AFPM Q&A - 3: Refiners address FCC operations

During the 2018 American Fuel and Petrochemical Manufacturers Operations & Process Technology Summit (formerly Q&A and Technology Forum), Oct. 1-3, 2018, in Atlanta, Ga., US domestic and international refiners addressed questions about fluid catalytic cracking (FCC) operations selected and answered by industry experts from refining companies and other technology specialists.

The respondents

Alec Klinghoffer, Coffeyville Resources Refining & Marketing LLC
Minaz Makhania, Honeywell UOP LLC
Tiffany Clark, BASF SE
Saeed Alalloush, Saudi Aramco
Michael Federspiel, W.R. Grace & Co.
Raul Romero, Nalco Champion
Michael Talmadge, Johnson Matthey PLC
Ken Bryden, W.R. Grace & Co.
Bob Riley, W.R. Grace & Co.
Matthew Wojtowicz, Honeywell UOP LLC

This annual meeting addresses real problems and issues refiners face in their plants and provides an opportunity for members to sort through potential solutions in a discussion with panelists and other attendees.

This is the third of three installments based on edited responses in the 2018 official answer book. Part 1 in the series (OGJ, Sept. 2, 2019, p.66) highlighted discussion surrounding hydroprocessing operations, while Part 2 (OGJ, Oct. 7, 2019, p.34) addressed gasoline processes.

The only disclaimer for respondents was that they discuss their own experiences, their own views, and the views of their companies. What has worked for them in their plants or refineries might not be applicable to every situation, but it can provide sound guidelines for what would work to address specific issues.

Safety

Describe your procedures for placing and operating the FCC in hot standby-safe park mode. What safety concerns do you consider, and what safeguards should be in place?

Klinghoffer: The first question you should ask when considering what safe-park mode means would be: Is the air blower running or not? If the air blower isn’t running, things you should consider are ensuring all ignition sources are shutdown, bypassed, or isolated (electrostatic precipitators, ESPs; torch oil; direct-fired air heater, DFAH; carbon monoxide (CO), boiler). Verify all hydrocarbon sources to the reactor, riser, and regenerator are closed (fresh feed, any recycles, torch oil, fuel-gas purges, light-cycle oil quench, etc.) and all sources of air aren’t being injected (main air, plant-instrument air, fluffing air, etc.). Consider isolating by blocking in, double block and bleed (DBB), or blind as a stabilizing action. Confirm slide valves are closed and maintaining a positive delta-P (ΔP, 3-5+ psig). Close fuel gas to furnaces unless feed or fractionator-bottoms oil circulation is maintained. Confirm you have steam to air blower(s) discharge, feed riser, feed, and torch oil nozzles. Verify steam purging of reactor to the fractionator is controlling the reactor pressure greater than the fractionator and greater than the regenerator by 2+ psig. Verify steam to main air blower (MAB) discharge is being injected at the prescribed rate. Verify steam to feed nozzles matches design-curve specification. 

If steam isn’t available, then use nitrogen (very large volumes of nitrogen will be required for cooling due to its very low heat capacity). If neither steam nor site nitrogen are available, then source a contract nitrogen vaporizer as soon as practical. Verify the MAB discharge line is clear and that the MAB check-valve has closed. Check status of the wet gas compressor (WGC)—on some units the WGC auto trips on blower shutdown; the reactor pressure must be higher than the fractionator and regenerator to prevent flow reversal. If the unit has a power-recovery turbine, ensure that flue-gas quench system is preventing the hot-gas expander (PRT) inlet temperatures from exceeding limits. The vapor spaces of the regenerator and flue-gas system must be purged before restarting the MAB. Use N2 or steam in enough quantities for three volume changes from the purge point to the stack. Sample gas in regenerator and flue-gas system for combustibles. Ensure steam is dry, preferably superheated. Oil soaked on catalyst will crack to form explosive gases. Additionally, steam passing through catalyst at temperatures above 1000° F. will make hydrogen and CO via the water-gas shift reaction. When purge is completed and gases are verified to be free of combustibles, air can be introduced to the unit and catalyst circulation restarted. This must be done as slowly as possible. Temperatures must be monitored to ensure no rapid rises. If there are rapid temperature rises, you must go back and repeat purge and combustible check.

Makhania: If the air blower is running, confirm feed has been bypassed with isolation valves closed, again DBB or blind as a stabilizing action. Deenergize ESP. ESP should not be reenergized until the flue gas has been verified to contain no combustibles and is below the acceptable CO levels. Supply nitrogen or steam to riser at rates adequate to keep reactor as high-pressure point (reactor > main fractionator and regenerator). Set reactor at the highest pressure in the system. We specify pressure differential because required flow rates will be different under varied conditions. Pressure conditions are constant under all scenarios. Use steam to feed nozzles at design-curve specification, as feed nozzles are very prone to plugging in the posture. Monitor vessel velocities to prevent exceeding velocity and temperature limits. Confirm slide valves are closed and maintaining a positive ΔP (3-5+ psig). Use main fractionator to steam pressure control. Use torch oil to control regenerator temperature. Maintain levels in the main fractionator and gas-recovery unit (GRU). Conduct extensive catalyst-loss monitoring. 

Clark: In a Standby #3 situation, the air blower is running, and catalyst circulation is being maintained. This can be a difficult operation to control in the long term and has major associated risks, including detonation of ESPs. 

Major things to consider in this mode of operation to ensure safety are:

• Deenergizing ESPs until stable operation is reached, and monitoring the flue-gas system upstream of the ESP regularly for hydrocarbon and CO.
• While circulating catalyst, always maintaining isolation between the regenerator and reactor-main fractionator by pressure balance.
• Though most emergency shutdown (ESD) systems will trip feed-isolation valves, always verifying feed has been bypassed and that isolation valves are closed and not leaking.

Other things to verify include adequate oil flow through main-fractionator bottoms system to prevent catalyst buildup. You should also maintain sufficient levels in the main fractionator at all times, with adequate flush to tankage.

Nitrogen or steam to the riser-stripper should be maintained at rates adequate to hold reactor pressure as the highest pressure in the system by at least 2+ psi. You should closely monitor the reactor-regenerator pressure balance, fluidization, and catalyst-circulation stability to prevent oxygen from entering the reactor. The main fractionator should be on steam or nitrogen pressure control. Adequate steam to the feed nozzles is required to prevent plugging the nozzles and should be consistent with design specifications.

You also want to monitor velocities and catalyst levels to prevent excessive catalyst carryover during this mode of operation. Monitor your slurry ash content and regen fines or scrubber-solids content frequently and conduct extensive catalyst-loss monitoring.

When torch oil is used to maintain the regenerator temperature, it should be monitored very closely to prevent exceeding temperature design limitations and major catalyst deactivation. You should adjust catalyst-circulation rates to prevent temperature excursions or excessive thermal cycles in the reactor. The main-fractionator overhead system should be monitored routinely for excessive oxygen buildup and steam-condensate acidity. Also, maintaining levels in the vapor-recovery unit will help tremendously on unit start-up.

Are there any operational parameters that can be manipulated to improve operation of the slurry circuit and minimize fouling? Can you outline the slurry exchanger circuit recommended design practices to minimize fouling, plugging, and erosion? 

Makhania: Fouling in the slurry circuit is commonly caused by coking. Three main variables affecting coking in the slurry circuit are composition, residence time, and temperature. 

From experience, the rate of coke buildup increases greatly when the bottoms temperature is > 700° F. (370° C.). It’s recommended to maintain the bottoms temperature below 680° F. (360° C.) to be on the safe side. Often while processing highly paraffinic feedstock, 660° F. (350° C.) is a good starting point for main-column bottoms (MCB) temperature, and it can be optimized based on monitoring of heat-transfer coefficients of the exchangers in the slurry circuit. Use quench if very low-slurry API is required (e.g., for carbon-black feedstock, CBFS).

Minimize the level in the MCB to reduce residence time. Typically, the minimum circulation rate required to prevent coking is 6 gpm/sq ft (14.67 cu m/hr/sq m) of column cross-sectional area but not less than 150% of the design feed rate for disc and donut trays to ensure the trays stay wetted. UOP provides a hot bypass to help maintain the pumparound rate without affecting main-column heat balance.

More saturated slurries decompose at lower temperatures, so increase reactor temperature and catalyst activity to eliminate coking through this mechanism.

When running heavy feeds with a low-matrix activity catalyst at low conversion, it’s still possible to minimize coking with control of operation in the main column. As an initial guideline, UOP recommends that the light cycle oil (LCO) draw rate be controlled to limit the slurry gravity at a minimum of -2° API or the slurry viscosity at a maximum of 25 cst at 99° C.

Slurry catalyst content usually doesn’t directly cause coking problems. To prevent catalyst plugging or erosion in the exchangers, UOP recommends maintaining a maximum velocity of 8.0 ft/sec (2.44 m/sec) and a minimum velocity of 4.5 ft/sec (1.37m/sec) in the slurry exchanger tubes.

In general, the optimum velocity is 5.7 ft/sec (1.75 m/sec). Straight tube construction is recommended.

Alalloush: There are a couple of operating parameters that could improve the fouling rate in the slurry loop like slurry fluid velocity inside the slurry exchangers. Also, the design of the slurry exchangers can improve the fouling rate. Additives can also be used to minimize fouling.

Federspiel: Slurry exchanger fouling comes in several forms, which can be broken down into either organic or inorganic fouling. Inorganic fouling can be caused by corrosion products, precipitated metals, or catalyst particulates in the slurry circuit. Organic fouling, which is more common, can be caused by coke deposits or asphaltenes that have precipitated from the slurry. 

Understanding and addressing the root causes of the different types of fouling can help minimize their impact on FCC operations. Using the correct metallurgy in the main fractionator and slurry circuit will greatly reduce corrosion. By closely monitoring antimony injection, a refiner can reduce the risk of antimony accumulation in the main fractionator.

By maintaining cyclone physical integrity and operating at proper cyclone-inlet velocities, a refiner can reduce the contribution of catalyst particles to slurry fouling. It’s also worthwhile to pay attention to catalyst and additive attrition index and particle-size distribution, as these can both impact losses to the main fractionator.

Time, temperature, and composition of the slurry all contribute to coke formation, and steps can be taken with each of these parameters to help minimize slurry fouling. Ensuring proximity of slurry exchangers and avoiding unnecessarily long slurry piping runs can reduce the amount of time slurry spends at elevated temperatures. The temperature in the slurry circuit can be reduced using slurry quench. It’s recommended to calculate and monitor the bubble-point temperature of the slurry while using slurry quench as an indication of the slurry composition. Ensuring good distribution of the slurry circuit return to the main fractionator and maintaining a slurry pumparound rate such that the wash trays are always sufficiently wetted will also reduce the chances of coke formation. Finally, undercutting LCO into the slurry product will both reduce the temperature and lead to a directionally lighter slurry composition.

Asphaltene precipitation can occur when the asphaltene concentration increases (which can be due to feed type) or if the solubility of those asphaltenes is reduced. Asphaltenes are more soluble in highly aromatic environments, while the presence of more saturated compounds reduces this solubility and leads to fouling. Loss of conversion due to lower catalyst activity or reactor severity can lead to more saturated compounds in the slurry, so addressing loss of conversion is a solid strategy for reducing slurry fouling.

Grace published a thorough paper on this topic in 2007.¹ In it, the causes of the above types of fouling are discussed in more detail, along with mitigation strategies and design considerations.

Romero: Comparison of current operating process variables on slurry-bottom pumparound systems should be a first step to assess operating gaps and justify and understand any changes. Some parameters like temperature profile through washing, sheds area and bottom systems, bottom residence time, pumparound flowrates, and heat exchanger velocities are key. Control stability of flows and temperature associated with above-mentioned sections mitigate coke-spalling situations leading to a sudden plugging of heat exchanger and pump filters. A thorough antifouling chemical program has demonstrated good success in preventing coking formation and catalyst deposit. This program should be preceded by a specific lab test program to identify the best solution for specific slurry properties. 

Operations

As demand for higher-octane gasoline components increases and lobbying for a 95 RON gasoline standard continues, how are you adjusting your operations to meet market demand? What FCC-specific changes do you make to produce higher-octane gasoline components?

Clark: Refiners are optimizing their gasoline blend components to maximize high-octane components. Alkylate and reformate yields are becoming increasingly important, as is minimizing low-octane component yields. The emergence of tight oil feeds has created an increase in low-octane natural gasoline and LPG saturate production. 

While some operational changes, such as raising the riser-outlet temperatures, may seem like an obvious response to increase gasoline octane, current economics in the US penalize incremental production of dry gas due to its big discount to crude and other product slates; they limit octane gain via higher riser-outlet temperature. Not only this, increased dry gas production limits wet-gas compressor throughput, which is a typical FCC constraint.

On the other hand, base catalyst reformulations that promote production of higher-octane gasoline and more LPG olefins production to increase feed to the alkylation unit have become mainstream.

Opportunistic use of ZSM-5 and other butylene-increasing additives has become an important avenue for FCCs to increase the amount of high-octane gasoline blending components via alkylation. They also increase the octanes of the remaining gasoline as an added benefit.

Talmadge: Many refiners are adding ZSM-5 and C4-selective additives to the FCC catalyst inventory to maximize FCC-derived gasoline octane. Gasoline octane is increased with these zeolite components from the following yield-shifting mechanisms in the FCC: 

• Cracking straight chain C6-C10 molecules, which are low-octane, gasoline-range components. Gasoline octane is increased when these molecules are cracked to smaller molecules and the higher-octane molecules are increased in concentration.
• Isomerization activity, which converts remaining straight-chain, gasoline-range molecules to higher-octane, branched isomers.
• Increased C4 (and C3) olefins and i-C4 resulting from selective cracking and isomerization reactions increases the potential for producing high-octane alkylate.
• Decreasing octane loss from gasoline hydrotreating due to lowering the concentrations of olefins (from cracking to C3-C4) in the untreated FCC gasoline. Substantial percentages of olefins are saturated to lower octane paraffins in the hydrotreating reactor.

Bryden, Riley: Octane is a relative measure of the knocking characteristics of a fuel in an internal combustion engine. Knocking is caused by autoignition of fuel ahead of the flame front. Different hydrocarbon molecules have different resistance to autoignition, related to their role in hydrogen peroxide formation under combustion conditions.² Hence, gasoline octane is governed by the types and relative concentrations of the individual hydrocarbon molecules that comprise the fuel. Lighter molecules have higher octane.³ RON and MON values trend by hydrocarbon type as follows: Aromatics ~ olefins > naphthenes > isoparaffins > paraffins. 

Also, for olefins and isoparaffins, octane increases as the degree of branching increases. To increase gasoline octane, the composition of the molecular types in the stream must be changed. Changes that can be made specific to the FCC to produce higher refinery gasoline octane fall into two main categories: Changes inside the FCC unit that change the composition of the FCC gasoline, and adjustments to FCC operation that improve overall refinery gasoline pool octane.

Changes inside the FCC unit

Gasoline cutpoints. Changing the distillation range of the gasoline from the FCC can influence octane. Butane is part of the light end of gasoline and possesses a high-octane number. Increasing amounts of butane will increase RON. This, however, must be balanced against vapor-pressure considerations. For the heavy end, the effect of increasing gasoline endpoint on octane can vary. For aromatic gasolines, increasing endpoint will generally increase octane as more higher-boiling point aromatic molecules are included in the gasoline. For other gasolines, the effect of endpoint on octane will vary with the feedstock to the unit, the conversion level, and the catalyst. Detailed hydrocarbon analysis of FCC gasoline via gas chromatography and application of gasoline-octane prediction models can be used to simulate how octane will change with gasoline endpoint.^4-5

Feedstock. The feedstock to the FCC will have a major effect on octane. Feedstock paraffins generally crack to form low-octane, gasoline-range paraffins. Feed naphthenes crack to form high-octane, gasoline-range aromatics and olefins. Aromatics with side chains present in the feed generally crack to form high-octane, gasoline-range aromatics. As the feed becomes less paraffinic, octane increases. As a rule of thumb, a 0.2 number decrease in the UOP K factor of the feed will result in a one number increase in RON.6 Similarly, a 0.1 number increase in the ratio of naphthenic to paraffinic carbons (Cn/Cp) in the feed will generally result in a one number RON increase.⁷

Operating variables. Increasing riser outlet temperature will increase RON by increasing the number of olefins in the gasoline. As a rule of thumb, at a base RON of 90, an 18° F. increase in riser temperature will result in a one number increase in RON.8 The octane gains with increasing riser-outlet temperature will diminish as reactor temperature is increased. More precise values can be determined by FCC operators through observations made on their own units.

Increasing conversion will increase octane. As conversion increases, cracked products increase, which means that the number of olefins and aromatics in the gasoline increases. As a rule of thumb, a 10 LV % increase in conversion will result in a one number increase in RON at constant riser-outlet temperature.

Decreasing hydrocarbon partial pressure will increase FCC gasoline octane. Gasoline olefin content increases when the rate of bimolecular hydrogen transfer reactions drops, which happens as hydrocarbon partial pressure drops.

Catalyst, additives. The molecular composition of FCC gasoline is governed by the relative rates of cracking and hydrogen-transfer reactions. Lowering zeolite-unit cell size will lower hydrogen transfer and increase gasoline-range olefins, thus increasing octane. Increasing matrix content of the catalyst will help to crack side chains off aromatic cores and increase octane by increasing gasoline-range aromatics. Dual-zeolite catalysts that incorporate both faujasite and pentasil-type zeolites will lead to increased rates of isomerization and result in higher octane from the greater number of branched hydrocarbons.

ZSM-5 based additives and butylene-selective additives can also be used to increase octane. These additives can increase isomerization reactions. Also, by cracking some gasoline-range olefins to LPG olefins, they concentrate aromatics in the FCC gasoline, resulting in increased octane.

Refinery adjustments

Increasing alkylate production. With a typical RON of 95+, alkylate is one of the highest-octane blend streams in the gasoline-blending pool. For refineries with alkylation capacity, FCC adjustments that increase the amount of LPG olefins used as alkylation feedstock will increase alkylate production and refinery gasoline-pool octane. LPG olefins from the FCC can be increased by adjustments to reactor conditions, base catalyst, and use of ZSM-5 based additives. For units desiring a higher ratio of butylene to propylene in their LPG, butylene-selective additives can be used instead of conventional ZSM-5 type additives. Variables that affect LPG olefin production in the FCC have been covered in detail in previous AFPM Q&A sessions.^9-10

Reducing FCC gasoline hydrotreating severity. Refiners report losses between one and five numbers of octane when FCC gasoline is hydrotreated to remove sulfur. Hydrotreater severity can be lowered when the FCC gasoline contains less sulfur. Lower FCC gasoline sulfur can be achieved through use of gasoline sulfur-reducing catalysts and additives that convert gasoline-range sulfur to hydrogen sulfide. A detailed discussion of preserving octane with gasoline desulfurization technology can be found in an earlier AFPM publication.¹¹

In summary, there are many ways FCC operations can be adjusted to increase octane. Inside the FCC unit, octane can be increased through feedstock selection, choice of operating conditions, tuning of base-catalyst properties, and use of specialty additives. Outside of the FCC, the amount of alkylation feed derived from the FCC can be increased through careful FCC catalyst and additive selection, and octane loss during FCC gasoline hydrotreating can be reduced by lowering FCC gasoline sulfur through use of gasoline sulfur-reducing catalysts and additives.

As always in FCC, changes to influence one variable (octane) will result in changes to other FCC unit yield objectives. Refiners should work closely with their catalyst supplier to understand the options available to increase octane and how to balance these with other yield objectives.

Wojtowicz: If a potential future scenario occurs, where specifications require production of higher-octane gasoline, US refiners have several options to enable production of greater amounts of higher-octane gasoline. Each refinery would require a unique solution, depending on the existing configuration, future target-gasoline specifications, and future target-gasoline production rates. The extent of any modifications would also depend on the existing refinery complexity and process unit capabilities. UOP has studied ways to increase gasoline RON for our customers with different types of refineries, and lower-complexity refineries would likely require major investment to enable higher-octane gasoline production compared with more complex refineries. Production of 95+ RON gasoline is being achieved throughout most of the world. The US is one of the few countries in the world that doesn’t currently consume gasoline with a RON of 95 or higher. 

UOP offers up the following sampling of solutions for improving octane:

Continuous catalytic reformer (CCR). Increase existing CCR severity to produce higher-octane reformate for blending (~101* RON could potentially be achieved based on typical designs). This will lower yield but enable refiners to use latent octane capability in existing units. With the introduction of ethanol (120+ RON) into the US gasoline blend, refiners have dialed back reforming severity and reformate octane. Revamp an existing CCR to achieve ~105* RON, or add a CCR if a reformer doesn’t currently exist in the refinery.

Steam reformer (SR). Increase existing SR severity to produce higher-octane reformate for blending (~95* RON could potentially be achieved based on typical designs). This will lower yield but enable refiners to use latent octane capability in existing units. Revamp an existing SR to achieve ~101* RON.

Light naphtha isomerization. Add an isomerization unit if one does not currently exist. This will increase octane of the light naphtha (~85* RON could be achieved). Add a deisohexanizer to an existing isomerization unit. This will further increase octane of the light naphtha (~88* RON could be achieved). Add a deisohexanizer and a deisopentanizer to an existing isomerization unit to achieve ~91* RON.

FCC. Increasing FCC operating severity (increased conversion) in conjunction with the addition of ZSM-5 additive will increase the light-olefin yield (propylene, butylenes and amylenes), but at the expense of overall FCC naphtha production. These lighter olefins can be sent to an alkylation unit, where the increased alkylate production, along with its inherently higher RON, will help offset the loss of naphtha production from the FCC. The remaining FCC naphtha will also experience an increase in the RON.

As you increase the severity of operation in the FCC unit for higher light-olefin production, more naphtha-range olefins are converted to light olefins, and aromatics are concentrated in the FCC naphtha, thereby increasing the overall RON.

Alkylation. Additional alkylate with ~95 RON can be purchased. Increase alkylation capacity to produce additional alkylate. Increasing FCC operating severity as discussed above increases propylene and butylenes that can be converted to alkylate. The ultimate potential capacity of the alkylation unit is generally limited by the propylene and butylenes that can be produced by the FCC unit. In the US, this could be overcome by leveraging low-cost C4s (from the natural gas fields) and by adding a butane dehydrogenation unit. Increased alkylation capacity is enabled through production of additional butylenes from the low-cost C4s.

*Typical RON representations. These may vary for a specific refiner.   

References

1. Hunt, D., Minyard, B., and Koebel, J., “Understanding and Minimizing FCC Slurry Exchanger Fouling,” Grace Davison Catalagram 101, Spring 2007, p. 30, https://grace.com/catalysts-and-fuels/en-us/catalagram.
2. Westbrook, C.K., “Chemical Kinetics of Hydrocarbon Ignition in Practical Combustion Systems,” Proceedings of the Combustion Institute, Vol. 28, 2000, p. 1563.
3. American Petroleum Institute Research Project 45, “Knocking Characteristics of Pure Hydrocarbons,” ASTM Special Technical Publication No. 225, 1958
4. Cotterman, R.L. and Plumlee, K.W., “Effects of Gasoline Composition on Octane Number,” Proceedings of the Symposium of the Division of Petroleum Chemistry, American Chemical Society Meeting, Miami Beach, Fla., September 1989, p. 165.
5. Haas, A., McElhiney, G., Ginzel, W., and Buchsbaum, A., “Gasoline Quality—The Measurement of Compositions and Calculation of Octanes,” Petrochem/Hydrocarbon Technology, Vol. 43, 1990, p. 21.
6. Magee, J.S., Ritter, R.E., Wallace, D.N., and Blazek, J.J, “How Cat-Cracker Feed Composition Affects Catalyst Octane Performance AM-80-48,” National Petroleum Refiners Association annual meeting, New Orleans, Mar. 23-25, 1980.
7. Andreasson, H.U. and Upson, L.L., “Four Main FCC Factors Affect Octane,” Oil and Gas Journal, Vol. 83, No. 31, Aug. 5, 1985, p. 91.
8. “FCC Operation,” Grace Davison Guide to Fluid Catalytic Cracking, 1993.
9. Question 101, FCC Q&A Session, American Fuel & Petrochemical Manufacturers Q&A and Technology Forum, Denver, Colo., Oct. 6-8, 2014.
10. Question 10, Process Q&A Session, American Fuel & Petrochemical Manufacturers Cat Cracker Seminar, Houston, Aug. 23-24, 2016.
11. Cheng, G., “Preserving Octane for a Tier 3 Gasoline Market CAT-16-23,” American Fuel & Petrochemical Manufacturers Cat Cracker Seminar, Houston, Aug. 23-24, 2016.