the activity, selectivity and deactivation in propane dehydrogenation”, Appl. Catal. A- Gen., 183, 189-198 (1999). Aguilar-Rio, G., Salas, P., “Propane dehydrogenation activity Pt and Pt-Sn catalysts supported on mangenesium aluminate: Influence of steam and hydrogen”, Catal. Lett., 60, 21-25 (1999). De Miguel, S.R., Jablonski, E.L., “Highly selective and stable multimetallic catalysts for propane dehydrogenation”, J. Chem. Technol. Biotechnol, 75, 596-600 (2000). Tasbihi, M., Feyzi, F., Amlashi, M. A., Abdullah, A.Z., Mohamed, A.R., “Effect of the addition of potassium and lithium in Pt-Sn/Al2O3 catalysts for the dehydrogenation of isobutene, Fuel Process. Technol., 88, 883-889 (2007). Sint Van, M., Kuipers, J.A.M., Van Swaij, W.P.M., “A kinetic rate expression for the time-dependent coke formation rate during pro-
11 12 13 14 15 16 17
18 19 20
pane dehydrogenation over a platinum alumina monolithic catalyst”, Catal. Today, 66, 427-436 (2001). Meriaudeau, P., Thangaraj, A., Dutel, J.F., Naccache, C., “Studies on PtxSny bimetallic in NaY”, J. Catal., 167, 180-186 (1997). Sault, A.G., Martino, A., “Novel sol-gel based Pt nanocluster catalysts for propane dehydrogenation”, J. Catal., 191, 474-479 (2000). Sanfilippo, D., Miracca, I., “Dehydrogenation of paraffins: Synergies between catalyst design and reactor engineering”, Catal. Today, 111, 133-139 (2006). Moulijn, J.A., Diepen, A.E., Van Kapteijn, F., “Catalyst deactivation: Is it predictable? What to do”, Appl. Catal. A- Gen., 212, 3-16 (2001). Larry Barrentine, B., An Introduction to Design of Experiments: A simplified Approach, ASQ Quality Press, Milwaukee (1999). Breck, D.W., Zeolite Molecular Sieves, John Wiley, New York, 636-641 (1974). Mohagheghi, M., Bakeri, G., Saeedizad, M., “Study of the effect of external and internal diffusion on the propane dehydrogenation reaction over Pt-Sn/Al2O3 catalyst”, Chem. Eng. Techn. J., 30, 1721-1725 (2007). Resasco, D.E., Dehydrogenation-Heterogenous, Encyclopedia of Catalysis, John Wiley, New York (2003). Sahebdelfar, S., Moghimpour Bijani, P., “Modeling of a radial-flow moving-bed reactor for dehydrogenation of isobutane”, Kinet. Catal., 49, 625-632 (2008). Praserthdam, P., Grisdanurak, N., Yuangsawatdikul, W., “Coke formation over Pt-Sn-K/Al2O3 in C3, C5-C8 alkane dehydrogenation”, Chem. Eng. J., 77, 215-219 (2000).