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

Подождите немного. Документ загружается.

412 CHAPTER 9

Net

Hydrogen

Stabilizer

Offgas

Stabilizer

Bottoms

Stabilizer

Column

Product

Separator

Recycle

Compressor

Reforming

Reactors

Heavy

Naphtha

Feed

Charge

Heater

Light

Naphtha

Feed

Par-Isom

Reactor

Figure 9.3.9. Par-Isom and reformer revamp ﬂow scheme.

units is shown in Figure 9.3.9. In this conﬁguration, a new Par-Isom reactor and charge

heater are added to operate in parallel with the existing reforming unit reactors.

The hydrogen required for the Par-Isom reactor is taken from the reforming unit

recycle compressor, combined with the light-naphtha charge, and sent to the Par-

Isom reactor. The Par-Isom reactor efﬂuent is returned to the reforming unit. The

integrated units share a common product condenser, product separator, and stabilizer.

The isomerization unit at the Flying J Reﬁnery has operated successfully in this mode

since 1986 and is currently operating as a Par-Isom unit.

Isomerization process economics

The economics of a process are the most-important consideration in selecting a tech-

nology for a given application. The Par-Isom process offers an economic alternative

to conventional isomerization processes. Two case studies illustrate the beneﬁts of

this technology: one for reloading existing once-through zeolitic units (Case study

A) and the other for integration of a Par-Isom reactor into a reforming unit to add

isomerization capacity (Case study B).

Case study A

The revamp of an existing once-through zeolitic isomerization unit is an excellent

application for the Par-Isom process because both increased product quality and

throughput can be realized. This increase is shown in Table 9.3.2 as a comparison of

the base case (continued operation with current zeolitic catalyst) and the unit after

revamp to the Par-Isom process. This example was selected to represent a typical

GASOLINE COMPONENTS 413

Table 9.3.2. Par-Isom revamp opportunities

for once-through isomerization units

Base case Revamp

Catalyst Zeolitic LPI-100

Capacity, bpd 5,000 6,250

KMTA 190 238

Product RONC 79.5 81.7

MONC 77.1 79.8

Yield, LV-% 98.0 98.0

Incremental

Catalyst cost, $MM Base 1.33

Net income, $MM/yr Base 1.28

Shutdown case

Payback, years Base 1.7

IRR, % Base 58

Incremental case

Payback, years Base 1.0

IRR, % Base 96

zeolitic process unit. The Par-Isom process allowed throughput to be increased by

25% along with an increased product octane of 2 RONC. An alternative would be to

realizea3RONupgrade by keeping the capacity constant. However, the economics

of this option are generally not as attractive as increasing throughput. Octane was

valued at $0.25 per octane-barrel, and no substantial changes to existing equipment

were considered in this case.

Two scenarios were considered: a shutdown case and an incremental case. In the

shutdown case, the current zeolitic catalyst was assumed to be operating satisfactorily,

and so the revamp would have to justify replacement of the existing catalyst. The

payback for this case was 1.7 years, which is equivalent to an internal rate of return

(IRR) of 58%; typically a viable alternative for most reﬁners. In the incremental

case, the zeolitic catalyst was assumed to be deactivated and new catalyst required

for continued operation of the unit. Revamp to the Par-Isom process is an attractive

option in this situation. The payback was 1 year, or an IRR of 96% on the incremental

investment required for the revamp.

Case study B

New isomerization capacity is being considered in many areas of the world, such as

South America, Russia, and the Far East, as lead is being phased out of the gasoline

pool or regulations on benzene or aromatics content are enacted. Often the reﬁneries

involved are hydroskimming-type reﬁneries that have limited conversion capacity

and one or more semi-regenerative reforming units to upgrade the heavy naphtha

414 CHAPTER 9

and provide a source of hydrogen. These reﬁneries are ideal candidates for adding

isomerization capacity by integrating a Par-Isom unit with the existing reforming unit

(Figure 9.3.9).

The revamp in this example involved reloading a 600 kMTA (15,200 bpd) reforming

unit with UOP’s R-56™ catalyst in place of the existing unit’s less-stable reform-

ing catalyst so that excess recycle hydrogen was available for the Par-Isom reactor.

The Par-Isom reactor section is installed to operate in parallel with the reforming

unit reactors. The addition of the Par-Isom unit involved installing a charge heater

using high-pressure steam, a light-naphtha feed pump, the Par-Isom reactor, and the

necessary controls. The balance of the existing equipment, including the product

condenser, product separator, and stabilizer, was adequate for the integrated unit. The

estimated erected cost (EEC) for the Par-Isom reactor system, including an allowance

for royalties, was $2.3 million U.S. (4th Quarter 1997, U.S. Gulf Coast basis). The

catalyst cost was $2.1 million U.S., excluding the cost of the platinum, and the total

investment cost was $4.4 million U.S.

Adding isomerization capacity is a low-cost means of increasing gasoline pool octane

(Table 9.3.3). The base case, without isomerization, produces a gasoline product

consisting of 494 kMTA (11,750 bpd) of 96 RONC reformate and 300 kMTA (8,780

bpd) of 66 RONC light naphtha. The combined streams produce a gasoline product at

Table 9.3.3. Integration of a Par-Isom unit with a semiregenerative reforming unit

Heavy Reforming Light Par-Isom

naphtha unit naphtha unit

Catalyst R-56 LPI-100

Product, bpd 15,200 11,750 8,780 8,325

kMTA 600 494 300 287

Product RONC – 96 66 79

MONC – 85 65 78

Blended gasoline without Par-Isom unit with Par-Isom unit

RONC – 84.6 – 89.8

Bpd – 20,530 – 20,075

kMTA – 794 – 781

 RON-bpd – – – 65,897

Revamp

Catalyst cost, $MM – – – 2.10

Equipment cost, $MM – – – 2.30

Total cost, $MM – – – 4.40

Net income, $MM/yr – – – 5.50

Revamp economics

Payback, years – – – 0.8

IRR, % – – – 125

GASOLINE COMPONENTS 415

84.6 RONC and would require the addition of lead to make regular-grade gasoline. In

the revamp case, the light naphtha is upgraded in the Par-Isom reactor to a 79 RONC.

Because of the nature of the feedstocks, achieving this high isomerate octane with

conventional zeolitic catalysts is not possible. Some yield loss occurs in the Par-Isom

reactor, and so 287 kMTA (8,325 bpd) of isomerate is available for blending with

the reformate. Adding the Par-Isom reactor provides sufﬁcient octane to produce a

total gasoline product at 781 kMTA (20,075 bpd) with a 89.8 RONC. Regular-grade

gasoline can be produced in the integrated reforming and Par-Isom units without the

addition of lead.

The economics involved in the revamp are also shown in Table 9.3.3. The required

investment of $5.5 million U.S. includes catalyst, equipment, and royalty costs. The

Par-Isom unit provides an additional 65,897 octane-barrels to the gasoline pool, or

$5.5 million U.S. if the octane is valued at $0.25 U.S. per octane-barrel. The simple

payback for the revamp is less than a year, and the internal rate of return is attractive

at 125%. Integrating the Par-Isom reactor with an existing reforming unit is clearly

an attractive solution to the challenges posed by lead phaseout or benzene reduction.

Other applications

The Par-Isom process can be considered for other applications, particularly for new

units. In general, chlorided alumina catalysts, such as the I-80 catalyst used in the

Penex process provide the most cost-effective isomerization technology for grassroots

reﬁneries or for installing new units without revamping of existing equipment. The

additional octane provided by these catalysts results in the most-value-added and

most-attractive economics.

In some cases, the Par-Isom technology should be considered for new units. Since

the Par-Isom process is the lowest capital cost option for once-through isomerization

units, it is the process of choice when capital cost is a major concern. Another situation

is when the primary processing objective is to minimize the benzene content in the

gasoline pool, and sufﬁcient octane to meet the gasoline product blending require-

ments already exists. Because the additional octane achieved by a chlorided alumina

catalyst may not have any value in these situations, the lower-cost Par-Isom process

is the most economically attractive option.

Conclusions

Light-naphtha isomerization processes have become increasingly important in help-

ing reﬁners improve octane and meet the production demands for reformulated or

low-benzene-content gasoline. The discussion in this chapter has been based on the

relative performance of UOP’s catalysts. Similar conclusions could be drawn if using

equivalent catalyst from other suppliers.

416 CHAPTER 9

The I-80 catalyst is one of the highest-activity chlorided alumina catalysts available to-

day. This catalyst provides a means to improve the proﬁtability of Penex and Butamer

units by improving octane or allowing for increased capacity or both. The 0.5–1 RONC

additional octane that can be provided by the I-80 catalyst makes this catalyst an obvi-

ous choice for reloading existing units or for new installations that require high octane.

The commercialization of the LPI-100 catalyst and the Par-Isom process provides op-

portunities for upgrading existing zeolitic isomerization units or the low-cost addition

of isomerization capacity in reﬁneries by revamping other process units or integrat-

ing with semiregenerative reforming units. The Par-Isom process provides a low-cost,

economically attractive route for reﬁners who must meet demands for higher octane or

reformulated gasoline. In most cases, the revamp can be accomplished with minimal

changes to existing equipment. The high activity of the LPI-100 catalyst can result in

up to 3 RON higher isomerate octane than can be achieved with zeolitic catalysts as

well as substantially higher throughput in revamp situations.

References

1. Physical Constants of Hydrocarbons C

-C

, ASTM Committee D-2 on Petroleum Products

and Lubricants and API Research Project 44 on Hydrocarbons and Related Compounds,

American Society of Testing Materials, 1971.

2. D. R. Stull, E. F. Westrom, Jr., and G. C. Sinke, The Chemical Thermodynamics of Organic

Compounds, Robert E. Krieger Publishing Co., Malabar, FL, 1987.

3. H. Pines, The Chemistry of Catalytic Hydrocarbon Conversion, Academic Press, New York,

1981, pp. 12–18.

4. C. D. Gosling, R. R. Rosin, P. Bullen, T. Shimizu, and T. Imai, Revamp Opportunities for

Isomerization Units, Petroleum Technol. Q. Winter 1997–1998 55–59.

5. P. J. Kuchar, and others, Parafﬁn Isomerization Innovations, Fuel Proc. Technol. 35 (1993)

183–200.

Bibliography

N. A. Cusher and others, Isomerization for future gasoline requirements, Proceedings of the

NPRA Annual Meeting, Mar. 25–27, 1990.

R. A. Meyers, Handbook of Petroleum Reﬁning Processes, 2nd ed, McGraw-Hill, New York,

1996.

A. S. Zarchy and others, Impact of desulfurization on the performance of zeolite isomerization

catalysts, Proceedings of the AIChE Annual Meeting, Nov. 11–16, 1990.

C. L. Moy and others, Benzene reduction for reformulated gasoline, Proceedings of the AIChE

Spring National Meeting, Mar. 29–Apr. 2, 1992.

Chapter 10

Reﬁnery gas treating processes

D.S.J. Jones

Introduction

Reﬁnery gas sweetening is the process used to remove the so called ‘Acid Gasses’

which are hydrogen sulﬁde and carbon dioxide from the reﬁnery gas streams. These

acid gas removal processes used in the reﬁnery are required either to purify a gas

stream for further use in a process or for environmental reasons. Clean Air legislation

now being practiced through most industrial countries require the removal of these

acid gases to very low concentrations in all gaseous efﬂuent to the atmosphere. Hy-

drogen Sulﬁde combines with the atmosphere to form very dilute sulfuric acid and

carbon dioxide to form carbonic acid both of which are considered injurious to per-

sonal health. These compounds also cause excessive corrosion to metals and metallic

objects.

The following sections of this chapter describe the processes, their chemistry and a

method of calculating the design of the more common processes.

The process development and description

The use of chemically ‘basic’liquids to react with the acidic gases was developed in

1930. The chemical used initially was Tri ethanol amine (TEA). However, as Mono

ethanol amine (MEA) became commercially more available it became the preferred

liquid reactant due to its high acid gas absorption on a unit basis. The molecular weight

of MEA is 61 while that for TEA is 149. As both react mole for mole with the acid

gases 2.44 lbs of TEA must be circulated to achieve the same absorptive capacity as

1 lb of MEA. TEA also suffers degradation at temperatures above its boiling point and

as this is its regeneration temperature MEA has replaced TEA in this service. Since

1955 numerous alternative processes to MEA have been developed. These have fewer

corrosion problems and are to a large extent more energy efﬁcient. Inhibitor systems

have however been developed which have eliminated much of the MEA corrosion

417

418 CHAPTER 10

Sweet Gas out

Contactor

Wash Water

Stripper

Reflux Drum

Sat Steam

Cond

Lean Amine

Circulating Pump

Rich Amine

Sour Gas In

Acid Gas to Sulfur Recover

To Sewer

Figure 10.1. Schematic drawing of a typical amine treating unit.

problems. Some of these newer processes also are designed to selectively remove the

S leaving the CO

to remain in the gas stream.

The process ﬂow and description of the more common processes are essentially the

same. This process ﬂow is shown in Figure 10.1.

For product gas streams which must meet lower than 1 grain per scf of H

S, MEA

must be used. This amine, however, is degraded by certain sulﬁde compounds found

in gas from thermal crackers. The most common compound that degrades MEA is

carbonyl sulﬁde (COS). MEA can, however, be regenerated by batch or continuous

vaporization and disposal of the sludge formed by the degeneration.

Referring to the above ﬂow sheet, sour gas (rich in H

S) enters the bottom of the

trayed absorber (or contactor). Lean amine is introduced at the top tray of the absorber

section to move down the column. Contact between the gas and amine liquid on the

trays results in the H

S in the gas being absorbed into the amine. The sweet gas is

water washed to remove any entrained amine before leaving the top of the contactor.

Rich amine leaves the bottom of the contactor to enter a surge drum. If the contactor

pressure is high enough a ﬂash stream of H

S can be routed from the drum to a trayed

stripper. The liquid from the drum is preheated before entering a stripping column on

REFINERY GAS TREATING PROCESSES 419

the top stripping tray. This stripper is reboiled with 50 psig saturated steam. Saturated

50 psig steam is used because higher temperatures cause amines to break down. The

S is stripped off and leaves the reﬂux drum usually to a sulfur production plant.

Sulfur is produced in this plant by burning H

S with a controlled air stream.

The lean amine leaves the stripper bottom and is cooled. The cooled stream is routed

to the contactor.

Common processes

There are several liquid solvents in commercial use for the removal of H

S and CO

from reﬁnery gases. Among the more common are the amines. These include:

MEA Mono ethanol amine

DEA Diethanol amine

DGA Diglycol amine

In addition to these amine base solvents there are also the hot potassium carbonate

process, sulﬁnol and ADIP. These latter two processes are marketed by the Shell

company and are quite common in world wide usage.

It is not the intent here to detail all of these processes with their various properties.

The three amine solvents will however be described and discussed and some mention

will be made of the other processes.

Mono ethanol amine

This is the most common acid gas absorption process. Normally 15–20 wt% MEA in

water is circulated down through a trayed absorber to provide intimate contact with

the sour gas. The rich solution is routed to a steam stripping column where it is heated

to about 250

◦

F at 10 psig to strip out the acid gases. The lean MEA solution is then

returned to the absorber.

MEA is the most basic (and thus reactive) of the ethanol amines. MEA will completely

sweeten sour gases removing nearly all acid gases if desired. The process is well proven

in reﬁnery operations.

Like all of the amine solvents used for acid gas removal MEA depends upon its amino

nitrogen group to react with the acidic CO

and H

S in performing its absorption. The

particular amines are selected with a hydroxyl group which increases their molecular

weight and lowers their vapor pressures yielding minimum solvent losses to the gas

stream. MEA is considered a chemically stable compound. If there are no other

chemicals present it will not suffer degradation or decomposition at temperatures up

to its normal boiling point.

420 CHAPTER 10

The process reactions are given below:

HOCH

= RNH

Low temp

2 RNH

+ H

−−−−

−−−−

(RNH

)

High temp

Low temp

(RNH

)

S + H

−−−−

−−−−

2RNH

High temp

Some of the degradation products formed in these systems are highly corrosive.

They are usually removed by ﬁltration or reclaimer operations. Filtration will remove

corrosive by products such as iron sulﬁde. Reclaiming is designed to remove heat

stable salts formed by the irreversible reaction of MEA with COS, CS

(carbonyl

sulﬁde and carbon disulﬁde). The reclaimer operates on a side stream of 1–3% of the

total MEA circulation. It is operated as a stream stripping kettle to boil water and MEA

overhead while retaining the higher boiling point heat stable salts. When the kettle

liquids become saturated at a constant boiling point with the degradation products

it is shut in and dumped to the drain. Union Carbide has developed a well proven

corrosion inhibitor system for MEA that allows solution strengths of 28–32%. The

inhibitor requires payment of a royalty. The inhibitor chemicals are both expensive

and hazardous for personnel to handle. The system does reduce corrosion problems to

nearly zero and allows much higher system capacity for the same size equipment. In

addition to the chemical degradation mentioned above, MEA oxidizes when exposed

to air. Storage and surge tanks must be provided with inert blanket gases such as N

or sweet natural gas to avoid this degradation.

Amine systems foam rather easily resulting in excessive amine carryover in the con-

tactor. Foaming can be caused by solids such as carbon or iron sulﬁde; condensed

hydrocarbon liquids from the gas stream; degradation products; almost any foreign

material introduced to the system such as valve grease, excess corrosion inhibitor, etc.

Some of these items such as iron sulﬁde or carbon particles are removed by cartridge

ﬁlters. Hydrocarbon liquids are usually removed by the use of a carbon bed ﬁlter on a

lean amine sidestream (about 10% of total ﬂow). Corrosion byproducts are removed

by reclaiming as noted above.

High skin temperatures on the reboiler/reclaimer tubes promote amine degradation.

Steam or hot oil used for the reboiler should be limited to a maximum of 285

◦

(140

◦

C) to avoid excessive temperatures. The reclaimer should not see hot oil or

steam above 415

◦

F (213

◦

C).

MEA is nonselective in absorbing acid gases. It will absorb H

S faster than CO

but

the difference is not signiﬁcant enough to allow its use to separate them. With the

lowest molecular weight of the common amines, it has a greater carrying capacity for

REFINERY GAS TREATING PROCESSES 421

acid gases on a unit weight or volume basis. This generally means less pure amine

circulation to remove a given amount of acid gases.

Because the solvent is in solution with water, the gas with which it comes in intimate

contact will leave the contactor at its water saturation point. If dehydration is necessary

it must be done after the MEA system.

MEA has a vapor pressure 30 times that of DEA (300 times that of TEA and DGA) at

the same temperature. This causes MEA losses of 1 to 3 lbs/mmscf compared to 1/4 to

1/2 lb/mmscf for the other amine systems using the same system design parameters.

Entrainment and leakage losses prevent low vapor pressure amines from attaining the

predicted losses.

Diethanol amine

DEA does not degrade when contacted with CS

, COS, and mercaptans as does MEA.

However for product gas streams which must meet lower than 1 grain per scf of H

MEA must be used. Because of this, DEA has been developed as a preferred solvent

when these chemicals are present in the stream to be treated.

The reaction with acid gas for any of the amines is a mole to mole reaction. As shown

in Table 10.1 the molecular weight of DEA is 1.7 times that of MEA. Even after

correcting for density it requires 1.6 lb of DEA to react with the same amount of acid

gas as 1 lb of MEA.

DEA is a weaker base (less reactive) than MEA. This has allowed DEA to be circulated

at about twice the solution strength of MEA without corrosion problems. DEA systems

are commonly operated at strengths up to 30 wt% in water and it is not unusual to

see them as high as 35 wt%. This results in the DEA solution circulation rate usually

being a little less than MEA for the same system design parameters.

The process reactions are shown below.

HOCH

NHCH

OH = R

NH = DEA

Low temp

NH + H

−−−−

−−−−

)

High temp

Low temp

)

S + H

−−−−

−−−−

High temp

Because the system has much fewer corrosion problems and removes acid gases to

nearly pipeline speciﬁcations it has been installed as the predominant system in recent

years.