A. D. ReichleBaton Rouge, La.
As we approach the fiftieth anniversary of the start-up of the first commercial fluid catalytic cracker, it seems appropriate that we examine some of the background and details of that auspicious event and how it came about.
Not only was this the birth of what rapidly became the world's major refining process, but also the beginning of fluidization technology that reached outside the oil industry. The advantages of a fluidized bed, with or without circulation, over fixed-bed, gas/solid contacting were quickly recognized within the chemical industry. And practical applications, particularly as a chemical reactor, were soon developed in diverse areas other than catalytic cracking.
By 1956, over 500 U.S. patent applications for fluidized bed processes had been granted. By 1967, fluidized reactors were in commercial use for naphtha hydroforming, fluid coking, coal carbonization, coal coking, chemicals production, iron ore reduction, roasting of sulfide ores, limestone calcination, uranium processing, and other processes.
Fluidization continues today as a major component of chemical engineering technology, with a steady stream of improved understanding of this complex operation, new developments, and commercial operations.
BEGINNINGS
In October 1938, four companies-Standard Oil of New Jersey, Standard Oil of Indiana, M.W. Kellogg Co., and I.G. Farben (the former German chemical cartel) organized a group called the Catalytic Research Associates (CRA) to develop a cracking process that would not infringe on the Houdry patents related to fixed-bed cracking.
Anglo-Iranian Oil Co. Ltd., Royal Dutch-Shell Co., The Texas Co., and Universal Oil Products Co. (UOP) joined the group. I.G. Farben was dropped from the group in April 1940.
A wide variety of cracking schemes was under investigation by the CRA members until mid-1940. At that time a fluidized solids transfer system involving a standpipe had been successfully tested on Jersey's 100 b/d pilot plant in Baton Rouge, La., and was being incorporated into the commercial designs. Jersey Standard later changed its name to Exxon Corp.)
The other CRA members were aware of this advance and now redirected their research away from using catalyst pellets and towards using powdered catalyst systems.
However, Standard Oil Development Co. (Jersey), with its large research and development staff, wide range of pilot plant facilities, and momentum, continued its dominant role in the development of a continuous powdered catalyst cracking system, with important contributions from the CRA members, especially M.W. Kellogg Co.
A critical date in the commercialization of fluid catalytic cracking was May 9, 1940 (about 2 years before the first commercial unit start-up), when Jersey Standard's board of directors authorized money for a first "powdered catalyst" cracking plant.
This event, hastened by the critical need for butadiene, aviation gasoline, and other war materials, marked a turning point in Jersey's long history of research and development on catalytic cracking. But there were still several new twists and turns to be taken before commercial fluid cat cracking was to be a reality.
The history of FCC research, development, and early process innovation has been well documented. 1-13
By May 1940, Jersey had been activity involved in cat cracking research and development studies since 1936 and had progressed from fixed beds to a pipe coil reactor and regenerator using powdered catalyst in a 100 b/d pilot plant, PECLA-Powdered experimental catalyst, Louisiana (Fig. 1). When the board decided to proceed, 53 days of operation with such a "snake" reactor (450 ft of 4-in. pipe) and a screw pump for moving catalyst had been logged since December 1939.
The system was operable, and the yields and product qualities were as expected, but circulation of the catalyst with a mechanical screw pump was unsatisfactory due to agglomeration and activity loss. However, the broad program on "powdered catalyst" phenomena, including circulation, that Jersey had been carrying out since 1936 in its own laboratories and with others (notably researchers at the Massachusetts Institute of Technology and M.W. Kellogg Co.) provided the answer.
Building on this background, an important experiment was carried out on the 100 b/d unit in late April 1940-namely, a test on one of the tower strippers and "standpipes" in which an air injector was used to circulate the catalyst and overcome the pressure drop in the system.
Thus, the board's decision to proceed with construction of a commercial unit was made in an environment of unsatisfactory circulation on the 100 b/d "demonstration" plant, but of encouraging results with an air injection/standpipe system to replace the mechanical screw pump for moving the powdered catalyst around the system.
Construction of a 120-ft standpipe to test this concept further began immediately and was completed on May 24. By June 6, a pressure of 20 psi had been developed at the bottom of the hopper/standpipe with a flowing catalyst density of 26 lb/cu ft and the "catalyst flowed like water from the bottom valve under a head of 25 psi."
Conversion of the 100 b/d pilot plant to a completely new system utilizing standpipes, catalyst star valves, and hindered settling reactor/regenerators (upflow of both catalyst and oil) proceeded rapidly.
The pilot plant was shut down on June 19 for the turnaround. A new design was developed by Kellogg by June 22. Circulation with a Super Filtrol clay catalyst was started on July 17 and was under satisfactory control by Aug. 3 after replacing the star valves with slide valves.
Oil was introduced into the new system the next day and the unit operated almost continuously from Aug. 13 through June 7, 1941, providing process and engineering data for the ongoing design of the first commercial FCC unit, PCLA No. 1 (powdered catalyst, Louisiana No. 1). At the beginning of this operation in August 1940 it was clear that Jersey had demonstrated a successful commercial catalytic cracking process basis using a powdered catalyst.
Design and construction of PCLA No. 1 continued with several unanticipated changes. A preliminary design based on pipe reactors was under way in March 1940, but this was soon switched to the new standpipe/hindered settling system, Model 1, in which both catalyst and gases flowed up through the reactor or regenerator and out the top to a system of cyclones for separation of the catalyst from the products.
Process and engineering results from the 100 b/d unit and other pilot plant and laboratory units were immediately incorporated into the ongoing design. Construction of PCLA No. 1 was started on Sept. 16, 1940, and was completed on May 1, 1942, just 19 months later.
By mid-1941, it was apparent that a bottoms draw-off reactor/regenerator in which the catalyst exited from the bottom of the vessel while the cracked products/gases flowed out the top would have significant advantages over the upflow reactors. Operation of the 100 b/d unit on this new system, Model II, was initiated on Sept. 3, 1941, but the design of PCLA No.1 remained with the hindered settling-type reactor and regenerator.
Simultaneously with the design of PCLA No. 1 for Baton Rouge, designs were being made for similar units at Bayway, N.J. (Standard Oil of New Jersey) and at Baytown, Tex. (Humble Oil & Refining Co.).
These three were to be the only units constructed with flow of both catalyst and gases up through the reactor/regenerator (Model I). All other FCC units of that period employed a dense bed with bottom withdrawal of the catalyst (Model II).
Fig. 2 illustrates the remarkable simultaneous operation of the 100 b/d pilot plant, gathering development and process data, along with the design, engineering, and construction of both the Model I and Model II units.
PCLA NO. 1
PCLA No. 1 was a gigantic unit for those times-19 stories tall, 6,000 tons of steel, 85 miles of pipe, 3,500 yards of concrete, 209 instruments, and 63 electric motors. Fig. 3 shows PCLA No. 1 as it was built in 1942.
The fluid catalyst cracking section of the PCLA No. 1 unit consisted of an upflow reactor and regenerator, cyclone separators and hoppers for both regenerated and spent catalyst, a Cottrell precipitator, and fresh catalyst hoppers. A flow plan of the catalyst section is shown in Fig. 4.
The cold oil feed (typically 28 API South Louisiana reduced crude), after being preheated in various exchangers, including catalyst coolers and a waste gas cooler, was passed with steam through the coils of the vaporizer furnace. It was then flashed in the vaporizer tower, where about 85% was taken overhead.
The overhead vapors and steam were passed through the superheater furnace and injected at 800 F. and 17 psi into a stream of catalyst at 1,050 F. from the regenerated catalyst standpipe at an approximate catalyst-to-oil weight ratio of 4. Cracking took place at 900-925 F. in the reactor, a cylindrical vessel, 15 ft in diameter and 28 ft high.
Catalyst holdup in the reactor was about 19 tons at a density of 6.5 lb/cu ft. The cracked vapors, from which the catalyst had been removed in the cyclones, was then passed to the fractionator for separation into finished products.
About 99.9% of the catalyst was separated from the cracked gas oil vapors at about 10 psig by means of three cyclone separators in series, which dropped the catalyst through aerated dip legs into the spent catalyst hopper. The last traces of catalyst in the products were scrubbed out by condensing a heavy cut in the bottom of the fractionator and pumping this slurry to either the inlet or outlet of the reactor.
Final traces of cracked oil were removed from the catalyst by stripping with steam in the primary cyclone dip leg and the spent catalyst hopper and standpipe. A pressure differential of about 27 psi was built up in the spent catalyst standpipe by the static head of aerated catalyst in the hopper and 45 ft of standpipe.
Flow of spent catalyst to the injector was regulated by a slide valve. At the injector, the spent catalyst was picked up by a stream of air and carried to the regenerator.
In the regenerator, a cylindrical vessel 19.5 ft in diameter and 37 ft high, the catalyst density was about 9 lb/cu ft, and the holdup was about 68 tons. Regeneration was conducted at about 1,050 F., which was controlled by recycling catalyst from the regenerated catalyst hopper through three catalyst coolers (fresh feed preheaters) in parallel back to the regenerator.
A similar recycling system which bypassed the coolers was used to control catalyst density in the regenerator. The recycling rates were controlled by slide valves, and recycling was accomplished by air injection.
Regenerated catalyst and gas flowed out the top of the regenerator to three cyclone separators connected in series, where, at about 1 psi pressure, approximately 99.7% of the catalyst was separated from the flue gases and deposited in the regenerated catalyst hopper.
The stack gas was then passed through a waste-gas fresh feed heat exchanger to a Cottrell precipitator of 42,000 cfm capacity, where 96-98% of the fine catalyst not separated by the cyclones was recovered and returned by air injection to the inlet of the tertiary regenerated catalyst cyclone.
Recycled catalyst and catalyst to the reactor were drawn from the regenerated catalyst hopper through standpipes in which slide valves also controlled the flow rates. A pressure of about 22 psi was developed in the 88 ft of recycle standpipe at a density of about 28 lb/cu ft.
In the 113 ft of regenerated catalyst standpipe, a pressure of about 31 psi was obtained because of less aeration (higher density) and the greater standpipe height.
The Model I units were built at an on site cost of $2.28 million (1939 dollars) or for $179/bbl of charge. Total cost for PCLA No. 1 was $3.8 million.
This does not, of course, reflect the 5 million man-hours of research, development, and engineering that went into creating the process. But it does include about 125,000 man-hours of engineering for the specific unit. 14 In terms of 1991 dollars, a unit of this size would cost approximately $37 million, or $2,882/bbl of charge.
Public announcements of new or critical technology were limited in those tense dan,s preceding World War II. On Feb. 11, 1941, Standard Oil Co. (New Jersey) and Standard Oil Development Co. announced the forthcoming operation of the three above-mentioned plants using a new "continuous catalytic cracking process." 15
Few details were given, but it was disclosed that it used a powdered catalyst, that the gasoline yield was about 50 vol %, and that the gasoline was ideally suited for motor gasoline blending with its research and motor clear octane numbers of 92-96 and 78-82, respectively. 16
Also announced were licensing arrangements with M.W. Kellogg Co. and Universal Oil Products Co.
START-UP OF NO. 1
During the final stages of construction of the cracking section, the oil system was started up on Apr. 22, 1942, to test the equipment and to familiarize the operators with its operation.
These shake-down operations continued for the next month as the construction of the unit was completed and minor equipment deficiencies were located and corrected. A brief log of these activities:
- Apr. 22-May 1: Hot oil through vaporizer and fractionator towers. Auxiliary heater test.
- May 1: Mechanical completion.
- May 2-6: Pressure testing, heat up of catalyst sections with air/steam. Circulation through exchangers and accumulators. Oil to bottom of fractionator, feed preheaters, and steam-generating system.
- May 6-9: Catalyst addition to regenerator. Catalyst recycle through by pass and recycle lines.
- May 11: Torch oil to regenerator. Auxiliary heater off.
- May 14: Regenerator at 950 F.; reactor at 825 F. Catalyst to the reactor. Burn out of rectifier tube on Cottrell precipitator. Plant shut down for repair of Cottrell and minor repairs (wear plates, aeration taps, orifices, brick work).
- May 20-23: Regular heat up procedure started. Check of Cottrell insulators. Catalyst addition to regenerator and recycle. Torch oil added.
- May 24: Catalyst introduced to the reactor at 1:00 a.in. and complete circulation successfully effected. Torch oil line plugged, line cleared, and torch oil reintroduced with reactor at 500 F.
- May 25, 2:25 a.m.: With the unit filled to the desired catalyst levels, the regenerator temperature at 1,050 F., and the reactor at 925 F., reduced mixed crude was charged to the vaporizer coils at a rate of 7,500 b/d.
After the upper and lower recirculation streams of the vaporizer tower had smoothed out, the feed rate was increased over 24 hr to 12,000 b/d to the reactor.
At this time, temperatures of 915-925 F. in the reactor and 1,040-1,050 F. in the regenerator were being held with a catalyst-to-oil weight ratio of 4 and a W/Hr/W in the reactor of 4.5. Torch oil addition was continued until May 28 when carbon deposition on the catalyst in the reactor was sufficient to maintain the regenerator temperature.
- May 26-June 3: Feed rate to reactor gradually increased to 16,600 b/d (128% of design), limited by availability of reduced crude to the unit and Cottrell precipitator gas velocities.
- June 5: After 12 days of successful operation, the unit was voluntarily shut down for a thorough inspection and to make modifications necessary to furnish information for the coming conversion of the unit to liquid feed injection operation for aviation gasoline production with synthetic catalyst.
In keeping with the times, only a brief announcement of this major accomplishment and advance in petroleum refining was made 3 days later. Details were missing but it was noted that a 10-day run at 120% of design capacity had been achieved and "...that the first large-scale units of such intricate design and revolutionary principle were usually subject to 'children's diseases,' but in this case only the mildest forms occurred and have already been remedied." 17
It was also reported that 30 similar units (6 for Standard Oil Co. [New Jersey]) were being built at a cost of over $100 million and that these fluid cat crackers "...might be operated by a new technique which would enable production of the anticipated volumes of 100-octane aviation gasoline and at the same time turn out substantial volumes of the raw material required for the manufacture of synthetic rubber."
This referred to a prior report released in late May by Jersey Standard disclosing the simultaneous operation of the "fluid catalyst" process (maximum production of butenes) along with sulfuric acid alkylation and butene dehydrogenation for increased yields of either 100-octane gasoline or synthetic rubber raw materials."
In the June 25 issue of Oil & Gas Journal, just 3 weeks after the successful completion of Run 1 in PCLA No. 1, Kellogg advertised as a licensing and construction agent for a "Catalytic Cracking Continuous Process."
After a short turnaround of Only 10 days, PCLA No. 1 was brought back on stream for Run 2 at essentially the same conditions. A maximum feed rate of 18,500 b/d was reached in this 43-day run at a cracking temperature of 900-930 F.
Following a turnaround of 3 weeks to make modifications for liquid feed injection, Run 3 was initiated for avgas production with the DA-1 synthetic catalyst and a light (36 API) paraffinic gas oil.
This run extended for 154 days, well past the target of 4 months, and was terminated by failure of a slide valve stem. Table 1 reports early run length histories and catalysts losses.
FIRST PRODUCTS
From the first balances made on PCLA No. 1, the yields and product properties were consistent with expectations based on the 100 b/d pilot plant operations. Table 2 shows operations and product data from a typical 24-hr balance period on May 31.
Reactor feed was a 31.3 API overhead fraction (5%/95% Ca, 510/755 F.) from a South Louisiana reduced crude, excluding recycle of a 17 API slurry oil stream at a rate of about 4% on feed. At a reactor temperature of 910 F., W/Hr/W of 5.0, and a cat/oil ratio of 3.5, a conversion to 400 F. of 53.5 vol % was obtained with Super Filtrol natural clay catalyst.
The regenerator operated at 1,052 F., reducing carbon content from 1.64% (spent) to 0.50% (regenerated). Excess oxygen was 4.4%.
Dry gas (C3-) yield was 6.7 wt % and excess C4 yield was 8.9 vol %. The 10-lb Rvp, 400 F.-E.P. naphtha yield was 43.6 vol %; research and motor clear octanes were 93 and 79, respectively; the yield of 620 F.-E.P. heating oil at an API gravity of 30 was 39.9 vol %.
In addition, there was a 6.6 vol % production of a 28 API cycle gas oil. Carbon make was 3.8 wt %.
By late August the unit had made two runs and had been converted to avgas production. Typical yields for this type of operation with a synthetic DA-1 catalyst and a light Coastal gas oil feed (31 API, 5%/95% Ca 419/614 F.) are shown in Table 3 for comparison.
At these more severe conditions of 975 F., 2.6 W/Hr/W, and 10.5 cat/oil ratio, conversion of the feed was 65 vol % with dry gas and carbon yields of 11.0 and 4.7 wt %, respectively. Yield of the 7 lb Rvp, 325 F.-E.P. avgas was 25.7 vol % along with 14 vol % heavy (430 F. -E.P.) naphtha and 35 Vol % catalytic gas oil. This type of operation was continued on PCLA No. 1 until the end of the war.
EXPLOSION OF CAPACITY
"Capacity exploded... this process was born running." 19
In addition to PCLA No. I at Baton Rouge, two other Model 1 units were being constructed in 1942 and early 1943, one for Standard Oil of New Jersey at Bayway, N.J., and one for its Humble Oil & Refining Co. affiliate at Baytown, Tex.
As soon as these projects on the Model Is were well under way, designs were started for the Model II types, using a downflow, bottom withdrawal for the catalyst from the reactor and regenerator. Five of these newer design units were under way for Jersey before the first Model I unit, PCLA No. 1, went on stream in Baton Rouge.
The urgency of the times and the competence of the development and design engineers at Standard Oil Development Co. and the CRA associates is reflected in the fact that work was proceeding on the design of the Model II units in May 1941, even though operation of the 100 b/d pilot plant to demonstrate this new downflow bottom-withdrawal system was not started until September of that year (Fig. 2).
Throughout the war, the pace of design and construction never slowed. At the end of the war, 34 FCC units (8 for Standard Oil and affiliates) were operating, with a capacity exceeding 500,000 b/d in the U.S. for 20 different companies.
At Baton Rouge, the second FCC unit was a 17,000 b/d Model 11 (PCLA No. 3), starting up on June 25, 1943, slightly over a year after the Model I PCLA No. I came on stream. This unit was followed 3 months later by start-up of a twin Model II unit on Sept. 22, 1943.
The three Model I units (Baton Rouge, Baytown, and Bayway) operated continuously throughout the war on avgas operations, after which they switched to motor gasoline-type operation.
Prior to October 1963, PCLA No. 1 had operated at feed rates as high as 41,000 b/d. At that time it was shutdown and dismantled except for the main fractionator which remains in service today splitting fractionator products.
At Baton Rouge, PCLA No. 3, the world's oldest operating FCC unit, and PCLA No. 2 (Fig. 5) remain in operation today, after some 49 years of continuous service, with a current combined feed rate of 188,000 b/sd, over five times their original design feed rate.
THE FUTURE
Growth of FCC as the major petroleum refining process continues at a rate of about 1.7% per year (1989-92). Total world fresh feed capacity now stands at some 10.6 million b/d-not insignificant for a 50 year old process sometimes mistakenly referred to as a "mature" process.
Improvements in process engineering and catalyst technology continually appear as the FCC process continues to evolve. 20 By 1990 there were about 350 commercial fluid crackers in operation. 21
Growth of FCC's offspring-the "fluidization industry"-is difficult to quantify, but the steady stream of new understandings, research, development, and applications of this basic technology indicates another 50 years of continued progress for fluidization processes.
REFERENCES
1. Murphree, E.V., Brown, C.L., Fischer, H.G.M., Gohr, E.J., and Sweeney, W.J., "Catalytic Cracking of Petroleum," Ind. Eng. Chem, 35, No. 7, pp.768-773, 1943.
2. Murphree, E.V., Fischer, H.G.M., Gohr, E.J., Sweeney, W.]., and Brown, C.L., "Fluid Catalytic Cracking of Premium Fuels," Pet. Ref., 22 No. 11, pp. 357-364, 1943.
3. Carlsmith, L.E., and Johnson, F.B., "Pilot Plant Development of Fluid Cat Cracking," Ind. Eng. Chem., 37, No. 5, pp. 451-455, 1945.
4. Gohr, E.J., "Background, History, and Future of Fluidization," in D.F. Othmer (Ed.), "Fluidization," pp. 102-116, Reinhold, N.Y., 1956.
5. Larsen, H.M., Knowlton, E.H., and Popple, C.S., "Fluid Catalytic Cracking, Toluene and Rubber," in "History of Standard Oil Company (N.J.), New Horizons, 1927-1950," pp. 166-169, Harper & Row, N.Y., 1971.
6. Williamson, H.F., Andreano, R.L., Daum, A.R., Klose, G.C., Aduddell, R., and Seidenstat, P., "Refining Technology and the Development of Catalytic Cracking," in "The American Petroleum Industry," Chap. 17, pp. 603-626, Northwestern University Press, N.Y., 1959.
7. Jahnig, C.E., Campbell, D.L., and Martin, H.Z., "History of Fluidized Solids Development at Exxon," in J.R. Grace and J.M. Matsen (Eds.), "Fluidization," pp. 3-24, Plenum Press, N.Y., 1980.
8. Jahnig, E.E., Martin, H.Z., and Campbell, D.L., "The Development of Fluid Catalytic Cracking," CHEMTECH, 14, pp. 106-112, 1984.
9. Squires, Arthur M., "The Story of Fluid Cat Cracking: The First Circulating Fluid Bed," in "Proceeding of the First International Conference on Circulating Fluid Beds," pp. 1-19, Nov. 18-20, 1985, University of Nova Scotia, Halifax, Pergamon Press, N.Y., 1986.
10. Wrench, R.E., Wilson, J.W., Logwinuk, A.K., and Kendrick, H.P.S., "Fifty Years of Catalytic Cracking," in M.W. Kellogg Company presentation, Sept. 21, 1987, Qatar, U.A.E.
11. Enos, J.L., "Petroleum Progress and Profits," Chap. 6, The M.I.T. Press, Mass., 1962.
12. Squires, A.M., "Contribution Toward A History of Fluidization," in "Proceedings, Vol. 1, Joint Meeting of Chemical Industry & Engineering Society of China and American Institute of Chemical Engineering," Sept. 19-22, 1982, Beijing, China.
13. Reichle, A.D., "Early Days of Cat Cracking at Exxon," presented at Ketjen Catalyst Symposium, Kurhaus, Scheveningen, The Netherlands, May 29-June 1, 1988.
14. Russell, R.P., "The Genesis of A Giant," Pet. Ref., 23, No. 1, pp. 91-95, 1944.
15. "Powdered Catalyst Used in New Cracking Process," OGJ, 39, No. 4, P. 23, Feb. 13,1941.
16. "Continuous Catalytic Cracking Process Announced," Ref. Nat. Gas. Mfgr., 20, No. 3, pp. 66 (90), March 1941.
17. "First Fluid Catalyst Cracking Unit Starts Operating," OGJ, 41, No. 5, June 11, 1942.
18. O'Donnell, J.P., "New Technique Raises Yield of Refinery War Materials," OGJ, 41, No. 3, pp. 34-36, May 28, 1942.
19. "Catalytic cracking-its advent was dramatic-and it revolutionized an industry," OGJ, 57, F-14, 1959.
20. Bienstock, M.G., Draemel, D.C., Terry, P.H., and Shaw, D.F., "Modernizing FCCU's For Improved Profitability," AIChE Annual Meeting, Chicago, Nov. 11-16, 1990.
21. Geldart, G., "Challenges in Fluidized Bed Technology," AIChE Symposium Series No. 270, 85, p. 111, 1989.
Copyright 1992 Oil & Gas Journal. All Rights Reserved.
