Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
I
HIGHER CONTAINMENT VSS WITH MULTI ZONE STRIPPING
FIELD OF THE INVENTION
The present invention discloses engineering design modifications to
the vortex separation system (VSS) exit, stripper entrance and the primary
cyclone diplegs that can significantly reduce the underflow of reactor riser
products into the stripper and reactor vessel and thereby produce higher
desired product selectivities, improved stripping efficiency and a stripper
vent
gas, that continuously flows through the reactor vessel, with a low coke
forming potential due to its low concentration of ethylene and higher
molecular
weight material, that could, if desired, be recovered separately from the
primary riser products.
BACKGROUND OF THE INVENTION
Over the years, the FCC reactor section has been developed from Bed
and Riser Systems to various current High Containment Configurations.
These High Containment Reactor Systems should include:
Rough-Cut Cyclones
Direct Coupled Cyclones
VSSNDS
With these Reactor Systems, the vast majority of the catalytic reactions
and conversion now take place in a highly selective dilute phase transport
regime with short contact times and essentially plug flow conditions.
However, by using a combination of various published commercial
data, along with fundamental catalytic cracking mechanisms and fluidization,
one can show that further significant selectivity improvements can still be
achieved by making some key design modifications that will produce even
higher product containment, with improved operability and better overall
stripping efficiencies.
Here it must be understood that various equations are presented in
support of the findings disclosed in the application and a table defining the
nomenclature used in such would be helpful in the understanding and is
presented below.
J TURNOVER RATE OF SOLIDS lb/Ft2/sec
Date Recue/Date Received 2023-03-28
2
Kea EFFECTIVE AXIAL THERMAL CONDUCTIVITY OF THE BED
Btu/Ft/F/Hr
a RATIO OF WAKE TO BUBBLE VOLUME
ps SKELETAL DENSITY OF SOLIDS lb/Ft3
pmf MINIMUM FLUIDIZATION DENSITY lb/Ft3
emf VOID FRACTION AT MINIMUM FLUIDIZATION
e VOID FRACTION
U SUPERFICIAL GAS VELOCITY Ft/Sec
UL & Ue SUPERFICIAL EMULSION PHASE VELOCITY Ft/Sec
UB BUBBLE RISE VELOCITY OF A CROWD OF BUBBLES Ft/Sec
Umf MINIMUM FLUIDIZATION VELOCITY Ft/Sec
Umb MINIMUM BUBBLING VELOCITY Ft/sec
fl" BUBBLE FREQUENCY RELATIVE TO PACKET OR EMULSION
IT BUBBLE FREQUENCY RELATIVE TO .A STATIONARY OBSERVER
VB BUBBLE VOLUME Ft3
A CROSS SECTIONAL AREA Ft2
W CATALYST FLUX lb/Ft2/Sec
With the nomenclature now defined, turning to Table 1, it is found to
show published commercial vapor samples taken from a modern Direct
Coupled Cyclone System. Sample (1) represents the final FCC products and
sample (2), taken from the reactor vessel, represents the combined
hydrocarbon composition of the stripper effluent and the voidage or underflow
material flowing with the catalyst down the cyclone diplegs. Unlike the bulk
of
the contained riser products, the underflow and stripper material suffers
additional catalytic cracking in a now less than desirable pseudo Dense Bed
reactor configuration with a high degree of backmixing, low space velocity and
high residence times. The change in not only selectivity, but product
composition between the two samples is enormous and very significant.
TABLE 1
FCC Products
YIELDS SAMPLE 1 SAMPLE 2
Date Recue/Date Received 2023-03-28
3
DRY GAS 2.7 15.6
C3 LV% 8.9 17.2
C4 LV% 14.9 20.4
GASOLINE LV% 54.2 46.6
LCO LV% 20.9 10.0
CO LV% 8.1 5.8
With the pseudo Dense Bed reactor, the catalytically cracked products
in sample (2) will be rich in C3, C4, C5, and Iso-C6 branched paraffins rather
than olefins. Significant gasoline yield has been lost to these products; but
under these pseudo Dense Bed conditions, additional production of iso-
paraffins from iso-olefins via hydrogen transfer and the generation of
aromatics by cyclization and dehydrogenation will occur. The gasoline
paraffin, isomer, naphthene, and aromatic composition will be totally
transformed; now being much richer in toluene and xylenes, it will have a
research octane number approaching 100.
The still unconverted cycle oils are now essentially all highly de-
alkylated, two, three and four ring aromatics; with all the remaining alkyl
groups being primarily methyl and a few ethyl.
However, all this undesirable secondary catalytic cracking of the riser
products does not explain the high 15,6 Wt % dry gas production. The vast
majority of this dry gas production is actually being generated from the
chemically adsorbed material on the catalyst's surface, which is often
referred
to as "soft coke". This material is composed primarily of two or more highly
condensed ring aromatics and plays a significant role in the complex
sequence of the final "hard coke" formation. The coke forming tendency of
these compounds correlates well with their basicity due to the catalytic
surfaces acid-base interaction.
First, the quantity of this undesirable riser underflow material can be
determined from calculations of the flowing catalyst voidage down the primary
and secondary cyclone diplegs. With the void fraction being represented by:
e = (1-pB/ps) where Ps is the skeletal density
Date Recue/Date Received 2023-03-28
4
Under normal FCC reactor conditions and assuming an average riser
product molecular weight of 100, Table 2 shows, the estimated voidage or
underflow material could be anywhere from 3Wt% to 8Wt% on a fresh feed
basis; and since all the catalyst flows with the underflow, this Wt% "soft
coke"
could easily be equivalent to the Wt% hydrocarbon underflow.
Note: In Table 2 the voidage calculations and their estimated Wt % of
fresh feed do not include this soft coke term since it is chemically adsorbed
material on the catalyst's surface.
TABLE 2
HYDROCARBON UNDERFLOW AND DIPLEG DENSITY
DI PLEG DENSITY HCBN's ACF/IbFF HCBN's Wt% FF
50 0.0973 2.16
40 0.1323 2.93
30 0.1907 4.23
0.307 6.81
Secondly, we can now determine the light cracked gas that's being
produced in these secondary pseudo Dense Bed reactors and strippers.
15 Since this adsorbed "soft coke" material undergoes further dealkylation
and
condensation reactions with higher aromatic ring formations. In the process,
they will produce relative equal amounts of hydrogen and methane on a
molecular basis, due to the alkyl groups that are present. Similar type
reactions will also produce significant quantities of hydrogen sulfide from
this
20 strongly adsorbed "soft coke". The highly, more condensed multi-ring
aromatics become even more strongly adsorbed on the surface of the
catalyst; and ultimately finish up as "hard coke" in the regenerator. With
these
reactions, the resulting dry gas composition is therefore quite different from
that produced in the primary riser. If we assume a typical 50/50 Mol%
Hydrogen/Methane blend in sample 2's dry gas fraction. Using a typical dipleg
density and a 5Wt% of fresh feed underflow, one can then calculate the weight
and volume percent dry gas that's produced from the "soft coke" on a Wt'' net
underflow or fresh feed basis. See Figure 11.
Date Recue/Date Received 2023-03-28
5
The estimated numbers in Figure 11 are for the secondary dry gas
production from the "soft coke" in the DCC's reactor vessel, diplegs and
stripper sections. They are large, particularly on a volume basis, and become
very significant.
Normal stripping steam rates are set around 2 pound's steam/1000
pound's of catalyst. In a unit with 7 cat/oil this steam usage is 1.4 Wt % of
fresh
feed (FF) or 7.77 Vol% of Rx Effluent. The total estimated dry gas production
from the cracked "soft coke" along with that produced with the additional
underflow conversion, equates to 0.81 Wt% of FF. As shown in Table 3, with
a 9 to 18 molecular weight advantage, this cracked gas can become higher
than the normal stripping steam rates on a volume basis.
TABLE 3
CRACKED "SOFT COKE"
Mol WT Wt% FE Vol% Rx Eff
STRIPPING STEAM 18 1.4 7.77
CRACKED DRY GAS 9 0.81 9
HCBN UNDERFLOW 100 5 5
The "soft coke" dry gas production on a typical 5.5wt% coke yield would
be 11.50 Wt% of the overall units enthalpy coke yield. An additional review of
the "soft coke", "hard coke" hydrogen balance yields:
SOFT COKE HARD COKE DRY GAS
111.50(X)= 100(0.06) + 11.50(0.30)
X = 8.48 Wt % Hydrogen.
All these numbers make sense and are very significant. The "soft coke"
underflow is calculated having a 8.48Wt% hydrogen content but the final "hard
coke" entering the regenerator as a much lower 6.0 Wt % hydrogen content.
The simple magnitude of this cracked gas on a volume basis (9 verses
7.77) will have a significant impact on all commercial stripping efficiencies
particularly on a volume basis. Yet none of the previous, cold flow modeling
work associated with FCC strippers appear to have taken into account the
Date Recue/Date Received 2023-03-28
6
magnitude of this inert, extremely low molecular weight, material that's being
continuously produced from the catalyst's surface.
In reviewing these various none-selective post riser reactions it does
need to be emphasized that many will occur almost instantaneously in the low
space velocity, pseudo beds; however, the condensation "soft coke" reactions
will take significantly longer and will be highly dependent on reactor
temperature.
Unlike the catalyst flow in the DCC's primary cyclone diplegs the
fluidized state in the reactor stripper and lower VSS chamber is ideally that
of
a flowing counter current, dense phase, gently bubbling bed. A dense
fluidized bed has many unique and beneficial characteristics; but the high
degree of axial solids mixing along with low contacting efficiency between the
bubbling gas phase and the solid emulsion phase, can be quite detrimental.
The reactor stripping section therefore typically contains various internals
in
order to limit the overall backmixing and approach a more desirable "plug
flow" stripper via the use of multiple "backmixed" stages. Table 4, shows the
calculated volume percent hydrocarbon displacement that can be expected
for various steam/hydrocarbon ratios along with the number of theoretical
"backmixed" stages.
TABLE 4
REACTOR STRIPPER & THEORETICAL BACKMIXED STAGES
STM/HCBN Ratio 1.0 1.0 1.0 1.0 1.0
STAGE 1 2 3 4 6
Vol% HCBN REMOVED 50.0 67.0 75.0 80.0 85.7
STM/HCBN Ratio 2.0 2.0 2.0 2.0
STAGE 1 2 3 4
Vol% HCBN REMOVED 67.0 85.7 93.4 96.8
STM/HCBN Ratio 4.0 4.0
STAGE 1 2
Date Recue/Date Received 2023-03-28
7
Vol% HCBN REMOVED 80.0 95.2
Using the classic bubbling bed model and some calculated fluidization
parameters, one can review and discuss the various design and operating
conditions presently used in the lower VSS chamber, stripper and its internals
and the cyclone diplegs. Kunni and Levenspiel proposed a bubbling bed
model that views a vigorously bubbling fluidized bed to consist of a crowd of
uniformly sized bubbles rising through the continuous phase, called the
emulsion. Each bubble is surrounded by its cloud of circulating gas that is
followed by a wake of material. Thus, solids are carried up the bed in the
bubble wakes and move downwards elsewhere. In a stationary bed, there is
no net flow of solids across the plain X-X. The mixing between the steam rich
bubble phase and the hydrocarbon rich emulsion phase is therefore limited
and set by diffusion.
This axial turnover rate, or backmixing in a vigorously bubbling and
flowing bed, can become extremely high and very detrimental to high
containment of the reactor products and efficient counter current "plug flow"
stripping. Van Deemeter, Lewis and May developed various expressions to
relate this axial mixing, dispersion coefficient and effective thermal
conductivity for such bubble induced circulation of solids.
Turnover rate of solids:
= aps (1- emf) (U-Umf) lb/Ft2/sec
For FCC type material J= 20-30 lb/Ft2/sec at 1ft/sec superficial gas
velocity.
The high "J" values generated within the stationary bubbling bed are
very significant. For a moving bubbling bed, these expressions should be
based on the relative bubble frequency rather than the superficial gas
velocity.
Nicklin for a liquid/gas system showed this relative bubble frequency to be:
(VB/A) = U (1+ (UL/ (UB-U)))
fl"= fl'= (ANB) U
Date Recue/Date Received 2023-03-28
8
Or that the bubble frequency relative to the emulsion phase is
equivalent to the superficial velocity when UL is zero and the bed is
stationary.
Since the lower VSS chamber and reactor stripper operate at gas velocities
>>Umf, for a Geldart type "A" solids, these liquid/gas relationships of
Nicklin
are very applicable to these FCC flowing systems.
Since e= U/ (U+UB-UL), the void fraction e will approach 1.0 when the
velocity of the emulsion phase approaches the bubble rise velocity, and this
can lead to flow instability in lower strippers and standpipes.
In summary when:
UL<UB counter current flow will exist
UL=UB e goes to 1.0 and flow instability results
UL >UB co-current flow will exist
Therefore, for a stable dense phase, counter current flow UL should be
< U. In the case of the FCC VSS and stripper where UL is the emulsion
phase velocity, this is set by the design's "open area" catalyst flux
lb/ft2/sec.
Table 5, shows typical operating conditions used in various cold flow modeling
studies of the FCC reactor stripper system.
TABLE 5
TYPICAL FCC REACTOR STRIPPER OPERATING CONDITIONS
EQUIVALENT AIR (lbSTM/1000Ib CAT) 0.5-2.5
STAGES 7
CATALYST FLUX lb/Ft2/SEC 17-33
CATALYST BED DENSITY lb/Ft3 40-50
SUPERFICIAL GAS VEL Ft/Sec 1-1.5
At superficial gas velocities much greater than lft/sec, the bubble rise
velocity can become limiting. Yerushalmi reported this transition between the
bubbling and turbulent bed with respect to the relative pressure fluctuations
at the beds surface. Much beyond this velocity, more and more of the gas
Date Recue/Date Received 2023-03-28
9
phase starts to flow through high voidage gas channels rather than distinct
bubbles. The bubble frequency eventually plateaus; as do the beds, solids
turnover, axial diffusivity, effective conductivity and the bed to surface
heat
transfer coefficient.
The catalyst physical properties, such as particle size distribution,
angle of repose, and the <40p fines content, play a significant role in
setting
Db and the various characteristics of the fluidized bed. All the major FCC
licensors have conducted extensive cold flow modeling, using helium tracer
gas, in order to study the performance of various internal designs. As
predicted, flow instability occurred along with reduced stripping efficiencies
above certain catalyst fluxes. All report internal designs that can achieve
overall stripping efficiencies >95Vol% at the operating conditions shown in
Table 5. However, none of these studies appear to have considered the
significant impact of the secondary "cracked gas" reactions, "J" values and
the beds freeboard activity would have on riser product underflows and
stripping efficiency.
Since the stripping steam is usually based on pound's/1000 pound's of
catalyst, the relationship between superficial gas velocity and catalyst flux
can
generate some interesting trends due to the prevailing "J"s. At the higher gas
rates, the backmixing "J" values can be equal to, or significantly higher
than,
the net flowrate of catalyst. As some of the modeling data suggests, with
moderate to high gas rates as you slow the catalyst down, you are actually
increasing the degree of backmixing (JAN) and the stripping efficiency can
decline.
When compared to the theoretical stages in Table 4, the cold flow
modeling results do not exhibit any great removal efficiencies. At a
relatively
low 1lbsteam/1000Ibcatalyst, which is equivalent to a 3.6 steam/hydrocarbon
volume ratio, a stripper with only two theoretical stages would achieve 94.3%
removal. There is also no published data on actual commercial stripper
efficiencies, where these slow secondary reactions and the
hydrogen/methane production could drastically lower the volume percent
hydrocarbon removal.
Date Recue/Date Received 2023-03-28
10
AD these high containment systems now have small amounts of still
reactive riser products that spend considerable time at temperature in the
reactor vessel. Their concentrations have often been significant enough to
form highly undesirable and often problematic coke depositions.
SUMMARY OF INVENTION
Based on this detailed review of current high containment designs, one
can clearly see the potential for various significant and patentable design
improvements particularly in the VSS and stripper system. This patent
discloses various engineering design modifications to the VSS exit, stripper
entrance and the primary cyclone diplegs. The design of a three zone stripper
can significantly reduce the fluidized beds freeboard activity, superficial
velocity, and J value. These novel design modifications can be applied to a
new or existing high containment VSS system and thereby significantly
reduce the underflow of reactor riser products into the stripper and reactor
vessel, producing higher desired product selectivities and improved stripping
efficiency. The stripper vent gas, which continuously flows through the
reactor
vessel, will now have a low coke forming potential due to its low
concentration
of ethylene and higher molecular weight material that could now, if desired,
be recovered separately from the primary riser products that flow into the
main
column and gas concentration units.
Additionally, for a ZSM-5 type petrochemical operation, space velocity
and J values can now be controlled independently within the VSS chamber
via additional steam injection to independently control superficial gas
velocities and catalyst transport rates to the primary cyclone. In this
invention,
the stripper vent gas can not only be recovered and treated separately but
with its high hydrogen and methane content it can be used in a regenerator
combustion chamber to augment the enthalpy balance, increase liquid
volume yield and reduce regenerator "Green House" gases, NO and SO,
emissions.
Also, the catalyst residence times in the first and second stripping
zones can be controlled via the reactor level and the spent slide valve
opening. The residence time in the first zone is now particularly significant
in
Date Recue/Date Received 2023-03-28
11
that it will set the degree of dry gas production from the "soft coke" which
will also vary
with, and be dependent on, reactor temperature. This improved RTD in the three
zone
stripper leads to lower regenerator SOx levels via the increased conversion of
"soft coke"
sulfur to hydrogen sulfide followed by its higher removal efficiency from
within the
catalyst voidage going to the regenerator.
In some implementations, there is provided an improved vortex separation
system, comprising: a vortex separation system (VSS) separation chamber; a
stripper
positioned below the VSS separation chamber; and a baffle located between the
stripper
and the VSS separation chamber, wherein the baffle is configured to direct
stripper vapor
away from the VSS separation chamber.
In some implementations, there is provided An improved vortex separation
system that generates significantly higher plug flow containment of primary
riser
products due to reduced back mixing within a chamber and a lower final vapor
product
underflow within a catalyst void, the system comprising: a vortex separation
system
(VSS) separation chamber; a baffle located between an original stripping zone
and the
VSS separation chamber, thereby creating a three-zone stripping section that
includes
the original stripping zone, a cracked gas stripping zone and a transition
zone, the baffle
being configured to direct stripper vapors that exit the original stripping
zone and the
transition zone away from the VSS separation chamber; and at least one vent
pipe
through which the stripper vapors from the original stripping zone and the
transition zone
that rise through the original stripping zone and the transition zone and
bypass the VSS
separation chamber are passed into a VSS riser.
In some implementations, there is provided an improved vortex separation
system, comprising: a vortex separation system (VSS) separation chamber; a
stripper
positioned below the VSS separation chamber; and a cyclone separator in fluid
communication with the VSS separation chamber, the cyclone separator having a
dipleg
with a circumference that is expanded to match a circumference of the cyclone
separator.
In some implementations, there is provided an improved vortex separation
system, comprising: a vortex separation system (VSS) separation chamber; a
stripper
positioned below the VSS separation chamber; a baffle configured to divert
vapor in the
stripper away from the VSS separation chamber; and a cyclone separator in
fluid
communication with the VSS separation chamber, the cyclone separator having a
dipleg
with a circumference configured to equal a circumference of the cyclone
separator.
Date Recue/Date Received 2023-03-28
11a
In some implementations, there is provided an improved vortex separation
system, comprising: a vortex separation system (VSS) separation chamber;
a stripper positioned below the VSS separation chamber, the stripper having a
catalyst
bed at least partially disposed therein; a cyclone separator in fluid
communication with
the VSS separation chamber, the cyclone separator having a dipleg with a
circumference
that is expanded to match a circumference of the cyclone separator; a baffle
located
between the VSS separation chamber and the stripper, the baffle being
configured to
direct vapor in the stripper around the VSS separation chamber; and an
additional baffle
located on top of the catalyst bed to reduce catalyst movement.
In some implementations, there is provided an improved vortex separation
system, comprising: a vortex separation system (VSS) separation chamber; a
stripper
positioned below the VSS separation chamber; a catalyst bed disposed at least
partially
within the stripper; and a baffle located on the catalyst bed.
In some implementations, there is provided an improved vortex separation
system comprising: a vortex separation system (VSS) separation chamber; a
stripper
positioned below the VSS separation chamber; a catalyst bed disposed at least
partially
within the stripper; and a baffle configured to divert steam in the stripper
away from the
VSS separation chamber, the baffle being located on the catalyst bed.
In some implementations, there is provided an improved vortex separation
system, comprising: a reactor vessel having a bottom section; a vortex
separation
system (VSS) chamber within the reactor vessel; a stripping chamber within the
bottom
section of the reactor vessel below the VSS chamber; and a baffle positioned
to
segregate the stripping chamber into three separate stripping zones that
include a
cracked gas stripping zone, a transition zone, and a main stripper.
In some implementations, there is provided an improved crossflow vortex
separation system, comprising: a chamber; a vortex separation system (VSS)
separation
chamber disposed within the chamber, the VSS separation chamber having an
outer
wall; a stripper disposed below the VSS separation chamber; a baffle
positioned
between the VSS separation chamber and the stripper and configured to
segregate the
stripper into three zones; and at least one deflection shield engaged with the
outer wall
and angularly disposed with respect to the outer wall, the defection shield
being
configured to deflect catalyst falling through the VSS separation chamber
toward a
center of the VSS separation chamber.
In some implementations, there is provided an improved vortex separation
system, comprising: a chamber; a vortex separation system (VSS) separation
chamber
disposed within the chamber, the VSS separation chamber having at least two
outer
Date Recue/Date Received 2023-03-28
lib
walls; a stripper disposed below the VSS separation chamber and configured to
have a
main stripping zone; and a baffle positioned between the VSS separation
chamber and
the stripper; wherein catalyst falls from the VSS separation chamber into the
main
stripping zone and is further stripped by steam that passes from a pipe, rises
through
the main stripping zone, engages with the baffle, and is directed away from
the VSS
chamber.
In some implementations, there is provided an improved vortex separation
system comprising: a chamber; a vortex separation system (VSS) separation
chamber
having at least one outer wall; a stripping chamber positioned beneath the VSS
separation chamber; and a baffle that segregates the stripping chamber into
three
separate stripping zones, the three separate stripping zones including a
cracked gas
stripping zone, a transition zone, and a main stripper; wherein as catalyst
falls from the
VSS separation chamber, the baffle directs the falling catalyst into the
cracked gas
stripping zone, and the falling catalyst then passes into the transition zone
before
entering the main stripper.
In some implementations, there is provided an improved vortex separation
system having multiple stripping zones, comprising: a reactor vessel; a vortex
separation
system (VSS) separation chamber located within the reactor vessel; a stripping
chamber
located beneath the VSS separation chamber in a bottom portion of the reactor
vessel;
and a baffle located between the stripping chamber and the VSS separation
chamber to
define multiple stripping zones; the baffle being configured to direct
stripper vapors
exiting a main stripping zone and a transition stripping zone away from the
VSS
separation chamber.
In some implementations, there is provided an improved vortex separation
system, comprising: a vortex separation system (VSS) separation chamber; a
stripper
positioned below the VSS separation chamber; at least one vent pipe through
which
stripper vapors exit the system; and a baffle positioned between the VSS
separation
chamber and the stripper, the baffle having an inverted V-shape with distal
ends that
extend outwardly beyond a circumference of a bottom portion of the VSS
separation
chamber to increase stripper vapor diversion around the VSS separation chamber
and
into the at least one vent pipe.
In some implementations, there is provided an improved vortex separation
system, comprising: a vortex separation system (VSS) separation chamber; a
stripper
positioned below the VSS separation chamber; and a cyclone separator having a
dipleg
with an expanded circumference that reduces flowing catalyst flux (Ibift2/sec)
in the
dipleg to produce a flow velocity less than bubble rise velocity.
Date Recue/Date Received 2023-03-28
11c
In some implementations, there is provided an improved vortex separation
system, comprising: a vortex separation system (VSS) separation chamber; a
stripper
positioned below the VSS separation chamber; at least one vent pipe through
which
stripper vapors exit the system; a baffle positioned between the VSS
separation chamber
and the stripper, the baffle being configured to divert vapor in the stripper
away from the
VSS separation chamber and into the at least one vent pipe; and a cyclone
separator
having a dipleg with a circumference that reduces flowing catalyst flux
(Ib/ft2/sec) in the
dipleg to produce a flow velocity less than bubble rise velocity.
In some implementations, there is provided an improved vortex separation
system comprising: a vortex separation system (VSS) separation chamber; a
baffle
located between a first stripping zone and the VSS separation chamber, thereby
forming
a stripping section that includes at least three zones that include the first
stripping zone,
a second stripping zone and a third stripping zone, the baffle being
configured to direct
stripper vapors that exit the first stripping zone and the third stripping
zone away from
the VSS separation chamber; and at least one vent pipe through which the
stripper
vapors that rise through the first stripping zone and the third stripping zone
and bypass
the VSS separation chamber are passed into a VSS riser.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG.1 is a schematic representation of a conventional high containment VSS,
stripper and primary cyclone system.
FIG.2 illustrates a steam/hydrocarbon voidage balance across the VSS
freeboard and a single stage reactor strippers.
FIG.3 shows the catalyst and voidage flow in a typical primary cyclone dipleg
for
either a DCC or VSS high containment system.
FIG.4 is a schematic representation of the "soft coke" reactions from the
catalyst's interstitial voids and surface into the reactor stripper's emulsion
and bubble
phase.
FIG.5 shows a modified higher containment VSS with 3 zone reactor stripping.
FIG.6 and 6A show a primary reactor cyclone with a modified barrel dipleg for
either a high containment DCC Rough Cut Cyclone or VSS application.
FIG.7 shows a modified higher containment VSS with a 3-zone reactor stripper
along with the modified primary cyclone barrel dipleg.
FIG.8 shows a modified higher containment VSS with a 3-zone reactor stripper
along with the modified primary cyclone barrel dipleg and packing in the
freeboard.
Date Recue/Date Received 2023-03-28
11d
FIG.9 shows a modified higher containment VSS with a crossflow 3 zone reactor
stripper along with the modified primary cyclone barrel dipleg.
FIG.10 shows a modified higher containment VSS with a crossflow 3 zone reactor
stripper along with the modified primary cyclone barrel dipleg and packing in
the
freeboard.
Date Recue/Date Received 2023-03-28
12
FIG. 11 illustrates a material balance on dry gas in a typical primary
cyclone dipleg for either a DCC or VSS high containment system.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT(S)
FIG.1 is a schematic representation of a typical high containment VSS
and reactor stripper configuration. Feed injection nozzles 15 inject feed into
a
flowing stream of catalyst. The resulting suspension reacts rapidly and flows
through a plug flow riser reactor 16 into the VSS separation chamber 1 where
most of the catalyst is separated. The catalyst then flows down into the
stripper section 23 along with some hydrocarbon underflow 21 that is still
contained within the catalyst voids. The majority of the hydrocarbon vapor
products and any entrained catalyst then pass through several single stage
cyclones 5 housed within the reactor vessel 27 for final catalyst separation
prior to entering the combined vapor product line 3. This entrained catalyst
then flows from the cyclones through diplegs 26 into the catalyst bed 19
located in the outer annulus section of the reactor vessel 27. The vapor flow
8 in the reactor vessel 27 contains some but not all of the entrained
hydrocarbons 25 flowing with this dipleg catalyst is ultimately vented from
the
reactor vessel via several vent pipes 7 back into the vapor transport line.
The
dipleg catalyst and any residual hydrocarbon underflow then passes into the
stripper section via a series of openings 11 within the VSS. All the catalyst
along with any adsorbed or entrained hydrocarbons enter the upper dense
bed stripping section 23. Stripping steam 13 is injected into the lower
section
of the stripper and rises counter-current to the downward flowing catalyst for
displacement of these adsorbed and entrained hydrocarbons from the
catalyst prior to it entering the regenerator via standpipe 17. The stripping
section 23 contains baffles or packing in order to facilitate effective
hydrocarbon removal. Excess steam and any displaced hydrocarbons leave
the upper section of the stripper through the dense bed freeboard section 29
prior to reentering the VSS 1. The main objectives in any high containment
system is operability, flexibility, with minimum underflow of reactor products
and efficient stripping of whatever underflow persists. In FIG.1, this would
Date Recue/Date Received 2023-03-28
13
involve minimizing the hydrocarbon flow in streams 21 and 25 along with a
low voidage, high percent steam, net catalyst flow to the regenerator through
standpipe 17. The following shows the significant impact high "J" values, the
stripper freeboard and the "soft coke" dry gas reactions are having on a
typical
commercial VSS, reactor stripper configuration. Unlike the Direct Coupled
System, the VSS stripper is connected directly to the VSS chamber via the
beds freeboard 29 which essentially contains a well mixed 100% reactor
effluent environment. In the DCC system the content in the reactor vessel is
essentially inert steam, cyclone dipleg underflow and cracked dry gas.
FIG.2 illustrates a steam/hydrocarbon voidage balance across the
VSS freeboard 29 and a single stage reactor stripper configuration. With the
typical FCC processing conditions listed in Table 6 and the corresponding
hydrocarbon, steam entrainment rates shown in Table 7 it clearly
demonstrates the significant impact the prevailing "J" values, freeboard
activity and the "soft cokes" cracked gas production has on the hydrocarbon
underflows 21 and overall stripping efficiencies.
TABLE 6
TYPICAL FCC PROCESSING CONDITIONS
BASE CONDITIONS
lbSTM/1000IbCAT 2.0
STRIPPING STEAM 1.4wt%ff
CAT/OIL 7.0
REACTOR TEMPERATURE 1000F
REACTOR PRESSURE 20PSIG
STRIPPER CATALYST FLUX 20Ib/ft2/sec
"J" VALUE 20Ib/ft2/sec
STRIPPER CSA 0.35ft2/Ibff
Catalyst Flowing Density 30Ib/Ft3
Catalyst Emulsion Density 50Ib/Ft3
Date Recue/Date Received 2023-03-28
14
"SOFT COKE" Dry Gas 0.81wt%ff
Production
Dry Gas Molecular Weight 9
TABLE 7
HYDROCARBON AND STEAM ENTRAINMENTS
STRIPPER/DIPLEG HCBN'S HCBN'S STEAM STM/HCBN
DENSITY ACF/IbFF Wt% FF ACF/IbFF Vol%
50 0.0973 2.16 0.3506 3.61
40 0.1326 2.93 0.3506 2.64
30 0.1907 4.23 0.3506 1.84
20 0.3073 6.81 0.3506 1.14
FIG.2 VOLUME FLOWS:
Steam to stripper = 0.3506 ACF/IbFF
Catalyst Voids = 0.0973 ACF/IbFF @ 50Ib/Ft3 and 7 Cat/oil
= 0.1907 ACF/IbFF @ 30Ib/Ft3 and 7 Cat/oil
Using e = (1-pB/164)
VOIDAGE BALANCE (ACE/lb FF):
0.3506 +2(0.1907) + 0.406 = 2(0.0973) + 0.3506 + 0.406+ 2(0.0973)
VOIDAGE PERCENT HDROCARBON BALANCE:
2(0.1907)100+ (0.406)100 = 2(0.0973) X+ 0.3506X +0.406X+
2(0.0973) X
X = 69.17 Vol% HCBN's
Without the "J" flow and dry gas reactions a single theoretical
backmixed stage, with a catalyst inlet density of 30Ib/ft3 and a 1.84
steam/hydrocarbon volume ratio, would have a much higher stripping
efficiency and produce a significantly lower hydrocarbon mix.
X= 35.20 Vol% HCBN's
The "J" bubble wake and freeboard entrainment not only effects the
stripper efficiency; but, more importantly, the Wt % of riser products or
underflow that's going into the stripper, where it will rapidly undergo the
undesirable secondary reactions that were illustrated in Table 1, Table 7,
Date Recue/Date Received 2023-03-28
15
shows that with a 30Ib/ft3 flowing density into the stripper and a
20Ib/ft2/sec
"J", the VSS underflow, stream 21 in FIG 1, increases from 2.93 to 8.46wt% if.
Also, the dry gas production that is being generated from the "soft
coke" and additional conversion reactions, now combines with the steam to
produce a much higher superficial velocity of 2.16Ft/sec in the upper section
of the commercial stripper.
STEAM VELOCITY 1.0 Ft/sec
LIGHT GAS VELOCITY 1.16Ft/sec
COMBINED VELOCITY 2.16Ft/sec
In an actual commercial stripper, operating at these much higher
combined superficial velocities in the upper stripper section, the bubbling
bed
model no longer applies. The upper bed has become very turbulent with an
extensive freeboard region and high catalyst entrainment into the lower
section of the VSS. If the system was to be designed and operated at even
higher catalyst fluxes, like 301bif12/sec, and some are, the combined
superficial velocity would be 3.24ft/sec. As shown in FIG.1 there is still a
need
to account for the primary cyclone dipleg underflow 25.
FIG.3, shows a schematic representation of a typical primary cyclone
5 and dipleg 26. The catalyst flow in the upper section 53 is streaming flow.
In commercial operations dipleg catalyst fluxes run between 50-80Ib/ft2/sec.
In the sealed lower section 55, these fluxes unfortunately generate emulsion
phase velocities that are greater than the maximum bubble riser velocity.
Since UL or Ue > lib, the gas and catalyst flow 57 become co-current. This
produces a flowing medium with a high voidage, low density that results in
further highly undesirable hydrocarbon underflow 25.
Unfortunately, in the VSS reactor stripper system the "J" catalyst voids
will always leave the stripper 23 via the freeboard 29 partially stripped but
return 100% loaded with riser products as unwanted under-flows 21 or 25.
With the prevailing upper stripper velocities, "J" values and freeboard
activity the VSS catalyst separation efficiency will probably drop
significantly
from 95% to say 90%. At 90% separation efficiency and a 20Ib/ft3 dipleg
density, this would generate a 0.681wt%ff hydrocarbon underflow in stream
25, bringing the total underflow in streams 21 and 2510 9.141wt%ff.
Date Recue/Date Received 2023-03-28
16
Unlike the DCC, the VSS reactor vessel is essentially 100% full of
relatively stagnant, high molecular weight, hydrocarbons with little or no
steam. A perfect environment for coke formation.
The present day high containment VSS stripper configuration shown
in FIG.1 has been determined to have significant design issues. The following
simple, yet highly effective, design modifications disclosed herein should
produce significantly high riser product containment and thereby generate
further desired selectivity improvements, higher stripping efficiencies, and
greater operating flexibility with the elimination of any future potential
coke
formation within the reactor vessel.
The design modifications can vary but most can be easily incorporated
into existing VSS units and eliminate all of the issues discussed throughout
this review. The three zone stripper modifications actually utilize the
significant "soft coke" dry gas production to the reactor stripper's
advantage.
FIG.4 shows a schematic representation of the "soft coke" reactions
from the catalyst's interstitial voids and surface into the strippers emulsion
and bubble phase. These products, are continuously being produced from the
flowing catalyst. Since they are primarily hydrogen and methane, they have
extremely low molecular weights and are essentially inert. With the migration
of these products from the catalysts inner pores and surface area, through
the emulsion phase to the bubble phase, they offer an excellent initial
stripping medium for the higher molecular weight, riser product, underflow
material. The continuous but declining production of these inert hydrocarbons
from this source makes the overall catalyst residence time distributions in
the
VSS stripper critical in achieving high overall stripping efficiencies.
FIG. 5 shows the invention of a commercial design for a 3 zone VSS
stripper configuration. The design essentially eliminates most of the negative
issues discussed throughout this review. This invention includes a baffle
addition 35 that segregates the stripper into three zones 31, 33 and 23. The
first zone 31 is a "cracked gas" stripping zone with a low 0.2 to 0.5 Ft/sec
superficial gas velocity. The only gas entering this zone is the underflow 21
along with the "cracked gas" being produced. The freeboard activity 29 and
the "J" values are therefore low. The average catalyst residence time, 10 to
Date Recue/Date Received 2023-03-28
17
20sec's at 1000 F reactor temperature, is sufficient to complete over 50% of
the "soft coke" reactions. A small amount of additional steam 37 is available
to maintain minimum bubbling velocity if required. The partially stripped
catalyst then flows into the, higher residence time, transition zone 33 in the
reactor vessel 27. Where any residual "soft coke" reactions are completed
and stripped along with most of the remaining riser product underflow using
the gas from the main stripper 23. The catalyst then enters the existing main
stripper 23 for final "cracked gas" displacement. This system could also
include stripper packing at the entrance of zone 1 to further suppress
freeboard activity and "J" values. Such design modifications will reduce the
net product underflow from 8.46wt%ff to 2.50wt%ff and thereby achieve
significant additional improvements in desired product yields, selectivity and
overall stripper efficiency.
FIG.6 and 6A shows the invention of a primary reactor cyclone 5 with
a modified barrel dipleg 37 for either a high containment DCC, Rough-cut
Cyclone or VSS application. The invention has an outer sleeve addition 37
that extends the cyclone barrel down to the stripper bed 19 and shortens the
previous dipleg to just beyond the vortex. Any excess gas 39 can now be
vented back into the cyclone 5 or as in Fig.6A into the reactor plenum 45 via
additional vent pipes 46. In the DCC or Rough Cut Cyclone systems the vent
gas could be discharged to any location downstream of the primary cyclone
separator. The vortex and high cyclone separation efficiency is still
maintained. The catalyst flux, in the modified barrel dipleg 37, will be
reduced
to less than 20Ib/ft2/sec; thereby producing a countercurrent bubble/emulsion
phase flow 41, since Ue is now less than Ub. The flowing catalyst density in
the modified dipleg will approach minimum fluidization density, which will be
close to 50Ib/ft3. This reduces the underflow 25 to 0.108wt%ff. The "soft
coke"
cracked gas production will always maintain minimum fluidization and achieve
some displacement of riser products.
FIG.7 shows the combined invention of a 3-zone stripper and barrel
cyclone dipleg. The total combined underflow with these two inventions as
now been reduced from 9.141wt%ff to 2.27wt%ff. This type of cyclone dipleg
modification can also be used in either a Rough-Cut or +1- pressure DCC
Date Recue/Date Received 2023-03-28
18
application. The potential coke formation in the reactor vessel has now been
eliminated. Assuming only 50% conversion of the "soft coke" in zone 1, Table
8 shows the new volume percent vapor composition in the reactor vessel. It's
dropped from almost 100vol% reactive hydrocarbons, with low flow in FIG.1,
to 6.3vo1% reactive hydrocarbons, with high flow, in the new FIG 7 design.
TABLE 8
VAPOR COMPOSITION IN RX VESSEL
Mol Wt Wt% Vol% Rx Vol% In Rx
FE Eff Vessel
STRIPPING STEAM 18 1.40 7.77 59.64
CRACKED DRY GAS 9 0.41 4.44 34.08
HCBN UNDERFLOW 100 0.81 0.81 6.28
With such a vapor composition and flow, vapor stream 8 now as the
potential for a separate cyclone/condensation and recovery system:
STEAM 59.6Vol%
H2/CH4 34.1Vol%
UNDERFLOW 6.3Vol%
TOTAL FLOW 13.028 Vol% Rx effluent
F1G.8 shows the combined invention of a 3-zone stripper and barrel
cyclone dipleg along with the addition of stripper packing 36 at the entrance
of the stripping zone one 31 to further suppress freeboard activity and "J"
values.
FIG.9 shows a combined 3 zone crossflow stripper and barrel cyclone
dipleg invention. The two baffles 41 and 42 are arranged to initiate catalyst
crossflow in zone one 31 for better residence time distribution and more "soft
coke" reactions per linear foot of stripper.
FIG.10 shows a combined 3 zone crossflow stripper and barrel cyclone
dipleg invention along with the addition of stripper packing 40 at the
entrance
of the stripping zone one 31 to further suppress freeboard activity and "J"
values.
Other embodiments of this invention are:
Date Recue/Date Received 2023-03-28
19
If a high ZSM-5, petrochemical operation is desired, one should still
use these design innovations but increase the steam flow 38 to zone one 31.
Thereby, in a controlled fashion, increasing the "J" valve and lowering the
LHSV in the VSS chamber and increasing the catalyst entrained to the
primary cyclone system 5. This would effectively increase the catalytic
severity and light olefin yields for a given reactor temperature. The "cracked
gas" from the reactor vessel could still be recovered separately and sent to a
combustion chamber in the regenerator to augment the enthalpy coke.
The catalyst residence times in stripping zones 31 and 33 can be
controlled via the reactor level 19 and the spent slide valve opening. The
residence time in zone one 31 is now particularly significant in that it will
set
the degree of dry gas production from the "soft coke" which will also vary
with,
and be dependent on, reactor temperature. This improved RTD in the three
zone stripper also leads to lower regenerator SOx levels via the increased
conversion of "soft coke" sulfur to hydrogen sulfide followed by its higher
removal efficiency from within the catalyst voidage going to the regenerator.
In summary, it will be understood by those skilled in the art that the
present invention shows how the addition of stripper baffles 35 or 41 and 43
and a modified, low catalyst flux, primary cyclone dipleg 37 can significantly
reduce the under-flow of reactor riser products 21, and 25 into an existing
VSS
stripper and reactor vessel. The addition of these baffles establishes three
distinct stripping zones 31, 33 and 23 with no backmixing of catalyst between
the zones. Zone 31 is a low velocity bubbling bed that uses the "soft cokes"
light cracked products to displace and strip the VSS underflow 21. The "soft
coke" reactions are completed in zone 33 prior to a conventional stripping in
zone 23. This increased containment and multi zone stripping will produce
higher desired product selectivities, improved stripping efficiency, lower
regenerator SOx levels and a stripper vent gas 8, that continuously flows
through the reactor vessel 27, that now has a significantly lower coke forming
potential due to its low concentration of ethylene and higher molecular weight
material, shown in table 8, that could, if desired, be recovered separately
from
the primary riser products. These stripper baffles could be used with or
Date Recue/Date Received 2023-03-28
20
without the packing 40 at the freeboard zone 29 or the modified diplegs 37.
The modified, low catalyst flux, primary cyclone dipleg 37 can be used
independently on either a DCC or Rough Cut cyclone system to significantly
reduce their underflow of reactor riser products 25, shown in Table 1, into
the
reactor vessel. The catalyst bed level in zone 31 can be further controlled to
achieve the desired residence time and more conversion of the "soft coke"
material to hydrogen and methane. A high ZSM-5, petrochemical type
operation can now be established by controlling the "J" values, LHSV in the
VSS chamber and increasing the catalyst to oil ratio in the vapor line to the
primary cyclone. With a separate recovery and treatment, another option for
the hydrogen and methane "cracked gas" 8 from the reactor vessel 27 could
be in a combustion chamber in the regenerator to augment the enthalpy coke.
Date Recue/Date Received 2023-03-28
