AU605188B2 - Processing of activated heavy hydrocarbon feeds - Google Patents

Description

7 AU-AI79681/ 8 7 WORLD INTELLECTU PRO ERTY ORGANI ATI& International Bureau PCT INTERNATIONAL APPLICATION PUBLISHED UNDER THE PATENT COOPERATION TREATY (PCT) (51) International Patent Classification (11) International Publication Number: WO 88/ 01638 55/02 Al (43) International Publication Date: 10 March 1988 (10.03.88) (21) taternational Application Number: PCT/US87/02227 (72) Inventors; and Inventors/Applicants (for US onv) DWYER, Francis, Gerard (22) International Filing Date: 3 September 1987 (03.09.87) [US/US]; 1128 Talleyrand Road, West Chester, PA 19380 HERBST, Joseph, Anthony [US/US]; 60 Bryant Road, (31) Priority Application Numbers: 903,183 Turnersville, NJ 08012 HUANG, Yun-Yang [US/US]; 903,314 98 William Feather Drive, Voorhees, NJ 08043 OW- 903,341 EN, Hartley [US/US]; 5 Riverview Terrace, Belle Mead, NJ 08502 SCHIPPER, Paul, Herbert [US/US]; 1506 Ber- (32) Priority Dates: 3 September 1986 (03.09.86) wyn Road, Wilmington, DE 19810 SCHWARTZ, Al- 3 September 1986 (03.09.86) bert, B. [US/US]; 1901 John F. Kennedy Boulevard, Apart- 3 September 1986 (03.09.86) ment 1204, Philadelphia, PA 19103 (US).

(33) Priority Country: US (74) Agents: STONE, Richard, D. et al.; Mobil Oil Corporation, 3225 Gallows Road, Fairfax, VA 22037 (US).

Parent Applications or Grants (63) Related by Continuation (81) Designated States: AT (European patent), AU, BB, BE (Euro- US 903,183 (CIP) pean patent), BG, BJ (OAPI patent), BR, CF (OAPI patent), Filed on 3 September 1986 (03.09.86) CG (OAPI patent), CH (European patent), CM (OAPI pa- US 903,314 (CIP) tent), DE (European patent), DK, FI, FR (European pa- Filed on 3 September 1986 (03.09.86) tent), GA (OAPI patent), GB (European patent), HU, IT US 903,341 (CIP) (European patent), JP, KP, KR, LK, LU (European patent), Filed on 3 September 1986 (03.09.86) MC, MG, ML (OAPI patent), MR (OAPI patent), MW, NL (European patent), NO, RO, SD, SE (European patent), SN (71) Applicant (for all designated States except US): MOBIL OIL (OAPI patent), SU, TD (OAPI patent), TG (OAPI patent), CORPORATION [US/US]; 150 East 42nd Street, New York, US, US, US.

NY 10017 (US).

iPublished S I With international search report.

28 APR 1988

AUSTRALIAN

(54) Title: PROCESSING OF ACTIVATED HEAVY HYDROCARBON FEEDS 24 A1988 2 4 MAR I988 (57) Abstract 22 0 m PATENT OFFICE Activation of heavy hydrocarbon feeds, preferably by thermal treatment, improves o 242 downstream catalytic processing. One embodiment of such thermal treatment is shown in fi- 212aaJ gure 2. A hydrocarbon resid (209) is first pre- treated in a microwave heater (211). The pre-1 treated resid is then admixed with catalyst and 214 light hydrocarbons wherein the mixture is passed up through a riser reactor (206). In the 224 riser reactor the pretreated resid is catalytical- 2221 22 ly cracked. Another mode of thermal pretreat- 0 ment of the hydrocarbon feed is visbreaking.

This is shown in figure 4. s- 206 -274 i I 2 204 1

LIGHT

HYDRCCAReON SF 11CE.A-,C.T i 1 WO 88/01638 PCT/US87/02227 -1- PROCESSING OF ACTIVATED HEAVY HYDROCARBON FEEDS CROSS REFERENCE TO RELATED APPLICATIONS This application is a continuation-in-part of prior co-pending applications Serial Nos. 903,183, 903,314 and 903,341, all of which are incorporated herein by reference.

BACKGROUND OF THE INVENTION Upgrading of heavy hydrocarbon chargestocks, such as resids, to lighter, more valuable products has been a goal of refinery operators for years. Because of their high viscosity, high pour point and high boiling range they have generally been suitable for use only as residual fuel oil, a relatively low value product.

Refiners have tried to minimize the amount of residual fuel oil produced from a given volume of crude by both catalytic and thermal approaches.

Most refiners have adopted thermal approaches (non-catalytic) such as visbreaking and delayed coking. Visbreaking is a relatively mild thermal cracking process primarily used to reduce the viscosity and pour point of vacuum tower bottoms enough to meet the specifications of No. 6 fuel oil, or at least to reduce the amount of refinery cutting stock required to dilute the resid to meet the specifications.

Coking, especially delayed coking, is a popular way of recovering lighter, more valuable products from resids. In delayed coking the residual feed is heated in a furnace and then charged to one or more coke drums where coke (which is believed to be a polymer) forms. The net efftct of coking is to convert a hydrogen deficient heavy fuel (resid) into two fractions, a very hydrogen deficient material (coke) and a relatively hydrogen enriched materials (coke or gas oil). Coking produces valuable liquid products, but converts a good portion of the feed to low value coke, frequently 20 to Swt coke is produced.

SWO 88/01638 PCT/US87/02 2 2 7 -2- Some early work on non-catalytic upgrading using microwave energy on heavy, hydrogen deficient feeds was reported in US 3,503,865 to Stone.

Coal was liquefied using microwave energy.

Microwaves have been used to heat hydrocarbon streams. US 4,230,448 to Ward et al discloses an oil burner having a microwave energy source connected to a fuel supply line to heat the fuel.

US 4,279,722 to Kirkbride discloses subjecting hydrocarbon reactants in contact with catalytic material, such as fluid catalytic cracking catalysts, with microwaves.

Catalytic routes to upgrading of heavy oil generally involve hydrotreating, hydrocracking, or catalytic cracking. In general these processes have not been too successful because the residual feeds contain a lot of sulfur and nitrogen, and high levels of metals which poison catalyst. In addition, the large amount of asphaltenes present in the feed tend to rapidly poison the catalyst.

Some work has been reported in the literature on charging of whole crudes or resid fractions.to catalytic cracking units. Resids regenerate too much heat in conventional FCC units so it is difficult to "heat balance" an FCC unit operating with a resid feed. These problems can be overcome to some extent by providing cooling coils in the regenerator section to remove heat and keep the unit in heat balance.

As resids make a lot of coke, and conventional FCC's can only tolerate small amounts of coke, the amount of resid and similar materials in the FCC feed is usually restricted.

Although modern zeolite cracking catalysts, using zeolites X and Y, are low coke producing catalysts, FCC's still do not tolerate much resid in the feed.

Although the problems of upgrading resids are substantial, the rewards are substantial enough to justify continued attempts to develop a way to economically and efficiently upgrade such residual fractions.

Some attempts have been made to combine one or more conventional upgrading steps, by charging a resid to a coker, and charging the coker gas oil to an FCC unit. Unfortunately, such coker gas oils are generally more unreactive than virgin gas oils, as reported in Veneuto and SWO 88/01638 PCT/CS87/02227 Habib, Fluid Catalytic Cracking, at page 27. This reference also reports that some other integration, of a sort, has occurred in that slops from chemical plants, heavy pyrolysis oils and heavy oils from the visbreaker may also be routed to an FCC. This is believed to represent a common refinery process to collect off-spec, or unwanted, product fractions in a tank farm, and slowly eliminate these materials by blending them with the FCC feeds.

In general, catalytic approaches to upgrading of resids have not been too successful and refiners usually resort to the thermal processes discussed above to upgrade resids.

In conventional fluidized catalytic cracking (FCC) processes, many improvements have been made, but usually not directed to resid processing.

Typically, in FCC processes, a relatively heavy hydrocarbon feed, a gas oil, is mixed with a hot regenerated cracking catalyst in the base of an elongated riser reactor and cracked to lighter hydrocarbons. The cracked products and spent catalyst are discharged from the riser and separated into a vapor phase and a catalyst phase. The catalyst passes through a stripper to remove entrained hydrocarbons from the catalyst, then catalyst is regenerated. The catalyst circulates between the reactor and the regenerator and transfers heat from the regenerator to the reactor, supplying heat for the endothermic cracking reaction.

In 4,051,013, a naphtha feed and a gas oil feed are converted in the presence of amorphous or zeolite cracking catalyst in a riser reactor to high octane gasoline.

Several FCC processes use a mixture of catalysts having different catalytic properties, U.S. 3,894,934 uses a mixture of a large pore zeolite cracking catalyst such as zeolite Y and shape selective zeolite such as ZSM-5. The combined catalyst system (or mixture) produces a gasoline product of relatively high octane rating.

J In U.S. 4,116,814, Zahner teaches use of two different kinds of catalyst, with catalyst separation in the fluidized regenerator. This approach will work, but when a less coke sensitive catalyst containing is used, the catalyst unnecessarily will spend a lot of time in the L i WO 1638 PCT/US87/022 2 7 WO 89/01638 -4regenerator. It would be beneficial if additive catalysts like this would be separated outside the regenerator.

The approach taken in U.S. 4,490,241, Chou, to keeping the additive out of the regenerator is to make the additive very light, so that it can be collected in secondary cyclones downstream of the riser reactor.

Use of very small, or light, particles of ZSM-5 additive which is recycled from secondary cyclones will work but will result in rapid loss of additive with catalyst fines. Use of light, or low density, ZSM-5 additive will also minimize the residence time of the ZSM-5 in the riser reactor because the light catalyst will not "slip" in the riser as much as the II conventional catalyst. The light ZSM-5 will be largely kept out of the regenerator, but at the price of less residence time in the riser reactor.

U.S. 4,336,160 reduces hydrothermal degradation of conventional FCC h catalyst by staged regeneration. However, all the catalyst from the reactor still is regenerated, thus providing opportunity for hydrothermal degradation.

Although FCC processes using very active zeolite based catalysts, Sor mixtures of two or more zeolite catalysts are known, they have not been used much for cracking of hydrogen-deficient feeds such as resids.

In U.S. 4,035,285, a low molecular weight carbon-hydrogen contributing material and a high molecular weight feedstock, a gas oil, are combined and reacted in the presence of zeolite catalysts, e.g., izeolite Y with ZSM-5. The resulting cracking and carbon-hydrogen additive I products are superior to those formed in the absence of the low molecular weight carbon-hydrogen contributing material. Advantages of the process include improved crackability of heavy feedstocks, increased gasoline yield and quality, and better fuel oil with less sulfur and nitrogen. The need ii for high pressure hydrotreaters and hydrocrackers is reduced or eliminated.

If A similar process in which whole crude and naphtha are cracked in J the presence of low molecular weight carbon-hydrogen contributing material and zeolites in separate risers of a catalytic cracking unit is described in U.S. 3,974,062.

In spite of the many advances made, there is still a need for processes which can upgrade heavy, hydrogen-deficient feeds.

I

SN"W O 88/01638 PCT/US87/02217 A way has now been discovered to upgrade these refractory stocks.

At its most basic level, the feed is made reactive by pre-treatment, and then, while still reactive, subjected to additional contact with catalyst, hydrocracking, hydrotreating or catalytic cracking. When FCC or catalytic cracking is practiced, the reactive feed readily reacts with low molecular weight carbon-hydrogen fragments which may be generated in the base of the riser, to react with heavy feed added higher up in the riser.

This promotes catalytic cracking of the feed to products which contribute to gasoline boiling range material.

Use of an elutriable catalyst, a mixture of low coke-forming, long lasting additive catalyst such as ZSM-5 with conventional FCC catalyst allows refiners to break the chains that heretofore made regeneration of additive catalysts proceed in lockstep with the conventional catalyst regeneration. Catalyst elutriation allows more efficient fragment generation, and more efficient cracking of the heavy feed.

Accordingly, the present invention provides a feed activation catalytic process for upgrading a heavy, hydrogen deficient feed comprising I catalytic cracking, which in eases the activity of the feed for subsequent catalytic upgrading by adding nergy to the feed prior to catalytic treatment; and, passing the activated eed, without intermediate storage thereof, into a catalytic processing zon to produce a catalytically upgraded product.

Each important p rameter of the process will be discussed. Those parts of the process which re conventional will receive only brief mention. Feed activation is iscussed first. Then various catalytic processes which can be used to onvert the activated feed are discussed.

The following will be considered.

1. Activation treatments e.g. visbreaking) 2. Hydrotreating 3. Hydrocracking 4. Conventional FCC catalysts REY in matrix) S. Additive FCC catalysts ZSM-5 in a matrix) 6. H 2 deficientfedse.g Resid iL o

I'

subjecting the feed to at least one of liquid phase visbreaking at a visbreaking severity of 500-2000 ERT seconds, and microwave treatment; and, cascading the visbroken or microwaved feed, without intermediate storage thereof, into a catalytic processing zone to produce a catalytically upgraded product.

Each parameter of the process will be discussed.

Those parts of the process which are conventional will receive only brief mention. Feed activation is discussed first. Then various catalytic processes which can be used to convert the activated feed are discussed. The following possible parameters will be considered: 1. Activation treatments visbreaking) 2. Hydrotreating 3. Hydrocracking 4. Conventional FCC catalysts REY in matrix) 5. Additive FCC catalysts ZSM-5 in a matrix) 6. H 2 deficient feeds Resids) *o i 4 SWO 88/01638 PCT/US87/02227 -6- 7. H 2 rich feeds propane) S8. FCC cracking conditions single or dual riser) S9. FCC fragment generation cracking propane) FCC riser elutriation i 11. FCC stripper elutriation 12. FCC sieve stripper 13. FCC exothermic stripping 14. FCC catalyst regeneration w/air) FCC catalyst reactivation w/olefins) Although each parameter is discussed, not all embodiments of the i invention will require all elements discussed above. All elements can cooperate to upgrade heavy feeds, but economics will determine if, e.g., there is a reasonable return on investment from having an FCC with 1 elutriable ZSM-5 and stripper elutriation of ZSM-5 downstream of a I visbreaker.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a schematic diagram of a dual riser FCC system of a first embodiment of the invention; Figure 2 is a schematic diagram of a single riser FCC system of a second embodiment of the invention; and Figure 3 is a schematic diagram of a dual riser FCC system of a third embodiment of the invention.

Figure 4 is another embodiment of the invention, using an elutriable catalyst mixture, an elutriating riser reactor, and elutriating stripper, and a visbroken resid feed.

ACTIVATION PRETREATMENT OF FEED The activation pretreatment can be any treatment which imparts enough energy into the heavy feeds to make them more susceptible to upgrading in downstream conventional catalytic processes.

Activation energy can be added to the feed via radiation, e.g., microwave radiation, laser radiation, or ultrasound, or by severe thermal pretreatment, heating.

WO 88/01638 PCT/US87/02227 S-7- Commercially, heating the feed is preferred, because it is easy to heat a refinery stream in a fired heater, heat exchanger, or the like.

The more exotic energy sources such as lasers and microwaves may also be used, preferably in conjunction with heating. Although these exotic sources are far more costly than simple high temperature, the exotic sources can be fine tuned to activate selected portions of the charge stock, whereas much less selective activation is possible with a thermal pretreatment.

Regardless of the method used to activate the feed, enough energy must be added to activate the feed for the desired downstream processing reaction.

VISBREAKING PROCESS Visbreaking, or viscosity breaking, is a well known petroleum refining process in which reduced crudes are pyrolyzed, or cracked, under comparatively mild conditions to provide products having lower viscosities and pour points, thus reducing the amounts of less viscous and more valuable blending oils required to make the residual stocks useful as fuel oils. The visbreaker feedstock usually consists of a mixture of two or more refinery streams derived from sources such as atmospheric residuum, vacuum residuum, furfural-extract, propane-deasphalted tar and catalytic cracker bottoms.

Most of these feedstock components, except the heavy aromatic oils, behave relatively independently in the visbreaking operation. Consequently, the severity of the operation for a mixed feed is limited greatly by the least desirable (highest coke-forming) components. In a typical visbreaking process, the crude or resid feed is passed through a heater and heated to about 425 to about 525 0 C at about 450 to about 7000 kPa. Light gas-oil may be recycled to lower the temperature of the effluent to about 260 to about 370 0 F. Cracked products from the reaction are flash distilled with the vapor overhead being fractionated into a light distillate overhead product, for example gasoline and light gas-oil bottoms, and the liquid bottoms are vacuum fractionated into heavy gas-oil distillate and residual tar.

Examples of such visbreaking methods are described in Beuther et al, 'Thermal Visbreaking of Heavy Residues," The Oil and Gas Journal, 57:46, November 9, 1959, pp. 151-157; Rhoe et al, "Visbreaking: A Flexible L SWO 88/01638 PCT/US87/02227.

-8- Process," Hydrocarbon Processing, January 1979, pp. 131-136; and U.S. Patent 4,233,138.

When thermal pretreatment, such as visbreaking, is used, the severity of processing can be conveniently expressed as equivalent reaction time at 800 0 F, or ERT. The concept of ERT was originally developed as a way to predict what would happen in visbreakers and delayed cokers. It allows comparison of one visbreaker operating at a relatively low temperature with a long feed residence time to another visbreaker operating at a higher temperature and shorter residence time.

Conventionally, visbreakers operate at reaction severities of 250 to 1500 ERT seconds. Cokers operate typically at 2000 to 5000 ERT.

When a relatively non-selective, thermal activation treatment is i used, the lower limit on severity is about 100 ERT, but this brings about i such a small amount of activation that it will be difficult to see it in Sdownstream catalytic processing.

SIt is preferred that severities similar to those used in classical visbreaking operations be used, 500 to 1500 ERT seconds.

An upper limit on severity, especially when .using a non-selective |i thermal activation pretreatment, is solids formationor coke formation. Some of the reactive heavy intermediates formed by the pretreatment process of the present invention may react with other reactive heavy intermediate species, or with some other part of the heavy feed to form solids. Thermal pretreatment should not be so severe as to generate any significant amount of coke or solids. Preferably, the upper limits on solids generation is 1-2 wt. though operation with generation of less than 0.5 wt. solids, and more preferably less than 0.1 wt solids is preferred.

Temperatures which may be used in thermal processing preferably range from 800 to 1500 0 F, more preferably from 900 to 1200 0 F, measured at the outlet of the thermal process.

The upper limit on severity is about 200 to 5000. This material would readily form coke if sent to a coking drum, but will be exceedingly reactive if promptly fed to an FCC or other catalytic unit.

Generation of solids, as used herein, refers to solids generation between activation and downstream catalytic process. Severe thermal 1~ MUO 38/01638 PCT/US87/02227 -9pretreatment, as used herein, could result in formation of solids after several days or weeks of standing. This is not too severe, because preferably only a few minutes, elapse between pretreatment and catalytic treatment.

The optimum thermal pretreatment severity is believed to corresponding to a visbreaking pretreatment which is severe enough to cause inscdbility problems in the visbroken fuel in storage. This severely visbroken charge stock does not immediately form sediment, the sediment forms gradually in a storage tank. This severely treated material, which contains many reactive fragments, is an ideal feedstock for downstream catalytic processing units, but is unsuitable for storage as fuel, because it will develop sediment after standing for one or two weeks in a storage tank.

It is essential that thermal processing stop well short of significant coke formation. Operation of a fluid coker in cascade relationship with a downstream FCC has nothing to do with the present invention. The fluid coker undoubtedly generates a lot of reactive fragments, which then go on to react with one another to form coke.

Preferably the severe thermal treatment is conducted at sufficient pressure to maintain at least SO wt of the heavy feed in liquid phase.

The presence of a hydroaromatic solvent during the thermal treatment, such as is disclosed in US 4,615,791, to Choi, et al, incorporated herein by reference, is helpful.

Other conventional additives, used to permit higher severity visbreaking or related thermal processing, may also be used.

Although we do not know what goes on during the activation pre-treatment of the present invention, it is believed that relatively short lived free radicals are generated which can react with acid sites on a catalyst. Some examination of visbroken product over several weeks time indicates that free radicals can exist for days and weeks.

Fairly elegant equations and tables have been developed to calculate ERT. Typical of such presentations is the discussion of "soaking factor" in Petroleum Refinery Engineering--Thermocracking and Decomposition L II I i i la- rri~ lr~ IIICII11-~l~ WO 8/01638 PCT/US87/02227 SProcess--Equation 19-23 and Tahle 19-18, in Nelson Modern Refining Technology, Chapter 19.

Although that text uses the term "soaking factor", the term "ERT" or "Euivalent Reaction Time" in seconds as measured at 427 0 C is used herein to express visbreaking severity; numerically, soaking factor is the same as

ERT.

ERT refers to the severity of the operation, expressed as seconds of residence time in a reactor operating at 427°C. In very general terms, Sthe reaction rate doubles for every 12 to 13'C increase in temperature.

j Thus, 60 seconds of residence time at 427 0 C is equivalent to 60 ERT, and increasing the temperature to 456 0 C would make the operation five times as severe, i.e. 300 ERT. Expressed in another way, 300 seconds at 427 0 C is Seauivalent to 60 seconds at 456 0 C, and the same product mix and distribution should be obtained under either set of conditions.

i Some visbreaker units operate with 20-40% vaporization at the Svisbreaker coil outlet. Light solvents will vaporize more and the vapor will not do much good towards improving the cracking of the liquid phase material. Accordingly, liquid phase operation is preferred, but significant amounts of vaporization can be tolerated.

The pressures commonly encountered in visbreakers range from 170 to 10450 kPa, with a vast majority of units operating with pressures of 1480 to 7000 kPa. Such pressures will usually he sufficient to maintain linuid phase conditions and the desired degree of conversion.

Most visbreakers operate with a coil, some operate with a combination of a coil and a drum, and a few operate primarily with a drum.

As far as product distribution is concerned, it is believed that it is of no great significance whether the residence time is obtained in a coil, drum, or combination of both.

Frequently, visbreaker units are built with a coil, and when an expansion of the unit's capacity is desired it is cheaper to add a soaking drum (and increase the oil's residence time) than to build and operate a bigger furnace and achieve a higher reactor temperature.

Typical of the coil/soaking drum combinations is the process described in U.S. Patent 4,247,387.

Ohl WO 88/01638 PCT/US87/02227 -11- The preferred hydro-aromatic solvents which may be used in the visbreaking process are thermally stable, polycyclic, aromatic/hydroaromatic distillate hydrogen donor.materials, preferably resulting from one or more petroleum refining operations. The hydrogen-donor solvent nominally has an average boiling point of 200 to 500 0 C, and a density of 0.85 to 1.1 g/cc.

Examples of suitable hydrogen-donors are highly aromatic petroleum refinery streams, such as fluidized catalytic cracker (FCC) "main column" bottoms, FCC "light cycle oil," and thermofor catalytic cracker (TCC) "syntower" bottoms, all of which contain a substantial proportion of polycyclic aromatic hydrocarbon constituents such as naphthalene, dimethylnaphthalene, anthracene, phenanthrene, fluorene, chrysene, pyrene, perylene, diphenyl., benzothiophene, tetralin and dihydronaphthalene, for example. Such refractory petroleum materials are resistant to conversion to lighter (lower molecular weight) products by conventional non-hydrogenative procedures. Typically, these petroleum refinery residual and recycle fractions are hydrocarbonaceous mixtures having an average carbon to hydrogen ratio above about 1:1, typically around 1.5:1, and an average boiling point above 230 0

C.

An FCC main column bottoms refinery fraction is a highly preferred hydrogen donor solvent. A typical FCC main column bottoms (or FCC clarified slurry oil (CSO)) contains a mixture of constituents as represented in the following mass spectrometric analysis: I1 iU *i WO 88/01638 PCT/US87/02227 WO 88/01638 PCT/US87/02227 -12- Naphthenic/ Aromatics LabileH2 Comp~ounds Aromat ics Alkyl -Benz ene Naphthene- Benzenes 0 inapthene- Ben zenes Naphthalenes Acenaphthenes( biphenyls) FlUorenes Phenanthrenes Naphthene-phenanthrenes Pyrenes, fluoranthenes Chrysenes Benzofluoranthenes Pe rylenes Benzothiophenes Dibenzothiophenes Naphthobenzothiophenes Total 0.00 0.03 0.16 0.00 0.08 0.11 7.4 10.1 13.1 11.0 20.5 10.4 0.18 0.00 0.00 0.00 0.00 2.4 35.6 64.4 0.60

I

/038/01638 PCT/US87/02227 13- A tv~pical FCC main columnrn bottoms or clarified slurry oil has the following analysis and properties: Elemental Analysis, Weight

C

H

0

N

S

Total 89.93 7.35 0.99 0.44 1.09 99.80 Pour Point, CCP., 9.96 Distillation: IBP, IC: 254 OC: 338 9S%, OC: 485

I

WO 88/01638 PCT/US87/02227 -14- Closely related to FCC is TCC, or Thermofor catalytic cracking.

Thermofor catalytic cracking is roughly analogous to FCC; both processes operate without addition of hydrogen, both operate at relatively low pressure, and both require frequent regeneration of catalyst. The products of thermofor catalytic cracking will have hydrogen contents and distribution very similar to those obtained as a result of FCC. Accordingly, light cycle oils obtained as product streams from a TCC process, or main column bottoms streams obtained as a result of a TCC process, are also suitable for use as hydroaromatic solvents.

Another source of hydrogen donor solvent is the heavy fraction normally associated with lubricating oil. The lubricating oil may be either a paraffin based oil or a naphthenic based oil. Preferably the lubricating oil is first subjected to aromatics extraction, so that the extract will have more ideal properties.

The aromatic extract from a lube oil plant is highly aromatic and not a good hydrogen-donor; however, it may he hydrogenated to produce a hydrogen-donor diluent with the right hydrogen content and distribution.

Diluents or solvents with the right hydrogen content and distribution are produced also by the catalytic dewaxing of lubricating oil stocks and the catalytic dewaxing of fuels.

Another suitable hydrogen donor solvent source is the highly aromatic tars produced in olefin crackers.

Further sources of suitable hydrogen donor solvents are the various coal liquifaction processes.

Normally it will be most economic to use whatever solvents can be found in a refinery, without hydrotreating, for use in the process of the invention. It is also possible, although usually more expensive, to take a hydrogen-donor solvent which is not entirely satisfactory, and hydrogenate it.

Critical features of the hydrogen-donor solvent are its particular proportions of aromatic, naphthenic and paraffinic moieties and the type and Quantity of hydrogen associated therewith. A high content of aromatic and naphthenic structures together with a high content of alpha hydrogen provides a superior hydrogen-donor material.

-e 088/01638 PCT/US87/02227 The solvents preferred are hydro-aromatic solvents.

The hydrogen transfer ability of a donor material can be expressed in terms of specific types of hydrogen content as determined by proton nuclear magnetic resonance spectral analysis. Nuclear magnetic resonance characterization of heavy hydrocarbon oils is well developed. The spectra are divided into four bands according to the following frequencies in Hertz (Hz) and chemical shift (ppm): Halpha Hbeta Hgamma HAr Freq. (Hz) 0-60 60-100 120-200 360-560 Shift (ppm) 0-1.0 1.0-1.8 2.0-3.3 6.0-9.2 The HAt protons are attached directly to aromatic rings and are a measure of aromaticity of a material. Halpha protons are attached to non-aromatic carbon atoms themselves attached directly to an aromatic ring structure, for example alkyl groups and naphthenic ring structures. Hbeta protons are attached to carbon atoms which are in a second position away from an aromatic ring, and Hgamma protons are attached to carbon atoms which are in a third position or more away from an aromatic ring structure. This can be illustrated by the following: Some alpha hydrogens are not donatable, for example the alpha hydrogen in toluene.

The HAr protons are important because of their strong solvency power. A high content of Halph a protons is particularly significant because Halpha protons are labile and are potential hydrogen-donors.

The hydrogen-donor materials if used, should have an HAr proton content is at least 20 percent, preferably from 20 to 50 percent, and an Halpha proton content is at least 20 percent, preferably from 20 to percent. For example, in H-donor streams containing 9.5 weight total hydrogen, the alpha-hydrogen content should be at least 1.9 weight (20% of total hydrogen content). The balance of the hydrogen is non- alpha-hydrogen.

Hydrogen-donors possessing the desired hydrogen content distribution may frequently be obtained as a bottoms fraction from the i -e WO 88/01638 PCT/US87/02227 -16catalytic cracking or hydrocracking of gas oil stocks in the moving bed or fluidized bed reactor processes. In general, depending upon such conditions as temperature, pressure, catalyst-to-oil ratio, space velocity and catalyst nature, a high severity cracking process results in a petroleum residuum solvent having an increased content of HAr and Halpha protons and a decreased content of the less desirable non- alpha-hydrogen.

The proton distribution of examples of various highly aromatic hydrocarbon by-product streams are shown below.

.~e PCT/US8/02227 Wo 98/01638 -17- PROTON ANALYSIS H-DONORS H alpha HAr %H (%Solv) %1 Solv.) HOther %H Solv.)

FCC/LCO

No. 1 22.2 No. 2 34.1 No. 3 34.3

FCC/CSO

No. 1 34.0 No. 2 30.0 No. 3 19.4

FCC/MCB

No. 1 36.0 No. 2 36.4 No. 3 18.5 No. 4 18.1 TCC/Syntower Bottoms No. 1 39.8 No. 2 18.2 No. 3 16.3

TCC

Distillate No. 1 21.5 No. 2 20 No. 3 6.9 SRC oil 27.1 (2.07) (3.18) (3.19) (2.43) (2.15) (1.39) (2.65) (2.68) (1.36) (1.33) (2.78) (1.70) (1.52) (2.39) (2.07) (0.89) 20.0 29.1 30.2 33.0 35.0 5.0 32.0 44.8 17.2 14.2 41.4 23.0 15.6 20.1 22 8 (1.87) (2.71) (2.81) (2.36) (2.51) (0.03) (2.36) (3.30) (1.26) (1.04) (3.86) (2.15) (1.45) (2.23) (2.28) (1.03) 57.8 36.8 35.5 33.0 35.0 65.0 32.0 18.8 64.3 67.7 28.8 58.8 68.1 58.4 58 85.1 21.6 (5.40) (3.43) (3.30) (2.36) (2.36) (33.3) (2.36) (1.38) (4.73) (4.96) (2.69) (5.49) (6.35) (6.49) (6.00) (10.98) H Total Solv.

9.34 9.32 9.30 7.15 7.17 7.16 7.36 7.36 7.35 7.35 9.33 9.34 9.32 11.12 10.35 12.90 46.3 where: LCO Light Cycle Oil CSO Clarified Slurry Oil MCB Main Column Bottoms TCC Thermofor Catalytic Cracking Syntower Synthetic Crude Tower SRC Oil Solvent Refines Coal Recycle Oil WO 88/01638 PCT/US87/02227 -18- The upper limit on thermal treatments is coking. The thermal treatment should not be severe enough to result in significant coke generation upstream of, or in, the catalytic cracking unit. Close coupling of a delayed coking furnace with an FCC unit may permit very severe thermal pretreatment of the feed without significant coke generation, as long as the material coming out of the coking furnace is charged directly to the FCC unit, before it has a time to form coke.

The present invention does not involve recovery of byproduct streams from thermal processing and charge of these streams to the catalytic cracking unit. Operation of a conventional delayed coker to form coke and lighter materials, such as coke or gas oil, with passage of coke or gas oil to an FCC unit is not part of the present invention.

It is essential that the activated feed be promptly subjected to catalytic processing while still activated. It is believed that thermal treatment generates short lived reactive species, perhaps free radicals, which are very amenable to catalytic cracking, but which disappear within a few minutes. Direct, cascade operation of a thermal treatment unit into an FCC unit, preferably with a heated and insulated transfer line, fs preferred. Preferably no more than 10 minutes should elapse between thermal treatment and catalytic treatment, and most preferably no more than one minute separates the two processes.

The thermal treatment process used preferably is a liquid phase process, such as visbreaking, which does not result in formation of any substantial amount of coke or solid material. It is believed that the reactive species formed in thermal treatment are the coke precursors which form coke in, a delayed coking drum. Allowing the reactive species to react with one another and form coke defeats the whole point of thermal processing. Thus operation of a fluid coker upstream of an FCC unit would not achieve any benefit because although reactive species would undoubtedly form, they would react with themselves to form coke rather than remain reactive for better upgrading in the FCC reactor.

;1

V

I JM WO 88/01638 PCT/US87/02227 -19- ACTIVATION MICROWAVE TREATMENTS When the heavy, H, deficient feed, residuum, is to be activated by microwave treatment exposure to microwave energy in the range from 1.0 to 1000 gigacycles per second (1.0 x 10 to 101 cycles per second, respectively) can be used. Preferably, the microwave energy heats a preheated resid to 600°-1100F (316° 5930C) in, at most, 5 minutes.

Preferably, the microwave energy heats the resid to 8200 1100°F (4380 593°C) in 1 minute, or less than, less than 30 seconds. Most preferably, the microwave energy heats the resid to 9000 1100 0 F (4820 593°C) in less than 5 seconds.

Preferably, enough microwave energy is added to heat the feed at least 25 0 F (140C), more preferably by over 50OF and most preferably by over 100°F (560C).

HYDROTREATING

Any conventional hydrotreating process can be used. Such processes typically operate with relatively high hydrogen partial pressures, on the order of 100-1000 psig (790 to 7,000 kPa), and preferably at 150-450 psig (1100 to 3200 kPa).

Conventional hydrotreating catalyst comprise a catalyst support, usually a high surface area material such as alumina, containing one or more hydrogenation/dehydrogenation components. Any hydrotreating catalyst now known or hereafter developed can be used. The thermally activated heavy feeds are much easier to hydrotreat than feeds which have not been preactivated.

HYDROCRACKING CATALYST The activated heavy feed may be charged to any conventional hydrocracking unit. Such units usually operate at relatively high hydrogen partial pressures and elevated temperatures. Auitable catalysts and operating conditions are disclosed in U.S. Patent 4,435,275 and in European Patent 0098040, both of which are incorporated by reference.

1- Wo 88/01638 PCT/US87/02227 The hydrocracking catalyst can be all amorphous, but preferably contains some zeolite, such as zeolite-Y, along with a hydrogenation/dehydrogenation component. Any hydrocracking catalyst now known or hereafter developed can be used.

Because heavy feeds used herein are still very viscous, difficult to handle materials, even after thermal or other pretreatment to activate the feeds, it is preferred that the catalysts, whether hydrotreating or hydrocracking, have a relatively large percent of their pore volume in i relatively large pores.

FCC CATALYST Conventional FCC catalysts are either amorphous or zeolitic. Most FCC's use zeolitic catalyst, typically a large pore zeolite, in a matrix which may or may not possess catalytic activity. Most zeolites typically used have crystallographic pore dimensions of 7.0 angstroms and above for their major pore opening. Zeolites usually used in cracking catalysts are zeolite X 2,882,244) and zeolite Y 3,130,007).

Silicon-substituted zeolites, described in U.S. 4,503,023 can also be used.

Two or more of the foregoing amorphous and/or large pore crystalline cracking catalysts can be used as the conventional catalyst.

Preferred conventional catalysts are the natural zeolites mordenite and I faujasite and the synthetic zeolites X and'Y with particular preference given zeolites Y, REY, USY and RE-USY.

Such conventional FCC catalysts, and those hereafter developed for FCC can be used in practicing the invention.

it ADDITIVE FCC CATALYSTS The present invention permits use of an optional additive catalyst, with different properties than the conventional catalyst, when catalytic cracking of activated resid is practiced.

Preferred additives comprise the shape selective medium pore zeolites exemplified by ZSM-5, ZSM-11, ZSM-12, ZSM-23, ZSM-35, ZSM-48 and similar materials.

tI ti

I

ki

V

VO 88/01 .638 P(T/US87/02227 -21is described in U.S. 3,702,886, U.S. Reissue 29,948 and in U.S. 4,061,724 (describing a high silica ZSM-5 as "silicalite").

ZSM-11 is described in U.S. 3,709,979.

ZSM-12 is described in U.S. 3,832,449.

ZSM-23 is described in U.S. 4,076,842.

is described in U.S. 4,016,245.

ZSM-48 is described in U.S. 4,375,573.

is particularly preferred.

The additive zeolites can be modified in activity by dilution with a matrix component of significant or little catalytic activity. The matrix may act as a coke sink. r~talytically active, inorganic oxide matrix material is preferred because of its porosity, attrition resistance and stability under the cracking reaction conditions encountered particularly in a fluid catalyst cracking operation. The additive catalyst may contain up to 50 wt crystalline material and preferably from 0.5 to 25 wt in a matrix.

The matrix may include, or may be, a raw or natural clay, a calcined clay, or a clay which has been chemically treated with an acid or an alkali medium or both.

Zeolites in which some other framework element which is present in partial or total substitution of aluminum can be advantageous. For example, such catalysts may convert more feed to aromatics with higher octanes.

Elements which can be substituted for part or all of the framework aluminum are boron, gallium, zirconium, titanium and other metals. Specific examples of such catalysts include ZSM-5 or zeolite beta containing boron, gallium, zirconium and/or titanium. In lieu of, or in addition to, being incorporated into the zeolite framework, these and other catalytically active elements can also be deposited upon the zeolite by any suitable procedure, impregnation. Thus, the zeolite can contain a hydrogen-activating function, a metal such as platinum, nickel, iron, cobalt, chromium, thorium (or other metal capable of catalyzing the Fischer-Tropsch or water-gas shift reactions) or rhenium, tungsten, molybdenum (or other metal capable of catalyzing olefin disproportionation).

SI' ii ri WO 88/01638 PCT/US87/02227 -22- Other additives can also be present, So x or NO x removal additives, metals removing additives and the like.

H

2 RICH FEEDS Suitable hydrogen-rich hydrocarbon feeds are those containing 12 to wt hydrogen, CHI, C 2

H

6

C

3

H

8 light virgin naphtha, and similar materials. Any or all of the C 1 to C 5 hydrocarbons recovered from the process can be used as H 2 rich feed to the lower region of the riser where these and other hydrogen-rich hydrocarbon materials undergo thermal cracking due to the hot, freshly regenerated cracking catalyst and/or shape selective catalytic cracking and other reactions due to the additive, the medium pore zeolite catalyst. The H 2 -rich feed when cracked in the base of the riser generates gasiform material contributing mobile hydrogen species and/or carbon-hydrogen fragments.

FCC RISER CRACKING Conventional FCC riser cracking conditions may be used.

Preferably, the resid has a relatively short residence time in the riser.

The activated resid quickly cracks when it contacts catalyst, and thus has little time for coking. Preferably, the resid residence time in the riser is less than 1 second, and most preferably between 0.5 and 1 second, to provide sufficient time for cracking while minimizing coking.

FRAGMENT GENERATION The light, H 2 -rich feed may be converted into light reactive fragments thermally, catalytically, or both. Contact of H 2 -rich feed with hot, regenerated conventional FCC catalyst will both thermally and catalytically crack the feed into reactive fragments.

The concept of generating light fragments is different from the concept of generating heavy, reactive intermediate species by resid feed pretreatment.

I .WO 88/01638 -23- Temperatures can range from preferably 677 to 732 0 C (1250 to 1350 50:1 to 200:1 and preferably is 100:1 can be 10 to 50 seconds and preferabl production is maximized by less sevei The easiest way to generate amounts of hot regenerated conventior amounts of light, hydrogen-rich feeds

H

2

DEFICIE

Suitable charge stocks for a hydrocarbons generally and, in parti( fractions, gas oils, thermal oi crudes, tar sand oils, shale oils, s fractions derived from the destructi asphalts, hydrotreated feedstocks de like. In short, any heavy conventior hydrogen-deficient feedstock can be PCT/US37/02227 593 to 816 0 C (1100 to 1500'F) and The catalyst to feed ratio can be to 150:1. The catalyst contact time y is 15 to 35 seconds. Light olefin re operation.

a lot of fragments is to have large ial catalyst contact relatively small such as propane.

.F FEEDS ctivation comprise the heavy :ular, conventional heavy petroleum *ls, residual oils, cycle stocks, whole ynthetic fuels, heavy hydrocarbon ie hydrogenation of coal, tar, pitches, rived from any of the foregoing, and the hal feedstock, and preferably a used in the process of this invention.

FCC RISER ELUTRIATTON When an elutriatable catalyst is used in a downstream FCC process it is preferred to have a riser elutriation zone. This can be a zone at either the bottom or top of the riser of increased cross-sectional area.

The increased cross-sectional area results in lower superficial vapor velocity in the riser, which allows the catalyst (preferably the additive) with the highest settling velocity to remain longer in the riser.

As a general guideline, the feed rate, riser cross-sectional area, and additive catalyst properties should be selected so that the additive catalyst settling rate approaches the superficial vapor velocity expected in the riser.

Riser expansion to handle increased molar volumes in the riser reactor would not change superficial vapor velocity and would not produce WO 88/01638 PCT/US87/02227 -24significant elutriation. Conversely, a constant diameter riser would provide elutriation at the base of the riser.

FCC STRIPPER ELUTRIATION When a elutriatable catalyst mixture is used, it is preferred to operate with a catalyst stripper which separates more from less elutriatable catalyst.

Separation in the stripper can be achieved by particle size difference alone, a sieve action. Preferably a stripper is used which separates conventional catalyst from additive catalyst by exploiting differences in settling velocity.

A closely related approach is one which relies to some extent on settling properties and to some extent on catalyst density, is shown in Figure 4.

Stripper elutriation separates additive catalyst from conventional catalyst upstream of the catalyst regenerator. If elutriation occurred in the catalyst regenerator, then the additive catalyst (which may not need regeneration and may be damaged by regeneration) is unnecessarily subjected to regeneration. Thus stripper elutriation significantly reduces additive catalyst residence time in the FCC regenerator.

Additional elutriation may occur in the FCC regenerator, by means not shown, however, it is the goal of the present invention to keep the additive out of the regenerator. Regenerator elutriation would minimize damage to additive catalyst and may be a beneficial way of quickly removing from the regenerator small amounts of additive which will spill over into the regenerator.

Riser elutriation, and stripper elutriation, preferably with the additive catalyst being a denser catalyst which settles rapidly, minimize loss of additive catalyst with catalyst fines. If elutriating cyclones were used to separate a light, readily elutriable additive from reactor effluent prior to discharge from the reaction vessel, then there would be a significant increase in loss of additive catalyst with fines. There would also be a significant dilution effect caused by accumulation of conventional catalyst fines with more elutriatable additives. Finally use of a light I i WO 8/01638 PCT/US87/02227 LU additive would reduce the residence time of the additive in the FCC riser reactor, because it would tend to be blown out of the reactor faster than the c ,'nvenrional catalyst.

FCC EXOTH-ERMIC STRIPPING Stripping efficiency can be improved by adding one or more light olef ins to a stripping zone. The light olef ins form higher molecular weight products (which are valuable) and heat (which aids stripping).

If exothermic stripping is to be practiced enough olefins should be added to heat the catalyst at least 281C (5O1F) and preferably at least 56 0

C

Ii(1000F) or more.

FCC CONVENTIONAL CATALYST REGENERATION q The conditions in the FCC catalyst regenerator are conventional.

U.S. 4,116,814 (and many other patents) discuss regent_ .ation conditions.

Internal or external heat exchangers may be used to remove heat.

FCC CATALYST REACTIVATION II Reactivation of additive catalyst, or conventional catalyst, with hydrogen or hydrogen-rich gas may be practiced herein. Operation with H 2 at 600-14001F, preferably at 800-1200*F gives 'good results.

Figure 1 shows integration of conventional FCC processing in one 4 riser with FCC upgrading of resid, activated by microwave radiation, and with reactive fragment generation in a second riser.

'1 As shown in figure 1, light hydrocarbon in line 4 enters a first FCC riser reactor 6 and combines with a hot regenerated catalyst, from a regenerator not shown, via conduit 8. Reactive fragments and catalyst pass up riser 6 and contact activated resid from line 13. Heated resid from line 1 9 passes through microwave heater 11 to become activated resid. The heated resid combines with the reactive fragments and catalyst and passes up riser 6.

The microwaved resid contains reactive compounds which combine with reactive fragments of the light hydrocarbons in the riser. Riser 6 discharges into a riser cyclone 10 in reactor vessel 7. Cyclone

I

Y

WO 88/01638 PCT/US87/02227 1 -26discharges catalyst down through a dipleg 14 into a catalyst bed 16. Gas exists cyclone 10 via outlet 12 into an atmosphere of vessel 7. The gas then passes through cyclone 40, which recovers entrained catalyst from gas.

Catalyst is discharged via dipleg 44 into bed 16. Gas passes via outlet 42 to a plenum chamber 50, and line 52 to a conduit 150.

The catalyst in bed 16 enters stripper 18 in a lower portion of reactor vessel 7. Catalyst passes down and countercurrently contacts stripping gas from line 20 and header 22. Optional trays (baffles) 24 may enhance contact. Stripped catalyst exits via conduit 30 and passes to the regenerator.

V Vacuum gas oil from line 102 and a heavy cycle oil recycle stream from line 174 pass via line 104 into a second, and conventional, FCC riser reactor 106. Hot regenerated catalyst, from a regenerator (not shown), is added to the base of the riser. Catalyst and feed pass through riser 106 and discharge into riser cyclone 110 in vessel 107. Catalyst is discharged via dipleg 114 to bed 116. Vapor exits via outlet into vessel 107. Vapor then passes into cyclone 140. Catalyst passes through dipleg 144 to bed 116. Vapor passes via line 142 to plenum chamber 150, line 152, and line 160, where it combines with vapor EiJm line 52, and enters fractionators 170.

Bed 116 passes down through stripper 118 and countercurrently contacts stripping gas from line 120 and header 122. Trays 124 enhance stripping. Stripped catalyst passes via conduit 130 to a regenerator.

Fractionators 170, comprising one or more towers or flash drums (not shown), separate hydrocarbons in line 160 into one or more light products in line 172, and a heavy cycle oil stream of 6S0°F (343C+) liquid, which is recycled via line 174.

The advantages of the Figure 1 embodiment of the invention include quick initial cracking of resid and separating the heaviest portion of the cracked resid and recycling it to a second riser reactor with a conventional feed, vacuum gas oil. The process enhances yields by alkylating activated resid with reactive light hydrocarbon fragments. The invention may also be practiced in a single riser FCC as shown in Figure 2.

i .WO 88/01638 PCT/US87/02227 -27- In Figure 2, a light hydrocarbon enters a riser 206 via line 204 and combines with hot regenerated catalyst from conduit 208 to form reactive fragments.

Resid in line 209 passes through microwave heater 211, is activated, and enters riser 206 via line 213 to combine with the catalyst and reactive light fragments. The resulting mixture passes up the riser, for a resid residence time of preferably less than 1 second, and contacts more regenerated catalyst and hydrocarbon from lines 203, 205 respectively.

The hydrocarbons in line 205 are preferably vacuum gas oil from line 22 and recycled heavy cycle oil from line 274. The mixture continues up through riser 206 into riser cyclone 210 vessel 207.

Cyclone 210 separates catalyst from vapor. Catalyst passes via dipleg 214 into bed 216. Vapor passes via outlet 212 into vessel 207.

Vapor from vessel 207 enters cyclone 240 which recovers entrained catalyst which passes via dipleg 244 into bed 216. The vapor passes via outlet 242, plenum chamber 250, vessel outlet 252, and line 260 to separator 270.

Separator 270 separates cracked products into a lighter product stream 272, and a heavy cycle oil stream comprising 650°F+ (343°C+) hydrocarbons, recycled via line 205 into riser 206.

The catalyst in bed 216 passes down through stripper 218 and contacts stripping gas from line 220 and header 222. Baffles 224 enhance stripping. Stripped catalyst exits vessel 218 via outlet 230 and passes to a regenerator, not shown.

The invention shown in Figure 2 uses a single riser to react microwave-activated resid with light hydrocarbons fragments, and catalyst, before adding more catalyst, vacuum gas oil and recycle heavy cycle oil.

The vacuum gas oil and recycled heavy cycle oil dilute the resid to minimize coking. The vacuum gas oil and cycle oil may also quench the riser as much as 300 0 F (167°C), preferably between 50° and 300°F (28° 167°C) to minimize coking.

In Figure 3, light hydrocarbons from line 304 and hot regenerated catalyst, from conduit 308, enter a first riser reactor 306 to form reactive fragments.

WO 88/01638 PCT/US87/02227 -28- Resid passes via line 309 into a microwave heater 311, and then into the riser via line 313.

The activated resid and reactive fragments, react in riser 306.

Riser 306 discharges into cyclone 310 in reactor vessel 307. Cyclone 310 discharges catalyst via dipleg 314 into bed 316. Vapor exits via outlet 312 into an atmosphere of vessel 307. Vapor leaves vessel 307 via a cyclone 340. Entrained catalyst is recovered and discharged via dipleg 414. Vapor passes via outlet 342, plenum chamber 350, and lines 352 and 360, to separator 370.

Separator 370, comprising one or more towers or flash drums (not shown), recovers a light product stream in line 372, and a heavy cycle oil stream in line 374, which combines with vacuum gas, oil in line 302, and passes via line 305 to riser 406.

Vacuum gas oil, from line 302 and a heavy cycle oil stream from line 374 pass via line 304 into a second riser 406. Hot regenerated catalyst is added via conduit 408 to the riser. Riser 406 discharges into a riser cyclone 410 in vessel 307. Cyclone 410 discharges catalyst via dipleg 414 to bed 316. Vapor via exits outlet 412 into vessel 307. Vapor exits vessel 307 via cyclone 340, which recovers entrained catalyst and passes it via dipleg 344 to bed 316. The vapor passes through outlet 342 into plenum 350, outlet 352, and line 360, into separators 370.

The catalyst in bed 316 passes down to stripper 318 and countercurrently contacts stripping gas from line 320 and header 322. Trays 334 enhance stripping. Stripped catalyst passes via outlet 330 to a regenerator Cnot shown).

Figure 4 shows use of an elutriable catalyst mixture used in conjunction with an elutriating riser, an elutriating stripper, a reactivation zone, a resid feed and fragment generation.

Riser reactor 410 receives C 3 and C 4 paraffins in lower region 411 through line 413 and stripped, reactivated catalyst via line 480 and valve 481. The stripped catalyst contains a lot of ZSM-5. Conditions in region 411 can be varied to maximize production of aromatics or light olefins, by varying the ZSM-5 content, and using fragment generation conditions previously discussed.

MI

i' 1.

.WO 88/01638 PCT/LUS87/02227 -29- A feed in line 401 is visbroken in visbreaker 402 and cascaded via line 415 into regiofi 412 of riser 410 via line 415. The feed combines with the ascending catalyst-hydrocarbon vapor suspension from region 4al.

Addition of hot, regenerated conventional catalyst from the regenerator via conduit 460 and valve 461 permits some control of catalyst composition in region 412 and also some control of the temperature. Preferably the zeolite Y concentration is 2 to 50, most preferably 5 to 25 wt The temperature can be 482 to 621 0 C (900 to 1150°F) and preferably 496 to 566 0 C (925 to 1050°F). The preferred catalyst to visbroken feed ratio is 3:1 to 20:1 and most preferably 4:1 to 10:1. The visbroken feed combines with reactive fragments and also cracks in the riser to lower boiling products. The riser discharges into cyclone separator 414 which separates catalyst from Ras.

Catalyst is discharged via dipleg 420 into bed 422. Vapor enters plenum chamber 416.

Descending catalyst bed 422 in an outer region of the stripper encounters stripping gas, steam, added via lines 427 and 428 which lifts the less dense particles of catalyst, the conventional catalyst, up concentrically arranged vertical lines 460 and 461 which lead to cyclones 470 and 471. These cyclones discharge stripped, conventional FCC catalyst, via conduitss 472 and 473, to a conventional FCC regenerator, not shown. A light olefin feed, a gas rich in ethylene and/or propylene, may be added to the bottom of bed 222 via line 250 to make higher molecular weight products and generate heat. The more dense particles, ZSM-5 additive catalyst, flow down via conduit 465 to be reactivated and in vessel 450 with H? from line 451 then returned to riser 410.

Although Figure 4 shows use of a dense, or less elutriable, additive it is also possible to reverse the relative settling rates and use an additive which is less dense than the conventional catalyst. In that case the additive will be removed overhead in the stripper.

In another embodiment, not shown, a dual riser FCC unit with a slightly different configuration is used. A conventional feed contacts a conventional catalyst (it may also contain a minor amount of additive rich in ZSM-5) from a conventional FCC regenerator in the base of a conventional FCC riser reactor which discharges into a cyclone separator within a vessel.

n WO 88/01638 PCT/US87/02227 In the second riser, an elutriable mixture of conventional catalyst and additive catalyst (rich in ZSM-5, and with a faster settling rate than the conventional catalyst) and a light H 2 -rich gas form reactive fragments in the base of the riser. The base of the riser preferably has an enlarged diameter lower portion, which results in lower superficial vapor velocities in the base of the riser than in the top of the riser. Preferably the settling velocity of the additive catalyst approaches the superficial vapor velocity in the bottom of the riser. This results in a longer residence time for the additive catalyst, rich in ZSM-5, in the base of the riser.

Additional hot regenerated catalyst, and activated heavy feed such as visbroken or microwaved resid, are added about half way up the second riser. The activated resid reacts readily with the reactive fragments generated in the base of the riser. The riser preferably discharges into a cyclone which removes vapor overhead and discharges catalyst via a dipleg into an elutriating catalyst stripper, such as that shown in Figure 4.

Conventional catalyst is lifted out of the central stripper by the stripping gas. The heavier, or less elutriable, additive catalyst passes down through the stripper and is recycled to the base of the elutriating riser, as previously discussed. The conventional catalyst, displaced from the central region by stripping gas preferably flows into an auxiliary stripper, and then to conventional regeneration.

In a preferred embodiment, not shown in the drawing, a single-riser FCC operates with multiple feed-point injection. Added to the base of the riser is a light, preferably olefins hydrogen-rich stream, which contacts hot regenerated catalyst to form reactive fragments. The fragment-catalyst mixture passes up the riser and contacts freshly visbroken resid. The resulting mixture passes up the riser and contacts additional hot regenerated catalyst. Finally, a conventional gas oil feed is added at about the midpoint of the riser. Preferably, a mixed catalyst system comprising a conventional FCC catalyst and a shape selective zeolite additive, such as ZSM-S, is used.

When an elutriatable additive is used, it is beneficial to provide an elutriating riser, i.e. an unusually wide portion of the riser at either the base or the top of the riser.

O 88/01638 PCT/US87/02227 -31- It is also beneficial, when an elutriable catalyst mixture is used, to provide an elutriating catalyst stripper. The catalyst stripper may separate additive from conventional catalyst by sieving or by relying on differences in settling velocity.

Catalyst may be reactivated prior to use. This is especially advantageous when a shape selective zeolite additive catalyst is used. To implement this, shape selective additive, after stripping, may be charged to a reactivation zone for contact with a reactivation gas such as hydrogen, prior to re-use in the FCC riser.

EXPERIMENTAL

The experiments are based on laboratory tests. The tests roughly simulated severe visbreaking immediately followed by catalytic cracking.

The severe visbreaking FCC process was simulated by heating the feed to 1100°F for 2-5 seconds and then immediately passing this over cracking catalysts at 960°F in a fixed-fluidized bed reactor. The test is reliable for predicting most product yields and octane number of product but is not as reliable for predicting coke yields. There is probably laminar flow in some portions of the preheater, and coking rapidly occurs in such conditions. In commercial units, this is not a problem because velocities through the furnace tubes are higher.

With these caveats in mind, the experimental apparatus feed, test conditions and results will be discussed in more detail below.

The activation pretreatment of the feed was conducted in the preheater. The preheater was a coil 385 cm long, by 1.0 cm ID, with a volume of 302 cc, heated by conventional means. Feed was added at 700 0

F,

and left the preheater at 1100°F when activation was needed, and 960°F when i conventional FCC operation was simulated.

The FCC reactor was a conventional, pilot plant fixed-fluidized bed reactor.

This fixed-fluidized bed reactor has proven reliable in the past at dirlicating what would happen in commerical units, which of course do not operate with fixed-fluidized beds, but instead have riser reactors.

The feedstock was an Arab light atmospheric resid with the following properties: L WO 88/01638 PCT/US87/02227, -32- Properties of Raw Arab Light Atmospheric Resid A.PT Gravity 20.0 IlPour Point, 0

P/

0 C 35/2 K.V. 401C, cs 108.5 K.V. 100'C, cs 11.46 Bromine Number 2.8 IiCCR, wt 6.64 Sulfur, wt 2.76 Hydrogen, wt 11.64 Total nitrogen, ppm 1300 Basic Nitrogen, ppm 370 Molecular weight 397 UNickel, ppm. 6.4 Vanadium, ppm 24 Iron, ppm 1.7 Sodium, ppm 4.3 Distillation 0D1160) OF p IBP 391 199 Vol 531 277 597 314 20664 351 715 379 773 412 s0 838 448 906 486 978 526 )YO 38/01638 PCT/LS87/02227 -33- The FCC catalyst, given F19260 used was a commercially available catalyst, and had the following properties: Chemical and Physical Properties of Fresh FCC Catalyst Mobil I.D. #F 19260 Chemical Analysis, Wt. Si02 A1 2 0 3 RE203 Na Physical Properties Packed Density, g/cc Surface Area, m2/g Pore Vol, cc/g By

N

2 Pore Size Distribution L.T. 30 Angstroms 30-50 50-100 100-150 150-200 200-400 400-600 Attrition Index Fresh Calcined Total Size Distribution, 0-20 microns 20-40 microns 40-60 microns 60-80 microns microns Median 62.3 32.1 0.2 0.89 188 0.15 11.5 30.7 18.5 7.3 5.4 13.8 12.8 13 8 21 1 11 29 29 67 microns 1 I I I~ I II WO 88/01638 PCT/US87/02227 -34- Before use, the catalyst was steam deactived at 788 0

C

(14500F) for ten hours in an atmosphere of 45% steam and 55% air to simulate commercial aging. The properties shown are for fresh catalyst, while what was actually used was steam deactivated.

i A ;i i ii ill x 88/01638 PCT/US87/02227 Experimental results are shown in the following table: FCC Cracking of Arab Light Atmospheric Resid Fixed-Fluidized Bed, 1.0 min. on-stream Laboratory deactivated FCC catalyst Invention Comparison Higher Higher Preheater Reactor Base Case Temperature Temperature Preheater, °F/oC 960/516 1100/593 980/527 Reactor, OF 960/516 960/516 980/527 Measured T mix, °F 955/5130 960/516 975/524 Conversion, vol 70 70 Delta 70 Delta C Gasoline, vol 52.4 50.2 -2.2 48.7 -3.7 Dry Gas, wt 9.0 11.8 +2.8 10.2 +1.2 Coke, wt 7.5 4.7 -2.8 7.7 +0.2 G+A, vol 78.7 81.2 +2.5 76.9 -1.8 RON+O, C 5 Gasoline 87.6 89.7 +2.1 89.6 LFO, wt 25.8 26.1 +0.3 25.2 -0.6 HFO, wt 6.2 6.7 +0.5 6.2 0.0 where: G C5+ gasoline A Alkylate made by alkylating olefins with isobutane RON+0 Research octane number of the gasoline LFO Light fuel oil HFO Heavy fuel oil The yields and octane, etc. reported above are based on the average of four runs at slightly different conversions. There were four runs of the IL I _no i W 0 88/0 1638 PCT/US87/02227 -36hase case, and the conversions were all around 70%. The data were graphed as a function of the conversion, and the results at 70% conversion used to generate the table. Four runs were also conducted at 1100 0 F preheater temperature, and four runs at 980 F reactor temperature.

These experiments show that a very high feed preheat temperature (5930C) 1100°F equivalent to a severe visbreaking operation increases gasoline plus alkylyate yields by 2.5 volume and gasoline octane by 2.1 octane numbers compared to the base case in which the feed was preheated at 960°F. The data also show a reduction in coke yield, by 2.8 wt The comparison test, with a higher reactor temperature, was designed to see if simply increasing the reactor temperature would improve a resid cracking process. In a sense it did, in that there was some increase in gasoline octane number (which would be expected from a higher reactor temperature), but there was a corresponding loss in gasoline yield. This confirms what hapnens in most FCC units, that it is possible to increase the gasoline octane, but only at the expense of gasoline yield.

The severity of visbreaking at 593 0 C, 1100°F in the pretreater can be estimated. Based on the residence time of the liquid feed in the pretreater of 2-5 seconds. The reaction severity was 606 to 1515 ERT.

1 Conventional correlations used to calculate reaction severities in visbreakers and cokers were used, however, the temperature in the preheater was quite a bit higher than that in conventional visbreaking operation. It is also difficult to accurately calculate the residence time of small amounts of liquid in a small preheater operated at such a high temperature.

Accordingly, there is a good deal of uncertainty about the severity of the pretreatment step.

To check on the validity of the coke yields, the reactor preheater was "regenerated" by burning off the carbon in the preheater and absorbing D CO 2 in the flue gas, in a scrubber. From the CO 2 yields, the amount of coke deposited in the preheater could be determined, and the plant material balance could he re-calculated to account for this. When this was done, as shown in the following table, the yields changed. The G+A yields are now about the same whether the preheat temperature is 960°F or 1100°F, and coke yields get worse with an 593 0 C (1100°F) higher preheat temperature. The I i_ i .WO88/01638 PCT/US87/02227 gasoline octane numher advantage from resid activation remains. The octane increase is significant. Because bf the uncertainties about coke deposition in the preheater, we have not proven if the practice of the invention, when applied to visbreaking before FCC, also leads to increased yields of gasoline plus alkylate and reduced coke make. This probably happens, hut the experiments do not unequivocably prove this.

I'

I

WO 88/01638 PCT/US87/02227 YIELD ADJUSTED FOR PREHEATER COKING Fixed-fluidized hed reactor, 960 0 F, 4 C/O, 1.0 min on-stream Arab Light atmospheric resid Preheat Temp., "F/°C Carbon-on-cat., g Carbon in preheat, g Total carbon, g 960/516 Initial Adjusted 3.244 3.584 0.34 3.584 1100/593 Initial Adjusted 2.650 4.530 1.88 4.530 Conversion, vol%

C

5 Gasoline, vol% Total C 4 vol% Dry Gas, wt% Coke, wt% RON+0, C5 Gaso.

G+A, vol% Recovery, wt% 72.1 49.5 16.5 9.7 8.6 88.1 76.2 93.5 72.3 49.1 16.3 9.6 9.4 88.0 75.6 94.4 73.0 46.7 17.0 14.2 7.1 90.6 79.8 92.8 74.3 44.5 16.2 13.5 11.5 90.6 76.0 97.5 Adjusted Yields are recalculated from initial yields to reflect coke deposition in the preheater.

I

I

r 5 -e WO 88/01638 PCT/US87/02227 -39- The above yields and weight balances are based on a single test (Of 960F) compared to a single test (at 1100°F). The data could be adjusted to a constant 70% conversion, but this was not done.

These experiments show that activation pretreatment of the feed improves octane number a significant amount (over 2.0 research octane numbers). It is difficult to understand or rationalize why the invention works. Visbreaking is normally a process of last resort in a refinery, used only for feeds which are so heavy that they cannot profitably be processed in, catalytical cracking units. That visbreaking a heavy feed just before subjecting it to the cat cracking would increase the octane number of the FCC gasoline by two or more octane numbers is a surprise.

It is surprising that in a mature process such as FCC that it is possible to achieve an increase in gasoline octane number by thermal pretreatment of the resid feed.

Commercially, it would be fairly easy to practice the invention, especially the visbreaking-FCC embodiment. A mixture of hot, freshlv visbroken feed and conventional feeds, such as gas oils, could be fed to the FCC. The reason for using a feed mixture is because most conventional FCC units could not accommodate a 100% resid feed, because such feeds. Resids make too much coke. Adding as much visbroken feed as the unit would tolerate to conventional FCC feed would avoid this problem. Probably, quite a lot of visbroken feed can be accommodated in conventional FCC units both because of the anticipated (but unproven) reduction in coke yields expected from severe thermal pretreatment, but also because increasing the feed preheat temperature will reduce the coke make reauired to keep the unit in balance. This means that less catalyst circulation is needed to keep the riser top temperature constant, and reduced catalyst circulation reduces apparent coke make.

The process of the present invention will work especially well in an FCC unit with additive catalysts, such as ZSM-5, which make very little coke, in addition to conventional FCC catalyst. Use of a catalyst with perhaps 5-10% conventional, large pore zeolite and 5-10 wt preferably 2-10 wt ZSM-5 zeolite may be the ideal catalyst for use in upgrading 7 WO 88/01638 PCT/US87/02227 severely visbroken feed. Elutriating risers, elutriating strippers, etc.

which can increase the residence time of ZSM-5 additive in the riser and minimize the residence time of the ZSM-5 in the regenerator could also be beneficial. Although the invention has been discussed in relation to the combination visbreaking-FCC, it can be used in many other ways. Activation pretreatment of resid will render resid more amenable to conventional hydrotreating and hydrocracking processes. Alternative pretreatment procedures can also be used, tunable microwaves, lasers, and the like, provided they make resid feed as reactive as a corresponding thermal pretreatment.

I

i i

Claims (4)

1. A process for upgrading a heavy, hydrogen deficient feed comprising: subjecting the feed to at least one of liquid phase visbreaking at a visbreaking severity of 500-2000 ERT seconds and microwave treatment and cascading the visbroken or microwaved feed, without intermediate storage thereof, into a catalystic processing zone to produce a catalytically upgraded product. -6h4th~.

9. 9 S hvdrog prior subsec catal' there( upgrac compr feed compr -c-OMP- catal hydro catal *too 0 9 a. S 9. *49 that ERT s unit, s-. Afl- A i (7 WO 88/01638 PCT/US87/02227 -41- A-eed- c.t-i-vaton-caa--y-t-ic-p--ocess for upgrading a heavy,- hydrogen deficient fe d comprising: subje ting the heavy, hydrogen deficient feed to treatment prior to catalytic, which\ increases the re-activity of the feed for subsequent catalytic upgr ding by adding energy to the feed orior to catalytic treatment passing th activated feed, without intermediate storage thereof, into a catalytic pr essing zone to produce a catalytically upgraded product 2. The process of cl im 1 wherein the feed pretreatment process comprises thermal activation, wi a severity of at least 500 ERT seconds. 3. The process of claim l or 2 further characterized in that the feed pretreatment process is visbre king conducted at 500-2000 ERT seconds. 4. The process of claim 1 wherein the feed activation treatment comprises microwaving the feed. The process of claim 1 wh rein the feed activation treatment -compri ses irradiating the Feed with lsp 2- The process of claim 1 wherein the catalytic treatment is catalytic cracking. 3 The process of claim 1 wherein the catalytic treatment is hydrotreating. 4 X' The process of claim 1 further characterized in that the catalytic treatment is hydrocracking. The process of any preceding claim further characterized in that the feed is a resid. JO. A process for upgrading a resid by I visbreaking the resid at a reaction severity of 500-1000 ERT seconds, cascading the visbroken resid into a catalytic cracking unit, and mm__ m SWO 88/01638 PCT/US87/0222 7 -42- recovering cracked products including gasoline with an octane number at least 1.0 research octane number higher than could be obtained without absent the visbreaking feed pretreatment. The process of Claim 4 further characterized by use of a riser reactor having an enlarged base portion whereby elutriation of catalyst occurs in the base of the riser reactor. 7 A 3 The process of Claim further characterized by use of an elutriable mixture of a conventional FCC catalyst and a shape selective additive in an FCC unit with a catalyst stripper. i7 A The process of Claim/I4 further characterized in that the i riser reactor has an enlarged portion near the outlet of the riser whereby I catalyst elutriation occurs in the top of the riser. 7 3 Io 10 14. The process of any of ClaimsL"r to I-5 further characterized by elutriating catalyst in the catalyst stripper. to I j I IJ The process of Claim 4ifurther characterized in that the additive catalyst has a different size than the conventional catalyst and the stripper acts as a sieve to separate additive from conventional catalyst.

12- 1- The process of Claim yS further characterized in that the stripper operates with a superficial vapor velocity sufficient to effect a separation between additive catalyst and conventional catalysts based on different settling rates. -7 k S13 1 The process of any of Claims J- to J] further characterized by reactivation of the additive catalyst with hydrogen or hydrogen containing gas in a reactivation zone intermediate the catalyst stripper and the riser reactor. -7 3 1 18-. The process of any of Claims i4 to 1 further characterized in that conversion of light olefins to gasoline and distillate occurs in the base of the riser reactor. The process of any preceeding claim further characterized in that regenerated catalyst and a light, H 2 rich hydrocarbon is also added to the base to generate reactive fragments of the riser of the FCC reactor. i i- n -43-

16. A process substantially as hereinbefore described with reference to any one of Figures 1 to 4. DATED: 3 September, 1990 MOBIL OIL CORPORATION By their Patent Attorneys: PHILLIPS ORMONDE FITZPATRICK q t~ 41 5558m a* a a a. -4 a