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

Подождите немного. Документ загружается.

CATALYTIC REFORMING 231

The hydrogen-to-hydrocarbon (H

/HC) mol ratio is the ratio of mols of hydrogen in

the recycle gas to mols of naphtha charged to the unit. The recycle gas is a mixture of

hydrogen and light gases, typically 75–92 mol% hydrogen. The ratio of total recycle

gas to hydrocarbon is sometimes called the gas-to-oil ratio. Recycle hydrogen is

necessary to maintain catalyst-life stability by sweeping coke precursors from the

catalyst metals. The exact mechanism is proposed to be hydrogenation and inhibition

of polymerization. The rate of coke formation on the catalyst is a function of the

hydrogen partial pressure present. Increasing the H

/HC ratio increases the hydrogen

partial pressure and removes coke precursors from the metal sites, thereby increasing

stability with little effect on product quality or yields.

Except for units designed for continuous regeneration through the circulation of the

catalyst between the reactors and the regenerator, catalytic reforming units normally

will require a shutdown for regeneration every 6–12 months. This relatively long

cycle can be obtained by operating under milder conditions of high partial pressure of

hydrogen, lower reactor temperatures, and lower octane products. Continuous units

operate under severe conditions to yield high octane, high aromatics production, at

low hydrogen partial pressures, and higher reactor temperatures. Different catalysts

are used depending on the application.

Semi regenerative reformers make use of catalysts the contain platinum or plat-

inum modiﬁed by rhenium or, to a lesser extent, iridium. The support is most of-

ten gamma alumina, although there have been uses of eta alumina (4). Rhenium

or iridium is used to enhance the life of the catalyst over that observed for Pt-only

catalysts. All these catalysts are typically sulﬁded to minimize metal-catalyzed hy-

drogenolysis reactions that produce light gases and reduce gasoline yield. Additional

components were used on catalysts commercialized in the 1990’s. The use of two

catalysts in a SR unit; one catalyst in the front reactors and another catalyst in the

back reactors to provide maximum yield, activity, and stability was commercialized in

1994 (8).

There are two main shapes of catalysts, cylindrical and spherical. The cylindrical

catalysts are usually extruded alumina. The spherical catalysts may be formed through

a dropping method or by rolling wet, soft alumina dough. In some instances, factors

such as the resistance to ﬂow or ﬂow distribution concerns may cause one form to

be chosen over the other. The density of the catalysts may vary from approximately

0.5–0.8 g/cm

. The variability in density allows the reﬁner to load more pounds of

catalyst in a unit, should additional catalyst activity or stability be desired.

The process of moving catalyst from the reactors to the regenerator and back re-

quires the use of spherical catalysts, rather than extrudate, to avoid catalyst dusting

and breakage. Continuous regeneration units are operated at high severity and low

pressures to produce the greatest amount of aromatics and hydrogen possible. The

232 CHAPTER 5

catalyst is circulated at a rate such that it corresponds to about one regeneration per

week or even at a greater frequency if needed due to the rapid deactivation under these

conditions.

Typical catalysts used in these units had a composition of platinum and tin on gamma

alumina. The tin was used to reduce the hydrogenolysis activity of the platinum and

to improve yields. The reduction in metal-catalyzed cracking is also considered to

stabilize the catalyst relative to platinum only. Currently, new proprietary catalysts

are used to increase yields, lower coke make, or allow higher throughput (9).

Process ﬂow schemes

Fixed bed semiregenerative reforming

A typical SR Platforming ﬂow diagram is presented in Figure 5.6. Feed to the unit

is mixed with recycled hydrogen gas, raised to the reaction temperature ﬁrst by a

feed-efﬂuent combined feed exchanger and then by a ﬁred heater, and then charged to

Figure 5.6. Semiregenerative reforming process (reprinted with permission from UOP LLC).

CATALYTIC REFORMING 233

the reactor section. Because most of the reactions that occur in the Platforming process

are endothermic, the reactor section is separated into several stages, or reactors.

Interheaters are installed between these stages to maintain the desired temperature

range across the catalyst in the reactor section. Efﬂuent from the last reactor is cooled

by the feed-efﬂuent heat exchanger for maximum heat recovery. Air or water cooling

provides additional cooling to near-ambient temperature. The efﬂuent is then charged

to the separation section, where the liquid and gas products are separated. A portion of

the gas from the separator is compressed and recycled back to the reactor section. The

net hydrogen produced is sent to hydrogen users in the reﬁnery complex or for use as

fuel. The separator liquid is pumped to a product stabilizer, where the more-volatile

light hydrocarbons are fractionated from the high-octane liquid product.

Fixed bed cyclic reforming

Cyclic reforming is similar to SR reforming, but an additional reactor replaces one

of the primary reactors while that reactor is being regenerated. The frequency with

which a particular primary reactor is replaced and regenerated depends upon its rate

of deactivation. Large diameter valves and piping are used to vary the process ﬂow

between reactors.

Platforming process with continuous catalyst regeneration

In parallel with bimetallic catalyst improvements and other process and regeneration

advances, UOP began to develop the CCR Platforming™ process (Figure 5.7). In the

CCR Platforming unit, partially aged catalyst in the reactors is continuously replaced

with catalyst that has been freshly regenerated in an external regenerator (CCR section)

to maintain a low average age for the reactor catalyst. Thus, the high selectivity

and high activity characteristics associated with new catalyst can be maintained at

signiﬁcantly higher severities than with the SR Platforming process. For example, a

SR Platforming unit operates at a severity that steadily builds coke up on the catalyst

surface over the length of a cycle (6–18 months), at which point the unit is shut

down and the catalyst regenerated. Throughout the cycle, yields decline. Instead, in a

modern CCR Platforming unit, the catalyst is regenerated approximately every three

to seven days and the yield does not decline.

The ability to continuously regenerate a controlled quantity of catalyst is the signiﬁcant

innovation of the CCR Platforming unit. The catalyst ﬂows by gravity from the last

reactor into a catalyst collector vessel. The catalyst is lifted by either nitrogen or

hydrogen lifting gas to a catalyst hopper above the regeneration tower. Catalyst then

ﬂows to the regeneration tower, where the catalyst is reconditioned. Regenerated

catalyst is returned to the top of the reactor stack by a transfer system similar to

that used in the reactor-to-regenerator transfer. Thus, the reactors are continuously

supplied with freshly regenerated catalyst, and product yields are maintained at fresh

234 CHAPTER 5

Figure 5.7. CCR Platforming process (reprinted with permission from UOP LLC).

catalyst levels. The regeneration and reactor sections of the unit are easily isolated to

permit a shutdown of the regeneration system for normal inspection or maintenance

without interrupting of the Platforming operation.

A few years after the introduction of the UOP CCR Platforming process, another con-

tinuously regenerable process design was offered by the Institut Fran¸cais du P´etrole.

Though similar to CCR Platforming, the continuous reforming units designed by the

Institut Fran¸cais du P´etrole differ most notably in that the reactors are located side-

by-side and the catalyst transfer is effected through transfer piping between reactors.

Advantages of CCR Platforming

From both an economic and technical standpoint, the CCR Platforming process is

superior to the SR and cyclic reforming processes. The CCR Platforming unit allows

for low-pressure operation, leading to higher yields. At these conditions, the SR

Platforming catalyst is completely deactivated after only a few days of operation.

Both the hydrogen and C

+yields are maximized with the CCR Platforming process.

Since the number of cyclic reformers is small relative to CCR Platforming process

units and SR process units, the following comparison will focus on contrasting CCR

Platforming units and SR units.

CATALYTIC REFORMING 235

Table 5.6. Relative severities of CCR versus SR

Platforming units

Operating mode SR CCR

Charge rate, barrels/day 20,000 20,000

LHSV, h

−1

Base Base × 1.8

/HC Base Base × 0.5

RONC 97 102

Reactor pressure, psig Base Base-50

Separator pressure, psig Base Base-145

Cycle life, months 12 Continuous

High yields and constant yields are important in the economics of reforming. As

the catalyst is deactivated by coke deposition in the SR Platforming process, the

yields begin to decline. With the CCR Platforming process, the reformate, aromatics,

and hydrogen yields remain consistent and constant. This is particularly important

for downstream users. The CCR section ensures proper redispersion of the metals

and chloride balance to maintain fresh catalyst activity. CCR Platforming units have

higher on-stream efﬁciency and are able to handle upset scenarios without long-term

shutdown or signiﬁcant decline in performance.

Table 5.6 shows the relative operating severities for the SR and CCR Platforming

units. The CCR Platforming unit operates at higher severity and lower reactor catalyst

inventory. In addition, the CCR unit runs continuously compared to 12-month SR

Platforming cycle lengths.

Typical product yields for the SR and CCR Platforming units operating at the con-

ditions presented in Table 5.6 are shown in Table 5.7. Many of the beneﬁts of

CCR Platforming are demonstrated in Table 5.7. More and higher-purity hydro-

gen is produced. The higher severity of the CCR Platforming unit results in simi-

lar liquid volume for the two units. However, the reformate produced by the CCR

Platforming is more valuable than that produced by the SR Platforming unit. Tak-

ing into account both the higher octane value and the increased on-stream efﬁciency

of the CCR Platforming unit, 80 million more octane-barrels, or 11.4 million more

Table 5.7. Yield comparison of CCR versus SR

Platforming units

SR CCR Delta

Hydrogen yield, SCF/bbl 1,085 1,709 +624

Hydrogen purity, mol% 80 92.6 +12.6

+ yield, LV% 79.3 79.4 +.1

+ yield, wt% 85.2 88.2 +3

Octane-barrel, 10

bbl/yr 513 583 +80

236 CHAPTER 5

Table 5.8. Economic summary

Description SR CCR

Gross key product value, $MM/yr 120 141

Raw materials less by-products, $MM/yr 98 103

Consumables, MM$/y 0.3 0.75

Utilities, $MM/yr 2.8 6.2

Total ﬁxed costs, $MM/yr 5.5 6.5

Capital charges, $MM/yr 3.5 5.2

Net cost of production, $MM/yr 110 122

Pretax proﬁt, $MM/yr 10 20

Pretax ROI, % 30 41

Payout period, (gross) years 1.5 1.3

metric octane-tons, are produced per year with the CCR Platforming unit than with

the SR Platforming unit. Octane-yield is deﬁned as the product of the reformate yield,

octane, and operating days.

A summary of the operating revenues and costs expected for the SR and CCR Plat-

forming units in shown in Table 5.8. The nomenclature follows standard deﬁnitions.

The economics of the CCR Platforming process are superior as a direct result of the

differences in operating severity and ﬂexibility of the two modes of operations. The

CCR Platforming unit produces more valuable reformate at 102 RONC versus the SR

Platforming reformate at 97 RONC. On-stream efﬁciency of the CCR Platforming

unit is 8,640 hr per year compared to about 8,000 hr per year for the SR Platforming

unit. Although the CCR Platforming utility costs are higher than those for the SR

Platforming unit, these costs are offset by the increase in both product quantity and

value as demonstrated by pretax proﬁt and return on investment.

Catalysts and suppliers

For detailed updated lists of catalysts and suppliers consult the periodic reviews

published by the Oil and Gas Journal.

The main catalyst suppliers are:

Axens/IFP Group Technologies

Criterion Catalyst Co.

Exxon Research & Engineering Co. (ERECO)

Indian Petrochemicals Corp., Ltd.

Instituto Mexicano del Petr´oleo (IMP)

UOP LLC

Some of the catalyst suppliers may restrict availability to process licensees only.

CATALYTIC REFORMING 237

References

1. A. L. Huebner, “Tutorial: Fundamentals of Naphtha Reforming,” AIChE Spring Meeting

1999, Houston, TX, 14–18 March 1999.

2. American Petroleum Institute Research Project 45, Sixteenth Annual Report, 1954.

3. E. L. Marshall and K. Owen, eds., Motor Gasoline, The Royal Society of Chemistry,

London, 1995, p. 8.

4. G. A. Mills, H. Heinemann, T. H. Milliken, and A. G. Oblad, Ind. Eng. Chem., 1953:45;134–

137.

5. D. R. Stull, E. F. Westrum, and G. C. Sinke, The Chemical Thermodynamics of Organic

Compounds. John Wiley & Sons, New York, 1969.

6. S. M. Augustine, G. N. Alameddin, and W. M. H. Sachtler, J. Catal., 1989:115(1); 217–232.

U.S. Pat. 4,469,812 (September 4, 1984) C. M. Sorrentino, R. J. Pellet, R. J. Bertolacini

(to Standard Oil Company—Indiana).

7. J. A. Weiszmann, In Meyers, ed., Handbook of Petroleum Reﬁning Processes, McGraw-

Hill, New York, 1986, p. 31.

8. M. D. Moser, D. H. Wei, R. S. Haizmann, CHEMTECH, October, 1996, pp. 37–41.

Chapter 6

Fluid catalytic cracking

Warren Letzsch

∗

Crude oil comprises hundreds of molecules that boil over a wide temperature range.

The lighter products can be separated directly by distillation into LPG, gasoline,

naphtha, kerosene, and diesel fuels. Heavier products (BP > 650

◦

F/344

◦

C) include

vacuum gas oils and resids. Thermal and catalytic cracking processes in petroleum

reﬁning reduce the molecular weight of these heavier constituents and produce more

valuable lighter products such as LPG, gasoline and diesel fuels.

Catalytic cracking was ﬁrst commercialized in 1936 by Eugene Houdry. This ﬁxed bed

process was a major improvement over the thermal cracking processes it replaced due

to the improved yield distribution and superior product properties. Multiple vessels

were utilized that alternated between cracking, stripping, regeneration, and purge

cycles. This conﬁguration was quickly replaced by a moving bed reactor and a separate

regenerator or kiln that ﬁrst used a bucket lift to move the pelleted catalyst followed

later by a pneumatic air lift system. The last of these units was built around 1960.

Standard Oil of New Jersey developed their own cracking process rather than pay the

large royalty being asked at the time. They commercialized the ﬂuid catalytic cracking

(FCC) process in three years, starting in 1939 and culminating in 1942 with the start-

up of PCLA#1 at their Baton Rouge, Louisiana reﬁnery. The inherent superiority of

the ﬂuid process to transfer both heat and catalyst ultimately made it the catalytic

cracking process of choice.

Many different designs of ﬂuid catalytic crackers have been introduced over the years.

Table 6.1 is a list of the various FCCU conﬁgurations and the approximate year of

their commercial introduction.

Fluid catalytic cracking has evolved considerably over the more than 60 years since

its inception. As seen in Figure 6.1 these changes have encompassed all aspects of

*Senior Reﬁning Consultant, Stone & Webster Inc—A Shaw Group Company.

239

240 CHAPTER 6

Table 6.1. Evolution of ﬂuid catalytic crackers

Commercial ﬂuid catalytic crackers

1942 Model I upﬂow

1943 Model II downﬂow

1945 Sinclair design

1947 Model II side by side

1951 Kellogg orthoﬂow A

1952 Exxon Model IV

1953 Kellogg orthoﬂow B

1955 Shell two stage reactor

1956 UOP straight riser (SBS)

1958 Exxon riser cracker

1961 HOC cracker (Phillips)

1962 Kellogg orthoﬂow C

1967 Texaco design

1971 Gulf FCC process

1972 Exxon ﬂexicracker

1972 Amoco ultracracking

1973 UOP high efﬁciency design

1973 Kellogg orthoﬂow F

1981 Total petroleum resid cracker

1982 Ashland/UOP RCC unit

1985 IFP R2R

1990 Kellogg/Mobil HOC

1991 Deep catalytic cracking (RIPP and S&W)

1993 Exxon ﬂexicracker III

1996 MSCC UOP/Coastal

2002 Catalytic pyrolysis process

the process as it has adapted to meet ever-changing demands and to accommodate

new technologies.

The initial units (1940s) were tall, had dense bed reactors and were made of carbon

steel. Dilute phase catalyst coolers and regenerator steam coils were employed to limit

the regenerator temperatures. Recycle rates of 100–150% were needed to achieve the

desired conversions. Later designs (late 1940s to early 1950s) were undertaken to

reduce the height of the crackers, make them more compact and cater to the many

small reﬁners around the United States.

In the late 1950s side by side designs with straight feed risers were introduced to im-

prove gasoline selectivity. Residue cracking in ﬂuid cracking units was ﬁrst practiced

in the early 1960s. This unit was designed for 100% atmospheric bottoms and had

large amounts of heat removal.

Catalytic cracking was truly revolutionized in the early 1960s with the advent of zeo-

lite containing ﬂuid cracking catalysts. Catalyst activities were raised by an order of

FLUID CATALYTIC CRACKING 241

Processing Objectives

Gasoline

Resid

Olefins

Octane

Environmental

CATALYSTS

EQUIPMENT

Carbon Steel Stainless Steel

Dilute Coolers & Bed Coils

Dense Bed Coolers

Dense Bed Reactor Riser Cracking

Decreasing Recycle

Advanced Feed Injection

Reduce Dilute

Phase Cracking

Partial CO Burn

Full CO Burn

Amorphous Zeolite

Low Alumina High Alumina

Spray DriedGround

Increasing Concentrations

Shape Selectivity

DeSoxCO Promotion

1942 1952 1972 1982 19921962

YEAR

OPERATING MODE

Controlled Recycle

Figure 6.1. Fluid catalytic cracking development.

magnitude and units needed to be redesigned to take full advantage of the new catalyst

technology. These design changes included the elimination of reactor dense beds and

the use of the feed riser as the sole conversion vessel. Recycle was greatly reduced

and replaced with more fresh feed. It was found that coke left on the regenerated

catalyst impaired the catalyst’s activity and selectivity and the average carbons on

regenerated catalyst were reduced from 0.3–0.5 wt% to 0.1 wt% or less. Complete

CO combustion in the regenerator was introduced to provide the needed regener-

ator burning conditions which necessitated higher regenerator temperatures. Alloy

internals replaced the carbon steel and chrome-moly hardware in the regenerator and

the regenerated catalyst standpipe. Catalyst inventories were minimized due to the

favorable coke burning kinetics and modiﬁcations in the regenerator design (1970s).

In the early 1980s several new resid cracking designs were introduced. The spent

catalyst regeneration was staged in these designs and dense bed catalyst coolers were

optional. These coolers have also been placed with single stage regenerators for

residual catalytic cracking units. The growth of this segment of the catalytic cracking

process is shown in Figure 6.2. Over two million barrels per day of on purpose

resid crackers have now been licensed and another million barrels of FCC capacity

processes some resid along with their normal gas oils. Typical yields from resid

cracking along with the feedstock properties are given in Table 6.2.