BISMUTH NICKEL PASSIVATION EFFECTIVE IN FCCU

June 4, 1990
Richard S. Heite Mapco Petroleum Inc. Memphis, Tenn. A.R. English Chevron Research & Technology Co. Richmond, Calif. G. Andrew Smith Intercat, Inc. Sea Girt, N.J. Bismuth-based nickel passivation has been effective in Mapco Petroleum Inc.'s fluid catalytic cracking unit (FCCU) at its Memphis, Tenn., refinery for the past 2 years. Mapco switched to the bismuth passivator in 1988 after using antimony as a passivator since the early 1980s,
Richard S. Heite
Mapco Petroleum Inc.
Memphis, Tenn.
A.R. English
Chevron Research & Technology Co.
Richmond, Calif.
G. Andrew Smith
Intercat, Inc.
Sea Girt, N.J.

Bismuth-based nickel passivation has been effective in Mapco Petroleum Inc.'s fluid catalytic cracking unit (FCCU) at its Memphis, Tenn., refinery for the past 2 years. Mapco switched to the bismuth passivator in 1988 after using antimony as a passivator since the early 1980s,

Metals (nickel and vanadium) passivators help reduce the catalyst activity suppression that occurs from contamination of the catalyst with feed-borne metals.

With the switch to bismuth, a hazardous material has been eliminated.

Antimony is on the U.S. Environmental Protection Agency's list of hazardous chemicals.

The bismuth also reduced the deleterious effects of high nickel content in the feed to the FCCU, at a bismuth quantity equal to, or slightly greater than, the amount of antimony previously used. Trouble-free operation of the unit was maintained at a reduced passivation cost.

MAPCO'S OPERATIONS

Mapco Petroleum Inc. conducts refining and marketing operations in Alaska and the Mid-South. The Alaska system consists of a refinery located outside of Fairbanks, capable of producing 36,000 b/d of refined products. Those products are marketed wholesale with retail marketing of heating oil, diesel fuel, and gasoline.

The Mid-South system includes a fully integrated refinery in Memphis, with a crude capacity of 62,000 b/d. Petroleum products are marketed on a wholesale and retail basis, including 300 retail gasoline/convenience store outlets.

The Memphis refinery operates a 30,000 b/d UOP high-efficiency fluid catalytic cracking unit, constructed in 1980. The unit charges a residual feedstock derived from Louisiana, North Sea, and Nigerian crude oils.

During most of the year, the feed rate to the FCCU is limited by both the air and gas compressors on the unit.

THE SWITCH TO BISMUTH

The refinery had used various antimony formulations for metals passivation since the early 1980s when antimony passivation became available. The use of a nickel passivator enabled additional charge to be cracked by reducing the activity-suppressing effects of nickel on the catalyst.

In mid-1988, Mapco evaluated several alternatives to its antimony passivation program. The evaluation was directed at four goals: to maintain or reduce the cost of the passivation program, to eliminate antimony, to maintain or improve passivator effectiveness, and to maintain trouble-free operation.

By mid-1988, Chevron Corp.'s metals passivation process had been commercially evaluated in six operating units.1 Results of those evaluations caused Mapco to select the same technology for its FCCU.

Chevron's passivation technology is based on bismuth. Bismuth and its compounds are not on the EPA's list of hazardous chemicals.2 In Mapco's case, the Chevron process also offered significant cost savings over the antimony program.

The switch from antimony to bismuth was straight forward and went smoothly. Before the switch, Mapco was using a water-based antimony passivator. Based on an evaluation of the feed and products, and on the desired FCCU operating conditions, Intercat CMP-112 was selected as the best bismuth formulation for Mapco's application.

Intercat delivers the additive as needed, usually every 4-6 weeks. The passivator is fed into the FCCU by Intercat's skid-mounted injection system and holding tanks.

After the inventory of the existing antimony was exhausted, the holding tank was thoroughly flushed, all piping was either cleaned or replaced, and a new injection pump was installed. Disposal of the antimony was done under appropriate safety and environmental requirements.

All of the system modifications and preparations were done in accordance with instructions in Chevron's process operating manual.

Mapco opted not to utilize a diluent flush in the injection system to prevent plugging of the injection point because the injection system is located close to the feed lines. To date, there has been no plugging, except when trying to inject a vanadium passivator through the same injection quill (the injection lance or pipe that enters the FCCU).

In most systems, pre-dosing with a large dose of passivator is usually not required, but due to the high catalyst turnover in the Mapco unit and high feed nickel content, an initial higher base load of passivator was applied for several days after the switch.

Following the base loading, the injection rate was reduced to the rate prescribed in the operating manual.

Although the antimony had been very effective in passivating nickel, deposition efficiency, the amount of passivator that deposits on the circulating FCC catalyst, had been only between 36 and 44% of the amount of antimony added to the unit.

After switching to bismuth, overall deposition efficiency for an 11-month average has been better than 60% of the bismuth added to the unit. As expected, the deposition efficiency was only about 50% during the initial high-dosage period. Deposition since then has been about 65%.

EFFECTIVENESS MEASUREMENTS

Nickel passivation is expected to provide the benefits of reduced hydrogen selectivity, reduced coke selectivity, and increased gasoline selectivity by reducing catalyst poisoning by the nickel.3 Indirect benefits, caused by favorable shifts in unit heat balance or unloading of the gas compression/treating system, may also result.

If unit feed, catalyst, and operation remain constant during a passivator change, the relative effectiveness of the passivators can be gauged by comparing hydrogen yield, liquid yields, and coke production. Unfortunately, few commercial units have sufficiently constant feed quality over the time period required to purge the old passivator from the unit. This makes direct comparison of before and after test run data difficult.

Figs. 1 and 2 show feed and catalyst inspections for a period before and a period 1 year after the switch to bismuth passivator in the Mapco FCCU. The variations in feed nickel and vanadium content correspond to variations in the crude mix delivered to the refinery.

Minor changes took place throughout the comparison periods, For instance, a major change occurred between May and December 1988, when the refinery charged a significant amount of Nigerian crude oils. The periods of early 1988 and early 1989 are similar. These changes had a large effect on the FCC operation, liquid product yields, and coke make. Fortunately, the circulating FCC catalyst type was not changed during the period.

HYDROGEN YIELD

Hydrogen yield is fairly independent of changes in feed and operating conditions, so it provides the best measure of relative passivator performance. Fig. 3 compares the daily observed hydrogen yield with both passivators. The data scatter is typical of commercial operations, but both sets of data compare well.

The regression curve fits both data sets. This indicates that both passivators performed similarly.

Fig. 4 shows the total FCC catalyst makeup rate and composition during the comparison periods. Catalyst makeup includes fresh catalyst and equilibrium catalyst added to the unit.

Total catalyst usage increased by about 20% during the period of bismuth use. The amount of equilibrium catalyst, which brings in a small amount of very old nickel, also increased.

Table 1 shows that despite the increase in equilibrium catalyst usage, the catalyst nickel was significantly fresher during the bismuth period than during the antimony period. Fresher nickel has more dehydrogenation activity.4 5

Although we were unable to estimate the amount of additional hydrogen this would produce, it is significant that the bismuth is working as well as the antimony despite the increased nickel freshness.

A comparison of the unit hydrogen yield for each passivator as measured during test runs conducted periodically during the comparison periods is also shown in Fig. 3.

These measurements were slightly more accurate than the daily observed yields because the presence of extraneous gas streams were accounted for.

Equivalent results were seen for each passivator, even with the increased nickel freshness during the bismuth period.

EFFECTS ON OTHER YIELDS

It is more difficult to reach conclusions on the effect of passivation on yields other than hydrogen, because the other yields are affected to a greater extent by feed type and unit operating conditions. However, it is worth investigating these yields to be sure that no unexpected adverse effects are taking place.

Fig. 5 shows daily methane and test run methane yields, respectively, as a function of catalyst nickel level. Methane yield decreases slightly with increasing metals. It appears that methane yield is slightly lower during the bismuth period, but the difference is insignificant considering data scatter and unit changes.

Fig. 6 shows the fuel gas production as a function of catalyst nickel. Only test run data are available. As expected, the results are similar to those seen in Fig. 5.

Figs. 7 and 8 show conversion and gasoline as functions of catalyst nickel. At first glance, it appears that conversion and gasoline may be slightly improved with bismuth. However, the differences are again insignificant considering data scatter and unit changes.

Fig. 9 shows the yield of coke and the coke/feed carbon-residue ratio as a function of catalyst nickel. The two passivators compare very well.

As mentioned before, there was a higher total catalyst makeup rate during the bismuth period. However, the ratio of fresh to equilibrium makeup was lower.

Fig. 10 shows catalyst micro activity test (MAT) activity with respect to nickel content. No difference is shown between the bismuth and antimony periods, indicating that the conclusions drawn from Figs. 5 through 9 are not clouded by a shift in MAT activity. If these yields are plotted against catalyst MAT activity instead of catalyst nickel content, the same trends are seen.

REFERENCES

  1. Ramamoorthy, P., English, A.R., Kennedy, J.V., Jossens, L.W., and Krishna, A.S., "A New Metals Passivator in Fluid Catalytic Cracking," National Petroleum Refiners Association annual meeting, San Antonio, Mar. 20-22, 1988.

  2. Federal Register, Vol. 52, No. 23, Feb. 4, 1987.

  3. Bohmer, R.W., McKay, D.L., and Knopp, K.G., "Metals Passivation Past, Present, and Future," National Petroleum Refiners Association annual meeting, San Francisco, Mar. 19-21, 1989.

  4. Grane, H.R., Connor, J.E., and Mesologites, G.P., "Predict Poisoning Effects of Metals," Petroleum Refiner, Vol. 40, No. 5, May 1961, p. 168.

  5. Cimbalo, R.N., Foster, R.L., and Wachtel, S.J., "Metal Poisoning of Zeolite Cracking Catalysts," American Petroleum Institute 37th midyear refining meeting, New York, May 10, 1972.

Copyright 1990 Oil & Gas Journal. All Rights Reserved.