Lober A 2008
Lober A 2008
Lober A 2008
A two-zone fluidized bed reactor (TZFBR) is employed for propane dehydrogenation over a Pt-Sn-K/γ-
Al2O3 catalyst. This type of reactor constitutes an alternative to the use of two separate reactors with solid
circulation between them or of a single reactor with periodic operation and allows continuous operation with
in situ catalyst regeneration. The effect of the main TZFBR operating variables is studied and the reactor
performance compared with results obtained in other works using Pt-Sn based catalysts, showing that a
steady state is reached with similar selectivity to that reported in the literature for other reactors which suffer
the problem of catalyst deactivation.
Figure 5. Propene selectivity vs propane conversion for experiments Figure 7. Influence of relation between length of the reacting and the
performed with different O2 flows fed into the reactor and different relative regenerating zones using the same amount of catalyst in the bed. T ) 500
gas velocityies in the reaction zone, ur1. T ) 500 °C, WT ) 60 g, 50% °C, WT ) 60 g, 50% C3H8, H ) 14 cm, and ur1) 2.8.
C3H8, and ha/H ) 0.5.
multitubular reactor to provide the reaction heat, as is
currently used in some industrial processes.
3.4. Effect of the Ratio between the Heights of the
Reaction and the Regeneration Zones. The ratio between the
heights of the reaction and regeneration zones, (H - ha)/ha,
has two effects on the operation of the TZFBR. An increase
in this ratio involves a higher Wr/F ratio in the reaction zone
and a smaller size of the regeneration zone.
Figure 7 shows the propene yield for the same catalyst
weight in the reactor (WT) but for different propane feeding
points (ha), that is, for the same total bed length (H) but
different ratios between the height of the reaction and re-
generation zones (different Wr/F ratios). The flows of oxygen,
propane, and argon fed to the reactor were the same in all
cases. This procedure was repeated with different percentages
of oxygen fed to the bottom of reactor. For 5% of oxygen,
Figure 6. Coke concentration along the catalyst bed for two different O2 it can be seen that the propene yield increases with (H -
flows fed into the reactor, operating conditions are the same as Figure 5. ha)/ha, but for higher values of (H - ha)/ha, the propene yield
For the 5% O2 run, the profile of oxygen along the bed is shown. decreases. So an optimum propene yield was found when
(H - ha)/ha ) 1; that is, the reaction zone and the
regeneration zone had the same size. Furthermore, stable
the oxygen fed is consumed in the lower reactor zone where operation was not achieved when (H - ha)/ha was higher
the regeneration of the catalyst takes place (Figure 6). than 2 (the height of the regeneration zone was not sufficient
Moreover, the propene selectivity decreases because of the to compensate for the catalyst coking in the reaction zone).
increased formation of COx. It is also possible that if there However, when these experiments were carried out with a
is more oxygen than the amount necessary to burn the coke larger oxygen flow rate (10% of oxygen), the optimum
formed in the reaction zone, some oxygen will reach the up- propene yield was found at higher values of (H - ha)/ha.
per zone of the reactor and new oxidation reactions could Moreover, when the height of the regeneration zone decreases
take place. As Pt/Al 2O 3 catalyst is able to promote (higher values of (H - ha)/ha), the coke content in the catalyst
combustion,15,16 the hydrocarbon will react with the remain- along the bed increases (Figure 8), as may be expected from
ing oxygen to produce COx. the previous reasoning.
As discussed before, dehydrogenation of propane and These results can be explained with the help of a previous
cracking are both very endothermic reactions, so it is ne- kinetic study8 since a higher coke content, as was found in the
cessary to bring energy to the system. The burning of coke experiments with a larger reaction zone, has a greater effect on
formed during the propane dehydrogenation brings part of the dehydrogenation reaction than on the cracking. Therefore,
this energy. A Hysys Plant simulator has been used to more catalyst in contact with the propane feed increases the
calculate the heat involved in the process for two tests with propane conversion (approaching the equilibrium value) but
different percentages of oxygen fed into the reactor (2.5 and decreases the propene selectivity, and for higher values of
5% O2), while maintaining the rest of the operating condi- (H - ha)/ha the propene yield decreases, due probably to the
tions: T ) 500 °C, Wr/FC3H8 ) 90 s kg mol-1, 50% C3H8, higher amount of coke formed per unit of time and the smaller
and ha/H ) 0.5. Taking into account the experimentally size of the regeneration zone.
obtained yields to the main reaction products, the net heat 3.5. Influence of Relative Gas Velocity (ur). Figure 5 shows,
produced in the experiment with 5% O2 is 4940 kJ (kmol besides the effect of percentage of oxygen in the feed, the
C3H8)-1 while with 2.5% O2 it is just 344 kJ (kmol C3H8)-1; propene selectivity versus propane conversion for experiments
that is, in these cases the combustion of coke brings more performed with a constant oxygen percentage (5%) and different
energy than that necessary to carry out the dehydrogenation relative gas velocities, ur. It can be seen that both propane
reaction. This could avoid having to use an expensive conversion and propene selectivity decrease with relative gas
9318 Ind. Eng. Chem. Res., Vol. 47, No. 23, 2008
Figure 8. Coke concentration along the catalyst bed for different ratios of
length of the reacting zone/length of the regenerating zone. T ) 500 °C,
WT ) 60 g, 50% C3H8, H ) 14 cm, and ur1 ) 2.8.
Figure 10. Comparison between flammability region of propane-oxy-
gen-nitrogen at 500 °C and operating zone in TZFBR, and the operation
conditions are the same as in Figure 9.
(17) Kunii, D.; Levenspiel, O. Fluidization Engineering; Elsevier: New (21) Hullmann, D.; Wendt, G.; Singliar, U.; Ziegenbalg, G. Propane
York, 2005. dehydrogenation over supported platinum silicon nitride catalysts. Appl.
(18) Zabetakis, M. G. Flammability Characteristics of Combustible Catal. Part. A 2002, 225, 261.
Gases and Vapours; Bulletin 627; U.S. Department of Interior, Bureau of (22) Jablonski, E. L.; Castro, A. A.; Scelza, O. A.; de Miguel, S. R.
Mines: Washington, DC, 1965. Effect of Ga addition to Pt/Al2O3 on the activity, selectivity and deactivation
(19) Salmones, J.; Wang, J.; Galicia, J. A.; Aguilar-Rı́os, G. H2 in the propane dehydrogenation. Appl. Catal., A 1999, 183, 189.
reduction behaviours and catalytic performance of bimetallic tin-modified
platinum catalysts for propane dehydrogenation. J. Mol. Catal. A 2002, ReceiVed for reView May 27, 2008
184, 203. ReVised manuscript receiVed September 15, 2008
(20) Assabumrungrat, S.; Jhoraleecharnchai, W.; Praserthdam, P.; Goto, Accepted September 28, 2008
S. Kinetics for dehydrogenation of propane on Pt-Sn-K/γ-Al2O3 catalyst.
J. Chem. Eng. Jpn. 2000, 33, 529. IE800848C