Shutting Down A Sulfuric Acid Alkylation Unit Reactor

Dimerization „

Alkylation

Self-

Alkylation

H-Transfer

H-Transfer

Figure 12.5 Key reaction paths for alkylation and related reactions [9].

Low temperature (e.g., 70 °C and below) yields a higher TMP product slate. This is a kinetic effect, and isomerization reactions are lowered with decreased temperature. The distribution of TMPs is highly dependent on temperature. The preferred high octane, 2,2,4 isomer is abundant at lower temperatures. Figure 12.5 [9] covers many reactions: alkylation, dimerization, self-alkylation, b-scission and H-transfer. Figure 12.6 [14] shows that the C8 yield goes through a maximum versus T Other reactions, including oligomerization (at low T) and sequential cracking to C5-C7 (at higher T), deplete the C8 yield.

As mentioned with liquid acid catalysts, selectivity and yield remain at an almost constant high level when the olefin space velocity (OSV) is increased. However, a critical OSV can be identified above which the polymerization reactions rapidly start to increase. This is related to the available number of acid sites and the rate

343 70 °C Temperature

Figure 12.6 Alkylate products versus temperature (i-C4/1-C4= via REY catalyst) [14].

at which these sites are regenerated by hydrogen transfer reactions. At a certain OSV, all sites will be covered with hydrocarbon species, and polymerization will become dominant.

Overview of Zeolites in Alkylation

Successful alkylation catalysts have been limited to large-pore zeolites (e.g., zeolite X, zeolite Y, zeolite b, ZSM-20, etc.), probably due to the adsorption/desorption capability of TMPs. Some surprisingly good activity has been reported for medium-pore MCM-22, along with excellent selectivity; however, we surmise that stability was fairly poor. Otherwise, it would have received far more attention.

Although researchers have long sought a solid catalyst substitute for HF and H2SO4, they were not successful in finding one that met the common requirements of activity, selectivity and stability. Historically, selectivity was usually thought of as determining yields and product quality only. When researchers eventually understood that the major aging mechanism was a selectivity effect, that is, formation of polyalkylates, they began to assess a way of redefining performance criteria and devising an operating strategy to mitigate the aging issue.

12.4

AlkyClean Alkylation Process: A True Solid Acid Catalyst (SAC) Process

Many attempts have been made to apply a SAC for isobutane alkylation [15-17]; however, in most cases the catalyst and process conditions used resulted in very rapid catalyst deactivation. These attempts also failed because of the absence of viable regeneration procedures. As such, in addition to a suitable catalyst, suitable process and regeneration conditions had to be developed.

Albemarle and ABB Lummus Global began collaboration in 1996 to develop a catalyst/process combination that addresses the aforementioned product selectivity and deactivation/regeneration issues. The following targets were the focal points of our efforts:

• develop a true SAC process, not merely a hybrid process (i.e., not utilizing toxic and/or corrosive volatile compounds adsorbed onto a solid carrier as catalyst);

• ensure no or low environmental hazards compared to current H2SO4/HF processes;

• investment and operating costs equal to, but preferably lower than, current processes;

• product quality and yields equal to, but preferably higher than, current processes;

• refinery-compatible catalyst and process technology;

• a robust catalyst with regard to feed impurities;

• no halogens in the catalyst in order to reduce unit corrosion problems and maintenance.

Catalyst Selection and Development

After prescreening of potential candidates in a laboratory alkylation micro-reactor, a "true" SAC was selected as the preferred prototype. By true SAC we mean that the catalytic acid function is intrinsic to the solid itself rather than being a separate species, such as an immobilized liquid deposited on a solid substrate. Starting with this prototype, application research and development work led to novel discoveries, which yielded significant improvements in catalyst performance. The resulting AlkyClean catalyst, which has now been successfully manufactured in commercial-scale trials, is of a type well known and proven in the refining industry. It is based on zeolite Y and the porosity was tailored using a binder [18]. Other zeolites were investigated as well, but they either have an insufficient number of acid sites (e.g., zeolite beta) or a small pore aperture (e.g., ZSM-5).

The AlkyClean catalyst contains no halogens, has acid sites with sufficient strength for alkylation, yields high quality alkylate with minimal side reactions, and exhibits the required activity, stability and regenerability characteristics necessary for a successful process. It is promoted with a low Pt content to assist regeneration and hydrogen transfer.

Process Development Activities

Many bench-scale tests have been carried out to determine favorable process operating and preferred regeneration procedures. To obtain a high I/O at the catalytic sites, the test unit utilized a fixed bed recycle reactor design (Figure 12.7). With this reactor configuration, the feed I/O (i.e., external I/O) could be kept close to that used in commercial liquid acid units (e.g., 8-12). However, the internal I/O (i.e., the I/O at the inlet of the catalyst bed) can be increased to 250 or higher by recycling the isobutene-rich reactor effluent. This higher level of internal I/O, relative to the external I/O, is analogous to the situation for H2SO4 and HF units. In those units, the effective (i.e., internal) I/O at the active liquid acid sites is orders-of-magnitude higher than the external I/O. This is due to the effects of intensive mixing and the differences of isobutane and olefins in solubility and reactivity with the acid sites. In the case of H2SO4, the effective I/O may be as high as 1000.

Initially, test runs were carried out until the catalyst was deactivated to such an extent that breakthrough of olefins occurred (Figures 12.8 and 12.9) [19]. Unless otherwise stated, pure cis-2-butene was used as olefin feed. At the reaction conditions used (reaction temperature 90 °C, I/O = 250), olefin breakthrough occurred after about 8-10 hours. At this point, selectivity, as represented by RON, also started to deteriorate. Alkylate yield was 204 wt% (theoretical yield, based on equimolar reaction of C4= and i-C4) or higher until olefin breakthrough.

With reference to Table 12.7 [19], it was found that catalyst activity could be fully recovered repeatedly by vapor phase stripping with hydrogen at 250 ° C. In commercial

Figure 12.7 Alkylation bench-scale unit [19].

practice, however, such a regeneration procedure, considering the requirements for draining/heating and cooling/filling, would take too much time compared to the alkylation time period prior to olefin breakthrough. As such, less time-consuming procedures had to be developed. Regeneration conditions that are close to reaction conditions would reduce any heating and cooling periods required at the beginning and end, respectively, of a regeneration cycle.

Prior art/literature claimed that some catalysts could be regenerated in the liquid phase with dissolved hydrogen at alkylation reaction conditions. We investigated this procedure in the case of our catalyst and process combination (cf. Table 12.7, exp. 2-3). After catalyst deactivation (as indicated by olefin breakthrough) and regeneration using this procedure, the regenerated catalyst's life before olefin breakthrough could be restored only to about 40-65% of the initial fresh catalyst period. Thus, a novel regeneration method had to be developed. To this end, various alkylation/regeneration cycles were investigated. A time period with olefin addition was followed by a time period with dissolved hydrogen addition, and so on. It was

Table 12.7 Effect of regeneration procedures [19].

Exp.

Medium

T ("C)

P (bar)

Time (h)

Cat. life (h)

0

Fresh catalyst

10

1

H2 gas

250

21

1

10

1a

H2 gas

250

21

1

10

1b

H2 gas

250

21

1

10

2

i-C4 liquid with dissolved H2

90

21

66

6.5

3

i-C4 liquid with dissolved H2

115

30

18

4

Figure 12.8 RON versus time.
Figure 12.9 Olefin concentration versus time.

observed that this cyclic operation of the catalyst, at alkylation reaction conditions, could be maintained from a week to more than a month, depending on the operating severity. Under this operation, the alkylation step was halted well before any significant olefin breakthrough. A patent covering this new procedure was obtained in 1999 [20].

Figure 12.10 [19] illustrates some of the results under cyclic operation. After every hour of olefin addition, the catalyst was regenerated with dissolved H2 for one or two hours, depending on the operating I/O. Evidently, good alkylate quality, as represented by RON, can be maintained by utilizing this frequent "mild regeneration" (MR) procedure. Furthermore, high internal I/O and low Tare favorable to product quality and catalyst stability.

The catalyst slowly deactivated during the cyclic operation and, after an extended period (two to four weeks), olefin breakthrough occurred during an alkylation cycle. At this point, the catalyst activity was fully restored by treatment with vapor phase H2 at 250 °C, as described previously. Even after this "high temperature regeneration" (HTR) procedure was carried out 15 times during a pilot run of more than six months with the same catalyst sample, catalyst activity could be fully recovered. A "coke" burn with air was not required.

| I/O = base value

I/O = higher than base value

V. !

Time (hrs)

Figure 12.10 Cyclic operation; RON versus time.

Optimization of Process Conditions

After optimizing the catalyst and process conditions, RON numbers as high as 97 to 98 could be obtained for a prolonged time period. Figure 12.11 and Table 12.8 [19] illustrate some of the results.

At sufficiently high I/O, selectivity was optimized by varying reaction temperature. When the temperature was lowered, less C5-C7 and more of the desired C8 compounds were formed. Also, at lower temperature, isomerization of high-octane TMPs to low-octane DMHs was reduced. However, when the temperature was reduced too much at a given olefin space velocity, overall olefin conversion was

Figure 12.11 Cyclic run with optimized catalyst; RON versus time [19].
Table 12.8 Product distribution and properties versus temperature [19].

T ("C)

C4= conv.

RON

Yield

C5-C7

C8

C9+

TMP/DMH

(wt%)

(wt%)

(wt%)

(wt%)

(wt%)

80

100

95.4

213

22.0

74.2

3.8

6.0

70

100

96.5

208

17.5

78.2

4.2

7.5

65

100

97.0

207

16.0

79.4

4.5

8.2

60

100

97.6

205

14.3

80.7

5.0

9.2

55

100

98.1

204

13.4

81.4

5.1

10.3

reduced and formation of C9 + compounds sharply increased, leading to more rapid deactivation of the catalyst. Alkylate yield (based on olefin feed) was always somewhat higher than the theoretical 204 wt%. This can be explained by dispropor-tionation of C8 with i-C4 via carbenium ions on the catalyst surface, leading to consumption of more than one i-C4 molecule per C4 olefin molecule. One drawback is that disproportionation leads to the increased formation ofthe less desirable C5-C7 compounds. At lower temperature, disproportionation activity is lowered, leading to less C5-C7 products and therefore lower yields.

Other observations of this test work, with respect to key alkylate product properties, were that neither the Reid vapor pressure (RVP) nor density deviated significantly from values that would be obtained via liquid acid alkylation. Further, acid-soluble oils (ASO), formed as contaminant side products in the case of liquid acid processes, could not be detected among the reaction products in our SAC testing. Compared with the liquid acid technologies, this effect results in both lower feed consumption per unit of alkylate production and eliminates generation of a by-product that can be difficult to dispose of.

Effect of Feedstock Variation

In commercial practice, there will be significant differences in olefin feed composition. Under the AlkyClean process cyclic operation, high RON is obtained over a prolonged time period with various feeds. The use of a refinery-sourced MTBE raffinate gave similar results (alkylate yields and product quality) to a pure cis-2-butene feed (Table 12.9). The MTBE raffinate contained about 26wt% trans-2-butene, 15 wt% cis-2-butene, 12 wt% 1-butene, 2 wt% isobutene, 40 ppmw (parts per million by weight) of various oxygenates and 3 ppmw of sulfur (balance isobutane and n-butane).

In HF alkylation, the processing of 1-butene leads to a significantly lower RON. Therefore, to avoid this octane loss, HF units often employ upstream selective hydrogenation/isomerization to isomerize 1-butene to 2-butene.

The addition of about 25% isobutene on olefins resulted in a loss of less than 0.5 RON. This loss may be attributed to a somewhat higher formation level of C5-C7 and C9 + compounds. For H2SO4 alkylation, the same amount of feed isobutene would lead to a loss of about 1 RON.

Table 12.9 Effect of various butene feeds on the AlkyClean process performance at identical process conditions [19].

Feed cis-2-Butene MTBE raffinate 25/75 i-C4=/c/s-2-butene

Feed cis-2-Butene MTBE raffinate 25/75 i-C4=/c/s-2-butene

Table 12.9 Effect of various butene feeds on the AlkyClean process performance at identical process conditions [19].

RON

95.5

95.6

95.1

Yield (wt%)

211

210

215

C5-C7 (wt%)

20.2

20.3

25.3

C8 (wt%)

74.5

74.4

67.7

C9+ (wt%)

5.3

5.3

7.0

TMP/DMH

6.1

6.2

6.5

Table 12.10 Estimated ii

mpact of feedstock variation [19].

AlkyClean Process

H2SO4

HF

1-Butene (up to 100vol.%) —

up to -4.0 RON

Isobutene (25 vol.%)

-0.5 RON

-1.0 RON

-0.5 RON

Propylene (30 vol.%)

-1.0 RON

-1.5 RON

-1.0 RON

Similarly, after blending about 30vol.% of propylene with cis-2-butene, the RON loss was less than 1 number. With H2SO4 alkylation, similar amounts of propylene would lead to a RON about 1.5 lower. Table 12.10 summarizes the estimated impact of feedstock variation on RON relative to a pure cis-2-butene feedstock for the AlkyClean process and liquid acid technologies. Based on these results, it can be concluded that our new SAC technology is less sensitive to feedstock variation regarding product quality than either liquid acid technology.

Effect of Impurities

According to the open literature, other solid acid alkylation catalysts are generally susceptible to poisoning/deactivation by water and other common feed impurities (e.g., oxygenates, sulfur compounds, dienes, etc.), thus necessitating (potentially costly) feedstock pretreatment for their removal. In some cases, this requirement is further mandated by the potential corrosion problems associated with the use of halogens in the catalyst system.

In liquid acid units, these impurities increase both liquid acid consumption and side product formation. The feed may also require drying and additional pretreat-ment, depending on the acid employed and the level of contamination.

In contrast, the AlkyClean solid acid catalyst contains no halogens and it is very robust with regard to water and other potential feed impurities. This was observed even after exposure of the catalyst to high concentrations of oxygenates (250-700 ppmw), sulfur compounds (200-1200 ppmw) and butadiene (400-1800 ppmw). Moreover, after any observed deactivation from these impurities, the catalyst could always be restored to full activity via HTR (i.e., treatment with H2 at 250 °C). Furthermore, in

Figure 12.12 Simplified block diagram of the AlkyClean process [19].

commercial practice, after a feed upset requiring HTR, the catalyst does not require additional treatment (e.g., addition of halogens) before being put back on-line. This simplicity reduces turnaround time and eliminates procedures that carry the risk of corrosion problems. In conclusion, because of these catalyst attributes, the AlkyClean process feed pretreatment requirements are significantly less than those for processes based on alternative solid acid catalysts. Required pretreatment levels are projected to be no more than equivalent to that normally associated with sulfuric acid units.

The process flow scheme for the AlkyClean process is similar to that employed for current liquid acid technologies. As illustrated in the block flow diagram in Figure 12.12, the process consists of four main sections: feedstock pretreatment (optional, depending on contaminant level), reactor system, catalyst regeneration and product distillation.

Reactor System/Catalyst Regeneration

AlkyClean reactors operate in the liquid phase in the temperature range 50-90 °C, thereby eliminating the costly refrigeration requirements associated with liquid sulfuric acid technologies. To achieve a high octane alkylate and limit by-product production, H2SO4 units typically utilize a total reaction section feed (external) I/O of between 8/1 and 10/1, while HF units run at an I/O of about 12-15/1. In comparison, without any alkylate octane debit, this I/O for the AlkyClean process is in the range 8-10/1, comparable to the H2SO4 process, which operates at a significantly lower temperature (Table 12.11). The ability to operate the AlkyClean process at this low I/O is important for two reasons. First, it enables a cost competitive process, as the fractionation requirement associated with isobutane recycle is a major capital investment and operating cost component. Second, it facilitates incorporation of AlkyClean technology in the revamp/de-bottlenecking of an existing liquid acid unit, without major modification to the "back-end" fractionation/recycle facilities.

Key to the AlkyClean technology's superior performance is the coupling of a newly developed catalyst with a novel alkylation reactor system, which minimizes the peak

AlkyClean process

H2SO4

Operating temperature (°C) External I/O

olefin concentration in the reaction zone (i.e., maximizes the internal I/O) without requiring extremely high and economically non-viable reactor effluent recycle rates. This is accomplished by utilizing serial reaction stages and a unique (but mechanically uncomplicated) reactor design, which allow for distributed olefin feed injection and the operating conditions essential to both prevent rapid catalyst deactivation and attain high product quality.

In the AlkyClean process, a reactor may remain on-stream for up to 12 hours before olefin breakthrough. In practice, a reactor is regenerated safely before the expected olefin breakthrough time. Multiple reactors enable continuous alkylate production, as individual reactors cycle back and forth between on-line alkylation and mild regeneration, following the inventive procedure established during our process development effort. During mild regeneration, olefin addition is stopped and hydrogen is added to achieve a low reactor concentration of dissolved hydrogen, while maintaining liquid phase alkylation reaction conditions. This enables a seamless switchover between operations and minimizes energy consumption requirements. Over time, however, there is a gradual loss ofcatalyst activity. To recover this activity, with a frequency depending on the operating severity, a reactor is taken off-stream for high temperature (250 °C) gas-phase regeneration with hydrogen. To allow for high temperature regeneration while maintaining full continuous alkylate production, an additional "swing" reactor is provided.

Figure 12.13 depicts this cyclic operation of the AlkyClean reactor section. Notably, our processing scheme does not require any transfer of catalyst, either between reactor stages or to a separate regeneration vessel. For operability reasons, this was a fundamental design choice made early in the process development effort. Furthermore, the use of a swing reactor provides additional maintenance flexibility and enables the unit to stay on-stream when catalyst replacement (after years of operation) inevitably becomes necessary. At the end of its useful life, the catalyst is returnable to the manufacturer, Albemarle, eliminating any potential catalyst disposal problem for the refiner. The noble metal Pt is reclaimed and the remaining material can be used in the construction industry.

AlkyClean Process Demonstration Unit

Construction of an AlkyClean process demonstration unit at Fortum's facilities in Porvoo, Finland, was completed in 2002. Figure 12.14 shows the process flow schematic of the demo unit, which contains all of the key elements of our proposed commercial design. Three reactors are included - two under cyclic operation (i.e., alternating between alkylation and mild regeneration) allow for continuous production

' Olefin

Reactor effluent w \Continuously

Occasionally

Mild regeneration

Mild regeneration

Regeneration at 250 °C (1 reactor)

Figure 12.13 AlkyClean reactor operating scheme [21, 28].

of alkylate and the third allows for swing reactor high temperature regeneration. The process design of the demo reactor section has been set to achieve operating conditions and compositional profiles analogous to those for a commercial design.

Figure 12.15 is a photograph of the installed demo reactor section. The demo reactors are sufficiently large and proportioned to allow for reliable scale-up. As such, each demo reactor represents a "core" of a much larger reactor and provides for the necessary hydrodynamic similarity (e.g., equivalent superficial velocities) to a full commercial-scale reactor system. Equally important, these reactors use AlkyClean catalyst produced in commercial manufacturing trials, not "developmental-scale" catalyst with characteristics that may not be fully duplicable under commercial-scale production conditions.

Demo Unit Operation

The AlkyClean process demonstration unit was built and operated with the following objectives:

• first and foremost, to demonstrate the operability and performance of our new catalyst and process technology;

• to optimize the process through parametric studies;

• to confirm key process, reactor and performance parameters;

• to verify its correspondence to bench scale unit performance;

• to tune key computer models;

• to test alternate olefin feedstocks.

After mechanical completion, the demo unit went through a shakeout and start up period of about one month. During this period, procedures were refined and proven for the in situ activation of the catalyst and the reliable start up of the reactor section.

Alkylation Reactor Alkylation Reactor Alkylation Reactor

Mild Regeneration Alkylation High Temp Regeneration

V » r V ■ ■ I ■ l ■ ■■■■■■■■■■

Light Ends Separation

Tower

Mild Regeneration Alkylation High Temp Regeneration

V » r V ■ ■ I ■ l ■ ■■■■■■■■■■

Light Ends Separation

LJ.J

Hot Oil

Olefin 11 Hydrogen^

H Closed 1X1 Open

High Temp Regeneration

Alkylation

Mild Regeneration

Alkylate

Make-up iC,

\l-butan«

|_1 W|

ght nds

Figure 12.14 Process flow schematic of the demo unit [22],

Figure 12.15 Demo reactor section [22].

Subsequently, over several months, the demo unit reliably operated on a nearly continuous basis (except for holiday shut-down periods) while producing high quality product. The unit utilizes piped slipstreams of actual refinery feeds. To date, we have tested both predominantly C4 olefin and mixed C3/C4 olefin feed streams, and the demo has produced alkylate of comparable quality to that from Fortum's Porvoo Refinery HF alkylation unit.

In addition to proving the operability of the process, key aspects of the technology have also been demonstrated. First and foremost is continuous cyclic operation -alternating reactors between periods of alkylation and mild regeneration - for periods of up to four weeks before taking a reactor off-line for HTR. In doing so, the durability of the AlkyClean catalyst has been demonstrated over hundreds of cycles of mild regeneration and multiple high-temperature regenerations. Regenerated catalyst samples from the demo unit have also been tested in Albemarle's bench-scale unit under benchmark conditions, further confirming our ability to repeatedly regenerate the catalyst and re-establish fresh catalyst activity and performance.

After obtaining performance data over a wide range of conditions, which supported our identified objectives, the demo unit operation was suspended during late 2003. This enabled a more economically efficient and intensive bench-scale unit effort to focus on identified opportunities for catalyst and process optimization that stemmed from insights gained through the analysis of the demo unit's performance. The result was a significantly improved catalyst with an activity advantage that is discussed below. Operational changes were also tested and refined, resulting in the further optimization of process performance.

Figure 12.16 provides a performance comparison of the second-generation catalyst relative to the original (i.e., old) prototype catalyst. This catalyst provided for about a 15-20 °C activity advantage over the old catalyst (i.e., the difference in operating

55 60 65

TEMPERATURE °C

Figure 12.16 Comparison of first- and second-generation AlkyClean catalyst [22].

temperatures for equivalent conversion). In commercial practice, this substantial activity advantage can be used to operate at higher olefin throughputs and/or enable lower operating temperatures that will result in a higher octane alkylate (by up to about 1 RON). Albemarle has made an even more significant improvement in catalyst performance with its new third-generation catalyst.

Based on this successful bench scale program, operation of the demo unit resumed in 2004, following commercial trial manufacture of the newly improved catalyst, demo reactor catalyst replacement, and the completion of required unit modifications to incorporate operational improvements. Several months after its restart, the demo unit continued to run smoothly and continuously under full cyclic operation, with periodic rotational HTR of the reactors.

To date, the benefits of the operational improvements have been verified and the improved activity of the second-generation catalyst has been confirmed, along with stability and full activity recovery over multiple HTRs. Representative performance data from the current demo unit operation processing refinery C4 olefins is presented in Figures 12.17 to 12.19 - for alkylate RON, RVP and C5 + yield, respectively. This data, based on automated sampling from the on-line analytical system, show both the high product quality achieved and the stability of the cyclic process over time.

The demo unit successfully operated to fully verify the process's improved performance under targeted commercial design basis conditions. Further, it allowed for parametric optimization and confirmation of bench-scale unit correspondence, and provided necessary support for our correlation and modeling effort. Based on all the positive progress to date, this demo unit campaign was successfully completed after three months.

Finally, after many years of dedicated research carried out to develop a process that fully fulfills the targets set at the start of the quest, commercialization of the technology was started [23-25].

t-

3 -S

OT s

n 0 0

«

0/ % \

CC0

o

o Rx inlet o Inter. Rx

Figure 12.17 Demo unit performance processing refinery C4 olefins: RON [22].

3.000

3.000

Start :arted i

to o3

T

o5

«5

O

o Rx inlet o Inter. Rx • Rx outlet

Figure 12.18 Demo unit performance processing refinery C4 olefins: RVP [22].

Time

Figure 12.18 Demo unit performance processing refinery C4 olefins: RVP [22].

Examples of other inventive concepts that were developed during this quest have been mentioned in the patent literature, such as the use of a high-performance zeolite [26], combined use of zeolites imbedded in a mesoporous material [27] and continuous alkylation with intermittent regeneration [28].

Competitiveness versus Liquid Acid Technologies [25]

Table 12.12 provides an overview of the competitiveness of the AlkyClean process versus the established liquid acid technologies. The performance and economics ofthe

200.000

150.000

100.000

50.000

0.000

Yield C5+ on olefin

Start " tarted 6

«•o© *ioC

ci»

CO *CD

O

o Rx inlet ® Inter. Rx • Rx outlet

Figure 12.19 Demo unit performance processing refinery C4 olefins: C5 + yield based on olefin [22].

AlkyClean process are fully competitive with current liquid acid technologies. High quality product has been produced in an operation that has proven to be reliable and robust. Sensitivity to feedstock variation is low and tolerance to impurities is high. The economic competitiveness of the new SAC process is enhanced by its low mechanical complexity and the use of common (i.e., non-proprietary) refinery process equipment.

Table 12.12 Comparison of AlkyClean process with liquid acid technologies.

Parameter Modern sulfuric acid Modern HF acid AlkyClean process

Base conditions: C4= feedstock

Product RON

95

95

95

Product MON

92.0-92.5

92.5

92.5

Alkylate yield

Base

Base

Base or better

Total installed cost, ISBL

Base

85% of base

85% of base

Total installed cost,

Base

70% of base

< 50% of base

including OSBL

(regeneration facilities,

and/or safety installations)

Feed treatment

Base

Higher

Base

Product treatment

Yes

Yes

No

ASO yield

Up to 2 wt%

Less

None

on olefins

Equipment maintenance

High

High

Very low

Corrosion problems

Yes

Yes

None

Reliability and on

Average

Average

Expected above

stream factor

average/high

Turnarounds frequency/

Varies/longer

Varies/longer

Match FCC

duration

or better/shorter

OPEX

Base

Site specific

Base

Typically lower

The low pressure and mild temperatures employed, along with the absence of either a corrosive or erosive environment, allow for the use of carbon-steel construction.

Based on an in-house benchmarking effort, we estimate that the total installed cost (TIC) for an AlkyClean process unit is about 10-15% lower than that for the equivalent H2SO4 unit. This comparison excludes off-site costs. When acid regeneration facilities for the H2SO4 unit are included, the SAC unit's cost becomes substantially less. We project that this level of investment for the SAC process is about on par with the cost for an HF unit. With respect to total cost of production, the results indicate that the requirements for the new AlkyClean process are comparable to that of the H2SO4 process. For both technologies, catalyst consumption cost is a significant component of the variable cost. For the AlkyClean process, this cost was conservatively based on a minimum ultimate catalyst life. Thus, there is considerable upside potential, which would result in the reduction of the AlkyClean process's total production cost. Compared to the AlkyClean and H2SO4 processes, production costs for HF units may be judged, on the surface, to still be somewhat lower; however, the increased costs for maintenance, mitigation and monitoring, among others, that the HF technology requires offset any perceived advantage.

12.4.10

Environmental, Cross-Media Effects

Table 12.13 shows the environmental benefits of the AlkyClean process.

Table 12.13 Summary of cross-media effects, waste and safety.

Sulfuric acid

Hydrofluoric acid

AlkyClean process

Spent catalyst

High

About 100x lower

About 1000X lower

production

(Pt of spent catalyst can be reclaimed after many years of operation)

Product

Acid removal

Acid removal needed

None

treatment

needed

With caustic

Creates waste

Creates waste water

None

and/or lime

water and sludge

and sludge

Acid-soluble oil

Up to about

Less then sulfuric,

None

production

2 wt% on olefin feed

but still significant

Corrosion

Yes

Yes

None

issues

Maintenance

High

High

Low

Reliability

Moderate

Moderate

High

Safety

Unit-specific safety

Very specific safety

No special precautions

precautions as well as

precautions required

other than those for

transport of H2SO4

that extend throughout

any refinery process

precautions

refinery and adjoining

unit (inert catalyst)

(accidental acid spills)

neighborhoods (accidental acid spills)

Conclusion

The refining and petrochemical industry has seen many acid-catalyzed reactions evolve from liquid acids to solid acids, and each time the benefits were multi-fold. This is one more example of this evolution. As industry adjusts to the psychological hurdle of "first-of-its-kind" technology, the demise of liquid acid alkylation processes will be the cornerstone in making refining a very safe operation.

References

1 Weast, R.C. Astle, M.J. and Beyer, W.H. (eds) (1986) Handbook of Chemistry and Physics, 67th edn, CRC Press, Boca Raton, FL.

2 Hyman, H.H., Kirkpatrick, M. and Katz, J.J. (1957) J. Am. Chem. Soc., 79, 3668.

3 Paul, M.A. and Long, F.A. (1957) Chem. Rev., 57, 1.

4 Simons, J.H. and Dresdner, R.D. (1944) J. Am. Chem. Soc., 66, 1070.

5 Simmons, M.C. and Kelly, T.R. (1991) Gas Chromatography, Second International Symposium, Academic Press, New York.

6 Cupit, C.R., Gwyn, J.E. and Jernigan, E.C. (1962) Petro/Chem. Eng., 34, 49.

7 Pfile, M.E. (1987) Alkylation process fundamentals, refining process services. Presented at Symposium on Technology for Gasoline Production and Octane Enhancement, Houston, TX, October 8.

9 Corma, A. and Martinez, A. (1993) Chemistry, catalysts, and processes for isoparaffin-olefin alkylation: actual situation and future trends. Catal. Rev.-Sci. Eng., 35 (4), 483-570.

10 Feller, A. and Lercher, J.A. (2004) Chemistry and technology of isobutane/ alkene alkylation catalyzed by liquid and solid catalyst. Adv. Catal, 48, 229-295.

13 Weitkamp, J. (1980) Proc. Int. Conf Zeolites, 5, 858.

14 Gardos, Gy., Pechy, L., Redey, A. and Sokorai, I. (1980) Hung. J. Ind. Chem., Veszprem, 8, 371.

15 Juguin, B. et al. (1988) French Patent 2,631,956.

18 van Broekhoven, E.H. and Mas Cabré F.R. (2005) US 6,855,856.

19 van Broekhoven, E.H. et al. (2001) A new solid acid isobutane alkylation technology: AlkyClean. Akzo Nobel Catalysts Eco-Magic Symposium, Noordwijk, The Netherlands, paper G9, June 10-13.

20 van Broekhoven, E.H., Mas Cabré, F.R., Bogaard, P., Klaver, G. and Vonhof, M. (1999) US 5,986,158.

21 D'Amico, V.J. et al. (2002) The AlkyClean process: a new solid acid catalyst gasoline alkylation technology.

NPRA Spring 2002 Meeting, paper AM-02-19.

22 D'Amico, V.J. et al. (2004) AlkyClean solid acid alkylation: will it finally become a reality? Akzo Nobel Catalysts Scope Symposium, Florence, Italy, June 21-23.

23 D'Amico, V. et al. (2006) Consider new methods to debottleneck clean alkylate production. Hydrocarbon Processing, February, p. 65.

24 D'Amico, V.J. et al. The AlkyClean process - demonstrated new standard for alkylation technology. AIChE Spring National Meeting 2006, Orlando, FL.

25 D'Amico, V. et al. (2006) The AlkyClean 27 Shan, Z. et al. US Appl 20060128555 alkylation process - new technology (filed 2/8/06).

eliminates liquid acids. NPRA Spring 28 van Broekhoven, E.H., Sonnemans, 2006 Meeting, paper AM-06-41. J.W.M. and Zuijdendorp, S. (2007)

26 Yeh, C.Y. et al. (2005) US 6,844,479. US 7,176,340.

10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook


Post a comment