Ultrium and Millennium Utilize Different Mechanisms to Improve Metals Tolerance
by Dr. Gerald Woltermann
Modern cracking catalysts are required to deliver excellent selectivity in the presence of ever increasing amounts of contaminant metals such as nickel and vanadium. These contaminants are present in the feed oil as porphyrins or other organic salts(Ref. 1). Upon introduction into the FCC unit these compounds will either thermally degrade or condense to coke, laying down on the catalyst Surface. Upon regeneration the organic portion is removed leaving the oxidized metal on the catalyst. The metal is then free to migrate further into the microsphere or transfer to another particle during subsequent oxidation / reduction cycles in the unit. These metals can and do promote unwanted side reactions such as coke and hydrogen formation and also result in pore blockage and zeolite destruction. Mitigating the deleterious effects of these metals is a major accomplishment of ever improving FCC technology.
The two major metal contaminants have significantly different redox chemistry and effect catalyst selectivity in different ways. Nickel exists, under FCC conditions, in either the +2 or 0 valence state. The latter form is the most active for promoting dehydrogenation reactions leading to coke and hydrogen formation. In the zero oxidation state, the nickel is also quite mobile and readily agglomerates under reducing conditions found in the FCC reactor. Agglomeration reduces the available surface and decreases dehydrogenation activity per unit of nickel. The oxidized or +2 valence of nickel is much less likely to agglomerate into large particles and on a surface such as alumina can embed itself in the alumina structure forming nickel aluminate. In this case the nickel can stay relatively highly dispersed over the course of many redox cycles. Cadet et. al.(Ref. 2) have shown that nickel aluminate can also promote dehydrogenation reactions, albeit less readily than nickel metal. Promotion requires that the nickel atoms be present near the surface of the nickel aluminate such as to be accessible to reactant molecules. Nickel does not readily transfer between particles(Ref. 3).
Vanadium also exists in several oxidation states between +3 and +5. However, it does not reduce to the 0 valence state under FCC conditions. It, therefore, is not as likely to agglomerate into large particles. Even at vanadium levels of greater than 2%, formation of crystalline V2O5 has not been observed(Ref. 4). Unlike nickel, vanadium can readily transfer between particles either by vapor phase transfer of vanadic acid or by solid - solid transfer during particle collision(Ref. 3,5). This fact allows potential passivation of vanadium by solid state additives; a route not available in nickel passivation. Vanadic acid is highly mobile due to its high vapor pressure at FCC temperatures and is known to form low melting eutectics with a number of solid state compounds. Because of its acidity and formation of eutectics, vanadium can cause significant loss in zeolite or matrix surface area. In the +5 oxidation state vanadium is a moderate promoter of dehydrogenation reactions(Ref. 6).
ULTRIUM Passivates Nickel by Promoting Surface Agglomeration
The pore size and chemistry of ULTRIUM were designed to promote agglomeration of nickel on the microsphere surface, thereby significantly reducing its potential to promote dehydrogenation reactions. Secondary Ion Mass Spectrometry (SIMS) mapping of nickel location was done on an ULTRIUM sample impregnated with nickel using a modified Mitchell method(Ref. 7). By nature, Mitchell method impregnation exaggerates nickel dispersion compared to metal deposition in an FCC unit. Figure 1 illustrates results from this study. The white crust on the particle exterior represents area of high nickel concentration and is approximately 3-5 microns thick.
Another method used to ascertain the state of nickel on ULTRIUM is Temperature Programmed Reduction (TPR). In this technique the temperature at which nickel is reduced to Ni (0) in a reducing (hydrogen) atmosphere is measured. By comparing results to those found for nickel in well characterized environments, the state of nickel on the catalyst can be clarified. As can be seen from Figure 2, the ULTRIUM has a significant amount of nickel in easily reduced form similar to nickel on silica. In this environment nickel readily agglomerates to large crystallites as is observed on a silica support.
ULTRIUM passivates vanadium via a trapping component added to the formulation. This magnesium based component traps the vanadium as magnesium vanadate, which prevents its migration and eliminates its ability to destroy zeolite. TEM elemental mapping shows high concentrations of vanadium associated with the magnesium containing trap after lab metallation.
MILLENNIUM Passivates Nickel and Vanadium by Absorbing onto Alumina Rich Sites
BASF's MILLENNIUM incorporates a large pore, surface modified, alumina rich component to easily absorb the large metal porphyrins present in the feed. Because of the alumina and special surface treatment given to it, both nickel and vanadium are immobilized on the surface and cannot migrate to zeolitic sites. This prevents vanadium from destroying zeolite. Because of the stability and low surface area of these pores the metals are buried in the structure and prevented from promoting dehydrogenation reactions. The TPR shown in Figure 2 clearly shows the nickel to be present in sites which render it very difficult to reduce. Bolt et. al.(Ref. 8) showed reduction temperatures of nickel in the 850-900°C region are characteristic of nickel aluminates with the nickel in the tetrahedral holes of the spinal structure. Elemental mapping studies done by Transmission Electron Spectroscopy (TEM) clearly demonstrated that high concentrations of nickel and vanadium on impregnated and equilibrium MILLENNIUM were coincident with high areas of alumina concentration. Figure 3 rounds out the picture by demonstrating loss of surface area of these large pores with increasing vanadium concentration, thereby burying contaminant metal present in the pores.
ULTRIUM and MILLENNIUM Show Lower Coke and Hydrogen Yield and Better Activity Maintenance in the Presence of Metals
In order to demonstrate the catalytic effect of metal passivation by ULTRIUM and MILLENNIUM, we used a cyclic metals deposition unit (CMDU) to impregnate both catalysts plus two well known competitive resid catalysts with 3000 Ni and 3000 V. The metallated catalyst were post steamed in 90% steam and 10% air at 1450°F for 4 hours and tested on a microactivity unit (MAT). Coke and hydrogen yields are shown in Figure 4. The competitive catalyst data is represented as a composite because of the similarity of the data. Hydrogen and coke yield, when using a light feed, are indicative of the metals activity.
ULTRIUM and MILLENNIUM make significantly less coke and hydrogen than the competitive catalyst. Also significant is the activity retention of these two catalysts in the presence of vanadium as shown in Figure 5. Both have a clear advantage over the competition.
Metals Passivation by ULTRIUM and MILLENNIUM are Demonstrated by Commercial Performance
MILLENNIUM was put in a resid unit Cracking heavy feed with high nickel concentration over a popular resid catalyst. The unit was able to significantly reduce makeup rate while maintaining conversion due to the superior activity maintenance Of MILLENNIUM. Moreover the reduced coke yield at the high nickel levels experienced in this unit allowed the refiner to increase throughput. Figure 6 demonstrates the ability of MILLENNIUM to handle very high nickel levels. As the proportion of MILLENNIUM in the inventory increases, coke decreases in spite of a huge increase in nickel on equilibrium catalyst. Hydrogen yields likewise decreased over the same period.
In a commercial trial of ULTRIUM, coke yield decreased on the equilibrium catalyst despite increasing levels of metal on the equilibrium catalyst. These results are represented graphically in Figure 7. The improved metals passivation was likewise mirrored in decreased H2 yields.
Using a variety of analytical tools we have shown ULTRIUM and MILLENNIUM to effectively passivate nickel and vanadium by two different methods. MILLENNIUM primarily passivates both by absorbing the metals onto a large pore size, surface treated, alumina rich component of the catalyst. During aging the trapped metals are buried in the structure making them unavailable to promote dehydrogenation reactions or to destroy zeolite. ULTRIUM, on the other hand, causes nickel to agglomerate into large crystallites on the microsphere surface, which severely limits the nickel's promotion of coke and hydrogen formation. ULTRIUM also has a magnesium based vanadium trapping component which effectively traps and immobilizes the vanadium as magnesium vanadate. The ability of these two catalysts to passivate metals is reflected in MAT selectivity data on lab treated and equilibrium catalyst. Commercial data has also proven their effectiveness in high metals resid operations.
1. Nielsen, R.H. et. al., 'Metals Passivation' in Fluid Cracking Science and Technology p339, Elsevier (1993)
2. Cadet. V. et. al., Appl. Catal., 68, 263 (1991)
3. Wormsbecher, R.F. et. al., Symposium on Deactivation and Testing of Hydrocarbon Conversion Catalysts Preprint, Div. Petr. Chem., ACS Meeting Chicago Ill. 1995
4. Occeli, M. 'Laser Raman and X-Ray Photoelectron Characterization of Vanadium Contaminated Components of Fluid Cracking Catalysts', in Fluid Catalytic Cracking II: Concepts in Catalyst Design, p252, ACS (1991)
5. Lerner, Bruce and Deeba, Michelle, Symposium on Deactivation and Testing of Hydrocarbon Conversion Catalysts Preprint, Div. Petr. Chem., ACS Meeting Chicago Ill. 1995
6. Bock, L.T., et. al., Symposium on Deactivation and Testing of Hydrocarbon Conversion Catalysts Preprint, Div. Petr. Chem., ACS Meeting Chicago Ill. 1995
7. Mitchell, B.R., Ind. Chem. Prod. Res. Dev., 19, 209 (1980)
8. Bolt, P.H., et. al., J. Catal., 151, 300 (1995)
In a previous Catalyst Report, 'The Impact of the FCC Catalyst Matrix on Bottoms Upgrading When Processing Heavy Feed', we neglected to cite an influential paper on the effect of matrix on bottoms upgrading. Interested readers are referred to Alerasool, Doolin, and Hoffman in I&EC RESEARCH, Vol. 34, No. 2, pg. 434 (1995), which was released by Ashland Petroleum. In this publication, robust circulating pilot unit LCO/HCO results using a heavy topped crude feed were shown to correlate well with pseudo-equilibrium matrix acidity. The value in this is that the matrix acidity test, like the MAT in our own work, is a far easier measurement to make than the pilot unit evaluation, yet they can be calibrated to be good predictive tools. Our MAT activities have been found to correlate well with matrix acidity results obtained using Ashland's procedures. Some of the results we reported were obtained under steam deactivation conditions recommended by Alerasool et al.