MATRIX SIEVE, BINDER DEVELOPMENTS IMPROVE FCC CATALYSTS

Oct. 1, 1990
Lawrence L. Upson, R. Joe Lawson UOP Des Plaines, III. William E. Cormier Katalistiks Baltimore Fred J. Baars Katalistiks B.V. Leiderdorp, The Netherlands Several FCC catalyst developments have taken place since the merger of UOP and Katalistiks. These developments include catalyst matrix improvements, zeolite improvements, and better attrition resistance. During the last 2 years, these developments have been tested in the laboratory, pilot plants, and in operating refineries.

Lawrence L. Upson, R. Joe Lawson
UOP
Des Plaines, III.
William E. Cormier
Katalistiks
Baltimore
Fred J. Baars
Katalistiks B.V.
Leiderdorp, The Netherlands

Several FCC catalyst developments have taken place since the merger of UOP and Katalistiks. These developments include catalyst matrix improvements, zeolite improvements, and better attrition resistance.

During the last 2 years, these developments have been tested in the laboratory, pilot plants, and in operating refineries.

Katalistiks has evolved from a small group of FCC experts who gathered together in 1979 to start a new FCC catalyst company to its current status as a member of the technology company known as the new UOP. In the early days, this small nucleus of individuals provided technology support in the areas of catalyst design, development, and evaluation as well as in the area of the FCC process and its interrelationship with other refinery processes.

In mid-1984, Katalistiks became a part of the Union Carbide organization. Catalyst development then concentrated primarily on zeolite improvements to take advantage of the expertise in molecular sieve technology available within Union Carbide. As a result, in 1986 Katalistiks introduced catalysts containing Carbide's LZ210 chemically stabilized and dealuminated sieve.1

The Katalistiks Alpha and Beta series catalysts were a major part of Katalistik's octane catalysts. In 1988 the molecular sieve catalyst and process equipment portion of Union Carbide (CAPS) was combined with a portion of UOP Inc. to form a new legal entity referred to informally as the new UOP.

Programs were immediately started at three laboratory sites on physical property improvement, principally in the area of attrition resistance. At the same time, efforts were initiated to use UOP technology to improve matrix activity and bottoms cracking performance.

Research also was initiated to characterize the LZ-210 zeolite and its performance.

PHYSICAL PROPERTY IMPROVEMENT

Modern FCC catalysts are mixtures of active zeolite molecular sieves and a matrix consisting of one or more catalytically active components. These components promote reactions, such as bottoms cracking or metals passivation, that are preferably or more selectively accomplished outside the zeolite structure.

Typically, clay is also added as a weighing agent or diluent to modify overall cracking activity. Historically, the relative amounts of zeolite and active matrix components within the catalyst formulation have increased over the years as the FCC unit has been asked to provide more upgrading capability, i.e., higher octane, increased conversion, reduced bottoms, or a greater degree of resid processing. At the same time environmental pressure has been increasing to reduce particulate emissions from FCC units.

The combination of these two developments has placed increasing demands on the binding portion of the FCC catalyst matrix to provide the intrinsic particle integrity necessary for effective operation. Intensive studies in 1988 of the factors affecting the formation of strong catalyst particles resulted in substantial changes in manufacturing procedures to achieve attrition improvement.

These changes were implemented in the catalyst manufacturing plants in stages, starting in mid-1989, along with improvements in control techniques. Fig. 1 shows the commercially realized attrition resistance improvement achieved in 198990 for an otherwise constant composition product.

The attrition improvement is indicated by a decrease in Katalistiks' attrition index (Ks). Improvements in the period from May to June 1989 were achieved by modification to existing process conditions and equipment.

The later July-August improvement was accomplished with the infusion of significant capital investment to both modify existing operations and add new process unit operations. By the fall of 1989, these improvements had reduced the attrition index by a factor of 2.

With the improvement in attrition well established, studies shifted to the binder itself.

Development work indicated that a further 30-40% improvement in attrition resistance could be achieved. These improved binder procedures were commercialized in late February 1990.

The results of the new procedures are shown in Fig. 1. The expected improvement resulted in a typical Katalistiks attrition Index (Ks) of 0.35 for this high zeolite, high-activity product.

ZEOLITE DEVELOPMENT

The introduction of Union Carbide's LZ-210 molecular sieve into FCC use in 1986 marked a major FCC catalyst sieve development since the ultra stable Y (USY) sieve was introduced in the early 1970's. As with USY, the LZ210 sieve is a Y-type sieve that has been partially dealuminated in the catalyst manufacturing process to produce a higher SiO2/Al2O3 ratio and thus is a more stable sieve than the starting NaY material.

The difference is that an LZ-210 sieve is chemically dealuminated by inserting silica quantitatively into the vacancies in the crystal lattice created by the dealumination and removing from the crystal structure all of the alumina debris that was produced by this process .2 In the USY process, dealumination is carried out at high temperature in a steam environment.

The alumina removed from the framework remains in the crystal structure as debris, which functions as a nonselective catalytic agent .3

Since the introduction of the LZ-210 sieve, much R&D work has been done to refine the properties of this material and further enhance its utility in FCC applications. This work led to an LZ-210 modification designated as LZ-210K, which provides FCC performance advantages over the standard LZ-210 sieve.

The LZ-210K stabilization procedure changes the acidsite characteristics of the zeolite. Thus, hydrogen transfer capability can be altered to achieve enhanced gasoline octane without the usual penalty of reduced catalyst activity.

Laboratory results from testing a zero rare earth catalyst containing the LZ-210K sieve compared to a similar catalyst (approximately the same amount of sieve and the same amount and type of matrix) containing the original LZ-210 sieve are given in Tables 1 and 2. When compared with a catalyst containing an LZ-210 Beta zeolite, the catalyst containing the LZ-21 OK sieve achieved better bottoms cracking and produced a higher octane gasoline.

Comparing total potential gasoline (FCC + net alkylate production), the LZ-21 OK catalyst produced equal gasoline yield at 0.5 RON higher and 0.3 MON higher. Compared to an LZ-210 Alpha catalyst, the LZ-21 OK catalyst was equivalent in bottoms cracking and total potential gasoline yield but produced 1.3 higher RON and 1.8 higher MON.

A catalyst containing the new LZ-210K sieve was put into commercial use in a North American refinery (Refinery A) in December 1989. The previous catalyst was also a Katalistiks catalyst containing an LZ-210 zeolite, intermediate in SiO2/Al2O3 ratio between a Beta LZ-210 zeolite and an Alpha LZ-210 zeolite.

The zeolite content and the zeolite rare-earth level were the same in both cases. As seen in Table 3, the RON and MON increased despite a feed quality change that directionally would have been expected to reduce octanes.

After correcting the octanes to constant conditions (including constant feed quality), the data indicate that RON was increased by 0.6 and MON increased by 0.2.

These commercial octane gains, which are somewhat less than anticipated from the lab study, may be due to the feed quality variable having a larger impact than it was adjusted for. Close monitoring of the results from this refinery continues to provide a better understanding of the effect of feed quality.

MATRIX DEVELOPMENT

Katalistiks began a major R&D program in 1988 to develop new matrix materials for FCC catalyst use. The goal was to substantially improve bottoms cracking and LCO selectivity and maintain excellent low coke and gas selectivity. These improvements had to be achieved without any detrimental effect on catalyst attrition resistance.

By the end of 1989, two new matrix systems that satisfied these objectives had emerged from this program. They are designated as Matrix I and Matrix II.

The development work on Matrix I is complete, and the catalyst has been commercialized in several refineries.

Laboratory data on this matrix system are shown in Figs. 2-4. These data were generated in a MAT unit at 950 F. reactor temperature using a naphthenic feed that enhances the differences in LCO selectivity between catalysts.

Conversion was varied by changing cat/oil ratio. Catalyst formulations, including amount of active matrix and zeolite type and amount, were identical, except for matrix type.

For rating purposes, the new matrix is compared to the standard Katalistiks bottoms cracking matrix. The new matrix system significantly improves LCO selectivity, especially at conversions lower than 70 wt %.

The improvements are typified in Fig. 2: at a constant 65 wt % conversion, the LCO yield is relatively more than 5% greater than a similar Katalistiks catalyst containing the standard bottoms cracking matrix. At constant operating conditions, the new matrix catalyst increases catalyst activity approximately 10% over the catalyst containing the standard matrix (Fig. 3).

Since all catalysts contain the same zeolite type and amount, the activity increase must be attributed solely to matrix activity. As may be suspected from a more active matrix coke yield (Fig. 4) is up slightly.

The first commercial trial with the Matrix I material was carried out early in 1990 at Refinery B. Refinery B was using Beta 640 catalyst, which contains Katalistiks standard matrix.

In early February 1990, Matrix I replaced the standard matrix in the Beta 640 formulation. All other components in the catalyst remained constant. The results of the trial are shown in Table 4.

The matrix surface area in the equilibrium catalyst increased 20% as a result of the catalyst reformulation. The zeolite surface area remained constant, which is consistent with the fact that no change was made in either type or content of zeolite in this catalyst reformulation.

The increased matrix activity resulted in a two number increase in equilibrium catalyst activity (71 to 73) and a corresponding increase of 1.3 vol % in conversion.

Product selectivities were essentially unchanged except for the LCO-HCO split.

The LCO selectivity, LCO/(LCO+HCO), increased from 49.4 to 57.5 wt %.

The increase in LCO selectivity as the catalyst transition occurred is further illustrated in Fig. 5. The largest gain in LCO selectivity was obtained after two thirds of the matrix transition had occurred.

This trial thus confirmed the improved matrix performance, i.e., improved activity and improved LCO selectivity, that was found in the laboratory tests.

Matrix 11 incorporates an approach to matrix technology that is based on concepts recently developed in UOP's Des Plaines research center. A catalyst using these concepts has been produced commercially, and this catalyst is now in use in a Western European refinery to verify the expected benefits.

One of the unique features of this new matrix material is its enhanced stability. This stability is illustrated in Table 5 by comparing Matrix II to a typical commercial boehmite-type alumina that is frequently used in FCC catalyst formulations.

Not only is surface area stability improved (53% surface area retention after 1,500 F. for Matrix II vs. 32% for commercial boehmite), but also the catalytic activity actually increases upon steaming.

A MAT comparison of two Beta 440-11 catalysts containing the same sieve input and the same quantity of active matrix but differing in matrix type is shown in Figs. 6-8. As seen in Fig. 6, the catalyst containing Matrix II is approximately 20% more active.

At constant conversion (60 wt %), the Matrix II catalyst produced about 10% more LCO (Fig. 7). Perhaps of most interest is the fact that this increased matrix activity was achieved without the usual coke-selectivity penalty.

Fig. 8 shows that the coke selectivity actually was better than the already good selectivity of the standard matrix.

Refinery C has been using Beta 500-11 since November 1988. Beta 500-11 is a low H-transfer octane catalyst containing an LZ-21 0 sieve and a higher than typical amount of the standard active alumina matrix.

In December 1989, the Matrix II material replaced the standard matrix in the catalyst formulation. To compensate for the anticipated increase in the activity of the new matrix the zeolite content of the new catalyst was reduced.

Table 6 summarizes the operating data available from Refinery C. The table compares data related to the performance of the Beta 500-11 catalyst containing Matrix II with previous results from the standard matrix.

At the time that these data were obtained, approximately 75% of the inventory had been replaced by the new matrix system. Operating conditions and feed quality are similar for the two cases shown in Table 5, and so detecting catalyst effects is relatively easy.

In these data, the increased activity and stability of the new matrix are becoming noticeable. After a catalyst replacement of 75%, the matrix surface area at that point has gone up more than 14%.

This activity more than compensates for an 11 % decrease in zeolite content and results in a two-number increase in equilibrium catalyst activity. This activity increase translated into the expected increase in refinery FCCU conversion of approximately 1 %.

These data strongly indicate the enhanced bottoms cracking and LCO selectivity. The 0.9 vol % increase in conversion came entirely out of the unconverted bottoms, and so the bottoms yield was reduced by nearly 15%, from 6.9 to 5.9 vol % of fresh feed.

The LCO selectivity increased from 72.5 to 75.6%. The data also show that this increase in matrix activity is not accompanied by a deterioration in coke selectivity.

The 4 F. increase in regeneration temperature shown in Table 6 for the Matrix 11 period is consistent with the expected increase in temperature that results from a 0.9% increase in conversion and a slight decrease in feed quality. The catalyst coke selectivity is at least as good as the previous matrix.

Changes in the characteristics of the equilibrium catalyst used in Refinery C were followed closely during the entire transition period (Fig. 9). The matrix surface area increased 25%, and the zeolite surface area decreased nearly 20%.

The zeolite surface area drop is consistent with the nearly 20% decrease in zeolite input in the reformulated catalyst. The more active matrix is expected to compensate for the decreased zeolite content.

In fact, the matrix activity increase actually is larger than anticipated, as the equilibrium catalyst activity at constant catalyst addition increased significantly (Table 7).

All of the developments discussed have been utilized in the Katalistiks Omega series FCC catalysts. All catalysts in the series contain the Z-21 OK matrix.

REFERENCES

  1. Pieper, R. E., Valeri, F., and Letzsch, W. S., "LZ-210 Update Commercial Performance," paper presented at Katalistiks 8th annual FCC Symposium, Budapest, June 1987.

  2. Breck, D. W., and Skeels, G. W., U.S. Patent 4,503,023, Mar. 5,1985.

  3. Pellet, R. J., Blackwell, C. S., and Rabo, J. A., "Catalytic Cracking Studies and Characterization of Steamed Y and LZ-210 Zeolites," Journal of Catalysis, Vol. 114, Nov. 1, 1988, pp. 71-89.

Copyright 1990 Oil & Gas Journal. All Rights Reserved.