Tests Show Effects of Nitrogen Compounds on Commercial Fluid Cat Cracking Catalysts

Introduction

BASF presented at the NPRA meeting held on March 23-26, 1986, a technical paper entitled "Matrix Effects in Catalytic Cracking" authored by Messrs L. D. Silverman, W.S. Winkler, J.A. Tiethof, and A. Witoshkin. This paper discussed the role which the catalyst matrix plays in fluidized catalytic cracking operations.

A section of this paper discussed the role that a catalyst's matrix plays in improving its tolerance to FCC catalyst poisons such as heavy metals and nitrogen, which are concentrated in the heavier end of gas oils, especially in the residuum. These poisons contained within large hydrocarbon molecules deposit on the catalyst matrix. Each type of contaminant affects catalyst performance differently and the catalyst matrix plays an important role in trapping these contaminants, and inhibiting a poison's effects on both catalyst activity and selectivity.

The following article was authored by Messrs. Julius Scherzer and Dennis P. McArthur of Unocal Corporation, Brea, California. It is an excellent technical discussion on the effects nitrogen compounds can have on the performance of an FCC operation including the FCC catalyst's activity and selectivity. Among the conclusions reached at the end of this article are the following:

  1. - At constant steam deactivation temperature, an increase in feed-nitrogen content results in a decrease of conversion and gasoline yields and an increase in LCO and Decant Oil (DO) yields.

    - At constant conversion, an increase in feed-nitrogen content results in a decrease in gasoline and DO yields and an increase in LCO, coke, and hydrogen yields.

    - High zeolite content and matrix type play a role in enhancing the nitrogen resistance of FCC catalysts.

The above conclusions are consistent with independent research conducted at BASF R & D laboratories. For additional background on the role played by catalyst matrix in nitrogen tolerance, we suggest reading the above-mentioned BASF Technical paper and also the BASF Catalyst Report (T.l. 784) entitled "Improve Catalyst Resistance To Feedstock Poisons By Increasing Matrix Surface Area."

A copy of both publications can be obtained directly from BASF Corporation either by asking your local BASF Technical Service representative or calling 1-800-932-0444, in N.J. call 1-800-624-0818.

Our thanks are extended to Dennis McArthur, Julius Scherzer, Unocal Corp. and the Oil and Gas Journal for granting us permission to reprint this article for the benefit of our readers.


Six commercial fluid catalytic cracking (FCC) catalysts have been evaluated for their resistance to the effects of nitrogen compounds on their performance.

The results of the evaluations are important because of the shift in the quality of FCC feedstocks to heavier types containing higher levels of sulfur, metals, and nitrogen.

The impact of feedstock quality on the operation of FCCU's has been reviewed in several papers(Ref. 1,2). The deleterious effects of nitrogen compounds on the performance of cracking catalysts has been known for many decades(Ref. 3,4,5).

Significant progress has been made more recently in the identification and characterization of nitrogen compounds present in different oil fractions and feeds(Ref.6,7,8). Furthermore, it was shown that hydrotreatment can significantly change the nitrogen-type distribution in a feedstock. Fig. 1 shows the different types of nitrogen-containing compounds found in petroleum fractions.

Effects Minimized

Several approaches can be used to minimize the deleterious effects of nitrogen compounds during the catalytic cracking of high-nitrogen feedstocks:

  1. 1. Hydrotreating. This is a wellknown technique used to decrease the nitrogen content in feedstocks. This process requires high pressures and temperatures.

    2. Use of solid adsorbents for the removal of nitrogen compounds from feedstocks. The solid adsorbent is usually acidic and "captures" the basic nitrogen compounds. This method has been used to separate basic asphaltenes from feedstocks(Ref. 11,12) and for syncrude denitrogenation(Ref. 13).

    3. Use of immiscible solvent to extract nitrogen compounds(Ref. 14,15,16). The method has been recommended to separate nitrogen compounds from shale oil.

    4. Neutralization of basic-nitrogen compounds in feedstocks using acid additives. The products resulting from neutralization can be separated(Ref. 16) or left in the feedstock(Ref. 17).

    5. Use of nitrogen-resistant FCC catalysts. The major advantage of this method is elimination or greatly reduced cost of a pretreatment process.

Post Studies

A number of papers have been published describing the cracking of model compounds over amorphous cracking catalysts(Ref. 3,4) and hydrogen Y-zeolites(Ref. 18,19) in the presence of various nitrogen-containing compounds. The relationship between the nitrogen content of an FCC feedstock and its effect on cracking catalysts has also been discussed in the literature(Ref. 20,21).

In a recent publication, Fu and Schaffer(Ref. 22) reported the effects of a variety of individual nitrogen compounds on the activity and selectivity of two commercial FCC catalysts. The authors found a correlation between the gas-phase proton affinity of the nitrogen-containing molecule and its poisoning effect on cracking catalysts.

Although the poisoning effects of different nitrogen compounds on cracking catalysts have been described in the literature, there is little information available with regard to the nitrogen resistance of different types of cracking catalysts. Nor are there any correlation between catalytic activity and the distribution of nitrogen (N) compounds in the liquid products and spent catalyst.

Feedstocks

We used a variety of feedstocks with different levels and types of nitrogen compounds. The properties of the feedstocks are shown in Table 1.

Feedstock F-1 is a vacuum gas oil (200 to 540 C. cut) similar to feedstocks used by California refineries. It contains about 0.3 wt % N. Two other feedstocks, F-2 (0.5 wt % N) and F-3 (0.75 wt % N), were obtained by blending F-1 with shale oil, F-4.

FCC Catalysts

The commercial catalysts used in our tests were selected on the basis of catalytic activity, selectivity, and physical properties. Analytical data and properties of the commercial catalysts are shown in Table 2. The catalysts were selected from different manufacturers and differed significantly in their compositions and properties.

The data in Table 2 show that commercial catalysts K and L have high silica content, while catalysts N, O, and P have high alumina content.

Peak height measurements, chemical analysis, and microactivity test (MAT) data obtained with a low-nitrogen feedstock, show that catalysts L and P have higher contents of rare earth Y-zeolite (REY). Catalysts K, N, and O have lower zeolite contents and lower cracking activity.

Catalyst M is an octane-enhancing catalyst and contains ultrastable Y-zeoiite (USY). These catalysts also have significant differences in surface area, pore volume, and bulk density.

Testing Method

Steam-deactivated catalyst samples were tested with different feedstocks in a MAT unit. The steam deactivation was done with 100% steam at 760, 788, and 815C, respectively, for 5 hr.

The MAT runs were done under the following conditions: 510C reaction temperature, 14.5 weight hourly space velocity (WHSV), and 3.5 cataIyst/oil ratio. The cut point for gasoline was 232C and for LCO (light cycle oil), 355C. DO (decanted oil) was the 355C + liquid fraction.

 

Results and Discussion

MAT data obtained for six commercial FCC catalysts with feedstocks F-1, F-2, and F-3 are plotted in Figs. 2-7. Fig. 2 shows conversion and gasoline yields as a function of feed nitrogen for catalysts steam-deactivated at 788 C for 5 hr.

The plots show a general decline in conversion and gasoline yield with increasing feed nitrogen content. Such a decline is due to increased poisoning of the catalysts by the nitrogen compounds in the feedstock.

The figure also shows that the most active among the catalysts tested are catalysts P and L. These two catalysts show a smaller decline in conversion and gasoline yields compared to the other catalysts.

Fig. 3 shows a general increase in LCO and DO yields with increasing feed nitrogen content. Catalysts P and L give the lowest LCO and DO yields.

MAT data obtained with feedstock F-2 at constant coke make of 5.6 wt % are shown in Table 3. The data show that, under these conditions, catalyst P is the most active among the catalysts tested, giving the highest conversion and gasoline yield. However, the hydrogen yield obtained with catalyst P is also very high.

The high activity and gasoline selectivity of catalysts P and L are most likely due to the high content in rare-earth-exchanged Y-zeolite, as indicated by the high x-ray diffraction (XRD) peak height values and high rare-earth content (Table 2). The differences in hydrogen selectivity between catalyst P and L (Table 3) are likely due to differences in the nonzeolitic components of the two catalysts. Such differences could also explain the higher activity of catalyst P vs. catalyst L, in spite of the higher peak height and rare earth content of the latter.

The effect of feed nitrogen content on gasoline, coke, LCO, and DO selectivities, as a function of conversion, is illustrated in Figs. 4 and 5, using catalyst K as an example. The plots in Fig. 4 show that at constant conversion, an increase in feed-nitrogen content results in an increase in LCO and a decrease in DO yields.

The decrease in DO yield with higher feed nitrogen content at constant conversion illustrates an interesting selectivity effect caused by the nitrogen compounds in the feedstock. It suggests that a higher nitrogen con-tent favors DO conversion.

The increase in DO conversion can be understood if one considers that the reaction severity—or catalyst activity—has to be increased when cracking a feed with higher nitrogen content in order to maintain constant conversion.

Correlations similar to those described for catalyst K have been found for other catalysts investigated. The effect of feed-nitrogen content on catalytic selectivity can be seen for catalysts N and O in Table 4.

The data show that at constant conversion (64 vol %), an increase in feed-nitrogen content results in a decrease in gasoline and DO yields and an increase in LCO, coke, and hydrogen yields. Also, the bigger the change in feed-nitrogen content, the more significant are the selectivity changes observed. Fig. 6 shows the effect of feed nitrogen content on C3 and C4 olefin selectivities as a function of conversion for catalyst K. The plots show that the C3 and C4 olefin yields are strongly affected by both conversion and feed nitrogen content results in lower C3 and C4 olefin yields.

Fig. 7 shows gasoline and coke yields as a function of conversion for the catalysts investigated, using feeds with different nitrogen content. The plots show that the relationships established for catalyst K apply to all the catalysts investigated.

At constant conversion, an increase in feed nitrogen results in a decrease in gasoline yield and an increase in coke make. In spite of the significant differences in composition and properties of the catalysts tested, at constant feed nitrogen content, the gasoline-vs.-conversion plot shows a linear relationship. The nitrogen content of the liquid products, as a function of conversion for different catalysts, is shown in Fig. 8. At constant feed-nitrogen content, the nitrogen content of the liquid products decreases with increasing conversion.

Furthermore, at constant conversion, an increase in feed-nitrogen content results in an increase of nitrogen content in the liquid products. Mass spectroscopic analysis has shown that most of the nitrogen compounds are in the DO and LCO fractions.

Nitrogen analysis of spent catalysts has shown that the amount of nitrogen on the catalyst is considerably smaller than in the corresponding liquid product. For example, catalyst K tested with feed F-1 gave, at 74 vol % conversion, a liquid product containing 1,100 ppm nitrogen, while the corresponding catalyst contained only 190 ppm nitrogen. At constant feed-nitrogen content, an increase in conversion results in an increase in nitrogen content of the catalyst.

Fig. 9 is a plot of percent nitrogen recovered in the liquid product vs. conversion for the catalysts investigated. Percent nitrogen recovered is defined as the ratio of percent nitrogen in liquid product and percent nitrogen in feed, times 100.

The plot shows a general decline in percent nitrogen recovered with increasing conversion, regardless of catalyst tested. All catalysts line up on the same curve, regardless of feed-nitrogen content. The lower concentration of nitrogen in the liquid product obtained with the more active catalysts, and the corresponding higher concentration of nitrogen on the spent catalysts, indicates stronger conversion of feedstock-nitrogen compounds into coke components.

Conclusions

The conclusions drawn from the data presented can be summarized as follows:

- At constant steam-deactivation temperature, an increase in feed-nitrogen content results in a decrease of conversion and gasoline yields and an increase in LCO and DO yields (Figs. 2 and 3).

- At constant conversion, an increase in feed-nitrogen content results in a decrease in gasoline and DO yields and an increase in LCO, coke, and hydrogen yields (Figs. 4, 5, and 7 and Table 4).

- At constant feed-nitrogen content, the nitrogen content of the liquid products decreases with increasing conversion (Fig. 8).

- The percent nitrogen recovered in the liquid products decreases with increasing conversion, regardless of feed nitrogen content (Fig. 9).

- Most of the nitrogen compounds in the liquid products are in the LCO and DO fractions.

- Catalysts P and L have the highest activity among the catalysts tested, regardless of steaming temperature of feed-nitrogen content (Fig. 2).

- Catalysts P and L gave the highest gasoline and coke yields, as well as the lowest LCO and DO yields.

- At constant coke make (5.6 wt %), catalyst P gave the highest conversion and gasoline yield.

- High zeolite content and matrix type play a role in enhancing the nitrogen resistance of FCC catalysts.

References

1, Magee, J.S., Ritter, R.E. and Rheaume, L., Hydrocarbon Processing, 1979, Vol. 58, No. 9,1979, p. 123

2. Rider R.E., and Rheaume, L., Welsh, W.A., and Magee, J.S., "A look at new FCC catalysts for resid", OGJ, Julv 6, 1981, p. 103.

3. Mills, G A., Boedecker, E.R., and Oblad, A.G., J. Am. Chem. Soc., 1950, 72, p. 1554.

4. Voge H. H., Good G.M. and Greensfelder B.S., Proceedings of the 3rd World Petrochemical Congress, Vol. 4, 1951, p. 124.

5. Villand, C.K., ACS Petroleum Division, Preprint, Vol. 2, No. 4, 1957, A-41.

6. Raith, J., and Lanik, A., Erdoel-Erdgas Zeitschrift, 98 Jg, May 1982, p. 169.

7. Snyder, L.R., Preprint Symposium Heterocomp. in Petroleum ACS meeting, Houston, Feb. 22-27 1979, C-44.

8. Holmes S.A., and Thompson, L.F., Oil Shale Symposium Proceedings, Vol. 14, 1981, p. 235.

9. Rollman, L.D., Journal of Catalysis, Vol. 46, 1971, p.243

10. Katzer J.R, and Silvasubramanian, R., Cat. Rev. Sci. Eng., Vol. 20, 1979, p. 155.

11. Environmental Protection Agency, 76,090, June 4, 1983

12. U.S. Patent No. 4,419,219, Dec. 6, 1983.

13. U.S. Patent No. 4,090,951, May 23, 1978.

14. U.S. Patent No. 4,518, 489, May 21, 1985.

15. U.S. Patent No. 4,287,051, Sept. 1, 1981.

16. U.S. Patent No. 4,081,351, Mar. 28, 1978.

17. Department of Energy, 3244-376, June 9, 1983.

18. Jacobs, P.A., and Heylen, C.F., Journal of Catalysis, Vol. 34, 1974, p. 267.

19. Jacobs, P.A., Leeman, H.E. and Uytterhoeven J.B., Journal of Catalysis, Vol. 33, 1974 p 17.

20. Jacob, S.M., Gross, B., Voltz, S.E., and Weekman, V.W., AlChE Journal, Vol. 22, 1976, p 701.

21. Schwab, R.F., and Baron, K., 2nd Katalistiks FCC Symposium, Amsterdam, 1981.

22. Fu, Chia-Min, and Schaffer, A.M., Industrial Engineering Chemistry, Product Research & Development, Vol., 24, 1985, p. 68.