Jones D.S.J., Pujado P.R. Handbook of Petroleum Processing

Подождите немного. Документ загружается.

262 CHAPTER 6

High Quality Parafrfinic Resids

Commercial Design

RCC Process

Reactor product

Reactor riser

Residue diluents

Diluents

Secondary

air

Primary air

Flue gas

Two stage

regenerator

Catalyst

cooler

High Eff Regen

Total Combustion

Sox Acceptor

Catalyst Cooler

Lift Gas/Feed Distributor

Feed Temp

Diluents

Vented Riser

•

Figure 6.18. UOP resid designs.

Figure 6.19. Shell resid cracker.

FLUID CATALYTIC CRACKING 263

DISENGAGER

EXTERNAL PLENUM

REGENERATOR

DENSE PHASE CATALYST COOLER

SPENT CATALYST DISTRIBUTION

AIR DISTRIBUTOR

CLOSED CYCLONE

SYSTEM

STAGED

STRIPPER

SPLIT FEED

QUENCH

ATOMIZING

FEED INJECTION

LATERAL

CATALYST PLUG VALVE

Figure 6.20. Kellogg resid cracker.

On the regenerator side, the two approaches are to use a single stage regenerator and

add a catalyst cooler or split the regenerator into two stages and make the catalyst

coolers optional. The two stage designs offered differ in the sequence of catalyst ﬂow

and how the air is introduced and utilized.

The Ashland/UOP design has the ﬁrst regenerator on top of the second. Spent catalyst

ﬂows into regenerator one, is partially regenerated and ﬂows to the second regenerator

where the carbon burn is completed. Air is introduced into both regenerators but the

ﬂue gas from number two passes up through arms into regenerator one. Entrained

catalyst is carried with the ﬂue gas. All of the regeneration air goes through the top

regenerator and a CO boiler to reduce the CO content of the ﬂue gas to permitted

levels.

Stone & Webster/Axens have reversed the regenerators so that the spent catalyst enters

the bottom regenerator and the partially regenerated catalyst is vertically conveyed

with lift air to regenerator two where the coke burn is ﬁnished. Air goes to both

regenerators and the catalyst lift line. The ﬂue gas from regenerator one contains CO

since the regenerator is run with no excess oxygen. A CO incinerator or CO boiler is

used to produce an acceptable ﬂue gas composition. Lift gas and the air to regenerator

two are burned and exit the second regenerator and is recombined with the ﬂue gas

from regenerator one. Several ﬂue gas combination schemes have been used.

264 CHAPTER 6

SPENT CATALYST

J-BEND

SPENT CATALYST RISER

UPPER

REGENERATOR

REGENERATED

CATALYST

STANDPIPE

LOWER

REGENERATOR

REGENERATED

CATALYST J-BEND

REGENERATED

CATLAYST

STANDPIPE

REGENERATED

CATALYST

STANDPIPE

PLAN VIEW

REACTOR

RISER

REACTOR

STRIPPER

VESSEL

CYCLONE

STRIPPER

REACTOR

RISER

COKE

SPENT

CATALYST

STANDPIPE

THROTTLING

SUDE VALVE

SPENT

CATALYST J-BEND

REGENERATOR

FEED

PKL

REGENERATED

CATALYST J-BEND

SLIDE

VALVE

SPENT

CATALYST

STANDPIPE

Figure 6.21. Exxon ﬂexicracking IIIR unit.

The reason for splitting the regeneration is to produce CO rather than CO

so that

the heat removal can be eliminated or reduced. Depending on the capacity of the

FCC Unit, the heat removal by this technique can reach 100 million BTUs per hour.

Another beneﬁt is that the hydrogen in the coke burns faster than the carbon as shown

in Figure 6.22. This hydrogen is the chief source of moisture in the regeneration

process that has been shown to deactivate the catalyst. The carbon burn is typically

adjusted so that 45–75% is accomplished in the ﬁrst stage.

FLUID CATALYTIC CRACKING 265

100

1 st

Stage

2 nd

Stage

Hydrogen Burned,

% of Initial

Hydrogen

Carbon Burned, % of Initial Carbon

Either

Stage

Figure 6.22. Carbon and hydrogen burning rates of coke.

If even more heat removal is required one or more catalyst coolers can be added to

the top regenerator. The two basic designs used are shown in Figure 6.23.

Both are dense bed catalyst coolers. Dilute phase coolers were used in the past but

frequent leaks made them too unreliable to use for commercial applications. One of

the designs features a shell and tube exchanger while the other has tube clusters that

have isolation valves. The latter is more expensive but allows isolation of the leaking

cluster without shutting down the catalyst cooler. If steam leaks into the regenerator

unabated, excessive catalyst deactivation occurs. The advantage of a catalyst cooler

UOP/Ashland BDI/S&W

Figure 6.23. Commercial catalyst coolers.

266 CHAPTER 6

is that its duty can be varied over virtually its entire heat load range. This allows the

reﬁner to adjust the coke make to the feedstock and desired reaction severity.

Many early FCC units had steam coils but these must be run at full load to prevent a

mechanical failure. All of these systems require a high water to steam ratio to ensure

vaporization does not occur in the regenerator coils or tubes, which leads to hot spots

and subsequent holes. Older heat removal systems were designed with coils and trim

coolers, but this concept has been rejected in modern resid crackers.

The demand for residual fuels has steadily declined and it is unlikely that it will ever

return. Consequently, residual catalytic cracking is a high growth area as shown in

Figure 6.2.

Heavy fuel oil will not be sold without hydroprocessing in the future. Concern over

acid rain and the severe desulfurization required for transportation fuels will focus

attention to all the other fuels (off-road diesel, bunkers, etc.) and make desulfuriza-

tion mandatory. Reﬁners will ﬁnd feedstock preparation, while costly, will greatly

improve overall yields and install hydroprocessing where the crude warrants. For

high metals laden crude or those deﬁcient in hydrogen, coking will be the preferred

bottoms processing route followed by hydroprocessing of the coker gas oils or the

FCC products.

Fluid cracking catalysts

The FCC process has been shaped and reshaped to accommodate the advances made

in ﬂuid cracking catalysts. Early catalysts were relatively inactive and amorphous

in nature and required a lot of recycle of the uncracked feed to achieve the desired

conversions. Carbon on regenerated catalyst was usually around 0.3–0.6 wt% and

had little effect on unit performance. In the early 1960s zeolite containing catalysts

were introduced that were much more active and selective than previous catalysts but

required the removal of residual coke for optimum commercial performance. This

allowed the reﬁner to substitute fresh feed for the large amounts of recycle being used

and resulted in greatly expanded capacity and gasoline yields.

The preferred FCC zeolite is a crystalline silica-alumina compound that has the

sodium removed. The Type Y or Ultrastable Y zeolite commonly employed has a fau-

jasite structure and as produced formula as shown in Figure 6.24. The important prop-

erties of these zeolites that make them suitable for use in ﬂuid cracking catalysts are:

High stability (>1,600

◦

F) to heat and steam

Three-dimensional structure

High activity (acidity)

Large pores (7.5

FLUID CATALYTIC CRACKING 267

Figure 6.24. Fautasite zeolite structure.

Without all these characteristics the zeolite will not standup to the high temperatures

in the regenerator or let in large molecules so that the interior of the crystal can be

utilized. Rapid coking is mitigated by the three-dimensional structure. The acidity or

activity of the zeolite is associated with the hydroxyl groups attached to the aluminum

atoms in the crystal structure.

The type of zeolites contained in ﬂuid cracking catalysts are variations of the basic

faujasite or Type Y zeolite structure. These are made and characterized as illustrated

in equation (8). The products are referred to as Ultrastable Y’s (US–Y), Hydrogen Y’s

(H-Y), Calcined Rare Earth Y (CREY) or Rare Earth Ultrastable Y (RE–US–Y). Each

of these has been used as the primary cracking component in commercial cracking

catalysts. Variations of the structures and compositions of the above products are

made by new methods of zeolite syntheses and secondary treatment. These include

controlling the amount of alumina in the crystal structure and occluded in the zeolites

pores, substituting other cat ions for alumina in the zeolite framework and using other

cat ions for ion exchange to replace the sodium.

Na-Y

Exc

−→

NaHY

Cal

−→

Cal

−→

US-Y

Exc. Stm

NaREY CREY RE-US-Y

O13.03.0 <1 <1

(8)

where

Exc = Exchange (NH

or RE)

Cal = Calcine

Stm = Steam

268 CHAPTER 6

Table 6.12. Effect of rare earth exchange of Y zeolite on

activity and selectivity

Rare earth on zeolite 0 4 8 16

Activity (C/(100-C)) 2.55 2.8 3.0 4.0

Conversion 71.8 73.7 75 80

Coke 2.55 3.5 4.8 6.1

=/TC

0.81 0.75 0.72 0.65

=/TC

0.39 0.32 0.29 0.23

In Table 6.12 the effect of rare earth exchange of the Type Y zeolite used in FCC

catalysts is shown. Rare earths stabilize the zeolite and results in higher activity and

more hydrogen transfer. Coke selectivity declines along with the oleﬁnicity of the

LPG and catalytic gasoline.

In commercial use, the zeolites undergo dealumination due to contacting with steam

at elevated temperatures such as those encountered in the regenerator. This is shown

in Figure 6.25 where the dealumination is measured by x-ray defraction to obtain the

corresponding unit cell size of the zeolite crystal.

The commercial performance of the cracking catalyst depends on a number of factors,

but the equilibrium unit cell size of the catalyst is a principle variable. As illustrated

24.20

24.30

24.40

24.50

24.60

24.70

030506070 80 85

604030

Unit Cell Size (a

) of Nay Zeoliete

SODIUM-Y TYPE

SiO

/AL

MOL RATIO IN UNIT CELL

% DEALUMINATION (STARTING WITH 5.0 SiO

)

24.66

24.52 (US-Y)

Figure 6.25. Unit cell size versus SiO

/Al

ratio for NaY zeolite.

FLUID CATALYTIC CRACKING 269

Figure 6.26. Effect of unit cell size on catalytic properties.

in Figure 6.26 many of the important properties of the catalyst are determined by this

number. The zeolite types tend to equilibrate around the levels shown but it should be

understood that both hydrogen and rare earth can be used for exchange on the same

zeolite to give to a mixed result.

As the aluminum atoms are removed from the zeolite structure the activity goes down

and much more zeolite needs to be used to give an equivalent conversion. Lower unit

cell sizes increases C

and C

oleﬁnicity, gasoline octane, and reduces coke formation.

The catalyst’s activity is measured by a standard laboratory (MAT) test. The base

catalyst without zeolite can have an activity as low as four MAT though numbers

ranging from 20 to 45 MAT are more typical. Commercial FCC operations have

activities that range from 58 to 77 MAT with 62 to 72 being the most common range.

The matrix is the rest of the catalyst and contains clay, additives and/or a binder that

holds all the components together. Important properties such as the catalyst’s attrition

characteristics, density, CO burning rate, coke selectivity, bottoms cracking and dry

gas make are a direct function of the matrix composition.

Clay is used as ﬁller and provides some pore structure. Additives such as alumina

and silica-alumina are used to increase the matrix cracking activity and crack large

molecules. Binders can consist of silica, alumina or silica-alumina, and can also

enhance activity. In the new catalysts, more than one additive can be incorporated

into the catalyst and small amounts of secondary compounds such as titanium or

phosphorous can modify the catalyst’s performance.

270 CHAPTER 6

Amorphous

Cracking

Gasoline Yield Wt% vs Z/M Surface Area

Gasoline Yield Wt%

Zeolite/Matrix Surface Area

Of Steamed Catalst

Zeolite

Cracking

204

Constant 60 wt% Conversion

MAT Operating Conditions

930 deg F 3c/o 16 WHSV

MAT Feed 225 API/11.5K

Coke Yield vs Z/M Surface Area

Coke Yield Wt%

Zeolite/Matrix Surface Area

Of steamed Catalyst

Amorphous

Cracking

Zeolite

Cracking

Constant 60 wt% Conversion

MAT Operating Conditions

930 deg F 3c/o 16 WHSV

MAT Feed 225 API/11.5K

Figure 6.27. (A–C) Effect Zeolite/Matrix SA ratio on gasoline, coke, and 640+ bottoms yield.

The ratio of zeolite to matrix is used to vary the yields from an FCC and must be

optimized for each matrix and zeolite system. Each feedstock with different proper-

ties such as molecular weight and aromaticity require a unique catalyst formulation.

Figures 6.27A–C show how the ratio of zeolite to matrix affects yields for a 22.5

◦

API (11.5 K) feed.

FLUID CATALYTIC CRACKING 271

640 + Bottoms Yield vs Z/M Surface Area

640 + Bottoms Yield Wt%

Zeolite/Matrix Surface Area

Of steamed Catalyst

Amorphous

Cracking

Zeolite

Cracking

Constant 60 wt% Conversion

MAT Operating Conditions

930 deg F 3c/o 16 WHSV

MAT Feed 225 API/11.5K

Figure 6.27. (Cont.)

In addition to FCC catalysts, there is now a long list of FCC additives to meet speciﬁc

processing needs. These include products for:

removal

Octane enhancement

Metals passivation

Bottoms cracking

Fluidization aids

Oleﬁn generation

A compilation of all the commercially available products is given in Appendix A.

Cracking for light oleﬁns and aromatics

The demand for propylene as a petrochemical feedstock is outpacing the need for

ethylene. As a result, the traditional source of propylene, i.e. as byproduct of the

steam cracker, is not sufﬁcient for future propylene projections.

Catalytic cracking is the other major source of the propylene with propane dehydro-

genation making up the balance. Worldwide market shares (2001) are 68:28:4, though

recently in the United States, catalytic cracking has overtaken steam cracking as the

largest single source of propylene used for petrochemicals.