The first fluid catalytic cracking (FCC) process was
introduced almost 72 years ago; yet, this technology remains a principle
conversion process in a modern refinery. Much has changed since the
original FCC unit (FCCU) became operational. Properly operated,
an FCCU can produce a variety of valuable refined products,
including olefins and aromatics in addition to high-quality
transportation fuel products. Various refined products can be
produced via an FCCU only with changes in operating condition,
different feedstocks and advanced/specialty
Maximizing individual products
Numerous papers have been presented that focus on maximizing
the various products produced by an FCCU. These topics usually
appear when the economics favor a particular yield. This
article can be used as a reference for FCC operators to apply
as a guide on what is necessary to maximize (or minimize) the
yield of the many diverse FCC products.
To achieve top performance from an FCCU, several factors
must be considered. First, the design must be conducive to the
making of the particular product. Second, the properties of the
feedstock are important since the molecules entering the unit
will determine what can potentially be made.
Optimization of the many operating variables is required,
along with using proper catalyst. Using the right catalyst in
the proper way is the single most important action that a
refiner can take to maximize FCC performance.
From the beginning, the FCCU was primarily a gasoline
machine. Feedstocks were atmospheric gasoil
(AGO) and vacuum gasoil (VGO), and the objective was to produce
high yields of gasoline. Originally, the gasoline yield was
maximized by recycling both light- and heavy-cycle oils. The
coke make was almost twice as high as necessary since the
recycle rates could be as much as 100% of the fresh feedrate.
Some designs had two risers that were of equal size.
The present-day design parameters for an FCCU to maximize
- A straight and vertical reaction system that operates in
a gas/solid mode with minimum slip and an optimum residence
- Feed injectors that vaporize the feed as fast as possible
and reduce the incoming regenerated catalyst temperature to
the mix temperature quickly.
Rapid separation of the spent catalyst from the reaction
products. This quickly terminates the reactions, thus
minimizing the dilute-phase residence time and/or the
temperature of the product vapors.
Stripping should be as efficient since hydrocarbons
carried over to the regenerator are primarily gasoline
and diesel. Excessive residence time of the hydrocarbons
in the stripper will convert them to light gases, thus
reducing gasoline yield.
Regeneration should be essentially complete since the
left on the catalyst is
associated with the large-pore molecular sieves used in
Feedstock has a very marked impact on gasoline
yields.1 Since gasoline typically has a hydrogen
content of about 13.5 wt%, a feedstock that is higher in
hydrogen will produce more gasoline yield. Table
1 shows results from a cracking study done in a
circulating pilot plant with a Middle East GO, a severely
hydrotreated Louisiana GO, and the atmospheric bottoms from a
US shale oil.2 The shale oil data was adjusted to
better simulate the expected commercial results.
The data show a direct correlation between hydrogen content
of the feed and gasoline yield. Even though the tight oil
contains some 1,050°F plus material, the gasoline yield
exceeds 57 wt% or 67 vol% on fresh feed.
Operating variables must be manipulated to attain high
conversion while minimizing coke and light gases
make.3 Maximum gasoline yield usually occurs at
conversion levels between 80 vol% and 85 vol%. The conversion
will be lower when processing aromatic feeds. Highest gasoline
yields are achieved by maximizing catalytic activity within the
parameters of the reaction system. Too much activity will give
too low catalyst-to-oil (C/O) ratio, and it could lead to
catalyst deactivation due to higher Δ coke.
As summarized in Table 2, the reactor
temperature will usually be between 960°F and 985°F.
Lower temperatures adversely affect the stripper operation, and
the higher temperatures will overcrack the gasoline
Table 3 summarizes the effects of the major
operating variables on an FCC operation.4 In this
instance, reaction temperatures above 980°F will have lower
gasoline yields as did the heavy-cycle oil recycle. Recycle at
high conversion is usually used for bottoms cracking rather
than for producing more gasoline. Increasing catalyst activity
and the C/O ratio (by lowering feed temperature), and reducing
the reactor pressure, will increase gasoline yield. While
lowering the feed temperature will make more coke, the dry gas
yield may be reduced. If heavy feeds are processed, the ability
of the feed injection system to vaporize the feed may set a
minimum feed temperature.
Catalyst properties must be tuned to the particular FCC
operation. Both the feedstock and equipment limitations impact
the choice of catalyst and additives. Gasoline and conversion
may not be maximized if the unit is operating against multiple
constraints, such as the air blower, wet-gas compressor and
The main catalyst variables that can be controlled are
faujasite zeolite content and type and degree of exchange of
the zeolite. Matrix activity, pore structure and total pore
volume, and metals passivators are all matrix components, which
are varied to optimize the FCCU.
Catalysts containing intermediate pore-size zeolites (ZSM-5)
need to be excluded. This is sometimes forgotten when an
equilibrium catalyst is used.
The equilibrium unit cell size needs to be optimized as
well. For maximum gasoline, values ranging from 24.32 to 24.40
are used. Feeds with few coke precursors would benefit from
larger numbers, while heavier or more aromatic feeds normally
require lower values. Coke and gas selectivities are usually
limited in that case. The starting unit cell size should be as
close to the equilibrium value as possible since the fresh
catalyst will play a significant role in the overall cracking
A few years ago, the only products making money for US
refiners were the middle distillates, i.e., diesel, kerosine,
heating oil and jet fuel. The entire refinery became focused on
maximizing middle-distillate products including the FCCU. As
with gasoline maximization, all aspects of the FCCU must be
Diesel is the first product from the cracking reactions, and
it reaches a peak before the gasoline yield, as shown in
When diesel-range material cracks, the primary product is
gasoline. A riser would be the preferred reactor design with no
back-mixing. Contact time should be short, and the recycle of
unconverted feed is required since the bottoms yield would be
1. Effect of zeolite content on lco
at constant operating severity.
Feed injection systems must vaporize the
feedstock quickly while quenching the hot catalyst from the
regenerator. This minimizes Δ coke and dry gas make. A
quick separation of the hydrocarbons and spent catalyst
minimizes dry gas make. While bottoms cracking occurs in the
dilute phase of the reactor, it is more effective to recycle
the unconverted feed.
Efficient stripping minimizes the amount of
middle distillates that are burned in the regenerator, thus
reducing the loading to the gas plant. Better stripping
provides more operating flexibility to optimize other operating
Regenerator. Regeneration of the catalyst
should be efficient. However, striving for the lowest carbon on
catalyst may not be desirable. The strongest acid sites on the
catalyst tend to crack the feed all the way to gasoline. Since
the residual carbon is associated with these sites, a carbon
level of about 0.1 wt% to 0.2 wt% may be desirable, and it
would depend on the catalyst (formulation) used.
Diesel-free feed. The feed to the cracker
should not have any diesel present. This material is
preferentially cracked to gasoline, and there is a large cetane
loss. If a cat-feed hydrotreater is being used, then the
operator should consider operating it as a mild hydrocracker.
This action is more selective to diesel than an FCCU, and it
provides a high-quality product for the diesel pool.
Hydrocracker. If the refinery has a hydrocracker, it
should be operated at capacity. This may require additional
hydrogen for the plant, but it should be economical. At lower
FCC conversions, the light-cycle oil (LCO) is a higher-quality
product and requires less-severe post-treating.5
Feedstocks with two and three aromatic
rings will make more LCO than paraffins, since the aromatic
nuclei are resistant to cracking reactions. Recycle streams and
coker GOs are relatively rich in these molecules, and they will
produce more middle distillates. These can be processed in the
fresh feed riser or in a separate dedicated reactor riser.
Coker GOs boiling in the diesel range would go to a
middle-distillate hydrotreater rather than the FCCU.
Main fractionators should have a
heavy-cycle oil (HCO) draw so that a more coke-selective stream
can be sent to the FCC.6 Decant oil should go
through dedicated nozzles to preserve the integrity of the
regular feed nozzles. Refiners have reported better overall
yields when the decant oil is injected well downstream of the
Catalysts used to maximize LCO should be
very low in activity.7 As shown in Fig.
2, this is true for rare earth Y. Type Y or
ultra-stable zeolites have large pores, and they are very
effective at cracking diesel-sized molecules. Type Y or
ultra-stable zeolites must be limited to about 5%10% in a
catalyst formulation to prevent overcracking of the LCO.
Studies also suggest that smaller crystal sizes would help by
allowing the LCO produced to diffuse more rapidly out of the
2. Effect of zeolite concentration
on LCO yields.
Matrix activity should be maximized to give
higher first-pass cracking, and the catalyst should have mainly
intermediate or low-acid strength sites. Strong-acid sites
produce gasoline. An open-pore structure is desired to minimize
LCO overcracking. Compositions that include magnesium have been
proven to make more diesel due to their unique acid-strength
Operating variables are manipulated to give
low conversions. Table 4 summarizes various
operating variables to control. Reactor temperatures of
930°F950°F are used for GO, while resids may
require 950°F960°F to avoid excessive hydrocarbon carryover from the stripper.
Feed preheat may be maximized, and a fired heater would be
required for the more crackable feeds.
Recycle is essential to improve bottoms cracking with low
conversions. Rates of 1%30% would be required to give
LCO/GO ratios of at least 3, and, in many cases, over 5, for
these middle-distillate operations.
The iso-C4 hydrocarbons are very valuable.
Refiners frequently want to maximize or, at least, increase one
of them. These molecules are isobutylene and
isobutane.3 Both are important feeds to an
alkylation unit, and the isobutylene is used to make
methyl-tertiary butyl ether (MTBE), the preferred oxygenate
used globally except in North America. Isobutylene can also
serve as a feed to a catalytic polymerization unit or as a
similar process that makes gasoline from light olefins.
Isobutane is required for alkylation. Some refiners are short
of this material due to a lack of local field butane
The C4 hydrocarbons are generally one of the
ultimate products from a catalytic cracking unit due to the
beta-scission carbenium ion cracking mechanism. Equilibrium
concentrations of the various C4s are rarely
achieved due to the reactivity of the butylenes. The
isobutylene equilibrium sets the maximum amount of isobutylene
that can be produced in a typical catalytic
cracker.8 At equilibrium conditions, the percentage
of isobutylene in the butenes stream varies from 46% at
950°F to 44% at 1,050°F.9 The actual
percentage of isobutylene in the butene stream depends on the
unit design operating conditions, feedstocks and catalyst. While
isobutylene is initially produced, hydrogen transfer reactions
can diminish its yield.
Unit design features that help preserve isobutylene are: a
short reactor contact time, and rapid separation of the spent
catalyst and reactor effluent. Reducing the dilute phase
temperature and contact time also is important. Low reactor
hydrocarbon partial pressures are essential for minimizing
hydrogen transfer, and FCCUs operating above 30 psig may not be
able to make the needed isobutylene. Dispersion and/or riser
stream will lower the hydrogen transfer reactions.
Like every other product from an FCCU, feedstock plays an
important role in isobutylene manufacturing. In general, high
K factor feeds will give more isobutylene than more
aromatic stocks. More paraffinic feeds can operate at higher
conversions and generate more LPG. Typically, LPG has a
hydrogen content well above 15 wt%, so hydrogen-deficient feeds
cannot make as many barrels of isobutylene. The percentage of
the olefins may be lower, depending on the operating variables
and conversion levels. Key feedstock parameters would include
the amount and configuration of the naphthenes. The length of
the side chains on the ring compounds will determine whether
C3 or C4 olefins are formed. The main
operating variable for producing more isobutylene is reactor
temperature. The amount of the C4 iso-olefin
increases significantly when the gasoline is over-cracked.
Other variables that increase conversion can also increase
isobutylene yield. These include higher catalyst activity and
higher C/O ratios. Both can promote hydrogen transfer
reactions; thus, the amount of isobutylene will not be as high
as that yielded by increasing reactor temperature. If the
conversion is taken too high, then the higher reactivity of the
isobutylene will diminish its yield.
Catalyst properties can have a very large effect on
isobutylene yield. The goal is to get conversion, but with
minimum hydrogen transfer. The large-pore zeolite used should
be an ultra-stable Y (US-Y) rather than a rare earth Y (ReY).
Minimal RE should be used for stabilization. A lower unit cell
size is desirable, and it is important that the fresh catalyst
also be low in unit cell size. High unit cell sizes mean the
acid sites on the zeolite are closer together, and that
promotes hydrogen transfer.
An active matrix will provide much needed catalytic activity
and has minimal hydrogen transfer activity. Bronsted acid sites
would be preferred to Lewis acids since they would promote more
skeletal isomerization of the olefins. ZSM-5 tends to increase
the isobutylene concentration since the gasoline that cracks
within the medium-pore zeolite produces a concentration of
isobutylene near the equilibrium value (40% or higher).
Isobutane yield will increase with conversion but can be
cracked thermally in the feed injection zone. When reactor
temperatures reach 1,030°F, the base of the riser is near
1,100°F. Since isobutane is a function of hydrogen transfer
reactions, the opposite factors for maximizing isobutylene
To minimize thermal cracking, the best conditions are good
feed injection system, low slip factor riser, rapid and high
separation of the spent catalyst and reactor products, short
dilute phase contact times and/or low temperatures, and a
highly effective stripper and regenerator.
Feedstocks higher in hydrogen work
best for the same reasons outlined in the previous section
(isobutylenes). Tight oils would be expected to produce high
quantities of isobutane with the proper unit design and
While high conversions are desired, catalytic conversion is
preferred to thermal conversion (reactor temperature). Higher
catalyst activity can be achieved by increasing the activity of
the catalyst or the C/O ratio by reducing feed temperature. It
is normally more economical to first increase the RE content of
the catalyst to raise activity, then increase the large-pore
zeolite content, and finally raise catalyst additions. A higher
UCS is usually desired. Using ZSM-5 should be reduced or
Butylenes are the most desirable light olefins in a
gasoline-oriented refinery. These often have a value of
gasoline or higher since they are the preferred feed to an
alkylation unit. Butylenes give the highest octanes and consume
less isobutane than propylene. Amylenes are usually only
processed to reduce gasoline olefinicity and/or vapor
All of the caveats that applied to making more isobutylene
apply to maximizing butylenes. The unit design and feedstock properties would not be
different. Reactor temperature is the key operating variable.
Normally, a minimum reactor temperature of 980°F would be
used to make light olefins. Many refiners will run over
1,000°F reactor temperature when operating in this mode.
Low-hydrogen transfer and controlled activity are desired with
an ultra-stable zeolite with a unit cell size of 24.32
angstroms or less. Some units might run as low as 24.27
angstroms, but dry gas or low conversion may become a
Some catalyst additives are aimed at increasing butylenes
vs. propylene. Since the additive suppliers have more
experience with their products than a typical refiner, they
should be consulted as to a product recommendation and a yield
estimate. Increased matrix activity is also desired for
maximizing butylenes and modifications to the ZSM-5 carrier may
also help. HP
This is an updated version of an earlier presentation at the
American Fuels and Petrochemical Association Annual
Meeting, March 16-18, 2014, at Orlando, Florida.
1 Bryden, K. J. and E. T. Habib, Jr.,
Processing shale oils in FCC: Challenges and
opportunities, Hydrocarbon Processing, pp.
5964, September 2013.
2 Huovie, C., et. al., Solutions for FCC
Refiners in the Shale Oil Era, AFPM Annual Meeting, March
3 Alternative Routes to High Conversion,
The Catalyst Report, TI-805 Engelhard (BASF).
4 Gim, S., W. Letzsch, H. McQuiston and C. Santner,
FCC Opportunities at Lower Throughputs, AFPM Annual
Meeting, March 2010.
5 Unzelman, G., Potential Impact of Cracking
on Diesel Fuel Quality, Katalistics 4th Annual FCC
6 Niccum, P., Maximizing diesel production in
an FCC-centered refineryPart 1,
Hydrocarbon Processing, September 2012.
7 Pillai, R. and P. Niccum, Select new
production strategies for FCC light cycle oil,
Hydrocarbon Processing, February
8 McLean, J. B. , G. S. Koermer and R. J. Madon,
Maximizing catalytic isobutylene selectivity, paper
9 McLean, J. B. and A. Witsoshkin, Iso-olefins
for oxygenate production using Isoplus, NPRA Annual
Meeting, March 1993.
Warren S. Letzsch has 46 years of
experience in petroleum refining including
petroleum catalysts, refining, and engineering and
design. His positions have included R & D,
technical service and sales, which led to senior
management positions in sales, marketing and technology development and
oversight. He was one of the developers of the
Technip/Axens R2R process and has authored over 80
technical papers. Mr. Letzsch holds eight patents in
the field of fluid catalytic cracking. He was the
FCC/DCC Program manager at Stone & Webster for 10
years and is now a senior refining consultant for
Technip as well as a private consultant to the
Christopher Dean is an independent
process engineering consultant with over 37 years in
the worldwide refining business with an emphasis on
high olefin fluid catalytic cracking (HOFCC) with petrochemical integration. He is the
founder and principal consultant for High Olefins FCC
Technology Services LLC.
His worldwide refining background
includes the development and commercialization of the
High Severity-FCC Process, the development of several
integrated refinery and petrochemical projects, catalyst
technical service, process engineering, design and
unit operations on a variety of refinery units. He has
published or presented over 30 papers and has been
issued two patents on FCC gasoline desulfurization
and has three other FCC pending process