Chapter 11

Cracking

As mentioned previously, cracking larger molecules produces gasoline, kerosene,

jet fuel, diesel, heating oils and fuel oils. It also generates light gases (including
olefins), LPG, and butanes. Cracking can be carried out thermally through free radical
chemistry or by catalysis (catalytic cracking) through carbocation chemistry. Modern
thermal cracking processes include steam cracking, coking and visbreaking. Catalytic
processes include fluid catalytic cracking and hydrocracking. Cracking requires heat
to break C–C bonds. In hydrocracking, hydrogenation of cracked fragments generates
heat, so on a net basis hydrocracking is exothermic.

11.1 Thermal Cracking

Thermal cracking was first invented and patented by a Russian engineer, Valdimir
Shukhov, in 1891. However, its development was not pursued beyond the laboratory.
American engineers, William Merriam Burton and Robert E. Humphreys indepen-
dently developed a similar process in 1908, but Burton filed for a patent in 1912; the
patent, which is of great significance in the history petroleum refining, was granted
in 1913 (U.S. 1,049,667). The first battery of twelve stills used in thermal cracking
went into operations at the Witting Refinery of Standard Oil of Indiana (now BP) in
1913. Early versions were batch processes. However continuous processes eventu-
ally took over in the 1920s. Once catalytic cracking was invented in the early 1940s,
it became much more popular than thermal cracking. More sophisticated forms of
thermal cracking have been developed for various purposes, including steam crack-
ing, visbreaking and coking. Modern high-pressure thermal crackers, which operate
at absolute pressures of about 7000 kPa, are the basis for economically important
production of olefins for polymers.
Thermal cracking is similar to what occurs below the surface of the Earth when
kerogen is broken down into lighter components. The actual reaction mechanism
behind cracking is very complex, and computers model hundreds or thousands of
reactions for the process, but the main reactions are all related to free radical chem-
© Springer Nature Switzerland AG 2019 211
C. S. Hsu and P. R. Robinson, Petroleum Science and Technology,
https://doi.org/10.1007/978-3-030-16275-7_11
212 11 Cracking

Chain Reaction Initiation Propagation

X2 2 X• (X can be H, Cl, peroxide, etc.) H2 + X• H• + HX

RCH2CH3 + H• RCH2CH2• + H2
CH3CH3 2 CH3•
RCH2CH2• + CH4 RCH2CH3 + CH3•
Radical Addition
CH3
Radical Decomposition
RCH=CH2 + •CH3 RCHCH2•
RCH2CH2• H2C=CH2 + R•
CH3 CH3
CH3
RCHCH2• + R’CH=CH2 RCHCH2R’CH2CH2•
RCHCH2• H2CCH=CH2 + R•

Cyclization Chain Reaction Termination

CH2 CH3• + H• CH4
=

HC(CH2)3CH2• CH3 + H• CH3• + CH3• CH3CH3

CH3• + CH3• H2C=CH2 + H2
R• + R• R-R
CH3 + 2 H2 R• + H• R-H
H• + H• H2

+ + 4 H2

Fig. 11.1 Thermal cracking mechanism: radical chain reaction

istry. At the high temperatures, free radicals are more likely to form. A free radical
is a molecule with an unshared electron attached to it. This unshared electron can
break other bonds leading to a series of chain reactions. Each subsequent reaction will
either create another free radical, which can react further, or combine with another
free radical to end the chain.
Figure 11.1 outlines mechanisms postulated by Greensfelder, et al [1]. for the
thermal cracking of hydrocarbons. This chain-reaction mechanism includes the fol-
lowing steps:
(1) Chain Initiation: Radicals—neutral atoms or compounds with a free electron
and no charge—are formed due to direct thermal rupture of a chemical bond.
Common initiators (other than heat) are organic sulfur and oxygen compounds,
which are present in delayed coker feeds. H2 O2 and Cl2 are potent initiators.
In steam cracker feed, heteroatom initiators are absent, so the chain reaction is
initiated by the rupture of a C–C bond at very high temperature.
(2) Propagation: A radical abstracts hydrogen from another compound, turning it
into a different radical.
(3) Radical addition: A radical attacks an olefin group, forming a C–C bond and a
larger radical.
(4) Isomerization: A primary radical isomerizes into a more stable secondary radi-
cal.
11.1 Thermal Cracking 213

(5) Radical decomposition (cracking): A radical decomposes into an olefin and a

smaller radical.
(6) Cyclization: radicals with 5 or more carbons and an olefin group react within the
molecules to form ring compounds. At the high temperatures at which thermal
cracking occurs, aromatics and polyaromatics are thermodynamically favored,
so saturated or partially saturated ring compounds readily undergo dehydro-
genation and form aromatics.
(7) Chain reaction termination: Two radicals react to produce a saturated product
or an olefin and hydrogen.
The radical mechanism is consistent with the fact that thermal cracking produces
significant amounts of hydrogen, methane, ethane, ethylene and higher olefins. Feeds
to delayed coking units contain organic sulfur, nitrogen, and oxygen compounds,
which form free radicals more easily than hydrocarbons. These heteroatoms end up
in two types of molecules—light ones (H2 S, ammonia and water) and heavy ones,
from which they are hard to remove.
Figure 11.2 represents a thermal cracking unit for gas oils and residues. The feed
is heated to 700–1100 °F in a furnace before entering a reactor under 50–1500 psig
pressure. For residues, 2-stage thermal crackers are often used; the 1st stage produces
gas oils for 2nd stage cracking. The reactor product is introduced into a flasher to sep-
arate residue from the product mixture, which is fractionated into olefin-containing
gas, naphtha, distillates, and gas oil products. The fractionator bottom product is
mainly residue, which is mixed with fresh feed and charged to the reactor. Lin-
ear alpha olefins can be produced from wax by thermal cracking. The cracking is
accompanied with simultaneous removal of distillates as a semi-continuous process.

Fig. 11.2 Residue thermal cracking unit

214 11 Cracking

Fig. 11.3 Links between a refinery and a steam cracking unit

11.2 Steam Cracking

Steam cracking, shown in Fig. 11.3, is a process of breaking down saturated hydro-
carbons into smaller, often unsaturated, hydrocarbons by reactions with steam in a
bank of pyrolysis furnaces. It is the principal industrial method employed at olefin
plants for producing ethylene and propylene, important feedstocks for polyolefins
(polyethylenes, polypropylenes, etc.), which account for 50–60% of all commer-
cial organic chemicals. Propylene is also produced by other methods, such as propane
dehydrogenation [2]. Fluid catalytic cracking (FCC) is an important commercial
source of propylene and butylene. Steam cracking in olefin plants co-produces hydro-
gen, which is often used in refineries.
The feed can be gaseous or liquid hydrocarbons, such as ethane, LPG (propane
and butane), naphtha, and unconverted oil from hydrocracking units. C4 olefins,
including butadiene, and benzene-rich pyrolysis gasoline are produced when heavy
liquid feeds, such as naphtha and hydrocracker gas oils, are used. The feeds are
diluted with steam and briefly heated to 1050 °C in a furnace without the presence of
oxygen and fed to Cr–Ni reactor tubes. The hydrocarbon chains preferentially crack
at the center of molecules at 400 °C. Cracking shifts toward the end of the molecule
with increasing temperature, leading to larger quantities of preferred low molecular
weight olefin products. Thus, the reaction temperature is very high, with a reactor
outlet temperature around 850 °C. The residence time is very short to improve yield
and to avoid coke formation and oligomerization. In modern furnaces, reaction time
can be reduced to milliseconds, with velocities faster than the speed of sound. The
reaction is stopped by rapid quenching to 300 °C. The product mixture is scrubbed
to remove H2 S and CO2 and then dried before it is sent to a series of separation
columns to separate and recover methane, ethane, ethane, propane, propylene and
C4 hydrocarbons. Figure 11.4 shows a steam cracking unit outside London with tall
11.2 Steam Cracking 215

1. A debutaniser which separates the C4 hydrocarbons from the C1 - C3 hydrocarbons

2. A depropaniser which separates out the C3 hydrocarbons
3. A deethaniser which separates out the C2 hydrocarbons
4. A demethaniser which separates out the methane
5. A C3 splitter which separates propene from propane
6. A C2 splitter which separates ethene from ethane

Fig. 11.4 A steam cracking unit outside London [3]

separation columns in series for the production and separation of light hydrocarbon
gases.
Steam cracking product distribution depends on the feed composition, hydrocar-
bon to steam ratio, cracking temperature and residence time in the furnace. Light
hydrocarbon gas feeds, such as natural gas, ethane or LPG yield product streams
rich in the light olefins, essentially ethylene and propylene. In addition to light
olefins, heavier liquid feedstocks can also produce butylenes, butadiene and prod-
ucts rich in aromatic hydrocarbons suitable for blending into gasoline and fuel oil,
or routed through an extraction process to recover BTX aromatics (benzene, toluene
and xylenes).

11.3 Catalytic Cracking [4]

Thermal cracking was used prior to 1925 for producing naphtha through thermal
decomposition of larger molecules into smaller molecules. In the late 1920s, Eugene
Houdry demonstrated that a catalytic cracking process yielded more gasoline of
higher octane-number, with less heavy fuel oils and light gases. Thermally cracked
216 11 Cracking

naphtha is quite olefinic, while catalytically cracked naphtha contains fewer olefins
and large amounts of aromatics and branched compounds. The first commercial
fixed bed cat cracking unit began production in 1937. The catalysts are covered by
a deactivating layer of coke in a short time during the process. The catalysts can
be regenerated by burning off the coke, but the time is relatively slow compared to
reaction time. It is more efficient to move the catalyst from one reactor for cracking
hydrocarbons to another reactor for catalyst regeneration.
In refining, catalytic cracking falls into two categories: catalytic cracking in the
absence of external hydrogen—primarily fluid catalytic cracking (FCC)—and cat-
alytic cracking in the presence of external hydrogen—hydrocracking.
Catalysts for both kinds of catalytic cracking contain strong acid sites. The mecha-
nism involves carbocations, also known as carbenium ions or carbonium ions. Acidity
is provided by amorphous silica/alumina or a crystalline zeolite. Commercial cata-
lysts for these and other processes are discussed in a subsequent section.
FCC units produce aromatics and other heavy products via cyclization, alkyla-
tion and polymerization of intermediate olefins. Polyaromatics can grow into larger
polyaromatics, which eventually can form coke.
Key features of catalytic cracking chemistry include a preponderance of branched-
chain paraffin products and low yields of methane and ethane. If there is a deficiency
of hydrogen, such as in the FCC process, hydrogen is produced and significant
amounts of small olefins are formed.

11.3.1 FCC Feed Pretreating

These days, most refiners pretreat FCC feeds in a fixed-bed hydrotreater. The
hydrotreater removes trace metal contaminants such as nickel and vanadium. Oth-
erwise, nickel would increase coke formation and decrease liquid yields. Vanadium
reduces conversion, decreases liquid yields, and destroys the catalyst. In addition to
removing Ni and V, the pretreater decreases concentrations of sulfur, nitrogen, and
aromatics. In the FCC regenerator, sulfur on the coked catalyst is converted to sulfur
oxides (SOx) in the flue gas. Clean air regulations restrict SOx emissions, which
cause acid rain. Therefore, removing sulfur from the FCC feed—thereby reducing
SOx formation—is highly beneficial. Removing nitrogen is beneficial, too, because
basic feed nitrogen suppresses the activity of highly acidic FCC catalysts. Pretreating
also saturates aromatics. As we have seen, saturating aromatics makes them more
crackable, so pretreating increases FCC conversion, often by more than 10 vol.%.

11.3.2 Catalytic Cracking Mechanism [4, 5]

The underlying mechanism for catalytic cracking is essentially the same for both fluid
catalytic cracking and hydrocracking. Process differences are due to differences in
11.3 Catalytic Cracking 217

Table 11.1 Zeolitic catalytic Silica-Alumina Zeolite

cracking versus cracking
catalyzed by an amorphous Coke, wt% 4 4
silica-alumina [6] Conversion, vol.% 55 65
C5 + gasoline, vol.% 38 51
C3 gas, vol.% 7 6
C4 ’s, vol.% 17 16

catalysts, equipment, and operating conditions. FCC produces a high yield of naph-
tha (gasoline) and LPG. It is considered as the heart and workhorse of a fuels refin-
ery. FCC produces more than half the world’s gasoline. It generates middle distil-
late streams (cycle oils) for further refining or blending. It also produces, via heat
exchange, a large quantity of high-quality steam.
FCC chemistry is a complex mixture of many reactions. The strong acid sites, both
Lewis and Brønsted, are supplied by zeolites, which are key components of modern
FCC catalysts. Influenced by their unique pore geometry, zeolites (usually H–Y) are
far more efficient at generating gasoline and middle distillate products than either
thermal cracking or cracking catalyzed by amorphous silica-alumina. A comparison
of zeolite and amorphous Al–Si cracking is shown in Table 11.1.

11.3.2.1 Reaction Initiation

There are several possible initiation steps, shown in Fig. 11.5 [7]. One type of initia-
tion step (Reaction 11.1) is mild thermal cracking to generate free radicals, followed
by radical recombination to generate olefins. FCC reactions are mainly catalytic as
evidenced by the fact that they produce relatively small amounts of methane, the
generation of which is significant in thermal cracking. Methane can be formed by
hydrogen abstraction by a methyl radical or by combining a methyl radical with
hydrogen radical; the latter is a termination step of free-radical reactions. The main
initial steps involve Brønsted acid protonation of olefins to generate secondary (2°)
carbocations, shown in Reaction 2, and hydride abstraction from alkanes at Lewis
base catalyst sites, shown in Reaction 3, also to generate (2°) carbocations. Of minor
importance is the reversible dehydrogenation of alkanes at metal-containing catalyst
sites.

11.3.2.2 Carbocation Ion Rearrangement

Figure 11.6 shows how 2° carbocations rearrange. There can be 2°–2° rearrangement
and 2°–3° (tertiary) rearrangement; 3° carbocations are more stable.
218 11 Cracking

Thermal
cracking
R CH2 CH2 CH2 CH3 R CH2 CH2 CH2 + CH3
1

R CH2 CH2 CH2 R CH2=CH2 + CH3

Radical
decomposition

R CH=CH CH2 R’ R CH2 CH2 CH2 R’ + L R CH2 CH2 CH2 R‘

Brønsted Lewis
H+ addition H- removal 3
2 acid site acid site H2 / +H2 4 Metal Site

R CH=CH CH2 R’
+ +
R CH2 CH CH2 R’ R CH2 CH CH2 R’ + LH

2° carbocation 2° carbocation

Fig. 11.5 Catalytic cracking mechanism: initiation steps

H
|
| + +
R C C H R CH CH3
| |
| | 2° carbocation
H H

1° carbocation

H
|
| + +
R C C CH2 R C CH2 CH2
| | |
| | |
CH3 H CH3
2° carbocation 3° carbocation

+ +
CH3 C CH2 CH2 R CH3 CH2 C CH2 R
| |
H H
2° to 2° skeletal isomerization

Fig. 11.6 Catalytic cracking mechanism: carbocation rearrangement

11.3 Catalytic Cracking 219

+ +
(1) CH3CHCH2CH2CH3 X CH3CH=CH2 + CH2 CH3 1° carbocation produced:
less favored
CH3 CH3
| + |
(2) CH3 C CH2 CH CH3 CH3 CH=CH2 + CH3 C+ 3° carbocation produced:
| | favored
CH3 scission CH3

+
+
CH3 CH2 CH2 CH CH3 + RH CH3 CH2 CH2 CH2 CH3 + R

hydride transfer

Fig. 11.7 Catalytic cracking mechanism: beta scission and hydride transfer

11.3.2.3 Cracking of C–C bonds via beta scission

Beta scission involves cleavage of a C–C bond in the position beta to the carbon
atom that carries the positive charge. It is the primary cracking reaction in FCC. It is
endothermic and favored at higher temperatures; however, if the temperature is too
high, thermal cracking can become significant. Beta scission, shown in Fig. 11.7, is
characteristic of mechanisms with carbocation intermediates.

11.3.2.4 Hydride Transfer (Propagation)

Hydride transfer also is shown in Fig. 11.6. If hydride transfer is substantial, paraf-
fins (normal and branched) will be the predominant products: FCC catalysts can be
designed for greater or lesser hydrogen transfer.

11.3.2.5 Termination Reactions

Termination reactions involve proton transfer to the catalyst surface or desorbing

an olefin while regenerating a Brønsted site. Proton transfer to a carbocation ion
generates a paraffin and regenerates a Lewis acid site. Radical termination reactions
involve radical-radical reactions, such as:

H • +H• → H2
H • +R• → RH
R • +R • → R − R
220 11 Cracking

11.3.3 Fluid Catalytic Cracking Processes

In FCC, a full range of smaller molecules is formed from the breakup of large
molecules. Due to the overall deficiency of hydrogen, significant amounts of olefins
are formed. Since the feed contains large aromatic and naphthenic molecules with
long side chains attached, side chain cleavage, which can also initiate the chain of
reactions, is common. The molecules from which side chains have been removed
have higher specific gravity, i.e., lower API gravity.
Cracking also generates coke via aromatic ring growth by successive cyclization,
polymerization and dehydrogenation. The coke coats the catalyst, rendering it inac-
tive. Hence, the catalyst needs to be regenerated by burning off the coke in order to
restore activity.
To allow onstream catalyst regeneration, a moving-bed unit, Thermofor catalytic
cracking (TCC), was developed in 1941 with catalyst cycled between the reactor and
regenerator. It has become obsolete and replaced by more advanced fluidized bed
units with greatly improved catalysts.
A simplified diagram of the TCC process is shown in Fig. 11.8 [8]. Its essential
elements are a reactor for continuously contacting hydrocarbon feed with a moving
bed of granular catalyst for conversion of the hydrocarbons, and a kiln (regenera-
tor) for removing carbon deposit from the catalyst during the cracking operation by
controlled combustion with air. The catalyst flows by gravity through both vessels,
which stand side by side, and is transferred from the bottom of one to the top of
the other by means of bucket elevators. All required process heat is supplied by
the highly exothermic combustion of coke in the regenerator. Catalyst temperatures
reach 1200–1500 °F. In contrast, the cracking reaction is endothermic. The feed is
preheated by heat exchange to 500–800 °F before entering the reactor. There, addi-
tional heat is supplied by mixing with the hot recycled catalyst to reach a temperature
of 900–1050 °F. The reactor outlet vapor is sent to a fractionator, where it is separated
into different fractions—gas (offgas), cracked naphtha, fuel oil, and slurry oil, which
can be mixed with fresh feed and recycled back to the reactor for further processing.
A typical FCC unit comprises three major sections—riser/reactor, regenerator,
and disengaging vessel. In the riser/reaction section, preheated oil is mixed with hot,
regenerated catalyst. The mixture acts as a fluid because the catalyst particles are
about the size of sifted flour. The hot catalyst vaporizes the oil, and the vaporized oil
carries the catalyst up the riser/reactor. The cracking reaction is very fast, achieving
completion in just a few seconds or even less. It produces light gases, high-octane
gasoline, and heavier products called light cycle oil (LCO), heavy cycle oil (HCO),
slurry oil, and decant oil. It also leaves a layer of coke on the catalyst particles,
rendering them inactive.
There are two basic types of FCC units: “side-by-side,” where the reactor and the
regenerator adjacent to each other, shown in Fig. 11.9, and the stacked type, in which
the reactor is mounted on top of the regenerator, shown in Fig. 11.10.
At the top of the riser, the riser outlet temperature (ROT) can reach 900–1020 °F
(482–549 °C). The ROT determines conversion and affects product selectivity, so
11.3 Catalytic Cracking 221

Bucket elevators
Regenerated
Catalyst
hopper

Product out Flue gas out

Reactor Regenerator

Feed in
Air in

Fig. 11.8 Simplified diagram of Thermofor catalytic cracking system

Fig. 11.9 FCC process flow

222 11 Cracking

Fig. 11.10 FCC Unit—M. W. Kellogg Design

FCC operators control it as tightly as possible. Higher temperatures favor production

of olefin-rich light gases, especially propylene, at the expense of gasoline; many FCC
units maximize refinery-grade propylene for purification and use in olefin plants.
Moderate temperatures favor gasoline production. Lower temperatures decrease
gasoline yields and increase heavier products—light cycle oil (LCO) and heavy
cycle oil (HCO).
In the disengaging section, steam helps separate the now-deactivated catalyst from
the reaction products. The spent catalyst goes to the regenerator by gravity, where
the coke is burned away by fluidized combustion in the presence of air or oxygen-
enriched air. The regenerated catalyst is hot, with temperatures up to 1350 °F (732 °C).
It returns to the riser/reactor, where the cycle begins again.
In a 60,000 barrels-per-day unit processing a typical mixture of vacuum gas oils,
the total catalyst in the unit (the “inventory”) is 400–500 tons. To maintain activity,
about 0.5–1 wt% of the inventory is replaced each day. If the feed to the unit contains
significant amounts of residue, the replacement rate is higher. The discharged catalyst
is cooled and shipped either to a land fill for disposal or to another refiner, which
might need “conditioned” FCC catalyst.
The stacked type FCC unit is represented by a Kellogg design, shown in Fig. 11.8.
Again, it constitutes a riser/reactor, a regenerator and a disengager. The spent catalysts
from the reactor drop from the disengager to the regenerator by gravity, with less
hardware, space and maintenance required than the side-by-side type. In the figure, a
11.3 Catalytic Cracking 223

fractionator to separate product components is also shown. The top effluent contains
light gases that are sent to a recovery system to remove sour gases for recovery of
flue gas, LPG, light olefins, etc. The bottom stream can be recycled by mixing with
feed for further processing.

11.3.4 FCC Heat Balance

FCC units must be heat-balanced, or they won’t run. The burning of coke deposited
on the catalyst in the regenerator provides all of the heat required by the process. In
fact, FCC units are significant sources of high-quality steam for other refinery units.
Table 11.2 gives a representative breakdown of FCC heat requirements.

11.3.5 Residue FCC

Residue FCC (RFCC) units, also known as heavy oil crackers, can process significant
amounts of atmospheric residue (650 °F+) and vacuum residue to produce gasoline
and lighter components [9, 10]. It generates substantially more coke than conven-
tional FCC feeds. Excess heat is generated when the extra coke is burned in the
regenerator, and residues contain high amounts of trace metals, particularly nickel
and vanadium. Those metals destroy FCC catalysts. Residue FCC units must handle
both challenges.
The metals are removed in upstream hydrodemetalation (HDM) units [11]. Cat-
alyst coolers and supplemental regenerators recover the excess heat. The catalyst
coolers (e.g., steam coils) are installed usually on the second-stage regenerator. The
UOP catalyst cooler is an external vertical shell-and-tube heat exchanger [12]. Cat-
alyst flows across the tube bundle in the dense phase. UOP’s air lance distribution

Table 11.2 Representative FCC heat consumption

Factor Portion of total
Heat up and vaporize fresh feed 40–50%
Heat recycled oil 0–10%
Heat of reaction (endothermic) 15–30%
Heat steam 2–8%
Heat losses 2–5%
Heat air to regenerator temperature 15–25%
Heat coke from the reactor to regenerator temperature 1–2%
Total heat duty 500–1000 Btu/lb
1160–2325 kJ/kg
224 11 Cracking

system ensures uniform air distribution within the tube bundle and a uniform heat
transfer coefficient. The generation of steam (up to 850 psig) from the circulating
water is used to remove heat from the regenerated catalyst.
Three different styles of catalyst coolers—flow-through, back-mix and
hybrid—have been designed and commercialized to accommodate a wide range
of heat removal duties as well as physical and plot-space constraints [13].

11.3.6 FCC Gasoline Post-treating

Conventional hydrotreating does a good job of removing sulfur from FCC feed,
which leads to lower sulfur in FCC products. Unfortunately, despite pretreatment,
FCC gasoline can still contain up to 150 ppmw sulfur—far more than the present
specification of 10 ppmw. Hydrotreating removes sulfur from the gasoline, but it also
reduces octane by saturating C6 –C10 olefins.
In processes such as Prime-G+ [14], offered for license by Axens, full-range
naphtha is split into light and heavy fractions. The light fraction contains most of
the high-octane olefins but not much of the sulfur. After diolefins are removed via
selective hydrogenation, the light cut is ready for gasoline blending. The heavy
fraction contains most of the sulfur but not much of the olefins. It is hydrotreated
conventionally.
The S-Zorb process [15], invented by ConocoPhillips, uses selective adsorption
to remove sulfur from FCC gasoline. The feed is combined with a small amount
of hydrogen, heated, and injected into an expanded fluid-bed reactor, where a pro-
prietary sorbent removes sulfur from the feed. A disengaging zone in the reactor
removes suspended sorbent from the vapor, which exits the reactor as a low-sulfur
stock suitable for gasoline blending. The sorbent is withdrawn continuously from
the reactor and sent to the regenerator section, where the sulfur is removed as SO2
and sent to a sulfur recovery unit. The clean sorbent is reconditioned and returned to
the reactor. The rate of sorbent circulation is controlled to help maintain the desired
sulfur concentration in the product.

11.4 Petroleum Refining Catalysts

Many refining processes require catalysts. Catalysts do not change thermodynamics,

which govern whether or not a reaction can proceed, but an appropriate catalyst
increases the rate of the desired reaction. A catalyst facilitates chemical reactions,
either by increasing the rate at a given set of process conditions, or by lowering
required temperatures and/or pressures and/or residence times, thereby reducing the
size of process equipment. Catalysts do this by decreasing activation energy. The
reaction heat might vary with temperature and pressure, but ultimately, reaction heat
is governed by thermochemistry.
11.4 Petroleum Refining Catalysts 225

In practice, catalysts do change as time goes by and more materials are processed.
They degrade due to coking, attrition, feed contamination and/or process upsets.
Some catalysts last for years before they have to be replaced. In other processes,
such as FCC and CCR catalytic reforming, they are regenerated and reused inside the
process during normal operation. In the FCC process, degraded catalyst is removed
and replaced with fresh catalyst as needed. The chemistry of coke formation on
catalysts is similar to the coke formation mechanism presented in Chap. 9, Sect. 4.
Catalysts facilitate reactions by decreasing activation energies. Consider ammonia
synthesis:

N2 + 3H2 → 3NH3

The reaction is favored by thermodynamics, but molecular nitrogen is exceedingly

stable due to the strength of the N≡N triple bond. To break the N≡N bond with heat
alone requires 3000 °C (5400 °F). Before the catalytic Haber-Bosch process, ammo-
nia was synthesized from nitrogen and hydrogen in electric arc reactors, which in
essence generate mini-lightning bolts while consuming immense amounts of energy.
In catalytic ammonia synthesis, the N≡N bond is weakened when it “dissolves” in
the catalyst, allowing it to react with hydrogen at <750 °F (<400 °C).
Most petroleum refinery catalysts are solids, with active metals supported on high-
strength supports such as γ-alumina, silica or aluminosilicates. The major exceptions
are alkylation catalysts, which employ liquid-phase acids—HF or H2 SO4 or ionic
liquids. The separation of liquid and vapor products from solids is straightforward,
so it is easier to recover the catalysts and regenerate or dispose of them.
As mentioned in the FCC chemistry discussion, FCC and hydrocracking (dis-
cussed below) employ solid-acid catalysts, primarily based on synthetic zeolites.
Hydrocracking employs both zeolites and amorphous silica/alumina catalysts. Zeo-
lites possess microporosity, which gives them high surface area and contributes to
their high activity. They also possess a mesopore structure, which affects feed flexi-
bility and product selectivity.
Related materials are used in other processes. For example, metal-promoted silica
alumina phosphates such as SAPO-11 are used to isomerize normal paraffins in
catalytic dewaxing units. Figure 11.11 shows six zeolite structures [16]. In cracking
catalysts, HY zeolite is the most common, but beta and ZSM-5 are used as well. The
building blocks of A, X, and other zeolites are tetrahedral units of Si and Al oxides.
In the ultra-stable Y (USY) zeolites employed in cracking catalysts, the Si/Al ratio
is >10.
The acidity of zeolites (and amorphous silica/aluminas) comes from their structure
[17]. Fig. 11.12 shows how one can visualize these materials as a silica (SiO2 )
superstructure, in which every so often an aluminum atom replaces a silicon atom.
The silicon atoms have a valence of +4, and each one binds to four oxygen atoms.
Replacing Si (+4) with Al (+3) creates electron-accepting Lewis acid sites. Car-
bocations can be formed from olefins by hydride transfer to the empty d-orbital on
aluminum. The oxygen atoms in associated hydroxyl groups are Brønsted acids (pro-
226 11 Cracking

Fig. 11.11 Structures of zeolites: ZSM-5 (a), mordenite (b), beta (c), MCM-22 (d), zeolite Y (e),
and zeolite L (f)

Fig. 11.12 Brønsted (B) and Lewis acid sites (L) in zeolites and amorphous silica/aluminas

ton donors). The H+ can be replaced by other positive counter-ions such as Na+ , K+ ,
and NH4 + . The counter ions can be swapped via ion exchange. For example, when
Na-Y zeolite is exchanged with an ammonium salt, the Na+ ion is replaced by NH4 + .
When NH4 -Y is heated to the right temperature, the ammonium ion decomposes,
releasing NH3 (gas) and leaving behind highly acid H-Y zeolite. A typical Si/Al ratio
for a zeolite with high cracking activity is >5. In hydrocracking catalysts with less
activity but greater selectivity for production of middle distillates, the Si/Al might
be about 30.
ZSM-5 is a shape-selective zeolite made by including a soluble organic template
in the mix of raw materials. Templates for this kind of synthesis include quarternary
ammonium salts. ZSM-5 enhances distillate yields in FCC units and catalytic dewax-
ing in hydroprocessing units, where due to its unique pore structure, it selectively
cracks waxy n-paraffins into lighter molecules.
The amorphous silica/alumina (ASA) catalysts used for hydrocracking are less
active, but they do a better job of converting straight-run VGO into diesel with
11.4 Petroleum Refining Catalysts 227

Mesopores
ASA Zeolite

Fig. 11.13 Schematic comparison of amorphous alumina silica (ASA) and zeolite structures

minimal co-production of naphtha and lighter products. Figure 11.13 provides a

conceptual comparison of amorphous ASA catalysts with crystalline zeolite catalysts.
The pore diameters of H-Y zeolite catalysts are uniform and relatively small, around
7.5 Å. Mesopores have larger diameters, which admit mid-sized molecules such as
those found in heavy vacuum gas oils. ASA catalysts include small, medium and large
pores. The width of ASA pores can exceed 100 Å. The larger pores can accommodate
larger molecules, which explains why ASA catalysts do a better job of cracking feeds
with endpoints >1000 °F (538 °C). The pore diameter for ZSM-5 is 6.3 Å.

11.4.1 FCC Catalysts and Additives

Finished FCC catalysts are powders with the consistency of sifted flour. They are
given their final form by spray drying in which a slurry of catalyst components is
converted into a dry powder by spraying into hot air. The method gives a consistent
particle size distribution. The particles are roughly spherical, with bulk densities of
0.85–0.95 g/cm3 and average diameters of 60–100 μm.
The catalysts include four major components—one or more zeolites (up to 50%), a
matrix based on non-crystalline alumina, a binder such as silica sol, and a clay-based
filler such as kaolin. The zeolite is ultra-stable (US) H-Y (faujasite) or very ultra-
stable (VUS) H-Y. The extra stability is provided by de-alumination with hydrother-
mal or chemical treatment. The H-Y zeolite is sometimes augmented with ZSM-5
to increase propylene yield. To provide thermal stability and optimize the relative
amounts of different active sites, a mixture of rare earths (RE), such as lanthanum-
rich mixture of rare earth (RE) oxides, is incorporated into the zeolite structure by
ion exchange. The RE mixture can contain up to 8 wt% ceria, up to 80% La2 O3 , up
to 15 wt% Nd2 O3 , with traces of other rare earths.
228 11 Cracking

Other components can be part of the core catalyst or introduced as external addi-
tives. The extra components can provide NOx and/or SOx reduction, Noble metal
combustion promoters enhance the conversion of CO to CO2 in the regenerator.

11.4.2 Catalysts for Fixed-Bed Reactors

Fixed-bed catalysts are shaped into spheres, cylinders, hollow cylinders, lobed extru-
dates, pellets. Some look like small wagon wheels. Figure 11.14 shows some of these.
A cross-section of a lobed extrudate can look like a 3-leaf or 4-leaf clover without the
stem. Compared to cylindrical extrudates, shaped extrudates have a higher surface-
to-volume ratio, and the average distance from the outside of a particle to the center
is shorter. This increases activity by decreasing the average distances traveled by
molecules to reach active catalyst sites. To make extrudates, a paste of support mate-
rial forced through a die. The resulting spaghetti-like strands are dried and broken
into short pieces with a length/diameter ratio of 2–4; for main-bed hydrotreating cat-
alysts, diameters range from 1.3 to 4.8 mm. The particles are calcined, which hardens
them and removes additional water and volatile molecules such as ammonia.
Spherical catalysts are made by (a) spray-drying slurries of catalyst precursors, (b)
spraying liquid onto powders in a tilted rotating pan, or (c) dripping a silica-alumina
slurry into hot oil. Pellets are made by compressing powders in a dye. FCC catalysts
are made by spray drying.
Impregnation distributes active metals within the pores of a catalyst support. Like
sponges, calcined supports are especially porous. Far more than 99% of the surface
area is inside the pores. When the pores are exposed to aqueous solutions containing
active metals, capillary action pulls the aqueous phase into them. After drying, the
catalyst might be soaked in another solution to increase the loading of the same
(or a different) active metal. Catalysts can also be made by co-mulling active metal
oxides with the support. Co-mulling tends to cost less because it requires fewer steps.
It also produces materials with different activities—sometimes higher, sometimes
lower—than impregnation.
Eventually, refinery catalysts deactivate and must be replaced. The major causes
of deactivation are feed contaminants (trace metals, particulates, etc.) and catalyst
coking; the latter is discussed above in some detail. In fluid catalytic cracking (FCC),
continuous catalyst replacement (CCR) processes, and ebullated-bed hydrocracking,
aged catalysts are continuously removed and replaced with fresh. But in fixed-bed
units, catalyst replacement requires a shutdown. For a 40,000 barrels-per-day hydro-
cracker with an upgrade value of $15-20 per barrel, every day of down time for a
cost $600,000 to $800,000. Lost production during a 4-week catalyst changeout can
amount to $18 to $24 million.
11.5 Comparison of Catalytic and Thermal Cracking 229

Fig. 11.14 Catalyst loading scheme showing size/shape grading. Photo used with kind permission
from Haldor Topsoe Inc

11.5 Comparison of Catalytic and Thermal Cracking

Table 11.3 compares thermal cracking and catalytic cracking. The fundamental dif-
ferences are: thermal cracking does not use a catalyst. It operates at higher tempera-
ture; the pressure can be higher, as in steam cracking, or lower as in delayed coking.
The mechanism involves free radical chemistry. Catalytic cracking uses a catalyst
at lower temperature and pressure, and the mechanism involves ionic (carbocation)
reaction chemistry.
Table 11.4 compares the yields on similar topped crude feed by thermal and cat-
alytic cracking. Catalytic cracking produces more gasoline and olefins than paraf-
fins of the same carbon number. Thermal cracking produces more residual oil than
catalytic cracking. Catalytic cracking gives volume swell—the volume of liquid
products is greater than the volume of feed—but thermal cracking decreases liquid
230 11 Cracking

Table 11.3 Comparison between thermal cracking and catalytic cracking

Thermal cracking of VGO Catalytic cracking of VGO with FCC
No catalyst Use a catalyst
Higher temperature Lower temperature
Higher Pressure Lower pressure
No regeneration of catalysts needed More flexible in term of product slates
Free radical reaction mechanism Ionic reaction mechanism
Moderate thermal efficiency Higher thermal efficiency
Gas yield is relatively high and feedstock Good integration of cracking and regeneration
dependent
Moderate yields of gasoline and other Higher yields of gasoline and other distillates
distillates
Alkanes produced but yields are feedstock Low gas yield
dependent
Low to moderate C5 + product selectivity High C5 + product selectivity
Low octane number gasoline High octane number gasoline
Some chain branching in alkanes Low n-alkane yield
Low to moderate yield of C4 olefins Chain branching and high yield of C3 and C4
olefins
Low to moderate yield of aromatics High yield of aromatics

volume. Also, catalytic cracking produces coke that can deactivate the catalyst, but
thermal cracking of VGO does not.

11.6 Hydrocracking [18]

From the crude distillation complex, straight-run gas oils and especially VGO can
be sent to hydrocracking units. Hydrocrackers produce LPG, light gasoline, heavy
naphtha, middle distillate fuels (jet and diesel), FCC feed, lube basestock, olefin
plant feed, and isobutane, which is an important feed component for alky plants.
Hydrocrackers also process cracked stocks, such as coker gas oils and FCC cycle
oils. Typically, if a refinery has both a hydrocracker and an FCC unit, straight-run
VGO goes to the FCC while cracked stocks go to the hydrocracker. In North America,
most hydrocrackers process cracked stocks.
Refineries which own recycle hydrocrackers and switch from maximum produc-
tion of middle distillate fuels in the winter to maximum production of naphtha in the
summer to meet heavier gasoline demands. Kerosene is a swing product, which can
be minimized or maximized by adjusting cut points. Hydrocracker middle distillate
are excellent blendstocks for diesel and jet. Hydrocracker naphtha has low octane,
but due to high naphthene content, it is a superb feed for catalytic reformers.
11.6 Hydrocracking 231

Table 11.4 Thermal versus Catalytic Cracking Yields on Similar Topped Crude Feed
Thermal cracking Catalytic cracking
wt% vol.% wt% vol.%
Fresh feed 100.0 100.0 100.0 100.0
Gas 6.6 4.3
Propane 2.1 3.7 1.3 2.2
Propylene 1.0 1.8 6.8 10.4
Isobutane 0.8 1.3 2.6 4.0
n-Butane 1.9 2.9 0.9 1.4
Butylene 1.8 2.6 6.5 10.4
C5 + gasoline 26.9 32.1 48.9 59.0
Light cycle oil 1.9 1.9 15.7 15.0
Decant oil 8.0 7.0
Residual oil 57.0 50.2
Coke 0 5.0
Total 100.0 96.5 100.0 109.9

Friedrich Bergius developed hydrocracking technology for converting lignite coal

into synthetic fuels with hydrogen at high pressure and temperature [19]. His lab-
scale unit operated at about 450 °C (840 °F) and up to 130 bars. Bergius received
a patent for this work in 1913. In 1931, he and Karl Bosch shared the Nobel Prize
in Chemistry for their respective contributions in high-pressure chemistry. Together
with Matthias Pier, Bergius designed a demonstration plant, which was constructed
by Goldschmidt AG near Essen. Due to World War I, production didn’t begin until
1919. Subsequently, Bergius sold his patent to BASF.
Commercial plants based on improved versions of the Bergius-Pier process were
operated between 1927 and 1944 to supply Germany with fuels and lubricants from
coal. Meanwhile, between 1925 and 1930, Standard Oil of New Jersey (Esso) collab-
orated with I.G. Farben to extend Bergius-Pier technology into a process for heavy
oil conversion.
In 1977, with support from the German government, Ruhr Kohle and Veba Oel,
existing coal liquefaction technology evolved into the Veba COMBIcracking pro-
cess, which was demonstrated commercially in a 200 tons/day unit in Bottrop, Ger-
many. The process is now known as VCCTM and is licensed jointly by BP Refining
Technology and KBR Technology. The integrated process includes a slurry-phase 1st
stage hydrocracking section directly linked to a fixed-bed 2nd stage hydroprocessing
section. The first commercial integrated unit, which was designed by KBR, started
operation in Yulin, China in January 2015.
Modern fixed-bed hydrocracking was commercialized by Chevron in 1958. The
UOP UnicrackingTM process, based on zeolite-containing hydrocracking catalysts,
was co-invented by Union Oil, Esso, and Union Carbide. The first Unicracker started
232 11 Cracking

operation in 1964 at the Unocal refinery in Wilmington, California. Acquired from

Unocal by UOP in 1995, Unicracking is now world’s leading fixed-bed hydrocracking
process. Hydrocracking is also licensed by Shell Global Solutions, Haldor Topse,
and Axens.
Ebullated bed (e-bed) hydrocracking was commercialized in 1968. The leading
e-bed processes are H-Oil RC, now offered by Axens Technology Licensing, and
LC-Fining® , now licensed by CB&I.
In 1993, Isodewaxing technology, invented by Chevron, revolutionized lube base-
stock production. Together with Isofinishing, it replaced aromatics extraction and
solvent dewaxing with hydroprocessing, thereby increasing overall base stock yields.
Catalytic hydrocracking uses catalytic cracking in the presence of hydrogen to
convert high boiling and heavy hydrocarbons into lower-boiling and lower molec-
ular weight hydrocarbons. Breaking C–C bonds in paraffins and naphthenes (with
ring opening) is endothermic; aromatic C–C bonds are not broken. Hydrogenation
of reaction intermediates (olefins) is exothermic. Hence, the heat released by hydro-
genation more than compensates for all the heat required for cracking. In addition,
hydrogenation minimizes coke formation. As with FCC, isoparaffins are produced
preferentially.
Table 11.5 lists length and energy of selected bonds, and Table 11.6 compares
heat of reaction calculated from bond energies with that from heat of formation in
hydrocracking of n-hexane (Reaction 11.1) and hydrodesulfurization of ethyl sulfide
(Reaction 11.2).

C6 H14 + H2 → 2C3 H8 (11.1)

C2 H5 SH + H2 → C2 H6 + H2 S (11.2)

Table 11.5 Average bond lengths and energies of selected bonds

Bond Length Energy (kJ/mol) Bond Length Energy (kJ/mol)
(pm) (pm)
C–H 109 413 H–H 74 436
C–C 154 348 H–N 101 391
C=C 134 614 H–O 96 366
C≡C 120 839 H–S – 347
C–N 147 308 N–N 145 170
C–O 143 360 N≡N 110 945
C=O – 745 O–O 148 145
C=O in CO2 803 O=O 121 498
C–S 182 272 S–S – 266
11.6 Hydrocracking 233

Fixed-bed catalytic hydrocracking employs high hydrogen partial pressures

(1000–3000 psi) and, relative to FCC, moderate temperatures (600–800 °F). Other
than hydrogenation, the role of hydrogen is to prevent buildup of coke on the catalyst.
In addition to fixed-bed hydrocracker, there are slurry-phase thermal hydrocrack-
ers such as the VCC units in China. Since catalysts cannot economically recovered
from slurry, a slurry hydrocracker uses low cost catalysts that elute off with bot-
toms streams or non-catalytic solid additives. Process conditions for slurry-phase
and fixed-bed hydrocracking are as follows:
Slurry-phase Thermal Hydrocracking
• Feedstocks: coal, vacuum residue, FCC slurry oil
• Normal operating temperature: 460–490 °C (860–920 °F)
• Normal operating pressure: 170–200 barg (2500–3000 psig)
• Conversion: up to 95 wt% conversion of vacuum residue
• Non-catalytic additive used in the VCC process. Low-metal catalysts used in Uni-
flex and other
• No need to shut down for catalyst or additive replacement.
• Products must be stabilized by subsequent hydrotreating.
Fixed-bed Catalytic Hydrocracking
• Feedstocks: gas oil, FCC light cycle oil, coker gas oil, vacuum gas oil, FCC heavy
cycle oil
• Normal operating temperature: 315–440 °C (600–825 °F)
• Normal operating pressure: 68–200 barg (1000–3000 psig)
• Liquid hourly space velocity (LHSV): 0.5–2.0
• Conversion: up to 99 wt% conversion of VGO
• Catalyst cycle life: 1–5 years, typically 2–3 years.
Ebullated-bed Catalytic Hydrocracking
• Feedstocks: atmospheric and vacuum residue
• Normal operating temperature: 420–450 °C (730–840 °F)
• Normal operating pressure: 136–200 barg (2000–3000 psig)
• Liquid hourly space velocity (LHSV): 0.1–1.5
• Conversion: up to 70 wt% conversion of vacuum residue
• No need to shut down for catalyst or additive replacement.

Table 11.6 Comparison of heats of reaction (kJ/mol) calculated from bond energies with that heats
of formation for Reactions 11.1 and 11.2
Bonds broken Bonds formed Net bond energy Calc. from Hf
Reaction 11.1 C–C, H–H 2 C–H −42 −42.3
Reaction 11.2 C–S, H–H C–H, S–H −52 −58
1. C6 H14 + H2 → 2 C3 H8
2. C2 H5 SH + H2 → C2 H6 + H2 S
234 11 Cracking

Hydrocracking can use hydrogen from a variety of sources: catalytic reformers,

steam/hydrocarbon reformers (including steam/methane reformers (SMR)), olefin
manufacturing plants, partial oxidation facilities, and electrolysis plants.
Compared to the FCC process, hydrocracking has several advantages: greater
feedstock flexibility, greater product naphtha boiling flexibility, relatively high ratio
of iso- to normal butane, and vastly superior middle distillate products. FCC works
best with straight-run paraffinic atmospheric and vacuum gas oils. Due to the lack of
external hydrogen, it makes substantial quantities of olefins. Fixed-bed hydrocrack-
ers can be designed for a greater variety of feedstocks. Most units in Asia, and a large
percentage of the units in Europe, process straight-run vacuum gas oil. Most units in
North America also process aromatic cycle oils and coker vacuum distillates, which
are relatively unreactive in catalytic cracking. Instead of making olefins, hydrocrack-
ing produces light paraffins with iso/normal ratios that are higher than predicted by
thermodynamics. For example, the isobutane:n-butane ratio is >2.0.
In hydrocracking units, abundant high-pressure hydrogen quickly saturates olefin
intermediates, suppressing the growth of polyaromatic rings. The reaction mech-
anism is similar to catalytic cracking, but with concurrent hydrogenation. For the
hydrocracking of fluorene (C13 H8 ), Lapinas, et al [20]. proposed a mechanism that
might apply generally to the hydrocracking of polyaromatics.
Figure 11.15 outlines the reaction sequence for hydrocracking a heptyl ethyl
naphthalene isomer into lighter compounds. The reactions include:
• Removal of part of a side chain by dealkylation
• Saturation of an aromatic ring
• Isomerization of the saturated ring
• Opening the saturated ring
• Paraffin hydrocracking.
The first dealkylation reaction removes hexanes from the heptyl group through
β-cleavage, not heptanes. The ethyl group is not removed, because the production of
methane by the β-scission of a two-carbon C–C bond isn’t possible thermodynami-
cally. Due to resonance stabilization, it is difficult to break C–C bonds in an aromatic
ring. Saturation is required before rings can be opened.
It is important to note that hydrocracking is a bifunctional process, combining
acid-catalyzed catalytic cracking and metals-catalyzed hydrogenation. Hydrogena-
tion is exothermic and cracking is endothermic. However, the heat released from
hydrogenation is much greater than the heat consumed by cracking. Hence, cold
hydrogen is added in quench decks to lower temperature, thereby controlling reac-
tion rates. Heat control is the primary concern during the design of hydrocracking
units, because loss of control can lead to loss of containment, causing fires and even
explosions [21]. This issue is discussed in detail later.
11.6 Hydrocracking 235

C7H15 H2
CH3
+ C6H14
CH2CH3 Dealkylation CH2CH3

CH3 3 H2 Saturation

CH2CHCH2CH2CH3 H2 CH3

CH3 Ring Opening CH2CH3

H2 Dealkylation

CH3 H2
C6H14 2 C 3H 8
+ C5H12
Paraffin
CH3 Hydrocracking

Fig. 11.15 Reaction sequence for the hydrocracking of polyaromatics

11.6.1 Fixed-Bed Catalytic Hydrocracking

The chemical mechanism for catalytic hydrocracking is essentially the same as that
for FCC catalytic cracking. The processes themselves are considerably different
due to different catalysts, equipment, and operating conditions. An integral part of
hydrocracking is hydrotreating, which in fixed-bed units removes almost all oxygen,
sulfur, nitrogen and trace elements from the feed before it reaches the cracking
catalyst. Organic nitrogen poisons acidic cracking sites, so its removal is essential.
Along the way, hydrocracking also isomerizes n-paraffins and saturates olefins and
aromatics. With respect to equipment and process flow, fixed-bed hydrocrackers are
similar to fixed-bed hydrotreaters.
As shown in Fig. 11.16, hydrocrackers can have many different configurations; for
simplicity, pumps, heaters and heat exchangers are not shown. Most commercial units
have two reactors, but at least one unit (at Chevron El Segundo) has 6 reactors. Some
reactors have two beds. Other reactors have seven beds. Some units are once-through;
the unconverted oil goes off-plot. Others recycle unconverted oil to near-extinction.
Some have two flash drums, others have four. Some have amine treaters, others do
not. In some, only the gas is heated—hot gas is mixed with warm feed just before
the 1st reactor. In others, gas and oil are mixed before heating.
• Sketch 1 shows a once-through unit in which the oil flows in series through the
pretreat catalysts directly to the cracking catalysts.
• Sketch 2 shows two options for recycle of unconverted oil (UCO). The UCO can
go either to R1 (first reactor) or R2 (2nd reactor).
236 11 Cracking

Feed Makeup
Treat Gas Feed Treat Gas Makeup Gas
Gas

H2S Removal
H2S
Optional
PT Removal
1 PT 3

Recycled UCO
HC Seps
Seps HC
Strip
Strip
Products
Products
Frac
Frac
Products
Products

UCO
UCO

Feed Makeup Feed Treat Gas

Treat Gas Makeup Gas
Gas

H2S Removal H2S

Optional Removal
PT PT Makeup Gas
2
Treat
Gas
HC 1
HC
Seps
Strip
4
Products
Strip
Frac
HC 2
Products Stripper
Overhead

Recycled UCO UCO Seps

To PT or HC Products

Strip, Frac

Products

Fig. 11.16 Hydrocracker Configurations (PT: pretreater, HC: hydrocracker, HP: high pressure, LP:
low pressure, UCO: unconverted oil). The makeup gas is hydrogen-rich with H2 ranging from 80
to 99.9%

• Sketch 3 shows a unit in which the pretreated oil is stripped or fractionated before
moving on to a hydrocracking reactor.
• Sketch 4 shows a unit with two independent recycle gas loops, one for PT and HC-1
and another for HC-2. This configuration has two major process advantages: in R3,
the pressure can be lower and the environment is nearly sweet—sweet because H2 S
and ammonia are stripped, and almost all of the nitrogen and sulfur are removed
in PT and HC-1. Temperatures required by uninhibited sweet hydrocracking are
lower. This leads to increased saturation of aromatics, which improves the quality
of middle distillate products. Some refiners still use noble-metal catalysts, due
to the higher activity and lower gas production over such catalysts. The Sketch
4 option seems like it would be more expensive, because it contains additional
equipment. But in one recent head-to-head comparison, due to the lower pressure
and less expensive metallurgy, the estimated installation cost of a Sketch 4 unit
was nearly the same as that for a Sketch 2 unit.
Figure 11.17 shows a two-reactor once-through hydrocracker with four catalyst
beds, along with some typical oil properties at various points in the process. The
liquid feed in this case is a typical straight-run VGO.
The makeup hydrogen can come from a steam-methane reformer (SMR), a cat-
alytic reformer, purified refinery purge gas, an olefins plant, or even an electrolysis
11.6 Hydrocracking 237

Fig. 11.17 Once-through single-stage (series flow) hydrocracking flow sketch with selected stream
properties

plant. The highest purity makeup (99.9 + % H2 ) comes from an SMR in which purifi-
cation is achieved with a pressure swing adsorption (PSA) unit. Makeup gas from an
SMR equipped with Benfield purification can contain 4 vol.% methane. Olefin plant
hydrogen has up to 5 vol.% methane, and electrolysis hydrogen can contain trouble-
some traces of HCl. Purified refinery purge gas can have purities ranging from 90 to
95% H2 for a membrane unit to 99.9 + % H2 for a PSA. FCC and thermal cracking
units produce a lot of low-purity hydrogen containing olefins; it is uneconomical to
purify such gases.
Hydrotreating catalysts are described above. They remove metals, organic sulfur,
and organic nitrogen. Sulfur doesn’t harm the hydrocracking catalysts in R2, but
nitrogen inhibits acid-induced cracking.
Hydrocracking catalysts are bifunctional, containing metals for saturation and
a solid acid for cracking. The active metals can be either palladium, which is an
expensive noble metal, or a base-metal sulfide (MoS2 or WS2 ) promoted by NiS.
The acid is either an amorphous aluminosilicate or a crystalline zeolite. These were
discussed in the refining catalyst section.
A mixture of liquid feed and hydrogen-rich gas enter the first reactor at 550 °F
(start-of-run) to 750 °F (end-of-run); these temperatures correspond to 290 and
400 °C, respectively. Temperatures increase by as much as 100 °F (56 °C) but in
no case are they allowed to exceed the metallurgical limit (usually 825 °F, 440 °C).
Between beds, relatively cold quench gas is added to reduce the temperature prior to
the subsequent bed. The effluent from the last first-stage reactor is sent through a heat
exchanger train, to a series of two to four separators (flash drums). Before the cold
238 11 Cracking

high-pressure separator (CHPS), wash water is added to dissolve NH4 SH and NH4 Cl,
which otherwise would foul tubes in the reactor effluent air cooler (REAC). The
resulting sour water is drained from the boot of the CHPS. In the cold high-pressure
separator, hydrogen laden with H2 S, methane, and other light gases is recovered
overhead and recycled. Sometimes, an amine unit is used to remove H2 S from the
recycle gas. The high-pressure separator bottom stream goes to low-pressure sepa-
rator(s), which remove residual gases before the liquid product proceeds to stripping
and fractionation. Some of the recycle gas is purged to maintain purity. The purge
stream, which contains 80–85% hydrogen, can go to hydrotreaters. As mentioned
above, purge gases can be purified with a membrane unit which boosts the purity of
diffused (low-pressure) gas to >95%. Purification with a PSA provides makeup with
99.9 + % H2 .
From the separators, the liquids go to a fractionator, which cuts the full-range
product into individual streams—light naphtha, heavy naphtha, middle distillates,
and unconverted oil (UCO).
There can be one middle distillate draw (diesel) or two (kerosene and diesel). The
fractionator can be operated to give sales-quality diesel. Kerosene quality is highly
feed dependent. It often meets sales specifications, but toward the end of a catalyst
cycle, high aromatics might keep it from meeting smoke-point specifications.
As mentioned, the UCO can go off-plot to an FCC unit or olefins plant. It can
also be used as lube base stock. The UCO from hydrocracking is a premium product.
Compared to the VGO from the vacuum distillation column, it has higher isoparaffin
concentration, high viscosity index, less sulfur, less nitrogen, lower aromatics, and
higher hydrogen content. When recycled to the reactors for additional conversion,
overall conversions can exceed 98 wt%.
The hydrocracker is a pivot point in the refinery—the “process in between.” It
takes straight-run VGO from the crude distillation complex, but it also takes feed
from other conversion units. It generates finished products—light naphtha and middle
distillates—but it also provides feeds to other units.
Feeds Straight-run VGO, coker gas oil, FCC cycle and slurry oils, DAO,
extract.
Intermediates Isobutane => alkylation
Light naphtha => gasoline blending
Heavy naphtha => catalytic reforming
Middle distillates => jet, diesel
Unconverted oil (UCO) => recycle, FCC, olefins plant, lube bases-
tock
As shown in Table 11.7, a fixed-bed recycle hydrocracking unit can have signifi-
cant product flexibility, producing either large amounts of C4 -plus naphtha or large
amounts of middle distillates. In petroleum refining, this kind of process flexibility
is unique.
Catalyst cycles last from 1 to 4 years, typically for two years. Units run to achieve
specified targets, such as “this much” conversion or “that much” production of FCC
11.6 Hydrocracking 239

Table 11.7 Hydrocracking product flexibility

Feed Straight-run vacuum gas oil
Boiling range, °C 340–550
Boiling range, °F 644–1020
API gravity 22.0
Specific gravity 0.9218
Nitrogen, wppm 950
Sulfur, wt% 2.5
Primary product objective Naphtha Kerosene Gas oil
Weighted average reactor temp, °C base −6 −12
Weighted average reactor temp, °F base −11 −22
Product yields, vol.%
Butanes 11 8 7
Light naphtha 25 18 16
Heavy naphtha 90 29 21
Kerosene or gas oil – 69 77
Total C4 -plus 126 124 121
Chemical H2 consumption
Nm3 /m3 345 315 292
Scf/bbl 2050 1870 1730
Product qualities
Light naphtha (C5 -82 °C)
RON Clear 79 79 80
Heavy naphtha
P/N/A 45/50/5 44/52/4 –
RON clear 41 63 67
End point, °C (°F) 216 (421) 121 (250) 118 (244)
Kerosene
Flash point, °C (°F) – 38 (100) –
Freeze point, °C (°F) – −48 (−54) –
Smoke point, mm – 30 –
FIA aromatics, vol.% – 7 –
End point, °C (°F) – 282 (540) –
Gas oil
Cloud point, °C (°F) – – −15 (5)
API gravity – – 44
Cetane number – – 55
Flash point, °C (°F) – – 52 (126)
End point, °C (°F) – – 349 (660)
240 11 Cracking

or olefin-plant feed. As a catalyst cycle progresses, it’s necessary to raise the average
temperature about 1–3 °F per month to compensate for loss of catalyst activity.
Higher hydrogen partial pressure improves almost every aspect of hydrocracker
operation. The maximum operating pressure of an existing unit is fixed, but hydrogen
partial pressure can be increased by purging recycle gas or improving makeup gas
purity.
Conversion is a function of residence time, i.e., feed rate. In an existing unit,
increasing the feed rate decreases conversion. To compensate, it’s necessary to
increase temperature.
An increase in feed nitrogen content might decrease conversion, because organic
nitrogen inhibits catalyst activity. However, pretreat reactors are operated to maintain
constant nitrogen in the feed to the hydrocracking catalyst, typically 10–30 ppmw,
so the impact of feed nitrogen on the cracking catalyst is dampened.
If a unit is not designed for high levels of feed sulfur, corrosion of equipment
could impair unit performance. Feed sulfur affects the pretreat catalyst more than it
does the hydrocracking catalyst. For both catalysts, removing hydrogen sulfide from
the recycle gas with an amine treater diminishes the impact of feed sulfur content.
The buildup of heavy polynuclear aromatics from recycled fractionator bottoms
can cause fouling of heat exchanger. Reducing feed end point or removing 2% of the
UCO in a drag stream may be necessary.

11.6.2 Ebullated Bed Hydrocracking

Heavy feeds, such as residual fuel oils and reduced crudes contain high concentrations
of asphaltenes and ash, and they contain more sulfur, nitrogen, and metal-containing
components (such as metalloporphyrins) than gas oils. Typically, catalytic residue
upgrading processes are applied to atmospheric residues (AR). The type of catalysts
and operation conditions in AR hydrocracking are different than those used for
gas oils. Vacuum residues (VR) have low hydrogen/carbon ratio and high metal
content, which deactivates catalysts rapidly. They typically are processed in non-
catalytic processes, such as solvent extraction, delayed coking, Flexicoking, and
thermal hydrocracking. (However, certain thermal hydrocracking additives can have
catalytic activity.)
In contrast to fixed-bed hydrocrackers, ebullated bed (e-bed) units can process
large amounts of atmospheric and vacuum residues (AR and VR) with high metals,
sulfur, nitrogen, asphaltenes and solid contents. They employ catalysts with both
hydrotreating and hydrocracking activity; in such units, it is impossible to segregate
catalysts, so the catalyst accomplishes both hydrotreating and hydrocracking. As dis-
cussed, hydrotreating removes sulfur and nitrogen and hydrogenates aromatic rings.
Hydrocracking entails catalytic breaking of C–C bonds. E-bed processes convert
residue-containing feeds into distillates and upgraded bottoms for FCC and other
conversion units.
11.6 Hydrocracking 241

Fig. 11.18 H-Oil Reactor as designed by IFP

Process-wise, catalyst life does not limit these units, because fresh catalyst is
continually added as spent catalyst is removed. Economics determine whether or not
the benefits of residue conversion offset the cost of catalyst replacement.
In ebullated bed reactors, hydrogen-rich recycle gas is bubbled up through a mix-
ture of oil and catalyst particles. This provides three-phase turbulent mixing, which
is needed to ensure a uniform temperature distribution. The process can tolerate sig-
nificant differences in feed quality, because in addition to manipulating temperature,
operators can change catalyst addition rates. Catalyst consumption is determined by
the concentrations of trace metals—particularly Fe, Ni and V—in the feed.
Figure 11.18 shows an H-Oil reactor that uses ebullated-bed hydrocracking tech-
nology to process heavy feedstock residues such as vacuum gasoils (VGO), deas-
phalted oils (DAO), and Coal derived oils. A fresh catalyst is continuously added and
the spent catalyst withdrawn to control the level of catalyst activity in the reactor,
enabling constant yields and product quality over time.
At the top of the reactor, catalyst is disengaged from the process fluids, which are
separated in downstream flash drums. Most of the catalyst is returned to the reactor.
Some is withdrawn and replaced with fresh catalyst. When compared to fixed-bed
processes e-bed technology offers the following advantages:
• The ability to achieve more than 70 wt% conversion of atmospheric residue.
242 11 Cracking

Fig. 11.19 LC-fining expanded-bed hydroprocessing unit [22]

• Ample free space between catalyst particles, which allows entrained solids to pass
through the reactor without accumulation, plugging, or build-up of pressure drop.
• Better liquid-product quality than delayed coking.
Disadvantages versus fixed-bed processes include high catalyst attrition, which
leads to high rates of catalyst consumption; higher installation costs due to larger
reactor volume and higher operating temperatures; and sediment formation. Recent
improvements include second-generation catalysts with lower attrition; catalyst reju-
venation, which allows the reuse of spent catalysts; improved reactor design leading
to higher single-train feed rates; and two-reactor layouts with inter-stage separation.
Another version of fluidized (or extended) bed hydrocracking is LC-fining devel-
oped by Lummus (now CB&I), shown in Fig. 11.19.

11.6.3 Slurry Phase Hydrocracking

Slurry-phase hydrocracking (Fig. 11.20) achieves up to 95 wt% conversion of vacuum

residue, FCC slurry oil, coal or coal tar. Reaction temperatures are about 840 °F
(450 °C) and pressures range from 2000 to 3000 psig (14,000–20,800 kPa). Slurry-
phase hydrocracking employs finely divided solid additives, which are proprietary
and may or may not be infused with catalytic metals.
11.6 Hydrocracking 243

Off Gases
Sulfur
Etc.
Stage 2
Hot HP Reactor
Vacuum Resid Separator

Stage 1 Gas
Reactor Processing

Light
Ends
Additive

Cold Naphtha
Separator
Vacuum Middle
Flash Distillates

Hydrogen Residue

UCO

Fig. 11.20 Two-stage slurry-phase hydrocracking process flow. Based on drawings supplied by
KBR Technology. Used with kind permission from KBR, Inc. [23]

Inside the reactor, the liquid/additive mixture behaves as a single phase due to the
small size of the additive particles. The additives prevent bulk coking by providing
highly dispersed nucleation sites for “micro coking.” The additive isn’t recovered.
Instead, it ends up in a pitch fraction, which comprises <5 wt% of the feed.
Slurry-phase hydrocracking has several advantages:
• It can achieve >95 wt% conversion of vacuum residue and high conversion of coal.
• In two-stage designs, which incorporate fixed-bed hydrotreating and hydrocrack-
ing, product quality is excellent.
• Feeds can include vacuum residue, FCC slurry oil, and even coal.
• The additive is low-cost and disposable.
• For a given volume of residue feed, total slurry phase reactor volume is lower than
the reactor volume of e-bed processes.
The main disadvantage of the KBR slurry-phase hydrocracking is the unconverted
pitch. The pitch quality is so low that it is exceptionally difficult to dispose.
A better process is LC-MAX, offered by CB&I. LC-MAX is a combination of
LC FINING and solvent deasphalting. The conversion to fuels is somewhat lower,
but all products, including the asphalt, can be sold conventionally.

References

1. Greensfelder BS, Voge HH, Good GM (1949) Catalytic and thermal cracking of pure hydro-
carbons. Ind Eng Chem 41(11):2573–2584
2. Zhu G, Xie C, Li Z, Wang X (2017) Catalytic processes for light olefin production (Chapter 36).
In: Hsu CS, Robinson PR (eds) Springer handbook of petroleum technology. Springer, New
York
244 11 Cracking

3. https://www.quora.com/What-does-a-steam-cracker-look-like-and-what-are-its-essential-
components. Accessed 6 Aug 2014
4. Hsu CS, Robinson PR (2017) Gasoline production and blending (Chapter 17). In: Hsu CS,
Robinson PR (eds) Springer handbook of petroleum technology. Springer, New York
5. Speight J (2017) Fluid-bed catalytic cracking (Chapter 19). In: Hsu CS, Robinson PR (eds)
Springer handbook of petroleum technology. Springer, Heidelberg
6. Avidan A, Edwards M, Owen H (1980) Experiments performed at constant coke production.
Oil Gas J 88(1):52
7. von Ballmoos R, Harris DH, Magee JS (1995) Catalytic cracking (Section 3.10). Encyclopedia
of Catalysis. Wiley, London, pp 1955–1985
8. Van Antwerpen FJ (1944) Thermofor catalytic cracking. Ind Eng Chem 36(8):694–698
9. AIChE (2009) Chemical engineers and energy: In: Chemical engineers in action—innovation
10. Palmas P (2009) Traces of the history of RFCC and provides guidelines for choosing the appro-
priate regenerator style: https://www.uop.com/?document=uop-25-years-of-rfcc-innovation-
tech-paper&download=1. A reprint from hydrocarbon engineering
11. Park JI, Mochida M, Marah AMJ, Al-Mutairi A (2017) Modern approaches to hydrotreat-
ing catalysis (Chapter 21). In: Hsu CS, Robinson PR (eds) Springer handbook of petroleum
technology. Springer, New York
12. Resid Upgrading, By Honey UOP: https://www.uop.com/processing-solutions/refining/
residue-upgrading/#resid-fcc. Accessed 31 Aug 2016
13. Honeywell UOP, “Catalyst Cooler,” https://www.uop.com/equipment/fcc/catalyst-cooler/.
Accessed 1 Nov 2018
14. Prime-G +: The benchmark technology for ultra-low sulfur gasoline by Axens: https://zh.
scribd.com/doc/209103920/Prime-G. Accessed 20 Sep 2106
15. Laan JV ConocoPhillips S Zorb Gasoline Sulfur Removal Technology: http://www.icheh.com/
Files/Posts/Portal1/S-Zorb.pdf. Accessed 20 Sep 2016
16. Schroder M (2010) Functional metal organic frameworks: gas storage. Sep Catal 293:175–205
17. Olah GA (1994) My search for carbocations and their role in chemistry. In: Nobel Lecture
18. Robinson PR, Dolbear GE (2017) Hydrocracking (Chapter 22). In: Hsu CS, Robinson PR (eds)
Springer handbook of petroleum technology. Springer, New York
19. Bergius Process, Project Gutenberg, http://self.gutenberg.org/articles/Bergius_process.
Retrieved 4 Aug 2015
20. Lapinas AT, Klein MT, Gates BC, Macris A, Lyons JE (1991) Catalytic hydrogenation and
hydrocracking of fluorene: reaction pathways, kinetics, and mechanisms. Ind Eng Chem Res
30(1):42–50
21. https://archive.epa.gov/emergencies/docs/chem/web/pdf/tosco.pdf. Retrieved 9 Oct 2018
22. Chevron Technology Marketing: LC Fining: http://www.chevrontechnologymarketing.com/
CLGtech/lc_finishing.aspx Retrieved 4 Aug 2015
23. Motaghi M, Ulrich B, Subramanian A (2011) Slurry-phase hydrocracking: possible solution
to refining margins. Hydrocarbon Proces 90(2):37