IMPROVED FCC YIELDS THROUGH A COMBINATION OF CATALYST AND PROCESS TECHNOLOGY ADVANCES
Joseph B. Mclean,
Executive Technology Specialist
Edward B. Bovo, Technical Service Engineer
Petroleum Catalysts Group
1800 St.James Place, Suite 501
Houston, TX 77056
Presented at the
March 20-22, 1994
San Antonio, Texas
Dramatic yield improvements have been demonstrated using the latest FCC process technology advances. These include improvements in feed dispersion, stripper and regenerator design, and riser termination for short contact time cracking. Riser termination designs in particular have received much attention, for both new units as well as revamps. Effective systems have been developed to rapidly separate catalyst from hydrocarbon vapors, reducing non-selective post-riser cracking. Improved selectivities result, reducing coke and dry gas formation and allowing unit operation at higher severity. Proper catalyst selection is essential to realize the maximum potential benefit from these hardware improvements.
This paper discusses the types of yield changes which can be expected in a revamp scenario incorporating the latest riser termination designs, particularly when catalyst change effects are included. Both fresh catalyst properties and catalyst management policy are important variables. Process model studies for several hypothetical operating cases are presented, including the use of a high activity catalyst for maximum conversion or throughput, a high matrix catalyst for reducing slurry yield, a metals tolerant catalyst for processing resid feed, and a high olefin selectivity catalyst for a reformulated gasoline scenario. In all cases, improved yields are achieved when incorporating catalyst change effects which may not have been possible without the advanced hardware due to unit constraints. Commercial data from several operating units which have implemented these systems is presented which also highlights the interaction between process and catalyst improvements.
Advanced Riser Termination Systems
Early FCC units traditionally used dense phase bed reactors, with one or two-stage cyclones above the bed to recover catalyst from hydrocarbon product vapors. As more active catalysts were developed, the industry has moved away firom dense bed cracking to all-riser configurations.
This is an excellent example of how catalyst and process technology advances have worked together to improve overall performance. It has been recognized for several years that significant post riser cracking occurs in commercial FCC units resulting in substantial production of dry gas and other low valued products(Ref. 1). There are two mechanisms by which this occurs, thermal and dilute phase catalytic cracking. Thermal cracking results from extended residence times of hydrocarbon vapors in the reactor disengaging area, and leads to high dry gas yields via non-selective free radical cracking mechanisms. Dilute phase catalytic cracking results from extended contact between catalyst and hydrocarbon vapors downstream of the riser. While much of this was eliminated in the transition from bed to riser cracking, there is still a substantial amount occurring in the dilute phase due to significant catalyst holdup which occurs without an advanced termination system design. Dilute phase cracking increases dry gas production as well, and also causes additional coke formation (higher "delta coke") on the surface of the catalyst. High hydrogen transfer selectivities are also characteristic of this secondary reaction process, resulting in gasoline octane loss and reduced LPG olefinicity through formation of C3/ C4 saturates, especially isobutane. Reaction mix sampling studies for various units have shown that up to 50% of the total dry gas and C3/C4 saturates can be formed from post-riser cracking, in most cases with little increase in more valuable yields of gasoline or C3/C4 olefins(Ref. 2). Additional conversion which occurs in the dilute phase varies among units, but in general is no more than a few percent. Because most units operate against constraints which are related to the total coke and gas production, unit operating severity is typically restricted as a result of these non-selective reaction mechanisms.
The major FCC process licensers have developed advanced riser termination systems to minimize post-riser cracking(Refs. 3-7), and many units have implemented these in both new unit and revamp applications. In addition, some refiners have implemented their own "in-house" designs for the same purpose. Due to the complexity and diversity of existing FCC units as well as new unit design differences, there are many variations of these systems such as "closed" cyclones, "close-coupled" cyclones, "direct coupled" cyclones, "high containment systems", "short contact time", etc. There are differences in the specific designs, and some may be more appropriate for specific unit configurations than others, but all serve the same fundamental purpose of reducing the undesirable post-riser reactions.
There are many options for taking advantage of these new systems to improve yields. Some combination of higher reactor temperature, higher cat/oil ratio, higher feed rate, and/or poorer quality feed is typically employed. It is interesting that very little notice in the references cited is given to what types of modifications to the catalyst should be made to complement these designs, particularly for revamp applications. FCC catalyst formulations today are highly specialized and customized to each refiner's specific needs. Therefore it is not surprising that significant catalyst modifications should be appropriate when a change of this magnitude in a unit's yield pattern occurs due to hardware effects. In fact, this is what has happened in most cases where revamps have been performed incorporating advanced riser termination systems. Typical catalyst selection objectives such as low coke and dry gas selectivity are reduced in importance due to the process changes, while other features such as activity stability and bottoms cracking selectivity become more important for the new unit constraints. Changes in catalyst management policy - balancing addition rates with fresh catalyst properties may also be appropriate.
Process Model Studies
The study conducted here considered a hypothetical unit which incorporates an advanced riser termination system in a revamp. Heat balanced process model studies were conducted using a combination of the Profimatics FCC model and BASF's catalyst model to predict process operating condition and catalyst change effects. Parameters for the base case are summarized in Table 1. These conditions were chosen to be typical for a unit feeding VGO and operating in full combustion. Since none of the base case conditions were considered extreme, flexibility was allowed for the post-revamp studies to change conditions significantly while still staying in the range of typical commercial operations. The base case catalyst was considered to be an octane barrel oriented RE-USY catalyst, with properties estimated to be typical of the average catalyst in use in VGO cracking units based on BASF's equilibrium catalyst data base. This also left room for considerable flexibility in modifying catalyst properties for the alternative model cases.
The changes due to the revamp were estimated based on published reports of typical changes observed for elimination of dilute phase contributions via introduction of an advanced riser termination system. Some realistic constraints were placed on the post-revamp operation to help define the effects of process and catalyst changes. These included assumptions that the original base case values for total wet gas, air rate, and regenerator temperature could not be exceeded. Riser and feed preheat temperatures and catalyst makeup rate (for the base case catalyst) were the primary independent variables considered. Most of the cases were run with a maximum catalyst circulation constraint as well, equivalent to 12% over the initial base case rate. Since circulation flexibility is a parameter that varies widely for different units, some cases were run where this constraint was removed.
Yield changes due to the revamp without changing catalyst or conditions are shown in Table 2. Two cases are shown corresponding to whether or not the circulation constraint was imposed. The lost conversion due to elimination of the dilute phase contribution consists primarily of dry gas reduction, with much smaller changes in gasoline and LPG yields. Slurry yield increases slightly, and there is a minor change in C4 olefins since the lost dilute phase contribution produced mainly saturates.
Of course, in a real refinery situation other operating conditions will be changed to take advantage of the unloading of the unit constraints. Different refiners may choose to respond in different directions based on their specific objectives. Typically some combination of higher conversion, higher feedrate, poorer quality feed, and/or selectivity shifts to more valuable products are desired. Based on the new operation indicated in Table 2, studies were conducted for several potential scenarios to accomplish one or more of these objectives. Each scenario was evaluated first by considering process changes only, then expanded to include the effects of catalyst changes as well.
Performance for cases involving catalyst changes is predicted by integrating relative catalyst activity and selectivity responses into the heat-balanced model and varying conditions subject to the unit constraints. The relative properties for the alternative catalysts are shown in Table 3. Note that the coke and dry gas selectivities for the high activity, high matrix, and high olefin catalysts are higher than the base catalyst, so that these catalyst work best in conjunction with the delta coke and gas reductions achieved via the hardware modifications. The high activity catalyst was considered for higher conversion and higher feedrate cases. The high matrix catalyst was also considered for the higher feedrate case, with a secondary objective to minimize slurry yield. The high olefin catalyst was used for a reformulated gasoline oriented application where a maximum yield of isobutylene for MTBE production was desired. The resid catalyst differs from the others in that it features lower coke and dry gas selectivities than the base, as well as a more favorable metals response factor which becomes important for resid feed applications.
Maximum Conversion Scenario
In a situation where feed availability is fixed, it is desirable to maximize conversion subject to the wet gas and air constraints. Table 4 shows results for this scenario. If the base catalyst is used, addition rates increase from 5 to 12 T/D. Alternatively, the reaction temperature can be increased to 1020 F while maintaining catalyst additions at 5 TPD. The conversion in this case is much lower, due to the stronger wet gas response to temperature. A combined case is also shown, which may be considered a reasonable compromise. Finally, the high activity catalyst case is shown, which allows a much higher conversion at the base reactor temperature without increasing addition rates. The higher activity catalyst would be more expensive on a $/ton basis than the base catalyst, but overall catalyst cost would be lower since the lower addition rates could be maintained.
Figure 1 shows the results for these cases graphically. The highest conversion is obtained for the high catalyst addition case, with the response for the high activity catalyst close behind. High reactor temperature is the least effective way to maximize conversion. While the results improve for the high temperature case if there is no circulation constraint, it is still not as effective as the catalyst change.
Increased Feedrate Scenarios
Another common scenario is that more feed of similar quality may be available, and it is desirable to increase feedrate up to the unit constraints, but without losing conversion relative to the original base case. Table 5 shows results for increasing feedrate by 10 %. The high activity catalyst is again compared with a combination of increased temperature and catalyst makeup for the base catalyst. A higher conversion and more favorable selectivity pattern is indicated for the high activity catalyst. Figure 2 shows that, as feedrate is increased, a higher conversion can be achieved at the wet gas limit for the high activity catalyst. Conversely, for a given conversion target, a higher incremental feedrate can be achieved with the high activity catalyst.
Another common objective would be to increase feedrate while at the same time minimizing production of low value slurry. The slurry production rate is a function of feedrate, conversion, and bottoms upgrading selectivity. The high activity catalyst maximizes conversion as discussed above, and thus reduces slurry yield relative to the base catalyst. Alternatively, a high matrix activity catalyst featuring enhanced bottoms upgrading selectivity can be used. Table 6 and Figure 3 show that, for a given feedrate and conversion target, the high matrix catalyst achieves the lowest overall slurry yield. The dry gas and regenerator temperature limitations which would have resulted from using this catalyst in the original pre-revamp case have been mitigated by the hardware changes.
Resid Feed Scenario
Another common situation is that a refiner would take advantage of the revamp by running heavier and lower cost resid feed to the unit. Table 7 shows cases where 10 % of the VGO feed has been replaced by vacuum resid. The combined higher temperature / higher addition case for the base catalyst was compared with the more selective resid catalyst. The resid catalyst achieves a higher conversion and more favorable yields, with the difference accentuated due to the higher coke producing tendency of the feed and the higher metals loading on the catalyst relative to the VGO operation. Figure 4 shows that a higher quantity of resid feed can be added at a given conversion target when the resid catalyst is used.
Maximum C4 Olefin Scenario
For a refiner interested in producing reformulated gasoline, it is desirable to maximize the total production of C4 olefins for MTBE and alky feed. It is common practice to run at high riser temperatures when light olefins are desired. BASF has previously reported(Ref. 8) that it is more effective to use a high olefin selectivity catalyst rather than temperature to maximize potential MTBE production. Table 8 shows how such a catalyst can be used in combination with the riser termination system to dramatically increase C4 olefin production compared with riser temperature alone. The economics for this type of catalyst and hardware combination have been reported to be potentially favorable for situations where high levels of reformulated gasoline are produced(Ref. 9).
Commercial Performance Summary
Data from four operating units which have incorporated advanced riser termination designs is summarized in Table 9. The feed quality did not change appreciably in any of these cases, but each refiner did use a combination of higher feedrate, temperature, and catalyst property changes to produce overall yield improvements compared with the pre-revamp operation. Conversion is increased in all cases, as is overall gasoline production when feedrate changes are considered. Actual coke yields are variable depending on what other parameters were changed to heat balance the unit, but the relative coke selectivity improvements can be noted by comparing regenerator temperature responses to what would normally be expected for the indicated feedrate and reactor temperature increases.
Yield changes of this nature have been reported before, but few results have been published showing the comparison of catalyst properties. The general trend shows increases in fresh and equilibrium activity. The average equilibrium MAT activity increased by nearly 5 MAT numbers for these four refiners. Some refiners have increased activity via both surface area and rare earth. The effects of higher rare earth on octane have been mitigated by the higher reactor temperatures employed. Increases in matrix surface area are also indicated as these refiners have opted to take advantage of improved bottoms upgrading catalysts. In many cases these formulation changes were adopted over a period of time as the refiner made progressive moves to adapt to the dramatic changes in unit operation as a result of the revamp. Overall it is apparent that the types of catalyst changes discussed in the case studies above have in fact been implemented in commercial FCCUs with these advanced technology systems.
Dramatic yield improvements have been demonstrated by implementing the latest FCC hardware technology, in particular advanced riser termination designs to minimize non-selective post-riser cracking. The potential yield benefits are magnified when combined with the proper catalyst choice for the new operation. In nearly all cases the optimum catalyst choice will change when advanced riser termination designs are implemented. Depending on how an individual refiner chooses to optimize the new operation, beneficial catalyst changes may include higher activity for increased conversion and throughput, higher matrix activity for enhanced bottoms cracking while minimizing coke and gas penalties, higher olefin selectivity for reformulated gasoline production, or improved resid tolerance for operation with poorer quality feeds. Potential catalyst formulation changes and the combined impact of process and catalyst changes on performance should be addressed by any refiner considering the implementation of these new technologies.
1. A.A. Avidan and F.J. Krambeck, "FCC Closed Cyclone System Eliminates Post Riser Cracking", NPRA 1990 Annual Meeting
2. W. Letzsch, R.J. Campagna, and D.C. Kowalczyk, "New Optimization Tools and Technologies for the FCCU", NPRA 1993 Annual Meeting Paper AM-93-66
3. R.E. Wrench and P.E. Glascow, "FCC Hardware Options for the Modern Cat Cracker", AlChE Symposium Series #29, 1992
4. L.L. Upson and D.E. Wegerer, "Rapid Disengager Techniques in Riser Design", ACS 3rd International Symposium on Advances in Fluid Catalytic Cracking, August 1993
5. S.L. Long, A.R. Johnson, and D. Dharia, "Advances in Residual Oil FCC", NPRA 1993 Annual Meeting Paper AM-93-50
6. D. C. Draemel, "Flexicracking IIIR ER& E's Latest Cat Cracking Design", JPI Petroleum Refining Conference 1992
7. F.H.H. Khouw, "Shell Residue FCC Technology: Challenges and Opportunities in a Changing Environment", JPI Petroleum Refining Conference 1992
8. J.B. McLean. G.S. Koermer, R.J. Madon, and W.S. Winkler, "Reformulated Gasoline Catalyst's Impact on FCCU", NPRA 1992 Annual Meeting Paper AM-92-45
9. J.B. McLean and A. Witoshkin, "Iso-Olefin Production for Oxygenates Using IsoPlus", NPRA 1993 Annual Meeting Paper AM-93-17