Lloyd L. Handbook of Industrial Catalysts

Подождите немного. Документ загружается.

228 Chapter 6

to the active mixed sulﬁde. The subsurface cobalt/aluminate does not sulﬁde.

The same mechanism probably applies to nickel/molybate and the metal ratio

should be controlled to avoid the formation of nickel sulﬁde (Ni

Catalyst can be presulﬁded in the reactor in a controlled manner using the

following procedure:

• The reactor is freed from moisture by purging with dry air, before purg-

ing with nitrogen to remove oxygen. Hydrogen is then circulated at the

operating pressure of the process and at 175

• Feed containing 1–2% sulfur is circulated and the temperature is in-

creased to 200–250

C to start the sulﬁding process.

• Some of the sulfur in the feed is incorporated into the catalyst as metal

sulfides. The level of sulfur in the feed leaving the catalyst bed reaches a

constant level when this stage has come to quasi-equilibrium. The tem-

perature is then increased to 300

C and the process repeated, giving fur-

ther sulfiding of the catalyst. Normal operation can be started after this

stage.

Presulﬁding normally takes 10–20 h to complete. Sulfur uptake depends on

the metal oxide content of the catalyst used but is normally in the range 7–10%.

Good gas distribution through the catalyst bed is very important to sulﬁde even-

ly all of the catalyst. Pre-sulfurised catalysts were introduced to avoid problems

associated with poor gas distribution and temperature control. They also had the

added bonus of a quicker and simpler start-up procedure. The early pre-

sulfurised catalysts were prepared by treating green catalyst with elemental sul-

fur to fill the pores within the catalyst particle. The active sulﬁdes were formed

subsequently, when the catalyst was heated in a stream of hydrogen in the reac-

tor. This procedure led to a signiﬁcant temperature rise during start-up, which

resulted in the vaporization and subsequent deposition of sulfur in downstream

equipment.

Improved pre-sulfurised catalysts containing polysulﬁdes or metal oxysul-

ﬁdes were later developed to overcome these problems.

Polysulﬁdes still react

rapidly with hydrogen during conversion of the cobalt, nickel, and molybdenum

oxides to sulﬁdes and produce a temperature rise in the catalyst bed. On the oth-

er hand, metal oxysulﬁdes react slowly over a much wider temperature range

and do not produce an exotherm.

Presulfurized catalysts can be commissioned satisfactorily with a ﬂow of

hydrogen in the temperature range 150–300

C. There is no signiﬁcant tempera-

ture rise in the catalyst bed and water evolves gradually between 200–300

Refinery Catalysts 229

6.4.2.4. Catalyst Operation

The early processes used to hydrotreat naphthas only required single beds, but

modern plants treating heavier feeds are more complicated. Several catalyst beds

with a variety of catalysts are usually required.

Interbed cooling with cold gas

quench or liquid feed may also be necessary. Hydrotreating reactions are exo-

thermic and cooling is essential to avoid catalyst damage.

Different catalysts are used in separate beds, because several different reac-

tions are taking place at the same time. Top beds operate as guards that remove

metals from vacuum gas oils and residues, as well as promoting sulfur and ni-

trogen removal reactions. Guard catalysts consist of large particles which have

large pores to absorb metal porphyrins and Conradson carbon. The large particle

size is needed to limit any increase in pressure drop caused by trapped impuri-

ties. There is usually an improvement in hydrotreating performance if the mo-

lybdenum and nickel content of these catalysts are high. Pressure drop is still the

most signiﬁcant operating problem.

A selection of the different sulfurous and nitrogenous impurities in a typical

feed and the reaction sequences involved in their removal is shown in Table

6.11.

Coke is gradually deposited onto the catalyst surface during operation, even

when using naphtha and middle distillates as feed, and this leads to a reduction

in catalyst activity and an increase in pressure drop throughout the reactor. The

plant may need to be shut down for catalyst regeneration to restore activity and

to reduce pressure drop. Some liquid feeds do not vaporize completelyunder

operating conditions and uneven liquid distribution through the beds also leads

to poor performance.

6.4.2.5. Catalyst Regeneration

A typical procedure to regenerate a hydrotreating catalyst in the reactor is as

follows:

• Isolate feed to the reactor and purge all volatile organic material from the

catalyst with hydrogen.

• Purge hydrogen from the reactor with steam and increase temperature to

350

• Add air to the steam, keeping the maximum bed temperature to 400

◦

There is a localized hot zone passing through the bed as the residual car-

bon is burnt to carbon dioxide, and as the hot spot passes through the bed,

increase the temperature to 450

• Maintain temperature at 450

C until no carbon dioxide is detected at the

bed outlet and then purge residual oxygen from the reactor with steam.

230 Chapter 6

TABLE 6.11. Hydrotreating Reactions.

Impurity

Hydrodesulfurization (HDS)

Mercaptans

RSH + H

→ RH + H

Sulfides

RSR + 2 H

→ 2 RH + H

Disulfides

RSSH + 3 H

→ 2 RH + 2 H

Thiophene

+ 4 H

→ C

+ H

Benzothiophene

+ 3 H

→ + H

Dibenzothiophene

+ 2 H

→ + H

Hydrodenitrogenation (HDN)

Amines

RNH

+ 2 H

→ RH + NH

Pyrrole

+ 4 H

→ C

+ NH

Pyridine

+ 5 H

→ C

+ NH

Indole

+ 4 H

→ + NH

Quinoline

+ 4 H

→ + NH

Carbazole

+ 4 H

→ + NH

Note: Olefins are also saturated during its hydrotreating process.

The catalyst can be reused following regeneration provided that it has not

been deactivated by metals such as vanadium, nickel, sodium from the feed or

iron scale. Deactivated catalysts can contain between 10 and 30% of metal de-

posits, which may be recovered for economic or environmental reasons. Oxida-

tion of cobalt or nickel sulﬁdes to sulfate, during in situ regeneration, can gradu-

ally lead to deactivation of the catalyst.

Refinery Catalysts 231

With the large volumes of catalyst now used it has become usual to regen-

erate catalyst externally in special rotary kilns to save time and energy.

In situ

regeneration can often take up to 20 days. External regeneration can be quicker

and gives better temperature control and gas circulation, so that regenerated cat-

alyst can be cleaner and almost as active as new catalyst. At the same time, dis-

charged catalyst can be separated into cobalt or nickel types and different sized

particles.

Apart from the possible need to hold a spare catalyst charge, external regen-

eration is more economic. However, as most reﬁneries operate several hy-

drotreating units, it is usual to cycle regenerated catalysts from the less severe

naphtha hydrotreating to middle distillate and ﬁnally to residue treatments.

External regeneration in the rotary kiln requires two stages:

• Sulfur is removed from the catalyst at 200–250

C by gentle oxidation of

sulﬁdes to sulfur dioxide in a ﬂow of air. Temperature control is im-

portant to avoid complete oxidation of sulﬁde to sulfate at high tempera-

tures.

• Carbon can then be oxidized in air at 400

C, taking care to limit the oxy-

gen content of the hot gas and avoid temperature runaway, which would

deactivate catalyst.

The operating cycle time before regeneration is needed varies considerably

depending on the quality of the feed being used, as shown in Table 6.9. The type

and size of catalyst is also important. For example, a naphtha hydrotreater can

probably operate for at least two years before regeneration, whereas diesel oil

and residue hydrotreaters only operate for a maximum of about 350 and 250

days, respectively.

6.5. HYDROCRACKING

Hydrocracking is a process for the conversion of heavy gas oils, including low-

value fractions not suitable for catalytic cracking, into light gases, gasoline, jet

fuel, and diesel oil. The process operates at a high hydrogen pressure (60–200

atm) and moderate temperatures in the range 350–450

C. Dual-function cata-

lysts are used to combine the hydrogenation activity of hydrotreating catalysts

with an acidic cracking support.

The process was introduced in modern reﬁneries by the Union Oil Company

(Unocal) in 1964,

but dates back to the 1920s when coal hydrogenation pro-

cesses were developed in Germany and the United Kingdom. Gas oil was hy-

drocracked by Standard Oil at their Baton Rouge Reﬁnery, during the 1930s

using the German catalysts. The process produced aviation gasoline during

World War II in both the United States and the United Kingdom.

232 Chapter 6

The nickel catalysts supported on silica/alumina introduced by Chevron

the low-temperature isocracking process in 1959 were used to upgrade distillates

and to produce low-molecular-weight isoparafﬁns for high-octane gasoline.

When the Union Oil Company used a more acidic zeolite support in 1964 the

process became more versatile and a wider range of high-boiling-point feeds

including middle distillates could be converted to high-quality products.

Or-

ganic nitrogen and sulfur compounds present in these feeds were also hydrogen-

ated, forming ammonia and hydrogen sulﬁde, and removed from the products.

Since the 1970s there has been a continuous improvement to the process and

catalysts, as lower-quality feeds have been upgraded.

6.5.1. Hydrocracking Processes

Low-value distillates, including heavy-cycle oils from FCC units, thermal and

coker gas oils, and other heavy-vacuum gas oils, are cracked to produce naph-

tha, jet fuel, and diesel oils. The reaction mechanism is the same as in catalytic

cracking and some aromatic products are also hydrogenated.

There are two steps in the hydrocracking process. Following hydrotreatment

to remove organic nitrogen and sulfur compounds, the feed is cracked. The main

hydrocracking reactions in the second step which proceed via a carbenium ion

mechanism are as follows:

• Dehydrogenation of long-chain paraffins; isomerization and cracking of

the resulting oleﬁns at the third carbon atom; hydrogenation of oleﬁns to

isoparaffins.

• Stepwise ring hydrogenation and hydrocracking of polycyclic aromatics.

• Dealkylation of alkyl aromatics and naphthenes.

• Saturation and cracking of aromatics.

• Hydrogenation of coke precursors formed during the hydrocracking reac-

tions.

The proportion of different products formed depends on the catalysts used.

Weak hydrogenation catalysts with a strongly acid support produce low-boiling

gasoline with a high iso-to-normal paraffin ratio. Catalysts with high hydrogena-

tion activity and low support acidity provide middle distillates. The hydrogen

added for the hydrogenation reactions suppresses coke formation, which would

quickly deactivate the catalyst. Catalysts operate for 2–6 years before being re-

generated or replaced.

Feed and hydrogen are circulated through the catalyst at the conditions se-

lected for the range of products required. Reaction temperatures in the range

250–500

C and pressures in the range 60–200 atm have been used. Hydrotreat-

ing and hydrocracking reactions are exothermic and the catalyst is loaded into

several beds so that temperature can be controlled by the addition of a cold hy-

drogen quench. Several different process designs have been used since 1959.

Refinery Catalysts 233

TABLE 6.12. Hydrocracking Processes.

Feed Aromatic cycle oil, heavy gas oil, thermal/coker

gas oil

Product LPG, gasoline, middle distillates

Single-stage process/with recycle:

One or two beds, relatively low sulfur and nitrogen in feed

Feed temperature (

C) 350–425

Pressure (atm) 60–200

Conversion (%) 40–70

Aromatics (vol%) ~15

Catalyst (sulﬁded) Hydrotreating plus ammonia tolerant hy-

drocracking types.

Two-stage process:

Two reactors, nitrogen and/or sulfur compounds can be removed between beds.

First stage:

Feed temperature (

C) 350–425

Pressure (atm) 60–200

Conversion (%) 40–50

Catalyst (sulﬁded) Early units: nickel molybdate/alumina;

high nitrogen conversion;

little cracking.

Later units: nickel tungstate/alumina;

High nitrogen conversion

up to 40% cracking.

Second stage:

Feed temperature (

C) ~260–375

Pressure (atm) 60–200

Conversion (%) 40–50 (better aromatic conversion than single

stage)

Aromatics (vol%) ~2

Catalyst Early units: sulﬁded nickel molybdate/silica

alumina catalysts.

Later units: Catalysts with high zeolite content.

These can be divided into two different groups,

and details are given in Table

6.12.

6.5.1.1. Single-Stage Processes

In the early hydrocrackers, the feed was hydrotreated in a separate unit to re-

move organic nitrogen compounds, which are catalyst poisons. A nick-

el/tungsten/silica/ alumina hydrocracking catalyst was then used in a single ves-

sel operating in the range 350–450

C. Conversion was in the range 40–70%, and

unconverted feed was recirculated after fractionation. These units operate with

relatively low levels of impurity in the feed, but recycle gas contains hydrogen

sulﬁde. However, early units still operating now use the new zeolite-based cata-

lysts to treat higher-boiling feeds.

234 Chapter 6

Design was improved by combining the two stages. Feed from the hy-

drotreating catalyst passed directly to the hydrocracking catalyst for conversion

and unconverted feed was recycled back to the hydrocracking section. Hy-

drotreating and hydrocracking catalysts were used so both feed and hydrogen

passed directly from one reactor to the other. This meant that after hydrotreat-

ment, the hydrogen still contained ammonia and hydrogen sulﬁde when it en-

tered the hydrocracker. This was usually acceptable because ammonia did not

poison the zeolite-based catalysts being used at the time.

6.5.1.2. Two-Stage Processes

In two-stage processes mainly hydrotreating and some cracking takes place in

the ﬁrst reactor at 350–425

C. The cracked products are then fractionated before

the unconverted feed passes to the second reactor which contains hydrocracking

catalyst. Recycled gas is also scrubbed to remove ammonia and hydrogen sulﬁde

if palladium-based hydrocracking catalysts are used in the second reactor. Nick-

el molybdate or tungstate catalysts must be sulﬁded before use, so there is no

need to remove hydrogen sulﬁde from recycle hydrogen. Hydrocracking cata-

lysts can operate with or without sulfur at 260–375

C. Feeds with high nitrogen

content can therefore be hydrocracked at a lower second-stage temperature and

higher space velocity than in single-stage processes.

By using a high-activity hydrotreating catalyst in the ﬁrst reactor, with a ze-

olite catalyst in the second, overall conversion can be increased to 50–80%. Un-

converted feed is still recycled, but a purge of about 5–10% may be needed to

remove accumulated polynuclear aromatics from the circulating feed. These can

cause deactivation of the catalyst, and foul the reactors. Typical compounds in-

clude coronene (seven fused rings) and ovalene (ten fused rings), both of which

are not very soluble. They are brightly colored and have been called the red

death!

6.5.1.3. Once-Through Process

More recently a very ﬂexible hydrocracking process has been operated, based on

the single-stage process.

Conversion of the feed can range between 40–90%,

but unconverted feed is not recycled. Feeds including heavy-vacuum gas oils,

heavy-cycle oil, coker gas oil, and deasphalted oil are cracked to produce high-

quality middle distillates and naphtha for catalytic reformers. Residue from the

fractionator is used as catalytic cracker and ethylene plant feed, lube oil stocks,

or high-quality fuel oil. Organic nitrogen is almost completely removed from the

product and sulfur content is very low. Moderately active zeolite catalysts are

used.

Refinery Catalysts 235

6.5.2. Hydrocracking Catalysts

It is important that hydrocracking catalysts have both high cracking and hy-

drogenation activities. Typical catalysts are generally made with an acid support

containing either nickel/tungstate or palladium as the hydrogenation catalyst.

Typical examples are shown in Table 6.13. All nickel/molybdate and nick-

el/tungstate catalysts can only work effectively in the form of sulﬁdes in hy-

drotreating and hydrocracking reactions. For this reason the hydrogen sulﬁde

content of recycle hydrogen must be controlled at an appropriate level. Ammo-

nia is normally scrubbed from recycle gas.

6.5.2.1. Acid Supports

The ﬁrst acid support to be used in hydrocracking catalysts was an amorphous

silica/alumina, but following the successful application of zeolites in catalytic

cracking processes in 1964, this has generally been replaced by Y-zeolite. The

most important feature of the support, apart from the high acidity needed to

crack high-molecular-weight hydrocarbons, is high thermal stability to with-

stand the process conditions. Pore size is not as critical when using heavy feed

for hydrocracking as is the case for hydrotreating, provided that the surface area

is sufﬁciently high. Many different support formulations have been patented,

with many of them similar to FCC catalysts. All are produced as extrudates, held

together with binders such as peptized alumina.

TABLE 6.13. Hydrocracking Catalysts.

Design Composition

First stage/single stage

Catalyst 5–10% NiO/10–20% MoO

/1–2% P

Support Alumina containing varying amounts of silica

Porosity (cm

-1

) 0.4–0.6

Shape Extrudates

Rings

Second stage

Catalyst 5–10%NiO/10–20% MoO

5–10% NiO/10–20% WO

Palladium 0.5%

Support Amorphous silica/alumina 0–80%

Ultrastable Y-zeolite 20–80%

Mixed zeolite/silica/alumina (zeolite 0–100%)

Binders such as peptized alumina (0–20%)

Porosity (cm

-1

) 0.4–0.6

236 Chapter 6

Zeolites, often combined with amorphous alumina or silica/alumina and a

suitable binder, are now the most widely used acidic component to hydrocrack

heavy feeds. Hydrogen-exchanged and dealuminated Y-zeolites are normally

preferred. Zeolites are more active than amorphous materials because of their

uniform and acidic structure. Less coke is formed during operation, resulting in

less deactivation of the catalyst so that operational life can be longer than six

years. With nitrogen-free feed the zeolite is active at lower operating tempera-

tures and heavier feeds can be cracked.

Modification of the silica/alumina ratio of the Y-zeolite used in catalytic

cracking catalysts can also improve its hydrocracking performance. It was found

that dealuminated Y-zeolites give high diesel selectivity, whereas untreated Y-

zeolite provides a hydrocracking catalyst with high selectivity for gasoline.

A hydrocracking catalyst support can therefore be tailor-made depending on

how the catalyst is to be operated and the product type needed. To obtain maxi-

mum conversion to gasoline at the lowest operating temperature, up to 80% Y-

zeolite and about 20% of a peptized alumina binder is used as support. High

middle-distillate conversion is obtained at higher operating temperatures with a

support containing about 10% dealuminated Y-zeolite, 70% alumina or sili-

ca/alumina in varying proportions, and 20% of the binder. The appropriate Y-

zeolite and the proportion to be used in the catalyst can be selected for the prod-

ucts, as required.

6.5.2.2. Hydrogenation Catalysts

Sulﬁded nickel/molybdate catalysts supported on a suitable γ-alumina are used

for ﬁrst-stage hydrocracking reactions and they have a higher nickel molybdate

content than typical hydrotreating catalysts. Sulﬁded nickel/tungstate or palladi-

um, supported on zeolite or silica/alumina, is used in the second hydrocracking

stage. Palladium catalysts are active if the feed contains residual sulfur com-

pounds but are not active for the hydrogenation of benzene rings. Hydrogen sul-

fide need not, therefore, be removed from the recirculating hydrogen unless it is

necessary to remove all aromatics.

6.5.2.3. Catalyst Preparation

Nickel/molybdate hydrocracking catalysts are made by impregnating preformed

supports with solutions of nickel and molybdenum salts.

31,32

The addition of

ammonia or phosphoric acid to the solutions before impregnation is claimed to

simplify the procedure and to improve activity of the catalyst.

Palladium catalysts are made by ion exchanging the zeolite with a solution

of palladium chloride containing ammonia followed by careful washing and

drying.

31,32

The tetramine complex can be decomposed in air at temperatures

above 800

C to produce palladium metal in the form of ﬁnely divided crystal-

Refinery Catalysts 237

lites. Direct reduction with hydrogen can lead to agglomeration of the palladium

crystals, which are inactive. The zeolite is extruded with a binder and any other

necessary component of the support.

6.5.2.4. Catalyst Activity

Catalysts are activated before use. Nickel catalysts are carefully presulﬁded ei-

ther before or after loading to the reactor with the same organic sulfur com-

pounds and procedures used for hydrotreating catalysts. Palladium catalysts are

briefly activated in hydrogen at about 350

C, taking care not to overheat the bed,

so that the crystallites do not sinter.

Silica/alumina hydrocracking catalysts are severely poisoned by ammonia.

The feed must therefore be hydrotreated and hydrogen recycle must be scrubbed

to eliminate as much organic nitrogen and ammonia as possible. Zeolites are

slightly less affected by typical ammonia levels and are widely used in the more

efﬁcient two-reactor single-stage process.

Catalysts should, however, contain

more zeolite if the ammonia content is variable. The active metal content must

also be increased to achieve the same conversion in the presence of ammonia.

The cracking activity of the support and the hydrogenation activity of the metal

component must therefore be carefully optimized to obtain maximum conver-

sion with different operating conditions and feeds.

6.5.2.5. Catalyst Reactivation

A typical catalyst can operate for a number of years, and any slow loss of activi-

ty can be compensated for by a gradual increase in process operating tempera-

ture until the reactor temperature limit is reached. Regeneration is then possible,

either in the reactor or externally, by burning accumulated coke in dilute air.

This restores the activity of most nickel/molybdate or nickel/tungstate catalysts,

which must then be resulﬁded before further use.

Deactivated palladium supported on zeolite can also be regenerated, and the

catalyst must also be reactivated before being reused. This is because palladium

in the zeolite framework sinters. By treating the catalyst with an excess of am-

monium hydroxide solution, however, the agglomerated palladium dissolves to

form a tetramine complex. Calcination in air then decomposes the complex and

the original palladium distribution and activity are restored.

The ﬁrst Y-zeolite hydrocracking catalysts contained residual sodium ions

in the sodalite cages that were mobile during operation and they entered the su-

percage. This led to a loss of cracking activity. Treatment of the zeolite with an

ammonium salt solution removed the mobile sodium ions and restored acidity.

The redistribution of palladium with ammonia solution could be combined with

an exchange of sodium ions to rejuvenate the catalyst in one step.

This was

done before reactivation by burning off the carbon deposits.