FCC Profitability Assessment via Advanced Modeling

7/23/2019 FCC Profitability Assessment via Advanced Modeling

http://slidepdf.com/reader/full/fcc-profitability-assessment-via-advanced-modeling 1/10

FCC profitability assessment viaadvanced modelling

The FCC unit at Puget

Sound was originally

installed in 1958. Over theintervening years, the unit was

modied several times. The last

revamp in 1999 installed a new

external riser as well as the

current reactor vessel and with

the unit congured as a two-ves-

sel side-by-side FCC unit with

an external riser. The reactor

vessel contained the reactor

cyclones with a stripper incor-

porated within the bottomsection of the reactor vessel.

The basic layout of the FCC unit

is illustrated in the process ow

diagram used in the project

scope (see Figure 1).

During the renery’s 2009

turnaround, Shell determined

the unit’s riser refractory to be

at end of life and in need of

renewal. Replacing the refrac-

tory would have been a majoreffort, potentially extending the next turnaround

in 2014. Therefore, the company decided that

replacement would be a more cost-effective unit

upgrade and life extension strategy.

Technical review of the riser determined that

the riser residence time was too short

for the current demand. Replacement of the

riser simply to increase riser residence time

would not have met the required return on

investment criteria. However, once the company

Sayantan Chatterjee, Cian Carroll, Michael Basden and Kevin Kunz Shell Global Solutions (US) Inc Charles Burton MOTIVA RefiningSteve Nelson Shell Oil Products US

decided to replace the riser refractory, the

incremental engineering and fabrication costs

were low enough to justify the increase in riser

residence time.

Likewise, the as-found condition of the strip-

per in the 2009 inspection indicated that

replacement would be cost effective as an alter-

native to the required repairs that would have

had to be scheduled for the 2014 turnaround.

For example, the existing stripper was a disk and

www.digitalrefining.com/article/1001180 PTQ Q4 2015 1

Process and computational fluid dynamic modelling were applied to validatethe technology used in an FCC unit revamp at Shell’s Puget Sound refinery

To COBsand WGS

Lift airex-LAB

Combustionair to regen

To MF

Figure 1 Process flow diagram of Shell’s Puget Sound FCC unit (revampproject scope items are outlined in red)

http://www.digitalrefining.com/article/1001180






donut design, which was prone to ooding at

high catalyst circulation rates. As with the riser,

the incremental cost of replacing the stripper with a new Shell design was justied when

compared to stripper repair and refurbishment.

FCC unit turnaroundScope and incentives

Project premise: driver for change

The 2009 equipment inspection revealed the

riser refractory to be at end of life condition and

anticipated repairs would result in a signicant

extension to the renery’s planned 2014 turna-

round duration. Similarly, the amount of repairto the stripper disk and donuts, combined with

the health, safety and environmental risks of

extremely tight working conditions, could also

not be completed within the planned turnaround

window. Thus, replacement of the riser and disk

and donut stripper became the preferred and

more cost effective alternative to repair. Once

the decision to replace the riser and stripper was

made, the incremental cost of the upgrade versus

an in-kind replacement was easily justied based

on the incremental margin improvement.Originally, an expansion joint was not included

as part of the regenerated catalyst standpipe

design. Increasing the riser volume increased

system stiffness. The riser upgrade resulted in

system rigidity under start-up and shutdown

conditions, which contributed to both high

regenerator nozzle stresses and regenerated

catalyst slide valve sticking. As a result, an

expansion joint was added above the regener-

ated catalyst slide valve. At the onset of the

2 PTQ Q4 2015 www.digitalrefining.com/article/1001180

revamp scope denition, instal-

lation of an expansion joint was

in the project scope for the

spent catalyst standpipe as well.

However, after applying value

engineering practices, it was

removed with just minor regen-

erator nozzle reinforcement,

resulting in signicant savings.

The regenerated catalyst slide

valve had a history of sticking,

identied as an on-going relia-

bility threat for the unit, and

therefore was replaced with a

new valve incorporating the best

of Shell’s current FCC technol-

ogy design elements.

The company justied the incremental cost of

the riser and stripper replacement partially on

the basis of the anticipated margin improvementfrom the unit revamp. Since the riser and strip-

per replacement were primarily justied to

minimise turnaround duration, the incremental

cost for upgrading this equipment was relatively

small, amounting to some additional engineering

and fabrication cost.

Project planning and execution

The project was kicked off in early 2011 and

involved Puget Sound renery staff and Shell

Global Solutions jointly working through theoptions evaluation, economic reviews, and

preliminary project scoping. Shell utilised a third

party rm for detailed engineering and project

management. Concurrently, the same rm

managed a parallel upgrade project for the FCC

unit’s instrumented protective function (IPF). A

fabrication contractor was then brought in as

equipment fabricator, with eld construction

completed by a construction contractor as part

of the overall turnaround. Part of the project’s

execution strategy was to engage the construc-tion contractor during revamp scope design to

ensure planning for constructability and

sequencing of delivery and lifting within the

overall turnaround schedule.

Overall, the project was well developed and

executed. All potential challenging issues were

quickly identied and resolved with minimal

impact to the overall project or the turnaround.

An outstanding team effort was involved in the

safe and successful completion of the project.

Figure 2 Installing the new stripper into structure and its final position atShell’s Puget Sound FCC unit







Puget Sound FCC unit todayToday, Puget Sound’s FCC unit operates on aver-age at rates up to 52000 b/d, processing

vacuum gasoil and unhydrotreated heavy coker

gasoil. The unit includes Shell’s technology

designs installed during the latest revamp in

March and April of 2014. The present

unit conguration following the revamp includes

the following plant upgrades (see Figure 2):

• New riser of larger diameter, new HIB rings to

inner lining and increased residence time

• New J-Bend

• New feed nozzles including associated piping• New stripper, lengthened and upgraded from

disk and donuts to Shell’s PentaFlow bafe design

• New regenerated catalyst slide valve incorpo-

rating current best practice design

• Regenerator standpipe expansion bellows to

alleviate stresses

• Upgraded IPF system with new instrumenta-

tion and logic system

• Completion of new DCS cutovers.

The revamp was successfully completed on

time, within the turnaround window and theplanned budget. Post turnaround, the unit has

operated very well, realising an estimated benet

matching the expected cost versus forecasted

gains as the business basis for the revamp.

New hardware installedShell has led active research in FCC technology

development for over seven decades since its rst

FCC unit start-up in 1942. As an operator, the

company has safely and successfully operated


FCC units for over 1200 unit-years, and has

designed 33 grassroots FCC units. Since 2006,Shell Global Solutions has successfully completed

over 40 FCC unit revamps across Shell owned, JV

operated and third party licensed reneries. The

portfolio of Shell’s FCC technologies that have

been implemented in these revamp projects

include: reactor and regenerator vessels with

internals; feed injection nozzles; close-coupled

reactor and regenerator cyclones; catalyst circula-

tion enhancement technology (CCET); catalyst

stripper PentaFlow bafes; air grid and spent

catalyst distributor (SCID); and third stageseparators (TSS). Most of these FCC equipment

specic technologies were incorporated in a major

riser replacement revamp project in 2012 at

Shell’s Deer Park renery.1,2 Among the technolo-

gies listed above, some were already present in

Puget Sound renery’s FCC unit prior to the turn-

around (for instance, reactor and regenerator

cyclone technology and air grid). Most of the

remaining technologies were incorporated during

the 2014 riser and stripper revamp project

described in this article.

Riser Internals

Shell developed riser internals known as HIB

rings, which have been installed and operation-

ally validated in several units, including the

Puget Sound FCC unit (see Figure 3a). The HIB

rings deployed in the riser reduce the catalyst

slip or back mixing and help the catalyst particles

and hydrocarbon vapours to reach ideal plug-

ow conditions. HIB rings are well documented

20

10

R i s e r h e i g h t , m

0

0 50 100 150 2 00

Pressure, mbar

2

1

D i m

e n s i o n l e s s

v e l o c i t y

0

Centreline Wall

Riser location

Figure 3 (a) A schematic illustration of Shell’s HIB rings installed in a riser and (b) validation studies showingpresence of HIB rings lead to superior flatness of the radial catalyst velocity profile, and (c) reduced catalyst slip andpressure drop over the riser

a b c







for delivering benets such as improved overall

unit performance plus greater exibility to oper-

ate over a wide range of conditions. Independent

benchmarking studies show that HIB rings offer

more uniform catalyst distribution, more even

velocity prole (see Figure 3b), and better catalyst

mixing, which consequently produce lower pres-

sure drop (see Figure 3c) and enhanced product yields when compared to an open riser.3

For the Puget Sound renery FCC revamp,

CFD modelling tools were used to illustrate and

validate the enhanced benets of HIB rings

installed in FCC risers.

Feed injection nozzles

Shell’s FCC feed injection nozzles have a unique

design where atomisation occurs just before

nozzle exit to create the nest drop sizes with

the highest energy efciency. This atomisationtechnique produces small droplets at the nozzle

tip and avoids droplet re-agglomeration and

formation of stratied, two-phase ow. Shell

offers single-wide-slot (SWS) feed nozzles, which

can usually be customised over a wide operating

range for greater exibility (see Figure 4a).

Furthermore, the newest design offers a single

ow exit with improved wear reliability and

robustness compared to past Shell designs. In

addition, the nozzles are accompanied by a

proprietary, shield-like shroud that covers mostof the nozzle tip surface and is incorporated in

the latest nozzle designs (see Figure 4b). The

shroud allows feed injection to pass through it

and further minimises tip erosion from owing

catalyst in the riser. Shell nozzles have been

installed in scores of units, and in many sites

they have been retained for multiple runs. Puget

Sound installed the latest version of the feed

nozzle shroud in 2014; this does not mix the

steam and feed with catalyst until it exits the

protective shroud. This is

expected to further improve the

life of the feed nozzle as well as

reduce exit velocity for lower

catalyst attrition at the nozzle

mix zone of the riser.

In addition to using feed

nozzles in actual operating

assets, Shell benchmarks its

feed nozzles in cold ow condi-

tions at its in-house FCC

technologies research facility in

Houston, Texas. The FCC feed atomisation test-

ing facility has been in operation for many years

and is the main vehicle to test and develop full-

scale feed nozzles spraying a mixture of air and

water (simulating steam and oil) under ambient

conditions. Using an existing, advanced method

to measure droplet sizes, the company renes

and improves the feed nozzle’s development,coupled with direct access to the renery data

from its operating assets.

The key objective is to minimise the droplet size

exiting the feed nozzles as it increases the total

contact surface area in feed, reduces the feed

vaporisation volume in the riser, and further

minimises coke yield and coke fouling. Shell eval-

uates feed nozzles using a novel technique that

captures the full droplet size distribution, deter-

mines the associated droplet size characteristics

and distinguishes the formation of large drops inthe spray. This technique helps in developing

newer feed nozzle designs with rigorous testing at

the in-house facility prior to commercial deploy-

ment and subsequent validation by running in the

actual operating units.

PentaFlow stripper

Shell developed and implemented structured

packing for FCC strippers in the 1990s. While

the company experienced good initial perfor-

mance, the packing was found to progressivelycause plugging, limited catalyst circulation and

led to unplanned shutdowns. To address the

reliability concerns in structured packing in FCC

applications, Shell developed an improved strip-

per internals design with high ux PentaFlow

bafes combining superior performance, higher

reliability, improved mechanical strength, and

easier access for equipment maintenance at

lower cost (see Figure 5). PentaFlow bafe tech-

nology has been successfully implemented in


A B

Figure 4 Shell’s (a) feed nozzle design;4 with a (b) sacrificial shroud5







over a dozen units globally,

including at the Puget Sound

FCC unit, and its application is

documented for the resid FCC at

Shell’s Pulau Bukom renery.6

Prior to the unit revamp at

Puget Sound, the stripper vessel

had conventional disk and donut

bafes installed in the unit

where the steam often got

trapped underneath the bafes

and led to poor steam distribu-

tion. On the other hand, the

PentaFlow bafe design redis-

tributes the stripping steam

through holes perforated along

the bafe length to guarantee

optimal stripping performance

throughout the run. In fact,

PentaFlow bafes maintainhigher performance even at high

catalyst ux rates in comparison

to conventional disk and donut

design.7 In addition to the supe-

rior reliability from these bafes, the design

offers exceptional stripping quality, reducing the

hydrogen in coke to 6 wt% or less. At Puget

Sound, the old stripper with disk and donuts

would ood at high circulation rates. The new

stripper with PentaFlow technology is not

expected to show the same behaviour. This, inturn, provides an opportunity to change catalyst,

lower activity, increase catalyst circulation rates

and provides more exible oper-

ating options to the unit.

Through CFD simulations, Shell

compared and demonstrated the

performance of PentaFlow strip-

pers versus conventional disk

and donuts.

Catalyst circulation enhancementtechnology

Traditionally, catalyst circulation

at the inlet of both the regener-

ated and spent catalyst

standpipes is enhanced using

hoppers, by minimising the

amount of excess gas bubbles

entering the standpipes.

However, conventional hoppers

of a given capacity are only

effective over narrow operating

ranges and inefcient in elimi-

nating all gas bubbles. Shell

developed a new CCET,

deployed in many units, which

prevents gas bubbles from

entering the standpipe in the

rst place (see Figure 6).

The device is responsible for

efciently degassing the cata-

lyst, removing excess gas and re-

introducing a controlled

amount of gas with uidised

catalyst into the standpipe to

optimise both catalyst ow and

high pressure gain in the stand-

pipe.8 This further improves

catalyst circulation rates and

offers higher reliability and ex-

ibility in performance.

Existing hardware from pre-turnaround eraThe 2014 Puget Sound FCC unit

turnaround included selected Shell technologies

in the project scope. Some of the other Shell

hardware options already existed from the

pre-turnaround era at Puget Sound. This article

briey reviews these technologies and discusses

use of CFD modelling tools to validate the perfor-

mances of these equipment technology additions.

Reactor and regenerator cyclones

Cyclones were found to be a

chief cause of unscheduled shut-

downs and slowdowns in the

1980s. To address this reliability

issue, in the early 1990s, Shell’s

FCC R&D initiated an advanced

cyclone design and development

programme, and reduced

unplanned shutdowns by 90% inless than a decade. Some of the

unique features in a Shell

cyclone include its close-coupled

design with a gas outlet tube, a

coke catcher (in a reactor) or a

trash guard (in regenerator

applications), and a vortex stabi-

liser. The close-coupled design

optimises performance,

improves reliability and facili-

www.digitalrefining.com/article/1001180

PTQ Q4 2015 5

Figure 5 Generic design of stripperinternal with high flux PentaFlowbaffles

Figure 6 Catalyst circulationenhancement technology





tates easy maintenance and repair. The special

gas outlet tube design and the coke catcher/trash

guard minimise coke and debris falling into the

cyclone and prevent plugging of the diplegs,

enhancing reliability. A vortex stabiliser is used to

minimise erosion between the bottom of the

cyclone barrel and the top of the dipleg.9 It also

reduces the threat of efciency loss from high

dipleg backup, a condition resulting from high

bed level operations. Typically, the secondary

cyclones employ the vortex stabiliser for improved

cyclone reliability. Shell cyclones offer higher

separation than conventional cyclones of similar

size and have been installed in about 20 units.

Two primary cyclones are deployed in the

Puget Sound FCC reactor, connected with four

secondary cyclones using an internal primary

cyclone plenum. The third-party reactor cyclones

are now effective and reliable after modications

were made.The Puget Sound FCC regenerator houses nine

primary cyclones and nine secondary cyclones.

The 1999 regenerator cyclones were replaced in

2004 with Shell designed cyclones having vortex

stabilisers in the secondary cyclones. The catalyst

loss rate was reduced from a design of 4 t/d with

the old cyclones to a loss rate of 1 t/d with the

new cyclones. The previous cyclones were not

capable of making a ve-year run while the new

Shell cyclone design does so. Thus, they were not

included as part of the scope for the 2014 revamp.

Air grid and spent catalyst distributors

The key to good mixing in a regenerator vessel is

to ensure good air and spent catalyst distribu-

tion. Shell develops air grid distributors and

spent catalyst inlet devices (SCIDs) to address

the need for effective mixing. These air grids use

refractory lined, two-stage nozzles that retain

excellent conditions even after a runtime of ve

years.

SCIDs have been installed to deliver increasedstability and reduced after burn. Currently, Shell

uses a CFD code that has been validated both

experimentally and through process data to

determine the optimum location, sizing and

shape of such distributors.

Post revamp start-up: reliability assurancesand concernsNo issues were found on start-up with the new

equipment installed during the revamp. The unit

has operated well, meeting site expectations in

terms of feed rate, riser temperature, and NOx

emissions. Even though the unit coke burn

capacity was effectively reduced by new limits on

NOx requiring a lower CO in the regenerator ue

gas, the apparent conversion has remained at or

above pre-turnaround levels.

Post-turnaround, CFD was used to model the

performance of SCID to help understand catalyst

distribution in the regenerator, more specically

to determine the effect of the SCID on cyclone

loading and couple this to cyclone wear observed

during the turnaround.

SHARC modelling to simulate unitperformanceOverview

Shell Advanced and Rigorous Cat Cracking

Model (SHARC) is Shell Global Solutions’ uid

catalytic cracking simulation and optimisationtool programmed in AIMMS, a mathematical

modelling platform.10 SHARC can be used as a

standalone program for process studies, unit

optimisation and troubleshooting, or in inte-

grated modelling environments such as renery

simulators, real time optimisation applications,

and renery economics and scheduling applica-

tions. The model is easy to set up, use and has

several unique features. First, it encompasses

understanding of FCC process fundamentals

developed through R&D programmes, designexperience, and pilot plant and commercial unit

operations. Secondly, Shell developed and vali-

dated SHARC using a database of thousands of

pilot plant and commercial precision test run

datasets. Current R&D efforts seek to expand the

scope of SHARC and to further develop it by

incorporating newer manufacturing technologies

to meet demand in fuels and petrochemicals.

Thirdly, the model can accurately deal with a

wide range of feedstocks such as crude sources,

vacuum gasoil (VGO), residue, straight run andpre-treated cracked hydrocarbons. Fourthly, the

model accurately predicts operational changes

due to small perturbations in input data away

from known or well-established commercial

operation data, which makes the model impor-

tant in on-line optimisation and LP modelling.

Finally, the model has an open-equation archi-

tecture which makes it easy to maintain, develop

and integrate with other process models and

simulation tools.








Post-turnaround model tuning

Shell maintains and manages

SHARC models for all the FCC

units it supports. These unit

specic models are referred to

as ‘base cases’. Post-turnaround,

Puget Sound collected test runs

to tune the unit specic SHARC

base case. Upon completion, the

model was updated by Shell

Global Solutions and approved

by the site before being incorpo-

rated into the online optimiser

and LP model. The new base

case (post-turnaround) clearly

characterises the current opera-

tion more closely than the

preceding 2012 base case

(pre-turnaround). SHARC was used for the

Puget Sound FCC unit revamp to evaluate pre-and post-turnaround operation and quantify the

yield shifts and associated margin benets from

the revamp.

Pre- and post-turnaround yield shifts and margin benefits

The post-turnaround model shows improved

yields and margin benets reecting installation

of the new hardware. Comparing the pre- and

post-turnaround models at constant feed, cata-

lyst and operating conditions shows a yield

improvement with an estimated positive margin benet and signed off by the renery (see Figure

7). These margin benets were in agreement with

the proposed margin gain dened in the scope.

Minor corrections were made to the pre-turna-

round model prior to comparing it with the post-

turnaround model. Both models were optimised

and compared at constant feed (rate and proper-

ties) and catalyst addition (and metals),

constraining the model at the same regenerator

conditions (CO in ue gas and air ow rate) into

two optimisation cases. Case 1 included varyingcoke burn at constant riser outlet temperature

(see Figure 7a), and Case 2 varied riser outlet

temperature at constant coke burn (see Figure

7b). Simulations show increasing margin with

increasing coke burn and riser outlet tempera-

tures, and the delta margin benet was

considerably positive estimated using the new

model. Testing and comparing the performance

of the pre- and post-turnaround models in the

online unit optimiser showed the same trends

that the standalone analysis predicted for

projected unit operation swings and margin.

Prediction, assessment and validation of FCC unit revamps

One of the main capabilities of the SHARC

model is that the post-turnaround model can be

used to simulate improved individual yields as a

result of the equipment upgrade. The differential

in model predictions at Puget Sound’s FCC unit

suggests the benet was obtained from the hard-

ware changes during the recent revamp. The

model is then used to quantify the unit perfor-

mance benets as a result of these hardwarechanges. The key nding of such modelling is

used to assess actual margin uplift at Puget

Sound FCC unit revamp and was found to be

slightly higher than that proposed in the project

scope.

A series of data analyses was performed during

comparison of the pre- and post-turnaround

models. For example, SHARC can calculate aver-

age offsets, where offset is the difference

between model predictions and actual plant

measurements. For Puget Sound, the modelcompared average offsets between the pre- and

post-turnaround models. This analysis can be

used further to evaluate the yield and margin

shifts attributed to the combined hardware

upgrades implemented during the revamp at

Puget Sound FCC unit. Comparing such analysis

with similar studies completed at other units

may be used to parse impacts due to individual

hardware components (feed nozzles, riser resi-

dence time, and so on).


Pre-TA Post-TA

M a r g i n

B

Base Base+30

Riser outlet

temperature, ºF

M a r g i n

A

Base−5 Base+5Base

Coke burn,

Mlb/h

Figure 7 Model comparison showing pre- and post-turnaround models atsame feed (rate and properties), catalyst addition (metals), and regeneratorconditions (flue gas composition and so on) across a range of (a) coke burnand (b) riser outlet temperatures







CFD modelling to showenhanced solid flow in newhardwareComputational uid dynamics

has evolved dramatically since

the development of the rst nite

volume models in the 1980s. It is

only with recent advances in

numerical techniques that

multiphase systems have begun

to be accurately solved on a large

enough scale to be applied in

industry. For example, a typical

FCC unit contains upwards of

1015 particles. The modelling code

of choice at Shell for FCC hybrid

multiphase CFD is the CPFD

Barracuda software package. As

CFD used in this manner is a

relatively new approach, initialstudies focused on applying these

techniques to both experimental

and site problems that can be

validated objectively against

existing data. Given the successes

in recreating operating data,

Shell now uses CFD modelling in

support of hardware design and

scale-up.

Stripper internalsCFD simulations were performed on two different

stripper geometries to evaluate the benet of

PentaFlow bafes over the previous disk and

donut bafes installed at Puget Sound. Designs

are compared on the basis of particle residence

time and gas-solid contacting.

From the CFD simulation perspective,

PentaFlow bafes offer the following benets:

• Less inactive volume (see Figure 8); disk and

donut bafes have signicant volumes under

bafes where catalyst and steam do not contact.Elimination of these regions allows for more

steam and hydrocarbon mixing, increasing the

efciency of the stripper

• Less prone to plugging and ooding when

compared to disk and donut bafes

• More efcient operation under higher catalyst

uxes than disk and donut bafes

• Improved distribution of steam and catalyst

demonstrated by near-uniform bed density

throughout the stripper.

HIB rings

FCC risers feature complex

internal ow hydrodynamics.

Such core annulus ows

comprise peak centreline uid

velocity coupled with slow or

reversed ow at the riser walls.

HIB rings, designed to reduce

the wall backow in the riser

while improving the gas-solids

contacting, are key to driving

effective gas-solids contacting in

the riser.

From this standpoint, HIB

rings serve several benets to

the riser system, demonstrated

through CFD computations:

• Improved gas-solids contact-

ing in the riser

o More uniform velocitydistribution (see Figure 9)

o More uniform catalyst

density across the riser

• Improved riser performance at

reduced feed rates

• Identical or lower riser pres-

sure drop due to elimination of

riser backow, especially under

reduced rate conditions, but

with improved riser performance

• Decreased standard deviation in both gas andcatalyst residence time in the riser.

Generic spent catalyst inlet device

Recently, Shell has performed extensive CFD

modelling work at other FCC units (not applica-

ble to Puget Sound) to examine the performance

of different types of spent catalyst inlet devices

(SCIDs). To test the CFD modelling method, a

unit with ski-jump style inlet device was

modelled (see Figure 10). Prior radioactive tracer

experiments on this unit provided baselineexperimental data for comparison against the

model results. By reproducing the experimental

outcomes in CFD (see Figure 10), a baseline

model is established, which can be used for

future modications and design analysis.

The CFD results illustrate the motion of spent

catalyst upon entering the regenerator vessel via a

tangential inlet SCID (see Figure 10). The simula-

tion images appended to the vessel drawing show

the progress of the resulting spent catalyst swirl;


Particle volumefraction

Disk and doughnutpre-turnaround

PentaFlowpost-turnaround

0.6

0.5

0.4

0.3

0.2

0.1

0

Figure 8 Particle volume fractionsfor the stripper pre-turnaround

(left) and post-turnaround (right).Disk and donut baffles showsignificant areas of inactive volumeunder the baffles where steamand FCC catalyst cannot contact.Furthermore, the disk and donutbaffles were prone to flooding, withcatalyst de-fluidisation observed ontop of the disk sections. PentaFlowbaffles showed no such weaknesses







each snapshot is taken at the

corresponding breakthrough

time as measured experimentally

using radioactive probes.

Catalyst enters the regenerator

and travels in a counter-clock-

wise direction around the

circumference of the regenera-

tor. These results show a very

uneven particle residence time

across the regenerator. Part of

the area on the opposite side of

the SCID sees no catalyst parti-

cles at all after 30 seconds.

Ideally, the vessel would recreate

the residence time distribution

of a well-mixed reactor, wherein

the volume averaged residence

time distribution is identical

throughout the uidised bed.This would allow for even coke

burn across the entire cross-sec-

tion of the regenerator and

minimise local hot spots and

afterburn.

CFD analysis is also used to

assess, compare and validate

bottom and side entry SCIDs.

The CFD simulations illustrate a

FCC unit with a bottom entry

SCID (see Figure 11a) andanother FCC unit with a modi-

ed side entry SCID (see Figure

11b). The important parameter

to look at is how even the resi-

dence times are across the

regenerator. For the bottom

entry SCID, it is seen that the

distribution is quite even in the

centre of the vessel. However,

there is poor mixing along the

outside of the regenerator dueto the large vessel diameter, as

evidenced by the higher resi-

dence times.

In comparison, the CFD results

for the modied side entry SCID

show that distribution of spent

catalyst is focused at the vessel

centreline. Radial mixing is

stronger in this case driven by

the jet penetration of spent cata-


Without HIB rings With HIB rings

Riser backflow

0

0.2

0.4

0.6

0.8

1.0

V* = V / V max

Figure 9 Non-dimensionalised velocity (V*=V/Vmax) plot at a riser cross-section 15m (50ft) above the feed nozzles. Left image shows the simulationdata without HIB rings; note the high velocity gradients across the riserwidth as well as downward flowing catalyst on the riser wall. The rightimage shows the simulation with HIB rings; the velocity profile is now moreuniform. Catalyst density is consequently also more uniform across the

riser since total flux is constant. The improvement in catalyst density profileimplies that high density catalyst flowing down at the riser walls has beenreduced, and this acts as a driver for improved gas-solids contacting in theriser and reduced total riser pressure drop

Catalyst inlet

Manhole

Draw-off bin

Stripperstandpipe

0º

180º

270º90º

30s3.0s

3.4s

6.8s

9.4s

Figure 10 CFD model of a regenerator unit with ski jump style SCID (notapplicable to Puget Sound). Central image illustrates the regeneratorvessel configuration with tangential spent catalyst inlet along with arrowsindicating the observed catalyst motion from radioactive tracer experiments.The radioactive detector locations (red circles) measure the ‘breakthroughtime’ when the tracer material reaches each probe. Along the vesselwall, model simulation snapshots correspond to the recorded catalystbreakthrough time at each probe. Comparison between the radioactiveprobe data and the simulation results shows close correspondence for boththe dynamics of the trace catalyst as well as the overall bed mixing that tookplace over approximately 30 seconds






catalysis technology, Shanghai, China,

2014.

3 Ludolph R, Question Answer Session,

AFPM Cat Cracker Meeting, Houston TX,

Question 24, 2012.

4 Brosten D J, Chen Y, Feed nozzle assembly,

2014, US Patent 20140145356 A1.

5 Chen Y, Geertshuis B M, Horwege J A,

Samson R, Protective shroud for FCC feednozzles and method for designing such

shroud, 2006, US Patent 7108195 B2.

6 Dries, et al, First-rate FCC technology

design for sustainable reliability: The Pulau

Bukom LRCCU in Singapore, Singapore,

ARTC, 2004.

7 Chen Y, et al, Stripper technology - how to

get more profits from FCC units, NPRA AM

San Francisco CA, AM-05-25, 2005.

8 Chen Y, McIntyre, Catalyst circulation

enhancement technology provides new

opportunity for debottlenecking FCC unit,

NPRA AM, San Antonio TX, AM-04-08,

2004.

9 Chen, et al, Keeping FCC units on

track: winning the operation race with

an innovative cyclone technology, NPRA

Annual Meeting, Phoenix AZ, AM-10-108,

2010, 12-14.

10 Mo W, Hadjigeorge G, Khouw F H H, van der Werf R P, Muller F,

FCCU process modeling and development in Shell, Hydrocarbon

Asia, 2002.

Sayantan Chatterjee is an FCC Research Engineer with Shell

Global Solutions (US) inc in Houston, Texas.

Cian Carroll is an FCC Research Engineer with Shell Global

Solutions (US) inc in Houston, Texas.

Michael Basden is an FCC Research Engineer with Shell Global

Solutions in Houston, Texas.

Kevin Kunz is FCC Licensing & Design Team Lead with Shell Global

Solutions (US) Inc in Houston, Texas.

Charles Burton is a FCC/Alkyl Subject Matter Expert with MOTIVA

Refining, previously Senior FCC Technologist with Shell Global

Solutions (US) Inc., in Houston, Texas.

Steve Nelson is Senior Staff Engineer Projects, Shell Oil Products

US, in Anacortes, Washington.

lyst into the bed. Moreover, the

breakthrough of spent catalyst

into the vessel freeboard can be

quantied by CFD (see Figure

11b). The blue dots spaced evenly

around the regenerator are the

cyclone dipleg exit.

Shell uses CFD for SCID

design to achieve improved

spent catalyst distribution. SCID

redesign should reduce total

afterburn and minimise

localised peak afterburn temper-

atures. Coupling the ow

hydrodynamics with combustion

chemistry allows Shell to illus-

trate afterburn mitigation in the

regenerator.

ConclusionShell’s Puget Sound renery

replaced its old FCC unit end-of-

life riser with a larger riser to

increase residence time, conver-

sion, and to offset the penalty by

moving to a lower coke burn

(driven by NOx emission limits). Furthermore, it

replaced the stripper from existing disk and

donuts to PentaFlow bafes, feed injection

nozzles and associated hardware.

Margin improvement was slightly above thepredicted level for the project. Shell uses the

SHARC model to characterise FCC unit opera-

tions and to assess margin benets derived from

installing the new hardware options at Puget

Sound FCC unit. In addition, CFD modelling

illustrates and validates performance enhance-

ments as a consequence of installing new

hardware.

This paper was presented at the AFPM Annual meeting 2015 in

San Antonio, Texas.SHARC and PentaFlow are registered trademarks of Shell Global

Solutions.

References

1 Hunt D S, Chatterjee S, Munsch C B, Sanborn, Implementation

of state-of-the-art FCC technology for improved reliability and

profitability at Deer Park refinery, AFPM AM, Orlando FL, AM-14-

28, 2014.

2 Hunt D S, Chatterjee S, Munsch C B, Sanborn R, Implementation

of state-of-the-art FCC technology for improved reliability and

profitability at Deer Park refinery, International conference on

Bottom entry SCIDs

Modified side entrySCIDs

Particleresidence

time

60

54

48

42

36

30

24

18

12

6

0

Figure 11 CFD models of regeneratorunit with bottom and modified sideentry SCIDs

LINKS

More articles from: Shell Global Solutions

International

More articles from the following categories:

Fluid Catalytic Cracking

Revamps, Shutdowns and Turnarounds



http://www.digitalrefining.com/150?action=articles


http://www.digitalrefining.com/articles.html?categoryId=16







FCC Profitability Assessment via Advanced Modeling

Documents