MAXIMIZE OLEFINS THROUGH CATALYTIC CRACKING INDMAX FCC (I-FCCSM) PROCESS

Dalip Soni, Philip J Angevine, Rama Rao Lummus Technology Inc., Bloomfield, NJ, USA G Saidulu, D Bhattacharyya, V Krishnan

Indian Oil Corporation Ltd, R&D Centre, Faridabad, India

Overview

The primary purpose of the fluid catalytic cracking (FCC) process historically has been to convert low-valued, heavier petroleum streams into gasoline, alkylation feed (e.g., isobutane and butene) and, to a lesser extent, other distillate products. While other light olefins were always produced, the low volumes did not attract much attention because of the separation costs and modest margins. Production of propylene from FCC units has grown significantly in the last decade and is

now approaching 20-30% of the total propylene produced worldwide1. The price spread between propylene and gasoline and increase in demand for propylene have been the key drivers. Significant improvements have been made in the FCC catalyst formulation and process technology to maximize propylene yield from the FCC units. This relatively new strategy for on purpose production of FCC propylene will help meet approximately 5% of the annual growth rate in worldwide propylene demand1. Early attempts to increase light olefins from FCC were based primarily on process variables,

but poor selectivity of this approach resulted in excess dry gas and coke make. By the 1970s, researchers found that non-Y zeolites could also co-produce light olefins (C2= to C5=), often at the expense of gasoline. Table 1 below shows the chronology of catalyst and additive developments for light olefins enhancement in FCC units.

Table 1. Timeline of Zeolites use for Higher FCC Olefins Production

Year Event

1970 ZSM-5 in FCC invented 1974 USY use in FCC invented 1983 ZSM-5 commercialized in cracking 1986 Zeolite Beta in FCC invented 1990 First on-purpose olefins production

via ZSM-5 1995 IndmaxSM FCC process invented 2003 Indmax process commercialized

This paper discusses the growth of on purpose light olefins production as an FCC

objective and the approaches employed. Specifically, the IndmaxSM FCC (I-FCCSM) process will be addressed, including its benefits, the underlying concepts, the chronology of its development, and its commercialization.

575

Approaches to Higher Olefins Production

Process Variables Some of the FCC process variables employed for olefin production include: (a) higher

severity operation, (b) re-cracking of the cracked material, and (c) partial catalyst recycle. Table 2 summarizes the benefits and drawback of such variables.

Table 2. Benefits and Drawbacks of Process Options

Process Pros Cons

High severity operation Short contact time Long contact time

Low-cost route to increased olefins

Excess dry gas/coke make

Re-cracking of cracked material (naphtha)

Product selectivity optimization

Excess dry gas at the expense of gasoline product

Partial catalyst recycle Increased catalyst-to-oil ratio

Poor selectivity

High severity operation is achieved by high reaction temperature with or without higher

residence time. The general concept for higher severity operation is to overcrack (beyond maximum gasoline mode) and produce increased quantities of light olefins. The major drawback of this approach is the production of excess dry gas and subsequent overloading of the gas plant (compressor and separator). Also, higher severity results in higher coke make2 and higher diene content in gasoline that can lead to gum formation. This whole approach is somewhat of a brute force method with little control over the desired products. Re-cracking of cracked material like naphtha can yield incremental light olefins. This re-

cracking can be done in the same or a separate riser/reactor. The concept of injecting in the same riser/reactor has been used for upgrading visbreaker or coker naphtha in FCC risers. Recycle of FCC naphtha to the same riser reactor has also been implemented; however, dry gas yield increases significantly. In the partial-catalyst-recycle approach, part of the spent catalyst is recycled back to the

reaction section to increase the catalyst-to-oil ratio. The drawback is that the average catalyst activity is low compared to the case when higher catalyst-to-oil ratio is achieved with all regenerated catalyst. Therefore the partial-catalyst-recycle approach results in lower conversion and selectivity to light olefins. Catalyst Variables Since the 1960s, the refining industry has used various X and Y zeolites as its primary

cracking catalyst in FCC3, 4 . Stability and selectivity benefits quickly led to the use of Zeolite Y exclusively. The phase-out of lead caused researchers to seek alternate routes to increased gasoline octane. Several versions of Y zeolites were used, and these are listed in Table 3.

576

Table 3. Benefits and Drawbacks of Catalyst Options

Catalyst Pros Cons

Ultrastable Y (USY) More C3=/C4=; less coke Poor stability RE-USY Improved stability versus

USY Limited olefin potential

ZSM-5 Excellent C3= yields Poor conversion activity; additive dilution effect

Zeolite Beta Enhanced C4= High cost; poor stability Many methods were tried to dealuminate zeolite Y to balance activity, stability, gasoline

yield, and octane. The use of zeolite Y for increased olefins focused on ultrastable Y, achieved mainly by its decreased zeolite unit cell size and decreased framework Al content.5, 6 At constant conversion, USY catalysts yield more C3-C4 olefins, higher-octane gasoline, and less gasoline and coke. One patent7 showed that olefins yield reached a maximum at unit cell size a0 of 24.15-24.20 . Figure 1 shows olefin yield as a function of unit cell size. Unfortunately, USY is less active and stable than its REY counterpart. RE-USY can achieve benefits similar to USY, but with about 35% higher activity than USY.8

Figure 1. Incremental olefin yields for various USY catalysts Currently, the popular approach to olefin production is the use of ZSM-5 additive.9 ZSM-5

was first explored as a stand-alone FCC catalyst, but found to be lacking in conversion activity not surprising for a medium-pore-size zeolite. It was later proved that ZSM-5 enhances gasoline octane when used as an additive to the main FCC catalyst, due to the ability of ZSM-5 to produce more olefins. While early ZSM-5 use focused on octane enhancement, the use of larger quantities of

ZSM-5 for on purpose propylene production started in the 1990s. Refiners have since increased their reliance upon ZSM-5 to the point where it now is used in almost 20% of worldwide FCC capacity. This approach has proven to be a relatively inexpensive route to propylene. ZSM-5 can be used as a separate additive or as an integral part of the FCC catalyst. Each

approach has its benefits and drawbacks. Since the two zeolites (Y and ZSM-5) age at different rates, the separate additive method can adjust better for this effect. However, the

0

1

2

3

4

24.36 24.33 24.25 24.19 24.18 24.15 24.02

Unit Cell Size

Del

ta w

t% o

f C

4-C

5 O

lefi

ns

C C4 5()

0

1

2

3

4

24.36 24.33 24.25 24.19 24.18 24.15 24.02

Unit Cell Size

Del

ta w

t% o

f C

4-C

5 O

lefi

ns

C C4 5()

577

additive can build up in the FCC inventory to the point where this dilution can reduce the overall conversion performance of the catalyst mix. Using the integral, single particle approach, the Y zeolites has to have some extra activity so-called activity giveaway which may not be easily achieved. This approach also does not allow the flexibility to change the percentage of ZSM-5 in the total catalyst inventory in the event of feedstock changes or fluctuations in market product demand. The generally accepted mechanism for ZSM-5s utility is the secondary conversion of

heavier gasoline olefins to lighter (e.g., C3 and C4) olefins as well as isobutane.10, 11, 12 The

gasoline paraffins have been shown to be a hundred-fold less reactive than the olefins.13 Feedstock plays a key role in the abundance of gasoline olefins. For example, highly paraffinic crudes (e.g., Minas, etc.) are the most responsive to ZSM-5, while aromatic crudes (e.g., San Joaquin) are significantly less effective for octane boost and olefin production. The I-FCC, I-RFCC Process The Indmax process, developed by the Indian Oil Corporation (IOCL), is a breakthrough in

FCC technology for converting heavy feeds, including residue, to light olefins14. This process overcomes the drawbacks or limitations of just adding ZSM-5. The Indmax process employs a riser reactor system along with a catalyst stripper and

catalyst regenerator, just like a conventional FCC unit. The process uses a multi-component, multi-functional, proprietary catalyst formulation that is very selective in cracking molecules of different shapes and sizes to yield very high quantities of light olefins. At the optimum formulation, different components work synergistically to provide high catalyst activity, and hence conversion, along with high selectively to light olefins. Figure 2 clearly reflects these attributes of high coke and dry gas selectivity, as the Indmax process produces much less coke and dry gas (at constant conversion) when compared with resid FCC catalyst.

Figure 2. Indmax catalyst performance versus conventional FCC/RFCC route with respect to coke and dry gas selectivity

The Indmax catalyst formulation is unique and very different from what has been used in FCC or RFCC operations. The multi-component catalyst approach provides flexibility to change the catalyst formulation quickly and meet the challenges of fluctuations in feedstock properties or product demand. The catalyst formulation is specific to each operation and depends on the feedstock characteristics and product objectives. Extensive research studies have been conducted for various feedstocks to determine the light olefins yields and selectivity. Table 4

578

summarizes the range of olefins yield and selectivities that can be obtained from different feedstocks and operations.

Table 4. Typical Olefin Yields for the I-FCC Process

Wt% of feed

Wt% in DG/LPG

Ethylene 3-7 40-50 Propylene 12-24 40-50 i-Butylene 4-5 8-12 Other butylene

5-12 20-25

The Indmax catalyst formulation is also highly tolerant of metals and can operate with high vanadium concentration on the equilibrium catalyst. This ability to accept feeds with high metals level is very important for residue operations.

The Indmax process utilizes higher riser reactor temperature (560-600C), higher catalyst-to-oil ratio (12-20), and lower hydrocarbon partial pressure to achieve high conversion and selectivity for light olefins production. Since all the cracking reactions take place in the short-contact-time riser reactor, and with very high catalyst-to-oil ratio and all high activity regenerated catalyst, the selectivity to light olefins is very high. The LPG produced contains up

to 50 wt% propylene and 12 wt% isobutylene. Total olefins in LPG can be as high as 80 wt%. The Indmax process can accept feedstocks ranging from hydrotreated vacuum gas oil to residue. The first commercial unit (Figure 3) to

utilize the Indmax process was commissioned in 2003 at the Guwahati Refinery (Assam, India). With a capacity of 100,000 metric tons per annum, it has been running smoothly since startup, producing products with the desired specifications. The plant performance test run data is presented in Table 5. The unit operates on a heavy feed with CCR of 3.75 wt%. The regenerator operates in complete combustion without a catalyst cooler and

coke of regenerated catalyst (CRC) is less than 0.05 wt%. The unit achieves very high conversion with propylene yield of about 17.2 wt% of total feed. Excellent coke selectivity of the catalyst makes it possible for this unit to be run in complete combustion without a catalyst cooler.

Figure 3. 100,000 Indmax Unit, Guwahati, India

579

The I-FCC unit is designed by Lummus Technology, a CB&I company (Lummus) with basic process information developed by IOCL. Lummus is a Licensor of state-of-the-art FCC and RFCC technologies, while the IOCL R&D Center is a premium research and development

organization in the field of catalyst and process development. This combination provides an in-depth and extensive knowledge base that is used in the design of each I-FCC unit. The reactor regenerator section equipment

and hardware pieces are designed to utilize the maximum potential of the Indmax catalyst and the given feedstock to produce light olefins. The key process features include:

Micro-Jet Feed Injectors: These injectors maximize selective catalytic cracking reactions by thoroughly contacting the feed with the hot regenerated catalyst and rapidly vaporizing it. In the Micro-Jet injector design, the inherent energy of the feed dispersion steam is used to shear the feed into small droplets. Figure 4 is a schematic of the Micro-Jet injector. It consists of a double pipe wherein the oil flows through the center pipe and steam flows through the annulus. The cylindrical flow of oil is turned into thin films as

it flows past a device attached at the end of the center pipe. The thin films of oil are then hit by several jets of dispersion steam, which is flowing in the annulus. The small droplets formed are then quickly ejected into the riser through a specially designed tip that generates several micro jets designed for the optimum exit velocity when entering the riser.

Table 5: Indmax Commercial plant performance data

Parameter Indmax Feed

Feed Composition Mix of Atm. Residue, Delayed Coker Liquid

Products Feed Density (ATB+CHGO) Coker Naphtha

0.9456 0.7164

CCR (ATB+CHGO) 3.75 wt% Yield, wt%(fresh feed basis) Dry gas/ethylene 7.4/3.3 LPG/Propylene 36.3/17.2 Gasoline (C5-200

oC) 34.7 Conversion, wt% ( up to 200 C D86)

86

Gasoline Octane (RON)

99

Steam

Oil Feed

Figure 4 Schematic of Micro-JetTM Injectors

580

Each individual jet is angled such that all the jets collectively form a flat fan spray pattern

for maximum coverage of the riser cross section. This design ensures uniform distribution of feed across the riser while preventing riser erosion and catalyst attrition. The angle of injection is optimized to achieve thorough contact of oil and catalyst. The injection velocity is optimized so that it is high enough to penetrate the catalyst mass yet does not cause catalyst attrition and riser erosion.

Riser Design: The riser residence time and velocity are optimized to maximize catalytic cracking reactions, minimize undesirable thermal cracking reactions, and maximize selectivity to olefins. Riser residence time is optimized through pilot plant tests and kinetic model studies. The riser diameter, and hence velocity, is optimized by tracking the volume expansion as the reaction mixture flows up the riser by means of a computer model. The riser design approaches plug flow conditions with minimum slip, which enhances catalytic cracking reactions and hence selectivity to light olefins. Reaction Termination Device (RTD): Direct-coupled cyclones are used at the end of the riser reactor to quickly and efficiently recover product vapors at the end of the riser and terminate the reactions (Figure 5). Due to the strong cyclonic forces, most of the catalyst is thrown against the cyclone wall immediately after exiting the riser reactor and separates from the product vapors. There is essentially no time for the bimolecular, post-riser, hydrogen-transfer reactions that re-saturate the olefins made in the riser, to take place. Another unique feature of Lummus Technologys RTD design is that the primary/riser cyclone operates at a pressure lower than that in the cyclone containment vessel or reactor. As a result, the recovery of products at the end of the riser is almost complete and post-riser residence time is minimal. This further helps in preserving olefins produced in the riser reactor by selective catalytic cracking . In addition, entrainment of the product vapors with the spent catalyst is minimal. This reduces delta coke and hence regenerator temperature which in turn helps to increase catalyst-to-oil ratio that is critical for achieving high yield of light olefins. In summary, this RTD design preserves and promotes the selective component cracking characteristic of the I-FCC reaction system.

Modular Grid (MG) Catalyst Stripper: This stripper design uses flat baffles that are angled and oriented such that the entire cross-sectional area of the stripper is available for catalyst flow and maximum surface area is available for catalyst and stripping steam contact (Figure 6). The baffles are assembled and built as modular grids that can be installed in the stripper vessel by simply stacking them over one another. These grids can be easily inspected in place through manholes provided in between the stacks.

P2

P2 >P1

P1

To Fractionator

Vents

P3

StripperGas

Figure 5

581

Figure 6 Highly efficient modular grid stripper internals

The MG Stripper is highly efficient compared to other designs. It can accept very high catalyst flux rates and still maintain excellent efficiency. Advantages of the MG Stripper include reduced dry gas, reduced delta coke (and hence higher catalyst-to-oil ratio), increased conversion/yield of light products, and higher selectivity to light olefins. Lummus Technologys FCC reactor-regenerator design is complimented by a very stable

catalyst circulation system and an efficient catalyst regenerator design, including highly effective spent catalyst and air distributors. All these design features are successfully operating in several FCC units worldwide. Performance of FCC units improved significantly after these were revamped by incorporating some or all of the hardware design features described above. When the objective is to produce light olefins from heavier feedstocks having high CCR and

metals, major concerns are excessive coke make, high regenerator temperature, high dry gas make, and high catalyst makeup rates. The low delta coke and dry gas selectivity, and high metals tolerance of the Indmax catalyst, in conjunction with the hardware design improvements, allow I-FCC and I-RFCC to easily handle these difficult feedstocks. It is worth noting that I-RFCC can process heavy feedstocks with CCR as high as 5.0 wt% without a catalyst-cooler. Lummus Technology and the IOCL R&D Center have recently completed the design of an

85,000 BPSD I-FCC unit with the objective of maximizing propylene and ethylene from hydrotreated VGO feed. The unit, scheduled for startup in 2010, will maximize propylene yield and also produce an increased quantity of ethylene as feed to the downstream petrochemical complex. The I-FCC process has the flexibility to either maximize production of propylene plus ethylene, or maximize propylene alone, without an increase in ethylene yield. The design of another 15,000 BPSD unit with heavier feed is in progress, with the objective to maximize LPG/propylene plus gasoline. The design of these two units demonstrates the full flexibility of the I-FCC process with regard to product slate optimization and /or processing a wide range of feedstocks. Conclusion The I-FCC/I-RFCC process is a very significant advance in the field of light olefins

production by catalytic cracking. This process can upgrade heavy feeds (including residue)

582

with high yield of light olefins such as propylene, ethylene, and butylenes. This is due to the combination of unique catalyst formulation, hardware design, and operating strategy. The I-FCC/I-RFCC process is highly flexible as it involves only riser cracking. The operation

can be easily adjusted depending on the demand and pricing of different products. The multi-component catalyst approach offers further flexibility by easily adjusting the catalyst formulation to meet any changes in feedstock properties or market demand of the various products. I-RFCC offers technological advantages in terms of low delta coke, and dry gas production. These factors assume added importance when processing heavier feedstocks. The need for a catalyst cooler in many situations is eliminated. References

1. CMAI News, Nov.12, 2001. 2. Hutchinson, D.; Hood, R.; Catalytic Cracking to Maximize Light Olefins, PETROLE ET TECHNIQUES, March April 1996, 29.

3. U.S. 3,140,249 (to Socony Mobil) July 7, 1964. 4. U.S. 3,140,251 (to Socony Mobil) July 7, 1964. 5. U.S. 3,595,611 (to WR Grace) July 27, 1971; 6. U.S. 3,957,623 (to WR Grace) May 18, 1976. 7. U.S. 5,549,813 (to Texaco) August 27, 1996 8. Buchanan, J. Scott; Adewuyi, Applied Catalysis A: General 134 (1996) 247-262. 9. U.S. 3,758,403 (to Mobil) September 11, 1973. 10. Rajagopalan, K.; Young, G.W.; Am. Chem. Soc. Symp. Ser., 375 (1988) 34. 11. Madon, R.J.; J. Catal. 129 (1991) 275. 12. Buchanan, J.S.; Appl. Catal. 74 (1991) 83. 13. Haag, W.O.; Lago, R.M.; Weisz, P.B.; Discuss. Faraday Soc. 72, 317 (1981). 14. Upadhyay et al; Successful Demonstration of Indmax Process for Conversion of Residues to LPG and High Octane Gasoline, International conference Petrotech-2005 held at New Delhi in January 2005.

583