+ All Categories
Home > Documents > 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both...

6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both...

Date post: 10-Aug-2018
Category:
Upload: lekhue
View: 231 times
Download: 1 times
Share this document with a friend
28
6 CATALYTIC CONVERSION PROCESSES
Transcript
Page 1: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

6

CATALYTICCONVERSIONPROCESSES

Page 2: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the
Page 3: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

6.1.1 Principles and development

IntroductionAfter about 70 years of activity, at the beginning of

the new millennium catalytic cracking still remains theprincipal process used to convert heavy oil fractionsinto lighter products, especially gasoline.

Historically, the distinction between simple-cyclerefineries and conversion refineries is based on theabsence or presence of catalytic cracking in theproduction cycle.

The first true catalytic process in the refiningindustry, cracking is still one of the most importantcatalytic processes in the petrochemical industry, bothin terms of plantsize and the amount of catalyst.

The advent of cracking has significantlycontributed to the understanding of the acid catalysismechanisms that leads to the formation ofcarbocations starting from hydrocarbon molecules.

Compared to its predecessor (i.e. the thermalcracking process), catalytic cracking presentsnumerous advantages, including higher gasoline yields(�50% in weight with respect to the feed), the higherquality of the gasoline produced and a lowerproduction of gaseous, liquid and solid by-products(coke). The gaseous fractions can be used as feed foralkylation processes (see Chapter 4.3), for theproduction of methyl tert-butyl ether or MTBE (seeChapter 4.4) and for polypropylene plants; heavyliquid fractions (cycle oil) are excellent feedstocks forthe production of carbon black (Fig. 1). The use ofalkylates and ethers has been encouraged, from the lastdecade of the Twentieth century onwards, by thereduction of the aromatics and benzene content ofcommercial gasolines.

Typical feedstocks for catalytic cracking are thehigh boiling distillates obtained from vacuum

distillation, and deasphalted or hydrogenatedresidues.

The most recent developments in the process alsoallows the partial feed of atmospheric residues, albeitmixed with the distilled feedstock, since the processtakes place in the vapour phase; moreover, residuesdeactivate the catalyst more rapidly.

Thanks to its versatility and capacity forcontinuous renewal and development, catalyticcracking has long withstood ‘competition’ from otherconversion processes, especially hydrocracking (see Chapter 6.2).

It cannot be ruled out that cracking’spredominant role among catalytic processes willbe downscaled, due to changing marketrequirements (lower demand for gasoline withrespect to other fuels), the need to obtain sulphur-free products directly, and to the establishment ofprocesses capable of converting residues directly.However, catalytic cracking will remain afundamental process in the refining industry formany years to come.

Development of the processesDespite some earlier attempts to improve the

thermal process with the addition of varioussubstances (which cannot always be described ascatalysts), it was only in the 1930s that catalyticcracking became commercially important, thanks tothe work of Eugène Houdry (see Chapter 1.1).

The first unit, equipped with three fixed bedreactors, came on line in the United States in 1936; thecatalyst consisted of a natural clay based onmontmorillonite. In the same year, the first plant wasbuilt to supply ‘activated earths’ (with acid) to thecatalytic plants and, in 1940, the Houdry Corporationstarted up a plant for the production of syntheticaluminium-silicates.

247VOLUME II / REFINING AND PETROCHEMICALS

6.1

Catalytic cracking

Page 4: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

The fixed bed process was difficult to manage atthat time since the three reactors alternated reactionphases with regeneration phases, with intermediatepurging. The process remained complex and demandingdespite an increase in the number of reactors (six) andthe introduction of electrical cycle timers to control theopening and closing of all the valves in the variouscircuits (oil, vacuum, air and steam). Motivated by thewar, experiments began to introduce moving bedplants (TCC, Thermofor Catalytic Cracking) and fluidbed plants (FCC, Fluid Catalytic Cracking), whichcame on line in the United States almostsimultaneously in the years 1942-43.

In moving bed reactors, the catalyst was initiallymoved using mechanical bucket elevators, andsubsequently with air; this allowed for the continuousregeneration of the catalyst, leading to improved yieldsand product quality. The same benefits were obtainedby fluidizing the bed with the vaporized feed (in thereactor) and with air (in the regenerator).

The catalyst consisted of spheres approximately 3 mm in diameter, and microspheres (powder) for theTCC and FCC processes, respectively.

After competing for several decades, the fluid bedtechnology supplanted the moving bed process so that,already at the end of the Twentieth century, TCC unitswere extremely rare.

The FCC process, in turn, has undergone continuousdevelopment over the years, maintaining it constantlyup-to-date. One of the most important developmentswas the introduction in the early 1960s of zeolitecatalysts (able to ‘select’ the reacting molecules), whichconsiderably revitalized the process.

The improved efficiency and stability of catalystshas led to the elimination of the traditional reactor andthe introduction of the riser reactor; more efficientcatalyst regeneration systems have also been developed.A further step forward was made possible thanks tocatalysts, again zeolites (ZSM-5), able to improve theoctane number of the gasolines produced, and to theintroduction of a series of new additives and passivators.

Catalytic cracking reactionsThe relatively high process temperatures

(�450°C) lead to the formation of free radicals andto thermal reactions. These reactions have lowselectivity and produce light gas molecules, such asmethane and ethane, and lead to the formation ofolefins.

Although the latter may be precursors to theformation of carbocations, thermal reactions should belimited by operating at temperatures which are as lowas possible.

Catalytic cracking reactions includeisomerization, the b-scission of paraffins,dehydrogenation, hydrogen transfer and varioustypes of condensation reactions. The main reactions,according to the various classes of hydrocarbons, aresummarized in Table 1.

Catalysts of acid type promote the formation ofcarbocationic intermediates rather than free radicals,improving yields and selectivity.

Carbocations may form starting from an olefin,in the presence of Brönsted acid sites in thecatalyst, or by the protonation of a paraffin ornaphthene:

The first of these mechanisms is universallyaccepted and, in comparison with others, issignificantly faster. However, the hypothesis thatcarbocations may also be formed starting from

248 ENCYCLOPAEDIA OF HYDROCARBONS

CATALYTIC CONVERSION PROCESSES

carbon black

propylene

alkylatednaphtha(C7-C8)

MTBE(or other ether)

crackingnaphtha

heavydistillate

feed

CH3OH(or other alcohol)

aromatic oil

gas

Fig. 1. Processesdownstream ofcatalytic cracking.The most commonconfigurationinvolves alkylationimmediately aftercracking.

R1 C R1 C C+ R2C R2�H�

H H H

H H

R1 C C R2�H+ R1 C C R2

H H H H

H

H

HH

H

R1 C C+ R2�H2

H

H H

Page 5: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

Lewis-type sites, present on the catalyst togetherwith Brönsted sites, is also commonly accepted:

Since they are deficient in electrons, Lewis-typesites can stabilize one of the hydrogens in the H�

form, and form the complementary carbocation.The carbocations that form on the surface of the

catalyst tend to isomerize towards the more stableform (from a primary to secondary to tertiarycarbocation); in the latter state, the carbon containingthe charge is linked to three other carbon atoms. Withreference to a paraffin chain, after the formation of thecarbon ion, there are various possibilities. The first isisomerization towards a more stable form; the second,endothermic, involves the rupture of the C�C bondin the b position with respect to the charge, forming anolefin and an unstable paraffinic carbocation, whichsubsequently isomerizes:

The probability of b-scission increases if theconfiguration of the original carbocation is favourable

(tertiary or secondary, rather than primary). There arealso other possibilities: the carbocation frees a protonand turns into an olefin, or saturates by taking a protonfrom the catalyst’s active site, or reacts with an olefinto alkylate it.

Olefins behave in a comparable way, with thedifference that they crack much faster, given theirhigher tendency to form carbocations; however, theymay also oligomerize and cyclize, contributing,alongside aromatics, to the formation of coke.

The b-scission mechanism leads to a preferentialrupture of the bonds inside the molecule; non-condensible gases such as methane, ethane andethylene, which would be formed by the rupture ofterminal bonds, are thus only present in smallquantities, in contrast to what occurs in thermalprocesses. The olefins that form have 3 or 4 carbonatoms and are excellent feedstocks for the processesdownstream (see again Fig. 1).

However, the formation of olefins is on averagelower than that predicted by the mechanisms describedabove. This is due to exothermic reactions involvingthe transfer of hydrogen from cycloalkane donormolecules to unsaturated molecules, with theformation of aromatic compounds and paraffins.

This reaction is probably as important as therupture of the naphthene ring, with the formation ofisoalkanes.

The reactivity of the naphthene ring increases withthe degree of substitution, in other words the potentialfor forming tertiary carbocations on the ring. Only thelonger side-chains are broken; the methyl and ethylgroups are generally unaffected, given the high

249VOLUME II / REFINING AND PETROCHEMICALS

CATALYTIC CRACKING

R1 C C R2�L R1 C C+ R2�LH−

H H

H

H

HH H

R1 C+

H CH2 CH2 CH2 CH2 R2

R1 CH CH2�R2 CH2 CH2 C+

H2

R2 CH2 CH2 C+H2 R2 CH2 C

+H CH3

R2 C+H CH3

CH3

Table 1. Main cracking reactions for various classes of hydrocarbons

Hydrocarbons Schematic structure Main reactions Main products

ParaffinsRupture of the molecule in different points: difficult rupture of C�C bonds

Paraffins and olefins with at least 3 or 4 C atoms;small quantities of methane

NaphthenesRupture of the ring and side chains

Paraffins and olefins; small quantities of aromatics

Naphthene-aromaticsOpening of the naphthene ring;rupture of the side chains of the aromatic rings

Paraffins, olefins and aromatics

Aromatics without side chains

Negligible cracking Coke

Aromatics with side chains

Rupture of side chains Olefins and aromatics

OlefinsRupture of the chains at various points

Branched olefins; paraffins; diolefins

Page 6: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

formation energy of the corresponding ions. In thiscase, a ring with 5 carbon atoms may isomerize to themore reactive ring with 6 carbon atoms.

The above discussion also applies to thedealkylation of aromatics, whose ring is extremelystable and is not ruptured; however, it may beinvolved in condensation reactions with theformation of coke. Methyl aromatics may undergodisproportionation; thus, benzene and xylene can bemade from toluene.

The formation of cokeThe term coke is used to describe the material

deposited on the catalyst during the process, andoxidized during the regeneration phase, producing theenergy needed for the cracking reactions. It consists ofa series of components with a high carbon content(�90%), mainly in the form of condensed aromaticrings. Its composition depends on the type of feed, thecontent of contaminants (such as V, Ni, and Fe, whichcatalyze the dehydrogenation reactions), the nature ofthe catalyst and the operating conditions. Like itscomposition, the mechanism of coke formation iscomplex and involves cyclization andpolycondensation reactions starting from precursorssuch as olefins, diolefins and aromatics. Thesereactions, though not encouraged by the conditionsadopted in current processes (high temperatures andlow pressures), may occur anyway. The polyaromaticcompounds formed are resistant to cracking, andgradually accumulate in the heavy liquid fractions andon the catalyst.

CatalystsAs seen above, the catalytic cracking mechanism

involves the formation of carbocations and is activatedby acid functions.

In the earliest processes, the catalysts wereessentially natural clays (aluminium-silicates) activatedwith an acid treatment and then calcinated. Thistreatment had the aim of creating acid centres,responsible for catalytic activity, by replacing thealkaline and alkaline-earth ions saturating the negativecharges.

The clay most widely used was montmorillonite,which can be described with the general formula:

Si8Al4 O20(OH)4 �nH2O

Silica and alumina, taken separately, do not haveacid properties; however, if the alumina is dispersedwithin a silica matrix, strong acidity can be observed.Silica consists of SiO4

�4 tetrahedrons; substituting asilicon atom with an aluminium atom is accompaniedby the formation of a negative charge, which must bebalanced; if this is done by a proton (rather than, for

example, a sodium ion), the result is a strongly acidicmaterial:

Nevertheless, silica-alumina based catalysts havetwo types of acidity linked to the aluminium atom:Lewis acidity, characteristic of a tricoordinate Al(capable of acquiring a pair of electrons to form thestable octet), and Brönsted, or protonic acidity. Duringheating, protonic acidity tends to become Lewisacidity; this, in turn, tends to turn into protonic aciditydue to the action of small quantities of water.

The need for greater control over chemicalcomposition and morphology led, as early as the 1930s,to the appearance of synthetic catalysts (silica-aluminagel), obtained using spray drying processes; these had amore regular physical form and improved performance.

However, the genuine revolution occurred duringthe mid-1960s with the introduction of catalysts basedon zeolites (faujasite).

Unlike the natural aluminium-silicates previouslyused, which were amorphous, zeolites are porouscrystalline materials whose properties are governedand defined principally by their chemical compositionand crystalline structure, consisting of a three-dimensional grid with regular pores. Chemically, theseare also aluminium-silicates with negative charges onthe tetrahedrons [AlO4]

�; therefore, they must containpositive external M ions (e.g. H�, Na�, K�, Ca��,Mg��, etc.) to balance the charge of the anions. Thesematerials can be described with the general formula:

Mu(AlO2)x(SiO2)y�zH2O

If M is a monovalent positive cation, u�x; if it isbivalent, u�x/2.

The cations are positioned near the anions in thecrystal tunnels, whose diameters range, in naturalminerals, from 0.25 nm (sodalite) to around 0.8 nm(faujasite).

Although zeolites exist as natural minerals,synthetic products are now used for catalysis.

Zeolites are acid catalysts, but are often consideredseparately, given their unusual properties and, in particular,their ability to carry out shape-selective catalysis: the sizeand shape of the internal cavity determine those of theproducts, whereas the diameter of the pores determines thetype of molecules which can enter them.

250 ENCYCLOPAEDIA OF HYDROCARBONS

CATALYTIC CONVERSION PROCESSES

Si O Al O Si

O

O

Si

Si

H�

Page 7: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

Zeolites are often classified on the basis of the Si-Al ratio (Table 2); this ratio is equal to or greater thanone.

The fact that the [AlO4]� group in zeolites of the

Linde ‘A’ and ‘X’ types is an unstable site to acidattack, in the presence of steam at high temperatures,created the basis for the synthesis of products with alower Al content.

The Linde ‘Y’ zeolite, with a Si/Al ratio ofbetween 1.5 and 3.0, and the same structural andskeletal typology as Linde ‘X’ (and identical to therare natural zeolite faujasite), was introduced in 1964and immediately became the preferred product forindustrial processes.

The basic unit of faujasite is the sodalite cage,consisting of 24 (SiO4) or (AlO4) tetrahedrons;depending on how the basic units are joined, twodifferent structures can be obtained: the Linde ‘A’ typezeolite and the Linde ‘X’ or ‘Y’ type zeolite (Fig. 2). Inall cases, a three-dimensional grid ofintercommunicating tunnels containing larger cavities(or cages) inside is obtained; the diameter of the poresis determined in part by the type of cation, whichbalances the structure’s negative charges. These arethus highly porous structures, within which gasmolecules can circulate.

The linear dimensions of the broadest cages areabout 1.15-1.20 nm; the ‘entrance windows’, however,are extremely small in Linde ‘A’ type zeolites, andwider in Linde ‘X’ and ‘Y’ type zeolites (see againTable 2 and Fig. 3); this property makes it possible touse Linde ‘A’ type zeolites as molecular sieves forsmall molecules, whereas Linde ‘X’ and ‘Y’ typezeolites, which allow larger molecules to pass throughthem (naphthenes, branched hydrocarbons etc.), areideal for the cracking process.

When the zeolites are synthesized, the negativecharges are neutralized by sodium ions, which are thenexchanged with ammonium ions or the ions of rareearths (Ce�3, La�3); during calcination, the NH4

� ionsare turned into ammonia and H� ions, which createthe necessary protonic acidity, while the rare earthscontribute to both Brönsted-type and Lewis-typeacidity:

The exchange with rare earths confers greateracidity; the result is an increase in conversion and adecrease in octane quality (Fig. 4).

251VOLUME II / REFINING AND PETROCHEMICALS

CATALYTIC CRACKING

Table 2. Classification of zeolites on the basis of the Si/Al ratio

Si/Al Ratio Types Origin Å Properties

Low (1.0-1.5)Linde ‘A’Linde ‘X’

SyntheticSynthetic

3-57-8

Unstable at high T and to acid attackLow resistance to steam

Medium (1.5-5.0)FaujasiteLinde ‘Y’Mordenite

NaturalSyntheticSynthetic

7-87-86-7

Rare in natureLike faujasite but more stable; for cracking and isomerizationDifferent structure

High (6-100)ZSM-5Silicalite

SyntheticSynthetic

5-65-6

Highly selectiveBasically siliceous, hydrophobic

O O

O O

O�

O�

O�

H�

O

O

O

O

O O

O

Ce (OH)

O O

O

O

Si Al Si Si Si Al SiAl

Brönsted-typeacid site

Lewis-typeacid site

Fig. 2. Cubic-octahedralunit of sodalite (A),structure of Linde ‘A’type (B) and ‘X’ and ‘Y’type zeolites (C).

A

B

C

Page 8: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

Increasing the Si/Al ratio leads to greater thermalstability and a higher resistance to steam, fundamentalproperties for cracking catalysts given the significantuse of steam (stripping during the transitions fromreactor to regenerator and vice versa) and the hightemperatures in regeneration.

Ultrastable zeolites, up to 1,000°C (such asUltraStable ‘Y’ zeolites or USY) are also made with ahydrothermal treatment, moving Al towards positionswhich are no longer reticular.

Table 3 provides an indication of the selectivity ofsome types of Linde ‘Y’ zeolites. The need to favourquantity, rather than the octane quality of the gasolineproduced, leads to the selection of zeolites containingrare earths. To increase the Octane Number (ON),alongside the usual Linde ‘Y’-type zeolites, smallerquantities of ZSM-5 zeolites (pore diameter around0.5 nm) can be used, more selective for linearcompounds; these are in part isomerized and in partcracked, eliminating them (in the form of gas) fromthe gasoline fraction, but at the expense of gasolineyields.

Industrial catalysts are composed of crystallinezeolites dispersed in an amorphous matrix that acts asa binder, as a vehicle for diffusion and as a heatdisperser; this also carries out the precracking of thelarger components, and thus prepares the moleculesfor the zeolite. The matrix contains various

252 ENCYCLOPAEDIA OF HYDROCARBONS

CATALYTIC CONVERSION PROCESSES

6.1.1 Giavarini fig 03

H

H innercage

(12 nm)

0.7-

0.8

nm

1 1

2

23

3

4

4

Fig. 3. Structure of synthetic Linde ‘Y’ faujasite with the formula Na56[(AlO2)56(SiO2)136]�250 H2O,giving a partial indication of the acid sites, the positionof the oxygen (yellow circles) and the non-reticular cations (red circles).

92 70

60

50

91

89

90

0.5 1.0 1.5 2.0

Res

earc

h O

ctan

e N

umbe

r (R

ON

)

conv

ersi

on (

wei

ght %

)

conversion

RON

rare earth (%)

Fig. 4. Effect of the rare earth content on conversionand the Research Octane Number (RON).

Table 3. Selectivity of some Linde ‘Y’-type zeolites to different products

USY � ultrastable Y zeoliteREUSY � ultrastable Y zeolite containing rare earths

REHY � Y zeolite containing rare earths and hydrogenREY � Y zeolite containing rare earths

Product USY REUSY REHY REY

Saturated C3-C4 High Moderate Moderate Low

C3-C4 olefins High Moderate Moderate Low

Coke/conversion Very low Very low Low Moderate

Gasoline Moderate High High High

Octane yield High Moderate Low Low

Activity vs feedstocks

340-480°C High High High High

�480°C Moderate Moderate Low Low

Page 9: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

components: cohesion between the microspheres ofthe catalyst is ensured by silica-alumina gel; kaolin ora similar material has the function of dispersing heat(during regeneration) and acts as macroporousmaterial for diffusing and precracking the reagents.

The use of selective matrices (SAM, SelectiveAlumina Matrix) is especially important in processesthat are also fed with a significant quantity of residue(30-50%) in addition to heavy distillates (O’Connor etal., 1991), as shown in the example given in Fig. 5.

Various additives are dispersed in the matrix,particularly alumina, with the aim of promoting thecracking activity of specific catalysts destined for thetreatment of particularly heavy feedstocks; otheradditives can be added as separate microspheres.

Among the additives, it is worth mentioningcombustion promoters (from CO to CO2); aluminiumand magnesium compounds to fix sulphur oxides in theform of sulphates (in the regenerator), releasing them asH2S in the reactor and in the strippers; and passivatorsto neutralize the action of vanadium and nickel.

Fig. 6 shows the structure of a typical crackingcatalyst.

The structure of cracking catalysts makes it easy tosee how these can be deactivated by potentially basiccompounds (Na or compounds containing nitrogen)and by the metals present in the feed, which in the longrun lead to permanent deactivation.

Specifically, vanadium is oxidized to V2O5 in theregenerator and may form vanadates of rare earths, inaddition to catalyzing dehydrogenation (undercracking conditions).

Nickel is a far stronger dehydrogenator thanvanadium, and thus encourages the formation of coke;the addition of organometallic compounds of antimonyand bismuth partially neutralizes this effect with theformation of intermetallic compounds.

The formation of coke, which, as seen above,originates from various secondary dehydrogenation,condensation and polymerization reactions, leads tothe gradual obstruction of the active centres of thecatalyst, which loses activity and must therefore beregenerated by the combustion of the deposit andrecycled into the reactor.

In the cracking process, the catalyst performs atleast three functions (see below): activator for crackingreactions, support for coke, heat transporter; all occursunder extremely severe conditions since itcontinuously circulates in high friction zones (fluidbed) and high velocity zones (cyclones, riser: 20-30m/s). Furthermore, every 8-10 minutes, the catalystpasses from a reducing atmosphere at about 500°C(riser-stripper) to an oxidizing atmosphere at700-800°C (regenerator).

Thermodynamic aspectsAn accurate thermodynamic analysis of catalytic

cracking would require the acquisition of data on thehydrocarbons and other compounds contained withinthe heavy fractions that feed the process; this isimpossible. It may be useful to refer to some typical,simpler hydrocarbons in order to roughly define thethermodynamic aspects of the process.

253VOLUME II / REFINING AND PETROCHEMICALS

CATALYTIC CRACKING

6.1.1 Giavarini fig 05

H�

asphaltene-type molecule

large porecomponent of the

catalyst matrix

zeoliticcomponent

H�

i-C4, C5, C4=, C3

=, etc.

Fig. 5. Model showing the stages of the partial cracking of an asphaltene molecule. Only the side chain,broken by the acid components with wide pores (�100 nm)of the matrix, comes into contact with the zeolitecomponent, which carries out further cracking.

10-40% 60-90%

Fig. 6. Typicalcompositionof a FCC catalystfor the production ofhigh ON gasoline.The octane promotersand additives can beincorporated inindependentmicrospheres.

Page 10: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

Table 4 (Raseev, 2003) shows the heats of reactionand entropy variations for some typical crackingreactions involving paraffins, olefins, cycloalkanesand aromatics under typical cracking conditions (i.e.about 500°C and pressures slightly above atmosphericpressure).

The differences between the various possiblereactions of an individual hydrocarbon series are notsignificant. With the exception of secondarycondensation and polymerization reactions (whichare not desired, but occur anyway) andisomerizations, typical cracking reactions (i.e. therupture of carbon-carbon bonds anddehydrogenation) are all endothermic.

Isomerizations are weakly exothermic (∆H between�4 and �20 kJ/mol).

Overall, the cracking process is thus endothermic,with values of the reaction ∆H moderately influencedby the type of catalyst, and generally in the range of900 to 1,000 kJ/kg (Pekediz et al., 1997).

Approximate thermodynamic calculations, basedon simplified expressions of free energy, show that thebalance which leads to a paraffin and an olefin startingfrom a generic paraffin

Cm�nH2(m�n)�2����CmH2m�2�CnH2n

moves to the right at temperatures above 300°C(Giavarini, 1999).

254 ENCYCLOPAEDIA OF HYDROCARBONS

CATALYTIC CONVERSION PROCESSES

Table 4. Values of DH° e DS° at 800 K, for some typical cracking reactions (Raseev, 2003)

Reaction DH°800 K (kJ/mol) DS°800 K (kJ/mol)

79.13 140.50

79.09 139.70

77.79 140.49

77.75 143.18

62.24 66.71

84.10 94.90

77.25 141.53

72.90 140.50

220.29 402.58

90.71 126.40

79.71 138.27

78.58 135.47

�77.71 �143.08

C6H14 C3H8�C3H6

C20H42 C3H8�C17H34

C6H12 2C3H6

C20H40 2C10H20

C6H12

C5H10

C3H6�

C4H9 CH3

�3H2

C3H6�C4H9 CH3

2C8H16 C16H32

C5H11

�C4H8

CH3

�C3H6

C3H7

�C4H8

C5H11 CH3

Page 11: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

Industrial processes must therefore operate athigher temperatures (above 450°C), at which it canbe assumed that the cracking reactions arecomplete.

A similar calculation, referring to the transitionfrom cyclohexane to benzene, gives a temperatureabove 550°C.

As such, under the conditions adopted in industrialprocesses (470-520°C and pressures slightly aboveatmospheric pressure), thermodynamic aspects limitthe decomposition of alkanes to alkenes in the gasphase.

The same can be said for the rupture of the sidechains of naphthene and aromatic rings, whoseconversions are determined by the relative kinetics.Again in the gas phase, the cracking of naphthenerings, which is less favoured thermodynamicallythan that of aromatic rings, preferably occurs inrings with 5 terminals, whereas in rings with 6terminals dehydrogenation is favoured(Raseev, 2003).

The decomposition of non-cyclic heteroatomiccompounds (which also contain sulphur, nitrogenor metals) does not have thermodynamiclimitations.

If the aim is to obtain isomers with a highoctane number (exothermic reactions), the processshould take place at the lowest possibletemperatures, compatible with cracking reactions(endothermic).

It is more difficult to describe the reactions thatoccur on the chemiadsorbed layer in the catalyst andlead to the formation of coke. Under processconditions, the polymerization of alkenes and othercondensation reactions are not favoured: these cantake place only in the liquid phase or on the surfaceof the catalyst; in the gas phase, they can occur onlyat very high pressures (Raseev, 2003).

The formation of compounds with a highmolecular weight (by polymerization, condensation,dehydrogenation) takes place inside the cages andpores, where they cannot be desorbed (given their

size) from the catalyst, leading to the formation ofcoke.

In the regenerator, the coke deposited on thecatalyst is oxidized to reactivate it and to supplyheat to the process. The heat of combustion of thecoke depends on its hydrogen content and theCO2/CO ratio in the combustion gases; the H2content generally falls within the interval 4-10%.Table 5 (Raseev, 2003) shows the heats ofcombustion of cokes with different H/C ratios anddifferent CO2/CO ratios in the fumes.

The thermal balance, based on two oppositeenergy exchanges, depends mainly on the quality ofthe feed: for ‘easy’ feeds with a low Conradsoncarbon content, such as vacuum gas oils, the cokeyield may be insufficient to balance therequirements of the unit, whereas for residues, theenergy produced by the regenerator is excessive andsome of it must be disposed of.

The heat Qcat (kJ/h) transferred to the catalyst inthe regenerator can be expressed by the equation(Bonifay and Marcilly, 2001):

Qcat�∆coke �Wcat �Qcoke �R

where: ∆coke (kg of coke/kg of catalyst) is thedifference, referring to the unit of weight of thecirculating catalyst, between the weight of the cokedeposited on the catalyst (from the reactor) and theweight of the residual coke on the catalyst (from theregenerator); Wcat (kg of catalyst/h) is the weight flowrate of the catalyst in circulation; Qcoke (kJ/kg of coke)is the combustion heat of the coke; R is the efficiencyof combustion in the regenerator, in other words, theratio of heat absorbed by the catalyst to heat producedby the combustion of the coke. Qcat is an extremelyimportant parameter that depends on the properties ofthe feedstock and those of the catalyst. For heavyfeedstocks, such as residues, it is thus important tochoose catalysts characterized by low ∆coke values, toreduce problems of overheating; if there is an energydeficit (lighter feedstocks), a catalyst with high ∆cokeshould be used.

255VOLUME II / REFINING AND PETROCHEMICALS

CATALYTIC CRACKING

Table 5. Thermal effects of coke combustion (kJ/kg of coke) (Raseev, 2003)

CO2/CO ratio in the fumes

Content (% in weight) of hydrogen in the coke

4.0 8.0 12.0

0 14,590 19,010 23,400

1 24,075 28,070 32,090

2 27,215 31,085 34,960

4 29,745 33,535 37,300

10 31,820 35,505 39,210

Page 12: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

As a rough indication, a 0.1% increase in ∆cokemay lead to variations of 20°C or more in theregenerator.

Kinetic aspectsA complete analysis of the process should also take

into consideration both the phenomena of the diffusionof the reagents and the products, and the gradualdecrease in the activity of the catalyst, in systems withan extremely high number of unknown componentssuch as heavy petroleum fractions; this is thusextremely complex.

Among the first attempts was that of Voorhies(1945), who correlated mean conversion on astationary bed with the feed rate and the contact time.The following expression for the kinetic constant kwas then proposed:

1K�w �ln11�x�1�x

where w is the volume flow rate and x the conversionrate.

A more recent method for obtaining a kineticrepresentation involves grouping the molecules andconsidering ‘pseudoreactions’ between groups orlumps of components (Fig. 7; Lee et al., 1989).

It is assumed that the constant refers to secondorder kinetics for gas oil and first order kinetics forgasoline.

A series of equations (Ancheyta-Juarez andMurillo-Hernandez, 2000) takes into considerationthe function of the deactivation of the catalyst,kinetic parameters and the variations in yields overtime.

The kinetic parameters thus obtained can be usedas starting values for estimating the kinetic constantsof models with more than four lumps.

The combustion reaction (regeneration phase)is assumed to be of the first order with respect toboth carbon and oxygen (Bonifay and Marcilly,2001):

r �ke�E/RT [C] [O2]

where r is the reaction velocity, [C] is theconcentration of carbon in the catalyst, [O2] the partialpressure of the oxygen, T the absolute temperature, Ethe activation energy, R the gas constant and k thekinetic constant.

Combustion in a real process is far more complexthan that shown in the schematic equation given above.However, the expression does allow for a roughestimate of the time required to reach a givenreduction of the coke in the catalyst; in turn, this timedepends exponentially on the temperature, which thusdetermines the size of the regenerator.

The development of fluid bed technologiesCurrent catalytic cracking processes are all

fluidized bed processes (FCC, Fluid CatalyticCracking) using powder catalysts (20-50 m).

Regeneration takes place continually, with someof the catalyst being sent from the reactor to theregenerator, and then from the regenerator to thereactor (Fig. 8). Some of the heat serves tocompensate for the strongly endothermic nature ofcracking, and some exits with the fumes in the formof sensitive heat; a third part (any excess) can berecovered in cooling coils or preferably in externalexchangers. As already mentioned, the partial use ofresidues increases the production of coke and theexcess heat produced by its combustion. Theproduction of coke generally ranges from 3 to 5% inweight of the feedstock.

256 ENCYCLOPAEDIA OF HYDROCARBONS

CATALYTIC CONVERSION PROCESSES

k21k22 k31

k32

k2k3

k1

k1

Fig. 7. Kinetic models for cracking with three or four lumps.

product catalyst

feedcatalyst+ coke

flue gases

air

Fig. 8. Functional diagram of a FCC unit.

Page 13: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

The catalyst is maintained under fluidized bedconditions by the vaporized hydrocarbon phase (in thereactor) and by air (in the regenerator). The movementof the catalyst between these two sections takes placeby a gravity or pressure differential; this movementcan be controlled by varying the apparent density ofthe catalyst with the intake of steam or air.

Steam also functions as a stripping fluid during thetransitions from the oxidizing phase (regenerator) tothe purely hydrocarbon phase of the reactor. The needto use large quantities of steam at high temperaturesexplains the constant efforts made to improve thethermal stability and the stability to water of zeolitecatalysts.

The heat required for the process decreases as afunction of the degree of conversion since at lowconversion rates strongly endothermic reactionsinvolving the rupture of the C�C bond predominate,whereas at high conversion rates, exothermiccondensation and polymerization reactions becomesignificant.

The combustion temperature needed to regeneratethe catalyst is usually 100-200°C higher than that ofthe cracking process.

Depending on the conditions, especially as afunction of the excess of air, combustion may be moreor less complete, with the formation of CO2 or CO;combustion to CO allows for a better control oftemperature, but makes it essential to install a system

(CO boiler) for the subsequent oxidation of CO to CO2.

In addition to the deposition of coke (reversibledeactivation), the catalyst is also poisoned by thedeposition of metals and basic compounds; it alsoundergoes mechanical erosion and the alteration ofporosity. Some of it is therefore continuously purged,with the addition of fresh catalyst.

There are various versions of the crackingprocess, which differ in terms of the relativepositions of the reactor and the regenerator, and theirshape.

These technologies have always evolved constantly,starting from the installation of the first FCC unit in1942 in Baton Rouge (Louisiana, USA).

Historically, the pioneers of these processesinclude Standard Oil (later Esso and then ExxonMobil), which through a series of developmentscreated the well-known Model IV in 1952, and thenthe various versions of Flexicracking; this group alsoincludes UOP (Universal Oil Products), which as earlyas 1945 introduced the version with a reactor andregenerator side by side, and which currently alsooffers processes for residues and for deep conversion(see Chapter 7.2).

Fig. 9 shows a diagram of Esso’s Model IV, the firstto introduce ‘U’ piping for the transfer of the catalyst.Weighed down by the coke, the spent catalyst fallstowards the bottom of the reactor where it is stripped

257VOLUME II / REFINING AND PETROCHEMICALS

CATALYTIC CRACKING

steam

reactor

steam

steam

steam

stripper

catalystmake up

fuel gas

maincompressor

secondarycompressor

air

air

steam

regenerator

steam

distributorgrid

effluent tofractionator

cyclones

preheatedfeed

Fig. 9. The historicModel IV FCCby Esso.

Page 14: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

with steam and then transported to the regenerator.From the regenerator, the regenerated (lighter) catalystis separated by ‘skimming’ from above, and sent backto the reactor, whose temperature is regulated by theflow rate of the hot catalyst. This movement isfacilitated by maintaining a slight pressure differencebetween the reactor and regenerator, and by varyingthe intake of air into the piping leading back into theregenerator (increasing the amount of air decreases thedensity, thus slowing the circulation of the catalyst).The two-stage cyclones on the regenerator and reactorcycle have the function of capturing the particles ofcatalyst entrained in the effluent vapours and gases;the diagram shows only one, but in fact there are atleast six or eight.

In the later Flexicracking process, introduced in1979, Exxon positioned the ‘reactor’ above theregenerator, with an external vertical riser (riserreactor).

Fig. 10 shows a classic scheme by Kellogg(Orthoflow F) equipped with two regeneration zonesand an external tubular reactor. As said earlier, theintroduction in the 1960s of far more active catalystsled to the elimination of the catalytic bed, since thereactions took place already in the piping (riser). Thevarious riser cracking technologies differ in terms ofthe shape and positioning of the riser (outside or insidethe regenerator).

If the system is fed with particularly heavyfeedstocks, an exchanger to recover heat from the fluidcatalyst may be included.

The circulation of the catalyst is controlled byspecial valves linked to the temperature controlsystem. Worth noting, again in Fig. 10, is theconnector at right angles on the upper part of the riser,equipped with a ‘cap’, which can be easily replacedsince it is subjected to the erosion caused by thecatalyst. It should be remembered that these plants,since they must operate at high temperatures, areprotected internally by fire-proof materials, which areeasily eroded.

Other companies, in addition to those namedabove, have developed viable cracking processes;these include Stone & Webster, the Institute Françaisdu Pétrole (Axens), and Shell (Shell Global Solutions).

The times required for the various phases in atypical process are roughly as follows:• Residence time of the catalyst: in the riser 3 s; in

the stripper 2 min; in the regenerator 6-10 min.• Residence time of the hydrocarbons in the

riser 1.5 s.The most recent developments of the process have

aimed at reducing the contact time between catalystand hydrocarbon vapours, in order to limitnon-selective post-riser cracking; this can be done byseparating the vapours from the catalyst with theinstallation of ‘prestripping’ cyclones at the exit fromthe riser reactor (McAuley and Dries, 2001). Thisexpedient leads to higher yields, lower ∆coke andcracking which is more selective for gasolines andlight olefins.

The coexistence of a combustion phase and a phaseconsisting of hydrocarbon vapours, where the onlybarrier is represented by the steam strippers and thecatalyst transfer piping, makes the whole processextremely delicate; the instrumentation and safetysystems must therefore be adequate.

The effluent from the cracking reactor contains awide range of products (Table 6); therefore, it must befractionated in a plant with characteristics similar to

258 ENCYCLOPAEDIA OF HYDROCARBONS

CATALYTIC CONVERSION PROCESSES

steam

plug valve

steam

riser

disengager

stripper

effluent todistillation

flue gas

airdistributor

combustion air

preheated feed

Fig. 10. Diagram of the Kellog F system with an externalriser reactor and two-zone regenerator.

Page 15: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

those of the topping column (see Chapter 2.2), withthe difference that the feed is not a liquid-vapourmixture, but a superheated vapour, which musttherefore be cooled rather than heated as in topping.Some of the residue and the heavy distillate from themain distillation column is usually recycled to thereactor to increase conversion yields.

The cracking of residuesAlthough many normal FCC units are often

partially fed with residues, the direct cracking ofresidues is possible only in purposely-designed units,and when the content of metals (especially Ni and V)and Conradson carbon does not exceed specific values(e.g. 30 ppm and 5-10% in weight, respectively).Feedstocks that are especially rich in thesecomponents must first be hydrogenated(hydrocracking/hydrofining).

The units for treating residues are equipped withspecial exchangers to cool the catalyst; theseexchangers must be able to work in a particularlyerosive environment. The high C/H ratio of thefeedstock leads to the production of larger amounts ofcoke on the catalyst and, consequently, to an increasedproduction of heat in the regenerator.

The feedstock injection systems must also beappropriate, as must the types of catalysts, particularly‘robust’ with respect to contaminants and capable ofpretreating large-sized molecules in the matrix.

UOP’s MSCC process (MilliSecond CatalyticCracking) involves charging the feedstockperpendicularly to the descending flow of the catalyst(Fig. 11). The reaction products and the catalyst aretransported horizontally through the reaction zonetowards the separator. The riser reactor is thus alsoeliminated (see Section 6.1.2).

Other techniques have been developed (for example,by IFP-SWEC-Total) based on the regeneration of thecatalyst in two separate stages (Bonifay and Marcilly,2001). In the first stage, 60-70% of the coke is burnedat low temperature in a deficiency of air to give CO; inthe second stage, the combustion of the remaining cokeis completed in an excess of air. The gases rich in COfrom the first stage must be sent to a boiler (CO boiler)in which high-pressure steam is generated. The COboiler is also characteristic of other FCC processes

259VOLUME II / REFINING AND PETROCHEMICALS

CATALYTIC CRACKING

Table 6. Typical yields (% in weight) of a FCC unit operating at two different temperatures and of a deep conversion (DCC, Deep Catalytic Cracking) unit operating with different feeds

Entrance temperatureto the reactor 490°C 530°C DCC

Typical feedDCC

Paraffinic feed

H2, CH4, EthaneEthylene

0.670.42

1.320.59

3.52.3

3.25.8

PropanePropylene

1.282.72

1.154.81

2.114.5

2.022.0

i-Butanen-Butane

5.171.24

3.861.12 14.6

5.0

Butenes 3.52 6.78 14.0

Gasolines 57.32 56.14 42.5 28.1

Light oilHeavy oil

15.207.65

15.805.85

15.99.45.0

Coke 4.85 2.58 4.6 5.5

catalyst fromregenerator

mainseparator

feed

reactor

stripper

Fig. 11. Operating principlesof the MSCC reactor (UOP).

Page 16: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

which are not necessarily devoted to residues; thecombustion of CO makes it possible to work at lowertemperatures in the regenerator, to the benefit ofcatalyst life and of safety. However, the investmentsrequired for the unit are higher.

Deep conversion processesAdopting more severe operating conditions (high

conversion temperatures with low contact times and lowpartial pressures) and the use of shape-selectivecatalysts allow the process to produce larger quantitiesof olefins, when these are required by specific marketconditions. The early years of the Third millennium, likethose immediately preceding it, have seen a strongincrease in the consumption of propylene for theproduction of polypropylene, an extremely versatilepolymer with one of the highest growth rates. Propyleneis also supplied by thermic cracking plants, whoseprincipal product is ethylene (see Chapter 10.5).

Estimates made at the beginning of thismillennium attributed about 30% of the worldwideproduction of propylene to catalytic crackingprocesses. However, since refineries were not yet wellintegrated with petrochemical plants, only about 50%of the propylene made in refineries was used forpetrochemical productions. Over the years, the FCCprocess has revealed itself to be extremely flexible;although it is aimed mainly at the production ofgasolines, the process has frequently been managed inorder to maximize other products, such as LPG andolefins. In the immediate future, the FCC process islikely to play a greater role in petrochemicalproductions, at the expense of the production ofgasoline.

Specially designed Deep Catalytic Cracking (DCC)units may, if fed with paraffinic feedstocks, supplyover 25% in weight of propylene. The Stone &Webster/Sinopec unit can operate so as to maximizepolypropylene or iso-olefin yields (Refining […],2002). UOP’s Petro FCC unit is described in Section6.1.2.

The layout of a DCC unit is essentially similar tothat of a conventional FCC with a riser reactor. Asidefrom the catalyst, the differences concern the choice ofoperating variables, the severity and the design of thegas plant. Table 6 compares the typical yields of aDCC plant with those of an FCC unit.

Operating variablesBefore examining the impact of the main operating

variables on the cracking process, it is worth reviewingthe concepts of conversion and recycling.

In general, it is assumed that all products with amolecular weight similar to the feedstock have notundergone conversion. Traditionally, all products that

boil above 220°C are considered heavy fractions; thistemperature can be considered the end boiling point ofgasoline. Conversion can thus be defined as follows:

C% �100(1 �Qpes/Qalim)

where Qpes and Qalim are the volume flow rates ofheavy products and the liquid feedstock respectively.

The products are classified as dry gases, LiquefiedPetroleum Gas or LPG (C3 and C4), gasoline, lightcycle oil (LCGO, Light Cycle Gas Oil), heavy cycleoil (HCGO, Heavy Coker Gas Oil), residue (decantoil) and coke.

Decant oil, sometimes considered alongsideHCGO, is highly aromatic and suitable for producingblack carbon.

If conversion is too high (�76-81%, depending onthe type of feedstock) some of the gasoline producedis transformed into lighter products and coke bysecondary cracking reactions, with a consequentdecrease in gasoline yields.

Conversion is influenced by: a) the type of feed;b) the activity of the catalyst; c) the amount of coke onthe regenerated catalyst (Dcoke); d) the reactiontemperature; e) the flow rate of the combined feed;f ) the oil/catalyst ratio; g) the nebulization vapour of the feedstock.

The oils (LCGO and HCGO) obtained fromcracking are fairly resistant to subsequent conversionsdue to their prevalently aromatic nature; however, ifpartially recycled to the reactor, they produce anadditional quantity of gasoline, increasing yields ascompared with a process without recycling.

The recycling ratio R is defined by the ratio ofcycle oil to fresh feed (R�Qric/Qalim); the total volumefed to the reactor is thus Qalim�Qric.

An example will help to clarify the concept ofrecycling. Suppose that 100 m3/h of liquid feedstockis sent to the reactor, with a conversion of 60%without recycling; thus, remembering the conversionequation, 40 m3/h of non-converted oil is obtained.If 20 m3/h of non-converted residual oil is recycledto the reactor, the feed to the reactor must belowered to 80 m3/h, and the recycling ratio will be25%. Suppose, for the sake of simplicity, thatconversion is identical for the fresh feedstock andfor the cycle oil (although in practice this is not thecase) and remembering that recycling is a closedloop inside the reaction section, the net productionof heavy oil exiting the bottom of the column will be20 m3/h rather than 40 m3/h, and conversion willthus rise to 75% rather than 60%, sinceC�100 [1�(20/80)]�75%.

Even assuming that the cracking of the cycle oil ishalf that of the fresh feedstock, the increase inconversion (15/2�7.5%) is still significant.

260 ENCYCLOPAEDIA OF HYDROCARBONS

CATALYTIC CONVERSION PROCESSES

Page 17: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

The main variables of catalytic cracking thatinfluence conversion are the quality of the feedstock,the temperature of the reactor, the pressure of thereactor, the space velocity, the catalyst/oil ratio and therate of recycling.

The main variables of the regenerator are the flowrate of air, the temperature and the pressure differencefrom the reactor. The entire process is also affected bythe coke balance.

Quality of the feedstockAs already noted, cracking feedstocks containing

mostly aliphatic and naphthenic hydrocarbons is mucheasier and more selective than cracking aromaticfeedstocks. Within an individual class of compounds,the greater or lesser predisposition towards crackingdepends on molecular weight: high boiling paraffinsare the easiest to treat; cycle oils, rich inpolyaromatics, give low gasoline yields, even thoughthey are high boiling fractions. Olefins promote theformation of coke and aromatics act as precursors.

Sulphur and nitrogen compounds are partiallydecomposed with the production of H2S and NH3; thegasolines and gas oils obtained must nevertheless bedesulphurized with subsequent treatments. Ammoniatends to neutralize the acidity of the catalyst. Highsulphur contents increase SO2 emissions duringregeneration.

Organometallic compounds tend to release heavymetals (V, Ni, Fe) under cracking conditions,depositing them on the catalyst and reducing itsactivity; they (especially nickel) may also triggerdehydrogenation reactions, leading to a higherproduction of light products and coke. Vanadium alsocauses sintering phenomena. The difficulty of treatingresidues results from their high metal contents.

Other problems linked to the treatment of residuesinclude: the high concentration of asphalteniccompounds that deactivate the catalyst too quickly(due to the formation of coke); the presence ofalkaline metals, such as Na and K, which permanentlyneutralize the acid sites of the zeolite; and the fact thata large proportion of the feed does not vaporize undernormal cracking conditions. If these residues derivefrom hydrotreating processes (such as hydrocracking),these problems are significantly reduced.

Temperature of the reactorThe feed is preheated to 300-400°C to vaporize as

large a quantity as possible and to lower thecatalyst/oil ratio. The cracking temperature usuallyranges from 480 to 550°C; this influences the reactionvelocity, which almost doubles every 80°C. Increasingthe temperature thus increases conversion, andtherefore the amount of gasoline, gas, LPG and coke.

Above a given conversion level, the gasoline yielddecreases and its octane number increases, due to thehigher olefins and aromatics content.

The temperature is defined by the enthalpiccontent of the combined feed and the amount of hotcatalyst in the riser; it is varied by regulating the intakeof catalyst. The temperature is the most importantvariable since this controls the progress of the processand the thermal balance between reactor andregenerator.

Pressure of the reactorThe total pressure of the process is only slightly

above atmospheric pressure; the regenerator isgenerally at a higher pressure than the reactor. In theusual interval of values (1.5-3.5 bar), the effect of thepartial pressure of the vaporized hydrocarbonfeedstock on conversion and selectivity is extremelysmall.

An increase in pressure leads to a decrease in theoctane number, given the lower production of olefins;at low-pressure values, a modest increase encouragesthe contact between oil and catalyst, as well asconversion and the formation of coke.

Space velocityThe space velocity in itself is not particularly

significant; it varies even if the flow rate of the feed(understood as the sum of fresh feed and cycle oil)remains constant, since the amount of catalyst in thereactor may vary. In any case, it must be expressed interms of weight and not volume (Weight Hourly SpaceVelocity, WHSV) given the variability of the volumeof catalyst in the fluid phase: WHSV�(weight offeed/h)/(weight of catalyst in the reactor).

The circulation of the catalyst may be varied byincreasing or decreasing the pressure in theregenerator.

Experience has shown that the best results usingmodern catalysts are obtained with extremely shortcontact times: a very few seconds (or fractions of asecond) in the riser are sufficient to convert almosttotally the non-aromatic portion of the feed. With shortcontact times, gasoline yields increase, limitingsecondary cracking phenomena (overcracking); theformation of coke is also decreased.

Catalyst/oil ratioThis is defined as the ratio of the unit weight flow-

rate of the circulating catalyst to the weight flow-rateof the feedstock to the reactor.

The amount of catalyst in circulation depends onthe thermal balance of the reaction section. At constanttemperature, any variation in the temperature of thefeed leads to a variation in the catalyst/oil ratio. If the

261VOLUME II / REFINING AND PETROCHEMICALS

CATALYTIC CRACKING

Page 18: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

temperature of the feedstock is low, the ratio increasesbut vaporization worsens; if the temperature of thefeedstock is too high, the ratio decreases, with aconsequent decrease in cracking reactions.

A variation in the temperature of the regeneratedcatalyst also leads to changes in the catalyst/oil ratio:if it decreases, the ratio increases to maintain thereactor at the predetermined temperature. An increasein the catalyst/oil ratio, given an identical reactiontemperature, leads to an increase in conversion, cokeand aromatics.

Recycle rateAn increase in the recycle rate causes a decrease in

the residual oil from the plant and an increaseddeposition of coke; as a consequence, the temperatureof the regenerator increases and the circulation of thecatalyst decreases. The amount of oil to be recycleddepends on the working conditions: the more severethese are, the smaller the amount of oil to be recycled.

Before the advent of zeolite catalysts, highrecycling ratios were adopted to compensate for thelow selectivity of the catalyst.

Temperature of the regeneratorGenerally speaking, the temperature of the

regenerator (700-750°C) is not controlled directly,but depends on the conditions in the reactor, thequality of the feedstock and the amount of air. Theupper temperature limit depends on the thermalresistance of the catalysts and the materials used inthe regenerator.

The heavier the feedstock, the greater theproduction of coke, with the consequent increaseddevelopment of heat during regeneration; the amount

of catalyst needed in the reactor is thus lowered; thelower circulation of catalyst limits conversion.

Low temperatures due to a lower coke content, bycontrast, increase the amount of circulating catalyst.

References

Ancheyta-Juarez J., Murillo-Hernandez J.A. (2000) Asimple method for estimating gasoline, gas and coke yieldsin FCC process, «Energy and Fuels», 14, 373-379.

Bonifay R., Marcilly C. (2001) Catalytic cracking, in: P.Leprince (edited by) Conversion processes, Paris,Technip.

Giavarini C. (1999) Guida allo studio dei processi diraffinazione e petrolchimici, Roma, Siderea.

Lee L.S. et al. (1989) Four-lump kinetic model for fluid catalyticcracking pricess, «The Canadian Journal of ChemicalEngineering», 67, 615-619.

McAuley R., Dries H. (2001) FCC cyclones. A vital element inprofitability, «Petroleum Technology Quarterly», Spring, 21.

O’Connor P. et al. (1991) Improve resid processing,«Hydrocarbon Processing. International edition», November,76-84.

Pekediz A. et al. (1997) Heats of catalytic cracking. Determinationin a riser simulator, «Industrial & Engineering ChemistryResearch», 36.

Raseev S. (2003) Thermal and catalytic processes in petroleumrefining, New York, Marcel Dekker.

Refining processes handbook 2002 (2002), «HydrocarbonProcessing», November.

Voorhies A. Jr. (1945) Carbon formation in catalytic cracking,«Industrial & Engineering Chemistry», 37, 318-322.

Carlo GiavariniDipartimento di Ingegneria Chimica,

dei Materiali, delle Materie Prime e MetallurgiaUniversità degli Studi di Roma ‘La Sapienza’

Roma, Italy

262 ENCYCLOPAEDIA OF HYDROCARBONS

CATALYTIC CONVERSION PROCESSES

Page 19: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

6.1.2 Industrial technologies

UOP Fluid Catalytic Cracking (FCC) and related processes

The UOP (Universal Oil Products) licensedcatalytic cracking processes include the FluidCatalytic Cracking (FCC) process, the ResidFCC (RFCC) process, the MilliSecond CatalyticCracking (MSCC) and the PetroFCC process.The above processes convert gas oil and heavierstreams into lighter, more valuable products viahigh-temperature catalytic cracking. A fluidizedcatalyst system is used to facilitate catalyst andheat transfer between the reactor and regenerator.The system is heat-balanced: the combustion ofcoke in the regenerator provides all of the heatnecessary for the reactor. The main products fromthe processes include: a) light gas: primarily H2,C1, and C2s (normally an undesirable by-productof thermal cracking), where ‘s’ means saturatedhydrocarbon; b) LPG (Liquefied Petroleum Gas):C3s and C4s, including light olefins valuable foralkylation or petrochemical production (thePetroFCC process targets maximum light olefinproduction); c) gasoline: C5� high octanecomponent for gasoline pool or light fuel; d) LCO (Light-Cycle Oil): light-cycle oil blendcomponent for diesel pool or light fuel; e) HCO(Heavy-Cycle Oil): optional heavy-cycle oilproduct for fuel oil or cutter stock; f ) CLO(Clarified Oil): clarified oil or slurry for fuel oil(potential carbon black feedstock); g) coke:by-product consumed in the regenerator toprovide reactor heat demand.

Based on conventional and proventechnology, the FCC process features Optimixfeed distributors, the Vortex Separation System(VSS) riser termination device, AF-spent-catalyst-stripping technology and a combustorstyle regenerator. Catalyst coolers, the RxCattechnology and selective recycle can be added toincrease flexibility for feed and productdemands.

The feedstock to the FCC unit is typicallyVGO (Vacuum Gas Oil), but can also includemany other heavy streams, such as straight rungas oil, coker gas oil, hydrocracked gas oil and

deasphalted or demetallized oil. Some residuumcan be fed, up to 4 wt% Conradson carbon.

The RFCC process uses similar reactortechnology as the FCC process and is targeted forresidual feeds containing over 4 wt% ofConradson carbon. A two-stage regenerator withcatalyst cooling is typically used to control thehigher coke production and resulting heat.

Mechanically similar to a conventional FCC, thePetroFCC process incorporates additional reactionseverity along with the RxCat technology, Optimixfeed distribution, AF spent catalyst strippingtechnology and VSS technology to enhance lightolefin and/or aromatics production. Depending onthe feedstock, propylene yields of 20-25 wt% arepossible; paraffinic feed stocks and hydrocrackedgas oils facilitate light olefin selectivity.

Developed by BARCO and licensedexclusively by UOP, the MSCC process reactortechnology utilizes an ultra-short contact time ina proprietary design contact zone, without theuse of a conventional reactor riser. It can beapplied to all feeds, but its advantages growwhen resid feeds are being processed.

The FCC processes

Feed distribution systemThe cracking process starts with the injection

of feed to the riser. Radially installed Optimixdistributors are an integral part of UOP’s feeddistribution system. Unlike other systems stillusing dense-phase feed injection, one of thefeatures that differentiates the Optimix feeddistribution system is that the acceleration zonebelow the Optimix distributors produces amoderate catalyst density to achieve goodpenetration and mixing of the atomized feedspray. Other benefits of the Optimix feeddistribution system include reduced dry gas andDcoke (delta coke, see Section 6.1.1) as well asincreased gasoline yield.

The newest generation of Optimix feeddistributors allows for the processing of VGOquality feeds with a reduction in steamconsumption of 50% or more, compared withprevious generations of feed distributors.

263VOLUME II / REFINING AND PETROCHEMICALS

Page 20: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

In every case, regardless of the feed quality orprocessing severity, a uniform and controlledcatalyst environment and proper feed distributionare key factors for good mixing and quickvapourization of the raw oil feed in the FCCreactor riser. For this reason, UOP’s feeddistribution technology is focused on both thecatalyst environment into which the feed isinjected and the feed distributor itself. UOP’scatalyst acceleration zone at the base of thereactor riser establishes a uniform catalystenvironment of moderate velocity and densityprior to feed injection. Optimix feed distributors,the heart of the UOP feed system, are used toatomize the feed and inject it into the conditionedcatalyst in the reactor riser.

The base of the riser, commonly referred toas the wye section, is a very turbulent zone wherethe flow of regenerated catalyst changes directionand moves up the riser. It is important that thecatalyst moves evenly up the riser prior to feedinjection in order to minimize backmixing andthe inefficient contact of the feed and catalyst.UOP’s feed distribution system injects eithersteam and/or dry gas at the base of the wye toaccelerate the catalyst to a moderate velocity andto achieve an even, plug-flow catalystdistribution at a moderate density. This uniformand moderate density enhances penetration andquick vapourization of the feed. Commercial unitperformance testing has confirmed the benefitsof the acceleration zone, with conversions over80% by volume and naphtha yields over 66% byvolume.

During testing, gamma scans were used tomeasure the resulting catalyst density profilesthrough the acceleration zone and up beyond thefeed injection point. The catalyst density in theacceleration zone can vary between 15 and 20lb/ft3, depending on the amount of accelerationmedia used. Within 1.5 ft diameters downstreamof the feed injection point, the density decreasesto approximately 5 lb/ft3. This confirms that thefeed is vapourized very rapidly and efficiently bythe moderate density catalyst phase. Densitymaps also show that the Optimix feeddistribution system facilitates an even plug-flowregime in the riser both before and after the feedinjection point.

The Optimix feed distributor is asophisticated mechanical device that efficientlyatomizes the raw-oil feed. Only a small amountof steam (0.5-3.0 wt% based on fresh feed) isrequired for good atomization and distributionof the feed through the use of an innovative

three-stage atomization technique. One of themost important features of the Optimix feeddistributor is that atomization occurs close to thedistributor’s tip and prevents drop coalescing andpulsating flow.

The Optimix feed distributor tip generates auniform, flat, fan spray pattern. This flat spraypattern is generated using a series of specialorifices. These orifices (instead of larger slots)are designed and oriented to create the flat fanfor each installation based on the feed quality,feed rate and riser dimensions. The spray patternof the Optimix feed distributor generates a veryuniform spray, containing small oil droplets witha narrow range of droplet sizes. The Optimixfeed distributor generates a controlled sprayangle, even at 50% turndown, to providecomplete coverage of the reactor riser’s crosssectional area. In the vertical direction, the flatfan pattern remains thin, facilitating quick anduniform feed vapourization for superior yieldsand selectivity. Each Optimix feed distributor isindividually designed and tested for each specificinstallation. This allows UOP to design thedistributors in order to provide full riser coverageat moderate pressure drop and exit velocities,which avoids catalyst attrition and riser erosionproblems. It also allows the maximum reuse ofexisting feed system equipment.

Maintenance and inspection requirements of the Optimix feed distributors are nearly non-existent. The Optimix feed distributor’s tip is ofcast material, which provides improved resistanceto erosion. In addition, UOP uses Dur-O-Lokcouplings to facilitate easy replacement of theinternal and external components, should therefiner’s processing objectives or operating capacitychange substantially.

Separation technology (VSS and VDS) The patented Vortex Separation System (VSS)

for internal riser reactors and the VortexDisengager Stripper (VDS) for external riserreactors represent state-of-the-art risertermination technology. Both systems havecritical pre-stripping features and offer thehighest post-riser hydrocarbon containmentavailable in the industry. These systems capturethe vapour-catalyst mixture at the outlet of theriser and efficiently separate the catalyst withoutallowing the vapour to enter the reactor vessel.The vapour stream is fed into cyclones for finalclean up. Over 99.5% of the vapours passthrough the disengager/cyclone system withoutentering the reactor. In this way, post-riser

264 ENCYCLOPAEDIA OF HYDROCARBONS

CATALYTIC CONVERSION PROCESSES

Page 21: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

cracking is virtually eliminated, resulting in animproved product distribution. Benefits includereduced dry gas, reduced Dcoke, increasedolefinicity, increased gasoline yield and reducedclarified oil yield.

UOP’s Vortex separation technology is thelatest in a long line of fluid catalytic crackingriser termination improvements for moreeffective separation of the catalyst andhydrocarbon phases in the FCC reactor. UOPoffers two options: the VDS design and the VSSdesign.

No riser termination device contains 100% ofthe hydrocarbon vapours flowing to it.Containment refers to the amount ofhydrocarbon vapour that exits the primaryseparation device without spending time in thereactor vessel. In all systems, the catalyst flowsinto a primary separation device, either cyclonicor inertial. Most of the captured vapours quicklypass out of the device and into downstream,secondary cyclones. As a rule, fluidizing gasmust be entrained in the catalyst stream exitingthe primary separation device. These vapoursescape into the reactor/stripper vessel. In thedilute phase of the reactor, the hydrocarbonvapours spend an extended period in thepresence of catalyst and high temperature.Overcracking of this hydrocarbon vapour leadsto higher dry gas production and loss ofselectivity.

UOP’s Vortex separation technologyminimizes the vapour passing into the reactorvessel to provide the greatest reactionselectivities. In other words, this technologymaximizes hydrocarbon containment, which can exceed 99%. In addition to achieving rapid separation and high containment, UOP’sVortex separation technology also provides atermination device that is flexible to operationalchanges and tolerant in upset conditions.

In direct-connected cyclones and othercyclonic separation systems, considerablehydrocarbon vapours escape the cycloneenvironment and enter the reactor vessel. The gasthat keeps the catalyst fluidized travels down thecyclone diplegs with the catalyst. Since almostthe entire catalyst circulation travels through theprimary cyclone diplegs, the gas phase carriedwith it can amount to 5-6 wt% of the feed. As aresult, a cyclonic system without pre-strippingcan achieve only 94-95% hydrocarboncontainment at best. When released into thestripper or reactor vessel, the remaininghydrocarbon content can spend 30 seconds or

more in direct contact with hot, active catalyst. Asubstantial amount of this material eventuallyleaves the reactor as light ends, condensed ringaromatics and coke on the catalyst.

To capture and recover the useful products inthis hydrocarbon stream and preventovercracking, some form of pre-stripping of thecatalyst is needed before it discharges into thereactor. To strip flowing catalyst effectively, thedownward velocity of the catalyst must be lessthan the bubble rise velocity within the fluidizedcatalyst phase. The velocity difference isnecessary so that the hydrocarbon gas phase canrise out of the catalyst phase and be removedquickly from the system. To minimize thestripping diluent (in most cases, steam), a dense phase of catalyst is desirable. A densecatalyst phase also reduces interstitial volume byeffectively ‘squeezing’ gas from the catalyst.UOP’s efforts were focused on converting thecatalyst stream, flowing down the dipleg, into aslower moving dense phase so that thepre-stripping step could be accomplished. Theseefforts resulted in the Vortex separationtechnology.

The first commercial application of Vortexseparation technology suited smaller stackedreactor systems that are easily revamped to anexternal riser. This design is known as the VDSdesign. A single VDS system is fitted at the endof an external side-entry riser. Although the VDSsystem uses the same principles of centrifugalseparation as a cyclone, it functions somewhatdifferently and has a special section at the base toslow catalyst flow and form a dense phase.Stripping steam is injected at the base of thischamber, below the dense phase of catalyst. Thestripped hydrocarbons rise up into the disengagerand exit through the gas tube with the rest of thevapour phase. A set of secondary cyclones isconnected to the vapour outlet to complete thecatalyst separation.

In the VSS design, the catalyst-vapourmixture travels up the reactor riser through thecenter of the chamber and exits through specialdisengaging arms. These arms generate acentrifugal flow pattern that separates thecatalyst from the vapour inside the chamber. Theflow mechanism is similar to that of traditionalcyclone inlet horns. The catalyst falls and formsa dense phase at the base of the chamber, whereit is pre-stripped prior to flowing into the reactorstripper. The stripped hydrocarbon vapours arefully contained in the chamber and exit with therest of the riser effluent vapours to the secondary

265VOLUME II / REFINING AND PETROCHEMICALS

CATALYTIC CRACKING

Page 22: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

cyclones. The only hydrocarbon lost to thereactor vessel is that which accompanies thesmall amount of catalyst disengaged by thecyclones. In both the VSS and VDS systems,hydrocarbon rising from the stripper vessel iscaptured in the chamber and is prevented fromspending excessive time in the surroundingreactor space. The overall hydrocarboncontainment for this system exceeds 99%.

Compared to other riser termination systemsoffered today, the VSS and VDS systems reducethe potential for non-selective post-risercracking. The calculated post-riser residence timeof the hydrocarbon phase within the chamber isless than one second. By creating a dense phaseof catalyst and stripping it within the primaryseparation device, the hydrocarbon ‘leaking’ into thereactor vessel is minimized. In comparison, the dilute phase stripping of other systems isflawed because the interstitial volume is difficultto displace (i.e. the catalyst downward velocityexceeds the stripping bubble rise velocity).UOP’s Vortex separation technology hassuccessfully overcome the shortcomings ofearlier systems, while maintaining operatingflexibility.

For larger units and units with internal risers,the VDS design has some mechanical drawbacks.For these layouts, UOP developed the VSSdesign. The VSS design retains hydrocarboncontainment and pre-stripping characteristics ofthe VDS system; it is compact to fit into thewidest possible range of reactor styles and sizesand has a lower investment cost than the VDSsystem.

Spent-catalyst-stripperUOP’s riser termination systems incorporate a

pre-stripping zone. The design achieves superiorcontacting and stripping efficiency. Vapours fromthe stripper enter the riser termination device anddo not spend any time inside of the reactorvessel. A combination of pre-stripping and primaryzone stripping, with due attention to catalystresidence time, provides the best possible catalyst stripping. UOP’s AF (AdvancedFluidization) stripping technology has resulted inincreased hydrocarbon displacement efficiency, even at very high catalyst flux rates (over120,000 lb/h/ft2).

Many of today’s state-of-the-art FCC unitsutilize UOP’s advanced technologies, such asOptimix feed distributors and VSS risertermination devices. An outcome of the enhancedcatalytic cracking from these technologies is

increased catalyst circulation. As a result, thespent-catalyst-stripper frequently operates wellabove the original catalyst flux design value,which can compromise hydrocarbondisplacement efficiency and unit performance.AFspentcatalyststripping technology wasdeveloped not only to improve the FCC unit yieldperformance, but also its catalyst circulation(hydraulic) performance.

The spentcatalyststripper is a very importantelement of the FCC unit. Its duty is to strip theentrained and adsorbed hydrocarbons from thespent catalyst before it enters the regeneratorvessel. These hydrocarbons are commonlyreferred to as coke on catalyst. In general, fourdifferent types of coke can be associated with thespent catalyst: catalytic coke, contaminant coke,additive coke and cat-to-oil coke. Catalytic,contaminant and additive coke are functions ofthe feed quality, catalyst type, and operatingseverity, and do not present much improvementopportunity for the catalyst stripper. However,cat-to-oil coke is entrained, strippablehydrocarbon, which is directly linked to thecatalyst circulation rate. This type of coke isstrongly impacted by the stripper performance.

AF stripper technology was developed duringa three-year optimization program that includedcomputational fluid dynamic modelling andextensive cold flow modelling work. The coldflow modelling effort tested numerouscommercial and experimental tray designs at awide range of fluxes (up to 140,000 lb/h/ft2), anda wide range of stripping media rates. Theseefforts made it quite clear that optimal stripperperformance is a result of proper control of thestripping media to ensure maximum use of thestripper’s cross sectional area. The outcome of this optimization program wasthe AF stripper tray, grid, and packingtechnology. All styles bring improvedperformance by creating a superior fluidizationand contacting regime throughout the entirestripper vessel (Fig. 1).

Catalyst coolerCatalyst coolers are relatively easy to add to

all styles of regenerators. The dense-phase, low-velocity shell-and-tubetype cooler waspioneered by UOP in the 1980s for operation onheavy feeds, where high Dcoke would causeexcessive regenerator temperatures. The UOPcatalyst cooler uses bayonet-style tubes andgenerates medium- or high-pressure saturatedsteam. Benefits include reduced regenerated

266 ENCYCLOPAEDIA OF HYDROCARBONS

CATALYTIC CONVERSION PROCESSES

Page 23: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

catalyst temperature, higher catalyst/oil ratio, theability to process heavy feeds and better yields.

Many refiners are choosing to process lowercost, residual feed stock components in theirFCC units. The conversion of these heavier feedsto lighter, more valuable products results in ahigher operating temperature in the regenerator.Removing this heat not only retains the catalyst’seffectiveness, but also increases the catalyst/oilratio for better yields, improved productselectivity and enhanced profitability.

The UOP FCC catalyst cooler (Fig. 2) is anexternal vertical shell-and-tube heat exchanger.The catalyst flows over the entire cross sectionalarea of the tube bundle in the dense phase. UOP’sair lance distribution system ensures uniform airdistribution within the tube bundle and a uniformheat transfer coefficient. The generation of steam(up to 850 psi) from the circulating water is usedto remove heat from the regenerated catalyst.Three different styles of catalyst coolers

(flow-through, back-mix and hybrid) have beendesigned and commercialized to accommodate awide range of heat removal duties as well asphysical and plot space constraints.

The key features of the UOP catalyst coolerinclude the following:• Vertical heat transfer surface provides uniform

heat transfer and catalyst contact with thetube bundle, and reduces localized stressesresulting from uneven temperaturedistributions.

• Variable heat-removal capacity facilitatesFCC unit operation when raising thetemperature of the reactor and regeneratorduring start-ups. When necessary, the coolercan even be shut down and effectivelyisolated while the FCC unit continues tooperate.

• Mechanical reliability is achieved throughdense-phase operation with low catalystvelocity to reduce erosion.

• The cooler shell is separate from theregenerator vessel, which provides ultimaterevamp flexibility plus easy, quick coolermaintenance or bundle replacement duringturnarounds.

267VOLUME II / REFINING AND PETROCHEMICALS

CATALYTIC CRACKING

Fig. 1. Spent-catalyst-stripper with AF trays (courtesy of UOP).

hot catalyst

cooledcatalyst

circulating water

steam and water

air

Fig. 2. UOP FCC catalyst cooler.

Page 24: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

All of these features provide a reliable andflexible operation to enhance profitability anddeliver an attractive return on investment.

Catalyst recycle (RxCat technology)Good catalyst circulation is a key factor in

FCC unit reliability. An innovative technology isused to recycle carbonized catalyst back from thereactor to the feed contacting zone (Fig. 3), takingadvantage of inherent activity of moderncatalysts that are not really ‘spent’ when leavingthe reactor. The result is a much highercatalyst/oil ratio than is possible via a heatbalanced regenerator. RxCat is intended forexisting low Dcoke operations, where coldregenerators can become a problem and for thoseoperations where increased light olefin yield isdesired. Regenerator temperature increases whenthe RxCat design is applied. Benefits includevery low dry gas and improved overall yieldselectivity. RxCat technology is an integral partof UOP’s PetroFCC process.

The recycled carbonized catalyst is mixedwith the regenerated catalyst in a mixing vessel(M�R chamber) at the base of the riser. Sincethe recycle of this carbonized catalyst back to theriser is a heat balance neutral process, thecatalyst circulation rate up the riser can be variedwidely without increasing the coke yield.Essentially, RxCat technology breaks the linkbetween catalyst circulation and coke yield. Theresult is a much higher catalyst/oil ratio than ispossible via the normal heat balanced operation.The use of RxCat technology provides significantadvantages:• It allows much shorter contact times to be

used, thereby reducing olefin oligimerizationand hence olefin loss.

• It allows the reactor riser to operate withrelatively cool inlet catalyst temperatures, asthe regenerated catalyst is mixed with thecooler carbonized catalyst, while maintainingan elevated riser outlet temperature.

• It provides more catalytic reaction sites in thereaction zone.Practical benefits include: increased

conversion, improved product selectivity (gasolineand propylene increase), reduced dry gas yield,increased throughput and process flexibility.

Oil selective recycleSelective recycle is the application of

once-through recycles of CLO, HCO, LCO ornaphtha to reduce undesirable products and toimprove yields in selective products. A separate

reaction zone prevents products from comminglingwith riser effluent. Selective recycle is suitable forlow-severity distillate operations.Benefits include improvement in selective yields(e.g. gasoline�LCO) and reduction inundesirable product (e.g. CLO in distillateoperations).

Combustor and regeneratorIntroduced in the late 1970s, UOP’s

combustor regenerator uses a fast-fluidizedcombustion zone to achieve the best cokeburning and full combustion of CO to CO2.Benefits include full combustion without the useof a promotor, minimum afterburn, lowest carbonon regenerated catalyst, no possibility of spentcatalyst bypassing the regeneration zone andlower catalyst inventory.

The two-stage regenerator is used in unitswhere full combustion would result in excessive regenerator temperatures (e.g. RFCC).The upper regenerator (first stage) operates inpartial combustion, while the lower regenerator(second stage) operates in full combustion, withflue gas and excess O2 rising into the upperregenerator. The two-stage regenerator has onlyone flue gas line, which exits from the uppervessel. Catalyst cooler (or coolers) is included inthe design for control of combustion heat. Benefitsinclude clean catalyst from the second stage (lessthan 0.05% carbon on regenerated catalyst) and the

268 ENCYCLOPAEDIA OF HYDROCARBONS

CATALYTIC CONVERSION PROCESSES

spentcatalystrecycle

MxRchamber

Fig. 3. Catalyst recycle.

Page 25: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

ability to process heavy and contaminated residues(up to 10 wt% Conradson carbon).

A power recovery system recovers usableenergy (in the form of electricity from theregenerator flue gas) and typically uses it to drivethe main air blower. Economic justificationdepends on local power costs. UOP’s Third StageSeparator (TSS) technology is employed toprotect rotating equipment and reduce emissions.

The PetroFCC process for petrochemicalfeedstock production

The PetroFCC process targets the productionof petrochemical feedstocks rather than fuelproducts. The new process, which utilizes auniquely designed FCC unit, can produce veryhigh yields of light olefins and aromatics whencoupled with an aromatics complex (Fig. 4).

Driven by an increased demand forpolyethylene and polypropylene, future demandis expected to increase for petrochemicalfeedstocks, particularly the light olefins andespecially propylene. The additional propyleneproduced from the increase in steam-crackerethylene production is expected to be insufficientto meet the demand, and thus propylene fromother sources will be required. Although itsprincipal function has been to produce gasoline,the FCC unit is frequently operated to maximize

other products, such as distillates or LPG. TheLPG mode can be considered a step towards apetrochemical mode of operation since itprovides enhanced yields of the petrochemicalfeedstocks. However, to be considered a majorcomponent in the petrochemical complex, anFCC style unit must produce substantially greaterquantities of these light olefins, produce otherpetrochemical feedstocks of interest, andminimize or eliminate the yield of gasoline andheavier liquid fuels. This is the targeted area forthe PetroFCC process.

The PetroFCC process starts with a uniquelydesigned FCC unit and is capable of producing aproduct slate, in liquid volume, of 35%propylene, 12% ethylene, 20% butylenes, and20% of benzene and paraxylene. This is asubstantial increase in the yield ofpetrochemicals compared to a conventional FCCunit. The catalyst section of the PetroFCCprocess uses a high-conversion, short-contacttime reaction zone that operates at elevatedreactor riser outlet temperatures and low partialpressures. It also incorporates a relatively highlevel of a shape selective zeolite catalyst additivewith selected standard FCC catalyst as thebalance. The PetroFCC process incorporatesOptimix feed distribution system, VSS Vortexseparation technology, AF-spent-catalyst-

269VOLUME II / REFINING AND PETROCHEMICALS

CATALYTIC CRACKING

main column and gas

concentrationsection

LPGtreating

‘sponge’gas

treating

ethylenerecovery

C3/C4splitter

C4 torefinery

heavyrecycle

propylenerecovery

aromaticcomplex

FCC catalystsection

freshfeed

light ends

benzene

p-xylene

raffinate

C3�

C3

C2�

C2

H2�CH4Fig. 4. UOP’s PetroFCCcomplex. C2, ethane;C3, propane; C2�, ethylene;C3�, propylene.

Page 26: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

stripping technology and RxCat technology toenhance light olefin and/or aromatics production.RxCat is a key differentiating technology of thePetroFCC process when compared to alternativetechnology.

The components of PetroFCC technology canalso be applied to existing FCC units that processmore conventional VGO-based feedstocks and desire incremental propylene yieldimprovement. The extent of this light olefin yield improvement is usually dependent on thelimitations in the existing light ends recovery section of the FCC.

Propylene recovery unitPropylene recovery from refinery streams is

increasingly used to supply downstream userswith polymer-grade propylene for petrochemicalapplications. To meet polymer-gradespecifications, the stream must be concentratedto a minimum of 99.5% propylene and beessentially free of diolefins and acetylenicspecies. Incorporating proprietary UOP processequipment, UOP’s Propylene Recovery Unit(PRU) allows the refiner to recover polymer-grade propylene in the most economic manner available today (Fig. 5).

About 30% of world propylene production forthe petrochemical industry currently is containedin FCC and RFCC by-product streams.The UOP PRU gives the refiner the opportunityto cost-effectively upgrade the low-value C3/C4by-product to high-value polymer-grade propylene. Additionally, the PRU can be used to upgradeexisting refinery-grade and chemical-gradepropylene to polymer grade.

The PRU configuration depends on the feedexpected to be processed in the unit. Assumingthat the feed is the C3/C4 by-product from anFCC, the feed would first be depropanized andthen de-ethanized. De-ethanizer overhead istypically sent to the refinery fuel gas system,while the destination of the depropanizer bottomsis refinery specific. The C3 splitter thenfractionates propane and propylene; propane issent from the bottom of the column to storage,while the propylene is taken overhead and sentdownstream for further processing. This is adifficult separation, historically requiring twotowers to accommodate the high reflux and alarge number of distillation trays. However, thePRU has three proven technological advantagesthat allow the fractionation to be accomplished ina single tower, significantly reducing capitalinvestment:

• UOP’s MD (Multiple Downcomer) distillationtrays, which can be installed with minimumtray spacing, enable the reduction of bothtower height and diameter without sacrificingproduct purity.

• UOP’s High Flux tubing installed in thereboiler/condenser reduces the size of thatexchanger by promoting greater heat transferefficiency.

• A heat pump compressor system functions asboth the reboiler and the condenser. Thisreduces the overall equipment count andsignificantly lowers the fractionationpressure.The combination of these technologies results

in the most economic separation of propylenefrom propane available.

If necessary, propylene is further processed toremove carbonyl sulphide (COS) in a solventsystem. After drying in an adsorbent chamber,the propylene stream is treated to remove traceamounts of arsine, phosphine, and antimony (andother metals, if present). The resulting productmeets industry specifications for polymer-gradepropylene.

The MilliSecond Catalytic Cracking (MSCC)process

Catalytic cracking continues to be thecornerstone of most petroleum refineries. It hasproven to be one of the most efficient processesavailable for the conversion of gas oils and residueinto lighter, more valuable hydrocarbons. Recentdevelopments in the FCC unit design have focusedon reducing the contact time between the catalystand the hydrocarbon vapour. Improvements in risertermination devices have led to significantdecreases in post-riser residence time andpost-riser cracking. The benefits of shorter catalystand oil contact times have been lower dry gasyields, lower Dcoke on catalyst and more selectivecracking to gasoline and light olefins. Aninnovative reaction system has been developed toextend the capabilities of short contact time byeliminating the reactor riser altogether. The MSCCdesign has proved to be a robust and easy-to-operate system that can provide distinctyield and process advantages when compared totraditional FCC reactor systems.

The MSCC reactor design is a novel departurefrom traditional riser systems. In the MSCC, feedis injected perpendicular to a down-flowing curtainof catalyst (Fig. 6). Reaction products move across thereaction zone and are quickly separated from the catalyst in a primary separation device.

270 ENCYCLOPAEDIA OF HYDROCARBONS

CATALYTIC CONVERSION PROCESSES

Page 27: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

Following the primary separation device, theremaining catalyst is further separated in a singlestage of external cyclones. The reactor vapours arethen carried over to the main column section of theMSCC complex. The combination of rapidcatalyst-vapour separation and a small-volumereaction zone allows to maximize catalyticreactions and minimize thermal reactions. Thisleads to a notable reduction in undesired secondaryreactions and a more selective product slate. TheMSCC process also incorporates a uniqueapproach to spent catalyst stripping. Due to theposition of the reactor relative to the regenerator, itis possible to take a stream of hot catalyst from theregenerator and inject it directly into the stripper.The hot catalyst acts to increase the strippertemperature and thus the stripping efficiency. Theincreased stripping efficiency leads to higher liquidvolume recovery and lower hydrogen content in thecoke. The result is a lower regenerator temperatureand a higher catalyst/oil ratio.

With the incentive to process more residues inthe FCC feed, refiners are being forced to operatetheir FCC units at a greater severity to maintainyields and conversion. However, this can have anadverse effect on the product selectivity: in fact,an increase in reaction severity of a riser FCCunit will concern all of the feed, including thematerial that is easily converted. The latter willhave the tendency to overcrack and to producedry gas and coke. The MSCC unit design makesit possible to utilize two reaction zones. Theprimary reaction zone, with its inherent shortcontacting time between the catalyst and thehydrocarbon, is the basis of the improvedselectivity and lower Dcoke. The secondaryreaction zone, which takes place in the more

severe hot stripping section, provides theconversion of the more refractory components ofthe feed. This allows effective conversion ofMSCC feed, while avoiding the potential forovercracking in a typical riser FCC unit runningat high severity.

271VOLUME II / REFINING AND PETROCHEMICALS

CATALYTIC CRACKING

6.1.2 Houdek fig 05

C4by-product

propaneproduct

propyleneproduct

Fig. 5. UOP’sPropyleneRecovery Unit(PRU).

feedinjector

hotstripper

regeneratedcatalyst

primaryseparator

externalcyclone

Fig. 6. MSCC reactor.

Page 28: 6 CATALYTIC CONVERSION PROCESSES - Treccani catalytic processes in the petrochemical industry, both in terms of plantsize and the amount of catalyst. ... Catalysts As seen above, the

The MSCC unit expands on the modern FCCunit’s ability to handle a variety of complexcatalytic cracking reactions by further controllinghydrogen transfer and dehydrogenation. Theincreased control of these reactions is animportant consideration since hydrogen transferreduces the olefinicity of the LPG components(valuable alkylation and petrochemicalfeedstocks), while dehydrogenation allowsheavier components to form coke on the catalyst.

Bibliography

Johnson J.A. (1986) Aromatics complexes, in: Meyers R.A.(editor in chief) Handbook of petroleum refining processes,New York, McGraw-Hill, Chapter 2.1.

Mark HoudekCopyright 2004 UOP LLC

All rights reservedUsed with permission

272 ENCYCLOPAEDIA OF HYDROCARBONS

CATALYTIC CONVERSION PROCESSES


Recommended