Chapter 11
Downloaded by UNIV OF CALIFORNIA SANTA BARBARA on September 2, 2015 | http://pubs.acs.org Publication Date: September 12, 1988 | doi: 10.1021/bk-1988-0375.ch011
Effects of Ni and V in Catalysts on Contaminant Coke and Hydrogen Yields 1
Paul F. Schubert and Carol A. Altomare Engelhard Corporation, Edison, NJ 08818 During cracking, low levels (10 m /g surface area were prepared by steaming commercial FCC catalysts u n t i l no z e o l i t e was detected by X-ray d i f f r a c t i o n (100% steam, 1600°F, t y p i c a l l y 4 hours). Non-zeolitic, low matrix surface area « 1 0 m2/g) s i l i c a alumina p a r t i c l e s were prepared to test the e f f e c t s of catalyst composition at constant surface area and s i l i c a to alumina r a t i o (S1O2/AI2O3 = 1.15). Unpromoted silica-alumina p a r t i c l e s were prepared by spray drying kaolin followed by c a l c i n a t i o n for 1 hour at 1800°F. Rare earths were added to these unpromoted p a r t i c l e s by i n c i p i e n t wetness impregnation with mixed rare earth n i t r a t e s to obtain approximately 4.3% rare earth oxides on the f i n a l p a r t i c l e s . Following impregnation the p a r t i c l e s were dried at 250°F for 16 hours, and then calcined for one hour at 1100°F. Microspheres containing magnesium (19.5% MgO) were prepared by adding magnesium compounds to the kaolin slurry prior to spray drying. These were then processed i n the same manner as the unpromoted microspheres. To eliminate the influence of the steam treatment used to destroy the z e o l i t e i n the higher surface area
In Fluid Catalytic Cracking; Occelli, M.; ACS Symposium Series; American Chemical Society: Washington, DC, 1988.
183
Downloaded by UNIV OF CALIFORNIA SANTA BARBARA on September 2, 2015 | http://pubs.acs.org Publication Date: September 12, 1988 | doi: 10.1021/bk-1988-0375.ch011
184
FLUID CATALYTIC CRACKING: R O L E IN M O D E R N REFINING
non-zeolitic p a r t i c l e s , a l l low surface area microspheres were also given a f i n a l steam treatment for 4 hours at 1600°F prior to further t e s t i n g . Catalysts were contaminated with n i c k e l and vanadium according to the method of M i t c h e l l (8), using metal naphthenates. Prior to blending, a l l contaminated materials were steamed (1450°F, 4 hrs, 90% steam, 10% a i r ) to age the metals. The s e l e c t i v i t y e f f e c t s of the metals on the non-zeolitic component were determined by blending impregnated non-zeolitic components with 20% of the steamed, uncontaminated high a c t i v i t y z e o l i t i c component such that the o v e r a l l blend yielded 70% conversion. To show the e f f e c t of having z e o l i t e present i n the contaminated p a r t i c l e s , a REY commercial cracking catalyst with a matrix surface area of ca. 85 m^/g was also contaminated with n i c k e l and vanadium, and steamed (1450°F, 4 hrs, 90% steam, 10% a i r ) to age the metals. I t s s e l e c t i v i t i e s were compared to the non-zeolitic additive having the same surface area and chemical composition blended with s u f f i c i e n t metals-free active cracking component to give the same conversion. C a t a l y t i c evaluations were conducted using microactivity tests (MAT) (4) at 910°F i n i t i a l temperature, 15 WHSV, 6.0 g c a t a l y s t , and a 5.0 c a t - t o - o i l r a t i o . The feedstock was a metalsfree mid-continent gas o i l . Each data point shown i s the average of two MAT runs. Only MAT runs with acceptable mass balance were used (96 to 101%). Additionally, MAT data was normalized to 100% mass balance. Extensive error analysis of conversion, coke, and hydrogen yields indicates the following respective standard deviations: 1.62, 0.29, 0.025. The e f f e c t s of n i c k e l and vanadium on the hydrogen and coke make were calculated by obtaining the difference between the yields obtained with uncontaminated catalysts and that of the contaminated catalyst at the same conversion. Temperature Programmed Reduction. To determine whether impregnated nickel reduces more readily on z e o l i t i c p a r t i c l e s than on non-zeolitic p a r t i c l e s , Temperature Programmed Reduction (TPR) experiments of n i c k e l on z e o l i t i c and non-zeolitic p a r t i c l e s were carried out. In order to emphasize any differences i n the n i c k e l r e d u c i b i l i t y due to the presence of z e o l i t e , a high z e o l i t e containing material and a low matrix surface area, non-zeolitic material were compared. Prior to running the TPR, the samples were impregnated with n i c k e l naphthenates as previously described, and then steamed (1450°F, 4 hours, 90% steam, 10% a i r ) . The f i n a l n i c k e l concentrations on the z e o l i t i c and non-zeolitic samples were 10,860 and 10,100 ppm respectively. Attempts at obtaining TPR r e s u l t s at n i c k e l l e v e l s of 1000 to 4000 ppm as used i n the c a t a l y t i c portions of t h i s study were unsuccessful. Samples tested were pretreated by c a l c i n i n g at 500°C for 30 minutes. In running the TPR, a 10% hydrogen i n argon gas mixture was passed through the samples at a rate of 20 ml/min. The heating rate was 10°C/minute. Hydrogen consumption was measured. Hydrocarbon Adsorption. Hydrocarbon adsorption experiments were carried out to determine the e f f e c t of the z e o l i t e ' s acid s i t e s on
In Fluid Catalytic Cracking; Occelli, M.; ACS Symposium Series; American Chemical Society: Washington, DC, 1988.
Downloaded by UNIV OF CALIFORNIA SANTA BARBARA on September 2, 2015 | http://pubs.acs.org Publication Date: September 12, 1988 | doi: 10.1021/bk-1988-0375.ch011
11. SCHUBERT AND A L T O M A R E
Effects ofNi and V on Catalytic Activity
the a c t i v i t y of n i c k e l . Hexane and 1-hexene adsorption on uncontaminated and n i c k e l contaminated z e o l i t i c and non-zeolitic components were studied at n i c k e l l e v e l s from 1000 to 10,000 ppm. Henry's Law c o e f f i c i e n t s and i n i t i a l heats of adsorption were determined by a gas chromatographic technique employing the test samples as the e f f e c t i v e packing i n four inch gas chromatographic columns. Samples were activated i n the column at 425°C and studied over the range of 150 to 425°C. The Henry's Law constant was taken as an i n d i c a t i o n of the adsorptivity of test molecules towards the n i c k e l contaminated c a t a l y s t p a r t i c l e s when compared to the same quantities for their metals-free equivalents. Results Our i n i t i a l experiments on contaminant s e l e c t i v i t i e s were designed to compare the e f f e c t s of n i c k e l and vanadium. Comparison at equal metals l e v e l s (e.g. 2000 ppm) would allow determination of the e f f e c t s of catalyst parameters on the i n d i v i d u a l metals. However, i n c a t a l y t i c cracking of heavy crudes, equal l e v e l s of metals are not deposited, and the concentration of vanadium on catalyst p a r t i c l e s i s t y p i c a l l y 1.5 to 2 times that of n i c k e l (9). Thus, determination of the r e l a t i v e contributions of these metals to the s e l e c t i v i t y of actual catalysts requires comparing them at r e a l i s t i c , but unequal metals l e v e l s (e.g. 2000 ppm Ni vs 4000 ppm V or 1000 ppm Ni vs 2000 ppm V). I t has generally been quite d i f f i c u l t to make these comparisons due to the propensity of vanadium to migrate from one p a r t i c l e to another, while n i c k e l generally remains on the p a r t i c l e i t i s deposited on (6). Furthermore, z e o l i t e destruction due to vanadium makes comparisons at both equal a c t i v i t y and equal metals l e v e l quite d i f f i c u l t . By blending presteamed components to constant a c t i v i t y (70% conversion), and varying which components were metals impregnated, we were able to overcome some of these d i f f i c u l t i e s . Our current work on non-zeolitic p a r t i c l e s shows that the contaminant coke and hydrogen y i e l d s due to both n i c k e l and vanadium increase with increasing matrix surface area at constant conversion (70%) as shown i n Figure 1. I t i s i n t e r e s t i n g to note that above 25 m^/g, there i s l i t t l e , i f any, e f f e c t of surface area on contaminant yields due to n i c k e l . Surprisingly, on the n o n - z e o l i t i c , r e l a t i v e l y l o w - s i l i c a content p a r t i c l e s i n t h i s study, Ni was consistently less active than vanadium at equal metals l e v e l s over the entire range of surface area tested (5 to 140 m^/g). However, the matrix composition may s i g n i f i c a n t l y a l t e r both the magnitude of the coke and hydrogen yields due to n i c k e l and vanadium, and the contribution of n i c k e l r e l a t i v e to vanadium. At very low surface areas (about 5 m^/g) and constant conversion (70%), the contaminant s e l e c t i v i t i e s are dominated by the matrix composition (Table I ) . Rare earth and magnesiumcontaining microspheres were prepared to examine the e f f e c t s of these metal oxides on catalyst s e l e c t i v i t i e s i n the presence of n i c k e l and vanadium. These oxides were chosen because the l i t e r a t u r e (3,5,10-15) has shown them to be e f f e c t i v e at reducing the deleterious e f f e c t s of vanadium i n cracking c a t a l y s t s .
In Fluid Catalytic Cracking; Occelli, M.; ACS Symposium Series; American Chemical Society: Washington, DC, 1988.
185
FLUID CATALYTIC CRACKING: R O L E IN M O D E R N REFINING
Downloaded by UNIV OF CALIFORNIA SANTA BARBARA on September 2, 2015 | http://pubs.acs.org Publication Date: September 12, 1988 | doi: 10.1021/bk-1988-0375.ch011
186
Figure 1. Contaminant hydrogen (a) and contaminant coke (b) y i e l d s as weight percent of feed from non-zeolitic p a r t i c l e s : 2000 ppm Ni (+); 2000 ppm V (x), at constant conversion (70%).
In Fluid Catalytic Cracking; Occelli, M.; ACS Symposium Series; American Chemical Society: Washington, DC, 1988.
Downloaded by UNIV OF CALIFORNIA SANTA BARBARA on September 2, 2015 | http://pubs.acs.org Publication Date: September 12, 1988 | doi: 10.1021/bk-1988-0375.ch011
11. SCHUBERT AND A L T O M A R E
Effects ofNi and V on Catalytic Activity
Although non-equivalent metal oxide l e v e l s were tested, approximately 5% rare earth oxide and 20% magnesium oxide give e s s e n t i a l l y equivalent l e v e l s of vanadium migration o f f nonz e o l i t i c p a r t i c l e s of t h i s composition (Schubert, P. F.; Altomare, C. Α.; Koermer, G. S., Engelhard Corporation; W i l l i s , W. S.; Suib, S. L., University of Connecticut, manuscript i n preparation). Additional work has shown that increasing the rare earth l e v e l does not substantially improve the contaminant vanadium e f f e c t s (Martins, E., Engelhard Corporation, unpublished data). Furthermore, neither of these metal oxides s i g n i f i c a n t l y a f f e c t the nickel s e l e c t i v i t y . At these low surface areas, nickel contributes very l i t t l e to the contaminant coke and hydrogen y i e l d s , while vanadium shows s i g n i f i c a n t v a r i a b i l i t y i n i t s contribution depending on the matrix. At 2000 ppm n i c k e l , the contaminant coke and hydrogen y i e l d s were equal to or less than 0.4% and 0.06% respectively f o r a l l of the matrices studied. On the unpromoted matrix at 2000 ppm vanadium, contaminant hydrogen y i e l d s were s i g n i f i c a n t l y higher than the n i c k e l y i e l d s , and contaminant coke was nearly f i v e times as high. In contrast, on the magnesium promoted matrix, 2000 ppm vanadium made e s s e n t i a l l y no contaminant hydrogen, and less contaminant coke than the same l e v e l of n i c k e l .
Table I.
Effect of Matrix Composition on the Contaminant S e l e c t i v i t y Yields of Low Surface Area Non-Zeolitic Materials
Si0?-Al90q S i 0 (wt%) 52.3 A 1 0 (wt%) 45.1 Rare Earth Oxides (wt%) 0.0 MgO (wt%) 0.0 Surface Area 5 2
2
3
5% RE0 on Si09-Al90^ 50.1 43.2 4.3 0.0 6
20% MgO i n Si0?-Al9(h 40.6 36.1 0.0 19.5
1000 2000 2000 4000
Contaminant Hydrogen Yields (wt% of feed) ppm Ni fi 0.04 ppm Ni 0.06 0.04 ppm V 0.15 0.07 ppm V 0.24 0.17
1000 2000 2000 4000
ppm Ni ppm Ni ppm V ppm V
Contaminant Coke Yields (wt% of feed) 0.04 0.20 0.40 0.96 0.35 1.25 1.44
8
0.01 0.04 0.00 0.02
0.01 0.39 0.15 0.60
Contaminant s e l e c t i v i t y yields measured at 70% conversion Si09/Al90^ equals 1.15
In Fluid Catalytic Cracking; Occelli, M.; ACS Symposium Series; American Chemical Society: Washington, DC, 1988.
187
FLUID CATALYTIC CRACKING: R O L E IN M O D E R N REFINING
188
Downloaded by UNIV OF CALIFORNIA SANTA BARBARA on September 2, 2015 | http://pubs.acs.org Publication Date: September 12, 1988 | doi: 10.1021/bk-1988-0375.ch011
The data shown i n Table II show interactions between matrix surface area and chemical composition at constant conversion. As expected, the increased matrix surface area of Matrix C caused increased coke and hydrogen y i e l d s compared to the lower surface area Matrix B, at constant s i l i c a content. However, the very low surface area of Matrix A does not compensate for the poorer s e l e c t i v i t y due to i t s lower s i l i c a content r e l a t i v e to Matrices Β and C.
Table I I .
Effect of Matrix Composition and Surface Area on Contaminant S e l e c t i v i t y Yields of Non-Zeolitic Silica-Aluminas
Matrix Comp Si0 /Al9(h 9
A
Surface Area (m /g) 2
V Effects Cont Cont Coke H2 (wt%) (wt%) 1
Ni E f f e c t s Cont Cont Coke H2 (wt%) (wt%)
1
1
1
Mod Si0 =52%
5
0.5
0.08
0.2
0.05
High Si0 =60%
10
0.3
0.04
0.1
0.04
High Si0 =60%
25
0.3
0.07
0.4
0.07
Low Si0 =42%
85
1.1
0.15
0.6
0.10
2
Β
2
C
2
D
2
Contaminant s e l e c t i v i t y e f f e c t per 1000 ppm metal Contaminant s e l e c t i v i t y yields measured at 70% conversion
The e f f e c t of the presence of z e o l i t e s on the r e l a t i v e contaminant s e l e c t i v i t y y i e l d s due to n i c k e l and vanadium was studied using a low s i l i c a content (42% S i 0 ) silica-alumina with about 85 m^/g surface area (see Table I I I ) . Since the contaminated non-zeolitic p a r t i c l e s constituted only 80% of the catalyst blend, the contaminant s e l e c t i v i t y yields for these materials were divided by 0.8 to estimate coke and hydrogen y i e l d s . While t h i s methodology may not give completely accurate r e s u l t s , we believe that the r e l a t i v e comparisons w i l l be reasonable and useful. The data c l e a r l y show that at constant conversion (70%), both contaminant coke and contaminant hydrogen due to n i c k e l are greater on z e o l i t i c than non-zeolitic p a r t i c l e s of r e l a t i v e l y low s i l i c a content. However, at 4000 ppm vanadium, the presence or absence of z e o l i t e does not a f f e c t the contaminant s e l e c t i v i t y y i e l d s . Thus, the a c t i v i t y of n i c k e l , but not vanadium, for dehydrogenation and dehydrocyclization i s enhanced due to the presence of z e o l i t e s i n the same p a r t i c l e . 2
In Fluid Catalytic Cracking; Occelli, M.; ACS Symposium Series; American Chemical Society: Washington, DC, 1988.
11. SCHUBERT AND A L T O M A R E
Table I I I .
Effects of Ni and Von Catalytic Activity 189
E f f e c t of the Presence of Z e o l i t e on Contaminant S e l e c t i v i t y Yields P a r t i c l e Type Non-Zeolitic Zeolitic
Downloaded by UNIV OF CALIFORNIA SANTA BARBARA on September 2, 2015 | http://pubs.acs.org Publication Date: September 12, 1988 | doi: 10.1021/bk-1988-0375.ch011
Nickel E f f e c t (2000 ppm) Contaminant Coke (wt% of feed) Contaminant H (wt % of feed) 2
2.3 0.34
1.2 0.19
3.6 0.52
3.8 0.52
Vanadium E f f e c t (4000 ppm V) Contaminant Coke (wt% of feed) Contaminant H (wt % of feed) 2
Contaminant s e l e c t i v i t y yields at 70% conversion
The TPR comparison (Figure 2) shows that n i c k e l on nonz e o l i t i c p a r t i c l e s reduces at a lower temperature than n i c k e l on z e o l i t i c p a r t i c l e s . The peak at 591.3°C i n the TPR for the z e o l i t i c sample i s believed to be due to the reduction of cerium. Hydrocarbon adsorption experiments show s i g n i f i c a n t differences between the n i c k e l contaminated z e o l i t i c and nonz e o l i t i c p a r t i c l e s at metals l e v e l s comparable to those of the c a t a l y t i c experiments. Neither hexane nor 1-hexene showed any interaction with nickel on the low surface area, non-zeolitic p a r t i c l e s (the unpromoted material of Table I) at temperatures up to 425°C. Additionally, no i n t e r a c t i o n between hexene and the n i c k e l on the z e o l i t i c p a r t i c l e s was observed over the temperature range studied. However, the n i c k e l on the z e o l i t i c component did cause s i g n i f i c a n t retention of hexane at temperatures as low as 200°C with generation of what appeared to be higher molecular weight products. No cracking products were observed. With the uncontaminated z e o l i t i c p a r t i c l e s , hexane retention only occurred at temperatures above 300°C. Thus, the lower temperature retention for the contaminated p a r t i c l e s appears to be due to the presence of n i c k e l . Discussion We have found that the a c t i v i t y of n i c k e l towards dehydrogenation and dehydrocyclization i s increased on p a r t i c l e s which contain z e o l i t e s r e l a t i v e to non-zeolitic p a r t i c l e s of the same matrix composition and surface area. However, the coke and hydrogen making c a p a b i l i t y of vanadium does not depend on the presence or absence of z e o l i t e . The increase i n hydrogen and coke production by n i c k e l when the z e o l i t e i s present i n the matrix suggests that the z e o l i t e increases the metal a c t i v i t y i n catalyzing secondary
In Fluid Catalytic Cracking; Occelli, M.; ACS Symposium Series; American Chemical Society: Washington, DC, 1988.
FLUID CATALYTIC CRACKING: R O L E IN M O D E R N REFINING
190
Downloaded by UNIV OF CALIFORNIA SANTA BARBARA on September 2, 2015 | http://pubs.acs.org Publication Date: September 12, 1988 | doi: 10.1021/bk-1988-0375.ch011
0.50 Ï
TEMPERATURE
(C)
Figure 2. Temperature Programmed Reduction of Ni contaminated catalyst components: a) non-zeolitic p a r t i c l e s with 10,100 ppm Ni; b) z e o l i t i c p a r t i c l e s with 10,860 ppm N i . These materials were impregnated using nickel naphthenate and then steamed (1450°F, 4 hrs, 90% steam, 10% a i r ) prior to running the TPR. The Ni on the non-zeolitic p a r t i c l e s reduced at a lower temperature than that on the z e o l i t i c p a r t i c l e s .
In Fluid Catalytic Cracking; Occelli, M.; ACS Symposium Series; American Chemical Society: Washington, DC, 1988.
Downloaded by UNIV OF CALIFORNIA SANTA BARBARA on September 2, 2015 | http://pubs.acs.org Publication Date: September 12, 1988 | doi: 10.1021/bk-1988-0375.ch011
11. SCHUBERT AND A L T O M A R E
Effects of Ni and Von Catalytic Activity
cracking reactions. Several possible explanations f o r the source of t h i s a c t i v i t y enhancement for n i c k e l are apparent. F i r s t , the z e o l i t e could represent a physical b a r r i e r to n i c k e l migration, thus enhancing the metal dispersion. A l t e r n a t i v e l y , the z e o l i t e could possibly f a c i l i t a t e reduction of n i c k e l oxides to the more active n i c k e l metal. This could greatly increase contaminant coke and hydrogen y i e l d s due to the increased contact time between the hydrocarbons i n the feed and the reduced n i c k e l . A t h i r d p o s s i b i l i t y i s that the a c t i v i t y enhancement involves the proximity of the metal s i t e s to the z e o l i t e ' s acid s i t e s (16). This might a l t e r the e l e c t r i c f i e l d of the metal or that experienced by the incoming hydrocarbon, leading to enhanced dehydrogenation on the metal s i t e , and c y c l i z a t i o n on the nearby acid s i t e s . Neither of the f i r s t two routes would be expected to increase the a c t i v i t y associated with vanadium, since vanadium i s much more mobile than n i c k e l , and appears to remain as 5+ V
in these systems (Schubert, P. F.; Altomare, C. Α.; Koermer, G. S., Engelhard Corporation; W i l l i s , W. S.; Suib, S. L., University of Connecticut, manuscript i n preparation). While vanadium might be expected to be affected by proximity to the z e o l i t e , t h i s e f f e c t could be masked by the destruction of the z e o l i t e . The second of these hypotheses (more f a c i l e reduction of n i c k e l on z e o l i t i c p a r t i c l e s ) i s contradicted by the results of our TPR experiments. In f a c t , the TPR r e s u l t s on both n i c k e l contaminated z e o l i t i c and non-zeolitic p a r t i c l e s suggest that none of the n i c k e l on these materials i s reduced under normal MAT testing conditions, since the onset temperature of n i c k e l reduction (1100-1150°F) i s considerably higher than the operating temperature of the MAT (910°F). Our t h i r d hypothesis, i . e . , that the a c t i v i t y enhancement involves the proximity of the z e o l i t e ' s acid s i t e s , appears to be consistent with the hydrocarbon adsorption experiments, but may also be due to differences i n the n i c k e l dispersion a r i s i n g from surface area differences between the two types of p a r t i c l e s . C l e a r l y , the adsorption of hexane at lower temperature on the n i c k e l contaminated z e o l i t i c p a r t i c l e s suggests a s i g n i f i c a n t l y altered environment from both the uncontaminated and the nonz e o l i t i c materials. On the r e l a t i v e l y low s i l i c a non-zeolitic p a r t i c l e s studied at length, vanadium had higher coke and hydrogen producing tendencies than nickel at 70% conversion. The higher dehydrogenation/dehydrocyclization a c t i v i t y for vanadium r e l a t i v e to n i c k e l on non-zeolitic p a r t i c l e s for both coke and hydrogen formation was surprising, given the higher a c t i v i t y of n i c k e l reportedly found when present on early amorphous silica-alumina cracking catalysts (17,18). I t was expected that the dehydrogenation and dehydrocyclization a c t i v i t y of both n i c k e l and vanadium would increase with increasing surface area, evidenced by an increase i n coke and hydrogen production. While at very low surface areas poor dispersion of n i c k e l could account for i t s low a c t i v i t y , we expected that, as the surface area was increased and the dispersion of n i c k e l improved, dehydrogenation and dehydrocyclization a c t i v i t y would surpass that of vanadium.
In Fluid Catalytic Cracking; Occelli, M.; ACS Symposium Series; American Chemical Society: Washington, DC, 1988.
191
FLUID CATALYTIC CRACKING: R O L E IN M O D E R N REFINING
Downloaded by UNIV OF CALIFORNIA SANTA BARBARA on September 2, 2015 | http://pubs.acs.org Publication Date: September 12, 1988 | doi: 10.1021/bk-1988-0375.ch011
192
However, even at higher surface areas the nickel showed no increase i n r e l a t i v e a c t i v i t y compared to vanadium. At e s s e n t i a l l y constant surface area and conversion, the chemical composition of the non-zeolitic p a r t i c l e s determines the coke and hydrogen making c a p a b i l i t y of the two metals. By introducing materials into the matrix which interact with the contaminant metals, the n i c k e l and vanadium contributions can be s i g n i f i c a n t l y a l t e r e d . Both rare earth (19) and magnesium compounds (_5) are known to reduce z e o l i t e destruction by immobilizing vanadium. Other work (Schubert, P. F.; Altomare, C. Α.; Koermer, G. S., Engelhard Corporation; W i l l i s , W. S.; Suib, S. L., University of Connecticut, manuscript i n preparation) has shown that at about 4.5% rare earth and 20% magnesium oxide, e s s e n t i a l l y equal l e v e l s of vanadium immobilization are achieved. However, i n spite of t h i s equivalent e f f e c t on mobility, there i s considerable difference i n t h e i r e f f e c t on s e l e c t i v i t y . Thus, the mechanisms a f f e c t i n g vanadium's migration and dehydrogenation and dehydrocyclization a c t i v i t y are not necessarily the same. Rare earth at 4.3% causes a 30-50% reduction i n the contaminant hydrogen y i e l d , while magnesium oxide addition at 19.5% r e s u l t s i n nearly complete quenching of vanadium induced dehydrogenation, and a substantial reduction i n coke production. Further addition of rare earth does not s i g n i f i c a n t l y a l t e r the l e v e l of vanadium migration, nor the s e l e c t i v i t y at constant metals l e v e l s (Martins, E., Engelhard Corporation, unpublished data). These immobilizers were present i n low surface area matrices. Matrix composition can also overcome surface area e f f e c t s . High s i l i c a content silica-aluminas generally produce less contaminant coke and hydrogen than lower s i l i c a content silica-alumina matrices of the same surface area, since s i l i c a - r i c h surfaces favor n i c k e l s i n t e r i n g while alumina-rich surfaces favor n i c k e l dispersion (16). Our r e s u l t s indicate that a high s i l i c a content silica-alumina had as good or better contaminant coke and hydrogen s e l e c t i v i t i e s than a lower s i l i c a content silica-alumina having o n e - f i f t h of i t s matrix surface area. However, these matrices which reduce the detrimental a c t i v i t y of these metals toward coke and hydrogen production may not be as resistant to vanadium attack of the z e o l i t e as those that contain more alumina (14). I t i s , therefore, not necessarily desirable to use high s i l i c a content matrix catalysts when cracking gas o i l s that contain both n i c k e l and vanadium. Conclusions The results of t h i s work suggest that the greatest contaminant metals e f f e c t s are due not only to the most recently deposited metals, but to those recently deposited metals which are present on the most recently added z e o l i t i c p a r t i c l e s ( i . e . , those containing the most z e o l i t e ) . At constant metals aging then, the contaminant s e l e c t i v i t i e s due to n i c k e l and vanadium are i n a large part determined by: 1) the presence or absence of z e o l i t e i n the p a r t i c l e ; 2) the non-zeolitic surface area of the p a r t i c l e ; and 3) the chemical composition of the p a r t i c l e .
In Fluid Catalytic Cracking; Occelli, M.; ACS Symposium Series; American Chemical Society: Washington, DC, 1988.
Downloaded by UNIV OF CALIFORNIA SANTA BARBARA on September 2, 2015 | http://pubs.acs.org Publication Date: September 12, 1988 | doi: 10.1021/bk-1988-0375.ch011
11. SCHUBERT AND A L T O M A R E
Effects of Ni and Von Catalytic Activity
On n o n - z e o l i t i c p a r t i c l e s i n the absence of a vanadium passivator, vanadium (when present at the 0.4 wt% l e v e l ) makes a greater contribution to contaminant coke and hydrogen y i e l d s than n i c k e l at constant surface area and metals loading. Incorporation of a vanadium passivator into the catalyst matrix can greatly a l t e r the s e l e c t i v i t y effects of vanadium, and can e s s e n t i a l l y negate i t s effect on n o n - z e o l i t i c p a r t i c l e s as i n the case of magnesium. Matrix composition can also overcome surface area e f f e c t s . A high s i l i c a matrix of moderate surface area was found to have e s s e n t i a l l y the same contaminant s e l e c t i v i t i e s as a moderate s i l i c a matrix having only very low surface area. The presence of z e o l i t e i n the p a r t i c l e greatly enhances the a c t i v i t y of n i c k e l towards dehydrogenation and dehydrocyclization, and causes i t to become more active than vanadium. Therefore, i t i s advantageous to concentrate the metals on n o n - z e o l i t i c p a r t i c l e s not only to minimize the a c t i v i t y loss r e s u l t i n g from vanadium attack on the z e o l i t e , but also to minimize the s e l e c t i v i t y effects of the contaminant metals. Thus, catalysts consisting of high a c t i v i t y p a r t i c l e s ( i . e . , high z e o l i t e ) mixed with p a r t i c l e s having no z e o l i t e should help i n l i m i t i n g the deleterious effects of the metals. Acknowledgements The authors wish to acknowledge the invaluable assistance of S. M. Kuznicki who conducted the hydrocarbon adsorption experiments, D. R. Anderson who ran and assisted i n the interpretation of the TPR, and H. Furbeck who prepared the many samples tested.
Literature Cited 1. Parks, G. D., et al.; "Surface Studies of the Interaction of Nickel...," Symposium of the Petrol. Div. of the ACS, Houston March, 1980, p. 334. 2. Cimbalo, R. N., et al. Oil and Gas J. 1972, 70, (2), 112. 3. Jaras, S. Appl. Catal. 1982, 2, 207. 4. Brown, S. M.; Durante, V. Α.; Reagan, W. J . ; Speronello, B. K. U.S. Patent 4 493 902, 1985. 5. Occelli, M. L.; Kennedy, J. V. U.S. Patent 4 465 588, 1984. 6. Occelli, M. L.; Swift, Η. E. U.S. Patent 4 466 884, 1984. 7. Haden, W. L.; Dzierzanowski, F. J. U.S. Patent 3 657 154, 1972. 8. Mitchell, B. R. Ind. Eng. Chem, Prod. Res. Dev. 1980, 19, 209-213. 9. Speronello, Β. K., Reagan, W. J. Oil Gas J. 1984, 139. 10. Ritter, R. E.; Rheume, L.; Welsh, W. Α.; Magee, J. S. Oil Gas J. 1981, 103. 11. Wormsbecher, R. F.; Peters, A. W.; Maselli, J. M. J. Catal. 1986, 100, 130-137. 12. Pompe, R.; Jaras, S.; Vannerberg, N. Appl. Catal. 1984, 13, 171-179. 13. Occelli, et al. U.S. Patent 4 466 886, 1984. 14. Occelli, et al. U.S. Patent 4 650 564, 1987.
In Fluid Catalytic Cracking; Occelli, M.; ACS Symposium Series; American Chemical Society: Washington, DC, 1988.
193
194
FLUID CATALYTIC CRACKING: R O L E IN M O D E R N REFINING
Downloaded by UNIV OF CALIFORNIA SANTA BARBARA on September 2, 2015 | http://pubs.acs.org Publication Date: September 12, 1988 | doi: 10.1021/bk-1988-0375.ch011
15. Occelli, et al. U.S. Patent 4 615 996, 1986. 16. Occelli, M. L.; et al. J. Catal. 1985, 96, 363. 17. Rothrock, J. J . ; Birkhimer, E. R.; Leum, L. N. I&EC 1957, 49 (2), 272-6. 18. Grane, H. R.; Proc. Am. Petrol. Inst, Sec. III, 1961, 41, 241-6. 19. Horvath, Ε. B.; Martins, E.; Tiethof, J. Α.; Durante, V. Α.; and Speronello, Β. K. Eur. Pat. Appl. EP 189267 A2, 1986. RECEIVED June 22, 1988
In Fluid Catalytic Cracking; Occelli, M.; ACS Symposium Series; American Chemical Society: Washington, DC, 1988.