Training material

Training module: Residue fluid catalytic cracking unit

Duration: 8 days

Delivery: L=4 days; S=4 days;

Prepared for:

NGHI SON REFINERY & PETROCHEMICAL LLC

Prepared by:

PETROVIETNAM MANPOWER TRAINING COLLEGE

Issued by PVMTC Approved by TMC/ NSRP

Prepared Checked Approved Reviewed Approved

Version
(date) (date) (date) (date) (date)

A.1

Page 1 of 83
 Table of content

I. Process Description ....................................................................................... 5

II. Residue Fluid Catalytic Cracking ............................................................ 9
1. Purpose of the unit ..................................................................................... 9
2. About RFCC R2R ...................................................................................... 9
3. Feed preheat description ......................................................................... 10
4. Reaction section description ................................................................... 11
a. Feed injection zone ............................................................................... 11
b. Riser/Disengager ................................................................................ 13
c. Stripper .................................................................................................. 15
d. Spent catalyst transfer ...................................................................... 15
5. Regeneration section ................................................................................ 16
a. Air blower and air heaters ................................................................... 16
b. First stage regenerator ...................................................................... 16
c. Second stage regenerator ..................................................................... 17
6. Main Fractionation section ..................................................................... 18
7. RFCC R2R’s features .............................................................................. 20
8. Control philosophy of the process .......................................................... 23
9. Process variables ...................................................................................... 25
a. General................................................................................................... 25
b. Riser Outlet Temperature ................................................................ 26
c. Regenerators temperature ................................................................... 27
d. Regenerators residence time............................................................. 28
e. Regenerators velocities ......................................................................... 29
f. Stripper operation ................................................................................ 29
g. Regenerators air balance ..................................................................... 29
III. Chemical reactions and catalysts ........................................................... 31
1. Objective of this section........................................................................... 31
2. Thermodynamics and kinetics ................................................................ 31

Page 2 of 83
 3. Catalyst activity, selectivity, stability ..................................................... 32
4. Types of reactions .................................................................................... 32
a. Thermal cracking ................................................................................. 33
b. Catalyst cracking ............................................................................... 34
5. Desired reactions ...................................................................................... 37
a. Cracking reactions ................................................................................ 37
b. Isomerization reaction....................................................................... 38
6. Reactions to be limited (but not eliminated) ......................................... 39
a. Hydrogen transfer ................................................................................ 39
b. Condensation reactions ..................................................................... 39
7. Undesirable reactions to be reduced to the minimum ......................... 39
a. Hydrogen formation ............................................................................. 39
b. C1 and C2 hydrocarbons ................................................................... 40
8. Conversion selectivity of various hydrocarbon families ...................... 40
9. Definitions of conversion ......................................................................... 40
a. Conversion ............................................................................................. 40
b. Industrial conversion rate................................................................. 41
c. Adjusted conversion rate ..................................................................... 42
IV. Catalyst ..................................................................................................... 42
1. Catalyst characteristics ........................................................................... 42
2. FCC catalyst fines .................................................................................... 45
3. Catalyst mechanism ................................................................................. 45
4. Catalyst contaminants ............................................................................. 46
a. Definition ............................................................................................... 46
b. Inhibitors or activity moderators ..................................................... 46
c. Temporary poisons ............................................................................... 47
d. Permanent poisons............................................................................. 48
5. Catalyst regeneration .............................................................................. 49
V. Process variables ...................................................................................... 51
1. Introduction .............................................................................................. 51
2. FCC reaction variables............................................................................ 52

Page 3 of 83
 3. Feed temperature effects ......................................................................... 53
4. Riser Outlet Temperature effects ........................................................... 55
5. Effects of contact time ............................................................................. 57
6. Effects of catalyst activity ....................................................................... 59
7. Effects of fresh feed quality .................................................................... 60
VI. Operation of the unit ............................................................................... 62
1. Reaction and Regeneration Section ....................................................... 62
a. Control of the process........................................................................... 62
b. Operating parameters ....................................................................... 64
c. Adjustment of operating conditions.................................................... 75
d. Troubleshooting ................................................................................. 77
2. Main Fractionation and Gas Recovery Sections................................... 80
a. Control of the Main Fractionation section. ............................................ 80
b. Operating parameters for several specifications of products .............. 80
c. Operating parameters for Slurry section................................................. 81
d. Wet Gas Compressor – Absorption section of Gas Recovery ............ 82

Page 4 of 83
I. Process Description

Global demand for transportation fuels will continue to grow and this demand
will be met largely by gasoline and diesel fuels. The fluid catalytic cracking
(FCC) process continues to play a key role in an integrated refinery as the
primary conversion process of crude oil to lighter products. In the next two
decades, the FCC process will be likely used for biofuels and possibly for
reducing CO2 emissions. For many refiners, the cat cracker is the key to
profitability because the successful operation of the unit determines whether or
not the refiner can remain competitive in today’s market.

The FCC unit uses a “microspherical” catalyst that behaves like a liquid when it
is properly fluidized. The main purpose of the FCC unit is to convert high-
boiling petroleum fractions called gas oil to high-value transportation fuels
(gasoline, jet fuel, and diesel). FCC feedstock is often the gas oil portion of
crude oil that commonly boils in the 330 – 550oC range.

Figures 1-2 contain diagram of typical unit configurations offered by the FCC
technology licensors. Although the mechanical configuration of individual FCC
units may differ, their common objective is to upgrade low-value feedstock to
the more valuable products used for transportation and petrochemical industries.
Worldwide, about 45% of all gasoline comes from FCC and ancillary units such
as the alkylation unit.

Page 5 of 83
Figure 1: Example of KBR Orthoflow (left) & a side-by-side design FCC unit.

Page 6 of 83
Figure 2: Example of The Shaw Group Inc. design FCC unit

Page 7 of 83
Figure 3: A typical high-conversion refinery.

Before proceeding, it is helpful to understand how a typical cat cracker fits into
the refining process. A petroleum refinery is composed of several processing
units which convert the raw crude oil into usable products such as gasoline,
diesel, jet fuel, and heating oil (Figure 3). The crude unit is the first unit in this

Page 8 of 83
refining process. Here, the raw crude is distilled into several intermediate
products: naphtha, kerosene, diesel, and gas oil.

II. Residue Fluid Catalytic Cracking

1. Purpose of the unit

The main purpose of the Residue Fluid Catalytic Cracking Process is to convert
various reduced crudes to lower boiling, high value products, primarily C3-C4
LPG, gasoline, and light cycle oil. Using vapor phase chemical reactions in the
presence of specialized RFCC cracking catalyst, the high molecular weight
RFCC feedstock is cracked to shorter chain molecules.

Heat for the cracking process is supplied by the hot regenerated catalyst which
vaporizes the finely atomized oil feed and sets the stage for the rapid but
selective cracking process. The vaporization and cracking reactions occur in the
“reactor-riser” in roughly two seconds.

As by-products of the reaction, fuel gas, slurry oil, and coke are also generated
in the “reactor-riser”. The majority of the RFCC equipment handles catalyst /
vapor product separation and removal of the coke from the catalyst, while only a
small part of the system is directly used for the cracking reaction.

2. About RFCC R2R

The R2R Process is a fluid catalytic cracking process incorporating a two-stage

regeneration system, a unique feed injection system, mixed temperature control,
back flush oil injection, an efficient riser termination system, and effective air /
steam distribution devices a proprietary catalyst separator in the disengager and
a catalyst cooler. This proven process eliminates or greatly reduces many of the
constraints of previous FCC configurations and offers maximum flexibility for
converting reduced crudes and mixtures of gas oils and vacuum residues.
The core of the reaction and regeneration section of this process unit consists in
feed injection system, riser, riser outlet separator system, disengager/ stripper,
Page 9 of 83
first stage regenerator, second stage regenerator, catalyst cooler, catalyst
withdrawal well, catalyst transfer lines and control systems.

3. Feed preheat description

Hydrotreated Atmospheric Residue from storage and/or from the Residue

Hydrodesulphurisation unit enters the Feed Surge Drum at a temperature of
80°C. A water boot on the Feed Surge Drum allows for draining of any free
water and/or sludge which may accumulate during start-up or upset conditions
to storage with the Oily Water/Sludge Pump.

Figure 4: Typical feed preheat system (FC=flow control, LC=level control,

TC=temperature control, LCO=light cycle oil).

From the Feed Surge Drum, the feed is pumped by the Feed Pumps to the feed
exchangers. The feed is preheated to 180°C by heat exchange against the Top
Pumparound in, against the LCO product in, against the LCO pumparound and
against the HCO pumparound.

Page 10 of 83
The preheat train must be flexible in order to be able to adjust the temperature of
the feed so that:
 a proper oil viscosity is ensured (around 10 to 15 cSt maximum at the
injector inlet) for correct atomization of the feed ;
 a minimum temperature is reached to avoid steam condensation in the
feed injectors.
The feed temperature must also be optimized depending on the heat balance.
The feed temperature affects significantly the coke production and the second
regenerator temperature.

4. Reaction section description

Feed is brought to the base of the riser and divided into equal flows, to each of
the feed nozzles. Feed is finely atomized and mixed with dispersion steam in the
feed nozzles and injected into the riser. The small droplets of feed contact hot
regenerated catalyst in a co current way and vaporize immediately. The
vaporized oil intimately mixes with the catalyst particles and cracks into lighter,
more valuable products along with slurry oil, coke and gas.
The product vapors travel up the riser while carrying the catalyst. Residence
time in the riser is approximately 2 seconds at design conditions.

a. Feed injection zone

Oil feed to the riser is preheated before entering the reaction system. Preheat
temperature along with regenerated catalyst temperature is controlled to result in
an optimum catalyst to oil ratio. Shutdown valves will stop the flow of feed to
the riser and divert it back to the feed surge drum in case of emergency. Flow
controllers regulate the flow to each of the feed nozzles.
The feed should be split evenly between the feed nozzles as observed by
individual flow indicators. Pressures on each feed nozzle should be monitored as
a verification of flow and indication of nozzle condition.

Page 11 of 83
Figure 5: Typical riser Wye feed section.

Dispersion steam is supplied to each of the feed nozzles to promote atomization

and vaporization of the feed. Upstream the feed injection, stabilization steam is
injected in the riser, through the stabilization steam injectors, in order to

Page 12 of 83
promote a smooth and homogeneous catalyst flow at the feed injection point.
The flow to each of the injectors is adjusted by flow controllers.
MTC injectors are provided above the feed injection zone to inject the recycled
hydrocarbons from main fractionator. MTC plays an important role in the heat
balance control and heavy feed vaporization. The key is to achieve a higher
temperature in the fresh feed mixing zone. As for the feed injectors, MTC flow
and dispersion steam flow to each injector are adjusted by flow controllers.
Upon emergency shutdown the steam flow control valves on all injectors
automatically open to clear the riser of oil and catalyst with steam.

b. Riser/Disengager

The sensible heat, heat of vaporization and heat of reaction required by the feed
is supplied by the hot regenerated catalyst. The riser outlet temperature is
controlled by the amount of regenerated catalyst admitted to the riser through
the regenerated catalyst slide valve. In the wye section at the base of the riser,
steam is injected via a steam ring to keep the regenerated catalyst in a fluid state
at all times.
The cracking reactions take place during the two second residence time in the
riser as the reaction mixture accelerates toward the Riser Separator System
(RSS).
The catalyst is quickly separated from the hydrocarbon/steam vapors in the RSS
separator located at the end of the riser. This separation is necessary to discour
age the undesirable continuation of reactions which produce gas at the expense
of gasoline. This system drastically reduces the post riser catalyst/vapors contact
time and consequently the fuel gas production due to thermal cracking.
After exiting the RSS separator, the vapors pass through high efficiency single
stage cyclones to complete the separation of catalyst from vapors, thus
minimizing the amount of catalyst lost into the product. The disengager product
vapors, containing a small amount of inert gases and steam, flow to the quench

Page 13 of 83
zone of the main fractionator. The small amount of catalyst contained in the
product vapors is carried away from the fractionator in the bottom slurry
product.
The disengager pressure "floats” on the main fractionator pressure and as such is
not directly controlled at the converter section. A pressure control at the main
column overhead receiver allows achieving a steady operating pressure in the
reaction system.
The RSS separator and disengager cyclones separate the product vapors from
spent catalyst and return the catalyst to the stripper bed. On one hand, in order to
reach a good separation efficiency and to minimize catalyst losses, RSS inlet
velocity and disengager cyclones inlet velocity have to be higher than 10 m/s.
This corresponds to the minimum required energy for correct cyclones
performance. On the other hand, in order to avoid erosion of RSS and cyclones,
the inlet velocities to those equipments have to be maintained below 21 m/s.

Figure 6: Typical trickle valve drawing

The cyclone diplegs are equipped with trickle-valves to prevent reverse flow of
gas up the diplegs. Also, the RSS separator is equipped with diplegs fitted on its

Page 14 of 83
prestripping chambers. These diplegs are sealed into the stripper catalyst bed in
order to avoid any possibility of vapors entrainment in the stripper. As
disengager cyclones are coupled to the RSS separator and catalyst diplegs are
sealed in the stripper bed, gas residence time in the disengager is very short.
This leads to very low dry gas production caused by thermal cracking.

c. Stripper

Catalyst exiting the RSS separator flows in the stripper bed. Stripping of
hydrocarbons is achieved by steam injected in the main ring. Fluffing ring is
located in the bottom head of the stripper to achieve a stable fluidization at the
inlet of the spent catalyst standpipe and a good flow conditions of spent catalyst
to the spent catalyst stand pipe nozzle.

Figure 7: An example of a catalyst stripper.

d. Spent catalyst transfer

The stripped spent catalyst flows down the spent catalyst standpipe and through
the spent catalyst slide valve. Aeration by fuel gas (or nitrogen during start-up)
is added to the standpipe at several elevations to maintain proper density and
fluid characteristics of the spent catalyst emulsion. The spent catalyst slide valve

Page 15 of 83
controls the level in the stripper by controlling the flow of spent catalyst from
the stripper. The spent catalyst then flows into the regenerator.

5. Regeneration section

The first stage regenerator burns 50 to 80% of the coke and the remainder is
burned in the second stage regenerator. This two stage approach to regeneration
adds considerable flexibility to the process as potential heat is rejected in the
first stage regenerator in the form of CO.
When running heavy feeds, the need for heat rejection is higher, the amount of
coke burned in the first stage regenerator is increased, thereby lowering the final
temperature of the regenerated catalyst. When running lighter feeds the amount
of coke burned in the first stage regenerator is reduced, thus increasing the
regenerated catalyst temperature. The amount of coke burned in the first stage
can be modified by adjusting the air flowrate in order to achieve operating
flexibility on the unit for different feedstocks.

a. Air blower and air heaters

The combustion air required is supplied by an air blower, commonly driven by a

steam turbine. The steam supply to the turbine is throttled on cascade air flow
trim control/compressor speed.
The blower air is distributed to a header system providing combustion air to first
regenerator rings, second regenerator ring, lift air, withdrawal well ring. Low air
flow to either regenerator will trigger the emergency shutdown circuit in the
case of blower failure. The air heaters are used during start-up to heat-up the
equipment including dry out of refractory.

b. First stage regenerator

Spent catalyst containing roughly 0.8 to 1.5 wt % coke flows from the spent
catalyst distributor and is spread across the bed in the first stage regenerator.

Page 16 of 83
Part of the coke is burned by combustion air from the air rings. This regenerator
operates in a counter current (air in at bottom and spent catalyst in at top) mode
which helps prevent catalyst overheating and NOx formation.
First stage regenerator total combustion air is controlled to limit the temperature
in the first stage to maximum 730°C.
The partially regenerated catalyst flows down through the first stage regenerator
bed to the entrance of the air lift. Aeration is supplied in this area to ensure
smooth flow of catalyst to the lift.
Two stage cyclones separate entrained catalyst from the flue gas exiting the first
stage regenerator. At the exit of the regenerator, the flue gas pressure is sent to
the CO incinerator where CO incineration in the flue gas is accomplished.
A continuous catalyst withdrawal is necessary to maintain the catalyst quality in
the normal operating region.
Torch oil is used to heat the process to its operating temperature during start-up.

c. Second stage regenerator

The partially regenerated catalyst flows up the lift and enters the second stage
regenerator below the air ring. A distributor at the end of the lift provides
efficient distribution of catalyst and air from the lift. Catalyst is then completely
regenerated to less than about 0.03% carbon at more severe conditions than in
the first stage regenerator. Very little carbon monoxide is produced in the
second stage and excess oxygen is controlled by flow control of the second
regenerator combustion air for efficient and complete combustion. Because most
of the hydrogen in coke was removed in the first stage, very little water vapor is
produced in the second stage. This limits hydrothermal deactivation of the
catalyst as higher regeneration temperatures are experienced.
The first stage cyclone diplegs are external to the regenerator. The second stage
cyclones diplegs are entering the second stage regenerator though a
hemispherical top head. Catalyst recovered in the cyclones is returned to the
regenerator bed below the normal operating level by way of the diplegs.

Page 17 of 83
Aeration is supplied to the first stage cyclone diplegs to provide smooth
fluidized catalyst flow. The second stage regenerator pressure is controlled by a
flue gas double disc slide valve, through differential pressure between the first
and second stage regenerators.

6. Main Fractionation section

Purpose of Main Fractionation and Gas Recovery sections is to fractionate hot

vapor effluent coming from Reaction section into different products. Main
Fractionation consists in separation of product at different levels in the column.
Products are clarified Oil and LCO. Heat input by the hot feed from reaction
section has to be removed in the column. Pumparounds sections are used to
recover this heat input.

Page 18 of 83
Figure 8: A typical FCC main fractionator circuit (HCO=heavy cycle oil,
P/A=pumparound).

Intermediate products from the Main Fractionator are wet gas, overhead liquid
distillate (which is used for absorption in the Gas Recovery Section and further
processed) and sour water. Products from Gas Recovery unit, after recontacting
of gas and liquid coming from Main Fractionator, are Fuel Gas, LPG, full range
Gasoline and sour water. This section consists in absorption and fractionation

Page 19 of 83
processes. Heavy and Light Cracked Gasoline are products from the Gasoline
Splitter section.

Figure 9: A typical FCC gas plant (HPS=high-pressure separator)

Absorption takes place to absorb C3+ from Fuel Gas which is amine treated
before going to Network. Absorption is also used to increase C 3, C4 recovery for
LPG Cut. C2– and H2S are stripped to be removed from C3+ cut which is
fractionated into LPG Cut sent to FCC Propylene Recovery Unit.

7. RFCC R2R’s features

MTC system
The MTC (Mix feed Temperature Control) has been developed to further
improve the R2R process capabilities and performances with respect to the
cracking of increasingly heavier and more refractory feedstocks.
It is designed to face the two main challenges encountered when processing very
heavy, highly contaminated feeds, namely:
 Achieve a satisfactory vaporization of the feed so as to eliminate the
unnecessary coke production resulting from an incomplete vaporization.
Page 20 of 83
  To keep the desired heat balance while maintaining the conversion at the
optimum level.
The MTC system basically consists of the injection of an appropriate stream into
the riser, downstream of the feed injection point, at a location and under
conditions selected to obtain the desired quenching effect and the best yield
selectivity.
The optimum MTC flowrate depends on the feed quality (the heavier the feed
the higher the recycle flowrate) and on the desired yields structure.
MTC flowrate should be limited to 20% maximum of fresh feed.

Riser Outlet Separation System

The separation chambers are connected in their upper section to the riser and the
vapor + catalyst mixture follows a U turn downwards around an inlet baffle. The
vapor outlets to the collecting chambers are located underneath the inlet baffle in
the perpendicular direction of the incoming stream. This geometry provides
simultaneously both the centrifugal effect and the inertial effect which are key
factors for catalyst separation.
The catalyst separated is collected in a conical section prolongated by a dipleg
which achieves the necessary seal between the riser and the stripper vessel. The
vapor exiting the separation chamber is then mixed in the collecting chamber
with the steam evolved from the stripper vessel. The collecting chambers are
then connected to a center pipe collector which distributes the vapor to the
disengager cyclones for final catalyst separation.
This system achieves in a single device an extra-short time of vapor
disengagement of about 1 second (U turn, no vortex) and an efficient catalyst
separation in two chambers arranged in series. The main chamber which is the
separation chamber itself and the secondary chamber which is the vapor
collecting chamber providing an additional safety buffer volume for further
catalyst disengagement in case of catalyst carry-over.

Page 21 of 83
Two-stage regeneration
Residue cracking differs from gas oil cracking in that the feed contains more
asphaltenes, and is more hydrogen deficient. Coke make, which is usually
dictated by heat balance requirements can become a critical factor in the
operability of the process. Coke make increases as residual oil is added to the
feed and regenerator temperatures rise as more heat of combustion from the
coke is released. A conventional single stage regenerator is limited in regard to
cracking residues because of metallurgical limits within the regenerator vessel.
The R2R technology splits the regeneration into two stages. The first stage burns
part of the coke from the catalyst at mild conditions and completes the
regeneration in the second stage. Several benefits are realized as a result of this
configuration. Most of the hydrogen in coke is removed in the first stage at mild
conditions, i.e. a maximum operating temperature limited at 730°C (usually 650
to 690°C). The resulting water vapor is in contact with catalyst at lower
temperatures than those necessary for complete regeneration, leading to much
less catalyst deactivation. The relatively dry atmosphere of the second
regenerator allows raising the temperature while the regeneration is completed.
Thus the R2R two stage regeneration results in less hydrothermal deactivation of
the catalyst.
As the residue content of the feed increases to higher levels, the coke make
increases and the heat balance of the unit must be adjusted to maintain catalyst
circulation rates and conversion. In the two stage regenerator configuration,
potential energy, in the form of carbon monoxide, is rejected from the
regenerator system by increasing the amount of total coke burned in the first
stage regenerator.
Complete regeneration of the catalyst is accomplished in the second stage
regenerator. Excess oxygen is maintained to ensure complete regeneration to
essentially carbon free catalyst. Since the second stage regenerator cyclones are
external to the vessel and lined with refractory, the metallurgical limitations
previously encountered with residue feeds are circumvented (the maximum

Page 22 of 83
operating temperature can reach 810°C relatively to the catalyst while the
mechanical design is 840°C).

Special valves
Two types of catalyst valves are used in the R2R process: slide valves and plug
valve. Slide valves are carefully designed with abrasion resistant protection for
improving the valves reliability. Internal insulation allows to use carbon steel for
the body of the valves.
The valves are provided with appropriate purges, on the stem and body, to clear
the valves from any catalyst particles. Internal inlet shapes are designed to
provide smooth operation.
The plug valve design includes a stuffing box, an insulation sleeve and an air
purge which provides efficient protection against catalyst blockage. Thrust
limiter and seat and plug angles have been optimized to overcome the thermal
expansion of the air lift.

Riser wye steam ring

Its function is to straighten out the catalyst flow pattern as it makes the transition
at the "wye" from its downward flowing, partially deaerated state, to an upward
flowing, evenly aerated state.
In aerating the catalyst, this ring also provides the reverse seal needed to protect
against oil flow reversals. The vertical column of catalyst below the feed nozzles
provides a seal against such upsets. This ring is designed for a normal flowrate
that should be maintained at all times when the catalyst is circulating.

8. Control philosophy of the process

The most critical control loop on the converter is the riser outlet temperature.
The temperature should be controlled within 1°C of the set point. A
thermocouple located near the outlet of the riser measures the reaction

Page 23 of 83
temperature. The temperature is a function of the amount of catalyst admitted to
the riser by the regenerated catalyst slide valve. Reaction temperature is basic to
the conversion of RFCC feedstocks.
To enhance fresh feed vaporization and ultimately the product yields, mixed
temperature control (MTC) technology is used. MTC flowrate is set to maintain
the desired temperature at the fresh feed injection point.
The spent catalyst slide valve controls the catalyst stripper level by modulating
the flow of spent catalyst from the catalyst stripper to the first stage regenerator.
The catalyst level in the stripper is measured by differential pressure instruments
and sends a signal to the controller which sets the position of the slide valve. A
minimum level is required in the catalyst stripper to ensure good stripping of
hydrocarbons from the catalyst and a seal at the diplegs outlet of the Riser
Separation System.
The level in the first stage regenerator is measured by a differential pressure
instrument and signals a level controller which resets the position of the plug
valve. The plug valve modulates the flow of catalyst from the first stage
regenerator into the lift to the second stage regenerator. A minimum level must
be maintained in the first stage regenerator to ensure good regeneration and to
seal the cyclone diplegs. A low level in the regenerator may unseal the diplegs
which could result in backflow of flue gas up the diplegs and loss of catalyst
fines with the flue gas. A high level in the first stage regenerator can also result
in catalyst carryover from the cyclones because of higher entrainment from the
bed and re entrainment from the cyclone dust bowls.
During normal operation, the second stage regenerator level is controlled and the
first stage regenerator follows the unit inventory. This level is monitored and
adjusted by continuous withdrawal as the level builds through catalyst addition.
The level in the second stage regenerator must be held within certain limits for
the same reasons as stated above for the first stage regenerator.
The regenerators pressures are controlled by flue gas slide valves which throttle
the flow of flue gas from the vessels. The pressure of the first stage regenerator

Page 24 of 83
is directly controlled by the first regenerator flue gas slide valve. The differential
pressure between the first stage and second stage regenerators is controlled by
the second regenerator flue gas slide valve.
This differential pressure is set to provide adequate pressure drop across the plug
valve for stable control of the regenerators levels. The disengager pressure rides
on the main fractionator pressure which is controlled at the main fractionator
overhead receiver. The objective in operation is usually to set the main
fractionator pressure to a base level for efficient operation and the regenerator
vessel pressures to result in approximately equal differentials across the catalyst
slide valves.

Override controls are provided for action in low slide valve differential upsets
which close slide valves to prevent dangerous reverse flow. Normal catalyst
slide valve differential is around 30 to 100 kPa to provide stable control. Slide
valve differentials above 100 kPa are to be avoided because they may cause
valve erosion. Negative differentials should never be permitted and the PDIC
controls should be set to override the main controllers at 10 kPa. Smooth
catalyst circulation is paramount for successful operation of R2R and is
achieved by proper catalyst aeration and fluidization in the transfer lines, as well
as control of the unit pressure balance.
An emergency shutdown circuit will close the catalyst slide valves automatically
upon loss of feed and loss of combustion air.

9. Process variables

a. General

The R2R unit will operate in a stable manner over a wide range of conditions.
The operating mode chosen will affect the distribution and quality of products
produced. The unit operator chooses a mode of operation to maximize desirable
products for different feedstocks or to conform to the demands and limitations of
Page 25 of 83
the refining complex. Careful attention to unit performance and process
variables will result in an operation of maximum profitability with few problems
and upsets.
The quality of the reaction which determines the products quality and quantity
depends on the :
 riser outlet temperature,
 reactor pressure,
 catalyst activity.
The quality of the regeneration depends on the :
 regenerators air balance,
 regenerators temperatures,
 regenerators residence time,
 regenerators fluidization velocities.
The quality of stripping will affect the yields pattern through the delta-coke
which controls the regeneration conditions and further the catalyst circulation.
Reaction and regeneration are interdependent through the overall heat balance.
The heat balance depends on the following variables :
 feedstock properties,
 feed temperature,
 coke production / delta-coke / catalyst to oil ratio.
Heat is carried from the regeneration zone to the reaction zone by the catalyst
circulation. A proper pressure balance is required to ensure a smooth catalyst
circulation.

b. Riser Outlet Temperature

The operating temperature of the riser outlet will normally be set to achieve the
desired degree of conversion. Typically, a temperature of 510°C will correspond
to a maximum distillate operation, a temperature of 525 to 530°C to a maximum
gasoline operation and 550°C to a maximum propylene operation. This
temperature is controlled by the regenerated catalyst slide valve position

Page 26 of 83
allowing more or less hot regenerated catalyst to contact and mix with the
incoming feed.
When the hot catalyst contacts the feed in the bottom of the riser the oil
vaporizes almost instantaneously. The homogeneous mixture of oil and catalyst
reaches a temperature approximately 30 to 40°C higher than the top of the riser.
The initial temperature shock causes thermal cracking, while catalytic cracking
begins after the oil is converted to vapor, and the molecules contact the active
catalyst sites. As the catalytic cracking progresses and the catalyst/oil mixture
flows up the riser, the temperature drops since the heat of cracking is
endothermic. At the top of riser the moles of products are 3.5 - 5.0 times more
than the moles of fresh feed.
The riser temperature has a complex interrelation with the other parameters in
cracking. An increase in the riser temperature is generally accompanied by:
 an increase in conversion,
 an increase in dry gas yield,
 an increase in LPG production,
 an increase or a decrease (over cracking) in gasoline production
depending on
 the reaction severity,
 an increase in gasoline octane,
 a decrease in LCO and slurry yield,
 a slight increase in coke production.
These are only trends. A quantitative definition of the changes requires heat and
material balances as well as yield correlations. The catalyst quality also affects
the extent of the changes.

c. Regenerators temperature

a) Dense phase temperatures

First stage

Page 27 of 83
The first stage regenerator dense phase temperature is a function of the riser
temperature, the amount of coke burned, and the catalyst circulation rate. The
temperature is controlled by varying the air flow to the vessel. The air rate
should be adjusted so as not to exceed a temperature of 730°C.

Second stage
The second stage regenerator dense bed temperature is also a function of the
amount of coke burned and the catalyst circulation rate. The normal operating
temperature is around 100°C higher than the first stage regenerator dense bed.
Control of this temperature is not independent because the air rate to the second
stage regenerator is adjusted to achieve complete CO combustion (i.e. 1 - 3 %
mole O2 in flue gas) under normal operation.
In any case, it is important to keep a minimum temperature in the catalyst dense
bed, which is required for a proper combustion (around 680°C).

b) Dilute phase temperatures

The dilute phase temperature will normally run within 10°C of its corresponding
dense phase temperature. The catalyst used, while not promoted, has properties
which enhance combustion in the dense bed. These properties, together with
proper inventory of catalyst in the dense phase, ensure that all oxygen required
for combustion will be burnt away in the dense bed, removing the possibility of
after burning.

d. Regenerators residence time

The catalyst residence time in the regenerators is a key parameter for the
regeneration quality. The catalyst levels in the two regenerators are optimized
depending on the regeneration temperatures to achieve the required inventories.
Typically, a total residence time around 6 minutes for the two regenerators is
sufficient to achieve a carbon on regenerated catalyst leaving the second

Page 28 of 83
regenerator of less than 0.03 wt %, which is considered as a target value for a
good regeneration.
During normal operation, care must be taken to keep the catalyst levels at their
normal levels. Low catalyst levels will affect the regeneration quality, whereas
high catalyst levels will affect the catalyst entrainment to the cyclones and
consequently will increase the catalyst losses.

e. Regenerators velocities

The quality of the combustion depends on the catalyst fluidization. A good

mixing between catalyst and air is required for the combustion and therefore
turbulent flow is desired in the regenerators dense bed. The corresponding
superficial velocities are in the range of 1 - 1.3 m/s. A minimum velocity of 0.5
m/s should be kept in any case (especially for reduced capacity operation).

f. Stripper operation

The removal of any light hydrocarbons which may remain with the catalyst after
disengagement in the disengager is accomplished by steam injection into the
stripper dense bed via different steam rings. The amount of steam is adjusted
depending on the catalyst circulation rate and feed rate. The flowrate is adjusted
on the main and pre-stripping steam rings, keeping the flowrate on the lower
rings constant. Stripping efficiency is measured by the hydrogen content on
coke. Stripping is considered as efficient when hydrogen content in coke is
around 6 wt %.

g. Regenerators air balance

The reduction of the coke on the catalyst to less than 0.03 wt % requires a
predetermined amount of air regardless of how the air is distributed between the
two regenerators. In a unit with one regenerator the required air rate would be
the sum of what the two regenerators require as long as coke yield and the
average flue gas composition are the same. Having two regenerators provides a
Page 29 of 83
flexibility of how to split the total air required. When certain constraints, such as
the maximum regenerator bed temperature and desirable flue gas composition
are imposed, the freedom to arbitrarily split the required air decreases. However,
there is a great advantage in assigning a specific air rate to each regenerator.
The first stage regenerator design temperature is 770°C, therefore the coke that
should be burned must be controlled so that this temperature limits is not
exceeded. The combustion of carbon monoxide is more rapid as the temperature
increases over 650°C. This results in lower carbon monoxide concentration and
greater air requirement as the regenerator bed temperature rises. When coke
burns to form carbon monoxide and steam the heat generated is about half as
much as when the same amount of coke burns to form carbon dioxide and
steam. The higher temperature generated by greater carbon dioxide production
further facilitates the burning of carbon monoxide. If the air rate is limited, then
the conversion of CO to CO2 is controlled and the regenerator bed temperature is
kept lower. The objective is to determine an air rate to the first stage regenerator
that will limit the bed temperature and yields carbon monoxide in the flue gas.
The objective in the second stage regenerator is to burn all the remaining coke
completely to carbon dioxide. The bed temperature is allowed to rise around
810°C (mechanical design : 840°C). The air rate is adjusted to give 1 - 3 mole %
of excess oxygen in the flue gas which ensures that the carbon monoxide
concentration in the flue gas is less than 0.05 mole %.
The above split of the required air also has the advantage of minimizing catalyst
deactivation. Most of the hydrogen contained in the coke burns in the first stage
regenerator. The steam thus generated is at a lower temperature and causes less
catalyst deactivation. Since only a small part of the total hydrogen contained in
the coke is burnt in the second stage regenerator, the steam concentration is
reduced in the higher temperature environment. The flue gas from each
regenerator is routed separately which segregates the flue gases according to
their carbon monoxide concentration.

Page 30 of 83
III. Chemical reactions and catalysts

1. Objective of this section

The objective is to provide the operation team with a better understanding of the
FCC chemistry and operational philosophy.

This section describes the following:

 The various chemical reactions involved in the process as well as the

effect of the operating conditions.
 The catalyst characteristics.
 The catalysis mechanism.
 The catalyst contaminants.
 The process variables.

2. Thermodynamics and kinetics

For any chemical reaction the thermodynamics dictates the possibility of its
occurrence and the amount of products and unconverted reactants. In fact, some
reactions are 100% completed i.e. all the reactants are converted into products.
Others are in equilibrium i.e. part of the reactants only are converted. The
amount of products and reactants at equilibrium depends upon the operating
conditions and is dictated by the thermodynamics.

Note that the thermodynamics does not mention the time required to reach the
equilibrium or the full completion of a reaction.

In other words, thermodynamics dictates the ultimate equilibrium composition

assuming the time is infinite. Kinetics enables to forecast the composition after a
finite time. Since time is always limited, when reactions are competing, kinetics
is generally predominant.

Page 31 of 83
A catalyst is a substance (might be solid, liquid, gas), usually used in small
amounts relative to the reactants, that speeds up (or modifies) the rate of a
chemical reaction, but is not consumed by the reaction; hence a catalyst can be
recovered chemically unchanged at the end of the reaction it has been used to
speed up, or catalyze.

A catalyst is not consumed but can be deactivated either by impurities in the

feed or by some of the products of the chemical reactions involved, resulting in
coke deposit on the catalyst.

3. Catalyst activity, selectivity, stability

The main characteristics of a catalyst other than its physical and mechanical
properties are:

The activity which is the catalyst ability to increase the rate of the reactions
involved. For FCC it is measured by the conversion performed by the catalyst
with a specific feed and in specific operating conditions.

The selectivity expresses the catalyst ability to favour desirable reactions rather
than others. It is measured by the yield of desired product.

The stability characterizes the change with time of the catalyst performance
(i.e., activity, selectivity) when operating conditions and feed are stable. It is
chiefly the polymers or coke deposit which affects stability, because it decreases
the metal contact area. Traces of metals in the feed also adversely affect
stability.

4. Types of reactions

The reactions involved in FCC process are of two types: thermal cracking and
catalytic cracking.

Page 32 of 83
Table 6.1: Important reactions occurring in the FCC Unit.

a. Thermal cracking

Chemical cracking reactions are characterized by the breaking of chemical

bonds within the molecules of hydrocarbons or other compounds contained in
petroleum fractions. They mainly involve the breaking of carbon-carbon bonds
(–C–C–), carbon-sulfur bond (–C–S–) and also, although to a lesser extent,
carbon-hydrogen bonds (–C–H).

These reactions are the result of the heating effect, whether or not in the
presence of a catalyst. However, cracking of a petroleum cut is not confined to
these breakages of bonds, since both shorter and longer molecules than those
originally present in the treated cut are simultaneously obtained. This can be
explained by a rather complex mechanism that occurs after the first cracking
reactions. To simplify matters, one can distinguish primary cracking reactions
and progress reactions.

Primary cracking reactions

Page 33 of 83
This type of reaction concerns various molecules in the feed. In the case of the
main groups of hydrocarbons, paraffins, naphthenes and aromatics, the
following conversions are mostly observed.

Starting with olefins gives a product with higher hydrogen content (paraffin) and
another with lower hydrogen content (diolefin, aromatic). Additional reactions
between certain intermediate products, mainly involve progressive attachment of
diolefins to aromatics, resulting in condensed and highly aromatic polycyclic
heavy products; this is how heavy fuel oils and coke appear.

b. Catalyst cracking

The use of a catalyst in chemical reactions during hydrocarbon cracking makes

a significant change in the chemical conversions, affecting reaction mechanisms
and selectively speeding up some of them. Catalytic cracking differs from
thermal cracking in the following main ways:

 Lower yields of C2 gases, since catalysts theoretically do not allow the

formation of short molecules with fewer than 3 carbon atoms.

 Higher yields of gasoline, which also have better stability and a higher
octane number.

 Smaller coke production for the same feed processing, thereby

indicating better catalyst selectivity.

 Products of different chemical composition: few olefins, many isomerized

structures, high aromatic contents, etc.

Since their discovery, catalytic cracking catalysts have changed a great deal, and
now consist mainly of mixtures of amorphous silica-alumina and crystallised
silica-alumina, the latter better known as zeolite or molecular sieves.

Page 34 of 83
They take the form of a solid in powder form, the catalyst activity of which
during cracking is related to the acid properties of their surface.

Acid catalyst cracking

The cracking mechanism involving catalyst acidity is related to the appearance

of intermediate ionic compounds, which possess very specific properties
governing the progress of the cracking process. Such compounds usually appear
during adsorption of an olefin produced by primary cracking on the acid
catalyst.

Whereas thermal cracking of a heavy petroleum cut involves a fairly complex

mechanism, which rather unselectively generates unsaturated lighter products,
and heavier products, than the original feed (in particular producing a great deal
of gas, rather unstable gasolines and a high proportion of coke), cracking
reactions can be improved to move product yields and qualities in the desired
direction, by using a catalyst. This explains the great benefit of catalytic
cracking. Indeed, the catalyst gives one H+ acid ion to the olefin, thereby partly
saturating the double bond. The olefin is thereupon converted to a hydrocarbon
cation, which is conducive to the following conversions:

Structural change in the hydrocarbon chain in the direction of branched-chain

isomerized structures. These new forms represent greater stability for the
hydrocarbon cation. This stage explains the high presence of such molecules in
catalytic cracking products.

The structural change takes place spontaneously and leads to this more stable
carbocation: this is isomerization.

Secondary cracking resulting in the formation of a smaller olefin and new

hydrocarbon cation with a shorter chain, which may then undergo the same
process as the previous one. This corresponds to the standard progression of
cracking, resulting in the formation of light products. Hydrocarbon cations,

Page 35 of 83
however, very strongly discourage the formation of short chains with fewer than
3 carbon atoms, explaining the lower yields of C2- gases obtained by catalytic
cracking

Hydrogen transfer reactions

Hydrogen transfer reactions from olefins, which are highly encouraged by

zeolite catalysts, result in the parallel formation of:

 Hydrocarbons with more hydrogen, benefiting from the transfer; these are
therefore mainly paraffins.

 Compounds with less hydrogen, such as aromatics.

Hydrogen transfer is influenced by:

 Feed quality.

 Riser Outlet Temperature.

 Cat to oil ratio.

 RE2O3 on catalyst.

 Catalyst composition.

More hydrogen transfer means:

 More coke.

 Less LPG.

 More gasoline.

 Less product olefins.

 Lower RON.

Page 36 of 83
A glance at the general scheme of the parallel and consecutive cracking
reactions shows that all the products (gas, gasoline, LCO and coke) are primary.
The LCO usually results from a single cracking operation of the large molecules
in the feed. Most of the gasoline and part of the gas also result from primary
cracking. However, a large share of the gases, especially C3 and C4, result from
the secondary cracking of gasoline and LCO.

A small part of the coke appears to form instantaneously by thermal

dehydrogenation of large probably naphteno-aromatic molecules. C1 and C2
hydrocarbons are essentially produced by thermal cracking. This set of reactions
involved can be divided into three categories

Table 6.2: Comparison of products of thermal and catalytic cracking.

5. Desired reactions

a. Cracking reactions

Hydrocarbon reactivities

A very rough order of reactivity of the main hydrocarbon categories is:

olefins > alkylaromatics > alkylnaphtenes and isoparaffins >

n-paraffins and naphtenic cycle >> aromatic ring (very stable)

Page 37 of 83
In all cases, cracking results from the scission of a carbocation (or carbenium
ion) located in the neighborhood of an acid site on the catalyst. The more easily
the carbocation is formed and occupies the catalyst surface, the more it
undergoes scission. This helps to define an additional reactivity rule valid within
the first four categories above:

 Crackability increases with size (or number of carbon atoms) and the
degree of branching of the hydrocarbon category.

How are hydrocarbons cracked ?

A few simple rules can be set forth:

 Heavy linear alkanes and linear alkenes are always isomerized before
cracking. The cracking products of the molecules hence contain a
majority of 1 to 2 methyl branches (or, to a lesser degree, ethyl).

 Cracking by scission at of these molecules mainly occurs in mid-chain

and practically never at less than three carbon atoms from the end. C 1 and
C2 are formed by a thermal homolytic mechanism different from -
scission.

 The alkyl chains fixed to a cycle are cut at the edge of the ring if the cycle
is aromatic, and at any point located at three or more carbon atoms from
its end if the cycle is naphthenic.

b. Isomerization reaction

Given the short residence times employed in cracking, this reaction only affects
the olefins with roughly more than four or five carbon atoms and the heavy
paraffins of the feed. The light paraffins produced by a first cracking
subsequently do not have the time to isomerize. The C4 paraffins are thus
present in proportions quite different from those of thermodynamic equilibrium.
C4 olefins are much closer to equilibrium but never reach it.

Page 38 of 83
6. Reactions to be limited (but not eliminated)

a. Hydrogen transfer

This bimolecular reaction consists in the transfer of a hydrogen molecule from

one naphthene or olefin to another olefin, by a heterolytic acid mechanism. It
leads essentially to the formation of saturated hydrocarbons, mainly paraffins, as
well as aromatic compounds (mono and essentially polyaromatics).
Polynaphthenoaromatics and other heavy coke precursors are very reactive
molecules with respect to hydrogen transfer.

This reaction plays a very important role in the quality of the gasoline obtained
and the formation of coke. Hence considerable hydrogen transfer decreases the
olefinicity of the gasoline and increases the coke production in the installation. It
simultaneously improves the stability of the gasoline (i.e. its ability to withstand
overcracking) by lowering the reactive molecule content.

b. Condensation reactions

The Diels-Alder cyclo-addition is the main reaction. It consists in combining a

linear or cyclic olefin with a linear or cyclic diolefin, leading to more condensed
and partly unsaturated cyclic structures. These are transformed into aromatics
(mono or poly) by hydrogen transfer, or continue to undergo cyclo-addition. The
final stage of the successive cyclo-addition and hydrogen transfer reactions is
the deposition of very heavy condensed polyaromatics constituting coke on the
catalyst.

7. Undesirable reactions to be reduced to the minimum

a. Hydrogen formation

The dehydrogenation of condensed and partly unsaturated cyclic molecules by

metallic contaminants, in particular Ni and V, is responsible for most of the

Page 39 of 83
production of hydrogen (Ni and V) and a substantial share of coke (mainly Ni).
The production of methane is virtually unaffected by the metal content, therefore
this effect is monitored through the H2/methane ratio.

b. C1 and C2 hydrocarbons

These are mainly due to non-selective thermal cracking. Hence their production
is highly sensitive to the riser temperature, and residence time.

8. Conversion selectivity of various hydrocarbon families

The two families most selectively converted to gasoline are alkylmonoaromatics

and condensed polynaphthenes (mainly with two cycles).

Among the simple saturated hydrocarbons, the order of gasoline selectivities is:

naphthene > isoparaffins > normal paraffins

Polyaromatics, especially with four and five rings, are not converted to gasoline
but are excellent coke precursors.

9. Definitions of conversion

a. Conversion

The conversion rate is defined by the following ratio:

products converted of rate Flow

Conversion = x 100
feed fresh of rate Flow

Depending on whether flow rates are expressed by weight or volume, the

conversion rate is a weight percentage or a volume percentage.

Converted and non-converted products are as follows:

Page 40 of 83
Converted products:

 Dry gas

 C3 cut

 C4 cut

 gasoline and swing cut (C5 – 190oC cut)

 Coke

Unconverted products:

 LCO (190oC – 350oC cut)

 Slurry (350 oC + cut)

b. Industrial conversion rate

It is not easy to measure the flow rates of converted products. The flow rate of
coke is not directly measurable, and volume flow rates of dry gas are
meaningless. In contrast, it is possible to measure flow rates of non-converted
products.

The flow rate of converted product is obtained by subtracting the flow rate of
non-converted products from the flow rate of fresh feed:

Converted products flow rate = fresh feed flow rate – non converted products
flow rates

rates flow products converted non − rate flow feed Fresh

Industrial conversion rate = x
rate flow feed Fresh

100

Page 41 of 83
 c. Adjusted conversion rate

It can be understood that the flow rate of converted products depends on the
cutpoint between gasoline and LCO. This cutpoint is conventionally defined by
the ASTM FBP of the gasoline at a constant cutpoint of 221 oC. If the initial
point of the fresh feed is less than 221 oC, the flow rate of fresh feed is adjusted
as follows:

Adjusted fresh feed flow rate = fresh feed flow rate - fraction with boiling point
lower than 221 deg. C

IV. Catalyst

1. Catalyst characteristics

The performance of the reaction system and the regeneration is highly

dependent on the kinetic properties of the catalyst. Performance of the catalyst
circulation system is also dependent on the physical properties of the catalyst.

The FCC catalyst is a solid complex composite acid. Modern FCC catalysts
consist of several ingredients, such as:

Zeolites are microporous crystalline aluminium silicates. Procedures exist for

the synthesis of zeolites with a structure similar to known minerals as well as
zeolites without a natural counterpart, such as for example the ZSM-5 that is
used in octane boosting additives. For FCC catalysts mainly the Y-type
(Faujasite) zeolite is of interest, with a silica to alumina ratio of about 5.
Page 42 of 83
The framework of the zeolite is composed of silicon, aluminium and oxygen
atoms, forming a rigid structure of tetrahedra, linked together in cubo-octahedra.
The silicon atoms in the SiO4 tetrahedra are partly replaced with aluminium
atoms and a corresponding number of charge-compensating sodium ion. These
sodium ions are rather mobile and can be exchanged with NH4 + or rare earth
(RE3+). This ion exchange is essential for zeolites used as cracking catalysts,
because it generates active sites for cracking.

Figure 10: Geometry of USY and REY zeolites

RE-exchanged zeolites (REY and REUSY) are favored in cracking catalysts

because they show an excellent stability and activity. However, as the rare earth
enhances hydrogen transfer reactions, the product olefinicity and research octane
numbers are depressed. During the eighties the rare earth contents were reduced
somewhat, while other techniques were developed to compensate for the loss in
zeolite stability and activity. The openings of the zeolite pores (0.8 nm) are too
narrow for the average feedstock molecules, so that precracking is required.
Active « matrices » are applied to give the catalyst a proper pore size and
activity distribution, and a good accessibility to the highly active zeolite sites.
These matrices also protect the zeolite from catalyst poisons, such as vanadium

Page 43 of 83
and sodium. The pore size and activity distribution of the catalyst should be
tailored to the feedstock quality and the required product yields.

Figure 11: Comparison of Y-faujasite and ZSM-5 zeolites

The average particle size is around 70 microns in the fresh catalyst and slightly
higher in the equilibrium, depending on the efficiency of the cyclones in the
disengager/ stripper and regenerator vessels.

Page 44 of 83
Figure 12: PSD of a typical FCC catalyst.

The performance of FCC catalysts is the result of the concentration, catalytic

activity of the functional ingredients and the intraparticle accessibility of the
catalyst. The catalytic activity is determined by e.g., the type of zeolite, the rare
earth concentration, the type of matrix and the stability of the active
components. The accessibility depends not only on the matrix and the zeolite but
also on the non-active components in the catalyst.

2. FCC catalyst fines

FCC catalyst fines are in general not suitable for re-use in the RFCC unit and
should therefore be disposed of as spent catalyst. However certain qualities of
fines are suitable as flow improver for units with circulation problems.

3. Catalyst mechanism

It is the selection of the right catalyst together with the operating conditions
(mainly the temperature) which determines the rate of the various reactions and
enables to meet the required selectivity. The performance of RFCC catalysts is
the result of the concentration and catalytic activity of the functional ingredients
and the intraparticle accessibility of the catalyst. The catalytic activity is
determined by e.g., the type of zeolite, the rare earth concentration, the type of
matrix and the stability of the active components.

Page 45 of 83
The accessibility depends not only on the matrix and the zeolite but also on the
non-active components in the catalyst. The catalyst support must meet the
following criteria:

 improve the metal dispersion within its structure as well as the reactants
diffusion to the active sites,

 be mechanically resistant.

 withstand steam, air treatments at temperature higher than 750 deg. C,

 and keep an excellent behaviour towards thermal shocks and steam during
the course of regenerations.

Catalyst used in residue cracking operations should be double calcinated to

enhance thermal stability. Particle size distribution should be examined on the
fresh catalyst. The zero to 20 micron fraction should be low to minimize losses
through the cyclones. The cyclone system is designed to retain enough of the
equilibrium 0-40 micron fraction to maintain good fluidization properties.

4. Catalyst contaminants

a. Definition

Three different types of contaminants are considered: inhibitors or activity

moderators, temporary poisons, permanent poisons.

b. Inhibitors or activity moderators

Inhibitors are compounds which compete with the reactants for the catalyst
active surface resulting in a reduction of the available active surface. They
adsorb strongly on the catalyst metal but this adsorption is perfectly reversible.

Page 46 of 83
Carbon on catalysts is a typical inhibitor. In order to monitor the regenerator
efficiency of the RFCC unit, the remaining carbon on regenerated catalyst is
measured. The carbon on catalyst flowing from the stripper into the regenerator
(spent catalyst) can be measured as well, allowing direct analysis of the delta
coke level.

The carbon is converted to carbon dioxide, which is analyzed using an infrared

detector. Apart from measuring the carbon on regenerated catalyst (CRC), the
carbon content can be estimated by the refiner by comparing the color of the
equilibrium catalyst with those of reference samples taken from the same unit.

A high carbon content does not affect the measured activity but results in a loss
of effective activity of the catalyst flowing into the reactor riser (typically 1-2 wt
% MAT per 0.1 wt % CRC). As a consequence the unit conversion and
selectivities may change. An improved air and/or catalyst distribution results in
a reduced carbon content. The efficiency of carbon removal also benefits from
increasing the dense bed temperature for units operating in a partial combustion
mode.

c. Temporary poisons

Temporary poisons are contaminants which are so strongly adsorbed that they
accumulate over the active surface. Removal of poisons and activity recovery
are obtained only by using specific procedures.

Antimony (Sb)

Antimony is only present on equilibrium catalyst if an antimony passivator is

used to reduce nickel activity. Approximately 30-50% of the value of the feed
nickel content is sufficient to reduce hydrogen yields to an acceptable level.

Miscellaneous

Page 47 of 83
Other metals may also affect catalyst performance. For example, iron and
copper can increase the hydrogen make while calcium and magnesium may
affect catalyst activity or stability. Usually these metals are not present in such a
concentration that a significant impact on the performance is noticed.

The equilibrium catalyst may also contain minor amounts of non-metallic oxides
from e.g., S, P, Cl and elements like Ti, originating from the kaolin or a metal
trap. The balance usually is silica.

d. Permanent poisons

Permanent poisons are not removable by procedures available on site, i.e.

usually steam-and- air decoking. The catalyst must be dumped and replaced by a
fresh load. Nickel, vanadium and sodium are typical permanent poisons.

Nickel (Ni)

Nickel is introduced with the feed and deposited on the equilibrium catalyst.
Nickel is not mobile under normal regeneration conditions and acts as a
dehydrogenation catalyst. In FCC nickel enhances non-selective cracking
reactions, particularly those producing more hydrogen and coke.

Vanadium (V)

Vanadium is introduced with the feed and is deposited on the equilibrium

catalyst. Under regenerator conditions, vanadium migrates and is able to enter
the fresh catalyst and destroy the zeolite. As a consequence, catalyst activity and
conversion suffer. For conventional catalysts (a typical rule is that 2 points
catalyst activity are lost per 1000 ppm vanadium at constant unit conditions and
catalyst consumption). As with sodium, the deactivation rate strongly depends
on the highest temperature and the vapor pressure of water in the regenerator.
The damaging form of vanadium is the fully oxidized V2O5 form which

Page 48 of 83
requires water to form a vanadium acid, destroying the zeolite and hence
catalyst activity.

Vanadium in the feed and deposited on the catalyst will promote collapse of the
zeolite structure and loss of active surface area at the elevated regenerator
temperature.

Sodium (Na)

Small amounts of sodium (typically 0.1-0.4 wt %) are present in the fresh

catalyst. Sodium can also be introduced by the feed, especially if heavy cuts is
processed. In case of high sodium levels on equilibrium catalyst, the feed quality
should be checked. Malfunctioning of crude desalters or processing of imported
feeds contaminated with seawater are the typical causes of high sodium levels in
most cases.

Sodium is a catalyst poison which neutralizes acid sites and destroys the zeolite.
Approximately 6 points MAT activity are lost per wt % sodium originating from
the FCC feed. At high sodium levels the catalyst is more sensitive to high
temperature, due to increased rates of sintering, and surface destruction. Feed
sodium should be kept imperatively below 2 wt ppm.

5. Catalyst regeneration

The catalyst that leaves the stripper usually contains up to 1 wt % coke with a
C/H ratio of about unity. The main objectives of regeneration are to burn off the
coke from the catalyst in order to restore its activity and to maintain the heat
balance of the unit. The heat generated by the major combustion reactions is
shown in table.

Page 49 of 83
The total heat generated largely depends on the hydrogen content of the coke,
the level of CO combustion and the coke yield. Depending on the stripper
conditions and the strippability of the catalyst, the H content of the coke can
vary from 6 to 8 wt %.

It should also be noted that, in addition to carbon and hydrogen, coke contains
sulphur and nitrogen. These two impurities originate from sulphur and nitrogen
compounds in the feed which combine with the coke during the cracking
process. During combustion in the regenerator, the sulphur and nitrogen produce
sulphur oxides (SO2 and SO3), and nitrogen oxides (NOx) in amounts that vary
with the regenerator operating conditions. All these compounds present in the
regenerator flue gas are atmospheric pollutants and the allowable amounts
released are governed by statutory regulations.

Normal Regeneration leads to a CRC of 0.1 % wt. The regeneration temperature

is a function of the coke concentration on spent catalyst. In the R2R process,
catalyst regeneration is achieved in two independent regenerators. The main
objectives are:

 Handling, when required, high coke laden catalyst without exceeding any
catalyst or metallurgical constraints,

 Regenerating the catalyst with a complete coke removal < 0.05 wt %,

 Favoring CO formation over CO2 to minimize the heat of coke

combustion. The heat released from the combustion of CO (carbon
Page 50 of 83
 monoxide) to CO2 (carbon dioxide) is about 2.5 times greater per
kilogram of carbon than that from the combustion of carbon to CO.

 Letting the regeneration temperature free to float and to equilibrate to the

level corresponding to the coke deposition on spent catalyst.

 Minimizing catalyst deactivation mainly with vanadium by preventing the

vanadium acid destruction of the zeolite embedded in the catalyst.

 Obtaining a hot catalyst in short contact time designs to favor rapid heat
transfer to feed by radiation. This is particularly advantageous in high
conversion operations.

In the first stage regenerator production of CO is encouraged, which limits the

heat release and thereby rejecting potential heat from the process to the flue
gases. The increase of the CO combustion not only results in a higher heat of
combustion, but also in a higher consumption of air per ton of coke.

V. Process variables

1. Introduction

The performance of the RFCC unit is determined by the relationship between

the feedstock, the process variables, the catalyst, process design and unit control.
Proper control of the RFCCU requires careful balancing of the many process
variables. The unit can then be optimized within certain limitations to provide
the best possible performance. These restrictions include available feedstock,
mechanical and operational equipment limitations, and environmental
constraints.

The operating costs and product values for a given yield distribution will be the
major factors in any optimization effort. The process variables are interrelated,

Page 51 of 83
so many of the effects may not be immediately obvious. The effects of key
process variables will be reviewed. Examples will be given illustrating the role
of adjusting these variables to achieve a particular process objective.

2. FCC reaction variables

Process variables can be classified in two categories:

 Independent variables,

 Dependent variables.

Table 1 lists some of the key independent variables in the RFCC process. Of
these, the most critical are the feed temperature, the Riser Outlet Temperature,
the catalyst activity, the contact time, and the feed quality. The dependent
variables are presented in Table 2 and include the conversion (and selectivities),
the regenerator temperature and the catalyst to oil ratio.

Page 52 of 83
Reaction and regeneration are interdependent through the overall heat balance.
Independent variables that affect the heat balance can be classified depending on
whether they affect the coke production or the delta coke on catalyst.
Independent variables that influence the heat balance.

The coke yield and delta coke are examples of dependent variables that can only
be changed indirectly, i.e. by manipulation of independent variables.

3. Feed temperature effects

Table 3 summarizes the effects of increasing the feed preheat temperature at a

constant Riser Outlet Temperature. If the feed temperature is increased, and no
change is made to the catalyst circulation rate, the riser temperature would rise
along with the feed temperature. This is because the catalyst is supplying the
heat needed to heat the feed from the feed preheat temperature to the riser
temperature. So if the feed temperature is increased, the catalyst circulation rate
will have to decrease to prevent the riser temperature from getting too high.
Since the riser temperature is automatically set and is controlled by the action of
the regenerated catalyst slide valve, the slide valve should close slightly as the
feed temperature is increased.

Page 53 of 83
This reduction in catalyst circulation rate (Cat/Oil) causes a number of other
changes to occur. First, conversion is usually reduced, while delta coke tends to
increase, because the feed coke is spread over fewer catalyst particles. As delta
coke increases, the regenerator temperature tends to increase, making it easier to
burn the coke off the catalyst. As a result, the concentration of coke on the
regenerated catalyst tends to decrease, while the concentration of coke on the
spent catalyst will increase slightly. It is important to remember the equation:

Coke Yield = (Cat/Oil) * Delta Coke

As the feed temperature increases, delta coke increases slightly, but (Cat/Oil) is
significantly reduced, so that the coke yield drops significantly as the feed
temperature is increased. Probably the most important effect of increasing feed
temperature is that it reduces the coke yield. Conversely, lowering the feed
temperature increases the coke yield. Thus, the required air rate drops as the
feed temperature is increased. For a given air rate, more barrels of feed can be
processed as the feed temperature is increased.

Following are reasons why feed temperature might be increased:

 If more feed needs to be processed at a constant air rate.

 If regenerator temperature needs to be increased.

 If more feed needs to be processed at a constant catalyst circulation rate.

 If regenerators gas velocities need to be decreased to reduce catalyst

losses from the regenerators.

 If catalyst circulation rate or air rate need to be reduced at a constant feed

rate.

 If conversion needs to be lowered.

Following are reasons why feed temperature might be reduced:

Page 54 of 83
  If conversion needs to be increased.

 If dry gas or wet gas rates need to be reduced.

 If regenerators temperature needs to be lowered.

 If catalyst circulation rate needs to be increased.

4. Riser Outlet Temperature effects

Table 4 summarizes the effects of increasing Riser Outlet Temperature at

constant feed temperature. The automatic controls on the regenerated catalyst
slide valve is set to control the disengager temperature. Usually, the disengager
temperature is a few degrees lower than the Riser Outlet Temperature due to
heat loss and the continuation of endothermic cracking reactions.

In most cases, the Riser Outlet Temperature is changed by changing the catalyst
circulation rate. If a ROT increase is called for, the regenerated catalyst slide
valve is opened slightly to allow more hot catalyst to mix with the feed. The
FCC conversion will increase for two reasons, the higher temperature and the
higher catalyst-to-oil ratio. The coke yields must increase to generate the extra
heat required to heat the oil to the higher reactor temperature. The delta coke
declines slightly because the coke is spread over more catalyst particles.

Page 55 of 83
The regenerator temperature increases, but by less than the reactor temperature
is increased. The carbon on the regenerated catalyst declines because the coke is
easier to burn as the regenerator temperature rises. However, since the coke
yield increases, more air is required to burn the coke.

The gasoline octane numbers will increase, as will the fuel gas and wet gas rates.
Gasoline yields will increase until the overcracking point, which usually occurs
at ROT between about 525 deg. C and 535 deg. C. Above the overcracking
point, the gasoline yields will decline as ROT is increased, while LPG and dry
gas (fuel gas) yields will increase rapidly.

In summary, Riser Outlet Temperature is the fastest and simplest method of

changing conversion in the RFCCU. Raising reaction temperature is appropriate
when conversion and/or gasoline octane numbers must be increased and:

 Wet gas and dry gas rates are not at their maximum.

 Catalyst circulation rate can be increased.

 Slide valve pressure differentials are above minimum.

 Air rate can be increased.

 Regenerator temperatures are not at maximum limits.

Other methods of increasing conversion (such as increasing catalyst activity)

may be more appropriate if:

 Gas compressor capacity is at a maximum.

 Catalyst circulation rate is limited.

 Air blower capacity is limited.

 Feed rate must be increased.

Page 56 of 83
The choice of the Riser Outlet Temperature will depend on many factors, and
optimizing the reaction temperature is often a difficult and complicated
procedure that requires a computer model to simulate various effects. The above
guidelines, however, can be used to at least indicate the proper direction in
which the reaction temperature should be moved.

5. Effects of contact time

Table 5 summarizes the effects of contact time. Contact time is defined as the
length of time while the catalyst and the oil vapor are together in the riser. It
depends on the feed rate, the steam rate and both the length and the diameter of
the riser. Contact time can be changed by:

 Changing the feed rate.

 Changing the dispersion steam rate.

 Changing the length and/or the diameter of the riser (in case of revamp).

 Changing the location of the feed injectors (in case of revamp).

Page 57 of 83
  Changing the riser/reactor pressure.

The first four methods of changing contact time are obvious. Disengager/
stripper pressure affects contact time because it affects the volume that is
occupied by a given amount of vapor. Increasing the pressure by 10 kPa will
reduce the vapor volume (and will increase the contact time by approximately 5
%).

Reducing contact time will lower conversion because the oil molecules simply
have less time to react. The major benefit of reducing contact time, however, is
that the tendency to make coke, or the delta coke, is substantially reduced. As a
result, the regenerator temperature drops significantly and the amount of catalyst
required to heat the oil increases. Conversion becomes more selective to
gasoline, with less production of dry gas and coke.

Because refiners do not generally like to see conversion decrease, when contact
time is reduced, some other adjustment is usually made to recover any
conversion that may be lost. The adjustment may be an increase in Riser Outlet
Temperature or an increase in catalyst activity, depending on the other unit
limitations.

An important point is that the volume of vapor changes as conversion takes

place. The volume of vaporized products is about 3.5 to 4 times as great as the
volume of vaporized feed. As molecules crack, the volume is proportional to the
number of moles. As the volume increases, the velocity of flow up the riser also
increases. This is why the velocity at the end of the riser is about 20 m/s. For
reasons of simplicity, contact time may be calculated based on the volume of
vaporized products.

In summary, shortening contact time will improve product selectivity by

reducing the time available for undesirable coking and thermal reactions. The
result is lower delta coke, a lower regenerator temperature and a higher catalyst

Page 58 of 83
circulation rate, with less production of dry gas. These benefits are offset
somewhat by a lower conversion, which must be regained by other means, such
as higher catalyst activity or higher Riser Outlet Temperature.

6. Effects of catalyst activity

Table 6 summarizes the effects of FCC catalyst activity. The activity of the
RFCCU equilibrium catalyst is measured by the catalyst manufacturers in a
laboratory microactivity (MAT) test. The test indicates how active the catalyst is
for cracking a standard feed at standard conditions. Changes in the MAT activity
will reliably predict the activity of the catalyst in the refinery RFCCU.

The equilibrium catalyst activity is controlled by how much fresh catalyst is

added to the unit each day. As the fresh catalyst addition rate is increased, the
equilibrium catalyst activity will rise. Changes in the equilibrium catalyst
activity usually do not happen quickly. It may take several days to see noticeable
changes in activity after a change in catalyst addition rate is made. If, however, a
unit upset occurs and regenerators temperature or steam concentration increases
significantly, for even one or two hours, the activity of the entire catalyst
inventory can be reduced very rapidly.

Most importantly, as catalyst activity increases, conversion rises by about 0.5 to

0.7 vol. % on fresh feed for each 1 number increase in MAT activity. The
Page 59 of 83
conversion increases by less than the activity because the delta coke increases as
the catalyst gets more active. This causes regenerators temperature to increase,
which, in turn, causes the Cat/oil to be reduced, offsetting some of the
conversion gain.

Increasing catalyst activity also increases the yields of coke, wet gas and dry
gas, but these yields increase by smaller amounts than they would if the same
conversion increases were achieved by increasing riser temperature. Gasoline
octane numbers are not greatly affected by catalyst activity.

7. Effects of fresh feed quality

Table 7 shows the most important RFCC feed characteristics. The quality of the
RFCC feed is probably the single most important factor in determining the
yields that can be obtained from the RFCCU. The structure of the feed
molecules is critical not only in defining the reactions that can take place, but
also in determining the quality of the RFCC products. The following paragraphs
discuss the key effects of the key feed properties.

Page 60 of 83
API gravity, hydrocarbon type and boiling range

These properties are discussed as a group, because they are correlated. For a
given boiling range feed (for example, 370 deg. C to 540 deg. C), the API
gravity will increase as the feed becomes more paraffinic, and will decrease as
the feed becomes more aromatic. Also, for a constant feed composition, the API
will increase as the feed becomes lighter (lower boiling points) and will decrease
as the feed gets heavier.

Two key properties indicate how aromatic the feed will be. High values of the K
Factor and the Aniline Point indicate high paraffin concentrations, with lower
levels of aromatics. Low K Factors and Aniline Points indicate high levels of
aromatics. Feeds with K Factors above 12.0 and Aniline Points above 90 deg. C
are considered very paraffinic. Feeds with K Factors below 11.7 and Aniline
Points below 77 deg. C are considered very aromatic.

The more aromatic feeds give lower conversions and lower yields of gasoline,
C3 and C4, with higher yields of LCO, slurry oil, coke and dry gas. Gasoline
octane numbers generally increase as the feed becomes more aromatic.

Feed sulfur and nitrogen contents

Page 61 of 83
The sulfur content of the feed has only minor effects on the conversion and other
FCC yields, except that as sulfur increases, the yield of H2S will increase. In
addition, the SOx emissions from the regenerator will rise, as will the levels of
sulfur in all of the liquid products. This may significantly affect the ability of the
refiner to make environmentally acceptable products. This is one reason that the
refiner may consider feed hydrotreating, since it will remove up to 99% of the
FCC feed sulfur.

Nitrogen can significantly reduce FCC conversion. Usually, nitrogen levels are
highest in aromatic feeds. These feeds tend to make high octane gasoline, which
is high in olefins (as indicated by a high bromine number). High activity
catalysts are usually needed to overcome the effects of the high feed nitrogen.

VI. Operation of the unit

Then control is presented with the operating parameters and process variables on
which the operator can act in order to:

 Ensure efficient and safe operation of the unit.

 Meet the required product specifications.

1. Reaction and Regeneration Section

a. Control of the process

The RFCC unit will operate in a stable manner over a wide range of conditions.
Careful attention to unit performances and process variables will result in an
operation of maximum profitability with few problems and upsets.

As a general rule any change in the vessels should be done slowly in gradual
increments for stable catalyst circulation.
Page 62 of 83
In the present section we will look at these variables again from more practical
standpoints. The most critical control loop on the converter is the Riser Outlet
Temperature. This temperature should be controlled within 1 deg. C of the set
point. The reaction temperature is measured by three thermocouples located at
the reactor outlet. The temperature is a function of the amount of catalyst
admitted to the riser by the regenerated catalyst slide valve. Reaction
temperature is the most important parameter for the conversion of feedstocks.

The spent catalyst slide valve controls the catalyst stripper level by modulating
the flow of spent catalyst from the catalyst stripper to the regenerator. The
catalyst level in the stripper is measured by differential pressure instruments and
sends a signal to the controller which sets the position of the slide valve. A
minimum level is required in the catalyst stripper to ensure good stripping of
hydrocarbons from the catalyst.

A minimum level must be maintained in the first regenerator to ensure good

regeneration and to seal the cyclone diplegs. A low level in the regenerators may
unseal the diplegs which could result in backflow of flue gas up the diplegs and
loss of catalyst fines with the flue gas. A high level in regenerators can also
result in catalyst carryover to the cyclones because of higher entrainment from
the bed and re-entrainment from the cyclone dust bowls. During normal
operation, first regenerator bed level is not controlled, and follows the unit
inventory. Second regenerator bed level is controlled by the plug valve opening.

First and second regenerator pressure are controlled by the flue gas slide valves
which throttles the flow of flue gas from those vessels. The differential pressure
between the disengager and the first regenerator is controlled by the first
regenerator flue gas slide valves. This differential pressure is set to provide
adequate pressure drops across the catalyst slide valves for stable catalyst
circulation. The disengager pressure rides on the main fractionator pressure
which is controlled at the main fractionator overhead receiver. The objective in
operation is usually to set the main fractionator pressure to a base level for
Page 63 of 83
efficient operation and regenerators vessel pressures to result in approximately
equal differentials across the catalyst slide valves.

As a general rule any change in the vessels should be done slowly in gradual
increments to allow slide valves to reposition properly for stable catalyst
circulation. The unit is designed to provide adequate slide valve pressure
differential for safe operation. Override controls are provided for action in low
slide valve differential upsets which close slide valves to prevent dangerous
reverse flow. Normal catalyst slide valve differential is around 30 to 50 kPa to
provide stable control. Slide valve differentials above 70 kPa are to be avoided
because they may cause valve erosion. Negative differentials should never be
permitted and the PDIC controls should be set to override the main controllers at
10 kPa. Smooth catalyst circulation is paramount for successful operation and is
achieved by proper catalyst aeration and conveying in the transfer lines, as well
as control of the unit pressure balance. An emergency shutdown circuit will
close the catalyst slide valves automatically upon loss of feed and loss of
combustion air.

b. Operating parameters

In the previous section has listed the process variables, i.e. the variables
(pressures, temperatures, catalyst activity, regenerators air balance, feedstock
quality, coke yield, delta coke, catalyst to oil ratio) which according to the
thermodynamics and the kinetics have an impact on the reaction involved in the
process. This rather theoretical approach did not outline whether the operators
could actually change the considered variable. In the present section we will
look at these variables again from a more practical standpoint such as operating
parameters and how the operators can actually use them to adjust the
performance of the unit.

Page 64 of 83
Riser Outlet Temperature

The operating temperature of the riser outlet will normally be set to achieve the
desired degree of conversion. Typically, a temperature of 550Deg. C will
correspond to a maximum propylene operation and a temperature of 525 to
530Deg. C will correspond to a maximum gasoline operation.

This temperature is controlled by the regenerated catalyst slide valve position

allowing more or less hot regenerated catalyst to contact and mix with the
incoming feed. When the hot catalyst contacts the feed in the bottom of the riser
the oil vaporizes almost instantaneously. The homogeneous mixture of oil and
catalyst reaches a temperature approximately 30 to 40Deg. C higher than the top
of the riser. The initial temperature shock causes thermal cracking, while
catalytic cracking begins after the oil is converted to vapor, and the molecules
contact the active catalyst sites.

As the catalytic cracking progresses and the catalyst/oil mixture flows up the
riser, the temperature drops since the heat of cracking is endothermic. At the top
of riser the moles of products are 3.5 - 5.0 times more than the moles of fresh
feed. The riser temperature has a complex interrelation with the other parameters
in cracking. An increase in the riser temperature is generally accompanied by:

 an increase in conversion,

 an increase in dry gas yield,

 an increase in LPG production,

 an increase or a decrease (over cracking) in gasoline production

depending on the reaction severity,

 an increase in gasoline octane,

 a decrease in LCO and slurry yield,

Page 65 of 83
  a slight increase in coke production.

These are only trends. A quantitative definition of the changes requires heat and
material balances as well as yield correlations. The catalyst quality also affects
the extent of the changes.

Catalyst activity

In conventional gas oil cracking the catalyst activity is measured in two different
ways, from actual operation and in the laboratory.

In the current practice, an equilibrium catalyst sample is forwarded each week to

the catalyst supplier laboratory. There, a standard test (feedstock and operating
conditions) is performed to check catalyst activity. Other analyses such as
surface area, density, pore volume, particle size distribution, metal content, are
also reported. The catalyst activity, thus measured, relates to the conversion that
the actual operating unit may experience with a different feedstock.

The correlation of the laboratory measurement is somewhat loose on account of

the difficulty in properly quantifying the differences not only in the feedstock,
but also the relationship of the method to unit operation. This leads more to a
directional relationship in the sense that higher activity catalyst in gas oil
cracking leads to greater conversion.

The best way of measuring catalyst activity is done by analyzing unit operation.
The conversion is calculated from unit yields, which include gas and coke
productions. Close monitoring of the yields, changes in metal deposition and
catalyst surface area are the best methods for maintaining the desirable level of
activity. Catalyst activity should not only be interpreted as the level of
conversion, but also as the ability of the catalyst to produce the maximum

Page 66 of 83
amount of valuable products as high octane number gasoline, while at the same
time minimizing the coke and dry gas productions.

Catalyst activity is maintained constant by the addition of fresh catalyst on a

continuous basis to compensate for losses through cyclones. Frequent
monitoring of the equilibrium catalyst properties, MAT activity, surface area
and level of metal contamination (i.e. Ni, V, Na, etc.) is recommended (see
section 7.9).

Regenerators air balance

The reduction of the coke on the catalyst to less than 0.05 wt % requires a
predetermined amount of air regardless of how the air is distributed between the
two regenerators. In a unit with one regenerator the required air rate would be
the sum of what the two regenerators require as long as coke yield and the
average flue gas composition are the same. Having two regenerators provides a
flexibility of how to split the total air required. When certain constraints, such as
the maximum regenerator bed temperature and desirable flue gas composition
are imposed, the freedom to arbitrarily split the required air decreases. However,
there is a great advantage in assigning a specific air rate to each regenerator.

The first stage regenerator temperature is 770Deg. C, therefore the coke that
should be burnt must be controlled so that this temperature limit is not exceeded.
The combustion of carbon monoxide is more rapid as the temperature increases
over 650 Deg. C. This results in lower carbon monoxide concentration and
greater air requirement as the regenerator bed temperature rises. When coke
burns to form carbon monoxide and steam the heat generated is about half as
much as when the same amount of coke burns to form carbon dioxide and
steam. The higher temperature generated by greater carbon dioxide production
further facilitates the burning of carbon monoxide. If the air rate is limited, then
the conversion of CO to CO2 is controlled and the regenerator bed temperature is
Page 67 of 83
kept lower. But at the same time more coke remains on the catalyst. The
objective is to determine an air rate to the first stage regenerator that will limit
the bed temperature and yields carbon monoxide in the flue gas.

The objective in the second stage regenerator is to burn all the remaining coke
completely to carbon dioxide. The bed temperature is allowed to rise 730Deg. C
(mechanical design : 840Deg. C). The air rate is adjusted to give 2 - 3 mole % of
free oxygen in the flue gas which ensures that the carbon monoxide
concentration in the flue gas is less than 0.05 mole %.

The above split of the required air also has the advantage of minimizing catalyst
deactivation. Most of the hydrogen contained in the coke burns in the first stage
regenerator. The steam thus generated is at a lower temperature and causes less
catalyst deactivation. Since only a small part of the total hydrogen contained in
the coke is burnt in the second stage regenerator, the steam concentration is
reduced in the higher temperature environment. The flue gas from each
regenerator is routed separately which segregates the flue gases according to
their carbon monoxide concentration.

Regenerator temperatures

Dense phase temperatures

First stage

The first stage regenerator dense phase temperature is a function of the riser
temperature, the amount of coke burned, and the catalyst circulation rate. The
temperature is controlled by varying the air flow to the vessel. The air rate
should be adjusted so as not to exceed a temperature of 730Deg. C.

Second stage

Page 68 of 83
The second stage regenerator dense bed temperature is also a function of the
amount of coke burned and the catalyst circulation rate. The normal operating
temperature is around 100Deg. C higher than the first stage regenerator dense
bed. Control of this temperature is not independent because the air rate to the
second stage regenerator is adjusted to achieve complete CO combustion (i.e. 2 -
3 % mole O2 in flue gas) under normal operation. In any case, it is important to
keep a minimum temperature in the catalyst dense bed, which is required for a
proper combustion.

Dilute phase temperatures

The dilute phase temperature will normally run within 10Deg. C of its
corresponding dense phase temperature. The catalyst used, while not promoted,
has properties which enhance combustion in the dense bed. These properties,
together with proper inventory of catalyst in the dense phase, ensure that all
oxygen required for combustion will be burnt away in the dense bed, removing
the possibility of after burning.

Regenerators residence time

The catalyst residence time in the regenerators is a key parameter for the
regeneration quality. The catalyst levels in the two regenerators are optimized
depending on the regeneration temperatures to achieve the required inventories.

Typically, a total residence time around 6 minutes for the two regenerators is
sufficient to achieve a carbon on regenerated catalyst leaving the second
regenerator of less than 0.05 wt %, which is considered as a target value for a
good regeneration.

Page 69 of 83
During normal operation, care must be taken to keep the catalyst levels at their
normal levels. Low catalyst levels will affect the regeneration quality, whereas
high catalyst levels will affect the catalyst entrainment to the cyclones and
consequently will increase the catalyst losses.

Stripper operation

The removal of any light hydrocarbons which may remain with the catalyst after
disengagement in the disengager / stripper is accomplished by steam injection
into the stripper dense bed via different steam rings. The amount of steam is
adjusted depending on the catalyst circulation rate and feed rate. The flow rate is
adjusted on the main steam ring, keeping the flow rates on the lower ring and
upper ring constant. Stripping efficiency is measured by the hydrogen content on
coke. Stripping is considered as efficient when the hydrogen on coke is around 7
wt %.

Feedstock quality

In catalytic cracking a hydrogen deficiency develops as the hydrocarbon

molecule splits and requires that a hydrogen joins a cracked molecule. The
higher molecular weight hydrocarbons have lower concentration of hydrogen
than the lower molecular weight hydrocarbons. When the cracking process
produces lighter hydrocarbons than the feed, the hydrogen is made available
from the hydrogen of hydrocarbons with higher molecular weight.

The molecule that gives up hydrogen can become deficient in hydrogen to the
point that it turns into coke. Hence, it is easy to see that the yield of lighter
hydrocarbons or the conversion depends on the total amount of hydrogen
contained in the feed. API gravity and distillation determine the hydrogen
availability to yield lighter products. Even when the distillation is not properly
Page 70 of 83
defined, the API gravity gives an indication of feed quality, since lower API
gravity feedstock generally has a higher boiling range and less hydrogen.

The metals, which are part of the RFCC feed, adversely affect performance.
Nickel, vanadium, copper and iron carried by the feed will deposit on the
catalyst sites and through a complex mechanism, lead to catalyst deactivation.
The process is gradual up to a certain metal level on the catalyst, but above
about 10 000 ppm of metals, deactivation becomes more rapid. If the metal
concentration of the feed cannot be reduced to control this, then fresh catalyst
addition replacing catalyst in the unit must be at a rate high enough to keep the
metal concentration on the catalyst at the required level.

Sodium and other alkali metals also act as catalyst poisons. Exceeding 1 ppm of
sodium in the feed should be avoided. Also sodium decreases catalyst melting
point making it more sensible to high temperatures.

The nitrogen and the sulphur in the feed adversely affect cracking, but are less
harmful. Both the nitrogen and the sulphur tie up hydrogen that could be used in
the formation of valuable hydrocarbons. In addition to this, the ammonia that is
formed is basic, and has a neutralizing effect on the acidic catalyst sites.

The Conradson carbon in the feed was in the past believed to convert fully to
coke. The coke on the catalyst surface must be burned off in the regenerator, and
more coke produces higher regenerator temperatures. Hence, increasing
Conradson carbon in the feed was expected to lead to inoperable conditions. In
the development of the R2R process, the temperature limitation was raised to be
able to handle higher coke burning rate. At the same time, it was noted that
previous beliefs stating that 100% of the Conradson carbon converts to coke
were false. Only about 50% of the Conradson carbon converts to coke, while the
rest turns into gaseous products. As the Conradson carbon concentration
increases, the second stage regenerator temperature has a tendency to increase.
Adjusting certain operating parameters such as feed temperature, disengager

Page 71 of 83
pressure, atomisation steam flow, stripping steam flow, may compensate for this
up to a point.

Feed temperature

The preheat temperature of the feed must be adjustable:

 to ensure a proper oil viscosity (around 10 to 15 cSt maximum at the

injector inlet) for proper atomization of the feed,

 to ensure a minimum temperature to avoid steam condensation in the feed

injectors.

The feed temperature must also be optimized depending on the heat balance.
The feed temperature affects significantly the coke production and the second
regenerator temperature.

Note that an increase of the feed temperature will result in:

 a decrease of coke production,

 an increase of the second regenerator temperature.

Coke yield / delta coke / catalyst to oil ratio

The catalyst to oil ratio (C/O) is defined as the rate of catalyst divided by the
rate of fresh feed. The delta coke is defined as the difference between the coke
percentage on spent catalyst and the coke percentage on the regenerated catalyst.

The coke yield corresponds to the percentage of feed transformed into coke.

These three parameters are correlated by the following equations:

Page 72 of 83
These relations reflect the heat balance between the reaction section and
regeneration section. The three process variables will automatically achieve the
conditions where the heat balance is satisfied. For a given feed quality and
catalyst type, the coke production depends upon nothing except Riser Outlet
Temperature and feed temperature. The unit will be optimized with the highest
possible C/O and the lowest possible delta coke for the following reasons:

a higher C/O will provide:

 more active catalyst sites for the reaction,

 a better contact between catalyst and oil,

 a higher heat transfer efficiency.

This leads to higher conversion, i.e. larger total liquid, LPG, gasoline yields
while slurry yield declines.

a lower delta coke will provide:

 lower regeneration temperature.

C/O is increased by:

 raising the Riser Outlet Temperature

 decreasing the feed temperature.

But the adjustment of the feed temperature and Riser Outlet temperature is
limited:

Page 73 of 83
  the feed temperature cannot be lowered below a limit value in order to
maintain an acceptable feed viscosity for proper atomization and to
maintain a minimum temperature for good vaporization,

 the Riser Outlet Temperature is fixed by the operation mode (maxi

gasoline or high propylene mode).

Catalyst circulation / pressure balance

Catalyst circulation results from different vessels elevations and from

differential pressure created by various catalyst densities.

Fluidized catalyst behaves very similarly to normal liquid fluids.

Smooth catalyst circulation requires precise control of the unit pressure balance,
through the precise control of the pressure in the vessels and the proper control
of the densities in the catalyst dense beds and standpipes.

The pressure in the disengager is controlled by the main column overhead

receiver. The pressure in the disengager is higher than the receiver pressure by
the amount of pressure drop through the main column, the condenser and the
lines between top of the disengager and the receiver. The pressure in the
disengager is normally kept as low as possible within the limits of the wet gas
compressor.

The first stage regenerator pressure is controlled by a double disc slide valve
followed by a variable orifice and a differential pressure controller holds a
constant differential pressure (around 90 kPa) between the first and the second
regenerator. The differential pressure between the two regenerators should be
maintained constant for a smooth operation of the air lift.

Page 74 of 83
Flue gas slide valves should operate with a pressure differential between 30 to
70 kPa for good control and minimal disk erosion. These valves are designed to
never fully close by means of a mechanical stop or a slot cut into the disks.
Complete closure of these valves could potentially over-pressure the disengager
/ stripper / regenerator system.

Pressure of the first regenerator must be controlled so that the differential

pressures on the regenerated and spent catalyst slide valves are balanced
satisfactorily, i.e. to obtain similar pressure drops through the two slide valves.
The minimum required pressure drop through the slide valves for steady control
of the valve is around 30 kPa. Maximum pressure drop should be limited to 70
kPa for avoiding erosion problems.

It is obviously essential for steady circulation that the pressure difference

between the vessels be as consistent as possible. Toward this end, it should be
noted that the stripper and first stage regenerator levels are kept steady. The
second stage regenerator is allowed to float between its minimum and maximum
levels.

Catalyst circulation is controlled by the opening of the regenerated catalyst slide

valve. The circulation rate cannot be measured directly; it must be calculated
from the heat balance or estimated from the openings and pressure drops of the
regenerated and spent catalyst slide valves.

c. Adjustment of operating conditions

Monitoring and optimization of the performance of a RFCC can add

significantly to the overall refinery margins. It requires careful data collection
and validation procedures to ensure that a proper evaluation is made and the
right conclusions are drawn with respect of the catalyst effects. The relevant
data for this exercise include the unit conditions, feedstock properties, product
yields and qualities as well as equilibrium catalyst analyses.

Page 75 of 83
Unit monitoring includes data collection, validation and interpretation. The
number of available measurements will determine the extent of the evaluation.
The required data can be grouped as follows:

 Feedstock properties (density, sulfur, metals etc..),

 Mass balance data (product flows, densities),

 Essential product properties (distillation, octanes, sulfur, viscosity),

 Heat balance data (temperatures, flue gas composition, air rates),

 Pressure balance data (vessels pressures, valves P, standpipe operation,

catalyst levels),

 Equilibrium catalyst analyses (activity, surface area, metals, etc…).

In most cases heat and mass balance data are available on a continuous basis
from the process computer. Feed and product analyses are made periodically.
The equilibrium catalyst analyses are normally available once per week. A
careful survey of the operating conditions must be constantly performed.

To minimize analyses, a good definition of essential product properties is

required for the refinery engineer. Apart from that it is also important that the
essential analyses are from samples taken at the same time.

All this together allows a proper monitoring of the RFCC unit. The modification
of the feedstock properties, operation mode (maxi distillate / maxi gasoline...)
and catalyst activity (metals content, new catalyst) requires to adjust the
operating variables.

Feedstock properties

Page 76 of 83
The feed must be analyzed on a daily basis. The metals content must be
carefully checked, especially the sodium content which should be limited to 2 wt
ppm maximum to avoid catalyst poisoning and deactivation. Malfunctioning of
crude desalters or processing of imported feeds contaminated with seawater are
the typical causes of high sodium levels in most cases.

If applicable, for feeds with high metal concentration, nickel introduced with the
feed is partially passivated by passivator injection (Sb solution recommended).
The passivation efficiency can be checked by measuring the ratio H2/C1 in the
dry gas. This ratio should not exceed 1.0. Otherwise, the amount of passivator
should be increased accordingly.

d. Troubleshooting

Troubleshooting situations

The problems often encountered in an RFCC unit are mainly due to changes in
the feedstock, catalyst, operating variables, and mechanical equipment. As
stated previously, the solution can take the form of improving yields, avoiding
shutdowns or, increasing unit reliability.

Troubleshooting consists of investigating and correcting the causes of

unsatisfactory operations (typically off-specifications products or unexpected
operating conditions) before they deteriorate any further. Troubleshooting
actions must be undertaken when:

 The product is off-specifications,

 Unexpected operating conditions are noticed.

The main causes of concern, related to process are:

 Catalyst circulation problems,

Page 77 of 83
  Excessive catalyst losses,

 Poor quality of regeneration,

 Poor spent catalyst stripping,

 Product quantity and quality.

Catalyst circulation problems

Catalyst circulation problems are mainly correlated with:

 Inadequate equilibrium catalyst properties,

 Improper catalyst fluidization and aeration.

In case of unstable circulation (unstable catalyst valve P, sudden loss of pressure

above the slide valve, ragged ROT temperature and/or stripper level control,
levels swings in the regenerator), the problem must first be localized by
checking the complete pressure balance of the unit. Improper catalyst
fluidization and/or aeration will result in erratic pressure profile, i.e. no pressure
gain or limited pressure gain in the catalyst standpipes or catalyst dense beds. In
that case, the adjustment of the fluidization rates or aeration rates will help to
restore the proper circulation. Especially aeration and fluidization nozzles and
lines must be checked for cleanliness and flow conditions.

The catalyst circulation stability depends often on the equilibrium catalyst

properties. The following factors appear to improve the fluidization:

 A lower particle size distribution and lower catalyst density,

 A higher fines content (lower than 40 microns).

Page 78 of 83
Typically, no circulation problems are observed with catalyst containing more
than 10 wt % of fines (0-40 microns). In case of too low fines content (lower
than 10 wt %) adding more fresh catalyst and/or using a softer catalyst can help
to improve the catalyst properties. Adding fines directly in the unit is inefficient
and can overload the cyclones. The circulation can also be improved by
increasing the pressure level in the regenerator if pressure margin is available on
the catalyst slide valves.

Erratic circulation occurs when catalyst is not developing a smooth and uniform
static head over the entire length of the stand pipe. When this appears, the
catalyst packs and bridges across the stand pipe section.

Excessive catalyst losses

Excessive losses can occur through the disengager cyclones or through the
regenerators cyclones. In the first case, the fines content in the slurry product
will become too high, causing plugging of the slurry circuit equipments:
exchangers, pumps. In the second case, flue gas dust emissions will be
excessive. Monitoring the catalyst losses requires periodic analyses of fines
content in the flue gases and in the slurry product.

Excessive losses can be due to:

 Catalyst attrition,

 High or low catalyst levels in the vessels,

 Cyclone performances deterioration.

Page 79 of 83
2. Main Fractionation and Gas Recovery Sections.

a. Control of the Main Fractionation section.

Control of the Main Fractionation section rests on heat removal coupled with
sufficient liquid / vapor contacting to achieve the desired fractionation in order
to have product streams on specification. The main parameters to operate the
column are those which control the following product qualities:

 Final boiling point.

 Gap between products.

 Flash point when applicable.

Even though each section of the column has a relatively independent system for
control, changes in pumparound or in product rate have an impact on the
regulation in one or more sections or on the product rates to maintain the global
heat and material balance of the column constant. The basic criteria used for
setting the Fractionator operation is related to operating pressure and
temperature profile.

b. Operating parameters for several specifications of products

The Main Fractionator pressure is controlled at the Overhead Receiver by speed

control of the Wet gas Compressor in the Gas Recovery section. Pressure is not
a process variable from an operation point of view and is not adjusted to correct
product specifications.

The unit operates with a pressure of 20 kPa g in the Main Fractionator Overhead
Receiver. This pressure sets the pressure in the Reactor, where low pressure is
desirable to achieve good feed vaporization. For stable feed to the Reactor and
stable Reactor operating conditions, the heat input to the Main Fractionator

Page 80 of 83
remains constant. The heat removal in the pumparounds and the product draw
rates are the variables available to the operator in the operation of the column.

Final boiling point is obtained by adjusting the flow of the product withdrawn.
Therefore, as the pressure is not a control parameter and is kept constant, the
temperature of the withdrawn product is a good indication of the final boiling
point.

Main Fractionator top temperature, adjusted by the reflux rate, controls the end
point of the naphtha to the Primary Absorber. If a product draw rate is increased,
the end point increases. Separation requirements, gap between products are
achieved by adjusting the internal liquid and gas traffics which are related to the
pumparound flows and temperatures.

If the heat exchanged on a pumparound is too important, there will be less

internal reflux in the section of the column above this pumparound. The gap
between the product withdrawn above the too cold pumparound return and the
product withdrawn at the pumparound level will decrease. If the heat exchanged
on a pumparound is not enough the reflux in the sections above increases and
therefore, separation increases. Reducing pumparound duties results in better
separation at the expense of high level heat recovery. Each pumparound should
be set to provide adequate fractionation between products while maximizing
heat recovery. Finally the flash point is adjusted by the flowrate of stripping
steam.

c. Operating parameters for Slurry section

Great care should be applied to the operation of the bottom of the Main
Fractionator and of the associated steam generators in order to minimize fouling
of the steam generators and coke deposit in the Main Fractionator bottom. In
order to avoid the above problems the following parameters are to be respected:

 Low temperature in the bottom (320 deg. C max)

Page 81 of 83
  Short residence time in the bottom.

 Sufficient flow of washing liquid above the bottom packing section of bed
1 (700 t/h) whatever the unit capacity.

 Velocity in steam generators tubes between 1 and 2 m/s.

If the bottom temperature is too high, coking on the grid and in the bottom of the
column can occur. Coking is a function of both temperature and residence time.
The column bottom has a reduced diameter to limit residence time and the
bottom pumparound and quench are designed to limit temperature.

It is essential to have an adequate Slurry Pumparound return flowrate to the grid

to ensure good distribution. The hot slurry by-pass and the flow control system
allow constant flowrate to the grid and heat removal flexibility. At low capacity
a by-pass of the steam generators allows to maintain a constant flow of washing
liquid above bed 1.

Also, to maintain velocity in the steam generators it is possible to by-pass one or

two steam generators at low capacity operation.

d. Wet Gas Compressor – Absorption section of Gas Recovery

The aim of this section is to separate the fuel gas from a cut LPG + Gasoline by
maximizing the recovery of C3/C4. Control of the absorption process rests on
having high and stable pressure in the columns and cooled process fluid to
improve absorption.

LPG Amine Absorber Section

The LPG amine absorber removes H2S from the LPG prior to LPG treatment for
the removal of mercaptans. The LPG should have a maximum of 50 ppm wt H 2S
at the column outlet. The column operates liquid-filled, with the LPG rate from
the column controlled by MDEA/LPG interface level control.

Page 82 of 83
Foaming is in this case not as strong an issue as for Fuel Gas Amine Absorber
since both LPG and amine are under liquid phase. Nevertheless, foaming may
still occur in case of accumulation of heavier hydrocarbon compounds (at the
interface between LPG and amine). An intermittent injection of anti-foaming
agent is provided also in the lean amine line to the lean amine absorber.

Moreover, amine entrainment in LPG is to be minimized in order to minimize

losses (for economical reasons) and to avoid perturbations in downstream units.
A water wash of LPG cut is provided to eliminate remaining amine.

Gasoline Splitter Section.

The purpose of the Gasoline Splitter section is to separate Light Cracked

Gasoline from Heavy Cracked Gasoline. Temperature of sensitive tray controls
the reboiling of Gasoline Splitter by adjusting the flowrate of MP steam in the
Gasoline Splitter MP Steam Reboiler.

The reboiling and the reflux rate of the Gasoline Splitter are the key parameters
to control the gap between heavy cracked and light cracked gasoline.

Page 83 of 83