2012 Catalysis

catalysis
2012
ptq
cover and spine copy 6.indd 1 23/2/12 20:44:33
Are you looking to step up plant performance?
Topses next generation BRIM catalysts offer refiners the opportunity to increase
performance through an increase in catalyst activity.
Using the original BRIM technology Topse has developed several new catalysts, resulting
in higher activity at lower flling densities.
The next generation BRIM catalysts display
- high dispersion
- high porosity
- high activity
We look forward to stepping up your performance!
WWW. TOPSOE. COM
Stepping up performance
next generation BRIM

technology
haldor.indd 1 23/2/12 11:45:41
2012. The entire content of this publication is protected by copyright full details of which are available from the publishers. All rights
reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic,
mechanical, photocopying, recording or otherwise without the prior permission of the copyright owner.
The opinions and views expressed by the authors in this publication are not necessarily those of the editor or publisher and while every care
has been taken in the preparation of all material included in Petroleum Technology Quarterly the publisher cannot be held responsible for any
statements, opinions or views or for any inaccuracies.
3 Onwards and upwards
ChrisCunningham
5 ptq&a
17 Evaluation of a low rare earth resid FCC catalyst

SabeethSrikantharajahandColinBaillieGrace Catalysts Technologies
BernhardZahnbrecherandWielandWacheBayernoil
23 Refnery fuel gas in steam reforming hydrogen plants
PeterBroadhurstandGrahamHintonJohnson Matthey Catalysts
31 Estimating silicon accumulation in coker naphtha hydrotreaters
ThienanTran,PatrickGripkaandLarryKraus
Criterion Catalysts & Technologies
35 FCC catalyst coolers in maximum propylene mode
RahulPillaiandPhillipNiccumKBR
45 Decrease catalyst costs by regeneration, analysis and sorting
PierreDufresne Eurecat
FrancoisLocatelli Eurecat France

53 Optimisation of integrated aromatic complexes
AxelDkerSd-Chemie AG
59 Troubleshooting a FCC unit
ChiranjeeviThota,ShaliniGupta,DattatrayaTammannaGokak,
RavikumarVoolapalli,PVCRaoandViswanathanPoyyamani
SwaminathanBharat Petroleum Corporation
MarathonOilsCatlettsburgrefnery,Kentucky,USA Photo: Marathon Oil

2012
www.eptq.com
Edit or
Ren G Gon zalez
edit or@pet roleum t echnology.com
Product ion Edit or
Rach el Zamorski
product ion@pet roleum t echn ology.com
Graph ics Edit or
Moh ammed Samiuddin
graphics@pet roleum t echn ology.com
Edit orial
PO Box 11283
Sprin g TX 77391, USA
t el +1 281 374 8240
fax +1 281 257 0582
Advert isin g Sales Man ager
Paul Mason
sales@pet roleum t echnology.com
Advert isin g Sales
Bob Aldridge
sales@pet roleum t echnology.com
Advert isin g Sales Ofce
t el +44 870 90 303 90
fax +44 870 90 246 90
Publish er
Nic Allen
publisher@pet roleum t echnology.com
Circulat ion
Jacki Wat t s
circulat ion@pet roleum t echn ology.com
Crambet h Allen Publish in g Lt d
Hopesay, Craven Arms SY7 8HD, UK
t el +44 870 90 600 20
fax +44 870 90 600 40
ISSN 1362-363X
Pet roleum Tech n ology Quart erly (USPS
0014-781) is publish ed quart erly plus
an n ual Cat alysis edit ion by Crambet h Allen
Publish in g Lt d an d is dist ribut ed in t h e USA
by SPP, 75 Aberdeen Rd, Emigsville, PA 17318.
Periodicals post age paid at Emigsville PA.
Post mast er: sen d address ch an ges t o
Pet roleum Tech n ology Quart erly c/ o PO
Box 437, Emigsville, PA 17318-0437
Back n umbers available from t h e Publish er
at $30 per copy in c post age.
espit e sign s in 2007 of a slowdown in various sect ors of t h e econ omy,
ren ers remain a big play for prospect ive in vest ors. It used t o be
con ven t ion al wisdom t h at h igh er fuel prices an d a slowin g econ omy
would curb deman d an d in crease supply, but for t h e past seven years
t h at h as n ot proved t o be t h e case. Wh ile t h e rat e of in crease in world oil deman d
h as declin ed sin ce t h e surprisin g 4% surge in 2004, it n evert h eless appears t h at
deman d beyon d 2008 will grow, alon g wit h prices. It is a safe bet t h at rapidly
in creasin g oil con sumpt ion by Ch in a, In dia an d even t h e Middle East producers
t h emselves will con t in ue. It is also safe t o assume t h at ren ery an d pet roch emical
con version un it capacit y will n eed t o expan d.
No massive n ew sources of en ergy are expect ed t o come on st ream for t h e
foreseeable fut ure. Th e world will remain depen den t on oil an d gas for decades t o
come even t h ough t h e upst ream in dust ry faces in creasin g ch allen ges in t h e
discovery an d product ion of n ew sources. In fact , some well-placed in dust ry
an alyst s t h in k 2008 may be t h e year wh ere t h ere is n o in crease in crude supply at
all from region s out side of OPEC. For t h is reason , we will con t in ue t o see sign ican t
in vest men t in ren ery upgrades despit e surgin g cost s securit y of feedst ock
supply, albeit un con ven t ion al low-qualit y feedst ock, t akes preceden ce over t h e
qualit y of feedst ock supply.
Feedst ock opt ion s such as biomass (for biofuels product ion ), Can adian t ar san ds
(for dist illat e product ion ) an d ot h er t ypes of un con ven t ion al crude sources require
react or t ech n ology t h at allows for t h e in t egrat ion of t h ese operat ion s in t o exist in g
process con gurat ion s. Th e qualit y of t h ese t ypes of feedst ock are on e import an t
reason wh y a wider array of cat alyst s h as been in t roduced in t o t h e market . For
example, as ren ers cut deeper in t o t h e vacuum t ower, t h e con cen t rat ion of
met als in t h e VGO requires a properly design ed guard bed syst em t o prot ect act ive
cat alyst s in t h e h ydrocracker. Th e ch aract erist ics of feedst ock wit h low API gravit y
(eg, <10), h igh met als, n it rogen an d ot h er un desirable compon en t s is on e of t h e
main reason s wh y h ydrot reat ers an d h ydrocrackers are becomin g larger t o
accommodat e n ot on ly h igh er volumes of cat alyst , but also a wider variet y of
cat alyst wit h specic formulat ion s.
Non -cat alyt ic processes are also playin g a sign ican t role in t h e ren ers abilit y
t o process wh at ever un con ven t ion al crude sources become available. For example,
some ren ers processin g h igh er volumes of resid an d at mosph eric t ower bot t oms
h ave con sidered addin g cert ain t ypes of solven t -ext ract ion processes in addit ion
t o overall improvemen t s t o crude un it (eg, vacuum t ower revamps) an d delayed
coker operat ion s. Improvemen t s in furn ace t ech n ology, such as wit h olen st eam
cracker operat ion s, h ave result ed in sign ican t in creases in worldwide et h ylen e
capacit y.
However, an y expan sion of t h e value ch ain (eg, et h ylen e-t o-propylen e via
deh ydrogen at ion ) requires in vest men t in cat alyt ic-based processes, as discussed in
t h e followin g art icles aut h ored by expert s in t h e eld of down st ream process
t ech n ology. PTQ wish es t o ext en d it s grat it ude t o t h e aut h ors wh o provided
edit orial an d respon ded t o t h e Q&A publish ed in t h is issue of PTQ Catalysis, as
well as t o t h ose respon den t s wh o addressed t h e on lin e quest ion s (www.ept q.com)
t h at addressed t h e specics of cert ain react or an d cat alyt ic issues of import an ce t o
t h e in dust ry.
Securit y of
feedstock supply
Ren G Gon zalez
Vol 13 No 2
2008
PTQ CATALYSIS 2008 5
D
catalysis
ptq
contents/ed com copy 8.indt 1 24/2/12 09:54:43
EVEk WONDEk WHAI MAKE5
OUk CAIALY5I5 5O ADVANCED?
lNDU5IkY-LEADlNG MlND5, OF COUk5E.
|vor v|| o voo |orgo o| p|ovor co|o|,s|s ||o C||||/
r o| po|||o|o oro roo||, o00 c,c|os

o| couuo|co| U|SD opo|o|ors o|oro ||o vo||o, o| C|||||C|, vo s||| ||r| ||o ||uo|o |o, |o
po||o|uorco s o| poop|o. C| |osoo|c| oro oovo|opuor| |oou |op|osor|s o so|oc| |o|co o| oosossvo|,
oooco|oo ||r|o|s ros||,|ooorg scor|s|s v|| ||o oo||, |o ||ors|o|u or ooo r|o o o|oo||||og|
so||or. |os| oss|oo, ||o rox| goro|o|or o| co|o|,s| |oc|ro|og, s r gooo |oros oro |ooos}.
www.CRITERION Catalysts.com
CRITERIO N : Leadi ng mi nds. Advanced technologi es.
criterion.indd 1 23/2/12 11:52:11
CATALYSIS 2012 3
Editor
Chris Cunningham
editor@petroleumtechnology.com
Production Editor
Rachel Storry
production@petroleumtechnology.com
Graphics Editor
Rob Fris
graphics@petroleumtechnology.com
Editorial
tel +44 844 5888 773
fax +44 844 5888 667
Business Development Director
Paul Mason
sales@petroleumtechnology.com

Advertising Sales
Bob Aldridge
sales@petroleumtechnology.com
Advertising Sales Offce
tel +44 844 5888 771
fax +44 844 5888 662
Publisher
Nic Allen
publisher@petroleumtechnology.com
Circulation
Jacki Watts
circulation@petroleumtechnology.com
Crambeth Allen Publishing Ltd
tel +44 844 5888 776
fax +44 844 5888 667
ISSN 1362-363X

Petroleum Technology Quarterly (USPS 0014-781)
is published quarterly plus annual Catalysis edition
by Crambeth Allen Publishing Ltd and is distributed
in the USA by SPP, 75 Aberdeen Rd, Emigsville, PA
17318. Periodicals postage paid at Emigsville PA.
Postmaster: send address changes to Petroleum
Technology Quarterly c/o PO
Back numbers available from the Publisher
at $30 per copy inc postage.
Vol 17 No 2
2012
Onwards and
upwards
T
he refning catalysts business is nothing if not responsive. Sometimes
that response has to be in the immediate term, reference some signifcant
price hikes for FCC catalyst in particular during 2011, although there
is a continuing and strong technical riposte from catalyst developers to the
unlegislated rise in rare earth metals prices that chiefy caused the hikes. For the
most part, though, the catalyst frms technical development and business focus
is determined in the longer term by the twin drivers of economic development
and environmental regulation.
In the US and Western Europe, engineering standards are delivering increas-
ingly effcient road vehicles, while the post-recession market is applying a
more general brake to growth in demand for transportation fuels. As a result,
the worlds developing economies are determining the future shape of growth
in demand for petroleum products.
To illustrate, demand for transport fuels in developing economies may rise
by as much as 300% by 2050, according to the World Energy Council. Half a
decade ago, vehicle ownership was at 11 cars per 1000 people in China and
about 20% higher in India. The world average was around 110 cars per 1000
capita. But Chinas car ownership has been growing by 12% per annum in
recent years, while the equivalent rate in India is 9%. More immediately, China
is expected to cut its sulphur limit for vehicle fuels from a street-chocking
350 ppm to 50 ppm, while Brazil, India and the Arabian Gulf area are moving
towards ultra-low sulphur diesel regulation, all of which should deliver a
strong upsurge in demand for hydrotreating catalysts in the near term. Taken
together, these trends are strongly infuencing both catalyst sales effort and the
siting of new catalyst production centres.
If not quite on the scale of demand for car fuels in developing countries,
growth in demand for ships bunker fuels is strong and, geographically, more
even. Most of the worlds trade is done by ship and the global feet continues
to grow in line with populations and their trade. Although bunkers share of
the total market for fuel oils continues to increase, there is uncertainty about
the rate of increase. The reasons are of special interest to catalyst suppliers. The
International Maritime Organisations Marpol Annex VI regulations are deliv-
ering a steady reduction in the level of sulphur oxide emissions from ships. In
its initial stages, the legislation chiefy affected coastal and semi-enclosed
seawaters, but now it is taking effect in the open seas.
From the start of this year, a reduction in the global cap on the sulphur
content of bunker fuels, from 4.50% to 3.50%, came into force. A progressive
reduction in the allowable level of sulphur in ships fuel will see the cap fall to
0.5% in 2020, subject to a review in 2018. For coastal waters and sulphur emis-
sion control areas, the allowable sulphur level is set to fall to 0.10%, from a
current 1%, in 2015. Uncertainty arises in how the IMO expects to apply the
more stringent levels of Annex VI. The IMO does not favour stack emissions
cleaning on board ships, but ship owners are not especially in favour of the
price premium implied by a radical drop in fuel sulphur levels. Depending on
the price balance, operators of new-build ships may opt for distillate as their
fuel of choice. In any event, a whole lot more hydrotreating to meet maritime
demand is implied.
CHRIS CUNNINGHAM
Edit or
Ren G Gon zalez
edit or@pet roleum t echn ology.com
Product ion Edit or
Rach el Zamorski
product ion@pet roleum t echnology.com
Graph ics Edit or
Moh ammed Samiuddin
graphics@pet roleum t echnology.com
Edit orial
PO Box 11283
Sprin g TX 77391, USA
t el +1 281 374 8240
fax +1 281 257 0582
Advert isin g Sales Man ager
Paul Mason
sales@pet roleum t echn ology.com
Advert isin g Sales
Bob Aldridge
sales@pet roleum t echn ology.com
Advert isin g Sales Ofce
t el +44 870 90 303 90
fax +44 870 90 246 90
Publish er
Nic Allen
publisher@pet roleum t echn ology.com
Circulat ion
Jacki Wat t s
circulat ion@pet roleum t echn ology.com
Crambet h Allen Publish in g Lt d
t el +44 870 90 600 20
fax +44 870 90 600 40
ISSN 1362-363X
Pet roleum Tech n ology Quart erly (USPS
0014-781) is publish ed quart erly plus
an n ual Cat alysis edit ion by Crambet h Allen
Publish in g Lt d an d is dist ribut ed in t h e USA
by SPP, 75 Aberdeen Rd, Emigsville, PA 17318.
Periodicals post age paid at Emigsville PA.
Post mast er: sen d address ch an ges t o
Pet roleum Tech n ology Quart erly c/ o PO
Back n umbers available from t h e Publish er
at $30 per copy in c post age.
espit e sign s in 2007 of a slowdown in various sect ors of t h e econ omy,
ren ers remain a big play for prospect ive in vest ors. It used t o be
con ven t ion al wisdom t h at h igh er fuel prices an d a slowin g econ omy
would curb deman d an d in crease supply, but for t h e past seven years
t h at h as n ot proved t o be t h e case. Wh ile t h e rat e of in crease in world oil deman d
h as declin ed sin ce t h e surprisin g 4% surge in 2004, it n evert h eless appears t h at
deman d beyon d 2008 will grow, alon g wit h prices. It is a safe bet t h at rapidly
in creasin g oil con sumpt ion by Ch in a, In dia an d even t h e Middle East producers
t h emselves will con t in ue. It is also safe t o assume t h at ren ery an d pet roch emical
con version un it capacit y will n eed t o expan d.
No massive n ew sources of en ergy are expect ed t o come on st ream for t h e
foreseeable fut ure. Th e world will remain depen den t on oil an d gas for decades t o
come even t h ough t h e upst ream in dust ry faces in creasin g ch allen ges in t h e
discovery an d product ion of n ew sources. In fact , some well-placed in dust ry
an alyst s t h in k 2008 may be t h e year wh ere t h ere is n o in crease in crude supply at
all from region s out side of OPEC. For t h is reason , we will con t in ue t o see sign ican t
in vest men t in ren ery upgrades despit e surgin g cost s securit y of feedst ock
supply, albeit un con ven t ion al low-qualit y feedst ock, t akes preceden ce over t h e
qualit y of feedst ock supply.
Feedst ock opt ion s such as biomass (for biofuels product ion ), Can adian t ar san ds
(for dist illat e product ion ) an d ot h er t ypes of un con ven t ion al crude sources require
react or t ech n ology t h at allows for t h e in t egrat ion of t h ese operat ion s in t o exist in g
process con gurat ion s. Th e qualit y of t h ese t ypes of feedst ock are on e import an t
reason wh y a wider array of cat alyst s h as been in t roduced in t o t h e market . For
example, as ren ers cut deeper in t o t h e vacuum t ower, t h e con cen t rat ion of
met als in t h e VGO requires a properly design ed guard bed syst em t o prot ect act ive
cat alyst s in t h e h ydrocracker. Th e ch aract erist ics of feedst ock wit h low API gravit y
(eg, <10), h igh met als, n it rogen an d ot h er un desirable compon en t s is on e of t h e
main reason s wh y h ydrot reat ers an d h ydrocrackers are becomin g larger t o
accommodat e n ot on ly h igh er volumes of cat alyst , but also a wider variet y of
cat alyst wit h specic formulat ion s.
Non -cat alyt ic processes are also playin g a sign ican t role in t h e ren ers abilit y
t o process wh at ever un con ven t ion al crude sources become available. For example,
some ren ers processin g h igh er volumes of resid an d at mosph eric t ower bot t oms
h ave con sidered addin g cert ain t ypes of solven t -ext ract ion processes in addit ion
t o overall improvemen t s t o crude un it (eg, vacuum t ower revamps) an d delayed
coker operat ion s. Improvemen t s in furn ace t ech n ology, such as wit h olen st eam
cracker operat ion s, h ave result ed in sign ican t in creases in worldwide et h ylen e
capacit y.
However, an y expan sion of t h e value ch ain (eg, et h ylen e-t o-propylen e via
deh ydrogen at ion ) requires in vest men t in cat alyt ic-based processes, as discussed in
t h e followin g art icles aut h ored by expert s in t h e eld of down st ream process
t ech n ology. PTQ wish es t o ext en d it s grat it ude t o t h e aut h ors wh o provided
edit orial an d respon ded t o t h e Q&A publish ed in t h is issue of PTQ Catalysis, as
well as t o t h ose respon den t s wh o addressed t h e on lin e quest ion s (www.ept q.com)
t h at addressed t h e specics of cert ain react or an d cat alyt ic issues of import an ce t o
t h e in dust ry.
Securit y of
feedstock supply
Ren G Gon zalez
Vol 13 No 2
2008
PTQ CATALYSIS 2008 5
D
catalysis
ptq
contents/ed com copy 8.indt 2 23/2/12 12:35:32
Rare earth price inflation is the most serious issue facing the global refining
industry. Grace, with our long history of innovation and strong R&D, leads the
industry with the first line of commercially successful zero/low rare earth FCC
catalysts: the REpLaCeR family.
Launched in the first quarter of 2011, the REpLaCeR family includes five new
catalysts for both hydrotreated and resid feed processing with zero and low
rare earth content. The REpLaCeR family of catalysts utilizes proprietary
zeolites and state-of-the-art stabilization methods to deliver performance similar
to current rare earth-based FCC technologies.
Were also investing in our plants to bring these products to the refining industry
quickly and globally.
So if youre concerned about rare earth pricing and availability, but need optimal
FCC performance, call the technical experts at Grace. Well customize a solution
using one of our new zero/low rare earth catalysts that delivers the yields you expect.
REp R
TM
Worried about the cost of rare earth?
Grace has the solution:
Grace Catalysts Technologies
7500 Grace Drive
Columbia, MD USA 21044
+1.410.531.4000
www.grace.com
www.e-catalysts.com
grace.indd 1 23/2/12 11:54:00
Q
Istheresignifcantcommercialexperiencewithsolidacid
alkylationcatalysts?Whatsortsofadvantagesareexperienced
orexpectedoverliquidacidcatalysts?
A
Edwin van Rooijen, Business Manager, Albemarle,
Edwin.vanRooijen@albemarle.com
We are still experiencing a considerable amount of
interest in our AlkyClean technology, especially in the
emerging economies of the world. This technology and
the associated solid acid catalyst AlkyStar were jointly
developed by Albemarle, Lummus Technology and
Neste Oil. AlkyClean technology was honoured by the
American Chemical Society with a 2010 Award for
Affordable Green Chemistry.
The AlkyClean process signifcantly improves the
safety of refnery alkylation over conventional liquid
acid-based processes. It reduces potential hazards asso-
ciated with the transportation and handling of liquid
acids. Relying on patented technology, combined with
Albemarles durable AlkyStar catalyst, the AlkyClean
process gives refners a competitive, cleaner and inher-
ently safer alkylation technology. No acid-soluble oils
or spent acids are produced, and there is no need for
product post-treatment of any kind to remove traces of
acid. In addition to these environmental advantages,
the AlkyClean process has proven to be economic and
robust and requires minimal maintenance.
Q
Whatcatalysttypesarebestforminimisingoctanelosses
inFCCgasolinehydrotreaters?
A
Brian Watkins, Manager of Technical Service and
Laboratory Evaluations, Advanced Refning Technologies,
brian.watkins@grace.com
Octane loss in FCC gasoline hydrotreaters occurs with
the saturation of the olefns present in the oil coming
from the FCC unit. This saturation readily occurs over
hydrotreating catalyst in the presence of heat and
hydrogen, so a low-metals cobalt molybdenum
(CoMo/ Al
2
O
3
) catalyst selective for hydrodesulphurisa-
tion is recommended. The goal is to be able to provide
the required sulphur removal with limited olefn and
aromatic saturation. Nickel molybdenum (NiMo/
Al
2
O
3
) catalyst, although having high hydrodesulphuri-
sation activity, also has a much higher olefn and
aromatics conversion activity, making it unsuitable for
this application. Generally, to minimise olefn satura-
tion, lower pressure and high liquid space velocity are
recommended in order to limit octane loss.
A
Steven Mayo, Global Manager Hydroprocessing
Applications,Albemarle,steven.mayo@albemarle.com
To minimise octane loss in FCC gasoline hydrotreaters,
both a catalyst and a process are needed that selec-
tively maximise sulphur removal while minimising
olefn saturation and mercaptan recombination reac-
tions. The best catalysts for the application are
formulated to maximise direct-route desulphurisation
with minimum hydrogenation activity. Cobalt-
molybdenum catalysts are the preferred catalyst type.
Olefns are readily saturated under typical naphtha
hydrotreating conditions, so catalysts alone are usually
insuffcient to prevent signifcant loss of octane in these
units. Licensed FCC gasoline post-treatment process
technology combined with proprietary catalyst technol-
ogy allows for very high levels of sulphur removal
(>95%) with minimum octane loss. RT-235 is the latest
catalyst development by ExxonMobil Research and
Engineering and Albemarle for their SCANfning,
selective FCC gasoline desulphurisation, process. This
catalyst offers exceptionally high HDS activity with
even better octane retention than the frst-generation
SCANfning catalyst, RT-225. RT-235 can be used in
any SCANfner and is also available for use in selective
FCC naphtha desulphurisation units licensed by
others.
Q
How effective are NO
x
-reducing additives at cutting
regeneratorstackemissions,andisusingthemacost-effective
option?
A
Alan Kramer, Global FCC Additives Specialist,
Albemarle,alan.kramer@albemarle.com
There are two additive types that can lower FCC regen-
erator stack NO
x
emissions. The frst type is low-NO
x

combustion promoters such as Albemarles ElimiNO
x
.
These replace conventional platinum-based promoters
used in full-combustion FCC units. ElimiNO
x
has been
shown to be very effective over the past 15 years in
lowering NO
x
emissions while maintaining CO and
afterburn control in the regenerator. In the 2007 NPRA
annual meeting,
1
it was reported that US refneries
using low-NO
x
promoters, in accordance with EPA
consent decrees, usually saw between 20% and 80%
reductions in NO
x
after switching promoter type.
The second type of additive is a non-promoting NO
x
reduction additive, such as Albemarles DuraNO
x
.
These additives are also only used in full-combustion
FCC units. The performance of these additives varies
www.eptq.com Catalysis 2012 5
ptq&a
AdditionalQ&Acanbefoundatwww.eptq.com/QandA
Q&A copy 10.indd 1 23/2/12 12:51:18
greatly from unit to unit and is often diffcult to
predict. During the 2007 NPRA annual meeting, it was
also reported that NO
x
reduction additives used by US
refners engaged in consent decree trials averaged 26%
reduction in NO
x
. Of the full-burn FCC units reporting
results, fve saw no effect, 10 observed reductions up
to 30%, and eight saw reductions between 50% and
80% when NO
x
additives were used.
When combined with tighter controls on regenerator
excess oxygen levels, Albemarles NO
x
reduction addi-
tives allow refners to completely avoid the capital
expenditure of installing hardware to reduce NO
x
emis-
sions, proving once again that additives can be a very
cost-effective option.
1 Sexton, Joyal, Foley, EPA Consent Decrees: Progress on FCC
Implementation and Future Challenges, NPRA AM-07-44, 2007.
A
Jason Smith, Refning Additives Manager, BASF,
Jason.k.smith@basf.com
NO
x
-reducing additive performance and cost effective-
ness is highly dependent on the unit, as both
equipment characteristics and operational variables
play a role in NO
x
formation and reduction. Controlling
NO
x
emissions is probably one of the most diffcult
applications in the FCC unit.
There are several interacting factors that infuence
NO
x
emissions. These include: type and level of CO
promotion, air fow distribution, excess oxygen, regen-
erator temperature, regenerator pressure, regen bed
level, stripping steam rate, catalyst circulation, and
type and quantity of NO
x
reduction additive.
There are two options to reduce NO
x
emissions. The
frst approach is to reduce the level of NO
x
generated.
This can be achieved by replacing the platinum CO
promoter with one that generates lower levels of NO
x

while maintaining the capability of oxidising CO to
CO
2
. Low NO
x
Promoter (LNP) is used by many refn-
eries in controlling afterburn and CO emissions, with a
limited amount of NO
x
produced. Typically, switching
from platinum to LNP will result in a reduction of NO
x

of approximately 30%.
A second approach is to use a NO
x
additive specif-
cally designed to chemically reduce NO
x
to inert
nitrogen. Several NO
x
-reducing products are being
offered in the marketplace, including CleaNO
x
, BASFs
NO
x
reduction additive. CleaNO
x
has been most
successful in the US, where it has been used to address
EPA consent decrees. In one example, a US refnery in
the midwest was able to reduce NO
x
by more than 70%
from an average of 200 ppm to 60 ppm using 1.4 wt%
of CleaNO
x
in its inventory.
CleaNO
x
has also been used and proven in applica-
tions where a refnery wanted to reduce NO
x
from an
already low average base of 27 ppm on the East Coast
of the US. Even in such a demanding application,
CleaNO
x
demonstrated a 33% reduction in NO
x
.
A
Eric Griesinger, Marketing Manager, Environmental
Additives, Grace Catalysts Technologies, eric.griesinger@
grace.com
NO
x
reduction additives generally fall under two cate-
gories: standalone NO
x
reduction additives and low
NO
x
combustion promoters.
Standalone NO
x
reduction additives are catalytic-
based NO
x
control technologies that provide NO
x

reduction without any combustion promotional activ-
ity. Generally, this NO
x
control technology has
provided a slow response to mitigating elevated NO
x

concentrations. Grace Davison has developed a cata-
lytic NO
x
reduction additive, GDNO
x
1, which shows
prospect of providing a quicker ability to curb NO
x

emissions. Further, GDNO
x
1 technology, which has
been patented, provides greater NO
x

reduction with a correspondingly
greater dosing rate (see Figure 1),
yet with diminished FCC unit yield
penalties often encountered when
utilising previous-generation NO
x

reduction additives. Additionally,
GDNO
x
1 has not been vulnerable to
material surcharges, thus making the
product a cost-effective option.
Current generation of low NO
x

combustion promoters are typically
formulated with a noble metal other
than platinum. Historically, the use
of platinum has been demonstrated
to exhibit a correlation with elevated
and prolonged NO
x
concentrations
in regenerator fue stack gases.
Applications of Grace Davisons
current-generation low NO
x
combus-
tion promoter, CP P, when dosed in
higher than normal rates, whether
intentionally to correct other FCC
unit conditions or unintentionally,
has shown a shortened duration of
6 Catalysis 2012 www.eptq.com
300
350
250
200
150
100
50
0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
N
O
,

p
p
m
Ti me, hrs
0
5.0% GDNOX 1
2.5% GDNOX 1
10.0% GDNOX 1
GDNO
X
1 NO
x
after GDNO
X
1 Percentage NO
x

Addition rate % of inventory Base line NO
x
, ppm addition, ppm reduction, %
2.5% GDNO
X
1 292 139 50
5.0% GDNO
X
1 287 110 60
10.0% GDNO
X
1 287 62 80
Figure 1 Pilot plant testing: NO
x
reduction with multiple GDNO
x
1 additions
Q&A copy 10.indd 2 23/2/12 12:51:35
R
e
f
n
e
r
s
w
o
r
ld
w
ide use our
h
y
d
r
o
p
r
o
c
e
s
s
i
n
g
t
e
c
h
n
o
lo
g
ie
s
and catalyst
s
t
o
d
e
l
iv
e
r
c
l
e
a
n
e
r
p
r
o
d
ucts from low
-
q
u
a
lit
y
f
e
e
d
s
.
L
e
t
u
s
s
h
o
w
you how. (00
1)5
10
.2
4
2
.3
17
7
www.clg
-
c
le
a
n
.c
o
m
Learn how CLGs tailored hydroprocessing catalyst systems
maximize coversion to clean products visit clg-catalysts.com
ICR 512
ICR 180
ICR 185
ICR 250
ISOCRACKING
VGO
clean transportation fuels
ultra-low sulfur diesel (<5 ppm)
high smoke point jet (25-30 mm)
Ultra-low sulfur naphtha (<0.5 ppm)
FCC feed
clg.indd 1 23/2/12 11:55:22
elevated NO
x
emissions is likely. This observation of a
shortened NO
x
emission excursion interval can provide
beneft to refners when striving to satisfy a rolling day
average or other time-based NO
x
emission limit
constraints, while still providing CO promotion
performance similar to prior medium-activity plati-
num-formulated CO promoters.
The US Environmental Protection Agency (EPA)
concluded that newly adopted emission limits utilising
additives and combustion controls were achievable,
cost effective and had fewer secondary impacts than
more costly hardware-oriented control technologies.
1
The EPA issued fnal amendments to its New Source
Performance Standards for Petroleum Refneries
(NSPS)
1
on 24 June 2008. Within this amendment, the
EPA states that the currently Best Demonstrated
Technology (BDT) to NO
x
emission control now
includes the use of additives in conjunction with an
upwardly revised NO
x
emission limit of 80 ppmv
based on a seven-day rolling average. Typically, under
EPA Consent Decree proceedings, FCC unit operations
have been restricted to a NO
x
emission limit of 20
ppmv based on a 365-day rolling average and 40 ppmv
based on a seven-day rolling average. This NSPS
amendment now also recognises the secondary envi-
ronmental impact that many of the hardware solutions
infict upon the environment, inherent in their opera-
tion to achieve a 20 ppmv maximum NO
x
emission
limit. These secondary impacts include PM (Particulate
Matter) as well as additional SO
2
and NO
x
emissions
resulting from increased electrical demand. In addition,
many of the hardware solutions require supplementary
chemical reactants that add hazards and emission
problems of their own.
2
As such, non-platinum formu-
lated oxidation promoters and advanced oxidation
controls typically are anticipated to provide the least
overall environmental impact, as they generally do not
generate further secondary environmental emissions,
and do so cost effectively by EPA measures.
Grace Davison continues offering catalytic NO
x

control technologies to the refning industry in agree-
ment with the EPAs NSPS
1
conclusions, whereby the
combination of non-platinum-formulated oxidation
promoters and advanced oxidation controls typically
are anticipated to provide the least overall environ-
mental impact, and do so at a reasonable cost in many
applications.
1 New Source Performance Standards (NSPS) for Petroleum Refneries,
at 40 C.F.R. Part 60, Subpart J/Ja. 73 Fed. Reg. 35838 (24 June 2008). The
amendments were proposed in 2007 as the outcome of the periodic
review of NSPS standards required under the Clean Air Act - Section
111(b)(1). 72 Fed. Reg. 27278 (14 May 2007). The rules provide technical
corrections to the existing Subpart J standards and create a set of new
emissions for fuid catalytic cracking units (FCCU), fuid coking units
(FCU), sulphur recovery plants (SRP), and fuel gas combustion devices
for facilities that were newly constructed, modifed or reconstructed
after 14 May 2007. The new rules became effective on 24 June 2008.
2 Roser F S, Schnaith M W, Walker P D , Integrated View to Understanding
the FCC NO
x
Puzzle, UOP LLC, Des Plaines Illinois, 2004 AIChE Annual
Meeting.
Q
Are there any rule-of-thumb indications of the trade-off
in price and performance where low rare earth catalysts have
replaced conventional rare earth-containing FCC catalysts?
A
Raul Arriaga Global FCC Applications Technology
Specialist, Albemarle, raul.arriaga@albemarle.com
Ken Bruno Global Applications Technology Manager, FCC,
Albemarle, ken.bruno@albemarle.com
The total price of an FCC catalyst is the combination of
a catalysts base price plus rare earth surcharges. The
adjustment formula to calculate the rare earth
surcharge is the result of mutual agreement between
suppliers and refners and typically depends on the
market price of lanthanum oxide. The base price of the
catalyst depends on the type of technology used and
the amount of active components in the formulation.
Special low rare earth technologies have been devel-
oped by catalyst suppliers to compensate for the
selectivities and zeolite stabilisation provided by rare
earth. Additionally, the amount of active components
may need to be increased if the rare earth is reduced.
Therefore, it naturally follows that the base price of a
low rare earth technology catalyst will be higher than
a conventional rare earth-containing FCC catalyst.
However, the lower rare earth catalyst results in a
reduced rare earth surcharge, making the total price of
the catalyst economically attractive.
The changes expected in the performance of FCC
catalysts at lower rare earth depend on the gap
between the technology used in the original catalyst
and the new lower rare earth catalyst. A simple exam-
ple is reducing the rare earth on an originally high
rare earth catalyst while keeping all other catalyst
parameters constant. In this case, the zeolite stability
would deteriorate and the equilibrium activity of
the catalyst would decline. At the same time, LPG
selectivity would increase and gasoline selectivity
would drop. Under these circumstances, the
catalyst will also produce less coke at constant conver-
sion, resulting in lower delta coke. These changes in
selectivities are usually not acceptable because each
FCC unit is typically operating against multiple
constraints.
Based on the above, very rarely would a catalyst
supplier recommend only a reduction in the catalysts
rare earth content without compensating via advanced
low rare earth technology modifcations. For example,
Albemarle can make use of various features developed
for a new family of Low Rare Earth Technology (LRT)
catalysts. These features include a new zeolite stabilisa-
tion technology, improved porosity, reduced mass
transfer limitations (higher accessibility), advanced
active matrices and zeolites grown to a high silica-to-
alumina ratio, which results in improved structural
integrity and lower amounts of non-framework
alumina. The right combination of these features
will recover most, if not all, of the activity lost
with the lower rare earth content and will modify
selectivities to keep the FCC unit operating within
constraints. The result is maximum proftability for the
FCC unit.
Q&A copy 10.indd 3 23/2/12 16:21:58
Engineering Solut ions . . . Delivering Result s
Lummus Technology o ers a mult it ude of performance
cat alyst s for re ning and pet rochemical processes, including:
t Hydroprocessing
t Alkylat ion
t Hydrogenat ion
t Hydrocracking
t Ole ns Met at hesis
t FCC
Visit our websit e for a complet e list of cat alyst s and services.
CB&Is performance cat al yst s
t hey cause quit e t he react ion.
www.CBI.com
C
M
Y
CM
MY
CY
CMY
K
PTQ_Ad_CBI- jan-press.pdf 1 1/ 20/ 2012 11:18:33 AM
cbi.indd 1 24/2/12 12:56:10
A
Solly Ismail, Modeling Specialist, BASF, solly.ismail@
basf.com
While the reduction of rare earth in catalyst formula-
tions can deliver immediate operating budget cost
savings, it is important to consider the impact this will
have on product slate margins. Based on proprietary
simulation modelling, BASF has shown that as rare
earth levels decrease, the conversion of feed to higher
valued products will also drop (assuming all other
variables are held constant). In order to assist custom-
ers with evaluating the impact of lower rare earth
catalytic options and to limit its downside, BASF works
closely with each refner to understand specifc unit
parameters and objectives. A customised strategy can
include a combination of levers such as increasing cata-
lyst addition rates, increasing total surface area or a
combination of both in order to restore activity and
achieve desired product specifcations and margin
targets. By doing so, BASF is able to determine if low
rare earth catalyst formulations are appropriate for the
customer and, if so, develop a customised strategy to
implement a low rare earth catalytic option that fts the
needs of the specifc user.
As of the end of Q4 2011, 40% of our customers had
made the switch to a lower rare earth formulation. Of
these, fve went through multiple reductions. All
customers were happy with BASFs approach of tailor-
ing new solutions based on either increased surface
area with a minimum of additional catalyst usage. The
company also worked closely with refners in monitor-
ing the changes to proactively mitigate surprises.
To get a fuller understanding of the impact of lower
REO, the reader is referred to the more detailed article
in the Q4 2011 issue of PTQ (FCC catalyst optimisation
in response to rare earth prices).
A
Rosann Schiller, Senior Marketing Manager, rosann.
schiller@grace.com and Colin Baillie, Marketing Manager,
EMEA, Grace Catalysts Technologies, colin.baillie@grace.
com
The REpLaCeR series of low and zero rare earth cata-
lysts from Grace is being used in over 50 applications
globally. First and foremost, there has been no trade-off
in performance with respect to either product yields or
catalyst additions. However, users of the REpLaCeR
series of catalysts have experienced signifcant catalyst
cost savings associated with high rare earth prices.
Refneries moving to zero rare earth REpLaCeR cata-
lysts for low metal feed applications have seen catalyst
costs reduced by up to 500 000/y per 1 t/d of catalyst
used, while users of low rare earth REpLaCeR catalysts
for resid processing have been able to reduce catalyst
costs by between 250 000 and 750 000/y per 1 t/d
of catalyst used.
Q
What catalyst formulations will maximise LCO yield from
the FCC unit with minimum effect on bottoms yield?
A
Yen Yung, Global Technical Specialist, Albemarle, yen.
yung@albemarle.com
The biggest challenge in fuid catalytic cracking is
converting as much material in the feed with an atmo-
spheric boiling point above 370C bottoms to more
valuable LPG, gasoline (hydrocarbon molecules boiling
between about 40C and 221C) and light cycle oil
(LCO hydrocarbon molecules boiling between 221C
and 370C). To maximise the yield of LCO, it is impera-
tive to maximise the conversion of bottoms to LCO
while minimising the conversion of LCO to lighter
products and coke. It is generally accepted that meso-
pore and macropore activity, the so-called alumina
matrix activity, favours bottoms cracking, while zeolites
provide higher LPG and gasoline selectivity. Therefore,
middle distillate production is generally favoured by
higher matrix cracking (as evidenced by a higher meso
surface area) and reduced zeolite cracking. In other
words, middle distillate production increases as the
zeolite-to-matrix ratio decreases.
For greatest bottoms conversion, the feed molecules
need to quickly reach the active sites. Conversion of
the desirable products in the diesel boiling range and
other secondary reactions, such as hydrogen transfer,
aromatisation and condensation, must be avoided. This
is achieved by increasing the accessibility of the cata-
lyst. Accessibility is the property that allows primary
products to escape promptly from the reaction sites.
Outstanding performance of highly accessible cata-
lysts, as measured by our internally developed
Albemarle Accessibility Index (AAI) method, has been
confrmed in several applications.
Albemarle has a full line of MD catalysts. Amber MD
and Upgrader MD feature very high matrix cracking
activity and AAI. Amber MD is recommended for gas
oil feed applications and Upgrader MD is recom-
mended for cracking residual feedstocks. For
applications requiring fexibility, the companys
bottoms conversion additive, BCMT-500, is recom-
mended for all types of feedstocks. In addition,
Albemarles technical specialists have special tools for
optimising unit operations and selecting the proper
FCC catalysts grades, including those that utilise Low
Rare Earth (LRT) technology.
Once an FCC catalyst is selected, Albemarles techni-
cal specialists will assist their customer in optimising
their operating strategy for maximum LCO production,
as discussed in the following two examples.
The frst example includes an FCC unit processing
vacuum gas oil with a typical API of 23C and a
sulphur content of about 1 wt%. The catalyst used in
this example is Amber MD. The FCC unit was operat-
ing at a unit riser outlet temperature of 518C, a
combined feed temperature of 226C and a catalyst-to-
oil ratio of 7.0 kg/kg. In this example, the gasoline end
point is minimised to an ASTM D-86 end point of
149C, while the LCO end point is very high at an
ASTM D-86 end point of 379C. With these cut points
and the use of Amber MD, a yield of 44 wt% LCO is
obtained with a typical cetane index of 34. This unit
applies no bottoms recycle.
The second example also consists of an FCC unit that
is processing vacuum gas oil. Like the frst example,
Q&A copy 10.indd 4 23/2/12 12:51:57
action loves reaction
Chemical reactions require chemical catalysts. As the
global leader in chemical catalysts, BASF acts through
continuous product and process innovations in collaborative
partnerships with our customers. The result is a broad
chemical catalyst portfolio backed by dedicated customer
and technical service and enabled through the strength of
BASF - The Chemical Company.
At BASF, we create chemistry.
www.cat alyst s.basf.com/process
n
Adsorbents
n
Fine Chemical Catalysts
n
Environmental Catalysts
n
Catalysts for Fuel Cells
n

Catalysts for Oleochemicals & Other Biorenewables
n

Oxidation & Dehydrogenation Catalysts
n

Petrochemical Catalysts
n

Polyolef n Catalysts
n

Ref ning Catalysts
n

Syngas Catalysts
n

Custom Catalysts
basf.indd 1 23/2/12 11:58:16
Amber MD is used. In this example, there is a low
reaction temperature (499C), high combined feed
temperature (368C) and low catalyst-to-oil ratio (4.0
kg/ kg). Bottoms recycle (the recycle rate/ fresh feed
ratio varied between 0.5-1.0 vol/ vol) is applied to
enhance the production of LCO. Note that the volume
of recycle can be as high as the fresh feed intake. The
gasoline end point is also minimised. LCO yield and
cetane index are very high at 42.4 wt% and 34, respec-
tively. Bottoms yield (21.2 wt %) is higher due to
extremely low severity. Despite the high bottoms yield
unit, the economics were much improved as the market
favoured a high LCO yield.
A
Rosann Schiller, Senior Marketing Manager, rosann.
schiller@grace.com and Colin Baillie, Marketing Manager,
EMEA, Grace Catalysts Technologies, colin.baillie@grace.
com
Such LCO maximisation catalysts obviously need to
have good bottoms-cracking performance. Therefore,
Grace catalysts for LCO maximisation incorporate high
matrix activity, including the option of utilising a new
technology that provides a controlled deposition of a
thin layer of reactive alumina on the surface of the
zeolite crystals to facilitate the pre-cracking of large
feed molecules. In addition, an FCC catalyst for LCO
maximisation must also have the ability to maintain
the cracked HCO molecules within the LCO boiling
range fraction, which requires limiting the cracking of
LCO to gasoline. Therefore, LCO maximisation cata-
lysts from Grace incorporate proprietary pore
restructuring functionality, which results in more pores
with the diameter range of 100-600 (see Figure 1).
This boost in porosity enables a more effective release
of LCO molecules from the acid sites, minimising the
undesired cracking of LCO into gasoline. Graces LCO
maximisation catalyst brands include DieseliseR, Midas
and Rebel FCC catalysts.

A
Stefano Riva, Technical Service Manager, BASF,
stefano.riva@basf.com
For maximising LCO, an intermediate product in the
cracking reaction sequence, focus should be on the
matrix cracking activity of the catalyst. While the
zeolite can achieve good bottoms cracking in a coke-
selective way (low delta coke), the amount of zeolite
required for that objective will rapidly crack the
desired LCO to lighter products. Due to this trade-off,
the bottoms cracking has to come from an increase in
matrix. It is generally recognised that, at constant
conversion, a lower Z/ M (zeolite-to-matrix surface
area) catalyst may have a higher delta coke. However,
this will not necessarily result in a hotter regenerator
(typically the opposite is true) when a unit moves from
maximum conversion to maximum distillate modes.
Not only should the catalyst Z/ M be adjusted for
maximum LCO, but also the catalyst activity and the
FCC operating conditions. FCC units should operate at
lower reactor severity (lowering the heat demand), and
with lower equilibrium catalyst activity (reducing delta
coke). This leaves plenty of room to accommodate a
moderate increase in higher delta coke that can be
derived from a lower Z/ M catalyst. With that said,
attention should still be paid to selecting both the right
amount of matrix and the associated technology, with
preference for the best coke-selective low Z/ M catalyst.
This will provide ample fexibility to swing between
maximum conversion and maximum distillate opera-
tions with the same catalyst should the market change
rapidly. BASFs Prox-SMZ (Proximal Stable Matrix and
Zeolite) technology is an example that addresses all of
0.6
0.7
0.5
0.4
0.3
0.2
0.1
10 100 1000
P
o
r
e

v
o
l
u
m
e
,

c
m
3
/
g
Pore di amet er,
0
Figure 1 Porosity of LCO maximisation catalysts
Unit composition Early trial Late trial
Competitor, % 85 31
Stamina, % 15 69
Unit operating conditions
Total feed rate, ton/d 1723 1672
Feed preheat temp, C 178 232
ROT, C 520 515
Regenerator bed temp, C 721 720
C/O, wt/wt 7.0 5.8
Feed quality
Specifc gravity 0.93 0.92
Conradson carbon, wt% 0.71 1.14
Basic nitrogen, ppm 450 410
TBP90, C 547 571
Equilibrium catalyst properties
V + Ni, ppm 4100 4240
TSA, m
2
/g 120 127
MSA, m
2
/g 53 64
Z/M 1.3 1
FACT activity, wt% 73 70
FCC unit yields (with cutpoint adjustments)
Gasoline (C
5
-160C), wt% 37.19 34.06
LCO (160-340C), wt% 29.03 33.87
Slurry (340C+), wt% 13.57 13.14
Coke, wt% 5.79 5.09
LCO/slurry, % 68.1 72.1
Summary:
Achieved >4.8 wt% greater distillate yields at partial turnover
Maintained lower bottoms at reduced reactor severity and dirtier
feed conditions.
European maximum distillate trial of BASF Stamina catalyst
Table 1
Q&A copy 10.indd 5 23/2/12 12:52:08
Value adding catalysts, absorbents, additives and
process technology for oil refning processes.
www.jmcatalysts.com/refneries
UK
Tel +44 (0)1642 553601
USA
Tel +1 630 268 6300
Tel +1 732 223 4644
Naphtha HDS
Diesel HDS
Hydrotreating catalysts
Vacuum tower
Hydrogen catalysts
FCC unit
FCC additives
SOx NOx removal
Light olefn production
Bottoms conversion
Metals traps
Activity boosters
CO oxidation
Gas / liquid
purifcation
S, Hg, Cl removal
Crude unit
VGO HDS
Hydrogen plant
Process
diagnostics
JM_2485_RefineriesAd_ART_210x297.indd 1 09/02/2012 15:59
j matthey.indd 1 23/2/12 11:59:38
the above (see NPRA-AM-09-34 and
NPRA-AM-10-17).
With BASFs unique manufacturing process,
the zeolite and the matrix are not physically
blended, as in a traditional low Z/ M catalyst,
but are created in-situ during a single manu-
facturing step. This process creates
unprecedented proximity of matrix and zeolite,
enabling reduced diffusion path length and
best-in-class coke selectivity among the low
Z/ M family of catalysts. Further, the extremely
low sodium content (below 0.1 wt%) achiev-
able in this manufacturing process not only
enables high stability of both zeolite and
matrix to reduce the opex, but also helps to
reduce the hydrogen transfer reactions,
improving the LCO cetane. Maximum LCO
catalysts from the Prox-SMZ technology have
been commercially established and are avail-
able for both VGO (under the name of HDXtra)
and resid feeds (under the name of Stamina).
In several classic distillate maximisation
trials with BASFs Stamina catalyst, LCO yields
have increased while either holding steady or
dropping slurry yields. In one case, a European
FCC unit was processing resid feed and
wanted to move to LCO maximum mode while
dropping slurry yields and improving coke
selectivity. The unit severity was dropped by
lowering the reactor outlet temperature by 5C
and increasing the feed preheat. The reduced
severity conditions together with the highly
stable Stamina MSA generated a 4.84 wt%
improvement in LCO yields while maintaining
bottoms conversion. All this was achieved in
spite of dirtier feed conditions (ie, higher
Conradson carbon and feed metals).
The second case was an Asian trial that also
underwent reduced reactor severity to maxi-
mise LCO yields. While using Stamina catalyst,
the lowest bottom yields on record were
achieved together with record throughputs due
to the coke-selective bottoms upgrading.
Q
Are there any dedicated catalyst developments
geared towards favouring FCC propylene yield?
A
Stuart Foskett, Regional Technology Manager,
BASF, stuart.foskett@basf.com
BASF is continually developing new catalyst
technologies aimed at enhancing propylene
production for maximum propylene operations
(upwards of 10 wt% or 17.5 vol% propylene
yield). A high level of ZSM-5 is always a
prerequisite for maximum propylene; however,
it is the characteristics of the FCC catalyst itself
that defne how much propylene can ulti-
mately be produced. Propylene yields
eventually reach a plateau as ZSM-5 content is
increased to high levels; therefore, a holistic
10500
10000
9500
9000
8500
8000
T
o
t
a
l

f
e
e
d

r
a
t
e
,

t
o
n
s
/
d
a
y
Ti me
7500
20
18
16
14
12
10
52 54 56 58 60 62 64 66
S
l
u
r
r
y
,

w
t
%
Conversi on, wt %
8
Competitor
Stamina
516
520
518
514
512
510
508
506
T
e
m
p
e
r
a
t
u
r
e
,

C
Ti me
504
Figure 2 Asian Stamina trial. Record low slurry yields were achieved at
reduced reactor severity. Simultaneously the coke selective bottoms
upgrading of Stamina allowed for record throughputs
40
45
35
30
25
20
15
L
C
O

(
1
6
0
-
3
4
0
C
)
,

w
t
%
Ti me
10
LCO
HCO + slurry 26
30
22
18
14
10
6
S
l
u
r
r
y

(
3
4
0
C
+
)
,

w
t
%
2
Figure 1 LCO and slurry yields with BASF Stamina. LCO yields were increased
with reduced reactor severity without impacting the slurry
Q&A copy 10.indd 6 23/2/12 12:52:21
approach to catalyst design requires
attention to additional factors
beyond the ZSM-5. It is the genera-
tion and preservation of gasoline
olens, as precursors to ZSM-5
cracking, that is the dening factor
for ultimate propylene potential.
The high percentage of active
ingredient in the catalyst, enabled by
our in-situ zeolite synthesis, allows
us to offer maximum propylene
catalysts featuring reduced rare
earth content, without any penalty
in terms of activity and required
catalyst addition rate. Lower rare
earth helps to minimise hydrogen
transfer, preserving more gasoline
olens for cracking to propylene by
ZSM-5. We have also invested heav-
ily in technology upgrades for our
plants to allow extreme levels of
ultra-stabilisation, to produce
zeolites with very low and stable
unit cell size (UCS). BASFs MPS
(maximum propylene solution) tech-
nology was rst introduced in 2005
(see NPRA-AM-05-61) and has
undergone continuous improvement
since then. MPS has been operating
continuously in the worlds largest
propylene-focused FCC unit since
2006. Based upon extensive circulat-
ing pilot plant evaluations, we are
anticipating the latest MPS develop-
ments will achieve incremental gains
in propylene yield in the range of
0.5 to 1.0 wt% compared to the
previous state-of-the-art technology.
Meanwhile, the development of
additional technologies aimed at
maximising propylene are progress-
ing. This includes technologies
aimed at offering high propylene
yields with resid feeds combined
with leading coke selectivity and
metals resistance.
A
Carel Pouwels, Global FCC Resid
Specialist, Albemarle, carel.pouwels@
albemarle.com
Achieving record-high propylene
and conversion from wide ranges
of feed qualities offers considerable
challenges to catalyst design. Key to
this is good understanding of the
mechanisms involved and then
responding with the proper catalyst
technology and design to meet a
units objectives. Crucial to the
success of reaching record propyl-
ene yields is the ability to minimise
hydrogen transfer while having
sufcient cracking activity.
Albemarles AFX catalyst has
been developed to meet the desired
objectives, through its unique
features of high catalyst accessibil-
ity and strong matrix activity.
Hereby, maximum slurry conver-
sion is achieved while generating a
maximum of gasoline precursors,
which are converted in record
propylene yields. The high accessi-
bility of AFX enables fast diffusion
of primary cracking products away
from the acid sites, thus minimising
unwanted hydrogen transfer.
Q
What is the impact of vanadium level
on E-cat affecting FCC gasoline sulphur?
A
Alan Kramer, Global FCC
Additives Specialist, Albemarle, alan.
kramer@albemarle.com
Generally, reners want to avoid
loading their catalyst with vanadium
due to the known negative effects it
has on zeolite stability and catalyst
activity. However, increased levels of
vanadium in catalysts with higher
alumina contents (which typically
are more resistant to vanadium)
directionally yield lower levels of
gasoline sulphur. Testing has indi-
cated the vanadium mechanism
primarily reduces the saturated
sulphur content of the gasoline and
has little to no effect on benzothio-
phene, which often comprises the
bulk of gasoline sulphur.
Commercially, vanadium levels need
to be increased by at least 1000-2000
ppm on E-cat before differences can
be measured. The losses in catalyst
activity and negative yield effects
related to these large levels of extra
vanadium on the catalyst are rarely
justiable. Depending on local regu-
lations, equilibrium catalyst with
high levels of vanadium may be
classied as hazardous or toxic
waste and can be very difcult and
expensive to dispose of properly.
Reners do have other options
besides increasing E-cat vanadium
or using vanadium-based products
for reducing gasoline sulphur. For
example, Albemarles R-975 and
Scavenger catalyst additives are
designed to remove gasoline sulphur
and do not contain any vanadium.
1 DURANIT
inert ceramic balls

2 special reformed packings
6 liquid distrubutors / collectors
7 random packings made of plastic
8 random packings made of metal
9 random packings made of ceramic
10 software and consulting
4 support plates / grids
5 feed devices: gas / liquids
3 droplet separators / demisters
P. O. Box 552, D - 56225 Ransbach-Baumbach
Phone +49 26 23 / 895 - 0, info@vff.com
Tower packings,
catalyst support
material and
column equipment.
For further information
please visit:
www.vff.com
1
1
2
2
3
6
6
4
4
4
4
5
7
7
8
10
10
8
9
9
Please visit us
Hall: 4.0
Stand: D66
Q&A copy 10.indd 7 23/2/12 12:52:32
dupont.indd 1 23/2/12 15:04:26
Evaluation of a low rare earth resid
FCC catalyst
R
are earth metals have played
an important role in the refn-
ing industry since the 1970s,
when it was discovered that they
could be used to stabilise the
zeolite-Y component of FCC cata-
lysts to provide higher activity, as
well as being used to infuence
product selectivity. Rare earth
metals play an additional role in
resid processing applications, as
they have proven to be until now
the most effective vanadium trap,
helping to maintain stability and
activity.
The two main rare earths used in
FCC catalysts are lanthanum and
cerium, and these metals have
historically been readily available
for under $5/ kg. However, a reduc-
tion in Chinese export quotas
resulted in rare earth prices rising
dramatically in 2010, with the price
of lanthanum reaching $140/ kg
around May 2011. Since then, rare
earth prices have subsided some-
what, but remain signifcantly
higher than historical levels.
The rare earth market is incredi-
bly unpredictable and is expected
to remain highly volatile. Against
this backdrop of uncertainty with
respect to availability and pricing,
zero and low rare earth catalysts
will continue to play an important
role in the FCC industry. Grace
Catalysts Technologies provides the
REpLaCeR series, the frst commer-
cially successful zero and low rare
earth FCC catalysts.
Zero and low rare earth FCC
catalysts
Simply removing rare earth from
an FCC catalyst would result in a
considerable detrimental effect in
A zero rare earth catalyst blended with a rare earth-based resid catalyst enabled a
refnery to reduce its FCC catalyst rare earth requirement by 80%
SABEETH SRIKANTHARAJAH and COLIN BAILLIE Grace Catalysts Technologies
BERNHARD ZAHNBRECHER and WIELAND WACHE Bayernoil
most FCC operations due to the
lower activity and worsening prod-
uct yield slate obtained. To develop
FCC catalysts containing lower rare
earth content, it is necessary for
alternative materials and processing
techniques to be used that stabilise
the zeolite component. Grace has
considerable experience developing
zero and low rare earth FCC cata-
lysts. During the 1990s, it developed
Z-21, a rare earth-free stabilised
zeolite-Y, which was the basis for
the Nexus catalyst family. This was
commercialised in 1997 as a rare
earth-free catalyst family for low-
metal feed applications, and has
since been successfully used in 10
applications.
In 2010, the company developed
the REpLaCeR series of zero and
low rare earth FCC catalysts, which
are based on the existing Z-21
zeolite technology, as well as a new
Z-22 zeolite technology. State-of-
the-art methods are used to stabilise
the rare earth-free Z-21 and Z-22
zeolites, involving proprietary
stabilising compounds and unique
manufacturing processes. FCC cata-
lysts incorporating these new
zeolites provide similar and even
improved performance compared
to conventional rare earth-contain-
ing catalysts. Based on the Z-21 and
Z-22 technologies, the REpLaCeR
series of zero rare earth catalysts
for low-metal hydrotreating and
VGO applications includes
REsolution and REactoR, which are
currently being used in more than
15 applications.
For resid applications, the devel-
opment of rare earth-free catalysts
is much more challenging due to
the additional demands placed on
zeolite stability. However, signif-
cant advances have been made by
applying processing technology
involving metals resistance func-
tionality to catalyst systems
containing the Z-21 and Z-22
zeolites. This has resulted in the
rare earth-free REduceR catalyst,
which can be blended with rare
earth-based resid FCC catalysts,
thus reducing the overall rare earth
requirement and the costs associ-
ated. There are currently 22
refneries using the REduceR cata-
lyst, and typically they are applying
a stepwise approach to implement
the rare earth-free catalyst. Refners
are starting with a blending level of
30% REduceR catalyst and, upon
confrming its performance, many
are then moving to a blending level
of 50%. Bayernoil is using the
REduceR catalyst in both of its two
FCC units with blending levels
even higher than 50%.
Commercial experience of low rare
earth resid catalysts
The Bayernoil Vohburg refnery is
located in the Bavarian region of
southern Germany and, along with
the nearby Bayernoil Neustadt
refnery, contributes to a total refn-
ing capacity of 10.3 million t/ y. The
two locations combined contain
three crude units, two vacuum
towers, two FCC units, one mild
hydrocracker and hydrogen plant,
one visbreaker, three reformers and
one ether plant. The FCC unit at
Vohburg is a UOP side-by-side
model and was built in 1967. It is a
resid unit with a typical throughput
of 14 000 b/ d, operates in deep
partial burn and processes 80-90%
atmospheric residue. The feedstock
grace.indd 1 23/2/12 12:56:58
require higher catalyst additions,
which has not been the case in any
application of the REduceR catalyst.
Figure 1 shows the catalyst addition
rate and e-cat microactivity at
Bayernoil Vohburg before and after
using the 30% blend. It can be seen
that good activity retention was
achieved after the switch at a simi-
lar or even slightly lower catalyst
addition rate, highlighting the high
vanadium tolerance of the REduceR
catalyst.
Figure 2 shows the e-cat coke and
gas factors of Nektor and the 30%
REduceR catalyst against nickel
equivalents to compare nickel
resistance. The 30% REduceR cata-
lyst shows lower gas factors and
similar coke factors, further demon-
strating its suitability for high metal
resid feeds. The FCC unit data
provided in Figure 3 show that the
REduceR catalyst blend provided
improved bottoms conversion
compared with the Nektor catalyst.
In addition, a lower delta coke was
obtained, which reduced the regen-
erator bed temperature by about
10C. This allowed the refnery to
achieve higher conversion at
constant feed atmospheric residue
content, or to process an increased
amount of atmospheric residue at
constant conversion.
The refnery considered the
performance of the REduceR cata-
lyst to be such a success that they
increased the blending ratio from
30% to 50%, thus reducing the over-
all rare earth content of the catalyst
to 1.5 wt%. Table 1 shows the FCC
unit product yields obtained with
the 50% REduceR catalyst blend
compared with the Nektor catalyst.
During the REduceR catalyst trial,
feed quality deteriorated and feed
throughput decreased; therefore,
for the purposes of evaluating the
actual catalyst performance, the
yields shown are calculated on the
basis of constant feed properties
and independent operating condi-
tions. The key objective of the
refnery was to maintain conversion
and bottoms upgrading while
reducing rare earth content. As can
be seen, these key objectives were
met, and in addition conversion
and bottoms upgrading were even
increased. The REduceR catalyst
has a Conradson carbon content of
3 wt%, and the e-cat metals levels
are approximately 4500 ppm vana-
dium, 3500 ppm nickel, 6000 ppm
Fe and 5000 ppm sodium.
This Vohburg FCC unit was
previously using a Nektor catalyst
from Grace containing 3.1 wt% rare
earth, which performed well. In
April 2011, the refnery began to
blend 30% of REduceR catalyst with
the Nektor catalyst, with the simple
objective of reducing rare earth
while maintaining high perform-
ance. A certain misconception about
rare earth-free catalysts is that they
5
7
6
4
3
3
1
/
1
2
/
2
0
0
9
2
5
/
0
2
/
2
0
1
0
2
2
/
0
4
/
2
0
1
0
1
7
/
0
6
/
2
0
1
0
1
2
/
0
8
/
2
0
1
0
0
7
/
1
0
/
2
0
1
0
0
2
/
1
2
/
2
0
1
0
2
7
/
0
1
/
2
0
1
1
2
4
/
0
3
/
2
0
1
1
1
9
/
0
5
/
2
0
1
1
1
4
/
0
7
/
2
0
1
1
0
8
/
0
9
/
2
0
1
1
C
A
R
,

t
/
d
2
66
70
68
64
62
E
-
C
a
t

M
A
T
,

w
t
%
76
78
74
72
CAR
E-Cat MAT REduceR
Figure 1 Activity retention of the REduceR catalyst blend at 30%
5
7
6
4
3
3000 4000 5000 6000 7000
r
o
t
c
a
f

s
a
G
Ni equivalents, mg/kg
2
REduceR
Nektor
2.3
1.9
1.5
1.1
3000 4000 5000 6000 7000
r
o
t
c
a
f

e
k
o
C
Ni equivalents, mg/kg
REduceR
Nektor
Figure 2 Coke and gas factors of the REduceR catalyst blend at 30%
grace.indd 2 23/2/12 12:57:07
pcs 1.indd 1 23/2/12 20:37:57
provided a similar coke yield but
an improved delta coke, and
allowed the regen bed temperature
to be decreased by 15C. The higher
LPG yield at the expense of gaso-
line is a consequence of the lower
hydrogen transfer from REduceR.
This is a positive yield shift for the
refnery and was anticipated.
Bayernoil Vohburg was highly
satisfed with the REduceR catalyst
trial, and subsequently became the
frst refnery to move to a 70%
blending level, further reducing the
rare earth content to 1.0 wt%. In
December 2011, the refnery
increased the blending ratio of the
REduceR catalyst up to 80%, thus
reducing rare earth to 0.6 wt%.
Despite high nickel and vanadium
73
77
75
71
69
67
65
50 55 60 65 70 75 80 85 90
F
F
%
t
w

,
n
o
i
s
r
e
v
n
o
c

t
n
e
r
a
p
p
A
AtRes in feed, wt%
Nektor
REduceR
720
725
715
710
705
700
695
508 510 512 514 516 518 520 522 524

,
e
r
u
t
a
r
e
p
m
e
t

d
e
b

r
o
t
a
r
e
n
e
g
e
R
C
ROT, C
690
Nektor
REduceR
0.87
0.93
0.91
0.89
0.85
0.83
0.81
0.79
0.77
60 62 64 66 68 70
%
t
w

,
e
k
o
c

a
t
l
e
D
Apparent conversion, wt%FF
0.75
Nektor
REduceR
14
16
15
13
12
11
10
9
65 67 69 71 73 75 77
%
t
w

,
d
l
e
i
y

y
r
r
u
l
S
Conversion, wt%FF
8
Nektor
REduceR
levels, the refnery continues to see
excellent performance. It is esti-
mated that the switch to the
REduceR catalyst has provided the
refnery with over 2 million /y
cost savings when taking into
account catalyst costs and product
yields obtained.
GRACE, GRACE CATALYSTS TECHNOLOGIES,
REPLACER, RESOLUTION NEKTOR, NEXUS,
REDUCER and REACTOR are marks of W R
Grace & Co.-Conn.
Colin Baillie is Marketing Manager with Grace
Catalysts Technologies EMEA and holds a
masters degree and PhD in chemistry from the
University of Liverpool, UK.
Email: Colin.Baillie@grace.com
Sabeeth Srikantharajah is Technical Service
Manager at Grace Catalysts Technologies
EMEA. He holds a diploma in chemical
engineering from the University of Munster.
Bernhard Zahnbrecher is Process
Development Engineer for FCC with Bayernoil.
He holds a university diploma in chemical
engineering.
Wieland Wache is a Process Engineer in the
Production department at the Bayernoil
Refnery in Vohburg. He holds a diploma in
chemistry from the Technical University
Aachen and PhD in chemical engineering from
University Bayreuth.
Figure 3 FCC unit data of the REduceR catalyst blend at 30%
Nektor 50% Nektor
50% REduceR
Cat-to-oil, g/g Base Base + 0.4
Conversion, wt% Base Base + 0.5
Hydrogen, wt% Base Base + 0.02
C
1
+C
2
s, wt% Base Base + 0.2
Propylene, wt% Base Base + 0.4
C
4
olefns, wt% Base Base + 0.6
LPG, wt% Base Base + 2.0
Gasoline, wt% Base Base - 1.6
LCO, wt% Base Base - 0.2
Slurry, wt% Base Base - 0.2
Coke, wt% Base Base - 0.1
Delta coke, wt% Base Base - 0.09
CAR, MT/D Base Base
Feed Ni, ppm Base Base
Feed V, ppm Base Base
Regen bed temp, C Base Base - 15C
Calculated FCC unit data of the REduceR catalyst blend at 50%
Table 1
grace.indd 3 23/2/12 12:57:19
pcs2.indd 1 23/2/12 20:39:00
eurecat.indd 1 23/2/12 12:02:36
Refnery fuel gas in steam reforming
hydrogen plants
O
perators of steam reform-
ing-based hydrogen plants
want feedstock options to
minimise operating costs and
maximise operational fexibility.
Consequently, new-build hydrogen
plants are often designed for a
number of hydrocarbon feeds. It is
common to have three or four feed-
stocks ranging from offgases
through to naphtha requiring full
operational fexibility across the
range.
1,2
Operators of existing plants
are also evaluating alternatives to
the original feedstock slate and in
some cases implementing changes
that may necessitate modifcation of
the plants operating conditions,
hardware, equipment, catalyst
selection and so forth.
3-6
The quest
for cheaper feedstock options is
undoubtedly heightened by the
signifcant increases in both natural
gas and crude oil prices in the last
few years, although the emergence
of shale gas production has reversed
this trend in certain areas. A feed-
stock option being considered
increasingly by hydrogen plant
operators associated with oil refn-
eries is the refnery fuel gas (RFG)
pool. Relative to imported natural
gas and many other hydrocarbon
streams and offgases in the refnery,
RFG has a comparatively low value.
Thus, RFG represents an attractive
feedstock option where there is
excess RFG available.
RFG is not widely used as a
hydrogen plant feed. This is
because its composition leads to a
number of diffculties in process-
ing in the feedstock purifcation
and steam reforming sections of
the hydrogen plant fowsheet. In
this article, we will explore these
Fuel gas is an attractive feedstock for hydrogen production, but appropriate
catalysts and temperature control are needed to address high olefn levels
Peter BrOadhurst and Graham hintOn
Johnson Matthey Catalysts
diffculties and the strategies for
hydrogen plant design and opera-
tion, which may be used to allow
processing of RFG as a feedstock.
This will include some recently
developed catalytic and control
solutions developed jointly by
Johnson Matthey and Air Products
& Chemicals.
rFG composition and processing
diffculties
RFG is a combination of refnery
unit waste or by-product gases,
often referred to as offgases. The
offgases that are sent to the RFG
pool are those that cannot be proc-
essed elsewhere in the refnery
either because of the composition
or because there is an excess availa-
ble. The offgases in the RFG come
from various refnery unit opera-
tions (catalytic reforming, FCC,
hydrotreating, coking), the availa-
bility and amount of which will
depend on the refnery operation.
Hydrogen-containing offgases may
be partly or fully used in hydrogen-
consuming units or may be treated
to recover the hydrogen in a
membrane or PSA unit so that a
hydrogen lean gas is available to
the RFG pool. Also, offgases or
offgas blends with high olefn levels
may be treated to recover olefns,
with the olefn lean gas going to the
RFG pool. Thus, RFG can differ
signifcantly between refneries.
Examples of RFGs, which have
been proposed for hydrogen plant
feeds, are shown in Table 1. This
shows the substantial variations in:
hydrogen, from 11-35 mol%; meth-
ane, from 2665 mol%; and olefns,
from 2.615.9 mol%.
RFG feeds often contain quite
high levels of sulphur compounds.
Up to 100 ppmv can be found and
this can contain quite a signifcant
proportion of mixed organic
sulphur compounds. The level
and speciation are necessarily
mol% example 1 example 2 example 3 example 4 example 5 example 6
Hydrogen 34.9 10.8 25.5 34.1 13.9 14.4
Methane 26.3 64.9 34.8 45.3 43.2 42.3
Ethane 11.0 13.4 23.5 8.2 12.2 13.7
Propane 9.5 2.5 5.2 3.2 8.3 6.8
Butanes 7.1 1.4 3.2 1.4 0.3 4.4
Pentanes 0.0 0.2 0.5 0.6 0.3 1.5
Hexanes 0.0 0.3 0.9 0.2 0.3 0.6
Ethene 3.8 3.1 3.4 1.5 7.2 6.6
Propene 2.1 0.5 0.6 0.9 8.3 6.7
Butenes 3.0 ~ ~ 0.2 0.3 0.3
Pentenes 0.0 ~ ~ ~ 0.1 Trace
Nitrogen 1.9 1.8 1.9 3.7 4.5 2.2
Argon ~ ~ ~ Trace Trace ~
Oxygen Trace ~ ~ ~ ~ ~
Carbonmonoxide ~ 0.3 0.3 0.3 1.1 Trace
Carbondioxide 0.4 0.8 0.2 0.4 ~ 0.5
Total 100.0 100.0 100.0 100.0 100.0 100.0
rFG compositions considered for hydrogen plant feed
table 1
j matthey.indd 1 23/2/12 13:02:52
the inlet olefn level to an accepta-
ble level. This approach has some
disadvantages. Depending on the
inlet olefn level and therefore the
extent of dilution required, the
recycle can be substantial. This
increases capital expenditure in a
number of ways: the recycle fow
usually increases the vessel size to
stay within design space velocity
design limits and additional equip-
ment is needed such as a recycle
cooler and circulator.
Without a recycle system, there is
limited fexibility in terms of the
olefn level using standard HDS
catalyst types, so approaches that
widen the operating envelope with-
out installation of a recycle system
will be benefcial. One such
approach that Johnson Matthey has
recommended is to use a Katalco
higher activity catalyst. This has a
higher active metals loading and
allows operation at inlet tempera-
tures down to ~250C (462F) with
the same exit temperature limit of
400C (752F). Thus, the amount of
olefns that can be processed with-
out installation of a recycle system
is increased to 78 mol%, depend-
ing on the precise composition of
the feed in terms of heat release
and specifc heat.
This concept has been extended
to form part of the technology
claimed in recent patent applica-
tions fled jointly by Johnson
Matthey and Air Products &
Chemicals
7
and which is available
for licence. Detailed evaluation of a
typical range of RFG feed composi-
tions established that the extremely
active pre-sulphided Katalco prod-
uct allows the olefn hydrogenation
reaction to strike below 150C
(302F). This widens the operating
temperature envelope to over 250C
(450F) and allows the processing
of feeds containing well in excess of
10 mol% olefns.
It is imperative, however, that the
consequences of the changes in
operating temperatures are fully
considered if this is to be retroftted
into an existing reactor to accom-
modate a higher olefn feed.
An alternative approach is to use
a tube-cooled converter.
8
The heat
of reaction is removed on the shell
side of the converter using water,
dependent on the blend of gases
going to the RFG pool.
Additionally, the RFG composi-
tion can fuctuate signifcantly in a
given refnery as rates on different
units change and particularly if a
unit comes off line. The amount of
offgas available to the RFG pool
changes and so impacts on the
composition of the blend, which
comprises the RFG. This presents
control issues in some cases.
In terms of incorporating RFG
into the hydrogen plant feed slate,
the aspects that may cause diffcul-
ties can be summarised:
High olefn levels
Variability in the RFG composi-
tion as the blend of offgases
changes
High hydrogen levels
Signifcant sulphur levels
Substantial levels of higher
hydrocarbons, which may include
naphthenes and/or aromatics
Less usual trace and minor
components.
Not every RFG will present each
of these diffculties and each case
must be considered separately. If a
new-build hydrogen plant is being
considered, it must be designed to
include any RFG feed case(s). When
considering using RFG on an exist-
ing plant, the extent to which there
is a problem will be infuenced by
the original design basis.
Feedstock purifcation section
RFG feeds can cause various issues
in the feedstock purifcation
section.
High olefn levels
Olefns need to be removed from the
hydrocarbon feed in a hydrogen
plant to a level below 1 mol% to
minimise possible olefn-derived
carbon formation in the steam
reformer, although higher levels
may be acceptable where there is a
pre-reformer in the fowsheet. The
hydrodesulphurisation (HDS) cata-
lyst is also an effective olefn
hydrogenation catalyst and removes
olefns almost completely as long as
there is suffcient hydrogen present.
Thus, for RFGs with signifcant
olefn levels, the hydrogen available
in the feed and as recycle must
be suffcient for the olefn
hydrogenation reaction, other hydro-
gen-consuming reactions and to
provide the target hydrogen level
specifed at the HDS converter exit.
Johnson Matthey recommends a
different level of hydrogen be
present in the HDS effuent, depend-
ing on the feed composition and
how heavy it is. For the gases given
as examples 5 and 6 in Table 1,
insuffcient hydrogen is present in
the RFG on its own to hydrogenate
the signifcant olefn content. This
means that additional hydrogen
must to be added, usually recycled
to the purifcation section of the
plant, of approximately 5 mol% of
the RFG feed rate for example 5, and
approximately 3 mol% of the RFG
feed rate in example 6.
Olefns hydrogenate exothermi-
cally over the HDS catalyst and the
temperature rise can be 20+C
(36F) per mol% olefn, depending
on the heat capacity of the feed gas.
The inlet temperature must be
adjusted to ensure that the maxi-
mum HDS bed temperature remains
below 400C (752F). However,
using standard HDS catalysts, such
as Katalco 41-6T or Katalco 61-1T,
the inlet temperature needs to be
above 300C (572F) to provide
suffcient activity for reactions to
initiate. Given the need for some
operating margin inside these
restrictions, this limits the olefn
that can be processed to a few
mol% in a once-through reactor
system.
To process higher olefn levels, a
recirculation system is usually
employed around the HDS reactor
(see Figure 1). The reactor effuent
is olefn free so the recycle dilutes
Cooler
Recirculator
Figure 1 HDS converter with recirculation
j matthey.indd 2 23/2/12 13:03:03
cat tech1.indd 1 23/2/12 12:04:42
and the temperature in the catalyst
packed tubes is maintained in the
required range. The reactor is not
truly isothermal, as the heat release
is suffciently fast for the tempera-
ture to rise in the reaction zone
within the tubes before the coolant
brings the temperature back down.
This approach has been commer-
cialised in at least one plant.
Variability in the RFG composition
As refnery unit operations change
rate, come off line or back on line,
the RFG composition can fuctuate.
This can make step changes in the
gas going forward to the hydrogen
plant and can cause control issues,
particularly where there is a
substantial olefn content in the
feed. If the olefn level increases
sharply, it is possible that the
400C (752F) upper limit for HDS
operating temperature might be
exceeded. Also, if the inlet temper-
ature has been lowered to well
below 300C (572F) to allow for
the hydrogenation exotherm, and
the olefn level decreases sharply,
the exotherm will collapse, which
may lead to organo-sulphur slip.
In both cases, the HDS inlet
temperature needs to be adjusted
quickly in response to the change
in RFG composition.
A control system concept has
been developed to allow this to
happen.
7
There is a rapid feedback
system from changes in the HDS
bed exotherm into the upstream
feed preheat so that the tempera-
ture in the catalyst bed can be
controlled within acceptable limits.
Another possible problem is if the
hydrogen level in the RFG decreases
suddenly to a point where there is
insuffcient to hydrogenate olefns
(and any traces of oxygen and the
expected organo-S compounds) in
the feed. Additionally, the recom-
mended HDS exit hydrogen level
should ideally be maintained. If a
rapid decrease in feed hydrogen
level occurs, another source of
hydrogen, probably recycle hydro-
gen, needs to be rapidly established.
If not, the consequences could be
slippage of olefns and/or sulphur
compounds to the steam reforming
section, with resulting carbon
formation and/or poisoning.
Some RFG blends may contain a
signifcant amount of hydrogen. In
some circumstances, this might
provide a driving force for side
reactions over the HDS catalyst if
the RFG unusually happened to
contain extremely low sulphur
levels. The undesirable side reac-
tions are methanation, which would
also need a substantial amount of
carbon oxides present, and hydroc-
racking, which would need heavier
alkanes present. Both are very
rarely observed and are associated
with operation at transient or
abnormal operating conditions (for
instance, low fow, very low
sulphur and excess temperature).
Methanation and hydrocracking
activity can be signifcantly
suppressed by pre-sulphiding the
HDS, an effect that is also achieved
if there is a reasonable level of
sulphur in the feed.
RFG blends can contain signifcant
sulphur levels, of which much may
be present as organo-sulphur
compounds. As long as the maxi-
mum and typical levels are known
and there is speciation of the
sulphur, new plants can be
designed appropriately.
Existing plants may have been
designed for a lower level of and/
or less diffcult organic sulphur
compounds so that conversion in
the existing HDS converter may not
be possible. There are three retroft
options depending on the addi-
tional HDS capacity needed:
Replace the existing HDS catalyst
with a higher activity version (for
instance, replace Katalco 41-6T or
Katalco 61-1T with Katalco 61-2F)
Install a combined product, which
delivers HDS activity and H
2
S
absorption (such as Katalco 33-1) in
the downstream H
2
S removal beds
Replace the existing HDS
converter with a larger vessel.
If the total sulphur level is higher
than the original design, the life of
the H
2
S removal beds will be short-
ened. If the design features lead-lag
H
2
S removal beds, allowing change-
out on-line, the consequence is
more frequent change-out. Selection
of the highest capacity H
2
S removal
absorbents (for instance, Katalco 32-
5) can lessen the change-out
frequency. The problem can be
greater if the design has only a
single H
2
S removal bed built to last
between turnarounds, where it is
possible the increased sulphur level
will not allow operation for the
normal interval. Higher capacity
absorbents may help, but a retroft
of a second vessel may be required.
Substantial level of higher
hydrocarbons
In the event that the RFG contains
higher hydrocarbons, there may be
an increased tendency for thermal
cracking in the feed preheat coil.
The carbon formed can carry
forwards and deposit on top of the
HDS catalyst, leading to an
increased pressure drop over time
and some deactivation if the top
catalyst becomes coated with
carbon. If this problem is observed,
it can be mitigated by replacing the
hold-down balls on top of the cata-
lyst with a high-voidage shape,
such as Dypor 604, which has more
capacity for the carbon foulant,
before the pressure drop becomes
critical.
components
There are a number of other species
that can be present in a refnery
offgas stream and which, therefore,
could fnd their way into the RFG.
These include chlorides, arsine,
oxygen, acetylenes and dienes.
Chloride can be dealt with by
hydrogenation of any organo-
chlorides over the HDS catalyst
followed by the removal of HCl
using an HCl removal absorbent
such as Katalco 59-3. This must be
placed upstream of the ZnO-based
H
2
S removal absorbent. In existing
plants, it may be necessary to retro-
ft this into the purifcation section,
usually by putting it on top of the
H
2
S removal absorbent.
Arsine will be removed on the
surface of the HDS catalyst and acts
as a poison. If present, levels are
likely to be sub-ppm and the effect
on the HDS catalyst is minimal. For
a new plant design, the volume of
HDS catalyst would be increased
slightly to allow for the absorption
j matthey.indd 3 23/2/12 13:03:22
cat tech2.indd 1 23/2/12 12:06:09
temperature as the PSA unit goes
through its cycles. Additionally, if
RFG is being used as the feed to the
plant, it is likely that the trim fuel
type will also have changed. This
will necessitate that the suitability
of the burners is checked to fre this
new fuel type with good control
and fame stability.
With proper design and operation,
sulphur should be removed in the
feed purifcation section.
Substantial level of higher
hydrocarbons
In new plant designs, the selection
of steam reforming technology and
catalysts should take account of the
RFG composition. In existing plant
with non pre-reformer fowsheets,
the impact of a heavier hydrocar-
bon component in the feed must be
assessed in case a change in steam
reforming catalyst is required to
cope with the heavy component in
the RFG. As mentioned above, this
may necessitate an increase in the
steam-to-carbon ratio for the plant,
a reduction in reformer outlet
temperature or even a change in
steam reforming catalyst type.
components
these should be removed in the
feed purifcation section.
Variability in the RFG composition
This presents possible problems for
the steam reforming section if there
is not suffcient control. If there is a
sudden decrease in olefns causing
the HDS exit temperature to
decrease, slip of organo-sulphur is
possible, which would poison the
pre-reformer or steam reforming,
whichever is frst in the fowsheet.
If there is a sudden decrease in
hydrogen, there may be insuffcient
to hydrogenate olefns and/or
organo-sulphur. The resulting slip
could lead to sulphur poisoning
and carbon formation in the pre-
reformer or steam reforming,
whichever is frst in the fowsheet.
Conclusions
While RFG is an attractive economic
of arsenic and associated poisoning
of the HDS catalyst at the bed inlet.
Oxygen will be removed by
hydrogenation over the HDS
catalyst. This is a very exothermic
reaction and so the temperature rise
and hydrogen consumption must
be taken into account in addition to
any from olefn hydrogenation.
Acetylenes and dienes may be
present at low levels. There is some
evidence that these may polymerise
in the feed preheat or HDS
converter, leading to carbon deposi-
tion. This is unlikely to be a major
issue as long as the level remains at
low ppm levels so that any pressure
drop rise and catalyst deactivation
caused by the carbon is very slow.
Steam reforming section
Some considerations and impacts
for the steam reformer and associ-
ated equipment need to be
considered.
High olefn levels
olefns should be converted in the
feed purifcation section. However,
levels of up to 1 vol% olefns can be
accommodated in the feed to a
steam reformer.
The process duty in the reformer
may be lower for feeds with high
hydrogen levels compared to other
feeds such as natural gases. This is
because there is less endothermic
steam reforming reaction in order
to generate the same quantity of
hydrogen. Additionally, since the
feeds are more hydrogen rich than
a natural gas, less feed fow may be
needed in order to produce the
same quantity of hydrogen.
However, even though the feed is
more hydrogen rich than a typical
natural gas, the average carbon
number can be higher, meaning
that the tendency for carbon forma-
tion within the reformer is also
higher. This may necessitate either
an increase in steam to carbon, a
reduction in outlet temperature, or
a change in the steam reforming
catalyst type. This is discussed
further below.
All of these effects are shown in
Table 2. In the table, the reformer
heat load for constant hydrogen
production has been calculated,
assuming that no other process
parameters are changed (inlet and
outlet temperatures, steam-to-
carbon ratio and reformer outlet
pressure). The second row of the
table shows how the molar fow of
the RFG in question will compare
to the molar fow of a typical natu-
ral gas for the same hydrogen
production rate.
In a new plant design, all these
factors can be taken into account. If
introducing such a feed to an exist-
ing plant, however, these factors
and implications arising from them
need to be taken into account and
reviewed. The reduction in reformer
heat load will have a number of
consequences. The frst, and most
obvious, is that the fring on the
reformer box must be reduced. This
means that the bridge wall temper-
ature will reduce, the fow rate of
combustion gases will reduce and
the amount of heat available to be
recovered in the fue gas duct will
be less. Hence, there is the possibil-
ity that some of the coils may
become pinched, leading to a reduc-
tion in steam production or in
combustion air preheat tempera-
ture. Second, if the plant is a
modern PSA hydrogen plant in
which the PSA tail gas is used as
the major fuel source on the
reformer, it is possible that the trim
fuel requirement is dropped to a
level where there are control issues
in maintaining the reformer outlet
Table 2

Natural Example Example Example Example Example Example
gas 1 2 3 4 5 6
Reformer heat load, % 100 93.4 97.8 94.2 89.6 99.9 99.5
Feed fow, % 100 76.2 88.4 81.6 109.7 69.6 64.5
Average carbon no. 1.06 2.09 1.34 1.76 1.42 1.71 1.85
Impact of varying feed composition on feed fow and reformer heat load
Table 2
j matthey.indd 4 23/2/12 13:03:34
option for hydrogen plants
associated with refnery operations,
its processing presents challenges
in the feedstock purifcation and
steam reforming sections of the
fowsheet. These challenges result
from the composition of the RFG
and also from its variability.
In the feedstock purifcation
section, the range of options for
handling the high olefn levels often
found in the RFG pool has been
expanded in recent years by consid-
ering more active catalysts and heat
exchange reactors. In conjunction,
control systems have been devel-
oped to allow the HDS converter
operating temperature to be
adjusted as the feed composition
changes to maintain the tempera-
ture within the acceptable range for
successful operation.
In the steam reforming section,
the impact of the change in feed
and fuel type needs to be modelled
and understood. There may be
minimal impact associated with the
change but, conversely, there may
be changes required in operating
parameters, catalyst type, heat
exchanger performance or even
heat exchanger design, as well as
consideration given to any neces-
sary changes in control schemes.
KATALCO is a mark of the Johnson Matthey
Group of Companies
References
1 Cromarty B J, Hydrogen Market Review,
Synetix Technical Paper, 1999.
2 Ratan S, Hydrogen Technology Overview,
Proceedings of 6th International Conference
on Refnery Processing, AIChE Spring National
Meeting, New Orleans, Apr 2003.
3 Chlapik K, Slemp B, Alternative Lower Cost
Feedstock for Hydrogen Production, NPRA
Annual Meeting, San Antonio, 21-23 Mar
2004.
4 Winsper A J, Irizar I C, The Study on the use
of Butane Feed for the Repsol-YPF La Coruna
Hydrogen Plant, ERTC 11th Annual Meeting,
Paris, 13-15 Nov 2006.
5 Cotton W J, Singh V, Feedstock Conversion
for Indian Ammonia Plants. A Review of
the Challenges, Proceedings of Fertilizer
Association of India (FAI) Seminar 2003, New
Delhi, Dec 2003.
6 Broadhurst P V, Cotton W J, Taking feedstock,
Hydrocarbon Engineering, Mar 2005, 41.
7 Davis R A, Macleod N, Wilson G E, Process
for Hydrogenating Olefns, World Patent WO
2009/123909 A2, 8 Oct 2009.
8 Musich N, Natarajan R S, Klein H, Process
and System for Reducing the Olefn Content of
a Hydrocarbon Feed Gas and Production of a
Hydrogen-enriched Gas Therefrom, US Patent
Application 20080237090, 2 Oct 2008.
Peter Broadhurst leads the ammonia synthesis
and inert support product teams of Johnson
Matthey Catalysts and is Technical Consultant
for the syngas purifcation product team.
He was previously Technical Sales Manager
for hydrogenation catalysts, hydrogen plant
catalysts and refnery absorbent products, and
Technical Manager for oil refnery and syngas
products. He holds a BSc in chemistry from
Bristol University, UK, and a PhD in inorganic
chemistry from Cambridge University.
Graham Hinton is the Senior Process Engineer
leading a team of engineers at Billingham, UK,
supporting Johnson Matthey Catalysts sales
into new syngas plants. He holds a masters
degree in engineering science, specialising in
chemical engineering, from the University of
Oxford, UK.
Det erminat ion of Sulf ur and ot her element s
at -line and in t he laborat ory
El ement al Anal ysi s
of Fuel s
To keep pace w i t h t he demandi ng qual i t y requi rement s of
modern f uel s, advanced, preci se and easy t o use anal yt i cal
t echnol ogy i s requi red. Wi t h a compl et e range of XRF and
ICP spect romet ers, SPECTROs uni que sol ut i ons f or at -l i ne
and l aborat or y el ement al anal ysi s are capabl e of meet i ng
t he most demandi ng product speci f i cat i on t est i ng
requi rement s.
Measure your fuels
- at sub 10-15 ppm levels of sulfur t o ensure federal
government agency compliance
- at t race ppm levels for met al element s such as Cu, Ca, Mg,
Na, K, and P, t o ensure low engine emissions as well as
t rouble-f ree mot oring
- w i t h reliabili t y, speed and accuracy for any part icular applicat ion
- manually or fully aut omat ically
Discover more excit ing det ails,
visit SPECTROs
e-Learning cent er
or cont act us for
addit ional informat ion
about t he SPECTRO
solut ions for fuels
analysis at w w w.spect ro.com/f uels
spect ro.info@amet ek.com and
Tel + 49.2821.892-2102.
j matthey.indd 5 23/2/12 16:23:05
ACS-AMISTCO Catalyst Bed Supports
w w w. acs-amlstco. com - 24hr LMLPGLNC SLPv|CL - l-800-23l-0077
For separat ion or mass t ransfer applicat ions you will not nd a
capabilit y t hat can deliver like ACS-AMISTCO, Inc.
Wit h in-house engineering support and fabricat ion, we can use
your exist ing drawings, or modify t hem t o improve your process.
Catalyst Bed Supports
Basket Strainers
Nozzles
Outlet/Inlet Baskets
Distributors
Hub and Header Laterals
D|STP|8UTOPS & SUPPOPTS
Manufactured to customer
specihcations or engineered to
meet performance requirements
L|QU|D COALLSCLPS
0il water separations haze removal
from fuels. Removal of tower wet reux
Caustic treater applications
STPUCTUPLD PACK|NG
Woven wire, sheet metal, and knitted
structured packing. 8uilt to spec
or performance requirement.
PANDOM PACK|NG
Fixed & Floating Valve 7ype
3ieve or Perforated
8ubble cap
0ual flow, 8affle
TPAS
Wide variety of random packing types,
sizes and materials in stock! Ask us how
our patented 3uper8lend 2-Pac can
increase your capacity and efhciency. Vanes [Chevron} 3ingle & 0ouble Pocket
Plate Pack Multi Pocket Vanes
lnsertion Mistfix Mist Eliminator
M|ST LL|M|NATOPS
Mesh numerous alloy & plastics
manufact urer wit h a wider range of int ernals and fact ory
amistco.indd 1 23/2/12 12:08:58
Estimating silicon accumulation in coker
naphtha hydrotreaters
S
ilicon poisoning is a major
concern in coker naphtha
hydrotreaters. The source of
silicon in coker naphtha can be
traced back to the delayed coking
process, which typically uses silicon-
containing oils, polydimethylsiloxane
(PDMS), to suppress foaming in the
coker drums. At the elevated
temperature inside the coker drums,
these high molecular weight, silicon-
containing oils crack to form lighter
silicon oil fragments, such as dimers
and trimers of the dimethylsiloxane.
The majority of these silicon oil frag-
ments boil in naphtha range and
therefore are routed to the down-
stream naphtha hydrotreaters
together with the coker naphtha.
Under the operating conditions of
the naphtha hydrotreaters, the sili-
con oil fragments present in the
feed transform to modifed silica
gels and absorb onto the catalyst
surface.
1
As silicon accumulates on
the catalyst surface, it covers active
sites and restricts catalyst pores; the
latter process eventually blocks
access to the active sites. Once sili-
con is bound to the catalyst surface
it cannot be removed and results in
an irreversible loss of catalyst activ-
ity. Without silicon in the feed, the
typical cycle length of a naphtha
hydrotreater is three or more years.
When processing coker naphtha,
the cycle length is typically 12
months. In extreme cases, the cycle
length can be six months or less.
The cycle life of a coker naphtha
hydrotreater is dictated by the sili-
con capacity of the selected catalyst
system and the silicon accumulat-
ing rate. The silicon capacity of a
selected catalyst system is known.
However, the silicon accumulating
Improved sampling and analysis of silicon in the feed enable a signifcant gain in
the cycle life of coker naphtha hydrotreater catalysts
ThIenan Tran, PaTrIck GrIPka and Larry krauS
Criterion Catalysts & Technologies
rate often cannot be determined
due to a lack of accurate feed char-
acterisation data.
Refners normally collect feed
samples for silicon analysis on a
frequent basis. However, because of
the transient nature of the delayed
coking process, the frequency of the
feed sampling is often not suffcient
to determine the actual amount of
silicon being fed to the unit. In addi-
tion to the unrepresentative feed
sampling issue, the commonly used
inductively coupled plasma (ICP)
test method does not accurately
measure the silicon species present
in the coker naphtha feed. These
issues cause the calculated silicon
accumulating rate to be unreliable
and therefore the cycle life of coker
naphtha hydrotreaters often cannot
be predicted. This results in refners
changing out coker naphtha hydrot-
reater catalyst based on fxed cycle
length or silicon slippage. This either
under-utilises the silicon capacity of
the catalyst system or results in an
unplanned shutdown.
In this case study, a hot loop feed
sampling station was installed on a
commercial coker naphtha hydrot-
reater to obtain composite samples.
The weekly composited samples
were tested for silicon using a Shell
proprietary ICP direct injection
nebuliser (ICPDIN) analysis. The
results were used to estimate the
amount of silicon accumulated on
the catalyst as the cycle progressed.
At the end of the cycle, spent cata-
lysts from the unit were analysed to
determine the amount of silicon
accumulated on the catalyst. Results
indicated the silicon deposition esti-
mated using hot loop sampling and
ICPDIN was within 10% of the Si
deposition determined from spent
catalyst analysis. Due to the accu-
racy of the estimate, the cycle life of
the unit could have been extended
up to 4.5 months beyond the sched-
uled 12-month cycle length if the
unit was not shut down due to
furnace fouling.
case study
The coker naphtha hydrotreater in
this case study consists of a guard
reactor followed by two main reac-
tors, which are in parallel. The guard
reactor contains OptiTrap grading
materials and DN-200. The primary
function of the guard reactor is to
saturate diolefns. The main reactors
contain 45% MaxTrap[Si], 20% DN-
140 and 35% DN-3531. All of these
materials are Criterion catalyst
grades. MaxTrap[Si] is a silicon trap
catalyst. DN-140 is a dual functional
NiMo catalyst, which was used to
provide both signifcant silicon
uptake capacity and hydrotreating
activity for the load. DN-3531 is a
high-activity NiMo hydrotreating
catalyst, which provides the major-
ity of the HDS and HDN activity
requirements for the unit. The cata-
lyst system had a silicon capacity of
Feed
composition 18 vol% coker naphtha
82 vol% Sr naphtha
Sulphur, wppm 580
Nitrogen, wppm 18
Operating conditions
WABT, F >600
LHSV in the main reactors 4.1
Product properties
Sulphur, ppmw 0.2
Nitrogen, ppmw <0.2
unit description
Table 1
criterion copy.indd 1 23/2/12 13:08:39
can be off by a factor of 10 to 20.
2
To obtain reliable silicon content
data, the weekly composite samples
were sent to the Shell Global
Solutions Westhollow Technology
Center for silicon analysis. A Shell
proprietary ICP analysis was used
to measure the silicon present in
the weekly composite sample. This
analysis uses a custom design DIN
to introduce the sample into the
ICP. Based on results of extensive
research, the error margin of the
ICPDIN test is typically less than
10%.
Spent catalyst analysis
After 10 months online, the unit was
shut down due to furnace fouling
issues. The decision was made to
change out the catalyst while the
unit was down. The spent catalyst
was unloaded by vacuuming, which
allowed an accurate silicon deposi-
tion profle to be constructed.
Samples of the spent catalyst were
collected to determine the silicon
uptake capacity of each type of cata-
lyst and the total amount of silicon
accumulated on the catalyst system.
Results and discussion
Results of the weekly silicon analy-
sis indicated the silicon content in
the feed varied from 0.26 to 6.5
wppm, with 62% of the data popu-
lation in the 1.5 and 2.5 wppm
range (see Figure 1). Without
the composite feed sampling
programme, the periods with high
or low silicon content could have
been unaccounted for. This could
have signifcantly affected the esti-
mated average feed silicon content.
This shows that a good feed
sampling programme is required to
capture the changes in the silicon
content of the feed and accurately
estimate the total silicon fed to a
naphtha hydrotreater unit.
Based on the weekly test results of
the silicon content in the feed, the
silicon accumulation was calculated
as the cycle progressed. At the time
the unit was changed out, the total
silicon accumulation was estimated
to be 4770 lb (~ 55% of the total
catalyst system silicon capacity).
Results of the spent catalyst anal-
yses indicated the actual amount of
silicon accumulated on the catalyst
8750 lb and was scheduled for a 12-
month run length. The unit process
conditions and feed properties are
shown in Table 1.
Feed sampling
The majority of coker naphtha
hydrotreaters sample feed once
per week for silicon analysis.
Fluctuations in the amount of coker
naphtha in the feed and the amount
of foam-suppressing oils used in
the coker drums will affect the sili-
con content in the feed. Recognising
the fact that frequent feed sampling
is important to capture the changes
in the silicon content, a hot loop
feed sampling system was installed.
This sampling system was designed
to collect and composite a defned
number of feed samples throughout
the week to obtain weekly compos-
ite samples.
Silicon analysis of coker
naphtha feed
The silicon present in the coker
naphtha is in the form of dimers
and trimers of the dimethylsiloxane.
These dimethylsiloxane molecules
are volatile. Due to their volatility,
the standard ICP method, which is
used by most refners, cannot
accurately measure the silicon
content in the coker naphtha.
Depending on the form of the sili-
con compound and the dynamics of
the sample introduction system of
the ICP method being used, the
silicon concentration determined
6
7
5
4
3
2
1
0 5 10 15 20 25 30 35 40
S
i

c
o
n
t
e
n
t
,

w
t

p
p
m
Week on st ream
0
6000
10000
9000
8000
7000
5000
4000
3000
2000
1000
0 0
3
0
6
0
9
0
2
1
0
5
1
0
8
1
0
1
2
0
4
2
0
7
2
0
0
3
0
3
3
0
6
3
0
9
3
0
2
4
0
5
4
0
8
4
0
1
5
0
4
5
s
b
l

,
n
o
c
i
l
i
S
Days on stream
0
The estimated Si accumu|ation
is 9 |ower than the actua|
gure. This difference is within
the error margin of the ana|ytica|
test used in this study case.
Si breakthrough
capacity (8750 |bs}
If the unit had not been shut down due to
furnace fou|ing, the cyc|e cou|d have been
extended up to 4 months beyond the
schedu|ed 12-month cyc|e |ength.
Linear actua| Si accumu|ation
Budgeted Si accumu|ation
Estimated accumu|ated Si
Actua| Si accumu|ation
Linear estimated accumu|ated Si
Figure 1 Weekly average Si content in feed
Figure 2 Silicon accumulation
was 5220 lb, which equated to 9%
higher compared to the estimated
number. Results of the spent cata-
lysts also conrmed that the
ICPDIN test method used in this
case study can accurately measure
the silicon content in the coker
naphtha. If the unit was not shut
down due to furnace fouling, the
cycle length could have been
extended up to 4.5 months beyond
the scheduled 12-month cycle
length (see Figure 2).
In addition, results of the spent
catalyst analyses showed that the
guard reactor picked up a minimal
amount of silicon (see Figure 3).
This was expected due to the low
operating temperature (<450F) in
the guard reactor. The MaxTrap[Si]
in the main reactors picked up as
much as 21 wt% silicon. Almost no
silicon was deposited in the DN-140
and DN-3531 layer due to the
premature shutdown of the unit.
Conclusions
In coker naphtha hydrotreaters, the
quality of the estimates of silicon
deposition on a catalyst system
depends on the quality of the feed
sampling program and the accuracy
of the analytical method used to
determine the silicon content in the
feed samples. A hot loop sampling
system can be used to obtain
composite feed samples, which
capture the changes in the silicon
content of the feed. The ICPDIN
analysis used in this case study
accurately measured the silicon
content in coker naphtha feed. The
combination of the good feed
sampling program and the accurate
silicon analysis resulted in a more
reliable estimate of silicon deposi-
tion and could have extended the
cycle length up to 4.5 months
beyond the scheduled 12-month
cycle length.
Acknowledgment
The expertise provided by Michael Shepherd,
Research Chemist in Elemental Analysis
department at Shell Global Solutions
Westhollow Technology Center, is gratefully
acknowledged.
References
1 Kellberg L, Zeuthen P, Jakobsen H J, Journal of
Catalysis, 1993, 143, 45-51.
2 Sanchez R, Todoli J-L, Charles-Philippe, Mernet
J M, J. Anal. At. Spectrom., 2009, 24, 391-401.

Larry Kraus is Hydroprocessing Product
Manager with Criterion Catalysts &
Technologies. He holds BS degrees in chemistry
and chemical engineering from Kansas State
University and MS and PhD degrees in chemical
engineering from Northwestern University.
Email: lawrence.kraus@cri-criterion.com
Patrick Gripka is a Senior Technical Service
Engineer for Criterion Catalysts & Technologies,
primarily supporting the DHT and FCC PT
applications in the Americas. He holds BSChE
and MSChE degrees from the University of
Missouri Rolla.
Email: pat.gripka@cri-criterion.com
Thienan Tran is Senior Technical Service
Engineer with Criterion Catalysts &
Technologies. She holds a BS degree in chemical
engineering from the University of Houston,
Texas. Email: thienan.tran@cri-criterion.com
15
25
20
10
5
0 10 20 30 40 50 60
S
i
l
i
c
o
n

u
p
t
a
k
e
,

w
t
%
Feet down Rx syst em Guard i nl et = 0
React or i nl et = 30
0
Guard reactor silicon
Main reactors silicon
OptiTrap
products
DN-200 MaxTrap (Si) DN-140 DN-3531
Figure 3 Spent catalyst silicon prole
Our new clean-fuels
plant is straining the
auxiliary units!
How can we boost their capacity
without major construction?
Read more on this topic at
www.amistco.com
TO MEET CLEAN FUELS require-
ments for gasoline, a medium-sized
Midwestern renery added a low-
sulfur fuels technology plant. Gasoline
throughput was unchanged. However,
the sharp increase in sulfur removal
required more hydrogen from the
hydrogen unit and sent more sulfur
gases to the amine treaters and
downstream sulfur units. These
auxiliary units became bottlenecks,
overdriven at the cost of product
purity and amine consumption.
On studying the hydrogen, amine, and
sulfur units, ACS-Amistco found many
opportunities for improving separation
efciency and capacity. The problems
were solved without major construction
by applying modern technology to mist
eliminators, liquid-liquid separators,
and tower trays and packing. Results
included haze-free product and
reduced amine consumption. Now a
diesel clean-fuels plant is being added.
Prore: 281-331-595 Fax: 281-585-1Z80
amistco@amistco.com
21-rr Erergercy 3erv|ce: 800-839-3Z1
ACS-AMISTCO
Reaching our customers globally
through innovation
At Sd-Chemie, we create original solutions that shape the way you see the future.
Innovation is core to our business and has been our focus through our 150-year history. Our catalysts
boost the performance and value of our customers operations, limiting their impact on the environment
and ensuring that nite raw materials and energy are used efciently.
Di scover what t he fut ure hol ds for your busi ness at www.sud-chemi e.com
SC Syngas Anz_A4_2012-01-18 f01.indd 1 18.01.12 14:39
sud chemie.indd 1 23/2/12 12:11:02
FCC catalyst coolers in maximum
propylene mode
M
arket demand for propylene
has placed a strong empha-
sis on many FCC units to
run in maximum propylene mode.
Increasing reactor temperatures in
pursuit of higher propylene, without
regenerator heat removal, can raise
the regenerator temperature to unac-
ceptable levels, resulting in
accelerated catalyst deactivation,
degraded cracking selectivity and a
need for exotic mechanical design to
avoid equipment damage.
This article presents FCC model-
ling that demonstrates the utility of
continuous heat removal from the
regenerator for maximum propylene
operations. Developments in FCC
catalyst cooling technology have
given reners a exible and reliable
option to confront the heat balance
challenges of maximum propylene
FCC operations.
In an unconstrained environment,
increasing FCC propylene produc-
tion can be as easy as increasing the
reactor temperature. However, in
most cases, increasing the propylene
yield is not that easy, as most FCC
units are already operating against
several physical and economic
constraints. More commonly, regen-
erator coke burning and the vapour
recovery unit limit capacity, increas-
ing the reactor temperature without
rst reducing the FCC feed rate.
In grassroots FCC installations,
coke burning and vapour recovery
capacity can be built into the unit
design, and existing FCC units can
be revamped to include the requisite
coke burning and vapour recovery
capacity for increasing propylene
production. However:
Even with abundant coke burning
and vapour recovery unit capacity, a
Catalyst cooling technology for continuous heat removal from the regenerator in
maximum propylene operations can avoid damage to FCC catalyst and equipment
RAHUL PILLAI and PHILLIP NICCUM
KBR
high regenerator temperature can
emerge as a major constraint to
increasing reactor temperature
because of the impact of the higher
temperature on the unit heat
balance
1
FCC operators can effect a reduc-
tion in equilibrium catalyst activity
to offset the increasing regenerator
temperature that would naturally
come from increasing the reactor
temperature, but reducing catalyst
activity runs counter to the more
basic objective of increasing propyl-
ene production.
History of FCC propylene production
The rst commercial FCC unit was
built by The M W Kellogg Company
in Standard Oil of New Jerseys
Baton Rouge, Louisiana, renery
and commissioned in May 1942.
Between 1942 and 1944, Kellogg
built 22 of 34 FCC units constructed
throughout the US, and the FCC
process quickly became a major
contributor to worldwide propylene
and butylene production.
Rare earth-exchanged Y zeolite
catalyst was rst synthesised by
Mobil in 1959. By the late 1960s,
over 90% of US FCC units were
operating with the Mobil-invented
zeolite catalyst. The high activity of
the zeolite catalysts, compared to the
earlier amorphous catalysts, greatly
improved the gasoline yield and
reduced coke and dry gas yields
Primary feed riser
Disengager with internal
reactor cyclone system
2nd riser for
naphtha recycle
Recycle naphtha injection
Regenerator
Atomax-2
fresh feed injection
Figure 1 Maxon unit
kbr.indd 1 23/2/12 13:15:37
it simultaneously relieved constraints
on both vapour recovery unit capac-
ity and regenerator operating
temperature.
The Maxon FCC Process intro-
duced by M W Kellogg and Mobil
in 1985 (see Figure 1) is designed to
maximise the production of propyl-
ene, ethylene and aromatics from
traditional FCC feedstocks by
combining the effects of FCC
catalyst, ZSM-5 additive and a high-
severity second riser designed to
crack surplus naphtha and C
4
s into
incremental light olens and
aromatic naphtha.
6
Like closed
cyclones, the Maxon FCC Process
also provides some relief to the heat
balance while operating at high reac-
tor temperatures because of the
limited delta coke from recracking
the recycled naphtha and C
4
feed-
stocks. The recycled naphtha and
C
4
s essentially act as a regenerator
refrigeration system while simulta-
neously serving to increase
propylene production in the high-
severity second riser.
KBR dense-phase catalyst cooler
For many decades and until recently,
FCC catalyst coolers have been
considered only as a means to effec-
tively process high-carbon-
residue FCC feedstocks, where the
impact of Conradson carbon residue
(CCR) on delta coke is a fundamen-
tal driver of the FCC heat balance.
CCR in the feed increases the
amount of coke deposited on the
catalyst as it passes through the
riser. The increased concentration of
coke on the catalyst as it passes
through the reactor is referred to as
delta coke. Without intervention,
increasing delta coke leads to a high
regenerator temperature, reducing
FCC feed conversion due to lower-
ing of the catalyst-to-oil ratio and
accelerated catalyst deactivation. To
mitigate the impact of increasing
feed CCR on regenerator tempera-
ture, the heat released during
catalyst regeneration must be
controlled or heat must be removed
from the system. The best option for
controlling the heat balance with
increasing delta coke is often the use
of a catalyst cooler and/ or regenera-
tor operation in partial CO
combustion mode.
from the FCC units, but the cata-
lysts high hydrogen transfer
characteristic greatly reduced the
light olen yield and gasoline
octane.
In the 1970s, after the introduction
of zeolite catalyst, FCC unit design
and operation evolved to regain
some of the lost octane and light
olen yield, primarily with a higher
reactor operating temperature and
riser cracking.
2
Increasing reactor
temperatures increased the light
olen yield, but this came at the
expense of an increased yield of dry
gas a lower-valued FCC product.
During the 1980s, Mobil intro-
duced two new technologies with
application to increasing the produc-
tion of light olens and octane while
limiting incremental dry gas produc-
tion: Mobil developed the ZSM-5
catalyst additive to crack low-octane
(linear) gasoline-boiling-range olens
and parafns into light olens, and
invented closed cyclones, which
minimise product vapour residence
time between the riser outlet and the
main fractionator.
3, 4
In addition to the reduction in dry
gas, the closed cyclone riser termina-
tion system reduced delta coke,
especially on units that previously
employed low catalyst separation
efciency riser termination devices.
Therefore, the closed cyclone system
was especially adept at increasing
FCC propylene production because
Catalyst in
Catalyst
return
Inner tube
Slide
valve
Fluidisation
air
Tubesheet
Water in
Tubesheet
Scabbard
outer tube
Water and
steam out
High heat
transfer coefficient
Flow-through design,
high mean temperature
differential
No fluidisation impingement
on tubes
Upflow boiling with natural
boiler feed water circulation
Tube bundle easily removed
High turndown capability
Commercially proven design
Water in
Water and
steam out
Catalyst in
Catalyst out
Fluidisation
air
Figure 2 KBR dense-phase catalyst cooler
Figure 3 Key features of catalyst cooler
kbr.indd 2 23/2/12 13:15:45
Main Products
Regeneration of spent catalysts
XpresS ex-situ activation (sulphiding) of fresh and regenerated catalysts
Rejuvenation of catalysts
Resale of once regenerated catalysts
Sale of GUARDIAN material (guard bed catalysts)
Additional Services
Laboratory expertise of spent catalysts
Pilot plant ex-situ activation
Technical support
Storage of catalysts
Tricat GmbH Catalyst Service Bitterfeld

OT Greppin
Tricat-Str. (ChemiePark Areal B-Ost)
06803 Bitterfeld-Wolfen
Germany
Phone: +49 3493 75910
Fax: +49 3493 75999
Internet: www.tricatgroup.de
E-mail: info@tricatgroup.de
Tricat, Inc.
260 Schilling Circle
Hunt Valley, MD 21031
USA
Phone: (410) 785 7900
Fax: (410) 785 7901
Internet: www.tricatgroup.com
E-mail: sales@tricatgroup.com
- TRICAT
tricat.indd 1 23/2/12 12:15:06
The KBR dense-phase catalyst
cooler (see Figures 2 and 3) was
commercialised in 1991 based on
extensive KBR experience in high-
temperature ammonia applications
and cold fow modelling of the cata-
lyst side at KBRs Houston
Technology Development Center.
7

Two distinguishing features of the
KBR dense-phase catalyst coolers
impart fexibility in heat removal
duty and resistance to tube failure
from erosion by the catalyst. The
frst feature is a gas vent line at the
top of the cooler fuid bed that
prevents catalyst backmixing
between cooler and regenerator
whenever the cooler catalyst circula-
tion is stopped, thereby providing
complete heat removal turndown
capability. Without the vent, cold
fow modelling has demonstrated
that backmixing between the cooler
and regenerator (and therefore heat
transfer in a commercial unit) will
occur due to fuidisation of catalyst
in the inlet duct by aeration gas
travelling from the cooler back into
the regenerator bed. With the vent
in place, cooler aeration gas returns
to the regenerator through the vent
rather than through the catalyst inlet
duct, allowing catalyst in the inlet
duct to defuidise whenever catalyst
circulation is stopped. The other
distinguishing feature of the KBR
dense-phase catalyst cooler design is
that the tube sheet is located above
the tubes, which has several impor-
tant ramifcations:
Downward-hanging tubes allow
the cooler shell fuidisation air to be
introduced well below the tubes,
preventing any possibility of cooler
fuidisation air jet impingement on
the tubes, which could cause an
erosion-related tube failure
Since steam is generated upfow
between the inner and outer tubes,
the cooler can utilise natural boiler
feed water circulation, eliminating
the need for forced boiler feed water
circulation pumps, along with their
associated cost and reliability issues
The orientation of the tube bundle
also facilitates maintenance and
inspection of the cooler because the
tube bundle can be pulled from the
top of shell.
There are now 16 KBR dense-
phase catalyst coolers in operation,
and there have been no reports of
erosion-related tube failure in any of
these installations.
Use of catalyst cooler for propylene
production
Increasing reactor temperatures in
pursuit of higher propylene increases
the regenerator bed temperature
and, without regenerator heat
removal, raises this temperature to
unacceptable levels, resulting in
accelerated catalyst deactivation,
degraded cracking selectivity and a
need for exotic mechanical design to
avoid equipment damage.
The utility of FCC regenerator
heat removal for maximising propyl-
ene production is demonstrated by
way of a hypothetical example using
KBRs proprietary FCC yield model-
ling software. The base case for the
exercise is a hypothetical 40 000 b/ d
FCC operation on a good-quality
hydrotreated VGO feedstock. A
summary of the base case feedstock
quality is shown in Table 1. The
study cases are based on a catalyst
activity of 72 and a constant feed
preheat temperature of 650F
(343C).
Figure 4 shows that, while oper-
ating at the base case 980F (527C)
riser outlet temperature, the
1420
1400
1380
1360
1340
1320
980 990 1000 1010 1020 1030 1040 1050
R
e
g
e
n
e
r
a
t
o
r

t
e
m
p
e
r
a
t
u
r
e
,

F
Ri ser out l et t emperat ure, F
1300
Figure 4 Impact of ROT on regenerator temperature (w/o catalyst cooler)
Feed rate, BPSD 40 000
API 28.00
Molecular weight 445
Sulphur, wt% 0.05
Total nitrogen, ppmw 2
Watson K 12.27
Conradson Carbon, wt% 0.20
Distillation type D2887, F
10% 700
30% 754
50% 804
70% 883
90% 997
Feedstock properties
Table 1
60
90
80
70
50
40
30
20
10
980 990 1000 1010 1020 1030 1040 1050
C
o
o
l
e
r

d
u
t
y
,

m
m
B
T
U
/
h
r
0
Figure 5 Required catalyst cooler duty for 1350F max regenerator temperature
kbr.indd 3 23/2/12 16:24:15
regenerator temperature heat
balances at about 1310F (710C), a
temperature that is considered very
compatible with good FCC catalyst
activity maintenance and minimal
generation of dry gas and other
thermal cracking products in the
feed injection zone. As would be
expected, based upon the basic
tenets of the FCC heat balance, the
regenerator heat balanced tempera-
ture increases with the increasing
riser outlet temperature.
For the sake of this example,
considering the deleterious impact of
regenerator temperature on catalyst
activity, product yield selectivity and
mechanical reliability, the regenerator
bed temperature will be limited to a
maximum of 1350F (732C). As
Figure 5 shows, the maximum allow-
able riser outlet temperature will be
approximately 1010F (543C) when
limiting the regenerator temperature
to 1350F (732C).
The increasing regenerator temper-
ature is primarily driven by the
increasing temperature of the spent
catalyst being returned to the
regenerator.
Utilising a variable-duty dense-
phase catalyst cooler, it is possible to
keep the regenerator temperature
from exceeding 1350F (732C) at
riser outlet temperatures exceeding
the 1010F (543C) value. For
purposes of this example, the riser
outlet temperature is increased to as
high as 1050F (566C). Figure 5
shows the catalyst cooler duty
required to maintain the regenerator
Table 2
bed temperature of 1350F (732C)

at riser outlet temperatures greater
than 1010F (543C).
The point of this exercise is shown
in Figure 6, which demonstrates
Alternativesfor CatalystcoolerLimitedROT Reducedactivity
achievinga1350F1050FROT 1010FROT 1050FROT
regenerator & & &
temperature 72MAT 72MAT 58MAT
FCC yields Wt% Vol% Wt% Vol% Wt% Vol%
C
2
and lighter 2.18 -- 1.77 -- 2.30 --
Propane 2.91 5.09 2.16 3.77 1.64 2.87
Propylene 7.77 13.21 6.10 10.37 6.24 10.61
Total C
4
s 15.61 23.35 14.30 21.48 11.64 17.32
FCC gasoline (C
5
-204C) 51.03 61.22 52.73 63.35 50.61 60.89
Light cycle oil (204-360C) 10.48 10.01 13.52 13.01 16.50 15.99
Slurry oil (360C+) 4.68 3.72 5.36 4.48 6.70 5.88
Coke 5.34 -- 4.06 -- 4.37 --
Total liquid product 116.60 116.46 113.56
Conversion @ 204C 84.8 86.3 81.1 82.5 76.8 78.1
Alternativeregeneratortemperaturecontrolstrategies
Table2
that, starting from a base case
propylene yield just below 9.0 vol%,
the propylene yield can only be
increased to about 10.5 vol% with-
out the use of regenerator heat
11.50
13.00
13.50
12.50
12.00
11.00
10.50
10.00
9.50
9.00
980 990 1000 1010 1020 1030 1040 1050
P
r
o
p
y
l
e
n
e

p
r
o
d
u
c
t
i
o
n
,

v
o
l
%
8.50
Figure6Impact of ROT on propylene
HANDLING THE WORLDS MOST
IMPORTANT PRODUCTS EVERYDAY
CHEP Catal yst and Chemi cal Contai ners provi des Catal yst-Bi ns to the petrol eum reni ng,
gas processi ng and petrochemi cal manufacturi ng i ndustri es.
Our Catal yst-Bi ns offer you the fol l ow i ng advantages:
- Efci ent transportati on - Opti mi zati on of storage around reactor
- Efci ent handl i ng - Opti mi zed HSSE
- Reduced reactor dow n ti me - Mi ni mi zati on of w aste generati on
For more i nformati on pl ease vi si t our w ebsi te w w w.chepccc.eu or contact us:
CHEP CATALYST & CHEMICAL CONTAINERS B.V. - Scheepmakerstraat 13-15 - 2984 BE Ridderkerk - The Netherlands
Tel +31 180 33 11 44 - Fax + 31 180 41 40 90 - ccceur@chepccc.com
kbr.indd 4 23/2/12 16:25:06
TRANSFORMING TODAYS PRODUCTS
FOR TOMORROWS INNOVATIONS
Market Leader in Catalyst Services, Products and Tolling
Cat alyst Services:
Regeneration Reactivation Presulding Density Grading Length Grading
Catalyst Acquisition Metal Reclaim Drying Resale Quanta
Product s:
Claus Catalysts Adsorbents Purication Dehydration FCC-additives
Inert Support Media
Tolling:
Extrusion Calcination Impregnation Forming Drying Milling Blending
Classifying
GLOBAL PRESENCE, WORLDWIDE SATISFACTION
Porocel_adv_FC_210x297mm.indd 1 20-02-12 09:26
porocel.indd 1 23/2/12 12:16:42
removal while honouring the 1350F
(732C) regenerator temperature
constraint. It also shows that the
propylene yield can be increased to
as high as 13.5 vol% while using the
catalyst cooler to control the regen-
erator temperature.
The lessons learned from the
examples presented here can be
further demonstrated by comparing
some yield cases representing alter-
native routes for controlling
regenerator temperature in search of
increasing the propylene yield. Table
2 shows two cases taken from the
examples presented above, contrast-
ing the difference between
controlling the regenerator tempera-
ture with a catalyst cooler versus
simply limiting the reactor tempera-
ture. A third case has also been
developed and included in the table.
It shows the expected results for just
reducing the catalyst activity as
needed to keep the regenerator
temperature in the desired range
without a catalyst cooler as the riser
outlet temperature is increased to
1050F (566C).
Comparing the three cases,
depending on a refners objectives,
different cases may be the preferred
alternative:
If the only objective is the maximi-
sation of propylene, the frst case
with the catalyst cooler is the clear
winner
If the primary objective is the
maximisation of propylene and the
secondary objective is the maximisa-
tion of total liquid product yield, the
third case with the low catalyst
activity would clearly be the least
favourable among the three
alternatives
If the production of light cycle oil
and the production of propylene are
both primary objectives, the case
utilising low catalyst activity rather
than the catalyst cooler may be
preferred.
One of the broader conclusions
that can be drawn from this study is
that a refner might want to invest
in an FCC catalyst cooler if a large
investment is being made to increase
FCC coke burning and vapour
recovery unit capacity for the
purpose of increasing propylene
production.
History of catalyst coolers
The early FCC catalysts produced
coke yields as high as 12 wt%. This
resulted in a reactor-regenerator
system, where the amount of heat
liberated by coke combustion was
more than could be removed by the
process streams. In 1942, M W
Kellogg introduced Recycle Catalyst
Coolers on a Kellogg Model II FCC
unit to solve this heat balance issue.
By 1948, there were 22 commercial
Recycle Catalyst Coolers in opera-
tion. Most of these coolers had 988,
1 1/ 2in O.D. tubes, which were 22 ft
long. The design duties of these
coolers were as high as 100 million
Btu/ h each. These coolers had cata-
lyst at 25 ft/ s moving through the
tubes, transferring heat to water/
steam in the shell. A catalyst stand-
pipe withdrew catalyst from the
bottom of the regenerator bed that
was conveyed upwards through the
cooler tubes at high velocity. The
catalyst cooler duty was varied by
adjusting the catalyst circulation rate
with a slide valve.
These early catalyst coolers were
mechanically complex, with large
expansion joints in the shell to
accommodate the difference between
the inlet channels/ tubes and the
shell. The problems encountered in
these coolers were mainly related to
erosion due to internal solids mixing,
which included:
Erosion of cooler outlet cone
transition
Erosion of tubes and tube sheets
at the cooler inlet
Failure of linings in the stand-
pipes and carrier lines
Erosion of the carrier line at the
air/ catalyst mixing point.
The run lengths obtained with
these dilute-phase coolers were typi-
cally three to four months, at the
end of which tube bundle repair or
replacement was required. As the
fowing catalyst was in dilute phase
(density ~3 lb/ ft
3
), the heat transfer
coeffcient was relatively low. An
analysis of commercial catalyst
cooler data showed the heat transfer
coeffcient ranging from 9-35 Btu/h-
ft
2
-F, with most of the coeffcients
falling between 18 and 34 Btu/ h-ft
2
-
F.8 From 1950, cracking catalyst
with better coke selectivities became
available, and these catalyst coolers
were no longer used.
The advent of resid cracking in the
early 1960s paved the way for regen-
erator bed coils. The frst residue
cracker, built in 1960 for Phillips
Petroleum, had 15 layers of horizon-
tal hairpin coils, which covered 320
degrees of the vessel circumference
near the wall of the regenerator. The
high-pressure steam for use in the
refnery was generated by passing
boiler feed water through the coils.
In the original design of these bed
The basic science of heat transfer between a fuidised bed and a confning wall was
described by Chin-Yung Wen and Max Leva in a 1956 paper:
8

For a fuidised bed confned in a pipe, there will always be a laminar fuid flm in
immediate contact with the pipe. The heat transfer mechanism from the pipe wall
to the laminar fow is via conduction, and this constitutes the major resistance
between the fuid bed and the pipe wall
The heat transfer from the laminar flm to an adjacent buffer layer is through
turbulent mixing due to the eddy movement of the fuid
From the boundary layer to the core of the fuid bed, the prevalant heat transfer
mechanism is by turbulent mixing of the solid particles
The higher the thickness (or width) of the laminar layer, higher will be the
resistance for heat transfer from tube walls to the fowing fuid. Therefore, up to a
point, an increase in velocity of the fuid in the bed that decreases the laminar layer
thickness will leads to an increase in heat transfer coeffcient
As the velocity of the fuid in the bed is increased futher, the heat transfer
increases to a max heat transfer value, beyond which it decreases with velocity.
This is because the bed expands with higher gas fow and increases the solid particle
spacing, reducing the rate of heat transfer via turbulent mixing of the particles.
Fundamentals of fuid bed to tube heat transfer
kbr.indd 5 23/2/12 13:16:13
coils, a heat transfer coefcient of 60
Btu/ h-ft
2
-F was used.
7
This proved
to be conservative and the steam
generated was greater than expected.
This high heat transfer coefcient
was the result of the coils immersed
in a relatively high-density (25 to 35
lb/ ft
3
) uidised bed in the regenera-
tor. Unlike the dilute phase, the low
velocity in the regenerator (2 to 3 ft/
s) resulted in a lower erosion of the
coils.
The main problem encountered by
the regenerator coils was on the
inside (water) side. The use of low-
quality boiler feed water resulted in
a mineral deposition and eventual
failure due to pitting. Also, the low
velocity in the coils resulted in a
slug or stratied ow, ultimately
leading to failure due to thermal
stresses. To avoid this problem, the
ow inside the tube was maintained
in the bubble ow regime, restrict-
ing the exibility of these coolers to
cope with changes in feed rate and
feed quality.
In a 1963 paper, W J Danzinger presented
the development of a heat transfer
coefcient correlation for a uidised
catalyst bed moving vertically based on
commercial data from a 1940s vintage
FCC recycle catalyst cooler designed
by M W Kellogg.
9
The reported study
was conducted for a recycle catalyst
cooler operating in dilute phase as well
as dense phase:
A correlation of the heat transfer
coefcients of air-uidised cracking
catalyst of about 50-micron average
particle diameter in vertical transport
is presented. The correlation, based on
commercial data for cooling and the
data of Farber and Morley
10
for heating,
is:
(h*d/k) = 0.0784 * ((D*G)/)
0.68
*
(Wa/Wg)
0.45
The correlation covers Reynolds
numbers from 178 to 25 400, solids-to-
gas weight ratios of 2 to 446, and tube
ID from 0.689 to 1.497 inches
Data were obtained on recycle
catalyst coolers of two designs, both
vertical, single-tube pass, removable-
bundle, re-tube boilers with the
air-catalyst mixture owing upward
through the tubes. The steam drum
was elevated sufciently to limit
vapourisation in the boiler shell to
about 10% of the thermosyphon water
ow. In one design, the single-section
cooler tube bundle contained 988 steel
tubes, 1 1/2 - inch OD by 22 ft long. The
second design provided two sections in
parallel, each bundle containing 580
steel tubes, 1 7/8-inch OD by 19 ft
long
The arrangement and typical
instrumentation is shown in Figure
7. Catalyst ow was controlled by a
slide valve, which was positioned by a
temperature element in the regenerator
bed
Figure 8 shows the variation of heat
transfer coefcient with Reynolds
number.
Mixtures in vertical transport
Air
FR
T1
T1
T1
FRC
PR
PR
PR
TR
LLR
PR
TRC T1
Feed
water
Steam
Flue gas
Catalyst
to reactor
Catalyst
from reactor
Regenerator
Steam
drum
Standpipe
Recycle
catalyst
cooler
Figure 7 Arrangement and instrumentation of recycle catalyst cooler
10
6
8
4
3
2
100
60
80
40
30
20
1 2 3 4 6 8
10
2
10
3
10
4
(
h
D
/
k
)
/
(
w
s
/
w
g
)
0
.
4
5
DG/
1 2 3 4 6 8 1 2 3 4 6 8
Kellog 1
1
/
2
in. tubes, cooling
Kellog 1
7
/
8
in. tubes, cooling
Farbar and Morley heating
(averaged ordinates)
Figure 8 Correlation of data on heat transfer of air-cracking catalyst mixtures in
vertical upward transport
kbr.indd 6 23/2/12 13:16:24
The research continued for a cata-
lyst cooler that was more exible and
operated with a low catalyst velocity
to avoid the erosion, which compro-
mised the dilute-phase coolers. In
1991, this resulted in the modern-day
KBR dense-phase catalyst cooler with
water and steam on the tube side and
hot, slow-moving catalyst on the shell
side. Among the 16 KBR dense-phase
catalyst coolers in operation, there
have been no reports of erosion-
related tube failure.

Conclusions
A rener should consider investing
in an FCC catalyst cooler if a large
investment is being made to increase
FCC coke burning and vapour recov-
ery unit capacity for the purpose of
increasing propylene production.
Investing in coke burning and vapour
recovery expansion without a catalyst
cooler can result in under-utilised
investments if a high regenerator
temperature prevents the FCC unit
from producing the propylene
production targeted by the debottle-
necking project.
Once a decision has been taken to
install an FCC catalyst cooler, care-
ful consideration should be given to
aspects of the cooler technology that
impact the coolers operating exi-
bility and on-stream reliability.
References
1 Pillai R, Niccum P K, FCC catalyst coolers
open window to increased propylene, Grace
Davison FCC Conference, Munich, Sept 2011.
2 Whittington E L, Murphy J R, Lutz I
H, Catalytic cracking modern designs,
Symposium on Advances in Gasoline
Technology, Division of Petroleum Chemistry,
Inc, American Chemical Society New York
Meeting, 27 Aug-1 Sept 1972.
3 Andersen C D, Dwyer F G, Koch G, Niiranen
P, 9th Ibero American Symp. Catal., Lisbon,
Portugal, 1984.
4 Miller R B, Johnson T E, Santner C R, Avidan
A A, Johnson D L, FCC reactor product
catalyst separation ten years of commercial
experience with closed cyclones, 1995 NPRA
Meeting.
5 Miller R B, Niccum P K, Claude A, Silverman
M A, Bhore N A, Chitnis G K, McCarthy S J, Liu K,
MAXOFIN: a novel FCC process for maximizing
light olens using a new generation ZSM-5
additive, 1998 NPRA Meeting, Mar 1998.
6 Niccum P K, Gilbert, M F, Tallman M J,
Santner C R, Future renery FCCs role in
renery/petrochemical integration, 2001 NPRA
Meeting, Mar 2001.
7 Johnson T E, Improve regenerator heat
removal, Hydrocarbon Publishing, 55-57, Nov
1991.
8 Wen C Y, Leva M, Fluidized-bed heat
transfer: a generalized dense-phase correlation,
A.I.Ch.E. Journal, December 1956, 482-488.
9 Danzinger W J, Heat transfer to uidized
gas-solids mixtures in vertical transport, I&EC
Process Design and Development, 2, 269-276,
1963.
10 Farbar L, Morley M J, Heat transfer to
owing gas-solid mixtures in a circular tube,
Ind. Eng. Chem., 1957, 49(7), 1143-1150.
Independent Catalyst Test
Reports 2010, 2011 & 2012
We have tested catalysts from all major suppliers. Our
reports provide full test data, analysis, and activity rank-
ings. Our reports are available for immediate delivery.
Our team brings the industrys leading independent
lab, innovative test methods, and sound, objective
analysis. Our real-world experience includes 150
catalyst testing runs and 100 catalyst decisions.
Contact us today for
independent test data,
reports, and catalyst
rankings.
Hoekstra Trading LLC
www.hoekstratrading.com
george.hoekstra@hoekstratrading.com
+1 630 330-8159
OUR REPORT WILL HELP YOU:
t4JNQMJGZDBUBMZTUTFMFDUJPO
t$IPPTFUIFSJHIU
DBUBMZTUXJUI
DPOEFODF
t4BWFIVOESFETPG
UIPVTBOETPO
ZPVSOFYUDBUBMZTU
QVSDIBTF
Rahul Pillai is Process Engineering Associate,
KBR Fluid Catalytic Cracking Technology,
performing process engineering design
activities for grassroots FCC units, FCC revamp
projects, technology proposals, technical
service, and plant start-up assignments. He
holds a MS degree in mechanical engineering
from Texas A&M University.
Phillip Niccum, Director, KBR Fluid Catalytic
Cracking Technology, joined KBRs Fluid
Catalytic Cracking (FCC) team in 1989
following nine years of related work for a major
oil company.
kbr.indd 7 23/2/12 13:16:34
Ai m for More Di esel , Better Di esel
ExxonMobils MIDW
TM
*
technology is a commercially
proven process using an advanced proprietary
catalyst for the production of ultra-low sulfur and
low cloud-point diesel.
Benefits include:
Low-pressure, xed bed process |mproves product
y|e|d and qua||ty
Increased su|fur remova| and reduced naphtha y|e|ds
Propr|etary cata|yst** |somer|zes waxy n-parafns |n
gas o||, to|erant to H
2
S, ammon|a, and organ|c S and N
Stand-a|one or |ntegrate w|th others
* MIDW is a trademark of ExxonMobil Research and Engineering Company
** Available from ExxonMobil Catalyst Technologies LLC
ExxonMob|| Research and Eng|neer|ng
Company's MIDW* Techno|ogy
Improves Product Y|e|d and Qua||ty
3225 Gallows Road, Fairfax, Virginia 22037-0001, USA
www.exxonmobil.com/refningtechnologies +1-703-846-2568 fax +1-703-846-3872 tsl.email@exxonmobil.com
exxon.indd 1 23/2/12 12:18:28
Decrease catalyst costs by regeneration,
analysis and sorting
O
il refning and petrochemi-
cal plants are currently
undergoing serious chal-
lenges from a technical and
economical point of view, as well
as from an environmental point of
view. The volume of installed
capacity of hydroprocessing catalyst
is increasing in order to cope with
the development of nearly sulphur-
free clean fuels and the need for
deeper heavy ends conversion. At
the same time, low refning margins
are driving industry to cost optimi-
sation, while environmental
regulations are getting tougher.
In addition to the growing fresh
hydroprocessing catalyst market,
regenerated catalyst usage is
increasing, with an estimated
worldwide consumption of roughly
40 000 t/y. This additional tonnage
of regenerated catalysts is extremely
useful for refneries hoping to
reduce their procurement expenses
as well as limit their generation of
chemical waste. Unfortunately, the
ratio between the regenerated and
the fresh catalyst market in distil-
lates hydroprocessing is below 50%,
meaning that not all spent catalysts
are regenerated at least once. This
low reuse rate could certainly be
increased by careful evaluation of
the spent catalyst.
Regeneration of hydroprocessing
catalysts restores activity
The end-of-cycle for hydroprocess-
ing units is usually determined by:
a scheduled unit shutdown; a unit
upset, such as too high a pressure
drop or compressor failure; or a
catalyst activity that is too low for
meeting product specifcations. In
Use of regenerated catalyst can be maximised by careful sampling and analysis of
spent catalyst to construct a reactor contamination profle
PIERRE DUFRESNE Eurecat
FRANCOIS LOCATELLI Eurecat France
this last case, activity decrease is
mainly due to partial blockage of
the active sites by coke and some-
times contaminants. Carbon content
on spent HDS catalysts largely
varies from 5 to 25 wt%, with an
average for diesel units of 8-15
wt%. It is common knowledge now
that the regeneration of spent
hydroprocessing catalysts does
restore activity. By using an oxidis-
ing atmosphere at a temperature of
450-550C, carbonaceous species
can be removed from the catalyst.
Regeneration has the beneft of
eliminating the frst cause of deacti-
vation: coke deposit. But it does
more by converting the sulphide
phase, which may have become
partially sintered, back to an oxide
phase similar to the fresh catalyst.
This works well for conventional
catalysts, while the latest generation
of catalysts may need an additional
treatment to achieve better recovery
of activity.
How are hydroprocessing catalysts
contaminated?
Various contaminants are found on
spent hydroprocessing catalysts at
the end of cycle. They originate
either from the feed (vanadium,
nickel, arsenic and sodium), from
additives used during transport or
refning operations (silicon, lead
and phosphorous) or from corro-
sion (iron).
Ni + V
The nickel- and vanadium-contain-
ing molecules present in the heavier
fraction of vacuum gas oil and resid
feeds are readily decomposed
to nickel and vanadium sulphides
on the catalyst surface. After
regeneration, the sulphides are
converted back to oxides, also
promoting some alumina sulphata-
tion. Depending on the hydrotreating
application, it is usually not advised
to reuse catalysts containing more
than 1 to 3 wt% of vanadium (often
expressed as Ni+V, as these two
metals generally exist together, even
if nickel is not itself considered a
poison).
As
Organoarsenic compounds contained
in some crudes are extremely reac-
tive under hydroprocessing
conditions and are thus readily
decomposed and adsorbed on the
catalyst surface. Therefore, a very
steep arsenic gradient is always
observed between the top and the
bottom of the reactor. Arsenic
is considered a severe poison
for HDS activity, as a quantity of 0.2
to 0.4 wt% can prevent catalyst
reuse.
Na
Sodium is usually not present in
the various hydroprocessing feeds
in normal operation. However, it
could be present on the catalyst due
to desalter malfunction, because of
various upsets leading to upstream
introduction of caustic soda
(for instance, upsets of Merox
units), or when seawater is intro-
duced to the unit via heat exchanger
leakage. Sodium then adsorbs
on the catalyst surface and has a
signifcant deactivation effect,
which is usually more noticeable
after regeneration. The maximum
tolerated poison level for sodium
is typically between 0.1 and 0.2
wt%.
eurecat.indd 1 23/2/12 13:18:42
Si
Silicon originates essentially from
lighter fractions of coker or
visbreaker operations, where poly-
dimthylsiloxanes are used as
anti-foaming agents. Silicon could
also originate from silicon-based
chemicals injected into pipelines for
a reduction in pressure drop. Silicon
is not a very strong poison during
the cycle, and deactivation seems
higher after regeneration.
Pb, P
Lead is no longer seen in spent
catalysts in western countries, due
to the general phase-down of tetra-
ethyl lead as an octane booster.
Poisoning by phosphorous may be
seen sometimes, but not often, as
phosphorous-containing additives
are not so common yet.
Fe
In distillate hydrotreaters, iron
contamination comes from the
corrosion of upstream equipment.
As such, iron scale or fne particles
usually do not penetrate deeply
into the catalyst pores and do not
have any strong poisoning effect.
Its contribution to pressure drop
build-up is more problematic, as
iron particles may accumulate at
the top of the bed or in the inter-
stices between catalyst granules.
How to maximise the quantity
of reusable regenerated catalyst.
Sorting is a solution.
Regeneration, however, will not
cure anything as regards deactiva-
tion by contaminants, which remain
adsorbed after thermal treatment.
This does not mean that a contami-
nated batch cannot be regenerated.
What is necessary is to proft from
the concentration gradient that is
always observed in fxed-bed reac-
tors in case of contamination.
SAS (sampling, analysis, segregation)
service can save a great deal of good
catalyst
When catalyst is contaminated, the
analysis of a global representative
sample does not tell us anything
regarding the concentration gradi-
ents throughout the reactor bed and
is thus not suffcient. It could even
be counter productive, as it could
lead to wrong decisions, such as
sending the whole batch for metal
reclaim when clever sorting would
have saved a great deal of good
catalyst.
Contamination issues must be
considered before any reactor shut-
down. The sorting strategy simply
starts by good-quality labelling of
the drums or containers of unloaded
catalyst at the foot of the reactor.
This will later enable accurate inter-
pretation of the analysis of each
container by drawing the contami-
nant profle and then deciding on
the subsequent segregation of good
product from contaminated mate-
rial. The catalyst owner may decide
on a segregation strategy: the entire
catalyst batch is divided into several
lots of varying quality, which can
be either reloaded in the same unit,
1.20
1.60
1.40
1.00
0.80
36
0.60
0.40
0.20
0 27 30 33 39 42 45 48 3 6 9 12 15 18 21 24 54 57 60 63 66 69 72 51
%
t
w
Bin number
0.00
As
Si
Na
Figure 1 SAS service. Contaminants analysis (sodium, arsenic, silicon) vs bin number on a
load of spent HDS catalyst
used in another unit of similar serv-
ice, cascaded down to a less severe
service, or sent for metals
reclamation.
Case study 1 : SAS saved around 75%
of the catalyst batch (equivalent to
1 million savings as fresh catalyst)
The following is an interesting
example of Eurecats SAS service,
which resulted in considerable
savings for the refner. A reactor
containing 96 tons of catalyst was
gravity unloaded in well-labelled
containers. Upon arrival at the
Eurecat France site, the lot was
sampled in the standard way, and a
representative sample of the whole
reactor was analysed and tested.
Contaminant levels were pretty
high: 0.11 wt% arsenic; 0.35 wt%
sodium; and 0.5 wt% silicon.
Moreover the HDS activity was
found to be 83% that of fresh cata-
lyst. With these results in hand, the
customers initial decision was to
send the entire 96-ton batch for
metals reclamation.
Eurecat then proposed to examine
the lot in more detail and perform
a complete analysis on every third
container to determine the contami-
nation profle. Figure 1 shows the
contamination profle, with a rather
clear cut between clean and contam-
inated fractions.
The high level of contaminants in
the frst containers illustrates the
so-called chimney effect, where
top layers come frst just after the
very bottom part.
A composite sample of containers
1 to 3 and 24 to 74 was obtained,
analysed and tested. In these
composites, the level of contami-
nants is low (0.02 wt% As; 0.0 5
wt% Na; 0.1 wt% Si) and the HDS
activity came back at 96% that of
fresh catalyst. This detailed contam-
ination analysis coupled with HDS
testing made the customers deci-
sion much easier: recycle the
contaminated fraction of only 22
tons and regenerate the remaining
74 tons of good material.

Case study 2: a saving of 65%
(equivalent to 700,000)
The SAS procedure was applied on
another case of a 75-ton reactor,
where silicon and arsenic contami-
eurecat.indd 2 23/2/12 13:19:06
For more information about UOP, visit www.uop.com
2010 UOP. All Rights Reserved.
refning your proft
UOP helps you exceed your goals with innovative technology,
catalysts and optimization solutions specifcally designed to
meet your needs.
UOP hydroprocessing solutions and optimization services are designed to help
you maximize your return on investment and grow your business. As regional
market demands shift, we provide the process technologies, catalysts and
services that will meet your changing business needs. Our experts work closely
with you to meet your desired yields and product specifcations including
ultra-low sulfur diesel standards while improving your operational effciency.
Backed by over 50 years of hydroprocessing innovations, UOP offers the best and most advanced solutions
to keep your business one step ahead.
UOP_Refining Your Profit_Ad_A4_PRINT.indd 1 5/18/10 2:10:44 AM
uop.indd 1 23/2/12 12:23:31
ad copy 2.indt 1 23/2/12 16:19:20
nation was suspected. The arsenic
concentration of the whole batch is
0.11 wt%, not high enough to
decide to eliminate all of the batch,
but still preventing the batch from
performing at the ideal level for an
activity-sensitive unit. The labelling,
sampling, analysis procedure
resulted in a very regular contami-
nation profle.
The cut was decided as follows:
49 tons were recovered with an
arsenic level below 0.05 wt%, and
26 tons with 0.21 wt% arsenic and
0.51 wt% silicon were sent for metal
reclaim. HDS activity was upgraded
from 89% to 94% RVA compared
with fresh catalyst.
Spent catalyst analysis can assess
average liquid feed quality over
the run
The following example demon-
strates that the SAS service is useful
for understanding the average feed
quality with regards to contami-
nants over the run. A simple way
of assessing average feed contami-
nant content is the rule of 4 one:
1 ppm of contaminant in the feed
processed by a catalyst over one
year at a liquid hourly space veloc-
ity of one will result in 1 wt%
deposit of contaminant.
Figure 3 shows the arsenic
concentration profle of an HDS
reactor for two consecutive cycles.
Arsenic is repeatedly present, but
with average quantities, which vary
from one run to another. Thus, the
quantity of uncontaminated catalyst
varies each time. The good news is
there is always something to
recover for regeneration.
Some catalyst can be saved even in
the event of highly polluted feed
Another example illustrates why
the SAS procedure is always justi-
fed. Spent catalyst from an FCC
pretreat unit was analysed at 1.2
wt% arsenic. It is seldom that such
an arsenic content is found on a
catalyst batch. So far, a decision
was easy to make for the catalyst
owner: send the whole batch for
metals reclaim. Actually, the reac-
tivity of arsenic species contained
in crude are so high that, even in
the event of highly polluted feed,
adsorption takes place in the frst
layers of the bed. In this particular
case, the arsenic concentration
peaks at 3.7 wt% (see Figure 4). Still
slightly less than half of the batch
can be recovered.
SAS service can also establish carbon
profle within the reactor
Unit parameters, such as tempera-
ture, hydrogen and hydrogen
sulphide partial pressures, as well
as feed quality, vary all along the
various catalyst beds, and thus the
coke formation kinetics also vary
with those parameters. SAS service
can generate a precise carbon
profle of the reactor, which then
can be useful for helping to defne
the best strategy in terms of catalyst
choice and unit operation.
Usually coke increases from top to
bottom, mainly due to the highest
temperature and lowest hydrogen
partial pressure at the bottom. But
the gradient is more or less
pronounced for each specifc case.
In the same example, Figure 4
shows the coke profle: a high
initial value, corresponding to
the reactor bottom, then the
top with a lower carbon content,
the lowest corresponding to
the arsenic peak, then carbon
increases, as expected, from top to
bottom.
Catalyst reuse is not always
possible
Figure 5 shows a case where cata-
lyst recovery is not possible. This
example is an FCC pretreat unit
with a dirty vacuum gas oil, rich in
vanadium and sodium. Vanadium
and sodium concentrations are very
high at the top of the bed, but
remain above 1 wt% at
0.6
0.7
0.5
0.4
0.3
0.2
0.1
0 10 20 30 40 50 60
w
t
%
Bi n number
0
Fe
As
Si
Na
Recycling Regeneration
Figure 2 Contaminants analysis (Fe, As, Si, Na) vs bin number on a load of spent HDS
catalyst
1.2
1.0
0.8
0.6
0.4
0.2
0 50 100 150 200 250 300
t
w

,
c
i
n
e
s
r
A
0
As 2009
As 2008
Figure 3 Arsenic profle of an HDS reactor for two consecutive cycles
eurecat.indd 3 23/2/12 13:19:20
the bottom. Organic compounds
of those two elements are much
less reactive than arsenic
compounds and diffuse more easily
increase from top to bottom (6 to 14
wt%)
Activity testing is necessary for
assessing catalyst reuse
Until a couple of years ago, contam-
inant and surface area analysis were
suffcient for assessing safe reuse.
Eurecat has numerous statistical
data showing the correlation
between surface area and HDS
activity. One example, published in
2006 in Applied Catalysis, shows that
conventional catalyst surface area
recovery of more than 90%, or
better 95% (versus fresh catalyst),
would guarantee an activity recov-
ery of 90%. This corresponds to a
delta start-of-run temperature of
around 3C.
However, the situation has since
changed drastically with the
progressive replacement of conven-
tional catalysts with new-generation
catalysts. As a result, even if surface
area is still an important parameter
in catalyst characterisation, it is no
longer perfectly related to activity.
In other words, there are cases
where surface area is high enough
but activity is below the limit of
reuse, at 85 or 90%. But the reverse
can also apply, where surface area
is surprisingly lower than 80%
compared with fresh catalyst, yet
activity is above 90%.
This is shown in Figure 6 for vari-
ous new-generation CoMo catalysts.
Some points are clearly outside
direct correlation. Using the crite-
rion of surface recovery only would
have led to two types of mistakes:
accepting some batches with low
activity recovery, and discarding
some good batches despite their
low surface area.
This shows that true catalytic test-
ing is absolutely necessary for
assessing the safe reuse of regener-
ated catalyst. Eurecat has an
in-house capacity of up to 80 tests
per month for this purpose.
Assessing catalyst reuse by pressure
drop evaluation
Catalyst reuse can be achieved
safely as soon as two criteria are
met: activity recovery and mechani-
cal properties. The frst of these is
assessed by contamination studies
followed by sorting, as previously
6
5
4
3
2
1
0 100 200 300 400 500 600 700 800
%
t
w

,
t
n
e
m
e
l
E
Bin number
0
2
0
6
4
10
8
14
16
12
Vanadium
Coke
Sodium
Arsenic
Si|icon
Figure 5 Analysis (V, Na, As, Si and C) vs bin number on spent FCC pretreat catalyst:
partial recovery is not possible
3.0
4.0
3.5
2.5
2.0
1.5
1.0
0.5
0
4
0
8
0
1
2
0
1
6
0
2
0
0
2
4
0
2
8
0
2
0
6
0
1
0
0
1
4
0
1
8
0
2
2
0
2
6
0
3
0
0
E
l
e
m
e
n
t
,

w
t
%
Bi n number
0.0
Sodium
Silicon
6
8
7
5
10
12
11
9
4
3
2
1
C
a
r
b
o
n
,

w
t
%
0
6
8
7
5
10
12
11
9
4
3
2
1
C
a
r
b
o
n
,

w
t
%
0
Coke
Arsenic
Iron
Figure 4 Analysis (As, Fe, Si, Na and C) vs bin number on a highly contaminated spent
catalyst: partial recovery is possible
into the entire bed. In this particu-
lar case, the whole catalyst lot had
to be sent for metal reclaim. The
carbon profle showed a strong
95
105
100
90
85
130 140 150 160 170 180 190 200 210
R
W
A
,

%
SA, m
2
/ g
80
Figure 6 Variation of Surface Area (SA) and HDS Activity (RWA) for various regenerated
batches of a new generation CoMo catalyst.
eurecat.indd 4 23/2/12 13:19:31
discussed, and activity testing is
performed on the segregated clean
catalyst component. With this in
place, catalyst activity is well
defned for safe reuse. As regards
the second requirement concerning
mechanical properties, these are
evaluated by various analyses, such
as strength (bulk crushing or side
crushing), average length and parti-
cle size distribution, attrition level
and content of various sizes of
fnes. All of this helps to answer the
question: can this be reused without
any risk of pressure drop? A
complementary way to look at the
question is to measure directly the
pressure drop of a representative
sample at the laboratory scale.
Eurecat has developed and imple-
mented such equipment in order to
better assess catalyst reusability.
Figure 7 shows how fnes content
infuences the pressure drop of a
catalyst batch. Catalyst batches
have been selected with fnes (<1
mm) contents of 2, 5 and 10 wt%.
Pressure drop increases by about
60% and 130% respectively for a
fnes contents of 5 and 10 wt%.

Recycling spent non-reusable
catalyst
Used catalyst has to be recycled
when its performance cannot be
brought back to the desired level or
if the catalyst particles are too
damaged for reuse. (They are either
too weak or too short for safe reuse,
the risk being excess pressure drop
in the reactor). They are then
considered to be chemical waste.
Metals prices fuctuate a lot and
are now high enough to give spent
catalysts a positive value. Thus, the
debate between landflling and
reclaiming has ended, and all spent
hydroprocessing catalysts can now
be sent to metallurgical recovery
plants.
Two main routes are used for
metals reclamation: hydrometal-
lurgy and pyrometallurgy. Eurecat
has developed a pyrometallurgical
route with its partner Valdi, now
part of Eramet. This method has
several advantages over competi-
tive processes: the recycling is
performed not far from the regen-
eration plant in France, thus
limiting transport and guaranteeing
that the recycling operation meets
the highest environmental stand-
ards. In addition, this process
brings value to all of the catalyst
components: molybdenum, cobalt
and nickel are used to produce
speciality steels, while the recov-
ered alumina is used in the
production of rock wool.
Conclusion
Sound regeneration practice can
greatly contribute to an overall
reduction in hydroprocessing cata-
lyst costs. Careful analysis of spent
catalyst makes it possible to cherry-
pick the best-quality material for
reuse. An assessment of catalyst
quality needs to be done by direct
measurement of activity using pilot
plant testing. Finally, beyond all the
necessary analyses related to
mechanical properties, a direct
measure of bed pressure drop is a
guarantee of totally safe reuse of
catalyst.
Pierre Dufresne is Vice President, Research
and Development, for Eurecat SA. He holds a
degree from the Chemistry School of Mulhouse
and a PhD from the University of Lille. He
occupied various positions at Institut Franais
du Ptrole dealing with hydroprocessing and
zeolites before joining Eurecat. He has co-
authored around 70 patents.
Email: p.dufresne@eurecat.fr
Franois Locatelli is Director Sales and
Marketing of Eurecat France SAS. He holds a
degree in engineering from the Lyon School
of Chemistry and a PhD in catalysis from
the University of Lyon. He was previously an
R&D engineer, then a commercial engineer
at Eurecat France, then at Al Bilad Catalyst in
Saudi Arabia, until returning as Director Sales
and Marketing at Eurecat France.
Email: f.locatelli@eurecat.fr
60000
50000
40000
30000
20000
10000
0 1 2 3 4 5 6 7
G
a
s

p
h
a
s
e

p
r
e
s
s
u
r
e

d
r
o
p
,

P
a
/
m
Ai rf l ow rat e, Nm
3
/ hr
0
Base case
Base case + 2% fines
Figure 7a Effect of fnes of less than 1 mm up to 2, 5 and 10 wt% on pressure drop
generation
6
10
9
8
7
5
4
3
2
1
0
.
0
-
0
.
2
0
.
6
-
0
.
8
1
.
2
-
1
.
4
1
.
8
-
2
.
0
2
.
4
-
2
.
8
3
.
0
-
3
.
2
3
.
6
-
3
.
8
4
.
2
-
4
.
4
4
.
8
-
5
.
0
5
.
4
-
5
.
8
6
.
0
-
6
.
2
6
.
6
-
6
.
8
7
.
2
-
7
.
4
7
.
8
-
8
.
0
8
.
4
-
8
.
6
9
.
0
-
9
.
2
9
.
6
-
9
.
8
0
Figure 7b Particle size distribution of a catalyst containing 10% fnes below 1 mm
Length (mm)
eurecat.indd 5 23/2/12 16:26:58
Experience
t he Sabi n di f f er ence
for PGM cat alyst recovery
and re ning.
We turn science into art for hi ghest
possi bl e ret ur ns and added value.
The sci ence of recovering and rening precious metal
catalysts is straightforward: state of the art technology. The
ar t of this process, however, is what makes Sabin different
from all others: thats the knowledge, experience, and expertise
gained from seven decades of successfully serving thousands of
organizations around the world. Wed be pleased t o count you
among t hem.
Learn more at sabinmetal.com
Sabin ExpTheDif.indd 2 1/15/12 11:12 AM
Experience
t he Sabi n di f f er ence
for PGM cat alyst recovery
and re ning.
We turn science into art for hi ghest
possi bl e ret ur ns and added value.
The sci ence of recovering and rening precious metal
catalysts is straightforward: state of the art technology. The
ar t of this process, however, is what makes Sabin different
from all others: thats the knowledge, experience, and expertise
gained from seven decades of successfully serving thousands of
organizations around the world. Wed be pleased t o count you
among t hem.
Learn more at sabinmetal.com
Sabin ExpTheDif.indd 2 1/15/12 11:12 AM
sabin.indd 1 23/2/12 12:26:26
Optimisation of integrated aromatic
complexes
B
enzene, toluene and xylene
(BTX) are typically produced
by distillation and/ or solvent
extraction processes from reformate
streams. The various raw aromatic
streams contain small amounts of
non-aromatic olefns and other
impurities, which are removed by
Tonsil products. This article
discusses a new generation of
Tonsil catalysts that are already in
use in the industry. The feedback
from large unit operations confrms
superior performance compared to
standard, granular-activated clay
products.
The general fow scheme of an
aromatic complex is shown in
Figure 1. To simplify the overview,
only three major sections are
shown: the treatment of the refor-
mate bottom; the purifcation of the
benzene/ toluene stream; and the
fne purifcation of the p-xylene
rerun. All three services are differ-
ent with regard to catalyst activity,
which is important for the reduc-
tion in the amount of olefns, and
catalyst selectivity, which is critical
for the purifcation of the p-xylene
rerun.
Before having a closer look at the
services, an understanding of the
mechanism and the specifc differ-
ences of available catalytic systems
is needed. The conversion of naph-
tha into BTX streams in the reformer
is always accompanied by the
formation of non-aromatic olefns
(see Figure 2).
These olefns are either poisons to
downstream units, such as p-xylene
extraction units or benzene alkyla-
tion units, or they make the
produced BTX streams off spec and
lower their value. The removal
A novel class of purifcation catalysts provides improved selectivity in aromatic
complex service
Axel Dker
Sd-Chemie AG
of olefns is done by acid
catalysed alkylation with aromatic
compounds, mainly benzene and
toluene, yielding molecules with
higher boiling range, which are
then separated in the next down-
stream distillation unit. Granular
Tonsil CO products have been
widely used in this application for
many years. Activation with
mineral acid converts natural clay
into highly effective catalysts of the
Tonsil CO series.
The alkylation reaction of olefns
and benzene or toluene needs
primarily Lewis acids and little
Brnsted acids. An excess of
Brnsted acids catalyses the polym-
erisation reaction of olefns into
gum, which blocks the catalyst
pores and reduces its effective oper-
ating lifetime (see Figure 3).
Depending on the operating
temperature, the catalysts can form
both species of acids. At tempera-
tures below 140C, mainly Brnsted
acidity prevails, whereas at temper-
atures above 165C the desired
Reformate
bottom
p-Xylene
rerun
BT
treatment
Crude
reformate
p-Xylene
extraction
BT
extraction
Xylene
isomerisation
Raffinate
BT
C
9+
Light
ends
p-Xylene
Figure 1 Schematic overview of an aromatic complex
Benzene
Toluene
Aliphatic olefins
Impurities:
Mixed
xylenes
H
2
Figure 2 Formation of olefns in the
catalytic reformer
sud chemie.indd 1 23/2/12 13:51:07
Standard catalytic clay products
are granules of 0.5 mm on average,
as are almost all competitive prod-
ucts. A typical reactor loading
requires sand grading to assure
proper loading without any loss of
product during operation and for
the avoidance of channelling. Due
to the small particle size, the pres-
sure drop over the reactors is not
negligible.
Tonsil APT are new types of cata-
lysts that not only look different,
but also show more active and
more selective performance. They
are in the shape of 4 mm
extrusions.
In Table 1, the most suitable
Tonsil APT products for the three
major services in aromatic
complexes are shown in compari-
son to the standard products. For
easier reference, the designations of
the products refect the major serv-
ice they are designed for: BT for
benzene/ toluene; mX for mixed
xylenes; and pX for para-xylene.
The use of Tonsil APT products is
benefcial with regard to loading
pattern, performance and cycle life.
A typical loading pattern of a reac-
tor loaded with Tonsil CO
compared to one loaded with Tonsil
APT is shown in Figure 6. The most
striking difference is the replace-
ment of the sand grading by Tonsil
APT extrusions, which leads to
approximately 30% more catalyst
volume loaded in the reactor.
The major advantages of Tonsil
APT products over Tonsil CO prod-
ucts can be demonstrated by the
example of the treatment of refor-
mate bottom cuts. This service is
the most demanding (see Table 1)
and the catalyst lifetime is normally
the shortest of all services.
Lewis acidity is active. During
normal operation at temperatures
well above 170C, the reaction
mechanism favours alkylation cata-
lysed by Lewis acids (see Figure 4).
During startup conditions, espe-
cially during the dry-out phase, the
operating temperature remains
below 140C and the undesired
polymerisation reaction occurs. This
can only be avoided by increasing
the operating temperature during
the dry-out phase as fast as possi-
ble. The use of a special Tonsil
product can help to considerably
shorten the dry-out phase.
The characteristics of three major
services in integrated aromatic
complexes are determined by the
concentration and molecular weight
of non-aromatic olefns and the
necessity of providing high selectiv-
ity. The olefn concentration is
measured as bromine index (BI);
the selectivity is mainly determined
by the need to minimise the forma-
tion of benzene by the dealkylation
of toluene and xylene. This is
particularly crucial in the fne puri-
fcation of p-xylene rerun. Table 1
summarises the different features
and categorises the three services
from high-activity service to high-
selectivity service.
Table 1 lists the Tonsil CO prod-
ucts that are the most suitable for
the preferred services, with Tonsil
CO 616GS the most active product
and Tonsil CO 610G the most selec-
tive one.
Catalysed by
Lewis acids
Catalysed by
Brnstedt acids
+
2
Alkylation
Oligomerisation
Figure 3 Alkylation versus oligomerisation equations
40 60 80 100 120 140 160 180 200 220
Temperat ure, C
Polymerisation
Alkylation
Figure 4 Alkylation versus oligomerisation as a function of temperature
Reformate bottom BT unit p-Xylene rerun
Type of service High activity Medium activity High selectivity
BI of feed 500-1200 70-200 20-50
BI of product <20 <10 <5
Tonsil CO product Tonsil CO 616GS Tonsil CO 616G Tonsil CO 610G
Tonsil APT product Tonsil APT mX Tonsil APT BT Tonsil APT pX
Characteristics of services for Tonsil CO + APT products
Table 1
The cost blocks
may vary with plant
location, but the
message is clear: any
catalyst replacement
is accompanied by
high costs
sud chemie.indd 2 23/2/12 13:51:19
www.hpa-usa.com www.hpa.sg
Reactor Turnkey Services
HPA offers all react or t urnkey services from
blinds t o blinds using st at e of t he ar t equipment .
We specialize in iner t -ent ry using t he lat est life
suppor t unit s. All our cat alyst t echnicians are
t rained in vessel rescue, rst aid, and CPR.
Hydropac Catalyst Dense Loading
The Hydropac allows t he sprinkling of cat alyst
in a cont inually uniform pat t ern at a rat e slow
enough t o let each par t icle set t le, but fast
enough for accept able loading t ime. The
Hydropac sit s just six inches below t he t rays,
is able t o rot at e bot h direct ions, and can
load around t ransfer t ubes and ot her int ernal
obst ruct ions.
Catalyst Unloading Services
Our cat alyst unloading services make use of
power ful vacuum unit s wit h 26 inches of vacuum.
Vessel Repairs and Retro-Fits
We have our U and R st amp allowing us t o do
complet e vessel welding, repairs, and inspect ion.
Hydroprocessing Associates is ISO 9001 and
OHSAS 18001 accredit ed, and ISNet world and
PICS compliant .
Hydroprocessing Associates is locat ed in t he
U.S.A. and Singapore, and can mobilize t o any
count ry.
Phone: + 1 832-794-7942
E-mail: pet er@hpa-usa.com
515004adhydrocarbonworld100415.indd 1 15-4-2010 9:02:48
hpa.indd 1 23/2/12 12:28:06
It is therefore of utmost impor-
tance to extend the lifetime in this
service, as each change-out is a
costly operation. The reactor system
is a lead/ lag system that allows for
changing the catalyst of one reactor
while having the entire aromatic
plant on stream. The reactor that
needs the replacement is placed off
line and is purged by nitrogen and
steam free of aromatics. Once that
is done, the catalyst is unloaded,
disposed and fresh material is
loaded. Thereafter, time-consuming
heating up and drying of the cata-
lyst is necessary.
Cost for this operation can be as
high as 30% of the costs of the
catalytic clay. An approximate
breakdown of the major costs is
shown in Figure 7. The cost blocks
may vary with plant location, but
the message is clear: any catalyst
replacement is accompanied by
high costs.
With a typical lifetime for Tonsil
CO 616 GS of four months, the costs
of change-out have to be paid, on
average, three times per year. That
means that in only one year, more
than the cost of one charge of
catalyst is spent by replacement
of the same. Tonsil APT mX can
signifcantly help to reduce the
costs of the treatment of reformate
bottoms.
The lifetime of any Tonsil product
is expressed in tonnes of feed that
can be processed per tonne of cata-
lyst. Consequently, the more weight
loaded into the reactor, the longer
the lifetime. This is easy, provided
the catalyst is of comparable or
better performance.
Tonsil APT mX has approxi-
mately 15% more active acidic sites
compared to Tonsil CO 616 GS,
which means it can process approx-
imately 15% more feed on a weight
basis. This is a breakthrough in the
technology of acid-activated clay
products, as Sd-Chemie was able
to change the shape while also
increasing the catalytic activity.
Any APT loading results in about
30% more catalyst volume (see
Figure 6). As all of these products
have approximately 4% higher bulk
density than the respective granular
products, it is possible to load about
35% more weight into a reactor.
Taking the higher catalyst activity
into this equation, a total of up to
60% longer lifetime results.
Table 2 shows the combined
advantages of Tonsil APT mX in
the service of treating the reformate
bottom cut.
The net result of the change from
a granular clay product to extru-
sions is a signifcant extension of
Tonsil APT
mX
Tonsil APT
pX
Tonsil APT
BT
Crude
reformate
p-Xylene
extraction
BT
extraction
Xylene
isomerisation
Raffinate
BT
C
9+
Light
ends
p-Xylene
Figure 5 Tonsil APT catalysts in aromatic plant service
Figure 6 Advantageous reactor loading
with Tonsil APT
Utilities
(nitrogen/steam
Disposal
Loss of product
Manpower
R
e
l
a
t
i
v
e

c
o
s
t
s
Figure 7 Typical cost breakdown of the
catalyst change
Tonsil CO 616 GS Tonsil APT mX Variance
Relative catalyst activity, ton feed/ton Tonsil 100 115 +15%
Bulk density, kg/m
3
720 750 +4%
Relative catalyst activity, ton feed/m
3
Tonsil 720 863 +20%
Loaded volume, m
3
100 130 +30%
Catalyst lifetime, months 4 6.3 +60%
Catalyst consumption, batches per year 3 2
Catalyst consumption, ton per year 288 195 -30%
Comparison of Tonsil CO 616 GS and Tonsil APT mX
Table 2
sud chemie.indd 3 23/2/12 13:51:33
the service life. In this example, the plant operator has
to change out the catalyst only twice per year and not
three times, as the lifetime was extended from four
months to more than six months. The yearly catalyst
consumption using Tonsil APT mX is 30% lower, the
number of shutdowns is less and, consequently, the
overall yearly cost of operation is greatly reduced (see
Figure 7).
The situation in the other services of any aromatic
complex is similar; the change from a granular clay to
the respective APT grade results in up to 50% longer
service life.
Many factors infuence the choice of the optimum
catalytic clay. An additional factor is the duration of
the startup, meaning how long it takes to produce with
full capacity a product that meets the defned specifca-
tion. Here, we have to consider two major issues: the
formation of water during the dry-out phase, and the
formation of benzene in the treatment of p-Xylene
rerun cuts.
It was mentioned earlier that the catalytically active
acid species has to be of mainly Lewis type, which is
achieved at temperatures above 170C. When drying out
large catalyst volumes, large volumes of water are
released from the reactor. Due to the heat integration of
the downstream distillation columns, water can be
handled only in smaller amounts. Therefore, the opera-
tor cannot increase the catalyst temperature as fast as
needed. As a negative consequence, it is not possible to
run the plant at full capacity during this period of time.
To overcome this, a catalyst grade with low moisture is
applied. Due to the low water content of the fresh cata-
lyst, the dry-out period is very short and the unit can be
put into full-capacity operation in less than three days.
The selectivity in the fne purifcation of p-xylene
rerun streams is defned by the formation of benzene.
Benzene needs to be avoided, as it reduces the eff-
ciency of the downstream p-xylene extraction unit and
is typically limited to a maximum of 50 ppmwt at the
inlet of the extraction unit. Tonsil APT pX is the most
selective catalyst among the series and delivers this
specifcation within a very short time. Figure 9 shows
the evolution of the benzene concentrations down-
stream different catalytic clay products.
Table 2
Cata|yst
Change-out
Cata|yst
Change-out
Cata|yst
Tonsi| CO 616 GS
3 batches per year
Change-out
s
t
s
o
c

e
v
i
t
a
|
e
R
Cata|yst
Change-out
Cata|yst
Tonsi| APT mX
2 batches per year
Change-out
Figure8Cost comparison of Tonsil CO 616GS and Tonsil APT mX
150
175
125
100
75
50
25
t
w

m
p
p

,
e
k
a
m

e
n
e
z
n
e
B
Time on-stream
0
Tonsi| APT pX
Product A Product A Product A
Figure9 Benzene make over Tonsil APT pX
It is now proven on a large industrial scale that
Tonsil APT catalysts achieve more than 50% higher
productivity at much lower operating costs in any
existing clay tower of any aromatic complex. The
change from standard granular clay products to
Tonsil APT catalysts can be done easily.
Tonsil is a registered trademark of Sd-Chemie.
Axel Dker is Director of Sales EMEA/Refnery, Sd-Chemie AG,
Munich. He has over 20 years experience in catalysis and holds a PhD
in chemistry from the University of Munich.
-).)6!0/NLINE
NEW!
0rabner nstruments Messtechnik 0mbH / A-1220 Vienna/Austria / Dr. 0tto-Neurath-0asse 1
Phone +3/1/282 1 27-0 / Fax +3/1/280 73 3 / inIo.grabner-instrumentsDametek.at
4.-0.0#1130#
.0-!#11,*78#0
- VP oI gasoline, crude oil and LP0
- A5TM D511, 5188, 377, 378,
87, EN 1301-1+2, P3, 0, 81
- Highest accuracy Ior best
possible blending to oIIicial limits
- Up to 2 sample streams
- Automatic calibration
- Fast and easy maintenance
- Automatic lubrication
570002.indd 1 12/7/2010 10:30:28 AM
sud chemie.indd 4 23/2/12 13:51:47
14 - 16 May 2012
COLOGNE, GERMANY
www.wraconferences.com/globalpetrochemicals
GLOBAL
PETROCHEMICALS
CONFERENCE
9
th
Annual Meeting
Sponsored by Supported by
STRATEGIES TO ACHIEVE PROFIT OPTIMISATION IN A VOLATILE MARKET
e 9
th
Annual Global Petrochemicals Conference is the
only event to bring together the senior decision makers
from petrochemical producer companies from the major
players in the market.
KEY PANEL MEMBERS INCLUDE
Fernando Iturrieta, Chief Executive Ofcer, CEPSA Quimica
Andreas Gupper, Strategic Planning Manager - Strategy & Group Development,
Borealis
Fred du Plessis, Executive Advisor, Corporate Strategy & Planning, Sabic
Pierre de Kettenis, Executive Director, CEFIC Petrochemistry Programme
Junnan Sun, European Manager, China Petrochemical Corporation (SINOPEC)
Rob Parry, Executive Account Manager, Chemical Catalysts, BASF, UK
Saleh Ali Al-Zahrani, Manager, Global Technology Marketing, Licensing, Sabic,
Saudi Arabia
To view the latest programme and speakers line up visit
www.wraconferences.com/ globalpetrochemicals
To conrm your place, please contact:
Anca Sterian, Marketing Manager
+44 (0) 20 7067 1800 a.sterian@theenergyexchange.co.uk
Register before 2 March 2012 and save up to 380. Quote PTQ.
R
E
G
I
S
T
E
R

N
O
W

A
N
D

S
A
V
E

3
8
0
WORKSHOP
COCKTAIL
RECEPTION
14 May,
Radisson Blu Hotel
NETWORKING
LUNCH
15 May,
Radisson Blu Hotel
GPC GALA
DRINKS
15 May,
Radisson Blu Hotel
wra.indd 1 23/2/12 12:29:50
Troubleshooting a FCC unit
F
luid catalytic cracking (FCC) is
one of the major processes in
the refning industry for
converting heavier hydrocarbons to
useful middle distillates. Such a
process demands the continuous
addition of fresh FCC catalyst and
results in the generation of tonnes of
spent equilibrium FCC catalyst. To
understand the FCC units opera-
tion, the process can be divided into
six sections: feed preheater, reactor,
regenerator, main fractionator, gas
plant and treating facilities. The
cracking reactions occur in the reac-
tor zone and this leads to coking.
Coked catalyst is circulated back to
the regenerator to burn the coke at
high temperature in the presence of
air. The regenerator hot fue gas
(containing CO and CO
2
) holds an
appreciable amount of energy. In
most units, the fue gas is routed
through a steam generating boiler
(referred to as a CO boiler), where
the carbon monoxide in the fue gas
is burned as fuel to provide steam
for use in the refnery as well as to
comply with any applicable environ-
mental regulatory limits on carbon
monoxide emissions.
FCC units generally experience
catalyst-related problems such as
circulation, catalyst loss and activity
The source of a fouling problem in a FCC CO boiler was identifed by systematic
analysis of fresh catalysts, additives and equilibrium catalysts
Chiranjeevi ThOTa, Shalini GupTa, DaTTaTraya Tammanna GOkak, ravi kumar vOOlapalli, p v C raO
and viSwanaThan pOyyamani SwaminaThan Bharat Petroleum Corporation
decline. The FCC units studied were
encountering low yield problems
due to a decrease in catalyst activity.
During the same period, it was also
observed that the CO boiler became
fouled externally, and a lot of fnes
were found in the CO boiler stack
during cleaning. Subsequent to star-
tup after cleaning, the CO boiler
started to experience fouling again.
There was also a steady decline in
activity of the e-cat inventory.
methodology
The problem of fnes deposition in
the CO boiler stack section, as well
as catalyst activity loss, was investi-
gated systematically by analysing
particle size, attrition index, surface
area and chemical composition in
terms of metals, to arrive at a possi-
ble solution to enable smooth
operation of the plant.
Samples of e-cat, fresh catalysts
and CO boiler samples from FCC
units were obtained from refneries,
and these samples were analysed in
the laboratory. The experimental
techniques used to characterise the
catalyst samples and the results are
discussed in this article.
results
In order to understand the reasons
for CO boiler fouling in the FCC
unit, a systematic study was carried
out by measuring properties such as
particle size distribution, attrition
index and magnesium metal concen-
tration for fresh FCC catalysts,
additive and e-cat samples collected
at different time intervals.
particle size analysis
In Table 1, it is clear that the fraction
of the catalyst coded -45 increased
from 12 to 20 wt%. Particle size
distribution was determined in
accordance with the ASTM D4513-
97 method using a sieving
procedure. Typically, about 50g of
FCC catalyst sample was sieved
through a test sieve column compris-
ing 180, 150, 106, 90, 75, 63, 53, 45,
32, and <20m sieves for 30 minutes
using an auto-sieve shaker. After
sieving, fractions collected over each
sieve were weighed and PSD calcu-
lated as per the ASTM method.
Average particle size decreased
from 72 microns to 67 microns. This
drop corresponds to an increased
amount of fnes in the catalyst. In
order to substantiate this further,
e-cat samples collected at different
time intervals were also character-
ised for particle size (see Figures 1
and 2).
Sample code, wt% -180 -150 -106 -90 -75 -63 -53 -45 -32 -20 (apS)
Fresh Cat-1 99 98 83 69 34 37 25 12 5 1 72
Fresh Cat-2 100 98 87 75 61 44 31 20 10 3 69
Fresh Cat-3 99 98 88 77 61 41 24 15 3 0.1 69
Fresh Cat-4 100 99 90 75 58 44 25 16 3.5 0.27 71
Fresh Cat-5 99 99 90 77 62 44 28 19 5.7 0.6 67
Fresh Add-1 98 97 81 53 33 19 12 6 1.2 0.1 88
particle size distribution of fresh catalysts and additives (aSTm D4513)
Table 1
bpcl.indd 1 23/2/12 16:28:11
21
st
- 23
rd
May 2012 - Hesperia Tower, Barcelona, Spain
www.refningsummit.com
In Association With:
The 2012 Summit will look at:
Leading experts include:
Researched and produced by:
How to achieve operational excellence
Best practices for bottom of the barrel
lmproving energy effciency in refnery
Excelling clean fuel production
Maximising diesel production
B.R. Mehta, Senior Vice President, Reliance lndustries Limited
Paulo Roberto Costa, Member of the Executive Board and
Downstream Offcer, Petrobras, Brazil
Philippe Marchand, Director of Biofuels and Regulations for the
Refning and Marketing, Total
Frigyes Lestak, General Manager CO2 Effciency, Shell
Pierre Dechamps, Adviser for Energy, Climate Change and the
Environment, European Commission
His Excellency Abbas Ali Al-Naqi, Secretary General, OAPEC,
Kuwait
For more information
and to save 200 on your booking (attend for 1295) please visit
www.refiningsummit.com or contact Richard Jones on +44 (0) 207 202 7690
or email richard.jones@wtgevents.com
The Global Refning Summit 2012 the must attend conference for anyone involved in
the refning and processing of hydrocarbons. The event may also be of interest to those
involved in the health, safety and environment aspects of the refning sector.
Great opportunity to meet and learn from other refning
people who share their views and approaches to current
and future challenges
Head of Technical Services, Petrobrazi Refnery, OMV Petrom
Fruitful opportunity to exchange views and learn new
ideas for optimising refnery operations.
Senior Expert, Refning, Organisation of Arab Petroleum
Exporting Countries (OAPEC)
global ref.indd 1 23/2/12 12:32:32
be in the range 30-40 m
2
/ g. A sample
from the CO boiler was found to
have a surface area of 9 m
2
/ g. The
very low surface area of CO boiler
fnes was due to their exposure to
very high temperatures (>950C),
which could have resulted in pore
sintering and formation of non-
porous material; hence, it could not
be attributed to either additive or
The average particle size distribu-
tion of fractions -45 and -63 were
analysed over a period of one year.
Figures 1 and 2 show that average
particle size is more or less constant
around 64 microns before the addi-
tion of additive. After the
introduction of additive, the average
particle size increased initially (from
64 to 75 microns), then stabilised
around 65 micron. This increase in
average particle size is an indication
of less retention of fnes in the unit.
The -45 fraction also fuctuated in
average particle size after the intro-
duction of additive, before
stabilising. During this period, a lot
of fuctuation was noted in the unit
with respect to catalyst. Hence, the
decrease in particle size leads to less
retention of catalyst in the unit and
ultimately leads to fouling of the
boiler.
Attrition index
Attrition index was measured for
fresh FCC catalysts, which were
used at different intervals of time,
and also e-cat samples and catalyst
additive samples (see Table 2).
Attrition index was measured as per
the ASTM D5757 procedure. It was
observed that the additive attrition
index was on the high side (10 wt%).
Table 2 shows that the attrition loss
for fresh catalyst batches was also
on the high side. Further, it was
observed that the attrition value for
the e-cat-1 sample (4.7%) was in the
acceptable range, while for the e-cat-
2 and e-cat-3 samples attrition losses
(5.7 and 7.1 wt%, respectively) were
on the high side. This observation
was attributed to attrition-prone
catalyst and additive usage in the
unit at that time.
Surface area measurements
To further understand the source of
fnes, the surface areas of fresh FCC
catalyst fnes (<40 micron), additive,
attrited fresh catalyst and CO boiler
fnes were analysed. The surface
areas of e-cat, fresh catalyst and fresh
additive fnes were comparable with
those of the original sample. Fresh
additive fnes and attrited fresh cata-
lyst were heated to 1000C, and the
measured surface area was found to
74
78
70
66
62
58
54
1
4

M
a
r

2
0
0
4
3

M
a
y

2
0
0
4
2
2

J
u
n

2
0
0
4
1
1

A
u
g

2
0
0
4
3
0

S
e
p

2
0
0
4
1
9

N
o
v

2
0
0
4
8

J
a
n

2
0
0
5
2
7

F
e
b

2
0
0
5
A
P
S
50
Figure 1 APS variation with respect to time
24
28
20
16
12
8
4
1
4

M
a
r

2
0
0
4
3

M
a
y

2
0
0
4
2
2

J
u
n

2
0
0
4
1
1

A
u
g

2
0
0
4
3
0

S
e
p

2
0
0
4
1
9

N
o
v

2
0
0
4
8

J
a
n

2
0
0
5
2
7

F
e
b

2
0
0
5
4
5

f
r
a
c
t
i
o
n
0
Figure 2 Variation in -45 fraction with respect to time
Catalyst Attrition loss, wt%
Fresh Cat-1 5.0
Fresh Cat-3 6.9
Fresh Cat-6 9.3
Fresh Add-1 10.0
E-cat-1 4.7
E-cat-2 5.7
E-cat-3 7.1
Attrition index data (ASTM D5757)
Table 2
S.No Catalyst S.A, m
2
/g
1 Fresh Cat-1 349
3 Fresh Add-1 100
5 E-cat-4 204
7 CO boiler fnes surface area 9
8 Additives fresh fnes, heated to 1000C 30
9 Fresh Cat-3, attrited fnes, heated to 1000C 40
Surface area data for fresh catalyst, additive, e-cat and CO boiler samples
Table 3
bpcl.indd 2 23/2/12 14:10:09
FCC catalyst fnes. Surface area meas-
urements for catalyst samples were
carried out using nitrogen adsorp-
tion/desorption measurements in an
Autosorb-1MP unit. Nitrogen adsorp-
tion/desorption isotherms were
measured at -196C after degassing
about 50 mg of sample below 10-3
torr at 300C for three hours. BET
specifc surface area was estimated
by following ASTM method D4365
using adsorption data in a relative
pressure range from 0.008 to 0.08 bar.
Surface area values for various
samples are shown in Table 3 as per
ASTM D4365.
Chemical analysis
Chemical analysis was carried out
on various samples to ascertain the
source of CO boiler fnes by measur-
ing the magnesium content. Additive
contains magnesium metal and its
concentration should rise with the
inventory build-up in the unit. A
microwave digestion procedure has
been standardised for quantifying
magnesium in different fractions of
CO boiler fnes. A microwave diges-
tion system equipped with 14 vessels
and designed for pressures up to
800 psi was used. Each vessel is
charged with 0.2-0.5g of sample, to
which is added 10 ml of nitric acid,
3 ml of distilled water and 1 ml of
hydrogen peroxide. After about 15
minutes, when the frst vigorous
reaction has taken place, the pres-
sure vessels are closed and put into
a microwave oven to complete the
one-stage digestion. Metal content is
measured using the ICP technique.
Various parameters of microwave
digestion system were optimised for
obtaining a clear solution for sample
analysis. The optimised program
and parameters used in the one-
stage digestion are shown in
Table 4.
To understand additive retention
in the unit, an e-cat mass balance
over a period of four months was
done by analysing magnesium
content at different time intervals.
Figure 3 compares the expected and
actual magnesium content of the e-
cat. The actual magnesium content
of the e-cat is lower than the
expected values. The mass balance
assumes both catalyst and additive
are removed from the system
(either through cyclone losses or
catalyst purging) at the same rate.
Ideally, the estimated and actual
Mg concentrations should be the
same. The difference suggests that
additive retention is 85% of the
main catalyst. Perhaps excessive
generation of fnes with additive
may be contributing to this lower
retention. Hence, it may be neces-
sary to improve the attrition
resistance of the additive.
Based on magnesium analysis of
CO boiler fnes alone, the problem
cannot be attributed solely to either
catalyst or additive, as the fnes
could have accumulated over a
period of time and the addition of
additive may have started mean-
while. Magnesium and rare earth
oxide distribution in different-sized
fractions of the e-cat was analysed.
Magnesium and RE
2
O
3
distribution
with different-sized fractions of e-
cat are shown in Table 5. The table
shows that, with respect to particle
size, magnesium and rare earth
content decreases, which indicates
that a loss of fnes leads to a
decrease in activity. The effect is
shown in Figure 4.
Systematic analysis of fresh FCC
catalyst, equilibrium catalysts and
fnes collected from the CO boiler
at different time intervals, and anal-
ysis of fresh additive samples, was
employed to identify problems and
enable corrective actions to be
0.8
0.6
0.4
0.7
0.5
0.3
0 30 60 90 120
M
a
,

%
Days
0.2
Actual
Mass balance
Figure 3 Comparison of actual and calculated Mg content in e-cat
60
80
70
50
40
30
20
10
Dec May Aug Dec May
A
c
t
i
v
i
t
y
0
Figure 4 Activity variation with respect to time
Stage Power RAMP minuets Control pressure, psi Hold time, min
Max %
I 1200 W 100 30.00 120 30.00
Microwave system parameters for digestion
Table 4
bpcl.indd 3 23/2/12 14:11:06
to fnes, regular monitoring of the attrition index as well
as particle size analysis of the FCC catalyst and additive
would be good practice
Systematic physico-chemical characterisation of fresh
catalyst, e-cat and CO boiler fnes helped to diagnose
the CO boiler fnes and fouling problem, and enabled
suitable remedial measures to be suggested to
operations.

Acknowledgement
The authors gratefully acknowledge BPCLs management for
encouragement and permission to publish the paper.
Chiranjeevi Thota is a Deputy Manager at the Corporate R&D Centre,
Bharat Petroleum Corporation, India. He holds a MSc in chemistry from
Andhra University, Visakhapatnam, and a PhD from Indian Institute of
Petroleum, Dehradun. Email: chiranjeevit@bharatpetroelum.in
Shalini Gupta is a Deputy Manager (R&D) with Bharat Petroleum
Corporation. She holds a masters in chemistry.
Email: guptashalini@bharatpetroleum.in
Dattatraya Tammanna Gokak is a Senior Manager at the Corporate R&D
Centre, Bharat Petroleum Corporation. He holds a MSc in chemistry from
Karnatak University, Dharwad, and a PhD from University Of Baroda.
Ravi Kumar Voolapalli is a Chief Manager at Corporate R&D Centre,
Bharat Petroleum Corporation. He holds a BTech in chemical engineering
from Andhra University, Visakhapatnam, a MTech in chemical engineering
from the Indian Institute of Technology, Kanpur, and a PhD in chemical
engineering from Imperial College of Science Technology and Medicine.
P V C Rao works on refning processes, biofuels, crude assay and
compatibility, fuel additives and new products development. He holds
a PhD in chemistry from Indian Institute of Technology, Bombay.
Viswanathan Poyyamani Swaminathan is a Chief Manager and Head
of the Corporate R&D Centre, Bharat Petroleum Corporation. He holds a
MSc and PhD in chemistry from Indian Institute of Technology, Delhi.
recommended for the smooth operation of the FCC
unit.
Conclusions
Systematic studies were conducted on different refnery
samples to troubleshoot operational problems in the
FCC unit. The studies led to the following conclusions:
The CO boiler fnes problem confrmed that in-use
fresh catalysts and additive are more prone to attrition,
so they contributed more to the fnes loss from the unit
and ultimately to the fouling of the boiler. More fnes
loss also affected the conversion loss, because fnes have
more zeolite and additive content
Even though the attrition index of additive is on the
high side, which may contribute to the total fnes, the
amount of additive used is relatively small compared to
the catalyst; hence, the additive alone cannot contribute
to the total fnes generated. This was established by
correlating the magnesium content of CO boiler fnes
with additive
Since both fresh catalyst and additive can contribute
Table 5
E-cat -2 Mg, wt% RE
2
O
3
, wt%
32 to 45 0.5850 1.62
45 to 63 0.4855 1.55
63 to 90 0.4384 1.45
>90 0.2780 1.02
Mg and RE
2
O
3
analysis of different fractions of e-cat
bpcl.indd 4 23/2/12 14:10:44
ACS-Amistco 30 & 33
www.amistco.com
Axens OBC
www.axens.net
BASF Corporation, Catalysts Division 11
www.catalysts.basf.com
Buchen-ICS 63
www.buchen-ics.com
Cat-Tech 25 & 27
www.cat-tech.com
CB&I 9
www.cbi.com
Chep Catalyst & Chemical Containers 39
www.chepccc.eu
Chevron Lummus Global 7
www.clg-catalysts.com
Criterion Catalyst & Technologies 2
www.criterioncatalysts.com
Digital Refeining IBC
www.digitalrefning.com
DuPont Sustainable Solutions 16
www.cleantechnologies.dupont.com
Eurecat France 22
www.eurecat.fr
Euro-Petroleum 48
www.europetro.com
ExxonMobil Research and Engineering Company 44
www.exxonmobil.com/refningtechnologies
Global Petrochemicals 2012 58
www.wraconferences.com
Grabner Instruments 57
www.grabner-instruments.com
Grace Davison 4
www.grace.com
Haldor Topse IFC
www.topsoe.com
Hoekstra Trading 43
www.hoekstratrading.com
Hydroprocessing Associates 55
www.hpa.sg
Johnson Matthey Catalysts 13
www.jmcatalysts.com
Porocel 40
www.porocel.com
Process Consulting Services 19 & 21
www.revamps.com
Sabin Metal Corporation 52
www.sabinmetal.com
Spectro Analytical Instruments 29
www.spectro.com
Sd-Chemie, a Clariant Group Company 34
www.sud-chemie.com
Tricat 37
www.tricatgroup.de
UOP 47
www.uop.com
Vereinigte Fllkrper-Fabriken 15
www.vff.com
Alphabetical list of advertisers
For more information on these advertisers, go to www.ptqenquiry.com
ad index copy 8.indd 1 24/2/12 13:44:48
www.digitalrening.com is a new,
easy-to-use website for the rening, gas
and petrochemical processing industries.
A simple search facility enables you to

nd and display relevant information
across the whole spectrum of articles,
literature, news, events and companies.
Readers can download technical articles
and other information from over 40
categories.

Find relevant industry events worldwide.

Read news items focusing on contracts,
company news and new products.
dig ref.indd 1 24/2/12 13:30:07
The winning catalyst combination
for your hydrocracker
Single source lSC 9001 lSC 14001 CHSAS 18001
www.axens.net
HRK, HDK, HK series: the expert trio to maximize
cycle length, activity and middle distillates selectivity
Hydro_cats_ad_fin.indd 1 26/7/11 12:35:06
axens.indd 1 23/2/12 12:33:46

2012 Catalysis

Hochgeladen von

Dokumentinformationen

Originalbeschreibung:

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

2012 Catalysis

Hochgeladen von

Copyright:

Verfügbare Formate

catalysis

17 Evaluation of a low rare earth resid FCC catalyst

r o| po|||o|o oro roo||, o00 c,c|os

inert ceramic balls

Tricat GmbH Catalyst Service Bitterfeld

bed temperature of 1350F (732C)

A simple search facility enables you to

Das könnte Ihnen auch gefallen